Olives and Olive Oil in Health and Disease Prevention [1 ed.] 0123744202, 9780123744203
Long used in sacred ceremonies and associated with good health, the nutritional and health promoting benefits of olives
681 57 17MB
English Pages 1520 Year 2010
Front Cover......Page 1
Olives and Olive Oil in Health and Disease Prevention......Page 4
Copyright Page......Page 5
Contents......Page 6
Preface......Page 16
List of Contributors......Page 18
Abbreviations......Page 30
Section 1 General Aspects of Olives and Olive Oil......Page 42
1.1 The Plant, Production, Olives and Olive Oil and Their Detailed Characterization......Page 44
1.2 TYPES OF OLIVE ACCORDING TO RIPENESS......Page 46
1.4 MAJOR PROCESSING METHODS......Page 49
1.5 CHARACTERISTICS OF FINAL PRODUCTS......Page 54
REFERENCES......Page 55
2.2 ORIGIN AND SYSTEMATICS OF THE OLIVE......Page 58
2.3 MOLECULAR MARKERS IN OLIVE GENOME ANALYSIS......Page 59
2.4 MOLECULAR MARKER APPLICATIONS......Page 62
2.5 FUNCTIONAL GENOMICS......Page 63
2.6 CONCLUSIONS......Page 64
REFERENCES......Page 65
3.2 BOTANICAL FEATURES OF OLIVE TREE......Page 66
3.3 PROTEOMICS......Page 67
3.5 PROTEIN EXTRACTION FROM OLIVE TISSUES FOR PROTEOMICS ANALYSIS......Page 68
3.6 CURRENT INITIATIVES IN OLIVE PROTEOMICS......Page 70
REFERENCES......Page 72
4.1 INTRODUCTION......Page 74
4.2 FEATURES OF CHEMOMETRIC CLASSIFICATION......Page 75
4.4 UNSUPERVISED CLASSIFICATION......Page 77
4.5 LINEAR DISCRIMINANT ANALYSIS (LDA)......Page 81
REFERENCES......Page 82
5.2 THE EFFECTS OF CLIMATE CONDITIONS ON QUALITY OF OLIVE OIL......Page 84
REFERENCES......Page 90
6.2 IRRIGATION MANAGEMENT......Page 92
6.4 SENSORY CHARACTERISTICS......Page 93
6.6 NATURAL ANTIOXIDANTS CONTENT......Page 95
6.7 VOLATILE COMPOUNDS......Page 97
SUMMARY POINTS......Page 98
REFERENCES......Page 99
7.2 EFFECT OF THE RIPENING PROCESS OF THE OLIVE FRUIT ON THE CHLOROPHYLL AND CAROTENOID FRACTIONS OF DRUPES......Page 100
7.3 PIGMENT COMPOSITION OF VIRGIN OLIVE OIL RELATED WITH THE RIPENING STAGE OF THE OLIVE FRUIT OF ORIGIN......Page 103
7.4 EFFECT OF THE OLIVE-RIPENING STAGE ON CHLOROPHYLL AND CAROTENOID PARTITIONING DURING THE OIL EXTRACTION PROCESS......Page 107
REFERENCES......Page 109
8.2 MACHINES FOR OLIVE CRUSHING......Page 110
8.4 THE INFLUENCE OF OLIVE PASTE PREPARATION MACHINES ON OLIVE OIL CHARACTERISTICS......Page 112
8.5 EFFECT OF DE-STONING ON OLIVE OIL QUALITY......Page 113
REFERENCES......Page 116
9.1 INTRODUCTION......Page 118
9.3 MALAXER MACHINE AND ITS EVOLUTION IN ENHANCING OLIVE OIL QUALITY......Page 119
9.4 MIXING CONDITIONS AND THEIR EFFECT ON THE OLIVE OIL QUALITY......Page 120
REFERENCES......Page 123
10.1 INTRODUCTION......Page 126
10.2 VIRGIN OLIVE OIL EXTRACTION SYSTEMS......Page 127
10.4 DIFFERENCES BETWEEN TWO- AND THREE-PHASE CENTRIFUGAL DECANTERS......Page 128
10.5 THREE- AND TWO-PHASE CENTRIFUGAL DECANTERS: OPERATION AND TECHNOLOGICAL INNOVATION......Page 129
10.6 EFFECT OF THREE- AND TWO-PHASE CENTRIFUGAL DECANTERS ON VIRGIN OLIVE OIL QUALITY......Page 130
SUMMARY POINTS......Page 132
REFERENCES......Page 134
11.2 OLP EXPRESSION......Page 136
REFERENCES......Page 141
12.2 CHROMATOGRAPHIC METHODS......Page 142
12.3 SPECTROSCOPIC/SPECTROMETRIC METHODS......Page 144
REFERENCES......Page 146
13.3 AIM OF THE WORK AND STATE OF THE ART......Page 150
13.4 DESCRIPTION OF THE EXPERIMENTAL WORK......Page 151
13.6 RESULTS OF [sup(1)]H HR-MAS NMR ANALYSIS......Page 152
13.7 STRUCTURE OF OLIVES......Page 153
13.8 DISCUSSION OF MRI RESULTS......Page 154
13.9 EVOLUTION OF WATER AND OIL DISTRIBUTION IN OLIVE WITH RIPENING......Page 155
13.11 FINAL REMARKS......Page 156
REFERENCES......Page 157
14.2 PRACTICAL ASPECT OF THE NMR-STATISTICAL PROTOCOL......Page 158
14.3 GEOGRAPHICAL CHARACTERIZATION OF OLIVE OILS......Page 161
REFERENCES......Page 164
15.2 [sup(13)]C SPECTRUM OF AN OLIVE OIL......Page 166
15.3 THE EFFECT OF THE CULTIVAR ON OLIVE OIL COMPOSITION......Page 167
REFERENCES......Page 169
16.1 INTRODUCTION......Page 170
16.2 INTRODUCTION TO CAIMAN THEORY......Page 171
16.3 EVALUATING THE CLASSIFICATION PERFORMANCES......Page 172
16.6 EXAMPLE OF CLASSIFICATION WITH CAIMAN: OLIVE OILS FROM SOUTH APULIA......Page 173
16.7 CAIMAN CLASSIFICATION ON THE OTHER CLASSES......Page 174
16.8 CLASSIFICATION WITH PLS-DA......Page 176
REFERENCES......Page 177
17.1 INTRODUCTION......Page 180
17.2 CONSIDERATIONS ON THE ANALYTICAL METHODOLOGY TO APPLY......Page 182
17.3 EVALUATION OF THE OXIDATION AND POLYMERIZATION SUBSTANCE CLASSES DURING REFINING OF LAMPANTE OLIVE OIL......Page 184
17.4 INVESTIGATION ON THE DEGREE OF OXIDATION AND HYDROLYSIS OF REFINED OLIVE OILS......Page 186
17.5 COMPARISON BETWEEN TRADITIONAL AND RECENT NONCONVENTIONAL PARAMETERS IN THE ASCERTAINMENT OF THE QUALITY OF OLIVE OIL COMMERCIAL CLASS......Page 187
17.6 QUALITY COMPARISON BETWEEN THE OLIVE OIL COMMERCIAL CLASS AND MARKETED REFINED SEED OILS......Page 191
SUMMARY POINTS......Page 193
REFERENCES......Page 194
18.2 SICILIAN OLIVE OIL......Page 196
18.4 TRACE METALS ANALYSIS: DERIVATIVE STRIPPING CHRONOPOTENTIOMETRY......Page 197
18.5 CANONICAL DISCRIMINANT ANALYSIS USED FOR THE CLASSIFICATION OF SICILIAN OLIVE OILS......Page 200
SUMMARY POINTS......Page 203
REFERENCES......Page 204
1.2 Components of Olives and Olive Plant Product and Uses......Page 206
19.1 INTRODUCTION......Page 208
19.3 HEALTH ASPECTS LINKED TO PHENOLICS IN VOO......Page 212
19.5 SENSORY PROPERTIES AFFECTED BY PHENOLICS IN VOO......Page 213
SUMMARY POINTS......Page 214
REFERENCES......Page 215
20.2 OLEA EUROPAEA CULTIVARS AND THEIR DIFFERENTIATION......Page 218
20.5 CHARACTERIZATION OF CULTIVARS BY THEIR PHENOLIC PROFILE......Page 219
20.6 PHENOLIC PROFILES OF PORTUGUESE CULTIVARS......Page 221
REFERENCES......Page 226
21.2 METHODOLOGICAL CONSIDERATIONS......Page 228
21.3 SELECTED ION ANALYSIS FROM THE FULL SCAN DATA: A POWERFUL AND VERSATILE TOOL TO DETECT AND IDENTIFY PHENOLS......Page 231
21.4 PHENOLIC CONTENTS IN SICILIAN OLIVE OILS......Page 238
REFERENCES......Page 240
22.1 INTRODUCTION......Page 242
22.2 MEASUREMENT OF OLIVE OIL VOLATILES......Page 243
22.3 VOLATILE PROFILES OF OLIVE OILS......Page 245
22.4 COMPARISON WITH AUSTRALIAN OLIVE OILS......Page 246
22.6 CONCLUSION......Page 248
REFERENCES......Page 249
23.2 THE OLIVE FRUIT CONTAINS A VAST ARRAY OF STEROLS AND NON-STEROIDAL TRITERPENOIDS......Page 252
23.3 CHANGES IN THE CONTENT OF FREE AND ESTERIFIED STEROLS AND NON-STEROIDAL TRITERPENOIDS THROUGHOUT FRUIT DEVELOPMENT......Page 254
23.4 HOW IS CARBON FLUX REGULATED BETWEEN BOTH TRITERPENIC PATHWAYS IN THE OLIVE FRUIT?......Page 257
SUMMARY POINTS......Page 258
REFERENCES......Page 259
24.2 DIFFERENCES BETWEEN VOLATILE COMPOUNDS IN VIRGIN OLIVE OIL AND IN TABLE OLIVES......Page 260
24.4 FERMENTATION......Page 261
24.7 INVESTIGATION OF THE VOLATILE COMPOUND CONTENTS IN SPANISH-STYLE, GREEK-STYLE AND CASTELVETRANO-STYLE GREEN OLIVES OF THE NOCELLARA DEL BELICE CULTIVAR......Page 262
SUMMARY POINTS......Page 268
REFERENCES......Page 271
25.1 INTRODUCTION......Page 274
25.3 CHANGES DURING RIPENING......Page 276
25.4 OLEUROPEIN CONCENTRATION IN FRUIT AND LEAF OF OLIVE DURING RIPENING......Page 277
REFERENCES......Page 278
26.2 MONOVARIETAL VIRGIN OLIVE OILS COMPOSITION AND CULTIVAR CHARACTERIZATION......Page 280
26.5 TOTAL FATTY ACID COMPOSITION......Page 281
26.7 ALCOHOL COMPOSITION......Page 283
26.9 CORRELATION BETWEEN OXIDATIVE STABILITY AND PHENOLIC COMPOSITION......Page 286
REFERENCES......Page 287
27.2 DEFINITION OF PHYTOSTEROLS......Page 290
27.5 ANALYSIS OF PHYTOSTEROL CLASSES IN OLIVE OIL......Page 291
27.6 THIN-LAYER CHROMATOGRAPHY......Page 292
27.9 GC AND GC-MS ANALYSIS OF PHYTOSTEROL CLASSES......Page 293
27.10 LEVEL OF PHYTOSTEROL CLASSES IN OLIVE OIL......Page 294
27.11 FREE AND ESTERIFIED PHYTOSTEROLS......Page 296
SUMMARY POINTS......Page 297
REFERENCES......Page 298
28.2 EXTRACTION METHODS......Page 300
28.3 COMPARISON OF PHENOLIC CONTENTS OF OLIVE MATERIALS......Page 302
28.4 DISCRIMINATING AND CLASSIFYING POWER OF OLIVE LEAF EXTRACTS......Page 303
28.5 INFLUENCE OF THE STORAGE CONDITIONS ON BIOPHENOLS FROM OLIVE LEAVES......Page 304
28.6 POTENTIAL OF PHENOLS FROM OLIVE LEAVES......Page 305
REFERENCES......Page 314
29.1 INTRODUCTION......Page 316
29.3 MELATONIN IN EDIBLE PLANTS: CORRELATION WITH BLOOD LEVELS OF MELATONIN......Page 317
29.4 MELATONIN AS A NEW PHYTOCHEMICAL IN OLIVE OIL......Page 319
29.5 CONCLUDING REMARKS: IMPLICATIONS OF THE PRESENCE OF MELATONIN IN OLIVE OIL HEALTH AND DISEASE PREVENTION......Page 321
REFERENCES......Page 322
30.1 INTRODUCTION......Page 324
30.3 OLIVE PHENOLS FROM OLIVE-OIL MILL BYPRODUCTS......Page 326
30.5 OLIVE BIOPHENOLS AS FOOD ANTIOXIDANT ADDITIVE......Page 327
30.8 FINAL SAFETY CONSIDERATION......Page 328
REFERENCES......Page 329
31.2 MAJOR COMPONENTS OF RAW OLIVES......Page 332
31.3 SPANISH-STYLE GREEN OLIVES......Page 333
31.4 UNTREATED GREEN OLIVES IN BRINE......Page 335
REFERENCES......Page 337
32.1 INTRODUCTION......Page 340
32.2 OLIVE FRUITS......Page 341
32.3 PROCESSED OLIVES......Page 342
SUMMARY POINTS......Page 345
REFERENCES......Page 346
33.1 INTRODUCTION......Page 348
33.2 DISCUSSION......Page 349
SUMMARY POINTS......Page 350
REFERENCES......Page 351
34.2 DISCUSSION......Page 354
REFERENCES......Page 356
35.1 INTRODUCTION......Page 358
35.2 INORGANIC ANIONS: NUTRITIONAL AND TOXICOLOGICAL ASPECTS......Page 359
REFERENCES......Page 365
36.2 OLIVE POD PURIFICATION......Page 366
36.3 OLIVE POD CHARACTERIZATION......Page 367
SUMMARY POINTS......Page 371
REFERENCES......Page 372
37.2 OLIVE BP MECHANISMS......Page 374
37.3 THE BPs IN MAC OLIVE DRUPES......Page 376
37.4 MAC OLIVE ENZYMES......Page 377
37.5 CONVENTIONAL BIOTECHNOLOGY ON MAC OLIVES......Page 378
REFERENCES......Page 380
38.2 INDUCTION OF OLIVE CELL CULTURE......Page 382
38.4 IMPLICATIONS OF STUDY RESULTS......Page 383
38.5 cDNA CLONING OF OSC......Page 385
38.7 CONCLUSION......Page 386
REFERENCES......Page 387
39.1 INTRODUCTION......Page 390
39.2 SAMPLING......Page 391
39.5 METHODS OF DETERMINATION......Page 392
39.7 SECOIRIDOIDS, SIMPLE PHENOLS AND RELATED COMPOUNDS......Page 393
REFERENCES......Page 396
40.2 OLIVE OIL MILL WASTEWATER......Page 398
40.3 PHENOLIC COMPOUNDS......Page 400
40.4 OMW MANAGEMENT......Page 402
REFERENCES......Page 404
1.3 Stability, Microbes, Contaminants and Adverse Components and Processes......Page 408
41.3 DE-BITTERING AND FERMENTATION......Page 410
41.4 LAB, NATURALLY OCCURRING MICROORGANISMS DURING TABLE OLIVE FERMENTATION......Page 411
41.5 USE OF LAB STARTER CULTURES IN THE TABLE OLIVE FERMENTATION......Page 412
SUMMARY POINTS......Page 415
REFERENCES......Page 416
42.2 MICROBIAL LIPASES: AN OVERVIEW......Page 418
42.4 LIPASE PRODUCTION: INDUCTION AND SECRETION......Page 419
42.6 IMPACT OF OIL DROPLET SIZE......Page 421
42.7 HYDROLYSIS VS OXIDATION OF FATTY ACIDS......Page 422
42.9 GROWTH INHIBITION BY LONG-CHAIN FATTY ACIDS......Page 424
42.11 APPLICATION OF LIPOLYTIC MICROBES AND LIPASES IN WASTEWATER TREATMENT......Page 425
REFERENCES......Page 426
43.2 PHENOLIC COMPOUNDS AND LACTOBACILLUS PLANTARUM......Page 428
43.3 METABOLISM OF PHENOLIC COMPOUNDS BY LACTOBACILLUS PLANTARUM......Page 429
REFERENCES......Page 436
44.2 MICROBIAL FLORA OF OLIVES......Page 438
SUMMARY POINTS......Page 445
REFERENCES......Page 446
45.1 INTRODUCTION......Page 448
45.3 METHODOLOGICAL CONSIDERATIONS FOR THE DETERMINATION OF AFB[sub(1) IN OLIVE OIL......Page 449
45.4 AFB[sub(1)] OCCURRENCE IN OLIVE OIL......Page 451
45.5 AFB[sub(1)] DAILY EXPOSURE: RISK ASSESSMENT......Page 452
REFERENCES......Page 453
46.1 INTRODUCTION......Page 456
46.2 CRITICAL POINTS FOR THE DETERMINATION OF PESTICIDE RESIDUES IN OLIVE OIL......Page 457
46.3 CROP PROTECTION PRODUCTS (PESTICIDES) RESIDUES IN OLIVE OIL FROM ORGANIC AND CONVENTIONAL CULTIVATION......Page 458
SUMMARY POINTS......Page 464
REFERENCES......Page 465
47.1 INTRODUCTION......Page 466
47.2 SIMULTANEOUS DETERMINATION OF PESTICIDES AND PAHS IN OLIVE AND OLIVE-POMACE OILS BY GC-MS/MS......Page 470
SUMMARY POINTS......Page 475
REFERENCES......Page 476
48.2 ACEPHATE AND BUPROFEZIN RESIDUES IN OLIVES AND OLIVE OIL......Page 478
REFERENCES......Page 480
49.2 THE INTERNATIONAL OLIVE OIL COUNCIL (IOOC) RECOMMENDED METHOD......Page 482
49.3 DETECTION OF VIRGIN HAZELNUT OIL INTO VIRGIN OLIVE OIL......Page 483
49.4 DETECTION OF REFINED HAZELNUT OIL IN VIRGIN OLIVE OIL......Page 485
SUMMARY POINTS......Page 488
REFERENCES......Page 490
50.1 INTRODUCTION......Page 492
50.2 METHODOLOGICAL CONSIDERATIONS......Page 493
50.4 APPLICATION OF MOLECULAR FLUORESCENCE SPECTROSCOPY TO OLIVE OIL ADULTERATION......Page 496
ACKNOWLEDGMENT......Page 500
REFERENCES......Page 501
51.1 INTRODUCTION......Page 504
51.2 DETERMINATION OF BTEXS IN OLIVE OIL SAMPLES BY THE DIRECT COUPLING HEADSPACE-MASS SPECTROMETRY......Page 506
51.3 EVALUATION OF DIFFERENT EXTRACTION TECHNIQUES FOR THE DETERMINATION OF BTEXS IN OLIVE OIL SAMPLES......Page 507
REFERENCES......Page 510
52.3 ACUTE STUDIES......Page 512
52.4 SUB-CHRONIC (90-DAY) TOXICITY STUDIES IN RATS......Page 513
52.5 REPRODUCTIVE AND DEVELOPMENTAL TOXICITY STUDIES IN RATS......Page 515
52.6 GENOTOXICITY/MUTAGENICITY STUDIES......Page 517
SUMMARY POINTS......Page 518
REFERENCES......Page 519
53.2 PLASTICIZERS......Page 522
53.3 CONCLUSION......Page 528
REFERENCES......Page 529
54.1 INTRODUCTION......Page 530
54.2 OCCURRENCE OF PAH IN FOODS......Page 532
54.3 OCCURRENCE OF PAH IN OLIVE OILS......Page 533
54.4 OCCURRENCE OF PAH IN OTHER VEGETABLE OILS......Page 534
54.5 CARCINOGENESIS OF PAH......Page 537
REFERENCES......Page 538
55.1 INTRODUCTION......Page 540
55.2 TOXICITY ASPECTS......Page 541
55.3 CONFIRMING THE MINERAL ORIGIN OF PARAFFINS......Page 545
REFERENCES......Page 546
1.4 Analytical Methods......Page 548
56.2 SAMPLE PREPARATION......Page 550
56.3 ANALYTICAL TECHNIQUES FROM: SPECTROPHOTOMETRIC DETERMINATIONS TO POTENT METHODS BASED UPON MS AND NMR, WHICH ARE ABLE TO DISCOVER AND CHARACTERIZE NEW PHENOLIC COMPOUNDS IN VOO......Page 552
56.4 CONCLUSIONS......Page 556
REFERENCES......Page 562
57.2 ELECTRONIC TONGUES......Page 566
57.3 ELECTRONIC TONGUES DEDICATED TO THE ANALYSIS OF OLIVE OILS......Page 569
SUMMARY POINTS......Page 571
REFERENCES......Page 572
58.2 OLIVE FRUIT ANALYSIS......Page 574
58.3 OLIVE OIL ANALYSIS......Page 576
58.6 CONCLUSIONS......Page 581
REFERENCES......Page 583
59.2 METHODS BASED ON FLOW INJECTION ANALYSIS......Page 586
59.3 METHODS BASED ON ELECTROCHEMICAL DETECTIONS......Page 588
59.4 SPECTROSCOPIC METHODS......Page 589
59.5 SEPARATION METHODS......Page 590
REFERENCES......Page 592
60.1 ELECTRONIC NOSE APPARATUS......Page 594
60.2 EVALUATION OF DIFFERENT STORAGE PERIODS AND CONDITIONS IN EXTRA VIRGIN OLIVE OILS......Page 595
60.3 EVALUATION OF GEOGRAPHICAL ORIGIN OF EXTRA VIRGIN OLIVE OILS......Page 597
REFERENCES......Page 599
61.2 SQUALENE......Page 602
61.3 TOCOPHEROLS......Page 605
REFERENCES......Page 607
62.1 INTRODUCTION: TRACING THE ORIGIN OF OLIVE OIL......Page 610
62.2 CHEMOMETRICS: WHAT IT IS AND WHY WE USE IT......Page 611
62.3 USING CHEMOMETRICS TO AUTHENTICATE ITALIAN OLIVE OILS: SOME EXAMPLES......Page 614
SUMMARY POINTS......Page 618
REFERENCES......Page 619
63.2 COMPOSITIONAL CHARACTERISTICS OF PORTUGUESE MONOVARIETAL OLIVE OIL FROM TRÁS-OS-MONTES......Page 622
63.3 CHEMOMETRICS APPLIED TO MONOVARIETAL OLIVE OIL CHARACTERIZATION......Page 627
REFERENCES......Page 629
64.1 INTRODUCTION......Page 632
64.3 EXTRACTION AND ISOLATION OF STEROLS IN OLIVE OIL......Page 633
64.4 ANALYSIS OF STEROLS BY LC......Page 635
64.5 CONCLUSIONS AND FUTURE TRENDS......Page 639
REFERENCES......Page 641
65.2 HIGH-RESOLUTION [sup(13)]C NMR OF OLIVE OIL TRIACYLGLYCEROLS......Page 644
65.3 [sup(13)]C NMR FOR CARRYING OUT FATTY ACID POSITIONAL ANALYSIS OF OLIVE OIL TRIACYLGLYCEROLS......Page 646
65.4 [sup(13)]C NMR OF TRIACYLGLYCEROLS FOR DETERMINING OLIVE OIL AUTHENTICITY......Page 651
REFERENCES......Page 654
66.2 FEATURES OF VIRGIN OLIVE OIL AROMA......Page 656
66.3 HEADSPACE EXTRACTION TECHNIQUES......Page 657
66.4 DIRECT THERMAL DESORPTION (DTD)......Page 660
66.5 DISTILLATION AND FLUID-BASED EXTRACTION TECHNIQUES......Page 662
REFERENCES......Page 663
67.1 GENERAL CONCEPTS RELATING TO ANTIOXIDANT ACTIVITY OF PHENOLIC MOLECULES......Page 666
67.2 GENERAL CONCEPTS ON DIRECT AND INDIRECT METHODS TO DETERMINE THE ANTIOXIDANT ACTIVITY OF PHENOLS IN VIRGIN OLIVE OIL......Page 667
67.3. RECENT APPLICATIONS OF DIFFERENT METHODS FOR THE EVALUATION OF ANTIOXIDANT ACTIVITY IN VIRGIN OLIVE OIL......Page 670
SUMMARY POINTS......Page 674
REFERENCES......Page 675
68.2 TOXICOLOGY......Page 678
68.4 METHODS OF ANALYSIS: CURRENT STATUS......Page 680
REFERENCES......Page 683
69.2 AFLATOXINS......Page 686
69.4 AFLATOXIN AND OCHRATOXIN A CONTAMINATION IN OLIVE OIL......Page 687
69.6 ANALYTICAL METHODS FOR AFLATOXINS AND OCHRATOXIN A DETERMINATION IN OLIVE OIL......Page 689
REFERENCES......Page 692
70.1 INTRODUCTION......Page 694
70.3 ANALYTICAL METHODS FOR PESTICIDE RESIDUE DETERMINATION......Page 695
SUMMARY POINTS......Page 697
REFERENCES......Page 706
71.1 INTRODUCTION......Page 708
71.2 DESCRIPTION OF THE MULTIRESIDUE METHOD......Page 709
SUMMARY POINTS......Page 711
REFERENCES......Page 723
Section 2 Nutritional, Pharmacological and Metabolic Properties of Olives and Olive Oil......Page 726
2.1 General Nutrition......Page 728
72.1 INTRODUCTION......Page 730
72.2 OLIVE OIL PRODUCTION IN MEDITERRANEAN COUNTRIES......Page 731
72.3 OLIVE OIL SUPPLY IN MEDITERRANEAN COUNTRIES......Page 732
72.4 OLIVE OIL CONSUMPTION IN MEDITERRANEAN COUNTRIES......Page 735
72.6 CONCLUSION......Page 737
REFERENCES......Page 738
73.2 BIOAVAILABILITY OF OLIVE OIL PHENOLIC COMPOUNDS......Page 740
REFERENCES......Page 743
74.2 PROTEIN AND AMINO ACID CONTENTS IN TABLE OLIVES......Page 746
74.3 VITAMIN E PROFILE IN TABLE OLIVES......Page 748
74.4 PROVITAMIN A CAROTENOIDS PROFILE IN TABLE OLIVES......Page 750
74.5 VITAMIN B[sub(6)] IN TABLE OLIVES......Page 752
SUMMARY POINTS......Page 754
REFERENCES......Page 755
75.2 PROXIMATE COMPOSITION OF TABLE OLIVES......Page 756
75.3 LIPID CONTENT IN TABLE OLIVES......Page 757
75.4 STEROLS AND FATTY AND TRITERPENIC ALCOHOLS IN TABLE OLIVES......Page 761
REFERENCES......Page 764
76.2 METHODOLOGICAL CONSIDERATIONS......Page 766
76.3 GENETIC MATRIX AND NUTRITIONAL PROPERTIES OF MONOVARIETAL OLIVE OILS......Page 767
76.4 INFLUENCE OF OLIVE RIPENING STAGE ON EXTRA VIRGIN OLIVE OIL NUTRITIONAL PROPERTIES......Page 769
76.5 INFLUENCE OF RAW MATERIAL STATUS ON EXTRA VIRGIN OLIVE OIL NUTRITIONAL PROPERTIES......Page 770
SUMMARY POINTS......Page 772
REFERENCES......Page 773
77.1 INTRODUCTION......Page 776
77.3 EFFECTS OF PROBIOTICS ON THE IMMUNE RESPONSE......Page 777
77.5 ANTI-CANCER ACTIVITIES OF PROBIOTICS IN COLORECTAL NEOPLASMS......Page 778
77.7 BENEFICIAL BACTERIA AND TABLE OLIVES: DEVELOPING A NEW PROBIOTIC FOOD......Page 780
77.8 CONCLUSIONS......Page 782
REFERENCES......Page 783
78.2 LIPIDS IN FOOD......Page 786
78.3 LIPIDS IN BAKERY PRODUCTS: PROPERTIES......Page 787
78.5 EFFECTS OF TEMPERATURE ON THE LIPID FRACTION OF OVEN-COOKED FOCACCIAS......Page 788
78.6 CONCLUSIONS......Page 792
REFERENCES......Page 793
79.1 INTRODUCTION......Page 796
79.2 COMPOSITIONAL CHANGES OF FISH PAN-FRIED IN VOO......Page 797
79.3 NUTRITIONAL EVALUATION OF PAN-FRIED FISH......Page 804
REFERENCES......Page 805
80.1 INTRODUCTION......Page 808
80.2 COMPOSITIONAL CHANGES IN VEGETABLES AFTER FRYING......Page 809
SUMMARY POINTS......Page 814
REFERENCES......Page 816
2.2 Cardiovascular......Page 818
81.1 INTRODUCTION......Page 820
81.3 AVAILABLE EPIDEMIOLOGICAL EVIDENCE......Page 821
81.4 THE CASE-CONTROL STUDY IN THE UNIVERSITY OF NAVARRA......Page 822
81.5 DISCUSSION......Page 823
SUMMARY POINTS......Page 825
REFERENCES......Page 826
82.2 BLOOD PRESSURE......Page 828
82.3 CARDIAC STRUCTURE......Page 830
82.5 KIDNEY STRUCTURE......Page 831
SUMMARY POINTS......Page 833
REFERENCES......Page 834
83.3 HEALTH BENEFITS OF OLIVE OIL CONSUMPTION......Page 836
83.5 DESIGN OF CARDIO2000 STUDY......Page 837
83.7 AN ANALYSIS OF THE ROLE OF OLIVE OIL ON ACUTE CORONARY SYNDROME......Page 838
REFERENCES......Page 840
84.2 OLIVE OIL AND HYPERTENSION......Page 842
84.4 OLIVE OIL AND HYPERTENSION IN THE SUN STUDY: RESULTS......Page 843
84.5 OLIVE OIL AND HYPERTENSION IN OTHER STUDIES......Page 844
SUMMARY POINTS......Page 845
REFERENCES......Page 846
85.1 INTRODUCTION......Page 848
85.4 BIOLOGICAL MECHANISMS FOR THE HYPOTENSIVE EFFECTS OF VIRGIN OLIVE OIL......Page 849
85.5 VIRGIN OLIVE OIL MINOR COMPONENTS AND HYPERTENSION......Page 850
SUMMARY POINTS......Page 851
REFERENCES......Page 852
86.3 IMPORTANCE OF THE ENDOTHELIUM ON VASCULAR FUNCTION......Page 854
86.4 VASODILATOR EFFECTS OF OLIVE OIL TRITERPENOIDS IN ISOLATED RAT AORTA......Page 856
86.5 VASODILATOR EFFECTS OF OLIVE OIL TRITERPENOIDS IN ISOLATED RAT MESENTERIC ARTERIES......Page 857
86.6 MOLECULAR BASIS OF THE VASOPROTECTIVE EFFECTS OF OLIVE OIL TRITERPENOIDS......Page 858
86.8 CONCLUSION......Page 859
REFERENCES......Page 860
87.3 ATHEROSCLEROSIS AS AN INFLAMMATORY DISEASE......Page 862
87.4 PATHOGENESIS OF ATHEROSCLEROSIS: ROLE OF ENDOTHELIAL ACTIVATION......Page 863
87.6 EFFECTS OF OLIVE OIL ON ENDOTHELIAL ACTIVATION......Page 864
87.8 OLIVE OIL POLYPHENOLIC COMPOUNDS AND ENDOTHELIAL ACTIVATION......Page 865
87.9 CONCLUSIONS......Page 867
REFERENCES......Page 868
88.1 INTRODUCTION......Page 870
88.2 CHEMICAL COMPOSITION OF OLIVE OIL......Page 871
88.4 EFFECTS OF THE COMPONENTS OF POMACE OLIVE OIL ON ENDOTHELIAL FUNCTION......Page 872
REFERENCES......Page 874
89.2 GASTROINTESTINAL AND CARDIOVASCULAR DISEASES......Page 876
89.3 PHARMACOLOGY OF CHOLINERGIC AND CALCIUM CHANNEL BLOCKERS......Page 878
89.4 NATURALLY OCCURRING CHOLINERGIC AND CA[sup(++)] ANTAGONIST COMBINATION IN OLIVES......Page 879
SUMMARY POINTS......Page 882
REFERENCES......Page 883
90.2 EFFECTS OF DIFFERENT WD ON PLASMA LIPIDS......Page 886
90.6 THE SEARCH FOR NEW PLASMA BIOMARKERS TO EXPLAIN THE ATHEROSCLEROTIC CHANGES......Page 888
90.7 APOLIPOPROTEIN DISTRIBUTION AND ARYLESTERASE ACTIVITY AMONG LIPOPROTEIN SUBCLASSES......Page 889
REFERENCES......Page 891
91.3 OLIVE OIL AND ANGIOTENSINASES......Page 894
REFERENCES......Page 901
92.3 CAN THE TYPE OF DIETARY FAT AFFECT THE RATE OF FAT OXIDATION?......Page 904
92.4 OLIVE OIL, SATIETY AND FOOD INTAKE......Page 906
92.5 CONCLUSIONS......Page 909
REFERENCES......Page 910
93.2 LIPIDEMIC PROFILE......Page 912
93.3 OLIVE OIL’S IDEAL NUTRITIONAL AND ANTIOXIDATIVE PROPERTIES......Page 913
93.4 THE DOCUMENTED BENEFICIAL ROLE OF OLIVE OIL IN CAD PREVENTION AND GOOD HEALTH......Page 914
REFERENCES......Page 917
94.2 DIETARY FAT AND CORONARY ARTERY DISEASE......Page 920
94.3 OLIVE OIL AND POSTPRANDIAL LIPEMIA......Page 921
94.5 OLIVE OIL AND TRIGLYCERIDE-RICH LIPOPROTEIN NUMBER......Page 922
94.6 OLIVE OIL AND TRIGLYCERIDE-RICH LIPOPROTEIN SIZE......Page 923
SUMMARY POINTS......Page 924
REFERENCES......Page 925
95.2 USE OF PROTEOMICS IN NUTRITION RESEARCH......Page 928
95.3 EFFECTS OF OLIVE OIL CONSUMPTION ON HEPATIC LIPID AND GLUCOSE METABOLISM......Page 929
95.4 EFFECTS OF OLIVE OIL CONSUMPTION ON ANTIOXIDANT DEFENSE MECHANISMS......Page 931
SUMMARY POINTS......Page 933
REFERENCES......Page 934
96.2 AVAILABLE EPIDEMIOLOGICAL EVIDENCE......Page 936
96.3 POSSIBLE MECHANISMS......Page 940
REFERENCES......Page 941
2.3 Oxidative Stress......Page 944
97.1 INTRODUCTION......Page 946
97.3 THE INFLUENCE OF THE NUMBER OF METHOXY GROUPS IN THE PHENOL RING......Page 947
97.4 THE INFLUENCE OF ACYL GROUPS IN THE PHENOL RING......Page 949
97.5 THE INFLUENCE OF THE LENGTH AND NATURE OF THE ALKYL CHAIN......Page 950
97.6 THE INFLUENCE OF FUNCTIONAL GROUP AT THE END OF THE ALKYL CHAIN......Page 951
97.8 THE INFLUENCE OF OTHER ESTER GROUPS AT THE ALKYL CHAIN: OLEUROPEIN, SECOIRIDOIDS AND OTHER OLIVE OIL COMPONENTS......Page 952
REFERENCES......Page 954
98.1 INTRODUCTION......Page 956
98.3 STUDY IN HUMAN VOLUNTEERS......Page 958
98.4 ANTIOXIDANT CAPACITY METHODOLOGY......Page 959
SUMMARY POINTS......Page 960
REFERENCES......Page 962
99.1 INTRODUCTION......Page 966
99.3 ESTIMATION OF TOTAL POLYPHENOL CONTENT......Page 967
99.4 IDENTIFICATION AND QUANTIFICATION OF PHENOLIC COMPOUNDS......Page 968
99.5 ANTIOXIDANT CAPACITY OF TABLE OLIVES......Page 971
99.6 NUTRITIONAL EVALUATION OF THE POLYPHENOL CONTENT......Page 972
99.7 CONCLUSIONS......Page 973
REFERENCES......Page 974
100.2 OLIVE OIL AND CELLULAR OXIDATIVE STRESS......Page 976
100.3 EFFECTS OF OLIVE OIL COMPONENTS ON AA AMOUNT IN MEMBRANES AND AA RELEASE......Page 978
100.4 EFFECTS OF OLIVE OIL COMPONENTS ON AA METABOLISM......Page 979
100.5 CONCLUSIONS......Page 980
REFERENCES......Page 981
101.1 INTRODUCTION......Page 984
101.2 ANTIOXIDANT ACTIVITY OF CORATINA SOR – CELL-FREE SYSTEMS......Page 985
101.3 ANTIOXIDANT ACTIVITY IN CELL SYSTEMS......Page 986
REFERENCES......Page 989
102.1 INTRODUCTION......Page 992
102.2 METHODOLOGICAL CONSIDERATIONS......Page 993
102.3 ANALYSIS OF STRUCTURAL BASES OF ANTIOXIDANT AND RADIOPROTECTIVE ACTIVITY OF OLIVE LEAF PHENOLICS......Page 995
SUMMARY POINTS......Page 998
REFERENCES......Page 999
2.4 Cancer and Immunology......Page 1000
103.1 INTRODUCTION......Page 1002
103.4 OLIVE OIL EFFECT GIVEN IN COMBINATION WITH DRUGS......Page 1003
103.5 OLIVE OIL AND COLON CANCER......Page 1004
103.8 OLIVE OIL PHENOLICS AS CHEMOPROTECTIVE AGENTS AGAINST COLORECTAL CANCER......Page 1005
SUMMARY POINTS......Page 1007
REFERENCES......Page 1008
104.2 DIETARY FAT, OLIVE OIL, AND CANCER......Page 1010
104.3 N-METHYL NITROSOUREA (NMU) MODEL OF RAT BREAST CANCER AND DIETARY MANIPULATION......Page 1011
SUMMARY POINTS......Page 1018
REFERENCES......Page 1019
105.1 INTRODUCTION......Page 1022
105.2 HUMAN INTERVENTION STUDIES WITH OLIVE OIL PHENOLS......Page 1023
SUMMARY POINTS......Page 1027
REFERENCES......Page 1028
106.1 INTRODUCTION......Page 1030
106.3 THE AMES TEST IS USED TO DETERMINE THE MUTAGENIC ACTIVITY......Page 1031
106.4 FACTORS THAT DETERMINE THE PRESENCE OF MUTAGENS IN MEAT......Page 1032
106.5 MUTAGENIC ACTIVITY IN FRIED MEAT......Page 1033
106.6 CONCLUSIONS......Page 1034
SUMMARY POINTS......Page 1035
REFERENCES......Page 1036
107.2 WNT SIGNALING PATHWAY AND COLON CANCER......Page 1038
107.4 DIETARY FATTY ACID COMPOSITION AND COLON CANCER......Page 1039
REFERENCES......Page 1043
108.2 CARCINOGENESIS, DNA DAMAGE AND METASTASIS......Page 1046
108.3 OLIVE OIL, PHENOLICS AND DNA DAMAGE......Page 1048
108.4 OLIVE OIL, PHENOLICS AND METASTASIS-RELATED EVENTS......Page 1050
REFERENCES......Page 1052
109.1 INTRODUCTION......Page 1054
109.2 MAIN ANTIMICROBIAL COMPOUNDS IN OLIVE OIL......Page 1055
109.3 MAIN ANTIMICROBIAL COMPOUNDS IN TABLE OLIVES......Page 1057
REFERENCES......Page 1059
110.1 INTRODUCTION......Page 1062
110.2 OLE E 1 AS A MARKER FOR SENSITIZATION TO OLEACEAE POLLENS......Page 1065
110.4 OLE E 3 AND OLE E 8: CA[sup(2+)-BINDING ALLERGENS......Page 1066
110.6 OLE E 9 AND POLLEN-LATEX-FRUIT SYNDROME......Page 1067
110.8 THE ROLE OF N -GLYCANS IN OLIVE POLLEN ALLERGY......Page 1068
110.10 NEW CONCEPTS FOR SPECIFIC IMMUNOTHERAPY USING OLE E 1 AS A MODEL......Page 1069
REFERENCES......Page 1071
111.3 LIPID ABSORPTION AND ACTIVATION OF THE CHOLINERGIC ANTIINFLAMMATORY PATHWAY......Page 1074
111.5 CONCLUSION......Page 1077
REFERENCES......Page 1078
112.2 IMMUNE SYSTEM: A BRIEF DESCRIPTION......Page 1080
112.3 FATTY ACIDS AND IMMUNE SYSTEM FUNCTIONS......Page 1081
112.5 OLIVE OIL AND ALTERATION OF IMMUNE FUNCTIONS......Page 1082
112.6 OLIVE OIL AND INFECTIOUS DISEASE RESISTANCE: AN INTRODUCTION......Page 1085
SUMMARY POINTS......Page 1086
REFERENCES......Page 1087
113.1 INTRODUCTION......Page 1090
113.2 INTESTINAL ANTI-INFLAMMATORY ACTIVITY OF DIETARY OLIVE OIL......Page 1091
REFERENCES......Page 1095
114.3 PATHOPHYSIOLOGY OF INFLAMMATION......Page 1098
114.4 OLIVE OIL AND INFLAMMATION......Page 1099
114.5 ARE THERE DIFFERENCES BETWEEN OLIVE OIL AND EXTRA VIRGIN OLIVE OIL USE IN THE INFLAMMATORY PROCESS?......Page 1100
114.7 STUDIES USING OLIVE OIL AS PLACEBO IN RHEUMATOID ARTHRITIS PATIENTS......Page 1101
114.8 FUTURE PERSPECTIVES......Page 1103
REFERENCES......Page 1104
2.5 Other Effects, Uses and Diseases......Page 1106
115.1 INTRODUCTION......Page 1108
115.2 DIETARY FAT AND NF-κB IN MONONUCLEAR LEUKOCYTES......Page 1109
ACKNOWLEDGMENTS......Page 1110
REFERENCES......Page 1111
116.2 CHROMOSOME STUDIES IN GENETIC TOXICOLOGY......Page 1112
116.4 OLIVE OIL ANTIMUTAGENICITY......Page 1113
116.5 POSSIBLE MECHANISMS OF OLIVE OIL ANTIMUTAGENICITY......Page 1115
ACKNOWLEDGMENTS......Page 1117
REFERENCES......Page 1118
117.1 INTRODUCTION......Page 1120
117.2 FEATURES OF MINOR POLAR COMPOUND (MPC) OLIVE OIL EXTRACT......Page 1121
117.3 MPC-OLIVE OIL EXTRACT INHIBITS NF-κB ACTIVATION IN HUMAN MONOCYTES/MACROPHAGES......Page 1122
REFERENCES......Page 1126
118.1 INTRODUCTION......Page 1128
118.2 DIETARY LIPID SOURCE AND UCPs......Page 1129
REFERENCES......Page 1132
119.1 INTRODUCTION......Page 1136
119.2 GLUTATHIONE AND GSH-RELATED ENZYMES......Page 1137
119.3 EVOO BIOPHENOLS AND GSH CYCLE......Page 1138
119.4 EVOO BIOPHENOLS AND GENE EXPRESSION......Page 1140
REFERENCES......Page 1142
120.2 MEDITERRANEAN DIET AND OLIVE OIL......Page 1144
120.3 EVALUATION OF THE PROTECTION OFFERED BY EXOGENOUSLY ADDED COMPOUNDS AGAINST HYDROGEN PEROXIDE-INDUCED DNA DAMAGE......Page 1145
120.4 THE ROLE OF IRON IN HYDROGENPEROXIDE-INDUCED DNA DAMAGE......Page 1146
120.6 FLAVONOIDS PROTECT CELLS BY CHELATING INTRACELLULAR IRON......Page 1147
120.7 CONCLUDING REMARKS......Page 1148
REFERENCES......Page 1149
121.3 CYTOSOLIC CALCIUM......Page 1152
121.5 MOLECULAR EFFECTS OF PHENOLS......Page 1153
121.6 PHENOL–NITRIC OXIDE INTERACTION......Page 1154
SUMMARY POINTS......Page 1155
REFERENCES......Page 1156
122.2 CHEMISTRY......Page 1158
122.3 DIETARY PROTECTION......Page 1159
122.4 PHOTOPROTECTION......Page 1161
122.5 TOPICAL APPLICATIONS FOR DERMATOLOGIC CONDITIONS......Page 1162
122.6 COSMECEUTICALS......Page 1163
REFERENCES......Page 1164
123.3 DERMIS AND EPIDERMIS......Page 1166
123.4 PATHOPHYSIOLOGY OF WRINKLES, PRURITUS AND XEROSIS......Page 1167
123.5 BENEFICIAL PROPERTIES OF OLIVE OIL......Page 1168
123.6 THE EFFECTS OF OLIVE OIL ON THE SKIN......Page 1169
REFERENCES......Page 1172
124.1 INTRODUCTION......Page 1174
124.2 OLIVE OIL (OLEUM OLIVAE )......Page 1175
124.3 OLIVE OIL DERIVATIVES......Page 1177
124.4 SKIN PATHOLOGIES AND THE TREATMENT WITH OLIVE OIL......Page 1178
124.5 TOPICAL PHARMACEUTICAL FORMULATIONS MADE WITH OLIVE OIL......Page 1179
124.6 OZONATED OLIVE OIL......Page 1180
SUMMARY POINTS......Page 1181
REFERENCES......Page 1182
125.3 METHODOLOGICAL CONSIDERATION......Page 1184
125.4 GLOBAL CHANGES IN GENE EXPRESSION IN LIVERS FED THE DIFFERENT DIETS USING MICROARRAYS......Page 1185
125.5 PLASMA PRESENCE OF UNSAPONIFIABLE-ACTIVATED GENE PRODUCTS......Page 1186
125.6 EFFECTS OF MOUSE GENETIC BACKGROUND ON THE RESPONSE TO UNSAPONIFIABLE FRACTION-ENRICHED OLIVE OIL......Page 1187
REFERENCES......Page 1189
126.2 COMPOSITION OF MONOUNSATURATED FATTY ACID (MUFA), OLIVE OIL......Page 1192
126.3 PATHOPHYSIOLOGY OF NAFLD, DIETARY FAT AND HEPATIC LIPIDS......Page 1193
126.4 MONOUNSATURATED FATTY ACIDS (MUFA) AND NAFLD......Page 1194
126.5 THE SPECIAL MECHANISM OF OLIVE OIL......Page 1195
REFERENCES......Page 1196
127.1 INTRODUCTION......Page 1198
127.2 UPTAKE AND METABOLISM OF HTy IN CELL CULTURE......Page 1199
127.3 ANTIOXIDANT EFFECT OF HTy IN CELL CULTURE......Page 1203
REFERENCES......Page 1205
128.1 INTRODUCTION......Page 1208
128.2 OLIVE OIL AND APOPTOSIS IN AGING......Page 1209
128.3 CONCLUSIONS AND PERSPECTIVES......Page 1212
SUMMARY POINTS......Page 1213
REFERENCES......Page 1214
129.2 OLIVE OIL AND DIGESTIVE SECRETION IN DOGS......Page 1216
129.3 OLIVE OIL AND DIGESTIVE SECRETION IN HUMANS......Page 1219
REFERENCES......Page 1223
130.1 INTRODUCTION......Page 1226
130.2 DIETARY LIPIDS AND PANCREATIC SECRETION......Page 1227
130.3 PANCREATIC SECRETION IN ANESTHETIZED RATS......Page 1228
130.4 EXPERIMENTS IN ISOLATED PANCREATIC ACINI......Page 1230
130.5 AR42J STUDIES......Page 1231
SUMMARY POINTS......Page 1233
REFERENCES......Page 1235
131.1 INTRODUCTION......Page 1236
131.3 NUTRITIONAL VALUE OF OLIVES AND OLIVE OIL......Page 1237
131.4 OLIVES, OLIVE OIL AND BONE HEALTH......Page 1238
131.5 BIOLOGICAL EFFECTS OF OLIVES AND OLIVE OIL COMPONENTS......Page 1239
131.6 APPLICATION OF BIOLOGICAL EFFECTS OF OLIVES AND OLIVE OIL COMPONENTS TO BONE PHYSIOLOGY......Page 1241
131.7 CONCLUSION......Page 1242
REFERENCES......Page 1243
132.1 INTRODUCTION......Page 1246
132.2 HOMA, CLAMP AND INSULINOGENIC INDEX......Page 1247
132.4 REVERSIBILITY OF FRUCTOSEINDUCED INSULIN RESISTANCE......Page 1248
132.5 EFFECT OF OLIVE OIL AND GUAR ON GLUCOSE HOMEOSTASIS......Page 1249
132.7 EFFECT OF OLIVE OIL AND GUAR ON INSULIN RESISTANCE RATS......Page 1250
REFERENCES......Page 1251
133.2 GLUCAGON-LIKE PEPTIDE-1......Page 1254
133.3 MONOUNSATURATED FATTY ACIDS AND GLP-1......Page 1256
133.4 OLIVE OIL DIET IN NORMAL RATS......Page 1257
133.5 OLIVE OIL DIET IN TYPE 2 DIABETIC RATS......Page 1258
133.6 COMPARISON BETWEEN NORMAL AND DIABETIC RATS......Page 1259
REFERENCES......Page 1260
Section 3 Specific Components of Olive Oil and Their Effects on Tissue and Body Systems......Page 1262
3.1 Tyrosol and Hydroxytyrosol......Page 1264
134.2 PROPERTIES......Page 1266
134.5 OXIDATION REACTIVITY......Page 1268
134.6 MECHANISMS OF THE OXIDATION REACTIONS......Page 1269
134.7 REACTIVITY TOWARDS REACTIVE NITROGEN SPECIES......Page 1270
134.8 OVERVIEW OF CONCEPTS AND IMPLICATIONS FOR OXIDATIVE STRESS......Page 1271
REFERENCES......Page 1272
135.1 INTRODUCTION......Page 1274
135.2 LIPOPHILIC HYDROXYTYROSOL ANALOGUES......Page 1275
SUMMARY POINTS......Page 1282
REFERENCES......Page 1283
136.2 BIOLOGICAL EFFECTS OF HYDROXYTYROSOL......Page 1286
136.3 CYCLOSPORINE AND ROS......Page 1288
136.4 EFFECTS OF HYDROXYTYROSOL ON CYCLOSPORINE CYTOTOXICITY IN RAT RENAL TUBULAR CELLS......Page 1289
136.5 CYCLOSPORINE NEPHROTOXICITY IN RATS: EFFECT OF HYDROXYTYROSOL IN VIVO......Page 1290
136.6 CONCLUSIONS......Page 1291
SUMMARY POINTS......Page 1292
REFERENCES......Page 1293
137.2 THE PROTOCOL OF THE INTERVENTION STUDY......Page 1294
137.3 THE ‘ANTI-PLATELET’ CAPACITY OF THE HT-RICH EXTRACT......Page 1295
REFERENCES......Page 1297
138.2 HYDROXYTYROSOL AND ENDOTHELIAL CELLS......Page 1298
138.4 DO HYDROXYTYROSOL’S ACTIONS DEPEND ON THE PRESENCE OF PROINFLAMMATORY STIMULI?......Page 1299
138.6 OTHER WAYS OF INFLUENCING ENOS......Page 1300
REFERENCES......Page 1301
139.1 INTRODUCTION......Page 1304
139.2 TYROSOL......Page 1305
139.3 TYROSOL PREVENTS MACROPHAGE ACTIVATION BY IFN-γ AND GLIADIN......Page 1306
REFERENCES......Page 1308
140.2 HYDROXYTYROSOL DERIVATIVES APPEAR IN URINE AFTER TREATMENT......Page 1310
140.5 EFFECT OF HYDROXYTYROSOL ON HIGH-DENSITY LIPOPROTEINS......Page 1311
140.8 ANALYSIS OF THE EXPRESSION OF SURFACE MOLECULE, MAC-1, IN BLOOD MONOCYTES......Page 1312
140.12 ANALYSIS OF THE ROLE OF PHENOLIC COMPOUNDS IN ATHEROSCLEROSIS......Page 1313
REFERENCES......Page 1314
141.1 INTRODUCTION......Page 1316
141.2 MACROPHAGE ACTIVATION......Page 1317
141.3 HT INHIBITS MACROPHAGE ACTIVATION......Page 1319
REFERENCES......Page 1321
142.1 INTRODUCTION......Page 1324
142.3 MYCOPLASMAS, ANTIMYCOPLASMAL ACTIVITY......Page 1325
142.5 THE RATIONALE FOR THE USAGE OF HYDROXYTYROSOL IN ANTIMYCOPLASMAL THERAPY......Page 1326
142.6 CONCLUSION......Page 1327
REFERENCES......Page 1328
143.1 INTRODUCTION......Page 1330
143.2 ROS PRODUCTION BY NEUTROPHILS AND THEIR ROLE IN PATHOLOGY......Page 1331
143.3 ANTIOXIDANT EFFECTS OF OLIVE OIL PHENOLIC COMPOUNDS......Page 1332
REFERENCES......Page 1333
144.2 THE ROLE OF HYDROXYTYROSOL IN HEALTH AND DISEASE......Page 1336
144.3 ANTIPROLIFERATIVE ACTIVITY OF HYDROXYTYROSOL......Page 1337
144.5 HYDROXYTYROSOL IN THE DIET......Page 1339
REFERENCES......Page 1340
145.2 PENTACYCLIC TRITERPENES FROM OLEA EUROPAEA......Page 1342
145.3 APOPTOSIS IN COLORECTAL CANCER......Page 1343
145.4 PENTACYCLIC TRITERPENES INDUCE APOPTOSIS......Page 1345
REFERENCES......Page 1350
3.2 Oleuropein......Page 1352
146.2 THE EFFECT OF OLEUROPEIN ON ISCHEMIC MYOCARDIUM......Page 1354
146.3 THE EFFECT OF OLEUROPEIN ON OXIDATIVE REPERFUSION INJURY......Page 1356
146.4 THE EFFECT OF OLEUROPEIN ON DXR’S INDUCED ACUTE CARDIOTOXICITY......Page 1357
REFERENCES......Page 1359
147.2 PAST AND PRESENT OF IMMUNOINTERVENTION IN SEPSIS......Page 1362
147.4 OLEUROPEIN AS A PROMISING IMMUNOMODULATOR IN EXPERIMENTAL SEPSIS......Page 1363
147.5 PROPOSED MECHANISMS OF ACTION OF OLEUROPEIN......Page 1364
REFERENCES......Page 1365
148.1 INTRODUCTION......Page 1368
148.2 EXPERIMENTAL CONSIDERATIONS......Page 1369
148.3 FINAL CONSIDERATIONS ON THE ROLE OF OLEUROPEIN IN METABOLISM......Page 1371
REFERENCES......Page 1373
149.1 INTRODUCTION......Page 1376
149.2 THE ESSENTIAL LINK BETWEEN CELLULAR HOMEOSTASIS, AGING AND THE FUNCTION OF THE PROTEASOME......Page 1377
REFERENCES......Page 1383
150.2 EFFECTS OF OLE ON MEMBRANES USING PHOSPHOLIPID MODEL MEMBRANES......Page 1386
150.3 ANTIVIRAL ACTIVITY OF OLEUROPEIN AGAINST FISH RHABDOVIRUS......Page 1390
REFERENCES......Page 1394
151.2 ANTIMICROBIAL PROPERTIES......Page 1396
151.3 MYCOPLASMAS AND ANTIMYCOPLASMAL ACTIVITY......Page 1397
151.5 THE FATE OF OLEUROPEIN AFTER INGESTION......Page 1399
SUMMARY POINTS......Page 1400
ACKNOWLEDGMENTS......Page 1401
REFERENCES......Page 1433
3.3 Oleic Acid......Page 1404
152.2 IN VITRO STUDIES......Page 1406
152.3 IN VIVO STUDIES......Page 1408
152.4 CONCLUSIONS......Page 1412
REFERENCES......Page 1413
153.2 CARIES DEVELOPMENT......Page 1416
153.3 GLUCOSYLTRANSFERASE AS A TARGET OF ANTICARIES APPROACH......Page 1417
153.5 OLEIC ACID AND ITS DERIVATIVES AS A GLUCOSYLTRANSFERASE INHIBITOR......Page 1418
153.6 ANTIBACTERIAL ACTIVITIES OF LONG-CHAIN UNSATURATED FATTY ACIDS......Page 1419
153.7 CONCLUSION......Page 1421
SUMMARY POINTS......Page 1422
REFERENCES......Page 1423
154.2 OLEIC ACID ON POSTPRANDIAL THROMBOGENESIS......Page 1426
154.3 OLEIC ACID ON POSTPRANDIAL FIBRINOLYSIS......Page 1428
154.5 POSSIBLE MECHANISMS BY WHICH OLEIC ACID IS ACTING ON POSTPRANDIAL GLUCOSE HOMEOSTASIS......Page 1429
155.1 INTRODUCTION......Page 1436
155.3 ROLE OF PPARγ AND FATTY ACIDS IN CARBOHYDRATE METABOLISM......Page 1437
155.4 CHARACTERISTICS OF A POPULATION WITH A HIGH INTAKE OF OLEIC ACID AND THE PPARG2 GENE......Page 1438
REFERENCES......Page 1442
156.1 INTRODUCTION......Page 1446
156.3 TRANSCYTOSIS OF ALBUMIN INDUCES OLEIC ACID SYNTHESIS AND RELEASE BY ASTROCYTES......Page 1447
156.4 OLEIC ACID IS A NEUROTROPHIC FACTOR FOR NEURONAL DIFFERENTIATION......Page 1448
156.7 CLINICAL RELEVANCE OF OLEIC ACID IN THE BRAIN......Page 1449
ACKNOWLEDGMENTS......Page 1451
REFERENCES......Page 1452
3.4 Other Components Found in Olive Plants and Products......Page 1454
157.1 INTRODUCTION......Page 1456
157.2 MASLINIC ACID ADDED TO THE DIET STIMULATES THE GROWTH OF RAINBOW TROUT......Page 1458
157.4 MASLINIC ACID STIMULATES LIVER AND WHITE-MUSCLE PROTEIN-TURNOVER RATES......Page 1459
157.5 MASLINIC ACID STIMULATES GLYCOGEN ACCUMULATION IN HEPATOCYTES OF RAINBOW TROUT......Page 1460
SUMMARY POINTS......Page 1461
REFERENCES......Page 1462
158.1 INTRODUCTION......Page 1464
158.2 EFFECTS OF OLEANOLIC ACID AND MASLINIC ACID ON HYPERGLYCEMIA......Page 1465
158.3 EFFECTS OF OLEANOLIC ACID AND MASLINIC ACID ON THE PLASMA LIPID PROFILE......Page 1466
SUMMARY POINTS......Page 1468
REFERENCES......Page 1469
159.1 INTRODUCTION......Page 1472
159.2 CARDIOVASCULAR PROPERTIES......Page 1474
159.3 ANTI-INFLAMMATORY EFFECTS......Page 1475
159.4 ANTICANCER EFFECTS......Page 1476
159.7 OTHER FUNCTIONAL PROPERTIES......Page 1477
REFERENCES......Page 1478
160.2 THE FLAVANOIDS......Page 1480
160.4 THE DIMERIC LIGNANS......Page 1487
160.5 ELENOLIC ACID AND THE SECOIRIDOIDS......Page 1493
160.6 ELENOLIC ACID......Page 1494
160.7 OLEUROPEIN......Page 1496
160.8 OLEOCANTHAL: A ‘NATURAL’ NONSTEROIDAL ANTIINFLAMMATORY DRUG (NSAID) IN EXTRA VIRGIN OLIVE OILS......Page 1498
160.9 DECARBOMETHOXY OLEUROPEIN AGLYCONE......Page 1502
REFERENCES......Page 1504
A......Page 1506
C......Page 1507
F......Page 1509
G......Page 1510
H......Page 1511
L......Page 1512
M......Page 1513
N......Page 1514
P......Page 1515
R......Page 1517
S......Page 1518
T......Page 1519
Z......Page 1520
Recommend Papers
![Olives and Olive Oil in Health and Disease Prevention [2 ed.]
9780128195284](https://ebin.pub/img/200x200/olives-and-olive-oil-in-health-and-disease-prevention-2nbsped-9780128195284.jpg)
![Olives and Olive Oil in Health and Disease Prevention [2 ed.]
0128195282, 9780128195284](https://ebin.pub/img/200x200/olives-and-olive-oil-in-health-and-disease-prevention-2nbsped-0128195282-9780128195284.jpg)
![Olives and Olive Oil in Health and Disease Prevention [2 ed.]
0128195282, 9780128195284](https://ebin.pub/img/200x200/olives-and-olive-oil-in-health-and-disease-prevention-2nbsped-0128195282-9780128195284-a-5865992.jpg)
![Olive Oil and Health [First ed.]
1845930681, 9781845930684](https://ebin.pub/img/200x200/olive-oil-and-health-firstnbsped-1845930681-9781845930684.jpg)
![Vegetarian and Plant-Based Diets in Health and Disease Prevention [Illustrated]
012803968X, 9780128039687](https://ebin.pub/img/200x200/vegetarian-and-plant-based-diets-in-health-and-disease-prevention-illustrated-012803968x-9780128039687.jpg)
![Olive Oil: Minor Constituents and Health [1 ed.]
1420059939, 9781420059939](https://ebin.pub/img/200x200/olive-oil-minor-constituents-and-health-1nbsped-1420059939-9781420059939.jpg)
![Olive Oil Chemistry and Technology [2 ed.]
189399788X, 9781893997882](https://ebin.pub/img/200x200/olive-oil-chemistry-and-technology-2nbsped-189399788x-9781893997882.jpg)
![Nutritional Health: Strategies for Disease Prevention (Nutrition and Health) [4th ed. 2023]
3031246624, 9783031246623](https://ebin.pub/img/200x200/nutritional-health-strategies-for-disease-prevention-nutrition-and-health-4th-ed-2023-3031246624-9783031246623.jpg)
![Olives and Olive Oil in Health and Disease Prevention [1 ed.]
0123744202, 9780123744203](https://ebin.pub/img/200x200/olives-and-olive-oil-in-health-and-disease-prevention-1nbsped-0123744202-9780123744203.jpg)
- Author / Uploaded
- Victor R. Preedy
- Ronald Ross Watson
- Similar Topics
- Technique
- Oil and Gas Technologies
File loading please wait...
Citation preview
Olives and Olive Oil in Health and Disease Prevention
This page intentionally left blank
Olives and Olive Oil in Health and Disease Prevention Edited by Victor R. Preedy Dept Nutrition and Dietetics, Nutritional Sciences Division, School of Biomedical & Health Sciences, King’s College London, Franklin-Wilkins Building, London, UK
Ronald Ross Watson University of Arizona Division of Health Promotion Sciences Mel and Enid Zuckerman College of Public Health, and School of Medicine, Arizona Health Sciences Center, Tucson, AZ, USA
AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD • PARIS SAN DIEGO • SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier 32 Jamestown Road, London NW1 7BY, UK 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA First edition 2010 Copyright © 2010 Elsevier Inc. All rights reserved with the exception of: Chapter 68, Crown Copyright © 2009. Published by Elsevier Inc. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (⫹44) (0) 1865 843830; fax (⫹44) (0) 1865 853333; email: [email protected]. Alternatively, visit the Science and Technology Books website at www.elsevierdirect.com/rights for further information Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN : 978-0-12-374420-3 For information on all Academic Press publications visit our website at www.elsevierdirect.com Typeset by Macmillan Publishing Solutions www.macmillansolutions.com Printed and bound in United States of America 10 11 12 13
10 9 8 7 6 5 4 3 2 1
Contents
Preface xv List of Contributors xvii Abbreviations xxix
7. The Effect of the Ripening Process of the Olive Fruit on the Chlorophyll and Carotenoid Fractions of Drupes and Virgin Oils 59 Maria-Jose Motilva and Maria-Paz Romero
Section 1 General Aspects of Olives and Olive Oil 1 1.1 The Plant, Production, Olives and Olive Oil and Their Detailed Characterization 3 The Plant and Production 1. Table Olives: Varieties and Variations 5
Luis Rejano, Alfredo Montaño, Francisco Javier Casado, Antonio Higinio Sánchez and Antonio de Castro
2. Olive Genomics 17
Corrado Fogher, Matteo Busconi, Luca Sebastiani and Tania Bracci
3. Current Initiatives in Proteomics of the Olive Tree 25 Wei Wang, Fuju Tai and Xiuli Hu
4. Chemometric Classification of Cultivars of Olives: Perspectives on Portuguese Olives 33 Joaquim C.G. Esteves da Silva
Olives and Olive Oil 8. Influence of the Crushing System: Phenol Content in Virgin Olive Oil Produced from Whole and De-stoned Pastes 69 Paolo Amirante, Maria Lisa Clodoveo, Antonia Tamborrino, Alessandro Leone and Alistair G. Paice
9. The Malaxation Process: Influence on Olive Oil Quality and the Effect of the Control of Oxygen Concentration in Virgin Olive Oil 77 Antonia Tamborrino, Maria Lisa Clodoveo, Alessandro Leone, Paolo Amirante and Alistair G. Paice
10. Influence of Different Centrifugal Extraction Systems on Antioxidant Content and Stability of Virgin Olive Oil 85 Paolo Amirante, Maria Lisa Clodoveo, Alessandro Leone, Antonia Tamborrino and Vinood B. Patel
5. Effect of Climatic Conditions on Quality of Virgin Olive Oil 43
11. A Marker of Quality of Olive Oils: The Expression of Oleuropein 95
6. Influence of Irrigation Management and Ripening on Virgin Olive Oil Quality and Composition 51
12. Olive Oil Authenticity Evaluation by Chemical and Biological Methodologies 101
María-Paz Romero and María-José Motilva
Giuseppe Fregapane, Aurora Gómez-Rico and Maria Desamparados Salvador
Giovanni Sindona
Miguel A. Faria, Sara C. Cunha, Alistair G. Paice and Maria Beatriz P.P. Oliveira v
vi
Contents
The Detailed Characterization of Olives and Olive Products 13. Ripening of Table Olives: Use of Magnetic Resonance Imaging (MRI) 109
Maria Antonietta Brescia and Antonio Sacco
14. NMR and Olive Oils: A Geographical Characterization 117 Luisa Mannina and Anna Laura Segre†
15. NMR and Olive Oils: A Characterization According to the Cultivar 125 Luisa Mannina and Anna Laura Segre†
16. Geographical Characterization of Olive Oil by Means of Multivariate Classification: Application of CAIMAN 129 Davide Ballabio and Roberto Todeschini
17. Non-conventional Parameters for Quality Evaluation of Refined Olive Oil and Olive Oil Commercial Classes 139 Tommaso Gomes, Vito Michele Paradiso and Debora Delcuratolo
18. Classification of Sicilian Olive Oils According to Heavy Metal and Selenium Levels Using Canonical Discriminant Analysis (CDA) 155 Lara La Pera, Giacomo Dugo, Vincenzo Lo Turco, Rossana Rando and Giuseppa Di Bella
1.2 Components of Olives and Olive Plant Product and Uses 165 Lipids, Phenolics and Other Organics and Volatiles 19. Polyphenols in Olive Oil: The Importance of Phenolic Compounds in the Chemical Composition of Olive Oil 167 Antonio Segura-Carretero, Javier Menéndez-Menéndez and Alberto Fernández-Gutiérrez
†
deceased.
20. Phenolic Profiles of Portuguese Olives: Cultivar and Geographics 177
Rosa M. Seabra, Paula B. Andrade, Patrícia Valentão, Miguel Faria, Alistair G. Paice and Maria Beatriz P.P. Oliveira
21. Low-level Free Phenols in Sicilian Olive Oils 187
Marcello Saitta, Giuseppa Di Bella, Vincenzo Lo Turco, Giovanna Loredana La Torre and Giacomo Dugo
22. Volatile Compounds in Australian Olive Oils: How Different Are They From Other Oils 201 Curtis Kalua, Paul Prenzler, Danielle Ryan and Kevin Robards
23. Sterols and Non-steroidal Triterpenoids of the Developing Olive Fruit 211
Naïm Stiti, Saïda Triki and Marie-Andrée Hartmann
24. A Comparison of the Volatile Compounds, in Spanish-style, Greek-style and Castelvetrano-style Green Olives of the Nocellara del Belice Cultivar: Alcohols, Aldehydes, Ketones, Esters and Acids 219 Nadia Sabatini
25. Polyphenol Oxidase and Oleuropein in Olives and their Changes During Olive Ripening 233 Francisca Ortega-García, Santos Blanco, M. Ángeles Peinado and Juan Peragón
26. Major and Minor Lipid Constituents of Cornicabra Virgin Olive Oil and the Influence of Crop Season Changes 239 M. Desamparados Salvador and Giuseppe Fregapane
27. Phytosterol Classes in Olive Oils and their Analysis by Common Chromatographic Methods 249
Sodeif Azadmard-Damirchi and Paresh C. Dutta
28. Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches 259 M.D. Luque de Castro and F. Priego Capote
vii
Contents
29. The Occurrence of the Biogenic Amine Melatonin in Olive Oil: Implications in Health and Disease Prevention 275 Rafael Fernández-Montesinos, Cristina de la Puerta, Pedro P. García-Luna, Russel J. Reiter and David Pozo
30. Olive Biophenols as Food Supplements and Additives 283 Antonella De Leonardis and Vincenzo Macciola
Metals, Electrolytes and Other Components
Alfredo Montaño, Antonio Higinio Sánchez, Antonio López-López, Antonio de Castro and Luis Rejano
Yasemin Sahan
Ganapathy Sivakumar and Nicola A. Uccella
Non Fruit Aspects Including Mill Wastewater 38. Production of Triterpene Acids by Cell-suspension Cultures of Olea europaea 341 Yutaka Orihara and Yutaka Ebizuka
31. Chemical Composition of Fermented Green Olives: Acidity, Salt, Moisture, Fat, Protein, Ash, Fiber, Sugar, and Polyphenol 291
32. Some Metals in Table Olives
37. Olive Biophenols and Conventional Biotechnology from Mediterranean Aliment Culture 333
299
33. Olive Cultivar, Period of Harvest, and Environmental Pollution on the Contents of Cu, Cd, Pb, and Zn: Italian Perspectives 307 Alberto Angioni
34. Trace Components in Italian Virgin Olive Oils 313 Giovanni Sindona and Antonio Tagarelli
35. Inorganic Anions in Olive Oils: Application of Suppressed Ion Exchange Chromatography (IEC) for the Analysis of Olive Oils Produced from De-stoned Olives and Traditional Extraction Methods 317 Lara La Pera, Teresa Maria Pellicanò, Pellicano Vincenzo Lo Turco, Giuseppa Di Bella and Giacomo Dugo
36. Purification and Characterization of Olive (Olea europaea L.) Peroxidases 325
Jorge A. Saraiva, C.S. Cláudia, S. Nunes and Manuel A. Coimbra
39. Bioactive Ingredients in Olive Leaves 349
Maria Z. Tsimidou and Vassiliki T. Papoti
40. Phenolic Compounds in Olive Oil Mill Wastewater 357 José S. Torrecilla
1.3 Stability, Microbes, Contaminants and Adverse Components and Processes 367 Bacterial and Fungal and Other Microbial Aspects 41. Lactic Acid Bacteria in Table Olive Fermentation 369
Cinzia L. Randazzo, Rajkumar Rajendram and Cinzia Caggia
42. Understanding and Optimizing the Microbial Degradation of Olive Oil: A Case Study with the Thermophilic Bacterium Geobacillus thermoleovorans IHI-91 377 Peter Becker
43. Degradation of Phenolic Compounds Found in Olive Products by Lactobacillus plantarum Strains 387 José María Landete, Héctor Rodríguez, José Antonio Curiel, Blanca de las Rivas, Félix López de Felipe and Rosario Muñoz
44. Microbial Colonization of Naturally Fermented Olives 397 C.C. Tassou, E.Z. Panagou and G.-J.E. Nychas
viii
Contents
45. Occurrence of Aflatoxin B1 in the Greek Virgin Olive Oil: Estimation of the Daily Exposure 407
54. Polycyclic Aromatic Hydrocarbons (PAH) in Olive Oils and Other Vegetable Oils; Potential for Carcinogenesis 489
Pesticides and Adulterants
55. Mineral Paraffins in Olives and Olive Oils 499
Panagiota Markaki
46. Fenthion, Dimethoate and Other Pesticides in Olive Oils of Organic and Conventional Cultivation 415 Aristidis M. Tsatsakis and Ioannis N. Tsakiris
47. Residues of Pesticides and Polycyclic Aromatic Hydrocarbons in Olive and Olive-Pomace Oils by Gas Chromatography/Tandem Mass Spectrometry 425
Evaristo Ballesteros and Natividad Ramos-Martos
48. Acephate and Buprofezin Residues in Olives and Olive Oil 437 Pierluigi Caboni and Paolo Cabras
49. Olive Oil Adulteration with Hazelnut Oil and Analytical Approaches for Its Detection 441 Dimitrios Zabaras
50. Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy 451
Betül Öztürk, Aysun Ankan and Durmus¸ Özdemir
Toxicology and Contaminants 51
Benzene, Toluene, Ethylbenzene, (o-, m- and p-) Xylenes and Styrene in Olive Oil 463 Silvia López-Feria, Soledad Cárdenas and Miguel Valcárcel
52. The Toxicity Profile of Hydrolyzed Aqueous Olive Pulp Extract 471 Robert M. Diener and Mildred S. Christian
53. Plasticizer in Olive Oils
481
Giuseppa Di Bella, Lara La Pera, Vincenzo Lo Turco, Donatella Pollicino and Giacomo Dugo
Isabel Mafra, Joana S. Amaral and M. Beatriz P.P. Oliveira
Sabrina Moret, Tiziana Populin and Lanfranco S. Conte
1.4 Analytical Methods
507
Natural Components 56. Analytical Determination of Polyphenols in Olive Oil 509
Antonio Segura-Carretero, Alegría CarrascoPancorbo, Alessandra Bendini, Lorenzo Cerretani and Alberto Fernández-Gutiérrez
57. Electronic Tongues Purposely Designed for the Organoleptic Characterization of Olive Oils 525 María L. Rodríguez-Méndez, C. Apetrei and José A. De Saja
58. Determination of Olive Oil Parameters by Near Infrared Spectrometry 533
Sergio Armenta, Javier Moros, Salvador Garrigues and Miguel de la Guardia Cirugeda
59. Determination of Olive Oil Acidity 545 Marcone Augusto Leal de Oliveira, Manoela Ruchiga Balesteros, Adriana Ferreira Faria and Fernando Antonio Simas Vaz
60. Application of the Electronic Nose in Olive Oil Analyses 553
M. Stella Cosio, Simona Benedetti, Susanna Buratti, Matteo Scampicchio and Saverio Mannino
61. Squalene and Tocopherols in Olive Oil: Importance and Methods of Analysis 561 Maria Z. Tsimidou
62. An Overview of the Chemometric Methods for the Authentication of the Geographical and Varietal Origin of Olive Oils 569
Federico Marini, Remo Bucci, Antonio L. Magrì and Andrea D. Magrì
ix
Contents
63. Characterization of Three Portuguese Varietal Olive Oils Based on Fatty Acids, Triacylglycerols, Phytosterols and Vitamin E Profiles: Application of Chemometrics 581
71. Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS 667
64. Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil 591
Roberto Romero-González, Antonia Garrido Frenich and José Luis Martínez Vidal
Section 2 Nutritional, Pharmacological and Metabolic Properties of Olives and Olive Oil 685
13
2.1 General Nutrition 687
Joana S. Amaral, Isabel Mafra and M. Beatriz P.P. Oliveira
65.
C Nuclear Magnetic Resonance Spectroscopy as a New Quantitative Method for Determining Fatty Acid Positional Distribution in Olive Oil Triacylglycerols: Applications to Olive Oil Authenticity 603 Giovanna Vlahov
66. Extraction Techniques for the Analysis of Virgin Olive Oil Aroma 615 Stefania Vichi
67
Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil 625 Lorenzo Cerretani and Alessandra Bendini
68. Polycyclic Aromatic Hydrocarbons (PAHs) in Olive Oil: Methodological Aspects of Analysis 637 Martin Rose
Adverse Components 69. Determination of Aflatoxins and Ochratoxin A in Olive Oil 645
Chiara Cavaliere, Patrizia Foglia, Roberto Samperi and Aldo Laganà
70. Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils 653 Sara C. Cunha, Steven J. Lehotay, Katerina Mastovska, José O. Fernandes and M. Beatriz P.P. Oliveira
Alberto Marinas, Fernando Lafont, María A. Aramendía, I.M. García, José M. Marinas and Francisco J. Urbano
General Aspects and Changes in Food Processing 72. Trends in Olive Oil Production, Supply and Consumption in Mediterranean Countries from 1961 to the Present Day 689 Genevieve Buckland and Carlos A. González
73. The Bioavailability of Olive Oil Phenolic Compounds 699 María-Isabel Covas, Montserrat Fitó, Olha Khymenets and Rafael de la Torre
74. Nutrient Profiles of Commercial Table Olives: Proteins and Vitamins 705 Antonio López-López, Alfredo Montaño and Antonio Garrido-Fernández
75. Nutrient Profiles of Commercial Table Olives: Fatty Acids, Sterols, and Fatty Alcohols 715 Antonio López-López, Alfredo Montaño and Antonio Garrido-Fernández
76. Nutritional Properties of Extra Virgin Olive Oils from the Emilia-Romagna Region: Profiles of Phenols, Vitamins and Fatty Acids 725 Annalisa Rotondi and Chiara Lapucci
77. Table Olives: A Carrier for Delivering Probiotic Bacteria to Humans 735
Paola Lavermicocca, Mauro Rossi, Francesco Russo and Rajaventhan Srirajaskanthan
x
Contents
78. The Oxidative State of Olive Oil Used in Bakery Products with Special Reference to Focaccia 745
86. Vasorelaxant Effects of Oleanolic Acid and Erythrodiol in Pomace Olive Oil 813
79. Recovery and Distribution of Macroand Selected Microconstituents after Pan-frying of Mediterranean Fish in Virgin Olive Oil 755
87. Endothelial Activation and Olive Oil 821
Tommaso Gomes, Debora Delcuratolo, Vito Michele Paradiso and Raffaella Nasti
Rosalia Rodriguez-Rodriguez and Valentina Ruiz-Gutiérrez
Maria Annunziata Carluccio, Marika Massaro, Egeria Scoditti and Raffaele De Caterina
Nick Kalogeropoulos and Antonia Chiou
80. Recovery and Distribution of Macroand Selected Microconstituents after Pan-Frying of Vegetables in Virgin Olive Oil 767 Nick Kalogeropoulos
2.2 Cardiovascular
777
Cardiac Aspects 81. Myocardial Infarction and Protection with Olive Oil 779
Miguel A. Martínez-González, Moises RodríguezManero and Félix Valencia-Serrano
82. Beneficial Effects of Olive Oil Compared with Fish, Canola, Palm and Soybean Oils on Cardiovascular and Renal Adverse Remodeling due to Hypertension and Diabetes in Rat 787 Marcia Barbosa Aguila and Carlos Alberto Mandarim-de-Lacerda
83. Olive Oil and Acute Coronary Syndromes: The CARDIO2000 Case-control Study 795
Demosthenes Panagiotakos and Rena Kosti
Vascular Aspects Including Hypertension 84. Olive Oil Consumption and Reduced Incidence of Hypertension: The SUN Study 801 Alvaro Alonso, Javier S. Perona, Valentina RuizGutiérrez and Miguel A. Martínez-González
85. Virgin Olive Oil and Blood Pressure in Hypertensive Elderly Subjects 807
Javier S. Perona, Alvaro Alonso, Miguel A. Martínez-González and Valentina Ruiz-Gutiérrez
88. Pomace Olive Oil and Endothelial Function 829
Javier S. Perona, Rosana Cabello-Moruno and Valentina Ruiz-Gutiérrez
89. Medicinal Value of Combination of Cholinergic and Calcium Antagonist Constituents in Olives 835
Anwarul Hassan Gilani and Arif-ullah Khan
90. Olive Oil Cultivars and Atherosclerotic Protection in Apolipoprotein E-knockout Mice 845 José Miguel Arbonés-Mainar and Jesús Osada
91. Angiotensinase Activity and Olive Oil Supplementation 853
María Jesús Ramírez-Expósito, María Pilar Carrera and José Manuel Martínez-Martos
Lipid Aspects 92. The Effect of Olive Oil on Postprandial Thermogenesis, Fat Oxidation and Satiety: Potential Implications for Weight Control 863 Mario J. Soares
93. The Influence of Olive Oil on the Lipidemic Profile in Stress 871
Antonia Kotsiou and Christine Tesseromatis
94. Postprandial Triglyceride-rich Lipoprotein Composition and Size after Olive Oil 879
Rosana Cabello-Moruno, Javier S. Perona and Valentina Ruiz-Gutiérrez
xi
Contents
95. The Effects of Olive Oils on Hepatic Lipid Metabolism and Antioxidant Defense Mechanisms: Insights from Proteomics Studies 887
Baukje de Roos, Jose Miguel Arbones-Mainar and Guillermo Rodriguez Gutiérrez
96. Olive Oil Consumption and Weight Gain 895 Maira Bes-Rastrollo, Mario J. Soares and Miguel A. Martinez-Gonzalez
2.3 Oxidative Stress 903 97. Structure–Activity Relationship of Phenolic Antioxidants and Olive Components 905 Juan C. Morales and Ricardo Lucas
98. Antioxidant Capacity of Blood after Extra Virgin Olive Oil Intake in Human Volunteers 915 Cristina Samaniego-Sánchez, Jose Javier Quesada-Granados, Maria Rosa Sánchez-Navarro, Herminia López-Garcia de la Serrana and Maria Carmen López-Martinez
99. Antioxidant Capacity and Phenolic Profile of Table Olives from the Greek Market 925 George Boskou
100. Olive Oil Components on Oxidative Stress and Arachidonic Acid Metabolism 935 Maria Teresa Mitjavila and Juan José Moreno
101. Antioxidant Activity of Solid Olive Oil Residues from Olea europaea ‘Coratina’ Cultivar 943 Giangiacomo Beretta, Giancarlo Aldini and Roberto Maffei Facino
102. Antioxidant and Radioprotective Effects of Olive Leaf Extract 951 J. Julián Castillo, Miguel Alcaraz and Obdulio Benavente-García
2.4 Cancer and Immunology
959
Cancer 103. Olive Oil Prevents Experimentally Induced Breast and Colon Carcinogenesis 961 Betty Schwartz and Zecharia Madar
104. Dietary Fat Including Olive Oil and Breast Cancer in the N-methyl Nitrosourea (NMU) Animal Model 969
María Jesús Ramírez-Expósito, María Pilar Carrera, Pedro Cortés and José Manuel Martínez-Martos
105. Anticarcinogenic Properties of Olive Oil Phenols: Effects on Proliferation, Apoptosis and Differentiation 981 Roberto Fabiani and Guido Morozzi
106. Mutagenic Activity in Meat Samples after Deep-frying in Olive Oil: Comparison with other Oils 989 Adela López de Cerain, Amaya Azqueta and Ariane Vettorazzi
107. Azoxymethane-induced Colon Carcinogenesis through Wnt/betacatenin Signaling and the Effects of Olive Oil 997
Takehiro Fujise, Ryuichi Iwakiri, Ryosuke Shiraishi, Bin Wu and Kazuma Fujimoto
108. Olive Oil and its Phenolic Components and their Effects on Early- and Late-stage Events in Carcinogenesis 1005 Chris I.R. Gill, Yumi Z.H.-Y. Hashim, Maurizio Servili and Ian R. Rowland
Immunology and Inflammation 109. Olives and Olive Oil Compounds Active Against Pathogenic Microorganisms 1013
Manuel Brenes, Eduardo Medina, Aranzazu García, Concepción Romero and Antonio de Castro
xii
Contents
110. Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy 1021
Eva Batanero, Rosalía Rodríguez and Mayte Villalba
111. Olive Oil and Septic Pulmonary Dysfunctions 1033 J. Glatzle
112. Olive Oil and Immune Resistance to Infectious Microorganisms 1039 María A. Puertollano, Elena Puertollano, Gerardo Álvarez de Cienfuegos and Manuel A. de Pablo
113. Intestinal Anti-inflammatory Activity of Dietary Olive Oil 1049 Julio Gálvez, Desiree Camuesco, Maria Elena Rodríguez-Cabezas and Antonio Zarzuelo
114. Use of Olive Oil in Patients with Rheumatoid Arthritis 1057
Décio Sabbatini Barbosa, Andréa Colado Simão and Isaias Dichi
2.5 Other Effects, Uses and Diseases 1065 Cells and Cellular Effects 115. The Beneficial Effects of Virgin Olive Oil on Nuclear Transcription Factor kappaB and Other Inflammatory Markers 1067
Pablo Perez-Martinez, Francisco Perez-Jimenez and Jose Lopez-Miranda
119. Extra Virgin Olive Oil Biophenols and mRNA Transcription of Glutathionerelated Enzymes 1095 Rosaria Varì, Beatrice Scazzocchio, Claudio Giovannini and Roberta Masella
120. Protective Effects of Olive Oil Components Against Hydrogen Peroxide-induced DNA Damage: The Potential Role of Iron Chelation 1103
Alexandra Barbouti, Evangelos Briasoulis and Dimitrios Galaris
121. Olive Oil Phenols and Nitric Oxide Affect Lymphomonocyte Cytosolic Calcium 1111 Giuseppe Arienti, Michela Mazzoni and Carlo A. Palmerini
Skin and Cosmeceuticals 122. Olive Oil in Botanical Cosmeceuticals 1117
Leslie Baumann and Edmund Weisberg
123. Effect of Olive Oil on the Skin 1125 Diana Badiu, Rafael Luque and Rajkumar Rajendram
124. Skin Creams Made with Olive Oil 1133 M. Adolfina Ruiz, José L. Arias and Visitación Gallardo
Major Organ Systems Including Liver and Metabolism
116. In vivo Cytogenetic Effects of Multiple Doses of Dietary Vegetable Oils: Position of Olive Oils 1071
125. Microarray Analysis of Hepatic Genes Altered in Response to Olive Oil Fractions 1143
117. Minor Polar Compounds in Olive Oil and NF-κB Translocation 1079
126. Monounsaturated Fat Enriched with Olive Oil in Non-alcoholic Fatty Liver Disease 1151
Lusânia Maria Greggi Antunes, Maria de Lourdes Pires Bianchi
Sandra Brunelleschi, Angela Amoruso, Claudio Bardelli, Annalisa Romani, Francesca Ieri and Flavia Franconi
118. Olive Oil and Uncoupling Proteins 1087
Alfredo Fernández-Quintela, Itziar Churruca and María P. Portillo
María Victoria Martínez-Gracia and Jesús Osada
Nimmer Assy, Faris Nassar and Maria Grosovski
127. Uptake, Metabolism and Biological Effect of the Olive Oil Phenol Hydroxytyrosol in Human HepG2 Cells 1157 Luis Goya, Raquel Mateos, M. Angeles Martín, Sonia Ramos and Laura Bravo
xiii
Contents
128. Modulation of Hepatic Apoptotic Pathways by Dietary Olive and Sunflower Oil 1167
María I. Burón, Mónica Santos-González and José M. Villalba
129. Influence of Olive Oil on Pancreatic, Biliary and Gastric Secretion: Role of Gastrointestinal Peptides 1175
María Dolores Yago, Nama’a Audi, Mariano Mañas and Emilio Martínez-Victoria
130. Effects of Olive Oil on Fatty Acid Composition of Pancreatic Cell Membranes: Modulation of Acinar Cell Function and Signaling 1185 María Dolores Yago, María Alba Martínez, José Antonio Pariente, Emilio Martínez-Victoria and Mariano Mañas
131. Olives and Olive Oil in the Prevention of Osteoporosis 1195 Véronique Coxam, Caroline Puel and Marie-Jeanne Davicco
132. Effects of Olive Oil and Guar on Fructose-induced Insulin Resistance 1205
María L. Villanueva-Peñacarrillo, Pablo G. Prieto, Jésus Cancelas, Verónica Sancho, Paola Moreno, Willy J. Malaisse and Isabel Valverde
133. Effects of an Olive Oil-enriched Diet on Glucagon-like Peptide-1 1213
Isabel Valverde, Paola Moreno, Jesús Cancelas, Pablo G. Prieto, María L. Villanueva-Peñacarrillo and Willy J. Malaisse
Section 3 Specific Components of Olive Oil and Their Effects on Tissue and Body Systems 1221 3.1 Tyrosol and Hydroxytyrosol 1223 134. The Chemistry of Tyrosol and Hydroxytyrosol: Implications for Oxidative Stress 1225 Alessandra Napolitano, Maria De Lucia, Lucia Panzella and Marco d’Ischia
135. Hydroxytyrosol Lipophilic Analogues: Synthesis, Radical Scavenging Activity and Human Cell Oxidative Damage Protection 1233 Rosa Chillemi, Sebastiano Sciuto, Carmela Spatafora and Corrado Tringali
136. Effects of Hydroxytyrosol on Cyclosporine Nephrotoxicity
1245
Vincenzo Zappia, Patrizia Galletti, Caterina Manna, Stefania D’Angelo, Daniela Napoli, Maria Luigia De Bonis and Giovambattista Capasso
137. Investigation of the Inhibition of Platelet Activation and Anti-thrombotic Action of a Hydroxytyrosol-rich Olive Oil Wastewater Extract in Diabetic Subjects 1253 Claude Louis Léger
138. Nitric Oxide Synthase and Olive Oil Hydroxytyrosol in Endothelial Cells 1257
Christoph A. Schmitt and Verena M. Dirsch
139. Effects of Tyrosol on RAW 264.7 Macrophages Activated by Interferon-γ and Gliadin 1263 Daniela De Stefano, Maria Chiara Maiuri and Rosa Carnuccio
140. Effects of Hydroxytyrosol on Atherosclerotic Lesions in apoEDeficient Mice 1269
María Victoria Martínez-Gracia, Valentina Ruiz-Gutiérrez and Jesús Osada
141. Effects of Hydroxytyrosol on Macrophage Activation 1275
Daniela De Stefano, Maria Chiara Maiuri and Rosa Carnuccio
142. Usage of Hydroxytyrosol for Antimycoplasmal Activity 1283
Pio Maria Furneri and Giuseppe Bisignano
143. Antioxidant Effect of Hydroxytyrosol, a Polyphenol from Olive Oil by Scavenging Reactive Oxygen Species Produced by Human Neutrophils 1289 Fathi Driss and Jamel El-Benna
xiv
Contents
144. Cancer Chemopreventive Activity of Hydroxytyrosol: A Natural Antioxidant from Olives and Olive Oil 1295 M. Emília Juan, Uwe Wenzel, Hannelore Daniel and Joana M. Planas
145. Olive Fruit Extracts and HT-29 Human Colon Cancer Cells 1301
153. Oleic Acid and Inhibition of Glucosyltransferase 1375
Shin-Geon Choi, Se-Ra Won and Hae-Ik Rhee
154. Oleic Acid: The Main Component of Olive Oil on Postprandial Metabolic Processes 1385 Sergio Lopez, Beatriz Bermudez, Yolanda M. Pacheco, Almudena Ortega, Lourdes M. Varela, Rocio Abia and Francisco J.G. Muriana
M. Emília Juan, Uwe Wenzel, Hannelore Daniel and Joana M. Planas
3.2 Oleuropein
1311
146. The Use of Oleuropein on Myocardium 1313
Ioanna Andreadou, Efstathios K. Iliodromitis, Emmanuel Mikros, Alexios-Leandros Skaltsounis and Dimitrios Th. Kremastinos
147. Use of Oleuropein in Experimental Sepsis by Pseudomonas aeruginosa 1321 Evangelos J. Giamarellos-Bourboulis and Taxiarchis Geladopoulos
148. Modulatory Effect of Oleuropein on Digestive Enzymes 1327 Valeria Polzonetti, Paolo Natalini, Silvia Vincenzetti, Alberto Vita and Stefania Pucciarelli
149. Anti-aging Properties of the Olive Constituent Oleuropein in Human Cells 1335 Niki Chondrogianni, Ioanna Chinou and Efstathios S. Gonos
150. The Relationship between Oleuropein Antimicrobial Activity and its Effects on Biological Membranes 1345 Nuria Caturla, Amparo Estepa and Vicente Micol
151. Antimycoplasmal Activity of Oleuropein 1355
Pio Maria Furneri, Anna Piperno and Giuseppe Bisignano
3.3 Oleic Acid
1363
152. Oleic Acid as an Inhibitor of Fatty Acid and Cholesterol Synthesis 1365 Gabriele V. Gnoni, Francesco Natali, Math J.H. Geelen and Luisa Siculella
155. Characteristics of a Population with a High Intake of Oleic Acid and the PPAR gamma 2 Gene (PPARG2) 1395 Sonsoles Morcillo and Federico Soriguer
156. The Neurotrophic Effect of Oleic Acid: Implications for Olive Oil in Health and Disease 1405 José M. Medina and Arantxa Tabernero
3.4 Other Components Found in Olive Plants and Products 1413 157. Maslinic Acid: A Component of Olive Oil on Growth and Protein-turnover Rates 1415 Mónica Fernández-Navarro, Juan Peragón, Francisco J. Esteban, Victoria Amores, Manuel de la Higuera and José A. Lupiáñez
158. Effects of Oleanolic Acid and Maslinic Acid on Glucose and Lipid Metabolism: Implications for the Beneficial Effects of Olive Oil on Health 1423 Jun Liu, Rajkumar Rajendram and Luyong Zhang
159. Functional Properties of Pentacyclic Triterpenes Contained in Pomace Olive Oil 1431 Rosalia Rodriguez-Rodriguez and Valentina Ruiz-Gutierrez
160. Chemical Synthesis of Diverse Phenolic Compounds Isolated From Olive Oils 1439 Jeffrey B. Sperry and Amos B. Smith III
Index
1465
Preface
The olive tree has been nurtured and cultivated for well over 7000 years. In ancient times olive oil was used in sacred ceremonies and rituals: from birth to death. Olive oil has also long been associated with health promoting properties as well as used in culinary preparations and the diet in general. It is now apparent that the nutritional and health promoting benefits of olives and olive oil may have their foundation in scientific fact. There is an increasing body of evidence to suggest that usage of olive oil not only improves cardiovascular function but has therapeutic potential in a variety of conditions as well. Other properties include antioxidant activity and effects on macrophages and apoptosis as well as numerous cellular and pathophysiological processes. Novel compounds have been isolated and characterized. However, contamination occasionally occurs, potentially causing harm. Thus, a vast amount of scientific material and a comprehensive understanding of olives and olive oils is essential. Olives and Olive Oil in Health and Disease Prevention is divided into 3 main sections: 1. General aspects of olives and olive oil 2. Nutritional, pharmacological and metabolic properties of olives and olive oil 3. Specific components of olive oil and their effects on tissue and body systems.
The book includes a diverse range of topic areas such as varieties and variations, ripening, crushing, extraction, sensory qualities, authenticity and the chemometric classification of cultivars. Geographical characterizations, quality, and the chemical composition of olive oil including the major and minor lipid constituents are also described. Uniquely there are specific sections of the biomedical properties of specific compounds from olives and includes coverage of tyrosol, hydroxytyrosol, 3,4-dihydroxyphenyl acetic acid, oleuropein, and other compounds. The effects of individual components or groups of bioactive compounds are described in terms of cell lines, organs or whole populations. Contributors are authors of international and national standing, leaders in the field. Emerging fields of science and important discoveries relating to olive products are also incorporated in the book. This represents a comprehensive source of material related to olives and their constituents. Olives and Olive Oil in Health and Disease Prevention will be essential reading for nutritionists, pharmacologists, health care professionals, research scientists, cancer workers, pathologists, molecular or cellular biochemists, general practitioners as well as those interested in olives or olive oil or the olive industry in general. Professors Victor R. Preedy and Ronald Ross Watson
xv
This page intentionally left blank
List of Contributors Rocio Abia Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain Marcia Barbosa Aguila Laboratory of Morphometry and Cardiovascular Morphology, Biomedical Centre, Institute of Biology, State University of Rio de Janeiro, Brazil Miguel Alcaraz Radiology and Physical Medicine Department, Faculty of Medicine, University of Murcia, Murcia, Spain
Lusânia Maria Greggi Antunes Departamento Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto – USP, São Paulo, Brasil C. Apetrei Department of Chemistry, Faculty of Sciences, ‘Dunaˇrea de Jos’ University of Galati, Romania María A. Aramendía Department of Organic Chemistry, University of Córdoba, Spain
Giancarlo Aldini Istituto di Chimica Farmaceutica e Tossicologica ‘Pietro Pratesi’, Faculty of Pharmacy, University of Milan, Italy
Jose Miguel Arbones-Mainar Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Veterinaria, Universidad de Zaragoza, Spain
Alvaro Alonso Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, MN, USA
José L. Arias Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, University of Granada, Spain Giuseppe Arienti ˙Izmir Institute of Technology, Faculty of Science, Department of Chemistry, Gülbahçe, Urla, Izmir, Turkey; Dipartimento di Medicina Interna, Perugia 06127, Italy
Gerardo Álvarez de Cienfuegos Unit of Microbiology, Department of Health Sciences, Faculty of Experimental Sciences, University of Jaén, Spain Joana S. Amaral Escola Superior de Tecnologia e de Gestão, Instituto Politécnico de Bragança, Portugal Paolo Amirante Department of Engineering and Management of the Agricultural, Livestock and Forest Systems, Agriculture Faculty, University of Bari, Via Amendola 165/a, 70126 Bari, Italy Victoria Amores Institute of Water, University of Granada, Spain Angela Amoruso Department of Medical Sciences, University of Piemonte Orientale ‘A. Avogadro’, Novara, Italy Paula B. Andrade Serviço de Farmacognosia, Faculdade de Farmácia, Universidade do Porto, Portugal I. Andreadou Department of Pharmaceutical Chemistry School of Pharmacy, University of Athens, Panepistimioupolis, Zografou, Greece Alberto Angioni Department of Toxicology, Food and Environmental Unit, University of Cagliari, Italy . Aysun Ankan Izmir Institute of Technology, Faculty of Science, Department of Chemistry, Gülbahçe, Urla, . Izmir, Turkey
Sergio Armenta Department of Analytical Chemistry, Edificio Jeroni Muñoz (Research Building), University of Valencia, Spain Nimmer Assy Liver Unit, Ziv Medical Center, Safed, Israel Nama’a Audi Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain Sodeif Azadmard-Damirchi Department of Food Science, Swedish University of Agricultural Sciences, SLU, Uppsala, Sweden Amaia Azqueta Department of Nutrition, Institute of Basic Medical Science, University of Oslo, Norway Diana Badiu Department of Biochemistry, Ovidius University of Constanza, Constanza, Romania Davide Ballabio Milano Chemometrics and QSAR Research Group, Department of Environmental Sciences, University of Milano-Bicocca, P.za della Scienza, 1–20126 Milano, Italy xvii
xviii
Manoela Ruchiga Balesteros Grupo de Química Analítica e Quimiometria, Departamento de Química, Universidade Federal de Juiz de Fora, MG, Brazil Evaristo Ballesteros Department of Physical and Analytical Chemistry, EPS of Linares, University of Jaén, Linares (Jaén), Spain Décio Sabbatini Barbosa Department of Pathology, Clinical Analysis and Toxicology - University of Londrina, Paraná, Brazil Alexandra Barbouti Laboratory of Biological Chemistry, University of Ioannina Medical School, Ioannina, Greece Claudio Bardelli Department of Medical Sciences, University of Piemonte Orientale ‘A. Avogadro’, Novara, Italy Eva Batanero Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Químicas, Universidad Complutense, Madrid, Spain Leslie Baumann University of Miami Cosmetic Group, Miami Beach, FL, USA Obdulio Benavente-García Research and Development Department, Furfural Español S.A. Murcia, Spain Peter Becker Novo Nordisk A/S, BioProcess Technologies, Måløv, Denmark Alessandra Bendini Department of Food Science, University of Bologna, Cesena (FC), Italy Simona Benedetti Department of Food Science and Technologies, University of Milan, Italy Giangiacomo Beretta Istituto di Chimica Farmaceutica e Tossicologica ‘Pietro Pratesi’, Faculty of Pharmacy, University of Milan, Italy Beatriz Bermudez Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain Maira Bes-Rastrollo Department of Preventive Medicine and Public Health, University of Navarra, Pamplona (Navarra), Spain Maria de Lourdes Pires Bianchi Departamento Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto – USP, São Paulo, Brasil Giuseppe Bisignano Dipartimento Farmacobiologico, Università degli Studi di Messina, Italy Santos Blanco Cell Biology Section, Department of Experimental Biology, University of Jaén, Spain George Boskou Department of the Science of DieteticsNutrition, Harokopio University, Athens, Greece Tania Bracci Scuola Superiore Sant’Anna, Pisa, Italy Laura Bravo Departamento de Metabolismo y Nutrición, Instituto del Frío (CSIC), Madrid, Spain Manuel Brenes Food Biotechnology Department, Instituto de la Grasa (CSIC), Seville, Spain
List of Contributors
Maria Antonietta Brescia Dipartimento di Chimica, Campus Universitario, Università di Bari, Italy Evangelos Briasoulis Department of Oncology, University of Ioannina Medical School, Ioannina, Greece Sandra Brunelleschi Department of Medical Sciences, University of Piemonte Orientale ‘A. Avogadro’, Novara, Italy Remo Bucci Department of Chemistry, Sapienza University of Rome, Italy Genevieve Buckland Unit of Nutrition, Environment and Cancer, Cancer Epidemiology Program, Catalan Institute of Oncology (ICO), Barcelona, Spain Susanna Buratti Department of Food Science and Technologies, University of Milan, Italy María I. Buróníó Departamento de Biología Celular, Fisiología e Inmunología, Facultad de Ciencias, Universidad de Córdoba, Spain Matteo Busconi Istituto di Botanica e Genetica Vegetale, Università Cattolica del Sacro Cuore, Piacenza, Italy Rosana Cabello-Moruno Nutrition and Lipid Metabolism, Instituto de la Grasa, CSIC, Seville, Spain Pierluigi Caboni Department of Toxicology, University of Cagliari, Italy Paolo Cabras Department of Toxicology, University of Cagliari, Italy Cinzia Caggia DOFATA – Dipartimento di Orto Floro Arboricoltura e Tecnologie Agroalimentari, Catania, Italy Desiree Camuesco Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Department of Pharmacology, School of Pharmacy, University of Granada, Spain Jesús Cancelasú Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain Giovambattista Capasso Chair of Nephrology, Department of Internal Medicine, Second University of Naples, Italy F. Priego Capote Marie Curie Annex Building, Department of Analytical Chemistry, Campus of Rabanales, University of Córdoba, Spain Soledad Cárdenasá Department of Analytical Chemistry, University of Córdoba, Spain Maria Annunziata Carluccio C.N.R. Institute of Clinical Physiology, Lecce Section, Italy Alegria Carrasco-Pancorbo Department of Analytical Chemistry, Faculty of Sciences, University of Granada, Spain Rosa Carnuccio Dipartimento di Farmacologia Sperimentale, Università degli Studi di Napoli ‘Federico II’, Napoli, Italy María Pilar Carrera Experimental and Clinical Physiopathology Research Group. Department of Health
xix
List of Contributors
Sciences, Faculty of Experimental and Health Sciences, University of Jaén, E-23071, Jaén, Spain Antonio Segura Carretero Department of Analytical Chemistry, Faculty of Sciences, University of Granada, Spain
Manuel A. Coimbra Department of Chemistry, Universidade de Aveiro, Campus Universitario de Santaigo, Aveiro, Portugal Lanfranco S. Conte Department of Food Science, University of Udine, Italy
Francisco Javier Casado Food Biotechnology Department, Instituto de la Grasa CSIC, Seville, Spain
M. Stella Cosio Department of Food Science and Technologies, University of Milan, Italy
Julián Castillo Research and Development Department, Furfural Español S.A. Murcia, Spain
Maria Isabel Covas Cardiovascular Risk and Nutrition and Human Pharmacology and Clinical Neurosciences Research Groups. Institut Municipal d´Investigació Mèdica (IMIM-Hospital del Mar), Barcelona, Spain. CIBER de Fisiopatología de la Obesidad y Nutrición (CIBEROBN)
Nuria Caturla Instituto de Biología Molecular y Celular, Universidad Miguel Hernández, Avda. de la Universidad s/n. 03202-Elche, Alicante, Spain Lorenzo Cerretani Department of Food University of Bologna, Cesena (FC), Italy
Science,
Rosa Chillemi Dipartimento di Scienze Chimiche, Università di Catania, Viale A. Doria 6, 95125 Catania, Italy Antonia Chiou Laboratory of Chemistry – Biochemistry – Physical Chemistry of Foods, Department of Science of Dietetics-Nutrition, Harokopio University, Athens, Greece Shin-Geon Choi Department of Bioengineering and Technology, Kangwon National University, Chuncheon, Republic of Korea Mildred S. Christian Argus International, Horsham, Pennsylvania, USA Niki Chondrogianni National Hellenic Research Foundation, Institute of Biological Research and Biotechnology, Laboratory of Molecular and Cellular Aging, Athens, Greece Itziar Churruca Department of Nutrition and Food Science. University of País Vasco, Paseo de la Universidad, Vitoria, Spain Miguel de la Guardia Cirugeda Department of Analytical Chemistry, Edificio Jeroni Muñoz (Research Building), University of Valencia, Spain Antonio de Castro Food Biotechnology Department, Instituto de la Grasa CSIC, Seville, Spain Raffaele De Caterina Institute of Cardiology and Center of Excellence on Aging, ‘G. d’Annunzio’ University, Chieti, Italy
Véronique Coxam Unité de Nutrition humaine, UMR1019 (INRA/Université), INRA Theix, Saint Genès Champanelle, France Sara C. Cunha REQUIMTE/Serviço de Bromatologia, University of Porto, Portugal Jose Antonio Curiel Departamento de Microbiología. Instituto de Fermentaciones Industriales. CSIC, Madrid, Spain Stefania D’Angelo Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy; Faculty of Motor Sciences, Parthenope University, Naples, Italy Hannelore Daniel Molecular Nutrition Unit, Department of Food and Nutrition, Technical University of Munich, Freising, Germany Marie-Jeanne Davicco Unité de Nutrition humaine, UMR1019 (INRA/Université), INRA Theix, Saint Genès Champanelle, France Maria Luigia De Bonis Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy M.D. Luque De Castro Marie Curie Annex Building, Department of Analytical Chemistry, Campus of Rabanales, University of Córdoba, Spain Manuel de la Higuera Department of Animal Biology, University of Granada, Faculty of Sciences, Granada, Spain
Chiara Cavaliere Dipartimento di Chimica, ‘Sapienza’ Università di Roma, Italy
Cristina de la Puerta Department of Medical Biochemistry and Molecular Biology, University of Seville Medical School, Spain
Ioanna Chinou University of Athens, School of Pharmacy, Division of Pharmacognosy and Chemistry of Natural Products, Zografou Campus, Athens, Greece
Blanca de las Rivas Departamento de Microbiología. Instituto de Fermentaciones Industriales. CSIC, Madrid, Spain
Maria Lisa Clodoveo Department of Engineering and Management of the Agricultural, Livestock and Forest Systems, Agriculture Faculty, University of Bari, Via Amendola 165/a, 70126 Bari, Italy
Rafael de la Torre Cardiovascular Risk and Nutrition and Human Pharmacology and Clinical Neurosciences Research Groups. Institut Municipal d´Investigació Mèdica (IMIM-Hospital del Mar), Barcelona, Spain
xx
List of Contributors
CIBER de Fisiopatología de la Obesidad y Nutrición (CIBEROBN) Debora Delcuratolo Dipartimento di Progettazione e Gestione dei Sistemi Agro-zootecnici e Forestali, University of Bari, Italy Antonella De Leonardis Department of Agricultural, Food, Environmental and Microbiological Science and Technologies (DiSTAAM), Campobasso, Italy Maria De Lucia Department of Organic Chemistry and Biochemistry, University of Naples Federico II, Naples, Italy Marcone Augusto Leal de Oliveira Grupo de Química Analítica e Quimiometria, Departamento de Química, Universidade Federal de Juiz de Fora, MG, Brazil Manuel A. de Pablo Unit of Microbiology, Department of Health Sciences, Faculty of Experimental Sciences, University of Jaén, Spain Baukje De Roos Rowett Research Institute, Division of Vascular Health, Aberdeen, UK Jose A. De Saja Department of Condensed Matter Physics, Faculty of Sciences, University of Valladolid, Spain Daniela De Stefano Dipartimento di Farmacologia Sperimentale, Università degli Studi di Napoli ‘Federico II’, Napoli, Italy Giuseppa Di Bella Department of Food and Environmental Science, University of Messina, Italy Isaias Dichi Department of Internal Medicine – University of Londrina, Paraná, Brazil Robert M. Diener Argus Pennsylvania, USA
International,
Horsham,
Verena M. Dirsch University of Vienna, Department of Pharmacognosy, Vienna, Austria Marco d’Ischia Department of Organic Chemistry and Biochemistry, University of Naples Federico II, Naples, Italy Fathi Driss INSERM, U773, Centre de Recherche Biomédicale Bichat Beaujon CRB3, Paris, France; Université Paris 7 site Bichat, UMRS 773, Paris, France Giacomo Dugo Department of Food and Environmental Science, University of Messina, Italy Paresh C. Dutta Department of Food Science and Technology, Faculty of Agriculture, University of Tabriz, Iran Yutaka Ebizuk Experimental Station for Medicinal Plant Studies, Graduate School of Pharmaceutical Sciences, The University of Tokyo, Japan Jamel El-Benna Laboratoire de biochimie, CHU Xavier Bichat, Paris, France Francisco J. Esteban Cell Biology Section, Department of Experimental Biology, University of Jaén, Spain
Amparo Estepa Instituto de Biología Molecular y Celular, Universidad Miguel Hernández, Avda. de la Universidad s/n. 03202-Elche, Alicante, Spain Joaquim C.G. Esteves da Silva Centro de Investigação em Química (UP), Chemistry Department, Universidade do Porto, Portugal Roberto Fabiani Dipartimento Di Specialita’ MedicoChirurgiche E Sanita’ Pubblica, Sezione Di Epidemiologia Molecolare E Igiene Ambientale, University of Perugia, Italy Roberto Maffei Facino Istituto di Chimica Farmaceutica e Tossicologica ‘Pietro Pratesi’, Faculty of Pharmacy, University of Milan, Italy M.A. Faria REQUIMTE - Serviço de Bromatologia, University of Porto, Italy Miguel Faria Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal José O. Fernandesé REQUIMTE/Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal Alberto Fernandez-Gutierrez Department of Analytical Chemistry, Faculty of Sciences, University of Granada, Spain Rafael Fernández-Montesinos Department of Cell Therapy and Advanced Therapies, CABIMER-Andalusian Center for Molecular Biology and Regenerative Medicine (CSIC-University of Seville-UPO-Junta de Andalucia), Seville, Spain Mónica Fernández-Navarroó Department of Biochemistry and Molecular Biology I, Faculty of Sciences, University of Granada, Spain Alfredo Fernández-Quintela Department of Nutrition and Food Science. University of País Vasco, Paseo de la Universidad, Vitoria, Spain Adriana Ferreira Faria Grupo de Química Analítica e Quimiometria, Departamento de Química, Universidade Federal de Juiz de Fora, MG, Brazil Montserrat Fitó Cardiovascular Risk and Nutrition and Human Pharmacology and Clinical Neurosciences Research Groups. Institut Municipal d´Investigació Mèdica (IMIM-Hospital del Mar), Barcelona, Spain. CIBER de Fisiopatología de la Obesidad y Nutrición (CIBEROBN) Corrado Fogher Istituto di Botanica e Genetica Vegetale, Università Cattolica del Sacro Cuore, Piacenza, Italy Patrizia Foglia Dipartimento di Chimica, ‘Sapienza’ Università di Roma, Italy Flavia Franconi Department of Pharmacology, University of Sassari, Sassari, Italy Giuseppe Fregapane Departamento de Tecnología de Alimentos, Universidad de Castilla-La Mancha, Ciudad Real, Spain
List of Contributors
Kazuma Fujimoto Department of Internal Medicine and Gastrointestinal Endoscopy, Saga Medical School, Japan Takehiro Fujise Department of Internal Medicine and Gastrointestinal Endoscopy, Saga Medical School, Japan Pio Maria Furneri Dipartimento di Scienze Microbiologiche e Scienze Ginecologiche, Università degli Studi di Catania, Italy Visitación Gallardo Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, University of Granada, Spain Patrizia Galletti Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy Julio Galvez Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Department of Pharmacology, School of Pharmacy, University of Granada, Spain
xxi
Claudio Giovannini Nutrition Unit, Department of Veterinary Public Health and Food Safety, Istituto Superiore di Sanità, Rome, Italy Jörg Glatzle Dept of General and Transplantation Surgery, University Hospital of Tübingen, University of Tübingen, Germany Gabriele V. Gnoni Laboratory of Biochemistry and Molecular Biology, Department of Biological and Environmental Sciences and Technologies, University of Salento, Lecce, Italy Tommaso Gomes Dipartimento di Progettazione e Gestione dei Sistemi Agro-zootecnici e Forestali, University of Bari, Italy Aurora Gómez-Rico Departamento de Tecnología de Alimentos, Universidad de Castilla – La Mancha, Ciudad Real, Spain
Aranzazu Garcia Food Biotechnology Department, Instituto de la Grasa (CSIC), Seville, Spain
Efstathios S. Gonos National Hellenic Research Foundation, Institute of Biological Research and Biotechnology, Laboratory of Molecular and Cellular Aging, Athens, Greece
I.M. García Servicio Central de Apoyo a la Investigación (SCAI), Unidad de Espectrometría de Masas, Universidad de Córdoba, Spain
Carlos A. Gonzalez Unit of Nutrition, Environment and Cancer, Cancer Epidemiology Program, Catalan Institute of Oncology (ICO), Barcelona, Spain
Pedro P. García-Luna Department of Endocrinology and Nutrition, Clinical Nutrition Unit, Virgen del Rocio University Hospital, Seville, Spain
Luis Goya Departamento de Metabolismo y Nutrición, Instituto del Frío (CSIC), Madrid, Spain
Antonio Garrido-Fernández Departamento de Biotecnología de Alimentos, Instituto de la Grasa (CSIC), Sevilla, Spain Salvador Garrigues Department of Analytical Chemistry, Edificio Jeroni Muñoz (Research Building), University of Valencia, Spain Math J.H. Geelen Laboratory of Veterinary Biochemistry, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands
Maria Grosovski Department of Biotechnology, OrtBraude College, Karmiel, Israel Guillermo Rodriguez Gutiérrez Rowett Research Institute, Division of Vascular Health, Aberdeen, UK Marie-Andrée Hartmann lnstitut de Biologie Moleculaire des Plantes (CNRS UPR 2357), Universite Louis Pasteur, 28 rue Goethe, 67083 Strasbourg, France Yumi Z. Hashim UCD Institute of Food and Health, University College Dublin, Belfield, Dublin 4, Ireland
Taxiarchis Geladopoulos AlterChem Co, Athens, Greece
Xiuli Hu College of Life Sciences, Henan Agricultural University, Zhengzhou, China
Evangelos J. Giamarellos-Bourboulis 4th Department of Internal Medicine, University of Athens, Medical School, Greece
Francesca Ieri Department of Pharmaceutical Sciences, University of Florence, Firenze, Italy
Antonia Garrido Frenich Research Group ‘Analytical Chemistry of Contaminants’, Department of Analytical Chemistry, Almeria University, Spain Dimitrios Galaris Laboratory of Biological Chemistry, University of Ioannina Medical School, Ioannina, Greece Anwarul Hassan Gilani Natural Product Research Division, Department of Biological and Biomedical Sciences, Aga Khan University, Karachi, Pakistan Chris I.R. Gill Northern Ireland Centre for Food and Health (NICHE), University of Ulster (Coleraine), Cromore Road, Coleraine, Northern Ireland, UK
Efstathios K. Iliodromitis Second University Department of Cardiology, Medical School, Attikon General Hospital, University of Athens, Athens, Greece Ryuichi Iwakiri Department of Internal Medicine and Gastrointestinal Endoscopy, Saga Medical School, Japan M. Emilia Juan Department de Fisiologia (Farmàcia) and Institut de Recerca en Nutrició i Seguretat Alimentaria (INSA), Universitat de Barcelona, Spain Nick Kalogeropoulos Laboratory of Chemistry – Biochemistry – Physical Chemistry of Foods, Department of Science of Dietetics-Nutrition, Harokopio University, Athens, Greece
xxii
Curtis Kalua CSIRO Plant Industry and Food Futures Flagship, Glen Osmond, SA, Australia Arif-ullah Khan Natural Product Research Division, Department of Biological and Biomedical Sciences, Aga Khan University, Karachi, Pakistan Olha Khymenets Cardiovascular Risk and Nutrition and Human Pharmacology and Clinical Neurosciences Research Groups. Institut Municipal d´Investigació Mèdica (IMIM-Hospital del Mar), Barcelona, Spain. CIBER de Fisiopatología de la Obesidad y Nutrición (CIBEROBN) Rena Kosti Unit of Human Nutrition, Department of Food Science and Technology, Agricultural University of Athens, Greece Antonia Kotsiou Department of Pharmacology, Medical School, University of Athens, Athens, Greece Dimitrios Th. Kremastinos Second University Department of Cardiology, Medical School, Attikon General Hospital, University of Athens, Athens, Greece Fernando Lafont Servicio Central de Apoyo a la Investigación (SCAI), Unidad de Espectrometría de Masas, Universidad de Córdoba, Spain Aldo Laganà Dipartimento di Chimica, ‘Sapienza’ Università di Roma, Italy Jose Maria Landete Departamento de Microbiología. Instituto de Fermentaciones Industriales. CSIC, Madrid, Spain Lara La Pera Department of Food and Environmental Science, University of Messina, Italy Chiara Lapucci LaMMa (Laboratory of Monitoring and Environmental Modelling for the sustainable development), Sesto Fiorentino (Florence), Italy Paola Lavermicocca Institute of Sciences of Food Production, National Research Council, Bari, Italy Claude Louis Léger EA ‘Nutrition Humaine et Athérogénèse’, Faculté de Médecine, Université Montpellier 1, France Steven J. Lehotay U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, Wyndmoor, PA, USA Alessandro Leone Production Sciences, Engineering and Economics for Agricultural Systems Department, University of Foggia, Italy Jun Liu Jiangsu Center of Drug Screening, China Pharmaceutical University, Nanjing, China Vincenzo Lo Turco Department of Food and Environmental Science, University of Messina, Italy Sergio Lopez Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain Adela Lopez de Cerain Department of Food Science and Nutrition, Physiology and Toxicology, Faculty of Pharmacy, University of Navarra, Pamplona, Spain
List of Contributors
Felix Lopez de Felipe Grupo en Biotecnología de Bacterias Lácticas de Productos Fermentados, Instituto del Frío, CSIC, Madrid, Spain Silvia López-Feria Department of Analytical Chemistry, University of Córdoba, Spain Herminia López-Garcia de la Serrana Department of Nutrition and Bromatology, Faculty of Pharmacy, University of Granada, Spain Antonio López-López Departamento de Biotecnología de Alimentos, Instituto de la Grasa (CSIC), Sevilla, Spain Maria Carmen López-Martinezó Department of Nutrition and Bromatology, Faculty of Pharmacy, University of Granada, Spain Jose Lopez-Miranda Reina Sofia University Hospital, Lipids and Atherosclerosis Research Unit, University of Cordoba; CIBER Fisiopatologia de la Obesidad y Nutricion (CIBEROBN), Spain Giovanna Loredana La Torre Dipartimento di Scienze degli Alimenti e dell’Ambiente ‘G. Stagno d’Alcontres’, Università di Messina, Italy Ricardo Lucas Instituto de Investigaciones Químicas, CSIC, 49 Seville, Spain José A. Lupiáñez Department of Biochemistry and Molecular Biology I, Faculty of Sciences, University of Granada, Spain Rafael Luque Green Chemistry Centre of Excellence, The University of York, Heslington, York, UK Vincenzo Macciola Department of Agricultural, Food, Environmental and Microbiological Science and Technologies (DiSTAAM), Campobasso, Italy Zecharia Madar Institute of Biochemistry, Food Science and Nutrition, Faculty of Agricultural, Food and Environmental Quality Sciences, The Hebrew University of Jerusalem, Israel Isabel Mafra REQUIMTE/Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal Andrea D. Magrì Department of Chemistry, Sapienza University of Rome, Italy Antonio L. Magrì Department of Chemistry, Sapienza University of Rome, Italy Maria Chiara Maiuri Dipartimento di Farmacologia Sperimentale, Università degli Studi di Napoli ‘Federico II’, Napoli, Italy Willy J. Malaisse Laboratory of Experimental Hormonology, Brussels Free University, Brussels, Belgium Mariano Mañas Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain Carlos Alberto Mandarim-de-Lacerda Laboratory of Morphometry and Cardiovascular Morphology, Biomedical Centre, Institute of Biology, State University of Rio de Janeiro, Brazil
xxiii
List of Contributors
Caterina Manna Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy Luisa Mannina STAAM Department, University of Molise, Campobasso, Italy Saverio Mannino Department of Food Science and Technologies, University of Milan, Italy Alberto Marinas Department of Organic Chemistry, University of Córdoba, Spain Jose M. Marinas Department of Organic Chemistry, University of Córdoba, Spain Federico Marini Department of Chemistry, Sapienza University of Rome, Italy Panagiota Markaki Laboratory of Food Chemistry, University of Athens, Greece M. Angeles Martíní Departamento de Metabolismo y Nutrición, Instituto del Frío (CSIC), Madrid, Spain María Alba Martínez Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain Miguel A. Martinez-Gonzalez Department of Preventive Medicine and Public Health, Medical School-Clinica Universitaria, University of Navarra, Pamplona, Spain María Victoria Martínez-Gracia Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Veterinaria, Universidad de Zaragoza, Spain José Manuel Martinez-Martos Experimental and Clinical Physiopathology Research Group. Department of Health Sciences, Faculty of Experimental and Health Sciences, University of Jaén, E-23071, Jaén, Spain Emilio Martínez-Victoria Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain Roberta Masella Nutrition Unit, Department of Veterinary Public Health and Food Safety, Istituto Superiore di Sanità, Rome, Italy
Oncology, Dr. Josep Trueta University Hospital of Girona, Spain Eduardo Medina Food Biotechnology Department, Instituto de la Grasa (CSIC), Seville, Spain Jose M. Medina Department of Biochemistry and Molecular Biology, Institute of Neurosciences of Castilla y León (INCYL), University of Salamanca, Spain Vicente Micol Instituto de Biología Molecular y Celular, Universidad Miguel Hernández, Avda. de la Universidad s/n. 03202-Elche, Alicante, Spain Emmanuel Mikros Department of Pharmaceutical Chemistry School of Pharmacy, University of Athens, Panepistimioupolis, Zografou, Greece Maria Teresa Mitjavila Department of Physiology, Faculty of Biology, University of Barcelona, Spain Alfredo Montaño Food Biotechnology Instituto de la Grasa CSIC, Seville, Spain Juan Carlos Morales Instituto de Químicas, CSIC, 49 Seville, Spain
Department,
Investigaciones
Sonsoles Morcillo Servicio de Endocrinología y Nutrición, Hospital Universitario Carlos Haya, CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Málaga, Spain Juan José Moreno Department of Physiology, Faculty of Pharmacy, University of Barcelona, Spain Paola Moreno Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain Sabrina Moret Department of Food Science, University of Udine, Italy Javier Moros Department of Analytical Chemistry, Edificio Jeroni Muñoz (Research Building), University of Valencia, Spain Guido Morozzi Dipartimento Di Specialita’ MedicoChirurgiche E Sanita’ Pubblica, Sezione Di Epidemiologia Molecolare E Igiene Ambientale, University of Perugia, Italy Maria-Jose Motilva Department of Food Technology, University of Lleida, Lleida, Spain
Marika Massaro C.N.R. Institute of Clinical Physiology, Lecce Section, Italy
Rosario Muñoz Departamento de Microbiología, Instituto de Fermentaciones Industriales, CSIC, Madrid, Spain
Katerina Mastovska U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, Wyndmoor, PA, USA
Francisco J.G. Muriana Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain
Raquel Mateos Departamento de Metabolismo Nutrición, Instituto del Frío (CSIC), Madrid, Spain
y
Michela Mazzoni Dipartimento di Medicina Interna, Perugia 06127, Italy Javier Menéndez-Menéndez Catalan Institute of Oncology (ICO)-Health Services Division of Catalonia, Girona Biomedical Research Institute (IdIBGi), Medical
Daniela Napoli Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy Alessandra Napolitano Department of Organic Chemistry and Biochemistry, University of Naples Federico II, Naples, Italy Faris Nassar Department of Medicine, Western Galilee Hopital, Nahariya, Israel
xxiv
List of Contributors
Raffaella Nasti Dipartimento di Progettazione e Gestione dei Sistemi Agro-zootecnici e Forestali, University of Bari, Italy
V. T. Papoti Laboratory of Food Chemistry and Technology, School of Chemistry, Aristotle University of Thessaloniki, Greece
Francesco Natali Laboratory of Biochemistry and Molecular Biology, Department of Biological and Environmental Sciences and Technologies, University of Salento, Lecce, Italy
Vito Michele Paradiso Dipartimento di Progettazione e Gestione dei Sistemi Agro-zootecnici e Forestali, University of Bari, Italy
Paolo Natalini Dipartimento di Scienze Morfologiche e Biochimiche Comparate, Camerino (MC), Italy Cláudia S. Nunes Department of Chemistry, Univeridad de Aveiro, Campus Universitario de Santaigo, Aveiro, Portugal George Nychas Department of Food Science and Technology, Laboratory of Microbiology and Biotechnology of Foods, Agricultural University of Athens, Greece M. Beatriz P.P. Oliveira REQUIMTE - Serviço de Bromatologia, University of Porto, Italy Yutaka Orihara Experimental Station for Medicinal Plant Studies, Graduate School of Pharmaceutical Sciences, The University of Tokyo, Japan Almudena Ortega Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain Francisca Ortega-García Biochemistry and Molecular Biology Section, Department of Experimental Biology, University of Jaén, Spain Jesús Osada CIBER de Fisiopatología de la Obesidad y Nutrición, Instituto de Salud Carlos III, Spain; Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Veterinaria, Universidad de Zaragoza, Spain Durmus¸ Özdemir ˙Izmir Institute of Technology, Faculty of Science, Department of Chemistry, Gülbahçe, Urla, Izmir, Turkey Betül Öztürk ˙Izmir Institute of Technology, Faculty of Science, Department of Chemistry, Gülbahçe, Urla, Izmir, Turkey Yolanda M. Pacheco Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain Alistair G. Paice Department of Clinical Biochemistry, Nutrition and Dietetics, King’s College London, UK Carlo A. Palmerini Dipartimento di Medicina Interna, Perugia 06127, Italy Demosthenes Panagiotakos Department of Nutrition – Dietetics, Harokopio University, Athens, Greece E.Z. Panagou Department of Food Science and Technology, Laboratory of Microbiology and Biotechnology of Foods, Agricultural University of Athens, Greece Lucia Panzella Department of Organic Chemistry and Biochemistry, University of Naples Federico II, Naples, Italy
José Antonio Pariente Departamento de Fisiología, Facultad de Ciencias, Universidad de Extremadura, Badajoz, Spain Vinood B. Patel Department of Biomedical Sciences, University of Westminster, London, UK M. Ángeles Peinado Cell Biology Section, Department of Experimental Biology, University of Jaén, Spain Teresa Maria Pellicanò Department of Chemistry (cube 12th), University of Calabria (UNICAL), Arcavacata of Rende-Cosenza, Italy Juan Peragon Biochemistry and Molecular Biology Section, Department of Experimental Biology, University of Jaén, Spain Francisco Perez-Jimenez Reina Sofia University Hospital, Lipids and Atherosclerosis Research Unit, University of Cordoba, CIBER Fisiopatologia de la Obesidad y Nutricion (CIBEROBN), Spain Pablo Perez-Martinez Reina Sofia University Hospital, Lipids and Atherosclerosis Research Unit, University of Cordoba, CIBER Fisiopatologia de la Obesidad y Nutricion (CIBEROBN), Spain Javier Sanchez Perona Nutrition and Lipid Metabolism, Instituto de la Grasa, CSIC, Spain Anna Piperno Dipartimento Farmaco-Chimicoo, Università degli Studi di Messina, Messina, Italy Joana M. Planas Department de Fisiologia (Farmàcia) and Institut de Recerca en Nutrició i Seguretat Alimentaria (INSA), Universitat de Barcelona, Spain Donatella Pollicino Department of Food and Environmental Science, University of Messina, Italy Valeria Polzonetti Dipartimento di Scienze Morfologiche e Biochimiche Comparate, Camerino (MC), Italy Tiziana Populin Department of Food Science, University of Udine, Italy Maria P. Portillo Department of Nutrition and Food Science, University of País Vasco, Paseo de la Universidad, Vitoria, Spain David Pozo Department of Cellular and Structural Biology, University of Texas Health Science Center, San Antonio, TX, USA Paul Prenzler School of Agricultural and Wine Sciences, EH Graham Centre, Charles Sturt University, Wagga Wagga, NSW, Australia Pablo G. Prieto Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain
xxv
List of Contributors
Stefania Pucciarelli Dipartimento di Scienze Morfologiche e Biochimiche Comparate, Camerino (MC), Italy Caroline Puel Unité de Nutrition humaine, UMR1019 (INRA/Université), INRA Theix, Saint Genès Champanelle, France Elena Puertollano Unit of Microbiology, Department of Health Sciences, Faculty of Experimental Sciences, University of Jaén, Spain María A. Puertollano Unit of Microbiology, Department of Health Sciences, Faculty of Experimental Sciences, University of Jaén, Spain Jose Javier Quesada-Granados Department of Nutrition and Bromatology, Faculty of Pharmacy. University of Granada, Spain
María L. Rodríguez-Méndez Department of Inorganic Chemistry, E. T. S. Ingenieros Industriales, University of Valladolid, Spain Rosalía Rodríguez Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Químicas, Universidad Complutense, Madrid, Spain Rosalia Rodriguez-Rodriguez Instituto de la Grasa (CSIC), Seville, Spain Annalisa Romani Department of Pharmaceutical Sciences, University of Florence, Firenze, Italy Concepción Romero Food Biotechnology Department, Instituto de la Grasa (CSIC), Seville, Spain Maria-Paz Romero Department of Food Technology, University of Lleida, Lleida, Spain
Rajkumar Rajendram Nutritional Sciences Research Division, School of Life Sciences, King’s College London, UK
Roberto Romero-Gonzalez Research Group ‘Analytical Chemistry of Contaminants’, Department of Analytical Chemistry, Almeria University, Spain
María Jesús Ramírez-Expósitoí Experimental and Clinical Physiopathology Research Group. Department of Health Sciences, Faculty of Experimental and Health Sciences, University of Jaén, E-23071, Jaén, Spain
Martin Rose Central Science Laboratory, Sand Hutton, York, UK
Sonia Ramos Departamento de Metabolismo y Nutrición, Instituto del Frío (CSIC), Madrid, Spain
Annalisa Rotondi Institute of Biometeorology, National Research Council, Bologna, Italy
Natividad Ramos-Martos Department of Physical Chemistry, Faculty of Sciences, University of Jaén, Spain
Ian R. Rowland Department of Food Biosciences, University of Reading, Whiteknights, PO Box 226 Reading, UK
Cinzia L. Randazzo DOFATA - Dipartimento di Orto Floro Arboricoltura e Tecnologie Agroalimentari, Catania, Italy Rossana Rando Department of Food and Environmental Science, University of Messina, Italy Russel J. Reiter Department of Cellular and Structural Biology, University of Texas Health Science Center, San Antonio, TX, USA Luis Rejano Food Biotechnology Department, Instituto de la Grasa CSIC, Seville, Spain Hae-Ik Rhee Division of Biotechnology, Kangwon National University, Chuncheon, Republic of Korea Kevin Robards School of Agricultural and Wine Sciences, EH Graham Centre, Charles Sturt University, Wagga Wagga, NSW, Australia Hector Rodriguez Departamento de Microbiología, Instituto de Fermentaciones Industriales, CSIC, Madrid, Spain Maria Elena Rodríguez-Cabezas Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Department of Pharmacology, School of Pharmacy, University of Granada, Spain Moises Rodríguez-Manero Department of Cardiology, Medical School-Clinica Universitaria, University of Navarra, Pamplona, Spain
Mauro Rossi Institute of Food Sciences, National Research Council, Avellino, Italy
Adolfina Ruiz Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, University of Granada, Spain Valentina Ruiz-Gutiérrez Group of Nutrition and Lipid Metabolism, Instituto de la Grasa, Sevilla, Spain Francesco Russo Laboratory of Experimental Biochemistry I.R.C.C.S. ‘Saverio de Bellis’, National Institute of Digestive Diseases, Castellana Grotte (Ba), Italy Danielle Ryan School of Agricultural and Wine Sciences, EH Graham Centre, Charles Sturt University, Wagga Wagga, NSW, Australia Nadia Sabatini Cra-Centro Di Ricerca per L’olivicoltura E L’industria Olearia, Pescara, Italy Antonio Sacco Dipartimento di Chimica, Universitario, Università di Bari, Italy
Campus
Yasemin Sahan Department of Food Engineering, Faculty of Agriculture, Uludag University, Bursa, Turkey Marcello Saitta Dipartimento di Scienze degli Alimenti e dell’Ambiente ‘G. Stagno d’Alcontres’, Università di Messina, Italy Maria Desamparados Salvador Departamento de Tecnología de Alimentos, Universidad de Castilla - La Mancha., Ciudad Real, Spain
xxvi
Roberto Samperi Dipartimento di Chimica, ‘Sapienza’ Università di Roma, Italy Cristina Samaniego-Sánchez Department of Nutrition and Bromatology, Faculty of Pharmacy, University of Granada, Spain
List of Contributors
Andréa Colado Simão Department of Pathology, Clinical Analysis and Toxicology – University of Londrina Paraná, Brazil Giovanni Sindona Dipartimento di Chimica, Università della Calabria, Arcavacata di Rende (CS), Italy
Antonio Higinio Sánchez Food Biotechnology Department, Instituto de la Grasa CSIC, Seville, Spain
Ganapathy Sivakumar Arkansas Biosciences Institute, Arkansas State University, Jonesboro, USA
Maria Rosa Sánchez-Navarro Hospital Universitario San Cecilio, Granada, Spain
Alexios-Leandros Skaltsounis Department of Pharmacognosy, School of Pharmacy, University of Athens, Panepistimioupolis, Zografou, Greece
Verónica Sancho Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain Mónica Santos-González Departamento de Biología Celular, Fisiología e Inmunología, Facultad de Ciencias, Universidad de Córdoba, Spain Jorge A. Saraiva Department of Chemistry, Univeridad de Aveiro, Campus Universitario de Santaigo, Aveiro, Portugal Matteo Scampicchio Department of Food Science and Technologies, University of Milan, Italy Beatrice Scazzocchio Nutrition Unit, Department of Veterinary Public Health and Food Safety, Istituto Superiore di Sanità, Rome, Italy Christoph A. Schmitt University of Vienna, Department of Pharmacognosy, Vienna, Austria Betty Schwartz Institute of Biochemistry, Food Science and Nutrition, Faculty of Agricultural, Food and Environmental Quality Sciences, The Hebrew University of Jerusalem, Israel Sebastiano Sciuto Dipartimento di Scienze Chimiche, Università di Catania, Viale A. Doria 6, 95125 Catania, Italy Egeria Scoditti C.N.R. Institute of Clinical Physiology, Lecce Section, Italy
Amos B. Smith III Department of Chemistry, Center for Neurodegenerative Disease Research and the Marian S. Ward Alzheimer Drug Discovery Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania and the Monell Chemical Senses Center, Philadelphia, PA, USA Carmela Spatafora Dipartimento di Scienze Chimiche, Università di Catania, Viale A. Doria 6, 95125 Catania, Italy Jeffrey B. Sperry Department of Chemistry, Center for Neurodegenerative Disease Research and the Marian S. Ward Alzheimer Drug Discovery Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania and the Monell Chemical Senses Center, Philadelphia, PA, USA Mario J. Soares Program of Nutrition, School of Public Health, Curtin Health Innovation Research Institute, Curtin University of Technology, Perth, WA, Australia Federico Soriguer Servicio de Endocrinología y Nutrición, Hospital Universitario Carlos Haya, CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Málaga, Spain
Rosa M. Seabra Serviço de Farmacognosia, Faculdade de Farmácia, Universidade do Porto, Portugal
Rajaventhan Srirajaskanthan Centre of Gastroenterology, Royal Free Hospital, London, UK
Luca Sebastiani Scuola Superiore Sant’Anna, Pisa, Italy
Naïm Stitia lnstitut de Biologie Moleculaire des Plantes (CNRS UPR 2357), Universite Louis Pasteur, 28 rue Goethe, 67083 Strasbourg, France
Anna Laura Segre Institute of Chemical Methodologies of CNR, Research Area of Rome I, Italy Antonio Segura Department of Analytical Chemistry, Faculty of Sciences, University of Granada, Spain Maurizio Servili Dipartmento di Scienze degli Alimenti, Sezione di Tecnologie e Biotecnologie degli Alimenti, Via S. Costanzo, 06126 Perugia, Italy
Arantxa Tabernero Department of Biochemistry and Molecular Biology, Institute of Neurosciences of Castilla y León (INCYL), University of Salamanca, Spain Antonio Tagarelli Dipartimento di Chimica, Università della Calabria, Italy
Ryosuke Shiraishi Department of Internal Medicine and Gastrointestinal Endoscopy, Saga Medical School, Japan
Fuju Tai College of Life Sciences, Henan Agricultural University, Zhengzhou, China
Luisa Siculella Laboratory of Biochemistry and Molecular Biology, Department of Biological and Environmental Sciences and Technologies, University of Salento, Lecce, Italy
Antonia Tamborrino Department of Engineering and Management of the Agricultural, Livestock and Forest Systems, Agriculture Faculty, University of Bari, Via Amendola 165/a, 70126 Bari, Italy
List of Contributors
C.C. Tassou National Agricultural Research Foundation, Institute of Technology of Agricultural Products, Lycovrissi, Greece Christine Tesseromatis Department of Pharmacology, Medical School, University of Athens, Athens, Greece Roberto Todeschini Milano Chemometrics and QSAR Research Group, Department of Environmental Sciences, University of Milano-Bicocca, P.za della Scienza, 1–20126 Milano, Italy
xxvii
Stefania Vichi Departament de Nutrició i Bromatologia, Facultat de Farmàcia, Universitat de Barcelona, Spain José Luis Martínez Vidal Research Group ‘Analytical Chemistry of Contaminants’, Department of Analytical Chemistry, Almeria University, Spain José M. Villalba Departamento de Biología Celular, Fisiología e Inmunología, Facultad de Ciencias, Universidad de Córdoba, Spain
Jose S. Torrecilla Department of Chemical Engineering, Universidad Complutense de Madrid, Spain
Mayte Villalba Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Químicas, Universidad Complutense, Madrid, Spain
Saïda Trikib Faculte des Sciences de Tunis, Departement des Sciences Biologiques, Campus Universitaire, 2092 Tunis, Tunisia
María L. Villanueva-Peñacarrillo Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain
Corrado Tringali Dipartimento di Scienze Chimiche, Università di Catania, Viale A. Doria 6, 95125 Catania, Italy
Silvia Vincenzetti Dipartimento di Scienze Morfologiche e Biochimiche Comparate, Camerino (MC), Italy
Ioannis N. Tsakiris Centre of Toxicology Science and Research, Department of Medicine, University of Crete, Greece Aristidis M. Tsatsakis Centre of Toxicology Science and Research, Department of Medicine, University of Crete, Greece M.Z. Tsimidou Laboratory of Food Chemistry and Technology, School of Chemistry, Aristotle University of Thessaloniki, Greece Nicola A. Uccella IRESMO Foundation Group, Chemistry Department, Calabria University, Rende (CS), Italy
Alberto Vita Dipartimento di Scienze Morfologiche e Biochimiche Comparate, Camerino (MC), Italy Giovanna Vlahov CRA – OLI Centro per l’Olivicoltura e l’Industria Olearia, Sede Scientifica Città S. Angelo, Angelo (PE), Italy Wei Wang College of Life Sciences, Henan Agricultural University, Zhengzhou, China Edmund Weisberg Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, USA
Francisco J. Urbano Department of Organic Chemistry, University of Córdoba, Spain
Uwe Wenzel Molecular Nutrition Research, Interdisciplinary Research Center, Justus-Liebig-University of Giessen, Giessen, Germany
Miguel Valcarcel Department of Analytical Chemistry, University of Córdoba, Spain
Se-Ra Won Division of Biotechnology, Kangwon National University, Chuncheon, Republic of Korea
Félix Valencia-Serrano Cardiology Department, Hospital Virgen de la Victoria, University of Malaga, Spain
Bin Wu Department of Internal Medicine and Gastrointestinal Endoscopy, Saga Medical School, Japan
Patrícia Valentãoí Serviço de Farmacognosia, Faculdade de Farmácia, Universidade do Porto, Portugal Isabel Valverde Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain
María Dolores Yago Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain
Lourdes M. Varela Cellular and Molecular Nutrition, Instituto de la Grasa (CSIC), Seville, Spain
Dimitrios Zabaras CSIRO/Food Science Australia, Food Quality and Safety, NSW, Australia
Rosaria Varì Nutrition Unit, Department of Veterinary Public Health and Food Safety, Istituto Superiore di Sanità, Rome, Italy
Vincenzo Zappia Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy
Fernando Antonio Simas Vaz Grupo de Química Analítica e Quimiometria, Departamento de Química, Universidade Federal de Juiz de Fora, MG, Brazil
Antonio Zarzuelo Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Department of Pharmacology, School of Pharmacy, University of Granada, Spain
Ariane Vettorazzi Department of Food Science and Nutrition, Physiology and Toxicology, Faculty of Pharmacy, University of Navarra, Pamplona, Spain
Luyong Zhang Jiangsu Center of Drug Screening, China Pharmaceutical University, Nanjing, China
This page intentionally left blank
Abbreviations
γ: ω-3: μm: [Ca2⫹]c: [O]⫹: [OO]⫹: ⬍r⬎: ΔAbs: μg: μM: 1,2-DG: 1,3-DG: 10-H-Ol Agl: 13 C NMR: 16-NS: 1 H HR-MAS NMR: 1 1
H HR-MAS:
H NMR:
2-DE: 3,4 DHPEA-EDA: 3,4-DHPEA: 3,4-DHPEA-AC: 3,4-DHPEA-EA: 3,4-DHPEA-EDA: 3,5-DB-4-HBA: 3-D: 4,8-DiMeIQx: 5-ASA: 5-HMF:
Gamma is the third carbon atom from the methyl chain end of fatty acid micrometer cytosolic Ca2⫹ concentrations Monoleoylglycerol Fragment Dioleoylglycerol Fragment steady-state fluorescence anisotropy change in absorbance micrograms Micromolar 1,2-diacylglycerol 1,3-diacylglycerol 10-hydroxy-oleuropein aglycon carbon-13 nuclear magnetic resonance 16-doxyl-stearic acid 1 H (Proton) High Resolution Magic Angle Spinning Nuclear Magnetic Resonance 1 H (Proton) High Resolution Magic Angle Spinning proton nuclear magnetic resonance two-dimensional gel electrophoresis (3,4-dihydroxyphenylethyl 4-formyl-3-formylmethyl4-hexenoate) 3,4-dihydroxyphenethyl alcohol 4-(acetoxyethyl)-1,2-dihydroxybenzene isomer of oleuropein aglycon 2-(3,4-dihydroxyphenyl)ethyl ester of elenolic acid dialdehyde 3,5-di-t-butyl-4-hydroxybenzyl alcohol three dimensional structure 2-amino-3,4,8-trimethylimidazo(4,5-f)-quinoxaline 5-aminosalicylic acid 5-hydroxy-2-methylfurfural
5MC: 5-NS: 7α-nAch receptor: 7,8-DiMeIQx: 8-iso-PGF2α: 8-OHdG: A: A: A: A: aa: AA: AAPH: AAS: AAT: ABTS: Ac Pin: Ac: Ac2O: ACAT: ACC: ACE: Ach: CAN: ACP: ACS: ACTH: ACY: AD: AD: ADH: ADI: ADP: AES: AFB1: AFLP: AFs: AG: AI: ALA: ALARA: ALT:
5-Methylchrysene 5-doxyl-stearic acid 7α nicotinic acetylcholine receptor 2-Amino-3,7,8-trimethyl-imidazo (4,5-f)-quinoxaline 8-isoprostaglandinF2α 8-hydroxy-deoxyguanosine araquidate Aspergillus arachidoyl Arrhenius pre-exponential factor amino acid arachidonic acid 2,2’-azobis(2-amidinopropane) hydrochloride atomic absorption spectrometry alcohol acyl transferase 2,2-azino-bis-3-ethylbenzothiazoline6-sulphonic acid (⫹)-1-acetoxypinoresinol acetyl acetic anhydride cholesterol acyltransferase acetyl-CoA carboxylase angiotensin-converting enzyme acetylcholine acetonitrile acenaphthene acute coronary syndrome adrenocorticotropin acenaphthylene absolute protein-degradation rate atopic dermatitis alcohol dehydrogenase acceptable daily intake adenosine diphosphate atomic emission spectrometry aflatoxin B1 amplified fragment length polymorphism aflatoxins absolute protein-accumulation rate articular index alpha-linolenic acid as low as reasonably achievable alanine aminotransferase xxix
xxx
AMI: AMPK: Ang: ANN: ANOVA: ANT: AOAC, Int.: AOCS: AOTF: AP-1: APA: Apaf-1: APB: APCI: API: Apig: APN: Apo: APPI: ARA: ARE: Art: AS: AspAP: AST: ATP: ATR: ATSDR: AUC: AV: AVP: AVT: B: b.w.: BA: BaA: Bak: BaP: BAW: Bax: BbF: BcL: Bcl-2: bcl-xL: BDE: BeP: BgP: BH4: BHT: Bid: BjF: BkF:
Abbreviations
acute myocardial infarction AMP-activated protein kinase angiotensin artificial neural network ANalysis Of VAriance anthracene International Association of official Analytical Chemists American Oil Chemists Society acousto-optical tunable filter c-Jun part of activating protein-1 aminopeptidase A apoptotic protease activating factor 1 aminopeptidase B atmospheric pressure chemical ionization atmospheric pressure ionization apigenina aminopeptidase N apolipoprotein atmospheric pressure photoionization water-saving decanter antioxidant responsive elements article absolute protein-synthesis rate Aspartyl aminopeptidase aspartate aminotransferase adenosine triphosphate attenuated total reflectance Agency for Toxic Substances and Disease Registry area under the curve p-anisidine value vasopressin alkaline volumetric titration break body weight bile acids benz[a]anthracene BCL-2 antagonist killer 1 benzo[a]pyrene bulk acustic wave BCL-2 associated x protein benzo[b]fluoranthene benzo[c]fluorene B-cell CLL/Lymphoma 2 BCL-2 related gene, long isoform bond dissociation enthalpy benzo[e]pyrene benzo[ghi]perylene tetrahydrobiopterin butylhydroxytoluene BCL-2 interacting domain death agonist benzo[j]fluoranthene benzo[k]fluoranthene
BMCP1:
brain mitochondrial carrier protein-1 BMD: bone mineral density BMI: body mass index Bn: benzyl BnBr: benzyl bromide BP: back-propagation BP: blood pressure BPC: base peak chromatogram BPE: base peak electropherogram BPs: biophenols Brij 35®: polyoxyethylene 23 lauryl ether BSTFA: (bis(trimethylsilyl)trifluoroacetamide) BTEXS: benzene, toluene, ethylbenzene, xylene isomers, styrene bw: body weight C: catechol C: centigrade C: stratum corneum C10: fatty acids with a chain length of ten carbon atoms C12: fatty acids with a chain length of twelve carbon atoms C14:0: myristic acid C15:0: pentadecanoic acid C16: fatty acid with 16 carbon atoms (palmitic, palmitoleic acids) C16:0: palmitic acid C16:1: palmitoleic acid C17:0: margaric acid C17:1: heptadecenoic acid C18: fatty acids with 18 carbon atoms (estearic, oleic, linoleic, linolenic acids) C18:0: stearic acid C18:1: oleic acid C18:1 trans: trans oleic acid C18:1(9cis): oleic acid C18:1/C18:2: ratio between acid oleic and acid linoleic C18:1t: elaidic acid C18:2 n-6: linoleic acid C18:2(9cis,12cis): linoleic acid C18:2t: trans-linoleic acid C18:3 n-3: linolenic acid C18:3t: trans-linolenic acid C20:0: arachidic acid C20:1: eicosenoic acid C20:2: n-6 eicosadienoic acid C21:0: heneicosanoic acid C22:0: behenic acid C22:2: n-6 docosadienoic acid C23:0: tricosanoic acid C24:0: lignoceric acid
xxxi
Abbreviations
C24:1: CA: Ca⫹⫹: CaBP: CAD: CAIMAN: CAL: cAMP: CAO: CAR: Carb: CAS: CAT: CBB: CBM: CC50: CCK: CCK-8: CDA: cDDP: CDK: CDKi: CDs: CE: CECT:
nervonic acid cluster analysis calcium Ca2⫹-binding protein coronary artery disease Classification And Influence Matrix ANalysis Candida antarctica lipase cyclic adenosine monophosphate canola oil carboxen carboxen Chemical Abstracts Service catalase Coomassie blue carbohydrate-binding module 50% cytotoxic concentration cholecystokinin cholecystokinin-octapeptide canonical discriminant analysis cisplatin cyclin-dependent kinases inhibitors of cyclin-dependent kinases conjugate dienes capillary electrophoresis Colección Española de Cultivos Tipo (Spanish Collection of Type Cultures) CEN: European Committee for Standardization CERCLA: Comprehensive Environmental Response, Compensation, and Liability Act CETP: cholesterol ester transfer protein CH: cumene hydroperoxide CHD: coronary heart disease Ch-L: chymotrypsin-like activity CHO: carbohydrate CHO: Chinese hamster ovary (assay) CHR: chrysene CI: chemical ionization cIAP: cellular inhibitor of apoptosis protein CiC: citrate carrier CID: collision-induced dissociation CLA: conjugated linoleic acid isomers CLAMP: hyperinsulinemic euglycemic clamp; GE, gastric emptying CLS: classical least squares CLSA: closed-loop stripping apparatus CM: chylomicrons CM: carboxymethyl Cmax: maximum plasma concentration CMr: chylomicron remnant CN: total carbon number CNS: central nervous system CNT: carbon nanotube CoA: Coenzyme A Con A: concanavalin A
COO: CoQ: COX: COX-2: CP: CP: Cp: CPD: CPE: CPK: CPK-MB: CPP: CREB: CRF: CRP: Cs: CSA: CSF: CSI: CSIC:
corn oil coenzyme Q cyclo-oxygenase cyclo-oxygenase-2 counter-propagation conducting polymers cyclopentadienyl cumulative population doublings carbon paste electrode creatine phosphokinase creatine phosphokinase-MB cyclopenta[cd]pyrene cAMP response element binding protein Corticotropin-releasing factor C-reactive protein protein-synthesis capacity cyclosporin A cerebrospinal fluid cholesterol saturation index Consejo Superior de Investigaciones Científicas CtD: C-terminal domain Cv: cultivar CVA: canonical variate analysis Cvv: cultivars CW: carbowax CYP: cytochrome P450 enzymes Cys: cysteine residue CysNO: S-nitrosocysteine D.Lgs.: legislative decree D: amount of reversibly denatured enzyme D1: relaxation delay D2O: deuterium oxide D6D: Δ6-desaturase Da: Dalton DA: discriminant analysis DAD: diode-array detector DAG: diacylglycerol DALI: dialdehydic form of the ligstroside DAOL: dialdehydic form of the oleuropein DB: number of double bonds DBU: 1,8-diazabicyclo[5.4.0]undec-7-ene DC: direct current D-CAIMAN: discriminant CAIMAN DCC: dicyclohexylcarbodiimide DCF: dichlorofluorescein DEAE: diethylaminoethyl Decarbox-Lig Agl: decarboxilated derivatives of Lig Agl DEHA: bis-(2-ethylhexyl)adipate DEHP: bis-(2-ethylhexyl)phthalate DEI: dermal–epidermal interface DEJ: dermal–epidermal junction DELFIA: dissociation-enhanced lanthanide fluoroimmunoassay
xxxii
DeP: DEPE: DEPT: DFT: DG: DhA: DHA: DHE: DHP: DhP: DHS: DIBAL-H: DIGE: DiP: DISC: DIT: DL: DL-DOPA: DlP: DM: DM2: DMAP: DMEM: DMF: DMPA: DMPC: DMPD: DMPE: DMPG: DMPO: DMPS: DMSO: DNA: DOA: DOP: DOP: DOPET: DPE: DPH: DPLS: DPPH: DSC: DSS: DTD: DXR: e.p.: E: EA: Ea: EC:
Abbreviations
Dibenzo[a,e]pyrene 1,2-dielaidoyl-sn-glycero-3-phosphoethanolamine distortionless enhancement by polarization transfer density functional theory diglycerides dibenz[a,h]anthracene docosahexaenoic acid dihydroethidium dihydropyran dibenzo[a,h]pyrene dynamic headspace diisobutylaluminum hydride fluorescence difference gel electrophoresis Dibenzo[a,i]pyrene death-induced signaling complex diet-induced thermogenesis detection limit DL-dihydroxyphenylalanine Dibenzo[a,l]pyrene diabetes mellitus type 2 diabetes mellitus dimethylaminopyridine Dulbecco’s modified Eagle’s medium N,N-dimethylformamide 1,2-dimyristoyl-sn-glycero-3-phosphate 1,2-dimyristoyl-sn-glycero3-phosphocholine N,N-dimethyl-p-phenylenediamine 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine 1,2-dimyristoyl-sn-glycero-3- [phosphorac(1-glycerol)] 5,5-dimethyl-1-pyrroline-N-oxide 1,2-dimyristoyl-sn-glycero-3- [phosphoL-serine] dimethylsulfoxide deoxyribonucleic acid decarboxylated oleuropein aglycon denomination of protected origin di-n-octyl phthalate hydroxytyrosol 2-(3,4-dihydroxyphenyl)ethanol 1,6-diphenyl-1,3,5-hexatriene discriminant partial least squares 1,1-diphenyl-2-picrylhydrazyl differential scanning calorimetry dextran sodium sulfate direct thermal desorption doxorubicin edible portion eosine elenolic acid activation energy (kJ mol⫺1) European Community
EC: EC: EC: EC50: ECD: ECN: eCVRSD: ED: EDA: EDCI: EDHF: EDI: EDTA: EEC: EEF: EEM: EET: EFA: EG: EI: EIA: EIE: EIS: EKC: ELISA: ELSD: EMM: EMSA: eNOS: Eo: EP: EPA: EPA: EPA: EPC: EPIC: EPO: EPR: EpRE: ER: ERK: ESI: ESI-MS: EST: ESYD: ETA: EtOAc: EtOH: EU: EVO: EVOO:
efficient concentration endothelial cells ethyl catechol half maximal effective concentration electron capture detector equivalent carbon number external cross-validation relative standard deviation electrochemical detection dialdehydic form of decarboxymethyl elenolic acid N-ethyl-N’-(3-dimethylaminopropyl) carbodiimide hydrochloride endothelial-derived hyperpolarizing factor estimated daily intake ethylenediaminetetracetic acid European Economic Community excitation and emission fluorescence excitation-emission matrix epoxyeicosatrienoic acid essential fatty acids ethyl guaiacol electron ionization enzyme immunoassay extracted ion electropherogram electrospray ionization electrokinetic chromatography enzyme-linked immunosorbent assays evaporative light-scattering detection excitation emission matrices electrophoretic mobility shift assay endothelial nitric oxide synthase total amount of enzyme (Eo ⫽ N ⫹ D) ethyl phenol eicosapentaenoic acid Environmental Protection Agency eicosapentaenoic acid Epithelioma papulosum cyprini cells European Prospective Investigation into Cancer and Nutrition evening primrose oil electron paramagnetic resonance electrophile responsive elements endoplasmic reticulum extracellular signal-regulated kinase electrospray ionization electrospray ionization mass spectrometry expressed sequence tag Hellenic Accreditation System eicosatrienoic acid ethyl acetate ethanol European Union virgin olive oil enriched with its own unsaponifiable fraction extra virgin olive oil
xxxiii
Abbreviations
F: FA: FAD: FADD: FADH2: FAME: FAO:
fragment fatty acid flavin adenine dinucleotide Fas-associated death domain protein flavin adenine dinucleotide (reduced form) fatty acids methyl esters Food and Agriculture Organization of the United Nations FAO/WHO: Food and Agriculture Organization/World Health Organization FAS: fatty acid synthase FASL: Fas ligand FD: fluorescence detector FDA: Food and Drug Administration FDA: factorial discriminant analysis FER: feed–efficiency ratio FFA: free fatty acids FFA: free fatty acidity FIA: flow injection analysis FIA-ECD: flow injection analysis with electrochemical detection FID: flame ionization detector FLD: fluorescence detection FLIP: FADD-like interleukin-1β-converting enzyme (FLICE)-like inhibitory protein FLR: fluorene FLT: fluoranthene fMLP: N-formyl-methionyl-Leucyl-phenylalanine FMN: flavin adenine mononucleotide FPD: flame photometric detector FPLC: fast performance liquid chromatography FR: France FRAP: ferric-reducing antioxidant power FSIVGTT: frequently sampled intravenous glucose tolerance test FT: Fourier transform FT-IR: Fourier-transformed infrared spectroscopy FTIR: Fourier transform infrared FTIR-ATR: Fourier transform infrared–attenuated total reflectance spectroscopy FT-Raman: Fourier transform Raman spectroscopy FVIIa: activated factor VII g: grams G: gliadin G: glucose G: stratum glanulosum GA: genetic algorithm GAGs: glycosaminoglycans GAP-43: growth associated protein-43 GC: gas chromatography GCB: graphitized carbon black GC-CI-MS: gas chromatography-chemical ionization mass spectrometry GC-ECD: gas chromatography-electron capture detection GC-FPD: gas chromatography-flame photometric detection
GC-MS: GC-MS: GC-MS:
gas chromatography-mass spectrometry high pressure liquid chromatography mass spectrometry coupled to gas chromatography GC–MS/MS: gas chromatography–tandem mass spectrometry. GC-MS: coupled gas chromatography-mass spectrometry GC-NPD: gas chromatography-nitrogen-phosphorus detection GC-O: gas chromatography-olfactometry GFAAS: graphite furnace atomic absorption spectrometry GFR: glomerular filtration rate GILS: genetic inverse least squares GLA: gamma-linolenic acid GLP-1: glucagon-like peptide 1 Glu: glucosyl GluAP: glutamyl aminopeptidase Glut-4: glucose transporter GMP: good manufacturing practice GO: glucose oxidase GPC: gel-permeation chromatography GPx: glutathione peroxidase GR: glutathione reductase GR: whole-body growth rate GR: Greece GSFS: German Society for Fat Science GSH: glutathione GSS: genome survey sequence GSSG: oxidized glutathione GST: glutathione S-transferase GTFs: glucosyltransferase GTP: guanosine triphosphate GTTTM: glucose and triglyceride tolerance test meal H⫹: hydrogen ion H: haematoxiline H2O2: hydrogen peroxide HbA1c: glycated hemoglobin A1c HCA: hierarchical cluster analysis HCA: heterocyclic amines HDFs: human diploid fibroblasts HDL: high-density lipoproteins HDLc: high-density-lipoprotein cholesterol HDL-cholesterol: high-density lipoprotein-cholesterol HER2: Human Epidermal growth factor Receptor 2 HETE: hydroxy-eicosatetraenoic acid HII: inverted hexagonal-HII phase HL: human leukemia HMGCR: 3-hydroxy-3-methylglutaryl-CoA reductase HMPA: hexamethylphosphoramide HOAc: acetic acid HOCl: hypoclorous acid
xxxiv
HOMA: HOMA-ir:
homeostasis model assessment homeostasis model analysis-insulin resistance HOMA-IR: homeostasis model assessment insulin resistance index HOSO: high oleic sunflower oil HPA: hypothalamic-pituitary-adrenocortical HPETE: hydroperoxy-eicosatetraenoic acid H-Pin: hydroxy-pinoresinol HPL: hydroperoxide lyase HPLC: high-performance liquid chromatography HPLC/GC-MS: high-pressure liquid chromatography/ gas chromatography-mass spectrometry HPLC-APcI-MS: high-performance liquid chromatography-atmospheric pressure chemical ionization-mass spectrometry HPLC-APCI-MS: HPLC-atmospheric-pressure chemicalionization mass spectrometry HPLC-DAD: high-performance liquid chromatography-diode array detection HPLC-ECD: high performance liquid chromatography with electrochemical detection HPLC-ES-MS: high-performance liquid chromatography-electrospray ionization-mass spectrometry HPSEC: high-performance size-exclusion chromatography HPTA: hydroxy pentacyclic triterpenic acid hr: hour HRGC: high-resolution gas chromatography HRMS: high-resolution mass spectrometry HRP: horseradish peroxidise HS: headspace HSSE: headspace sorptive extraction HS-SPME: headspace- solid-phase microextraction HT: hydroxytyrosol HT: 3,4-(dihydroxyphenyl)ethanol (also: hydroxythyrosol) HTY: hydroxytyrosol-supplemented yogurt HUVEC: human umbilical vascular endothelial cells HyEDA: dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol HYTY: hydroxytyrosol HYTY-Ac: 2-(4-hydroxyphenyl)ethyl acetate or hydroxytyrosol acetate I: insulin i.p.: intrapenitoneal I/R: infarct to risk area IAC: immunoaffinity columns IAP: inhibitor of apoptosis protein IARC: International Agency for Research on Cancer IBD: inflammatory bowel disease IBIMET-CNR: Biometeorology Institute, National Research Council
Abbreviations
IC: IC50: ICAM-1,-2,-3: ICP: IcP: ICP-AES: ICP-MS: ICR: iCVRSD: IEC: IEF: IFG: IFN-γ: IFR-1: IGI: IGT: IL: IL-6: ILS: imid: iNOS: IOC: IOOC: IP: IP3: IPCS: IPG: IQR: IR: IR: IRMS: IRS: IS: ISI: ISO: ISTD: IT: IT: IT-MS: ITS: IUPAC: JECFA: JNK-1: K⫹: K232: K270: Kat: kcat (T): KD: KDNA:
ion chromatography 50% inhibitory concentration Intercellular adhesion molecule-1,-2,-3 inductively coupled plasma indeno[1,2,3-cd]pyrene inductively coupled plasma-atomic emission spectrometry inductively coupled plasma mass spectrometry ion cyclotron resonance internal cross-validation relative standard deviation ion exchange chromatography isoelectric focusing impaired fasting glucose interferon-γ interferon regulatory factor-1 insulinogenic index impaired glucose tolerance interleukin interleukin 6 inverse least squares imidazole inducible nitric oxide synthase International Olive Council International Olive Oil Council gel–fluid intermediate phase inositol trisphosphate International Programme on Chemical Safety immobilized pH gradient interquartile range infrared spectroscopy insulin resistance isotope ratio mass spectrometry insulin receptor substrate internal standard insulin sensitivity index International Standards Organization Internal Standard induction time Italy ion-trap mass spectrometry internal transcribed spacer International Union of Pure and Applied Chemistry Joint FAO/WHO Expert Committee on Food Additives c-Jun N-terminal kinase-1 Potassium specific absorption at 232 nm specific absorption at 270 nm units of enzyme activity (mol/s) catalytic rate constant of the enzyme fractional protein-degradation rate protein-synthesis rate per cell unit
xxxv
Abbreviations
Keap1: kg: KG: Km: Kmax: KNN: Kp: KRD: KRNA: KS: L: L: L: LA: LAH: LC: LCAT: LC-LC-UV: LC-MS: LD50: LDA: LDH: LDL: LDL-cholesterol: LExt: LFA-1: LFA-1: LIF: Lig Agl: LiHMDS: LLE: LLL: Ln: Ln: LNA: L-NAME: LnPc2: LOD: LOO: LOQ: LOX: LPL: LPO: LPS: Ls: LT: LTA: LTB4: Lut: M: m/z:
Kelch-like ECH-associated protein-1 kilogram fractional protein-accumulation rate Michaelis-Menten constant maximum specific absorption close to 270 nm K Nearest Neighbors phospholipid/water partition coefficient equilibrium denaturation constant protein-synthesis efficiency fractional protein-synthesis rate linoleate linoleoyl stratum lucidum linoleic acid lithium aluminum hydride liquid chromatography lecithin cholesterol acyltransferase coupled-column liquid chromatography with UV detection liquid chromatography-mass spectrometry lethal dose 50 lithium diisopropylamide lactate dehydrogenase low-density lipoprotein low-density lipoprotein-cholesterol olive leaf extract lymphocyte function antigen-1 leukocyte function associated antigen-1 laser-induced fluorescence ligstroside aglycon lithium hexamethyldisilazide liquid–liquid extraction 1,2,3-Trilinoleylglycerol linolenate linolenoyl linoleic acid N-Nitro-L-Arginine Methyl Ester lanthanide bisphthalocyanine limit of detection 2,3-dioleyl-1-linoleyglycerol limit of quantification lipoxygenase lipoprotein lipase lipoxygenase lipopolysaccharide lignans leukotriene lipoteichoic acid leukotriene B4 luteolin margarate Mass to charge ratio
M1, M2, M3, M4, M5: MABP: MAC: Mac-1: MALDI:
Muscarinic receptors subtypes mean arterial blood pressure Mediterranean aliment culture macrophage antigen-1 matrix-assisted laser desorption ionization MALDI-TOF: matrix-assisted laser desorption ionization time-of-flight MALI: monoaldehydic form of the ligstroside MAOL: monoaldehydic form of the oleuropein MAP-2: microtubule associated protein-2 MAPK: mitogen-activated protein kinase MAS: marker assisted selection MAS: magic angle spinning MAT: modified atmosphere Max.: maximum MCDD: methionine-choline deficient diet Mcl-1: myeloid cell leukemia 1 MCP (1): monocyte chemoattractant protein MCP-1: monocyte chemoattractant protein-1 m-CPBA: meta-chloroperoxybenzoic acid M-CSF: macrophage-colony stimulating factor MCT: medium-chain triglycerides MD: Mediterranean diet MDA: malondialdehyde MDM: monocyte-derived macrophages MED: minimal erythema dose MeIQx: 2-amino-3,8-dimethylimidazo [4,5-f]-quinoxaline MeOH: methanol MES: 4-morpholineethanesulfonic acid MESI: membrane extraction with sorbent interface mg/mL: milligram per liter mg: milligrams MHz: the unit megahertz indicates the spectrometer frequency MI: maturation index MI: mitotic index MI: myocardial infarction MIC: minimum inhibitory concentration Min: minute MIP (1-alpha, 2, 1-beta): macrophage inflammatory protein mL: milliliter MLF-NN: multilayer feedforward neural networks MODS: multiple-organ dysfunction syndrome MOM: methyloxymethyl MOMCl: methyloxymethyl chloride
xxxvi
MOS: MOSFET: MPC: MPLC: MPO: MRI: MRL: MRM: mRNA: MRP: MS: Ms: MS/MS: MSC: MsCl: MSD: MSn: MSPD: MSPDE: MUFA: MW: MWNT: n: N: NAD: NADH: NADPH: NaOAc: NaOEt: NaOH: NAP: NAS: NCBI: ND: NF-κB: ng: NGF: N-glycan: NIF: NIR: NK: NL: NLM: Nm: NMO: NMR: NO: NO2⫺: NOAEL: NOE: NOESY:
Abbreviations
Metal Oxide Semiconductor metal oxide semiconductor field effect transistors minor polar compounds middle pressure liquid chromatography myeloperoxidase magnetic resonance imaging maximum residue limit/level multiple reaction monitoring messenger-ribonucleic acid Maillard reaction product mass spectrometry methanesulfonyl tandem mass spectrometry multiplicative signal correction methanesulfonyl chloride Mediterranean style diets multiple-stage mass spectrometry matrix solid-phase dispersion matrix solid phase dispersion monounsaturated fatty acids molecular weight multiwalled carbon nanotube number of observations amount of native enzyme nicotinamide adenine dinucleotide nicotinamide adenine dinucleotide, reduced form nicotin adenine dinucleotide phosphate reduced form sodium acetate sodium ethoxide sodium hydroxide naphtalene net analyte signal National Center for Biotechnology Information not detected nuclear factor κB nanogram nerve growth factor asparagine-linked glycan nifedipine near infrared natural killer Netherlands nonlinear maps nanometer N-methylmorpholine-N-oxide nuclear magnetic resonance nitric oxide (also: nitrogen monoxide) nitrite no-observed-adverse-effect level nuclear Overhauser effect Nuclear Overhauser Enhancement SpectroscopY
NOS: NOx: NPD: Nrf2: NSAID: nsLTP: nSMase: NT: NtD: O: o/w: O: O2.⫺: OA: OA: OAc: OAS: ODS: OeF.Aq: OeF.CHCl3: OeF.Cr: OeF.EtAc: OeF.Pet: OGIS: OGTT: Ol Agl: Ole e: Ole: OLO: Olp: OMW: OMW: OMWW: OMWW: OO: OOE: OOO: OOWW: OP: OPE: OPO: ORAC assay: os: OSC: OSI: OTOS: oxLDL: ox-TG: p-AV: P: P: P: P/S ratio: P: P:
nitric oxide synthase nitrates ⫹ nitrites nitrogen phosphorus detector NF E2-related factor 2 non-steroidal anti-inflammatory drug non-specific lipid transfer protein Mg2⫹-dependent neutral sphingomyelinase nitrotyrosine N-terminal domain oleate oil-in-water oleoyl anion superoxide occupational asthma oleic acid acetate oral allergy syndrome octadecylsilica Olea europea fruit aqueous fraction Olea europea fruit chloroform fraction Olea europea fruit crude extract Olea europea fruit ethylacetate fraction Olea europea fruit petroleum spirit fraction oral glucose insulin sensitivity oral glucose tolerance test oleuropein aglycon Olea europaea allergens oleuropein olive oil oleuropein olive mill wastewater oil and olive mill wastewater olive mill waste water olive oil mill wastewaters olive oil olive oil extract 1,2,3-trioleoylglycerol olive oil waste waters olive phenol olive pulp extract olive pomace oil oxygen radical absorbance capacity oxidosqualene orthogonal signal correction Oil Stability Index organic table olives oxidized low-density lipoprotein oxidized triglycerides p-anisidine value palmítate pasteurization pirogallol polyunsaturated/saturated fatty acid ratio palmitoleoyl palmitoyl
xxxvii
Abbreviations
P: PA: PAB: PAD:
Pacini corpuscle palmitic acid Propionibacterium phenolic acid decarboxylase, also known as PDC PAF: platelet activating factor PAGE: polyacrylamide gel electrophoresis PAH: polycyclic aromatic hydrocarbons PAI-1: plasminogen activator inhibitor-1 pAOC: plasma antioxidant capacity PAS: photo acoustic spectrometry p-AV: p-Anisidine Value PBMC: peripheral blood mononuclear cells PC: polar compounds PC: principal component PC: protein carbonyl PC1: principal component 1 PC2: principal component 2 PCA: principal component analysis PCC: pyridinium chlorochromate PCR: polymerase chain reaction PCR: principal components regression PDC: pyridinium dichromate PDH: pyruvate dehydrogenase PDMS: polydimethylsiloxane PDO: protected denomination of origin PE: phenylephrine PE: olive oil phenolic extract PECAM-1: platelet endothelial cells adhesion molecule-1 PEG: polyethylene glycol PER: protein-efficiency ratio PF: protection factor PG: prostaglandin pg: picograms pG: G glycoprotein PG: phosphatidylglycerol PGE2: prostaglandin E2 PGH2: prostaglandin H2 PGI: protected geographical indication PGI2: prostacyclin PGPH: peptidylglutamyl-peptide hydrolyzing activity PHAP: (polyhydroxyalkyl)pyrazine PHE: phenanthrene PhIP: 2-amino-l-methyl-6-phenylimidazo [4,5-b]pyridine p-HPEA: p-hydroxyphenyl-ethanol or tyrosol p-HPEA: 2-(p-hydroxyphenyl)ethanol p-HPEA: tyrosol p-HPEA-EA: aldehydic form of elenolic acid linked to hydroxytyrosol p-HPEA-EDA: dialdehyde form of elenolic acid linked to tyrosol p-HPEA-EDA: oleuropein aglycone
pI: PI3K: Pin: PKC: PLA: PLA2: PLG: PLS: PLS-DA: PLTP: PMA: PMA: PMN: PMTDI: Po: POD: PON: PON-1: POO: PP: PPAR: ppm: PPO: PR: PR: PRE: PREDIMED: PSA: PTFE: p-TSA: p-TsOH: PTV: PUFA: PV: pyr: PYR: PYY: Q10: QDA: QDA: QL: QqQ: QqQ-MS: QTL: QuEChERS: R: R1 CSA: R1 D: RA: RAPD: RBC:
Isoelectric point phosphatidylinositol-3-kinase (⫹)-pinoresinol protein kinase C plasma linoleic acid phospholipase A2 poly (lactide-co-glycolide) partial least square partial least squares discriminant analysis phospholipid transfer protein phorbol-myristate acetate phorbol 12-myristate 13-acetate polymorphonuclear neutrophils provisional maximum tolerable daily intake palmitoleate peroxidase paraoxonase paraoxonase 1 second centrifugation-pomace olive oil pancreatic polypeptide peroxisome proliferator-activated receptor parts per million polyphenoloxydase pathogenesis-related protein preservatives protein-retention efficiency Prevencion con Dieta Mediterranea, Prevention with Mediterranean Diet primary secondary amine polytetrafluoroethylene p-toluenesulfonic acid p-toluenesulfonic acid programmed temperature vaporized polyunsaturated fatty acids peroxide value pyridine pyrene peptide YY quotient between the enzyme activity at a temperature (T ⫹ 10) K and the activity at T K quadratic discriminant analysis quantitative descriptive analysis quantitation limit triple quadrupole triple-quadrupole mass spectrometry quantitative trait loci quick, easy, cheap, effective, rugged, safe refrigeration relaxation rate of a chemical shift anisotropy mechanism of relaxation relaxation rate of a dipolar mechanism of relaxation rheumatoid arthritis random amplified length polymorphism red blood cells
xxxviii
RDA: REP: RER: RF: RF: RF: RFLP: RFU: RH: RI: RI: RMSEC: RMSEP: RNS: ROO: ROS: RP: RPE: RP-HPLC:
recommended daily allowance relative error of prediction rough endoplasmic reticulum rheumatoid factor response factor radio frequency restriction fragment length polymorphism relative fluorescence unit relative humidity refractive index ripeness index root mean square error of correlation root mean square error of prediction reactive nitrogen species refined olive oil reactive oxygen species reversed-phase R-phycoerythrin reversed-phase high-performance liquid chromatography RPTc: immortalized renal proximal tubule cells RQ: respiratory quotient rQUICKI: revised quantitative insulin sensitivity check index RSD: relative standard deviation RSECV: relative standard errors of cross-validation RT: reverse transcriptase RT-PCR: reverse transcriptase polymerase chain reaction RTW: retention time window RV: resveratrol S: stearate S: sterilization S: stearoyl SAA: serum amyloid A SAR: structure–activity relationship SAW: surface acoustic wave SBSE: stir bar sorptive extraction SC: scavenging concentration SCAR: sequence characterized amplified regions SCC: specific chemical characteristics SC-CNT: surfactant coated carbon nanotube SCD 1: stearoyl-CoA desaturase 1 SCD: stearoyl-CoA 9-desaturase SCE: saturated calomel electrode SCF: Scientific Committee for Food SCI: Science Citation Index SCOOP: Scientific Cooperation SCP: stripping chronopotentiometry ultraviolet SD: standard deviation SDA: stearidonic acid SDBS: sodium dodecylbenzenesulfonate SDE: simultaneous distillation/extraction SDME: single drop microextraction SDS: sodium dodecyl sulfate
Abbreviations
SDS-PAGE:
sodium dodecyl sulfate polyacrylamide gel electrophoresis SE: standard error SEC: standard error of calibration SECO: secoiridoid SEP: standard error of prediction SER: smooth endoplasmic reticulum SERCA: Ca2⫹-dependent ATPase SF: synchronous fluorescence SFA: saturated fatty acids SFC: supercritical fluid chromatography SFE: supercritical fluid extraction SHR: spontaneous hypertensive rat SHS: static headspace sICAM-1: soluble intercellular adhesion molecule SIDs: secoiridoids SIH: salicylaldehyde isonicotinoyl hydrazone SIM: single ion monitoring SIMCA: soft independent modeling of class analogy SIN-1: 3-morpholinosydnonimine siRNA: short interfering RNA SL: sepsis lymph SLDA: stepwise linear discriminant analysis SLM: supported liquid membrane extraction SL-OO: sepsis lymph collected during enteral olive oil resorption SLS: synchronous luminescence spectroscopy Smac/DIABLO: second mitochondrial-derived activators/ Direct IAP-bind protein with low pI SMC: smooth muscle cells SNP: single nucleotide polymorphism SNS: sympathetic nervous system SNV: standard normal variation SO: sunflower oil SOCCs: store-operated Ca2⫹ channels SOD: superoxide dismutase SOR: solid olive residues SP: Spain SPE: solid-phase extraction SPIA: spring pitting apparatus SPME: solid-phase microextraction SPs: simple phenols SPSS: Statistical Package for the Social Science SR-BI: scavenger receptor B1 SREBP: sterol regulatory element binding protein SREBP-1: sterol regulatory element binding protein 1 SREBP-2: sterol regulatory element binding protein 2 SSPs: seed storage proteins SSR: simple sequence repeat STZ rats: model of type 2 diabetes, obtained by streptozotocin injection during the neonatal period STZ: streptozotocin
xxxix
Abbreviations
SUN: sVCAM-1: SVM: T: T1: T3 : TAG: TAS: TBA2: TBAF: TBARS: t-BHP: t-BOOH: TBS: TBSCl: TC: TCA: TCA: TDI: TDNA: TEM: TF: TF: TFA: TG: TG: TGF: TGP: THE: THF: THP: TK: T-L: TLC: TLP: TLR: TMCS: TMS: TNF: TNFR: TNF-α: TOF: TOF-MS: Topt: TOTOX: TP: TPA: tPA:
Seguimiento Universidad de Navarra soluble vascular cell adhesion molecule support vector machines p-(hydroxyphenyl)ethanol (also: thyrosol) spin-lattice relaxation time triiodothyronine triacylglicerol total antioxidant status thromboxane A2 tetrabutylammonium fluoride lipoperoxidation products t-butylhydroperoxide tert-butyl hydroperoxide tert-butyldimethylsilyl t-butylchlorodimethylsilane total cholesterol trichloroacetic acid tree cluster analysis tolerable daily intake percentage of fragmented DNA Toscana Enologica Mori tissue factor transfer factor trans fatty acid triglycerides thapsigargin transforming growth factor triglyceride oligopolymers total hydroalcoholic extract tetrahydrofuran tetrahydropyranyl toxicokinetics trypsin-like activity thin-layer chromatography thaumatin-like protein toll-like receptor trimethylchlorosilane trimethylsilyl tumor necrosis factor tumor-necrosis factor receptor tumor necrosis factor-α time-of-flight time-of-flight mass spectrometry temperature for maximum activity (2PV ⫹ p-AV), a measure of total oxidation calculated as a combined limit of peroxide and p-anisidine values total oil phenols 12-O-tetradecanoylphorbol acetate tissue plasminogen activator
TPAP: TRADD: TRAIL: TRL: Trp-P-1: TUNEL:
tetrapropylammonium perruthenate TNF receptor 1-associated death domain protein TNF-related apoptosis inducing ligand triglyceride-rich lipoproteins 3-mino-1,4-dimethyl-5H-pyrido(4,3-b)-indole terminal deoxynucleotidyl transferasemediated dUTP-biotin nick end labeling, a marker for programmed cell death TX: tromboxane TY: tyrosol TyEDA: dialdehydic form of decarboxymethyl elenolic acid linked to tyrosol TYR: tyrosol UA: unstable angina UAA: utilized agriculture area UCP: uncoupling protein UDP: uridin diphosphate UEOO: unsaponifiable fraction-enriched olive oil UFA: unsaturated fatty acids UI: unsaturation index UK: United Kingdom UNEP/MAP: United Nations Environment Program/ Mediterranean Action Plan UNEP: United Nations Environment Program UNEQ: unequal class-modeling USDA: United States Department of Agriculture USERIA: ultrasensitive enzyme radioimmunoassay UV: ultraviolet UVR: ultraviolet radiation VC: vinyl catechol VCAM-1: vascular cell adhesion molecule-1 VDP: variable dynamic press decanter VG: vinyl guaiacol VHSV: viral hemorrhagic septicemia virus VLA4: very late antigen-4 VLDL: very low-density lipoprotein Vmax: maximum velocity vo (T): initial enzymatic activity VOCCs: voltage-operated Ca2⫹ channels VOO: virgin olive oil VP: vinyl phenol W: Watt w/o: water-in-oil WAF: week after flowering WCOT: wall coated open tubular WD: Western diet WHO: World Health Organization wk: week XIAP: X chromosome-linked inhibitor of apoptosis
This page intentionally left blank
Section 1
General Aspects of Olives and Olive Oil
This page intentionally left blank
1.1
The Plant, Production, Olives and Olive Oil and Their Detailed Characterization The Plant and Production Olives and Olive Oil The Detailed Characterization of Olives and Olive Products
This page intentionally left blank
Chapter 1
Table Olives: Varieties and Variations Luis Rejano, Alfredo Montaño, Francisco Javier Casado, Antonio Higinio Sánchez and Antonio de Castro Food Biotechnology Department, Instituto de la Grasa CSIC, Seville, Spain
1.1 INTRODUCTION
and resistant to blows and to the action of alkalis and brine. At the international trade level, the most important table olive varieties are Manzanilla, Gordal Sevillana, Hojiblanca, Kalamata and Conservolea, and to a lesser extent Bella de Cerignola, Ascolana Tenera, and Picholine. A characteristic common to almost all olive varieties is their extreme bitterness when tasted fresh. The glucoside oleuropein is responsible for this, and the different processing methods are aimed at removing this compound in order to obtain fruits with more-palatable attributes. It could be said that there are as many processing methods as places where olives are consumed. In an attempt to normalize the different products, the International Olive Council has a Trade Standard Applying to Table Olives (IOOC, 2004a), in which the types, trade preparations, quality factors, and other properties are described. The objective of this chapter is to describe in detail the different kinds or classifications applicable to table olives, explaining the distinctive traits for each case.
Table olives are the products prepared from sound fruits of the cultivated olive tree (Olea europaea L.). Consumption of table olives dates back to antiquity, with Columela, in the year 42 CE, being the first author to describe several methods to prepare edible olives according to their variety and degree of ripeness. Table olive production was initially restricted to the producing regions, mainly around the Mediterranean Sea. Today, however, olive preparation has extended to both North and South America, and even Australia. The world production of table olives is around 1.7–1.8 million tonnes. The main producers are the European Union, Turkey, Egypt, Syria, and Morocco. The United States of America and Argentina are also important. Inside the European Union, Spain is the main producer, followed by Greece and Italy. Table 1.1 presents detailed data on production, export and import, and consumption for the main countries involved in the table olive trade (IOOC, 2007). The first group of countries is that with the highest production and exportation (75% world total). The second group is formed by countries with 25% of production and exportation, but 40% consumption (always referred to world total), which means that this group of countries imports 80% of the total importations. Each olive-growing country has its own typical olive varieties. Of all the olive varieties that exist, only those having suitable characteristics (Table 1.2) are used for table olive processing, and even fewer varieties are used for industrial preparation and international trade. The suitability of olives for table consumption depends on size, shape, flesh-to-stone ratio, flesh finesse, taste, firmness, and ease of stone detachment. Olives weighing between 3 and 5 grams are considered to be medium-sized; over 5 grams they are large. Olives that are more or less spherical in shape facilitate processing operations and have a better market, although some elongated fruits also find favor. The stone should separate easily from the flesh; the higher the flesh-to-stone ratio, the better the commercial value of the olives – a ratio of 5:1 is acceptable. The skin of the fruit should be fine, yet elastic Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1.2 TYPES OF OLIVE ACCORDING TO RIPENESS Table olives are classified as one of the following types, depending on the degree of ripeness of the fresh fruits: (a) Green olives: Fruits harvested during the ripening period, prior to coloring and when they have reached normal size. Once processed, green olive color may vary from green to straw-yellow. (b) Turning-color olives: Fruits harvested before the stage of complete ripeness is attained, at color change. After processing, this type of olive may vary from pink to rosé wine or brown. (c) Black olives: Fruits harvested when fully ripe, or slightly before full ripeness is reached. Once processed, black olives may range from reddish black to violetblack, deep violet, greenish black, or deep chestnut. 5
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
6
SECTION | I The Plant and Production
TABLE 1.1 World table olive production, exportation, importation, and consumption along the last six seasons (IOOC, 2007). This table shows data on production, export and import, and consumption for the main countries involved in the table olive trade. Countries are grouped into two groups (A and B). The group A is that with the highest production and exportation (75% world total), whereas the group B is formed by countries with 25% of production and exportation. Country
Production Average (%)
Exports Average (%)
Imports Average (%)
Consumption Average (%)
Algeria
67.6 (3.9)
0.2 (0.0)
0.2 (0.0)
68.4 (4.0)
European Union
706.3 (41.2)
233.4 (49.2)b
75.7 (16.4)b
561.6 (32.5)
Cyprus
9.8 (0.6)
Croatia
0.9 (0.1)
0.1
0.1 (0.0)
0.9 (0.1)
Egypt
195.8 (11.4)
30.2 (6.4)
0.4 (0.1)
160.5 (9.3)
Iran
23.8 (1.8)
1.0 (0.2)
24.8 (1.4)
Israel
16.1 (0.9)
0.9 (0.2)
6.3 (1.4)
21.8 (1.3)
Jordan
23.9 (1.4)
4.6 (1.0)
2.8 (0.6)
22.1 (1.3)
Lebanon
6.1 (4)
0.8 (0.2)
1.6 (0.3)
6.8 (0.4)
Libya
3.3 (0.2)
0.1 (0.0)
3.8 (0.8)
7.0 (0.4)
Morocco
93.3 (5.4)
61.6 (13.0)
0.0 (0.0)
31.8 (1.8)
0.3 (0.1)
Serbia-Montenegro
0.5 (0.0)
0.2 (0.0)
0.7 (0.0)
Syria
148.3 (8.7)
21.1 (4.4)
0.0 (0.0)
129.1 (7.5)
Tunisia
15.6 (0.9)
0.5 (0.1)
0.0 (0.0)
14.9 (0.9)
Total A
1294.9 (75.5)
353.3 (74.5)
91.8 (19.9)
1043.7 (60.4)
19.7 (4.3)
19.7 (1.1)
Saudi Arabia Argentina
63.0 (3.7)
47.7 (10.1)
0.0 (0.0)
15.3 (0.9)
Australia
3.4 (0.2)
0.3 (0.0)
14.3 (3.1)
17.4 (1.0)
Brazil
0.5 (0.0)
52.5 (11.4)
52.9 (3.1)
Bulgaria
8.2 (1.8)
8.2 (0.5)
Canada
23.8 (5.1)
23.8 (1.4)
1.5 (0.3)
9.8 (0.6)
Chile
9.1 (0.5)
0.8 (0.2)
USA
87.0 (5.1)
4.5 (0.9)
Japan Mexico
11.2 (0.7)
Palestine
6.9 (0.4)
Peru
29.8 (1.7)
5.0 (1.1)
210.8 (12.2) 2.3 (0.1)
4.6 (1.0)
10.9 (0.6)
0.6 (0.1)
7.8 (0.5)
0.1 (0.0)
19.7 (1.1)
Romania
17.7 (3.8)
17.7 (1.0)
Russia
28.8 (7.7)
45.0 (2.6)
Switzerland
20.8 (4.5)
4.5 (0.3)
Turkey
192.5 (11.2)
9.9 (2.1)
122.5 (26.5) 2.3 (0.1)
52.7 (11.1)
Venezuela Other prod. count.
16.2 (0.9)
Other non-prod.
0.0 (0.0)
147.7 (8.6)
2.7 (0.6)
2.7 (0.2)
12.8 (3.5)
29.0 (1.7)
42.0 (9.1)
42.0 (2.4)
Total B
419.5 (24.5)
120.8 (25.5)
370.3 (80.1)
682.8 (39.6)
Total world
1714.4 (100)
474.1 (100)
462.2 (100)
1726.5 (100)
a
Average data (1000 tonnes) corresponding to harvest seasons 2001/2 to 2006/7, this last season’s data are provisional. Without intra-Community trade.
b
7
CHAPTER | 1 Table Olives: Varieties and Variations
TABLE 1.2 World olive varieties suitable for table olive or both table olive and oil extraction (IOOC, 2000). This table shows the major olive varieties grown worldwide suitable for table olive processing as well as those suitable for both table olive and oil extraction. The most important table olive varieties are Manzanilla, Gordal Sevillana, Hojiblanca, Kalamata and Conservolea, and to a lesser extent Bella de Cerignola, Ascolana Tenera, and Picholine. Country
Table
Table and oil
Albania
Kalinjot
Algeria
Azeradj, Blanquette de Guelma, Sigoise
Argentina
Arauco
Chile
Azapa
Cyprus
Ladoelia
Croatia
Oblica
Egypt
Aggezi Shami, Hamed, Toffahi
France
Lucques
Aglandau, Grossane, Picholine Languedoc, Salonenque, Tanche
Greece
Chalkidiki
Amigdalolia, Kalamón, Konservolia, Mastoidis, Megaritiki
Israel
Kadesh, Merhavia
Barnea
Italy
Ascolana Tenera, Giarraffa, Nocellara del Belice, Oliva di Cerignola, Sant´Agostino, Santa Caterina
Carolea, Cassanese, Cellina di Nardò, Cucco, Itrana, Majatica di Ferrandina, Nocellara Etnea, Pizz´e Carroga
Jordan
Rasi´i
Lebanon
Soury
Morocco
Meslala
Haouzia, Menara, Picholine Marocaine
Palestine
Nabali Baladi
Portugal
Carrasquenha, Cordovil de Castelo Branco, Cordovil de Serpa, Galega Vulgar, Maçanilha Algarvia, Redondal
Spain
Aloreña, Gordal de Granada y Sevillana, Loaime, Manzanilla de Sevilla, Mollar de Cieza, Morona
Hojiblanca, Manzanilla Cacereña, Manzanilla Prieta, Morisca, Rapasayo, Villalonga
Syria
Abou-Satl, Kaissy
Doebli, Sorani
Tunisia
Meski
Chétoui, Gerboui, Oueslati
Turkey
Domat, Izmir Sofralik, Uslu
Çekiste, Çelebi, Erkence, Gemlik Memecik, Memeli
USA
Mission
Yugoslavia F.R.
Zutica
8
1.3 TABLE OLIVES ACCORDING TO TRADE PREPARATIONS For thousands of years, olives have been an important foodstuff, possibly essential, for inhabitants living around the Mediterranean basin and in the Middle East. Methods of processing are diverse, and include not only traditional, but also those derived from them and improved by new technologies. The bitterness of olives may be removed by alkaline treatment, by immersion in a liquid to dilute the bitter compound oleuropein, or by biological processes. The product so obtained may be preserved in brine (depending on its specific characteristics), in dry salt, in a modified atmosphere, by heat treatment, by preservatives, or by acidifying agents. Generally, the complete name includes information on the type of raw material, the procedure used for eliminating the bitterness, and the method of preserving the product. Olives may be subjected to the following trade preparations (IOOC, 2004a): (a) Treated olives: Green olives, turning-color olives, or black olives that have undergone alkaline treatment, then been packed in brine in which they undergo complete or partial fermentation, and preserved or not by the addition of acidifying agents: (a-1) Treated green olives in brine. Examples are Spanish-style green olives, Picholine style, and Castelvetrano style; (a-2) Treated olives turning color in brine; (a-3) Treated black olives. Bitterness of treated olives is removed by alkaline treatment. (b) Natural olives: Green olives, turning-color olives, or black olives placed directly in brine in which they undergo complete or partial fermentation, preserved or not by the addition of acidifying agents: (b-1) Natural green olives; (b-2) Natural olives turning color; (b-3) Natural black olives. Bitterness of natural olives is mainly removed by dilution. (c) Dehydrated and/or shrivelled olives: Green olives, turning-color olives, or black olives that have undergone or not mild alkaline treatment, preserved in brine or partially dehydrated in dry salt and/or by heating or by any other technological process: (c-1) Dehydrated and/or shrivelled green olives; (c-2) Dehydrated and/or shrivelled turning color olives; (c-3) Dehydrated and/or shrivelled black olives. Black olives in dry salt are a principal example of this preparation. (d) Olives darkened by oxidation: Green olives or turningcolor olives preserved in brine, fermented or not,
SECTION | I The Plant and Production
darkened by oxidation in an alkaline medium, and preserved in hermetically sealed containers subjected to heat sterilization; these should be of a uniform black color. (d-1) Black olives. Other denominations for these olives are canned ripe olives, or Californian-style olives. (e) Specialities: Olives may be prepared by means distinct from, or additional to, those set forth above. Such specialities retain the name ‘olive’ as long as the fruit used complies with the general definitions laid down above.
1.4 MAJOR PROCESSING METHODS Although numerous processing methods are used around the world, only some of them are economically important from a global standpoint. At the same time, some local methods are highly valued. The most important of them are explained below.
1.4.1 Treated Green Olives Green olives are olives harvested during the ripening cycle when they have reached normal size, but prior to color change. Manual picking is still mainly used for table olive fruit harvesting, in spite of the high cost of this method (Figure 1.1). Table olives are machine-harvested in some cases, but – owing to the large proportion of bruised fruit – the catching frames (Figure 1.2) have to be handled with care, and the fruit may even have to be immersed in a dilute alkaline solution while still in the orchard (Vega et al., 2005). Freshly harvested, the olives are taken to the processing plant, if possible on the same day. There are two main ways of processing treated green olives: one with fermentation (Spanish-style) and the other without fermentation (Picholine and Castelvetrano styles).
1.4.1.1 Spanish-style Green Olives Also known as Sevillian-style green olives, this is one of the three main table olive products in the world. A flowchart with the steps of the process is presented in Figure 1.3. The olives, once carefully harvested and transported to the factory, are treated in a dilute lye solution (sodium hydroxide) with concentrations ranging from 2.0% to 3.5% (w/v, NaOH in water), depending on the variety and ripeness of the olives, the temperature, and the water quality. This alkaline treatment has several effects, such as the hydrolysis or elimination of the oleuropein, an increase of the fruit permeability, and other changes which aid subsequent fermentation (Rodríguez de la Borbolla, 1981; Brenes and de Castro, 1998). Treatment takes place in varying sizes of container, usually 10 tonnes capacity (Figure 1.4), in which the solution completely covers the fruits. The olives remain in this solution until the lye has penetrated two-thirds of the
9
CHAPTER | 1 Table Olives: Varieties and Variations
FIGURE 1.1 Manual harvesting. (Illustration courtesy of Dr. Rejano.) This figure shows the traditional procedure of picking olives. It involves handpicking the fruit and letting it fall into baskets which workers have hanging from their necks and suspended in front of them at waist level.
FIGURE 1.2 Mechanical harvesting. (Illustration courtesy of Dr. Rejano.) In the mechanical harvester shown, the picking up of the olives occurs through a mechanic method with the use of an umbrella-shaker directly from the tree without falling on the ground.
way through the flesh. The lye is then replaced by water, removing part of the NaOH. Lengthy or numerous washing steps adequately eliminate alkali, but also drag over soluble sugars needed for fermentation (Rodríguez de la Borbolla and Rejano, 1978). After washing, the olives are covered with brine and put into suitable containers. Nowadays, similarly to lye treatment, fermentation is carried out in large containers (Figure 1.5) with an inert inner covering (Rejano et al., 1977; Rodríguez de la Borbolla and Rejano, 1979, 1981). The brine triggers the release of the fruit cell juices, forming a culture medium suitable for fermentation. Brine concentrations are 9–10% NaCl to begin with, but rapidly drop to 5% owing to the high content of interchangeable
water in the olives. In this broth culture, a complex and variable microbiota grows. The origin of this microbiota is diverse: fruits, water, brine, pipes, containers, equipment in general, and atmosphere each play a role. Raw material control, cleaning and disinfection, and hygienic practices are the recommended methods to reduce contamination that can spoil the product. Initially, only alkali-tolerant organisms (Enterobacteriaceae and other Gram-negative bacteria, and Enterococci) multiply actively, but these disappear when the brine pH drops as a consequence of their own metabolism (de Castro et al., 2002). In fact, acid production by these early microorganisms encourages the growth of lactobacilli, the indispensable bacteria in all normal
10
SECTION | I The Plant and Production
SPANISH-STYLE GREEN OLIVES Harvesting Transport Grading (optional) Lye (NaOH) treatment Washing Brining Fermentation Preservation Sorting Grading
Whole
Pitting Stuffing (optional) Packing
FIGURE 1.3 Process steps for Spanish-style green olive production. This figure shows the main steps of Spanish-style green olive processing. It includes an alkaline treatment, a washing step to remove the excess alkali, a stage in brine, where the fruits undergo a spontaneous fermentation, carried out mainly by lactic acid bacteria.
Spanish-style green olive brines. Lactobacilli are hardly present in brines during the first days – they grow exponentially only when pH values are around 7.0 (Sánchez et al., 2001). Lactobacillus plantarum has always been considered the species mainly responsible for the fermentative process. However, taking into account the current status of
the genera, Lactobacillus pentosus should be considered equally or even more important. As a consequence of the homolactic fermentation by this species, lactic acid concentration increases, causing the pH to fall below 4.5. Actually, a pH below 4.0 is preferable, and can be attained when the fermentation proceeds properly. This is essential, as Enterobacteriaceae, spoiling Clostridia, and other problematic organisms are killed or fail to grow at these low pH values. Acid formation ceases when all the fermentable carbohydrates are consumed. Yeasts appear together with the lactobacilli. Fermentative yeasts do not cause deterioration, but oxidative yeasts forming films on the brine surface consume lactic acid and raise the pH, and may compromise the fermented product. Normal fermentation processes can be altered by the presence of undesirable microorganisms that can transmit poor sensorial properties to the olives, or impair their keeping properties. Gas pocket formation is usually caused by Enterobacteriaceae during the first stage of fermentation. Clostridia may cause butyric or putrid spoilage during the first days as well. In all cases, fermentation is controlled by ensuring the right pH and salt level. Obviously, it is also crucial to maintain good hygiene with containers and the rest of the equipment, and to use good-quality water. When properly fermented, olives can be kept for a long time. However, the spoilage known as zapateria may arise during preservation of the fermented product. Zapateria produces an unpleasant taste and odor, often coinciding with rising temperatures in the spring or early summer. Cyclohexanecarboxylic acid, in combination with other volatile acids, seems to be responsible for the unpleasant odor (Montaño et al., 1992, 1996). The bacteria responsible belong to the genera Clostridium and Propionibacterium (Kawatomari and Vaughn, 1956; Plastourgos and Vaughn, 1957). Again, the right combination of brine concentration (which must be increased before summer to more than 8%)
FIGURE 1.4 Alkaline treatment area in an olive processing plant. (Illustration courtesy of Dr. Rejano.) Alkaline treatment of olives in the Spanish-style processing usually takes place in large containers (10 tonnes capacity) made of fiber-glass.
11
CHAPTER | 1 Table Olives: Varieties and Variations
FIGURE 1.5 Fermentation area in an olive processing plant. (Illustration courtesy of Dr. Rejano.) This figure shows a fermentation yard in an olive processing plant with underground 10-tonnes capacity fermenters.
and pH (values below 4.2) helps to guarantee the correct keeping conditions. When olives are going to be marketed, the fruits are sorted and graded for the first or second time (Figure 1.3). The original brine is replaced by a new one, and the olives are packed in barrels, cans, or glass jars. Sometimes they are stoned (pitted), sliced, or stuffed with diverse stuffing materials. The levels of acidity and salt determine the product stability. Traditionally, this has been obtained by setting high values of free acidity and salt (0.5–0.7% and 5–7%, respectively) and low pH (⬍3.5). However, the progressive preference of consumers for milder conditions (low salt and acid percentages) has modified such conditions. In these cases the stability of the product is guaranteed by pasteurization (González-Pellissó et al., 1982; GonzálezPellissó and Rejano, 1984; Sánchez et al., 1991).
1.4.1.2 Picholine-style Green Olives Figure 1.6A shows the process diagram for this product, an example of NaOH-treated, but not fermented, fruits. Olives belonging to the Picholine Languedoc and Lucques varieties are prepared in this manner in the south of France, as are other varieties in Morocco (Picholine Marocaine) and Algeria. The bitterness of the olives is removed by treating them in lye (2.0–2.5% NaOH) in which they are left for 8–12 hours until the lye has penetrated three-quarters of the way through the flesh. They are rinsed several times over one or two days and then placed in 5–6% brine for two days. A second brine is used at 7%, and the acidity is corrected by adding citric acid (pH 4.5). After 8–10 days, the olives retain their bright green color, and are ready to be eaten. In the case of delayed shipment, it is necessary to store the olives.
PICHOLINE
CASTELVETRANO
Harvesting
Harvesting
Transport
Transport
Lye (NaOH) treatment
Grading
Washes (3 in 1 day)
Lye (NaOH) treatment
Washes (2 in 1 day)
Salt addition
Brining (2 or 3)
Fresh conservation
Cold storage (5-7°C)
Washing
Packing
Packing
(A)
(B)
FIGURE 1.6 Process steps for Picholine- (A), and Castelvetrano-style (B) green olive production. In this figure, the main steps of the Picholinestyle processing are shown in comparison with those of Castelvetranostyle processing. In both cases, fermentation does not constitute a key step.
This is straightforward as long as the temperature does not rise. The olives can be left in 8% brine until the spring, but then the concentration has to be raised to 10%. In large-scale facilities, they are kept in 3% brine in cold store, with the temperature maintained between 5°C and 7°C. Before shipping, the olives are washed repeatedly, and packed in suitable containers in brine at 5–6% (IOOC, 2004b).
1.4.1.3 Castelvetrano-style Green Olives A flowchart for this product is presented in Figure 1.6B. As with Picholine-style, these are olives where fermentation
12
SECTION | I The Plant and Production
does not constitute a key step. This is a production method used in Italy, almost exclusively in the Castelvetrano region and with the variety Nocellara del Belice, and the product is mainly consumed in central and southern Italy. Once the olives arrive at the processing plant, they are graded, since only fruits of more than 19 mm in diameter are used. The selected olives are put into plastic vessels and covered with 1.8–2.5% NaOH solution, depending on the fruit ripeness and size. These vessels have 220 L total capacity, and are filled with around 140 kg of fruits. One hour after the lye treatment begins, 5–8 kg salt are added to each container, and the olives are kept in this alkaline brine for 10–15 days. A mild washing step, carried out before marketing, does not totally eliminate the soda, whose taste is appreciated by the consumers of these olives (Salvo et al., 1995).
1.4.2 Natural Olives The designation ‘natural olives’ is applied to those fruits placed directly in brine, without any lye treatment to remove their bitterness. Although natural olives can be prepared from green, turning-color, or black fruits, the latter are more common. In fact, natural black olives in brine, Spanish-style green olives, and olives darkened by oxidation are the three main preparations globally (Garrido et al., 1995). A flow diagram for this method of olive processing is outlined in Figure 1.7. Natural black olives in brine are typical of the eastern Mediterranean and northern African countries. In Greece they are made with the Conservolea variety, which grades at around 200 fruits per kilogram, and in Turkey they are made with the Gemlik variety (IOOC, 2004b). To prepare natural black olives, the fruit is picked by hand when black ripe, but before the olives over-ripen or are shrivelled by frost. They have to be transported as quickly as possible to the processing plant, where they are washed, and immersed in 8–10% NaCl brine. NATURAL BLACK OLIVES Harvesting Transport Brining (Aerating) (Ferrous gluconate/lactate) Packing (vinegar-kalamata) FIGURE 1.7 Process steps for natural black olive production. Operations in brackets are optional. This figure shows the main steps of natural black olive processing. The traditional method of fermentation is carried out in anaerobic conditions. A diverse microbiota grows in brine, although yeasts are the microorganisms always present throughout the process.
Large-scale plants use big (10–20 ton) tanks, while smallscale processors continue to use wooden vats. At the start of fermentation, the tanks are tightly sealed because the olives must not be exposed to air. The brine stimulates the microbial activity for fermentation and reduces the bitterness of the oleuropein. Fermentation of these olives takes a long time because diffusion of soluble components through the epidermis, in fruits not treated with alkali, is slow. A diverse microbiota grows in these brines, although yeasts are the microorganisms always present throughout the process. Enterobacteriaceae can be found during the first 7–15 days, but they disappear as the brine characteristics do not support their growth. The presence of lactic acid bacteria depends on the salt concentration and the polyphenol content of the variety used. The traditional brining is carried out under anaerobic conditions. However, an aerobic method can be applied, using a central column in the fermenter through which air is bubbled. This system changes the ratio between fermentative and oxidative yeasts, and a final product with better quality is attained (García et al., 1985; Garrido et al., 1995). When the bitterness has been sufficiently weakened, the fruit can be marketed. The color fades during the process, but is corrected by aerating the olives for two or three days; sometimes they are treated with 0.1% ferrous gluconate or lactate to make them a deeper black. Lastly, the olives are selected and packed in barrels or internally varnished cans, which are filled with 8% fresh brine. They are popular with consumers because of their slightly bitter taste and aroma. Natural black olives can also be packed in vinegar (25% of brine volume) and may even be heat-processed; a few grams of oil are then added to each can to form a surface layer. The Kalamata variety is prepared in this way (Fernández-Díez et al., 1985).
1.4.3 Black Olives in Dry Salt Also of Greek origin, dehydrated black olives are encountering a great consumer acceptance in many producing areas. They are prepared using overripe olives. They are vigorously washed and placed in baskets with alternating layers of dry salt equivalent to 15% of the weight of the olives. The final product is not bitter, but salty, and looks like a raisin; it is for local consumption. The flowchart for these olives is outlined in Figure 1.8.
1.4.4 Olives Darkened by Oxidation These olives are also known as Californian-style black olives, ripe or semi-ripe olives, or simply black olives (USDA, 1983; IOOC, 2004b). The production flowchart for these olives is outlined in Figure 1.9. Fruits are harvested when their color is starting to change, before full maturity. Once in the production plant, olives are selected and may
13
CHAPTER | 1 Table Olives: Varieties and Variations
BLACK OLIVES IN DRY SALT Harvesting Transport Washing Dry salt layers Local consumption FIGURE 1.8 Process steps for black olives in dry salt production. These olives are obtained from fruit harvested when fully ripe. Olives are vigorously washed and placed in baskets with alternating layers of dry salt.
OLIVES DARKENED BY OXIDATION Harvesting Transport Previous handlings Washing (Preservation in brine, fermentation) Lye treatment and air oxidation Washing (alkali neutralization) Brining (pasteurization) Sorting and grading (Pitting, slicing, etc.) Packing Sterilization FIGURE 1.9 Process steps for olives darkened by oxidation. Operations in brackets are optional. These olives are obtained from fruit which, when not fully ripe, has been darkened by oxidation and whose bitterness has been removed by lye treatment. They are packed in brine and preserved by heat sterilization.
be directly processed or – more commonly – preserved before oxidation. Preservation is usually in brine, and a fermentative process comparable to that of natural black olives takes place. Nevertheless, this preservation can also be done in acidified water (de Castro et al., 2007). With this method, the discharge of sodium chloride into wastewater streams is reduced strikingly. From a microbiological point of view, the addition of acetic acid results in a pH incompatible with Enterobacteriaceae growth, might favor lactic acid bacteria in some instances, and – in any case – yeasts continue being the most important microorganisms
in these solutions. As this preservation step is not necessary, a complete fermentation is not required. The essential operation is the oxidation. In general, fruits are treated successively with sodium hydroxide solutions for varying periods of time to achieve a progressive penetration of the lye into the flesh. After each alkaline treatment, the olives are put into water and oxidized by injecting air under pressure into the water. This oxidation of the phenolic compounds permits a complete blackening of the fruit skin and a uniform coloration of the flesh. The promoters of the polymerization involved have been identified as hydroxytyrosol (3,4-dihydroxyphenyl acetic acid) and caffeic acid, the decrease of which in the flesh is strongly correlated with fruit darkening (BrenesBalbuena et al., 1992). Olives are darker and oxidation rates higher with higher pH values (García et al., 1992; Garrido et al., 1995). The number of lye treatments is generally between 3 and 5, although some processors apply two or even only one. Penetration into the fruit is controlled so that the sodium hydroxide of the first treatment passes merely through the skin. Subsequent treatments are chosen so that they penetrate increasingly deeper into the pulp. The final lye treatment must reach the pit. The concentration of sodium hydroxide in the lye solution depends on the ripeness of the fruit, its variety, the environmental temperature, and the desired penetration speed. It varies between 1% and 2% NaOH (w/v). The highest concentration is usually used for the first treatment. The blackened olives are washed several times with water to remove most of the sodium hydroxide and lower the pH in the flesh to around 8.0 units. Generally, 0.1% (w/v) of ferrous gluconate or lactate is added to the last wash to stabilize the color. The final canned product has sensorial properties very different from those of fermented fruits obtained by other processes. The pH values are between 5.8 and 7.9, and the NaCl content is between 1% and 3%. Due to these chemical characteristics, which do not guarantee safety, olives darkened by oxidation have to be sterilized to prevent any possibility of foodborne pathogen growth.
1.5 CHARACTERISTICS OF FINAL PRODUCTS The numerous procedures to prepare table olives imply a broad range of characteristics in the different final products. However, to be marketed they have to comply with the limits displayed in Table 1.3 (IOOC, 2004a). The limits vary depending on both the preparation system and the way preservation is guaranteed. Sodium chloride concentration, pH, and titratable acidity (as lactic acid) are the parameters to monitor. They can be analyzed in the brine or from the fruit juice, but always once the osmotic balance between olives and packing brine has been attained. Olives darkened by oxidation have no requirement regarding the cited
14
SECTION | I The Plant and Production
TABLE 1.3 Physicochemical characteristics of the packing brine or of the juice after osmotic balance (IOOC, 2004a). This table shows the limits for the physicochemical parameters established in the ‘Trade Standard Applying to Table Olives’ of the International Olive Council according to processing type and preservation method. Processing
Minimum sodium chloride content (%)
Maximum pH limit
Minimum lactic acidity (% lactic acid)
SCC, MAT
PR, R
P, S
SCC, MAT
PR, R
P, S
SCC, MAT
PR, R
P, S
Treated olives
5
4
GMP
4.0
4.0
4.3
0.5
0.4
GMP
Natural olives
6
6
GMP
4.3
4.3
4.3
0.3
0.3
GMP
Dehydrated and/or shrivelled olives
10
10
GMP
GMP
GMP
GMP
GMP
GMP
GMP
Olives darkened by oxidation
GMP
GMP
GMP
GMP
GMP
GMP
GMP
GMP
GMP
SCC: Specific chemical characteristics; MAT: Modified atmosphere; PR: Addition of preservatives; R: Refrigeration; P: Pasteurization; S: Sterilization; GMP: Good manufacturing practice.
parameters, since they have to be sterilized to guarantee safety. Dehydrated olives must contain a minimum of 10% NaCl unless they are thermally treated. Natural or NaOHtreated olives vary in their requirements, according to the preservation method. Apart from brine (water and food-grade salt) and olives, other possible ingredients are vinegar, olive oil, sugars, spices or aromatic herbs or natural extracts, and authorized additives. Furthermore, any single or combination of edible material used as an accompaniment or stuffing is also allowed. Typical examples (among many others) are pimiento, capers, and onions.
SUMMARY POINTS ●
●
●
●
●
Fruits from olive trees may be prepared following many different methods, usually related to the various producing areas. The different methods may, or may not, include a fermentation stage, and most of them are aimed at eliminating the natural bitterness of the fresh fruit. Spanish-style green olives are the only ones that include fermentation by lactic acid bacteria as a necessary step. Olives not treated with alkali usually support a fermentation by yeasts while maintained in brine. Keeping quality may be guaranteed by physicochemical characteristics or thermal treatment.
REFERENCES Brenes, M., de Castro, A., 1998. Transformation of oleuropein and its hydrolysis products during Spanish-style green olive processing. J. Sci. Food Agric. 77, 353–358.
Brenes-Balbuena, M., García-García, P., Garrido-Fernández, A., 1992. Phenolic compounds related to the black color formed during the processing of ripe olives. J. Agric. Food Chem. 40, 1192–1196. Columela, L.J.M. (42). In “De re rustica”. Sociedad Nestlé, Santander (Spain), 1979. de Castro, A., Montaño, A., Casado, F.-J., Sánchez, A.-H., Rejano, L., 2002. Utilization of Enterococcus casseliflavus and Lactobacillus pentosus as starter cultures for Spanish-style green olive fermentation. Food Microbiol. 19, 637–644. de Castro, A., García, P., Romero, C., Brenes, M., Garrido, A., 2007. Industrial implementation of black ripe olive storage under acid conditions. J. Food Eng. 80, 1206–1212. Fernández-Díez, M.J., de Castro, R., Garrido, A., González-Cancho, F., González-Pellissó, F., Vega, M.N., Moreno, A.H., Mosquera, I.M., Rejano, L., Durán, M.C., Roldán, F.S., García, P., de Castro, A., 1985. Biotecnología de la Aceituna de Mesa. Consejo Superior de Investigaciones Científicas, Madrid. García, P., Durán, M.C., Garrido, A., 1985. Fermentación aeróbica de aceitunas maduras en salmuera. Grasas y Aceites 36, 14–20. García, P., Brenes, M., Vattan, T., Garrido, A., 1992. Kinetic study at different pH values of the oxidation processes to produce ripe olives. J. Sci. Food Agric. 60, 327–331. Garrido, A., García, P., Brenes, M., 1995. Olive fermentations. In: Rehm, H.-J., Reed, G. (eds), Biotechnology, Vol. 9. VCH, Weinheim, pp. 593–627. González-Pellissó, F., Rejano, L., 1984. La pasterización de aceitunas estilo sevillano II. Grasas y Aceites 35, 235–239. González-Pellissó, F., Rejano, L., González-Cancho, F., 1982. La pasterización de aceitunas estilo sevillano I. Grasas y Aceites 33, 201–207. IOOC, 2000. Catálogo Mundial de Variedades de Olivo. IOOC, Madrid. IOOC, 2004a. International Olive Council. Trade Standard Applying to Table Olives. Document COI/OT/NC no. 1. Madrid. IOOC, 2004b. International Olive Council. Table Olives. Madrid. IOOC, 2007. World table olive figures. ⬍http://www.internationaloliveoil. org/⬎ Kawatomari, T., Vaughn, R.H., 1956. Species of Clostridium associated with zapatera spoilage of olives. Food Res. 21, 481–490. Montaño, A., de Castro, A., Rejano, L., Sánchez, A.-H., 1992. Analysis of zapatera olives by gas and high-performance liquid chromatography. J. Chromatogr. 594, 259–267.
CHAPTER | 1 Table Olives: Varieties and Variations
Montaño, A., de Castro, A., Rejano, L., Brenes, M., 1996. 4-Hydroxycyclohexanecarboxylic acid as a substrate for cyclohexanecarboxylic acid production during the “zapatera” spoilage of Spanish-style green table olives. J. Food Protect. 59, 657–662. Plastourgos, S., Vaughn, R.H., 1957. Species of Propionibacterium associated with zapatera spoilage of olives. Appl. Microbiol. 5, 267–271. Rejano, L., González-Cancho, F., Rodríguez de la Borbolla, J.-M., 1977. Estudio sobre el aderezo de aceitunas verdes. XXIV Nuevos ensayos sobre el control de la fermentación. Grasas y Aceites 28, 255–265. Rodríguez de la Borbolla, J.-M., 1981. Sobre la preparación de la aceituna estilo sevillano. El tratamiento con lejía. Grasas y Aceites 32, 181–189. Rodríguez de la Borbolla, J.-M., Rejano, L., 1978. Sobre la preparación de la aceituna estilo sevillano. El lavado de los frutos tratados con lejía. Grasas y Aceites 29, 281–291. Rodríguez de la Borbolla, J.-M., Rejano, L., 1979. Sobre la preparación de la aceituna estilo sevillano. La fermentación I. Grasas y Aceites 30, 175–185.
15
Rodríguez de la Borbolla, J.-M., Rejano, L., 1981. Sobre la preparación de la aceituna estilo sevillano. La fermentación II. Grasas y Aceites 32, 103–113. Sánchez, A.-H., Rejano, L., Montaño, A., 1991. Kinetics of the destruction by heat of colour and textura of pickled green olives. J. Sci. Food Agric. 54, 379–385. Sánchez, A.-H., Rejano, L., Montaño, A., de Castro, A., 2001. Utilization at high pH of starter cultures of lactobacilli for Spanish-style green olive fermentation. Int. J. Food Microbiol. 67, 115–122. Salvo, F., Cappello, A., Giacalone, L., 1995. In: L’Olivicoltura nella Valle del Belice. Istituto Nazionale di Economia Agraria. Ministero delle Risorse Agricole, Alimentari e Forestali, Italy. USDA, 1983. United States Department of Agriculture. United States Standards for Grades of Canned Ripe Olives. Washington, D.C. (USA). Vega, V., Rejano, L., Guzmán, J.-P., Sánchez, A.-H., Díaz, J.-M., 2005. Recolección mecanizada de la aceituna de verdeo. Agricultura Revista Agropecuaria LXXIV, 376–379.
This page intentionally left blank
Chapter 2
Olive Genomics Corrado Fogher1, Matteo Busconi1, Luca Sebastiani2 and Tania Bracci2 1 2
Istituto di Botanica e Genetica Vegetale, Università Cattolica del Sacro Cuore, Piacenza, Italy Scuola Superiore Sant’Anna, Pisa, Italy
2.1 INTRODUCTION
Domestication of the olive primarily occurred through vegetative multiplication of selected individuals with allele combinations giving favorable traits, and probably began in the near-east of the Mediterranean basin between 5500 and 5700 years ago (Zohary and Hopf, 1994) although olive use was also reported in Spain during the Neolithic period (Terral and Arnold-Simard, 1996). Due to its allogamous reproduction, a high level of heterozygosity has been maintained in olive, and since ancient times, humans have largely contributed to its dissemination in the Mediterranean region and more recently into new geographic areas (i.e., Australia, North and South America). For this reason, the genetic structure of the olive probably results from several factors related to the occurrence of refugia zones (where olive plants survived during Quaternary glaciations), the bio-geographic conditions of the Mediterranean Basin and human influence (Besnard et al., 2002). There are four main hypotheses on the origin of the Mediterranean olive tree: (1) migration, establishment and evolution from populations of subsp. cuspidata in Tropical Africa or southern Asia; (2) isolation of populations of subsp. laperrinei in the Mediterranean after Quaternary glaciations; (3) hybridization between two taxa (subsp. europaea and cuspidata) co-existing in eastern Mediterranean mountains; (4) divergence from an O. europaea lineage in the Mediterranean. Molecular data have given new insights about the history and domestication of the olive, suggesting that the cultivated olive most likely originated from a pre-Quaternary Mediterranean ancestor, with no evidence for a recent hybrid origin (Besnard et al., 2006). The actual number of cultivars in the world is estimated at around 1500, and more than 3000 cultivar synonyms have been recorded (Bartolini, 2008). Italy, Spain, France, Tunisia and Greece have the largest numbers with estimated values of 610, 280, 100, 70 and 40, respectively. The high levels of genetic and morphological diversity present today may be due to the continuous process of olive domestication through local hybridization events of cultivated trees with multiple wild Mediterranean
Genomics is the study of genes and their function and aims to understand the structure of the genome, including the mapping genes and sequencing the DNA. Genomics includes functional genomics (the characterization of genes and their mRNA and protein products), structural genomics (the architectural features of genes and chromosomes), and comparative genomics (the evolutionary relationships between the genes and proteins of different species).
2.2 ORIGIN AND SYSTEMATICS OF THE OLIVE Olive (Olea europaea L.) is one of the oldest agricultural tree crops and has been an important source of oil since ancient times. Olive belongs to the family Oleaceae, which includes about 30 genera and 600 species (Cronquist, 1981). The genus Olea L. consists of more than 30 species, which are distributed in Europe, Asia, Oceania and Africa, with only Olea europaea L. being cultivated. Two co-existing forms have been described for this species, the wild olive or oleaster (Olea europaea subsp. europaea var. sylvestris) and the cultivated olive (Olea europaea subsp. europaea var. europaea). Relatives of the Mediterranean olive tree are clustered in five subspecies: (a) laperrinei, present in Saharan massifs; (b) cuspidata, present from South Africa to southern Egypt and from Arabia to northern India and southwest China; (c) guanchica present in the Canary Islands; (d) maroccana present in south-western Morocco; (e) cerasiformis present in Madeira (Green, 2002). The cultivated species is a perennial evergreen tree bearing hermaphroditic flowers that show various degrees of self-incompatibility. Asexual vegetative reproduction has been used for several thousand years in cultivar propagation. Olive breeding approaches have been undertaken rarely, due to the long juvenile phase (more than 10 years), and the lack of phylogenetic relationships among Olea species. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
17
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
18
SECTION | I The Plant and Production
FIGURE 2.1 Olive cultivar identification by microsatellite analysis. Electropherograms showing the single sequence repeats (SSR) or microsatellite profiles generated in two different olive cultivars (upper panel and lower panel) by using five different microsatellite markers. The five markers are marked with different letters: (1) a, (2) b, (3) c, (4) d, and (5) e. The differences, detected as fragments with different sizes, are visible. The arrows point to the peaks scored for four out of five markers (a, c, d, and e). The presence of multiple scored peaks for each SSR with the exception of the last one (e) reflects the high heterozygosis found in olive. The first SSR (a) shows the presence of at least three scored peaks within a single cultivar (Taggiasca, lower panel); one possible explanation could be the presence of regions of partial polyploidy also within Olea europaea subsp. europaea.
genomes (Contento et al., 2002). This hypothesis seems to be compatible with the molecular similarity of wild and cultivated olives within the same country (Besnard et al., 2006). Moreover, Natali et al. (2007) establish that the large olive variability could be related to mutations within retroelements and/or to insertions/deletions of retro-elements. Olive has 2n ⫽ 2x ⫽ 46 chromosomes (Breviglieri and Battaglia, 1954). The nuclear DNA content of olive cultivars in Leccino and Frantoio was determined using cytometric methods by Rugini et al. (1996), resulting in 2.26 and 2.20 pg of DNA per haploid nucleus, respectively. More recently, Loureiro et al. (2007) estimated the genome size of six Portuguese cultivars of olive (Olea europaea subsp. europaea var. europaea) and wild olive (Olea europaea subsp. europaea var. sylvestris) by flow cytometry. The nuclear DNA content of Portuguese cultivars ranged between 2.90 ⫾ 0.020 pg/2C and 3.07 ⫾ 0.018 pg/2C, while the nuclear DNA of wild olive was estimated as 3.19 ⫾ 0.047 pg/2C DNA. Based on the flow cytometry data of Louriero et al. (2007) the olive genome size can be estimated as between 2.90 to 3.07 billion base pairs, given that 1 pg of DNA corresponds to approximately 109 base pairs (Bennett and Smith, 1976).
Multiple polyploidy in olive tree relatives was recently demonstrated by Besnard et al. (2008) using flow cytometry and highly variable nuclear microsatellites in six olive subspecies. This study showed that four subspecies appear to be diploids, while subsp. cerasiformis is tetraploid and subsp. maroccana hexaploid. Concerning the subspp. europea and cuspidata, SSR data suggest the possibility of a partial polyploidy (Figure 2.1; Rallo et al., 2003) although Besnard et al. (2008) did not confirm this hypothesis.
2.3 MOLECULAR MARKERS IN OLIVE GENOME ANALYSIS Studies of genetic diversity and variability have traditionally been carried out using agronomic, morphological, and biochemical characters. Although these markers provide a very useful tool for cultivar identification, they have limitations because of the small number of polymorphisms considered and because of the influence of the environment and cultivation. In contrast, DNA-based markers are independent from agronomic practice and the environment, they are numerous, highly polymorphic, and widely distributed across the
19
CHAPTER | 2 Olive Genomics
FIGURE 2.2 Olive cultivar identification by amplified fragment length polymorphism (AFLP) analysis. Electropherograms show the AFLP profiles generated in two different olive cultivars by a single selective primer combination. Each fluorescence peak corresponds to a DNA fragment with a definite size in base pairs. The difference between the two cultivars, evaluated as the presence/absence of a peak, is visible and constitutes the polymorphic signal that may be used for subsequent analyses.
genome both in coding and non-coding regions. Molecular markers derive from DNA polymorphisms present at any site, bearing DNA sequence variations between individuals within a family or population. Sequence data from non-translated regions are widely used for several applications, mainly for phylogenetic analysis, in order to reconstruct the evolutionary history and the origin of the olive and to define the relationships between different cultivars. Molecular markers are also used to obtain genetic maps, which are important tools for several subsequent applications such as marker-assisted selection in olive breeding, and physical maps that actually have a key role for sequence assembly in whole genome sequencing projects. Nowadays the most commonly used types of markers in plants are: AFLPs (amplified fragment length polymorphisms), SCARs (sequence characterized amplified regions), SSRs (simple sequence repeats or microsatellites), SNPs (single nucleotide polymorphisms), sequence variation in nuclear ribosomal DNA, and cytoplasmic DNA variations. All these classes of markers have in common that they are based on PCR (polymerase chain reaction) analysis. Following a brief description of the main typologies of markers, an insight into some of the different applications of olive genetic markers will be given.
analysis of plant DNA, it has successfully been applied to several other organisms belonging to different kingdoms such as bacteria and animals. Genomic DNA is doubledigested with two restriction enzymes and, to reduce the number of fragments generated to a level that can be resolved on a gel or with capillary chromatography, these are ligated with adaptors (short double-stranded oligonucleotides with known sequence) and amplified with arbitrarily selected primers in one or two steps. The amplification products are resolved by length using polyacrylamide gels or automated DNA genetic analyzers (Figure 2.2). The resulting genome fragments are usually considered as dominant markers and the polymorphic signals are mainly seen by the presence/ absence of a given band (acrylamide gel) or peak (automated DNA sequencing systems). Sometimes co-dominant alleles may be present but they are rarely visible without a segregation analysis of family or population data. Since the first application of AFLP analysis in olive for cultivar differentiation (Angiolillo et al., 1999), several authors have used the same technique, for example, to establish genetic relationships among different cultivars, to construct a linkage map of the olive, to analyze the genetic structure of wild and cultivated olive (Baldoni et al., 2006), and to identify cultivars used in olive oil production (Busconi et al., 2003).
2.3.1 AFLPs
2.3.2 SCARs
The AFLP technique is a powerful method for revealing genetic diversity. Developed by Vos et al. (1995) for the
The conversion of AFLP and other molecular markers to SCAR, by sequencing specific DNA bands and defining
20
SECTION | I The Plant and Production
FIGURE 2.3 Single nucleotide polymorphism (SNP) in two olive cultivars. Electropherograms show the sequence of the same DNA fragment in two different cultivars. The polymorphic site at position 99 (SNP) is emphasized.
the delimiting primers (Busconi et al., 2006), significantly improves the reproducibility and the reliability of PCR assays and therefore their utility for several applications including breeding programs, marker-assisted selection (MAS) for quantitative trait loci (QTLs), mapping and food traceability. SCAR markers are delimited by the sequence of the primers and they can be both dominant (presence/absence of a given band) and co-dominant (bands with different sizes in different samples), and usually they are considered single locus, while AFLP is a multi-locus method. The analysis of these markers is simple and relatively cheap, consisting of a PCR reaction performed at the annealing temperature defined by the sequence of the primers.
2.3.3 SSRs Microsatellites consist of short basal motifs (with length generally between 1 to 6 bp) repeated tandemly several times (e.g. the bases AT repeated 12 times). Variation in the number of basal motifs causes the polymorphism of these markers. They are numerous and present in every eukaryotic genome and they are very useful for several applications
such as establishing genetic relationships, constructing linkage maps, or confirming a pedigree. Nowadays, microsatellites are the markers of choice for genetic studies in olive because of their high polymorphism level. Many authors have reported on the development of SSRs in olive and several different sets of markers are actually available for DNA analysis (Sabino Gil et al., 2006). As for SCAR, the primers designed on the repeat flanking regions delimit SSRs. They are co-dominant markers and the analysis consists of a PCR reaction. The amplification products are resolved by length either using gel electrophoresis or automated high-throughput genotyping DNA sequencing systems (Figure 2.1).
2.3.4 SNPs Compared to the markers described above, SNPs are the most recent and are an outcome of sequencing analysis; their advantages were discovered only recently. They consist of those single nucleotide sequence variations present between individuals within a family or population (Figure 2.3). These markers are widely used in human genetics and in the major crop species, but in olive the identification of
21
CHAPTER | 2 Olive Genomics
SNPs was documented only recently in a few studies (Reale et al., 2006). Compared with other genetic markers, SNPs are more abundant and can occur in the coding region of genomes, leading to occasional amino acid changes in the encoded polypeptide. They are, in most cases, bi-allelic, co-dominant, and amenable to high-throughput genotyping and automation.
2.3.5 Ribosomal DNA The internal transcribed spacer (ITS1 and ITS2) regions of the nuclear rRNA genes, 18S-5.8S-26S, are one of the most popular choices for phylogenetic analysis. Indeed, as reviewed in Alvarez and Wendel (2003), the use of these sequences should be carefully evaluated because, beside a series of characteristics that are potentially useful for this kind of application (biparental inheritance, universality, simplicity, high intergenomic variability, low functional constraint), there are also aspects that could falsify the phylogenetic analyses, in particular the presence of multiple rDNA arrays in eukaryotic genomes, the concerted evolution of repetitive regions, and the presence of pseudogenes. Base changes and/or indel (insertion/deletion of nucleotides) determine the sequence variations in these regions, detection of which can be carried out: by direct sequencing, by DNA restriction of the fragments of interest obtained via PCR, and by designing selective primers based on the sequence variations.
2.3.6 Cytoplasmic DNA The inheritance of mitochondrial and chloroplast DNAs does not follow Mendelian laws, but proceeds by vegetative segregation and maternal inheritance. In the last ten years, several studies dealing with cytoplasmic DNA variation (both chloroplast and mitochondrial) in olive have been carried out. The sequence polymorphisms in organelle DNA can be detected in several ways, for example by using SSR, SCAR and SNP markers specific to cytoplasmic DNA, RFLP of a fragment amplified via PCR, and direct sequencing of a region of interest. Besnard et al. (2002) used chloroplast markers to analyze the structure of the olive complex. The authors studied the geographic distribution of different chlorotypes of the O. europaea complex.
2.4 MOLECULAR MARKER APPLICATIONS 2.4.1 Characterization of Olea europaea Germplasm Knowledge about genetic variability in the olive and the discovery of the genetic relationships between cultivars and
wild genotypes are important tasks for germplasm preservation and olive improvement. Unlike other crops, olive has conserved a wide genetic patrimony, across more than 1500 cultivars (Bartolini, 2008) and in some wild populations around the Mediterranean basin. Efforts have initially been made to review the genus Olea and to colocate cultivated and wild olive trees in the Olea europaea complex. This analysis resolved all six subspecies within the complex as monophyletic groups, although some uncertainty remains for subspecies cuspidata, europaea and guanchica (Rubio de Casas et al., 2006). Wild olive populations are a potential source of genetic traits for the improvement of cultivated forms and their genetic variability has been studied by multiple approaches. The results show that the population structure of wild olives from the north-western Mediterranean partially reflects the evolutionary history of these populations, although hybridization between wild olive and cultivated forms can occur in areas of close contact. The large number of cultivars present in all olive-growing countries raises several problems for germplasm management and preservation. Evaluation and characterization of olive genetic resources is therefore crucial, since identification of cultivars is complicated by the large number of synonyms and homonyms and by the presence of clones.
2.4.2 Genetic Improvement The length of the juvenile period and the wide spacing necessary for the management of progeny in selection programs makes the genetic improvement of the olive very difficult. These are the main reasons why, up to now, few results have been obtained in the development of new olive varieties. Molecular markers provide a useful tool to overcome these drawbacks, introducing the possibility of precociously selecting the cross-progeny and consequently reducing the time and the cost for developing new genotypes. The first critical step in a breeding program is to identify with certainty the parental origins of the progeny. Among the molecular markers available in olive, SSRs are the most suitable for this purpose because they are co-dominant and highly polymorphic. They are thus useful for tracing the genetic contribution of alleles from the parents to the offspring. Another difficulty in progeny selection is that the characters of agronomic interest are typically expressed in the mature plants, these being related to the fruit and the size of the tree. It is thus necessary to wait for the full development of the plants before selecting those showing the most interesting traits. Molecular markers offer the possibility to develop early selection strategies since they can be associated with specific agronomic traits and used to select plantlets having such characteristics from the first stages of development, when the traits are not yet expressed. However, this technique, named marker-assisted selection (MAS), needs some knowledge
22
SECTION | I The Plant and Production
50
100
1600 1200 800 400 0 Taggiasca DNA da foglie - Eco-ACA_Mse-CAG
Taggiasca DNA da olio - Eco-ACA_Mse-CAG
Moraiolo DNA da foglie - Eco-ACA_Mse-CAG
Taggiasca DNA da olio - Eco-ACA_Mse-CAG
1600 1200 800 400 0 FIGURE 2.4 Amplified fragment length polymorphism (AFLP) analysis of DNA from leaves and oil of the same or different olive cultivars. Capillary electrophoresis electropherogram shows the AFLP profile generated by the following primer combinations: EcoRI ⫹ ACA and MseI ⫹ CAG. In the upper panel, the profiles of DNA of the cultivar Taggiasca, isolated from leaves (blue) and oil (red) are superimposed. The perfect coincidence of the two profiles for all the fragments present can be seen. In the lower panel the AFLP profile of the leaf DNA of the cultivar Moraiolo (blue) is superimposed on the AFLP profile of the oil DNA of the cultivar Taggiasca (red) and the differentiating bands can be seen (Busconi et al., 2003).
about the co-segregation of molecular markers and genetic characters in the progeny. Several efforts to build a map of association in olive have recently produced results.
2.4.3 Olive Oil Traceability Food forensics refers to the possibility of using DNA analysis in the food industry to identify the raw materials used to make processed food, and thus to identify mislabeling and fraud. DNA is the best choice for traceability purposes since it is not dependent on environmental and processing conditions, unlike other biomolecules such as proteins or metabolites. Olive oil is commercialized at different grades of quality depending on the raw material and its origin, the production technology and knowledge of the production process, which give products with significantly different nutritional and health values. As for many high economic value products, PDO (Protected Designation of Origin) and PGI (Protected Geographical Indication) olive oils can be subject to fraudulent practices such as admixture with oils from other species (mainly sunflower and hazelnut), or by the use of lower-quality olives. As a result, the availability of procedures to recover DNA of the original raw materials from processed food is important. After a preliminary report that olive DNA was present in processed oils (Cresti et al., 1997) several authors developed purification methods (Busconi et al., 2003; Pasqualone et al., 2007), and many commercial DNA extraction kits have protocols adapted for this purpose. By using different classes of markers (SCARs and AFLPs), Busconi et al. (2003) were able to demonstrate that the DNA purified from oil was of organelle and nuclear
origin, and that the DNA purified from fresh monovarietal oil contains molecular marker profiles corresponding with the profiles of DNA purified from the leaves of the same cultivar (Figure 2.4). However, for traceability purposes, we have to consider that the length of oil storage has a great impact on the quality of the recovered DNA. Pafundo et al. (2005) used AFLPs to trace the plants’ contribution in monovarietal olive oils and they suggested that DNA extraction is the most critical step affecting reliability in AFLP analyses. In summary, single locus markers like SCARs and SSRs seem to be more useful in olive oil traceability than AFLPs. This is probably because AFLPs are strongly influenced by the quality and quantity of the starting DNA used in the analysis. It is important to add a cautionary note on the use of DNA markers for provenance testing, since the nonconcordance between the genetic profiles of DNA from olive oil and from fruit could be due to the contribution of pollen donors in DNA extracted from the paste obtained by crushing the whole fruits.
2.5 FUNCTIONAL GENOMICS Functional genomics is a field of molecular biology that attempts to make use of the vast wealth of data produced by genomics projects to describe gene and protein function. The understanding of protein function in olive provides insights that may have practical implications for human health. For example, pollinosis caused by olive pollen consists of a respiratory allergy that is an important health problem in several geographical areas worldwide (Mediterranean basin,
23
CHAPTER | 2 Olive Genomics
Genus Olea 1037 core nucleotide 24 EST 2 GSS
Olea europaea Olea spp. 1029 (99.23%) core nucleotide 8 (0.77%) core nucleotide
24 EST 2 GSS
Olea europaea Olea europaea ssp. europaea 931 (90.48%) core nucleotide 24 EST
ssp. cuspidata 63 (6.12%) core nucleotide ssp. laperrinei 12 (1.17%) core nucleotide ssp. cerasiformis 11 (1.07%) core nucleotide ssp. guanchica 6 (0.58%) core nucleotide ssp. maroccana 6 (0.58%) core nucleotide
2 GSS FIGURE 2.5 Olive genomics information present on database. Typology of sequence records recovered from database screening. The largest amount of data belongs to Olea europaea ssp. europaea, with just a few records belonging to different Olea species or subspecies within the Olea europaea complex. EST refers to Expressed Sequence Tags data, while GSS refers to Genome Survey Sequences. (http://www.ncbi.nlm.nih.gov/dbGSS/index.html).
North and South America, and Australia), affecting more than 30% of Mediterranean populations during the pollination season. Olive oil is a functional food with exceptional nutritional properties due to the balanced fatty acid composition and to the presence of natural antioxidants and vitamins as minor components. Besides their important physiological role for the plant, these components are crucial for human health because they cannot be synthesized in the body and therefore must be provided by the diet (Hernandez et al., 2005). Given the importance of these two aspects it is not surprising that a large number of sequenced genes recoverable from sequence databases are related to allergens and to genes involved in fatty acid biosynthesis, modification, and storage. To date, ten different allergens (named Ole e 1 to Ole e 10) have been found in olive pollen (Hamman-Khalifa et al., 2008), and for almost all of these genes, with the exception of Ole e 7 and Ole e 8, the nucleotide sequences are available. Other studies still in progress concern the high levels of fitness of olive in different environmental conditions, and response to different biotic and abiotic stresses.
2.6 CONCLUSIONS GenBank data show that Zea mays and Oryza sativa are the most well-studied plant species, having 3.6 billion and 1.5 billion bases of sequence in the database respectively (Benson et al., 2008). The situation is completely different for the genus Olea. In fact only a few sequences have been submitted in the last few years and only 1037 core nucleotide, 24 EST (expressed sequence tag), and two GSS (genome survey sequence) sequences were actually recovered from Entrez, the NCBI’s retrieval system, which integrates the main DNA sequence databases (Figure 2.5 summarizes this information). The core nucleotide records belonging to O. europaea can be further divided into: sequences from genomic non-translated regions (608 sequences, approximately 59% of total) and translated regions (421, approximately 41%). This situation will probably change in the near future as a consequence of the development of the latest generation of sequencing platforms such as the 454 (Life Sciences – Roche), Illumina Genome Analyzer (Illumina), and SOLID (Applied
24
SECTION | I The Plant and Production
Biosystems), which enable the large-scale sequencing of billions of bases in a few days. These technologies make it possible to quickly generate new sequence data for any given organism and can save time and money compared to the present system if the aim is whole genome sequencing. Moreover, more data will be available soon since several groups are working on the identification of ESTs in olive that are related to fruit development and biotic and abiotic stress responses.
SUMMARY POINTS ●
●
●
●
●
●
Genomics is a field of molecular biology that makes use of data from genomics projects and in olive provides insights that may have practical implications for human health. Domestication of the olive occurred through vegetative multiplication of selected individuals. The actual number of cultivars in the world is estimated at around 1500. The estimation of genetic variability in the olive and the discovery of the genetic relationships between cultivars are important tasks for germplasm preservation. Wild olive populations are a potential source of genetic traits. Olive forensics refers to the possibility of using DNA analysis to identify fraud.
REFERENCES Angiolillo, A., Mencuccini, M., Baldoni, L., 1999. Olive genetic diversity assessed using amplified fragment length polymorphisms. Theor. Appl. Genet. 98, 411–421. Álvarez, I., Wendel, J.F., 2003. Ribosomal ITS sequences and plant phylogenetic inference. Mol. Phylogenet. Evol. 29, 417–434. Baldoni, L., Tosti, N., Ricciolini, C., Belaj, A., Arcioni, S., Pannelli, G., Germana, M.A., Mulas, M., Porceddu, A., 2006. Genetic structure of wild and cultivated olives in the central Mediterranean basin. Ann. Botany 98, 935–942. Bartolini, G., 2008. Olea databases. http://www.oleadb.it/ Bennett, M.D., Smith, J.B., 1976. Nuclear DNA amounts in angiosperms. Philos. Trans. R. Soc. B274, 228–274. Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., Wheeler, D.L., 2008. GenBank. Nucleic Acids Res. 36 Database issue D25–D30. Besnard, G., Khadari, B., Baradat, P., Bervillé, A., 2002. Combination of chloroplast and mitochondrial DNA polymorphism to study cytoplasm genetic differentiation in the olive complex (Olea europaea L.). Theor. Appl. Genet. 105, 139–144. Besnard, G., Henry, P., Wille, L., Cooke, D., Chapuis, E., 2006. On the origin of the invasive olives (Olea europaea L., Oleaceae). Heredity 99, 608–619. Besnard, G., Garcia-Verdugo, C., Rubio De Casas, R., Treier, U.A., Galland, N., Vargas, P., 2008. Polyploidy in the olive complex (Olea europaea): evidence from flow cytometry and nuclear microsatellite analyses. Ann. Botany 101, 25–30. Breviglieri, N., Battaglia, E., 1954. Ricerche cariologiche in Olea europaea L. Caryologia 6, 271–283.
Busconi, M., Foroni, C., Corradi, M., Bongiorni, C., Cattapan, F., Fogher, C., 2003. DNA extraction from olive oil and its use in the identification of the production cultivar. Food Chem. 83, 127–134. Busconi, M., Sebastiani, L., Fogher, C., 2006. Development of SCAR markers for germplasm characterization in olive tree (Olea europaea L.). Mol. Breeding 17, 59–68. Contento, A., Ceccarelli, M., Gelati, M.T., Maggini, F., Baldoni, L., Cionini, P.G., 2002. Diversity of Olea genotypes and the origin of cultivated olives. Theor. Appl. Genet. 104, 1229–1238. Cresti, M., Linskens, H.F., Mulchay, D.L., Bush, S., Di Stilio, V., My, X., Vignani, R., Cimato, A., 1997. Preliminary communication about the identification of DNA in leaves and in olive oil of Olea europaea. Olivae 69, 36–37. Cronquist, A., 1981. An integrated system of classification of flowering plants. Columbia University Press, N.Y. Green, P.S., 2002. A revision of Olea L. (Oleaceae). Kew Bulletin 57, 91–140. Hamman-Khalifa, A.M., Castro, A.J., Jímenez-López, J.C., Rodríguez-García, M.I., de Dios Alché, J., 2008. Olive cultivar origin is a major cause of polymorphism for Ole e 1 pollen allergen. BMC Plant Biol. 8, 10. Hernández, M.L., Mancha, M., Martínez-Rivas, J.M., 2005. Molecular cloning and characterization of genes encoding two microsomal oleate desaturases (FAD2) from olive. Phytochem. 66, 1417–1426. Loureiro, J., Rodriguez, E., Costa, A., Santos, C., 2007. Nuclear DNA content estimations in wild olive (Olea europaea L. ssp. europaea var. sylvestris Brot.) and Portuguese cultivars of O. europaea using flow cytometry. Genet. Resour. Crop. Evol. 54, 21–25. Natali, L., Giordani, T., Buti, M., Cavallini, A., 2007. Isolation of Ty1copia putative LTR sequences and their use as a tool to analyse genetic diversity in Olea europaea. Mol. Breeding 19, 255–265. Pafundo, S., Agrimonti, C., Marmiroli, N., 2005. Traceability of plant contribution in olive oil by amplified fragment length polymorphisms. J. Agric. Food Chem. 53, 6995–7002. Pasqualone, A., Montemurro, C., Summo, C., Sabetta, W., Caponio, F., Blanco, A., 2007. Effectiveness of microsatellite DNA markers in checking the identity of protected designation of origin extra virgin olive oil. J. Agric. Food Chem. 55, 3857–3862. Rallo, P., Tenzer, I., Gessler, C., Baldoni, L., Dorado, M., Martin, A., 2003. Transferability of olive microsatellite loci across the genus Olea. Theor. Appl. Genet. 107, 940–946. Reale, S., Doveri, S., Diaz, A., Angiolillo, A., Lucentini, L., Pilla, F., Martin, A., Donini, P., Lee, D., 2006. SNP-based markers for discriminating olive (Olea europaea L.) cultivars. Genome 49, 1193–1205. Rubio de Casas, R., Besnard, G., Schonswetter, P., Balanguer, L., Vargas, P., 2006. Extensive gene flow blurs phylogeographic but not phylogenetic signal in Olea europaea L. Theor. Appl. Genet. 113, 573–583. Rugini, E., Pannelli, G., Ceccarelli, M., Muganu, M., 1996. Isolation of triploid and tetraploid olive (Olea europaea L.) plants from mixoploid cv. ‘Frantoio’ and ‘Leccino’ mutants by in vivo and in vitro selection. Plant Breed. 115, 23–27. Sabino Gil, F., Busconi, M., Da Camara Machado, A., Fogher, C., 2006. Development and characterization of microsatellite loci from Olea europaea. Mol. Ecol. Notes 6, 1275–1277. Terral, J.F., Arnold-Simard, G., 1996. Beginnings of olive cultivation in eastern Spain in relation to holocene bioclimatic changes. Quaternary Res. 46, 176–185. Vos, P., Hogers, R., Bleeker, M., Reijans, M., Vande Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M., Zabeu, M., 1995. AFLP: a new technique for DNA fingerprinting. Nucl. Acids Res. 23, 4407–4414. Zohary, D., Hopf, M., 1994. Domestication of plants in the Old World, 2nd edn. Oxford Clarendon Press, Oxford, U.K., pp. 137–142.
Chapter 3
Current Initiatives in Proteomics of the Olive Tree Wei Wang, Fuju Tai and Xiuli Hu College of Life Sciences, Henan Agricultural University, Zhengzhou 450002, China
3.1 INTRODUCTION
spreading crowns (Figure 3.1A). Leaves, arranged in opposite, are linear with entire margins and acute tips, silvergreen in color, and thick (Figure 3.1B). They can live for 2–3 years before abscission. After exposure to low temperature, lateral buds of olive tree develop into inflorescences while terminal buds continue to grow vegetatively. Usually, 15–30 small, off-white flowers consist of a racemose inflorescence (Figure 3.1C). A perfect flower has four sepals and four petals, two stamens and a superior ovary. Olive fruit is a drupe, which is oblong with smooth, waxy surfaces. It is green or yellowish-green when immature, and turns red, purple, or black at full maturity (requiring 6–8 months) (Figure 3.1D). Mature olive fruit exhibits a typical drupe structure with a thin protective exocarp, a fleshy mesocarp and a stony endocarp which surrounds the seed (Figure 3.1E). Most table olives are harvested in mid-autumn when they change from green to yellowish-green and are firm. Oil olives are harvested in late autumn or winter when they have turned black, with a reduction in chlorophyll content and an increase in anthocyanin content, and have maximal oil content. The oil content of fruit mesocarp tissue is about 75% of fresh weight, accounting for more than 95% of the total oil in the fruit, compared to less than 3% of seed oil (Haralampidis et al., 1998).
Olive tree is the most economically important oil-producing crop in many Mediterranean countries. Currently, about 90% of olive trees are grown in these countries, especially in Spain, Italy and Greece. The cultivation of olive tree has expanded into Australia, China, Latin America, South Africa and the USA (Hatzopoulos et al., 2002). Over time, research on olive trees has become well established, particularly with respect to oil biosynthesis in olive drupe. Proteomics is the analysis of the protein complement of a cell or an organism. As a novel tool for protein identification and gene function analysis, it has become widely applicable in plants. In olive tree, proteomics of pollen allergen, storage proteins and stress responses is recently initiated. This chapter introduces the specific aspects associated with olive proteomics (e.g., protein extraction and interfering compound removal), and summarizes the current state and initiatives in olive proteomics.
3.2 BOTANICAL FEATURES OF OLIVE TREE Olive trees (Olea europaea L.) in the Oleaceae family are large, evergreen shrubs in their native state, but are usually trimmed as stout trees on massive trunks with round, A
B
C
D
E
exocarp mesocarp endocarp
FIGURE 3.1 Olive tree and its leaf, flower and fruit. (A) A nearly 50-year-old olive tree grown in an olive orchard, Sichuan, China. (B) Young (left) and aged (right) olive leaves, bar ⫽ 1 cm. (C) Olive inflorescences (racemose panicle) consisting of small flower. (D) An olive fruit turning purple. (E) A drupe showing its anatomy structure. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
25
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
26
SECTION | I The Plant and Production
Olive is a highly variable species with more than 2600 different cultivars, including synonyms, or ecotypes (Hatzopoulos et al., 2002). Each olive-planting country has its unique cultivars used for oil or table olives. For example, ‘Leccino’ and ‘Frantoio’ are the widespread cultivars for oil production in Italy; ‘Koroneiki’ is a common oil cultivar in Greece.
3.3 PROTEOMICS
Intensity
Proteomics is one of the fastest growing areas of biological research, largely because the global-scale analysis of proteins is expected to yield more direct understanding of function and regulation than analysis of genes. Currently, the availability of the complete Arabidopsis, rice, poplar and grapevine genomes, along with increasing nucleotide databases,
and improvements in protein separation, imaging and identification technologies, have all contributed to the wide application of proteomics in plant biology (Agrawal et al., 2005; Rossignol et al., 2006). Over years, proteomics has evolved from the early qualitative, protein cataloging towards a quantitative, profiling technology useful for making metabolic comparisons and predictions (Hajduch et al., 2006). The flowchart of proteomics approaches is illustrated in Figure 3.2. Proteins are first separated with gel-free or gel-based approaches, the separated proteins are subjected to enzymatic proteolysis, analyzed by mass spectrometry (MS), usually matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF), and recognized by bioinformatics techniques (Table 3.1). Typically, the approach for protein separation is two-dimensional gel electrophoresis
2-DE
Protein sample
Enzyme digestion
m/z
Database search
Intensity
MS analysis
Time
LC FIGURE 3.2 Flow chart of proteomics. Protein samples are separated by 2-DE or LC, the protein spots or fractions of interest are subjected to enzymatic digestion. The attributes measured by MS analysis are used to search protein-sequence databases for protein identification.
TABLE 3.1 Key facts of proteomics. The common concepts and techniques involved in proteomics are outlined, with emphasis on 2-DE-based proteomics. 1. Proteomics is the study of the protein complement in a cell, tissue or organism. The term ‘proteome’ was first coined in 1994, and refers to all the proteins in a cell, tissue, or organism. Proteomics refers to the study of the proteome 2. 2-DE, MS and bioinformatics tools are the key components of classical proteomics. Traditionally, proteomics experiments have been done using 2-DE. The steps of 2-DE-based proteomics approach include: protein extraction and solubilization, separation of the proteins by 2-DE, protein visualization and image analysis, MS analysis of specific attributes of the proteins of interest, and searching of databases 3. In 2-DE, protein samples are first separated by their charge in IPG strips, and then are separated by size using SDS-PAGE. For visualization, proteins in the gel are stained using a variety of different methods. The most widely used methods are silver staining, Coomassie blue staining, fluorescence detection systems, etc. 2D gel images are analyzed with specialized software. Proteins resolved by 2-DE can be are enzymatically digested and identified by MS analysis 4. MS analysis measures the unique attributes of proteins. The first attribute is the peptide-mass fingerprint. The second attribute is fragmentation of selected peptides into series of ions. The peptide mass fingerprint (PMF) for every digested protein is analyzed by MALDI-TOF 5. PMF, product-ion data or peptide-sequence tags are used to search a protein-sequence database to identify the protein of interest. The identification is made by comparing the experimental data with theoretical data calculated for each database entry. Several protein sequence databases are available in the public domain, e.g., SWISSPROT database 6. Database search programs are often in commercial software packages affiliated with mass spectrometers or can be accessed free-ofcharge. Commonly used programs include the MASCOT, SEQUEST, PeptIdent, MUtiIdent, etc.
CHAPTER | 3 Current Initiatives in Proteomics of the Olive Tree
(2-DE) or liquid chromatography (LC), followed by MS analysis. Despite rapid advances in gel-free proteomics, 2-DE coupled to MS currently remains the dominant proteomic technique (Rossignol et al., 2006). One advantage of this approach is the ability to separate different isoforms of the same protein due to post-translational modifications. Protein samples are first separated by isoelectric focusing (IEF) according to the isoelectric point (pI) in an immobilized pH gradient (IPG), followed by the second dimensional separation of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) based on molecular weight (MW). The use of IPG has markedly increased the reproducibility and resolution of 2-DE. Protein visualization after electrophoresis represents a critical step in quantitative proteome analysis. Available methods using colloidal Coomassie blue (CBB), silver reagents, fluorescent dye, or Pro-Q Diamond phosphoprotein stain vary in limit of detection, dynamic range, and compatibility with MS analysis. An excellent manual for 2-DE with IPG strips is openly accessible at: http://www.weihenstephan. de/blm/deg/manual/manualwork2html02test.htm. Besides, the newly developed fluorescence difference gel electrophoresis (DIGE) labels protein samples with fluorescent dyes before 2-DE, enabling accurate analysis of differences in protein abundance between pair samples within the same gel, thus avoiding gel-to-gel variance (Amme et al., 2006). Afterwards, software (e.g., PDQuest, ImageMaster) is used to analyze protein profiles, and spots of interest are subjected to MS or N-terminal sequencing. At present, plant proteomics is focused on model plants Arabidopsis, rice, maize, etc., to address the biochemical, physiological, metabolic and developmental processes. The advances in plant proteomics were recently reviewed (Rossignol et al., 2006). Systematic proteomic analysis of protein expression during oilseed development has recently been carried out in soybean and oilseed rape. Especially, 103 phosphorylated proteins expressed during oilseed rape seed filling have been identified by high-resolution 2-DE/ MS, of which approximately 80 are novel phosphoproteins, and 45% are involved in metabolism or energy production (Hajduch et al., 2006). However, olive proteomics has just been initiated, with few research articles available, mainly due to the limited genome resources and the great difficulty in protein extraction from olive tissues.
27
secondary metabolites displayed are species/tissue-specific and vary with age or developmental stage. They mainly accumulate in vacuoles in various soluble forms, and are richer in adult, green tissues than in young, etiolated tissues. Plants use secondary metabolites for protection against pests, as coloring, scent, or attractants and as plant hormones. Plant secondary metabolites can severely interfere with protein extraction and separation. Particularly, phenolics can form irreversible complexes with proteins, and the oxidation of phenolics by phenoloxidases and peroxidases can cause streaking and generate artificial spots on 2-DE gels (Vâlcu and Schlink, 2006). Olive leaves contain large amounts of phenolic compounds, ranging from 15 to 70 mg g⫺1 fresh weight (Niaounakis and Halvadakis, 2004), mainly in the form of oleuropein (Briante et al., 2002). When leaf is destroyed by herbivores or tissue homogenization, oleuropein becomes a very strong protein denaturant by activating enzymes (e.g. β-glucosidase) localized in different organelles (Konno et al., 1999). So, this may explain why direct homogenization of olive leaf in aqueous extraction buffers, followed by protein precipitations, always results in brownish pellets, which are difficult for further processing (Wang et al., 2003). Likely, olive drupes contain a large amount of oleuropein, and the majority of the polyphenols found in olive oil or table olives are products of its hydrolysis (Stupans et al., 2002). Besides, olive drupe often contains significant amounts of organic acids, pigments, and storage compounds (lipids and polysaccharides) that interfere with protein extraction and 2-DE analysis. Pigments, polysaccharides and lipids can also cause severe disturbances in 2-DE gels. Therefore, olive leaf and fruit mesocarp can be considered as recalcitrant tissues for proteomic analysis. Phenolics content in pollen or seed is low compared to olive leaf and mesocarp tissues. Oil bodies are the predominant organelles in seeds and pollen and contain a triacylglycerol matrix surrounded by a layer of phospholipids embedded with oleosins (Alché et al., 2006). The main problem in protein extraction from pollen and seeds is the disturbance of excess lipids. Two strategies used for removal of secondary metabolites were recently summarized (Wang et al., 2008). For olive leaf and fruit tissues, 10% trichloroacetic acid (TCA)/ acetone cleanup of tissue powder is an effective step for removing secondary compounds prior to protein extraction (Wang et al., 2003, 2006).
3.4 SECONDARY METABOLITES POTENTIALLY AFFECTING PROTEIN EXTRACTION
3.5 PROTEIN EXTRACTION FROM OLIVE TISSUES FOR PROTEOMICS ANALYSIS
Compared to other organisms, plants contain high levels of secondary metabolites (more than 40 000 kinds) (Crozier et al., 2006). Among them, phenolics are the common constituents, including phenols, flavonoids, stilbenes, tannins, lignins, etc. (Stalikas, 2007). The occurrence and kinds of
High-quality protein sample is a prerequisite for successful proteomic analysis. Over recent years, many efforts have been made to develop protein extraction protocols for enhanced plant proteomic analysis, especially on olive leaf and fruit tissues (Wang et al., 2003, 2006; Malik and Bradford, 2005).
28
SECTION | I The Plant and Production
Plant tissue powder
A: TCA/acetone method
B: Phenol-based method
C: TCA/acetone/phenol method
Precipitation In TCA/acetone (−20°C)
Extraction In aqueous medium (4°C)
Precipitation In TCA/acetone (−20°C)
Wash and dry Acetone (−20°C) wash twice Air-dry
Phenol extraction Mixing phenol with crude extract (1:1)
Wash and dry Acetone (−20°C) wash twice, Air-dry
Resuspension In 2-DE lysis buffer, Vortexing for 1h
Precipitation Ammonium acetate in methanol, −20°C for 2h
Resuspension SDS extraction buffer/phenol (1:1), vortexing for 1h
Wash and dry Methanol and acetone wash Air-dry
Precipitation Ammonium acetate in methanol, −20°C for 2h
Resuspension In 2-DE lysis buffer Vortexing for 1h
Wash and dry Methanol & acetone wash Air-dry Resuspension In 2-DE lysis buffer Vortexing for 1h
FIGURE 3.3 The scheme of three protocols for protein extraction from recalcitrant plant tissues. Plant tissue powder is routinely used as starting material in protein extraction. (A) The most common protocol is based on precipitating proteins from homogenized tissue or cells with TCA in acetone. (B) An alternative protocol is based on the solubilization of proteins in phenol, followed by their precipitation with ammonium acetate in methanol. (C) The TCA/acetone/phenol method integrates phenol-based extraction with TCA/acetone cleanup steps prior to protein extraction. These basic protocols might have to be modified to achieve optimal results for different plant tissues.
TABLE 3.2 Protein extraction from olive tissues. Selected examples of protein extraction from various olive tissues outlined, with emphasis on recent developed methods. Tissue and protein Protein extraction Examples Total leaf protein TCA/acetone/phenol extraction Wang et al. (2003, 2006) TCA/acetone precipitation Malik and Bradford (2005) Fruit mesocarp TCA/acetone/phenol extraction Wang et al. (2006) Total seed proteins Chloroform/TCA/acetone precipitation Wang et al. (2007) Pollen allergen Aqueous bicarbonate extraction Barral et al. (2004)
The three protocols suitable for protein extraction from recalcitrant plant tissues are outlined in Figure 3.3. Due to the presence of high levels of secondary metabolites, protein extraction from olive tissues (esp. leaf and fruit) should be optimized to guarantee good results of proteomics analysis. Selected examples are given to highlight protein extraction from olive tissues (Table 3.2).
3.5.1 Leaf Evergreen and thick leaves of olive tree are hard to disrupt. Only in liquid N2 can aged leaves be ground into a fine powder in mortar. The finer olive leaf powder can be obtained by additional grinding of dry tissue powder (Wang et al., 2003). Alternatively, homogenizing olive leaves in TCA/acetone in a device that generates strong shearing forces (e.g. a Polytron) (Malik and Bradford, 2005) also produces a fine powder. Using fine tissue powder as a starting material has become routine practice in sample preparation of plant proteomics. Previously, extracting protein from olive leaves by homogenizing leaf material in aqueous medium, followed by various protein precipitations, has not resulted in good results (Garcia et al., 2000; Wang et al., 2003). In the case of olive leaf, it is necessary to remove secondary metabolites prior to protein extraction. Wang et al. (2003) developed a protocol by combination of TCA/acetone precipitation and phenol extraction, which provides the first successful extraction of olive leaf protein for 2-DE. It involves extensive cleanup steps with cold acetone, 10% TCA/acetone, 10% aqueous TCA to remove pigments, phenolics, etc. from tissue powder. The protocol takes 1–2 days with a protein yield 2.49 mg g⫺1 of aged leaf, and allows obtaining well-resolved 2-DE patterns of olive leaves for the first time (Figure 3.4). Its simplified version, consuming less time, compatible with
29
CHAPTER | 3 Current Initiatives in Proteomics of the Olive Tree
4 −
pI
A
74
pI
7
B
+ FIGURE 3.4 Representative 2-DE gels of olive tissue proteins extracted using TCA/acetone/phenol method. (A) CBB-stained gel of aged leaf (7 cm, pH 4–7, 100 μg). (B) CBB-stained gel of fruit mesocarp (7 cm, pH 4–7, 100 μg). Grateful acknowledgment is made to Wiley-VCH Verlag GmbH & Co for permission to reproduce material from Wang et al. (2006).
silver staining, could be used for routine protein extraction from recalcitrant plant tissues for proteomic analysis (Wang et al., 2006). Additionally, powdered olive leaf tissue can be homogenized with a polytron in cold 10% TCA/acetone plus 0.07% ME and then centrifuged several times, and the dry, fine leaf tissue is used for protein extraction in the 2-DE lysis buffer. By this way, 2-DE map comparable to phenolbased one can be obtained (Malik and Bradford, 2005).
3.5.2 Fruit Different from oilseed crops, olive fruit mesocarp is the main tissue of oil accumulation. Global characterization of proteins expressed during olive fruit development will provide a foundation for uncovering temporal relationships between metabolic enzyme expression and oil production. Fruit mesocarp is characterized by a high content of oil, pigments, phenolics, etc. Proteins in mesocarp (cv. Arbequina and Picual) constitute 1.3–1.8% of the dry weight of the olive fruit (Zamora et al., 2001), but protein composition is not very well known at present. The protocol based on TCA/acetone/phenol extraction could be applicable in mesocarp, producing a good 2-DE map (Figure 3.4; Wang et al., 2006).
3.5.3 Seed Mature olive seeds contain high amounts of lipids and proteins. A protocol based on the chloroform/TCA/acetone precipitation works well in oilseed (including olive seeds) protein extraction (Wang et al., 2004). By the protocol, oilseed extract is mixed well with an equal volume of chloroform/methanol (2:1), and lipids are sequentially washed away using 10% TCA/acetone, aqueous 10% TCA and acetone. Protein samples prepared by the protocol allow obtaining well-resolved spots on 2-DE gel. Likely, olive
seed proteins extracted by a simple protocol relying on chloroform/acetone are excellently resolved in the gel, with clear bands and improved resolution (Wang et al., 2007).
3.6 CURRENT INITIATIVES IN OLIVE PROTEOMICS Compared to the numerous proteomics research efforts in Arabidopsis, rice and maize, the dearth of olive proteomics is particularly noticeable. To date, there are only several studies on olive pollen allergen, seed storage protein and stress responses involving proteomic approaches.
3.6.1 Pollen Allergen More than 20% of the population in the Mediterranean countries is affected by type I allergy during the pollination season (Florido et al., 1999), and allergies constitute a serious global medical problem (Tichá et al., 2002). Olive pollen is one of the main causes of seasonal respiratory allergy in these countries. The proteomic analysis of pollen allergens is important to deepen our understanding of their allergenic nature and basic aspects of pollen biology, and to reveal the biological functions. Ten allergens (Ole e 1–10) from olive pollen have been described to date (Barral et al., 2004; Alché et al., 2007). The properties of isolated and characterized olive pollen allergens are summarized in Table 3.3. Ole 1 is considered the major allergen, affecting more than 70% of the olive-pollen-allergic patients (Rodríguez et al., 2002). It was suggested that Ole e 1, homologous to tomato LAT52 protein, might play a role in signal transduction during pollen germination (Alché et al., 2004). For pollen allergen extraction, the high solubility of aqueous medium is an important prerequisite. The partition coefficient of allergens and the antigenic profile of olive pollen are strongly related to the solvents used for the
30
SECTION | I The Plant and Production
TABLE 3.3 Properties of isolated and characterized olive pollen allergens. Ten allergens (Ole e 1–10) from olive pollen have been described to date. Their identified properties are listed with respect to identity, MW, pI and the number of isoforms. Allergen kDa pI Isoforms Identity References Ole e 1 18–21 5.5–6.6 3 Glycoprotein Rodríguez et al. (2002) Ole e 2 13–16 5.06–5.21 At least five Profilin Alché et al. (2007) Ole e 3 9.2 4.49 Polcalcin Batanero et al. (1996) Ole e 4 32 4.6–5.1 At least two superoxide dismutase Carnés and Fernández-Caldas (2002) Ole e 5 16 5.1–6.5 At least five superoxide dismutase-like protein Carnés and Fernández-Caldas (2002) Ole e 6 10 4.2 Cysteine-enriched protein Batanero et al. (1997)
In other plants, Corti et al., (2005) used a proteomic approach to detect grass allergens from a natural protein extract. Consequently, six out of eight expected clinically relevant allergens in the natural grass extract are identified with 2-DE/immunoblot and MS analysis. Petersen et al. (2006) made the proteome analysis of maize pollen for allergy-relevant components associating with 2-DE immunoblotting. They confirmed that major cross-reacting allergens in pollen belong to groups 1 and 13 (Zea m 1 and 13), both having high IgE prevalence and sequence identities of 72 and 70%, respectively, to the corresponding Phi p 1 and Phl p 13 allergens of timothy grass pollen. Besides, rice pollen proteome analysis reveals five putative pollen allergens, whose relative levels are reduced on pollen germination (Dai et al., 2007). In summary, the recent proteomics of pollen allergens shows promise to identify new allergens, allergen composition, allergenic potency of pollen, and the sensitization profile of allergic patients.
Ole e 7 9–11 3.56 At least two lipid transfer protein Tejera et al. (1999) Ole e 8 21 4.51 Ca2⫹-binding protein Ledesma et al. (2002) Ole e 9 46 4.8–5.4 4 1,3-β-Glucanase Barral et al. (2004) Ole e 10 10.8 5.8 Homologous with C-terminal of Ole e 9 Barral et al. (2004)
extraction (Carnès et al., 2002). Various methods have been employed for the isolation and characterization of the allergens, including SDS-PAGE/immunoblot (Lauzurica et al., 1988), gel filtration (Boluda et al., 1999) and 2-DE/immunoblot (Rodríguez et al., 2002), and high-performance liquid chromatography (HPLC) coupled with MS (Rubio et al., 1987; Napoli et al., 2006). By using 2-DE/immunoblot analysis, Rodríguez et al., (2002) found that Ole e 1 appears as many spots of different pI values (5.5, 5.9, and 6.6) but identical MW, which is consistent with the polymorphic character of the protein. However, other olive pollen allergens, especially those of high MW, have been neither isolated nor characterized. Alternatively, Barral et al. (2004), using the reverse phase (RP) HPLC/MS approach, found that allergen Ole e 9 can be resolved by column chromatography and RP-HPLC as two different forms and by IEF as four components with pI values 4.8, 4.9, 5.1, and 5.4, exhibiting a low but significant polymorphism. A recent study on a number of olive pollen extracts of different cultivars detected significant differences in their allergenic compositions. 2-DE profiles obtained from the pollen of two cv. Picual and Arbequina are quite different in Ole e 1 content, suggesting that pollen allergen polymorphism is closely related to the cultivar origin (Napoli et al., 2006).
3.6.2 Seed and Fruit Proteomics Seed storage proteins (SSPs) are formed during seed development and deposited predominantly in specialized storage tissues, like cotyledon or endosperm, and used for early seedling growth. SSPs from soybean and cereal crops are important protein resources for humans. Due to their high abundance in seeds, SSPs can be selected for molecular markers in breeding programs and for describing plant cultivars and lines (Liang et al., 2006). Since whole fruits are processed for oil production, the knowledge of olive SSPs is ignored for a long time. Wang et al. (2001) first reported the analysis of olive SSPs by SDS-PAGE, immunoblotting and N-terminal sequencing. Mature olive seeds contained five prominent protein bands (20–30 kDa), which represent the reduced forms of olive SSPs. Based on sequence homology, olive SSPs are classified to 11S legumin (globulin) family, whereas they show similar in solubility to prolamins in cereal seeds, with a high solubility in aqueous alcohol, a limited solubility in water and dilute salt. The synthesis of SSPs is highly conserved in olive species, and no visible differences exist in subunit composition among six olive cultivars examined (Wang et al., 2007). Moreover, by using immunoelectroscopy technique, olive SSPs are found to accumulate in conspicuous protein bodies present in both the endosperm and the cotyledon (Alché et al., 2006). The basic character of each subunit of olive SSPs is further verified by 2-DE, and the isoforms of SSPs dominate protein 2-DE profiles. Besides, the putative presence of highly similar isoforms or posttranslational modifications of these polypeptides are detected (Alché et al., 2006). The proteomic analysis of global protein changes during fruit development will be important to elucidate key
31
CHAPTER | 3 Current Initiatives in Proteomics of the Olive Tree
enzymes for olive oil biosynthesis. However, there is only a conference abstract on this topic up to now. Motamed et al. (2007) briefly reported both qualitative and quantitative changes in protein content and their expression during olive fruit maturation using 2-DE. A close association between growth, ripening, expression and appearance of specific polypeptides in fruit was found. These polypeptides could be considered as molecular markers for developmental stages and full fruit maturity.
3.6.3 Stress Proteomics As a Mediterranean tree, olive trees are subjected to severe climate factors such as high radiation, drought, salinity and drastic temperature changes, which can induce oxidative stress situations and significant losses in crop productivity. They are generally cultivated in areas where water is the main limiting factor in agricultural production (Hatzopoulos et al., 2002). Until present, the mechanisms of olive adaptation to drought (Sofo et al., 2004; Secchi et al. 2007), cold (D’Angeli and Altamura, 2007), and salinity (Valderrama et al., 2006) are mainly characterized using physiological, biochemical and molecular techniques (SDS-PAGE and immunoblot analysis), designed for some special enzymes. Proteomics is promising in its validity for in-depth monitoring the overall changes in proteomes of selected organs exposed to stress conditions, as in Arabidopsis (Amme et al., 2006). However, proteomic analysis of global protein changes of olive tree exposed to various stress conditions are still lacking. A study is under way to explore the differences in protein patterns existing in two olive clones (tolerance to cold vs. nontolerant) after cold treatments (Bartolini et al., 2007).
SUMMARY POINTS ●
●
●
●
As a novel tool for protein identification and gene function analysis, proteomics has become widely applicable in plants, including olive tree. Olive tissues contain high amounts of interfering compounds, thus bringing specific difficulties in proteomic analysis. Removal of these non-protein compounds should be considered before protein extraction. Protein extraction protocols for olive tissues (leaf, fruit, seed and pollen) have been established and are expected to be applicable in future proteomic analysis. For protein extraction from olive leaf and fruit tissues, a TCA/ acetone/phenol extraction protocol is recommended. Olive proteomics is still at its infancy. Current olive proteomic analysis is mainly on pollen allergens. The proteomics of pollen allergens shows the great potential to identify new allergens, allergen composition and allergenic potency of pollen. Stress response proteomics in olive tree has attracted many efforts.
●
There is currently no large-scale proteomic analysis on oil biosynthesis in olive fruit. It is expected that the application of proteomic approach on the oil-producing process in olive fruit will produce promising results.
ACKNOWLEDGMENTS We wish to thank Dr. Monica Scali (University of Siena, Italy) for valuable advice and help during manuscript writing.
REFERENCES Agrawal, G.K., Yonekura, M., Iwahashi, Y., Iwahashi, H., Rakwal, R., 2005. System, trends and perspectives of proteomics in dicot plants. Part I: Technologies in proteome establishment. J. Chromatogr. B. 815, 109–123. Alché, J.D., M’rani-Alaoui, M., Castro, A.J., Rodríguez-García, M.I., 2004. Ole e 1, the major allergens from olive (Olea europaea L.) pollen, increases its expression and is released to the culture medium during in vitro germination. Plant Cell Physiol. 45, 1149–1157. Alché, J.D., Jimenez-Lopez, J., Wang, W., Castro-Lopez, A.J., RodríguezGarcía, M.I., 2006. Biochemical characterization and cellular localization of 11S type storage proteins in olive (Olea europaea L.) seeds. J. Agric. Food Chem. 54, 5562–5570. Alché, J.D., Castro, A.J., Jiménez-López, J.C., Morales, S., Zafra, A., Hamman-Khalifa, A.M., Rodríguez-García, M.I., 2007. Differential characteristics of olive pollen from different cultivars: biological and clinical implications. J. Investig. Allergol. Clin. Immunol. 17 (Suppl.), 69–75. Amme, S., Matros, A., Schlesier, B., Mock, H.P., 2006. Proteome analysis of cold stress response in Arabidopsis thaliana using DIGE-technology. J. Exp. Bot. 57, 1537–1546. Barral, P., Batanero, E., Palomares, O., Quiralte, J., Villalba, M., Rodríguez, R., 2004. A major allergen from pollen defines a novel family of plant proteins and shows intra- and interspecies crossreactivity. J. Immunol. 172, 3644–3651. Bartolini, G., Di Monte, G., Rea, E., Toponi, M. A., 2007. Protein patterns in response to cold stress on clones of Olea europaea L. cv. Leccino. Acta Horticulturae International Symposium on Olive Growing. 474, III. Batanero, E., Ledesma, A., Villalba, M., Rodríguez, R., 1997. Purification, amino acid sequence and immunological characterization of Ole e 6, a cysteine-enriched allergen from olive tree pollen. FEBS Lett 410, 293–296. Batanero, E., Villalba, M., Ledesma, A., Puente, X.S., Rodriguez, R., 1996. Ole e 3, an olive-tree allergen, belongs to a widespread family of pollen proteins. Eur. J. Biochem. 241, 772–778. Boluda, L., Sastre, J., Casanovas, M., Fernandez-Caldas, E., 1999. Determination of Ole e 1 by enzyme immunoassay and scanning densitometry. Validation by skin-prick testing. J. Immunol. Methods 223, 17–26. Briante, R., Patumi, M., Terenziani, S., Bismuto, E., Febbraio, F., Nucci, R.J., 2002. Olea europaea L. leaf extract and derivatives: Antioxidant properties. J. Agric. Food Chem. 50, 4934–4940. Carnès, J., Fernandez-Caldas, E., Boluda, L., Casanovas, M., Sastre, J., Lluch-Bernal, M., Blanca, M., 2002. Rapid release of Ole e 1 from Olive pollen using different solvents. Allergy 57, 798–804. Carnés, J., Fernández-Caldas, E., 2002. Ole e 4 and Ole e 5, important allergens of Olea europaea. Allergy 57 (Suppl. 71), 24–28.
32
Corti, V., Cattaneo, A., Bachi, A., Rossi, R.E., Monasterolo, G., Paolucci, C., Burastero, S.E., Alessio, M., 2005. Identification of grass pollen allergens by two-dimensional gel electrophoresis and serological screening. Proteomics 5, 729–736. Crozier, A., Clifford, M., Ashihara, H., 2006. Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet. Blackwell Publishing, London. Dai, S., Chen, T., Chong, K., Xue, Y., Li, S., Wang, T., 2007. Proteomic identification of differentially expressed proteins associated with pollen germination and tube growth reveals characteristics of germinated Oryza sativa pollen. Mol. Cell. Proteomics 6, 207–230. D’Angeli, S., Altamura, M.M., 2007. Osmotin induces cold protection in olive trees by affecting programmed cell death and cytoskeleton organization. Planta 225, 1147–1163. Florido, J.F., Gonzalez-Delgado, P., Saenz de San Pedro, B., Quiralte, J., Arias de Saavedra, J.M., Peralta, V., Ruiz Valenzuela, L., 1999. High levels of Olea europaea pollen and relation with clinical findings. Int. Arch. Allergy Immunol. 119, 133–137. Garcia, J.L., Avidan, N., Troncoso, A., Sarmiento, R., Lavee, S., 2000. Possible juvenile-related proteins in olive tree tissues. Sci. Hort. 85, 271–284. Hajduch, M., Castee, J.E., Hurrelmeyer, K.E., Song, Z., Agrawal, G.K., Thelen, J.J., 2006. Proteomic analysis of seed filling in Brassica napus. Developmental characterization of metabolic isozymes using highresolution two-dimensional gel electrophoresis. Plant Physiol. 141, 32–46. Haralampidis, K., Milioni, D., Sanchez, J., Baltrusch, M., Heinz, E., Hatzopoulos, P., 1998. Temporal and transient expression of stearoylACP carrier protein desaturase gene during olive fruit development. J. Exp. Bot. 49, 1661–1669. Hatzopoulos, P., Banilas, G., Giannoulia, K., Gazis, F., Nikoloudakis, N., Milioni, D., Haralampidis, K., 2002. Breeding, molecular markers and molecular biology of the olive tree. Eur. J. Lipid Technol. 104, 574–586. Konno, K., Hirayama, C., Yasui, H., Nakamura, M., 1999. Enzymatic activation of oleuropein: A protein crosslinker used as a chemical defense in the privet tree. Proc. Natl. Acad. Sci. 96, 9159–9164. Lauzurica, P., Gurbindo, C., Maruri, N., Galocha, B., Diaz, R., Gonzalez, J., Garcia, R., Lahoz, C., 1988. Olive (Olea europaea) pollen allergens. I. Immunochemical characterization by immunoblotting, CRIE and immunodetection by a monoclonal antibody. Mol. Immunol. 25, 329–335. Ledesma, A., Villalba, M., Vivanco, F., Rodriguez, R., 2002. Olive pollen allergen Ole e 8: identification in mature pollen and presence of Ole e 8-like proteins in different pollens. Allergy 57, 40–43. Liang, X.Q., Luo, M., Holbrook, C.C., Guo, B.Z., 2006. Storage protein profiles in Spanish and runner market type peanuts and potential markers. BMC Plant Biol. 6, 24–32. Malik, N.A., Bradford, J.M., 2005. A simple protein extraction method for proteomic studies on olive leaves. J. Food Agric. Environ. 3, 246–248. Motamed, N., Rastgar-Jazi, F., Karkhaneh, A., Ebrahimzadeh, H., 2007. Two dimensional gel electrophoresis and analysis of polypeptides in developmental stages of olive fruit ripening. Mol. Cell. Proteomics, Abstract 34.2, 2.9, p. 743. Niaounakis, M., Halvadakis, C.P., 2004. Olive-mill waste management. Literature review and patent survey. Typothito-George Dardanos Publications, Athens, Greece, pp. 13–44. Napoli, A., Aiello, D., Di Donna, L., Sajjad, A., Perri, E., Sindona, G., 2006. Profiling of hydrophilic proteins from Olea europaea olive pollen by MALDI TOF mass spectrometry. Anal. Chem. 78, 3434–3443.
SECTION | I The Plant and Production
Petersen, A.., Dresselhaus, T., Grobe, K., Becker, W.-M., 2006. Proteome analysis of maize pollen for allergen-relevant components. Proteomics 6, 6317–6325. Rodríguez, R., Villalba, M., Batanero, E., González, E.M., Monsalve, R. I., Huecas, S., Tejera, M.L., Ledesma, A., 2002. Allergenic diversity of the olive pollen. Allergy 57 (Suppl. 71), 6–16. Rossignol, M., Peltier, J.B., Mock, H.P., Matros, A., Maldonado, A.M., Jorrin, J.V., 2006. Plant proteome analysis: A 2004-2006 update. Proteomics 6, 5529–5548. Rubio, N., Brieva, A., Alcazar, B., 1987. Purification of allergens by highperformance liquid chromatography. IV. Purification of the allergen of olive pollen (Olea europaea). J. Chromatogr. 403, 312–318. Secchi, F., Lovisolo, C., Uehlein, N., Kaldenho, R., Schubert, A., 2007. Isolation and functional characterization of three aquaporins from olive (Olea europaea L.). Planta 225, 381–392. Sofo, A., Dichio, B., Xiloyannis, C., Masia, A., 2004. Lipoxygenase activity and proline accumulation in leaves and roots of olive trees in response to drought stress. Physiol. Plant 121, 58–65. Stalikas, C.D., 2007. Extraction, separation, and detection methods for phenolic acids and flavonoids. J. Sep. Sci. 30, 3268–3295. Stupans, I., Kirlich, A., Tuck, K.L., Hayball, P.J., 2002. Comparison of radical scavenging effect, inhibition of microsomal oxygen free radical generation, and serum lipoprotein oxidation of several natural antioxidants. J. Agric. Food Chem. 50, 2464–2469. Tejera, M.L., Villalba, M., Batanero, E., Rodríguez, R., 1999. Identification, isolation, and characterization of Ole e 7, a new allergen of olive tree pollen. J. Allergy Clin. Immunol. 104, 797–802. Tichá, M., Pacáková, V., Štulík, K., 2002. Proteomics of allergens. J. Chromatogr. B 771, 343–353. Vâlcu, C.M., Schlink, K., 2006. Reduction of proteins during sample preparation and two-dimensional electrophoresis of woody plant samples. Proteomics 6, 1599–1605. Valderrama, R., Corpas, F.J., Carreras, A., Gómez-Rodríguez, M.V., Chaki, M., Pedrajas, J.R., Fernández-Ocana, A., Del Río, L.A., Barroso, J.B., 2006. The dehydrogenase-mediated recycling of NADPH is a key antioxidant system against salt-induced oxidative stress in olive plants. Plant Cell Environ. 29, 1449–1459. Wang, W., Alché, J.D., Castro-Lopez, A.J., Rodríguez-García, M.I., 2001. Characterization of seed storage proteins and their synthesis during seed development in Olea europaea. Int. J. Dev. Biol. 45 (s), 63–64. Wang, W., Scali, M., Vignani, R., Spadafora, A., Sensi, E., Mazzuca, S., Cresti, M., 2003. Protein extraction for two-dimensional electrophoresis from olive leaf, a plant tissue containing high levels of interfering compounds. Electrophoresis 24, 2369–2375. Wang, W., Vignani, R., Scali, M., Sensi, E., Tiberi, P., Cresti, M., 2004. Removal of lipid contaminants by organic solvents from protein extracts of plant oilseeds prior to electrophoresis. Anal. Biochem. 329, 139–141. Wang, W., Vignani, R., Scali, M., Cresti, M., 2006. A universal and rapid protocol for protein extraction from recalcitrant plant tissues for proteomic analysis. Electrophoresis 27, 2782–2786. Wang, W., Alché, J.D., Rodríguez-García, M.I., 2007. Characterization of olive seed storage proteins. Acta Physiol. Plant. 29, 439–444. Wang, W., Tai, F.J., Chen, S.N., 2008. Optimizing protein extraction from plant tissues for enhanced proteomics analysis. J. Sep. Sci. 31, 2032–2039. Zamora, R., Alaiz, M., Hidalgo, F.J., 2001. Influence of cultivar and fruit ripening on olive (Olea europaea) fruit protein content, composition, and antioxidant activity. J. Agric. Food Chem. 49, 4267–4270.
Chapter 4
Chemometric Classification of Cultivars of Olives: Perspectives on Portuguese Olives Joaquim C.G. Esteves da Silva Centro de Investigação em Química (UP), Chemistry Department, Universidade do Porto, Portugal
4.1 INTRODUCTION
natural bitterness and softening of the tissue – one of the most used industrial methods to process olives is the black oxidizing or Californian processing (Mafra et al., 2001, 2006). European agriculture policies led to the establishment of limited olive-oil production regions and to the implementation of olive-oil ‘Denomination of Protected Origin (DOP)’ (European Community, 1992, 1993). DOP olive oil has a certified high quality, therefore an increased market value when compared to others. There are six DOP areas certified in Portugal: ‘Moura olive oil’; ‘Trás-os-Montes olive oil’; ‘Beira-Interior olive oil’; ‘North-Alentejo olive oil’; ‘Ribatejo olive oil’; ‘Alentejo-Interior olive oil’ (Casa do Azeite, 2008).
4.1.1 Portuguese Olive Cultivars Olive trees (Olea europaea) have been cultivated for thousands of years in the Mediterranean area (Leitão et al., 1986; Zamora et al., 2001; Bartolini and Petruccelli, 2002). Although the cultivation of olive tree has been extended to many other regions of the world, olive fruits remain a typical Mediterranean crop, where they play an important role in diet of the people in the area as well as in their economy and culture (Zamora et al, 2001; Bartolini and Petruccelli, 2002; Pinheiro and Esteves da Silva, 2005). In Portugal the olive trees are distributed in small groves in the Trás-os-Montes region located in the north of the country and in large plantations in the center and south of the country (Gemas et al., 2004). The most important olive cultivar in Portugal is the ‘Galega’, representing about 60% of the olive trees, and the fruit is useful as a table olive or for olive oil production – it shows a weak to average olive oil yield poorer in linoleic acid (Leitão et al., 1986). The other most abundant olives cultivars are Carrasquenha, Cordovil, Cobrançosa and Verdeal. The olive fruit is a drupe, becoming generally blackishpurple when fully ripe. Olives contain an alkaloid (oleuropein), which makes them bitter and unpalatable, a low sugar content compared with other drupes (2–5% against about 12%) and high oil content (20–30% against 1–2%) depending on the time of year and variety (Leitão et al, 1986; Mafra et al., 2006; Casa do Azeite, 2008). These characteristics make it a fruit that cannot be consumed directly from the tree and it has to undergo a series of processes that differ considerably from region to region, and which also depend on variety. Olives are usually consumed as table olives or as olive oil. Table olives result from the processing of raw olive with the objective of eliminating its Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
4.1.2 Chemometric Classification of Olive Cultivars The great importance of distinguishing and identifying olive cultivars has been recognized since Roman times (Leitão et al., 1986; Ganino et al., 2006). Several classification schemes have been tried based on botanic, ecological, morphological, commercial and agronomical characters (Leitão et al., 1986; Ganino et al., 2006). Presently there are a great number of different olive cultivars showing a diversity of morphological and physiological characteristics that result in quite different qualities and uses. However, the olive quality cannot be directly measured and several indirect measurements, for example fruit and endocarp features, must be correlated with the olive quality and cultivar (Cantini et al., 1999; Rotondi et al., 2003; Pinheiro and Esteves da Silva, 2005; Ganino et al., 2006). Also, the objective of these studies is the identification of the most discriminating features for the characterization of olive cultivars in order to reduce the parameters 33
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
34
necessary for olive cultivar identification (Leitão et al., 1986). This task usually corresponds to a multivariate analysis problem that can be solved using chemometric classification methods (Muhammad et al., 1986; Lavine, 1992; Vandeginste et al., 1998a, 1998b). These methods have already been successfully applied to olive classification studies (Cantini et al., 1999; Rotondi et al., 2003; Pinheiro and Esteves da Silva, 2005; Ganino et al., 2006). This chapter shows the great usefulness of chemometric classification techniques for olive cultivar study. This conclusion is illustrated using a data set constituted by 22 Portuguese olive cultivars and the chemometric techniques of unsupervised (non-linear maps and cluster analysis) and supervised (linear discriminant analysis) classification. Forty morphological characteristics of olive fruits, endocarps, trees, branches, leaves and flowers are used in the classification of the 22 cultivars with the following objectives: (i) to show that these characteristics are able to discriminate the cultivars; (ii) to demonstrate similarities among the cultivars under investigation; and (iii) to show that some characteristics have a larger discrimination power than others.
4.2 FEATURES OF CHEMOMETRIC CLASSIFICATION 4.2.1 Similarity of Olive Cultivars Olive cultivars (samples) are defined by a list of distinctive features (variables) that allow the discrimination of different samples and the development of classification rules that can be used to control the origin of the olives. In this work, 40 morphological variables (fruit, endocarp, trees, branches, flowers and leaves) will be analyzed (Tables 4.1 and 4.2). Morphological features of olive cultivars are easily and quickly obtained, even for a well-trained nonspecialist, and can be used as a screening analysis of the quality of olive cultivars. However, humans are not trained for the analysis of large data tables having a large number of samples and/or variables. Indeed, humans don’t have the capacity to interpret graphical representations with more than three dimensions (hyperspace). The objective of multivariate chemometric techniques is to reduce the hyperdimensional experimental data into a small number of bidimensional graphics that can be analyzed by the well-trained human eye. Morphological measurements (variables) are expected to be indirectly related to the cultivar and are used to obtain an empirical relationship or classification rule. These variables can be numeric or categorical – in this case a number (for example one, two, etc.) is assigned to each different category. A similarity index among the samples is naturally defined using the proximity (distance) in the spatial representation. The distance between two samples, i and j
SECTION | I The Plant and Production
TABLE 4.1 Variables of the olive fruit and olive endocarp and their codes. Fruit
Endocarp
Size of the fruit – F_size 1 – small (up to 1.9 g) 2 – medium (2 to 3.9 g) 3 – large (4 to 7.9 g) 4 – very large (greater than 8 g)
Size of the fruit – E_size 1 – small 2 – medium 3 – large 4 – very large
Shape – F_shape 1 – spherical 2 – ovoid 3 – asymmetric ovoid 4 – elliptic 5 – elongated
Shape – E_shape 1 – spherical 2 – ovoid 3 – asymmetric ovoid 4 – elliptic 5 – elongated
Position maximum transverse diameter – F_pmtd 1 – next to base 2 – central 3 – next to apex
Position of maximum transverse diameter – E_pmtd 1 – next to base 2 – central 3 – next to apex
Apex (position A) – F_apex 1 – rounded 2 – sharp-pointed
Apex – E_apex 1 – rounded 2 – sharp-pointed
Nipple – F_nipple 1 – with mild depression 2 – less evident 3 – absent 4 – with nipple
Apex termination – E_apexts 1 – without or less evident mucro 2 – with mucro
Base – F_base 1 – depressed 2 – rounded 3 – truncated
Base – E_base 1 – rounded 2 – truncated 3 – sharp-pointed 4 – nipple shaped
Size of the pedicel cavity – F_sds 1 – small 2 – ample
Surface – E_surf 1 – smooth 2 – rough 3 – very rough
Shape of the pedicel cavity – F_sdsh 1 – circular 2 – elliptical-oval
Suture line – E_sl 1 – reduced 2 – medium 3 – elevated
Depth of pedicel cavity – F_sdd 1 – little deeper 2 – medium deep 3 – very deep Color of epicarp at the turnover – F_bct 1 – reddish 2 – violet Color of epicarp at maturity – F_bcm 1 – reddish 2 – dark red 3 – violet 4 – dark Epicuticular wax coating – F_pc 1 – without 2 – with
CHAPTER | 4 Chemometric Classification of Cultivars of Olives: Perspectives on Portuguese Olives
Perceptibility of lenticels at maturity – F_lensp 1 – not detected 2 – less visible 3 – very visible
TABLE 4.2 Variables of the olive trees, branches, leaves and flowers and their codes. Trees
Branches
Lenticels density – F_lensd 1 – few numerous 2 – very numerous
Size of the tree – T_size 1 – small 2 – medium 3 – large
Type – B_b 1 – short 2 – medium 3 – elongated
Lenticels size – F_lenss 1 – little 2 – big
Tree structure – T_ar 1 – weeping 2 – median 3 – upright
Rugousity – B_r 1 – flat 2 – median 3 – rough
Mesocarp consistency – F_mc 1 – soft 2 – hard
Color – B_c 1 – greenish 2 – grayish 3 – brownish
Mesocarp clinginess – F_ma 1 – do not adhere 2 – adhere
(dij – each one representing a sample), in the original measurements space (hyperspace with the selected number of variables – n) can be computed using for example the Euclidean distance metric: dij = ⎡⎢ ∑ (k⫽1,n ) (xik ⫺ x jk )2 ⎤⎥ ⎣ ⎦
1/ 2
(4.1)
where the variables are represented by x. The similarity between two samples i and j (Sij) can be calculated using all the computed distances: Sij ⫽ 1 ⫺ dij / dmax
35
(4.2)
Leaves
Flowers
Size – L_ls 1 – small (⬍550 mm2) 2 – medium (550 mm2 to 700 mm2) 3 – large (⬎700 mm2)
Length of the inflorescence – Fl_l 1 – short (⬍2.5 cm) 2 – medium (2.5 cm to 3.5 cm) 3 – long (⬎3.5 cm)
Length and width – L_ll 1 – short and wide (⬍65 mm; ⬎13 mm) 2 – short and narrow (⬍65 mm; ⬍13 mm) 3 – long and wide (⬎65 mm; b ⬎ 13 mm) 4 – long and narrow (⬎65 mm; b ⬍ 13 mm)
Flowers number/ inflorescence– Fl_n 1 – reduced (⬍16) 2 – medium (16 to 20) 3 – abundant (⬎20)
Surface – L_lsurf 1 – flat 2 – with curly border 3 – curly Consistency – L_cons 1 – flaccid 2 – medium 3 – rigid
where dmax is the observed maximum distance between two samples of the original data set. Using this similarity definition, the most similar pair of samples have S ⫽ 1 and the most dissimilar pair of samples S ⫽ 0.
Apex angle – L_aa 1 – open 2 – medium 3 – closed
4.2.2 Non-linear Map (NLM)
Base angle – L_ba 1 – open 2 – medium 3 – closed
Graphical representations are the most common tool to analyze and explore data. If there are only two or three variables per sample the data can be displayed as points in a two- or three-dimensional plot – one axis per variable. The analysis of these graphs shows in a straightforward way the most similar samples (nearest points – relatively small dij and close to one Sij), the most dissimilar (furthest points – relatively high dij and close to zero Sij), and if there are clusters of points that reveal the presence of classes of samples.
Size of the swollen bud – Fl_b 1 – small 2 – medium 3 – large Frequency of supernumerary flowers – Fl_f 1 – nil (0) 2 – rare (⬍4) 3 – frequent (⬎4)
If the number of variables is higher than three a strategy should be used in order to obtain a two-dimensional representation of the original hyperspace, which can be achieved by non-linear mapping (Muhammad et al., 1986; Lavine, 1992; Vandeginste et al., 1998b). If the distances between the data points in the non-linear map
36
SECTION | I The Plant and Production
(NLM) is dij* tries to mimic the original distances dij, they can be obtained by minimization of the following error function (E): E⫽
∑ (i⬎ j) (dij − d jj* )2 / dij2
(4.3)
The objective of the NLM is to obtain a graph that shows, as reliably as possible, the raw structure of the data.
4.2.3 Hierarchical Cluster Analysis (HCA) Like NLM, cluster analysis techniques, and particularly the hierarchical methods, are used to obtain information about the structure of multivariate data (Muhammad et al., 1986; Lavine, 1992; Vandeginste et al., 1998b). The starting point of HCA is the similarity data calculated between all samples using Eq. (4.2) (similarity matrix) and the detection of the smallest Sij. The two data points corresponding to this maximum similarity are combined (clustered) to form a new point and the similarity matrix recomputed. This clustering process is repeated until every point has been linked. There are several different methods of HCA that show differences in the way the new distance among clusters is calculated. For example, in the nearest neighbor method the new distance between two clusters corresponds to the shortest distance between two points, each one in a different cluster. The clustering process can be visualized in the form of a dendogram, and clusters of similar samples (corresponding to a defined similarity) can be detected.
4.2.4 Linear Discriminant Analysis (LDA) NLM and HCA are unsupervised pattern recognition techniques because data are analyzed without any previous information about the existence of classes of samples (or about the intrinsic structure of the data matrix). When this type of information exists and the objective is the development of classification rules, supervised pattern recognition techniques, like LDA, should be used. LDA is used to find explicit boundaries between classes and to evaluate the discriminating capacity of the variables (Muhammad et al., 1986; Lavine, 1992; Vandeginste et al., 1998a).
4.3 OLIVE CULTIVARS DATA Morphological data of 22 olive cultivars were obtained from Leitão et al. (1986). The olive cultivars under analysis are (cultivar code and region of origin under parenthesis): Galega (var1 – dispersed all over Portugal), Carrasquenha (var2 – Alentejo), Redondil (var3 – Elvas region), Azeiteira (var4 – Elvas and Campo Maior region), Blanqueta (var5 – Spain), Conserva de Elvas (var6 – Elvas region), Negrita
(var7 – northeast Trás-os-Montes region), Madural (var8 – Trás-os-Montes region), Cobrançosa (var9 – Trás-os-Montes region), Verdeal Transmontana (var10 – Trás-os-Montes region), Redondal (var11 – Trás-os-Montes region), Galega Grada de Serpa (var12 – Serpa-Moura region), Cordovil de Serpa (var13 – Serpa-Moura region), Verdeal Alentejana (var14 – Serpa-Moura region), Cordovil de Castelo Branco (var15 – Beira-Interior region), Bical de Castelo Branco (var16 – Beira-Interior region), Maçanilha Algarvia (var17 – Algarve region), Maçanilha Carrasquenha (var18 – Alentejo region), Picual (var19 – Spain), Maçanilha (var20 – Spain), Hojiblanca (var21 – Spain) and Gordal (var22 – Spain). Tables 4.1 and 4.2 show the morphological variables under analysis. The 40 variables were qualitative, or they were categorized into qualitative features, and one integer number (1, 2, etc.) was assigned to each feature as shown in Tables 4.1 and 4.2. A small number of olive cultivars show some variables with mixed features and, when this was observed, the average of the integer numbers assigned to the detected features was used. Experimental data were assembled into a 22 samples ⫻ 40 variable table (global data set). The cultivars Bical de Castelo Branco and Maçanilha Carrasquenha have no Flowers data. Each sample corresponds to an olive cultivar and the correspondent Fruit (17 variables), Endocarp (8 variables), Trees (2 variables), Branches (3 variables), Flowers (4 variables) and Leaves (6 variables) feature (Tables 4.1 and 4.2). The data analysis focused on four data sets (cultivars ⫻ variables): (i) Fruit data set (22 ⫻ 17); (ii) Endocarp data set (22 ⫻ 8); (iii) Fruit ⫹ Endocarp data set (22 ⫻ 25); (iv) Trees ⫹ Branches ⫹ Leaves ⫹ Flow ers data set (20 ⫻ 15) (this data set will be named Trees data set). NLM (ALSCAL algorithm), HCA and LDA calculations were done on raw data using SPSS 16.0 for Windows (SPSS Inc., Chicago, USA).
4.4 UNSUPERVISED CLASSIFICATION 4.4.1 Non-linear Maps (NLM) Figure 4.1 shows the NLM of the Fruit (Figure 4.1A), Endocarp (Figure 4.1B) and Trees (Figure 4.1C) data sets. All data sets under analysis allow discrimination among olive cultivars but, when olives quality assessment is the objective, only the Fruit and/or Endocarp data sets are of practical usefulness. The analysis of the three NLM in Figure 4.1 shows the existence of clusters of similar cultivars – the point that represents one cultivar is close to others and several clusters of points are detected. The borders represented in the NLM highlight the clusters of similar olive cultivars and the comparison of the clusters composition in the three NLM show that they are not identical. However,
37
CHAPTER | 4 Chemometric Classification of Cultivars of Olives: Perspectives on Portuguese Olives
VAR1
2
VAR8
VAR15
VAR5
1 Dimension 2
VAR18
VAR19
VAR9
VAR11 VAR4
VAR21
0
VAR10
VAR20
VAR12 VAR7
VAR14
VAR16
VAR13 VAR22
VAR2
−1
VAR6 VAR3 VAR17
−2 −3
−2
−1
A
0
1
2
Dimension 1
3 VAR17
2 Dimension 2
VAR15 VAR16
1
VAR22
VAR11 VAR2
VAR3 VAR14
VAR19
0
VAR6
VAR21 VAR9
VAR5
VAR13
−1
VAR20 VAR18 VAR10 VAR4 VAR8
VAR7
VAR1 VAR12
−2 −3
−2
−1
B
0
1
2
Dimension 1
1.5 VAR5
VAR16 VAR19
1.0
VAR2
Dimension 2
VAR20
0.5
VAR6 VAR3
0.0
VAR9 VAR10 VAR11
VAR17
VAR13 VAR18
−0.5
VAR8
VAR12
VAR19
VAR14 VAR7
−1.0
VAR1 VAR4
−1.5 −3
−2
−1
0
1
2
Dimension 1 C FIGURE 4.1 Non-linear maps of the cultivars using the Fruit (A), Endocarp (B) and Tree ⫹ Branches⫹ Leaves ⫹ Flowers (C) data sets.
38
SECTION | I The Plant and Production
FIGURE 4.2 Dendograms obtained using the Wards method of the hierarchical cluster analysis of the cultivars using the Fruit (A), Endocarp (B) and Fruit ⫹ Endocarp (C) data sets.
a detailed analysis of the NLM should be done with caution because they are a bidimensional projection of the real hyperspace graph. Indeed, the error functions (Young’s S-stress formula) of NLM of Figure 4.1 are: Figure 4.1A, S-stress ⫽ 0.19644; Figure 4.1B, S-stress ⫽ 0.21118; Figure 4.1C, S-stress ⫽ 0.21980. These error values show that the NLM of Figure 4.1 are only rough estimations of the relative distribution of the cultivars in the corresponding
hyperspace. Although the NLM clearly show that the olive cultivars show tendency to cluster, a more rigorous analysis of the cluster composition should be done with HCA.
4.4.2 Hierarchical Cluster Analysis (HCA) Figures 4.2 and 4.3 show the dendograms of the Fruit (Figure 4.2A), Endocarp (Figure 4.2B), Fruit ⫹ Endocarp
CHAPTER | 4 Chemometric Classification of Cultivars of Olives: Perspectives on Portuguese Olives
39
FIGURE 4.3 Dendograms obtained using the Wards method of the hierarchical cluster analysis of the cultivars using the Tree ⫹ Branches ⫹ Leaves ⫹ Flowers data set.
TABLE 4.3 Cultivars that constitute the classes detected by hierarchical cluster analysis. This table presents the classes of similar olive cultivars detected by hierarchical cluster analysis. Class
Data set Fruit
Endocarp
Fruit ⫹ Endocarp
Tree ⫹ Branches ⫹ Leaves ⫹ Flowers
1
Azeiteira Negrinha Redondal Maçanilha Galega Grada de Serpa Cordovil de Serpa Blanqueta Madural Galega
Azeiteira Verdeal Transmontana Verdeal Alentejana Cordovil de Serpa Negrinha Madural
Azeiteira Negrinha Madural Cordovil de Serpa Galega Galega Grada de Serpa
Redondil Maçanilha Cordovil de Serpa Galega Grada de Serpa
2
Cobrançosa Picual Hojiblanca Conserva de Elvas Maçanilha Carrasquenha Cordovil de Castelo Branco
Maçanilha Carrasquenha Maçanilha Redondil Galega Galega Grada de Serpa
Redondil Maçanilha Maçanilha Carrasquenha Maçanilha Algarvia
Verdeal Transmontana Hojiblanca Redondal Cobrançosa
3
Redondil Maçanilha Algarvia Gordal
Carrasquenha Redondal Hojiblanca Cordovil de Castelo Branco Bical de Castelo Branco Maçanilha Algarvia
Verdeal Transmontana Verdeal Alentejana Carrasquenha Cordovil de Castelo Branco Bical de Castelo Branco
Galega Verdeal Alentejana Carrasquenha Blanqueta
4
Carrasquenha Verdeal Transmontana Verdeal Alentejana Bical de Castelo Branco
Blanqueta Cobrançosa Picual Conserva de Elvas Gordal
Blanqueta Redondal Hojiblanca Cobrançosa Picual Conserva de Elvas Gordal
Conserva de Elvas Picual Madural Cordovil de Castelo Branco Gordal Maçanilha Algarvia
5
Azeiteira Negrinha
40
(Figure 4.2C) and Trees (Figure 4.3) data sets. The analysis of these figures shows that most olive cultivars cluster at distances close to zero (or similarities close to one) and, at a rescaled distance of about ten, four and five clusters are clearly detected for the Fruit and/or Endocarp data sets (Figure 4.2) and for the Trees (Figure 4.3) data set, respectively. Table 4.3 presents the olive cultivars that compose the detected classes. Each class is composed by olive cultivars that are similar taking into consideration the variables used in the characterization and, as observed in Table 4.3, different data sets originate classes with different olive cultivars. The qualitative composition of the classes obtained using the fruit and/or endocarp and tree, branches, leaves and flowers characteristics are particularly different. This result shows that there is a quite small correlation between these two data sets and that the quality of the fruit cannot be easily predicted from the measured olive trees characteristics. The analysis of the geographic origin of the cultivars belonging to each class of Table 4.3 reveals that there is no regional clustering. This observation shows that the cultivars from a particular DOP can easily be discriminated using the fruit or trees characteristics under analysis. The HCA of the fruit and trees characteristics showed that the cultivars Azeiteira and Negrinha should be very similar because they are always in the same class and, when trees characteristics are under analysis these two cultivars form one class (Table 4.3). Indeed, this result was expected (Leitão et al., 1986) because the Azeiteira cultivar from the Elvas-Campo Maior region (south of Portugal) seems to be a variation of the Negrinha cultivar from the Trás-os-Montes region (north of Portugal). Nevertheless, as shown in Figure 4.2B, these two cultivars can be easily discriminated by HCA of the endocarp characteristics. Taking into consideration that the data sets Fruit ⫹ Endocarp and Trees originated NLM and dendograms with well-defined classes, the discrimination capacities of the characteristics that compose these data sets will be analyzed by LDA.
4.5 LINEAR DISCRIMINANT ANALYSIS (LDA) In order to assess the discriminating capacity of the Fruit ⫹ Endocarp data set to the detected four classes (Table 4.3) of olive cultivar the Wilks’ Lambda and F-tests were performed (Table 4.4). The smallest value of Wilks’ Lambda and the highest F-tests are the highest discriminating capacity of the variable (Pinheiro and Esteves da Silva, 2005). The analysis of Table 4.4 shows that either the Fruit or Endocarp variables contribute markedly to the olive cultivars classification. The most discriminating variables (F-test ⬎3) are the following: shape, nipple, size of the pedicel cavity, color of epicarp at maturity, lenticels size,
SECTION | I The Plant and Production
TABLE 4.4 Wilks’ Lambda and F-tests of group means for the linear discriminant analysis of the Fruit ⫹ Endocarp data set. This table shows that the most discriminating variables are: lenticels size (F_lenss), color of epicarp at maturity (F_bcm), shape (F_shape), mesocarp consistency (F_mc), nipple (F_nipple) and size of the pedicel cavity (F_sds) of the fruit; shape (E_shape), surface (E_surf), base (E_base), apex termination (E_apexts), size (E_size) and position of maximum transverse diameter (E_pmtd) of the endocarp. Variable type
Variable
Wilks’ Lambda
F-test
Fruit
F_lenss F_bcm F_shape F_mc F_nipple F_sds F_sdd F_base F_ma F_size F_bct F_apex F_pmtd F_sdsh F_pc F_lensd
0.229 0.480 0.491 0.494 0.538 0.572 0.699 0.727 0.736 0.761 0.804 0.834 0.841 0.914 0.919 0.993
20.217 6.510 6.222 6.149 5.146 4.483 2.583 2.283 2.152 1.888 1.462 1.194 1.135 0.562 0.527 0.043
Endocarp
E_shape E_surf E_base E_apexts E_size E_pmtd E_sl E_apex
0.327 0.418 0.477 0.589 0.605 0.648 0.720 0.785
12.328 8.354 6.566 4.182 3.912 3.254 2.330 1.641
mesocarp consistency of the fruit; size, shape, base, apex termination, position of maximum transverse diameter and surface of the endocarp. Although previous works have suggested that the endocarp morphological variables would play an important role in the classification of olive cultivars (Leitão et al., 1986), the results obtained in this work confirm previous observations that the incorporation of both endocarp and fruit morphological variables improves the discrimination (Pinheiro and Esteves da Silva, 2005). The LDA of the Fruit ⫹ Endocarp data set using the 12 most discriminating variables (F-test ⬎3) allows the discrimination of the previously detected four olive cultivar classes (Figure 4.4). The analysis of Figure 4.4 also shows that olive cultivars that compose classes three (Verdeal Transmontana, Verdeal Alentejana, Carrasquenha, Cordovil de Castelo Branco e Bical de Castelo Branco) and four
41
CHAPTER | 4 Chemometric Classification of Cultivars of Olives: Perspectives on Portuguese Olives
TABLE 4.5 Wilks’ Lambda and F-tests of group means for the linear discriminant analysis of the Trees ⫹ Branches ⫹ Leaves ⫹ Flowers data set.
10
This table shows that the most discriminating variables are: size of the tree (T_size); length and width (L_ll), base angle (L_ba) and size (L_ls) of the leaves; length of the inflorescence (Fl_l) and frequency of supernumerary flowers (Fl_f).
5
Function 2
4
3
0 1 2
−5
Class_F_E 1
−10 −10
−5
2
3 0
4
Variable
Wilks’ Lambda
Trees
T_size T_ar
0.519 0.787
3.477 1.017
Branches
B_r B_c B_b
0.774 0.914 0.915
1.096 0.353 0.348
Leaves
L_ll L_ba L_ls L_aa L_lsurf
0.060 0.237 0.344 0.625 0.876
58.828 12.104 7.153 2.254 0.530
Flowers
Fl_l Fl_f Fl_b Fl_n
0.172 0.280 0.570 0.814
18.091 9.659 2.828 0.855
Group centroid 5
10
Function 1 FIGURE 4.4 Linear discriminant plots of the classes detected by hierarchical cluster analysis of the Fruit ⫹ Endocarp data using the most discriminating features (F-test ⬎ 3). Class number is defined in Table 4.3.
(Blanqueta, Redondal, Hojiblanca, Cobrançosa, Picual, Conserva de Elvas and Gordal) have well-defined features (they are quite well separated in the plot) while olive cultivars that compose classes one (Azeiteira, Negrita, Madural, Cordovil de Serpa, Galega and Galega Grada de Serpa) and two (Redondil, Maçanilha, Maçanilha Carrasquenha and Maçanilha Algarvia) are relatively similar, because they somewhat overlap in the LDA plot. This result shows that mixing a small number of fruit and endocarp morphological parameters, and analyzing the data table with multivariate data analysis techniques, shows great potential for the development of olive cultivar classification rules. In order to assess the discriminating capacity of the Trees data set to the detected five classes (Table 4.3) of olive cultivar the Wilks’ Lambda and F-tests were performed (Table 4.5). The analysis of Table 4.5 shows that some leaves and flowers morphological variables contribute markedly to the olive cultivars classification and the trees and branches variables have no or small discrimination ability. The most discriminating variables (F-test ⬎3) are the following: length and width, base angle and size of the leaves; length of the inflorescence and frequency of supernumerary flowers.
4.6 CONCLUSIONS The chapter showed that from morphological features of olives (fruit and endocarp) and olive trees (including trees, branches, leaves and flowers) coupled to chemometric multivariate classification techniques, classes of olives from
F-test
Variable type
different cultivars are naturally observed. Mixing morphological features from olive fruit and endocarp, classification rules with potential to be applied in the quality control of olives and, consequently, in the DOP assessment can be developed. Also, it was observed that olive trees can be characterized using classification rules based on leaves and flowers features.
SUMMARY POINTS ●
●
●
Chemometric classification of olive cultivars originated classes of similar morphologic features of olives. Olive fruit and endocarp characteristics have enough discrimination capacity to allow olive cultivar classification. Olive trees can be classified according to leaves and flowers characteristics.
REFERENCES Bartolini, G., Petruccelli, R., 2002. Classification, Origin, Diffusion and History of the Olive. Food and Agriculture Organization of the United Nations, Rome. Cantini, C., Cimato, A., Sani, G., 1999. Morphological evaluation of the olive germplast present in Tuscany region. Euphytica 109, 173–181.
42
Casa do Azeite, 2008. www.casadoazeite.pt. (1/April/2008). European Community (EC), 1992. Official Journal of the Commission of European Communities, Regulation Nº 2081/92, July 14. European Community (EC), 1993. Official Journal of the Commission of European Communities, Regulation Nº 2037/93, July 27. Ganino, T., Bartolini, G., Fabbri, A., 2006. The classification of olive germplast – A review. J. Hort. Sci. Biotechn. 81, 319–334. Gemas, V.J.V., Almadanim, M.C., Tenreiro, R., Martins, A., Fevereiro, P., 2004. Genetic diversity in the Olive tree (Olea europaea L. subsp. Europaea) cultivated in Portugal revealed by RAPD and ISSR markers. Genet. Resour. Crop. Ev. 51, 501–511. Lavine, B.K., 1992. Signal processing and data analysis. In: Haswell, S.J. (ed.), Practical Guide to Chemometrics. Marcel Dekker, Inc., New York, pp. 211–269. Leitão, F., De Fátima Potes, M., Leonilde Calado, M., José de Almeida, F., 1986. Descrição de 22 variedades de oliveira cultivadas em Portugal. Ministério da Agricultura, Pescas e Alimentação, Direcção Geral de Planeamento e Agricultura, Lisboa. Mafra, I., Lanza, B., Reis, A., Marsilio, V., Campestre, C., De Angelis, M., Coimbra, M.A., 2001. Effect of ripening on texture, microstructure and cell wall polysaccharide composition of olive fruite (Olea europaea). Physiol. Plantarum 111, 439–447.
SECTION | I The Plant and Production
Mafra, I., Barros, A.S., Coimbra, M.A., 2006. Effect of black oxidising table olive process on the cell wall polysaccharides of olive pulp (Olea europaea L. var Negrinha do Douro). Carboh. Polym. 65, 1–8. Muhammad, A.S., Illman, D.L., Kowalski, B.R., 1986. Chemometrics. Wiley, New York. Pinheiro, P.B.M., Esteves da Silva, J.C.G., 2005. Chemometric classification of the biometric parameters of olives from three Portuguese cultivars of Olea europaea L. Anal. Chim. Acta. 544, 229–235. Rotondi, A., Magli, M., Ricciolini, C., Baldoni, L., 2003. Morphological and molecular analysis for the characterization of a group of Italian olive cultivars. Euphytica 132, 129–137. Vandeginste, B.G.M., Massart, D.L., Buydens, L.M.C., De Jong, S., Lewi, P.J., Smeyers-Verbeke, L., 1998a. Handbook of Chemometrics and Qualimetrics: Part A. Elsevier, Amsterdam. Vandeginste, B.G.M., Massart, D.L., Buydens, L.M.C., De Jong, S., Lewi, P.J., Smeyers-Verbeke, L., 1998b. Handbook of Chemometrics and Qualimetrics: Part B. Elsevier, Amsterdam. Zamora, R., Alaiz, M., Hidalgo, F.J., 2001. Influence of cultivar and fruit ripening on olive (Olea Europaea) fruit protein content, composition, and antioxidant activity. J. Agric. Food Chem. 49, 4267–4270.
Chapter 5
Effect of Climatic Conditions on Quality of Virgin Olive Oil María-Paz Romero and María-José Motilva Department of Food Technology CeRTA-TPV, Escuela Téccnica Superior de Ingeniería Agraria, Universidad de Lleida, Lleida, Spain
5.2 THE EFFECTS OF CLIMATE CONDITIONS ON QUALITY OF OLIVE OIL
5.1 INTRODUCTION The Protected Designation of Origin (PDO) ‘Les Garrigues’ is located in southern Lleida, a province of north-eastern Spain in the autonomous community of Catalonia. The climate of the area is continental Mediterranean, characterized by warm summers and long, cold winters. The important thermal difference between winter and summer, with annual minimum temperatures of ⫺8°C and maximum temperatures above 40°C, distinguishes this area from other olive-growing areas with a more temperate climate due to their proximity to the Mediterranean Sea. Rainfall in the area is scarce and very irregular, being more abundant in the spring and lowest in summer (July) and winter (January and February). The rainfall pattern differs from one year to another, a feature of the Mediterranean climate. Arbequina is the native olive tree cultivar in this area, and it is characterized by frost resistance, low vigor, small fruit size, and high productivity. Oil obtained from Arbequina is well known in the international oil market for its excellent taste and flavor. The olive harvest period in this area takes place from November to January, with the aim of finishing harvesting before the arrival of the frosts that are very common in this region. The Oil Laboratory of the Department of Food Technology has monitored olive oils from PDO Les Garrigues since 1995 and has analyzed more than 2000 olive oil samples and obtained more than 500 olive oils from the pilot plant. Throughout these years there has been a wide range of climatic conditions that have permitted the study of the effects of climatic conditions on virgin olive oil composition and quality indices as well as effects of damage to oil quality due to the freezing of olives, a cause of concern in this area where high-quality oils are produced. Of special relevance are the changes in minor components, such as chlorophylls, carotenoids and phenolic compounds and the sensory attributes that play an important role in the organoleptic characteristics and antioxidant properties of virgin olive oils. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Olive oil composition has been studied for Arbequina cultivar over nine years to develop an understanding of seasonal variation.
5.2.1 Influence of Crop Season on the Composition of Virgin Olive Oil The first experiment was designed to characterize oils from several locations in the PDO Les Garrigues. A total of 130 olive oils from Arbequina cultivar obtained in 13 industrial oil mills were analyzed through two consecutive seasons. Results had shown that oils from the same crop season and variety, extracted with similar technologies but obtained from olive trees grown at different locations inside this area, could present important differences in composition. The traditional oil quality parameters like peroxide value, acidity and K270 of olive oils did not present differences within crops seasons, however an important difference has been observed in the fatty acids profile. Taking the most extreme observations for location variable (Table 5.1) these differences could be over 5% for oleic acid. The differences for fatty acid profile between locations have been observed every crop season. In general, oils from location A contain more oleic acid than those from location B. Agroclimatic studies could explain the differences. So, Llasat (1997) splits the PDO Garrigues area into three clusters and while location A belongs to the coldest cluster, with the lowest minimum and maximum temperatures, location B belongs to a hot area with long summers. According with other studies (Aguilera et al., 2005) the high temperature and low altitude could be associated with an increased polyunsaturation and the lowest oleic percentage. A complete analysis was published in Motilva et al. (2001). A second experience was designed from 1996 to 1999. These years presented several differences in patterns 43
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
44
SECTION | I The Plant and Production
TABLE 5.1 Percentage of several fatty acids in two Garrigues locations. This table summarizes the differences in the main fatty acids of olive oil. Each value is the mean of fatty acid concentration in five olive oils. Crop season
1995/96
Fatty acid
Palmitic acid
1996/97 1995/96
Oleic acid
1996/97 1995/96 1996/97
Linoleic acid
Location A
Location B
%
Standard error
%
Standard error
11.6
0.2
14.3
0.2
11.9
0.3
14.4
0.4
75.9
0.3
70.0
0.2
76.3
0.4
69.4
0.4
8.7
0.1
11.2
0.2
8.3
0.1
11.5
0.2
for air temperatures and rainfall. The weather in 1996 was characterized by a very dry summer and occasional frost in November and December. Total cumulative rainfall in 1996 was 513 mm, markedly higher than the average for this area, and accumulated at harvest period. The crop season in 1997/98 was characterized by low temperatures and less sunshine than other seasons and there was also an excess of rain during the maturation period. The main climatic characteristics in 1998 were the persistent frost during November and December and the scarce rainfall, especially during the summer. Accumulative rainfall in this year was 342 mm, 64 mm of which fell in summer. The weather in 1999 was characterized by very severe frost (below ⫺5°C) and a total absence of rainfall during the harvest period (November and December). Maximum accumulated rainfall was during September (total accumulated rainfall in this year was 397 mm). Oils from the 1997/98 season showed a significant difference ( p ⬍ 0.001) in fatty acid composition in relation to oils from the other seasons studied (Table 5.2). The percentage of estearic and oleic acids were lowest, and on the other hand palmitic, palmitoleic and linoleic acids were highest. This could be attributed to a modification of lipid biosynthesis coinciding with the wet summer. The metabolism and lipid levels of the olive fruit are affected by environmental factors, such as light, temperature, and water stress, and it has been observed by some authors that oleic acid, some triacylglycerides, and the oleic/linoleic acid ratio are connected to the rainfall in the summer period. The rainfall regime in 1997 could have affected the de novo fatty acid biosynthesis that occurs in plant plastids and that needs the concerted activity of two enzymes, acetyl-CoA carboxylase and fatty acid synthase, to regulate a further chain elongation cycle in de novo fatty acid biosynthesis. This step is particularly relevant because it determines the C16/C18 ratio, and this is directly related to the degree of unsaturation of the final oil product. The highest C16/C18
ratio was observed in oils from 1997/98. This ratio seems to be related to the accumulated rainfall regime in the summer of each year. In considering the accumulated rainfall during June–August, the maximum corresponds to 1997, with 167 mm, in front of 39, 33 and 29 mm for 1996, 1998 and 1999, respectively. These climatological conditions could cause the modification in the degree of unsaturation of the oils. Also the influence of crop season on color has been studied. The color of the oil affects the consumer’s perception of quality. It is related to pigment content and decreases with the olives’ ripeness. In order to analyze the influence of climatic conditions, the oils analyzed were grouped together by harvest period, first and second. The effect of harvest period on the loss of chlorophyll and carotenoid pigments does not show the same pattern in every crop season (Figure 5.1). Thus, the main differences were observed in oils from the 1997/98 and 1999/00 seasons. This is probably a consequence of frosts in November 1997 and 1999 that led to deterioration of the olive fruit and pigment degradation, mainly in the chlorophyll fraction. Oils from the 1998/99 crop season showed no significant differences between the first and second harvest periods. The minimum air temperature during the harvest period (November 1998) remained around 0°C without heavy frosts that could have produced pigment degradation in the olive fruit. Oils from the 1999/00 crop season showed the lowest chlorophyll content. As mentioned above, the 1999 harvest period was characterized by frequent, heavy frosts that may have initiated the degradation of the olive pigments, mainly the chlorophyll fraction. Thus, with respect to pigment content the main effect was the minimum air temperature during the harvest period (November– December); rainfall regime was a secondary effect. The amount of phenolic compounds in virgin olive oil is another important factor when evaluating its quality,
CHAPTER | 5
45
Effect of Climatic Conditions on Quality of Virgin Olive Oil
TABLE 5.2 Influence of the crop season on fatty acid profile (%) of monovarietal virgin olive oil from Arbequina cultivar at two locations. This table shows the profile of some fatty acid in two of 13 locations studied. Each value is the mean of fatty acid concentration of five olive oils. Fatty acid (%)
Palmitic acid
1996/97
1997/98
1998/99
1999/2000
Location A
Location B
Location A
Location B
Location A
Location B
Location A
Location B
11.9
14.4
14.5
15.7
11.6
14.3
11.7
14.2
Palmitoleic acid
1.20
1.41
1.4
1.7
1.1
1.5
1.0
1.4
Estearic
1.68
2.09
1.6
1.6
2.1
2.2
2.0
2.1
Oleic acid
76.3
69.4
71.1
67.9
75.9
70.0
76.0
70.1
Linoleic acid
8.3
11.5
10.7
12.3
8.7
11.2
8.3
11.0
C16/C18 ratio
0.152
0.191
0.191
0.213
0.146
0.189
0.147
0.188
C16 is palmitic and palmitoleic acids and C18 is estearic, oleic, linoleic and linolenic acids.
Pigments content (mg kg−1) 14
Total phenols (mg kg−1)
12
250
10
200
8
150
6
100
4
50
2
0
300
1996/97 0
1996/97
1997/98
1998/99
1999/00
FIGURE 5.1 Oil pigments content (chlorophyll and carotenoid) in relation to crop season and harvest period. This figure shows the average value for pigment content. Chlorophylls are expressed as mg of pheophytin kg⫺1 of oil and carotenoids are expressed as mg of luteolin kg⫺1 of oil. The order of serials is: chlorophyll first harvest, chlorophyll second harvest, carotenoid first harvest and carotenoid second harvest.
given that the natural phenols improve its oxidative stability and, to a certain extent, are responsible for its bitter taste. The total phenols in the oils analyzed in this study varied considerably and a significant ( p ⬍ 0.001) effect of the season could be observed (Figure 5.2). Thus, oils from the 1997/98 crop season had the lowest total phenol concentration (106 to 84 mg kg⫺1), and the highest values were in oils from the 1996/97 and 1998/99 seasons (from 272 to 215 and from 242 to 172 mg kg⫺1 for each year and harvest period, respectively). A relationship was observed between the total phenols content of the oil and accumulated rainfall by year. Thus, the weather in 1997, which corresponded to the year with the lowest polyphenol content in the olive oils, was characterized
1997/98
1998/99
1999/00
FIGURE 5.2 Total phenol content (mg kg⫺1 as caffeic acid) in relation to crop season and harvest period. The first column corresponds to first harvest, and the second one to second harvest.
by nearly 400 mm accumulated rainfall during the summer period (June–August), above the average for the area. In contrast, during this period, there was very low rainfall accumulation in 1996 and 1998, and the oils from these years showed the highest polyphenol contents. It has long been known that the level of phenolics in plant tissues can be influenced by environmental factors such as ambient temperature and water availability. With regard to the latter factor, a water deficiency generates a stress situation that induces the production of phenolics, and this factor could be related to the increase in the polyphenol content of the oils from the 1996/97 and 1998/99 seasons. Oils from the second period of harvest in 1999/00 season showed a significant decrease in phenolic compounds that could be attributed to frost. Bitter taste is one of the characteristic attributes of virgin olive oil. In our study, the bitter index (K225) was analyzed in oils from the 1998/99 and 1999/00 crop seasons with values from 0.19 to 0.15 and from 0.17 to 0.12 in
46
the first and second harvest period and each crop season, respectively. There were no differences in K225 between seasons, and the effect of harvest period was only significant ( p ⬍ 0.001) during the 1999/00 season, similar to that observed in oil phenol content. The mentioned decrease in phenols gave rise to a loss in bitterness. Gutiérrez et al. (1992) suggested that K225 values of the order of 0.14 or lower correspond to oils with slight bitterness intensity, corresponding to oils from the area of the PDO ‘Les Garrigues’, and values close to 0.36 correspond to quite bitter oils.
5.2.2 Effect of Freeze Injuries in Olive on Virgin Olive Oil Composition The harvest period for the 2001–02 crop season is remarkable because a cold front from Siberia moved to the south of Europe and arrived in Catalonia (Spain) on 15 December bringing snowfalls, together with an important drop in temperature. Trees were covered by snow for more than 10 days because of the persistent cold. Climatic data obtained from the weather station situated at La Granadella, in the geographical center of ‘Les Garrigues’, showed that maximum temperatures were below ⫺5°C for more than 110 consecutive hours, reaching minimum values of ⫺12.5°C. As a consequence of these exceptional weather conditions, about 40% of Arbequina olive oil production in the PDO ‘Les Garrigues’ was affected and negative effects have also been forecast for the coming crop seasons as a result of the high number of olive trees damaged by the extreme temperatures. These climatic conditions allow us to analyze oils derived from olives growing in the same orchard before and after the frost. The values for the quality indices, free fatty acid content and peroxide value of after-frost oils showed slightly higher values. In fact, they are related to the degree of deterioration of the oil, determined by the state of the fruit and its manipulation before and during oil processing. However, it is not significant because the average values of free fatty acid content, peroxide value and K270 were considerably below the limit established by EEC legislation for virgin olive oil. A possible explanation for the low increase observed in the peroxide value of oils obtained after frost could be that the olives were harvested and processed within 24 h after thaw, as soon as it was possible to harvest them. Neither α-tocopherol content nor fatty acid profile have shown significant differences between oils obtained before and after freeze damage. However, the chlorophyll and carotenoid concentration and chromatic ordinates of oils obtained before and after frosts were different. Oil pigment content decreased slightly after freeze damage ( p ⬍ 0.05). The chlorophyllase and lypoxygenase enzymes could be involved in this loss of chlorophyllic and carotenoid pigments, favored by deterioration of the olive fruit (Mínguez-Mosquera, 1997). The chromatic ordinates a* and b* were not affected by the
SECTION | I The Plant and Production
low temperatures reached while luminosity (L*) negatively correlated with total pigment content, and increased in oils obtained after freeze damage. In addition to pigment differences, a remarkable difference between oils obtained before and after frost was the total phenol content, showing significant decrease in those oils obtained from fruits under freeze conditions. The olive tissue destruction caused by the ice crystals formed inside parenchyma cells may encourage the oxidative degradation of phenolic compounds in PPO-catalyzed reaction. It has been reported that the main agent responsible for enzymatic browning in fruits and vegetables is polyphenol oxidase (PPO; 1.14.18.1). It catalyzes two different reactions in the presence of molecular oxygen: the hydroxylation of monophenols to o-diphenols and the oxidation of o-diphenols to o-quinones which polymerize non-enzymatically and give rise to heterogeneous black, brown or red pigments, commonly called melanins (Tomás-Barberán and Espín, 2001). The bitter index (K225) and oxidative stability of oils followed the same pattern as total phenol content. Figure 5.3 shows the chromatographic profile of the phenolic extracts from Arbequina virgin olive oils from the same mill but obtained before and after frost, respectively. Table 5.3 shows the average concentrations of the quantified phenolic compounds in the oils examined in this study. Peaks 1–8 and Peak 10 were identified according to a previous article (Tovar et al., 2001). There is a larger number of compounds in the phenolic fraction than in those already defined. Peaks 9, 11 and 12 represent unknown complex phenolic compounds found in all olive oils analyzed, with a spectrum similar to that of secoiridoid derivatives, showing two maxima at 210 and 278 nm. Peak 13, however, shows a UV spectrum similar to that of trans-cinnamic acid, with a maximum at 276 nm. The main phenolic compounds found in oils from PDO ‘Les Garrigues’ analyzed in this trial were the secoiridoid derivatives, a dialdehydic form of elenolic acid linked to hydroxytyrosol (3,4-DHPEA-EDA), oleuropein aglycone (3,4-DHPEA-EA) and the dialdehydic form of elenolic acid linked to tyrosol (p-HPEA-EDA), followed in order of quantitative importance by lignans and the phenolic compound 4-(acetoxyethyl)- 1,2-dihydroxybenzene (3,4-DHPEA-AC). Low levels of the simple phenols 3,4-DHPEA, p-HPEA, vanillin and vanillic acid were found, while ferulic and pcoumaric acids were identified but could not be quantified. The concentrations of most olive oil phenolic compounds were affected by the freeze conditions. The three secoiridoid derivatives and the compound 3,4-DHPEA-AC decreased in the oils obtained after the freeze damage suffered by the olive fruit. Secoiridoid derivatives are compounds of major interest since their antioxidant activity has already been evaluated and has been shown to extend the shelf-life of olive oil (Baldioli et al., 1996; Gennaro et al., 1998). Good correlations have also been found between them and the bitter index (Tovar et al., 2001). The concentration of the unknown phenolic compounds, Peak 9 and Peak 12,
CHAPTER | 5
47
Effect of Climatic Conditions on Quality of Virgin Olive Oil
0.20
0.15
7 6
8
AU 0.10 13 12 1011
5
0.05 4 2
1
9
3
0.00 10.00
20.00
30.00
40.00
50.00
60.00
50.00
60.00
Minutes
A 0.20
8 0.15
AU 0.10
12
0.05
1 10.00
2
3
13
10
4 0.00
11
7
6
9
5 20.00
B
30.00
40.00
Minutes
FIGURE 5.3 HPLC chromatograms (at 278 nm) of phenolic extracts from olive oil. (A) Before frost damage; (B) After frost damage. (Reprinted from Morelló et al., 2003, with permission). This figure showed two chromatograms of olive oil phenolic profile. See Table 5.3 to identify the peaks.
decreased, as did the earlier-mentioned compounds, which may indicate that they are biochemically related. Wounding of olive fruit, caused by the formation of ice crystals, induces cellular recompartmentalization which would allow mixing of phenolic substrates and PPO. Thus, the decrease in the level of those compounds could be due to an enzymatic oxidation that would result in the browning of the olive fruit. Despite showing a UV spectrum similar to that of Peaks 9 and 12, Peak 11 followed a different pattern. Its content in the oils increased significantly after freeze damage, as it was a compound resulting from the oxidative process. The contents of lignans and Peak 13 remained practically unchanged while the 3,4-DHPEA, vanillic acid and vanillin concentrations increased in oils obtained after freeze damage. The increase in 3,4-DHPEA content may be a consequence of the degradation of secoiridoid derivatives in their structure. On the other hand, vanillic acid and vanillin are considered to be lignin degradation compounds (Fernández-Bolaños et al., 1998) (Figure 5.4). One of the aims of this study was to examine the relationship between the sensory attributes of virgin olive oil
and the freeze conditions and to connect these to chemical composition. Differences are detected between oils in aroma, and mouthfeel perceptual differences are detected between oils. The sensory notes of artichoke, tomato and almond were perceived in before-frost oils, and no unpleasant aroma or flavor was detected. After-frost olive oils were qualified as nonextra virgin olive oil. Defects were defined as frozen olives by some panelists, while others defined these oils as thicker, softer, and with the term rancid tallow. A less bitter and pungent taste should be related to the noticeable reduction of oil phenolic compounds, especially secoiridoid compounds.
5.2.3 Influence of Seasonal Conditions on Phenolic Profile Of Olive Oils During four successive seasons, corresponding to 2000/01 to 2003/04, the effect of the climatological conditions of the olive crop season on the composition of phenolic fractions and related properties such as bitterness and oxidative stability were studied. Seventy virgin oils obtained in various industrial olive mills were analyzed. Profile of the phenolic
48
SECTION | I The Plant and Production
TABLE 5.3 Phenolic compounds (mg kg⫺1) of virgin olive oil obtained from olive fruits harvested before and after frost damage (n ⫽ 10). This table shows the concentration of phenolic compounds of olive oil before and after frost damage. Peak
Phenolic compounds
Before frost damage
After frost damage
Significance level
1
3,4-DHPEA
0.12
0.21
**
2
p-HPEA
0.95
0.42
NS
3
Vanillic acid
0.08
0.15
*
4
Vanillin
0.17
0.42
**
5
3,4-DHPEA-AC
22
1
**
6
3,4-DHPEA-EDA
116
17
**
7
p-HPEA-EDA
44
28
NS
8
Lignans
80
80
NS
9
Peak 9
5.21
*
1.20
79
33
**
22
*
10
3,4-DHPEA-EA
11
Peak 11
6.50
12
Peak 12
9.30
5.24
**
13
Peak 13
7.36
4.54
NS
*
**
Significance level: NS Non significant (p ⬎ 0.05), p ⬍ 0.05, p ⬍ 0.01. (Reprinted from Morelló et al., 2003, with permission.)
80 DHPEA-EDA
DHPEA-EA
p-HPEA-EDA
DHPEA-AC
Lignans
Others
70
Percentage (%)
60
50
40
30
20
10
0
2000/01
2001/02
2002/03
2003/04
Season
FIGURE 5.4 Percentage distribution of the main phenolic compounds on monovarietal olive oil phenolic fraction for crop seasons 2000/01 to 2003/04. This figure shows the average concentration of main secoiridoids.
fraction is characterized by low levels of simple phenols [hydroxytyrosol(3,4-DHPEA), tyrosol (p-DHPEA), vanillic acid, and vanillin], with concentrations between 0 and 2 ppm. These values are different from those obtained by García et al.
(2003) for Arbequina grown in southern Spain, which averaged about 5 ppm, and other varieties, such as Picual, Hojiblanca, and Cornicabra having values of 17, 12, and 9 ppm, respectively (García et al., 2002; Beltrán et al., 2005).
CHAPTER | 5
Effect of Climatic Conditions on Quality of Virgin Olive Oil
The main phenols present in olive oil are the 4-(acetoxyethyl)-1,2-dihydroxybenzene (3,4-DHPEA-AC) and secoiridoid derivatives, such as the dialdehydic form of elenolic acid linked to hydroxytyrosol (3,4-DHPEA-EDA); the dialdehydic form of elenolic acid linked to tyrosol (p-HPEAEDA); the aldehydic form of elenolic acid linked to tyrosol (p-HPEAEA); the aldehydic form of elenolic acid linked to hydroxytyrosol (3,4-DHPEA-EA; oleuropein aglycone) and lignans. Flavonoid aglycones such as apigenin and luteolin were also present in small amounts. Although no qualitative differences were observed in the phenolic fraction between crop seasons, quantitative differences were observed between phenolic compounds. The main differences were observed in secoiridoid derivatives (3,4-DHPEA-EDA, p-HPEA-EDA, p-HPEA-EA, and 3,4-DHPEAEA). The main phenolic compound quantified in the olive oil phenolic fraction, 3,4-DHPEA-EDA, showed a significantly different concentration between crop seasons 2000/01 and 2001/02 and crop seasons 2002/03 and 2003/04. The oils from the 2000/01 and 2001/02 crop seasons showed average concentrations of between 30 and 223 ppm, whereas the values for oils from the 2002/03 and 2003/04 crop seasons rose to between 427 and 610 ppm. In December 2001 olive drupes froze on the trees because low temperature (below ⫺5°C). The large decrease in the 3,4DHPEA-EDA concentration observed in olive oils from the second harvest period of the 2001/02 crop season may be due to this. The concentrations of the remaining secoiridoid derivatives, such as p-HPEA-EDA, p-HPEA-EA, and 3,4-DHPEA-EA, increased in the 2002/03 and 2003/04 crop seasons after the frost damage that occurred in the 2001/02 crop season. However, no significant differences in lignans (pinoresinol and acetoxypinoresinol) were observed between the first and second harvesting period during the four crop seasons studied. The percentages of the main phenols in the total fraction showed that the secoiridoid derivatives varied between crop seasons as reflected in the level of 3,4-DHPEA-EDA, the main secoiridoid derivative in all the crop seasons, and this percentage depended on the year’s weather changes, including the frost damage from December 2001. In the 2001/02 crop season, a significantly different percent composition of the phenolic profile was observed. This was characterized by a lower amount of secoiridoid derivatives and especially of 3,4-DHPEA-EDA, which decreased significantly, while the percentage of lignans increased owing to their high stability, although their concentration remained practically constant over the four crop seasons. The decrease in the percentage of 3,4-DHPEA-EDA in olive oils extracted from olives after frost damage was probably caused by its antioxidant activity, as reported in Morelló et al. (2003). This antioxidant activity is due to the presence of a 3,4-dihydroxy moiety linked to an aromatic ring, and the effect depends on the polarity of the compound (open or closed ring in elenolic acid). Another change observed in the phenolic fraction of the oil was the increase in the
49
3,4-DHPEA-EDA percentage in the 2002/03 and 2003/04 crop seasons. Olive trees that were frost damaged in December 2001 were more sensitive to stress caused by water deficit during the summer in the two last crop seasons, particularity in 2002/03. This fact may have caused this increase. Although temperatures and rainfall during the four crop seasons were similar (except for December 2001), higher levels of phenolic compounds were detected in olive oils from the two last crop seasons.
SUMMARY POINTS ●
●
●
●
●
●
The virgin olive oil composition is greatly influenced by climatic conditions. In a limited area, small differences in maximum and minimum temperatures could be the cause of significant differences in fatty acid profile. The high cumulate rainfall causes a significant decrease in oleic acid. No differences were found in the traditional quality indices of oils affected by freeze injuries, since olives were harvested and processed in a short time period. Frost damage caused two main changes in virgin olive oil composition. These were slight decreases in chlorophyll and carotenoid contents and an important decrease in the concentration of secoiridoid derivatives and 3,4-DHPEA-AC which play an important role in oil stability and the sensory attributes. After frost, oils showed lower stability and suffered important changes in their sensory attributes, leading to absence of the bitter taste and less pungent taste. After the freeze conditions, there were slight rises in the concentrations of simple phenols, such as vanillic acid and vanillin, giving rise to sweeter oils.
ACKNOWLEDGMENTS This work was supported by the Centre de Referencia de Tecnologia d’Aliments (CeRTA) of the Catalan Government and the grants ALI19990760 and AGL2002-00289 from the Ministerio de Ciencia y Tecnologia (Spanish Government). We wish to thank the Regulator Organism of the PDO ‘Les Garrigues’ (Catalonia, Spain).
REFERENCES Baldioli, M., Servili, M., Perretti, G., Montedoro, G.F., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oil. J. Am. Oil Chem. Soc. 73, 1589–1593. Beltrán, G., Aguilera, M.P., Del Rio, S., Sánchez, S., Martínez, L., 2005. Influence of fruit ripening process on the natural antioxidant content of Hojiblanca virgin olive oils. Food Chem. 89, 207–215. Fernández-Bolaños., Felizón, B., Brenes, M., Guillen, R., Heredia, A., 1998. Hydroxytyrosol and tyrosol as the main compounds found in the phenolic fraction of steam-exploded olive stones. J. Am. Oil Chem. Soc. 75, 1643–1649.
50
García, A., Brenes, M., Romero, C., García, P., Garrido, A., 2002. Study of phenolic compounds in virgin olive oils of the Picual variety. Eur. Food. Res. Technol. 215, 407–412. García, A., Brenes, M., García, P., Romero, C., Garrido, A., 2003. Phenolic content of commercial olive oils. Eur. Food. Res. Technol. 216, 520–525. Gennaro, L., Piccioli-Bocca, A., Modesti, D., Masella, R., Coni, E., 1998. Effect of biophenols on olive oil stability evaluated by Thermogravimetric analysis. J. Agric. Food Chem. 46, 4465–4469. Gutiérrez, F., Perdiguero, S., Gutiérrez, R., Olías, J.M., 1992. Evaluation of the bitter taste in virgin olive oil. J. Am. Oil Chem. Soc. 69, 394–395. Llasat, 1997. Meteorologia agricola i forestal a Catalunya: conceptes, estacions i estadístiques. Ed. DARP. Generalitat de Catalunya. Spain. Mínguez-Mosquera, 1997. Clorofilas y carotenoides en Tecnología de Alimentos. Serie Ciencias, n 47. Secretariado de Publicaciones de la Universidad de Sevilla. Morello, J.R., Motilva, M.J., Ramo, T., Romero, M.P., 2003. Effect of freeze injuries in olive fruit on virgin olive oil composition. Food Chem. 81, 547–553.
SECTION | I The Plant and Production
Morelló, J.R., Romero, M.P., Motilva, M.J., 2003. Influence of seasonal conditions on the composition and quality parameters of monovarietal virgin olive oils. J. Am. Oil Chem. Soc. 83, 683–690. Motilva, M.J., Ramo, T., Romero, M.P., 2001. Geographical characterisation of virgin olive oils of the denomination of protected origin les garrigues by their fatty acid profile. Grasas y Aceites 52, 26–32. Romero, M.P., Tovar, M.J., Ramo, T., Motilva, M.J., 2003. Effect of crop season on the composition of virgin olive oil with protected designation of origin “Les Garrigues”. J. Am. Oil Chem. Soc. 80, 423–430. Tomás-Barberán, F.A., Espín, J.C., 2001. Phenolic compounds and related enzymes as determinants of quality in fruits and vegetables. J. Sci. Food Agric. 81, 853–876. Tovar, M.J., Motilva, M.J., Romero, M.P., 2001. Changes in the phenolic composition of virgin olive oil from young trees (Olea europaea L. cv. Arbequina) grown under linear irrigation strategies. J. Agric. Food Chem. 49, 5502–5508.
Chapter 6
Influence of Irrigation Management and Ripening on Virgin Olive Oil Quality and Composition Giuseppe Fregapane, Aurora Gómez-Rico and Maria Desamparados Salvador Departamento de Tecnología de Alimentos, Universidad de Castilla – La Mancha, Spain
6.1 INTRODUCTION
6.2 IRRIGATION MANAGEMENT
The olive tree is generally grown under rainfed conditions, especially in Castilla – La Mancha, a region with limited water resources. Nevertheless, since irrigation increases the yield of the olive orchard, even with a low amount of water, there is increasing interest in irrigated agriculture. This has led to a situation in which some of the traditional olive groves and the majority of the new ones are being adapted to irrigation techniques. However, a satisfactory compromise between the amount of water applied and the improvement in the production of the olive crop and characteristics of virgin olive oil must be reached. Some recent research has shown differences in the chemical makeup and sensory characteristics of virgin olive oil from irrigated and rainfed olive trees (Aparicio and Luna, 2002). The chemical components most influenced by irrigation are the phenolic compounds which affect both the oxidative stability and the sensory characteristics, especially the bitterness attribute, showing in both cases an inverse relationship with the amount of water applied to the olive trees (D’Andria et al., 1996; Motilva et al., 1999, 2000; Tovar et al., 2001). This aspect is important in olive cultivars that produce virgin olive oils with high bitterness and pungency, like the Cornicabra variety in Castilla – La Mancha and therefore just the right level of irrigation could enhance its sensory characteristics. To optimize sustainable irrigation conditions in the Cornicabra olive cultivar grown in Castilla – La Mancha, a region where aquifers are over-exploited, and to study the effect of different irrigation management and ripening on the composition of Cornicabra virgin olive oil, different irrigation treatments (based on 100% crop evapotranspiration, ETc, also known as the FAO method, 125 FAO, two different regulated deficit irrigation strategies and rainfed) were applied to a traditional olive orchard (Cornicabra cv).
The experimental olive orchard of Cornicabra cv., located in Ciudad Real (Spain), is constituted by three hundred 50-year-old trees (spaced 12 ⫻ 12 m2) structured in a randomized complete block design with four different irrigation treatments (Table 6.1): rainfed (RF) conditions, regulated deficit irrigation (RDI), FAO and 125 FAO. Rainfed conditions were used as a control to compare the results obtained with the irrigation treatments. In FAO treatment the water requirements were calculated using a methodology based on the crop evapotranspiration (ETc) proposed by the United Nations Food and Agriculture Organization. In 125 FAO treatments a total irrigation dosage 25% higher than the FAO treatment was applied. For regulated deficit irrigation (RDI), a maximum of 75 mm of water was established since in many Spanish irrigated olive areas there is a legal limit of 100 mm. Two different strategies were evaluated. In 2003 (RDI-1), water was applied throughout the entire season with different rates of application (10% FAO in May and June, 4% FAO in July and August and 18% FAO in September); whereas in 2004 (RDI-2), based on the results obtained during the previous crop season, water was applied only from the beginning of August, when the oil starts to form in the fruit, for the purpose of investigating which RDI treatment is more effective in achieving similar olive production and olive oil quality to that obtained by the FAO method while considerably reducing the total amount of water applied. The total water applied in 2003/04 for the different irrigation treatments was: 56 mm for RDI-1, 148 mm for FAO and 206 mm for 125 FAO; and in 2004/05: 60 mm for RDI-2, 124 mm for FAO and 154 mm for 125 FAO. More detailed data on the irrigation management have been previously reported (Gómez-Rico et al., 2006, 2007).
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
51
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
52
SECTION | I The Plant and Production
TABLE 6.1 Key facts of irrigation management. This table lists the key facts of irrigation management describing the basic characteristics of the different irrigation treatments studied in the Cornicabra olive orchard. 1. Irrigation management in olive orchard increases the olive and olive oil production 2. Water applied to olive trees may affect virgin olive oil quality and composition 3. According to the irrigation methodology proposed by the FAO (Food and Agriculture Organization), the total water requirements of olive trees is calculated by subtracting the effective precipitation from the crop evapotranspiration (ETc) 4. ETc is calculated using the effective crop coefficient (Kc), the reference crop evapotranspiration (ETo), and a reducing coefficient (Kr) 5. The regulated deficit irrigation (RDI) is based on a lower dose of water to the olive orchard than the 100% ETc, established according to the weather and phenological stage of the trees, assuring an adequate water status during full bloom and oil accumulation in the fruit 6. The olive trees in rainfed (RF) condition only receive the effective precipitations and therefore are used to determine the ‘highest water stress’ situation 7. Olive trees are generally irrigated by compensating drippers placed around the trees
Olive fruit samples from rainfed and irrigation treatment trees were harvested throughout ripening, from immature stage to normal harvest period for the Cornicabra variety (from the beginning of November to the middle of January). The influence of fruit ripening and irrigation management on the production of the olive grove, which was significantly lower under rainfed conditions than under irrigation (about 35%), and the characteristics and composition of the olive fruit have been discussed elsewhere (Gómez-Rico et al., 2007).
6.3 VIRGIN OLIVE OIL QUALITY INDICES Free acidity, given as % of oleic acid, peroxide value (PV) expressed as milliequivalents of active oxygen per kilogram of oil (meq O2 kg⫺1), and K232 and K270 extinction coefficients calculated from absorption at 232 and 270 nm, were measured following the analytical methods described in European Regulation EEC 2568/91 and subsequent amendments. The observed free acidity ranging from 0.09 to 0.20%, and peroxide value, from 1.7 to 3.4 meqO2 kg⫺1, of the different types of virgin olive oils (VOO) studied in this assay
in the crop season 2003/04 were considerably lower than the upper limit of 0.8% as oleic acid and 20 meqO2 kg⫺1, respectively, established by EU legislation for extra virgin olive oil. Moreover, these two quality indices were not influenced by irrigation, since no statistically significant differences in oil from rainfed and irrigation treatments in the crop season 2003/04 were obtained. This was also observed by Tovar et al. (2001) in virgin olive oils from Arbequina cultivar, Dettori and Russo (1993) in Leccino, Nociara and Ogliarola Salentina cultivars and Patumi et al. (1999) in Nocellara del Belice and Ascolana Tenera cultivars. On the contrary, in crop 2004/05 a statistical difference for free acidity and peroxide value was indeed obtained between RF and the irrigation treatments due to the higher degree of fruit damage as a consequence of the olive fly attack. Nevertheless, the values of free acidity and the peroxide value of the olive oil obtained from partially damaged fruit were not high from an olive oil quality point of view: a maximum acidity of 0.4% and a 5.4 peroxide value were observed.
6.4 SENSORY CHARACTERISTICS All the virgin olive oils obtained using the different irrigation treatments of the trees were classified as ‘extra virgin’ oil by means of the organoleptic evaluation as reported in Table 6.2. Sensory evaluation was assessed by an International Olive Oil Council-recognized panel of assessors from the Protected Designation of Origin ‘Montes de Toledo’ (Toledo, Spain) according to Annex XII of Regulation EC 796/2002 (amending ECC 2568/91). Sensory attributes affected by irrigation were ‘bitterness’, ‘pungency’ and ‘fruitiness’, according to what has previously been described for other olive cultivars (Salas et al., 1997; Tovar et al., 2001, 2002). As is known, the intensity of sensory pungency and especially bitterness are related to the phenol content in the olive oil, which, as expected, was higher in oils obtained under rainfed conditions. In all cases, a slight decrease in the intensity of these positive attributes was observed, more marked in the case of bitterness, by increasing the amount of water delivered through irrigation. This observation is very relevant from the olive quality and marketing point of view since, although bitterness is a positive sensory attribute in virgin olive oil, a high level of bitterness could cause consumers to reject the oil. A high level of bitterness is a typical characteristic of the Cornicabra variety virgin olive oils’ sensory profile, and therefore the use of irrigation could produce a desirable descent in the intensity of this attribute and hence increase consumer preference. However, in the 2004/05 crop season no statistically significant differences were obtained in the positive sensory attributes, including bitterness, between the olive oils obtained under rainfed and irrigation conditions.
CHAPTER | 6 Influence of Irrigation Management and Ripening on Virgin Olive Oil Quality and Composition
TABLE 6.2 Virgin olive oil organoleptic evaluation as affected by the different irrigation treatments and the ripeness index of the fruits. Ripeness Index
Grade
2003/2004
Sensory attributes
K225
Fruity
Bitterness
Pungency
RF
2.7 ⫾ 0.3a,x
Extra Virgin
6.1 ⫾ 0.2b,x
8.2 ⫾ 0.4b,w
7.8 ⫾ 0.3a,w
0.78 ⫾ 0.01c,x
RDI-1
2.8 ⫾ 0.4a,x
Extra Virgin
6.2 ⫾ 0.5b,w
8.0 ⫾ 0.3b,wx
8.0 ⫾ 0.3a,wx
0.67 ⫾ 0.07b,x
FAO
3.1 ⫾ 0.3a,x
Extra Virgin
5.5 ⫾ 0.1b,w
7.7 ⫾ 0.5ab,w
7.7 ⫾ 0.3a,w
0.66 ⫾ 0.04b,yz
125 FAO
2.8 ⫾ 0.2a,w
Extra Virgin
4.9 ⫾ 0.3a,w
7.2 ⫾ 0.3a,wx
7.6 ⫾ 0.3a,w
0.56 ⫾ 0.07a,xy
RF
3.7 ⫾ 0.2a,y
Extra Virgin
5.4 ⫾ 0.2ab,w
8.5 ⫾ 0.2c,w
8.4 ⫾ 0.2ab,w
0.77 ⫾ 0.01c,x
RDI-1
3.8 ⫾ 0.3a,y
Extra Virgin
6.2 ⫾ 0.3b,w
8.3 ⫾ 0.2c,x
8.5 ⫾ 0.1b,w
0.64 ⫾ 0.07b,wx
FAO
4.0 ⫾ 0.4a,y
Extra Virgin
5.5 ⫾ 0.2ab,w
7.7 ⫾ 0.3b,w
8.0 ⫾ 0.2ab,w
0.56 ⫾ 0.04ab,x
125 FAO
3.9 ⫾ 0.1a,y
Extra Virgin
5.3 ⫾ 0.2a,w
6.9 ⫾ 0.4a,w
8.0 ⫾ 0.2a,w
0.49 ⫾ 0.08a,wx
RF
5.7 ⫾ 0.4b,z
Extra Virgin
5.4 ⫾ 0.3ab,wx
8.3 ⫾ 0.2a,w
8.1 ⫾ 0.2ab,w
0.66 ⫾ 0.05c,w
RDI-1
5.4 ⫾ 0.5ab,z
Extra Virgin
6.2 ⫾ 0.1b,w
7.5 ⫾ 0.5a,w
7.7 ⫾ 0.1a,x
0.57 ⫾ 0.03b,w
FAO
5.5 ⫾ 0.3ab,z
Extra Virgin
5.0 ⫾ 0.3a,w
7.7 ⫾ 0.3a,w
7.9 ⫾ 0.2a,w
0.46 ⫾ 0.03a,w
125 FAO
4.9 ⫾ 0.1a,z
Extra Virgin
6.0 ⫾ 0.1b,x
8.0 ⫾ 0.3a,x
8.0 ⫾ 0.2b,w
0.41 ⫾ 0.09a,w
RF
2.8 ⫾ 0.2b,w
Extra Virgin
6.4 ⫾ 0.3a,w
7.6 ⫾ 0.2a,w
8.0 ⫾ 0.3a,w
0.66 ⫾ 0.09a,w
RDI-2
2.3 ⫾ 0.1a,w
Extra Virgin
6.0 ⫾ 0.2a,x
7.4 ⫾ 0.5a,w
7.9 ⫾ 0.6a,wx
0.70 ⫾ 0.00a,x
FAO
2.5 ⫾ 0.2ab,w
Extra Virgin
5.6 ⫾ 0.4a,w
6.7 ⫾ 0.3a,w
7.5 ⫾ 0.2a,w
0.67 ⫾ 0.00a,x
125 FAO
2.4 ⫾ 0.1a,w
Extra Virgin
5.2 ⫾ 0.6a,w
7.3 ⫾ 0.2a,w
7.7 ⫾ 0.3a,w
0.60 ⫾ 0.00a,y
RF
3.4 ⫾ 0.0a,x
Extra Virgin
5.5 ⫾ 0.2ab,w
7.0 ⫾ 0.5a,w
7.4 ⫾ 0.5a,w
0.60 ⫾ 0.08a,w
RDI-2
3.4 ⫾ 0.0a,x
Extra Virgin
4.9 ⫾ 0.4a,w
7.0 ⫾ 0.3a,w
7.1 ⫾ 0.3a,w
0.60 ⫾ 0.03a,x
FAO
3.4 ⫾ 0.0a,x
Extra Virgin
5.5 ⫾ 0.2ab,w
6.9 ⫾ 0.4a,w
7.6 ⫾ 0.2a,w
0.59 ⫾ 0.03a,x
125 FAO
3.5 ⫾ 0.1a,x
Extra Virgin
5.5 ⫾ 0.5b,w
6.6 ⫾ 0.3a,w
7.4 ⫾ 0.2a,w
0.50 ⫾ 0.02a,x
RF
4.1 ⫾ 0.1a,y
Extra Virgin
6.0 ⫾ 0.4a,w
7.4 ⫾ 0.3ab,w
7.6 ⫾ 0.4a,w
0.59 ⫾ 0.12b,w
RDI-2
4.2 ⫾ 0.0a,y
Extra Virgin
5.4 ⫾ 0.3a,wx
7.6 ⫾ 0.2b,w
8.2 ⫾ 0.2a,x
0.57 ⫾ 0.01b,w
FAO
4.2 ⫾ 0.0a,y
Extra Virgin
5.4 ⫾ 0.3a,w
7.4 ⫾ 0.3b,w
7.5 ⫾ 0.3a,w
0.54 ⫾ 0.01ab,w
125 FAO
4.2 ⫾ 0.0a,y
Extra Virgin
5.9 ⫾ 0.2a,w
6.6 ⫾ 0.4a,w
7.4 ⫾ 0.3a,w
0.36 ⫾ 0.04a,w
2004/2005
Sensory attributes affected by irrigation were ‘bitterness’, ‘pungency’ and ‘fruitiness’. In all cases, a slight decrease in the intensity of these positive attributes was observed, more marked in the case of bitterness, by increasing the amount of water delivered through irrigation. RF, rainfed; RDI, regulated deficit irrigation; FAO, Food and Agriculture Organization method. Different letters within a column (a–c) indicate significant differences (p ⬍ 0.05) with respect to irrigation treatment in each sampling. Different letters within a column (w–y) indicate significant differences (p ⬍ 0.05) with respect to ripeness index for each treatment.
53
54
Olive oil bitterness can also be measured by the instrumental K225 parameter called bitterness index (GutiérrezRosales et al., 1992). In the 2003/04 crop, a great decrease in the bitterness index was observed as the water dose applied to olive trees increased (Table 6.2), varying from 0.77 to 0.49 respectively for RF and 125 FAO for the sampling close to a ripeness index of 4.0. However, in the 2004/05 crop no statistically significant differences were obtained.
SECTION | I The Plant and Production
were observed between the irrigation treatments studied. However, Tovar et al. (2002) reported significant increases in the α-tocopherol values in Arbequina VOO as the irrigation doses applied in the trees increased. On the other hand, these rises did not imply changes in the oxidative stability of the oils.
6.6.1 Total Phenol Content 6.5 FATTY ACID COMPOSITION Fatty acid composition is performed according to the European Regulations EEC 2568/91 and subsequent amendments, corresponding to the AOCS Method Ch 2-91. In both crop seasons studied and in all irrigation treatments studied, the palmitic acid content slightly decreased as fruit ripened, i.e. from 10.4% down to 9.1% and from 11.4% to 9.7% respectively for RF and FAO irrigation treatments; whereas oleic and linoleic acids showed an opposite trend, i.e. the oleic acid content varied from 78.4% to 79.5% and the linoleic acid from 3.7% to 4.6% under the FAO conditions. The increase in oleic acid content is due to the triacylglycerols’ active biosynthesis which takes place throughout fruit ripening, involving a fall in the relative percentage of the oil’s palmitic acid content. On the other hand, the increase in linoleic acid content is due to the transformation of oleic acid into linoleic acid by the oleate desaturase activity which is active during triacylglycerol biosynthesis (Sanchez and Harwood, 2002). The content of the other fatty acids remained practically unchanged during fruit ripening. In the 2003/04 crop, rainfed olive oils always showed a statistically significant higher content in oleic acid, whereas olive oils from irrigated trees had a higher content in palmitic and linoleic acids. As a consequence, the unsaturated/ saturated and MUFA/PUFA ratios were significantly higher in oils obtained in rainfed conditions, in line with the results obtained by Salas et al. (1997). However, these changes are very slight and do not possess any nutritional value relevance.
6.6 NATURAL ANTIOXIDANTS CONTENT The values of the α-tocopherol and total phenol content and the oxidative stability of the virgin olive oils obtained from the different treatments studied are shown in Table 6.3. Phenolic compounds were quantified by HPLC at 280 nm using syringic acid as internal standard and the response factors determined by Mateos et al. (2001), and described in GomezAlonso et al. (2002). Tocopherols were evaluated following the AOCS Method Ce 8-89. Oxidative stability was evaluated by the Rancimat Method (Laübli and Bruttel, 1986). The α-tocopherol content decreased slightly during ripening, whereas insignificant differences in its concentration
The total phenol content of the oils was significantly affected by the irrigation such that as the water dose applied to olive trees increased, the amount of the phenolic compounds in the virgin olive oil obtained decreased significantly (Table 6.3). For example, in crop 2003/04 in the case of rainfed virgin olive oil samples the total phenol content decreased from 1700 mg kg⫺1 to 900 mg kg⫺1 through fruit ripening, whereas for olive oil samples under FAO treatment, the phenol content decreased from 1080 to 650 mg kg⫺1. Panelli et al. (1989), Salas et al. (1997), and Patumi et al. (1999, 2002) observed similar behavior for other olive cultivars like Picual, Nocellara del Belice, Kalamata and Ascolana Tenera. As was previously mentioned, the concentration of phenolic compounds affects the sensory bitterness attribute with the beneficial and important consequences earlier discussed in the case of the Cornicabra olive oil variety, as well as oxidative stability. In terms of the latter, the observed decrease in the oxidative stability does not affect the Cornicabra virgin olive oil shelf-life or quality since this is a very stable and phenol-rich olive oil variety, but could significantly reduce the shelf-life of other varieties like Arbequina, due to its naturally poor phenol content.
6.6.2 Phenolic Profile (Reprinted in part with permission from J Agric Food Chem 54 (2006) 7130–7136. Copyright 2006 American Chemical Society.) There is a considerable difference in the concentrations of secoiridoid derivatives of hydroxytyrosol and tyrosol observed in the VOO in the course of fruit ripening and under the various irrigation treatments studied. In fact the compounds most affected by irrigation scheduling of the olive grove and by ripening of the fruit were the complex phenol chemical forms, the levels of which decreased significantly in the VOO during ripening and as the water supplied increased. For example, in crop 2003/04 the 3,4-DHPEA-EDA content decreased from 770 mg kg⫺1 to 450 mg kg⫺1 and the 3,4-DHPEA-EA diminished from 300 mg kg⫺1 to 170 mg kg⫺1 in the course of fruit ripening in the rainfed (RF) VOO samples, while from RF conditions to FAO irrigation, at a ripeness index of approximately 4.0, the 3,4-DHPEA-EDA content decreased from
CHAPTER | 6 Influence of Irrigation Management and Ripening on Virgin Olive Oil Quality and Composition
TABLE 6.3 Virgin olive oil antioxidants content and oxidative stability as affected by the different irrigation treatments studied and the ripeness index of the fruits. 2003/2004
Ripeness index
α-tocopherol (mg kg⫺1)
Total phenols (mg kg⫺1)
Oxidative stability (h)
RF
1.5 ⫾ 0.5a,w
283 ⫾ 64a,x
1719 ⫾ 130c,y
–
ab,w
a,w
b,y
RDI-1
1.8 ⫾ 0.6
284 ⫾ 59
1354 ⫾ 42
–
FAO
2.5 ⫾ 0.4b,w
222 ⫾ 25a,w
1076 ⫾ 122a,z
–
125 FAO
2.0 ⫾ 0.3ab,v
273 ⫾ 33a,x
968 ⫾ 254a,y
–
a,x
a,wx
38.3 ⫾ 0.5d,x
RF
2.7 ⫾ 0.3
235 ⫾ 43
RDI-1
2.8 ⫾ 0.4a,x
259 ⫾ 35a,w
1084 ⫾ 146b,x
34.0 ⫾ 0.4c,w
FAO
3.1 ⫾ 0.3a,x
212 ⫾ 25a,w
998 ⫾ 85b,yz
31.1 ⫾ 1.3b,x
125 FAO
2.8 ⫾ 0.2a,w
254 ⫾ 26a,wx
805 ⫾ 125a,xy
27.1 ⫾ 0.1a,w
RF
3.2 ⫾ 0.3a,xy
226 ⫾ 41ab,wx
RDI-1
3.5 ⫾ 0.4a,xy
FAO
3.6 ⫾ 0.4a,xy a,x
1380 ⫾ 62
c,x
1294 ⫾ 64c,x
–
252 ⫾ 29b,w
946 ⫾ 40b,x
–
201 ⫾ 25a,w
868 ⫾ 78b,xy
–
237 ⫾ 8
ab,wx
699 ⫾ 139
a,wx
125 FAO
3.4 ⫾ 0.3
RF
3.7 ⫾ 0.2a,y
225 ⫾ 47a,wx
1364 ⫾ 107c,x
38.4 ⫾ 0.5d,x
RDI-1
3.8 ⫾ 0.3a,y
242 ⫾ 44a,w
1004 ⫾ 160b,x
31.9 ⫾ 1.3c,w
FAO
4.0 ⫾ 0.4a,y
204 ⫾ 21a,w
824 ⫾ 56ab,x
30.1 ⫾ 1.3b,x
125 FAO
3.9 ⫾ 0.1a,y
227 ⫾ 25a,w
651 ⫾ 124a,wx
24.6 ⫾ 0.4a,w
RF
5.7 ⫾ 0.4b,z
193 ⫾ 32a,w
905 ⫾ 189b,w
34.4 ⫾ 0.3b,w
RDI-1
5.4 ⫾ 0.5ab,z
226 ⫾ 31a,w
757 ⫾ 12ab,w
28.5 ⫾ 3.2ab,w
FAO
5.5 ⫾ 0.3ab,z
202 ⫾ 11a,w
654 ⫾ 108a,w
24.3 ⫾ 0.5a,x
125 FAO
4.9 ⫾ 0.1a,z
233 ⫾ 19a,w
536 ⫾ 124a,w
22.2 ⫾ 3.4a,w
RF
2.8 ⫾ 0.2b,w
298 ⫾ 17b,w
1019 ⫾ 216a,w
29.8 ⫾ 2.2a,w
RDI-2
2.3 ⫾ 0.1a,w
250 ⫾ 7a,w
905 ⫾ 10a,x
32.5 ⫾ 2.0a,w
FAO
2.5 ⫾ 0.2ab,w
272 ⫾ 10ab,x
877 ⫾ 11a,x
30.3 ⫾ 0.9a,x
125 FAO
2.4 ⫾ 0.1a,w
271 ⫾ 19ab,w
724 ⫾ 38a,x
28.7 ⫾ 1.0a,x
RF
3.4 ⫾ 0.0a,x
280 ⫾ 14b,w
921 ⫾ 183b,w
29.7 ⫾ 2.5a,w
RDI-2
3.4 ⫾ 0.0a,x
238 ⫾ 11a,w
691 ⫾ 11ab,w
28.2 ⫾ 1.6a,w
FAO
3.4 ⫾ 0.0a,x
263 ⫾ 2b,wx
724 ⫾ 57ab,w
28.5 ⫾ 1.6a,wx
125 FAO
3.5 ⫾ 0.1a,x
238 ⫾ 3a,w
551 ⫾ 14a,wx
25.5 ⫾ 0.1a,x
RF
4.1 ⫾ 0.1a,y
269 ⫾ 12b,w
818 ⫾ 224b,w
27.2 ⫾ 3.4b,w
RDI-2
4.2 ⫾ 0.0a,y
226 ⫾ 3a,w
739 ⫾ 51b,w
27.4 ⫾ 1.1b,w
FAO
4.2 ⫾ 0.0a,y
241 ⫾ 8ab,x
679 ⫾ 19b,w
23.9 ⫾ 1.8ab,w
125 FAO
4.2 ⫾ 0.0a,y
241 ⫾ 17ab,w
423 ⫾ 102a,w
18.3 ⫾ 3.6a,w
–
2004/2005
The total phenol content in VOO decreased clearly as the irrigation dose applied in olive trees increased; as expected a similar behavior was observed in the oxidative stability of the oils. RF, rainfed; RDI, regulated deficit irrigation; FAO, Food and Agriculture Organization method. Different letters within a column (a–d) indicate significant differences (p ⬍ 0.05) with respect to irrigation treatment in each sampling. Different letters within a column (w–z) indicate significant differences (p ⬍ 0.05) with respect to ripeness index for each treatment. Reprinted with permission from Food Chemistry 100 (2007) 568–578. Copyright 2007 Elsevier Science.
55
56
SECTION | I The Plant and Production
FIGURE 6.1 Evolution of hydroxytyrosol and its complex secoiridoid forms, and of tyrosol and its derivatives, in the course of fruit ripening as affected by irrigation management in crop season 2003/04. The complex phenols were not affected in the same way by irrigation, since the secoiridoid derivatives of hydroxytyrosol decreased more than those of tyrosol. 䊊, rainfed; 䉱, regulated deficit irrigation, RDI-1; 䊐, FAO; 䉬, 125 FAO. Reprinted with permission from J Agric Food Chem 54 (2006) 7130–7136. Copyright 2006 American Chemical Society.
676 mg kg⫺1 to 388 mg kg⫺1 and the 3,4-DHPEA-EA from 258 mg kg⫺1 to 123 mg kg⫺1. Tovar et al. (2001) observed similar behavior for the Arbequina cultivar, since the levels of secoiridoids diminished as the irrigation dose of olive trees increased. As far as the 2004/05 crop season is concerned, although the levels of complex phenols in the VOO samples were lower, the trend was similar to that of the previous crop season. It is important to note that the differences in phenol contents between the oils from FAO and the second regulated deficit irrigation (RDI-2) strategy were not statistically significant; this indicates that the RDI scheduling employed in the second crop season (RDI-2), where water was applied from the beginning of August only, produced a VOO with a phenolic composition more similar to that of oil from FAO-treated olives than that from olives grown under RDI-1 water scheduling of the previous year. Indeed, the phenolic and volatile composition related to the quality of VOO comparable to FAO management was achieved with less demand in water supply. Moreover, the complex phenols were not affected in the same way by irrigation, since the secoiridoid derivatives of hydroxytyrosol decreased more than those of tyrosol, as clearly shown in Figure 6.1. This is very important, since hydroxytyrosol and its complex derivative forms are known to possess much greater antioxidant activity and organoleptic influence than the tyrosol group (Baldioli et al., 1996; Gennaro et al., 1998). These results were similar to those reported by Tovar et al. (2001, 2002) for the Arbequina variety.
6.7 VOLATILE COMPOUNDS 6.7.1 Evolution Along Fruit Ripening Figure 6.2 depicts the evolution of the main volatile compounds found in Cornicabra virgin olive oil in the course of fruit ripening as affected by RF and FAO irrigation conditions in crop season 2003/04. Solid phase microextraction (SPME) followed by GC is used to analyze the volatiles (Vichi et al., 2003). In all the Cornicabra VOO samples analyzed, the major volatile component was the C6 aldehyde fraction, the content of which decreased as ripening progressed. For example, the E-2-hexenal content ranged between 4.2 mg kg⫺1 and 2.6 mg kg⫺1 (as internal standard; IS) over fruit maturation for VOOs under RF conditions and between 8.0 mg kg⫺1 and 3.5 mg kg⫺1 for those under FAO scheduling; the amount of hexanal was lower and varied between 1.10 mg kg⫺1 and 0.45 mg kg⫺1 over fruit ripening under RF conditions and between 0.90 mg kg⫺1 and 0.50 mg kg⫺1 under FAO treatment. These compounds, which are responsible for the positive green sensory notes in VOO, are produced through the LOX pathway that takes place during crushing of the olive fruit and olive paste malaxation and are incorporated into the oily phase (Sanchez and Harwood, 2002). Aparicio et al. (1998) reported similar behavior, but other researchers (Angerosa et al., 2001) have shown that during olive ripening the amount of volatile compounds, especially E-2-hexenal, increased up to a maximum
CHAPTER | 6 Influence of Irrigation Management and Ripening on Virgin Olive Oil Quality and Composition
57
FIGURE 6.2 Evolution of main volatile compounds in virgin olive oils under rainfed conditions and FAO irrigation scheduling in the course of fruit ripening in crop season 2003/04. A great fall of C6 aldehydes in VOO along the fruit maturation, as well as a higher content of E-2-hexenal, hexan-1-ol and Z-3-hexen-1-ol in VOO from FAO methodology rather than in rainfed conditions is observed. 䊊 (open circle), E-2-hexenal; 䊏, hexanal; 䉫, Z-3hexen-1-ol; 䉱, hexan-1-ol; 䉬, E-2-hexen-1-ol. Reprinted with permission from J Agric Food Chem 54 (2006) 7130–7136. Copyright 2006 American Chemical Society.
concentration, which occurred when fruit skin color turned from yellow-green to purple; beyond that point the volatile content decreased. With respect to the evolution of C6 alcohols, there was a significant decrease in E-2-hexen-1-ol content, while hexan-1-ol and Z-3-hexen-1-ol increased slightly during fruit ripening. It is worth noting that this observed increase was statistically significant in VOO from olives under FAO and 125 FAO treatments, which received the highest irrigation doses. A similar trend was observed in the crop season 2004/05. As regards C6 esters such as hexyl-acetate and Z-3-hexyl-acetate, these were present in very small amounts in Cornicabra VOO, indicating that there was also little activity of the alcohol acyl transferase.
under FAO conditions, but with a considerable reduction in the total amount of water used in the olive grove.
SUMMARY POINTS ●
●
●
●
6.7.2 Effect of Irrigation The volatile compounds most affected by the irrigation were E-2-hexenal, Z-3-hexen-1-ol and hexan-1-ol, in the sense that the increase in the water applied to the olive trees produced an increase in these volatiles, mainly in oils from fruits whose ripeness index was higher than 2.5–3.0. RDI-1 resulted in a VOO with a similar volatile composition to those from FAO treatment. On the other hand, the total volatile levels in VOOs produced under RDI-2 conditions were higher than in VOOs produced under FAO conditions and similar to the levels found in VOOs produced under 125 FAO conditions. It is therefore very important to note that the VOOs produced by the second RDI strategy were richer in volatile compounds than those produced
●
●
●
The quality indices are generally not influenced by irrigation, since no statistically significant differences in oil from rainfed and irrigation treatments are observed. The different water stress levels in olive trees affect not only the total amount of phenolic and volatile compounds in the VOO but also their profile. The total polar phenol content, which affects the sensory bitterness in the oils, decreases significantly as the amount of supplied water increases. Complex phenols are not affected in the same way by irrigation, since the secoiridoid derivatives of hydroxytyrosol decreased more than those of tyrosol. The volatile compounds most affected by the irrigation are E-2-hexenal, Z-3-hexen-1-ol and hexan-1-ol; the increase in the water applied to the olive trees produces an increase in these volatiles. The selection of an optimal irrigation treatment for traditional Cornicabra olive orchards in Castilla – La Mancha calls for the establishment of a suitable compromise between olive production, virgin olive oil characteristics and water consumption. The best irrigation treatment is regulated deficit irrigation (RDI), and apparently better results are obtained applying water only from the beginning of August when the accumulation of oil begins in the fruit.
58
REFERENCES Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2001. Influence of malaxation temperature and time on the quality of virgin olive oils. Food Chem. 72, 19–28. Aparicio, R., Morales, M.T., 1998. Characterization of olive ripeness by green aroma compounds of virgin olive oil. J. Agric. Food Chem. 46, 1116–1122. Aparicio, R., Luna, G., 2002. Characterisation of monovarietal virgin olive oil. Eur. J. Lipid Sci. Tech. 104, 614–627. Baldioli, M., Servili, M., Peretti, G., Montedoro, G.F., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oil. J. Am. Oil Chem. Soc. 73, 1589–1593. D’Andria, R., Morelli, G., Martuccio, G., Fontanazza, G., Patumi, M., 1996. Valutazione della produzione e della qualita’dell’olio di Giovanni piante di olivo allevate con diversi regimi idrici. Italus Hortus 3, 23–31. Dettori, S., Russo, G., 1993. Influencia del cultivar y del régimen hídrico sobre el volumen y la calidad del aceite de oliva. Olivae 49, 36–43. Gennaro, L., Piciola Bocca, A., Modesti, D., Masella, R., Coni, E., 1998. Effect of biophenols on olive oil stability evaluated by thermogravimetric analysis. J. Agric. Food Chem. 46, 4465–4469. Gómez-Alonso, S., Salvador, M.D., Fregapane, G., 2002. Phenolic compounds profile of Cornicabra virgin olive oil. J. Agric. Food Chem. 50, 6812–6817. Gómez-Rico, A., Salvador, M.D., La Greca, M., Fregapane, G., 2006. Phenolic and volatile compounds of extra virgin olive oil (Olea europea L. cv. Cornicabra) with regards to fruit ripening and irrigation management. J. Agric. Food Chem. 54, 7130–7136. Gómez-Rico, A., Salvador, M.D., Moriana, A., Pérez, D., Olmedilla, N., Ribas, F., Fregapane, G., 2007. Influence of different irrigation strategies in a traditional Cornicabra cv. olive orchard on virgin olive oil composition and quality. Food Chem. 100, 568–578. Gutiérrez-Rosales, F., Perdiguero, S., Gutiérrez, R., Olías, J.M., 1992. Evaluation of bitter taste in virgin olive oil. J. Am. Oil Chem. Soc. 69, 394–395. Laübli, M.W., Bruttel, P.A., 1986. Determination of the oxidative stability of fats and oils by the Rancimat method. J. Am. Oil Chem. Soc. 63, 792–795. Mateos, R., Espartero, J.L., Trujillo, M., Ríos, J.J., León-Camacho, M., Alcudia, F., Cert, A., 2001. Determination of phenols, flavones
SECTION | I The Plant and Production
and lignans in virgin olive oil by solid-phase extraction and highperformance liquid chromatography with diode array ultraviolet detection. J. Agric. Food Chem. 49, 2185–2192. Motilva, M.J., Romero, M.P., Alegre, S., Girona, J., 1999. Effect of regulated déficit irrigation in olive oil production and quality. Acta Hort. 474, 377–380. Motilva, M.J., Tovar, M.J., Romero, M.P., Alegre, A., Girona, J., 2000. Influence of regulated deficit irrigation strategies applied to olive trees (Arbequina cultivar) on oil yield and oil composition during the fruit ripening. J. Sci. Food Agric. 80, 2037–2043. Panelli, G., Famiani, F., Servili, M., Montedoro, G.F., 1989. Agro-climatic factors and characteristics of the compostion of virgin olive oils. Acta Hort. 286, 477–480. Patumi, M., d’Andria, R., Fontanazza, G., Morelli, G., Giori, P., Sorrentino, G., 1999. Yield and oil quality of intensively trained trees of three cultivars of olive under different irrigation regimes. J. Hort. Sci. Biotech. 74, 729–737. Patumi, M., d’Andria, R., Marsilio, G., Fontanazza, G., Morelli, G., Lanza, B., 2002. Olive and olive oil quality after intensive monocone olive growing (Olea europaea L., cv. Kalamata) in different irrigation regimes. Food Chem. 77, 27–34. Salas, J., Pastor, M., Castro, J., Vega, V., 1997. Influencia del riego sobre la composición y características del aceite de oliva. Grasas y Aceites 48, 74–82. Sanchez, J., Harwood, J.L., 2002. Byosinthesis of triacylglycerols and volatiles in olives. Eur. J. Lipid Sci. Tech. 104, 564–573. Tovar, M.J., Romero, M.P.J., Motilva, M.J., 2001. Changes in the phenolic composition of olive oil from young trees (Olea europaea L. cv. Arbequina) grown under linear irrigation strategies. J. Agric. Food Chem. 49, 5502–5508. Tovar, M.J., Romero, M.P., Alegre, S., Girona, J., Motilva, M.J., 2002. Composition and organoleptic characteristics of oil from Arbequina olive (Olea europaea L) trees under déficit irrigation. J. Sci. Food Agric. 82, 1755–1763. Vichi, S., Castellote, A.I., Pizzale, L., Conte, L.S., Buxaderas, S., LopezTamames, E., 2003. Analysis of virgin olive oil volatile compounds by headspace solid-phase microextraction coupled to gas chromatography with mass spectrometric and flame ionisation detection. J. Chromatograp. A 983, 19–33.
Chapter 7
The Effect of the Ripening Process of the Olive Fruit on the Chlorophyll and Carotenoid Fractions of Drupes and Virgin Oils Maria-Jose Motilva and Maria-Paz Romero Department of Food Technology CeRTA-TPV, Escuela Técnica Superior de Ingeniería Agraria, Universidad de Lleida, Spain
7.1 INTRODUCTION
The composition of the chlorophyll fraction of the olive fruit comprises chlorophylls a and b, and the carotenoids that typically accompany the chlorophylls in the chloroplast: lutein, β-carotene, violaxanthin, neoxanthin, and antheraxanthin (Table 7.1). During the olive ripening, the skin color changes from green (G) at the beginning of the harvest period, light-green (LG), small reddish spots (SRS), turning-color (TC), purple (P) to black (B) at the end of the harvest period. During the ripening process, the gradual disintegration of the chloroplasts involves the disappearance of the chloroplast pigments. At the beginning of the ripening period, when the olive drupe color turns from green to light green, chlorophylls a and b are the major pigments, decreasing during the process. This decrease of the chlorophylls is more marked than in the carotenoid fraction, although both pigment fractions reach similar concentration at the end of the ripening period (black stage). The pigment content of plants depends, both quantitatively and qualitatively, on their genetic makeup between other factors. Some exclusive pigments have been detected in olive fruits from Arbequina variety (Table 7.2). Besides the chlorophylls a and b, chlorophyllide a is quantified at all the ripening stages of the fruit, and chlorophyllide b, at the green and light-green stages. Chlorophyllides are the derivative forms from the hydrolysis of chlorophylls obtaining phytol and the respective chlorophyll. This transformation is the consequence of the chlorophyllase activity, an enzyme localized in the membranes of chloroplasts. The presence of these chlorophyll derivatives in fruits of Arbequina cv could demonstrate the more active presence of chlorophyllase in this cultivar in relation to other olive cultivars.
Color is one of the major attributes that affects the consumer perception of quality of virgin olive oil. The chloroplast pigments (chlorophylls and carotenoids) are mainly responsible for the oil color, ranging from yellow-green to greenish gold (Escolar et al., 2007). In addition to the color, chlorophylls and carotenoids play an important role in the oxidative stability of virgin olive oil due to their antioxidant nature in the dark and pro-oxidant activity in the light.
7.2 EFFECT OF THE RIPENING PROCESS OF THE OLIVE FRUIT ON THE CHLOROPHYLL AND CAROTENOID FRACTIONS OF DRUPES Maturity is one of the most important factors associated with the quality evaluation of fruits and vegetables. One of the attributes for judging maturity is the skin color. In general, during fruit ripening, the chlorophylls, which are present in all unripe fruit, break down following the transformation of chloroplasts into chromoplasts, and carotenogenesis takes over (Kozukue and Friedman, 2003). The olive fruit, Olea europaea, a well-known and widespread species of the Oleaceae family, is a green, fleshy, edible drupe. During the ripening process, photosynthetic activity decreases and the concentrations of chloroplast pigments, chlorophylls and carotenoids, decrease progressively. At the end of the maturation process, the violet or purple color of the olive fruit is due to the formation of anthocyanins (Roca and Mínguez-Mosquera, 2001a). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
59
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
60
SECTION | I The Plant and Production
TABLE 7.1 Pigment composition (chlorophylls and carotenoids) of olive fruit, expressed as mg kgⴚ1 dry weight olive pulp, from Farga cv at different ripening stages. Pigment
Ripening stage (fruit epidermis color) G
LG
SRS
TC
P
B
Chlorophyll a
287
135
43
37
2.9
1.2
Chlorophyll b
105
50
16
12
1.1
0.41
2.6
2.8
4.0
1.6
Chlorophyll a/chlorophyll b Total chlorophylls Neoxanthin Violaxanthin Antheraxanthin
2.7 392 9.5 11 3.8
2.7 186
2.7 59
3.2 48
3.7
0.96
0.86
0.03
0.02
2.7
0.86
0.85
0.09
0.04
1.8
0.65
0.36
0.07
0.04
9.7
7.7
1.4
1.1
2.8
2.2
0.66
0.53
Lutein
34
all-trans-β-carotene
11
Total carotenoids
70
37
15
12
2.3
1.8
462
222
74
60
6.3
3.3
2.6
2.8
Total pigment Chlorophylls/carotenoids
2.7
22 6.5
2.7
3.2
3.2
G, green; LG, light-green; SRS, small reddish spots; TC, turning-colour; P, purple; B, black; nq, not quantified; tr, trace amounts. Reprinted from Criado, MN., et al., 2007. Food Chem 100, 748–755, with permission.
Exclusive carotenoids have been found too in Arbequina fruit; these include cis-α-carotene, β-cryptoxanthin and esterified xanthophylls (Table 7.2). The presence of these exclusive pigments has been observed in this cultivar growing in different areas in the north and the south of Spain (GandulRojas et al., 1999; Roca and Mínguez-Mosquera, 2001b; Criado et al., 2006), corresponding to olive-growing areas with different edaphologic characteristics and climatological conditions. Thus, these exclusive pigments can be considered as chemical-taxonomic differentiators of the Arbequina variety independent of the growing area of the olive tree. Besides the exclusive pigments, a carotenogenic process has been observed in the fruits of Arbequina cv during the first stages of the maturity. In relation to other vegetables, such as tomatoes (Ronen et al., 1999) and peppers (HorneroMéndez et al., 2000), the carotenogenic process in the fruit of the Arbequina cv is expressed at a lower level and there is no great increase in carotenoids. When the skin fruit color changes from green to light-green color (Table 7.2), the concentration of esterified xanthophylls increases, and the concentration of esterified neoxanthin increases, too, when the skin color changed from small reddish spots to turning-color. This synthesis of carotenoids was not found in other olive
cultivars. The carotenogenic process observed during ripening allows affirming that fruit from Arbequina cv behave similarly to carotenogenic fruit, in which chloroplast changes into chromoplast during ripening and while chlorophylls disappear, a synthesis of carotenoids occurs (Gross, 1991). Moreover, the presence of esterified xanthophylls in the olive fruit indicates that the chloroplast does not remain intact during the ripening process, but degenerates into chromoplast, characteristic of carotenogenic fruits. The presence of exclusive pigments, the different rate of pigment disappearance, in addition to the carotenogenic process, in fruits of the Arbequina variety, could be related with a singular structure of the photosynthetic apparatus in this cultivar. The gradual disintegration of the chloroplasts, during the ripening process of the olive fruit, involves the disappearance of the chloroplast pigments, mainly the major pigments chlorophylls a and b. The two chlorophylls disappear together throughout the ripening process, showing no differences in the chlorophyll a/chlorophyll b relationship (Tables 7.1 and 7.2). This relation is usually 3:1 in higher plants and constitutes a parameter of the physiological status. Chlorophyll b is a constituent of the light-harvesting system and the reaction centers
61
CHAPTER | 7 The Effect of the Ripening Process on the Chlorophyll and Carotenoid Fractions
TABLE 7.2 Pigment composition (chlorophylls and carotenoids) of olive fruit, expressed as mg kgⴚ1 dry weight olive pulp, from Arbequina cv at different ripening stages. Pigment
Ripening stage (fruit epidermis color) G
LG
SRS
TC
P
B
Chlorophyll a
215
106
43
21
3.2
2.4
Chlorophyll b
69
35
15
6.9
1.1
0.66
0.83
0.10
nq
Chlorophyllide a
1.1
1.7
Chlorophyllide b
0.28
0.06
1.6 nq
nq
nq
nq
Pheophytin a
nq
nq
nq
nq
nq
nq
Pheophorbide a
nq
nq
nq
nq
nq
nq
3.0
3.6
4.4
3.1
Chlorophyll a/chlorophyll b Total chlorophylls
3.1 285
3.1 143
2.9 60
3.0 28
Neoxanthin
6.1
3.9
1.8
0.82
0.13
0.06
Violaxanthin
5.3
5.1
2.8
0.91
0.08
0.02
Antheraxanthin
4.6
3.0
2.2
1.2
0.36
0.15
6.2
2.4
1.8
Lutein
30
all-trans-β-carotene
11
20
10
6.2
3.8
2.8
1.1
0.83
Violaxanthin monoesterified
nq
0.13
0.09
0.04
0.01
nq
Neoxanthin esterified
nq
0.13
0.11
0.13
0.03
nq
cis-α-carotene
tr
Total carotenoids Total pigment Chlorophylls/carotenoids
tr
tr
tr
tr
tr
57
39
21
12
4.2
2.9
342
182
82
40
8.6
5.9
1.1
1.1
5.0
3.7
2.8
2.3
G, green; LG, light-green; SRS, small reddish spots; TC, turning-colour; P, purple; B, black; nq, not quantified. tr, trace amounts. Reprinted from Criado, MN., et al., 2007. Food Chem 100, 748–755, with permission.
are rich in chlorophyll a (Thomas, 1997). Studies carried out in the olive-growing area of Andalusia (southern Spain), with fruit from the Arbequina cv, have showed that the chlorophyll a/chlorophyll b ratio is between 5 to 6 (Roca and MínguezMosquera, 2001c), values higher than showed with Arbequina cv growing in the northeast of Spain (Criado et al., 2007a). According to historical climatic data the northeast area is characterized by extensive periods of foggy days between September and December, coinciding with the olive maturity process in this area and a mean of 666 hours of sunlight; in the southern area the foggy days are scarce with a mean of
773 hours of sunlight. The lower sunlight availability in olive trees in the northeast could be one of the factors responsible for a greater development of the light-harvesting complex, relatively rich in chlorophyll b. During the olive maturity process the degradation ratio of the carotenoids is lower than chlorophylls. At the green color stage, the concentration of chlorophylls is approximately five times higher than that of the carotenoids (chlorophyll/carotenoid ratio) (Tables 7.1 and 7.2). When the olive drupes reach the black stage, coinciding with the habitual harvest period for industrial virgin olive oil production, this ratio is near to
62
SECTION | I The Plant and Production
one, indicating a major retention of carotenoids, probably as a consequence of their lower rate of degradation by the enzymatic system of the chloroplasts. The main enzymes involved in fruit pigment catabolism in chloroplast are chlorophyllase, lipoxygenase and peroxidase (Yamauchi and Minamide, 1985; Gandul-Rojas and Mínguez-Mosquera, 1996; Salas et al., 1999; GallardoGuerrero et al., 2003). The chlorophyllase enzyme (EC 3.1.1.14) catalyzes the hydrolysis of the ester bond of the chlorophyll molecule into chlorophyllide and phytol, suggesting a key role in chlorophyll degradation during the olive fruit ripening (Matile and Hörtensteiner, 1999). Additionally lipoxygenase (EC 1.13.11.12) catalyzes the oxidation of fatty acids containing a cis,cis-1-4 pentadiene system (mainly linoleic and linolenic acids) by addition of molecular oxygen to unsaturated fatty acid hydroperoxide (Feussner and Wasternack, 2002). The presence of a thylakoid-associated peroxidative activity in olives has been demonstrated in vitro, which performs chlorophyll degradation to 132-hydroxychlorophylls a and b (132-OH-Chl a and 132-OH-Chl b) and suggests the participation of this enzyme system in the formation of these compounds in olives (Gandul-Rojas et al., 2004). The activity of these catabolic enzymes varies with the olive cultivar and with the ripening stage of the olive fruit. So, differences have been observed between fruits of Farga and Arbequina cvs (Table 7.3), mainly in the chlorophyllase activity. The highest chlorophyllase activity has been observed in Arbequina cv during the first steps of the olive ripening that could explain the presence of chlorophyllides a and b in fruits from this cultivar (Table 7.2). During ripening, when the skin color of the fruit changes from green to lightgreen, both the chlorophyll content and chlorophyllase and lipoxygenase activities decrease progressively, and at the end of the ripening process, a sensible increase of the enzymatic activity is observed coinciding with harvest period for olive oil extraction. By contrast during ripening chlorophyllase
activity is not detected in fruits from Farga cv and the lipoxygenase activity remains almost constant (Table 7.3). The rapid disappearance of chlorophylls in this cultivar during fruit ripening can be attributed to this lipoxygenase activity more than to chlorophyllase activity. This peroxidative activity associated with the thylakoid membranes affects not only chlorophylls but also other accessory pigments in the photosynthetic process, such as carotenoids. The reaction of the carotenoids with the peroxy radicals generated during lipid peroxidation sets off a process of oxidation finally leading to pigment decoloration. According with this mechanism, lipoxygenase can catalyze carotenoid decoloration indirectly as a consequence of the radical species generated in its specific hydroperoxidation of fatty acids having a cis,cis-1,4- pentadiene structure (Gandul-Rojas et al., 2004). As a consequence the degradation of the carotenoid fraction during the olive fruit-ripening process could be related with the lipoxygenase activity in the olive fruit (Tables 7.1 and 7.2).
7.3 PIGMENT COMPOSITION OF VIRGIN OLIVE OIL RELATED WITH THE RIPENING STAGE OF THE OLIVE FRUIT OF ORIGIN Virgin olive oil is the oil obtained from the fruit of the olive tree solely by mechanical or other physical means under conditions, particularly thermal conditions, which maintain the composition and other organoleptic characteristics of the fruit of origin. Therefore the changes in drupes during the ripening period are directly reflected in the pigment composition of virgin olive oils. Some olive components as pigments are fairly stable in their natural environment in the fruit tissue. The oil is obtained from the olive fruit by the following operations: crushing, malaxation, extraction and decantation. The olives
TABLE 7.3 Chlorophyllase and lipoxygenase activity, expressed as nmols sⴚ1 kgⴚ1 dry weight, in olive fruit of Farga and Arbequina cvs at different ripening stages of the olive fruit. Cultivar
Farga
Arbequina
Enzymatic activity
Ripening stage (fruit epidermis color) G
LG
SRS
TC
P
B
Chlorophyllase
nd
nd
nd
nd
nd
nd
Lipoxygenase
14
13
12
11
10
8
G
LG
SRS
TC
P
B
Chlorophyllase
37
22
18
19
20
40
Lipoxygenase
12
10
11
12
14
15
G, green; LG, light-green; SRS, small reddish spots; TC, turning-color; P, purple; B, black; nd, not detected. Reprinted from Criado, MN., et al., 2006. J Amer Soc Hort Sci 131, 593–600, with permission.
63
CHAPTER | 7 The Effect of the Ripening Process on the Chlorophyll and Carotenoid Fractions
are crushed to break the cell walls and release small drops of oil included in the fruit tissue. After that the malaxation favors the dispersion of the small drops of oil in the olive paste forming a continuous oily phase, making its extraction easier. Mixing is usually done in horizontal or vertical thermo-mixers. The crushing and malaxation processes break cell walls and expose the olive matrix to enzymes (as chlorophyllase and lipoxygenase), oxygen, and mild heat. As a consequence, some modifications in olive fruit pigments take place. After the malaxation, the centrifugation in a horizontal centrifuge, or two phases decanter permets obtain the olive pomace (byproduct) and the oil. Finally the oil is separated from the wastewater by centrifugation. During the final separation phases, the native olive fruit pigments and their derivatives are distributed and partitioned between the oil and the solid fractions. During the extraction process, the acidic conditions of the olive paste favor that a proportion of the native chlorophylls
are transformed into pheophytins. At the end of the extraction process a great proportion of the chloroplast pigments of the fruit, chlorophylls and carotenoids, are transferred to the oil (Tables 7.4 and 7.5). In the case of the Arbequina cv the high chlorophyllase activity added to the acidic conditions meant that pheophorbide a is only present in the oils from this cultivar. Additionally its exclusive carotenoids, such as cis-α-carotene and esterified xanthophylls, are transferred too from the fruit to the virgin oil (Table 7.5). As a consequence the pigment profile can be used as chemicaltaxonomic identifier of virgin olive oils from Arbequina cv. The pigment concentration of virgin olive oil is clearly affected by the ripening stage of the olive fruit of origin. The concentration of pigments of the oils diminishes with the ripeness of the fruits, being the chlorophyll fraction the more affected. Analyzing the relation of chlorophyll and carotenoid fractions, at the beginning of the ripening (green and light-green stages) the corresponding oils
TABLE 7.4 Pigment composition (chlorophylls and carotenoids) of virgin olive oil, expressed as mg kgⴚ1 oil, extracted from Farga cv at different ripening stages of the olive fruit. Pigment
Ripening stage (fruit epidermis color) G
Chlorophyll a
LG
18
SRS
TC
P
6.9
1.3
0.81
0.48
0.17 0.01
Chlorophyll b
2.3
1.2
0.16
0.11
0.04
Pheophytin a
1.8
1.5
0.16
0.04
0.03
Pheophytin b
0.02
0.01
Chlorophyll a/chlorophyll b
7.9
5.8
7.9
7.8
9.6
1.6
0.96
0.54
Total chlorophylls
23
B
nq
nq
tr
nq
nq
12
12 0.18
Neoxanthin
0.55
0.21
0.07
0.03
0.02
Violaxanthin
2.4
0.80
0.20
0.11
0.03
0.03
Anteraxanthin
0.56
0.40
0.21
0.21
0.18
0.15
Lutein
4.0
2.7
1.4
0.91
0.56
0.54
all-trans-β-caroteno
3.1
1.9
0.37
0.34
0.19
0.12
Provitamin A#
508
Total carotenoids
11
Total pigment
33
Chlorophylls/carotenoids
2.1
315 6.0 16 1.6
62
57
32
tr
20
2.2
1.6
0.97
0.83
3.8
2.6
1.5
1.0
0.72
0.60
0.56
0.22
G, green; LG, light-green; SRS, small reddish spots; TC, turning-color; P, purple; B, black; nq, not quantified. tr, trace amounts. # expressed as μg retinol kg⫺1 oil. Reprinted from Criado, MN., et al., 2007. Food Chem 100, 748–755, with permission.
64
SECTION | I The Plant and Production
TABLE 7.5 Pigment composition (chlorophylls and carotenoids) of virgin olive oil, expressed as mg kg⫺1 oil, extracted from Arbequina cv at different ripening stages of the olive fruit. Pigment
Ripening stage (fruit epidermis color) G
Chlorophyll a
LG
31
SRS
24
TC
P
B
4.6
2.1
0.82
0.40 0.03
Chlorophyll b
5.1
3.8
0.94
0.35
0.09
Pheophytin a
8.7
4.9
2.3
0.88
0.04
tr
Pheophorbide a
1.7
0.25
0.07
0.02
0.01
nq
Chlorophyll a/chlorophyll b
6.0
6.5
4.9
6.0
8.8
13
7.9
3.4
0.97
0.43
Total chlorophylls
46
33
Neoxanthin
1.3
1.0
0.38
0.14
0.08
0.02
Violaxanthin
2.1
1.6
1.0
0.60
0.11
0.03
Antheraxanthin
2.2
2.0
1.0
0.58
0.29
0.23
Lutein
8.1
6.5
3.1
2.1
1.3
0.71
all-trans-β-carotene
4.2
2.4
1.0
0.98
0.42
0.28
Violaxanthin monoesterified
0.14
0.08
0.05
0.02
0.01
Neoxanthin esterified
0.26
0.22
0.10
0.06
0.02
nq 0.02
Cis-α-carotene
tr
tr
tr
tr
tr
tr
Provitamin A#
697
400
167
164
70
47
Total carotenoids
18
14
6.7
4.5
2.3
1.3
Total pigment
64
47
15
7.8
3.2
1.7
1.2
0.7
0.4
Chlorophylls/carotenoids
2.5
2.4
0.3
G, green; LG, light-green; SRS, small reddish spots; TC, turning-color; P, purple; B, black; nq, not quantified. tr, trace amounts. # expressed as μg retinol kg⫺1 oil. Reprinted from Criado, MN., et al., 2007. Food Chem 100, 748–755, with permission.
present the double of chlorophyll pigments that of carotenoids. Nevertheless, when the olive fruits acquire a reddish coloration (small reddish spots) the content of carotenoids in the oils is higher than chlorophylls (Tables 7.4 and 7.5). From the nutritional point of view, the provitamin A value is obtained from the carotene contents with β-ionone rings and the results are expressed as μg of retinol by kg of oil, taking into account that μg retinol ⫽ 0.167 ⫻ μg β-carotene ⫹ 0.083 ⫻ (μg α-carotene ⫹ μg γ-carotene ⫹ μg β-cryptoxanthin) (Bauernfeind et al., 1971). According to the gradual loss of pigments with the fruit ripening, the more
highly pigmented oils from the green and light-green olives show a higher content of provitamin A (Tables 7.4 and 7.5). Additionally to the ripening stage of the olive fruit there is a clear effect of olive cultivar on the chlorophyll and carotenoid content in the oil, related with the differences in the percentages of pigment transferred from the fruit to the oil during extraction process (Figure 7.1). The percentage of transference of pigments seems to be related with the olive paste moisture (Table 7.6). At the first stages of ripeness, the transference of pigments from the fruit to the oil is lower than the last stages according with the higher
65
CHAPTER | 7 The Effect of the Ripening Process on the Chlorophyll and Carotenoid Fractions
60
14 50
12 10
40
8
30
6
20
4 10
2
0
0 G
LG
SRS
TC
P
20
10
Oil yield (%)
Pigment transference (%)
16
Arbequina cv
Olive paste moisture (%)
70
18
0
B
Ripening stage 70 Farga cv 60
14 50
12 10
40
8
30
6 20 4 10
2 0
0 G
LG
SRS
TC
P
20
10
Oil yield (%)
16
Olive paste moisture (%)
Pigment transference (%)
18
0
B
Ripening stage Chlorophylls
Olive paste moisture
Carotenoids
Oil yield
FIGURE 7.1 Percentage of pigment transference from fruit to oil during the oil extraction process in relation to the olive ripening. Ripening stage: G, green; LG, light-green; SRS, small reddish spots; TC, turning-color; P, purple; B, black. The histograms show the mean of three sets of analysis of five lots of three kilograms of olive fruits with a determined ripening index. The oil was extracted using an Abencor system simulating the industrial processes of milling, malaxing, centrifuging and decanting.
TABLE 7.6 Pit percentage and pulp water content of the olive fruit, and oil yield (expressed as olive paste) from Arbequina and Farga cvs at different ripening stages of the olive fruit. Cultivar
Arbequina
Parameter G
LG
SRS
TC
P
B
% Pit
24.6
22.2
19.8
18.2
18.2
16.9
% Water content
63.2
58.8
58.7
56.0
48.8
49.2
9.5
15.7
14.4
16.2
16.5
21.6
G
LG
SRS
TC
P
B
% Pit
22.4
19.5
18.3
17.3
17.2
15.9
% Water content
66.5
64.2
60.5
59.9
53.1
47.8
5.7
9.6
12.0
12.6
18.5
21.9
% Oil yield#
Farga
Ripening stage (fruit epidermis color)
% Oil yield#
G, green; LG, light-green; SRS, small reddish spots; TC, turning-color; P, purple; B, black; nq, not quantified. tr, trace amounts. # Oil yield of the olive paste (%).
66
SECTION | I The Plant and Production
7.4 EFFECT OF THE OLIVE-RIPENING STAGE ON CHLOROPHYLL AND CAROTENOID PARTITIONING DURING THE OIL EXTRACTION PROCESS
olive paste moisture that results in a more fluid olive paste. During the crushing operation the more water content of the fruit results in a rapid crossing of the hammer-crush sieves, suffering less tissue crushing and in consequence a reduction in the extraction of pigments. Additionally the higher percentage of pigment transference observed in Arbequina cv could be related with the pit percentage in fruits of this cultivar (Table 7.6). The presence of the high proportion of the pit pieces in the olive paste could favor the breakdown of the cellular membranes and thus favor the pigment transfer from the olive paste to the oil during the malaxation and centrifugation steps. Additionally to the effect of the olive paste moisture on the pigment transference, during the extraction process, the destruction of chlorophylls is greater than that of carotenoids at all the stages of ripeness (Figure 7.1). Probably the major destruction of chlorophylls is the consequence of the chlorophyllase and lipoxygenase enzymes which activity is favored by the temperature conditions during the malaxation step (Table 7.3).
During the first steps of the oil extraction process, the crushing of the olives in order to break down the cellular membranes and thus release small drops of oil results in an olive paste that is a multiphasic system and the pigment partitioning into phases is thermodynamically according to their affinities toward these phases (Rodis et al., 2002). The lipophylic nature of the chloroplast pigments, chlorophylls and carotenoids favors their transference from the olive paste to the oil. However, the anthocyanin fraction, responsible for the purple and black color of the olive fruit at the end of the ripening process, remains in the solid phase (pomace) as a consequence of its hydrophylic nature. Therefore additionally to the ripening stage of the olive fruit and the enzymatic degradation during the oil extraction
Total chlorophylls
75
Ripening stage
LG
42
TC
8
13
45
68
B
0%
20%
16
40%
17
60%
16
80%
100%
LG
TC
B
75
42
8
13
68
Chlorophyll b
45
16
LG
17
16
Ripening stage
Ripening stage
Chlorophyll a
B
61
2
4
78
22
35
4 18
0% 20% 40% 60% 80% 100%
0% 20% 40% 60% 80% 100% Pomace
TC
76
Oil
Losses
FIGURE 7.2 Percentage of the olive paste chlorophylls (100%) retained in the pomace, transferred to the olive oil, and degraded (losses) during the virgin olive oil extraction process in three ripening stages of the olive fruit. Ripening stage: LG, light-green; TC, turning-color; B, black. The histograms show the mean of three sets of analysis of five lots of three kilograms of olive fruits with three ripening indexes. The percentage of the chlorophylls transferred from the olive paste to the oil and the percentage retained in the pomace are calculated considering the total mass balance (inputs and outputs) during the olive oil extraction process. The difference between the chlorophyll content of the pomace (byproduct) and oil corresponds to the percentage of chlorophyll losses by oxidative and enzymatic degradations.
67
CHAPTER | 7 The Effect of the Ripening Process on the Chlorophyll and Carotenoid Fractions
process, the pigment partition between the different phases determines the pigment composition of virgin olive oil. The oil extraction process has one input, olive paste, and two outputs: pomace (byproduct) and oil. The percentage of the pigments transferred from the olive paste to the oil, and the percentage retained in the pomace, are calculated considering the total mass balance (inputs and outputs) during the olive oil extraction process (Criado et al., 2007b). The difference between the pigment content of the pomace (byproduct) and oil corresponds to the percentage of pigment losses by oxidative and enzymatic degradations (Figures 7.2 and 7.3). In general, pigment retention in pomace is high, and this means important losses of the olive fruit pigment together with the pigment degradation. As a consequence the percentage of pigments from olive fruit in virgin olive oils is relatively low. The percentage of the pigments, chlorophylls and carotenoids, transferred from the olive paste (100% as reference) to the oily phase (oil) and the percentage of pigments retained in the pomace (byproduct) are shown in Figures 7.2 and 7.3 respectively, considering three ripening stages of the olive fruit (raw material). When the ripening stage of the olive fruit advances the pigment transference from olive paste to the oil increases, mainly the carotenoid fraction. The more important pigment losses
in pomace are observed with fruits in the first stages of the ripening (light-green) probably related to the higher water content in the olive paste (Table 7.6) that causes the pigment transference to the oil. At the same time, important losses from degradation are observed, mainly chlorophylls, in the process that corresponds to turning-color ripening stage. This higher degradation could be related to the slight increase in the chlorophyllase activity at the latest stages of ripening (Table 7.3). In relation to the carotenoid fraction, while the percentage of transference of β-carotene is higher than lutein in oils from light-green fruits in oils from the more advanced ripening stages (black) the percentage of transference of lutein is higher. In conclusion the total pigment content in olive drupe suffers a decrease when the ripening process begins, and the chlorophyll degradation is more marked than that of the carotenoid pigment. Additional to the ripening effect, the transference of chloroplast pigments, chlorophylls and carotenoids from fruit to oil is conditioned by the olive paste moisture, and by the destruction of pigments as a consequence of the chlorophyllase and lipoxygenase enzymatic activities favored by the temperature conditions during the oil extraction process.
Total carotenoids
70
Ripening stage
LG
20 20
Pomace 52 52
TC
24 24
24
Oil Losses
54
B 0%
20%
30
40%
60%
16
80%
Lutein 73 73
LG
13 14 13 14
TC
61 61
17 22 17 22
B
63 63
30 77 30
0%
100%
β -carotene
20% 40% 60% 80% 100%
Ripening stage
Ripening stage
10
72 72
LG
61 61
TC
47 47
B 0%
20% 40%
24 44 24 22 17 17 22 28 28
25 25
60% 80% 100%
FIGURE 7.3 Percentage of the olive paste carotenoids (100%) retained in the pomace, transferred to the olive oil, and degraded (losses) during the virgin olive oil extraction process in three ripening stages of the olive fruit. Ripening stage: LG, light-green; TC, turning-color; B, black. The histograms show the mean of three sets of analysis of five lots of three kilograms of olive fruits with three ripening indexes. The percentage of the carotenoids transferred from the olive paste to the oil and the percentage retained in the pomace are calculated considering the total mass balance (inputs and outputs) during the olive oil extraction process. The difference between the carotenoid content of the pomace (byproduct) and oil corresponds to the percentage of carotenoid losses by oxidative and enzymatic degradations.
68
SECTION | I The Plant and Production
SUMMARY POINTS ●
●
●
●
●
●
●
●
●
The composition of the chlorophyll fraction of the olive fruit comprises chlorophylls a and b, and the carotenoids that typically accompany the chlorophylls in the chloroplast: lutein, β-carotene, violaxanthin, neoxanthin, and antheraxanthin. During the ripening process, the gradual disintegration of the chloroplasts involves the disappearance of the chloroplast pigments, mainly chlorophylls. Olive fruits from Arbequina cultivar have exclusive carotenoids: these include cis-α-carotene, β-cryptoxanthin and esterified xanthophylls, that can be considered as chemical-taxonomic differentiators of this olive cultivar. The chlorophyllase enzyme (EC 3.1.1.14) catalyzes the hydrolysis of the ester bond of the chlorophyll molecule into chlorophyllide and phytol, suggesting a key role in chlorophyll degradation during the olive fruit ripening. The highest chlorophyllase activity has been observed in Arbequina cv during the first steps of the olive ripening that could explain the presence of chlorophyllides a and b in fruits from this cultivar. The degradation of the carotenoid fraction during the olive fruit-ripening process could be related to the lipoxygenase activity in the olive fruit more than with the chlorophyllase activity. During the extraction process, the acidic conditions of the olive paste favor that a proportion of the native chlorophylls are transformed into pheophytins. When the ripening stage of the olive fruit advances the pigment transference from olive paste to the oil increases, mainly the carotenoid fraction. Additional to the ripening effect, the transference of chloroplast pigments, chlorophylls and carotenoids, from fruit to oil is conditioned by the olive paste moisture, and by the destruction of pigments as a consequence of the chlorophyllase and lipoxygenase enzymatic activities favored by the temperature conditions during the oil extraction process.
REFERENCES Bauernfeind, J.C., Brucbacher, G.B., Kläui, H.M., Marusich, W.L., 1971. Use of carotenoids. In: Birkhäuser Verla, I.O. (ed.), Carotenoids. Basel, Switzerland, pp. 744–750. Criado, M.N., Motilva, M.J., Ramo, T., Romero, M.P., 2006. Chlorophyll and carotenoid profile andenzymatic activities (Chlorophyllase and lipoxigenase) in olive drupes from the fruit-setting period to harvest time. J. Am. Soc. Hortic. Sci. 131, 593–600. Criado, M.N., Motilva, M.J., Goñi, M., Romero, M.P., 2007a. Comparative study of the effect of the maturation process of the olive fruit on the
chlorophyll and carotenoid fractions of drupes andvirgin oils from Arbequina and Farga cultivars. Food Chem. 1000, 748–755. Criado, M.N., Romero, M.P., Motilva, M.J., 2007b. Effect of the technological and agronomical factors on pigment transfer during olive oil extraction. J. Agric. Food Chem. 55, 5681–5688. Escolar, D., Haro, M.R., Ayuso, J., 2007. The color space of foods: Virgin olive oil. J. Agric. Food Chem. 55, 2085–2093. Feussner, I., Wasternack, C., 2002. The lipoxygenase pathway. Annu. Rev. Plant Biol. 53, 275–297. Gallardo-Guerrero, L., Jarén-Galán, M., Hornero-Méndez, D., MínguezMosquera, M.I., 2003. Evidence for the involvement of lipoxygenase in the oxidative processes associated with the appearance of green staining alteration in the Gordal olive. J. Sci. Food Agric. 83, 1487–1492. Gandul-Rojas, B., Mínguez-Mosquera, M.I., 1996. Chlorophyllase activity in olive fruits and its relationship with the loss of chlorophyll pigments in the fruits and oils. J. Sci. Food Agric. 72, 291–294. Gandul-Rojas, B., Roca, M., Mínguez-Mosquera, M.I., 1999. Chlorophyll and carotenoid patterns in olive fruits, Olea europaea cv. Arbequina. J. Agric. Food Chem. 47, 2207–2212. Gandul-Rojas, B., Roca, M., Mínguez-Mosquera, M.I., 2004. Chlorophyll and carotenoid degradation mediated by thylakoid-associated peroxidative activity in olives (Olea europaea) cv. Hojiblanca. J. Plant Physiol. 161, 499–507. Gross, J., 1991. Pigments in Vegetables: Chlorophylls and Carotenoids. Van Nostrand Reinhold, New York. Hornero-Méndez, D., Gómez-Ladrón, R., Mínguez-Mosquera, M.I., 2000. Carotenoid biosynthesis changes in five red pepper (Capsicum annuum L.) cultivars during ripening. Cultivar selection for breeding. J. Agric. Food Chem. 48, 3857–3864. Kozukue, N., Friedman, M., 2003. Tomatine, chlorophyll, β-carotene and lycopene content in tomatoes during growth and maturation. J. Sci. Food Agric. 83, 195–200. Matile, P., Hörtensteiner, S., 1999. Chlorophylls degradation. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 67–95. Roca, M., Mínguez-Mosquera, M.I., 2001a. Changes in chloroplast pigments of olive varieties during fruit ripening. J. Agric. Food Chem. 49, 832–839. Roca, M., Mínguez-Mosquera, M.I., 2001b. Unusual carotenogenesis in fruits with pronounced anthocyanic ripening (Olea europaea var, Arbequina). J. Agric. Food Chem. 49, 4414–4419. Roca, M., Mínguez-Mosquera, M.I., 2001c. Change in the natural ratio between chlorophylls and carotenoid in olive fruit during processing for virgin olive oil. J. Am. Oil Chem. Soc. 78, 133–138. Rodis, P.S., Karathanos, V.J., Mantzavinou, A., 2002. Partitioning of olive oil antioxidants between oil and water phases. J. Agric. Food Chem. 50, 596–601. Ronen, G., Cohen, M., Zamir, D., Hirschberg, J., 1999. Regulation of carotenoid biosynthesis during tomato fruit development: expression of the gene for lycopene epsilon ciclase is down-regulating during ripening and is elevated in the mutant. Delta Plant J. 17, 3857–3864. Salas, J.J., Williams, M., Harwood, J.L., Sánchez, J., 1999. Lipoxygenase activity in olive (Olea europaea) fruit. J. Am. Oil Chem. Soc. 76, 1163–1168. Thomas, H., 1997. Chlorophyll a: a symptom and a regulator of plastid development. New Phytol. 136, 163–181. Yamauchi, N., Minamide, T., 1985. Mechanism of chlorophyll degradation in harvested leaf vegetables. 1. Chlorophyll degradation by peroxidase in parsley leaves. J. Jpn. Soc. Hortic. Sci. 54, 265–271.
Chapter 8
Influence of the Crushing System: Phenol Content in Virgin Olive Oil Produced from Whole and De-stoned Pastes Paolo Amirante1, Maria Lisa Clodoveo1, Antonia Tamborrino1, Alessandro Leone2 and Alistair G. Paice3 1
Department of Engineering and Management of the Agricultural, Livestock and Forest Systems, University of Bari, Italy Department of Production and Innovation in Mediterranean Agriculture and Food Systems (PriMe), University of Foggia, Italy 3 Department of Clinical Biochemsitry, Nutrition and Dietetics, King’s College London, UK
2
8.1 INTRODUCTION
results in an increase in oxidative stability and nutritional value of virgin olive oil (Servili and Buonaurio, 1999).
The mechanical processes used to extract virgin olive oil from olive fruit include the crushing of the olives, malaxation of resulting pastes and separation of the oily phase essentially by pressure or centrifugation. All these operations affect the quality of virgin olive oil. Olive paste preparation is the most important phase of the process when oil is mechanically extracted from the olives. The use of differing machines in olive oil production has inevitable repercussions on the cost-effectiveness of the oil-making process, on the amounts of oil extracted and especially on the quality of the oil obtained (Servili et al., 2004). In order to obtain the best virgin olive oil quality olives must be processed as quickly as possible after harvesting from the trees. Normally they should be delivered to the mill within a day or so of picking, in order to keep down oxidation and acidity. When extracting oil from olives, it is very important to clean them properly first, in order to ensure the levels of hygiene required for a high-quality olive oil product. Once cleaned, the olives must be crushed, in order to make the olive paste that itself is the first stage in extracting the oil. How this is done is crucial for both the quantity and the quality of the olive oil product. Mechanical oil extraction from de-stoned pastes can improve the oil phenolic concentration. The quality characteristics of virgin olive oil also depend on oxidative enzyme reactions that take place in olive paste during the extraction process. Two enzymes, polyphenol oxidase (PPO) and peroxidase (POD), are highly concentrated in the olive kernel. PPO and POD can oxidize phenolic compounds resulting in a reduced phenolic concentration of oil. The destoning process, excluding the olive seed before malaxation, partially removes the peroxidase activity in the pastes. This Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
8.2 MACHINES FOR OLIVE CRUSHING Crushing is normally carried out using a traditional stone mill or by means of a hammer or disk crusher. Characteristically the traditional stone mills have batch-processing potential whilst the mechanical crushers have varying structural features. For example, they have crushing devices that are designed and built a variety of ways to ensure continuous on-line olive processing and decanting for oil extraction. The stone mill consists of three stone rollers or wheels, which roll in circles on a slab of granite to grind the olives into a paste. In hammer crushing machines, a three- or four-lobe rotor with wear-resistant metal plates crushes the olives against a stationary grid. The diameter of the grid holes determines the thickness of the paste. Disk crushing machines, on the other hand, crush the olives between two toothed disks – one stationary and one that rotates. A new technology in virgin olive oil production is olive de-pitting. This ensures that the paste consists solely of the fleshy part of the olive (mesocarp), without the stone or pit (endocarp) that holds the seed. The de-stoner consists of a cylindrical perforated stationary grill and a rotary shaft. The olives are pushed by centrifugal force towards the perforated grill. Olive tissue crosses the grill whilst the kernel remains inside the cylinder. Using this method the grinding of pulp tissues is not drastic. Table 8.1 illustrates the machines employed in olive oil production for the crushing step. Each machine has its advantages and disadvantages. The choice of crushing system appears to depend on subjective circumstances. 69
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
70
SECTION | I Olives and Olive Oil
TABLE 8.1 Advantages and disadvantages of the machines employed in the crushing step of the extraction process for olive oil production. Machines for olive crushing
Advantages
Disadvantages
Can be adapted to olive and pit size to optimize paste characteristics Doesn’t cut the skin, releasing less chlorophyll Formation of larger size drops of oil, minimizing mixing times Paste isn’t heated Less phenols so less bitter oil
Bulky Stone is difficult to clean Olives will prolong grinding time Costly Discontinuous process – time consuming High labor cost Lower level of phenols extracted Shorter life
Continuous Less costly Precise regulation of olive paste fragment size High throughput may give the oil a longer shelf-life
Olive paste fragment size not easily modified May form emulsion which impedes oil–water separation Organoleptic characteristics affected – oil can have stronger, spicy taste (this may be an advantage with mild olives) One stone can break an expensive tooth on the grinder Oil paste may heat up
Continuous High throughput Tolerates debris such as rocks and grit Known, perfected technology Extracts more phenols so oil has longer shelf-life
May form emulsion which impedes oil–water separation Organoleptic characteristics affected – oil can have bitter, stronger, spicy taste due to more phenols Wear and tear of metal parts Oil paste may heat up
Higher phenol level Oils subjected to less heating due to stone fracturing De-stoned pomace is easier to use as animal feed The stones can be burned to create heat to dry a watery pomace for easier disposal
May form emulsion which impedes oil–water separation Requires longer mixing times Requires additional machinery at additional expense Slightly lower oil yields
Stone mill
Disk crusher
Hammer crusher
De-stoner
CHAPTER | 8 Influence of the Crushing System
8.3 THE ROLE OF POLYPHENOLS IN VIRGIN OLIVE OIL QUALITY Virgin olive oil contains phenyl-acids, phenyl-alcohols and several secoiridoid derivatives such as the dialdehydic form of elenolic acid linked to 3,4-DHPEA, or p-HPEA (3,4-DHPEA-EDA or p-HPEA-EDA) and an isomer of the oleuropein aglycon (3,4-DHPEA-EA), which is the most concentrated phenolic compound found in virgin olive oil (Servili et al., 2002; Servili and Montedoro, 2002). Furthermore, recently lignans have also been found in virgin olive oils. The antioxidant activity of 3,4-DHPEA, p-HPEA and phenyl acids has been studied, and the high antioxidant activity of 3,4-DHPEA, 3,4-DHPEA-EDA and 3,4-DHPEAEA has been demonstrated. Several nutritional properties of secoiridoids have also been recognized: (a) inhibition of blood platelet aggregation and involvement in the synthesis of thromboxane in the human cells; (b) inhibition of phospholipids and LDL oxidation; (c) protection of the human erythrocytes from the oxidative damage. Taken together, these factors indicate a role for secoiridoids in prolonging the shelf-life of olive oil and suggest a possible protective effect with respect to the risk of thrombosis and the onset of atherosclerotic damage in consumers. Minor dietary constituents of virgin olive oil would seem to have major biochemical importance. The sensory quality of virgin olive oil is strictly related to phenolic and volatile compounds. Phenolic compounds are directly responsible for any bitter and pungent taste in the oil and their concentration affects the shelf-life of the product. Several volatile compounds, particularly aliphatic alcohols, low-molecular-weight terpens, chetons, ethers, furan and thiophene derivatives, and, most importantly, C5, C6 and C9 saturated and unsaturated alcohols and aldehydes, determine the aroma of the oil. The release of phenols and the formation of volatiles are two basic components of virgin olive oil quality that directly relate to the mechanical extraction process itself. In this ambit, control of endogenous enzymes of olive fruit during processing is the most critical point in the mechanical extraction process of olive oil. In fact, the secoiridoid concentration in the virgin olive oil is largely due to the activation of the glycosidases of olive fruit that activate the formation of aglycon, while the oxidoreductases such as polyphenoloxidase (PPO) and peroxidase (POD) can catalyze their oxidation during the oil mechanical extraction process and subsequently trigger the autoxidation mechanism.
8.4 THE INFLUENCE OF OLIVE PASTE PREPARATION MACHINES ON OLIVE OIL CHARACTERISTICS Catalano and Caponio (1996) compared the use of stonemills with hammer crushers. Their comparisons offered significant insights into the phenomena occurring in paste
71
preparation and the subsequent interactions taking place amongst the components of the drupes. They also indicated the most rational use of either of the two machines in industrial practice. When the paste is prepared using hammer crushers the oils produced contain greater amounts of polyphenols than by preparation with a stone mill. This frequently corresponds to an increased resistance to autoxidation and causes a marked ‘bitterness’ and ‘pungency’. This is particularly true for oils extracted from certain cultivars, for example Coratina, which can lead to an unpleasant experience for the inexpert consumer. In order to upgrade the organoleptic quality of EVOO and to enhance their preservation, two systems have been suggested for processing procedures. For olives yielding oils rich in total polyphenols (of the Coratina cultivar and non-blackened or slightly blackened drupes) it is best to use the stone mill or any innovative system that reproduces the milling process such as the disk crusher. However, the hammer crusher is more suitable in processing olives yielding very ‘sweet’ oils with a low content of polyphenols and that are thus not very preservable. These findings have been confirmed by Di Giovacchino et al. (2002). When olive oil mills are equipped with pressure systems, olive crushing is generally carried out by granite millstones (with 2–6 stones) for 20–30 minutes. The resulting olive paste is squeezed by a hydraulic press. This approach ensures that good oil extraction yields are obtained. When olive oil mills are equipped with centrifugation systems olive crushing is generally carried out using metallic crushers such as mobile or fixed hammers with toothed disks. These crushers have a high working capacity and exert a violent action that breaks the cells of the olive flesh containing oil. The resulting paste leads to good extraction yields after a suitable malaxation step. Olive crushing can increase the temperature of the olive paste due to the frictional energy generated by the rotating crushers. Table 8.2 illustrates the effect of crushing method on assorted qualitative characteristics of the oils extracted using a three-phase centrifugal decanter. The data indicate that the crushing method, i.e. using a stone mill (gentle) or a metallic crusher (very violent), does not influence qualitative parameters such as free fatty acid percentage, peroxide value, specific spectrophotometric absorptions in the UV region or the organoleptic assessment. The crushing method, however, does have a clear influence on the total phenol content of oils (Table 8.2). The use of the more violent metallic crushers results in an oil with a total phenol content higher than that obtained using a stone mill. This is due to the more complete disruption of olive flesh that liberates higher quantities of phenolic substances. These then bond to the different cellular tissues of the olive flesh, and hence increase their concentration in the olive paste. Caponio and Catalano (2001) used a hammer crusher and a disk crusher in order to evaluate the effect of differing processing temperatures on the quality of the virgin
72
SECTION | I Olives and Olive Oil
TABLE 8.2 Qualitative characteristics of virgin olive oils extracted with three-phase centrifugal decanters from olive paste obtained using differing crushing methods. Olive cultivar
Crushing method
Free Peroxide value acidity [%] [meqO2 kg⫺1]
K232
Organoleptic Total phenols Induction Bitterness assessment [mg/L] time [h] intensity (score) (score)
Coratina
Stone mill Fixed-hammers metallic crusher
0.40 0.37
6.5 5.4
1.18 1.20
– –
228 411
9.2 11.9
– –
Peranzana
Stone mill Disks metallic crusher
0.23 0.23
11.5 11.7
1.87 1.90
7.4 7.2
133 247
7.8 10.6
1.8 2.4
Adapted from: Di Giovacchino, L., Sestili, S., Di Vincenzo, D., 2002. Influence of olive processing on virgin olive oil quality. Eur. J. Lipid. Sci. Tech. 104, 587– 601. The influence of the crushing method on assorted qualitative characteristics of oils extracted by a three-phase centrifugal decanter. The crushing method, using stone mill (gentle) or metallic crushers (very violent), does not influence qualitative parameters such as free fatty acid percentage, peroxide value, specific spectrophotometric absorptions in the UV region or organoleptic assessment. The crushing method, however, does have a clear influence on the total phenol content of oils. Utilizing more violent metallic crushers helps to obtain oil with a higher total phenol content than that obtained using a stone mill. This is due to the more complete breakage of olive flesh liberating higher quantities of phenolic substances. These are then bonded to the different cellular tissues of the olive flesh, which increases their concentration in the olive paste. The data confirm that virgin olive oils obtained from the centrifugation of olive paste prepared by a metallic crusher have a higher induction time (h) as measured by the Rancimat apparatus.
1.0
0.8 Frequency
olive oils obtained. The results showed that the hammer crusher produced more intense fragmentation of the olive pits than the disk crusher (Figure 8.1), resulting in a substantial increase in output temperature. Higher temperatures in the crusher during olive processing lead to a shorter shelf-life for the resulting oils. All of the analysis performed so far demonstrates that the oils obtained from hammer-crushed pastes degrade at a greater rate than those from disk-crushed pastes. The data suggest that the temperatures reached during fast olive crushing – either with traditional hammer crushers or with disk crushers – influence the quality and preservation of olive oils and that these oils are more susceptible to auto-oxidation if they are produced with a hammer crusher rather than with a disk crusher.
0.6
0.4
0.2
0.0 0
1
2
3
4
5
Olive-pit fragment size (mm)
8.5 EFFECT OF DE-STONING ON OLIVE OIL QUALITY Mechanical extraction of the olive oil from de-stoned paste has several potential advantages. For example, the technology used might improve virgin olive oil quality. The associated machinery might improve the working capacity of the mill plant as the seeds, which constitute about 25% of the total paste volume, would be excluded. The oil seed produced as a byproduct could be used in the cosmetics and pharmaceutical industries; as could the stones in the production of coal for pharmaceutical and food industries. However, thus far, several disadvantages have emerged. De-stoning of the olive paste leads to a reduction of the oil yield to about 1.5 kg of oil per 100 kg of olives in comparison to the traditional process (Amirante et al., 1987). About 30% of this reduction is due to the oil fraction
FIGURE 8.1 Olive-pit fragment sizes in the pastes obtained with the ‘hammer crusher’ and ‘disk crusher’ crushing systems. Stone sizes obtained when hammer crushers and disk crushers are employed to crush olives. In both cases the distribution curves are Guassian. The mechanical thrust exerted by the hammer crusher is stronger than that of the disk crusher, as evidenced by the smaller size of stone fragment produced by the latter. Adapted with permission from: Caponio F. and Catalano P. (2001). Hammer crushers vs disk crushers: the influence of working temperature on the quality and preservation of virgin olive oil. European Journal of Lipid Sciences and Technology 213: 219–224.
present in the kernel and thus lost in the de-stoning process. The remaining losses reflect extraction processes that, over the course of time, have been improved with respect to the rheological characteristics of the whole olive paste. In order to correct for these disadvantages, Amirante et al. (2006) employed a heat exchanger coupled to the stoner to improve the olive oil yields and to render the malaxation process more efficient. The absence of stone fragments
73
Yield (kg of oil/kg of olives)
CHAPTER | 8 Influence of the Crushing System
20 18 16 14 12 10 8 6 4 2 0
TABLE 8.3 Standard quality parameters, induction time and polyphenol content of the olive oils obtained by three different paste preparation techniques during three harvesting years (Amirante et al., 2006). SM m Stone mill
De-stoner
De-stoner + heat exchanger
Paste preparation technique
FIGURE 8.2 Mean olive oil extraction yields obtained by three different paste preparation techniques during three crop seasons. The use of the de-stoner causes a lower olive oil extraction yield. The introduction of the heat exchanger between the de-stoner and the malaxer represents a system improvement to increase the efficiency of malaxation of de-stoned paste. The heat exchanger corrected for the lower yields obtained using the de-stoner alone. Adapted with permission from: Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A. and Tamborrino, A. (2005). Advance technology in virgin olive oil production from traditional and de-stoned pastes: Influence of the introduction of a heat exchanger on oil quality. Food Chemistry 98(4): 797–805.
causes a change of the olive paste viscosity and makes the malaxation effect less efficient. The aim of malaxation is to merge the small droplets of the oil formed during the milling into large drops that can be easily separated when passing through the mechanical system. Introduction of the heat exchanger increases the efficiency of malaxation of de-stoned paste. Figure 8.2 shows the mean olive oil extraction yields obtained from three different paste preparation techniques during three crop seasons. The analytical characteristics of the oils are shown in Table 8.3. In general, acidity, peroxide index and UV spectrophotometric indices were not influenced by the type of crushing method used to prepare olive pastes. The healthy quality of virgin olive oil is related to the concentration of phenolic antioxidants, which is in turn strongly influenced by the parameters of oil mechanical extraction process. The oils obtained from de-stoned pastes had a higher polyphenol content compared to those obtained from whole paste (Table 8.3). The increase in phenolic compounds lengthens the time until the oxidation induction. De-stoned oils had a higher content of C5 and C6 volatile compounds, responsible for positive flavor notes, in comparison to oils obtained by the stone mill. The de-stoner oil had a higher sensory score than the traditional one and consequently a higher market value. Luaces et al. (2007) assessed the effect of fruit de-stoning on the phenolic profile of three Spanish olive oils and studied the role of olive seed in the modification of phenolic compounds during the production of virgin olive oil. De-stoned pulp was immediately processed to extract ‘de-stoned’ oil. Oils from whole paste were produced by mixing olive pulp and stones fragments. Olive oil samples from whole paste were then extracted, increasing the seed proportion. They
DS SD
m
DH SD
m
SD
Olive paste preparation techniques 2000/2001 A
0.21 ns
0.02
0.20 ns
0.03
0.19 ns
0.03
PV
4.7 a
0.05
4.5 b
0.04
4.5 b
0.06
K232
1.71 a
0.02
1.68 b
0.01
1.66 b
0.02
K270
0.12 a
0.00
0.12 a
0.00
0.10 b
0.01
IT
15.2 c
0.4
18.2 b
0.3
19.1 a
0.5
TP
213 c
43
382 b
28
445 a
20
SS
6.4 c
0.2
7.9 b
0.1
8.7 a
0.2
2001/2002 A
0.33 ns
0.04
0.30 ns
0.02
0.31 ns
0.02
PV
6.0 b
0.07
5.8 c
0.05
6.1 a
0.06
K232
1.83 b
0.01
1.84 a
0.02
1.82 ab
0.02
K270
0.14 a
0.00
0.12 b
0.01
0.14 a
0.00
IT
15.4 c
0.5
18.2 b
0.7
19.2 a
0.5
TP
235 b
43
399 a
59
447 a
21
SS
7.0 c
0.2
7.6 b
0.2
8.4 a
0.3
2002/2003 A
0.25 ns
0.01
0.26 ns
0.02
0.24 ns
0.02
PV
5.5 b
0.04
5.3 c
0.05
5.6 a
0.04
K232
1.71 ns
0.03
1.75 ns
0.01
1.73 ns
0.02
K270
0.12 a
0.00
0.11 b
0.00
0.11 b
0.01
IT
15.8 c
0.8
17.7 b
0.8
18.8 a
0.3
TP
237 b
39
388 ab
38
418 a
46
SS
6.2 c
0.2
6.8 b
0.2
8.2 a
0.2
Data represent mean value (m) and standard deviation (SD). Significant differences in the same row are shown by different letters ( p ⬍ 0.05). Abbreviations: SM, stone mill; DS, de-stoner; DH, destoner equipped with a heat exchanger; A, acidity (% of oleic acid); PV, peroxide value (meq of oxygen/kg of oil); IT, induction time (h at 120°C); TP, total phenols (mg kg⫺1); SS, sensory score. Adapted with permission from: Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A., Tamborrino, A., 2005. Advance technology in virgin olive oil production from traditional and de-stoned pastes: Influence of the introduction of a heat exchanger on oil quality. Food Chem. 98 (4), 797–805.
74
SECTION | I Olives and Olive Oil
TABLE 8.4 The POD activity in extracts of olive pulp and seed, and the effect of virgin olive oil phenolics on olive seed POD activity. VOO phenolics POX activity (nmol) (U g⫺1 FW)a 0.10 ⫾ 0.01
Pulp Seed
TABLE 8.5 Effect of olive seed POD (POX) on virgin olive oil phenolics. Control
POX-incubateda
HTyr
0.065
0.000
0.0
DGO
0.137
0.000
0.0
AGO
0.741
0.000
0.0
∑ o-diphenols
0.943
0.000
0.0
Tyr
0.022
0.020
90.9
DGL
0.045
0.010
22.2
AGL
0.035
0.033
94.3
∑ monophenols
0.102
0.063
61.8
∑ other phenols
0.027
0.006
22.2
∑ PHENOLICS
1.072
0.069
6.4
%
% activity
0.13
0
72.40 ⫾ 0.01
318
48.38 ⫾ 0.01
66.82
636
22.20 ⫾ 0.01
30.66
100
a Mean of three extractions. Abbreviations: VOO: Virgin olive oil; POX (POD) peroxidase. Adapted with permission from: Luaces, P., Romero, C., Gutierrez, F., Sanz, C., G Perez A., 2007. Contribution of olive seed to the phenolic profile and related quality parameters of virgin olive oil. J. Sci. Food Agric. 87, 2721–2727.
a
found that there was an increase in the total phenolic compounds of de-stoned fruit oils in the three olive cultivars under study. This was higher for Picual (34%) than for Manzanilla and Hojiblanca (around 18%). The higher phenolic content of oils from de-stoned fruits indicated an effective role for the seed in the catabolism of phenolic compounds during the crushing and/or malaxation processes. The authors concluded that the role of olive seed in determining the phenolics profile during the extraction process seems to be associated with the high levels of POD activity observed (Table 8.4). This hypothesis is supported by the observed decrease in phenolic content when the seed proportion is increased during the olive oil extraction process, and by the observed modification of the phenolics profile by seed POD enzymatic extracts (Table 8.5). Servili et al. (2007) studied olive stoning during the virgin olive oil mechanical extraction process to determine its effect on the phenolic and volatile composition of the oil. To elucidate the impact of the constitutive parts of olive fruit on the composition of pastes during processing, phenolic compounds and several enzymatic activities such as PPO, POD, and lipoxygenase (LPO) in the olive pulp, stone, and seed were also studied. The olive pulp had large amounts of oleuropein, demethyloleuropein, and lignans, but the contribution of the stone and the seed to overall phenolic composition in the fruit was very low. The presence of crushed stone in the pastes during malaxation increased the peroxidase activity (Table 8.6) of the pastes, reducing the phenolic concentration in virgin olive oil and, at the same time, modifying the composition of volatile compounds produced by the LOX. The oil obtained from stoned olive pastes contained higher amounts of secoiridoid derivatives (Table 8.7) such
lncubations were carried out with 2 U of olive seed POX and 1.072 µmol VOO phenolics. Abbreviations: VOO: Virgin olive oil; POX (POD): peroxidase; HTyr: hydroxytyrosol; DGO: dialdehydic forms of decarboxymethyl oleuropein; DGL: dialdehydic forms of ligstroside aglycones; AGO: aldehydic forms of oleuropein; AGL: aldehydic forms of ligstroside aglycones; Tyr: Tyrosol. Adapted with permission from: Luaces, P., Romero, C., Gutierrez, F., Sanz, C., G Perez A., 2007. Contribution of olive seed to the phenolic profile and related quality parameters of virgin olive oil. J. Sci. Food Agric. 87, 2721–2727.
TABLE 8.6 Enzymatic activities (U/mg d.w.) in crushed and malaxed pastes obtained from whole and stoned Frantoio and Coratina olive fruits. Crushed paste Traditional
Malaxed paste Stoned
Traditional
Stoned
Frantoio Cultivar PODa 37.75 (2.05) a 12.45 (1.06) b 14.05 (0.35) b PPO
6.40 (0.28) c
17.00 (0.28) a 17.35 (1.48) a 10.30 (0.70) b 10.70 (1.41) b Coratina Cultivar
POD
16.05 (0.80) a 5.35 (0.40) b
7.25 (0.21) c
2.60 (0.14) d
PPO
5.35 (0.49) a 5.95 (0.21) a
4.20 (0.42) b
5.20 (0.42) ab
PPO: polyphenol-oxidase; POD: peroxidase;: lipoxygenase. a The enzymatic activity is the mean value of three independent experiments; the standard deviation is reported in parentheses. Values in each row having different letters (a–d) are significantly different from one another at p ⬍ 0.01. Adapted with permission from: Servili, M., Taticchi, A., Esposto, S., Urbani, S., Selvaggini, R., Montedoro, G., 2007. Olive stoning and virgin olive oil quality. J. Sci. Food Agri. 55 (17), 7028–7035.
75
CHAPTER | 8 Influence of the Crushing System
TABLE 8.7 Phenolic composition (mg kg⫺1) in oils obtained by mechanical extraction with traditional process and with stoning. Traditional
●
3,4-DHPEAa
1.9 (0.3) a
1.8 (0.2) a
p-HPEA
4.6 (0.4) a
4.9 (0.4) a
3,4-DHPEA-EDA
88.8 (7.2) a
112.4 (10.1) b
p-HPEA-EDA
43.8 (3.1) a
54.8 (3.1 } b
3,4-DHPEA-EA
30.3 (2.7) a
42.8 (3.6) b
(⫹)-1acetoxypinoresinol
40.7 (2.3) a
42.8(1.1) a
●
●
●
4.2 (0.4) a
4.5 (0.5) a ●
Coratina Cultivar 3,4-DHPEA
●
Stoned
Frantoio Cultivar
(⫹)-pinoresinol
SUMMARY POINTS
3.1 (0.2) a
2.8 (0.3) a
14.9 (1.1) a
15.0 (0.8) a
365.2 (18.6) a
507.1 (19.9) b
●
92.8 (7.8) a
115.9 (6.2) b
●
101.2(5.8) a
125.9 (7.1) b
41.1(1.1) a
44.0 (1.8) a
8.4 (0.6) a
10.4 (0.6) b
●
p-HPEA 3,4-DHPEA-EDA p-HPEA-EDA 3,4-DHPEA-EA (⫹)-1acetoxypinoresinoI (⫹)-pinoresinol
●
●
a
The phenolic content is the mean value of three independent experiments; the standard deviation is reported in parentheses. Values in each row having different letters (a–b) are significantly different from one another at p ⬍ 0.01. 3,4-DHPEA: 3,4-Dihydroxyphenyl-ethanol; p-HPEA: p-hydroxyphenylethanol; 3,4-DHPEA-EDA: dialdehydic forms of elenolic acid linked to 3,4-DHPEA; p-HPEA-EDA: dialdehydic forms of elenolic acid linked to p-HPEA; 3,4-DHPEA-EA: isomer of oleuropein aglycon. Adapted with permission from: Servili, M., Taticchi, A., Esposto, S., Urbani, S., Selvaggini, R., Montedoro, G., 2007. Olive stoning and virgin olive oil quality. J. Sci. Food Agric. 55 (17), 7028–7035.
as the dialdehydic forms of elenolic acid linked to (3,4dihydroxyphenyl)- ethanol and (p-hdroxyphenyl)ethanol (3,4-DHPEA-EDA and p-HPEA-EDA, respectively) and the isomer of the oleuropein aglycon (3,4-DHPEA-EA) and, at the same time, did not show significant variations of lignans. The stoning process also modified the volatile profile of virgin olive oil by increasing the C6 unsaturated aldehydes that are strictly related to the cut-grass sensory notes of the oil (Amirante et al., 2005).
●
The use of different machines in olive oil production has inevitable repercussions on the cost-effectiveness of the oil-making process, on the amounts of oil extracted and especially on the quality of the oil obtained. When the paste is prepared by resorting to hammer crushers the oils produced contain greater amounts of polyphenols than when prepared with a stone mill. Hammer crushing produces a more intense fragmentation of the olive pits than the disk crusher, resulting in a substantial increase in output temperature. Higher temperatures in the crusher during olive processing lead to decreased preservation of the oils. The oils obtained from de-stoned pastes had a higher polyphenol content in comparison to the oils obtained from whole paste. Polyphenoloxidase (PPO) and peroxidase (POD) are endogenous oxidoreductases that can catalyze phenolic oxidation during processing. The control of these enzymes during the oil mechanical extraction process is a very important component strictly related to the phenolic concentration of virgin olive oil. The seed, in particular, showed the highest POD activity compared to other constitutive parts of fruit. The de-stoning of olives prior to malaxation reduces oxidative reactions catalyzed by POD. Oil extracted from de-stoned pastes has increased nutritional and sensory quality. The byproducts of this mechanical extraction process are highly marketable. De-stoned olive pomace, rich in antioxidants and high-value monounsaturated fatty acids, can then be employed in animal feeding. The large concentration of antioxidants in vegetation waters could be recovered and converted, using environmentally sustainable processes, into compounds of high biological value, such as fine chemicals and bioactive compounds. The overall value of the olive seed could be increased through the recovery of high-value ‘healthy’ molecules such as phospholipids, nuzenide, squalene and terpens, as a byproduct of the oil recovery process. Resulting seedcoats can be used as fuel.
REFERENCES Amirante, P., Arena, G., Clodoveo, M.L., Dugo, G., Leone, A., Lo Turco, V., Pollicino, D., Tamborrino, A., 2006. Virgin olive oil production from de-stoned pastes: a new technology to improve the shelf life of the product. Ital. J. Food. Sci. (Special Issue), 116–120. Amirante, P., Baccioni, L., Bellomo, F., Di Renzo, G.C., 1987. Installations pour l’extraction d’huile d’olive a’ partir de pastes d’olives de’noyante’es. Olivae 17. Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A., Tamborrino, A., 2005. Advance technology in virgin olive oil production from traditional and
76
de-stoned pastes: Influence of the introduction of a heat exchanger on oil quality. Food Chem. 98 (4), 797–805. Caponio, F., Catalano, P., 2001. Hammer crushers vs disk crushers: the influence of working temperature on the quality and preservation of virgin olive oil. Eur. J. Lipid Sci. Technol. 213, 219–224. Catalano, P., Caponio, F., 1996. Machines for olive paste preparation producing quality virgin olive oil. Fett-Lipid 98 (12), 408–412. Di Giovacchino, L., Sestili, S., Di Vincenzo, D., 2002. Influence of olive processing on virgin olive oil quality. Eur. J. Lipid Sci. Technol. 104, 587–601. Luaces, P., Romero, C., Gutierrez, F., Sanz, C., G Perez, A., 2007. Contribution of olive seed to the phenolic profile and related quality parameters of virgin olive oil. J. Sci. Food Agr. 87, 2721–2727. Servili, M., Buonaurio, R., 1999. Involvement of lipoxygenase, lipoxygenase pathway volatiles, and lipid peroxidation during the hypersensitive
SECTION | I Olives and Olive Oil
reaction of pepper leaves to Xanthomonas campestris pv. vesicatoria. Physiol. Mol. Plant P. 54, 155–169. Servili, M., Piacquadio, P., De Stefano, G., Taticchi, A., Sciancalepore, V., 2002a. Influence of a new crushing technique on the composition of the volatile compounds and related sensory quality of virgin olive oil. Eur. J. Lipid Sci. Technol. 104, 483–489. Servili, M., Montedoro, F., 2002b. Contribution of phenolic compounds to virgin olive oil quality. Eur. J. Lipid Sci. Technol. 104, 602–613. Servili, M., Selvaggini, R., Esposto, S., Taticchi, A., Montedoro, G., Morozzi, G., 2004. Health and sensory properties of virgin olive oil hydrophilic phenols: agronomic and technological aspects of production that affect their occurrence in the oil. J. Chromatogr. A. 1054, 113–127. Servili, M., Taticchi, A., Esposto, S., Urbani, S., Selvaggini, R., Montedoro, G., 2007. Olive stoning and virgin olive oil quality. J. Sci. Food Agr. 55 (17), 7028–7035.
Chapter 9
The Malaxation Process: Influence on Olive Oil Quality and the Effect of the Control of Oxygen Concentration in Virgin Olive Oil Antonia Tamborrino1, Maria Lisa Clodoveo1, Alessandro Leone2, Paolo Amirante1 and Alistair G. Paice3 1
Department of Engineering and Management of the Agricultural, Livestock and Forest Systems (PROGESA), University of Bari, Italy Department of Production and Innovation in Mediterranean Agriculture and Food Systems (PriMe), University of Foggia, Italy 3 Department of Clinical Biochemistry, Nutrition and Dietetics, King’s College London, UK
2
disk crusher). In order to preserve the kneading operation that occurred using rotating wheels, a new machine was introduced: the malaxer machine. Thus, a new phase in the procedure was introduced: the malaxation phase. Malaxing is an extremely important phase in olive oil extraction. During the malaxing phase the olive paste is subjected to a slow continuous kneading, aimed at breaking off the emulsions formed during the crushing process and facilitating adequate coalescence. It is necessary to heat the olive paste at a carefully monitored temperature during malaxation in order to diminish the viscosity of the product and to stimulate its enzymic activity, therefore increasing the extraction yields. This operation facilitates high extraction yields, by helping small oil droplets to coalesce. These can be separated subsequently using a decanter centrifuge (Di Giovacchino, 2000). The malaxing process determines the balance between the quality and the quantity of the oil extracted, by varying a range of parameters (time, temperature and atmosphere in contact with the olive paste), as the olive paste is gradually heated and the enzymes within are activated. All this must be done without affecting the biochemical structure of the olive paste, as this would affect the flavor, shelflife and nutritional properties of the oil (Amirante et al., 2006). This operation is one of the critical points in olive oil extraction. Many studies have been carried out to investigate its influence on the olive oil quality. Nowadays the olive oil consumer asks for healthy products. There has been a large increase in demand for high-quality virgin olive oil, attributed not only to its potential health benefits,
9.1 INTRODUCTION Olive oil was discovered by early man, as he accidentally crushed fallen olive seeds and noticed that the segregated oil moistened. Since ca. 5000 BC, people have collected and squeezed olives in stone mortars, but the Romans expedited the crushing operation with a millstone crusher, the trapetum, and improved the separation system with the introduction of presses (Kiourellis, 2005). Following the fall of the Roman Empire there were no innovations in olive oil processing for many years, which continued to be based on the screw press (Di Giovacchino, 2000). In 1795 the hydraulic pressing system was invented (Balatsouras, 1986). Since the second half of the 20th century many technological improvements and innovations have occurred (Kapellakis et al., 2008). Olive oil separators have replaced the traditional methods, and productivity has been increased with the widespread adoption of hydraulic pressing systems. The pressing process has been the most widespread method for processing olive fruit to obtain olive oil. The use of stone rollers or wheels on a slab of stone allows the grinding of olives and kneading the resulting olive paste. The olive paste was drenched with hot water, in order to achieve better separation of olive oil, and then placed in oil diaphragms. In 1960 a new process for separation of oil from olive paste was introduced, based on the density differences of the olive paste constituents (olive oil, water and insoluble solids). Separation is accomplished through a horizontal centrifuge. At the same time, the rotating wheels were replaced with metallic crushers (hammer crusher and Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
77
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
78
but also to its particular organoleptic properties. In fact, the sensory quality plays an important role in customer preferences (Angerosa et al., 1998). The operating environment during malaxation affects the volatile and phenolic composition of virgin olive oil and, as a consequence, its sensory and healthy qualities. The sensory and health-related properties of virgin olive oil are closely linked to its volatile and phenolic composition (Reiners and Grosch, 1998; Servili et al., 2002). The presence of these substances in the oil is the end result of various endogenous enzymatic activities activated during processing of the olive fruit (Morales et al., 1999; Angerosa et al., 2001). Research on olive oil quality has also shown the effect of the oxygen concentration in the head space of malaxer on the activities of these enzymes and hence on the olive oil quality itself. Processing conditions strongly influence rheological properties of the olive paste. These in turn determine the behavior of the paste during the sedimentation and can thereby influence the extraction yield (Di Renzo and Colelli, 1997). Extension of the malaxation time improves the yield but can damage the final quality of the product. In order to correct for this potential disadvantage modifications have to be made to the malaxing system (Amirante et al., 2002). Much research has targeted the efficiency of this innovative continuous system, with many changes to the malaxer machine, aimed at improving the eventual olive oil quality (Amirante et al., 2005, 2006, 2008a, 2008b).
9.2 ENZYMES INVOLVED DURING MALAXATION AND THEIR INFLUENCE ON PHENOLIC CONTENT AND VOLATILE COMPOUNDS Fresh and good-quality virgin olive oil is appreciated by consumers for its delicious taste and aroma. These qualities are partially due to non-volatile compounds, i.e. complex phenols (bitter taste) (Angerosa et al., 1998), and a number of volatile chemical compounds generated during the crushing-malaxation steps of oil production (Morales et al., 1994). Phenolic compounds originate from endocellular oil; malaxation modifies their qualitative and quantitative composition (Montedoro et al., 1992; Servili et al., 1992). The distribution of phenolic compounds in oil and its byproducts during mechanical extraction is strictly linked to endogenous enzyme activities (Servili et al., 2003). During extraction, the content of some components is significantly modified, depending on the extraction technique employed (Amirante et al., 2005). New components such as volatile compounds can be formed as a result of chemical and/or enzymatic pathways (Ranalli et al., 2001). The relationship between the volatile and the phenolic components in virgin olive oil is based on the activity of the polyphenoloxidase (PPO), peroxidase (POD) and lipoxygenase (LOX) pathways. Polyphenoloxidase (PPO) and peroxidase (POD) can
SECTION | I Olives and Olive Oil
oxidize phenolic compounds (i.e., secoiridoids), resulting in reduced phenolic concentration of the oil. Potential effects on product quality characteristics include a decrease in the ‘bitter’ and ‘pungent’ sensory attributes, oxidative stability and a reduction in the nutritional value of virgin olive oil (Georgalaki et al., 1998; Ranalli et al., 2001). Volatile compounds, which help make up the aroma of many fruits and vegetables, are produced from polyunsaturated fatty acids via a cascade of enzymatic reactions known as the lipoxygenase pathway. Lipoxygenase is a non-heme iron-containing enzyme that catalyzes the oxygenation of the 1,4-pentadiene sequence of polyunsaturated fatty acids to produce their corresponding hydroperoxides (Hatanaka, 1993). The lipoxygenase pathway involves the oxidative degradation of the polyunsaturated linoleic and a-linolenic acids, which are split into volatile C6 or C9 carbonyl fragments. These fragments can be further modified by isomerization, reduction and esterification. The variety of volatile compounds thus produced constitutes the volatile fraction of the oil, so this pathway is important in helping determine the quality of the oil produced (Angerosa et al., 2001). Many of the volatiles produced are incorporated into the oil phase, adding to its characteristic aroma. The mentioned enzymes are triggered by the milling of olives, and are active during the malaxation step. Both classes of enzymes have oxygen as a co-factor. Hence, the presence of oxygen in the head-space of the mixer might favor oxidation of phenols during malaxation, reducing their concentration in the pastes and in the oil. On the other hand, oxygen may facilitate the LPO pathway activity, determining the sensory note intensity in virgin olive oil. The latest research indicates that the choice of crushing system can limit the presence of enzymes such as PPO and POD. Strategic use of oxygen concentration during the malaxing phase helps to obtain high phenol concentrations. This in turn leads to well-balanced sensory characteristics in the resulting olive oil (Amirante et al., 2008a).
9.3 MALAXER MACHINE AND ITS EVOLUTION IN ENHANCING OLIVE OIL QUALITY The malaxer machine consists of a stainless steel tank containing the olive paste, a malaxing central screw stirring the paste slowly and continuously and a mono screw pump conveying the paste to the separation phase utilizing a decanter machine. The malaxer is equipped with a circulating hot water jacket to heat the paste. Every tank is also equipped with a temperature switch. For many years the malaxer machine was characterized by a non-hermetic closure, i.e. an inox grill. The traditional tank has typically been a cradle shape. There was considerable loss of the phenolic and volatile compound content of the oil using this type of machine due to the lack of an hermetic seal
79
CHAPTER | 9 The Malaxation Process
(Amirante et al., 2006). In fact the inox grill allowed the loss of volatile compounds into the air above the tank. At the same time, stirring the olive paste whilst in contact with the air caused oxidation of the phenolic content and thus their loss to the oil phase (Amirante et al., 2006). The malaxing phase has been the subject of much research, which has stimulated the search for new technologies (Angerosa et al., 2001; Di Giovacchino et al., 2002; Amirante et al., 2005). In recent years improvements in malaxer machine technology have included new models with inert gas processing (nitrogen or argon) and oxygen concentration control using a hermetic cover cap. From an analytical point of view, processing in an atmosphere of inert gas has lessened the amount of peroxide, drastically reducing oxidation phenomena during the malaxing phase of the extraction process. Continuous nitrogen flow during malaxing reduces the oxidation of phenolic compounds, increasing oxidative stability of the product (Figure 9.1) (Amirante et al., 2006). Studies have shown that extending the malaxation time improves yields but can damage the final quality of the product, causing dispersion of phenols into waste water and increasing degradative and oxidative phenomena (Amirante et al., 2002; Di Giovacchino et al., 2002). A few years ago, a heat exchanger was introduced between the crushing machine and the malaxing machine, bringing the olive paste instantaneously to the malaxation temperature and therefore reducing the malaxation time needed to warm the olive paste. This improved the phenol and volatile compound content of the product (Amirante
22 Induction time (at 120°C)
20
R2 = 0.8962
18 16 14 12 10 8 6 200
Mixing with nitrogen Mixing without nitrogen
250
300 350 Total phenols (mg kg−1)
400
450
FIGURE 9.1 Correlation between total phenol content and induction time. The data show mean values (m) and standard deviations (sd). Phenolic compounds obtained from the oils extracted with a malaxer utilizing a continuous flow of nitrogen. Dates were correlated with the oxidative stability. Oils with a higher phenol content show reduced oxidation. Each point on the graph represents the mean of five values. Adapted with permission from: Amirante et al. (2006). ‘New mixer equipped with control atmosphere system: influence of malaxation on the shelf-life of extra virgin olive oil’, in Italian Journal of Food Science. Special Issue, 215–220.
et al., 2005). Recently, an innovative mixer for the malaxation process has been proposed. This new machine has been developed to improve the kneading and heating process of the olive paste, increasing the heat transfer surface in order to reduce the mixing time. The mixer was designed to guarantee perfect control of the atmosphere in contact with the paste. In order to improve the heating surface per volume with respect to standard cradle scutching machines, this innovative mixer has been designed with a circular and spiral-shaped interspace that covers the whole internal longitudinal surface of the tank. This means that paste can be conveyed and maintained at the desired temperature more quickly and effectively (Figure 9.2). The new malaxer is equipped with a more efficient reel designed to provide bidirectional thrust to paste, causing it to continuously rotate and bring new sections of paste into contact with the heating walls. Initial studies carried out on this machine have highlighted its importance when used in continuous processing regarding the health and organoleptic properties of the resulting olive oil. The new mixer allows better control of processing parameters such as temperature, time, and the atmosphere in contact with the olive paste, facilitating a combination that enhances the quality of the olive oil.
9.4 MIXING CONDITIONS AND THEIR EFFECT ON THE OLIVE OIL QUALITY Malaxation temperature and time are processing-related parameters that can be controlled to change the sensory properties of the oil and increase its phenol content. Increasing the time and temperature of the malaxing process decreases the concentration of secoiridoid aglycons and phenolic alcohols in olive pastes. These changes are attributed to the oxidative reactions catalyzed by endogenous oxidoreductases such as polyphenoloxidase and peroxidise as well as of the distribution of hydrophilic phenols between the oil and the water phase (Table 9.1) (Angerosa et al., 2001). Malaxation conditions play a crucial role in the formation and degradation of phenolic compounds and also influence the volatile compound levels. Malaxing olive paste at 30°C for at least 45 min (Ranalli et al., 2003) produces both pleasant ‘green’ extra virgin olive oil and acceptable oil extraction yields, but malaxing at 35°C introduces several defects into the oil without increasing oil yield (Ranalli et al., 2001). Volatile compounds were the most common discriminating variables for malaxation time and temperature. The increase of alcohols and of C6, C5 and carbonyl compounds, especially of hexanal, contribute significantly to the olive oil flavor resulting from the longer malaxation time, whereas higher malaxation temperatures lead to a reduction of C6 esters and C5 metabolites and an accumulation of hexan-1-ol and trans-2-hexen-1-ol,
80
SECTION | I Olives and Olive Oil
FIGURE 9.2 An innovative malaxer in a industrial plant. This new machine has been used in experimental tests carried out during the 2007/08 season in order to study its effect on the healthy and organoleptic properties of olive oil.
TABLE 9.1 Total secoiridoid aglycons in the oils extracted at different malaxation times and temperatures (as ppm resorcin). Changes in secoiridoid compounds with respect to time and temperature changes in paste malaxation. Oils were extracted from the fruits of two Italian cvs (Coratina and Frantoio). Several possible beneficial properties of secoiridoids have been identified. These include enhancing the shelf-life of olive oil and a possible protective effect against thrombotic risk and atherosclerotic damage. Cultivar
15⬘
30⬘
45⬘
25°C
35°C
25°C
35°C
25°C
Coratina
651
357
511
357
470
Frantoio
179
105
93
87
78
60⬘ 35°C
71
90⬘
25°C
35°C
25°C
35°C
426
263
363
219
61
30
31
27
Adapted with permission from: Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2001. Influence of malaxation temperature and time on the quality of virgin olive oils. Food Chemistry. 72 (1), 19–28.
both considered by some authors to elicit unpleasant odors (Morales et al., 1999; Angerosa et al., 2001). In addition, high temperatures in the malaxation step activate the amino acid conversion pathway with production of considerable amounts of 2-methyl-butanal and 3-methyl-butanal, but without accumulation of corresponding alcohols correlated with a ‘fusty’ defect (Angerosa et al., 1998). A malaxing time of between 30 and 45 minutes at the lower temperature is recommended to obtain a good olive oil quality with acceptable yields. In recent years several studies have investigated how the head-space composition changed during the course of malaxation, investigating the change in phenol and volatile compound content. Selective control of some endogenous oxidoreductases during the malaxing phase seems to be key to obtaining well-balanced olive oil. In this ambit previous studies tried using inert gas to remove oxygen in the
headspace of the malaxer (Amirante et al., 2006, 2008a). They showed that using N2 during malaxation reduced the oxidative degradation of phenolic antioxidants with little alteration to the volatile composition of oil. At the same time the O2 level during the malaxation phase was monitored using an innovative malaxer with complete control of the mixing conditions such as time and temperature. Initially there was a rapid decrease in concentration of O2 followed by a gradual reduction in its consumption rate. The O2 concentration fell to below 5% after roughly 20 minutes, then gradually fell to 0. The overall effect is illustrated by the trend line in Figure 9.3. Initially the O2 concentration decreased by roughly 2 ml L⫺1 min⫺1, probably caused by the oxidative activity of enzymes such as lipoxygenase, polyphenoloxidase and peroxidase. Some of this oxygen might be being consumed by cellular respiration resulting from the breakdown of
81
CHAPTER | 9 The Malaxation Process
Oxygen (ml I−1)
25 20
y = 20.94e−0.13x R2 = 0.963
15 10 5 0
0
5
10
15 20 Minutes
25
30
35
0
5
10
15 20 Minutes
25
30
35
A
Carbon dioxide (MI I-1)
35 30 25 20 15 10 5 0 B
FIGURE 9.3 (A) Oxygen depletion during olive paste malaxation in the innovative malaxer. (B) Carbon dioxide emission during olive paste malaxation in the innovative malaxer. O2 consumption and CO2 emission measured during a set of trials. Each point represents the mean of five measurements. The O2 represents concentration changes (ml L⫺1) in the head-space of the malaxing machines. In these trials, an initial rapid decrease in concentration of O2 was observed followed by a gradual decrease in consumption rate. The Innovative malaxer was used. Adapted with permission from Amirante et al. (2008a). ‘Influence of three different atmosphere composition of head space of mixer on total phenol content of de-stoned virgin olive oil’, Proceedings book of the International Symposium on Food and Bioprocess Technology. Brazil, 31 August to 4 September 2008. ISSN 1982-3797. CD printed.
cellular structures by crushing and successive malaxation. The CO2 levels emitted by paste malaxation are demonstrated in Figure 9.2 as changes in the head-space of the sealed chamber. The natural increase of CO2 concentration coupled with the decrease of O2 in the head-space of the hermetic malaxer might feasibly improve the quality of resulting oil. When air was used in the head-space of the mixer, the oils extracted contained much higher amounts of pleasant volatile compounds than the oils obtained using the nitrogen (Table 9.2). In fact, oxidative phenomena that take place during transformation result in the loss of compounds such as polyphenols, tocopherols and chlorophylls, which are important in both human health and conservation of the olive oil itself. In this experiment oxygen and carbon dioxide were measured in a full-scale plant. In a hermetic industrial mixer, oxygen was absent after 20 minutes of malaxation in the head-space. The atmosphere was saturated by carbon dioxide, facilitating a reduction in oxidative phenomena without using inert gases. This considerably reduces the cost of the extraction process. In order to obtain oil rich in the pleasant volatile compounds responsible for its fruity flavor, partial oxidation of fatty acid chains is necessary. These reactions largely take place in the initial stage of malaxation, where the lipoxygenase pathway generates volatile compounds that help make up the oil’s aroma. The presence of a small concentration of oxygen is important, especially in the initial part of malaxation, to develop the typical aromas appreciated in the best extra virgin olive oil. Amongst the most important olive oil volatiles that arise from the lipoxygenase pathway include C5 and C6 compounds. The sensory qualities of unsaturated C6 aldehydes and alcohols are related to the so-called ‘green odor’ of this product. The presence of these compounds
TABLE 9.2 The amount of C6 and C5 compounds, derived from the lipoxygenase pathway, found in virgin olive oil generated by the two different paste malaxation conditions being studied. The effect of the two different paste malaxation conditions on volatile compounds. When air was used in the head-space of the mixer, the oils extracted contained much higher amounts of ‘pleasant ’ volatile compounds than when nitrogen was used. Volatile compounds (mg kg⫺1)
100% air
100% nitrogen
m
SD
m
SD
Hexanal
11.6
0.5
3.1
0.1
trans-2-Hexenal
147.3
5.8
43.5
2.1
1-Hexanol
45.2
3.1
3.9
0.7
trans-2-Hexen-1-ol
11.9
0.3
4.6
0.9
cis-3-Hexen-1-ol
7.7
0.4
1.4
0.1
cis-2-Penten-1-ol
3.6
0.1
nd
Nd
Data represent mean (m) and standard deviation (SD) values. Significant differences in the same row are statistically significant (p ⬍ 0.01). Adapted with permission from: Amirante, P., Clodoveo, M. L., Leone, A., Tamborrino, A. 2008a. ‘Influence of three different atmosphere composition of head space of mixer on total phenol content of de-stoned virgin olive oil’. Proceedings book of the International Symposium on Food and Bioprocess Technology. Brazil, 31 August to 4 September 2008. ISSN 1982-3797. CD printed.
82
SECTION | I Olives and Olive Oil
in good-quality virgin olive oils is critical to provide the typical fresh ‘green notes’ that they are known for. The natural increase of CO2 concentration, coupled to the decrease of O2, in the head-space of the innovative hermetic malaxer generates an inert malaxer head-space, potentiating its beneficial effect.
SUMMARY POINTS ●
●
●
●
●
●
●
●
●
Malaxing is an extremely important phase in olive oil extraction in which the olive paste is subjected to a slow, continuous kneading to disperse the emulsions formed during the crushing process and to facilitate adequate coalescence. In recent years this operation has become one of the most critical points in the mechanical extraction process for olive oil. Many studies have been carried out to investigate this crucial phase and its influence on olive oil quality. Mixing conditions such as time, temperature and the composition of the atmosphere in contact with the olive paste can influence the activity of the enzymes that are responsible for the healthy and organoleptic properties of the product. A low malaxing temperature and a process time between 30 and 45 minutes are recommended to obtain good olive oil quality without compromising the yield. Much research has been carried out to optimize the malaxing phase. This in turn has stimulated the development of new technologies. In recent years new models of the malaxer machine have evolved with improvements that include the use of inert gases, optimized kneading and heating processes and optimized control of the atmosphere that is in contact with the paste. Using N2 during malaxation reduces the oxidative degradation of phenolic antioxidants with little modification to the volatile composition of olive oil. The natural increase of CO2 concentration coupled with the decrease of O2 concentration in the head-space of the innovative hermetic malaxer generates an inert head-space of the malaxer potentiating its benefit effect. The strategic manipulation of oxygen concentration during the malaxing phase facilitates the generation of high phenol contents. This leads to well-balanced sensory characteristics in the resulting olive oil.
REFERENCES Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A., Tamborrino, A., 2005. Advance technology in virgin olive oil production from traditional and de-stoned pastes: influence of the introduction of a heat exchanger on oil quality. Food Chem. 98 (4), 797–805.
Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A., Salvo, F., Tamborrino, A., 2006. New mixer equipped with control atmosphere system: influence of malaxation on the shelf life of extra virgin olive oil. Ital. J. Food Sci. 215–220 (special issue). Amirante, P., Clodoveo, M.L., Leone, A., Tamborrino, A., 2008a. Influence of three different atmosphere composition of head space of mixer on total phenol content of de-stoned virgin olive oil. Proceedings book of the International Symposium on Food and Bioprocess Technology. Brazil, 31 August to 4 September 2008. ISSN 1982-3797. CD printed. Amirante, P., Clodoveo, M.L., Leone, A., Tamborrino, A., 2008b. Assessment of the viscosity value in olive oil paste using different blade rotation speed in an innovative mixer. Proceedings book of the International Symposium on Food and Bioprocess Technology. Brazil, 31 August to 4 September 2008. ISSN 1982-3797. CD printed. Amirante, P., Dugo, G., Gomez, T., 2002. Influence of technological innovation in improving the quality of extra virgin olive oil. Olivae 93, 34–42. Angerosa, F., D’Alessandro, N., Basti, C., Vito, R., 1998. Biogeneration of volatile compounds in virgin olive oil: their evolution in relation to malaxation time. J. Agric. Food Chem. 46, 2940–2944. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2001. Influence of malaxation temperature and time on the quality of virgin olive oils. Food Chem. 72 (1), 19–28. Balatsouras, G., 1986. Elaiolado, Sporelaia, Lipi. (Olive Oil, Seed Oil and Fats.) Karaberopoulos, Athens, Greece (in Greek). Di Giovacchino, L., 2000. Technological aspects. In: Hardwood, J., Aparicio, R. (eds), Handbook of Olive Oil. Analysis and Properties. Aspen Publications, pp. 17–59. Di Giovacchino, L., Costantini, N., Ferrante, M.L., Serraiocco, A., 2002. Influence of malaxation time of olive paste on oil extraction yields and chemicals and organoleptic characteristics of virgin olive oil obtained by a centrifugal decanter at water saving. Grasas Aceites 53, 179–186. Di Renzo, G.C., Colelli, G., 1997. Flow behavior of olive paste. Appl. Eng. Agric. 13 (6), 751–755. Georgalaki, M.D., Bachmann, A., Sotiroudis, T.G., Xenakis, A., Porzel, A., Feussner, I., 1998. Characterization of a 13-lipoxygenase from virgin olive oil and oil bodies of olive endosperms. Fett-Lipid 100 (12), 554–560. Hatanaka, A., 1993. The biogeneration of green odour by green leaves. Phytochemistry 34, 1201–1218. Kapellakis, I.E., Tsagarakis, K.P., Crowther, J.C., 2008. Olive oil history, production and by-product management. Rev. Environ. Sci. Biotechnol. 7, 1–26. Kiourellis, A., 2005. I Tehnologia Paragogis Elaioladou sti Lesvo kata tin Arhaiotita. (The Technology of Olive Oil Production in the Island of Lesvos in Antiquity.) Prefecture of Lesvos, p. 148. Morales, M.T., Angerosa, F., Aparicio, R., 1999. Effect of the extraction conditions of virgin olive oil on the lipoxygenase cascade: chemical and sensory implications. Grasas Aceites 50, 114–121. Morales, M.T., Aparicio, R., Rios, J.J., 1994. Dynamic headspace gas chromatographic method for determining volatiles in virgin olive oil. J. Chromatogr. A. 668, 455–462. Ranalli, A., Contento, S., Schiavone, C., Simone, N., 2001. Malaxing temperature affects volatile and phenol composition as well as other analytical features of virgin olive oil. Eur. J. Lipid Sci. Tech. 103, 228–238. Ranalli, A., Pollastri, L., Contento, S., Iannucci, E., Lucera, L., 2003. Effect of olive paste kneading process time on the overall quality of virgin olive oil. Eur. J. Lipid Sci. Tech. 105, 57–67.
CHAPTER | 9 The Malaxation Process
Reiners, J., Grosch, W., 1998. Odorants of virgin olive oils with different flavor profiles. J. Agric. Food. Chem. 46, 2754–2763. Servili, M., Montedoro, G.F., 2002. Contribution of phenolic compounds to virgin olive oil quality. Eur. J. Lipid Sci. Tech. 104, 602–613. Servili, M., Baldioli, M., Montedoro, G.F., 1992. I meccanismi che influenzano la concentrazione in polifenoli dell’olio vergine di oliva. Proceedings of the international congress on olive oil quality. 375 ⫾ 376. Firenze (Italy) 1–3 December.
83
Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G., 2003. Volatile compounds and phenolic composition of virgin olive oil: optimization of temperature and time of exposure of olive pastes to air contact during the mechanical extraction process. J. Agric. Food Chem. 51 (27), 7980–7988.
This page intentionally left blank
Chapter 10
Influence of Different Centrifugal Extraction Systems on Antioxidant Content and Stability of Virgin Olive Oil Paolo Amirante1, Maria Lisa Clodoveo1, Alessandro Leone2, Antonia Tamborrino1 and Vinood B. Patel3 1
Department of Engineering and Management of the Agricultural, Livestock and Forest Systems (PROGESA), University of Bari, Italy Department of Production and Innovation in Mediterranean Agriculture and Food Systems (PriMe), University of Foggia, Italy 3 Department of Biomedical Sciences, University of Westminster, London, UK
2
The final manufacturing innovation regarding the decanter centrifuges is called ‘third generation’. They have an innovative design, which ensures greater strength and reliability due to a longer cylindrical part of the bowl and a shorter beach section, and a special bowl with a variable dynamic pressure (VDP) cone system. The VDP cone decanters allow for the adjustment of the process parameters for optimum extraction yield. This innovation makes the machine extremely flexible in order to adapt to the heterogenic rheological characteristics of the raw material. The different centrifugal decanters employed in olive processing influence oil yields, qualitative characteristics such as total phenols and induction time values, and composition of volatile compounds such as aldehydes, alcohols, esters, hydrocarbons, ketones, furans, and other compounds that are responsible for the unique and delicate flavor of olive oil (Di Giovacchino et al., 2001). Total phenols as well as induction time values are higher in oils obtained by the centrifugal decanter of two-phases. The induction time is the length of time before the rate of lipid oxidation of an oil sample rapidly accelerates (Pike, 1998). The induction time of olive oil samples, as measured by the Rancimat instrument, showed a significant correlation to the concentration of total phenolic compounds (Baldioli et al., 1996). Dual-phase decanters work on the same principle as the three-phase decanters, except little or no water is added prior to centrifugation. This allows the retention of more polyphenols and volatiles (Di Giovacchino et al., 2001). The ‘third-generation’ three-phase decanters allow for the improvement of oil yields without compromising the quality of the product. This is the best solution with respect to the two-phase decanters because it is possible to extract the oil without adding water to the process, thus
10.1 INTRODUCTION In the 1970s and 1980s, olive processing using the continuous centrifugation system, called a three-phase system, expanded to many countries of the Mediterranean area. The success of this system is due to the high working capacity and automation of the industrial plants leading to a reduction of manual labor and olive-processing costs (Kapellakis et al., 2008). This system is called three-phase because the centrifugal decanter allows for the separation of three flows of matter; the olive oil, pomace (solid remains of olive) and vegetable waste water. However, this process requires lukewarm water to be added to dilute the olive paste. This causes the reduction of natural antioxidants in the oil and a considerable volume of vegetable waste water (80–100 L/100 kg of olives). At the beginning of the 1990s, olive oil plant manufacturers launched new models of decanters in the market. These were able to separate the oily phase from the olive paste without requiring the addition of lukewarm water and without producing vegetable waste water. These decanters have two exits producing oil and pomace only, and for this reason are called ‘two-phase decanters’. They produce a very wet pomace, with water content between 65 and 70% by weight. However, the centrifugal three-phase decanters were improved to be able to separate oil employing only a small quantity of warm water (0–20 L/100 kg of olives) to dilute the olive paste. These decanters have the ability to separate a small volume of vegetable waste water (5–25 L/100 kg of olives) and a less wet pomace, with a water content between 55 and 60%. These decanters are called ‘ARA’ (Italian acronym meaning ‘water saving decanter’). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
85
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
86
SECTION | I Olives and Olive Oil
obtaining dry pomace, which is more easily transportable and workable.
10.2 VIRGIN OLIVE OIL EXTRACTION SYSTEMS Olive oil is present in the form of small drops in the vacuoles of mesocarpal cells in the olive fruit. It is also scattered to a lesser extent in the colloidal system of the cell’s cytoplasm and to a lesser degree, in the epicarp and endosperm (Balatsouras, 1999). The main processing steps needed to obtain olive oil include: feeding, leaf removal and washing, crushing, mixing, separating the olive oil and centrifuging the oil. There are three different systems to distinguish the three different phases (olive oil, pomace and waste water): (a) pressure, (b) percolation process and (c) centrifugation. A typical layout of the extraction process, based on centrifugation, together with the main processes is given in Figure 10.1 (Kapellakis et al., 2008). Pressure process is the oldest method for processing olive fruit to obtain olive oil. This system was widely used until the 1980s. The olive paste of 2–3 cm thickness obtained from the previous stages is placed uniformly in synthetic fiber draining diaphragms which operate as filters. The diaphragms are then placed in moving units (trolleys) with a central shaft. A metal tray and a diaphragm without paste are placed after every 3–4 diaphragms to obtain uniform application and a more stable load. Then the moving unit along with its load is placed under a hydraulic pressure unit. When applying pressure, the liquid phases (oil and water) run through the olive cake. The advantages of this method include the use of simple, reliable machinery and little initial investment; the low energy requirement; a resulting pomace that is low in water content. The disadvantages of pressure process include a high labor cost and the production is not continuous. In the discontinuous process the raw materials are introduced to the production
Process Addition Disposal
Leaf removal Olives
Washing Water
Leaves
Waste water
line in the first instance. Sequentially the finished products and byproducts are released. However, the production line must be arrested, often cleaned, to allow the upload of new raw materials before commencing again. A major problem with this form of processing, which is still used in some countries, is maintaining the cleanliness of the diaphragms. The diaphragms, which act like sponges, retain oil which can become rancid or even ‘ferments’. Olives that are subsequently processed are very likely to become contaminated with the undesirable taste retained by the diaphragms. Finally, the pressure process which exposes the olive paste to air for a long time increases the degree of oxidation of the olive oil, leading to a product with reduced quality. The percolation process combined with centrifugation is used for the separation of olive oil from the olive paste. It is based on the different interfacial tension of oil and water coming in contact with a steel plate. When the steel plate is plunged into olive paste, it will be coated with oil due to the lower interfacial tension of the oil than that of the water. Five to seven thousand steel sheets are dipped into the paste; the oil preferentially wets and sticks to the metal sheets and is removed with scrapers in a continuous process. The different physical behaviors allow the olive oil to adhere to a steel plaque while the other two phases stay behind. This process is not completely efficient, leaving a large quantity of oil still in the paste, so the remaining paste has to be processed by centrifugal decanter, as discussed below. Centrifugation is a relatively new process for the separation of oil from olive paste. This technology was introduced at the end of the 1980s and currently is the most applied extraction process. It is based on the differences in density of the olive paste constituents (olive oil, water and insoluble solids). Separation is accomplished through a horizontal centrifuge (decanter). Decanters consist of a cylindrical conical bowl. Inside the bowl there is a hollow, similarly shaped component with helical blades. A slight difference between the speed at which the bowl rotates and
Centrifugation
Crushing Malaxation
Separation
Water Waste water and Waste water olive pomace
FIGURE 10.1 An olive mill based on the centrifugation process. The various steps involved in the production of olive oil. Taken with permission from Kapellakis et al. (2008).
87
CHAPTER | 10 Influence of Different Centrifugal Extraction Systems
that at which the inner screw gyrates results in the movement of the pomace to one end of the centrifuge, while the two other constituents of the olive paste (oil and olive mill wastewater (OMW) are pushed to the other end. In this case, when this capacity is exceeded, decanter performance is not efficient. A factor that affects the oil yield is the amount of water added to the paste. The optimum amount of water needed to dilute the paste is determined by the characteristics of the olive paste and the centrifuge. The main advantages of centrifugal processing systems are (Kiritsakis, 1998): ●
●
●
●
Limited labor is needed, since the process is continuous and automated. Stainless steel materials are always used and thus the oil is well protected from contamination. Since no diaphragms are used, the risk of defects in the oil is eliminated. Improved yield performance, as the majority of oil is collected.
The main disadvantages of centrifugal processing systems are (Kiritsakis, 1998): ●
●
Water and energy demanding: a significant amount of phenols (natural antioxidants) are lost during the centrifuge process in the OMW. Increased production of OMW, which is approximately 50% more than the pressure process.
10.3 DECANTER CENTRIFUGE TECHNOLOGY Decanters consist of a cylindrical conical bowl. Inside there is a hollow, similarly shaped component with helical blades. A slight difference between the speed at which the bowl rotates and that at which the inner screw gyrates
results in the movement of the pomace to one end of the centrifuge, while the two other constituents of the olive paste (oil and OMW) are pushed to the other end. Six crucial factors determine the performance of decanter centrifuges: 1. the centrifugal force required for sedimentation of the solids; 2. the clarification area necessary to ‘capture’ the solids; 3. the differential speed required to transport the solids out of the decanter; 4. the hydrodynamic design, which determines the exact parameters for the turbulence; 5. the height of the escape levels of the liquid phases; 6. the design of the conveyor and beach sections, which are important for efficient solid transportation.
10.4 DIFFERENCES BETWEEN TWO- AND THREE-PHASE CENTRIFUGAL DECANTERS The first main step in olive fruit processing is the crushing. The purpose of crushing is to tear the flesh cells to facilitate the release of the oil from the vacuoles. After the olive fruit has been crushed, the resulting paste is malaxed. Malaxation entails stirring the olive mash slowly and constantly for about 30 min. This operation aids in the coalescence of small oil drops into larger ones, thereby facilitating separation of the oil and water phases. After the malaxation phase the paste could be pumped to a two-phase centrifugal decanter or a three-phase centrifugal decanter for the separation. In the three-phase centrifugal decanter, paste is divided into oil, vegetation water and solids (olive pomace), i.e. kernel and pulp fragments (Figure 10.2). During the path to the three-phase centrifugal decanter, water is added to dilute the incoming paste. In the twophase process, paste instead is separated in oil as a liquid
Feed
Feed tube
DD Gearbox Wall of the bowl
Inlet distributor
Solids
Light liquid phase Heavy liquid phase
Conical end Screw conveyor
FIGURE 10.2 Three-phase centrifugal decanter. The vertical section of a three-phase centrifugal decanter is illustrated. The heavy liquid phase represents the water. The light liquid phase represents the olive oil. The black phase represents the olive pomace.
88
SECTION | I Olives and Olive Oil
Feed
Feed tube
DD Gearbox Wall of the bowl
Inlet distributor
Solids and heavy liquid phase
Light liquid phase
Conical end Screw conveyor
FIGURE 10.3 Two-phase centrifugal decanter. The vertical section of a two-phase centrifugal decanter is shown. The light liquid phase represents the olive oil. The black phase represents the olive pomace that is very humid.
phase and a solid phase, composed of fragments and kernels, pulp and vegetation water (humid olive pomace) (Figure 10.3). The two-phase process requires no dilution or only a little dilution during the malaxation phase. The oil coming from the decanter is then processed with a disk centrifuge (vertical centrifuge), which separates the residual water and the solid impurities to get a clear oil. In the three-phase process, a second disk centrifuge is used to recover the oil residues from the water phase. The two-phase process has a low water consumption and a low waste water production. From the solid phase a very humid paste is obtained that is difficult to treat further (thermal drying and extraction of the oil residue with solvents). Instead, in the threephase process, the solid phase is much drier and much more easily transportable and workable. In any case, water consumption and subsequently waste water production is higher than in the two-phase process. The selection of a treatment instead of another treatment depends on subjective circumstances.
10.5 THREE- AND TWO-PHASE CENTRIFUGAL DECANTERS: OPERATION AND TECHNOLOGICAL INNOVATION The most important part of these two machines (threephase and two-phase centrifugal decanter) is the rotary section composed of a cylindrical-conic drum where there is the screw feeder, rotating at differential speed. The rotary section is driven by electric motors through the belt transmission. Separation by using the centrifugal force occurs in the drum. In a two-phase decanter the product is separated into a liquid phase (oil) and a solid phase (kernel fragments, pulp and vegetable water). In a three-phase centrifugal decanter the product is separated into a light liquid phase (oil), into a heavy liquid phase (water) and into
a solid phase (kernel fragments and pulp). The separated oil is discharged by gravity in both cases, while in the threephase centrifugal decanter the separated water phase is discharged by pressure by a centripetal or gravity pump. The screw feeder transports the separated solids to the drum’s conic terminal, where they are discharged. Three-phase and two-phase centrifugal decanters are equipped with gear drives: in the first, an inverter activates the drum and controls the motor (thanks to the inverter it is possible to adjust the drum’s speed in a continuous way and fix a transmission ratio for the differential speed), while in the second, drum and screw drive are powered by motors through many inverters, which allow for a continuous adjustment of the drum’s speed as well as the differential speed for both. At the beginning of the 1990s, olive oil plant manufacturers launched new models of decanters in the market called ‘ARA’ (water saving decanter). In this three-phase extraction, a special design ensures that less water is used than with other comparable designs, so it produces, consequently, a small volume of vegetable waste water and an oil richer in polyphenol. As stated earlier, the decanter centrifuge called ‘of third generation’ has an innovative geometry, which ensures greater performance due to a longer cylindrical part of the bowl and shorter beach sections, and a special bowl with a variable dynamic pressure (VDP) cone system. Figure 10.4 shows the traditional decanter and the innovative VDP decanter. This makes it possible to electronically adjust the speed of the screw according to the torque on the conveyor screw. Moreover, with the VDP the automatic control of differential speed depends on the twisting moment, thus obtaining drier discharged solids, with lower oil residue contents. The centripetal pump can be adjusted, for an optimum liquid–liquid separation and it ensures a good separation of the oil and water phase that can be obtained by accurately adjusting the liquid discharge device. In 2003 Catalano et al. found that the variable dynamic pressure cone decanters allowed
89
CHAPTER | 10 Influence of Different Centrifugal Extraction Systems
Feed zone DD Gearbox
Feed
Light liquid phase
Normal cone
A
Solids
Feed zone DD Gearbox
Feed
B
Light liquid phase
Active baffle
Steep cone
Solids
FIGURE 10.4 Traditional decanter with a long drainage cone (A) and innovative VDP decanter with a short drainage cone (B). A vertical section of (A) the traditional decanter and (B) the innovative centrifugal decanter. In this figure the innovative design of the ‘third-generation’ decanter (VDP) is shown. This decanter presents a longer cylindrical part of the bowl and shorter beach sections. This new geometry improves the performance because the separation of the solid particles occurs principally in the cylindrical zone.
for the adjustment of the process parameters for optimum extraction yield. These results led to the following conclusions: the regulation of the differential velocity screw/bowl allows for a better performance of the decanter at low dilutions of the olive paste; it is possible to obtain high efficiency at a low dilution of the olive paste achieving a higher content of minor compounds; and the more advanced system of regulation leads to better results according to the variations of the ratio between liquid and solid phases, thus optimizing operational performances in relation to the rheological features of olive pastes. Before the introduction of the new decanter with short and variable and dynamic pressure cone, the employment of the de-stoner in the industrial processing line was not possible. In fact, a new technology in virgin olive oil production consists in the de-stoned olives. This means that the paste consists solely of the fleshy part of the olive (mesocarp), without the stone or pit (endocarp) that holds the seed. The absence of kernel fragments in olive paste causes a change in the olive paste viscosity, reducing the action of drainage of the liquid phases and the mechanical action of the malaxation step, resulting in a serious loss of oil yields. However, regulation of the differential speed of the conveyor ensured the machine was extremely flexible and also able to optimize the instrument performances in relation to the rheological features of olive pastes. Amirante et al. (2005) tested an innovative industrial continuous processing line made up of a de-stoner (instead of the usual metal crusher) and a new-generation VDP cone decanter (water-saving decanter with variable
speed of the conveyor). The results demonstrated that this decanter provides an effective optimization of oil yields and from de-stoned olive paste, and that it is possible to improve oil yields without compromising the quality of the product (in contrast to the traditional strategy of water dilution of the olive paste that causes losses of polyphenols).
10.6 EFFECT OF THREE- AND TWO-PHASE CENTRIFUGAL DECANTERS ON VIRGIN OLIVE OIL QUALITY Ranalli and Angerosa in 1996 studied the qualitative characteristics of virgin olive oil obtained from three olive varieties (Coratina, Nebbio, and Grossa di Cassano) comparing the two-phase centrifugal decanter with conventional threephase equipment. The results obtained, relating to the analytical characteristics of the oils, are given in Table 10.1. In general, results indicated that acidity, peroxide index, UV spectrophotometric indices, were not influenced or were only slightly influenced by the type of decanter adopted for the centrifugation of olive pastes. The oils obtained after extraction by the two-phase centrifugal system always exhibited (for all processed olive varieties) higher contents of polyphenols and ortho-diphenols. When water is added to the pastes, which are rich in polyphenols (that are soluble in both water and oil), a significant amount of such constituents is carried away from the oily phase because of partitioning between two non-mixable liquids. All oils produced by the two-phase centrifugal decanter received a higher
90
SECTION | I Olives and Olive Oil
TABLE 10.1 Analytical characteristics of oils extracted by a two- or three-phase decanter. This table reports that acidity, peroxide index, UV spectrophotometric indices, were not influenced or were only slightly influenced by the type of decanter adopted for the centrifugation of olive pastes. The oils obtained after extraction by the two-phase centrifugal system always exhibited (for all processed olive varieties) higher contents of polyphenols than the oils obtained after extraction by the three-phase centrifugal system. Taken with permission from Ranalli et al. (1996). Coratina
Acidity (% oleic acid) Peroxide index
Nebbio
Grossa di Cassano
2 phases
3 phases
2 phases
3 Phases
2 phases
0.25
0.27
0.33
0.34
0.22
12.0
14.6
9.1
9.8
8.0
3 phases 0.23 8.8
Total polyphenols (caffeic acid)
328
270
272
129
183
60
O-diphenols (caffeic acid)
226
178
188
92
121
43
Rancimat stability (h)
16.9
15.1
15.5
10.0
13.4
8.3
K232
1.699
1.734
1.510
1.430
1.347
1.260
K270
0.144
0.148
0.091
0.083
0.089
0.091
Panel test (score)
8.1
7.8
7.7
7.2
7.5
6.8
sensory score. In fact, sensory properties of virgin olive oil are largely affected by the phenolic composition. In particular, these compounds were associated with the typical bitter and pungent sensory aspects of olive oil. Several studies showed the relationships between the ‘bitter’ and the ‘pungent’ taste of virgin olive oil and the total phenol concentration (Gutiérrez Rosales et al., 1992; Tsimidou, 1998; Servili and Montedoro, 2002). Even oxidation resistance was greater, so that a positive influence of the two-phase continuous system on the shelf-life of the product could be hypothesized. These results were confirmed by Di Giovacchino et al. (2001). Their research has been carried out to ascertain the influence of different centrifugal decanters employed in olive process on oil yields and qualitative characteristics and composition of volatile compounds of virgin olive oil. In this research, the coefficient of the partition equilibrium of total phenols between oil and vegetable water was also assessed. Tests were performed in an olive oil mill equipped with centrifugal decanters at two or three phases. Results showed that oil yields were similar and oils extracted from good-quality olives do not differ in free fatty acids, peroxide value, or UV absorptions. Total phenols and o-diphenols content as well as induction time values are higher in oils obtained by the centrifugal decanter at two phases, because it requires less quantity of water added to olive paste in comparison to the three-phase centrifugal decanter. The amount of water added determines the dilution of the aqueous phase and lowers the concentration of the phenolic substances that are more soluble in vegetable waste water. Due to the partition equilibrium law the concentration of the same substances consequently diminishes in the oil. Water addition, in fact, modifies the concentrations of the soluble
substances in oily and aqueous phases achieved during the malaxation step. In fact, during this operation, the phenolic substances dissolve in oily and aqueous immiscible phases in contact, according to the corresponding value of the constant (K) of partition equilibrium, which is expressed by the following equation: [ A]
K⫽
aqueous phase
[ A]
oily phase
(10.1)
where K is the constant of partition equilibrium and [A] is the concentration of compound A, expressed as mol L⫺1 (or mg L⫺1). The value of K, in conditions of chemical-physical equilibrium, depends only on the temperature and, therefore, is constant at constant temperature. Of course, the law of the partition equilibrium is strictly valid for the individual compound only, and it is not valid for the total phenol content of oil and vegetable water. Many phenolic substances, in fact, are present in olive fruit, olive oil and vegetable waste water and each of them has a specific value of the constant of partition equilibrium (K) between the concentrations in the aqueous and oily phases in contact. However, it is possibly an approach to the study of this phenomenon considering the total phenol content (mg L⫺1) of vegetable waste water and olive oil, as reported in Figure 10.5. The data indicated that the total phenol content of vegetable waste water and olive oil were very different when the centrifugal decanter at two or three phases was used. The ratios of concentrations of the same substances in the aqueous and oily phase, in contrast, were very similar as reported in Table 10.2. The average values of the ratios of o-diphenol concentrations for two-phase and three-phase centrifugal decanters
91
CHAPTER | 10 Influence of Different Centrifugal Extraction Systems
Total phenols oil [mg/L]
600 500 400 300 200 100 0 3000
4000
5000
6000 7000 8000 Total phenols veg. wat. [mg/L]
9000
10000
11000
FIGURE 10.5 Relationship between total phenols content (as gallic acid) of oils and the corresponding vegetable waters (䊊: two phase; ◆: three phase). The total phenols content (mg L⫺1) of vegetable waste water and olive oil is shown. The data indicate that the total phenol contents of vegetable waste water and olive oil are very different when the centrifugal decanter at two or three phase was used. Taken with permission from Di Giovacchino et al. (2001).
TABLE 10.2 Average values of ratios between total phenols and o-diphenols Concentrations (mg L⫺1) of vegetable waste water (VWW) and oil (O.O.) obtained by centrifugal decanter at two or three phases. This table reports the ratios of concentrations of the same substances in the aqueous and oily phase. The data show that the average values of the ratios of total phenol concentrations was not significantly different using the centrifugal decanter at two or three phases. Taken with permission from Di Giovacchino et al. (2001). Ratios TotalphenolVWW TotalphenolO.O. Totalo-diphenolVWW Totalo-diphenolO.O.
Centrifugal decanter at two phases
Centrifugal decanter at three phases
26.0a
29.0a
38.3a
59.2a
Values with the same letter are not statistically different (p ⬍ 0.05, t-test).
a
were not statistically different (the data were tested by ttest to ascertain the statistical significance of the difference). The average values of the induction time of oils obtained by centrifugal decanter were higher at two phases (14.2 h) than at three phases (11.0 h) and they were significantly different ( p ⬍ 0.05 with t-test). Table 10.3 shows the average technological results obtained from olive processing by the centrifugal decanter employed at two or three phases. The data indicate that no significant differences in oil extraction yields, moisture and oil content of fresh pomace or oil content of vegetable waste water occurred, due to the different centrifugal decanters used. Also, Piacquadio et al. (1998) found that the twophase decanter leads to oils with high contents of phenolic components, which are natural antioxidants and as a consequence yield a good quality of oils with an increased resistance to autoxidation. This is confirmed by kinetic curves of peroxide accumulation shown in Figure 10.6, where the
two kinetic curves are the mean of three independent experiments. The kinetic results (Figure 10.6) indicate that oils extracted with the centrifugation system using a two-phase decanter show lower amounts of hydroperoxides, which are the only product during the initial stage of lipid oxidation in a kinetic regime, compared to oils obtained with the current centrifugation system with three-phase decanter.
SUMMARY POINTS ●
●
There are three different extraction systems, depending on the extraction method: (a) pressure, (b) percolation process and (c) centrifugation. The main advantages of centrifugal processing systems are that limited labor is needed and the oil yield performance is better than when the other different extraction systems are used.
92
SECTION | I Olives and Olive Oil
TABLE 10.3 Average technological results obtained from olive processing by centrifugal decanter at two or three phases. The average technological results obtained from olive processing by the centrifugal decanter employed at two or three phases are shown. The data indicate that no significant differences in oil extraction yields, moisture and oil content of fresh pomace or oil content of vegetable waste water occurred, due to the different centrifugal decanters used. Very important differences, in contrast, were ascertained in oil content (kg/100 kg olives) of fresh pomace, in quantity (L/100 kg olives) and oil content (kg/100 kg olives) of vegetable waste water. Taken with permission from Di Giovacchino et al. (2001). Determinations
Centrifugal decanter at two phases
Centrifugal decanter at three phases
86.1a
85.3a
74.0a
54.0b
a
54.0a
Oil extraction yield [%] Olive pomace Pomace quantity [kg/100 kg olives] Moisture [% on fresh pomace]
57.3
Oil [% on fresh pomace] Oil [kg/100 kg olives]
3.32a
3.37a
a
1.81b
2.48
Vegetable waste water Quantity [L/100 kg olives] Dry matter [% of the vegetable waste water]
6.1A
93.6B
15.0a
8.8b
a
1.2a
Oil [% of vegetable waste water]
1.4
Oil [kg/100 kg olives]
0.10A
1.13B
a
2.92a
Oil lost with byproducts [kg/100 kg olives]
2.58
* Values with the same letters are not statistically different (t-test). Different capital letters: significance p ⬍ 0.001; different lower case letters: significance p ⬍ 0.05.
B Peroxide value (meq O2 kg−1)
300
200
A
100
0
0
30
60
90
120
150
180
210
Time (min) FIGURE 10.6 Kinetic curves of peroxide accumulation during the oxidation of oils extracted with two-phase (A) and three-phase (B) decanters. This figure reports kinetic curves of peroxide accumulation. The kinetic results indicate that oils extracted with the new centrifugation system using a twophase decanter show lower amounts of hydroperoxides, which are the only product during the initial stage of lipid oxidation in a kinetic regime, compared to oils obtained with the current centrifugation system with a three-phase decanter. Taken with permission from Piacquadio et al. (1998).
CHAPTER | 10 Influence of Different Centrifugal Extraction Systems
●
●
●
●
During the extraction process, after the malaxation phase the paste can be pumped to a two-phase decanter or a three-phase decanter for separation. The oils obtained after extraction by the two-phase centrifugal system always exhibit higher contents of polyphenols than the oils obtained after extraction by the three-phase centrifugal system. The decanters called ‘ARA’ (water-saving decanter) have a special design which ensures that less water is used than with other comparable designs, so it produces, consequently, a small volume of vegetable waste water and an oil richer in polyphenol. The third-generation ‘variable dynamic pressure cone’ decanter (water-saving decanter with variable speed of the conveyor) provides an effective optimization of oil yields without compromising the quality of the product.
REFERENCES Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A., Pollicino, D., Tamborrino, A., Lo Turco, V., 2005. Virgin olive oil from de-stoned paste: Introduction of a new decanter with short and variable dynamic pressure cone to increase oil yield. Proceedings of EFFoST 2005 INTRADFOOD 2005 – Innovations in Traditional Foods (Valencia, SPAIN) Vol 2, 1183–1186. Balatsouras, G., 1999. In “I Elaiourgia” (The olive mill) Vol 4. Athens, Greece.
93
Baldioli, M., Servili, M., Perretti, G., Montedoro, G.F., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oil. J. Am. Oil. Chem. Soc. 73, 1589–1593. Catalano, P., Pipitone, F., Calafatello, A., Leone, A., 2003. Productive efficiency of decanters with short and variable dynamic pressure cones. Biosyst. Eng. 86 (4), 459–464. Di Giovacchino, L., Costantini, N., Serraiocco, A., Surricchio, G., Basti, C., 2001. Natural antioxidants and volatile compounds of virgin olive oils obtained by two or three-phases centrifugal decanters. Eur. J. Lipid Sci. Tech. 103, 279–285. Gutiérrez Rosales, F., Perdiguero, S., Gutiérrez, R., Olías, J.M., 1992. Evaluation of bitter taste in virgin olive oil. J. Am. Oil Chem. Soc. 69, 394–395. Kapellakis, I.E., Tsagarakis, K.P., Crowther, J.C., 2008. Olive oil history, production and by-product management. Rev. Environ. Sci. Biotechnol. 7, 1–26. Kiritsakis, A., 1998. In ‘Olive oil’, 2nd edn. From the tree to the table. Food and Nutrition Press, Inc., Trumbull, Connecticut, 006611, USA. Piacquadio, P., De Stefano, G., Sciancalepore, V., 1998. Quality of virgin olive oil extracted with the new centrifugation system using a twophases decanter. Fett/Lipid. 100 (10), 472–474. Pike, O.A., 1998. Fat characterization. In ‘Food analysis’ (SS Nielsen), 2nd edn. Aspen Publishers, Gaithersburg, pp. 217–235. Ranalli, A., Angerosa, F., 1996. Olive oil extraction with integral centrifuges. J. Am. Oil Chem. Soc. 3 (4), 417–422. Servili, M., Montedoro, G.F., 2002. Contribution of phenolic compounds to virgin olive oil quality. Eur. J. Lipid Sci. Tech. 104, 602–613. Tsimidou, M., 1998. Polyphenols and quality of virgin olive oil in retrospect. Ital. J. Food Sci. 10, 99–116.
This page intentionally left blank
Chapter 11
A Marker of Quality of Olive Oils: The Expression of Oleuropein Giovanni Sindona Dipartimento di Chimica, Università della Calabria, Arcavacata di Rende (CS), Italy
11.1 INTRODUCTION
Figure 11.1), carrying a hydroxytyrosol moiety of known antioxidant properties (Del Boccio et al., 2003; Puel et al., 2004), is considered the archetype of the so-called olive oil ‘polyphenols’ and is present in all olive tissues, being its relative amount higher in leaves and drupes. Its assay has become a need also in biological fluids, owing to the widespread distribution of food integrators based on this natural active principle and to the many beneficial effects reported for it. The assay of Olp is also becoming important in the veterinary field since the Mediterranean tradition of feeding goats and sheep with olive leaves has been recently extended to cows, aiming at providing a constant intake of a natural anti-inflammatory/antioxidant molecule which could promote animal health (Regione Calabria Project, 2004) and wellbeing, of particular importance for dairy farms.
Olive oil is a typical Mediterranean food whose quality depends on a number of parameters that are not always easily controlled by conventional chemical analysis. Therefore over the years highly sophisticated physicochemical methods have been introduced to set up protocols that could assess both the quality and the origin of the aliment. Nuclear magnetic resonance spectroscopy (NMR) is a good source of information on the major constituents of oil, but fails in the direct structural characterization of trace components from the complex matrices represented by the oil itself or by its extracts in organic solvents (Mannina et al., 2003). Therefore even if NMR is the method of choice for a full structure determination of pure compounds it can not be applied in the assay of a minor component present in the oil at the ppb level. Olive oil is celebrated not only for its nutritional value but also for the content of minor amounts of pharmacologically active principles, belonging to the nutraceutical family, otherwise known as functional foods, such as the polyhydroxylated phenolic and catecolic species. Oleuropein (Olp 1,
11.2 OLP EXPRESSION The many applications briefly outlined before call for a reliable and affordable method of assaying Olp in various matrices. This molecule moreover can be considered
HO HO O H
O
O R
O O
HO O
H HO
H
H OH
H H
OH
FIGURE 11.1 Oleuropein structure. The catechol antioxidant moiety is shown in red in the formula. R is equal to CH3 (1) or CD3 (2). Oleuropein (1) can be enzymatically degraded by esterases at the ester functions (red arrows), by β-glucosidase at the glycosidic linkage (blue arrow), as well as by acid-base and thermal treatments. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
95
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
96
SECTION | I Olives and Olive Oil
TABLE 11.1 New trace antioxidants found in olive leaves. [M ⴙ NH4]ⴙ
Structures
m/z 348
H3CO O
OH
HO O
OH OH
OH m/z 334 HO O
OH
HO O
OH OH
OH m/z 364
OCH3 HO O
OH
HO O
OH OH
OH m/z 562 HO
H3COOC O
HO
O
O O OH
O
OH
HO OH m/z 426
H3COOC O
O
O
O OH
O
OH
HO OH m/z 586 H3CO
H3COOC O
H3CO
O
O O OH
O
OH
HO OH
The glycosides reported in the table have been found in leaves. Their presence in olive oil could prove the illegal addition of leaves during milling procedures. The presence of Olp in oil should only be due to the milled drupes.
97
CHAPTER | 11 A Marker of Quality of Olive Oils: The Expression of Oleuropein
a marker of quality and as such it could represent an alternative way of labeling the production of good quality of extra virgin olive oils. It undergoes, in fact, degradation if exposed to enzymatic or chemical treatments, hence its relative amount in oil can be correlated to those pedoclimatic changes, and food processing techniques which can affect the integrity of the molecule. Fraudulent addition of leaves during olive milling, which inevitably raises the content of oleuropein, can be confidently verified by extending the investigation to those Olp-related molecules which have been recently found in leaves only (Table 11.1) (Di Donna et al., 2007). Olp is mainly dispersed in the fatty matrix of oil, therefore its assay may pose recovery problems. Chromatographic (Tan et al., 2003; Tsarbopoulos et al., 2003) as well as mass spectrometric (De Nino et al., 1997, 1999, 2000; Perri et al., 1999; Caruso et al., 2000) methods have been applied to detect its presence in different natural and biological fluids. LC-MS/MS has definitely emerged as the method of choice for its high specificity and sensitivity. Although the mentioned mass spectrometric methodologies perfectly apply to qualitative and quantitative assay of 1, the maximum recovery at the ppb level, needed in the case of oil, can be achieved by use of the isotope dilution method only. A proper deuterium-labeled Olp (2, Figure 11.1), sharing identical gas-phase chemistry with the natural one, was therefore devised on the ground of the fragmentation pattern of 1 (Table 11.2). The latter, ionized by the formation of a protonbound hetero-dimer with ammonia, gives rise to the
A 100%
137.1
TABLE 11.2 Olp content in filtered and not-filtered extra virgin olive oil from different cultivars grown in Calabria (Italy). Cultivar
Procedure
mg kg⫺1
Carolea
Filtered
357 ⫾ 14
not filtered
263 ⫾ 4
Filtered
175 ⫾ 2
not filtered
132 ⫾ 5
Filtered
344 ⫾ 8
not filtered
214 ⫾ 2
Coratina
Frantoio
The table show that the final treatment of the oil by a conventional manufacturing procedure can be assessed by determining by means of the proposed method, the relative amount of oleuropein.
[M ⫹ NH4]⫹species at m/z 558 (Figure 11.2(A) undergoing gas-phase fragmentation which leads to the breakdown of the glycosyl linkage, product ions at m/z 361 and 379, and to the releasing of the hydroxytyrosol moiety, at m/z 137. The labeled compound 2 was therefore planned in order to preserve the H/D isotopic distribution within the product ion at m/z 137. In other words, to allow a simultaneous detection of 558→137 and 561→137 transitions, due to the analyte and the reference, respectively (Figure 11.2B).
361.1
Rel. InL (%)
379.1
0% 100 B 100%
150
200
558.2
347.1
225.0
165.0
287.3 250
541.1
329.2 300
350
400
450
500
m/z, u
550
137.1
Rel. InL (%)
364.1
[M+NH4]+
d3 = 100,0, d2 = 5.5, d1 = 2.2 228.0
168.2 0% 100
150
200
250
347.0 287.0 300
350
382.2 400
561.1
438.0 450
500
550
m/z, u
FIGURE 11.2 ESI-MS/MS spectra of (A) oleuropein (1) and (B) of the d3 homologue 2. The formation of the production at m/z 137 does not involve the heavy isotopes. The distribution of the deuterium atoms is also shown. The [M ⫹ NH4]⫹ ammoniated ions of oleuropein and its d3 analogue fragments through common processes. The formation of m/z 137 does not involve the deuterium ions and corresponds to the base peak of the spectra. Natural samples of oil are spiked with known amount of 2 and from the area of the two transitions 561→137 and 558→137, the known amount of 1 is determined with high precision and accuracy.
98
SECTION | I Olives and Olive Oil
The isotope absolute dilution method was applied, in connection with ESI, or APCI-MS/MS, to the assay of Olp in olive oil produced in different manufacturing conditions, providing the sensitivity and specificity needed when the active principle is dispersed in fatty matrices. One of the first applications of the method was represented by the investigation of the effect of the ripening phase of the olives on the content of this secoiridoid glycoside. It is generally accepted that the quality of the oil is inversely proportional to pigmentation of the milled drupes; different parameters have been correlated to this observation. The assay of Olp content represents the scientific approach, at the molecular level, which can better relate quality of oil with ripening stage of the drupes. The molecule becomes, in fact, progressively exposed to endogenous enzymes as a consequence of the increasing breakdown of cell membranes with maturation. Actually, the amount of 1 drops down by nearly 80% on going from full green to 100% superficial pigmented drupes, as shown for oils produced from Cassanese cultivar (Figure 11.3) (De Nino et al., 2005). Although oleuropein is present in olive oil in the range of hundreds of ppb, nonetheless the availability of a reliable detection method confidently allowing the evaluation of tens of ppb provides unique tools which can be useful for other correlations, always in the direction of the quality control of the foodstuff. The method was, in fact, applied to a survey of virgin olive oils produced from different cultivars grown in different Italian regions as a function of different agronomical and processing parameters within a joint national project. A systematic investigation of Olp content as a function of cultivar type, pedoclimatic conditions and production
0.45 0.4 0.35 0.3
ppm OLP
0.25 0.2 0.15 0.1 0.05 0 1
2
3
FIGURE 11.3 Ripening-phase-dependent oleuropein content in virgin olive oils of Cassanese cultivars. Oleuropein assay in olive oil. (1) 100% green; (2) 100% superficial pigmentation; (3) 100% deep pigmentation. The results presented in this figure show that as soon as the enzymes present in the drupes are released, as a consequence of the progressive ripening of the fruit, they cause a degradation of the original content of Olp. This phenomenon is directly proportional to the degree of pigmentation which could take place when the drupes are still not harvested or, even worse, after a prolonging stocking of the crops.
technology was carried out with experimental oils. A general tendency was observed in the enhancement of Olp content with the latitude of the harvested drupes, independently of the examined cultivars. Olives from Carolea showed a variation of Olp ranging from 0.357 to 0.245 ppm when grown in different Italian regions from 38° 57⬘ N to 42° 28⬘ N latitude, respectively. In the case of Coratina, a typical cultivar from Apulia, the maximum amount of 1 was, actually, found in oils produced from olives grown in this region at 40° 37⬘ N latitude, with respect to those produced from the same cultivar in other Italian regions (De Nino et al., 2008). A correlation was also established by the absolute amount of 1 and the manufacturing procedures. Properly processed oils are, as expected, rich in secoiridoid anti-oxidant molecules such as oleuropein; the relative amount of these ingredients drops, in some cases, by nearly 50% on going from filtered to non-filtered products. In this case it can be assumed that the presence of traces of β-glucosidase in the less-processed foodstuff causes a progressive degradation of oleuropein. The presence of oleuropein can, therefore, be related to food processing, thus, to the quality and safety of the total olive oil chain. The absolute content of 1 was also a significant parameter to match the quality of extra virgin olive oils produced by the conventional and an innovative stoning procedure using the same olives in prototype machinery. In the new approach the flesh of the drupes was separated from the stone by a Spring Pitting Apparatus (SPIA) and directly used with conventional mills for the preparation of oil. The amount of Olp was consistently higher in oils produced from the flesh than from the entire fruit (Amirante et al., 2000; Patumi et al., 2003; Lavelli and Bondesan, 2005). The observed differences were also related to the cultivar type; in fact, the ratio of 1 between conventional and stoned oil varied from 4.0 to 1.5 on going from Carolea to Cassanese cultivars, respectively. The higher amount of this important secoiridoid glucoside in oils prepared from stoned olives should necessarily be correlated with the adopted procedure. Stoned olive oil can also be prepared in commercial plants; a general agreement on quality improvement when stoned drupes are processed was, however, not reached (Amirante et al., 2000; Patumi et al., 2003; Lavelli and Bondesan, 2005). The observed variation in oleuropein concentration was clearly established by the isotope dilution method; the presence of the crushed pit in conventional production of oil should, therefore, be important. If this effect is due to the action of enzymes freed from the pit, it should be confirmed in blank experiments carried out on pure samples of oleuropein. When oleuropein is exposed to the action of the water extracts of crushed Carolea pits, after 60 minutes of incubation, the ESI spectrum reported in Figure 11.3 was obtained. The [M-H]⫺ species of deglycosylated 1 and of its open methyl acetal derivative (not shown in Figure 11.4) were about the only detectable species.
99
CHAPTER | 11 A Marker of Quality of Olive Oils: The Expression of Oleuropein
100
377.14
O %
O
OH O
O H3COOC
HO
378.15 379.21 0 240
260
280
300
320
340
360
380
400
420
440
460
480
m/z u
FIGURE 11.4 ESI spectrum (negative ions) from aqueous solution of standard Olp (1) incubated for 60 minutes with water extracts of Carolea pits. Pure Olp extracted from olive leaves was incubated in water with the enzymes released by pits. The ESI-MS spectrum of the water solution did not show any peak corresponding to 1. The only detectable species were that displayed in the ESI negative spectrum and an artefact derived by the semiacetal ring opening of the methanol used as solvent for recording the mass spectrum.
The results provide clues, at the molecular level, that olive pits, at least for the investigated cultivars, release when crushed β-glucosidase enzymes, which inevitably interact with the glucosylated species there present, such as oleuropein. No evidence of esterases activity was detected by MS/MS. The two ester functions of 1 were, apparently, not hydrolyzed or at least no traces of the species that should be obtained were displayed in the MS/MS spectra. Other peculiar information, which was difficult to be achieved with other methodologies, was the indication that the kinetic of Olp degradation was a function of the cultivar. Some differences were, in fact, observed on the relative amount of Olp present in conventional and stoned oils produced from different cultivars. Since the experimental oils were prepared from the same drupes, by varying the mechanical procedures only, the observed effect should be ascribed to the different enzyme content of the pits of different cultivars. When pure Olp is exposed to water extracts of crushed stones from Carolea and Cassanese cultivars, for 10 minutes, 51 and 36% of the original glycosides were consumed, respectively (Figure 11.5). Oleuropein has been considered throughout the different applications described above as a good marker of all the steps of olive oil production. The application of the method has allowed, in fact, the correlation of the relative amounts of this secoiridoid in oil with (i) the ripening phase and cultivar of the processed olives, and (ii) the adopted experimental procedures, (a) in the preparation of the oil and (b) in the final purification process. Olp is recognized as a good antioxidant and antiinflammatory molecule. Its presence in olive oil accounts for a few hundreds of ppb, an amount which hardly can reach any toxicity limit, if it is taken into account an intake of oil in the range of grams per day, as happens in Mediterranean countries. Nevertheless the role of oleuropein should not be underestimated as a pro-oxidant, in vitro at least (Figure 11.6). As many other popular nutraceuticals, such as vitamin C and
β-carotene, Olp can undergo electron capture, in a Fentonlike environment (Podmore et al., 1998). In this case the redox role of the catecolic moiety was demonstrated by the lack of activity when one or both the phenolic hydroxyl groups were methylated. The dual behavior of catecols in vivo may represent a drawback when overdoses of nutraceuticals are assumed with diet. 90 80 70 B
60 50 40
Olp ppm
30 A
20 10 0 0
2
4
6
8
10
12
Time (min) FIGURE 11.5 Oleuropein metabolism in water extracts of crushed pits from (A) Carolea and (B) Cassanese cultivars. The oil obtained from stoned olives showed a higher oleuropein content than that conventionally produced. The average distribution of Olp in different cultivars reflects the kinetic of its metabolism in the presence of stone enzymes.
OGLu OH
O
HO HO
OGLu O2• –
O H O
O
R
O
HO
O
R
O + H2O2 O
FIGURE 11.6 Oxidation of the Olp catecol moiety by superoxide anion. The superoxide anion formed in the Fenton reaction is reduced by the catecol moiety of Olp thus producing deuterium which characterizes a typical oxidant environment. In this redox process Olp acts as a pro-oxidant molecule. The process is blocked when the hydroxyl groups are methylated.
100
SECTION | I Olives and Olive Oil
Oleuropein, a secoiridoid glucoside present in olive tissues, is a biomarker which can be used to trace the entire chain of production of extra virgin olive oil. The choice is based on the enzymatic and chemical lability of the molecule whose absolute amount can be correlated with the introduction of best practice principles in all steps, from the production to the stocking and commercialization of the oil.
SUMMARY POINTS ●
●
●
●
●
●
Oleuropein is a secoiridoid produced by secondary metabolism of plants and is present in all olive tissues. As a glycoside bearing in the aglycon two ester functions it is amenable to chemical and enzymatic degradations. The absolute amount of oleuropein in any tissue or fluid can be determined with the required sensitivity and accuracy by APCI-MS/MS by MRM applying the isotope dilution method. Oleuropein can be considered a trace element in extra virgin olive oil, accounting for a few hundreds of ppb, and is a marker of quality and of best practice along the entire chain of production. Stoned oil produced with a prototype SPIA apparatus is richer than the conventional one of oleuropein. The amount of oleuropein is higher in conventional extra virgin olive oil filtered at the end of the manufacturing process. Oleuropein can act as a pro-oxidant in the presence of ferric ions, interacting with the superoxide produced in the Fenton-like redox process from oleuropein iron complexes.
REFERENCES Amirante, P., Clodoveo, M.L., Dugo, G., Leone, A., Tamborrino, A., 2000. Advance technology in virgin olive oil production from traditional and de-stoned pastes: Influence of the introduction of a heat exchanger on oil quality. Food Chem. 98, 797–805. Caruso, D., Colombo, R., Patelli, R., Giavarini, F., Galli, G., 2000. Rapid evaluation of Phenolic component profile and analysis of Oleuropein Aglycon in olive oil by atmospheric pressure chemical Ionization-Mass Spectrometry (APCI-MS). J. Agric. Food Chem. 48, 1182–1185. De Nino, A., Mazzotti, F., Perri, E., Procopio, A., Raffaelli, A., Sindona, G., 2000. Virtual freezing of the hemiacetal–aldehyde equilibrium of the aglycones of oleuropein and ligstroside present in olive oils from
Carolea and Coratina cultivars by ionspray ionization tandem mass spectrometry. J. Mass Spectrom. 35, 461. De Nino, A., Mazzotti, F., Morrone, S.P., Perri, E., Raffaelli, A., Sindona, G., 1999. Characterization of Cassanese olive cultivar through the identification of new trace components by Ionspray Tandem Mass Spectrometry. J. Mass Spectrom. 34, 10. De Nino, A., Di Donna, L., Mazzotti, F., Sajjad, A., Sindona, G., Perri, E., Russo, A., De Napoli, L., Filice, L., 2008. Oleuropein expression in olive oils produced from drupes stoned in a spring pitting apparatus (SPIA). Food Chem. 106, 677–684. De Nino, A., Di Donna, L., Mazzotti, F., Muzzalupo, E., Perri, E., Sindona, G., Tagarelli, A., 2005. Absolute method for the assay of Oleuropein in olive oils by atmospheric pressure chemical Ionization Tandem Mass Spectrometry. Anal. Chem. 77, 5961. De Nino, A., Lombardo, A., Perri, E., Procopio, A., Raffaelli, A., Sindona, G., 1997. Direct identification of Phenolic Glucosides from olive leaf extracts by atmospheric pressure Ionization Tandem Mass Spectrometry. J. Mass Spectrom. 32, 533. Del Boccio, P., Di Deo, A., De Curtis, A., Celli, N., Iacoviello, L., Rotilio, D., 2003. Liquid chromatography–tandem mass spectrometry analysis of oleuropein and its metabolite hydroxytyrosol in rat plasma and urine after oral administration. J. Chromatogr. B 785, 47–56. Di Donna, L., Mazzotti, F., Salerno, R., Tagarelli, A., Taverna, D., Sindona, G., 2007. Secondary metabolism of olive secoiridoids. New microcomponents detected in drupes by electrospray ionization and high-resolution tandem mass spectrometry. Rapid Commun. Mass Spectrom. 21, 3653–3657. Lavelli, V., Bondesan, L., 2005. Secoiridoids, tocopherols, and antioxidant activity of monovarietal extra virgin olive oils extracted from destoned fruits. J. Agric. Food Chem. 53, 1102L–1107L. Mannina, L., Sobolev, A.P., Segre, A., 2003. Olive oil as seen by NMR and chemometrics. Spectroscopy Europe 15, 6–14. Patumi, M., Terenziani, S., Ridolfi, M., Fontanazza, G., 2003. Effect of fruit stoning on olive oil quality. J. Am. Oil Chem. Soc. 80, 249–255. Perri, E., Raffaelli, A., Sindona, G., 1999. Quantitation of Oleuropein in Virgin olive oil by Ion spray mass Spectrometry-selected reaction monitoring. J. Agric. and Food Chem. 47, 4156. Podmore, I.D., Griffiths, H.R., Herbert, K.E., Mistry, N., Mistry, P., Lunec, J., 1998. Vitamic C exhibits pro-oxidant properties. Nature 392, 559. Puel, C., Quintin, A., Agalias, A., Mathey, J., Obled, C., Mazur, A., Davicco, M.J., Lebecque, P., Skaltsounis, A.L., Coxam, V., 2004. Olive oil and its main phenolic micronutrient (oleuropein) prevent inflammation-induced bone loss in the ovariectomised rat. Br. J. Nutr. 92, 119–127. Regione Calabria project POR Mis. 3.16-2004 Tan, H.-W., Tuck, K.L., Stupans, I., Hayba, P.J., 2003. Simultaneous determination of oleuropein and hydroxytyrosol in rat plasma using liquid chromatography with fluorescence detection. J. Chromatograph. B 785, 187–191. Tsarbopoulos, A., Gikas, E., Papadopoulos, N., Aligiannis, N., Kafatos, A., 2003. Simultaneous determination of oleuropein and its metabolites in plasma by high-performance liquid chromatography. J. Chromatograph. B 785, 157–164.
Chapter 12
Olive Oil Authenticity Evaluation by Chemical and Biological Methodologies Miguel A. Faria1, Sara C. Cunha1, Alistair G. Paice2, Maria Beatriz P.P. Oliveira1 1 2
REQUIMTE-Serviço de Bromatologia, University of Porto, Portugal Department of Clinical Biochemistry and Nutrition and Dietetics, King’s College School of Medicine, London, UK
12.1 INTRODUCTION Olive oil authenticity continues to be the target of many studies reflecting its high economic value, not only for the producers in Mediterranean countries, but also for consumers worldwide. The latter expect products that correspond to the information on the packaging label. A vast amount of legislation relating to the accurate characterization and classification of olive oil has boosted scientific work on this subject. The main aim of these studies is to control changes of olive oil composition in order to protect both the consumer from adulteration and the industry from unfair competition, as well as preserving the prestigious image of olive oil itself. In fact, the high market price of olive oil might lead to the unscrupulous blending of this product with cheaper vegetable oils such as pomace or refined oils in an unregulated manner. High-quality olive oils such as those labeled as Protected Designation of Origin (PDO) or mono-varietal olive oils can provide opportunities for very sophisticated types of adulterations that are very difficult to detect. Numerous attempts have been made to find suitable methods to authenticate the composition of olive oils. The number of analytical techniques used to face this challenge approaches the number of ways of actually adulterating it, ranging from classical determination of chemical parameters to highly sophisticated instrumental and biological techniques. Generally, the analytical techniques used investigate one or more of the constituents of olive oil, such as triacylglycerols, fatty acids, sterols, alkanes, waxes and aliphatic alcohols. These can prove the presence/absence of adulterants, or verify geographical or cultivar origin by comparing with known samples or legally established limits. In recent years several reviews have been published using general techniques to authenticate olive oils (Aparicio and Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Aparicio-Ruíz, 2000; Arvanitoyannis and Vlachos, 2007) or utilizing special techniques such as chromatographic techniques (Lercker and Rodriguez-Estrada, 2000). Rather than being an exhaustive revision of published works, the aim of the present chapter is to summarize the analytical methods utilized so far to authenticate and provide quality control for olive oils, especially those categorized as extra virgin olive oil and virgin olive oil. Methods referred to are further divided into the type of technique used in the analytical methodology as follows: (i) chromatographic, (ii) spectroscopic/spectrometric and (iii) DNA-based techniques (Figure 12.1).
12.2 CHROMATOGRAPHIC METHODS Gas chromatography (GC) and liquid chromatography (LC) are undoubtedly the prevailing techniques in the determination of macro- and micro-components within foodstuffs. Due to the rapid and reliable separation of the analytes, these techniques have long been established as the ‘gold standard’ for olive oil authenticity and traceability.
12.2.1 Gas Chromatography The high number of theoretical plates and the variety and capabilities of the detectors that can be coupled to GC make it the first choice in the analysis in a large number of fatty acids with similar physico-chemical characteristics. In general, the use of ester derivatives, especially methyl esters, is recommended in the analysis of fatty acids within olive oils. These esters are more volatile than the corresponding free fatty acids and therefore more suitable for analysis in the gaseous form (Casal and Oliveira, 2006). The detection of fatty acids in routine analysis is usually performed
101
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
102
SECTION | I Olives and Olive Oil
FIGURE 12.1 Analytical approaches used in olive oil authenticity evaluation. Several analytical approaches have been used in the last few years for the evaluation of olive oil authenticity. Analytical methodologies (round boxes) are mainly based on chromatographic, spectroscopic/spectrometric and DNA-based techniques.
with a flame ionization detector (FID) that responds to non-oxidized carbon in a linear relationship. Fatty acid composition varies from year to year and with the harvest date, which probably reflects differences in the amount of summer rainfall in the growing region as well as temperature variation during the olive fruit ripening and oil biosynthesis (Beltrán et al., 2004). Several studies have demonstrated that comparison of fatty acid composition detects virgin olive oil adulteration with sunflower seed oil (Gamazo-Vázquez et al., 2003), deodorized olive oil (Saba et al., 2005) and vegetable refined oils (Hajimahmoodi et al., 2005). GC-FID has also been applied in the analysis of sterols in olive oils. In most studies the sterols were analyzed as free sterols after hydrolysis from their conjugates and derivatization to their trimethylsilyl (TMS) esters (Cercaci et al., 2003). In addition some studies have been suggested that involve the separation of the free sterol and steryl fatty acid ester fractions in olive oils (Cunha et al., 2006a). Despite sterol composition and content of olive oil being affected by factors such as the cultivar, crop year, degree of fruit ripeness, storage time of fruits before oil extraction and method of oil extraction (Boskou, 2006), the fraction of sterols is frequently used to track commercial frauds utilizing other vegetable oils such as hazelnut oil (Damirchi et al., 2005). The peculiar content of sterols is also used to characterize monovarietal virgin olive oil (Alves et al., 2005).
Alcohols constitute a minor component of olive oil that can also be analyzed by GC. This alcoholic fraction includes both aliphatic alcohols (docosanol, tetracosanol, hexacosanol and octacosanol) and triterpene alcohols (β-amyrin, butyrospermol, cycloartenol and 24-methylenecycloartanol), the determination of which is useful in verifying the virgin olive oil (Boskou, 2006). The other class of substances usually analyzed by GC are the volatile components of olive oil – mainly aldehydes, esters, alcohols and ketones (Aparicio et al., 1997). Volatile components can be used to check the quality of olive oil (Angerosa, 2002), to detect adulteration (Flores et al., 2006), in the detection of possible ‘off ’ flavors (Morales and León-Camacho, 2000) and to determine the variety of the olive fruit used (Baccouri et al., 2007). This analysis is generally performed by solid-phase micro-extraction (SPME) and GC coupled to mass spectrometry (MS), which allows the simultaneous identification and quantification of numerous substances. Detailed studies on the chemical structure of wax esters occurring in olive oils have also been performed using GC-MS (Morales and León-Camacho, 2000). Virgin olive oil has a higher content of C36 and C38 waxes than C40, C42, C44, and C46 waxes, whereas the reverse is true in both olive pomace oil and refined olive oils (Morales and León-Camacho, 2000). This is used to distinguish these products.
CHAPTER | 12 Olive Oil Authenticity Evaluation by Chemical and Biological Methodologies
12.2.2 Liquid Chromatography Liquid chromatography is useful in the separation of all types of organic chemicals independent of polarity or volatility, especially in the analysis of polar, thermolabile and/or non-volatile chemicals not easily performed by GC. The predominant component of the olive oil, triacylglycerols (TAG), has been investigated using high-performance liquid chromatography (HPLC) with a reversed-phase (RP) column (Cunha and Oliveira, 2006). This can utilize ultraviolet (UV), refractive index (RI) or evaporative light-scattering detectors (ELSD). ELSD is the best method out of these as no baseline drift occurs and there are no limitations on the use of mobile phase solvents compared with the other detection methods (Cunha and Oliveira, 2006). The TAG profile allows the geographic origin or monovarietal identification of olive oils utilizing multivariate statistical analysis (Cunha et al., 2005). Antioxidant components such as the tocopherols, which play a key role in preventing oil from becoming rancid during storage, have been measured by HPLC by either normal- or reversed-phase columns. Whilst normalphase columns provide better selectivity and allow separation of β- and γ-tocopherols, the RP-HPLC shows higher column stability, reproducibility of retention times, and shorter analysis time (Cunha et al., 2006b). When UV, ELSD and fluorescence (FLD) connected diode array (DAD) detectors are compared in terms of analyte determination the FLD approach provided the most information and confirmed the identity of tocopherols and tocotrienols (Cunha et al., 2006b). Even though α-tocopherol is traditionally considered the major antioxidant of olive oil, other tocopherols such as β-, γ- and δ- plus α-, β-, and γ-tocotrienols have also been reported in it (Cunha et al., 2006b; Matos et al., 2007). The concentration of tocopherols and tocotrienols in olive oil varies according to the geographic origin or the cultivar used. These compounds can be employed as markers of authenticity for vegetable oils. Phenols are another minor component in olive oil. Their levels and profiles reflect agronomic factors, maturity of the olives, processing, packaging and storage (Boskou et al., 2005). These substances might be potential markers of geographic origin or olive fruit variety. Phenolic acids such as gallic, caffeic, vanillic, p-coumaric, syringic, ferulic, homovanillic, p-hydroxybenzoic and protocatechuic acids, tyrosol and hydroxytyrosol can be identified and quantified in olive oils (Boskou et al., 2005). In general, analysis of the phenolic fraction is performed by RP-HPLC with different detectors such as UV, FLD, coulometric electrode array detection or amperometric detectors. More recently LC coupled to MS has been applied, allowing better sensitivity and selectivity in the analysis of total and individual phenolic acids (Bendini et al., 2007).
103
12.3 SPECTROSCOPIC/SPECTROMETRIC METHODS 12.3.1 Nuclear Magnetic Resonance (NMR) Despite its inferior sensitivity compared to both GC and HPLC techniques, NMR has several advantages. It is nondestructive, selective and can simultaneously detect several low-mass components in complex mixtures utilizing simple sample preparation procedures. Ogrinc et al. (2003) published a revision study examining the use of NMR and mass spectrometry (MS) in the authentification of olive oils and other food products. Using 1H, 13C and 31P NMR spectroscopy, several authors have detected adulteration with seed oils (soybean, peanut, maize, hazelnut, sunflower, walnut, coconut, almond, rapeseed, etc.) by measuring the levels of n-3 linolenic acid (Sacchi, 2001) or the ratio of 1,2-diglycerides to total diglycerides (1,2-diglycerides and 1,3-diglycerides) together with acidity, iodine value and fatty acid composition (Vigli et al., 2003). McKenzi, in a 2004 work, has developed a rapid (less than 20 minutes) 13 C NMR methodology to authenticate extra virgin olive oil based on the determination of the major fatty acids (oleic, linoleic and saturated acids). This demonstrates the potential of this technique. In order to detect the adulteration of olive oils with low percentages of hazelnut oils, GarcíaGonzalez et al. (2004) utilized an artificial neural network (ANN) based on 1H and 13C NMR data which resulted in a mathematical model capable of detecting the admixture of 0.8% hazelnut oil in olive oil. 13C NMR is particularly useful in the evaluation of the fatty acid mixtures including trans fatty acid isomers. The constituents of the unsaponifiable fraction of olive oils can also be analyzed by 13C NMR as a means of discriminating virgin olive oils from refined and olive pomace oils (Zamora et al., 1994). Extra virgin, refined and lampante olive oils were investigated using multivariate statistical analysis applied to 31P NMR spectra. Admixtures of 5% of refined and lampante grade oils could be detected within virgin ones (Fragaki et al., 2005).
12.3.2 Fluorescence Spectroscopy Fluorescence spectroscopy has been used to determine the authenticity of olive oils. This has several advantages including speed of analysis, the small amount of sample required and the fact that there is no requirement for reagents. In the last few years instrumental improvements, together with the availability of software that allows the information contained in fluorescence spectra to be extracted, have contributed to the development of this technique (Guimet et al., 2005). In particular, the possibility of using excitation emission matrices (EMM) – a set of emission spectra recorded at several excitation wavelengths – have boosted research in this area. The ability to detect hazelnut crude or refined oil
104
was tested using this methodology (Sayago et al., 2007) in virgin and refined olive oils. The degree of adulteration of virgin olive oils (VOO) with refined olive pomace oil (OPO) was also studied by interpreting EMM data using several statistical discrimination techniques (Guimet et al., 2005). The potential of these techniques has been demonstrated, but a strong statistical background is needed to achieve satisfactory results.
12.3.3 Mass Spectrometry (MS) In recent years advanced MS methods such as HeadspaceMS (HS-MS), Isotope Ratio MS (IRMS), Electrospray Ionization MS (EIS-MS) and Inductively Coupled Plasma MS (ICP-MS) have been applied in the authentication of olive oils. No chromatographic or any other kind of separation process is required when using HS-MS. Thus a global signal corresponding to the volatile fraction of the sample is obtained. This is then employed for the characterization of samples by chemometric methods. Peña et al. (2005) used this methodology to detect the addition of hazelnut, sunflower and OPO to VOO. The IRMS approach entails measuring the isotope ratio (2H/1H, 13C/12C, 15N/14N and 18 16 O/ O) of an analyte isotopically representative of the original sample. The pattern of naturally occurring isotopes is influenced by biochemical properties of plants and geoclimatic conditions making it dependent on the geographical and botanical origin of plants (Ogrinc et al., 2003). IRMS has been used in this way to identify the geographic origin of VOO sold with certified origin appellation, based on the 13C values of palmitic and oleic acids. It has also been used to detect the blending of other edible oils based on the 13C values of the same fatty acid co-variations. The 13C values of the aliphatic alcoholic fraction of oils have been used to detect the adulteration of both VOO and refined olive oils with OPO down to a level of 5% (Angerosa et al., 1997). Soft ionization MS methods are required to analyze both small molecules and biomacromolecules. This approach has been used in the very informative ‘wholefood fingerprinting techniques’ utilized in metabolomics. ESI-MS can provide soft ionization and has been used to produce informative and discriminating mass spectra of VOO and its common adulterants (Goodacre et al., 2002). Catharino et al. (2005) used the same approach applied to the polar components of several oils extracted with a 1:1 mixture of methanol to water. The authors refer to the ability of the method not only to detect adulterants (soybean oil in olive oil) but also the aging of vegetable oils. Using a further development of ESI-MS technique, the ESI Fourier transform ion cyclotron resonance MS (ESI FT-ICR MS), Wu et al. (2004) resolved and identified fatty acids, di- and tri-acylglycerols and tocopherols in canola, olive and soybean oils. One of the advantages of this
SECTION | I Olives and Olive Oil
technique is that it permits the chemical characterization of vegetable oils without any prior sample extraction or separation. The authors have proven the efficiency of this technique in detecting the admixture of soybean oil to olive oil. Trace elements can also be used to determine the geographical origin of olive oil. For example, Benincasa et al. (2007) applied the ICP-MS technique in the quantitative analysis of 18 trace elements within olive oils produced in four different regions of Italy. The results indicated that discrimination might be possible using this approach but databases for each producing zone being studied need to be generated.
12.3.4 Atomic Absorption/Emission Spectrometry (AAS/AES) The most common techniques used in the determination of trace metals within olive oils are ICP-AES and AAS. Zeiner et al. (2005) used both approaches to detect, respectively, Ca, Fe, Mg, Na and Zn, and Al, Co, Cu, K, Mn and Ni. Geographical assignment of olive oils was only possible by using the levels of the latter, present in small amounts within the samples.
12.3.5 Raman Spectroscopy The technique of Raman spectroscopy is based on vibrational transitions occurring in the ground electronic state of molecules. It has been used in the detection of adulterants in olive oils, and does not require laborious sample pretreatments. Heise et al. (2005) applied this technique to the discrimination of vegetable oils and the detection of hazelnut and sunflower oils in VOO. This technique can provide unambiguous results when combined with adequate statistical analysis. Heise and collaborators, using Fourier transform Raman spectroscopy, concluded that sunflower oil could be detected even at a level of 1% by weight. In a similar vein, Baeten et al. (2005) used FT-Raman to detect hazelnut oil in percentages greater than 8%.
12.3.6 Infra-red Spectroscopy Fourier transform infra-red (FT-IR) spectroscopy has emerged as a rapid food analysis tool using minimal sample preparation. FT-IR methods have been used successfully to detect the adulteration of VOO with refined olive oils and other vegetable and nut oils such as sunflower, corn and soybean oil (Allam and Hamed, 2007). Spectroscopic analysis of monovarietal olive oils obtained from three different cultivars also permitted differentiation amongst monovarietal oils and between pure oils and mixtures containing more than one variety (Gurdeniz et al., 2007).
CHAPTER | 12 Olive Oil Authenticity Evaluation by Chemical and Biological Methodologies
The potential of Near-Infrared (NIR) Spectroscopy to discriminate adulterated VOO was investigated by Downey et al. (2002). They detected sunflower oils at levels as low as 1%. Kasemsumran et al. (2005) obtained good results with low error limits using the same approach in the quantification of VOO adulteration with oils from soya, sunflower, corn, walnut and hazelnut.
12.4 DNA-BASED METHODS With a few exceptions referred to above, the analysis of molecular DNA markers is the only analytical approach that can provide information about the cultivar identity of olive oils. This is mainly due to the environmental effects on the phenotype of plants and the resulting chemical composition. Genetic identity appears in the analytical methodologies applied to VOO authenticity evaluation for varietal composition certification. The major constraint in the use of these molecular methods is obtaining satisfactory DNA extracts. One of the earliest works regarding DNA extraction from olive oils was published by Cresti et al. (1997), in which the nucleic acids were obtained from oil sediments. Since then several works have been published evaluating DNA extraction methods (Breton et al., 2004). Techniques detecting molecular polymorphisms have been used in the varietal authentication of VOO. Approaches include random amplified length polymorphisms (RAPD) (Cresti et al., 1997), amplified fragment length polymorphisms (AFLP) and sequence characterized amplified regions (SCAR) developed from AFLP specific fragments (Pafundo et al., 2007). Although these techniques can provide useful information, the most promising and most utilized approach is the detection of polymorphisms in SSR (simple sequence repeat) loci or microsatellites, due to their high polymorphism and loci specificity (Breton et al., 2004; Pasqualone et al., 2007). Notwithstanding their usefulness, these markers should be used with caution since, as stated by Doveri et al. (2006), non-concordance between the genetic profiles of olive oil and fruits of origin can occur due to the concomitant extraction of DNA from embryos of olive fruits and the consequent presence of alleles originating from cross pollination in olive orchards.
SUMMARY POINTS ●
●
●
Authenticity of olive oil continues to be the target of many studies mainly due to its economic value. Consumers expect olive oil to be a healthy product that conforms to the information on the label. There is a vast amount of legislation regarding the accurate characterization and classification of olive oils. This has boosted the volume of scientific work on this subject.
●
●
●
105
The number of analytical techniques addressing these questions approaches the number of ways of adulterating the product itself. The present chapter aims to summarize the analytical methods utilized so far, rather than being an exhaustive revision of published works. Methods are described divided by the type of analytical methodology used as follows: (i) chromatographic, (ii) spectroscopic/spectrometric and (iii) DNA-based techniques.
ACKNOWLEDGMENTS M.A. Faria and S.C. Cunha are grateful to Fundação para a Ciência e a Tecnologia – FCT, ‘Ciência e Inovação 2010’ for grants SFRH/ BPD/41854/2007 and SFRH/BPD/20725/2004, respectively.
REFERENCES Allam, M.A., Hamed, S.F., 2007. Application of FTIR spectroscopy in the assessment of olive oil adulteration. J. Appl. Sci. Res. 3, 102–108. Alves, M.R., Cunha, S.C., Amaral, J.S., Pereira, J.A., Oliveira, M.B., 2005. Classification of PDO olive oils on the basis of their sterol composition by multivariate analysis. Anal. Chim. Acta. 549, 166–178. Angerosa, F., 2002. Influence of volatile compounds on virgin olive oil quality evaluated by analytical approaches and sensor panels. Eur. J. Lip. Sci. Techn. 104, 639–660. Angerosa, F., Camera, L., Cumitini, S., Gleixner, G., Reniero, F., 1997. Carbon stable isotopes and olive oil adulteration with pomace oil. J. Agric. Food Chem. 45, 3044–3048. Aparicio, R., Aparicio-Ruíz, R., 2000. Authentication of vegetable oils by chromatographic techniques. J. Chromatogr. A 881, 93–104. Aparicio, R., Morales, M.T., Alonso, V., 1997. Authentication of European Virgin Olive Oils by their chemical compounds, sensory attributes, and consumers’ attitudes. J. Agric. Food Chem. 45, 1076–1083. Arvanitoyannis, I.S., Vlachos, A., 2007. Implementation of physicochemical and sensory analysis in conjunction with multivariate analysis towards assessing olive oil authentication/adulteration. Crit. Rev. Food Sci. Nut. 47, 441–498. Baccouri, B., Temime, S.B., Campeol, E., Cioni, P.L., Daoud, D., Zarrouk, M., 2007. Application of solid-phase microextraction to the analysis of volatile compounds in virgin olive oils from five new cultivars. Food Chem. 102, 850–856. Baeten, V., Pierna, J.A.F., Dardenne, P., Meurens, M., Garcia-Gonzalez, D.L., Aparicio-Ruiz, R., 2005. Detection of the presence of hazelnut oil in olive oil by FT-Raman and FT-MIR spectroscopy. J. Agric. Food Chem. 53, 6201–6206. Beltrán, G., Del Rio, C., Sánchez,, S., Martínez, L., 2004. Influence of harvest date and crop yield on the fatty acid composition of virgin olive oils from cv. Picual. J. Agric. Food Chem. 52, 3434–3440. Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gómez-Caravaca, A. M., Segura-Carretero, A., Fernández-Gutiérrez, A., Lercker, G., 2007. Phenolic molecules in virgin olive oils: A survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules 12, 1679–1719. Benincasa, C., Lewis, J., Perri, E., Sindona, G., Tagarelli, A., 2007. Determination of trace element in Italian virgin olive oils and their
106
characterization according to geographical origin by statistical analysis. Anal. Chim. Acta. 585, 366–370. Boskou, D., 2006. Polar phenolic compounds. In: Boskou, D. (ed.), Olive Oil: Chemistry and Technology. AOCS Press, Champaign, pp. 73–92. Boskou, D., Blekas, G., Tsimidou, M., 2005. Phenolic compounds in olive oil and olives. Curr. Top. Nutraceut. Res. 3, 125–136. Breton, C., Claux, D., Metton, I., Skorski, G., Berville, A., 2004. Comparative study of methods for DNA preparation from olive oil samples to identify cultivar SSR alleles in commercial oil samples: Possible forensic applications. J. Agric. Food Chem. 52, 531–537. Casal, S., Oliveira, B., 2006. Fatty acids analysis by gas chromatography (GC). In: Cazes, J. (ed.), Encyclopedia of Chromatography. Taylor & Francis, New York. Catharino, R.R., Haddad, R., Cabrini, L.G., Cunha, I.B.S., Sawaya, A.C.H.F., Eberlin, M.N., 2005. Characterization of vegetable oils by electrospray ionization mass spectrometry fingerprinting: Classification, quality, adulteration, and aging. Anal. Chem. 77, 7429–7433. Cercaci, L., Rodriguez-Estrada, M.T., Lercker, G., 2003. Solid-phase extraction-thin-layer chromatography-gas chromatography method for the detection of hazelnut oil in olive oils by determination of esterified sterols. J. Chromat. A 985, 211–220. Cresti, M., Linskens, H.F., Mulcahy, D.L., Bush, S., di Stilio, V., Xu, M.Y., Vignani, R., Cimato, A., 1997. Preliminary communication about the identification of DNA in leaves and in olive oil of Olea europaea. Olivae 69, 36–37. Cunha, S.C., Oliveira, M.B.P.P., 2006. Discrimination of vegetable oils by triacylglycerols evaluation of profile using HPLC/ELSD. Food Chem. 95, 518–524. Cunha, S.C., Amaral, J.S., Fernandes, J.O., Oliveira, M.B.P.P., 2006b. Quantification of tocopherols and tocotrienols in Portuguese olive oils using HPLC with three different detection systems. J. Agric. Food Chem. 54, 3351–3356. Cunha, S.C., Casal, S., Oliveira, M.B.P.P., 2005. Triacylglycerol profile by HPLC/ELSD as a discriminant parameter of varietal olive oils from Portugal. Ital. J. Food Sci. 17, 447–454. Cunha, S.S., Fernandes, J.O., Oliveira, M.B.P.P., 2006a. Quantification of free and esterified sterols in Portuguese olive oils by solid-phase extraction and gas chromatography-mass spectrometry. J. Chromat. A 1128, 220–227. Damirchi, S.A., Savage, G.P., Dutta, P.C., 2005. Sterol fractions in hazelnut and virgin olive oils and 4,4⬘-dimethylsterols as possible markers for detection of adulteration of virgin olive oil. JAOCS 82, 717–725. Doveri, S., O’Sullivan, D.M., Lee, D., 2006. Non-concordance between genetic profiles of olive oil and fruit: a cautionary note to the use of DNA markers for provenance testing. J. Agric. Food Chem. 54, 9221–9226. Downey, G., McIntyre, P., Davies, A.N., 2002. Detecting and quantifying sunflower oil adulteration in extra virgin olive oils from the Eastern Mediterranean by visible and near-infrared spectroscopy. J. Agric. Food Chem. 50, 5520–5525. Flores, G., Ruiz Del Castillo, M.L., Blanch, G.P., Herraiz, M., 2006. Detection of the adulteration of olive oils by solid phase microextraction and multidimensional gas chromatography. Food Chem. 97, 336–342. Fragaki, G., Spyros, A., Siragakis, G., Salivaras, E., Dais, P., 2005. Detection of extra virgin olive oil adulteration with lampante olive oil and refined olive oil using nuclear magnetic resonance spectroscopy and multivariate statistical analysis. J. Agric. Food Chem. 53, 2810–2816. Gamazo-Vázquez, J., García-Falcón, M.S., Simal-Gándara, J., 2003. Control of contamination of olive oil by sunflower seed oil in bottling
SECTION | I Olives and Olive Oil
plants by GC-MS of fatty acid methyl esters. Food Control 14, 463–467. Garcia-Gonzalez, D.L., Mannina, L., D’Imperio, M., Segre, A.L., Aparicio, R., 2004. Using H-1 and C-13 NMR techniques and artificial neural networks to detect the adulteration of olive oil with hazelnut oil. Eur. Food Res. Technol. 219, 545–548. Goodacre, R., Vaidyanathan, S., Bianchi, G., Kell, D.B., 2002. Metabolic profiling using direct infusion electrospray ionisation mass spectrometry for the characterisation of olive oils. Analyst 127, 1457–1462. Guimet, F., Ferre, J., Boque, R., 2005. Rapid detection of olive-pomace oil adulteration in extra virgin olive oils from the protected denomination of origin “Siurana” using excitation-emission fluorescence spectroscopy and three-way methods of analysis. Anal. Chim. Acta. 544, 143–152. Gurdeniz, G., Tokatli, F., Ozen, B., 2007. Differentiation of mixtures of monovarietal olive oils by mid-infrared spectroscopy and chemometrics. Eur. Food Res. Technol. 109, 1194–1202. Hajimahmoodi, M., Vander Heyden, Y., Sadeghi, N., Jannat, B., Oveisi, M.R., Shahbazian, S., 2005. Gas-chromatographic fatty-acid fingerprints and partial least squares modeling as a basis for the simultaneous determination of edible oil mixtures. Talanta 66, 1108–1116. Heise, H.M., Damm, U., Lampen, P., Davies, A.N., McIntyre, P.S., 2005. Spectral variable selection for partial least squares calibration applied to authentication and quantification of extra virgin olive oils using Fourier transform Raman spectroscopy. Appl. Spectrosc. 59, 1286–1294. Kasemsumran, S., Kang, N., Christy, A., Ozaki, Y., 2005. Partial least squares processing of near-infrared spectra for discrimination and quantification of adulterated olive oils. Spectrosc. Lett. 38, 839–851. Lercker, G., Rodriguez-Estrada, M.T., 2000. Chromatographic analysis of unsaponifiable compounds of olive oils and fat-containing foods. J. Chromat. A 881, 105–129. Matos, L.C., Cunha, S.C., Amaral, J.S., Pereira, J.A., Andrade, P.B., Seabra, R.M., Oliveira, B.P.P., 2007. Chemometric characterization of three varietal olive oils (Cvs. Cobrançosa, Madural and Verdeal Transmontana) extracted from olives with different maturation indices. Food Chem. 102, 406–414. Morales, M.T., León-Camacho, M., 2000. Gas and liquid chromatography: Methodology applied to olive oil. In: Harwood, J., Aparicio, R. (eds) Handbook of Olive Oil: Analysis and Properties. Aspen Publishers, Inc., Gaithersburg, MD, USA, pp. 159–208. Ogrinc, N., Kosir, I.J., Spangenberg, J.E., Kidric, J., 2003. The application of NMR and MS methods for detection of adulteration of wine, fruit juices, and olive oil. A review. Anal. Bioanal. Chem. 376, 424–430. Pafundo, S., Agrimonti, C., Maestri, E., Marmiroli, N., 2007. Applicability of SCAR markers to food genomics: olive oil traceability. J. Agric. Food Chem. 55, 6052–6059. Pasqualone, A., Montemurro, C., Summo, C., Sabetta, W., Caponio, F., Blanco, A., 2007. Effectiveness of microsatellite DNA markers in checking the identity of protected designation of origin extra virgin olive oil. J. Agric. Food Chem. 55, 3857–3862. Pena, F., Cardenas, S., Gallego, M., Valcarcel, M., 2005. Direct olive oil authentication: Detection of adulteration of olive oil with hazelnut oil by direct coupling of headspace and mass spectrometry, and multivariate regression techniques. J. Chromatogr. A. 1074, 215–221. Saba, A., Mazzini, F., Raffaelli, A., Mattei, A., Salvador, P., 2005. Identification of 9(E),11(E)-18:2 Fatty acid methyl ester at trace level in thermal stressed olive oils by GC coupled to acetonitrile CI-MS and CI-MS/MS, a possible marker for adulteration by addition of deodorized olive oil. J. Agric. Food Chem. 53, 4867–4872.
CHAPTER | 12 Olive Oil Authenticity Evaluation by Chemical and Biological Methodologies
Sacchi, R., 2001. In ‘Magnetic Resonance in Food Science’, Webb, G.A., Belton, P.S., Gil, M.A., Delgadillo, I. (eds), Royal Society of Chemistry, Cambridge, p. 213. Sayago, A., Garcia-Gonzalez, D.L., Morales, M.T., Aparicio, R., 2007. Detection of the presence of refined hazelnut oil in refined olive oil by fluorescence spectroscopy. J. Agric. Food Chem. 55, 2068–2071. Vigli, G., Philippidis, A., Spyros, A., Dais, P., 2003. Classification of edible oils by employing P-31 and H-1 NMR spectroscopy in combination with multivariate statistical analysis. A proposal for the detection of seed oil adulteration in virgin olive oils. J. Agric. Food Chem. 51, 5715–5722.
107
Wu, Z.G., Rodgers, R.P., Marshall, A.G., 2004. Characterization of vegetable oils: Detailed compositional fingerprints derived from electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry. J. Agric. Food Chem. 52, 5322–5328. Zamora, R., Navarro, J.L., Hidalgo, F.J., 1994. Identification and classification of olive oils by high-resolution C-13 nuclear magnetic resonance. J. Am. Oil Chem. Soc. 71, 361–364. Zeiner, M., Steffan, I., Cindric, I.J., 2005. Determination of trace elements in olive oil by ICP-AES and ETA-AAS: A pilot study on the geographical characterization. Microchem. J. 81, 171–176.
This page intentionally left blank
Chapter 13
Ripening of Table Olives: Use of Magnetic Resonance Imaging (MRI) Maria Antonietta Brescia and Antonio Sacco Dipartimento di Chimica,Campus Universitario, Università di Bari, Italy
13.1 INTRODUCTION Nuclear magnetic resonance (NMR) analysis techniques are considered the most advantageous in the investigations on food products. In fact, the techniques are non-prejudicial, that is target molecules or structures do not need to be preselected for analysis: all relevant nuclei are affected by the experiment and can be detected. Generally, ‘high resolution’ NMR is applied to liquid food (oil, wine, milk, beverages, etc.), whereas in the case of solid matrices (wheat, cheese, meat, etc.), in which the molecular rigidity is the key characteristic, ‘solid state’ NMR can be applied. If the environment prevents the isotropic reorientation of the mobile molecules, causing the spectra to appear very broad and therefore of no diagnostic value, magic angle spinning (MAS) can be used for averaging such anisotropies to zero. Such an approach has been widely used recently due to the availability of commercial 1H HR-MAS probe and has been applied to solid food products (Shintu et al., 2004; Sacco et al., 2005; Brescia et al., 2007). In the last ten years, interest in the application of magnetic resonance imaging (MRI) to food products has grown, since the technique can provide useful insights into many structural aspects of both raw and processed foods.
13.2 USE AND ADVANTAGES OF MRI TECHNIQUE MRI is a technique for looking inside the body noninvasively, which is widely used in medical diagnosis from images. It affords anatomic images in multiple planes and may provide information on tissue characterization. The MR images are obtained by placing the area of interest (body or object) within a powerful static magnetic field. Magnetized protons (hydrogen nuclei) within the area align like small magnets in this field. Radiofrequency pulses are then utilized to create an oscillating magnetic field perpendicular to the main field, from which the nuclei absorb Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
energy and move out of alignment with the static field, in a state of excitation. As the nuclei return from the excitation to the equilibrium state, a signal induced in the receiver coil of the instrument by the nuclear magnetization can be transformed by a series of algorithms into diagnostic images. Images based on different tissue characteristics can be obtained by varying the number and sequence of pulsed radiofrequency fields in order to take advantage of magnetic relaxation properties of tissues. Several researchers have attempted to bring the numerous benefits of medical MRI digital image analysis into the area of food products. In food science MRI techniques allow the interior of foods to be imaged non-invasively and non-destructively. These images can then be quantified to yield information about several processes and material properties, such as mass and heat transfer, fat and ice crystallization, gelation, water mobility, composition and volume changes, food stability and maturation, flow behavior, and temperature. In particular, water and fat content distribution is one of the applications that is widely found in food-related literature. Two reasons can be proposed. Firstly, the gravimetric technique for water determination and the chemical extraction method for fat determination are known to be time consuming, and this limits their use for the determination of spatial distribution in the food matrix. Secondly, water content is one of the most relevant factors to the quality of food and its distribution is rarely uniform in a product. This therefore explains why considerable efforts have been made to devise effective techniques to quantify the spatial distribution of fat and water non-destructively.
13.3 AIM OF THE WORK AND STATE OF THE ART We present here routine 1H HR-MAS NMR and MRI analysis on the structural modifications that take place during the ripening of Bella della Daunia table olives
109
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
110
SECTION | I The Detailed Characterization of Olives and Olive Products
produced in a restricted geographical area of Apulia in Southern Italy. Bella della Daunia table olives were awarded Protected Designation of Origin (PDO) status in 2000 (European Communities, 2000), in accordance with community regulation 2081/92. PDO is awarded to typical agricultural foods whose characteristic features are fundamentally or exclusively due to geographical environment, including both natural and human factors; besides, production, transformation and processing must occur in the place of origin according to the corresponding production disciplinary (European Communities, 1999). Bella della Daunia table olives are produced in a restricted area in the province of Foggia, Apulia (Southern Italy). Environmental factors, together with the variety and the methods of cultivation, harvesting and processing, define the typical and organoleptic features of the olive, regarded as one of the best cultivars for the production of table olives. The purpose of this study was to investigate the changes of some compounds found in Bella della Daunia olives during ripening and after processing. To this purpose, the amount of oil in the pulp, the moisture and reducing sugars content, the composition of fatty acids were determined. While in literature there are extensive data on the composition of olives during ripening (Nergiz and Engez, 2000; Finotti et al., 2001; Rial and Falqué, 2003), there is no information about structural distribution of oil and water in the olives’ tissue, and on its changes due to ripening and processing. MRI has allowed to monitor non-invasively the quality and ripening-related processes in agricultural products (Gussoni et al., 1993; Jagannathan et al., 1995; Clark et al., 1997; Joyce et al., 2002). In plant systems with a high oil content as olive, the physical, chemical and physiological properties are determined by water and triacylglycerols which are present as major constituents. Since olive fruit has a high oil content with a resonance signal from oil protons that resonates about 3.5 ppm away from water, chemical shift selective MRI experiments could be carried out in order to identify oil and water distribution separately and to estimate their variation as ripening progressed. The objective of combining traditional, NMR and MRI analyses is to obtain a complete chemical and structural study of the olive ripening.
13.4 DESCRIPTION OF THE EXPERIMENTAL WORK Sufficient amounts of olives were hand-picked from all sides of one olive tree from October 2002 to January 2003. The samples analyzed are reported in Table 13.1. Sample 6 was made up of pickled green olives processed with Siviglian method, while sample 7 was made up of pickled black olives processed using the Californian method. The samples were immersed in liquid nitrogen, and stored at ⫺80°C prior to analysis.
TABLE 13.1 List of the analyzed samples. Sample number
Picking date
1
30/10/02
2
11/11/02
3
27/11/02
4
18/12/02
5
13/01/03
6
Green pickled olives
7
Black pickled olives
Adapted with permission of Food Chemistry.
The samples were analyzed for moisture, oil, fatty acids and reducing sugars determination and investigated with 1 H HR-MAS NMR and MRI techniques. Moisture content was determined by measuring the weight difference between 10 g of homogenized olives dried in a stove at 105°C for 5 h. Oil content was determined by Soxhlet extraction, using petroleum ether on homogenized olives dried in an oven at 50°C. Fatty acid composition was determined by gas chromatography (GC), using previously described procedures and instrumentation (Sacco et al., 2000). Reducing sugars were twice extracted from 1 g of homogenized freezedried olives with 20 ml of water – ethanol solution (1:1) for 30 minutes. The extracts were filtered using a funnel with fritted glass filter support and evaporated at 40°C to eliminate the ethanol. The solution was brought to 100 ml with water. This solution was used to titrate the Fehling solution (1 ml of reagent A: copper sulfate, 1 ml of reagent B: sodium and potassium tartrate and NaOH, 15 ml of water), using methylene violet as indicator. Samples for 1H HR-MAS measurements were prepared by mixing 40 mg of the freeze-dried olives with 33 mg of D2O. Spectra were acquired at 300 K on an AVANCE 500 MHz (Bruker Analytik GmbH, Rheinstetten, Germany) spectrometer equipped with a high-resolution HR-MAS probehead suitable for 4 mm rotors. The following experimental conditions were applied: spectral width ⫽ 10 ppm; time domain ⫽ 16384 points; number of transients ⫽ 128 and spinning rate 4200 Hz. Spectra were acquired by using the mono-dimensional version of NOESY sequence, with water signal suppression by presaturation and were processed by applying a 0.3 Hz line broadening factor. The imaging experiments were performed at 4.7 T (200 MHz for 1H) on a Bruker Avance 200SWB NMR spectrometer operated with PARAVISION. The Micro2.5 gradient system has an inner diameter of 40 mm and produces maximum gradients of 1 T/m. The radio-frequency birdcage resonator has an inner diameter of 25 mm, restricting the sample diameter to about 23 mm. A spin-echo imaging
111
CHAPTER | 13 Ripening of Table Olives: Use of Magnetic Resonance Imaging (MRI)
sequence, with a chemical shift selective excitation pulse and a slice selective refocusing pulse, was used to obtain selective images. The field of view of 25 mm ⫻ 25 mm was divided into 128 ⫻ 128 voxels; the repetition time was 2 s and the echo time 8.2 ms. Images of the three olives were obtained using the Mini0.36 gradient system with an internal diameter of 85 mm and maximum gradients of 0.14 T/m. The corresponding birdcage resonator has an inner diameter of 64 mm.
13.5 RESULTS OF ANALYTICAL DETERMINATIONS A detailed discussion of the experimental results is here presented. Figure 13.1 shows moisture, oil and reducing sugars content for the analyzed samples. The decreasing moisture content during ripening is typical of olives of other cultivars (Nergiz et al., 2000). The moisture content is due to biological development, but also to human and climatic factors, such as irrigation, rainfall and temperature. The high moisture content of sample 5 compared to sample 4 can probably be attributed to picking after rainfall or irrigation. The amount of oil increased over the course of the sampling period, while reducing sugars content decreased, except for the last sample. A relationship was found between the oil accumulation and sugar content. This correlation can be explained by considering that sugars are precursors of fatty acid biosynthesis in the olive fruit (Wodner et al., 1988). Water contents in samples 6 and 7 were higher than those in the other samples probably because of the washings they undergo during processing; in fact the osmotic process
13.6 RESULTS OF 1H HR-MAS NMR ANALYSIS The assignments of the signals resonating in the 1H HR-MAS NMR spectrum (Figure 13.2) were performed according to the literature (Sacco et al., 1998). The spectrum contains strong resonances from olive fatty acids in the spectral zone between 0.5 and 3 ppm. These signals are due to terminal methyl hydrogens (0.9 ppm); methylene protons (1.3 ppm) of aliphatic chains of fatty acids; methylene protons β to carboxyl glycerides (1.5 ppm); diallylic CH2 hydrogens of the linoleic acid (2 ppm) and methylene protons α to carboxyl glycerides (2.2 ppm). The resonances between 3 and 4 ppm are attributable to sugar protons. Due to the high overlapping of signals in this area, it was difficult to make an assignment. Triplets at 3.22 and 3.51 ppm are respectively due to H2 of β-glucose and α-glucose. Peaks from 4 ppm to 4.5 ppm are due to glycerol CH and CH2, signals at 4.6 and 5.2 ppm are due to the anomeric protons of β-glucose and α-glucose respectively,
Oil content Reducing sugars content
75
60
74
50
73
40
72
30
71
20
70
10
69
30/10/02
11/11/02
27/11/02 18/12/02 Time (date)
13/01/03
0
FIGURE 13.1 Moisture, oil, and reducing sugars content of the analyzed olives (with permission of Food Chemistry).
% W/W oil and reducing sugars contents
% Water content
Water content
raises the water content of olive tissue (Marsilio et al., 2001; Bianchi, 2003). The oil content of these samples was unchanged, while reducing sugars content decreased, probably due to being solubilized in washing water. Table 13.2 reports the percentage content of fatty acids. The most common fatty acids in these olives were oleic, palmitic, linoleic, stearic and behenic acid. Fatty acid content does not present any particular trend with ripening; however it is noticeable that, with ripening, the oil fraction sensibly enriches in longer-chain fatty acids. Samples 6 and 7 have the same fatty acid composition as the other samples; showing that processing does not influence oil composition.
112
SECTION | I The Detailed Characterization of Olives and Olive Products
TABLE 13.2 Fatty acids (%) of oil extracted from the analyzed olives. Fatty acids C14:0 C16:0
Sample 1
Sample 2
0
Sample 3
0
17.06
0
18.83
16.31
Sample 4
Sample 5
0.28
0.22
18.85
19.40
Sample 6
Sample 7
0
1.04
15.46
18.28
C16:1
0.96
0.93
0.56
0.57
0.64
0.73
0.33
C17:1
0
0.21
0
6.63
0
0
0
C18:0
3.60
3.95
4.09
4.25
5.10
3.48
3.87
60.75
62.36
54.94
57.84
68.23
54.51
9.22
8.37
7.22
8.86
9.44
6.20
C18:1 C18:2
62.3 7.81
C20:0
0
0.90
0
0.22
0
0
0
C18:3
0.65
0.47
0.61
0.60
0.54
0
0.87
C20:1
1.09
0.42
0.78
0.51
0.77
0.79
0
C22:0
3.01
2.10
3.68
5.01
3.93
1.08
3.56
C24:1
1.86
0.98
2.23
0
1.13
0.77
1.63
C22:6
0.60
0.38
0.47
0.47
0.47
0
9.29
C14:0 (myristic acid); C16:0 (palmitic acid); C16:1 (palmitoleic acid); C17:1 (heptadecenoic acid); C18:0 (stearic acid); C18:1 (oleic acid); C18:2 (linoleic acid); C20:0 (arachidic acid); C18:3 (linolenic acid); C20:1 (eicosenoic acid); C22:0 (behenic acid); C24:1 (nervonic acid); C22:6 (docosahexaenoic acid). Adapted with permission of Food Chemistry.
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
ppm FIGURE 13.2 1H HR-MAS NMR spectrum of a freeze-dried olive.
while resonance at 5.3 ppm is due to olefinic protons of unsaturated fatty acids (principally oleic and linoleic). 1 H HR-MAS NMR spectra were obtained for all the olive samples and from their comparison (figure not shown) an increase of all the fatty acids signals was observed with maturation, while a decrease of sugar signals was observed, in agreement with the results of routine analyses. In the spectra of processed olives the intensities of carbohydrate signals were very small.
13.7 STRUCTURE OF OLIVES The olive is structurally constituted of different parts: epicarp, mesocarp and endocarp (Figure 13.3). The middle layer of the drupe is known as the mesocarp, or pulp, and makes up 70–80% of its weight. The pulp, or flesh, is made up of lacunar parenchyma cells which act as a store for its constituents: water, polysaccharides – which determine the texture of the olive – proteins and oil.
113
CHAPTER | 13 Ripening of Table Olives: Use of Magnetic Resonance Imaging (MRI)
13.8 DISCUSSION OF MRI RESULTS
Epicarp
Mesocarp
Endocarp Germ (tegument, albumen, embryo)
FIGURE 13.3 Olive section (with permission of Food Chemistry).
Oil is found in 3–5 μm diameter drops (the diameter varies according to temperature, season and cultivar), which are located in the intercellular spaces and in quite large roundish cells with very thin and highly elastic walls; oil takes up nearly all of the cell, even pushing the nucleus out towards the cell walls. The inner layer is made up of the endocarp, the outside of the pit, which makes up 22% of the weight of the drupe. This stony pit is formed by lignified parenchyma cells, which gradually harden completely. Inside this is the seed, which makes up 2–4% of the weight of the drupe. It is made up of an external membrane (tegument or episperm), a whitish fleshy part (albumen or endosperm) and by the embryo. The seed contains a considerable amount of oil (22–27%), whereas the endocarp only contains about 1%. Before carrying out the MRI analysis a mono-dimensional NMR spectrum of olive sample was carried out. The spectrum contains strong and well-resolved resonances due to water and fat; the water protons signal resonates at 5.3 ppm, while the signal arising from oil methylene protons resonates at 1.3 ppm.
CIW-NMR olives27Nov02 as_030417 “ F 5.0 kg
A
BRUKER spect Date: 17 Apr 2003 Time: 10:13
In MRI images clearer and darker areas are visible. The clearer areas indicate a higher concentration of the investigated type of proton. Chemical shift selective images were obtained setting the resonances. In order to have a resolution test and to illustrate our approach to chemical shift imaging when two liquids are present in the core, a phantom of water and olive oil was used. Figure 13.4 shows respectively, chemical shift selective images of water protons (A) and oil protons (B) of one olive forming sample 3. These images were taken using two NMR tubes containing respectively the water phantom made of 0.688 g of a mixture of water containing a gadolinium salt and 0.696 g of D2O and the olive oil phantom made of a 1:5 mixture of olive oil in chloroform. Oil and water were diluted to prevent obtaining intense signals compared with those of the olive. It can be seen that selective excitations works well, because water and oil signals are separated in the two images. The water-selective image shows that the intensity of the reference phantom tube, visible as a circle upside the olive, is less intense than the water signal due to the olive. This is due to the different concentration of water in the reference (around 50%) and in the olive (about 70%). The chemical shift selective MRI images of samples 2, 3 and 5 are shown in Figure 13.5 (sample 5 had to be slightly cut in order to fit into the resonator). Figure 13.5A shows water presence in the pulp near the epicarp, but also around the endocarp, with a radial crown distribution. The signal intensities show that the water quantities near the epicarp and the endocarp are comparable. Water is also present in the seed, but in a minor quantity. A detailed analysis of Figure 13.5A shows that there is more water in the tegument and in the embryo than in the albumen. The oil selective image (Figure 13.5B) presents a less evident distribution. Oil is located principally around the
CIW-NMR olives27Nov02 as_030417 “ F 5.0 kg
cm
1
L
W 3128 L 1693
0 Scan: 4 Slice;1/1 Echo: 1/20 m_chess(pvm) TR: 2000.0 ms TE:7.6 ms NEX 4
A
BRUKER spect Date: 17 Apr 2003 Time: 11:01
cm
1 W 3951 L 2121
A
SI 1.0/1.0 mm FOV 3.2 cm MTX 128 Pos −4.0 mm F olives27Nov02 + Oil/CHCl3 + H20/D20: 1 4:1
L
0 Scan: 5 Slice;1/1 Echo: 1/32 m_chess(pvm) TR: 750.0 ms TE:7.6 ms NEX 64
SI 1.0/1.0 mm FOV 3.2 cm MTX 128 Pos −4.0 mm F olives27Nov02 + Oil/CHCl3 + H20/D20: 1 5:1
B
FIGURE 13.4 Axial images of sample 3 with water and oil references: (A) water-selective image; (B) oil-selective image.
114
SECTION | I The Detailed Characterization of Olives and Olive Products
FIGURE 13.6 Axial images of processed olives: (A) water-selective image and (B) oil-selective image of green pickled olives (sample 6); (C) water-selective image and (D) oil-selective image of black pickled olives (sample 7) (with permission of Food Chemistry).
FIGURE 13.5 Axial images of olives: (A) water-selective image and (B) oil-selective image of sample 2; (C) water-selective image and (D) oil-selective image of sample 3; (E) water-selective image and (F) oilselective image of sample 5 (with permission of Food Chemistry).
endocarp and becomes less abundant towards the mesocarp, again with a radial crown distribution. The seed presents a well-defined oil distribution, with oil appearing to be concentrated in the center of the embryo. Moreover, the intensity of the oil signal in the seed is higher than that in the pulp. Figure 13.5C shows the distribution of water in the olive in sample 3. It is comparable to that observed in sample 2, but the radiant crown distribution around the endocarp is not evident. Water is concentrated in the outer pulp. This could indicate the water trend with ripening. The oil distribution observed in Figure 13.5D is analogous to that of sample 2. From a comparison of the signal intensities, it seems evident that oil is concentrated near the endocarp. In this image, the germ signal is absent, which is surprising, since it is known that the germ contains a certain quantity of oil. This result has been investigated more accurately in further experiments and will be discussed later. The characteristics of Figures 13.5E and 13.5F are analogous to those of sample 3, except for the fact that the seed is evident and assumes the same structural characteristics of sample 2. Moreover, Figure 13.5E has a sufficiently high resolution to show that the distribution of water in the embryo is not uniform. Intensity comparison shows that the water content in the seed from sample 5 is lower than that in the one from sample 2, while oil content increases and is more homogeneously distributed in the pulp, in agreement with previous studies (Gussoni et al., 1993). This trend is consistent with the observation that water content, both in the seed and in the pulp, is reduced.
The images of the processed products are shown in Figure 13.6. In the water-selective Figure 13.6A the water signal is almost completely located in the entire mesocarp region and in a thin layer around the epicarp, while its intensity decreases towards the endocarp. Particularly, in this image, there are alternating bright and dark areas close to the epicarp. These areas are close to the rods of the birdcage resonator and show artefacts due to inhomogeneities in the radio-frequency field. The seed structure is not evident but there is a water-rich region in its lower right corner. In the oil-selective image, the signal intensity is higher around the endocarp and decreases towards the flesh. The distribution of oil in the seed is the same as in previous samples.
13.9 EVOLUTION OF WATER AND OIL DISTRIBUTION IN OLIVE WITH RIPENING These images introduce a few considerations. First, in raw olives, water and oil are both concentrated around the endocarp. A tentative explanation for this behavior, assuming that there is a high porosity region around the endocarp, is that both water and fat will concentrate in this area, resulting in an undifferentiated distribution of these two components in the pulp. During ripening, water seems to move from the inner mesocarp towards the outer mesocarp. To explain this phenomenon some authors (Joyce et al., 2002) considered that, with ripening an increase in permeability or the rupture of a cell membrane could lead to the displacement of air. For this reason, signals due to water protons that were previously adjacent to spaces occupied by air are no longer dephased and give a stronger signal. One of the most relevant questions emerging from this study regards the absence of the seed signal in the oil-selective
CHAPTER | 13 Ripening of Table Olives: Use of Magnetic Resonance Imaging (MRI)
115
FIGURE 13.7 Axial images of three olives: (A) water-selective image and (B) oil-selective image of sample 3 (with permission of Food Chemistry).
image. This was observed on one olive picked on 27/11/02 (Figure 13.5D) and one on 13/01/02 (images not shown). This is not easily explainable, since it is known that the seed contains water and oil at the same time. The signal was also not visible using volume-selective spectroscopy. To finally ascertain this result, a simple one-pulse spectrum from the entire isolated kernel was recorded, and still there was no signal about 3.5 ppm downfield from water. For this reason, it was supposed that the oil components of the seed are supported on woody constituents of the seed, thus losing mobility and relaxing too fast to give narrow signals. In contrast with previously reported data (Gussoni et al., 1993), this phenomenon does not seem to be dependent on ripening. In order to study the differences between fruits picked on the same day, and to evaluate whether the observations performed on the samples can be generalized, three olives picked on 27/11/02 were inserted into a gradient system with a 64 mm internal probe diameter. The images obtained are shown in Figure 13.7. Although there are no evident differences in the waterselective image, in the oil-selective image, the top olive does not show a band of high intensity around the seed that is present in the other two olives. This leads to the conclusion that ripening in the different fruits can occur at different times and the factor that influences this behavior may be a different position on the tree. This suggests that it is not possible to generalize our considerations, unless MRI investigations are carried out on a larger number of different fruits picked on the same day. Also evident in Figure 13.7 is a darkened region around the air–olives interface in the center of the picture, again showing a susceptibility artefact. This artefact will be studied carefully when quantitative analysis is performed.
13.10 QUANTITATIVE ANALYSIS ON MRI IMAGES Since the selective chemical shift images are acquired using different experimental parameters, quantitative analysis can be performed using references of known concentration. The following image was taken with the same standards and olive sample used for Figure 13.3. The average intensity of
FIGURE 13.8 Axial water-selective image of sample 3 with water reference and selected areas for quantitative analysis.
selected areas were evaluated by the instrument software (Figure 13.8). From the comparison of water signal of the standard (that must be multiplied by two, since water concentration is about 50%) and the two selected areas of the olive, it was possibile to calculate that for the area closer to the seed (ISA_3), water content was equal to 68% while in the other section (ISA_4) it was equal to 82%. The average content was 75%, which is in agreement with the result obtained with routine analyses. The same approach was used to calculate fat content in the oil-selective image.
13.11 FINAL REMARKS The obtained results provide information on the variations that occur in the Bella della Daunia olive with ripening. They confirm the correlation existing between decreases in sugars and water and increases in oil content. Routine analyses carried out on processed olives showed how processing affects components. 1 H HR-MAS NMR experiments confirmed the results of routine analyses giving an interesting insight into the olive composition and showing the compositional changes occurring during the maturation of the fruit. MRI highlighted local movements of macro-components in the pulp during ripening. Evidence was obtained of a wave of apparent high water activity moving out from the inner to the outer mesocarp during the ripening process. The questions raised from this work (absence of the oil signal in some seeds and differences between olives picked on the same day) warrant extended work on a larger number of olives picked in different years in order to show the relationship between these behaviors and the degree of ripening.
116
SECTION | I The Detailed Characterization of Olives and Olive Products
The MRI approach could be used to characterize olives of different cultivars to provide information about the NMR parameters that differentiate the olives and explain characteristics such as pulp consistency and detachment of the pulp from the kernel. The results of such a study should be of value in the agricultural industry for quality control and for characterizing typical agricultural products.
SUMMARY POINTS ●
●
●
●
●
●
MRI can be used to identify and localize the major components within the seeds, the fruit or other plant organs without destroying the sample. Table olives of Bella della Daunia cultivar were picked from one tree at different ripening stages from October 2002 to January 2003. The amount of oil increased during ripening, while reducing sugars content and moisture decreased. MRI images showed that water and oil are concentrated around the olives’ endocarp, due to the high porosity of this region. During ripening water seems to move from the inner mesocarp to outer mesocarp. Ripening in the different fruits can occur at different times, maybe due to a different position on the tree.
REFERENCES Bianchi, G., 2003. Lipids and phenols in table olives. Eur. J. Lipid Sci. Tech. 105, 229–242. Brescia, M.A., Sacco, D., Sgaramella, A., Pasqualone, A., Simeone, R., Peri, G., Sacco, A., 2007. Characterisation of different typical Italian breads by means of traditional, spectroscopic and image analyses. Food Chem. 104, 429–438. Clark, C.J., Hockings, P.D., Joyce, D.C., Mazucco, R.A., 1997. Application of magnetic resonance imaging to pre- and post-harvest studies of fruits and vegetables. Postharvest Biol. Tec. 11, 1–21. European Communities., 1999. Application 2/99, Official Journal of the European Communities C358.
European Communities., 2000. Regulation 1904/00, Official Journal of the European Communities L228. Finotti, E., Beye, C., Nardo, N., Quaglia, C.B., Milin, C., Giacometti, G., 2001. Physico-chemical characteristics of olives and olive oil from two mono-cultivars during various ripening phases. Nahrung/Food 45, 350–352. Gussoni, M., Greco, F., Consonni, R., Molinari, H., Zannoni, G., Bianchi, G., Zetta, L., 1993. Application of NMR microscopy to the histochemistry study of olives (Olea europea L.). Magn. Reson. Imaging 11, 259–268. Jagannathan, N.R., Govindaraju, V., Raghunathan, P., 1995. In vivo magnetic resonance study of the histochemistry of coconut (Cocos nucifera). Magn. Reson. Imaging 13, 885–892. Joyce, D.C., Hockings, P.D., Mazucco, R.A., Shorter, A.J., 2002. 1HNuclear magnetic resonance imaging of ripening “Kensington pride” mango fruit. Funct. Plant Biol. 29, 873–879. Marsilio, V., Campestre, C., Lanza, B., 2001. Sugar and polyol compositions of some European olive fruit varieties (Olea europea L.) suitable for table olive purposes. Food Chem. 74, 55–60. Nergiz, C., Engez, Y., 2000. Compositional variation of olive fruit during ripening. Food Chem. 69, 55–59. Rial, D.J., Falqué, E., 2003. Characteristics of olive fruits and extra-virgin olive oils obtained from olive trees growing in Appellation of Controlled Origin ‘Sierra Mágina’. J. Sci. Food Agr. 83, 912–919. Sacco, A., Neri Bolsi, I., Massini, R., Spraul, M., Humpfer, E., Ghelli, S., 1998. Preliminary investigation on the characterization of durum wheat flours coming from some areas of south Italy by means of 1H high-resolution magic angle spinning nuclear magnetic resonance. J. Agric. Food Chem. 46, 4242–4249. Sacco, A., Brescia, M.A., Liuzzi, V., Remiero, F., Guillou, C., Ghelli, S., Van der Meer, P., 2000. Characterization of the geographical origin and variety of Italian extra virgin olive oils based on analytical and NMR determinations. J. Am. Oil Chem. Soc. 77, 619–625. Sacco, D., Brescia, M.A., Buccolieri, A., Caputi Jambrenghi, A., 2005. Geographical origin and breed discrimination of Apulian lamb meat samples by means of analytical and spectroscopic determinations. Meat Sci. 71, 542–548. Shintu, L., Ziarelli, R., Caldarelli, S., 2004. Is high-resolution magic angle spinning NMR a practical speciation tool for cheese samples? Parmigiano Reggiano as a case study. Magn. Reson. Chem. 42, 396–401. Wodner, M., Lavee, S., Epstein, E., 1988. Identification and seasonal changes of glucose, fructose and mannitol in relation to oil accumulation during fruit development in Olea europaea (L). Sci. Hortic. 36, 47–54.
Chapter 14
NMR and Olive Oils: A Geographical Characterization Luisa Mannina1,2 and Anna Laura Segre2,‡ 1 2
STAAM Department, University of Molise, Campobasso, Italy Institute of Chemical Methodologies of CNR, Research Area of Rome I, Italy
14.1 INTRODUCTION
14.2.1.2
The definition of the geographical origin of extra virgin olive oils is a controversial question: the composition of an extra virgin olive oil is in fact the result of different factors such as pedoclimatic conditions, the genotype and agronomic practices. Therefore, for the careful definition of the geographical origin based on chemical composition, many factors need to be taken into account. Here, a review of our papers regarding the geographical characterization of olive oils by Nuclear Magnetic Resonance (NMR) is reported. It includes the description of the NMR-statistical protocol as well as the more significant results in the characterization of olive oils.
In the case of 13C spectra, a strong dependence of chemical shifts from the concentration has been previously reported (Mannina et al., 2000a). We suggest 100 μL of oil and 600 μL of CDCl3; the spectrum has to be run immediately after the sample preparation.
14.2 PRACTICAL ASPECT OF THE NMR-STATISTICAL PROTOCOL
C Spectrum
14.2.1.3 Instruments The experimental conditions, previously reported, have been tested on a Bruker AVANCE AQS600 spectrometer operating at the proton frequency of 600.13 MHz. However, the same experimental conditions can be used on other 600 and 500 MHz instruments.
14.2.2 The 1H and 13C NMR Spectra
14.2.1 Sample Preparation The NMR technique allows the sample to be analyzed directly without any work-up or extraction procedure.
14.2.1.1 1H Spectrum
The basic information observable in an NMR spectrum is simply a spectral line characterized by its intensity (and sometimes line width) and its spectral position (the spectral frequency). The standard measurable parameters of a given X spectral line are: ●
Olive oils (20 μL) are placed into 5 mm NMR tubes and are dissolved in a mixed solvent (CDCl3/CD3SO2CD3 in the volume ratio 700/20 μL, respectively) with vigorous shaking. The spectrum has to be run immediately after the sample preparation.
●
●
‡
Prof. Anna Laura Segre passed away on 25th April 2008. Some weeks before her decease, Anna Laura and I began to write this chapter together. Therefore, I dedicate this paper to her, my teacher and beloved friend. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
13
●
117
The center of a line: it refers to a standard line and is called chemical shift and it is usually given in ppm (ppm ⫽ X frequency – standard frequency/carrier frequency; all frequencies are given in Hz). The multiplicity for split lines: it derives from neighbor interactions and it is due to the presence of coupling parameters called coupling constants. The signal integral: it is properly measured as the area under the line and reflects accurately the number of nuclei in each environment. Intensity (height) of a given line: it refers to a standard resonance. It can be used rather than signal integral Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
118
●
SECTION | I The Detailed Characterization of Olives and Olive Products
as an input for statistical analytical procedures. In this case, careful attention must be given to the resolution, which must be kept always the same in all samples. When an accurate intensity measurement is necessary, as in the case of olive oils, it is sometimes necessary to perform a careful correction of the baseline. This correction is absolutely necessary in 1H spectra when a large number of scans is performed. In fact, the empty probe itself and the system as a whole can give rise to a large hump under the real spectrum. Suitable mathematical routines for the baseline correction exist directly available in all commercial spectrometers; these routines use polynomial terms which, subtracted from the baseline, act as perfect flatteners without producing any appreciable error; more sophisticated functions usually are not necessary. The line width: it is usually measured at half the height of the signal and contains information on rate processes, including the rates of molecular motion.
●
chosen. It is important to note that, depending on the specific problem, these 27 NMR variables have a different discriminating power; to perform an accurate statistical analysis. In our papers (Mannina et al., 2003a), the NMR data have been treated using a statistical procedure which requires the following analyses: 1. The Box plot analysis to sort out anomalous samples. 2. Analysis of variance (ANOVA) to select the most selective resonances. The discriminant power of the variables has been valued using F and p level parameters. The F value is the ratio between within-group variability and between groups variability. Larger is this ratio, larger is the discriminant
14.2.3 The NMR-Statistical Protocol The NMR protocol for olive oil characterization requires the 1H and 13C NMR spectrum of the olive oil (see Figure 14.1A and 14.1B respectively) to be acquired at high field and the intensity (height) of a few selected normalized resonances to be measured. 1H and 13C NMR techniques allow olive oils to be characterized. The list of our papers is reported in the references. 1H NMR spectrum allows to have information about major and minor olive oil components whereas 13C NMR can provide valuable information about the acyl distribution and the acyl positional distribution of glycerol tri-esters of different oils. The spectral assignment of 1H and 13C spectra has been reported elsewhere (Sacchi et al., 1998; Mannina et al., 1999a). In order to perform an accurate study, it is necessary: ● ●
●
to perform an accurate baseline correction; to perform a signal normalization. The normalization must be done setting a chosen height value to a particular resonance. The choice of this normalizing resonance is not trivial since it must be in a clearly delimitated range, it must be insensitive to composition, acidity, and possibly to defects. A perfect choice is the resonance at 2.25 ppm of CH2 due to methylenic protons bound to C2 normalized to 1000; to select NMR resonances for the statistical analysis. It must be considered that the selected resonance must be in an accessible spectral range, must be due to a component not related with oxidation or decomposition products and should show large intensity variability in different samples. Following the above criteria, ten 13C resonances and 17 proton resonances, all fully assigned to a particular compound (see Table 14.1) have been
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 ppm
A
190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 ppm
B FIGURE 14.1 (A) 1H spectrum of olive oil in a mixed solvent; (B) 13C spectrum of olive oil in CDCl3. The analysis of 1H and 13C NMR spectra together with a suitable statistical analysis allow olive oils to be grouped according to the geographical area and the cultivar.
119
CHAPTER | 14 NMR and Olive Oils: A Geographical Characterization
TABLE 14.1 Assignment and chemical shifts of 1H and 13C resonances selected for the statistical analysis. NMR resonances (27) are used for the statistical characterization of olive oils according to the geographical origin and/or their cultivar. Since the variability of these resonances depend on many factors, it is important to repeat the statistical protocol for any specific problem. Selected resonances
NMR resonances (ppm)
1: Carbonyl signals of sn 1,3 saturated fatty chain
173.27
2: Carbonyl signals of sn 1,3 eicosen-11-oic and vaccenic fatty chains
173.26
3: Carbonyl signals of sn 1,3 oleic fatty chains
173.24
4: Carbonyl signals of sn 1,3 linoleic fatty chains
173.23
5: Carbonyl signals of sn 2 oleic fatty chains
172.83
6: Carbonyl signals of sn 2 linoleic fatty chains
172.82
7: Methyl of palmitic and stearic fatty chains
14.14
8: Methyl of oleic fatty chains
14.13
9: Methyl of eicoseonic and vaccenic fatty chains
14.12
10: Methyl of linoleic fatty chains
14.09
11: Hexanal
9.704
12: Trans 2-Hexenal
9.454
13: Terpene 4
4.885
14: Terpene 3
4.661
15: Terpene 2
4.609
16: Terpene 1
4.541
17: Methylenic protons in α glycerol moiety of sn 1,3 diglycerides
3.988
18: Methylenic protons in α glycerol moiety of sn 1,2 diglycerides
3.636
19: Diallylic protons of linolenic fatty chains
2.746
20: Diallylic protons of linoleic fatty chains
2.710
21: Squalene
1.620
22: Methylenic protons of all unsaturated fatty chains
1.244
23: Methylenic protons of palmitic and stearic fatty chains
1.197
24: Wax
0.978
25: Methyl of linolenic fatty chains
0.910
26: Methyl of linoleic fatty chains
0.843
27: Methyl-18 of β-sitosterol
0.623
power of the corresponding variable. The p level represents a decreasing index of the reliability of a result and, specifically gives the probability of error involved in accepting the result: the lower the p level (⬍0.05) the larger is the probability that the relationship between the variables is reliable. 3. Principal component analysis (PCA) or/and tree clustering analysis (TCA) on the selected resonances without any a priori hypothesis.
The PCA analysis is the mathematical procedure of a data matrix, which allows one to represent the variation present in many variables using a small number of linear combinations (called principal components). The TCA method joins together objects into successively larger clusters, using some measures of distance. All possible classification in a prescribed number of groups can be obtained by cutting the tree at a suitable level. Different distance measures can be used to form clusters. Moreover, a linkage or amalgamation rule is necessary to
120
SECTION | I The Detailed Characterization of Olives and Olive Products
determine when two clusters are sufficiently similar to be linked together. PCA and TCA allow us to see the natural grouping and give information which can be used as a priori hypothesis for further analysis as LDA.
according to the specific problem. For instance a resonance due to a specific compound may be important in the discrimination of extra-European olive oils but may be not relevant in the discrimination of European oils.
1. Linear discriminant analysis (LDA) on selected resonances with an a priori hypothesis regarding the number of groups also suggested by PCA or TCA. The selected variables are used to build up a data matrix and to give rise to a discriminant (canonical) linear equation. The discriminant power of each variable can be evaluated by measuring the value of the Wilks’ lambda factor for the overall model after removing selected variables. The Wilks’ lambda factor ranges from 0.0 (perfect discriminatory power) to 1.0 (no discriminatory power). 2. Reliability of the system. In order to prove the reliability of the system, some randomly selected oils are removed from the data-set and are introduced to system as unknown samples. The model is again calculated and if the selected samples are well classified, the system is stable and can be use for real samples.
14.3 GEOGRAPHICAL CHARACTERIZATION OF OLIVE OILS
Since the variability of the selected resonances is dependent on many different factors such as environment, cultivar, particular defects of the olive oil and the year, it is important to repeat this statistical protocol each time for any specific problem. This means that the ‘correct’ resonances with the high discriminating power must be identified
An important act of legislation, the PDO (Protected Designation of Origin), allows the identification of some Mediterranean extra virgin olive oils with the names of the geographical areas where they are produced. This certification improves the value of Mediterranean olive oils and assures a product of high quality and defined geographical origin. We have carried out many studies on the olive oil characterization obtained using NMR spectroscopy. Here, we report some important results.
14.3.1 The Geographical and Cultivar Effects In order to verify the potentiality of the NMR methodology in the geographical discrimination of olive oils, a systematic study was performed on olive oils (Mannina et al., 2001b) coming from three Tuscany districts, all within 100 km. In two of these districts, Lucca and Arezzo, the same cultivar are present (Frantoio and Leccino cultivar), whereas in a third borderline district between Lazio and Tuscany named Seggiano only monovarietal oils are present
Euclidean distances 35
30
Linkage distace
25
1st level
20 2nd level 15
10
3rd level
5
S8 S7 S4 S13 S12 S10 S9 S6 S3 S15 S14 S11 S5 S2 S1 A9 A11 A14 A4 A7 A3 A8 A2 A13 A12 A6 A5 A10 A1 AR2 AR7 AR13 AR12 AR8 AR10 AR9 AR6 AR11 AR5 AR4 AR3 AR1
0
FIGURE 14.2 Tree clustering analysis (TCA) applied to selected resonances of extra virgin olive oils from three different areas of Tuscany (Italy); Ar ⫽ Arezzo, A ⫽ Lucca, S ⫽ Seggiano. The dendrogram can be cut at different levels: cutting the dendrogram at first level, olive oils from Seggiano are fully separated from the others. Cutting the dendrogram at the second level, olive oils from Lucca and Arezzo are also well distinguished.
121
CHAPTER | 14 NMR and Olive Oils: A Geographical Characterization
7 6
North TUAR
TUS
TUA
LA
GA
5 4 3 Root 3
2 VITERBO
1
RIETI
0 −1
South
−2 ROMA
−3 −4 −5 −6 −20
FROSINONE −15
−10
0
−5
5
10
15
Root 1
FIGURE 14.3 LDA of extra virgin olive oils coming from Lake Garda (䊉, GA), Tuscany-Arezzo (䊊, TUAR), Tuscany-Lucca (䉫, TUA), Tuscany-Seggiano (䊐, TUS) and Lazio (䉭, LA). LDA map shows that olive oils from different Italian areas can be well separated. Ellipses represent the 95% confidence regions for each group.
(cultivar Seggianese). The NMR-statistical protocol has been applied to these olive oils. The TCA results are reported as a dendrogram in Figure 14.2. The dendrogram can be cut to different levels. Cutting the dendrogram at a high level, olive oils from Seggiano, coming from a specific cultivar grown in a well-defined area, are well grouped together and separated from all the other samples. Therefore, samples characterized from particular cultivar and territory parameters are fully separated from the other samples which differ only for their geographical origin. However, if the dendrogram is cut to a second level even the samples which differ only for the geographical origin are fully separated. Therefore, if particular cultivars are present in a particular environment we will get a strong effect resulting in a full separation. If only the pedoclimatic effects are present we still are able to separate oils coming from different geographical areas, even when these regions are contiguous. On this regard the NMR capability to distinguish oils on the basis of their geographical origin seems unique. The NMR-statistical protocol has also been applied to olive oils coming from different areas of Italy (Mannina et al., 2001a), that is Tuscany (Lucca, Arezzo and Seggiano area), Lazio and Garda Lake. Again, a classification of olive oils coming from different geographic areas has been obtained by LDA analysis as shown in Figure 14.3.
14.3.2 The Case of Olive Oils from Lazio (D’Imperio et al., 2007a) 14.3.2.1 Geographical Characterization ‘Lazio’ is an Italian region, see Figure 14.4, which can be divided into three well-defined geographical areas, namely,
Center LATINA
FIGURE 14.4 Map of Lazio region. The Lazio region can be divided into three geographical areas (South, Center and North) and the five provinces (Viterbo, Rieti, Roma, Frosinone and Latina).
the Northern area, the Center, and the Southern area, characterized by slightly different pedoclimatic conditions. For instance, the Southern area is characterized by a weather wetter than the other two areas. In this study, 72 multivarietal olive oils, namely 11 samples from the Northern area, 41 samples from the Center and 20 samples from the Southern area have been submitted to the NMR-statistical protocol. The normalized intensities of the 27 selected NMR resonances were submitted to a suitable statistical analysis. Some samples were removed as anomalous by the Box plot analysis. The PCA map, labeled according to the three geographical areas of Lazio, is shown in Figure 14.5. The PC1 and PC2 scores, explaining the 51% of the total variance, allow a good separation of the three geographical areas to be observed. PC1 contributes mainly to the discrimination between olive oils from the South and the Center. Linolenic acid, βsitosterol, oleic acid and squalene are the most discriminant parameters: in fact, samples coming from the Center (䊐) have a high value of linolenic acid and β-sitosterol; PC2 mostly to the separation between the Northern area and the Center. The most discriminant parameters correspond to terpenes and wax, having a high value in the samples from the Center and to linolenic acid having a high value in samples from the North. The oleic acid content decreases slightly from South to North; this observation might be related to the presence of different cultivars, to the agronomical practice or to the rainfalls which give water to the soils. It is important to analyze the meaning that, in this particular study, the 27 selected variables have. The variable
122
SECTION | I The Detailed Characterization of Olives and Olive Products
3
3 North
Center
South
3
2
450–600
> 300–450
150–300
< 150 0
100
200
300
400
500
600
700
800
900
ppm Caffeic acid
FIGURE 19.1 Total polyphenols content of 25 cultivated varieties. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
167
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
168
SECTION | I
COOH
COOH HO
O OGluc Deoxy loganic acid
OH
Lipids, Phenolics and Other Organics and Volatiles
O
O
O
O
OGluc Loganic acid
OGluc Secologanic acid
OGluc Secologanin HOOC
COOH OH
CHO COOMe
O
O
O
OGluc Secoxy loganin
OGluc 7-Epi-loganic acid
HOOC COOH O
COOMe
COOMe
COOMe O
O
OGluc 7-Ketologanic
O OGluc Oleoside11-methyl ester
O
OGluc 7-Ketologanin
{elenolic acid glucoside} GlucOOC
COOMe
O
HO
O
O
COOMe
HO Oleuropein
O
H
O
O
OGluc 7-β-D-Glucopyransyl 11-methyl oleoside COOMe
HO
OGluc
Ligstroside
O OGluc
FIGURE 19.2 Pathway of synthesis of oleuropein in Oleaceae.
1992a, b, 1993) (Figure 19.2). Olive-oil secoiridoids in aglyconic forms derive from the glycosides in olives via the hydrolysis of endogenous glucosidases during crushing and malaxation. These newly formed amphiphilic substances are to be found both in the oily layer and the water although they are more concentrated in the latter fraction because of their polar functional groups. During the storage of VOO hydrolytic mechanisms may be involved in the release of simple phenolics such as hydroxytyrosol and tyrosol from the more complex secoiridoids (Gutierrez Gonzales-Quijano et al., 1977; Tsimidou, 1998). The most abundant secoiridoids in VOO, identified for the first time by Montedoro et al. (1992a, b, 1993; Montedoro and Cantarelli, 1969) and confirmed by other authors (Cortesi et al., 1995a, b; Angerosa et al., 1996) are the dialdehyde form of elenolic acid linked to hydroxytyrosol or tyrosol ( p-HPEA), known respectively as 3,4DHPEA-EDA and p-HPEA-EDA, and an isomer of the oleuropein aglycon (3,4-DHPEA-EA) (Table 19.1). In 1999 another hydroxytyrosol derivative, hydroxytyrosol acetate (3,4-DHPEA-AC), was found in VOO (Brenes et al., 1999). Phenolic acids are naturally occurring secondary aromatic plant metabolites found widely throughout the plant
kingdom (Exarchou et al., 2003; Prim et al., 2003). They contain two distinguishing constitutive carbon frameworks, the hydroxycinnamic and hydroxybenzoic structures. Their various contributions to plant life are currently being subject to intense scrutiny, one aspect of which deals specifically with their role in food quality (Hakkinen et al., 1999; Barthelmebs et al., 2000). Phenolic acids have been associated with color and sensory qualities as well as with the health-related and antioxidant properties of foods (Maga, 1978; Mergiz and Unal, 1991). One impetus for analytical investigation has been the role of phenolics in the organoleptic properties (flavor and astringency) of foods (Peleg et al., 1991; Harborne, 1994). Additionally, the content and profile of phenolic acids, their effect on fruit ripening, prevention of enzymatic browning and their role in food preservation have been evaluated (Shahidi and Nacsk, 1995). Recent interest in phenolic acids stems from their potential protective role through fruit and vegetables against diseases that may be related to oxidative damage (coronary diseases, strokes and cancers) (Gomes et al., 2003; Robbins, 2003). In particular several phenolic acids such as gallic, protocatechuic, p-hydroxybenzoic, vanillic, caffeic, syringic, p- and
CHAPTER | 19 Polyphenols in Olive Oil: The Importance of Phenolic Compounds in the Chemical Composition
169
TABLE 19.1 Phenolic compounds in virgin olive oil: compound name, general chemical structure and molecular weight. Compound
Substituent (MW)
Structure
Benzoic and derivative acids 3-Hydroxybenzoic acid
3-OH (138)
p-Hydroxybenzoic acid
4-OH (138)
3,4-Dihydroxybenzoic acid
3,4-OH (154)
Gentisic acid
2,5-OH (154)
Vanillic acid
3-OCH3, 4-OH (168)
Gallic acid
3,4,5-OH (170)
Syringic acid
3,5-OCH3, 4-OH (198)
5
6
4
1
COOH
2
3
Cinnamic acids and derivatives o-Coumaric acid
2-OH (164)
p-Coumaric acid
4-OH (164)
Caffeic acid
3,4-OH (180)
Ferulic acid
3-OCH3, 4-OH (194)
Sinapinic acid
3,5-OCH3, 4-OH (224)
5
6 COOH
4
1 2
3
Phenyl ethyl alcohols Tyrosol [(p-hydroxyphenyl)ethanol] or p-HPEA 4-OH (138) Hydroxytyrosol [(3,4dihydroxyphenyl)ethanol] or 3,4-DHPEA
5
6 OH
3,4-OH (154)
4
1 3
2
5
6
Other phenol acids and derivatives p-Hydroxyphenylacetic acid
4-OH (152)
3,4-Dihydroxyphenylacetic acid
3,4-OH (168)
4-Hydroxy-3-methoxyphenylacetic acid
3-OCH3, 4-OH (182)
3-(3,4-Dihydroxyphenyl)propanoic acid
4
1 3
(182)
COOH
2 COOH
HO
HO
Dialdehydic forms of secoiridoids Decarboxymethyloleuropein aglycon (3,4DHPEA-EDA)
R1-OH (304)
Decarboxymethyl ligstroside aglycon (pHPEA-EDA)
R1-H (320)
R*—O
dialdehydic form of Elenolic Acid (EDA) CHO CHO
(Continued)
170
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 19.1 (Continued) Compound
Substituent (MW)
Structure
Secoiridoid aglycons Oleuropein aglycon or 3,4-DHPEA-EA
R1-OH (378)
Ligstroside aglycon or p-HPEA-EA
R1-H (362)
Aldehydic form of oleuropein aglycon
R1-OH (378)
Aldehydic form of ligstroside aglycon
R1-H (362)
OCH3 O
R1
O
C
O
HO p-HPEA or 3,4-DHPEA
O
O
Elenolic Acid (EA)
Aldehydic form of Elenolic Acid (EA)
R* OH
O
Flavonols (304)
(⫹)-Taxifolin
OH OH O
HO
OH OH
O
Flavones Apigenin
R1-OH, R2-H (270)
Luteolin
R1-OH, R2-OH (286)
OH R2 O
HO
H R1
O
Lignans (⫹)-Pinoresinol
R-H (358)
(⫹)-1-Acetoxypinoresinol
R-OCOCH3 (416)
(⫹)-1-Hydroxypinoresinol
R-OH (374)
OCH3 OH O H
R
HO O
H3CO
Hydroxyisochromans 1-Phenyl-6,7-dihydroxyisochroman
R1,R2-H (242)
1-(3⬘-Methoxy-4⬘-hydroxy)phenyl-6,7dihydroxy-isochroman
R1-OH,R2-OCH3 (288)
O R2
HO
R1 OH
CH3
CHAPTER | 19 Polyphenols in Olive Oil: The Importance of Phenolic Compounds in the Chemical Composition
o-coumaric, ferulic and cinnamic have been identified and quantified in VOO (in quantities lower than 1 mg of analyte kg⫺1 of olive oil). Two research groups have been involved in extensive analyses of VOO for these types of compounds (Buiarelli et al., 2004; Carrasco-Pancorbo et al., 2004, 2005). In one of their papers the authors found that trans-cinnamic acid, sinapinic acid, caffeic acid and 3,4dihydroxyphenylacetic acid were present in several monovarietal VOOs of the six Spanish olive cultivars analyzed (Carrasco-Pancorbo et al., 2004) and so these compounds might be potential markers of geographical origin or olive fruit variety. (⫹)-Pinoresinol is a common component of the lignan fraction of several plants, such as Forsythia species (Davin et al., 1992) and Sesamum indicum seeds, whilst (⫹)-1acetoxypinoresinol and (⫹)-1-hydroxy-pinoresinol and their respective glucosides have been detected in the bark of Olea europaea L. According to Owen et al. (2000), the quantity of lignans in VOO may be as high as 100 mg kg⫺1, but as with the simple phenolics and secoiridoids, considerable variation exists between different oils. As suggested by Brenes et al. (2002), the quantity of lignans may be used as a varietal marker, and they reported a method to authenticate VOO produced by Picual olives based on the very low content of the lignan (⫹)-1-acetoxypinoresinol in these oils. A few years ago Bianco et al. (2001) investigated the presence of hydroxy-isochromans in VOOs. In fact, during the malaxation step of oil extraction the quantities of hydroxytyrosol and carbonylic compounds are increased by hydrolytic processes via the activity of glycosidases and esterases, thus favoring the presence of all the compounds necessary for the formation of isochroman derivatives. Two hydroxy-isochromans, formed by the reaction between hydroxytyrosol and benzaldehyde or vanillin, have been identified by HPLC-MS/MS and quantified in commercial VOOs. Flavonoids are widespread secondary plant metabolites. During the past decade an increasing number of publications highlighting the beneficial effects of flavonoids upon health have appeared, including some related to cancer and coronary diseases (Le March, 2002; Nestel, 2003; Kanadaswami et al., 2005). Flavonoids are largely planar molecules and their structural variation comes in part from the pattern of modification by hydroxylation, methoxylation, prenylation or glycosylation. Flavonoid aglycones are subdivided into flavones, flavonols, flavanones, and flavanols, depending upon the presence of a carbonyl carbon at C-4, an OH group at C-3, a saturated single bond between C-2 and C-3 or a combination of a non-carbonyl at C-4 with an OH group at C-3, respectively. Several authors have reported that flavonoids such as luteolin and apigenin are also present in VOO (Rovellini et al., 1997; Murkovic et al., 2004; Morelló et al., 2005; CarrascoPancorbo et al., 2006). Luteolin may originate from rutin or luteolin-7-glucoside, and apigenin from apigenin glucosides. Some interesting studies have also been published
171
describing how several flavonoids have been found in olive leaves and fruit (Romani et al., 2000; Ryan et al., 2003; Bouaziz et al., 2005).
19.2 WHY ARE THE PHENOLIC COMPOUNDS IN VIRGIN OLIVE OIL SO IMPORTANT? Boskou published an interesting review in 2006 (Boskou, 2006) in which the sources of natural phenolic antioxidants were discussed, and the following idea was highlighted, ‘Widely distributed in the plant kingdom and abundant in our diet, plant phenolics are today among the most talked about classes of phytochemicals’. To answer the question, ‘Why are phenolic compounds so interesting?’ the author of the review summarized several issues which have been studied in depth over the last decade: ●
●
●
●
●
●
The levels and chemical structure of antioxidant phenolics in different plant foods, aromatic plants and various plant materials. The probable role of plant phenolics in the prevention of various diseases associated with oxidative stress, such as cardiovascular and neurodegenerative diseases and cancer. The ability of plant phenolics to modulate the activity of enzymes, a biological action not yet understood. The ability of certain classes of plant phenolics such as flavonoids (also called polyphenolics) to bind to proteins. Flavonol–protein binding, such as binding to cell receptors and transporters, involves mechanisms that are not related to their direct activity as antioxidants. The stabilization of edible oils, protection from formation of off-flavors and the stabilization of flavors. The preparation of food supplements.
19.3 HEALTH ASPECTS LINKED TO PHENOLICS IN VOO VOO is an integral ingredient of the Mediterranean diet and accumulating evidence suggests that it may have health benefits which include the reduction of risk factors of coronary disease, prevention of several types of cancer and the modification of immune and inflammatory responses. VOO can be considered as an example of a functional food containing a variety of components that may contribute to its overall therapeutic characteristics (Stark and Madar, 2002). Its nutritional and health values and pleasant flavor have contributed to an increase in its consumption, which has fostered the cultivation of olives outside the traditional olive-oil-producing regions of the Mediterranean basin into relatively new areas such as Australia, Argentina and South Africa. The nutritional value of VOO can be attributed to its high levels of oleic acid and other minor components such as phytosterols, carotenoids, tocopherols and hydrophilic phenolics (Perez-Jimenez, 2005) (Figure 19.3).
172
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
VIRGIN OLIVE OIL
Saponifiable fraction
Non-saponifiable fraction
Lipidic pattern of the VOO-rich ‘Mediterranean Diet’: Low ω-6/ω-3 PUFAs ratio + High levels of ω-9 MUFA
+PHENOLIC BIOCOMPOUNDS
Phenolics PHENOLICS
Cell viability
Cell viability
ω-6 PUFAs
ω-9 OA ω-3 PUFAs HER2
Healthy properties
FIGURE 19.3 Relation between the olive oil composition and the health properties.
19.4 PHENOLIC CONTRIBUTION TO THE OXIDATIVE STABILITY OF VOO Oxidation is an inevitable process that starts after the oil has been extracted and leads to a deterioration that becomes more pronounced during storage. Initially lipids are radically oxidized to hydroperoxides, which are odorless and tasteless (Frankel, 1982) and do not bring about any sensory changes. Nevertheless, decomposition occurs through homolytic cleavage of the hydroperoxide group, generating various volatile compounds known as secondary oxidation products, which are responsible for typical unpleasant sensory characteristics. Oxygen, light, temperature, metals, pigments, unsaturated-fatty-acid composition and the quantity and kind of natural antioxidants are all factors that can influence the free-radical mechanism of the autoxidation process in different ways (Montedoro, 1972; Frankel, 1985). Natural antioxidants behave in very different, complex ways at air–oil and oil–water interfaces, which significantly affect their relative activities in different lipid systems. The presence of hydrophilic phenolic compounds in VOO and their high antioxidant activity can be explained by the so-called ‘polar paradox’ (Porter et al., 1989), which maintains that ‘polar antioxidants are more effective in nonpolar lipids whereas non-polar antioxidants are more active in polar lipid emulsions’. According to Frankel (1996), in a bulk oil system hydrophilic antioxidants such as polar phenolics are located at the air–oil interface (a small quantity
of air is always trapped in the oil) and thus protect more effectively against oxidation than lipophilic antioxidants such as tocopherols, which remain in solution in the oil. By studying the phenolic profiles of the oils after heating them to 180°C and then using HPLC-UV, HPLC-MS and CE-UV techniques, Carrasco-Pancorbo et al. (2007) found several hitherto unknown compounds that were probably linked to phenolic oxidation. In particular, seven peaks increased significantly when the thermal treatment was kept up for 1–3 hours, and their presence has also been confirmed in refined olive oils. The concentrations of hydroxytyrosol, elenolic acid, 3,4-DHPEA-EDA and 3,4-DHPEA-EA decreased more rapidly with the thermal treatment than did other phenolic compounds present, confirming their high antioxidant power. Moreover, 3,4-DHPEA-AC and p-HPEAEA were more resistant to heat treatment, whereas the quantities of (⫹)-pinoresinol and (⫹)-1-acetoxypinoresinol remained almost unchanged (Figure 19.4).
19.5 SENSORY PROPERTIES AFFECTED BY PHENOLICS IN VOO Virgin olive oil is a natural fruit juice obtained directly from olives without any further refining process. Its flavor is characteristic and is markedly different from those of other edible fats and oils. The combined effects of taste, odor (directly via the nose or indirectly through the retronasal path via the mouth) and chemical responses (pungency,
173
CHAPTER | 19 Polyphenols in Olive Oil: The Importance of Phenolic Compounds in the Chemical Composition
Olfactory fruity
OSI vs Total Phenols (by spectrophotometric assay) 700
5.0
y = 12.133x + 51.756 R2 = 0.6523 r=0.8077 N=76
600 500
4.0 3.0
Pleasant flavors (2)
400
2.0
300
1.0
200
0.0
100
Pleasant flavors (1)
Sweet
Gustatory fruity
0 A
0
5
10
15
20
25
30
35
40
OSI vs DPPH (antiradical test) Pungent
2.5 y = 0.0615x − 0.2062 R2 = 0.707 r=0.8408 N=64
2.0
0.5
10
15
20
25
OSI vs o-diphenols (by spectrophotometric assay) 350 y = 5.9318x + 8.5033 R2 = 0.7302 r=0.8545 N=58
300 250
mg 3,4-DHPAA/kg of oil
1.0
5
250 200 150 100 50
HPh
HPh LPh
8 6 4 2
200
0
B
150 100 50 0 C
LPh
A
1.5
0.0 B 0
Bitter
0
5
10
15
20
25
30
35
40
FIGURE 19.4 Correlations between OSI values (in hours), phenol quantities and antioxidant activity (DPPH test) by spectrophotometric assays. (A) OSI vs total phenols (mg gallic acid kg⫺1 VOO); (B) OSI vs DPPH (mmol trolox kg⫺1 VOO); (C) OSI vs o-diphenols (mg gallic acid kg⫺1 VOO). Analyses were carried out over three years; the number of samples is given in each figure (N). Three replicates were prepared and analyzed for each sample.
astringency, metallic, cooling or burning) give rise to the sensation generally perceived as ‘flavor’ (Kilcast, 2003). Virgin olive oil, when extracted from fresh and healthy olives (Olea europaea L.) and properly processed and stored, is characterized by a unique, highly appreciated combination of aroma and taste (Kiritsakis, 1998; Angerosa et al., 2000). The sensory aspect, due to the use of VOO as a seasoning in both cooked and raw foods, has great repercussions on its acceptability. Thus, since sensory quality plays an important role in directing the preference of consumers, many attempts have been made to clarify the relationships between the sensory attributes in a VOO as perceived by assessors and its volatile and phenolic profiles, which are responsible for aroma and taste respectively.
A
B
C
D
E
F
G
H
I
L
Total
FIGURE 19.5 Sensory profile and phenolic content of two different VOOs (HPh, high phenol oil and LPh, low phenol oil). (A) Sensory profiles of samples by quantitative descriptive analysis (QDA); the intensity of each descriptor is evaluated on a scale of 0–5; different perception routes: (1) orthonasal, (2) retronasal; (B) single and total phenolic content of samples; A, hydroxytyrosol; B, tyrosol; C, vanillic acid; D, unknown phenolic compound with a retention time of 30.69 min; E, unknown phenolic compound with a retention time of 36.27 min; F, 3,4-DHPEA-EDA; G, (⫹)-pinoresinol; H, (⫹)-1-acetoxypinoresinol ⫹ p-HPEA-EDA; I, 3,4-DHPEA-EA; L, p-HPEA-EA.
The bitterness and pungency perceived by taste are positive attributes for a VOO. These two sensory characteristics are closely connected by the qualitative–quantitative phenolic profile of the product (Figure 19.5).
SUMMARY POINTS ●
●
●
Phenolic compounds of virgin olive oil (VOO) belong to different classes: phenolic acids, phenyl ethyl alcohols, hydroxy-isochromans, flavonoids, lignans and secoiridoids. Secoiridoids are the main compounds of the phenolic fraction of Oleaceae plants. The composition of the phenolic fraction is affected by many agronomical and technological factors.
174
●
SECTION | I
The phenolic fraction has beneficial effects on health, influences long shelf-life compared to other vegetable oils and contributes to the typical sensory properties of VOO, such as bitterness and pungency, all of which made its study very important.
REFERENCES Angerosa, F., D’Alessandro, N., Konstantinou, P., Di Giacinto, L., 1996. Characterization of phenol and secoiridoid aglycons present in virgin olive oil by gas chromatography-chemical ionization mass spectrometry. J. Chromatogr. A 736, 195–203. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2000. Virgin olive oil odour notes: their relationships with volatile compounds from the lipoxygenase pathway and secoiridoid compounds. Food Chem. 68, 283–287. Barthelmebs, L., Divies, C., Cavin, J.F., 2000. Knockout of the pcoumarate decarboxylase gene from lactobacillus plantarum reveals the existence of two other inducible enzymatic activities involved in phenol acid metabolism. Appl. Environ. Micro. 66, 3368–3375. Bianco, A., Coccioli, F., Guiso, M., Marra, C., 2001. The occurrence in olive oil of a new class of phenol compounds: hydroxy-isochromans. Food Chem. 77, 405–411. Boskou, D., 2006. Sources of natural antioxidants. Trends Food Sci. Technol. 17, 505–512. Bouaziz, M., Grayer, R.J., Simmonds, M.S.J., Damak, M., Sayadi, S., 2005. Identification and antioxidant potential of flavonoids and low molecular weight phenols in olive cultivar Chemlali growing in Tunisia. J. Agric. Food Chem. 53, 236–241. Brenes, M., Garcıa, A., Garcıa, P., Rios, J.J., Garrido, A., 1999. Phenolic compounds in spanish olive oils. J. Agric. Food Chem. 47, 3535–3540. Brenes, M., Garcıa, A., Rios, J.J., Garcıa, P., Garrido, A., 2002. Use of 1-acetoxypinoresinol to authenticate Picual olive oils. Int. J. Food Sci. Technol. 37, 615–625. Buiarelli, F., Di Berardino, S., Coccioli, F., Jasionowska, R., Russo, M.V., 2004. Determination of phenol acids in olive oil by capillary electrophoresis. Anal. Chim. 94, 699–705. Carrasco-Pancorbo, A., Cruces-Blanco, C., Segura Carretero, A., Fernández Gutiérrez, A., 2004. Sensitive determination of phenol acids in extra-virgin olive oil by capillary zone electrophoresis. J. Agric. Food. Chem. 52, 6687–6693. Carrasco-Pancorbo, A., Cerretani, L., Bendini, A., Segura-Carretero, A., Lercker, G., Fernández-Gutiérrez, A., 2007. Evaluation of the influence of thermal oxidation in the phenolic composition and in antioxidant activity of extra-virgin olive oils. J. Agric. Food Chem. 55, 4771–4780. Carrasco-Pancorbo, A., Gómez-Caravaca, A.M., Cerretani, L., Bendini, A., Segura-Carretero, A., Fernández-Gutiérrez, A., 2006. Rapid quantification of the phenol fraction of Spanish virgin olive oils by capillary electrophoresis with uv detection. J. Agric. Food Chem. 54, 7984–7991. Carrasco-Pancorbo, A., Segura-Carretero, A., Fernández-Gutiérrez, A., 2005. Co-electroosmotic capillary electrophoresis determination of phenol acids in commercial olive oil. J. Sep. Sci. 28, 925–934. Cortesi, N., Azzolini, M., Rovellini, P., Fedeli, E., 1995a. I componenti minori polari degli oli vergini di oliva: ipotesi di struttura mediante LC-MS. Riv. Ital. Sost. Grasse 72, 241–251.
Lipids, Phenolics and Other Organics and Volatiles
Cortesi, N., Azzolini, M., Rovellini, P., Fedeli, E., 1995b. Dosaggio dei componenti minori polari (CMP) in oli vergini di oliva. Riv. Ital. Sostanze Grasse 72, 333–337. Davin, B.D., Bedgar, D.L., Katayama, T., Lewis, N.G., 1992. On the stereoselective synthesis of (1)-pinoresinol in Forsythia suspensa from its achiral precursor, coniferyl alcohol. Phytochemistry 31, 3869–3874. Exarchou, V., Godejohann, M., van Beek, T.A., Gerothanassis, I.P., Vervoort, J., 2003. LC-UV-solid-phase extraction-NMR-MS combined with a cryogenic flow probe and its application to the identification of compounds present in Greek Oregano. Anal. Chem. 75, 6288–6294. Frankel, E.N., 1982. Volatile lipid oxidation products. Prog. Lipid Res. 22, 1–33. Frankel, E.N., 1985. Chemistry of autoxidation: mechanism, products and flavor significance. In: Min, D.B., Smouse, T.H. (eds) Flavor Chemistry of Fats and Oils. AOCS Press, Champaign, IL (USA), pp. 1–37. Frankel, E.N., 1996. Antioxidants in lipid foods and their impact on food quality. Food Chem. 57, 51–55. Gomes, C.A., Girao da Cruz, T., Andrade, J.L., Milhazes, N., Borges, F., Marques, M.P.M., 2003. Anticancer activity of phenolic acids of natural or synthetic origin: A structure-activity study. J. Med. Chem. 46, 5395–5401. Gutierrez Gonzales-Quijano, R., Janer del Valle, C., Janier del Valle, M.L., Gutierrez Rosales, F., Vazquez Roncero, A., 1977. Relacion entre los polifenoles y la calidad y estabilidad del aceite de oliva virgen. Grasas Aceites 28, 101–106. Hakkinen, S., Heinonen, M., Karenlampi, S., Mykkanen, H., Ruuskanen, J., Torronen, R., 1999. Screening of selected flavonoids and phenol acids in 19 berries. Food Res. Int. 32, 345–353. Harborne, J.B., 1994. The Flavonoids: Advances in Research since 1986. Chapman and Hall, London, U.K. Kanadaswami, C., Lee, L.T., Lee, P.P.H., Hwang, J.J., Ke, F.C., Huanh, Y.T., Lee, M.T., 2005. The antitumour activities of flavonoids. In vivo 19, 895–909. Kilcast, D., 2003. Sensory science. Chem. Brit. 39, 62. Kiritsakis, A.K., 1998. Flavor components of olive oil—a review. J. Am. Oil Chem. Soc. 75, 673–681. Le March, L., 2002. Cancer preventive effects of flavonoids—a review. Biomed. Pharmacother. 56, 296–301. Maga, J.A., 1978. Simple phenol and phenol compounds in food flavor. Crit. Rev. Food Sci. Nutr. 10, 323–372. Montedoro, G., 1972. Costituenti fenolici presenti negli oli vergini di oliva Nota I: Identificazione di alcuni acidi fenolici e loro potere antiossidante. Sci. Tecnol. Aliment. 3, 177–186. Montedoro, G.F., Servili, M., Baldioli, M., Miniati, E., 1992a. Simple and hydrolyzable phenolic compounds in virgin olive oil. 1. Their extraction, separation, and quantitative and semiquantitative evaluation by HPLC. J. Agric. Food. Chem. 40, 1571–1576. Montedoro, G.F., Cantarelli, C., 1969. Indagini sulle sostanze fenoliche presenti negli oli d’oliva. Riv. Ital. Sost. Grasse 46, 115–124. Montedoro, G.F., Servili, M., Baldioli, M., Miniati, E., 1992b. Simple and hydrolyzable phenolic compounds in virgin olive oil. 2. Initial characterization of the hydrolyzable fraction. J. Agric. Food. Chem. 40, 1577–1580. Montedoro, G.F., Servili, M., Baldioli, M., Selvaggini, R., Miniati, E., Macchioni, A., 1993. Simple and hydrolyzable compounds in virgin olive oil. 3. Spectroscopic characterizations of the secoiridoid derivatives. J. Agric. Food Chem. 41, 2228–2234.
CHAPTER | 19 Polyphenols in Olive Oil: The Importance of Phenolic Compounds in the Chemical Composition
Morelló, J.R., Vuorela, S., Romero, M.P., Motilva, M.J., Heinonen, M., 2005. Antioxidant activity of olive pulp and olive oil phenol compounds of the arbequina cultivar. J. Agric. Food Chem. 53, 2002–2008. Murkovic, M., Lechner, S., Pietzka, A., Bratacos, M., Katzogiannos, E., 2004. Analysis of minor components in olive oil. J. Biochem. Methods 61, 155–160. Nergiz, C., Unal, K., 1991. Determination of phenol acids in virgin olive oil. Food Chem. 39, 237–240. Nestel, P., 2003. Isoflavones: their effects on cardiovascular risk and functions. Curr. Opin. Lipidol. 14, 3–8. Owen, R.W., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2000. Identification of lignans as major components in the phenolic fraction of olive oil. Clinical Chem. 46, 976–988. Peleg, H., Naim, M., Rouseff, R.L., Zehavi, U., 1991. Distribution of bound and free phenol acids in oranges (Citrus sinensis) and grapefruits (Citrus paradisi). J. Sci. Food Agric. 57, 417–426. Perez-Jimenez, F. (coordinator), 2005. International conference on the healthy effect of virgin olive oil. Eur. J. Clinic. Invest. 35, 421–424. Porter, W.L., Black, E.D., Drolet, A.M., 1989. Use of polyamide oxidative fluorescence test on lipid emulsions: contrast in relative effectiveness of antioxidants in bulk versus dispersed systems. J. Agric. Food Chem. 37, 615–624.
175
Prim, N., Pastor, F.I.J., Diaz, P., 2003. Biochemical studies on cloned Bacillus sp. BP-7 phenol acid decarboxylase PadA. Appl. Microbiol. Biot. 63, 51–56. Robbins, R.J., 2003. Phenolic acids in foods: An overview of analytical methodology. J. Agric. Food Chem. 51, 2866–2887. Romani, A., Pinelli, P., Mulinacci, N., Vincieri, F.F., Gravano, E., Tattini, M., 2000. HPLC analysis of flavonoids and secoiridoids in leaves of Ligustrum vulgare L. (Oleaceae). J. Agr. Food Chem. 48, 4091–4096. Rovellini, P., Cortesi, N., Fedeli, E., 1997. Analysis of flavonoids from Olea europaea by HPLC-UV and HPLC-electrospray-MS. Riv. Ital. Sost. Grasse 74, 273–279. Ryan, D., Prenzler, P.D., Lavee, S., Antolovich, M., Robards, K., 2003. Quantitative changes in phenol content during physiological development of olive (Olea europaea) cultivar Hardy’s Mammoth. J. Agric. Food Chem. 51, 2532–2538. Shahidi, F., Nacsk, M., 1995. Food Phenolics: Sources, Chemistry, Effects, and Applications. Technomic Publishing Company, Inc., Lancaster, PA. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food: Epidemiology and nutritional approaches. Nutr. Rev. 60, 170–176. Tsimidou, M., 1998. Polyphenols and quality of virgin olive oil in retrospective. Ital. J. Food Sci. 10, 99–115.
This page intentionally left blank
Chapter 20
Phenolic Profiles of Portuguese Olives: Cultivar and Geographics Rosa M. Seabra1, Paula B. Andrade1, Patrícia Valentão1, Miguel Faria2, Alistair G. Paice3 and Maria Beatriz P. P. Oliveira2 1
REQUIMTE Serviço de Farmacognosia, Faculdade de Farmácia, Universidade do Porto, Portugal REQUIMTE Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal 3 REQUIMTE Department of Clinical Biochemistry and Nutrition and Dietetics, King’s College School of Medicine, London, United Kingdom
2
20.1 INTRODUCTION According to FAO, olive is the most extensively cultivated fruit crop in the world. This can be attributed to the high nutritional value of its products, its tolerance to drought and salinity, and to its minimal maintenance requirements. Although it can be found in the USA, Argentina and Mexico, it is in countries from the Mediterranean basin that olive groves have their main impact. Therefore, table olives and olive oil are characteristic components of the Mediterranean diet and are considered one of the factors responsible for the low incidence of coronary heart disease and prostate, breast and colon cancers in this area (Covas, 2007). The aforementioned health benefits of olives are partially attributed to their characteristic fatty acid composition but also micronutrients such as the tocopherol antioxidants and, above all, phenolic compounds. In fact, studies have shown that diets containing olive oil achieve better results than other diets containing equivalent amounts of oleic acid but lower amounts of phenolic compounds (Medeiros, 2001). Aside from its recognized role as a health promoter, the phenolic composition of olive oil has also been assayed as a tool to indicate the oil’s quality and to differentiate olive cultivars. Portugal, although possessing a large Atlantic coast, is still considered a Mediterranean country; especially its eastern region, which is characterized by a climate and soil very similar to those observed in Spain, and where olive groves are a characteristic feature of the landscape. Although its oil production is much less than that of some other EU countries, it is comparable to that of Algeria and double that of Palestine (http:// www.internationaloliveoil.org). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
This chapter intends to give an overview of Portuguese olive production and cultivars and to evaluate the role of phenolic profiles as possible markers.
20.2 OLEA EUROPAEA CULTIVARS AND THEIR DIFFERENTIATION Olea europaea is a highly variable species with a total of 1250 cultivars dispersed over 54 countries as referred to in the OLEA DATABASE (http://www.oleadb.it/). This large number of olive cultivars is partially explained by the fact that olive plants can survive for a long time, thus retaining their genetic characteristics for thousands of years. Furthermore, open crossing between individuals, environmental pressure and its long history as a crop have resulted in numerous cultivars (Hatzopoulos et al., 2002). Traditionally, cultivars have been distinguished by their phenotypical characteristics. These include countless traits from the size, weight, shape and color of fruit to the precocity of maturation, the pulp/stone ratio, size of leaves, yield of oil, resistance to disease, etc. (Pinheiro and Silva, 2005). Notwithstanding their major role and usefulness in characterizing cultivars, these markers can be affected by the environment, which may present a problem if they are to be used for identification. With the advent of specific chemical methodologies, the chemical composition of leaves, olives and olive oil began to be used as possible markers. Chemical constituents that have been used for this purpose include tocopherols, sterols, fatty acids (Matos et al., 2007), volatile compounds (Tura et al., 2004) and phenolic compounds (Vinha et al., 2005). Possible correlations between cultivar and biochemical (Briante et al., 2002) and physiological (Bacelar et al., 2007) behavior have also been studied.
177
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
178
SECTION | I
In the search for new differentiation methodologies, researchers have over the last two decades begun to use molecular markers in olive characterization. Subsequently practically all the available molecular markers for plants have been applied in olive trees (Owen et al., 2005; Sarri et al., 2006).
20.3 PHENOLICS IN OLIVE FRUITS In olives, phenolics may represent simple substances distributed into several subgroups such as benzoic or cinnamic acids, glycosylated or free flavonoids (mainly luteolin, quercetin or apigenin based) or phenolic alcohols represented by tyrosol and hydroxytyrosol (and their glycosides and derivatives like cornoside and halleridone). Verbascoside is a slightly more complex molecule synthesized by grouping caffeic acid with rutinose and hydroxytyrosol. However, the phenolic profile of olives is predominantly characterized by the presence of a group of compounds with a mixed biosynthetic origin (shikimate/mevalonate) where tyrosol or hydroxytyrosol are linked to an iridoid moiety giving rise to the so-called secoiridoid phenolics. Oleuropein and demethyloleuropein are the main examples of this. Olives and olive oil also characteristically contain the iridoidic component of the latter class of compounds but, as they are devoid of phenolic character, these will not be referred to in this chapter.
20.4 OLIVES IN PORTUGAL: CULTIVARS AND PRODUCTION A database organized by FAO (http://apps3.fao.org/wiews/ olive/intro.jsp) includes 21 cultivars of Portuguese origin (data collected from the Portuguese Agronomical Stations). Table 20.1 displays those cultivars together with their main synonyms. The geographic distribution indicated refers to the provinces where the cultivars are grown which, as mentioned above, are located in the inner part of Portugal (see Figure 20.1). However, reference to other Portuguese cultivars such as Roupuda (Vinha et al., 2005) or Douro (Bianco and Uccella, 2000) can also be found in the literature. As can be seen in Table 20.2, Portuguese olive groves are mainly located in Alentejo. In Portugal, olives are grown in small parcels, which represent a hindrance to high productivity but favor the creation of small, high-quality niches. Consumers wish for variety; products considered as safe, more palatable and produced by traditional methods (regarded as ‘natural’) are now more desirable. In spite of its relatively small production, Portugal has seven olive oils registered as PDO (Protected Denomination of Origin) spread within the main producing areas (Figure 20.1). According to data from EUROSTAT (2005) (http://www. oliveoil.eu), only 2% of the Portuguese olive production is devoted to the generation of table olives. As can be seen in
Lipids, Phenolics and Other Organics and Volatiles
Table 20.3 the Portuguese eat more table olives than they produce. During the 1990s the average annual consumption was in the order of 15 200 tons, which equates to 1.5 kg/ inhabitant/year (http://www.internationaloliveoil.org). Table olives can be prepared from almost all the cultivars growing in Portugal but some cultivars are almost exclusively devoted to table olive. Negrinha is one example, with a low amount of oil production but very good characteristics for table olive. This cultivar is the only one used to produce ‘Negrinha do Freixo Table Olives’ PDO. The only other Portuguese table olive PDO is ‘Axeitona e Elvas e Campo Maior’. The Azeiteira, Carrasquenha, Conserva de Elvas and Redondil cultivars can be used for this. Although olives as a fruit crop are probably the oldest cultivated fruit trees, the maintenance of their intraspecific diversity is a concern. Several factors threaten the existence of certain cultivars. Undoubtedly one of them is the substitution of rustic cultivars with more productive ones. Portugal is no exception. In 2005 80% of the total acreage of Portuguese olive groves was planted with Galega, a cultivar spread throughout the country. This cultivar is characterized by great productivity. However, when compared to Picual, a Spanish cultivar that is being introduced in Portugal, it has considerably lighter fruits (1.48 g compared to 4.20 g for Picual), a lower resistance to ‘Gafa’ (infection by Gloeosporium olivarum) and a smaller fat content (about 17% compared to 23% for Picual). Although the oil obtained from Galega is thought to be more acceptable to the public, there are obvious disadvantages for the Portuguese olive growers using it. Picual is not the only Spanish cultivar that can be found in Portuguese olive groves; other cultivars include Santulhana, Blanqueta, Cornicabra and probably others in small quantities.
20.5 CHARACTERIZATION OF CULTIVARS BY THEIR PHENOLIC PROFILE If, as we believe, cultivars have slight variations in their genomes, then the factor ‘cultivar’ must be the main source of variation in the composition in Olea europaea. However, we also know that external factors such as the climate, soil composition or agricultural practices can modulate gene expression (Romero et al., 2002; Morelló et al., 2003). Apart from genetic factors, one of the most influential intrinsic factors on the set of compounds found in a given cultivar is the degree of ripening. During the development of olives a large set of biochemical reactions takes place (Amiot et al., 1986; Vinha et al, 2005). Most reports point to a decrease of total phenolic content as ripening occurs (Gimeno et al., 2002; Beltrán et al., 2005; Morelló et al., 2005). This decrease does not occur at the same rate for all compounds and in the same way in different cultivars (Morelló et al., 2005). Some authors believe these differences could even be used to differentiate cultivars.
179
CHAPTER | 20 Phenolic Profiles of Portuguese Olives: Cultivar and Geographics
TABLE 20.1 Olea europaea Portuguese cultivars. Cultivar
Main synonyms
Distribution
Uses
Azeitoneira
Azeiteira
Alentejo
Table/oil
Bical*
Bicuda Bical de castelo Branco Longal Cordovil-bical
Beira Alta Alentejo Algarve
Table/oil
Branquita
Blanqueta Blanqueta de Elvas
Ribatejo Alto e Baixo Alentejo
Table/oil
Carrasquenha
Carrasca Carrasquinha Redonda Salola
Beiras Alto Alentejo Baixo Alentejo
Table/oil
Cobrançosa*
Coutinia Quebrançosa Verdeal cobrançosa
Trás-os-Montes Alentejo
Oil
Alto e Baixo Alentejo
Table/oil
Conserva de Elvas Cordovesa*
Cordovil Cordovil de Elvas Vermelhal Bico de corvo Cordovil de Serpa
Algarve Alto Alentejo Baixo Alentejo
Table/oil
Cordovil*
Cordovil de Castelo Branco Cordovil Grosso Cordovil Nocal
Trás-os-Montes Beira Alta Beira Baixa
Table/oil
Galega*
Galega vulgar Galega comum Molar Molarinha Negra molar
Trás-os-Montes Beira Ribatejo Alentejo
Table/oil
Galega grada de Serpa
Galega grossa redonda Galega grada
Alentejo Algarve
Table/oil
Golosinha
Glozinha Golosinha mançanica Pele de sapo
Alentejo
Table/oil
Lentisca*
Borreira Borrenta Brunhenta Carrasca Bical Carrasca Negral Corlideira Zambulha
Trás-os Montes Estremadura Ribatejo Alto Alentejo Baixo Alentejo
Oil
Madural*
Comum Cornicabra de Trás-os-Montes Madural fina Madural grossa Negra volumosa
Trás-os-Montes Beira-Alta
Table/oil
(Continued)
180
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 20.1 (Continued) Cultivar
Main synonyms
Distribution
Uses
Mançanilha algarvia
Mançanilha acuminada
Alentejo Algarve
Table/oil
Mançanilha carrasquenha
Maçanilha carrasca
Alto Alentejo
Table/oil
Negrinha*
Galego negrão Negruxa
Trás-os-Montes
Table
Redondal
Redonda Redondil Grosso Roupudo
Trás-os-Montes
Table/oil
Redondil
Mançanilha meuda
Alto e Baixo Alentejo Ribatejo
Table/oil
Tentilheira
Amendoeira Tentilheira acuminata
Alto Alentejo
Table/oil
Verdeal transmontana*
Verdeal
Trás-os-Montes Beira Alta
Table/oil
Verdeal alentejana
Verdeal Verdeal de Serpa Verdeal tinto Verdeais
Alentejo Algarve
Table/oil
*
Cultivars that were subjected to phenolic analysis (Vinha et al., 2005).
Given that all these potential influences exist, extreme care must be taken when choosing and/or collecting the samples to be analyzed if a ‘chemical fingerprint’ (in this case a phenolic fingerprint) is to be generated. A heterogeneous sampling requires wide sampling and the use of substantial statistical analysis of any results produced. A ‘chemical fingerprint’ (or a ‘chromatographic fingerprint’ as chromatography is the usual technique used) represents a qualitative and quantitative chromatographic pattern of characteristic compounds within a sample. According to Gong and co-workers (Gong et al., 2003) this chromatographic profile should feature the fundamental attributes of ‘integrity’ and ‘fuzziness’ or ‘sameness’ and ‘differences’. These attributes suggest that, with the help of a constructed chromatographic fingerprint, the authentication and identification of a natural (complex) product can be accurately conducted (‘integrity’) even if the number and/or concentration of chemically characteristic constituents are dissimilar from sample to sample (‘fuzziness’) or, chromatographic fingerprints could demonstrate both the ‘sameness’ and ‘differences’ between samples. Given what is summarized above with respect to the variability of natural products, the construction of a reliable chromatographic fingerprint is no trivial undertaking. Too much analysis and
mathematical work are required! From the extensive literature on this subject the nature of the phenolic compounds present in olives is very important. However, as far as we know there is no work defining a general quantitative profile for Olea europaea fruits or for their numerous cultivars. The main factor preventing further understanding in this field is the differing methodologies used in its analysis. Different experimental procedures have been used to extract, separate and detect different compounds and this is the main reason why, in any given work, the authors cannot find the complete set of phenolics described in the review papers (Table 20.4). Bearing in mind all these limitations it is likely that collating data from different papers to try to compare cultivars will be met with very limited success.
20.6 PHENOLIC PROFILES OF PORTUGUESE CULTIVARS As far as we know, there is only one paper reporting the phenolic profiles of Portuguese cultivars (Vinha et al., 2005). Using a simple methodology previously developed
181
CHAPTER | 20 Phenolic Profiles of Portuguese Olives: Cultivar and Geographics
TABLE 20.2 Data for Portuguese olive groves.
SPAIN
Region
A
1
4
SPAIN
3
D
70 000
21
Beira Alta (B*) and Beira Baixa (C*)
60 000
18
Ribatejo (D*)
40 000
12
150 000
44
20 000
6
340 000
100
Remaining
C
Area (%)
Trás-os-Montes (A*)
Alto Alentejo (E*) and Baixo Alentejo (F*)
2
B
Area (ha)
Total
Capital letters in brackets refer to provinces listed in legend of Figure 20.1.
5 E TABLE 20.3 Data for Portuguese table olives. 6 7
F G Capital letters refer to provinces (areas surrounded by full lines); numbers refer to locations of PDO olive oils (areas surrounded by dashed lines). A: Trás-os-Montes; B: Beira Alta; C: Beira Baixa; D: Ribatejo; E: Alto Alentejo; F: Baixo Alentejo; G: Algarve;
1: Trás-os-Montes olive oil; 2: Beira Alta olive oil; 3: Beira Baixa olive oil; 4: Ribatejo olive oil; 5: Norte Alentejano olive oil; 6. Alentejo-Interior olive oil; 7: Moura olive oil.
FIGURE 20.1 Distribution of olive groves and PDO olive oils in Portugal. The figure illustrates the provinces where olive groves are predominantly located together with the regions where PDO olive oil production occurs.
for this (Vinha et al., 2002), the authors analyzed samples from nine of the 21 cultivars listed in Table 20.1 plus three cultivars of Spanish origin cultivated in Portugal. Picual, Cornicabra and Santulhana and Roupuda were labeled as Portuguese but not registered by FAO (Table 20.1). The dry weight results obtained indicated an overall profile characterized by high levels of hydroxytyrosol and oleuropein (Table 20.5 and Figure 20.2) with very low amounts of chlorogenic acid (below 12 mg kg⫺1) and low amounts of verbascoside (usually below 0.02% of total phenolics). The colored phenolics were those usually described, namely cyanidin 3-glucoside and cyanidin 3-rutinoside,
Tonnes
% in EU
Production
1995/6–2000/1 2001/2–2006/7
9500 10 700
2.0 1.5
Imports
1995/6–2000/1 2001/2–2006/7
300 300
0.5 0.4
Exports
1995/6–2000/1 2001/2–2006/7
4800 5100
3.1 2.2
Consumption
1995/6–2000/1 2001/2–2006/7
11 800 13 100
3.1 2.3
with the latter always in higher amounts than the former. As expected, there was a good (but not strict) correlation between the amounts of these compounds and the maturation indices. Five non-colored flavonoids were also found: luteolin 7-glucoside, rutin, apigenin 7-glucoside, quercetin 3-rhamnonise and luteolin. In 25 out of 29 samples analyzed, luteolin 7-glucoside and rutin were the predominant flavonoids and, in general, rutin was present in higher amounts than the luteolin derivative. This seems to be a characteristic of the general olive profile since these two compounds are always reported in cultivars from other countries, even when other flavonoids are not found (Table 20.4). Although it is generally reported that the phenolic content of cultivars decreases during ripening, we did not find that this was true for the cultivars under discussion. For example, the theory that hydroxytyrosol is a degradation
182
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 20.4 Non-colored phenolic profiles of olives. Reference
Amiot et al. (1986)
Number of cultivars
Geographic origin
Extractive method
11
France
Identified phenolics Phenolic alcohols (glucosides, esters and derivatives)
Flavonoids
Phenolic acids and derivatives
80% ethanol; cleaning of water phase by petroleum ether; LLE with ethyl acetate
Oleuropein
Lut 7-Glu Rutin
Verbascoside
Esti et al. (1998)
6
Italy
As in Amiot et al. (1986)
Hydroxytyrosol Demethyloleuropein Oleuropein
Lut 7-Glu Rutin
Romani et al. (1999)
5
Italy
80% ethanol; cleaning of water phase by n-hexane; cleaning by SPE and recover of compounds by ethyl acetate
Tyrosol Hydroxytyrosol Demethyloleuropein Oleuropein
Lut 7-Glu Rutin Api 7-Rut and Api 7-Rut Quer 3-Rut Luteolin Homoorientin
Verbascoside Vanillic acid p-Coumaric acid vanillin
Lut 7-Glu Rutin Api 7-Glu Quer 3-Rham Luteolin
5-CaffeoylQui Verbascoside
Vinha et al. (2005)
10 3
Portugal Spain
Methanol; cleaning of methanolic extract by SPE and recover of compounds by methanol
Tyrosol Hydroxytyrosol Oleuropein
Bianco and Ucella (2000)
1 1 1
Portugal Greece Spain
Simple biophenols (BP) (ext by 6 N HCl)
Hydroxytyrosol Tyrosol
Caffeic acid p-Coumaric acid Hydroxycaffeic acid
Alkali hydrolysable biophenols (extracted by 2N NaOH)
Hydroxytyrosol Tyrosol
Caffeic acid p-Coumaric acid Hydroxycaffeic acid
Cytoplasmic biophenols (extracted by CH2Cl2 and study of the aqueous fraction)
Hydroxytyrosol Tyrosol Demethyloleuropein Oleuropein Oleuropein derivatives Three Hydroxytyrosol Glucosides Tyrosol-1-glucoside Cornoside Halleridone
Soluble BP ⫹ soluble esterified BP ⫹ insoluble-bound BP fraction (extracted by methanol:acetone 1:1)
Hydroxytyrosol Tyrosol Oleuropein Demethyloleuropein
Protocatechuic acid 3,4-diOH-phenylacetic acid p-OH-benzoic acid
Vanillic and homovanillic acids Cinnammic and caffeic acids Syringic acid o-, m- and p-Coumaric acids Ferulic and sinapic acids
183
CHAPTER | 20 Phenolic Profiles of Portuguese Olives: Cultivar and Geographics
TABLE 20.5 Characteristics of Portuguese samples under discussion. Cultivars
Geographical origin
Maturation index
Sum of identified phenolics (mg kg⫺1) Hydroxytyrosol ⫹ oleuropein
Flavonoids
Bical
Macedo de Cavaleiros (A) Mogadouro (A) Fundão (B) Castelo Branco (C)
3.4 2.5 2.5 3.3
20 539 10 357 9684 8264
468 773 859 386
Cobrançosa
Valpaços (A) Mirandela (A) Mogadouro (A) Fundão (C)
4.0 3.9 2.4 3.1
4228 6743 9426 16 166
1367 1461 383 1224
Cordovesa
Macedo de Cavaleiros (A)
3.6
7422
1532
Cordovil
Fundão (C) Castelo Branco (C)
3.8 3.3
15 802 3487
1330 1380
Galega
Fundão (C) Castelo Branco (C)
4.1 4.2
4255 2791
472 1537
Lentisca
Valpaços (A) Valpaços (A) Mirandela (A) Mogadouro (A)
2.8 3.8 3.3 3.3
49 354 8087 3169 14 774
1627 1152 574 2399
Madural
Valpaços (A) Mirandela (A) Mogadouro (A)
3.3 4.3 3.3
11 686 6703 3093
1619 2091 1016
Madural fina
Mogadouro (A)
3.2
73 467
1673
Negrinha
Figueira de Castelo Rodrigo (A)
6.2
12 764
1452
Roupuda*
Mogadouro
1.5
13 712
1937
Verdeal Transmontana
Mirandela (A) Valpaços (A) Macedo de Cavaleiros (A)
4.1 2.2 4.6
5960 4276 29 406
364 443 1172
Capital letters in brackets refer to provinces where samples were collected. * Cultivar not listed by FAO in Table 20.1.
product of oleuropein and, therefore, hydroxytyrosol increases as oleuropein decreases, was not observed in this study. In fact, samples with higher maturation indices did not show any increase in the hydroxytyrosol/oleuropein ratio and no correlation was found between maturation index and hydroxytyrosol content, even for the same cultivar. The same can be said for the content of luteolin, since no correlation was found between the maturation index and the content of luteolin. It is generally accepted that free flavonoids appear at the end of the maturation process as a consequence of hydrolytic phenomena.
The differentiation of cultivars was only possible at the quantitative level. Some trends were seen but, as found in other matrices (Dopico-García et al., 2008), the influence of geographical origin often surpassed that of the cultivar. Different cultivars with the same geographical origin showed very similar profiles. Of course, this may indicate that extrinsic factors had a strong influence on the expression of the genome. Alternatively, these cultivars may have had very similar genomes, or they may have been mislabeled. The heterogeneity of the sampling (Table 20.5) ensures that this work must be considered as preliminary. As such
184
SECTION | I
A
B
Hydroxytyrosol and Oleuropein
Verdeal 1200
80000
mg kg−1
mg kg−1
60000 40000
Lipids, Phenolics and Other Organics and Volatiles
800
400
20000 0
0 Lentisca
Madural fina
Average
4.1
4.6
Cultivars
D
Madural
Cobrançosa
1200
1200
800
800
mg kg−1
mg kg−1
C
400
400
0
0 4.3
3.3
3.2
4.0/Valpaços
3.3*
3.9/Mirandela
MI
E
3.1/Fundão
2.4/Mogadouro
MI/origin
F
Mogadouro
Cobrançosa/Cordovil
1200
1200
800
800
mg kg−1
mg kg−1
2.2
MI
400
0
400
0 2.5/Bical
3.1/Fundão
2.5/Madural MI/cultivar
Hydroxytyrosol
Luteolin 7-glucoside
3.3/Castelo Branco MI/origin
Apigenin 7-glucoside
Luteolin
Oleuropein
Rutin
Quercetin 3-rhamnoside
FIGURE 20.2 Selected data for phenolics in Portuguese samples. This figure depicts the major components of the non-flavonoid (A) or the flavonoid (B–F) profile for Portuguese cultivars exhibiting intracultivar or intrageographical similarities.
one cannot draw any sound conclusions about the individual profiles of each cultivar, but this approach does raise possibilities that might be confirmed by further studies. For example, with respect to hydroxytyrosol and oleuropein, the samples (and therefore cultivars) are very homogeneous, with the notable exceptions of Madural Fina and Lentisca (Figure 20.2A). In these cultivars the levels are very high. Two notes of caution must be added. Madural Fina is not recorded as a synonym of Madural (Table 20.1). This may indicate that it is an independent cultivar. We must also add that we analyzed samples of Lentisca (Table 20.5); however, three of these reached the laboratory labeled as Borrenta or Borreira (Table 20.1) and only one as Lentisca (Valpaços, MI 2.8). As can be seen from Table 20.5 it is unlikely that this difference in the level of hydroxytyrosol and oleuropein can be attributed to different maturation index or geographical origin. Again we propose that this cultivar deserves further attention, and that it may be heterogeneous. From the analysis of oleuropein and hydrxytyrosol content no other conclusions can be drawn.
Flavonoids may be more useful in cultivar differentiation. In Verdeal Transmontana, three of the four samples analyzed had characteristically low amounts of flavonoids (Figure 20.2B). The observed exception may reflect a higher maturation index or a differing geographical origin. The cultivar Bical had slightly more flavonoids and seems to be reasonably homogeneous with respect to the quantities of flavonoids. Madural and Cordovil have consistently higher flavonoid contents. Madural is the only cultivar where luteolin-7-glucoside content resembles that of rutin (Figure 20.2C). Cobrançosa (n ⫽ 4) (Figure 20.2D), Galega (n ⫽ 2) and Lentisca (n ⫽ 4) were more heterogeneous with respect to the proportion of flavonoids. To evaluate the influence of geographical origin on the characteristics of the phenolic profile, the pair of samples collected in Mogadouro but from different cultivars is noteworthy (Figure 20.2E). Of course, similarities are observable among samples from different cultivars with different geographical origins (Figure 20.2F), but these are exceptions.
CHAPTER | 20 Phenolic Profiles of Portuguese Olives: Cultivar and Geographics
20.7 CONCLUSION In conclusion, the phenolic composition of olives is an important concept. However, although a lot of knowledge has already been gained in this area, the many factors influencing phenolic profile have thus far prevented the establishment of a general ‘phenolic profile of olive fruits’. Of course, the phenolic profiles for each cultivar area are not yet fully characterized. There are a large number of registered cultivars and the number of cultivars analyzed in the literature is always very small. To study the effect of the cultivar’s character on the phenolic profile in isolation it is necessary to correct for all the other main variables that may influence this. Examples of these confounding variables might include geographical origin, maturation index and the analytical methodology. Since it is also known that climatic factors can affect the composition of organisms, it is also necessary to monitor compositional changes over several years to monitor the effect of differing weather conditions. This would be a Herculean task, involving a huge number of samples – discouraging for a single laboratory to undertake. Overall, despite the considerable number of techniques applied and the large amount of published literature on the subject, olive cultivars are far from being fully characterized and continuing efforts will have to be made in order to solve problems such as mislabeling, homonyms and synonyms.
SUMMARY POINTS ●
●
●
●
●
●
●
Portugal, although having only an Atlantic coast, is still a Mediterranean country and is responsible for 2% of olives produced in the EU. Twenty-one cultivars are registered in Agronomical Stations as having Portuguese origin. Chemical profiles have revealed themselves as a somewhat useful tool for species and cultivar differentiation. Since phenolics are recognized as health promoter compounds, it would be doubly useful to use the phenolic profile as a differentiation tool for olive cultivars. At a qualitative level, the phenolic compounds identified so far in olive fruits have revealed some constancy, in spite of the numerous factors that can affect them. However, little information is still available for depicting olive fruit quantitative phenolic profile, exactly because those factors exert influence on the levels of compounds. The possibility of using the phenolic profile as a marker for olive cultivars is still under study and many studies are still needed to attain such a goal.
REFERENCES Amiot, M.-J., Fleuriet, A., Macheix, J.-J., 1986. Importance and evolution of phenolic compounds in olive during growth and maturation. J. Agric. Food Chem. 34, 823–826.
185
Bacelar, E., Moutinho-Pereira, J.M., Gonçalves, B.C., Ferreira, H., Correia, C.M., 2007. Changes in growth, gas exchange, xylem hydraulic properties and water use efficiency of three olive cultivars under contrasting water availability regimes. Environ. Exper. Bot. 60, 183–192. Beltrán, G., Aguilera, M.P., Del Rio, C., Sanchez, S., Martinez, L., 2005. Influence of fruit ripening process on the natural antioxidant content of Hojiblanca virgen olive oils. Food Chem. 89, 207–215. Bianco, A., Uccella, N., 2000. Biophenolic components of olives. Food Res. Inter. 33, 475–485. Briante, R., Patumi, M., Limongelli, S., Febbraio, F., Vaccaro, C., Di Salle, A., La Cara, F., Nucci, R., 2002. Changes in phenolic and enzymatic activities content during fruit ripening in two Italian cultivars of Olea europaea L. Plant Sci 162, 791–798. Covas, M.-I., 2007. Olive oil and the cardiovascular system. Pharmacol. Res. 55, 175–186. Dopico-García, M.S., Fique, A., Guerra, L., Afonso, J.M., Pereira, O., Valentão, P., Andrade, P.B., Seabra, R.M., 2008. Principal components of phenolics to characterize red Vinho Verde grapes: anthocyanins or non coloured compounds? Talanta 75, 1190–1202. Esti, M., Cinquant, L., La Notte, E., 1998. Phenolic compounds in different olive varieties. J. Agric. Food Chem. 46, 32–35. Gimeno, E., Castellote, A.I., Lamuela-Raventós, R.M., De La Torre, M.C., López-Sabater, M.C., 2002. The effect of harvest and extraction methods on the antioxidant content (phenolics, α-tocopherol and β-carotene) in virgin olive oil. Food Chem. 78, 207–211. Gong, F., Liang, Y.-Z., Xie, P., Chau, F.-T., 2003. Information theory applied to chromatographic fingerprint of herbal medicine for quality control. J. Chromat. A 1002, 25–40. Hatzopoulos, P., Banilas, G., Giannoulia, K., Gazis, F., Nikoloudakis, N., Milioni, D., Haralampidis, K., 2002. Breeding, molecular markers and molecular biology of the olive tree. Eur. J. Lipid Sci. Technol. 104, 574–586. Matos, L.C., Cunha, S.C., Amaral, J.S., Pereira, J.A., Andrade, P.B., Seabra, R.M., Oliveira, B.P.P., 2007. Chemometric characterization of three varietal olive oils (Cvs Cobrançosa, Madural and Verdeal Transmontana) extracted from olives with different maturation indices. Food Chem. 102, 406–414. Medeiros, D.M., 2001. Olive oil and health benefits. In: Wildman, R.E.C. (ed.), Handbook of Nutraceuticals and Functional Foods. CRC Press, London, pp. 261–266. Morelló, J.-R., Romero, M.-P., Ramo, T., Motilva, M.-J., 2005. Evaluation of L-phenilalanine ammonia-lyase activity and phenolic profile in olive drupe (Olea europaea L.) from fruit setting period to harvesting time. Plant Sci. 168, 65–72. Morelló, J.-R., Motilva, M.-J., Ramo, T., Romero, M.-P., 2003. Effect of freeze injuries in olive fruit on virgin olive oil composition. Food Chem 84, 547–553. Owen, C.A., Bita, E.-C., Banilas, G., Hajjar, S.E., Sellianakis, V., Aksoy, U., Hepaksoy, S., Chamoun, R., Talhook, S.N., Metzidakis, I., Hatzopoulos, P., Kalaitzis, P., 2005. AFLP reveals structural details of genetic diversity within cultivated olive germplasm from the Eastern Mediterranean. Theor. Appl. Genet. 110, 1169–1176. Pinheiro, P.B.M., Silva, E., 2005. Chemometric classification of olives from three Portuguese cultivars of Olea europaea L. Analyt. Chim. Acta 544, 229–235. Romani, A., Mulinacci, N., Pinelli, P., Vincieri, F., Cimato, A., 1999. Polyphenolic content in five Tuscany cultivars of Olea europaea L. J. Agric. Food Chem. 47, 964–967.
186
SECTION | I
Romero, M.P., Tovar, M.J., Girona, J., Motilva, M.J., 2002. Changes in the HPLC phenolic profile of virgin olive oil from young trees (Olea europeae L. Cv Arbequina) grown under different deficit irrigation strategies. J. Agric. Food Chem. 50, 5349–5354. Sarri, V., Baldoni, L., Porceddu, A., Cultrera, N.G.M., Contento, A., Frediani, M., Belaj, A., Trujillo, I., Cionini, P.G., 2006. Microsatellite markers are powerful tools for discriminating among olive cultivars and assigning them to geographically defined populations. Genome 49, 1615–1616. Tura Jr., D., Prenzler, P.D., Bedgood, D.R., Antolovich, M., Robards, K., 2004. Varietal and processing effects on the volatile profile of Australian olive oils. Food Chem. 84, 341–349.
Lipids, Phenolics and Other Organics and Volatiles
Vinha, A., Silva, B., Andrade, P., Seabra, R., Pereira, J.A., Oliveira, M.B.P.P., 2002. Development and evaluation of a HPLC/ DAD method for the analysis of phenolic compounds from olive fruits. J. Liq. Chrom. & Rel. Technol. 25, 151–160. Vinha, A.F., Ferreres, F., Silva, B.M., Valentão, P., Gonçalves, A., Pereira, J.A., Oliveira, M.B., Seabra, R.M., Andrade, P.B., 2005. Phenolic profiles of Portuguese olive fruits (Olea europaea L.): influence of cultivar and geographical origin. Food Chem. 89, 561–568.
Chapter 21
Low-level Free Phenols in Sicilian Olive Oils Marcello Saitta, Giuseppa Di Bella, Vincenzo Lo Turco, Giovanna Loredana La Torre and Giacomo Dugo Dipartimento di Scienze degli Alimenti e dell’Ambiente ‘G. Stagno d’Alcontres’, Università di Messina, Italy
21.1 INTRODUCTION The phenolic fraction of olive oils can be analyzed in different manners: the simple total quantitative determination is carried out by a colorimetric method and the single phenols qualitative and quantitative analysis by using chromatographic or electrophoretic methods. Many methods have been set by high-performance liquid chromatography (HPLC) (Tsimidou et al., 1992; Montedoro et al., 1992a, b, 1993; Akasbi et al., 1993; Mannino et al., 1993; Brenes et al., 1995; Cortesi et al., 1995; Tsimidou et al., 1996; Pirisi et al., 1997; Bianco et al., 1998; Owen et al., 2000); the quantification is simple and derivatizing reagents are not necessary, as in gas chromatography (Janer del Valle and Vazquez Roncero, 1980; Solinas, 1987; Romani et al., 2001). GC methods do however have some advantages with respect to HPLC: lower detection limits and better separations. Hyphenated methods like GC-MS or GC-MS/MS (Angerosa et al., 1995, 1996; Tasioula-Margari and Okogeri, 2001; Saitta et al., 2002) add a direct qualitative information and further sensitivity due to ion current analysis or selected ion monitoring. Even trace compounds can be detected and identified with this technique (Saitta et al., 2002); the aglycons of ligstroside and oleuropein can be detected too (Angerosa et al., 1995); only the whole glycosides, ligstroside and oleuropein, cannot be detected in GC. The importance of the analysis of the phenols is due to their antioxidant action (Montedoro, 1972; YanishlievaMaslarola, 1984; Papadopoulos and Boskou, 1991; Baldioli et al., 1996; Servili et al., 1996), but their amount depends on many factors, like the growth and maturation of the olives (Amiot et al., 1986; Alloggio and Caponio, 1997), the oil preparation (Alloggio et al., 1996; Catalano and Caponio, 1996), the variety and the climate (Solinas et al., 1978; Montedoro and Garofolo, 1984; Frega et al., 1997). The determination of low-level free phenols in olive oils is an aid to evaluate differences among samples from different varieties and some compounds can be related to Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
the autoxidation stages of olive oils. We herein report the chromatographic and spectroscopic approach to GC-MS and GC-MS/MS analysis and the results obtained on 76 Sicilian monovarietal virgin olive oils (six different cultivars), produced in the years 1998–2000.
21.2 METHODOLOGICAL CONSIDERATIONS 21.2.1 Standard Phenols Standards are not always commercially available; sometimes it is necessary to synthesize them. Thus, 4-(acetoxyethyl)1-hydroxybenzene can be synthesized according to the procedure of Posner and Oda (1981), and 3,4-dihydroxyphenylacetaldehyde can be synthesized according to the procedure of Fellman (1958) (Saitta et al., 2002).
21.2.2 Extraction Methods are essentially based on liquid–liquid or solid-phase extraction. None of the two methods guarantees a complete recovery of the phenols; authors often report conflicting data and thus the better way is always to evaluate own recoveries. One common liquid–liquid extraction employs methanol/ water (80:20 v/v) and then hexane to eliminate residues of triglycerides (Montedoro et al., 1992a).
21.2.3 Derivatization Bis(trimethylsilyl)trifluoroacetamide (BSTFA) or bis(trimethylsilyl)trifluoroacetamide – trimethylchlorosilane (BSTFA–TMCS 99:1), with or without solvent, are strong silylating agents. A reaction time of 30 minutes at room temperature is enough to obtain the trimethylsilyl derivatives in a complete way (Saitta et al., 2002).
187
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
188
SECTION | I
21.2.4 Gas Chromatography Different stationary phases are used in the GC analysis of the TMS derivatives of the phenols. A 5% diphenyl–95% dimethyl polysiloxane is a suitable stationary phase in this kind of analysis and a program of temperature of the column oven, sometimes very complex too, is essential to get a good separation of the aglycons (Angerosa et al., 1995). Figure 21.1 shows two chromatograms of the phenolic compounds detected in Sicilian virgin olive oils; peaks identification is reported in Table 21.1 (Saitta et al., 2002). Some interferences (fatty acids, monoglycerides) are always present in the chromatograms. Only the major peaks
Lipids, Phenolics and Other Organics and Volatiles
are easily identified: tyrosol, hydroxytyrosol and their esters with acetic acid and elenolic acid; usually, no other free phenol can be observed in the total ion chromatogram.
21.2.5 Mass Spectrometry The mass spectra of the phenols TMS derivatives usually show the molecular ion and a loss of 15 Da (methyl group). Losses of 30 Da (HCHO from methoxyphenyl derivatives), 60 Da (CH3COOH from acetoxyethylphenyl derivatives), 89 Da (-OTMS), 90 Da (HOTMS), 103 Da (-CH2OTMS from phenylethyl and benzyl derivatives) can also be
100 %
TOT
A
200 6:22
400 9:42
600 13:02
800 16:22
1000 19:42
Time (min)
200 6:23
400 9:43
600 13:03
800 16:23
1000 19:43
Time (min)
100 %
TOT
B
FIGURE 21.1 Chromatogram of the phenolic compounds (TMS derivatives) of a Nocellara del Belice sample (A) and of a Cerasuola sample (B) both produced in the year 1998. Peaks identification as in Table 21.1; I. S.: Internal Standard (sinapinic acid). This figure shows the differences between two Sicilian samples. Reprinted from Analytica Chimica Acta, Vol. 446, Saitta, M., Lo Curto, S., Salvo, F., Di Bella, G., and Dugo, G., Gas chromatographictandem mass spectrometric identification of phenolic compounds in Sicilian olive oils, pp. 335–344, ©, 2002, with permission from Elsevier.
189
CHAPTER | 21 Low-level Free Phenols in Sicilian Olive Oils
TABLE 21.1 Compounds detected in Sicilian olive oils produced in 1998, retention times, molecular weight and main daughters of M⫹. of the TMS derivatives. Compound (1) Salicylic acid (2-hydroxybenzoic acid) (2) Vanillin (3-methoxy-4-hydroxybenzaldehyde) (3) Tyrosol (4-hydroxyphenylethanol) (4) 4-(Acetoxyethyl)-1-hydroxybenzene (5) 4-Hydroxybenzoic acid (6) Vanillic alcohol (3-methoxy-4-hydroxybenzyl alcohol) (7) 3,4-Dihydroxyphenylacetaldehyde (8) Homovanillic alcohol (3-Methoxy4-hydroxyphenylethanol) (9) Syringaldehyde (3,5-dimethoxy4-hydroxybenzaldehyde) (10) Hydroxytyrosol (3,4-dihydroxyphenylethanol) (11) Vanillic acid (3-methoxy-4-hydroxybenzoic acid) (12) 2-Coumaric acid (trans-2-hydroxycinnamic acid) (13) 4-(Acetoxyethyl)-1,2-dihydroxybenzene (14) Protocatechuic acid (3,4-dihydroxybenzoic acid) (15) Syringic acid (3,5-dimethoxy-4-hydroxybenzoic acid) (16) cis-Ferulic acid (cis-3-methoxy-4-hydroxycinnamic acid) (17) 4-Coumaric acid (trans-4-hydroxycinnamic acid) (18) Ferulic acid (trans-3-methoxy-4-hydroxycinnamic acid) (19) Caffeic acid (trans-3,4-dihydroxycinnamic acid) (20) Ligstroside aglycon – dialdehydic forma (21) Ligstroside aglycon (elenolic acid linked to tyrosol)a (22) Dialdehydic form of elenolic acid linked to 3-methoxy-4-hydroxyphenylethanola,b (23) Oleuropein aglycon – dialdehydic forma (24) Oleuropein aglycon (elenolic acid linked to hydroxytyrosol)a
RT(m)
MW
Daughters of M⫹.
Confirmation
7:39
282
267, 193
Standard
7:57
224
209, 194
Standard
8:12
282
267, 193, 179
Standard
8:34
252
237, 192, 177
Synthesized
8:38
282
267, 193
Standard
8:42
298
283, 268, 253
Standard
8:44
296
281, 267, 179
Synthesized
9:16
312
297, 282, 267
Standard
9:20
254
239, 224
Standard
9:38
370
355, 267, 193, 179
Standard
9:42
312
297, 282, 267
Standard
9:53
308
293, 219
Standard
10:02
340
325, 280, 265, 193, 179
Reference spectrum
10:05
370
355, 193
Standard
10:37
342
327, 312, 297
Standard
10:48
338
323, 308, 293
Standard
10:58
308
293, 219
Standard
11:59
338
323, 308, 293
Standard
12:15
396
381, 219
Standard
14:44
376
361, 192, 177
Reference spectrum
15:01
448
433, 358, 343, 192, 177
Reference spectrum
15:36
406
391, 222, 207, 192
—
15:54
464
449, 280, 265, 193, 179
Reference spectrum
16:12
536
521, 446, 280, 265, 193, 179
Reference spectrum
This table summarizes the characteristics of the phenolic compounds detected: the elution order and time, some ions in the daughters’ analysis of molecular ions, the method used to confirm the identification. Reprinted from Saitta, M., Lo Curto, S., Salvo, F., Di Bella, G., Dugo, G., 2002. Gas chromatographic-tandem mass spectrometric identification of phenolic compounds in Sicilian olive oils. Anal. Chim. Acta. 446, 335–344. ©, with permission from Elsevier. a With no carbomethoxy group; b tentative assignment.
190
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
0.34%
TOT
0.00% 1
267 + 193
240 7:03
260 7:23
280 7:43
(min)
300 8:03
FIGURE 21.2 Total ion chromatogram and ion current at m/z 267 ⫹ 193 of a 1998 Santagatese sample. Peak identification as in Table 21.1. Example of difficult detection of salicylic acid.
observed. Typical intense ions occur at m/z 73 [(CH3)3Si]⫹, 179 [TMSOC6H4CH2]⫹ and 267 [(TMSO)2C6H3CH2]⫹; ions at m/z 192 and 280 originate from a McLafferty rearrangement typical of β-phenylethyl esters (Angerosa et al., 1995). The phenolic compounds found in olive oils are aromatic alcohols, aldehydes, acids and esters; in these substances, the first phenolic group is commonly in the para position and a methoxy group, if present, is in the meta position. In Sicilian olive oils, an accurate selection of the spectra allows to identify up to 23 phenolic compounds and to hypothesize a further one (Saitta et al., 2002). Identification method of unknown compounds consists of a few steps. 1. Spectral examination. All the spectra in a sample data file are examined; the spectra with interesting features are selected and reported. Ion current chromatograms are eventually extracted from the full scan data and sample minus background spectra are obtained to improve spectral quality. The same work is done for other sample files: same spectral data found in more samples are chosen. 2. Spectral interpretation. Spectra with characteristic features are analyzed to individuate a possible structure. 3. Comparison with standard or synthesized compounds. Standard compound is purchased or synthesized to compare retention time and spectrum. The detection of minor compounds can be done in ion current, if the phenols have known spectra and retention times. MS/MS experiments (daughters’ analysis of molecular ion) are useful to verify the correct identification in
doubtful cases: the daughter ions must be at the expected m/z values (Saitta et al., 2002). The ion current detection also increases the sensitivities, allowing to reach values better than 0.1 mg kg⫺1.
21.3 SELECTED ION ANALYSIS FROM THE FULL SCAN DATA: A POWERFUL AND VERSATILE TOOL TO DETECT AND IDENTIFY PHENOLS The full scan data are essential to evaluate the presence of unknown compounds. Selected ion analysis is necessary to detect low concentrations. Thus, a good compromise to obtain both whole qualitative information and excellent sensitivity is to extract the required ion current values from the total ion chromatogram. The first compound in elution order is salicylic acid (peak 1). It has a very low molecular ion and its spectrum is always hidden by background; main fragments occur at m/z 267 and 193 (loss of .OSi(CH3)3). It is one of the more difficult compounds to detect; even in ion current, the corresponding chromatographic peak cannot be clearly shown (Figure 21.2). Vanillin (peak 2) shows a characteristic and simple spectrum and can be detected even in the total ion chromatogram (Figure 21.3). Molecular ion is at m/z 224 and the main daughters are at m/z 209 and 194 (loss of HCHO from the group Ph-OCH3). Tyrosol (peak 3) is one of the largest compounds detected, its spectrum shows ions at m/z 282 (M⫹.), 267, 193, 179 (base peak) and 73.
191
CHAPTER | 21 Low-level Free Phenols in Sicilian Olive Oils
12%
194
75
TOT
3
224 299
401
0.04% 2
224 + 194
260 7:23
270 7:33
280 7:43
290 7:53
300 8:03
310 8:13
320 8:23 (min)
FIGURE 21.3 Total ion chromatogram, ion current at m/z 224 ⫹ 194 and vanillin spectrum (TMS der.) of a 1998 Cerasuola sample. Peaks identification as in Table 21.1. Easy detection of vanillin.
11%
4
TOT
0.03% 5
282 + 267
320 8:23
330 8:33
340 8:43
350 8:53
(min)
FIGURE 21.4 Total ion chromatogram and ion current at m/z 282 ⫹ 267 of a 1998 Santagatese sample. Peaks identification as in Table 21.1. Detection of 4-hydroxybenzoic acid.
The spectrum of 4-(acetoxyethyl)-1-hydroxybenzene (peak 4) does not show a significative molecular ion. The even base peak (m/z 192) results from a neutral loss; a weak ion at m/z 237 is due to a methyl group loss. To confirm the presence of 4-(acetoxyethyl)-1-hydroxybenzene in Sicilian sample, it was necessary to synthesize the compound (Saitta et al., 2002). 4-Hydroxybenzoic acid (peak 5) can be partially hidden by 4-(acetoxyethyl)-1-hydroxybenzene: ion current detection at m/z 282 ⫹ 267 (Figure 21.4) or MS/MS is required to detect this compound.
Vanillic alcohol spectrum (peak 6) has characteristic features, with a molecular ion at m/z 298 (base peak) and daughter ions at m/z 283, 268 and 253. Its low concentration and the presence of the close peak 7 require the use of ion current analysis (Figure 21.5) or MS/MS at m/z 298 to detect the compound. The spectrum of peak 7 presents a molecular ion at m/z 296; the base peak at m/z 267 shows the loss of 29 Da. An intense ion is at m/z 179. MS/MS performed on m/z 296 shows the main daughters at m/z 281, 267 and 179. We supposed that this compound was similar to hydroxytyrosol for the
192
SECTION | I
1.32%
Lipids, Phenolics and Other Organics and Volatiles
4
TOT
0.01% 6
298 + 283
330 8:33
340 8:43
350 8:53
360 9:03 (min)
FIGURE 21.5 Total ion chromatogram and ion current at m/z 298 ⫹ 283 of a 1998 Cerasuola sample. Peaks identification as in Table 21.1. Detection of vanillic alcohol.
2.51%
4
TOT
0.02% 7
296 + 267
5
320 8:23
330 8:33
340 8:43
350 8:53
(min)
FIGURE 21.6 Total ion chromatogram and ion current at m/z 296 ⫹ 267 of a 1998 Nocellara del Belice sample. Peaks identification as in Table 21.1. Detection of 3,4-dihydroxyphenylacetaldehyde.
typical benzyl ions at m/z 179 and 267; the loss of 29 Da could be due to the presence of an alkylic aldehyde. We observed that the synthesized 3,4-dihydroxyphenylacetaldehyde showed the same retention time and spectrum of peak 9 (Saitta et al., 2002). In Figure 21.6 is shown the detection in ion current at the values 296 ⫹ 267: the peak of 4-hydroxybenzoic acid is visible too, because of the common ion at m/z 267. Homovanillic alcohol (peak 8) is easily identified by its characteristic spectrum. An ion current analysis can be performed at m/z 312 ⫹ 209 (Figure 21.7).
The spectrum of peak 9 shows a molecular ion at m/z 254 and fragments at m/z 239 and 224 (base peak). This spectrum looks like the vanillin spectrum and the difference in molecular weight could be due to the presence of an additional methoxy group. Standard syringaldehyde (3,5-dimethoxy-4hydroxybenzaldehyde) has retention time and spectrum identical to peak 9 (Saitta et al., 2002). A routine detection can be done in ion current at m/z 254 ⫹ 224 (Figure 21.8). Hydroxytyrosol (peak 10) is one of the largest compounds detected and is easily identified. In its spectrum,
193
CHAPTER | 21 Low-level Free Phenols in Sicilian Olive Oils
4.50%
209
4 73
312
179
TOT
345 419 488
0.12%
8
312 + 209
330 8:33
340 8:43
350 8:53
360 9:03
370 9:13
380 9:23
390 9:33 (min)
FIGURE 21.7 Total ion chromatogram, ion current at m/z 312 ⫹ 209 and spectrum of homovanillic alcohol (TMS der.) of a 1998 Nocellara del Belice sample. Peaks identification as in Table 21.1. Easy detection of homovanillic alcohol.
2.63%
4
TOT
0.01% 9
254 + 224
320 8:23
340 8:43
360 8:63
380 9:23
(min)
FIGURE 21.8 Total ion chromatogram and ion current at m/z 254 ⫹ 224 of a 1998 Cerasuola sample. Peaks identification as in Table 21.1. Detection of syringaldehyde.
intense ions occur at m/z 370 (molecular ion), 267 (base peak), 193 and 179. Vanillic acid (peak 11) has a retention time close to hydroxytyrosol, but an ion current analysis can detect it. In Figure 21.9 is shown the detection in ion current at the values 312 ⫹ 297: the peak of homovanillic alcohol is visible too, because of the common ions at m/z 312 and 297. 2-Coumaric acid (peak 12) spectrum is usually hidden by background, so an ion current analysis (Figure 21.10)
or an MS/MS experiment on molecular ion (m/z 308) is required to detect this phenol. 4-(Acetoxyethyl)-1,2-dihydroxybenzene (peak 13) spectrum shows the molecular ion at m/z 340 and fragments at m/z 325, 280 (loss of acetic acid), 265, 193, 179 and 73 (base peak). This spectrum can be easily observed in the total ion chromatogram. Protocatechuic acid (peak 14) is very close to peak 13 and usually hidden by it: MS/MS on m/z 370 (molecular
194
SECTION | I
8.11%
Lipids, Phenolics and Other Organics and Volatiles
10
TOT
13
0.28% 11
312 + 297 8
360 9:03
380 9:23
400 9:43
420 10:03
440 10:23 (min)
FIGURE 21.9 Total ion chromatogram and ion current at m/z 312 ⫹ 297 of a 1998 Santagatese sample. Peaks identification as in Table 21.1. Detection of vanillic acid.
8.11%
10
TOT
13
0.04% 17
308 + 293
380 9:23
12
400 9:43
420 10:03
440 10:23
460 10:23
480 10:43
500 11:23 (min)
FIGURE 21.10 Total ion chromatogram and ion current at m/z 308 ⫹ 293 of a 1998 Santagatese sample. Peaks identification as in Table 21.1. Detection of two coumaric acids.
ion) can show the major daughter ions at the m/z values of 355 and 193. A routine detection can be done in ion current at m/z 370 ⫹ 355 (Figure 21.11): the base peak (m/z 193) cannot be used, because this ion is present in the spectrum of 4-(acetoxyethyl)-1,2-dihydroxybenzene too. Syringic acid (peak 15) spectrum is usually hidden by background; MS/MS on molecular ion (m/z 342) is able to detect the characteristic daughters at m/z 327, 312 and 297. A routine detection can be done in ion current at m/z 342 ⫹ 327 (Figure 21.12).
Peak 16 is always hidden by background too. The spectrum shows a molecular ion at m/z 338 and fragments at m/z 323, 308 and 293. This spectrum is similar to that of trans ferulic acid, but with a different retention time. We hypothesized that the compound was cis ferulic acid and, in fact, it showed the same retention time and spectrum (Saitta et al., 2002). A routine detection can be done in ion current at m/z 338 ⫹ 323 (Figure 21.13). 4 -Coumaric acid (peak 17) can be detected and identified even in the total ion chromatogram. A routine
195
CHAPTER | 21 Low-level Free Phenols in Sicilian Olive Oils
58%
13
TOT
0.02% 14
370 + 355
410 9:53
420 10:03
430 10:13
440 10:23
450 10:33
(min)
FIGURE 21.11 Total ion chromatogram and ion current at m/z 370 ⫹ 355 of a 1998 Nocellara del Belice sample. Peaks identification as in Table 21.1. Detection of protocatechuic acid.
4.77%
TOT
0.01% 15
342 + 327
440 10:23
460 10:43
480 11:03
500 11:23
(min)
FIGURE 21.12 Total ion chromatogram and ion current at m/z 342 ⫹ 327 of a 1998 Cerasuola sample. Peak identification as in Table 21.1. Detection of syringic acid.
detection can be done in ion current at m/z 308 ⫹ 293 (Figure 21.10). Trans ferulic acid (peak 18) is not easily detected in total ion chromatogram. In ion current the characteristic ions at m/z 338 and 323 can show its presence (Figure 21.13). For caffeic acid (peak 19) ion current analysis can be necessary. The ions at m/z 396 (molecular ion) and 219 are characteristic for this compound (Figure 21.14). Aglycons appear late in the chromatogram. Aglycons of tyrosol and hydroxytyrosol with elenolic acid are well
known: these esters are usually in equilibrium between open dialdehydic and closed alcoholic forms (Angerosa et al., 1995). In our studies we found only decarbomethoxy aglycons in the Sicilian samples: in Figure 21.15 is reported part of a chromatogram of a sample. The total ion current and the current at the m/z values of 192 (typical of the ligstroside aglycons, esters of tyrosol) and 280 (typical of the oleuropein aglycons, esters of hydroxytyrosol) are shown. Aglycons are identified comparing literature and samples spectra (Angerosa et al., 1995).
196
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
10%
FA
TOT
FA 13
0.01%
18
338 + 323
16
450 10:33
500 11:23
550 12:13
600 13:03
(min)
FIGURE 21.13 Total ion chromatogram and ion current at m/z 338 ⫹ 323 of a 1998 Santagatese sample. FA ⫽ Fatty acid. Peaks identification as in Table 21.1. Detection of two ferulic acids.
10%
FA
TOT
FA 13
0.02% 19
396 + 219
450 10:33
500 11:23
550 12:13
600 13:03
(min)
FIGURE 21.14 Total ion chromatogram and ion current at m/z 396 ⫹ 219 of a 1998 Santagatese sample. FA ⫽ Fatty acid. Peaks identification as in Table 21.1. Detection of caffeic acid.
Peak 20 has a very low molecular ion at m/z 376 and fragments at m/z 361, 192 and 177; it is the decarbomethoxy ligstroside aglycon in the dialdehydic form. Peak 21 has a low molecular ion at m/z 448 and fragments at m/z 433, 358, 192 and 177; it is the decarbomethoxy ligstroside aglycon. Peak 22 shows a molecular ion at m/z 406 and fragments at m/z 222, 207 and 192. The fragmentation led to a β-phenyl ethyl ester: the dialdehydic form of decarbomethoxy elenolic acid linked to 3-methoxy-4-hydroxyphenylethanol satisfied all the features. This hypothesis is
also supported by the presence of the free alcohol (peak 8). Unfortunately, a confirmation could not be obtained, so this structural assignment is tentative (Saitta et al., 2002). The typical ion at m/z 222 can detect this compound (Figure 21.15). Peak 23 shows a molecular ion at m/z 464 and fragments at m/z 280, 193 and 179; it is the decarbomethoxy oleuropein aglycon in the dialdehydic form. Peak 24 shows a molecular ion at m/z 536 and fragments at m/z 521, 446, 280, 193 and 179; it is the decarbomethoxy oleuropein aglycon.
197
CHAPTER | 21 Low-level Free Phenols in Sicilian Olive Oils
71% TOT 22% 20 192 21 0.33% 22 222 15% 23 280 24 600 13:03
650 13:53
700 14:43
750 15:33
800 16:23
850 900 16:23 (min) 18:03
FIGURE 21.15 Total ion chromatogram and ion currents at m/z 192, 222 and 280 of a 1998 Nocellara del Belice sample. Peaks identification as in Table 21.1. Detection of the aglycons: tyrosol esters m/z 192, homovanillic alcohol ester m/z 222, hydroxytyrosol esters m/z 280.
21.4 PHENOLIC CONTENTS IN SICILIAN OLIVE OILS The characteristics of the monovarietal Sicilian virgin olive oil samples are shown in Table 21.2. The quantitation of each compound was performed with an internal standard (sinapinic acid) in ion current, using two selected ions and calibration curves for all the standards; for the quantitation of the aglycons, tyrosol and hydroxytyrosol were used instead of the respective esters, because the aglycon standards were not available (Saitta et al., 2002; Dugo et al., 2004). In Table 21.3 the total amount of the phenols detected in the Sicilian samples is reported. Some considerations about the differences among the varieties and the production years can be done. In the year 1998, the mean of the total values was 390 mg kg⫺1 for the Nocellara del Belice samples; the Santagatese and the Cerasuola samples reached only 130 and 117 mg kg⫺1, respectively. These differences are not simply due to the varieties’ characteristics. In fact, different olive varieties can have different qualitative and quantitative phenol composition, but it is important to evaluate the compounds evolution during the fruits maturation: the glycosides oleuropein and ligstroside are predominant in olive fruits and their quantities decrease with maturation (Amiot et al., 1986); aglycon and total phenol concentrations decrease with maturation too (Brenes et al., 1999). By taking a look at the year 1999 and 2000 levels, it is possible to see that the Cerasuola samples reached a total value of 328 and 271 mg kg⫺1, respectively: the low-content variety factor can be excluded. The differences in the years can be explained in terms of olive maturation: in the 1998 Cerasuola samples, even the total aglycon levels were low (46 mg kg⫺1). These
TABLE 21.2 Sicilian virgin olive oils examined. Variety
Production year
Number Abbreviation of samples
Nocellara del Belice 1998
6
No98
Santagatese
1998
5
Sa98
Cerasuola
1998
5
Ce98
Nocellara del Belice 1999
6
No99
Biancolilla
1999
23
Bi99
Cerasuola
1999
4
Ce99
Tonda Iblea
1999
3
To99
Nocellara del Belice 2000
10
No00
Biancolilla
2000
4
Bi00
Cerasuola
2000
8
Ce00
Crastu
2000
2
Cr00
Cultivar names and samples analyzed in the years 1998–2000.
low levels can be found in oils from overripe fruits, when already fallen olives are picked up or mature fruits are left for some time before being used for oil production. The same considerations can be done for the Santagatese samples, even if other data are unavailable. Also in this case, total aglycon mean is only 60 mg kg⫺1.
198
SECTION | I
The Nocellara del Belice variety always shows high phenols contents: 390 mg kg⫺1 in 1998, 277 mg kg⫺1 in 1999 and 257 mg kg⫺1 in 2000. Also the total aglycon means are high: 324 mg kg⫺1 in 1998, 237 mg kg⫺1 in 1999 and 194 mg kg⫺1 in 2000. These values demonstrate good practice in the harvest of the olives and in the oils production. The few samples of Tonda Iblea and Crastu do not permit speculation on the matter; the first appear as a low total content and low aglycons content samples, while the second have values of 202 mg kg⫺1 (mean of total phenols) and 160 mg kg⫺1 (mean of total aglycons). Great differences are in the 1999 and 2000 samples of Biancolilla. The samples produced in 1999 have total phenol contents of 133 mg kg⫺1 and total aglycon contents of 63 mg kg⫺1, while in 2000 the levels are 239 and 173 mg kg⫺1, respectively. This points out an improvement in the oil quality production. An indication of the ripeness factor can be done considering the ratio total aglycons/total phenols. As is possible to see in Table 21.3, these values are ⬎0.7 when the oils have good characteristics in terms of correct maturation of the olives; values ⬍0.6 can indicate incorrect practices in olive harvest and oil production. Phenols type and amount are directly related to the oil autoxidation resistance. Every phenolic compound has different antioxidant characteristics; in particular, the
Lipids, Phenolics and Other Organics and Volatiles
ortho-diphenols like caffeic acid, protocatechuic acid, 3,4dihydroxyphenylacetic acid and hydroxytyrosol possess the best antioxidative qualities (Blekas et al., 1995; Fki et al., 2005). Among these compounds, only hydroxytyrosol and its esters are always present in significant quantities. A measure of these substances can give indications on the autoxidation stage of the samples: high amounts of total hydroxytyrosol (free alcohol ⫹ its esters) indicate good protection against oil oxidation. Thus, in the Nocellara del Belice samples, the total hydroxytyrosol quantities are 175 mg kg⫺1 in 1998, 128 mg kg⫺1 in 1999 and 108 mg kg⫺1 in 2000 (Table 21.3). These quantities are high enough to guarantee good resistance to the autoxidation, for a correctly preserved oil. The Biancolilla samples increase from 15 mg kg⫺1 in 1999 to 107 mg kg⫺1 in 2000; for the Cerasuola samples the values are 52 mg kg⫺1 in 1998, 133 mg kg⫺1 in 1999 and 104 mg kg⫺1 in 2000; the samples of Santagatese, Tonda Iblea and Crastu reach 57, 39 and 51 mg kg⫺1, respectively. In this case, all the values lower than 60 mg kg⫺1 can be considered a sign of badly preserved samples, but a more complete evaluation of the autoxidation risk must consider also the total phenol amount: a useful parameter can be the ratio total hydroxytyrosol/total phenols (Table 21.3). Thus, the ratios for the Nocellara del Belice samples range from a minimum of 0.42 in 2000 to a maximum of 0.46 in 1999: this demonstrates the
TABLE 21.3 Amounts of the phenols (mg kg⫺1) in the Sicilian virgin olive oils examined. Abbreviation
Total phenols (mean ⫾ s.d.)
Total aglycons (mean ⫾ s.d.)
R1
Total hydroxytyrosol (mean ⫾ s.d.)
R2
No98
390 ⫾ 58
324 ⫾ 46
0.83
175 ⫾ 28
0.45
Sa98
130 ⫾ 38
60 ⫾ 23
0.46
57 ⫾ 19
0.44
Ce98
117 ⫾ 30
46 ⫾ 14
0.39
52 ⫾ 11
0.44
No99
277 ⫾ 106
237 ⫾ 83
0.86
128 ⫾ 48
0.46
Bi99
133 ⫾ 22
63 ⫾ 24
0.47
15 ⫾ 18
0.11
Ce99
328 ⫾ 108
263 ⫾ 92
0.80
133 ⫾ 48
0.41
To99
137 ⫾ 33
79 ⫾ 26
0.58
39 ⫾ 22
0.28
No00
257 ⫾ 63
194 ⫾ 61
0.75
108 ⫾ 37
0.42
Bi00
239 ⫾ 82
173 ⫾ 62
0.72
107 ⫾ 47
0.45
Ce00
271 ⫾ 52
211 ⫾ 54
0.78
104 ⫾ 15
0.38
Cr00
202 ⫾ 13
160 ⫾ 11
0.79
51 ⫾ 8
0.25
s. d.: standard deviation; R1 ⫽ total aglycons/total phenols; R2 ⫽ total hydroxytyrosol/total phenols. This table summarizes the differences among the varieties and the production years. High values of the ratios R1 and R2 show the samples with the best qualitative characteristics.
CHAPTER | 21 Low-level Free Phenols in Sicilian Olive Oils
good and constant characteristics of these oils. The Cerasuola samples have ratios of 0.44 in 1998, 0.41 in 1999 and 0.38 in 2000, and this means that a relatively good protection against autoxidation is present in the 1998 samples too. The same conclusion can be reached for the Santagatese samples, with a total hydroxytyrosol amount of 57 mg kg⫺1, but a ratio of 0.44. The Biancolilla samples have ratios of 0.45 in 2000, but only 0.11 in 1999. This is further demonstration of badly preserved samples, probably exposed to oxygen and/or to sources of heat or sunlight during the phases of production and storage.
SUMMARY POINTS ●
●
●
●
●
Free phenols in Sicilian virgin olive oils can be analyzed by gas chromatography-mass spectrometry (GC-MS). Mass spectra permit the identification of 23 different compounds. Four phenols were identified in olive oils for the first time and a further compound was hypothesized. Low-level compounds can be detected in ion current analysis. Monovarietal oils produced in the years 1998–2000 from six different cultivars show quantitative differences based on olive maturation, variety and oxidation stage.
REFERENCES Akasbi, M., Shoeman, D.W., Saari Csallany, A., 1993. High performance liquid chromatography of selected phenolic compounds in olive oils. J. Am. Oil Chem. Soc. 70, 367–370. Alloggio, V., Caponio, F., 1997. Influenza delle tecniche di preparazione della pasta di olive sulla qualità dell’olio. Nota 1. Evoluzione delle sostanze fenoliche e di alcuni parametri di qualità in funzione della maturazione delle drupe in olio di oliva vergine della cv. Coratina. Riv. Ital. Sostanze Grasse 74, 443–447. Alloggio, V., Caponio, F., De Leonardis, T., 1996. Influenza delle tecniche di preparazione della pasta di olive sulla qualità dell’olio. Nota 1. Profilo quali-quantitativo delle sostanze fenoliche mediante HPLC, in olio di oliva vergine della cv. Ogliarola Salentina. Riv. Ital. Sostanze Grasse 73, 355–360. Amiot, M.T., Fleuriet, A., Macheix, J.T., 1986. Importance and evolution of phenolic compounds in olive during growth and maturation. J. Agric. Food Chem. 34, 823–826. Angerosa, F., d’Alessandro, N., Konstantinou, P., Di Giacinto, L., 1995. GC-MS evaluation of phenolic compounds in virgin olive oil. J. Agric. Food Chem. 43, 1802–1807. Angerosa, F., d’Alessandro, N., Corana, F., Mellerio, G., 1996. Characterization of phenolic and secoiridoid aglycons present in virgin olive oil by gas chromatography-chemical ionization mass spectrometry. J. Chromatogr. A 736, 195–203. Baldioli, M., Servili, M., Perretti, G., Montedoro, G., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oil. J. Am. Oil Chem. Soc. 73, 1589–1593.
199
Bianco, A., Mazzei, R.A., Melchioni, C., Romeo, G., Scarpati, M.L., Soriero, A., Uccella, N., 1998. Microcomponents of olive oil – III. Glucosides of 2-(3,4-dihydroxy-phenyl)-ethanol. Food Chem. 63, 461–464. Blekas, G., Tsimidou, M., Boskou, D., 1995. Contribution of α-tocopherol to olive oil stability. Food Chem. 52, 289–294. Brenes, M., Garcia, A., Garcia, P.G., Rios, J.J., Garrido, A., 1999. Phenolic compounds in Spanish olive oils. J. Agric. Food Chem. 47, 3535–3540. Catalano, P., Caponio, F., 1996. Machines for olive paste preparation producing quality virgin olive oil. Fett. Lipid 98, 408–412. Cortesi, N., Azzolini, M., Rovellini, P., Fedeli, E., 1995. I componenti minori polari degli oli vergini di oliva: ipotesi di struttura mediante LC-MS. Riv. Ital. Sostanze Grasse 72, 241–251. Dugo, G., Lo Turco, V., Pollicino, D., Mavrogeni, E., Pipitone, F., 2004. Caratterizzazione degli oli di oliva cergini siciliani. Variazione qualitativa di oli da cv ‘Biancolilla’, ‘Nocellara del Belice’, ‘Cerasuola, ‘Tonda Iblea’ e ‘Crastu’ in funzione delle tecniche e del periodo di raccolta delle olive. Olivae 101, 44–52. Fellman, J.H., 1958. Rearrangement of adrenaline. Nature 182, 311–312. Fki, I., Allouche, N., Sayadi, S., 2005. The use of polyphenolic extract, purified hydroxytyrosol and 3,4-dihydroxyphenyl acetic acid from olive mill wastewater for the stabilization of refined oils: a potential alternative to synthetic antioxidants. Food Chem. 93, 197–204. Frega, N., Caglioti, L., Mozzon, M., 1997. Chemical composition and quality parameters of oils from stoned olives. Riv. Ital. Sostanze Grasse 74, 241–245. Janer del Valle, C., Vazquez Roncero, A., 1980. Estudio de los componentes polares del aceite de oliva por cromatografia gaseosa. Grasas Aceites 31, 309–316. Mannino, S., Cosio, M.S., Bertuccioli, M., 1993. High performance liquid chromatography of phenolic compounds in olive oils using amperometric detection. Ital. J. Food Chem. 4, 363–370. Montedoro, G., 1972. Phenolic substances present in virgin olive oil. Note I. Identification of phenolic acids and their antioxidant power. Sci. Tecnol. Aliment. 2, 177–186. Montedoro, G., Garofolo, L., 1984. Caratteristiche quali-quantitative degli oli vergini di oliva. Influenza di alcune variabili: varietà, ambiente, conservazione, estrazione, condizionamento del prodotto finito. Riv. Ital. Sostanze Grasse 61, 157–168. Montedoro, G., Servili, M., Baldioli, M., Miniati, E., 1992a. Simple and hydrolizable compounds in virgin olive oil, 1. Their extraction, separation and quantitative and semiquantitative evaluation by HPLC. J. Agric. Food Chem. 40, 1571–1576. Montedoro, G., Servili, M., Baldioli, M., Miniati, E., 1992b. Simple and hydrolizable compounds in virgin olive oil, 2. Initial characterization of the hydrolizable fraction. J. Agric. Food Chem. 40, 1577–1580. Montedoro, G., Servili, M., Baldioli, M., Selvaggini, R., Miniati, E., Macchioni, A., 1993. Simple and hydrolizable compounds in virgin olive oil, 3. Spectroscopic characterization of the secoiridoid derivatives. J. Agric. Food Chem. 41, 2228–2234. Owen, R.W., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2000. Identification of lignans as major components in the phenolic fraction of olive oil. Clin. Chem. 46, 976–988. Papadopoulos, G., Boskou, D., 1991. Antioxidant effect of natural phenols in olive oil. J. Am. Oil Chem. Soc. 68, 669–671. Pirisi, F.M., Angioni, A., Cabras, P., Garau, V.L., Sanjust di Teulada, M.T., Kaim dos Santos, M., Bandino, G., 1997. Phenolic compounds in virgin olive oils – I. Low-wavelength quantitative determination of complex phenols by high-performance liquid chromatography under isocratic elution. J. Chromatogr. A 768, 207–213.
200
SECTION | I
Posner, G.H., Oda, M., 1981. Organic reactions at alumina surfaces. An extremely simple, convenient and selective method for acetylating primary alcohols in the presence of secondary alcohols. Tetrahedron Lett. 22, 5003–5006. Romani, A., Pinelli, P., Mulinacci, N., Galardi, C., Vincieri, F.F., Liberatore, L., Cichelli, A., 2001. HPLC and HRGC analyses of polyphenols and secoiridoid in olive oil. Chromatographia 53, 279–284. Saitta, M., Lo Curto, S., Salvo, F., Di Bella, G., Dugo, G., 2002. Gas chromatographic-tandem mass spectrometric identification of phenolic compounds in Sicilian olive oils. Anal. Chim. Acta 446, 335–344. Servili, M., Baldioli, M., . Miniati, E., Montedoro, G., 1996. Antioxidant activity of new phenolic compounds extracted from virgin olive oil and their interaction with α-tocopherol and β-carotene. Riv. Ital. Sostanze Grasse 73, 55–59. Solinas, M., 1987. Analisi HRGC delle sostanze fenoliche di oli vergini di oliva in relazione al grado di maturazione e alla varietà delle olive. Riv. Ital. Sostanze Grasse 64, 255–262.
Lipids, Phenolics and Other Organics and Volatiles
Solinas, M., Di Giovacchino, L., Mascolo, A., 1978. The polyphenols of olives and olive oil. Note III. Influence of temperature and kneading time and their polyphenols content. Riv Ital. Sostanze Grasse 55, 19–23. Tasioula-Margari, M., Okogeri, O., 2001. Isolation and characterization of virgin olive oil phenolic compounds by HPLC/UV and GC-MS. J. Food Sci. 66, 530–534. Tsimidou, M., Papadopoulos, G., Boskou, D., 1992. Phenolic compounds and stability of virgin olive oil. Part 1. Food Chem. 45, 141–144. Tsimidou, M., Lytridou, M., Boskou, D., Pappa-Louisi, A., Kotsifaki, F., Petrakis, C., 1996. On the determination of minor phenolic acids of virgin olive oils by RP-HPLC. Grasas Aceites 47, 151–157. Yanishlieva-Maslarola, N.V., 1984. On the antioxidative effect of some polyphenols in lipid systems. Bull. Liois. Groupe Polyphenols 11, 470–480.
Chapter 22
Volatile Compounds in Australian Olive Oils: How Different Are They From Other Oils Curtis Kalua1, Paul Prenzler2, Danielle Ryan2 and Kevin Robards2 1 2
CSIRO Plant Industry and Food Futures Flagship, Glen Osmond, SA, Australia School of Agricultural and Wine Sciences, EH Graham Centre, Charles Sturt University, Wagga Wagga, NSW, Australia
22.1 INTRODUCTION The term ‘volatile compound’ is used differently in various areas and even within a single disciplinary area. For example, US environmental legislation defines a volatile organic compound in terms of its reactivity, whereas corresponding European legislation is based on evaporation into the atmosphere. In food science, the theoretical underpinning of any definition is more closely related to the concept of evaporation. Thus, in the present context we may regard a volatile compound as one that has a high vapor pressure and that easily vaporizes at normal temperature and pressure. The relationship to evaporation links the concept of a volatile compound to aroma and flavor (Jimenez et al., 2006). Nevertheless, the definition of a volatile compound in food science is largely methodological and as gas chromatography (GC) is historically the most significant technique for measurement of volatiles, by default ‘volatiles’ has come to mean compounds most amenable to measurement by GC techniques. At the present time, this will mean that the definition encompasses a range of compounds that includes hydrocarbons (Vichi et al., 2007b), aldehydes and ketones, alcohols, esters; furans and carboxylic acids (Luna et al., 2006). Some compounds such as hydroxytyrosol are sufficiently volatile to be measured by GC but are not traditionally included in the definition of volatile compounds (Servili et al., 2007b). This is presumably related to the fact that they are more amenable to measurement by HPLC than GC. The volatile component of olive oil is a complex mixture of chemicals that determines the oil’s aroma (Bianco et al., 2006) and influences quality and consumer acceptability. The relationship between number and quantity Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
of volatiles and the aroma (Garcia-Gonzalez et al., 2007) is not simple and depends on synergistic and antagonistic effects. Moreover, exogenous volatile compounds such as aromatic hydrocarbons and halogenated solvent residues may occur in an edible oil such as olive oil consequential to the extraction procedure (Arrebola et al., 2005; CarrilloCarrion et al., 2007). The situation is complicated as many volatile compounds and particularly hydrocarbons can be present in virgin olive oils either naturally or as contaminants (Pena et al., 2004). Moreover, volatile compounds may also be lost from an oil either to the atmosphere or due to absorption by packaging materials (Fiselier et al., 2005). This chapter examines the differences between Australian and other oils, principally European, in terms of the volatile fraction (Kalua et al., 2007). This follows a brief discussion of the measurement of the volatile fraction. The latter is important as any differences in the volatile profile, quantitative or qualitative, may be artefactual and relate to methodological differences in the measurement step. Generalized comparisons between Australian olive oils and those from other regions are further complicated by such factors as the variety of cultivars, harvesting time, climatic and regional weather conditions, pre- and post-processing handling practices, and of course annual changes that take place in a number of these variables. At competitions such as the Annual Der Feinschmecker and Sol Doro which are held in Hamburg, Germany and Verona, Italy, respectively, there is great diversity in the submitted oils which number, in the case of ‘Der Feinschmecker’, 700 oils from most oilproducing countries of the world. Sadly, a high percentage of oils submitted for competition are faulted (Robert Harris, Organoleptic Assessor, personal communication). This, together with the diversity of cultivars, makes it very
201
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
202
SECTION | I
difficult to make any accurate assessment about the sensory comparison of oils from various countries.
22.2 MEASUREMENT OF OLIVE OIL VOLATILES 22.2.1 Sensory Although volatile and aroma compounds are not synonymous, there is a close relationship. The compounds that produce aroma must be sufficiently volatile to stimulate the human sensory receptors. The use of instrumental techniques for ‘sensory’ analysis has recently been reviewed (Escuderos et al., 2007). Instrumental methods generally measure the volatile profile, although techniques such as gas chromatography-olfactometry (GC-O) and the electronic nose are usually related to aroma via statistical analysis. Indeed, correlation of other instrumental data, such as gas chromatography-mass spectrometry and aroma, is widely utilized. For example, virgin olive oils extracted from Italian, Greek and Spanish cultivars were submitted to sensory evaluation and to determination of phenols and volatile compounds derived from lipoxygenase pathways (Angerosa et al., 2000). Correlations between the measured variables were assessed by univariate analysis and a linear regression analysis. Nevertheless, despite the significant advances in instrumentation, sensory evaluation remains the gold standard for assessing aroma. Olive oils are the only foods where sensory evaluation is a legal requirement (Procida et al., 2005). The IOC has developed guidelines for sensory analysis of olive oil and specifically the organoleptic assessment of virgin olive oil (International Olive Council, 2007). Assessment of olive oil is undertaken by 8–12 trained tasters, and includes organoleptic (smell) and buccal assessment (assessment of overall retronasal olfactory, gustatory and tactile sensations). Oils are scored and classified according to six negative attributes (fusty/muddy sediment; musty-humid-earthy; winey-vinegary-acid-sour; metallic; rancid; others) and three positive attributes (fruity; bitter; pungent). For example, classification of EVOO arises when the median of the defects is equal to 0 and the median of the fruity attribute is more than 0. Such classification protocols are adhered to in Australia in recognition of the IOC’s standing as the international authority on olive oil. Nevertheless, the sensory guidelines developed by the IOC are somewhat limiting since their main goal lies in the identification of oil defects. In order to monitor small but meaningful changes in oil sensory profiles, a greater number of sensory descriptors should be available (Angerosa, 2000). This has led to the development of a statistical sensory wheel for representing the global flavor matrix of EVOO which consists of 103 sensory attribute evaluations (Morales and Tsimidou, 2000). Sensory data
Lipids, Phenolics and Other Organics and Volatiles
are statistically treated and plotted as x,y coordinates to produce a wheel corresponding to the particular EVOO. Although this method of sensory analysis is more subjective and variable compared to that described by the IOC, significantly more fine detail is gathered which is fundamental for differentiation of olive oil samples. The question as to whether the sensory characteristics of Australian olive oils are unique in comparison to oils from around the world cannot be unequivocally answered as relevant data are not available. Whilst Australian olive oils are very highly regarded (Robert Harris, personal communication) definitive information with respect to sensory uniqueness does not exist. What is known is that concentrations of volatile compounds in olive oils may vary due to climate, cultivar and growing conditions (e.g. irrigation). Certainly Australia will present unique and varied climatic regions, and genetically different cultivars exist. It can therefore be expected that volatile profiles of Australian olive oils may differ to Mediterranean oils, for example. Whether such differences impact upon sensory properties is yet to be established, and will require fundamental, scientifically designed sensory studies to be undertaken, incorporating both IOC and statistical sensory wheel approaches. By necessity, these studies will need to adopt a global focus to compare the sensory characteristics of olive oils from various regions around the world, and also distinct regions within Australia. Ultimately, these studies should be conducted in parallel with chemical analysis of olive oil volatiles for a comprehensive understanding of olive oil aroma and sensory quality.
22.2.2 Instrumental The instrumentally derived volatile profile is obtained in a two-step process. First, the volatiles must be separated from the bulk oil and secondly, the profile is obtained on suitable instrumentation, which invariably utilizes GC with different detectors. As recognized by Vichi et al. (2007a) ‘the volatile profile of virgin olive oil closely depend(s) upon the method of extraction used’. The question ‘Are Australian oils different?’ must be examined in this light. Volatile ‘extracts’ can be obtained in a number of ways. The simplest is to sample the headspace above the oil and directly inject this onto the gas chromatograph (static headspace analysis (Ridolfi et al., 2002). However, most researchers employ some form of concentration step in order to assist in detecting volatiles present in low concentrations. Dynamic headspace extraction was utilized widely by Angerosa et al. (e.g. (Angerosa et al., 2001) for Italian oils and Morales et al. (e.g. (Luna et al., 2006) for Spanish oils). More recently, solid phase micro-extraction (SPME) has come into widespread use (e.g. (Contini and Esti, 2006) and references therein) due to it being a ‘solvent-free’ technique and its ease of use. We have used it extensively to
CHAPTER | 22 Volatile Compounds in Australian Olive Oils: How Different Are They From Other Oils
study Australian oils beginning with the work of Tura et al. (2004) (Figure 22.1). However, with SPME it is important that fiber performance is closely monitored and controlled (Kalua et al., 2006c). The separation of volatile compounds occurs via GC. Ideally the separation should be complete, i.e. one signal response for each compound. But because the headspace is a complex mixture, it is not always possible. Volatile compounds can be detected in numerous ways. If the aim of the work is to find the volatiles that are important to sensory perception, then an olfactometer (‘sniffing port’) may
203
be utilized where the ‘detector’ is a human nose (Reiners and Grosch, 1998). The most common form of instrumental detector is the flame ionization detector (FID). A mass-spectrometric detector offers the advantage of being able to identify the volatile compounds at the same time as separating them. Often this is done based on a match of the mass spectrum of the compound with one in a commercial library. It should be noted that this is a tentative identification and guidelines for full identification of volatile compounds have recently been published (Molyneux and Schieberle, 2007).
A E-hex-2-enal
50
E,Z-deca2,4-dienal E-4,5-epoxy-E-dec-2-enal Damascenone 4-ethyl guaiacol Unknown
40 mVolts
Heptanal 30 20
Heptan-1-ol Octanal
aldehyde I Unknown aldehyde II
Hexyl acetate E-oct-2-enal Nonanal
Hexanal
10
B 50
mVolts
40 30 20
Nona-2,4-dienal Hexan-1-ol
10
C 50
mVolts
40 30 20 10
5
10
15
20
25
Minutes FIGURE 22.1 The SPME-GC volatile profiles of three contrasting Australian olive oils: (A) from Corregiola; (B) from Manzanillo; (C) rancid oil (Tura et al., 2004). These chromatograms show that the volatile profile of oils from different cultivars are different to each other. Corregiola (A) is more like European oils with the dominance of E-2-hexenal, but extra volatiles are shown that are not typically reported in other work.
SECTION | I
Notwithstanding the issues raised above regarding methodology, many factors have been reported to influence the volatile composition of olive oil. As these factors may vary across small geographical distances, e.g. microclimate, soil, etc. and have been reported to be reflected in the volatile composition of oils, then it might be expected that Australian oils would indeed be different to their northern hemisphere counterparts. A short review of recent studies on the effect of these various parameters on volatile profiles is followed by a closer look at volatile profiles of Australian oils. Olive variety or cultivar has perhaps been most studied of all the factors that influence volatile composition. Recent work by Kalua et al. (2006d) on five cultivars grown in Australia showed that hexanol, hexanal, and 1-penten-3-ol were volatiles that discriminated cultivar, irrespective of the ripening stage of the fruit. Baccouri et al. (2007) were also able to find varietal differences in volatile composition and their analysis included terpenes, which historically have not been investigated for olive oil. Further, they found that in one variety the volatile profile was dominated by C6 alcohols (nearly 100%). Again, this is unusual in that most work reports E-2-hexenal as the volatile compound in highest concentration. Many of Australia’s large olive groves are under irrigation to cope with this country’s historically unreliable rainfall patterns. No studies have been performed on the volatile profiles of Australian oils with respect to irrigation, but reports have recently appeared from Spain (Gomez-Rico et al., 2006) and Italy (Servili et al., 2007a). In the Spanish study, the levels of major volatile components decreased in the course of ripening but were higher in irrigated olive oils: for example, the E-2-hexenal content ranged between 4.2 and 2.6 mg kg⫺1 (expressed as 4-methyl-2-pentanol) over fruit maturation under rain-fed conditions and between 8.0 and 3.5 mg kg⫺1 under Food and Agricultural Organization scheduling. The volatiles E-2-hexenal and Z-3-hexen-1-ol were the most useful variables for classification of the samples obtained under the different types of irrigation scheduling. Similar results were found in the Italian study, but more volatile compounds were reported to be significantly affected by irrigation, viz. hexanal, E-2-hexenal, E-2-hexen-1-ol, Z-2-hexen-1-ol, and 1-hexen-3-ol. Recently, much interest has focused on assessing the composition of oils for the purpose of establishing markers to define ‘Protected Designation of Origin’ oils. To this end Berlioz et al. (2006) examined 53 oils and by PCA were able to separate those originating from Nice, from other French or commercial oils. They found that E-2-hexenal and pentan-2-one had the highest discriminating power. Another approach has been to use pre-determined compounds as markers to separate oils from different growing regions as reported by Cavaliere et al. (2007). The five
Lipids, Phenolics and Other Organics and Volatiles
compounds were hexanal, (E)-2-hexenal, (E)-2-hexen-1-ol, 1-hexanol, and (Z)-3-hexen-1-yl acetate; and PCA discriminated 16 oils from four different regions in Calabria. Variation in the volatile compounds of olive oil can also be due to the processing of the fruit to obtain the oil. In Australia, most olive oil is processed using a two-phase system. In the northern hemisphere, processing conditions are more varied – two-phase, three-phase, hydraulic press, stone-crushing, etc. Even with a two-phase process, volatile compounds can vary depending on the time and temperature of processing. For example, malaxation time and temperature were discriminated by hexanal (amongst other variables) (Figure 22.2) whereas 1-penten-3-ol, E-2-hexenal, and octane (Figure 22.3) were volatiles that changed with temperature only, and Z-2-penten-1-ol changed with time only (Kalua et al., 2006b). A new development in olive processing is the use of modified atmosphere conditions – specifically the introduction of nitrogen to 30 27 ppm
22.3 VOLATILE PROFILES OF OLIVE OILS
24 21 18 60 50 40 Degrees 30 celsius
100 75 50 20
min
25 0
FIGURE 22.2 Variation in concentration of hexanal as malaxation time and temperature are changed in the processing of olive oil (Kalua et al., 2006b). The concentration of hexanal is dependent on both the time and temperature of malaxation during oil processing. Maximum concentrations are found at 60 minutes malaxation time, and at temperatures below 30°C and above 50°C.
1.2 ppm
204
0.9 0.5 0.3
50 40 Degrees 30 celsius
100 75 50
20
25
min
0
FIGURE 22.3 Variation in concentration of octane as malaxation time and temperature are changed in the processing of olive oil (Kalua et al., 2006b). Temperature has the most influence on the concentration of octane produced during processing.
CHAPTER | 22 Volatile Compounds in Australian Olive Oils: How Different Are They From Other Oils
limit spoilage reactions (Servili et al., 2003). The authors are not aware of this technology being adopted on a commercial scale in Australia. Unlike wine, olive oil is best consumed fresh and changes during storage, e.g. rancidity, only leads to a poorerquality product. Despite an enormous amount of literature on the chemistry of rancidity, there seems to be no universally recognized marker of this condition. Moreover, studies conducted on Australian oils have pointed to different markers of rancidity. In one study on Australian oils (Kalua et al., 2006a), the presence of oxygen and light led to hexanal as a marker (Figure 22.4), whereas during low-temperature storage acetic acid and pentanal were associated with this treatment. In the absence of oxygen, octane was the marker for storage in the light. On the other hand, for European oils suggested markers for storage are nonanal (Vichi et al., 2003a), the ratio of hexanal to nonanal (Morales et al., 1997), and 2-pentenal and 2-heptenal (Solinas et al., 1987).
22.4 COMPARISON WITH AUSTRALIAN OLIVE OILS Comparison of Australian olive oils with other oils is based on extra virgin oil as Australia predominantly produces 10
Function 2 (20% variance explained)
8 6 4 2 0 −2 −4 −6 10 −8 −20
−10 0 10 Function 1 (80% variance explained)
20
205
EVOO (Mailer, 2005). Anecdotal evidence on olive oils from different regions indicates that most EVOO have similar volatile compounds occurring at different concentrations. Flavor differences in olive oils can be attributed to either different concentrations of potent odor compounds or the presence/absence of compounds in olive oils under comparison (Reiners and Grosch, 1998). For Australian olive oils, (E)-2-hexenal has been found not always to be the dominant volatile compound as is usually the case with European oils (Tura et al., 2004) (Figure 22.1). The major and common volatile compounds detected in most virgin olive oils from Europe are (E)-2-hexenal, hexanal, hexan-1-ol and 3-methylbutan-1-ol (Aparicio et al., 1997). In contrast to European EVOO, damascenone and an unknown aldehyde were identified as major components in Australian EVOO where (E)-2-hexenal was not dominant (Tura et al., 2004). Additionally, some Australian EVOO have reported hexanal levels higher than (E)-2hexenal and characterized by longer-chain volatile compounds such as octanal and nona-2,4-dienal (Tura et al., 2004; Kalua et al., 2005). A study of Italian, Spanish and Moroccan EVOO (Reiners and Grosch, 1998) confirmed the abundance of C6 volatile compounds in Italian oils but showed they were low in fruity esters. The fruity esters, ethyl isobutyrate, ethyl butyrate, ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, and ethyl cyclohexylcarboxylate were abundant in Moroccan EVOOs (Reiners and Grosch, 1998). These fruity esters detected in Moroccan oils, are uncommon in Australian olive oils. Apart from hexyl acetate and ethyl isobutyrate, esters are rarely detected in most of the Australian olive oils (Tura et al., 2004; Kalua et al., 2005). The occurrence of minor volatile components detected in Australian EVOO and rarely reported in European EVOO can be a point of differentiation between these regions. Volatile compounds detected in Australian EVOO and rarely reported in European oils are (E)-2-octenal, nonanal, (E,Z)-2,4-nonadienal, 4-ethyl guaiacol, (E,Z)-2,4decadienal, (E)-2,5-epoxy-(E)-2-decenal and damascenone (Tura et al., 2004); octane, octanal, octanol, 6-methyl-5hepten-2-one, 5-methyl-5-hepten-2-one, and 2-pentylfuran (Kalua et al., 2005). In European oils, longer-chain volatile compounds are rarely reported; in cases where they are detected, their presence does not allow differentiation of samples and the volatile compounds are generally at low concentrations (Vichi et al., 2003b).
Storage condition Cold-with air
Dark-with air
Light-with air
FIGURE 22.4 Discriminant plot showing that different storage conditions lead to oils that are measurably different (Kalua et al., 2006a). Volatile compounds are markers for these conditions, e.g. hexanal for storage in the light. Storage of oil induces measurable changes in the oil. These changes are closely correlated with the storage conditions regardless of the time of storage and are readily detected by instrumental and statistical analyses.
22.4.1 Sources of Differences between Australian Olive Oils and other Oils The differences in volatile compounds for Australian olive oils and other oils can be attributed to the influence of geographic region. This has also been associated with differences in olive oils within Europe (Vichi et al., 2003b).
206
SECTION | I
The dominance of (E)-2-hexenal in volatile profiles of European EVOOs (Vichi et al., 2003b) is not always the case for all European regions. For instance, a study on two different geographic regions of Italy demonstrated that olive oils from one region showed a higher content of volatile compounds derived from linolenic acid than the other region (Vichi et al., 2003b). The differences between Australian EVOO and other olive oils based on major C6 volatile compounds and longer chain compounds are further complicated with the basis of comparison. Comparison of oils at the same time and conditions is logistically challenging for Australia, being farremoved from other regions. Monitoring volatile compounds in olive oils at point of production and point of consumption produces different results (Kalua et al., 2006a). In addition to the point of reference for comparison and the geographic influence on the production and occurrence of volatile compounds in Australian EVOO, other factors can be crucial in any comparison of Australian olive oils with other oils.
22.4.2 Methodologies Used in the Determination of Volatile Compounds Most work on European olive oils has focused on C5/C6 compounds from the lipoxygenase pathway and is rarely extended to less-volatile compounds with retention indices exceeding 1400 on an SE-54 column (Tura et al., 2004). It is not apparent whether the dominance of C5/C6 compounds is representative of most of the European EVOO or culminates from the subjective selection of compounds associated with the lipoxygenase pathway based on prior knowledge of researchers. Additionally, the difference between Australian EVOO and oils from other regions could reflect the instrumental methods used to determine the volatile profiles (see above). In general, the C8–C10 volatile compounds have been overlooked in most of the European fresh virgin olive oils, regardless of their frequently high and detectable concentrations (Reiners and Grosch, 1998). These C8–C10 volatile compounds are usually linked to rancid off-flavors with minimal association with fresh olive oils (Reiners and Grosch, 1998). However, the presence of C8–C10 volatile compounds in Australian EVOO did not introduce sensory defects (Tura et al., 2004). The reported C8–C10 compounds in Australian EVOO could be due to the nondiscriminatory reporting of volatile compounds in current studies (Tura et al., 2004; Kalua et al., 2005).
22.4.3 Effect of Phenolic Compounds and other Minor Constituents The way olive oils withstand different oxidative conditions during transportation and packaging can vary. Most of the
Lipids, Phenolics and Other Organics and Volatiles
Australian EVOOs have adequate antioxidant activity due to phenolic compounds to preserve the freshness and hinder potential development of off-flavors during transportation for at least two months under different storage conditions (Kalua et al., 2006a). The phenolic profile of Australian olives can be distinctly different to those from other regions (Ryan et al., 2002). This can potentially result in different antioxidant capacities and mechanisms generating different flavor compounds for the oils that are not monitored at the point of production. Additionally, the differences in phenolic compounds and minor components have a direct impact on the sensory quality of olive oils by binding the volatile compounds (Jung et al., 2000). These differences in composition imply that different volatile compounds are released and perceived by the consumer adding to the complexity of differences between Australian olive oils and other oils.
22.4.4 Fatty Acid Composition Differences The fatty acid variability of Australian oils has been reported and, in particular, linolenic acid in some oils was higher than the recommended maximum level of 1.0% (Mailer, 2005). This has implications of generating volatile compounds linked more to the linolenic than linoleic acid in the lipoxygenase pathway (Gardner, 1995). In Australian olive oils, this has produced oils with considerable amounts of (E)-2-hexenal, though not always dominant (Tura et al., 2004; Kalua et al., 2005). However, the variable environment and cultivars in the Australian olive industry (Mailer, 2005) has resulted in some Australian olive oils having higher levels of hexanal than (E)-2-hexenal (Tura et al., 2004; Kalua et al., 2005) – an indication of higher linoleic acid than linolenic acid (Aparicio et al., 1997). This is different from most of the European EVOO where (E)-2hexenal is usually the major volatile compound and more abundant than hexanal (Aparicio et al., 1997).
22.4.5 Enzyme Activity Differences In addition to the effect of different proportions of volatile compounds from linoleic and linolenic acid branches of the lipoxygenase pathway, the volatile profile can also be affected by the proportion of active enzymes. For instance, most of the Australian oils are low in esters (Tura et al., 2004; Kalua et al., 2005) compared to some Tunisian and Moroccan olive oils (Reiners and Grosch, 1998; Dhifi et al., 2005). The apparent difference between most of the Tunisian/Moroccan and Australian ester levels can be attributed to the different expressions of AAT activities in cultivars common in these two countries (Angerosa et al., 1999) bearing in mind that AAT activity is less dependent on climatic conditions and geographic regions (Vichi et al., 2003b). In general, aldehydes are abundant in most of the Australian EVOO (Tura et al., 2004; Kalua et al., 2005) indicating a higher hydroperoxide lyase (HPL) than alcohol
207
CHAPTER | 22 Volatile Compounds in Australian Olive Oils: How Different Are They From Other Oils
dehydrogenase (ADH) activity. The higher aldehyde levels in Australian EVOO is consistent with most EVOO from other regions (Angerosa et al., 1999; Vichi et al., 2003b) apart from cases that have reported higher concentrations of alcohols with respect to aldehydes attributed to overexpression or over-activity of ADH (Cavaliere et al., 2007).
22.5 HEALTH ASPECTS Table 22.1 compares weekly per capita consumption rates of olive oil for Greece, Italy and Australia (Anon, 2004). Such data demonstrate that the Mediterranean countries maintain their traditional role as the highest consumers of olive oil, and per capita consume more than ten times the amount of olive oil compared to that of Australia. Australia represents a new market with respect to both consumption and production of olive oil. Based on the per capita consumption data, weekly consumption of specific virgin olive oil volatiles can be calculated (Table 22.2). Potential toxicity risks associated with consumption of olive oil volatiles can then be evaluated using toxicity data derived from MSDS (material safety data sheets; Table 22.2). Using (E)-2-hexenal as an example, and its LD50 data, a toxic dose for an average human weighing 70 kg would be 54 600 mg, or 54.6 g. Consumption of this quantity of (E)-2-hexenal from a single dose of virgin olive oil would require ingestion of approximately 8000 kg of oil. Ingestion of such a significant quantity of oil is not feasible either as a single dose or even as an accumulated consumption over many days. In fact, the calculated dose of 8000 kg is 340 times the annual
per capita consumption rate in Greece! Such data demonstrate that the toxicity of olive oil volatiles is negligible based on current consumption trends, known volatiles and known toxicity data. Even excessive consumption of 1 kg week⫺1 (i.e. approximately twice the per capita consumption rate in Greece) would not pose any toxicity risks with respect to consumption of olive oil volatiles.
22.6 CONCLUSION As far as the authors are aware, there has not been a controlled study on the question of the uniqueness or not of Australian oils. Such a study would in fact be logistically challenging for reasons elaborated above, not to mention the fact that the Australian production season is temporally separated by 6 months from the northern hemisphere. This TABLE 22.1 Per capita consumption of olive oil for selected countries. Country
Per capita consumption of olive oil (kg week⫺1)
Greece
0.46
Italy
0.24
Australia
0.023
This table presents weekly per capita consumption of olive oil for selected countries representing traditional markets (Italy and Greece) and emerging markets (Australia).
TABLE 22.2 Toxicity and consumption data for olive oil volatiles. Selected compounds
Toxicity (LD50 rat; mg kg⫺1)
Concentration in virgin olive oil (μg kg⫺1)*
Weekly per capita volatile consumption in μg Greece
Hexanal
Italy
Australia
4890
1770
806.7
420.3
40.8
(E)-2-Hexenal
780
6770
3085.6
1607.8
156.2
(Z)-3-Hexenol
4700
684
311.7
162.5
15.8
Octanal
5630
382
174.1
90.7
8.8
661
587
267.5
139.4
13.5
Propanal
1410
409
186.4
97.1
9.4
2-Phenylethanol
1790
843
384.2
200.2
19.5
Acetaldehyde
*
Based on an Italian olive oil (Reiners and Grosch, 1998). LD50; Median lethal dose. Lethal dose required to kill half the members of a tested population. This table summarizes toxicity data for selected volatile compounds and corresponding concentrations in an Italian virgin olive oil. The latter together with the oil consumption data in Table 22.1 are used to calculate consumption data for the selected volatile compounds.
208
SECTION | I
fact alone means that oils would need to be stored prior to sensory analysis, and such storage would be unlikely to be beneficial for the oil. In the absence of such a study, anecdotal evidence suggests that Australian oils can be regarded very highly, but as yet, the jury is out on whether they have unique (or ‘regional’) characteristics. The thrust of our chapter is that there are many factors that affect the volatile profile of an oil – climate, cultivar, irrigation, etc. – and that because Australia has differences in some of these, it would be expected that the volatile profile would be different. Whether or not that translates to a difference in sensory profile is yet to be established.
SUMMARY POINTS ●
●
●
●
●
●
There is no controlled study on the question of the uniqueness or not of Australian oils. Anecdotal evidence suggests that Australian oils compare favorably to those from elsewhere. The unique or regional characteristics of Australian oils have yet to be shown. Studies from different countries have shown that climate, cultivar, horticultural practices and processing inter alia, contribute to differences in the volatile profile of an oil. Data from studies in Australia show that for many of these, differences are observed. Further research is needed to establish if these differences translate to a difference in sensory profile.
REFERENCES Angerosa, F., Basti, C., Vito, R., 1999. Virgin olive oil volatile compounds from lipoxygenase pathway and characterization of some Italian cultivars. J. Agric. Food Chem. 47, 836–839. Angerosa, F., 2000. Sensory quality of olive oils. In: Handbook of Olive Oil: Analysis and Properties. Aspen Publishers, Inc., Maryland, USA. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2000. Virgin olive oil odour notes: their relationships with volatile compounds from the lipoxygenase pathway and secoiridoid compounds. Food Chem. 68, 283–287. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2001. Influence of malaxation temperature and time on the quality of virgin olive oils. Food Chem. 72, 19–28. Anon, California and World Olive Oil Statistics: Production Short Course. 2004. Accessed on April 20 2008, http://ucce.ucdavis.edu/files/ filelibrary/2161/17343.pdf. Aparicio, R., Morales, M.T., Alonso, V., 1997. Authentication of European virgin olive oils by their chemical compounds, sensory attributes, and consumers’ attitudes. J. Agric. Food Chem. 45, 1076–1083. Arrebola, F.J., Gonzalez-Rodriguez, M.J., Frenich, A.G., Marin-Juan, A., Vidal, J.L.M., 2005. Determination of halogenated solvents content in olive oil by two completely automated headspace techniques coupled to gas chromatography-mass spectrometry. Anal. Chim. Acta 552, 60–66.
Lipids, Phenolics and Other Organics and Volatiles
Baccouri, B., Ben Temime, S., Campeol, E., Cioni, P.L., Daoud, D., Zarrouk, M., 2007. Application of solid-phase microextraction to the analysis of volatile compounds in virgin olive oils from five new cultivars. Food Chem. 102, 850–856. Berlioz, B., Cordella, C., Cavalli, J.F., Lizzani-Cuvelier, L., Loiseau, A.M., Fernandez, X., 2006. Comparison of the amounts of volatile compounds in French protected designation of origin virgin olive oils. J. Agric. Food Chem. 54, 10092–10101. Bianco, A., Dezzi, S., Bonadies, F., Romeo, G., Scarpati, M.L., Uccella, N., 2006. The variability of composition of the volatile fraction of olive oil. Nat. Prod. Res. 20, 475–478. Carrillo-Carrion, C., Lucena, R., Cardenas, S., Valcarcel, M., 2007. Liquid-liquid extraction/headspace/gas chromatographic/mass spectrometric determination of benzene, toluene, ethylbenzene, (o-, mand p-)xylene and styrene in olive oil using surfactant-coated carbon nanotubes as extractant. J. Chromatogr. A 1171, 1–7. Cavaliere, B., De Nino, A., Hayet, F., Lazez, A., Macchione, B., Moncef, C., Perri, E., Sindona, G., Tagarelli, A., 2007. A metabolomic approach to the evaluation of the origin of extra virgin olive oil: a convenient statistical treatment of mass spectrometric analytical data. J. Agric. Food Chem. 55, 1454–1462. Contini, M., Esti, M., 2006. Effect of the matrix volatile composition in the headspace solid-phase microextraction analysis of extra virgin olive oil. Food Chem. 94, 143–150. Dhifi, W., Angerosa, F., Serraiocco, A., Oumar, I., Hamrouni, I., Marzouk, B., 2005. Virgin olive oil aroma: characterization of some Tunisian cultivars. Food Chem. 93, 697–701. Escuderos, M.E., Uceda, M., Sanchez, S., Jimenez, A., 2007. Instrumental technique evolution for olive oil sensory analysis. Eur. J. Lipid Sci. Technol. 109, 536–546. Fiselier, K., Biedermann, M., Grob, K., 2005. Injector-internal thermal desorption from edible oils. Part 2: chromatographic optimization for the analysis of migrants from food packaging material. J. Sep. Sci. 28, 2144–2152. Garcia-Gonzalez, D.L., Tena, N., Aparicio, R., 2007. Characterization of olive paste volatiles to predict the sensory quality of virgin olive oil. Eur. J. Lipid Sci. Technol. 109, 663–672. Gardner, H.W., 1995. Biological roles and biochemistry of the lipoxygenase pathway. HortScience 30, 197–205. Gomez-Rico, A., Salvador, M.D., La Greca, M., Fregapane, G., 2006. Phenolic and volatile compounds of extra virgin olive oil (Olea europaea L. cv. Cornicabra) with regard to fruit ripening and irrigation management. J. Agric. Food Chem. 54, 7130–7136. International Olive Council, Sensory Analysis of Olive Oil: Method for the organoleptic assessment of virgin olive oil. 2007. Accessed on April 20 2008, http://www.internationaloliveoil.org/downloads/orga6.pdf. Jimenez, A., Aguilera, M.P., Beltran, G., Uceda, M., 2006. Application of solid-phase microextraction to virgin olive oil quality control. J. Chromatogr. A 1121, 140–144. Jung, D.M., De Ropp, J.S., Ebeler, S.E., 2000. Study of interactions between food phenolics and aromatic flavors using one- and twodimensional H-1 NMR spectroscopy. J. Agric. Food Chem. 48, 407–412. Kalua, C.M., Allen, M.S., Bedgood, D.R., Bishop, A.G., Prenzler, P.D., 2005. Discrimination of olive oils and fruits into cultivars and maturity stages based on phenolic and volatile compounds. J. Agric. Food Chem. 53, 8054–8062. Kalua, C.M., Allen, M.S., Bedgood, D.R., Bishop, A.G., Prenzler, P.D., Robards, K., 2007. Olive oil volatile compounds, flavour development and quality: a critical review. Food Chem. 100, 273–286.
CHAPTER | 22 Volatile Compounds in Australian Olive Oils: How Different Are They From Other Oils
Kalua, C.M., Bedgood, D.R., Bishop, A.G., Prenzler, P.D., 2006a. Discrimination of storage conditions and freshness in virgin olive oil. J. Agric. Food Chem. 54, 7144–7151. Kalua, C.M., Bedgood, D.R., Bishop, A.G., Prenzler, P.D., 2006b. Changes in volatile and phenolic compounds with malaxation time and temperature during virgin olive oil production. J. Agric. Food Chem. 54, 7641–7651. Kalua, C.M., Bedgood, D.R., Prenzler, P.D., 2006c. Development of a headspace solid phase microextraction-gas chromatography method for monitoring volatile compounds in extended time-course experiments of olive oil. Anal. Chim. Acta. 556, 407–414. Kalua, C.M., Mailer, R.J., Ayton, J., Allen, M.S., Bedgood, D.R., Bishop, A.G., Prenzler, P.D., 2006d. Discrimination of olive oils and fruits into cultivars and maturity stages based on phenolic and volatile compounds (Vol 53, p. 8054, 2005). J. Agric. Food Chem. 54, 8390. Luna, G., Morales, M.T., Aparicio, R., 2006. Characterisation of 39 varietal virgin olive oils by their volatile compositions. Food Chem. 98, 243–252. Mailer, R.J., 2005. Variation in oil quality and fatty acid composition in Australian olive oil. Aust. J. Exp. Agric. 45, 115–119. Molyneux, R.J., Schieberle, P., 2007. Compound identification: a Journal of Agricultural and Food Chemistry Perspective. J. Agric. Food Chem. 55, 4625–4629. Morales, M.T., Rios, J.J., Aparicio, R., 1997. Changes in the volatile composition of virgin olive oil during oxidation: flavors and off-flavors. J. Agric. Food Chem. 45, 2666–2673. Morales, M.T., Tsimidou, M., 2000. The role of volatile compounds and polyphenols. In: Handbook of Olive Oil: Analysis and Properties. Aspen Publishers, Inc., Maryland, USA. Pena, F., Cardenas, S., Gallego, M., Valcarcel, M., 2004. Combining headspace gas chromatography with mass spectrometry detection for confirmation of hydrocarbon residues in virgin olive oil following automatic screening. J. Chromatogr. A 1052, 137–143. Procida, G., Giomo, A., Cichelli, A., Conte, L.S., 2005. Study of volatile compounds of defective virgin olive oils and sensory evaluation: a chemometric approach. J. Sci. Food. Agric. 85, 2175–2183. Reiners, J., Grosch, W., 1998. Odorants of virgin olive oils with different flavor profiles. J. Agric. Food Chem. 46, 2754–2763. Ridolfi, M., Terenziani, S., Patumi, M., Fontanazza, G., 2002. Characterization of the lipoxygenases in some olive cultivars and determination of their role in volatile compounds formation. J. Agric. Food Chem. 50, 835–839.
209
Ryan, D., Antolovich, M., Herlt, T., Prenzler, P.D., Lavee, S., Robards, K., 2002. Identification of phenolic compounds in tissues of the novel olive cultivar hardy’s mammoth. J. Agric. Food Chem. 50, 6716–6724. Servili, M., Esposto, S., Lodolini, E., Selvaggini, R., Taticchi, A., Urbani, S., Montedoro, G., Serravalle, M., Gucci, R., 2007a. Irrigation effects on quality, phenolic composition, and selected volatiles of virgin olive oils cv. Leccino. J. Agric. Food Chem. 55, 6609–6618. Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G.F., 2003. Volatile compounds and phenolic composition of virgin olive oil: optimization of temperature and time of exposure of olive pastes to air contact during the mechanical extraction process. J. Agric. Food Chem. 51, 7980–7988. Servili, M., Taticchi, A., Esposto, S., Urbani, S., Selvaggini, R., Montedoro, G.F., 2007b. Effect of olive stoning on the volatile and phenolic composition of virgin olive oil. J. Agric. Food Chem. 55, 7028–7035. Solinas, M., Marsilio, V., Angerosa, F., 1987. Behaviour of some components of virgin olive oil flavour in relation to maturity. Riv. It. Sost. Grasse 64, 475. Tura, D., Prenzler, P.D., Bedgood, D.R., Antolovich, M., Robards, K., 2004. Varietal and processing effects on the volatile profile of Australian olive oils. Food Chem. 84, 341–349. Vichi, S., Guadayol, J.M., Caixach, J., Lopez-Tamames, E., Buxaderas, S., 2007b. Comparative study of different extraction techniques for the analysis of virgin olive oil aroma. Food Chem. 105, 1171–1178. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., Lopez-Tamames, E., 2003a. Solid-phase microextraction in the analysis of virgin olive oil volatile fraction: modifications induced by oxidation and suitable markers of oxidative status. J. Agric. Food Chem. 51, 6564–6571. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., Lopez-Tamames, E., 2003b. Solid-phase microextraction in the analysis of virgin olive oil volatile fraction: characterization of virgin olive oils from two distinct geographical areas of northern Italy. J. Agric. Food Chem. 51, 6572–6577. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., Lopez-Tamames, E., 2007a. The occurrence of volatile and semi-volatile aromatic hydrocarbons in virgin olive oils from north-eastern Italy. Food Control 18, 1204–1210.
This page intentionally left blank
Chapter 23
Sterols and Non-steroidal Triterpenoids of the Developing Olive Fruit Naïm Stiti1,2, Saïda Triki2 and Marie-Andrée Hartmann1 1 2
lnstitut de Biologie Moleculaire des Plantes (CNRS UPR 2357), Université de Strasbourg, 28 rue Goethe, 67083 Strasbourg, France Faculté des Sciences de Tunis, Département des Sciences Biologiques, Campus Universitaire, 2092 Tunis, Tunisia
23.1 INTRODUCTION Olive oil as well as olive leaves have been known for a long time to contain a wide range of sterols and non-steroidal triterpenoids, including erythrodiol, oleanolic acid and maslinic acid (Power and Tutin 1908; Caputo et al., 1974; Itoh et al., 1981), which are oxygenated derivatives of β-amyrin (olean-12-en-3β-ol), one of the most commonly occurring triterpenes. Sterols and non-steroidal triterpenoids, which belong to the group of terpenoids or isoprenoids, the largest family of natural products, are synthesized via the cytoplasmic acetate/mevalonate pathway and share common precursors up to (3S)-2,3-oxidosqualene (OS) (Seo et al., 1988; Benveniste, 2002). Then, OS serves as a substrate for various OS cyclases, also called triterpene synthases, to form C30 compounds (i.e., comprising six C5-isoprene units). Cycloartenol synthase catalyzes the cyclization of OS folded in the pre-chair-boat-chair conformation, via the protosteryl cation, into cycloartenol, the first cyclic precursor of the sterol pathway (Figure 23.1). About 20 steps are needed to convert cycloartenol in end pathway sterols (Benveniste, 2002). Non-steroidal triterpenoids are assumed to be formed from OS folded in the all-pre-chair conformation, through a series of carbocationic intermediates (Abe et al., 1993) (Figure 23.1). They are then often metabolized into more oxygenated compounds, which serve as precursors for the synthesis of triterpenic saponins (Mahato et al., 1988). As the cyclization of OS into sterols and non-steroidal triterpenoids represents a branch point between primary and secondary metabolisms, OS cyclases are attractive tools for investigating the physiological roles of non-steroidal triterpenoids. The present study sheds more light on biosynthetic relationships occurring between the sterol and non-steroidal triterpenoid pathways in Olea europaea L. throughout olive fruit ontogeny. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
23.2 THE OLIVE FRUIT CONTAINS A VAST ARRAY OF STEROLS AND NON-STEROIDAL TRITERPENOIDS Sterols and non-steroidal triterpenoids were isolated from total lipid extracts of olive drupes (i.e., the whole fruit comprising the epicarp, the mesocarp and the endocarp or pit with the seed) as previously described (Hartmann and Benveniste 1987; Stiti et al., 2007). Free sterols, tetracyclic and pentacyclic triterpenes, triterpenic diols as well as the compounds released after hydrolysis of ester conjugates were identified as acetate derivatives by their relative retention time in gas chromatography and their mass spectrometry fragmentation pattern in gas chromatography coupled to mass spectrometry (Rahier et al., 1989; Stiti et al., 2007 and references herein). Mono- and dihydroxy pentacyclic triterpenic acids (HPTAs) were isolated according to Pérez-Camino and Cert (1999) and Stiti et al. (2007) and identified as acetate derivatives of the corresponding methylesters (Stiti et al., 2007).
23.2.1 Sterols Olive drupes were shown to contain a mixture of sterols, with sitosterol as the largely predominant compound (between 70 and 95% of total sterols), and 24-methylcholesterol, stigmasterol and isofucosterol (Δ5-avenasterol) (1–3%). We also identified brassicasterol, 24-methylenecholesterol, Δ5,24-stigmastadienol, Δ7-avenasterol and cholesterol. All the usual intermediates of the sterol pathway: squalene, 4α-dimethylsterols (cycloartenol and 24-methylenecycloartanol) and 4a-methylsterols (obtusifoliol, cycloeucalenol, 24-methylene and 24-ethylidenelophenol) were found. The occurrence of some less usual sterols such as 24-methyl- and 24-ethyl-lophenol, (24S)24-ethylcholesta-5,25-dien-3β-ol (clerosterol) and sterols
211
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
212
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
H
+
+ H
20
H H
+
17
2,3-oxidosqualene
CCC
dammarenyl cation
C B C
protosteryl cation
H 19
HO
HO
H
+
H H
H H
H 3 2 O
+
HO
HO
lupenyl cation
baccharenyl cation
20 + H
HO
oleanyl cation
-12αH
ursanyl cation
-12βH
H HO
HO
9
HO
1
12
7
HO
-9βH
-19H
CH2OH
HO
HO
HO
parkeol
HO
9
3
14 HO
CH2OH
CH2OH
cycloartenol COOH
HO
HO
15
COOH
COOH HO
10
4
Sterols COOH HO HO
5
Non-steroidal triterpenoids FIGURE 23.1 Postulated biosynthetic pathway of non-steroidal triterpenoids from 2,3-oxidosqualene cycIization in the Olea europaea fruit. This figure is adapted from the previously published Scheme 1 (Stiti et al., 2007). CBC and CCC refer respectively to the pre-chair-boat-chair and allpre-chair conformations of oxidosqualene (OS). OS serves as a substrate for the synthesis of either sterols or non-steroidal triterpenes. Cycloartenol synthase catalyzes the formation of cycloartenol, the first cyclic precursor of sterols, via the protosteryl cation. The OS cyclization reaction by non-steroidal triterpene synthases is thought to proceed through generation of several four or five ring-carbocationic intermediates. These carbocationic intermediates are represented in brackets. Oleanane-type triterpenoids, which arise from the oleanyl cation, are by far the predominant compounds, as shown by the widest arrows. The names of the different compounds, which are designated by a number, are given in the legend of Figure 23.2.
with a double bond at C-23 (Δ5,23-stigmastadienol and 24ethyl E-23-dehydrolophenol); at C-11 (5α-lanosta-9(11), 24-dien-3β-ol) or parkeol and 24-methylene-lanost-9(11)en-3β-ol has to be mentioned. All these sterols were also present as esters. However, it is interesting to note that no free parkeol could be detected.
23.2.2 Non-steroidal Triterpenoids Besides sterols, the olive fruit contains a great diversity of triterpenoids. Their structures are shown in Figure 23.2. They include 19 pentacyclic triterpenoids arising from four different carbon skeletons: oleanane-type (1–7) (β-amyrin 1, 28-nor-β-amyrin 2, erythrodiol 3, oleanolic acid 4, maslinic acid 5, β-amyrone 6 and β-amyrin 7), ursane-type (8–12) (α-amyrin 8, 28-nor-α amyrin 9, uvaol 10, ursolic acid 11 and α-amyrone 12), lupane-type (13–17) (lupeol 13, 3epi-lupeol 14, 3-epi-betulin 15 and 3-epi-betulinic acid 16
and lupenone 17) and taraxane-type (18–19) (taraxerol 18, taraxer-14-ene-3β,28-diol 19), as well as two tetracyclic triterpenes with euphane-type (butyrospermol 20) and baccharane-type (bacchar-12,21-dien-3β-ol 21) carbon skeletons. Oleanane triterpenoids were largely predominant, with oleanolic and maslinic acids representing by far the major compounds. Pentacyclic triterpenes also occurred as esters, but no acylated triterpenic diols have been found. Thus, more than 40 sterols and non-steroidal triterpenoids have been identified in the olive fruit. Our results are consistent with previous reports about the sterol and triterpenoid composition of olive oil or fruit (Itoh et al., 1981; Chryssafidis et al., 1992; Bianchi et al., 1994; Reina et al., 1997; Ranalli et al., 2002; Stiti et al., 2002; Azadmard-Damirchi et al., 2005; see Chapter 27). However, the occurrence in the olive fruit of taraxer-14-ene-3β,28-diol, 3-epi-lupeol and its metabolites, 3-epi-betulin and 3-epi-betulinic acid, had not been reported before.
213
CHAPTER | 23 Sterols and Non-steroidal Triterpenoids of the Developing Olive Fruit
R1
R1 R2
(1) (2) (3) (4) (5) (6)
(13) (14) (15) (16) (17)
R
HO
R
(7)
R = α-H, β-OH, R1 = CH3, R2 = H R = α-H, β-OH, R1 = H, R2 = H R = α-H, β-OH, R1 = CH2OH, R2 = H R = α-H, β-OH, R1 = COOH, R2 = H R = α-H, β-OH, R1 = COOH, R2 = OH R = O, R1 = CH3, R2 = H R = α-H, β-OH, R1 = CH3 R = α-OH, β-H, R1 = CH3 R = α-OH, β-H, R1 = CH2OH R = α-OH, β-H, R1 = COOH R = O, R1 = CH3
(8) (9) (10) (11) (12)
R = α-H, β-OH, R1 = CH3 R = α-H, β-OH, R1 = H R = α-H, β-OH, R1 = CH2OH R = α-H, β-OH, R1 = COOH R = O, R1 = CH3
(18) R1 = CH3 (19) R1 = CH2OH R1
R1 H HO
R
H
H HO
HO
(20)
(21)
FIGURE 23.2 Structures of the non-steroidal triterpenoids identified in the olive fruit. Oleanane-type: (1), β-amyrin; (2), 28-nor-β-amyrin; (3), erythrodiol; (4), oleanolic acid; (5), maslinic acid; (6), β-amyrone; (7), δ-amyrin; Ursane-type: (8), α-amyrin; (9), 28-nor-α-amyrin; (10), uvaol; (11), ursolic acid; (12), α-amyrone; Lupane-type: (13), lupeol; (14), 3-epi-lupeol; (15), 3-epi-betulin; (16), 3-epi-betulinic acid; (17), lupenone; Taraxane-type: (18), taraxerol; (19), taraxer-14-ene-3β,28-diol; Euphol-type: (20), butyrospermol; Baccharanetype: (21), bacchar-12,21-dien-3β-ol.
23.3 CHANGES IN THE CONTENT OF FREE AND ESTERIFIED STEROLS AND NON-STEROIDAL TRITERPENOIDS THROUGHOUT FRUIT DEVELOPMENT
Drupes from the different batches were analyzed for their content in free and esterified sterols and non-steroidal triterpenoids, but only data corresponding to olives harvested at the stages 12, 18, 21 and 30 WAF are presented here.
Olive fruit were handpicked from all the sides of one olive tree, Olea europaea L. cv Chemlali, at 13 distinct stages of fruit growth and ripening corresponding to 12, 13, 15, 16, 18, 21 23, 25, 27, 29, 30, 32 and 33 weeks after development (WAF). At the time of the first harvest, the lignification of the olive endocarp had ended. Between 12 and 18 WAF, olives were green and progressively increased in size and fresh weight, but in the case of the Chemlali cultivar, these changes were of limited amplitude compared to other olive varieties. At the end of this period, the final fruit size was almost fixed and from the 21st WAF, epidermal color gradually turned from green to purple. Complete maturity was observed after 29 WAF. The 33 WAF stage corresponded to an ‘over maturation’ stage.
23.3.1 Sterols from the 12th to the 18th WAF In the young olive fruit (i.e. between 12 and 18 WAF), free sterols were present as a mixture in which sitosterol was largely predominant (95%), but the usual sterol intermediates, 4,4-dimethyl- and 4α-methylsterols, were barely detectable (Table 23.1). A relatively high amount of squalene (700 μg g⫺1 dry wt) was detected at 13 WAF, but then rapidly decreased to 40 μg g⫺1 at 18 WAF (data not shown). During this period of time, a slight decrease in the total free sterol content of the olive fruit was observed (Table 23.1). The young olive drupes were also found to contain sterols as ester conjugates, with a sterol profile slightly
214
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 23.1 Changes in free sterols during olive fruit development.
TABLE 23.2 Changes in esterified sterols during olive fruit development.
Developmental stage Sterol classes
12 WAF
18 WAF
Developmental stage Sterol classes
12 WAF
18 WAF
21 WAF
30 WAF
4,4-dimethylsterols cycloartenol
nd
4,4-dimethylsterols cycloartenol
nd
nd
2.7
1.4
24-methylenecycloartanol nd
parkeol
nd
nd
16.4
13.4
24-methylenecycloartanol
4.7a
6.6
7.7
10.4
21 WAF
30 WAF
nd
0.4
3.2
nd
3.1
26.6
4α-methylsterols obtusifoliol
nd
nd
0.6
0.2
24-methylenelophenol
nd
nd
1.0
0.7
24-methyl-lophenol
nd
nd
0.2
0.1
4α-methylsterols obtusifoliol
0.5
0.7
1.4
0.7
0.4
24-methylenelophenol
2.7
2.0
0.8
0.4
1.6
0.9
0.6
0.6
cycloeucalenol
nd
nd
0.7
24-ethyl-lophenol
nd
nd
0.2
0.1
cycloeucalenol
24-ethylidenelophenol
nd
nd
0.9
1.1
24-ethylidenelophenol
1.1
1.3
2.7
3.3
0
0
4-demethylsterols 24-methylcholesterol
2.1
2.1
2.7
2.3
7.2
7.5
2.4
0.7
4-demethylsterols brassicasterol
0.4a
0.4
24-methylcholesterol
1.5
1.4
2.9
1.9
stigmasterol
stigmasterol
1.4
2.5
1.3
0.8
clerosterol
0.8
0.7
0.8
0.7
0.6
sitosterol
62.5
62.5
59.2
62.0
isofucosterol
16.2
15
2.6
4.1
Δ5,24-stigmastadienol
0.5
0.6
nd
nd
Total amount (μg/g dry wt)
44
100
140
305
clerosterol sitosterol isofucosterol Total amount (μg/g dry wt)
0.7 95.9
1.2 94.4
0
0
250
230
0.8 87.5
61.3
0.2
2.7
500
960
The standard deviation for quantitative determinations was ⫾ 5%. a
% of total free sterols; nd: not detectable.
different from that of free forms. In particular, sitosterol remained the major compound (63%), but significantly higher relative proportions of 24-methylcholesterol, stigmasterol and isofucosterol were found (Table 23.2). Low amounts of acylated sterol intermediates, especially 24methylenecycloartanol, were present (Table 23.2). Between 12 and 18 WAF, a 2.3-fold increase in the total amount of sterol esters was observed, an increase that equally affected sterol intermediates and end products, indicating that some sterol biosynthesis took place in the very young fruit. However, these newly synthesized sterols were immediately conjugated to a fatty acid and thus removed from the free sterol pathway.
23.3.2 Sterols from the 21st to the 30th WAF From the 21st WAF, dramatic changes were observed in the free sterol pathway. Early biosynthetic intermediates, i.e., squalene, 4,4-dimethyl- (cycloartenol and 24-methylenecycloartanol) and 4α-methylsterols (cycloeucalenol, obtusifoliol, 24-methylene- and 24-ethylidenelophenol) as well as late precursors (isofucosterol) began to be detectable
The standard deviation for quantitative determinations was ⫾ 5%. a
% of total sterols; nd: not detectable.
and a progressive increase in sterol end products was concomitantly observed (Table 23.1 and Figure 23.3A, 21st WAF). Throughout the fruit-ripening process, sterols continued to accumulate, with sitosterol remaining the major compound. At 30 WAF, the free sterol content of the olive fruit amounted to 950 μg g⫺1 dry wt, corresponding to a four-fold total increase from the 12th WAF. At this stage, a significant accumulation of some early intermediates, especially squalene (data not shown) and 24-methylenecycloartanol, was observed (Table 23.1 and Figure 23.3A, 30th WAF), indicating a slowing down of the metabolic flux through the sterol pathway. Between 21 and 30 WAF, the content of the olive fruit in sterol esters continued to rise, particularly between 27 and 29 WAF, to give a total amount corresponding to a seven-fold increase during the whole period of fruit development (Table 23.2). At the end of the ripening process (30 WAF), a significant accumulation of 24-methylenecycloartanol and 24-ethylidenelophenol was observed, in agreement with previous data on olive oil (Chryssafidis et al., 1992). It is interesting to note the occurrence of a new compound in the fraction of 4,4-dimethylsterols, which has been identified as parkeol (Table 23.2). This compound was formed concomitantly with cycloartenol and represented 60% of the total esterified 4,4-dimethyl sterols.
CHAPTER | 23 Sterols and Non-steroidal Triterpenoids of the Developing Olive Fruit
215
FIGURE 23.3 Changes in the content and the composition of free sterols (A) and free non-steroidal triterpenoids (B) throughout olive fruit development. The central part of the figure corresponds to the total content in free sterols and non-steroidal triterpenoids at different developmental stages of the olive fruit (12, 18, 21 and 30th WAF). Sterol pathway: 䊐 4,4-dimethylsterols; 䊏 4α-methylsterols; 4-demethylsterols. Non-steroidal pathway: pentacyclic triterpenes; triterpenic diols; mono- and di-HPTAs.
During the whole period of olive fruit development, it should be pointed out that free sterols remained predominant compared to ester conjugates (Tables 23.1 and 23.2).
23.3.3 Non-steroidal Triterpenoids from the 12th to the 18th WAF At the beginning of fruit ontogeny, besides sterols, the olive fruit were found to contain high levels of the pentacyclic triterpenes α- and β-amyrins, in a 3:2 ratio, as well as several more oxygenated compounds with additional hydroxymethyl or carboxylic groups (Table 23.3 and Figure 23.3B, 12th WAF). The introduction of a hydroxyl group in C-28 position of α- and β-amyrins gives rise to uvaol and erythrodiol,
respectively (Figures 23.1 and 23.2). A further oxidation of this hydroxyl group leads to the corresponding ursolic and oleanolic acids. Finally, the introduction of an additional hydroxyl group at the C-2 position of oleanolic acid results in the formation of maslinic acid. In addition to these oleanane- and ursane-type triterpenoids, lupane (3-epi-betulin and 3-epi-betulinic acid) and taraxane (taraxen-14-ene3β,28-diol) derivatives were also formed (Table 23.3). The enzymes involved in all these oxidation reactions have not been characterized yet, but are likely cytochrome P-450 monooxygenases. According to this hypothesis, the pathway should implicate the intermediate formation of the aldehydes 3β-hydroxy-5α-urs-12-en-28-al and 3β-hydroxy5α-olean-12-en-28-al. The presence of these compounds has not been checked. However, the occurrence of significant
216
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 23.3 Changes in free non-steroidal triterpenoids throughout olive fruit development.
TABLE 23.4 Changes in esterified non-steroidal triterpenes.
Developmental stage Triterpenoid classes
12 WAF
18 WAF
21 WAF
30 WAF
Developmental stage
Pentacyclic triterpenes ß-amyrin
4.3a
2.5
0.1
0.2
28-nor-β-amyrin
2.3
1.5
–
–
α-amyrin
6.0
3.3
–
–
28-nor-α-amyrin
0.8
0.4
–
–
Pentacyclic diols taraxerol
12 WAF
18 WAF
21 WAF
30 WAF
α-amyrin
5.5
5.5
nd
Nd
ß-amyrin
25.1
13.9
10.9
11.9
δ-amyrin
10.0
21.5
15.9
22.6
taraxerol
33.6a
21.7
11.6
5.6
butyrospermol
nd
nd
61.6
59.8
25.7
37.3
nd
nd
7.0
8.8
39
35
–
–
0.1
0.1
lupeol
taraxer-14-ene-3ß-28diol
0.2
0.3
–
–
Total amount (μg/g dry wt)
erythrodiol
12.5
9.7
0.4
0.5
The standard deviation for quantitative determinations was ⫾ 5%. a
uvaol
7.6
6.2
0.1
–
3-epi-betulin
0.3
0.2
–
–
Mono-HPTAs 3-epi-betulinic acid
0.7
0.6
0.6
0.6
oleanolic acid
39.2
35.8
41.2
37.6
Di-HPTAs ursolic acid
0.2
0.2
0.2
0.2
maslinic acid
25.9
39.3
57.3
60.8
Total amount (μg/g dry wt)
3210
2610
3930
2470
The standard deviation for quantitative determinations was ⫾ 10%. a
% of total free non-steroidal triterpenoids.
amounts of 28-nor-α-amyrin and 28-nor-β-amyrin, which have no methyl group at C-17 (Figure 23.2), in the pentacyclic triterpene fraction (Table 23.3), might result from the decarbonylation of such aldehydes (Hota and Bapuji, 1994). The somewhat delayed accumulation of maslinic acid (Table 23.3) suggests that the hydroxylation step at C-2 may involve another type of cyt P450-monooxygenase. Between 12 and 18 WAF, the non-steroidal triterpenoid pathway was very efficient as attested by the synthesis of very high levels of these compounds (i.e. 3 mg g⫺1 dry wt) (Table 23.3 and Figure 23.3B18th WAF). During the same period, ester conjugates of pentacyclic triterpenes were barely detectable (0.2% of the corresponding free forms) (Table 23.4). Traces of α- and β-amyrins, taraxerol, δ-amyrin and lupeol were found, with no change in the total amount of these esters between 12 and 18 WAF (Table 23.4).
% of total esterified non-steroidal triterpenes; nd, not detectable.
23.3.4 Non-steroidal Triterpenoids from the 21st to the 30th WAF Between 21 to 30 WAF, a dramatic decrease in the content of α- and β-amyrins and pentacyclic triterpenic diols was observed (Table 23.3 and Figure 23.3B, 21st WAF). Nonsteroidal triterpenoids were constituted almost exclusively of mono- and di-HPTAs, with oleanolic and maslinic acids as the largely predominant compounds (98% of total triterpenoids) (Table 23.3 and Figure 23.3B, 21st WAF). In the mature olive fruit (30 WAF), a significant decrease in the content of all HPTAs was observed (Table 23.3), indicating that these compounds might be further metabolized, maybe into triterpenic saponins via the involvement of specific glycosyltransferases (Achnine et al., 2005). To our knowledge, saponins from olive tree have not yet been identified. Between 21 and 30 WAF, when amyrins were not any longer formed, esterified conjugates began to progressively accumulate, especially δ-amyrin (23% at 30 WAF) (Table 23.4). Concomitantly, a change in the profile of triterpenes could be noticed, consisting in the appearance of a new tetracyclic triterpene, identified as butyrospermol, which represented up to 60% of total esterified triterpenes.
23.4 HOW IS CARBON FLUX REGULATED BETWEEN BOTH TRITERPENIC PATHWAYS IN THE OLIVE FRUIT? Evidence is presented here for the occurrence in the olive fruit of a vast array of sterols and non-steroidal triterpenoids. More than 40 different compounds have been found and the composition of this complex mixture was found to
CHAPTER | 23 Sterols and Non-steroidal Triterpenoids of the Developing Olive Fruit
be strongly dependent on the fruit developmental stage as illustrated in Figure 23.3. Throughout fruit ontogeny, two periods can be clearly distinguished: from 12 to 18 WAF and from 21 to 30 WAF. In the young green olive fruit (between 12 and 18 WAF), most of the available squalene molecules are almost exclusively devoted to the synthesis of α- and βamyrins. These non-steroidal pentacyclic triterpenes were rapidly metabolized into more oxygenated compounds, first into triterpenic alcohols, then into mono- and di-HPTAs. During the same period, no free sterols were formed. However, the sterol pathway remained functional as attested by the formation of sterol esters. Between 21 and 30 WAF, when the epidermal color gradually turned from green to purple, α- and β-amyrins and their hydroxylated derivatives were not present any longer, while the already-formed oxygenated intermediates were converted into mono- and diHPTAs (see the postulated biosynthetic pathway in Figure 23.1). Interestingly, our data also indicate that although α-amyrin was present in excess compared to β-amyrin, oleanane-type compounds as a whole were produced in far higher amounts than ursane-type compounds (Table 23.3). In early stages of fruit development, oleanane-type compounds represented from 84–89% of total non-steroidal triterpenoids and ursane-type compounds, only 10–15%. Between 21 and 30 WAF, ursane-type compounds completely disappeared. Thus, the question of the metabolic fate of α-amyrin in the olive fruit remains to be solved. From the 21st WAF, free and esterified sterols began to be formed and accumulated until the complete maturity of the fruit, but whatever the fruit developmental stage, nonsteroidal triterpenoids remained the major triterpenic compounds, with maslinic acid as the most represented one in the mature fruit (Table 23.3). Taken together, these results clearly indicate that a complex regulation process takes place at the oxidosqualene cyclization step, which represents a branch point between the sterol pathway and the non-steroidal triterpenoid pathway (Figure 23.1). OS serves indeed as a substrate for cycloartenol synthase, the first enzyme of the sterol pathway, but also for various OS cyclases involved in the synthesis of the different classes of pentacyclic triterpenes. These OS cyclases are designated as mono- or multifunctional enzymes, depending on whether they produce single or several cyclization products (Ebizuka et al., 2003). OS cyclases of the olive fruit have not been characterized yet, but might include a lupeol synthase, a mono-functional enzyme, similar to that identified in the olive leaf (Shibuya et al., 1999) as well as a multifunctional triterpene synthase, such as the OS cyclase recently identified in Olea cell suspension cultures and able to form mainly α-amyrin, but also β-amyrin and butyrospermol (Saimaru et al., 2007). Mechanisms underlying regulation of the carbon flux through both pathways remain to be elucidated. Such a regulation clearly involves interplay between several partners, including the different OS cyclases (cycloartenol synthase
217
and triterpenes synthases) but also several acyltransferases and maybe glycosyltransferases. Squalene and OS are synthesized in the membranes of the endoplasmic reticulum where cycloartenol synthase (as well as the other enzymes of the sterol pathway) and also probably triterpenes synthases are located. As suggested by the present work, the expression patterns of the different OS cyclases appear to be closely dependent on the stage of the fruit developmental process. Our results clearly indicate that, in early stages, synthesis of pentacyclic triterpenes, by one or several triterpene synthases, occurs concomitantly with acylation of sterols whereas the opposite situation (i.e. free sterol biosynthesis and esterification of pentacyclic triterpenes) is observed in later stages (after the 21st WAF). Thus, these acylation reactions, leading to the removal of the newly synthesized sterol intermediates or pentacyclic triterpenes from each respective pathway appears as a means to direct most of the available OS molecules toward only one pathway. Very little attention has been paid to acyltransferases involved in these reactions. In Arabidopsis, two different enzymes catalyzing the formation of sterol esters, via either a phospholipid (Banas et al., 2005) or a fatty acyl CoA as the acyl donor (Chen et al., 2007), have been recently characterized. The first enzyme seems to be specific to the sterol pathway as it is able to acylate various sterol end products as well as sterol intermediates, but not lupeol or ß-amyrin. The best substrate of the second enzyme was found to be cycloartenol, but whether or not this enzyme is also capable of forming acylated pentacyclic triterpenes has not been determined. It should be pointed out that sterol esters, which are not membrane components, are synthesized concomitantly with triacylglycerols (Stiti et al., 2007) and thus participate with them in the formation of olive fruit oil droplets. In conclusion, further work is needed to investigate more deeply the relationships between both triterpenic pathways in the olive fruit. The elucidation of the roles played in planta by the non-steroidal triterpenoids also appears to be a challenging objective. For example, these compounds are known to be constituents of epicuticular wax crystals and might be involved in plant–insect interactions (Guhling et al., 2006). The rising interest in the valuable biological properties for human health of non-steroidal triterpenoids, including maslinic acid (see Liu et al., 2007 and Chapter 158), constitutes an additional motivation to address these questions.
SUMMARY POINTS ●
Evidence is given here for the occurrence in the olive fruit of a vast array of sterols and non-steroidal triterpenoids, among which oleanane-type compounds are largely predominant. These two classes of compounds are synthesized via the mevalonate pathway and share common precursors.
218
●
●
●
●
SECTION | I
The composition of this complex mixture of triterpenoids was found to be closely dependent on the fruit developmental stage. From the 12th to the 18th WAF, the young green olive fruit contained high amounts of α- and β-amyrins along with hydroxylated pentacyclic alcohols and acids, but no new free sterols were formed. From the 21st WAF, when the epidermal color progressively turned from green to purple, α- and β-amyrins were not present any longer whereas free sterols began to be synthesized, indicating a re-direction of the carbon flux from the non-steroidal pathway toward the sterol pathway. Between 21 and 30 WAF, a two-fold increase in the content of free and esterified sterols was observed. Concomitantly, non-steroidal triterpenoids were represented almost exclusively by oleanolic and maslinic acids. These data clearly indicate that a complex regulation process takes place at the oxidosqualene cyclization step.
REFERENCES Abe, L., Rohmer, M., Prestwich, G.D., 1993. Enzymatic cyclization of squalene and oxidosqualene to sterols and triterpenes. Chem. Rev. 93, 2189–2206. Achnine, L., Huhman, D.V., Farag, M.A., Sumner, L.W., Blount, J.W., Dixon, R.A., 2005. Genomics-based selection and functional characterization of triterpene glycosyltransferascs from the model legume Medicago truncatula. Plant J. 41, 875–887. Azadmard-Damirchi, S., Savage, G.P., Dutta, P.C., 2005. Sterol fractions in hazelnut and virgin olive oils and 4,4-dimethylsterols as possible markers for detection of adulteration of virgin olive oil. J. Am. Oil Chem. Soc. 82, 717–725. Banas, A., Carlsson, A.S., Huang, B., Lenman, M., Banas, W., Lee, M., Noiriel, A., Benveniste, P., Schaller, H., Bouvier-Nave, P., Stymne, S., 2005. Cellular sterol ester synthesis in plants is performed by an enzyme (phospholipid:sterol acyltransferase) different from the yeast and mammalian acyl-CoA:sterol acyltransferases. J. Biol. Chem. 280, 34626–34634. Benveniste, P., 2002. Sterol metabolism. American Society of Plant Biologists, Rockville. http://www.bioone.org/archive/i1543-8120-38-1.pdf/ Bianchi, G., Pozzi, N., Vlahov, G., 1994. Pentacyclic triterpene acids in olives. Phytochemistry 37, 205–207. Caputo, R., Mangoni, L., Monaco, P., Previtera, L., 1974. New triterpenes from the leaves of Olea europaea. Phytochemistry 13, 2825–2827. Chen, Q., Steinhauer, L., Hammerlindl, J., Keller, W., Zou, J., 2007. Biosynthesis of phytosterol esters: identification of a sterol-Oacyltransferase in Arabidopsis. Plant Physiol. 145, 974–984. Chryssafidis, D., Maggos, P., Kiosseoglou, V., Boskou, D., 1992. Composition of total and esterified 4α-monomethylsterols and triterpene alcohols in virgin olive oil. J. Sci. Food Agric. 58, 581–583.
Lipids, Phenolics and Other Organics and Volatiles
Ebizuka, Y., Katsube, Y., Tsutsurni, T., Kushiro, T., Shibuya, M., 2003. Functional genornics approach to the study of triterpene biosynthesis. Pure Appl. Chem. 75, 369–374. Guhling, O., Hobl, B., Yeats, T., Jetter, R., 2006. Cloning and characterization of a lupeol synthase involved in the synthesis of epicuticular wax crystals on stem and hypocotyl surfaces of Ricinus communis. Arch. Biochem. Biophys. 448, 60–72. Hartmann, M.A., Benveniste, P., 1987. Plant membrane sterols: isolation, identification and biosynthesis. Methods Enzymol. 148, 632–650. Hota, R.K., Bapuji, M., 1994. Triterpenoids from the resin of Shorea Robusta. Phytochemistry 35, 1073–1074. Itoh, T., Yoshida, K., Yatsu, T., Tamura, T., Matsumoto, T., 1981. Triterpene alcohols and sterols of Spanish olive oil. J. Am. Oil Chem. Soc. 58, 545–550. Liu, J., Sun, H., Wang, X., Mu, D., Liao, H., Zhang, L., 2007. Effects of oleanolic acid and maslinic acid on hyperlipidemia. Drug Dev. Res. 68, 261–266. Mahato, S.B., Sarkar, S.K., Poddar, G., 1988. Triterpenoid saponins. Phytochemistry 27, 3037–3067. Perez-Camino, M.C., Cert, A., 1999. Quantitative determination of hydroxy pentacyclic triterpene acids in vegetable oils. J. Agric. Food Chem. 47, 1558–1562. Power, F.B., Tutin, F., 1908. The constituents of olive leaves. J. Chern. Soc. Trans. 93, 891–904. Rahier, A., Benveniste, P., 1989. Mass spectral identification of phytosterols. In: Nes, W.D., Parish, E. (eds) Analysis of Sterols and Other Biologically Significant Steroids. Academic Press, New York, pp. 223–250. Ranalli, A., Pollastri, L., Contento, S., Di Loreto, G., Iannucci, E., Lucera, L., Russi, F., 2002. Sterol and alcohol components of seed, pulp and whole olive fruit oils. Their use to characterise olive fruit variety by multivariates. J. Sci. Food Agric. 82, 854–859. Reina, R.J., White, K.D., Jahngen, E.G., 1997. Validated method for quantitation and identification of 4,4-desmethylsterols and triterpene diols in plant oils by thin-layer chromatography-high resolution gas chromatography-mass spectrometry. J. AOAC Int. 80, 1272–1280. Saimaru, H., Orihara, Y., Tansakul, P., Kang, Y.-H., Shibuya, M., Ebizuka, Y., 2007. Production of triterpene acids by cell suspension cultures of Olea europaea. Chem. Pharm. Bull. (Tokyo) 55, 784–788. Seo, S., Yoshimura, Y., Uomori, A., Takeda, K., Seto, H., Ebizuka, Y., Sankara, U., 1988. Biosynthesis of triterpenes, ursolic acid, and oleanolic acid in tissue cultures of Rabdosia japonica Hara fed [5-13C2H2 J mevalonolactone and [2-13C2H3] acetate. J. Am. Chem. Soc. 110, 1740–1745. Shibuya, M., Zhang, H., Endo, A., Shishikura, K., Kushiro, T., Ebizuka, Y., 1999. Two branches of the lupeol synthase gene in the molecular evolution of plant oxidosqualene cyclases. Eur. J. Biochem. 266, 302–307. Stiti, N., M’Sallem, M., Triki, S., Cherif, A., 2002. Etude de la fraction insaponifiable de l’huile d’olive de differcntcs varietes tunisiennes. Riv.ltal. Sostanze Grasse 79, 357–363. Stiti, N., Triki, S., Hartmann, M.A., 2007. Formation of triterpenoids throughout Olea europaea fruit ontogeny. Lipids 42, 55–67.
Chapter 24
A Comparison of the Volatile Compounds, in Spanish-style, Greek-style and Castelvetrano-style Green Olives of the Nocellara del Belice Cultivar: Alcohols, Aldehydes, Ketones, Esters and Acids Nadia Sabatini CRA-Centro Di Ricerca Per L’Olivicoltura E L’Industria Olearia, Pescara, Italy
24.1 INTRODUCTION Volatile compounds are low-molecular-weight compounds (less than 300 Da) which readily vaporize at room temperature. Some volatile compounds reach the human olfactory epithelium, dissolve into the mucus and may bond with olfactory receptors to give an odor sensation (Angerosa, 2002). Volatile organic compounds give much information about the components which characterize a specific product and they represent specific markers of product and process quality. Thus, volatile compounds describe all production pathways, from the harvest to the market. Aroma compound composition changes in relation to species, cultivar, physiological state of primary product, climacteric conditions, production areas, process technology and storage. Olive product, essentially olive oil, but also table olives, are the principal source of mono-unsaturated lipids of Mediterranean diet. The interest of scientific research about table olives has increased strongly in recent years, above all for the minor compounds. Table olives are currently the most important fermented vegetable products in the developed world. Worldwide production is around 1.5 million tons, of which nearly half is produced in the European Union, predominantly in Spain, Greece, Italy and Portugal (IOOC, 2005). For table olive consumption, the fruits are opportunely processed and served as an appetizer or as a complement to salads, pasta, pizza and other foods (Marsilio et al., 2008; Sabatini et al., 2008). In general, any processing method aims to remove the natural bitterness of the fruit, caused by the glucoside oleuropein. The primary purpose of table olive fermentation is to achieve Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
a preservation effect and to enhance the sensory properties of the final product (Panagou and Tassou, 2006). High-quality table olives are characterized by a pleasant fruit flavor. Table olive flavor is closely connected with the quali/quantitative composition of volatile compounds and it is considered as a quality index of olive products, playing an important role in consumer’s acceptability (Koprivnjak et al., 2002). Aroma compound composition in table olives depends on several factors, such as genetic factors, climatic conditions, ripening degree of fruits, and processing conditions (Garrido-Fernandez et al., 1997; Ruiz et al., 2005). A pleasant fragrance derives from the equilibrium of a number of volatile substances such as hydrocarbons, alcohols, aldehydes, ketones, esters and other compounds. Furthermore, changes in olive aroma allow a comparison between different cultivars and processing methods as well as to follow the evolution of quality during processing and to check off-flavors produced during fruit storage. Volatile compounds are not produced in significant amounts during fruit growth but arise during the climacteric stage of ripening and processing technology (Kalua et al., 2007).
24.2 DIFFERENCES BETWEEN VOLATILE COMPOUNDS IN VIRGIN OLIVE OIL AND IN TABLE OLIVES Although, in the literature, a lot is known about aroma compounds in virgin olive oil, very little is known about the quali/quantitative composition of volatile compounds
219
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
220
SECTION | I
in table olives, whose matrix is deeply different from that of olive oil. Olive oil volatile compounds are produced by enzymatic pathways during the milling process. Thus, in a good-quality olive oil, well-flavors are produced only by physical processes which activate endogenous enzymes, and not by a fermentative process as well as in table olives (Sabatini and Marsilio, 2008). Flavor compound formation in table olives is a dynamic process that develops mainly during the olive fermentation by lactic acid bacteria and yeasts together with a variety of contaminating microorganisms, which produce volatile compounds from major fruit constituents through various biochemical pathways (McFeeters, 2004; Sabatini and Marsilio, 2008). Thus, it is important not only to investigate olive fermentation by a microbiological point of view, but also to determine analytically volatile and semi-volatile compound contents and changing on their qualitative and quantitative profile during the entire making process until consumption (Sabatini and Marsilio, 2008).
24.3 AN INSIGHT ON VOLATILE COMPOUNDS BIOFORMATION The typical flavor of fruit such as bananas, peaches, pears, and cherries is not present during early fruit formation, but develops entirely during the ripening period. This flavor development period, or ripening, occurs during climacteric rise in respiration. During this period, metabolism of the fruit changes to catabolism and flavor formation begins (Heat, 1981). Aroma compounds are formed from major plant constituents (e.g. carbohydrates, lipids and protein) through different biochemical pathways. The most important ways by which volatiles are known to be formed are (Heat, 1981): ●
●
●
●
Volatiles formed during normal plant metabolism which remain in situ when the plant is harvested (e.g. essential oils, fruit and vegetable flavors). Volatile compounds resulting from endogenous enzyme-catalyzed reactions during physical processes (e.g. olive oil milling process). Volatiles produced as a result of microbiological action and fermentation (wine, tea, table olives). Volatile compounds produced during heat processing or cooking.
24.4 FERMENTATION Fermentation is a technique that has been employed for generations to preserve food for consumption at a later date and to improve food security. The lowering of the pH operated from lactic acid bacteria inhibits the growth of food spoiling or poisoning bacteria and destroys certain pathogens (FAO document repository; Hammes and Tichaczek, 1994).
Lipids, Phenolics and Other Organics and Volatiles
Furthermore, fermentation can improve the flavor and appearance of food. In food fermentations the byproducts play a beneficial role in preserving and changing the texture and flavor of the food substrate. Olive fermentation involves the breaking down of complex organic substances into simpler ones. Fermenting olives are typically very complex ecosystems with active enzyme systems from the ingredient material interacting with the metabolic activities of microorganisms (Sabatini and Marsilio, 2008). The changes that occur during fermentation of foods are the result of enzymatic activity. Enzymes are complex proteins produced by living cells to carry out specific biochemical reactions. Lactic acid bacteria directly influence the flavor of fruits, contributing to the development of typical characteristics of fermented olives. The lactic acid bacteria are a group of Gram-positive bacteria, non-respiring non-spore-forming, cocci or rods, which produce lactic acid as the major end product of the fermentation of carbohydrates. The lactic acid bacteria belong to two main groups: the homofermenters and the heterofermenters. The pathways of lactic acid production differ for the two. Homofermenters produce mainly lactic acid, via the glycolytic (Embden-Meyerhof) pathway. Heterofermenters produce lactic acid plus appreciable amounts of ethanol, and carbon dioxide, via the 6-phosphoglucanate/phosphoketolase pathway (Eqs. 24.1 and 24.2, FAO corporate document repository). HOMOLACTIC FERMENTATION C6 H12 O6 → 2 CH3 CHOHCOOH GLUCOSE
(24.1)
LACTIC ACID
HETEROLACTIC FERMENTATION C6H12O6 → CH3CHOHCOOH ⫹ C2H 5OH ⫹ CO2 GLUCOSE
LACTIC ACID
ETHANOL
CARBON DIOXIDE
(24.2) The inhibitory effect of lactic acid bacteria is due to the accumulation of the main primary metabolites (lactic and acetic acids, ethanol and carbon dioxide) as well as the production of other antimicrobial compounds, such as formic and benzoic acids, hydrogen peroxide, diacetyl, acetoin and bacteriocins (Delgado et al., 2001). By lowering the pH of the medium, lactic acid bacteria may reduce the activity or completely inactivate endogenous enzymes of the plant material, generating either flavor components or flavor precursor compounds. Yeasts and yeast-like fungi are widely distributed in nature. They are present in orchards and vineyards, in the air, the soil and in the intestinal tract of animals. Like bacteria and molds, yeasts can have beneficial and non-beneficial effects in foods. Yeasts produce ethyl-alcohol and carbon dioxide from simple sugars such as glucose and fructose (Eq. 24.3; FAO corporate document repository). The production of vinegar (acetic acid) depends on mixed fermentation, which involves both yeasts and bacteria. The fermentation is usually initiated by
221
CHAPTER | 24 A Comparison of the Volatile Compounds
yeasts which break down glucose into ethyl alcohol with the liberation of carbon dioxide gas. ALCOHOLIC FERMENTATION C6 H12 O6 → 2C2 H 5 OH ⫹ 2CO2 GLUCOSE
ETHANOL
(24.3)
CARBON DI OXIDE
Following on from the yeasts, Acetobacter and Clostridium acetobutylicum oxidize the alcohol to acetic acid and water (Eq. 24.4; FAO corporate document repository). These bacteria are commonly found in foodstuffs, water, and soil. OXIDATION OF ETHANOL TO ACETIC ACID O2
C2 H 5 OH → CH3 COOH ⫹ H 2 O ETHANOL
ACETIC ACID
(24.4)
WATER
The acetyl group derived from acetic acid when bound to Coenzyme A (CoA) is central to the metabolism of carbohydrates and fats. Propionic acid is produced biologically by Propionibacterium (PAB) species as the product of the metabolic breakdown of fatty acids and/or some amino acids (http://en.wikipedia.org/wiki; Marsilio et al., 1996; Jimenez et al., 1997; Bianchi, 2003). Enzymes produced by cited microorganisms may also directly metabolize precursors of aroma compounds (McFeeters, 2004). Moreover, some alcohols, esters, aldehydes, and ketones as well as other acids are known to be formed by microorganisms which are competitive with lactic acid bacteria in brined olives (Fleming et al., 1969). The biosynthesis of higher alcohols is considered to be linked to amino acid metabolism (Herrero et al., 2006).
24.5 LIPOXYGENASE PATHWAY Among the catabolic pathways related to lipids, the lipoxygenase (LOX) pathway plays an important role in the case of olive oil from the point of view of its quality (Gardner, 1991; Salas, 2000; Williams and Harwood, 2000). Unsaturated six carbon aldehydes and alcohols are present in high portions in the volatile fractions of some other fruits and vegetables and their sensory qualities are related to the so-called ‘green odor’ or ‘green notes’ (Salas, 2000). Although there are surely different biogenesis pathways of volatile compounds between table olives and olive oil, it is indeed important to consider some enzymatic pathways which occur in olive oil flavor compound biosynthesis (Sabatini and Marsilio, 2008). Lipoxygenases are enzymes that are ubiquitous in the plant and animal kingdoms. They catalyze the oxygenation of polyunsaturated fatty acids containing a 1,4-Z,Z-pentadiene moiety using molecular oxygen (Williams and Harwood, 2000). C5 and C6 aldehydes and alcohols, responsible for green attributes of virgin olive oils, have been proved to be produced through the enzymatic oxidation of linolenic and linoleic acids
(Hatanaka, 1993; Olias and Pérez, 1993). Once unsaturated fatty acids have been oxidized to their hydroperoxides, the next step of the LOX pathway consists of their scission by hydroperoxide lyase (HPL). HPL catalyzes the cleavage of the C–C bond located between hydroperoxide function and the double bond with trans (E) configuration in the 13(S) or 9(S) hydroperoxides of linolenic and linoleic acids. The products of this reaction are volatile aldehyde and nonvolatile oxoacid (Salas et al., 2000; Angerosa, 2002; Ridolfi et al., 2002). The metabolism of 13-hydroperoxides of linolenic acid gives rise to cis-3-hexenal which can isomerize to trans-2-hexenal or it is either quickly enzymatically reduced to cis-3-hexen-1-ol. 13-Hydroperoxides of linoleic acid are cleaved by hydroperoxide lyses producing hexanal, which is reduced to 1-hexanol by alcohol dehydrogenases (ADHs) (Kiritsakis, 1998; Angerosa et al., 1999, 2000; Salas et al., 2000; Feussner and Wasternack, 2002; Salas, 2004) (Figure 24.1).
24.6 VOLATILE ESTERS BIOSYNTHESIS Esters are very important flavor compounds and occur widespread in nature in a great variety of foodstuffs. Esters are compounds with high perception thresholds and typically exhibit fruity, sweet and green notes (Salas, 2004). There are many reports in the literature which show that esters are biosynthetic products of many microorganisms (Patterson et al., 1992). Volatile esters are major components of the aroma of fruits such as strawberries, bananas, apples, sometimes being the compounds mainly responsible for the well-flavor appreciated by consumers (Seimour, 1993; Sanz et al., 1997). It is well known that acetate esters are synthesized by an enzyme alcohol-acyl-transferase (AAT) which catalyzes the esterification of volatile alcohols with acetyl CoA molecules to produce volatile esters and free CoA-SH (Salas, 2004). Furthermore, volatile alcohols can be formed by aldehyde reduction by action of ADHs enzymes, broadly distributed among the different plant species that catalyze the reversible reduction of aldehydes to alcohols in a reaction dependent on pyridine nucleotides.
24.7 INVESTIGATION OF THE VOLATILE COMPOUND CONTENTS IN SPANISH-STYLE, GREEK-STYLE AND CASTELVETRANO-STYLE GREEN OLIVES OF THE NOCELLARA DEL BELICE CULTIVAR 24.7.1 Nocellara del Belice Cultivar Fruit Characteristics This cultivar originates from the Italian island of Sicily and is considered the best green table olive in Italy. It has a firm crisp texture with a creamy, nutty taste.
222
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
LIPOXYGENASE PATHWAY Olive lipids Acyl-hydrolase
Linoleic acid
Linolenic acid
Lypoxigenase
9-hydroperoxide
13-hydroperoxide
13-hydroperoxide
9-hydroperoxide
Hydroperoxide lyase Cis 3: trans 2 enal-isomerase
hexanal
Cis-3-hexenal
Trans-2-hexenal
1-hexanol
Cis-3-hexen-1-ol
Trans-2-hexen-1-ol
hexyl-acetate
Cis-3-hexenil-acetate
Trans-2-hexenilacetate
Alcohol dehydrogenase
Alcohol acyltransferase
FIGURE 24.1 Lipoxygenase pathway in virgin olive oils (adapted from Olias et al., 1993 and Ridolfi et al., 2002).
FEATURES OF NOCELLARA DEL BELICE CULTIVAR
24.8 MAIN TABLE OLIVE PREPARATIONS FOR NOCELLARA DEL BELICE CULTIVAR 24.8.1 Greek-style Method
Nocellara del Belice Almost spherical-shaped fruits used both for table olives and olive oil extraction • • • • •
Sizes of 150 fruits kg−1 Flesh-to-pit ratio around 7:1 Flesh 86–88% Polar diameter: 24.0 Equatorial diameter: 21.0
FIGURE 24.2 Most principal features of olive fruits Nocellara del Belice cultivar.
24.7.2 Diffusion Area ●
●
West Sicily in the Trapani province (Belice valley) mainly in the territories of Castelvetrano, Campobello di Mazara and Partanna’ Crop production 25 000 tonnes of which about 10 000– 15 000 tonnes is processed as table olives (Figure 24.2).
Washed olives, green-ripe or black-ripe, are directly placed in the brine (8–10% w/v sodium chloride in potable water) where they undergo a spontaneous fermentation at room temperature (IOOC, 2004).
24.8.2 Spanish-style Method Washed olives, generally green-ripe, are placed into tanks and soaked with lye solution (NaOH 1–2% w/v food-grade sodium hydroxide in potable water) for up to 8–12 hours to debitter olives. The lye is allowed to penetrate through three-quarters of the flesh, leaving a small volume of flesh around the stone unaffected. Then the olives are washed with potable water and filled with brine (8–10% w/v foodgrade sodium chloride in potable water). A lactic fermentation step proceeds (Zervakis).
24.8.3 Castelvetrano-style Method Washed olives are treated with a mixture of lye (NaOH) and solid salt (NaCl). Olives are spontaneously fermented at room temperature.
CHAPTER | 24 A Comparison of the Volatile Compounds
Dynamic headspace technique Nitrogen (N2)
Volatile compounds Charcoal
Water bath Temperature 33°C
FIGURE 24.3 Description of dynamic headspace technique for olive fruits. Source: Sabatini, N. and Marsilio, V. (2008). Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chemistry, 107, 1522–1528.
24.9 DYNAMIC HEADSPACE ANALYSIS 24.9.1 Volatile Compounds Extraction from Fruit Matrix of Nocellara del Belice table olives (source: Sabatini and Marsilio, 2008) Dynamic headspace analysis is a convenient method for the enrichment of volatiles from complex matrices, combining the advantage of low detection limits with minimal sample preparation (Ruther and Hilker, 1998). Dynamic headspace method (Solinas et al., 1987), largely used to analyze quality and quantity flavor molecules of olive oil, has been updated and used to extract volatile compounds of Nocellara del Belice olives. Chemical–physical characteristics of table olives are different from those of olive oil. Fermented olive is a solid phase made up by hydrophilic and hydrophobic portions, while olive oil is only a hydrophobic liquid phase. For this reason it has been necessary to change some parameters linked with the technique described for olive oil. Temperature of extraction was diminished to 30–33°C (37°C for olive oil) so as to avoid excessive water evaporation, which could inactivate charcoal (table olives contain a high concentration of water); charcoal concentration was increased to 100 mg (30 mg for olive oil); volume of diethyl ether (elution solvent) was diminished to 1 mL (1.5 mL for olive oil). So, 60 grams of stoned olive fruits were put into a 120 mL Drechsel gas washing bottle, volatiles were stripped with N2 (1.0 dm3 min⫺1) at 33°C for 2 h, trapped on 100 mg of activated charcoal and then eluted with 1 mL of diethyl ether (Figure 24.3).
24.9.2 Analytical Aspects (source: Sabatini and Marsilio, 2008) The volatile compounds in Spanish-style, Greek-style and Castelvetrano-style green olives of Nocellara del Belice cultivar after 6, 7 and 8 months brining were extracted by
223
dynamic headspace technique and analyzed by gas chromatography and GC/mass spectrometry. Figure 24.4 shows a typical gas chromatogram of olive fruit’s flavor compounds obtained by dynamic headspace technique. Details on peak identities and volatile compounds amounts expressed in μg(compound)/Kg(olive fruit) are shown in Table 24.1. Among identified organic compounds there were alcohols, aldehydes, ketones, esters as well as acids of low molecular weight. In Figure 24.5 the molecular structures of volatiles identified by gas chromatography and GC/mass spectrometry are represented (Figure 24.5). Figure 24.6 (data from Table 24.1) displays the meaningful differences in volatile composition of three olive samples found after 6 months’ brining. High contents of ethanol, acetic acid and 2-butanol were detected in all samples. 1Propanol, propyl-acetate, ethyl-propanoate and propionic acid were found at greater levels in Spanish-style olives. Good amounts of propyl-propanoate were detected only in this commodity. Ethyl-acetate was found in all samples but particularly in Spanish-style olives. Cis-3-hexen-1-ol and 2-butanone were detected above all in Castelvetrano-style olives, while 1-hexanol and isopentanol were mainly found in Spanish- and Greek-style olives. Major levels of 2-pentanol and 3-pentanol were found in Castelvetrano- and Spanishstyle with respect to Greek olives. Furthermore a lower content of 3-hydroxy-2-butanone (acetoine) was revealed in Spanish-style olives. Figure 24.7 (data from Table 24.1) discloses the time evolution of some volatile compounds found in Greek- and Spanish-style olives after 6, 7, and 8 months’ brining. In Greek-style samples there was a meaningful increase on time of ethyl-propanoate, 2-butanone and propionic acid contents (Figure 24.7) but, on the other hand, many molecules increased their contents at 7 months and diminished at 8 months (Table 24.1). Also Castelvetrano-type olives showed an increase in the contents of most volatile compounds at 7 months’ processing and a lessening thereafter (Table 24.1), while Spanish-style samples disclosed a meaningful decrease of most molecules such as 2-butanol, ethanol and cis-3-hexen-1-ol amounts (Figure 24.7). Results showed that during olive processing all three samples underwent alcoholic, heterolactic, and propionic fermentation, which led to the production of great amounts of ethanol, acetic acid, propionic acid and other alcohols and esters. After 6 months’ processing, Castelvetranostyle olives showed higher contents of ethanol and lower amounts of acids, suggesting a major alcoholic fermentation. Furthermore, levels of 3-hydroxy-2-butanone (acetoine), which is a normal product of alcoholic fermentation (Romano and Suzzi, 1996), were significantly lower in Spanish-style olives with respect to Castelvetrano ones. On the other hand, Spanish-style olives disclosed higher contents of acids such as acetic acid and propionic acid, suggesting a probably major proliferation of Acetobacter
224
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
Typical GC chromatogram of volatile compounds in table olives 1
6
3
7
14
18 19
21
22
I.S.
FID response
2 10
17
4 1213 11
5
15 + 16 20
8 9 15
20
25
30
35
40
45
50
55
Time (min) FIGURE 24.4 Typical gas chromatogram of flavor compounds in Greek-style sample. See Table 24.1 for each number’s description. Source: Sabatini, N. and Marsilio, V. (2008). Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chemistry, 107, 1522–1528.
and PAB in the brine medium. These evidences were confirmed by higher contents of propionic and acetic acid esters. In addition, such results were in agreement with microbiological analyses which evidenced the presence of higher concentration of yeasts in Castelvetrano samples (1.7 ⫻ 109 CFU mL⫺1 brine), while a very lower value (4 ⫻ 106 CFU mL⫺1 brine) was detected in Spanish-style samples. Also, isobutanol, isopentanol and 1-propanol, which were contained in all three samples, were surely products of yeasts fermentation (Arrizon et al., 2006). High concentration of 2-butanol indicated that appreciable microbial activity in addition to lactic acid bacteria occurred in all samples (Fleming, 1969). As previously assessed, volatile esters are major components (in different extent) of the aroma of all fruits, sometimes being the compounds mainly responsible for the well-flavor appreciated by consumers (Seimour, 1993; Sanz et al., 1997). The volatile alcohols produced by the ADHs can be esterified with acetyl-CoA molecules to produce volatile esters by the action of AAT (Ke et al., 1994). This enzyme has been studied in fruits like apple, strawberry and banana, in which esters are important components of their aroma (Myers et al., 1970; Tressl and Drawert, 1973; Salas, 2004). In the same way propanoate esters could be synthesized by esterification of volatile alcohols (e.g. ethanol, 1-propanol) with propionyl CoA molecules. From these results and the cited references, a proposed pathway for esters biosynthesis in table olives is presented in Figure 24.8.
A LOX pathway may also be hypothesized (Figure 24.1) for three samples, since some aroma molecules which are formed in this have been found, as shown by the presence of 1-hexanol, 1-hexanal and cis-3-hexen-1-ol. In particular, after 6 months’ brining, higher levels of 1-hexanol in Spanish-style olives disclosed a probably prevalent linoleic acid oxidation pathway, while higher amounts of cis3-hexen-1-ol showed a probably prevalent linolenic acid oxidation pathway in Castelvetrano-style olives. On the other hand, in the Greek-style sample it seems to be a balance between linolenic and linoleic oxidation. However, a ‘lipoxygenase-like’ metabolism of polyunsaturated fatty acids affected by enzymes produced (in brine medium) by lactic acid bacteria, yeasts together with other contaminating microorganisms could be also considered. Referring to cited literature, an overall view of volatile compounds biogenesis during ripening and processing of table olives is shown in Figure 24.9. Furthermore, the strong differences observed between 6, 7, and 8 months of brining (Figure 24.7) are explained by time evolution of room temperature which changed dynamics of proliferative activity of microorganisms and consequently the volatile compounds’ metabolic pathways. This research showed that processing technology affected significantly the volatile compounds of table olives of Nocellara del Belice cultivar. Therefore, quali/quantitative evaluation of aroma compounds is important in order to determine a quality index of final products. In these samples, only wellodors have been found because olives underwent a good
After 6 months’ brining
After 7 months’ brining
After 8 months’ brining
Compound
Greek style
Spanish style
Castelvetrano style
Greek style
Spanish style
Castelvetrano style
Greek style
Spanish style
Castelvetrano style
1.
Ethyl-acetate
26.7 ⫾ 1
45 ⫾ 3.4
40 ⫾ 2.5
315 ⫾ 15
37 ⫾ 1.5
240 ⫾ 1.5
61 ⫾ 3.6
14 ⫾ 0.8
57 ⫾ 3
2.
2-Butanone
30.6 ⫾ 2
91 ⫾ 6.1
182 ⫾ 15
152 ⫾ 10
50 ⫾ 3
440 ⫾ 17
117 ⫾ 1.4
32 ⫾ 1.4
107 ⫾ 5
3.
Ethanol
474 ⫾ 30
701 ⫾ 20
1119 ⫾ 69
2326 ⫾ 105
369 ⫾ 18
2020 ⫾ 89
740 ⫾ 5.2
200 ⫾ 10
580 ⫾ 22
4.
Propyl-acetate
0.8 ⫾ 0.07
83 ⫾ 15
3.3 ⫾ 0.2
17 ⫾ 1.1
53 ⫾ 2.7
25 ⫾ 1.4
17 ⫾ 0.9
33 ⫾ 1
9 ⫾ 0.4
5.
Ethyl-propanoate
2.2 ⫾ 0.15
130 ⫾ 11
2.2 ⫾ 0.1
49 ⫾ 2.3
77 ⫾ 5.5
0.9 ⫾ 0.03
53 ⫾ 1
43 ⫾ 1
5 ⫾ 0.1
6.
2-Butanol
1020 ⫾ 70
2408 ⫾ 90
1496 ⫾ 99
6208 ⫾ 220
1270 ⫾ 68
3375 ⫾ 115
2133 ⫾ 65
670 ⫾ 21
1220 ⫾ 70
7.
1-Propanol
45.3 ⫾ 3
1527 ⫾ 11
109 ⫾ 8
1153 ⫾74
825 ⫾ 16
323 ⫾ 19
30 ⫾ 1.7
534 ⫾ 2
110 ⫾ 6
8.
Propyl-propanoate
n.d.
73 ⫾ 4
1.1 ⫾ 0.08
n.d.
58 ⫾ 3
n.d.
3.6 ⫾ 0.15
42 ⫾ 1.4
n.d.
9.
1-Hexanal
2.2 ⫾ 0.3
0.8 ⫾ 0.02
1.7 ⫾ 0.05
2.2 ⫾ 0.1
n.d.
n.d.
1.1 ⫾ 0.6
0.3 ⫾ 0.01
0.5 ⫾ 0.05
10.
Isobutanol
9.4 ⫾ 0.4
36 ⫾ 2.3
7.1 ⫾ 0.65
68 ⫾ 4.2
8.3 ⫾ 3
16 ⫾ 0.6
26 ⫾ 1.2
2.2 ⫾ 0.1
6 ⫾ 0.04
11.
3-Pentanol
2.5 ⫾ 0.14
14 ⫾ 1.2
14 ⫾ 0.12
12 ⫾ 0.9
8.3 ⫾ 4.5
28 ⫾ 1
7.4 ⫾ 0.3
4 ⫾ 0.1
10 ⫾ 0.8
12.
2-Pentanol
1.7 ⫾ 0.15
10.5 ⫾ 0.9
12 ⫾ 0.9
16 ⫾ 1.2
6.6 ⫾ 2.2
25 ⫾ 0.7
10 ⫾ 0.2
3.3 ⫾ 0.08
9 ⫾ 0.09
CHAPTER | 24 A Comparison of the Volatile Compounds
TABLE 24.1 Volatile compounds amounts of Nocellara del Belice cultivar, after 6, 7, 8 months’ brining extracted by dynamic headspace technique and analyzed by gas chromatography/mass spectrometry, expressed in μg(compound)/Kg(olive fruit).
(Continued)
225
226
TABLE 24.1 (Continued ) After 6 months’ brining
After 7 months’ brining
After 8 months’ brining
Greek style
Spanish style
Castelvetrano style
Greek style
Spanish style
Castelvetrano style
Greek style
Spanish style
Castelvetrano style
13.
1-Butanol
3 ⫾ 0.1
25 ⫾ 1.3
9.3 ⫾ 0.4
15 ⫾ 0.9
17 ⫾ 1
19 ⫾ 0.8
5.2 ⫾ 2.3
12 ⫾ 0.06
31 ⫾ 2
14.
Isopentanol
178 ⫾ 11
138 ⫾ 11
41 ⫾ 0.02
692 ⫾ 26
194 ⫾ 13
94 ⫾ 3
507 ⫾ 25
96 ⫾ 4
34 ⫾ 3
15.
1-Pentanol
4.7 ⫾ 0.5
0.5 ⫾ 0.01
6.9 ⫾ 0.04
0.8 ⫾ 0.04
0.5 ⫾ 0.01
11 ⫾ 0.9
0.8 ⫾ 0.02
1.7 ⫾ 0.08
3.6 ⫾ 0.1
16.
4-Penten-1-ol
0.5 ⫾ 0.04
18.5 ⫾ 1.8
17 ⫾ 0.13
25 ⫾ 1.8
14 ⫾ 1.2
36 ⫾ 1.6
15 ⫾ 0.8
4.4 ⫾ 0.14
10 ⫾ 1
17.
3-Hydroxy-2butanone
13.2 ⫾ 1.4
1.7 ⫾ 1.1
15 ⫾ 0.09
3 ⫾ 2.5
n.d.
37 ⫾ 2.3
2.2 ⫾ 0.08
0.5 ⫾ 0.03
7 ⫾ 0.2
18.
1-Hexanol
36 ⫾ 2.5
84.5 ⫾ 5
21 ⫾ 0.13
165 ⫾ 9
61 ⫾ 3.2
34 ⫾ 1.2
8 ⫾ 0.3
24 ⫾ 1.5
10 ⫾ 0.9
19.
Cis-3-hexen-1-ol
32.3 ⫾ 2
124 ⫾ 9
154 ⫾ 12
113 ⫾ 11
94 ⫾ 5.1
262 ⫾ 1.2
65 ⫾ 3.2
34 ⫾ 1.4
74 ⫾ 3
20.
Nonanal
10 ⫾ 0.8
3 ⫾ 0.01
3 ⫾ 0.06
12 ⫾ 0.6
7 ⫾ 2.1
12 ⫾ 0.8
9 ⫾ 0.4
3 ⫾ 0.1
4 ⫾ 0.2
21.
Acetic acid
3050 ⫾ 150
6940 ⫾ 195
2688 ⫾ 165
13700 ⫾ 660
6500 ⫾ 165
3953 ⫾ 150
4561 ⫾ 250
3347 ⫾ 120
1192 ⫾ 56
22.
Propionic acid
39.5 ⫾ 3
3442 ⫾ 230
111 ⫾ 8.5
560 ⫾ 24
3645 ⫾ 1.5
159 ⫾ 9
817 ⫾ 35
1944 ⫾ 54
120 ⫾ 6
Each value is the mean of triplicate analyses expressed in μg(compound)/Kg(olive fruit) ⫾ SD (Standard Deviation); n.d.: not determined. Source: Sabatini, N., Marsilio, M., 2008. Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chem. 107, 1522–1528.
SECTION | I Lipids, Phenolics and Other Organics and Volatiles
Compound
227
CHAPTER | 24 A Comparison of the Volatile Compounds
Molecular structures of volatile compounds in table olives Alcohols OH OH Ethanol
Isobutanol
OH OH 2-pentanol
1-propanol
HO
OH OH
Cis-3-hexen-1-ol
1-butanol OH
2-butanol OH OH 1-pentanol
4-penten-1-ol
Isopentanol OH
1-hexanol OH 3-pentanol
Esters O
Ketones
Aldehydes
O
1-hexanal
O
O O ethyl-acetate O
1-nonanal
2-butanone
Acids
Hydroxy-Ketones
O O propyl-acetate
O OH Acetic acid
O
O O Ethyl-propanoate O
OH 3-hydroxy-2-butanone
OH Propionioc acid
O Propyl-propanonate FIGURE 24.5 Molecular structures of volatile compounds of Spanish-, Greek- and Castelvetrano-style green olives of Nocellara del Belice cultivar identified by gas chromatography and GC/mass spectrometry. Source: Sabatini, N. and Marsilio, V. (2008). Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chemistry, 107, 1522–1528.
fermentation process (no coliforms have been revealed in brine medium). Thus, it will be very useful, in the future, to identify also off-odors formed by anomalous fermentations, in order to reveal them, just in small traces, in premature times, so to obtain the recovery of the fermentation process.
SUMMARY POINTS ●
High-quality table olives are characterized by a pleasant fruit flavor. Table olive flavor is tightly connected with the quali/quantitative composition of volatile compounds and it is considered as a quality index
228
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
Meaningful differences in volatile composition after 6 months’ brining A 250 Greek style
Spanish style
Castelvetrano style
200 150 100 50
oa
yl -p
ro p
an
an ro p Pr op
l-p hy
Et
te
te oa
te ta ce
1-
yl -a
nPr op
xe he 3-
is -
3-
hy
C
dr ox
ol
l no he
1-
ta
y-
2-
bu
pe Is o
xa
no
an nt
nt pe
ne
ol
ol an
ol an 2-
3-
pe
nt
ta bu 1-
no ta
2-
bu
no
ne
te ta ce l-a hy Et
l
0
Greek style
B 7000
Spanish style
Castelvetrano style
6000 μ 5000 4000 3000 2000 1000 0 Ethanol
2-butanol
1-propanol
Acetic acid
Propionic acid
FIGURE 24.6 Meaningful differences (data from Table 24.1) in volatile composition of Spanish-, Greek- and Castelvetrano-style green olives of Nocellara del Belice cultivar after 6 months’ processing. (A) Ethyl-acetate; 2-butanone; 1-butanol; 3-pentanol; 2-pentanol; Isopentanol; 3-hydroxy-2-butanone; 1-hexanol; Cis-3-hexen-1-ol; propyl-acetate; ethyl-propanoate; propyl-propanoate. (B) Ethanol; 2-butanol; 1-propanol; acetic acid; propionic acid. Source: Sabatini, N. and Marsilio, V. (2008). Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chemistry, 107, 1522–1528.
●
of olive products. Volatile and semi-volatile organic compounds present both in the sample matrix and in the headspace aroma are responsible for the olive fruit aroma influencing the consumer’s compliance. Aroma compounds composition change in relation to species, cultivar, physiological state of primary product, climacteric conditions, production areas, process technology and storage. Aroma compounds are formed from major plant constituents (e.g. carbohydrates, lipids and protein) through different biochemical pathways: (a) Volatiles formed during normal plant metabolism which remain in situ when the plant is harvested (e.g. essential oils, fruit and vegetable flavors). (b) Volatile compounds resulting
●
from endogenous enzyme-catalyzed reactions during physical processes (e.g. olive oil milling process). (c) Volatiles produced as a result of microbiological action and fermentation (wine, tea, table olives). (d) Volatile compounds produced during heat processing or cooking. Fermentation is a technique that has been employed for generations to preserve food for consumption at a later date and to improve food security. Furthermore, fermentation can improve the flavor and appearance of food. In food fermentations the byproducts play a beneficial role in preserving and changing the texture and flavor of the food substrate. Lactic acid bacteria, yeasts and other microorganisms directly influence
229
CHAPTER | 24 A Comparison of the Volatile Compounds
Evolution on time of some volatile compounds A
900
Greek style
800 700 600
6 months’ brining 7 months’ brining 8 months’ brining
500 400 ●
300 200 100 0 2-butanone B 3000
Ethyl-propanoate
Propionic acid
Spanish style
2500
2000
6 months’ brining
1500
7 months’ brining 8 months’ brining ●
1000
500
0 2-butanol
Ethanol
Cis-3-hexen-1-ol
FIGURE 24.7 Evolution on time, after 6, 7 and 8 months of brining (data from Table 24.1) of some volatile compounds contained in Greek- and Spanish-style green olives of Nocellara del Belice cultivar. (A) 2-butanone; ethyl-propanoate; propionic acid. (B) 2-butanol; ethanol; cis-3-hexen-1-ol. Source: Sabatini, N. and Marsilio, M. (2008). Volatile compounds in table olives (Olea europaea L., Nocellara del Belice cultivar). Food Chemistry, 107, 1522–1528.
●
the flavor of fruits, contributing to the development of typical characteristics of fermented olives. The lactic acid bacteria belong to two main groups; the homofermenters and the heterofermenters. Homofermenters produce mainly lactic acid, while heterofermenters produce lactic acid plus appreciable amounts of ethanol and carbon dioxide. Yeasts produce ethyl-alcohol and carbon dioxide from simple sugars such as glucose and fructose. Following on from the yeasts, Acetobacter and Clostridium acetobutylicum oxidize the alcohol to acetic acid and water. Among the catabolic pathways related to lipids, the lipoxygenase (LOX) pathway plays an important role in olive fruit. Lipoxygenases are enzymes that are
●
●
ubiquitous in the plant and animal kingdoms. Unsaturated five and six carbon aldehydes and alcohols are present in high percentage in the volatile fractions of virgin olive oils and some other fruits and vegetables and their sensory qualities are related to the so-called ‘green odor’ or ‘green notes’. Furthermore, C5 and C6 aldehydes and alcohols have been proved to be produced through the lipoxygenase-mediated enzymatic oxidation of linolenic and linoleic unsaturated fatty acids. This chapter contains a recent published study concerning the volatile compounds’ quali/quantitative composition in Spanish-style, Greek-style and Castelvetrano-style green olives of the Nocellara del Belice cultivar. Olives, hand harvested at the green ripening stage, were processed by Spanish, Greek and Castelvetrano methods. Briefly, the olives were Lyetreated to remove the natural bitterness, washed with water to remove the excess of alkali and then placed in brine where they underwent a spontaneous lactic fermentation (Spanish-style method) at room temperature. In the Greek method the olives were directly brined (8–10% w/v food grade NaCl in potable water), while in the Castelvetrano the olives were treated with a mixture of Lye and solid salt (NaCl). Also in the Greek and Castelvetrano methods olives were spontaneously fermented at room temperature. The volatile compounds in Spanish-style, Greek-style and Castelvetrano-style green olives of Nocellara del Belice cultivar after 6, 7 and 8 months’ brining were extracted by dynamic headspace technique and analyzed by gas chromatography and gc/mass spectrometry. Dynamic headspace method, largely used to analyze quality and quantity flavor molecules of virgin olive oil, has been updated by authors and used to extract volatile compounds of Nocellara del Belice olives. Formed during olive fermentation (after 6, 7, 8 months’ brining) aroma compounds comprised alcohols, aldehydes, ketones, esters, as well as acids (of low molecular weight). The different processing technologies affected significantly the volatile compounds of three samples. In particular, meaningful differences in aromatic profiles not only qualitatively, but above all, from a quantitative point of view have been found. For example 1-propanol, propyl-acetate, ethyl-propanoate, propyl-propanoate, acetic acid and propionic acid were found at greater levels in Spanish-style olives. Cis-3hexen-1-ol and 2-butanone were detected above all in Castelvetrano-style olives, while isopentanol were mainly found in Greek-style samples. Authors also hypothesized a LOX pathway for three samples, since some aroma molecules which are formed in this process have been found, as shown by the presence of 1-hexanol, 1-hexanal and cis-3-hexen-1-ol. In particular, after 6 months’ brining, higher levels of
230
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
Proposed pathway for volatile esters biosynthesis in table olives Ethyl-propanoate PROPIONYL CoA
PYRUVATE DECARBOXYLASE
Pyruvate
ALCOHOL DEHYDROGENASE
Acetaldehyde
ACETYL CoA
Ethanol
Ethyl-acetate AAT
PYRUVATE DEHYDROGENASE
Acetyl CoA ACETYL CoA
Byproduct formed from certain amino acids
1-Propanol
Propyl-acetate AAT
PROPIONYL CoA
Propyl-propanoate FIGURE 24.8 Proposed pathway for ester biosynthesis in table olives. AAT ⫽ alcohol acetyl transferase, ADH ⫽ alcohol dehydrogenase, PDC ⫽ pyruvate decarboxylase, PDH ⫽ pyruvate dehydrogenase. Source: Sabatini, N. and Marsilio, M. (2008). Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chemistry, 107, 1522–1528.
Proposed source of origin of volatile compounds in table olives Lipids
Polysaccharides
Proteins, enzymes
Fatty acids metabolism
Carbohydrates metabolism
Amino acids metabolism
Aliphatic
Aliphatic
Methyl-branched
Saturated Alcohols 1-Hexanol Carbonyls 1-Hexanal
Saturated Alcohols Ethanol 1-Propanol 1-Butanol 1-Pentanol 1-Hexanol Esters Ethyl-acetate Ethyl-propanoate Propyl-acetate Carbonyls Acetic acid Propionic acid 1-Hexanal 1-Nonanal
Saturated Alcohols 2-Butanol 2-Pentanol 2-Methyl-1-propanol (isobutanol) 2-Methyl-2-butanol (isopentanol) Carbonyls Methyl-ethylcheton (2butanone) 3-Hydroxy-2-butanone
Unsaturated Carbonyls Cis-3-Hexen-1-ol
Unsaturated 4-Penten-1-0l
FIGURE 24.9 Proposed sources of origin of volatile compounds formation in Nocellara del Belice table olives.
1-hexanol in Spanish-style olives disclosed a probably prevalent linoleic acid oxidation pathway, while higher amounts of cis-3-hexen-1-ol showed a probably prevalent linolenic acid oxidation pathway in Castelvetranostyle olives. On the other hand, in Greek-style samples it seems to be a balance between linolenic and linoleic oxidation.
REFERENCES Angerosa, F., Basti, C., Vito, R., 1999. Virgin olive oil volatile compounds from lipoxygenase pathway and characterization of some Italian cultivars. J. Agric. Food Chem. 47, 836–839. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2000. Virgin olive oil odour notes, their relationship with volatile compounds from the lipoxygenase pathway and secoiridoid compounds. Food Chem. 68, 283–287.
CHAPTER | 24 A Comparison of the Volatile Compounds
Angerosa, F., 2002. Influence of volatile compounds on virgin olive oil quality evaluated by analytical approaches and sensor panels. Eur. J. Lipid Sci. Technol. 104, 639–660. Arrizon, J., Fiore, C., Acosta, G., Romano, P., Gschaedler, A., 2006. Fermentation behaviour and volatile compound by agave and grape must yeasts in high sugar Agave tequilana and grape must fermentations. Ant. Van Lee. 89, 181–189. Bianchi, G., 2003. Lipids and phenols in table olives. Eur. J. Lipid Sci. Technol. 105, 229–242. Delgado, A., Brito, D., Fevereiro, P., Peres, C., Figueiredo-Marques, J., 2001. Antimicrobial activity of L. plantarum, isolated from traditional lactic acid fermentation of table olives. Natl. Inst. Agr. Res. 81, 203–215. En.wikipedia.org/wiki. FAO corporate document repository: http://www.fao.org/docrep/x0560e/ x0560e08.htm Feussner, I., Wasternack, C., 2002. The lipoxygenase pathway. Ann. Rev. Plant. Biol. 53, 275–297. Fleming, H.P., Etchells, J.L., Bell, T.A., 1969. Vapor analysis of fermented Spanish-type green olives by gas chromatography. Food Chem. 34, 419–422. Gardner, H.W., 1991. Characterization of a C-5,13-cleaving enzyme of 13(S)-hydroperoxide of linolenic acid by soybean seed. Biochim. Biophys. Acta 1084, 221–239. Garrido-Fernandez, A., Fernandez-Diez, M.J., Adams, M.R., 1997. Table olives. Chapman and Hall; London, UK. Hammes, W.P., Tichaczek, P.S., 1994. The potential of lactic acid bacteria for the production of safe and wholesome food. Z. Lebensm. Unters. Forsch. A. 198, 193–201. Hatanaka, A., 1993. The biogeneration of green odour by green leaves. Phytochemistry 34, 1201–1218. Heath, H.B., 1981. Source book of flavors. Springer, p. 79. Herrero, M., Garcia, L.A., Diaz, M., 2006. Volatile compounds in cider, inoculation time and fermentation temperature effects. J. Inst. Brewing 112, 210–214. International Olive Oil Council (IOOC). 2005. 93a International Olive Oil Council Session, Madrid. International Olive Oil Council (IOOC). 2004. Trade standard applying to table olives. Jimenez, A., Guillen, R., Fernandez, B., Heredia, A., 1997. Factors affecting the Spanish green olives process, their influence on final texture and industrial losses. J. Agric. Food Chem. 45, 4065–4070. Kalua, C.M., Allen, M.S., Bedgood, D.R., Bishop, A.G., Prenzler, P.D., Robards, K., 2007. Olive oil volatile compounds, flavor development and quality: a critical review. Food Chem 100, 273–286. Ke, D., Zhou, L., Kader, A., 1994. Mode of oxygen and carbon dioxide action on strawberry ester biosynthesis. J. Am. Soc. Hort. Sci. 119, 971–975. Kiritsakis, A.K., 1998. Flavor components of olive oil-a review. J. Am. Oil Chem. Soc. 75, 673–681. Koprivnjak, O., Conte, L., Totis, N., 2002. Influence of olive fruit storage in bags on oil quality and composition of volatile compounds. Food Technol. Biotechnol. 40, 129–134. Marsilio, V., Lanza, B., Pozzi, N., 1996. Progress in table olive debittering, degradation in vitro of oleuropein and its derivatives by Lactobacillus plantarum. J. Am. Oil Chem. Soc. 73, 593–597.
231
Marsilio, V., Russi, F., Iannucci, E., Sabatini, N., 2008. Effects of alkali neutralization with CO2 on fermentation, chemical parameters and sensory characteristics in Spanish-style green olives (Olea europaea L.). LWT Food Sci. Technol. 41, 796–802. McFeeters, R.F., 2004. Fermentation microorganisms and flavor changes in fermented foods. J. Food Sci. 69, 35–37. Myers, M.J., Issenberg, P., Wick, E.L., 1970. L-leucine as a precursor of isoamyl alcohol and isoamyl acetate, volatile aroma constituents of banana fruit discs. Phytochemistry 9, 1693–1700. Olias, J.M., Pérez, A.G., 1993. Aroma of virgin olive oil, biogenesis of green odour notes. J. Agric. Food Chem. 41, 2368–2373. Panagou, E.Z., Tassou, C.C., 2006. Changes in volatile compounds and related biochemical profile during controlled fermentation of cv. Conservolea green olives. Food Microbiol 23, 738–746. Patterson, R.L.S., Charlwood, B.V., Macleod, G., Williams, A.A., 1992. Bioformation of flavors. The Royal Society of Chemistry, p. 103. Ridolfi, M., Terenziani, S., Patumi, M., Fontanazza, G., 2002. Characterization of the lipoxygenases in some olive cultivars and determination of their role in volatile compounds formation. J. Agric. Food Chem. 50, 835–839. Romano, P., Suzzi, G., 1996. Origin and production of acetoin during wine yeasts fermentation. Appl. Env. Microbiol. 62, 309–315. Ruiz, J.J., Alonso, A., Garcia-Martinez, S., Valero, M., Blasco, P., RuizBevia, F., 2005. Quantitative analysis of flavor volatiles detects differences among closely related traditional cultivars of tomato. J. Sci. Food Agric. 85, 54–60. Ruther, J., Hilker, M., 1998. A versatile method for on-line analysis of volatile compounds from living samples. J. Chem. Ecol. 24, 525–534. Sabatini, N., Marsilio, V., 2008. Volatile compounds in table olives. (Olea europaea L., Nocellara del Belice cultivar). Food Chem. 107, 1522–1528. Sabatini, N., Mucciarella, M.R., Marsilio, V., 2008. Volatile compounds in uninoculated and inoculated table olives with Lactobacillus Plantarum. (Olea europaea L., cv. Moresca and Kalamata). LWT Food Sci Technol. 41, 2017–2022. Salas, J.J., 2004. Characterization of alcohol acyltransferase from olive fruit. J. Agric. Food Chem. 52, 3155–3158. Salas, J.J., Sanchez, J., Ramli, U.S., Manaf, A.M., Williams, M., Harwood, J.L., 2000. Biochemistry of lipid metabolism in olive and other olive fruits. Prog. Lipid Res. 39, 151–180. Sanz, C., Olias, J.M., Perez, A.G., 1997. Aroma biochemistry of fruits and vegetables. In: Tomas-Barberan, F.A., Robins, R.J. (eds), Phytochemistry of Fruits and Vegetables. Clarendon Press, Oxford, pp. 125–155. Seimour, G.B., 1993. Biochemistry of fruit ripening. In: Taylor, J.E., Tucker, G.A. (eds), Chapman and Hall, London, U.K Solinas, M., Marsilio, V., Angerosa, F., 1987. Behavior of some components of virgin olive oils flavor in connection with the ripening of the olives. Riv. It. Sost. Gr. 64, 475–480. Tressl, R., Drawert, F., 1973. Biogenesis of banana volatiles. J. Agric. Food Chem. 21, 560–565. Williams, M., Harwood, J.L., 2000. Characterization of lipoxygenase isoforms in olive callus cultures. Biochem. Soc. Trans. 28, 830–831. Zervakis, G. Brief report on table olives cultivation and industry. http:// www.prochile.cl/tarapaca/promocion_aceitunas_zervakis.pdf.
This page intentionally left blank
Chapter 25
Polyphenol Oxidase and Oleuropein in Olives and their Changes During Olive Ripening Francisca Ortega-García1, Santos Blanco2, M. Ángeles Peinado2 and Juan Peragón1 1 2
Biochemistry and Molecular Biology Section, Department of Experimental Biology, University of Jaén, Spain Cell Biology Section, Department of Experimental Biology, University of Jaén, Spain
25.1 INTRODUCTION
TABLE 25.1 Key features of polyphenol oxidase and oleuropein.
25.1.1 Phenols, Types, Biological Significance and Presence in Olive Phenols are secondary metabolites of plants widely distributed throughout all plant organs and have important functions in the metabolism and physiology of plants. This complex group of substances has structures that vary from single phenolic molecules such as hydroxytyrosol [2-(3,4dihydroxyphenyl)ethanol] to highly polymerized compounds such as lignins. These compounds have diverse and important functions such as: the maintenance of plant integrity (lignins), floral pigmentation (flavonoids), antibiotics (phytoalexins), and plant defense against pathogens or symbionts (Dixon and Paiva, 1995; Yedidia et al., 2003; Ververidis et al., 2007). The occurrence of these substances in food is broadly variable and reaches high levels in the olive fruit and oil (Brenes et al., 1999). Currently, there is keen interest in dietary polyphenols due to their antioxidant capacity and consequent benefits to human health (Galli and Visioli, 1999; Dixon, 2004; Ververidis et al., 2007). Oleuropein, the main phenol of the olive, is a heterosidic ester of β-glucosylated elenolic acid and hydroxytyrosol (Table 25.1). Many of the nutritional and organoleptic properties of olive oil depend on the content of phenols in general and of oleuropein and hydroxytyrosol in particular. Oleuropein and hydroxytyrosol have a high antioxidant capacity with high free-radical scavenging activity (Visioli et al., 2002). The amount of phenols in olive oil is considered as an index of the quality of this product (Angerosa et al., 2004). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1. Polyphenol oxidase (PPO) is an enzyme that contains copper and catalyzes the hydroxylation of monophenols and the oxidation of o-diphenols to o-diquinones that accompany fruit ripening 2. Different isoforms of PPO have been found in different species of plants and seem to result from conformational changes, association–dissociation phenomena, covalent attachment of phenolic or carbohydrates 3. PPO has also been associated with the resistance of the plant against biotic and abiotic stress 4. Oleuropein is a heterosidic ester of β-glucosylated elenolic acid and hydroxytyrosol 5. Oleuropein is the main phenolic compound found in the olive and responsible for some of the nutritional properties of olive and olive oil 6. Oleuropein and hydroxytyrosol present a high antioxidant capacity with high free-radical scavenging 7. PPO and oleuropein have been linked to the chemical defense of the plant
25.1.2 Ripening, Polyphenol Oxidase, Structure and Biological Properties Fruit ripening brings major changes in the content and composition of phenols. In this process, phenols oxidize to quinones. These compounds are highly reactive and
233
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
234
SECTION | I
polymerize, bringing about the accumulation of browning pigments typical of mature fruits. The pigmentation of the fruit is often accompanied by a sensorial and nutritional change involving food-value loss. These oxidation reactions progress in the presence of oxygen and are catalyzed primarily by polyphenol oxidase (PPO, EC 1.10.3.1). PPO has been amply studied in many species of plants, fungi, and bacteria (Mayer, 2006; Marusek et al., 2006) and many of its properties have been extensively studied. Nevertheless, many aspects of its biological functions remain unknown. PPO has been located in chloroplast associated with the inner membrane of tylacoids, in the cytoplasm and in vesicles between the membrane and the cell wall (Obukowicz and Kennedy, 1981; Sommer et al., 1994; Murata et al., 1997). A great variety of kinetic and molecular properties and isoforms of PPO have been described in different species (Mayer and Harel, 1991). A comparative analysis of sequences of plant and fungal PPOs has shown several conserved structural features (Marusek et al., 2006). It has been suggested that the different isoforms of PPO result from conformational changes (Lerner et al., 1972), association– dissociation phenomena (Jolley et al., 1969), covalent attachment of phenolic material (Gregory and Bendall, 1966) or possible attachment of carbohydrate (Flurkey and Jen, 1980). The molecular mass differs widely among species, with values of 25 (Das et al., 1997) to 64 kDa (Marquès et al., 1995). With respect to its biological function, PPO has been related to the ripening of fruits and to resistance against pathogens, herbivores, and different biotic as well as abiotic stress (Mayer, 2006). In the olive, some studies have examined PPO but very few using the Picual variety. Ben-Shalom et al. (1977) reported the purification and properties of a catechol oxidase from green olives of cv Manzanillo. Shomer et al. (1979) showed that this enzyme is located at the inner face of the thylakoids and in the mitochondria. Several works have described the time course of fruit PPO activity in several varieties during ripening (Sciancalepore and Longone, 1984; Goupy et al., 1991; Ebrahimzadeh et al., 2003), and Goupy et al. (1991) related PPO activity to oleuropein concentration. To understand more about the biochemistry of phenols during olive-fruit maturation in this variety, we investigated the kinetics and molecular properties of PPO during fruit ripening in relation to total phenols and oleuropein concentration. Such information will help to identify the optimal time for harvesting the olive of Picual variety and thereby optimize product quality. As a result of this work, we provide the kinetic and molecular characterization of polyphenol oxidase in the fruits and leaves of olive trees of the Picual variety (Ortega-García et al., 2008). In this chapter we summarize the most representative results found in this study. The study was made in olive trees (Figure 25.1) located in an orchard at Torredonjimeno (Jaén, Spain, 37°45⬘61⬙ N, 3°57⬘12⬙W, 655 m a.s.l.). The trees were dryland farmed
Lipids, Phenolics and Other Organics and Volatiles
following traditional methods. Five trees selected in the orchard were sampled for leaves and fruits from all the positions on each tree. Five 25-cm-long branch samples with fruits near the apical end were collected and all leaves and fruits from each branch were pooled. The samples were frozen in liquid nitrogen and stored at ⫺20 °C until they were analyzed. From August to November, seven samples of fruits (from F1 to F7) and leaves (from L1 to L7) were picked from the same trees with an interval of 30 days for the first and second samples and 15 days otherwise. The fruitripeness index (IR) was calculated as proposed by Uceda and Frías (1975). Figure 25.2 presents the aspect of the fruits and leaves used. Fruit and leaf acetone-powder extracts were used to assay PPO activity, protein quantification, polyacrylamide gel electrophoresis (PAGE) and Western blotting. Also, an immunohistochemical procedure in early October samples was used to determine the tissue location of PPO. In the same samples, phenolic extracts were obtained with 80% methanol. The phenolic fraction
FIGURE 25.1 Photograph of an olive tree used in the experiment.
IR:
1
2
3
0
0.28
1.02
AUG
SEP
4
5
2.69
3.82
OCT
6
4.79
7
4.93 NOV
FIGURE 25.2 Changes in the aspect and fruit-ripeness index (IR) of Olea europaea L. cv Picual over the experiment.
235
CHAPTER | 25 Polyphenol Oxidase and Oleuropein in Olives and their Changes During Olive Ripening
of fruit and leaves was analyzed by high-performance liquid chromatography (HPLC) using a reverse phase column (Spherisorb ODS-2) and a UV-Vis and MS detector. Oleuropein was identified (Figure 25.3) and quantified in fruit and leaves during ripening. The total phenolic content of the methanol extracts was also determined by a colorimetric assay.
values in leaves, indicating that a different isoenzyme can be expressed in each organ (Ortega-García et al., 2008). The relative molecular mass of the polypeptide that showed PPO activity in gel was 50 and 55 kDa in leaf and fruit, respectively (Figure 25.5). Under denaturing conditions the molecular mass of this polypeptide determined by Western blotting was 27.7 kDa (Figure 25.5). This indicated that the enzyme can have a molecular mass of 50–55 kDa with two subunits each of 27.7 kDa. The immunohistochemical location of PPO showed that, in the fruit, it is present in the epidermis below the cuticle. In the leaf, PPO immunoreactivity was located in epidermal and lagunal parenchyma cells probably located in the plastids. Immunoreactivity was also shown in the vascular tissue in the zone around the phloem sieve cells and companion cells (Ortega-García et al., 2008). All these results indicate that PPO is a protein extensively distributed in both organs, probably it is expressed as two different isoenzymes in each tissue that also probably have different functions in the global metabolism of each organ.
25.2 KINETIC AND MOLECULAR PROPERTIES OF PPO IN THE FRUIT AND THE LEAF OF OLIVE TREES OF THE PICUAL VARIETY The study and characterization of PPO and oleuropein during fruit ripening is of great interest for a better characterization of fruit maturation in this Spanish cultivar that constitute 30% of the total olive-cultivation area in the world’s leading olive-oil-producing country (Barranco, 1998). In fruit and leaf, a hyperbolic kinetic behavior was found when catechol, 4-methylcatechol, catechin, DLDOPA, or chlorogenic acid was used as substrate (Figure 25.4). The highest specific activity was shown using catechol as the substrate. With practically all the substrates, the specific activity found in the fruit was significantly higher than in the leaf. The values of Michaelis constant (Km), maximum velocity (Vmax) and catalytic efficiency for the different substrates in fruits differed significantly from the
25.3 CHANGES DURING RIPENING During ripening, a significant and exponential increase of PPO activity was reported in the fruit (Figure 25.5). The Vmax found at the last stage of ripening studied (F7) was eight-fold higher than in the first stage (F1). The Km of PPO, when catechol was used as substrate, also changes during ripening. The values increased from 8 mM (found in the
C Intens. x 105
1.25
306.6 274.6
1.00
-MS2, 538.9 m/z
376.7
0.75 0.50 0.25
803.3 148.7190.7
0.00 100
200
468.8 300
400
500
600
700
800
900
B 538.9
Intens. x 105
3
A
2 1 216.6
0
0.20
Units of A280 nm
-MS1, 51.4 min
200
384.8 400
600
800
0.15
0.10
0.05
0 Oleuropein
0
10
20
30
40
50
60
70
80
90
Time of retention (minutes)
FIGURE 25.3 HPLC analysis of a methanol extract of fruit of Olea europaea L. cv Picual to identify oleuropein. (A) Chromatograph obtained recording the absorbance at 280 nm, (B) MS analysis of the compound eluted at 51.4 ⫾ 1.3 min. (C) MS-MS analysis of the fragment with m/z ratio ⫽ 538.9.
236
SECTION | I
PPO specific activity (units mg protein−1)
A
Lipids, Phenolics and Other Organics and Volatiles
B 160
30
120 20 80 10 40
0
0
0.05
0.1 0.15 [Substrate] (M)
0.20
0.25
0
0
0.05
0.1 0.15 0.20 [Substrate] (M)
0.25
0.30
FIGURE 25.4 Effect of the concentration of catechol (䊉), catechin (䊊), 4-methyl-catechol (䊐), chlorogenic acid (䊏), and dl-DOPA (䉭) on the PPO specific activity in fruit (A) and leaf (B) of Olea europaea L. cv Picual. Results are expressed as mean ⫾ S.E.M. of five data.
1500 40 1000
20 500
0
B
0
20 F1
40 F2
F3
60 F4
80 F5
0 120 days
100 F6
PPO specific activity (units mg protein−1)
Oleuropein concentration (mg g dry weight−1)
A
F7
55 kDa 36 kDa
C 27.7 kDa
FIGURE 25.5 Changes in oleuropein concentration, PPO activity and protein expression level in fruit of Olea europaea L. cv Picual during fruit ripening. (A) Evolution of oleuropein concentration (䊊, left y axis) and PPO specific activity (䊉, right y axis). (B) Partially denaturing SDS-PAGE analysis of PPO activity developed with DL-DOPA in fruits samples with different maturity index. (C) Immunoblot analysis of PPO in fruits with different maturity index. Reprinted from Ortega-García et al. (2008) with permission.
samples F1, F2 and F3) to 25 mM (found in F5, F6 and F7) (Ortega-García et al., 2008). Coinciding with these changes in the Vmax and Km values, a higher activity was found when assayed by partially denaturing SDS-PAGE and developing with DL-DOPA while a new band of 36 kDa was also reported in the samples F5, F6 and F7 (Figure 25.5).
This band can be used as a marker of the final phase of fruit maturation. The changes in the Km and the appearance of the new band in the last stages of ripening indicate that a new isoenzyme will also be present in these stages. Also, during ripening a significant rise in the protein-expression level of PPO was found by Western blotting (Figure 25.5). This result indicated that synthesis of this protein is stimulated in the last stages of fruit maturation. The combination of all these results indicated that two types of regulating mechanisms can coincide in this enzyme during fruit ripening, the induction of the synthesis of new molecules of protein and the differential expression of two isoenzymes. Both mechanisms would be part of a regulating mechanism of the gene expression of PPO in the fruit of olive trees of the Picual variety during ripening. The investigation of the nature of this mechanism will be the aim of future studies. In the leaf, PPO activity also exponentially increased during ripening although the changes were of a lower magnitude than in fruit. The Vmax value in H7 sample was twofold higher than in the H1 sample, whereas no changes in Km values nor the appearance of new bands have been reported (Ortega-García et al., 2008). The ample distribution of PPO in leaf and its changes during ripening agree with the existence of an important function of this enzyme during the normal leaf metabolism, probably related to the protection of the plant against biotic and abiotic stress (Shi et al., 2002).
25.4 OLEUROPEIN CONCENTRATION IN FRUIT AND LEAF OF OLIVE DURING RIPENING Oleuropein concentration was also determined along ripening in the fruit and the leaf of olive trees of Picual variety (Ortega-García et al., 2008). The results in the fruit are
CHAPTER | 25 Polyphenol Oxidase and Oleuropein in Olives and their Changes During Olive Ripening
shown in Figure 25.5. The oleuropein concentration significantly decreased in the fruit and increased in the leaf. The highest changes were found between the samples F1–F2 and L1–L2. By comparing the time courses of oleuropein concentration and PPO activity in both tissues, we found that the sharpest changes in oleuropein concentration during fruit ripening did not coincide with the increase in PPO activity. Thus, there was no direct relationship between the changes in oleuropein concentration and the PPO activity.
25.5 CONCLUSION Oleuropein and PPO are present and respond with different intensity and characteristics in the fruit and leaves during ripening. Probably this different response agrees with the existence of a different function of both compounds in both organs. In the leaf, PPO and oleuropein can be related to the response against biotic and abiotic stress whereas in the fruit, PPO and oleuropein can be directly linked to the mechanism of fruit ripening. Analyzing the changes in both parameters during ripening, we find that an early harvest provides fruit with a higher phenol content.
SUMMARY POINTS ●
●
●
●
●
●
Phenols and oleuropein are minor components of olive and olive oil, and have important antioxidant benefits for human health. Polyphenol oxidase catalyzed the oxidation of phenols to quinones that accompany the ripening of olives. Here, we describe the kinetic and molecular properties of PPO in the fruit and the leaf of olive of the Picual variety in relation to oleuropein concentration during ripening. Different PPO isoenzymes appear to be expressed in the fruit and the leaf during ripening. A high expression level of PPO was also found in the last stages of fruit maturation. The changes described in PPO expression are not responsible for the sharp fall in oleuropein concentration during the first stages of ripening.
REFERENCES Angerosa, F., Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G.F., 2004. Volatile compounds in virgin olive oil: occurrence and their relationship with the quality. J. Chromatogr. A 1054, 17–31. Barranco, D., 1998. Variedades y patrones. In: Barranco, D., FernándezEscobar, R., Rallo, L. (Eds.) El Cultivo del Olivo, second ed. Ediciones Mundi-Prensa, Junta de Andalucía, Sevilla, Spain, pp. 61–87. Ben-Shalom, N.B., Kahn, V., Harel, E., Mayer, A.M., 1977. Catechol oxidase from green olives properties and partial purification. Phytochemistry 16, 1153–1158.
237
Brenes, M., García, A., García, P., Ríos, J.J., Garrido, A., 1999. Phenolic compounds in spanish olive oils. J. Agric. Food Chem. 47, 3535–3540. Das, J.R., Bhat, S.G., Gowda, L.R., 1997. Purification and characterization of a polyphenol oxidase from the kew cultivar of Indian pineapple fruit. J. Agric. Food. Chem. 45, 2031–2035. Dixon, R.A., 2004. Phytoestrogens. Annu. Rev. Plant Physiol. Plant Mol. Biol. 55, 225–261. Dixon, R.A., Paiva, N.L., 1995. Stress-induced phenylpropanoid metabolism. Plant Cell 7, 1085–1097. Ebrahimzadeh, H., Motamed, N., Rastgar-Jazii, F., Montasser-Kouhsari, S., Shokraii, E.H., 2003. Oxidative enzyme activities and soluble protein content in leaves and fruits of olives during ripening. J. Food Biochem. 27, 181–196. Flurkey, W.H., Jen, J.J., 1980. Purification of peach polyphenoloxidase in the presence of added protease inhibitors. J. Food Biochem. 4, 29–41. Galli, C., Visioli, F., 1999. Antioxidant and other activities of phenolics in olives/olive oil, typical components of the Mediterranean diet. Lipids 34, S23–S26. Goupy, P., Fleuriet, A., Amiot, M.-J., Macheix, J.-J., 1991. Enzymatic browning, oleuropein content, and diphenol oxidase activity in olive cultivars (Olea europaea L.). J. Agric. Food Chem. 39, 92–95. Gregory, R.P.F., Bendall, D.S., 1966. The purification and some properties of the polyphenoloxidase from tea (Camellia sinensis L.). Biochem. J. 101, 569–581. Jolley, R.L., Robb, D.A., Mason, H.S., 1969. The multiple forms of mushroom tyrosinase. J. Biol. Chem. 244, 1593–1599. Lerner, H.R., Mayer, A.M., Harel, E., 1972. Evidence for conformational changes in grape catechol oxidase. Phytochemistry 11, 2415–2421. Marquès, L., Fleuriet, A., Macheix, J.J., 1995. Characterization of multiple forms of polyphenoloxidase from apple fruit. Plant Physiol. Biochem. 33, 193–200. Marusek, C.M., Trobaugh, N.M., Flurkey, W.H., Inlow, J.K., 2006. Comparative analysis of polyphenol oxidase from plant and fungal species. J. Inorg. Biochem. 100, 108–123. Mayer, A.M., 2006. Polyphenol oxidases in plants and fungi: Going places? A review. Phytochemistry 67, 2318–2331. Mayer, A.M., Harel, E., 1991. Phenoloxidases and their significance in fruit and vegetables. In: Fox, P.F. (ed.), Food enzymology, Vol. 1. Elsevier, London, United Kingdom, pp. 373–393. Murata, M., Tsurutani, M., Hagiwara, S., Homma, S., 1997. Subcellular location of polyphenol oxidase in apples. Biosci. Biotech. Biochem. 61, 1495–1499. Obukowicz, M., Kennedy, G.S., 1981. Phenolic ultracytochemistry of tobacco cells undergoing the hypersensitive reaction to Pseudomonas solanacearum. Physiol. Plant Pathol. 18, 339–344. Ortega-García, F., Blanco, S., Peinado, M.A., Peragón, J., 2008. Polyphenol oxidase and its relationship with oleuropein concentration in fruits and leaves of olive (Olea europaea) cv. “Picual” trees during fruit ripening. Tree Physiol. 28, 45–54. Shi, Ch., Dai, Y., Xu, X., Xie, Y., Liu, Q., 2002. The purification of polyphenol oxidase from tobacco. Protein Expres. Purif. 24, 51–55. Shomer, I., Ben-Shalom, N., Harel, E., Mayer, A.M., 1979. The intracellular location of catechol oxidase in the olive fruit. Ann. Bot. 44, 261–263. Sommer, A., Ne’eman, E., Steffens, J., Mayer, A., Harel, E., 1994. Import, targeting and processing of a plant polyphenol oxidase. Plant Physiol 105, 1301–1331. Sciancalepore, V., Longone, V., 1984. Polyphenol oxidase activity and browning in green olives. J. Agric. Food Chem. 32, 320–321.
238
SECTION | I
Uceda, M., Frías, L., 1975. Harvest dates. Evolution of the fruit of content, oil composition and oil quality. In: Internacional, Consejo Oleícola (ed.), Proceedings of II Seminario Oleícola Internacional. Córdoba, Spain, pp. 125–130. Ververidis, F., Trantas, E., Douglas, C., Vollmer, G., Kretzschmar, G., Panopoulos, N., 2007. Biotechnology of flavonoids and other phenylpropanoid-derived natural products. Part I: Chemical diversity, impacts on plant biology and human health. Biotechnol. J. 2, 1214–1234.
Lipids, Phenolics and Other Organics and Volatiles
Visioli, F., Poli, A., Galli, C., 2002. Antioxidant and other biological activities of phenols from olives and olive oil. Med. Res. Rev. 22, 65–75. Yedidia, I., Shoresh, M., Kerem, Z., Benhamou, N., Kapulnik, Y., Chet, I., 2003. Concomitant induction of systemic resistance to Pseudomonas syringae pv. lachrymans in cucumber by Trichoderma asperellum (T-203) and accumulation of phytoalexins. Appl. Environ. Microbiol. 69, 7343–7353.
Chapter 26
Major and Minor Lipid Constituents of Cornicabra Virgin Olive Oil and the Influence of Crop Season Changes M. Desamparados Salvador and Giuseppe Fregapane Departamento de Tecnología de Alimentos, Universidad de Castilla-La Mancha, Ciudad Real (España)
26.1 INTRODUCTION The Cornicabra olive cultivar (Table 26.1) covers an area of 300 000 Ha, mainly in the provinces of Ciudad Real and Toledo (Castilla-La Mancha, Spain), and accounts for more than 15% of the cultivated land under olive in Spain, the world’s largest olive-oil-producing country. Cornicabra virgin olive oil is valued for its high stability and good sensory characteristics, which have been described as a dense sensation and a balanced aroma, bitter and pungent (Salvador et al., 2001).
TABLE 26.1 Key facts of Cornicabra virgin olive oil. 1. The Cornicabra olive cultivar covers an area of 300 000 Ha, mainly in the provinces of Ciudad Real and Toledo (Castilla-La Mancha, Spain) 2. It accounts for more than 15% of the cultivated land under olive in Spain, the world’s largest olive oil-producing country 3. The Cornicabra fruit is medium to large with a characteristically elongated and asymmetric shape 4. The fat yield is 22–24% of fresh weight 5. Cornicabra virgin olive oil is valued for its high stability and good sensory characteristics 6. The ‘Montes de Toledo’ Protected Designation of Origin certifies the origin, authenticity and quality of the Cornicabra virgin olive oil produced in a specific geographic area
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
26.2 MONOVARIETAL VIRGIN OLIVE OILS COMPOSITION AND CULTIVAR CHARACTERIZATION The study of the chemical composition of virgin olive oils of a pure variety or from a specific production area is of great interest not only in terms of basic scientific knowledge but also (i) to the local industrial sector, which is composed of many small oil mills lacking adequate laboratory facilities and qualified staff for monitoring and improving the quality of their products; (ii) to the international olive oil business, since the exportation and commercialization of Spanish oil affects many countries; and (iii) to the final consumer, who demands more information on the characteristics and properties of high-quality traditional local products. To this end, a major effort has been made in recent years by the main olive-oil-producing countries to study the chemical composition of major and minor components and their relation with oil quality, and to establish analytical determinations that can effectively identify olive oil varieties or oil produced in a specific area. In Spain some of the most important olive varieties such as Picual, Hojiblanca and Arbequina from Andalucia and Cataluña have been exhaustively studied (Aparicio et al., 1990; Motilva et al., 1998; Cert et al., 1999). Moreover, chemometric and statistical procedures that employ series of chemical compounds and/or sensory descriptors are used to characterize or authenticate monovarietal virgin olive oils (Tsimidou and Karakostas, 1993; Aparicio and Luna, 2002; Bucci et al., 2002; Mannina et al., 2003). It has been proven that the ratio between saturated and unsaturated fatty acids can contribute to cultivar characterization, since it is known that the acidic profile
239
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
240
SECTION | I
of virgin olive oils is mainly affected by the fruit variety (Stefanoudaki et al., 1999; Cortesi et al., 2000). Nevertheless, other major factors such as climatic conditions, cultivar irrigation and the stage of ripeness of the fruit, can affect the TG and FA composition (Ranalli et al., 1997; Aparicio and Luna, 2002). Sterol and alcohol profiles are also used to characterize virgin olive oils and especially to detect the adulteration of olive oil with hazelnut oil (Mariani et al., 1999; Vichi et al., 2001). Recently, it has also been proposed that these profiles may be used to classify virgin olive oils according to their fruit variety (Aparicio et al., 1997; Bucci et al., 2002; Ranalli et al., 2002). In addition, it is known that virgin olive oil contains phenolic substances which affect its stability and contribute to oil flavor and aroma (Gutiérrez Gonzalez-Quijano et al., 1977; Vázquez Roncero, 1978). These compounds are different from those present in the olive fruit, since during ripening and processing several chemical and enzymatic reactions may take place yielding phenols of lower molecular weight (Angerosa and Di Giovacchino, 1996; Tsimidou, 1998).
26.3 QUALITY INDICES OF CORNICABRA VIRGIN OLIVE OIL Commercial Cornicabra virgin olive oil (n ⫽ 334) were collected from industrial oil mills located in the provinces of Toledo and Ciudad Real (Castilla-La Mancha, Spain) during a series of crop seasons from 1994/1995 to 2001/02. Most of these were obtained from oil mills belonging to the Montes de Toledo Protected Designation of Origin (EC 1187/2000). The rest were obtained from other oil mills located in Castilla-La Mancha, all exclusively processing the Cornicabra olive cultivar, in order to ensure chemical characterization of virgin olive oils of a single variety. Quality indices and organoleptic evaluation for this set of a large number of commercial Cornicabra virgin olive oils from several crop seasons are reported elsewhere (Salvador et al., 2001). Quality and genuineness criteria for various olive oil types are described in detail in the EU Regulations EC 2568/91 and later modification. For the majority of the Cornicabra oil analyzed the values of the analytical parameters fell within the ranges established for the highest-quality category ‘extra virgin’ olive oil.
26.4 TRIGLYCERIDE COMPOSITION The composition (%) of triglycerides (TGs), and of the TG fractions expressed as the equivalent carbon number (ECN) are shown in Table 26.2. The range and the 25, 50 and 75 percentiles are reported for better description of the distribution of values. As is known, the percentile is the percentage
Lipids, Phenolics and Other Organics and Volatiles
of the values with a lower score. Determination of triglyceride composition is performed according to Annex VIII of Regulation EC 2568/91. The main TG peaks in the Cornicabra virgin olive oil were OOO, SOL ⫹ POO, OLO ⫹ LnPP and OLA ⫹ SOO; these accounted for more than 85% of the total area of peaks in the chromatogram. The level of triolein (OOO), the main TG in all olive oil varieties, was remarkably high, with a mean concentration (⫾SD) of 51.68 ⫾ 1.84% and a range from 44.79 to 54.70%. Similar OOO content has been reported for the Picual variety (Graciani, 1988; Gouveia, 1997; Osorio et al., 2003); whereas it is greater than other Spanish varieties like Manzanilla, Verdial, Morisca and Lechin (Graciani, 1988; Osorio et al., 2003). The second peak in order of quantitative importance in the Cornicabra virgin olive oil corresponded to the SOL ⫹ POO TG mixture, with an average content of 20.77 ⫾ 1.33% and a range from 18.10 to 25.56%, as reported for other Spanish virgin olive oil varieties, like Picual (Graciani, 1988), Manzanilla Cacereña and Verdial (Osorio et al., 2003). The next two TG fractions are OLO ⫹ LnPP and OLA ⫹ SOO, with mean contents of 7.79 ⫾ 0.91% and 6.76 ⫾ 0.58%, and ranges from 5.26 to 10.72% and 5.50 to 8.61% respectively; similar to other Spanish monovarietal virgin olive oils like Picual and Verdial (Graciani, 1988), but which content is relatively high as compared to other varieties like Manzanilla, Lechin and Verdial (Graciani, 1988), Manzanilla Cacereña, Verdial, Morisca or Cornezuelo (Osorio et al., 2003). Finally, concentrations of trilinolein (LLL) and the ECN42 fraction (LLL, OLLn and PLLn) in the Cornicabra virgin olive oil were very low (0.06 ⫾ 0.02% and 0.18 ⫾ 0.04% respectively).
26.5 TOTAL FATTY ACID COMPOSITION Determination of total fatty acids is carried out as described in Annex XA of EC 2568/91 and XB of EC 796/2002. The fatty acid (FA) composition (%) of Cornicabra virgin olive oil (n ⫽ 224) in the crop seasons studied is depicted in Table 26.3. These analytical results are within EC Regulation limits for olive oils (myristic acid: ⱕ0.05%; linoleic: ⱕ0.9%; arachidic: ⱕ0.6%; gadoleic: ⱕ0.4%; behenic: ⱕ0.2%; and lignoceric: ⱕ0.2%). In Cornicabra olive oil, oleic acid content is especially high (80.41 ⫾ 0.96%) and linoleic acid content is particularly low (4.46 ⫾ 0.57%). Cornicabra and Picual from Andalusia are the two Spanish varieties with the highest oleic acid content of relevant nutritional interest, and with the lowest in linoleic and linolenic acids (Alba et al., 1996), which explains their high oxidative stability as determined by the Rancimat method (Salvador et al., 2001). Moreover,
241
CHAPTER | 26 Major and Minor Lipid Constituents of Cornicabra Virgin Olive Oil
TABLE 26.2 Triglyceride composition (%) of Cornicabra virgin olive oil from 1995/96 to 1999/00 crops (n ⫽ 194). Triglyceride
Mean ⫾ SD
Range
Percentiles 25
50
75
OLL ⫹ PoOL
0.75 ⫾ 0.16
0.34–1.35
0.65
0.73
0.84
OOLn ⫹ PLL ⫹ PoPoO
1.38 ⫾ 0.15
1.09–1.97
1.29
1.36
1.47
OLO ⫹ LnPP
7.79 ⫾ 0.91
5.26–10.72
7.21
7.84
8.35
PoOO
1.15 ⫾ 0.20
0.72–1.78
1.00
1.13
1.25
POL ⫹ SLL
2.69 ⫾ 0.32
2.10–4.19
2.49
2.62
2.81
OOO
51.68 ⫾ 1.84
44.79–54.70
50.72
52.24
52.92
SOL ⫹ POO
20.77 ⫾ 1.33
18.10–25.56
19.88
20.58
21.47
PSL ⫹ PPO
2.21 ⫾ 0.29
1.50–2.99
2.00
2.14
2.39
OOG
0.69 ⫾ 0.12
0.47–1.02
0.58
0.67
0.76
OLA ⫹ SOO
6.76 ⫾ 0.58
5.50–8.61
6.35
6.64
7.11
SOP ⫹ SLS
1.24 ⫾ 0.14
0.71–1.78
1.14
1.22
1.32
OOA
0.93 ⫾ 0.06
0.60–1.12
0.90
0.93
0.97
ECN42
0.18 ⫾ 0.04
0.07–0.32
0.16
0.18
0.20
ECN44
2.57 ⫾ 0.30
1.84–3.82
2.37
2.52
2.71
ECN46
12.46 ⫾ 1.08
9.13–16.66
11.78
12.37
13.09
ECN48
74.66 ⫾ 1.62
69.04–79.21
73.71
74.64
75.79
ECN50
8.68 ⫾ 0.70
7.10–11.09
8.19
8.52
9.04
ECN52
1.31 ⫾ 0.11
0.90–1.60
1.24
1.30
1.38
The main TG peaks in the Cornicabra virgin olive oil were OOO, SOL ⫹ POO, OLO ⫹ LnPP and OLA ⫹ SOO; these accounted for more than 85% of the total area of peaks in the chromatogram. The percentile values are reported for better description of the distribution of the triglyceride composition. SD, standard deviation. P, palmític; Po, palmitoleic; M, margaric; S, stearic; O, oleic; L, linoleic; Ln, linolenic; and A, araquidic acids. ECN42 ⫽ LLL ⫹ OLLn ⫹ PoLL. ECN44 ⫽ OLL ⫹ PoOL ⫹ OOLn ⫹ PLL ⫹ PoPoO ⫹ POLn ⫹ PPoL ⫹ PPoPo. ECN46 ⫽ OLO ⫹ LnPP ⫹ PoOO ⫹ POL ⫹ SLL ⫹ PoOP ⫹ SPoL ⫹ SOLn ⫹ PPL. ECN48 ⫽ OOO ⫹ SOL ⫹ POO ⫹ PSL ⫹ PPO. ECN50 ⫽ OOG ⫹ OLA ⫹ SOO ⫹ SOP ⫹ SLS. ECN52 ⫽ OOA ⫹ SOS ⫹ POA. Reprinted with permission from Food Chem. 86, 2004, 485–492. Copyright 2004 Elsevier Science.
the Cornicabra variety, like the Hojiblanca variety, also contains low levels of palmitic acid (9.22 ⫾ 0.17%). In addition to the cultivar, it should be considered that the other main known factors affecting total fatty acid composition and especially oleic acid content are latitude, climatic conditions, and the ripening stage of the fruit on harvesting (Ranalli et al., 1997; Aparicio and Luna, 2002). There were high statistically significant differences (p ⱕ 0.001) among main Spanish varieties (Picual, Hojiblanca, Arbequina and Cornicabra) in terms of triglycerides and total fatty acid composition (Aranda et al., 2004).
The highest statistically significant differences among the cultivars in terms of TGs were found in the peaks corresponding to OLL ⫹ PoOL (Anova-F ratio ⫽ 404), POL ⫹ SLL (F ⫽ 858), SPoL ⫹ SOLn (F ⫽ 749), PPL (F ⫽ 513) and OOO (F ⫽ 386) and the fraction ECN46 (F ⫽ 674). On the other hand, the most significant statistical significant differences (p ⱕ 0.001) in total fatty acids were those corresponding to C17:1 (F ⫽ 1014), C18:1 (F ⫽ 461), C18:2 (F ⫽ 482), MUFA (F ⫽ 457) and PUFA (F ⫽ 468). Therefore, several combinations of some TG and total FA variables, from 3 to 5, could be selected by
242
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 26.3 Total fatty acid composition (%) of Cornicabra virgin olive oil from 1995/96 to 1999/00 crops (n ⫽ 224). Fatty acid
Mean ⫾ SD
Range
Percentiles 25
50
75
Palmitic, C16:0
9.22 ⫾ 0.17
6.99–11.05
8.73
9.13
9.67
Palmitoleic, C16:1
0.77 ⫾ 0.11
0.49–1.11
0.68
0.76
0.85
Margaric, C17:0
0.06 ⫾ 0.01
0.04–0.07
0.05
0.06
0.06
Margaroleic, C17:1
0.10 ⫾ 0.01
0.08–0.11
0.09
0.10
0.10
Stearic, C18:0
3.36 ⫾ 0.29
2.61–4.43
3.15
3.29
3.56
Oleic, C18:1
80.41 ⫾ 0.96
76.52–82.49
79.92
80.59
80.97
Linoleic, C18:2
4.46 ⫾ 0.57
3.07–6.62
4.09
4.45
4.73
Linolenic, C18:3
0.62 ⫾ 0.08
0.48–0.95
0.56
0.60
0.66
Arachidic, C20:0
0.51 ⫾ 0.03
0.28– 0.62
0.49
0.50
0.51
SFA
13.29 ⫾ 0.67
11.40–15.13
12.85
13.24
13.71
MUFA
81.63 ⫾ 0.90
77.98–83.59
81.28
81.77
82.16
PUFA
5.08 ⫾ 0.58
3.67–7.22
4.66
5.03
5.37
In Cornicabra olive oil, oleic acid content is especially high and linoleic acid content is especially low. The percentile values are reported for better description of the distribution of the triglyceride composition. Reprinted with permission from Food Chem. 86, 2004, 485–492. Copyright 2004 Elsevier Science.
principal component analysis and discriminant analysis to satisfactorily classify the virgin olive oil varieties studied.
a median value of 1489 mg kg⫺1, and a range from 1125 to 1906 mg kg⫺1.
26.6 STEROL COMPOSITION
26.7 ALCOHOL COMPOSITION
The sterol composition (%; determined according to EC 2568/91, Annexes V and VI) of commercial Cornicabra virgin olive oil for the crop seasons from 1997/98 to 2001/02 (n ⫽ 334) is reported in Table 26.4. As expected, the main sterols found were β-sitosterol, Δ5-avenasterol and campesterol, with respective contents (mean ⫾ SD) of 84.4 ⫾ 2.4, 6.9 ⫾ 2.2 and 4.0 ⫾ 0.2%. The campesterol content was remarkably high, with a median value of 4.0, an interquartile range of 0.4, and a global range from 3.4 to 4.5% in the five crop seasons studied (from 1997/98 to 2001/02). More than half of the samples analyzed exceeded the upper limit of the 4% established by the EC Regulation (EC 2568/91 and later amendments). All of the Cornicabra olive oil samples analyzed contained more than 1000 mg kg⫺1 of total sterols, the minimum value established by EC Regulation for olive oil, with
The concentrations (mg/100 g) of triterpenic and higher aliphatic alcohols (EC 796/2002, Annex XIX) in Cornicabra virgin olive oil from the 2000/01 and 2001/02 crop seasons (n ⫽ 205) are shown in Table 26.5. The main aliphatic alcohol components found in commercial Cornicabra virgin olive oil were tetracosanol (C24), hexacosanol (C26) and dicosanol (C22), with respective median values of 3.8, 3.2 and 3.0 mg/100 g, and concentration ranges of 1.1–12.0, 1.3–7.0 and 0.8–7.3 mg/100 g respectively. As expected, the concentrations of odd carbon number moiety alcohols were much lower, with median values under 0.3 mg/100 g. The concentration of many sterol and alcohol compounds differed significantly (p ⱕ 0.01) among the main Spanish virgin olive oil varieties (Picual, Hojiblanca, Arbequina and Cornicabra; Rivera del Álamo et al., 2004). Campesterol was the sterol component with the highest
243
CHAPTER | 26 Major and Minor Lipid Constituents of Cornicabra Virgin Olive Oil
TABLE 26.4 Sterol composition (%) of commercial Cornicabra virgin olive oils from 1997/98 to 2001/2002 crop seasons (n ⫽ 334). Sterol
Mean ⫾ SD
Percentiles
Range 25
50
75
Cholesterol*
0.35 ⫾ 0.22
0.06–1.30
0.20
0.28
0.43
Brassicasterol*
0.08 ⫾ 0.12
0.00–0.56
0.00
0.03
0.08
24-Metilencolesterol
0.24 ⫾ 0.09
0.09–0.51
0.17
0.21
0.30
Campesterol*
4.01 ⫾ 0.21
3.42–4.50
3.83
4.03
4.17
Campestanol
0.34 ⫾ 0.06
0.03–0.64
0.31
0.33
0.36
Stigmasterol*
0.68 ⫾ 0.23
0.28–1.75
0.51
0.63
0.80
Δ7-Campesterol
0.14 ⫾ 0.12
0.00–0.87
0.07
0.11
0.18
Δ5.23-Stigmastadienol
0.03 ⫾ 0.07
0.00–0.30
0.00
0.00
0.01
Clerosterol
0.93 ⫾ 0.08
0.68–1.16
0.87
0.93
0.99
β-Sitosterol
84.41 ⫾ 2.41
74.76–87.90
82.71
85.31
86.19
Sitostanol
0.70 ⫾ 0.26
0.20–1.87
0.50
0.70
0.85
Δ5-Avenasterol
6.94 ⫾ 2.24
4.18–14.60
5.17
6.10
8.62
Δ5.24-Stigmastadienol
0.46 ⫾ 0.16
0.23–1.56
0.35
0.44
0.52
Δ7-Stigmastenol*
0.37 ⫾ 0.24
0.13–1.68
0.24
0.32
0.40
Δ7-Avenasterol
0.33 ⫾ 0.09
0.13–0.65
0.25
0.32
0.39
Apparent β-Sitosterol*
93.5 ⫾ 0.5
91.9–95.0
Total sterols* (mg kg⫺1)
1489 ⫾ 126
1125–1906
93.2 1413
93.5 1488
93.8 1567
As expected for virgin olive oil, the main sterols found were β-sitosterol, Δ5-avenasterol and campesterol. Reprinted with permission from Food Chem. 84, 2004, 533–537. Copyright 2004 Elsevier Science. * Limits established by the current European Legislation: cholesterol, ⱕ0.5; brassicasterol, ⱕ0.1; campesterol, ⱕ4.0; stigmasterol, ⬍campesterol; Δ7-Stigmastenol, ⱕ0.5; apparent β-Sitosterol, ⱖ93; Total sterols, ⱖ1000 mg kg⫺1.
TABLE 26.5 Alcohol composition (mg/100 g) of Cornicabra virgin olive oils from 2000/01 and 2001/2002 crop seasons (n ⫽ 205). Alcohol
Mean ⫾ SD
Percentiles
Range 25
50
75
Erythrodiol
3.56 ⫾ 1.18
0.37–7.79
3.08
3.77
4.18
Uvaol
0.64 ⫾ 0.21
0.14–1.25
0.54
0.67
0.76
Erythrodiol ⫹ Uvaol* (%)
2.68 ⫾ 0.69
0.32–4.09
2.48
2.78
3.09 (Continued)
244
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 26.5 (Continued ) Alcohol
Range
Mean ⫾ SD
Percentiles 25
50
75
Dicosanol (C22)
3.09 ⫾ 1.15
0.82–7.25
2.24
3.03
3.91
Tricosanol
0.24 ⫾ 0.09
0.08–0.56
0.18
0.23
0.29
Tetracosanol (C24)
4.11 ⫾ 1.84
1.13–11.99
2.88
3.75
4.93
Pentacosanol
0.30 ⫾ 0.10
0.11–0.59
0.22
0.29
0.36
Hexacosanol (C26)
3.53 ⫾ 1.27
1.32–7.03
2.71
3.24
4.39
Heptacosanol
0.23 ⫾ 0.09
0.08–0.47
0.17
0.21
0.29
Octacosanol (C28)
1.58 ⫾ 0.65
0.55–3.66
1.05
1.45
1.99
13.08 ⫾ 3.80
4.42–29.06
10.78
12.36
14.71
Total aliphatic alcohols
The main aliphatic alcohol components found in commercial Cornicabra virgin olive oil were tetracosanol (C24), hexacosanol (C26) and dicosanol (C22). Reprinted with permission from Food Chem. 84, 2004, 533–537. Copyright 2004 Elsevier Science. * Limit established by the current European Legislation: ⱕ4.5%.
TABLE 26.6 Natural antioxidants (mg kg⫺1) and oxidative stability (h) of commercial Cornicabra virgin olive oils from 1994/95 to 1999/00 crop seasons (n ⫽ 152 samples). Antioxidants and oxidative stability
Mean ⫾ SD
Range
Percentiles 25
50
75
α-Tocopherol
168 ⫾ 36
55–315
149
170
183
Total phenols (as caffeic acid)*
147 ⫾ 59
19–380
104
147
177
ortho-diphenols (as caffeic acid)*
7.8 ⫾ 5.5
0–27.2
4.0
6.7
4.25 ⫾ 3.76
0–18.04
1.57
3.04
5.46
3,4-DHPEA-EDA**
43.83 ⫾ 38.68
0–169.57
13.25
32.95
63.66
Oleuropein aglycon**
61.64 ⫾ 34.92
1.44–180.36
35.88
60.44
85.08
8.72 ⫾ 5.83
0.63–25.41
4.03
7.30
11.82
51.36
87.71
111.86
32.29
42.70
57.82
Hydroxytyrosol**
Tyrosol**
10.1
p-HPEA-EDA**
84.64 ⫾ 40.27
Ligstroside aglycon**
43.88 ⫾ 17.06
5.36–88.82
Total phenols** (HPLC)
308 ⫾ 139
36.6–680.0
200.0
314.0
408.3
Oxidative stability
61.0 ⫾ 24.6
8.8–143.4
43.8
60.8
77.7
0–202.09
The phenol content of the Cornicabra variety (and the Picual) is among the highest of all Spanish olive varieties. Concentration expressed as: * mg of caffeic acid per kg, by colorimetric method. ** Absolute concentration (mg kg⫺1) of phenols by HPLC, calculated according to the response factors determined by Mateos et al. (10). 3,4-DHPEA-EDA (dialdehyde form of elenolic acid linked to hydroxytyrosol); p-HPEA-EDA (dialdehyde form of elenolic acid linked to tyrosol). Adapted with permission from Food Chem. 74, 2001, 267–274 (Copyright 2001 Elsevier Science) and J. Agric. Food Chem. 50, 2002, 6812–6817 (Copyright 2002 American Chemical Society).
245
CHAPTER | 26 Major and Minor Lipid Constituents of Cornicabra Virgin Olive Oil
26.8 NATURAL ANTIOXIDANTS The α-tocopherol content in the Cornicabra variety oils studied ranged from 55 up to 315 mg kg⫺1, with a median value of 170 mg kg⫺1 (Table 26.6) and is evaluated following the AOCS Method Ce 8-89 (AOCS, 1989). Cornicabra olive oil apparently has slightly lower tocopherol content than other Spanish varieties (Hidalgo et al., 1993). The tocopherol content is highly variety-dependent, with concentrations ranging from 5 to 300 ppm. Usual values reported for good quality oils vary between 100 and 300 ppm (Angerosa and Di Giovacchino, 1996; Baldioli et al., 1996). Virgin olive oil contains phenolic substances which affect its stability and flavor. Colorimetric determination of total phenols and ortho-diphenol compounds is performed according to the method based on the Folin-Ciocalteau reagent (Gutfinger, 1981), whereas, individual phenolic compounds are quantified by HPLC at 280 nm using syringic acid as internal standard and the response factors determined by Mateos et al. (2001), and described by Gómez-Alonso et al. (2002). The median content of total polar phenol compounds in the samples analyzed was 147 mg kg⫺1 (as caffeic acid), although a wide range of concentrations was observed, from 19 up to 380 mg kg⫺1 (Table 26.6). Twenty per cent of the oils contained more than 200 ppm of phenols, and 10% contained less than 100 ppm. As found with respect to oxidative stability, the phenol content of the commercial Cornicabra variety (and the Picual) is among the highest of all Spanish olive varieties (Cert et al., 1996; Aparicio et al., 1999). Similar results were observed with o-diphenols: the median content in commercial olive oils was 29 mg kg⫺1 (as caffeic acid) respectively. The dialdehyde form of elenolic acid linked to tyrosol ( p-HPEA-EDA) was generally the main phenolic compound, with a median concentration of 87.7 ppm and an interquartile range (IQR, difference between the 75th and 25th percentiles) of 60.5 ppm (Table 26.6). Similarly, the dialdehyde form of elenolic acid linked to hydroxytyrosol (3,4-DHPEA-EDA) was also found in high concentration: 33.0 ⫾ 50.4 ppm (as median and IQR). Following in order of concentration were oleuropein and ligstroside aglycons, also derivatives of hydroxytyrosol and tyrosol, with contents of 60.4 ⫾ 49.2 ppm and 42.7 ⫾ 25.5 ppm respectively. The main simple phenols found in the Cornicabra virgin olive oil were tyrosol (7.3 ⫾ 7.8 ppm) and hydroxytyrosol (3.0 ⫾ 3.9 ppm), whereas the concentrations of other phenolic acids were very low.
26.9 CORRELATION BETWEEN OXIDATIVE STABILITY AND PHENOLIC COMPOSITION Oxidative stability of the commercial oils ranged from a minimum of only 9 h to a maximum of 143 h, with a median value of 61 h (Table 26.6). A quarter of the samples exhibited stability in excess of 78 h, while in another quarter stability was less than 44 h. According to these results and published data on other Spanish varieties (Cert et al., 1999), Cornicabra and Picual are the two Spanish olive varieties whose oils are most stable to oxidation. Oxidative stability is measured by the Rancimat method at 100°C with an air flow of 10 L h⫺1. A high correlation, consistent through the five seasons, was observed between total phenol content (from 0.873 to 0.964) and oxidative stability by Rancimat (Salvador et al., 2001), as depicted in Figure 26.1. The five resulting lines, one for each crop season, are almost parallel, with a slope ranging from 0.266 to 0.355. As shown in Table 26.7, the phenolic substances that showed the highest correlation with oxidative stability by 200
180
160
140 Oxidative stability (h)
Anova F-ratio (160) of the main Spanish olive oil varieties. Moreover, the average erythrodiol content was also the highest of the studied varieties, as also reported by Aparicio et al. (1997).
120
100
80
60
40
20
0 0
100
200
300
400
500
600
700
Total phenols (mg kg−1) FIGURE 26.1 Linear regression of oxidative stability versus total phenol content for Cornicabra virgin olive oil samples obtained from five crop seasons (n ⫽ 181). A high correlation, consistent through the five seasons, was observed between total phenol content (from 0.873 to 0.964) and oxidative stability by Rancimat. Crop season: 䊐, 94/95; 䊉, 95/96; 䉭, 96/97; 䉲, 97/98; 䉬, 98/99. Reprinted with permission from Food Chem. 74 (2001) 267–274. Copyright 200l Elsevier Science.
246
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 26.7 Regression coefficients of the correlation between oxidative stability and phenolic composition of Cornicabra virgin olive oil from 1995/96 to 1999/00 crop seasons (n ⫽ 130). r2
Compound 96/97
97/98
98/99
99/00
Total
Hydroxytyrosol [A]
0.209
0.340
0.285
0.197
0.061
3,4-DHPEA-EDA [B]
0.880
0.939
0.807
0.916
0.902
p-HPEA-EDA
0.883
0.828
0.515
0.833
0.786
Oleuropein aglycon [C]
0.889
0.829
0.740
0.669
0.697
Sum [B ⫹ C]
0.911
0.950
0.938
0.961
0.931
Sum [A ⫹ B ⫹ C]
0.954
0.896
0.947
0.966
0.878
HPLC total phenols
0.921
0.940
0.914
0.951
0.932
Total polyphenols*
0.774
0.928
0.862
0.949
0.894
o-diphenols
0.779
0.929
0.877
0.915
0.906
A higher correlation coefficient was generally obtained when the sum of 3,4-DHPEA-EDA and oleuropein aglycon or the sum of these two compounds and hydroxytyrosol were considered. Second grade polynomial regression. Reprinted with permission from J. Agric. Food Chem. 50, 2002, 6812–6817. Copyright 2002 American Chemical Society. * linear regression.
the Rancimat through the years studied were 3,4-DHPEAEDA (r2 ⫽ 0.81–0.94), p-HPEA-EDA (r2 ⫽ 0.52–0.88) and oleuropein aglycon (r2 ⫽ 0.74–0.89). In all cases a second-degree polynomial relationship was found. A higher correlation coefficient was generally obtained when the sum of 3,4-DHPEA-EDA and oleuropein aglycon (r2 ⫽ 0.91–0.96) or the sum of these two compounds and hydroxytyrosol (r2 ⫽ 0.90–0.97) were considered. As shown in Table 26.7, for each of the crop seasons studied, these values were often higher than the correlation coefficients obtained with total phenol content by HPLC (r2 ⫽ 0.91–0.95), and more so with other traditional analytical parameters like colorimetric determination of total polyphenols and o-diphenols (r2 ⫽ 0.77–0.95 and 0.78–0.92 respectively), because these methods are less specific and measure compounds that do not possess antioxidant activity.
SUMMARY POINTS ●
●
The chemical composition of Cornicabra virgin olive oils (334 samples) and the influence of crop season changes were examined. The main characteristics of Cornicabra olive oils are: high oleic acid (80.4 ⫾ 1.0%, as mean and standard deviation) and low linoleic acid content (4.5 ⫾ 0.6%); high campesterol level (4.0 ⫾ 0.2%), exceeding the
●
●
European Union Regulation upper limit of 4%; large total phenols content (up to 680 mg kg⫺1) and great Rancimat oxidative stability (up to 143 h). The main triglycerides found in the Cornicabra virgin olive oil variety are OOO, SOL ⫹ POO, OLO ⫹ LnPP and SOO ⫹ OLA (O ⫽ oleate; S ⫽ stearate; L ⫽ linoleate; Ln ⫽ linolenate; P ⫽ palmitate; A ⫽ araquidate), as expected from the total fatty acid profile observed. The main individual phenols found are the dialdehyde forms of elenolic acid linked to tyrosol (p-HPEA-EDA) and hydroxytyrosol (3,4-DHPEA-EDA), oleuropein and ligstroside aglycons.
REFERENCES Alba, J., Hidalgo, F., Ruíz, M.A., Martínez, F., Moyano, M.J., Cert, A., Pérez-Camino, M.C., Ruíz, M.V., 1996. Características de los aceites de oliva de primera y segunda centrifugación. Grasas y Aceites 47, 163–181. Angerosa, F., Di Giovacchino, L., 1996. Natural antioxidants of virgin oil obtained by two and tri-phases centrifugal decanters. Grasas y Aceites 47, 247–254. Aparicio, R., Luna, G., 2002. Characterization of monovarietal virgin olive oils. Eur. J. Lipid Sci. Technol. 104, 614–627. Aparicio, R., Ferreiro, L., Cert, A., Lazón, A., 1990. Caracterización de aceites de oliva vírgenes andaluces. Grasas y Aceites 41, 23–39.
CHAPTER | 26 Major and Minor Lipid Constituents of Cornicabra Virgin Olive Oil
Aparicio, R., Morales, M.T., Alonso, M.V., 1997. Authentification of European virgin olive oils by their chemical compounds, sensory attributes and consumers’ attitudes. J. Agric. Food Chem. 45, 1076–1083. Aparicio, R., Roda, L., Albi, M.A., Gutiérrez, F., 1999. Effect of various compounds on virgin olive oil stability measured by Rancimat. J. Agric. Food Chem. 47, 4150–4155. Aranda, F., Gómez-Alonso, S., Rivera del Alamo, R.M., Salvador, M.D., Fregapane, G., 2004. Triglyceride, total and 2-position fatty acid composition of Cornicabra virgin olive oil: Comparison with other Spanish cultivars. Food Chem. 86, 485–492. Baldioli, M., Servili, M., Perretti, G., Montedoro, G.F., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oils. J. Am. Oil Chem. Soc. 73, 1589–1593. Bucci, R., Magri, A.D., Magri, A.L., Marini, D., Marini, F., 2002. Chemical authentication of extra virgin olive varieties by supervised chemometric procedures. J. Agric. Food Chem. 50, 413–418. Cert, A., Alba, J., Camino, M.C., Ruiz, A., Hidalgo, F., Moreda, W., Moyano, M.J., Martinez, F., Tubaileh, R., Olias, J.M., 1999. Influencia de los sistemas de extracción sobre las características y los componentes menores del aceite de oliva virgen extra. Olivae 79, 41–50. Cert, A., Alba, J., León-Camacho, M., Moreda, W., Pérez-Camino, M.C., 1996. Effects of talc addition and operating mode on the quality and oxidative stability of virgin oils obtained by centrifugation. J. Agric. Food Chem. 44, 3930–3934. Cortesi, N., Fiorino, P., Ponzetti, A., 2000. Influencia de los cultivares y sistemas de extracción en la composición del aceite de oliva. Olivae 81, 36–39. EC 1187/2000. Register of protected designations of origin and protected geographical indications 2000. Off. J. Eur. Commun. L133, 19–20. EC 796/2002. Commission Regulation amending Regulation EEC 2568/91 on the characteristics of olive oil and olive-pomace oil and on the relevant methods of analysis. 2002 L128, 8–28. EC 2568/91. Characteristics of olive and olive pomance oils and their analytical methods 1991. Off. J. Eur. Commun. L248, 1–82. Hidalgo, F., Navas, M.A., Guinda, A., Ruiz, A., León, M., Lazón, A., Maestro, R., Janer, M.L., Pérez, M.C., Cert, A., Alba, J., Gutierrez, F., Dobarganes, M.C., Graciani, E., 1993. La calidad del aceite de oliva virgen: Posibles nuevos criterios para su evaluación. Grasas y Aceites 44, 10–17. Gómez-Alonso, S., Salvador, M.D., Fregapane, G., 2002. Phenolic compounds profile of Cornicabra virgen olive oil. J. Agric. Food Chem. 50, 6812–6817. Gouveia, J.M.B., 1997. Comparación de los Aceites de oliva producidos en el Norte del Alentejo. I. Principales Características Químicas y Sensoriales. Olivae 66, 34–45. Graciani, E., 1988. Caracterización del aceite de oliva virgen español. III. Posibilidad de caracterización por variedades de aceituna o por zonas de producción de acuerdo con su contenido en triacilgliceroles. Grasas y Aceites 39, 105–110. Gutiérrez Gonzalez-Quijano, R., Janer del Valle, C., Janer del Valle, M.L., Vázquez Roncero, A., 1977. Relación entre los polifenoles y la calidad y estabilidad del aceite de oliva virgen. Grasas y Aceites 28, 101–106.
247
Gutfinger, T., 1981. Polyphenols in olive virgin oils. J. Am. Oil Chem. Soc. 58, 966–968. Mannina, L., Dugo, G., Salvo, F., Cicero, L., Ansanelli, G., Calcagni, C., Segre, A., 2003. Study of the cultivar-composition relationship in Sicilian olive oils by GC, NMR, and statistical methods. J. Agric. Food Chem. 51, 120–127. Mariani, C., Bellan, G., Morchio, G., Pellegrino, A., 1999. Free and esterified minor components of olive and hazelnut oils: their potential utilisation in checking oil blend. Riv. Ital. Sostanze Grasse 76, 297–305. Mateos, R., Espartero, J.L., Trujillo, M., Rios, J.J., León-Camacho, M., Alcudia, F., Cert, A., 2001. Determination of phenols, flavones and lignans in virgin olive oils by solid-phase extraction and highperformance liquid chromatography with diode array ultraviolet detection. J. Agric. Food Chem. 49, 2185–2192. Motilva, M.J., Jaria, I., Bellart, I., Romero, M.P., 1998. Estudio de la calidad del aceite de oliva virgen de la Denominación de Origen “Les Garrigues” Lleida durante la campaña 1995/96. Grasas y Aceites 49, 425–433. Osorio, E., Sanchez, J.J., Martinez Cano, M., Montaño, A.M., 2003. Estudio del contenido en triglicéridos de aceites monovarietales elaborados a partir de aceitunas producidas en la región extremeña. Grasas y Aceites 54, 1–6. Ranalli, A., Pollastri, L., Contento, S., Di Loreto, G., Iannucci, E., Lucera, L., Russi, F., 2002. Sterol and alcohol components of seeds, pulp, and whole olive fruit oils. Their use to characterise olive fruit variety by multivariates. J. Sci. Food Agr. 82, 854–859. Ranalli, A., de Mattia, G., Ferrante, M.L., Giansante, L., 1997. Incidence of olive cultivation area on the analytical characteristics of the oil. Note 1. Riv. Ital. Sostanze Grasse 74, 501–508. Rivera del Álamo, R.M., Fregapane, G., Aranda, F., Gómez-Alonso, S., Salvador, M.D., 2004. Sterol and alcohol composition of Cornicabra virgin olive oil: the campesterol content exceeds the upper limit of 4% established by EU regulations. Food Chem. 84, 533–537. Salvador, M.D., Aranda, F., Gómez-Alonso, S., Fregapane, G., 2001. Cornicabra virgin olive oil a study of five crop seasons: composition, quality and oxidative stability. Food Chem. 74, 274–276. Stefanoudaki, E., Kotsifaki, F., Koutsaftakis, A., 1999. Classification of virgin olive oils of the two major cretan cultivars based on their fatty acid composition. J. Am. Oil Chem. Soc. 76, 623–626. Tsimidou, M., Karakostas, K.X., 1993. Geographical classification of Greek virgen olive oil by non-parametric multivariate evaluation of fatty acid composition. J. Sci. Food Agric. 62, 253–257. Tsimidou, M., 1998. Polyphenols and quality of virgin olive oil in retrospect. Ital J. Food Sci. 10, 99–116. Vázquez Roncero, A., 1978. Les polyphénol de l’huile d’olive et leur influence sur les caractéristiques de l’huile. Rev. Fr. Corps. Gras. 25, 6–21. Vichi, S., Pizzale, L., Toffano, E., Bortolomeazzi, R., Conte, L., 2001. Detection of hazelnut oil in virgin olive oil by assessment of free sterols and triacylglycerols. J. AOAC Int. 84, 1534–1541.
This page intentionally left blank
Chapter 27
Phytosterol Classes in Olive Oils and their Analysis by Common Chromatographic Methods Sodeif Azadmard-Damirchi1,2 and Paresh C. Dutta1 1 2
Department of Food Science, Swedish University of Agricultural Sciences, SLU, Uppsala, Sweden Department of Food Science and Technology, Faculty of Agriculture, University of Tabriz, Iran
27.1 INTRODUCTION According to the IUPAC recommendations from 1989, sterol molecules consist of four rings marked A, B, C and D with standard carbon numbering (Figure 27.1). Three of these rings (A, B and C) have six carbon atoms in a nonlinear structure and are fused to one 5-carbon atom ring (D). The various phytosterols found in plants differ in the number of carbon atoms in the side chain and the position and number of double bonds in the ring and side chain. Phytosterols comprise a major proportion of the unsaponifiables in all vegetable oils, including olive oil. Phytosterols are important from a nutritional point of view because they contribute to lowering serum cholesterol levels in humans and may enhance the oxidative and thermal stability and shelf-life of vegetable oils. They are also used to detect olive oil adulteration with other vegetable oils, particularly with hazelnut oil. Phytosterols are divided into three classes: 4-des-, 4-mono- and 4,4⬘-dimethylsterols,
and they can occur either in free form or esterified with fatty acids and other conjugates (Azadmard-Damirchi and Dutta, 2007). Analysis of phytosterols is very important in order to identify and characterize the vegetable oils. The major phytosterol class 4-desmethylsterols is rather easy to analyze, but the other classes are generally present in small amounts and require separation and enrichment from the unsaponifiables prior to further qualitative and quantitative analyses by gas chromatography (GC) and GC-mass spectrometry (MS). Preparative chromatography techniques such as thin-layer chromatography (TLC), high liquid chromatography (HPLC) and solid-phase extraction (SPE) are used for this purpose. This chapter mainly concentrates on the analysis of phytosterols in olive oils using traditional chromatographic methods, excluding HPLC-MS methods (Chapter 64). Readers are also referred to Chapters 23 and 63 for further in-depth discussion of other aspects relating to phytosterols in olive oils.
27.2 DEFINITION OF PHYTOSTEROLS 241 21
20
22
18
1
9
C
2 10
A HO
3 28
α
25
16
27
D 15
8 7
5 4
13 14
B
24
17
11 19
26 23
12
242
6 29
β
FIGURE 27.1 Basic structure of a sterol with standard carbon numbering according to the IUPAC system. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Phytosterols (plant sterols) are important components of plant cells in controlling membrane fluidity and permeability, although some have a specific function in signal transduction events and the activity of membrane-bound enzymes (Piironen et al., 2000). Phytosterols are biosynthetically derived from squalene and form a group of triterpenes, derivatives of a tetracyclic perhydro-cyclopentano-phenanthrene ring system with a flexible side chain at the C-17 atom and 3β-monohydroxy compounds (Hartmann, 1998). Most phytosterols contain 28 or 29 carbons and one or two carbon–carbon double bonds, typically one in the phytosterol nucleus and sometimes a second in the alkyl side chain (Moreau, 2005). 249
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
250
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
FIGURE 27.2 Chemical structure of a major 4-desmethylsterol, a 4-monomethylsterol and a 4,4⬘-dimethylsterol (triterpene alcohol).
27.3 BENEFITS OF PHYTOSTEROLS 27.3.1 Phytosterols in Cholesterol Lowering Phytosterols are structurally similar to cholesterol, but have an additional methyl or ethyl group in their side chain and are thus poorly absorbed in humans. Since the early 1950s, phytosterols have been shown to lower cholesterol absorption in animals. The average daily intake of phytosterols is in the range 150–400 mg, but to be effective as a cholesterol-lowering agent, phytosterol intake from enriched foods needs to be between 1.5–2.0 g per day (Normén et al., 2004). A short-term study has demonstrated that cholesterol levels increase considerably on a diet containing corn oil free from phytosterols, but when phytosterols are restored, cholesterol absorption decreases to normal (Ostlund et al., 2002). Further studies are needed on whether phytosterols in virgin olive oils contribute to cholesterollowering effects during long-term consumption.
27.3.2 Phytosterols as Anti-Carcinogens Cancer is one of the largest causes of mortality in Western societies. The types of cancer strongly related to dietary factors are colon, breast and prostate cancer. Results from some animal studies with phytosterol supplementation suggest that phytosterol can act as an anti-carcinogenic agent (Normén and Andersson, 2004). However, these authors caution that epidemiological studies and human intervention studies are needed to confirm the anti-carcinogenic effects of phytosterols. Other epidemiological data suggest that the phytosterol content of the diet may be associated with a reduction in common cancers, including cancers of the colon, breast and prostate. In addition, phytosterols may directly inhibit tumor growth, including the slowing of cell cycle progression, the induction of apoptosis and the inhibition of tumor metastasis (Bradford and Awad, 2007).
27.4 PHYTOSTEROL CLASSES Phytosterols can be classified into three classes based on the presence or absence of methyl groups at the C4 position
in the A ring (Figure 27.1), i.e. the presence or absence of C28 and C29. These three classes are: 4-desmethylsterols (without methyl group); 4-monomethylsterols (one methyl group); and 4,4⬘-dimethylsterols (triterpene alcohols, two methyl groups). The structural formulae of sitosterol (a 4desmethylsterol), citrostadienol (a 4-monomethylsterol), and 24-methylenecycloartanol (a 4,4⬘-dimethylsterol) are shown in Figure 27.2. The 4-desmethylsterols include all of the common phytosterols with a 28- or 29-carbon skeleton, but also cholesterol with a 27-carbon skeleton (Moreau et al., 2002). Cholesterol occurs as a major sterol in animal cells, although only as a few percent in plant cells. Chemically, it is an analog to the phytosterols, differing only in the side chain. Some common sterols from each class are given in Table 27.1.
27.5 ANALYSIS OF PHYTOSTEROL CLASSES IN OLIVE OIL For quantitative and qualitative analysis of phytosterol classes by GC and GC-MS, it is necessary to separate and enrich the phytosterol classes, especially the methylsterols (Azadmard-Damirchi et al., 2005). The main reason for the separate analyses of the three sterol fractions by GC is that some 4-desmethyl-, 4-monomethyl-, and 4,4⬘-dimethylsterols cannot be separated and can overlap during GC analysis (Azadmard-Damirchi et al., 2005). For example, sitosterol and β-amyrin [relative retention time (RRT) ⫽ 1.43]; Δ5-avenasterol, Δ7-(4)-monomethylsterol and taraxerol (RRT ⫽ 1.45); and Δ7- and Δ8-(4)-monomethylsterols and Δ7-(4,4⬘)-dimethylsterols (RRT ⫽ 1.58) have been found to overlap (Azadmard-Damirchi et al., 2005). Moreover, 4-monomethyl- and 4,4⬘-dimethylsterols in vegetable oils are generally present in much lower amounts than 4-desmethylsterols. For these reasons, it is necessary to enrich 4-monomethyl- and 4,4⬘-dimethylsterols from total sterols. Different methods are used to separate and enrich phytosterol classes and total sterols from unsaponifiables. The three main methods are TLC, HPLC and SPE, which are described in more detail below.
CHAPTER | 27 Phytosterol Classes in Olive Oils and their Analysis by Common Chromatographic Methods
251
TABLE 27.1 Classes of phytosterols generally reported in olive oil. Common name
IUPAC name
4-desmethylsterol Sitosterol
5α-Stigmast-5-ene 3β-ol
Campesterol
Ergost-5-ene 3β-ol
Stigmasterol
5α-Stigmasta-5,22-diene 3β-ol
Δ5-Avenasterol
5α-Stigmasta-5,24(28)-diene 3β-ol
4-monomethylsterol Citrostadienol
4α-Methyl-24-ethylidene-5α-cholest-7-ene 3β-ol
Obtusifoliol
4α,14α-Dimethyl-24-methylene-9β,19-cyclo-5α-cholest-8-ene 3β-ol
Gramisterol
4α-Methyl-24-methylene-5α-cholest-7-ene 3β-ol
Cycloeucalenol
4α,14α-Dimethyl-9β,19-cyclo-24-methylene-5α-cholestane 3β-ol
4,4⬘-dimethylsterol 24-Methylenecycloartanol
24-Methylen-9β,19-cyclo-5α-lanost-24-ene 3β-ol
Cycloartenol
9β,19-Cyclo-5α-lanost-24-ene 3β-ol
α-Amyrin
5α-Urs-12-ene 3β-ol
β-Amyrin
5α-Olean-12-ene 3β-ol
Methylsterols (4-monomethyl- and 4,4⬘-dimethylsterols) are synthesized at an early stage in the biosynthetic pathway and they are precursors of 4-desmethylsterols (adapted with permission from Hartmann, 1998).
27.6 THIN-LAYER CHROMATOGRAPHY TABLE 27.2 4-Desmethylsterol composition (% total sterols) of olive oil according to International Olive Oil Council trade standards (IOOC, 2003). Sterol
Limit
Cholesterol
⬍0.5
Brassicasterol
⬍0.1
Campesterol
⬍4.0
Stigmasterol
⬍campesterol
Δ7-Stigmastenol
⬍0.5
Apparent β-sitosterol
ⱖ93.0%a
a Apparent β-sitosterol comprises: β-sitosterol, Δ5-avenasterol, Δ5,23-stigmastadienol, clerosterol, sitostanol, Δ5,24-stigmastadienol.
TLC is the conventional method for separating and enriching phytosterol classes (Azadmard-Damirchi et al., 2005). The olive oil is saponified with KOH and the unsaponifiables are extracted with an appropriate solvent such as hexane, diethyl ether, etc. The unsaponifiable materials are then applied to the TLC plate (silica gel). To correctly identify the sterol bands, a reference sample of purified sterol fractions is applied alongside the sample band on the TLC plate. The plate is then developed twice in a suitable mobile phase, mostly hexane/diethyl ether/acetic acid (70:30:1, v/v/v). After development, the reference band is exposed to iodine vapor while the sample area is covered with a glass plate. Figure 27.3 shows the separation pattern of phytosterol classes by the TLC method. On the basis of the reference spots, three zones (4-desmethyl-, 4-monomethyl- and 4,4⬘-dimethylsterols) are identified and marked out. The zones on the TLC are then scraped off and after adding an internal standard, generally 5-α-cholestane, each fraction is extracted three times with
252
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
ether and ethyl acetate (9:1, v/v). The pressure before the column is 470.4 psi and the eluting speed is 0.05 L min⫺1 (Li et al., 2001). However, unlike TLC, use of HPLC may require a high solvent volume and also involve a higher cost. See Chapter 64 for details of HPLC-MS analysis of sterols in olive oils.
27.8 SOLID PHASE EXTRACTION
FIGURE 27.3 Separation of phytosterol classes by thin-layer chromatography showing quantitative differences in phytosterol classes of virgin olive oil (reproduced from Azadmard-Damirchi et al., 2005). Lane 1 ⫽ purified phytosterol classes of virgin olive oil; lane 2 ⫽ total unsaponifiables extracted from virgin olive; lane 3 ⫽ mixture of standard lipid samples containing 4-desmethylsterol (cholesterol) and free fatty acid. A ⫽ 4,4⬘-dimethylsterol class; B ⫽ 4-monomethylsterol class; C ⫽ 4-desmethylsterol class; D ⫽ free fatty acid; E ⫽ 4-desmethylsterol (cholesterol).
an appropriate solvent such as hexane, dichloromethane, etc. After separation and enrichment, phytosterol classes are generally derivatized to their trimethylsilyl (TMS) ether derivatives and analyzed by GC and GC-MS. However, the TLC method has some drawbacks. Different sterol fractions have close Rf values in TLC, which may cause mixing during scraping of TLC bands. The method is also time-consuming and laborious. In addition, preparative TLC is disadvantageous because it has a low recovery rate. It has been reported that using the TLC method, the actual recoveries are 61, 61 and 65% for campesterol, stigmasterol and sitosterol, respectively (Azadmard-Damirchi et al., 2005).
27.7 HIGH PERFORMANCE LIQUID CHROMATOGRAPHY Preparative HPLC has also been used to separate phytosterol classes of vegetable oils (Li et al., 2001). In this method, oil sample is saponified with alcoholic potassium hydroxide and unsaponifiables are extracted with diethyl ether. The ether solution is treated with anhydrous sodium sulfate to remove any residual water and 0.5 g of the unsaponifiables dissolved in ether is applied to the HPLC. For separation of phytosterol classes, Waters Prep LC 500A with normal phase silica gel column (Prep PAK-500/ Silica, 300 mm ⫻ 70 mm) and refractive index detector is used. The eluting mobile phase is a mixture of petroleum
SPE is a simple and inexpensive chromatographic method. It has been widely used in the preparation of lipid classes prior to further analyses by HPLC, GC and GC-MS. SPE has also been used to extract and purify total sterols from other unsaponifiables of vegetable oils. Moreover, SPE has been shown to separate total sterols in some vegetable oils more effectively and conveniently than traditional TLC methods (Ham et al., 2000). A rapid and reliable SPE method to separate and enrich sterol classes, particularly 4,4⬘-dimethylsterols in vegetable oils, has been developed (Azadmard-Damirchi and Dutta, 2006). Due to additional numbers of methyl groups, the polarity of the three sterol classes decreases in the order 4-desmethylsterols ⬎ 4-monomethylsterols ⬎ 4,4 ⬘ dimethylsterols. Because of the polarity of sterol fractions, 4,4⬘-dimethylsterols are a weakly retained isolate, followed by 4-monomethylsterols. The 4-desmethylsterols are retained more strongly than methylsterols on the silica sorbent and are eluted as the last sterol class in this method. Figure 27.4 shows the SPE work-up steps to separate 4,4⬘dimethylsterols from unsaponifiables of vegetable oils. Samples of hazelnut and virgin olive oil collected from different countries have been separated for their phytosterol classes using this new SPE method and the results have been compared with a previously published prep-TLC method. The amounts of 4,4⬘-dimethylsterols and 4-desmethylsterols separated with SPE in both hazelnut and virgin olive oils are at least 75% and 35% higher, respectively, than with TLC. Generally, both TLC and SPE produce similar results for 4-monomethylsterols of the two oils. The new SPE method to separate phytosterol classes is less time-consuming, simpler and can be used instead of preparative TLC to separate and enrich phytosterol classes in vegetable oils (Azadmard-Damirchi and Dutta, 2006).
27.9 GC AND GC-MS ANALYSIS OF PHYTOSTEROL CLASSES After separation and enrichment of phytosterol classes by one of the above-mentioned preparative methods, for GC and GC-MS analysis they are generally derivatized to TMS ether derivatives. For each class an appropriate amount (usually 100 μL) of silylating reagent is added, the mixture is well stirred and placed in an ultrasonic bath for 3 min,
253
CHAPTER | 27 Phytosterol Classes in Olive Oils and their Analysis by Common Chromatographic Methods
Unsaponifiables (dissolved in 5 mL hexane)
IS
8
Loading onto SPE cartridge (conditioned with 5 mL hexane)
Step 1. Washing apolar non-sterol compounds using 40 mL hexane: diethyl ether (99: 1) [Discarded]
10 3 1
Step 2. Eluting the pure 4,4’-dimethylsterols using 40 mL hexane: diethyl ether (99: 1) 10 mL hexane: diethyl ether (98: 2) [Collected for further analysis] Step 3. Washing with 10 mL hexane: diethyl ether (98: 2) [Discarded] Step 4. Eluting the pure 4-monomethylsterols using 20 mL hexane: diethyl ether (98: 2) [Collected for further analysis] Step 5. Washing with 5 mL hexane: diethyl ether (98: 2) [Discarded] Step 6. Eluting the pure 4-desmethylsterols using 10 mL hexane: diethyl ether (60: 40) [Collected for further analysis]
FIGURE 27.4 Flowchart of the solid phase extraction method to separate sterol classes in vegetable oils for further analysis by gas chromatography and gas chromatography-mass spectrometry (reproduced from Azadmard-Damirchi and Dutta, 2006).
incubated at 60 °C for 45 min, and thereafter resuspended again using the ultrasonic bath. The solvent is evaporated under a stream of nitrogen, and the TMS ether derivatives are usually dissolved in 200 μL hexane for further analysis by GC and GC-MS. Different methods can be used for GC and GC-MS analysis of phytosterol classes (AzadmardDamirchi et al., 2005). However, in different analyses we have used the following method and have obtained good results (Azadmard-Damirchi et al., 2005; AzadmardDamirchi and Dutta 2007). A fused-silica capillary column DB-5MS (30 m ⫻ 0.25 mm, 0.50 μm (J&W Scientific, Folsom, CA) is connected to a GC equipped with an FID. The analysis conditions are: (i) injector 260 °C, (ii) oven 60 °C for 1 min, rate 40°C min⫺1, final temperature 310 °C for 27 min, (iii) helium as a carrier gas and nitrogen as a makeup gas at a flow rate of 30 mL min⫺1, (iv) detector 310 °C. The peak areas are integrated, and quantification is done relative to 5α-cholastane as internal standard. It should be mentioned here that other suitable columns and GC and GC conditions can be used to perform the analysis of phytosterols. The column and conditions for the analysis by GC-MS are the same as used for GC analysis. The mass spectra are
2
4
5
6
7
9 11 12 13
FIGURE 27.5 Chromatogram of 4-desmethylsterols of virgin olive oil (reproduced from Casas et al., 2004). 1 ⫽ cholesterol, IS ⫽ α-cholestanol (internal standard, IS); 2 ⫽ methylene cholesterol; 3 ⫽ campesterol; 4 ⫽ campestanol; 5 ⫽ stigmasterol; 6 ⫽ Δ7-campesterol; 7 ⫽ clerosterol; 8 ⫽ sitosterol; 9 ⫽ sitostanol; 10 ⫽ Δ5-avenasterol, 11 ⫽ Δ5,24-stigmastadienol; 12 ⫽ Δ7-stigmastenol; 13 ⫽ Δ7-avenasterol.
recorded at an electron energy of 70 eV, and the ion source temperature is 200 °C. The spectra are scanned in the range m/z 50–600. The phytosterols are identified by comparing the mass spectra with pure samples of commercially available phytosterols, or otherwise their mass spectra are known from the literature and can be used for identification (Azadmard-Damirchi et al., 2005).
27.10 LEVEL OF PHYTOSTEROL CLASSES IN OLIVE OIL Considerable quantitative differences can occur in the relative proportions of phytosterol classes in olive oil (BenitezSánchez et al., 2003; Azadmard-Damirchi et al., 2005). Olive oils have the lowest amount of 4-monomethylsterols (9–11%) while 4-desmethyl and 4,4⬘-dimethylsterols constitute 51–57% and 32–40%, respectively, of total sterols. It is also visually apparent on the TLC plate that quantitative differences among phytosterol classes in olive oils are considerable (Figure 27.3).
27.10.1 4-Desmethylsterols From the 4-desmethylsterol class, cholesterol, campesterol, campestanol, stigmasterol, clerosterol, sitosterol, Δ5-avenasterol, Δ5,24-stigmastadienol, Δ7-stigmastenol and Δ7-avenasterol can be detected in olive oil (Paganuzzi and Leoni, 1979; Benitez-Sánchez et al., 2003; AzadmardDamirchi et al., 2005). Sitosterol is predominant, followed by Δ5-avenasterol and campesterol. Figure 27.5 shows the typical GC chromatogram of 4-desmethylsterols of olive oil. Phytosterol composition and content of olive oil are affected by cultivar, cropping year, degree of fruit ripeness, storage time of fruits before oil extraction and method of oil extraction. Table 27.2 shows the 4-desmethylsterol levels according to International Olive Oil Council trade standards (IOOC, 2003). Benitez-Sánchez et al. (2003) have
254
SECTION | I
reported the total 4-desmethylsterol content in European, North African and Turkish olive oils to be ⭓1000 ppm, 1800–2300 ppm and 1100–1700 ppm, respectively (Table 27.3). Sitosterol is the predominant 4-desmethylsterol, followed by Δ5-avenasterol and campesterol (Table 27.3). In virgin olive oil, there is a very good correlation between stability and concentration of total sterols, βsitosterol and Δ5-avenasterol (Gutiérrez et al., 1999). The 4-desmethylsterol level does not vary substantially during ripening of olive fruits, except for a reduction in total sterols and β-sitosterol and an increase in Δ5-avenasterol level. The decrease in total sterols is because sterols form in the first phases of ripening and as the oil content increases during this period, the sterols are therefore diluted. The decrease in β-sitosterol is exactly the same as the increase in Δ5-avenasterol, suggesting the presence of a desaturase enzyme that transforms β-sitosterol into Δ5-avenasterol (Gutiérrez et al., 1999). The influence of storage temperature of olive fruits on sterol composition is more important than the influence of storage time. The total sterol content increases gradually with olive storage time and this increase is greater for olive fruits stored at ambient temperature than those stored at low temperature (5 °C) (Gutiérrez et al., 2000). Stigmasterol is related to various quality parameters of virgin olive oil. High levels of this compound are correlated with high acidity and low organoleptic quality (Gutiérrez et al., 2000). Ranalli et al. (2002) compared the phytosterol classes of seed, pulp and whole olive fruit oil. Seed oil was found to have a higher content of total 4-desmethylsterols (2.3-fold higher), sitosterol, campesterol, clerosterol, Δ5-24-stigmastadienol, Δ7-stigmastenol and Δ7-avenasterol compared with the other extracted oils. Pulp and whole olive fruit oil generally had the same amounts of 4-desmethylsterols.
Lipids, Phenolics and Other Organics and Volatiles
2005). In these studies, the 4-monomethylsterol class includes obtusifoliol, Δ7-sterol, gramisterol, cycloeucalenol, cyclobranol, Δ7- and Δ8-sterol, Δ7,22-sterol and citrostadienol. In the 4-monomethylsterol class of olive oil the main compounds are citrostadienol (38–51%), obtusifoliol (11–13%) and cycloeucalenol (7–14%) (AzadmardDamirchi et al., 2005). Figure 27.6 shows the typical GC chromatogram of 4-monomethylsterols of olive oil. The 4-monomethylsterols, like other phytosterol classes in olive oils from different origins, can differ (Benitez-Sánchez et al., 2003). For example, gramisterol was not detected in Turkish olive oil but it was present at 1–48 ppm and 3– 26 ppm in North African and European olive oils, respectively (Benitez-Sánchez et al., 2003).
27.10.3 4,4⬘-Dimethylsterols α-, β-, and δ-amyrin, butyrospermol, taraxerol, cycloartenol, tirucalla-7,24-dienol, Δ7-sterol and 24-methylenecycloartanol have been detected in olive oil (Figure 27.7) (Paganuzzi and Leoni, 1979; Ntsourankoua et al., 1994;
4
1 2
27.10.2 4-Monomethylsterols
18
4-Monomethylsterols of olive oil from different countries have been studied in detail (Paganuzzi and Leoni, 1979; Benitez-Sánchez et al., 2003; Azadmard-Damirchi et al.,
7
IS
20
22
3
24
56
26
FIGURE 27.6 Chromatogram of 4-monomethylsterols of virgin olive oil. IS ⫽ internal standard (5α-cholestane); 1 ⫽ obtusifoliol; 2 ⫽ Δ7-sterol; 3 ⫽ gramisterol; 4 ⫽ cycloeucalenol; 5 ⫽ Δ7- and Δ8-sterol; 6 ⫽ Δ7,22-sterol; 7 ⫽ citrostadienol.
TABLE 27.3 4-Desmethylsterols content (parts per million) in olive oils from different geographical origins. Stigmasterol
Δ7-Stigmasterol
Δ7-Avanasterol
Total
34–266
Nr
Nr
ⱖ1000
1545–1851
158–214
1 –4
8–14
1800–2300
1000–2025
30–218
2
5–30
1100–1700
Olive oil
Campesterol
Sitosterol
European
25–114
5–67
683–2610
North African
59–62
16–26
Turkish
33–74
26–17
Nr, not reported. Data adapted from Benitez-Sánchez et al. (2003).
Δ5-Avenasterol
CHAPTER | 27 Phytosterol Classes in Olive Oils and their Analysis by Common Chromatographic Methods
Benitez-Sánchez et al., 2003; Azadmard-Damirchi et al., 2005; Azadmard-Damirchi and Dutta, 2007). Ntsourankoua et al. (1994) have determined the 4,4⬘-dimethylsterol content of olive oil. Identification of compounds has been carried out using GC-MS and authentic samples of α-amyrin, β-amyrin, lupeol, and also extracted cycloartenol and 24methylene-cycloartanol from sunflower oil. The compounds present include butyrospermol (4.1%), β-amyrin (2.1%), cycloartenol (9.7%), 7,24-tirucallalladienol (4.9%), 28-nor Δ17,18-oleanen-3β-ol (trace), 24-methylene-cycloartanol (74.3%) and some unknown compounds in virgin olive oil. Moreover, lupeol has not been detected in any sample of olive oil. Ranalli et al. (2002) have compared the phytosterol classes of seed, pulp and whole olive fruit oil. In the 4,4⬘dimethylsterols, β-amyrin, butyrospermol, cycloartenol and 24-methylene-cycloartanol were determined. Seed oil had a lower amount of total 4,4⬘-dimethylsterols and of cycloartenol and 24-methylene-cycloartanol, and a higher amount of β-amyrin and butyrospermol (not well separated) compared with other extracted oils. Pulp and whole olive fruit
7
5,6
3,4
IS
A
18
1
20
2
22
24
26
oil generally had similar levels of 4,4⬘-dimethylsterols. It was concluded that seed oil did not change the phytosterol classes of the whole fruit oil (mixture of seed and pulp oil). In general, there are many 4,4⬘-dimethylsterols in vegetable oils. Therefore, there is a possibility of overlap during GC and GC-MS analysis. Resolution of sterols by capillary column GC depends on many factors, e.g. length, internal diameter, film thickness, polarity of columns. In olive oil, tirucalla-7,24-dienol is co-eluted with cycloartenol using DB5-MS in GC analysis (Figure 27.7) (Azadmard-Damirchi and Dutta, 2007). It has been shown that by combining a non-polar DB5-MS column (10 m ⫻ 0.18 mm, 0.18 μm) and a mid-polar DB17-MS column (10 m ⫻ 0.18 mm, 0.18 μm) for GC-MS, these 4,4⬘-dimethylsterols are well separated compared with the single DB5-MS column (30 m ⫻ 0.25 mm, 0.50 μm) (Figure 27.7).
27.11 FREE AND ESTERIFIED PHYTOSTEROLS Phytosterols occur in free and esterified forms, i.e. as fatty acid esters, steryl glycosides or acylated steryl glycosides (Moreau et al., 2002). In free form, the hydroxyl group at C3 in the A ring is underivatized, whereas in the esterified form, the hydroxyl group is covalently bound to other constituents (Figure 27.8). The conventional method for total sterol analysis is saponification of the oil sample followed by extraction of the unsaponifiables with an organic solvent. On the other hand, separate determination of sterols in free and esterified forms provides detailed information on their distribution and stability (Phillips et al., 2002). The levels of free and esterified sterols in olive oil have been studied in detail (Grob et al., 1990). The concentration of free campesterol in pressed olive oil is below 40 ppm. In high-quality extra virgin olive oil, the concentration of free stigmasterol is below 10 ppm. Higher concentrations
7 5
O
6 IS
4 2 1
B
14
16
18
R
O
A
3 20
22
FIGURE 27.7 Gas chromatography-mass spectrometry total ion chromatograms of 4,4⬘-dimethylsterols from virgin olive oil with (A) nonpolar DB5-MS column, (B) combined non-polar DB5-MS and mid-polar DB17-MS (reproduced from Azadmard-Damirchi and Dutta, 2007). IS ⫽ internal standard (cholesterol); 1 ⫽ taraxerol; 2 ⫽ δ-amyrin; 3 ⫽ β-amyrin; 4 ⫽ butyrospermol; 5 ⫽ cycloartenol; 6 ⫽ tirucalla-7,24dienol; 7 ⫽ 24-methylene-cycloartanol.
255
OH B FIGURE 27.8 Chemical structure of 24-methylene-cycloartanol (a 4,4⬘-dimethylsterol): (A) esterified and (B) free form.
256
SECTION | I
are an indicator of low-quality olives (overripe or spoiled fruits). Raw lampante olive oil contains more free stigmasterol than extra virgin olive oil, which is also reflected by a lower campesterol/stigmasterol ratio. After refining, lampante olive oil contains free campesterol and stigmasterol at concentrations not very different from those in extra virgin olive oil. However, as both components are removed during refining at a similar ratio, the campesterol/stigmasterol ratio remains low (Grob et al., 1990). The concentration of sitosterol-C18-esters in highquality extra virgin olive oil is below 200 ppm, but up to 400 ppm must be considered acceptable. As refined solventextracted oil contains approximately 2500 ppm sitosterolC18-esters, the addition of 10% such oil increases the sitosterol ester concentration by about 250 ppm in extra virgin olive oil. The percentage of free sitosterol is a key parameter for assessing the quality of the olive oil. In highquality extra virgin olive oils, the percentage of free sitosterol exceeds 90%. The acceptable quality limit is around 80% and lower relative concentrations indicate the use of low-quality olives or forced extraction procedures. This parameter might be useful for setting a limit between extra virgin and lampante olive oil, particularly for those oils that appear to be extra virgin olive oil after gentle neutralization (Grob et al., 1990). Chryssafidis et al. (1992) have reported the amount of free and esterified 4-monomethylsterols and 4,4⬘dimethylsterols in virgin olive oil. Obtusifoliol, gramisterol, cycloeucalenol and citrostadienol were identified in both free and esterified forms of 4-monomethylsterols. Citrostadienol was the main sterol in this fraction, mostly occurring in esterified form. β-Amyrin, butyrospermol, cycloeucalenol and 24-methylene-cycloartanol were the sterols identified in the 4,4⬘-dimethylsterol class, in which 24-methylene-cycloartanol was the main sterol, occurring mostly in free form.
27.12 USE OF PHYTOSTEROLS IN DETECTION OF OLIVE OIL ADULTERATION WITH HAZELNUT OIL This section mainly discusses the issue of tracing hazelnut oil in olive oils. Readers are referred to Chapters 49 and 50 for further discussion on the issue of detecting adulteration of olive oils using various techniques. Hazelnut oil is used to adulterate olive oil due to its similar composition of triacylglycerols, fatty acids and major sterols (Cercaci et al., 2003; Christopoulou et al., 2004). Different methods have been proposed to detect this adulteration (Bøwadt and Aparicio, 2003). The sterol profile can be used as a means of differentiating between vegetable oils or detecting their authenticity (Itoh et al., 1973). Some esterified 4-desmethylsterols (campesterol, Δ7-stigmastenol and Δ7-avenasterol)
Lipids, Phenolics and Other Organics and Volatiles
have been used to detect olive oil adulteration with hazelnut oil using the Mariani ratio (RMAR) (Mariani et al., 1999): R MAR ⫽ (% campesterol ⫻ (% Δ7 -stigmastenol)2 ) % Δ7 -avenasterol (27.1) For non-adulterated olive oil, RMAR is not more than 1. This method can be used to detect adulteration at a level of 10% (Cercaci et al., 2003). 4,4⬘-Dimethylsterols are more variable in composition than 4-desmethylsterols, and therefore they are more effective for detecting vegetable oil adulteration. 4,4⬘Dimethylsterols have been separated from other unsaponifiables by TLC and analyzed by GC-MS (AzadmardDamirchi et al., 2005). The results showed that adulteration of virgin olive oil by hazelnut oil can be detected at a level of less than 4% by using lupeol and an unknown compound X (with a lupane skeleton) as possible potential markers. In a recent study, 4,4⬘-dimethylsterols of olive oil adulterated with refined hazelnut oil have been separated by SPE and the two above-mentioned marker compounds traced by GC-MS (Azadmard-Damirchi and Dutta, 2007). The results show that olive oil adulteration with refined hazelnut oil can be detected at levels as low as 2%. Phytosterols are interesting bioactive compounds and are now being used as functional food ingredients in order to improve quality of life in lowering cholesterol level and thus preventing coronary heart disease. A substantial body of research has been conducted on their potential anti-carcinogenic and other beneficial properties. Since the numbers of phytosterols are quite large, it is not always clear what the relative activities of the individual phytosterols are as regards various beneficial properties. In the future this area needs to be more closely studied and would also require better and more sensitive analytical techniques. The search for olive cultivars with higher levels of phytosterols is indeed an interesting area for the future.
SUMMARY POINTS ●
●
●
● ●
Phytosterols comprise a major proportion of the unsaponifiables in olive oil. Phytosterols are divided into three classes: 4-des-, 4-mono- and 4,4⬘-dimethylsterols. It is necessary to separate the methylsterols before gas chromatography and gas chromatography-mass spectrometry analysis. Phytosterols are important from a nutritional point of view. Olive oil adulteration with other vegetable oils can be detected by phytosterols analysis.
CHAPTER | 27 Phytosterol Classes in Olive Oils and their Analysis by Common Chromatographic Methods
REFERENCES Azadmard-Damirchi, S., Savage, G.P., Dutta, P.C., 2005. Sterol fractions in hazelnut and virgin olive oils and 4,4⬘-dimethylsterols as possible markers for detection of adulteration of virgin olive oil. J. Am. Oil Chem. Soc. 82, 717–725. Azadmard-Damirchi, S., Dutta, P.C., 2006. Novel solid-phase extraction method to separate 4-desmethyl-, 4-monomethyl-, and 4,4⬘-dimethylsterols in vegetable oils. J. Chrom. A. 1108, 183–187. Azadmard-Damirchi, S., Dutta, P.C., 2006. Rapid separation of methylsterols from vegetable oils by solid-phase extraction. Lipid Technol. 18, 231–234. Azadmard-Damirchi, S., Dutta, P.C., 2007. Free and esterified 4,4⬘-dimethylsterols in hazelnut oil and their retention during refining processes. J. Am. Oil Chem. Soc. 84, 297–304. Benitez-Sánchez, P.L., Camacho, L.M., Aparicio, R., 2003. A comprehensive study of hazelnut oil composition with comparisons to other vegetable oils, particularly olive oil. Eur. Food Res. Technol. 218, 13–19. Bradford, B.G., Awad, A.B., 2007. Phytosterols as anticancer compounds. Mol. Nutr. Food Res. 51, 161–170. Bøwadt, S., Aparicio, R., 2003. The detection of the adulteration of olive oil with hazelnut oil: A challenge for the chemist. Inform 14, 342–344. Casas, J.S., Bueno, E.O., García, A.M.M., Cano, M.M., 2004. Sterol and erythrodiol ⫹ uvaol content of virgin olive oils from cultivars of Extremadura (Spain). Food Chem. 87, 225–230. Cercaci, L., Rodriguez-Estrada, M.T., Lercker, G., 2003. Solid-phase extraction–thin-layer chromatography-gas chromatography method for the detection of hazelnut oil in olive oils by determination of esterified sterols. J. Chrom. A. 985, 211–220. Christopoulou, E., Lazaraki, M., Komaitis, M., Kaselimis, K., 2004. Effectiveness of determinations of fatty acids and triglycerides for the detection of adulteration of olive oils with vegetable oils. Food Chem. 84, 463–474. Chryssafidis, D., Maggos, P., Kiosseoglou, V., Boskou, D., 1992. Composition of total and esterified 4-monomethylsterols and triterpene alcohols in virgin olive oil. J. Sci. Food Agri. 58, 581–583. Grob, K., Lanfranchi, M., Mariani, C., 1990. Evaluation of olive oils through the fatty alcohols, the sterols and their esters by coupled LC-GC. J. Am. Oil Chem. Soc. 67, 626–634. Gutiérrez, F., Jímenez, B., Ruíz, A., Albi, M.A., 1999. Effect of olive ripeness on the oxidative stability of virgin olive oil extracted from the varieties Picual and Hojiblanca and on the different components involved. J. Agric. Food Chem. 47, 121–127. Gutiérrez, F., Varona, I., Albi, M.A., 2000. Relation of acidity and sensory quality with sterol content of olive oil from stored fruit. J. Agric. Food Chem. 48, 1106–1110.
257
Hartmann, M.A., 1998. Plant sterols and the membrane environment. Trend Plant Sci. 3, 170–175. International Olive Oil Council. 2003. COI/T.15/NC no. 3/Rev. 1, Trade standard applying to olive oils and olive pomace oils. Madrid IOOC. 2003. International Olive Oil Council activities: World Olive Oil Consumption. Available from: ⬍www.internationaloliveoil.org/⬎. Itoh, T., Tamura, T., Matsumoto, T., 1973. Sterol composition of 19 vegetable oils. J. Am. Oil Chem. Soc. 50, 122–125. Li, J., Ho, C., Li, H., Tao, H., Liu, L., 2001. Separation of sterols and triterpene alcohols from unsaponifiable fractions of three plant seed oils. J. Food. Lipids. 7, 11–20. Mariani, C., Bellan, G., Morchio, G., Pellegrino, A., 1999. I componenti minori liberi ed esterificati dellâolio di oliva e dellâolio di nocciola: loro possibile utilizzo nellâindividuazione di commistioni. Rivista Italiana delle Sostanze Grasse 76, 297–305. Moreau, R.A., Whitakerb, B.D., Hicksa, K.B., 2002. Phytosterols, phytostanols, and their conjugates in foods: Structural diversity, quantitative analysis, and health-promoting uses. Prog. Lipid Res. 41, 457–500. Normén, L., Andersson, S.W., 2004. Does phytosterol intake affect the development of cancer? In: Dutta, P.C. (ed.), Phytosterols as Functional Food Components and Nutraceuticals. Marcel Dekker, Inc., New York, pp. 191–242. Normén, L., Frohlich, J., Trautwein, E., 2004. Role of plant sterols in cholesterol lowering. In: Dutta, P.C. (ed.), Phytosterols as Functional Food Components and Nutraceuticals. Marcel Dekker, Inc., New York, pp. 243–315. Ntsourankoua, T., Artaud, J., Guerere, M., 1994. Triterpene alcohols in virgin olive oil and refined olive pomace oil. Annales des Falsifications de l’Expertise Chimique et Toxicologique 87, 91–107. Ostlund, J., Racette, S.R., Okeke, A., Stenson, W.F., 2002. Phytosterols that are naturally present in commercial corn oil significantly reduce cholesterol absorption in humans. Am. J. Clin. Nutr. 75, 1000–1004. Paganuzzi, V., Leoni, E., 1979. On the composition of Iranian olive oil. J. Am. Oil Chem. Soc. 56, 925–930. Phillips, K.M., Ruggio, D.M., Toivo, J.I., Swank, M.A., Simpkins, A.H., 2002. Free and esterified sterol composition of edible oils and fats. J. Food Comp. Anal. 15, 123–142. Piironen, V., Lindsay, D.G., Miettinen, T.A., Toivo, J., Lampi, A.M., 2000. Plant sterols: Biosynthesis, biological function and their importance to human nutrition. J. Sci. Food Agric. 80, 939–966. Ranalli, A., Pollastri, L., Contento, S., Di Loreto, G., Lannucci, E., Lucera, L., Russi, F., 2002. Sterol and alcohol components of seed, pulp and whole olive fruit oils. Their use to characterise olive fruit variety by multivariates. J. Sci. Food Agric. 82, 854–859.
This page intentionally left blank
Chapter 28
Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches M.D. Luque de Castro and F. Priego Capote Marie Curie Annex Building, Department of Analytical Chemistry, Campus of Rabanales, University of Córdoba, Spain
28.1 INTRODUCTION Interest in olive phenols (OPs) has grown substantially in recent decades and promoted multidisciplinary research into their composition, histological distribution and histochemical enzymatic localization. Some studies have exposed the biomolecular functions of both simple and conjugated phenol compounds such as oleuropein. The excellent properties of OPs have promoted an active search for effective raw materials for their isolation. The sources of oleuropein, its derivatives and various other OPs include the olive fruit and the waste from olive oil production, in addition to olive leaves and branches; the two primary sources are olive leaves and the high-water-content waste produced by the olive oil industry. Leaves are the organ containing the highest levels of these compounds in olive trees. For example, the OP model compound oleuropein is present in proportions of 0.005–0.12% in olive oil, up to 0.87% in alperujo and 1–14% in olive leaves (Luque de Castro and Japón-Luján, 2006). The high content in this substance of olive leaves is behind their former widespread use in infusions as a folk remedy against diseases such as malaria. In the 20th century, research into olive leaf extracts revealed that their healthy properties are a consequence of the functions served by OPs in olive trees (namely, protection from pathogen attacks and response to insect injury). Olive leaves and small branches are agricultural residues resulting from the beating of olive trees for fruit removal, and also from pruning. Spain produces a vast amount of olive leaves and small branches in mills prior to oil production (ca. 6 Mt of biomass each year). The interest of both raw materials as OP sources may also promote their obtainment as industrial byproducts from olive mills – in fact, they account for about 10% of the total weight of fruit processed. At present, however, most of these residues are simply disposed of by burning (Rada et al., 2007). The scientifically proven health properties of OPs, and their concentration levels in extracts from olive leaves and Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
small branches, which have usually been studied jointly, are promoting the development of new, faster, more efficient automatic extraction methods for their industrial exploitation.
28.2 EXTRACTION METHODS The particular technique used to isolate OPs depends on the characteristics of the sample. The extraction of OPs from solid samples has aroused some industrial interest and been the subject of several patents concerned with the separation of phenols (particularly oleuropein) from olive leaves based on traditional solid–liquid extraction methodology (Pinnell and Omar, 2004). In the laboratory, OPs are usually isolated by using traditional methods involving maceration and Soxhlet extraction with various extractants such as methanol–water mixtures (Montedoro et al., 1993; Agalias et al., 2003) or hexane (Guinda et al., 2002). It is important to avoid toxic extractants when the procedures are to be used on an industrial scale and the products targeted at human use. Also, it is desirable to shorten extraction times, which are most often in the region of 24–48 h. The extraction process can be expedited by using alternative procedures such as the following.
28.2.1 Extraction with Superheated Liquids Superheated liquids have proved effective for isolating natural products. Figure 28.1A depicts the extraction unit used to develop a specific method for extracting OPs from olive leaves and small branches (Japón-Luján and Luque de Castro, 2006, 2007). Extraction is done in two consecutive static and dynamic steps, using a 70:30 (v/v) ethanol–water mixture as extractant. Once assembled, the sample cell is placed in the oven, which is then pressurized to 6 bar by opening the inlet valve, closing the restrictor valve, propelling the extractant and heating it at the working
259
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
260
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
UP
EC
HPP
C
WB
SV
ER V1
V2
ER
PP
N2 CO
LC
EC O
C
PC
HPLC
Extract
A
B
Refrigerant Digestor
Controller Extractant
Microwaves C
Sample
FIGURE 28.1 Devices used for the extraction of olive phenols from olive leaves and branches. (A) Superheated liquid-based extractor. ER, extractant reservoir; HPP, high-pressure pump; CO, coil; EC, extraction cell; O, oven; C, cooler; V1, inlet valve; V2, restrictor valve. [Reproduced with permission of Elsevier, Luque-Rodríguez, J.M., Luque de Castro, M.D., Pérez-Juan, P. (2005) Extraction of fatty acids from grape seed by superheated hexane, 68, 126–130]. (B) Ultrasound-assisted extractor. C, extraction coil; EC, extraction chamber; ER, extractant reservoir; HPLC, high performance liquid chromatograph; LC, leaching carrier; PC, personal computer; PP, peristaltic pump; SV, switching valve; UP, ultrasonic probe; WB, water bath. (Reproduced with permission of Elsevier, Japón-Luján et al., 2006a). (C) Microwave-assisted extractor.
temperature (140°C). These conditions are kept for 6 min, after which dynamic extraction is started by opening the inlet valve and controlling the outlet restrictor in order to maintain the pressure, the extractant being pumped at 1 mL min⫺1 for 7 min. The whole extraction process takes 13 min.
28.2.2 Ultrasound-Assisted Extraction Using ultrasound is one other way of accelerating OP extraction. As known by users of ultrasound-based devices, cavitation (namely, the formation of tiny bubbles subject to fast adiabatic compressions and expansions which result in high temperatures and pressures on their inside while having a minimal effect on the temperature of the overall system) (Luque de Castro and Priego Capote, 2006) facilitates
penetration and mass transfer to extractants, thereby resulting in increased solubility and diffusivity (Luque de Castro and Priego Capote, 2006). Ultrasonic energy can be applied batchwise or, preferentially, in a continuous manner, using a dynamic system such as that reported by Japón-Luján et al. (2006a) (Figure 28.1B) for the separate extraction of OPs from leaves and small branches. In the process, 1 g of either milled leaves or branches is placed in the extraction chamber (EC), which is then assembled and filled with leaching carrier (LC, a 60:40 ethanol–water mixture) propelled by a peristaltic pump (PP). After filling, the extraction chamber is immersed in a water bath (WB) that is thermostated at 40°C throughout the extraction time. Next, the leaching carrier is circulated through the solid sample under ultrasonic irradiation (ultrasound duty cycle 0.7 s, output amplitude 30% of the converter, applied power 450 W with the ultrasonic probe
261
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
placed 4 cm from the top surface of the extraction cell) for a preset time of 25 min. During extraction, the direction of the leaching carrier, which is circulated at 5 mL min⫺1, is reversed at 90 s intervals in order to minimize dilution of the extract and compaction of the sample in the extraction cell, which might cause overpressure in the system.
28.2.3 Microwave-Assisted Extraction The use of microwaves additionally expedites OP extraction. The extraction method of Japón-Luján et al. (2006b) for olive leaves and small branches uses the device of Figure 28.1C to irradiate the sample–extractant suspension with focused microwaves in order to raise the leaching efficiency. The microwave-assisted extraction process completes within 8 (olive leaves) or 10 min (small branches). Table 28.1 compares the previous extraction procedures in terms of extraction time, ethanol–water ratio in the extractant and extractant volume. In contrast to superheated liquid extraction and US-assisted extraction, where the extract and solid sample residue are automatically separated, microwave-enhanced extraction requires manual removal of the residue – usually by centrifugation. Existing methods for the isolation of OPs have shown that the use of microwaves of appropriate energy, high temperatures and pressures or ultrasound of suitable power and frequency cause no degradation of phenols (Luque de Castro and Japón-Luján, 2006). Therefore, all three can and should be used as they dramatically shorten leaching times and ensure a very high extraction efficiency, both of which make the process amenable to industrial implementation. Alternative methods involving modern technologies for isolating OPs from olive leaves have also been reported.
One is based on counter-current supercritical fluid extraction with CO2 as supercritical fluid and ethanol as modifier – or a high pressure instead of the latter – to fractionate the raw extract obtained by macerating olive leaves with hexane. Although the method was optimized by using a multivariate approach, separation was poor – waxes and α-tocopherol remained in the raffinate fraction, whereas hydrocarbons were recovered in the separator (Tabera et al., 2004). One other method is based on ethanolic extraction of olive leaves, followed by incorporation of the extract into glycerine for short-path distillation. According to its proponents, this fractionation sequence allows phenol compounds in olive leaves to be extracted without degrading their natural properties. The vacuum conditions used reduce the boiling temperature of the extract, thereby preventing thermal decomposition of the target compounds (Rada et al., 2007).
28.3 COMPARISON OF PHENOLIC CONTENTS OF OLIVE MATERIALS All raw materials used by the olive oil industry contain substantially greater amounts of OPs than does olive oil, which is logical if one considers the polarity of OPs. The most concentrated phenol compound in such materials is oleuropein, which is present in olive leaves in proportions usually around 2% (w/w) but as high as 14% (w/w) in some instances. Therefore, olive leaves are an excellent source of oleuropein – and so are, to a lesser extent, small olive branches. Verbascoside, which is another precursor of hydroxytyrosol, is also present in leaves and branches, in contents around 1500 mg kg⫺1. The fact that olive leaves and small branches have traditionally been used jointly to extract phenols has led to the belief that the two materials have the same
TABLE 28.1 Comparison between reference and alternative extraction methods for isolation of OPs from olive leaves and branches. Extraction time
Extractant
Extract volume, mL
Maceration
24 h
ethanol–water at different ratios
Soxhlet extraction
16–24 h
Superheated liquid extraction
13 min
70:30 ethanol–water
11
No
Dynamic ultrasoundassisted extraction
25 min
60:40 ethanol–water
15
No
Microwave-assisted extraction
8 min
80:20 ethanol–water
24
Yes
8
Extract post-treatment Yes
Yes
Using auxiliary energies (microwaves, ultrasound, high temperatures and pressures) enhances extraction of OPs from olive leaves and branches, particularly as regards the leaching time.
262
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 28.2 Concentration of olive phenols found in extracts from olive oil and olive tree materials (n ⫽ 3, concentration ⫾ standard deviation mg kg⫺1). Sample preparation: olive oil, liquid–liquid extraction; alperujo, superheated liquid extraction; leaves, small branches and stones, microwave-assisted extraction. Determination: LC–MS–MS by multiple reaction monitoring ( Japón-Luján et al., 2008a). Sample/ Phenol (mg/kg)
Hydroxytyrosol
Luteolin-7glucoside
Apigenin-7glucoside
Verbascoside
Oleuropein
Apigenin
Luteolin
Diosmetin
Olive oil
3.0 ⫾ 0.2
n.d.
n.d.
0.08 ⫾ 0.02
n.d.
0.65 ⫾ 0.04
8.6 ⫾ 0.9
0.60 ⫾ 0.08
Alperujo
831 ⫾ 22
14.3 ⫾ 2.3
6.2 ⫾ 0.9
20.2 ⫾ 2.8
37 ⫾ 4
22.5 ⫾ 3.0
22.4 ⫾ 3.1
n.d.
Olive leaves
n.d.
155 ⫾ 10
207 ⫾ 10
1428 ⫾ 46
19050 ⫾ 880
n.d.
n.d.
n.d.
Olive small branches
22.2 ⫾ 2.0
175 ⫾ 8
10.9 ⫾ 0.8
1560 ⫾ 50
673 ⫾ 34
n.d.
n.d.
n.d.
Olive stones
18.1 ⫾ 1.9
6.2 ⫾ 0.8
0.09 ⫾ 0.01
0.15 ⫾ 0.03
0.06 ⫾ 0.02
n.d.
1.2 ⫾ 0.2
n.d.
n.d.: not detected. Alperujo is highly concentrated in hydroxytyrosol, whereas leaves and branches contain large amounts of the glycosylated flavones verbascoside and oleuropein. On the other hand, aglycone flavones are only present in olive oil, alperujo and olive stones which contain small amounts of luteolin.
phenol composition. Rather, separate extraction of the two has revealed marked differences in the most abundant OPs. Thus, branches can be used to obtain additional key phenolic compounds not present in leaves such as hydroxytyrosol, tyrosol or α-taxifolin at levels from 500 to 2000 mg kg⫺1 as shown for more than 13 olive tree varieties (Japón-Luján et al., 2008a). Table 28.2 shows the contents in various major OPs of various raw materials from picual olive trees. In short, leaves and branches contain large amounts of glycosilated flavones such as luteolin-7-glucoside and apigenin-7-glucoside. Therefore, these olive materials play a prominent role in storing phenol compounds as glucosides. In fact, no aglycone flavones are present in either raw material. The simultaneous extraction of small branches and leaves provides extracts containing a greater variety of OPs than do those obtained by separate extraction; also, it avoids the need to physically separate the two materials and facilitates industrial exploitation as a result (JapónLuján and Luque de Castro, 2007). Alternatively, leaves and branches can be used separately to obtain extracts containing hydroxytyrosol, tyrosol, and α-taxifolin (small branches) or increased amounts of oleuropein and verbascoside (leaves).
28.4 DISCRIMINATING AND CLASSIFYING POWER OF OLIVE LEAF EXTRACTS Hundreds of olive tree varieties have been selected over centuries for adaptation to various microclimates and soil types. Some cultivars are typical of specific areas, whereas
others can be found in several countries. Also, some variety names are used indifferently with similar, but clearly different varieties, and some identical varieties are given different names. As shown in other chapters, olive oil can be classified according to olive variety and cultivation area by its composition in fatty acids or other, minor compounds. Also, the contents in major OPs of olive leaves have been used as chemo-taxonomic markers developed by using chemometric models constructed by principal component analysis (PCA), hierarchical cluster analysis (HCA), the K-nearest neighbor method (KNN) or soft independent modeling of class analogy (SIMCA) methodology (Japón-Luján et al., 2006c). To this end, samplings of arbequina olive leaves were done in various Spanish areas including Córdoba, Majorca (north and south of the island), Ciudad Real, Lleida and Navarra. Samples of 13 olive varieties (viz. alameño, arbequina, azulillo, chorna, hojiblanca, lechín, manzanillo, negrillo, nevadillo, ocal, pierra, sevillano and tempranillo), all from Córdoba, were also collected. Each sample was obtained from at least five different, healthy trees in December 2005. All samples were dried, milled and kept at 4°C until use. Extraction and separation-determination of major biophenols in olive leaves (viz. oleuropein, verbascoside, apigenin7-glucoside and luteolin-7-glucoside) were done with the microwave-assisted method (Japón-Luján et al., 2006b) and HPLC–diode array detection (Japón-Luján et al., 2006a), respectively. The concentrations of the previous compounds in the samples were used to construct chemometric models. As can be seen from Figure 28.2, PCA plots clearly allowed arbequina samples to be discriminated in terms of geographical origin. Also, the plots of Figure 28.3 allowed
263
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
15000
PC2
Scores Sevillano
10000
5000
Pierra
Ocal
Lechin
Hojib
Azulillo
0
Negrillo
Manzanillo
−5000
Alameño
Chorna
−10000
A
4000
Arbequina
Nevadillo
Tempranillo
PC1
−5000
0
PC3
5000
10000
15000
Scores
Negrillo
3000 2000
Tempranillo Lechin
1000
Nevadillo
Ocal
Alameño
−1000
Arbequina
Pierra
0 Hojiblanca
Manzanillo Sevillano
−2000 B
Chorna
−10000
Azulillo
PC1
−5000
0
5000
10000
15000
FIGURE 28.2 Classification of olive tree varieties by Principal Components Analysis (PCA). PCA plots for thirteen olive varieties: PC1–PC2 (A) and PC1–PC3 (B). (Reproduced with permission of the American Chemical Society, Japón-Luján et al., 2006c.)
a well-defined area for each of the 13 olive varieties to be established. The HCA models also exhibited a very high classification– discrimination power as regards both variety and cultivation area (see dendrograms in Figure 28.4A and 28.4B, respectively). Also, KNN predictions (Table 28.3) allowed 93% of both varieties and cultivation areas to be accurately identified. On the other hand, SIMCA models provided slightly poorer predictions, with 85% of hits for tree varieties and 92% for cultivation areas (Table 28.4). Interested readers are referred to the ‘Supplementary Information’ of the paper by Japón-Luján et al. (2006c) for greater details.
for this seasonal material, which is obtained twice a year. The scant information available on this subject has come from a recent study in which olive leaves of the picual variety were stored for 7, 14, 30 or 45 days under conditions mimicking those typically used for storage after pruning or separation from the fruit in mills prior to OP extraction, namely: ●
●
●
28.5 INFLUENCE OF THE STORAGE CONDITIONS ON BIOPHENOLS FROM OLIVE LEAVES Efficient exploitation of OPs obtained by pruning olive trees and separation from the fruit at olive mills requires an accurate knowledge of the most suitable storage conditions
●
Room conditions. Samples were placed in rooms at 10 or 20°C in contact with light and air. Refrigeration or freezing conditions. Samples were placed in a refrigerator at 4°C or a freezer at ⫺18°C. Protection from open air and light by using aluminum bags under same temperature conditions as in above. Outdoor environmental conditions. Samples were placed outside the laboratory and exposed to weather changes (temperatures over the range 5–35°C).
After storage, the samples were dried and milled prior to microwave-assisted extraction, separation and detection by HPLC and triple quadrupole mass spectrometry, respectively (Japón-Luján and Luque de Castro, 2008a).
264
SECTION | I
2000
PC2
Lipids, Phenolics and Other Organics and Volatiles
Scores Mallorca (South)
1000 Navarra Mallorca (North)
0
Córdoba
Ciudad Real
−1000
Lleida
−2000
−3000 A 1000
PC1
−5000
0
PC3
Scores
5000
Mallorca (South) Ciudad Real
Mallorca (North)
500
0 Córdoba Lleida
−500
Navarra
−1000 B
PC1
−5000
0
5000
FIGURE 28.3 Discrimination of cultivation areas of the arbequina variety by Principal Components Analysis (PCA). PCA plots for the six cultivation areas: PC1–PC2 (A) and PC1–PC3 (B). (Reproduced with permission of the American Chemical Society, Japón-Luján et al., 2006c.)
As can be seen from Tables 28.5–28.9, the concentrations of oleuropein, apigenin-7-glucoside, luteolin-7-glucoside and verbascoside, all in mg kg⫺1, decreased with increasing storage time, whereas those of their respective metabolites (hydroxytyrosol, apigenin, luteolin and verbascoside) obviously exhibited the opposite trend. The effect of the storage time increased with increasing temperature from 4 to 20°C; thus, the phenolic composition remained virtually constant during the first month of storage in the samples kept at 4°C. The greatest extent of OP degradation – and hence largest increase in metabolite concentrations – was observed during the first week in the samples stored at ⫺18°C. This can be ascribed both to the abrupt change in temperature and to freezing of moisture in the leaves, which led to final concentrations of OPs and their metabolites in the samples stored at ⫺18°C in between those for the samples stored at 10°C and 20°C, thus breaking the trend of increasing stability with decreasing storage temperature. Abrupt changes in temperature can explain the changes in metabolic profile of OPs in leaves exposed to outdoor environmental conditions,
where both the decrease in the contents of oleuropein, apigenin-7-glucoside and luteolin-7-glucoside contents, and the resulting increase in those of their metabolites, peaked. Also, they may be the reason why the leaves protected from light and open air were the best preserved, and also those exhibiting the smallest decrease in OP contents and increase in metabolite contents.
28.6 POTENTIAL OF PHENOLS FROM OLIVE LEAVES The undeniable health properties of individual OPs from various parts of the olive tree and their mixtures are discussed in several chapters of this book. One potential, immediate additional use of phenols from olive leaves is for enriching edible oils. This is a growing industrial application which can be implemented in various ways, namely: (1) by liquid–liquid extraction, using the olive leaf extract as a polar donor phase; (2) by solid–liquid extraction, using the leaves themselves and an auxiliary form of energy such
265
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
Tempranillo Alameño Nevadillo Pierra Ocal Manzanillo Azulillo Hojiblanca Chorna Negrillo Lechín Sevillano Arbequina A
Arbequina (Navarra) Arbequina (Lleida) Arbequina (C.Real) Arbequina (Mallorca1) Arbequina (Mallorca2) Arbequina (Córdoba)
B FIGURE 28.4 Classification of olive tree varieties and discrimination between cultivation areas for the arbequina variety by Hierarchical Cluster Analysis (HCA). HCA dendrograms from the thirteen olive varieties (A) and six cultivation areas (B). (Reproduced with permission of the American Chemical Society, Japón-Luján et al., 2006c.)
as ultrasound to facilitate mass transfer to oil; and (3) by dissolving phenols following extraction from leaves, cleanup of the extract and removal of the extractant.
28.6.1 Enrichment of Edible Oils with Phenols from Olive Leaf Extracts: Liquid–Liquid Extraction This potential application has scarcely been explored. To the authors’ knowledge, only their own group has ever
reported a method for enriching oils with phenols from olive leaf extracts and calculated the partition coefficient between the polar and oil phases, both for individual phenols and for the combination of major ones in olive leaves (Japón-Luján and Luque de Castro, 2008b). To this end, extracts from three olive leaf varieties (picual, arbequina, and lechín from Seville) containing variable concentrations of phenols were used to study mass transfer of phenols to refined olive, sunflower and soya oils, calculate enrichment factors, and establish the phenol extract concentration required to obtain oils with
266
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 28.3 KNN model predictions. (A) Olive tree varieties
Sample
Best prediction
Real variety
150
Hojiblanca
Ocal
174
Tempranillo
Tempranillo
122
Negrillo
Negrillo
Best prediction
Real variety
70
Alameño
Alameño
99
Hojiblanca
Hojiblanca
Ocal
Ocal
Arbequina
Arbequina
80
Hojiblanca
Azulillo
178
Tempranillo
Tempranillo
136
Nevadillo
Nevadillo
170
Sevillano
Sevillano
106
Lechín
Lechín
71
Azulillo
Azulillo
154
Pierra
Pierra
169
Sevillano
Sevillano
139
Nevadillo
Nevadillo
115
Manzanillo
Manzanillo
117
Manzanillo
Manzanillo
Chorna
Chorna
63
Alameño
Alameño
157
Pierra
Pierra
127
Negrillo
Negrillo
101
Lechín
Lechín
98
Hojiblanca
Hojiblanca
88
Chorna
Chorna
Arbequina
Arbequina
7
84
4
Sample
143
Correctly classified samples
93%
Sample
Best prediction
Real cultivation zone
Córdoba
Córdoba
(B) Cultivation zones
Sample
Best prediction
Real cultivation zone
35
Ciudad Real
Ciudad Real
12
Mallorca (South)
Mallorca (South)
45
Lleida
Lleida
51
Navarra
Navarra
16
Mallorca (North)
Mallorca (North)
59
Navarra
Navarra
36
Ciudad Real
Ciudad Real
4
Córdoba
Córdoba
28
Mallorca (South)
Mallorca (South)
30
Mallorca (South)
Mallorca (North)
93%
46
Lleida
Lleida
Correctly classified samples
7
The proposed KKN models provided correct predictions of varieties and cultivation areas in the validation sets in 93% of cases. This testifies to its high predictive ability.
a given content in these compounds. The extraction step was assisted by focused microwaves (Japón-Luján et al., 2006b), and the initial and final concentrations of the target compounds in the donor and acceptor phases were monitored by using a diode array detector following isolation of the individual target compounds by HPLC (Japón-Luján et al., 2006a). Tests were performed at variable dilution of
the extracts from each leaf variety. Table 28.10 shows the dilutions used and the concentrations of major phenols in the diluted extracts. Based on the partition coefficients of oleuropein between the different oils and extracts (Table 28.11), the partition pattern depends on the particular type and concentration of fatty acids in the oil to which the biophenols are
267
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
TABLE 28.4 SIMCA model predictions. (A) Olive tree varieties
Sample
Best prediction
Second best prediction
Real variety
150
UD
UD
Ocal
174
Tempranillo
Tempranillo
Tempranillo
122
UD
UD
Negrillo
Best prediction
Second best prediction
Real variety
70
UD
UD
Alameño
99
Hojiblanca
Hojiblanca
Hojiblanca
143
Ocal
Ocal
Ocal
Arbequina
Arbequina
Arbequina
80
UD
UD
Azulillo
178
Tempranillo
Tempranillo
Tempranillo
136
Nevadillo
Nevadillo
Nevadillo
170
Sevillano
Sevillano
Sevillano
106
Lechín
Lechín
Lechín
71
Azulillo
Azulillo
Azulillo
154
Pierra
Pierra
Pierra
169
Sevillano
Sevillano
Sevillano
139
Nevadillo
Nevadillo
Nevadillo
115
Manzanillo
Manzanillo
Manzanillo
117
Manzanillo
Manzanillo
Manzanillo
Chorna
Chorna
Chorna
63
Alameño
Alameño
Alameño
157
Pierra
Pierra
Pierra
127
Negrillo
Negrillo
Negrillo
101
Lechín
Lechín
Lechín
98
Hojiblanca
Hojiblanca
Hojiblanca
88
Chorna
Chorna
Chorna
Arbequina
Arbequina
Arbequina
7
84
4
Sample
Correctly classified samples
85%
Sample
Best prediction
Second best prediction
Real cultivation zone
Córdoba
Córdoba
Córdoba
(B) Cultivation zones
Sample
Best prediction
Second best prediction
Real cultivation zone
35
Ciudad Real
Ciudad Real
Ciudad Real
12
Mallorca (South)
Mallorca (South)
Mallorca (South)
45
Lleida
Lleida
Lleida
51
Navarra
Navarra
Navarra
16
Mallorca (North)
Mallorca (North)
Mallorca (North)
59
UD
UD
Navarra
36
Ciudad Real
Ciudad Real
Ciudad Real
Córdoba
Córdoba
Córdoba
28
Mallorca (South)
Mallorca (South)
Mallorca (South)
30
Mallorca (North)
Mallorca (North)
Mallorca (North)
92%
46
Lleida
Lleida
Lleida
Correctly classified samples
4
7
UD ⫽ undefined. The proposed SIMCA models correctly predicted varieties and cultivation areas in 85 and 92% of cases, respectively. False predictions could not be classified into an erroneous category and remained undefined.
268
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 28.5 Phenols composition of olive leaves under room conditions stored at 20°C.
TABLE 28.6 Phenols composition of olive leaves under room conditions stored at 10°C.
Storage time
Phenol
Metabolite
Storage time
Phenol
Metabolite
(days)
Oleuropein
Hydroxytyrosol
(days)
Oleuropein
Hydroxytyrosol
0
34850.24 ⫾ 987.98
n.d.
0
34850.24 ⫾ 987.98
n.d.
7
34484.33 ⫾ 932.86
57.32 ⫾ 4.21
7
34687.65 ⫾ 954.02
37.54 ⫾ 2.31
14
33569.83 ⫾ 806.54
99.45 ⫾ 4.89
14
33978.34 ⫾ 844.03
66.53 ⫾ 3.01
30
31976.40 ⫾ 784.09
190.32 ⫾ 6.43
30
32568.11 ⫾ 799.41
99.23 ⫾ 5.07
45
29784.66 ⫾ 699.41
302.34 ⫾ 8.00
45
31976.43 ⫾ 702.30
165.34 ⫾ 8.56
(days)
Apigenin-7-glucoside
Apigenin
(days)
Apigenin-7-glucoside
Apigenin
0
1208.88 ⫾ 50.02
n.d.
0
1208.88 ⫾ 50.02
n.d.
7
1185.43 ⫾ 46.03
4.90 ⫾ 0.85
7
1195.20 ⫾ 47.33
4.45 ⫾ 0.97
14
1156.96 ⫾ 43.87
16.78 ⫾ 1.09
14
1169.05 ⫾ 44.44
11.34 ⫾ 1.15
30
1090.11 ⫾ 40.01
34.24 ⫾ 2.98
30
1111.65 ⫾ 42.40
27.34 ⫾ 4.01
45
960.72 ⫾ 39.23
94.95 ⫾ 6.07
45
1023.79 ⫾ 37.09
74.63 ⫾ 6.67
(days)
Luteolin-7-glucoside
Luteolin
(days)
Luteolin-7-glucoside
Luteolin
0
874.23 ⫾ 32.21
n.d.
0
874.23 ⫾ 32.21
n.d.
7
868.43 ⫾ 33.31
5.99 ⫾ 0.87
7
870.90 ⫾ 31.21
5.56 ⫾ 0.67
14
839.87 ⫾ 29.32
8.76 ⫾ 1.01
14
858.32 ⫾ 28.94
7.32 ⫾ 0.99
30
811.82 ⫾ 22.45
15.42 ⫾ 2.11
30
832.49 ⫾ 21.80
11.29 ⫾ 1.51
45
730.31 ⫾ 17.23
30.21 ⫾ 3.2
45
779.32 ⫾ 18.54
23.41 ⫾ 1.73
(days)
Verbascoside
Caffeic acid
(days)
Verbascoside
Caffeic acid
0
668.07 ⫾ 31.44
n.d.
0
668.07 ⫾ 31.44
n.d.
7
667.52 ⫾ 32.24
0.70 ⫾ 0.08
7
666.22 ⫾ 29.87
0.60 ⫾ 0.09
14
649.11 ⫾ 26.99
2.43 ⫾ 0.56
14
653.51 ⫾ 24.82
1.04 ⫾ 0.26
30
629.53 ⫾ 21.21
5.32 ⫾ 1.15
30
640.07 ⫾ 24.99
3.75 ⫾ 0.65
45
584.50 ⫾ 20.23
9.45 ⫾ 1.12
45
601.55 ⫾ 23.88
6.43 ⫾ 0.81
Phenol concentrations decreased within their first week of storage at 20°C (degradation, 1–2%), and more markedly after the second week (around 4%). The decrease after 45 days ranged from 13% for verbascoside to 21% for apeginin-7-glucoside; as a result, the concentration of metabolites ranged from 300 mg/kg for hydroxytyrosol to 9 mg kg⫺1 for caffeic acid. n.d.: not detected.
The concentrations of the four main OPs remained nearly constant during the first week of storage at 10°C – decreases never exceeded 1% – but were reduced after the second week. The decrease after 45 days ranged from 8.3% for oleuropein to 15.3% for apigenin-7glucoside; as a result, the concentration of metabolites ranged from 165 mg kg⫺1 for hydroxytyrosol to 6 mg/kg for caffeic acid. n.d.: not detected.
269
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
TABLE 28.7 Phenols composition of olive leaves under refrigeration conditions stored at 4°C.
TABLE 28.8 Phenols composition of olive leaves under refrigeration conditions stored at ⫺18°C.
Storage time
Phenol
Metabolite
Storage time
Phenol
Metabolite
(days)
Oleuropein
Hydroxytyrosol
(days)
Oleuropein
Hydroxytyrosol
0
34850.24 ⫾ 987.98
n.d.
0
34850.24 ⫾ 987.98
n.d.
7
34823.83 ⫾ 989.21
11.21 ⫾ 1.78
7
33660.21 ⫾ 940.11
87.55 ⫾ 3.89
14
34810.21 ⫾ 967.21
12.28 ⫾ 1.56
14
33490.55 ⫾ 832.00
96.69 ⫾ 4.55
30
34678.21 ⫾ 902.78
32.80 ⫾ 2.00
30
32007.41 ⫾ 790.07
199.79 ⫾ 6.41
45
34021.32 ⫾ 878.65
47.32 ⫾ 2.76
45
30876.42 ⫾ 777.09
278.34 ⫾ 11.90
(days)
Apigenin-7-glucoside
Apigenin
(days)
Apigenin-7-glucoside Apigenin
0
1208.88 ⫾ 50.02
n.d.
0
1208.88 ⫾ 50.02
n.d.
7
1204.21 ⫾ 46.90
2.09 ⫾ 0.21
7
1101.69 ⫾ 42.90
30.21 ⫾ 2.31
14
1200.89 ⫾ 43.60
2.78 ⫾ 0.67
14
1067.20 ⫾ 39.38
51.14 ⫾ 3.89
30
1191.20 ⫾ 46.84
3.90 ⫾ 1.01
30
1027.31 ⫾ 40.40
71.20 ⫾ 4.91
45
1160.21 ⫾ 42.91
10.62 ⫾ 1.75
45
1000.62 ⫾ 30.86
84.89 ⫾ 5.27
(days)
Luteolin-7-glucoside
Luteolin
(days)
Luteolin-7-glucoside
Luteolin
0
874.23 ⫾ 32.21
n.d.
0
874.23 ⫾ 32.21
n.d.
7
873.88 ⫾ 30.89
0.98 ⫾ 0.08
7
820.20 ⫾ 30.56
11.02 ⫾ 0.88
14
869.55 ⫾ 30.34
1.40 ⫾ 0.22
14
814.62 ⫾ 24.33
15.98 ⫾ 1.67
45
845.42 ⫾ 21.33
8.29 ⫾ 1.07
30
802.90 ⫾ 22.97
20.90 ⫾ 1.99
(days)
Verbascoside
Caffeic acid
45
758.48 ⫾ 16.66
25.90 ⫾ 2.83
0
668.07 ⫾ 31.44
n.d.
(days)
Verbascoside
Caffeic acid
7
667.92 ⫾ 30.32
0.58 ⫾ 0.07
0
668.07 ⫾ 31.44
n.d.
14
666.32 ⫾ 29.31
0.89 ⫾ 0.23
7
647.20 ⫾ 26.05
2.59 ⫾ 0.80
30
658.23 ⫾ 28.95
1.04 ⫾ 0.42
14
645.99 ⫾ 25.01
2.63 ⫾ 0.89
45
643.55 ⫾ 27.40
2.23 ⫾ 0.91
30
630.67 ⫾ 26.64
4.99 ⫾ 0.97
45
595.21 ⫾ 19.19
7.76 ⫾ 1.00
Storage at 4°C for a month caused almost no change in OP concentrations – decreases never exceeded 1.5%. After 45 days, degradation was slightly higher (the concentration of oleuropein, apigenin-7-glucoside, luteoin-7-glucoside and verbascoside decreased by 2.4, 4.0, 3.4 and 3.8%, respectively). Metabolite concentrations were the lowest found at the studied temperatures and ranged from 47 mg/kg for hydroxytyrosol to 2 mg kg⫺1 for caffeic acid. n.d.: not detected.
After one week of storage at ⫺18°C, the OP concentrations exhibited the greatest changes at all studied temperatures (from 3.5% for oleuropein to 9.0% for apigenin-7-glucoside). The decrease was smaller during the second week and ranged from 11.0% for verbascoside to 13.3% for apigenin-7-glucoside after 45 days. Concomitantly, the concentrations of metabolites also peaked after a week. However, those after 45 days were also higher than those observed during storage at 10°C, but lower than those obtained at 20°C; such concentrations ranged from 278 mg kg⫺1 for hydroxytyrosol to 3 mg/kg for caffeic acid. n.d.: not detected.
270
SECTION | I
TABLE 28.9 Phenols composition of olive leaves under exterior environmental conditions. Storage time
Phenol
Metabolite
(days)
Oleuropein
Hydroxytyrosol
0
34850.24 ⫾ 987.98
n.d.
7
34004.21 ⫾ 967.74
96.34 ⫾ 9.34
14
33178.12 ⫾ 910.99
199.31 ⫾ 23.94
30
31024.74 ⫾ 923.34
378.97 ⫾ 59.34
45
28890.33 ⫾ 896.54
597.60 ⫾ 58.98
(days)
Apigenin-7-glucoside
Apigenin
0
1208.88 ⫾ 50.02
n.d.
7
1165.55 ⫾ 49.54
13.45 ⫾ 3.04
14
1109.41 ⫾ 48.88
48.90 ⫾ 4.05
30
997.09 ⫾ 43.87
106.76 ⫾ 7.77
45
850.88 ⫾ 35.13
174.54 ⫾ 8.90
(days)
Luteolin-7-glucoside
Luteolin
0
874.23 ⫾ 32.21
n.d.
7
860.43 ⫾ 31.66
7.56 ⫾ 3.14
14
823.41 ⫾ 27.65
9.67 ⫾ 2.12
30
759.02 ⫾ 20.87
27.89 ⫾ 5.90
45
648.06 ⫾ 17.98
50.98 ⫾ 9.62
(days)
Verbascoside
Caffeic acid
0
668.07 ⫾ 31.44
n.d.
7
660.22 ⫾ 30.34
0.85 ⫾ 0.10
14
640.01 ⫾ 28.57
3.04 ⫾ 0.56
30
595.47 ⫾ 30.09
8.89 ⫾ 0.89
45
550.92 ⫾ 25.71
14.87 ⫾ 1.32
The reduction in OP concentrations and increase in metabolite concentrations peaked in the samples stored under outdoor environmental conditions relative to those stored under other conditions for 14, 30 or 45 days. Degradation after 7 days was only surpassed by storage at ⫺18°C. n.d.: not detected.
transferred. The ‘Supplementary information’ of the paper by Japón-Luján and Luque de Castro (2008b) provides more detailed information about the partition coefficients of other major biophenols.
Lipids, Phenolics and Other Organics and Volatiles
28.6.2 Enrichment of Edible Oils with Phenols from Olive Leaves: Solid–Liquid Extraction The authors have developed a faster way to enrich edible oils with major phenols from olive leaves involving bringing the leaves into contact with the oil and facilitating mass transfer between the two with the aid of auxiliary energy (ultrasound). A continuous, ultrasound-assisted procedure similar to that for the extraction of phenols from leaves of the picual variety (Figure 28.1B) was used for direct enrichment of edible oils (olive, sunflower and soya) with the major phenols from olive leaves (viz. oleuropein, verbascoside, apigenin-7-glucoside and luteolin-7-glucoside). Multivariate methodology was used to fully optimize the procedure, and HLPC in combination with tandem mass spectrometry in the multiple reaction monitoring mode was used to quantify transferred compounds. Under the optimum operating conditions, reaching solid–liquid partition equilibrium took only 20 min. The OP concentrations thus supplied to the oils were found to depend on the particular type of oil and fell in the range 14.45–9.92 μg mL⫺1 for oleuropein, 2.29–2.12 μg mL⫺1 for verbascoside, 1.91–1.51 μg mL⫺1 for apigenin-7-glucoside and 1.60–1.42 μg mL⫺1 for luteolin-7-glucoside (Japón-Luján et al., 2008b). One salient advantage of using ultrasound here, in addition to dramatic acceleration of the partition equilibrium, was the negligible increase in temperature observed, which prevented degradation of the target analytes and changes in the sensory properties of the oils. The rather disparate behavior of oils towards ultrasound observed by CañizaresMacías et al. (2004) (viz. fast oxidation of stable olive oil, which shortened the time needed to determine its oxidative stability) was a result of direct insertion of the probe in the oil facilitating formation of free radicals and oxidation of the oil as a result, which was not the case under the conditions used in our procedure. Direct transfer of phenols from leaves to oil avoids time-consuming preliminary extractions and contact of edible oils with organic solvents. The main shortcoming of this enrichment procedure is that no control over the concentration of OPs that is transferred to the oil can be exerted; such a concentration depends on the original contents in the leaves and the specific type of oil, which dictates the partition equilibrium of each phenol.
28.6.3 Dissolution of Phenols Following Extraction from Leaves, Clean-up of the Extract and Removal of the Extractant The industries currently using OPs to increase the value of their oils tend to obtain them from olive leaves and, occasionally, also from the waste of olive oil production (alperujo, mainly). Although each industry keeps its procedure secret, it usually involves the following steps: (1) extraction,
271
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
TABLE 28.10 Enrichment of oils after applying the proposed method as a function of type of leaf variety and overall oil – extract distribution factor of the main phenols, but oleuropein, as a function of the type of oil. Type of oil
Phenol
Phenol concentration in oil after applying the proposed method* (mg/L) Picual
Olive
Sunflower
Soya
Oil/extract distribution factor
Arbequina
Lechín
Apigenin-7-glucoside 5.93 ⫾ 0.12
15.13 ⫾ 0.25
10.00 ⫾ 0.14
0.031 ⫾ 0.001
Luteolin-7-glucoside
6.27 ⫾ 0.12
12.09 ⫾ 0.30
10.02 ⫾ 0.32
0.026 ⫾ 0.001
Verbascoside
5.18 ⫾ 0.19
10.97 ⫾ 0.35
8.16 ⫾ 0.36
0.021 ⫾ 0.003
Apigenin-7-glucoside 3.20 ⫾ 0.05
8.25 ⫾ 0.22
5.94 ⫾ 0.42
0.017 ⫾ 0.001
Luteolin-7-glucoside
5.85 ⫾ 0.17
11.32 ⫾ 0.06
9.80 ⫾ 0.33
0.025 ⫾ 0.002
Verbascoside
5.97 ⫾ 0.17
13.14 ⫾ 0.13
10.41 ⫾ 0.30
0.026 ⫾ 0.002
Apigenin-7-glucoside 3.24 ⫾ 0.06
8.94 ⫾ 0.19
6.61 ⫾ 0.09
0.018 ⫾ 0.001
Luteolin-7-glucoside
5.66 ⫾ 0.03
11.30 ⫾ 0.32
9.90 ⫾ 0.32
0.025 ⫾ 0.002
Verbascoside
5.80 ⫾ 0.11
13.19 ⫾ 0.01
10.19 ⫾ 0.11
0.025 ⫾ 0.002
The initial concentration of these phenols in the oils was under LOD. Mass transfer (expressed as a distribution factor) to the fatty phase was found to depend on the particular type of olive phenol and oil; transfer was specially effective in olive oil, but similar to sunflower and soya oils.
TABLE 28.11 Oil – extract distribution of oleuropein as a function of the type of oil, variety of the leaves extracted and concentration in the extracts. Variety
Oleuropein concentration in leaf extract (mg L⫺1)
Concentration of oleuropein in Oil–extract distribution factor oil after 15 min of liquid–liquid For each variety Overall extraction (mg L⫺1) Olive oil
Picual
Arbequina
Lechín
323
9.51 ⫾ 0.42
646
20.30 ⫾ 0.72
1293
40.37 ⫾ 0.67
2586
94.86 ⫾ 1.44
5173
187.29 ⫾ 3.19
808
24.47 ⫾ 0.92
1615
52.72 ⫾ 1.54
3230
112.68 ⫾ 2.91
6460
215.20 ⫾ 5.27
12921
442.44 ⫾ 16.62
498
15.29 ⫾ 0.57
997
31.41 ⫾ 1.01
1994
67.03 ⫾ 2.79
3989
127.50 ⫾ 1.77
7977
261.07 ⫾ 10.43
0.033 ⫾ 0.002
0.033 ⫾ 0.001
0.033 ⫾ 0.001
0.032 ⫾ 0.001
(Continued)
272
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 28.11 (Continued) Variety
Oleuropein concentration in leaf extract (mg L⫺1)
Concentration of oleuropein in Oil–extract distribution factor oil after 15 min of liquid–liquid For each variety Overall extraction (mg L⫺1) Sunflower oil
Picual
Arbequina
Lechín
323
3.03 ⫾ 0.48
646
7.48 ⫾ 0.86
1293
17.29 ⫾ 1.46
2586
35.50 ⫾ 0.47
5173
66.40 ⫾ 3.21
808
8.99 ⫾ 0.91
1615
20.96 ⫾ 0.81
3230
39.96 ⫾ 1.97
6460
78.23 ⫾ 3.41
12921
162.39 ⫾ 3.41
498
4.98 ⫾ 0.49
997
12.03 ⫾ 1.03
1994
24.85 ⫾ 1.58
3989
56.16 ⫾ 0.81
7977
100.38 ⫾ 5.21
0.012 ⫾ 0.002
0.012 ⫾ 0.001
0.012 ⫾ 0.001
0.012 ⫾ 0.001
Soya oil Picual
Arbequina
Lechín
323
3.34 ⫾ 0.57
646
7.55 ⫾ 0.87
1293
15.14 ⫾ 0.79
2586
36.87 ⫾ 1.59
5173
67.19 ⫾ 3.31
808
9.00 ⫾ 0.76
1615
19.66 ⫾ 2.07
3230
42.80 ⫾ 1.67
6460
79.99 ⫾ 2.89
12921
168.74 ⫾ 1.74
498
6.06 ⫾ 0.68
997
12.53 ⫾ 0.61
1994
25.82 ⫾ 0.72
3989
58.03 ⫾ 3.34
7977
101.22 ⫾ 1.68
0.013 ⫾ 0.001
0.012 ⫾ 0.001
0.013 ⫾ 0.001
0.013 ⫾ 0.001
The distribution factor for oleuropein – the most abundant phenol in olive leaves – was independent of the olive variety used to obtain the extract, but clearly influenced by the specific type of oil, olive oil being the most strongly enriched after the liquid– liquid extraction step (roughly three times more than the other oils).
CHAPTER | 28 Extraction of Oleuropein and Related Phenols from Olive Leaves and Branches
whether conventional or accelerated by auxiliary energies; (2) clean-up by retention of OPs on an appropriate sorbent material and elution with an organic solvent, preferably of low toxicity (e.g. ethanol) – for greater efficiency, the extract can be cleaned up by passage through activated charcoal in order to remove extraneous compounds such as chlorophylls; (3) calculation of the final extract richness in the target compounds; (4) extract-to-oil OP transfer by liquid–liquid extraction, using an appropriate volume ratio in order to obtain the concentration to be certified on the product label. This procedure is the most laborious, but also that ensuring the highest reproducibility. Olive oil industries marketing oils enriched with OPs buy these additives as solid products.
SUMMARY POINTS ●
●
●
●
●
Olive leaves and branches contain high amounts of the most important, very healthy hydrophilic phenols. Fast, automatic and efficient methods to extract phenols from olive leaves and branches have recently been developed. Phenols thus obtained allowed discrimination both between olive tree varieties and cultivation areas. Changes in phenols as a function of the storage conditions of leaves and branches were studied with a view to assessing the industrial potential of phenol extracts from these raw materials. The potential of olive phenols to enrich phenol-free oils and other foods either by mass transfer to them from leaf extracts or directly from leaves should be taken into consideration.
REFERENCES Agalias, A., Melliou, E., Magiatis, P., Mitaku, S., Gikas, E., Tsarbopoulos, A., 2003. Quantitation of oleuropein and related metabolites in decoctions of Olea europaea leaves from ten Greek cultivated varieties by HPLC with diode array detection (HPLC–DAD). J. Liq. Chromatogr. Relat. Technol. 28, 1557–1571. Cañizares-Macías, M.P., García-Mesa, J.A., Luque de Castro, M.D., 2004. Fast ultrasound-assisted method for the determination of the oxidative stability of virgin olive oil. Anal. Chim. Acta 502, 161–166. Pinnell, S.R., Omar, M.M., 2004. U.S. Pat. Appl. Publ. 6743449.
273
Guinda, A., Lanzón, A., Ríos, J.J., Albi, T., 2002. Aislamiento y cuantificación de los componentes de la hoja del olivo: extracto de hexano. Grasas Aceites 4, 419–422. Japón-Luján, R., Luque de Castro, M.D., 2006. Superheated liquid extraction of oleuropein and related biophenols from olive leaves. J. Chromatogr. A 1136, 185–191. Japón-Luján, R., Luque-Rodríguez, J.M., Luque de Castro, M.D., 2006a. Dynamic ultrasound-assisted extraction of oleuropein and related biophenols from olive leaves. J. Chromatogr. A 1108, 76–82. Japón-Luján, R., Luque-Rodríguez, J.M., Luque de Castro, M.D., 2006b. Multivariate optimisation of the microwave-assisted extraction of oleuropein and related biophenols from olive leaves. Anal. Bioanal. Chem. 385, 753–759. Japón-Luján, R., Ruiz-Jiménez, J., Luque de Castro, M.D., 2006c. Discrimination and classification of olive tree varieties and cultivation zones by biophenols contents. J. Agric. Food Chem. 54, 9706–9712. Japón-Luján, R., Luque de Castro, M.D., 2007. Small branches of olive tree: a source of biophenols complementary to olive leaves. J. Agric. Food Chem. 55, 4584–4588. Japón-Luján, R., Priego Capote, F., Marinas, A., Luque de Castro, M.D., 2008a. Liquid chromatography/triple quadrupole tandem mass spectrometry with multiple reaction monitoring for optimal selection of transitions to evaluate nutraceuticals from olive-tree materials. Rapid Comm. Mass Spectrom. 22, 855–864. Japón-Luján, R., Janeiro, P., Luque de Castro, M.D., 2008b. Solid–liquid transfer of phenols from olive leaves for the enrichment of edible oils by a dynamic ultrasound-assisted approach. J. Agric. Food Chem., 56, 7231–7235. Japón-Luján, R., Luque de Castro, M.D., 2008a. Influence of storage conditions on the metabolic profile of olive leaf biophenols. Unpublished results. Japón-Luján, R., Luque de Castro, M.D., 2008b. Liquid–liquid extraction for the enrichment of edible oils with phenols from olive leaf extracts. J. Agric. Food Chem. 56, 2505–2511. Luque de Castro, M.D., Japón-Luján, R., 2006. State-of-the-art and trends in the analysis of oleuropein and derivatives. Trends Anal. Chem. 25, 501–510. Luque de Castro, M.D., Priego Capote, F., 2006. Analytical applications of ultrasound. Elsevier, Amsterdam pp. 1–11. Montedoro, G.F., Servili, M., Baldioli, M., Selvaggini, R., Miniati, E., Macchioni, J., 1993. Simple and hydrolysable compounds in virgin olive oil. 3. Spectroscopic characterization of the secoiridoid derivatives. J. Agric. Food Chem. 41, 2228–2234. Rada, M., Guinda, A., Cayuela, J., 2007. Solid/liquid extraction and isolation by molecular distillation of hydroxytyrosol from Olea europaea L. leaves. Eur. J. Lipid Sci. Technol. 109, 1071–1076. Tabera, J., Guinda, A., Ruiz Rodríguez, A., Señoráns, F.J., Ibáñez, E., Albi, T., Reglero, G., 2004. Countercurrent supercritical fluid extraction and fractionation of high-added-value compounds from a hexane extract of olive leaves. J. Agric. Food Chem. 52, 4774–4779.
This page intentionally left blank
Chapter 29
The Occurrence of the Biogenic Amine Melatonin in Olive Oil: Implications in Health and Disease Prevention Rafael Fernández-Montesinos1, Cristina de la Puerta2, Pedro P. García-Luna3, Russel J. Reiter4 and David Pozo1 1
Department of Cell Therapy and Regenerative Medicine, CABIMER-Andalusian Center for Molecular Biology and Regenerative Medicine (CSIC-University of Seville-UPO-Junta de Andalucia), Seville, Spain 2 Department of Medical Biochemistry and Molecular Biology, University of Seville Medical School, Spain 3 Department of Endocrinology and Nutrition, Clinical Nutrition Unit, Virgen del Rocio University Hospital, Seville, Spain 4 Department of Cellular and Structural Biology, University of Texas Health Science Center, San Antonio, TX
29.1 INTRODUCTION Melatonin (N-acetyl-5-methoxy-tryptamine) was the first methoxyindole found in mammalian tissue. Melatonin is a biogenic amine with a chemical structure similar to serotonin (Figure 29.1). Melatonin was isolated from thousands of bovine pineal glands by Aaron B. Lerner (Lerner et al., 1958) in a remarkable attempt to characterize a lightening factor of melanosomes in tadpoles; this followed the initial observations made by McCord and Allen in 1917 (see the following reviews on melatonin for further details (Carrillo-Vico et al., 2006; Reiter et al., 2007a, 2007b; Szczepanik, 2007). Its biosynthesis from tryptophan involves four well-defined intracellular steps catalyzed by tryptophan hydroxylase, aromatic amino acid decarboxylase, arylalkylamine-Nacetyltransferase, and hydroxyndole-O-methyltransferase. The pineal gland under the control of postganglionic sympathetic fibers is the factory where melatonin is produced and released during darkness (Figure 29.2). Consequently, this temporal information is used as a circadian-based photoperiod message and a seasonal timer and the rhythm represents one of melatonin’s first and best-characterized physiological roles. Nobel Prize laureate Julius Axelrod performed many of these seminal experiments that elucidated the role of melatonin and the pineal gland in regulating sleep–wake cycles (circadian rhythms). The pineal indolamine melatonin is considered an important member of the complex neuroendocrine–immune Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
network (Carrillo-Vico et al., 2006; Szczepanik, 2007). In addition to its chronobiotic properties, melatonin and its metabolites show a remarkable functional versatility by exhibiting antioxidant (Martinez-Cruz et al., 2002; Tan et al., 2007b), oncostatic (Leon-Blanco et al., 2003, 2004; Lissoni et al., 2008), and anti-aging (von Gall and Weaver, 2008) properties. The molecular mechanisms of action responsible for its pleiotropic effects involved several paradigms: high-affinity G-protein-coupled receptors (GPCR) at membrane level, direct interaction with cytosol and nuclear proteins, and both direct radical scavenging of free radical species and modification of redox-dependent processes (Pozo et al., 1997, 2004; Reiter et al., 2007b; Tan et al., 2007b). Regarding the mechanisms mediated by GPCR receptors, it is worth mentioning the recent reports that show further possibilities to increase the signaling diversity by receptor heterodimerization (Levoye et al., 2006) and interaction with PDZ (PSD-95/Drosophila Disc large/ ZO-1 homology) domain-containing scaffolding proteins (Guillaume et al., 2008) (Figure 29.3). During the last decade, much attention has centered on melatonin, which was considered to be only a hormone of the pineal gland for many years. When highly sensitive methods and antibodies to indole alkylamines became available, melatonin was identified not only in the pineal gland, but also in extrapineal tissues and cells, e.g. the retina, Harderian gland, gut mucosa, cerebellum, enterochromaffin cells, airway epithelium, liver, kidney, adrenals, pancreas, thyroid gland and thymus. Melatonin has been
275
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
276
SECTION | I
Melatonin
O
CH2
CH3 N H
CH2
N H CO CH3
FIGURE 29.1 Melatonin chemical stucture. Melatonin is an indolic compound (biogenic indoleamine) related structurally to other important substances, including tryptophan, serotonin, and indole-3-acetic acid. Melatonin is found widely in nature, where it has been identified in vertebrates, invertebrates, plants, unicellular eukaryotes, algae and even bacteria.
found in some non-endocrine cells, e.g. mast cells, natural killer cells, eosinophilic leukocytes, platelets and endothelial cells among others. The production of melatonin in higher vertebrates has been reported in retina, Harderian glands and the enterochromaffin cells, but the contribution of extrapineal melatonin to blood melatonin concentrations is still a matter of debate. Currently, the increased number of reports defining the enzymatic machinery of melatonin synthesis and receptors able to transduce the melatonin message highlight the need for considerable research effort to decipher the key issue as to what cells endogenously produce melatonin, under what circumstances melatonin is released if it is in fact released, whether there exists some type of melatonin-concentrating mechanisms within cells, and to what extent this extra-pineal melatonin is related to the classical circadian-related processes (see general reviews on melatonin indicated in the Reference section). Table 29.1 summarizes some of the key facts of melatonin.
29.2 MELATONIN IN EDIBLE PLANTS: WHY MELATONIN EXISTS IN THE PLANT KINGDOM In 1995, melatonin was reported in a variety of vascular plant species, mainly in mono- and dicotyledon edible plants (Dubbels et al., 1995; Hattori et al., 1995). Plant research on melatonin is in its infancy. Although melatonin’s function in plants is not well known, yet a hypothesis can be put forward that it is a plant growth and development regulator that promotes vegetative growth in a similar way to auxin indole-3-acetic acid. The experimental research carried out so far in order to elucidate the role of melatonin in the plant kingdom could be included in the following categories: (a) reproductive development, including circadian
Lipids, Phenolics and Other Organics and Volatiles
rhythms, (b) cell protection, and (c) vegetative development (Kolar et al., 1999; Kolar and Machackova, 2005; Tan et al., 2007a). Regarding the roles in reproductive development, it seems to be that melatonin synthesis is not directly light-regulated but it does exhibit a circadian rhythm (Kolar and Machackova, 2005; Tan et al., 2007a). Also, whether or not melatonin affects flowering remains unclear (Kolar and Machackova, 2005). Studies by Van Tassel et al. concluded that there is no conclusive evidence that melatonin levels and photoperiod influence ripening (Van Tassel et al., 2001). Cell-protective effects have been proposed to be related with a role as a germ tissue protection factor and/ or as a dormancy maintenance agent in seeds due to the high concentration observed (Manchester et al., 2000). In this sense, it has been suggested that melatonin could act as a stress-protecting agent in flower development, providing an adaptative mechanism to ensure reproduction (Lei et al., 2004). As a regulator of vegetative development, Arnao and co-workers have recently demonstrated its role as a growth promoter (Hernandez-Ruiz et al., 2004, 2005). It is clear that further studies are required to identify additional physiological processes in which melatonin could be involved, with a specific role or rather in conjunction with other classical hormones such as auxin.
29.3 MELATONIN IN EDIBLE PLANTS: CORRELATION WITH BLOOD LEVELS OF MELATONIN Since substances normally found in foods are out of the jurisdiction of the US Food and Drug Administration, melatonin was considered as a dietary supplement, and thus available over-the-counter with great media impact. Moreover, its presence in edible plants, such as bananas, cherries, white sprouts, and cucumbers (Reiter and Tan, 2002), have shown that there is a correlation between dietary vegetable intake and blood levels of melatonin when these products are consumed (Reiter et al., 2001, 2005), demonstrating that this molecule is capable of raising the blood plasma concentration of melatonin when foodstuffs are eaten (Reiter et al., 2001; Nagata et al., 2005). As we mentioned before, among a number of actions, melatonin is a direct free radical scavenger and an indirect antioxidant (Tan et al., 2007b). Melatonin directly detoxifies the hydroxyl radical (⫺OH), hydrogen peroxide, nitric oxide, peroxynitrite anion, peroxynitrous acid, and hypochlorous acid (Tan et al., 2007b). Some of the products that are produced when melatonin detoxifies reactive species are also highly efficient scavengers (Tan et al., 2007b). As a result, a cascade of scavenging reactions may enhance the antioxidant capacity of melatonin. Additionally, melatonin increases the activity of several antioxidative enzymes, thereby improving its ability to protect macromolecules
N
H N
O
N CH3 H H OH
HN
OH C H2N
OH
OH
H2O C N
H2N
CH CH2
N OH
N
N O
OH
H2N
CH3
O OH
CH2
Tryptophan
N H 5-Hydroxytryptophan
TPH
CH2 CH2
B6
HO N H
H2N
CO2
CH
HO
N H
AADC
5-Hydroxytryptamine (Serotonin) O H3C
C
SCoA
NH2
N O C HO
AA-NAT
N
N
HSCoA
N NH2
CH CH2 CH2 S CH2 O H2 N
C
H N
OH
CH3 O C HO
O H3C
N
N OH
CH2
N
H3C
H2N OH
N H N-Acetyl-5-methoxytryptamine (Melatonin)
O
CH CH2 CH2 S CH2 O
CH2
H3CO
N
C
H N
OH
CH2
HO
HIOMT
CH2
CHAPTER | 29 The Occurrence of the Biogenic Amine Melatonin in Olive Oil
H2N
N H N-Acetyl5-hydroxytryptamine
FIGURE 29.2 Biosynthesis of melatonin. The biosynthesis of melatonin is initiated by the uptake of the essential amino acid tryptophan into pineal parenchymal cells. Tryptophan is the least abundant of the essential amino acids in normal diets. It is converted to another amino acid, 5-hydroxytryptophan, through the action of the enzyme tryptophan hydroxylase and then to 5-hydroxytryptamine (serotonin) by the enzyme aromatic amino acid decarboxylase. Serotonin concentrations are higher in the pineal than in any other organ or in any brain region. They exhibit a striking diurnal rhythm remaining at a maximum level during the daylight hours and falling by more than 80% soon after the onset of darkness as the serotonin is converted to melatonin, 5-hydroxytryptophol and other methoxyindoles. Serotonin’s conversion to melatonin involves two enzymes: 5-hydroxytryptamine-N-acetyltransferase, which converts the serotonin to N-acetylserotonin, and a hydroxyindole-O-methyltransferase which transfers a methyl group from S-adenosylmethionine to the 5-hydroxyl of the N-acetylserotonin. The activities of both enzymes rise soon after the onset of darkness because of the enhanced release of norepinephrine from sympathetic neurons terminating on the pineal parenchymal cells. Tryptophan hydroxylase, TPH; aromatic L-amino acid decarboxylase, AADC; 5-hydroxytryptamine-N-acetyltransferase, AA-NAT; hydroxyindole-O-methyltransferase, HIOMT; acetylcoenzyme A, HSCoA; vitamin B6, B6.
277
278
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
Melatonin O O CH3
MT1/MT2
Orphan GPCRs
H CH2 CH N CO 2 N H
CH3
Free radicals scavenging
I II III IV V β γ
Lipid rafts
α
γ β α
Free radicals scavenging
CaM
PKC
Quinone reductase 2
Nuclear receptors FIGURE 29.3 Melatonin’s mechanisms of action. Melatonin produces its cellular effects via a variety of mechanisms utilizing both receptor-dependent and receptor-independent means. Since melatonin is an amphiphilic molecule, it is capable of freely entering cells. Most of melatonin’s antioxidant effects are believed to be receptor-independent modes of action, i.e., direct free radical scavenging. Melatonin’s effects are also mediated by specific high-affinity G-protein-coupled receptors localized on the plasma membrane (MT1, MT2). A third binding site (MT3) belongs to the quinone reductase 2 family of detoxifying enzymes. Interactions between melatonin and nuclear receptors of the retinoid-related orphan receptor (RZR/ROR) subfamily of nuclear receptors and cytosolic proteins have also been described. G-protein-coupled receptor, GPCR; membrane melatonin receptor, MT; heterotrimeric G protein subunits, αβγ; calmodulin, CaM; protein kinase C, PKC.
from oxidative stress (Reiter et al., 2007b). Because physiologic concentrations of melatonin in the blood are known to correlate with the total antioxidant capacity of the serum, consuming foodstuffs containing melatonin may be helpful in lowering oxidative stress, among other potential healthy effects ascribed to melatonin (Reiter et al., 2007b). It is worth noting that experimental evidence shows no major toxic effects after chronic melatonin treatments (Cheung et al., 2006).
29.4 MELATONIN AS A NEW PHYTOCHEMICAL IN OLIVE OIL The Mediterranean diet has long accumulated scientific evidence regarding its property of contributing to a healthy life and of preventing disease. Virgin olive oil constitutes one of the featured items of this ancient diet (Stark and Madar, 2002;
Perez-Jimenez, 2005; Alexandratos, 2006). In this regard, numerous studies have demonstrated that it has relevant properties which convey benefits to an individual’s health, when taken as part of a habitual diet (Perez-Jimenez, 2005). Virgin olive oil contains trace amounts of a wide variety of substances such as different polyphenols, tocopherols, etc., which have proven to exert beneficial actions upon health, such as reduction of risk of coronary heart disease (Perona et al., 2006), prevention or reduction of risk of certain cancers (Wahle et al., 2004; Perez-Jimenez, 2005), as well as immunomodulatory effects and the ability of partially reverting some inflammatory conditions (de la Puerta et al., 2000; Miles et al., 2005; Bogani et al., 2006). Due to the presence of these minor constituents and the health benefits exerted by them, virgin olive oil is considered as a functional food, since it complies with all the requirements to be classified under this nutritional definition (Stark and Madar, 2002; Alexandratos, 2006).
279
TABLE 29.1 Features of melatonin. 1. Melatonin is produced in the pineal gland – a pea-sized gland located just above the middle of the brain – under the control of postganglionic sympathetic fibers, and it is released during periods of darkness 2. The amount of melatonin released at night varies among individuals, but it is somewhat related to age 3. The U.S. Dietary Supplement Health and Education Act of 1994 allows it to be sold as a dietary supplement (e.g., vitamins and minerals). However, safe and appropriate use of melatonin needs further testing 4. Melatonin, besides being a sleep hormone, is a very powerful antioxidant and immunoregulator. In some mammals with short mating periods (due to hibernation), melatonin also is related to the production of other hormones which control sexual activity 5. Melatonin presence was first reported outside the vertebrate class in the alga Gonyaulax. It is also present in insects, fungi, protists and in prokaryotes 6. Due to its antioxidant capacity, in plants melatonin is suggested to protect them from intrinsic and environmental oxidative stress. Melatonin plant research is in its early stages
Based upon the discussed facts in the sections above related to melatonin, it is now assumed that melatonin’s presence in extra virgin olive oil contributes to the chemical diversity of this product and to its functional value. We have recently determined that melatonin is present in olive oil, especially in the extra virgin varieties (De la Puerta et al., 2007). For its measurement, we performed an extraction based on sequential methanol and chloroform extraction from extra virgin olive oil. This approach was partly validated by performing an immunoprecipitation of melatonin as well as a competitive enzyme immunoassay (De la Puerta et al., 2007). Extra virgin olive oils from different regions of Spain, refined olive oil and sunseed oil samples commercially available were analyzed. Control quality requirements were fulfilled under national and EU olive oil registered designation of origins (RD 308/1983; EC 2472/97; EC 1176/2003). Melatonin extraction based on sequential methanol and chloroform extraction used in this report was confirmed by immunoprecipitation (Figure 29.4). The melatonin values in the case of the extra virgin olive oil samples were roughly about double those of both refined olive and sunflower oil samples (Table 29.2). Only the samples from the Designation of Origin (D.O.) Bajo Aragón had melatonin levels (71.48 ⫾ 15.11 pg mL⫺1) comparable to one of the refined olive oils analyzed (74.77 ⫾ 6.92 pg mL⫺1). Generally, the higher melatonin
Immunoprecipitated melatonin (% of max)
CHAPTER | 29 The Occurrence of the Biogenic Amine Melatonin in Olive Oil
100
75
50
25
0 Anti-Ig control Ab
Anti-Mel Ab
FIGURE 29.4 Determination of melatonin in samples of olive oil content by ELISA after immunoprecipitation. To validate the specificity of the ELISA method, melatonin was determined after immunoprecipitation with anti-melatonin antibodies (Mel I.P.) or isotype control antibodies (control I.P.). Extracted samples were incubated in PBS with anti-melatonin antibody (Biogenesis, Poole, England) (5 μg mL–1) or rabbit IgG control Ab (Sigma-Aldrich, St. Louis, MO, USA) for 1 hour at 4°C in a rotary shaker. Samples were incubated with protein A-Sepharose beads (Amersham Biosciences) for 30 min at 4°C. After PBS washing, the immunoprecipitated complexes were discarded and the supernatants recovered in ELISA buffer for melatonin measurement. Reprinted from De la Puerta et al., Food Chem. (2007) 104, 609–612, with permission.
TABLE 29.2 Melatonin values (pg mL⫺1) in refined olive oil and extra virgin olive oil. Oil sample
Mean ⫾ S.E.M.
D.O. Sierra Mágina
107 ⫾ 36.2
D.O. Siurana
95 ⫾ 20.3
D.O. Bajo Aragón
71 ⫾ 15.11
D.O. Montes de Toledo
108 ⫾ 9.76
D.O. Baena
119 ⫾ 2.93
D.O. Sierra de Segura
89 ⫾ 2.93
D.O. Les Garrigues
98 ⫾ 3.89
D.O. Toscano
108 ⫾ 17.32
Refined olive oil sample 1
53 ⫾ 5.6
Refined olive oil sample 2
75 ⫾ 6.92
Refined sunflower oil sample
50 ⫾ 12.2
This table shows the concentration of melatonin in different samples of olive oil from different Designation of Origin (D.O.). Reprinted from De la Puerta et al., Food Chem. (2007) 104, 609–612, with permission.
280
SECTION | I
levels were found in those samples with no heat-treatment or chemical processing, i.e., extra virgin olive oil. Epidemiological studies provide evidence to support the concept that a Mediterranean diet has beneficial effects on diseases associated with oxidative damage, such as coronary heart disease, cancer, cardiovascular and neurodegenerative diseases (Perez-Jimenez, 2005; Alexandratos, 2006). A common key in the dietary habits of Mediterranean populations is the consumption of olive oil (Alexandratos, 2006). Growing experimental evidence demonstrates that olive oil phytochemicals are potent bioactive compounds responsible for the functional properties of olive oil, particularly extra virgin olive oil (Stark and Madar, 2002). Given its antioxidant capacity (Tan et al., 2007b), its presence in edible plants (Hattori et al., 1995; Reiter and Tan, 2002; Reiter et al., 2005) and its role in key physiological responses with therapeutic implications (Carrillo-Vico et al., 2006; PandiPerumal et al., 2006), we tested whether melatonin is present in olive oil. The highest levels of melatonin in plants are in Chinese herbal products (Murch et al., 1997; Chen et al., 2003). It is remarkable to note that purslane (Portulaca oleracea), which is the most important source of terrestrial α-linolenic acid in edible wild plants (Simopoulos and Salem, 1986), is also rich in melatonin (Simopoulos et al., 2005). Moreover, it is eaten as a leaf vegetable as part of the Mediterranean diet (Simopoulos, 2001). Unfortunately, there have been few systematic evaluations of melatonin levels in plant materials with the walnut being the first common tree nut in which melatonin has been studied from a nutritional perspective (Reiter et al., 2005).
29.5 CONCLUDING REMARKS: IMPLICATIONS OF THE PRESENCE OF MELATONIN IN OLIVE OIL HEALTH AND DISEASE PREVENTION Recent studies demonstrate that dietary combinations of phytochemicals show enhancing health benefits due to their additive and synergistic effects (Jacobs and Steffen, 2003). It is suggested that melatonin and other new bioactive compounds previously found in other physiological contexts account for the positive effects of dietary Mediterranean habits. The parallels between some of the reported positive effects ascribed to melatonin and to olive oil are quite remarkable. Thus, exogenous melatonin has been reported to have a clear immunomodulatory role in different in vivo paradigms (see the following review on melatonin for further details to this matter) (Carrillo-Vico et al., 2006; Szczepanik, 2007). Most studies published on this subject have confirmed that melatonin administration promotes a clear immunoenhancement in terms of immune tissue morphology. For example, melatonin causes an increase in weight of thymus and spleen of several rodents, both
Lipids, Phenolics and Other Organics and Volatiles
under basal and immunosuppressive conditions. Melatonin administration also increases the proliferate capacity of mouse splenocytes and rat lymphocytes. Moreover, melatonin affects non-specific responses that promote an increase in the number of natural killer (NK) cells and monocytes in the bone marrow as well as in antibodydependent cellular cytotoxicity. In humans, a melatonininduced enhancement of NK activity has been described (Carrillo-Vico et al., 2006; Szczepanik, 2007). Regarding the humoral response, pineal extracts increase both the number of antibody-forming cells generated and the response against sheep red blood cell immunization, in mice spleen. Administration of melatonin to birds also induces a significant increase in the humoral immune responses without prior immunosuppression. An additional function of melatonin in the immune system is the modulation of several immune mediators via regulation of gene expression and production when administered in vivo. Melatonin also participates in the apoptosis regulation of T and B cells (Radogna et al., 2006). Overall, these data show that situations in which the immunostimulatory effects of melatonin are best demonstrated are those in which the immune system is depressed (Carrillo-Vico et al., 2006; Szczepanik, 2007). Mechanistically, Maestroni et al. and Pandi-Perumal et al. have postulated that all these melatonin effects may be mediated by an opiatergic mechanism, since the use of naltrexone, a specific opioid antagonist, prevented the melatonin immunoenhancing properties, while beta-endorphin and dynorphin mimicked melatonin effects (Maestroni, 2001; Pandi-Perumal et al., 2006). These works show an in vivo immunoenhancing action of melatonin which seems most pronounced in those situations in which the immune system is depressed and/or when melatonin is administered in the late afternoon or evening. In the last decade, melatonin has been shown to improve the survival of mice and rats from a lethal dose of lipopolysaccharide through the inhibition of pro-inflammatory factors such as cytokines and nitric oxide as well as decreased lipid peroxidation levels and apoptosis (Carrillo-Vico et al., 2006; Szczepanik, 2007). A clinical study has shown the relation between abnormalities in the circadian melatonin secretion in septic patients and the presence of severe sepsis (Mundigler et al., 2002). In newborn infants as well, melatonin has been shown to improve clinical outcome and prevent death due to septic shock (Gitto et al., 2001). Over the last 15 years, several studies have shown that the concomitant administration of melatonin with interleukin-2 (IL-2) amplified the lymphocytosis associated with antitumoral efficiency of IL-2 in several kinds of tumors (see the following review on melatonin for further details to this matter) (Carrillo-Vico et al., 2006; Lissoni et al., 2008). Moreover, the simultaneous administration of melatonin enhances the lymphocytosis induced by the IL-2/IL-12 combination and reduces thrombocytopenia
CHAPTER | 29 The Occurrence of the Biogenic Amine Melatonin in Olive Oil
(toxicity) levels. Several reports have also suggested that melatonin modulates the biological activity and toxicity of tumor necrosis factor-α (TNF-α), another important antitumor cytokine (Szczepanik, 2007; Lissoni et al., 2008). Melatonin therapy has also been shown to induce a decrease in IL-6 levels in patients with advanced solid tumors, which was associated with an improvement in their general well-being; it also induces a reduction in TNF-α serum levels (Lissoni et al., 2008). One of the mechanisms that tumors use for evading the immune system is the production of factors which suppress immune Th1 responsemediated cell immunity against tumor cells, promoting a Th2 response. Melatonin could counteract this Th2 effect, since it increases IL-12 production by monocytes driving T cell differentiation towards the Th1 phenotype and causing an increase of interferon-γ production (Carrillo-Vico et al., 2006; Szczepanik, 2007). Further research is needed to address whether the reported benefits of olive oil could, in part, relate to its content of the phytochemical melatonin as well as other elements contained in the Mediterranean diet. If so, it would be an additional argument to explain the reported health benefits of the Mediterranean diet which is rich in olive oil.
ACKNOWLEDGMENTS This research was supported in part by research grants from ICAI-Madrid (P1264/2005 to D.P), Consejería de Innovación-Universidad de Sevilla, Junta de Andalucía (P957/2005 to D.P), and PAI-BIO323 (to D.P). The authors state no conflict of interest. The author apologizes to other colleagues whose contributions could not be cited here because of space limitations.
REFERENCES Alexandratos, N., 2006. The Mediterranean diet in a world context. Public Health Nutr. 9, 111–117. Bogani, P., Galli, C., Villa, M., Visioli, F., 2006. Postprandial anti-inflammatory and antioxidant effects of extra virgin olive oil. Atherosclerosis 190, 181–186. Carrillo-Vico, A., Reiter, R.J., Lardone, P.J., Herrera, J.L., FernandezMontesinos, R., Guerrero, J.M., Pozo, D., 2006. The modulatory role of melatonin on immune responsiveness. Curr. Opin. Investig. Drugs 7, 423–431. Chen, G., Huo, Y., Tan, D.X., Liang, Z., Zhang, W., Zhang, Y., 2003. Melatonin in Chinese medicinal herbs. Life Sci. 73, 19–26. Cheung, R.T., Tipoe, G.L., Tam, S., Ma, E.S., Zou, L.Y., Chan, P.S., 2006. Preclinical evaluation of pharmacokinetics and safety of melatonin in propylene glycol for intravenous administration. J. Pineal. Res. 41, 337–343. De la Puerta, C., Carrascosa-Salmoral, M.P., Garcia-Luna, P.P., Lardone, P.J., Herrera, J.L., Fernandez-Montesinos, R., Guerrero, J.M., Pozo, D., 2007. Melatonin is a phytochemical in olive oil. Food Chem. 104, 609–612. De la Puerta, R., Martinez-Dominguez, E., Ruiz-Gutierrez, V., 2000. Effect of minor components of virgin olive oil on topical antiinflammatory assays. Z. Naturforsch. 55, 814–819.
281
Dubbels, R., Reiter, R.J., Klenke, E., Goebel, A., Schnakenberg, E., Ehlers, C., Schiwara, H.W., Schloot, W., 1995. Melatonin in edible plants identified by radioimmunoassay and by high performance liquid chromatography-mass spectrometry. J. Pineal. Res. 18, 28–31. Gitto, E., Tan, D.X., Reiter, R.J., Karbownik, M., Manchester, L.C., Cuzzocrea, S., Fulia, F., Barberi, I., 2001. Individual and synergistic antioxidative actions of melatonin: studies with vitamin E, vitamin C, glutathione and desferrioxamine (desferoxamine) in rat liver homogenates. J. Pharm. Pharmacol. 53, 1393–1401. Guillaume, J.L., Daulat, A.M., Maurice, P., Levoye, A., Migaud, M., Brydon, L., Malpaux, B., Borg-Capra, C., Jockers, R., 2008. The PDZ protein Mupp1 promotes Gi coupling and signaling of the Mt1 melatonin receptor. J. Biol. Chem. 283, 16762–16771. Hattori, A., Migitaka, H., Iigo, M., Itoh, M., Yamamoto, K., OhtaniKaneko, R., Hara, M., Suzuki, T., Reiter, R.J., 1995. Identification of melatonin in plants and its effects on plasma melatonin levels and binding to melatonin receptors in vertebrates. Biochem. Mol. Biol. Int. 35, 627–634. Hernandez-Ruiz, J., Cano, A., Arnao, M.B., 2004. Melatonin: a growthstimulating compound present in lupin tissues. Planta 220, 140–144. Hernandez-Ruiz, J., Cano, A., Arnao, M.B., 2005. Melatonin acts as a growth-stimulating compound in some monocot species. J. Pineal Res. 39, 137–142. Jacobs Jr., D.R., Steffen, L.M., 2003. Nutrients, foods, and dietary patterns as exposures in research: a framework for food synergy. Am. J. Clin. Nutr. 78, 508–513. Kolar, J., Johnson, C.H., Machackova, I., 1999. Presence and possible role of melatonin in a short-day flowering plant, Chenopodium rubrum. Adv. Exp. Med. Biol. 460, 391–393. Kolar, J., Machackova, I., 2005. Melatonin in higher plants: occurrence and possible functions. J. Pineal. Res. 39, 333–341. Lei, X.Y., Zhu, R.Y., Zhang, G.Y., Dai, Y.R., 2004. Attenuation of coldinduced apoptosis by exogenous melatonin in carrot suspension cells: the possible involvement of polyamines. J. Pineal. Res. 36, 126–131. Leon-Blanco, M.M., Guerrero, J.M., Reiter, R.J., Calvo, J.R., Pozo, D., 2003. Melatonin inhibits telomerase activity in the MCF-7 tumor cell line both in vivo and in vitro. J. Pineal. Res. 35, 204–211. Leon-Blanco, M.M., Guerrero, J.M., Reiter, R.J., Pozo, D., 2004. RNA expression of human telomerase subunits TR and TERT is differentially affected by melatonin receptor agonists in the MCF-7 tumor cell line. Cancer Lett. 216, 73–80. Lerner, A.B., Case, J.D., Takahasi, Y., Lee, T.H., Mori, W., 1958. Isolation of melatonin, the pineal gland factor that lightens melanocytes. J. Am. Chem. Soc. 80, 2587. Levoye, A., Dam, J., Ayoub, M.A., Guillaume, J.L., Couturier, C., Delagrange, P., Jockers, R., 2006. The orphan GPR50 receptor specifically inhibits MT1 melatonin receptor function through heterodimerization. EMBO J. 25, 3012–3023. Lissoni, P., Brivio, F., Fumagalli, L., Messina, G., Vigore, L., Parolini, D., Colciago, M., Rovelli, F., 2008. Neuroimmunomodulation in medical oncology: application of psychoneuroimmunology with subcutaneous low-dose IL-2 and the pineal hormone melatonin in patients with untreatable metastatic solid tumors. Anticancer Res. 28, 1377–1381. Maestroni, G.J., 2001. The immunotherapeutic potential of melatonin. Expert. Opin. Investig. Drugs 10, 467–476. Manchester, L.C., Tan, D.X., Reiter, R.J., Park, W., Monis, K., Qi, W., 2000. High levels of melatonin in the seeds of edible plants: possible function in germ tissue protection. Life Sci. 67, 3023–3029. Martinez-Cruz, F., Pozo, D., Osuna, C., Espinar, A., Marchante, C., Guerrero, J.M., 2002. Oxidative stress induced by phenylketonuria in
282
SECTION | I
the rat: Prevention by melatonin, vitamin E, and vitamin C. J. Neurosci. Res. 69, 550–558. Miles, E.A., Zoubouli, P., Calder, P.C., 2005. Differential anti-inflammatory effects of phenolic compounds from extra virgin olive oil identified in human whole blood cultures. Nutrition 21, 389–394. Mundigler, G., Delle-Karth, G., Koreny, M., Zehetgruber, M., SteindlMunda, P., Marktl, W., Ferti, L., Siostrzonek, P., 2002. Impaired circadian rhythm of melatonin secretion in sedated critically ill patients with severe sepsis. Crit. Care Med. 30, 536–540. Murch, S.J., Simmons, C.B., Saxena, P.K., 1997. Melatonin in feverfew and other medicinal plants. Lancet 350, 1598–1599. Nagata, C., Nagao, Y., Shibuya, C., Kashiki, Y., Shimizu, H., 2005. Association of vegetable intake with urinary 6-sulfatoxymelatonin level. Cancer Epidemiol. Biomarkers Prev. 14, 1333–1335. Pandi-Perumal, S.R., Srinivasan, V., Maestroni, G.J., Cardinali, D.P., Poeggeler, B., Hardeland, R., 2006. Melatonin. FEBS J. 273, 2813–2838. Perez-Jimenez, F., 2005. International conference on the healthy effect of virgin olive oil. Eur. J. Clin. Invest. 35, 421–424. Perona, J.S., Cabello-Moruno, R., Ruiz-Gutierrez, V., 2006. The role of virgin olive oil components in the modulation of endothelial function. J. Nutr. Biochem. 17, 429–445. Pozo, D., Delgado, M., Fernandez-Santos, J.M., Calvo, J.R., Gomariz, R.P., Martin-Lacave, I., Ortiz, G.G., Guerrero, J.M., 1997. Expression of the Mel1a-melatonin receptor mRNA in T and B subsets of lymphocytes from rat thymus and spleen. FASEB J. 11, 466–473. Pozo, D., García-Mauriño, S., Guerrero, J.M., Calvo, J.R., 2004. mRNA expression of nuclear receptor RZR/ROR, melatonin membrane receptor MT1, and hydroxyndole-O-methyltransferase in different populations of human immune cells. J. Pineal. Res. 37, 48–54. Radogna, F., Paternoster, L., Albertini, M.C., Accorsi, A., Cerella, C., D’Alessio, M., De Nicola, M., Nuccitelli, S., Magrini, A., Bergamaschi, A., Ghibelli, L., 2006. Melatonin as an apoptosis antagonist. Ann. N.Y. Acad. Sci. 1090, 226–233. Reiter, R.J., Manchester, L.C., Tan, D.X., 2005. Melatonin in walnuts: influence on levels of melatonin and total antioxidant capacity of blood. Nutrition 21, 920–924.
Lipids, Phenolics and Other Organics and Volatiles
Reiter, R.J., Tan, D.X., 2002. Melatonin: an antioxidant in edible plants. Ann. N.Y. Acad. Sci. 957, 341–344. Reiter, R.J., Tan, D.X., Burkhardt, S., Manchester, L.C., 2001. Melatonin in plants. Nutr. Rev. 59, 286–290. Reiter, R.J., Tan, D.X., Korkmaz, A., Erren, T.C., Piekarski, C., Tamura, H., Manchester, L.C., 2007a. Light at night, chronodisruption, melatonin suppression, and cancer risk: a review. Crit. Rev. Oncog. 13, 303–328. Reiter, R.J., Tan, D.X., Manchester, L.C., Pilar Terron, M., Flores, L.J., Koppisepi, S., 2007b. Medical implications of melatonin: receptormediated and receptor-independent actions. Adv. Med. Sci. 52, 11–28. Simopoulos, A.P., 2001. The Mediterranean diets: What is so special about the diet of Greece? The scientific evidence. J. Nutr. 131, 3065–3073. Simopoulos, A.P., Salem Jr. N., 1986. Purslane: a terrestrial source of omega-3 fatty acids. N. Engl. J. Med. 315, 833. Simopoulos, A.P., Tan, D.X., Manchester, L.C., Reiter, R.J., 2005. Purslane: a plant source of omega-3 fatty acids and melatonin. J. Pineal Res. 39, 331–332. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food: epidemiology and nutritional approaches. Nutr. Rev. 60, 170–176. Szczepanik, M., 2007. Melatonin and its influence on immune system. J. Physiol. Pharmacol. 58, 115–124. Tan, D.X., Manchester, L.C., Di Mascio, P., Martinez, G.R., Prado, F.M., Reiter, R.J., 2007a. Novel rhythms of N1-acetyl-N2-formyl-5methoxykynuramine and its precursor melatonin in water hyacinth: importance for phytoremediation. FASEB J. 21, 1724–1729. Tan, D.X., Manchester, L.C., Terron, M.P., Flores, L.J., Reiter, R.J., 2007b. One molecule, many derivatives: a never-ending interaction of melatonin with reactive oxygen and nitrogen species? J. Pineal. Res. 42, 28–42. Van Tassel, D.L., Roberts, N., Lewy, A., O’Neill, S.D., 2001. Melatonin in plant organs. J. Pineal Res. 31, 8–15. von Gall, C., Weaver, D.R., 2008. Loss of responsiveness to melatonin in the aging mouse suprachiasmatic nucleus. Neurobiol. Aging 29, 464–470. Wahle, K.W., Caruso, D., Ochoa, J.J., Quiles, J.L., 2004. Olive oil and modulation of cell signaling in disease prevention. Lipids 39, 1223–1231.
Chapter 30
Olive Biophenols as Food Supplements and Additives Antonella De Leonardis and Vincenzo Macciola Department of Agricultural, Food, Environmental and Microbiological Science and Technology (DiSTAAM), Campobasso, Italy
30.1 INTRODUCTION Food, cosmetic and pharmaceutical industries are drawn together to promote new products, named ‘functional food’ (food with scientifically recognized health and therapeutic properties), food supplements, nutraceuticals or cosmeceuticals (Peschel et al., 2006). Functional food products meet the consumer’s demand for a healthy lifestyle. These foods are not intended only to satisfy hunger and provide humans with necessary nutrients, but also to prevent nutrition-related diseases and increase physical and mental well-being of consumers. Besides, the modern customers, especially those living in the industrial countries, demand ‘non-chemical’ ingredients in health products. The most important and dynamic market of the functional foods represent USA and Japan (Menrad, 2003). Also in Europe, the functional foods market is increasing, especially in Germany, France, the United Kingdom and the Netherlands. Since ancient times, in the Mediterranean countries, most of the plant parts of Olea europaea have been used as an effective medical treatment in traditional medicine and the principal common uses are listed in Table 30.1 (Yaseen Khan et al., 2007). In recent decades, the positive effects of some olive phenols on human health have been scientifically demonstrated numerous times. The olive phenols with biological activity are also called ‘biophenols’. Table olives and extra virgin olive oil, both typical ethnic products of the Mediterranean food culture, are certainly the principal food source of olive biophenols (Uccella, 2001). Nevertheless, the demand of olive biomolecules is quickly increasing across the world, including in countries far-off in the Mediterranean area. In fact, the American calibrated Food Guide Pyramid (USDA, 2000) also suggests eating foods containing olive biophenols, which combat dangerous free radicals and prevent the accumulation of cholesterol. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
TABLE 30.1 Traditional medical uses of plant parts of Olea europaea (Yaseen Khan et al., 2007). ●
Oil with lemon juice to treat gallstones
●
Leaves taken orally for stomach and intestinal pathologies
●
Hot water extract of fresh leaves is taken orally to treat hypertension, for bronchial asthma and to induce diuresis
●
Seed oil is taken orally as a cholagogue and laxative; it is also applied externally as an emollient and pectoral and to prevent hair loss
●
Decoction of dried leaves is taken orally for diabetes
●
Tincture of leaves is taken orally as a febrifuge
●
Fruit is applied externally to fractured limb and as a skin cleanser
●
Hot water extract of dried plant is taken
●
Infusion of the fresh leaf is taken orally as an anti-inflammatory
Nowadays, besides the traditional olive foods (table olives and olive oil), a lot of olive biophenol formulations are already available commercially. Some recent patents relative to the extraction techniques and potential applications of olive biophenols are listed in Table 30.2. In the industrial process, olive biophenols are extracted from different olive-based starting materials, including olives, olive pulps, olive oil, olive oil mill wastewaters and finally, olive leaves. Olive oil mill wastewaters, but also olive leaves, are certainly the most convenient and environmentally friendly
283
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
284
SECTION | I
Lipids, Phenolics and Other Organics and Volatiles
TABLE 30.2 Some recent patents relative to the production and utilization of olive biophenols. Patent number
Title
US Patent 6358542/2002
Antioxidant compositions extracted from olives and olive byproducts
US Patent 5714150/1998
Method for producing extract of olive leaves
US Patent 6162480/2000
Fortification of a vegetable fat with antioxidants
EP1790234/2007
Natural antioxidant additive for feed and drinking water
WO/1999/038383
Method for producing extract of olive leaves and extract produced thereby
WO/1999/032589
Olive oil containing food composition
WO/1999/055349
Substance mixture for topical application comprising olive oil and honey
WO/2000/038541
Fortification of food products with olive fruit ingredients
WO/2001/076579
Treatment of skin damage using olive oil polyphenols
WO/2002/032387
Use of olive oil in the preparation of a product for oral hygiene for eliminating or reducing bacterial plaque and/or bacteria in the mouth
WO/2003/079794
Use of olive tree extracts in detergents, rinsing agents and cleaning agents
WO/2004/110171
Olive powder
WO/2004/005228
An hydroxytyrosol-rich composition from olive vegetation water and method of use thereof
WO/2005/123603
Process for recovering the components of olive mill wastewater with membrane technologies
WO/2006/020588
Olive compositions and methods for treating inflammatory conditions
WO/2006/043117
Processing of olive marc into alimentary meal
WO/2006/005986
Olive polyphenols concentrate
WO/2007/013032
Method of obtaining a natural hydroxytyrosol-rich concentrate from olive tree residues and subproducts using clean technologies
WO/2007/042742
Cured olive powder
WO/2007/074490
Process for producing triacetyl-hydroxytyrosol from olive oil mill wastewaters for use as stabilized antioxidant
WO/2007/096446
Use of olive extract as a pronutrient in animal feed
WO/2008/040550
Olive juice extracts for promoting muscle health
raw material to produce bio-technologically olive biophenols. In fact, owing to their potential health benefits, olive biophenols represent an opportunity to add value to the waste products that are an environmental problem in oliveproducing countries (Schieber et al., 2001). The principal ‘recovery-systems’ are based on the following techniques: extraction with solvents; resin chromatography; selective concentration by ultrafiltration and inverse osmosis; solid–liquid or liquid–liquid extraction;
supercritical fluid extraction (Vásquez et al., 1987; Capasso et al., 1999; Fernández-Bolanˇos et al., 2002; Agalias et al., 2007; Japón-Luján and de Castro, 2007). The olive biophenols, that are generally present in the commercial formulates, can be grouped as follows: ●
●
oleuropeosides (oleuropein, demethyloleuropein, ligustroside, verbascoside) flavonoids
CHAPTER | 30 Olive Biophenols as Food Supplements and Additives
●
●
simple phenols (tyrosol, hydroxytyrosol, vanilic, p-coumaric, ferulic and caffeic acids, catechol) others (elenolic acid).
The composition of commercial formulates is extremely variable in relation to the olive-based starting materials and the extraction techniques. Specific technologies, e.g. supercritical fluid extraction, nanofiltration and reverse osmosis, are used to concentrate singular phenols, especially oleuropein and hydroxytyrosol. The olive biophenol extracts can be used in liquid form or in powder manufactured by spray drying or selective absorption. Olive biophenols can be formulated into dietary supplements, food, beverages, cosmetics and pharmaceutical products and health fortificant for feed. One qualitative system to discriminate the olive biophenol commercial products could be the percentage and the purity grade of the free hydroxytyrosol (HT). In nature, native HT is rarely found free in common foods, with the exception of ripened olives, in which it is present in quantities ranging from 1.0 to 3.0 g per 100 g of dried weight. On the contrary, free HT is present in considerable amounts in olive mill solid–liquid wastes from two-phase olive oil processing (Fernández-Bolanˇos et al., 2002), as well as in olive-mill waste waters from traditional and industrial threephase plants (Capasso et al., 1999; Allouche et al., 2004). The natural mechanism that occurs when the olive tree forms free HT is enzymatic hydrolysis, and specific native β-glycosidase and esterase are implicated. On the contrary, in the industrial processes, acid or alkaline hydrolysis is generally the mechanism used to produce free HT. Finally, with the aim to produce a more stable and effective antioxidant ingredient, HT can also be transformed into triacetylhydroxytyrosol (Capasso et al., 2006). The principal health-promoting activities that are claimed for olive biophenol commercial products (supported on the bases of a copious scientific literature) are listed in Table 30.3. In particular, the antioxidative, antimicrobial and antivirus activities are certainly the most important properties of olive biophenols that are advertised throughout the marketing strategy.
30.2 OLIVE BIOPHENOLS FROM OLIVE LEAVES The most popular olive phenol dietary supplements are those produced from olive leaves. Commercial formulates of olive leaves are as follows: ● ●
●
dried olive leaves olive leaf extracts as gel capsule or tablets with phenol standardized composition liquid olive leaf extracts.
The dried olive leaves are used to make a beverage similar to tea.
285
TABLE 30.3 The principal health-promoting activities that are claimed for olive biophenol commercial products. Protection of human erythrocytes against oxidative damage Cardioprotective and antiatherogenic Scavenges and reduces superoxide anion production in human promonocyte cells Inhibition of the proliferation of tumor cells Antimicrobial human pathogens Anti-inflammatory prostaglandin sparing and antioxidant Inhibition of leukocytes leukotriene B4 Radical scavenging activity within biomembrane Inhibition of LDL oxidation and platelets aggregation Hypoglycemic activity Antihypertensive vasodilator Antimicrobial and antiviral activity Anti-HIV activity
The olive leaf extracts contain several different types of phytochemicals, but generally oleuropein is the most abundant, comprising 20–40% of total components (De Leonardis et al., 2008).
30.3 OLIVE PHENOLS FROM OLIVE-OIL MILL BYPRODUCTS During the process of olive oil extraction, the native olive phenols migrate into the fat phase or aqueous phase in function of their lipophilic/hydrophilic nature. At the end of the process, only about 200 mg kg⫺1 of total phenols could be found in the virgin olive oil, 5000 mg kg⫺1 in the wastewaters and about 10 000 mg kg⫺1 in the solid waste (Rodis et al., 2002). The phenol substances found in olive oil mill wastewaters (OMWW) are different from those of olive fruits. In fact, olives are very rich in secoiridoid glucosides, while OMWW actually contain secoiridoid derivates, especially hydroxytyrosol and dialdehydic form of decarboxymethyl oleuropein aglycon (Macciola and De Leonardis, 2006; De Marco et al., 2007). The phenol composition of raw material could vary with the cultivars, ripening status and quality of the olives and with the industrial ‘oil-extraction’ technology, but also with the storage conditions of the same wastes (Feki et al., 2005).
286
SECTION | I
Consequently, characteristics of the raw material and extraction techniques have a significant effect on the composition of the final extracts, and so in the industrial process, the olive phenol extract composition must be standardized. Finally, the oil-mill byproducts phenol extracts are generally available in the market as powder, gel capsules or tablets.
30.4 OLIVE BIOPHENOLS DIRECTLY FROM OLIVES Sometimes, a fine dried powder can be directly obtained from olives. Generally, green olives are used because they are richer in phenol substances. The solid matter obtained from olive fruits has variable particle sizes, ranging from 0.1 μm to 5.0 mm. Olive fruit powder can not be de-bittered or can be de-bittered, by a treatment with sodium hydroxide solution, before it is used as an ingredient in other foods.
30.5 OLIVE BIOPHENOLS AS FOOD ANTIOXIDANT ADDITIVE It is well known that olive phenols have antioxidant properties, so they could be useful in improving the oxidative stability of different edible fats as an alternative to synthetic compounds. However, not all the olive phenol substances are effective as food antioxidants and especially, concentration of free HT is demonstrated to be fundamental in improving the antioxidant effectiveness of olive extracts. The graph of Figure 30.1 shows the results of laboratory tests in which the antioxidant effectiveness of four
Induction time (minutes)
4 3.5
IT control IT fortified PF
3 2.6 2.2
2
1.8
1
0 E1. (TP: 7.0 g/L; HT/TP: 1.1%)
E2. (TP: 2.5 g/L; HT/TP: 4.0%)
E3. (TP: 1.5 g/L; HT/TP: 8.4%)
E4. (TP: 0.8 g/L; HT/TP: 18.8%)
Phenol extracts from olive oil mill wastewaters
FIGURE 30.1 Antioxidant effectiveness of different olive oil mill wastewater extracts measured on lard by Rancimat test (under 120°C temperature and 20 L h⫺1 air flow). IT, induction time; PF, protection factor; TP, total phenols; HT/TP, free hydroxytyrosol on total phenols.
Lipids, Phenolics and Other Organics and Volatiles
different olive oil mill wastewater phenol extracts (E1., E2., E3., E4.) are measured on lard by using a Rancimat apparatus, under operative conditions of 120°C temperature and 20 L h⫺1 air flow (De Leonardis et al., 2009). Antioxidant effectiveness has been calculated on the protection factor (PF) that is the percentage ratio of the fat induction times, with and without each OMWW phenol extract. Specifically, the OMWW extracts were different for content and composition of phenols. Total phenols were measured by the colorimetric Folin-Ciocalteu methods (Gutfinger, 1981), while phenol composition was determined by HPLC analysis. It has clearly emerged that antioxidant effectiveness is scarcely correlated to the values of total phenols, while it is strongly correlated with the percentage of free HT (Figure 30.1). In conclusion, antioxidant effectiveness of an olive phenol extract is affected more from its phenol composition, than from its total phenol contents. Without doubt, Folin-Ciocalteu’s reagent reacts also with several other compounds, especially condensed phenols, that have no antioxidant properties. HT is the best antioxidant component among the OMWW olive phenols. Nevertheless, native content of HT into the OMWW could be extremely variable; in addition, it is only partially in free form. Therefore, adequate and standardized processes are needed to produce a good antioxidant extract, preferably enriched with major amounts of free hydroxytyrosol. In Table 30.4, the antioxidant effectiveness of an aqueous HT-rich olive leaf extract (at least 92% of HT on total phenols) on different edible lipids is shown (De Leonardis et al., 2008). In particular, the antioxidant effect has been evaluated on butter, lard and cod liver oil, by using the Rancimat apparatus and a dosage of 50 mg kg⫺1 of HT on fat. For each lipid sample, HT effectiveness has been measured on the protection factor (PF). In the above-mentioned experiment, HT confirmed to be a good antioxidant for food lipids and in fact, there has been a significant increase in the induction time in all the lipid samples fortified with the HT extract, if compared with the relative control. More specifically, at 120°C, the induction times were 10.7 and 3.6 times higher than those of the control in lard and in butter, respectively; for the cod liver oil at 100°C the induction time was 1.7 times than that of the control. In the specific case of the lard, the HT-rich extract can delay the oxidative stability by at least 7–15 times, when under low and high temperature (De Leonardis et al., 2007). Such treatments comprising crushing, freezing and thawing, sonication, heat shock, drying and high-pressure shock, can simplify the fortification of the edible fats with the hydrophilic HT. Lard or other edible fats enriched with olive phenols, especially HT, have the following advantages: improved oxidative stability and increased nutritional value.
287
CHAPTER | 30 Olive Biophenols as Food Supplements and Additives
TABLE 30.4 Oxidation stability, measured by Rancimat test, of different food lipids fortified with and without hydroxytyrosol-rich olive leaf extract. Operative temperature °C
120
120
100
Edible lipids
Lard
Butter
Cod liver oil
Determinations
IT
Control
0.41
Proof with 50 mg kg⫺1 of HT extract
4.37
PF
IT
PF
5.02 10.7
18.04
IT
PF
1.80 3.6
3.07
1.7
IT: induction time (minutes); PF: protection factor.
30.6 OLIVE BIOPHENOLS AS FUNCTIONAL FOOD Nutritional value of a food product, particularly vegetable oils, spreads, mayonnaises, salad dressings and sauces, can be enhanced by the addition of olive biophenol ingredients. In recent years, new food formulations, derived from the combination of nutraceutical compounds and probiotic microorganisms, are setting quite a trend. Indeed, co-cultures of Streptococcus thermophilus and Lactobacillus delbruechii ssp. bulgaricus are generally used in yogurt and similar fermented milk. The effects of olive leaf extracts at high content of free HT (at least 92% on total phenols) has been observed on the growth of co-cultures of Streptococcus thermophilus and Lactobacillus delbruechii ssp. bulgaricus, at very variable concentrations, ranging from 25 to 3200 μg mL⫺1 (as HT) (De Leonardis et al., 2008). At all the tested concentrations, the antimicrobial activities of olive leaf extract showed no growth inhibition and minimum antimicrobial activity (MIC) was a value ⬎3200 μg mL⫺1 of HT. In conclusion, olive phenols, especially HT, can be added as an integrator or antioxidant to fermented milk, to increase both the quality and the nutritional value of the final milk product, without inducing any negative effects on the viability of the lactic acid bacteria.
30.7 HEALTH FORTIFICANT FOR FEED A recent return to ‘natural medicine’ to treat animals is also an increasing trend (Viegi et al., 2003). The use of antibiotic or chemotherapeutic substances, referred to as ‘growth promoters’, in the production of livestock is well known. In recent years, attention has increasingly focused on problems related to the widespread use of antibiotics or chemotherapeutics as ‘growth promoters’. As a consequence, in various countries a complete cessation of the use of classical growth promoters in swine production
has been imposed. New alternative methods are sought to ensure the livestock’s growth conditions, free of both disease and use of antibiotics. Olive biophenols, especially those extracted from leaves, can be used as an antioxidant into products for feed for domestic animals. Olive phenols in feed (5% in weight) improve the productivity and health status in animals by having antimicrobial and antiviral properties, improving feed conversion, the utilization of nutrients and the health status of the animals. Olive biophenols can be used for monogastric animals and ruminants but also for fish, crustaceans, and pets.
30.8 FINAL SAFETY CONSIDERATION The claimed effects of olive biophenols are often studied in vitro, while in the same way they are not adequately welldocumented in humans. Indeed, several questions are still open about the correct use of olive biophenol (Pokorný, 2007). The increased use of fortified foods and functional supplements has increased the intake of nutrient substances around the world. In turn, there has been growing interest on an international basis for determining the levels of intake that may pose risk. ‘Habitual intake’ is the long-term average daily intake of the nutrient substance. The upper level of intake is the maximum level of ‘habitual intake’ from all sources of a nutrient or related substance, judged to be unlikely to lead to adverse health effects in humans. It is approximately estimated that the average daily intake of native antioxidants present in the diet consisting of normal food is totally about 1000 mg, of which about 50–200 mg are polyphenols (Murkovic, 2003). In the case of olive biophenols, the safety and physiological limits are mostly not known. Theoretically, an adequate intake of olive phenol supplements should be determined on the quantity of extra virgin olive oil eaten usually by Mediterranean people. By considering a typical
288
SECTION | I
Mediterranean diet, the daily quantity eaten of extra virgin olive oil, containing presumably 200 mg kg⫺1 of olive phenols, is about 30 g, so the calculated daily intake of olive phenols should be about 6 mg. Another nutritional doubt is the very different composition between the olive oil phenols and that of the commercial olive biophenol formulates. The greatest part of natural phenols often shows, in vitro, a strong inhibitor effect on cell proliferation. It is known that the degree of toxicity of a phenol varies in relation to its concentration but also with the chemical structure. In human cells, the cytotoxicity of several phenols typically present in virgin olive oil, is in the following order of toxicity: oleuropein aglycone ⬎ oleuropein glycoside ⬎ caffeic acid ⬎ o-coumaric acid ⬎ cinnamic acid ⬎ tyrosol and syringic ⬎ protocatechuic and vanillic acids (Babich and Visioli, 2003). Specific studies on the cytotoxicity of free HT are not numerous. However, it is known that at high concentrations (higher than 1 mM) olive phenols exert a cytotoxic effect; nevertheless, at lower concentrations increased cell proliferation has been observed more than once (Manna et al., 1997; Owen et al., 2000). Mouse fibroblasts NIH/3T3 and human umbilical vein endothelial cells (HUVEC) have been exposed to an olive phenol extract at the concentrations of 0.01, 0.16, 0.32, 0.64 and 1.28 mM (as hydroxytyrosol) for 12, 24 and 48 h (De Leonardis et al., 2008). The tested cell types gave similar results; cytotoxicity varied with olive phenol extract concentration, and at constant concentration it varied with the stimulation time. At concentrations higher than 0.32 mM, the cytotoxic effect of olive phenol extract emerged on both cell samples in the first hours. At 0.32 mM, the number of stimulated cells equalled those of the control. Finally, at concentrations lower than 0.32 mM there was little antiproliferative activity and during the first 12 h, the cell numbers were higher than in the control in both cases. Therefore, at concentrations less than 0.32 mM, an olive phenol extract can be considered non-cytotoxic. In conclusion, the information on safety and health advantages of olive biophenols is still insufficient. Major specific toxicity tests and nutritional studies are needed to regulate an appropriate use of olive biophenols as antioxidant and health-promoting supplements, ingredients and additives.
SUMMARY POINTS ●
●
Since ancient times, in the Mediterranean countries, most of the plant parts of Olea europaea have been used as effective medical treatments in traditional medicine. Olive phenols with biological activity (also called ‘biophenols’) are sold in several formulations, in liquid form or in powder.
●
●
●
Lipids, Phenolics and Other Organics and Volatiles
In the industrial process, olive biophenols are extracted from different olive-based starting materials, including olives, olive pulps, olive oil, olive oil mill wastewaters, and finally, olive leaves. Composition of the commercial formulates are extremely variable in relation to the olive-based starting materials and the extraction techniques. Olive ‘biophenols’ can also be used as a functional ingredient and antioxidant additive for food and feed.
REFERENCES Agalias, A., Magiatis, P., Skaltsounis, A.L., Mikros, E., Tsarbopoulos, A., Gikas, E., Spanos, I., Manios, T., 2007. A new process for the management of olive oil mill waste water and recovery of natural antioxidants. J. Agric. Food Chem. 55, 2671–2676. Allouche, N., Fki, I., Sayadi, S., 2004. Toward a high yield recovery of antioxidants and purified hydroxytyrosol from olive mill wastewaters. J. Agric. Food Chem. 52, 267–273. Babich, H., Visioli, F., 2003. In vitro cytotoxicity to human cells in culture of some phenols from olive oil. Il Farmaco 58, 403–407. Capasso, R., Evidente, A., Avolio, S., Solla, F., 1999. A highly convenient synthesis of hydroxytyrosol and its recovery from agricultural waste waters. J. Agric. Food Chem. 47, 1745–1748. Capasso, R., Sannino, F., De Martino, A., Manna, C., 2006. Production of triacetylhydroxytyrosol from olive mill waste waters for use as stabilized bio-antioxidant. J. Agric. Food Chem. 54, 9063–9070. De Leonardis, A., Macciola, V., Lembo, G., Aretini, A., Nag, A., 2007. Studies on oxidative stabilisation of lard by natural antioxidants recovered from olive-oil mill wastewater. Food Chem. 100, 998–1004. De Leonardis, A., Aretini, A., Alfano, G., Macciola, V., Ranalli, G., 2008. Isolation of a hydroxytyrosol-rich extract from olive leaves (Olea Europaea L.) and evaluation of its antioxidant properties and bioactivity. Eur. Food Res. Technol. 226, 653–659. De Leonardis, A., Macciola, V., Nag, A., 2009. Antioxidant activity of different phenol extracts from olive-oil mill wastewaters. Acta Alim. 38, 77–86. De Marco, E., Savarese, M., Paduano, A., Sacchi, R., 2007. Characterization and fraction of phenolic compounds extracted from olive-oil mill wastewaters. Food Chem. 104, 858–867. Fernández-Bolanˇos, J., Rodríguez, G., Rodríguez, R., Heredia, A., Guillén, R., Jiménez, A., 2002. Production in large quantities of highly purified hydroxytyrosol from liquid–solid waste of two-phase olive oil processing or ‘‘Alperujio’’. J. Agric. Food Chem. 50, 6804–6811. Feki, M., Allouche, N., Bouaziz, M., Gargoubi, A., Sayadi, S., 2005. Effect of storage of olive mill wastewaters on hydroxytyrosol concentration. Eur. J. Lipid Sci. Technol. 107, 497–504. Gutfinger, T., 1981. Phenols in olive oil. J. Am. Oil Chem. Soc. 11, 966–968. Japón-Luján, R., de Castro, M.D.L., 2007. Static-dynamic superheated liquid extraction of hydroxytyrosol and other biophenols from alperujo (a semisolid residue of the olive oil industry). J. Agric. Food Chem. 55, 3629–3643. Macciola, V., De Leonardis, A., 2006. Natura e proprietà antiossidanti di fenoli estratti da un olio extra-vergine d’oliva e suoi sottoprodotti. Ingredienti Alimentari 5, 6–14. Manna, C., Galletti, P., Cucciolla, V., Moltedo, O., Leone, A., Zappia, V., 1997. The protective effect of the olive oil polyphenols (3-4-dyhydroxyphenyl)-ethanol counteracts reactive oxygen metabolite induced cytoxicity in caco-2-cells. J. Nutrit. 127, 286–292.
CHAPTER | 30 Olive Biophenols as Food Supplements and Additives
Menrad, K., 2003. Market and marketing of functional food. J. Food Eng. 56, 181–188. Murkovic, M., 2003. Phenolic compounds. In: Caballero, B., Trugo, C., Finglas, P.M. (eds) Encyclopedia of Food Sciences and Nutrition, 2nd edn. Academic Press, Amsterdam (The Netherlands), pp. 4507–4514. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Spiegelhalder, B., Bartsch, H., 2000. The antioxidant/anticancer potential of phenolic compounds isolated from olive oil. Eur. J. Cancer 36, 1235–1247. Peschel, W., Sànchez-Rabaneda, F., Diekmann, W., Plescher, A., Gartzìa, I., Jimènez, D., Lamuela-Raventçò, R., Buxaderas, S., Codina, C., 2006. An industrial approach in the search of natural antioxidants from vegetable and fruit wastes. Food Chem. 97, 137–150. Pokorný, J., 2007. Are natural antioxidants better-and safer-than synthetic antioxidants? Eur. J. Lipid Sci. Technol. 109, 629–642. Rodis, P.S., Karathanos, V.T., Mantzavinos, A., 2002. Partitioning of olive oil antioxidants between oil and water phases. J. Agric. Food Chem. 50, 596–601.
289
Schieber, A., Stintzing, F.C., Carle, R., 2001. By-products of plant food processing as a source of functional compounds–recent developments. Trend Food Sci. Tech. 12, 401–413. Yaseen Khan, M.D., Panchal, S., Vyas, N., Butani, A., Kumar, V., 2007. Olea europaea: a phyto-pharmacological review. Pharmacognosy Reviews 1, 114–118. Uccella, N., 2001. Olive biophenols: novel ethnic and technological approach. Trend Food Sci. Tech. 11, 328–339. USDA, 2000. ‘The Food Guide Pyramid’ in Center for Nutrition Policy and Promotion. US Department of Agriculture, Washington, DC. Vásquez, R., Maestro Durán, R., Graciani, C., 1987. Phenols components of olive oil mill waste water. Grasas y Aceites 25, 341–345. Viegi, L., Pieroni, A., Guarrera, P.M., Vangelisti, R., 2003. A review of plants used in folk veterinary medicine in Italy as basis for a databank. J. Ethnopharmacol. 89, 221–244.
This page intentionally left blank
Chapter 31
Chemical Composition of Fermented Green Olives: Acidity, Salt, Moisture, Fat, Protein, Ash, Fiber, Sugar, and Polyphenol Alfredo Montaño, Antonio Higinio Sánchez, Antonio López-López, Antonio de Castro and Luis Rejano Food Biotechnology Department, Instituto de la Grasa (CSIC), Seville (Spain)
31.1 INTRODUCTION The worldwide production of table olives is around 1.8 ⫻ 106 t year⫺1, the largest part of which (37%) comes from the European Union (mainly Spain, Greece, and Italy), followed by Turkey (15%) and Egypt (12%) (Essid, 2008). Fermented green olives are the most popular fermented food in the European Union. Because of its organoleptic properties and its composition, the fermented green olive is identified in many countries with a high standard of living, where it is used as an appetiser with many kinds of alcoholic and non-alcoholic beverages and as a decorative or nutritional element of various dishes. In other countries, the same properties make it a basic food for the majority of people (Fernández-Díez, 1983). Of fermented green olives, alkali-treated green olives in brine, also known as ‘Spanish-style green olives’ or ‘pickled green olives’, are the most widely distributed and investigated type of table olive. However, there are other traditional preparations of fermented green olives which are of lesser economic importance in the international market, but highly appreciated by consumers in the Mediterranean region. One of these involves the direct brining of green olives without prior de-bittering with NaOH solution. This preparation is known as ‘untreated green olives in brine’, ‘naturally green olives’, ‘directly brined olives’, or ‘Sicilian-style green olives’ (Fleming and Moore, 1983; Fernández-Díez et al., 1985). The taste of untreated green olives is completely different from that of alkali-treated fruits, mainly due to the residual bitterness they retain even after a long period of storage in brine. This olive type (green or turning-color) is one of the bases for many other commercial products, such as ‘seasoned’ olives, which are very popular in Spain (Fernández-Díez et al., 1985). When Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
this preparation is made with natural black olives, it is known as Greek-style table olives, and has been extensively studied (Garrido et al., 1997; Tassou et al., 2002; Oliveira et al., 2004; Servili et al., 2006). This chapter reviews the two above-mentioned processing types of fermented green olive – that is, Spanish-style and untreated (green or turning-color) olives – with special emphasis on the characteristics (physicochemical parameters, chemical composition) of the final product. It is introduced with a summary of information in the literature (Fernández-Bolaños et al., 1983; Fernández-Díez, 1983; Amiot et al., 1986; Guillén et al., 1992; Garrido et al., 1997; Visioli and Galli, 1998; Patumi et al., 1999; Marsilio et al., 2001; Romero et al., 2002; Ryan et al., 2002; Bianchi, 2003) regarding the major components in the fresh fruit (raw material). Detailed information on minor components related to the nutritional value of the final product, such as vitamins, amino acids, and fat components, is presented in other chapters of this book.
31.2 MAJOR COMPONENTS OF RAW OLIVES The chemical composition of raw olives depends on several factors. Of these, the cultivar and the state of ripeness at the time of harvest are the most important. The main constituents of the olive flesh are water (60–75%) and lipids (10–25%). There is an inverse relationship between moisture content and lipid (oil) content, for the same degree of ripeness. Most of the lipids are triglycerides, but there are also diglycerides and free fatty acids. Olive oil makes a significant contribution to the fact that table olives are considered a product of high biological and nutritive value.
291
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
292
Soluble reducing and non-reducing sugars (3–6% fresh pulp) are the most important components with regard to the fermentation and preservation stages in all types of table olive processing. The main sugars in the flesh are glucose, followed by fructose, galactose, and mannitol and, at a lower level, sucrose. The distribution of the hexoses varies with the cultivar. During olive ripening, there is continuous decrease of sugars in the flesh. The protein content of the fresh pulp is relatively low, generally between 1% and 3%, and remains almost constant during growth and ripening of the fruits. It seems that the protein content is similar among cultivars and that its amino acid composition does not differ significantly, despite ecological and physiological differences. Crude fiber content varies between 1% and 4% of the fresh pulp, with the major components being cellulose, lignin, and hemicellulose. Phenolic compounds are important for the sensory characteristics of olives, and have significant nutritional, physiological, and pharmaceutical effects on human health. Their levels range from 1% to 3% of the fresh pulp. The composition of the phenolic fraction in olive fruit is very complex, depending on cultivar, development stage, and season. The main polyphenol in raw olives is oleuropein, a glucoside ester of 3,4-dihydroxyphenyl ethanol (hydroxytyrosol) and elenolic acid; its concentration decreases with fruit ripening and olive tree irrigation. Other natural phenols that have been identified in the olive fruit are verbascoside, ligstroside, salidroside, rutin, luteolin-7-glucoside, cyanidin-3-glucoside, and cyanidin-3-rutinoside. Ash content varies from 0.6% to 1% of the fresh pulp, with the major elements being (in decreasing order of importance) K, Ca, P, Na, Mg, and S. Organic acids, with concentrations in raw olives between 0.5% and 1% fresh weight, are of great importance, owing to their buffering capacity during the fermentation stage and subsequent storage. Malic and citric acids are the two major acids in raw olives. The polysaccharides and pectic substances, major constituents of intercellular lamellae, have a cementing function and play an important role in the texture of the olive flesh. Levels of these compounds range between 0.3% and 0.6% of the fresh pulp.
31.3 SPANISH-STYLE GREEN OLIVES A flow diagram as well as detailed information about this method of green olive processing is presented in Chapter 1 of this book. Briefly, this method includes an alkaline treatment with NaOH solution (lye) to chemically remove bitterness, a washing step to eliminate the excess of alkali, and a stage in brine, where the fruits undergo the typical fermentation by lactic acid bacteria (Fernández-Díez et al., 1985).
SECTION | I
Metals, Electrolytes and Other Components
Information on the evolution of the physicochemical and microbiological characteristics, as well as of the main fermentation substrates and end-products, during both spontaneous and controlled Spanish-style green olive fermentation is abundant (Montaño et al., 1993; Garrido et al., 1997; Sánchez et al., 2000, 2001; de Castro et al., 2002; Leal-Sánchez et al., 2003; Panagou and Tassou, 2006). However, there is little information with regard to the quality profile of the product in the market.
31.3.1 Product in Bulk Table 31.1 shows the chemical profile of brines from industrially fermented green olives of different olive cultivars (Manzanilla, Hojiblanca, Gordal) marketed in Spain (Montaño et al., 2003). Ranges in pH, titratable acidity, and combined acidity are higher than those previously reported by Fernández-Díez (1983). The major compounds in brine were lactic acid, acetic acid, ethanol, methanol, formic acid, and succinic acid. Other compounds, such as propanol, 2-butanol, and acetaldehyde, were also detected in small amounts (mean concentration ⬍1 mM). Residual fermentation substrates, namely, glucose, sucrose, mannitol,
TABLE 31.1 Chemical composition of the brines of Spanish-style green olives in bulk (adapted from Montaño et al., 2003). Chemical composition of brine
Range
Meana
pH
3.65–4.40
4.04 (4)
Titratable acidity (% lactic acid)
0.35–1.41
0.93 (18)
Combined acidity (N)
0.06–0.257
0.129 (24)
Salt (% NaCl)
4.0–9.9
6.3 (15)
Lactic acid (mM)
54–245
138 (23)
Acetic acid (mM)
16–83
42 (27)
Ethanol (mM)
4.6–53
21 (49)
Methanol (mM)
0.4–36
19 (35)
Formic acid (mM)
0–28
12 (54)
Succinic acid (mM)
0–9.5
4 (49)
This table shows the physicochemical characteristics and major compounds in brine after fermentation of Spanish-style green olives. Values were obtained from an extensive survey of industrially processed green olives in Spain. a
Relative standard deviation in parentheses. N ⫽ 167.
CHAPTER | 31 Chemical Composition of Fermented Green Olives
and citric acid, were detected in a limited number of samples, where their mean concentrations were 0.8, 0.7, 8.4, and 1.3 mM, respectively. Propionic acid, which is characteristic of growth of Propionibacterium species, was also detected in a limited number of samples, in a concentration range of 0.4–11 mM. Mean values of both the physicochemical characteristics and the major compounds are significantly (p ⬍ 0.05) affected by olive cultivar. Thus, Hojiblanca cultivar usually shows higher values of pH and combined acidity than those in Manzanilla or Gordal cultivars. Mean values for pH, combined acidity, and salt content shown in Table 31.1 are practically identical to those found by Panagou et al. (2006), who examined Spanishstyle green olives marketed in bulk in Greece. However, the mean value for titratable acidity found by those authors was considerably lower (0.53% versus 0.93%), which may be attributed to a lower sugar content of the raw material and/or differences in processing conditions. The proximate composition of the fermented fruit (Gordal, Manzanilla, Hojiblanca, and Verdial cultivars) in bulk compared to that of raw fruit is shown in Table 31.2 (Fernández-Díez, 1983). The following observations can be made: (1) as in the fresh fruit, the moisture and oil contents range widely in the fermented product; (2) sugars are practically absent, due to losses both by solubilization during the lye treatment and washing step and by fermentation; (3) the protein and fiber contents tend to decrease slightly; and (4) the ash content increases as a consequence of the alkaline treatment, fermentation, and storage in brine.
TABLE 31.2 Proximate composition of the pulp of Spanish-style green olives in bulk in comparison with that of raw olives (adapted from Fernández-Díez, 1983). Proximate composition of pulp (g/100 g)
Raw olives
Spanish-style green olives
Water
60–75
61.0–80.6
Fat
10–25
9.1–28.2
Protein
1–3
1.0–1.5
Ash
0.6–1
4.2–5.5
Fiber
1–4
1.4–2.1
Sugars
3 –6
⬍0.1
This table shows the concentration ranges of the major components in olives processed according to the Spanish style in comparison with those found in raw olives. It can be noted that sugars are practically absent from the final product. As a consequence of processing and storage in brine, the percentage of ash increases significantly.
293
Researchers have paid particular attention to olive polyphenols because of their nutritional and sensory properties, and the changes occurring in this class of compounds during processing (Amiot et al., 1990; Brenes et al., 1995; Brenes and de Castro, 1998; Bianchi, 2003). The loss in total polyphenols during Spanish-style processing can reach some 90% (Marsilio et al., 2005). During the alkaline step in Spanish-style green olive processing, oleuropein is hydrolyzed into hydroxytyrosol and elenolic acid glucoside. During the fermentation step, there is a slow acid hydrolysis of the elenolic acid glucoside, with the production of elenolic acid and glucose. Elenolic acid is comparatively unstable, and tends to degrade due to the acid conditions of the medium. The glucose formed is used as substrate by the microorganisms present in the brine and maintains the microbial activity for a longer period of time. At the end, hydroxytyrosol and tyrosol are the major phenols in the fermented product.
31.3.2 Packed Product Prior to packing, green fermented olives must undergo a series of complementary operations (sorting, size-grading, washing, pitting, slicing, stuffing) to adapt them to the different commercial presentations. Stuffing materials are numerous: almond, cucumber, onion, garlic, caper, hazelnut, natural hot pepper, ham, lemon, cheese, formulated red pepper strips, formulated tuna strips, formulated anchovy strips, etc. Table 31.3 shows the ranges and mean values of the main physicochemical characteristics and proximate composition of packed Spanish-style green olives marketed in Spain (López-López et al., 2004, 2007; López-López, 2006; López et al., 2007). As expected, the physicochemical parameters showed lower mean values than those reported in Table 31.1 for the fermented product in bulk. Mean values for pH, titratable acidity, and salt content comply with those established by the International Olive Oil Council (the maximum pH value is established at 4.0 when the product is preserved by its own physicochemical characteristics, or 4.3 when it is preserved by pasteurization; the minimum value for titratable acidity, expressed as lactic acid, is established at 0.5% (w/v) for olives preserved by their own physicochemical characteristics, or 0.4% in the case of using preservatives or refrigeration; the minimum NaCl content is established at 5% if olives are preserved by their own physicochemical characteristics, or 4% in the case of using preservatives or refrigeration) (IOOC, 2004). Combined acidity (or residual lye) is an indicator of the buffer capacity of fermenting brines. As a general approach, the higher the combined acidity, the greater the amount of acid needed to reach a specific pH value. The mean value of combined acidity in Table 31.3 (0.046 N) is higher than that considered adequate for packed Spanish-style green olives (namely, 0.025 N or lower) in order to preserve the product
294
SECTION | I
by its own physicochemical characteristics (Garrido et al., 1997). This indicates that alternative preservation methods, such as pasteurization, are widely used today. Similarly to what is mentioned above for the product in bulk, the contents of moisture and fat in the packed product (Table 31.3) are quite variable, and depend principally on the olive cultivar used as raw material. The protein content in the packed product varies widely – the range (0.7–3.7%) being higher than that for the product in bulk. This is not surprising considering the many types of stuffing material or ingredient currently used in the packed product. Thus, relatively high protein contents have been found in presentations with almond (3.4–3.7%), hazelnut (1.8%), and those with stuffing materials of animal origin, such as ham (1.8%) or tuna (1.7%). The ranges of ash and fiber contents (2.5–7.0% and 1.8–5.9%, respectively) are also higher than those for green olives in bulk. It can be said that, on average, Spanish-style green olives have a moderate content (2.6%) of dietary fiber, which is interesting from a nutritional point of view.
TABLE 31.3 Range and mean values of the main physicochemical characteristics and proximate composition of commercial packed Spanish-style green olives (adapted from López-López, 2006). Range
Meana
Physicochemical characteristic pH
3.25–4.19
3.69 (0.03)
Titratable acidity (% lactic acid)
0.42–1.77
0.64 (0.03)
Combined acidity (N)
0.021–0.095
0.046 (0.22)
Salt (% NaCl)
3.5–8.2
5.5 (0.17)
Proximate composition (g/100 g edible portion) Water
65.2–83.4
75.7 (0.52)
Fat
8.7–23.2
15.1 (0.47)
Protein
0.7–3.7
1.2 (0.08)
Ash
2.5–7.0
4.5 (0.08)
Fiber
1.8–5.9
2.6 (0.05)
This table shows the ranges and mean values of both physicochemical characteristics of brine and major components of pulp in Spanish-style green olives. Values were obtained from a survey of different commercial presentations of packed product in Spain. As expected, the mean values of physicochemical parameters are lower than those for the fermented product in bulk. On the other hand, the ranges of protein, ash, and fiber contents are higher than those for green olives in bulk. a Standard error in parentheses. N ⫽ 48.
Metals, Electrolytes and Other Components
Fiber intake can decrease the incidence of various disorders, such as cardiovascular complications, cancer (colon, breast, and prostate), and hypercholesterolemia (Heredia et al., 2002). It has been found that the proportion of soluble fiber is always negligible with respect to the insoluble fraction, so the total fiber in green table olives (or table olives in general) is, in practice, insoluble fiber. The fiber content in green table olives is comparable to that reported for other vegetables or fruits, and is exceeded only by that in dried fruits such as figs, apricots, or raisins (USDA, 2007). Due to the consumption of sugars during the fermentation process and additional losses in later operations prior to packing, the content of sugars in the packed product can be expected to be fairly low. Average glucose, fructose, and mannitol contents in Spanish-style olives on the Spanish market have been determined as 8.89, 17.87, and 13.12 mg/100 g per edible portion (López-López et al., 2007). Sucrose is not detected in any commercial presentation. The organic acids present in raw olives are practically removed during processing. Therefore, the main acids found in the packed product (0.3% lactic acid and 0.1% acetic acid, on average) are those produced during fermentation or storage or added to the cover liquid (López-López et al., 2007). Total polyphenol content depends on the olive cultivar; it ranges between 200 and 1000 mg kg⫺1 in the pulp of packed Spanish-style olives (Romero et al., 2004). This concentration is similar to the total polyphenol concentration (about 400 mg kg⫺1) of commercial virgin olive oils (García et al., 2003).
31.4 UNTREATED GREEN OLIVES IN BRINE The traditional method for processing untreated green olives (Figure 31.1) includes harvesting, transportation to the factory, sorting to remove damaged fruits, washing to remove surface dirt, and finally brining at a salt concentration of around 6% (w/v) at equilibrium, where the fruits undergo a spontaneous fermentation by a mixture of lactic acid bacteria and yeasts. This fermentation is similar to that of untreated turning-color olives (Garrido et al., 1997). Spontaneous fermentation is typically the result of competitive activities of the indigenous flora together with a variety of contaminating microorganisms from fermentation vessels, pipelines, pumps, and other devices in contact with the olives and brine. Those microorganisms best adapted to the food substrate and to technical control parameters during fermentation eventually dominate the process. However, the indigenous flora of the fruits will vary depending on the quality of the raw material, harvesting conditions, and post-harvest treatments, thus leading to variations in the sensory and organoleptic characteristics of the final product. Inoculation of the brine with a commercial starter culture of Lactobacillus pentosus
CHAPTER | 31 Chemical Composition of Fermented Green Olives
reduces the probability of spoilage and helps to achieve an improved and more-predictable fermentation process (Panagou et al., 2003).
31.4.1 Product in Bulk An extensive survey on the chemical profile of the product in bulk has not been carried out to date. Both the fermentation process (which in turn is affected by NaCl concentration of the brine) and the olive cultivar have been demonstrated to affect the final chemical composition (Nosti Vega et al., 1979). As there is no previous lye treatment, the epidermis of the fruits remains intact, and acts as a barrier, so that diffusion of sugar and polyphenols is slow. As a consequence, levels of these compounds in the pulp after fermentation are higher in untreated green olives than in Spanish-style green olives. The major phenolics in the pulp after fermentation are hydroxytyrosol and tyrosol in both processing types, but an appreciable level of oleoside 11-methylester has also been detected in untreated olives (Marsilio et al., 2005). Lactic and acetic acids are the major fermentation end-products. Citric, tartaric and malic acids can also be detected in small amounts at the end of process (Panagou et al., 2003). The final pH is 4.0 or lower (Panagou et al., 2003; Marsilio et al., 2005, 2006).
Untreated green olives in brine Harvesting
Transport
Sorting
Brining
Culture addition (optional)
Fermentation
Grading
Packing FIGURE 31.1 Flow diagram for untreated green olive processing. This figure shows the main steps of untreated green olive processing. The traditional processing method is based on spontaneous fermentation by a mixture of lactic acid bacteria and yeasts. However, the use of lactic starter cultures as inoculants can be also applied, with positive effects on fermentation.
295
31.4.2 Packed Product With regard to the chemical profile of the packed product, ranges and mean values for physicochemical characteristics and proximate composition of untreated (green and turningcolor) olives sampled from the Spanish market are shown in Table 31.4 (López-López et al., 2004, 2007; LópezLópez, 2006; López et al., 2007). Mean values of physicochemical parameters for the packed product are similar to those for packed Spanish-style green olives. Limits for these parameters in untreated table olives have been established by the IOOC: the maximum pH is established at 4.3, irrespective of the preservation technique used; the minimum value for titratable acidity, expressed as lactic acid, is established at 0.3% (w/v), except for pasteurized olives; and the minimum salt content is established at 6% (w/v), except for pasteurized olives (IOOC, 2004). Therefore, the mean values of pH and titratable acidity shown in Table 31.4 are in compliance with the legislation, but salt content is below the limit. This may be a reflection of a tendency towards the use of pasteurization to stabilize this product, thus satisfying consumer demand for a foodstuff with a lower level of salt. Presentations of untreated olives have lower moisture (⬇6% lower, on average) and higher fat (⬇7% higher, on average) contents than those of Spanish-style olives. However, the mean protein content is identical (1.2%) in the two products. The ash content ranges from 2.5% to 5.0%, with a mean value of 3.8%. Therefore, ash values are lower in untreated olives than in treated green olives. This may be attributed to the lower concentrations of salt usually used in the latter product, but also to the lower permeability of the epidermis of the fruit or to a lower capacity of the olive tissue to absorb salts. On the other hand, the fiber content is significantly (p ⬍ 0.05) higher in untreated olives than in Spanish-style green olives. This can be attributed to the lye treatment, which affects the cell wall components of olives in the latter product (Jiménez et al., 1995). As in the case of green olives, the sugar content is very low in the packed untreated product. Thus, average glucose, fructose, and mannitol contents have been determined as 9.47, 10.97, and 54.90 mg/100 g per edible portion (LópezLópez et al., 2007). Therefore, on average, untreated olives exhibit higher residual sugars than do treated green olives. This can be attributed to partial fermentation and the more difficult solubilization of sugars (due to absence of lye treatment) in the former product. The major organic acids found in the packed product (0.1% lactic acid and 0.2% acetic acid, on average) are those produced during fermentation or storage or added to the cover liquid (López-López et al., 2007). Because the polyphenol content, which is positively correlated with the bitter taste of fermented fruit
296
SECTION | I
TABLE 31.4 Range and mean values of the main physicochemical characteristics and proximate composition of commercial packed untreated (green and turning-color) table olives (adapted from López-López, 2006). Range
●
Meana
Physicochemical characteristic
●
pH
3.46–4.07
3.92 (0.11)
Titratable acidity (% lactic acid)
0.27–1.26
0.70 (0.09)
Combined acidity (N)
0.030–0.074
0.044 (0.53)
Salt (% NaCl)
3.91–5.72
4.98 (0.22)
Proximate composition (g/100 g edible portion) Water
58.5–73.0
68.7 (1.5)
Fat
18.7–30.7
21.8 (1.2)
Protein
0.9–1.4
1.2 (0.06)
Ash
2.5–5.0
3.8 (0.12)
Fiber
2.2–4.8
3.3 (0.13)
This table shows the ranges and mean values of both physicochemical characteristics of brine and major components of pulp in untreated (green and turning-color) olives. Values were obtained from a survey of different commercial presentations of packed product in Spain. Mean values of physicochemical parameters are similar to those for packed Spanish-style green olives. On average, untreated olives have lower moisture and ash contents, but higher fat and fiber contents than those of Spanish-style olives. a
Standard error in parentheses. N ⫽ 9.
(Marsilio et al., 2005, 2006), is higher in untreated than in treated olives, it is reasonable to find this difference in the packed products. In fact, it has been found that packed untreated turning-color olives have the highest polyphenol concentration (as much as 1200 mg kg⫺1) when compared with packed Spanish-style green olives or black ripe (Californian-style) olives (Romero et al., 2004). Hydroxytyrosol, tyrosol, and salidroside are the major phenols found by those authors in untreated turning-color olives. These phenols, especially hydroxytyrosol, make a significant contribution to the antioxidant activity of olives (Owen et al., 2003; Pereira et al., 2006).
SUMMARY POINTS ●
●
In general, there are significant differences between the chemical composition of Spanish-style green olives and that of untreated (green or turning-color) olives.
●
Metals, Electrolytes and Other Components
On average, untreated olives have lower moisture and ash contents, but higher fat and fiber contents. Also, untreated olives exhibit a higher residual sugar content. With regard to health and the prevention of disease, both Spanish-style green olives and untreated (green or turning-color) olives have interesting properties, because of their moderately high percentage of total dietary fiber and their high polyphenol content. Because of their higher contents of polyphenols and fiber, untreated olives are nutritionally superior to treated olives. The consumption of fermented green olives in addition to olive oil will be a more efficient means of obtaining phenolic antioxidants, reinforcing the health-promoting properties of the Mediterranean diet.
REFERENCES Amiot, J.M., Fleuriet, A., Macheix, J.J., 1986. Importance and evolution of phenolic compounds in olive during growth and maturation. J. Agric. Food Chem. 34, 823–826. Amiot, M.J., Tacchini, M., Fleuriet, A., Macheix, J.J., 1990. The technological debittering process of olives: characterization of fruits before and during alkaline treatment. Sci. Aliments. 10, 619–631. Bianchi, G., 2003. Lipids and phenols in table olives. Eur. J. Lipid Sci. Technol. 105, 229–242. Brenes, M., de Castro, A., 1998. Transformation of oleuropein and its hydrolysis products during Spanish-style green olive processing. J. Sci. Food Agric. 77, 353–358. Brenes, M., Rejano, L., García, P., Sánchez, A.H., Garrido, A., 1995. Biochemical changes in phenolic compounds during Spanish-style green olive processing. J. Agric. Food Chem. 43, 2702–2706. de Castro, A., Montaño, A., Casado, F.J., Sánchez, A.H., Rejano, L., 2002. Utilization of Enterococcus casseliflavus and Lactobacillus pentosus as starter cultures for Spanish-style green olive fermentation. Food Microbiol. 19, 637–644. Essid, H., 2008. New strategies for the production and commercialization of table olives. II International Table Olive Conference, Dos Hermanas, Spain. Fernández-Bolaños, J., Fernández-Díez, M.J., Moreno, M.R., Serrano, A.G., Romero, T.P., 1983. Azúcares y polioles en aceitunas verdes. III. Determinación cuantitativa por cromatografía gas-líquido. Grasas y Aceites 34, 168–171. Fernández-Díez, M.J., 1983. Olives. In: Rehm, H.J., Reed, G. (eds) Biotechnology, 8th edn., Vol. 5. Verlag Chemie, Weinheim, Germany, pp. 379–397. Fernández-Díez, M.J., Castro, R., Fernández, A.G., Cancho, F.G., Pellissó, F.G., Vega, M.N., Moreno, A.H., Mosquera, I.M., Navarro, L.R., Quintana, M.C.D., Roldán, F.S., García, P.G., Castro, A., 1985. Biotecnología de la aceituna de mesa. (CSIC, ed.). CSIC, Madrid. Fleming, H.P., Moore, W.R., 1983. Pickling. In: Fuller, G. and Dull, G.G. (eds), Processing of Horticultural Crops in the United States. CRC Handbook of Processing and Utilization in Agriculture. Vol. II, part 2: Plant Products. I.A. Wolff, ed., CRC Press Inc., Boca Raton, FL, pp. 397–463. García, A., Brenes, M., García, P., Romero, C., Garrido, A., 2003. Phenolic content of commercial olive oils. Eur. Food Res. Technol. 216, 520–525.
CHAPTER | 31 Chemical Composition of Fermented Green Olives
Garrido, A., Fernández-Díez, M.J., Adams, R.M., 1997. Table Olives. Production and Processing. Chapman and Hall, London. Guillén, R., Heredia, A., Felizón, B., Jiménez, A., Montaño, A., FernándezBolaños, J., 1992. Fibre fraction carbohydrates in Olea europaea (Gordal and Manzanilla var.). Food Chem. 44, 173–178. Heredia, A., Jiménez, A., Fernández-Bolaños, J., Guillén, R., Rodriguez, R., 2002. Fibra Alimentaria. Biblioteca de las Ciencias, Vol. 4, Consejo Superior de Investigaciones Científicas, Madrid. IOOC (International Olive Oil Council), 2004. Trade Standard Applying to Table Olives. Res-2/91-IV/04. OOC, Madrid. Jiménez, A., Guillén, R., Sánchez, C., Fernández-Bolaños, J., Heredia, A., 1995. Changes in texture and cell wall polysaccharides of olive fruits during “Spanish Green Olive” processing. J. Agric. Food Chem. 43, 2240–2246. Leal-Sánchez, M.V., Ruiz-Barba, J.L., Sánchez, A.H., Rejano, L., Jiménez-Díaz, R., Garrido-Fernández, A., 2003. Fermentation profile and optimisation of green olive fermentation using Lactobacillus plantarum LPCO10 as a starter culture. Food Microbiol. 20, 421–430. López-López, A., 2006. Composición y valor nutricional de la aceituna de mesa. PhD Thesis, University of Seville. López-López, A., García-García, P., Durán-Quintana, M.C., GarridoFernández, A., 2004. Physicochemical and microbiological profile of packed table olives. J. Food Prot. 67, 2320–2325. López, A., Garrido, A., Montaño, A., 2007. Proteins and amino acids in table olives: relationship to processing and commercial presentation. Ital. J. Food Sci. 19, 217–228. López-López, A., Jiménez-Araujo, A., García-García, P., GarridoFernández, A., 2007. Multivariate analysis for the evaluation of fiber, sugars, and organic acids in commercial presentations of table olives. J. Agric. Food Chem. 55, 10803–10811. Marsilio, V., Campestre, C., Lanza, B., De Angelis, M., 2001. Sugar and polyol compositions of some European olive fruit varieties (Olea europaea L.) suitable for table olive purposes. Food Chem. 72, 485–490. Marsilio, V., Seghetti, L., Iannucci, E., Russi, F., Lanza, B., Felicioni, M., 2005. Use of a lactic acid bacteria starter cultura during green olive (Olea europaea L. cv. Ascolana tenera) processing. J. Sci. Food Agric. 85, 1084–1090. Marsilio, V., d´Andria, R., Lanza, B., Russi, F., Iannucci, E., Lavini, A., Morelli, G., 2006. Effect of irrigation and lactic acid bacteria inoculants on the phenolic fraction, fermentation and sensory characteristics of olive (Olea europaea L. cv. Ascolana tenera) fruits. J. Sci. Food Agric. 86, 1005–1013. Montaño, A., Sánchez, A.H., de Castro, A., 1993. Controlled fermentation of Spanish-type green olives. J. Food Sci. 4, 842–844. Nosti Vega, M., Vazquez Ladrón, R., de Castro Ramos, R., 1979. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. II. Aceitunas verdes en salmuera. Grasas y Aceites 30, 93–100. Oliveira, M., Brito, D., Catulo, L., Leitao, F., Gomes, L., Silva, S., Vilas-Boas, L., Peito, A., Fernández, I., Gordo, F., Peres, C., 2004.
297
Biotechnology of olive fermentation of “Galega” Portuguese variety. Grasas y Aceites 55, 219–226. Owen, R.W., Haubner, R., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2003. Isolation, structure elucidation and antioxidant potential of the major phenolic and flavonoid compounds in brined olive drupes. Food Chem. Toxicol. 41, 703–717. Panagou, E.Z., Tassou, C.C., Katsaboxakis, C.Z., 2003. Induced lactic acid fermentation of untreated green olives of the Conservolea cultivar by Lactobacillus pentosus. J. Sci. Food Agric. 83, 667–674. Panagou, E.Z., Tassou, C.C., 2006. Changes in volatile compounds and related biochemical profile during controlled fermentation of cv. Conservolea green olives. Food Microbiol. 23, 738–746. Panagou, E.Z., Tassou, C.C., Skandamis, P.N., 2006. Physicochemical, microbiological, and organoleptic profiles of Greek table olives from retail outlets. J. Food Prot. 69, 1732–1738. Patumi, M., Dándria, R., Fontanazza, G., Morelli, G., Giorio, P., Sorrentino, G., 1999. Yield and oil quality of intensively trained trees of three cultivars of olive (Olea europaea) under different irrigation regimes. J. Hortic. Sci. Biotechnol. 74, 729–737. Pereira, J.A., Pereira, A.P.G., Ferreira, I.C.F.R., Valentao, P., Andrade, P.B., Seabra, R., Estevinho, L., Bento, A., 2006. Table olives from Portugal: phenolic compounds, antioxidant potential, and antimicrobial activity. J. Agric. Food Chem. 54, 8425–8431. Romero, C., Brenes, M., Yousfi, K., García, P., García, A., Garrido, A., 2004. Effect of cultivar and processing method on the contents of polyphenols in table olives. J. Agric. Food Chem. 52, 479–484. Romero, C., García, P., Brenes, M., García, A., Garrido, A., 2002. Phenolic compounds in natural black Spanish olive cultivars. Eur. Food Res. Technol. 215, 489–496. Ryan, D., Antolovich, M., Herlt, T., Prenzler, P.D., Lavee, S., Robards, K., 2002. Identification of phenolic compounds in tissues of novel olive cultivar Hardy’s Mammoth. J. Agric. Food Chem. 50, 6716–6724. Sánchez, A.H., de Castro, A., Rejano, L., Montaño, A., 2000. Comparative study on chemical changes in olive juice and brine during green olive fermentation. J. Agric. Food Chem. 48, 5975–5980. Sánchez, A.H., Rejano, L., Montaño, A., de Castro, A., 2001. Utilization at high pH of starter cultures of lactobacilli for Spanish-style green olive fermentation. Int. J. Food Microbiol. 67, 115–122. Servili, M., Settanni, L., Veneziani, G., Esposto, S., Massitti, O., Taticchi, A., Urbani, S., Montedoro, G.F., Corsetti, A., 2006. The use of Lactobacillus pentosus 1MO to shorten the debittering process time of black table olives (cv. Itrana and Leccino): a pilot-scale application. J. Agric. Food Chem. 54, 3869–3875. Tassou, C.C., Panagou, E.Z., Katsaboxakis, K.Z., 2002. Microbiological and physicochemical changes of naturally black olives fermented at different temperatures and NaCl levels in the brines. Food Microbiol. 19, 605–615. USDA, Agricultural Research Service. 2007. USDA Nutrient database for standard reference, release 20. Nutrient data laboratory home page, http://www.nal.usda.gov/fnic/foodcom. Visioli, F., Galli, C., 1998. Olive oil phenols and their potencial effects on human health. J. Agric. Food Chem. 46, 4292–4296.
This page intentionally left blank
Chapter 32
Some Metals in Table Olives Yasemin Sahan Department of Food Engineering, Faculty of Agriculture, Uludag University, Bursa, Turkey
32.1 INTRODUCTION
Others 15%
The domesticated olive tree is of very ancient origin probably arising at the dawn of agriculture. It is thought to have originated in the Middle East spreading south and west to the rest of the Mediterranean basin through movement and trading activities of the Phoenicians and ancient Greeks. Historically and to the present time olives have been culturally and economically significant for the Mediterranean and Middle Eastern regions (Kailis and Harris, 2004). The olive agrifood chain is the largest agrindustrial sector in some countries such as Spain, Turkey, Greece and Italy. Olive products, mainly table olives, have been processed as plant foods since prehistoric times. The consumption of table olives is widespread in the Mediterranean region, but increasing in other non-producing countries (IOOC, 2005). The worldwide production of table olives has been estimated to be 1 796 000 tons in the 2007/2008 season (IOOC, 2007a) (Figure 32.1). Metals in olives are important both from nutritional and toxicological viewpoints. Some metals, particularly iron, copper and zinc, are essential substances for the human body and their deficiency can have chronic and acute effects. But even these elements can have toxic effects depending on the chemical form, dose, route of absorption, and a host of other factors. Other metals, especially heavy metals, such as lead and cadmium, are well known as potentially toxic elements. Diseases caused by improper nutrition, including the consumption of food contaminated by trace elements, constitute serious problems in today’s world. Heavy metal toxicity can result in damaged or reduced mental and central nervous function, lower energy levels, and damage to blood composition, lungs, kidneys, liver, and other vital organs. Long-term exposure may result in slowly progressing physical, muscular, and neurological degenerative processes that mimic Alzheimer’s disease, Parkinson’s disease, muscular dystrophy, multiple sclerosis and cancer (Szteke, 2002). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Spain 27% Italy 4% USA 5% Argentina 5% Greece 5%
Turkey 15%
Syria 6% Morocco 6%
Egypt 12%
FIGURE 32.1 Countries with significant interests in table olive production in 2007–2008 (IOOC, 2007a).
Metal contamination can take place during the handling and processing of foods, from farm to the point of consumption. Hence, the importance of being able to monitor low-level concentrations of metals in food to ensure that levels are not exceeded (Nasreddine and Parent-Massin, 2002). Foods have been analyzed for different elements up to μg kg⫺1 levels using different techniques such as atomic emission spectrometry (AES), atomic absorption spectrometry (AAS), induced coupled plasma atomic emission (ICP-AES) and induced coupled plasma mass spectrometry (ICP-MS). Owing to the peculiar characteristics of ICP-MS (low detection limits, multi elemental capacity, wide linear range, etc.) the number of papers dealing with the analysis of food samples by ICP-MS has increased in recent years (Alam et al., 2003; Barbaste et al., 2003; Nikkarinen and Mertanen, 2004).
299
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
300
SECTION | I
The metal content of table olives could originate from depending on the mineral constitution of the soils where the olive trees are located, weather condition, resulting from environmental pollution, contamination by phytochemical products, cultivars differences, maturation, processing method, packing material and the chemicals used (Garcia et al., 2002; Soares et al., 2006).
32.2 OLIVE FRUITS Metals in olive fruit vary markedly among the cultivars. Metal concentrations in Turkish cultivars are given in Table 32.1. Potassium is the most abundant element in the olive fruit, followed by calcium, magnesium and sodium. The variations of metal concentrations in the olive flesh during ripening were not uniform. A metal concentration in the olive flesh varies during ripening; while Na, K and Mg increased at all intervals, Ca and Mn decreased. These variations may originate from olive varieties or the distribution of metals in the soil, as well as environmental and weather conditions during the ripening period (Nergiz and Engez, 2000; Soyergin and Katkat, 2002). Pollution of the environment with toxic metals has increased dramatically since the onset of the industrial revolution. Food pollution with heavy metals, such as cadmium, lead, copper, etc., is a widespread problem of concern.
TABLE 32.1 Metal contents of Turkish olive fruit (Nergiz and Engez, 2000; Biricik and Basoglu, 2006). Metals*
Samanlı
Domat
Memecik
Na
107.0–150.0
89.0–111.0
32.9–43.0
K
5070.0–7470.0
4410.0– 4760.0
14318.0– 15820.0
Ca
127.0–266.0
178.0–255.0
413.0–650.0
Mg
159.0–291.0
85.0–372.0
269.0–344.0
Mn
2.37–2.50
1.80–2.34
0.67–1.21
Fe
5.67–14.88
5.62–9.61
5.60–11.40
Zn
3.21–4.11
2.00–5.06
4.50–10.00
Cu
5.25–7.19
3.59–4.79
2.30–8.40
Pb
0.07–0.14
0.03–0.04
ND
Cd
0.01–0.02
0.01–0.05
ND
As
0.01–0.03
0.01–0.03
0.06–0.10
ND: Not detected. * mg kg⫺1.
Metals, Electrolytes and Other Components
Although heavy metals are naturally present in soil contamination from local sources, mostly industry (mainly nonferrous industries, but also power plants, iron and steel and chemical industries), agriculture (irrigation with polluted waters, use of fertilizers, especially phosphates, contaminated manure, sewage sludge and pesticides containing heavy metals), waste incineration, combustion of fossil fuels and highway traffic. Long-range transport of atmospheric pollutants adds to the metal load and is the main source of heavy metals in agricultural areas (Celik et al., 2005). Olive fruits can be affected by all these factors. The concentrations of metals found in the samples collected from a field situated near to busy roads and industrial areas are shown in Table 32.2 (Sahan and Basoglu, 2004). Lead is toxic to humans, animals and plants. Hence, its concentration has been limited in food by various regulations (Anonymous, 2003, 2004). Sahan and Basoglu (2004) reported the lead values of post-harvest olive samples vary between 778.11 and 3680.90 μg kg⫺1 wet weight. Agricultural uses of phosphate fertilizers and sewage sludge and industrial uses of cadmium have been identified as a major cause of widespread dispersion of the metal at trace levels into the environment and food (Satarug et al., 2003). Cadmium induces serious peroxidation in membrane structures and has been associated with several diseases. Moreover, cadmium is mutagenic and has been linked with carcinogenic activity (Casalino et al., 2002). The content of cadmium in olives varied from 61.47–356.44 μg kg⫺1 wet weight (Sahan and Basoglu, 2004). Iron is an essential metal in plants and is a constituent of many enzyme systems regulating oxidation-reduction processes. Levels of Fe in olive varied between 14.36– 118.55 mg kg⫺1 ww. The samples with higher Fe concentration were harvested from contaminated areas described above (Sahan and Basoglu, 2004). These pollution factors increase soil acidity. Acid conditions enhance the mobility of iron when the pH value of soil is less than 4.0 and the leaching rate is about double of that in neutral conditions (Dierkes and Geiger, 1999). The incidence of the copper studied does not show a regular gradient as the distance from the motor road and industrial area (Caselles, 1998). However, this metal is affected by using metal-based fungicides and pesticides (Vavoulidou et al., 2004). The Zn in olive samples is derived mainly from abrasion of tires and brake disks (Nimis et al., 1999). Copper and zinc are essential elements in all higher plants and play important functions in several physiological processes, such as photosynthesis, respiration and protein metabolism. However, high doses of these metals in food have a detrimental outcome on human health. Sahan and Basoglu (2004) indicated that the accumulation values for Cu and Zn in olives were 2.85–13.01 mg kg⫺1 ww, 2.19–11.53 mg kg⫺1 ww, respectively. Ozturk and Turkan (1993) observed a mean content of zinc of 40 mg kg⫺1 in olives grown near a highway.
301
CHAPTER | 32 Some Metals in Table Olives
TABLE 32.2 The metal contents of post-harvest olive samples (Sahan and Basoglu, 2004). Metals
Provinces
Sample counted
Lead (μg kg⫺1 ww)
Mudanya
30
861.18 ⫾ 3.98
589.42–1376.23
Gemlik
30
1610.63 ⫾ 6.08
857.18–3695.21
Orhangazi
30
1608.14 ⫾ 4.67
778.11–2657.48
Total
90
1362.23 ⫾ 4.94
589.42–3695.21
Mudanya
30
112.43 ⫾ 0.92
59.34–275.28
Gemlik
30
151.07 ⫾ 0.87
65.78–369.14
Orhangazi
30
182.98 ⫾ 0.71
70.08–348.75
45.56 ⫾ 0.22
Control
Iron (mg kg⫺1 ww)
Total
90
158.29 ⫾ 1.12
59.34–369.14
Mudanya
30
22.11 ⫾ 2.03
6.85–51.33
Gemlik
30
56.72 ⫾ 1.97
9.63–129.38
Orhangazi
30
102.14 ⫾ 2.11
10.05–146.32
5.97 ⫾ 0.34
Control
Copper (mg kg⫺1 ww)
Total
90
66.37 ⫾ 2.03
Mudanya
30
4.76 ⫾ 0.43
2.85–7.64
Gemlik
30
7.14 ⫾ 0.85
3.78–10.22
Orhangazi
30
6.98 ⫾ 0.54
3.41–13.01
6.85–146.32
2.10 ⫾ 0.38
Control
Zinc (mg kg⫺1 ww)
Range
425.51 ⫾ 2.86
Control
Cadmium (μg kg⫺1 ww)
Mean ⫾ SD
Total
90
6.51 ⫾ 0.79
2.85–13.01
Mudanya
30
5.44 ⫾ 0.82
3.64–9.02
Gemlik
30
5.69 ⫾ 1.11
3.79–11.53
Orhangazi
30
6.78 ⫾ 0.47
2.19–10.25
2.18 ⫾ 0.09
Control Total
90
32.3 PROCESSED OLIVES Table olives are prepared from specifically cultivated olive varieties picked at the right maturation stage and their quality, after appropriate processing, is that of an edible wellpreserved product. The most common processing methods
5.76 ⫾ 0.79
2.19–11.53
for table olives are the Spanish for green olives and the Californian for oxidized black olives and naturally black olives (untreated black olives in brine and untreated black olives in dry salt). In the Spanish and Californian procedures, olives are treated with a diluted aqueous NaOH solution that brings
302
SECTION | I
Metals, Electrolytes and Other Components
TABLE 32.3 Average values of metals on table olive samples (Sahan et al., 2007). Black olives (mg kg⫺1)
Metals Minimum
Maximum
Mg
36.12 ⫾ 3.46*
125.11 ⫾ 5.02
Cr
0.35 ⫾ 0.13
Fe
Green olives (mg kg⫺1) Mean value
Minimum
Maximum
Mean value
79.28 ⫾ 19.58
21.61 ⫾ 3.24
57.91 ⫾ 8.62
37.64 ⫾ 13.14
0.88 ⫾ 0.14
0.56 ⫾ 0.11
0.39 ⫾ 0.18
1.30 ⫾ 0.43
0.59 ⫾ 0.20
6.11 ⫾ 1.45
48.58 ⫾ 2.53
12.65 ⫾ 8.44
4.45 ⫾ 1.36
12.27 ⫾ 4.67
7.08 ⫾ 1.56
Co
0.05 ⫾ 0.01
0.08 ⫾ 0.01
0.06 ⫾ 0.01
0.05 ⫾ 0.01
0.09 ⫾ 0.01
0.06 ⫾ 0.03
Ni
0.18 ⫾ 0.06
0.53 ⫾ 0.04
0.30 ⫾ 0.06
0.20 ⫾ 0.09
0.50 ⫾ 0.05
0.37 ⫾ 0.06
Cu
0.73 ⫾ 0.08
2.55 ⫾ 0.22
1.48 ⫾ 0.38
0.54 ⫾ 0.53
1.37 ⫾ 0.46
0.78 ⫾ 0.20
Zn
4.25 ⫾ 1.03
13.33 ⫾ 1.94
8.50 ⫾ 1.74
5.59 ⫾ 2.01
14.30 ⫾ 1.87
10.58 ⫾ 2.01
Cd
0.08 ⫾ 0.01
0.15 ⫾ 0.01
0.11 ⫾ 0.01
0.09 ⫾ 0.01
0.16 ⫾ 0.02
0.12 ⫾ 0.04
Sn
14.40 ⫾ 2.31
53.62 ⫾ 3.56
35.48 ⫾ 7.65
33.34 ⫾ 4.21
47.58 ⫾ 3.85
39.06 ⫾ 6.42
Pb
0.57 ⫾ 0.04
0.91 ⫾ 0.06
0.71 ⫾ 0.07
0.56 ⫾ 0.03
0.86 ⫾ 0.06
0.75 ⫾ 0.12
*Mean value (n ⫽ 6) of standard deviation (p ⬍ 0.05).
about several changes in the susceptible classes of compounds in the fruit. After treatment the olives are rinsed to remove the alkali and the fruit is then left to ferment in brine for several months. The production of natural black olives is a simple natural process, which does not use chemicals (Dura´n Quintana et al., 1999; Uccella, 2001; Tassou et al., 2002; Bianchi, 2003). All of the different treatments may produce some changes in the metal composition of the processed olives. Sahan et al. (2007) determined the Mg, Cr, Fe, Co, Ni, Cu, Zn, Cd, Sn, and Pb content in the most commonly consumed table olives from different locations in Bursa, Turkey. The range of the concentrations of the metals in black and green olives is given in the Table 32.3. Biricik and Basoglu (2006) and Lopez et al. (2008) reported metal content in brined olives over a range of processing styles and cultivars (Table 32.4). As seen Table 32.3, Mg levels were found to be higher than other tested metals (Sahan et al., 2007). The amount decreased in the order of Sn, Fe, Zn, Cu, Pb, Cr, Ni, Cd and Co. In the comparison of the concentration of trace elements among black and green olives, differences were observed. These variations could be from different olive varieties, the distribution of elements in the soil, the maturation and processing methods or packing material as well as environmental and weather conditions. Sahan et al. (2007) showed Mg was the most abundant among the elements quantified. Mg concentrations were determined between 36.12 ⫾ 3.46 and 125.11 ⫾ 5.02
in black table olives, whereas in green table olives levels were between 21.61 ⫾ 3.24 and 57.91 ⫾ 8.62. Nergiz and Engez (2000), Yasar and Gücer (2004) and Lopez et al. (2008) reported Mg concentrations in green olives in the range of 114–372 mg kg⫺1, 132.0–223.3 mg kg⫺1 and 98.6–146.7 mg kg⫺1, respectively. In addition, the concentration of metal differed significantly among the three processing methods of the black table olives. These differences are related to the differences in the processes. Thus, the use of NaOH in the Californian style may decrease the mineral concentration in the black olive samples. The Food Composition and Nutrition Tables recorded an average of 190 mg kg⫺1 for this element in green marinated olives. The daily value for magnesium is 400 mg. Olives may then contribute to fill its requirements (Lopez et al., 2008). Sodium is the only element habitually added during processing (Garrido Fernandez et al., 1997) and its concentration was the highest in all commercial presentations. Biricik and Basoglu (2006) and Lopez et al. (2008), determined the Na levels (11 490–17 840 mg kg⫺1) in table olives. This high Na concentration is because the preservation of most olive products still relies on salt. However, there are big differences among elaboration styles. The highest concentrations were found in green olives followed by directly brined olives and ripe olives. Ripe olives have only a moderate Na content because they are preserved by sterilization. Ca is used in the preparation of the stuffing strips from green olives and is sometimes added during the storage
303
CHAPTER | 32 Some Metals in Table Olives
TABLE 32.4 Metal content in brined olives. Metals (mg kg⫺1)
Gordal
Ascolana
Manzanilla
Hojiblanca
Arbequina
Alorena
Verdial
Samanlı
Domat
Na
12675
16030
11490
13797
13152
16600
16020
17040
17840
K
1176
6171
750
684
571
902
766
7300
4123
Ca
379
462
447
337
613
691
684.7
653.7
850.3
P
129.4
Nd
121.9
111.9
105.4
99
144.1
ND
ND
Mg
119.0
77.4
143.4
56.1
67.7
53.8
197.3
149.9
76.8
Mn
0.24
1.40
0.55
0.92
0.29
1.04
1.09
2.33
1.77
Fe
7.7
3.23
5.51
7.62
4.09
5.73
4.90
5.46
5.14
Cu
5.09
4.38
5.22
3.99
10.93
4.68
4.35
6.67
3.50
Zn
3.20
2.23
1.98
2.90
1.55
2.03
2.94
3.05
2.18
ND: Not detected
phase of ripe olive processing, so its high concentration in these elaboration styles may be justified. However, its concentration in directly brined olives, in which it is not intentionally incorporated, is also high. In general, olive flesh can absorb Ca and retain it. This Ca is not released during the lye or water washing treatments involved in processing (Garrido Fernandez et al., 1997). The concentration of Ca in green, directly brined and ripe olives ranged from 476 to 850, 337 to 691, and 363 to 731 mg kg⫺1, respectively (Lopez et al., 2008). Other values of Ca concentration in olives from the literature are 270–450 mg kg⫺1 or 110–230 mg kg⫺1 for Kalamata or natural black olives after fermentation, respectively (Unal and Nergiz, 2003) and 422–850 mg kg⫺1 in green table olives (Biricik and Basoglu, 2006). Potassium is the most abundant element in fresh olives; however, this monovalent metal is not as strongly fixed as Ca in the pulp of the olive and is progressively lost during processing (Garrido Fernandez et al., 1997). Lopez et al. (2008), reported potassium concentration is higher in directly brined olives (especially in Gordal and Alorena, 1176 and 902 mg kg⫺1, respectively), which are not subjected to lye treatments. Successive immersions in lye or washing cycles with water partially diminish the potassium content (mainly in ripe olives). But, final concentrations in ripe olives were still high (82–223 mg kg⫺1). Values reported by Unal and Nergiz (2003) found 560–1130 mg kg⫺1, 1140–1820 mg kg⫺1, and 3260–3760 mg kg⫺1 for green, Kalamata and natural black olives, respectively. Biricik and Basoglu (2006) illustrated a concentration of 4230–7410 mg kg⫺1 in green table olives. These differences may result from different cultivars.
The three elements Fe, Zn and Cu are essential in human nutrition and have been quantified in olives. Iron salts such as ferrous lactate and ferrous gluconate are utilized for fixing the final color of ripe olives (Brenes et al., 1995). The maximum concentration of total Fe allowed is 150 mg kg⫺1 (IOOC, 2004). The presence of Fe in green and directly brined olives is unfavorable because it may cause browning due to the formation of complexes with the olive polyphenols (Garrido Fernandez et al., 1997). As shown in Table 32.3, the mean Fe levels of olive samples varied between 4.45 and 48.58 mg kg⫺1 depending on the type, processing method, brand and packing type. Several studies have monitored the levels of Fe in olives. Fe concentrations reported by other authors were 3.23– 15.10 mg kg⫺1 in green table olives (Biricik and Basoglu, 2006), 10.76–180.06 mg kg⫺1 (Ziena et al., 1997) and 3.49–7.70 mg kg⫺1 (Lopez et al., 2008). According to the results of other studies, maturation, variety properties and production conditions influenced the level of Fe in olives (Ziena et al., 1997; Biricik and Basoglu, 2006; Sahan et al., 2007; Lopez et al., 2008). Ripe olives may be a good source of Fe for which the daily recommended value is 18 mg (Lopez et al., 2008). Zinc is an essential constituent of more than two hundred metalloenzymes. Zinc plays a key role in the synthesis and stabilization of genetic material and is necessary for cell division and the synthesis and degradation of carbohydrates, lipids and proteins (EVM, 2003). Sahan et al. (2007) analyzed Zn levels in olive samples as shown in Table 32.3. The mean Zn values belonging to black and green types were found to be 8.50 ⫾ 1.74 mg kg⫺1 and 10.58 ⫾ 2.01 mg kg⫺1, respectively. In statistical
304
evaluation, it was found that the differences were important ( p ⬍ 0.05) between cultivars (Sahan et al., 2007). Zinc levels were always low in brined olives (1.55– 3.20 mg kg⫺1) and differences were insignificant ( p ⬍ 0.05) among cultivars (Biricik and Basoglu, 2006; Lopez et al., 2008). It might be that raw material (the origin of the fruit and the maturation period) and different processing methods caused these differences. Among several metals, Cu is of great importance because formulations containing this element are usually used as fungicides to fight fungal diseases of olive trees. Furthermore Cu is a transition metal that even in small concentrations is a very potent oxidation catalyst. Taking into account the high content of lipids in olives, their Cu residues should be controlled to determine their influence on the quality of the final product (Soares et al., 2006). Cu values in table olives from the literature are 0.73–2.55 mg kg⫺1 in black olives (Sahan et al., 2007), 3.50–6.67 mg kg⫺1 in green olives (Biricik and Basoglu, 2006) and 3.99–10.93 mg kg⫺1 in directly brined olives (Lopez et al., 2008). The safe upper level allowed is 10 mg for a 60 kg adult (EVM, 2003). Sahan et al. (2007) determined Sn concentration (14.40–53.63 mg kg⫺1) in the table olives. None of the olive samples were found to be above the limit of 250 mg kg⫺1 recommended by the Codex Alimentarius (1987). It was observed that Sn values showed a relationship to packing materials. This result supported the findings of another reported study showing increasing concentration of Sn in food with respect to use of tinplate for food packaging (Blunden and Wallace, 2003). Manganese is assumed to be an essential nutrient for humans, with an estimated safe and adequate dietary intake being 2–5 mg day⫺1. In addition manganese may catalyze the darkening reaction that occurs in ripe olive processing and chemical oxidation of olive orthodiphenols in model systems (Romero et al., 1998, 2000, 2001; Sahan and Basoglu, 2008). Lopez et al. (2008) and Biricik and Basoglu (2006) determined that manganese content ranged from 0.24–1.09 mg kg⫺1 in directly brined olives and 1.40– 2.72 mg kg⫺1 in green olives. Sahan et al. (2007) found the contents of Cr, Co and Ni were similar in the table olive samples. Also no significant differences of these metal levels were found between olive types (p ⬍ 0.05). Nergiz and Engez (2000) reported that the concentration of Cr decreases over the maturation period in Domat variety olives. On the other hand, there were no differences in the level of Co in the same olives. Ni levels found by Madejan et al. (2006) in wild olives were in the range of 1–3 mg kg⫺1. These differences might be explained by different values in air and soil composition (Konarski et al., 2006; Micó et al., 2006; Tasdemir et al., 2006). Sahan et al. (2007) reported the levels of the toxic metals Cd and Pb were low, being much less than or just
SECTION | I
Metals, Electrolytes and Other Components
about 1 mg kg⫺1 in almost all table olives analyzed. Cd values in the black olives were found to range between 0.08 and 0.15 mg kg⫺1, similar to those in green olives (Table 32.3). Cd values reported by other authors were 3.2–8.1 mg kg⫺1 (Ziena et al., 1997) and 0.06 mg kg⫺1 (Madejan et al., 2006). Sahan et al. (2007) determined that the mean value of Pb was 0.71 mg kg⫺1 in black olives and 0.75 mg kg⫺1 in green olives, which were below the safe limits specified for table olives by Codex Alimentarius. Contact between food and metal, such as processing equipment, storage and packaging containers, is a significant source of metal in food. Once metals are present in food, their concentrations are rarely modified by traditional preparation and processing techniques, although in some cases washing may decrease the metal content (Morgan, 1999). In addition it has been reported that olives treated with chemicals contain more Pb than natural processed olives (Ziena et al., 1997). This case might explain the Pb concentration in Californian- and Spanish-style olives. Sahan et al. (2007) reported that olive consumption of Turkey (185 000 tonnes) is higher than that of many countries (IOOC, 2007b). Therefore this product can contribute significantly to the overall intakes of these metals. The daily consumption of olive for an adult is about 15–45 g per capita. Table 32.5 shows that dietary intake of olive has no significant role in mean toxic metal levels. The levels of the various metals are mostly below the safe limits specified for table olives by the Codex Alimentarius (1987) and Turkish local food standards (TSE, 2003).
SUMMARY POINTS ●
●
●
●
●
●
●
●
●
Heavy metals, such as Cd, Pb, Cu and Zn are commonly present in the environment. Human activities have dramatically increased the amount of heavy metals in the environment and food. The adsorption processes of metals differ not only between plant species, but also between cultivars within one species. Heavy metals cannot be degraded or destroyed in the environment. Metals in olives are important both from nutritional and toxicological viewpoints. Metal contamination can take place during the handling and processing of foods, from farm to the point of consumption. A direct correlation was assessed between the contamination mineral constitution of the soil and the amount of heavy metals in olive fruits. Table olive contamination by metals could be related to the cultivar, location, environmental pollution, usage of phytochemical products, maturation, processing method, packing material and the chemicals used. The presence of metals in olives may affect oil quality.
305
CHAPTER | 32 Some Metals in Table Olives
TABLE 32.5 Estimated intake values of metals based on consumption of table olives. Metal (mg kg⫺1)
Mean value (mg kg⫺1)
Daily estimated intake values* Calculated value from mean
The daily intake value established from JECFA and EVM 280 mg
Rate (%)** Calculated value from mean
Mg
79.28
1.182–3.561 mg
Cr
0.56
0.010–0.023 mg
Fe
12.65
0.180–0.564 mg
Co
0.06
0.001–0.002 mg
0.012 mg
8.30–16.60
Ni
0.30
0.004–0.013 mg
0.13 mg
3.07–10.00
Cu
1.48
0.022–0.073 mg
30 mg
0.06–0.23
Zn
8.50
0.131–0.380 mg
22 mg
0.59–1.72
Cd
0.11
0.002–0.005 mg
0.06 mg
3.33–8.33
Sn
35.48
0.534–1.601 mg
1.8 mg
29.44–88.88
Pb
0.71
0.010–0.030 mg
0.21 mg
4.76–14.28
0.10 mg 20 mg
0.42–1.27 10.00–20.00 0.90–2.80
*The dietary intake of each metal was calculated by multiplying the concentration of these metals in olive by the weights of that group consumed per capita. **The rate (%) of each metal was calculated by dividing the estimated intake value by the daily intake value established from JECFA and EVM and by multiplying by 100.
REFERENCES Alam, M.G.M., Snow, E.T., Tanaka, A., 2003. Arsenic and heavy metal contamination of vegetables grown in Samta Village, Bangladesh. Sci. Total Environ. 308, 83–96. Anonymous, 2003. COT statements on twelve metals and other elements in the 2000 total diet study. Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment. 13 p. Anonymous, 2004. Safety evaluation of certain food additives and contaminants. World Health Organization, Genova. Barbaste, M., Medina, B., Perez-Trujillo, J.P., 2003. Analysis of arsenic, lead and cadmium in wines from the Canary Islands, Spain, by ICP/ MS. Food Add. Contam. 20, 141–148. Biricik, G.F., Basoglu, F., 2006. Determination of mineral contents in some olives (Samanlı, Domat, Manzanilla, Ascolana) varieties. Gida 2, 67–75. Blunden, S., Wallace, T., 2003. Tin in canned food: a review and understanding of occurrence and effect. Food Chem. Toxicol. 41, 1651–1662. Brenes, M., Romero, C., Garcia, P., Garrido, A., 1995. Effect of pH on the colour formed by Fe –phenolic complex in ripe olives. J. Sci. Food Agric. 67, 35–41. Bianchi, G., 2003. Lipids and phenols in table olives. Eur. J. Lipid. Sci. Technol. 105, 229–242. Casalino, E., Calzaretti, G., Sblano, C., Landriscina, V., Tecce, M.F., Landriscina, C., 2002, Antioxidant effect of hydroxytyrosol (DPE) and Mn⫹2 in liver of cadmium-intoxicated rats. Comp. Biochem. Phys. C. 133, 625–632.
Caselles, J., 1998. Levels lead and other metals in citrus alongside a motor road. Water Air Soil Poll. 105, 593–602. Celik, A., Kartal, A.A., Akdog˘an, A. Kaska, Y., 2005. Determining the heavy metal pollution in Denizli (Turkey) by using Robinio pseudoacacia L. Environ. Int. 31, 105–112. Codex Alimentarius, 1987. Codex Standard For Table Olıves, Codex Stan 66, 19 p. Dierkes, C., Geiger, W.F., 1999. Pollution retention capabilities of roadside soils. Water Sci. Technol. 39, 201–208. Dura’n Quintana, M.C., Garc’ıa Garc’ıa, P., Garrido Ferna’ndez, A., 1999. Establishment of conditions for green table olive fermentation at low temperature. Int. J. Food Microbiol. 51, 133–143. EVM, 2003. Safe upper levels for vitamins and minerals. Report of the Expert Group on Vitamins and Minerals. Food Standards Agency, ISBN 1- 904026-11-7. Garcia, P., Romero, C., Brenes, M., Garrıdo, A., 2002. Validation of a method for the analysis of iron and manganese in table olives by flame atomic absorption spectrometry. J. Agric. Food Chem. 50, 3654–3659. Garrido Fernandez, A., Fernandez- Diez, M.J., Adams, R.M., 1997. Table olives. Production and processing. Chapman & Hall. IOOC, 2004. Trade Standard applying to table olives. International Olive Oil Council, COI7OT7NC No 1. Madrid, Spain. IOOC, 2005. Table olive. International Olive Oil Council. IOOC, 2007a. Production of table olives. International Olive Oil Council. IOOC, 2007b. Consumption of table olives. International Olive Oil Council. Kailis, S.G., Harris, D., 2004. Establish Protocols and Guidelines for Table Olive Processing in Australia. A report for the Rural Industries Research and Development Corporation. 117 p.
306 Konarski, P., Hałuszka, J., C´ wil, M., 2006. Comparison of urban and rural particulate air pollution characteristics obtained by SIMS and SSMS. Appl. Surface Sci. 252, 7010–7013. Lopez, A., Garcia, P., Garrido, A., 2008. Multivariate characterization of table olives according to their mineral nutrient composition. Food Chem. 106, 369–378. Madejan, P., Moranon, T., Murillo, J.M., 2006. Biomonitoring of trace elements in the leaves and fruits of wild olive and holm oak trees. Sci. Total Environ. 355, 187–203. Micó, C., Recatalá, L., Peris, M., Sánchez, J., 2006. Assessing heavy metal sources in agricultural soils of a European Mediterranean area by multivariate analysis. Chemosphere 65, 863–872. Morgan, J.N., 1999. Effects of processing on heavy metal content of foods. Adv. Exp. Medicine Biol. 459, 195–211. Nasreddine, L., Parent-Massin, D., 2002. Food contamination by metals and pesticides. Should we worry? Toxicol. Lett. 127, 29–41. Nergiz, C., Engez, Y., 2000. Compositional variation of olive fruit during ripening. Food Chem. 60, 55–59. Nikkarinen, M., Mertanen, E., 2004. Impact of geological origin on trace element composition of edible mushrooms. J. Food Comp. Analy. 17, 301–310. Nimis, P.L., Skert, N., Castello, M., 1999. Biomonitora aggio di metallic in traccia tramite licheni in aree a rischio del friuli-venezia giuliu. Studi. Geob. 18, 3–49. Ozturk, M.A., Turkan, I., 1993. Heavy metal accumulation by plants growing alongside the motor roads: a case study from Turkey. In: Markert, B. (ed.), Plants as biomonitors. Indicators for heavy metals in the terrestrial environment. VCH, Weinheim, 515–522. Romero, C., Garcia, P., Brenes, M. Garrido, A., 1998. Use of manganese in ripe olive processing. Z. Lebensm Unters Forsch A. 206, 297–302. Romero, C., Brenes, M., Garcia, P., Garrido, A., 2000. Optimization of simulated ripe olive darkening in presence of manganese. J. Food Sci. 65, 254–258. Romero, C., Garcia, P., Brenes, M. Garrido, A., 2001. Colour improvement in ripe olive processing by mangenese cations: industrial performance. J. Food Engin. 48, 75–81. Sahan, Y., Basoglu, F., 2004. Determinations of some heavy metal contents of Black olives cv. Gemlik. Institude of Natural and Applied Sciences, Uludag University. PhD Thesis. 167 s. Sahan, Y., Basoglu, F., Gucer, S., 2007. ICP-MS analysis of a series of metals (Namely: Mg, Cr, Co, Ni, Fe, Cu, Zn, Sn, Cd and Pb) in
SECTION | I
Metals, Electrolytes and Other Components
black and green olive samples from Bursa, Turkey. Food Chem. 105, 395–399. Sahan, Y,. Bas¸og˘lu, F., 2008. Use of mangenese in black olive processing. Proceedings of the Fifth International Symposium on Olive Growing. Acta Hortic. 79, 725–727. Satarug, S., Baker, J.R., Urbenjapol, S., Haswell-Elkins, M., Reilly, P.E.B., Williams, D.J., Moore, M.R., 2003. A global perspective on cadmium pollution and toxicity in non-occupationally exposed pollution. Toxicol. Lett. 137, 65–83. Soares, M.E., Pereira, J.A., Bastos, M.L., 2006. Validation of a method to quantify copper and other metals in olive fruit by ETAAS. application to the residual metal control after olive tree treatments with different copper formulations. J. Agric. Food Chem. 54, 3923–3928. Soyergin, S., Katkat, A.V., 2002. Studies on nutrient contents and seasonal element fluctuations of the olive variety Gemlik in Bursa area. Acta Horticulturae 586, 405–407. Szteke, B., 2002. Trace elements in food. Food Add. Contam. 19, 905. Tassou, C.C., Panagou, E.Z., Katsaboxakis, K.Z., 2002. Microbiological and physicochemical changes of naturally black olives fermented at different temperatures and NaCl levels in the brines. Food Microbiol. 19, 605–615. Tasdemir, Y., Kural, C., Cindoruk, S.S., Vardar, N., 2006. Assessment of trace element concentrations and their estimated dry deposition fluxes in an urban atmosphere. Atmosp. Res. 81, 17–35. TSE, 2003. TS 774 Table olives. Turkish local food standards. Ankara, Turkey, 16 p. Uccella, N., 2001. Olive biophenols: novel ethnic and technological approach. Trends Food Sci. Tech. 11, 328–339. Unal, K., Nergiz, C., 2003. The effect of table olive preparing methods and storage on the composition and nutritive value of table olives. Grasas y Aceites 54, 71–76. Vavoulidou, E., Avramides, E.J., Papadopoulos, P., Dimirkou, A., 2004. Trace metals in different crop cultivation systems in Greece. Water Air. Soil. Poll. 4, 631–640. Yasar, B.S., Gücer, S., 2004. Fractionation analysis of magnesium in olive products by atomic absorption spectrometry. Analyt. Chim. Acta. 505, 43–49. Ziena, H.M.S., Youssef, M.M., Aman, M.E., 1997. Quality attributes of black olives as affected by different darkening methods. Food Chem. 60, 501–508.
Chapter 33
Olive Cultivar, Period of Harvest, and Environmental Pollution on the Contents of Cu, Cd, Pb, and Zn: Italian Perspectives Alberto Angioni Department of Toxicology, Food and Environmental Unit, University of Cagliari, Italy
33.1 INTRODUCTION Among the parameters to be monitored in biological matrices, in order to assess food hygiene and quality, is the analytical determination of heavy metals, such as Cd, Cu, Pb, and Zn (Underwood et al., 1987; Payne et al., 1998). Human activities have dramatically increased the rate of heavy metals in the environment, and these elements present in ground, water and in air may contaminate aliments, drinking water, and consequently humans (Payne et al., 1998). Data from the UNEP classify Italy in the first position in marine pollution for Cd, Cu, Pb, and Zn level (UNEP/MAP, 2003). The adsorption processes of heavy metals differ not only between plant species, but also between cultivars within one species. Moreover, the accumulation varies in the different parts of the plant, being lower in leaves than in roots and in fruits. Heavy metals cannot be degraded or destroyed in the environment and can enter our bodies via food, drinking water and air. Despite the fact that heavy metals are essential to maintain the metabolism of the human body (Underwood et al., 1987), in high concentrations they can lead to poisoning, interfering with many biochemical processes such as the Krebs cycle and steroid synthesis (Reijnders and Brausseur, 1997). Free heavy metals can also break molecular binding sites increasing the production of free radicals. The list of 14 toxic metals published by Morgan and Stumm (1991) includes: arsenic, lead, mercury, cadmium, chromium, nickel, zinc, antimony, silver, bismuth, indium, copper, selenium, tin, and thallium. The Agency for Toxic Substances and Disease Registry (ATSDR) of the United States of America published a list of 275 organic and inorganic substances hazardous for humans and the Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
environment (CERCLA Priority List of Hazardous Substances, 2007). Among the 20 most dangerous compounds, five were heavy metals: arsenic, lead, mercury, cadmium and chromium were at 1st, 2nd, 3rd, 7th, and 18th positions, respectively (Table 33.1). Data from the
307
TABLE 33.1 CERCLA Priority List of Hazardous Substances, 2007. 2007 Rank
Substance name
Total points
1
Arsenic
1672.58
2
Lead
1534.07
3
Mercury
1504.69
7
Cadmium
1324.22
18
Chromium, hexavalent
1149.98
53
Nickel
1005.40
74
Zinc
932.89
77
Chromium
908.52
128
Copper
804.86
214
Silver
612.19
219
Antimony
605.37
This table reported an abstract from CERCLA Priority List of Hazardous Substances, 2007. In the table are evidenced the rank position and the points of the heavy metal by the commission. Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
308
SECTION | I
World Health Organization (WHO), report that 60–70% of all acute and chronic diseases can be associated with heavy metal pollution (FAO/WHO, 1993). In Europe the Council Regulation (EEC) No 315/93 laid down Community procedures for contaminants in food, subsequently Commission Regulation (EC) No 1881/2006 defined the maximum levels of certain contaminants in foodstuffs, and the Commission Regulation (EC) No 333/2007 laid down the methods of sampling and analysis for the official control of the levels of lead, cadmium, mercury, in foodstuffs.
33.2 DISCUSSION The presence of heavy metals in olives may affect oil quality. For instance, trace levels of some heavy metals may
Metals, Electrolytes and Other Components
catalyze oxidation reaction, with deleterious effects on oil flavor and storage (Fedeli, 1968). Lead is the only heavy metal having a fixed maximum residue level (MRL) for olive oil (0.10 mg L⫺1; Commission Regulation (EC) No 1881/2006). Dugo et al. (2002) studied the presence of heavy metals in Sicilian olive oils. The average level found was below the MRL (0.09 mg L⫺1), but eight of the 34 samples analyzed exceeded the MRL, having maximum values of 0.28 mg L⫺1 (Figure 33.1). The higher levels of Pb were correlated with the use of organophosphoric pesticides. The literature on heavy metals on olives and olive oil mainly involves analytical methods for their determination in oils (Castillo et al., 1999; La Pera et al., 2002a, 2002b; O’Connor et al., 2002; Jimenez et al., 2003, 2004; Llbet et al., 2003; Lo Coco et al., 2003; Roca et al., 2004) but lacks in quantitation and in field experiments.
Section 3 : Metals Foodstuffs (1)
Maximum levels (mg kg−1 wet weight)
3.1
Lead
3.1.1
Raw milk (6), heat-treated milk and milk for the manufacture of milk-based products
0,020
3.1.2
Infant formulae and follow-on formulae
(4) (8)
0,020
3.1.3
Meat (excluding offal) of bovine animals, sheep, pig and poultry (6)
0,10
3.1.4
Offal of bovine animals, sheep, pig and poultry (6)
0,50
3.1.5
Muscle meat of fish (24) (25)
0,30
3.1.6
Crustaceans, excluding brown meat of crab and excluding head and thorax meat of lobster and similar large crustaceans (Nephropidae and Palinuridae) (26)
0,50
3.1.7
Bivalve molluscs (26)
1,5
3.1.8
Cephalopods (without viscera) (26)
1,0
3.1.9
Cereals, legumes and pulses
0,20
3.1.10
Vegetables, excluding brassica vegetables, leaf vegetables, fresh herbs and fungi (27). For potatoes the maximum level applies to peeled potatoes
0,10
Foodstuffs (1)
Maximum levels (mg kg−1 wet weight)
3.1.11
Brassica vegetables, leaf vegetables and cultivated fungi (27)
0,30
3.1.12
Fruit, excluding berries and small fruit (27)
0,10
3.1.13
Berries and small fruit (27)
0,20
3.1.14
Fats and oils, including milk fat
0,10
3.1.15
Fruit juices, concentrated fruit juices as reconstituted and fruit nectars (14)
0,050
3.1.16
Wine (including sparkling wine, excluding liqueur wine), cider, perry and fruit wine (11)
0,20 (28)
3.1.17
Aromatized wine, aromatized wine-based drinks and aromatized wine-product cocktails (13)
0,20 (28)
FIGURE 33.1 Lead limits on food and raw materials. This figure is an abstract from the annex to the Commission Regulation (EC) No 1881/2006 regarding the maximum levels of contaminants in foodstuffs.
CHAPTER | 33 Olive Cultivar, Period of Harvest, and Environmental Pollution on the Contents of Cu, Cd, Pb, and Zn
Among the experimentation that can be performed to assess olive oil contamination by heavy metals, there is its relation with the cultivar and the period of harvest. In Italy there is an unusual situation with different names for the same cultivar, depending on the region. This can create confusion also among the experts. Here the reported names are those used in Sardinia. Nine cultivars: Bosana, Pizz’e Carroga, Mallocrina, Manna, Nera di Gonnos, Nera di Oliena, Ogliastrina, Semidana, Tonda of Cagliari were tested to assess cultivar influence, and two harvests, the first when olives turn dark colored and the second after 1 month, were performed, a successive test using only the cultivar Bosana was carried out to evaluate the contamination in orchards at a different distance from an iron foundry. The data reported in Table 33.2 showed that the levels of cadmium and copper in olive oil from different cultivars are similar, with an average value of 5.3 μg L⫺1, and 4.3 μg L⫺1, respectively. Lead showed a significant variability, the oil from Nera di Oliena (10.3 μg L⫺1) showed the lowest amount, whereas the oil from Tonda di Cagliari (36.3 μg L⫺1) the highest. However, considering the MRL adopted for oil and fats (100 μg L⫺1) and the daily per capita consumption of oil (25 mL), all oils showed amounts of lead notably under the toxicological risk. Also zinc showed a high variability with minimum and maximum values obtained for Bosana (0.1 μg L⫺1) and Tonda di Cagliari (27.2 μg L⫺1), respectively. The time of harvest did not affect the content of lead, copper and cadmium in the analyzed oils (Table 33.3). The reported values are those recovered in olive oils from different cultivars. The comprehensive examination of the results showed no statistical differences for all heavy
309
metals related to olive growth. Some samples of Tonda di Cagliari showed higher values for lead and zinc in the first harvest compared to the second. Levels of zinc were for all cultivars higher in the first harvest than in the second (ripe olives), except for Ogliastrina and Semidana. The data from the ground composition in heavy metals of three soils (A, B, and C) at three different distances from a foundry, and the composition of the olive oils obtained from the relative olives are shown in Table 33.4. All orchards were of the cultivar Bosana. The data reported showed a positive correlation between soil concentration and oil composition for lead, zinc, and cadmium, while no influence of soil composition could be ascribed for Cu amounts. Copper showed the same contamination also with a soil content six times higher. All Sardinian olive oils analyzed showed a low content of heavy metals. Levels of lead were below MRLs fixed for olive oil and fats; zinc was the only metal with a huge variability within the first and the second harvest. Soil contamination was directly correlated to olive oil contamination for some heavy metals.
SUMMARY POINTS ●
●
A recent decree of the Italian Ministry of Agriculture requires that Italian producers show in the label the geographical zone where olives are grown as well as the location of the oil mill factory. The controls of the reliability of the olive oil labels are, at the moment, not based on any trustable scientific methodology.
TABLE 33.2 Influence of the cultivar. Cultivar
Cadmium
Copper
Lead
Zinc
Bosana
5.4 ⫾ 0.4
4.3 ⫾ 0.1
19.1 ⫾ 3.7
3.0 ⫾ 4.9
Mallocrina
5.4 ⫾ 0.2
4.3 ⫾ 0.0
19.6 ⫾ 1.2
2.2 ⫾ 2.0
Manna
5.3 ⫾ 0.3
4.2 ⫾ 0.2
14.7 ⫾ 1.6
2.2 ⫾ 2.3
Nera di Gonnos
5.3 ⫾ 0.4
4.4 ⫾ 0.1
17.5 ⫾ 5.4
6.4 ⫾ 6.2
Nera di Oliena
5.3 ⫾ 0.5
4.4 ⫾ 0.1
13.8 ⫾ 4.9
1.4 ⫾ 1.1
Ogliastrina
5.0 ⫾ 0.3
4.3 ⫾ 0.1
27.9 ⫾ 1.3
0.7 ⫾ 0.1
Pitz’e Carroga
5.8 ⫾ 1.2
4.6 ⫾ 0.3
17.7 ⫾ 4.3
4.8 ⫾ 6.6
Semidana
5.0 ⫾ 0.2
4.3 ⫾ 0.2
14.1 ⫾ 1.4
1.5 ⫾ 0.9
Tonda di Cagliari
5.3 ⫾ 0.3
4.3 ⫾ 0.2
22.7 ⫾ 10.4
10.3 ⫾ 12.8
This table shows the residues level (μg L⫺1 ⫾ S.D.) of the four heavy metals studied in olive oils of different cultivars. Cd and Cu showed no differences, while lead and zinc showed a huge variability with average maximum and minimum levels of 13.8 and 27.9 μg L⫺1 for lead, and 0.7 and 10.3 μg L⫺1 for zinc.
310
SECTION | I
Metals, Electrolytes and Other Components
TABLE 33.3 Influence of the time of harvest. Cultivar Time of harvest
Lead 1st
Copper 2nd
1st
Cadmium 2nd
1st
Zinc
2nd
1st
2nd
Bosana
19.3 ⫾ 4.1
18.8 ⫾ 3.5
4.4 ⫾ 0.1
4.3 ⫾ 0.2
5.4 ⫾ 0.5
5.3 ⫾ 0.3
4.4 ⫾ 6.3
1.3 ⫾ 1.2
Mallocrina
18.7 ⫾ 3.9
20.4 ⫾ 2.7
4.3 ⫾ 0.0
4.3 ⫾ 0.1
5.5 ⫾ 0.2
5.2 ⫾ 0.2
3.6 ⫾ 0.1
0.8 ⫾ 0.0
Manna
13.6 ⫾ 2.6
15.8 ⫾ 2.1
4.3 ⫾ 0.1
4.0 ⫾ 0.0
5.5 ⫾ 0.2
5.1 ⫾ 0.1
3.8 ⫾ 0.1
0.5 ⫾ 0.0
Nera di Gonnos
17.4 ⫾ 6.1
17.6 ⫾ 6.6
4.4 ⫾ 0.1
4.5 ⫾ 0.1
5.3 ⫾ 0.5
5.2 ⫾ 0.3
4.8 ⫾ 5.1
4.2 ⫾ 5.1
Nera di Oliena
10.3 ⫾ 2.1
17.2 ⫾ 4.3
4.3 ⫾ 0.0
4.4 ⫾ 0.1
4.9 ⫾ 0.2
5.6 ⫾ 0.4
2.2 ⫾ 0.0
0.6 ⫾ 0.0
Ogliastrina
29.7 ⫾ 2.1
27.9 ⫾ 3.4
4.2 ⫾ 0.1
4.3 ⫾ 0.1
4.8 ⫾ 0.1
5.2 ⫾ 0.3
0.6 ⫾ 0.0
0.7 ⫾ 0.0
Pitz’e Carroga
19.3 ⫾ 4.5
15.6 ⫾ 3.3
4.5 ⫾ 0.2
4.6 ⫾ 0.4
6.1 ⫾ 1.5
5.4 ⫾ 0.6
5.9 ⫾ 8.6
3.1 ⫾ 2.1
Semidana
13.3 ⫾ 0.1
15.7 ⫾ 1.9
4.3 ⫾ 0.2
4.3 ⫾ 0.1
5.0 ⫾ 0.2
5.1 ⫾ 0.2
1.1 ⫾ 0.7
2.4 ⫾ 0.1
Tonda di Cagliari
28.0 ⫾ 12.8
17.4 ⫾ 4.9
4.4 ⫾ 0.3
4.1 ⫾ 0.1
5.6 ⫾ 0.3
5.1 ⫾ 0.1
18.4 ⫾ 14.5
2.2 ⫾ 0.2
This table reported the influence of olive ripening on heavy metal amounts (μg L⫺1 ⫾ S.D.). The reported values are those recovered in olive oils from different cultivars. The comprehensive examination of the results showed no statistical differences for all heavy metals related to olive growth, except for zinc which generally showed values lower in ripe olives at the 2nd harvest.
TABLE 33.4 Influence of ground composition. A Metal
B
C
Soil
Oil
Soil
Oil
Soil
Oil
mg kgⴚ1
μg Lⴚ1
mg kgⴚ1
μg Lⴚ1
mg kgⴚ1
μg Lⴚ1
Cu
120.2
70.0
19.9
96.0
22.4
87.0
Zn
255.3
98.0
96.2
64.0
58.8
53.0
Pb
1038.9
270.0
96.9
26.0
23.2
36.0
Cd
13.6
3.0
1.0
⬍0.01
0.6
⬍0.01
This table reported the ground composition in heavy metals of three soils (A, B, and C) and the composition of the related olive oils, obtained from olive orchards grown on the corresponding soils. All orchards were of the cultivar Bosana. The data reported showed a positive correlation between soil concentration and oil composition for lead, zinc and cadmium, while no influence of soil concentration could be ascribed for Cu.
●
●
●
●
The metabolomic approach considers the evaluation of trace compounds present in the oil aroma originating from the lipoxygenase (LOX) cascade. The metabolomic–LDA approach was applied in the evaluation of the origin of extra virgin olive oils produced in different zones of Calabria region, in Italy. It can be assumed that the trace elemental distribution in olive oils varies according to their origin. Trace elements–LDA approach shows a clean separation of the production areas of Rossano, Andria, Lamezia, Spoleto and Pescara.
REFERENCES Castillo, J.R., Jimenez, M.S., Ebdon, L., 1999. Semiquantitative simultaneous determination of metals in olive oil using direct emulsion nebulization. J. Anal. Atom. Spectrom. 14 (9), 1515–1518. Dugo, G., Lo Curto, S., Lo Turco, V., La Torre, L., Salvo, F., 2002. Valutazione del contenuto di Cu(II), Zn (II), Cd (II) e Pb (II) in oli di oliva prodotti nella valle del belice. Riv. Ital. Sostanze Grasse 79, 157–160. CERCLA Priority List Of Hazardous Substances, 2007. Commission Regulation (EC) No 333/2007. Commission Regulation (EC) No 1881/2006.
CHAPTER | 33 Olive Cultivar, Period of Harvest, and Environmental Pollution on the Contents of Cu, Cd, Pb, and Zn
Council Regulation (EEC) No 315/93. FAO/WHO, 1993. Evaluation of Certain Food Additives and Contaminants. Technical Report, series 837, World Health Organization, Geneva. Fedeli, E., 1968. In “Metals Catalysed Lipid Oxidation”. Marcuse, R. (ed.), SIK, Goteborg, Sweden, p. 105. Gaines, P., 2004. ICP Operations. http://www.ivstandards.com/tech/ icp-ops/part10.asp. Jimenez, M.S., Velarte, R., Gomez, M.T., Castillo, J.R., 2004. Multielement determination using on-line emulsion formation and ICP-MS/FAAS for the characterization of virgin olive oils by principal component analysis. At. Spectrosc. 25 (1), 1–12. Jimenez, M.S., Velarte, R., Castillo, J.R., 2003. On-line emulsions of olive oil samples and ICP-MS multi elemental determination. J. Anal Atom. Spectrom. 18 (9), 1154–1162. La Pera, L., Lo Curto, S., Visco, A., La Torre, L., Dugo, G., 2002a. Derivative potentiometric stripping analysis (dPSA) used for the determination of cadmium, copper, lead, and zinc in Sicilian olive oils. J. Agric. Food Chem. 50 (11), 3090–3093. La Pera, L., Lo Coco, F., Mavrogeni, E., Giuffrida, D., Dugo, G., 2002b. Determination of copper (II), lead (II) and zinc (II) in virgin olive oils produced in Sicily and Apulia by derivative potentiometric stripping analysis. Ital. J. Food Sci. 14 (4), 389–399. Llbet, J.M., Falco, G., Casa, C., Teixido, A., Domingo, J.L., 2003. Concentrations of arsenic, cadmium, mercury, and lead in common foods and estimated daily intake by children, adolescents, adults, and seniors of catolina, Spain. J. Agric. Food Chem. 51, 838–842.
311
Lo Coco, F., Ceccon, L., Circolo, L., Novelli, V., 2003. Determination of cadmium(II) and zinc(II) in olive oils by derivative potentiometric stripping analysis. Food Control 14, 55–59. Morgan, J.J., Stumm, W., 1991. Chemical processes in the environment, relevance of chemical speciation. In: Merian, E. (ed.), Metals and Their Compounds in the Environment. VCH, Weinheim, pp. 67–103. O’Connor, G., Rowland, S.J., Evans, E.H., 2002. Evaluation of gas chromatography coupled with low pressure plasma source mass spectrometry for the screening of volatile organic compounds in food. J. Sep. Sci. 25 (13), 839–846. Payne, J.F., Alins, D.C., Gunselman, S., Rahimtula, A., Yeats, P.A., 1998. DNA oxidative damage and vitamin: a reduction in fish from a large lake system in Labrador, Newfoundland, contaminated with iron-ore mine tailings. Mar. Environ. Res. 46, 289–294. Reijnders, P.J.H., Brausseur, S.M.J.M., 1997. Xenobiotic induced hormonal and associated developmental disorders in marine organisms and related effects in humans. An overview. J. Clean Technol. Invironm. Toxycol. Occup. Med. 6, 367–380. Roca, A., Cabrera, C., Lorenzo, M.L., Lopez, M.C., 2004. Levels of calcium, magnesium, manganese, zinc, selenium and chromium in olive oils produced in Andalusia. Grassa y Aceites 51 (6), 393–399. Underwood, E., Mertz, W., 1987. Trace elements needs and tolerances. In: Mertz, W. (ed.), Trace Elements in Human and Animal Nutrition. Academic Press, London, pp. 11–19. UNEP/MAP Assessment of Transboundary Pollution Issues in the Mediterranean Sea, 2003. Athens: UNEP/MAP, 292 p. Mediterranean shelved at: 153/K2003/MED.POL.
This page intentionally left blank
Chapter 34
Trace Components in Italian Virgin Olive Oils Giovanni Sindona and Antonio Tagarelli Dipartimento di Chimica, Università della Calabria, Italy
34.1 INTRODUCTION The production of olive oil in Italy reaches 500 000 tonnes per year, ca. 70% of which is represented by high-quality extra virgin foodstuff, worth 2.2 billion euros, approximately. Consumers of Italian olive oil are spread worldwide and, obviously, they trust the information reported in the label. Up to the beginning of the year 2008 the labeling system approved by European directives did not include the origin of the olives used for preparing the oil. A recent decree of the Italian Ministry of Agriculture requires that Italian producers show on the label the geographical zone where the olives were grown as well as the location of the oil mill factory. A warning was issued by the appropriate European Commission against the application of the Italian decree, which is highly supported by consumer and producer associations. The controls of the reliability of the olive oil labels are, at the moment, not based on any trustable scientific methodology.
was correlated to monovarietal oils produced from drupes at different ripening stages (Maestro Duran, 1990; Angerosa et al., 1996; Morales et al., 1996; Aparicio and Morales, 1998). The metabolomic–LDA approach was applied in the evaluation of the origin of extra virgin olive oils produced in different zones of Calabria region, in Italy (Figure 34.1). Samples from different production areas are grouped in quite distinct different zones of the plot. An approach which could be less affected by the cultivar type as well as the ripening stage of the olives being processed should be the correlation of the origin with the composition of trace elements.
16 14 12 10
34.2 DISCUSSION
8
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
6 Rad. 2
Two innovative procedures have recently been presented, aiming at the creation of free databases for both dealers and consumers enabling the self checking of the identity of this particular foodstuff (Benincasa et al., 2007; Cavaliere et al., 2007). Both approaches are based on the application of the statistical method known as linear discriminant analysis (LDA), whereby the classification of the origin of unknown olive oils is performed after having verified the possible differences among samples of known origin. The metabolomic approach considers the evaluation of trace compounds present in the oil aroma originating from the lipoxygenase (LOX) cascade (Benincasa et al., 2003), the principal components of this aerobic secondary metabolism being mainly represented by aliphatic C6 species (Guth and Grosh, 1991; Morales et al., 1996; Angerosa et al., 1999, 2000). The distribution of the same biomarkers
4 2 0 −2 −4 −6 −8 −10 −15
−10
−5
0
5
10
15
20
25
Rad. 1 Nocera terinese
Catanzaro
Rende
Corigliano
FIGURE 34.1 LDA plot of 16 olive oil samples based on quantitative values of the five selected compounds. Oil samples from different areas of production are growing in distinct zones of the plot.
313
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
314
SECTION | I
Metals can be incorporated into the oil from the soil or be introduced during the manufacturing of the foodstuff. Therefore, it can be assumed that the trace elemental distribution in olive oils varies according to their origin and then it can be supposed that a suitable statistical treatment on trace element data could allow a geographical characterization of different olive oils. The analysis has been applied to the concentration of 18 elements of each single sample present in Table 34.1, and using eight groups as input a priori.
Metals, Electrolytes and Other Components
The bidimensional plot of the first two roots shows a clean separation of the production areas of Rossano, Andria, Lamezia and Spoleto-Pescara (Figure 34.2), on the basis of the first root, only. The plot in Figure 34.2 suggests that the trace element distribution is mainly a function of the geographical zone. A deeper analysis of the obtained variance, performed with the ANOVA test, has provided statistical parameters allowing one to rule out the contribution of the cultivar on the cluster separation is negligible.
TABLE 34.1 Distribution of 18 elements in different oils, determined by Inductively coupled plasma mass spectrometry.
RSD(%)
Carolea
RSD(%)
Coratina
RSD(%)
Carolea
RSD(%)
Coratina
Coratina
RSD(%)
Be
0.118
0.8
0.119
1.7
0.178
50.6
0.146
63.7
0.118
0.8
0.181
51.4
n.d.
0.182
50.5
Mg
1032
13.7
446
15.9
467
26.8
651.6
0.4
463.2
1.1
319.9
20.5
267.6
26.0
56
35.7
Ca
12 220
12.7
14 969
22.8
9797
50.5
26 887 63.4
1853
2.7
5757
34.6
3274
35.2
5679
24.2
Sc
63.2
17.1
66.1
17.2
747.9
0.6
80.9
21.8
57.8
33.9
54.9
19.9
52.7
31.9
49.94
17.1
Cr
309
30.7
368
3.0
182.7
2.0
437.4
18.6
281.2
68.4
256
43.4
116.49
33.7
138
79.0
Mn
10.3
35.9
12.78
2.0
5.33
118
25.2
66.3
8.47
77.7
4.4
53.0
n.d.
9.4
51.0
Fe (Standard)
1614
44.0
1348
30.0
463.5
2.7
1284
2.3
662
46.5
1099
82.5
710
41.7
593.1
5.1
Fe (DRC)
404
30.9
246.6
20.9
458.9
5.4
550.9
4.9
203.1
36.6
157
60.5
89.3
25.6
96.6
11.4
Co
0.413
31.0
0.173
19.7
0.404
33.7
0.318
16.7
0.056
32.1
0.226
12.8
0.023
21.7
0.084
39.3
Ni
46.9
5.5
39.9
29.8
15.0
50.7
37.4
4.8
10.6
47.2
25.5
59.2
n.d.
As
9.9
17.2
6.04
34.4
26.65
1.2
12.24
21.2
1.248
7.5
7.74
66.7
1.72
18.6
4.89
85.7
Se
6.24
9.9
5.96
16.1
4.06
45.8
6.78
5.9
5.33
78.8
2.13
42.3
1.47
24.5
2.27
32.6
Sr
24.49
0.2
23.49
26.6
19.5
55.4
48.9
82.2
1.52
30.3
n.d.
n.d.
13.84
20.7
Y
0.112
0.9
0.136
3.7
0.135
8.1
0.331
81.9
0.082
79.3
0.092
65.2
n.d.
0.107
41.1
Cd
0.228
2.6
0.327
17.4
0.145
3.4
0.366
19.7
0.236
72.5
0.235
22.1
0.088
12.5
0.184
72.8
Sb
0.338
10.9
0.229
70.7
0.25
18.4
0.242
10.3
0.233
73.4
0.411
77.9
0.194
33.0
0.253
13.8
Sm
0.023
0
0.019
47.4
0.226
1.8
0.042
28.6
0.004
25.0
0.047
4.3
0.019
73.7
0.027
3.7
Eu
0.007
28.6
0.013
15.4
0.004
75.0
0.021
23.8
0.01
70.0
0.008
12.5
n.d.
0.006
33.3
Gd
0.023
60.9
0.031
9.7
0.094
72.3
0.047
17.0
0.003
33.3
0.023
69.6
0.004
0.015
60.0
The amounts of each element are given in µg kg⫺1.
RSD(%)
Carolea
Pescara
RSD(%)
Spoleto
Carolea
Andria
RSD(%)
Lamezia
Coratina
Rossano
n.d.
25.0
315
CHAPTER | 34 Trace Components in Italian Virgin Olive Oils
On the ground of these considerations, the linear discriminating analysis was, therefore, performed on the concentrations of the selected 18 elements using as input a priori the five groups corresponding to the oil-producing areas only (Figure 34.3). The stepwise approach generated Rad. 1 vs. Rad. 2 35
a plot in which it is possible to clearly distinguish the five clusters corresponding to the production regions of Rossano, Andria, Lamezia, Spoleto and Pescara, respectively. It can be confidently assumed that the selection of appropriate biomarkers and, among these, the trace element distribution represents a valid method to assess the origin of olive oils based on the application of statistical methods using mass spectrometric methodologies.
30 25
SUMMARY POINTS
20 ●
Rad. 2
15
●
10 ●
5 0
●
−5 ●
−10
The control of the reliability of olive oil labels. The metabolomic approach in the evaluation of trace compounds. The metabolomic–LDA approach in the evaluation of the origin of extra virgin olive oils. Trace elemental distribution in olive oils varies according to their origin. Trace elements–LDA approach shows a clear separation of the different Italian production areas.
−15 −20 −50
−40
−30
−20
−10
0
10
20
30
REFERENCES
Rad. 1 Rossano Coratina
Anria Carolea
Anria Coratina
Lamezia Carolea
Spoleto Carolea
Spoleto Coratina
Pescara Carolea
Pescara Coratina
FIGURE 34.2 LDA plot for 36 olive oil samples based on concentration of 18 elements and using eight groups as input a priori. The matching of olive oils from two different cultivars produced in five different zones shows a significant differentiation of production zones. Cultivar is not a parameter to disaggregate the samples. Rad. 1 vs. Rad. 2 30 25 20
Rad. 2
15 10 5 0 −5 −10 −15 −30
−20
−10
0
10
20
Spoleto
Pescara
30
Rad. 1 Rossano
Andria
Lamezia
FIGURE 34.3 LDA plot for 36 olive oil samples based on concentration of 18 elements and using as input a priori five groups, i.e. those corresponding to the oil-producing areas. The plot shows that the elimination of the cultivar parameter allows an easy distinction of the production areas.
Cavaliere, B., De Nino, A., Hayet, F., Lazez, A., Macchione, B., Moncef, C., Perri, E., Sindona, G., Tagarelli, A., 2007. A metabolomic approach to the evaluation of the origin of extra virgin olive oil: a convenient statistical treatment of mass spectrometric analytical data. J. Agric. Food Chem. 55, 1454–1462. Benincasa, C., Lewis, J., Perri, E., Sindona, G., Tagarelli, A., 2007. Determination of trace element in Italian virgin olive oils and their characterization according to geographical origin by statistical analysis. Anal. Chimica Acta 585, 366–370. Benincasa, C., De Nino, A., Lombardo, N., Perri, E., Sindona, G., Tagarelli, A., 2003. Assay of aroma active components of virgin olive oils from southern Italian regions by SPME-GC/ion trap mass spectrometry. J. Agric. Food Chem. 51, 733–741. Angerosa, F., Basti, C., Vito, R., 1999. Virgin olive oil volatile compounds from lipoxygenase pathway and characterization of some Italian cultivars. J. Agric. Food Chem. 47, 836–839. Guth, H., Grosh, W., 1991. A comparative study of the potent odorants of different virgin olive oils. Fat Sci. Technol. Fat Sci. Technol. 93, 335–339. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2000. Virgin olive oil odour notes: their relationships with volatile compounds from the lipoxygenase pathway and secoiridoid compounds. Food Chem. 68, 283–287. Morales, M.T., Calvente, J.J., Aparicio, R., 1996. Influence of olive ripeness on the concentration of green aroma compounds in virgin olive oil. Flavour Fragrance J. 11, 171–178. Aparicio, R., Morales, M.T., 1998. Characterization of olive ripeness by green aroma compounds of virgin olive oil. J. Agric. Food Chem. 46, 1116–1122. Angerosa, F., Di Giacinto, L., Basti, C., Serraioco, A., 1996. Influenza della variabile “ambiente” sulla composizione degli oli vergini di oliva. Riv. It. Sost. Grasse 73, 461–467. Maestro Duran, R., 1990. Relationship between the composition and ripening of the olive oil and quality of the oil. Acta Hort. 286, 441–451.
This page intentionally left blank
Chapter 35
Inorganic Anions in Olive Oils: Application of Suppressed Ion Exchange Chromatography (IEC) for the Analysis of Olive Oils Produced from De-stoned Olives and Traditional Extraction Methods Lara La Pera1, Teresa Maria Pellicanò2, Pellicano Vincenzo Lo Turco1, Giuseppa Di Bella1 and Giacomo Dugo1 1 2
Department of Food and Environmental Science, University of Messina, Messina, Italy Department of Chemistry (cube 12th), University of Calabria (UNICAL), Arcavacata of Rende-Cosenza, Italy
35.1 INTRODUCTION Olive oil (OO) plays a vital role in the Mediterranean diet and in recent years a great of importance has been addressed not only to its organic composition, but also to the presence of inorganic species including heavy metals and anions. The presence of inorganic elements in vegetable oils depends on many factors, such as type of soil, climatic condition, fruit maturity, use of pesticides, extraction procedures, storage and technological factors (La Pera and Dugo, 2005). It’s important to determine trace levels of some inorganic anions including fluoride, chloride, bromide, nitrate, nitrite, sulfate, or phosphate because they are naturally present in the crude matter and may also be introduced during industrial manipulations. (Buldini et al., 1997a,b; Lopez-Ruiz, 2000). Therefore the analysis of inorganic anion olive samples is important from the nutritional, toxicological and technological points of view. Ion exchange chromatography (IEC) was introduced in 1975 as a chromatographic method for the determination of inorganic ions, which consisted of a low ion-exchange capacity resin as the stationary phase, a conductimeter as detector and a suppressor column to increase separation speed and analytical response. Suppressed conductivity detection requires the use of eluents like sodium hydroxide or carbonate-bicarbonate buffers that can be converted into species of low conductance like water or H2CO3 after exchanging the cations of the eluent for hydrogen ions by a suitable cation-exchange device. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Nowadays IC has become a routine analytical method for the determination of inorganic ions in various matrices including food samples. In recent years the anion-exchange columns have been improved through the use of macroporous substrates with chemically grafted ion-exchange sites which reduce the interferences from organic anions, thus allowing the determination of inorganic anions in very complex matrices, such as food samples (Table 35.1). Also the sample preparation procedure is of great importance when an IEC method is developed for inorganic species analysis in food samples. The traditional sample preparation procedures are based on wet digestion, dry ashing or alkaline fusion. These techniques are time consuming, require a high consumption of reagents, and often result in solutions that are easily contaminated from reagents involved in the sample preparation procedure. Buldini et al. have reported the simultaneous analysis of chloride, phosphate and sulfate in vegetable fats by IEC after saponification and UV photolysis (Buldini et al., 1997a). This method is sensitive and accurate but does not allow the determination of fluoride, nitrite, nitrate, bromide and iodide; furthermore the sample preparation procedure takes about 1.5 h. This study describes the use of ion exchange chromatography (IEC) with conductivity detection and chemical suppression for the simultaneous analysis of F⫺, Cl⫺, Br⫺, NO2⫺, NO3⫺, PO43⫺, SO42⫺ and I⫺ in olive oil samples, after extraction with hot water applying ultrasound (Dugo et al., 2007). Furthermore this study aims to bring some novelty
317
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
318
SECTION | I
Metals, Electrolytes and Other Components
TABLE 35.1 Key facts for Ion Exchange Chromatography (IEC). Ion exchange chromatography (IEC) was introduced in 1975 as a chromatographic method for the determination of inorganic ions, which consisted of a low ion-exchange capacity resin as the stationary phase, a conductimeter as detector and a suppressor column to increase separation speed and anaytical response Nowadays IEC has become a routine analytical method for the determination of inorganic ions, in various matrices including food samples Suppressed conductivity detection requires the use of eluents like NaOH or carbonate-bicarbonate buffers that can be converted into species of low conductance like water or H2CO3 after exchanging the cations of the eluent for hydrogen ions by a suitable cationexchange device The modern anion-exchange columns have been improved through the use of macroporous substrates with chemically grafted ion-exchange sites which reduce the interferences from organic anions, thus allowing the determination of inorganic anions in very complex matrices as food samples
concerning the inorganic composition of olive oils obtained both from the traditional extraction method (whole olives) and from de-pitted fruits since there are many publications about the influence of stone removing on olive oil quality, with no regard to its inorganic composition (Angerosa et al., 1999; Saitta et al., 2003).
35.2 INORGANIC ANIONS: NUTRITIONAL AND TOXICOLOGICAL ASPECTS Fluoride is a trace element found widely in the environment. Animals contain relatively small amounts of fluorides; however, some plants can accumulate fluorides in their leaves. A certain daily intake of fluoride is necessary for skeletal bone integrity, but excess consumption causes toxicity (fluorosis). An adequate intake level has been set for fluoride at 3 milligrams (mg) daily for women and 4 mg daily for men (USDA, 2004). However, until now, scant data existed on the quantity of fluoride in the national food supply, meaning the actual level of the mineral being consumed by individuals is relatively unknown. Good food sources include tea and fish. Drinking water can be a rich source and we also take in some fluoride when we use products such as toothpaste and mouthwash with added fluoride. Chloride is an essential element and is the main extracellular anion in the body. It is also one of the most common inorganic anions in foods and in the form of NaCl is employed as a preservative, therefore its determination in foods is essential to fulfill legal regulations and to meet quality control requirements. Iodine is found in foods as iodide. It is important for essential hormone development in the human body. Inadequate intake of dietary iodine can lead to an enlarged thyroid gland (goiter) or other iodine-deficiency disorders. Iodine is found in seawater, so any type of seafood is a
rich source of this element. The RDA was set at 120 μg for women and 150 μg for men. Bromide, nitrite and nitrate are all used as food additives, of all the studied inorganic anions they are the most important in terms of food chain contamination because of their potential toxicity. Potassium bromate has long been used to increase the volume of bread and to produce bread with a fine crumb (the non-crust part of bread) structure. Most bromate rapidly breaks down to form innocuous bromide. However, bromate itself causes cancer in animals. The small amounts of bromate that may remain in bread pose a small risk to consumers. Bromate can be also found as a disinfection byproduct of the ozonation of drinking water derived from source water containing bromide. The bromide in the source water is oxidized to bromate by the ozone. It is a potential 21 carcinogen to rats and mice at mg L⫺1 levels. New toxicological studies have led the International Agency for Research on Cancer (IARC) to classify bromate as a group 2B carcinogen to humans with renal tumor risks at concentrations of 0.05 μg L⫺1. High bromide values in vegetable food may derive from soil fumigation with methyl-bromide which is used to combat nematodes, soilborne fungi, phytophtora, weeds and diseases which attack roots and thus destroy and kill the plants. Sodium nitrite and sodium nitrate are two closely related chemicals used for centuries to preserve meat. While nitrate itself is harmless, it is readily converted to nitrite. The reduction in the human intestine of nitrate to nitrite can induce metahemoglobinemia and under physiological conditions, nitrite reacts with secondary amines forming carcinogenic nitrosamines (Pennington, 1998). Nitrite, as a food additive, serves a dual purpose in the food industry since it both alters the color of preserved fish and meats and also prevents growth of Clostridium botulinum, the bacteria which causes botulism.
319
CHAPTER | 35 Inorganic Anions in Olive Oils
TABLE 35.2 Key facts for inorganic anions. The halides fluorides, iodides and particularly chlorides are widely diffused in the environment and are present in human body playing many important rules Bromide, nitrite and nitrate are all used as food additives; of all the studied inorganic anions, they are the most important in terms of food chain contamination because of their potential toxicity. Phosphorous compounds are present in most foods. Inorganic phosphates are used as fertilizers; phosphate and condensed phosphate compounds are used as food acidulants. The presence of sulfate in food is usually related to the remarkable use of sulfur and its compounds (copper sulfate) in agriculture, due to their good efficacy and low toxicity, as pesticides It’s important to determinate trace levels of some inorganic anions in food as fluoride, chloride, bromide, nitrate, nitrite, sulfate, or phosphate because they are naturally present in the crude matter and may also be introduced during industrial manipulations The presence of inorganic elements in vegetable oils depends on many factors as the type of soil, climatic condition, fruits maturity, use of pesticides, extraction procedures, storage and technological factors
Phosphorus is one of the primary nutrients essential for plant growth and crop production; it occurs in nature as phosphate. Phosphorous compounds are present in most foods. Inorganic phosphates are used as fertilizers; phosphate and condensed phosphate compounds are used as food acidulants. Polyphosphates are widely used as additives in meat, fruit and cheese. The presence of sulfate in food is usually related to the remarkable use of sulfur and its compounds (copper sulfate) in agriculture, due to the good efficacy and low toxicity, as pesticides. The key facts for inorganic anions are given in Table 35.2.
35.3 OLIVE OILS FROM DE-STONED OLIVES AND TRADITIONAL EXTRACTION METHODS For experimental tests olives belonging to Nocellara del Belice variety harvested in olive groves of the same area near Trapani (Sicily-Italy) during the crop year 2001–2002 were used. Harvesting was done by hand using rakes. An homogeneous batch of olives (about 600 kg) were divided into three lots (A, B, C) and taken to an industrial oil mill where a laboratory mill was used to extract oil. Two lots (A, B) of the de-leafed olives were subjected to stone removal using a suitable tool (de-stoner composed from a cylindrical perforated stationary grill and a rotary shaft), crushed and placed into a water-saving decanter with variable speed of the conveyor, equipped with sensors and automatic adjustment of process parameters: process water volume, oil paste discharge, and differential speed of the conveyor respect to the bowl. Both pastes from de-stoned fruit and from the whole fruits were malaxed at 27°C for 30 minutes. Lots A, B, and C also differed for the processing water volume added to the paste entering the decanter (expressed as % with respect to the pulp weight), 21.2%,
28.1% and 31.6% respectively. De-leafed fruits from lot C were submitted, without further treatments, to the extraction process using the same plant. Each lot yielded about 57–59 kg of olive oil which was bottled in dark glass bottles. Particularly 20 samples of olive oil from lots A, 22 from lot B and 25 from lot C were analyzed. Furthermore, commercial refined seed oils produced in Italy were studied. Peanut oils (n ⫽ 5), sunflower oils (n ⫽ 5), soy oils (n ⫽ 5), maize oils (n ⫽ 5) were analyzed. All the studied seed oils were purchased from a supermarket on February 2003 and were stored in cans. All the studied vegetable oil samples were stored at ⫺20°C till the analysis.
35.4 METHODOLOGICAL CONSIDERATIONS 35.4.1 A New Approach for Inorganic Anion Extraction from Fat Matrices The determinations of inorganic elements such as fluoride, chloride, bromide, nitrite, nitrates, phosphates and sulfates in vegetable oils is of great concern both from the technological and the toxicological points of view. Testing for inorganic anions in edible oils has always been very problematic, because they are present in small quantities within a complex organic matrix. Sample preparation is a critical step in the whole analytical procedure. Ion exchange chromatography is the most effective method for anion analysis owing to its precision, high sensitivity, rapidity coupled with the advantage of simultaneous determinations. In 1997 Buldini, Cavalli and Trifirò, developed a procedure for the ion chromatographic determination of chloride, phosphates and sulfates in edible vegetable oil; the sample pretreatment was based on saponification
320
followed by oxidative UV photolysis which took about three hours. The saponification-photolytic pretreatment was accurate and achieved a high sensitivity, but it did not allow the determination of fluoride, nitrite, bromide and iodide. This chapter describes the use of IEC with conductivity detection and chemical suppression, for the simultaneous analysis of fluoride, chloride, bromide, nitrite, nitrate, phosphate, sulfate and iodide in vegetable oil after extraction in a pH 8 carbonate/hydrogen carbonate buffered solution at 70°C, applying ultrasound. In particular, the extraction procedure was carried out as follows: a 0.5 g aliquot of olive oil and 25.0 mL of HPLC grade water (previously subjected to a blank analysis) pH 8 with carbonate (3 mM)/hydrogen carbonate (3 mM) buffer, were introduced in a teflon beaker fitted with PTFE stoppers to avoid sample loss and contamination. The extraction was carried out for about 30 minutes under magnetic stirring at the temperature of 70°C. The mixture was then placed in an ultrasonic bath (Soltec, Milan, Italy) operating at the frequency of 50 KHz at 70°C for 10 minutes, and after cooling it was centrifuged at 1000 rpm for 5 minutes; the aqueous phase was directly collected into a 50 mL volumetric flask and the oily one was extracted again as previously described. Aqueous phase was again recovered and added to the first extract, up to the mark with ultra-pure water. In order to remove organic residues, the samples were filtered through 0.2 μm glassmicrofiber chromatographic filter, before the IEC analysis. This extraction procedure employs the principles of traditional solvent extraction, exploiting the greater affinity of inorganic anions towards an aqueous medium in respect to an oily one. The aqueous extract was injected into a Compact IC 761 ion chromatograph (Metrohm, Switzerland) equipped with a double piston pump, a thermostatted (20°C) conductivity detector and a continuously regenerable suppressor containing three cartridges which are in turn used for suppression, regenerated with 20 mM H2SO4 and rinsed with distilled water. Anion separation was achieved with a Metrosep Anion Dual 1 column (3.0 ⫻ 150 mm; 10.0 μm) and ion guard column, both packed with quaternary ammonium polymethacrylate. The isocratic elution was carried out using a solution of carbonate (3.12 mM)/hydrogen carbonate (3.25 mM)/2% acetone. The flow rate was 0.50 mL min⫺1. Samples were injected using a 20 μΛ loop injector.
35.4.2 Validation of the Ion Exchange Chromatographic Method for Inorganic Anions Analysis in Vegetable Oils The method described was subjected to a validation process in order to prove that it is acceptable for its intended purpose. In particular, studies on specificity, linearity, accuracy,
SECTION | I
Metals, Electrolytes and Other Components
precision, range, detection limit and quantitation limit were performed. For chromatographic methods, developing a separation involves demonstrating specificity, which is the ability of the method to accurately measure the analyte response in the presence of all potential sample components. The response of the analyte in real samples (containing the analyte and all potential sample components), is compared with the response of a solution containing only the analyte. The comparison of the chromatogram standard oily mixture containing 10.0 mg kg⫺1 of each assayed anion (Figure 35.1A) with that of an olive oil extract (Figure 35.1B), gave evidence that no co-elution occurred and a good resolution and peak purity were obtained in the chromatographic conditions described. The linearity test, which allows one to obtain the calibration graph for the quantitative analysis, was performed by preparing standard solutions of anions at five concentration levels, from 0.1 to 100 mg kg⫺1. In this concentration range good linearity was obtained (R2 ⫽ 0.9991) for all the studied anions (Table 35.3). The injection precision of the method, expressed as the rsd% range of nine measurements performed on an anion free olive oil spiked at different concentration levels (0.25, 0.5, 1.0, 10 mg kg⫺1), for a total of 36 runs, was within 6.5% (Table 35.3). The limit of detection was calculated according to the signal-to-noise ratio approach: S/N was 3:1 for LOD and 10:1 for LOQ, proportions usually acceptable for these limits (Winefordner, 1983). The method described shows excellent sensitivity for fluorides, chlorides, nitrite and bromide, in fact concentrations lower than 10 μg kg⫺1 can be detected; for all the other studied anions the LODs were within 31 μg kg⫺1. The limits of quantification ranged from 28.1 to 103.3 μg kg⫺1 (Table 35.3). The accuracy of a method is the closeness of the measured value to the true value for the sample. The best way to assess accuracy is by analyzing standard reference matrices, since edible oil or fat standards certified for inorganic anions were not found in commerce. The accuracy was evaluated using an anion free standard olive oil spiked at different concentration levels (0.25, 0.5, 1.0, 10 mg kg⫺1) with F⫺, Cl⫺, Br⫺, NO2⫺, NO3⫺, PO43⫺, SO42⫺ and I⫺ sodium salts and left under magnetic stirring overnight in order to obtain a completely homogeneous mixture which was then subjected to the extraction procedure described and analyzed by IEC. The accuracy of the method described, expressed as recovery factor from the standard oily matrix, ranged from 88 to 102%. Furthermore, to verify whether the extraction procedure may cause anion loss, a separate spike-and-recovery test on a virgin olive oil was performed, spiking the sample with appropriate amounts of F⫺, Cl⫺, Br⫺, NO2⫺, NO3⫺, PO43⫺, SO42⫺ and I⫺ sodium salts, as described earlier. The recovery test from the virgin olive oil shows that anion quantification remained unaffected by extraction treatment
321
CHAPTER | 35 Inorganic Anions in Olive Oils
30
Fluoride
A uS cm
25
Iodide
5
Sulfate
10
Phosphate Nitrate
Nitrate
15
Bromide
Chloride
20
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Min
Chloride
B uS cm 8 7 6 5 4
Sulfate
1
Phosphate
2
Fluoride
3
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Min
FIGURE 35.1 IEC chromatogram of (A) inorganic anions standard solution containing 10 mg L⫺1 of each essayed anion and of (B) virgin olive oil. Column: Metrosep Anion Dual 1 column (3.0 ⫻ 150 mm; 10.0 μm ); eluent: carbonate (3.12 mM)/hydrogen carbonate (3.25 mM)/2% acetone; flow rate 0.5 mL min⫺1.
followed by filtration on a carbon column: mean obtained recoveries spanned from 88% to 97.6% (Table 35.4). The reproducibility of the proposed method, expressed as long-term stability of the method, was evaluated by extracting and analyzing a standard olive oil containing 1.0 mg kg⫺1 of each assayed anion over five consecutive days. Reproducibility, which is very important when establishing routine methods to be used over extended periods, obtained for F⫺, Cl⫺, Br⫺, NO2⫺, NO3⫺, PO43⫺, SO42⫺ and I⫺ IEC analysis was 94%, 99%, 94%, 91%, 99%, 95%, 95%, and 91% respectively. This method couples rapidity with high performances, and therefore is suitable for routine analysis.
35.5 INORGANIC ANION LEVELS IN OLIVE OILS Olive oil is the most important fat of the Mediterranean diet and in recent years its production and consumption have
increased worldwide. Till now very few data about the presence of inorganic anions in olive oils were found in literature. Buldini et al. (1997a) reported chloride levels of 35.8 ⫾ 0.09 μg kg⫺1, whereas sulfate and phosphate were lower than the LOD. In this study the mean anion concentrations found in olive oil samples produced from the whole fruits (traditional method) were: 1.41 ⫾ 0.45 mg kg⫺1 fluorides, 18.2 ⫾ 5.5 mg kg⫺1 chlorides, 2.51 ⫾ 0.66 mg kg⫺1 phosphates and 3.13 ⫾ 0.75 mg kg⫺1 sulfates; the concentrations of bromide, nitrite, nitrate and iodide were lower than the LOD (Figure 35.2, lot C). Olive oil production technology has been modified in recent decades in order to improve olive oil quality. Some researches about the influence of de-stoning the olives on oil composition and quality were carried out (Angerosa et al., 1999; Saitta et al., 2003). To verify whether the concentration of inorganic anions in virgin olive oils was influenced by the olive fruit stone, oils from whole olives and from pulp-only tissue, respectively,
322
SECTION | I
Metals, Electrolytes and Other Components
TABLE 35.3 Calibration curves (y ⫽ Kx, r ⱖ 0.999), detection limits (LOD)and quantification limits (LOQ) for inorganic ion analysis in vegetable oils determined by ion exchange chromatography. This table reports all the parameters regarding the validation of the chromatographic method described. Method validation is the process of proving that an analytical method is acceptable for its intended purpose. Method validation must include studies on specificity, linearity, accuracy, precision, range, detection limit and quantitation limit. F⫺
Cl⫺
NO2⫺
Br⫺
NO3⫺
PO43⫺
SO42⫺
I⫺
Retention times (minutes)
4.12
6.77
7.36
8.70
10.47
11.67
14.75
26.24
y⫽
0.720x
1.015x
1.656x
3.190x
1.953x
3.130x
1.464x
3.342x
RSD%
0.852
1.171
9.170
10.84
4.110
1.927
4.040
0.456
Linearity range (mg kg⫺1)
0–100
0–100
0–100
0–100
0–100
0–100
0–100
0–100
Noise (nS cm⫺1)
7.0
6.0
2.0
1.0
2.5
2.5
3.0
6.0
LODa(Mg kg⫺1)
9.3
10
8.5
8.4
20.2
29
25
31
QOLb(Mg kg⫺1)
31.1
33.3
28.3
28.1
67.4
96.7
83.3
103.3
Precision (rsd%)c
3.7–5.3
3.9–4.2
4.5–6.5
4.3–6.5
4.1–5.0
4.2–6.2
4.1–5.8
4.5–6.5
RF(%)d
88–95
96–103
92–94
88–94
92–102
96–102
96–99
88–95
a
Defined as the signal height at a signal/noise ratio S/N ⫽ 3. Defined as the signal height at a signal/noise ratio S/N ⫽ 10. Precision was expressed as the rsd% range of nine runs performed at different concentration levels (0.25, 0.5, 1.0, 10, mg kg⫺1). d Recovery Factor ranges obtained at different concentration levels (0.25, 0.5, 1.0, 10, mg kg⫺1). b
c
TABLE 35.4 Spike and recovery text from a virgin olive oil. This table reports data about the accuracy of the analytical method described. The accuracy of a method is the closeness of the measured value to the true value for the sample. Usually, the accuracy of an analytical method is valued by performing measures on standard reference matrices; National Institute of Standards and Technology (NIST) reference standards are often used. If such a well characterized sample is not available, one of the most common approaches to assess the accuracy is to spike the sample matrix with different amounts of the analyte of interest and then to perform the extraction and the analysis according to the method developed. Recovery, which represents the accuracy, is calculated according to the following equation: C found ⭈100 Recovery% ⫽ Cexpected Concentration mg kg⫺1
F⫺
1.4
Cl⫺
17
Br⫺
⬍
Added mg kg⫺1 0 2 5 0 10 20 0 0.5 2
Expected mg kg⫺1 1.4 3.4 6.4 17 27 37 0 0.5 2
Founda mg kg⫺1 1.25 ⫾ 0.06 3.00 ⫾ 0.13 6.00 ⫾ 0.25 16.10 ⫾ 0.7 25.90 ⫾ 1.0 36.20 ⫾ 1.5 0 0.43 ⫾ 0.03 1.75 ⫾ 0.11
Recovery % 89.3 91.2 93.0 94.7 95.9 97.8 – 86.0 87.5
323
CHAPTER | 35 Inorganic Anions in Olive Oils
NO2⫺
⬍
0 0.5 2
0 0.5 2
0 0.44 ⫾ 0.03 1.81 ⫾ 0.12
– 88.0 90.5
NO3⫺
⬍
0 0.5 2
0 0.5 2
0.46 ⫾ 0.04 1.90 ⫾ 0.14
– 92 95
PO43⫺
2.5
0 5 10
2.5 7.5 12.5
2.30 ⫾ 0.14 7.11 ⫾ 0.46 12.20 ⫾ 0.73
92.0 94.7 97.6
SO42⫺
3.1
0 5 10
3.1 8.1 13.1
2.97 ⫾ 0.15 7.85 ⫾ 0.39 12.68 ⫾ 0.64
95.8 96.9 96.8
I⫺
⬍
0 0.5 2
0 0.44 ⫾ 0.03 1.87 ⫾ 0.15
– 88.0 93.5
0 0.5 2
a
Mean of three analyses.
35.5.1 Comparison with Commercial Seed Oils Seed oils are widely used in cooking and alimentary, cosmetic, pharmaceutical and chemical industries. Seed oils are characterized by a high content of polyunsatured fatty acids (PUFAs), linoleic and linolenic acids, and α tocopherol, whereas olive oil is high particularly in the mono-unsatured oleic acid as well as in α tocopherol. In recent years their consumption has been increasing because of their cholesterol-reducing effect, protecting against cardiovascular pathologies. Therefore it seems to be of great concern to
Chloride
Phosphates
Sulfate 18.2+5.5
Fluoride
20 18 16 14
3.13+0.75
2.51+0.66
1.41+0.45
0.83+0.05
2
0.22+0.08
4
0.31+0.11
6
1.93+0.98
8
0.8+0.1
4.55+1.03
10
7.6+1.0
12
0.41+0.10
Anions (mg kg−1)
were compared. Furthermore, the influence of the volume of processing water used in the production of olive oil from de-stoned fruit (lot A and B) on anion levels was assessed. Figure 35.2 gives evidence that the mean levels of anions in samples from lots A and B, respectively, were: 0.41 ⫾ 0.10 and 1.93 ⫾ 0.98 mg kg⫺1 fluorides, 4.55 ⫾ 1.03 and 7.6 ⫾ 1.0 mg kg⫺1 chlorides, 0.22 ⫾ 0.08 and 0.31 ⫾ 0.11 mg kg⫺1 phosphates, 0.8 ⫾ 0.1 and 0.23 ⫾ 0.05 mg kg⫺1 sulfates. The comparison of the obtained results showed that the content of chloride, phosphates and sulfates was significantly higher in oils obtained from the whole fruits (lot C), indicating that the stone contributes to the presence of inorganic anions in the oil more than the pulp tissue. Among the oils obtained from de-stoned fruits, those extracted using the highest amount of processing water (lot B) had a higher content of fluoride, chloride, phosphates and sulfates, giving evidence that water volume added to the paste affects inorganic anion levels in olive oils.
0 A
B
C
FIGURE 35.2 Mean levels of inorganic anions in olive oil samples from Nocellara del Belice variety produced from de-stoned olives (A and B) and from whole fruits (C).
compare the concentration of inorganic anions found in olive oil obtained from the whole fruit (traditional method), with those found in commercial seed oils (Figure 35.3): maize, sunflower, soy and peanuts. Fluoride levels in seed ranged from 0.92 mg kg⫺1 (maize oils) to 1.4 mg kg⫺1 (soy oils), being slightly lower than those found in olive. Chlorides, phosphate and sulfate concentrations were significantly lower than those found in olive oil. In particular, chlorides ranged from 1.04 mg kg⫺1 (peanut oils) to 6.31 mg kg⫺1 (maize oils),
324
SECTION | I
Chloride
Phosphates
●
Sulfate
6.31±2.51
Fluoride
7
●
5
0.22±0.1 0.29±0.15
1.02±0.55 1.04±0.31
0.52±0.12 0.47±0.2
1.4±0.63 1.87±0.9
1.75±0.81
1
0.83±0.22 0.49±0.15
2
0.87±0.27 0.75±0.29
3
1.21±0.75
4 0.92±0.35
Anions (mg kg−1)
6
0 Maize
Sunflower
Soy
Peanut
FIGURE 35.3 Mean levels of inorganic anions in commercial seed oils.
phosphates from 0.22 mg kg⫺1 (peanut oils) to 0.87 mg kg⫺1 (maize oils), and sulfates to 0.29 mg kg⫺1 (peanut oils) to 0.75 mg kg⫺1 (maize oils). The obtained results gave evidence that seed oils had lower amounts of all the studied inorganic anions with respect to olive oils from the whole fruit; this might be explained from the difference between the production processes of olive oil, which are mechanical, and seed oils, mainly based on solvent extraction.
SUMMARY POINTS ●
●
This chapter describes a study regarding the determination of several inorganic ions in olive oils from whole and de-stoned olives by ion chromatography with conductivity detection and chemical suppression. Sample preparation is a critical step in the whole analytical procedure; this study describes the ultrasoundassisted extraction of inorganic anions from the oil anions in a moderately alkaline solution (pH 8).
Metals, Electrolytes and Other Components
The analytical method was validated in terms of linearity, sensitivity, accuracy and precision. The obtained results have shown that the content of nitrite, nitrate, iodide and bromide was lower than the detection limits in all the samples. Chloride, phosphate and sulfate levels were significantly higher in oils obtained from the whole fruits, indicating that the stone contributes to the presence of inorganic anions in the oil more than the pulp tissue.
REFERENCES Angerosa, F., Basti, C., Vito, R., Lanza, B., 1999. Effect of stone removal on the production of virgin olive oil volatile compounds. Food Chem. 67, 295–299. Buldini, P.L., Cavalli, S., Trifirò, A., 1997a. State of the art ion chromatographic determination of inorganic ions in food. J. Chromatogr. A 789, 529–548. Buldini, P.L., Ferri, D., Lal Sharma, J., 1997b. Determination of some inorganic species in edible vegetable oils and fats by ion chromatography. J. Chromatogr. A 789, 549–555. Dugo, G.mo, Pellicanò, T.M., La Pera, L., Lo Turco, V., Tamborrino, A., Clodoveo, M.L., 2007. Determination of inorganic anions in commercial seed oils and in virgin olive oils produced from de-stoned olives and traditional extraction methods, using suppressed Ion Exchange Chromatography (IEC). Food Chem. 102, 599–605. La Pera, L., Dugo, G.mo, 2005. Metalli pesanti e selenio in oli vegetali In: Determinazione di metalli pesanti negli alimenti e nell’ambiente, Chiriotti ed. (Turin, Italy), chapt. 4, pp. 47–95. Lopez-Ruiz, B., 2000. Advances in the determination of inorganic anions by ion chromatography. J. Chromatogr. A 881, 607–627. Pennington, J.A.T., 1998. Dietary exposure for nitrate and nitrite. Food Control 9, 385–395. Saitta, M., Lo Turco, V., Pollicino, D., Dugo, G.mo, 2003. Oli d’oliva da pasta denocciolata ottenuta da cv. Coratina e Paranzana. Riv. Ital. delle Sost. Grasse 80, 27–34. USDA/Agricultural Research Service. (2004, November 22). Tracking Fluoride In The National Food Supply. ScienceDaily. Retrieved January 29, 2008, from http://www.sciencedaily.com/releases/2004/1 1/041122093050.htm
Chapter 36
Purification and Characterization of Olive (Olea europaea L.) Peroxidases Jorge A. Saraiva, C.S. Cláudia, S. Nunes and Manuel A. Coimbra Departamento de Química, Universidade de Aveiro, Campus Universitário de Santiago, Aveiro, Portugal
36.1 INTRODUCTION Peroxidases (donor: H2O2 oxidoreductase, EC 1.11.1.7; POD) are glycoproteins with ubiquitous distribution in the plant kingdom, showing generally several isoenzyme forms, high thermo-resistance and activity regeneration after heat inactivation (Vámos-Vigýazó, 1981). Physiological functions of POD include lignification, suberization and wound healing, general stress response, and protection against pathogen attack (Veitch, 2004). In raw and processed foods POD activity has been associated with adverse changes of flavor, color, texture, and nutritional value (Fils et al., 1985). The multiple isoperoxidase forms found within the same plant source can differ with respect to molecular mass, isoelectric point, pH and temperature optima, substrate specificity, amino acid and sugar compositions, and heat stability. Due to its high thermal stability and easiness of activity quantification, peroxidase residual activity is frequently used to determine whether the heat processing of vegetables has been adequate. Peroxidases are also becoming increasingly attractive catalysts to generate free radicals to synthesize a variety of polymers, to promote stereospecific biotransformation of organic molecules, and for bioremediation (Klibanov et al., 1983). Olive fruit is used for oil extraction and also for consumption as table olives after processing. Due to the moderate degree of saturation of olive oil fatty acids, table olives and olive oil are considered products of high biological and nutritive values (Fernández et al., 1997). Also, these products have been claimed to contain significant amounts of simple and complex phenols that are bioactive antioxidants, and their health-promoting benefits support the recommendation to include them in the human diet. Texture and consistency are important quality characteristics of table olives that can be influenced by the catalytic action of peroxidase, since it has been associated with cell wall stiffening Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
(Shah et al., 2004). The Portuguese Douro cultivar is one of the few cultivars that produce fruits that can be harvested at the black ripening stage, to be processed for consumption as black oxidized table olives, resulting in a final product with good texture properties. This feature was attributed to an increase of both skin strength and stiffness caused by ripening (Georget et al., 2001) that prevents detrimental effects on texture caused by processing. POD has also been linked to fatty acid oxidation (Adams et al., 2003) and it was reported that POD present in palm fruit can initiate the oxidation of palm oil (Deepa and Arumughan, 2002), with consequent quality deterioration. Texture, consistency, color, and fatty acid oxidation are important quality characteristics of table olives that can also be influenced by the catalytic action of peroxidase. Olive POD from naturally black olives of the Portuguese Douro cultivar was recently purified and biochemically characterized (Saraiva et al., 2007), as an initial step to study the possible contribution of this enzyme to the peculiar texture characteristics of Douro cultivar black ripened olive.
36.2 OLIVE POD PURIFICATION The crude enzyme extract was obtained by extraction with 0.05 M sodium phosphate buffer, pH 7.0 (Ingham et al., 1998), from the acetone powder prepared based on the method described by Civello et al. (1995). Four different cationic fractions with POD activity were found using cationic exchange CM-Sepharose chromatography (Figure 36.1A), named PODc1-4, with the (c) standing for cationic and the number indicating the order of elution. The unbound fraction was further purified by anionic exchange DEAE-Sephacel chromatography resulting in four anionic fractions showing POD activity (Figure 36.1B), named PODa1-4, with the (a) standing for anionic
325
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
326
SECTION | I
A 3.5
1.0
1.5 PODc1
1.0
PODc2 PODc3
PODc4
0.5
20.0 [NaCl] (M)
Excluded fraction
2.0
15.0 10.0
Activity (nkat)
25.0
2.5 Abs 280 nm
β-mercaptoethanol, indicating that all fractions were homogeneous and composed of a single polypeptide chain.
30.0 1.5
3.0
5.0
0.5 0.0
0.0
0.0 0
10
20
30
40
50
60
Fraction nº
B 3.5
12.0 1.5
2.0
8.0
PODa3
PODa1
1.0
1.5 PODa2
0.5
1.0 0.5
0.0
0.0 10
20
30
4.0 2.0
PODa4
0
6.0
40
50
Activity (nkat x 10−2)
10.0
2.5 [NaCl] (M)
Abs 280 nm
3.0
0.0 60
Fraction number
FIGURE 36.1 Ion-exchange chromatography of olive peroxidase. (A) CM-Sepharose chromatography; (B) DEAE-Sephacel chromatography of excluded fraction from (A). Peroxidase activity (䊐), absorbance at 280 nm (䊉) and, NaCl gradient (full straightline). These figures present the results obtained for purification of olive peroxidase using cation- (A) and anion-exchange (B) chromatographies. CM: carboxymethyl; DEAE: diethylaminoethyl; PODa1-4 and PODc1-4: POD stands for peroxidase, c for cationic, a for anionic and the number indicates the order of elution. Figure 36.1B is reprinted from Saraiva J. A. et al. Food Chem 2007; 101: 1571–1579, with permission.
and the number indicating the order of elution. Peroxidase activity (nkat) was measured spectrophotometrically (Worthington, 1993), with phenol as substrate (Eq. 36.1), using the extinction coefficient (6.58 M⫺1 cm⫺1), , at 510 nm for quinol red (antipyrilquinoneimine), the colored reaction product, and revealed that the four anionic fractions were responsible for 92% of the recovered activity, with PODa4 accounting for about 70% of the recovered activity. O
OH
CH3
NH2
CH3
N
Peroxidase +
2 H2O2 +
H3C
Phenol
+ 4 H 2O
N
N N
O
4-Aminoantipyrine
H3C
N
O
Antipyrilquinoneimine
Metals, Electrolytes and Other Components
(36.1)
SDS-PAGE of each of the eight POD fractions revealed a single protein band either in the presence or absence of
36.3 OLIVE POD CHARACTERIZATION 36.3.1 Molecular Weight and Isoelectric Point The molecular mass of the purified fractions, determined by SDS-PAGE, under reducing and non-reducing conditions, yielded an average of 66.4, 68.6, 64.3, and 60.2 kDa, for the cationic fractions (PODc1-4), and 18.4, 20.3, 20.1, and 20.5 kDa for the anionic fractions (PODa1-4, Figure 36.2A). The molecular masses of peroxidases from several fruits and vegetables usually range from 30 to 54 kDa (VámosVigýazó, 1981), but higher values have been reported by Padiglia et al. (1995) for opuntia (58 kDa) and by Civello et al. (1995) for strawberry (58 and 66 kDa) peroxidases. Although the low molecular mass for olive PODa1-4 anionic fractions found in this work is rather unusual, similar values were also reported by Khan and Robinson (1993) for two mango isoperoxidases (22 and 27 kDa) and more recently, a 6 kDa peptide from Raphanus sativus was reported as showing peroxidase activity (Omumi et al., 2001). Ebrahimzadeh et al. (2003) reported that POD activity in olive increased gradually during fruit development and concomitantly the same authors verified the appearance and increment of peroxidase protein bands at 18–22 kDa, thus pointing to the occurrence of POD isoenzymes with molecular masses in this range in olive fruit. The four olive POD anionic fractions showed isoelectric points of 6.9 for PODa1 and PODa2 and 4.6 and 4.4 for, respectively, PODa3 and PODa4 (Figure 36.2B).
36.3.2 Optimum pH and Km values for Phenol and H2O2 PODa4 that comprised about 70% of the total recovered activity showed a pH optimum of 7.0, in 0.1 M sodium phosphate buffer (Figure 36.3). Horseradish peroxidase, the most studied peroxidase, which main isoenzyme was already cloned, shows also an optimum pH of 7.0 using phenol as substrate (Worthington, 1993). PODa4 showed a Michaelis-Menten kinetics with both phenol and H2O2, up to a concentration of, respectively, 150 mM and 4.0 mM (Figure 36.4). Apparent Km values estimated using least-squares non-linear regression were 41.0 (r2 ⫽ 0.99) and 0.53 mM (r2 ⫽ 0.99) for phenol and H2O2, respectively. The value obtained for Km for H2O2 was similar to that obtained by Dong (2002) for horseradish peroxidase (0.43 mM), using phenol as hydrogen donor, and comparable to those reported for anionic peroxidases from melon (Rodríguez-López et al., 2000) and from turnip
327
CHAPTER | 36 Purification and Characterization of Olive (Olea europaea L.) Peroxidases
A
M PODa1 PODa2 PODa3 PODa4 M
PODa4*
Mr(kDa)
200.0 45.0 31.0 21.5
Relative activity (%)
100 80 60 40 20 0 3.0
14.4 6.5
4.0
5.0
6.0 pH
7.0
8.0
9.0
FIGURE 36.3 Effect of pH on enzymatic activity of olive peroxidase anionic fraction PODa4, using phenol as substrate. This figure shows the effect of pH on the activity of olive peroxidase PODa4 (results are shown relative to the activity at pH 7.0). Vertical bars indicate ⫾ standard deviation (3 determinations). Reprinted from Saraiva J. A. et al. Food Chem 2007; 101: 1571–1579, with permission.
B pH 9.6 8.0 7.8 7.5 7.0 6.8 6.5
M PODa1 PODa2 M PODa3 PODa4
6.0 5.1 4.2
FIGURE 36.2 (A) SDS-polyacrylamide gel electrophoresis under nonreducing conditions and (B) isoelectric focusing of olive POD purified anionic fractions (PODa1, PODa2, PODa3, and PODa4) (M stands for markers with known molecular mass or isoelectric point values, in (A) or (B), respectively). PODa4* refers to PODa4 fraction with one fourth of the protein content used in the gel, compared to lane PODa4. These figures present the results obtained for purification of olive peroxidase using cation- (A) and anion-exchange (B) chromatographies. SDS: sodium dodecyl sulfate; Mr: molecular mass; M: marker; for the meaning of PODa1-4 and PODc1-4 see the legend to Figure 36.1. Reprinted from Saraiva J. A. et al. Food Chem 2007; 101: 1571–1579, with permission.
roots (Duarte-Vázquez et al., 2000), with other hydrogen donors. For horseradish peroxidase, a lower Km value (13 mM) for phenol was reported (Liu et al., 2002).
36.3.3 Denaturation and Optimum Temperatures Denaturation (unfolding) temperature of olive PODa4 purified fraction was determined based on the procedure reported by Hei and Clarck (1993), from the experimentally
determined activity–temperature profile (Figure 36.5), in the range of 15–60°C, using Equations 36.2–36.4. vo (T ) ⫽ kcat (T ) N ⫽
kcat (T )Eo 1 ⫹ K RD (T )
(36.2)
N and D are the amounts of native and reversibly denatured enzyme, respectively, Eo is the total amount of enzyme (Eo ⫽ N ⫹ D), KRD ⫽ [D]/[N] is the equilibrium denaturation constant, vo(T), initial enzymatic activity, relates to the amount of active, native enzyme (N), by the catalytic rate constant of the enzyme, kcat (T). ⎛ E ⎞ kcat (T ) ⫽ Aexp ⎜⎜⫺ a ⎟⎟ ⎜⎝ RT ⎟⎠
(36.3)
A is the Arrhenius pre-exponential factor, Ea is the activation energy (kJ mol⫺1), R is the universal gas constant (8.314 ⫻ 10⫺3 mol kJ⫺1 K⫺1), and T is the absolute temperature (K). ⎪⎧⎪ ΔH o ⎪⎫⎪ ΔC op m ⎪⎪ ⎪⎪ ⫺ (1 ⫹ ln(Tm )) ⫹ ⎪⎪ RTm ⎪⎪ R K RD (T ) ⫽ exp ⎨⎛ ⎬ o o ⎪⎪⎜ ΔC pTm ⎪ ΔH mo ⎞⎟⎟ 1 ΔC p ⎪⎪⎜⎜ ln(T )⎪⎪⎪ ⫺ ⎟⎟ ⫹ R ⎠⎟ T ⎪⎪⎜⎜⎝ RT ⎪⎪ R ⎪⎩ ⎪⎭ (36.4) ΔH mo , ΔC op , and Tm are, respectively, the change in enthalpy for thermal unfolding at Tm and 1 atm (Hreversible denaturated ⫺ Hnative), the change in heat capacity for thermal unfolding at 1 atm (Cp, reversible denaturated ⫺ Cp, native), and the unfolding temperature (the temperature at which KRD ⫽ 1 and [D] ⫽ [N]).
328
SECTION | I
A 14.0
10.0 1/Activity (nkat−1)
Activity (nkat)
12.0
8.0 6.0 4.0 2.0
0.8 0.6 0.4 0.2 0.0 0.00 0.04 0.08 0.12 0.16 0.20 −1 1/[Phenol] (mM )
0.0 0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0 160.0
[Phenol] (mM) B 12.0
1/Activity (nkat−1)
Activity (nkat)
10.0 8.0 6.0 4.0
0.8 0.6 0.4 0.2 0.0 0.0
2.0
5.0
10.0
15.0
1/[H2O2] (mM−1)
0.0 0.0
1.0
2.0
3.0
4.0
5.0
[H2O2] (mM)
FIGURE 36.4 Effect of (A) phenol and (B) hydrogen peroxide concentration on activity of olive peroxidase anionic fraction PODa4 (insets: Lineweaver-Burk plots). These figures present the effect of phenol (A) and hydrogen peroxide (B) concentration on activity of olive peroxidase anionic fraction PODa4. Vertical bars represent ⫾ standard deviation (3 determinations). Reprinted from Saraiva J. A. et al. Food Chem 2007; 101: 1571–1579, with permission.
Relative activity (%)
100
Metals, Electrolytes and Other Components
MathWorks, Inc., MA, USA) (for details see also Saraiva et al., 2007). Using the Arrhenius equation, a value of 99.1 kJ mol⫺1 (Table 36.1) was obtained for the activation energy (Ea) for phenol oxidation. A value of 4 was found for the temperature coefficient (Q10 – the quotient between the activity at a temperature (T ⫹ 10) K and the activity at T K), meaning that PODa4 activity doubles for each 5°C increase. For most enzymes in homeothermic species (e.g. mammals) the value of Q10 is approximately 2 (Price and Stevens, 1996). However, enzymes of poikilothermic organisms, as for example, species that have to adapt to cold conditions, usually have Q10 values lower than 2. This permits these organisms to avoid an excessive slow down of their essential metabolic reactions. Obviously, temperature increase does also cause smaller effects on metabolic reactions in these organisms (Price and Stevens, 1996). Organisms with enzymes with Q10 values higher than 2 might experience considerable increase/decrease in metabolic reactions, when temperature increases/decreases. This will only happen if the Q10 determined in vitro reflects the actual in vivo Q10 of an enzyme and does not take into account other possible mechanisms of enzyme activity control in vivo. The results obtained for the thermodynamic parameters for the reversible thermal unfolding of PODa4 (Table o 36.1), revealed a value of 411.2 kJ mol⫺1 for ΔH m , which is within the range obtained for many other proteins, and a o negative value of ⫺13.6 kJ mol⫺1 K⫺1 for ΔC p , (several o proteins were reported to show negative values for ΔC p for thermal unfolding) (Pfeil, 1998). The value obtained for the unfolding temperature, Tm, was 36.5°C and the temperature for maximum activity, Topt, was 34.7°C (Figure 36.5 shows the experimental points and the model fitted curve, in the whole activity–temperature range).
80
36.3.4 Thermal Stability
60 40 20 0 280
290
300
310
320
330
340
Temperature (K)
FIGURE 36.5 Activity–temperature profile obtained for the oxidation of phenol by PODa4. Effect of temperature on activity of olive peroxidase anionic fraction PODa4. The model curve was obtained by fitting the experimental data to the equation obtained by combining Eqs 36.2–36.4. Vertical bars represent ⫾ standard deviation (at least 2 determinations). Reprinted from Saraiva J. A. et al. Food Chem 2007; 101: 1571–1579, with permission.
Curve fitting was also carried out based on the procedure described by Hei and Clarck (1993), by leastsquare non-linear regression (MATLAB software, The
Thermal stability of olive POD was undertaken using the partially purified enzymatic extract obtained after ammonium sulfate fractionation and desalting, which is representative of all eight POD fractions found in olive fruit. Heating at 40°C for 5 and 10 min caused 60% and 85% reduction of activity, respectively, and no measurable activity could be detected upon heating at 50°C and 60°C for 5 min, indicating a moderate thermal stability of olive POD. The enzyme was not able to regain activity upon inactivation and storage at 4°C for 24 h, which indicates that the inactivation observed was irreversible.
36.3.5 Carbohydrate Content and Glycosidic Linkage Analysis Neutral and amine sugars resulting from hydrolysis of the carbohydrate moiety of PODa4 were analyzed as their
329
CHAPTER | 36 Purification and Characterization of Olive (Olea europaea L.) Peroxidases
TABLE 36.1 Parameters obtained from the Arrhenius fit (Ea and A), for the fit of the thermal denaturation (Eq. 36.4), o o ΔHm , ΔC p , and Tm), respectively, for the low- and high-temperature ranges of the activity-temperature profile of phenol oxidation by PODa4, and for the estimation of Topt (errors indicated are ⫾ 95% confidence interval). Ea (kJ mol⫺1)
A
o ΔHm (kJ mol⫺1)
ΔC po
99.1 ⫾ 16.4
1.54 ⫻ 1019
411.2 ⫾ 62.2
⫺13.6 ⫾ 4.5
(r2 ⫽ 0.97)
(kJ mol⫺1 K⫺1)
Tm , K (°C)
Topt , K (°C)
309.7 ⫾ 0.5
307.9 ⫾ ⬍ 0.1
(36.5)
(34.7)
This table shows the results obtained for the activation energy for oxidation of phenol and for the thermodynamic parameters characterizing the thermal o o denaturation of olive PODa4. Ea: the activation energy (kJ mol⫺1); A: the Arrhenius pre-exponential factor; ΔHm , ΔC p , Tm, and Topt are, respectively, the change in enthalpy for thermal unfolding at Tm and 1 atm (Hreversible denaturated ⫺ Hnative), the change in heat capacity for thermal unfolding at 1 atm (Cp, reversible denaturated ⫺ Cp, native), the unfolding temperature, and the temperature for maximum activity. Reprinted from Saraiva, J. A. et al. 2007. Food Chem. 101, 1571–1579, with permission.
alditol acetates by GC (Harris et al., 1988). Uronic acids were quantified by a colorimetric method as described by Coimbra et al. (1996). Glycosidic linkage analysis was carried out by methylation studies as described by Coimbra et al. (1996) and the partially methylated alditol acetates were analyzed and characterized by GC-MS. The glycosidic fraction of PODa4 was constituted mainly by arabinose (39 mol%) and uronic acids (38%). Glucose (10%), rhamnose (6%), galactose (4%), fucose (1%), xylose (1%) and mannose (1%) were also detected. The presence of galactose, fucose, xylose, and mannose, and the relative amounts of the latter three sugars, points to PODa4 having a glycan moiety characteristic of the ‘complex type’ of glycoproteins (Dwek, 1996), as was reported for petunia (Hendriks and van Loon, 1990), and peanut (Sun et al., 1997) peroxidases. Arabinose and glucose were also reported to be present in petunia (Hendriks and van Loon, 1990) and horseradish (Clark and Shannon, 1976) peroxidases. The origin of these sugars (and of rhamnose and uronic acids) may be related to the presence of cell wall pectic polysaccharides, as was proposed for petunia peroxidase (Hendriks and van Loon, 1990). Such a possibility is supported by FTIR analysis of the purified PODa4 that revealed pectic polysaccharides characteristic bands at 1150, 1100 and 1020 cm⫺1 (Coimbra et al., 1998), indicating the presence of pectic polysaccharides in the purified PODa4 fraction. The presence of pectic polysaccharides was also confirmed by methylation analysis of the arabinan moiety, which allows determination of the type of glycosidic linkages present. The results revealed an arabinosyl composition that, although more branched, is characteristic of the arabinan that occurs as side chains of olive pulp pectic polysaccharides (Cardoso et al., 2002). To eliminate the possibility of the presence of these pectic polysaccharides as an artefact of the purification procedure, namely, to discard the possible presence of polysaccharides in the gels obtained by isoelectric focusing electrophoresis, the phenol-sulfuric acid method (Dubois et al., 1956) that allows the detection of sugars was applied.
The results indicated the presence of polysaccharides only at the positions coincident with the four olive POD anionic protein bands, as seen by the appearance of a brown color. These results point to the presence of pectic polysaccharides, in olive PODa1-4 purified fractions, possibly linked to the enzyme, since it is not expected that the pectic polysaccharides, due to their heterogeneity, would behave the same way as the enzyme during ammonium sulfate fractionation, ionic chromatographies, and isoelectric focusing electrophoresis.
36.3.6 Peroxidases Binding to Pectins Some peroxidase isoenzymes from lupin (Ros Barceló et al., 1988), zucchini (Penel and Greppin, 1994), and Arabidopsis leaves (Shah et al., 2004), have been reported to show a specific binding affinity for low methylesterified pectins in the presence of calcium. These peroxidases bind specifically to the homogalacturonan domains of pectin chains, in their Ca2⫹-pectate conformation, a structure called ‘egg-box’ or junction zone. The Ca2⫹-pectate affinity exhibited by these peroxidases was proposed to possibly function as a spatial distribution control mechanism within the cell wall matrix (Carpin et al., 2001), since the Ca2⫹pectate conformation of pectins occurs mainly in middle lamella and cell corners. The interaction between zucchini peroxidase and pectin was found to be of electrostatic nature, although resisting relatively high concentrations of NaCl (Carpin et al., 2001). The results obtained for fraction PODa4 support the hypothesis that olive peroxidase fraction is a peroxidase that binds specifically to pectic polysaccharides, with a very low molecular mass and very low degree of methyl esterification. Ripening of olive fruit was found to cause significant depolymerization of olive fruit cell wall pectic polysaccharides (Mafra et al., 2001), with the consequent formation of a relatively higher proportion of lower molecular mass pectic material. In addition, the degree of methyl
330
esterification of olive pectic polysaccharides decreases from 71% to 35%, from green to black olives (Mafra et al., 2001). In this way, low methylesterified and low molecular mass pectic material is present in black ripened olives that can bind to olive POD.
36.4 ACTIVITY OF PEROXIDASE IN TABLE OLIVES
POD activity (ΔAbs510 nm/min/g olive pulp)
SECTION | I
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 C-u
36.4.1 Effect of Ripening Peroxidase activity was detected in table olives of the Portuguese Douro variety already at the green color stage (Table 36.2). At the turning-color stage the activity was only about 32% that at the green color stage. At the black stage the activity increased to reach a value of around 67% that of the green stage. At the three color stages, the activity of peroxidase was found to be mainly in the soluble and ionically bound to the cell wall forms (Table 36.2).
36.4.2 Effect of Processing Table olives are processed by different methods, based mostly on traditional and empirical procedures, in European countries (Table 36.3). Peroxidase activity was present in unprocessed and processed table olives of four TABLE 36.2 Peroxidase activity in the pulp of the Portuguese olive Douro variety (⫾ standard deviation for two determinations) at three different color stages. ⫺1
⫺1
Peroxidase activity (ΔAbs 510 nm min pulp)
g
of fresh olive
Green
Turning color
Black
0.79 ⫾ 0.19
0.26 ⫾ 0.06
0.53 ⫾ 0.18
(56/40/4)
(30/52/18)
(81/15/4)
This table presents the results for the quantification of the total peroxidase activity present in the pulp of the Portuguese olive Douro variety, at the green, turning, and black color stages. In parenthesis is shown the relative distribution (%) of the quantified peroxidase activity for the soluble, ionically-, and covalently-bound cell wall forms. ΔAbs: change in absorbance.
Metals, Electrolytes and Other Components
C-p
Ca-u
Ca-p
Th-u
Th-p
T-u
different varieties, processed by different methods (Figure 36.6; FAIR Project CT97 3053, 2000). Because peroxidase activity remains in processed table olives, its action may influence the texture quality of the olives, since POD is considered to be involved in the formation of cross-links among cell wall polymers, participating in this way in the construction of cell walls and in the control of cell wall plasticity (Shah et al., 2004).
ACKNOWLEDGMENTS The authors acknowledge financial support from Universidade de Aveiro (UA), Fundação para a Ciência e Tecnologia (Grant SFRH/BM/2399/ 2000), and European Union (OLITEXT project, FAIR CT97-3053).
SUMMARY POINTS ●
Peroxidase activity is present in olives of the Portuguese Douro variety in the green, turning, and black color stages, mainly soluble in the cell and ionically linked to the cell wall.
TABLE 36.3 Processing applied to Conservolia, Taggiasca, Cassanese, and Thasos olive varieties for the production of table olives.
Processing
T-p
FIGURE 36.6 Effect of different types of processing on activity of olive peroxidase in different varieties of table olives. Effect of different types of processing on activity of olive peroxidase of olives of the varieties Conservolia, Cassanese, Thasos, and Taggiasca. Legend: C-u and C-p, Naturally black fermented Conservolia olives; Ca-u and Ca-p, Boiled and salted Cassanese olives; Th-u and Th-p, Dry-salted Thasos olives; T-u and T-p, Naturally black fermented Taggiasca olives (letters u and p indicate, respectively, unprocessed and processed olives). Vertical bars represent ⫾ standard deviation (at least 2 determinations).
Conservolia
Taggiasca
Cassanese
Thasos
Fermented in 6% NaCl
Fermented in 12% NaCl
Heated at 100°C for 5–10 min for de-bittering, followed by oven drying
Dried with solid NaCl
CHAPTER | 36 Purification and Characterization of Olive (Olea europaea L.) Peroxidases
●
●
●
●
●
●
●
●
●
●
Olives at the black color stage show eight peroxidase isoenzymes, being four cationic and four anionic. A single anionic fraction, PODa4, accounts for 70% of total recovered activity. PODa4 has a molecular weight of 20.5 kDa and an isoelectric point of 4.4. PODa4 has an optimum pH of 7.0 and Km values of 41.0 and 0.53 mM, for phenol and H2O2, respectively. The activation energy for oxidation of phenol by PODa4 is 99.1 kJ mol⫺1 (corresponding to a Q10 value of 4, i.e., the activity changes four-fold when the temperature changes 10°C). PODa4 thermal unfolding temperature (Tm), and optimum temperature (Topt) for activity are, respectively, 36.5°C and 34.7°C. PODa4 is a glycoprotein, showing a glycosidic fraction constituted mainly by arabinose and uronic acids. Several experimental evidences indicate PODa4 as being a possible pectin binding peroxidase. Peroxidase activity is also found in four other olive varieties (Conservolia, Taggiasca, Cassanese, and Thasos) at the harvesting stage and resists several types of processing (fermentation in 6% and 12% NaCl, heating at 100°C for 5–10 min, followed by oven drying, and drying with solid NaCl). Activity of peroxidase during processing and storage of processed table olives may influence the texture quality of the olives, since peroxidase activity can contribute to cell wall plasticity.
REFERENCES Adams, J.B., Brown, H.M., Ledward, D.A., Turner, R., 2003. Heatinactivated peroxidases and the role of calcium abstraction as a cause of their enhanced lipid oxidation activity: potential effects on the flavour quality of heat-processed vegetables. Food Chem. 80, 499–510. Cardoso, S.M., Silva, A.M.S., Coimbra, M.A., 2002. Structural characterisation of the olive pomace pectic polysaccharide arabinan side chains. Carbohydr. Res. 337, 917–924. Carpin, S., Crèvecoeur, M., de Meyer, M., Simon, P., Greppin, H., Penel, C., 2001. Identification of a Ca2 ⫹ -pectate binding site on an apoplastic peroxidase. Plant Cell 13, 511–520. Civello, P.M., Martínez, G.A., Chaves, A.R., Añón, M.C., 1995. Peroxidase from strawberry fruit (Fragaria ananassa Duch.): Partial purification and determination of some properties. J. Agric. Food Chem. 43, 2596–2601. Clark, J., Shannon, L.M., 1976. The isolation and characterisation of the glycopeptides from horseradish peroxidase isoenzyme C. Biochim. Biophys. Acta, 427, 428–442. Coimbra, M.A., Delgadillo, I., Waldron, K.W., Selvendran, R.R., 1996. Isolation and analysis of cell polymers from olive pulp. In: Linskens, H.F., Jackson, J.F. (eds) Modern Methods of Plant Analysis, Vol. 17. Springer-Verlag, Berlin, pp. 18–44. Coimbra, M.A., Barros, A., Barros, M., Rutledge, D.N., Delgadillo, I., 1998. Multivariate analysis of uronic acid and neutral sugars in
331
whole pectic samples by FT-IR spectroscopy. Carbohydr. Polym. 37, 241–248. Deepa, S.S., Arumughan, C., 2002. Purification and characterization of soluble peroxidase from oil palm (Elaeis guineensis Jacq.) leaf. Phytochemistry 61, 503–511. Dong, H., 2002. Organic-Inorganic Hybrid Mesoporous Materials and Their Application as Host Matrix for Protein Molecules. PhD. Thesis, pp. 136–137. Drexel University, Philadelphia. Duarte-Vázquez, M.A., García-Almendárez, B., Regalado, C., Whitaker, J.R., 2000. Purification and partial characterization of three turnip (Brassica napus L. Var. esculenta D.C.) peroxidases. J. Agric. Food Chem. 48, 1574–1579. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350–356. Dwek, R.A., 1996. Glycobiology: Toward understanding the function of sugars. Chem. Rev. 96, 683–720. Ebrahimzadeh, H., Motamed, N., Rastgar-Jazii, F., Montasser-Kouhsari, S., Shokraii, E.H., 2003. Oxidative enzyme activities and soluble protein content in leaves and fruits of olives during ripening. J. Food Biochem. 27, 181–196. FAIR Project CT97 3053. 2000. OLITEXT – Improvement of texture characteristics of some European olive-fruit varieties suitable for table-olive purposes. Fils, B., Sauvage, F.X., Nicolas, J., 1985. Tomato peroxidase purification and some properties. Sci. Aliment. 5, 217–232. Fernández, G.A., Fernández Díez, M.J., Adams, M.R., 1997. Physical and chemical characteristics of the olive fruit. In: Fernández, A.G., Díez, M.J.F., Adams, M.R. (eds) Table Olives–Production and Processing. Chapman and Hall, London, pp. 66–109. Georget, D.M.R., Smith, A.C., Waldron, K.W., 2001. Effect of ripening on the mechanical properties of Portuguese and Spanish varieties of olive (Olea europaea L.). J. Agric. Food Chem. 81, 448–454. Harris, P.J., Blakeney, A.B., Henry, R.J., Stone, B.A., 1988. Gas chromatographic determination of the monosaccharide composition of plant cell wall preparations. J. Assoc. Off. Anal. Chem. 71, 272–275. Hei, D.J., Clarck, D.S., 1993. Estimation of melting curves from enzymatic activity-temperature profiles. Biotechnol. Bioeng. 42, 1245–1251. Hendriks, T., van Loon, L.C., 1990. The nature of the heterogeneity of petunia peroxidase a. J. Plant Physiol. 136, 526–531. Ingham, L.M., Parker, M.L., Waldron, K.W., 1998. Peroxidase: Changes in soluble and bound forms during maturation and ripening of apples. Physiol. Plant. 102, 93–100. Khan, A.A., Robinson, D.S., 1993. Purification of an anionic peroxidase isoenzyme from mango (Mangifera indica L. var. Chaunsa). Food Chem. 46, 61–64. Klibanov, A.M., Tu, T.M., Scott, K.P., 1983. Peroxidase-catalysed removal of phenols from coal-conversion waste waters. Science 221, 259–261. Liu, J.Z., Song, H.Y., Weng, L.P., Ji, L.N., 2002. Increased thermostability and phenol removal efficiency by chemical modified horseradish peroxidase. J. Mol. Catal. B: Enzym. 18, 225–232. Mafra, I., Lanza, B., Reis, A., Marsilio, V., Campestre, C., De Angelis, M., Coimbra, M.A., 2001. Effect of ripening on texture, microstructure and cell wall polysaccharide composition of olive fruit (Olea europaea). Physiol. Plant. 111, 439–447. Omumi, A., Morshedi, D., Samadi, A., 2001. Isolation of a novel stable peptide from cultivated Raphanus sativus with peroxidase activity. World J. Microbiol. Biotechnol. 17, 827–828.
332
Padiglia, A., Cruciani, E., Pazzaglia, G., Medda, R., Floris, G., 1995. Purification and characterization of opuntia peroxidase. Phytochemistry 38, 295–297. Penel, C., Greppin, H., 1994. Binding of plant isoperoxidases to pectin in the presence of calcium. FEBS Lett. 343, 51–55. Pfeil, W., 1998. Protein stability and folding. In: A Collection of Thermodynamic Data. Springer-Verlag, Berlin. Price, C.N., Stevens, L., 1996. An introduction to enzyme kinetics. In: Fundamentals of Enzymology. Oxford University Press, Oxford, pp. 158–160. Rodríguez-López, J.N., Espín, J.C., Amor, F., Tudela, J., Martínez, V., Cerdá, A., García-Cánovas, F., 2000. Purification and kinetic measurement of an anionic peroxidase from melon (Cucumis melo L.) cultivated under different salinity conditions. J. Agric. Food Chem. 48, 1537–1541. Ros Barceló, A., Pedreño, M.A., Muñoz, R., Sabater, F., 1988. Lupin peroxidases. II. Binding of acidic isoperoxidases to cell walls. Physiol. Plant. 73, 238–244.
SECTION | I
Metals, Electrolytes and Other Components
Saraiva, J.A., Nunes, C.S., Coimbra, M.A., 2007. Purification and characterization of olive (Olea europaea L.) peroxidase – Evidence for the occurrence of a pectin binding peroxidise. Food Chem. 101, 1571–1579. Shah, K., Penel, C., Gagnon, J., Dunand, C., 2004. Purification and identification of a Ca2⫹-pectate binding peroxidase from Arabidopsis leaves. Phytochemistry 65, 307–312. Sun, Y., Lige, B., van Huystee, R.B., 1997. HPLC determination of the sugar compositions of the glycans on the cationic peanut peroxidase. J. Agric. Food Chem. 45, 4196–4200. Vámos-Vigýazó, L., 1981. Polyphenol oxidase and peroxidase in fruits and vegetables. Crit. Rev. Food Sci. Nut. 15, 49–127. Veitch, N.C., 2004. Horseradish peroxidase: a modern view of a classic enzyme. Phytochemistry 65, 249–259. Worthington Biochemical Corporation, 1993. Worthington Enzyme Manual. In: Enzymes and Related Biochemicals. Millipore Corporation, Bedford, pp. 293–294.
Chapter 37
Olive Biophenols and Conventional Biotechnology from Mediterranean Aliment Culture Ganapathy Sivakumar1 and Nicola A. Uccella2 1 2
Arkansas Biosciences Institute, Arkansas State University, Jonesboro, USA IRESMO Foundation Group, Chemistry Department, Calabria University, Rende (CS), Italy
37.1 INTRODUCTION The olive plant (Olea europea L.), an important agro-industrial crop in Mediterranean countries, produces extra virgin olive oil (evoo) and organic table olives (otos). The global success of these typical components in everyday meals, consumed in the Mediterranean diet (Trichopoulou et al., 2009), relies on the mythical food, selected as a goodness triad. Spermo, the wheat carbohydrates, Oinos, the grape phytomolecules, like quercetin and resveratrol, and Elais, olive lipids and BPs, were Zeus divinities on Olympus. The Mediterranean diet is a sort of therapeutic way to reduce developing cardioand cerebro-vascular pathologies and degenerative diseases, cancer and aging, but does not reflect the eating habits of the Mediterranean population. Their good health and well-being depend on seasonal fruit, vegetables, herbs, and their wise mixture and manipulators. The Portulaca oleracea used in salads, omelettes and potato dumplings, contains eicosapentaenoic acid, an ω-3 fatty acid, in extraordinary amounts relative to other plant sources and normally found in fish and algae (Simopouloas, 2004). The millenarian benefits, experimented by Mediterraneans, stands on the famous dish sapience as well as pasta, pizza and bread being seasoned by evoo and otos, which slows down fatty acid penetration into arterial walls and exerts antioxidant activity by phytomolecules, like tocopherols and BPs. More then a simple diet, it is a diaità or a way of life, and, adopting a coherent Latin synonym, not the Saxon one, becomes MAC, not just food, a nutriment to survive, but a well-living approach, on health and well-being (Saija and Uccella, 2001). Evoo and otos biomolecules are vital to MAC for nutritional benefit and excellent eating quality, within the hedonic-sensory descriptors of modern functional food for aroma, odor, taste, texture, and convenience (Uccella, 2003a). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
The evoo and otos phytomolecules are influenced by agronomics, genomics, proteomics, pre- and post-harvest techniques and processing, till to the table as MAC traditional recipes. Complex bio- and techno-molecules in olive functional products are significant in determining crù and cuvée characteristics of evoos and otos. Some are largely responsible for the strong, fragrant, bitter and pungent character of Mediterranean dishes, because of volatile aldehydes, ketones, aliphatic acids, and non-volatile secoiridoid-(seco)monoterpenes (Borzillo et al., 2000; Uccella, 2001a, 2003b; Gutierrez-Rosales et al., 2003). Oleuropein 1, the major BP from olives (Figure 37.1), its analogs and derivatives, play anticancer (Trichopoulou et al., 1995), anti-HIV (Lee-Huang et al., 2003), antioxidant (Saija et al., 1998), anti-inflammatory (Beauchamp et al., 2005) and antibacterial (Bisignano et al., 1999) roles. Olive BPs and seco-BPs, derivatives, i.e. the cathecol-like hydroxytyrosol 2, as pro-oxidants, and polymers, cross-reacting with proteins in must and veiled evoos, impart a peculiar haze, typical for fresh products, with their peculiar content of asclepic acid, the oleic positional isomer, belonging to the n-7 family, i.e., the Z-11-octadecenoic acid, a fundamental component for cell membrane fluidity, whose amount is largely affected by the chilling effect occurring in the olive growing orchard (Filippelli et al., 2008).
37.2 OLIVE BP MECHANISMS Olive trees are constantly biosynthesizing vast and diverse bio-organic molecules, for their essential metabolic roles found in every plant. They, besides nucleotides, amino acids, carbohydrates, organic acids, acyl lipids, and phytosterols (Sivakumar et al., 2006), produce several other phytomolecules too which are in the great majority, widely perceived
333
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
334
SECTION | I
Metals, Electrolytes and Other Components
OH O
RO
RO
R1
O COOMe
8 COOMe
10 9
2
3
1
4
6
7
O
O
H
5
O glc
OH
O
Hydroxytyrosol, 2; R1=OH Tyrosol, 3; R1=H
Oleuropein, 1; R=2 Ligstroside, 4; R=3
Me
Hydroxytyrosilelenolate, 5a; R=2 Tyrosilelenolate, 5b; R=3
FIGURE 37.1 Major biophenols and secoiridoids from olive fruits.
as biologically insignificant and not appearing to participate directly in growth and development. Therefore, BPs are classified as secondary metabolites (Uccella, 2002). These olive phytomolecules avoid pathogen invasion by using defense systems, even if a specific disease resistance is lacking. In fact, olives and pathogens are in a delicate relationship controlled by the environment, temperature and humidity. When favorable for the pathogen, the invasion of the olive orchard easily occurs (Uccella, 2001c). The olive otherwise overcomes pathogen attack with phytomolecular defense against external biological injuries, using combined strategies of quite different characteristics. When of the structural type, a physical barrier prevents pathogen entrance and invasion. The alternative defense is through phytomolecular reactions, occurring in the plant tissue and/or cell, with a great diversity of antimicrobial and antiviral phytoagents. They work by inhibiting pathogen growth in olive orchard trees. Relevant plant-formed anti-pathogens are phytoalexins, initially indicated by de novo synthesis after plant tissue infection (Muller and Börger, 1940). Phytoalexins, lowweight phytomolecules synthesized by and accumulated in plants after pathogen exposure, later receive consensus and gain general acceptance as a new working definition (Paxton, 1981). Phytomolecules, already in plant tissues prior to infection and/or produced from preformed constituents during infection, were excluded. The phytoalexin hypothesis induced the phytoanticipin concept of lowweight phytomolecules in plants before pathogen challenge RO
O COOMe
RO
or produced after infection solely from pre-existing constituents (VanEtten et al., 1994). The important distinction between phytoalexin and phytoanticipin is not the molecular structure, but the production. Some bio-product behaves as phytoalexin and phytoanticipin, even in olive trees. From preformed 1, its aglycone 6 is released from injured olive tissue by preformed glucosidases during tissue decompartmentalization (Figure 37.2), and the phytomolecule 1 becomes a phytoanticipin. Some olive phytoanticipins have two or more functional groups, especially if they belong to single phytomolecules, like seco-BPs. These imply hybrids, the combination of two or more functional groups, with the goal of creating an entity, more defense effective than their individual structural groups. These combo-phytomolecules, that is with phytoanticipin and phytoalexin activity, are indeed more powerful than their precursor groups. Reasons aren’t always well known. Most tantalizing is seco-BPs, such as 1. The hybrid phytomolecules are much better than their respective building blocks. The olive drupe, just by combining two phytomolecules, seco 1b and BP 2, in the proper way by the ester functional group sometimes has an activity increase by a factor 102, 103 or even more. Although combo-phytomolecules are often the combination of natural biosynthesized phyto-precursors, hybrid phytoanticipins must be considered as new molecular species, since they have themselves a novel functional combo-activity, which surpasses components and sometimes works in unexpected ways, preventing O COOMe
RO
O COOMe
OH
RO
O COOMe OH
β-glucosidase
CHO
Oleuropein, 1
OH
O CHO
CHO
CHO
OH Oleuropeinaglycon 6a; R=2
Enololeuropeindiale 6b; R=2
FIGURE 37.2 Formation of aglycons 6a–d from oleuropein 1 by β-glucosidase.
Oleuropeindiale 6c; R=2
Oleuropein monohydrate 6d; R=2
CHAPTER | 37 Olive Biophenols and Conventional Biotechnology from Mediterranean Aliment Culture
or relieving diseases in olive fruits. A major driving force in the hybrid phytoanticipin bio-development is to overcome one of the worst things that can happen to a phytomolecule, the growth of resistance in its target pathogen population. In most such hybrids, the two phytoanticipin-like portions, also called biodefense-phore, have independent modes of action that make the emergence of defense resistance less likely. In the seco-BP 1, the two molecular parts are connected with a linker no more elaborate than an ester bond, which combines the o-diphenolic moiety of 2, fast acting against free radicals, and the slow-acting oleoside 1b, the agent able to inactivate all myxoviruses (Heinze et al., 1975; Renis, 1975) after mild acid hydrolysis to elenolate 5. Therefore, the first part behaves like a phytoalexin, and the second as a phytoanticipin, releasing both carbonyl groups, which interact with nucleophiles found in the attacking pathogens (Konno et al., 1999; Uccella, 2001c). The defense treatment with the fast-acting antiradical groups from the BP 2 part may not be efficient enough and lead to reoccurrence of the pathogenic effect, which causes disease in the olive drupes. But, if in addition, the olive tissues have a slow-acting anti-pathogen agent like the secoglucoside, releasing the seco-dialdehydes 6c by enzymatic reaction, there is a chance that it will kill the remaining parasites that have not been killed by the previous acting agents. The olive biosynthesis is very careful to ensure that the activity of each phytomolecular part is not compromised by the site at which the two are linked together. Ester derivatives and glucosides, such as the phytomolecule seco-BP 1, do not adversely affect the parent molecular structures that have anti-pathogen activity. The hybrid phytoanticipin is more effective against parasite-sensitive and -resistant than the individual precursor phytomolecules alone, i.e. 2 and 1b, or a cocktail made of a 1:1 molar ratio of the two. There are several reasons why the hybrid version improves the phyto-treatment’s efficacy in olive drupes. The hybrid may increase cellular uptake, noting that the lipophilic seco-moiety of 1 can have problems getting into the cell’s aqueous interior that the polar glucose and 2 can alleviate, and it is easy to make water-soluble salts of the combo-phytomolecules via the o-diphenolic group. The 1b molecular structure with its glucoside serves to prevent the rearrangement of the otherwise released 6c into the 5. The o-quinone, formed from 2 after the free radical quenching, is able to undergo addition to nucleophyles from pathogen molecules, when they invade olive fruits. This is an interesting example of nature’s adaptation of imine reaction to produce derivatives possessing a carbon–nitrogen double bond for the defense of olives against invading pathogens. The very stable imines thus formed kill the enemies of live dupes by releasing four carbonyl groups from a single phytoanticipin molecule. The anti-parasite hybrid, designed by nature in the olive tissue, combines the two carbonyl groups, originated from the initial structure of 2, with the other two from the oleoside 11-methylester 1b, when this is activated by a
335
β-glucosidase, taking on the role of blocking four distinct amino groups through a long bridge of 11–12 atoms apart. The seco-BP hybrid phytoanticipin is thus more effective against pathogens and has an advantage over a simple cocktail of the two individual parts, the 2 and the 1b. In a cocktail of the two phytomolecules, the separate agents are unlikely to make a very strict chelation with four branches among distinct amino groups. In the hybrid seco-BP, the defensive agent tags along with the reactive sites into the attacking cells on four different and distant sensitive centers in a concerted manner. The o-quinone part of the hybrid phytomolecule gets a large chance to act as cytotoxic moiety, when the novel phytoalexin is generated by quick β-glucosidase hydrolysis of the 1b, elimination of glucose and formation of the 6c, with the spacer serving now as an active agent for toxicity against invaders of olive drupes. During its long life, the olive tree must survive and react to the attack of hundreds of thousands of flies. Although damaged, it shows an excellent resistance and recovery ability, being able to generate every two years a good yield of olive drupes for the MAC evoos and otos. The helpful strategy is achieved by the combo-phytomolecules of BP and seco-structure (Uccella, 2001c), i.e., the direct action provided by BPs themselves and the induced one on the masked functional group of the acetal-glucoside, which, by instant excitation and due to the presence of pathogen elicitors, manifests the phytoalexins 6, under the endogenous control of β-glucosidase activity (Bianco et al., 1999a). The take-home message is that olive oil experts, using conventional biotechnology in processing olive fruits, should consider carefully what is the fate of bioactive phytomolecules, such as seco-BP phytoanticipin, for the production of olive functional food with MAC reminiscence. After all, however familiar the individual building blocks of seco-BP 1 may be, the hybrid combo-phytomolecule is at its core an important entity for the olive drupe and for the health of the consumer, with an identity more complex than its precursor parts, the seco and BP ones.
37.3 THE BPs IN MAC OLIVE DRUPES The seco-moiety in Olea europaea is derived from oleosides, which are characterized by an exocyclic 4,5-ene functionality. This type of phytomolecule is limited in distribution to the Oleaceae family, without any structural relation to coumarine-like molecules (Bianco et al., 2001). It is biosynthesized from mevalonic acid as the source of ten carbon skeletons, followed by geraniol, 10-hydroxygeraniol, iridodial, iridotrial-hemiacetal, deoxyloganic acid, 7-epiloganin, and ketologanin. The conversion of the monoterpenoid can be a single-step reaction through a Baeyer-Villiger-type intermediate. The reaction is initiated by rupture of the peroxide bond, followed by cleavage of the 7,8-bond and simultaneous abstraction of H-9 to give rise to 6c. BP 1 and ligstroside 4 are the major BPs and both have similar
336
SECTION | I
Metals, Electrolytes and Other Components
COOR COOR O
HO HO
O O O
O
HO
O
HO Demethyloleuropein, 1a Oleuropein, 1
OGlu
OGlu COOH HO
OH
HO + O
O HO OGlu
Oleoside-11-methylester, 1b
Hydroxytyrosol, 2
FIGURE 37.3 Oleuropein metabolites from MAC olives.
molecular structures, differing only in one OH group on the aromatic ring, i.e. moieties of BP 2 and tyrosol 3, esterified with the same carboxylic group of the seco-unit of oleoside 11-methylester 1b (Figure 37.3). Seco-BP 1 results and is the most abundant component of olive mesocarp, with the maximum concentration reached before the maturity stage, followed by a decline associated with drupe maturity. When its decline begins, more soluble derivatives, such as 2 and 1b, and demethyloleuropein 1a (Sivakumar et al., 2005) giving rise to a lower-reacting aglycone 6c are formed. Infection of olive drupes could initiate a multicomponent defense response which gives rise to the accumulation of peculiar BP molecules such as 3-glucoside, i.e. tyrosol 8-O-β-D-glucopyranoside, known as salidroside and 2-glucosides (Bianco et al., 1999b) crucial for successful antagonism to causal agents of diseases. Thus, the reproduction of flies could be drastically reduced suggesting that the high levels of such BPs, present in olive fruits, exert a negative effect on feeding insects. The biosynthesis of salidroside might be channeled into a preferential metabolic route starting from tyrosine to tyramine, via tyrosine decarboxylase, with an enzyme catalyzed conversion through the tyramine oxidase, a mono-amine oxydase, leading to 4-hydroxyphenylacetaldehyde, followed by reduction to 4-hydroxyphenylethanol with its glucosylation by 3-glucosyltransferase to the final BP glucoside (Bianco and Ramunno, 2006). A similar metabolic pathway might lead to the formation of 2-glucosides. As part of the response to pathogen infection, olive drupe accumulate soluble and cell-wall-bound BPs because incorporation of these phytomolecules into the olive cell leads to a fortified barrier against pathogens and raising the amounts of BPs might positively affect the drupe resistance response. The genomic expression of a tyrosine decarboxylase in olive drupes channels tyramine into the production of BP glucosides as potential storage
phytomolecules. During the growth and ripening of olive drupes, several enzymatic endo-activities are in operation, with cytoplasmatic β-glucosidases still active at the green ripening stage (Sivakumar et al., 2007) and decreasing at the black stage. Activation of a defense-related phenylpropanoid pathway in olive plays an important role in the drupe defense system. When the olive is being wounded, lignin is synthesized and acts as a physical barrier against pathogen invasion, and phytomolecules with anti-microbial activity are produced (Trombetta et al., 2002). Enzyme activities in the phenylpropanoid pathway, such as phenylalanine ammonia-lyase, peroxidase and phenoloxidase, contribute to the limitation of pathogen invasion in the olive tissue.
37.4 MAC OLIVE ENZYMES In olive research, it is not clear how bio-active molecules such as seco-BPs 1 and 4 and at a late stage of maturity their demethylated derivatives and hydrolytic enzyme systems, i.e. β-glucosidases and esterases, might coexist in the fruit cells. During pathogen response, stored olive phytomolecules may be mobilized either by fusing the two vacuole moieties or by delivering specific enzymes to storage vacuoles. Evidence in support of this interpretation has come from investigations using several types of probes, including colorsensitive reactions (Mazzuca and Uccella, 2002) and lytic enzyme activity (Konno et al., 1999). In fact, while triacylglycerols are initially stored in olive cell vacuoles, they are separated after mechanical processes, such as drupe crushing and kneading, and at the same time, all the other biomolecules, which are compartmentalized in the olive cell, become available for biomolecular reactions after the physical mixing process (Uccella, 2001b). Most of the phytomolecular reactions are brought about by vacuolar fusion when
337
CHAPTER | 37 Olive Biophenols and Conventional Biotechnology from Mediterranean Aliment Culture
the conventional biotechnology occurring within the extraction system.
37.5 CONVENTIONAL BIOTECHNOLOGY ON MAC OLIVES Since the regulation of BPs accumulation determines the final quality of evoos and otos, an understanding of basic BP metabolism in the olive fruit would be clearly advantageous. Olive callus cultures can be used as a model, standardizing the growth conditions of the olive callus, compared to working with field-grown material (Williams et al., 2000). Olive callus cultures have proven to be very stable and are free from pests and temperature effects and, therefore, are a suitable supply of experimental material for biochemical experiments (Rutter et al., 1998). Olive callus has an acyl composition that is comparable to developing fruit (Williams et al., 2000). It has been shown that olive callus is a reliable material for the investigation of olive lipid metabolism (Williams et al., 2000), for its ability to accumulate triacylglycerols, rich in oleate (Williams et al., 1993), substantial amounts of the polyene fatty acid, linoleate and α-linolenate (Williams et al., 1998), as well as lipoxygenase isoforms (Williams and Harwood, 2000). In cell culture, elicitors have been shown to induce the biosynthetic pathways of BPs, in order to elucidate the mechanisms that underlie the steps of their biosynthesis and accumulation (Saigne-Soulard et al., 2006). From biomimetic experiments on olive coratina cultivar, the calli, fed with phenylalanine and methyl jasmonate, the elicitor, induced the BP pathways. Under optimum callus conditions, HPLC chromatogram of soluble BP fraction shows the presence of 1 and 4, 2, 3, 1b and 1a (Figure 37.4), which may be formed via the shikimate and/or malonate pathways. Seco-BP 1 appeared to be the main BP component in olive callus, with its known structural subunits (Gikas et al., 2007). Seco-BP 1a is a hydrolysis product of the
Oleuropein
mAU 2500
1000 500
Ligstroside
Hydroxytyrosol
1500
Demethyloleuropein
2000 Tyrosol Oleoside-11-methyl ester
enzymes are allowed to come in contact with their reagent substrates. During crushing and kneading steps, detectable amounts of proteins are transferred into the lipidic phase of evoos showing marked enzymatic activity (Georgalaki et al., 1998), such as lipoxygenase, catechol oxidase, tyrosinase or polyphenoloxidase, monophenol, dihydroxyphenylalanine: oxygen oxidoreductase. The catechol oxidase present in the olive drupe plays an important role as a bifunctional, coppercontaining enzyme (Valgimigli et al., 2001). This enzyme uses molecular oxygen to catalyze two different reactions, the hydroxylation of monophenols, such as tyrosine to the corresponding o-diphenol by monophenolase-o-cresolase activity, and the subsequent oxidation of colorless o-diphenols to reddish-brown o-quinones with diphenolase or catecholase activity. The o-quinones thus generated condense to polymeric forms of the brown melanin pigments through a series of subsequent enzymatic and non-enzymatic reactions. Although the physiological function of catechol oxidase in olive drupes is only partially understood, melanin synthesis can be correlated with differentiation of reproductive organs, virulence of pathogens, and tissue protection after injury. In addition, catechol oxidase is responsible for the enzymatic browning that takes place during maturation and senescence or damage during post-harvest handling. In fact, the volatile aroma and odor phytomolecules are generated by enzymatic reactions occurring on free fatty acids, which contain multiple double bonds, while triacylglycerols and pure water dispersion of free fatty acids do not reveal any aroma or bitterness. Some seco-BPs release bitter and enhanced mouth-feel characteristics when subject to βglucosidases activity with the elicitation of pungent taste due to the seco-dialdehyde moieties 6c. Seco-BPs become biologically active only in response to olive tissue damage by the activity of endogenous enzyme β-glucosidases (Bianco et al., 1996). In healthy olive cells, sub-cellular compartmentalization allocates β-glucosidases and seco-BP phytomolecules (Konno et al., 1999). The olive drupe response to external stress generates the unstable aglycon intermediates 6c by β-glucosidase activity, which is converted into various bio- and technomolecular products, including dialdehydes 6c and rearranged elenolates 5a and 5b (Uccella, 2001b). Wounding not only activates β-glucosidase reactivity, but also causes olive cellular de-compartmentalization, which brings the first and rapid process of contact between the activated β-glucosidase enzyme and seco-BP glucosides. Thus, this increases the proportion of their bio- and techno-molecular derivatives with enhanced palatability of the MAC producing evoos and otos. A number of olive phytomolecules contributes to the fruttato gusto of evoos (Uccella, 2001c), but the relationship between the levels of these bio- and techno-components and their original molecular structures (Bastoni et al., 2001) and content into olive drupes is mostly unclear (Bianco et al., 2001). The production system must be still considered merely due to physics and also to
0 0
10
20
30
40
50
min
FIGURE 37.4 HPLC profile of soluble biophenols from coratina callus (λ 240).
338
SECTION | I
conjugated methyl ester present in 1 (Figure 37.3). Moreover, the β-glucosidase activity has been found in olive callus. Figure 37.5 shows changes of β-glucosidases in coratina olive cultivar callus. Even though β-glucosidase activity has been detected in callus, some of the degradation aglycone
molecules, such as 6c, demethyloleuropeindiale, oleacein and other aldehydic derivatives, are not found. The HPLC profile reveals the absence of degradation molecules, explained by enzymatic complexes, which, in a non-pathogenic environment, are not activated to generate the phytoalexins 6, via hydrolysis on seco-glucosides, typical of MAC foods. In fact, the olive callus culture system biosynthesizes major olive BPs (Figures 37.3 and 37.4), but does not produce aldehydes, such as 6. In a pathogenic environment, β-glucosidase has been highly active in coratina olive fruits, releasing the more defensive aglycone of 1a, as a pre-protection against pathogens (Sivakumar et al., 2006). In a non-pathogenic environment, the β-glucosidase hydrolysis on the seco-BPs is latent, being involved in other biological processes, growth and development, rather than aldehyde formation from the seco-BP phytoanticipin. The callus culture with elicitors does not precisely describe the rather complex interplay between olive plant and external stress, although it meets the basic requirement to elucidate elicitor impact on BP biosynthesis (Saigne-Soulard et al., 2006) and gives a rational interpretation of MAC synergistic effect on human well-being and health.
3.5
mol mg protein−1 h−1
3.0 2.5 2.0 1.5 1.0 0.5 0 1
2 3 Culture periods (weeks)
4
FIGURE 37.5 β-Glucosidase activity in olive coratina cultivar callus culture.
CHO
CHO
1
Metals, Electrolytes and Other Components
CHO
2 O
O
CHO HO
Iridodial
HO
Iridohemiacetal
Hemiacetal of iridotrial
2
COOH
3
OH
COOH
COOH
2 O
O
OGlu
O
HO
OGlu
Loganic acid
Deoxyloganic acid
Deoxyloganic acid aglicone
3
COOCH3
COOCH3
COOCH3
COOH HO
3
OH
OHC
3
O
O
OGlu
OGlu
Loganin
Secologanin
FIGURE 37.6 The seco-biophenol biosynthetic pathway.
HOOC
3 O
OGlu
Secoxyloganin
O
O
OGlu
Oleoside-11-methylester
CHAPTER | 37 Olive Biophenols and Conventional Biotechnology from Mediterranean Aliment Culture
Only at a late maturity stage or in some peculiar cultivars, i.e. the coratina, the olive drupe activates the specific pathway leading to 1a and then by β-glucosidase to the demethyloleuropeindiale 6e, a molecular structure similar to 6c. 3,4-Dihydroxyphenylethyl-4-formyl-3-formylmethyl-4hexenoate, also known as 3,4-DHPEA-EDA, as metabolite of 1, elicits antioxidant activity and a strong ACE-inhibition, therefore named oleacein (Hansen et al., 1996). This is a distinction from seed oils, together with the relative triacylglycerol ratio, which contributes directly to the characteristic taste, and indirectly to the aroma and odor of evoos, because of the high resistance against auto-oxidation, well-being and health benefits. When coratina is sampled at different maturity stages, it reveals 1a at early and late harvesting, with oleacein at a late stage only. Hardy’s mammoth green olives show oleacein as the major seco-BP, as cultivar-dependent and precursor of 1, via a novel bio-synthesis, different and alternative to that widely accepted and reported in Figure 37.6. The new route avoids mevalonic sequence for the seco biosynthesis and leads to seco-BP 1 via oleacein, then 1-aglicone (Ryan et al., 2002). The metabolic relationship between the two seco-BPs is straightforward and requires an enzymatic hydrolysis through a cultivar-dependent, specific enzyme which, at the late maturity stage and under stress, releases 1a, as shown in Figure 37.3, and by the experiment on olive leaves in plastic bags, stored at 37°C without air for several hours (Paiva-Martins and Gordon, 2001). The required decarboxylation resembles the free radical decomposition, well known in simple α,β-unsaturated organic acids. The decarboxylation to oleacein could occur either in olive leaves and drupes during milling and malaxation of evoos, or even during olive storage before processing from the olive orchard to the extraction factory. As originally found in ripe leccino olive and in other different Italian cultivars (Ragazzi and Veronese, 1973; Regazzi et al., 1973) the oleacein precursor is present in ripe and stressed olives, and harvested at a late maturity stage, although in variable concentrations, thus becoming a territorial marker of olive agri-management for late crop harvesting and for long storage of olive drupes without air before processing. Oleacein, linked to its precursor 6e, undergoes a slow rearrangement to a ring closure similar to 5, because the carboxyl group lacking in the molecule does not activate the enol-formation, such as 6b. The phytomolecular relationship between the hydrolytic derivatives of seco-BP 1 is linked to different esterase reactivity for the methyl ester scission, as cultivar characteristics greatly influence the hedonic-sensory attributes of olive products, i.e. MAC evoos and otos.
SUMMARY POINTS ●
Natural olive products and ingredients from MAC have an important role in human health and well-being. The modern technology of olive transformation into evoos
●
●
339
and otos should rely on MAC traditional procedures, with the rational understanding based on the food science. The biomimesis with olive callus has inherited the potential biosynthesis for seco-BPs, towards the interpretation of olive metabolism, disease resistance and regulation, key hedonic-sensory response in evoos and otos. The phytomolecular mechanisms of soluble BPs and seco-BPs in bio- and techno-processes provide insights to improve olive-growing and olive-mill and storage technology, for the best competitive quality of final products and a positive effect on the MAC food chain, as consumers, better informed about well-being and health aspects, are more demanding in product purity, quality and territorial identity. Experts, renowned for their emphasis on exquisitely fresh and flavored ingredients, linked to the territoir, are fussy about evoos and otos from MAC, with especial favor local products, which have a deep, fruity flavor with a slight spicy and bitter bite. The MAC gusto of evoos and otos, like wine, depends not only on the variety and ripeness of the olive drupe, but also on the essential phytomolecular composition of some of the famous members in the mythical Mediterranean triad of food crops, Spermo, the wheat, Oinos, the grape, and Elaias, the olive, fusion of modern functional foods for the well-being and health of consumers.
ACKNOWLEDGMENTS The authors thank Dr. C. Briccoli Bati and G. Godino, Istituto Sperimentale per l’Olivicoltura, c/da Li Rocchi, Rende (CS), Italy for providing olive callus establishing facilities. Also we thank Dan Vail for critical reading of the manuscript.
REFERENCES Bastoni, L., Bianco, A.D., Piccioni, F., Uccella, N., 2001. Biophenolic profile in olives by nuclear magnetic resonance. Food Chem. 73, 145–151. Beauchamp, G.K., Keast, R.S.J., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A.S., 2005. Ibuprofen-like activity in extravirgin olive oil. Nature 437, 45–46. Bianco, A., Ramunno, A., 2006. The chemistry of Olea europea. Studies Nat. Prod. Chem. 33, 859–903. Bianco, A.D., Buiarelli, F., Cartón, G.P., Coccioli, F., Muzzalupo, I., Polidori, A., Uccella, N., 2001. Analysis by HPLC-MS/MS of biophenols in olives and oils. Anal. Lett. 34, 1033–1051. Bianco, A.D., Piperno, A., Romeo, G., Uccella, N., 1999a. NMR experiments of oleuropein biomimetic hydrolysis. J. Agric. Food Chem. 47, 3665–3668. Bianco, A.D., Muzzalupo, I., Piperno, A., Romeo, G., Uccella, N., 1999b. Bioactive derivatives of oleuropein from olive fruits. Ibid, 47, 3531–3534. Bianco, A.D., Mazzei, R.A., Melchioni, C., Romeo, G., Uccella, N., 1996. Molecular composition and hedonic-sensory quality of olive oil: Monoterpenes and biophenols. In: Porretta, S. (ed.), Research and Innovation in Food Industry, Vol. 2. Ciriotti, Pinerolo, Italy, pp. 92–103.
340
Bisignano, G., Tomaino, A., Lo Cascio, R., Crisafi, G., Uccella, N., Saija, A., 1999. On the in vitro antimicrobial activity of oleuropein and hydroxytyrosol. J. Pharm. Pharmacol. 51, 971–974. Borzillo, A., Iannotta, N., Uccella, N., 2000. Oinotria table olives: Quality evaluation during ripening and procession by biomolecular components. Eur. Food Res. Tech. 212, 113–121. Filippelli, L., Oliviero Rossi, C., Uccella, N., 2008. Z, E and PositionalMonoenoic Phyto-Fatty Acids in MAC Functional Food Influencing Membrane Fluidity by DSC and NMR Experiments. First European Food Congress, Ljubljana, Slovenia, 4–9 Nov. Elsevier Ltd Inc, Amsterdam, 2008, Vol. 1, 125–126. Georgalaki, M.D., Sotiroudis, T.G., Xenakis, A., 1998. The presence of oxidising enzyme activities in virgin olive oil. J. Am. Oil Chem. Soc. 75, 155–159. Gikas, E., Bazoti, F.N., Tsarbopoulos, A., 2007. Conformation of oleuropein, the major bioactive compound of Olea europea. J. Mol. Struct. Theochem. 821, 125–132. Gutierrez-Rosales, F., Rios, J.J., Gomez-Rey., Ma, L., 2003. Main polyphenols in the bitter taste of virgin olive oil. Structural composition by on-line high-performance liquid chromatography electrospray ionisation mass spectrometry. J. Agric. Food Chem. 51, 6020–6025. Hansen, K., Adsersen, A., Brogger Christensen, S., Rosendal Jensen, S., Nyman, U., Wagner Smitt, U., 1996. Isolation of an angiotensin converting enzyme (ACE) inhibitor from Olea europea and Olea lancea. Phytomed 2, 319–325. Heinze, J.E., Hale, A.H., Carl, P.L., 1975. Specificity of the antiviral agent calcium elenolate. Ibid 8, 421–425. Konno, K., Hirayama, C., Yasui, H., Nakamura, M., 1999. Enzymatic activation of oleuropein: A protein crosslinker used as a chemical defence in the privet tree. Proc. Natl. Acad. Sci. 96, 9159–9164. Lee-Huang, S., Zhang, L., Lin Huang, P., Chang, Y.T., Huang, P.L., 2003. Anti-HIV activity of olive leaf extract (OLE) and modulation of host cell gene expression by HIV-1 infection and OLE treatment. Biochem. Biophys. Res. Commun. 307, 1029–1037. Mazzuca, S., Uccella, N., 2002. β-Glucosidase releasing of phytoalexin derivatives from seco-biophenols as defence mechanism against pathogenic elicitors in olive drupes. Acta Horticol 586, 529–531. Muller, K.O., Börger, H., 1940. Experimentelle untersuchungen über die phytophthora-resistenz der kartoffel. Arb. Bio. Reichsasnstalt. Landw. Forstw. Berlin. 23, 189–231. Paiva-Martins, F., Gordon, M.H., 2001. Isolation and characterization of the antioxidant component 3,4-Dihydroxyphenylethy-4-formyl-3formilmethyl-4-hexenoate from olive (Olea europea) leaves. J. Agric. Food Chem. 49, 4214–4219. Paxton, J.D., 1981. Phytoalexins, a working redefinition. Phytopathol. Z. 101, 106–109. Regazzi, E., Veronese, G., Gmatto, A., 1973. Demethyloleuropein, a new glucoside isolated from ripe olives. Annali di Chimica 63, 13–20. Regazzi, E., Veronese, G., 1973. Investigation on the presence of demethyloleuropein in different olive cultivars. Ibid. 63, 21–27. Renis, H.E., 1975. Inactivation of myxoviruses by calcium elenolate. Antimicrob. Agents Chemother. 8, 194–199. Rutter, A.J., Sanchez, J., Harwood, J.L., 1998. The effect of dimethoate on lipid biosynthesis in olive (Olea europaea) callus cultures. Pytochemistry 47, 735–741. Ryan, D., Antolovich, M., Herlt, T., Prenzler, P.D., Lavee, S., Robards, K., 2002. Identification of phenolic compounds in tissues of the novel olive cultivar Hardy’s Mammoth. J. Agric. Food Chem. 50, 6716–6724. Saigne-Soulard, C., Richard, T., M’erillon, J.M., Monti, J.P., 2006. 13C NMR analysis of polyphenol biosynthesis in grape cells: Impact of various inducing factors. Analytica Chimica Acta 563, 137–144.
SECTION | I
Metals, Electrolytes and Other Components
Saija, A., Uccella, N., 2001. Olive biophenols: Functional effects on human wellbeing. Trends Food Sci. Technol. 11, 357–363. Saija, A., Trombetta, D., Tomaino, A., Lo Cascio, R., Princi, P., Uccella, N., Bonina, F., Castelli, F., 1998. “In vitro” evaluation of the antioxidant activity and biomembrane interaction ofthe plant phenols oleuropein and hydroxytyrosol. Int. J. Pharm. 166, 123–133. Simopouloas, A., 2004. Omega-3 fatty acids and antioxidants in edibles wild plants. Biol. Res. 37, 263–267. Sivakumar, G., Briccoli Bati, C., Uccella, N., 2007. Demethyloleuropein and β-glucosidase activity in olive fruits. Biotechnol. J. 2, 381–385. Sivakumar, G., Briccoli Bati, C., Perri, E., Uccella, N., 2006. Gas chromatography screening of bioactive phytosterols from mono-cultivar olive oils. Food Chem. 95, 525–528. Sivakumar, G., Briccoli Bati, C., Uccella, N., 2005. HPLC-MS screening of the antioxidant profile of Italian olive cultivars. Chem. Natl. Comp. 41, 588–591. Trichopoulou, A., Bamia, C., Trichopoulou, D., 2009. Anatomy of health effects of Mediterranean diet: Greek EPIC prospective cohort study. BMJ 338:b, 2337–2340. Trichopoulou, A., Katsouyanni, K., Stuver, S., Tzala, L., Gnardellis, C., Rimm, E., Trichopoulos, D., 1995. Consumption of olive oil and specific food groups in relation to breast cancer risk in Greece. J. Natl. Cancer Inst. 87, 110–116. Trombetta, D., Saija, A., Bisignano, G., Arena, S., Caruso, S., Mazzanti, G., Uccella, N., 2002. Study on the mechanism of antibacterial action of some α,β-unsaturated aldehydes. Lett. App. Microbiol. 35, 285–290. Uccella, N., 2003a. Evolution of the Mediterranean Aliment Culture. Inform, Int. News Fats Oils Rel. Mat. 14, 494–495. Uccella, N., 2003b. Ulivo Alchemy: The bio- and techno-molecular approach to Mac-Mediterranean Aliment Culture. Ann. Chim. 93, 169–180. Uccella, N., 2002. The secoiridoid biophenols of olea europea L. drupes and the role of their metabolites. Acta Horticol. 586, 489–492. Uccella, N., 2001a. The olive biophenols: Hedonic-sensory descriptors of EVOO and WOTO in Mediterranean Aliment Culture. In: Spanier, A. (ed.), Food Flavour and Chemistry: Advanced in the New Millennium, Vol. 274. RSC, Cambridge, pp. 253–265. Uccella, N., 2001b. Olive biophenols: Novel ethnic and technological approach. Trends Food Sci. Technol. 11, 328–339. Uccella, N., 2001c. Olive biophenols: Biomolecular characterisation, distribution and phytoalexin histochemical localisation in the drupes. Trends Food Sci. Technol. 11, 315–327. Valgimigli, L., Sanjust, E., Curreli, N., Rinaldi, A., Pedulli, G.F., Rescigno, A., 2001. Photometric assay for polyphenol oxidase activity in olives, olive pastes, and virgin olive oils. J. Am. Oil Chem. Soc. 78, 1245–1248. VanEtten, H.D., Mansfield, J.W., Balley, J.A., Farmer, E.E., 1994. Two classes of plant antibiotics: Phytoalexins versus phytoanticipins. Plant Cell 1191–1192. Williams, M., Salas, J.J., Sanchez, J., Harwood, J.L., 2000. Lipoxygenase pathway in olive callus cultures (Olea europaea). Pytochemistry 53, 13–19. Williams, M., Harwood, J.L., 2000. Characterization of lipoxygenase isoforms in olive callus cultures. Biochem. Soc. Trans. 28, 830–831. Williams, M., Morales, M.T., Aparicio, R., Harwood, J.L., 1998. Analysis of volatiles from callus cultures of olive Olea europaea. Pytochemistry 47, 1253–1259. Williams, M., Sanchez, J., Hann, A.C., Harwood, J.L., 1993. Lipid biosynthesis in olive cultures. J. Exermental Bot. 268, 1717–1723.
Chapter 38
Production of Triterpene Acids by Cell-suspension Cultures of Olea europaea Yutaka Orihara1 and Yutaka Ebizuka2 1 2
Experimental Station for Medicinal Plant Studies, Graduate School of Pharmaceutical Sciences, The University of Tokyo, Japan Laboratory of Natural Products Chemistry, Graduate School of Pharmaceutical Sciences, The University of Tokyo, Japan
38.1 INTRODUCTION The olive (Olea europaea L.) is a well-known evergreen tree, native to the Mediterranean coast, of which the fruit and oil are used for food and cooking. Olives contain numerous triterpene acids including oleanolic acid (1) as the major one (Bianchi et al., 1992a, b). Recently, much attention has been paid to triterpene acids from the pharmaceutical viewpoint because of their anti-HIV (Kashiwada et al., 2000), anti-inflammatory (Safayhi and Sailer, 1997), and antitumor-promoting (Banno et al., 2005) activities, endothelin receptor-antagonist activity (Sakurawi et al., 1996), etc. The wax of the olive fruit contains predominantly oleanolic acid (Bianchi et al., 1992a), whereas that of the leaves contains a mixture of oleanolic acid and betulinic acid in a ratio of 7:2 (Bianchi et al., 1992b). In addition to these triterpene acids, triterpene diols, such as erythrodiol (oleanane type) and uvaol (ursane type) are also present, though in smaller quantities than the corresponding acids (Bianchi et al., 1992b). Triterpenes are concentrated mainly in the skin of the fruit. Their content in pomace olive oil (Orujo olive oil), which is an olive residue, is about 10-fold higher than in other types of olive oil. Key features of triterpene acids are summarized in Table 38.1. This chapter describes the induction of cell cultures, production of triterpene metabolites, and cDNA cloning of oxidosqualene cyclases (OSCs) from O. europaea cell cultures.
38.2 INDUCTION OF OLIVE CELL CULTURE The callus was induced from the young leaf stalks of the olive plant cultivated in the Experimental Station for Medicinal Plant Studies, The University of Tokyo (Chiba, Japan). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
TABLE 38.1 Key features of triterpene acids 1. Triterpenes are biosynthesized from 2,3-oxidosqualene by product-specific oxidosqualene cyclases 2. Lupeol, α-amyrin, and β-amyrin are the most common triterpene alcohols found in plants 3. Their C-28 methyl groups are easily oxidized to the corresponding carboxylic acids to form the stable triterpene acids betulinic acid, ursolic acid, and oleanolic acid, respectively 4. Further modifications including hydroxylation, oxidation, and glycosylation lead to biologically active triterpenoids and saponins in plant tissues, especially of medicinal plants 5. The main triterpene constituent of the olive plant (Olea europaea L.) is oleanolic acid accumulated in the fruit skin
Four types of agar media (DK agar medium: Murashige and Skoog’s agar media (Murashige and Skoog, 1962) supplemented with 2,4-dichlorophenxyacetic acid 1 mg L⫺1 and kinetin 0.1 mg L⫺1; CM agar medium; DK agar medium supplemented with 7% coconut milk, DK-NH4 and CM-NH4 agar media; DK and CM agar media supplemented with K⫹ in place of the original NH4⫹) were prepared for callus induction. DK-NH4 agar media gave the best results for callus induction. Callus induction of four other species of Oleaceae plants (Forsythia suspensa, Ligustrum japonicum, Ligustrum lucidum, and Jasminum mudiflorum) was attempted for comparison. In these trials, the same tendency for the removal of NH4⫹ ion from the medium to result in better callus induction and cell growth was observed.
341
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
342
FIGURE 38.1 Agar (left) and liquid (right) cultures of O. europaea. Left: Three-week agar olive cell culture. Right: Three-week olive cell suspension culture (in a rotary shaker, 120 rpm). Both cultures were maintained at 25 °C in the dark.
Cell-suspension cultures were established from the callus in the DK-NH4 liquid medium in a rotary shaker (120 rpm) at 25 °C in the dark. After 3 weeks’ culture, the callus and the cell suspension cultures of olive showed good growth (Figure 38.1).
38.3 ISOLATION AND STRUCTURE DETERMINATION OF TRITERPENE ACIDS From 4-week liquid culture, the cells (2 kg f.w.) were harvested and extracted with methanol at room temperature. The methanol extract was first partitioned between ethyl acetate and water, and the water phase was further partitioned with n-butanol. The ethyl acetate fraction (3.7 g) and n-butanol fraction (6.4 g) were separated by silica gel column chromatography and HPLC to yield 11 compounds. Spectroscopic analysis, mainly 1H- and 13C- NMR (Table 38.2) easily identified the structures of these isolated compounds as oleanolic acid (3β-hydroxyolean-12-en-28-oic acid (1), 6 mg, Mahato and Kundu, 1994), maslinic acid (2α, 3β-dihydroxyolean-12-en-28-oic acid (2), 5 mg, Tchivounda et al., 1991), ursolic acid (3β-hydroxyurs-12en-28-oic acid (3), 5 mg, Maillard et al., 1992), pomolic acid (3β, 19α-dihydroxyurs-12-en-28-oic acid (4), 2 mg, Kakuno et al., 1992), rotundic acid (3β, 19α, 23-trihydroxyurs-12-en-28-oic acid (5), 17 mg, Nakatani et al., 1989), tormentic acid (2α, 3β, 19α-trihydroxyurs-12-en-28-oic acid (6), 2 mg, Numata et al., 1989), 2α-hydroxyursolic acid (2α, 3β-dihydroxyurs-12-en-28-oic acid (7), 3 mg, Numata et al., 1989), 19α-hydroxyasiatic acid (2α, 3β, 19α, 23-tetrahydroxyurs-12-en-28-oic acid (8), 10 mg,
SECTION | I
Non Fruit Aspects Including Mill Wastewater
Seto et al., 1984), isofucosterol (9, 7 mg), campesterol (10, 5 mg), and β-sitosterol (11, 83 mg). Compounds 3–8 are ursane-type triterpene acids, whereas compounds 1 and 2 are oleanane-type triterpene acids (Figure 38.2) and compounds 9–11 are phytosterols (Figure 38.3). The isolation yields of ursane-type triterpene acids were higher than those of the oleanane type. In good agreement with these isolation yields, direct HPLC analysis of the ethyl acetate fraction and n-butanol fraction demonstrated a higher content of ursane triterpenes than that of oleananes in this cell culture. As mentioned above, it is well known that among triterpene constituents, oleananes dominate in intact olive plants, with trace amounts of ursane-type triterpenes. HPLC analysis of the extracts prepared from fresh leaves or fruit of the mother olive plant showed only a trace amount of ursane-type triterpenes, i.e., the cultured cells of O. europaea produce more ursane-type triterpenes, which are minor in the mother plant, than oleanane-type triterpenes. Pomolic acid (4), rotundic acid (5), tormentic acid (6), 2α-hydroxyursolic acid (7), and 19α-hydroxyasiatic acid (8) are common plant triterpene acids but have not been reported from the olive. It is interesting to note that the olive is a potential producer of these triterpene acids. The biological activities of these ursane-type triterpene acids are well documented. Pomolic acid (4) induces apoptosis in cell lines from patients with chronic myeloid leukemia exhibiting the multidrug-resistance phenotype (Fernandes et al., 2007; Vasconcelos et al., 2007). Rotundic acid (5) shows significant broad antimicrobial activity against bacteria, yeasts, and filamentous fungi, due to its perturbation of membrane permeability (Haraguchi et al., 1999). Tormentic acid (6) shows hypoglycemic effects by initiating insulin secretion in a concentration-dependent manner (Ivorra et al., 1989). Tormentic acid is isolated from bruised strawberry fruit as a phytoalexin (Hirai et al., 2000). 2α-Hydroxyursolic acid (7) also shows antibacterial activities and is isolated together with ursolic acid (3), euscaphic acid, and tormentic acid (6) from Rostellularia procumbens (Zhang et al., 2007). Especially notable is that 2α-hydroxyursolic acid (7) shows antibacterial activity against meticillinresistant Staphylococcus aureus (MRSA) (Filho et al., 2007). 19α-Hydroxyasiatic acid (8) shows antibacterial activity against the Gram-negative bacteria Pseudomonas aeruginosa (Richard et al., 1994).
38.4 IMPLICATIONS OF STUDY RESULTS Successful production of triterpene acids by plant cell cultures has been documented. The production of bryonolic acid, an antiallergic triterpene acid, by cultured cells of several cucurbitaceous plants (Cho et al., 1992) is a good example. Cell cultures of plants, however, do not always produce the same secondary metabolites as those in the mother plants. Some callus cultures produce no trace of the
343
CHAPTER | 38 Production of Triterpene Acids by Cell-suspension Cultures of Olea europaea
TABLE 38.2 C
13
C NMR data of triterpene acids (1–8) isolated from cultured cells of O. europaea
1
2
3
4
5
6
7
8
1
38.3
47.7
38.4
39.0
38.8
47.9
47.9
47.8
2
27.0
68.5
26.9
28.1
27.6
68.5
68.5
68.8
3
78.4
83.8
78.0
78.1
73.6
83.8
83.9
78.3
4
38.6
39.8
38.3
39.3
42.8
39.8
40.0
43.6
5
55.1
55.8
54.9
55.8
48.7
55.9
55.8
48.0
6
18.2
18.8
18.0
18.9
18.7
18.9
18.8
18.6
7
32.5
33.1
32.7
33.6
33.3
33.5
33.4
33.1
8
39.1
39.8
39.1
40.3
40.3
40.3
39.8
40.4
9
47.5
48.1
47.2
47.7
47.8
47.8
48.0
47.8
10
36.9
38.5
36.5
37.3
37.2
38.4
38.4
38.3
11
23.1
23.9
22.9
24.0
24.0
24.1
23.7
24.1
12
121.8
122.4
124.8
128.0
127.8
127.7
125.5
127.9
13
144.2
144.8
138.2
140.0
140.1
140.1
139.2
139.9
14
41.6
42.1
41.7
42.1
42.1
42.1
42.5
42.1
15
27.7
28.2
27.8
29.3
29.3
29.3
28.6
29.2
16
23.2
23.6
24.0
26.4
26.4
26.4
24.9
26.3
17
46.2
46.6
47.3
48.3
48.3
48.3
48.0
48.2
18
41.1
41.9
52.6
54.6
54.6
54.6
53.5
54.5
19
46.0
46.4
38.8
72.7
72.7
72.7
39.3
72.6
20
30.6
30.9
38.5
42.3
42.3
42.4
39.4
42.3
21
33.9
34.1
30.4
26.9
27.0
27.0
31.0
26.9
22
32.5
33.1
36.5
38.5
38.5
38.5
37.4
38.4
23
28.0
29.3
27.9
28.7
68.1
29.3
29.3
66.6
24
15.5
17.6
15.0
16.7
13.0
17.6
17.4
14.3
25
15.1
17.4
15.4
15.5
17.3
16.7
16.9
17.3
26
16.7
16.7
16.5
17.1
16.7
17.3
17.4
17.3
27
25.7
26.1
23.2
24.6
24.6
24.7
23.9
24.6
28
180.6
180.1
179.9
180.7
181.0
181.2
179.9
180.7
29
33.0
33.2
16.7
27.1
27.1
27.1
17.6
27.0
30
23.5
23.7
20.8
16.4
16.0
16.8
21.3
16.7
Condition: CDCl3 (1, 3), C5D5N (others), each at 125 MHz.
344
SECTION | I
Non Fruit Aspects Including Mill Wastewater
FIGURE 38.2 Structures of triterpene acids isolated from cultured cells of O. europaea. Compounds 1 and 2 have oleanane skeletons, whereas compounds 3–8 have ursane skeletons. The amount of ursane-type triterpene acids produced by cultured cells of O. europaea is greater than that of the oleanane type.
38.5 cDNA CLONING OF OSC
FIGURE 38.3 Structures of phytosterols isolated from cultured cells of O. europaea. β-Sitosterol (11) is the main phytosterol of O. europaea cultured cells, and isofucosterol (9) and campesterol (10) comprise less than 10% of compound 11.
metabolites found in the mother plants, but produce metabolites not produced by the mother plants. This situation is exemplified by Glycyrrhiza glabra cell cultures, which do not produce glycyrrhizin, a major triterpene saponin of the mother plant, but produce betulinic acid (Hayashi et al., 1988) and soyasaponin I (Hayashi et al., 1990). It is interesting and important to determine which types of triterpenes cultured plant cells can produce in which amounts. If they produce specific triterpenes in large quantities, that production system would be beneficial for the supply of biologically active triterpene acids for clinical use, since the purification and isolation of triterpenes from complex mixtures are generally laborious. Understanding all of the enzymatic steps in the biosynthetic pathway leading to these biologically active compounds is also important, as once they are known and all the genes become available, reconstitution of the pathway in rapidly growing microorganisms could be expected to offer an alternative method for the production of the metabolites in the future.
Triterpenes are biosynthesized from the common precursor (3S)-oxidosqualene. More than 100 diverse triterpene skeletons are reported from nature (Xu et al., 2004) and are constructed at the cyclization step of oxidosqualene catalyzed by OSC. The different spectra of triterpenes produced by cultured cells and the mother plants can be explained by differential expression of OSCs. β-Amyrin synthase is likely to be highly expressed in the mother olive plant, whereas α-amyrin synthase is expressed in the cultured cells. Among olive OSCs, only two, lupeol synthase OEW and cycloartenol synthase OEX, have been reported from its leaves (Shibuya et al., 1999). cDNA cloning from the cultured cells by homology-based PCRs was conducted following the reported method (Kushiro et al., 1998). PCR products with degenerate primers were subcloned into a plasmid vector and 18 colonies were selected for sequencing. They consisted of 14 cycloartenol synthases (identical to OEX (Shibuya et al., 1999)), two lupeol synthases (identical to OEW (Shibuya et al., 1999)), and two new sequences. These new sequences were identical to each other and named OEA. The full-length sequence of OEA was obtained using the RACE method (Frohman et al., 1988). OEA is composed of a 2289-nucleotide openreading frame (ORF) that encodes a protein of 763 amino acids in length. The sequence was deposited in the DDBJ (accession number AB291240). OEA shows high sequence identity (74%) to the recently cloned dammarenediol-II synthase PNA from Panax ginseng (Tansakul et al., 2006). The ORF of OEA was amplified by PCR and ligated to the cloning site of the yeast expression vector pYES2. The resulting plasmid was transferred to lanosterol synthasedeficient Saccharomyces cerevisiae strain GIL77, and
CHAPTER | 38 Production of Triterpene Acids by Cell-suspension Cultures of Olea europaea
FIGURE 38.4 Structures of OEA products as determined in LC-APCIMS analysis. Compound 12 is the main product (about 60% of total triterpene products), followed by compound 13 (about 35%). LC-APCIMS: liquid chromatography-atmospheric pressure chemical ionization mass spectrometry.
products from the induction culture were analyzed using LC-APCIMS, which showed the presence of α-amyrin (12), β-amyrin (13), ψ-taraxasterol (14), and butyrospermol (15) (Figure 38.4). It is noteworthy that these products were identical to those of PSM, a multiproduct OSC from Pisum sativum (Morita et al., 2000). The major product of OEA is α-amyrin, in good agreement with the high production of ursane-type triterpene acids in the cultured cells. These results strongly suggest that OEA is highly expressed in the cultured cells and is a major contributor OSC for the production of ursane-type triterpenes.
38.6 ORIGIN OF URSANE-TYPE TRITERPENES IN HIGHER PLANTS More than 100 triterpene skeletons are constructed at the cyclization step of oxidosqualene catalyzed by OSCs (Xu et al., 2004). So far, more than 30 OSCs with different product specificity have been cloned from higher plants (Phillips et al., 2006). Some OSCs show high catalytic fidelity in producing a single cyclization product, suggesting the existence of more than 100 different OSCs. However, this estimate should be greatly reduced due to the rather common occurrence of multiproduct synthases with leaky product specificity (Phillips et al., 2006). No OSC yielding α-amyrin as the sole product has been reported, and all OSCs from which the products include α-amyrin are multifunctional. They include OEA from O. europaea (Saimaru et al., 2007), PSM from Pisum sativum (Morita et al., 2000), At1g78960 (Kushiro et al., 2000) and At1g78500 (Ebizuka et al., 2003; Shibuya et al., 2007)
345
from Arabidopsis thaliana, KcMS from Kandelia candel (Basyuni et al., 2006), and two OSCs from Taraxacum officinale (our unpublished results). Furthermore, most phytochemical studies reported the presence of ursanetype triterpenes together with oleanane-type triterpenes. These findings strongly suggest that α-amyrin is produced only by multifunctional enzymes like OEA, and productspecific α-amyrin synthase may not occur in nature. This is in sharp contrast to oleanane and lupane triterpenes, as β-amyrin and lupeol are produced not only by monofunctional enzymes but also by multifunctional ones, although ursane-type triterpenes are widely distributed in higher plants along with oleanane- and lupane-type triterpenes. It is interesting to note that OEA shares significant amino acid sequence homology (74%) with PNA, a product-specific dammarenediol-II synthase from P. ginseng (Tansakul et al., 2006). In contrast, it shows less homology (55%) with PSM, a multifunctional OSC from P. sativum (Morita et al., 2000), which yields the same products as OEA. Considering the occurrence of dammarenediol-II in trace amounts in the fruit of O. europaea (Itoh et al., 1981) and in the closely related species Olea madagascariensis (Bianchini et al., 1988), OEA may be an intermediate OSC in the evolutionary process from β-amyrin synthase to dammarenediol-II synthase.
38.7 CONCLUSION Six ursane-type triterpene acids that have not been detected in intact plants were isolated and identified from cell cultures of O. europaea. They may be produced as phytoalexin-like stress compounds of the olive, because one of them, tormentic acid, was reported as a phytoalexin of the strawberry. This leads to the general hypothesis that most secondary metabolites produced by plant cell cultures are phytoalexin-like compounds, as the cultured cells are under heavy stress conditions similar to wound-healing tissue. The development of plant tissue culture systems for the production of pharmaceutically important compounds should be pursued further. An OSC cDNA, OEA, was obtained from cultured olive cells using homology-based PCRs and its enzymatic function was identified by expression in yeast. OEA is a multifunctional OSC producing a mixture of α-amyrin (12), β-amyrin (13), ψ-taraxasterol (14), and butyrospermol (15) (Figure 38.4). This result suggests that OEA is a main contributor OSC for the production of ursane-type triterpene acids in cell-suspension cultures of the olive. Three OSCs, OEX, OEW, and OEA, have so far been cloned from the olive plant, leaving the product-specific β-amyrin synthase responsible for the production of oleanolic acid, the major triterpene of the intact olive plant, undiscovered. Analysis of the transcription of each OSC is necessary for a better understanding of the metabolism of triterpenes in the olive.
346
SECTION | I
SUMMARY POINTS ●
●
●
●
●
●
Cell cultures of the olive plant (Olea europaea L.) were induced from leaf stalks in DK-NH4 medium (Murashige and Skoog’s medium (Murashige and Skoog, 1962) supplemented with K⫹ in place of the original NH4⫹, and as plant growth regulators, supplemented with 2,4-dichlorophenxyacetic acid 1 mg L–1 and kinetin 0.1 mg L–1) at 25 °C in the dark. Six ursane-type triterpene acids (ursolic acid, pomolic acid, rotundic acid, tormentic acid, 2α-hydroxyursolic acid, and 19α-hydroxyasiatic acid), and two oleananetype acids (oleanolic acid and maslinic acid) were isolated from the olive cell-suspension cultures. The amount of ursane-type triterpene acids produced by cell cultures was greater than that of the oleanane type. A multifunctional OSC named OEA was cloned using homology-based PCRs from the same cultured olive cells. The major product of OEA is α-amyrin (ursane skeleton), in agreement with the higher content of ursanetype triterpene acids in cultured olive cells. OEA is suggested to be a major contributor OSC for triterpene acid production in cultured olive cells.
REFERENCES Banno, N., Akihisa, T., Tokuda, H., Yasukawa, K., Taguchi, Y., Akazawa, H., Ukiya, M., Kimura, Y., Suzuki, T., Nishino, H., 2005. Anti-inflammatory and antitumor-promoting effects of the triterpene acids from the leaves of Eriobotrya japonica. Biol. Pharm. Bull. 28, 1995–1999. Basyuni, M., Oku, H., Inafuku, M., Baba, S., Iwasaki, H., Oshiro, K., Okabe, T., Shibuya, M., Ebizuka, Y., 2006. Molecular cloning and functional expression of a multifunctional triterpene synthase cDNA from a mangrove species Kandelia candel (L.) Druce. Phytochemistry 67, 2517–2524. Bianchi, G., Murelli, C., Vlahov, G., 1992a. Surface waxes from olive fruits. Phytochemistry 31, 3503–3506. Bianchi, G., Vlahov, G., Anglani, C., Murelli, C., 1992b. Epicuticular wax of olive leaves. Phytochemistry 32, 49–52. Bianchini, J.P., Gaydou, E.M., Rafaralahitsimba, G., Waegell, B., Zahra, J.P., 1988. Dammarane derivative in the fruit lipids of Olea madagascariensis. Phytochemistry 27, 2301–2304. Cho, H.J., Tanaka, S., Fukui, H., Tabata, M., 1992. Formation of bryonolic acid in Cucurbitaceous plants and their cell cultures. Phytochemistry 31, 3893–3896. Ebizuka, Y., Katsube, Y., Tsutsumi, T., Kushiro, T., Shibuya, M., 2003. Functional genomics approach to the study of triterpene biosynthesis. Pure Appl. Chem. 75, 369–374. Fernandes, J., Weinlich, R., Castilho, R.O., Amarante-Mendes, G.P., Gattass, C.R., 2007. Pomolic acid may overcome multidrug resistance mediated by overexpression of anti-apoptotic Bcl-2 proteins. Cancer Lett. 245, 315–320. Filho, A.A.daS., de Sousa, J.P.B., Soares, S., Furtado, N.A.J.C., Andrade e Silva, M.L., Cunda, W.R., Gregorio, L.E., Nanayakkara, N.P.D., Bastos, J.K., 2007. Antimicrobial activity of the extract and isolated compounds from Baccharis dracunculifolia D. C. (Astraceae). Z. Naturforsch. 63c, 40–46.
Non Fruit Aspects Including Mill Wastewater
Frohman, M.A., Dush, M.K., Martin, G.R., 1988. Rapid production of full-length cDNAs from rare transcripts. Amplification using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85, 8998–9002. Haraguchi, H., Kataoka, S., Okamoto, S., Hanafi, M., Shibata, K., 1999. Antimicrobial triterpenes from Ilex integra and the mechanism of antifungal action. Phytother. Res. 13, 151–156. Hayashi, H., Fukui, H., Tabata, M., 1988. Examination of triterpenoids produced by callus and cell suspension cultures of Glycyrrhiza glabra. Plant Cell Rep. 7, 508–511. Hayashi, H., Sakai, T., Fukui, H., Tabata, M., 1990. Formation of soyasaponin in licorice cell suspension culture. Phytochemistry 29, 3127–3129. Hirai, N., Sugie, M., Wada, M., Lahlow, E.H., Kamo, T., Yoshida, R., Tsuda, M., Ohigashi, H., 2000. Triterpene phytoalexins from strawberry fruit. Biosci. Biotechnol. Biochem. 64, 1707–1712. Itoh, T., Yoshida, K., Yatsu, T., Tamura, T., Matsumoto, T., Spencer, G.F., 1981. Triterpene alcohols and sterols of Spanish olive oil. J. Am. Oil Chem. Soc. 58, 545–550. Ivorra, M.D., Paya, M., Villar, A., 1989. Effect of tormentic acid on insulin secretion in isolated rat islets of Langerhans. Phytother. Res. 3, 145–147. Kakuno, T., Yoshikawa, K., Arihara, S., 1992. Triterpenoid saponins from Ilex crenata fruit. Phytochemistry 31, 3553–3557. Kashiwada, Y., Nagao, T., Hashimoto, A., Ikeshiro, Y., Okabe, H., Cosentino, L.M., Lee, K.H., 2000. Anti-AIDS agents 38. Anti-HIV activity of 3-O-acyl ursolic acid derivatives. J. Nat. Prod. 63, 1619–1622. Kushiro, T., Shibuya, M., Ebizuka, Y., 1998. β-Amyrin synthase. Cloning of oxidosqualene cyclase that catalyzes the formation of the most popular triterpene among higher plants. Eur. J. Biochem. 256, 238–244. Kushiro, T., Shibuya, M., Masuda, K., Ebizuka, Y., 2000. A novel multifunctional triterpene synthase from Arabidopsis thaliana. Tetrahedron Lett. 41, 7705–7710. Mahato, S.B., Kundu, A.P., 1994. 13C NMR spectra of pentacyclic triterpenoids—A compilation and some salient features. Phytochemistry 37, 1517–1575. Maillard, M., Adewunmi, C.O., Hostettmann, K., 1992. A triterpene glycoside from the fruits of Tetrapleura tetaptea. Phytochemistry 31, 1321–1323. Morita, M., Shibuya, M., Kushiro, T., Masuda, K., Ebizuka, Y., 2000. Molecular cloning and functional expression of triterpene synthases from pea (Pisum sativum). New α-amyrin-producing enzyme is a multifunctional triterpene synthase. Eur. J. Biochem. 267, 3453–3460. Murashige, T., Skoog, F., 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473–497. Nakatani, M., Miyazaki, Y., Iwashita, T., Naoki, H., Hase, T., 1989. Triterpenes from Ilex rotunda fruits. Phytochemistry 28, 1479–1482. Numata, A., Yang, P., Takahashi, C., Fujiki, R., Nabae, M., Fujita, E., 1989. Cytotoxic triterpenes from a Chinese medicine, Goreishi. Chem. Pharm. Bull. 37, 648–651. Phillips, D.R., Rasbery, J.M., Bartel, B., Matsuda, S.P., 2006. Biosynthetic diversity in plant triterpene cyclization. Curr. Opin. Plant Biol. 9, 305–314. Richard, R.M.E., Durham, D.G., Liu, X., 1994. Antibacterial activity of compounds from Rubus pinfaensis. Planta Med. 60, 471–473. Safayhi, H., Sailer, E.R., 1997. Anti-inflammatory actions of pentacyclic triterpenes. Planta Med. 63, 487–493. Saimaru, H., Orihara, Y., Tansakul, P., Kang, Y., Shinuya, M., Ebizuka, Y., 2007. Production of triterpene acids by cell suspension cultures of Olea europaea. Chem. Pharm. Bull. 55, 784–788.
CHAPTER | 38 Production of Triterpene Acids by Cell-suspension Cultures of Olea europaea
Sakurawi, K., Yasuda, F., Tozyo, T., Nakamura, M., Sato, T., Kikuchi, J., Terui, Y., Ikenishi, Y., Iwata, T., Takahashi, K., Konoike, T., Mihara, S., Fujimoto, M., 1996. Endothelin receptor antagonist triterpenoid, myriceric acid A, isolated from Myrica cerifera, and structure activity relationships of its derivatives. Chem. Pharm. Bull. 44, 343–351. Seto, T., Tanaka, T., Tanaka, O., Naruhashi, N., 1984. β-Glucosyl esters of 19α-hydroxyursolic acid derivatives in leaves of Rubus species. Phytochemistry 23, 2829–2834. Shibuya, M., Zhang, H., Endo, A., Shishikura, K., Kushiro, T., Ebizuka, Y., 1999. Two branches of the lupeol synthase gene in the molecular evolution of plant oxidosqualene cyclases. Eur. J. Biochem. 266, 302–307. Shibuya, M., Xiang, T., Katsube, Y., Otsuka, M., Zhang, H., Ebizuka, Y., 2007. Origin of structural diversity in natural triterpenes. Direct synthesis of seco-triterpene skeletons by oxidosqualene cyclase. J. Am. Chem. Soc. 129, 1450–1455.
347
Tansakul, P., Shibuya, M., Kushiro, T., Ebizuka, Y., 2006. Dammarenediol-II synthase, the first dedicated enzyme for ginsenoside biosynthesis, in Panax ginseng. FEBS Lett. 580, 5143–5149. Tchivounda, H.P., Koudogbo, B., Besace, Y., Casadevall, E., 1991. Triterpene saponins from Cylicodiscus gabunensis. Phytochemistry 30, 2711–2716. Vasconcelos, F.C., Gattass, C.R., Rumjanek, V.M., Maia, R.C., 2007. Pomolic acid-induced apoptosis in cells from patients with chronic myeloid leukemia exhibiting different drug resistance profile. Invest. New Drugs. 25, 525–533. Xu, R., Fazio, G.C., Matsuda, S.P., 2004. On the origins of triterpenoid skeletal diversity. Phytochemistry 65, 261–291. Zhang, Y., Bao, F., Hu, J., Liang, S., Zhang, Y., Du, G., Zhang, C., Cheng, Y., 2007. Antibacterial lignans and triterpenoids from Rostellularia procumbens. Planta Med. 73, 1596–1599.
This page intentionally left blank
Chapter 39
Bioactive Ingredients in Olive Leaves Maria Z. Tsimidou and Vassiliki T. Papoti Laboratory of Food Chemistry and Technology, School of Chemistry, Aristotle University of Thessaloniki, Greece
39.1 INTRODUCTION The evergreen olive tree (Olea europaea L.) treasured over the centuries for its fruits and the oil derived from them, is becoming a source of natural antioxidants and other bioactive ingredients from another part of its organism, the leaves. Olive fruit endogenous bioactive ingredients, mainly hydroxytyrosol (3,4-dihydroxyphenyl ethanol), its ester with elenolic acid glycoside (oleuropein or oleuropeoside) and related derivatives (Boskou et al., 2005) were possibly the reason why leaves attracted the interest of scientists in the food, medical and cosmetic sectors. This chapter is devoted to recent findings concerning olive leaves as a source of natural antioxidants and the factors that influence their levels (Table 39.1). Phenolic compounds identified and reported repeatedly in literature are presented in Tables 39.2–39.4. These compounds are grouped with regard to major molecular characteristics as simple phenols and acids, secoiridoids and flavonoids. The compounds are either secondary metabolites or compounds formed during leaf storage, extraction method or in the course of analytical examination (artefacts). Within each group compounds
are alphabetically ordered. Other bioactive ingredients are presented in Table 39.5. Subsections on sampling requirements, methods of extraction and determination precede those on levels and in vivo and in vitro antioxidant activity of the major ingredients. The chapter ends with reference to traditional uses of olive leaves, current status and future prospects of bioactive ingredients exploitation.
TABLE 39.2 Simple phenols and acids and other related compounds present in olive leaves. Simple phenols and acids1 Hydroxytyrosol and glucosides Tyrosol and glucosides Benzoic acids: Gallic, vanillic, syringic, salicylic, hydroxybenzoic, protocatechuic acids, vanillin Cinnamic acids:
TABLE 39.1 Key influencing features of olive leaf bioactive ingredients.
Cinnamic, caffeic, coumaric, ferulic, chlorogenic acids Homovanillic acid
1. Cultivation zone, agronomical practices, environmental conditions, cultivar, tree and leaf age, leaf development stage, other biotic and abiotic parameters
Other related compounds1 Elenolic acid and derivatives
2. Plant material post-harvest treatment (drying process and storage conditions)
Verbascoside
3. Extraction means and techniques
This table summarizes the most important simple phenolic and other related compounds frequently reported to be present in olive leaf extracts. 1
Indicative references: Gariboldi et al., 1986; Heimler et al., 1992; Ryan et al., 2002a.
This table lists the key factors that may influence the final bioactive content, composition and potential of olive leaf preparations. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
349
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
350
SECTION | I
TABLE 39.3 Secoiridoids present in olive leaves. Secoiridoids1
Non Fruit Aspects Including Mill Wastewater
TABLE 39.5 Other bioactive compounds reported in olive leaves. Bioactive compounds1
Demethyloleuropein Amyrin 3,4-DHPEA-EDA (3,4-dihydroxyphenylethyl 4-formyl-3formylmethyl-4-hexenoate)
β-carotene
Oleoside
Erythrodiol
Oleuropein
Maslinic acid
Oleuropein aglycone
Oleanolic acid
Oleuroside
β-sitosterol
Ligstroside
Squalene
Ligstroside aglycone
Stigmasterol
This table summarizes the most important secoiridoids frequently reported to be present in olive leaf extracts.
Tocopherol
1
Indicative references: Gariboldi et al., 1986; Kuwajima, H., Uemura, T., Takaishi, K., Inoue K., and Inoue, H., 1988. A secoiridoid glucoside from Olea europaea. Phytochemistry 27, 1757–1759; Paiva-Martins and Gordon, 2001; Ryan et al., 2002a & b; 2003; Cayuela et al., 2006; Silva et al., 2007.
Ursolic acid Uvaol This table summarizes frequently reported bioactive ingredients of olive leaf extracts, other than phenolics. 1
Indicative references: Bianchi, G., Vlahov, G., Anglani C., and Murelli, C., 1993. Epicuticular wax of olive leaves. Phytochemistry 32, 49–52; Albi et al., 2001; de Lucas et al., 2002; Tabera et al., 2004; Cayuela et al., 2006; Sánchez Ávila et al., 2007.
TABLE 39.4 Flavonoids present in olive leaves. Flavonoids1 Apigenin Apigenin 7-O-glucoside Apigenin 4-O-rutinoside Apigenin 7-O-rutinoside Hesperidin Luteolin Luteolin 4’-O-glucoside Luteolin 7-O-glycoside, Luteolin 7-O-rutinoside Quercetin Quercitrin Rutin This table summarizes the most important flavonoids frequently reported to be present in olive leaf extracts. 1
Indicative references: Heimler, D., Pieroni, A., Tattini, M., and Cimato, A., 1992. Determination of flavonoids, flavonoid glycosides and biflavonoids in Olea europaea L. leaves. Chromatographia 33, 369–373; Ryan et al., 2002a; Liakopoulos et al., 2006.
39.2 SAMPLING In short, olive leaf physiology has the following characteristics (Lavee, 1996; Therios, 2005). Leaves of different age can be found concurrently all over the tree canopy
throughout the year. For an adult tree, leaf life span is up to three years long; however, the majority of leaves fall during the second year, especially when situated in shadow. Olive leaves can be distinguished into current and old season ones. The former are usually further subdivided to new and mature leaves. New leaves are defined as those growing towards the extreme tip of current year shoots, with mature ones as those found between the middle and inner end of the same shoots. Apart from leaf age, other factors such as cultivar, period of leaf development, tree age and vitality as well as surrounding milieu affect leaf morphological characteristics. Leaf shape and size varies. Its upper side is light to dark green, while the opposite one is ash gray to green. Typical cultivar attributes (length, width and shape) referred to mature leaves. Mean leaf axial dimensions (length and width) vary between 5–11 cm and 1–3 cm, respectively. Sampling period, cultivar, leaf age and alternate-bearing affect leaf content and composition in inorganic elements, fats, sugars, proteins, phenolics, uronic acids and Krebs cycle acids (Donaire et al., 1975; Fernández-Escobar et al., 1999; Ryan et al., 2003; Ranalli et al., 2006). Alternatebearing and growth stage periods vary among cultivars. Abiotic parameters may notably affect qualitatively and quantitatively composition. Considering the above-mentioned
351
CHAPTER | 39 Bioactive Ingredients in Olive Leaves
parameters sampling becomes of major importance in the determination of the presence and levels of bioactive ingredients such as antioxidants. Leaf sampling strategy has to comply with objectives of the investigation. In all cases some basic rules should be fulfilled. It is useful in studies related with bioactive ingredient content and composition to describe sampling period, collection procedure (usually from the whole perimeter of trees or shoots with certain orientation, from within arm reach height), leaf type (mature, new, etc.), cultivar and cultivation zone. Data for agronomical practices and climatic conditions, when available, can be given additionally.
39.3 POST-HARVEST TREATMENT Upon collection, post-harvest treatment of this plant tissue has to follow general principles and precautions suggested for natural products (Launert, 1989). Immediate handling (e.g. freeze dried in liquid nitrogen) or, alternatively, transportation to the laboratory in a portable refrigerator are convenient means to avoid alterations in phenolic composition till analysis (e.g. Paiva-Martins and Gordon, 2001). Material should be washed with de-ionized water to remove dust while still green. After being thoroughly drained off, if not to be freshly used, leaves should be dried as soon as possible. The latter is needed to eliminate biotransformations. Drying conditions may change initial leaf phenolic content and composition (e.g. Silva et al., 2006); for this reason, a complete description of drying steps has to be given. Drying in the air, in conventional or ventilated ovens, under microwave irradiation or most commonly by freeze drying is accomplished over different lengths of time and heat conditions. Obviously avoidance of high temperatures and long time air exposure restricts chemical changes. Dried material is preferably kept in sealed containers (glass) and stored in a dry and dark place. Plastic bags are not recommended.
39.4 EXTRACTION PROCEDURES Following recent advances in extraction of nutraceuticals from plants (Wang and Weller, 2006) in most of the numerous olive leaf studies a range of means and techniques have been applied. However, few are the publications that are devoted to development and validation of extraction conditions. The majority adopts procedures that are repeatable but not necessarily adequate for quantitative recovery of bioactive ingredients. Publications at preparative scale (⬎1 kg) are rather limited (Gariboldi et al., 1986, 2.0 kg; Somova et al., 2003, 1.5 kg; Tabera et al., 2004, 10.0 kg). On the contrary, in most of the works quantities ⬍5 g are treated. Efforts to isolate bioactive compounds are increasing (Guinda et al., 2004; Rada et al., 2007).
Careful literature search indicated that special attention has been paid to phenolic polar compounds. Water, methanol, ethanol, as well as aqueous alcohol mixtures, are the usual solvents for phenol extraction from dry or fresh leaf material. Moreover, acetone, ethyl acetate or a sequence of all of the above solvents have also been reported for the same compounds. Less polar bioactive compounds (tocopherols, carotenoids, β-sitosterol, squalene) have been identified in n-hexane extracts, whereas terpenoids (e.g., oleanolic acid, ursolic acid, maslinic acid, uvaol and erythrodiol) in ethanol, ethyl acetate or methylene chloride extracts. Supercritical fluid extraction has been also used for both polar and non-polar compounds. Extraction techniques are critical parameters in the process depending upon objectives. Except for preparation of infusions and decoctions, imitating traditional practices, maceration, agitation, shaking, blending, sonication, hydrodistillation and Soxhlet extraction have been reported in combination to one or more solvents. Weaknesses such as low efficiency, degradation of thermolabile bioactive ingredients, long extraction periods and high solvent consumption characterize many of them. Therefore, in recent works aiming at future industrial applications experimentation involved alternatives such as ultrasound-assisted, superheated liquid and microwave-assisted extraction. This trend is demonstrated for phenolics and terpenoids in the works of JapónLuján and Luque de Castro (2006a) and Sánchez Ávila et al. (2007). Oleanolic acid (87% purity) was obtained at a satisfactory yield (⬃2%) from olive leaves according to the flow diagram described in Figure 39.1 (Guinda et al., 2004). Tocopherols and other high-added-value non-polar compounds attracted less interest from some researchers (de Lucas et al., 2002; Tabera et al., 2004).
39.5 METHODS OF DETERMINATION Estimation of total polar phenol content of olive leaf extracts is almost exclusively carried out via the colorimetric Folin-Ciocalteu assay. Identification and quantitative determination of individual polar phenolic compounds is usually accomplished by HPLC, mainly in the reversedphase mode. Gas chromatography-MS techniques are preferred for the determination of minor leaf phenolics when identification is needed. A typical RP-HPLC system involves a C18, 5 μm stationary phase [usually 125–250 mm length, 4.0–4.6 mm id]. Gradient elution is more frequently applied, considering the complexity of phenolic profile. Nevertheless, isocratic conditions (e.g. Savournin et al., 2001) can be used when selective monitoring of compounds is needed. Eluents comprised of acidified water, acetonitrile and/or methanol. Initial organic modifier content ranges from 5–10%. Reported elution order among complex species is not always the same, but this may be attributed to data misinterpretation, too.
352
SECTION | I
Non Fruit Aspects Including Mill Wastewater
Extraction 1/20 g dry leaf mL−1 ethanol 20°C, 1 h shaking
coated with 35% dimethyl-65% diphenyl polysiloxane or 5% phenyl 95% dimethylpolysiloxane and similar type bonded and cross-linked phases, medium-size columns (30 m) at relatively high temperatures (250–300 °C) have been adopted by several research groups (e.g. Cayuela et al., 2006; Sánchez Ávila et al., 2007).
Bleaching, activated charcoal
39.6 LEVELS OF MAJOR INGREDIENTS
Air-dried leaves Ambient temperature, 2 months
Charcoal removal by filtration Solution concentration to allow precipitation
Filtration
Precipitate washing with chilled (5–8°C) ethanol, in case of residual pigments (indicated by yellow color precipitate)
Repetition till oleanolic acid yield up to 90%
Drying at 160°C in a furnace
Recovery (~80%) of oleanolic acid (87% purity) FIGURE 39.1 Steps of oleanolic acid isolation (Guinda et al., 2004). The figure depicts the experimental conditions and procedure applied by Guinda et al. (2004) for the recovery of oleanolic acid, a bioactive ingredient, from olive leaves.
Detection is based on ultraviolet spectra characteristics so that photodiode array became an indispensable tool in leaf phenolic studies. The general use wavelength (280 nm) is preferred in most works, but one has to consider that no universal absorbance maximum exists for olive leaf phenolics. Introduction of fluorimetric detection and mass spectrometry (MS) widens available means for complete characterization of structurally related compounds. Gas chromatography is currently the preferred methodology for triterpenoid bioactive ingredients such as erythrodiol, uvaol, oleanolic, ursolic and maslinic acids. Liquid chromatography may advance in the near future due to the increasing application of LC-MS systems. Nevertheless, gas chromatography remains a tedious and demanding procedure at the analyte preparation step (derivatization). In their recent work Sánchez Ávila et al. (2007) present useful data for the reduction in the length of the silylation step, using ultrasound assistance. Reaction time can, thus, be reduced to a few minute processes. Efficiency of the proposed procedure was equivalent to that using a conventional silylation method (80 °C, 2 h). Capillary columns
As presented in Tables 39.2–39.4 leaf phenolics are numerous and of diverse nature. However, only a few of them are found in considerable levels that are worthy of commercial exploitation. Secoiridoids and flavonoids are the dominant groups sharing the total phenol content (Makris et al., 2007). Simple phenols and acids are present in lower amounts. Consequently major olive leaf phenolics seem to be 3,4-DHPEA-EDA (3,4-dihydroxyphenylethyl 4-formyl3-formylmethyl- 4-hexenoate), hydroxytyrosol, luteolin 7 glucoside and 4⬘ glucoside, oleuropein, and verbascoside. From the compounds shown in Table 39.5, oleanolic acid (3β-hydroxyolean-12-en-28-oic acid) is the dominant one. In terms of total phenol content, olive leaf seems to be a promising source for exploitation. Total polar phenol or individual phenol content has been expressed in terms of different external standard curves, namely, as caffeic, tannic and gallic acids, oleuropein or tyrosol. This fact is expected to influence magnitude of reported levels (Blekas et al., 2002). Our records indicate that the average total phenol content of extracts prepared under moderate extraction conditions (freeze-dried leaf material to methanol ratio: 1/40, 5 min sonication at ambient temperature, n ⫽ 35) via the Folin-Ciocalteu assay varied from approximately 6 to 20 mg per g dry leaf weight, expressed as caffeic acid. Further extraction (48 h agitation at ambient temperature) resulted in a two-fold increase of yield. Our results are in accordance with published data expressed in a similar way (Le Floch et al., 1998; Cayuela et al., 2006). Olive leaf can be considered as a source of phenolic antioxidants of similar potency to olive fruit (Malik and Bradford, 2006) and exceptional among agri-food byproducts (Makris et al., 2006) regarding availability throughout the year and composition.
39.7 SECOIRIDOIDS, SIMPLE PHENOLS AND RELATED COMPOUNDS Oleuropein has been frequently reported to be the major olive leaf compound. As mentioned by Ryan and co-workers (2003), for extracts prepared from leaves of different age, sampling period and alternate-bearing period under moderate extraction conditions, oleuropein content could reach 264 mg per g dry leaf, expressed as tyrosol equivalents. This finding does not preclude other compounds (hydroxytyrosol, oleuropein aglycone and 3,4-DHPEA-EDA) to predominate
353
CHAPTER | 39 Bioactive Ingredients in Olive Leaves
(Ryan et al., 2003; Rada et al., 2007) due to interconversions among the above or other structurally related compounds (e.g. oleuroside, oleuropeindials). The latter was suggested by different groups (e.g. Paiva-Martins and Gordon, 2001; Ryan et al., 2003). In certain cases the content of oleuropein has been found to be lower than that of major flavonoids (e.g. Japón-Luján et al., 2006b). Oleuropein content seems to decline with leaf aging (Ranalli et al., 2006). Oleuropein to oleuroside ratio ranges from two to ten (Savournin et al., 2001; Ryan et al., 2003). Despite the arguments on the precursors of 3,4-DHPEAEDA (Paiva-Martins and Gordon, 2001; Ryan et al., 2002a, 2003), its levels seem to be influenced by leaf age and sampling period and more importantly by post-harvest treatment and storage of the plant material. For the same cultivar fluctuations of the ratio 3,4-DHPEA-EDA to oleuropein were from 0.1 to 8.5, indicating that at certain times the dialdehyde form prevails. Similar observations apply to hydroxytyrosol levels as reported by the Australian researchers (Ryan et al., 2003). Verbascoside presence can not be underestimated. For example, verbascoside to oleuropein content ratio examined for 13 Spanish cultivars varied from 0.05 to 0.63 as illustrated in Table 39.6 (Japón-Luján and Luque de Castro, 2007).
39.8 FLAVONOIDS Flavonoid quantification is achieved between 330–365 nm and not at 280 nm or lower wavelengths. In this view direct comparison between the levels of individual flavonoids with oleuropein content is not accurate. Liakopoulos and co-workers (2006) studying distribution of flavonoids in various parts of the leaf found that luteolin 7-O-glucoside and luteolin 4⬘-O-glucoside are the most abundant forms. Their finding was confirmed for six different cultivars. The authors sampled fully expanded leaves, the same day and time, from an experimental olive plantation. The reported levels varied between ⬃0.4–1.6 and 0.1–0.8 mg g⫺1 fresh weight, respectively. These results are pointed out because quantification at 365 nm was achieved using the respective pure compounds.
39.9 TRITERPENOIDS Oleanolic acid content prevails with regards to erythrodiol, uvaol, ursolic and maslinic acids. These levels seem to be influenced by cultivar and sampling period (Albi et al., 2001; Somova et al., 2003, 2004; Cayuela et al., 2006; Sánchez Ávila et al., 2007). However, to our knowledge, no systematic study concerning the effects of leaf age, sampling period, etc., has been reported yet. Their levels seem to coincide with the lowest ones reported for oleuropein (⬍ 35 mg g⫺1 dry leaf).
TABLE 39.6 Verbascoside to oleuropein content ratio from different Spanish cultivars based on information presented by Japón-Luján and Luque de Castro (2007). Cultivar
Verbascoside/oleuropein content ratio
Alamenõ
0.07
Arbequina
0.42
Azulillo
0.46
Chorna
0.34
Hojiblanca
0.08
Lechín
0.18
Manzanillo
0.24
Negrillo
0.11
Nevadillo
0.17
Ocal
0.05
Pierra
0.14
Sevillano
0.63
Tempranillo
0.06
Data in this table indicate that the content of verbascoside was found markedly to vary among the different Spanish cultivars studied by Japón-Luján and Luque de Castro (2007). Moreover, when verbascoside levels are compared to oleuropein ones, a major olive leaf component, its content can not be underestimated in certain cases.
39.10 ANTIOXIDANT ACTIVITY Bioactivity has been attributed to all of the above-mentioned compounds and also to derivatives obtained from them such as tyrosol, elenolic acid and flavonoid aglycones. Moreover, the presence of other compounds shown in Tables 39.2–39.5 can not be underestimated when bioactivity of olive leaf extracts is examined. Synergistic activity between or among classes of bioactive ingredients has been repeatedly suggested in literature. Among the various health effects antioxidant activity seems to be an important inherent property of olive leaves (Soler-Rivas et al., 2000; Visioli et al., 2002; Boskou et al., 2005; Khan et al., 2007; Nenadis and Tsimidou, 2009). Therefore, special attention is paid to antioxidant properties of both individual compounds and various types of olive leaf extracts. The antioxidant activity of oleuropein and related secoiridoids has been recently reviewed by our group (Nenadis and Tsimidou, 2009). Oleuropein, except for playing an important protective role against certain herbivores, pests and microbes that may damage olive tree and its fruits, has
354
been repeatedly tested by common in vitro assays for reactive oxygen and nitrogen species scavenging properties. Moreover, protection against cardiovascular disorders, diabetes, certain cancers and even anti-HIV activity has been linked to this secoiridoid. Its counterpart moieties, hydroxytyrosol and elenolic acids are known for radical scavenging and antimicrobial properties, respectively. Data from the parallel examination of olive leaf extract (⬃25% oleuropein) and pure compounds, namely, oleuropein (~90%), verbascoside, tyrosol, hydroxytyrosol, luteolin 7-O-glucoside, and luteolin towards ABTS˙⫹ indicated the superiority of the extract compared with that of oleuropein (⬃ half TEAC value) and the similarity with the performance of hydroxytyrosol. The study gave evidence for the contribution of verbascoside, luteolin glucoside and aglycon to the overall antioxidant activity of the extract tested (Benavente-García et al., 2000), despite the fact that they constituted less than 3% of the total phenol content. A hypothesis for a possible synergism demands further substantiation. Oleuropein performance towards the DPPH radical differs from that of α-tocopherol in terms of kinetics. The latter reached the steady state within 15 min after initiation of the reaction and presented a stoichiometry of 2 over an experimental period of 250 min. Oleuropein showed a reduction in EC50 value of 45% between 15 and 250 min (Paiva-Martins and Gordon, 2001). Oleanolic acid and its regio-isomer ursolic acid are well-established multifunctional ingredients of many natural products (Liu, 1995). Recently, Yin and Chen (2007) tested the activity of both acids towards 2, 2⬘-azobis-(2amidinopropane) dihydrochloride and 2, 2⬘-azobis (2, 4dimethylvaleronitrile) in a phosphatidylcholine liposome system and compared it with that of α-tocopherol under different heating and pH conditions. The activities varied upon experimental conditions and were proved to be even higher than that of vitamin E in certain cases. The two triterpenes exhibited a concentration-dependent effect using superoxide anion scavenging, chelating effect, xanthine oxidase inhibition activity and reducing power assays. No structure–activity relationships have been established for differences in the activity of the two isomers. Considering that the characteristic health effects of the above two acids, as well as of erythrodiol, uvaol and maslinic acids are antihypertensive, antiatherosclerotic, antioxidant, cardiotonic or antidysrhythmic (Somova et al., 2003, 2004), it is expected that the interest in this fraction of olive leaf will increase.
SECTION | I
olive leaf extracts are mentioned in traditional medicine as effective against malaria, hypertension and diabetes. The broad traditional use of olive leaf as a folk medicine is frequently presented in ethnobotanical and other studies (Guarrera, 2005; Gião et al., 2007; Khan et al., 2007; Tahraoui et al., 2007). Olive leaf is also registered in certain pharmacopeias (e.g. the European Pharmacopoeia, 6th edn.). Exploitation of antioxidant properties of leaf extracts – concentrated or not – within the context of functional foods was the basis of some recent applications such as oil supplementation (Guinda et al., 2004; Farag et al., 2007; Bouaziz et al., 2008) and encapsulation with the aid of proteins (Bayçin et al., 2007) or hydrocolloids (Mourtzinos et al., 2007). Fractionation of extracts and purification of active ingredients, such as oleuropein, hydroxytyrosol, oleanolic acid and tocopherols, in order to be further used in food and food supplements were also the objectives of other publications from 2000 onwards (e.g. Briante et al., 2002; de Lucas et al., 2002; Tabera et al., 2004; Rada et al., 2007). Commercial products in the form of herbal teas or food supplements are found all over the world, as whole leaves, powders, liquid extracts or tablets. Depending on legislation framing these products have to be consistent with safety and other specifications and requirements (e.g. Hanekamp and Bast, 2007). Toxicological studies on oleuropein, hydroxytyrosol, triterpenoids and leaf extracts (Petkov and Manolov, 1972; Liu, 1995; Farag et al., 2003; Soni et al., 2006) are rather promising for the safe use of decoctions, infusions and commercial preparations containing the above materials. Nevertheless, toxicological research seems to be an ongoing process.
39.12 CONCLUSION Olive leaf is a promising source for exploitation in terms of its bioactive content. Olive-oil-producing countries have to consider seriously commercialization of olive leaf in various forms. Olive leaves or in combination with other olive tree byproducts (wood material after pruning, olive seeds) are without doubt a source for multifunctional bioactive ingredients.
SUMMARY POINTS ●
39.11 APPLICATIONS AND PROSPECTS Careful search in ethnopharmacological sources led us to the observation that olive leaf use is rather restricted even in the major olive-producing countries (Spain, Italy, Greece, Tunisia, Morocco). Reports, many of them cited to specific electronic web pages, are not easily accessible. However,
Non Fruit Aspects Including Mill Wastewater
●
●
●
Olive leaf is a natural source rich in bioactive simple phenols, secoiridoids, flavonoids and terpenoids. Olive leaf is a folk medicine against malaria, hypertension and diabetes. Recent studies aim at substantiating and extending traditional applications. Olive leaf bioactive ingredient content is influenced by a number of biotic and abiotic parameters. Sampling, post-harvest treatment and extraction means affect to
CHAPTER | 39 Bioactive Ingredients in Olive Leaves
●
●
●
a certain extent content and composition of bioactive ingredients. Among the various olive leaf attributes, antioxidant activity has potential in pharmaceutical, food and cosmetic sectors. Olive leaf is a promising source for annual industrial exploitation with regard to its antioxidant potential, arising from synergism of its bioactive constituents. Olive-oil-producing countries have to consider seriously commercialization of olive leaf in various forms.
ACKNOWLEDGMENTS V.T.P. thanks Greek General Secretariat of Research and Technology (PENED 03EΔ596) for financial support.
REFERENCES Albi, T., Guinda, Á., Lanzón, A., 2001. Obtaining procedure and determination of terpenic acids of olive leaf (Olea europea). Grasas y Aceites 52, 275–278. Bayçin, D., Altıok, E., Ülkü, S., Bayraktar, O., 2007. Adsorption of olive leaf (Olea europaea L.) antioxidants on silk fibroin. J. Agric. Food Chem. 55, 1227–1236. Benavente-García, O., Castillo, J., Lorente, J., Ortuño, A., Del Rio, J.A., 2000. Antioxidant activity of phenolics extracted from Olea europaea L. leaves. Food Chem. 68, 457–462. Blekas, G., Psomiadou, E., Tsimidou, M., Boskou, D., 2002. On the importance of total polar phenols to monitor the stability of Greek virgin olive oil. Eur. J. Lipid Sci. Technol. 104, 340–346. Boskou, D., Blekas, G., Tsimidou, M., 2005. Phenolic compounds in olive oil and olives. Curr. Top. Nutr. Res. 3, 125–136. Bouaziz, M., Fki, I., Jemai, H., Ayadi, M., Sayadi, S., 2008. Effect of storage on refined and husk olive oils composition: Stabilization by addition of natural antioxidants from Chemlali olive leaves. Food Chem. 108, 253–262. Briante, R., La Cara, F., Febbraio, F., Patumi, M., Nucci, R., 2002. Bioactive derivatives from oleuropein by a biotransformation on Olea europaea leaf extracts. J. Biotechnol. 93, 109–119. Cayuela, J.A., Rada, M., Rios, J.J., Albi, T., Guinda, Á., 2006. Changes in phenolic composition induced by Pseudomonas savastanoi pv. savastanoi infection in olive tree: presence of large amounts of verbascoside in nodules of tuberculosis disease. J. Agric. Food Chem. 54, 5363–5368. de Lucas, A., Martinez de la Ossa, E., Rincón, J., Blanco, M.A., Gracia, I., 2002. Supercritical fluid extraction of tocopherol concentrates from olive tree leaves. J. Supercrit. Fluids 22, 221–228. Donaire, J.P., Sánchez, A.J., López-Gorgé, J., Recalde, L., 1975. Metabolic changes in fruit and leaf during ripening in the olive. Phytochemistry 14, 1167–1169. Farag, R.S., El-Baroty, G.S., Basuny, A.M., 2003. Safety evaluation of olive phenolic compounds as natural antioxidants. Int. J. Food Sci. Nutr. 54, 159–174. Farag, R.S., Mahmoud, E.A., Basuny, A.M., 2007. Use crude olive leaf juice as a natural antioxidant for the stability of sunflower oil during heating. Intl. J. Food Sci. Technol. 42, 107–115. Fernández-Escobar, R., Moreno, R., García-Creus, M., 1999. Seasonal changes of mineral nutrients in olive leaves during the alternate-bearing cycle. Sci. Hort. 82, 25–45.
355
Gariboldi, P., Jommi, G., Verotta, L., 1986. Secoiridoids from Olea europaea. Phytochemistry 25, 865–869. Gião, M.S., González-Sanjosé, M.L., Rivero-Pérez, M.D., Pereira, C.I., Pintado, M.E., Malcata, F.X., 2007. Infusions of Portuguese medicinal plants: dependence of final antioxidant capacity and phenol content on extraction features. J. Agric. Food Chem. 87, 2638–2647. Guarrera, P.M., 2005. Traditional phytotherapy in central Italy (Marche, Abruzzo, and Latium). Fitoterapia 76, 1–25. Guinda, Á., Albi, T., Pérez-Camino, Ma. C., and Lanzón, A. (2004). Supplementation of oils with oleanolic acid from the olive leaf (Olea europaea). Eur. J. Lipid Sci. Technol. 106, 22–26. Japón-Luján, R., Luque de Castro, M.D., 2006a. Superheated liquid extraction of oleuropein and related biophenols from olive leaves. J. Chromatogr. A. 1136, 185–191. Japón-Luján, R., Ruiz-Jiménez, J., Luque de Castro, M.D., 2006b. Discrimination and classification of olive tree varieties and cultivation zones by biophenol contents. J. Agric. Food Chem. 54, 9706–9712. Japón-Luján, R., Luque de Castro, M.D., 2007. Small branches of olive tree: a source of biophenols complementary to olive leaves. J. Agric. Food Chem. 55, 4584–4588. Hanekamp, J.C., Bast, A., 2007. Food supplements and European regulation within a precautionary context: a critique and implications for nutritional, toxicological and regulatory consistency. Crit. Rev. Food Sci. Nutr. 47, 267–285. Khan, Md.Y., Panchal, S., Vyas, N., Butani, A., Kumar, V., 2007. Olea europaea: A phyt -pharmacological review. Phcog. Rev. 1, 114–118. Launert, E., 1989. The Hamlyn guide to edible and medicinal plants of Britain and northern Europe. The Hamlyn Publishing Group Ltd., London. Lavee, S. 1996. Biology and physiology of olive tree. In: The World Olive Encyclopaedia. IOOC (eds). Madrid, Spain, pp. 61–110. Le Floch, F., Tena, M.T., Ríos, A., Valcárcel, M., 1998. Supercritical fluid extraction of phenol compounds from olive leaves. Talanta 46, 1123–1130. Liakopoulos, G., Stavrianakou, S., Karabourniotis, G., 2006. Trichome layers versus dehaired lamina of Olea europaea leaves: differences in flavonoid distribution, UV-absorbing capacity, and wax yield. Environ. Exp. Bot. 55, 294–304. Liu, J., 1995. Pharmacology of oleanolic acid and ursolic acid. J. Ethnopharmacol. 49, 57–68. Makris, D.P., Boskou, G., Andrikopoulos, N.K., 2007. Polyphenolic content and in vitro antioxidant characteristics of wine industry and other agri-food solid waste extracts. J. Food Comp. Anal. 20, 125–132. Malik, N.S.A., Bradford, J.M., 2006. Changes in oleuropein levels during differentiation and development of floral buds in ‘Arbequina’ olives. Sci. Hort. 110, 274–278. Mourtzinos, I., Salta, F., Yannakopoulou, K., Chiou, A., Karathanos, V.T., 2007. Encapsulation of olive leaf extract in β-Cyclodextrin. J. Agric. Food Chem. 55, 8088–8094. Nenadis, N., Tsimidou, M.Z., 2009. Oleuropein and related secoiridoids. Antioxidant activity and sources other than Olea europaea L. (Olive tree). Recent Prog. Med. Plants 25(6), 53–74. Paiva-Martins, F., Gordon, M.H., 2001. Isolation and characterization of the antioxidant component 3,4-dihydroxyphenylethyl 4-formyl3-formylmethyl-4-hexenoate from olive (Olea europaea) leaves. J. Agric. Food Chem. 49, 4214–4219. Petkov, V., Manolov, P., 1972. Pharmacological analysis of the iridoid oleuropein. Arzneim-Forsch. (Drug Res.) 22, 1476–1485. Rada, M., Guinda, Á., Cayuela, J., 2007. Solid/liquid extraction and isolation by molecular distillation of hydroxytyrosol from Olea europaea L. leaves. Eur. J. Lipid Sci. Technol. 109, 1071–1076.
356
Ranalli, A., Contento, S., Lucera, L., Di Febo, M., Marchegiani, D., Di Fonzo, V., 2006. Factors affecting the contents of iridoid oleuropein in olive leaves (Olea europaea L.). J. Agric. Food Chem. 54, 434–440. Ryan, D., Antolovich, M., Herlt, T., Prenzler, P.D., Lavee, S., Robards, K., 2002a. Identification of phenolic compounds in tissues of the novel olive cultivar Hardy’s Mammoth. J. Agric. Food Chem. 50, 6716–6724. Ryan, D., Prenzler, P.D., Lavee, S., Antolovich, M., Robards, K., 2003. Quantitative changes in phenolic content during physiological development of the olive (Olea europaea) cultivar Hardy’s Mammoth. J. Agric. Food Chem. 51, 2532–2538. Sánchez Ávila, N., Capote, F.P., Luque de Castro, M.D., 2007. Ultrasoundassisted extraction and silylation prior to gas chromatography–mass spectrometry for the characterization of the triterpenic fraction in olive leaves. J. Chromatogr. A 1165, 158–165. Savournin, C., Baghdikian, B., Elias, R., Dargouth-Kesraoui, F., Boukef, K., Balansard, G., 2001. Rapid high-performance liquid chromatography analysis for the quantitative determination of oleuropein in Olea europaea leaves. J. Agric. Food Chem. 49, 618–621. Silva, S., Gomes, L., Leitão, F., Coelho, A.V., Vilas Boas, L., 2006. Phenolic compounds and antioxidant activity of Olea europaea L. fruits and leaves. Food Sci. Tech. Int. 12, 385–395. Soler-Rivas, C., Espín, J.C., Wichers, J.H., 2000. Oleuropein and related compounds. J. Sci. Food Agric. 80, 1013–1023. Somova, L.I., Shode, F.O., Ramnanan, P., Nadar, A., 2003. Antihypertensive, antiatherosclerotic and antioxidant activity of triterpenoids isolated from
SECTION | I
Non Fruit Aspects Including Mill Wastewater
Olea europaea, subspecies africana leaves. J. Ethnopharmacol. 84, 299–305. Somova, L.I., Shode, F.O., Mipando, M., 2004. Cardiotonic and antidysrhythmic effects of oleanolic and ursolic acids, methyl maslinate and uvaol. Phytomedicine 11, 121–129. Soni, M.G., Burdock, G.A., Christian, M.S., Bitler, C.M., Crea, R., 2006. Safety assessment of aqueous olive pulp extract as an antioxidant or antimicrobial agent in foods. Food Chem. Toxicol. 44, 903–915. Tabera, J., Guinda, A., Ruiz-Rodríguez, A., Señoráns, F.J., Ibáñez, E., Albi, T., Reglero, G., 2004. Countercurrent supercritical fluid extraction and fractionation of high-added-value compounds from a hexane extract of olive leaves. J. Agric. Food Chem. 52, 4774–4779. Tahraoui, A., El-Hilaly, J., Israili, Z.H., Lyoussi, B., 2007. Ethnopharmacological survey of plants used in the traditional treatment of hypertension and diabetes in south-eastern Morocco (Errachidia province). J. Ethnopharmacol. 110, 105–117. Therios, I., 2005. Olive Production, 1st edn. Gartaganis Publications, Thessaloniki, Greece (in Greek). Visioli, F., Galli, C., Galli, G., Caruso, D., 2002. Biological activities and metabolic fate of olive oil phenols. Eur. J. Lipid Sci. Technol. 104, 677–684. Wang, L., Weller, C.L., 2006. Recent advances in extraction of nutraceuticals from plants. Trends Food Sci. Tech. 17, 300–312. Yin, M.-C., Chen, K.-C., 2007. Nonenzymatic antioxidative and antiglycative effects of oleanolic acid and ursolic acid. J. Agric. Food Chem. 55, 7177–7181.
Chapter 40
Phenolic Compounds in Olive Oil Mill Wastewater José S. Torrecilla Department of Chemical Engineering, Universidad Complutense de Madrid, Spain
40.1 INTRODUCTION The oxidative resistance of olive oil is well known. This is mainly because of the quantity of phenolic compounds. In addition to this resistance, olive oil possesses beneficial properties for the human circulatory system (Visioli and Galli 1998). In particular, olive oil phenols have been beneficially linked to processes that contribute to the pathogenesis of heart disease (in particular hydroxytyrosol), atheroscelosis and cancer (this point requires further research) among others (Salami et al., 1995). Although there are a wide variety of reported methods to quantify phenols in olive oil, up to 1 gram of phenolic substances per kg of olive oil can be found (Visioli and Galli, 1998). As can be seen in Table 40.1, the largest amount of phenolic compounds is present in the olive pulp (Aragón and Palancar, 2001). During the manufacturing process of olive oil, mainly in the malaxation stage, a considerable quantity of water is used (Kapellakis et al., 2008) and due to enzymatic or degradation reactions, the phenolic substance terminates in the water phase. The concentration values of each phenolic compound in water and oil phases depends on its partition coefficient (Kp) (Eq. 40.1) (Rodis et al., 2002; Obied et al., 2005).
Kp =
Coil Cwater
in water than oil, resulting in concentrations ranging from 0.5 to 25 g L⫺1 of phenolic substances in water (McNamara et al., 2008).
40.2 OLIVE OIL MILL WASTEWATER During 2008, more than 2.8 ⫻ 106 tons of olive oil and 6.9 ⫻ 105 tons of table olives will be produced in the world and 76.4% and 41.6% of this quantity respectively will be produced in the European Community (International Olive Council, 2008). In this process, estimated volumes of 3 ⫻ 107 m3 of olive oil waste are generated from early
(40.1)
In Equation 40.1, Coil and Cwater are the concentrations of a given phenolic substance in olive oil and water phases, respectively. As can be seen in Table 40.2, oleuropein, hydroxytyrosol and protocatechuic acid compounds are much more soluble in the water phase than in the oil phase. Thus, the water phase will be mainly composed of these phenolic substances. In general, most of the phenolic compounds that are in olive pulp tend to be more soluble Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
357
TABLE 40.1 Olive composition (Aragón and Palancar, 2001). The chemical composition of each part of the olive is shown. Pulp
Stone
Seed
Water (%)
50–60
9.3
30
Oil (%)
15–30
0.7
27.3
Constituents containing nitrogen (%)
2 –5
3.4
10.2
Sugar (%)
3–7.5
41
26.6
Cellulose (%)
3 –6
38
1.9
Minerals (%)
1– 2
4.1
1.5
Polyphenols (%)
2–2.25
0.1
0.5–1
3.4
2.4
Others (%)
–
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
358
SECTION | I
TABLE 40.2 Partition coefficients of olive oil phenolic compounds between oil and water phases (Rodis et al., 2002). The concentration of the phenolic compound in the oil phase divided by this in the water phase is shown. The lower the number, the more soluble the phenolic compound in the water phase. Phenolic compounds
Partition coefficient
Oleuropein
0.0006
4-(2-Hydroxyethyl)-1,2benzenediol (hydroxytyrosol)
0.0100
3,4-Dihydroxybenzoic acid (protocatechuic acid)
0.0390
4-(2-Hydroxyethyl)phenol (tyrosol)
0.0770
3-(3,4-Dihydroxyphenyl)-2propenoic acid (caffeic acid)
0.0890
Elenolic acid
0.1890
Oleuropein aglycon
1.4900
Non Fruit Aspects Including Mill Wastewater
November to late February. In the Mediterranean area, it is one of the principal environmental problems. To realize the importance of its environmental impact, this is equivalent to the pollution produced by more than 22 million people per year (Isidori et al., 2005). The quantity of olive oil waste produced depends on the manufacturing process used. Table 40.3 shows that for every 1000 kg of olives, in pressure processes 1000 kg, in three-phase processes 1500–1800 kg and in two-phase processes 800–950 kg of olive waste are produced (Kapellakis et al., 2008). From these quantities, 600, 1000–1200 and 0 kg of olive oil mill wastewater (OMW) are produced, respectively. In the two-phase method, the water is mainly contained in the solid waste (olive pomace). The two-phase method produces an olive pomace with a moisture content of around 63 wet basis-% (Torrecilla et al., 2006). The OMW is called aqua reflue in Italy; katsigaros in Greece; alpechín in Spain; zebra in Arab countries (Kapellakis et al., 2008), and is mainly formed by the aforementioned water phase (80–95% by weight). It is characterized by a dark-colored liquid caused by lignin polymerized with phenolic compounds, increased acidity (pH about 5), and high conductivity, among other factors. Quantitatively, OMW is formed by 15% of volatile
TABLE 40.3 Comparative data for olive oil extraction processes (Aragón and Palancar, 2001). The influence of processing systems and raw material used on the production of olive oil and waste are shown. Process
Input
Input amount
Output
Output amount
Pressing
Olives
1000 kg
Oil
200 kg
Washing water
0.1–0.12 m3
Solid waste (25% water ⫹ 6% oil)
400 kg
Energy
40–63 kWh
Wastewater (88% water)
600 kg
Olives
1000 kg
Oil
200 kg
Washing water
0.1–0.12 m3
Solid waste (50% water ⫹ 4% oil)
500–600 kg
Fresh water for decanter
0.5–1 m3
Water to polish the impure oil
10 kg
Energy
90–117 kWh
Olives
1000 kg
Washing water
0.1–0.12 m3
Energy
⬍90–117 kWh
Three-phase
Two-phase
Wastewater (94% water ⫹ 1% oil)
1000–1200 kg
Oil
200 kg
Solid waste (60% water ⫹ 3% oil)
800–950 kg
359
CHAPTER | 40 Phenolic Compounds in Olive Oil Mill Wastewater
solid, 2% of inorganic matter and the remainder formed by organic load. Chemically, OMW is formed by phenols, sugars (fructose, mannose, glucose, saccharose, sucrose and pentose), tannins, polyalcohols, aromatic compounds, fermentable proteins, organic acid, vitamins, traces of pesticides, a small quantity of emulsified olive oil, etc. (Table 40.4; Bazoti et al., 2006; Kapellakis et al., 2008). Apart from the manufacturing process, the OMW chemical composition depends on the variety of olive, stage of maturity, storage time and climate, among other factors. The maximum and minimum ranges of the polluting load, organic and average values of inorganic constituents of OMW are shown in Table 40.4. The range can be explained by the aforementioned characteristics and because the OMW chemical composition changes depending on fermentation reactions during the storage time (Borja et al., 2006). Given that OMW comes from olives, the biodegradability of this waste can be assumed. But given that the OMW degradation rate, in particular the phenolic compounds, is much slower than for other substances (sugars or volatile acids with short chains), OMW is not easily biodegradable. On the contrary, tannins, for instance, are highly toxic but biodegradable. Therefore, the phenolic compounds are the main cause of OMW environmental impact. This is based on their high toxicity (see below) and their biodegradability characteristics (Table 40.4; Kapellakis et al., 2008).
TABLE 40.4 Olive oil mill wastewater characteristics and its chemical composition (Kapellakis et al., 2008). The chemical composition and parameters that characterize the environmental impact of olive oil wastewater depending on the processing system are shown. Processing system Pressure
Centrifugation
Chemical oxygen demand (COD) (g L⫺1)
120–130
60–180
Biological oxygen demand (BOD5) (g L⫺1)
90–100
20–55
Suspended solids (%)
0.1
0.9
Total solids (%)
12
6
Sugars
2.0–8.0
0.5–2.6
Nitrogen compounds
0.5–2.0
0.1–0.3
Organic acids
0.1–1.5
0.2–0.4
Polyalcohols
1.0–1.5
0.3–0.5
40.3 PHENOLIC COMPOUNDS
Pentoses, tannins
1.0–1.5
0.2–0.5
The phenolic compounds are derived from phenol. They are composed of one or more aromatic benzene rings, one or more hydroxyl groups and one or more functional side chains. At present, more than 50 different phenolic compounds have been identified in OMW (Obied et al., 2005). These substances have been classified using different methods depending on their molecular weights (Davies et al., 2004) or following their molecular structures (Miranda et al., 2001; Mantzavinos and Kalogerakis 2005; Kapellakis et al., 2008) among other classification methods. As can be seen in Figure 40.1, depending on their chemical structure, some representative phenolic compounds have been classified into three groups: compounds related to tyrosol, derivatives of benzoic acids and cinnamic acids (Miranda et al., 2001). Oleuropein [(4S,5E,6S)-4-[2-[2-(3,4-dihydroxyphenyl) ethoxy]-2-oxoethyl]-5-ethylidene-6-[[(2S,3R,4S,5S,6R)3,4,5-trihydroxy-6-(hydroxymethyl)-2-tetrahydropyranyl] oxy]-4H-pyran-3-carboxylic acid, methyl ester] is the major biophenol found in many olive varieties. During the olive oil extraction, hydroxytyrosol is produced as a result of hydrolysis of this compound by esterase action. The olives, leaves and olive pulp contain low levels of hydroxytyrosol. This compound has also been found in olive pomade (solid waste of the two-phase method) and OMW (FernándezBolaños et al., 2002), most of which can be recovered to
Polyphenols
2.0–2.4
0.3–0.8
Lipids
0.03–1.0
0.5–2.3
Phosphorus
0.11
0.03
Potassium
0.72
0.27
Calcium
0.07
0.02
Magnesium
0.04
0.01
Sodium
0.09
0.03
Chlorine
0.03
0.01
Polluting load
Organic constituents (%)
Inorganic constituents (%)
produce hydroxytyrosol extracts. This phenolic compound is a phytochemical with antioxidant properties. After gallic acid, hydroxytyrosol is believed to be one of the most powerful antioxidants and it is partially responsible for the bitter taste of extra virgin olive oil. Studies have shown that a low dose of hydroxytyrosol reduces the consequences of sidestream smoke-induced oxidative stress in rats (Visioli et al., 2000). Several procedures have been proposed to recover hydroxytyrosol from OMW. Among other methods,
360
SECTION | I
Tyrosol related R1
R3
Non Fruit Aspects Including Mill Wastewater
R1 Et-OH Et-OH
R3 H OH
R4 OH OH
Tyrosol Hydroxytyrosol
Et-COOH
H
OH
p-Hydroxyphenylacetic
R3 H H OH OMe OMe OH OMe R3 H H OH
R4 H OH OH OH OMe OH OH R4 H OH OH
R5 H H H H H OH OMe
Benzoic acid p-Hydroxybenzoic acid Protocatechuic acid Vanillic acid Veratric acid Gallic acid Syringic acid
OMe
OH
R4 Benzoic acid COOH
R5
R3 R4
Cinnamic acid COOH
R3
Cynnamic acid p-Coumaric acid Caffeic acid
Ferulic acid
R3 FIGURE 40.1 Structures of important families of phenolic compounds found in OMW. A classification of phenolic compounds depending on their chemical structure is shown.
this compound was recovered from OMW with a relatively high yield (⬎85%) using a three-staged continuous counter liquid–liquid extraction unit; 1.225 g of hidroxytyrosol was extracted from one liter of OMW and a gram was then purified using chromatographic methods (Allouche et al., 2004). Capasso et al. have extracted hydroxytyrosol by means of ethyl acetate and chromatographic methods, obtaining 0.091 g L⫺1 (Capasso et al., 1999). Recently, hydroxytyrosol has become commercially available for research purposes and costs in a range of US$1000–2000 per gram (Obied et al., 2005). Some other polyphenols found in OMW with an agricultural application are methylcatechol, catechol, acetylcatechol, o-quinone, and guaiacol. As these possess bactericidal activity in their original concentration, these compounds can be used as pesticides in agriculture for the protection of olive trees (Bazoti et al., 2006).
40.3.1 Identification and Quantification of Phenolic Compounds Even though there is a plethora of analytical methods for the determination of phenolic compounds, including
spectrometry, high-performance liquid chromatography (Zhao and Lee, 2001), gas chromatography, and gas chromatography-mass spectrometry (Tasioula-Margari and Okogeri, 2001), there is still a demand for relatively simple analytical devices, suitable for screening, and rapid assays of these type of compounds in complex real samples, such as OMW (Torrecilla et al., 2007). In this context, electrochemical biosensors with laccase as the biological recognition element for the analysis of phenols have been developed by immobilizing them on different electrode surfaces such as carbon fibers, glassy carbon, graphite, carbon paste, polyethersulfone membranes on a universal sensor base electrode, polyaniline-modified interdigitated Pt sensors, PVP-gel deposited on a Clark electrode, and gold surface (Torrecilla et al., 2007). This kind of determination commonly uses enzymatic reactions combined with amperometric detection of the resulting product. For the determination of phenolic compounds, laccase is used as the enzymatic recognition part. This enzyme is well known for reducing oxygen directly into water without the intermediate formation of hydrogen peroxide at the expense of oxidation of phenols. Amperometric reduction of the generated products is then used as the quantification method,
CHAPTER | 40 Phenolic Compounds in Olive Oil Mill Wastewater
by simply applying reduction potentials. Necessary reduction potentials for this process are very near 0.0 V, which presents a great advantage as few substances interfere at this potential. Currently, the possible interferences can be resolved by linear or non-linear algorithms (Torrecilla et al., 2007).
40.3.2 Toxicity of Phenolic Compounds During the last few years, the toxicity of phenolic compounds such as tyrosol, hydroxytyrosol, catechol, protocatechuic acid, caffeic acid have been studied on different seeds (Cucumis sativus, Lepidium sativum, Sorghum bicolour, etc.) and different organisms (Daphnia magna, Thamnocephaus platyurus, Brachionus calyciflorus, Pseudokirchneriella subcapitata, etc.) (Fiorentino et al., 2003; Isidori et al., 2005). As can be seen in Table 40.5, the two most toxic phenolic compounds are catechol and hydroxytyrosol. In the light of these results, in the case of accidental introduction of large quantities of OMW into urban sewage treatment plants, severe phytotoxic effects may occur in the germination and seedling development (Greco et al., 2006). As Mekki et al. stated, the elimination of monomeric phenolic compounds such as tyrosol and hydroxytyrosol has led to a significant decrease in toxicity (Mekki et al., 2008), and many treatments of OMW focus on the removal of the phenolic compounds. Taking into account the applications of isolate phenols (see above), proposing reliable methods to reduce the phenolic concentration is important and the study of OMW management is essential.
40.4 OMW MANAGEMENT From 1953, when Professor Fiestas Ros de Ursinos published his first work related to OMW treatment methods up to today (Fiestas, 1953), more than 1000 studies have been published. These treatments can be classified in three wide groups depending on their nature: (i) biological; (ii) physicochemical and (iii) natural treatments. From the point of view of the reduction of phenolic compounds, all these methods have been briefly revised here.
40.4.1 Biological Treatments These methods are mainly based on the aerobic and anaerobic digestion of organic matter of OMW. In particular, the methods used to reduce the phenolic compounds have been shown here.
40.4.1.1 Aerobic Microorganisms Aerobic bacteria have been tested primarily as a method for the removal of monoaromatic or simple phenolics from
361
OMW. Aerobic bacteria appear to be very effective against some phenolic compounds and relatively ineffective against others. For example, Bacillus pumilus 123 is able to completely degrade protocatechuic and caffeic acids, but has much less affect on tyrosol. In general, using this bacterium, a 50% reduction in the phenolic content of OMW was found (Ramos-Cormenzana et al., 1996; McNamara et al., 2008). On the other hand, aerobic bacteria consortia from activated sludge (Borja et al., 1995a; Benitez et al., 1997), commercial communities (Ranalli, 1992), soil, or wastewater (Zouari and Ellouz, 1996) have been used to bioremediate the OMW. In this bio-remediation process, the concentration of phytotoxic compounds and chemical oxygen demand (COD) decreases by up to 80%, and the simple phenolic compounds are completely removed (McNamara et al., 2008). Fungal species as Geotrichum candidum, Lactobacillus plantarum, Phanerochaete chrysisporium, Panus tigrinus, etc. are used to reduce the phenolic content of OMW (Kapellakis et al., 2008). Fungal remediation removes simple phenolic compounds and reduces the COD. Fadil et al. have found that using Geotrichum sp., Aspergillus sp. and Candida tropicalis enables the removal of polyphenols in percentages of 46.6%, 44.3%, and 51.7%, respectively (Fadil et al., 2003). Tomati et al. reported the total removal of phenolic compounds in OMW by means of Pleorotus ostreatus (Tomati et al., 1991). Other research groups have found that the percentage of phenolic compound can be decreased by over 90% using Pleorotus eyingii, Pleorotus floridae and Pleorotus sajor-caju (Sanjust et al., 1991).
40.4.1.2 Anaerobic Microorganisms By treating OMW using anaerobic methods, two important advantages over the aerobic processes have been found. Firstly, the generation of methane which can be used as an energy power in other processes (Dalis et al., 1996; Erguder et al., 2000), and secondly, this process produces much less waste sludge than aerobic processes. Besides, more than 75% of toxic phenols and fatty acids are also removed (Dalis et al., 1996; McNamara et al., 2008). An important disadvantage of anaerobic digestion is based on the presence of toxic compounds in the methanogens process of the OMW. In order to overcome this problem, these compounds can be removed by aerobic pre-treatment (McNamara et al., 2008).
40.4.1.3 Combination of Aerobic and Anaerobic Microorganisms Borja et al. have demonstrated that aerobic pre-treatments increase the production of methane and reduce by up to 23% the phenolic compounds (Borja et al., 1995b,c). This research group compared anaerobic digestion of
362
TABLE 40.5 Edian effective concentration expressed as half maximal effective concentration (EC50). How the phenolic compounds (first column) affect microorganisms (first row) is quantified by numbers. These figures signify the molar phenolic compound concentration which produces 50% of the maximum possible response for the phenolic compounds. Lepidium sativum1
Cucumis sativus1
Sorghum bicolour1
Sorghum bicolour1
Daphnia magna2
Thamnocephaus platyurus2
Brachionus calyciflorus2
Pseudokirchneriella subcapitata2
1.09
0.52
0.40
10
8
17
34
p-Hydroxybenzoic acid
5.36
3.55
2.56
5.36
446
983
225
256
Protocatechuic acid
5.32
4.95
3.22
6.31
413
589
385
344
Vanillic acid
2.65
2.04
2.05
8.79
386
431
1
255
Syringic acid
2.15
2.05
1.55
1.94
177
97
141
214
4-Hydroxyphenylacetic acid
0.86
2.68
1.04
4.13
391
689
273
486
3,4-Dihydroxyphenylacetic acid
1.10
4.40
1.97
3.87
331
390
136
33
Homovanillic acid
5.23
6.62
3.30
5.23
268
299
407
440
Tyrosol
5.37
5.95
6.33
5.37
861
296
47
210
Hydroxytyrosol
1.02
1.55
0.47
0.82
11
4
9
120
3,4-Dihydroxyphenylethylene glycol
2.22
2.30
1.08
4.14
208
65
144
137
p-Coumaric acid
2.03
2.11
1.15
4.05
290
591
108
225
Caffeic acid
9.35
11.59
1.94
2.29
326
626
359
120
Ferulic acid
1.88
1.64
1.22
3.65
249
300
247
413
Sinapic acid
5.39
3.66
2.68
5.51
208
628
398
254
1
in mmol L⫺1 (Isidori et al., 2005).
2
in μmol L⫺1 (Fiorentino et al., 2003).
Non Fruit Aspects Including Mill Wastewater
1.07
SECTION | I
Catechol
CHAPTER | 40 Phenolic Compounds in Olive Oil Mill Wastewater
OMW pre-treated by two different fungi and a bacterium: Geotrichum candidum, Aspergillus terreus and Azotobacter chroococcum. These organisms partially removed the phenolic concentration and therefore the toxicity of OMW decreased by 59%, 87% and 79%, respectively (Borja et al., 1998). Using Candida tropicalis to aerobically pre-treat OMW prior to anaerobic digestion resulted in a reduction of 54% of the phenolic content. Pre-treating OMW with Geotrichum candidum reduced the phenolic and volatile fatty acid content and increased substrate uptake during anaerobic digestion (Martin et al., 1993).
40.4.2 Physicochemical Treatments In this group, methods such as distillation, evaporation, combustion or incineration, and flocculation-clarification of OMW can be found. These methods can be used mainly to concentrate or as the final treatment of OMW. Techniques such as adsorption or ion exchange among others are methods that could be used to eliminate phenols and polyphenols but these are usually applied in combination with others (Adhoum and Monser, 2004; Kapellakis et al., 2008). Other methods related to the decrease in the phenolic content of OMW using oxidative processes have been summarized here. Fenton reaction is based on chemical oxidation and coagulation of organic compounds present in OMW by means of hydrogen peroxide and ferrous sulfate. Apart from the COD reduction, by Fenton reaction, the phenolic content is notably reduced (Azabou et al., 2007). Nevertheless, the total removal of phenolic compounds could be reached using a combination of Fenton reaction and the coagulation process (Vlyssides et al., 2003; Mantzavinos and Kalogerakis, 2005). Electrochemical oxidation processes: given the high conductivity of OMW, the pollutants of OMW can be destroyed by means of oxidation processes. Panizza et al. removed almost completely the fraction based on aromatic compounds (Panizza and Cerisola, 2006). Advanced oxidation processes: in this group, oxidation processes based on ozonation are included. Using this method in combination with photocatalysis, UV irradiation and their combinations the total removal of phenol compounds can be achieved (Benitez et al., 1999; Mantzavinos and Kalogerakis, 2005).
363
(Sierra et al., 2001, 2007), although Marsilio et al. argued that spreading OMW on cultivated soils does not generate problems (Marsilio et al., 1990). Moreover, some groups stated that OMW spread on soil has an enriching effect (Levi-Minzi et al., 1992; Rouina et al., 1999). Because of this, before applying this treatment, the OMW characteristics, soil properties and the crops involved should be studied in detail, and then the advantages and disadvantages of natural or conventional treatment should be evaluated (Kapellakis et al., 2008). To reduce the concentration of phenolic compounds, therefore reducing its environmental impact, and for appropriate management of OMW, the best method to apply requires further detailed study.
40.5 CONCLUSION Production of olive oil involves the generation of environmentally hazardous waste olive oil mill wastewater. The seasonal nature of olive oil production implies that more than 3 ⫻ 107 m3 of OMW will be produced in just over 4 months every year. The main factor of the environmental impact of the waste is the toxicity of phenolic fractions and their slow biodegradation. Therefore, in order to optimize the management of OMW, various methods to determine, quantify and treat this waste are being developed.
SUMMARY POINTS ●
●
● ●
Brief description of the phenolic compounds role in olive oil. Where the phenolic compounds come from and their disadvantages and advantages. Main olive oil mill wastewater characteristics. Brief description of phenolic compounds. Method to identify and quantify the phenolic compound in olive oil mill wastewater. Toxicity of phenolic compounds. Olive oil mill wastewater management. Biological treatments. Aerobic, anaerobic and combinations of aerobic and anaerobic methods. Physicochemical treatments. Fenton reaction. Electrochemical oxidation. Ozonation. Natural treatments. ●
●
●
●
●
●
●
40.4.3 Natural Treatments The use of OMW on soils has two opposite effects. On one hand, considering the fertilizing properties of the waste (potassium, phosphorus, and nitrogen content and that OMW contains neither pathogenic nor heavy metals), spreading OMW on soils enhances fertility. But when the OMW is put onto soil their salinity and phenolic composition increases and it negatively affects crop production
●
●
●
REFERENCES Adhoum, N., Monser, L., 2004. Decolourization and removal of phenolic compounds from olive mill wastewater by electrocoagulation. Chem. Eng. Process. 43, 1281–1287.
364
Allouche, N., Fki, I., Sayadi, S., 2004. Toward a high yield recovery of antioxidants and purified hydroxytyrosol from olive mill wastewaters. J. Agric. Food. Chem. 52, 267–273. Aragón, J.M., Palancar, M.C., 2001. IMPROLIVE 2000, present and future of alpeorujo, Ed. Complutense S. A., Madrid. Azabou, S., Najjar, W., Gargoubi, A., Ghorbel, A., Sayadi, S., 2007. Catalytic wet peroxide photo-oxidation of phenolic olive oil mill wastewater contaminants – Part II. Degradation and detoxification of low-molecular mass phenolic compounds in model and real effluent. Appl. Catal. B-Environ. 77, 166–174. Bazoti, F.N., Gikas, E., Skaltsounis, A.L., Tsarbopoulos, A., 2006. Development of a liquid chromatography–electrospray ionization tandem mass spectrometry (LC–ESI MS/MS) method for the quantification of bioactive substances present in olive oil mill wastewaters. Issue Series Title: Anal. Chim. Acta 573–574, 258–266. Benitez, F.J., Beltran-Heredia, J., Torregrosa, J., Acero, J.L., 1999. Treatment of olive mill wastewaters by ozonation, aerobic degradation and the combination of both treatments. J. Chem. Technol. Biotechnol. 74, 639–646. Benitez, J., Beltran-Heredia, J., Torregrosa, J., Acero, J.L., Cercas, V., 1997. Aerobic degradation of olive mill wastewaters. Appl. Microbiol. Biotechnol. 47, 185–188. Borja, R., Alba, J., Banks, C.J., 1995a. Activated sludge treatment of wash waters derived from the purification of virgin olive oil in a new manufacturing process. J. Chem. Technol. Biotechnol. 64, 25–30. Borja, R., Alba, J., Garrido, S.E., Martinez, L., Garcia, M.P., Monteoliva, M., Ramos-Cormenzana, A., 1995b. Effect of aerobic pretreatment with Aspergillus terreus on the anaerobic digestion of olive-mill wastewater. Biotechnol. Appl. Biochem. 22, 233–246. Borja, R., Alba, J., Garrido, S.E., Martinez, L., Garcia, M.P., Incerti, C., Ramos-Cormenzana, A., 1995c. Comparative study of anaerobic digestion of olive mill wastewater (OMW) and OMW previously fermented with aspergillus terreus. Bioprocess. Biosyst. Eng. 13, 317–322. Borja, R., Alba, J., Mancha, A., Martin, A., Alonso, V., Sanchez, E., 1998. Comparative effect of different aerobic pretreatments on the kinetics and macroenergetic parameters of anaerobic digestion of olive mill wastewater in continuous mode. Bioprocess. Biosyst. Eng. 18, 127–134. Borja, R., Raposo, F., Rincón, B., 2006. Treatment technologies of liquid and solid wastes from two-phase olive oil mills. Issue Series Title: Grasas Aceites 57, 32–46. Capasso, R., Evidente, A., Avolio, S., Solla, F., 1999. A highly convenient synthesis of hydroxytyrosol and its recovery from agricultural waste waters. Issue Series Title: J. Agric. Food Chem. 47, 1745–1748. Dalis, D., Anagnostidis, K., Lopez, A., Letsiou, I., Hartmann, L., 1996. Anaerobic digestion of total raw olive-oil wastewater in a two-stage pilot-plant (up-flow and fixed-bed bioreactors). Issue Series Title: Biores. Tech. 57, 237–243. Davies, L.C., Vilhena, A.M., Novais, J.M., Martins-Dias, S., 2004. Olive mill wastewater characteristics: modelling and statistical analysis. Issue Series Title: Grasas Aceites 55, 233–241. Erguder, T.H., Guven, E., Demirer, G.N., 2000. Anaerobic treatment of olive mill wastes in batch reactors. Issue Series Title: Process Biochem. 36, 243–248. Fadil, K., Chahlaouia, A., Ouahbib, A., Zaida, A., Borja, R., 2003. Aerobic biodegradation and detoxification of wastewaters from the olive oil industry. Int. Biodeterior. Biodegrad. 51, 37–41. Fernández-Bolaños, J., Rodríguez, G., Rodríguez, R., Heredia, A., Guillén, R., Jiménez, A., 2002. Production in large quantities of
SECTION | I
Non Fruit Aspects Including Mill Wastewater
highly purified hydroxytyrosol from liquid-solid waste of two-phase olive oil processing or “alperujo”. Issue Series Title: J. Agric. Food Chem. 50, 6804–6811. Fiestas, J.A., 1953. Estudio del alpechín para su aprovechamiento industrial I. concentración de los azúcares y demás substancias que lleva en emulsión y disolución por tratamiento con óxido de calico. Issue Series Title: Grasas Aceites 4, 63–67. Fiorentino, A., Gentili, A., Isidori, M., Monaco, P., Nardelli, A., Parrella, A., Temussi, F., 2003. Environmental effects caused by olive mill wastewaters: toxicity comparison of low-molecular-weight phenol components. Issue Series Title: J. Agric. Food Chem. 51, 1005–1009. Greco, G., Colarieti, M.L., Toscano, G., Iamarino, G., Rao, M.A., Gianfreda, L., 2006. Mitigation of olive mill wastewater toxicity. Issue Series Title: J. Agric. Food Chem. 54, 6776–6782. International Olive Council. 2008. www.internationaloliveoil.org. Isidori, M., Lacorgna, M., Nardelli, A., Parrella, A., 2005. Model study on the effect of 15 phenolic olive mill wastewater constituents on seed germination and vibrio fischeri metabolism. Issue Series Title: J. Agric. Food Chem. 53, 8414–8417. Kapellakis, I.E., Tsagarakis, K.P., Crowther, J.C., 2008. Olive oil history, production and by-product management. Rev. Environ. Sci. Biotechnol. 7, 1–26. Levi-Minzi, R., Saviozzi, A., Riffaldi, R., Falzo, L., 1992. Land application of vegetable water: effects on soil properties. Issue Series Title: Olivae 40, 20–25. Mantzavinos, D., Kalogerakis, N., 2005. Treatment of olive mill effluents Part I. Organic matter degradation by chemical and biological processes—an overview. Issue Series Title: Environ. Int. 31, 289–295. Marsilio, V., Di Giovacchino, L., Lombardo, N., Briccoli-Bati, C., 1990. First observations on the disposal effects of olive mills vegetation waters on cultivated soil. Issue Series Title: Acta. Hort. 286, 493–498. Martin, A., Borja, R., Chica, A., 1993. Kinetic study of an anaerobic fluidized bed system used for the purification of fermented olive mill wastewater. J. Chem. Technol. Biotechnol. 56, 155–162. McNamara, C.J., Anastasiouc, C.C., O’Flahertyd, V., Mitchell, R., 2008. Bioremediation of olive mill wastewater. Int. Biodeterior. Biodegrad. 61, 127–134. Mekki, A., Dhouib, A., Feki, F., Sayadi, S., 2008. Assessment of toxicity of the untreated and treated olive mill wastewaters and soil irrigated by using microbiotests. Ecotox. Environ. Safe 69, 488–495. Miranda, M.A., Galindo, F., Amat, A.M., Arques, A., 2001. Pyrylium saltphotosensitised degradation of phenolic contaminants present in olive oil wastewaters with solar light: Part II. Benzoic acid derivatives. Appl. Catal. B-Environ. 30, 437–444. Obied, H.K., Allen, M.S., Bedgood, D.R., Prenzler, P.D., Robards, K., Stockmann, R., 2005. Bioactivity and analysis of biophenols recovered from olive mill waste. J. Agric. Food Chem. 53, 823–837. Panizza, M., Cerisola, G., 2006. Olive mill wastewater treatment by anodic oxidation with parallel plate electrodes. Wat. Res. 40, 1179–1184. Ramos-Cormenzana, A., Juarez-Jimenez, B., Garcia-Pareja, M.P., 1996. Antimicrobial activity of olive mill waste-waters (alpechín) and biotransformed olive oil mill wastewater. Int. Biodeterior. Biodegrad. 38, 283–290. Ranalli, A., 1992. Microbiological treatment of oil mill waste waters. Grasas Aceites 43, 16–19. Rodis, P.S., Karathanos, V.T., Mantzavinou, A., 2002. Partitioning of olive oil antioxidants between oil and water phases. J. Agric. Food Chem. 50, 596–601. Rouina, B.B., Taamallah, H., Ammar, E., 1999. Vegetation water used as a fertilizer on young olive plants. Acta Horticulturae 474, 353–355.
CHAPTER | 40 Phenolic Compounds in Olive Oil Mill Wastewater
Salami, M., Galli, C., De Angelis, L., Visioli, F., 1995. Formation of F2isoprostanes in oxidized low density lipoprotein: inhibitory effect of hydroxytyrosol. Pharmacol. Res. 31, 275–279. Sanjust, E., Pompei, R., Rescigno, A., Rinaldi, A., Ballero, M., 1991. Olive milling wastewater as a medium for growth of four Pleurotus species. Appl. Biochem. Biotechnol. 31, 223–235. Sierra, J., Martí, E., Garau, M.A., Cruañas, R., 2007. Effects of the agronomic use of olive oil mill wastewater: Field experiment. Sci. Total Environ. 378, 90–94. Sierra, J., Martí, E., Montserrat, G., Cruañas, R., Garau, M.A., 2001. Characterisation and evolution of a soil affected by olive mill wastewater disposal. Sci. Total Environ. 279, 207–214. Tasioula-Margari, M., Okogeri, O., 2001. Isolation and characterization of virgin olive oil phenolic compounds by HPLC/UV and GC-MS. J. Food Sci. 66, 530–534. Tomati, U., Galli, E., Di Lena, G., Buffone, R., 1991. Induction of laccase in Pleurotus ostreatus mycelium grown in olive oil waste waters. Agrochimica 35, 275–279. Torrecilla, J.S., Aragón, J.M., Palancar, M.C., 2006. Improvement of fluidized bed dryers for drying olive oil mill solid waste (olive pomace). Eur. J. Lipid Sci. Technol. 108, 913–924.
365
Torrecilla, J.S., Mena, M.L., Yáñez-Sedeño, P., García, J., 2007. Application of artificial neural network to the determination of phenolic compounds in olive oil mill wastewater. J. Food Eng. 81, 544–552. Visioli, F., Galli, C., 1998. Olive oil phenols and their potential effects on human health. J. Agric. Food Chem. 46, 4292–4296. Visioli, F., Galli, C., Plasmati, E., Viappiani, S., Hernandez, A., Colombo, C., Sala, A., 2000. Olive phenol hydroxytyrosol prevents passive smoking-induced oxidative stress. Circulation 102, 2169–2171. Vlyssides, A., Loukakis, H., Israilides, C., Barampouti, E.M., Mai, S., 2003. Detoxification of olive mill wastewater using a Fenton process. In: Kalogerakis, N. (ed.), 2nd European Bioremediation Conference. Crete, pp. 513-534. Zhao, L.M., Lee, H.K., 2001. Determination of phenols in water using liquid phase microextraction with back extraction combined with high-performance liquid chromatography. J. Chromatogr. A 931, 95–105. Zouari, N., Ellouz, R., 1996. Microbial consortia for the aerobic degradation of aromatic compounds in olive oil mill effluent. J. Ind. Microbiol. Biotechnol. 16, 155–162.
This page intentionally left blank
1.3
Stability, Microbes, Contaminants and Adverse Components and Processes Bacterial and Fungal and Other Microbial Aspects Pesticides and Adulterants Toxicology and Contaminants
This page intentionally left blank
Chapter 41
Lactic Acid Bacteria in Table Olive Fermentation Cinzia L. Randazzo,1 Rajkumar Rajendram,2 and Cinzia Caggia1 1 2
DOFATA-Dipartimento di Orto Floro Arboricoltura e Tecnologie Agroalimentari, Catania, Italy Nutritional Sciences Research Division, School of Life Sciences, King’s College London, UK
41.1 INTRODUCTION Freshly picked olives contain phenols and oleuropein, which makes the fruit bitter and unpalatable. There are many ways to de-bitter olives. Fermentation of olives in brine is one of the oldest methods of preserving and processing food known to man. Traditional methods of fermentation use the indigenous microflora (e.g. lactic acid bacteria; LAB), encouraging the growth of those that ferment the olive. The indigenous LAB involved in table olive fermentation mainly belong to the genus Lactobacillus. Lactic acid bacteria which belong to Leuconostoc, Streptococcus, Enterococcus, and Pediococcus genera are also present in smaller numbers. The main factor determining the quality of the olives produced is the LAB involved in fermentation. The use of starter cultures can accelerate and improve fermentation and also prevent the growth of undesirable microorganisms. Although table olive production remains an empirical process, interest in LAB starter cultures for processing olives is increasing. The selection of LAB for starter cultures focuses on several important criteria, including their ability to monitor the fermentation process and improve the safety of table olives. Bacteriocin production and the probiotic nature of LAB are also important in the selection of LAB starter cultures. This chapter provides an overview of the LAB species involved in the preparation of table olives. Methods available for the detection and identification of LAB species involved in table olive fermentation are also described. The selection and use of LAB starter cultures to enhance fermentation and prevent the growth of undesirable microorganisms are discussed.
●
●
treated green table olives in brine ‘Spanish-style,’ ‘green olives’ (de Castro et al., 2002) untreated ‘natural black olives,’ ‘Greek-style’ (Tassou et al., 2002).
The main difference between these preparations is the degree of ripeness of the raw olives used. The Spanishstyle olives are picked when they are a yellow-green color. Natural black olives are picked only when completely ripe. In each case a series of common processing steps are then performed prior to de-bittering and fermentation. A simple recipe is described here. The olives are washed by soaking in water, drained and stored in food-grade containers. Water leaves the olives entering the brine whilst the olives take up salt. The olives are held fully immersed by the brine within the container. Lightly sealing the olives allows the gaseous products of fermentation to escape. Although unpalatable, the phenols and oleuropein are not poisonous so the olives can be tasted at any time. The olives are usually edible within 2–4 weeks, but may be left for up to 3 months.
41.3 DE-BITTERING AND FERMENTATION
41.2 PRODUCTION OF TABLE OLIVES Table olives are a fermented plant product. They are considered by many food scientists to be a ‘food of the future’ Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
because of the beneficial polyphenolic antioxidants they contain (Buckenhüskes, 1993). Freshly picked olives contain phenols and oleuropein (glucoside) which are bitter and unpalatable but not toxic. The removal of these compounds from olives (de-bittering) by fermentation in brine is one of the oldest known methods of preserving and processing food. This process requires lactic acid bacteria (LAB) to promote fermentation by metabolizing the vegetable sugars to organic acids which, together with salt, preserve the olives. The two main commercial preparations of fermented table olives are:
Spanish-style green olives are initially de-bittered in an alkaline, sodium hydroxide (lye) solution. This hydrolyzes the bitter-tasting oleuropein and demethyloleuropein (Ciafardini
369
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
370
SECTION | I Bacterial and Fungal and Other Microbial Aspects
et al., 1994), improving the taste of the final product. The olives then undergo lactic fermentation in brine. Fermentation leaches out and breaks down oleuropein and phenolic compounds. Fermentation produces lactic acid, a natural preservative, and also produces a complex of other flavorsome byproducts. If fermentation is complete, the olives only need the appropriate brine for preservation (i.e. pH ⬍ 3.5 and NaCl ⬎ 5.0%) and can be stored with or without refrigeration. This process is generally known as Spanish-style preparation of green olives (de Castro et al., 2002). Preservation of partially fermented olives requires sterilization, pasteurization, preservatives, refrigeration or treatment with inert gas (without brine). Untreated natural black olives are put directly into brine and the secoiridoid glucosides are hydrolyzed by the enzymatic activities of indigenous microorganisms (Tassou et al., 2002). The olives are preserved by natural fermentation in brine alone, sterilization, pasteurization or adding preservatives. This process is generally known as Greekstyle preparation (Tassou et al., 2002). Fermentation can be ‘lactic’ (Spanish) or ‘alcoholic-lactic’ (Greek or naturally green olives). Lactic fermentation is
performed almost exclusively by LAB, mainly lactobacilli and to a lesser degree Leuconostoc and Pediococcus genera (Fleming et al., 1984). Key features of LAB are shown in Table 41.1. During lactic fermentation several species (Lactobacillus plantarum, Lactobacillus pentosus, Lactobacillus rhamnosus, Lactobacillus paracasei) flourish within a complex ecosystem (Ruiz-Barba et al., 1994). This system is influenced by the indigenous microbial flora present on the olives initially (Tassou, 1993; Garrido Fernández et al., 1997) as well as intrinsic and extrinsic factors (Garrido Fernández et al., 1995). These factors are listed in Table 41.2.
41.4 LAB, NATURALLY OCCURRING MICROORGANISMS DURING TABLE OLIVE FERMENTATION The microbial succession in spontaneous olive fermentation is generally characterized by distinct phases (Garrido Fernández et al., 1997). The initial stage is characterized by the growth of Gram-negative bacteria (Enterobacteriaceae), yeasts and molds. The flora present at this stage also
TABLE 41.1 Key features of lactic acid bacteria (LAB). 1. LAB are Gram-positive, non-motile, catalase-negative, non-spore-forming, rod- and coccus-shaped bacteria 2. They are classified as Generally Recognized As Safe (GRAS) microorganisms 3. They ferment carbohydrates and higher alcohols to form lactic acid predominantly 4. Pathways of hexose metabolism divide LAB into two groups; homofermentative and heterofermentative 5. Homofermenters such as Pediococcus, Streptococcus, Lactococcus and some lactobacilli produce lactic acid as the major or sole end-product of glucose fermentation 6. Heterofermenters such as Weissella and Leuconostoc and some lactobacilli produce equimolar amounts of lactate, CO2 and ethanol from glucose 7. They are generally mesophilic but can grow at temperatures as low as 5°C or as high as 45°C. Similarly, while the majority of strains grow at pH 4.0–4.5, some remain active at pH 9.6 and others at pH 3.2 8. Most of the LAB strains are weakly proteolytic and lipolytic and require preformed amino acids, purine and pyrimidine bases and B vitamins for growth 9. LAB are generally found in a variety of natural and man-made habitats. Natural habitats include the mucosal membranes of humans and animals (oral cavity, intestine and vagina), plants and material of plant origin and manure. Man-made habitats include sewage and fermenting or spoiling food 10. LAB are used as starter cultures in the fermentation of several food products 11. The LAB found in food belong to the genera: Carnobacterium, Enterococcus, Lactobacillus, Lactococcus, Leuconostoc, Oenococcus, Pediococcus, Streptococcus, Tetragenococcus, Vagococcus, and Weissella 12. LAB produce bacteriocins and are probiotic. These properties are important in many aspects of food development, food safety and health This table lists the key features of LAB, including phenotypic traits, metabolism, habitat, and role in fermented food, including food safety and healthy attitude.
CHAPTER | 41 Lactic Acid Bacteria in Table Olive Fermentation
TABLE 41.2 Main factors influencing microbial ecosystem of table olive fermentation. Initial microbial population present on freshly harvested olives and therefore on olives in the field
Enterobacteriaceae Pseudomonas Leuconostoc Pediococcus Lactobacillus Yeasts and molds
Intrinsic factors
pH Water activity Availability of nutrients Olive structure
Extrinsic factors
Temperature Salt concentration
This table lists the factors which influence the metabolism of the developing microbial population. These factors determine the LAB population and their dynamics during olive fermentation.
includes small numbers of Gram-positive bacteria belonging to Leuconostoc, Streptococcus, Pediococcus and Bacillus genera. The growth of LAB increases from 1% in the initial fresh brine to 80% of the total population a few days later. Spoilage may occur if undesirable bacteria (e.g. Clostridia, coliforms and Pseudomonas) are not controlled by rapid growth of LAB. The salt concentration and pH of the environment are the main factors determining the growth of LAB. Currently, the most common methods used to facilitate the growth of indigenous LAB include: ● ● ●
●
reducing brine pH to less than 4.5 adding glucose maintaining salt concentration between 5–6%, during the initial phase increasing salt concentration to 7% by the end of fermentation.
The second and most important stage of olive fermentation is characterized by an increase in LAB of the genera Leuconostoc and Pediococcus. The concentration of yeasts and Gram-negative bacteria declines. The time between the first and the second phase should be as short as possible to encourage the rapid dominance of LAB over the epiphytic and contaminant microbial flora (Panagou and Katsaboxakis, 2006).
371
The third and longest phase can take up to two months after the olives are initially left in the brine to complete. The duration of this phase varies depending on the initial sugar and salt concentrations, temperature and LAB dominance. During this phase, lactobacilli proliferate. These are responsible for the flavor and texture of the final products (Montaño et al., 1993; Garrido Fernández et al., 1997; Sánchez et al., 2000). The dominant species during spontaneous fermentation are Lb. plantarum and Lb. pentosus. These can withstand and even grow in environments with pH as high as 8.5–9.5 (Balatsouras et al., 1983; Balatsouras, 1985). Lb. plantarum generally co-exists with the yeast population until fermentation is completed and during storage (Ruiz-Barba et al., 1994; Garrido Fernández et al., 1995), particularly in black and naturally green olive fermentation (Marquina et al., 1992; Tassou et al., 2002; Durán Quintana et al., 2003; Hernández et al., 2006). Lb. plantarum is the most common species found in both green and black olives. It is adaptable and is very metabolically active (Adam and Moss, 1995; Garrido Fernández et al., 1995; Durán Quintana et al., 1999; Sánchez et al., 2001). It can tolerate acid environments and is even able to grow at salt concentrations of 8%. In addition, some strains of Lb. plantarum which have been isolated from olive brines, produce β-glucosidase. This can break down oleuropein and several other substrates (Marsilio and Lanza, 1998). These strains can therefore resist the inhibitory effect of oleuropein and its derivates on population growth. Figures 41.1 and 41.2 show Lb. plantarum strains, isolated from olive brine and taken directly from brine respectively. Figure 41.3 shows the β-glucosidase activity of Lb. plantarum strains, isolated from green table olives. Several other lactobacilli and enterococci are also present on olives (Van den Berg et al., 1993; Garrido Fernández et al., 1997). Studies of Italian olives found Lb. plantarum, Lb. brevis, Lb. casei and Lb. fermentum on the olives and also in the brine (Giudici et al., 1997; Randazzo et al., 1999, 2004).
41.5 USE OF LAB STARTER CULTURES IN THE TABLE OLIVE FERMENTATION Until recently, the use of pure LAB starter cultures in the fermentation of vegetables was uncommon. Interest in the development and use of starter cultures for table olive fermentation has been steadily increasing. This is predominantly because industrial experience suggests that an appropriate inoculation of LAB can help to achieve a more controlled process. In most cases, the starter cultures used are produced from wild strains of LAB isolated from previous fermentations. This is known as back-slopping. Unfortunately, the metabolic activities of back-slopped starter cultures are inconsistent, varying even among strains. For example, there may be differences in growth
372
SECTION | I Bacterial and Fungal and Other Microbial Aspects
FIGURE 41.1 Lactobacillus plantarum isolated from olive brine.
FIGURE 41.3 β-Glucosidase activity of Lactobacillus plantarum, isolated from green table olives.
olive fermentation (Panagou et al., 2003; Panagou and Tassou, 2006). The use of LAB starter cultures affected the acidification of the brine, the survival of Gram-negative bacteria and improved the quality of the final product by accelerating metabolic activity (Panagou et al., 2008).
41.5.1 Criteria Used to Select LAB Starter Cultures for Table Olive Fermentation
FIGURE 41.2 Lactobacillus plantarum in olive brine.
rate, adaptation to a particular substrate, antimicrobial properties and competitive growth behavior in mixed cultures (Holzapfel, 1997). These differences may affect the flavor and quality of the table olives produced. The preparation of starter cultures specifically for the fermentation of table olives has been described (Comi et al., 2000) but predominantly for the Spanish process. Natural black olives are produced almost exclusively by natural spontaneous fermentation. There are very little available data on the use of starter cultures in this process. However, Panagou et al. (2008) recently evaluated the effect of selected LAB strains on the Corservolea fermentation of natural black olives. The mixed starter cultures used consisted of a Lb. plantarum strain with a Lb. pentosus strain. The Lb. plantarum strain which was isolated from fermented cassava was chosen for its β-glucosidase activity. The Lb. pentosus strain was previously used in green
The criteria used to select LAB to use as starter cultures for table olive fermentation are listed in Table 41.3. Firstly, the starter must be able to grow in the presence of oleuropein and verbascoside. These natural inhibitors would otherwise limit the adaptation of the LAB to the brine environment (Panagou et al., 2003). Lb. plantarum, the most suitable species to use as starter for green table olive fermentation (Ruiz-Barba et al., 1993) is able to grow in the presence of oleuropein and verbascoside (Ciafardini et al., 1994). Tolerance of high saline concentrations is also extremely important (Brenes and de Castro, 1998). Strains of LAB isolated from the brine of black olive samples could grow at 8% NaCl, and most Lb. plantarum species could even grow in 10% NaCl (Tassou et al., 2002). However, the optimum conditions for developing free acid levels as high as 1.2% were obtained at 25°C and salt concentration of 6% (Tassou et al., 2002). Successful fermentation requires relatively high temperatures (⬎18°C). Higher temperatures facilitate the diffusion of nutrients into the brine, resulting in a vigorous fermentation by LAB. This is particularly important during the initial period of brining. However, in most processing
373
CHAPTER | 41 Lactic Acid Bacteria in Table Olive Fermentation
plants, fermenters are exposed to the open air and so fermentation temperatures vary with ambient temperature. This is particularly relevant in winter when the prevailing low temperatures in olive-producing areas with cold climates retard microbial activity. Thus it is essential that the starter cultures are able to grow at low temperatures. Selected wild strains of Lb. plantarum, isolated from table olive cold fermentation brines, exhibited good performance at low temperature in brines which initially contained 3% NaCl and pH 5.0 (Durán Quintana et al., 1999). Similar results were obtained with a selected strain of Lb. pentosus inoculated in green olives lye-treated, at alkaline pH (Sánchez et al., 2001). In pilot-scale fermentations using the 1MO strain Lb. pentosus the time taken for the biological de-bittering of black olives was reduced from 6 months to 8 days (Servili et al., 2006).
TABLE 41.3 Selection criteria for lactic acid bacteria used for table olive fermentation. Rapid and predominant growth Homo- and heterofermentative metabolism Salt tolerance Acid production and tolerance Inability to metabolize organic acids Growth at low temperature Tolerance of phenolic glycosides Ability to hydrolyze oleuropein Ability to reduce the biogenic amine content
41.5.2 Concerns About Using Enterococci in Starter Cultures Enterococcus faecium and other enterococci have been isolated from the brines of olives produced in different countries (Asehraou et al., 1992; van den Berg et al., 1993, Lavermicocca et al., 1998). As a result the use of mixed starter cultures containing Lb. plantarum and E. faecium (Lavermicocca et al., 1998) or Lb. pentosus and Enterococcus casseliflavus (de Castro et al., 2002) have been proposed. However, enterococci can cause infections in humans and the determinants of virulence are not fully understood. As a result of these concerns enterococci have not been proposed by European Food Safety Authority as ‘Qualified Presumption of Safety’ (QPS) status (EFSA, 2007).
41.5.3 LAB Starters to Improve Table Olive Safety Many pathogenic microorganisms which can cause spoilage are able to survive and grow at low pH. This is clinically relevant because the highly pathogenic bacteria Escherichia coli O157:H7 can survive during the production of Spanish-style green table olives (Spyropoulou et al., 2001). Listeria monocytogenes has also been isolated from both commercially produced and home-made green olives (Caggia et al., 2004). Nevertheless, the rapid acidification (pH reduction) of brines induced by LAB starter cultures increases the prevalence of LAB in the epiphytic population and reduces the risk of pathogen growth. These actions significantly improve the safety of table olives. Furthermore, several bacteria produce toxic compounds (e.g. biogenic amines) in acidic conditions. The production of biogenic amines is influenced by the presence of microorganisms with amino acid decarboxylase activity. This is greater in acidic environments (pH 4.0–5.5; Garcia-Garcia
Formation of flavor precursors Bacteriophage resistance Bacteriocin production Probiotic attitude This table lists the features required for the application of LAB strains as starter cultures in table olive fermentation.
et al., 2004). Hence, the LAB decarboxylase activity should be considered when selecting organisms for use in starter cultures.
41.5.4 Use of LAB Starter Cultures to Improve the Beneficial Effects of Table Olive Consumption The acid production from starter cultures and the shorter fermentation time can reduce the oxidative degradation of naturally occurring phenolic antioxidants in olives (Servili et al., 2006). Antioxidants are at least partly responsible for the beneficial effect of consumption of olives on health. Other important functional considerations include: (i) the ability to produce antimicrobial compounds (i.e. bacteriocins) and (ii) the probiotic properties.
41.5.4.1 Production of Antimicrobial Compounds by LAB Bacteriocins are low-molecular-weight peptides or proteins produced by LAB. Bacteriocins have antimicrobial properties restricted to related Gram-positive bacteria. The production of bacteriocins in situ may increase the
374
SECTION | I Bacterial and Fungal and Other Microbial Aspects
competitiveness of the producing strain at the expense of other strains and so help prevent spoilage (Ross et al., 2002; Settani and Corsetti, 2008). Lactic acid bacteria starter cultures can produce their bacteriocins within the olive matrix and so inhibit pathogenic bacterial strains and spoilage of sensitive food (Ruiz-Barba et al., 1994). Bacteriocin-producing strains of Lb. plantarum have received great interest in recent years (Jimenez-Diaz et al., 1993, Ruiz-Barba et al., 1994; Leal Sánchez et al., 1998; Maldonado et al., 2002). Studies, carried out using molecular methods, demonstrated that the plantaricin S operon is widely distributed among wild-type Lb. plantarum strains isolated from olive fermentations (Maldonado et al., 2002). Bacteriocin-producing strains which can gain dominance over epiphytic bacteria and persist during the olive fermentation may be useful in the development of functional starter cultures, ensuring a higher uniformity of the final products (Leal Sánchez et al., 1998; Leal-Sánchez et al., 2003). Interesting bacteriocin-producing strains include Lb. plantarum LPCO10 (Jimenez-Diaz et al., 1993), Lb. pentosus B96 (Delgado et al., 2005) and E. faecium BFE 900 (Franz et al., 1996). The ability of Lb. plantarum strain LPCO10 to produce plantaricins S and T and proliferate over the epiphytic microflora of Spanish-style fermented olives has also been evaluated (Jimenez-Diaz et al., 1993; Ruiz-Barba et al., 1994). These bacteriocins are active against several natural competitors of Lb. plantarum in brine fermentation and are also active against spoilage bacteria. Leal-Sánchez et al. (2003) compared the fermentation profile of green olives produced with Lb. plantarum LPCO10 with the spontaneous process under various different conditions. They demonstrated that Lb. plantarum LPCO10 inoculated at 107 CFU (colony forming units) ml⫺1 was able to gain dominance over the natural LAB population. Furthermore Lb. plantarum LPCO10 accelerated the pH decrease during the first phase of fermentation and yielded higher free and correct total acidities. Bacteriocinproducing strains of Lb. plantarum ST23LD and ST341LD, E. faecium ST311LD and Leuc. mesenteroides subsp. mesenteroides ST33LD were also isolated from the brine of spoiled black olives (Todorov and Dicks, 2005).
41.6 DETECTION AND IDENTIFICATION OF LAB IN TABLE OLIVES AND CONCLUDING REMARKS LAB has a fundamentally important role in olive fermentation. The identification and characterization of LAB is one of the principal aims of modern applied microbiology. The development of a rapid and reliable method for the detection of LAB in natural ecosystems is actively being investigated. A schematic outline of the molecular approaches that may be used to study the microbial flora present on olives is illustrated in Figure 41.4. Traditionally bacterial number and diversity in table olives were assessed by culturing on selective media. The bacterial genus and species were subsequently identified by phenotypic characterization. The use of molecular biological techniques to identify LAB has increased in recent years. Randazzo et al. (2004) assessed the microbial flora of fermented olives using molecular methods such as restriction fragment length polymorphism (RFLP) analysis and demonstrated the dominance of Lb. casei in naturally fermented green olives. Figure 41.5 shows the RFLP profile of the 16S rRNA gene of several strains of lactobacilli that were isolated during table olive fermentation. However, taxonomists are aware that conventional methods do not always allow the precise detection of the phylogenetic relationships between certain groups of bacteria. Until recently the application of culture-independent approaches to the investigation of olive fermentation has been limited. However, in 2006, Ercolini et al. (2006) described the use of fluorescence in situ hybridization (FISH) to detect the presence of Lb. plantarum in samples of black olives from different areas within the Campania region of Italy. The microbial ecosystem which develops during the production of table olives is dynamic. It is therefore necessary to combine several different methods of investigation in a polyphasic approach to understand the microbial interactions which occur during olive fermentation. Further studies in this field are vital to improve the safety and quality of table olives.
SUMMARY POINTS 41.5.4.2 Probiotic Properties of LAB Probiotics are viable microorganisms that benefit the health of the host when ingested. Recently, Lavermicocca et al. (2005) evaluated the ability of probiotic strains of Lactobacillus and Bifidobacterium to survive on the surface of olives. Lactobacillus paracasei survived longer (up to 3 months) than other species, while Bifodobacteria were detected in the feces of human volunteers at high concentration (106 CFU g⫺1) for 30 days (Lavermicocca et al., 2005). Thus, table olives could be used to develop novel, tasty, probiotic vegetable foods.
● ●
●
●
Table olives are a fermented plant product. The microbial flora involved in table olive fermentation are part of a complex ecosystem, which is influenced by several intrinsic and extrinsic factors related to brine composition. The microbial population initially present on the olives is relevant. Lactic acid bacteria (LAB) are part of the indigenous microbial flora of olives and determine, at least in part, the quality of the fermentation. The indigenous LAB involved in table olive fermentation are predominantly lactobacilli. Species of the Leuconostoc,
375
CHAPTER | 41 Lactic Acid Bacteria in Table Olive Fermentation
FISH
●
Cultivation
Colony hybridization
Table olive samples
Isolates DNA/RNA
RT/PCR
DNA
Molecular fingerprinting
PCR
T/DGGE
Cloning and sequencing
Phylogeny
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
FIGURE 41.5 Endonuclease Hae III restriction fragment length polymorphism (RFLP) patterns of the 16S rRNA gene of lactobacilli, isolated during table olive fermentation. Lane 6: 100 bp DNA ladder.
●
●
●
●
●
●
ACKNOWLEDGMENTS The authors thank Giovanni Fava and Aldo Todaro for technical advice and assistance.
FIGURE 41.4 Flowchart of molecular methods that could be used alone or in combination to analyze table olive microbial communities. FISH, fluorescence in situ hybridization; DNA, deoxyribonucleic acid; PCR, polymerase chain reaction; RT, reverse transcriptase; T/DGGE, temperature/denaturing gradient gel electrophoresis.
●
A polyphasic approach using culture-dependent and culture-independent methods (e.g. FISH) is required to understand the interactions which occur between indigenous microbial flora and the environment during the fermentation of olives.
Streptococcus, Enterococcus, and Pediococcus genera are also present in lower concentrations. The main factors determining the growth of LAB population during spontaneous fermentation are salt concentration and environmental pH. Lactobacillus plantarum is the most common species found during fermentation of both Spanish and black olives because it is adaptable and very metabolically active. Lactobacillus pentosus, Lactobacillus brevis, Lactobacillus casei, Lactobacillus fermentum and some enterococci have also been found in olive brines. LAB starter cultures may be added to accelerate and improve the quality and consistency of fermentation. The selection of LAB for starter cultures depends on various criteria including the ability to promote fermentation and improve the safety of table olives. Bacteriocin production and probiotic properties are important features considered when LAB are selected for use as starter cultures. The methods available for the detection and identification of LAB species involved in table olive fermentation are very limited and generally culture-dependent.
REFERENCES Adam, M.R., Moss, M.O., 1995. Food Microbiology. The Royal Society of Chemistry, North Yorkshire. Asehraou, A., Faid, M., Jana, M., 1992. Physicochemical properties and the microflora of Moroccan black table olives. Grasas y Aceites 43, 130–133. Balatsouras, G., Tsibri, A., Dalles, T., Doutsias, G., 1983. Effects of fermentation and its control on the sensory characteristics of Conservolea variety green olives. Appl. Environ. Microbiol. 46, 68–74. Balatsouras, G., 1985. Taxonomic and physiological characteristics of the facultative rod type lactic acid bacteria isolated from fermenting green and black olives. Grasas Aceites 36, 239–249. Brenes, M., and de Castro, A., 1998. Transformation of oleuropein and its hydrolysis products during Spanish-style green olive processing. J. Sci. Food Agric. 77, 353–358. Buckenhüskes, H.J., 1993. Selection criteria for lactic acid bacteria to be used as starter cultures for various food commodities. FEMS Microbiol. Rev. 12, 253–272. Caggia, C., Randazzo, C.L., Di Salvo, M., Romeo, F., Giudici, P., 2004. Occurrence of Listeria monocytogenes in green table olives. J. Food Prot. 64, 2189–2194. Ciafardini, G., Marsilio, V., Lanza, B., Pozzi, N., 1994. Hydrolysis of oleuropein by Lactobacillus plantarum strains associated with olive fermentation. Appl. Environ. Microbiol. 60, 4142–4147. Comi, G., Manzano, M., Cocolin, L., Vailanti, P., Cantoni, C., Giomo, A., 2000. Valutazione dell’attività di starters microbici nella preparazione delle olive da mensa. Ind. Aliment. 39, 1258–1265. de Castro, A., Montaño, A., Casado, F.J., Sánchez, A.H., Rejano, L., 2002. Utilization of Enterococcus casseliflavus and Lactobacillus pentosus as starter culture for Spanish-style green olive fermentation. Food Microbiol. 19, 637–644. Delgado, A., Brito, D., Peres, C., Noe-Arrooyo, F., Garriso-Fernandez, A., 2005. Bacteriocin production by Lactobacillus pentosus B96 can be expressed as a function of temperature and NaCl concentration. Food Microbiol. 22, 521–528. Durán Quintana, M.C., García García, P., Garrido Fernández, A., 1999. Establishment of conditions for green table olive fermentation at low temperature. Int. J. Food Microbiol. 51, 133–143. Durán Quintana, M.C., García García, P., Garrido Fernández, A., 2003. Características del crecimiento de levaduras de aceitunas de mese a bajas temperaturas. Grasas Aceites 54, 264–271. EFSA, 2007. Qualified presumption of safety. In “The EFSA Journal”. Appendix A: Assessment of gram positive non sporuling bacteria. 587. pp. 4–6.
376
SECTION | I Bacterial and Fungal and Other Microbial Aspects
Ercolini, D., Villani, F., Aponte, M., Mauriello, G., 2006. Fluorescence in situ hybridization detection of Lactobacillus plantarum group on olives to be used in natural fermentations. Int. J. Food Microbiol. 112, 291–296. Fleming, H.P., Mc Feeters, R.F., Daeschel, M.A., 1984. The Lactobacilli, Pediococci and Leuconostoc: vegetable products. In: Gilliland, S.E. (ed.), Bacterial Starter Cultures for Foods. CRC Press Inc., Boca Raton, Florida, pp. 97–114. Franz, C.M.A.P., Schillinger, U., Holzapfel, W.H., 1996. Production and characterization of enterocin 900, a bacteriocin produced by Enterococcus faecium BFE 900 from black olives. Int. J. Food Microbiol. 29, 255–270. Garcia-Garcia, P., Romero Barranco, C., Duran Quintana, M.C., Garrido-Fernandez, A., 2004. Biogenic amine formation and “zapatera” spoilage of fermented green olives: effect of storage temperature and debittering process. J. Food Prot. 67, 117–123. Garrido Fernández, A., García García, P., Brenes Balbuena, M., 1995. Olive fermentation. In: Rehm & Reed (ed.), Biotechnology: Enzymes, “Biomass, Food and Feed”. VCH, New York, pp. 593–627. Garrido Fernández, A., Díez, M.J., Adams, M.R., 1997. In: Chapman & Hall (eds), Table Olives: Production and Processing. London, UK, pp. 134–197. Giudici, P., Strano, M.C., Pulvirenti, A., 1997. La sicurezza microbiologica nella preparazione casalinga delle olive verdi al naturale. Tecnica Agric. 1, 11–19. Hernández, A., Martín, A., Aranda, E., Pérez-Nevado, F., Córdoba, M.G., 2006. Identification and characterization of yeast isolated from the elaboration of seasoned green table olives. Food Microbiol. 24, 346–351. Holzapfel, W.H., 1997. Use of starter cultures in fermentation on a household scale. Food Control. 8, 241–258. Jimenez-Diaz, R., Rios-Sanchez, R.M., Desmaseaud, M., Ruiz-Barba, J.L., Piard, J.C., 1993. Plantaricins S and T, two new bacteriocins produced by Lactobacillus plantarum LPCO10 isolated from a green olive fermentation. Appl. Environ. Microbiol. 59, 1416–1424. Lavermicocca, P., Gobbetti, M., Corsetti, A., Caputo, L., 1998. Characterization of lactic acid bacteria isolated from olive phyllophane and table olive brines. Ital. J. Food. Sci. 10, 27–39. Lavermicocca, P., Valerio, F., Lonigro, S.L., De Angelis, M., Morelli, L., Callegari, M.L., Rizzello, C.G., Visconti, A., 2005. Study of adhesion and survival of lactobacilli and bifidobacteria on table olives with the aim of formulating a new probiotic food. Appl. Environ. Microbiol. 71, 4223–4240. Leal Sánchez, M.V., Baras, M., Ruiz-Barba, J.L., Floriano, B., JiménezDíaz, R., 1998. Bacteriocin production and competitiveness of Lactobacillus plantarum LPO10 in olive juice broth, a culture medium obtained from olives. Int. J. Food Microbiol. 43, 129–134. Leal-Sánchez, M.V., Ruiz-Barba, J.L., Sánchez, A.H., Rejano, L., Jiménez-Díaz, R., Garrido, A., 2003. Fermentation profile and optimization of green olive fermentation using Lactobacillus plantarum LPO10 as a starter culture. Food Microbiol. 20, 421–430. Maldonado, A., Ruiz-Barba, J.L., Floriano, B., Jimenez-Diaz, R., 2002. The locus responsible for production of plantaricin S, a class IIb bacteriocin produced by Lactobacillus plantarum LPO10, is widely distributed among wild-type Lact. plantarum strains isolated from olive fermentations. Int. J. Food Microbiol. 77, 117–124. Marquina, D., Peres, C., Caldas, F.V., Marques, J.F., Peinado, J.M., Spencer-Martins, I., 1992. Characterization of the yeast population in olive brines. Lett. Appl. Microbiol. 14, 279–283. Marsilio, V., Lanza, B., 1998. Characterization of an oleuropein degrading strain of Lactobacillus plantarum. Combined effects of compounds present in olive fermenting brines (phenol, glucose and NaCl) on bacterial activity. J. Sci. Food Agric. 76, 520–524.
Montaño, A., Sánchez, A.H., de Castro, A., 1993. Controlled fermentation of Spanish-type green olives. J. Food Sci. 58, 842–844. Panagou, E.Z., Tassou, C.C., Katsaboxakis, C.Z., 2003. Induced lactic acid fermentation of untreated green olives of the Conservolea cultivar by Lactobacillus pentosus. J. Sci. Food Agric. 83, 667–674. Panagou, E.Z., Katsaboxakis, C.Z., 2006. Effect of different brining treatments on the fermentation of cv. Conservolea green olives processed by the Spanish method. Food Microbiol. 23, 199–204. Panagou, E.Z., Tassou, C.C., 2006. Changes in volatile compounds and related biochemical profile during controlled fermentation of cv. Conservolea green olives. Food Microbiol. 23, 738–746. Panagou, E.Z., Schillinger, U., Franz, C.M.A.P., Nychas, G.J.E., 2008. Microbiological and biochemical profile of cv. Conservolea naturally black olives during controlled fermentation with selected strains of lactic acid bacteria. Food Microbiol. 25, 348–358. Randazzo, C.L., Caggia, C., Restuccia, C., Giudici, P., Torriani, S., 1999. Valutazione della biodiversità di ceppi di batteri lattici isolati da olive siciliane fermentate al naturale. Tecnica. Agric. 2, 3–9. Randazzo, C.L., Restuccia, C., Romano, D.A., Caggia, C., 2004. Lactobacillus casei, dominant species in naturally fermented Sicilian green olives. Int. J. Food Microbiol. 90, 9–14. Ross, R.P., Morgan, S., Hill, C., 2002. Preservation and fermentation: past, present and future. Int. J. Food Microbiol. 79, 3–16. Ruiz-Barba, J.L., Brenes Balbuera, M., Jiménez-Diaz, R., García García, P., Garrido-Fernández, R., 1993. Inhibition of Lactobacillus plantarum by polyphenols extracted from two different kinds of olive brine. J. Appl. Bacteriol. 74, 15–19. Ruiz-Barba, J.L., Cathcart, D.P., Warner, P.J., Jimenez-Diaz, R., Piard, J.C., 1994. Use of Lactobacillus plantarum LPCO10, a bacteriocin produces as starter culture in Spanish-style green olive fermentations. Appl. Environ. Microbiol. 60, 2059–2064. Sánchez, A.H., de Castro, A., Rejano, L., Montaño, A., 2000. Comparative study on chemical changes in olive juice and brine during green olive fermentation. J. Agric. Food Chem. 48, 5975–5980. Sánchez, A.H., Rejano, L., Montaño, A., de Castro, A., 2001. Utilization at high pH of starter cultures of lactobacilli for Spanish-style green olive fermentation. Int. J. Food Microbiol. 67, 115–122. Settani, L., Corsetti, A., 2008. Application of bacteriocins in vegetable food biopreservation. Int. J. Food Microbiol. 121, 123–138. Servili, M., Settanni, L., Veneziani, G., Esposto, S., Massitti, O., Taticchi, A., Urbani, S., Montedoro, G.F., Corsetti, A., 2006. The use of Lactobacillus pentosus 1MO to shorten the debittering process time of black olives (cv. Itrana and Leccino): a pilot-scale application. J. Agric. Food Chem. 54, 3869–3875. Spyropoulou, K.E., Chorianopoulos, P.N., Nychas, G.J.E., 2001. Survival of Escherichia coli O157:H7 during the fermentation of Spanish-style green table olives (conservolea variety) supplemented with different carbon sources. Int. J. Food Microbiol. 66, 3–11. Tassou, C.C., 1993. Microbiology of olives with emphasis on the antimicrobial activity of phenolic compounds. PhD Thesis, University of Bath, UK. Tassou, C.C., Panagou, E.Z., Katsaboxakis, K.Z., 2002. Microbiological and physiochemical changes of naturally black olives fermented at different temperatures and NaCl levels in the brines. Food Microbiol. 19, 605–615. Todorov, S.D., Dicks, L.M.T., 2005. Characterization of bacteriocins produced by lactic acid bacteria isolated from spoiled black olives. J. Basic Microbiol. 45, 312–322. Van den Berg, D.J.C., Smits, A., Pot, B., Ledeboer, A.M., Kersters, K., Verbakel, J.M.A., Verrips, C.T., 1993. Isolation, screening and identification of lactic acid bacteria from traditional food fermentation processes and culture collections. Food Biotechnol. 7, 189–205.
Chapter 42
Understanding and Optimizing the Microbial Degradation of Olive Oil: A Case Study with the Thermophilic Bacterium Geobacillus thermoleovorans IHI-91 Peter Becker Novo Nordisk A/S, BioProcess Technologies, Måløv, Denmark
42.2 MICROBIAL LIPASES: AN OVERVIEW
The microbial degradation of triglycerides from natural fats and oils is a complex process. Triglycerides, which are insoluble in water, are hydrolyzed by extracellular lipases to yield long-chain fatty acids and glycerol. After active or passive transportation into the cell (DiRusso and Black, 1999), the fatty acids are further degraded via the β-oxidation pathway, yielding acetyl-CoA, which in turn passes through the TCA cycle and respiratory chain for energy generation. Glycerol plays energetically an insignificant role in the overall balance for triglyceride degradation and contributes only with approximately 5% to the overall ATP generation. Aerobic growth on olive oil can be described by the following stoichiometric equation: aCH1.83O0.11 ⫹ bO2 ⫹ cNH3 → CH1.8O0.5 N 0.2 ⫹ dCO2 ⫹ eH 2O
1 Esterase Lipase
(42.1)
where CH1.83O0.11 is the elemental composition of olive oil and CH1.8O0.5N0.2 is a general elemental composition of microbial biomass. The stoichiometric coefficients were determined as a ⫽ 1.37, b ⫽ 0.87, c ⫽ 0.2; d ⫽ 0.37 and e ⫽ 0.50 for an exponentially growing culture of the strain G. thermoleovorans IHI-91 with olive oil as the only carbon source (Becker and Märkl, 2000). Thus, one gram of olive oil yields approximately 1.2 grams of dry biomass. In comparison, glucose yields only approximately 0.5 g g⫺1, highlighting the extremely high energy content of lipidic growth substrates. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Microbial lipases play an increasingly important role in many areas of modern biotechnology. Lipases (glycerol ester hydrolases; EC 3.1.1.3) are hydrolases acting on the carboxyl ester bonds in acylglycerols, liberating organic acids and glycerol. Lipases display the unique feature of acting at the interface between an aqueous and a nonaqueous (i.e. organic) phase; this feature distinguishes them from esterases (see Figure 42.1). The concept of interfacial activation arises from the fact that the catalytic activity of the lipase generally depends
Enzyme activity
42.1 INTRODUCTION TO THE BIOCHEMISTRY OF TRIGLYCERIDE DEGRADATION
Substrate saturation 0 0 Substrate concentration FIGURE 42.1 Schematic representation of the concept of interfacial activation. The figure depicts the general kinetic difference between esterases and lipases. The vertical horizontal line indicates the saturation point for the substrate. The lipase is activated upon appearance of a lipid–water interface above the saturation point of the substrate.
377
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
378
SECTION | I Bacterial and Fungal and Other Microbial Aspects
on the physical state of the substrate (see Figure 42.1). Generally, activation involves the liberation of the enzyme’s active site through conformational changes (opening of a ‘lid’), requiring the presence of an oil-in-water interface. A useful kinetic model describes the lipolytic reaction as a two-step process (Verger and de Haas, 1976): 1. adsorption of the enzyme at the lipid–water interface and activation of the lipase by a confirmational change (opening of the lid to liberate the active site) 2. formation of the enzyme/substrate complex and hydrolysis to yield the product and the enzymes for regeneration. Michaelis–Menten kinetics can be used to describe the latter step, when applying the units of (mol/surface) instead of (mol/volume). The rate of lipolysis is generally expected to be dependent on most or all of the following parameters: the interfacial area (droplet size of the emulsion or micelles), the presence of emulsifiers, general mass transfer properties and other physicochemical parameters such as pH, temperature and viscosity. The generalized feature of lipase interfacial activation was critically discussed by Verger (1997).
42.3 MICROBIAL SOURCES OF LIPASES Lipases are ubiquitously present in the microbial world, and most lipases applied in current biotechnological applications are derived from fungal or bacterial sources (Jaeger and Eggert, 2002). They are mass produced either from wild-type strains or through modern overexpression technology in hosts like Escherichia coli, Bacillus subtilis and Pichia pastoris. Some of the most widely used fungal lipases are derived from various species within genera such as Candida, Yarrowia, Aspergillus and Penicillium, while bacterial lipases often come from Pseudomonas sp., Bacillus sp., Staphylococcus sp., Burkholderia sp. and many others (Gupta et al., 2004). Bacterial lipases and esterases have been classified into eight families (and several subfamilies) based on sequence homology and biological properties such as the usage of specific secretion machinery and foldases (Arpigny and Jaeger, 1999). The bacterial isolate Geobacillus thermoleovorans IHI-91 is part of a taxonomically diverse group of thermophilic bacteria that display high lipolytic activity. It was originally described as Bacillus thermoleovorans IHI-91 (Markossian et al., 2000) and was later renamed to Geobacillus thermoleovorans IHI-91. Both lipase (AAN72417) and esterase (AAG53982) of G. thermoleovorans IHI-91 were cloned, purified and characterized. Its lipase belongs to lipase family I.5 of the classification system (Arpigny and Jaeger, 1999). Table 42.1 gives an overview of some physiological properties of various thermophilic, lipolytic Bacillus isolates.
42.4 LIPASE PRODUCTION: INDUCTION AND SECRETION Gupta et al. (2004) reviewed some of the important factors governing lipase production in bacterial systems. While it seems to be a universal feature that lipase production requires the presence of an inducer, the nature of the inducer can vary. Thus, oils, triglycerides, long-chain fatty acids, TWEENs, hydrolyzable esters, n-alkanes, bile salts and glycerol have all been reported as lipase inducers (see Gupta et al., 2004 and Table 42.1). For G. thermoleovorans IHI-91, olive oil was found to be a very potent inducer, while glycerol was amongst the carbon sources that have no inducing potential (see Figure 42.2). Apart from induction, lipase production is also regulated by repression. Studies with G. thermoleovorans IHI91 on mixed substrates revealed that glucose and glycerol are potent repressors, which down-regulate lipase production even in the presence of olive oil; glycerol though only at concentrations ⬎2 g L⫺1 (not shown). On mixed substrates, lipase-producing strains will therefore often display the classical diauxic behavior, metabolizing the most easily assimilated substrate first. Long-chain fatty acids have been reported to display both inducing and repressing capability, which seems to be species-dependent (Hsu et al., 1983). Inducers often serve as carbon sources during fermentation, coupling lipase production directly to biomass formation. Commercially interesting processes for cultivating microorganisms are fed-batch and continuous fermentations. These processes allow for generation of high cell densities under defined conditions for growth rate and substrate concentration. Geobacillus thermoleovorans IHI91 was extensively studied in chemostat (⫽ continuous) cultivation (Becker et al., 1997; Becker and Märkl, 2000) allowing to elucidate the factors determining lipase production. One such factor was the feed substrate concentration, where olive oil was used as the sole carbon and energy source. As shown in Figure 42.3, the maximal lipase activity was observed at a feed concentration of around 2 g L⫺1 of olive oil. This graph also underlines another feature of lipases: a large proportion of the total lipase activity can be located in the insoluble fraction, i.e. in cell-bound or aggregated form and/or attached to (partially) hydrolyzed substrate. Bacterial lipases are extracellular enzymes and must therefore be translocated through the cell wall of the bacterial cell. Some of the involved secretion mechanisms have been summarized by Jaeger and Eggert (2002). In Grampositive bacteria, which only display a single membrane, the Sec machinery is widely used, while the Tat-pathway has been described for both Gram-positive and Gramnegative bacteria. The secretion and folding process can be very complex, with some Pseudomonas strains requiring about 30 different proteins in order to functionally fold and secrete lipase into the medium. Since the requirements
Strain/isolate
Growth temperature
pH
Growth rate (h⫺1)
Complex substrates in media
Substrate/ Inducer
Repression by
References
n.d.
n.d.
0.1% Nutrient broth
1–1.5% Tween 80
Glucose Glycerol
Gowland et al. (1987)
n.d.
0.83a
0.4% Casamino acids 0.1% Nutrient broth
1% Tween 80
n.d.
Handelsmann and Shoham (1994)
Topt
Tmin
Tmax
Bacillus sp. CT42
65°C
40–45°C
70–73°C
Bacillus sp. H1
60–65°C
Bacillus sp. MC-7
62°C
45°C
72°C
pHopt 9.2
0.5b
0.05% Starch 0.2% Bacto casitone
0.5 ml/l Tween 80
Glucose
Kambourova et al. (1996)
Bacillus sp. A30-1
60°C
45°C
75°C
5.7–9.0
0.25–0.35c
0.1% Yeast extract
1% Corn oil 0.1% Tween 80
n.d.
Wang et al. (1995)
Bacillus sp. 393
55°C
40–45°C
65°C
n.d.
n.d.
Peptone, yeast extract, meat extract
1% Olive oil
n.d.
Kim et al. (1994)
Bacillus sp. Wai28A5
70°C
7.5
0.128
none
0.4 g/l Tripalmitin
n.d.
Janssen et al. (1994)
Bacillus thermoleovorans IHI-91
65°C
5–7.5
1.25d 2.7e
none
2 g/l Olive oil
Glucose Glycerol ⬎2g/l
Becker et al. (1997) Markossian et al. (2000)
37–50°C
75°C
CHAPTER | 42 Understanding and Optimizing the Microbial Degradation of Olive Oil
TABLE 42.1 Physiological properties of various lipolytic thermophilic Bacillus sp. isolates.
Topt , Tmin , Tmax refer to the optimal, minimal and maximal growth temperatures, respectively. a
data for 60°C;
b
estimated from growth curve at 55°C and pH 8.5 yeast extract;
c
estimated from growth curve at 60°C and start pH 9.0;
d e
T ⫽ 65°C and pH ⫽ 6.0, C-source olive oil;
T ⫽ 65°C and pH ⫽ 6.0, C-source;
n.d. no data.
379
380
SECTION | I Bacterial and Fungal and Other Microbial Aspects
4
transport limitation at higher dilution rates during chemostat cultivations of the yeast Candida rugosa on oleic acid as substrate.
Olive oil Glycerol Glucose Yeast extract
Lipase activity (U mL−1)
3
42.5 IDENTIFICATION OF RATE-LIMITING STEP
2
1
0 0
1
2
3
4
5
6
7
Cell density (109 cells mL−1) FIGURE 42.2 Measurement of the lipase-inducing property of various growth substrates. Lipase activity was measured for growth of G. thermoleovorans IHI-91 on various substrates. Olive oil leads to a significant induction of the lipase, while glucose, glycerol and yeast extract do not induce the lipase production.
6
Cell suspension Cell-free supernatant
Lipase activity (U mL−1)
5
4
3
2
1
0
0
1 2 3 4 5 6 Olive oil concentration in medium feed sF (g L−1)
7
FIGURE 42.3 Optimization of lipase production in a continuous fermentation process. Impact of the olive oil feed concentration on lipase production during chemostat cultivation of G. thermoleovorans IHI-91 at a dilution rate of 0.2 h⫺1. The lipase measurements were performed on cell suspension and cell-free supernatant.
for correct folding and translocation of lipases can be huge, many of the known lipases can only be produced in sufficient amounts from wild-type strain cultivations or in homologous hosts. Montesinos et al. (2003) found lipase
As schematically drawn in Figure 42.4, the process of lipid degradation may be limited at different levels due to: (a) the kinetics of lipid hydrolysis, (b) mass transfer limitation effects between the two liquid phases (or even between gas and aqueous phase, if oxygen transfer is disturbed through the presence of lipids), (c) uptake of and inhibition by long-chain fatty acids and (d) production, secretion and activation of the lipolytic enzymes. Also, the physicochemical properties of the primary hydrolysis products, namely long-chain fatty acids and their respective salts (chain length, degree of unsaturation, solubility, particle size) have been found to be an important factor (Loehr and Roth, 1968). The mass transfer situation is often governed by the power input into the system (e.g. through the stirrer speed in a reactor), giving rise to a specific interfacial area between oil and water phase. Mass transfer can be substantially improved, e.g. by addition of surfactants (Sekelsky and Shreve, 1999), by chemical or enzymatic pre-hydrolysis of fats (Masse et al., 2001) or by increasing process temperatures using thermophilic microorganisms (Becker et al., 1999). In the latter case, the major parameters that govern transport phenomena such as viscosity, diffusion rates, surface tension, water solubility of triglycerides and free fatty acids and the melting points of these compounds, are all positively affected by the temperature rise. A large variety of thermophilic, lipolytic microorganisms has recently become available (Gowland et al., 1987; Handelsmann and Shoham, 1994; Janssen et al., 1994; Kim et al., 1994; Wang et al., 1995; Kambourova et al., 1996), as shown in Table 42.1, but investigations on the kinetics of thermophilic oil utilization have been very rare (Dalmau et al., 1998; Becker and Märkl, 2000).
42.6 IMPACT OF OIL DROPLET SIZE The presence of a water–lipid interface is not only crucial for activation of the lipases; the total interfacial area will furthermore determine the rate of the lipolytic reaction, if no other limitation is present. However, also lipase production itself can be influenced by the interfacial area. Dalmau et al. (1998) found that lipase production by the yeast Candida rugosa was inversely proportional to the specific interfacial area, i.e. the biggest droplets led to the highest lipase production of the cells. Our own studies indicate a less pronounced effect of olive oil droplet size on lipase formation and growth kinetics of the thermophilic bacterium
CHAPTER | 42 Understanding and Optimizing the Microbial Degradation of Olive Oil
Lipid phase • [substrate] • [metabolites] • [biosurfactant] • CMC • interfacial area
381
Fatty acid uptake • active (protein-mediated) • passive (diffusional)
Mass transfer • stirrer speed • viscosity • surface tension
Growth kinetics • temperature, pH • inhibition
Lipase production • induction • (de-)repression • secretion • foldases
Lipolysis • interfacial activation • lipase activity • temperature, pH • inhibition • lipase half-life Micelle
Cell wall
FIGURE 42.4 Schematic overview of the cellular and biochemical processes and some important physicochemical parameters involved in the microbial degradation of natural oils. The ‘micelle’ represents the substrate in an aggregated state, i.e. in the form of an adsorbed monolayer, a micelle or an emulsion according to Verger (1997). The substrate is also present in a non-aggregated form as a ‘monomer’, below its critical micelle concentration (CMC). The ‘cell wall’ represents the transport barrier for transport of substrate into the cell and translocation of lipase out of the cell and is more complex in its structural composition than drawn. Note that Gram-negative bacteria, for example, have two distinct membranes and a periplasmic space in between, which is crucial for correct folding of the lipase (Jaeger and Eggert, 2002).
Geobacillus thermoleovorans IHI-91, indicating differences in the respective mechanisms for lipase production by yeast and bacteria (Figure 42.5). Whereas Dalmau et al. worked with oleic acid as the best lipase inducer for C. rugosa, olive oil was chosen as the inducer for G. thermoleovorans. The insignificant effect of stirrer speed on the maximum growth rate suggests that even the lowest lipase activity measured was sufficient to allow for maximal bacterial growth rate. This in turn indicates that lipid hydrolysis does not constitute a bottleneck for bacterial growth under the given conditions at a growth temperature of 65°C (see Figure 42.5). In a stirred-tank bioreactor, the stirrer speed (power input) is the single most important parameter, determining the size of the resulting interfacial area between the lipid and water phase. However, during microbial growth, the cells might actively produce biosurfactants, and/or some of the degradation products might act as emulsifiers, thereby influencing the particle size distribution tremendously. Thus, for a hexadecane degrading Pseudomonas aeruginosa strain, Sekelsky and Shreve (1999) describe a kinetic model that includes the influence of a biosurfactant (rhamnolipid) on the dissolution of the hydrophobic phase, on cellular uptake and thus on microbial growth kinetics. For long-chain fatty acids, the primary product of olive oil hydrolysis, both passive (diffusional) and active (proteinmediated) uptake mechanisms are described (DiRusso and Black, 1999).
42.7 HYDROLYSIS VS OXIDATION OF FATTY ACIDS Since the rate of triglyceride hydrolysis is ruled by the lipase activity, it is worth looking first at some of the factors influencing lipase activity. Temperature and pH are often considered the predominant factors influencing enzymatic activities, and the optimal conditions for enzyme activity can differ considerably from the optimal growth conditions. Every enzyme undergoes inactivation. The rate of inactivation is largely dependent on temperature, pH and some of the components in the surrounding medium. At 65°C, the growth temperature optimum for G. thermoleovorans IHI91, the half-life of the lipolytic activity was ⬍2 h (Figure 42.6). Since the inactivated lipase has to be replaced during growth, the short half-life indicates a high lipase-producing capacity of the strain. In fact, a maximum productivity of almost 900 U L⫺1 h⫺1 was found in an optimized continuous cultivation process (dilution rate 0.3 h⫺1, olive oil concentration 2.0 g L⫺1). Furthermore, CaCl2 was found to stabilize the lipolytic enzymes. It has also been reported that the presence of a competitive inhibitor, such as oleic acid, improves lipase stability by binding to the active site and reducing the denaturation rate of the enzyme (Tsai and Chiang, 1991). The hydrolysis of olive oil by the thermophilic bacterium G. thermoleovorans IHI-91 was studied in a substratelimited batch cultivation with olive oil as the only carbon
382
SECTION | I Bacterial and Fungal and Other Microbial Aspects
100
Stirrer speed 700 rpm 1500 rpm 2000 rpm 2500 rpm
60°C 80
50 Activity (%)
q3 -Distribution (μm−1)
75
25
65°C
60
75°C (1 mM CaCl2)
40
20 75°C
0 1 A
10 Drop radius (μm)
2
1,0
Lipase activity (U/mL)
0
350
Triolein Oleic acid
300
Diolein
1
0,5
0 0
1000 2000 Stirrer speed (rpm)
0 3000
FIGURE 42.5 Impact of mass transfer on growth and lipase production of Geobacillus thermoleovorans IHI-91 in batch culture on olive oil. (A) Droplet size distribution of olive oil (2 g L⫺1) dispersed in a laboratory stirred tank reactor was measured at the start of each fermentation. Coagulation of the olive oil droplets was prevented by using a fast-acting emulsifier (L01, a polyglycolether from Bayer AG, Germany) during sampling. The droplet size density distribution was measured using a laser diffraction spectrometer (Model Helos LA, Sympatec GmbH, Germany). (B) Maximum growth rate and maximum lipase activity of the strain G. thermoleovorans IHI-91 grown at various stirrer speeds, kept constant throughout the fermentation. Dissolved oxygen was controlled at ⬎20% air saturation using oxygen enriched air.
and energy source. As shown in Figure 42.7, the gradual decrease of triolein was accompanied by a transient accumulation especially of oleic acid and the diolein species. Cell growth was exponential throughout most of the cultivation, with a maximum specific growth rate of 1.17 h⫺1 (not shown).
2
1 Duration of incubation (h)
FIGURE 42.6 Stability of the lipolytic enzymes of G. thermoleovorans IHI-91. The residual lipase activity was measured after incubation at 60°C, 65°C and 75°C. Addition of 1 mM CaCl2 stabilizes the lipase activity markedly. Activities were measured in partially purified lipase extract using the substrate p-nitrophenyllaurate.
Concentration (mg L−1)
Maximum growth rate μmax (h−1)
0 3
1,5
B
100
Monoolein 250
200
150
100
50
0
1
2 3 4 Cultivation time (h)
5
FIGURE 42.7 Time course of olive oil hydrolysis during growth of G. thermoleovorans IHI-91. The concentrations of triolein, diolein species, monoolein species and oleic acid were measured during batch growth on olive oil as the only carbon and energy source at 65°C.
The accumulation of oleic acid suggests that the growthlimiting factor for this thermophilic bacterium is the oxidation of the long-chain fatty acids rather than the hydrolysis reaction. Hsu et al. (1983) studied the kinetics of olive oil degradation by activated sludge and concluded that the
383
CHAPTER | 42 Understanding and Optimizing the Microbial Degradation of Olive Oil
overall rate of degradation was dependent on the hydrolysis reaction at low olive oil concentrations and the fatty acid oxidation at high olive oil concentrations (⬎800 mg L⫺1).
Prior to the work with G. thermoleovorans IHI-91, it was difficult to find investigations describing olive oil consumption (or degradation) kinetics. Hsu et al. (1983) reported Michaelis–Menten type kinetics for the consumption of olive oil by activated sludge. The lipid removal from the suspension showed the following kinetic constants: maximum lipid removal rate ⫽ 1.20 mg mg⫺1 biomass day⫺1 and a half-saturation constant Km ⫽ 1290 mg L⫺1. The early studies with Bacillus thermoleovorans IHI-91 revealed ‘global’ kinetic parameters according to the Monod equation (Eq. 42.3) as: μmax ⫽ 1.0 h⫺1, Ks ⫽ 880 mg L⫺1 (Becker et al., 1997). Both studies show that the assumed substrate half-saturation constants are extremely high compared to non-lipidic substrates such as glucose, where values ⬍100 mg L⫺1 are commonly observed. Generally, the higher the half-saturation constant, the lower is the affinity to a given substrate. An explanation for these apparent high values can be seen in the attempt to combine several steps, such as lipase production, lipid hydrolysis, fatty acid uptake and microbial growth in one ‘global’ kinetic equation. As later studies revealed (see next paragraph), the Ks value for oleic acid was determined as 35 mg L⫺1, thus being more in the ‘normal’ range for microbial substrates.
42.9 GROWTH INHIBITION BY LONG-CHAIN FATTY ACIDS A growth-inhibitory action and cytotoxic effect of oleic acid has been reported widely in the literature. Fatty acids of varying chain lengths are generally known for their antimicrobial action, primarily against Gram-positive bacteria and yeasts at low pH. The observed inhibition is explained as a consequence of the passive uptake of the undissociated form of the fatty acid, which disturbs the transmembrane proton gradient and thereby affects ATPase activity. Also active uptake mechanisms have been described for many microbes. In Escherichia coli, the protein-mediated uptake mechanism consists of the outer-membrane-bound fatty acid transporter protein FadL and the innermembrane-bound fatty acyl CoA synthetase (FACS) (DiRusso and Black, 1999). In most lipid-rich wastewaters a large proportion of the total lipid contamination is hydrolyzed before entering the biological treatment, oleic acid often being the predominant long-chain fatty acid, rendering olive mill effluent with its high level of oleic acid as particularly problematic.
Chemostat experiments: Substrate olive oil sF = 2.0 g L−1, D variable
1.25 Specific growth rate μ (h−1)
42.8 MODELING KINETICS FOR BACTERIAL GROWTH ON OLIVE OIL
1.50
D = 0.2 h−1, sF variable D = 0.3 h−1, sF variable
1.00
0.75
0.50
100% inhibition
0.25
0
0
100
200 300 400 500 Oleic acid concentration (g L−1)
600
FIGURE 42.8 Oleic acid inhibition kinetics of G. thermoleovorans IHI-91. The graph shows various steady-state conditions from chemostat cultivations. Data as previously published (Becker and Märkl, 2000). Reprinted with permission from John Wiley & Sons Inc.
For G. thermoleovorans IHI-91 it was observed that increasing olive oil start concentrations in batch fermentations led to a saturation of the maximum observed cell density, although the medium components were not growth-limiting. Moreover, the spiking of oleic acid into running fermentations on glycerol led to a significant growth retardation already at a concentration of 50 mg L⫺1, while a spike of 1 g L⫺1 led to an immediate and irreversible stop of cell growth (not shown). Cells grown on olive oil are adapted to higher levels of oleic acid and kinetic measurements showed that they are completely inhibited at a concentration of 430 mg L⫺1 (see Figure 42.8 and Eq. 42.2). From the measurement of the steady-state oleic acid concentrations during various chemostat cultivations (as shown in Figure 42.8), it was possible to derive a kinetic expression for oleic acid utilization according to the following equation: μ ⫽ μmax ⭈
⎛ c(oleic acid ) c(oleic acid ) ⎞⎟ ⎟⎟ ⭈⎜⎜⎜1 ⫺ c(oleic acid ) ⫹ K S ⎜⎝ cmax (oleic acid) ⎟⎠ (42.2)
where μ and μmax represent the specific and maximum specific growth rate of G. thermoleovorans IHI-91 for growth on oleic acid, Ks represents the Monod half-saturation constant for oleic acid, c the concentration of oleic acid, and cmax the oleic acid concentration for which the growth rate becomes zero due to substrate inhibition. The experimental data were fitted with the following parameters: Ks ⫽ 35 mg L⫺1, cmax(oleic acid) ⫽ 430 mg L⫺1 and μmax ⫽ 2.1 h⫺1.
SECTION | I Bacterial and Fungal and Other Microbial Aspects
Recent investigations showed that oleic acid inhibition furthermore is dependent on the cell concentration and on the interfacial area of the dispersed oleic acid phase. Thus, cultures with higher biomass concentrations can tolerate higher concentrations of oleic acid, whereas a higher stirrer speed, i.e. a larger interfacial area, can worsen the inhibitory effect (Reimann, 2003).
42.10 MODEL PREDICTIONS The extensive studies of the growth kinetics of G. thermoleovorans IHI-91 on olive oil led to mathematical models that can be used to predict the system’s behavior. Especially two features were of special attention, namely 1. bi-stability due to the inhibition kinetics 2. oscillatory behavior due to coupling of the inhibition kinetics with the lipase production. The mathematical models dealt with in this chapter were described elsewhere (Becker and Märkl, 2000). As a consequence of Equation 42.2, there can be two stable steadystate conditions at each growth rate. One condition would have a high biomass concentration and a low product concentration, the other state would have a low biomass concentration (or complete washout of the cells) and a high residual substrate concentration in the chemostat cultivation. The simulations (see Figure 42.9) of the chemostat behavior, assuming oleic acid to be the growth substrate, were done according to model A (Becker and Märkl, 2000), using the inhibitory kinetic expression for oleic acid (Eq. 42.2). Oscillatory behavior was observed in experimental chemostat cultivations on olive oil, especially after a large step increase in olive oil concentration in the feed, sF. A decrease in sF always led back to stable operating conditions. Chemostat simulations with model C (Becker and Märkl, 2000) revealed that the reason for these oscillations can be seen in the rapid accumulation of oleic acid upon a step increase in the feed olive oil concentration (Figure 42.10). This is followed by a severe growth inhibition with biomass and, therefore, also lipase washout. Oleic acid levels drop again after most of the lipase has been washed out and biomass growth is resumed. These simulations support the conclusion that the thermophilic lipid degradation by G. thermoleovorans IHI-91 is limited by the capacity of the β-oxidation pathway to remove the long-chain fatty acids, generated during olive oil hydrolysis. The simulations underline that process stability is an important issue in lipid biodegradation processes, especially with fluctuating lipid concentrations and/or high concentrations of inhibitory long-chain fatty acids. In the case of oleic acid as the growth substrate, simulations show that a complete biomass washout can be expected if a critical oleic acid concentration, depending on the biomass concentration and the dilution rate, is exceeded.
1.2
1.0 Substrate concentration s (g L−1)
384
Substrate oleic acid sF = 1.0 g L−1 D = 0.5 h−1
0.8
0.6
0.4
0.2
0
0
0.3 0.6 0.9 1.2 Biomass concentration x (g L−1)
1.5
FIGURE 42.9 Model prediction of the bi-stability during the continuous cultivation of G. thermoleovorans IHI-91 on oleic acid. Prediction according to model A in Becker and Märkl (2000). The graph shows bi-stability of the system and its transient behavior in form of a substrate-biomass phase diagram (s–x phase diagram) for substrate feed concentration sF ⫽ 1 g L⫺1 and dilution rate D ⫽ 0.5 h⫺1. Arrows indicate the dynamics of the system before reaching steady state. Two stable steady states are reached: the washout point at x ⫽ 0 and a stable operating point at low s and high x.
42.11 APPLICATION OF LIPOLYTIC MICROBES AND LIPASES IN WASTEWATER TREATMENT High lipid loads in wastewaters create a wide range of problems including the production of unpleasant odors, the blockage of sewer lines and disturbance of the proper operation of treatment facilities. Removal of fats, oils and fatty acids from wastewater is thus critically important to ensure that wastewater can be treated efficiently and economically. Two strategies have recently been intensively studied and commercialized to overcome these problems, namely 1. bioaugmentation, i.e. the use of microbial supplements containing highly lipolytic species (Loperenaa et al., 2006; Brooksbank et al., 2007) and 2. direct enzymatic pre-treatment using lipases (Masse et al., 2001; Cammarotaa and Freire, 2006). A commercial multi-species supplement, named F69, was capable of significantly enhancing the degradation of several fats and oils by 37–62% (Brooksbank et al., 2007). The supplement displayed a broad spectrum of fatty acid utilization, while wastewater-associated bacteria preferentially degraded linolenic acid (18:3ω3) and linoleic acid (18:2ω6). Loperenaa et al. (2006) studied the kinetics of
CHAPTER | 42 Understanding and Optimizing the Microbial Degradation of Olive Oil
As an alternative to the use of microbial supplements, enzymes can be employed in wastewater pre-treatment operations. Using, for example, pancreatic lipase 250 and Rhizomucor miehi lipase G-1000, pre-treatment has been demonstrated to efficiently decrease the average particle size of animal fat particles in slaughterhouse wastewater and thereby accelerate the following degradation process (Masse et al., 2001). Cammarotaa and Freire (2006) suggest an enzymatic pre-treatment using an extract from the fungus Penicillium restrictum, while other fungal species of interest include Candida rugosa.
5
Concentration (g L−1)
4 Total fat
3
2
Oleic acid
1
Biomass (g L−1), Lipase (U mL−1)
0
SUMMARY POINTS Lipase activity Biomass concentration
2,5
●
2,0 1,5 ●
1,0 0,5 0
385
0
20
40
60 Time (h)
80
100
120
FIGURE 42.10 Model prediction of oscillations during the continuous cultivation of Geobacillus thermoleovorans IHI-91. Prediction according to model C in Becker and Märkl (2000). From a steady state at t ⫽ 20 h, the olive oil feed concentration is increased from 0.5 g L⫺1 to 5 g L⫺1 (at a constant dilution rate of 0.5 h⫺1).
butter oil degradation using two different inocula, a commercial product (Sybron Bi-Chem1003FG) and a ‘native’ microbial consortium derived from activated sludge from a dairy wastewater treatment plant. The commercial inoculum exhibited inhibition kinetics, while the ‘native’ inoculum followed simple Monod kinetics according to the following equation: μ ⫽ μmax ⭈
s s ⫹ KS
●
●
●
●
The microbial degradation of oil and fats comprises a combination of complex biological and physicochemical processes, such as the production of lipase, the lipolytic reaction, the mass transfer properties for the substrate phase, the fatty acid uptake and the growth kinetics of the cells. Microbial lipase production often depends on one or more of the following parameters: growth media composition (in particular the enzyme inducing compound), substrate concentration, growth rate, fermentation process design (i.e. batch, fed-batch or continuous operation mode) and interfacial area between lipid and water phase. The strain Geobacillus thermoleovorans IHI-91 is a member of a group of thermophilic bacteria, which show a high lipolytic activity. Kinetic studies with G. thermoleovorans IHI-91, growing aerobically on olive oil as the only carbon and energy source, show that oxidation of the liberated long-chain fatty acids rather than hydrolysis is the ratelimiting step. The main olive oil hydrolysis product, oleic acid, causes a significant growth-inhibitory effect. For G. thermoleovorans IHI-91, a concentration of 430 mg L⫺1 was found non-permissive for growth. Recently, researchers have focused on the use of lipolytic organisms and microbial lipases in the (pre-)treatment of lipid-rich wastewaters.
(42.3)
where the following parameters were obtained: μmax ⫽ 0.68 h⫺1 and Ks ⫽ 215 mg L⫺1. While a majority of the commercial supplements consist of bacterial species, a fungus, namely Yarrowia lipolytica ATCC20255, proved to be highly suitable for the (pre-)treatment of olive oil mill wastewater with a problematically high concentration of total fats (16 g L⫺1) (De Felice et al., 1997).
REFERENCES Arpigny, J.L., Jaeger, K.-E., 1999. Bacterial lipolytic enzymes: classification and properties. Biochem. J. 343, 177–183. Becker, P., Abu-Reesh, I., Markossian, S., Antranikian, G., Märkl, H., 1997. Determination of the kinetic parameters during continuous cultivation of the lipase-producing thermophile bacillus sp. IHI-91 on olive oil. Appl. Microbiol. Biotechnol. 48, 184–190. Becker, P., Köster, D., Popov, M.N., Markossian, S., Antranikian, G., Märkl, H., 1999. The biodegradation of olive oil and the treatment of
386
SECTION | I Bacterial and Fungal and Other Microbial Aspects
lipid-rich wool scouring wastewater under aerobic thermophilic conditions. Water Res 33, 653–660. Becker, P., Märkl, H., 2000. Modeling of olive oil degradation and oleic acid inhibition during chemostat and batch cultivation of Bacillus thermoleovorans IHI-91. Biotechnol. Bioeng. 70, 630–637. Brooksbank, A.M., Latchford, J.W., Mudge, S.M., 2007. Degradation and modification of fats, oils and grease by commercial microbial supplements. World J. Microbiol. Biotechnol. 23, 977–985. Cammarotaa, M.C., Freire, D.M.G., 2006. A review on hydrolytic enzymes in the treatment of wastewater with high oil and grease content. Bioresour. Technol. 97, 2195–2210. DiRusso, C.C., Black, P.N., 1999. Long-chain fatty acid transport in bacteria and yeast. Paradigms for defining the mechanism underlying this protein mediated process. Mol. Cell. Biochem. 192, 41–52. Dalmau, E., Sanchez, A., Montesinos, J.L., Valero, F., Lafuente, F.J., Cases, C., 1998. Study of the drop size frequencies in a microbial growth system with an aqueous-organic culture medium: lipase production from Candida rugosa. J. Biotechnol. 59, 183–192. De Felice, B., Pontecorvo, G., Carfagna, M., 1997. Degradation of waste waters from olive oil mills by Yarrowia lipolytica ATCC 20255 and Pseudomonas putida. Acta Biotechnol 17, 231–239. Gowland, P., Kernick, M., Sundaram, T., 1987. Thermophilic bacterial isolates producing lipase. FEMS Microbiol. Lett. 48, 339–343. Gupta, R., Gupta, N., Rathi, P., 2004. Bacterial lipases. An overview of production, purification and biochemical properties. Appl. Microbiol. Biotechnol. 64, 763–781. Handelsmann, T., Shoham, Y., 1994. Production and characterization of an extracellular thermostable lipase from a thermophilic Bacillus sp. J. Gen. Appl. Microbiol. 40, 435–443. Hsu, T.-C., Hanaki, K., Matsumoto, J., 1983. Kinetics of hydrolysis, oxidation and adsorption during olive oil degradation by activated sludge. Biotechnol. Bioeng. 25, 1829–1839. Jaeger, K.-E., Eggert, T., 2002. Lipases for biotechnology. Curr. Opin. Biotechnol. 13, 390–397. Janssen, P., Monk, C.R., Morgan, H.W., 1994. A thermophilic, lipolytic Bacillus sp. and continuous assay of its p-nitrophenyl-palmitate esterase activity. FEMS Microbiol. Lett. 120, 195–200.
Kambourova, M., Emanuilova, E., Dimitrov, P., 1996. Influence of culture conditions on thermostable lipase production by a thermophilic alkalitolerant strain of Bacillus sp. Folia Microbiol 41, 146–148. Kim, H.-K., Sung, M.-H., Kim, H.-M., Oh, T.-K., 1994. Occurrence of thermostable lipase in thermophilic Bacillus sp. strain 398. Biosci. Biotech. Biochem. 58, 961–962. Loehr, R.D., Roth, J.C., 1968. Aerobic degradation of long-chain fatty acid salts. J. WPCF 40, R385–R403. Loperenaa, L., Saraviaa, V., Murroa, D., Ferraria, M.D., Lareo, C., 2006. Kinetic properties of a commercial and a native inoculum for aerobic milk fat degradation. Bioresour. Technol. 97, 2160–2165. Markossian, S., Becker, P., Märkl, H., Antranikian, G., 2000. Isolation and characterisation of lipid-degrading Bacillus thermoleovorans IHI-91 from an Icelandic hot spring. Extremophiles 4, 365–371. Masse, L., Kennedy, K.J., Chou, S., 2001. Testing alkaline and enzymatic hydrolysis pretreatments for fat particles in slaughterhouse wastewater. Bioresour. Technol. 77, 145–155. Montesinos, J.L., Dalmau, E., Casas, E., 2003. Lipase production in continuous culture during chemostat culture of Candida rugosa. J. Chem. Technol. Biotechnol. 78, 753–761. Reimann, I., 2003. Thermophile Behandlungsverfahren zur biologischen Reinigung fetthaltiger Abwässer. PhD thesis. Technical University Hamburg-Harburg. Fortschritt-Berichte VDI Reihe 3 Nr. 789, VDI Verlag Düsseldorf. Sekelsky, A.M., Shreve, G.S., 1999. Kinetic model of biosurfactantenhanced hexadecane biodegradation by Pseudomonas aeruginosa. Biotechnol. Bioeng. 63, 401–409. Tsai, S.W., Chiang, C.L., 1991. Kinetics, mechanism and time-course of lipase-catalysed hydrolysis of high concentration olive oil in AOTisooctane reversed micelles. Biotechnol. Bioeng. 38, 206–210. Verger, R., de Haas, G.D., 1976. Interfacial enzyme kinetics of lipolysis. Annu. Rev. Biophy. Bioeng. 5, 77–117. Verger, R., 1997. “Interfacial activation” of lipases: Facts and artifacts. TIBTECH 15, 32–38. Wang, Y., Srivastava, K.C., Shen, G.-J., Wang, H.Y., 1995. Thermostable alkaline lipase from a newly isolated thermophilic Bacillus, strain A30-1 (ATCC 53841). J. Ferment. Bioeng. 79, 433–438.
Chapter 43
Degradation of Phenolic Compounds Found in Olive Products by Lactobacillus plantarum Strains José María Landete1, Héctor Rodríguez1, José Antonio Curiel1, Blanca de las Rivas1, Félix López de Felipe2 and Rosario Muñoz1 1 2
Departamento de Microbiología. Instituto de Fermentaciones Industriales. CSIC, Madrid, Spain Grupo en Biotecnología de Bacterias Lácticas de Productos Fermentados. Instituto del Frío. CSIC, Madrid, Spain
43.1 INTRODUCTION The olive tree (Olea europaea L.) is one of the most important fruit trees in the Mediterranean countries. Their products, olive oil and also table olives, are important components of the Mediterranean diet and are consumed in large quantities globally. Olives are the major fermented vegetable in Western countries. Homemade production of naturally fermented table olives is very common in Mediterranean countries and the production methods vary according to local tradition. The fermentation is fundamental in order to have an appropriate decrease in pH and the development of a lactic acid microbiota that can prevent the growth of spoilage microorganisms as well as potential pathogenic microbes. Lactic acid bacteria are recognized to play an important role in olive fermentation and Lactobacillus plantarum and Lactobacillus pentosus are, in fact, regarded as the main species leading this process, being often used as starters in guided olive fermentation (Ruíz-Barba et al., 1994; Vega Leal-Sánchez et al., 2003). Although L. plantarum is present in small numbers in the fresh products, the traditional process consisting of lye treatment, washing, and brining of the fruits favors its rapid and predominant growth over other microorganisms in the brine (Durán et al., 1993; Ercolini et al., 2006; Chamkha et al., 2008). Olives have been recognized as a source of biophenols. Despite the great importance of phenolic compounds in the health-beneficial properties of olive products, and the prevalence of L. plantarum during olive fermentations, there are only few reports studying the metabolism of olive phenolics by L. plantarum strains. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
43.2 PHENOLIC COMPOUNDS AND LACTOBACILLUS PLANTARUM Lactobacillus plantarum is a versatile bacterium found in a variety of ecological niches, ranging from vegetable and plant fermentations to the human gastrointestinal tract. L. plantarum cells are rods with rounded ends, straight, generally 0.9–1.2 μm wide and 3–8 μm long, occurring singly, in pairs or in short chains (Figure 43.1). The genome of the lactic acid bacterium L. plantarum strain WCFS1 has been sequenced (Kleerebezem et al., 2003), and its size of 3.3 Mb is among the largest known for lactic acid bacteria (Makarova et al., 2006; Makarova and Koonin, 2007). It is thought that such genome length is related to the diversity of environmental niches in which L. plantarum is encountered.
FIGURE 43.1 Transmission electron micrograph of L. plantarum CECT 748T grown in a defined media (⫻ 12 000).
387
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
388
SECTION | I Bacterial and Fungal and Other Microbial Aspects
Nonetheless this bacterium is most frequently found in the fermentation of plant-derived raw materials, which include several industrial and artisan food and feed fermentations, like must, olives, and a variety of vegetable fermentations. The effects of olive phenolic compounds on L. plantarum growth have been studied by several authors. Brines from non-alkali-treated green olives were tested for their antimicrobial properties against L. plantarum (RuízBarba et al., 1990). They showed a marked bactericidal effect toward L. plantarum. Aqueous solutions of the total phenolics extracted from these brines had the same bactericidal effect. It has been demonstrated that diffused olive polyphenols exerted a significant inhibitory growth effect on L. plantarum during brining only when they were associated with NaCl (Durán et al., 1993). Ruíz-Barba et al. (1993) studied the viability of L. plantarum in the presence of single or combined fractions of isolated phenolic compounds from NaOH-treated and untreated olive brines. When assayed at the concentrations found in brines, only the single phenolic fraction containing hydroxytyrosol strongly inhibited L. plantarum. However, tyrosol, vanillic acid, verbascoside, and luteolin-7-glucoside, when assayed as single fractions, had no bactericidal effect against L. plantarum. Recently, similar results were obtained by evaluating inhibitory activities of p-hydroxybenzoic, sinapic, syringic, protocatechuic, and cinnamic acids on L. plantarum growth (Landete et al., 2008). Hydroxytyrosol was the sole compound found to be bactericidal at low concentration (Ruíz-Barba et al., 1993). However, in the same study the inhibitory combined effect of some olive phenolics was also clearly demonstrated (Ruíz-Barba et al., 1993). Several authors have studied the effects of oleuropein and its hydrolysis products on the survival of L. plantarum. The results reported were different depending on the antibacterial test method used. Ruíz-Barba et al. (1991) found that oleuropein was bactericidal against L. plantarum strains isolated from green olive fermentation brines. Heattreated oleuropein also demonstrated a strong bactericidal effect but not the alkali-treated oleuropein, which allowed survival of most of the L. plantarum strains tested. In addition, the inhibition of L. plantarum growth was always observed when oleuropein was associated with another phenolic fraction or added directly in brines. Contrarily, Rozès and Peres (1996) reported that untreated oleuropein was not inhibitory to L. plantarum; however, when the aglycon was formed in the medium, cell viability decreased. Moreover, when oleuropein and NaCl were associated, the bactericidal effect on L. plantarum growth was very pronounced. The same authors also reported that the presence of glucose seems to delay the antimicrobial activity of these two compounds (Rozès and Peres, 1996). The bactericidal effect of oleuropein was accompanied by changes in the typical bacillary structure of L. plantarum (Ruíz-Barba et al., 1991). The morphology of L. plantarum strains incubated in heat-treated or untreated
oleuropein solutions changes in both size and shape. Cells become longer, often deformed in their bacillary structure and also become wider. This could indicate that oleuropein promotes peptidoglycan disruption, which could lead to cell death by destruction of the cell envelope. It has been described that oleuropein increased the leakage of inorganic phosphate, glutamic acid and potassium from L. plantarum cells (Juven et al., 1972). In addition, brines from nonalkali-treated green olives showed a marked bactericidal effect toward L. plantarum (Ruíz-Barba et al., 1990). The bactericidal effect was shown by certain alterations at two different levels of the cellular ultrastructure, the cell wall and the cytoplasmic membrane. Other changes in the cell ultrastructure were observed, such as the presence of mesosomal membranes. Fluorescence microscopy showed adsorption of phenolics to both whole cells and isolated cell walls from L. plantarum. These alterations possibly lead to the disruption of the cell envelope when the incubation time is extended, thus promoting cellular lysis (Ruíz-Barba et al., 1990). The effects of different phenolic compound concentrations on the fatty acid composition of L. plantarum isolated from traditional homemade olive brines were determined by Rozès and Peres (1998). They found that phenolic compounds altered L. plantarum fatty acid constitution. When caffeic and ferulic acids were added to the medium, lactobacillic acid production was reduced at the expense of unsaturated fatty acids, mainly palmitoleic acid. Consequently in stationary-phase cultures, an increase in the degree of unsaturation was found. The authors suggest that acidic phenols could modify the membrane fluidity, as a possible hypothesis for the changes observed in the unsaturated fatty acid content.
43.3 METABOLISM OF PHENOLIC COMPOUNDS BY LACTOBACILLUS PLANTARUM Table olives have a different qualitative and quantitative phenolic composition than the raw olive fruits from which they are prepared. There are also notable differences among the phenolic compounds present in olive fruit and oil. Ryan and Robards (1998) listed the phenolic compounds found in olives, in leaves, seed, pulp, as well as oil. Different phenolic groups are present in some olive products, such as phenolic acids, phenolic alcohols, flavonoids, and secoiridoids. Some of the phenolic acids present in olives are caffeic, p-coumaric, protocatechuic, vanillic, p-hydroxybenzoic, p-hydroxyphenylacetic, 3,4-dihydroxyphenylacetic, and ferulic acid. Among the phenolic alcohols, tyrosol and hydroxytyrosol are present in olive products. They can be found complexed to glucoside and acetate. The most frequently described flavonoids include luteolin
CHAPTER | 43 Degradation of Phenolic Compounds Found in Olive Products by Lactobacillus plantarum Strains
7-O-glucoside, rutin and apigenin 7-O-glucoside, and the anthocyanins, cyaniding 3-O-glucoside. The predominant secoiridoids of olive fruit pulp are oleuropein and ligstroside. Some oleuropein derivatives have also been described, namely, demethyloleuropein, oleuropein aglycone, and elenoic acid. Table 43.1 summarizes the current knowledge on the metabolism of some phenolic compounds found in olive products by L. plantarum strains.
43.3.1 Phenolic Acids The term ‘phenolic acids’, in general, describe phenols that possess one carboxylic acid functional group. The
phenolic acids found in olive products contain mainly two distinguishing constitutive carbon frameworks: the hydroxycinnamic and the hydroxybenzoic structures. Figure 43.2 summarizes the phenolic acids or related acids reported to be metabolized by L. plantarum strains.
43.3.1.1 Hydroxycinnamic Acids Hydroxycinnamic acids are common in olive products. The presence of acids such as p-coumaric, o-coumaric, caffeic, hydrocaffeic, ferulic, and sinapic acid, as well as cinnamic acid, has been reported in olive products (Ryan and Robards, 1998). Several studies were performed in order
TABLE 43.1 Metabolism of phenolic compounds by L. plantarum strains. Compound assayed
Compound produced
Enzymes involved
References
Caffeic acid
p-Vinyl catechol
PAD
Cavin et al. (1997b)
p-Ethyl catechol
Reductase
Cavin et al. (1997a) Barthelmebs et al. (2000) Rodríguez et al. (2008c) Rodríguez et al. (2008b)
Catechol
Not degraded
Whiting and Coggins (1971) Rodríguez et al. (2008c)
Chlorogenic acid
NDa
Rodríguez et al. (2008c)
Cinnamic acid
Not degraded
Rodríguez et al. (2008c)
o-Coumaric acid
Not degraded
Rodríguez et al. (2008c)
p-Coumaric acid
p-Vinyl phenol
PADb
Cavin et al. (1997b)
p-Ethyl phenol
Reductase
Cavin et al. (1997a) Barthelmebs et al. (2000) Rodríguez et al. (2008c) Rodríguez et al. (2008b)
Ferulic acid
p-Vinyl guaiacol
PAD
Cavin et al. (1997b)
p-Ethyl guaiacol
Reductase
Cavin et al. (1997a) Rodríguez et al. (2008c) Rodríguez et al. (2008b)
Gallic acid
Pirogallol
Hydroxybenzoic acid
Not degraded
389
Decarboxylase
Rodríguez et al. (2008c) Rodríguez et al. (2008c) (Continued)
390
SECTION | I Bacterial and Fungal and Other Microbial Aspects
TABLE 43.1 (Continued) Compound assayed
Compound produced
Hydroxytyrosol
Not degraded
Enzymes involved
References Ciafardini et al. (1994) Marsilio et al. (1996) Marsilio and Lanza (1998) Landete et al. (2008)
Oleuropein
Hydroxytyrosol
β-Glucosidase
Ciafardini et al. (1994)
Esterase
Marsilio et al. (1996) Marsilio and Lanza (1998) Landete et al. (2008)
Phloretic acid
Not degraded
Protocatechuic acid
Catechol
Quercitin
Not degraded
Quinic acid
Catechol
Rodríguez et al. (2008c) Decarboxylase
Rodríguez et al. (2008c) Landete et al. (2007)
Several enzymes
Whiting and Coggins (1971) Whiting and Coggins (1974) Whiting (1975)
Shikimic acid
Catechol
Several enzymes
Whiting and Coggins (1971) Whiting and Coggins (1974) Whiting (1975)
Sinapic acid
Not degraded
Rodríguez et al. (2008c)
Syringic acid
Not degraded
Rodríguez et al. (2008c)
Tyrosol
Not degraded
Landete et al. (2008)
Vanillic acid
Not degraded
Rodríguez et al. (2008c)
a
ND, not detected;
b
PAD, phenolic acid decarboxylase or PDC or PadA.
to discover whether L. plantarum strains have the ability to degrade hydroxycinnamic acids. L. plantarum cultures were grown in presence of each acid, so, if L. plantarum cells are able to metabolize the hydroxycinnamic acid, the dead-end degradation products could be detected in the culture media. In addition, cell-free extracts containing all the L. plantarum soluble proteins were incubated in presence of the same acids in order to obtain information on the induction of the enzymes involved.
Figure 43.3 showed the HPLC chromatograms of the hydroxycinnamic acids that are metabolized by L. plantarum CECT 748T cultures. As compared to the control (Figure 43.3, 1A), supernatants obtained from cell cultures grown in p-coumaric acid showed the presence of vinyl and ethyl phenol, resulting from the decarboxylation, and decarboxylation plus reduction of p-coumaric acid (Figure 43.3, 1B). A similar situation was observed in the caffeic acid sample as in the supernatants from the cultures, the products of the
CHAPTER | 43 Degradation of Phenolic Compounds Found in Olive Products by Lactobacillus plantarum Strains
O
HO
O
OCH3
OH OH p-Coumaric acid O
OH
OH
Ferulic acid
Caffeic acid
OH
HO
O
OH
OH
OH OH
OH Gallic acid
Protocatechuic acid
O HO
decarboxylation (4-vinyl catechol) as well as the decarboxylation plus reduction (4-ethyl catechol) of caffeic acid were identified (Figure 43.3, 2A and 2B). Like p-coumaric and caffeic acid, ferulic acid was found to be metabolized by L. plantarum cell cultures (Figure 43.3, 3A and 3B). Among the hydroxycinnamic acids assayed, only p-coumaric and caffeic acids were metabolized by both cell cultures and cell-free extracts of L. plantarum CECT 748T; however, the extracts were unable to degrade ferulic acid. Ferulic acid was only partially modified by cell cultures. These results seem to indicate that the enzymes involved in ferulic acid metabolism need to be synthesized after their induction by the presence of this phenolic acid. From these results it could be concluded that uninduced L. plantarum cell-free extracts contained decarboxylases able to decarboxylate p-coumaric and caffeic acids into their vinyl derivatives. In fact, a phenolic acid decarboxylase (PAD) has been identified in L. plantarum strains. The gene encoding a p-coumarate decarboxylase (PAD) from L. plantarum LPCHL2 has been cloned and the recombinant protein overexpressed in E. coli (Cavin et al., 1997b). Cavin et al. (1997a) analyzed the substrate specificity of the purified PAD by using several hydroxycinnamic acids (e.g., p-coumaric, o-coumaric, m-coumaric, ferulic, caffeic, hydrocaffeic, phloretic, 2-methoxycinnamic, and 3-methoxycinnamic acids). Only p-coumaric and
OH
O
OH
HO
O
OH
HO
HO
OH
OH OH
OH Quinic acid
Shikimic acid
FIGURE 43.2 Chemical structure of the phenolic or related acids reported to be metabolized by L. plantarum strains.
2000
150
1A
1B
mAU
100 1000 50
EP
VP 0
0
2A
2B
mAU
200 1000 100
VC 0
mAU
EC
0
3A
2000
3B
600 400
1000
VG
200
EG
0
0 0
20
40
Minutes
60
391
80
0
20
40
60
80
Minutes
FIGURE 43.3 HPLC chromatograms of the degradation of hydroxycinnamic acids by Lactobacillus plantarum. Chromatograms of supernatant from L. plantarum CECT 748T grown for 10 days in the presence of p-coumaric (1B), caffeic (2B), and ferulic acids (3B). The HPLC chromatograms from the control samples (1A, 2A, or 3A, respectively). VP, p-vinyl phenol; EP, p-ethyl phenol; VC, p-vinyl catechol; EC, p-ethyl catechol; VG, p-vinyl guaiacol; EG, p-ethyl guaiacol.
392
SECTION | I Bacterial and Fungal and Other Microbial Aspects
caffeic acids were metabolized, at the same rate as KM. The authors concluded that only the acids with a para hydroxyl group with respect to the unsaturated side chain and with a substitution –H or –OH in position meta were metabolized. In 2003, the complete genome sequence of Lactobacillus plantarum WCFS1 was available (Kleerebezem et al., 2003). This strain possesses a PAD enzyme that differs from the enzyme previously purified, mainly on its C-terminal region where the substrate specificity is determined. In addition, the PAD protein from the L. plantarum type strain CECT 748T was crystallized, and its amino acid sequence determined to be identical to that of L. plantarum WCFS1 (Rodríguez et al., 2007). The recombinant PAD from L. plantarum CECT 748T (Figure 43.4) is able to decarboxylate exclusively the hydroxycinnamic acids, p-coumaric, caffeic, and ferulic acids (Rodríguez et al., 2008b). Kinetic analysis showed that the enzyme has a 14-fold higher KM value for p-coumaric and caffeic acids than for ferulic acid. PDC catalyzes the formation of the corresponding 4-vinyl derivatives (vinyl phenol, vinyl catechol, and vinyl guaiacol) from p-coumaric, caffeic, and ferulic acids, respectively (Rodríguez et al., 2008b). Transcriptional studies of the gene encoding the PAD in L. plantarum LPCHL2 demonstrated that its transcription is phenolic-acid-dependent. A mutant deficient in the PAD activity was constructed to study phenolic acid alternate pathways in L. plantarum (Barthelmebs et al., 2000). The mutant strain retained the ability to metabolize weakly p-coumaric and ferulic acids into vinyl derivatives. Therefore, these results indicate the presence in L. plantarum of a second phenolic acid decarboxylase, greatly induced by ferulic acid and of a putative phenolic acid reductase activity, induced by p-coumaric and ferulic acids in the presence of glucose. The strong, rapid, and inducible PAD enzyme synthesis following exposure to phenolic acids can be considered a specific chemical stress response to overcome phenolic acid toxicity. This was proved by disruption of the PAD
1
2
3
kDa 45 31 21.5 14.5 FIGURE 43.4 Expression and purification of the PAD protein from L. plantarum CECT 748T. SDS-PAGE analysis of soluble E. coli cell extracts. Lane 1: Control plasmid. Lane 2: Plasmid containing the L. plantarum PAD gene. Lane 3: Fraction eluated after the affinity column. The 15% polyacrylamide gel was stained with Coomassie blue. The positions of molecular mass markers (Bio-Rad) are indicated on the left. The L. plantarum recombinant PAD protein is indicated by an arrow.
encoding gene, which rendered the bacterium sensitive to these substrates and makes it unable to be grown at a low pH in the presence of p-coumaric acid (Barthelmebs et al., 2000).
43.3.1.2 Hydroxybenzoic Acids Some hydroxybenzoic acids are found in olive products (Table 43.1). L. plantarum cells were grown in the presence of several hydroxybenzoic acids, such as gallic, syringic, p-hydroxybenzoic, protocatechuic, and vanillic acids. Among the hydroxybenzoic acids assayed, only gallic and protocatechuic acids were metabolized by both cell cultures and cellfree extracts from L. plantarum CECT 748T (Table 43.1). Figure 43.5 shows the HPLC chromatograms obtained with L. plantarum cells grown in the presence of gallic acids. As compared to the control (Figure 43.5, 1A), it could be observed that gallic acid was decarboxylated to pirogallol (Figure 43.5, 1B). Pirogallol was also obtained from gallic acid during the degradation of tannic acid by L. plantarum cell-free extracts (Rodríguez et al., 2008a). Previously, Osawa et al. (2000) suggested the occurrence of a gallate decarboxylase activity in L. plantarum strains. As early as 1971, Whiting and Coggins reported that L. plantarum cells grown in a medium containing protocatechuic acid completely metabolized it to catechol, and there was no indication of a further metabolism of catechol (Whiting and Coggins, 1971). Recently, Rodríguez et al. (2008c) described that protocatechuic acid was completely decarboxylated to catechol by cultures of L. plantarum grown in the presence of this hydroxybenzoic acid (Figure 43.5, 2A and 2B). These results indicate that catechol is a dead-end product of protocatechuate degradation in L. plantarum cultures. No information is available about the L. plantarum enzyme involved in the protocatechuic acid decarboxylation. However, similarly to gallic acid, cell-free extracts were able to decarboxylate partially protocatechuic acid (Rodríguez et al., 2008c). The same five hydroxybenzoic acids (gallic, syringic, p-hydroxybenzoic, protocatechuic, and vanillic acids) were assayed by using recombinant PAD from L. plantarum CECT 748T. As expected, the phenolic acids that were not metabolized by L. plantarum cultures were not metabolized by the purified enzyme. In addition, neither protocatechuic nor gallic acids were used as substrate by purified PAD (Rodríguez et al., 2008b). Therefore, these results indicated that a still unknown decarboxylase enzyme(s) must be the response of the decarboxylation of gallic and protocatechuic acids by L. plantarum CECT 748T.
43.3.1.3 Phenolic-related Acids As shown in Table 43.1, the metabolism of two additional olive phenolic acids, such as phloretic and chlorogenic acids,
CHAPTER | 43 Degradation of Phenolic Compounds Found in Olive Products by Lactobacillus plantarum Strains
393
1B
1A
mAU
100 1000 50 P 0
0
2A
2B 400
mAU
1000 200 500 C 0
0 0
20
40
60
80
Minutes
0
20
40
60
80
Minutes
FIGURE 43.5 HPLC chromatograms of the degradation of hydroxybenzoic acids by Lactobacillus plantarum. Chromatograms of supernatant from L. plantarum CECT 748T grown for 10 days in the presence of gallic (1B), and protocatechuic acids (2B). The HPLC chromatograms from the control samples (1A, or 2A, respectively). P, pirogallol; C, catechol.
have been studied previously (Rodríguez et al., 2008c). However, chlorogenic acid was not detected by the chromatographic method used, and phloretic acid was not metabolized by cell cultures as well as by the cell-free extracts. In addition, a number of acids present in olives (such as cinnamic, elenoic, shikimic, and quinic acids) could be considered as phenolics on the basis of metabolic considerations, although they lack a phenolic group or even an aromatic ring. The metabolism of quinic and shikimic acids by L. plantarum has been studied by Whiting and Coggins (1971). L. plantarum reduced quinic acid and the related shikimic acid under anaerobic conditions in the presence of suitable hydrogen donors including fructose, glucose and lactates (Whiting, 1975). L. plantarum shows two metabolic routes for these acids, one oxidative and one reductive. Both pathways proceed simultaneously under anaerobic conditions. The oxidative route gives catechol as end-product; there was no indication of further metabolism under anaerobic conditions. The pathway, that has been described to comprise 11 steps, has been elucidated by using several procedures, namely the use of growing cells, washed cells grown with different substrates, cellfree extracts, and the separation and purification of a NADspecific hydroaromatic dehydrogenase involved in five steps of the pathway (Whiting and Coggins, 1974). This enzyme, similarly to the enzymes catalyzing the second and sixth steps, was an inducible enzyme. In this pathway both D- and L-lactates act indistinctly as hydrogen donors for the reductive steps, while lactate was simultaneously oxidized to acetate. These oxidoreductions proceeded under anaerobic conditions. However, L. plantarum not only reduces quinic acid but at the same
time, even under anaerobic conditions, oxidizes a proportion and converts the product of dehydratase and decarboxylase action to catechol (Whiting and Coggins, 1971). It is not known whether dehydroshikimic acid is the branch point of the oxidative and reductive pathways or whether shikimic acid is oxidized to dehydroshikimate by a pyridine nucleotide-independent shikimate dehydrogenase found in cellfree extracts of this organism (Whiting and Coggins, 1971). The pathway suggested by these authors may be involved in raising the pH of their environment since all the metabolic changes involved bring about such an increase in pH.
43.3.2 Phenyl Alcohols The presence of several phenyl alcohols (such as hydroxytyrosol, tyrosol, or catechol) has been described in olive products. Hydroxytyrosol and tyrosol are among the prevailing phenols in olive products. Hydroxytyrosol, that results from the hydrolysis of oleuropein, killed L. plantarum completely within 2 h (Ruíz-Barba et al., 1990). However, an antimicrobial effect for tyrosol has not been reported (Ruíz-Barba et al., 1993; Landete et al., 2008). In relation to the bacterial metabolism of these phenols, it should be noted that none of them was degraded when L. plantarum strains were grown in the presence of hydroxytyrosol and tyrosol (Landete et al., 2008). Catechol was by the first time identified in natural black olive pulp by Romero et al. (2002). As mentioned in previous sections, no indication of a metabolism of catechol was reported by Whiting and Coggins (1971), and more recently, by Rodríguez et al. (2008c).
394
SECTION | I Bacterial and Fungal and Other Microbial Aspects
43.3.3 Glycosides
mAU
Phenolic structural complexity is introduced by the common occurrence of certain phenolics in the O-glycosides form, glucose being the most commonly encountered sugar. Oleuropein is the main bitter secoiridoid glucoside present in olives. During crushing and malaxing processes, oleuropein and demethyloleuropein are hydrolyzed by endogenous glycosidases. The aglycons become soluble in the oil phase while the glycosides remain in the wastewater phase. Oleuropein was also found to be hydrolyzed by the β-glucosidase produced by L. plantarum strains (Ciafardini et al., 1994). Cultures from L. plantarum containing oleuropein showed a significant decrease of oleuropein content over time with a concomitant increase of aglycone derivatives, which were further degraded to hydroxytyrosol (Marsilio et al., 1996). Due to its instability in aqueous solution, the first compound of the enzymatic degradation of the oleuropein rearranged to several aglycone structures before transforming into stable final compounds. Therefore, at various steps of the enzymatic reaction, different chemical structures were determined (Figure 43.6). The results obtained by Marsilio et al. (1996) suggest that the L. plantarum activity on oleuropein involves a two-step process: (i) an enzymatic hydrolysis of the glycosidic linkage by a β-glucosidase action to release the aglycone, the first observable intermediate in the process; (ii) after the β-glycosidic linkage is completely hydrolyzed, as deduced from the disappearance of oleuropein, other compounds originate, probably due to esterase activity that hydrolyzes the ester to an acid (elenoic acid) and an alcohol (hydroxytyrosol) (Marsilio et al., 1996). Metabolism of oleuropein seems to be carried out by inducible enzymes since a cell-free extract from a L. plantarum culture grown in the absence of oleuropein was unable to metabolize it (Landete et al., 2008). Marsilio and Lanza (1998) reported that oleuropein consumption is concomitant with bacterial growth. The concentration of glucose in the medium is a critical variable affecting oleuropein breakdown in the olives. In the
presence of 10–20 g L⫺1 glucose, oleuropein concentration decreased but a 30–40% residual was not hydrolyzed. The unhydrolyzed oleuropein concentration increased to 70% in the presence of 30–50 g L⫺1 glucose (Marsilio and Lanza, 1998). Concurrent to the decrease of oleuropein, aglycone was detected only at trace levels since this derivative was further degraded to hydroxytyrosol. This indicated that the esterase activity involved in the biodegradation process of oleuropein was not affected by glucose (Marsilio and Lanza, 1998). The presence of a β-glucosidase activity on L. plantarum strains has been also reported by Sestelo et al. (2004), who identified one active form of β-glucosidase in the culture filtrate of L. plantarum. This enzyme was able to hydrolyze several β-linked glucose dimers, with a hydrolysis rate order of (β-1,4) ⬎ (β-1,3) ⬎ (β-1,2) ⬎ (β-1,6). This protein seems to be a broad specificity β-glucosidase since it can hydrolyze β-diglucosides and also aryl-β-glucosides. The enzyme had a molecular mass of about 40 kDa. However, later, Spano et al. (2005) described a gene coding for a putative 61 kDa β-glucosidase in L. plantarum. They analyzed the expression of this β-glucosidase gene under several stresses, by RT-PCR and Northern blot analysis. The putative β-glucosidase gene was apparently regulated by abiotic stresses such as temperature, ethanol, and pH.
43.3.4 Flavonols Among the flavonols whose presence in olive products has been generally detected, quercitin and its related glycoside, rutin (quercitin-3-rutinoside), are included (Ryan and Robards, 1998). Landete et al. (2007) reported that quercitin was not degraded by L. plantarum strains. The fact that the quercitin is not degraded by L. plantarum strains after growth does not mean that it did not influence its growth performance. Among flavonols, quercitin showed high affinity by the lipid bilayer of membranes (Nakayama et al., 1998) and therefore could affect bacterial growth.
200
200
100
100
0
0 0
A
HT
20
40
Minutes
60
80
0
B
20
40
60
80
Minutes
FIGURE 43.6 HPLC chromatograms of the degradation of oleuropein by Lactobacillus plantarum. Chromatograms of supernatant from L. plantarum CECT 748T grown for 10 days in the presence of oleuropein (B). The HPLC chromatogram from the control oleuropein sample is showed in (A). HT, hydroxytyrosol.
CHAPTER | 43 Degradation of Phenolic Compounds Found in Olive Products by Lactobacillus plantarum Strains
It has been observed that in some L. plantarum strains quercitin promotes quicker growth upon inoculation and higher growth rates (López de Felipe et al., submitted manuscript).
43.4 TREATMENT OF OLIVE BYPRODUCTS BY LACTOBACILLUS PLANTARUM Generally the olive industry produces two residues, solids and olive mill wastewater (OMW). OMW is composed of the vegetation water of olive pulp tissue and oil in the form of a very stable emulsion. OMW is one of the most complex plant effluents. The ecological problem of OMW is due primarily to the presence of phenolic compounds, which make OMW toxic and resistant to biological degradation. Up to date, a large variety of methods have been suggested for the treatment of OMW; however, most of them are aimed at the decomposition/destruction of the contained polyphenols and not their exploitation (Arvanitoyannis et al., 2007). Lactic acid bacteria with their capacity to reduce oxygen pressure, redox potential and pH, offers a new promising approach to the bioconversion of phenolic compounds present in olive waste. In fact, the effects of L. plantarum growth on the reductive decolorization and biodegradation of olive phenolic compounds was evaluated by Lamia and Moktar (2003). OMW is unstable and turns black under aerobic conditions because of the auto-oxidation of phenolic compounds. L. plantarum growth on fresh OMW induced the depolymerization of phenolic compounds of high molecular weight, with a resultant decolorization of fresh OMW, and significant reduction of total phenols, in proportion to the dilution of OMW. They found that approximately 58% of the color, 55% of the chemical oxygen demand, and 46% of the phenolic compounds were removed when OMW was diluted ten times. The removal of phenolic compounds was associated with the depolymerization, the partial adsorption on the cells and the biodegradation of certain simple phenolic compounds. A similar phenolic depolymerization mechanism was observed in L. plantarum degrading tannins (Rodríguez et al., 2008a). In addition, the application of L. plantarum to the olive fruit during crushing could constitute a new microbiological process for olive oil quality improvement. Kachouri and Hamdi (2004) studied the transformation of phenolic compounds contained in OMW into valuable products using L. plantarum in order to increase their transportation from OMW to olive oil. Incubation of olive oil samples with fermented OMW by L. plantarum caused polyphenols to decrease in OMW and increase in oil. The lower total phenolic content in fermented OMW in comparison to OMW control resulted from the depolymerization of phenolic compounds of high molecular weight by L. plantarum. Fermentation with L. plantarum induced reductive
395
depolymerization of OMW which are more soluble in olive oil. The analysis of the phenolic compounds found in olive oil after storage showed that the application of L. plantarum favors the increase of all phenolic compounds in olive oil, especially by depolymerization and by reductive conversion of phenolic compounds of olive and oxygen fixation. The increase in phenolic content in olive oil was more prominent for simple polyphenols such as oleuropein, p-hydroxyphenylacetic, vanillic acid and ferulic acids and tyrosol. In summary, the authors concluded that olive oil mixed with the OMW and fermented by L. plantarum had a higher quality and stability because of a higher content of simple phenolic compounds.
SUMMARY POINTS ●
●
●
●
●
Lactobacillus plantarum is the main lactic acid bacterial species associated with olive processing. Olive phenolic compounds could inhibit L. plantarum growth depending on its concentration. L. plantarum metabolize some hydroxycinnamic acids (p-coumaric, ferulic, and caffeic acids) into their vinyl derivatives by the PAD protein. Among the hydroxycinnamic acids assayed, only gallic and protocatechuic acids were metabolized to pirogallol and catechol, respectively. Olive glycosides are converted to their corresponding aglycones by the action of L. plantarum glycosidases that still remains unknown.
REFERENCES Arvanitoyannis, I.S., Kassaveti, A., Stefanatos, S., 2007. Current and potential uses of thermally treated olive oil waste. Int. J. Food Sci. Technol. 42, 852–867. Barthelmebs, L., Divies, C., Cavin, J.F., 2000. Knockout of the p-coumarate decarboxylase gene from Lactobacillus plantarum reveals the existence of two other inducible enzymatic activities involved in phenolic acid metabolism. Appl. Environ. Microbiol. 66, 3368–3375. Cavin, J.F., Barthelmebs, L., Diviès, C., 1997a. Molecular characterization of an inducible p-coumaric acid decarboxylase from Lactobacillus plantarum: gene cloning, transcriptional analysis, overexpression in Escherichia coli, purification, and characterization. Appl. Environ. Microbiol. 63, 1939–1944. Cavin, J.F., Barthelmebs, L., Guzzo, J., Van Beeumen, J., Samyn, B., Travers, J.F., Diviès, C., 1997b. Purification and characterization of an inducible p-coumaric acid decarboxylase from Lactobacillus plantarum. FEMS Microbiol. Lett. 147, 291–295. Chamkha, M., Sayadi, S., Bru, V., Godon, J.J., 2008. Microbial diversity in Tunisian olive fermentation brine as evaluated by small subunit rRNA – single strand conformation polymorphism analysis. Int. J. Food Microbiol. 122, 211–215. Ciafardini, G., Marsilio, V., Lanza, B., Pozzi, N., 1994. Hydrolysis of oleuropein by Lactobacillus plantarum strains associated with olive fermentation. Appl. Environ. Microbiol. 60, 4142–4147.
396
SECTION | I Bacterial and Fungal and Other Microbial Aspects
Durán, M.C., García, P., Brenes, M., Garrido, A., 1993. Lactobacillus plantarum survival during the first days of ripe olive brining. System. Appl. Microbiol. 16, 153–158. Juven, B., Henis, Y., Jacoby, B., 1972. Studies on the mechanism of the antimicrobial action of oleuropein. J. Appl. Bacteriol. 35, 559–567. Kachouri, F., Hamdi, M., 2004. Enhancement of polyphenols in olive oil by contact with fermented olive mill wastewater by Lactobacillus plantarum. Process Biochem. 39, 841–845. Ercolini, D., Villani, F., Aponte, M., Mauriello, G., 2006. Fluorescence in situ hybridisation of Lactobacillus plantarum group on olives to be used in natural fermentations. Int. J. Food Microbiol. 112, 291–296. Kleerebezem, M., Boekhorst, J., van Kranenburg, R., Molenaar, D., Kuipers, O.P., Leer, R., Tarchini, R., Peters, S.A., Sandbrink, H.M., Fiers, M.W.E.J., Stiekema, W., Lankhorst, R.M.K., Bron, P.A., Hoffer, S.M., Nierop Groot, M.N., Kerkhoven, R., de Vries, M., Ursing, B., de Vos, W.M., Siezen, R.J., 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. USA 100, 1990–1995. Lamia, A., Moktar, H., 2003. Fermentative decolorization of olive mill wastewater by Lactobacillus plantarum. Process Biochem. 39, 59–65. Landete, J.M., Curiel, J.A., Rodríguez, H., de las Rivas, B., Muñoz, R., 2008. Study of the inhibitory activity of phenolic compounds found in olive products and their degradation by Lactobacillus plantarum strains. Food Chem. 107, 320–326. Landete, J.M., Rodríguez, H., de las Rivas, B., Muñoz, R., 2007. Highadded-value antioxidants obtained from the degradation of wine phenolics by Lactobacillus plantarum. J. Food Prot. 70, 2670–2675. López de Felipe, F., Curiel, J.A., Muñoz, R. Quercitin effects on the wine Lactobacillus plantarum RM71 strain growing in a chemically defined medium (submitted). Makarova, K.S., Koonin, E.V., 2007. Evolutionary genomics of lactic acid bacteria. J. Bacteriol. 189, 1199–1208. Makarova, K., Slesarev, A., Wolf, Y., Sorokin, A., Mirkin, B., Koonin, E., Pavlov, A., Pavlova, N., Karamychev, V., Polouchine, N., Shakhova, V., Grigoriev, I., Lou, Y., Rohksar, D., Lucas, S., Huang, K., Goodstein, D.M., Hawkins, T., Plengvidhya, V., Welker, D., Hughes, J., Goh, Y., Benson, A., Baldwin, K., Lee, J.H., Díaz-Muñiz, I., Dosti, B., Smeianov, V., Wechter, W., Barabote, R., Lorca, G., Altermann, R., Barrangou, R., Ganesan, B., Xie, Y., Rawsthorne, H., Tamir, D., Parker, C., Breidt, F., Broadbent, J., Hutkins, R., O’Sullivan, D., Steele, J., Unlu, G., Saier, M., Klaenhammer, T., Richardson, P., Kozyavkinn, S., Weimer, B., Mills, D., 2006. Comparative genomics of the lactic acid bacteria. Proc. Natl. Acad. Sci. USA 103, 145611–145616. Marsilio, V., Lanza, B., 1998. Characterization of an oleuropein degrading strain of Lactobacillus plantarum. Combined effects of compounds present in olive fermenting brines (phenols, glucose and NaCl) on bacterial activity. J. Sci. Food Agric. 76, 520–524. Marsilio, V., Lanza, B., Pozzi, N., 1996. Progress in table olive debittering: degradation in vitro of oleuropein and its derivatives by Lactobacillus plantarum. JAOCS 73, 593–597. Nakayama, T., Ono, K., Hashimoto, K., 1998. Affinity of antioxidant polyphenols for lipid bilayers evaluated with a liposome system. Biosc. Biotechnol. Biochem. 62, 1005–1007. Osawa, R., Kuroiso, K., Goto, S., Shimizu, A., 2000. Isolation of tannin-degrading lactobacilli from human and fermented foods. Appl. Environ. Microbiol. 66, 3093–3097. Rodríguez, H., de las Rivas, B., Gómez-Cordovés, C., Muñoz, R., 2008a. Degradation of tannic acid by cell-free extracts of Lactobacillus plantarum. Food Chem. 107, 664–670.
Rodríguez, H., de las Rivas, B., Muñoz, R., Mancheño, J.M., 2007. Overexpression, purification, crystallization and preliminary structural studies of p-coumaric acid decarboxylase from Lactobacillus plantarum. Acta Cryst. F. 63, 300–303. Rodríguez, H., Landete, J.M., Curiel, J.A., de las Rivas, B., Mancheño, J.M., Muñoz, R., 2008b. Characterization of the p-coumaric acid decarboxylase from Lactobacillus plantarum CECT 748T. J. Agric. Food Chem. 56, 3068–3072. (doi:10.1021/jf703779s). Rodríguez, H., Landete, J.M., de las Rivas, B., Muñoz, R., 2008c. Metabolism of food phenolic acids by Lactobacillus plantarum CECT 748T. Food Chem. 107, 1393–1398. Romero, C., García, P., Brenes, M., García, A., Garrido, A., 2002. Phenolic compounds in natural black Spanish olive varieties. Eur. Food Res. Technol. 215, 489–496. Rozès, N., Peres, C., 1996. Effect of oleuropein and sodium chloride on viability and metabolism of Lactobacillus plantarum. Appl. Microbiol. Biotechnol. 45, 839–843. Rozès, N., Peres, C., 1998. Effects of phenolic compounds on the growth and the fatty acid composition of Lactobacillus plantarum. Appl. Microbiol. Biotechnol. 49, 108–111. Ruíz-Barba, J.L., Brenes-Balbuena, M., Jiménez-Díaz, R., GarcíaGarcía, P., Garrido-Fernández, A., 1993. Inhibition of Lactobacillus plantarum by polyphenols extracted from two different kinds of olive brine. J. Appl. Microbiol. 74, 15–19. Ruíz-Barba, J.L., Cathcart, D.P., Warner, P.J., Jiménez-Díaz., 1994. Use of Lactobacillus plantarum LPCO10, a bacteriocin producer, as a starter culture in Spanish-style green olive fermentations. Appl. Environ. Microbiol. 60, 2059–2064. Ruíz-Barba, J.L., Garrido-Fernández, A., Jiménez-Díaz, R., 1991. Bactericidal action of oleuropein extracted from green olives against Lactobacillus plantarum. Lett. Appl. Microbiol. 12, 65–68. Ruíz-Barba, J.L., Rios-Sánchez, R.M., Fedriani-Iriso, C., Olias, J.M., Rios, J.L., Jiménez-Díaz, R., 1990. Bactericidal effect of phenolic compounds from green olives on Lactobacillus plantarum. System. Appl. Microbiol. 13, 199–205. Ryan, D., Robards, K., 1998. Phenolic compounds in olives. Analyst 123, 31R–44R. Sestelo, A.B.F., Poza, M., Villa, T.G., 2004. β-Glucosidase activity in a Lactobacillus plantarum wine strain. World J. Microbiol. Biotech. 20, 633–637. Spano, G., Rinaldi, A., Ugliano, M., Moio, L., Beneduce, L., Massa, S., 2005. A β-glucosidase gene isolated from wine Lactobacillus plantarum is regulated by abiotic stresses. J. Appl. Microbiol. 98, 855–861. Vega Leal-Sánchez, M., Ruíz-Barba, J.L., Sánchez, A.H., Rejano, L., Jiménez-Díaz, R., Garrido, A., 2003. Fermentation profile and optimization of green olive fermentation using Lactobacillus plantarum LPCO10 as a starter culture. Food Microbiol. 20, 421–430. Whiting, G.C., 1975. Some biochemical and flavour aspects of lactic acid bacteria in ciders and other alcoholic beverages. In: Carr, J.G., Cutting, C.V., Whiting, G.C. (eds) Lactic Acid Bacteria in Beverages and Food. Academic Press, London, pp. 69–85. Whiting, G.C., Coggins, R.A., 1971. The role of quinate and shikimate in the metabolism of lactobacilli. Ant. Leeuw. 37, 33–49. Whiting, G.C., Coggins, R.A., 1974. A new nicotinamide-adenine dinucleotide-dependent hydroaromatic dehydrogenase of Lactobacillus plantarum and its role in formation of (⫺)t-3, t-4-dihydroxycyclohexane-c-1-carboxylate. Biochem. J. 141, 35–42.
Chapter 44
Microbial Colonization of Naturally Fermented Olives C.C. Tassou1, E.Z. Panagou2 and G.-J.E. Nychas2 1
National Agricultural Research Foundation, Institute of Technology of Agricultural Products, Lycovrissi, Greece Department of Food Science and Technology, Laboratory of Microbiology and Biotechnology of Foods, Agricultural University of Athens, Greece
2
44.1 INTRODUCTION Fermentation is one of the oldest food-processing/preservation technologies known to mankind which is considered as an important determining factor to control microbial growth, improve digestibility and nutritional value of food, and enhance food safety (Nout and Rombouts, 1992). Nowadays, fermentation processes associated with animal products (e.g., meat, fish and dairy produce) have become increasingly sophisticated, in contrast with the fermentation of vegetables, particularly olives, which remains craft-based and empirical. This is surprising as the economic importance of olives in the European Union is undisputable, with 38% of total world production originating from the member countries Spain, Italy and Greece. After harvesting, olives are transported to the factory, sorted, washed, and finally brined in a salt solution (Balatsouras, 1990). Currently, both green and black olives are fermented ‘naturally’ on an industrial scale, without the addition of lactic acid bacteria starter cultures. Natural (or spontaneous) fermentations typically result from the competitive activities of the indigenous microflora, together with a variety of contaminating microorganisms from fermentation vessels, pipelines, pumps, and other devices in contact with the olives and brine. Those microorganisms best adapted to the food substrate and to technical control parameters eventually dominate the process. Today, fermentation control is limited to the maintenance of the olive ecosystem (Spyropoulou et al., 2001). In general, the ecosystem in olives is influenced by (1) the microbial association of harvested olives in the field (Tassou, 1993), (2) a range of intrinsic factors inherent to olives such as pH, water activity, availability of nutrients diffusing from the tissue, the structure of olive skin, the levels of indigenous antimicrobial compounds (phenolic compounds and organic acids), and (3) extrinsic factors such as temperature, storage conditions, and type of packaging. These intrinsic and extrinsic factors can influence the rate Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
as well as the type of metabolism of the developing microbial association and its response to the external environment (Boddy and Wimpenny, 1992; Fleet, 1999). The colonization (e.g., the spatial distribution of microorganisms on olives) is of great importance since the epidermis of the fruits is covered by a thick cuticle, which not only restricts the flow of nutrients from the fruit into the brine, but also hinders the penetration of microorganisms from the brine into the product. The present review focuses on the topography of the population of yeasts and specific groups of bacteria, as well as on the effect of microbial colonization on various physicochemical characteristics of the fermented product.
44.2 MICROBIAL FLORA OF OLIVES Pseudomonads, Enterobacteriaceae, lactic acid bacteria and yeasts/fungi are the principal members of the microbial association isolated from olive fruits (Table 44.1). The various bacterial and yeast groups isolated from olive fruits are shown in Tables 44.2 and 44.3, respectively.
44.2.1 Sources of Olive Contamination The origin of this microbial flora can be traced back to olive leaves (Table 44.4) as well as on stored olives (Table 44.5) and physiological condition of the fruits (healthy or damaged olives) (Table 44.6). Pseudomonas savastanoi is a major microorganism, which has been studied in detail, causing the endemic disease, olive knot or tubercle, which occurs in most regions of the world where the olive tree (Olea europaea) is cultivated. The disease takes its name from the woody outgrowths found most frequently on young stems and on branches and twigs as a consequence of wound
397
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
398
SECTION | I Bacterial and Fungal and Other Microbial Aspects
TABLE 44.1 Mean values (log10 CFU g⫺1*), range (min/max) of microorganisms of olive microflora, and number of surveyed samples. Microorganism
Number of samples
Mean value
Range
1st Year Pseudomonads
31
6.08
4.0–8.50
Yeasts – Fungi
35
5.92
4.0–8.84
Lactic acid bacteria
26
5.03
3.6–7.35
Enterobacteriaceae
15
4.24
0.0–9.47
TABLE 44.2 Bacterial groups isolated from raw olives. Pseudomonas spp. savastanoi
mesenteroides
fluorescens
subsp. mesenteroides
aeruginosa
dextranicum
putida
Lactobacillus
luteola
delbrueckii
cepacia
helveticus
Bacillus
2nd Year Pseudomonads
42
1.05
0.0–5.63
Yeasts – Fungi
42
4.95
2.7–6.41
Lactic acid bacteria
42
3.04
0.0–6.37
Enterobacteriaceae
42
0.68
0.0–4.32
*
Colony-forming units per gram.
infections. Extensive studies show that knot development involves colonization of the injection site, development of lysogenous cavities, and finally abnormal enlargement and proliferation of the host cells starting from the bacterial tissue. The development pattern of hypertrophy and hyperplasia suggests that they are induced by a substance with cytokininlike activity, synthesized by Pseudomonas savastanoi (Surico et al., 1976). The disease is related to a reduction of olive yield and quality. Pseudomonas savastanoi causes a similar disease on other plants, including Fraxinus excelsior (ash), Nerium oleander and Ligustrum japonicum.
44.2.2 Microbial Flora of Olives Related to Olive Oil Production Verona and Valleggi (1949) isolated many microorganisms, primarily yeasts but also bacteria and molds to a lesser extent, from heaps of pits (endocarps) from olive oil mills. The Italian Picci (1959) isolated bacteria, yeasts and molds from the site of oviposition of Dacus oleae of olives (Table 44.5).
44.2.3 Microbial Flora of Olives Related to Fermentation Procedure The natural fermentation of black and green olives involves a complex microflora (Figure 44.1) of lactic acid bacteria,
Leuconostoc
acidophilus
subtilis
casei subsp. rhamnosus
megaterium
casei subsp. alactosus
Serratia marcescens
xylosus
Erwinia carotovora
plantarum
Klebsiella
curvatus
pneumoniae
fermentum
planticola
brevis
Acetobacter aceti
hilgardii
Arthrobacter globiformis
leichmannii
Xanthomonas campestris
coryniformis subsp. coryniformis
Gluconobacter oxydans
Aeromonas spp.
Enterococcus faecium
sobria
Zygomonas mobilis
hydrophila
Cellulomonas flavigena Bacteria isolated from raw olives. Adapted from Tassou (1993) and Panagou (2002).
yeasts and Gram-negative bacteria (Fernández Díez, 1983; Balatsouras, 1990; Özay and Borcakli, 1996). The population dynamics of selected microbial groups in the brine during fermentation is well established in the literature. At the onset, the microflora of the brine consists of lactic acid bacteria, Enterobacteriaceae, yeasts, and Pseudomonas spp. (Table 44.7). These groups have been reported previously for both green and black Conservolea olives (Balatsouras, 1990; Spyropoulou et al., 2001) and for Manzanilla, Gemlik and Edincik green and black olives cultivated in Spain and Turkey (Borcakli et al., 1993;
399
CHAPTER | 44 Microbial Colonization of Naturally Fermented Olives
TABLE 44.3 Yeasts isolated from olive fruits. Brettanomyces spp.
Leucosporidium spp.
Bacteria
anomalus Candida guilliermondi olea parapsilopsis var. intermedia tenuis utilis
Metschnikowa pulcherrima
Cryptococcus albidus var. albidus laurentii var. laurenti macerans
Rhodotorula aurantiaca glutinis var. glutinis mucilaginosa rubra
Debaryomyces hanseii nicotianae
Saccharomyces bayanus cerevisiae var. ellipsoideus chevalieri exiguus fermentati globosus italicus kluyveri oleaceus rosei rouxii uvarum
Endomycopsis vini
TABLE 44.4 Bacteria isolated from olive leaves in two different years.
Hanseniaspora osmophila uvarum valbyensis
Sporobolomyces roseus
Hansenula anomala var. anomala anomala var. schneggii holstii
Torulopsis candida glabrata holmi magnoliae stellata
1991
67.86
51.00
Pseudomonas Savastanoi subsp. Syringae
Schizosaccharomyces spp. octosporus
Distribution (%) 1979
delafieldi
0.11
fluorescens
1.06
aeruginosa
–
0.04
Bacillus spp.
–
0.29
subtilis
0.34
0.57
megaterium
4.02
3.80
1.34
0.81
Serratia marcescens Erwinia spp. carotovora
–
0.37
–
0.08
Pantoea agglomerans
8.50
Trichosporon pullulans
Kluyveromyces veronae
Pichia fermentans membranefaciens pinus polymorpha terricola vini var. vini
Yeast species isolated from olives (Tassou, 1993).
6.00
Klebsiella 1.40
pneumoniae planticola
–
– 1.20
1.45
Lactobacillus plantarum
Kloeckera apiculata corticis
3.90
2.80
Leuconostoc 1.12
dextranicum
–
mesenteroides subsp. dextarnicum
–
3.10
Acetobacter aceti
1.23
4.70
Arthrobacter globiformis
2.07
1.40
“Micrococcus luteus”
3.63
2.20
Xanthomonas compestris
3.35
6.70
Gluconobacter oxydans
–
4.30
Curtobacterium plantarum
–
2.20
Enterococcus faecium
–
1.20 (Continued)
400
SECTION | I Bacterial and Fungal and Other Microbial Aspects
TABLE 44.4 (Continued ) Bacteria
Distribution (%)
TABLE 44.5 The microbiology of olives; stored olives (a), oviposition sites of Dacus oleae (b), and heaps of pits (c). a
1979
1991
Clavibacter spp.
–
0.98
Molds
Micrococcus spp.
–
0.82
Cellulomonas flavigena
–
0.40
Zymomonas mobilis
–
0.30
Alcaligenes faecalis
–
0.27
Rhizopus Aspergillus flavus* glaucus niger* sydowi* terreus* versicolor* Fusarium Penicillium notatum* purpurogenum Alternaria
Bacterial species isolated from olive leaves (Tassou, 1993).
Ruiz-Barba et al., 1994; Özay and Borcakli, 1996; Durán Quintana et al., 1999). The presence of yeasts and bacteria on and within fermented olives has been studied by electron microscopy (Nychas et al., 2002). On the surface of the olive epidermis, areas with a coating of epicuticular wax provide a site of attachment for both yeasts and bacteria (Figure 44.2A). Very few microorganisms are present on the smooth epidermis (e.g. around stoma, Figure 44.2B) although it is possible that they are removed during sample preparation. Each stomal opening is plugged with a mixed microbial colony, primarily of yeasts of elongate morphology, but also bacteria (Figure 44.3A). The epidermis is easily detached from the mesocarp allowing the microbial association in the substomal cavity to be observed. Yeasts of both spherical and elongate form are present on the underside of the stomal opening which is rich in epicuticular wax (Figure 44.3A). The intercellular spaces of sub-stomal cells are packed with bacteria in colonies (Figure 44.3B). Microorganisms are not detected deeper in the olive mesocarp.
44.2.4 Biochemical Characteristics of Microbial Association 44.2.4.1 Effect on Olive Oil There is considerable information about the microflora, especially its lipolytic activity on waste products from olive extraction mills. This lipolytic activity, especially with respect to determinant effect on the olive oil, has been studied in detail. As yet no clear-cut conclusion can be drawn. Oil hydrolysis mediated by lipases and the resultant oxidation (rancidity) affects mainly the organoleptic properties of the final product. Olive fruit contains both a lipase and lipoxidase, the latter occurring in the endocarp (Tassou, 1993). According to this researcher, the increase in titratable acidity of olive oil during storage is attributed to enzymic action rather than autocatalysis. Concurrently,
b
c
Rhizopus nigricans* Actinomucor* Oospora* Aspergillus glaucus* Fusidium* Geotrichum candidum* Penicillium purpurescens* Alternaria tenuis* cheiranthi* Pullalaria pullulans* Cladosporium avellaneum* Fusarium vasinfectum*
nd
Saccharomyces cerevisiae* Pichia fermentans Debaryomyces kloeckeri* sybglobosus* Candida guillermondi var. membranefaciens* intermedia* parapsilopsis* parapsilopsis var.intermedia* pulcherrima* tenuis* Rhodotorula glutinis* mucilaginosa* Trichosporon behrendii Cryptococcus laurentii
Rhodotorula mucilaginosa* Candida lipolytica* guillermondii* parapsilopsis var. intermedia* Trichosporon
Yeasts Pichia fermentans membranefaciens Saccharomyces italicus elegans Candida crusei* parapsilopsis var.intermedia* Trichosporon sericum*
Bacteria Aerobacter Escherichia Serratia marcescens* plymuthicum* Pseudomonas aeruginosa* Achromobacter Bacillus subtilis megaterium cereus
Micrococcus roseus*
nd
flavus* Sarcina lutea* Bacillus cereus* brevis* megaterium* pumilus* subtilis*
Microflora isolated from different parts of olives. Adapted from Verona and Valleggi (1949), González-Cancho (1957a, b), and Picci (1959). * Lipolytic strains; nd: not detected.
CHAPTER | 44 Microbial Colonization of Naturally Fermented Olives
TABLE 44.6 The influence of the physiological condition of olives (damaged or healthy, green or black) and leaves on the microflora in two different years. Microorganisms Olives
Yeasts
Lactic acid bacteria
Pseudomonads
1st Year Black healthy Green healthy Black damaged Green damaged Leaves 5.08 (6) F-test*
5.9# (5&) 4.92 (4) 6.57 (10) 6.51 (7) 4.56 (5) ns
4.92 (4) 4.62 (5) 5.51 (8) 4.92 (2) 5.27 (6) ns
6.10 (4) 5.46 (3) 6.54 (10) 6.41 (7)
2nd Year Green healthy Geen damaged Leaves F-test
2.86 (22) 3.26 (17) 3.72 (15) ns
4.74 (19) 4.39 (23) 5.66 (15) **
Microbial association of olive leaves and fruits (Tassou, 1993). *: significant difference at 5% level. **: significant difference at 1% level. ns: not significant. #: log10 CFU g–1. &: number of samples surveyed.
Borbolla et al. (1958) could not verify the presence of a lipase or a lipoxidase in healthy olive. Consequently, they doubted the presence of these enzymes in nature and concluded that the lipolytic enzymes were derived from microorganisms which grew in the olive paste at various stages of processing or even in the oil produced thereafter. While the existence of endogenous lipase and lipoxidase in the olive fruit is disputed, it is a fact that the microorganisms growing on stored olives before processing, in the olive paste before extraction, or on olives attacked by the insect Dacus oleae, are commonly lipolytic as suggested by studies in Spain and Italy (Balatsouras, 1990). Lipolytic activity of microbial origin increases the acidity of oil produced from pits vis a vis the virgin material. The yeasts isolated by these researchers are listed in Table 44.5. In 1954 Verona isolated a new genus Trichosporon sp. from pit samples from olive oil factories. It produces lipase and split the olive oil to glycerine and free fatty acids. Tassou (1993) reviewed the microflora at various stages of oil processing. She reports that molds, bacteria and yeasts are mainly lipolytic, while the majority of molds isolated from heaps of pits produce lipase. GonzálezCancho (1957a) isolated molds, yeasts and bacteria from olives stored in heaps prior to oil production (Table 44.5). He notes that Serratia and Pseudomonas are the most lipolytic genera followed by Trichosporon and Pichia. The remaining microorganisms are lipase-negative. In another study, González-Cancho (1957b) found geographical
401
differences in the composition of the microflora and he isolated new species of bacteria, yeasts and molds. The lipolytic strains are given in Table 44.6. He concluded that the majority of the isolates produce lipase with the exception of Pichia fermentans, a variety of Cryptococcus laurentii, Trichosporon behrendii, and some species of Penicillium and Alternaria. Gracián et al. (1961) analyzed chemically samples from olive pits, the same heap being sampled at various times before processing. They determined the quantity and quality of the oil by physical and chemical standards and found that by the time hydrolysis of fatty substances had occurred, a decrease in weight had also occurred. Both are attributed to microorganisms growing in and on the surface of the heap, splitting the glycerides to glycerine and free fatty acids. The micro-organisms use the hydrolysis products for growth thereby causing not only deterioration of the quality but also loss in the yield of the oil. They also report an increase in the acidity of the oil and formation of hydroxyacids as well as other metabolites which are transferred eventually to the pitting oil. The softening of olive is attributed partly to endogenous and microbial pectinolytic enzymes. The presence of pectinolytic enzymes, pectin methylesterase and polygalacturonase, has been confirmed in certain olive varieties (Castillo-Gomez et al., 1978a,b; Mínguez Mosquera et al., 1978) as well as the presence of a natural polygalacturonase inhibitor (Castillo-Gomez et al., 1979). The types of pectic enzymes and the changes they cause in pectic substances during ripening of the fruit are important and must be considered to determine the appropriate time of harvest and processing of certain table olive cultivars. The pectinolytic activity of microorganisms isolated from fermentation brines has been studied because it causes spoilage during and after fermentation. The lipolytic enzymes (lipases) produced by many microorganisms on the raw product are found in the final product even after UHT sterilization, and consequently, they can cause spoilage of the sterile product. To act optimally the lipase needs an olive oil of pH 8.0–8.5 and a temperature of 40 °C (Adams and Brawley, 1981a,b). There is high probability also of these enzymes surviving processing during the macerating process of olives (carried out at 35–40 °C) before oil extraction. Thus they can be transferred to the olive oil where they exacerbate the development of oxidative rancidity. Lipases from Candida rugosa (Han and Rhee, 1985), Saccharomycopsis lipolytica and Micrococcus caseolyticus (Jonsson, 1976) can hydrolyze olive oil to a great extent. The hydrolysis by the lipase of S. lipolytica is highly specific for oleic acid while that of M. caseolyticus is totally non-specific (Jonsson, 1976). Olive oil stimulates growth of Achromobacter lipolytica thereby increasing lipase production and hydrolysis (Kahn et al., 1967). The same is true for Pseudomonas mephitica (Kosugi and Kamibayashi, 1971). Olive oil stimulates lipase production initially by enhancing
402
SECTION | I Bacterial and Fungal and Other Microbial Aspects
A
B
7
7
_ 6 5 4 pH
log cfu ml−1
6
3 5 2 1 0
4 0
8
16
24
32
40
48
56
Days
0
8
16
24
32
40
48
56
Days
FIGURE 44.1 Changes in (A) total viable counts (䊐), lactic acid bacteria (䊊), Pseudomonas spp. (△), Enterobacteriaceae (▽), yeasts (✫), and pH (B) in black olive fermentation. Population dynamics in olive fermentation (Nychas et al., 2002).
TABLE 44.7 Microbial association during different fermentation stages of olives. Lactic acid bacteria
Spoilage bacteria Primary stage
Streptococcus spp. Pediococcus spp. Leuconostoc mesenteroides Leuconostoc dextranicum
Aeromonas spp. Aerobacter spp. Enterobacter spp. Escherichia coli Pseudomonas spp. Bacillus polymyxa Bacillus macerans Clostridium spp. Flavobacterium spp.
Intermediate stage Leuconostoc mesenteroides Lactobacillus plantarum Final stage Lactobacillus plantarum Lactobacillus brevis Lactobacillus pentosus Lactobacillus buchneri
Candida spp. Pichia spp. Hansenula spp. Rhodotorula spp. Saccharomyces spp. Debaryomyces spp. Propionibacterium spp.
Microbial association in olive fermentation. Adapted from Vaughn et al. (1969a,b), Tassou (1993), Garrido Fernández et al. (1997), and Panagou (2002).
growth of Geotrichum candidum but eventually the release of glycerol from the triglycerides diminishes the production (Nelson, 1953; Wouters, 1967). Lipases of Pseudomonas fragi, Staphylococcus aureus, Aspergillus niger can
FIGURE 44.2 Micrograph of the microflora on the epidermis of naturally fermented black olives: (A) yeasts and bacteria form a biofilm in areas with epicuticular surface wax, (B) yeasts and bacteria in stomal aperture. Micrograph of olive epidermis (Nychas et al., 2002).
also hydrolyze olive oil (Nashif and Nelson, 1953; Alford and Pierce, 1961; Fukumoto et al., 1963). Small amounts of oleic acid activate, but higher concentrations inhibit, the lipase of Candida lipolytica (Ota et al., 1972).
CHAPTER | 44 Microbial Colonization of Naturally Fermented Olives
FIGURE 44.3 Micrograph of the microflora inside naturally-fermented black olives: (A) yeasts and bacteria below a stomal opening, (B) bacteria colonizing the spaces between sub-stomal cells. Microflora inside olives (Nychas et al., 2002).
44.2.4.2 Effect on Fermentation The main metabolic products in the brine during fermentation are citric, tartaric, malic, succinic, lactic and acetic acids (Nychas et al., 2002; Panagou, 2002). Their accumulation in the brine results in a steady pH decrease to a value of 4.4 or lower (Figure 44.1B). Although the presence of these acids has been reported in previous studies for green table olives (Durán Quintana et al., 1999; Spyropoulou et al., 2001), this is believed to be the first report of their occurrence in the brine of naturally fermented Greek-style black olives. Fermentation of black olives at high salt concentration is slow and further hampered by the availability of nutrients needed to initiate fermentation (Balatsouras, 1990). The olive cuticle acts as a barrier to the diffusion of solutes into and out of olive fruit, as there is no lye pre-treatment to disrupt the cuticle, thus exchange can only take place through apertures on the surface of the fruit (stomata). Substances that are leached from the olives include inorganic nutrients (minerals), sugars, pectic substances, sugar alcohols,
403
hormones, vitamins, alkaloids and phenolics (Balatsouras, 1990; Tassou, 1993). In the early stages of olive fruit development, gaseous exchange takes place through the stomata, but later on some become occluded with wax which is removed during the washing procedure before fermentation (Proietti et al., 1999). The stomata of olive leaves are considered to be the site of entry of bacterial suspensions applied as sprays (Surico, 1993), and it is possible that some of the microorganisms on ripe olives might be associated with the stomata as well as the skin surface. Epicuticular waxes, present as a bloom on the surface of ripe olives, are known to influence the wettability of plant leaves (Leben, 1988), and provide initially a site for microbial attachment and a substrate on which biofilms of yeasts and bacteria develop during fermentation. By the end of fermentation, yeasts and lactic acid bacteria are the dominant microorganisms in the brine. Our observations (Nychas et al., 2002) of the olives themselves show that there is a spatial differentiation between the two groups. Specifically, yeasts tend to predominate on the skin surface and right under the stomal openings, whereas bacteria predominate in the intercellular spaces of the sub-stomal cavities. Sampling of the brine alone will tend to underestimate the numbers of microorganisms present. Similarly, pH and organic acid levels in the brine may not be representative of the local conditions in the mesocarp. Vegetable tissue will buffer changes in pH in food materials, and the pH in the center of microbial colonies can be lower than at the edge (Boddy and Wimpenny, 1992; Malakar et al., 2000). Organic acid end-products are affected by the extrinsic and intrinsic factors such as oxygen limitation, glucose and salt concentration, low pH (Bobillo and Marshall, 1991, 1992; Spyropoulou et al., 2001) prevailing at the beginning and end of fermentation. Further research is needed to investigate the development and significance of the microbial association on and within both green and black olives during fermentation.
44.2.4.3 Dry Salted Olives An important type of black olives is the so-called ‘dry salted olives’. The fruits are harvested at full ripeness and they are processed by coarse salt without adding any brine. As a result of the high osmotic pressure exerted by the salt the fruits become gradually de-bittered, shriveled and edible. The final physicochemical characteristics of the product are: water activity (aw), 0.75–0.85; pH, 4.5–5.5; oil content, 35–40%; moisture, 30–35%; reducing sugars, 2.0–2.5%; NaCl content, 4–10% (Panagou, 2006). Despite the relatively high pH, compared with fermented olives in brine, the low water activity/high salt content can ensure its microbiological safety during storage, although potential spoilage microorganisms, mainly fungi, can grow at these conditions (Panagou et al.,
404
SECTION | I Bacterial and Fungal and Other Microbial Aspects
TABLE 44.8 Microbial association (log10 CFU g⫺1)* during dry salting of black olives. Microorganism
Dry salting period (days)
Total viable counts Lactic acid bacteria Yeasts Enterobacteria Pseudomonads
0
20
40
60
80
6.5 ⫾ 0.7 4.1 ⫾ 0.3 5.7 ⫾ 0.6 3.7 ⫾ 0.9 4.0 ⫾ 0.5
5.9 ⫾ 0.4 ⬍10 5.6 ⫾ 0.2 ⬍10 ⬍100
4.7 ⫾ 0.6 ⬍10 4.7 ⫾ 0.5 ⬍10 ⬍100
5.6 ⫾ 0.5 ⬍10 5.6 ⫾ 0.4 ⬍10 ⬍100
6.0 ⫾ 0.4 ⬍10 6.0 ⫾ 0.5 ⬍10 ⬍100
Microbial association in dry salting. Results are expressed as mean ⫾ SD, n ⫽ 3 (Panagou, 2006). *
Colony-forming units per gram.
1.00
9 8
0.95
NaCl (g/100 g)
0.90
6 5
0.85 4 0.80
3
Water activity (aw)
7
2 0.75 1 0.70
0 0
10
20
30
40
50
60
70
80
Dry salting period (days) FIGURE 44.4 Changes in water activity (•) and salt concentration (䉱) in black olives during dry salting. Physicochemical changes during dry salting (Panagou, 2006).
than 20 days due to the low water activity of the product at that period (Figure 44.4). The dominant yeast species is Candida famata. Study of the processed olives with electron microscope reveals the presence of a biofilm of yeasts covering almost the whole surface of olives (Figure 44.5). In a cross-section of the fruits, large intercellular spaces are observed under the epidermal cells (Figure 44.6). The development of these spaces is due to the advanced stage of maturity that the olives are harvested as well as the dry salting process. In this area an extensive network of fungal hyphae is observed that penetrates deeper into the parenchyma cells of the mesocarp at prolonged storage of olives (Figure 44.7). No fungal growth is observed on the surface of olives; instead extensive mycelium is evident in the mesocarp that is initially restricted in the intercellular spaces of parenchyma cells but finally penetrates and disintegrates the cells of olives. Fungal growth may negatively affect both the nutritional and aesthetic value of olives. More importantly, there is increasing evidence linking fungal growth and mycotoxin production in dry salted olives but further research is needed for this purpose.
SUMMARY POINTS ●
●
FIGURE 44.5 Micrograph of olive epidermis after dry salting with yeast biofilm on the surface. Surface of dry salted olives (Panagou et al., 2003).
●
●
2003). The initial microflora consist of lactic acid bacteria, yeasts, pseudomonads, and enterobacteria (Table 44.8). Within the first 20 days, yeasts become the dominant species whereas no other microbial group survives for more
●
Pseudomonads, lactic acid bacteria, yeasts, fungi and Enterobacteriaceae are members of the microbial association of olives. Leaves, healthy or damaged olives, and storage conditions influence the microbial association of olives. The main biochemical activity of this microbial flora is related to lipolysis and fermentation of available sugars. Initially, on the surface of olive epidermis, areas with a coating of epicuticular wax provide a site of attachment for both yeasts and bacteria. The stomata of olive leaves provide a site of entry of microorganisms.
CHAPTER | 44 Microbial Colonization of Naturally Fermented Olives
FIGURE 44.6 Micrograph of dry salted olives: (a) epidermal cells, (b) parenchyma cells, (c) fungal hyphae. Cross-section of dry salted olives (Panagou et al., 2003).
FIGURE 44.7 Micrograph of parenchyma cells of dry salted olives with fungal growth. Cross-section of dry salted olives (Panagou et al., 2003).
●
●
During fermentation yeasts tend to predominate on the skin surface and right under the stomal openings, whereas bacteria predominate in the intercellular spaces of the sub-stomal cavities. Extensive network of fungal hyphae is observed deeper into the parenchyma cells of the mesocarp at prolonged storage of dry salted olives.
REFERENCES Adams, D.M., Brawley, T.G., 1981a. Factors influencing the heat resistance of a heat resistant lipase of Pseudomonas. J. Food Sci. 46, 673–676. Adams, D.M., Brawley, T.G., 1981b. Factors influencing the activity of a heat resistant lipase of Pseudomonas. J. Food Sci. 46, 677–680. Alford, J.A., Pierce, D.A., 1961. Lipolytic activity of micro-organisms at low and intermediate temperatures. III. Activity of microbial lipases at temperatures below 0°C. J. Food Sci. 26, 318. Balatsouras, G.D., 1990. Edible olive cultivars, chemical composition of fruit, harvesting, transportation, processing, sorting and packaging,
405
styles of black olives, deterioration, quality standards, chemical analysis, nutritional and biological value of the end product. In “Olio d’oliva e olive da tavola: tecnologia e qualità”, pp. 291–330. Istituto Sperimentale per la Elaiotecnica, Italy. Bobillo, M., Marshall, V.M., 1991. Effect of salt and culture aeration on lactate and acetate production by Lactobacillus plantarum. Food Microbiol. 8, 153–160. Bobillo, M., Marshall, V.M., 1992. Effect of acidic pH and salt on acid endproducts by Lactobacillus plantarum in aerated, glucose–limited continuous culture. J. Appl. Bacteriol. 73, 67–70. Boddy, L., Wimpenny, J.W.T., 1992. Ecological concepts in food microbiology. J. Appl. Bacteriol. 73, 23S–38S. Borbolla y Alcalá, J.M.R., Gomez Herrera, C., González Cancho, F., Fernández Díez, M.J., 1958. La primera fase de fermentacion. Grasas Aceites 9, 118. Borcakli, M., Özay, G., Alperden, I., Özsan, E., Erdek, Y., 1993. Changes in chemical and microbiological composition of olive during fermentation. Grasas Aceites 44, 253–258. Castillo-Gomez, J., Mínguez Mosquera, M.I., Fernández Díez, M.J., 1978a. Presencia de poligalacturonasa y su relacion con el ablandamiento en algunos productos empleados en la industria del aderezo (pimientos y aceitunas). Grasas Aceites 29, 97. Castillo-Gómez, J., Míinguez Mosquera, M.I., Fernández Díez, M.J. 1978b., Presencia de poligalacturonasa en la aceituna negra madura. Factores que influencian la actividad de diclia enzima. Grasas Aceites 29, 333. Castillo-Gómez, J., Mínguez Mosquera, M.I., Cabrera Martin, J., Fernández Díez, M.J., 1979. Presencia de inhibitores de polygalacturonasa en la aceituna negra madura. Grasas Aceites 30, 11. Durán Quintana, M.C., García García, P., Garrido Fernández, A., 1999. Establishment of conditions for green table olive fermentation at low temperature. Int. J. Food Microbiol. 51, 133–143. Fernández Díez, M.J., 1983. Olives. In: Reed, G. (ed.), Food and Feed Production with Micro-Organisms. Verlag Chemie, Deerfield Beach, Fla, pp. 379–397. Fleet, G.H., 1999. Micro-organisms in food ecosystems. Int. J. Food Microbiol. 50, 101–118. Fukumoto, J., Iwai, M., Tsujisaka, Y., 1963. Studies on lipase. I. Purification and crystallization of a lipase secreted by Aspergillus niger. J. Gen. Appl. Microbiol. 9, 353. Garrido Fernández, A., Fernández Díez, M.J., Adams, M.R., 1997. Table Olives: Production and Processing. Chapman & Hall, London. González-Cancho, F., 1957a. Investigaciones sobre la conservacion de aceituna de molino. III. Poblacion microbiana de los trojes. Grasas Aceites 8, 55. González-Cancho, F., 1957b. Investigaciones sobre la conservacion de la aceituna de molino. IV. Poblacion microbiana de los trojes. Grasas Aceites 8, 258. Gracián, J., Arevalo, G., Albi, Fca., 1961. Alteraciones del orujo graso de aceituna durante el transcursode su almacenamiento. I. Transformaciones quimicas. Grasas Aceites 12, 174. Han, D., Rhee, J.S., 1985. Batchwise hydrolysis of olive oil by lipase in AOTisooctane reverse micelles. Korea Biotechnol. Lett. 7, 651. Jonsson, V., 1976. Rates of hydrolysis of olive oil, soybean oil, and linseed oil, by Saccharomycopsis lipolytica and Micrococcus caseolyticus (Food preservation). Chem. Microbiol. Technol. Lebensm. 4, 139. Kahn, I.M., Dill, C.W., Chandan, R.C., Shahani, K.M., 1967. Production and properties of the extracellular lipase of Achromobacter lipolyticum. Biochim. Biophys. Acta 132, 68.
406
SECTION | I Bacterial and Fungal and Other Microbial Aspects
Kosugi, Y., Kamibayashi, A., 1971. Thermostable lipase from Pseudomonas sp. Cultural conditions and properties of the crude enzyme. J. Ferment. Technol. 49, 968. Leben, C., 1988. Relative humidity and the survival of epiphytic bacteria on buds and leaves of cucumber plants. Phytopathol. 78, 179–185. Malakar, P., Brocklehurst, T.F., Mackie, A.R., Wilson, P.D.G., Zwietering, M.H., Van’t Riet, K., 2000. Microgradients in bacterial colonies: use of fluorescent ratio imaging, a non-invasive technique. Int. J. Food Microbiol. 56, 71–80. Mínguez Mosquera, M.I., Castillo Gómez, J., Fernández Díez, M.J., 1978. Presencia de pectinesterasa y su relacion con el ablandamiento en aglunos productos del aderezo. Grasas Aceites 29, 29. Nashif, S.A., Nelson, F.E., 1953. The lipase of Pseudomonas fragi. I. Characterization of the enzyme. J. Dairy Sci. 36, 459. Nelson, W.O., 1953. Nutritional factors influencing growth and lipase production by Geotrichum candidum. Ibid. 36, 143. Nout, M.J.R., Rombouts, F.M., 1992. Fermentative preservation of plant foods. J. Appl. Bacteriol. 73, 136S–147S. Nychas, G.-J.E., Panagou, E., Parker, M.L., Waldron, K.W., Tassou, C.C., 2002. Microbial colonization of naturally-black olives during fermentation and associated biochemical activities in the cover brine. Lett. Appl. Microbiol. 34, 173–177. Ota, Y., Nakamiya, T., Yamada, K., 1972. On the substrate specificity of the lipase produced by Candida paralipolytica. Agr. Biol. Chem. 36 (11), 1895. Özay, G., Borcakli, M., 1996. Effect of brine replacement and salt concentration on the fermentation of naturally black olives. Food Res. Int. 28, 553–559. Panagou, E.Z., 2006. Greek dry-salted olives: Monitoring the dry-salting process and subsequent physicochemical and microbiological profile during storage under different packing conditions at 4 and 20°C. Lebensm. Wiss. Technol. 39, 322–329. Panagou, E.Z., Parker, M.L., Katsaboxakis, K.Z., 2003. Study of the indigenous microflora of dry-salted olives of the Thassos variety using scanning electron microscopy. Agric. Res. 26, 3–10 (in Greek).
Panagou, E.Z., 2002. Fermentation, preservation, and microbial ecology of table olives. Ph.D. Thesis, Agricultural University of Athens, GR. Picci, G., 1959. Ancora sulla microflora presenta nelle olive colpite de Dacus oleae. Ann. Facol. Agr. 20, 65. Proietti, P., Famiani, F., Tombesi, A., 1999. Gas exchange in olive fruit. Photosynthetica 36, 423–432. Ruiz-Barba, J.L., Cathcart, D.P., Warner, P.J., Jiménez-Díaz, R., 1994. Use of Lactobacillus plantarum LPCO10, a bacteriocin producer, as a starter culture in Spanish style green olive fermentation. Appl. Environl. Microbiol. 60, 2059–2064. Spyropoulou, K.E., Chorianopoulos, N.G., Skandamis, P.N., Nychas, G.-J.E., 2001. Control of Escherichia coli O157:H7 during the fermentation of Spanish-style green table olives (conservolea variety) supplemented with different carbon sources. Int. J. Food Microbiol. 66, 3–11. Surico, G., Sparapano, L., Lerario, P., Durbin, R.D., Iacobellis, N., 1976. Studies on growth-promoting substances by Pseudomonas savastanoi. Agric. Consp. Sci. 39, 449–458. Surico, G., 1993. Scanning electron microscopy of olive and oleander leaves colonized by Pseudomonas syringae subsp. savastanoi. J. Phytopathol. 138, 31–40. Tassou, C.C., 1993. Microbiology of olives with emphasis on the antimicrobial activity of phenolic compounds. Ph.D. thesis, University of Bath, UK. Vaughn, R.H., Jakubczyk, T., MacMillan, J.D., Higgins, T.E., Dave, B.A., Crampton, V.M., 1969a. Some pink yeasts associated with softening of olives. Appl. Microbiol. 18, 771–775. Vaughn, R.H., King, A.D., Nagel, C.W., Ng, H., Levin, R.E., MacMillan, J.D., York, G.K., 1969b. Gram negative bacteria associated with sloughing, a softening of California ripe olives. J. Food Sci. 34, 224–227. Verona, O., Valleggi, M., 1949. Il problema della conservazione delle sanse de oliva. Olearia 3, 639. Wouters, J.T.M., 1967. The effect of Tweens on the lipolytic activity of Geotrichum candidum. Anton. van Leeuw. 33, 365.
Chapter 45
Occurrence of Aflatoxin B1 in the Greek Virgin Olive Oil: Estimation of the Daily Exposure Panagiota Markaki Laboratory of Food Chemistry, Department of Chemistry, University of Athens, Panepistimioupolis Zografou, Greece
45.1 INTRODUCTION 45.1.1 Aflatoxins Aflatoxins (AFs) are secondary metabolites of the molds Aspergillus flavus and Aspergillus parasiticus, which are able to contaminate, already in the field, food commodities such as seeds. Once the crop is contaminated, these fungi can produce AFs, which can enter the human and animal food chains, not only by the direct ingestion of contaminated seeds or processed foods, but also by the consumption of animal products (meat, milk coming from livestock fed with contaminated feeds). AFs are potent carcinogens, teratogens, genotoxics and mutagens and pose severe hazards to animal and human health. The most potent of the four naturally occurring aflatoxins (AFB1, AFG1, AFB2, AFG2) is aflatoxin B1 (AFB1) (Chu, 1991). The aflatoxins of the B group are bifurano coumarins fused to cyclopentanone and the G group is bifurano coumarins fused to lactone. The B group fluoresces blue in the long-wavelength ultraviolet light (UV), while the G group fluoresces green. The subscripts 1 and 2 show the chromatographic mobility (Rf values) pattern of the compounds on thin-layer chromatography (TLC) (Gourama and Bullerman, 1995). The mode of action of AFB1 induces metabolic activation, DNA modification, cell deregulation and cell death/ transformation (Eaton and Gallager, 1994). It has been previously shown that the metabolism plays an important role in the action of aflatoxins. AFB1 is activated by the cytochrome P-450 to an 8,9-epoxide before having exerted its carcinogenic effect. So AFB1 could either be converted to other hydroxylated metabolites and exerted or bind to DNA, RNA and proteins to manifest its toxic carcinogenic and mutagenic effect (Chu, 1991). On Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
the various types of aflatoxin toxicity, carcinogenicity is the most serious and the International Agency for Research on Cancer has classified AFB1 as a ‘human carcinogen’ (Group I) (IARC, 1993). The optimum temperature for aflatoxins’ production is generally from 25 °C to 30 °C and the maximum production depends on the strain of fungus and the used substrate (Diener and Davis, 1966). However, aflatoxins may be found in several commodities such as peanut butter, fig paste, paprika powder (Stroka et al., 2000), cooked food components of whole meals (Midio et al., 2001), meat, cheese and spices (Gurbuz et al., 2000), barley and cornbased foods (Park et al., 2004). Nevertheless, the exposure to low levels of aflatoxins occurs mostly through the consumption of maize and peanuts (Kumar et al., 2008). A number of approaches have been taken to detoxify aflatoxins; however, only a few have practical applications. Among these, ammonization is an effective way of reducing the aflatoxin content of a variety of foods (Marth and Doyle, 1979).
45.1.2 Olive Oil in the Mediterranean Diet Since antiquity olive oil is obtained from the fruit of the olive (Olea europaea) and is considered to have many medicinal properties. Recent data have suggested that the components in olive oil may have health benefits. Various researchers have shown that polyphenols or secoiridoids present in olive oil act as antihypertensives, antithrombotics and antioxidants (Manna et al., 2002). It has been reported that the lower incidence of coronary heart disease as well as prostate and colon cancers in Greece, Italy and Spain is due to the Mediterranean diet. The Mediterranean diet is largely vegetarian in its nature
407
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
408
SECTION | I Bacterial and Fungal and Other Microbial Aspects
and the consumption of olive oil is the principal source of fat. The amount of vegetable fat obtained via olive oil in Greece is 71% (Rose et al., 1986), while the amount of olive oil consumed is 18 kg per capita (Quaranta and Rotundo, 2000). This protection against cardiovascular diseases and cancer is due to its fatty acid profile and the presence of phenolic constituents (Visioli and Galli, 1998). The major phenolic compounds in olive oil are oleuropein, hydroxytyrosol and tyrosol; however, its phenolic content depends mainly on the production and storage of the oil (Brenes et al., 2001). Additionally, many authors investigated the antifungal (A. parasiticus and A. flavus included) and the antibacterial effect of some phenolic compounds (Aziz et al., 1998).
45.2 OLIVES AS SUBSTRATE FOR MOLD GROWTH AND AFLATOXIN B1 PRODUCTION 45.2.1 Aflatoxin B1 Production in Olives According to Samane et al. (1991) olives can be contaminated by molds including Aspergillus parasiticus. Moreover, spores of these molds can occur naturally on fresh and processed olives (Tantaoui-Elaraki, et al., 1983). Reports by several authors (Tantaoui-Elaraki et al., 1983; Mahjoub and Bullerman, 1987; Paster et al., 1988, Yassa, 1995; Eltem, 1996) showed that olives could support aflatoxin and other mycotoxins production as often they are stored for extensive periods in conditions favorable for the mold growth. Obviously inadequate storage conditions may allow the proliferation of toxigenic fungi (Pardo et al., 2005). Therefore, toxinogenesis is possible and leads to the contamination of olives by the mycotoxins and their possible transfer into olive oil (Samane et al., 1991). Yassa et al. (1994) proved that a total of 40 mold strains were isolated from black table olives produced in Egypt. Among them, nine strains of A. flavus and five strains of A. parasiticus were found to produce AFB1 in synthetic media and olive paste. Tantaoui-Elaraki and Marmioui (1996) stated that among all kinds of olives the black ‘Greek style’ is the most exposed to mold contamination. Additionally, Gourama and Bullermann (1988) reported that molds isolated from moldy ‘Greek-style’ black olives are potentially toxigenic. On the other hand, Leontopoulos et al. (2003) showed that black olives of Greek origin are not a favorable substrate for AFB1 biosynthesis at hazardous levels. They reported that the production of AFB1 in yeast extract sucrose medium (YES) on the third, ninth and fifteenth day of incubation (T ⫽ 30 °C) was ⬃1000-, 2500- and 10 000-fold higher, respectively, compared to the corresponding production in olives. Low AFB1 production in olives could be explained by the presence of antimicrobial constituents such as caffeic acid, catechin, coumarins
(Paster et al., 1988) and phenolic compounds (Visioli and Galli, 1998). Many authors have investigated the antifungal and the antibacterial effects of the phenolic compounds. Precisely, oleuropein reduced the production of AFB1 greatly (reduction 83–93%) (Gourama and Bullerman, 1987). Furthermore, Ghitakou et al. (2006) reported that the AFB1 production in two varieties of black olives, after inoculation by A. parasiticus, was not significantly higher compared to the control samples. In contrast, the AFB1 production in green olives was stimulated. Before distributing green olives for consumption, they must be heated or treated with lye (Mahjoub and Bullerman, 1987). Such a treatment stimulates the mold growth and the AFB1 production. Furthermore, the processing of green olives increases the permeability of the olive skin resulting in the afflux of fresh nutrients; therefore, the invasion by molds and the AFB1 production is likely easier (Ghitakou et al., 2006).
45.2.2 AFB1 Occurrence in Olives Olive samples were found to be contaminated with AFB1 at concentrations 0.7 μg AFB1 kg⫺1 (Ollé and Furon, 1988), 15–37 μg AFB1 kg⫺1 (Tantaoui-Elaraki et al., 1983). Investigation of the occurrence of AFB1 in 30 samples of olives and olive pasta from the Athens market showed that AFB1 was found at levels 0.15–1.13 ng AFB1 kg⫺1. Previously, AFB1 was also detected at low concentrations in olives used as control (Tantaoui-Elaraki and Marmioui, 1996). However, even when olives are kept under favorable conditions, they turn out to be a poor substrate for the mold growth and the AFB1 production, as compared to other natural substrates such as rice, peanuts and wheat (TantaouiElaraki et al., 1983). Nevertheless, the fact of the matter is that the AFB1 accumulation in olives may happen even at very low levels and a transfer into the olive oil is possible. Although such a transfer is limited, this may be harmful to the consumer, since the olive oil is not submitted to refining procedures which are known to eliminate aflatoxins (Parker and Melnick, 1996).
45.3 METHODOLOGICAL CONSIDERATIONS FOR THE DETERMINATION OF AFB1 IN OLIVE OIL In Table 45.1 the principal data concerning the analytical methods for the AFB1 determination in olive oil are shown. Olive oil is usually dissolved with hexane and then AFB1 is extracted by using methanol in water in various ratios (Le Tutour et al., 1983; Magnarini et al., 1990; Daradimos et al., 2000, Ferracane et al., 2007) followed by extraction with chloroform (Hagan and Tietjen, 1975). According to Cavaliere et al. (2007) aflatoxins were extracted from the sample by means of matrix solid-phase
CHAPTER | 45 Occurrence of Aflatoxin B1 in the Greek Virgin Olive Oil: Estimation of the Daily Exposure
409
TABLE 45.1 Methods for determination of AFB1 in olive oil. This table summarizes the principal steps and characteristics of the methods applied for the determination of the AFB1 in olive oil. (AFB1: aflatoxin B1). Methods
Analytical methods Extraction
Clean-up
Partitioning Quantitation
Recovery %
DLa (μg AFB1 kg⫺1)
Hagan and Tietjen (1975)
H2O ⫹ chloroform
1)TLCb Benzene ⫹ hexane (3 ⫹ 1) 2) Benzene ⫹ acetic acid (2 ⫹ 1)
TLC-UVc
–
5
Le Tutour et al. (1983)
1) Hexane 2) Methanol ⫹ H2O (60 ⫹ 40)
Lead-acetate
TLC-UV
90
4
Miller et al. (1985)
Dissolved with hexane
Silica column
TLC or LCd
89.5–93.5
5
Magnarini et al. (1990)
Methanol ⫹ H2O (60 ⫹ 40)
Easi-Extract (SPE)e
HPLCf
72–99.7
0.5
Daradimos et al. (2000) A
Methanol ⫹ H2O (60 ⫹ 40) ⫹ chloroform
Sep-Pak (SPE)
HPLC-FDg
87.2
2.8 ⫻ 10⫺3
Daradimos et al. (2000) B Papachristou and Markaki (2004)
Methanol ⫹ H2O (80 ⫹ 20)
Aflaprep (IAC)h
HPLC-FD
84.8
56 ⫻ 10⫺3
Ferracane et al. (2007)
Dissolved with hexanemethanol ⫹ H2O (60 ⫹ 40)
SPE
LC-MSi
74.6
0.25
Cavaliere et al. (2007)
MSPDEj
C18
LC/ESI-MS/MS
96–98
0.04
a
e
i
b
f
j
Detection limit; Thin-layer chromatography; c Ultraviolet; d Liquid chromatography;
Solid-phase extraction; High performance liquid chromatography; g Fluorescence detector; h Immunoaffinity columns;
dispersion (MSPDE), utilizing a column C18 as a dispersing material. Therefore, no further purification step for the lipid removal was necessary. The cleaning of the extract was performed by using a thin-layer chromatographic (TLC) clean-up development with benzene-hexane (3 ⫹ 1) followed by a second development in the same direction using toluene-ethyl acetate-formic acid (6 ⫹ 3⫹1) or benzeneacetic acid (9 ⫹ 1) (Hagan and Tietjen, 1975). Le Tutour et al. (1983) used lead acetate for cleaning, while the washing of the silica column with ether, toluene and chloroform was applied by Miller et al. (1985). The solid phase extraction (SPE) simplified the cleanup procedure and toxic materials such as lead acetate and
Mass spectrometry; Matrix solid-phase dispersion.
toluene were omitted (Daradimos et al., 2000; Ferracane et al., 2007). Moreover, lower quantities of solvents were used as compared to those that have been applied in silica columns. The cost of the immunoaffinity minicolumn (IAC) (Papachristou and Markaki, 2004) is higher than the silica gel SPE cartridges either Sep-Pak (Daradimos et al., 2000) or Easi-Extract (Magnarini et al., 1990). Alternatively, the saving in time, solvents and the elimination of false positives because of the antibody specificity, more than makes up for the difference in cost. According to Daradimos et al. (2000) both minicolumns (SPE cartridges and immuoaffinity columns) are satisfactory for
410
SECTION | I Bacterial and Fungal and Other Microbial Aspects
routine analysis and research purposes. However, the good precision, the low cost and mostly the very low detection limit (2.8 ⫻ 10⫺3 μg kg⫺1) make the SPE cartridges more convenient for working out problems associated with the presence of AFB1 in olive oil. On the other hand, immunoaffinity columns are very popular, especially when AFB1 is analyzed in complex food matrices. The recoveries, 90% (Le Tutour et al., 1983) and 89.5–93.3% (Miller et al., 1985), are very satisfactory when AFB1 is quantitated with TLC. On the contrary, the sensitivity of the TLC is low as compared to the results taken from the determination with high performance liquid chromatography (HPLC) (Table 45.1). The detection limits are less satisfactory, when a TLC with UV detection is performed for the AFB1 quantitation. HPLC with fluorescence detector (λex ⫽ 365 nm, λem ⫽ 425 nm) still remain the most popular method for the determination of the AFB1. The lowest detection limit (DL) (2.8 ⫻ 10⫺3 μg kg⫺1) was performed, when AFB1 was quantitated by HPLC (Daradimos et al., 2000). Recently, Ferracane et al. (2007) combined the liquid chromatography (LC) with mass spectrometry (MS) for the AFB1 confirmation. The recovery factor and the detection limit were found to be 74.6% and 0.25 μg kg⫺1 respectively.
45.4 AFB1 OCCURRENCE IN OLIVE OIL 45.4.1 AFB1 Occurrence in the Mediterranean Olive Oil In the literature the data concerning the occurrence of AFB1 in olive oil are limited. Previously, Toussaint et al. (1977) showed that olive oil samples originating from Greece and Spain were found to contain AFB1 at levels
of 5–10 μg kg⫺1. Furthermore, Gracian and Arevalo (1980) pointed out that the Spanish oil was contaminated with total aflatoxins at levels of 13.4–155.4 μg kg⫺1. Recently, Ferracane et al. (2007) reported that only three out of 30 samples of olive oil originated from Italy and Morocco were contaminated with AFB1 (range 0.54– 2.50 ⫻ 10⫺3 μg kg⫺1). In addition, in 20 olive oil samples, originating from Southern Italy cultivations, no aflatoxins were detected when analyzed. Regarding the 15 samples of Italian commercial virgin olive oil only three were found to be contaminated with AFB1 at levels below the limit of quantification (0.04 μg kg⫺1) (Cavaliere et al., 2007).
45.4.2 AFB1 Occurrence in Greek Virgin Olive Oil Regarding the occurrence of AFB1 in Greek virgin olive oil, 100 samples produced from 1995 to 2001 were examined. Fifty of them were offered from a Greek oil company. They originated from olive fruits collected from different areas of Greece (1995–1998). The remaining 50 samples of virgin olive oil produced between 1998 and 2001 were collected during 2001. Twenty-five of these samples were collected from the producers and 25 from the Athens market. Among them four samples were labeled as ‘Organic Agricultural Products’. The presence of AFB1 was revealed in 72% of the samples (n ⫽ 50) of olive oil of Greek origin produced between 1995 and 1998 (mean ⫽ 8.3 ng kg⫺1, median ⫽ 7.4 ng kg⫺1) (Table 45.2). Eleven samples were found to be contaminated with AFB1 at levels of 2.8–4.7 ⫻ 10⫺3 μg kg⫺1. In 20 samples the contamination was between 5.2 ⫻ 10⫺3 and 9.4 ⫻ 10⫺3 μg kg⫺1 and in four samples between
TABLE 45.2 AFB1 concentrations in olive oil samplesa of Greek origin. This table shows that the AFB1 levels of contamination in Greek olive oil are not correlated to the origin of its production (DL: Detection limit, AFB1: Aflatoxin B1). Range of contamination (μg AFB1 kg⫺1) Origin of samples
N
⬍DL
DL ⫺ 56 ⫻ 10⫺3b
Contamination %
Northern Greece
3
–
3
100.0
Western Greece
11
1
10
90.9
Eastern Greece
5
3
2
40.0
Central Greece
5
2
3
60.0
26
9
17
65.3
Southern Greece a
Samples produced between 1995–1998 and determined with the method in Daradimos et al. (2000) A (DL ⫽ 2.8 ⫻ 10⫺3 μg AFB1 kg⫺1). DL ⫽ 56 ⫻ 10⫺3 μg AFB1 kg⫺1 of the Methods in Daradimos et al. (2000) B and Papachristou and Markaki (2004).
b
CHAPTER | 45 Occurrence of Aflatoxin B1 in the Greek Virgin Olive Oil: Estimation of the Daily Exposure
13.7 ⫻ 10⫺3 and 15.7 ⫻ 10⫺3 μg kg⫺1. The most contaminated samples were two from Messini (Southern Greece) (46.3 ⫻ 10⫺3 μg AFB1 kg⫺1 and 13.8 ⫻ 10⫺3 μg AFB1 kg⫺1), two from Western Greece (Corfu 15.7 ⫻ 10⫺3 μg AFB1 kg⫺1 and Mesologgi 13.7 ⫻ 10⫺3 μg AFB1 kg⫺1) and one from the island Lesvos (Eastern Greece) (15.1 ⫻ 10⫺3 μg AFB1 kg⫺1). It must be mentioned that the DL of the method is very low (2.8 ⫻ 10⫺3 μg kg⫺1). Table 45.2 shows that there were no significant differences in AFB1 connected to the origin of olive oil. Since Greece is a small country it would be inappropriate to correlate the AFB1 production of Southern Greece with the AFB1 production of Northern Greece. On the other hand, it might be taken under consideration that Peloponnesus and Crete (both Southern Greece) are large producers of olive oil compared to other regions of Greece. Analysis of the 50 samples of virgin olive oil produced between 1998 and 2001 revealed the presence of AFB1 traces in only 11 samples. Only one sample from Crete (Southern Greece) was contaminated with 60 ⫻ 10⫺3 μg AFB1 kg⫺1 olive oil. In the remaining 38 samples AFB1 was not detectable. It must be mentioned that these olive oil samples were analyzed with a method, where the DL was 56 ⫻ 10⫺3 μg AFB1 kg⫺1 (Table 45.1) (Papachristou and Markaki, 2004). The results from the determination of the AFB1 in the Greek virgin olive oil indicate that in recent years the levels of contamination in olive oil of Greek origin have diminished. Subsequently, the possible transfer of AFB1 from olives into olive oil could be considered as negligible.
45.5 AFB1 DAILY EXPOSURE: RISK ASSESSMENT The daily intake of AFB1 through food depends on both the concentration in the food and the amount consumed. As long as there is no regulation related to the maximum level allowed for AFB1 in olive oil, in this study we have used the regulation applied by the Commission of the European Communities which has agreed with a regulation level of 2 μg kg⫺1 for AFB1 in food commodities for human consumption (European Economic Community Council, 1998). In addition, the European maximum level allowed for AFB1 in baby foods and cereal-based foods for infants and young children is 0.10 μg kg⫺1 (European Commission, 2004). It is evident that in Greece and in the other Mediterranean countries olive oil is consumed by all ages, even by children. AFB1 is a genotoxic carcinogen, thus the safety factors used for non-genotoxic carcinogens cannot be applied. Therefore, most agencies, including the Joint Expert Committee on Food Additives (JECFA) and the U.S. Food and Drug Administration (FDA) (1998) have not set a tolerable daily intake (TDI) for AFB1. Despite its uncertainty, a provisional maximum tolerable daily intake (PMTDI) of 1.0 ng AFB1 kg⫺1 bw per day for adults and children has been quoted (Kuiper-Goodman, 1998) for estimating the risk in Korea associated with the consumption of food contaminated with AFB1. Table 45.3 shows that the daily intake of AFB1 occurred in the samples, analyzed for all ages, reaches
TABLE 45.3 Daily intake of AFB1a (ng AFB1 kg⫺1 bw). This table shows the daily exposure to AFB1 detected in Greek olive oil, of consumers of all ages, weighing between 20 and 70 kg according to their diet. The occurrence of AFB1 in Greek olive oil is not hazardous for consumers of all ages (AFB1: aflatoxin B1; PMTDI: provisional maximum tolerable daily intake). Age
Children
Adolescent
Adults
Body weight (kg)
20
55
70
Consumptionb of olive oil (kg)
0.005–0.05
0.015–0.1
0.015–0.1
Median (7.4 ng AFB1 kg⫺1)c
1.8 ⫻ 10⫺3–1.8 ⫻ 10⫺2
2 ⫻ 10⫺3–1.3 ⫻ 10⫺2
1.5 ⫻ 10⫺3–1 ⫻ 10⫺2
Most contaminated (60 ng AFB1 kg⫺1)d
1.5 ⫻ 10⫺2–0.15
0.01–0.1
0.01–0.08
Contaminated with 2 μg kg⫺1e
–
0.5–3.6
0.4–2.8
Contaminated with 0.10 μg kg⫺1f
0.02–0.25
–
–
a
PMTDI: 1 ng AFB1 kg⫺1 bw per day; The estimated daily consumption by consumers of virgin olive oil; c Occurrence of AFB1 in virgin olive oil samples; d Olive oil originated from Crete; e Regulation maximum level for AFB1 in food commodities for human consumption; f Regulation maximum level for AFB1 in baby foods and cereal-based foods for infants and young children.
b
411
412
SECTION | I Bacterial and Fungal and Other Microbial Aspects
TABLE 45.4 Key features of risk assessment. This table lists the key facts of the Risk Assessment including general information about the toxicity of a substance, the significance and application of the PMTDI, the regulation in force for AFB1 and the risk from the consumption of Greek olive oil contaminated with AFB1 (AFB1: aflatoxin B1; PMTDI: provisional maximum tolerable daily intake). 1. The toxicity of a food substance depends on its concentration in the food, the amount consumed, the body weight, the age, the sex, the diet and the overall metabolism of the consumer 2. The PMTDI is an index of safety. It has been established for estimating the risk associated with the consumption of food contaminated with AFB1 3. AFB1 is a genotoxic carcinogen. Despite its uncertainty the in force reported PMTDI is 1.0 ng AFB1 kg⫺1 bw per day. It denotes the maximum tolerable daily exposure to AFB1 of the consumer, being connected to its body weight, and the daily total amount of the food consumed, contaminated with AFB1 4. The in force regulation level of AFB1 in food commodities for human consumption is 2 μg kg⫺1. The regulation for AFB1 in food destined for infants and young children is 0.10 μg kg⫺1 5. The AFB1 occurrence in the Greek olive oil produced from 1995–2001 is not hazardous for consumers of all ages
from 1.8 ⫻ 10⫺3 ng AFB1 kg⫺1 bw to 0.15 ng AFB1 kg⫺1 bw. These levels are from ⬃6.6-fold to ⬃555-fold lower from the PMTDI (1 ng AFB1 kg⫺1 bw). On the other hand, a consumption of 0.1 kg of olive oil contaminated with 2 μg kg⫺1 (regulation maximum level for AFB1 in food commodities for human consumption) displayed a daily intake of 3.6-fold and 2.8-fold higher for adolescents and adults, respectively. It must be pointed out that because of the frequent consumption even the low levels of contamination can pose a hazard to public health and that the virgin olive oil is the principal fat in the diet of Greeks (adults and children). The AFB1 occurrence in olive oil is directly correlated to the aflatoxin presence in olives in the oil-making industry. Molds do not grow easily in olives and even when they do toxinogenesis would not necessarily be very high (Samane et al., 1991). However, even low levels of AFB1 may contribute to the daily consumption since olive oil is a main component in the Mediterranean diet. See Table 45.4 for the risk assessment.
45.6 CONCLUSION Since the biosynthesis of AFB1 in food is a very complicated phenomenon, the removal of the mycotoxins would also be difficult, once food has been contaminated (Aziz et al., 1998). Hence, it is necessary to ensure that olives are stored and handled in a manner that will prevent the growth of aflatoxigenic mold and the production of AFB1. The conditions during the storage of olives can be controlled to a greater extent than in the field which means in general that it is easier to prevent fungal growth during storage. On the other hand, excessive intake of AFB1 via olive oil could occur in the case of some Greek consumers and other people
of the Mediterranean basin as well (depending on their diet). Therefore, it is necessary to apply rigorous regulation for AFB1 as well as for other mycotoxins in virgin olive oil, ensuring in this way a negligible daily intake of AFB1.
SUMMARY POINTS ●
●
●
●
●
●
●
Aflatoxin B1 (AFB1) is produced by the molds Aspergillus flavus and Aspergillus parasiticus which are able to contaminate food commodities. AFB1 is a potent carcinogen and poses human serious health hazards. Olives could support AFB1 production therefore its transfer into olive oil is possible. The recent analytical methods for the determination of AFB1 in olive oil are accurate, efficient, reliable and sensitive. In the recent years the occurrence of AFB1 in Greek olive oil as well as in olive oil of other Mediterranean countries is limited. Therefore the exposure to AFB1, present in olive oil, is not hazardous. Low levels of AFB1 may contribute to the daily consumption of the toxin since olive oil is a main component in the Mediterranean diet. It is necessary to apply rigorous regulation for the AFB1 in virgin olive oil in order to secure restricted AFB1 daily intake.
REFERENCES Aziz, N.H., Farag, S.E., Mousa, L.A.A., Abo-Zaid, M.A., 1998. Comparative antibacterial and antifungal effects of some phenolic compounds. Microbios 93, 43–54.
CHAPTER | 45 Occurrence of Aflatoxin B1 in the Greek Virgin Olive Oil: Estimation of the Daily Exposure
Brenes, M., Garcia, A., Garcia, P., Garrido, A., 2001. Acid hydrolysis of secoiridoid aglycons during storage of virgin olive oil. J. Agric. Food Chem. 49, 5609–5614. Cavaliere, C., Foglia, P., Guarino, C., Nazzari, M., Samperi, R., Laganá, A., 2007. Determination of aflatoxins in olive oil by liquid chromatography–tandem mass spectrometry. Anal. Chem. Acta 596, 141–148. Chu, F.S., 1991. Mycotoxins: food contamination, mechanism, carcinogenic potential and preventing measures. Mutat. Res. 259, 291–396. Daradimos, E., Markaki, P., Koupparis, M., 2000. Evaluation and validation of two fluorometric HPLC methods for the determination of aflatoxin B1 in olive oil. Food Addit. Contam. 17, 65–73. Diener, U.L., Davis, N.D., 1966. Aflatoxin production by isolates of Aspergilus flavus. Phytopathology 56, 1390–1393. Eaton, D.L., Gallager, E.P., 1994. Mechanisms of aflatoxin carcinogenesis. Ann. Rev. Pharmacol. Toxicol. 34, 135–136. Eltem, R., 1996. Growth and aflatoxin B1 production on olives and olive paste by molds isolated from Turkish-style natural black olives in brine. Int. J. Food Microbiol. 32, 217–223. European Commission, Commission Regulation (ER) No 683/2004 of 13 April 2004 Amending Regulation (EC) No 466/2001 as regards aflatoxins and ochratoxin A in foods for infants and young children 2004. Off. J. Eur. Union 2, 3–5. European Economic Community Council, Commission Regulation (EC) No 1525/981998. Off. J. Eur. Comm. 201, 43–46. Ferracane, R., Tafuri, A., Logieco, A., Galvano, F., Balzano, D., Ritieni, A., 2007. Simultaneous determination of aflatoxin B1 and ochratoxin A and their natural occurrence in Mediterranean virgin olive oil. Food Addit. Contam. 24, 173–180. Ghitakou, S., Koutras, K., Kanellou, E., Markaki, P., 2006. Study of aflatoxin B1 production and ochratoxin A production by natural microflora and Aspergillus parasiticus in black and green olives of Greek origin. Food Microbiol 23, 612–621. Gourama, H., Bullerman, L.B., 1987. Effects of oleuropein on growth and aflatoxin production by Aspergillus parasiticus NRRL 299. Lebens Wiss. Tech. 20, 226–228. Gourama, H., Bullerman, L.B., 1988. Mycotoxin production by molds isolated from “Greek style” black olives. I. Food. Microbiol. 6, 81–90. Gourama, H., Bullerman, L.B., 1995. Aspergillus flavus and Aspergillus parasiticus: Aflatoxigenic fungi of concern in food and feeds: A review. J. Food Prot. 1395–1404. Gracian, J., Arevalo, G., 1980. Presencia de aflatoxinas en los productos del olivar. Grasas y Acetas 39, 167–171. Gurbuz, U., Nizmlioglu, M., Nizamlioglu, F., Dinc, I., Dogruer, Y., 2000. Examination of meat, cheeses and spices for aflatoxins B1 and M1. Veterinarium 10, 34–41. Hagan, S.N., Tietjen, W.H., 1975. A convenient thin layer chromatographic clean-up procedure for screening several mycotoxin in oils. Assoc. Offic. Anal. Chem. 58, 620–621. IARC (International Agency for Research on Cancer), 1993. Monographs on the evaluation of the carcinogenic risk of chemicals to humans: Some naturally occurring substances. Food items and constituents. Heterocyclic Aromatic Amines and Mycotoxins (Lyon: IARC), pp 397–444. Joint FAO/WHO Expert Committee on Food Additives, 1998. Forty-ninth Meeting of the Joint FAO/WHO Expert Committee on Food Additives. Safety Evaluation of Certain Food Additives and Contaminants in Food: Aflatoxins. WHO Food Additives Series, 40 (Geneva: WHO), pp 359–469. Kuiper-Goodman, T., 1998. Food safety: mycotoxins and phycotoxins in perspective. In: Miraglia, M., van Egmond, H.P., Brera, C.,
413
Gilbert, J. (eds) Mycotoxins and Phycotoxins–Developments in Chemistry. Toxicology and Food Safety. Ft Collins, Alaken, pp. 25–48. Kumar, V., Basu, M.S., Rajendran, T.P., 2008. Mycotoxin research and mycoflora in some commercially important agricultural commodities. Crop. Prot. 27, 891–905. Le Tutour, B., Tantaoui-Elaraki, A., Inlah, L., 1983. Simultaneous detection of aflatoxin B1 and ochratoxin A in olive oil. J. A.O.C.S. 60, 835–837. Leontopoulos, D., Siafaka, A., Markaki, P., 2003. Black olives as substrate for Aspergillus parasiticus and aflatoxin B1 production. Food Microbiol 20, 119–126. Mahjoub, A., Bullerman, L., 1987. Mold growth and aflatoxin production on whole olives and olive pastes. Sci. Aliments 7, 629–636. Magnarini, C., Mezzetti, T., Cossignani, L., Cecchetti, V., Cubbiotti, C., Santinelle, F., Burini, G., Damiani, P., 1990. RP HPLC determination of aflatoxins in edible oils. Rassegna Chemica 42, 273–276. Manna, C., D’Angelo, S., Migliardi, V., Loffredi, E., Mazzoni, O., Morrica, P., Galletti, P., Zappia, V., 2002. Protective effect of the phenolic fraction from virgin olive oils against oxidative stress in human cells. J. Agric. Food Chem. 50, 6521–6526. Marth, E.H., Doyle, M.P., 1979. Update on molds: degradation of aflatoxin. Food Technol. 33, 81–87. Midio, A.F., Campos, R.R., Sabino, M., 2001. Occurrence of aflatoxins B1, B2, G1 and G2 in cooked food components of whole meals marketed in fast food outlets of the city of Sao Paulo, SP, Brazil. Food Addit. Contam. 18, 445–448. Miller, N., Pretorius, H.E., Trinder, D.W., 1985. Determination of aflatoxins in vegetable oils. J. Assoc. Off. Anal. Chem. 68, 136–139. Ollé, M., Furon, D., 1988. Aspects récents de l’ analyse des huiles vierges. Rev. Fr. Corps Gras. 35, 63–65. Papachristou, A., Markaki, P., 2004. Determinaiton of ochratoxin A in virgin olive oils of Greek origin by immunoaffinity column clean-up and highperformance liquid chromatography. Food Addit. Contam. 21, 85–92. Pardo, E., Ramos, A.J., Sanchis, V., Marin, S., 2005. Modelling of effects of water activity and temperature on germination and growth of ochratoxigenic isolates of Aspergillus occhraceus on a green coffeebased medium. Int. J. Food Microbiol. 98, 1–9. Park, J.W., Kim, E.K., Kim, Y.B., 2004. Estimation of the daily exposure of Koreans to aflatoxin B1 through food consumption. Food Addit. Contam. 21, 70–75. Parker, W.A., Melnick, D., 1996. Absence of aflatoxins from refined vegetable oils. J. Am. Oil Chem. Soc. 43, 635–638. Paster, N., Juven, B.J., Harshemesh, H., 1988. Antimicrobial activity and inhibition of aflatoxin B1 formation by olive plant tissue constituents. J. Appl. Bacteriol. 64, 293–297. Quaranta, G., Rotundo, V., 2000. Economic and commercial prospects for olive oil in view of the changes in the common market organisation (CMO) (Part one). Olivae 91, 20–24. Rose, D.P., Boyar, A.P., Wynder, E.L., 1986. International comparison of mortality rates for cancer of the breast, ovary, prostate, and colon and per capita food consumption. Cancer 58, 2363–2371. Samane, S., Tantaoui-Elaraki, A., Essadaoui, M., 1991. Mycoflora of Moroccan “Greek style” black olives II toxigenesis. Microbiol. Alim. Nutr. 9, 335–352. Stroka, J., Anklam, E., Jörissen, U., Gilbert, J., 2000. Immunoaffinity column cleanup with liquid chromatography using post-column bromination for determination of aflatoxins in peanut butter, pistachio paste, fig paste and paprika powder: collaborative study. J. Assoc. Offic. Anal. Chem. Intern. 2, 320–340.
414
SECTION | I Bacterial and Fungal and Other Microbial Aspects
Tantaoui-Elaraki, A., Le Tutour, M., Bouzid, M., Keddani, M.J., 1983. Contamination des olives noirs “façon Grèce” par les sporeds d’ Aspergillus toxinogenes et leurs toxins. Ind. Alim. Agric. Cahier Sci. Tech. 100, 997–1000. Tantaoui-Elaraki, A., Marmioui, A., 1996. Antifungal treatment trial of “Greek style” black olives. Microbiol. Alim.. Nutr. 14, 5–14. Toussaint, G., Lafaverges, F., Walker, E.A., 1977. The use of high pressure liquid chromatography for determination of aflatoxin in olive oil. Arch. Inst. Pasteur, Tunis 3–4, 325–334.
Visioli, F., Galli, C., 1998. The effect of minor constituents of olive oil on cardiovascular disease: New findings. Nutrition Rev. 56 (5 Pt I), 142–147. Yassa, I.A., Abdala, E.A.M., Aziz, S.Y., 1994. Aflatoxin B1 production by molds isolated from black olives. Ann. Agric. Sci. (Cairo) 39, 525–537. Yassa, I.A., 1995. Some factors affecting mold growth and aflatoxin production in olives. Ann. Agric. Sci. (Cairo) 40, 59–65.
Chapter 46
Fenthion, Dimethoate and Other Pesticides in Olive Oils of Organic and Conventional Cultivation Aristidis M. Tsatsakis and Ioannis N. Tsakiris Centre of Toxicology Science and Research, Department of Medicine, University of Crete, Greece
46.1 INTRODUCTION Olive oil is of great economic importance for the Mediterranean countries. According to a FAO statistical survey, 2 459 000 tonnes of olive oil were produced in the European Union (of 12 members), during the production period 2003–2004. Fifty-eight per cent was produced in Spain, 24% in Italy and 16% in Greece, on an average producer’s price of 2.7 Euros per kg. Greece also holds a prominent position, third, in world olive oil production. It is noteworthy that up to 2007, olive oil production in Greece brought in a fair return in approximately 700 000 families. A fact of great significance is that 80% of Greek olive oil production is characterized as extra virgin olive oil. The great majority of the olive oil trees in Greece lie in southern latitudes. The prefecture of Iraklion is one of the most important olive oil producing areas (http://www.minagric. gr/greek/press/2007/06/g kostianis speech 5 6 07.doc). The total organic area in the European Union (of 15 members), fully converted and under conversion is 4.9 million hectares, representing 3.8% of the total Utilised Agriculture Area (UAA). The UAA area in Greece is 3 583 190 hectares, while 77 120 from these are the total organic area. In Northern Europe forage plants and cereal crops seem to be the most important organic crops. In Southern Europe the influence of olive groves is significant, especially in Greece. According to most recent statistics, the relative shares of total organic area for main food crops for olive plantations, other than fodder grass, in Greece represents approximately 38%, which is the uppermost percentage compared with other Mediterranean countries (Rohner-Thielen, 2005). Forty-six active ingredients are registered up to now in Greece for applications in olives (http://www.minagric.gr/ syspest/syspest_crops.aspx). Twenty-five are insecticides, 11 are fungicides, four are pesticides, while six are classified in other groups. The most important insect which Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
infects olives is Dacus olea. The damage caused by Dacus olea as reported, was not only quantitative, that is, fall of olives before harvest and reduction of olive flesh, but was also qualitative, as the infection caused an increase in olive oil acidity, denaturation of organoleptic characteristics and frowy taste of olive. By local practice, fenthion and dimethoate were the most widespread crop protection products used against Dacus olea. Fenthion was used from June to August, while dimethoate was used from September to October. Nowadays though, only nine formulations of dimethoate in the form of emulsifiable concentrates are commercially available. Fenthion shall not be included as an active substance in Annex I to Directive 91/414/EEC, while all authorizations for crop protection products containing fenthion were withdrawn from 11 August 2004. Only Greece, Spain, Italy and Portugal were allowed by the European Commission to maintain in force authorizations until 30 June 2007, to the effect that usage of plant protection products with fenthion within those countries was permissible until 31/12/2007 (European Commission, 2004). The preharvest interval for dimethoate is 21 days for coverage applications and 15 days for bail applications. The formulations of dimethoate are classified as a dangerous irritant to humans. The formulation of fenthion was classified as dangerous for humans. Suicide attempts and intoxications during spray applications have been reported by use of those two pesticides (Tsatsakis et al., 2002). Organic production pursues three general objectives: (a) establish a sustainable management system for agriculture, (b) aim at producing products of high quality, (c) aim at producing a wide variety of food and other agricultural products that respond to consumer demand for goods, produced by methods and processes which are not detrimental to humans, animals and plants and harmful to the environment and welfare at large. Council regulation (EEC) 2092/91 on organic production of agricultural products and indications
415
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
416
referring thereto on agricultural products and foodstuffs is about to be replaced according to Council regulation 834/2007 EC (European Commission, 2007). As stated at the 834/2007 EC, organic processed products should be produced by the use of processing methods which guarantee that organic integrity and the vital qualities of the product are maintained through all stages of the production chain. Nowadays, organic cultivation in Greece is being validated from Control and Certification private bodies accredited by the National Accreditation System. Control and Certification private bodies are responsible for the accurate implementation of this system, in compliance to European directives. These certification organizations are authorized from AgroCert® (Organization for Certification and Inspection of Agricultural Products), which is a Private Law Legal Entity operating for the public benefit under the supervision of the Ministry of Rural Development and Food. Similar structure is applied (or operative) in the rest of the European countries. The features and the overall principles of organic olives cultivation are presented in Table 46.1. According to regulation 396/2005 EC, where MRLs (maximum residue limits) were not set out for processed food, those provided for the relevant product shall be applicable (European Commission, 2005). Certification organizations for organic oil demand practically zero levels (lower than 0.001 mg kg⫺1). This is not easy to achieve, due to possible bad agricultural or bad industrial practice (Fontcuberta et al., 2008). Drift from neighboring cultivations, contamination from under- and groundwater, contamination from pesticide residues on ground, transfer from stone mill during extraction procedure and contamination
TABLE 46.1 Overall principles of organic olives cultivation. This table lists the overall principles of organic olives cultivation including the design of process, the use of external inputs, the limitations in chemical inputs and the adaption of rules where necessary. Principle 1. Appropriate design and management of biological processes based on ecological systems using natural resources which are internal to the system 2. Restriction of the use of external inputs 3. Strict limitation of the use of chemically synthesized inputs to exceptional cases 4. Adaptation where necessary of the rules of organic production taking account of sanitary status, regional differences in climate and local conditions, stages of development and specific husbandry practices
SECTION | I
Pesticides and Adulterants
from washing water, are only some of the critical points and likely perils which should be controlled, in order to keep biological olive oil free from residues. Hereby, the status of pesticide residues in olive oil is presented, focusing on fenthion and dimethoate residue data from olive oil samples from organic cultivation and conventional cultivation. Our data are based: 1. on the results of our laboratory during the years 1999– 2001 and 2005 2. on the results reported by certification organizations during the period 2006–2008 3. on the results cited in recent and current bibliography during the period 2000–2008. These results will be used so as to assess the relationship between the quality of organic and conventional olive oil and the levels of selected pesticide residues. Assessment of the factors that affect the levels of pesticide residues in olive oil during the cultivation and production procedures is a further objective of this chapter.
46.2 CRITICAL POINTS FOR THE DETERMINATION OF PESTICIDE RESIDUES IN OLIVE OIL Olive oil is indeed the most difficult vegetable oil to analyze, compared to sunflower oil, corn oil and soybean oil. This is not only because of the relatively high amount of lipids that elute from the clean-up system, but also because of the potential lipid interference at the GC determinate step (Lentza-Rizos, 1994). Thus, during the analytical procedure the clean-up stage plays an important role. The fate of pesticide residues during oil extraction is closely related to the fat-solubility of the compounds under consideration. The transfer factor is used as a measure of concentration. This factor is the quotient of concentration of the active ingredient in the oil divided by the concentration of the active ingredient in the olives. The transfer factor (TF) for dimethoate is 0.03 and for fenthion is 3.3 or 5.20 (Farris et al., 1992). When the TF is higher than 1, there is evidence that concentration of the residues is taking place. Another problem during olive oil analysis for pesticide residues is the fact that multiresidues are highly desirable, although the different natures, classes and physicochemical properties of pesticides used in olive groves eventually hamper the development of such methodologies (GarciaReyes et al., 2007). The method that we used for the determination of dimethoate and fenthion residues in olive oil samples is based on liquid–liquid and solid-phase extraction with subsequent gas chromatography and mass spectrometric analysis (Kidd and James 1991; Lentza-Rizos, 1994). Figure 46.1 shows the structures of dimethoate, fenthion, endosulfan and a-cypermethrin.
CHAPTER | 46 Fenthion, Dimethoate and Other Pesticides in Olive Oils of Organic and Conventional Cultivation
417
S S
H N
S
P
O O
S
P O
O Dimethoate
Cl
Fenthion
Cl
O
O
CH3
Cl
S
O O
Cl
O
H3C
CH3
CH3 COO
Cl Cl Endosulfan
CH CN
O
a-Cypermethrin
FIGURE 46.1 Structure of dimethoate, fenthion, endosulfan and a-Cypermethrin. This figure shows the chemical structure of dimethoate, fenthion, endosulfan and a-Cypermethrin which are the most frequently detected pesticides in olive oil samples.
46.3 CROP PROTECTION PRODUCTS (PESTICIDES) RESIDUES IN OLIVE OIL FROM ORGANIC AND CONVENTIONAL CULTIVATION 46.3.1 Analytical Data for Pesticide Residues in Olive Oil Originated from Biological Cultivations in Greece given by Control Certification Bodies (Bio-Hellas) During 2006–2008 These results were obtained from Bio-Hellas monitoring program (http://www.bio-hellas.gr). Bio-Hellas is the largest private Control and Certification Body in Greece which supervises the most important olive oil production areas such as Crete and Peloponnesus. The monitoring for pesticide residues in olive oil is a crucial part of the inspection system. Thus, the usage of crop protection products in organic cultivation is prohibited. Accordingly, the selection of the specific crop protection products for residual analysis was based on criteria dealing mostly with the conventional cultivation of olives. The criteria used in this case were: (1) the directives for the applications of National Authorities (Crop Protection Centers); (2) the specific directives for the applications of supervising agronomists; (3) local crop protection practice according to which application was decided depending the season; and (4) the factor of potential danger, due to neighboring cultivations. Monitoring was performed for the following, most frequently used in conventional cultivation, pesticides: Chlorpyriphos, Cyfluthrin, a-Cypermethrin, λ-Cyhalothrin,
Deltamethrin, Diazinon, Dimethoate, Endosulfan (Endosulfan-a, Endosulfan-b, Endosulfan sulfate), Fenthion (Fenthion oxon, Fenthion sulfone, Fenthion sulfoxide, Fenthion o sulfone, Fenthion o sulfoxide), Malathion/Malaoxon, Methidathion, Methomyl, Parathion. From these only a-Cypermethrin, λ-Cyhalothrin, Deltamethrin, Dimethoate, Fenthion and Methidathion are authorized for application in cultivation of olives. Endosulfan, Malathion and Methomyl authorizations are to be withdrawn by the end of 2008, while the authorization of Parathion has been withdrawn since 2003. Chlorpyrifos, Cufluthrin and Diazinon are authorized for other cultivations. During the period 2006–2008, some 597 samples were analyzed. The distribution of samples among these years was 267 in 2006, 164 in 2007 and 166 in 2008. The olive oil samples were analyzed in line with the National Accreditation System SA ESYD laboratories (EN ISO/IEC 17025:2005) in accordance with standard official analytical methods. Table 46.2 presents the main results for the period 2006–2008. Cufluthrin, λ-Cyhalothrin, Deltamethrin, Fenthion o-sulfone, Fenthion o-sulfoxide, Malathion, Methomyl and Parathion were not detected in the years 2006–2008. Methidathion and Dimethoate were detected only once in 2008 and 2006, respectively. Figure 46.2 shows the percentage of olive oil samples with detectable concentrations. Endosulfan appears to be the most frequently detected plant protection product followed by Fenthion, a-Cypermethrin and Chlorpyrifos. It is obvious that the levels of pesticide residues decreased from year to year. Data were subject to statistical analysis so as to define if there was statistically significant difference between the values of analyzed pesticide
418
SECTION | I
Pesticides and Adulterants
TABLE 46.2 Pesticide mean value and concentration range in mg kg⫺1 and number of positive samples for pesticides in olive oil samples originated from biological cultivations in Greece during 2006 (number of samples 267), 2007 (number of samples 164) and 2008 (number of samples 166) as given by the Control Certification Body, Bio-Hellas. This table presents the main results for pesticide residue in organic olive oil during 2006–2008 originating from the Bio-Hellas monitoring program. Mean value (mg kg⫺1)
Pesticide
Concentration range (mg kg⫺1)
Positive samples
2006
2007
2008
2006
2007
2008
2006
2007
2008
Chlorpyriphos
0.03
0.02
0.04
0.01–0.15
0.01–0.05
0.02–0.06
9
5
2
Cyfluthrin
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
a-Cypermethrin
0.04
0.03
0.09
0.02–0.05
0.01–0.11
0.02–0.2
2
12
3
l-Cyhalothrin
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Deltametrin
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Diazinon
0.01
0.01
–
0.01
0.01–0.02
0.02
2
3
1
Dimethoate
–
n.d.
n.d.
0.02
n.d.
n.d.
1
n.d.
n.d.
α-Endosulfan
0.03
0.01
0.01
0.01–0.06
0.01
0.01
4
2
1
β-Endosulfan
–
n.d.
–
0.01
n.d.
0.02
1
n.d.
1
Endosulfan sulfate
0.03
0.04
0.03
0.01–0.22
0.01–0.16
0.02–0.07
100
28
9
Endosulfan total
0.03
0.04
0.03
0.01–0.24
0.01–0.17
0.02–0.08
100
28
9
Fenthion oxon
–
–
n.d.
0.02
0.06
n.d.
1
1
n.d.
Fenthion sulfone
0.01
–
n.d.
0.01
0.01
n.d.
2
2
n.d.
Fenthion sulfoxide
0.04
0.01
–
0.01–0.16
0.01
0.02
7
6
1
Fenthion-osulfone
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Fenthion-osulfoxide
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Fenthion
0.04
0.05
0.02
0.01–0.25
0.01–0.30
0.01–0.03
30
15
4
Fenthion total
0.04
0.05
0.02
0.01–0.43
0.01–0.31
0.01–0.03
32
16
4
Malathion
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Methidathion
n.d.
n.d.
–
n.d.
n.d.
0.02
n.d.
n.d.
1
Methomyl
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Parathion
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.: not detected
CHAPTER | 46 Fenthion, Dimethoate and Other Pesticides in Olive Oils of Organic and Conventional Cultivation
419
40,00 35,00 30,00 25,00 20,00 15,00 2006
10,00
2006 2006
5,00
C
a-
py r or hl C
C
ip h y os yp flut er hri λ- me n C yh thri D alo n el ta thri m n et D hrin i En D azi do im non su eth lfa oa n t En su e do lfa t En sul e do fan Fe sul -a nt fan h Fe ion -b nt hi tota on l to Fe Fen tal Fe n t n th hio Fe thio ion n ox nt n Fe hio sul on nt n s ph h u o Fe ion lph ne nt O ox id h s M ion ulp e al h o o at hi sul ne on ph /M on M ala e et ox hi da on t M hio et n ho Pa my ra l th io n
0,00
FIGURE 46.2 Percentage of virgin olive oil samples with detectable concentrations of pesticides originated from biological cultivations in Greece during 2006–2008 (data attained by the Control Certification Body – Bio-Hellas). This figure showed which pesticides are detected in olive oil samples from biological cultivations in Greece, certified by Bio-Hellas, during 2006–2008 and the percentage of detection.
TABLE 46.3 Number and percentage of Greek biological olive oil samples with at least 1, 2 and 3 residual pesticides during the years 2006–2008. This table presents the number and the percentage of the organic olive oil samples with at least one, two and three residual pesticides originating from the Bio-Hellas monitoring program during 2006– 2008 considering all the isomers and metabolites as one active ingredient in cases of Endosulfan and Fenthion. 2006
2007
2008
N
P
N
P
N
P
Total of positive samples
124
46.44
51
31.10
17
10.24
Samples with 1 pesticide
104
38.95
37
22.56
12
7.23
Samples with 2 pesticides
20
7.49
13
7.93
4
2.41
Samples with 3 pesticides
0
1
0.6
1
0.6
0
N: number of samples; P: percentage of samples with residual pesticides.
residues in sequential years. SPSS was used for the statistical analysis and Pearson Chi-square was used for the estimation of statistical significant deference. Differences at p ⬍ 0.001 were considered as significant. According to these results there was a statistically significant difference between the following:
● ●
The levels of endosulfan in sequential years The levels of fenthion in sequential years.
Table 46.3 depicts the number and the percentage of biological olive oil samples with at least one, two and three residual pesticides considering all the isomers and metabolites as one active ingredient in cases of Endosulfan and Fenthion.
420
SECTION | I
46.3.2 An Overview of Studies for Residual Pesticides in Olive Oil Samples from Greece, Italy and Spain During 2000–2008 As far as the olive oil analysis for pesticide residues is concerned, the great majority of published papers during recent years focus mainly on new methods (Garcia-Reyes et al., 2007). Monitoring studies of pesticide residues in olive oil are few. Furthermore, these studies concentrate on specific categories of compounds and mostly on organophosphates and chlorinated organic compounds (Lentza-Rizos et al., 2001; Rasterlli et al., 2002; Dugo et al., 2005; Guardia-Rubio et al., 2006; Sanchez et al., 2006; Tsoutsi et al., 2006; Fontcuberta et al., 2008). Table 46.4 summarizes the results of monitoring studies in Greece from 2000 up to 2008 while Tables 46.5 and 46.6 summarize the results of monitoring studies from Italy and Spain for the same years, respectively. Fenthion was detected in every monitoring program and thus, it seems to be the most frequently detected crop protection product with a percentage of detection range from 14% up to 27%. Dimethoate was detected in fewer cases but with higher percentage of detection, ranging from 33% up to 40%. As far as chlorinated organic compounds are concerned, endosulfan seems to be the most frequently detected one and its residues occurred almost exclusively as sulfate metabolite. From pyrethrins, λ-cyhalothrin and Cypermethrin seem to be the most frequently detected, yet with a percentage of detection lower than 1%. Comparing the above results with results asserted by studies previous to the year 2000, we may conclude that organophosphates and mainly fenthion and dimethoate are the most frequently detected crop protection products in olive oil samples (Hiskia et al., 1998). Furthermore, the fact that samples in these studies were originated only from conventional cultivations is very important indeed.
46.3.3 Fenthion and Dimethoate Pesticides in Olive Oil from Organic and Conventional Cultivation in Crete, Greece Residues of fenthion and dimethoate were determined in organic and conventional virgin olive oil samples collected from Crete during 1997–1999 (Tsatsakis et al., 2003). According to these results, there was a statistically significant difference between the following: ●
●
●
●
Values of fenthion residues in organic and conventional olive oil samples in the same years. Values of dimethoate residues in organic and conventional olive oil samples in the same years. Values of fenthion residues in organic olive oil samples in sequential years. Values of dimethoate residues in organic olive oil samples in sequential years.
Pesticides and Adulterants
TABLE 46.4 Pesticide mean value and concentration range in mg kg⫺1 and percentage of detection for pesticides in olive oil samples from conventional cultivations in Greece (derived from the studies of Lentza-Rizos et al., 2006 (number of samples 338) and Tsoutsi et al., 2006 (number of samples 30)). This table summarizes the results from three different monitoring studies in conventional olive oil in Greece from 2000 up to 2008. Pesticide
Mean value (mg kg⫺1)
Concentration range (mg kg⫺1)
Percentage of detection
Bromophos methyl*
n.d.
n.d.
–
Cypermethrin
0.04
0.04
0.29
Deltamethrin
n.d.
n.d.
–
n.d.
n.d.
–
Dimethoate*
0.033
0.022–0.044
33.33
Endosulfan sulfate
0.03
0.02–0.57
22.00
a-Endosulfan
n.r
n.r
1.48
b-Endosulfan
n.r
n.r
2.07
Ethion*
0.051
0.011–0.093
6
Fenitrothion*
n.d.
n.d.
–
Fenthion*
0.093
0.055–0.13
20.00
Fenthion sulfone*
n.d.
n.d.
–
Fenthion sulfoxide*
0.093
0.055–0.13
2
Fenvalerate
n.d.
n.d.
–
Malaoxon*
0.022
0.022
3.33
n.d.
n.d.
–
Methidathion*
n.d.
n.d.
–
Omethoate*
0.033
0.022–0.044
6.66
Parathion ethyl*
n.d.
n.d.
–
Permethrin
n.d.
n.d.
–
λ-Cyhalothrin
0.02
0.02
0.29
Diazinon
*
Malathion
*
n.d.: not detected ; n.r: not reported ; Pesticides marked with * reported in Tsoutsi et al. (2006); pesticides without* reported in Lentza-Rizos et al. (2001).
CHAPTER | 46 Fenthion, Dimethoate and Other Pesticides in Olive Oils of Organic and Conventional Cultivation
TABLE 46.5 Pesticide mean value and concentration range in mg kg⫺1 and percentage of detection for pesticides in olive oil samples from conventional cultivations in Italy (derived from the studies of Rasterlli et al., 2002 (number of samples 65) and Dugo et al., 2005 (number of samples 79)). This table summarizes the results from two different monitoring studies in conventional olive oil in Italy from 2000 up to 2008. Pesticide
Mean Concentration value range (mg kg⫺1) (mg kg⫺1)
Percentage of detection
Acephate
n.d.
n.d.
–
Azinphos-ethyl
0.090
0.080–0.100
3.08
Carbophonothion
n.d.
n.d.
Chlorpyrifos
n.d.
Chlropyrifosmethyl
TABLE 46.6 Pesticide mean value and concentration range in mg kg⫺1 and percentage of detection for pesticides in olive oil samples from conventional cultivations in Spain (derived from the studies of Sanchez et al. (2006) (number of samples 26), GuardiaRubio et al. (2006) (number of samples 24), Fontcuberta et al. (2008) (number of samples 31)). This table summarizes the results from three different monitoring studies in conventional olive oil in Spain from 2000 up to 2008. Pesticide
Mean value (mg kg⫺1)
Concentration Percentage range of detection (mg kg⫺1)
a-Cypermethrin 0.026
0.017–0.076
52
–
a-Endosulfan
n.d.
n.d.
–
n.d.
–
a-Endosulfan**
n.d.
n.d.
–
0.080
0.05–0.090
6.15
Aldrin**
n.d.
n.d.
–
Diazinon
0.083
0.064–0.101
4.62
b-Endosulfan
0.017
0.009–0.039
24
Dimethoate
0.061
0.030–0.120
44.62
b-Endosulfan**
n.d.
n.d.
–
Dimethoate*
n.d.
n.d.
–
Carbaryl
n.d.
n.d.
–
Fenthion
0.073
0.055–0.085
27.69
n.d.
n.d.
–
Fenthion*
0.20
0.09–0.42
13.7
Chlordanane (oxy, alpha, gamma)**
Fenthion*
0.35
0.18-0.73
21.4
Chlorpyrifos
0.024
0.023–0.024
8
Formothion
0.082
0.082
1.54
n.d.
–
n.d.
n.d.
–
Chlorpyrifos methyl
n.d.
Malaoxon Malathion
n.d.
n.d.
–
DDTs**
n.d.
n.d.
–
Methamidophos
n.d.
n.d.
–
Deltamethrin
0.017
0.009–0.029
36
Methidathion
0.063
0.051–0.085
4.62
Dieldrin**
n.d.
n.d.
–
Omethoate
n.d.
n.d.
– Diflufenican
0.025
0.009–0.098
40
Paraoxon-methyl
n.d.
n.d.
–
Parathion
0.080
0.060–0.100
3.08
Dimethoate
n.d.
n.d.
–
Parathion-ethyl*
n.d.
n.d.
–
Diuron
0.052
0.010–0.107
96
Parathion-ethyl*
n.d.
n.d.
–
Diuron*
0.062
0.021–0.180
66.66
0.056
0.056
1.54
0.041
0.018–0.091
100
Parathion-methyl
n.d.
n.d.
–
Endosulfan sulfate
Pirimiphosmethyl
n.d.
n.d.
–
Endosulfan sulfate*
0.042
0.019–0.074
37.5
Endosulfan sulfate**
0.03
0.03
3.23
Parathion-methyl *
n.d.: not detected ; n.r: not reported in study; Pesticides marked with * reported in Rastrelli et al. (2006); Pesticides without * reported in Dugo et al. (2005).
421
(Continued)
422
SECTION | I
TABLE 46.6 (Continued) Pesticide
Mean value (mg kg⫺1)
Concentration Percentage of detection range (mg kg⫺1)
Endrin**
n.d.
n.d.
–
Fenoxycarb
0.013
0.013
4
Formothion
n.d.
n.d.
–
HCB**
n.d.
n.d.
–
HCHs**
n.d.
n.d.
–
Heptachlor/ Heptachlor epoxide**
n.d.
n.d.
–
Malathion
n.d.
n.d.
–
Methidathion
n.d.
n.d.
–
Oxyfluorfen
0.029
0.012–0.066
76
Parathion ethyl
n.d.
n.d.
–
Parathion methyl
n.d.
n.d.
–
Phosmet
0.015
0.015
4
Pirimiphos methyl
n.d.
n.d.
–
Promecarb
n.d.
n.d.
–
Simazine
0.120
0.120
4
Terbuthylazine
0.076
0.005–0.190
92
Terbuthylazine*
0.092
0.021–0.196
87.5
Terbutryn
0.086
0.080–0.130
16
Trichlorfon
0.011
0.011
4
λ-Cyhalothrin
0.011
0.010–0.126
8
n.d.: not detected ; Pesticides marked with * reported in GuardiaRubio et al. (2006); Pesticides without * reported in Sanchez et al. (2006); Pesticides marked with ** reported in Fontcuberta et al. (2008).
Furthermore, there was no statistical difference between the levels of fenthion residues in conventional olive oil samples in sequential years and the levels of dimethoate residues in conventional olive oil samples in sequential years. The average concentrations of fenthion in conventional olive oils were 0.12, 0.15 and 0.17 mg kg⫺1 and for dimethoate were 0.02, 0.03 and 0.03 mg kg⫺1 for 1997, 1998 and
Pesticides and Adulterants
1999, respectively. The average concentrations of fenthion in organic olive oils were 0.02, 0.01 and 0.003 mg kg⫺1 for 1997, 1998 and 1999 while for dimethoate they were 0.009, 0.004 and 0.001 mg kg⫺1, respectively. A new monitoring study for the same plant protection products for the year 2005 was performed in Crete. Forty-nine samples originated from conventional and 52 from biological cultivations. The results are presented in Table 46.7. The average concentration of fenthion in virgin olive oil samples from conventional cultivation was 0.13 mg kg⫺1 and for dimethoate 0.02 mg kg⫺1. The average concentration of fenthion in virgin olive oil samples from biological cultivations was 0.007 mg kg⫺1 and for dimethoate 0.003 mg kg⫺1. A statistical significant difference was observed between the values of dimethoate and fenthion residues in organic and conventional olive oil samples.
46.4 SYNOPSIS OF DATA PERTAINING TO PESTICIDES IN OLIVE OIL Olive oil is a very important component of the Greek diet, also known as the Mediterranean diet, while a high olive oil content in the diet is thought to be cardioprotective (Kafatos et al., 2000). Olive oil consumption figures based on household budget surveys showed that the olive oil consumption in Greece was indeed high, especially in rural areas, compared with olive oil consumption in the rest of the European Union. The European Union is the world’s leading consumer of olive oil. It accounts for an average 71.5% of the total world consumption. Italy, Spain and Greece alone account for more than 85% of the European Union total consumption (http://www.internationaloliveoil. org/downloads/consommation1_ang.PDF). The great majority of studies for residual pesticides in olive oil samples from Greece, Italy and Spain are concerned and correlate mostly with new analytical methods and their validation. In these studies the number of olive oil samples analyzed amounts from 20 up to 80 with an exception of one study which reported a number of 338 samples. Moreover, results reported from Greece, Italy and Spain indicate the detection of specific plant-protection products both in conventional and biological cultivations of olives but with different frequency and severity of detection. Appraising the findings in the analysis of conventional olive oil samples, it is observed that, in most cases, samples with levels of pesticide residues lower than the MRLs are considered as safe for consumers. In samples originating from biological cultivations, the levels of pesticide residues were always lower than MRLs. Detection of plant-protection products in organic olive oil samples is forbidden. In general, the presented results show how difficult it is to derive conclusions about pesticide residues among several countries, even when controlling for specific food groups and chemical compounds.
CHAPTER | 46 Fenthion, Dimethoate and Other Pesticides in Olive Oils of Organic and Conventional Cultivation
423
TABLE 46.7 Dimethoate and Fenthion mean value and concentration range and the detection percentage for residual pesticide concentrations in olive oil samples from organic and conventional cultivation in Crete during 2005 (according to unpublished data from Centre of Toxicology Science and Research, Department of Medicine, University of Crete, Heraklion, Greece). Fifty-two organic olive oil samples and 49 conventional olive oil samples from Crete were monitored during 2005 for Dimethoate and Fenthion from Centre of Toxicology Science and Research. The results presented in this table. Pesticide
Cultivation system
Mean value (mg kg⫺1)
N
Concentration range (mg kg⫺1)
Dimethoate
SOC
0.003
52
0.001–0.005
9.61
SCC
0.020
49
0.006–0.030
24.48
SOC
0.007
52
0.002–0.018
5.76
SCC
0.130
49
0.030–0.420
16.32
Fenthion
Percentage of detection
N: number of samples; SOC: samples of olive oil from organic cultivation; SCC: samples of olive oil from conventional cultivation.
SUMMARY POINTS ●
●
●
●
●
●
The most common organophosphorous plant protection products were dimethoate and fenthion. From chlorinated organic compounds endosulfan was most frequently detected. From pyrethrins, a-Cypermethrin and λ-Cyhalothrin were the most important (considerable). Drift is always a potential danger for the organic cultivation. In this cultivation system, protection measures were always taken so as to avoid drift from neighboring fields. Despite this, however, it was practically impossible to eliminate the effect of this factor. The hypothesis that the residues in organic oils arise from the contamination of the olives because of the drift is usually true, yet of low importance. In drift cases, data in relation to the existence of neighboring cultivations, the local application, common and frequent practice and knowledge of the weather conditions in the area, are essential in order to conclude for residues origin. Bad agricultural practice from a small fraction of organic farmers is also a potential danger for the system of biological cultivation. Thus, the criteria for sample collection include knowledge about the farmer’s customary agricultural practices and compliance with supervising agronomist counsels’ directives and instructions. Prevention is always better than cure. Agronomists and farmers should be aware of and pay attention to the cultivation techniques in order to eliminate the factors that contribute towards the creation of conditions that favor insect infestations. Nitrogenous fertilization and irrigation should be well balanced and great care should be taken especially during hot and moist years where insect infestation is very likely to occur.
●
●
●
●
●
Some contamination in organic olive oil samples may conclude as a consequence of bad industrial practice. Abiding properly to the new council regulation 834/2007 EC for biological cultivations and organic processed products will efficiently eliminate the impact of this specific factor on organic olive oil contamination with pesticide residues. The quality of water used for washing olives and the improper cleaning of machines are some of the critical control points which may be the cause of contamination of organic oil with pesticides. One of the great advantages of biological cultivation compared to conventional cultivation is the continuous monitoring applied for pesticide residues in combination with supervising agronomists’ strict controls and regular inspections. Bio-Hellas results from monitoring organic olive oil samples demonstrate that the levels of endosulfan and fenthion residues in sequential years denote a statistically significant difference (decrease) due to the aforementioned practice ensued. Same results with statistically significant difference are also reported in repeated monitoring studies in Crete for dimethoate and fenthion residues in organic olive. Furthermore, in this case the results indicate a statistically significant difference between dimethoate and fenthion residues in biological and conventional cultivations at the same period. Compared with conventional cultivation, organic cultivation has the advantage of olive oil production free from pesticide residues and thus a reduction in environmental pollution. Based on the results from Greek monitoring studies, fenthion was detected in higher concentrations and in higher frequency compared with dimethoate. This is probably due to higher water solubility of dimethoate. Because of this, a large fraction of active ingredient
424
●
●
●
SECTION | I
is withdrawn at the water phase during olive oil production. The higher frequency of fenthion detection suggests that it was usually preferred for plant protection, as, actually, fenthion usage was legal until 31/12/2007. Monitoring olive oil for pesticide residues is essential in order to control and restrict erroneous, bad agricultural as well as industrial practice. Full traceability of olive oil samples is crucial in order to scrutinize and detect samples with levels of pesticide residues above MRLs.
ACKNOWLEDGMENT The authors acknowledge BIO-HELLAS Control Certification Body for providing the data regarding the monitoring program biological olive oil residues.
REFERENCES Dugo, G., Di Bella, G., La Torre, L., Saita, M., 2005. Rapid GC-FPD determination of organophosphorus pesticide residues in Sicilian and Apulian olive oil. Food Control 16, 435–438. European Commission, 2004. Commission decision concerning the noninclusion of fenthion in Annex I to Council Directive 91/414/EEC and the withdrawal of authorizations for plant protection products containing this active substance. Official Journal, EC 313/2004, L46/32. European Commission, 2005. Regulation of the European Parliament and of the Council on maximum residues levels of pesticides in or on food and feed of plant and animal origin and amending Council Directive 91/414/EEC. Official Journal, EC 396/2005, L70/1. European Commission, 2007. Council Regulation on organic production and labeling of organic products and repealing Regulation (EEC) 2092/91. Official Journal, EC 834/2007, L189/1. Farris, G.A., Cabras, P., Spanedda, L., 1992. Pesticide residues in food processing. Ital. J. Food Sci. 3, 149–170. Fontcuberta, M., Arques, J.F., Villaldi, J.R., Martinez, M., Centrich, F., Serrahima, E., Pineda, L., Duran, J., Casas, C., 2008. Chlorinated organic pesticides in marketed food: Barcelona, 2001–2006. Sci. Total Environ. 389, 52–57. Garcia-Reyes, J.F., Ferres, C., Gomez-Ramos, M.J., Molina-Diaz, A., Fernandez-Alba, A.R., 2007. Determination of pesticide residues in olive oil and olives. Trends Anal. Chem. 26 (3), 239–251.
Pesticides and Adulterants
Guardia-Rubio, M., De Cordova, F.M.L., Ayora-Canada, M.J., RuizMedina, A., 2006. Simplified pesticide pesticide multiresidue analysis in virgin olive oil by gas chromatography with thermionic specific, electron-capture and mass spectrometric detection. J. Chromatog. A. 1108, 231–239. Hiskia, E.A., Atmajidou, E.M., Tsipi, F.D., 1998. Determination of organophosphorus pesticide residues in Greek virgin olive oil by gas chromatography. J. Agric. Food Chem. 46, 570–574. http://www.internationaloliveoil.org/downloads/consommation1_ang.PDF http://www.minagric.gr/greek/press/2007/06/g kostianis speech 5 6 07.doc http://www.minagric.gr/syspest/syspest_crops.aspx Kafatos, A., Verhagen, H., Moschandreas, J., Apostolaki, I., Van Westerop, J.J., 2000. Mediterranean diet of Crete: foods and nutrient content. J. Am. Diet. Assoc. 100 (12), 1487–1493. Kidd, H., James, D.R., 1991. The Agrochemicals Handbook, 3rd edn. Royal Society of Chemistry Information Services, Cambridge, UK pp. 5–14. Lentza-Rizos, Ch., 1994. Monitoring of pesticide residues in olive products: Organophosphorus insecticides in olives and oil. J. AOAC Int. 77, 5. Lentza-Rizos, Ch., Avramides, E.J., 1995. Pesticides residues in olive oil. Rev. Environ. Contam. Toxicol. 144, 111–134. Lentza-Rizos, Ch., Avramides, E.J., Visi, E., 2001. Determination of residues of endosulfan and five pyrethroid insecticides in virgin olive oil using gas chromatography with electron-capture detection. J. Chromatogr. A. 921, 297–304. Rasterlli, L., Totaro, K., De Simone, F., 2002. Determination of organophosphorus pesticide residues in Cilento (Campania, Italy) virgin olive oil by capillary gas chromatography. Food Chem. 79, 303–305. Rohner-Thielen, E., 2005. Statistics in focus: Organic farming in Europe. In: 4% of Utilised Agricultural Area Devoted to Organic Farming. Eurostat European Communities, pp. 2–5. Sanchez, A.G., Martos, N.R., Ballesteros, E., 2006. Multiresidue analysis of pesticides in olive by gel permeation chromatography followed by gas chromatography-tandem mass-spectometric determination. Anal. Chim. Acta 558, 53–61. Tsatsakis, A.M., Bertsias, G.K., Liakou, V., Mammas, I.N., Stiakakis, I., Tzanakakis, G.N., 2002. Severe fenthion intoxications due to ingestion and inhalation with survival outcome. Hum. Exp. Tox. 21, 49–54. Tsatsakis, A.M., Tsakiris, I.N., Tzatzarakis, M.N., Agourakis, Z.B., Tutudaki, M., Alegakis, A.K., 2003. Three-year study of fenthion and dimethoate pesticides in olive oil from organic and conventional cultivation. Food Addit. Contams. 20 (6), 553–559. Tsoutsi, C., Konstantinou, I., Hela, D., Albanis, T., 2006. Screening method for organophosphorus insecticides and their metabolites in olive oil samples based on headspace solid-phase microextracion coupled with gas chromatography. Anal. Chim. Acta 573–574, 216–222.
Chapter 47
Residues of Pesticides and Polycyclic Aromatic Hydrocarbons in Olive and Olive-Pomace Oils by Gas Chromatography/Tandem Mass Spectrometry Evaristo Ballesteros1 and Natividad Ramos-Martos2 1 2
Department of Physical and Analytical Chemistry, EPS of Linares, University of Jaén, Linares (Jaén), Spain Department of Physical Chemistry, Faculty of Sciences, University of Jaén, Spain
47.1 INTRODUCTION 47.1.1 Residues of Pesticides in Olive Products The expansion of agriculture to meet the demands of a rapidly growing population in the early 20th century and the inefficiency of the first generation of pesticides (inorganic substances) compelled farmers to find new, effective agents to control pests. This led to the development of secondgeneration, organic pesticides, application of which peaked in 1939 with dichloro-diphenyl trichloroethane (DDT). The vast use of DDT fostered the development of other organochlorine pesticides (Tomlin, 1994) that were used in an uncontrolled manner for many years until they were found to be highly persistent in the environment, soluble in fat tissue and biologically undegradable – which facilitated their accumulation in human organs (WHO, 1990). In response, a third generation of less persistent, more biodegradable, non-cumulative synthetic pesticides including organophosphorus, carbamate and pyrethroid compounds was developed (Hamilton and Crossley, 2004; Ballesteros, 2007). Other researchers focused on the development of alternative pest control methods based on natural and biological agents capable of altering pest–host interactions (Kling, 1996). Pesticide residues reaching consumers come essentially from four different sources, namely: (a) the use of pesticides on farms; (b) their application to harvested produce; (c) their presence in imported foods; and (d) banned substances discharged into the environment. In recent years, pesticide Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
levels in vegetable foods, fish, meat and fruit have risen to alarming levels according to the scientific community and food safety officials, both in Europe and elsewhere; this has been especially so in Spain, where these toxins can reach essential ingredients of consumers’ daily diet such as olive oil (Hajslová, 1999). This chapter is concerned with the determination of pesticide residues in olive oil and related products. Olive oil is in fact an indispensable ingredient of the Mediterranean diet on account of its nutritional and biological properties. As such, it is being increasingly used in other European countries, and also in eastern and western regions of the world which demand strict control of the presence of contaminants reaching oil by effect of the agricultural treatments used to protect olive trees from pests, weeds and undergrowth (Lentza-Rizos et al., 2001). The Codex Alimentarius Commission of the Food and Agriculture Organization of United Nations (FAO) and the World Health Organization (WHO) have jointly established maximum residue limits (MRLs) for pesticides in olives and olive oil (FAO-WHO, 1996). Also, the European Community has set maximum tolerated levels for pesticide residues in and on some products of plant origin including olives (EU Council Directives of 1976, 1986, 1990, 2002, 2003).
47.1.2 Residues of Polycyclic Aromatic Hydrocarbons in Olive Products Polycyclic aromatic hydrocarbons (PAHs) constitute a special hydrocarbon family consisting of compounds possessing
425
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
426
a certain number of aromatic rings condensed via two or more carbon atoms. Some PAHs are commercially available as intermediates for production processes (e.g. phenanthrene for obtaining pesticides, pyrene for pigments, anthracene and fluoranthene for dyes, naphthalene for plastics). All are used in small amounts, however, so the PAHs encountered in the environment come largely from the oxygen-deficient combustion at temperatures above 500°C of hydrocarbons present in organic materials such as wood, coal or fuel (Mastral and Callén, 2000). Polycyclic aromatic hydrocarbons have been the subject of much study in the last few decades owing to the carcinogenic and mutagenic properties of some members (Menzie et al., 1992). PAHs present in foods come from natural sources or are the result of environmental contamination (e.g. by deposition of air particulates on vegetable foods or intake of contaminated water by fish). PAHs in food can also arise from cooking, grilling and frying; from preservation treatments such as traditional curing and drying (Tamakawa, 2004); and from packaging in materials treated with contaminated mineral oils (Moffat and Whittle, 1999). The presence of PAHs in edible oils and fats is widely documented (Tamakawa, 2004). These products may be contaminated by environmental pollution and/or during processing steps preceding refining (Moret and Conte, 2000). Such processes usually include decolorization with activated carbon and clay to significantly reduce PAH levels in oil (Moreda et al., 2004). The International Olive Oil Council (IOOC) recommended in 2001 a value of 2 μg kg⫺1 as the maximum tolerated concentration of benzo(a)pyrene and other PAHs in olive-pomace oil (IOOC, 2001). Also, Spain’s Ministry of the Presidency has established MRLs for eight PAHs in olive-pomace oil (viz. 2 μg kg⫺1 for individual PAHs and 5 μg kg⫺1 for their combination) (Boletín Oficial del Estado, 2001).
47.1.3 Sample Treatment and Extraction of Pesticide and PAH Residues from Oil and Fat Samples The low detection levels required by official regulations in many countries and the high complexity of the matrix of many foods potentially contaminated with toxic residues (pesticides and PAHs) have promoted the development of simple, sensitive, selective, accurate methods for their routine analysis. Toxic organic residues in olive oil are usually determined by gas (GC) or liquid chromatography (LC) (Ballesteros, 2004). Directly injecting oil samples into a chromatograph, however, results in considerably decreased resolution by effect of the adverse effects of the high fat content of the oil on the chromatographic column; a cleanup pretreatment is therefore recommended prior to injection. Most available methods for removing fat from oil rely
SECTION | I
Pesticides and Adulterants
on liquid–liquid extraction with an appropriate solvent, solid-phase extraction (SPE) or gel-permeation chromatography (GPC) (García-Reyes et al., 2007). To date, GPC is the most widely used technique for routine multiresidue analyses of olive oil, usually after liquid–liquid extraction with acetonitrile (García Sánchez et al., 2006); some GPC methods, however, afford direct extraction and clean-up of pesticides from olive oil with no need for liquid–liquid partitioning (Patel et al., 2005). In any case, the lifetime of columns and chromatographic systems for both extraction and determination can be extended and cleaner extracts obtained by partitioning the samples prior to GPC. Solid-phase extraction has been successfully used for both extraction and clean-up of pesticide residues in olive oil (Lentza-Rizos et al., 2001). The most common method for extracting PAHs from oil and fat samples involves saponifying lipids with alcoholic KOH, followed by liquid–liquid partitioning and solid-phase purification on silica gel (Moret and Conte, 2000). PAHs have also been determined in smoked meat products by using an analytical method involving accelerated solvent extraction and GPC for efficient lipid removal without saponification, and GC quantitation (Jira, 2004).
47.1.4 Determination of Pesticides and PAHs in Olive Products Organic residues in olive oil are usually determined by gas or liquid chromatography, depending on their polarity, volatility and thermal lability. As noted earlier, most oil and fat samples require some treatment prior to their chromatographic analysis owing to the high complexity of their matrices and the need to obtain adequate sensitivity. Pesticides in food samples are most often determined by gas chromatography on account of the high sensitivity and selectivity of this technique, and also of its ready availability to most analytical laboratories. Detection is normally done with a flame ionization, nitrogen-phosphorus or electron capture instrument (Lentza-Rizos et al., 2001). In recent years, however, these detectors have been superseded by mass spectrometers (MS), which feature a high sensitivity and selectivity for pesticides (Patel et al., 2005; García Sánchez et al., 2006). The use of MS for detection is essential with a view to ensuring unambiguous identification of pesticides in food. In this respect, the European Commission has established an identification criterion for contaminants which involves the use of mass spectrometry to meet confirmation criteria based on preset identification points (European Commission, 2006). Table 47.1 summarizes available GCMS methods for determining pesticides in olive oil and olive products. As can be seen, the most widely accepted methods for this purpose involve GC-MS with a single-quadrupole analyzer or gas chromatography/tandem mass spectrometry (GC-MS/MS) with an ion-trap analyzer (Table 47.2).
Analytes and samples
Extraction and clean-up proceduresa
Determinationb
Analytical figures of meritc
Ref.
Dimethoate and fenthion in olive oil
Liquid–liquid partition of 15 g of olive oil in 50 ml of n-hexane and 100 ml of acetonitrile, followed by clean-up using a C18 cartridge
GC-MS (single-quadrupole analyzer) GC-NPD
DL: 1.1–1.8 μg kg⫺1 rec: 78–84 %
Tsatsakis et al. (2003)
Multiresidue of 30 pesticides in olive oil
0.5 ml of olive oil dissolved in 5 ml of tetrahydrofuran. Direct extraction and clean-up by GPC
GC-MS (single-quadrupole analyzer) LC-ESI-MS
DL: 1–1000 μg L⫺1 rec: 2.9–99.1% RSD: 0.5–29.6%
Barrek et al. (2003)
19 organochlorine pesticides in fats and oils (olive oil)
1.25 of sample dissolved in 10 ml of ethyl acetate–cyclohexane (1:1) followed by GPC clean-up
GC-MS/MS (triple-quadrupole analyzer)
DL: 0.1–2.0 μg kg⫺1 rec: 70–110% RSD: 1–18%
Patel et al. (2005)
Organophosphorus and organochlorine pesticides, and PCBs in virgin olive oil
15 g of olive oil is dissolved in 25 ml of n-hexane, followed by extraction on Extrelut-QE column and clean-up with two cartridges (C18 and alumina)
GC-MS/MS (ion-trap analyzer) GC-NPD/ECD
QL: 1–25 μg kg⫺1 rec: 60.1–119.20% RSD: 5–21%
Yagüe et al. (2005)
11 pyrethroid insecticides in vegetable oils (olive oil)
Liquid–liquid partition of 5 g of oil with 10 ml of n-hexane-acetonitrile (1:1) followed by clean-up by SPE using cartridge (C18 and alumina)
GC-MS/MS (ion-trap analyzer)
DL: 0.1–1.1 μg kg⫺1 rec: 91–102% RSD: 4–13%
Esteve-Turrillas et al. (2005)
Multiresidue of 12 pesticides in olives and olive oil
From olives: Extraction by MSPD (aminopropyl) of 1 g of sample followed by clean-up with Florisil column
GC-MS (single-quadrupole analyzer)
DL: 0.2–80 μg kg⫺1
Ferrer et al. (2005)
From olive oil: Liquid–liquid partition of 5 g of sample in 15 ml of petroleum ether (saturated with acetonitrile) followed by MSPD extraction with aminopropyl and clean-up with Florisil
LC-ESI-MS
rec: 73.2–129.7% RSD: 3.3–9.9%
Liquid–liquid partition of 2 g of olive oil with 10 ml of acetonitrile and 2 ml of n-hexane followed by GPC clean-up
GC-MS/MS (ion-trap analyzer)
DL: 0.1–1.6 μg kg⫺1 rec: 83.8–110.3% RSD: 4.9–8.1%
Multiresidue of 26 pesticides in virgin and refined olive oil
CHAPTER | 47 Residues of Pesticides and Polycyclic Aromatic Hydrocarbons
TABLE 47.1 GC–MS methods for the determination of pesticides in olive oil and other olive products.
García Sánchez et al. (2006)
(Continued)
427
428
Table 47.1 (Continued) Analytes and samples
Extraction and clean-up proceduresa
Determinationb
Analytical figures of meritc
Ref.
Multiresidue of 32 pesticides in olive oil
Liquid–liquid partition of 2 g of sample with 10 ml of n-hexane saturated with acetonitrile followed by GPC clean-up
GC-MS/MS (ion-trap analyzer) GC-ECD, GC-NPD
DL: 0.2–10 μg kg⫺1 rec: 82–124% RSD: 1–20%
Guardia-Rubio et al. (2006)
Multiresidue of 32 pesticides in olives
Liquid–liquid partition of 130 g of sample with 100 ml of light petroleum or 100 ml of ethyl acetate (low-fat content vegetables) followed by GPC clean-up
GC-MS/MS (ion-trap analyzer) GC-ECD, GC-NPD
DL: 0.05–5 μg kg⫺1 rec: 70–134% RSD ⬍ 23%
Guardia-Rubio et al. (2007)
Multiresidue of 16 pesticides in olives
QuEChERS method: Liquid–liquid partition of 10 g of sample with 10 ml of acetonitrile followed by a clean-up step with dispersive SPE (MgSO4 ⫹ C18 ⫹ graphitized carbon black ⫹ primary secondary amine)
GC-MS (ion-trap analyzer) LC-MS/MS (triple-quadrupole analyzer)
rec: 70–130% RSD ⬍ 20%
Cunha et al. (2007)
GPC: gel permeation chromatography; SPE: solid-phase extraction; MSPD: matrix solid-phase dispersion; QuEChERS: quick, easy, cheap, effective, rugged and safe method.
b
GC: gas chromatography; LC: liquid chromatography; MS: mass spectrometry; ESI: electrospray interface; ECD: electron capture detector; NPD: nitrogen-phosphorus detector.
c
DL: detection limit; QL: quantitation limit; rec: recovery; RSD: relative standard deviation.
SECTION | I
a
Pesticides and Adulterants
CHAPTER | 47 Residues of Pesticides and Polycyclic Aromatic Hydrocarbons
TABLE 47.2 Definitions of gas chromatography/ tandem mass spectrometry. 1. Gas chromatography with detection by tandem mass spectrometry 2. Tandem mass spectrometry summarizes the numerous techniques where mass-select ions (MS1) are subject to a second mass spectrometric analysis (MS2) 3. Tandem mass spectrometry can also be done in a single mass analyzer over time, as in a quadrupole ion trap 4. Methods for fragmenting molecules for tandem mass spectrometry: collision-induced dissociation (CID), surfaceinductive dissociation (SID), electron capture dissociation (ECD), electron transfer dissociation (ETD), infrared multiphoton dissociation (IRMPD) and blackbody infrared radiative dissociation (BIRD) 5. Tandem mass spectrometry is also known as MS/MS or MS2
Liquid chromatography has been less extensively used than gas chromatography to determine pesticides. However, some recent methods based on LC-MS have provided results on a par with those obtained with GC-MS (Barrek et al., 2003; Ferrer et al., 2005). The determination of PAHs has evolved dramatically ever since the earliest methods, based on paper or thinlayer chromatography with UV or fluorescence detection, were reported. Recent methods rely on LC or GC for this purpose (Moreda et al., 2004). Gas chromatography has been used in combination with various types of detectors including flame ionization, photoionization and mass spectrometric instruments (Tamakawa, 2004). Thus, PAHs in olive-pomace oil have been determined by GC-MS with a single-quadrupole analyzer; detection levels below the maximum tolerated values have thus been obtained (e.g. 0.1–0.4 μg kg⫺1 by Diletti et al. (2005) or 0.10–0.25 μg kg⫺1 by Guillen et al. (2004)). Also, Abdulkadar et al. (2003) determined benzo(a)pyrene by GC-MS/MS in 45 different olive oils with a detection limit of 0.5 μg kg⫺1 and a recovery of 88%. Because most PAHs exhibit fluorescence and specific emission and excitation spectra, they are usually detected by spectrofluorimetry, which is especially sensitive and selective for this purpose. The combined use of this spectroscopic technique and LC provides a powerful tool for PAH routine analyses (Tamakawa, 2004). Thus, benzo(a)pyrene present in vegetable oils has been determined by extraction with acetonitrile, concentration in a rotavapor, redissolution in n-hexane, purification of the extract in a silica gel cartridge and fluorimetric detection following liquid chromatography (Vázquez Troche et al., 2000). Also, high-molecular-weight PAHs in vegetable
429
and olive-pomace oils have been determined by CLC with fluorimetric detection following solid-phase extraction on silica gel; the detection limits thus obtained ranged from 0.01 to 0.2 μg kg⫺1 (Moreda et al., 2004). The concomitant presence of pesticides and PAHs in olive products has rarely been addressed, however, in multiresidue analyses. There is seemingly only one reference to their joint determination (in food (fish muscle) and sediments); the analytes were Soxhlet extracted from the samples with 1:1 acetone-dichloromethane and then cleaned up by GPC prior to their determination by GC/MS (Kong et al., 2005). The following section describes a method developed by the authors’ group which allowed pesticides and PAHs in olive and olive-pomace oils to be simultaneously determined by GC-MS/MS for the first time (Ballesteros et al., 2006).
47.2 SIMULTANEOUS DETERMINATION OF PESTICIDES AND PAHS IN OLIVE AND OLIVE-POMACE OILS BY GC-MS/MS 47.2.1 Methodological Considerations 47.2.1.1 Sample Preparation The sample preparation procedure (Figure 47.1) involved two steps, namely: liquid–liquid extraction and clean-up of the extract by gel-permeation chromatography. 47.2.1.1.1 Solvent Extraction Step An amount of ca. 2 g of filtered olive or olive-pomace oil was weighed into a 30 mL all-glass measuring flask containing 2 mL of n-hexane and 10 mL of acetonitrile, and supplied with 3 mg of anhydrous sodium sulfate, 40 μL of a solution containing 25 μg mL⫺1 (1 μg) bromophos-methyl (the internal standard for pesticides) and 50 μL of one containing 1 μg mL⫺1 (0.05 μg) p-terphenyl (the internal standard for PAHs). The resulting mixture was stirred for 30 min and allowed to stand for 30 min. Then, a volume of 10 mL of extract (the light phase) was dried in a rotary evaporator at a reduced pressure at 45°C, the residue being collected in 10 mL of dichloromethane for subsequent purification by GPC. 47.2.1.1.2 Clean-up Step The GPC system used comprised a Varian Prostar 220 isocratic pump and a Varian Prostar 410 autosampler furnished with two Waters Envirogel CPG clean-up columns, namely: a 150 ⫻ 19 mm column and a 300 ⫻ 19 mm main cleanup column packed with styrene-divinylbenzene copolymer of 200–400 μm mesh. A Prostar 704 fraction collector was used to collect extracts, which were passed through a syringe filter of 0.2 μm pore size attached to a disposable 5 mL plastic syringe prior to injection of a 5 mL aliquot into the GPC column. The mobile phase was dichloromethane at ambient temperature and the flow-rate 5.0 mL min⫺1. The
430
SECTION | I
Sa
Pesticides and Adulterants
m
pl
es Filtration Extraction Pre-concentration Clean-up by GPC
Analysis of residues by GC-MS/MS
FIGURE 47.1 Analytical procedure for determining pesticides and PAHs in olive and olive-pomace oils. PAHs: polycyclic aromatic hydrocarbons; GPC: gel permeation chromatography; GC-MS/MS: gas chromatography–tandem mass spectrometry.
eluent was collected between 14 and 24 min in fraction collection tubes. A 2487 UV-Vis Dual 1 detector from Waters (Milford, MA) was used to measure the absorbance at 220 and 254 nm of the eluate from the GPC column. The volume of fraction containing the pesticides and PAHs, 50 mL, was reduced in a rotary evaporator at a low pressure at a temperature below 50°C. Finally, the extract was diluted to 1 mL with cyclohexane.
47.2.1.2 Optimization of Variables Affecting Sample Preparation and Purification by GPC Tests with various organic solvents as extractants for pesticide and PAH residues in olive and olive-pomace oils (viz. acetonitrile, cyclohexane, dichloromethane, diethyl ether, methanol, n-hexane and petroleum ether) revealed a 1:5 n-hexane/acetonitrile mixture to be the most effective choice. The resulting extracts were purified by gel permeation chromatography (GPC), which has been used for decades to determine residues in foods (Balinova, 1998). The GPC technique uses a column packed with a porous polymer which only retains molecules smaller than the pore size of the material. Synthetic pesticides and most PAHs encountered in olive extracts have molecular masses of 200–400 dalton; by contrast, most fats are 600–1500 dalton in molecular mass. Because fat molecules too large to penetrate the pores of polymer beads are not retained by the column, they are the first to be eluted. This has led some authors to use GPC to purify oil and fat samples for the subsequent determination of pesticides (Patel et al., 2005; García Sánchez et al., 2006; Guardia-Rubio et al., 2006), and also in that of PAHs in foods (Jira, 2004).
We tested various organic solvents as GPC mobile phases and selected dichloromethane on the grounds of its increased efficiency in separating the fat fraction from that containing the pesticides and PAHs. As can be seen from Figure 47.2, pesticides and PAHs were detected in the fraction collected at times from 15 to 23 min. However, the time range was expanded one minute on each end (i.e. to 14–24 min) in order to ensure elution of all pesticides and PAHs potentially present. The phenols and minor components potentially collected at 14 and 15 min had no effect on the chromatographic determination of the analytes.
47.2.1.3 GC-MS/MS Determination Gas chromatography/tandem mass spectrometry was performed on a Varian 3800 gas chromatograph equipped with a Saturno 2200 mass spectrometer (electron ionization and chemical ionization with methanol and ion-trap analyzer with mass/mass option), a 1079 PTV Universal capillary injector, an EFC control and two columns, namely: a 2 m ⫻ 0.25 mm i.d. pre-column (a deactivated silica tube) and a 30 mm ⫻ 0.25 mm i.d. fused silica column (0.25 μm HP-5 5% cross-linked with phenylmethylpolysiloxane) to inject samples. The temperature of the PTV injector port was programmed as follows: 70°C, hold 0.5 min, ramp to 300°C at 150°C min⫺1, hold 15 min. The oven temperature was changed as follows: initial level, 70°C, hold 3.5 min, ramp to 180°C at 25°C min⫺1, hold 10 min, ramp to 300°C at 4°C min⫺1, hold 12 min. Samples were injected in the splitless mode, via a CP8400 autosampler (injected volume 7 μL). The detector temperature (GC-MS transfer line) was 280°C. Analyses were done with a filament-multiplier delay
431
CHAPTER | 47 Residues of Pesticides and Polycyclic Aromatic Hydrocarbons
14–24 min 1.50
Selected time for collection of fractions
E (V)
1.25
Fats
1.00
Phenols and other minor components
Pesticides + PAHs
0.75 0.50
Matrix
0.25 0.00 7.5
10.0
12.5
15.0
17.5
20.0
22.5
25.0
Time (min) FIGURE 47.2 Fractions obtained following gel permeation chromatographic clean-up of the olive oil extracts as determined spectrophotometrically at 220 and 254 nm.
of 7 min in order to avoid damaging the instrument. The MS operating conditions were as follows: electron impact ionization voltage 70 eV, scan rate 1.5 scans s⫺1 and scanned mass range 80–450 m/z. The ion-trap mass spectrometer was operated in both the electron ionization (EI) and chemical ionization (CI) modes; the respective filament emission currents were 90 and 30 μA. Chemical ionization was done with methanol as the reagent gas. The MS/MS process was conducted by collision-induced dissociation (CID) with non-resonant (pesticides) or resonant excitation (PAHs).
47.2.2 Sensitivity and Validation of the Method Calibration curves in the form of straight lines were obtained by injecting extracts of the studied pesticides and PAHs at variable concentrations into the gas chromatograph. Matrix effects commonly encountered in the determination of residues in foods (Patel et al., 2005) were avoided by constructing calibration graphs with refined olive oil (i.e. residue-free oil) as matrix. To this end, we used solutions containing 2 g of refined olive oil in addition to variable concentrations of the pesticides and PAHs, 40 μL of 25 μg mL⫺1 (1 μg) bromophos-methyl (the internal standard for pesticides) and 50 μL of 1 μg mL⫺1 (0.05 μg) p-terphenyl (the internal standard for the PAHs). As can be seen in Table 47.3, the curves were linear over the range 1–500 μg kg⫺1 for pesticides and 0.3–200 μg kg⫺1 for PAHs. The detection and quantitation limits of the target analytes were determined as the lowest concentrations providing chromatographic signals three and ten times, respectively, higher than background noise (viz. the blank signals obtained for extracts from residue-free refined olive oil) (Currie, 1999). In addition, run-to-run relative standard
deviations were all less than 8.6% for both olive and olivepomace oils. The proposed method was validated by analyzing olive and olive-pomace oil samples spiked with pesticide concentrations of 10, 50 or 250 μg kg⫺1, and PAH concentrations of 1, 2 or 10 μg kg⫺1. Each sample was analyzed in triplicate (n ⫽ 3). Recoveries ranged from 84 to 108% for pesticides and 84 to 110% for PAHs.
47.2.3 Analysis of Olive and Olive-Pomace Oils We analyzed samples of virgin and refined olive oil obtained from various producers in Andalucía (southern Spain). All olive-pomace oil samples were from the same area; some were analyzed as received, while others were pretreated with activated carbon to remove PAH residues introduced as contaminants during the production process. The validated method was used to determine 26 pesticides and four polycyclic aromatic hydrocarbons in five samples of each type of oil. As can be seen from Table 47.4, terbuthylazine, diuron and endosulfan sulfate were present in all virgin olive oil samples. The incidence and levels of the pesticides in refined olive oil were low relative to virgin olive oil. This was possibly a result of the pesticides being partly or completely removed during refining of the oil. On the other hand, the PAHs were present at higher concentrations in some refined olive oils than in the virgin olive oils. This may have resulted from the refining process involving heating at temperatures high enough to produce carcinogenic compounds. By way of example, Figure 47.3A shows the chromatogram for one of the virgin olive oils studied. The incidence of pesticides in olive-pomace oil was similar to that in virgin olive oil. Samples 1–3, which were
432
SECTION | I
Pesticides and Adulterants
TABLE 47.3 Analytical figures of merit of the determination of pesticides and PAHs. Residue
Linear range (μg kg⫺1)
R2
RSDa (%)
Limit of detection (μg kg⫺1)
Limit of quantification (μg kg⫺1)
MRL (μg kg⫺1)
Trichlorfon
5–500
0.9957
6.80
1.5
3.4
500c
Diuron
1–500
0.9997
7.46
0.3
0.9
200c
Carbaryl
3–500
0.9914
8.64
0.7
1.6
25 000b
Promecarb
2–500
0.9942
5.62
0.4
1.0
10c
Dimethoate
2–500
0.9916
6.74
0.3
0.8
50b
Simazine
4–500
0.9942
5.96
0.7
1.7
100c
Terbuthylazine
1–500
0.9937
7.80
0.2
0.7
50c
Formothion
1–500
0.9966
7.44
0.2
0.6
20c
Chlorpyrifos methyl
2–500
0.9965
7.60
0.4
1.1
50c
Parathion methyl
2–500
0.9986
7.23
0.3
0.8
200c
Pirimiphos methyl
1–500
0.9992
7.21
0.1
0.3
50c
Terbutryne
5–500
0.9952
6.33
0.8
2.2
50c
Malathion
2–500
0.9988
6.45
0.4
1.1
500c
Chlorpyrifos
1–500
0.9989
3.40
0.3
0.8
50c
Parathion ethyl
4–500
0.9989
6.36
0.7
1.8
50c
Methidation
2–500
0.9996
4.94
0.4
1.2
2000b
Endosulfan α
5–500
0.9987
5.86
1.1
2.6
50c
Oxyfluorfen
4–500
0.9978
6.96
0.7
1.5
20c
Endosulfan β
5–500
0.9982
5.91
1.0
2.4
50c
Endosulfan sulfate
3–500
0.9993
4.98
0.4
1.0
50c
Diflufenican
1–500
0.9996
4.19
0.1
0.4
20c
Phosmet
1–500
0.9984
6.72
0.1
0.3
2000c
Fenoxycarb
5–500
0.9931
7.21
1.7
3.8
1000c
L-Cyhalothryn
1–500
0.9998
8.61
0.1
0.3
20c
α-Cypermethrin
1–500
0.9994
6.40
0.2
0.5
50c
Deltamethrin
3–500
0.9996
6.39
0.2
0.6
100c
Benzo(k)Fluoranteno
0.3–200
0.9922
5.54
0.06
0.15
2d
Benzo(e)pireno
0.3–200
0.9956
7.03
0.05
0.10
2d
Benzo(a)pireno
0.3–200
0.9980
5.11
0.05
0.10
2d
Benzo(ghi)peryleno
0.4–200
0.9973
7.76
0.07
0.20
2d
*Relative standard deviation (n ⫽ 11) for 10 μg kg⫺1 (pesticides) and 1 μg kg⫺1 (PAHs). Reprinted from Journal of Chromatography A, 1111, E. Ballesteros, A. García Sánchez and N. Ramos Martos, Simultaneous multidetermination of residues of pesticides and polycyclic aromatic hydrocarbons in olive and olive-pomace oils by gas chromatography/tandem mass spectrometry, 89–96, 2006, with permission from Elsevier.
433
CHAPTER | 47 Residues of Pesticides and Polycyclic Aromatic Hydrocarbons
TABLE 47.4 Results obtained in the analysis of olive and olive-pomace oils by GPC-GC-MS/MS, all in μg kg⫺1. Residue
Virgin olive oil
Refined olive oil
Olive-pomace oil
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Trichlorfon
–a
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Diuron
46.2
44.2
2.5
24.2
6.5
–
2.1
–
1.4
1.8
2.5
1.6
–
–
–
–
6.5
10.7
3.8
–
–
–
–
1.4
4.7
–
5.8
–
–
Carbaryl
5.2
Promecarb
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Dimethoate
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Simazine
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Terbuthylazine
52.8
66.0
8.7
11.6
55.8
3.4
–
8.6
–
7.2
10.9
–
9.6
–
9.2
Formothion
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Chlorpyriphos methyl
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Parathion methyl
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Pyrimiphos methyl
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Terbutryn
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Malathion
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
6.1
–
–
–
–
–
–
1.1
1.1
1.9
–
–
Chlorpyriphos
5.8
Parathion ethyl
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Methidathion
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Endosulfan α
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Oxyfluorfen
–
–
–
–
–
–
–
–
1.4
–
3.8
–
6.0
–
3.3
Endosulfan β
–
–
–
–
16.1
–
–
–
–
–
–
–
–
–
–
Endosulfan sulfate
48.9
22.6
18.9
9.5
44.2
6.4
6.1
10.2
5.7
2.6
11.6
12.8
–
8.9
3.6
–
–
10.3
3.9
–
4.3
–
2.5
2.7
1.7
1.6
2.4
–
–
1.5
–
–
–
–
2.7
2.9
–
5.3
–
–
1.4
–
–
0.9
–
Diflufenican
8.9
Phosmet
–
–
–
–
–
–
–
–
–
–
–
–
Fenoxycarb
–
–
–
–
–
–
–
–
–
–
–
–
–
–
1.0
–
1.5
1.2
2.8
2.0
–
2.5
1.4
3.2
–
–
–
–
–
–
–
–
–
1.1
–
–
0.9
56.0
3.3
34.9
–
0.9
1.6
–
0.8
2.6
23.8
6.0
8.9
–
5.1
0.6
0.8
1.1
–
1.8
14.9
10.9
7.0
–
1.9
1.1
0.9
1.6
–
1.0
54.0
15.9
1.8
1.5
4.0
L-Cyhalothrin
1.6
1.8
α-Cypermethrin
–
Deltamethrin
–
–
–
Benzo(k)fluoranthene
–
–
–
Benzo(e)pyrene
–
–
–
Benzo(a)pyrene Benzo(g,h,i)perylene a
Not detected.
6.4
1.9
0.5 –
–
0.8 0.5
2.1 3.0 – 1.1
0.5 –
– –
7.2
0.4 0.7
–
3.0
–
– 5.0 –
434
SECTION | I
A
Pesticides and Adulterants
Virgin olive oil
150
0 10
20
B
30
Benzo(g,h,i)perylene
Benzo(a)pyrene 40
Time (min)
Olive-pomace oil
150
0 10
20
30
40
Benzo(g,h,i)perylene
α-Cypermethrin
L-Cyhalothrin
Endosulfan sulfate diflufenican
IS2
Oxyfluorfen
Chlorpyrifos
50
Terbuthylazine
100
Benzo(k)fluoranthene Benzo(e)pyrene Benzo(a)pyrene
IS1
Diuron Carbaryl
kCounts
α-Cypermethrin
IS2
L-Cyhalothrin
Diuron
50
Endosulfan sulfate
100
Terbuthylazine
kCounts
IS1
Time (min)
FIGURE 47.3 Gas chromatograms for virgin (A) and olive-pomace oil (B). IS1 internal standard for pesticides (bromophos-methyl); IS2 internal standard for PAHs (p-terphenyl).
not decolorized during refining, exhibited increased PAH concentrations relative to the other, decolorized samples; this testifies to the need to decolorize this type of oil with activated carbon and clay in order to reduce its PAH content (see Figure 47.3B, which shows the chromatogram for a non-decolorized olive-pomace oil sample).
●
●
●
SUMMARY POINTS ●
A number of analytical methods for pesticide and PAH residues in olive oil and related products are available, most of which use GPC or SPE to pretreat samples and chromatography for their determination.
Only the method described in Section 47.2, however, affords the simultaneous determination of pesticides and PAHs in olive and olive-pomace oil. Gas chromatography in combination with tandem mass spectrometry allows residual pesticides and PAHs at concentrations below their MRLs in olive and olivepomace oils to be sensitively determined. The proposed method has some advantages over existing alternatives. Thus, it affords the simultaneous determination of 26 pesticides and four polycyclic aromatic hydrocarbons in various types of olive oils. Also, it uses samples, reagents and organic solvents sparingly to pretreat samples. Finally, it expedites analyses in relation to other methods.
CHAPTER | 47 Residues of Pesticides and Polycyclic Aromatic Hydrocarbons
●
●
All samples analyzed with the proposed method contained pesticides or PAHs at concentrations below their legally tolerated levels. Based on the results, reducing the concentrations of PAHs in olive-pomace oils entails their decolorization during refining.
REFERENCES Abdulkadar, A.H.W., Kunhi, A.A.M., Jassim, A.J., Abdulla, A.A., 2003. Determination of benzo(a)pyrene by GC/MS/MS in retail olive oil simples available in Qatar. Food Addit. Contam. 20, 1164–1169. Balinova, A., 1998. Multiresidue determination of pesticides in plants by high-performance liquid chromatography following gel permeation chromatographic clean-up. J. Chromatogr. A 823, 11–16. Ballesteros, E., 2004. Carbamate pesticide residues in food. In: Nollet, L.M.L. (ed.), Handbook of Food Analysis. Marcel Dekker, New York, pp. 1177–1209. Ballesteros, E., 2007. Analysis of pesticides in water. In: Nollet, L.M.L. (ed.), Handbook of Water Analysis, 2nd edn. CRC Press, Taylor and Francis Group, Boca Raton, FL, pp. 449–490. Ballesteros, E., García Sánchez, A., Ramos Martos, N., 2006. Simultaneous multidetermination of residues of pesticides and polycyclic aromatic hydrocarbons in olive and olive-pomace oils by gas chromatography/ tandem mass spectrometry. J. Chromatogr. A 1111, 89–96. Barrek, S., Paisse, O., Grenier-Loustalot, M.F., 2003. Determination of residual pesticides in olive oil by GC–MS and HPLC–MS after extraction by size-exclusion chromatography. Anal. Bional. Chem. 376, 355–359. Boletín Oficial del Estado (BOE), 2001. Orden 14558 del 25 de Julio de 2001 (BOE 26/7/01): Establece los límites de determinados hidrocarburos aromáticos policíclicos en aceite de orujo de oliva. pp. 27397–27398. Cunha, S.C., Lehotay, S.J., Mastovska, K., Fernandes, J.O., Beatriz, M., and Oliveira, P.P., 2007. Evaluation of the QuEChERS sample preparation approach for the analysis of pesticide residues in olives. J. Sep. Sci. 30, 620–632. Currie, L.C., 1999. Nomenclature in evaluation of analytical methods including detection and quantification capabilities (IUPAC Recommendations 1995). Anal. Chim. Acta 391, 105–126. Diletti, G., Scortichini, G., Scarpone, R., Gatti, G., Torreti, L., Migliorati, G., 2005. Isotope dilution determination of polycyclic aromatic hydrocarbons in olive pomace oil by gas chromatography–mass spectrometry. J. Chromatogr. A 1062, 247–254. Esteve-Turrillas, F.A., Pastor, A., de la Guardia, M., 2005. Determination of pyrethroid insecticide residues in vegetable oils by using combined solid-phases extraction and tandem mass spectrometry detection. Anal. Chim. Acta 553, 50–57. EU Council Directives (European Community). 76/895/EEC, 86/362/EEC, 86/363/EEC, 90/642/EEC, 02/76/EEC, 03/60/EEC and other modifications. “On the fixing of maximum levels for pesticide residues”. European Commission, 2006. Method of validation and quality control procedures for pesticide residues analysis in food and feed, SANCO/10232/2006. (http://ec.europa.eu/food/plant/protection/ resources/qualcontrol_en.pdf) FAO-WHO, 1996. Codex Alimentarius Commision. Codex Alimentarius Pesticide Residues in Food–Maximum Residues Limits. 2nd edn., Vol. 2B. Rome.
435
Ferrer, C., Gómez, M.J., García-Reyes, J.F., Ferrer, I., Thurman, E.M., Fernández-Alba, A.R., 2005. Determination of pesticide residues in olives and olive oil by matrix solid-phase dispersion followed by gas chromatography/mass spectrometry and liquid chromatography/tandem mass spectrometry. J. Chromatogr. A 1069, 183–194. García-Reyes, J.F., Ferrer, C., Gómez-Ramos, M.J., Molina-Díaz, A., Fernández-Alba, A.R., 2007. Determination of pesticide residues in olive oil and olives. Trends Anal. Chem. 26, 239–251. García Sánchez, A., Ramos Martos, N., Ballesteros, E., 2006. Multiresidue analysis of pesticides in olive oil by gel permeation chromatography followed by gas chromatography–tandem mass-spectrometric determination. Anal. Chim. Acta 558, 53–61. Guardia-Rubio, M., Fernández-de Córdova, M.L., Ayora-Cañada, M.J., Ruiz-Medina, A., 2006. Simplified pesticide multiresidue analysis in virgin olive oil by gas chromatography with thermoionic specific, electron-capture and mass spectrometric detection. J. Chromatogr. A 1108, 231–239. Guardia-Rubio, M., Marchal-López, R.M., Ayora-Cañada, M.J., RuizMedina, A., 2007. Determination of pesticides in olives by gas chromatography using different detection systems. J. Chromatogr. A 1145, 195–203. Guillen, M.D., Sopelana, P., Palencia, G., 2004. Polycyclic aromatic hydrocarbons and olive pomace oil. J. Agric. Food Chem. 52, 2123–2132. Hajslová, J., 1999. Pesticides. In: Moffat, C.F., Whittle, K.J. (eds) Environmental Contaminants in Food. CRC Press, Boca Raton, FL, pp. 215–272. Hamilton, D., Crossley, S., 2004. Pesticide Residues in Food and Drinking Water: Human Exposure and Risks. John Wiley & Sons, Ltd., Chichester, UK. IOOC, 2001. Resolution no. RES-4/85-IV/O1, http://www.internationalo live.org . Jira, W., 2004. A GC/MS method for the determination of carcinogenic polycyclic aromatic hydrocarbons (PAH) in smoked meat products and liquid smokes. Eur. Food Res. Technol. 218, 208–212. Kling, J., 1996. Could transgenic supercrops one day breed superweeds? Science 274, 180–181. Kong, K.Y., Cheung, K.C., Wong, C.K.C., Wong, M.H., 2005. The residual dynamics of polycyclic aromatic hydrocarbons and organochlorine pesticides in fishponds of the Pearl River delta, South China. Water Res. 39, 1831–1843. Lentza-Rizos, Ch., Avramides, E.J., Visi, E., 2001. Determination of residues of endosulfan and five pyrethoid insecticides in virgen olive oil using gas chromatography with electron-capture detection. J. Chromatogr. A 921, 297–304. Mastral, A.M., Callén, M.S., 2000. A review on polycyclic aromatic hydrocabon (PAH) emissions from energy generation. Environ. Sci. Technol. 34, 3051–3057. Menzie, C.A., Potocki, B.B., Santodonato, J., 1992. Exposure to carcinogenic PAHs in the environment. Environ. Sci. Technol. 26, 1278–1284. Moffat, C.F., Whittle, K.J., 1999. Polycyclic aromatic hydrocarbons, petroleum and other hydrocarbon contaminants. In: Moffat, C.F., Whittle, J. (eds) Environmental Contaminants in Food. CRC Press, Boca Raton, FL, pp. 364–429. Moreda, W., Rodríguez-Acuña, R., Pérez-Camino, M.C., Cert, A., 2004. Determination of high molecular mass polycyclic aromatic hydrocarbons in refined olive pomace and other vegetable oils. J. Sci. Food Agric. 84, 1759–1764. Moret, S., Conte, L.S., 2000. Polycyclic aromatic hydrocarbons in edible fats and oils: occurrence and analytical methods. J. Chromatogr. A 882, 245–253.
436
Patel, K., Fussell, R.J., Hetmanski, M., Goodall, D.M., Keely, B.J., 2005. Evaluation of gas chromatography–tandem quadrupole mass spectrometry for the determination of organochlorine pesticides in fats and oils. J. Chromatogr. A 1068, 289–296. Tamakawa, K., 2004. Polycyclic aromatic hydrocarbons in food. In: Nollet, L.M.L. (ed.), Handbook of Food Analysis. Marcel Dekker, New York, pp. 1449–1483. Tomlin, C., 1994. The Pesticide Manual, 10th edn. British Crops Protection Council, Surrey Hill. Tsatsakis, A.M., Tsakiris, I.N., Tzatzarakis, M.N., Agourakis, Z.B., Tutudaki, M., Alegakis, A.K., 2003. Three-year study of fenthion and dimethoate pesticides in olive oil from organic and conventional cultivation. Food Addit. Contam. 20, 553–559.
SECTION | I
Pesticides and Adulterants
Vázquez Troche, S., García Falcón, M.S., González Amigo, S., Lage Yusty, M.A., Simal Lozano, J., 2000. Enrichment of benzo[a]pyrene in vegetable oils and determination by HPLC-FL. Talanta 51, 1069–1076. WHO, 1990. Public Health Impact of Pesticides Used in Agriculture. World Health Organization, Geneva. Yagüe, C., Bayarri, S., Conchello, P., Lázaro, R., Pérez-Arquillué, C., Herrerra, A., Ariño, A., 2005. Determination of pesticides and PCBs in virgin olive oil by multicolumn solid-phase extraction cleanup followed by GC-NPD-ECD and confirmation by ion-trap GC–MS. J. Agric. Food Chem. 53, 5105–5109.
Chapter 48
Acephate and Buprofezin Residues in Olives and Olive Oil Pierluigi Caboni and Paolo Cabras Department of Toxicology, University of Cagliari, Italy
48.1 INTRODUCTION The Mediterranean region is responsible for 95% of the world production of olive oil of which 92% comes from Spain (36%), Italy (25%), Greece (18%), Tunisia (8%) and Turkey (5%) (http://r0.unctad.org/infocomm/anglais/olive/market. htm). Production trends are growing due to expanded plantings of olives in Europe, Latin America, the USA, and Australia (http://r0.unctad.org/infocomm/anglais/olive/market.htm). The main consuming countries are also the main olive oil producers. The European Union accounts for 72% of world consumption. Mediterranean basin countries represent 77% of world consumption. Other consuming countries are United States, Canada, Australia and Japan. If we report daily per capita grams consumption of olive oil the rank will be: Greece (65), Spain (37), Italy (34), Tunisia (25), and Turkey (3). The most serious parasite afflicting olive cultivation in the Mediterranean area is the olive fruit fly (Bactrocera oleae). Larvae of the latter insect are monophagous, and feed exclusively on olive fruits determining important damages to the olive oil production. For this reason, to produce high-quality olive oil, the olive fruit should be protected from the attack of this parasite. Many pesticides are used for the control of this pest, and starting from 1999 in the EU, a high number of active ingredients were withdrawn from the market with particular attention given to organophosphates (OP). Only buprofezin, dimethoate and rotenone passed the EU revision step for the olive oil whereas the others were not approved. This study aims at contributing to the knowledge of fate of the residues of buprofezin and acephate from the olive to the oil.
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
48.2 ACEPHATE AND BUPROFEZIN RESIDUES IN OLIVES AND OLIVE OIL Field trials were carried out to compare the persistence of two insecticides, acephate and buprofezin, acetylcholinesterase and chitin synthesis inhibitors respectively, showing different water solubility to evaluate field disappearance and residue fate during olive processing to olive oil. Following pseudo-first-order kinetics, acephate olive residues after treatment decreased with half-life times between 12.0 and 12.7 days, those of buprofezin ranged from 9.5 and 13.4 days. These data are in agreement with half-life times of other pesticides used to control fruit flies. In Table 48.1, residue levels of acephate and buprofezin were compared with those of the most important insecticides used in olive production to control fruit fly (Ferreira, 1983; Cabras, 1997, 2000, 2002). Table 48.1 reports for each insecticide: the maximum residue level (MRL) (Table 48.2) together with the pre-harvest intervals (PHIs); residues on olives at different days after the treatment (DAT), residues in the processed olive oils, residue concentration factors during olive oil production (CF), the acceptable daily intake (ADI) (Table 48.3), and the partition coefficient octanol/water (log P). Moreover, Table 48.1 reports (in bold) pesticides actually registered in Italy for the control of the fruit fly. Due to the high water solubility of acephate, during olive processing, residues of this active ingredient are not transferred to olive oil but were found in the vegetation water. Similar considerations can be applied to dimethoate. On the other hand, due to the high lipophilicity of buprofezin (log P ⫽ 4.30), residues are transferred from olives to olive oil with a concentration factor of 3.7–4.6. More generally, pesticides with high log P values show
437
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
438
SECTION | I
Pesticides and Adulterants
TABLE 48.1 Residues (mg kg⫺1) of some insecticides in olives after treatment and in olive oil with some chemical and toxicological properties and Italian legal limits. Pesticide
PHI MRL* (mg kg⫺1) (days)
DAT (days)
Residues on olives (mg kg⫺1)
Residues in olive oil (mg kg⫺1)
CF
ADI (mg kg⫺1)
log P
Acephate
1.5
28
0.39 ⫾ 0.13
⬍0.01
–
0.01
⫺0.89
28
0.49 ⫾ 0.14
⬍0.01
1
1.82 ⫾ 0.63
4.57 ⫾ 0.88
2.5
0.005
2.96
8
1.03 ⫾ 0.13
3.10 ⫾ 0.58
3.0
14
0.69 ⫾ 0.18
1.62 ⫾ 0.32
2.3
28
0.19 ⫾ 0.09
0.87 ⫾ 0.35
4.6
0.01
4.30
28
0.38 ⫾ 0.16
1.40 ⫾ 0.60
3.7
60
2.75
1.97
0.8
0.01
4.7
1
1.34 ⫾ 0.11
4.43 ⫾ 1.26
3.3
0.002
3.30
8
1.11 ⫾ 0.47
3.78 ⫾ 0.32
3.4
13
0.68 ⫾ 0.36
2.15 ⫾ 0.46
3.2
20
0.35 ⫾ 0.10
1.95 ⫾ 0.80
5.6
1
1.60 ⫾ 0.11
0.53 ⫾ 0.18
0.3
0.002
0.70
8
1.08 ⫾ 0.01
0.24 ⫾ 0.01
0.2
14
0.17 ⫾ 0.00
⬍0.01
1
3.01 ⫾ 060
6.78 ⫾ 2.83
2.3
0.001
2.2
8
1.68 ⫾ 0.79
5.69 ⫾ 1.78
3.4
14
1.28 ⫾ 0.43
3.37 ⫾ 0.33
2.6
21
0.46
1.30
2.8
0.004
3.8
41
0.29
1.70
5.9
1
1.40 ⫾ 0.12
4.00 ⫾ 1.02
2.9
0.003
3.0
8
0.61 ⫾ 0.16
2.91 ⫾ 0.23
4.8
13
0.35 ⫾ 0.16
1.77 ⫾ 0.36
5.1
20
0.19 ⫾ 0.06
1.33 ⫾ 0.33
7.0
1
1.84 ⫾ 0.10
2.63 ⫾ 0.60
1.4
8
0.68 ⫾ 0.15
2.13 ⫾ 0.22
3.1
13
0.36 ⫾ 0.14
0.50 ⫾ 0.40
1.4
Azinphos methyl
Buprofezin
0.5
1.0
35
20
28
(3.0)** Chlorpyriphos Diazinon
Dimethoate
1.0
20
(0.2)**
Methidathion
1
100
Parathion
Parathion methyl
Quinalphos
0.2
20
–
4.44
439
CHAPTER | 48 Acephate and Buprofezin Residues in Olives and Olive Oil
Rotenone (5)
0.04
20
0.20 ⫾ 0.04
0.80 ⫾ 0.14
4.0
2
0.52 ⫾ 0.13
1.89 ⫾ 0.18
3.6
5
0.44 ⫾ 0.12
1.05 ⫾ 0.12
2.4
9
0.19 ⫾ 0.04
0.51 ⫾ 0.05
2.7
12
0.11 ⫾ 0.02
0.53 ⫾ 0.18
4.8
–
4.16
This table shows levels of some insecticide in olives and olive oil after filed treatment. Residues are expressed as mg kg⫺1. Residue level represents the mean of triplicate tests in each treatment. MRL represents the maximum residue level, PHI represents the pre-harvest interval; DAT represents residues on olives at different days after the treatment; CF represents the residue in the processed olive oils, residue concentration factor during olive oil production; ADI represents the acceptable daily intake, log P represents the partition coefficient octanol/water. * Relative to maximum residue levels in olives. ** Relative to maximum residue levels in olive oils.
TABLE 48.2 Definition of the maximum residue level (MRL) (http://www.pesticides.gov.uk/prc. asp?id⫽956). MRLs are defined as the maximum concentration of pesticide residue (expressed as milligrams of residue per kilogram of food/animal feeding stuff) likely to occur in or on food and feeding stuffs after the use of pesticides according to Good Agricultural Practice (GAP), i.e. when the pesticide has been applied in line with the product label recommendations and in keeping with local environmental and other conditions). There are a number of different types of MRLs depending on, e.g. whether they were set by the European Commission or by the local Committees on pesticides
TABLE 48.3 Definition of the acceptable daily intake (ADI) (http://www.medterms.com/script/main/art. asp?articlekey⫽30760). Estimate of the amount of a substance in food or drinking water, expressed on a body mass basis (usually mg kg⫺1 body weight), which can be ingested daily over a lifetime by humans without appreciable health risk. For calculation of the daily intake per person, a standard body mass of 60 kg is used. The acceptable daily intake is normally used for food additives (tolerable daily intake is used for contaminants)
higher residue levels in olive oil and for this reason it is necessary to harvest olives many days after treatment. MRLs are fixed on olives but the raw fruit is not consumed but transformed to olive oil or table olives. In the process of olive oil production we stated that hydrophilic compounds are not present in the olive oil, while residue levels of lipophilic compounds can increase up to a factor of 7. For the reason that the MRL is directly correlated to food safety it is not clear why the MRL is fixed on the fruit and not for the olive oil and table olives. For a correct evaluation of food safety the MRL should be fixed for olive oil or perhaps for table olives. A few years ago Italy fixed the MRL on olive oil for the following active ingredients: buprofezin (3 mg kg⫺1),
dimethoate (0.2 mg kg⫺1), fenoxicarb (1 mg kg⫺1), flazasulfuron (0.01 mg kg⫺1). For buprofezin, the MRL for olives is set to 1 mg kg⫺1 and 3 mg kg⫺1 for the olive oil. On the basis of our data, and taking into account the concentration factor that on average is higher than 4, with an MRL for olives of 1 mg kg⫺1 the MRL for olive oil should be fixed at least to a residue level of 5 mg kg⫺1. If we take into account the buprofezin ADI (0.01 mg kg⫺1) it means that for a man of 60 kg the maximum intake of buprofezin should be 0.6 mg per day. On average, in Italy the daily consumption of olive oil is 34 g, consuming olive oil with a residue level of 3 mg kg⫺1 the intake should be 0.10 mg corresponding to 14% of the ADI. Since buprofezin-containing olive oil has residues lower than the MRL (1.40 ⫾ 0.60 mg kg–1), the amount of buprofezin intake will be lower. Making similar considerations for other insecticides reported in Table 48.1, diazinon and methidathion, at the lowest olive oil residue levels, gave an intake greater than the ADI value. For this reason many OP insecticides were withdrawn from the market. These data clearly show that the insecticide intake through olive oil consumption is null or very small for hydrophilic pesticides. In contrast, when lipophilic compounds are used, olives must be harvested many days after pesticide treatment to ensure low residue levels in olive oil.
REFERENCES Cabras, P., Angioni, A., Garau, V.L., Melis, M., Pirisi, M.F., Karim, M., Minelli, E.V., 1997. Persistence of insecticide residues in olives and olive oil. J. Agric. Food Chem. 45, 2244–2247. Cabras, P., Angioni, A., Garau, V.L., Pirisi, F.M., Cabitza, F., Pala, M., 2000. Acephate and buprofezin residues in olives and olive oil. Food Addit. Contam. 17, 855–858. Cabras, P., Caboni, P., Cabras, M., Angioni, A., Russo, M., 2002. Rotenone residues on olives and in olive oil. J. Agric. Food Chem. 50, 2576–2580. Ferreira, J.R., Tainha, A.M., 1983. Organophosphorus insecticide residues in olives and olive oil. Pestic. Sci. 14, 167.
This page intentionally left blank
Chapter 49
Olive Oil Adulteration with Hazelnut Oil and Analytical Approaches for Its Detection Dimitrios Zabaras CSIRO/Food Science Australia, Food Quality and Safety, NSW, Australia
49.1 INTRODUCTION Adulteration of olive (Olea europaea) oil with vegetable, seed or nut oils has become a serious problem for regulatory agencies, consumers, oil producers and oil importers. Admixtures of olive oil with hazelnut (Corylus spp.) oil, in particular, represent one of the most recent challenges facing scientists and regulatory bodies. Although hazelnut oil is not associated with some of the health risks attributed to some of the other low-quality oils used to adulterate olive oil (for example, aniline-denatured rapeseed oil), the difference in economic value between the two oils causes an estimated loss of 4 million euros per year for countries in the European Union (EU, 2001). In addition, the dietary intake of health-benefiting components found naturally in virgin olive oil (e.g. polyphenols, sterols, tocopherols) can be greatly reduced following consumption of oils adulterated with low-quality refined hazelnut oils. Detection of hazelnut oil in admixtures with olive oil has always been very difficult to confirm due to very similar chemical profiles exhibited by the two oils, especially in terms of triacylglycerols and fatty acids (Parcerisa et al., 1997). In addition, fraudsters have been able to circumvent traditional methods and thus are no longer useful. Oil chemists, however, have responded well to the constant challenge of being able to outsmart fraudulent producers and reliably confirm olive oil admixtures with hazelnut oil. As a result, numerous chemical approaches have been proposed over the last 15 years that are able to detect various hazelnut oil/olive oil blends. Profitable olive oil adulteration by hazelnut oil, as with most other edible oils, can only be effected by one of three ways: virgin (also termed crude or pressed) hazelnut oil into virgin olive oil, refined hazelnut oil into virgin olive oil and refined hazelnut oil Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
into refined olive oil. An overview of the modern analytical approaches (proposed over the last two decades or so) and the current state of knowledge in each of the three adulteration scenarios is discussed in detail in the following text.
49.2 THE INTERNATIONAL OLIVE OIL COUNCIL (IOOC) RECOMMENDED METHOD The International Olive Oil Council (IOOC) has provisionally adopted a method, based on the determination of theoretical and experimental values of triacyglycerols, as a way of reliably detecting extraneous oil adulterants in olive oil (IOOC, 2006a,b). The method uses the overall ratio between the ratios of equivalent carbon number 42 and 44 (rECN42/rECN44) to ascertain the status of virgin and refined olive oils (IOOC, 2006b). Each rECN is calculated from the experimental triacylglycerol content (determined by liquid chromatography) over the theoretical triacylglycerol content (estimated from the gas chromatographic determination of the C16 and C18 fatty acids present in the oil) (IOOC, 2006b). However, the IOOC itself acknowledges that the recommended method has limitations (for example, the adulterant oil is not identified) and can be ineffective especially at low adulteration levels (equal or lower than 5%) as has been shown by additional studies (IOOC, 2006a,b; Christopoulou et al., 2004). The IOOC and other authorities affected by the olive oil adulteration (e.g. the European Union) regularly review and revise their recommendations in order to keep in pace with the constant analytical developments as well as the always-increasing sophistication of fraudulent practices.
441
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
442
SECTION | I
49.3 DETECTION OF VIRGIN HAZELNUT OIL INTO VIRGIN OLIVE OIL The major components of both olive and hazelnut oils are triacylglycerols (TAGs); they are normally present at levels greater than 95% (Cert et al., 2000). Additional minor components found in the oils include diacylglycerols (DAGs), free fatty acids, hydrocarbons, phenolics, carotenoids, tocopherols and a plethora of volatile compounds (Aparicio and Aparicio-Ruiz, 2000, and references therein). Some of these components, for example filbertone, have been investigated in detail over the years as potential markers for the detection of virgin olive/hazelnut oil admixtures (Blanch et al., 1999; Flores et al., 2006a,b).
49.3.1 The use of Filbertone as an Adulteration Marker Filbertone ((E)-5-methylhept-2-en-4-one), an enantiomeric aroma-impact compound in hazelnuts, is not found in olive oil (Blanch et al., 1999; Flores et al., 2006a). The presence of filbertone, in hazelnuts and hazelnut oil, as a mixture of two enantiomers (R- and S-forms; Figure 49.1) enhances its potential to be used as a hazelnut marker as the detection of both enantiomers in a suspect olive oil provides strong evidence of adulteration (Blanch et al., 1999). In a recent study, adulteration of virgin olive oil with virgin hazelnut oils derived from various regions (Turkey, France, Italy) was detected at levels as low as 7% using headspace extraction and multidimensional gas chromatographic determination of filbertone (Flores et al., 2006b). However, adulterations of virgin olive oil with refined hazelnut oils cannot be reliably detected at levels lower than 20–25% using filbertone (Flores et al., 2006b). The concentration of filbertone in hazelnuts increases by certain processing conditions such as roasting of the nuts prior to oil extraction (Ruiz del Castillo et al., 2003). As a result, hazelnut oils can vary significantly in filbertone levels (Ruiz del Castillo et al., 2003); this prohibits the use of filbertone as a quantitative marker for the estimation of adulteration.
49.3.2 The use of Maillard-reaction Products (MRPs) as Adulteration Markers Roasting is an important process for the hazelnut industry as it significantly enhances the flavor, color, texture and appearance of the nuts but it also removes the kernel pellicle and inactivates pathogenic microorganisms (Fallico et al., 2003). Volatile and non-volatile Maillard-reaction products (MRPs) generated during the heating process are responsible for the desirable changes in flavor and color of the nuts following roasting (Jung et al., 1997) (Table 49.1; Figure 49.2).
Pesticides and Adulterants
O
O
H3C
H R-form
H3C
H
S-form
FIGURE 49.1 Structures of filbertone. The structures of both forms (enantiomers) of filbertone found in hazelnut oil are shown.
TABLE 49.1 Key facts of the Maillard reaction. 1. The term Maillard reaction is used to describe a group of very complex non-enzymatic reactions between reducing sugars and amino acids that occur during the cooking of food 2. The reaction was named after the French chemist LouisCamille Maillard who began investigating the reaction in 1912 3. Hundreds of low- and high-molecular-weight compounds are formed as part of the overall reaction 4. Some of the low-molecular-weight products formed can have very strong flavor characteristics and are responsible for the desirable (or sometimes undesirable) aromas associated with cooked food 5. The amino acids participating in the reaction determine to a large extent the nature and flavor attributes of the products formed 6. The large molecules formed during the reaction, termed mellanoidins, are primarily responsible for the dark colors observed in cooked or roasted foods
Significant levels of known MRPs, including pyrrole and furanone derivatives, were detected in the polar fraction of 14 virgin hazelnut oils sourced from various oil mills (Spain, France, USA and Turkey) and retail outlets (UK) (Table 49.2) (Zabaras and Gordon, 2003). Two of those MRPs, 5-hydroxy-2-methylfurfural (5-HMF) and a (polyhydroxyalkyl)pyrazine (PHAP) (MW 288 as determined by electrospray mass spectrometry, Figure 49.3) were the most abundant MRPs across all virgin hazelnut oils investigated and were not detected in any of the ten virgin olive oils used as a reference (Figure 49.4) (Zabaras and Gordon, 2004). A simple, although laborious, liquid-chromatographic method was developed that used these two MRPs to detect virgin olive/hazelnut oil admixtures (Zabaras and Gordon, 2004). The method could reliably detect low-level adulteration (5% limit) and was ring-tested by six independent laboratories across Europe with positive results (Zabaras and Gordon, 2004). This approach, however, is only useful in a qualitative manner as the levels of MRPs were found to vary greatly between virgin hazelnut oils; roasting time and temperature are likely to be the major factors affecting generation of volatile and non-volatile MRPs in nuts
443
CHAPTER | 49 Olive Oil Adulteration with Hazelnut Oil and Analytical Approaches for Its Detection
A
800
ISTD
600
DAD response at 280 nm (mAu)
400
200
800 ISTD
1 600 2 400 6 200
min
−50 0.0
B
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
Retention time (min) FIGURE 49.2 RP-HPLC traces of the polar fractions from oils obtained by the Sohxlet extraction of (A) non-roasted and (B) roasted hazelnuts. This figure compares the profiles between non-roasted and roasted hazelnuts. The non-roasted profile is free from Maillard-reaction products (MRPs) (peaks 1, 2 and 6). Numbered peaks refer to MRPs in Table 49.1. Internal standard (ISTD) is syringic acid. DAD: diode array detector.
TABLE 49.2 Maillard-reaction products (MRPs) identified in the polar fraction of 14 virgin hazelnut oils from various sources. Peak no.a
MRPs
λmax (nm)b
Amount in oil (mg kgⴚ1)c min-max (ⴞSTD)
1
4-hydroxy-5-methyl-3-(2H)-furanone
285
nd-3.64 (⫾0.61)
2
5-hydroxymethyl-2-furaldehyde
285
nd-6.84(⫾1.19)
–
2-hydroxymethylfuran
279
nd-1.76 (⫾0.13)
–
2,5-dimethyl-4-hydroxy-3-(2H)-furanone
285
nd-0.51 (⫾0.04)
–
pyrrole-2-carboxaldehyde
295
nd-2.85 (⫾0.09)
6
a (polyhydroxyalkyl)pyrazine
295
nd-4.78 (⫾0.35)
–
2-acetylpyrrole
292
nd-1.83 (⫾0.12)
–
2-acetyl-1-methylpyrrole
288
nd-0.35 (⫾0.04)
This table presents the MRPs detected in the oils and their concentration range across all the virgin hazelnut oils investigated. The two most abundant MRPs in the virgin hazelnut oils examined were 5-hydroxymethyl-2furaldehyde and a (polyhydroxyalkyl)pyrazine. a
Numbers correspond to the peaks in Figure 49.2. Value of maximum absorption in UV spectrum. Syringic acid equivalents (mean ⫾ standard deviation, n ⫽ 3).
b c
444
SECTION | I
Pesticides and Adulterants
RT: 4.89 - 13.37 9.38
NL: 6.45E5 Channel A UV Avalon hazel2
600000
UV (λ = 295)
550000 500000 450000
uAU
400000 350000 300000 250000
7.00 6.71 6.14 6.52
200000 150000
4.92
100000
7.67
5.26 5.49
7.76 8.07
8.53
8.78
50000
A
10.07
8.90
10.22 10.84 11.14 11.60
12.06 12.30
12.85
0 NL: 1.66E7 m/z= 286.90000287.10000 F: - p Q1MS [ 270.90-287.10] MS ICIS hazel2
9.49 16000000 15000000
–ESI/MS (m/z 287)
14000000 13000000 12000000
Intensity
11000000 10000000 9000000 8000000 7000000 6000000 11.40
5000000 4000000
10.25
3000000
9.88
2000000 5.25 5.78 6.07 6.24
1000000
6.48
7.49 8.07
6.79 6.95
8.28
8.62
10.69 11.27
9.07
11.88 12.26
12.86
0 5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
B
9.0
9.5
10.0
10.5
11.0
11.5
12.0
12.5
13.0
Time (min)
FIGURE 49.3 (A) UV (λ ⫽ 295) and (B) negative electrospray mass spectrometric (m/z 287) trace of a polar fraction from a commercial virgin hazelnut oil. A (polyhydroxyalkyl)pyrazine (PHAP) gives rise to the major peak in both traces. A mass-to-charge ratio (m/z) of 287 in the negative mode suggests that the PHAP has a molecular weight of 288 daltons. UV: ultraviolet; ESI-MS: electrospray-mass spectrometry.
(Fallico et al., 2003). 5-HMF and PHAP were not detected in refined hazelnut (Zabaras and Gordon, 2003), likely removed during the refining step, and thus this approach is not applicable to admixtures where refined hazelnut oil is used as the adulterant. PHAPs are also known to be present in peanut oil (Magaletta and Ho, 1996) so this approach could be useful in detecting olive oil/peanut oil admixtures. Some other approaches utilizing minor components to detect virgin-into-virgin adulteration are summarized in Table 49.3.
49.4 DETECTION OF REFINED HAZELNUT OIL IN VIRGIN OLIVE OIL The detection of refined hazelnut oil into virgin olive oil is inherently more challenging than the detection of virgin
hazelnut oil into virgin olive oil as the refining process greatly reduces or eliminates most of the chemical compounds (shown in Table 49.2) that could serve as adulteration markers. Oil components usually unaffected by the refining process, such as triacylglycerols and sterols, have been the main focus of investigators attempting to develop methodologies that can reliably detect low levels of refined hazelnut oil in virgin olive oil. Obviously, these approaches are also useful in the detection of ‘virgin into virgin’ types of adulteration but they are discussed here as they play a critical role in the detection of ‘refined into virgin’ admixtures.
49.4.1 Sterols as Adulteration Markers Sterols or steroid alcohols are naturally occurring phytochemicals that are usually found in their free or esterified
445
CHAPTER | 49 Olive Oil Adulteration with Hazelnut Oil and Analytical Approaches for Its Detection
1400
dz128 #3 mAU
UV_VIS_2 WVL:280 nm
Virgin Hazelnut oil 5-HMF
1000
UV_VIS_2 WVL:280 nm
750
PHAP 500
A
DAD response (mAu)
250
0 min
−200 0.0 dz129 #2 1600 mAU
5.0
10.0
15.0
20.0
25.0
30.0
35.0
25.0
30.0
35.0
Virgin Olive oil
40.0 UV_VIS_2 WVL:280 nm
1250 1000
Tyrosol 750 500
Hydroxytyrosol
250
min
−200
B
0.0
5.0
10.0
15.0
20.0
40.0
Time (min) FIGURE 49.4 RP-HPLC traces (λ ⫽ 280 nm) of a polar fraction from (A) commercial virgin hazelnut oil and (B) commercial extra virgin olive oil. The two major components seen in the profile from the virgin hazelnut oil, 5-HMF and PHAP, are not detected in virgin olive oil and hence can be used to indicate adulteration. 5-HMF: 5-hydroxy-2-methylfurfural; PHAP: (polyhydroxyalkyl)pyrazine; DAD: diode array detector. Reprinted and modified from Zabaras, D. and Gordon, M.H., 2004. Food Chem. 84:475–483, with permission.
form in virgin edible oils (Mariani et al., 2006); refining is known to drastically reduce the level of free but not esterified sterols present in an oil (Cercaci et al., 2003). Attempts to utilize the amount and composition of total sterols (and other linear and triterpenic alcohols) to unequivocally confirm hazelnut-into-olive adulteration below 30% have not been successful (Cercaci et al., 2003). However, the distinct determination of free and esterified sterols could be useful, particularly for the detection of refined-into-virgin admixtures, as refining decreases the amount of total sterols detected without affecting the concentration of esterified sterols (Cercaci et al., 2003). Studies have shown that the esterified sterols, campesterol and Δ7-stigmastenol, are more abundant in hazelnut oil than olive oil; the values of Δ7-avenasterol were found to vary between studies (Cercaci et al., 2003) (Table 49.4). A couple of equations based on the percentage of the three esterified sterols in the oil have been developed that are able to reliably detect most hazelnut-into-olive oil blends at levels equal to or higher
than 6–8% (Mariani et al., 2006). The type of equation to be used in each case is determined by the total concentration of the three esterified sterols in the oil (Table 49.5). The major sterol in both olive and hazelnut oils is β-sitosterol (Parcerisa et al., 2000). Oil refining, and the use of bleaching earth in particular (Gordon and Firman, 2001), is known to catalyze the dehydration of sterols leading to the formation of steradiene hydrocarbons such as stigmastadienes (from β-sitosterol) (Lanzón et al., 1994) (Figure 49.5) and campestadienes (from campesterol). Therefore, elevated levels of these hydrocarbons, especially stigmastadienes as they are the most abundant in virgin olive oil can be a useful indicator of adulteration with a refined oil including hazelnut oil.
49.4.2 Use of Spectroscopic Methods Another useful method that has been applied for the detection of refined hazelnut oil into virgin olive oil is one based
446
SECTION | I
Pesticides and Adulterants
TABLE 49.3 Additional (to those described in the text) methodologies that have been useful in detecting adulteration of virgin olive oil with virgin hazelnut oil.
TABLE 49.5 Equations, constraints and decision rules for the detection of adulterated olive oil samples using free and esterifed sterols.
Equipment used
Analyte or other information used to indicate adulteration
Equation
Concentration of esterified sterols
R value
ICP-AES and GFAAS
Trace metal profilea
R1 ⫽ campesterol ⫻ [(Δ7stigmastenol)/(Δ7avenasterol)
⬍200 mg kg⫺1
R1 ⬍ 1.6 genuine
b
HPLC
Tocopherol ratios (γ/β and β/δ)
1
Total resonance datac
HS-MS
Total volatile signald
H and 13C NMR
This table provides information related to the major equipment used and the factor(s) used to indicate adulteration in selected approaches found in the literature. This set is shown as it indicates, together with the other methods described in the text, the great diversity of analytical approaches that have been used to detect oil adulteration. a Cindric, I.J., Zeiner, M., Steffan, I., 2007. Trace elemental characterisation of edible oils by ICP-AES and GFAAS. Microchem. J. 85: 136–139. b Cercaci, L,, Rodriguez-Estrada, M.T., Lercker, G., 2003. Solid-phase extraction-thin-layer chromatography-gas chromatography method for the detection of hazelnut oil in olive oils by determination of esterified sterols. J. Cromatogr. A 985: 211–220. c Garcia-Gónzalez, D.L., Mannina, L., D’Imperio, M., Segre, A.L., Aparicio, R., 2004. Using 1H and 13C NMR techniques and artificial neural networks to detect the adulteration of olive oil with hazelnut oil. Eur. Food Res. Technol. 219: 545–548. d Pena, F., Cárdenas, S., Gallego, M., Valcárcel, M., 2005. Direct olive oil authentication: Detection of adulteration of olive oil with hazelnut oil by direct coupling of headspace and mass spectrometry, and multivariate regression techniques. J. Cromatogr. A 1074: 215–221.
Sample status
R1 ⱖ 1.6 adulterated 200– 600 mg kg⫺1
R1 ⬍ 1.0 genuine R1 ⱖ 1.0 adulterated
⬎600 mg kg⫺1
R1 ⬍ 1.0 genuine R1 ⱖ 1.0 calculate R2
R2 ⫽ Δ7stigmastenol free (mg kg⫺1) ⫻ {[(Δ7stigmastenol free) (%)]/[(Δ7stigmastenol ester)(%)]}
⬎600 mg kg⫺1
R1 ⱖ 1.0 Genuine and R2 ⬍ 0.5
R1 ⱖ 1.0 adulterated and R2 ⱖ 0.5
TABLE 49.4 Mean and standard deviation (SD) of three individual esterified sterols quantified in 63 olive oil samples adulterated with hazelnut oil. Hazelnut oil in sample
These equations utilize the concentration of esterified and free sterols present in an oil to generate values, R, which can be used to confirm the status of the oil. Data reprinted from Mariani, C. et al., 2006. Eur. Food Res. Technol. 223: 655–661, with permission.
Esterified sterols (%) Mean ⫾ SD
Campesterol
Δ 7Stigmastenol
Δ7Avenasterol
0%
3.30 ⫾ 0.29
0.42 ⫾ 0.18
1.00 ⫾ 0.47
⬎0% and ⱕ8%
3.48 ⫾ 0.39
0.48 ⫾ 0.17
0.86 ⫾ 0.34
⬎8% and ⱕ20%
3.53 ⫾ 0.38
0.59 ⫾ 0.22
0.92 ⫾ 0.31
100%
5.32 ⫾ 0.53
1.71 ⫾ 0.70
1.15 ⫾ 0.42
The values of campesterol and Δ7-stigmastenol increase as the percentage of hazelnut oil in olive oil increases. This is because hazelnut oil contains greater concentrations of these components than olive oil. SD: standard deviation. Reprinted from Mariani, C., et al., 2006. Eur. Food Res. Technol. 223: 655–661, with permission.
-H2O HO β-Sitosterol
Stigmasta-3,5-diene
FIGURE 49.5 Formation of 3,5-stigmastadiene. Dehydration of β-sitosterol to form 3,5-stigmastadiene.
on 31P and 1H NMR spectroscopy in combination with statistical analysis (Vigli et al., 2003). Statistical analysis tends to be very important for spectroscopic methods as these usually rely on complex datasets with several variables for the discrimination of genuine and non-genuine oils. Data from 31P and 1H NMR spectra can successfully distinguish, following statistical treatment, low (down to
CHAPTER | 49 Olive Oil Adulteration with Hazelnut Oil and Analytical Approaches for Its Detection
5% w/w) adulteration of olive oil with virgin and refined hazelnut oil as long as the virgin olive oil in question has a high 1,2-diglycerides to total diglycerides ratio (D ⱖ 0.90) (Vigli et al., 2003). Good-quality, fresh oils are characterized by high D values whereas refined oils usually have much smaller D ratios due to the rapid conversion of 1,2 diglycerides (1,2-DGs) to 1,3-diglycerides (1,3-DGs) during oil refining (Vigli et al., 2003). The most useful variables for the detection of adulterated virgin olive oils with refined hazelnut or other seed oils based on 31P and 1 H NMR data were found to be 1,2- and 1,3-DGs, sterols, D ratio, iodine value and major fatty acids (Vigli et al., 2003).
49.5 DETECTION OF REFINED HAZELNUT OIL INTO REFINED OLIVE OIL The detection of refined hazelnut oil into refined olive oil is one of the most challenging tasks confronting oil chemists as the refining process removes many of the components that can assist as adulteration markers. Furthermore, because in this scenario both oils have been refined, use of parameters or markers that indicate refining has taken place (e.g. level of stigmastadienes) does not help as these are now expected to be present even in non-adulterated oils. One of the approaches used to detect refined-into-refined adulterations is the use of the campestadiene to stigmastadiene ratio. As seen in Table 49.3, hazelnut oil generally contains a greater level of campesterol than olive oil (which is a fruit oil) and thus is enriched with a greater amount of campestadiene upon refining. Therefore, an elevated level of campestadiene with respect to stigmastadiene may be indicative of the presence of refined hazelnut oil in refined olive oil. However, additional confirmation is usually needed in this case in order to obtain firm evidence of adulteration (MAFF, 1999).
with very little sample preparation and no chromatographic analysis thus saving considerable time. Goodacre and coworkers were one of the first groups to report that direct infusion ESI-MS was able to provide spectra that could be used, following principal component analysis (PCA), to distinguish between refined hazelnut oil and refined olive oil (Goodacre et al., 2002). The ESI-MS spectra of both oils are very similar and they are dominated by triacyl-, diacyland monoacylglycerols (Figure 49.6). This is to be expected as these components form the major fraction in both oils and they are not lost upon refining. However, the small peak at m/z 897 (Figure 49.6), attributed to trilinoleoylglycerol (LLL), could be useful in detecting adulteration as it is only seen in hazelnut oil. Other differences between the ESI spectra from the two oils are the higher relative abundance of trioleoylglycerol (OOO) and dioleyllinoleylglycerol (LOO) in the mass spectra from hazelnut oil compared to those from olive oil (Table 49.6) (Gómez-Ariza et al., 2006a). These differences are still apparent even at 10% adulteration and combined with the presence of LLL could be reliably used to identify admixtures (Gómez-Ariza et al., 2006a). APPI-MS offers additional information related to differences in the relative levels of monoacyl- and diacylglycerol fragments between the spectra from the two oils as APPI is more sensitive than ESI towards these smaller ions (Table 49.7) (Gómez-Ariza et al., 2006a). When data from these two sources are combined they can be a very powerful tool in the fight against olive oil fraud. Further work is, however, required if these MS-based approaches are to become the officially recommended testing methods to confirm olive oil adulteration by hazelnut and other seed oils.
SUMMARY POINTS ●
49.5.1 Mass Spectrometric Approaches Recent advances in mass spectrometry are starting to be very promising as they offer rapid, accurate and robust approaches for detecting even the most sophisticated adulterations. Soft ionization mass spectrometry techniques, such as matrix-assisted laser desorption ionization (MALDI), electrospray ionization (ESI), atmospheric pressure chemical ionization (APCI) and more recently atmospheric pressure photoionization (APPI), have become invaluable tools in many scientific fields as they are able to provide details related to the structure of molecules analyzed (Goodacre et al., 2002; Gómez-Ariza et al., 2006a,b). Coupled to modern high-resolution tandem mass spectrometers these ionization techniques can be carried out
447
●
●
●
Hazelnut oil is one of the most challenging adulterants to detect in olive oil due to the very similar chemical composition of the two oils. Financial gains by fraudulent producers/processors are the cause of the problem. Adulteration of olive oil by hazelnut oil can occur in three ways: virgin hazelnut into virgin olive, refined hazelnut into virgin olive, and refined hazelnut into refined olive. Adulterations that fit into the last category are usually the most challenging to identify and confirm. Although numerous approaches have been proposed for the detection and confirmation of olive oil adulteration by hazelnut oil, only few of them are able to reliably detect adulterations much lower than 10%. None of the methods available to date can be used to predict the adulteration level of an unknown sample due to the marked variability in oil components.
448
SECTION | I
Pesticides and Adulterants
FIGURE 49.6 Full scan ESI-MS spectra in the positive mode of an olive oil (A) and a hazelnut oil (B). The profiles from both oils are very similar. However, the arrow in the magnified 800–900 mass-to-charge (m/z) region of the hazelnut profile indicates the trilinoleoylglycerol peak at m/z 897 seen only in hazelnut oil. Reprinted and modified from Gómez-Ariza, J.L., et al., 2006. Talanta 70: 859–869, with permission.
CHAPTER | 49 Olive Oil Adulteration with Hazelnut Oil and Analytical Approaches for Its Detection
449
TABLE 49.6 Relative abundance (%) of triacylglycerols (TAGs) in ESI-MS spectra obtained from olive oil, hazelnut oil and their blends. TAGs
Hazelnut oil in sample (%) 0
10
30
50
100
OOO
26.92 ⫾ 0.52
28.64 ⫾ 0.20
29.87 ⫾ 0.28
30.21 ⫾ 0.33
31.16 ⫾ 0.63
LOO
11.76 ⫾ 0.22
8.75 ⫾ 0.10
9.33 ⫾ 0.05
10.61 ⫾ 0.03
18.23 ⫾ 0.20
LLL
n/d
0.39 ⫾ 0.03
0.44 ⫾ 0.01
0.59 ⫾ 0.02
1.44 ⫾ 0.07
The values of the three TAGs increase as the percentage of hazelnut oil in olive oil increases and hence they can be useful in detecting adulteration. TAGs: triacylglycerols; ESI-MS: electrospray-mass spectrometry; OOO: 1,2,3trioleoylglycerol; LOO: 2,3-dioleylinoleylglycerol; LLL: 1,2,3-trilinoleoylglycerol. Reprinted and modified from GómezAriza, J.L., et al., 2006. Talanta 70: 859–869, with permission.
TABLE 49.7 Relative abundance (%) of fragment ions from OOO, [OO]⫹ and [O]⫹ in APPIMS spectra obtained from olive oil, hazelnut oil and their blends. Fragment
Hazelnut oil in sample (%) 0
10
30
50
100
OOO
5.14 ⫾ 0.45
4.37 ⫾ 0.15
5.15 ⫾ 0.12
5.43 ⫾ 0.03
6.23 ⫾ 0.13
[OO]⫹
23.75 ⫾ 1.17
25.72 ⫾ 0.17
27.06 ⫾ 0.53
29.04 ⫾ 0.19
28.88 ⫾ 0.08
[O]⫹
25.15 ⫾ 0.02
27.29 ⫾ 0.09
29.12 ⫾ 0.21
30.70 ⫾ 0.35
28.31 ⫾ 0.41
The values of the three fragments increase as the percentage of hazelnut oil in olive oil increases and hence they can be useful in detecting adulteration. APPI-MS: atmospheric pressure photoionization-mass spectrometry; OOO: 1,2, 3-trioleoylglycerol; [OO]⫹: dioleoylglycerol fragment; [O]⫹: monoleoylglycerol fragment. Reprinted and modified from Gómez-Ariza, J.L., et al., 2006.Talanta 70: 859–869, with permission.
●
●
Most analytical approaches utilize natural or unnatural components in the oils as adulteration markers. Triacyglycerols and their fragments, fatty acids, sterols and their dehydration products are the oil components that are mostly employed as markers. Emerging high-resolution mass spectrometric approaches are likely to dominate future activities in the field due to their rapid, information-rich outputs that can be used to discriminate amongst the most complex matrices.
REFERENCES Aparicio, R., Aparicio-Ruiz, R., 2000. Authentication of vegetable oils by chromatographic techniques. J. Chromatogr. A 881, 93–104. Blanch, G.P., del Mar Caja, M., Ruiz del Castillo, M.L., Herraiz, M., 1999. A contribution to the study of the enantiomeric composition of a chiral constituent in hazelnut oil used in the detection of adulterated olive oil. Eur. Food Res. Technol. 210, 139–143. Cercaci, L., Rodriguez-Estrada, M.T., Lercker, G., 2003. Solid-phase extraction-thin-layer chromatography-gas chromatography method for
the detection of hazelnut oil in olive oils by determination of esterified sterols. J. Cromatogr. A 985, 211–220. Cert, A., Moreda, W., Perez-Camino, M.C., 2000. Chromatographic analysis of minor constituents in vegetable oils. J. Chromatogr. A 881, 131–148. Christopoulou, E., Lazaraki, M., Komaitis, M., Kaselimis, K., 2004. Effectiveness of determinations of fatty acids and triglycerides for the detection of adulteration of olive oils with vegetable oils. Food Chem. 84, 463–474. European Union Research Committee, 2001. Development and assessment of methods for the detection of adulteration of olive oil with hazelnut oil. Press Release. Brussels, Belgium. Fallico, B., Arena, E., Zappalà, M., 2003. Roasting of hazelnuts. Role of oil in colour development and hydroxymethylfurfural formation. J. Agric. Food Chem. 81, 569–573. Flores, G., Ruiz del Castillo, M.L., Blanch, G.P., Herraiz, M., 2006a. Detection of the adulteration of olive oils by solid phase microextraction and multidimensional gas chromatography. Food Chem. 97, 336–342. Flores, G., Ruiz del Castillo, M.L., Herraiz, M., Blanch, G.P., 2006b. Study of the adulteration of olive oil with hazelnut oil by on-line coupled high performance liquid chromatographic and gas chromatographic analysis of filbertone. Food Chem. 97, 742–749.
450
Goodacre, R., Vaidyanathan, S., Bianchi, G., Kell, D.B., 2002. Metabolic profiling using direct infusion electrospray ionisation mass spectrometry for the characterisation of olive oils. Analyst 127, 1457–1462. Gordon, M.H., Firman, C., 2001. Effects of heating and bleaching on formation of stigmastadienes in olive oil. J. Sci. Food Agric. 81, 1530–1532. Gómez-Ariza, J.L., Arias-Borrego, A., Garcia-Barrera, T., Beltran, R., 2006a. Comparative study of electrospray and photospray ionisation sources coupled to quadrupole time-of-flight mass spectrometer for olive oil authentication. Talanta 70, 859–869. Gómez-Ariza, J.L., Arias-Borrego, A., Garcia-Barrera, T., 2006b. Use of flow injection atmospheric pressure photoionisation quadrupole timeof-flight mass spectrometry for fast olive oil fingerprinting. Rapid Commun. Mass Spectrom. 20, 1181–1186. International Olive Oil Council, 2006a. Detection of hazelnut oil in olive oil, Resolution No. Res-3/94-V/06, Madrid, Spain. International Olive Oil Council, 2006b. Global method for the detection of extraneous oils in olive oil, COI/T.20/Doc. No. 25. Jung, M.Y., Bock, J.Y., Back, S.O., Lee, T.K., Kim, J.H., 1997. Pyrazine contents and oxidative stabilities of roasted soybean oils. Food Chem. 60, 95–102. Lanzón, A., Albi, T., Cert, A., Gracián, J., 1994. The hydrocarbon fraction of virgin olive oil and changes resulting from refining. JAOCS 71, 285–291. Magaletta, R.L., Ho, C.-T., 1996. Effect of roasting time and temperature on the generation of nonvolatile (polyhudroxyalkyl)pyrazine compounds in peanuts, as determined by high-performance liquid chromatography. J. Agric. Food Chem. 44, 2629–2635.
SECTION | I
Pesticides and Adulterants
Mariani, C., Bellan, G., Lestini, E., Aparicio, R., 2006. The detection of the presence of hazelnut oil in olive oil by free and esterified sterols. Eur. Food Res. Technol. 223, 655–661. Ministry of Agriculture, Fisheries and Forestry, UK, 1999. Authenticity of olive oil. Food surveillance information sheet 180. Parcerisa, J., Casals, I., Boatella, J., Codony, R., Rafecas, M., 2000. Analysis of olive and hazelnut oil mixtures by high-performance liquid chromatography-atmospheric pressure chemical ionisation mass spectrometry of triacylglycerols and gas-liquid chromatography of non-saponifiable compounds (tocopherols and sterols). J. Chromatogr. A 881, 149–158. Parcerisa, J., Richardson, D.G., Rafecas, M., Codony, R., Boatella, J., 1997. Fatty acid distribution in polar and non-polar lipid classes of hazelnut oil (Corylus avellana L.). J. Agric. Food Chem. 45, 3887–3890. Ruiz del Castillo, M.L., Flores, G., Herraiz, M., Blanch, G.P., 2003. Solid phase microextraction for studies on the enantiomeric composition of filbertone in hazelnut oils. J. Agric. Food Chem. 51, 2496–2500. Vigli, G., Philippidis, A., Spyros, A., Dais, P., 2003. Classification of edible oils by employing 31P and 1H NMR spectroscopy in combination with multivariate statistical analysis. A proposal for the detection of seed oil adulteration in virgin olive oils. J. Agric. Food Chem. 51, 5715–5722. Zabaras, D., Gordon, M. H., 2003. Detection of the adulteration of olive oil by hazelnut oil. Technical Report Q0158 (UK Food Standards Agency), School of Food Biosciences, University of Reading. Zabaras, D., Gordon, M.H., 2004. Detection of pressed hazelnut oil in virgin olive oil by analysis of polar components: improvement and validation of the method. Food Chem. 84, 475–483.
Chapter 50
Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy Betül Öztürk, Aysun Ankan and Durmus¸ Özdemir
. . I zmir Institute of Technology, Faculty of Science, Department of Chemistry, Gülbahçe, Urla, I zmir, Turkey
50.1 INTRODUCTION Olive oil is a valuable food product as compared with other vegetable oils due to its distinct taste, flower and possible health benefits. The economic value of olive oil is generally much higher than other seed oils. As a result, the adulteration of olive oil with cheaper vegetable oils becomes a real concern. For this reason, the analysis of edible oils for possible adulterants is very important for food safety and protection of consumers. Based on the extraction method used, there are various types of olive oil on the market today. Extra virgin olive oil is obtained from the olive by purely mechanical means, and the lower grade oils are obtained by solvent extraction, heat treatment, esterification or refining. The composition of the oils is based on the fatty acids present and their locations on the glycerol backbone. This composition varies not only with the type of oil and extraction method but also with the geographical origin and meteorological effects during the growth and harvest of the olives (Tay et al., 2002). This variation can be used for oil authentication and the identification of adulteration. Various physical and chemical tests have been used to establish the authenticity of olive oil and to detect the level of adulterants in it (Aparicio et al., 1997; Dennis, 1998; Christopouloua et al., 2004). Studies related to olive oil adulteration were mostly carried out with chromatographic methods in recent years (Wenzla et al., 2002; Ghosh et al., 2005; Hajimahmoodi et al., 2005). However, while chromatographic methods offer high sensitivity and accuracy, they are also time-consuming and expensive. On the other hand, spectroscopic methods may offer faster and cheaper analysis alternatives (Papadopoulos et al., 2002; Schulz et al., 2003; Guimet et al., 2005). Molecular fluorescence spectroscopy is a sensitive technique for differentiating various seed oils from olive oils as Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
they exhibit natural fluorescence (Kyriakidis and Skarkalis, 2000). However, when multicomponent oil mixtures need to be quantified, somewhat broad overlapping excitation and emission peaks make it necessary to use multivariate calibration methods. Nevertheless, there have been an increasing number of reports in recent years about the use of excitation-emission fluorescence (EEF) spectroscopy and synchronous fluorescence (SF) spectroscopy for the determination of olive oil adulteration with cheaper oils (Francesca et al., 2004a,b; Konstantina et al., 2006). In EEF, an excitation-emission matrix (EEM) which consists of emission spectra measured at different excitation wavelengths, is recorded for a sample. In SF mode, the excitation wavelength is increased with a constant wavelength increment while the emission spectrum is recorded in a predefined wavelength range, thus a constant wavelength interval is maintained between excitation and emission wavelengths (Δλ) as the scan progresses. As a result, the SF method produces a two-dimensional fluorescence spectrum for a given sample and the EEF method produces a three-dimensional fluorescence profile for each sample. The time required to collect an EEF spectrum is longer as it generates several emission spectra for a given sample when compared to SF mode but the data contain much richer information and could result in better characterizations and quantifications. Multivariate calibration methods make it possible to relate instrument responses that consist of several predictor variables to a chemical or physical property of a sample. Several classical multivariate calibration methods have been developed (Lindberg et al., 1983; Geladi and Kowalski, 1986; Haaland and Thomas, 1988; Wentzell et al., 1997) in the last couple of decades for the analysis of complex chemical mixtures, and the choice of the most suitable calibration method is very important in order to
451
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
452
SECTION | I
generate calibration models with high predictive ability for future samples. In some cases conventional methods may not offer a satisfactory solution to a given problem due to the complexity of the data and it may be necessary to apply some sort of variable selection. There have been many mathematical methods of variable selection and genetic algorithm is one of them offering a fast and effective solution for large-scale problems (Lucasius and Kateman, 1993; Horchner and Kalivas, 1995; Özdemir and Williams, 1999; Özdemir and Öztürk, 2004, 2007). Inverse least squares (ILS) method is based on the inverse of Beer’s Law where concentration of an analyte is modeled as a function of absorbance measurements. Genetic inverse least squares (GILS) is a modified version of the original ILS method in which a small set of wavelengths is selected from a full-spectral data matrix and evolved to an optimum solution using a genetic algorithm (GA) and has been applied to a number of wavelength selection problems. GAs are non-local search and optimization methods that are based upon the principles of natural selection. In this work, a genetic algorithm-based calibration method called genetic inverse least squares (GILS) was tested with the aim of establishing calibration models that have a high predictive ability for the determination of olive oil adulteration with sunflower oil and corn oil using EEF and SF spectroscopy.
50.2 METHODOLOGICAL CONSIDERATIONS The major drawback of the classical least squares (CLS) method is that all of the interfering species must be known and their concentrations included in the model. This need can be eliminated by using the inverse least squares (ILS) method which uses the inverse of Beer’s Law. In the ILS method, concentration of a component is modeled as a function of absorbance measurements. Because modern spectroscopic instruments are very stable and provide excellent signal-to-noise ratios, it is believed that the majority of errors lie in the reference values of the calibration sample, not in the measurement of their spectra. In fact, in many cases the concentration data of calibration set is generated from another analytical technique that already has its inherent errors which might be higher than those of the spectrometer (for example, Kjeldahl protein analysis used to calibrate near infrared spectra). The ILS model for m calibration samples with n wavelengths for each spectrum is described by: C ⫽ AP ⫹ EC
(50.1)
where C is the m ⫻ l matrix of the component concentrations, A is the m ⫻ n matrix of the calibration spectra, P is the n ⫻ l matrix of the unknown calibration coefficients relating l component concentrations to the spectral intensities
Pesticides and Adulterants
and EC is the m ⫻ l matrix of errors in the concentrations not fit by the model. In the calibration step, ILS minimizes the squared sum of the residuals in the concentrations. The biggest advantage of ILS is that Equation 50.1 can be reduced for the analysis of single component at a time since analysis based on an ILS model is invariant with respect to the number of chemical components included in the analysis. The reduced model is given as: c ⫽ Ap ⫹ e c
(50.2)
where c is the m ⫻ 1 vector of concentrations for the component that is being analyzed, p is n ⫻ 1 vector of calibration coefficients and ec is the m ⫻ 1 vector of concentration residuals not fit by the model. During the calibration step, the least-squares estimate of p is: pˆ ⫽ (A ′A)⫺1 A ′ ⭈ c
(50.3)
where pˆ are the estimated calibration coefficients. Once pˆ is calculated, the concentration of the analyte of interest can be predicted with the equation below. cˆ ⫽ a ′ ⭈ pˆ
(50.4)
where cˆ is the scalar estimated concentration and a is the spectrum of the unknown sample. The ability to predict one component at a time without knowing the concentrations of interfering species has made ILS one of the most frequently used calibration methods. The major disadvantage of ILS is that the number of wavelengths in the calibration spectra should not be more than the number of calibration samples. This is a big restriction since the number of wavelengths in a spectrum will generally be much more than the number of calibration samples and the selection of wavelengths that provide the best fit for the model is not a trivial process. Several wavelength selection strategies, such as stepwise wavelength selection and all possible combination searches, are available to build an ILS model which fits the data best. Genetic algorithms (GA) are global search and optimization methods based upon the principles of natural evolution and selection as developed by Darwin. Computationally, the implementation of a typical GA is quite simple and consists of five basic steps including initialization of a gene population, evaluation of the population, selection of the parent genes for breeding and mating, crossover and mutation, and replacing parents with their offspring. These steps have taken their names from the biological foundation of the algorithm. Genetic inverse least squares (GILS) is an implementation of a GA for selecting wavelengths to build multivariate calibration models with reduced data set. GILS follows the same basic initialize/breed/mutate/evaluate algorithm as other GAs to select a subset of wavelengths but is unique in the way it
CHAPTER | 50 Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy
encodes genes. A gene is a potential solution to a given problem and the exact form may vary from application to application. Here, the term gene is used to describe the collection of instrumental response at the wavelength range given in the data set. The term ‘population’ is used to describe the collection of individual genes in the current generation. In the initialization step, the first generation of genes is created randomly with a fixed population size. Although random initialization helps to minimize bias and maximize the number of possible recombinations, GILS is designed to select initial genes in a somewhat biased random fashion in order to start with genes better suited to the problem than those that would be randomly selected. Biasing is done with a correlation coefficient by plotting the predicted results of initial population against the actual component concentrations. The size of the gene pool is a user-defined even number in order to allow breeding of each gene in the population. It is important to note that the larger the population size, the longer the computation time. The number of instrumental responses in a gene is determined randomly between a fixed low limit and high limit. The lower limit was set to 2 in order to allow single point crossover whereas the higher limit was set to eliminate overfitting problems and reduce the computation time. Once the initial gene population is created, the next step is to evaluate and rank the genes using a fitness function, which is the inverse of the standard error of calibration (SEC). The third step is where the basic principle of natural evolution is put to work for GILS. This step involves the selection of the parent genes from the current population for breeding using a roulette wheel selection method according to their fitness values. The goal is to give a higher chance to those genes with high fitness so that only the best-performing members of the population will survive in the long run and will be able to pass their information to the next generations. Because of the random nature of the roulette wheel selection method, however, genes with low fitness values will also have some chance to be selected. Also, there will be genes that are selected multiple times and some genes will not be selected at all and will be thrown out of the gene pool. After the selection procedure is completed, the selected genes are allowed to mate top–down in pairs whereby the first gene mates with the second gene and the third one with the fourth one and so on as illustrated in the following example: Parents S1 ⫽ ( A347 , A951 , # A479 , A518 ) S2 ⫽ ( A625 , A378 , A568 , # A743 , A750 , A451 , A558 , A631 , A758 )
(50.5) (50.6)
The points where the genes are cut for mating are indicated by #.
453
Offspring S3 ⫽ ( A347 , A951 , A743 , A750 , A451 , A558 , A631 , A758 ) S4 ⫽ ( A625 , A378 , A568 , A479 , A518 )
(50.7) (50.8)
where A347 represents the instrument response at the wavelength given in subscript, S1 and S2 represent the first and second parent genes and S3 and S4 are the corresponding genes for the offspring. Here the first part of S1 is combined with the second part of the S2 to give the S3, likewise the second part of the S1 is combined with the first part of the S2 to give S4. This process is called the single point crossover and is common in GILS. Single point crossover will not provide different offspring if both parent genes are identical, which may happen in roulette wheel selection, when both genes are broken at the same point. Also note that mating can increase or decrease the number of instrument responses in the offspring genes. After crossover, the parent genes are replaced by their offspring and the offspring are evaluated. The ranking process is based on their fitness values following the evaluation step. Then the selection for breeding/mating starts all over again. This is repeated until a predefined number of iterations is reached. Mutation which introduces random deviations into the population was also introduced into the GILS during the mating step at a rate of 1% as is typical in GAs. This is usually done by replacing one of the responses in an existing gene with a randomly selected new one. Mutation allows the GILS to explore the search space and incorporate new material into the genetic population. It helps keep the search moving and can eject GILS from a local minimum on the response surface. However, it is important not to set the mutation rate too high since it may keep the GA from being able to exploit the existing population. Also, the GILS method is an iterative algorithm and therefore there is a high possibility that the method may easily overfit the calibration data so that the predictions for independent sets might be poor. To eliminate possible overfitting problems, cross validation is used in which one spectrum is left out of the calibration set and the model is constructed with m ᎐ 1 sample. Then this model is used to predict the concentration of left out sample. This process is continued until all samples are left out at least once in each iteration. As long as the number of spectra in the calibration set is not too large, cross validation is an effective method of eliminating overfitting. If the number of calibration spectra is very large, then the GILS method has the option of half validation approach in which the half of the spectra in the calibration set is used to validate the model in each iteration. In the end, the gene with the lowest SEC (highest fitness) is selected for the model building and this model is used to predict the concentrations of component being analyzed in the prediction (test) sets. The success of the model in the prediction of the test sets is evaluated using standard
454
SECTION | I
error of prediction (SEP). Because random processes are heavily involved in GILS as in all the GAs, the program has been set to run several times for each component in this study. The best run (i.e. the one generating the lowest SEC for the calibration set and at the same time producing SEPs for prediction sets that are in the same range with the SEC) is subsequently selected for evaluation and further analysis. The termination of the algorithm can be done in many ways. The easiest way is to set a predefined iteration number for the number of breeding/mating cycles. GILS has some major advantages over classical univariate and multivariate calibration methods. First of all, it is quite simple in terms of the mathematics involved in the model building and prediction steps, but at the same time it has the advantages of the multivariate calibration methods with a reduced data set since it uses the full spectrum to extract genes. By selecting a subset of instrument responses it is able to eliminate non-linearities that might be present in the full spectral region.
Pesticides and Adulterants
The GILS method was applied to the fluorescence spectra of ternary mixtures of olive oil, sunflower oil, and corn oil. The fluorescence spectra of 50 ternary samples of olive oil, corn oil and sunflower oil were measured in EEF and SF mode using a Varian Cary Eclipse spectrophotometer (Varian, Inc. Hansen Way, Palo Alto, CA) equipped with a xenon flash lamp. Olive oil, sunflower oil and corn oil samples were purchased from a local grocery store. The EEF spectra were collected between 320 and 800 nm emission wavelength by exciting the samples with a wavelength increment of 15 nm (Δλ) from 320 to 425 nm. The SF spectra of the samples were recorded between 250 and 750 nm with a Δλ of 20 nm. The slit widths of excitation and emission monochromators were set to 5 nm in both EEF and SF modes. All spectra were then transferred to a computer where the data-processing programs were installed. Among the 50 ternary mixtures, 34 of the sample were randomly selected with the exception that the samples with the lowest and highest concentration of each component were
TABLE 50.1 Percent composition of calibration set used in ternary mixtures of olive oil, corn oil and sunflower oil for both excitation-emission fluorescence (EEF) and synchronous fluorescence (SF) data sets. S. N.
Olive oil (w/w %)
Corn oil (w/w %)
Sunflower oil (w/w %)
S.N.
Olive oil (w/w %)
Corn oil (w/w %)
Sunflower oil (w/w %)
1
59.87
29.06
11.07
18
68.93
15.01
16.06
2
65.97
26.98
7.05
19
73.89
1.04
25.07
3
61.85
22.08
16.07
20
89.86
4.05
6.09
4
74.86
16.05
9.09
21
73.91
18.00
8.09
5
90.90
3.03
6.07
22
70.96
20.07
8.97
6
88.00
4.00
8.00
23
73.09
11.96
14.95
7
80.93
8.01
11.05
24
97.99
1.01
1.00
8
68.06
22.99
8.95
25
68.86
18.08
13.06
9
71.93
27.02
1.04
26
60.01
14.02
25.97
10
76.91
20.98
2.11
27
72.04
19.02
8.94
11
70.06
29.94
0.00
28
69.93
5.04
25.03
12
70.00
0.00
30.00
29
88.98
2.04
8.98
13
89.90
9.04
1.06
30
84.91
3.00
12.09
14
94.86
2.11
3.03
31
79.09
20.91
0.00
15
66.98
19.94
13.08
32
63.02
5.08
31.90
16
87.04
12.96
0.00
33
62.96
28.03
9.01
17
75.96
0.00
24.04
34
72.97
10.04
17.00
The same calibration set was used for both excitation-emission fluorescence (EEF) and synchronous fluorescence (SF) data. The total of 34 calibration samples were randomly selected from 50 samples.
CHAPTER | 50 Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy
455
TABLE 50.2 Percent composition of prediction set used in ternary mixtures of olive oil, corn oil and sunflower oil for both excitation-emission fluorescence (EEF) and synchronous fluorescence (SF) data sets. S. N.
Olive oil (w/w %)
Corn oil (w/w %)
Sunflower oil (w/w %)
1
77.99
3.97
18.04
2
90.98
2.05
3
59.93
4
S.N.
Olive oil (w/w %)
Corn oil (w/w %)
Sunflower oil (w/w %)
9
80.92
13.00
6.08
6.96
10
69.93
17.02
13.05
15.07
25.00
11
78.94
18.04
3.01
64.05
18.98
16.97
12
76.95
3.98
19.07
5
84.88
5.05
10.07
13
87.94
7.12
4.94
6
82.96
10.94
6.10
14
66.01
8.06
25.94
7
74.00
10.97
15.04
15
78.96
18.97
2.08
8
65.94
12.02
22.04
16
85.93
1.02
13.04
The same prediction set was used for both excitation-emission fluorescence (EEF) and synchronous fluorescence (SF) data. The total of 16 prediction samples were randomly selected from 50 samples.
intentionally put in the calibration set in order to construct calibration models as shown in Table 50.1. Table 50.2 shows the concentration profiles of the remaining 16 samples for the validation set as weight percent (Wt-%). The same calibration and validation sets were used in both EEF and SF data. The genetic inverse least squares (GILS) method was written in MATLAB programming language using Matlab 5.3 (MathWorks Inc, Natick, MA).
50.3 FEATURES OF MOLECULAR FLUORESCENCE SPECTROSCOPY Fluorescence spectroscopic measurements can be carried out from simple steady-state emission intensity to quite sophisticated time-resolved measurements. Although fluorescence measurements do not provide detailed structural information, fluorescence spectroscopy is gaining interest in many areas of science for quantitative analysis of complex mixtures with the help of advanced multivariate statistical tools. Fluorescence occurs in simple as well as in complex gaseous, liquid, and solid chemical systems. While fluorescence can be observed from almost all molecules with an excitation beam in adequate intensity only a small part of molecules demonstrates fluorescence characteristics which are desirable for analytical purposes. Therefore, fluorescence spectroscopy is less universal than absorption techniques although it is more selective. However, in some applications in terms of its lower detection limits and greater selectivity, fluorescence spectrometry is a preferred technique to molecular absorption spectrometry. Even if many species in a sample are fluorescent, selectivity is
always improved by a suitable choice of excitation and emission wavelengths. Fluorescence emission is a transition between electronic states of the same multiplicity and involves a singlet–singlet transition. Emission occurs from the ground vibrational level of excited electronic states (S1 or S2) to various vibrational levels in ground electronic state (S0). Fluorescence usually appears at longer wavelengths than absorption as absorption transitions are higher excited electronic states. For the analysis of multifluorophoric systems the widely used methods are the excitation emission fluorescence (EEF) spectroscopy and synchronous fluorescence (SF) spectroscopy. EEF spectroscopy is a rapid and inexpensive technique and is also known as total fluorescence spectroscopy (TFS) which provides a ‘fingerprint’ consisting of a 3-D emission/excitation intensity diagram. This ‘fingerprint’ along with multivariate calibration can be used for the qualitative and quantitative information about the multifluorophores present in the sample. Synchronous fluorescence spectroscopy is a highly sensitive and simple technique. In SF spectroscopy both the excitation and emission monochromators are simultaneously scanned at a constant wavelength interval between emission and excitation wavelengths (Δλ), so that spectral overlaps are reduced and the spectra is simplified.
50.4 APPLICATION OF MOLECULAR FLUORESCENCE SPECTROSCOPY TO OLIVE OIL ADULTERATION Fluorescence spectroscopy has become a popular spectroscopic technique due to its high sensitivity and selectivity.
456
SECTION | I
Corn oil
250
250
200
200 Intensity
Intensity
Olive oil
150 100
150 100
50
50
0 800
0 800 600
Em.(nm)
400 320
340
360
380
400
420
600 Em.(nm)
Ex.(nm)
250
200
200 Intensity
Intensity
250
150 100
0 800
0 800
400
320
340
360
420
Ex.(nm)
100 50
Em.(nm)
320
400
150
50
420 400 380 360 Ex.(nm) 340
400
380
Ternary mixture
Sunflower oil
600
Pesticides and Adulterants
600 Em.(nm)
400
320
420 400 380 360 Ex.(nm) 340
FIGURE 50.1 Excitation and emission fluorescence spectra show the maximum emission intensities between 320 and 800 nm emission wavelengths and between 320 and 425 excitation wavelengths for each component along with their ternary mixture. Olive oil when compared with corn oil and sunflower oil gives quite different emission profile.
Determination of olive oil adulteration with corn and sunflower oils was conducted using fluorescence spectroscopy coupled with genetic multivariate calibration. The EEF spectra of pure olive, sunflower and corn oil and their ternary mixture between the 320 and 800 nm emission wavelengths and excitation wavelengths ranging from 320 to 425 nm are shown in Figure 50.1. As seen from the figure, the EEF spectra of corn oil and sunflower oil are very much alike, showing maximum fluorescence emission intensity around 500 nm and strongest excitation around 380 nm. Pure olive oil has maximum fluorescence emission profile around 700 nm with an excitation wavelength of 410 nm. In addition, olive oil also gives a weaker emission peak around 500 nm which overlaps with corn oil and sunflower emission profile. Synchronous fluorescence spectra of olive oil, corn oil, and sunflower oil along with their ternary mixture between 250 and 750 nm are shown in Figure 50.2. The spectra
were divided into two parts in order to better illustrate the emission profile around 380 nm as the intensity at this lower wavelength is about 10 times lower compared to the peak around 670 nm which is only seen for olive oil. As can be seen from the figure, corn oil and sunflower oil have very similar synchronous fluorescence emission with a maximum intensity around 350 nm, whereas olive oil emission is distinctly different. A total of 50 ternary mixtures of olive oil, corn oil and sunflower oil were prepared in order to prepare multivariate calibration models. The calibration models were prepared with 34 samples as given in Table 50.1 and then these models were tested with 16 independent prediction samples shown in Table 50.2 for both EEF and SF data. Because of the random nature of the GILS method, the program was set to run 100 times with 30 genes and 50 iterations. Since the GILS program was iterated 50 times in each run, full cross validation was applied during the model building
CHAPTER | 50 Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy
100
450 Olive Oil
90
400
Corn Oil
80
Sunflower Oil
70
Ternary Mixture
350 Fluorescence intensity
Fluorescence intensity
457
60 50 40 30
300 250 200 150
20
100
10
50
0 250
350 450 Wavelength (nm)
550
0 550
600 650 700 Wavelength (nm)
750
FIGURE 50.2 Synchronous fluorescence emission spectra of olive oil, corn oil and sunflower oil along with their ternary mixture between 250 and 750 nm were divided into two parts in order to illustrate high fluorescence emission of olive oil around 670 nm where no significant emission is seen for corn oil and sunflower oil.
TABLE 50.3 Standard error of calibration (SEC) and standard error of prediction (SEP) results for both excitation-emission fluorescence (EEF) and synchronous fluoresecence (SF) data sets. Data sets
SEC and SEP
Olive oil
Corn oil
Sunflower oil
EEF data set
SEC (w/w %)
0.58
0.51
0.61
SEP (w/w %)
0.64
0.90
1.07
SEC (w/w %)
0.63
0.57
0.81
SEP (w/w %)
0.64
1.07
1.21
SF data set
The SEC and the SEP results for both excitation-emission fluorescence (EEF) and synchronous fluorescence (SF) data sets were given in order to compare the sucsess of GILS generating calibration models that have high predictive ability for the independent prediction sets. The results were ranged between 0.51% (w/w) and 1.21% (w/w).
step to avoid possible overfitting problems. The standard error of calibration (SEC) and standard error of prediction (SEP) results for both EEF and SF data sets are shown in Table 50.3. As seen in Table 50.3, the SEC and SEP values ranged between 0.51 and 1.27% by mass for both EEF and SF data sets. Considering the fact that any possible olive oil adulteration attempt may include up to 30% or more vegetable oil by volume or by mass, these values seem to be a good prediction for a fast identification. Furthermore, both calibration and prediction results for EEF data were slightly lower than SF data set indicating that better prediction results are obtained with EEF data. This seems
reasonable since the EEF data were obtained at eight different excitation wavelengths from 320 to 425 nm resulting in a richer fluorescence emission profile when compared with SF data. It is also evident from the table that the calibration model generated with both EEF and SF for olive oil gives better prediction results among the three oils used to prepare ternary mixtures. This is no surprise as the fluorescence spectra of olive oil are distinctly different from corn oil and sunflower oil whereas the latter two have very similar fluorescence emission spectra. The plot of actual versus predicted concentrations for olive oil, corn oil and sunflower oil is illustrated in Figure 50.3 for EEF (on the left) and SF (on the right) data. It is evident that the proposed method is able to predict adulteration of olive oil with corn oil and sunflower oil in a wide dynamic range from 1% to 35% by mass. While the results of olive oil are almost exactly the same for both EEF and SF data sets, the same is not true for corn oil and sunflower oil. The performance of GILS for corn oil and sunflower oil is slightly better for EEF data and this could be due to the partial resolution of their peaks around 500 nm as the excitation wavelength is changed. As a result, it is concluded that both excitation and emission fluorescence spectroscopy in conjunction with multivariate calibration can be used for the fast identification of olive oil adulteration with cheaper substitutes. Because GILS is a wavelength-selection-based method, it is interesting to observe the distribution of selected wavelengths in multiple runs over the entire full spectral region. The frequency distribution of selected wavelengths in 100 runs for olive oil, corn oil and sunflower oil is illustrated in Figures 50.4 and 50.5 for EEF and SF data, respectively.
458
SECTION | I
100 y = 0.997x + 0.2295 R2 = 0.997
Predicted olive oil (w/w%)
Predicted olive oil (w/w%)
100
90
80
70
Calibration Validation
y = 0.9963x + 0.2799 R2 = 0.9963
90
80
70
Calibration Validation
60
60 60
70
80
90
100
60
70
Actual olive oil (w/w%)
90
100
35 y = 0.9973x + 0.0362 R2 = 0.9973
30
Predicted corn oil (w/w%)
Predicted corn oil (w/w%)
80
Actual olive oil (w/w%)
35
25 20 15 10 Calibration
5
Validation
0
y = 0.9966x + 0.0445 R2 = 0.9966
30 25 20 15 10
Calibration
5
Validation
0 0
5
10
15
20
25
30
35
0
5
Predicted sunflower oil (w/w%)
35 y = 0.9955x + 0.0512 R2 = 0.9955
30 25 20 15 10
Calibration
5
Validation
0 0
5
10
15
20
25
10
15
20
25
30
35
Actual corn oil (w/w%)
Actual corn oil (w/w%) Predicted sunflower oil (w/w%)
Pesticides and Adulterants
30
35
Actual sunflower oil (w/w%)
35 y = 0.992x + 0.0905 R2 = 0.992
30 25 20 15 10
Calibration
5
Validation
0 0
5
10
15
20
25
30
35
Actual sunflower oil (w/w%)
FIGURE 50.3 Each graph illustrates the success of the GILS method for the construction of calibration models and the validation of these models with independent test sets which are not used in the model building step. In order to make comparison, actual vs. predicted percent olive oil, corn oil, and sunflower oil contents of the ternary mixtures for excitation-emission (on the left) and synchronous fluorescence (on the right) data are given side by side for each component.
When a close examination is done on Figure 50.4, emission spectra of eight different excitation wavelengths were concatenated in order to illustrate both pure component spectra and selection frequency distribution in a simple two-dimensional plot. The most frequently selected wavelengths for olive oil are located in the first two to three excitation wavelengths whereas the most frequently selected wavelengths for corn oil and sunflower oil are concentrated in the fourth and fifth excitation wavelengths and also a few other higher excitation wavelengths. It is also important to note that the selection frequencies were considerably low
in flat baseline portions of the spectra which indicate that the GILS method is able to focus on the information-rich regions of the spectra even though it starts with a completely random selection of wavelengths. In the case of Figure 50.5, an interesting selection frequency profile for olive oil is seen where the most frequently selected wavelengths are located around 300 nm where there is a small peak. The possible reason for this might be the better linearity of the signal at this wavelength region. The intensity scale is reduced to 25 in order to better illustrate this small peak for olive oil. The selection frequency distributions of
CHAPTER | 50 Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy
Olive oil
27
Selection frequency
22
150
17 100
12
50
7
0 0
500
1000
1500
2000 2500 Cell number
3000
3500
250
27
Selection frequency
150
22 17
100
12
50
7
0 0
500
1000
1500
2000 2500 Cell number
3000
3500
250
Selection frequency
32 Corn oil
200 Intensity
2 4000
2 4000
32 Sunflower oil
200
Selection frequency
150
27 22 17
100
12
50
7
0 0
500
1000
1500
2000 2500 Cell number
3000
3500
Selection frequency
Intensity
200
Selection frequency
32
250
Intensity
459
2 4000
FIGURE 50.4 Three-dimensional data generated from EEF data are concatenated in a way that EEF matrix data were converted to a vector for each sample resulting in a matrix representation of all samples in calibration and validation sets. This is required for GILS in order to generate multivariate calibration models. Distribution of the selected wavelengths for EEF data for olive oil, corn oil and sunflower oil along with their pure component concatenated EEF spectra for eight different excitation wavelengths from 320 to 425 nm shows that GILS selects different excitation and emission wavelengths for each component.
corn oil and sunflower oil samples resulted in expected profiles as shown in the figure. This is a strong indication that the genetic algorithm incorporated into the GILS method is focusing on the regions where most concentration-related information is contained.
●
SUMMARY POINTS
●
●
●
●
Adulteration of olive oil with cheaper substitutes such as sunflower and corn oil is a major concern for the public. Rapid analysis methods are required for a quick and easy screening of possible adulteration attempts. Fluorescence spectroscopy coupled with a genetic algorithm-based multivariate calibration method allows the
determination of olive oil adulteration with sunflower and corn oil. The fact that the standard error of prediction values are all below 1.30% (w/w) for the ternary set, fluorescence spectroscopy can be used as a fast screening method for possible olive oil adulteration with cheaper vegetable oils. In addition, the genetic algorithm used in the GILS method is able to select and extract the most relevant information to build successful calibration models that has high predictive ability for the independent test samples.
ACKNOWLEDGMENT This work was financially supported by the Scientific and Technological Research Council of Turkey (TUBI˙TAK) through Project No: 107T037.
Intensity
20
Olive oil Selection frequency
15 10 5 0 250
300
350
400
450 500 550 Wavelength (nm)
70
Intensity
60 40 30 20 10
Intensity
80 70 60 50 40 30 20 10 0 250
650
700
42 37 32 Selection frequency 27 22 17 12 7 2 600 650 700 750 Corn oil
50
0 250
600
42 37 32 27 22 17 12 7 2 750
300
350
400
450 500 550 Wavelength (nm)
42 37 32 Selection frequency 27 22 17 12 7 2 600 650 700 750 Sunflower oil
300
350
400
450 500 550 Wavelength (nm)
Selection frequency
25
Pesticides and Adulterants
Selection frequency
SECTION | I
Selection frequency
460
FIGURE 50.5 Distribution of the selected wavelengths for SF data set by the GILS method for olive oil, corn oil and sunflower oil along with their pure component SF spectra shows that the method focuses to the regions where corn, olive and sunflower oil gives fluorescence emission. On the other hand, higher selection frequencies for olive oil are seen around 300 nm where a smaller emission peak is given when compared with 670 nm region.
REFERENCES Aparicio, R., Morales, M.T., Alonso, V., 1997. Authentication of European virgin olive oils by their chemical compounds, sensory attributes and consumers’ attitudes, J. Agr. Food Chem. 45, 1076–1083. Christopouloua, E., Lazarakia, M., Komaitisb, M., Kaselimis, K., 2004. Effectiveness of determinations of fatty acids and triglycerides for the detection of adulteration of olive oils with vegetable oils. Food Chem 84, 463–474. Dennis, M.J., 1998. Recent developments in food authentication. Analyst 123, 151R–156R. Francesca, G., Ricard, B., Joan, F., 2004a. Cluster analysis applied to the exploratory analysis of commercial Spanish olive oils by means of excitation-emission fluorescence spectroscopy. J. Agric. Food Chem. 52, 6673–6679. Francesca, G., Joan, F., Ricard, F., Xavier, R., 2004b. Application of unfold principal component analysis and parallel factor analysis to the exploratory analysis of olive oils by means of excitation–emission matrix fluorescence spectroscopy. Anal. Chim. Acta. 515, 75–85. Geladi, P., Kowalski, B.R., 1986. Partial least squares regression: a tutorial. Anal. Chim. Acta. 185, 1–17.
Ghosh, P., Reddy, K.M.M., Sashidhar, R.B., 2005. Quantitative evaluation of sanguinarine as an index of argemone oil adulteration in edible mustard oil by high performance thin layer chromatography. Food Chem. 91, 757–764. Guimet, F., Fer e, J., Boqúe, R., 2005. Rapid detection of olive–pomace oil adulteration in extra virgin olive oils from the protected denomination of origin “Siurana” using excitation–emission fluorescence spectroscopy and three-way methods of analysis. Anal. Chim. Acta 544, 143–152. Haaland, D.M., Thomas, E.V., 1988. Partial least squares methods for spectral analyses. 1. Relation to other quantitative calibration methods and the extraction of qualitative information. Anal. Chem. 60, 1193–1202. Hajimahmoodi, M., Vander, H.Y., Sadeghi, N., Jannat, B., Oveisi, M.R., Shahbazian, S., 2005. Gas-chromatographic fatty-acid fingerprints and partial least squares modeling as a basis for the simultaneous determination of edible oil mixtures. Talanta 66, 1108–1116. Hörchner, U., Kalivas, J.H., 1995. Further investigation on a comparative study of simulated annealing and genetic algorithm for wavelength selection. Anal. Chim. Acta 311, 1–13. Konstantina, I.P., George, A.M., Constantinos, A.G., 2006. Synchronous fluorescence spectroscopy for quantitative determination of virgin olive oil adulteration with sunflower oil. Anal. Bioanal. Chem. 386, 1571–1575.
CHAPTER | 50 Olive Oil Adulteration with Sunflower and Corn Oil Using Molecular Fluorescence Spectroscopy
Kyriakidis, N.B., Skarkalis, P., 2000. Fluorescence spectra measurement of olive oil and other vegetable oils. J. AOAC Int. 83, 1435–1438. Lindberg, W., Persson, J.A., Wold, S., 1983. Partial least-squares method for spectrofluorimetric analysis of mixtures of humic acid and lignin sulfonate. Anal. Chem. 55, 643–648. Lucasius, C.B., Kateman, G., 1993. Understanding and using genetic algorithms. Part 1. Concepts, properties and context. Chem. Intell. Lab. Syst. 19, 1–33. Özdemir, D., Öztürk, B., 2007. Near infrared spectroscopic determination of olive oil adulteration with sunflower and corn oil. J. Food Drug Anal. 15 (1), 40–47. Özdemir, D., Öztürk, B., 2004. Genetic multivariate calibration methods for near infrared (NIR) spectroscopic determination of complex mixtures. Turk. J. Chem. 28, 497–514. Özdemir, D., Williams, R.R., 1999. Multi-instrument calibration in UV-visible spectroscopy using genetic regression. Appl. Spectrosc. 53, 210–217.
461
Papadopoulos, K., Triantis, T., Tzikis, C.H., Nikokavoura, A., Dimotikali, D., 2002. Investigations of the adulteration of extra virgin olive oils with seed oils using their weak chemiluminescence. Anal. Chim. Acta 464, 135–140. Schulz, H., Quilitzsch, R., Krüger, H., 2003. Rapid evaluation and quantitative analysis of thyme, oregano and chamomile essential oils by ATR-IR and NIR spectroscopy. J. Mol. Struct. 661, 299–306. Tay, A., Singh, R.K., Krishnan, S.S., Gore, J.P., 2002. Authentication of olive oil adulterated with vegetable oils using Fourier transform infrared spectroscopy. Lebensm.-Wiss. u.-Technol. 35, 99–103. Wentzell, P.D., Andrews, D.T., Kowalski, B.R., 1997. Maximum likelihood multivariate calibration. Anal. Chem. 69, 2299–2311. Wenzla, T., Prettnerb, E., Schweigerb, K., Wagnerc, F.S., 2002. An improved method to discover adulteration of Styrian pumpkin seed oil. J. Biochem. Bioph. Meth. 53, 193–202.
This page intentionally left blank
Chapter 51
Benzene, Toluene, Ethylbenzene, (o-, m- and p-) Xylenes and Styrene in Olive Oil Silvia López-Feria, Soledad Cárdenas and Miguel Valcárcel Department of Analytical Chemistry, University of Córdoba, Spain
51.1 INTRODUCTION The identification and quantitation of contaminants in waters, soils and air has systematically been carried out in order to minimize the toxic effect of these compounds in the environment. The presence of certain pollutants in cosmetics and foods is of great concern, however. They can produce irreversible consequences for human health by means of a cumulative effect. Among the potential toxics, single-ring aromatic hydrocarbons constitute priority pollutants because of their carcinogenic nature and wide distribution in environmental and food samples. The chemical structures of benzene, toluene, ethylbenzene, o-, m- and p-xylenes and styrene are shown in Figure 51.1 and their main features are summarized in Table 51.1. They can appear either naturally (viz. superior plants wax, natural oil seepage) or by anthropogenic sources (viz. combustion
products, industrial paints, adhesives, packages). Road transport makes the largest contribution to release benzene and toluene to the atmosphere while solvent usage is behind ethylbenzene and xylene release into the atmosphere. Considering that petrol and gasoline typically contain benzene, toluene, xylene and ethylbenzene in a 3-1-4-2 ratio, it seems easy to identify if the presence of such compounds in a given sample has an endogenous or external source. Several studies have been led by the European Union and the British Ministry of Agriculture, Fisheries and Food in the period 1993–1996 in order to determine the potential sources of human exposure to BTEXS. Both conclude that the main source is by inhalation from ambient air. The contribution of each contaminant varies depending on the atmosphere’s composition (rural or urban) but also on the diet. If the average exposure from ambient can be set at ca. H2C
CH3
Benzene
Toluene
CH3
Ethylbenzene
CH3
CH3
CH3
HC
CH3
CH3 CH3 (o+m+p) Xylene
CH3
Styrene
FIGURE 51.1 Chemical structures of the benzene hydrocarbons dealt with in this chapter. This figure shows the chemical structures for benzene, toluene, ethylbenzene, xylene isomers and styrene. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
463
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
464
SECTION | I Toxicology and Contaminants
TABLE 51.1 Key features of BTEXS as a group of chemicals. 1. 2. 3. 4.
BTEXS is the acronym of benzene, toluene, ortho, meta and para xylene isomers and styrene BTEXS are a subclass of volatile organic compounds with boiling points between 80–150°C Chemically, they are single aromatic hydrocarbons that can also present an alkyl substituent Natural sources of BTEXS include superior plant algae, plankton, oil seepage and decarboxylation of cinnamic acid (styrene) 5. Anthropogenic sources comprise combustion products of wood and fuels, industrial paints, adhesives, degreasing agents and aerosols 6. The effects of exposure to these substances comprise changes in the liver and harmful effect on the kidneys, heart, lungs and nervous system This table lists the key features of BTEXS including their definition, chemical characteristics, sources and negative effects on human health.
100 μg for benzene, toluene and xylene and up to 800 μg for ethylbenzene, the potential dietary intake for BTEXS varies enormously. For example, coffee has been identified to contain between 150 μg kg⫺1 (benzene) and 350 μg kg⫺1 (toluene). The average daily intake can be set at 5 μg g⫺1 person⫺1 day⫺1 for each BTEXS. Their lipophilic nature makes them especially harmful by accumulation in fatty foodstuffs and concretely edible oils and fats. As far as olive oil is concerned, several studies have been carried out in the European Union in order to determine the concentration of BTEXS in virgin olive oil samples from different countries. The results are listed in Table 51.2. However, taking into account that the analyses were carried out under different experimental conditions, it is difficult to extract definitive conclusions, although they can provide a general picture on the situation. Apart from this, different research articles have been published on the topic. The first reference on the correlation
TABLE 51.2 Intervals of concentration (μg kg⫺1) for BTEXS in olive oil samples in the European Union. Analyte
Germany
France
Greece
Benzene
0–10
0.1–294
0–158
Toluene
10–180
14–174
85–139
0–10
3–95
18–415
Ethylbenzene m ⫹ p-Xylene
10–280
15–548
o-Xylene Styrene
18–415
5–102 10–640
–
6–60
This table summarizes the concentration ranges in which BTEXS have been found in olive oil samples coming from the European Union according to the literature survey.
of the presence of benzene and the sample packing appeared in 1990 by Jickells et al. The authors established in collaboration with the suppliers of the material used for the cookware that the benzene originated from the use of t-butyl perbenzenoate as initiator in the polymer’s manufacturing. Its concentration ranged between 1.9 and 5.6 mg kg⫺1 in olive oil after extraction for 1 h at 175°C. Lower values were obtained (ⱕ0.1 mg kg⫺1) for samples containing non-aromatic initiator. Bieddermann et al. (1995) studied the presence of BTEXS in olives and the oils at different production steps. They compared the concentration of BTEXS in the fruits with those in the atmosphere from the olive grove and olive mills. They found that there was a systematic increase on the values as the likely result of the adsorption of BTEXS from contaminated air. Finally, they detected a significant increase on styrene concentration during storage of crushed olives at ambient temperature which was ascribed to the metabolic origin of this compound. In a further study, the same research group investigates the potential source of contamination of virgin olive oil by gasoline BTEXS (Bieddermann et al., 1996). The higher concentration of BTEXS in the olives on the tree and in the room of intermediate storage (by a factor of ca. 100), was attributed to contamination by gasoline vapors, since vehicles and other gasoline-driven engines could often be found there. The authors include some recommendation to avoid or minimize the contamination of olive oil with gasoline components. Finally, styrene was identified as marker of the olive oil production system (Ollivier and Guerere, 2001). The oils obtained following physical methods (pressing) were found to contain more styrene than those obtained by continuous processing, increasing this value with the storage time of the olives. It was attributed to the decarboxylation of a natural olive oil component (cinnamic acid) into styrene. Concerning the total benzene hydrocarbons concentration, the variability in the levels can be explained by the location, bad agricultural practices and the technological devices used.
CHAPTER | 51 Benzene, Toluene, Ethylbenzene (o-, m- and p-) Xylenes and Styrene in Olive Oil
51.2 DETERMINATION OF BTEXS IN OLIVE OIL SAMPLES BY THE DIRECT COUPLING HEADSPACE-MASS SPECTROMETRY The determination of volatile compounds in environmental samples and foodstuffs is usually carried out by headspacemass spectrometry (HS-MS). The headspace module permits the analysis of the volatile fraction of a given sample with minimal manipulation as, once in the vial, the sample is automatically heated at the selected temperature for a given time and injected into the gas chromatograph by the robotic arm. Depending on the temperature of the chromatographic oven, two approaches have been described. The simplest one uses the chromatographic column as a transfer line avoiding the interaction of the volatile compounds with the stationary phase by maintaining the oven at a high temperature. In this case, the mass spectrometer records the global signal corresponding to the volatile fraction of the olive oil sample. By using different chemometric techniques it is possible to extract the maximum analytical information from this signal. The so-called MS-based sensors have been successfully applied in the field of olive oil analysis. The direct screening of olive oil samples for BTEXS has also been the subject matter of this configuration (Peña et al., 2004a). The use of chromatographic separation increases the analysis time in ca. 15 min although it permits the detection of the individual compounds in the low nanogram-per-milliliter level (Peña et al., 2004b). The most relevant details of both methodologies are discussed below.
51.2.1 Global Determination of BTEXS by HS-MS The main objective of the developed method was to classify different olive oil samples as recommendable or nonrecommendable for human consumption on the basis of the normal levels previously commented on. For this purpose, aliquots of 10 mL of olive oil were analyzed by HS-MS after heating at 95°C for 25 min to enrich the headspace inside the vial with the volatile fraction of the olive oil sample. The range of the mass spectrometer was set between 75 and 110 m/z values, where all the molecular and base peaks of the BTEXS appear. The volatile fraction reaches the mass spectrometer simultaneously and the global signal obtained was used for classification purposes (Peña et al., 2004a). Two chemometric treatments were applied to the data. First of all, two unsupervised classification techniques were used, namely cluster analysis (CA) and principal components analysis (PCA). The main aim of both is to find clustering, natural grouping or internal structures of the samples. Then, supervised techniques such as k-nearest neighbors (KNN) or soft independent modeling of class analogy (SIMCA) were used to generate classification
465
models capable of predicting unknown samples. All the chemometric approaches were based on the analysis of 120 olive oil samples containing BTEXS concentration within the interval 0.03–1 μg mL⫺1. In a final step, the reliability of the proposed method was evaluated through the qualitative errors associated: false positive and false negative percentages. For this purpose, 80 olive oil samples spiked with a total BTEXS concentration above and below the threshold value (0.2 μg mL⫺1) were analyzed. No false negatives were obtained while 5% false positives were yielded by KNN. SIMCA provided the best results as the 80 samples were correctly classified. The application of the method to the analysis of commercial olive oil samples pointed out that four samples out of the 50 analyzed provided a BTEXS concentration over 0.2 μg mL⫺1.
51.2.2 Discriminated Determination of BTEXS by HS-GC-MS The main advantage of the implementation of gas chromatographic separation of the analytes under study is the unequivocal identification of the sample composition and the individual quantification of the different components of the family compounds. The analysis time and the investment in equipment seem to be a minor shortcoming considering the wide presence of gas chromatograph in routine laboratories. Two references can be found in the literature concerning the determination of BTEXS based on the sole use of HS-GC-MS hybridization. The first approach is a technical note, which provides the first reference to the determination of BTEXS in olive oil samples with high sensitivity and precision, claiming for its use as the official method of analysis (Ollivier and Guerere, 2000). Three years later, Peña et al. reported a research article with a systematic evaluation of the variables involved and thus providing a rigorous characterization of the analytical features of this technique for the proposed determination (Peña et al., 2004b). By maintaining the headspace conditions previously optimized, the separation of the BTEXS was accomplished on a 45 m ⫻ 0.32 mm i.d. ⫻ 0.25 μm film thickness capillary column coated with a 5% phenyl–95% methyl polysiloxane. The length of the column allows the chromatographic separation of styrene and o-xylene. A representative chromatogram for the analysis of BTEXS in a virgin olive oil sample is given in Figure 51.2. The temperature gradient was fixed between 40 and 200°C, with a total analysis time of ca. 16 min. The peak area of the base peak obtained in the mass spectrum of each compound was used for quantitation purposes. Table 51.3 lists the main analytical figures of the method in terms of detection limit, precision (as relative standard deviation, RSD) and recoveries from olive oil samples. In order to evaluate the potential of the optimized configuration, up
466
SECTION | I Toxicology and Contaminants
80000
1
2
Abundance
60000
3
40000
4 5 6
20000
5.00
6.00
7.00
8.00
9.00
Time (min) FIGURE 51.2 Typical chromatogram obtained for the headspace-gas chromatographic-mass spectrometric analysis of a virgin olive oil sample spiked with 50 ng mL⫺1 of benzene (1), toluene (2), ethylbenzene (3), m ⫹ p- xylene (4), styrene (5) and o-xylene (6). This figure presents the chromatographic separation obtained for the BTEXS in a real virgin olive oil sample. Adapted from Peña et al. (2004b).
TABLE 51.3 Analytical features of the HS/GC/MS determination of BTEXS in olive oil samples. Recovery (%)
Analyte
Detection limit (ng mL⫺1)
Precision RSD (%)
Benzene
2.8
4.5
96
Toluene
3.5
4.9
93
Ethylbenzene
6.0
6.5
95
m ⫹ p-Xylene
6.9
7.0
93
o-Xylene
7.4
7.5
99
Styrene
8.8
8.1
100
This table presents the minimum detectable concentration, reproducibility of the measurement and recovery for BTEXS obtained by the headspace-gas chromatography and mass spectrometry (adapted from Peña et al., 2004b). RSD, relative standard deviation.
to 25 commercial olive oil samples, previously analyzed by the direct coupling HS-MS described in the previous section, were processed. Two of the samples identified as positive were found to contain toluene at a concentration of ca. 300 ng mL⫺1, while in the other two, styrene was present at concentrations higher than 600 ng mL⫺1, being negligible to the contribution of m- and p-xylene while benzene, ethylbenzene and o-xylene were not detected. The higher concentration of styrene can be attributed to a migration of styrene monomers from the plastic bottle to the oil. This compound was not detected in glass-bottled olive oil samples.
51.3 EVALUATION OF DIFFERENT EXTRACTION TECHNIQUES FOR THE DETERMINATION OF BTEXS IN OLIVE OIL SAMPLES As it has been described in depth, the determination of BTEXS in olive oils is usually carried out by analyzing the sample’s headspace. However, new extraction strategies have been adopted prior to the gas chromatographic analysis of the volatile fraction of the oil. The main aims of the proposed approaches are to improve the sensitivity and selectivity of the determinations. Two different methodologies have been recently published. The first one, developed by Vichi et al. (2005), relies on the direct coupling headspace-solid phase microextraction. More recently, Carrillo-Carrión et al. (2007a) have evaluated the potential of surfactant-coated multiwalled carbon nanotubes as additive in the liquid–liquid extraction of BTEXS from olive oil samples. The most relevant characteristics of both methods are summarized below.
51.3.1 Determination of BTEXS in Olive Oil Samples by SPME-HS-GC-MS The so-called solventless extraction techniques such as solid-phase microextraction (SPME) and single-drop microextraction (SDME) have been gaining acceptance in recent years as alternatives to conventional extraction techniques as they provide equal or better analytical features with minimal sample manipulation and organic solvent consumption. These characteristics make them environmentally friendly, reducing the volume of residues to be managed by the laboratory. Both SPME and SDME have been used for the determination of BTEXS in environmental matrices (Przyjazny, 2002; Arambarri et al., 2004; Ezquerro et al., 2004).
CHAPTER | 51 Benzene, Toluene, Ethylbenzene (o-, m- and p-) Xylenes and Styrene in Olive Oil
However, only SPME has been applied to the analysis of the pollutants in vegetable oils (Page and Lacroix, 2000; Vichi et al., 2005). Vichi et al. afforded the simultaneous determination of volatile and semivolatile aromatic hydrocarbons in virgin olive oil by HS-SPME-GC-MS. The authors justified the very few applications found on the determination of volatiles in lipid matrices by the decrease in SPME efficiency due to the matrix effect. Therefore, in order to obtain better values of efficiency and selectivity, the variables affecting the SPME sampling process, namely type of fiber coating, temperature and extraction time, must be rigorously studied. The authors concluded that the divynilbenzene/carboxen/polydimethylsiloxane (DVB/Car/PDMS) coating provided the better efficiency on the grounds of its micropore dimensions, which allows extraction of small molecules. As is well-known, the adsorption of the analytes is an exothermic process and thus, it is favored at low temperatures; however, the higher the temperature, the higher the enrichment of the gaseous phase on the target compounds. The value of the variables finally selected should be a compromise between the two factors as a result. The determination of BTEXS can be carried out at temperatures between 40–60°C thanks to their high volatility. Less volatile compounds, such as PAHs, require the temperature to be set at 100°C to achieve acceptable sensitivity. The SPME extraction dramatically increases the analysis time, as for BTEXS it requires ca. 45 min to be completed. The authors jointly determined benzene hydrocarbons and PAHs and, therefore, the analysis time was fixed at 60 min, without reaching the equilibrium owing to the lower volatility of the latter. Table 51.4 summarizes the analytical features of
TABLE 51.4 Analytical parameters obtained for toluene, ethylbenzene and xylene isomers by HS/SPME/GC/MS (adapted from Vichi et al., 2005). Analyte
Detection limit (μg kg⫺1)
Quantitation limit (μg kg⫺1)
Precision RSD (%)
Toluene
0.4
1.4
12.3
Ethylbenzene
0.6
1.9
6.4
m-Xylene
0.7
2.2
4.2
p-Xylene
0.4
1.5
4.6
o-Xylene
0.6
1.9
7.5
This table shows the improvement on the detection and quantitation limits achieved through the implementation of a preconcentration technique such as solid-phase microextraction (adapted from Vichi et al., 2005). HS/SPME/GC/MS: headspace-solid phase microextraction-gas chromatography-mass spectrometry. RSD, relative standard deviation.
467
the method for toluene, ethylbenzene and xylene isomers under the optimized experimental conditions and using 2 g of sample. The vial was magnetically stirred for 2 min before exposure of fiber to the sample headspace. The article includes the data of the distribution constants for each analyte between (i) the sample matrix and fiber coating; (ii) sample-headspace; and (iii) headspacefiber as a mean to evaluate the individual efficiency of the SPME extraction. Finally, the proposed methodology was applied to the determination of the selected pollutants on virgin olive oils. They found that toluene was the most abundant component with a concentration of up to 32 μg kg⫺1; o- and m-xylenes and ethylbenzene were quantified up to 200 μg kg⫺1 while the highest concentration of p-xylene was 91 μg kg⫺1.
51.3.2 Liquid–Liquid Extraction of BTEXS from Olive Oils by Means of Dispersed Carbon Nanotubes Carbon nanotubes (CNTs) have been revealed in the last decade as a promising material for nanotechnological developments. Readers interested in the topic are referred to a recent prospective article (Valcarcel et al., 2007). Among their unique and extraordinary properties, CNTs present a strong adsorption affinity to aromatic compounds, being used in solid-phase extraction, as coating layer in SPME, stationary phase in gas chromatography and pseudophase in electrokinetic chromatography (Valcárcel et al., 2008). One of the main limitations reported as regards their use in SPE is their aggregation tendency, which clearly reduced their efficiency as the likely result of the lower surface area. This shortcoming can be overcome either by chemical derivatization or dispersion in a surfactant medium. The surfactact-coated carbon nanotubes (SC-CNTs) can be used as additives in liquid–liquid extraction. Carrillo-Carrión et al. (2007b) have published a rigorous study concerning the capability of single-walled and multiwalled carbon nanotubes, dispersed in sodium dodecyl sulfate (SDS) as extraction medium for benzene and toluene. The article points out that the presence of the carbon nanomaterial clearly improves the efficiency of the process with negligible synergic contribution of the surfactant. An additional advantage of the method is the recovery of the carbon nanotubes after use by gentle washing with methanol and water, which helps to reduce the cost of the analyses. The use of SC-CNTs as extractant for the determination of BTEXS in olive samples was evaluated by the same authors (Carrillo-Carrión et al., 2007a). The pseudophase was prepared by weighing 5.0 mg of MWNTs and 130 mg of SDS in a 50 mL volumetric flask; once filled with distilled water, it was sonicated for 20 min to obtain a homogeneous and stable dispersion which contains the nanotubes at a final concentration of 0.1 mg mL⫺1.
468
SECTION | I Toxicology and Contaminants
TABLE 51.5 Concentration interval (ng mL⫺1) for the BTEXS identified in olive oil commercial samples. Analyte
Extra virgin olive oil
Virgin olive oil
Olive oil
Refined olive oil
Benzene
75–174
16–23
7–36
45
Toluene
51–115
16–236
9–31
4–80
Ethylbenzene
–
17
8–34
5
m ⫹ p-Xylene
–
16
8–25
5
o-Xylene
–
15
8–20
5
8–32
4
Styrene
2–103
9–102
The results presented in this table were obtained by including a previous liquid–liquid extraction of BTEXS from the olive oil samples with surfactant coated carbon nanotubes (adapted from Carrillo-Carrión et al., 2007a).
The optimized procedure comprises the following steps. For the liquid–liquid extraction, 9 mL of olive oil plus 9 mL of SC-MWCTs were placed in a 20 mL glass vial. After the addition of 100 μL of methanol to enhance the transference between phases, the vial was sealed and manually shaken for 30 s. It should be noted that the presence of the organic solvent did not destabilize the nanotube dispersion. Once phase separation was completed, 6 mL of the aqueous extract was transferred to a 10 mL vial, hermetically sealed and placed in the headspace autosampler. The robotic arm transferred the vial to an oven heated to 80°C and maintained there for 15 min under magnetic stirring. Chromatographic separation was achieved after the injection of 2.5 mL of the headspace of the vial into the GC. The optimum value for each variable involved was established following a step-by-step design, considering the low number involved. To mention those concerning the headspace generation, the variables were studied by extracting a refined olive oil sample spiked with 50 μg mL⫺1 of toluene, selected as model analyte, with the pseudophase. It was concluded that the use of 6 mL of the SC-MWNTs phase in 10 mL glass vials provide the highest MS response as a result of the favorable ratio of sample: headspace. Moreover, the presence of an organic modifier was not necessary to obtain the quantitative release of toluene to the gaseous phase. Concerning the extraction temperature and time, the latter resulted to be more critical as the chromatographic area dramatically increased from 5 to 15 min (ca. three times higher), remaining constant over this value. By working under the chemical and instrumental optimum conditions, the analytical features of the method were established. The detection limits ranged between 0.25 ng mL⫺1 (ethylbenzene) and 0.43 ng mL⫺1 (benzene). Concerning the reproducibility (expressed as
relative standard deviation, n ⫽ 11), this analytical property was evaluated at two concentration levels and it ranged between 4.0% (benzene at 20 ng mL⫺1) and 1.5% (m ⫹ pxylene isomers at 50 ng mL⫺1). Under the chromatographic conditions employed in this work (mainly the dimensions of the chromatographic column) o-xylene and styrene overlapped. Notwithstanding this, they were identified and individually quantified by extracting from the whole mass spectrum, the m/z values that did not coincide in both spectra, namely ion fragment 91 for the xylene and ion fragment 104 for styrene. The optimized method was applied to the analysis of ten commercial samples for the determination of their BTEXS content. The samples were selected trying to maintain the geographical origin, olive oil variety and material package (glass, metal and plastic) as heterogeneous as possible. Table 51.5 summarizes the concentration range found for the BTEXS in the olive oil samples. As can be seen, benzene and toluene were present in all the samples analyzed while styrene was found in those samples stored in plastic bottles. It was also detected that its concentration was higher in those samples with a longer packing time. As a conclusion, a comparison between this alternative and the direct headspace analysis was carried out. Figure 51.3 shows the chromatographs obtained under both conditions. As can be seen, the implementation of a LLE step enhances the sensitivity of the determination by factors ranging between 20 (benzene) and 45 (ethylbenzene and xylene isomers). It can be justified by the fact that the aqueous nature of the matrix used to generate the headspace favored the release of the analytes. Moreover, the temperature used in this step (80°C) destabilized the MWNTS dispersion, which also helps to the analyte transference.
469
CHAPTER | 51 Benzene, Toluene, Ethylbenzene (o-, m- and p-) Xylenes and Styrene in Olive Oil
120 000 4 100 000
Abundance
80 000 5 +6 3
60 000
2 1
40 000
20 000 SD 0 0
2
4
6
8
10
12
14
16
18
Time (min) FIGURE 51.3 Total ion chromatograms obtained by direct headspace analysis (black line) and by the liquid–liquid extraction with SC-MWNTs (gray line) of a blank refined olive oil sample spiked with 20 ng mL⫺1 of benzene (1), toluene (2), ethylbenzene (3), m ⫹ p-xylene (4), styrene ⫹ o-xylene (5 ⫹ 6). This figure shows the sensitivity enhancement obtained as a result of the implementation of the liquid–liquid extraction of BTEXS from refined olive oil samples prior to their gas chromatographic analysis. SC-MWNTs, surfactant coated multiwalled carbon nanotubes; SD, solvent delay.
SUMMARY POINTS ●
●
●
●
●
●
●
●
BTEXS defines the mixture of benzene, toluene, ortho, meta and para xylene isomers and styrene. They are present in the environment from a wide variety of sources, including metabolic origin. Their lipophilic nature facilitates their accumulation in fatty matrices, including edible oils. Analytical methodologies developed for their determination includes total index and individual quantification of BTEXS. The global determination of BTEXS is carried out by means of the direct coupling headspace-mass spectrometry. Discriminated information about the BTEXS present in an olive oil sample is accomplished by headspace-gas chromatography-mass spectrometry. The sensitivity and selectivity enhancement is carried out by the implementation of extraction techniques such as solid-phase microextraction. Carbon nanotubes have also been proposed as extractant for BTEXS in olive oil samples with excellent results compared to the direct analysis of the headspace of the samples.
REFERENCES Arambarri, I., Lasa, M., Garcia, R., Millan, E., 2004. Determination of fuel dialkyl ethers and BTEX in water using headspace solid-phase
microextraction and gas chromatography–flame ionization detection. J. Chromatgr. A 1033, 193–2003. Bieddermann, M., Grog, K., Morchio, G., 1995. Origin of benzene, toluene, ethylbenzene and xylene in extra virgin olive oil. Zeitschrift fuer Lebensmittel-Untersuchung und Forschung 200, 266–272. Bieddermann, M., Grog, K., Morchio, G., 1996. On the origin of benzene, toluene, ethylbenzene and xylene in extra virgin olive oil. Further results. Zeitschrift fuer Lebensmittel-Untersuchung und Forschung 200, 266–272. Carrillo-Carrión, C., Lucena, R., Cárdenas, S., Valcárcel, M., 2007a. Liquid-liquid extraction/headspace/gas chromatographic/mass spectrometric determination of benzene, toluene, ethylbenzene, (o-, mand p-)xylene and styrene in olive oil using surfactant-coated carbon nanotubes as extractant. J. Chromatogr. A 1171, 1–7. Carrillo-Carrión, C., Lucena, R., Cárdenas, S., Valcárcel, M., 2007b. Surfactant coated carbon nanotubes as pseudophases in liquid-liquid extraction. Analyst 132, 551–559. Ezquerro, O., Ortiz, G., Pons, B., Tena, M.T., 2004. Determination of benzene, toluene, ethylbenzene and xylenes in soils by multiple headspace solid-phase microextraction. J. Chromatogr. A 1035, 17–22. Jickells, S.M., Crews, C., Castle, L., Gilbert, J., 1990. Headspace analysis of benzene in food contact materials and its migration into food from plastics cookware. Food Addit. Contam. 7, 197–205. Ollivier, D., Guerere, M., 2000. Detection and measurement of benzene hydrocarbons in virgin olive oils. Annals des Falsifications de l’Expertise Chimique et Toxicologique 93, 67–81. Ollivier, D., Guerere, M., 2001. Development of a benzene hydrocarbon measurement method for virgin olive oil. Am. Lab. 33, 18. Page, B.D., Lacroix, G., 2000. Analysis of volatile contaminants in vegetable oils by headspace solid-phase microextraction with Carboxenbased fibres. J. Chromatogr. A 873, 79–154.
470
Peña, F., Cárdenas, S., Gallego, M., Valcárcel, M., 2004a. Direct screening of olive oil simples for residual benzene hydrocarbon compounds by headspace-mass spectrometry. Anal. Chim. Acta 526, 77–82. Peña, F., Cárdenas, S., Gallego, M., Valcárcel, M., 2004b. Combining headspace gas chromatography with mass spectrometry for confirmation of hydrocarbon residues in virgin olive oil following automatic screening. J. Chromatogr. A 1052, 137–143. Przyjazny, A., Kokosa, J.M., 2002. Analytical characteristics of the determination of benzene, toluene, ethylbenzene and xylenes in water by headspace solvent microextraction. J. Chromatogr. A 977, 143–153.
SECTION | I Toxicology and Contaminants
Vichi, S., Pizzale, L., Conte, S.L., Buxaderas, S., López-Tamames, E., 2005. Simultaneous determination of volatile and semivolatile aromatic hydrocarbons in virgin olive oil by headspace solid-phase microextraction coupled to gas chromatography-mass spectrometry. J. Chromatogr. A 1090, 146–154. Valcarcel, M., Cárdenas, S., Simonet, B.M., 2007. Role of carbon nanotubes in Analytical Science. Anal. Chem. 79, 4788–4797. Valcárcel, M., Cárdenas, S., Simonet, B.M., Moliner-Martínez, Y., Lucena, R., 2008. Carbon nanostructures as sorbent materials in analytical processes. Trends Anal. Chem. 27, 34–43.
Chapter 52
The Toxicity Profile of Hydrolyzed Aqueous Olive Pulp Extract Robert M. Diener and Mildred S. Christian Argus International, Horsham, Pennsylvania, USA
52.1 INTRODUCTION HIDROX™ (Hydrolyzed Aqueous Olive Pulp Extract; OPE) is a water extract product of organically grown olives containing high concentrations of polyphenol antioxidants, especially hydroxytyrosol (HT) (Table 52.1). The polyphenols and hydroxytyrosol extracted from olive oil have in vitro and in vivo activity in preventing and/or reducing the deleterious effects of oxygen-derived free radicals associated with numerous inflammatory and stress-related human and animal diseases (Galli and Visioli, 1999; Manna et al., 1999; Visioli et al., 2002). Oral or topical dosages of 0.05,
TABLE 52.1 Overview of major features of olive pulp extract (OPE). This table summarizes important biological, pharmacological and toxicological characteristics of OPE. OPE is produced from the byproducts of olive oil processing OPE is a standardized freeze-dried powder prepared from olive pulp water OPE is rich in water-soluble, phenolic compounds (see Table 52.2) Polyphenols such as hydroxytyrosol have anti-inflammatory activity that is useful in preventing or reducing stress-related diseases (Galli and Visioli, 1999) Oral toxicity studies in rats have shown OPE to be non-toxic up to 5 g kg⫺1 In vivo mutagenicity studies have shown OPE to have no mutagenic potential OPE is not a reproductive or developmental hazard in oral doses up to 2 g kg⫺1 day⫺1
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
0.5 and 5 mg kg⫺1 day⫺1 OPE for 30 days also resulted in dose-dependent protection against UVR-induced responses in hairless mice (Arocena et al., 2003). All toxicology studies were conducted in strict compliance with Good Laboratory Practices (GLPs), as defined by the FDA (USFDA, 1987, GLP, Final Rule, 1987). Toxicological procedures in the acute studies reflect those described in the ‘Redbook 2000’ (USFDA, 2000). All animal husbandry practices and procedures were in compliance with the ‘Guide for the Care and Use of Laboratory Animals’ (Institute of Laboratory Animal Resources, 1996). Results from these studies and a recent (2007) scientific literature search indicated that OPE is a GRAS substance (Generally Recognized As Safe).
52.2 MATERIALS USED IN THE TOXICITY STUDIES For the oral toxicity studies, OPE was solubilized in deionized water to a concentration of 200 mg mL⫺1 for mice; concentrations of 0, 100, 150 or 200 mg mL⫺1 in aqueous 0.5% w/v methylcellulose were used for the rats (Table 52.2). Dosing formulations were prepared at room temperature and stirred continuously during administration to the animals. Stability test results for OPE demonstrated complete stability with 100% retention of the initial value of hydroxytyrosol (the major o-diphenol present in OPE) after 20 months of storage at 25 °C ⫾ 2 °C at 60% RH. The gavage route of administration was chosen for safety studies based on the fact that it simulated the human method of intake, and because high doses of OPE in feed are not palatable to rats.
52.3 ACUTE STUDIES In acute toxicity studies in CRL:CD1 ICR (BR) 7-weekold mice (five/sex/group), gavage administration or dermal 471
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
472
SECTION | I Toxicology and Contaminants
TABLE 52.2 The chemical composition of aqueous OPE. This table enumerates the various chemical entities detected in OPE. Chemical analysis
Value: g/100 g with range
Moisture
2.8 (1–3)
Ash
9.6 (6–10)
Protein
6.7 (4–7)
Fat
27.0 (20–30)
Carbohydrate (total)
53.5 (40–55)
Phenolics (total)
6.0
Total polyphenol
100%
Hydroxytyrosol
50–70%
Tyrosol
0.3%
Oleuropein
5–10%
Other polyphenols
20%
Gallic acid
Variable
Oleuropein aglycone
Variable
The analysis of the phenol fraction is given as a percent of total polyphenols.
application of a single dose of OPE at doses of 500, 1000 or 2000 mg kg⫺1 failed to produce any adverse effects. There was no mortality in any of the treatment groups, suggesting that the oral or dermal LD50 of the extract is greater than 2000 mg kg⫺1 (Christian et al., 2004). Single oral doses of 0, 1000, 1500 or 2000 mg kg⫺1 OPE were also administered by gavage to 70-day-old Crl: CD Sprague-Dawley rats (five/sex/dose). The animals were observed for 14 days and then euthanized. There were no deaths or adverse effects except some soft or liquid feces, indicating that the oral LD50 of OPE in rats is greater than 2000 mg kg⫺1 (Christian et al., 2004). Single doses of 2000 mg hydroxytyrosol kg⫺1 to six male Sprague-Dawley rats likewise produced no deaths or pathological changes in ‘main organs’ (D’Angelo et al., 2001). As part of a micronucleus mutagenicity study, Crl:CD Sprague-Dawley rats (five/sex) were gavaged with a single dose of 5000 mg kg⫺1 OPE. No mortality or signs of toxicity were noted during the following 6 days, and the rats were then given the same dose for 29 consecutive days. Again, no mortality or clinical signs were noted, indicating that OPE is essentially non-toxic (Christian et al., 2004). The acute toxicity of OPE can be summarized as follows: ●
The oral or dermal LD50 in mice is greater than 2000 mg kg⫺1
● ●
The oral LD50 in rats is greater than 5000 mg kg⫺1 Single oral doses of OPE as high as 5 g kg⫺1 are essentially non-toxic.
52.4 SUB-CHRONIC (90-DAY) TOXICITY STUDIES IN RATS The safety of OPE was investigated in a 90-day sub-chronic study (Christian et al., 2004). Groups of Crl:CD SpragueDawley rats (20/sex/group) were gavaged with 0, 1000, 1500 or 2000 mg kg⫺1 day⫺1 OPE (0, 60, 90 or 120 mg kg⫺1 day⫺1 of phenolics) for 90 consecutive days. The OPE was dissolved in aqueous 0.5% methylcellulose and administered at a volume of 10 mL kg⫺1 to male and female rats weighing 192 ⫾ 16 g and 155 ⫾ 16 g, respectively. Investigative parameters included: daily clinical signs; weekly body weights and feed consumption; hematology and serum chemistry determinations at termination; gross necropsy, organ weights and histopathology of selected tissues at termination. Satellite groups of six rats/sex were assigned to each dose level to determine toxicokinetics (TK). Toxic effects directly attributable to OPE did not occur in this 90-day toxicity study. Mortality or morbidity was not observed. There were no test-article-related clinical signs, except for excessive salivation and the presence of some test substance around the mouths of animals in the treated groups, which was attributed to the difficulty of administering the large quantity of relatively thick granular suspensions of OPE. Administration of OPE had no significant effect on body weights, body weight gains (Table 52.3) or feed consumption. A significant decrease in body weight gain that occurred in male rats at 1000 mg kg⫺1 day⫺1 on days 71–78 was considered unrelated to the test article because it was not dose-related. OPE did not produce any changes in organ weights or ophthalmologic parameters. Hematology parameters in the male and female rats did not reveal any untoward changes at doses as high as 2000 mg kg⫺1 day⫺1 (Table 52.4). However, a dose-related increasing trend in the number of red blood cells was noted in female rats. The increase was significant in the 2000 mg kg⫺1 day⫺1 dose group, although the values were within the range of historical control values. Increases in RBCs in the absence of changes in mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH) and mean corpuscular hemoglobin concentration (MCHC), were interpreted as a slight erythropoietic stimulation of the bone marrow without any toxicological consequences. In male rats, numbers of WBCs were increased significantly at doses of 1500 mg kg⫺1 day⫺1, compared to controls, but in the absence of a clear dose-relationship, this effect was also considered to be without any toxicological consequence. Serum chemistry determinations revealed that liver ‘leakage’ enzyme levels [alanine aminotransferase (ALT), aspartate aminotransferase (AST) and sorbitol dehydrogenase
473
CHAPTER | 52 The Toxicity Profile of Hydrolyzed Aqueous Olive Pulp Extract
TABLE 52.3 Initial body weights and body weight gains of male and female rats on the 90-day toxicity study. This table confirms that there were no significant differences in mean body weights among groups of male or female rats at initiation of the 90-day study, and gavaged administration of OPE had no significant effect on body weight gains after the conclusion of 90 days of treatment. Gavaged mg kg⫺1 day⫺1 doses of OPE for 90 consecutive days
20 rats/dosage level 0 (Vehicle)
1000
1500
2000
M Initial body weights (g)
193.6 ⫾ 15.0
192.6 ⫾ 16.0
192.4 ⫾ 14.4
189.7 ⫾ 19.3
F Initial body weights (g)
156.6 ⫾ 11.1
155.0 ⫾ 10.2
154.8 ⫾ 11.1
154.8 ⫾ 10.3
M Body weight gains (g, DS 90)
⫹345.6 ⫾ 62.0
⫹345.6 ⫾ 52.4
⫹345.4 ⫾ 68.4
⫹323.0 ⫾ 46.4
F Body weight gains (g, DS 90)
⫹145.2 ⫾ 35.4
⫹130.7 ⫾ 35.4
⫹133.0 ⫾ 31.7
⫹133.1 ⫾ 34.4
M: male; F: female; g: grams; DS: day of study.
TABLE 52.4 Leukocyte and erythrocyte counts of rats on 90-day toxicity study. Gavaged mg kg⫺1 day⫺1 doses of OPE for 90 consecutive days
20 rats/dosage level 0 (Vehicle)
1000
1500
2000
M WBC (thsn/mm3, DS 90)
14.3 ⫾ 2.98
14.9 ⫾ 2.18
17.4 ⫾ 3.91*
15.9 ⫾ 4.45
F WBC (thsn/mm3, DS 90)
12.2 ⫾ 2.69
13.5 ⫾ 4.94
14.0 ⫾ 3.52
13.7 ⫾ 2.51
M RBC (mill/mm3, DS 90)
7.78 ⫾ 0.42
7.78 ⫾ 0.39
7.85 ⫾ 0.38
7.76 ⫾ 0.45
F RBC (mill/mm3, DS 90)
6.85 ⫾ 0.93
6.98 ⫾ 0.41
7.04 ⫾ 0.29
7.22 ⫾ 0.42*
M: male; F: female; WBC: white blood cells; RBC: red blood cells; DS: day of study. * Significantly different from vehicle control group value (P⭐0.05). No untoward hematological effects were attributed to OPE in this table. The increased mean WBC count at 1500 mg kg⫺1 in male rats was not doserelated. The dose-related elevations in mean RBC counts in female rats were presumably due to a slight stimulation of the bone marrow, which was not considered an adverse effect.
(SDH)], as well as cholesterol values, were generally reduced (Table 52.5). The reductions were considered to be a beneficial, and not a toxic manifestation, because these cytosolic enzymes are sensitive indicators of hepatocellular injury when elevated (Loeb, 1999). The reason for the decreased activity could not be determined, although it may be associated with the large quantities of OPE that are excreted in the bile, which may also account for the decreased serum cholesterol levels, because primary bile acids are synthesized by the liver from cholesterol (Tennant, 1999). Other biomarkers of liver function (serum bilirubin, alkaline phosphatase, protein levels and histopathology) and serum chemistry determinations were unaffected by OPE. All rats used for TK blood sample collection survived to scheduled sacrifice. Mean hydroxytyrosol plasma concentration–time profiles on Day 90 were generally comparable for male and female rats and qualitatively similar for each dose group, although somewhat irregular. Plasma samples taken just before the last dose contained no measurable mean
concentrations of hydroxytyrosol, indicating minimal carryover from prior doses. Cmax was observed most frequently at 0.5 hours post-treatment, indicating a rapid rate of absorption, but Cmax did not consistently increase with dose. AUClast for males increased approximately in proportion to dose from 1000 to 1500 mg kg⫺1 day⫺1, but the increase from 1500 to 2000 mg kg⫺1 day⫺1 for both males and females resulted in a greater increase in AUClast than the proportional increase in dose. Terminal half-lives could not be estimated due to insufficient time points in the terminal phases. There were no gross or histological changes in any of the rats at the conclusion of the study that were considered to be due to OPE. Minimal or mild focal hyperplasia of the mucosal squamous epithelium of the limiting ridge of the forestomachs was observed in 11/20 male and 12/20 female rats in the 2000 mg kg⫺1 day⫺1 OPE dose group. Mild focal hyperplasia is often associated with irritation of this area of the gastric mucosa, and the causative factors in this case were presumed to be the large intubated volumes of viscous,
474
SECTION | I Toxicology and Contaminants
TABLE 52.5 Results of liver function tests from rats on 90-day toxicity study. This table reveals the reductions in mean serum liver ‘leakage’ enzyme levels, as well as cholesterol values, after 90 days of OPE treatment. Reduced serum levels are presumed to be beneficial effects of OPE, because elevated values reflect adverse hepatic conditions. Gavaged mg kg⫺1 day⫺1 doses of OPE for 90 consecutive days
20 rats/dosage level 0 (Vehicle)
1000
1500
2000
M cholesterol (mg dl⫺1, DS 90)
72 ⫾ 13.0
69 ⫾ 17.7
70 ⫾ 19.7
61 ⫾ 7.8
F cholesterol (mg dl⫺1, DS 90)
85 ⫾ 16.5
75 ⫾ 19.0
84 ⫾ 14.9
72 ⫾ 14.5*
M ALT (U/L, DS 90)
33 ⫾ 3.5
28 ⫾ 5.3*
26 ⫾ 5.8**
25 ⫾ 7.9**
F ALT (U/L, DS 90)
30 ⫾ 3.6
25 ⫾ 4.0**
25 ⫾ 3.7**
23 ⫾ 5.9**
M AST (U/L, DS 90)
74 ⫾ 6.8
71 ⫾ 6.7
69 ⫾ 9.8
67 ⫾ 9.5
F AST (U/L, DS 90)
76 ⫾ 9.0
72 ⫾ 10.5
69 ⫾ 7.4
69 ⫾ 12.8
M SDH (U/L, DS 90)
15.9 ⫾ 3.9
14.5 ⫾ 3.5
15.5 ⫾ 6.8
12.5 ⫾ 4.3
F SDH (U/L, DS 90)
15.5 ⫾ 6.5
13.5 ⫾ 8.8
9.5 ⫾ 2.4**
10.7 ⫾ 2.5*
M: male; F: female; DS: day of study. * Significantly different from vehicle control group value (P⭐0.05). ** Significantly different from vehicle control group value (P⭐0.01). ALT: alanine aminotransferase; AST: aspartate aminotransferase; SDH: sorbitol dehydrogenase.
granular formulation. Similar changes were noted in 3/20 rats from each sex treated with 1500 mg kg⫺1 day⫺1 and in 2/20 male and 3/20 female rats in the control group. All other microscopic changes that were noted in the examined organs and tissues were considered spontaneous in origin, incidental to treatment and not associated with any systemic toxicity due to the administration of OPE. The following conclusions summarize the 90-day toxicity study. ●
●
●
● ●
No mortality or morbidity at doses as high as 2000 mg kg⫺1 day⫺1 No significant changes of body weight, body weight gains or organ weights No adverse effects on hematology or serum chemistry parameters No gross or histological changes attributed directly to OPE No-Observed-Adverse-Effect Level (NOAEL): 2000 mg kg⫺1 day⫺1.
52.5 REPRODUCTIVE AND DEVELOPMENTAL TOXICITY STUDIES IN RATS In a study designed to provide an overall screening of potential reproductive and developmental toxicities, OPE was administered by oral gavage to Crl:CD SpragueDawley rats (Christian et al., 2004). The rats (eight/sex/
group) were assigned to five dose groups and administered 0 (aqueous 0.5% methylcellulose vehicle), 500, 1000, 1500 or 2000 mg kg⫺1 day⫺1 once daily for 14 days before cohabitation and continuation until the day before necropsy (males were euthanized after being administered a total of 49 daily doses; females were euthanized after completion of the 22-day postpartum period). The only clinical signs during OPE administration were occasional instances of excess salivation and non-dose-related increases in body weight gains. All F0 generation male rats survived to the scheduled euthanasia. One F0 female rat was found dead on the first day of lactation due to uterine torsion unrelated to OPE. Estrous cycling, mating and reproductive performance of the female rats were not affected by OPE. Small reductions (⬍10%) in pup body weights on lactation days 7, 14 and 21 in the 1000, 1500 and 2000 mg kg⫺1 day⫺1 dose groups were noted. All other delivery and litter observations were comparable among the five dose groups (Table 52.6). Evaluation of toxicokinetic parameters indicated that mean hydroxytyrosol plasma concentration–time profiles tended to be irregular and similar to those observed in the 90-day toxicity study. Mean concentrations of hydroxytyrosol were measurable through 1 to 4 hours (tlast) at the lowest dosage (500 mg kg⫺1 day⫺1) and through 8 hours at the higher dosages (1000, 1500 and 2000 mg kg⫺1 day⫺1). Terminal half-lives were variable and ranged from 5.2 to 22.8 hours post-treatment. Hydroxytyrosol was not detected in maternal milk or in plasma obtained from the
475
CHAPTER | 52 The Toxicity Profile of Hydrolyzed Aqueous Olive Pulp Extract
TABLE 52.6 Rat dose-range reproduction study: results in male and female rats and offspring. Treatment: daily gavage, 14 days prior to cohabitation until necropsy of males (48 days) and females (day 21 of lactation) Investigated parameters
0 (Vehicle)
500 mg kg⫺1 day⫺1
1000 mg kg⫺1 day⫺1
1500 mg kg⫺1 day⫺1
2000 mg kg⫺1 day⫺1
No. male rats tested
8
8
8
8
8
Terminal body weights (g)
489.5 ⫾ 36.6
499.8 ⫾ 24.5
497.0 ⫾ 30.3
508.4 ⫾ 38.7
491.6 ⫾ 34.2
Paired testes (g)
3.61 ⫾ 0.43
3.79 ⫾ 0.24
3.68 ⫾ 0.28
3.68 ⫾ 0.34
3.70 ⫾ 0.20
Paired epididymides (g)
1.52 ⫾ 0.20
1.56 ⫾ 0.12
1.50 ⫾ 0.10
1.49 ⫾ 0.09
1.50 ⫾ 0.15
No. female rats tested
8
8
8
8
8
No. pregnant
7
8
6
7
8
Body weight gains in gestation (g)
⫹136.6 ⫾ 21.9 ⫹143.3 ⫾ 21.9
⫹122.5 ⫾ 36.6
⫹134.3 ⫾ 17.7
⫹140.6 ⫾ 15.4
Gestation indexa
7/7
8/8
6/6
6/7
8/8
Mean no. pups delivered/litter
13.8 ⫾ 5.5
15.2 ⫾ 0.7
14.7 ⫾ 1.9
15.3 ⫾ 1.5
15.4 ⫾ 1.5
Viability indexb (%) (N/N)
98.9 (92/93)
99.2 (120/121)
100 (74/74)c
98.9 (90/91)
100 (120/120)
Pup weight/litter (g) at Day 0
6.5 ⫾ 0.4
6.8 ⫾ 0.4
6.4 ⫾ 0.3
6.6 ⫾ 0.7
6.4 ⫾ 0.7
Pup weight/litter (g) at Day 7
15.1 ⫾ 1.8
14.8 ⫾ 1.3
13.5 ⫾ 2.4
14.0 ⫾ 0.8
14.0 ⫾ 1.4
Pup weight/litter (g) at Day 14
25.7 ⫾ 3.4
24.5 ⫾ 2.3
23.3 ⫾ 3.0
24.0 ⫾ 1.1
23.5 ⫾ 2.6
Pup weight/litter (g) at Day 21
40.2 ⫾ 6.1
39.3 ⫾ 4.7
37.8 ⫾ 5.7
37.4 ⫾ 5.0
37.1 ⫾ 5.7
Small, not statistically significant, reductions (⬍10%) in pup body weights on lactation days 7, 14 and 21 in the 1000, 1500 and 2000 mg kg⫺1 day⫺1 dose groups were noted. All other delivery and litter observations were comparable among the five dose groups. a Number of rats with live offspring/number of pregnant rats. b Number of live pups on day 4 postpartum/number of live-born pups on day 0 postpartum. c One litter excluded because dam was found dead on postpartum day 1 due to torsion of the right horn of the uterus.
pups on Lactation Day 9. Based on the results of this study, OPE is not considered to be a reproductive or developmental toxicant. Developmental toxicity (embryo-fetal toxicity and teratogenic potential) of OPE was studied in four groups (25 rats/group) of time-mated Crl:CD Sprague-Dawley rats (Christian et al., 2004). Doses of 0 (0.5% methylcellulose), 1000, 1500 or 2000 mg kg⫺1 day⫺1 OPE were gavaged once daily to the pregnant rats on Gestation Days 6 through 20 (GD 6–20). On GD 21, one dam in the 2000 mg kg⫺1 day⫺1 group began premature delivery of its litter before scheduled cesarean sectioning and was euthanized. No abnormal findings were noted for this dam or its litter. All other rats survived until scheduled cesarean sectioning. No adverse clinical or necropsy observations or significant differences in maternal body weights, body weight gains, gravid uterine weights or absolute or relative feed consumption values were noted among the groups (Table 52.7).
Cesarean sectioning observations were based on 23, 22, 22 and 24 pregnant rats with one or more live fetuses in the four respective groups. OPE did not affect litter parameters at any of the doses. No treatment-related increases in gross external, soft tissue or skeletal fetal alterations (malformations or variations) were noted. A significantly increased mean number of corpora lutea at the 2000 mg kg⫺1 dose was well within the historical range of 14.5–20.1 per litter and was attributed to two females that had 27 and 30 corpora lutea. Mean hydroxytyrosol plasma concentration–time profiles from the pregnant rats on GD 20 in this study tended to be irregular, but were generally similar to those of the pregnant female rats and the non-pregnant and lactating female rats described in the previous study. Hydroxytyrosol was also detected in fetal plasma, indicating placental passage, but the values were below the limit of quantification. Based on the results of this study, the maternal and developmental NOAEL for OPE was 2000 mg kg⫺1 day⫺1, the highest dose administered.
476
SECTION | I Toxicology and Contaminants
TABLE 52.7 Rat developmental toxicity study results in female rats and offspring. This table shows that OPE did not affect litter parameters at any of the doses, or cause treatment-related increases in gross external, soft tissue or skeletal fetal alterations (malformations or variations). A significant increase in mean numbers of corpora lutea at 2000 mg kg⫺1 day⫺1 was within the historical range of 14.5–20.1 per litter. Treatment: daily gavage on days 6 through 20 of gestation; cesarean sectioned on gestation day 21 Investigated parameters
0 (Vehicle)
1000 mg kg⫺1 day⫺1
1500 mg kg⫺1 day⫺1
2000 mg kg⫺1 day
No. of female rats mated
25
25
25
25
No. of rats pregnant
23 (92%)
22 (88%)
22 (88%)
25 (100%)c
Mean body weight gain DG 6–21
⫹136.2 ⫾ 17.3
⫹138.7 ⫾ 19.4
⫹129.5 ⫾ 22.5
⫹130.0 ⫾ 23.9
Mean gravid uterine weight
101.8 ⫾ 10.9
107.7 ⫾ 13.4
100.9 ⫾ 14.2
104.1 ⫾ 12.5
Mean no. corpora lutea
15.3 ⫾ 2.2
17.2 ⫾ 2.3
17.3 ⫾ 3.2
17.9 ⫾ 3.7*
Mean no. implantations
14.3 ⫾ 1.8
15.5 ⫾ 2.2
15.2 ⫾ 2.0
15.3 ⫾ 1.6
Mean no. resorptions
0.3 ⫾ 0.5
0.1 ⫾ 0.4
0.7 ⫾ 1.0
0.4 ⫾ 0.9
Mean litter size (live fetusesa)
14.0 ⫾ 1.7
15.4 ⫾ 2.0
14.5 ⫾ 2.4
14.9 ⫾ 108
Mean fetal body weight (male)
5.36 ⫾ 0.31
5.33 ⫾ 0.23
5.37 ⫾ 0.42
5.20 ⫾ 0.29
Mean fetal body weight (female)
5.09 ⫾ 0.27
5.07 ⫾ 0.27
5.03 ⫾ 0.35
4.94 ⫾ 0.25
Fetuses with any alterationsb
7 (2.2%)
8 (2.4%)
10 (3.1%)
7 (2.0%)
6 (26.1%)
6 (27.3)
8 (36.4%)
6 (25.0%)
Litters with ‘altered’ fetuses a
⫺1
⫺1
One dead fetus at 2000 mg kg day . Alterations defined as gross, soft tissue or skeletal aberrations/abnormalities (malformations or variations). All data from this group are based on 24 dams; one dam began delivery of litter prior to cesarean section. * Significantly different from vehicle control group value (P⭐0.05). b
c
The following conclusions summarize the reproductive and developmental toxicity studies at oral doses up to 2000 mg kg⫺1 day⫺1 of OPE in rats: ●
●
●
●
●
No clinical signs of toxicity during mating, gestation and lactation No effects on reproductive organs, mating or pregnancy rate No significant effects on litter size, fetal weight or litter weights No developmental aberrations or teratogenic effect due to OPE No impairment of lactation or growth effects on weanlings.
52.6 GENOTOXICITY/MUTAGENICITY STUDIES OPE was subjected to three mutagenicity assays: an in vitro bacterial reverse mutation assay and a chromosome aberration assay and an in vivo rat micronucleus assay.
In the bacterial reverse mutation assay, Salmonella typhimurium strains TA97, TA98, TA100 and TA1535 and Escherichia coli strain WP2 uvrA (328) were used, in the presence and absence of S9 activation. OPE was tested at concentrations of 0, 5, 10, 50, 100, 500, 1000, 2500 and 5000 μg plate⫺1 (Christian et al., 2004). Evidence of mutagenic activity was detected in the plate incorporation test at concentrations of 1000 and 2500 μg plate⫺1 of OPE with S. typhimurium tester strains TA98 and TA100, but not in tester strains TA97a, TA1535 or Escherichia coli strain WP2 uvrA. The findings were confirmed in the more sensitive preincubation test, but only with metabolic activation. However, inconsistencies between the regular and repeat trials, the antibacterial properties of OPE, and observation of positive findings only at one or two concentrations where precipitates and toxicity were also present, confounded the interpretation of the mutagenic findings. The investigators concluded that under the conditions of the study, equivocal evidence of mutagenic activity of OPE was detected in S. typhimurium strains TA98 and TA100 (Christian et al., 2004).
CHAPTER | 52 The Toxicity Profile of Hydrolyzed Aqueous Olive Pulp Extract
The following conclusions summarize the bacterial reverse mutation assay: ●
●
●
Mutagenic potential detected only with tester strains TA98 and TA100 Mutagenic potential detected only with activation at the two highest doses Based on study conditions, mutagenic potential considered to be equivocal.
In the CHO chromosome aberration assay, the effect of 0, 10, 50, 100, 300, 600 or 1000 μg of OPE on chromosome aberrations in Chinese hamster ovary cells were investigated, in the presence and absence of an exogenous (S9) metabolic activation system (Christian et al., 2004). The solvent, diluent and negative control used in this assay was dimethyl sulfoxide (DMSO). The cell cultures were incubated with OPE for approximately 3 h, after which the treatment medium was washed and replaced with a fresh culture medium. Cells were sampled at approximately 20 h after initiation of treatment. Two hours prior to harvest, Colcemid® was added to arrest cells in metaphase. Concentrations of 100, 300 and 1000 μg mL⫺1 were then assessed for effects on mitotic index, polyploid cells and aberrations (chromatid and chromosome breaks/exchanges). No clear evidence of OPE-associated toxicity was observed at any concentration level, but a significant increase in the percentage of aberrant cells at 1000 μg mL⫺1 was observed in the presence of S9 activation. At this concentration, slight increases in the numbers of polyploid and/or endoreduplicated cells (numerical chromosome changes) were also noted, as well as some precipitation of OPE. Based on the results of this study, the investigators concluded that OPE was positive for the induction of chromosome aberrations (Christian et al., 2004). The following conclusions summarize the CHO chromosome aberration assay: ●
●
●
Increased % aberrant cells when tested with activation at the highest dose Slight increases in numerical chromosome changes at the highest dose CHO chromosome aberration assay demonstrated potential mutagenic activity.
The positive results in both in vitro assays were only observed at one or two of the highest concentrations, where slight amounts of precipitation also were present, and were confirmed only in the presence of S9 metabolic activation. These confounding factors present in the bacterial and tissue culture assays prompted a study in live animals, i.e., a micronucleus assay in rats. The mutagenicity potential was evaluated not only after single doses of 0, 1000, 1500 or 2000 mg kg⫺1, but also after repeated doses of this regimen and 5000 mg kg⫺1 day⫺1 for 29 consecutive days (Christian et al., 2004). TK assays from these concentrations in the 90day toxicity study verified that circulating plasma levels of OPE existed to properly expose the bone marrow tissues used
477
in the mutagenicity assay. A positive control (cyclophosphamide) was also added to the single-dose study. The rats were euthanized and femoral bone marrow samples were collected at 24 or 48 h after the single-dose regimen, and only at 24 h after the repeat-dose treatment. A minimum of 2000 polychromatic erythrocytes (PEs) was scored for micronuclei, and the number of polychromatic erythrocytes among 500 total erythrocytes also was determined for each animal. OPE did not produce adverse clinical or necropsy changes or affect absolute or relative feed consumption values. At 5000 mg kg⫺1 day⫺1, body weight gains for male and female rats were reduced during the third and fourth weeks of daily dosing, as compared with previous weeks. The number of micronucleated polychromatic erythrocytes (MN-PCEs) was not significantly increased in any of the OPE-treated groups, as compared to controls, and the ratio of polychromatic to normochromatic erythrocytes was not affected by OPE (Table 52.8). Results from this study indicate that OPE is not a mutagen under in vivo conditions, i.e., in the micronucleus assay (Christian et al., 2004). Conclusions on the mutagenicity potential of OPE can be summarized as follows: ●
●
●
●
●
Equivocal mutagenic potential elicited by OPE in bacterial reverse mutation assay Induction of chromosome aberrations with high dose of OPE in CHO assay with S9 Rat micronucleus assay negative after single doses of 1000–2000 mg kg⫺1 OPE Rat micronucleus assay negative after repeated doses of 5000 mg kg⫺1 day⫺1 OPE OPE not considered mutagenic under in vivo conditions.
52.7 CONCLUSIONS The studies summarized in this publication indicate that maximal concentrations of OPE (500 mg mL⫺1 of 0.5% w/v methylcellulose) at maximally feasible oral single or repeat dosages in rats (10 mL kg⫺1 body weight) are essentially devoid of toxicity. This 5000 mg kg⫺1 day⫺1 dose is equivalent to a dose of 350 grams in a 70-kg human on a mg kg⫺1 basis. In humans, the suggested daily dose of OPE is one or two 300 mg capsules taken twice daily, a dose that is 1200 mg day⫺1 or approximately 17 mg kg⫺1 day⫺1. Thus, even 2000 mg kg⫺1 day⫺1 (the NOAEL in the 90-day rat study and the reproductive and developmental studies) represents a safety factor that is 118 times the human daily dose.
SUMMARY POINTS ●
Hydrolyzed aqueous olive pulp extract (OPE) is a water extract product of organically grown olives containing high concentrations of polyphenol antioxidants, especially hydroxytyrosol.
478
SECTION | I Toxicology and Contaminants
TABLE 52.8 In vivo mutagenicity results from single and repeat dose micronucleus assays in rats. This table shows that the number of micronucleated polychromatic erythrocytes (MN-PCEs) was not significantly increased in any of the OPE-treated groups, as compared to negative controls, and the ratio of polychromatic to normochromatic erythrocytes was not affected by OPE. Only the CP positive controls had significantly elevated MNPCEs. It was concluded that OPE was not mutagenic in these stringent in vivo assays. Dosage group (5/group/sex)
Males Avg. PCE Fraction
Avg. No. MNPCEs/2000
Females Avg. No. MNNCEs/2000
Avg. PCE Fraction
Avg. No. MNPCEs/2000
Avg. No. MNNCEs/2000
Single-dose trial (assay performed 24 and 48 h after single dosage) Negative Control 24 h
0.636
1.2
0
0.614
1.6
0
1000 mg kg⫺1 24 h
0.653
1.6
0
0.598
2.6
0.2
1500 mg kg⫺1 24 h
0.588
2.4
0.2
0.620
2.4
0.4
2000 mg kg⫺1 24 h
0.655
1.8
0
0.601
4.0
0.2
Negative Control 48 h
0.554
0.6
0
0.504
0.2
0
2000 mg kg⫺1 48 h
0.564
0.0
0
0.664
1.2
0
Positive Control (CP)
0.497
58.2**
1.8
0.309
33.2**
1.6
Repeat-dosage trial (assay performed 24 h after 29 doses on consecutive days) Negative Control 24 h
0.521
0.6
0.2
0.494
0.2
0
5000 mg kg⫺1 24 h
0.368
0
0
0.480
0
0
(CP): Cyclophosphamide; * Statistically significant (P⭐0.05); ** Statistically significant (P⭐0.01). PCE: polychromatic erythrocyte; MN-PCE/2000: micronucleated polychromatic erythrocytes/2000 erythrocytes.
●
●
●
●
●
●
The toxicity of hydrolyzed aqueous olive pulp extract (OPE) was characterized in a series of toxicology studies to assess human safety parameters. Single oral doses of OPE to rodents resulted in a no-observed-adverse-effect-level (NOAEL) of 2000 mg kg⫺1. Daily oral doses of 1000, 1500 or 2000 mg kg⫺1 day⫺1 to rats for 90 days were rapidly absorbed, produced no significant adverse effects due to OPE, and resulted in a NOAEL of 2000 mg kg⫺1 day⫺1. The proposed human dose approximates 17 mg kg⫺1 day⫺1. Repeated oral doses of 500, 1000, 1500 or 2000 mg kg⫺1 day⫺1 did not adversely affect mating, fertility, delivery, or litter parameters in a rat reproduction study. Adverse effects were not observed in a rat developmental toxicity study in which pregnant dams were intubated with 1000, 1500 or 2000 mg kg⫺1 day⫺1 on days 6 through 20 of gestation. Plasma levels of hydroxytyrosol, the major phenolic of OPE, were comparable in pregnant, lactating or non-pregnant rats; minimal levels crossed the placenta;
●
quantifiable levels were not identified in maternal milk or plasma from nursing pups. Based on repeat doses of 5000 mg kg⫺1 day⫺1 to rats for 29 days, OPE was not considered to be a mutagen.
REFERENCES Arocena, L. C., Learn, D. B., Forbes, P. D., Hoberman, A. M., Crea, R., 2003. Protection against ultraviolet radiation-induced cutaneous inflammation with olivenol™ in hairless mice. Photomedicine Society Meeting presentation (abstract). Christian, M.S., Sharper, V.A., Hoberman, A.M., Seng, J.E., Fu, L., Covell, D., Diener, R.M., Bitler, C.M., Crea, R., 2004. The toxicity profile of hydrolyzed aqueous olive pulp extract. Drug Chem. Toxicol 27, 309–330. D’Angelo, S., Manna, C., Migliardi, V., Mazzoni, O., Morrica, P., Capasso, G., Pontoni, G., Galletti, P., Zappia, V., 2001. Pharmacokinetics and metabolism of hydroxytyrosol, a natural antioxidant from olive oil. Drug Metab Dispos 29, 1492–1498. Galli, C., Visioli, F., 1999. Antioxidant and other activities of phenolics in olives/olive oil, typical components of the Mediterranean diet. Lipids 34 (Suppl), S23–S26.
CHAPTER | 52 The Toxicity Profile of Hydrolyzed Aqueous Olive Pulp Extract
Institute of Laboratory Animal Resources, 1996. Guide for the Care and Use of Laboratory Animals. National Academy Press, Washington, DC. Loeb, W.F., 1999. The rat. In: Loeb, W.F., Quimby, F.W. (eds) The Clinical Chemistry of Laboratory Animals. Taylor & Francis, Philadelphia, PA, pp. 33–48. Manna, C., Ragione, F.D., Cucciolla, V., Borriello, A., D’Angelo, S., Galletti, Z.V., 1999. Biological effects of hydroxytyrosol, a polyphenol from olive oil endowed with antioxidant activity. Adv. Exp. Med. Biol. 472, 115–130. Tennant, B.C., 1999. Assessment of hepatic function. In: Loeb, W.F., Quimby, F.W. (eds) The Clinical Chemistry of Laboratory Animals. Taylor & Francis, Philadelphia, PA, pp. 501–517.
479
US Food and Drug Administration, 2000. Redbook 2000 – Toxicological Principles for the Safety of Food Ingredients. Section IV.C.2., Acute Oral Toxicity Tests. Center for Food Safety & Applied Nutrition, Washington, DC. US Food and Drug Administration, 1987. Good Laboratory Practice Regulations; Final Rule. 21 CFR Part 58. Visioli, F., Galli, C., Galli, G., Caruso, D., 2002. Biological activities and metabolic fate of olive oil phenols. European J Lipid Science and Technology 104, 677–684.
This page intentionally left blank
Chapter 53
Plasticizer in Olive Oils Giuseppa Di Bella, Lara La Pera, Vincenzo Lo Turco, Donatella Pollicino and Giacomo Dugo Dept. Food and Environmental Science, University of Messina, Italy
53.1 INTRODUCTION Humans have significant exposures to plasticizers, as these substances are ubiquitously present in flexible plastics. Polyvinyl chloride (PVC) is one of the most versatile plastics because of its blending ability with a variety of additives such as plasticizers and stabilizers to produce a wide range of products including packaging materials. Films and sheets followed by bottle production are the largest applications for PVC in food packaging. They usually possess a low molecular weight and thus have the tendency to migrate from the packaging material into the packaged food, thereby becoming a food contact substance. These plastics are called ‘Food Contact Substances’ by the US Food and Drug Administration (FDA), but until April 2002, they were called ‘Indirect Food Additives’ (Guidance for Industry, 2002). Because of this occurrence, both the US Food and Drug Administration and the EU have set regulations on plasticizer use in food-packaging materials. There are several hundred specific migration limits (SLM) in Directive 2002/72/EC (European Commission, 2002); in particular, 18 mg kg⫺1 food for bis-(2-ethylhexyl)adipate (DEHA) and 3 mg kg⫺1 food for bis-(2-ethylhexyl)phthalate (DEHP). Although DEHP is forbidden by the European legislation, it can sometimes be found in solvent-based inks. These inks are usually dedicated to non-food applications, but components and additives can be transferred to food-packaging inks. It is well documented that plasticizers such as adipates and phthalates from plasticized films readily migrate into fatty foods when there is direct surface contact between film and the food (Goulas et al., 1995, 1998; Goulas and Kontominas, 1996; Oi-Wah Lau and Siu-KayWong, 2000; Fankhauser-Noti and Grob, 2006; Zygoura et al., 2007). Food contamination from plasticizer might also occur during the production cycle since these chemicals are present as additives in the plastic components of the machinery used. For this reason, in recent years great importance has been directed towards the monitoring of alimentary products Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
for the presence of organic contaminants, including plasticizers. The presence of toxic residues in vegetable oils has been reported by several researchers (Di Muccio et al., 1990; Vreuls et al., 1996; Hiskia et al., 1998, Rastrelli et al., 2002; Dugo et al., 2004, 2005). Vegetable oils are one of the most important components of the human diet, therefore the presence of plasticizers constitutes a significant health risk.
53.2 PLASTICIZERS Plastics are made of two types of components. The main component is the polymer, or resin, which makes up the bulk of the plastic material. The second type of component is additives. Plasticizers are additives that increase the plasticity or fluidity of the material to which they are added, these include plastics, cement, concrete, wallboard and clay bodies. Plasticizer contents can vary from 3.0 to 80% by plastic weight which produces a considerable effect on material properties. With time, plasticizers can migrate to plastic article surface carrying other components of the composition (e.g., stabilizers) with them (Table 53.1). As mentioned above, a variety of compounds which are generally defined as plasticizers and which are used as technological additives by the plastic materials industry can contribute to the contamination of vegetables oils. These substances can be divided into five groups. Chloroparaffins are industrial mixtures of chloroalkanes of different chain length (C10–C23), with a chlorine content between 40 and 70% in weight (Figure 53.1). They are considered secondary plasticizers or extenders and are always employed together with a primary plasticizer. Chloroparaffins are divided in three different classes according to the chain length: short-chain, C10–13; medium-chain, C14–17; long-chain C18–23. Their toxicity strictly depends on the chain length; in particular short-chain chloroparaffins cause cancer in rats (Tomy et al., 1998). The United States and Europe have classified
481
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
482
SECTION | I Toxicology and Contaminants
short-chain chloroparaffins as possible carcinogenic compounds for humans. Triarylphosphates (composed of seven different classes of isomers: triphenylphosphates, diphenyltolylphosphates, phenylditolyphosphates, tritolylphosphates, ditolylxylylphosphates, tolydixylylphosphates, and trixylylphosphates), which are commonly known as phosphorated plasticizers (Figure 53.2). These compounds are inhibitors of colinhexterase, therefore they cause paralysis and exhibit a large number of neurotoxic effects (Fest and Schmidt, 1982). Phthalic acid esters represent the most important group of plasticizers. In particular, bis-(2-ethylhexyl)phthalate, commonly known as di-octyl phthalate (DOP) is extensively used as a plasticizer for PVC (Figure 53.3). Several factors make phthalates not ideal for use as plasticizers. They migrate to the surface of plastics, and can then evaporate or leach into the surrounding environment. This limits the usefulness
O n-C4H9
O
P
O
n-C4H9
O n-C4H9 O O
P
O
O
A O O H3C
P
O CH3
O
CH3
B
FIGURE 53.2 (A) Triphenylphosphates’ and (B) tritolylphosphates’ chemical structure.
TABLE 53.1 Key facts for plasticizers. 1. Plastics are made of two types of components: the main component is the polymer, or resin, the second type of component is additives. Plasticizers are additives that increase the plasticity or fluidity of the material to which they are added, these include plastics, cement, concrete, wallboard and clay bodies 2. Plasticizers can be divided into five groups: chloroparaffins, phosphoric acid esters, phthalates, adipates and sebacates 3. The most common type of plasticizer is chemicals called phthalates, chemically phthalic acid esters. In particular, di-octyl phthalate (DOP) is extensively used as a plasticizer for PVC 4. Recent epidemiological evidence indicates that plasticizers, and particularly phthalates, are toxic and have shown to damage the liver and testes and cause birth defects. Many phthalates, including DOP, are classified as toxic chemicals by the EPA’s Toxic Release Inventory. The use of some phthalates is restricted in the European Union for use in children’s toys 5. Plasticizers are present in food as a result of processing and storage procedure contamination
A CH2
CH
C O CH2 CH (CH2)3
CH3
CH2CH3 C O CH2 CH (CH2)3 CH3 O O
A
C O (CH2)3 CH3 C O (CH2)3 CH3 B
O O C O CH2CH3 C O CH2CH3 O
C
FIGURE 53.3 (A) Di-octyl phthalate, (B) di-butyl phthalate, (C) di-ethyl phthalate chemical structure.
Cl
Cl H3C
CH2CH3
O
CH2
CH
CH2
CH2
CH
CH2
CH
CH2
Cl
Cl
CH2 Cl
Cl
B H3C
CH
CH Cl
CH2
CH2
CH2
CH Cl
CH2
CH2
CH2
CH Cl
CH2
CH2
CH2
CH
CH2
CH2
Cl
FIGURE 53.1 (A) Short-chain (from C10 to C20) and (B) long-chain (from C18 to C23) chloroparaffins.
CH2
CH Cl
CH2
CH2
CH2
CH2 Cl
483
CHAPTER | 53 Plasticizer in Olive Oils
of phthalate plasticizers, as they eventually migrate out of plastics entirely, and as a result the plastics become brittle. The release of phthalates into the environment represents an environmental hazard as well. Because of the widespread use of phthalates, they have become one of the most abundant industrial pollutants in the environment. Many phthalates, including DOP, are classified as toxic chemicals by the EPA’s Toxic Release Inventory. Others commonly used are di-isobutyl and di-nbutylphthalate, (DiBP) and (DBP) respectively. Studies on rodents involving large amounts of phthalates have shown damage to the liver and testes and to cause birth defects. In addition, a recent British study showed that the phthalate di-butyl phthalate (DBP) or its metabolite mono-butyl phthalate (MBP) suppresses steroidogenesis by fetal-type Leydig cells in primates as in rodents (Ema et al., 1990; Parks et al., 2000; Hallmark et al., 2007). The use of some phthalates is restricted in the European Union for use in children’s toys. Bis-(2-ethylhexyl)phthalate (DEHP), butylbenzylphthalate (BBP), and di-butylphthalate (DBP) are restricted for all toys; di-iso-nonylphthalate (DiNP), di-iso-decylphthalate (DiDP), and di-octylphthalate (DOP) are restricted for all toys; di-iso-nonylphthalate (DINP), diiso-decylphthalate (DIDP), and di-n-octylphthalate (DOP) are restricted only in toys that can be taken into the mouth. The restriction states that the amount of phthalates may not be greater than 0.1% of the plasticized part of the toy. There are no other specific restrictions in the European Union, and the phthalates mentioned are allowed in any concentration in other products. Other phthalates are not restricted. Adipic acid esters. Esters with the diacid hexanedicarboxylic acid, trivial name adipic acid, and alcohols of varied carbon chain length are called adipates. The smallest adipates (with short-chain alcohols) are used as solvents for paint and in paint removers; they serve as solubility mediators since they can solve both organic substances and water, thus facilitating the blending of non-mixable materials like plastic and water. One of the most diffused adipate is bis-(2-ethylhexyl) adipate (DEHA) (Figure 53.4). The most widely used plasticizer for food contact applications in PVC is DEHA, often used in concentrations exceeding 20% by weight of the polymer. Also in cosmetics, different adipates are used as softeners. Sebacic acid esters. Esters with the diacid octanedicarboxylic acid, trivial name sebacic acid, and alcohols of varied carbon chain length are called sebacates. Dimethyl sebacate is used as a plasticizer and solvent for resins and rubbers. It is used as an intermediate to produce other organic compounds including UV stabilizers, pharmaceutical and colorants. Di-butyl sebacate and bis-(2-ethylhexyl) sebacate (DEHS) (Figure 53.5) are plasticizers permitted in the field of food contact materials, medical and pharmaceutical supplies. They are also used as a plasticizer for polymers and synthetic rubbers. Higher-chain compounds are used as components in metalworking fluids, surfactants, lubricants,
O CH2 Et
O
C
O (CH2)4
C
Et O
CH2
CH Bu n
CH Bu n
FIGURE 53.4 Bis-(2-ethylhexyl) adipate chemical structure.
Et
O
O
n Bu CH CH2 O C (CH2)8 C O
Et CH2 CH Bu n
FIGURE 53.5 Bis-(2-ethylhexyl) sebacate chemical structure.
detergents, oiling agents, emulsifiers, wetting agents, textile treatments and emollients. They are also used as intermediates for the manufacture of a variety of target compounds.
53.2.1 Analysis of Plasticizers: Methodological Considerations The analysis of different classes of plasticizers including chloroparaffins, phosphoric acid esters, phthalates, adipates and sebacates in olive oils is usually performed by highresolution gas chromatography (HR-GC), equipped with different detectors. In particular, the determination of chloroparaffins was managed using an electron capture detector (ECD), the determination of phosphorate plasticizers was carried out using a flame photometric detector (FPD), whereas a mass spectrometer detector (MS) has been used for the analysis of phthalates, adipates and sebacates. Before GC-ECD analysis of chloroparaffins, vegetable oil samples (0.2 mL) were passed through a glass column packed with silica gel, previously activated at 550 °C for 3 h. Chloroparaffins were eluted with 110 mL of n-hexane, concentrated to 0.2 mL and spiked with 0.2 mL of 1 ppm methyl bromophos, used as the internal standard. The separation of chloroparaffins was achieved using a RTX-5 (30 m ⫻ 0.32 mm, film 0.25 μm) capillary fused silica column from Restek; the column oven temperature program was 50 °C (held for 2 min)⫺150 °C at 25 °C min⫺1; 150– 270 °C (20 min) at 4 °C min⫺1. The injection was performed in the splitless mode at 250 °C, the sampling time was 2 min; the ECD temperature was 280 °C. Helium was the carrier gas at the linear velocity of 36 cm s⫺1. The chromatogram of a standard mixture of chloroparaffins (Figures 53.6–53.8) shows the characteristic profile of broadened peaks that correspond to many isomers. The limit of detection (LOD), calculated as three-fold the signal to noise ratio, was 0.370 mg kg⫺1. The accuracy of the method was assessed by spiking a standard mixture of chloroparaffins (C6–C18) dissolved in n-hexane, at different concentration levels in a ‘blank’ olive oil sample; spiked samples were subjected to the clean-up procedure described earlier, the obtained recoveries ranged from 77.5 to 91.4%.
484
SECTION | I Toxicology and Contaminants
A
5 3
4 6
7 9 8
2 1
10
5
10
15
20
25 Time
30
35
40
45
50
FIGURE 53.6 HRGC-ECD chromatogram of a standard mixture of chloroparaffins.
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 min
B 67 9
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 min
FIGURE 53.8 HRGC-MS chromatogram of (A) standard mixture and (B) Sicilian olive oil.
FIGURE 53.7 HRGC-FPD chromatogram of a standard mixture of (1) tributylphosphates, (2) methyl bromophos, (3) triphenylphosphates, (4) tritolylphosphates.
Phosphorated plasticizers were extracted from edible oil as follows: 2 g of oil were mixed with 2 mL of acetonitrile in a closed vial and manually shacked. The phases were separated, the supernatant (acetonitrile) was discharged and the oily one was extracted again as described. The acetonitrile phases were collected, centrifuged at 5000 rpm for 10 min and used for
the GC analysis. Phosphorate plasticizer separation (tributyl-, triphenyl- and triarylphosphates) was achieved using a MEGA 68 (cyano-methyl-phenyl polyoxan) (25 m ⫻ 0.32 mm, film 0.45 μm) capillary fused silica. The column oven temperature program was as follows: 75 (5 min)–100 °C at 7.5 °C min⫺1; 100–170 °C at 2 °C min⫺1; 170 (5 min)–250 °C at 10 °C min⫺1. A PTV injector was used with the following program: from 65 to 240 °C in 5 min at 999 °C min⫺1. The FPD temperature was 250 °C; N2, H2 and air were the detector gases at the pressure of 0.2, 2.8 and 0.4 bar. Helium was used as the carrier gas (0.3 bar). Triarylphosphates were further recognized using HRGC-MS. The LODs (S/N ⫽ 3) and the accuracy of the method, assessed by spike and recovery test, are reported in Table 53.2. Phthalates, adipates and sebacates were extracted from vegetable oils using the same method described for phosphorate plasticizers. All the solvents employed in the sample preparation procedure were bi-distilled in order to avoid contamination from phthalates.
485
CHAPTER | 53 Plasticizer in Olive Oils
TABLE 53.2 Limits of detection (LOD) and accuracy (recovery %) for phosphorated plasticizers determined in olive oils by HRGC-FPD. LOD (mg kg⫺1)
Recovery range (%)
Tributylphosphate
0.09
78.8–80.2
Triphenylphosphates
0.089
86.9–88.2
Triarylphosphates
0.180
76.8–78.8
This table reports limits of detections and accuracy for the determination of phosphorated plasticizers in olive oils by HRGC-FPD. The detection limit of a method is the lowest analyte concentration that produces a response detectable above the noise level of the system, typically, three times the noise level, whereas the accuracy of a method is the closeness of the measured value to the true value for the sample. Usually, the accuracy of an analytical method is valued by performing measures on standard reference matrices; National Institute of Standards and Technology (NIST) reference standards are often used. If such a well-characterized sample is not available, one of the most common approaches to assess the accuracy is to spike the sample matrix with different amounts of the analyte of interest and then to perform the extraction and the analysis according to the method developed. Recovery, which represents the accuracy, is calculated according to the Cfound Re cov ery % ⫽ ⭈ 100 following equation: Cexp ected
For the separation of phthalates, adipates and sebacates a Supelco MDN-5S (5% di-phenyl–95% di-methyl siloxane) capillary column (30 m ⫻ 0.25 mm, film 0.25 μm) was employed; the GC column oven temperature programme was from 60–275 °C (held for 14 min) at 15 °C min⫺1. The carrier gas was helium at a flow rate of 40 cm s⫺1, the interface temperature was 230 °C. The injector temperature was 250 °C and the splitless injection mode was used; the injection volume was 1 μL with a split ratio of 1:30 for 1 min. The acquisition was performed in the SIM mode; the ionization energy and emission current were 70 eV and 250 μA, respectively. Table 53.3 reports the m/z ratios and the time range of the monitored ions. The LODs (S/N ⫽ 3) and the accuracy of the method, assessed by spike and recovery test, are reported in Table 53.4.
53.2.2 Plasticizer Residues in Sicilian Olive Oils Phosphorated plasticizers, chloroparaffins, phthalates, adipates and sebacates contamination in extra virgin olive oils produced in Sicily in the crop years 2002–2003, 2006– 2007, and 2007–2008 was valued using the analytical methods described in the previous section. In particular, five samples from 2002–2003, 13 samples from 2006–2007 and 13 samples from 2007–2008. The samples were produced
TABLE 53.3 Time intervals and selected ion monitoring (SIM) of phthalate, adipate and sebacate determined in olive oils by HRGC-MS. Plasticizer
Time (min)
SIM (m/z)
Di-methyl adipate (DMA)
5– 7
101, 114, 143
Di-ethyl adipate (DEA)
7– 9
111, 157
Di-ethyl phthalate (DEP)
9–10.5
149, 177
Di-butyl adipate (DBA)
10.5–11.9
11, 129, 185
Bis(metoxyethyl)adipate
10.5–11.9
58, 11, 155
Di-isobutyl phthalate (DiBP)
11.9–12.6
149, 167, 223
Di-butyl phthalate (DBP)
12.6–14.5
149, 223
Butyl benzyl phthalate (BBP)
13–14.2
149, 206
Di-octyl phthalate (DOP)
14.2–17
149, 167
Bis-(2-ethylhexyl) adipate (DEHA)
14.5–16.2
129, 147
Di-iso-nonyl phthalate (DiNP)
16.2–19
149, 167
Bis-(2-ethylhexyl) sebacate (DEHS)
19–22
185, 203, 297
from Sicilian varieties in firms that export olive oils in all Italian regions. All the samples were stored in dark glass bottles. Table 53.5 gives evidence that chloroparaffin and phosphorated plasticizer residues were lower than the detection limits in all the studied samples. One out of five samples from 2002–2003 presented diethyl phthalate (DEP) contamination (0.04 ppm), whereas the 54% of samples from 2006–2007 had DEP mean level of 0.211 ppm; concentrations lower than the LOD were found in samples from 2007–2008. Di-isobutyl phthalate (DiBP) mean levels significantly decreased (p ⬍ 0.005) from 2002 to 2008, even though the number of contaminated samples increased from 60 to 100%. Also the mean concentrations of di-butyl phtalate (DBP) significantly decreased (p ⬍ 0.05) over the years; the percentage of contaminated samples increased from 60% (2002–2003) to 92% (2006–2007) and decreased to 46% in the last crop years. Data of butylbenzyl- and di-iso-nonylphthalate contamination in samples from 2002–2003 were not available; whereas a remarkable decrease (p ⬍ 0.005) of their concentration was observed from 2006 to 2008. One hundred per cent of the samples from both crop years showed BBP contamination; whereas di-iso-nonylphthalate was detected in 39% of samples from 2006–2007 and 46% from 2007–2008.
486
SECTION | I Toxicology and Contaminants
TABLE 53.4 Limits of detection (LOD) and accuracy (recovery %) for phthalates, adipates and sebacates determined in olive oils by HRGC-MS. Plasticizer
LOD (mg kg⫺1)
Recovery range (%)
Di-methyl adipate (DMA)
0.026
83.3–108.4
Di-ethyl adipate (DEA)
0.025
78.5–101.3
Di-ethyl phthalate (DEP)
0.031
75.7–101.3
Di-butyl adipate (DBA)
0.095
84.6–90.7
Bis(2-metoxyethyl)adipate (DMEA)
0.056
99.3–102.5
Di-isobutyl phthalate (DiBP)
0.190
84.3–86.6
Di-butyl phthalate (DBP)
0.190
81.5–94.6
Butyl benzyl phthalate (BBP)
0.020
91.5–98.6
Di-octyl phthalate (DOP)
0.010
Bis-(2-ethylhexyl) adipate (DEHA)
TABLE 53.5 Plasticizer concentration expressed in mg kg⫺1 (mean value and % of contaminated samples) in Sicilian extra virgin olive oils from 2002–2003 (n ⴝ 5), 2006–2007 (n ⴝ 13) and 2007–2008 (n ⴝ 13) crop years. 2002–2003
2006–2007
2007–2008
Chloroparaffins
nd*
nd
nd
Tri-butyl-phosphate
nd
nd
nd
Tri-phenyl-phosphate
nd
nd
nd
Tri-tolyl-phosphate
nd
nd
nd
Di-ethyl phthalate
0.040 (20%)
0.211 (54%)
nd
Di-isobutyl phthalate
0.447 (60%)
0.225 (100%)
0.187 (100%)
Di-butyl phthalate
0.241 (60%)
0.200 (92%)
0.160 (46%)
Butyl benzyl phthalate
na**
0.454 (100%)
0.260 (100%)
Di-octyl phthalate
nd
0.060 (15%)
0.260 (8%)
99.7–101.5
Di-iso-nonyl phthalate
na**
1.356 (39%)
0.812 (46%)
0.075
66.3–70.0
Di-ethylhexyl adipate
0.248 (80%)
1.157 (100%)
2.033 (92%)
Di-ethylhexyl sebacate
nd
nd
nd
Di-iso-nonyl phthalate (DiNP)
0.095
86.2–91.6
Bis-(2-ethyl exyl) sebacate (DEHS)
0.036
*
nd ⫽ not detectable. na ⫽ not available.
**
99.4–101.2
This table reports limits of detections and accuracy for the determination of phosphorated plasticizers in olive oils by HRGC-FPD. The detection limit of a method is the lowest analyte concentration that produces a response detectable above the noise level of the system, typically, three times the noise level. Whereas the accuracy of a method is the closeness of the measured value to the true value for the sample (see legend to Table 53.2).
Di-octyl phthalate (DOP) levels were lower than the LOD in samples from 2002–2003, the 15% of olive oils from 2006–2007 presented low contamination mean levels (0.06 mg kg⫺1), whereas a mean concentration four times higher than this was found in the 8% of samples from 2007–2008. Bis-(2-ethylhexyl) adipate (DEHA) residues significantly increased (p ⬍ 0.005) over the studied crop years: in 2007–2008 samples the average concentration level was eight times higher than that observed in samples from 2002–2003; also the percentage of contaminated samples increased. Bis-(2-ethylhexyl) sebacate (DEHS) levels were lower than the LOD in all the studied extra virgin olive oils.
53.2.3 Plasticizers in Commercial Vegetable Oils In 2003 Di Bella et al. reported the concentration of phosphorated plasticizers, chloroparaffins, phthalates, adipates and sebacates in 12 samples of different typology of Italian commercial vegetable oils (nut, maize, soy, sunflower, blend of seeds, castor, olive marc). Of all the studied samples, eight were stored in glass bottles and four were stored in plastic packaging (Table 53.6). No data regarding samples produced in 2006–2007 and 2007–2008 are available. The European Union has one of the highest rates of vegetable oil consumption in the world. As a result, the migration of plasticizers from plastic packaging into oil is of utmost importance to EU authorities, as well as to consumers because a significant portion of vegetable oils, particularly seed oils, are packaged in PVC bottles. As already observed for extra virgin olive oils, also in commercial vegetable oils among the studied plasticizers, phthalates were the most representative. In particular, diethyl phthalate (DEP) is present in all the studied samples, except nut oil, in concentrations ranging from 0.056 to 1.1 mg kg⫺1: maize oil and the blend of seeds oil stored in
487
CHAPTER | 53 Plasticizer in Olive Oils
TABLE 53.6 Mean concentration (mg kg⫺1) of plasticizers in commercial seed oils. Nut (n ⴝ 1)
Sunflower (n ⴝ 1)
Maize (n ⴝ 1)
Maize* (n ⴝ 1)
Blend (n ⴝ 1)
Blend* (n ⴝ 3)
Soy (n ⴝ 2)
Castor (n ⴝ 1)
Sansa (n ⴝ 1)
Chloroparaffins
nd
nd
nd
nd
nd
nd
nd
nd
nd
Tributylphosphate
nd
nd
nd
nd
nd
nd-0.092
nd
nd
nd
Triphenylphosphate
nd
nd
nd
nd
nd
nd
nd
nd
nd
Tritolylphosphate
nd
nd
nd
nd
nd
nd-0.189
nd
nd
nd
Di-methyl adipate
nd
nd
nd
nd
nd
nd
nd
nd
nd
Di-ethyl adipate
nd
nd
nd
nd
nd
nd
nd
nd
nd
Di-ethyl phthalate
nd
0.152
0.130
0.514
0.172
0.83–1.1
nd-0.197
0.056
0.424
Di-butyl adipate
nd
nd
nd
nd
nd
nd
nd
nd
nd
Bis-(2-metoxyethyl) adipate nd
nd
nd
nd
nd
nd
nd
nd
nd
Di-isobutyl phthalate
0.253
nd
0.254
nd
0.213
nd-0.20
0.224–0.49
nd
nd
Di-butyl phthalate
nd
nd
nd
0.214
0.228
nd
nd
nd
nd
Bis-(2-ethylhexyl) adipate
nd
nd
nd
nd
nd
nd
nd
nd
nd
Di-octyl phthalate
0.145
0.117
0.122
0.504
nd
0.32–0.51
0.110–0.129 0.155
0.099
Bis-(2-ethylhexyl) sebacate
nd
0.087
0.132
0.195
0.05
nd-0.176
nd
0.107
0.375
*
stored in plastic packaging; nd ⫽ non detectable.
plastic bottles showed significantly higher concentrations of DEP with respect to samples stored in glass bottles. Diisobutyl phthalate was found in seven out of 12 samples in concentrations within 0.50 mg kg⫺1. Di-butyl phthalate was detected only in maize samples packaged in plastic containers and in the blend of seed oil stored in glass bottles. Di-octyl phthalate (DOP) was found in all the samples, except the blend of seed stored in glass bottles; the highest level was measured in maize oil stored in plastic package (0.504 mg kg⫺1). The studied extra virgin olive oils from 2006–2007 and 2007–2008 showed DOP levels lower than those found in commercial vegetable oils. Bis-(2-ethylhexyl) sebacate (DEHS) was detected in concentrations within 0.40 mg kg⫺1 in five out of 12 samples: also in this case the blend of seed oils stored in plastic packaging presented higher levels of DEHS. In the studied extra virgin olive oils, bis-(2-ethylhexyl) adipate (DEHA) was the one most representative plasticizer contaminant; on the contrary, in all the studied commercial seed oils DEHA levels were lower than the limit of detection. Chloroparaffin levels were lower than the LOD in all the samples; among the studied phosphorated plasticizers
tributylphosphate and tritolylphosphate were detected in concentrations within 0.2 mg kg⫺1, in two of three samples of blends of seed oil stored in plastic bottles.
53.3 CONCLUSION Currently the European Community has not fixed any legal limit for plasticizer residues in edible oils but, as reported in the previous section, has adopted legal regulations designed to limit the amount of migration. In 2005 the Bundersverband NaturKost Naturwaren (BNN) Herstellung und Handel established safety values for phthalate and adipate residues in edible oils, taking into account they are ubiquitous substances: for DOP, DiBP, and BBP the critical value is of 10 mg kg⫺1, whereas for DEHP it is 6 mg kg⫺1. Even though all the studied Sicilian extra virgin olive oils and commercial seed oils presented phthalate concentrations lower than the safety limits established by the BNN, it is of primary importance to prevent food contamination by applying all the preventive measures along the supply chain.
488
SECTION | I Toxicology and Contaminants
SUMMARY POINTS ●
●
●
●
●
●
●
Plasticizers are present as additives in the plastic components of the machinery used to extract and/or decant the oils, therefore they are transferred to the product oils during the production cycle, or that of the packages. Plasticizers are mainly classified into five groups: phosphorated plasticizers, chloroparaffins, phthalates, adipates and sebacates The most common type is the phthalates. In particular, di-octyl phthalate (DOP) is extensively used as a plasticizer for PVC. Humans have significant exposures to plasticizers, as these substances are ubiquitously present in flexible plastics. For this reason, in recent years great importance has been directed towards the monitoring of alimentary products for the presence of organic contaminants including plasticizers. Recent epidemiological evidence indicates that plasticizers, and particularly phthalates, are toxic and have shown damage to the liver and testes and cause birth defects. The analysis of phosphorated plasticizers, chloroparaffins, phthalates, adipates and sebacates was performed on commercial samples of extra virgin olive oils produced in Sicily in different crop years (2002–2003, 2006–2007, 2007–2008) and on Italian commercial vegetable oils produced in 2003 (nut, sunflower, soy, maize, blend of seeds, castor, sansa). The obtained results gave evidence that of all the studied plasticizer, phthalates were the most representative both in extra virgin olive oils and in seed oils. The concentration of phthalates found in the studied oils were within the safety limits indicated in 2005 by the Bundersverband NaturKost Naturwaren (BNN).
REFERENCES Di Bella G., Condoleo C., Bruzzese A., Salvo F., Cappello A., Dugo, G.mo., 2004. Residui di plastificanti in oli vegetali: Nota I. La Rivista Italiana delle Sostanze Grasse, Vol. LXXXI (July–Aug.), 207–210. Di Muccio, A., Ausili, A., Vergori, L., Camoni, I., Dommarco, R., Gambetti, L., Santilio, A., Versori, F., 1990. Single-step multicartridge cleanup for organophosphate pesticide residue determination in vegetable oil extracts by gas chromatography. Analyst 115, 1167–1169. Dugo, G.mo., Di Bella, G., Condoleo, C., Saitta, M., Lo Turco, V., 2004. Residui di plastificanti in oli vegetali: Nota II. La Rivista Italiana delle Sostanze Grasse, Vol. LXXXI (Sept–Oct), 273–276. Dugo, G.mo., Di Bella, G., La Torre, L., Saitta, M., 2005. Rapid determination of organophosphorus pesticide residues in Sicilian and Apulian olive oil. Food Control 16, 435–438. Ema, M., Murai, T., Itami, T., Kawasaki, H., 1990a. Inizio modulo. Ema, M., Murai, T., Itami, T., Kawasaki, H., 1990b. Evaluation of the teratogenic potential of the plasticizer butyl benzyl phthalate in rats. J. Appl. Toxicol. 10, 339–343.
European Commission, 2002. Commission Directive No. 2002/72/EC, 6 August 2002 relating to plastic materials and articles intended to come into contact with foodstuffs (and amendments). Official Journal of the European Communities, L220/18,15.8.2002. Fankhauser-Noti, A., Grob, K., 2006. Migration of plasticizers from PVC gaskets of lids for glass jars into oily foods: Amount of gasket material in food contact, proportion of plasticizer migrating into food and compliance testing by simulation. Trends Food Sci. Technol. 17, 105–112. Fest, C., Schmidt, K.J., 1982. The Chemistry of Organophosphorus Pesticides, 2nd revised edition. Springer-Verlag, Berlin, pp. 282–289. Goulas, A.E., Kokkinos, A., Kontominas, M.G., 1995. Effect of γ-radiation on migration behaviour of dioctyladipate and acetylbutylcitrate plasticizers from food-grade PVC and PVDC/PVC films into olive oil. Z. Lebensm. Unters. Forsch. 201, 74–78. Goulas, A.E., Riganakos, K.A., Ehlermann, D.A.E., Demertzis, P.G., Kontominas, M.G., 1998. Effect of high-dose electron beam irradiation on the migration of DOA and ATBC plasticizers from food-grade PVC and PVDC/PVC films, respectively, into olive oil. J. Food Prot. 61, 720–724. Goulas, A.E., Kontominas, M.G., 1996. Migration of dioctyladipate plasticizer from food-grade PVC film into chicken meat products: effect of γ-radiation. Z. Lebensm. Unters. Forsch. 202, 250–255. Guidance for Industry, 2002. Preparation of Food Contact Notifications and Food Additive Petitions for Food Contact Substances: Chemistry Recommendations FINAL GUIDANCE U.S. Food and Drug Administration, Center for Food Safety & Applied Nutrition, Office of Food Additive Safety, April 2002. http://www.cfsan.fda.gov/~dms/ opa2pmnc.htmL. Hallmark, N., Walker, M., McKinnell, C., Mahood, I.K., Scott, H., Bayne, R., Coutts, S., Anderson, R.A., Greig, I., Morris, K., Sharpe, R.M., 2007. Effects of monobutyl and di(n-butyl) phthalate in vitro on steroidogenesis and Leydig cell aggregation in fetal testis explants from the rat: comparison with effects in vivo in the fetal rat and neonatal marmoset and in vitro in the human. Environ. Health Perspect. 115, 390–396. Hiskia, A.E., Atmajidou, M.E., Tsipi, D.F., 1998. Determination of organophosphorus pesticide residues in greek virgin olive oil by capillary gas chromatography. J. Agric. Food Chem. 46, 570–574. Lau, O.-W., Wong, S.-K., 2000. Contamination in food from packaging material. J. Chrom A 882, 255–270. Parks Jr., L.G., Ostby, J.S., Lambright, C.R., Abbott, B.D., Klinefelter, G.R., Barlow, N.J., Gray, L.E., 2000. The plasticizer diethylhexyl phthalate induces malformations by decreasing fetal testosterone synthesis during sexual differentiation in the male rat. Toxicol. Sci. 58, 339–349. Rastrelli, L., Totaro, K., De Simone, F., 2002. Determination of organophosphorus residues in Cilento (Campania, Italy) virgin olive oil by capillary gas chromatography. Food Chem. 79, 303–305. Tomy, G.T., Fisk, A.T., Westmore, J.B., Muir, D.C.G., 1998. In: Albert, L.A., Hutzinger, O., Knaak, J.B., Mayer, F.L., Morgan, D.P., Park, D.L., Tjeerdema, R.S., Voogt, P.D., Yang, R.S.H., Whitacre, D.M. (eds), Reviews of Environmental Contamination and Toxicology, Vol. 158, Springer, New York, pp. 53–128. Vreuls, J.J., Swen, R.J.J., Goudriaan, V.P., Kerkhoff, M.A.T., Jongenotter, G.A., Brinkman, U.A., 1996. Automated on-line gel permeation chromatography-gas chromatography for the determination of organophosphorus pesticides in olive oil. J. Chromatogr. 750, 275–286. Zygoura, P.D., Goulas, A.E., Riganakosa, K.A., Kontominas, M.G., 2007. Migration of bis(2-ethylhexyl)adipate and acetyltributyl citrate plasticizers from food-grade PVC film into isooctane: Effect of gamma radiation. J. Food Eng. 78, 870–877.
Chapter 54
Polycyclic Aromatic Hydrocarbons (PAH) in Olive Oils and Other Vegetable Oils; Potential for Carcinogenesis Isabel Mafra1, Joana S. Amaral1,2 and M. Beatriz P.P. Oliveira1 1 2
REQUIMTE/Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal Escola Superior de Tecnologia e de Gestão, Instituto Politécnico de Bragança, Portugal
54.1 INTRODUCTION Polycyclic aromatic hydrocarbons (PAH) comprise a family of more than 100 compounds, some of which are known or suspected to be mutagenic and/or carcinogenic to mammals. They are lipophilic organic contaminants composed by two or more fused aromatic rings. PAH containing up to four fused benzene rings are known as light PAH and those containing more than four benzene rings are called heavy PAH. Heavy PAH are more stable and considered to be more toxic than the light ones (Wenzl et al., 2006). These compounds are generated by incomplete combustion of organic matter arising, in part, from natural combustion (forest fires, volcanic eruptions) and mostly from human activities (engine exhausts, industrial production, coalderived products, petroleum distillates, waste incineration, tobacco smoke, among others). Due to their multiple potential sources of contamination, PAH are ubiquitously distributed in nature. Therefore, human exposition, attributable to occupational, environmental and dietary sources, is virtually unavoidable, raising an important public health concern due to their recognized carcinogenic activity (Table 54.1). In the 1970s, 16 PAH were identified as priority pollutants by the Environmental Protection Agency (EPA), based on their occurrence and carcinogenicity (Table 54.2) (Wenzl et al., 2006). Eight of these PAH are known to be mutagenic or carcinogenic and comprise part of the 15 European Union (EU) priority PAH (Table 54.2). In 2005 the EU introduced new legislation (European Commission, 2005a) in response to food-contamination problems based on data collected by the European Member States and risk assessment by the Scientific Committee for Food (SCF) in 2002. The SCF assessed 33 PAH and concluded that 15 PAH showed clear evidence of genotoxicity, and 14 of these were Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
carcinogenic to animals (European Commission, 2002). As measures are presently widely focused on benzo[a]pyrene (BaP), a compound classified as carcinogenic to humans (group 1 of IARC categories), the SCF concluded that BaP could be used as a marker. As BaP constitutes only 1–20% of the total concentration of carcinogenic PAH, the SCF also recommended monitoring another 14 PAH, both in food and the environment, to enable long-term exposure assessments and to verify the usefulness of BaP as a marker (European Commission, 2002). Of the 15 EU priority PAH, 12 were reasonably anticipated to be human carcinogens by the International Agency for Research on Cancer (IARC), which showed sufficient evidence of carcinogenicity in experimental animals (Wenzl et al., 2006). In 2005, the European Commission asked the EU Member States for further investigations on the 15 EU priority PAH
489
TABLE 54.1 Key features of PAH. 1. PAH comprise a family of more than 100 organic contaminants generated by incomplete combustion 2. Light PAH contain up to four benzene rings, while heavy PAH have more 3. PAH are known or suspected to be mutagenic and/or carcinogenic to mammals 4. Human exposure is attributable to occupational, environmental and dietary sources 5. Due to their lipophilic nature, PAH can easily contaminate oils and fats
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
490
SECTION | I Toxicology and Contaminants
TABLE 54.2 Molecular structures of PAHs of concern by different world organizations (adapted from Wenzl et al., 2006). This table presents the molecular structures of the PAH considered as being dangerous by different world organizations. Structure
US-EPAa
EU-SCFb
JECFAc
Groupd
Name
Abbreviation
Acenaphthene
ACP
X
3
Acenaphthylene
ACY
X
NSe
Anthracene
ANT
X
3
Fluoranthene
FLT
X
3
Fluorene
FLR
X
3
Naphthalene
NAP
X
2B
Phenanthrene
PHE
X
3
Pyrene
PYR
X
3
Benz[a]anthracene
BaA
X
X
X
2B
Benzo[b]fluoranthene
BbF
X
X
X
2B
Benzo[j]fluoranthene
BjF
X
X
2B
Benzo[k]fluoranthene
BkF
X
X
X
2B
Benzo[ghi]perylene
BgP
X
X
Benzo[a]pyrene
BaP
X
X
3
X
1
491
CHAPTER | 54 Polycyclic Aromatic Hydrocarbons (PAH) in Olive Oils and Other Vegetable Oils
Chrysene
CHR
Cyclopenta[cd]pyrene
CPP
Dibenz[a,h]anthracene
DhA
Dibenzo[a,e]pyrene
X
X
X
X
X
2B
2A
X
X
2A
DeP
X
X
3
Dibenzo[a,h]pyrene
DhP
X
X
2B
Dibenzo[a,i]pyrene
DiP
X
X
2B
Dibenzo[a,l]pyrene
DlP
X
X
2A
Indeno[1,2,3-cd]pyrene
IcP
X
X
2B
5-Methylchrysene
5MC
X
X
2B
Benzo[c]fluorene
BcL
X
3
X
a
US-Environmental Protection Agency. Scientific Committee on Food from European Union. Joint FAO/WHO Expert Committee on Food Additives. d IARC categories (IARC, 2006): Group 1: carcinogenic to humans; Group 2 A: probably carcinogenic to humans; Group 2B: possibly carcinogenic to humans; Group 3: not classified as to their carcinogenicity to humans. e Not studied.
b c
together with one PAH (benzo[c]fluorene) highlighted by the Joint FAO/WHO Expert Committee on Food Additives in the same year (Table 54.2). The areas for investigation were outlined in Commission Recommendation 2005/108/ EC (European Commission, 2005b).
54.2 OCCURRENCE OF PAH IN FOODS The presence of PAH has been reported in all types of foods, including raw or unprocessed products, processed and cooked food products (Phillips, 1999). Its origin in
492
foods has been widely studied and, in general, in addition to environmental contamination, certain types of food processing have been considered as the main cause of contamination by these compounds (Guillén et al., 1997). In a wide variety of plants, three possible sources of contamination by PAH have been considered: uptake as a result of atmospheric exposure, uptake from the soil and endogenous biosynthesis (Phillips, 1999). However, atmospheric pollution is considered by most investigators as the main source of contamination of unprocessed foods (Guillén et al., 2004; Rodríguez-Acuña et al., 2008). Around 70 different PAH or related compounds have been identified in foodstuffs, from which benzo[a]pyrene and benz[a]anthracene are the most abundant, existing in high quantities in cooked or smoked meat products (Smith et al., 2001). Regarding processed and cooked foods, some operations, such as smoke curing, cooking over charcoal and roasting, can significantly contribute to increasing the levels of PAH, while others, such as frying of vegetable oils, contribute only to a slight increase of these compounds (Purcaro et al., 2006). In opposition, the refining process of vegetable oils can lead to a decrease in PAH content in the final product (Cejpek et al., 1998; Teixeira et al., 2007). To evaluate PAH occurrence in foods consumed in the EU Member States, experts participating in the SCOOP (Scientific Cooperation) task collected data on the occurrence of PAH in foods and identified 44 food groups, from which five comprised more than 80% of total studied samples (8861): sausages and ham (27%), vegetable oils (24%), fish/fish products (13%), waters (excluding tap water) (11%), and meat (6%) (European Commission, 2004). The only consistently tested PAH (in 99% of the samples) was benzo[a]pyrene. The highest average levels for BaP found in foods were: 48.1 μg kg⫺1 (wet weight) in dried fruits, 17.1 μg kg⫺1 in olive pomace, 5.28 μg kg⫺1 in smoked fish, 4.2 μg kg⫺1 in grape seed oil, 3.27 μg kg⫺1 in smoked meat products, 3.09 μg kg⫺1 in fresh molluscs and 2.16 μg kg⫺1 in spices and condiments. Since diet is considered to be the major non-occupational source of PAH for non-smokers (Lodovici et al., 1995), several studies have been carried out to determine the level of intake associated with a normal or average human diet. Meat and meat products, cereals, and oils and fats have been suggested to be PAH main sources in the diet (Dennis et al., 1991). However, due to the numerous differences among diets, the levels and sources of PAH can be quite different (Phillips, 1999). The estimated average BaP intakes for a European adult range from 14 to 320 ng person⫺1 day⫺1 among the 11 states that provided intake data in the SCOOP task (European Commission, 2004).
54.3 OCCURRENCE OF PAH IN OLIVE OILS Virgin olive oil (VOO) is extracted from the olive fruit exclusively by mechanical processes without any further
SECTION | I Toxicology and Contaminants
treatment. Generally, the process comprises a series of steps including olive harvesting (manually or mechanically), transportation to olive mills, washing, crushing, mixing the olive paste in a thermobeater (with or without talc addition) and oil separation by centrifugation or pressing. Although VOO should be naturally free of PAH, contamination can occur either directly during the processing in the mill or indirectly due to olive skin contamination by environmental sources (Fromberg et al., 2007; RodriguézAcuña et al., 2008). In this last situation, PAH present in dust and particles from smoke and air pollution can contaminate olives via atmospheric fallout and this superficial contamination can be transferred to the final product (Rodriguéz-Acuña et al., 2008).
54.3.1 Sources of Contamination of Virgin Olive Oil Aiming to identify and evaluate the sources of PAH contamination during the processing of VOO, Rodriguéz-Acuña et al. (2008) studied the influence of factors such as the environmental pollution during olive growth, contamination during olive harvesting, contamination during extraction process and environmental pollution at the olive mill site. The authors identified nine PAH (benz[a]anthracene (BaA), chrysene (CHR), benzo[e]pyrene (BeP), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF), BaP, dibenz[a,h]anthracene (DhA), benzo[ghi]perylene (BgP), indeno[1,2,3-cd]pyrene (IcP)) and, comparing the total PAH concentration in olive oils with olive fruit surface extracted with hexane, they found identical values for both cases. This finding suggests that the contamination of olive oil is mainly from the olive skin. Rodriguéz-Acuña et al. (2008), when establishing the influence of different levels of environmental pollution during olive growth, showed that total PAH content in olives and, consequently, in respective oil, is related to the level of air pollution in the vicinity of the olive grove. The same authors compared mechanical and handpicked harvesting, concluding that exposure to diesel exhaust fumes from the combine is the most important source of olive skin contamination, since the highest values of PAH were found in the olives harvested mechanically. The influence of the olive washing step, micronized talc (hydrated magnesium silicate) addition during oil extraction, and the environmental pollution at the mill site, were also assessed (Rodriguéz-Acuña et al., 2008). The first two factors had no influence on PAH content of VOO, while the latter depends on several other issues, such as tank cleaning, installation of valves in the ventilation shafts to avoid intake of air pollution and the possibility of burning waste olive pomace in the facilities. According to the SCOOP task report of the European Commission (2004), from the 671 virgin and extra virgin
CHAPTER | 54 Polycyclic Aromatic Hydrocarbons (PAH) in Olive Oils and Other Vegetable Oils
olive oils (EVOO) analyzed, only 14 presented BaP levels above the maximum imposed by the EU (2 μg kg⫺1), from which two samples presented levels between 5–20 μg kg⫺1 and one was above 20 μg kg⫺1. The resumed published results of PAH content in olive oil during recent years are presented in Table 54.2. The majority of analyzed samples presented BaP levels below the maximum imposed by the EU; however, the reported results exceeding the limit reinforce the need for prevention and monitoring.
54.3.2 Sources of Contamination of Olive Pomace Oil During olive oil production, other low-quality oils, such as olive pomace oil (OPO), are sometimes produced. In OPO production, the dregs of crushed olives are dried and then extracted with organic solvents. For direct human consumption, this oil needs a refining process to remove unwanted minor components and undesirable organoleptic properties. Along OPO production, PAH contamination can occur during the drying and solvent extraction steps (León-Camacho et al., 2003). Sometimes, the olive pomace is dried by direct contact with combustion fumes. In such cases, the extent of PAH contamination is related to the type of fuel used and the exposure time necessary to eliminate water (Moreda et al., 2004). Thus, PAH content depends highly on the conditions used prior to the oil extraction process, and can be relatively high especially if harsh conditions are used. Nevertheless, PAH content is generally reduced during refining, either by the bleaching step, where the use of activated carbon and clay is recommended to remove the heaviest PAH, either by the deodorizing stage, where light PAH can be reduced together with other compounds, such as carotenoid pigments. The efficacy of the refining process can depend on the quality of the initial crude material, i.e., of the initial levels of PAH in the unrefined oil. A detailed study on the efficiency of the bleaching stage for the elimination of BaP in OPO was performed by León-Camacho et al. (2003). These authors reported a slight reduction of BaP to values above the legislation limits (2 μg kg⫺1) using earths in the bleaching stage, making necessary the use of active carbon in this step. Different procedures used along the refining process can possibly explain some differences in the values found in OPO (Table 54.3). In fact, for a group of ten refined OPO samples, lower levels of PAH were reported in the five samples submitted to decolorization during refining (Ballesteros et al., 2006).
54.4 OCCURRENCE OF PAH IN OTHER VEGETABLE OILS Due to their lipophilic nature, PAH can easily contaminate oils and fats, which are a significant dietary source, either
493
directly or indirectly by their incorporation into other foods such as cereal-based products (Dennis et al., 1991). Two main routes of PAH contamination in vegetable oils have been suggested, namely the contact of seeds with polluted surroundings and the drying process of oil seeds prior to oil extraction, by direct contact with combustion gases (Rodriguéz-Acuña et al., 2008). Another reported form of contamination may arise from direct migration of PAH to the oil seeds from jute bags treated with mineral oils, which are used for raw material storing and transporting (Gfrerer and Lankmayr, 2003). Recently, the EU has set maximum levels of 2.0 μg kg⫺1 for BaP in oils and fats intended for direct consumption or used as an ingredient in foods (European Commission, 2005a, 2006). Prior to that, some countries (Spain, Italy, Portugal and Greece) have established limits for the concentration of the following eight heavy PAH: BaA, BeP, BbF, BkF, BaP, DhA, BgP, IcP. The values established were a maximum limit of 2 μg kg⫺1 for each single PAH and 5 μg kg⫺1 for the sum of the referred eight heavy PAH (Teixeira et al., 2007). Some organizations establish their own recommendations, such as the German Society for Fat Science (GSFS), which suggests that total PAH in edible oils should not surpass 25 μg kg⫺1 and heavy PAH should be below 5 μg kg⫺1 (Cejpek et al., 1998). A high number of PAH with a wide range of molecular weights can be found in vegetable oils, of which many are alkylated compounds, although they are ignored by legal regulations (Guillén et al., 2004).
54.4.1 Influence of Refining PAH contamination in crude edible oils varies widely, but refined vegetable oils generally present lower levels than the crude oils, which can be attributed, at least in part, to the reduction observed through refining (Cejpek et al., 1998; Teixeira et al., 2007). The influence of different steps during the refining process (neutralization, bleaching and deodorization) on raw soybean and sunflower oils PAH content was evaluated by Teixeira et al. (2007). The authors observed an evident decrease of PAH content, especially light PAH. After refining, total PAH decreased 72% for sunflower oil and 87% for soybean oil. In both cases, the decrease of light PAH (71% and 88% for sunflower and soybean oils, respectively) was significantly higher than the decrease of heavy PAH (79% and 49% for sunflower and soybean oils, respectively) (Table 54.4). Regarding the different steps along refining, deodorization seems to have higher impact on decreasing total PAH levels, which agrees with other works (Cejpek et al., 1998). Moreover, the deodorization process seems to have little effect on heavy PAH removing mainly light PAH, while higher condensed heavy PAH are mainly removed by activated charcoal treatment (Dennis et al., 1991; Teixeira et al., 2007). The kind
494
TABLE 54.3 PAH content (μg kg⫺1) in olive oils and olive pomace oils. This table summarizes the PAH content in several olive oil types reported by different authors. Samplea
No. samples
Origin
BaP
Light PAH
Heavy PAH
Total PAH
Genotoxic PAH (average)
Reference
Italy
*
*
*
12.0–143.1b
*
Moret et al. (1997)
Olive oils EVOO
35
EVOO
9
Croatia
*
*
*
2.95–35.94b
*
Moret et al. (1997)
EVOO
5
Italian market
*
*
*
9.3–50.08b
*
Moret et al. (1997)
EVOO
54
Italy
*
38–260
*
*
*
Vichi et al. (2007)
1.6 (average)
16 (average)d
2.7e
Fromberg et al. (2007)
80 (average) EVOO
46
IT, SP, GR, FR, NLc
VOO
6
Spain (locality near airport)
0.5–1.3f
*
Rodríguez-Acuña et al. (2008)
VOO
2
Spain (urban locality)
0.5–1.0f
*
Rodríguez-Acuña et al. (2008)
VOO
2
Portuguese market
0.07–0.28
16.67–24.91
1.33–1.72
16.67–26.63g
*
Teixeira et al. (2007)
VOO
671
EU Countries
0.015–32
*
*
*
⬍0.2–0.4
15 (average)
Fromberg et al. (2007)
0.4 (average)
OO
7
European packed at origin
nd–1.2
*
*
*
*
Pupin and Toledo (1996)
OO
15
European packed at Brazil
nd–9.7
*
*
*
*
Pupin and Toledo (1996)
OO
10
Argentina
nd–164.4
*
*
*
*
Pupin and Toledo (1996)
OO
6
IT, SP
⬍0.2–0.2
6.8 (average)
1.3 (average)
8.1 (average)d
1.9e
Fromberg et al. (2007)
OO
280
EU Countries
0.03–89
*
*
*
*
Fromberg et al. (2007)
1.7 (average)
European Commission (2004)
SECTION | I Toxicology and Contaminants
European Commission (2004)
7
Brazilian market
2.2–9.2
*
*
*
*
Pupin and Toledo (1996)
Blend OO
2
Italian market
*
*
*
4.94–20.7b
*
Moret et al. (1997)
268
EU Countries
⬍0.1–206
*
*
*
*
Fromberg et al. (2007)
Olive pomace oils OPO
18 (average)
European Commission (2004)
OPO
3
Spanish refining industries
nd–0.34
*
*
0.0–3.2h
*
Moreda et al. (2004)
OPO
5
Spanish market
0.35–92.71
*
*
280.35–3199.79i
*
Guillén et al. (2004)
OPO
10
Spain
0.5–49.3
*
*
*
*
Ballesteros et al. (2006)
19.2 (average) OPO a
25
Spanish producers
⬍0.09–6.2
1.6–24.5j
Martinez-Lopez et al. (2005)
EVOO: extra virgin olive oil, VOO: virgin olive oil, OO: olive oil, OPO: olive pomace oil. Total of 13 PAH: FLR, PHE, ANT, FLT, PYR, BaA, CHR, BbF, BkF, BaP, DhA, BgP, IcP. IT: Italy, SP: Spain, GR: Greece, FR: France, NL: Netherlands. d Total of 17 PAH: ACY, ACP, FLR, ANT, FLT, PYR, BaA, CHR, BbF, perylene, BjF, BkF, BeP, BaP, IcP, DhA, BgP. e Sum of BaA, CHR, benzo[b ⫹ j]fluoranthene, BkF, BaP, IcP, DhA, and BgP. f BaA, CHR, BeP, BbF, BkF, BaP, DhA, BgP and IcP. g Sum of NAP, ACP, FLR, PHE, ANT, FLT, PYR, BaA, CHR, BbF, BkF, BaP, DhA, BgP, IcP. h Total of 9 PAH: BaA, CHR, BeP, BbF, BkF, BaP, DhA, BgP, IcP. i Sum of PAH and alkylderivatives: NAP, methyl-NAP isomers, dimethyl-NAP isomers, ethyl-NAP, ACY, ACP, fluorine, PHE, ANT, methyl-PHE isomers, dimethyl-PHE isomers, o-terphenyl, FLT, PYR, methyl-FLT, 11H-benzo[b]fluorine, 11H-benzo[c]fluorine, m-terphenyl, p-terphenyl, BaA, methyl-BaA isomers, dimethyl-BaA isomers,CHR, CHR isomers, methyl-CHR isomers, triphenylene, BbF, benzo[jk]fluoranthenes, benzo[a]fluoranthene, BeP, BaP, methylbenzopyrene or isomer, perylene, IcP, dibenz[a,h or a,c]anthracene, benzo[b]chrysene, picene, BgP, anthanthrene, coronene, dibenzopyrene and isomers, dibenzo[a,e]pyrene. j Sum of BaA, benzo[e]pyrene, BbF, BkF, BaP, DhA, BgP, IcP. b
c
CHAPTER | 54 Polycyclic Aromatic Hydrocarbons (PAH) in Olive Oils and Other Vegetable Oils
Blend OO
495
496
SECTION | I Toxicology and Contaminants
TABLE 54.4 PAH content (μg kg⫺1) during vegetable oil refining (adapted from Teixeira et al., 2007). Sunflower oil Crude
Neutralized
15.59
11.06
7.50
1.77
0.73
Total PAH
17.36
Total % of reduction
–
Light PAH Heavy PAH
Bleached
Soybean oil Deodorized
Crude
Neutralized
Bleached
Deodorized
4.53
63.59
42.56
43.45
7.79
0.46
0.37
1.74
1.60
1.25
0.89
11.80
7.96
4.90
65.33
44.16
44.71
8.67
32.0%
54.1%
71.8%
–
32.4%
31.5%
86.7%
of treatment used during the bleaching step seems to be of major importance. In some works, an increase in light PAH content was observed after bleaching, as associated with the use of contaminated clay (Cejpek et al., 1998; Teixeira et al., 2007). On the other hand, a greater reduction in heavy PAH content can be achieved when using activated charcoal compared to activated earth or clay (Teixeira et al., 2007).
54.4.2 PAH Content in Refined Vegetable Oils Generally, commercial refined vegetable oils show lower levels of PAH contamination when compared to olive oils. Van der Wielen et al. (2006) reported BaP concentrations up to 85 μg kg⫺1 in olive oils (n ⫽ 170), while for other edible vegetable oils (n ⫽ 170) they found values up to 9 μg kg⫺1. Teixeira et al. (2007) reported total PAH concentrations slightly higher for olive oil (18.0 to 26.3 μg kg⫺1, n ⫽ 2) when compared to soybean oil (9.3 to 10.8 μg kg⫺1, n ⫽ 3) and sunflower oil (8.8 to 9.7 μg kg⫺1, n ⫽ 3). The same authors reported a predominance of light PAH in all evaluated samples, which is in accordance with other works (Dennis et al., 1991; Lodovici et al., 1995). Conversely, the contribution of the more dangerous heavy PAH seems to show much lower significance. Nevertheless, there are reports on vegetable oils exceeding the maximum level admitted for BaP, although in other works the levels of this carcinogenic compound were found to be smaller than the limit imposed by the EU (2 μg kg⫺1) (Teixeira et al., 2007). For example, Dennis et al. (1991) reported BaP concentrations ranging from 1.4 to 64 μg kg⫺1 in rapeseed oil and Moret and Conte (2000) reported concentrations of the same compound ranging from 8.6 to 44 μg kg⫺1 in 20 grape seed oils evaluated. In the SCOOP task of the EU (European Commission, 2004), sunflower oils collected in seven EU Member States revealed an average value of 3.2 μg kg⫺1 for BaP, with 29% of all samples
presenting levels above the maximum limit established for this compound.
54.5 CARCINOGENESIS OF PAH PAH are themselves chemically inert and hydrophobic. However, as with other carcinogens, metabolic activation in an organism is required, which leads to the formation of electrophilic metabolites capable of binding to nucleophilic centers in DNA. PAH undergo metabolic activation in mammalian cells to diol-epoxides that bind covalently to cellular macromolecules, including DNA, thereby causing errors in DNA replication and mutations that initiate the carcinogenic process (Phillips, 1999). This mechanism of activation, with modifications in some cases, has been found to occur with all carcinogenic PAH studied (Phillips, 1999). When BaP is oxidized and hydroxylated it involves cytochrome P450 isoenzyme and epoxide hydrolase and is converted to epoxides, phenols, diols, tetrols and quinine derivatives (IARC, 1987). An initial step of tumorigenesis involves the metabolic conversion to compounds that may react covalently with DNA to yield DNA adducts. If those adducts are not repaired or misrepaired, they may initiate gene mutations and lead to adverse health effects in humans. This could be a biomarker of biological effective dose of BaP. Measuring the carcinogen–DNA adducts is thought to be a useful biomarker in molecular epidemiological studies that attempt to determine cancer risk (Phillips, 2005). Several studies reported the link between DNA adducts and colon cancer originated from exposure to PAH (Al-Saleh et al., 2008). Exposition to PAH has been epidemiologically associated with an increased risk of skin and lung cancer (IARC, 1987). There is strong evidence that the diol epoxide mechanism operates in the mouse lung tumorigenesis of many PAH evaluated. For some PAH, there is strong evidence
CHAPTER | 54 Polycyclic Aromatic Hydrocarbons (PAH) in Olive Oils and Other Vegetable Oils
that both radical cation and diol-epoxide mechanisms induce mouse skin carcinogenesis. Many of the pathways that lead to PAH carcinogenesis involve genotoxicity (IARC, 2006). Although studies in experimental animals on individual PAH, most notably BaP, have shown various toxicological effects, such as hematological effects, reproductive and developmental toxicity and immunotoxicity, it is the carcinogenic and genotoxic potential of these compounds that has attracted most attention. A number of PAH as well as coal-tar and some occupational exposures to combustion emissions containing these compounds have shown carcinogenicity in experimental animals and genotoxicity and mutagenicity in vitro and in vivo (IARC, 1987). As diet is one of the main sources of human and animal background exposure to PAH, the epithelial intestinal cells, the first to be in contact with contaminants, are of particular interest. In vivo studies suggest a transfer in intestinal epithelium by diffusion, which appears extensively governed by the physicochemical properties of PAH, particularly lipophilicity. However, a positive finding is that an intestinal transfer of PAH is not sufficient to characterize their availability and toxicity because intestinal metabolism can reshape the molecules, decreasing the bioavailability of contaminants (Cavret and Feidt, 2005).
SUMMARY POINTS ●
●
●
●
●
Polycyclic aromatic hydrocarbons (PAH) belong to a large class of organic compounds originated from incomplete combustion of organic matter, known or suspected to be carcinogenic and genotoxic compounds to mammals. Benzo[a]pyrene is the most studied PAH and is used as a marker. The main sources of PAH in foods are the environmental contamination and the food processing (such as smoke curing, cooking over charcoal and roasting). PAH in olive oil can occur due to environmental contamination of the fruit skin and in olive pomace oil due to the refining process. Refining process of vegetable oils may decrease the levels of PAH in the final products.
REFERENCES Al-Saleh, I., Arif, J. El-Doush, I., Al-Sanea, N., Jabbar, A. A., Billedo, G., Shinwari, N., Mashhour, A., Mohamed, G., 2008. Carcinogen DNA adducts and the risk of colon cancer: case-control study. Biomarkers 13, 201–216. Ballesteros, E., Sánchez, A.G., Martos, N.R., 2006. Simultaneous multidetermination of residues of pesticides and polycyclic aromatic hydrocarbons in olive and olive-pomace oils by gas chromatography/ tandem mass spectrometry. J. Chromatogr. A 1111, 89–96.
497
Cavret, S., Feidt, C., 2005. Intestinal metabolism of PAH: in vitro demonstration and study of its impact on PAH transfer through the intestinal epithelium. Environ. Res. 98, 22–32. Cejpek, K., Hajslova, J., Kocourek, V., Tomaniova, M., Cmolik, J., 1998. Changes in PAH levels during production of rapeseed oil. Food Addit. Contam. 15, 563–574. Dennis, M.J., Massey, R.C., Cripps, G., Venn, I., Howarth, N., Lee, G., 1991. Factors affecting the polycyclic aromatic hydrocarbon content of cereals, fats and other food products. Food Addit. Contam. 8, 517–530. European Comission, 2006. Commission Regulation (EC) No. 1881/2006 setting maximum levels of certain contaminants in foodstuffs. Off. J. Eur. Union L364/5-24. European Commission, 2002. Opinion of the Scientific Committee on Food on the risks to human health of Polycyclic Aromatic Hydrocarbons in food (http://ec.europa.eu/food/fs/sc/scf/out153_en.pdf). European Commission, 2004. Report of experts participating in Task 3.2.12. Collection of occurrence data on polycyclic aromatic hydrocarbons in food (http://ec.europa.eu/food/food/chemicalsafety/ contaminants/scoop_3-2-12_final_report_pah_en.pdf). European Commission, 2005a. Commission Regulation (EC) No. 208/2005 amending Regulation (EC) No. 466/2001 as regards polycyclic aromatic hydrocarbons. Off. J. Eur. Union. L 34/3-5. European Commission, 2005b. Commission Recommendation 2005/108/ EC on the investigation into the levels of polycyclic aromatic hydrocarbons in certain foods. Off. J. Eur. Union. L 34/43-45. Fromberg, A., Højgård, A., Duedahl-Olesen, L., 2007. Analysis of polycyclic aromatic hydrocarbons in vegetable oils combining gel permeation chromatography with solid-phase extraction clean-up. Food Addit. Contam. 24, 758–767. Gfrerer, M., Lankmayr, E., 2003. Microwave-assisted saponification for the determination of 16 polycyclic aromatic hydrocarbons from pumpkin seed oils. J. Sep. Sci. 26, 1230–1236. Guillén, M.D., Sopelana, P., Palencia, G., 2004. Polycyclic aromatic hydrocarbons and olive pomace oil. J. Agric. Food Chem. 52, 2123–2132. Guillén, M.D., Sopelana, P., Partearroyo, M.A., 1997. Food as a source of polycyclic aromatic carcinogens. Rev. Environ. Health 12, 133–146. IARC, 1987. Polynuclear Aromatic Compounds, Part 1: Chemical, environmental and experimental data in IARC Monographs on the evaluation of the carcinogenic risk of chemicals to humans, Vol. 32, Suppl. 7. International Agency for Research on Cancer: Lyon, France. IARC, 2006. Some Non-heterocyclic Polycyclic Aromatic Hydrocarbons and Some Related Industrial Exposures. Vol. 92 (http://monographs. iarc.fr/ENG/Meetings/92-pahs.pdf). León-Camacho, M., Viera-Alcaide, I., Ruiz-Méndez, M.V., 2003. Elimination of polycyclic aromatic hydrocarbons by bleaching of olive pomace oil. Eur. J. Lipid Sci. Technol. 105, 9–16. Lodovici, M., Dolara, P., Casalini, C., Ciappellano, S., Testolin, G., 1995. Polycyclic aromatic hydrocarbon contamination in the Italian diet. Food Addit. Contam. 12, 703–713. Martinez-Lopez, S., Morales-Noé, A., Pastor-Garcia, A., Morales-Rubio, A., Guardia, M., 2005. Sample preparation improvement in polycyclic aromatic hydrocarbons determination in olive oils by gel permeation chromatography and liquid chromatography with fluorescence detection. JAOAC Int. 88, 1247–1254. Moreda, W., Rodríguez-Acuña, R., Pérez-Camino, M.C., Cert, A., 2004. Determination of high molecular mass polycyclic aromatic hydrocarbons in refined olive pomace and other vegetable oils. J. Sci. Food Agric. 84, 1759–1764.
498
Moret, S., Piani, B., Bortolomeazzi, R., Contel., L.S., 1997. HPLC determination of polycyclic aromatic hydrocarbons in olive oils. Z. Lebensm. Unters. Forsch. A 205, 116–120. Moret, S., Conte, L.S., 2000. Polycyclic aromatic hydrocarbons in edible fats and oils: Occurrence and analytical methods. J. Chromatogr. A 882, 245–253. Phillips, D.H., 1999. Polycyclic aromatic hydrocarbons in the diet. Mutation Res. 443, 139–147. Phillips, D.H., 2005. DNA adducts as markers of exposure and risk. Mutation Res. 577, 284–292. Pupin, A.M., Toledo, M.C.F., 1996. Benzo(a)pyrene in olive oils on the Brazilian market. Food Chem. 55, 185–188. Purcaro, G., Navas, J.A., Guardiola, F., Conte, L.S., Moret, S., 2006. Polycyclic aromatic hydrocarbons in frying oils and snacks. J. Food Prot. 69, 199–204. Rodríguez-Acuña, R., Pérez-Camino, M.C., Cert, A., Moreda, W., 2008. Sources of contamination by polycyclic aromatic hydrocarbons in Spanish virgin olive oils. Food Addit. Contam. 25, 115–122.
SECTION | I Toxicology and Contaminants
Smith, C.J., Perfetti, T.A., Rumple, M.A., Rodgman, A., Doolittle, D.J., 2001. IARC group 2B carcinogen reported in cigarette mainstream smoke. Food Chem. Toxicol. 39, 183–205. Teixeira, V.H., Casal, S., Oliveira, M.B.P.P., 2007. PAHs content in sunflower, soybean and virgin olive oils: Evaluation in commercial samples and during refining process. Food Chem. 104, 106–112. Van der Wielen, J.C.A., Jansen, J.T.A., Martena, M.J., De Groot, H.N., In’t Veld, P.H., 2006. Determination of the level of benzo[a]pyrene in fatty foods and food supplements. Food Addit. Contam. 23, 709–714. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., López-Tamames, E., 2007. The ocurrence of volatile and semi-volatile aromatic hydrocarbons in virgin olive oils from north-eastern Italy. Food Control 18, 1204–1210. Wenzl, T., Simon, R., Klieiner, J., Anklam, E., 2006. Analytical methods for polycyclic aromatic hydrocarbons (PAHs) in food and the environment needed for new food legislation in the European Union. Trends Anal. Chem. 25, 716–725.
Chapter 55
Mineral Paraffins in Olives and Olive Oils Sabrina Moret, Tiziana Populin and Lanfranco S. Conte Department of Food Science, University of Udine, Italy
55.1 INTRODUCTION The term ‘mineral paraffins’ is usually used to indicate a complex mixture of branched and cyclic aliphatic hydrocarbons (in the range of C15–C45), whose presence in foods reveals a contamination with mineral oil. Foods may be contaminated with food-grade ‘white oils’ (from release agents, lubricating oils and food contact materials), but also with non-food-grade products containing variable amounts of aromatics and additives. Since analytical determination of mineral paraffins is less demanding than that of aromatics or additives, paraffins are usually used as a marker of the contamination in both cases. Mineral oil contamination is easily recognizable by the presence in the chromatographic trace of one or more ‘humps’ or ‘unresolved complex mixtures’ (UCMs) consisting of a large number of unresolved peaks, and/or n-alkanes with a balanced distribution between odd and even carbon number hydrocarbons, i.e. an odd-to-even ‘carbon preference index’ (CPI) close to 1. On the other side, vegetable oils, as well as all foods of plant and animal origin, contain also natural paraffins, primarily n-alkanes with a prevalence of odd carbon number hydrocarbons and olefins generated during oil refining, which must be distinguished or separated from hydrocarbons of mineral origin. Nevertheless, natural paraffins represent a useful
tool to characterize and differentiate virgin olive oils of different cultivars (Guinda et al., 1996; Koprivnjak et al., 1997, 2005). Wagner et al. (2001a), who analyzed more than 200 samples of vegetable oils, found concentrations of 30– 150 mg kg⫺1 of mineral paraffins in raw oils. After deodorization, which removes about two-thirds of the contamination (up to C25–C30), many commercial oils still contained 20– 80 mg kg⫺1. Most of the oils showed a Gaussian-like ‘hump’ centered at the C27–C29 n-alkanes, typical of lubricating oil; several oils presented more than one UCM, suggesting different sources of contamination. Only in a few cases it was possible to identify the origin of the contamination (technical oil used in the oil mill, batching oil from jute bags, diesel oil). In many instances the mineral paraffins were already in the seeds and had a composition resembling that in the atmospheric particulate matter, suggesting an environmental origin (Neukom et al., 2002). In this chapter, an overview of toxicity aspects of mineral paraffins, analytical methods for their determination, contamination levels found in olives and different types of olive oils, as well as possible sources of contamination, will be presented. Finally, a conclusive proof of the mineral identity of the contamination will be presented and discussed. Table 55.1 summarizes main features of mineral paraffins.
TABLE 55.1 Features of mineral paraffins. ●
Mineral paraffins are a complex mixture of branched and cyclic aliphatic hydrocarbons of petrogenic origin
●
The gas chromatographic trace of a contaminated sample usually shows an ‘unresolved complex mixture’ (UCM) forming a ‘hump’ and/or a balanced distribution between odd and even carbon number n-alkanes
●
A widespread contamination is observed in different food items and particularly in vegetable oils
●
Different sources of contamination (hydraulic and lubricating oils used in the extraction plant, atmospheric fall-out, vehicular exhausts, etc.) have been identified.
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
499
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
500
55.2 TOXICITY ASPECTS Adverse effects of paraffin oils were observed in rats at concentrations of 0.01–20 mg kg⫺1 body weight (bw), depending on the type of oil administered (SCF, 1995). Based on the assumption that paraffins, having a sufficiently high molecular mass, are not absorbed to a relevant extent, Scientific Committee on Food (SCF, 1995) assigned a group acceptable daily intake (ADI) of 0–20 mg kg⫺1 bw for microcrystalline waxes and a temporary group ADI of 0–4 mg kg⫺1 bw for mineral paraffin oils (for products characterized by a minimum viscosity, a maximum of 5% components below the n-alkane C25 and an average molecular mass of no less than 500 and 480 Da for paraffin waxes and mineral paraffin oils, respectively). In 2006, the European Food Safety Authority (EFSA, 2006), due to the lack of toxicity data, specified a restriction of 0.05 mg kg⫺1 food for ‘waxes, paraffinic, refined, derived from petroleum based or synthetic hydrocarbon feedstock’, with an average molecular weight of no less than 350 Da (about C32), a minimum viscosity, and not more than 40% of hydrocarbons below C25. A number of independent subchronic toxicity feeding studies reported the presence of granulomas in liver and histiocytosis in mesenteric lymph nodes of Fisher 344 rats, after ingestion of selected white mineral oils (Nash et al., 1996). Comparative studies with other rat strains indicated that Fisher 344 rats have an exceptionally high sensitivity to mineral paraffins. A limited amount of mineral paraffins (inversely related to the length of their carbon backbone) is absorbed after oral ingestion. According to Low et al. (1992), branched or cyclic hydrocarbons are absorbed more readily than straight chain compounds. Preferential accumulation of mineral hydrocarbons in the range C20–C35 was observed by Scotter et al. (2003) in all tissue analyzed (small intestine, heart, kidney and liver). Mineral paraffins probably represent the least toxic fraction present in mineral oils. Unless the contamination refers to a ‘white mineral oil’ (from which the aromatics are removed), substantial concentrations of additives and aromatics can be present. For example, high concentrations of carcinogenic polycyclic aromatic hydrocarbons are usually present in lubricating oils from engine exhaust (Wong and Wang, 2001; Moret et al., 2003). Lubricants contain up to 20% of additives, some of which are classified as dangerous by the European Community (Henry, 1998). A review on short-term effects of mineral oils, additives, and contaminants has been published by Hard (2000).
55.2.1 Bioaccumulation Particular concern arises from the evidence that mineral paraffins can be transferred through the food chain. According to Grob et al. (2001) hens transferred 1.5–3% of the mineral
SECTION | I Toxicology and Contaminants
paraffins (centered at C21–C24) added to the commercial feed into the eggs. Mineral paraffins were also found in body fat and human milk at concentrations exceeding those considered safe for foods by the EFSA by 2–3 orders of magnitude (Noti et al., 2003; Concin et al., 2008). Paraffins in body fat ranged from n-alkanes C17 to C32 and were centered at C23–C24. Since the mineral oil products we are exposed to range from much smaller to much higher molecular mass, the observed contamination must be the result of selective uptake, elimination by evaporation and metabolic degradation. Concentrations varied from 15 to 360 mg kg⫺1 fat (on average 60.7 mg kg⫺1). If food were the main source, exposure data would suggest the mineral paraffins are accumulated over a life-time. The average content of mineral paraffins in the milk fat on day 4 was 44.6 mg kg⫺1. It decreased with every day of breast feeding and was generally little above the detection limit on day 20 (Concin et al., 2008).
55.2.2 Analytical Methods for Detecting Mineral Paraffins Analysis of mineral paraffins (as well as natural n-alkanes) in edible oils involves a sample preparation step (enrichment/purification) aiming at isolating selectively mineral paraffins from triglycerides and other food components before gas chromatography (GC). This preparation step can be performed either off-line or on-line (coupled liquid chromatography (LC)-GC or LC-LC-GC). Because of the ubiquitous presence of mineral hydrocarbons in the environment, it is important to take some precautions in order to avoid spurious background contamination and avoid contact with plastic material. All glassware and apparatus must be accurately cleaned and the solvent used should be distilled or checked for the presence of impurities. A blank analysis should be included at regular intervals during sample analysis. Due to the large number of hydrocarbons with different alkylation degree present in vegetable oils contaminated with mineral oil residues, GC with flame ionization detection (FID) represents, without any doubt, the best choice for quantifying the contamination. Although it allows reaching lower sensitivities when compared to mass spectrometry (MS) detection, FID is the only detector able to provide approximately constant response for all hydrocarbons. Due to the lack of selectivity of FID, pre-separation methods must ensure that only paraffins enter the GC column. The sensitivity of the method depends on the amount of food components which interfere and the distribution of mineral paraffins. A broad ‘hump’, resulting from a mixture of different mineral oil products with a wide molecular weight distribution, is more difficult to detect and quantify (Wagner et al., 2001b). Quantitative data are usually obtained by the internal standard method. The area of the ‘hump’ can be determined manually by approximation with simple geometrical
CHAPTER | 55 Mineral Paraffins in Olives and Olive Oils
forms and compared with the area of the internal standard (usually C13). The position of the baseline needs to be assessed by blank runs performed on the same day. Natural n-alkanes present on the top of the ‘hump’ are determined through area counts and subtracted from the total area of the ‘hump’. When sample components interfere with the internal standard (as in the case of contamination with diesel oil), the external standard approach can be used to assess the contamination.
55.2.3 Off-line Techniques Several authors reported examples of off-line methods for GC determination of mineral paraffins and natural occurring n-alkanes. Traditional methods for hydrocarbon determination in edible oils involve a saponification step with strong alkali (to remove the bulk of triglycerides), followed by column chromatography or solid phase extraction (Guinda et al., 1996; Koprivnjak et al., 1997). Methods based on saponification allow to process high amounts of oil reaching high sensitivity, but are tedious, time-consuming, and liable to losses of the more volatile hydrocarbons during sample preparation. Nevertheless, hydrocarbons can be separated easily from other oil constituents by column chromatography on silica gel using n-hexane as the eluent (Tan and Kuntom, 1993). Paraffins are eluted at the dead time of the column, before other oil components and aromatics. Of course, the sensitivity of the methods is limited by the maximum amount of oil that can be loaded on the packed column. McGill et al. (1993) applied a two-step purification before GC analysis. First the oil (100 mg) was passed on a packed silica column (25 g) to retain triglycerides. Then the paraffin fraction was eluted with hexane, re-concentrated and isolated from aromatics on a LC silica column (25 ⫻ 0.46 cm). Since large amounts of olefins are formed during oil refining from isomerization of squalene and dehydration of sterols, it is important to ensure that only saturated hydrocarbons enter the GC column in order to have a correct quantification of the contamination. To this purpose, Wagner et al. (2001b) proposed a method involving the bromination of the oil, a passage through a small column of highly active aluminum oxide able to retain triglycerides and brominated olefins, and large-volume on-column injection into GC-FID. Due to the high volatility of some hydrocarbons, off-line methods usually demand the on-column injection technique. GC separation is carried out on nonpolar columns (methyl silicone).
55.2.4 On-line Coupled Techniques Coupled LC-GC techniques are completely automated, highly reproducible, allow minimizing sample manipulation
501
and transferring the entire LC fraction to GC (thus improving the detection limit). Manual sample preparation requires less than five minutes per sample, and 20 analyses per day can be carried out (using an autosampler). Both ‘loop-type interface’ with ‘concurrent eluent evaporation’ (Grob et al., 1991a,b,c) and ‘on-column interface’ with ‘partially concurrent eluent evaporation’ (Wagner et al., 2001a) were applied by Grob and collaborators to analyze mineral paraffins in vegetable oils or lipid extracts. To increase sensitivity, Grob and Bronz (1995) proposed a modified interface, involving LC transfer through the so-called ‘wire-interface’. Evaporation occurred concurrently in a heated capillary into which a piece of wire was introduced, and oven temperature was adjusted to the minimum preventing re-condensation. The retention of volatiles was improved by making better use of the ‘phasesoaking effect’ in the pre-column (involving temporary increase of retention power by swelling of the stationary phase with solvent). This allowed transferring larger eluent fractions (600 μL) with good recoveries of volatiles (complete recovery of C12). This interface was later used for LC-LC-GC determination of mineral oil hydrocarbons in the oil phase of canned sea foods (Grob et al., 1997). Two-dimensional LC (two silica columns, 250 ⫻ 0.2 cm i.d.) turned out particularly useful in preventing olefins, originated during oil refining, from reaching the GC column, thus complicating the correct quantification of the contamination. Hydrocarbons were isolated from the usual 20 mg oil injected on a first silica column. After their elution (300 μL), the column was backflushed with dichloromethane, while mineral paraffins were separated (from unsaturated hydrocarbons) on a second silica column and transferred into the GC (Wagner et al., 2001a). As a precautionary measure, a bromination step can be introduced before injection in order to enhance olefin retention on the silica column. LC-GC conditions used by Moret et al. (2003) and Koprivnjak et al. (2005) for olive oil analysis involved the injection of 60 mg of oil sample into a large silica column (100 ⫻ 4.6 mm i.d.). After the elution of the fraction of interest (600 μL), the column was backflushed with dichloromethane to remove the residual fat, while GC transfer occurred through the wire-interface described by Grob and Bronz (1995). As already discussed, mineral paraffins do not account for the overall toxicity of mineral oil. Moret et al. (1996) realized an automated on-line LC-LC-GC method for assessing both mineral paraffins and aromatics. It involved oil injection (150–250 mg) into a silica column, on-line solvent evaporation of the LC fraction, separation of aromatics on an aminosilane column (according to their ring number), and transfer to GC through the wire-interface. The ratio between mineral paraffins and total aromatics (obtained by backflushing the second column after the elution of paraffins) can help to identify the source of contamination.
502
SECTION | I Toxicology and Contaminants
55.2.5 Content of Mineral Paraffin in Olives and Different Classes of Olive Oils Even though different authors reported occasional contamination of olive oils with mineral paraffins during analysis of natural n-alkanes, to the best of our knowledge, only one work focused on olive and olive oil contamination. Table 55.2 summarizes results reported in this work by Moret et al. (2003) and other unpublished results from the same authors on mineral paraffin contents in different classes of olive oils.
55.2.6 Olives Moret et al. (2003) analyzed 27 samples of olives collected in two groves in Italy, one contiguous to a road with heavy traffic and the other in a rural location. The samples were dried, and oil extracted with solvent (hexane) before LC-GC determination of the paraffins. None of the samples contained mineral paraffins above the detection limit (1–2 mg kg⫺1 referring to the extracted oil). As all the plant material investigated contained detectable amounts of mineral paraffins (Neukom et al., 2002), and the fresh pomace contained about 10–20 mg kg⫺1 of mineral paraffins, it was hypothesized that their extraction from fresh olives is difficult and eased only after aging/partial digestion of the material. Only one of the five olive samples collected at five different oil mills resulted in contamination with an exceptional high amount of mineral paraffins (49 mg kg⫺1) forming a symmetric hump ranging from C21 to C33, and centered on C25. It is interesting to note that oil obtained by physical means (centrifugation) from these contaminated olives had a considerably lower content of mineral paraffins
(18 mg kg⫺1). It was calculated that less than 25% of the paraffins present in the olives were transferred into the virgin oil. This provides insight into the extractability of the paraffins. The mineral paraffins must be firmly included in solids poorly accessible by oil. Poor extraction by oil was further confirmed by the natural n-alkanes present at lower concentration in oil extracted by physical means with respect to the oil extracted with a solvent (Moret et al., 2003).
55.2.7 Virgin Olive Oils A number of authors investigated the natural n-alkane distribution in virgin olive oils, and only in a few cases the presence of an UCM, index of mineral oil contamination, was evidenced. Figure 55.1 shows the GC trace of a noncontaminated sample. During a survey on the presence of various contaminants in extra virgin olive oils, 46 samples taken directly from the oil mills (in six different Italian regions) were analyzed by the authors for the presence of mineral paraffins (unpublished data, 2005). Four samples were contaminated with mineral paraffins at amounts ranging from 25–120 mg kg⫺1. Figure 55.2 shows the GC traces of two contaminated samples. Trace A shows a ‘hump’ of paraffins centered on C31 from an unknown source, while trace B shows a large ‘hump’ of low molecular mass hydrocarbons centered on C17 with a balanced presence of odd- and even-numbered n-alkanes, typical of diesel oil. Natural nalkanes characterized by a prevalence of odd-numbered hydrocarbons are visible in both traces. The contamination, already present in the olives, was probably due to a diesel oil storage tank, present in the vicinity of the mill site.
TABLE 55.2 Mineral paraffin content (mg kg⫺1) found in different types of olive oils. Number of samples analyzed
Contaminated samples (%)
mg kg⫺1 mineral paraffins average (min–max)
Extra virgin olive oil
73
10
Olive oila
13
100
14 (6–30)
7
100
145 (121–250)
Olivepomace oila a
Based on Moret et al. (2003).
4 (1–120)
FIGURE 55.1 n-Alkane profile of a non-contaminated extra virgin olive oil. Natural n-alkanes are recognizable by the typical prevalence of oddnumbered terms.
503
CHAPTER | 55 Mineral Paraffins in Olives and Olive Oils
Only one of the 22 extra virgin olive oils (from both the market and the oil mills) analyzed by Moret et al. (2003) contained mineral paraffins (11 mg kg⫺1) above the detection limit (1–2 mg kg⫺1). The cause of the contamination was a leak in the press piston which contaminated the olive oil with lubricating oil (hump centered on C29). On the other side, the six lampante oils collected at the oil mills were also clean, while the re-extracted olive-pomace oils (three samples) contained 16–145 mg kg⫺1 of mineral paraffins. In order to dispose of this pomace, it must undergo an additional centrifugation step (performed up to several weeks later) and the virgin oil obtained in this way is classified as lampante or crude olive-pomace oil, depending on its chemical characteristics. Probably either the pomace was contaminated during storage, or the prolonged contact of residual oil with the partially aged/degraded pomace was responsible for a more efficient extraction of mineral paraffins which were not accessible in the first extraction.
55.2.8 Olive Oils Commercial olive oils (13 samples) were contaminated with amounts of mineral paraffins ranging between 6 and 30 mg kg⫺1. The ‘hump’ of mineral paraffins generally extended from C21 to beyond C35 and was centered at the C27–C29 n-alkanes, resembling the typical distribution of
paraffins in a motor oil (Moret et al., 2003). Figure 55.3 shows the LC-GC trace of an olive oil sample.
55.2.9 Olive-pomace Oils Very high concentrations of mineral paraffins (typically 100–300 mg kg⫺1) were found in olive-pomace oils collected from both the market and the refining plants. The data obtained from the samples collected in two pomace oil plants demonstrated that the contamination was already present in the pomace to be extracted and that there was no significant increase of the paraffins during the processing line. On the contrary, deodorization removed the more volatile part of the contamination (e.g. residues from diesel). As discussed above, most of the paraffins contained in the olives remain in the pomace and are more efficiently extracted with solvent after partial aging/degradation. There is also an important re-concentration effect, as the bulk of the paraffins are transferred into a small amount of oil (1–2% referring to the olive). In any case, it has been demonstrated that the concentration of mineral paraffins in pomace increases during storage of the pomace at the oil mill and/or the extraction plant (from about 15–25 mg kg⫺1 in fresh pomace to 100–400 mg kg⫺1 after 20–30 days, referring to the extracted oil), depending on the storage conditions (Moret et al., 2003). Figure 55.4 shows the effect of storage on pomace contamination, while Figure 55.5 shows the mineral paraffins of an olive-pomace oil from the market.
55.2.10 Sources of Contamination Even though olives and virgin olive oils rarely present contamination levels above the detection limit, the environment
FIGURE 55.2 Mineral paraffins in extra virgin olive oil. LC-GC traces of two contaminated samples: (A) contamination from unknown source (50 mg kg⫺1), (B) contamination with diesel oil (120 mg kg⫺1).
FIGURE 55.3 Mineral paraffins in olive oil. LC-GC trace of a contaminated sample from the market (22 mg kg⫺1).
504
FIGURE 55.4 Effect of storage on mineral paraffin content. LC-GC traces of: (A) newly delivered pomace with low contamination level, (B) pomace after a 3-week storage at the extraction plant (high contamination level).
SECTION | I Toxicology and Contaminants
the source of contamination to be established, but in most cases the source remains unknown. According to Moret et al. (2003), olive oil contamination probably occurs during transport of the lampante oil from the oil mill to the refining plant or fraudulent admixture of olive-pomace oil. The samples collected at different refining stages demonstrated that refining does not cause contamination with mineral paraffins. Also the solvent used for oil extraction was excluded from being the source of the contamination. The high amounts of mineral paraffins often found in pomace stored for a long time and in olive-pomace oil can be explained considering how the pomace is stored and handled. After the extraction of the virgin olive oil the pomace is typically removed by a conveyor belt (or a cochlea) and stored, for several weeks, outdoors (on asphalted squares often used as a parking area) until it is transported to the pomace oil plant. Pomace has no value for the oil mill and owners are glad to get rid of it free of charge. At the olive pomace plant it is stored again for a widely varying duration. Bulldozers used to move the pomace are often old and worn-out, and can spill hydraulic oil directly onto the pomace. It has been calculated that 1 g of hydraulic oil for every m3 of pomace is sufficient to generate a contamination of about 100 mg kg⫺1 in the pomace oil. Furthermore, the pomace is exposed to the exhaust of vehicles moving about the lot, which contains motor oil or incompletely combusted diesel fuel. These hypotheses are supported by the fact that the composition of mineral paraffins of stored pomace usually resembles that of motor oil, hydraulic oil and incompletely combusted diesel fuel. Atmospheric fall-out, lubricating oil used in the installations and asphalt debris may be other sources of contamination. Further investigations are required to verify if the high contamination levels found in olive-pomace oils primarily results from particulate matter deposition or incorrect practices of pomace storage (Moret et al., 2003).
55.3 CONFIRMING THE MINERAL ORIGIN OF PARAFFINS
FIGURE 55.5 Mineral paraffins in olive-pomace oil. LC-GC trace of a contaminated sample from the market (183 mg kg⫺1).
certainly represents a potential source of contamination. Accidental contact with mineral oil products (hydraulic and lubricating oils used in the extraction plant) can be responsible for high contamination levels. The molecular weight distribution of hydrocarbons forming an UCM may allow
Different issues, such as the presence of one or more ‘humps’ of highly isomerized paraffins with a Gaussiantype distribution (typical of products obtained by distillative fractionation), a CPI close to 1, and the presence of marker compounds such as isoprenoids (pristane and phytane), have been widely used to establish the mineral origin of an UCM. However, information about n-alkanes is sometimes lost due to weathering or because the contaminating mineral oil is deparaffinated. Furthermore, when the selective pre-separation fails, vegetable oils may contain UCMs of hydrocarbons originating from the raffination process, such as isomerized squalene or dehydrated sterols (sterenes).
505
CHAPTER | 55 Mineral Paraffins in Olives and Olive Oils
Moret et al. (2003) demonstrated that the presence of UCM material in olive pomace does not derive from bioconversion of organic matter (squalene, natural alkanes), but confirmed an environmental contamination. Pomace and olive paste stored for one month in a site far from contamination sources contained the same mineral paraffins as before, while another sample stored for one month in a window sill overlooking a parking area showed a significant increase in paraffins centered on C17, typical of incompletely combusted diesel oil. A conclusive proof of the mineral origin of the UCM in olive-pomace oils and other contaminated samples was provided in 2004 by Populin et al., who analyzed the triterpanes (hopanes), diagenetic products (derived from hopanoids) formed during geological times at the elevated temperatures reached during deep sediment burial. Since hopanes are persistent in the environment, they can be used as markers of petroleum pollution (Wang et al., 2001; Farrimond et al., 2002). Hopane analysis was performed by injecting the relevant LC fraction on an UltraTrace GC instrument equipped for large-volume on-column injection. The MS ionization was by electron impact; the MS detector was used in selected ion monitoring (SIM) mode at m/z 81 (for paraffins) and m/z 191 (for hopanes). The presence of the whole range of hopanes in the samples with UCMs analyzed by Populin et al. (2004) confirmed contamination with mineral oil, but it did not demonstrate that most or all of the UCM material was of mineral origin. To overcome this problem, the relative hopane content (RHC) was introduced, i.e. the area ratio of the sum of the hopanes and the paraffins (except n-alkanes) in the same segment of the UCM (comprised in the region from Kovacs Index 2860 to 3650). RHC values determined for a range of mineral oil products (motor oils, hydraulic oils, lubricating oils, vaseline, batching oil) possibly encountered in foodstuffs resulted fairly constant (on average 3.4%) and consistent with those found in contaminated samples, thus confirming the mineral origin of the UCM. On the basis of its calculation mode, RHC confirms the mineral origin only for the fraction of hump that contains the hopanes but, because of the homogeneous distribution in molecular weights of the UCM, it enables the same conclusion, with high probability, for the total material that formed it (it is sufficient for a small part of the hump to be in the relevant segment).
SUMMARY POINTS ●
●
●
Mineral paraffins are used as markers of contamination with mineral oils. Toxic effects are observed in Fisher 344 rats at concentrations of 0.01–20 mg kg⫺1 bw, depending on the type of paraffins administered. Concerns arise about possible presence (in contaminated foods) of toxic aromatics and additives and for
●
●
●
●
the ability of mineral paraffins to bioaccumulate also in human tissues. Olives and olive oils generally contain not detectable amounts of mineral paraffins, while all the olive oils analyzed were contaminated (6–30 mg kg⫺1 of mineral paraffins). Contamination levels exceeding 100 mg kg⫺1 were found in olive-pomace oils. Atmospheric fall-out and incorrect practices of pomace storage have been suggested to be the main sources of contamination. A conclusive proof of the mineral origin of the contamination was provided through hopane analysis.
ACKNOWLEDGMENT The authors wish to thank Dobrila Brausntein for the language revision and support.
REFERENCES Concin, N., Hofstetter, G., Plattner, B., Tomovski, C., Fiselier, K., Gerritzen, K., Fessler, S., Windbichler, G., Zeimet, A., Ulmer, H., Siegl, H., Rieger, K., Concin, H., Grob, K., 2008. Mineral oil paraffins in human body fat and milk. Food Chem. Toxicol. 46, 544–552. EFSA, 2006. Opinion of the Scientific Panel on food additives, flavourings, processing aids and materials in contact with food (AFC) on a request related to a 13th list of substances for food contact materials. Adopted November 29, 2006. Farrimond, P., Griffiths, T., Evdokiadis, E., 2002. Hopanoic acids in Mesozoic sedimentary rocks: their origin and relationship with hopanes. Organic Geochem. 33, 965–977. Grob, K., Artho, A., Biedermann, M., Egli, J., 1991a. Food contamination by hydrocarbons from lubricating oils and release agents: determination by coupled LC-GC. Food Addit. Contam. 8, 437–446. Grob, K., Biedermann, M., Artho, A., Egli, J., 1991b. Food contamination by hydrocarbons from packaging materials determined by coupled LC-GC. Z. Lebensm. Unters. Forsh A 193, 213–219. Grob, K., Lanfranchi, M., Egli, J., Artho, A., 1991c. Determination of food contamination by mineral oil from jute sacks using coupled LCGC. J. Assoc. Off. Anal. Chem. 74, 506–512. Grob, K., Bronz, M., 1995. On-line LC-GC transfer via Cortes interface and a vaporizer; increased sensitivity for the determination of mineral oil in foods. J. High Resol. Chromatogr. 7, 421–427. Grob, K., Huber, M., Boderius, U., Bronz, M., 1997. Mineral oil material in canned foods. Food Addit. Contam. 14, 83–88. Grob, K., Vass, M., Biedermann, M., Neukom, H.-P., 2001. Contamination of animal feed and food from animal origin with non-edible waste oils? Food Addit. Contam. 18, 1–10. Guinda, A., Lanzón, A., Albi, T., 1996. Differences in hydrocarbons of virgin olive oils obtained from several olive varieties. J. Agric. Food Chem. 44, 1723–1726. Hard, G.C., 2000. Short-term adverse effects in humans of ingested mineral oils, their additives and possible contaminants-a review. Hum. Experim. Toxicol. 19, 158–172. Henry, J.A., 1998. Composition and toxicity of petroleum products and their additives. Hum. Experim. Toxicol. 17, 111–123.
506
Koprivnjak, O., Procida, G., Favretto, L., 1997. Determination of endogenous aliphatic hydrocarbons of virgin olive oils of four autochthonous cultivars from Krk island (Croatia). Food Technol. Biotechnol. 35, 125–131. Koprivnjak, O., Moret, S., Populin, T., Lagazio, C., Conte, L.S., 2005. Variety differentiation of virgin olive oil based on n-alkane profile. Food Chem. 90, 603–608. Low, L.K., Shymansky, P.M., Kommineni, C., Naro, P.A., Mackerer, C.R., 1992. Oral absorption and pharmacokinetics studies of radiolabelled normal paraffinic, isoparaffinic and cycloparaffinic surrogates in white oil in Fisher 344 rats. In “Transcript of the Toxicology Forum Special Meeting on Mineral Hydrocarbons”, pp. 86–101, Oxford. McGill, A.S., Moffat, C.F., Mackie, P.R., Cruickshank, P., 1993. The composition and concentration of n-alkanes in retail samples of edible oils. J. Sci. Food Agric. 61, 357–362. Moret, S., Grob, K., Conte, L.S., 1996. On-line high-performance liquid chromatography-solvent evaporation-high performance liquid chromatography-capillary gas chromatography-flame ionisation detection for the analysis of mineral oil polyaromatic hydrocarbons in fatty foods. J. Chromatogr. A 750, 361–368. Moret, S., Populin, T., Conte, L.S., Grob, K., Neukom, H.-P., 2003. Occurrence of C15-C45 mineral paraffins in olives and olive oils. Food Addit. Contam. 20, 417–426. Nash, J.F., Gettings, S.D., Diembeck, W., Chudowski, M., Kraus, A.L., 1996. A toxicological review of topical exposure to white mineral oils. Food Chem. Toxicol. 34, 213–225. Neukom, H.-P., Grob, K., Biedermann, M., Noti, A., 2002. Food contamination by C20-C50 mineral paraffins from the atmosphere. Atmos. Environ. 36, 4839–4847. Noti, A., Grob, K., Biedermann, M., Deiss, U., Brüschweiler, B.J., 2003. Exposure of babies to C15–C45 mineral paraffins from human milk and breast salves. Reg. Tox. Pharm. 38, 317–325.
SECTION | I Toxicology and Contaminants
Populin, T., Biedermann, M., Grob, K., Moret, S., Conte, L.S., 2004. Relative hopane content confirming the mineral origin of hydrocarbons contaminating foods and human milk. Food Addit. Contam. 9, 893–904. SCF, 1995. Opinion expressed on 22 September 1995 (Annex 5 to Document III/5611/95). Scotter, M.J., Castle, L., Massey, R.C., Brantom, P.G., Cunninghame, M.E., 2003. A study of the toxicity of five mineral hydrocarbon waxes and oils in the F344 rat, with histological examination and tissue-specific chemical characterisation of accumulated hydrocarbon material. Food Chem. Toxicol. 41, 489–521. Tan, Y.A., Kuntom, A., 1993. Gas chromatographic determination of hydrocarbons in crude palm kernel oil. J. AOAC Int. 76, 371–376. Wagner, C., Neukom, H.P., Grob, K., Moret, S., Populin, T., Conte, L.S., 2001a. Mineral paraffins in vegetable oils and refinery by-products for animal feeds. Mitt. Lebensm. Hyg. 92, 499–514. Wagner, C., Neukom, H.P., Galleti, V., Grob, K., 2001b. Determination of mineral paraffins in feeds and foodstuffs by bromination and preseparation on aluminium oxide: method and results of a ring test. Mitt. Lebensm. Hyg. 92, 231–249. Wang, Z., Fingas, M., Owens, E.H., Sigouin, L., Brown, C.E., 2001. Long-term fate and persistence of the spilled Metula oil in a marine salt marsh environment. Degradation of petroleum biomarkers. J. Chromatogr. A 962, 275–290. Wong, P.K., Wang, J., 2001. The accumulation of polycyclic aromatic hydrocarbons in lubricating oils over time – a comparison of supercritical fluid and liquid-liquid extraction methods. Environ. Sci. Technol. 35, 306–311.
1.4
Analytical Methods Natural Components Adverse Components
This page intentionally left blank
Chapter 56
Analytical Determination of Polyphenols in Olive Oil Antonio Segura-Carretero1, Alegría Carrasco-Pancorbo1, Alessandra Bendini2, Lorenzo Cerretani2 and Alberto Fernández-Gutiérrez1 1 2
Department of Analytical Chemistry, Faculty of Sciences, University of Granada, Spain Department of Food Science, University of Bologna, Cesena (FC), Italy
56.1 INTRODUCTION In any evaluation of the quality of virgin olive oil the quantity of phenolic compounds it contains is an important parameter to bear in mind. It has been demonstrated that these compounds affect both the resistance of the oil to oxidation and its typical bitter and pungent tastes. Furthermore, some studies have shown that the quantity of phenols, together with a favorable monounsaturated-to-polyunsaturated fatty-acid ratio, is related to several health-giving properties. All these factors make VOO a valuable foodstuff and add importance to the determination of its phenolic fraction, both qualitatively and quantitatively. The phenolic fraction of VOO is very complex and despite its having been studied for decades and excellent progress having been made, it must be admitted that a considerable number of compounds present in it have still not been completely characterized and many problems remain to be resolved. The reason lying behind these difficulties is the complexity of the chemical nature of these compounds and the similar complexity of the matrix in which they are found. One of the current difficulties hindering rapid and reproducible analyses of phenolic compounds is the scarcity of suitable pure standards, in particular of secoiridoid and lignan compounds. In general, any analytical procedure for the determination of individual phenolic compounds in VOO involves three basic steps: extraction from the oil sample, analytical separation and identification and quantification. These steps will be explained in later sections of this chapter; Figure 56.1 lays out the whole scheme of the topics reviewed in this chapter. As far as the extraction step is concerned, we will focus our attention on LLE and SPE. Then we will explain the Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
evolution from spectrophotometric assays to separation techniques coupled with modern mass analyzers and NMR instruments. The purpose of this chapter, rather than to present the reader with an overwhelming number of references for each subject, is to provide an overview of the tools used to date and the new trends for analyzing phenols from olive oil, whilst emphasizing the idea that several of the techniques may be considered as complementing each other.
56.2 SAMPLE PREPARATION Preparation of the sample is often one of the most important steps in any method to analyze a fraction of compounds or a family of analytes from any matrix. It may be said that the isolation of phenolic compounds from the sample matrix is generally a prerequisite to any comprehensive analytic scheme although enhanced selectivity in the subsequent quantification step may reduce the need for sample manipulation. A great number of procedures for the isolation of the polar phenolic fraction of VOO using two basic extraction techniques, LLE and SPE, have been published in the literature. The systems not only vary in the solvents and/or solid-phase cartridges used but also in the quantities of sample needed for analysis, volumes of the solvents and other such details (Hrncirik and Fritsche, 2004). Even though this section will mainly describe the liquid– liquid protocols and solid-phase extraction methods, it is worth emphasizing that sometimes a hydrolysis step has been introduced to minimize interference in the subsequent analysis. New developments are widening our future possibilities with regard to the extraction of phenols from
509
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
510
SECTION | I Natural Components
HO
SPE
HO
LLE
OH R
Extraction
TLC
HPLC
HO
O
O
UV, MS, Fluorescence, sensors, NMR
Analytical separation Determination
Spectrophotometric determination
Identification/ Quantification
O O
GC FID, MS
MS CE UV, MS NMR
Other tools to be used Statistics
m/z FIGURE 56.1 Scheme in detail of any analytical procedure for the determination of individual phenolic compounds in VOO involving three basic steps: extraction from the oil sample, analytical separation, and identification and quantification.
oils with techniques such as supercritical fluid extraction, microwave-assisted extraction, simultaneous microwave-assisted solid–liquid extraction, solid-phase microextraction and pressurized liquid or fluid extraction, among others. It is difficult to assure the recovery of the extraction systems when we are working with phenols from olive oil since they are a heterogeneous mixture of compounds, which in many cases are not commercially available. Other similar phenolic compounds have frequently been used in the past. In recent studies, to overcome this obstacle, a refined phenol-free olive oil has been spiked with an exactly specified dose of a phenolic extract prepared by the extraction of VOO.
56.2.1 Liquid–Liquid Extraction Phenolic compounds of olive oil have traditionally been isolated by extracting an oil solution in a lipophylic solvent with several portions of methanol (Owen et al., 2000) or methanol/water (with different quantities of water ranging between 0% and 40%) (Vázquez Roncero, 1978; Solinas and Cichelli, 1981, 1982; Solinas,
1987; Tsimidou et al., 1996) followed by evaporation of the solvent from the aqueous extract and a clean-up of the residue by solvent partition (Tsimidou et al., 1992; Tasioula-Margari and Okogeri, 2001). The most widely used solvent has been hexane (petroleum ether and chloroform have also been proposed), although the addition of hexane or other organic solvents in the oil before extraction does not result in any significant differences in the efficiency of phenol recovery. Tensioactive substances such as Tween 20 (2% v/w) have often been used to liberate the phenolic compounds of the lipoprotein membranes. Extraction with tetrahydrofuran/water followed by centrifugation (Cortesi et al., 1995a, b) and extraction with N, N-dimethylformamide (Brenes et al., 2000) has also been assayed. Montedoro et al. (1992a, b) examined the extractive methods of simple and hydrolyzable phenolic compounds in virgin olive oil, studying different schemes where the volume and percentages of the solvents in which the olive oil was dissolved were altered. The best results were obtained using methanol/water (80:20 v/v), which is in accordance with data in literature (Montedoro and Cantarelli, 1969), and the optimum extraction was achieved by extracting 100 g of olive oil with two volumes of 20 mL
CHAPTER | 56 Analytical Determination of Polyphenols in Olive Oil
of solvent. Nevertheless, some years later Angerosa et al. (1995) reported contrasting results to these: the incomplete recovery of some components and the formation of considerable emulsions between the oil and the methanol/water layer led to their choosing neat methanol as extraction solvent. Cortesi et al. (1995a, b) assayed the extraction of the polar fraction of olive oil with tetrahydrofuran/water (80:20, v/v) followed by centrifugation, and concluded that recoveries were five-fold higher with this method in terms of hydroxytyrosol and twice as high for tyrosol than with methanol/water (60:40 v/v). The use of N, N-dimethylformamide has also seemed to show interesting results in terms of recovery efficiency and sample manipulation (Brenes et al., 2000). After liquid–liquid extraction, residual oil can be removed if necessary by overnight storage at subambient temperature, by centrifugation or by further extraction with hexane, although Shepadex column (Montedoro et al., 1992a, b) and Policlar AT:Celite 560 (1:2) (Solinas and Cichelli, 1981, 1982) chromatography have also been used to effect further clean-up.
56.2.2 Solid-phase Extraction Solid-phase extractions can use the same type of stationary phases that are used in LC columns and so the versatility of this kind of extraction has been taken advantage of for the recovery of phenolics from olive oil and various other systems employing SPE, either as an isolation or a cleanup step before using a chromatographic or other analytical method to quantify the analyte(s) in the sample. Some of the suitable adsorbents are alkylsilicas, such as C8 (Pirisi et al., 2000) or C18 (Gutiérrez et al., 1989). In principle, C18phase is less suitable for the isolation of polar components from a non-polar matrix than normal-phase SPE, although C18-cartridges have often been tested for isolating phenolics from VOO (Favati et al., 1994; Servili et al., 1999). Anionic exchange cartridges have also been used to isolate the phenolic fraction from various seed oils, but recoveries were low (53–62%) for some components (Andreoni and Fiorentini, 1995). Promising results were obtained by Mateos et al. (2001), who worked with amino-phase and diol-bond-phase SPE cartridges and achieved a high recovery rate with the latter (⬎90%) for all major olive phenolics. In doped, refined, olive oil samples recovery rates have been studied using a C18-cartridge with total suppression of the residual silanolic group C18 EC (end-capped C18) (Liberatore et al., 2001). In this study the discrimination between C18 and C18 EC was not as great as expected because the two stationary phases differ only in the presence of free silanolic groups. Nevertheless, their presence seems to improve the release mechanism and increases recovery rates.
511
56.3 ANALYTICAL TECHNIQUES FROM: SPECTROPHOTOMETRIC DETERMINATIONS TO POTENT METHODS BASED UPON MS AND NMR, WHICH ARE ABLE TO DISCOVER AND CHARACTERIZE NEW PHENOLIC COMPOUNDS IN VOO 56.3.1 Spectrophotometric Assays Total phenolics in VOO can be quantified by colorimetric assays based upon the reaction of Folin-Ciocalteu reagent with the functional hydroxy groups of phenolics (Singleton and Rossi, 1965; Gutfinger, 1981). The method consists of calibration with a pure phenolic compound, extraction of phenolics from the sample and the measurement of absorbance after the color reaction. This method is fast and simple (Blekas et al., 2002) and thus, in spite of the fact that this assay is not very specific (as the color reaction can occur with any oxidizable phenolic hydroxy group), it is still quite popular. The determination of the total phenolic content is based upon the procedure of Folin-Ciocalteu mentioned above (Gutfinger, 1981). For example, a typical protocol using this method might be as follows (Cerretani et al., 2003): an aliquot of the aqueous-methanolic solution of phenolics extracted from ⬇2 g of virgin olive oil is diluted in 6 mL of water, followed by the addition of 0.5 mL of Folin-Ciocalteu reagent. After 1 min, 2 mL of sodium carbonate solution (15% w/v) is added to the reaction mixture, which is finally mixed and diluted with water to 10 mL.The absorbance of the solution is measured after 2 h against a blank sample (the same protocol but using 0.1 mL of methanol/ water (50:50 v/v)) at a wavelength of 750 nm. In this case the calibration curve is constructed using standard solutions of gallic acid within the range of 0.01 to 1 mg mL⫺1. Different authors have, however, made several modifications to this method (Gutfinger, 1981; Shahidi and Naczk, 1995). The close relationship between olive oil stability and its total phenol and o-diphenol contents has long been known (Gutfinger, 1981) and consequently a number of research groups have focused their attention on this subject. Here we mention briefly a typical protocol for the determination of o-diphenols (Mateos et al., 2001): the methanolic extract obtained from ⬇2.5 g olive oil is evaporated, the residue dissolved in 10 mL methanol/water (50:50 v/v) and the solution filtered. A mixture of 4 mL of the solution and 1 mL of a 5% solution of sodium molibdate dihydrate in ethanol/water (50:50 v/v) is shaken vigorously. After 15 min the absorbance at 370 nm is measured. A blank is obtained by measuring a mixture of 4 mL of phenolic solution with 1 mL of ethanol/water (50:50 v/v) (Maestro-Durán et al., 1991).
512
56.3.2 Chromatographic Determination of Phenolic Compounds Even if spectrophotometric determinations can be useful in some cases, some of the drawbacks to this technique have already been mentioned. If in fact the analyst’s intention is to profile, identify and quantify individual phenolic compounds, he cannot rely on traditional methods alone. In other words, in order to use VOO as a source of phenolic compounds to develop complete compositional databases and obtain more accurate data about the intake of antioxidants further chemical characterization is needed. Identification and quantification, based traditionally on HPLC (with different detectors, such as UV, fluorescence, coulometric electrode array detection, amperometric detector, GC-FID and, more recently, CE-UV), can be helped nowadays by the complementary use of MS and NMR. Our prime focus in this section will be to afford the reader an overview of the chromatographic (GC and HPLC) determination of phenols from olive oil without burdening him with too many specific references. Janer del Valle and Vázquez-Roncero (1980) published the first paper about the separation of phenolic compounds of olive oil by GC more than 25 years ago. At the same time Solinas and Cichelli (1982) described the use of GC to identify mixtures of VOOs and refined oils. GC has several advantages, such as high chromatographic resolution, high peak capacity, easy quantification with FID, a single mobile phase (usually helium), few problems with solubility and separations that can be adjusted by electronic controls such as temperature and so on, but the major disadvantage is that to be suitable for GC analysis a compound must be sufficiently volatile and thermally stable. To achieve these ends we can use derivatization reactions, but, apart from introducing additional stages into the analysis, derivatization can result in a wide variability in the apparent recovery rate, thus rendering any quantitative evaluation completely unreliable. In addition, unwanted compounds may be formed during derivatization because of the presence of other extraneous compounds and their degradation products. Another problem of this technique is that the use of high temperature could damage the analytes. In 1987 Forcadell et al. (Forcadell et al., 1987) developed a protocol for the preparation of trimethylsilyl (TMS) derivatives. In the same year Solinas (1987) published a paper on the development of a GC method for the qualitative/quantitative evaluation of phenolic compounds in VOOs from different cultivars at various stages of ripeness. Nevertheless, the use of sophisticated analytical techniques such as GC-MS and GC-MS/MS has improved the identification of compounds in the phenolic fraction of VOO. GC/MS provides a considerable amount of structural information and the possibility of using commercial libraries, which make the identification of unknown compounds more feasible. For further information we recommend that
SECTION | I Natural Components
the reader consult Angerosa et al. (1995), Del Carlo et al. (2004) and Rios et al. (2005). Table 56.1 sets out the relative uses of the three most commonly used separation techniques in the analysis of polyphenols from olive oil, HPLC currently being the most popular and reliable. In HPLC detection has been routinely achieved by ultraviolet absorption, often involving a photodiode array detector, although other techniques have been tried. In recent years MS has become one of the most important detectors for LC, making it a high priority to have one of these detectors in the laboratory, depending of course on the aims of the research involved. In fact LC combined with MS has revolutionized the analysis of non-volatile species, as revealed by the numerous papers published on the subject. To cite the many important papers on HPLC coupled to different detectors for the analysis of phenols from olive oil would exceed our aim in this chapter so we direct the interested reader to other reviews (Carrasco-Pancorbo et al., 2005; Bendini et al., 2007). We will confine ourselves here to explaining some aspects of the coupling of LC to MS. Liquid chromatography/mass spectrometry (LC-MS) has been widely accepted as the main tool in the identification, structural characterization and quantitative analysis of phenolic compounds in olive oil. Using a mass spectrometer for detection offers undoubted advantages, such as independence from a chromo- or fluorophore, lower LOD than UV in most cases, the possibility of obtaining structural information and the easy separation of coeluting peaks by using the information about mass as a second dimension. The sensitivity of response in MS clearly depends upon the interface technology employed. In LC-MS analysis of phenolic compounds, atmospheric-pressure ionization
TABLE 56.1 Relative use of GC, HPLC and CE for the analysis of phenolic compounds in olive oil and detectors coupled to each technique. Analytical technique
Percentage considering the overall number of publications
Detectors used
GC
15–20%
FID, NMR and MS
HPLC
80–85%
UV, fluorescence, electrochemical detection, biosensors, NMR and MS
CE
1–5%
UV and MS
Percentages calculated after searching the ISI Web of Knowledge, Scopus and Scirus looking for publications containing “X and olive and oil and phenol*”. X⫽ GC, gas chromatography, HPLC, liquid chromatography, LC, capillary electrophoresis, and CE.
513
CHAPTER | 56 Analytical Determination of Polyphenols in Olive Oil
interfaces, i.e. APCI and electrospray ionization (ESI), are used almost exclusively today, and both positive and negative ionization are applied. In general, phenolic compounds are detected with higher sensitivity in the negative ion mode, but the results from positive and negative ion modes are complementary and the positive ion mode shows structurally significant fragments. On the other hand, optimum ionization depends not only upon the interface parameters but also upon the mobile phase of the liquid chromatography. As a primary rule, the use of non-volatile salts in the mobile phase (common in other chromatographic methods) should be avoided as they interfere with the ionization source. The mobilephase composition and its pH also need careful optimization as they may influence the ionization efficiency of the analytes. The selection of the analyzer, apart from its accessibility, is determined by the sensitivity and selectivity required and the general objectives involved. LC-atmospheric-pressure ionization (API)-MS typically yields a single strong ion, which reduces its ability to identify analytes accurately. In most cases single-stage MS is used in combination with UV detection to facilitate the identification of phenolic compounds in olive oil samples with the help of standards and/or reference data. Ion trap or QqQ offer the possibility of doing MS/MS or MSn, which can be used to elucidate structure or gain sensitivity and thus selectivity by reducing chemical noise. MS/MS and MSn involve two (or more) stages of mass analysis separated by a fragmentation step. TOF MS, which is one of the most advanced MS analyzers, provides excellent mass accuracy over a wide dynamic range if a modern detector technology is chosen. It also allows measurements of the correct isotopic pattern, providing important additional information for the determination of elemental composition. Table 56.2 contains the most important characteristics of several analyzers used
for determining phenols in olive oil, including scan speed, sensitivity, resolution, capacity for doing MS/MS and for quantifying, price and main applications. In other areas, and supposedly before long in this field, bidimensional chromatography (LC-LC, LC-GC or GC-GC) is becoming more and more important. Coupled-column separations or multidimensional chromatography can be considered to be a form of sample preparation, as one column is used to derive fractions for the second column. These methods are more commonly used in GC than LC; the complete combination is two-dimensional chromatography in which fractions from the first column are continuously passed to a second column to give a very high sample capacity.
56.3.3 Nuclear Magnetic Resonance If the analyst combines the power of several mass analyzers and obtains information regarding the molecular mass of an unknown compound (with high mass accuracy and taking into account the correct isotopic pattern) and also its fragmentation, he should be able to identify the whole structure of the analyte in question. Sometimes, however, this is impossible and the analyst should then resort to nuclear magnetic resonance (NMR). High-resolution spectroscopic techniques, particularly NMR spectroscopy, are finding interesting applications in the analysis of complex mixtures of various phenol-bearing food extracts. During the past decade proton nuclear magnetic resonance spectroscopy (NMR) has been successfully used in olive oil analysis (Sacchi et al., 1997; Vlahov, 1999). Currently available high-resolution spectroscopic techniques, coupled with the facilities of computerized mathematical or other treatment of data, have found interesting applications in the field of agricultural and food science
TABLE 56.2 Comparison among several analyzers used to determine phenolic compounds extracted from olive oil. Analyzer
Scan speed
Sensitivity Full Scan (SIM)
Resolution
MS/ MS
Quantification
Price
Main applications
Q
⫺
⫺ (⫹)
0
⫺
⫹
Low
Selective detector
QqQ
⫺
⫺ (⫹)
⫹
⫹
⫹⫹
High
Quantification in ‘target analysis’ for compounds in complex matrix
IT
⫹
⫹
⫹
⫹⫹
⫹
Medium
Unknown analytes in complex samples
TOF
⫹⫹
⫹
⫹⫹
( ⫹)
⫹
High
Unknown analytes in complex samples (quantification)
Q: quadrupole; QqQ: triple quadrupole; IT: ion trap; TOF: time of flight.
514
without the need for a separation technique coupled with NMR (Gerothanassis et al., 1998). In addition the usefulness of 1H NMR spectroscopy is ever more widely recognized for its non-invasiveness, rapidity and sensitivity to a wide range of compounds in a single measurement. Nevertheless, difficulties may arise with regard to the information obtained from the spectra of multicomponent mixtures such as olive oil. Strong signal overlap, dynamic range problems, diversity of intensities due to various concentrations of the food constituents and the inherent lack of scalar coupling information between different moieties can all lead to ambiguous or incomplete assignments, thus hindering detection even with the use of multidimensional NMR (Christophoridou et al., 2005). One possible approach to these problems involves combining the advantages of NMR spectroscopy with those of chromatography. Coupled techniques such as LC-NMR or LC-NMR/MS may provide information about overall composition and enable the identification of individual phenols in complex matrices. Moreover, on-line solid-phase extraction (SPE) in LC-NMR for peak storage after liquid chromatography separation prior to NMR analysis or similar techniques has recently been applied. Montedoro et al. (1993) were pioneers in this field, reporting as they did the characterization of the compounds in question by NMR, UV and IR and arriving at the identification of four new structures. They concluded that the newly identified compounds were an isomer of oleuropein aglycon, the dialdehydic form of elenolic acid linked to hydroxytyrosol and the dialdehydic form of elenolic acid linked to tyrosol. Subsequently Limiroli et al. (1995) and Gariboldi et al. (1986) contributed to a deeper understanding of the secoiridoid fraction of VOO. Christophoridou et al. (2005) recently reported the first application of the hyphenated LC-SPE-NMR technique using post-column, solid-phase extraction to the direct identification of new phenolic compounds in the polar fraction of VOO. The addition of a post-column SPE system to replace the loop system of LC-NMR resulted in higher sensitivity (significant increase in the signal to noise (S/N) ratio); in fact, S/N improvements by up to a factor of 4 could be demonstrated with this new technology. The spectra recorded were one-dimensional (1D) 1H-NMR and two-dimensional (2D) NMR. The presence of phenols was confirmed from the respective LC-SPE-NMR spectra, which were assigned on the basis of existing 1HNMR databases and with total correlation spectroscopy (TOCSY). The most interesting findings of this study were the verification of the presence of the lignan syringaresinol, the presence of two stereochemical isomers of the aldehydic form of oleuropein and the detection of homovanillyl alcohol. The reader is encouraged to find further interesting references in the summarized tables in Bendini et al. (2007).
SECTION | I Natural Components
56.3.4 Direct Infusion in MS and NMR Several interesting reports have been published describing the analysis of olive oil by ionspray-ionization tandem mass spectrometry (IS-MS/MS) and ESI-MS/MS with NMR, without the use of a previous separation technique (Capasso, 1999; De Nino et al., 2000). This is direct-injection analysis, which describes the injection or infusion of a sample into the ionization source of a mass spectrometer without prior chromatographic or electrophoretic separation. Descriptions of many important and pioneering biomedical assays without any preliminary separation have already been published and it is likely that in the near future the same will happen in this field, although we must not ignore possible problems caused by the co-infusion of molecules acting as suppression agents. The use of sophisticated computational analysis will be extremely important if analysts use this kind of approach. We consider it worthwhile including in this chapter a recent publication by Christophoridou and Dais (2006) in which the authors demonstrate the potential of 31PNMR spectroscopy to detect and quantify a large number of phenols in VOO extracts. This novel analytical method is based upon the derivatization of the hydroxyl and carboxyl groups of phenolic compounds with 2-chloro-4,4,5,5 tetramethyldioxaphospholane and the identification of the phosphitylated compounds on the basis of the 31P chemical shifts.
56.3.5 Capillary Electrophoresis Even when the phenolic compounds from VOOs have been successfully characterized and quantified by GC and HPLC with different detectors and with MS and NMR coupled to HPLC, without any previous separation, the use of faster analytical techniques and screening tools to allow a rapid screening of these compounds is strongly recommended. Separation by CE is highly efficient, requires only a small sample and little electrolyte consumption, and takes only a few minutes. This last characteristic represents the main advantage of CE over chromatographic methods and makes it useful for routine analysis as well as for controlling and monitoring processes in a number of industrial fields (Frazier, 2001). Moreover, CE is relatively well suited to analyzing samples with complex matrices such as VOO. CE can be coupled with different detectors, such as UV, LIF, electrochemical detectors or MS. As far as the analysis of phenolic compounds in VOO is concerned, several studies have reported the use of CE with ultraviolet detection, whilst only two describe the results obtained with CE-MS (all of them are reported in previous reviews: CarrascoPancorbo et al., 2005; Bendini et al., 2007). In general all the CE-UV methods published so far use simple CZE methods based on an alkaline borate run buffer. Differences can
CHAPTER | 56 Analytical Determination of Polyphenols in Olive Oil
be found amongst these methods with regard to the voltage applied, the internal diameter of the capillary, injection time (always hydrodynamic injection), the effective length of the capillary and the buffer concentration. Figure 56.2 sets out the electropherograms obtained for a methanol/ water extract of a Picual extra virgin olive oil using DAD detection at four wavelengths (200, 240, 280 and 340 nm) (Carrasco-Pancorbo et al., 2006). The method permits the analyst to identify simple phenols, lignans, complex phenols (isomeric forms of secoiridoids), phenolic acids and flavonoids in the SPE-Diol extracts from extra virgin olive oil in less than 10 min with satisfactory resolution. CE can achieve the aims traditionally achieved by HPLC, providing an alternative way of characterizing phenolic compounds from olive oil, and proved that in instances in which none of the HPLC methods provides enough resolution CE, with its flexible experimental conditions, should be assayed as a complementary second-choice technique. The use of CE as an analytical separation technique coupled with mass spectrometry as a detection method can provide important advantages in the analysis of phenolic compounds from olive oil because of the combination of the high separation capabilities of CE and the power of MS for identification and confirmation. When using mass spectrometric detection, differences in optical detection must be taken into account. Firstly, the separation electrolyte has to be volatile, thus reducing the choice of buffering system primarily to ammonia, acetate or formate. Whilst there are reports of non-volatile buffers being used with CE-UV, only low concentrations can be used, leading to lower sensitivity. Generally, non-aqueous solvents are well suited for coupling with MS and add another parameter to modifying selectivity. As mentioned above for HPLC coupled with MS, CE can also be coupled with different MS analyzers, such as quadrupole, ion trap and time-of-flight, and use several ionization methods such as APCI, ESI and so on. ESI is one of the most versatile ionization methods and is the natural method of choice for the detection of ions separated by CZE. With regard to the analyzers, ion trap (IT) and TOF systems are the two commonest analyzers to be found in a food analysis laboratory, although single-quadrupole MS is still often used as an easy and affordable detector. Of particular interest is the coupling of CZE to ESITOF MS. This coupling combines the above-mentioned benefits of CZE separation with the high selectivity, due to a mass accuracy of 5 ppm, which opens the possibility of determining elemental compositions. It might be worth highlighting one recent paper in which the results obtained by using CE and HPLC coupled to ESI-TOF MS are compared (Carrasco-Pancorbo et al., 2007). For HPLC the mobile phases A and B consisted of water plus 0.1% acetic acid and acetonitrile respectively; the chromatographic method consisted of a linear gradient
515
from 0 to 100% B for 30 min followed by a cleaning cycle of 8 min with 100% B and 7 min with 0% B (initial conditions). The flow-rate was 0.20 mL min⫺1. Optimum electrophoretic separation was obtained using a basic carbonate electrolyte (25 mM ammonium hydrogen carbonate at pH 9.0). Major phenolic compounds, previously observed in several studies, belonging to several important families (phenyl alcohols, phenyl acids, lignans, flavonoids and secoiridoids) were detected by HPLC- and CE-ESITOF MS in the polar fraction of the olive oil. These compounds are summarized in Table 56.3 (in particular those for HPLC-ESI-TOF, although the results obtained by both techniques agreed well), with their formulas, selected ions, experimental and calculated m/z values, errors, sigma values, tolerance and the first compound in the list of possibilities (sorted with respect to the sigma value). The BPE (50–800 m/z) and BPC (50–800 m/z) of a phenolic extract of Arbequina extra virgin olive oil are shown in Figure 56.3; in both cases several gray peaks can be seen, indicating well-known phenolic compounds included in Table 56.3. We also show by a point over the peak those compounds with a molecular mass not found before in olive oil. Table 56.4 (for CE-ESI-TOF) and Table 56.5 (for HPLCESI-TOF) summarize all the results for the unknown compounds. Some lines, indicating the compounds which were not found using HPLC or CE-ESI-TOF MS, are in bold print. As far as the unknown compounds are concerned, 26 were characterized by HPLC whilst 28 were detected with CE. HPLC provided information about two compounds that were not detected with CE (m/z experimental 291.1952 and 205.1599) and CE-MS gave a list of possible molecular formulae for four compounds which were not determined by LC-MS (m/z experimental 471.3457, 150.0562, 183.0304 and 201.0400). In a work published recently Aturki et al. (2008) achieved the simultaneous separation of ten phenolic compounds (protocatechuic, p-coumaric, o-coumaric, vanillic, ferulic, caffeic, syringic acids, hydroxytyrosol, tyrosol and oleuropein) in extra-VOO by isocratic, reversed-phase, capillary electrochromatography in less than 35 minutes after a suitable LLE procedure.
56.4 CONCLUSIONS Apart from our general interest in knowing about the composition of the polar fraction of VOO, an accurate determination of these compounds also helps us to understand their health benefits, which include the reduction of risk factors contributing to heart disease, and the prevention of several varieties of cancer and modifications to immune and inflammatory responses. It is equally interesting to distinguish which phenolic molecules are responsible for bitterness, pungency, astringency and metallic tastes and to evaluate the antioxidant activities of the polar fraction.
0.39
Absorbance (AU)
10*,11 0.29
2 1
12* 6
9
0.19
16
5* 0.09
17
78
3
200 nm
15 1314
4
19 20
26
−0.01 0
A
1
2
3
4
5
6
7
8
9
10
Time (min) 0.15
Absorbance (AU)
10*,11 0.11 12* 6
1
0.07
16
5*
0.03
240 nm
17
78 9
2
13* 15 14
3 4
26 1819 20 21 22
24 25 23
−0.01 0
B
1
2
3
4
5
6
7
8
9
10
Time (min) 10*,11
0.038
Absorbance (AU)
12* 6
0.028 12
5* 7
9 13*
0.018
0.008
15 14
8
3
16 17
280 nm 18 19 21 20
4 −0.002 0
C
1
2
3
4
5
6
7
8
9
10
Time (min) 0.022 21000 Absorbance(μAU)
0.017
Absorbance (AU)
15000
19
0.012
190 230 270 310 360 390 430
nm
20
190 230 270 310 360 390 430
nm
340 nm
0.007
0.002
−0.003
D
0
1
2
3
4
5
6
7
8
9
10
Time (min)
FIGURE 56.2 Electropherograms obtained for a methanol/water extract of a Picual extra virgin olive oil using the DAD detection at four wavelengths (200, 240, 280 and 340 nm). The Diol-SPE protocol, in D) (340 nm), was done using the same quantity of olive oil (60 g) and conditions as in the other cases, but the residue after the rotary evaporator was dissolved with 0.5 ml of methanol/water (50/50 v/v) to obtain more concentrated extract in terms of phenolic compounds. Separation conditions: capillary, 47 cm (40 cm of effective length); applied voltage, 28 kV; applied temperature, 22 °C; buffer, 45 mM sodium tetraborate (pH 9.30); hydrodynamic injection, 0.5 p.s.i. for 8 s. Peak identification numbers: 1, Lig Agl (a); 2, TY, 3, Pin; 4, Ac Pin; 5, Ol Agl (a) ⫹ DOA (a); 6, DOA (b); 7, Lig Agl (b); 8, Ol Agl (b); 9, EA (a); 10, Ol Agl (c) ⫹ Lig Agl (c) ⫹ DOA (c) ⫹ EA (b,c); 11, HYTY; 12, DOA (d) ⫹ EA (d); 13, EA (e); 14, trans-cinnamic acid; 15, 4-hydroxyphenylacetic acid; 16, sinapinic acid; 17, gentisic acid; 18, o-coumaric acid; 19, luteolin; 20, apigenin; 21, vanillic acid; 22, 4-hydroxybenzoic acid; 23, caffeic acid; 24, 3,4-dihydroxyphenylacetic acid; 25, gallic acid; and 26, protocatechuic acid. *Peaks overlapped. Reproduced with permission from Carrasco-Pancorbo et al. J. Sep. Sci. 2006, 29, 2221–2233.
Compound
HYTY TY EA HYTY-Ac DOA Ol Agl 10-H-Ol Agl DecarboxLig Agl Lig Agl Pin Lut Ac Pin Apig H-Pin
Selected ion
m/z experimental
m/z calculated
C8H10O3 C8H10O2 C11H14O6 C10H12O4 C17H20O6 C19H22O8 C19H22O9
[M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺
153.0562 137.0609 241.0718 195.0660 319.1184 377.1224 393.1183
C17H20O5 C19H22O7 C20H22O6 C15H10O6 C22H24O8 C15H10O5 C20H22O7
[M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺ [M-H]⫺
303.1239 361.1284 357.1338 285.0405 415.1389 269.0449 373.1283
Formula
Error
Sigma Value
Classification order considering other possibilities
Tolerance (ppm) in generate Molecular Formula
(ppm)
(mDa)
153.0557 137.0608 241.0718 195.0663 319.1187 377.1242 393.1191
⫺3.267 ⫺0.362 ⫺0.306 1.415 0.824 4.703 1.985
⫺0.50 ⫺0.05 ⫺0.07 0.28 0.26 1.77 0.78
0.0012 0.0048 0.0181 0.0129 0.0178 0.0272 0.0087
1st (1) 1st (1) 1st (3) 1st (3) 2nd (4) 2nd (8) 1st (4)
20 15 10 15 10 10 10
303.1238 361.1293 357.1344 285.0405 415.1398 269.0455 373.1293
⫺0.212 2.418 1.518 ⫺0.041 0.942 2.509 2.772
⫺0.06 0.87 0.54 ⫺0.01 0.39 0.67 1.03
0.0225 0.0105 0.0360 0.0118 0.0131 0.0478 0.0198
1st (3) 1st (6) 2nd (6) 2nd (5) 2nd (10) 2nd (4) 1st (5)
10 10 10 10 10 10 10
First compound in the list of possibilities
C13H15N6O4 C11H13N12O4
C15H21N2O8 C12H1N10 C13H19N8O8 C1H1N16O2
CHAPTER | 56 Analytical Determination of Polyphenols in Olive Oil
TABLE 56.3 Well-known phenolic compounds determined by HPLC-ESI-TOF-MS in an extract of Arbequina extra-virgin olive oil.
Other compounds in the list of possibilities Vanillin Vanillic acid Caffeic acid o-coumaric acid Ferulic acid
C8H8O3 C8H8O4 C9H8O4
[M-H]⫺ [M-H]⫺ [M-H]⫺
151.0406 167.0353 179.0354
151.0401 167.0350 179.0350
⫺3.828 ⫺1.996 ⫺2.409
⫺0.58 ⫺0.33 ⫺0.43
0.0089 0.0517 0.0580
1st (1) 1st (2) 1st (2)
10 10 10
C9H3N4 C10H3N4
C9H8O3 C10H10O4
[M-H]⫺ [M-H]⫺
163.0405 193.0512
163.0401 193.0506
⫺2.673 ⫺2.997
⫺0.44 ⫺0.58
0.0307 0.5746
1st (1) 1st (2)
10 10
C11H5N4
517
518
SECTION | I Natural Components
Intens. ×105
BPE 50–800 – Spectral Bkgrnd ‘Unknown’ peaks
CE-ESI-TOF
3
2
1
0 2
3
4
5
6
7
8
9
10
Intens. ×105
11 Time [min]
BPC 50–800 – All Ms ‘Unknown’ peaks
HPLC-ESI-TOF
5
2 1
Calibrant
3
Calibrant
4
0 0
5
10
15
20
25
Time [min]
FIGURE 56.3 Base peak electropherogram of the Diol-SPE extract obtained from an Arbequina extra virgin olive oil at the optima electrophoretic (running buffer 25 mM ammonium hydrogen carbonate at pH 9.0) and MS conditions, and base peak chromatogram of the same extract at optima HPLC (mobile phases A and B consisted of water with 0.1% acetic acid, and acetonitrile, respectively; and the chromatographic method consisted of a linear gradient from 0 to 100% B during 30 min, followed by a cleaning cycle of 8 using min 100% B and 7 min with 0% B (initial conditions). Flow-rate 0.20 mL/min) and MS conditions. Peaks in gray represent the well-known phenolic compounds. • With a point over the peak, those are compounds which have a molecular mass that has not previously been found in olive oil. Reproduced with permission from Carrasco-Pancorbo et al. Electrophoresis 2007, 28, 806–821.
TABLE 56.4 Unknown phenolic compounds determined by CE-ESI-TOF-MS in an extract of Arbequina extra virgin olive oil. m/z experimental
Selected ion
Tolerance (ppm) in generate Molecular Formula
List of possibilities in generate Molecular Formula (in increasing order of sigma)
Error (ppm) (for the first compound)
Sigma value (for the first compound)
155.0715
[M-H]⫺
20
C8H11O3
⫺0.853
0.0085
223.1331
[M-H]⫺
15
C8H19N2O5 / C13H19 O3 / C9H15N6O1 / C15H15N2
2.444 2nd(3.456)
0.0134 2nd(0.0265)
171.1389
[M-H]⫺
10
C10H19O2
1.187
0.0644
199.1704
[M-H]⫺
10
C12H23O2
⫺0.368
0.0772
315.1228
[M-H]⫺
5
C18H19O5
3.184
0.0360
315.2535
[M-H]⫺
10
C18H35O4/ C14H31N6O2 / C19H31N4
1.759
0.0228
471.3457
[M-H]⫺
5
C25H47N2O6 / C30H47 O4 / C26H43N6O2
ⴚ 3.760 2nd(4.774)
0.0257 2nd(0.0459)
TABLE 56.4 (Continued) m/z experimental
Selected ion
Tolerance (ppm) in generate Molecular Formula
List of possibilities in generate Molecular Formula (in increasing order of sigma)
Error (ppm) (for the first compound)
Sigma value (for the first compound)
299.2584
[M-H]⫺
10
C18H35O3 / C13H35N2O5/ C14H31N6O1
2.673
0.0179
297.2431
[M-H]⫺
15
C18H33O3 / C13H33N2 O5 / C14H29N6O1
1.346
0.0262
295.2272
[M-H]⫺
10
C18H31O3 / C13H31N2 O5 / C14H27N6O1
2.371
0.0563
311.2219
[M-H]⫺
10
C18H31O4/ C13H31N2O6 / C14H27N6O2 / C19H27N4
2.892
0.0611
407.1356
[M-H]⫺
5
C3H7N26 / C20H23O9 / C21H19N4O5 / C17H15N10O3
⫺0.865 2nd(-2.073)
0.0103 2nd(0.0123)
391.1404
[M-H]⫺
5
C20H23O8 / C16H19N6 O6 / C17H15N10O2 / C21H19N4O4
⫺1.511
0.0350
243.1966
[M-H]⫺
10
C14H27O3
⫺0.130
0.0157
199.0617
[M-H]⫺
15
C9H11O5 / C10H7N4O1
⫺2.488
0.0587
225.1488
[M-H]⫺
15
C13H21O3
3.553
0.0476
335.1134
[M-H]⫺
5
C17H19O7 / C14H11N10O1 / C18H15N4O3
0.587
0.0264
143.1082
[M-H]⫺
20
C8H15O2
⫺2.795
0.0629
257.0673
[M-H]⫺
10
C11H13O7 / C8H5N10 O1 / C12H9N4O3
⫺2.389
0.0720
215.0919
[M-H]⫺
15
C10H15O5 / C11H11N4O1
2.600
0.0652
150.0562
[M-H]⫺
20
C8H8N1O2 / C4H4N7
⫺0.838
0.0069
187.0970
[M-H]⫺
15
C5H11N6O2 / C9H15O4 / C10H11N4
⫺10.314 2nd(4.039)
0.0060 2nd(0.0108)
271.0608
[M-H]⫺
5
C15H11O5 / C1H3N16 O2 / C16H7N4O1
1.544
0.0349
157.1240
[M-H]⫺
20
C9H17O2
⫺3.496
0.0051
183.0304
[M-H]⫺
20
C8H7O5 / C9H3N4O1
ⴚ 2.732
0.0521
183.0667
[M-H]⫺
20
C9H11O4 / C10H7N4
2.197
0.0174
201.0400
[M-H]⫺
10
C8H9O6 / C5H1N10 / C9H5N4O2
2.457
0.0527
299.0561
[M-H]⫺
5
C2H3N16O3 / C5H11N6O9 / C16H11O6/ C13H3N10 / C17H7N4O2
2.803
0.0478
520
SECTION | I Natural Components
TABLE 56.5 ‘Unknown’ phenolic compounds determined by HPLC-ESI-TOF-MS in an extract of Arbequina extra virgin olive oil. m/z experimental
Selected ion
Tolerance (ppm) List of possibilities in in generate Molecular generate Molecular Formula Formula (in increasing order of sigma)
Error (ppm) (for the first compound)
Sigma value (for the first compound)
199.0610
[M-H]⫺
20
C9H11O5 / C10H7N4O1
3.463
0.0084
155.0714
[M-H]⫺
25
C8H11O3
⫺3.058
0.0069
215.0930
[M-H]⫺
15
C10H15O5 / C11H11N4O1
⫺2.299
0.0148
183.0667
[M-H]⫺
20
C9H11O4 / C10H7N4
2.197
0.0074
257.0668
[M-H]⫺
10
C11H13O7 / C8H5N10O1 / C12H9N4O3
⫺0.355
0.0126
187.097
[M-H]⫺
15
C9H15O4 / C5H11N6O2 / C10H11N4
1.273
0.0066
335.1134
[M-H]⫺
5
C17H19O7 / C14H11N10O1 / C18H15N4O3
0.587
0.0224
271.0608
[M-H]⫺
5
C15H11O5 / C1H3N16O2 / C16H7N4O1
1.544
0.0249
299.0561
[M-H]⫺
5
C16H11O6 / C13H3N10 / C17H7N4O2
0.197
0.0267
407.1340
[M-H]⫺
5
C20H23O9 / C16H19N6O7 / C17H15N10O3 / C3H7N26
1.828
0.0083
225.1485
[M-H]⫺
15
C13H21O3
4.951
0.0441
315.1240
[M-H]⫺
5
C18H19O5 / C19H15N4O1
0.082
0.0354
391.1383
[M-H]⫺
5
C20H23O8 / C16H19N6O6 / C17H15N10O2
3.866
0.0400
223.1344
[M-H]⫺
15
C13H19O3
⫺1.750
0.0224
143.1073
[M-H]⫺
15
C8H15O2
3.493
0.0094
311.2224
[M-H]⫺
10
C18H31O4/ C14H27N6O2 / C19H27N4
1.103
0.0485
157.1240
[M-H]⫺
20
C9H17O2
⫺3.990
0.0138
315.2532
[M-H]⫺
10
C18H35O4/ C14H31N6O2 / C19H31N4
2.718
0.0343
243.1966
[M-H]⫺
10
C14H27O3
0.025
0.0257
171.1391
[M-H]⫺
15
C10H19O2
⫺0.168
0.0146
291.1952
[M-H]⫺
10
C14H23N6O1 /C18H27O3/ C13H27N2O5
ⴚ6.599 2nd(2.623)
0.0097 2nd(0.0106)
295.2271
[M-H]⫺
10
C14H27N6O1 /C18H31O3
⫺8.209 2nd(0.887)
0.0070 2nd(0.0100)
521
CHAPTER | 56 Analytical Determination of Polyphenols in Olive Oil
TABLE 56.5 (Continued) m/z experimental
Selected ion
Tolerance (ppm) List of possibilities in in generate Molecular generate Molecular Formula Formula (in increasing order of sigma)
Error (ppm) (for the first compound)
Sigma value (for the first compound)
297.2427
[M-H]⫺
15
C14H29N6O1 / C18H33O3
⫺10.409 2nd(⫺1.375)
0.0124 2nd(0.0178)
205.1599
[M-H]⫺
20
C14H21O1
ⴚ0.463
0.0192
199.1700
[M-H]⫺
10
C12H23O2
1.837
0.0182
299.2594
[M-H]⫺
10
C18H35O3 / C14H31N6O1
0.607
0.0095
Although excellent progress has already been made it is to be hoped that the use of different potent techniques coupled with rapid, reliable, sophisticated detectors will become more common in the near future. There are still many unknown compounds present in the polar fraction of olive oil and it is very important that the scientific community should join efforts to carry out collaborative studies into this subject.
SUMMARY POINTS ●
●
●
●
●
A reliable determination of the phenolic compounds in virgin olive oil is not always straightforward because no official method exists, the availability of suitable reference standards for identification and quantification is limited and the compounds themselves are often present only in low concentrations. It is important to determine phenols in virgin olive oil since they are related to the resistance of the oil to oxidation, to some of its sensory properties and to several health-giving attributes. The separation techniques (GC, HPLC and CE) have to be coupled to powerful detectors such as MS and NMR to improve these kinds of study. These techniques should be considered as complementary. MS and NMR represent a good alternative even if a previous analytical technique is not used (directinfusion approaches). The phenolic fraction of olive oil has been studied for many years but there are still a great number of as yet unknown compounds present in its polar fraction and collaborative studies should be carried out by the
●
scientific community to discover their identity and characteristics. The scientific community has to try not only to study this fraction exhaustively but also to discover what effect these compounds might have on humans: What is the activity of any particular compound in vivo? What do we know about its metabolism? Might that compound be toxic at a certain concentration? And so on.
REFERENCES Andreoni, N., Fiorentini, R., 1995. Determinazione di composti fenolici in oil di oliva. Riv. Ital. Sost. Grasse 72, 163–164. Angerosa, F., D’Alessandro, N., Konstantinou, P., Di Giacinto, L., 1995. GC-MS evaluation of phenolic compounds in virgin olive oil. J. Agric. Food Chem. 43, 1802–1807. Aturki, Z., Fanali, S., D’Orazio, G., Rocco, A., Rosati, C., 2008. Analysis of phenolic compounds in extra virgin olive oil by using reversed-phase capillary electrochromatography. Electrophoresis 29, 1643–1650. Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gómez-Caravaca, A.M, Segura-Carretero, A., Fernández-Gutiérrez, A., Lercker, G., 2007. Phenolic molecules in virgin olive oils: a survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules 12, 1679–1719. Blekas, G., Psomiadou, E., Tsimidou, M., Boskou, D., 2002. On the importance of total polar phenols to monitor the stability of Greek virgin olive oil. Eur. J. Lipid Sci. 104, 340–346. Brenes, M., García, A., García, P., Garrido, A., 2000. Rapid and complete extraction of phenols from olive oil and determination by means of a coulometric electrode array system. J. Agric. Food Chem. 48, 5178–5183. Capasso, R., 1999. A review on the electron ionization and fast atom bombardment mass spectrometry of polyphenols naturally occurring in olive wastes and some of their synthetic derivatives. Phytochem. Anal. 10, 299–306.
522
Carrasco-Pancorbo, A., Cerretani, L., Bendini, A., Segura-Carretero, A., Gallina-Toschi, T., Fernández-Gutiérrez, A., 2005. Analytical determination of polyphenols in olive oils. J. Sep. Sci. 28, 837–858. Carrasco-Pancorbo, A., Gómez-Caravaca, A.M., Cerretani, L., Bendini, A., Segura-Carretero, A., Fernández-Gutiérrez, A., 2006. A simple and rapid electrophoretic method to characterize simple phenols, lignans, complex phenols, phenolic acids, and flavonoids in extravirgin olive oil. J. Sep. Sci. 29, 2221–2233. Carrasco-Pancorbo, A., Neusüß, C., Pelzing, M., Segura-Carretero, A., Fernández-Gutiérrez, A., 2007. CE- and HPLC-TOF-MS for the characterization of phenolic compounds in olive oil. Electrophoresis 28, 806–821. Cerretani, L., Bendini, A., Biguzzi, B., Lercker, G., Gallina Toschi, T., 2003. Evaluation of the oxidative stability of extra-virgin olive oils, obtained by different technological plants, with respect to some qualitative parameters [Stabilità ossidativa di oli extravergini di oliva ottenuti con diversi impianti tecnologici]. Industrie Alimentari 427, 706–711. Christophoridou, S., Dais, P., 2006. Novel approach to the detection and quantification of phenolic compounds in olive oil based on P-31 nuclear magnetic resonance spectroscopy. J. Agric. Food Chem. 54, 656–664. Christophoridou, S., Dais, P., Tseng, L.H., Spraul, M., 2005. Separation and identification of phenolic compounds in olive oil by coupling high-performance liquid chromatography with postcolumn solid-phase extraction to nuclear magnetic resonance spectroscopy (LC-SPENMR). J. Agric. Food Chem. 53, 4667–4679. Cortesi, N., Azzolini, M., Rovellini, P., Fedeli, E., 1995a. I componenti minori polari degli oli vergini di oliva: Ipotesi di struttura mediante LC-MS. Riv. Ital. Sost. Grasse 72, 241–251. Cortesi, N., Azzolini, M., Rovellini, P., Fedeli, E., 1995b. Minor polar components of virgin olive oils: a hypothetical structure by LC-MS. Riv. Ital. Sost. Grasse 72, 241–251. De Nino, A., Mazzotti, F., Perri, E., Procopio, A., Raffaelli, A., Sindona, G., 2000. Virtual freezing of hemiacetal-aldehyde equilibrium of the aglycones of oleuropein and ligstroside present in olive oils from Carolea and Coratina cultivars by ionspray ionization tandem mass spectrometry. J. Mass Spectrom. 35, 461–467. Del Carlo, M., Sacchetti, G., Di Mattia, C., Compagnone, D., Mastrocola, D., Liberatore, L., Cichelli, A., 2004. Contribution of the phenolic fraction to the antioxidant activity and oxidative stability of olive oil. J. Agric. Food Chem. 52, 4072–4079. Favati, F., Carporale, G., Bertuccioli, M., 1994. Rapid determination of phenol content in extra virgin olive oil. Grasas Aceites 45, 68–70. Forcadell, M.L., Comas, M., Miquel, X., de la Torre, MC., 1987. Tyrosol and hydroxy-tyrosol determination in virgin olive oils. Rev. Fr. Corps Gras. 34, 547–549. Frazier, R.A., 2001. Recent advances in capillary electrophoresis methods for food analysis. Electrophoresis 22, 4197–4206. Gariboldi, P., Jommi, G., Verotta, L., 1986. Secoiridoids from oleaeuropaea. Phytochemistry 25, 865–869. Gerothanassis, I.P., Exarchou, V., Lagouri, V., Troganis, A., Tsimidou, M., Boskou, D., 1998. Methodology for identification of phenolic acids in complex mixtures by high-resolution two-dimensional nuclear magnetic resonance. Application to methanolic extracts of two oregano species. J. Agric. Food Chem. 46, 4185–4192. Gutfinger, T., 1981. Polyphenols in olive oils. J. Am. Oil Chem. Soc. 58, 966–968. Gutiérrez, F., Albi, M.A., Palma, R., Ríos, J.J., Olías, J.M., 1989. Bitter taste of virgin olive oil – Correlation of sensory evaluation and instrumental HPLC analysis. J. Food Sci. 54, 68–70.
SECTION | I Natural Components
Hrncirik, K., Fritsche, S., 2004. Comparability and reliability of different techniques for the determination of phenolic compounds in virgin olive. Eur. J. Lipid Sci. Technol. 106, 540–549. Janer del Valle, C., Vázquez-Roncero, A., 1980. Estudio de los componentes polares del aceite de oliva por cromatografia gaseosa. Grasas Aceites 5, 309–316. Liberatore, L., Procida, G., D`Alessandro, N., Cichelli, A., 2001. Solidphase extraction and gas chromatographic analysis of phenolic compounds in virgin olive oil. Food Chem. 73, 119–124. Limiroli, R., Consonni, R., Ottolina, G., Marsilio, V., Bianchi, G., Zetta, L., 1995. H-1 and C-13 NMR characterization of new oleuropein aglycones. J. Chem. Soc-Perkin. 1, 1519–1523. Maestro-Durán, R., Borja, R., Martín, A., Fiestas, J.A., Alba, J., 1991. Biodegradación de los compuestos fenólicos presentes en el alpechín. Grasas Aceites 42, 271–276. Mateos, R., Espartero, J.L., Trujillo, M., Ríos, J.J., León-Camacho, M., Alcudia, F., Cert, A., 2001. Determination of phenols, flavones, and lignans in virgin olive oils by solid-phase extraction and highperformance liquid chromatography with diode array ultraviolet detection. J. Agric. Food Chem. 49, 2185–2192. Montedoro, G.F., Servili, M., Baldioli, M., Miniati, E., 1992a. Simple and hydrolyzable phenolic-compounds in virgin olive oil .1. Their extraction, separation, and quantitative and semiquantitative evaluation by HPLC. J. Agric. Food. Chem. 40, 1571–1576. Montedoro, G.F., Servili, M., Baldioli, M., Miniati, E., 1992b. Simple and hydrolyzable phenolic compounds in virgin olive oil. 2. Initial characterization of the hydrolyzable fraction. J. Agric. Food. Chem. 40, 1577–1580. Montedoro, G., Cantarelli, C., 1969. Investigation of olive oil phenolic components. Riv. Ital. Sost. Grasse 46, 3–12. Montedoro, G.F., Servili, M., Baldioli, M., Selvaggini, R., Miniati, E., Macchioni, A., 1993. Simple and hydrolyzable compounds in virgin olive oil. 3. Spectroscopic characterizations of the secoiridoid derivatives. J. Agric. Food Chem. 41, 2228–2234. Owen, R.W., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2000. Phenolic compounds and squalene in olive oils: the concentration and antioxidant potential of total phenols, simple phenols, secoiridoids, lignans and squalene. Food Chem. Toxicol. 38, 647–659. Pirisi, F.M., Cabras, P., Falqui Cao, C., Migliorini, M., Muggelli, M., 2000. Phenolic compounds in virgin olive oil. 2. Reappraisal of the extraction, HPLC separation, and quantification procedures. J. Agric. Food Chem. 48, 1191–1196. Rios, J.J., Gil, M.J., Gutiérrez-Rosales, F., 2005. Solid-phase extraction gas chromatography-ion trap-mass spectrometry qualitative method for evaluation of phenolic compounds in virgin olive oil and structural confirmation of oleuropein and ligstroside aglycons and their oxidation products. J. Chromatogr. A 1093, 167–176. Sacchi, R., Addeo, F., Paolillo, L., 1997. 1H and 13C NMR of virgin olive oil. An overview. Magn. Reson. Chem. 35, S5133–S5145. Servili, M., Baldioli, M., Selvaggini, R., Miniati, E., Macchioni, A., Montedoro, G., 1999. High-performance liquid chromatography evaluation of phenols in olive fruit, virgin olive oil, vegetation waters, and pomace and 1D-and 2D-nuclear magnetic resonance characterization. J. Am. Oil Chem. Soc. 76, 873–882. Shahidi, F., Naczk, M., 1995. In: Food Phenolic Sources. Chemistry. Effects. Applications. (ed.) Technomic Publishing Co. Inc., Lancenser, PA, USA, pp. 235–277. Singleton, V.L., Rossi, J.A., 1965. Colorimetry of total phenolics with phosphomolybdic-phosphotungstic acid reagents. Am. J. Enol. Viticult. 16, 144–158.
CHAPTER | 56 Analytical Determination of Polyphenols in Olive Oil
Solinas, M., 1987. HRGC analysis of phenolic components in virgin olive oils in relation to the ripening and the variety of olives. Riv. Ital. Sost. Grasse. 64, 255–262. Solinas, M., Cichelli, A., 1981. Sulla determinazione delle sostanze fenoliche dell’olio di oliva. Riv. Ital. Sost. Grasse. 58, 159–164. Solinas, M., Cichelli, A., 1982. Determination of phenolic substances in olive oil by GLC and HPLC; possible role of tyrosol in determination of the quantity of virgin oil in blends with refined. Riv. Soc. Ital. Sci. Aliment. 11, 223–230. Tasioula-Margari, M., Okogeri, O., 2001. Isolation and characterization of virgin olive oil phenolic compounds by HPLC/UV and GC-MS. J. Food Sci. 66, 530–538.
523
Tsimidou, M., Lytridou, M., Boskou, D., Pappa-Louis, A., Kotsifaki, F., Petrakis, C., 1996. On determination of minor phenolic acids of virgin olive oil by RP-HPLC. Grasas y Aceites 47, 151–157. Tsimidou, M., Papadopoulos, G., Boskou, D., 1992. Determination of phenolic compounds in virgin olive oil by reversed-phase HPLC with emphasis on UV detection. Food Chem. 44, 53–60. Vázquez Roncero, A., 1978. Les polyphenols de l’huile d’olive et leur influence sur les characteristiques de l’huile. Rev. Fr. Corps Gras. 25, 21–26. Vlahov, G., 1999. Application of NMR to the study of olive oils. Prog. Nucl. Magn. Reson. Spectrosc. 35, 341–357.
This page intentionally left blank
Chapter 57
Electronic Tongues Purposely Designed for the Organoleptic Characterization of Olive Oils María L. Rodríguez-Méndez1, C. Apetrei2 and José A. De Saja3 1
Dpt. Inorganic Chemistry, E. T. S. Ingenieros Industriales, University of Valladolid, Spain Dpt. Chemistry, Faculty of Sciences, ‘Duna˘rea de Jos’ University of Galati, Romania 3 Dpt. Condensed Matter Physics, Faculty of Sciences, University of Valladolid, Spain
2
57.1 INTRODUCTION Analysis of odor and taste in food has traditionally been addressed either by a trained sensory panel or using chemical methods (i.e. gas-chromatography). Unfortunately, these techniques are generally time-consuming, costly and labor-intensive. The food industry is demanding new methodologies for objective and automated measurements that can characterize odors and tastes with sufficient accuracy and reproducibility. During the last twenty years, the concept of an ‘electronic nose’ has been developed (Rock et al., 2008). These instruments provide a signal which is characteristic of the mixture of volatile compounds that form the headspace of a certain sample. For this reason, their main use is not the identification and quantification of the individual components of a sample. Instead, they perform a comparison of volatile patterns as the human nose does. Electronic noses are used in the food industry for odor analysis to evaluate the quality of a variety of products including wine, fish, meat, beer, milk, water, etc. (Rock et al., 2008). Electronic nose devices consist of three different blocks (Figure 57.1, Table 57.1): the sampling system, the sensory system and the pattern recognition system. The first block is devoted to sampling volatiles and its main purpose is to collect a representative headspace of the samples. The second block of the system is the multisensor array, a series of non-specific sensors with cross-sensitivity that respond to a wide variety of compounds. Finally, the third block concerns pattern recognition methods that are used to process the signals provided by the array of sensors. The success of electronic noses for the analysis of gases led researchers to develop an array of sensors able Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
to work in liquid surroundings. Such systems were called taste sensors or electronic tongues (Toko, 2000; Vlasov et al., 2005). Electronic tongues are defined as an array of non-specific chemical sensors with partial sensitivity (cross-sensitivity) to different components, able to analyze complex liquids. The sensor array produces signals which are not necessarily specific for any particular species in the liquid, but a signal pattern is generated which can be related to certain features or qualities of the sample using the appropriate software. Electronic tongues are normally used to give qualitative answers about the sample studied and only in some cases to predict the concentration of individual species in the sample.
57.2 ELECTRONIC TONGUES As described in previous paragraphs, the basic principle behind an electronic tongue is to analyze the signals obtained from an array of non-specific overlapping sensors with pattern recognition routines. So, it can be considered that electronic tongues consist of two different blocks: the first is formed by the array of sensors, the second by the data-processing techniques.
57.2.1 Multisensor Systems The heart of any electronic tongue is the multisensor system. Much effort has gone into developing new sensors with improved characteristics. Sensors used in electronic tongues can use several measurement principles including mass detection (Hossenlopp, 2006), optical transduction (Sohn et al., 2005) or electrochemical
525
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
526
SECTION | I Natural Components
FIGURE 57.1 Electronic nose vs. human nose.
TABLE 57.1 Definitions of several specific terms in electronic tongues. 1. Organoleptic testing involves inspection through visual examination, feeling and smelling of products. It is carried out by a panel of trained experts 2. An electronic tongue is an instrument which comprises an array of chemical sensors with partial specificity and an appropriate pattern recognition system, capable of recognizing simple and complex odors. It has the advantage of objectivity 3. A sensor is a device that measures a physical quantity and converts it into a signal which can be read by an observer or by an instrument 4. An array of sensors is a set of sensors that are non-specific but that show cross-selectivities. When the array is immersed in a certain sample, all the sensors respond, but each sensor provides a different response. The response of all the sensors forming the array is a fingerprint of a particular sample 5. A voltammetric sensor is a class of electrochemical sensor that obtains information of the system by measuring the current as the potential is varied 6. Polyphenols are a group of chemical substances found in plants, characterized by the presence of more than one phenol unit or building block per molecule 7. Pattern recognition is a branch of artificial intelligence concerned with the classification or description of observations
sensors. In particular, electrochemical sensors (voltammetric, impedimetric or potentiometric) are the most widely used sensing units (Legin et al., 1997; Arrieta et al., 2003; Ferreira et al., 2003; Winquist et al., 2005; Gutes et al., 2007). Most of the works in this field involved signal generation from potentiometric sensors where the potential value created by the diffusion of ions across a membrane is measured. These membranes can be prepared from different materials, which provide sensors with different selectivities. A search for ionophores that can recognize specific ions and obtain new or improved selectivity for different ions is currently in progress (Legin et al., 1997; Toko, 2000; Vlasov et al., 2005; Gutes et al., 2007). Potentiometric sensors show a relatively fast response, reasonable selectivity through the adequate selection of the active membrane material and a wide dynamic linear range. However, potentiometric sensors are limited to the detection of ions or charged species. Impedance spectroscopy has also been used as a transduction method to detect molecules responsible for basic tastes including bitterness and pungency (Ju et al., 2003). Electrodes covered with various organic materials (including conducting polymers, perylenes, phthalocyanines or carbon nanotubes) have been used to analyze complex solutions such as wines (Ferreira et al., 2003). Amperometric and voltammetric sensors have attracted considerable attention during recent years (Arrieta et al., 2003; Winquist et al., 2005; Parra et al., 2006). Using an array of amperometric sensors working at different potentials, it is possible to discriminate between several electroactive compounds, since, at low positive/negative potential, only strong reducing/oxidizing compounds may be detected.
CHAPTER | 57 Electronic Tongues Purposely Designed for the Organoleptic Characterization of Olive Oils
The modification of the electrode surface (i.e. with conducting polymers) can provide a range of electrodes with different selectivity (Nguyen et al., 1998). Amperometric biosensors take advantage of the specificity enzyme substrate. They consist of an enzyme mounted in a solid electrode. Biosensors have been used to detect specific substances and tastes. Glucose and ascorbic acid have been detected in fruit juices using an array of biosensors based on glucose oxidase (Gutes et al., 2006). In voltammetric sensors, a bias voltage is applied while the current is measured. Voltammograms show redox peaks associated to the oxidation and reduction of the molecules present in the solution. The intensity and position of the peaks depend on the experimental conditions (pH, ionic strength) and on the nature of the working electrode. In addition, different excitation functions can be applied (i.e. cyclic voltammetry, pulse voltammetry, square wave voltammetry). Depending on the technique used, different information can be obtained from the tested solution.
57.2.1.1 Electronic Tongues Based on Voltammetric Sensors The first electronic tongue based on voltammetry consisted of a number of different working electrodes made from different metals in a standard three-electrode configuration (Ivarsson et al., 2001; Winquist et al., 2005). Voltammetric A 400
I/μA
200 100
I
0 –100 –200 –1.0
–0.5
0.0
measurements are carried out using pulsed voltammetry. Metallic voltammetric electrodes have also been used by other groups to discriminate phenolic derivatives, wines and teas (Buratti et al., 2004; Gutes et al., 2006). Even if the sensitivity of metallic voltammetric electrodes is high, their selectivity is poor in some cases because (i) only electroactive compounds can be detected and (ii) electroactive compounds can undergo redox processes at similar potentials. The performance of a voltammetric multisensor system can be improved by using chemically modified electrodes, in particular when the sensing units are modified with electroactive materials. Voltammograms show peaks of two different origins: peaks associated with the oxidationreduction of the analytes present in the solution and transient responses associated with the electrode material (Arrieta et al., 2003; Parra et al., 2006). All these redox processes and interactions give rise to rich voltammograms with a high degree of selectivity. This is illustrated in Figure 57.2, where the response of an electrode chemically modified with a gadolinium bisphthalocyanine (GdPc2) towards different solutions containing antioxidants commonly found in foods and beverages is shown. As observed in Figure 57.2A, the voltammogram of a GdPc2 electrode immersed in KCl shows two redox pairs corresponding to the reversible reduction (peak I) and oxidation (peak II) of the phthalocyanine ring. When the electrode is immersed in solutions containing antioxidants, new peaks are observed B 500 400 300 200 100 0 –100 –200 –300 –400
II
300
0.5
1.0
527
–1.0
II I
–0.5
0.0
0.5
1.0
E/V vs Ag/AgCI C 400
D 300 II
300 200 I/μA
II
200 100
I
I
100 0
0
–100
–100
–200
–200 –1.0
–0.5
0.0
0.5
1.0
–1.0
–0.5
0.0
0.5
1.0
E/V vs Ag/AgCI FIGURE 57.2 Cyclic voltammetry of a GdPc2 modified electrode immersed in a solution of (A) 10−2 mol L−1 KCl and 10−3 mol L−1 of (B) vanillic acid; (C) pyrogallol; (D) ascorbic acid and catechin.
528
associated to the redox compounds present in the solution. In addition remarkable shifts of peak II are observed. These displacements to higher potentials are caused by the antioxidant effect of the solution. Different families of materials have been used as modifiers, including phthalocyanines (Arrieta et al., 2003), conducting polymers deposited electrochemically and doped with a variety of doping agents (Arrieta et al., 2004) and perylenes (Parra et al., 2004). As illustrated in Figure 57.3, the use of different sensitive materials with complementary reactivity greatly improves the cross-selectivity of the array and hence its capability of discrimination. Hybrid arrays formed by phthalocyanines, perylenes and conducting polymers have been successfully used to discriminate between wines with different organoleptic characteristics (Parra et al., 2006).
57.2.2 Multivariate Data Treatment Sensors produce signals which are not necessarily specific for any particular species. Instead, the pattern of responses generated is a fingerprint of the sample studied. This pattern can be related with certain features or characteristics of the samples by means of chemometrics. The pattern recognition techniques consist of four sequential stages: signal pre-processing, dimensionality reduction, prediction and validation (Beebe et al., 1998). The signal pre-processing prepares the feature vector for future processing. It includes compensation for sensor drift, scaling of the data and extracting representative parameters. A dimensionality reduction stage projects this initial feature
SECTION | I Natural Components
onto a lower dimensional space. This is usually done using a non-supervised technique such as principal component analysis (PCA). Using PCA, it is possible to discriminate between samples with different characteristics. The resulting low-dimensional feature vector is the one used to solve a given prediction problem, typically classification, regression or clustering. Classification tasks address the problem of identifying an unknown sample and to assign it to a certain set of previously learned categorized samples. Typical classification models used in electronic tongues are linear discriminant analysis (LDA) or artificial neural networks (ANN). In regression tasks, the goal is to establish a predictive model from a set of independent variables (e.g. sensor responses) to a second set of variables that are the properties of the sample analyzed (e.g. concentration, quality). Finally, in a clustering task, the goal is to learn the structural relationships between different samples. A final step, sometimes overlooked, is the selection of models and the estimation of the true error rates for a trained model by means of validation techniques.
57.3 ELECTRONIC TONGUES DEDICATED TO THE ANALYSIS OF OLIVE OILS Electronic tongues have demonstrated their capabilities to analyze and discriminate model solutions of basic tastes and a variety of foods such as mineral waters, milks, wines or beers. However, few works have been devoted to the analysis of olive oils. The main reason is the difficulty of carrying out electrochemical analysis in a non-conductive liquid with a high viscosity. To avoid the difficulties of electrochemical measurements in such a complex media, several solutions have been proposed.
57.3.1 Analysis of Simple Solutions: Basic Tastes and Antioxidants
FIGURE 57.3 Hybrid array of four electrodes based on (A) bisphthalocyanine, (B) perylene, (C) polypyrrole doped with ferrocyanide, (D) polypyrrole doped with toluensulfonic acid, immersed in 0.1 mol L⫺1 KCl. The array is formed by different sensing materials that give different responses when immersed in the same liquid.
Due to the complexity of olive oils, several works have analyzed simple solutions where compounds producing different tastes were solved (acids such as acetic or citric acid to produce sourness; NaCl and KCl to produce saltiness; MgCl2, quinine or caffeine to produce bitterness; sweetness produced by sucrose, glucose or aspartame; the last is umami that is produced by monosodium glutamate). The possibility to detect and discriminate bitterness and pungency is of particular interest for the olive oil industry. The first attempts to evaluate bitterness and pungency with an electronic tongue were carried out using sensors based on lipid membranes (Ju et al., 2003). The response produced by the sensors is related to the chemical properties and/or structure of the molecules analyzed. For instance, bitter-tasting amino acids such as tryptophan showed electric response patterns similar to a typical bitter substance, quinine.
CHAPTER | 57 Electronic Tongues Purposely Designed for the Organoleptic Characterization of Olive Oils
Electronic tongues have also been used for the detection and discrimination of phenolic compounds which are natural antioxidants present in virgin olive oils. They play an important role in their quality, since they contribute significantly to their stability towards oxidation (Mateos et al., 2003). In addition, polyphenols are the main contributors to bitterness, astringency and pungency of olive oils (Gutierrez et al., 2000). Due to their redox reactivity, the presence of such compounds can be detected by means of electrochemical methods (Campanella et al., 1999). An array formed by voltammetric electrodes modified with electroactive substances was used to detect model solutions of bitterness (Apetrei et al., 2004). Substances under study included MgCl2, quinine, and four phenolic compounds: the monoaldehydic and dialdehydic forms of the oleuropein and of the ligstroside (called MAOL, DAOL, DALI and MALI respectively). The phenolic compounds were extracted from olive oils and were selected because they are the main compounds responsible for the bitterness detected in olive oils (Gutierrez et al., 2000). As expected, the voltammetric responses depend on the chemical nature of the tastan. The principal component analysis of the signals obtained has allowed the clear discrimination of the bitter solutions according to their chemical nature (Figure 57.4). Arrays formed by biosensors have also been used to detect mixtures of polyphenols by using polyphenol oxidase enzyme biosensors that oxidize the phenolic compounds into their corresponding quinones. Artificial neural networks (ANN) were used for the extraction and quantification of each compound. Good prediction ability was attained, so the separate quantification of these three phenols was accomplished (Gutes et al., 2005).
529
57.3.2 Analysis of Polyphenolic Extracts An array of voltammetric electrodes chemically modified with electroactive materials (five phthalocyanines and six conducting polymers) has been used to analyze the phenolic fraction obtained from six extra virgin olive oils (Table 57.2) (Rodríguez-Mendez et al., 2008). As usually happens in voltammetric electrodes, the voltammograms obtained are complex and consist of peaks related to the electrochemical activity of the phenolic fraction under study, and redox peaks associated with the electroactive material. In addition, the antioxidant activity of polyphenols drastically influences the electrochemical behavior of the electrodic material. This complexity produced signals specific for each type of oil. Principal component analysis has demonstrated the capability of the array of electrodes to discriminate between olive oils according to their phenolic content and bitterness index (Figure 57.5). A correlation between the anodic peak potentials and the polyphenolic TABLE 57.2 Results of the chemical analysis of the six olive oils under study. Olive oil
Polyphenols (mg kg⫺1)
Bitterness index
Bitterness panel of experts
S1
475.36
2.61
3.2
S2
433.43
2.98
3.3
S3
403.06
3.23
3.33
S4
593.17
3.92
4.05
S5
856.79
5.71
4.6
S6
990.25
6.38
4.75
The polyphenolic content obtained by chemical methods (the Folin-Ciocalteau method) is well correlated with the bitterness index measured by chemical methods (Gutierrez et al., 2000) and the score given by a panel of experts.
FIGURE 57.4 PCA score plot of the response of the array of sensors towards bitter solutions. (1) Quinine; (2) MgCl2; (3) DALI – dialdehydic form of the ligstroside; (4) MAOL – monoaldehydic form of the oleuropein; (5) DAOL – dialdehydic form of the oleuropein; (6) MALI – monoaldehydic form of the ligstroside. The signals coming from the sensors can be treated using statistical tools. They permit to discriminate between bitter solutions.
FIGURE 57.5 PCA score plot of the measurements of the six olive oils with a hybrid sensor array. The electronic tongue allows discrimination of olive oils with different organoleptic characteristics.
530
content obtained by the Folin-Ciocalteau method was r2 ⫽ 0.9989. Two amperometric biosensors based on tyrosinase or peroxidase have successfully been applied to the detection of polyphenols in olive oils. The device includes a carbon working electrode, a carbon counter electrode, and a silver pseudo-reference electrode. The biosensors are built up onto disposable screen-printed carbon electrodes using a planar three-electrode strip. Such electrodes show different specificities towards different groups of phenolic compounds affecting bitterness and pungency (Busch et al., 2006).
57.3.3 Fabrication of Electrodes Modified with Oils An original method based on carbon paste electrodes (CPE) has been developed to discriminate oils of different origins and qualities. In this method, the carbon paste is prepared using the oil as a binder (Apetrei et al., 2005). The features observed in the voltammograms reflect the electroactive properties of the oils inside the carbon paste. In this way, the different content in polyphenols could be used to discriminate olive oil from sunflower oil or corn oil. The capability of discrimination is improved by immersing the CPEs in different electrolytic solutions. Changing the nature of the solution, (pH, nature and concentration of ions, etc.), the interactions between the electrode and the solution also change (Figure 57.6). Due to the different chemical composition of each oil, a variety of signals is obtained. Using PCA, this method has allowed a clear discrimination between olive oils and seed oils. Moreover, olive oils of different qualities have also been distinguished. In a recent work, nine olive oils (Table 57.3) with different degrees of bitterness have been analyzed using this method (Apetrei et al., 2007). The system has permitted
FIGURE 57.6 Cyclic voltammetry curves of an ordinary olive oil-based CPE immersed in 0.1 mol L⫺1 aqueous solutions. (A) KCl; (B) phosphate buffer pH ⫽ 4; (C) HCl. Electrodes fabricated with different oils produce different electrochemical signals due to their different chemical composition.
SECTION | I Natural Components
not only to discriminate but also to classify olive oils with different organoleptic characteristics. Using partial least squares discriminant analysis (PLS-DA), it has been possible to classify the olive oils in three groups according to their degree of bitterness (Figure 57.7). In addition, good correlations between redox processes observed in the electrodes and the analytical and sensorial characteristics of the virgin olive oils under study have been found. A correlation of 0.9205 was obtained with an RMSEC (root mean square error of correlation) of 0.324, which illustrates the correlations found between the bitterness measured using the sensors and the bitterness index given by a panel of experts.
57.3.4 Combination of an Electronic Nose and an Electronic Tongue Flavor perception is based on two components: taste and aroma. Flavor could be analyzed using a combination of an electronic nose and an electronic tongue. Several attempts have been made to combine e-noses and e-tongues and even with e-eyes (Rodríguez-Méndez et al., 2004). It has been demonstrated that the simultaneous use of electronic noses and tongues can increase the amount of information extracted from a certain sample. A combination of an electronic nose (based on MOS sensors) and an electronic tongue (based on amperometric sensors) has been used to evaluate the oxidation of extra virgin olive oils which was considered at different storage periods and conditions (Cosio et al., 2007).
57.4 FINAL REMARKS As a final remark, it is necessary to point out that the ‘electronic tongue’ does not exactly mimic the human system. In general, an electronic tongue bases its evaluation upon the analysis of chemical compounds including taste and tasteless species. In contrast, the human gustatory sense detects substances that possess taste, and also perceives mouthfeel (astringency, heat, viscosity, etc.) that also contributes to the perception. Thus, it is mandatory to keep in mind that an absolute taste description is not possible from an electronic tongue point of view. However, the advantages of these electronic systems are clear: they allow important substances that do not produce taste to be detected; they provide an objective response that does not depend on physiological conditions or personal preferences; they do not show fatigue (as human senses do) and are an alternative to the expensive human panel tests.
SUMMARY POINTS ●
Electronic tongues are one of the most rapidly emerging and exciting fields of non-classical analytics.
531
CHAPTER | 57 Electronic Tongues Purposely Designed for the Organoleptic Characterization of Olive Oils
TABLE 57.3 Bitterness of the nine extra virgin olive oils under study. Sensory attribute
S1
S2
S3
S4
S5
S6
S7
S8
S9
Bitterness measured by a panel of experts
0.55
0.7
1.05
1.39
2.0
2.35
2.22
2.88
2.83
Bitterness index measured by chemical methods
0.78
0.23
1.25
0.94
1.50
1.64
1.72
2.17
1.73
A scale of 1–5 was used to determine the intensity of the bitterness: 0–1 indicates imperceptible; 1–2 indicates slight; 2–3 indicates moderate; 3–4 indicates strong; 4–5 indicates very strong.
ACKNOWLEDGMENTS Financial support from CICYT (Grant nº. AGL2006-05501/ALI) and the Junta de Castilla y León (VA-052A06) is gratefully acknowledged.
REFERENCES
FIGURE 57.7 PLS-DA scores plot corresponding to the classification of oils according to their degree of bitterness. The electronic tongue is able to classify olive oils according to their degree of bitterness.
●
●
●
●
●
●
The use of electronic tongues for the analysis of olive oils has a degree of difficulty due to the high viscosity of oils and their low conductivity. This makes it necessary to design specific systems for this particular application. Sensor arrays have been used to provide information about simple solutions (molecules responsible for bitterness, pungency, and polyphenols). Electronic tongue devices have been able to correlate the electrochemical responses with the polyphenolic content measured by chemical methods. Specific sensors have been developed where oils have been inserted in carbon paste electrodes. Using this innovative approach, it has been possible to discriminate and classify oils of a different nature and quality. Good correlations with chemical analysis and human perceptions have been attained. According to these results, electronic tongue methods can provide useful information for qualitative and quantitative analysis of olive oils. New developments are expected in the near future. They will be mainly related to the development of new sensors with a higher degree of specificity and stability and to the combination of electronic noses and tongues.
Apetrei, C., Rodríguez-Méndez, M.L., Parra, V., Gutierrez, F., de Saja, J.A., 2004. Array of voltammetric sensors for the discrimination of bitter solutions. Sens. Actuators B 103, 145–152. Apetrei, C., Rodríguez-Méndez, M.L., de Saja, J.A., 2005. Modified carbon paste electrodes for discrimination of vegetable oils. Sens. Actuators B 111-112, 403–409. Apetrei, C., Gutierez, F., Rodríguez-Méndez, M.L., de Saja, J.A., 2007. Novel method based on carbon paste electrodes for the evaluation of bitterness in extra virgin olive oils. Sens. Actuators B 121, 567–575. Arrieta, A.A., Rodríguez-Méndez, M.L., de Saja, J.A., 2003. LangmuirBlodgett film and carbon paste electrodes based on phthalocyanines as sensing units for taste. Sens. Actuators B 95, 357–365. Arrieta, A.A., Apetrei, C., Rodríguez-Méndez, M.L., de Saja, J.A., 2004. Voltammetric sensor array based on conducting polymermodified electrodes for the discrimination of liquids. Electrochim. Acta 49, 4543–4551. Beebe, K.R., Pell, R., Seasholtz, M.B., 1998. Chemometrics: A Practical Guide. Wiley Interscience, New York. Buratti, S., Benedetti, S., Scampicchio, M., Pangerod, E.C., 2004. Characterisation and classification of Italian Barbera wines by using an electronic nose and amperometric electronic tongue. Anal. Chim. Acta 525, 133–139. Busch, J.L., Hrncirik, K., Bulukin, E., Boucon, C., Mascini, M., 2006. Biosensor measurements of polar phenolics for the assessment of the bitterness and pungency of virgin olive oil. J. Agric. Food Chem. 54, 4371–4377. Campanella, L., Favero, G., Pastorino, M., Tomassetti, M., 1999. Monitoring the rancidification process in olive oils using a biosensor operating in organic solvents. Biosens. Bioelectron. 14, 179–186. Cosio, M.S., Ballabio, D., Benedetti, S., Gigliotti, C., 2007. Evaluation of different storage conditions of extra virgin olive oils with an innovativerecognition tool built by means of electronic nose and electronic tongue. Food Chem. 101, 485–491. Ferreira, M., Riul, A., Wohnrath, K., Fonseca, F.J., Oliveira, O.N., Mattoso, L.H., 2003. High-performance taste sensor made from
532
Langmuir-Blodegtt films of conducting polymers and a ruthenium complex. Anal. Chem. 75, 953–955. Gutes, A., Cespedes, F., Alegret, S., del Valle, M., 2005. Determination of phenolic compounds by a polyphenol oxidase amperometric biosensor and artificial neural network analysis. Biosens. Bioelectron. 20, 1668–1673. Gutes, A., Ibañez, A.B., del Valle, M., Cespedes, F., 2006. Automated SIA e-tongue employing a voltammetric biosensor array for the simultaneous determination of glucose and asorbic acid. Electroanalysis 18, 82–88. Gutes, A., Cespedes, F., del Valle, M., 2007. Electronic tongues in flow analysis. Anal. Chim. Acta 600, 90–96. Gutierrez, F., Albi, M.A., Palma, R., Rios, J.J., Olias, J.M., 2000. Bitter taste of virgin olive oil: Correlation of sensory evaluation and instrumental HPLC analysis. J. Food Sci. 54, 68–70. Hossenlopp, J.M., 2006. Applications of acoustic wave devices for sensing in liquid environments. Appl. Spectros. Rev. 41, 151–164. Ivarsson, P., Holmin, S., Höjer, N.-E., Krantz-Rülcker, C., Winquist, F., 2001. Discrimination of tea by means of a voltammetric electronic tongue and different applied waveforms. Sens. Actuators B 76, 449–454. Ju, M.-J., Hayama, K., Hayashi, K., Toko, K., 2003. Discrimination of pungent-tasting substances using surface-polarity controlled sensor with indirect in situ modification. Sens. Actuators B 89, 150–157. Legin, A., Rudinitskaya, A., Vlasov, Y., Di Natale, C., Davide, F., D´Amico, A., 1997. Tasting of beverages using an electronic tongue. Sens. Actuators B 44, 291–296. Mateos, R., Dominguez, M.M., Espartero, J.L., Cert, A., 2003. Antioxidant effect of phenolic compounds, α-tocopherol, and other minor components in virgin olive oil. J. Agric. Food Chem. 51, 7170–7175.
SECTION | I Natural Components
Nguyen, T.A., Kobot, S., Ongarato, D.M., Wallace, G.G., 1998. The use of chronoamperometry and chemometrics for optimization of conducting polymer sensor arrays. Electroanalysis 11, 1327–1332. Parra, V., del Caño, T., Rodríguez-Méndez, M.L., de Saja, J.A., Aroca, R., 2004. Electrochemical characterization of two perylenetetracarboxilic diimides: Langmuir-Blodgett films and carbon paste electrodes. Chem. Mater. 16, 358–364. Parra, V., Arrieta, A., Fernández-Escudero, J.A., Iñiguez, M., RodríguezMéndez, M.L., de Saja, J.A., 2006. Monitoring of the ageing of red wines in oak barrels by means of an hybrid electronic tongue. Anal. Chim. Acta 563, 229–237. Rock, F., Barsan, N., Weimar, U., 2008. Electronic nose: Current status and future trends. Chem. Rev. 108, 705–725. Rodríguez-Méndez, M.L., Arrieta, A., Parra, V., Vegas, A., Villanueva, S., Gutierrez-Osuna, R., de Saja, J.A., 2004. Fusion of three sensory modalities for the multimodal characterization of red wines. IEEE Sensors J. 4, 348–354. Rodríguez-Mendez, M.L., Apetrei, C., de Saja, J.A., 2008. Evaluation of the polyphenolic content of extra virgin olive oils using an array of voltammetric sensors. Electrochim. Acta 53, 5867–5872. Sohn, Y.-S., Goodey, A., Anslyn, E.V., McDevitt, J.T., Shear, J.B., Neikirk, D.P., 2005. A microbead array chemical sensor using capillary-based sample introduction: toward the development of an “electronic tongue”. Biosens. Bioelectron. 21, 303–312. Toko, K., 2000. Taste sensor. Sens. Actuators B 64, 205–215. Vlasov, Y., Legin, A., Rudnitskaya, A., Di Natale, C., D’Amico, A., 2005. Nonspecific sensor arrays (“electronic tongue”) for chemical analysis of liquids. Pure Appl. Chem. 77, 1965–1983. Winquist, F., Bjorklund, R., Krantz-Rülcker, C., Lundström, I., Östergren, K., Skoglund, T., 2005. An electronic tongue in the dairy industry. Sens. Actuators B 111-112, 299–304.
Chapter 58
Determination of Olive Oil Parameters by Near Infrared Spectrometry Sergio Armenta, Javier Moros, Salvador Garrigues and Miguel de la Guardia Cirugeda Department of Analytical Chemistry, Edificio Jeroni Muñoz (Research Building), University of Valencia, Spain
58.1 INTRODUCTION Near infrared (NIR) covers the region between 780 and 2500 nm, and based on the absorption, transmission or reflection of the light NIR spectroscopy provides a fast and non-destructive technique very useful for the simultaneous determination of several compounds in the same sample. Food NIR spectra comprise broad bands corresponding to overtones and combinations of vibrational modes involving C–H, O–H and N–H chemical bonds, providing a great amount of information which properly treated by chemometrics is useful for classification and for the quantification of many parameters. As an example Figure 58.1 provides the NIR spectrum of different olive oils. The olive oil extraction process starts in the harvesting and transport of olive fruit and involves cleaning and washing
of olives, followed by milling, malaxation, solid-phase separation and press. After that, liquid-phase separation can be done by the so-called two-way or three-way systems, which involve the separation of pomace oil, dregs of oil and water. In all the aforementioned steps, the use of NIR provides data on the oil content, acidity and moisture of olives, moisture and fat content of pomace and the main characteristics of olive oil. In the following sections the reader will see the main applications of NIR for the fast characterization and quantitative analysis of olive oil and related products.
58.2 OLIVE FRUIT ANALYSIS The development of new olive cultivars that improve: (i) early bearing, (ii) high yield and oil content, (iii) resistance
4.0
Absorbance
3.0
2.0
1.0 C 0.0
B A 1000
1500 Wavelength (nm)
2000
2500
FIGURE 58.1 NIR spectrum of an extra virgin olive oil (A), virgin olive oil (B) and pomace olive oil (C). The NIR spectra of oils comprise broad bands corresponding to overtones and combinations of vibrational modes involving C–H, O–H and N–H chemical bonds and provides a great amount of information which is useful for classification and for the quantification of many parameters. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
533
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
534
to fungus, and (iv) high olive oil quality, is a hot topic of the oil industry (Leon et al., 2005) and involves the analysis of a large number of samples in breeding programs. Oil content, fatty acid composition, including oleic and linoleic acid concentration and moisture, are the main parameters useful to evaluate the quality of olives. Gravimetry after Soxhlet extraction is the official method to determine the oil content of olive fruit and gas chromatography is commonly used to determine fatty acid composition. These methods are tedious, time-consuming and impractical for processing a large number of samples, therefore new procedures that can simplify the aforementioned determinations are required. NIR spectroscopy offers a good alternative to evaluate olive fruits based on its rapid and non-destructive multicomponent analysis capability with a low analytical cost of both instrumention and operation, and without sample preparation. As an example Table 58.1 summarizes the different publications regarding olive fruit analysis by NIR spectroscopy compiled from the SciFinder Scholar® (Chemical Abstracts Service (CAS) and from the US National Library of Medicine) and the Science Citation Index (SCI) database of the Institute for Scientific Information (ISI, Philadelphia, PA, USA). Methods available can be classified as those based on: (i) the analysis of intact fruit and (ii) the determination of different parameters on the crushed olives. NIR, in combination with partial least squares (PLS) regression was used for the determination of oil, moisture and fatty acid composition in intact olive fruits. The best models provided regression coefficients of r2 ⫽ 0.88 for moisture, 0.83 for oil, 0.77 for oleic acid, and 0.81 for linoleic acid. These models were validated, showing standard error of prediction (SEP) values ranging from 0.6 to 4.5% for the different parameters evaluated (Leon et al., 2003). The same team evaluated the effect of parent and harvest year on the determination of oil, moisture, oleic acid and linoleic acid using fruit samples collected in 1996 and 1997 from seedling plants derived from three different female parents (Leon et al., 2004). PLS models were developed using samples for each year and for each female parent separately and were validated against the other groups. Calibration models were accurate to predict all constituents in new samples from different female parents but were not transferable across years. The usefulness of NIR to discriminate between juvenile and adult olive seedlings at an early stage and to predict oil content and fatty acid in olives was also evidenced (Leon et al., 2005). A NIR methodology was also employed for the determination of acidity and fat on intact olive fruits. The relative standard errors of cross-validation (RSECV) obtained for the whole sample group were 0.27 and 1.02%, for acidity and fat respectively (Gonzalez et al., 2004). Classification of olives has been done by diffuse reflectance near infrared spectroscopy combined with pattern
SECTION | I Natural Components
recognition techniques. The most common alterations of olives, such as freeze damage, harvest after falling on the ground, fermentation due to prolonged storage time and olive tree diseases, were considered by NIR. Discriminant analysis (DA) provided prediction abilities of 100% for sound, 79% for frostbite, 96% for ground, and 92% for fermented olives using cross-validation. Quantification of oil and water in olives was also approached by using PLS. RSECV, using all samples, were 7.2% and 3.4% for oil and moisture, respectively. Using only sound samples, relative errors were prediction of 3.8% and 2.8% for oil and water. The use of a previous classification of olive samples improved the accuracy of determinations (Ayora-Cañada et al., 2005). Samples milled by a hammer crusher were analyzed by NIR reflectance. SEP values, compared to Soxhlet and drying methods, were 0.811 and 0.928 for oil content and moisture, respectively (Jimenez et al., 2000). A portable NIR was used to collect in-field spectra of crushed olives and moisture values found ranged from 35% to 65% and fat content from 15% to 35% providing regression models to the reference values with correlation coefficients of 0.9 and 0.8, respectively (Nazarov et al., 2005). NIR and NMR were employed for the determination of moisture and fat of crushed olives by Garcia-Sanchez et al. (2005). The accuracy and precision of results obtained from the two techniques were similar to those provided by the official gravimetric method. The proposed methods involve less sample handling, use less reagents and solvents, and are faster than the official procedure. In another application, moisture, fat and acidity were determined in crushed olives by diffuse reflectance NIR spectrometry using multivariate methods, and relative errors of prediction (REP) of 7.04, 3.06 and 6.19%, were obtained respectively (NuñezSanchez et al., 2005). NIR has also been employed for the analysis of olive leaves to do the classification between juvenile or adult olive trees and the K and N content (Leon et al., 2005; Leon and Downey, 2006). The two-dimensional scores plot for PC1 and PC2 obtained from the spectral profile (400 to 2500 nm) of olive leaves describe most of the spectral variability and two groups can be clearly separated. This method provided a rapid screening technique for the selection in olive breeding programs. The hyperspectral reflectance curves of olive trees (Olea europaea L.) were used to discriminate between different nitrogen (N) or potassium (K) applications using DA. Field hyperspectral studies were carried out in two olive orchards. N treatment consisted of annual applications of N per tree of 0 kg (N0), 0.5 kg (normal), or 1 kg (high). The K treatment consisted of fertilizations with 0%, 2.5%, and 5% K2CO3. Hyperspectral measurements of leaf samples was made using a handheld field spectroradiometer working from 400 to 900 nm. Wavelength interval from 710 to 900 nm
CHAPTER | 58
535
Determination of Olive Oil Parameters by Near Infrared Spectrometry
TABLE 58.1 Analytical characteristics of NIR-based methodologies to analyze olive fruits. This table summarizes the analytical features of the different methodologies based on NIR spectroscopy developed to the olive fruit analysis. Parameters
Chemometric treatment
NIR region (nm)
Number of samples
Reference
Oil, moisture and fatty acid composition
PLS
400–1700
287
Leon et al. (2003)
Oil, moisture, oleic acid and linoleic acid
PLS
400–1700
437
Leon et al. (2004)
Oil and fatty acid composition
PLS
400–1700
434 (seedlings)
Leon et al. (2005)
Fat content and acidity
–
–
76
Gonzalez et al. (2004)
Moisture and fat content
HCA, PCA, DA, PLS
1010–2439
–
Ayora et al. (2005)
Moisture and fat content
Multivariate method
–
–
Jimenez (2000)
Moisture and fat content
PLS
–
–
Nazarov (2005)
Moisture and fat content
Artificial neural network
850–1050
2000
Garcia-Sanchez et al. (2005)
Moisture, acidity and fat content
Multivariate method
400–2500
1477
Nuñez-Sanchez et al. (2005)
Moisture and fat content
PLS
400–2500
287
Bendini et al. (2007)
PLS: partial least squares; HCA: hierarchical cluster analysis; PCA: principal component analysis; DA: discriminant analysis.
was selected for discriminating between N treatments, with an overall accuracy of up to 99.2%. Wavelength interval from 710 to 890 nm, and the normalized difference vegetation index were selected for discriminating K treatments with an overall accuracy of up to 94.4% (Gomez-Casero et al., 2007). It can be concluded that the direct analysis by NIR of olive leaves provides information on both the tree nature and the soil treatment, thus offering an interesting tool for complementary studies on the quality of oil to be expected.
58.3 OLIVE OIL ANALYSIS According to the European Regulation, there are four types of virgin olive oil from extra virgin, to virgin, ordinary virgin and lampante virgin. Olive oils labeled virgin cannot include oils obtained by using solvents and do not permit mixing with other oils. Free fatty acids (FFA) as indicator of free acidity, ultraviolet absorption at 225, 232 and 270 nm, known as K indexes related to oil bitterness, and oil oxidation, together with peroxide value, moisture and volatile matter, insoluble impurities in light petroleum and trace metal are the main physicochemical parameters used to classify olive oils, and international norms have been proposed for their determinations.
58.3.1 Olive Oil Parameters Table 58.2 reports the analytical conditions used for NIR determination of the analytical parameters of olive oils. As can be seen, acidity has been determined by NIR in mixtures of oils or liquid waxes using PCA or FA (Janosch and Ebel, 1993) as well as for the evaluation of used frying fats by optical light-guide-assisted NIR spectrometry with an analytical certainty of 2–3% (Kehraus et al., 1999). On the other hand, there is an available NIR sensor to determine acidity in oils (Moltó et al., 2003). It provides a precision of 0.075° in the range between 0° and 1°. Acidity was also determined using NIR through PLS on the wavenumber range between 1961 and 2212 nm allowing an uncertainty prediction of 0.06% (wt./wt.) within the 0.15 to 1.3% analytical range (Lagardere et al., 2004). Another PLS model using a nine-point smoothing previous and posterior to a three-point first derivative in the spectral range from 1100 nm to 2500 nm, was used for monitoring on-line the acidity during the processing of virgin olive oils (Jimenez-Marquez et al., 2005), and Bendini et al. extended the in-process monitoring to samples collected at three mill plants, located in Italy, which utilized different technological equipment (Bendini et al., 2007). Acidity was evaluated in edible oils of different types (olive, maize, seed and sunflower) and origins after their
536
SECTION | I Natural Components
TABLE 58.2 Analytical conditions of the NIR-based methodologies proposed for the determination of quality parameters of olive oils. This table summarizes the analytical features of the different methodologies based on NIR spectroscopy developed to the olive oil parameters determination, such as acidity, peroxide value and moisture. Parameter
Type of sample
Chemometric treatment NIR region (nm)
Acidity
Mixtures of oils or liquid waxes Frying fats Olive oils
PCA or FA
Virgin olive and refined sunflower oils
PLS
Virgin olive oils
PLS-smoothing, with 1100–2500 a nine points signal, previous and posterior to a three points first derivative on spectral ranges
190
Jimenez-Marquez et al. (2005)
Olive, maize, seed and sunflower oils
PLS – linear removed
126
Armenta et al. (2007)
Olive oils
PLS, first derivative and straight line subtraction
161
Bendini et al. (2007)
Edible oils
Second derivative
Peaks at 1468 and 2084
Takamura et al. (1996)
Virgin olive and refined sunflower oils
PLS
1333–1587
Lagardere et al. (2004)
Peroxide value
Reference
Janosch and Ebel (1993) Kehraus et al. (1999) Moltó et al. (2003) 1961–2212
853–2202
Olive oils from throughout the Australian olive-growing areas
Moisture
Number of samples
Lagardere et al. (2004)
216
South African extra virgin olive oils
PLS
PerkinElmer IdentiCheck spectrophotometer (1100– 2500) – Transmittance Buchi NIRLab N-200 instrument (978–2500) – Reflectance
Olive, maize, seed and sunflower oils
PLS – linear removed
853–2202
Mixtures of oils or liquid PCA or FA waxes
Mailer et al. (2004)
Manley (2006)
126
Armenta et al. (2007)
Janosch and Leben (1993)
PLS: partial least squares; PCA: principal component analysis; FA: factor analysis.
discrimination using hierarchical cluster analysis (HCA). Individual models were built for two groups defined as (I) olive oil and (II) sunflower, seed and maize oils, being obtained root mean square errors of prediction of 0.034 and 0.037% (wt./wt.) for both groups of oils, respectively. The limit of detection of the methodology was 0.03% for both groups (Armenta et al., 2007). In another study, the
intensity of the peak at 2084 nm (corresponding to the specific absorption for the hydroperoxyl group) on the second derivative spectra of oxidized edible oils was highly correlated with the peroxide value (PV) (Takamura et al., 1995). This parameter, jointly with acidity, is highly influenced, as stated Gardiman et al. (1999), by hypoxic and CO2-enriched atmospheres in which olive oils were stored.
CHAPTER | 58
Determination of Olive Oil Parameters by Near Infrared Spectrometry
PLS models were developed for the determination of PV in virgin olive and sunflower oils using wavenumber range between 7500 and 6300 cm⫺1. PV was predicted with an uncertainty of 1.0 meqO2 kg⫺1 in the 3–32 meqO2 kg⫺1 range and with an uncertainty of 1.1 meqO2 kg⫺1 in the range between 1.5 and 24 meqO2 kg⫺1 for olive and sunflower oils, respectively. It was necessary to build a calibration model for each type of oil to improve the prediction capability of PV (Lagardere et al., 2004). In this sense, PV determination in different types of oils made after discrimination of olive oils from maize, seed and sunflower, provided RMSEP values for PV of 1.87 and 0.79 meqO2 kg⫺1. The limits of detection for the developed methodology were 0.9 and 0.8 meqO2 kg⫺1 in olive oil and other edible oils, respectively (Armenta et al., 2007). NIR calibrations, using olive oil samples from Australia, were employed for routine screening of olive oil (Mailer, 2004). Moreover, PV, among other quality parameters, was used to make a comparison of PLS-NIR models for South African extra virgin olive oils. Reliable prediction results were obtained for PV with a SEP of 4.15 meqO2 kg⫺1 and a 0.87 value for correlation coefficient squared (r2) (Manley and Eberle, 2006). The NIR sensor developed by Moltó (Moltó et al., 2003), provided 0.63 meqO2 kg⫺1 precision values for PV in the analytical range between 0 and 14. Moisture has been evaluated by NIR in olive fruit (see Section 58.2), or on olive pomace (see Section 58.4), but there is one precedent about the determination of moisture in mixtures of oils or liquid waxes using PCA or FA (Janosch and Ebel, 1993). On the other hand, K232 and K270 parameters were evaluated by PLS-FT-NIR but reliable predictions were obtained only for K232 (SEP ⫽ 0.94, r2 ⫽ 0.94) (Manley and Eberle, 2006). Concerning organoleptic properties, such as flavor, aroma and color, related to the gustatory, olfactory and visual attributes of olive oils, Mignani et al. developed a microoptic sensor for the detection of olive oil aroma capable of distinguishing different aging levels of extra virgin olive oil (Mignani et al., 2007a). An optical NIR sensor was used to evaluate bitter taste (K225) using PLS regression on a total of 190 virgin olive oils. The validation set gave correlation coefficients and standard error of prediction of 0.936 and 0.058%, respectively (Jimenez-Marquez et al., 2005). Fatty index and saponification and ester numbers were determined on mixtures of oils and liquid waxes using PCA or FA (Janosch and Ebel, 1993). NIR transflectance spectroscopy scanning from 1100 nm to 2500 nm was applied for the determination of palmitic acid (C16:0), oleic acid (C18:1) and linoleic acid (C18:2) (Yang et al., 2005). A fiber-optic-based, cheap, and compact device for UVVIS-NIR was used to obtain the hyperspectral optical signature of olive oil in the 200–1700 nm spectral range, for authentication purposes, and also to correlate the spectral data with the content of fatty acids (Mignani et al., 2007b).
537
Oleic acid, linoleic acid, linolenic acid, saturated and unsaturated fatty acids, in edible oils and used frying fats were determined by optical light-guide-assisted NIR (Kehraus et al., 1999). The classification and quantification of the fatty acid composition of different oils and fats provided unique fingerprints for saturated fatty acids, cis and trans monounsaturated fatty acids, and all n-6 and n-3 polyunsaturated fatty acids (Azizian and Kramer, 2005). A rapid and accurate evaluation of fatty acids, with correlation coefficients for stearic acid (0.86), linolenic acid (0.85), oleic acid (0.99), linoleic acid (1.00), FFA (0.97), and chlorophyll (0.98) was carried out (Mailer, 2004). Several virgin olive oil quality parameters such as linoleic, oleic and saturated acid content, α:β ratios of linoleic and oleic acid, pigment, total polyphenol content, and thiobarbituric acid reactive substances value were evaluated by PLS-FT-NIR (Manley and Eberle, 2006) as well as for examination of the effect of using orthogonal signal correction (OSC) on low-scatter NIR spectra on the determination of analyte concentrations at different levels in samples containing additional components with very similar spectra (Bertran et al., 2001). So, it is clear that NIR offers a valuable tool for the quantification of olive oil parameters and, as will be discussed in Section 58.3.3, data obtained could be employed to evaluate the purity and authentication of olive oils.
58.3.2 Analytes in Olive Oils NIR has been employed for specific analyte determination in olive oil, as chlorophyll, carotenoid and tocopherol, which are the main pigments and play an important role as antioxidants (Gutierrez-Rosales et al., 1992), additionally than to determine the main quality parameters of oils. High-performance liquid chromatography (HPLC) has been the most widely employed technique for determining chlorophyll pigments, carotenoids and tocopherols in vegetable oils. However, there are some precedents in the literature about the evaluation of the aforementioned compounds by NIR. Visible near infrared spectroscopy has been used for the on-line determination of total levels of carotenoid and chlorophyll pigments during virgin olive oil processing. Correlation coefficient and standard error of prediction of 0.985 and 0.66 mg kg⫺1 for total carotenes and 0.993 and 0.96 mg kg⫺1 for total chlorophylls were obtained for the validation set (Jiménez-Márquez et al., 2003). On the other hand, NIR calibrations, using a set of 216 samples, were developed for a series of quality parameters of olive oil, including polyphenol content, induction time (Halbault et al., 1997) and chlorophyll, among others. Chlorophyll determination, with a multiple squared correlation coefficient (R2) of 0.98, provided high levels of accuracy (Mailer, 2004). Stability of virgin olive oil has been highly correlated with the polar phenol content, which may act against free
538
radical attack mainly by donating a hydrogen atom to the lipid radical formed during the propagation of lipid oxidation and alpha-tocopherol has been determined in edible oils after extraction with ethanol. Standard solutions (0.54– 53.54 mg mL⫺1) were employed for building the calibration model. Detection and quantification limits of 0.12 mg mL⫺1 (LOD) and 0.40 mg mL⫺1 (LOQ) as well as sensitivity and selectivity, were estimated using the net analyte signal calculation of the measured reflectance. Precision and accuracy of the proposed NIR methodology were comparable to those obtained by HPLC (Szlyk et al., 2005).
58.3.3 Classification and Authentication of Oils by NIR The potential of NIR spectrometry in combination with chemometrics has been exploited to identify the origin of samples, to carry out the authentication of oil and greases and to predict and to quantify adulterations in olive oils (see Table 58.3). The use of artificial neural networks and logistic regression in combination with NIR was succesfully used to discriminate samples of virgin olive oil of very similar and geographically close denominations of origin from Spain (Bertran et al., 2000). Neural networks provided a higher discrimination capacity than logistic regression, but the performance is dependent on the initial operating conditions. PLS, FDA and k-nn have been investigated in combination with visible and NIR spectra to classify extra virgin olive oils from Greece. Discriminant models were developed and evaluated using spectral data in the visible (400–750 nm), NIR (1100–2498 nm), and combined (400–2498 nm) wavelength ranges. The best results were obtained using FDA on raw spectral data over the combined wavelength range with a correct classification rate of 93.9% for a prediction sample set (Downey et al., 2003). Brazil nut, coconut, corn, sunflower, walnut, virgin olive, peanut, palm, canola, soybean, sunflower or animalbased (tallow and hydrogenated fish) oils, have been classified by NIR (Hourant et al., 2000). The NIR features of the most characteristic bands were studied to design a filtertype NIR instrument and to build an arborescent structure, based on stepwise linear discriminant analysis (SLDA), to classify samples according to their sources. The model was evaluated with a test set and 90% of samples were correctly classified. LDA and canonical variate analysis (CVA) were used for the discrimination and classification of ten different edible oils and fats based on spectral data. FTIR spectroscopy was found to be the most efficient in classification when used with CVA and yielded about 98% classification accuracy, followed by FT-Raman (94%) and FT-NIR (93%) methods (Yang et al., 2005). PCA data of vegetable oils in the region of 1600– 2200 nm were used for their classification, confirming that
SECTION | I Natural Components
wavelengths with a high loading weight correspond to specific fatty acid absorption regions (Sato, 1994). The multispectral digital signature of Italian extra virgin olive oils bearing labels of certified area of origin was measured by means of UV-VIS-NIR. Spectral data were processed by multivariate analysis and plotted on a 2D classification map, showing sharp clusters according to the geographical origin of oils. Spectral data were correlated with the content of the most important fatty acids, achieving an acceptable fitting (Mignani et al., 2007b). Detection of oil adulteration is a complex matter and highly sophisticated analytical techniques have been employed in identifying adulterated oils. Mass spectrometry (MS), gas chromatography (GC), HPLC, DNA-based methods and isotope ratio mass spectroscopy were used in this way. In contrast to the aforementioned techniques, NIR provides fast and non-destructive measurements with no sample pre-treatment. Evaluation of oil adulteration has been made based on their characteristic NIR spectra and a pioneering study in this field was the confirmation that shifts on the position of chlorophenyl band at 665 nm in the absorption spectra of olive oil were indicators of the amount of adulteration with soybean and cotton-seed oils in spite of the fact that refractive index measurements can not differentiate these adulterations (van der Lingen, 1937). The authentication of different oil samples was investigated by optical light-guide-assisted NIR and HCA (Kehraus et al., 1999). The authenticity was satisfactory for olive, rapeseed, sunflower and thistle oils. The spectral differentiation between soy and sunflower oil was possible by determination of the linolenic acid content. The fiber-optic-based UV-VIS-NIR device developed for the determination of fatty acids has also been employed for authentication of olive oils (Mignani et al., 2007b). As can be seen in Table 58.3, the chemometric treatment of a selected region of NIR spectra is nowadays the methodology of choice for identification of olive oil adulteration. PLS and PCA in combination with NIR were used for determining adulteration in extra virgin olive oil with corn, sunflower and raw olive residue oil. The classification and determination of the degree of adulteration were carried out by PLS using the first derivative spectral profiles from 1108–2200 nm. Correct classification was achieved for 98% of samples and the prediction of the type of adulterant was done with a prediction rate of 75% (Wesley et al., 1995). A similar procedure was used to determine sunflower, rapeseed and soybean adulteration in extra virgin olive oil. Over the range 5–100% of adulterants in olive oil, an internal cross-validation relative standard deviation (iCVRSD) of 0.9% and an external cross-validation relative standard deviation (eCVRSD) of 2.77% were obtained (Wesley et al., 1996). The adulteration of extra virgin olive oil with olive pomace oil has been studied with Raman. However, NIR,
CHAPTER | 58
539
Determination of Olive Oil Parameters by Near Infrared Spectrometry
TABLE 58.3 Analytical conditions of the NIR-based methodologies proposed for origin classification, authentication and detection of olive oil adulterations. This table summarizes the different methodologies based on NIR spectroscopy proposed for the classification of the olive oil as a function of their origin. Moreover, it resumes also the NIR methods developed for the authentication and olive oil quality control.
Origin classification
Authentication
Adulteration
Type
Chemometric treatment
NIR region
Spanish virgin olive oil
Artificial neural network and logistic regression
Crete and Greek olive oil
PCA, HCA, PLS, FDA, k-nn
400–2498
Italian extra virgin olive oil
Multivariate analysis and 2D classification map
200–1700
Soybean, corn, cottonseed, olive, PCA rice bran, peanut, rapeseed, sesame and coconut oils
Number of samples
Reference
Bertran et al. (2000) 65
Downey et al. (2003) Mignani et al. (2007b)
1600–2200
Sato (1994)
Olive, rape, soy, sunflower, and thistle oil
HCA
Kehraus et al. (1999)
Brazil nut, coconut, corn, sunflower, walnut, virgin olive, peanut, palm, canola, soybean, sunflower, tallow and hydrogenated fish oils
SLDA
1700–1800 2100–2400
104
Hourant et al. (2000)
Cod liver, extra virgin olive, corn, peanut, canola, soybean, safflower, coconut oils
PCA- PLS →LDA-CVA
1538–2500 1250–5000
–
Yang et al. (2005)
Extra virgin olive oil identification
Hyperspectral optical signature
Corn, sunflower and refined olive oils
PLS, PCA
1108–2200
99
Wesley et al. (1995)
Sunflower, rapeseed and soybean oils
PLS, PCA
1100–2500
18
Wesley et al. (1996)
Pomace oil
PLS
470–690 1145–1265 1355–1500
21
Yang (2001)
Sunflower oil
PCA, HCA, PLS, FDA, k-nn, SIMCA
400–2498
46
Downey et al. (2002)
Soya, sunflower, corn, walnut and hazelnut oils
PCA, PLS
833–2200
525
Christy et al. (2004)
Corn, hazelnut, soya and sunflower oils
DPLS, PLS
833–2200
280
Kasemsumran et al. (2005)
Soybean oil
PLS
2700–2900
30
Sunflower and corn oil
GILS
1000–2500
26
Mignani et al. (2007b)
Gonzaga and Pasquini (2005) Gonzaga et al. (2007)
PLS: partial least squares; PCA: principal component analysis; HCA: hierarchical cluster analysis; SIMCA: soft independent modeling of class analogy; FDA: factorial discriminant analysis; k-nn: k-nearest neighbors analysis; DPLS: discriminant partial least squares; GILS: genetic inverse least squares; SLDA: stepwise linear discriminant analysis; LDA: linear discriminant analysis; CVA: canonical variate analysis.
attenuated total reflectance (ATR)-FTIR and photo acoustic spectrometry (PAS)-FTIR also provided good results (Yang and Irudayaraj, 2001). Visible and NIR transflectance have been used to evaluate the adulteration of extra virgin olive oils with sunflower
oil. Different multivariate approaches, based on HCA, FDA, PCA, SIMCA and PLS were investigated. Complete classification accuracy was achieved using the first derivative data in the 400–2498 nm range. Prediction of adulterant content was possible with a standard error equal to 0.8%
540
using the first derivative data between 1100 and 2498 nm (Downey et al., 2002). The classification and quantification of adulterations of pure olive oil by soya oil, sunflower oil, corn oil, walnut oil and hazelnut oil was made based on a chemometric analysis of the NIR spectra of olive oil mixtures with different adulterants. The classification was done by PCA after multiplicative signal correction (MSC) of spectra. The MSC-corrected data were subjected to Savitzky-Golay smoothing and a mean normalization procedure before developing PLS calibration models. Corn, sunflower, soya, walnut and hazelnut oils adulterations were predicted with error limits of 0.57, 1.32, 0.96, 0.56 and 0.57% w/w, respectively. Furthermore, the PCA models were able to classify unknown adulterated olive oil mixtures with almost 100% certainty (Christy et al., 2004). The discrimination and determination of corn, hazelnut, soya and sunflower adulteration of olive oil was made by discriminant PLS (DPLS) and PLS. The classification method reported accuracies ranging from 96.7 to 100% and an error of prediction for the quantification of the adulterants lower than 1.05% (Kasemsumran et al., 2005). The determination of olive oil adulteration with soybean oil was made using an acousto-optical tunable filter (AOTF). The region between 2700 and 2900 nm of the emission spectra was baseline corrected, normalized by their maximum value and smoothed prior to being employed in the PLS model. The absolute root mean square error of prediction (RMSEP) of the methodology was 3.2% (Gonzaga and Pasquini, 2005). Multivariate calibration using genetic inverse least squares (GILS) in combination with NIR transmittance measurements were used to predict the concentration of sunflower and corn oil adulterants in olive oil. The SEP ranged between 2.49 and 2.88% for binary mixtures of olive and sunflower oil and between 1.42 and 6.38% for ternary mixtures.
58.4 OLIVE POMACE ANALYSIS NIR has been used in the laboratory or in processing mill lines for the analysis of olive by-products such as olive pomace, also called olive cakes or orujo, and pomace oil. Results for oil and moisture in olive pomace were satisfactory enough to suggest the development of an on-line NIR-based sensor for olive-pomace analysis (Garcia-Mesa et al., 1996). Table 58.4 summarizes studies reported in the literature for olive pomace analysis by NIR. As can be seen, PLSNIR reflectance spectrometry has been employed to determine oil and water in olive pomace. The RMSEP values were 0.20–0.21 and 1.8 for oil and water content, respectively (Muik et al., 2004).
SECTION | I Natural Components
The off-line analysis of moisture and fat in olive pomace has been evaluated comparing the accuracy and precision of NIR to those provided by NMR and the official gravimetric method (Garcia-Sanchez et al., 2005). Moreover, moisture and fat were also determined at-line in olive pomace. The system is not integrated in the two-phases decanter, but on the support structure, with the sampling made manually. Results were comparable with those obtained by NMR and reference methods (Garcia-Mesa, 2003). On the other hand, a NIR sensor was installed at the decanter cake passage. Results showed a good correlation between NIR and NMR methods, with a correlation coefficient (r ⫽ 0.96) for oil content. However, for moisture, the correlation between the NIR online method and the reference method provided an r ⫽ 0.60. A later revision of the methodology improved the results for moisture obtaining a correlation coefficient r ⫽ 0.79 maintaining the value for fat in 0.89 (Hermoso et al., 2002). A toal of 160 samples of olive pastes (from differing cultivar and maturity olives) and 122 samples of solid–liquid waste were used to build calibrations for the determination of moisture and fat by NIR (Gonzalez et al., 2005).
58.5 COMMERCIALLY AVAILABLE OIL ANALYZERS Oil analyzers based on NIR are nowadays commercially available and they can provide useful information about the oil production and the quality control of oil which unfortunately are not commonly published in international scientific journals as studies reported in previous sections. Regarding trademark analyzers available in the market, Table 58.5 shows typical examples which use NIR to determine moisture, fat and fatty acid content, and other parameters such as iodine index or phosphorous value in olive and olive oils. An example of oil sensor used in the mill industry is the multi-response instrument patented by Novedades Oleotecnologicas S.L. It is based on the infrared measurement of oil samples pumped from the different production stages through independent pipes, which are equipped with transparent material windows. The main problem of oil sensor systems remains in the acquisition and installation costs (acquisition cost of an oil and moisture sensor based on NIR is around 15 000 €), together with the need for trained staff. Unfortunately, data obtained every day by using some of the aforementioned equipments are scarcely available.
58.6 CONCLUSIONS The instrumental and chemometric developments presented throughout this review show the tremendous possibilities
CHAPTER | 58
541
Determination of Olive Oil Parameters by Near Infrared Spectrometry
TABLE 58.4 Analytical characteristics of NIR-based methodologies proposed to analyze olive pomace. This table summarizes the analytical features of the different methodologies based on NIR spectroscopy developed to the olive pomace analysis. Strategy
Analyte
Chemometric treatment
NIR region (nm)
Number samples
Reference
Off-line
Moisture and fat content
–
–
–
Garcia-Mesa et al. (1996)
PLS
1340–1633 1860–2197
132
Muik et al. (2004)
Moisture and fat content
1125–1276 1666–1835 2268–2395
Moisture and fat content
Artificial neural network
850–1050
2000
Garcia-Sanchez et al. (2005)
At-line
Moisture and fat content
–
–
18 (mills)
Garcia-Mesa et al. (2003)
On-line
Moisture and fat content
PLS
750–2500
–
Hermoso et al. (1999)
Moisture and fat content
PLS
750–2500
–
Hermoso et al. (2002)
Moisture and fat content
–
–
122
Gonzalez et al. (2005)
PLS: partial least squares.
TABLE 58.5 Characteristics of some commercially available oil analyzers. Name
Company
Applications
NIT-38 Olive Analyzer
Nirtech
Oil and moisture in ground olives
Range 720–1100
Chemometric method PLS
Acidity and moisture in virgin olive oil Oils and fat analyzer
Foss
Free fatty acid, phosphorous and iodine value in olive oil
1100–2500
ANN
400–2500 OliveScan
Foss
Moisture and oil content in olive paste
850–1050
ATA8000 Olive Analyzer
Unity Scientific
Oil and moisture in fruit and pomace
1200–2400
IR3000
Moisttech
Oil and moisture in olive paste
–
InfraScan
Wenk Labtec GmbH
Quality of oil
1400–2600
GT Olive Analyzer
Graintec
Oil and moisture in ground olives. Fatty acids and moisture in olive oil
Le Vigneron
Brimrose Corporation USA
Moisture and fat content in crushed olive
1100–2300
MultiComponent™ 2750 EP-NIR
Aspectrics
Quantifying corn, canola and olive oils in mixtures
2000–2600
ANN
720–1100
PCA-PCR
542
SECTION | I Natural Components
offered by NIR for the analysis and/or classification of olive fruits, oils and pastes. However, the major part of these studies have been made in a laboratory scale, far from the olive mills, and only a few of them have been applied online during the olive oil elaboration process. Automation in the olive mill industry is a key point to obtain a high-quality product through optimal process yields at low costs and thus it is necessary for close cooperation between research teams and the industrial sector. The challenge of producing a high-quality olive oil is of great concern and the selection of olive fruit with superior properties, which ensure positive attributes in the olive oil is, in our opinion, another hot topic in the NIR applications in olive oil production. Another step in the olive oil industries which needs to be improved through the incorporation of NIR measurements, is the control of wastes. Olive-foot is a highly contaminated residue and, to our knowledge, there is no precedent on the use of NIR for its characterization. So, we can expect new developments in this field in the near future.
SUMMARY POINTS ●
●
●
●
●
●
●
●
●
●
Olive oil NIR spectra comprise broad bands corresponding to overtones and combinations of vibrational modes providing a great amount of information, which properly treated by chemometrics is useful for classification and for the quantification of many parameters. NIR provides data on the oil content, acidity and moisture of olives, moisture and fat content of pomace and the main characteristics of olive oil. NIR offers a good alternative to evaluate olive fruits based on its rapid and non-destructive multicomponent analysis capability. Classification of olives has been done by diffuse reflectance near infrared spectroscopy combined with pattern recognition techniques. It has developed a NIR sensor to determine acidity with a precision of 0.075° in the range between 0° and 1°. PLS models were developed for the determination of PV in virgin olive and sunflower oils using wavenumber range between 7500 and 6300 cm⫺1. NIR transflectance spectroscopy scanning from 1100 nm to 2500 nm was applied for the determination of palmitic acid, oleic acid and linoleic acid. Linoleic, oleic and saturated acid content, α:β ratios of linoleic and oleic acid, pigment, total polyphenol content, and thiobarbituric acid reactive substances value were evaluated by PLS-FT-NIR. NIR has been employed for determination in olive oil of chlorophyll, carotenoid and tocopherol. The use of artificial neural networks and logistic regression in combination with NIR was successfully used to
●
●
●
●
●
discriminate samples of virgin olive oil of very similar and geographically close denominations of origin. PCA data of vegetable oils in the region of 1600– 2200 nm were used for their classification. PLS and PCA in combination with NIR were used for determining adulteration in extra virgin olive oil. The adulteration of extra virgin olive oil with olive pomace oil has been studied with NIR, attenuated total reflectance and photo acoustic spectrometry. Visible and NIR transflectance have been used to evaluate the adulteration of extra virgin olive oils with sunflower oil. The classification and quantification of adulterations of pure olive oil by soya oil, sunflower oil, corn oil, walnut oil and hazelnut oil was made based on a chemometric analysis of the NIR spectra.
REFERENCES Armenta, S., Garrigues, S., De la Guardia, M., 2007. Determination of edible oil parameters by near infrared spectrometry. Anal. Chim. Acta. 596, 330–337. Ayora-Cañada, M.J., Muik, B., García-Mesa, J.E., Ortega-Calderón, D., Molina-Díaz, A., 2005. Fourier-transform near-infrared spectroscopy as a tool for olive fruit classification and quantitative analysis. Spectrosc. Lett. 38, 769–785. Azizian, H., Kramer, J.K.G., 2005. A rapid method for the quantification of fatty acids in fats and oils with emphasis on trans fatty acids using Fourier Transform near infrared spectroscopy (FT-NIR). Lipids 40, 855–867. Bendini, A., Cerretani, L., Di Virgilio, F., Belloni, P., Lercker, G., Toschi, T.G., 2007. In-process monitoring in industrial olive mill by means of FT-NIR. Eur. J. Lipid Sci. Tech. 109, 498–504. Bertran, E., Blanco, M., Coello, J., Iturriaga, H., Maspoch, S., Montoliu, I., 2000. Near-infra-red spectrometry and pattern recognition as screening methods for the authentication of virgin olive oils of very close geographical origins. J. Near Infrared Spec. 8, 45–52. Bertran, E., Iturriaga, H., Maspoch, S., Montoliu, I., 2001. Effect of orthogonal signal correction on the determination of compounds with very similar near infrared spectra. Anal. Chim. Acta 431, 303–311. Christy, A.A., Kasemsumran, S., Du, Y.P., Ozaki, Y., 2004. The detection and quantification of adulteration in olive oil by near-infrared spectroscopy and chemometrics. Anal. Sci. 20, 935–940. Downey, G., McIntyre, P., Davies, A.N., 2002. Detecting and quantifying sunflower oil adulteration in extra virgin olive oils from the Eastern Mediterranean by visible and near-infra-red spectroscopy. J. Agric. Food Chem. 50, 5520–5525. Downey, G., McIntyre, P., Davies, A.N., 2003. Geographic classification of extra virgin olive oils from the eastern Mediterranean by chemometric analysis of visible and near-infrared spectroscopic data. Appl. Spectrosc. 57, 158–163. Gallardo-Gonzalez, L., Osorio-Bueno, E., Sánchez-Casas, J., 2005. Use of near-infrared spectroscopy (NIRS) for the real-time determination of moisture and fat content in olive pastes and solid-liquid wastes. Alimentacion, Equipos y Tecnologia 206, 85–89.
CHAPTER | 58
Determination of Olive Oil Parameters by Near Infrared Spectrometry
Garcia-Mesa, J.A., Hermoso-Fernández, M., Cáceres-Alonso, P., 1996. Determination of oil and moisture in two-phase olive-pomace by using near-infrared spectroscopy. Grasas Aceites 47, 317–322. García-Mesa, J.A., 2003. Analisis In-Situ de alpeorujo mediante un sistema NIR portatil. Expoliva 2003. Jaen, Spain. Garcia-Sanchez, A., Ramos-Martos, N., Ballesteros, E., 2005. Comparative study of different analytical techniques (NIR and NMR spectroscopy and Soxhlet extraction) for the determination of fat and moisture content in olives and pomace from Jaen (Spain). Grasas Aceites 56, 220–227. Gardiman, M., Tonutti, P., Pizzale, L., Conte, L., Carazzolo, A., 1999. The effect of hypoxic and CO2-enriched atmospheres on olive ripening and oil quality. Acta Horticulturae 474, 525–528. Gomez-Casero, M.T., Lopez-Granados, F., Peña-Barragan, J.M., JuradoExposito, M., Garcia-Torres, L., Fernandez-Escobar, R., 2007. Assessing nitrogen and potassium deficiencies in olive orchards through discriminant analysis of hyperspectral data. J. Am. Soc. Hortic. Sci. 132, 611–618. Gonzalez, J.M., Pérez, D., Garrido, A., Guerrero, J.E., 2004. Non destructive analysis of intact olive fruits by near infrared spectroscopy: prediction of fat yield and acidity. Near infrared spectroscopy. Proceedings of the 11th International conference, pp 373–375. Davies, A.M.C., Garrido, A. (eds), Córdoba (Spain). Gutierrez-Rosales, F., Garrido-Fernandez, J., Gallardo-Guerrero, M.L., Gandul-Rojas, B., Minguez-Mosquera, M.I., 1992. Action of chlorophylls on the stability of virgin olive oil. J. Am. Oil Chem. Soc. 69, 866–871. Halbault, L., Barbé, C., Aroztegui, M., De La Torre, C., 1997. Oxidative stability of semi-solid excipient mixtures with corn oil and its implication in the degradation of vitamin A. Int. J. Pharm. 147, 31–41. Hermoso, M., Uceda, M., Garcia-Ortiz, A., Jimenez, A., Beltran, G., 1999. Preliminary results of NIR “on-line” measure of oil content and humidity in olive cakes from the two phases decanter. Acta Horticulturae 474, 717–719. Hermoso, M., García-Ortíz, A., Jiménez, A., Uceda, M., Morales, J., 2002. Automatización de almazaras. Controles experimentales para la caracterización y regulación del proceso de elaboración. Actas I. Jornadas Técnicas del Aceite de Oliva. CSIC, Madrid, Spain. Hourant, P., Baeten, V., Morales, M.T., Meurens, M., Aparicio, R., 2000. Oil and fat classification by selected bands of near-infrared spectroscopy. Appl. Spectrosc. 54, 1168–1174. Janosch, J., Ebel, S., 1993. Determination of characteristic fat index values of fatty oils by NIR spectrometry. Pharmazie 48, 824–828. Jiménez-Márquez, A., 2003. Monitoring carotenoid and chlorophyll pigments in virgin olive oil by visible-near infrared transmittance spectroscopy. On-line application. J. Near Infrared Spec. 11, 219–226. Jiménez-Márquez, A., Molina-Diaz, A., Pascual-Reguera, M.I., 2005. Using optical NIR sensor for on-line virgin olive oils characterization. Sensor Actuat. B-Chem. 107, 64–68. Jimenez, A., Izquierdo, E., Rodriguez, F., Duenas, J.I., Tortosa, C., 2000. Determination of fat and moisture in olives by near-infrared reflectance spectroscopy. Grasas Aceites 51, 311–315. Kasemsumran, S., Kang, N., Christy, A., Ozaki, Y., 2005. Partial least squares processing of near-infrared spectra for discrimination and quantification of adulterated olive oils. Spectrosc. Lett. 38, 839–851. Kehraus, S., Harder, J., Buening-Pfaue, H., 1999. Optical light-guide assisted NIR -spectrometric food control. Investigations of edible oils and frying fats. Lebensmittelchemie 53, 9–11.
543
Lagardere, L., Lechat, H., Lacoste, F., 2004. Quantitative determination of peroxide value (PV) and acidity in virgin olive oils and in sunflower oils thanks to the rapid Fourier transform near-infrared (FT-NIR) spectroscopic method. Lipides 11, 70–75. León, L., Rallo, L., Garrido, A., 2003. Analysis of intact olive fruit by near-infrared spectroscopy (NIRS): A useful tool in olive breeding programs. Grasas Aceites 54, 41–47. León, L., Garrido-Varo, A., Downey, G., 2004. Parent and harvest year effects on near-infrared reflectance spectroscopic analysis of olive (Olea europaea L.) fruit traits. J. Agric. Food Chem. 52, 4957–4962. León, L., Garrido-Varo, A., Downey, G., 2005. Near Infrared Spectroscopy (NIRS) as a Promising Selection Tool in Olive Breeding Programs. FRUTIC 05. Montpellier, France. León, L., Downey, G., 2006. Preliminary studies by visible and nearinfrared reflectance spectroscopy of juvenile and adult olive (Olea europaea L.) leaves. J. Sci. Food Agr. 86, 999–1004. Mailer, R.J., 2004. Rapid evaluation of olive oil quality by NIR reflectance spectroscopy. J. Am. Oil Chem. Soc. 81, 823–827. Manley, M., Eberle, K., 2006. Comparison of Fourier transform near infrared spectroscopy partial least square regression models for South African extra virgin olive oil using spectra collected on two spectrophotometers at different resolutions and path lengths. J. Near Infrared Spec. 14, 111–126. Mignani, A.G., Ciaccheri, L., Cucci, C., Mencaglia, A.A., Cimato, A., Attilio, C., Thienpont, H., Ottevaere, H., Paolesse, R., Mastroianni, M., Monti, D., Buonocore, G., Del Nobile, A., Mentana, A., Grimaldi, M.F., Dall’Asta, C., Faccini, A., Galaverna, G., Dossena, A., 2007a. Eat-by-light fiber-optic and micro-optic devices for food quality and safety assessment. Proceedings of SPIE-The International Society for Optical Engineering. Environmental Monitoring and Ecological Applications; Optical Sensors in Biological, Chemical, and Engineering Technologies. 6733, 1K/1–1K/13. Mignani, A.G., Ciaccheri, L., Thienpont, H., Ottevaere, H., Attilio, C., Cimato, A., 2007b. Towards a ‘hyperspectral optical signature’ of extra virgin olive oil. Proceedings of SPIE-The International Society for Optical Engineering. Optical Sensing Technology and Applications. 6585, 2C/1–2C/6. Moltó-García, E., Blasco-Piquer, M., Aparicio-López, R., 2003. Desarrollo de sensores para la evaluación de la calidad del aceite de oliva. Agrícola vergel: Fruticultura, horticultura, floricultura 253, 32–38. Muik, B., Lendl, B., Molina-Diaz, A., Perez-Villarejo, L., Ayora-Canada, M.J., 2004. Determination of oil and water content in olive pomace using near infrared and Raman spectrometry. Anal. Bioanal. Chem. 379, 35–41. Nazarov, I., Wample, R.L., Kaye, O., Odair-Santos, A., Goulart, K., 2005. Near Infrared Laboratory on Shoulder. Portable NIR Solutions. FRUTIC 2005. Montpellier, France. Nuñez-Sanchez, N., Perez-Marin, D., Salas-Gomez, J., Gomez-Beltran, L., Soto-Motano, J., Garcia-Olmo, J., Fernandez-Cabanas, V.M., del Moral-Trujillo, R., 2005. Determinacion instantanea mediante NIRS de los parametros de calidad de la aceituna: Acidez, rendimiento graso y humedad. Expoliva 2005. Jaén, Spain. Sato, T., 1994. Application of principal-component analysis of near-infrared spectroscopic data of vegetable oils for their classification. J. Am. Oil Chem. Soc. 71, 293–298. Szlyk, E., Szydlowska-Czerniak, A., Kowalczyk-Marzec, A., 2005. NIR spectroscopy and partial least-squares for determination of natural alphatocopherol in vegetable oils. J. Agric. Food Chem. 53, 6980–6987.
544
Takamura, H., Hyakumoto, N., Endo, N., Matoba, T., Nishiike, T., 1995. Determination of lipid oxidation in edible oils by near infrared spectroscopy. J. Near Infrared Spec. 3, 219–225. van der Lingen, J.S., 1937. The absorption spectra in the near-infrared region of olive oils. S. Afr. J. Sci. 33, 199–202. Wesley, I.J., Barnes, R.J., McGill, A.E.J., 1995. Measurement of adulteration of olive oils by near-infra-red spectroscopy. J. Am. Oil Chem. Soc. 72, 289–292. Wesley, I.J., Pacheco, F., McGill, A.E.J., 1996. Identification of adulterants in olive oils. J. Am. Oil Chem. Soc. 73, 515–518.
SECTION | I Natural Components
Yang, H., Irudayaraj, J., 2001. Comparison of near-infrared, Fourier transform infrared and Fourier transform Raman methods for determining olive pomace oil adulteration in extra virgin olive oil. J. Am. Oil Chem. Soc. 78, 889–895. Yang, H., Irudayaraj, J., Paradkar, M.M., 2005. Discriminant analysis of edible oils and fats by FTIR, FT-NIR and FT-Raman spectroscopy. Food Chem. 93, 25–32.
Chapter 59
Determination of Olive Oil Acidity Marcone Augusto Leal de Oliveira, Manoela Ruchiga Balesteros, Adriana Ferreira Faria and Fernando Antonio Simas Vaz Grupo de Química Analítica e Quimiometria, Departamento de Química, Universidade Federal de Juiz de Fora, MG, Brazil
59.1 INTRODUCTION Generally, edible oils are prone to slow degradation, resulting in FFA, which are likely to increase as time passes and temperature is raised. Therefore, the determination of the FFA level is an important parameter to monitor the quality and validity of the products. As regards olive oils, controlling the raw material is a key aspect to be considered. Olives presenting an advanced stage of maturity and/or undergoing enzymatic degradation show higher FFA levels, resulting in a higher degree of acidity and demanding olive oil classification (Figure 59.1; Table 59.1). According to the European Commission Regulation (EEC, No. 2568/91), olive oil classification is based on the degree of acidity, expressed in terms of oleic acid weight percentage (g of C18:1 (9 cis) divided by 100 g of olive oil): extra virgin and virgin olive oil must present degrees of acidity lower than 1 and 3.3%, respectively, whereas lampante olive oil acidity is above 3.3%. The best olive oils are the ones that present up to 1% degree of acidity. Those with more than 3.3% acidity need to be submitted to refining processes before they are commercialized (Figure 59.2). As market value and health benefits are directly related to olive oil acidity level, the development and optimization of analytical methodologies aimed at determining the degree of acidity of the product is relevant. AVT is the standard methodology to determine oil acidity (ECC No. 2568/91, Off. J. Eur. Communities L248, 05/09/2001, p. 6; Official and Tentative Methods of the American Oil Chemist Society, Method Cd
3d 63, American Oil Chemists Society, Champaign 1989; AOCS Official Methods of Analysis, Chapter 41, p. 13 (Oil and Fat), 940.28; AOCS Official Method Ca 5a-40, Sampling and Analysis of Commercial Fats and Oils, reapproved 1997, pp. 1 and 2), in which the sample dissolved in alcohol or ether is neutralized with KOH or NaOH, using phenolphthalein as an indicator. Alternatively, other methods have been proposed in order to replace the conventional titration such as FIA in automated systems, methods based on voltammetric reduction of quinones, FT-Raman, FT-IR, NIR, NMR, HPLC, GC and CE (Figure 59.3). The following topics present further information as regards the application of analytical methodologies for determining olive oil acidity as found in the literature. Nevertheless, the reader is advised to refer to analytical chemistry texts which contain more details related to the chemical principles, calculations and techniques according to interest (Harris, nd; Baker, 1995; Skoog et al., 1996, 1998; Harvey, 2000; Brereton, 2003) (Table 59.2).
59.2 METHODS BASED ON FLOW INJECTION ANALYSIS Flow injection analyses in automated systems such as photometric titration in non-aqueous medium and flow-reversal injection liquid–liquid extractions are performed for acidity determination taking into account adaptations from AVT procedures.
O
O
H2C
O
C O
R1
HC
O
C O
R2
H2C
O
C
R3
hydrolysis
H2C
OH
HO
C O
R1
HC
OH
HO
C O
R2
H2C
OH
HO
C
R3
RiCOOH = saturated, mono- or polyunsaturated FIGURE 59.1 Schematic reaction for a triglyceride hydrolysis. This figure illustrates how the free fatty acids are formed due to degradation process. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
545
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
546
SECTION | I Natural Components
TABLE 59.2 Key features of olive oil acidity. This table lists the features of olive oil acidity including classification motivation, standard and alternative methodologies for analysis.
TABLE 59.1 Fatty acids composition present in olive oil. This table lists typical fatty acids composition present in olive oil under ecological normal conditions (CODEX STAN 33-1981) including symbols and fatty acids percentages expressed in fatty acids methylester (FAME). Fatty acid
Symbol
% m/m (FAME)
Lauric
C12:0
Not detectable
Myristic
C14:0
⬍ 0.1
Palmitic
C16:0
7.5–20.0
Palmitoleic
C16:1(9cis)
0.3–3.5
Hepatadecanoic
C17:0
⬍ 0.5
Hepatadecenoic
C17:1(9cis)
⬍ 0.6
Stearic
C18:0
0.5–5.0
Oleic
C18:1(9cis)
55.0–83.0
1. Olives presenting advanced stage of maturity and/or undergoing enzymatic degradation show higher FFA levels, resulting in a higher degree of acidity and demanding olive oil classification 2. Classification is based on the degree of acidity, expressed in terms of oleic acid weight percentage (g of C18:1cis by 100 g of olive oil): extra virgin and virgin olive oil must present degrees of acidity lower than 1 and 3.3%, respectively, whereas lampante olive oil acidity is above 3.3% 3. Development and optimization of analytical methodologies aimed at determining the degree of acidity of the product are relevant to monitor the quality and validity of the products 4. AVT is the standard methodology to determine oil acidity 5. Alternatively, other methods have been proposed in order to replace or comp the standard methodology such as FIA, FT-Raman, FT-IR, NIR, NMR, HPLC, GC and CE O
C 10
9 OH
FIGURE 59.2 Oleic acid chemical structure. This figure shows the chemical structure for oleic acid, the majority fatty acid present in olive oil. Degree of acidity in olive oil
Analytical techniques
Classics
Automated
Spectroscopics
Separations
AVT
FIA Voltametry
RAMAN IR, NIR NMR
GC HPLC CE
FIGURE 59.3 Schematic diagram for analytical techniques used for olive oil acidity determination. This diagram shows the analytical techniques most used for olive oil degree of acidity determination.
Mariotti et al. proposed a simple and fast methodology to determine FFA level in extra virgin olive oil. In this method KOH (1.0 mmol L⫺1) is used as stream solution with phenolphthalein indicator (5.0 mmol L⫺1) prepared in methanol. The standard solutions were prepared by addition of linoleic acid to the low acidity olive oil sample. The comparison between Mariotti’s method and the ECC was performed in samples and standards, resulting in 90–107% recovery for the proposed method. Statistic error between methods ranged from 0.3 to 4.1%, whereas the calculated error between methods ranged from 0.3 to 10% for the
FFA: free fatty acids; AVT: alkaline volumetric titration; FIA: flow injection analysis; FT-Raman: Fourier transform Raman spectroscopy; FT-IR: Fourier transformed infrared spectroscopy; NIR: near infrared spectroscopy; NMR: nuclear magnetic resonance spectroscopy; HPLC: high performance liquid chromatography; GC: gas chromatography; CE: capillary electrophoresis.
samples. Therefore, this methodology allows FFA determination in olive oil samples with low solvent volumes, small samples, short analysis time and in the absence of toxic solvents (Mariotti and Mascini, 2001). Bonastre et al. (2004) developed a new methodology based on the combination between FIA and distribution system for chemistry quality control in olive oil (degree of acidity determination and peroxides). Total acidity was determined by FIA-titrimetry in the absence of a mixing chamber. In the present case, the sample is diluted in an ethanol/ether solution and injected into a flow-stream containing alcoholic KOH solution with phenolphthalein as indicator. The absorbance decrease at 533 nm was measured. Standard traditional methods were used to make simultaneous analysis of the olive oil in order to ensure the suitability of the developed method. The statistical tests were carried out and no significant differences were verified in the 95% confidence interval between FIA-titrimetry and the AOAC standard method (Bonastre et al., 2004). Zhi et al. (1996) proposed a continuing liquid–liquid extraction system based on simple automatic injection mode under reverse flow and single channel in the absence of segmentation or isolated units for direct determination of FFA in olive oil samples expressed in oleic acid concentration. The methodology is based on the reaction between Cu(II)-pyridine and FFA, forming a new complex which is
CHAPTER | 59 Determination of Olive Oil Acidity
extracted into the organic phase in a reverse flow process. The Cu–FFA complex was monitored at 716 nm using a spectrophotometer. The method accuracy was evaluated in comparison with standard titrimetric methodology. Because olive oil has many FFA, which, in turn, have a molar absorptivity at a particular wavelength, the result obtained by the proposed method was 18% lower than the standard method. For this reason, a correction factor is used to make up for this limitation (Zhi et al., 1996). Nouros et al. (1997) proposed a titration method using FIA. In this method, a KOH carrier containing phenolphthalein as indicator diluted in n-propanol solution was used. The FIA-titration system was based on a single channel. In this methodology a calibration curve from 0.100 to 10 a.d. was performed. At the beginning, absorbance was measured by staining with KOH-phenolphthalein stream. After adding the sample, the reaction occurs and the medium becomes more acid, causing the phenolphthalein to change its color from cherry to colorless. In order to evaluate the method, oleic acid recovery tests in olive oil were carried out. The values obtained ranged from 97.9 to 101.6%. The precision obtained during the recovery study by triplicate injections ranged between 1.2 and 2.1 RSD%. The comparison between the proposed methodology, the standard method and FT-IR was carried out in order to evaluate its applicability. Thirty-two samples were tested and the results of the determination for FFA were obtained. The comparison between flow injection and the standard method was described by a linear regression which had a correlation coefficient r ⫽ 0.99996, specifying the accuracy of the method. A comparison between FIA and FT-IR presented an r ⫽ 0.9999. Therefore, the proposed method by Nouros et al. was shown to be fast, precise and accurate for FFA analysis in olive oil (Nouros et al., 1997).
59.3 METHODS BASED ON ELECTROCHEMICAL DETECTIONS Little attention has been paid to electrochemical analytical methods, as the majority of fatty acids (FA) have low electrochemical activity. However, some work on quinone reductions has been developed in the last 14 years. Kusu et al. (1994) developed a new method for determining the FFA content in fats and oils, based on the pre-peak voltammetric reduction of 2-methyl-1,4-naphthoquinone (VK3) due to the presence of excess FA. The total FFA amount in the samples was determined based on the proportionality of the pre-peak height with the acid concentration. Although this method is far superior to the conventional titration method for sensitivity and accuracy, the process spent more application time in quality control cases (Kusu et al., 1994). Takamura et al. (1995) proposed a method by flowinjection analysis–electrochemical detection (FIA-ECD) as
547
a quick way to determine the FFA content in fats and oils. The FIA analysis with electrochemical detection used an electrochemical cell fabricated from a glassy carbon working electrode, an Ag/AgCl, KCl (sat.) reference electrode, and a stainless-steel auxiliary electrode. The cell volume was 2.4 μL. The flow lines were made of polytetrafluoroethylene tubing (0.5 mm I.D.) shielded with aluminum foil to protect the sample from light. The tubing length from the injection of the sample up to the detector was 1 m. In the FIA-ECD method, the ethanol solution containing 3 mmol L⫺1 VK3 and 38 mmol L⫺1 LiClO4 was used as a carrier solution at a flow-rate of 0.6 mL min⫺1. Air was removed from this solution by degassing under reduced pressure. The detection potential for monitoring the acidity was kept at ⫺0.33 V versus Ag/AgCl. The detection system remained stable for about 6 months. The FIA-ECD method was applied for determining the FFA content in several fat and oil samples, among which we highlight the olive oil. Tests for recovery were made with the addition of standard linoleic acid in adequate quantities to the samples. The recovery of linoleic acid in olive oil was 98%. The results obtained from the FIA-ECD method showed good correlation with the values obtained by voltammetric and potentiometric titration (r ⫽ 0.999). The FIA-ECD method presented the same advantages as the voltammetric method, such as high sensitivity (detection limit: 25 pmol test⫺1), high reproducibility, lower consumption of sample compared to the titration and also presented a higher rate of sampling (about 30 samples h⫺1) showing itself to be very useful for determination of the degree of acidity of oils and fats in quality control processes (Takamura et al., 1995). Fuse et al. (1997a, b) proposed a high performance liquid chromatography–electrochemical detection (FIA-ECD) system, which used an electrochemical cell made from a glassy carbon working electrode, saturated calomel electrode (SCE), reference electrode and a stainless-steel auxiliary electrode. Carrier stream tubing of stainless-steel (0.5 mm I.D.) and polytetrafluoroethylene (0.5 mm I.D.) covered with aluminum foil (to shield the light) were used. The sample was dissolved in ethanol/acetonitrile (10:90). When the sample was not soluble, the mixture was centrifuged and the supernatant was used as a sample solution. An aliquot of 20 μL sample and standard solutions of acids were injected into an octadecylsilyl (ODS) column at room temperature. The detection potential for monitoring FFA, based on free fatty acid’s hydrodynamic voltammograms, was determined as ⫺415 mV versus SCE. The separation of higher FA was performed by using reverse-phase ODS column and a mobile phase consisted of mixing ethanol/ acetonitrile. The dissolved oxygen in the mobile phase and the solution of quinone was removed by degassing. Twenty μl of the solution containing FA was injected onto the column; the eluent was mixed with the solution of quinone and the FA were detected by ECD. The influence of the ratio of ethanol/acetonitrile on the separation and sensitivity was investigated and it was found that the higher the proportion
548
of acetonitrile, the better the separation of the peaks of the acids. The maximum peak height was achieved with 90% acetonitrile for all FA analyzed. Therefore, the proportion of ethanol/acetonitrile (10:90) was selected for the composition of the mobile phase. Linoleic, oleic, palmitic and stearic acids were separated in 15 minutes, the peak height was linear, with the amount injected over the range of 20–1200 pmol, the detection limit was 20 pmol and the relative standard deviation (n ⫽ 5) was within the range of 1.4 to 2.6% for the four analyzed acids. In order to evaluate the recovery, a standard mixture containing palmitic, stearic, oleic and linoleic acids was added to each sample. The recovery for an olive oil sample was 105%, 106%, 96% and 105% for palmitic, stearic, oleic and linoleic acid, respectively (Fuse et al., 1997a, b). FIA methods using electrochemical detection for FFA determination in various matrices were reviewed by Takamura et al. (1999) and methods involving HPLC-ECD systems to determine the degree of acidity in different samples were reviewed by Kotani et al. (2002).
59.4 SPECTROSCOPIC METHODS Molecular spectroscopy techniques, such as FT-Raman, FTIR, NIR and NMR, have been shown to be useful for quality control of olive oils and as alternative methods for quantitative determination of acidity. Moreover, these techniques associated with chemometric procedures have been used to identify adulteration of virgin olive oils by different vegetable oils or by olive pomace oils, to differentiate vegetable oils and to classify them according to their geographical origin (Baeten et al., 1996, 1998; Marigheto et al., 1998; Bertran et al., 2000; Mannina et al., 2001; Downey et al., 2003; Marquez et al., 2005; Yang et al., 2005; Wang et al., 2006).
59.4.1 Raman Spectroscopic Methods Baeten et al. used FT-Raman associated with multivariate procedures to predict the level of adulteration in a group of samples of olive oil that had been adulterated with soybean oil, cottonseed oil and the raw waste of olive oil (olive oil pomace) in a proportion that varied from 1.5 to 10%. The best result for the prediction of adulteration was a fit of R2 ⫽ 0.964 through the use of principal component regression (PCR). It was possible to discriminate between genuine and adulterated samples 100% correctly and 91.3% correctly as to classification between different grades of classification (Baeten et al., 1996). Muik et al. proposed a methodology using FT-Raman in combination with PLS for the determination of FFA in samples of olives and olive oil. The FA content predicted was used to classify olive oil and olives in different categories according to European Union regulations. Ninety percent of
SECTION | I Natural Components
the oil samples and 80% of the olives were correctly classified. These results demonstrate that the proposed procedures can be used for the screening of good-quality olives before processing, as well as for the on-line control of the produced oil (Muik et al., 2003).
59.4.2 Infrared Spectroscopic Methods Bertran et al. determined FFA content in olive oil, based on Fourier transformed infrared attenuated reflectance (FTIRATR). They used spiked samples in order to extend the range of FFA contents of samples by adding oleic acid to several virgin and pure olive oils, from 0.1 to 2.1% acidity. Different calibration models were tested using partial least squares (PLS) regression, in two wave number ranges (1775–1689 and 1480– 1050 cm⫺1) and with several data treatments (first and second derivative; SNV). The best results were obtained by splitting the calibration range into two concentration intervals (0.1–0.5 and 0.5–2.7%), using the SNV as a treatment, in the spectral range of 1775–1689 cm⫺1, with three PLS components. Sample throughput was around 30 h⫺1 (Bertran et al., 1999). Iñón et al. proposed a method for FFA determination in commercial olive oil samples of different origins by FTIRATR using hierarchical cluster analysis. The prediction capabilities of FTIR-ATR data were evaluated after using PLS multivariate calibration methods, net analyte signal preprocessing followed by PLS or classical least squares (CLS) regression methods. Using a calibration set of 16 samples, the properties of 28 samples, analyzed in triplicate, were predicted with a relative precision of ⫾0.017 wt.%. The mean difference between predicted and actual values and the standard deviation of mean differences were ⫺0.001 and 0.037 wt.%, respectively (Iñón et al., 2003). Cañada et al. proposed a flow injection system with FT-IR for the rapid determination of the FFA content in edible oils. The method is based on complete deprotonation of the FFA and evaluation of the intensity of the ηass.(COO⫺) band formed at 1570 cm⫺1 with respect to a baseline point at 1534 cm⫺1. Upon deprotonation, the absorption of the carboxylic acid moieties is shifted toward lower frequency, thus avoiding spectral interferences of the η –C⫽O from esters, ketones, or aldehydes also present in the sample. The developed method was applied to the determination of the total amount of FFA in olive, sunflower, and corn oils, delivering results that were in good agreement with those obtained by the official reference method proposed by the European Community. Due to its high degree of automation and throughput (40 samples h⫺1), the developed method is suitable for quality control in routine applications (Cañada et al., 2001). Armenta et al. proposed a chemometric method for the determination of acidity in olive oil of different types and origins by using near infrared (NIR) measurements. Different methods for selecting the calibration set, after a hierarchical cluster analysis, were applied. The detection limit of
549
CHAPTER | 59 Determination of Olive Oil Acidity
the methodology developed was 0.03% for acidity, requiring less than 30 s per sample without any previous treatment (Armenta et al., 2007).
59.4.3 Nuclear Magnetic Resonance Method Indirect determination of FFA in olive oil through a measurable amount of diglycerides can be achieved using 1H NMR spectrum in CDCl3/DMSO. The resonances of sn 1,2 and sn 1,3 diglycerides at 3.639 and 3.994 ppm have been monitored for this purpose (Mannina et al., 2003).
59.5 SEPARATION METHODS Most of the current work on edible oil analysis is based on HPLC, GC and CE analyses are widely used for the determination of individual FA content, sterols or pesticide residues in oils and to aid in the identification, quality control and to detect adulteration in edible oils (Minutilli, 1963; Hein and Isengard, 1997; Mannina et al., 1999; de Oliveira et al., 2001, 2003). However, only few papers are dedicated to the determination of FFA in olive oils by these methods. In general, chromatographic analyses of FA and FFA are similar, differing basically in the sample preparation. Depending on the information required, the FFA may be negligibly small, or they may be converted together with the lipid-bound FA to detectable species, or measured separately, often after preliminary fractionation into lipid classes. Hence, the procedure adopted for determining acidity may include extraction of the natural FFA from the oil before analysis. Afterward, the free oleic acid may be separated from other FFA through the separation method.
59.5.1 Gas Chromatography Minutilli (1963) used GC to determine total fatty acids (FFA and FA from the neutral fraction) of a lampante olive oil having 10% acidity. The amounts of palmitic, palmitoleic, stearic, oleic, and linoleic acids present in the different fractions were given separately. Lotti and Petronici (1967) determined FFA in olive, soybean, sunflower and pistachio oils. Although, to our knowledge, there are few reports on the GC determination of natural FFA in olive oils, the GC technique seems to be useful for its determination in those samples. Clearly, the overwhelming majority of FA profiles of lipids are determined by GC after derivatization, in which FA are usually converted to FAME. These are prepared by methylation in a methanolic medium that can be carried out with alkaline, acid, or alkaline and acid catalysis. With acid catalyst, both bound and FFA are converted concurrently to FAME. BF3 is the acid most commonly used, and FFA are converted to FAME in 2 min at 100°C. Recently, much shorter reaction times have been shown to suffice if conventional heating is replaced by microwave irradiation.
Other alternatives are methylation with diazomethane, CH2N2, which is used for FFA, and the formation of methyl esters without previous extraction of the fat (Boschelle et al., 1992). Diazomethane has long been used for the rapid esterification of FFA. Special reagents, procedures, and apparatus permit relatively safe operation with this compound despite its toxic and explosive nature. It has generally been assumed to react selectively with FFA and has been proposed for their determination in plasma. The carrier gas selected affects the resolution and analysis time. For a given degree of resolution, the shortest analysis times will be carried out with hydrogen gas, followed by helium, assuming one is operating at flow-rates above the optimum linear velocity, as is usual. Helium is often preferred, however, for safety reasons. The purity of the carrier gas is also an important factor to consider, as oxygen and water impurities can degrade stationary phases, which are exacerbated at higher temperatures. Thus, the installation of oxygen and water traps in the carrier gas line is usually a worthwhile investment. GC methods generally employ capillary columns. The wall-coated open-tubular (WCOT) columns typically used for the FA analysis are 25–30 m long, with an inner diameter of about 0.25 mm. Columns of 100 m may be essential for the most challenging separations when the highest resolution is required, such as the separation of the positional and geometrical isomers of unsaturated FA. In GC, the flame ionization detector (FID) is the most common detector used. To confirm the identity of an analyte FA it is also becoming increasingly routine to utilize gas chromatography mass spectrometry (GC-MS), in order to compare its mass spectrum with a reference spectrum stored in a computer databank. It constitutes one of the most definitive methods for the identification of complex organic compounds. Identification of the abundant, most commonly encountered FA should be based on the retention times of FAME and compared with authentic standards that are commercially available. These should cover the range C4 to C24, and include both saturated and unsaturated compounds.
59.5.2 High Performance Liquid Chromatography Fuse et al. (1997a, b) developed a system of HPLC for the separation and determination of FFA in olive, camellia, corn, rapeseed and soybean oils, which was found not only sensitive and reproducible but also a simple means for separating and determining FFA without derivatization. Kroumova and Wagner (1995) described a simple HPLC system to separate and quantify short-, medium- and long-chain FFA in plant-derived oils and an apparatus for the subsequent decarboxylation of these for a mechanistic study of the Schmidt degradation. Many other reports are devoted to the analysis of FFA in serum and plasma.
550
In HPLC a large number of different derivatives are proposed for FA analysis according to the detection mode, which is generally ultraviolet, fluorescence or chemiluminescence. A number of derivatives have been proposed for HPLC analysis of FA including benzyl, p-nitrobenzyl, phenacyl, p-bromophenacyl, p-methylthiobenzyl and 1naphthylamine esters, among others (Lima and Abdalla, 2002). Naphthacyl esters were studied with the objective of increasing the detection sensitivity. It was used for the routine determination of FFA in human serum. Substituted hydrazines have been used to derivatize FA, which were employed in the determination of FFA in serum. One of the most sensitive reagents for FFA is 3-bromomethyl-6,7-dimethoxy-l-methyl-2(H)-quinoxalinone, which has detection limits of 0.3–1 fmol and gives good resolution on account of its relatively small size. The 3-propionylcarboxylic acid hydrazide of this reagent was found to be more stable and to enable derivatization of FFA in an aqueous medium, without prior extraction. Another sensitive and specific reagent is 2-(2,3-naphthalimido)ethyl trifluoromethanesulfonate. Fluorescent derivatives have found broad applications in the biomedical field, particularly for the determination of low FA concentrations. Thus, it is one of the most widely used techniques for FFA analysis. Compared to UV detection, fluorometric and chemiluminescence detections via suitable derivatives of FA amplify the sensitivity by one to two orders of magnitude. The selectivity is also greatly enhanced but the resolution is generally diminished somewhat. Other techniques include ED, which is known to provide good sensitivity and specificity with the advantage that it can also be used without prior derivatization; evaporative light scattering detector (ELSD); and the most recent technique, liquid chromatography mass spectrometry (LCMS) that provides high sensitivity and specificity in the detection of FA. The C12-C18 FFA have been separated by reversed-phase ion-pair HPLC with conductivity detection (Tsuyama et al., 1992). Underivatized FA and their methyl esters have been separated by HPLC and monitored by refractive index or, more sensitively, by low-wavelength UV detection using water and acetonitrile as eluents. For analytical applications it is therefore advantageous to derivatize the FA with a reagent possessing a high molar absorptivity at longer UV wavelengths. Most of the separations by HPLC of FFA or their derivatives are performed on reversed-phase (RP) systems, which consist of alkyl chains of various lengths bonded onto a silica base. Both the retention times and the selectivity increase as the alkyl chain of the bonded phase is lengthened, as was convincingly demonstrated with a synthesized C30 RP. Because of their commercial availability, ODS phases have generally been favored for FA separations. Octylsilyl ones have also found some applications,
SECTION | I Natural Components
but shorter ones are seldom used. Since these are non-polar phases, the order of FA elution in HPLC corresponds to that of GC with non-polar phases, but with the unsaturated being eluted considerably ahead of their saturated analogs. For FA derivatives, it was found that acetonitrile (ACN), relative to methanol, increased retention times of carbon chains longer than C8, whereas methanol was more effective for the shorter chain lengths. In many cases a mixture of acetonitrile and methanol with water is used to achieve optimum resolution. Thus, the ability to modify retention times by adding other organic solvents and various buffers to the eluent gives HPLC more flexibility than GC in this respect.
59.5.3 Capillary Electrophoresis Recently, an alternative method for the determination of olive oil acidity using CE, a separation technique based on differentiate migration of neutral, ionic or ionizable compounds, under electric field application in a fused silica capillary tube containing a suitable electrolyte solution (Harris, nd; Baker, 1995) was proposed. The method is based on ethanolic extraction at 60°C of the long-chain FFA followed by electrokinetic chromatography (EKC) in pH 6.86 phosphate buffer (15 mmol L⫺1) containing: 4 mmol L⫺1 sodium dodecyl benzene sulfonate (SDBS), 10 mmol L⫺1 Brij 35, 2% v/v 1-octanol and 45% v/v ACN under indirect UV detection at 224 nm. This electrolyte achieved successful separation of myristic acid (C14:0) internal standard (IS) and olive oil major components C16:0, C18:1(9cis) and C18:2(9cis, 12cis) in less than 8 min (Balesteros et al., 2007). The procedure for the quantitative determination of oil acidity requires the calculation of the C18:1(9cis), RF, described by the following mathematical expression: A C18:1(9cis ) [C18:1(9cis)]
⫽ RF
A C14 :0 [C14:0]
(59.1)
where: AC18:1(9cis) is the C18:1(9cis) area, AC14:0 the C14:0 area, [C18:1(9cis)] the C18:1(9cis) concentration in mmol L⫺1 and [C14:0] is the C14:0 concentration, fixed at 0.50 mmol L⫺1. The oil acidity percentage is obtained after working Equation 59.1 and using the definition of % acidity (g C18:1(9cis) divided by 100 g oil) as follows: %acidity ⫽
MMC18:1(9cis ) [C14:0]VΣA x i R F A C14:0 m
102 (59.2)
where: Axi is the area of FA x in the sample with i varying from 1 to 3: (1) C18:1(9cis), (2) C16:0 and (3) C18:2(9cis,12cis), V the flask volume (L), AC14:0 the C14:0 IS area, m the oil sample mass (mg), RF the response factor of C18:1(9cis) (analytical curve slope), [C14:0] the IS
551
CHAPTER | 59 Determination of Olive Oil Acidity
59.6 FINAL CONSIDERATIONS
C14:0
16 mAU
C18:2cc C16:0 C18:1c
C18:0
A
C18:2cc C16:0 C18:1c
4
B
6
Several analytical methodologies for olive oil acidity determination expressed in FFA concentration have been considered. The scenario reported in the literature presents analytical methodologies as simple as volumetric titration up to the more sophisticated techniques such as spectroscopic and separation methods. Researchers, students and professionals interested in olive oil determinations have a wide variety of analytical methods available, from which they can choose, taking into account necessity, cost, accessibility, analysis time (number of samples analyzed per hour), sample preparation mode (with or without previous treatment) and sensitivity.
ACKNOWLEDGMENTS 8
Time (min)
FIGURE 59.4 Electropherograms of (A) FA standard solution and (B) extra virgin olive oil sample. Operational conditions: fused-silica capillary (Polymicro Technologies, Phoenix, AZ, USA) with 48.5 cm total length, 40.0 cm effective length and 75 μm i.d., 375 μm o.d. were used. Operational conditions: 4 s injection at 12.5 mBar pressure, ⫹20 kV applied voltage, 25°C cartridge temperature and indirect detection at 224 nm. Electrolyte: 15 mmol L⫺1 phosphate buffer at pH 6.86, 4 mmol L⫺1 SDBS, 10 mmol L⫺1 Brij 35®, 45% v/v ACN and 2% v/v 1-octanol. Reproduced from Balesteros et al. (2007) with permission.
concentration at 0.50 mmol L⫺1, and MMc18:1(9cis) is the molecular mass of C18:1(9cis). Figure 59.4 shows the separation profile of a typical extra virgin olive oil sample extract. The reliability of the proposed method was further investigated by the determination of the acidity of an extra virgin olive oil sample in comparison to the standard established methodology. No statistical differences were found within 95% confidence level. A % acidity of 0.39 ⫾ 0.02 was found for the olive oil sample under consideration.
59.5.4 Other Separation Methods Other methods, including supercritical fluid chromatography (SFC), have found only limited application for FA profiling. In many aspects SFC is a hybrid between GC and HPLC, especially in terms of the ability to use both capillary GC and packed HPLC columns and FID and UV detectors. Carbon dioxide, alone or with modifiers, is usually used as the mobile phase. SFC has been used in lipid analysis, but only to a limited extent for FA and their derivatives. The subject has been reviewed recently. FAME and FFA have been separated by SFC on capillary and packed columns, in both reversed-phase and normal-phase modes. The ELSD can be used with SFC.
The authors acknowledge Dr. Richard Grazul for proofreading, Dr. Heitor Abreu and Dra. Renata Diniz for revisions and suggestions. Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq - 476386/2007-1 and 300593/2008-2), Fundação de Amparo à Pesquisa do Estado de Minas Gerais of Brazil (FAPEMIG – CEX-APQ 1906-502/07, CEX APQ 01837/08) for fellowships and financial support and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) by electronic periodic availability.
REFERENCES Armenta, S., Garrigues, S., et al., 2007. Determination of edible oil parameters by near infrared spectrometry. Anal. Chim. Acta 596 (2), 330–337. Baeten, V., Hourant, P., et al., 1998. Oil and fat classification by FTRaman spectroscopy. J. Agric. Food Chem. 46 (7), 2638–2646. Baeten, V., Meurens, M., et al., 1996. Detection of virgin olive oil adulteration by Fourier transform Raman spectroscopy. J. Agric. Food Chem. 44 (8), 2225–2230. Baker, D.R., 1995. Capillary Electrophoresis. John Wiley & Sons, Inc. Balesteros, M.R., Tavares, M.F.M., et al., 2007. Determination of olive oil acidity by CE. Electrophoresis 28, 3731–3736. Bertran, E., Blanco, M., et al., 1999. Determination of olive oil free fatty acid by Fourier transform infrared spectroscopy. J. Am. Oil Chem. Soc. 76 (5), 611–616. Bertran, E., Blanco, M., et al., 2000. Near infrared spectrometry and pattern recognition as screening methods for the authentication of virgin olive oils of very close geographical origins. J. Near Infrared Spec. 8 (1), 45–52. Bonastre, A., Ors, R., et al., 2004. Advanced automation of a flow injection analysis system for quality control of olive oil through the use of a distributed expert system. Anal. Chim. Acta 506 (2), 189–195. Boschelle, O., Mozzon, M., et al., 1992. Determination method of free fatty acids in small samples. Application to the oil extracted from olive drupe. Riv. Ital. Sost. Grasse 69 (5), 257–261. Brereton, R.G., 2003. Chemometrics. Data Analysis for the Laboratory and Chemical Plant. John Wiley & Sons, Ltd, Chichester. Cañada, M.J.A., Medina, A.R., et al., 2001. Determination of free fatty acids in edible oils by continuous-flow analysis with FT-IR spectroscopic detection. Appl. Spectrosc. 55 (3), 356–360.
552
de Oliveira, M.A.L., Micke, G.A., et al., 2001. Factorial design of electrolyte systems for the separation of fatty acids by capillary electrophoresis. J. Chromatogr. A 924 (1–2), 533–539. de Oliveira, M.A.L., Solis, V.E.S., et al., 2003. Method development for the analysis of trans-fatty acids in hydrogenated oils by capillary electrophoresis. Electrophoresis 24 (10), 1641–1647. Downey, G., McIntyre, P., et al., 2003. Geographic classification of extra virgin olive oils from the eastern Mediterranean by chemometric analysis of visible and near-infrared spectroscopic data. Appl. Spectrosc. 57 (2), 158–163. Fuse, T., Kusu, F., et al., 1997a. Determination of acid values of fats and oils by flow injection analysis with electrochemical detection. J. Pharmaceut. Biomed. 15 (9–10), 1515–1519. Fuse, T., Kusu, F., et al., 1997b. Determination of higher fatty acids in oils by high-performance liquid chromatography with electrochemical detection. J. Chromatogr. A 764 (2), 177–182. Harris, D.C. nd. Quantitative Chemical Analysis. Harvey, D., 2000. Modern Analytical Chemistry. McGraw-Hill, Inc, Dubuque. Hein, M., Isengard, H.D., 1997. Determination of underivated fatty acids by HPLC. Z. Lebensm. Unters. F. A 204 (6), 420–424. Iñón, F.A., Garrigues, J.M., et al., 2003. Selection of calibration set samples in determination of olive oil acidity by partial least squaresattenuated total reflectance-Fourier transform infrared spectroscopy. Anal. Chim. Acta 489 (1), 59–75. Kotani, A., Kusu, F., et al., 2002. New electrochemical detection method in high-performance liquid chromatography for determining free fatty acids. Anal. Chim. Acta 465 (1–2), 199–206. Kroumova, A.B., Wagner, G.J., 1995. Methods for separation of free, short, medium, and long-chain fatty-acids and for their decarboxylation. Anal. Biochem. 225 (2), 270–276. Kusu, F., Fuse, T., et al., 1994. Voltammetric determination of acid values of fats and oils. J. AOAC Int. 77 (6), 1686–1689. Lima, E.S., Abdalla, D.S.P., 2002. High-performance liquid chromatography of fatty acids in biological samples. Anal. Chim. Acta 465 (1–2), 81–91. Lotti, G., Petronici, C., 1967. Glyceride and free fatty acid composition of vegetable oils. Glyceride and free fatty acid composition of vegetable oils, 67(3), 125–136 Mannina, L., Patumi, M., et al., 1999. Olive and hazelnut oils: A study by high-field H-1 NMR and gas chromatography. Ital. J. Food Sci. 11 (2), 139–149.
SECTION | I Natural Components
Mannina, L., Patumi, M., et al., 2001. Geographical characterization of Italian extra virgin olive oils using high-field H-1 NMR spectroscopy. J. Agric. Food Chem. 49 (6), 2687–2696. Mannina, L., Sobolev, A.P., et al., 2003. Olive oil as seen by NMR and chemometrics. Spectrosc. Eur. 15 (3), 6–14. Marigheto, N.A., Kemsley, E.K., et al., 1998. A comparison of midinfrared and Raman spectroscopies for the authentication of edible oils. J. Am. Oil Chem. Soc. 75 (8), 987–992. Mariotti, E., Mascini, M., 2001. Determination of extra virgin olive oil acidity by FIA-titration. Food Chem. 73 (2), 235–238. Marquez, A.J., Diaz, A.M., et al., 2005. Using optical NIR sensor for on-line virgin olive oils characterization. Sensor Actuat. B-Chem. 107 (1), 64–68. Minutilli, F., 1963. Free acidity in bright olive oil. Rass. Chim. 15 (4), 160–162. Muik, B., Lendl, B., et al., 2003. Direct, reagent-free determination of free fatty acid content in olive oil and olives by Fourier transform Raman spectrometry. Anal. Chim. Acta 487 (2), 211–220. Nouros, P.G., Georgiou, C.A., et al., 1997. Automated flow injection spectrophotometric non-aqueous titrimetric determination of the free fatty acid content of olive oil. Anal. Chim. Acta 351 (1-3), 291–297. Skoog, D.A., Holler, F.F., et al., 1998. Principles of Instrumental Analysis. Saunders College Publishing. Skoog, D.A., West, D.M., et al., 1996. Fundamentals of Analytical Chemistry. Saunders College Publishing. Takamura, K., Fuse, T., et al., 1999. A review of a new voltammetric method for determining acids. J. Electroanal. Chem. 468 (1), 53–63. Takamura, K., Fuse, T., et al., 1995. Determination of the free fatty-acid content in fats and oils by flow-injection analysis with electrochemical detection. Anal. Sci. 11 (6), 979–982. Tsuyama, Y., Uchida, T., et al., 1992. Analysis of underivatized c-12-c18 fatty-acids by reversed-phase ion-pair high-performance liquid-chromatography with conductivity detection. J. Chromatogr. 596 (2), 181–184. Wang, L., Lee, F.S.C., et al., 2006. Feasibility study of quantifying and discriminating soybean oil adulteration in camellia oils by attenuated total reflectance MIR and fiber optic diffuse reflectance NIR. Food Chem. 95 (3), 529–536. Yang, H., Irudayaraj, J., et al., 2005. Discriminant analysis of edible oils and fats by FTIR, FT-NIR and FT-Raman spectroscopy. Food Chem. 93 (1), 25–32. Zhi, Z.L., Rios, A., et al., 1996. An automated flow-reversal injection liquid-liquid extraction approach to the direct determination of total free fatty acids in olive oils. Anal. Chim. Acta 318 (2), 187–194.
Chapter 60
Application of the Electronic Nose in Olive Oil Analyses M. Stella Cosio, Simona Benedetti, Susanna Buratti, Matteo Scampicchio and Saverio Mannino Department of Food Science and Technologies, University of Milan, Italy
60.1 ELECTRONIC NOSE APPARATUS At the beginning of the 1990s the term ‘artificial’ or ‘electronic nose’ appeared. More extended research began, and applications, especially in the food industry, could be tested. Gardner and Bartlett (1993) defined the electronic nose as ‘an instrument, which comprises an array of electronic chemical sensors with partial specificity and appropriate pattern-recognition system, capable of recognising simple or complex odours’. Several commercial intelligent gas sensor array instruments are now available on the market covering a variety of chemical sensor principles, system design and data analysis techniques. Operationally, an electronic nose is a ‘sensing system’ comprised of three parts: a sampling system, an array of chemical gas sensors producing an array of signals when confronted with a gas, vapor, or odor, and an appropriate pattern-classification system. The ideal sensors to be integrated in an electronic nose should fulfill the following criteria: high sensitivity towards chemical compounds, that is, similar to that of the human nose (down to 10⫺12 g mL⫺1), low sensitivity towards humidity and temperature; medium selectivity, they must respond to different compounds present in the headspace of the sample; high stability; high reproducibility and reliability; short reaction and recovery time; robust and durable; easy calibration; easily processable data output; small dimensions (Schaller et al., 1998). By chemical interaction between odor compounds and the gas sensors the state of the sensors is altered giving rise to electrical signals which are registered by the instrument. In this way the signals from the individual sensors represent a pattern which is unique for the gas mixture measured and is interpreted by multivariate pattern recognition techniques like, for example, the artificial neural network (Figure 60.1). Samples with similar odors generally give rise to similar sensor response patterns and samples with different odors show differences in their patterns. The sensors of an Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
electronic nose can respond to both odorous and odorless volatile compounds. Various kinds of gas sensors are available, but only four technologies are currently used in commercialized electronic noses: metal oxide semiconductors (MOS); metal oxide semiconductor field effect transistors (MOSFET); conducting organic polymers (CP); piezoelectric crystals (Bulk Acoustic Wave–BAW, Surface Acoustic Wave– SAW). Others, such as fiber-optic, electrochemical and bi-metal sensors, are still in the developmental stage and may be integrated in the next generation of the electronic noses. In all cases the goal is to create an array of differentially sensitive sensing elements (Mannino et al., 2007). It must be pointed out that the electronic nose methodology is able to show the authenticity of the product only if supported by the multivariate statistical techniques useful not only to classify but also to validate and predict unknown samples. The data processing of the multivariate output data generated by the gas sensor array signals represents another essential part of the electronic nose concept. The statistical techniques used are based on commercial or specially designed software using pattern recognition routines like principal component analysis (PCA), cluster analysis (CA), partial least squares (PLS), linear discriminant analysis (LDA) and artificial neural network (ANN). There are striking analogies between the human nose and the electronic nose. Comparing the two is instructive. The human nose uses the lungs to bring the odor to the epithelium layer; the electronic nose has a pump. The human nose has mucus, hairs, and membranes to act as filters and concentrators, while the electronic nose has an inlet sampling system that provides sample filtration and conditioning to protect the sensors and enhance selectivity. The human epithelium contains the olfactory epithelium, which has millions of sensing cells, selected from 100–200 different genotypes that interact with the odorous molecules in unique ways. The electronic nose has a variety of sensors that interact differently with the sample. The human
553
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
554
SECTION | I Natural Components
(Buratti et al., 2005). Each sample has been evaluated in triplicate and the average of the results has been used for subsequent statistical analysis.
60.2 EVALUATION OF DIFFERENT STORAGE PERIODS AND CONDITIONS IN EXTRA VIRGIN OLIVE OILS
Single sensor change in resistance
Sensor responses
FIGURE 60.1 Schematic diagram of electronic nose during the experiment.
receptors convert the chemical responses to electronic nerve impulses. The unique pattern of nerve impulses is propagated by neurons through a complex network before reaching the higher brain for interpretation. Similarly, the chemical sensors in the electronic nose react with the sample and produce electrical signals. A computer reads the unique pattern of signals, and interprets them with some form of ‘intelligent’ pattern classification algorithm. In these studies the analyses have been conducted on extra virgin olive oil with a commercial electronic nose (model 3320 Applied Sensor Lab Emission Analyser, Applied Sensor Co., Linkoping, Sweden), comprising three parts: automatic sampling apparatus, detector unit containing the array of sensors, and software for data recording. The automatic sampling system supports a carousel of 12 sites for loading the samples and permits the control of internal temperature. Twenty-two different sensors compose the sensor array. Ten sensors are metal oxide Semiconductor Field Effect Transistors (MOSFET); twelve are Taguchi-type sensors (Metal Oxide Semiconductors – MOS). The MOSFET sensors are divided into two arrays of five sensors each, one array operating at 140 °C and the other at 170 °C, while the MOS are kept at 400–500 °C during all the process phases. Aliquots of 1 gram of each sample have been introduced in 40 mL Pirex® vials with pierceable Silicon/Teflon disks in the cap. The measurement sequence has started with the sample incubation at 40 °C for 10 min before injection. After a headspace generation, the volatile compounds have been sampled by an automatic syringe and have been pumped over the sensor surfaces for 60 s. During this time the sensor responses have been recorded. Then, sensors have been exposed to filtered air at a constant flow-rate (60 mL min⫺1) in order to keep the gas sensor signal back to the baseline. Each measurement cycle includes an internal standard as reference for the calibration method. In this work, the sample volume, the incubation temperature, time of sampling step, injection time and temperature optimization procedures were studied in order to obtain responses of suitable intensity and good reproducibility
Extra virgin olive oil (EVOO) is properly processed from fresh and mature fruit of the olive (Olea europea L.) of good quality and it presents a complex flavor which is greatly liked by native consumers and internationally appreciated by gourmets (Kiritsakis and Min, 1989). These sensorial properties decrease during storage because of oxidation processes affected by air, heat, light and metals. Oxidation is a very complex phenomenon and volatile compounds, responsible for the pleasant flavor, change in off-flavors (Morales et al., 1997; Angerosa et al., 1999). Consequently, it is a matter of great concern for the olive oil industry to preserve the positive attributes of extra virgin olive oil over time from production to bottling, up to purchasing. Nowadays, there are different methods used and/or proposed for evaluating the oxidative deterioration of olive oil. Among the routine methods, there are the peroxide value (PV), which determines the amount of primary oxidation products, UV absorbance at 232 and 270 nm and ΔK, which measures the formation of conjugated dienes and trienes due to the formation of secondary oxidation products. According to European Commission (EC/2568/91) and International Olive Oil Council (IOOC/2001) regulations, the extra virgin olive oil oxidation level is assessed by the PV and spectrophotometric absorbance, defining the following limits: PV ⱕ20 meq kg⫺1 and K270 ⱕ 0.22 and ΔK ⱕ 0.01. The EC legislation also considers the value of K232 that must be ⱕ2.4. Nevertheless these methods supply only limited information about the level of olive oil oxidation. In recent years, HPLC/GC-MS was applied to detect changes in the chemical composition of olive oil during the storage. HPLC with different detection systems has been used for hydroperoxide analysis (Oshima et al., 1996). GC/MS was used to detect hydroxy fatty acids and volatile compounds originated from hydroperoxide degradation (Morales et al., 1997) and to identify the products of triglyceride oxidation (Rovellini et al., 1998). However, these last techniques are complex, expensive and time-consuming and generally highlight only one or few aspects of the oxidation process, giving only partial information about the extent of the process. On the other hand, the olive oil industry needs to be able to know quickly the level of oil oxidation in order to predict its remaining shelf-life. Moreover, consumers expect manufacturers and retailers to provide products of high quality
555
CHAPTER | 60 Application of the Electronic Nose in Olive Oil Analyses
component was able to separate the oil samples belonging to class 3 (characterized by positive values) from all other samples, i.e. the first component was able to characterize the samples on the basis of the storage period. In fact, samples belonging to class 3 were characterized by a storage period of 2 years, while all the other samples by a storage period of only 1 year. On the second principal component, oil samples of class 2 had positive values, while all other samples had negative values, i.e. the second principal component was able to describe the samples on the basis of the storage conditions. In fact, class 2 samples were stored under light, with all other samples under dark. Finally, a sample belonging to class 1 appeared far from its class space in the score plot. This sample was characterized by the lower scores on the first and the second component. As described before, the meaning of each component is related to the quality of the storage period and conditions. The more negative scores on the two components were associated to the best storage situation, i.e. conservation under dark for 1 year. The behavior of sample no. 10 confirmed this hypothesis: in fact, all values of the classical chemical parameters for this sample respected the law limits and allowed it to be considered as an extra virgin olive oil. All other samples of class 1 could not be considered as extra virgin olive oils since they had just UV values or PV values higher than the law limit. Since the data structure analysis gave a good sample characterization, a classification model was built. LDA analysis was applied in order find a predictive classification model, able to separate the three described classes. LDA applied to the complete data set gave a recognition percentage of 100% for the three classes (error rate 0%), while during the leave one out cross validation, only one oil sample (class 3) was not correctly classified (cross validation error rate 3.4%). Even if this model performed a
3
2
1 PC2
and look for quality seals and brands. Therefore the development of innovative analytical tools able to execute quick and reliable quality checks on extra virgin olive oil is required. This study reports the potential of electronic nose, in combination with multivariate statistical analysis for evaluating the oxidation level, i.e. oil quality at bottling time, storage in real life conditions and not applying an accelerated thermo-oxidation process. For this study 61 virgin olive oils from typical cultivars of Garda region were packaged in glass bottles and stored in the light for 1 year and in the dark for 1 or 2 years, 1 year being generally considered the maximum storage period from bottling to consumption. This approach could represent a faster recognition tool for monitoring olive oil oxidation since it is characterized by simplicity of sample preparation. The oils were analyzed, before and after storage, using both chemical methods and electronic nose technique. In the literature, there are several examples that demonstrate the possibility of using an electronic nose for the characterization of vegetable oils (Martin et al., 1999; Gan et al., 2005) while information about the use of an electronic nose to predict shelf-life of vegetable oils or to monitoring oil oxidation under real life storage conditions is very limited (Shen et al., 2001; Cosio et al., 2007). The 61 extra virgin olive oils analyzed before the storage presented the acidity ranged from 0.1 to 0.3, the PV from 3 to 6, K232 from 0.8 to 1.4, K270 from 0.08 to 0.15, and ΔK was always lower than 0.01. Then the samples were analyzed after 1 year of storage under dark (class 1), under light (class 2), and after 2 years under dark (class 3). All samples of the three classes presented an acidity value lower than 0.4, PV from 16 to 22 (class 1), from 17 to 61 (class 2) and from 17 to 39 (class 3). Most samples had K232, K270 and ΔK above the low limits. At the end of their storage period, all virgin olive oils were also analyzed by using the electronic nose. The response of the electronic nose is characteristic of each sensor and depends on the concentration and the profile of the volatile compounds present in the olive oil. The signal obtained with the electronic nose (22 sensors) together with the classical chemical determinations (five parameters) calculated at the end of the sample storage period were considered all together and analyzed by means of PCA. The first principal component and the second principal component were enough to display the data structure, since they explained 61% of the total variance. Examining the score plot (Figure 60.2) in the area defined by the first two principal components, a separation of the samples into three groups was found according to the different storage conditions and storage periods. Only few samples belonging to class 3 were projected in the middle of class 1, but this does not affect the effectiveness of the plot. Furthermore, on the basis of the position of each group in the plot, it was possible to assign a particular meaning to each component. The first
0
−1
−2
−3
−2
−1
0 PC1
1
2
FIGURE 60.2 PCA on autoscaled data: score plot. Classes are shown with different symbols: # class 1, (class 2, ⫹ class 3).
3
556
SECTION | I Natural Components
TABLE 60.1 Confusion matrix of the LDA classification model with the electronic nose variables (fitting and validation results are both reported). Classes Fitting
Cross-validation
1
2
3
Total
1
16 (100%)
0 (0%)
0 (%)
16
2
0 (0%)
16 (100%)
0 (%)
16
3
0 (0%)
0 (%)
29 (100%)
29
1
16 (100%)
0 (%)
0 (%)
16
2
0 (%)
16 (100%)
0 (%)
16
3
0 (%)
1 (3.4%)
28 (96.6%)
29
Rows represent the true class, columns represent the assigned class. Percentages of correct classified samples appear in brackets.
Since an equal classification performance was obtained by considering only the electronic nose sensors, it is evident that chemical analyses were not required in order to achieve a better sample discrimination. The final classification model built by means of the electronic nose sensors was able to describe the sample storage conditions and could represent a simpler, faster, and cheaper recognition tool, since a minor number of variables must be determined.
6 4
Function 2
2 0 −2
60.3 EVALUATION OF GEOGRAPHICAL ORIGIN OF EXTRA VIRGIN OLIVE OILS
−4 −6 −10
0 Function 1
10
FIGURE 60.3 LDA classification model with the electronic nose sensors: discriminant scores. Classes are shown with different symbols: # class 1, (class 2, ⫹ class 3).
good classification result, in order to simplify once more the classification model, LDA was repeated by considering only the electronic nose data. In Table 60.1, the results of LDA and leave one out cross validation are reported. As expected, the classification model gave the same result as before; in particular a recognition percentage of 100%, and only one wrong assignation in the validation procedure was obtained. Figure 60.3 shows how the first two functions discriminated among classes. Function 1 (83.2% of the total variance) was able to discriminate between the oils stored under dark for 2 years (class 3) from the samples stored under dark for 1 year (class 1) and under light (class 2). Function 2 (16.2% of the total variance) was able to discriminate the oils stored under dark for 1 and 2 years (class 1 and 3) from the samples stored under light for 1 year (class 2).
The quality and uniqueness of specific extra virgin olive oils is the result of different factors such as cultivar, environment and cultural practices. Moreover, an important act of legislation (EC 2081/92) allows the PDO labeling of some European EVOOs with the names of the areas where they are produced. This designation guarantees that the quality of the product is closely linked to its geographical origin. PDO olive oils are considered the best among EVOOs on the basis of their authenticity and specified organoleptic characteristics. As a consequence, PDO olive oils have a much higher market price and therefore are subjected to frauds: the addition of refined oils and/or the marketing of oils from one region as those from another. Consumers are also more and more oriented towards purchasing food products of a certified genuineness and geographical origin. In the present study, EVOO Garda ‘Bresciano’ has been considered: this is a product made in the Garda Lake area, a circumscribed area in the north Italian region of Lombardia and distinguished as a PDO since 1997. Detailed percentages of specified cultivar olives, cultural practices, circumscribed geographical production areas, chemical and sensorial properties are required in order to obtain the PDO
557
CHAPTER | 60 Application of the Electronic Nose in Olive Oil Analyses
the law limits and consequently can be considered EVOOs. Total phenols have also been determined for all the samples, in order to verify a possible correlation with their geographical origin. For the same reason, the oil samples have been analyzed by means of electronic nose, which could be able to detect the presence of volatile compounds in olive oils. All data collected from the electronic nose were compared and elaborated together with the chemical parameters. As a first step, principal component analysis (PCA) was carried out using the complete data set. Then classification techniques (LDA) were applied. PCA was performed on the autoscaled data in order to provide partial visualization of the data set in a reduced dimension. The two principal components represent 79% of the total variance. Examining the score plot (Figure 60.4) in the area defined by the first two principal components a clear separation of the samples into five groups was found according to the region of origin. In order to characterize EVOO samples into the five mentioned classes a supervised pattern-recognition method was applied. LDA applied to the complete data set gave a recognition percentage of 100% for all EVOOs (error rate 0%), while during the leave one out cross validation, some samples were not correctly classified (cross validation error rate 7.55%). Finally the classification model was applied to only set of electronic nose data, i.e. the 19 commercial and multivarietal EVOOs. Figure 60.5 shows the predictive ability of LDA model. It can be seen that all the samples that came from Garda (commercial) were correctly classified while the multivarietal EVOOs were distributed among the other classes due probably to their characteristic of being a mixture of oils from different geographical areas.
2.5 Origin Spagia 2.0 Sardegia Garda
1.5
Campania 1.0
PC2
label, as indicated in the Production Disciplinary. However, at present no analytical parameters exist that enable the Garda PDO oil to be distinguished from similar products of other regions. The development of precise methods for the classification of oils is becoming very important for the assignment of a ‘denomination of origin’ trademark. Since official analysis of virgin olive oils involves a series of several determinations of chemical and physical constant that will be of little use in the geographical certification of the oil samples, reliable methods of authentication of oil geographical origin are essential. A variety of analytical methodologies have been proposed for the authentication of vegetables oils, including gas chromatographic analysis (Webster et al., 1999; Cert et al., 2000) nuclear magnetic resonance spectroscopy (Sacco et al., 2000; Rezzi et al., 2005), mass spectrometry (Caruso et al., 2000). These techniques usually require timeconsuming measurements, sample preparation and a qualified staff. The necessity of quick and simple methods has addressed the present study to the use of an electronic nose to characterize the origin of PDO Garda extra virgin olive oil. Moreover, the main compounds, mainly carbonyl compounds, alcohols, esters and hydrocarbons, were found in the volatile fraction of virgin olive oil (Flath et al., 1973). These volatile compounds, stimulating the olfactive receptors in the human nose, are responsible for the whole aroma of the virgin olive oil; similarly in the electronic nose a variety of sensors interact differently with the odors of sample. Volatile components of olive oil are considered as a key for quality and authentication control, consequently they are of great interest. In the literature, there are several examples that demonstrate the possibility of using an electronic nose for the quality control of olive oil aroma (Guadarrama et al., 2000). The combination of electronic nose fingerprinting with multivariate analysis also provides an original approach to study the profile of olive oil in relation to its geographical origin (Ballabio et al., 2006; Cosio et al., 2006). In the present study, EVOOs have been studied by means of an electronic nose and by classical chemical parameters. The data set has included 53 samples of monovarietal EVOOs obtained from several olive cultivars and grown in five different regions: Garda, 36 samples; Spagna, 6 samples; Sardegna, 5 samples; Campania, 4 samples; Abruzzo, 2 samples. The sampling has also included 19 commercial and multi varietal EVOOs: 3 samples labeled as Garda PDO, produced with cultivars allowed by the Garda Production Disciplinary; 3 samples of Garda, not labeled as PDO; 13 samples collected on the market, produced with unknown cultivars. All these commercial samples have been used only to test the classification model. The quality of extra virgin olive oil is determined by analytical parameters: free acidity, peroxide value, K232, K270, and ΔK, according to legislation (EEC/2568/91; IOOC/2001). All of the analyzed samples have respected
Abruzzo
5
0.0
−5
−1.0
−1.5 −2
−1
0
1
2
PC1
FIGURE 60.4 PCA on autoscaled data. EVOO samples are shown with different symbols according to the origin.
3
558
SECTION | I Natural Components
10
●
●
Function 2
0 ●
−10
●
Class Group centroids Ungrouped cases Spagia Sardegia
EVOO sensorial properties decrease during storage because of oxidation processes affected by air, heat, light and metals and the volatile compounds change in off-flavors. The development of innovative analytical tools able to execute fast and reliable quality checks on extra virgin olive oil is required. The electronic nose responds to flavor/odor using an array of simple and non-specific sensors and a pattern recognition software system. The electronic nose approach could represent a faster recognition tool for monitoring olive oil oxidation and uniqueness of specific extra virgin olive oil since it is characterized by simplicity of sample preparation.
Campania Garda Abruzzo
−20 −20
−10
0 10 Function 1
20
30
FIGURE 60.5 Projection of monovarietal EVOO samples of different origin, commercial and multivarietal EVOO samples (ungrouped cases) predicted by the LDA model with electronic nose variables. Classes are shown with different symbols.
This study clearly shows that it is possible to differentiate and classify EVOOs from different geographical areas by using a commercial electronic nose and by applying multidimensional chemometric techniques. In conclusion there is a growing emphasis and consensus that intelligent sensor arrays or electronic noses are most effective in the quality control of raw and manufactured products, for example for determination of food freshness and maturity monitoring, shelf-life investigations, authenticity assessments of products and even microbial pathogen detection and environmental control. This application area is particularly important because the e-nose can be trained to recognize hazardous chemicals as well as odors. Furthermore, with respect to the human nose the e-nose does not fatigue as easily, is less costly and can travel easily into outer space. It also holds the promise of being much cheaper, smaller and easier to use and maintain than a mass spectrometer.
SUMMARY POINTS ●
●
Extra virgin olive oil (EVOO) is processed from fruit of the olive (Olea europea L.) of good quality and it presents a complex flavor, which is greatly liked by native consumers and internationally appreciated by gourmets. The quality and uniqueness of specific extra virgin olive oils is the result of different factors such as cultivar, environment and cultural practices.
REFERENCES Angerosa, F., Basti, C., Vito, R., 1999. Virgin olive oil volatile compounds from lipoxygenase pathway and characterisation of some Italian Cultivars. J. Agric. Food Chem. 47, 836–839. Ballabio, D., Cosio, M.S., Mannino, S., Todeschini, R., 2006. A chemometric approach based on a nouvelsimilarity/diversità measure for the characterisation and selection of electronic nose sensors. Anal. Chim. Acta. 578, 170–177. Buratti, S., Benedetti, S., Cosio, M.S., 2005. An electronic nose to evaluate olive oil oxidation during storage. Ital. J. Food Sci. 2, 203–210. Caruso, D., Colombo, R., Patelli, R., Giavarini, F., Galli, G., 2000. Rapid evaluation of phenolic component profile and analysis of oleuropein aglycon in olive oil by atmospheric pressure chemical ionization mass spectrometry (ACPI-MS). J. Agric. Food Chem. 48, 1182–1185. Cert, A., Moreda, W., Perez-Camino, M.C., 2000. Chromatographic analysis of minor constituent in vegetables oils. J. Chromatogr. A 881, 131–148. Cosio, M.S., Ballabio, D., Benedetti, S., Gigliotti, C., 2006. Geographical origin and autentication of extra virgin olive oils by an electronic nose in combination with artificial neural networks. Anal. Chim. Acta. 567, 202–210. Cosio, M.S., Ballabio, D., Benedetti, S., Gigliotti, C., 2007. Evaluation of different storage conditions of extra virgin olive oils with an innovative recognition tool built by means of electronic nose and electronic tongue. Food Chem. 101, 485–491. E.C. Off. J. Eur. Communities, 1991. July 11, Regulation 2568/1991. Characteristics of olive and olive-pomace oils on their analytical methods. E.C. Off. J. Eur. Communities, 1992. July 11, Regulation 2081/1992. Flath, R.A., Forrey, R.R., Guadagni, D.G., 1973. Aroma components of olive oil. J. Agric. Food Chem. 21, 948–952. Gan, H.L., Che Man, Y.B., Tan, C.P., NorAini, I., Nazimah, S.A.H., 2005. Characterisation of vegetables oils by surface acoustic wave sensing electronic nose. Food Chem. 89, 507–518. Gardner, J.W., Bartlett, P.N., 1993. Brief history of electronic nose. Sens. Actuator B. 18, 211–220. Guadarrama, A., Mendz, M.L.R., Saia, J.A., Ros, J.L., Olas, J.M., 2000. Array of sensors based on conducting polymers for the quality control of the aroma of the virgin olive oil. Sens. Actuators B. 69, 276–282. IOOC (International Olive oil Council), 2001. T.15 n. 2/REV. 10/2001.
CHAPTER | 60 Application of the Electronic Nose in Olive Oil Analyses
Kiritsakis, A.K., Min, D.B., 1989. Flavour chemistry of olive oil. In: Min, D.B., Smouse, T.H. (eds) Flavor Chemistry of Lipid Foods. American Oil Chemists’ Society, Champaign, IL, pp. 196–221. Mannino, S., Benedetti, S., Buratti, S., Cosio, M.S., Scampicchio, M., 2007. Electrochemical sensors for food authentication. In: Alegret, S., Merkoci, A. (eds) Comprehensive Analytical Chemistry, Electrochemi-cal Sensor Analysis, Vol. 49. Elsevier Inc., San Diego, pp. 757–773. Martin, Y.G., Pavon, J.L.P., Cardero, B.M., Pinto, C.G., 1999. Classification of vegetable oils by linear discriminant analysis of electronic nose data. Anal. Chim. Acta. 384, 83–94. Morales, M.T., Rios, J.J., Aparicio, R., 1997. Changes in the volatile composition of virgin olive oil during oxidation: flavour and off-flavours. J. Agric. Food Chem. 45, 2666–2671. Oshima, T., Hopia, A., German, B., Frankel, E.N., 1996. Determination of hydroperoxides and structures by HPLC with post-column detection with diphenyl-1- pyrenylphosphina. Lipids 31, 1091–1098. Rezzi, S., Axelson, D.E., Heberger, K., Reniero, F., Mariani, C., Guillou, C., 2005. Classification of olive oils using high throughput flow 1H NMR
559
fingerprinting with principal components analysis, linear discriminant analysis and probabilistic neural networks. Anal. Chim. Acta. 552, 13–24. Rovellini, P., Cortesi, N., Fedeli, E., 1998. Profile oxidative and chemical structure of oxidation products of triglycerides by means HPLC-ES-MS. La Rivista Italiana delle Sostanze Grasse 76, 57–63. Sacco, A., Brescia, M.A., Liuzzi, V., Reniero, F., Guillou, C., Ghelli, S., 2000. Characterisation of Italian olive oils based on analytical and nuclear magnetic resonance determination. J. Am. Oil Chem. Soc. 77, 619–625. Schaller, E., Bosset, J.O., Escher, F., 1998. “Electronic noses” and their applications to food. Lebensmittel Technologies 31, 305–308. Shen, N., Moizuddin, S., Wilson, L., Duvich, P., White, P., Pollak, L., 2001. Relationship of electronic nose analyses and sensory evaluation of vegetable oils during storage. J. Am. Oil Chem. Soc. 78, 937–940. Webster, L., Simpson, P., Shanks, A.M., Moffat, C.F., 1999. The authentication of olive oil on the basis of hydrocarbon concentration and composition. Analyst 125, 97–104.
This page intentionally left blank
Chapter 61
Squalene and Tocopherols in Olive Oil: Importance and Methods of Analysis Maria Z. Tsimidou Laboratory of Food Chemistry and Technology, School of Chemistry, Aristotle University of Thessaloniki, Greece
61.1 INTRODUCTION Virgin olive oil, the juice of ripe olive fruit is endowed with a variety of minor compounds belonging to different classes of organic compounds, some of which are valuable functional ingredients. Among the latter, squalene and tocopherols are also considered important contributors to oil oxidative stability in the dark (autoxidation), under light exposure (photo-oxidation) or upon heating. Analysis of these compounds is not required by official organizations yet. Nevertheless, in a number of publications, determination of the above compounds in olive oil is reported using mainly chromatographic procedures. Recommended procedures can also be found in compilations of international scientific organizations [Codex Alimentarius, American Oil Chemists Society (AOCS), International Union of Pure and Applied Chemistry (IUPAC), Association of Official Analytical Chemists, International (AOAC, Int.), International Standardization Organzation (ISO), etc.]. Some of the analytical procedures developed, demand sample pretreatment, others do not. Since squalene and tocopherols confer to the nutritional and technological properties of virgin olive oil and products that contain it, a concise presentation of such aspects was considered important, especially for the less appreciated – within the olive oil industry – squalene.
61.2 SQUALENE 61.2.1 Occurrence and Importance The important precursor in sterol biosynthetic routes, squalene (C30H50, 2,6,10,15,19,23-hexamethyl-2,6,10,14, 18,22-tetracosahexaene) (Figure 61.1) is a terpenoid hydrocarbon originally identified at the beginning of the 20th century by the Japanese chemist Mitsumaru Tsujimoto in certain deep sea shark-liver oils. Depending upon age and sex, squalene content may reach 80% of the liver oil Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
FIGURE 61.1 Chemical structure of squalene. Squalene is an aliphatic unsaturated hydrocarbon having 30 carbon atoms and six isolated double bonds.
of such species (Wetherbee and Nichols, 2000 and references therein). High levels of squalene are also found in human skin lipids (⬃500 μg g⫺1 dry weight) and adipose tissue (⬃300 μg g⫺1). The unique composition of skin lipids attracted the interest of researchers, who hypothesized a particular functionality of squalene and other unusual lipids, e.g., exclusion of pathogens due to metabolic problems that lead to the survival of only compatible microorganisms (Nikolaides, 1974). Daily squalene excretion by human skin is strong evidence for its de novo synthesis there. Squalene’s role as an effective quencher of skin lipid photo-peroxidation is well established. This may explain the early commercial interest in squalene and its saturated counterpart, squalane, and the wide array of applications in cosmetics (e.g., sunscreens, moisturizing creams) and dyes (Eyres et al., 2002). On the other hand, dietary benefits through squalene consumption are not fully elucidated though it is regarded as offering some protection against certain types of cancer (Rao et al., 1998; Smith, 2000). Dietary squalene intake is reflected to triglyceride content increase more regularly than to plasma cholesterol levels (Ostlund et al., 2002). Non-conclusive is the evidence for the activity of the hydrocarbon as a radical scavenger (Psomiadou and Tsimidou, 1999). The excessive use of available marine sources led to a severe reduction in the population of certain sharks in the ocean so that alternative natural sources were sought next in the plant kingdom. The interest in plant sources is, thus, more of ecological importance than of comparable commercial potential, taking into account that world demand exceeds 1000 tonnes per year. Squalene content is related to
561
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
562
SECTION | I Natural Components
TABLE 61.1 Effect of squalene on oxidative stability of purified olive oil triacylglycerol fraction stored at 62 °C in the dark for 70 h (abstracted from Psomiadou and Tsimidou, 1999): a characteristic example of squalene antioxidant effectiveness is given. For a specific storage period (70 h, 60 °C), the presence of squalene at realistic level of addition protected oil from oxidation. Absorbance values at 232 nm (expressed as K232) Purified olive oil triacylglycerol fraction (Control)
Control ⫹ 7000 ppm squalene
Control ⫹ 7000 ppm squalene ⫹ 100 ppm αtocopherol ⫹ 10 ppm caffeic acid
4.72
2.93
2.18
the lipid content of the plant material. This finding does not imply that all plant fats are rich in squalene. There is diversity in levels (0–12 g kg⫺1). Virgin olive oil prevails. Other fruit-derived oils (palm oil, avocado oil) contain much lower levels. Among seed fats, corn oil is far richer than soybean, rapeseed or sunflower oils. Moreover, squalene is recovered from the distillates of the olive oil industry (10–30% yield, Bondioli et al., 1993), from Amaranthus seed oil (2–8% w/w, Jahaniaval et al., 2000) and other plant origin materials. Potential chemopreventive activity of squalene, though not conclusive, dressed this unsaturated hydrocarbon with the veil of a functional compound and attracted interest in new methods of its production (Bhattacharjee et al., 2001). Therefore, its determination in various matrices gained interest. Control of genuineness and origin of commercial squalene is also an important issue.
61.2.2 Squalene in Olive Oil Reports on the presence of squalene in virgin olive oil date back more than 50 years. A hypothesis on the relationship of squalene and cancer-risk-reducing effect of olive oil, suggested some years ago, had an impact on the renewed interest in this compound (Newmark, 1997). The hydrocarbon fraction of the oil is made almost exclusively of squalene (Eisner et al., 1965; Bastic´ et al., 1978; Lanzón et al., 1994). Ranges vary among reports (overall range 200–12 000 mg kg⫺1 oil, typical level ⬃5000 mg kg⫺1). The levels of squalene in virgin olive oil are related to cultivar characteristics (Guinda et al., 1996; De Leonardis et al., 1997; Manzi et al., 1998), whereas extraction technology and refining process cause a considerable reduction in them (Mariani et al., 1992; Lanzón et al., 1994). Bleaching gives rise to the formation of isomers ⫺ 3% of C30H50 content can be isomerized (Grob et al., 1992) – the determination of which was examined as a criterion for the addition of bleached olive oil to virgin olive oil (Amelio et al., 1998).
The contribution of squalene to olive oil stability under light exposure or in the dark was only recently investigated and the results are not conclusive (Albi et al., 1998; Manzi et al., 1998; Psomiadou and Tsimidou, 1999, 2002a, b). In the dark, phenolic antioxidants enhance effectiveness of squalene as shown in Table 61.1. Loss upon heating seems to be related to conditions and the presence of food that is cooked (Boskou and Morton, 1975; Kalogeropoulos and Andrikopoulos, 2004).
61.2.3 Methods of Determination The interest in the determination of this hydrocarbon in various matrices covers authentication issues, stability studies, dietary and clinical studies and applications to food, feed, cosmetics and pharmaceutical industrial-quality specifications. Literature presents variability in approaches closely related to the aim of study and sample preparation treatments. As we stressed a few years ago (Nenadis and Tsimidou, 2002), most of the existing methodologies have been developed for the determination of other lipids (fatty acid methyl esters, waxes, alcohols) so that the optimization process rarely had squalene as the target analyte. Gas or liquid procedures are the most important among analytical protocols available. In their review on hydrocarbons in edible vegetable oil, Moreda and coworkers (2001) presented characteristic chromatographic procedures published till 1995. In the present chapter, we focus on procedures designed and optimized for identification and quantitative determination of squalene in olive oil with emphasis on sample pretreatment.
61.2.3.1 Gas Chromatographic Procedures Saponification of the lipid matrix has been adopted in most of the gas chromatographic procedures (Eisner et al., 1965; Bastic´ et al., 1978; Lanzón et al., 1994) either according to IUPAC standard protocol or using milder conditions. Fractionation of unsaponifiables is preferred in
CHAPTER | 61 Squalene and Tocopherols in Olive Oil: Importance and Methods of Analysis
563
TABLE 61.2 Gas chromatographic procedures for quantitative determination of squalene in olive oil. Details for characteristic gas chromatographic conditions together with the groups of co-eluted compounds to facilitate choice of the reader. Column characteristics
Other elution conditions/coeluted compounds
De Leonardis et al. (1997) Supelcowax 10 30 m ⫻ 0.32 mm df: 0.50 μm
Oven: 235 °C; injector/detector: 250 °C; split system: 1 μL (ratio 35:1), He:1.7 mL min⫺1; mass spectra at electron impact mode: 70 eV; external standard: squalene in peanut oil/fatty acid methyl esters
Ackman et al. (2000) Omegawax-320, 30 m ⫻ 0.25 mm df: 0. 32 μm
Oven: 200 °C/hold 5 min/3 °C min⫺1 to 230 °C/hold 15 min; He/ hydrogenated fatty acid methyl esters and other hydrocarbons
Giacometti (2001) SPB-5 30 m ⫻ 0.53 mm df: 0. 32 μm 0
Oven:180–270 °C (0.8 °C min⫺1/hold 65 min); injector/detector 290/300 °C; He (100:1)/silylated aliphatic alcohols, α-tocopherol, sterols
certain studies before silylation. The former is carried out on Florisil columns (Eisner et al., 1965), silica columns (Guinda et al., 1996) or silica gel thin-layer plates (Basti et al., 1978). Cold alkaline methylation of the oil matrix, extraction with hexane, washing up of the hexane phase with ethanol:water mixtures and direct injection onto the chromatograph (Lanzón et al., 1995) is reported to have a good linearity and very good repeatability (⬍2%) for a wide range of squalene levels (500–19 000 mg kg⫺1). Quantification was based on the use of squalane as an internal standard. To a similar direction is the process reported by De Leonardis et al. (1997) using an external standard curve. Transesterification by boron trifluoride should be avoided in squalene analysis as pointed out by the wellknown lipid scientist (Ackman et al., 2000). The latter added a hydrogenation step – first reported by Lanzón and collaborators (1992) – to the sample preparation procedure to minimize loss due to oxidation. Briefly, 10–30 mg fatty acid methyl esters dissolved in CHCl3 (30 mL) are transferred into a 100 mL round-bottomed flask. After the addition of the catalyst (PtO2), a slow hydrogen flow is blown into the mixture, under magnetic stirring for 1 h and the esters are carefully removed with a pipette. Isolation of olefinic compounds on silica gel short HPLC column (10 cm ⫻ 5 mm i.d.) and subsequent on-line gas chromatographic examination was also developed in one laboratory (Grob et al., 1992). Villén et al. (1998) coupled on-line reversed-phase liquid chromatography and a gas capillary one using an especially designed programmed temperature vaporizer as an interface to develop a screening quantitative method for squalene determination suitable for routine quality control. The method allowed the simultaneous
analysis of free sterols and tocopherols, too. The on-line overall procedure required approximately one hour. The detection limit for squalene was 0.3 ppm and precision was 6% (n ⫽ 3). Nevertheless, such procedures are difficult to adopt in a wide range of analytical laboratories. Considering the changes in column properties over the years, only three recent characteristic gas chromatographic procedures on capillary columns are presented in Table 61.2. In two of them squalene is determined together with fatty acid methyl esters on polar columns whereas the third one allows codetermination of important minor compounds on a non-polar column.
61.2.3.2 Liquid Chromatographic Procedures Cortesi and coworkers (1996), presenting the merits of HPLC in the authenticity and quality control of edible oils, pointed out that on the reversed-phase chromatogram of triacylglycerols the presence of squalene among the peaks eluted prior to trilinolein. Elution was accomplished according to the official European Union method using acetone:acetonitrile (1:1, v/v) and refractive index detection. No further data were given by the authors on method validation. A normal-phase HPLC procedure using a rather complex gradient for silica columns using n-hexane:diethyl ether mixtures was also proposed by an industrial laboratory (Amelio et al., 1998) for quick screening of legal limits for steroid hydrocarbons in great numbers of samples. The proposed scheme included fractionation after LC and HPGC separation of stigmasta-3,5-diene, squalene isomers and wax esters. Tsimidou and coworkers (Nenadis and Tsimidou, 2002; Grigoriadou et al., 2007) introduced
564
fractional crystallization as a rapid, low-cost alternative in sample preparation prior to RP-HPLC determination of squalene. This approach can replace saponification in routine quality control laboratories. One of the advantages is that the pretreatment of a great number of samples is accomplished overnight so that working hours are devoted to actual analysis. Briefly, oil (0.1–0.5 g) is vortexed with 20 mL of methanol:acetone (7:3, v/v) in a 25 mL groundglass stoppered test tube for 1.5 min and then stored at ⫺22 °C overnight. Filtration through a coarse filter is accomplished rapidly, the solvent is removed under reduced pressure and the residue of the liquid fraction dissolved in the mobile phase is injected onto the liquid chromatograph. Due to the fact that the most saturated triacylglycerol species are eliminated from the analyte, reversed HPLC is shorter, especially when acetonitrile content increases ⫺ 60% acetonitrile instead of 40–50%. The use of both refractive index and UV detectors (at 208 nm) is feasible. These detectors are found in most lipid laboratories so that no extra capital cost is needed. Detection limits were 50 and 23 mg kg⫺1 oil, respectively. These limits are acceptable considering that the actual amounts of squalene in olive oil are some hundreds or thousands mg kg⫺1. Precision data were satisfactory (CV% ⫽ 3.76, n ⫽ 7). Recovery for two levels of addition (7000 and 700 mg kg⫺1) were 92.5 ⫾ 6.7 and 81.5 ⫾ 4.0, respectively. To our knowledge this was the first report on a liquid chromatographic method dedicated to squalene that provided validation data. The effort continued a few years later. Isolation of squalene from a silica cartridge with 10 mL n-hexane allowed chromatography on a similar column only with the aid of acetonitrile (Grigoriadou et al., 2007). The latter led to a three times improvement of detection limit (6.2 ng 10 μL⫺1) and limit of quantification (7.8 ng 10 μL⫺1). This was better than that reported by Lu et al. (2004). Recovery at two levels of addition (700 and 4000 mg kg⫺1) were 88 ⫾ 9% and 85 ⫾ 4%. The above-proposed procedures were comparatively as effective as when saponification (Manzi et al., 1998) was used for sample pretreatment and the injected material was the unsaponifiable fraction. Key facts for squalene are briefly presented in Table 61.3.
61.3 TOCOPHEROLS 61.3.1 Occurrence and Importance Natural occurrence, bioavailability, biopotency and technological importance of the eight compounds responsible for vitamin E activity, the E vitamers, have been extensively studied over many decades. α-Tocopherol II (Figure 61.2) is of particular interest for human and animal welfare in a multifunctional manner (Zingg and Azzi, 2004). The fact that it is a less active antioxidant in vitro than its regular counterparts in vegetable oils (β-, γ-, and δ-tocopherols)
SECTION | I Natural Components
TABLE 61.3 Key facts of squalene. 1. 2. 3. 4.
Squalene is a hydrocarbon usually found in marine fats Virgin olive oil is the only plant edible source of squalene Levels of squalene: 500–12 000 mg kg⫺1 oil Potential chemopreventive activity of squalene, though non-conclusive, attracted the interest in new methods of its production and determination 5. Gas liquid chromatography prevails as a method of quantitative determination of squalene in olive oil 6. Liquid chromatography has some advantages, especially if combined with solid phase extraction for sample pretreatment
CH3 HO CH3 H3C
CH3
CH3
O
CH3 CH3
CH3 FIGURE 61.2 Chemical structure of α-tocopherol. Eight compounds responsible for vitamin E activity are known, the E vitamers. Four of them bear a saturated carbon chain to the chroman ring and the other four an unsaturated carbon chain of the same size.
undermined its natural presence in virgin olive oil for many years. This view is now dramatically changed and α-tocopherol, the dominant form in olive oil, is determined in most recent studies of the field (e.g. Psomiadou and Tsimidou, 2000a, b). Vegetable oils provide humans with a significant part of their daily vitamin E dietary requirements, which increase when the diet is rich in unsaturated fatty acids. Virgin olive oil seems to present an optimum ratio of α-tocopherol/unsaturated fatty acids content.
61.3.2 Tocopherols in Olive Oil Research work concerning the occurrence and levels of αtocopherol in various sets of virgin olive oils from all over the world has augmented in the last two decades. Data indicate ⬎250 mg α-tocopherol per kg of high-quality virgin olive oils analyzed just after production and even unusually higher levels (⬎350 mg kg⫺1) in certain monovarietal products derived from healthy olives from different cultivars or regions under laboratory extraction conditions. Lower levels are expected in commercial samples or during domestic use (Psomiadou et al., 2000). The three other vitamers, β-, γand δ-tocopherols, are present in insignificant amounts (less than 5% of the tocopherol fraction). Therefore, the interest in tocopherol analysis in olive oil is focused on the determination of the α-vitamer unless authentication is the objective (Dionisi et al., 1995).
CHAPTER | 61 Squalene and Tocopherols in Olive Oil: Importance and Methods of Analysis
61.3.3 Methods of Determination Literature on the determination of all eight vitamers simultaneously or selectively with regards to matrix is abundant. Published procedures in a variety of analytical, food or pharmaceutical journals ‘compete’ with each other in sophistication and efficiency. Among them, tocopherol separation and quantitative determination in oils and fats has been adequately carried out using either normal-phase or reversed-phase HPLC for more than 20 years. The versatility in modes and detection systems (diode array, fluorescence, electrochemical) as well as the less demanding sample preparation steps conferred clear advantages over the time-consuming GC procedures (Rupérez et al., 2001). Normal-phase applications prevail in food analysis mainly because all eight vitamers are separated (Kamal-Eldin et al., 2000), whereas problems with repeatability in quantification have been thoroughly examined over the years (e.g. Hewavitharana, 2003). Within the frame of the book requirements and limitations we chose to present those methods that were developed for the simultaneous determination of α-tocopherol (and/or other tocopherols) and other bioactive endogenous ingredients of olive oil, namely carotenoids, chlorophylls, squalene, sterols, and phenols. Such methods are considered advantageous in official and routine quality control as well as in stability studies. We present next the most characteristic ones in chronological order. Psomiadou and Tsimidou (1998) developed a 20 min normal-phase protocol to determine simultaneously the four tocopherols (detection at 294 nm or fluorescence at 295 nm exc./330 nm emm.), β-carotene, lutein and chlorophyll b (453 nm), pheophytin a and derivatives (410 nm), pheophytin b and chlorophyll a (430 nm) using n-hexane:2-propanol (99:1, v/v) (solvent A) and 2-propanol (solvent B) in gradient mode (0%B for 10 min; 0–5%B in 4 min; 5%B for 6 min; 5–10%B in 4 min and 0% B for 6 min). Normal phase was selected to avoid sample pretreatment (e.g., saponification). The authors pointed out the importance of parallel determination of prooxidants and antioxidants. They presented all necessary validation data, which allowed them to determine levels of α-tocopherol in Greek virgin olive (Psomiadou et al., 2000) and monitor changes in the levels and composition of these analytes upon storage in the dark, light exposure or heating (Marinova et al., 2001; Psomiadou and Tsimidou, 2002a, b). A less advantageous procedure was proposed by Gimeno and collaborators (2002) for the parallel determination of α-tocopherol and β-carotene. Parcerisa and collaborators (2000) determined in the unsaponifiable fraction in parallel to tocopherols, β-sitosterol, Δ5-avenasterol, campesterol, stigmasterol and obtusifoliol using gas chromatography. Separation was on a 5% crosslinked phenylmethyl silicone fused silica column (30 m ⫻ 0.25 mm, df:0.25 μm) under gradient conditions (oven: 210–250 °C at a rate of 6 °C min⫺1; held
565
at 250 °C for 11 min, 250–310 °C at a rate of 3 °C min⫺1, 313 °C for 12 min). Eluted trimethylsilyl ether derivatives were detected by mass spectrometry at the electron impact mode. The procedure was developed for detection of the presence of hazelnut oil in olive oil. Tocopherols (295 nm or fluorescenece) and β-sitosterol (208 nm) were also separated (López Ortíz et al., 2006) on a reversed-phase narrow-bore Luna column (3 μm, 150 ⫻ 2.00 mm) using acetinitrile:water (95:5, v/v) at a flow rate of 0.4 mL min⫺1. Saponification of the sample (0.2 g oil) with ethanolic potassium hydroxide solution 2 M (5 mL) was carried out in the presence of ascorbic acid solution 0.1 M (5 mL) after heating and stirring at 60 °C for 45 min. A lengthy extraction procedure after neutralization required the addition of a second antioxidant (BHT). The overall procedure is shorter than the official method for sterol analysis. It can be considered when quantification of both functional lipids is required in olive oil or in products containing olive oil. Co-extraction of α-tocopherol and polar phenols from virgin olive oil and subsequent separation and quantification on a reversed phase column (C18, 250 ⫻ 4 mm) was the aim of the study by Tasioula-Margari and Okogeri (2001). Recovery of phenols exceeded 98% for the majority of selected peaks whereas recovery for α-tocopherol was 80%. Oil (10 g) was extracted in sequence with absolute methanol (2 ⫻ 25 mL) and methanol:2-propanol, 80:20, v/v (2 ⫻ 2 5 mL). The combined extracts were brought to dryness under vacuum (40 °C) and then dissolved in methanol: 2-propanol:hexane mixture (1:3:1, v/v/v) and transferred on to the column. Elution solvents were acetic acid in water (A), methanol (B), acetonitrile (C) and 2-propanol (D). The gradient was rather long (70 min) and complex (95% A/5% B in 2 min; 60% A/10% B/30% C in 8 min; 25% B/75% C in 22 min and then for 15 min; 25% B/75% C in 2 min and finally 95% A/5% B in 3 min). UV-detection wavelength was the same for all determinants (280 nm). Giacometti (2001) focused on the parallel study of αtocopherol, sterols, aliphatic alcohols and squalene using gas chromatography of the unsaponifiable fraction after silylation. TLC fractionation was avoided and total run time was 75 min. Elution conditions are presented in Table 61.1. The author pointed out that, under the same conditions, determination of both sterols and dehydration products (stigmasta-3,5-diene) can be also accomplished. In all the above methods sample preparation procedures included simple dilution (normal phase LC), saponification, fractionation and silylation prior to GC, saponification and direct injection (reversed-phase LC), saponification and silylation prior to GC. Fractionation is normally carried out on TLC plates using systems described in official and recommended procedures. Grigoriadou et al. (2007), in an effort to extend the use of solid-phase extraction (SPE) in the official EU methods on the characteristics of olive oil, managed to elute in sequence and quantify separately squalene (1 ⫻ 10 mL n-hexane) and α-tocopherol
566
SECTION | I Natural Components
TABLE 61.4 Key facts of α-tocopherol. 1. α-Tocopherol is valued for its vitamin E activity and antioxidant properties 2. Virgin olive oil is an important source of α-tocopherol 3. High-quality virgin olive oils contain ⬎200 mg αtocopherol kg⫺1 oil 4. Virgin olive oil seems to present an optimum ratio of αtocopherol/unsaturated fatty acid content 5. There is a great number of gas and liquid chromatographic methods suitable for their analysis
(2 ⫻ 10 mL n-hexane:diethyl ether, 99:1, v/v). The residual lipid fraction was then ready for triacylglycerol analysis. Recovery for tocopherol fraction was 88 ⫾ 6% and 91 ⫾ 3% (for 70 and 225 mg kg⫺1 levels of addition, n ⫽ 7). The above work together with that by PérezCamino and coworkers (2002) for the determination of esters of fatty acids with low-molecular-weight alcohols in olive oil may be the basis for further investigation of the application of SPE to the field of simultaneous analysis of minor components of olive oil. Key facts for α-tocopherol are briefly presented in Table 61.4.
SUMMARY POINTS ●
●
●
Virgin olive oil is a rich source of bioactive squalene and α-tocopherol. Squalene and α-tocopherol can be determined in olive oil by different gas and liquid chromatographic methods, and also can be simultaneously separated and quantified. Both gas and liquid chromatography offer alternatives that can be adopted.
REFERENCES Ackman, R.G., Macpherson, E.J., Timmins, A., 2000. Squalene in oils determined as squalane by gas–liquid chromatography after hydrogenation of methyl esters. J. Am. Oil Chem. Soc. 77, 831–835. Albi, T., Lanzón, A., Guinda, A., Pérez-Camino, M.C., León, M., 1998. Microwave and conventional heating effects on some physical and chemical parameters of edible oils. J Agric. Food Chem. 45, 3000–3003. Amelio, M., Rizzo, R., Varazini, F., 1998. Separation of stigmasta-3,5diene, squalene isomers and wax esters from olive oils by single HPLC run. J. Am. Oil Chem. Soc. 75, 527–530. Bastic´, M., Bastic´, Lj., Jovanovic´, J.A., Spiteller, G., 1978. Hydrocarbons and other weakly polar unsaponifiables in some vegetable oils. J. Am. Oil Chem. Soc. 55, 886–891.
Bhattacharjee, P., Shukla, V.B., Singhal, R.S., Kulkarni, P.R., 2001. Studies on fermentive production of squalene. World J. Microbiol. 17, 811–816. Bondioli, P., Mariani, C., Lanzani, A., Dedeli, E., Muller, A., 1993. Squalene recovery from olive oil deodorizer distillates. J. Am. Oil Chem. Soc. 70, 763–766. Boskou, D., Morton, I.D., 1975. Sterols in heated olive oil. J. Sci. Food Agric. 26, 537–538. Cortesi, N., Rovellini, P.A., Fedeli, E., 1996. Liquid chromatography in lipid analysis. Control of genuiness, research on quality. Riv. It. Sost. Grasse LXXIII, 397–400. De Leonardis, A., Macciola, V., De Felice, M., 1997. Rapid determination of squalene in virgin olive oils using gas-liquid chromatography. It. J. Food Sci. 1, 75–80. Dionisi, F., Prodolliet, J., Tagliaferri, E., 1995. Assessment of olive oil adulteration by reversed-phase high-performance liquid chromotography/amperometric detection of tocopherols and tocotrienols. J. Am. Oil Chem. Soc. 72, 1505–1511. Eisner, J., Iverson, J.L., Mozingo, A.K., Firestone, D., 1965. Gas chromatography of unsaponifiable matter. III. Identification of hydrocarbons, aliphatic alcohols, tocopherols, triterpenoid alcohols and sterols present in olive oils. J. AOAC. 48, 417–433. Eyres, L., Croft, J., McNeill, A., Nichols, P., 2002. Potential of squalene as a functional lipd in foods and cosmetics. Lipid Technolog. September, 104–108. Giacometti, J., 2001. Determination of aliphatic alcohols, squalene, αtocopherol and sterols in olive oils: direct method involving gas chromatography of the unsaponifiable fraction following silylation. Analyst 126, 472–475. Gimeno, E., Castellote, A.I., Lamuela-Raventós., De la Torre, M.C., López-Sabater, M.C., 2002. The effects of harvest and extraction methods on the antioxidant content (phenolics, α-tocopherol and βcarotene) in virgin olive oil. Food Chem. 78, 207–211. Grigoriadou, D., Androulaki, A., Psomiadou, E., Tsimidou, M.Z., 2007. Solid phase extraction in the analysis of squalene and tocopherols in olive oil. Food Chem. 105, 675–680. Grob, K., Artho, A., Mariani, C., 1992. Determination of raffination of edible oils and fats by olefinic degradation products of sterols and squalene using coupled LC-GC. Fat Sci. Techn. 94, 394–400. Guinda, A., Lanzon, A., Albi, T., 1996. Differences in hydrocarbons of virgin olive oils obtained from several olive varieties. J. Agric. Food Chem. 44, 1723–1726. Hewavitharana, A.K., 2003. Simple solutions to problems encountered in quantitative analysis of tocopherols and tocotrienols using silica columns. Anal. Biochem. 313, 342–344. Jahaniaval, F., Kakuda, Y., Marcone, M.F., 2000. Fatty acid and triacylglycerol compositions of seed oils of five Amaranthus accessions and their comparison to other oils. J. Am. Oil Chem. Soc. 77, 847–852. Kalogeropoulos, N., Andrikopoulos, N.K., 2004. Squalene in oils and fats from domestic and commercial fryings of potatoes. Int. J. Food Sci. Nutr. 55, 125–129. Kamal-Eldin, A., Görgen, S., Patterson, J., Lampi, A.M., 2000. Normalphase high-performance liquid chromatography of tocopherols and tocotrienols. Comparison of different chromatographic columns. J. Chromatogr. A 881, 217–227. Lanzón, A., Albi, T., Cert, A., Gracián, J., 1992. The hydrocarbon fraction of virgin olive oil and changes resulting from refining. Gracas y Aceites 43, 271–276. Lanzón, A., Albi, T., Cert, A., Gracián, J., 1994. The hydrocarbon fraction of virgin olive oil and changes resulting from refining. J. Am. Oil Chem. Soc. 71, 285–291.
CHAPTER | 61 Squalene and Tocopherols in Olive Oil: Importance and Methods of Analysis
Lanzón, A., Guinda, A., Albi, T., an de la Osa, C., 1995. Metodo rapido para la determinación de escualeno en aceites vegetales. Gracas y Aceites 46, 276–278. López Ortíz, C.M., Prats Moya, M.S., Berenguer Navarro, V., 2006. A rapid chromatographic method for simultaneous determination of β-sitosterol and tocopherol homologues in vegetable oils. J. Food Comp. Anal. 19, 141–149. Lu, H.-T., Jiang, Y., Chen, F., 2004. Determination of squalene using high performance liquid chromatography with diode array detection. Chromatographia 59, 367–371. Manzi, P., Panfili, G., Esti, M., Pizzoferrato, L., 1998. Natural antioxidants in the unsaponifiable fraction of virgin olive oils from different cultivars. J. Sci. Food Agric. 77, 115–120. Mariani, C., Venturini, S., Bondioli, P., Fedeli, E., Grob, K., 1992. Evaluation of the variations produced by bleaching process on more meaningful minor components free and esterified in olive oil. Riv. It. Sost. Grasse 69, 393–399. Marinova, E., Yanishlieva, N., Toneva, A., Psomiadou, E., Tsimidou, M., 2001. Changes in the oxidation stability and tocopherol content in oils during microwave heating. Riv. It. Sost. Grasse LXXVIII, 529–533. Moreda, W., Pérez-Camino, M.C., Cert, A., 2001. Gas and liquid chromatography of hydrocarbons in edible vegetable oils. J. Chromatogr. A 936, 159–171. Nenadis, N., Tsimidou, M., 2002. Determination of squalene in olive oil using fractional crystallization for sample preparation. J. Am. Oil Chem. Soc. 79, 257–259. Newmark, H.L., 1997. Squalene, olive oil and cancer: a review and hypothesis. Cancer Epidemiol. Biomarkers Prev. 6, 1101–1103. Nikolaides, N., 1974. Skin lipids: their biochemical uniqueness. Science 186, 9–26. Ostlund, R.E., Racette, S.B., Stenson, W.F., 2002. Effects of trace components of dietary fat on cholesterol metabolism: phytosterols. Nutr. Rev. 60, 349–359. Parcerisa, J., Casals, I., Boatella, J., Codony, R., Rafecas, M., 2000. Analysis of olive and hazelnut oil mixtures by high-performance liquid chromatography–atmospheric pressure chemical ionisation mass
567
spectrometry of triacylglycerols and gas–liquid chromatography of non-saponifiable compounds (tocopherols and sterols). J. Chromatogr. A 881, 149–158. Pérez-Camino, M.C., Moreda, W., Mateos, R., Cert, A., 2002. Determination of esters of fatty acids with low molecular weight alcohols in olive oils. J. Agric. Food Chem. 50, 4721–4725. Psomiadou, E., Tsimidou, M., 1998. Simultaneous HPLC determination of tocopherols, carotenoids and chlorophylls for monitoring their effect on virgin olive oil oxidation. J. Agric. Food Chem. 46, 5132–5138. Psomiadou, E., Tsimidou, M., 1999. On the role of squalene in olive oil stability. J. Agric. Food Chem. 47, 4025–4032. Psomiadou, E., Tsimidou, M., 2002a. Stability of virgin olive oil. I. Autoxidation studies. J. Agric. Food Chem. 50, 716–721. Psomiadou, E., Tsimidou, M., 2002b. Stability of virgin olive oil. II. Photo-oxidation studies. J. Agric. Food Chem. 50, 722–727. Psomiadou, E., Tsimidou, M., Boskou, D., 2000. α-Tocopherol levels of Greek virgin olive oil. J. Agric. Food Chem. 48, 1770–1775. Rao, C.V., Newmark, H.L., Reddy, B.S., 1998. Chemopreventive effect of squalene on colon cancer. Carcinogenesis 19, 287–290. Rupérez, F.J., Martin, D., Herrera, E., Barbas, C., 2001. Chromatographic analysis of α-tocopherol and related compounds in various matrices. J. Chromatogr. A 935, 45–69. Smith, T.J., 2000. Squalene: potential chemopreventive agent. Expert Opin. Investig. Drugs 9, 1841–1848. Tasioula-Margari, M., Okogeri, O., 2001. Simultaneous determination of phenolic compounds and tocopherols in virgin olive oil using HPLC and UV detection. Food Chem. 74, 377–383. Villén, J., Blanch, G.P., Ruiz del Castillo, M.L., Herraiz, M., 1998. Rapid and simultaneous analysis of free sterols, tocopherols, and squalene in edible oils by coupled reversed–phase liquid chromatography gas chromatography. J. Agric. Food Chem. 46, 1419–1422. Wetherbee, B.M., Nichols, P.D., 2000. Lipid composition of the liver oil of deep-sea sharks from the Chatman Rise, New Zealand. Comp. Biochem. Physiol. Part B 125, 511–521. Zingg, J.M., Azzi, A., 2004. Non-antioxidant activities of vitamin E. Curr. Med. Chem. 11, 1113–1133.
This page intentionally left blank
Chapter 62
An Overview of the Chemometric Methods for the Authentication of the Geographical and Varietal Origin of Olive Oils Federico Marini, Remo Bucci, Antonio L. Magrì and Andrea D. Magrì Department of Chemistry, Sapienza University of Rome, Italy
62.1 INTRODUCTION: TRACING THE ORIGIN OF OLIVE OIL During recent years, there has been an increasing demand for ‘quality’ as a primary criterion to access the market. The term ‘quality’ encompasses several different meanings, mainly being defined as ‘the totality of characteristics of an entity that bear on its ability to satisfy stated and implied needs’ (ISO 8402). It is apparent that several concepts can be listed and described by this term, such as genuineness, safety of use, tipicity, absence of adulterations and so on. In particular, when referring to a vegetable source foodstuff, it has been widely reported that its quality relies significantly on its origin (geographical and varietal). It is apparent that the authentication of the quality of a foodstuff can result in a relatively simple issue when it can be related to the objective determination of single analytical indices but a much more complex topic if the acquisition and the simultaneous evaluation of a greater host of chemical information is involved. The problem of protecting some high-quality foodstuffs from the competition of other products with the same commercial designation but a lesser value and hence, price can be considered within this framework. This issue has been recognized by the European Union which has been promulgating, since 1992, a series of norms aimed at protecting these high-quality productive identities, by the introduction of the Denominations of Origin (Protected Denomination of Origin, PDO, and Protected Geographical Indication, PGI) (The Commission of the European Communities, 1992). The foodstuffs which are labeled by these denominations must be produced in a welldefined geographic area and manufactured using only one or Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
a few specified botanical varieties. Consequently, to prevent fraud in this field, the quality control laboratories would need an analytical method to determine the origin of the sample. Unfortunately, even if a great host of instrumental analytical techniques are at present under investigation, none of those can be listed which can provide the quality control chemist with an index directly related to the origin of the product and that can consequently be involved in labeling compliance. On the other hand, chemometrics has been widely used in the literature to cope with classification problems involving the authentication of the origin of foodstuffs (Forina and Tiscornia, 1982; Derde et al., 1984; Forina and Lanteri, 1984; Zupan et al., 1994). This consideration led us, during the last 10 years, to inspect the possibility of using mathematical and statistical techniques to elaborate the multivariate data which are often measured on the samples for their commercial classification and routine quality control in order to classify the products according to their geographical or varietal origin (Marini et al., 2003a, b, 2004a, b). In particular, a significant part of our research has involved the authentication of extra virgin olive oil (Bucci et al., 2002; Marini et al., 2004c, 2006, 2007b), as a representative example of a food for which some denominations of origin have been defined, as geographical and/or botanical origin have been recognized as important variables regulating the overall quality of this product (tipicity). Therefore, in this chapter, the description of some chemometric approaches to the prediction and verification of the botanical and geographical origin of extra virgin oil samples will be presented together with some examples taken from the authors’ experience.
569
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
570
62.2 CHEMOMETRICS: WHAT IT IS AND WHY WE USE IT Chemometrics (Sharaf et al., 1986; Massart et al., 1988) is a relatively young discipline in chemistry, dating back to the 1970s, and can be defined as the ‘application of mathematical, statistical, logical and artificial intelligence methods to the optimization of chemical processes and to the elaboration and interpretation of chemical data’. Therefore, according to the definition reported above, all the aspects of the analytical process are covered by the different classes of chemometric methods. Practically speaking, these considerations mean that only a specific class of chemometric methods will be described and discussed in this section, i.e., the methods that are more directly involved in the authentication of the origin of olive oil. In statistically rigorous terms, the problem of assessing whether the origin of a sample is one or another falls into the realm of the so-called ‘pattern recognition’ methods. The scope of these methods is to find some criteria to assign an object to a specific category or, in other words, to ‘classify’ an individual (Kowalski and Bender, 1972). Coming back to our specific problem of authenticating a foodstuff, the term object or individual stands for sample while the word category or class is a synonym for the different geographical or varietal origins. For instance, if our goal is to assess whether an oil sample comes from Italy, Spain or Portugal, we will say that our problem considers three classes or categories, each class corresponding to a single origin. Already at this level, a first division among the various pattern recognition methods can be made based on whether the different categories among which the samples can be split are known a priori or not. Accordingly, one speaks of supervised or unsupervised methods, respectively. This division relies on the fact that when a priori information about classes exists, it can be used to define the criteria to be followed in order to predict the origin of future unknown samples, that is the ultimate scope of this kind of chemometric analysis. On the other hand, when no a priori information about the sample grouping is available, the first question to be ascertained is how many clusters of objects are present and which sample belongs to which. This step is usually accomplished by exploratory techniques, such as principal component analysis (PCA) or cluster analysis (CA). In the remainder of this methodological section, supervised techniques only will be described, since they are the most extensively used for authentication purposes.
62.2.1 Supervised Pattern Recognition Techniques As anticipated in Section 62.2, supervised pattern recognition techniques (see Table 62.1) actively use the additional information about the class of the samples to define the criteria that allow, whenever a new unknown sample
SECTION | I Natural Components
TABLE 62.1 Fundamentals of supervised pattern recognition 1. Classification of samples into groups based on the values of the experimental indices measured on them 2. Requires that, at least for a certain number of samples, the group (class, origin) is known a priori 3. These ‘sure-origin’ samples are used to build the models 4. The models are then used to predict the class (origin) of unknown samples (and usually also the probability of this prediction being correct) In this table the main issues of supervised pattern recognition techniques are described.
is analyzed, prediction of its category to a given accuracy (Vandeginste et al., 1998). In order to do so, a mathematical model linking the results of the measurement of a series of chemical and physical chemical properties of the samples (which will be referred to as variables in the remainder of this text) to the probability of the samples themselves belonging to a specific category has to be built. Consequently, the existing supervised pattern recognition techniques can be divided, for instance, into linear and non-linear, based on the functional form assumed by the model, or into parametric and non-parametric, according to whether a given distribution is assumed for the probability of class belonging or not. Together with these, particularly important is the distinction which can be made among pure classification (or discriminant) and class-modeling techniques.
62.2.1.1 Pure Classification (Discriminant) Techniques Pure classification techniques are mainly oriented in discriminating among the different groups, meaning that, if the samples that are used to build the model come only from four categories, whenever a new unknown sample will be analyzed, it will always be predicted as belonging to one of these four classes, even if it is totally different from all of them. In geometrical terms, they operate dividing the hyperspace of variables in as many regions as the number of classes so that, if a sample falls in the region of space corresponding to a particular category, it is classified as belonging to that category (Figure 62.1). (Since each sample is described by the numerical results of the measurements performed on it, these results can be organized into a vector, representing the coordinates of a point in a hyperspace (N-dimensional space) having as many axes as the number of variables.) This way, each sample is always assigned to one and only one class (Vandeginste et al., 1998). These methods are the most commonly used since they usually require a lower samples to variable ratio in order to provide accurate results but they have the problem
571
Variable 2
CHAPTER | 62 Chemometric Methods for the Authentication of Olive Oils
squares-discriminant analysis (PLS-DA) (Wold et al., 1983), back-propagation and counter-propagation artificial neural networks (BP- and CP-ANN) (Zupan and Gasteiger, 1999), support vector machines (Xu et al., 2006) and D-CAIMAN (Todeschini et al., 2007). In particular, linear discriminant analysis assumes that the conditional probabilities in Equation 62.1 are normally distributed with the same variance-covariance matrix for each group:
D C
A
B
Variable 1
FIGURE 62.1 Schematic representation of the principles of pure classification techniques in the case of samples from four classes described by two variables. The experimental region is divided in as many regions as the number of categories (four regions) so that if the values of the variables measured on a sample are such that it falls on the region A it will be predicted as belonging to class A and so on.
that the identification of an unknown sample coming from a category not considered in the modeling phase is not straightforward (since it will always be classified into one of the N modeled groups). Additionally, once a classification model is built, if a considerable number of samples from a not previously considered category become available, the only way of updating it is to build a brand new model including N ⫹ 1 groups. From a theoretical point of view, pure classification techniques use a set of samples whose true origin is known (training set) to build a probability-based classification rule, allowing any future sample to be assigned to one of the categories represented in this set. It can be shown (Duda et al., 2000) that whatever the technique chosen, the classification rule that provides the better prediction is ‘a sample should be assigned to the class for which it has the largest posterior probability’ (Bayes’ rule). The posterior probability, for each class, can be computed according to Bayes’ theorem: P(Gi | x, H 0 ) ⫽
p(x | Gi , H 0 )P0 (Gi | H 0 ) ∑i p(x | Gi , H 0 )P0 (Gi | H 0 )
(62.1)
where P(Gi|x,H0) is the posterior probability that a sample described by the random vector x belongs to group Gi, P0(Gi|H0) is the a priori probability of observing a sample from group Gi (prior), and p(x|Gi, H0) is the conditional probability of observing a random vector x for samples belonging to group Gi (likelihood). Therefore, pure classification is a two-stage process involving first the computation of the posterior probabilities that the sample belongs to each group (or any monotonically increasing function of these probabilities) and successively to assign the sample to the category for which the value of this probability is the highest. The various methods differ among themselves in the way the posterior probability (or its monotonically increasing function) is computed and include linear and quadratic discriminant analysis (LDA and QDA) (McLachlan, 1992), K-nearest neighbors (KNN) (Coomans and Massart, 1982), partial least
p(x | Gi ) ⫽
1 (2 π )
d /2
e⫺1 / 2( x⫺mi )
T
|Σ |
Σ⫺1 ( x⫺ mi )
(62.2)
where mi is the centroid of class i (i.e., a vector containing the means of each variable for that specific group) and Σ is the pooled variance-covariance matrix. Since probabilities are positive, taking the natural logarithm of posterior probabilities and considering that the quadratic term xTΣx is constant for each class, gives for each group a classification function – f(x|Gi) which is linear in the original variables, f (x | Gi ) ⫽ ln( p(x | Gi )) ⫽
∑ j a j x j ⫹ a0
(62.3)
hence the name of this method. Multilayer feed-forward artificial neural networks trained by the back-propagation algorithm (Rumelhart et al., 1986a, b; Zupan and Gasteiger, 1999) constitute a valid alternative to classical pattern recognition methods as, when Softmax Output is used (Bridle,1990), the outputs of a welltrained network can be interpreted as posterior probabilities. Roughly speaking, a feed-forward neural network operates a non-linear mapping between an input and an output space in an implicit way, via a combination of suitably weighted non-linear functions of the independent variables. Each of these functions is called a node or neuron, so that the variables it operates on are called the inputs of the neuron and the result is it produces its output. In a multilayer feed-forward network, the units are organized in layers so that all the neurons of a layer operate on the same inputs. In particular, feed-forward processing implies that the output of a layer becomes the input of the successive layer (Figure 62.2). Accordingly, three kinds of layers can be identified. The input layer contains as many units as the number of independent variables and it is used only to introduce the data in the network, while the ouput layer, made up of as many neurons as the number of categories, is used to provide an answer for a given set of input values. Indeed the output of the different output units represents the posterior probabilities of a sample belonging to the various classes. Together with these, there is a variable number of additional units, contained in one or more so-called hidden layers, that allow non-linearity in the data processing. In a fully connected artificial neural network, each neuron in a given layer is connected to each neuron in the following
572
SECTION | I Natural Components
y1
y2
y3 Output layer
Wjk Hidden layer Wij Input layer x1
x2
x3
x4
x5
x6
FIGURE 62.2 Schematic representation of a multilayer feed-forward neural network used for classification. This figure represents an ANN used to classify samples from three categories (three output neurons) using six variables (six input units). The four hidden neurons allow non-linearity in the modeling.
layer by an associated numerical weight. The weight connecting two neurons regulates the magnitude of signal that passes between them. In addition, each neuron possesses a numerical bias term, corresponding to an input of ⫺1 whose associated weight has the meaning of a threshold value. Information in an ANN is stored in these connection weights which can be thought of as the ‘memory’ of the system. The goal of back-propagation training is to change iteratively the weights between the neurons in a direction that minimizes the error E, defined as the squared difference between the desired and the actual outcomes of the output nodes, summed over the training patterns (training set data) and the output neurons according to the steepest descent method: Δwij (t ) ⫽ ⫺η
∂E ⫹ αΔwij (t ⫺ 1) ∂wij
(62.4)
The variation of a connection weight at the tth iteration depends on the partial derivative of the total error with respect to that weight through a proportionality constant termed learning rate (η) and on the variation of the same weight during the previous iteration by means of a momentum term (α). To compute the partial derivative of the error E with respect to the connection weights to the hidden layer(s) requires propagating backwards the prediction error E using the rules of chain derivation, hence the name ‘backpropagation’. Chain derivation acts as a way to ‘distribute’ the error E between the neurons of the hidden layer(s) in order to apply the iterative weight adjustment necessary for the learning of the network.
among the elements of a class rather than on discriminating among the different categories. In class-modeling each category is modeled separately. Objects fitting the model are considered part of the class, while objects which do not fit are rejected as non-members (Figure 62.3). When more than one class is modeled, three different situations can be encountered: each sample can be assigned to a single category, to more than one category or to no category at all (Vandeginste et al., 1998). With respect to pure classification techniques, class-modeling tools offer at least two main advantages. It is in principle possible to identify samples which do not fall in any of the examined categories and which, as a consequence, can be either simply outlying observations or members of a new class not considered during the modeling stage. Moreover, as each category is modeled separately, any additional class can be added without recalculating the already-existing class models. Therefore, it is apparent that in principle class-modeling techniques are the most suitable to be used in food authentication problems, particularly those concerning the origin of a foodstuff, where the number of the Denominations of Origin varies significantly with time and every year there are new categories entering the market. However, they usually require higher samples to variables ratio rather than the discriminant techniques. This issue can represent a problem even after a sensible variable reduction is performed. The most commonly used chemometric class-modeling techniques are SIMCA (Wold, 1976; Wold and Sjöström, 1977) and UNEQ (Derde and Massart, 1986). In particular, SIMCA describes the similarities among the samples of a category using a principal component model. The class space is then identified by combining information about leverage and residuals of the individuals into a distance to the model criterion. Depending on the way this information is combined, the resulting distance will follow a specific distribution that usually is related to Fisher’s F. In particular, samples having a distance-to-the-model corresponding to a probability of belonging to the class lower than 5% are rejected by the class model. On the other hand, UNEQ is based on the assumption of multivariate normality for each class population and can be considered as the modeling analog of quadratic discriminant analysis. In UNEQ, the class model is represented by the class centroid and the category space is defined on the basis of the Mahalanobis distance from this barycenter, corresponding to a desired confidence level (usually 95%). When a category is not homogeneous, the class space can be highly irregular. Recently, other non-linear class-modeling techniques have been developed, namely modeling CAIMAN (Todeschini et al., 2007) and two modeling versions of artificial neural networks (Marini et al., 2005; Marini et al., 2007a).
62.2.1.2 Class-Modeling Techniques
62.2.3 Validating the Models
Class-modeling techniques represent a different approach to pattern recognition, as they focus on modeling the analogies
As already stated, the aim of applying a pattern recognition technique to the authentication of the origin of a foodstuff
573
B
Variable 2
A
Variable 2
CHAPTER | 62 Chemometric Methods for the Authentication of Olive Oils
C A
A
A/B not A B
Variable 1
not A, B or C Variable 1
FIGURE 62.3 Schematic illustration of the principles of class modeling. (A) Each class is modeled separately and the class space is identified; (B) when more than one class is present, depending on whether the unknown sample falls, it can be assigned to one category, to more than one, or to no category at all.
is to be able to predict where the sample comes from to a given accuracy. Therefore, it is important to have an estimate of the prediction error of the chemometric methods and this is accomplished through a procedure called validation (Duda et al., 2000). The first way of estimating the predictive ability of a model is to analyze once again the samples used to build the model (training set) and compare the predicted origin to the true one to obtain a non-error rate. However, this prediction would be over-optimistic since the parameters of the model have been adjusted in order to obtain the minimum error on this set. It is then a normal practice to leave aside a set of samples of known origin to be used as an external set to check the true predictive ability of the model, called the validation set. The validation set should be large and representative enough to guarantee the accuracy of the estimate of the performances of the method. There are several algorithms that can be used, such as Kennard-Stone (Kennard and Stone, 1969) or Duplex (Snee, 1977), when the number of available samples is not sufficient. A good alternative is represented by cross-validation. In cross-validation, the data set is divided into k different splits so that in turn one is used as the validation set and the remaining k ⫺ 1 as the training set.
62.3 USING CHEMOMETRICS TO AUTHENTICATE ITALIAN OLIVE OILS: SOME EXAMPLES In Section 62.2.2, some of the most commonly used pattern recognition techniques used for the authentication of a foodstuff have been described together with their main characteristics. In this section, some examples of their application to the problem of predicting the varietal and geographical origin of extra virgin olive oil samples will be discussed.
62.3.1 Classification of Italian Monocultivar Oils The first example of application of different pattern recognition techniques to the classification of olive oils is a study on the possibility of discriminating monocultivar oil samples
based on the variety of the olives used in the manufacturing of the product. This work is paradigmatic because it shows how different techniques are needed depending on the complexity of the problem. Indeed in the first stage of our study (Bucci et al., 2002) only a set of 153 samples coming from five cultivars (Carboncella, Frantoio, Leccino, Moraiolo and Pendolino) all harvested in a restricted geographical area (Sabina, Lazio, Italy) were considered. Linear discriminant analysis was sufficient in order to obtain a 100% predictive ability as evaluated by four-fold cross-validation (Figure 62.4). It should be stressed that this correct classification rate was obtained using only some of the variables routinely used for the commercial classification of the product (fatty acids, triglycerides, sterols, UV extinction coefficients, acidity and peroxide number). Therefore, information on the varietal origin of the samples could be obtained without the need of measuring additional indices, but just by analyzing the customarily available results by chemometric techniques. When a non-linear technique such as multilayer feed-forward neural network was used for the same purpose, a 100% predictive ability was obtained using as descriptors only the concentration of six major fatty acids. On the other hand, when this data set was augmented with the results of the analyses on other monocultivar extra virgin olive oil samples from the same five varieties already considered but harvested in different regions and years (Marini et al., 2004c), then the simple linear discriminant analysis was not able to provide a 100% correct classification rate, even if the prediction errors were still below 10%. This means that the boundary between the different categories has become irregular and a linear model is not sufficient anymore to discriminate among the various cultivars. A perfect classification of all the training and validation samples was instead obtained when a non-linear technique – a multilayer feed-forward neural network – was used to discriminate the samples. This evidence suggested that in the last stage of this study, when the data set was further augmented with the measurements performed on 373 samples coming from another nine Italian varieties (Minuta (Nasitana) (50), Nocellara (del Belice) (118), Nociara (17), Ortice (40),
574
Scores on LV3 (8.58%)
SECTION | I Natural Components
4 3 2 1 0 −1 −2 −3 −4 6 5 4 3 2 1 0 −1 −2 −3 −4
5
4
Scores on LV1 (20.61%)
3
2
1
0
−1
−2
−3
−4
−5
Scores on LV2 (13.98%)
FIGURE 62.4 Discrimination of the varietal origin of monocultivar oils. Representation of the original four cultivar data set projected onto a discriminant subspace (䉲 Carboncella; * Frantoio; 䊏 Leccino; ⫹ Moraiolo; 䉫 Pendolino).
Ortolana (29), Ottobratica (18), Peranzana (Provenzale) (39), Racioppella (44) and Sinopolese (18)) harvested in various years and geographical regions, only neural networks had to be used to build a reliable classification model. Sixteen variables (linoleic, sitostanol, stigmasterol, campestanol, sitosterol, Δ5-avenasterol, linolenic, palmitic, clerosterol, cholesterol, oleic, campesterol, steraric, Δ5,24-stigmastadienol, palmitoleic and acidity) were necessary to achieve a 100% correct classification rate on the training set and a 100% correct prediction set both on the internal and external validation sets left aside to estimate the performances of the model.
62.3.2 Resolution of Binary Mixtures of Monocultivar Oils The previous study demonstrated that it is possible to accurately predict the varietal origin of extra virgin monocultivar olive oils using only a limited number of customarily determined chemical indices. Based on these results, in a successive stage of our research we have investigated whether it was possible to predict the composition of binary mixtures made of oils from two olive varieties (Marini et al., 2007b). Since we were lacking standard mixtures, a feasibility study was undertaken on computer-simulated binary mixtures calculated from the analytical data experimentally measured on the 153 monocultivar oils that constituted the first data set used in the research described in Section 62.3.1. The choice of using a simulated data set was justified by the fact that in this way we have been able to exploit a wide range of compositions and to take into account the intra-class and inter-class variability of the chemical indices considered. Indeed, almost 90 000 different mixtures were simulated taking into account all the possible
combinations between two samples coming from different cultivars and allowing for mixing ratios varying from 100:0 to 0:100 by steps of 10%. Particular care was then given to the selection of the samples to be used as training set to verify which approach would have resulted in the most representative set. The effect of using a random split, Kennard-Stone algorithm and Kohonen self-organizing maps (Zupan and Gasteiger, 1999) of different dimensions was evaluated and compared. In the modeling phase, a multilayer feed-forward neural network was used with respect to its use for classification problems. In this case, the outputs of the trained network are not interpreted anymore as probabilities of belonging to the different classes, but as the percentage of the individual varieties in the mixtures. The results for each of the ten investigated data sets (all the binary combinations two cultivars out of the five examined) are summarized in Figure 62.5, where the value of the predictive ability evaluated using each of the different ways of training/validation splitting (expressed as Q2, i.e. the R2 on the validation set) is reported. It can be immediately seen that using a Kohonen mapping to perform an ‘intelligent selection’ of the training set samples results in an improved predictive ability when confronted to the random choice. In particular, random selection of the samples resulted in an average difference between the actual and the desired percent composition of about 20%. Moreover, significantly different results between the different trials have been evidenced leading to a correspondingly high standard deviation for the final prediction error. As anticipated, definitely better results have been obtained using the ‘intelligent’ splitting of the data set according to the Kohonen-based mapping of the samples. In fact, all the 12 different mappings resulted in an improved accuracy of the results (in terms of prediction, expressed as Q2 computed for the validation set). In particular, depending on the
575
CHAPTER | 62 Chemometric Methods for the Authentication of Olive Oils
C/F
100
C/L
90
C/M F/L
80
F/M
70
L/M
60
C/P
Q^2 (%) 50
F/P L/P
40
M/P
30 20 10
random 1 random 2 random 3 random 4 random 5 random 6 random 7 random 8 random 9 random 10 random 11 random 12 random 13 random 14 random 15 Kohenen 20’20 (q) Kohenen 20’20 (h) Kohenen 25’25 (q) Kohenen 25’25 (h) Kohenen 30’30 (q) Kohenen 30’30 (h) Kohenen 35’35 (q) Kohenen 35’35 (h) Kohenen 40’40 (q) Kohenen 40’40 (h) Kohenen 45’45 (q) Kohenen 45’45 (h) Kennard-Stone
0
Set splitting technique
M/P L/P F/P C/P L/M F/M F/L C/M Cultivars C/L C/F
FIGURE 62.5 Resolution of binary mixtures of monocultivar oils. Predictive ability (Q2 on the validation set) of the investigated multilayer feedforward networks with different dataset splitting procedures. C ⫽ Carboncella; F ⫽ Frantoio; L ⫽ Leccino; M ⫽ Moraiolo; P ⫽ Pendolino.
data set, the observed Q2 varied between 0.65–0.75 (lowest dimension of the Kohonen layer) and 0.91–0.96 (largest networks), corresponding to the average difference between the predicted and the target values varying from about 13–15% to 5%–7.5%. Lastly, comparison with the outcomes resulting from Kennard-Stone splitting of the datasets shows that in all cases while resulting in an improved accuracy with respect to random division, the latter is significantly outperformed by the optimal Kohonen-based selection. The same results can also be graphically visualized in Figure 62.6 where the validation results of the neural networks analyses carried out on the binary mixture Frantoio/Leccino are reported as an example. The six panels represent the observed vs predicted values obtained after the best Kohonen and random splitting and after Kennard-Stone selection for both cultivars. It is apparent that the optimal Kohonen-based selection (panels A and B) leads to more accurate results (less dispersed around the diagonal of the graph).
62.3.3 Authentication of PDO Oils from Sicily The last example of the application of pattern recognition techniques to the authentication of the origin of extra virgin
olive oils is focused on the possibility of predicting or verifying labeling compliance of PDO samples (Marini et al., 2006). In particular, 200 samples from three PDOs of Sicily (Monti Iblei, Valli Trapanesi, Monte Etna, harvests 2002 and 2003) were analyzed for their fatty acid, triglyceride and sterol composition and these indices were used as variables for the successive modeling. Since the samples to variables ratio was allowing it, in this study it was possible to use class-modeling techniques to build the pattern recognition models to be used for the assessment of the origin of the samples. In particular, the two techniques, SIMCA and UNEQ, were used. Together with non-error classification and prediction rate (which are the results provided by all the classification methods), two additional figures of merit are usually computed to evaluate the performances of a class-modeling tool, sensitivity and specificity. Sensitivity is the percentage of samples from the modeled class which are accepted by the class model, while specificity is the percentage of samples from other classes which are rejected by the class model. SIMCA modeling was performed on the data set after separate category autoscaling (nine components were retained in each class model, accounting for more than 70% of the total variance). No sample resulted in an outlier for all the class models and only two or three samples
576
SECTION | I Natural Components
1.0
1.0 Predicted value (% Leccino)
B 1.2
Predicted value (% Frantoio)
A 1.2
0.8 0.6 0.4 0.2 0.0 −0.2 −0.2
0.8 0.6 0.4 0.2 0.0
0.0
0.2
0.4
0.6
0.8
1.0
−0.2 −0.2
1.2
0.0
Observed value (% Frantoio)
1.6
1.0 0.8 0.6 0.4 0.2 0.0 −0.2
0.8
1.0
1.2
1.0
1.2
1.0
1.2
1.2 1.0 0.8 0.6 0.4 0.2 0.0 −0.2 −0.4
−0.4
−0.6 0.0
0.2
0.4
0.6
0.8
1.0
−0.8 −0.2
1.2
0.0
Observed value (% Frantoio)
0.2
0.4
0.6
0.8
Observed value (% Leccino) F 1.2
1.0
1.0 Predicted value (% Leccino)
E 1.2
Predicted value (% Frantoio)
0.6
1.4
1.2
Predicted value (% Leccino)
Predicted values (% Frantoio)
D 1.8
1.4
0.8 0.6 0.4 0.2 0.0 −0.2 −0.2
0.4
Observed value (% Leccino)
C 1.6
−0.6 −0.2
0.2
0.8 0.6 0.4 0.2 0.0 −0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Observed value (% Frantoio)
−0.4 −0.2
0.0
0.2
0.4
0.6
0.8
Observed value (% Leccino)
FIGURE 62.6 Resolution of Frantoio/Leccino mixtures. Comparison between the actual and the predicted (validation set) ‘% Frantoio’ and values ‘% Leccino’ in the case of (A) and (B) Kohonen 45 ⫻ 45; (C) and (D) best random; (E) and (F) Kennard-Stone sample selections.
were rejected by their specific category models resulting in a high sensitivity for all the three classes. However, the three class models built using SIMCA showed a rather poor specificity. In particular, the model built for the class Monte Etna seems to be the only one with an acceptable specificity both versus Valli Trapanesi (78.85%) and versus Monti Iblei (93.68%). On the other hand, it seems to be
very difficult for these class models to distinguish among the samples from the two PDO Valli Trapanesi and Monti Iblei, which show a mutual specificity of about 40%. However, even with this rather low specificity, it is possible to achieve relevant modeling and prediction non-error rate (88% and 81% respectively). Also in this case, the number of misclassifications (both in the modeling and the
Distance to Monti Iblei
CHAPTER | 62 Chemometric Methods for the Authentication of Olive Oils
95%
Distance to Monte Etna FIGURE 62.7 Class-modeling of PDO from Sicily – SIMCA modeling on the complete dataset. Coomans plot for the classes Monte Etna (䊏) and Monti Iblei (䊊); Valli Trapanesi samples (䉱) are also shown.
validation phase) is not evenly distributed among the different classes, the highest number of errors being for Valli Trapanesi. The results of SIMCA modeling are displayed graphically in Figure 62.7, where the Coomans plot for a selected of categories is reported. In a Coomans plot (Coomans et al., 1984), the two axes represent the distance of each sample from a specific category, so that each class model is drawn as a rectangle corresponding to the critical distance (p ⫽ 0.05) from the class. Any sample having a distance to the corresponding centroid greater than the critical distance is considered as being outside the class model and, as a consequence, rejected as an outlier for the specific category (graphically, it is plotted outside the rectangle defining the class model). Moreover, the samples plotted onto the lower left square of the diagram are assigned to both classes. Successively, we have modeled the same data set with a different class-modeling technique, UNEQ, after variable reduction by stepwiseLDA. The three UNEQ class models showed a comparable sensitivity with respect to the models built with SIMCA (the only difference being three more Valli Trapanesi and one less Monti Iblei sample rejected by their category models); however, the specificity of the individual class models is rather poor. In the second stage of our study, we have tried to investigate if the year of harvesting could have an effect on our modeling of the three PDOs and perhaps be responsible for the low specificity of our models built on the whole data set. In order to do so, we have modeled in two separate stages the samples harvested in 2002 and those harvested in 2003, so that we could check whether the samples from a particular year were accepted by the model built on samples
577
from the other year or not. As far as the SIMCA modeling of the 2002 data set (104 samples) is concerned, while sensitivity was comparable to that obtained on the whole data set, the specificity values were significantly improved up to more than 20%. Additionally, a significant increase in the non-error rate was observed both in the modeling and the validation phase. When this model was successively used to analyze the 2003 samples, in order to investigate how samples harvested in a different year would have fitted, it was observed that only a small fraction (five of the 24) of the Valli Trapanesi samples harvested in 2003 were accepted by the 2002 model. The computations performed using UNEQ on the same 2002 data set led to similar results also in the prediction of samples harvested in 2003. Indeed, more than 50% of the 2003 Valle Trapanesi samples were rejected by their 2002 class model. Furthermore, it should be pointed out that both the models computed on the 2002 data set recognized a significant number of samples (19 in SIMCA and 13 in UNEQ) as outliers for all the classes. Finally, we repeated the computation using the 2003 data set (96 samples) to build the class model with the two techniques and then evaluating the performance of the models on the samples harvested in 2002. When considering the 2003 samples, the cross-validated sensitivities for all the categories are very good with both techniques (ranging from 88.46% to 97.82%); however, the specificities are rather poor, most of the values being lower than 50%.
62.4 CONCLUSIONS In this chapter, some examples of the application of different chemometric pattern recognition techniques to the authentication of the origin of extra virgin olive oils (in particular, from Italy). Attention was focused both on the geographical and on the varietal origin and it was shown when the complexity of the problem increases, it is necessary also that the complexity and non-linearity of the model used increases accordingly. Additionally, the last of the examples presented, involving the use of class-modeling techniques, has shown how these methods are more suited to be used in the problems involving the authentication of PDO products. They not only provide the usual non-error rate, but also additional figures of merit like sensitivity and specificity that can help understand which class is more likely to be confounded with which. Moreover, they allow the identification of samples not fitting any model and therefore being outlying for some reason (in the example, samples from 2003 were significantly different from samples from 2002 since summer 2003 in Italy was extraordinarily hot).
SUMMARY POINTS ●
The origin of olive oil is of capital importance in determining its quality.
578
●
●
●
●
●
●
SECTION | I Natural Components
At present, no single analytical technique allows the identification of the geographical or varietal origin of olive oil. A multivariate statistical approach is needed (supervised pattern recognition). Different supervised pattern recognition techniques can be used according to the complexity of the problem. Linear discriminant analysis is sufficient when the measured variables allow a good separation of the samples. Non-linear techniques such as artificial neural networks are necessary when the separation among the classes is more irregular. When a sufficient number of samples is available, classmodeling techniques are recommended, since they have greater flexibility, allowing a sample to be confused or not assigned to any class.
REFERENCES Bridle, J.S., 1990. Probabilistic interpretation of feedforward classification network outputs, with relationships to statistical pattern recognition. In: Fogleman Soulie, F., Herault, J. (Eds.) Neurocomputing: Algorithms, Architectures and Applications. Springer-Verlag, Berlin, pp. 227–236. Bucci, R., Magrì, A.D., Magri, A.L., Marini, D., Marini, F., 2002. Chemical authentication of extra virgin olive oil varieties by supervised chemometric procedures. J. Agric. Food Chem. 50, 413–418. Coomans, D., Massart, D.L., 1982. Alternative K-nearest neighbour rules in supervised pattern recognition. Part 1. K-nearest neighbour classification by using alternative voting rules. Anal. Chim. Acta 136, 15–27. Coomans, D., Broeckaert, I., Derde, M.P., Tassin, A., Massart, D.L., Wold, S., 1984. Use of a microcomputer for the definition of multivariate confidence regions in medical diagnosis based on clinical laboratory profiles. Comp. Biomed. Res. 17, 1–14. Derde, M.P., Coomans, D., Massart, D.L., 1984. SIMCA (Soft Independent Modeling of Class Analogy) demonstrated with characterization and classification of Italian olive oil. J. Assoc. Off. Anal. Chem. 67, 721–727. Derde, M.P., Massart, D.L., 1986. UNEQ: a disjoint modelling technique for pattern recognition based on normal distribution. Anal. Chim. Acta 184, 33–51. Duda, R.O., Hart, P.E., Stork, D.G., 2000. Pattern Classification, 2nd edn. Wiley Interscience, New York. Forina, M., Tiscornia, E., 1982. Pattern recognition methods in the prediction of Italian olive oil origin by their fatty acid content. Ann. Chim. 72, 143–155. Forina, M., Lanteri, S., 1984. Data analysis in food chemistry. In: Kowalski, B.R. (Ed.), Chemometrics. Mathematics and Statistics in Chemistry. Riedel Publishing, Dordrecht, pp. 305–351. Kennard, R.W., Stone, L.A., 1969. Computer-aided design of experiments. Technometrics 11, 137–148. Kowalski, B.R., Bender, C.F., 1972. Pattern recognition. A powerful approach to interpreting chemical data. J. Am. Chem. Soc. 94, 5632–5639. Marini, F., Magrì, A.L., Marini, D., Balestrieri, F., 2003a. Characterization of the lipid fraction of niger seeds (Guizotia Abyssinica Cass.) from different regions of Ethiopia and India and chemometric authentication of their geographical origin. Eur. J. Lipid Sci. Tech. 105, 697–704.
Marini, F., Balestrieri, F., Bucci, R., Magrì, A.L., Marini, D., 2003b. Supervised pattern recognition to discriminate the geographical origin of rice bran oils: a first study. Microchem. J. 74, 239–248. Marini, F., Zupan, J., Magrì, A.L., 2004a. On the use of counterpropagation artificial neural networks to characterize Italian rice varieties. Anal. Chim. Acta 510, 231–240. Marini, F., Magrì, A.L., Balestrieri, F., Fabretti, F., Marini, D., 2004b. Supervised pattern recognition applied to the discrimination of the floral origin of Italian honey samples. Anal. Chim. Acta 515, 117–125. Marini, F., Balestrieri, F., Bucci, R., Magrì, A.L., Marini, D., 2004c. Supervised pattern recognition to authenticate Italian olive oil varieties. Chemometr. Intell. Lab. Syst. 73, 85–93. Marini, F., Zupan, J., Magrì, A.L., 2005. Class-modeling using Kohonen artificial neural networks. Anal. Chim. Acta 544, 306–314. Marini, F., Magrì, A.L., Bucci, R., Balestrieri, F., Marini, D., 2006. Classmodeling techniques in the authentication of PDO Italian oils from Sicily. Chemom. Intell. Lab. Syst. 80, 140–149. Marini, F., Magrì, A.L., Bucci, R., 2007a. Multilayer feed-forward neural networks for class-modeling. Chemom. Intell. Lab. Syst. 88, 118–124. Marini, F., Magrì, A.L., Bucci, R., Magrì, A.D., 2007b. Use of different artificial neural networks to resolve binary blends of monocultivar Italian olive oils. Anal. Chim. Acta 599, 232–240. Massart, D.L., Vandeginste, B.G.M., Deming, S.N., Michette, Y., Kaufman, L., 1988. Chemometrics: A Textbook. Elsevier, Amsterdam. McLachlan, G., 1992. Discriminant Analysis and Statistical Pattern Recognition. Wiley, New York. Rumelhart, D.E., Hinton, G.E., Williams, R.J., 1986a. Learning internal representations by error back-propagation. In: Rumelhart, D.E., McClelland, J.L. (Eds.) Parallel Distributed Processing. Explorations in the Microstructure of Cognition. MIT Press, Cambridge, pp. 318–362. Rumelhart, D.E., Hinton, G.E., Williams, R.J., 1986b. Learning internal representations by back-propagating errors. Nature 323, 533–536. Sharaf, M.A., Illman, D.L., Kowalski, B.R., 1986. Chemometrics. Wiley, New York. Snee, R.D., 1977. Validation of regression models: methods and examples. Technometrics 19, 415–428. The Commission of the European Communities 1992. Regulation 2081/92. Off. J. Commission Eur. Commun. L208, 1–8. Todeschini, R., Ballabio, D., Consonni, V., Mauri, A., Pavan, M., 2007. CAIMAN (Classification and Influence Matrix Analysis): a new approach to the classification based on leverage-scaled functions. Chemom. Intell. Lab. Syst. 87, 3–17. Vandeginste, B.G.M., Massart, D.L., Buydens, L.M.C., De Jong, S., Lewi, P.J., Smeyers-Verbeke, J., 1998. Supervised pattern recognition. In: Handbook of Chemometrics and Qualimetrics: Part B. Elsevier, Amsterdam, pp. 207–241. Wold, S., 1976. Pattern recognition by means of disjoint principal components models. Pattern Recogn 8, 127–139. Wold, S., Sjöström, M., 1977. SIMCA: a method for analysing chemical data in terms of similarity and analogy. In: Kowalski, B. R. (Ed.), Chemometrics, Theory and Application. ACS Symposium Series No. 52, American Chemical Society, Washington DC, pp. 243–282. Wold III, S., Albano, C., Dunn, W.J., Esbensen Jr., K., Hellberg, S., Johansson, E., Sjöström, M., 1983. Pattern recognition: finding and using patterns in multivariate data. In: Martens, H., Russwurm, H. (Eds.) Food Research and Data Analysis. Applied Science, Barking, pp. 147–188.
CHAPTER | 62 Chemometric Methods for the Authentication of Olive Oils
Xu, Y., Zomer, S., Brereton, R.G., 2006. Support vector machines: a recent method for classification in chemometrics. Crit. Rev. Anal. Chem. 36, 177–188. Zupan, J., Novic, M., Li, X., Gasteiger, J., 1994. Classification of multicomponent analytical data of olive oils using different neural networks. Anal. Chim. Acta 292, 219–234.
579
Zupan, J., Gasteiger, J., 1999. Neural Networks in Chemistry and Drug Design, 2nd edn. Wiley-VCH, Weinheim.
This page intentionally left blank
Chapter 63
Characterization of Three Portuguese Varietal Olive Oils Based on Fatty Acids, Triacylglycerols, Phytosterols and Vitamin E Profiles: Application of Chemometrics Joana S. Amaral1,2, Isabel Mafra1,3 and M. Beatriz P.P. Oliveira1 1
REQUIMTE/Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal Escola Superior de Tecnologia e de Gestão, Instituto Politécnico de Bragança, Portugal 3 Escola Superior de Biotecnologia, Universidade Católica Portuguesa, Porto, Portugal
2
63.1 INTRODUCTION In Portugal, olive oil production is considered an ancient activity, where old olive groves can still be observed. In the last few years monovarietal groves seem to be increasing, though some disadvantages, such as the susceptibility to insects and diseases, can result from the growth of individual olive varieties (Aguilera et al., 2005). In some typical producer countries, the olive cultivation is being improved by renewing old trees, reducing the association with other crops, selecting the olive varieties suited to local agroclimates and planting new single variety orchards (Criado et al., 2008). This is leading to an increase in the prevalence of monovarietal olive oils. Portugal is one of the ten largest olive oil producers in the world, although in recent years (2004–2006) the olive oil production decreased from approximately 50 000 to 26 000 hL. Trás-os-Montes, located in the northeast of Portugal, is the main region of olive oil production. In this province, agriculture is still one of the major activities, where table olives and olive oil productions are considered essential to the regional economy. Several factors comprising the region climate conditions, the soils, the existence of dominant varieties, the traditional production techniques and the high quality of the final product led to the creation of a Protected Designation of Origin (PDO) olive oil. After the approval of the EU Regulation (EC) No. 2081/92 and the transposition to national laws, PDO olive oils with the designation of ‘Azeite de Trás-os-Montes DOP’ started being produced and commercialized. Four olive varieties are authorized for the production of this PDO olive oil, namely Cobrançosa, Madural, Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Verdeal Transmontana (also known as Verdeal) and Cordovil. The latter has no commercial expression in the region. The other three varieties account for more than 90% of olive cultivation in the region (Matos et al., 2007). Several factors are known to affect olive oil quality, namely: variety, stage of ripening, environmental conditions, cultural practices, method of oil extraction and storage conditions (Pereira et al., 2002; D’Imperio et al., 2007; Matos et al., 2007). Among them, the variety is considered to be of most importance, influencing olive oil quality and sensory characteristics. Its importance arises from the genetically defined enzymes involved in the lipoxygenase pathway, which constrain the production of flavor compounds and determine the inherent sensory quality of the virgin olive oil (Manai et al., 2008). Nevertheless, other authors state that environmental factors can sometimes exert a stronger influence than genetic factors regarding the chemical components that determine monovarietal virgin olive oil quality (Beltrán et al., 2005; Criado et al., 2008).
63.2 COMPOSITIONAL CHARACTERISTICS OF PORTUGUESE MONOVARIETAL OLIVE OIL FROM TRÁS-OS-MONTES The authenticity and traceability of olive oils has been the object of numerous studies in the past few years. Generally, characterization of monovarietal olive oils requires the study of several compounds, including major components, such as fatty acids (FA), triacylglycerols (TAG), and the minor ones, such as phytosterols, polyphenols, tocopherols,
581
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
582
SECTION | I Natural Components
63.2.1 Fatty Acids Composition
tocotrienols and pigments. FA have been successfully used for grouping olive oils according to the olive variety used. Besides, their knowledge is also important to estimate nutritional properties of oils as saturated FA are associated with a higher risk of cardiovascular diseases, contrarily to mono- and polyunsaturated FA. Several results have also proved the usefulness of TAG in grouping monovarietal olive oils, suggesting that the FA content and distribution on the glycerol moiety can contribute to the establishment of a cultivar chemical fingerprint (D’Imperio et al., 2007). Minor compounds, such as phenolics and tocopherols, important for their antioxidant activity, are also able to provide useful information for authenticity assessment. The geographical characterization of olive oils is another important aspect of concern, especially for those certificated as PDO, since they are added-value products. Chemical characterization followed by multivariate statistical analysis has been reported in several studies to be a useful tool for the classification of olive oils regarding their geographical origin (Aguilera et al., 2005; Ollivier et al., 2006) (Table 63.1). Known as high-quality products, the olive oils from Trásos-Montes region achieve very high market prices, which can propitiate unfair commercial practices. It is therefore important to use analytical techniques to ensure the assessment of identity and quality of these oils, and guarantee the proper product classification. Because of the importance of Cobrançosa, Madural and Verdeal varieties in Portuguese olive oil, especially concerning the ‘Azeite de Trás-osMontes’ PDO olive oil, several studies reporting their chemical composition have been conducted in recent years.
Fatty acid composition of edible oils is generally determined by gas–liquid chromatography coupled to flame ionization detection. This methodology has been applied in the determination of FA profile of Trás-os-Montes olive oil varieties (Figure 63.1). Several works report the FA profile of olive oil varieties together with the influence of different parameters on the oil composition, such as the effect of olive storage previous to oil extraction (Pereira et al., 2002) and the influence of the maturation index of the fruits (Matos et al., 2007). Table 63.2 shows the
TABLE 63.1 Key features of olive oil chemical characterization. 1. Triacylglycerols are the major compounds in olive oil and comprise three fatty acids linked to a glycerol molecule 2. According to the number of double bounds, fatty acids can be saturated (no double bond), monounsaturated (1) or polyunsaturated (⬎1) 3. Phytosterols and tocopherols are minor compounds present in the unsaponificable fraction of olive oil; they are thought to be beneficial health compounds 4. PDO olive oils are considered as added-value products 5. Principal component analysis (PCA) and discriminant analysis are multivariate statistical analysis tools frequently applied to chemical data exploitation
0.20
0.20 6
0.18
0.16
0.16
0.14
0.14
0.12
0.12
0.10
0.10 7
0.08
0.08
1
0.06
0.06
0.04 2
0.02
Value
Value
0.18
3
0.04
5 4
10 8
9
11
0.02
0.00
0.00 0
5
10
15 Minutes
20
25
FIGURE 63.1 Fatty acid profile of a virgin olive oil sample of Madural variety obtained by gas-chromatography. Peak legend: 1–C16:0; 2–C16:1n7; 3–C17:0; 4–C17:1n7; 4–C18:0; 6–C18:1n9; 7–C18:2n6; 8–C20:0; 9–C20:1n9; 10–C18:3n3; 11–C22:0.
583
CHAPTER | 63 Characterization of Three Portuguese Varietal Olive Oils: Application of Chemometrics
major FA and the total saturated (SFA), monounsaturated (MUFA), polyunsaturated fatty acids (PUFA), trans isomers and oleic/linoleic acid ratio reported for these cultivars (Pereira et al., 2002, 2004; Matos et al., 2007). It can be observed that, independently of the factor under study in each work, Cobrançosa variety consistently showed the highest levels of SFA and the lowest of unsaturated fatty acids (MUFA and PUFA). Verdeal variety showed the highest mean values of MUFA and the lowest of PUFA, whereas Madural variety was characterized by the lowest levels of MUFA and the highest of PUFA. According to Pereira et al. (2002), these considerable differences in saturated and unsaturated FA contents among varieties can explain the variation of oil stability measured on a Rancimat apparatus (4 h for Madural, 10 h for Verdeal and 12 h for Cobrançosa). Oleic/linoleic ratio is also a parameter frequently used to assess oil stability (Velasco and Dobarganes, 2002), as the varieties with the higher ratios are those with higher oxidative stability. Madural variety presented the lowest value (6.0), which agrees with the results reported by Pereira et al. (2002). Concerning the presence of trans isomers, whose high consumption is generally associated with cardiovascular diseases, only traces were reported for all varieties.
63.2.2 Triacylglycerol Composition Although triacylglycerol (TAG) profile can be determined by gas–liquid chromatography, non-aqueous reversed-phase high-performance liquid chromatography (RP-HPLC) is the most frequently used methodology, avoiding difficulties associated with the low volatility and the thermal stress of polyunsaturated TAG (Amaral et al., 2004). Figure 63.2 shows an RP-HPLC chromatogram of the TAG profile of an olive oil sample. Considering that in RP-HPLC TAG are mainly separated according to their equivalent carbon number (ECN ⫽ CN–2DB; CN – total carbon number; DB – number of double bonds), those presenting the same ECN can sometimes be difficult to resolve representing a critical pair, i.e., two compounds co-eluting in the same peak. Twelve peaks were quantified based on the relative percentage peak area by Cunha et al. (2005) (Table 63.3 and Figure 63.2). Other compounds, such as, PoLO, LnOO, PoOO, PPoO, SOL and AOL can also be found as possible critical pairs, although they are in minor amounts in olive oils (Ollivier et al., 2006). Table 63.3 reports the mean values of TAG composition for Cobrançosa, Madural
TABLE 63.3 Triacylglycerols composition of Cobrançosa, Madural and Verdeal monovarietal olive oils (adapted from Cunha et al. (2005)). TABLE 63.2 Main fatty acid composition (%) for Cobrançosa, Madural and Verdeal monovarietal olive oilsa (adapted from Matos et al. (2007)). Cobrançosa
Madural
Verdeal
Min
Max
Min
Max
Min
Max
C16:0
9.01
10.43
9.83
10.92
9.54
10.18
C18:0
3.21
4.84
2.19
2.33
2.69
2.97
C18:1c
73.72
78.81
71.23
72.56
79.74 81.56
C18:2cc
5.28
8.14
11.29
12.85
2.54
3.15
C18:3ccc
0.55
0.74
0.89
0.99
0.58
0.72
Σ SFAb
13.75
15.39
12.71
13.79
13.08 14.15
Σ MUFA
74.76
79.54
72.04
73.38
80.95 82.68
Σ PUFA
5.91
8.89
12.19
13.80
3.12
3.87
Σ transFA
0.05
0.11
0.10
0.12
0.07
0.10
C18:1/C18:2
9.06
14.48
5.57
6.36
25.31 32.11
a
Minimum and maximum values concerning the evaluation of samples with different maturation indexes: Cobrançosa - MI1 to MI7, Madural – MI3 to MI7, Verdeal – MI1 to MI6. MI – maturation index. b Σ: sum.
Triacylglycerol ECNa Cobrançosa Madural (%)
Verdeal
LLLb
42
0.03 (0.02)c
0.03 (0.01)
0.06 (0.02)
LLO ⫹ PoLO
44
0.38 (0.10)
2.09 (0.31)
0.05 (0.03)
LLP ⫹ LnOO
44
0.56 (0.20)
0.40 (0.15)
0.42 (0.03)
LnOP
44
0.05 (0.02)
0.76 (0.80)
0.07 (0.02)
LOO ⫹ PoPP
46
9.45 (1.43)
18.44 (1.23) 3.56 (0.15)
PLO ⫹ PPoO
46
3.08 (2.59)
4.31 (0.23)
OOO
48
57.57 (2.26) 50.52 (0.92) 66.81 (0.79)
POO ⫹ SOL
48
21.01 (1.04) 20.20 (1.07) 23.44 (0.70)
PPO
48
0.68 (0.16)
0.83 (0.45)
1.08 (0.42)
SOO ⫹ AOL
50
6.54 (0.87)
2.15 (0.37)
3.65 (0.29)
SOP
50
0.53 (0.09)
0.17 (0.03)
0.25 (0.03)
OOA
52
1.13 (0.15)
ndd
0.05 (0.01)
0.60 (0.10)
This table summarizes the TAG composition of monovarietal olive oils. a ECN: equivalent carbon number; b P: palmitoyl; Po: palmitoleoyl; S: stearoyl; O: oleoyl; L: linoleoyl, Ln: linolenoyl; A: arachidoyl; c mean values and standard deviation in parenthesis. d nd: not detected.
584
SECTION | I Natural Components
6.0E+05
OOO
5.0E+05
5.0E+05
5.0E+05
LOO
POO
5.0E+05
PLO
5.0E+05
POP OOA
SOO SOP 0.0E+05
5.0E+04
μV
10.00
20.00
30.00
40.00
50.00
[min]
LLO
4.0E+04
3.0E+04
LnOO
2.0E+04
LLP 1.0E+04
LL L LnOP
0.0E+04 12.00
14.00
16.00
18.00
20.00
[min]
FIGURE 63.2 Triacylglycerol profile of a virgin olive oil sample of Madural variety obtained by HPLC/ELSD. Legend as in Table 63.3.
and Verdeal varieties. All the varieties showed a similar qualitative profile with OOO as the major compound (50.5–66.8%), followed by POO (20.2–23.4%) and LOO (3.6–18.4%), reflecting the high oleic acid content in olive oils. Concerning the main triacylglycerols (OOO, POO and LOO), Verdeal variety presented the highest percentages of OOO and POO (66.8% and 23.4%, respectively, as mean values) and the lowest of LOO (3.6%). Conversely, Madural variety presented the lower proportion of OOO and POO and the highest of LOO.
63.2.3 Phytosterol Composition Clinical studies have demonstrated that dietary intake of phytosterols, due to their structural similarity with cholesterol, can inhibit its intestinal absorption, thereby lowering total plasma cholesterol and low-density lipoprotein levels (Wong, 2001). Phytosterols have also been reported to present antioxidant, antibacterial and anti-inflammatory activities and may offer protection against cancers, such as breast, colon and prostate (Awad and Fink, 2000). Furthermore, they can influence the oil stability at high temperatures since some phytosterols can act as polymerization
reaction inhibitors (Velasco and Dobarganes, 2002). For all these reasons, data concerning phytosterol composition are of high relevance for quality and nutritional evaluation of the oil. Phytosterols are a major portion of the unsaponifiable fraction of olive oils and present a more or less characteristic profile, making it an important tool for assessing the genuineness of the oil. According to Aparicio and Aparicio-Ruiz (2000), the qualitative and quantitative differences in phytosterol composition among several vegetable oils make them a suitable parameter to verify the botanical origin of vegetable oils and, thus, to detect eventual adulterations/contaminations. Other authors have suggested that sterol composition can be useful in virgin oil characterization, especially in detecting the adulteration with hazelnut oil and, more recently, in the classification of virgin olive oil according to its variety (Lazzez et al., 2008). Although the dominant phytosterols in olive oils are reported to be β-sitosterol, Δ5-avenasterol and campesterol, several other minor compounds, such as cholesterol, stigmasterol, clerosterol, sitosterol, Δ7-stigmastenol and Δ7-avenasterol have also been reported to occur in olive oils (Matos et al., 2007; Lazzez et al., 2008). To prevent olive oil adulteration several international organizations,
CHAPTER | 63 Characterization of Three Portuguese Varietal Olive Oils: Application of Chemometrics
585
TABLE 63.4 Phytosterol composition for Cobrançosa, Madural and Verdeal monovarietal olive oilsa (adapted from Matos et al. (2007) and Alves et al. (2005)). Phytosterols (mg kg⫺1)
Cobrançosa
Madural
Verdeal
Min
Max
Min
Max
Min
Max
Cholesterol
0.29
1.07
0.36
0.95
0.35
0.74
Campesterol
4.97
9.10
4.56
7.30
4.07
7.34
Stigmasterol
1.01
2.82
1.84
3.70
1.35
2.00
Clerosterol
1.51
2.38
1.70
2.83
1.24
1.78
β-Sitosterol
143.21
237.24
147.38
255.35
120.41 193.96
Δ5-Avenasterol
13.62
20.91
15.22
23.32
8.76
13.96
Δ7-Avenasterol (%)
Nd
0.43
0.37
0.66
nd
0.40
Apparent β-sitosterol
95.1
97.2
95.0
96.0
95.4
95.8
Cholesterol
0.16
0.56
0.16
0.52
0.22
0.35
Campesterol
1.20
3.48
2.49
2.66
2.87
3.36
a Minimum and maximum values concerning the evaluation of samples with different maturation indexes: Cobrançosa – MI1 to MI7, Madural – MI3 to MI7, Verdeal – MI1 to MI6. MI – maturation index.
such as the Codex Alimentarius of the FAO/WHO, the EU and the International Olive Oil Council have established characteristic values for some of these compounds. Table 63.4 presents the sterol composition of olive oils produced with the varieties Cobrançosa, Madural and Verdeal. Considering the aforementioned regulations, all samples studied by Matos et al. (2007) presented a total sterol content far above the threshold of 1000 mg kg⫺1 established for virgin olive oil, presenting also a higher value than that demanded (ⱖ93%) for apparent β-sitosterol (comprising the sum of clerosterol, β-sitosterol, Δ5-avenasterol, β-sitosterol and Δ5,24-stigmastenol). The results reported by Matos et al. (2007) and Alves et al. (2005) were under the upper limit of 4% established by the EU (Regulation No. 2568/91/EEC and later amendments) for olive oil campesterol content and were in general below the established upper limits of cholesterol (0.5%). Regarding the study of the effect of ripening in sterol composition, Matos et al. (2007) reported a decreasing tendency, although with some variations along the maturation.
63.2.4 Tocopherols and Tocotrienols Composition Vitamin E is the term used to designate a group of lipidsoluble compounds, comprising four tocopherols (α-, β-, γand δ-) and four tocotrienols (α-, β-, γ- and δ-). This family of compounds is particularly important in preventing
lipid oxidation processes in olive oils, mainly due to their antioxidant activity. Moreover, a large range of biological activities have been ascribed for these compounds since vitamin E vitamers have been associated with a preventive action against reactive oxygen species in biological systems (Woollard and Indyk, 2003). Although in the past α-tocopherol was probably the most studied vitamer since it was considered the most active vitamin E isoform, nowadays several studies have shown that the other vitamers also have important roles in the human organism, and thus, are considered to contribute to the total bioactivity in foods. For example, γ-tocopherol and tocotrienols are correlated with the reduction of blood cholesterol levels (Mishima et al., 2003) and may have a chemopreventive action (Campbell et al., 2003). Vitamin E determination has also been used for the authenticity and quality assessment of oils, based on qualitative and quantitative profiles. Tocopherol contents in monovarietal olive oils of Cobrançosa, Madural and Verdeal varieties (Table 63.5) were reported by Matos et al. (2007) and the influence of olive storage on their contents in olive oils of the same varieties was reported by Pereira et al. (2002). In both studies, only the three major tocopherols were quantified, possibly due to the low values of the other vitamers. α-Tocopherol was the major compound, followed by γ- and δ-tocopherols, respectively, which is in accordance with other authors (Cunha et al., 2006), although they reported also the presence of β-tocopherol and two tocotrienols (α and γ) in ‘Azeite de Trás-osMontes’ PDO olive oils. The results reported by Matos et al.
586
SECTION | I Natural Components
TABLE 63.5 Tocopherol composition of Cobrançosa, Madural and Verdeal monovarietal olive oilsa. Tocopherols (mg kg⫺1)
Cobrançosa min
max
Madural
MI ⫽ 4b MI ⫽ 4c
min
max
MI ⫽ 6b
Verdeal MI ⫽ 5.7c
min
max
MI ⫽ 3b MI ⫽ 3c
α
221.4 291.7 222.6
199.6 (17.2) 202.2 219.9 209.1
160.1 (26.2) 133.6 188.5 135.4
128.2 (26.6)
β
0.9
1
0.9
1.5 (0.1)
0.8
0.9
0.9
1.0 (0.14)
0.9
0.9
0.9
0.5 (0.1)
γ
5.4
16.1
12
4.0 (0.4)
5.1
7.6
6.3
1.7 (0.2)
nd
4.0
3.1
2.8 (0.4)
Total
235.5 298
235.5
205.1 (17.6) 208.8 226.8 216.3
162.7 (26.4) 138.5 189.4 139.4
131.6 (26.4)
a Minimum and maximum values concerning the evaluation of samples with different maturation indexes: Cobrançosa-MI1 to MI7, Madural-MI3 to MI7, Verdeal-MI1 to MI6; MI- maturation index. Adapted from b Matos et al. (2007) and c Pereira et al. (2002); nd: not determined.
(2007) showed, in general, higher α- and γ-tocopherol contents. However, a more accurate observation of results by comparing values for the same olive maturation index used by Pereira et al. (2002), shows that the differences among αtocopherol contents in both studies were not so large (Table 63.5). Beltrán et al. (2005) referred that tocopherol profile can change depending on the variety, fruit ripening stage, climatic conditions and olive growing techniques. Since samples studied by Pereira et al. (2002) were collected in the 1998/1999 crop year and those of Matos et al. (2007) in 2000/2001, this can explain the differences among results. Furthermore, in the latter only healthy fruits were collected, while in the former 30% of fruits were attacked by the olive fly.
63.3 CHEMOMETRICS APPLIED TO MONOVARIETAL OLIVE OIL CHARACTERIZATION For authenticity assessment of monovarietal olive oils the chemical data frequently need to be further explored by using statistical tools. Multivariate statistical analysis has been used to recognize the parameters able to discriminate the olive varieties. Among multivariate statistical analysis, principal component analysis (PCA) and discriminant analysis occupy a very important position. Nowadays, powerful statistical software packages are available, facilitating the application of complex and sophisticated algorithms, even by individuals with no special mathematical background (Alves et al., 2005). Several examples of statistical analysis applied to olive oil compositional data are available in the literature (Diaz et al., 2005; Cerretani et al., 2006; Ollivier et al., 2006; D’Imperio et al., 2007), sometimes coupling several statistical techniques to obtain high discrimination levels and reliable models (Alves et al., 2005). Generally, the multivariate statistical approach applied to olive oil analytical data allows recognizing the most significant descriptive variables among many, and to cluster
olive oil samples, which can lead to a first classification of olive variety (Cichelli and Pertesana, 2004). Recently, statistical tools have been applied to explore several data obtained from the chemical analysis of Trás-os-Montes monovarietal olive oils, namely FA and tocopherols (Matos et al., 2007), sterols (Alves et al., 2005; Matos et al., 2007) and TAG (Cunha et al., 2005).
63.3.1 Application to TAG Profile Cunha et al. (2005) compared the values for TAG by PCA after using a one-way analysis of variance to determine the differences among mean scores of each TAG profile of the three varieties. The PCA plot obtained (Figure 63.3) was able to explain approximately 61.5% of total variance, with the first dimension (PC1) representing 35.7% and the second dimension (PC2) representing 25.8%. The most important TAG defined by these two components were LLO, LOO and PLO for PC1 and SOO, SOP and OOA for PC2. In this plot, the three varieties could be perfectly distinguished, forming three groups. Madural and Verdeal were separated on PC1: the samples from Madural were located on the right side of the plot since they presented higher contents of LOO, LLO and PLO. The Verdeal olive oils presented lower values of the mentioned TAG and were, consequently, located on the left side of the plot. Cobrançosa differed from the other counterparts since it presented higher contents of SOP, SOO and OOA, thus was separated on PC2, being located at the top of the plot. The results reported by Cunha et al. (2005) are in accordance with other works, which showed that TAG composition could be useful to discriminate the varieties used in olive oil production (Cerretani et al., 2006).
63.3.2 Application to FA and Sterol Profiles Several authors have reported the usefulness of FA composition (Ollivier et al., 2006; D’Imperio et al., 2007) and sterol composition (Diaz et al., 2005) associated with
587
CHAPTER | 63 Characterization of Three Portuguese Varietal Olive Oils: Application of Chemometrics
OOA SOO SOP
C18
CB C6
2
C7
CB CB
PC2 (25.78%)
CB CB CB
CB CBCBCB CB CBCB CB CB
1
CB CB CB LOO PLO LLO
CB CB 0
MD MD MD MD MD MD MD MD MD MD MD MD MD MD MDMD
VD VD VD VD VD VD −1
C1 C4
1
C5 C3
C18:1 0
C2
C18:1 V1 V3 V6 V2 V5
M5 M7M6M3 M4 −1
CB VD VD VD VD VD VD VD
−1
Factor 2:31.42%
2
0
−3
−2
−1
0
1
2
Factor 1:68.16%
FIGURE 63.4 Scores plot of fatty acid data from virgin olive oil samples of Cobrançosa (C), Madural (M) and Verdeal (V) varieties (reprinted from Matos et al. (2007) with permission from Elsevier).
1
PC1 (35.66%)
FIGURE 63.3 Scores plot of triacylglycerol data from virgin olive oil samples of Cobrançosa (CB), Madural (MD) and Verdeal (VD) varieties (reprinted from Cunha et al. (2005) with permission from Chiriotti Editori).
statistical tools in olive oil characterization. A model for the characterization of monovarietal olive oils produced from Cobrançosa, Madural and Verdeal varieties, based on their main phytosterol composition, was presented by Alves et al. (2005). The developed model was considered suitable to classify commercial PDO olive oils produced from those varieties. A wider approach exploiting several chemical parameters (FA, tocopherols and phytosterols), was proposed by Matos et al. (2007). In this work, data obtained from Trás-os-Montes monovarietal olive oils were subjected to statistical analysis to evaluate the ability of single or multiple parameters as tools for variety discrimination. First, the authors used a multivariate analysis of variance for testing the significant differences between mean values of the evaluated parameters. The authors also performed the Student t-test for independent groups, and a PCA to detect structures in the relationship between variables (Matos et al., 2007). Regarding the individual exploitation of data obtained from FA analysis, the authors reported significant differences (p ⬎ 0.01) among the three varieties for all the quantified FA with the exception of C16:0. The analysis of PCA explained 99.6% of total variance, with PC1 and PC2 representing 68.2% and 31.4%, respectively. To reduce the number of variables, yet keeping the maximum information possible, three FA were chosen by the definition of these two components, namely C18:0, C18:1 and C18:2. In the PCA plot presented by Matos et al. (2007) the three varieties could be perfectly discriminated by their FA composition (Figure 63.4). In general, Madural variety was discriminated by its higher linoleic acid content and lower
oleic acid. Verdeal variety presented higher values of oleic acid and lower values of linoleic acid. Cobrançosa variety was the one showing higher stearic acid content. In the same study, Madural and Verdeal varieties appeared as very homogeneous groups independent of the stage of ripening considered, while Cobrançosa presented a considerable dispersion of values, attributed mainly to the increasing stearic acid content along the stage of ripening. The discrimination achieved with statistical analysis of tocopherol data was less evident than the one performed with FA data. Nevertheless, Cobrançosa was differentiated from the others by the higher α- and γ-tocopherol contents, while Verdeal presented much lower values for these vitamers. Data obtained from phytosterol analysis were also evaluated using statistical tools (Matos et al., 2007). The phytosterols chosen to perform a PCA, namely stigmasterol, clerosterol, β-sitosterol and Δ5-avenasterol, were selected by their statistical significance and also by their abundance in olive oil samples. Considering the multivariate analysis applied to the individual parameters, namely FA, tocopherols and phytosterols, the best level of discrimination for the three monovarietal olive oils was achieved using the FA data. Identical results were obtained by Cerretani et al. (2006), who stated that the use of minor components for the differentiation of olive oil variety is not reliable, since other external factors can influence minor compound composition beyond the genetic factor.
63.3.3 Global PCA Using Fatty Acids, Phytosterols and Tocopherols Matos et al. (2007) performed a global PCA considering simultaneously all data from FA, phytosterols and tocopherols. The most informative variables that represent the main data structures were selected avoiding co-linearity problems.
588
SECTION | I Natural Components
2
●
M5 M6
γ-tocopherol
Factor 2:23.10%
1
0
M3 V6
C18:2 D5-avenasterol α-tocopherol
C7 C4
C2
C5 −1
C3 M4
●
V4 V5 V3 C18:1
V2 V1
C18
REFERENCES
C6 C1 β-sitosterol
−2 −2
−1
0
1
2
Factor 1:55.62%
FIGURE 63.5 Principal component analysis based on all relevant chemical parameters from virgin olive oils of Cobrançosa (C), Madural (M) and Verdeal (V) varieties (reprinted from Matos et al. (2007) with permission from Elsevier).
Six variables, namely C18:0, C18:1, α-tocopherol, γ-tocopherol, β-sitosterol and Δ5-avenasterol, were extracted and the information condensed and simplified in a single graph. Two principal components were enough to describe the main features of the data, with PC1 and PC2 representing 55.6% and 23.1% of the information, respectively (Figure 63.5). Madural samples were very similar and formed a distinct group, with the exception of one sample with a maturation index of 4 that was relatively different from all other oils. Verdeal samples were characterized by their highest oleic acid contents. Cobrançosa samples were more dispersed, although the samples having maturation indexes (MI) of 2, 4 and 7 were closely related to each other, the same happening with the samples with MI of 1, 3 and 6. In general, better results were achieved considering the individual statistical analysis performed with the FA data as well as with the TAG data, although samples could be differentiated when all data parameters are used. Considering the results of the several studies performed on monovarietal ‘Trás-os-Montes’ olive oil, one can conclude that it is possible to apply analytical methods, which in this case are relatively inexpensive and easy to implement (such as FA or TAG analysis), coupled with statistical analysis to characterize the oils obtained from the varieties used in the production of extra virgin ‘Trás-os-Montes’ olive oil.
SUMMARY POINTS ●
●
For authenticity assessment, the chemical data are further explored by using multivariate statistical analysis, such as principal component analysis and discriminant analysis. In general, better results were obtained using individual statistical analysis coupled to fatty acid or triacylglycerol data, although differentiation could also be achieved with several parameters (fatty acids, phytosterols and tocopherols) simultaneously.
Cobrançosa, Madural and Verdeal are allowed for the PDO ‘Trás-os-Montes Olive oil’ production and account for more than 90% of the region total production. Several chemical parameters such as fatty acids, triacylglycerols, phytosterols and tocopherols can be used to characterize an olive oil regarding its variety.
Aguilera, M.P., Beltrán, G., Ortega, D., Fernández, A., Jiménez, A., Uceda, M., 2005. Characterisation of virgin olive oil of Italian olive cultivars: “Frantoio” and “Leccino” grown in Andalusia. Food Chem. 89, 387–391. Alves, R.M., Cunha, S.C., Amaral, J.S., Pereira, J.A., Oliveira, M.B.P.P., 2005. Classification of PDO olive oils on the basis of their sterol composition by multivariate analysis. Anal. Chim. Acta 549, 166–178. Amaral, J.S., Cunha, S.C., Alves, M.R., Pereira, J.A., Seabra, R.M., Oliveira, B.P.P., 2004. Triacylglycerols composition of walnut (Juglans regia L.) cultivars: characterization by HPLC/ELSD and chemometrics. J. Agric. Food Chem. 52, 7964–7969. Aparicio, R., Aparicio-Ruiz, R., 2000. Authentication of vegetable oils by chromatographic techniques. J. Chromatogr. A 881, 93–104. Awad, A.B., Fink, C.S., 2000. Phytosterols as anticancer dietary component: evidence and mechanism of action. J. Nutr. 130, 2127–2130. Beltrán, G., Aguilera, M.P., Rio, C.D., Sanchez, S., Martinez, L., 2005. Influence of fruit ripening process on the natural antioxidant content of Hojiblanca virgin olive oils. Food Chem. 89, 207–215. Campbell, S., Stone, W., Whaley, S., Krishnan, K., 2003. Development of gamma (γ)-tocopherol as a colorectal cancer chemopreventive agent. Crit. Rev. Oncol./Hematol. 47, 249–259. Cerretani, L., Bendini, A., Del Caro, A., Piga, A., Vacca, V., Caboni, M.F., Toschi, T.G., 2006. Preliminary characterisation of virgin olive oils obtained from different cultivars in Sardinia. Eur. Food Res. Technol. 222, 354–361. Cichelli, A., Pertesana, G.P., 2004. High-performance liquid chromatographic analysis of chlorophylls, pheophytins and carotenoids in virgin olive oils: chemometric approach to variety classification. J. Chromatogr. A 1046, 141–146. Criado, M.-N., Romero, M.-P., Casanovas, M., Motilva, M.-J., 2008. Pigment profile and colour of monovarietal virgin olive oils from Arbequina cultivar obtained during two consecutive crop seasons. Food Chem. 110, 873–880. Cunha, S.C., Casal, S., Oliveira, M.B.P.P., 2005. Triacylglycerol profile by HPLC/ELSD as a discriminant parameter of varietal olive oils from Portugal. Ital. J. Food Sci. 4, 447–454. Cunha, S.C., Amaral, J.S., Fernandes, J.O., Oliveira, M.B.P.P., 2006. Quantification of tocopherols and tocotrienols in Portuguese olive oils using HPLC with three different detection systems. J. Agric. Food Chem. 54, 3351–3356. D’Imperio, M., Dugo, G., Alfa, M., Mannina, L., Segre, A.L., 2007. Statistical analysis on Sicilian olive oils. Food Chem. 102, 956–965. Diaz, G., Merás, I.D., Casas, J.S., Franco, M.F.A., 2005. Characterization of virgin olive oils according to its triglycerides and sterols composition by chemometric methods. Food Control 16, 339–347. Lazzez, A., Perri, E., Caravita, M.A., Khlif, M., Cossentini, M., 2008. Influence of olive maturity stage and geographical origin on some
CHAPTER | 63 Characterization of Three Portuguese Varietal Olive Oils: Application of Chemometrics
minor components in virgin olive oil of the Chemlali variety. J. Agric. Food Chem. 56, 982–988. Manai, H., Mahjoub-Haddada, F., Oueslati, I., Daoud, D., Zarrouk, M., 2008. Characterization of monovarietal virgin olive oils from six crossing varieties. Sci. Horticult. 115, 252–260. Matos, L.C., Cunha, S.C., Amaral, J.S., Pereira, J., Andrade, P., Seabra, R. M., Oliveira, B.P.P., 2007. Chemometric characterization of three varietal olive oils (Cvs. Cobrançosa, Madural and Verdeal Transmontana) extracted from olives with different maturation indices. Food Chem. 102, 406–414. Mishima, K., Tanaka, T., Pu, F., Egashira, N., Iwasaki, K., Hidaka, R., Matsunaga, K., Takata, J., Karube, Y., Fujiwara, M., 2003. Vitamin E isoforms α-tocotrienol and γ-tocopherol prevent cerebral infarction in mice. Neurosci. Lett. 337, 56–60.
589
Ollivier, D., Artaud, J., Pinatel, C., Durbec, J.-P., Guérère, M., 2006. Differentiation of French virgin olive oil RDOs by sensory characteristics, fatty acid and triacylglycerol compositions and chemometrics. Food Chem. 97, 382–393. Pereira, J.A., Casal, S., Bento, A., Oliveira, M.B.P.P., 2002. Influence of olive storage period on oil quality of three Portuguese cultivars of Olea europaea, Cobrançosa, Madural and Verdeal Transmontana. J. Agric. Food Chem. 50, 6335–6340. Velasco, J., Dobarganes, C., 2002. Oxidative stability of virgin olive oil. Eur. J. Lipid Sci. Technol. 104, 661–676. Wong, N.C., 2001. The beneficial effects of plant sterols on serum cholesterol. Can. J. Cardiol. 17, 715–721. Woollard, D.C., Indyk, H.E., 2003. Tocopherols. In Encyclopedia of Food Science and Nutrition. Academic Press, London, pp. 5789–5796.
This page intentionally left blank
Chapter 64
Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil Roberto Romero-González, Antonia Garrido Frenich and José Luis Martínez Vidal Research Group ‘Analytical Chemistry of Contaminants’, Department of Analytical Chemistry, Almeria University, Spain
64.1 INTRODUCTION 64.1.1 Presence of Sterols in Olive Oil Nowadays, olive oil is widely appreciated in Europe and around the world for its nutritional, health and sensory properties. Its constituents can be divided into the saponifiable and the unsaponifiable fractions. The first one represents between 98.5 and 99.5% of the oil, whereas the unsaponifiable constituents of olive oil account for 0.5–1.5% (Gül and Seker, 2006). Plant sterols or phytosterols are a major portion of these unsaponifiable components in olive oil, and they are a group of naturally occurring substances obtained from the isoprenoid biosynthetic pathway (Piironen et al., 2000), having a characteristic three-dimensional arrangement of four rings (Figure 64.1). They can be divided into three
A
HO
HO
HO
E
D
HO
C
B
HO
F
HO
FIGURE 64.1 Chemical structure of legislated sterols in olive oil: (A) sitosterol; (B) campesterol; (C) cholesterol; (D) brasicassterol; (E) stigmasterol; (F) Δ7-avenasterol. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
main classes: 4-desmethylsterols, 4-monomethylsterols and 4,4⬘-dimethylsterols (triterpene alcohols), 4-desmethylsterols being the most predominant in vegetable oils. Phytosterols can exist as free alcohols, esters with fatty acids, glycosides and acylated glycosides (Phillips et al., 2002), although in edible oils they are primarily in the free and sterified forms, free sterols dominating in olive oil (Thanh et al., 2006). Basically, the main sterols found in olive oil are β-sitosterol (more than 90 %), Δ5-avenasterol and stigmasterol, although other compounds such as cholesterol and sitostanol are also present (Piironen et al., 2000). It is well known that phytosterols may help reduce blood cholesterol levels through the inhibition of its absorption from the small intestine (Abidi, 2001), as well as having other beneficial effects (Piironen et al., 2000). Despite their health benefits, the determination of these compounds is of major interest because their composition can be used to detect adulteration or to check authenticity, since they can be considered as a ‘fingerprint’ for lipid materials. It is well established that the presence of large quantities of stigmasterol indicate an adulteration with lower-priced soybean and/or cottonseed oil (Kamm et al., 2001), and the finding of campesterol, Δ7-stigmastenol and Δ7-avenasterol was enough to detect the presence of hazelnut oil in olive oil (Mariani et al., 2006). On the other hand, sterol contents can be related to various parameters of the quality of olive oil. High levels of stigmasterol are correlated with high acidity and low organoleptic quality (Gutiérrez et al., 1999). Furthermore, it has been found that levels of sterols decrease with increasing ripeness index (Sánchez-Casas et al., 2004). In order to characterize the olive oil and to prevent adulterations, international organizations such as the European Union (EC, 1991), the International Olive Oil Council, and the Codex Alimentarius of the FAO/WHO have regulated
591
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
592
the content of sterols in different types of olive oil. Thus, an upper legal limit for the sum of the total sterol content as well as several lower and upper limits for each individual sterol have been established. For example, total sterol content should be ⱖ1000 mg kg⫺1 for virgin olive oil, and ⱖ2500 mg kg⫺1 for crude olive pomace oil.
64.1.2 Analytical Determination of Sterols in Olive Oil Official analytical procedures for the analysis of phytosterols in olive oil involve several steps. First, the oil is saponified to remove triglycerides and the unsaponifiable compounds are extracted with an organic solvent. Then, the unsaponifiable matter is fractionated into several classes of compounds by TLC or preparative HPLC, and their subsequent analysis by GC as trimethylsilyl derivatives (Parcerisa et al., 2000; Rui-Alves et al., 2005). However these methods are laborious and time-consuming, and may also involve the loss of analytes. In consequence, new approaches such as SPE (Amelio et al., 1992; AzadmardDamirchi and Dutta, 2006), SFE (Medvedovici et al., 1997), and on-line coupling reversed-phase LC-GC (Cortés et al., 2006) have been proposed in the last few years in order to reduce or automate the extraction process. For subsequent quantification and characterization of sterols, the compounds were separated by chromatographic techniques, including GC or LC, using several detectors such as FID (Giacometti, 2001), UV detection (Sánchez-Machado et al., 2004), and MS (Martínez-Vidal et al., 2007). Since GC methods require laborious sample derivatization steps before analysis, LC is slowly replacing GC analysis. Furthermore, MS detection is being used instead of conventional detectors because it provides more reliable identification and confirmation of these analytes. In consequence, LC coupled with MS analyzers has been found to be a very suitable tool for sterol analysis in different matrices (Nagy et al., 2006). This chapter provides an update of the methods using LC coupled with MS detection for analysis of sterols in olive oil, showing their advantages and shortcomings in relation to traditional methods; Table 64.1 lists some definitions related to this technique.
64.2 PRINCIPLES OF MS DETECTION MS is based on the production of ions, which are subsequently separated or filtered and detected according to their m/z (Niessen, 2006). In the last few years, there have been several ionization techniques for organic compounds. API has been mainly used because of its sensitivity and
SECTION | I Natural Components
ruggedness, and ESI and APCI are the most common currently applied sources (Niessen, 2006), which provide a soft ionization process with only few ions, commonly the molecular ion and three or four fragments. Whereas ESI is the ionization technique recommended for polar, ionized and high-molecular-weight compounds, APCI is used for less polar compounds. The separation of ions according to their m/z can be achieved in different ways, such as quadrupoles, ion trap and time of flight (Niessen, 2006). The quadrupole mass filter (Figure 64.2) is the most widely applied analyzer in LC-MS because of its ease of handling, small size, and relatively low cost (Ruibal-Mendieta et al., 2004). This analyzer consists of four parallel hyperbolic rods and at a given combination of DF voltage and RF potential, the trajectory of an ion of particular m/z is stable and is transmitted towards the detector. Ions with other m/z have unstable trajectories and do not pass through the mass filter. The quadrupole can be used in two modes: Scan or SIM. In Scan mode, a mass spectrum is obtained over a selected mass range. In SIM mode, DC and RF are set to observe only a specific mass, or a selection of specific masses. This provides the highest sensitivity for users interested in specific ions or fragments, since more time can be spent on each mass. Depending on the type of analyzer, multiple stages of mass analysis separation can be accomplished. Thus MS/MS can be carried out in time or in space using ion trap or triple quadrupole analyzers, respectively. The second one is the most frequently used in LC-MS and basically consists of three quadrupoles, using the second one as collision cell. An ion of interest is selected with the first quadrupole, collisionally activated with argon in the collision chamber (second quadrupole), and the fragmentation products are analyzed with the third quadrupole. Bearing in mind the excellent characteristics of MS detectors, the application areas of LC-MS are diverse, encompassing both qualitative and quantitative determinations of both high- and low-molecular-weight materials (Ardrey, 2003; Niessen, 2006), becoming an interesting alternative to conventional GC methods for the determination of sterols.
64.3 EXTRACTION AND ISOLATION OF STEROLS IN OLIVE OIL For the analysis of sterols in olive oil, three steps are followed: saponification, liquid–liquid partition and chromatographic separation. Saponification has been mainly used by official methods, followed by liquid–liquid partition with organic solvents (methanol, hexane, acetonitrile, etc.) and TLC. These steps are described in Annexes V and VI of Regulation EEC/2568/91 of the European Union
CHAPTER | 64 Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil
593
TABLE 64.1 Features of liquid chromatography-mass spectrometry. Concept
Definition
Liquid chromatography (LC)
Physical separation method in which the components to be separated are selectively distributed between two immiscible phases, one of which is stationary (stationary phase), while the other (mobile liquid phase) moves in a definite direction
Mass spectrometry (MS)
Analytical technique that measures the molecular masses of individual compounds and atoms precisely by converting them into charged ions
Tandem mass spectrometry (MS/MS)
Any general method involving at least two stages of mass analysis, either in conjunction with a dissociation process or a chemical reaction that causes a change in the mass or charge of an ion
m/z
An abbreviation used to denote the dimensionless quantity formed by dividing the mass of an ion by the number of charges carried by the ion
Ionization
A process which converts the neutral sample molecules into gas-phase ions
Atmospheric pressure ionization (API)
Ionization of the sample at atmospheric pressure and then transferring the ions into the mass spectrometer
Electrospray ionization (ESI)
Ionization procedure where an electrospray is created by applying a large potential between a metal inlet needle and a skimmer in an API source
Atmospheric pressure chemical ionization (APCI)
Ionization process based on the application of a high voltage to a corona pin that leads to an electrical breakdown of the atmosphere surrounding the pin formed by the vapor generated from the eluent of LC, nitrogen and the analyte molecules
Atmospheric pressure photoionization (APPI)
APPI is a relatively new ionization technique in which the ionization process is initiated by a beam of photons from a UV lamp (usually a krypton discharge lamp emiting photons at 10 eV)
Precursor ion
An electrically charged molecular moiety that may dissociate to form fragments, of which one or more may be electrically charged, and one or more are neutral species
Product ion
Electrically charged product of reaction of a particular precursor ion. In general, such ions have a direct relationship to a particular precursor ion and indeed may relate to a unique state of the precursor ion
Quadrupole analyzer
A mass filter that creates a quadrupole field with a DC component and an RF component in such a manner as to allow transmission only of ions having a selected mass-to-charge ratio
Full scan
Scanning of all ions present in a wide mass range
Selected ion monitoring (SIM)
Describes the operation of a mass spectrometer in which the ion currents at one (or several) selected m/z values are recorded, rather than the entire mass spectrum
Multiple reaction monitoring (MRM)
Monitoring the fragments of a selected precursor ion to a selected product ion
Commission (EC, 1991). Briefly, this method implies the saponification of an oil sample with ethanolic potassium hydroxide solution. Then, the unsaponifiable fraction is extracted with ethyl ether and separated by TLC on silica gel plates, obtaining four separated bands. Separation and quantification of the silylated sterol fraction is carried out by capillary column GC, using FID.
64.3.1 Alternative Extraction Procedures to the Official Method However, this extraction procedure is not suitable for the analysis of some compounds such as sterol esters, since they are altered during this process. Likewise, this requires substantial manual operations, and the methodology is tedious,
594
SECTION | I Natural Components
Detector Non-resonant ion Resonant ion
Quadrupole
− +
−
+
Ion source
Z DC and RF voltages
FIGURE 64.2 Linear quadrupole mass spectrometer.
so several approaches have been described to replace the official methodology as can be observed in Table 64.2. For instance, transesterification provides a good alternative to the saponification step, being an easy procedure that avoids the tedious extraction of unsaponifiable material, and it is more environmentally friendly (Cunha et al., 2006). On the other hand, SPE has been proposed for the extraction of both free and sterified sterols (Ruiz-Gutiérrez and Pérez-Camino, 2000; Phillips et al., 2002; Cercaci et al., 2003; AzadmardDamirchi and Dutta, 2006). SPE technique has been used for the clean-up of the unsaponifiable fraction, although it has also been used without saponification to isolate free sterols from triacylglycerols (Ballesteros et al., 1995). Adsorbents like silica gel, alumina or C18 (Phillips et al., 2002; Cunha et al., 2006) have been tested, and they seem to be efficient and reproducible for free and esterified sterol separation, replacing solvent extraction (Toivo et al., 1998). Selecting different solvents, such as hexane and dihexyl ether, several fractions containing free and esterified sterols can be collected (Ruiz-Gutiérrez and PérezCamino, 2000). Sometimes, prior to SPE the free hydroxyl groups of sterols could be silylated to reduce their polarity (Lechner et al., 1999), finding more favorable conditions to the extraction process. Another extraction technique, which can be a good alternative to conventional techniques, is SFE, due to several advantages such as lack of toxicity, chemical inertness, low cost and ready availability. However, studies about its application for the determination of sterols in olive oil are very scarce (Medvedovici et al., 1997; Hurtado-Benavides et al., 2004) despite the fractionation of the sterols from oil matrix being obtained in less than 8 minutes, and the injection and collection of the sterol fraction is fully automated and time controlled. Preparative-LC has also been used to separate sterol fractions of vegetable oils (Li et al., 2000). However, the
use of LC may require a high solvent volume and also higher costs. Saponification can be avoided using on-line coupling of reversed-phase LC-GC (Señoráns et al., 1996; Villén et al., 1998). This technique also replaces the subsequent cleanup steps, because it removes triglycerides and other interfering compounds in an effective way, combining in one process sample preparation and chromatographic determination, allowing high separation, efficiency and sensitivity. One problem of this technique is that if normal phase is used, it may be deactivated by triacylglycerols, and a backflush is needed, so on-line reversed phase has been proposed using a PTV injector (Cortés et al., 2006). These new procedures considerably reduce the extraction time. For instance, the TLC method requires 2–3 hours per sample, whereas a conventional SPE method only takes 30–45 minutes to pass through the columns, remove the solvent and be ready for chromatographic analysis.
64.4 ANALYSIS OF STEROLS BY LC Traditionally, the separation and determination of sterols in olive oil has been carried out by GC, using FID or MS as detectors (López-López et al., 2008). When GC is used, sterols should be analyzed as trimethylsilyl or methyl derivatives in order to improve peak shape, resolution and sensitivity (Figure 64.3). The progress of the LC technology in recent years is provoking LC to replace GC in order to avoid this time-consuming step. When LC is used, it is possible to inject the unsaponifiable fraction from olive oil directly into the LC (Nagy et al., 2005), although an extraction and/or clean-up step is generally necessary.
64.4.1 Chromatographic Conditions The analysis of sterols in olive oil can be carried out by normal and reversed-phase LC techniques but the use of silica columns (normal phase) has several shortcomings, such as long equilibration times and the employment of hazardous volatile organic solvents (Careri et al., 2001). As can be observed in Table 64.3, current LC methods use C18 modified silica gel columns (Abidi, 2001), although less hydrophobic stationary phases such as hexylphenyl-modified silica gel have been proposed (Mezine et al., 2003), using a gradient of acetonitrile (90–100%) in water, improving the selectivity obtained by C18. When C18 is used, sterols can be eluted by the application of nonaqueous mobile phases, usually consisting of acetonitrile and stronger organic eluents such as propionitrile and 2-propanol (Mezine et al., 2003). Furthermore a threecomponent non-aqueous mobile phase can be used (Ferrari et al., 1997) or a combination of acetonitrile and an aqueous solution of acetic acid has also been applied (CañabateDíaz et al., 2007).
595
CHAPTER | 64 Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil
TABLE 64.2 Selected chromatographic methods for the analysis of sterols in olive oils. Sample treatment
Observations
Detection
Reference
SPE (quaternary amine) & LC
Elution with hexane:diethylether
GC-FID
Amelio et al., 1992
On-line SPE (Silica)
Elution with n-hexane Automatic extraction method
GC-FID
Ballesteros et al., 1995
SFE
Fully automated method
GC-MS
Medvedovici et al., 1997
Saponification & SPE (C18)
Elution with 5% MeOH in CHCl3
GC-FID
Toivo et al., 1998
Preparative LC coupled on-line with GC
PTV as interface
GC-FID
Villén et al., 1998
Saponification without TLC
TMS derivatives
GC-FID/MS
Parcerisa et al., 2000
SPE (Silica)
Elution with hexane and ethyl acetate MS is used for confirmation
LC-UV
Careri et al., 2001
Saponification without TLC
TMS derivatives
GC-FID
Giacometti, 2001
SPE (alumina) & saponification
Elution with diethyl ether/hexane MS is used for confirmation
GC-FID
Phillips et al., 2002
SPE (Silica)
Elution with hexane-diethyl ether MS is used for confirmation
GC-FID
Cercaci et al., 2003
Saponification
MS is used for confirmation
LC-UV
Sánchez-Casas et al., 2004
Dilution to 0.01% in methanol
Characterization study
GC-MS
Nagy et al., 2005
Saponification & TLC
Characterization study
GC-FID
Rui-Alves et al., 2005
SPE (Silica)
Elution with hexane-diethylether
GC-MS
Azadmard-Damirchi and Dutta, 2006
On-line RPLC
Automated extraction process
GC-FID
Cortés et al., 2006
SPE (Silica)
Elution with n-hexane/ethyl acetate/ethanol and diethylether Transesterification instead of saponification
GC-MS
Cunha et al., 2006
Saponification
Study of genotipic
GC-FID
Gül and Seker 2006
SPE (Silica)
Elution with hexane-diethylether
GC-FID
Mariani et al., 2006
Saponification
TMS derivatives
GC-MS
Thanh et al., 2006
Saponification & TLC
European official method
LC-MS
Cañabate-Díaz et al., 2007
Saponification
TLC was not necessary
LC-MS
Martínez-Vidal et al., 2007
Saponification & TLC
European official method
GC-FID
López-López et al., 2008
Related to particle size, in the last few years, packing lower than ⬍2 μm has been introduced to improve resolution in LC separations, reducing analysis time and increasing sample throughput, and it has been applied for the determination of sterols in food materials (Lu et al., 2007).
64.4.2 MS Analysis: Ionization Methods LC combined with UV detection (Careri et al., 2001; Sánchez-Machado et al., 2004) has been described for the analysis of sterols in olive oils because it is an accessible
596
SECTION | I Natural Components
Cholesterol 90 000
β-Sitosterol
80 000
Abundance
70 000 60 000
6
Δ -avenasterol
50 000
7
Δ - stigmasterol
40 000 campesterol stigmasterol
30 000 20 000
b-Sitosterol
cholesterol
10 000
4.00
6.00
8.0
10.00 12.0
14.0
16.0 18.0 Time (min)
20.0
22.0
24.0 26.00 28.0
FIGURE 64.3 Representative chromatograms of the determination of sterols by GC-MS of an olive oil, together with individual chromatograms in selective ion monitoring mode. Reprinted from Cunha et al. (2006), copyright 2006, with permission from Elsevier.
TABLE 64.3 Chromatographic conditions for the determination of sterols by LC-MS. Column
Mobile phase
Type of matrix
Reference
C8 narrow bore column
Acetonitrile/water
Oils
Careri et al., 2001
Luna hexyl-phenyl
Acetonitrile/water
Beverages and butter
Mezine et al., 2003
C18 Prevail column
Methanol/water with 1% acetonitrile
Cereals
Rozenberg et al., 2003
Alltech C18 Prevail
Methanol/water (1% acetonitrile)
Spelt and winter wheat
Ruibal-Mendieta et al., 2004
Kromasil 100 C18
Methanol/acetonitrile
Edible seaweeds
Sánchez-Machado et al., 2004
Purospher Star RP-18-e
Methanol/water (0.2% acetic acid)
Olive oils
Nagy et al., 2005
Xterra MS C8
Methanol/water (0.2 mM ammonium acetate)
Cultured Caco-2 cells
Palmgrén et al., 2005
Purospher Star RP-18e
Methanol/water/n-hexane
Plasma
Nagy et al., 2006
Atlantis dC18
Acetonitrile/water (0.01% acetic acid)
Olive oils
Cañabate-Díaz et al., 2007
Atlantis dC18
Acetonitrile/water (0.01% acetic acid)
Olive oils
Martínez-Vidal et al., 2007
Acquity BEH C18
Methanol/water with 1% acetonitrile
Food material
Lu et al., 2007
Chromolith SpeedRod RP-18e
Methanol/water/isoproponol
Serum
Lembcke et al., 2008
and low-cost technique (Figure 64.4A). However, due to the lack of selectivity, LC is being coupled with MS detection to provide a more reliable identification and confirmation of these analytes than conventional detectors, as well as solving the problems associated with coelution of sterols and neutral lipids (Figure 64.4B). Furthermore, when LC is coupled with MS, it is possible to provide structural
elucidation and confirmation of sterols, although few applications of LC-MS techniques in sterol analysis on olive oils have been found (Careri et al., 2001, Sánchez-Machado et al., 2004, Nagy et al., 2006; Cañabate-Díaz et al., 2007, Martínez-Vidal et al., 2007). Sterols are lipophilic molecules and they are poorly ionized by ESI, so several alternatives have been proposed,
CHAPTER | 64 Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil
597
Nevertheless, the highest sensitivity is obtained when APPI is used (Lembcke et al., 2008). Besides, APPI provides more precise measurements than those with APCI using identical chromatographic conditions, because the ionization efficiency is higher for APPI. However, APCI is still most widely used for sterol analysis, as can be observed in Table 64.4, because they are commonly available and can be easily coupled with the LC system.
V
1.2
1.0
0.8 3
64.4.3 MS Analysis: Ion Separation and Identification
0.4 2 1
0.2
A
2
4
6
8
10
12
14
min
B FIGURE 64.4 Representative chromatograms of the analysis of sterols in oils by (A) LC-UV: 6-ketocholestero (1), sitgmasterol (2), β-sitosterol (3), reprinted from Careri et al. (2001), copyright 2001, with permission from Elsevier, and (B) LC-APCI-MS extracted ion chromatograms of oil extracts: erythrodiol and uvaol m/z 425.30 (A), cholesterol m/z 369.20 and fucosterol m/z 395.30 (B), stigmasterol m/z 395.30 (C), β-sitosterol m/z 397.30 (D), sitostanol m/z 397.30 (E). Reprinted from Cañabate-Díaz et al. (2007), copyright 2006, with permission from Elsevier.
such as the derivatization with electrophoric labels like pentafluorobenzyl chloride (Trösken et al., 2004), but this step increases analysis time. Furthermore, ammonium acetate was added to the mobile phase and the acidification with acetic or formic acid has been tested (Rozenberg et al., 2003), but no significant improvement was observed, due to sterols being neutral compounds. Better results were obtained when APCI was used instead of ESI, which allows the ionization of lipophilic compounds (CañabateDíaz et al., 2007), as can be observed in Figure 64.5.
Independent of the ionization process (APCI or APPI), protonated sterol molecules [M ⫹ H]⫹ produce weak signal intensities, due to the loss of water [M ⫹ H–H2O]⫹, the most intense ion, which is usually used for quantification purposes. Other minor ions corresponding to the dehydrogenation of the molecule, such as [M ⫹ H ⫺ 2 H]⫹ can also be observed (Rozenberg et al., 2003), showing typical ions for sterols (Table 64.5). Basically, the normal mechanism is protonation at a basic site of the molecule, and in the case of sterols, the secondary alcohol is lost by dehydration, indicating that the facility of dehydration appears to be dependent on the ring conformation and hence, the proximity of adjacent hydrogen atoms. For the determination of sterols by LC-MS, single (Cañabate-Díaz et al., 2007) or triple quadrupoles have been mainly used (Nagy et al., 2005; Lu et al., 2007), although other analyzers such as ion trap can be applied (Palmgrén et al., 2005). Despite triple quadrupoles allowing the selective fragmentation of the precursor ion, improving the selectivity and the sensitivity of the study, it is difficult to distinguish different sterols using MS/MS, so most of the papers published so far use SIM or full scan mode (Table 64.4). However, in a recent work (Lembcke et al., 2008), sterols have been analyzed in serum using MRM, selecting the transitions indicated in Table 64.5. Furthermore, sterol esters can be analyzed using the same transitions because of ester bond cleavage during the ionization process. Bearing in mind that the use of MRM can enhance signal-to-noise ratio, increasing the selectivity and sensitivity, the sample extraction procedure can be reduced, diminishing the time required for this step. In LC-MS analysis, a complete resolution for all the compounds is not always necessary due to high specificity and selectivity of the detection method. However for compounds with the same molecular weight (isobaric compounds), effective chromatographic separation should be attained in order to decrease the interference of these type of compounds. For instance, stigmasterol and Δ5-avenasterol have the same molecular weight (412.7 g mol⫺1) and their mass spectra are practically the same, so they can interfere with each other in the case of coelution.
598
SECTION | I Natural Components
A
B
APCI (+) 369.3
100
ESI (+)
100
8.49x106
1.06x107
%
%
158.0 256.4 157.9 367.3 0
100
200
C
300 m/z
387.7 400
369.7 500
100
200
300
400
500
m/z D
APCI (+) 397.3
100
0
100
1.00x107
ESI (+) 104.9 1.82x107
158.0
%
%
413.2
0
100
200
300 m/z
400
500
0
159.1
100
200
300
400
500
m/z
FIGURE 64.5 Full scan ESI and APCI mass spectra of cholesterol (A and B) and β-sitosterol (C and D), using the conditions indicated by CañabateDíaz et al. (2007).
64.4.4 Matrix Effect When ESI or APCI are used, one of the main problems is the presence of matrix components which can affect the ionization of the analytes, reducing or enhancing the analytical signal (matrix effect). However, when APPI is used as the ionization source, no matrix effect is observed. Because olive oils are not standard reference materials and no blank oil samples are available, several approaches, such as internal standard or standard addition, have been proposed in order to overcome this problem. The best way to compensate the matrix effect is the use of isotope dilution. This approach allows the correction of signal suppression, as both labeled and native compounds will suffer the same suppression effect. This method is frequently used when only one or few analytes must be determined. For instance, there are few available standards for sterols, such as [2H7]-labeled cholesterol that has been used as internal standard for quantification (Careri et al., 2001; Lembcke et al., 2008), improving accuracy of the measurement.
However, for most of the sterols, these compounds are not available, so the addition of an internal standard, which is not present in the sample, has been indicated. For example, for the determination of sterols in food, 6-ketocholestanol has been proposed (Lu et al., 2007). Another alternative is the use of standard addition methodology. When this procedure is used, the analytical calibration is performed spiking the real sample at several concentrations. Nevertheless, this method is not very convenient when a high number of samples must be analyzed. Finally, external calibration seems to be the least suitable method and it can be used only if extensive clean-up procedures are applied to obtain sufficiently clean extracts (Careri et al., 2001).
64.5 CONCLUSIONS AND FUTURE TRENDS Reliable routine analysis of sterols in olive oils is still a difficult and lengthy procedure. Faster and more reliable
CHAPTER | 64 Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil
599
TABLE 64.4 Ionization conditions for the determination of sterols by LC-MS. Ionization source
Type of analyzer
Acquisition mode
Reference
APCI
Triple quadrupole
Full scan
Careri et al., 2001
APCI
Single quadrupole
Full scan
Mezine et al., 2003
APCI
Single quadrupole
SIM mode
Rozenberg et al., 2003
APCI
Triple quadrupole
SIM mode
Ruibal-Mendieta et al., 2004
APCI
Single quadrupole
SIM mode
Sánchez-Machado et al. (2004)
APCI
Triple quadrupole
SIM mode
Nagy et al., 2005
APCI
Ion trap
Full scan
Palmgrén et al., 2005
APCI
Triple quadrupole
SIM mode
Nagy et al., 2006
APCI
Triple quadrupole
SIM mode
Lu et al., 2007
APCI
Single quadrupole
SIM mode
Cañabate-Díaz et al., 2007
APCI
Single quadrupole
SIM mode
Martínez-Vidal et al., 2007
APPI
Triple quadrupole
MRM mode
Lembcke et al., 2008
TABLE 64.5 Molecular weight and principal ions of sterols during MS ionization. Sterol
Molecular weight (g mol⫺1)
[M ⫹ H ⫺ H20]⫹
[M ⫹ H-2 H]⫹a
MRM transitionsb
β-sitosterol
414.7
397
413
397.4 ⬎ 257.3
Brassicasterol
398.7
Campestanol
402.7
385
401
Campesterol
400.7
383
399
Cholesterol
386.7
369
385
Δ5-avenasterol
412.7
395
Δ7-avenasterol
412.7
395
Stigmasterol
412.7
395
Sitostanol
416.7
399
a
381.4 ⬎ 297.3
383.4 ⬎ 161.3
395.4 ⬎ 297.3 415
Obtained from Rozenberg et al. (2003) and Lu et al. (2007). Obtained from Lembcke et al. (2008).
b
methods must be selected only if they provide results that are not significantly different from those obtained using official methods. One of the weak points of the current analysis of sterols in olive oils is the sample treatment. Novel treatment
techniques prior to MS analysis could be particularly valuable tools in the isolation and purification of sterols. Although promising progress has been achieved, more work is mandatory in order to simplify this step. In this sense, new sample extraction procedures, such as SPME, which
600
SECTION | I Natural Components
could be a valuable technique, have been used for the determination of sterols in biological matrices (Domeño et al., 2005) although not for use in edible oils yet. Furthermore, SFE is a good technique for the extraction of compounds such as sterols, but so far few works have been published, so more work should be developed on this topic. The use of mass spectrometric detectors has improved the reliability of the detection of sterols. However, sterols are difficult to ionize with conventional sources such as ESI or APCI, so APPI can be a very powerful alternative, since the ionization efficiency is higher, providing better sensitivity. An interesting point can be the automation of the entire system, based on benchtop instrumentation, including online sampling treatment and LC-MS system, favoring the use of LC-MS for routine analysis.
SUMMARY POINTS ●
●
●
●
●
There is great concern about the determination of sterols in olive oil. Official methods for the determination of sterols in olive oil are tedious and time-consuming, so new alternatives are proposed. LC-MS methods are fast and provide a reliable identification and determination of sterols in olive oil. For the analysis of sterols, several ionization sources and strategies can be followed, obtaining the best results when atmospheric pressure chemical ionization and selected ion monitoring are applied. More work related to the automation of the extraction and determination of sterols should be carried out.
REFERENCES Abidi, S.L., 2001. Chromatographic analysis of plant sterols in foods and vegetable oils. J. Chromatogr. A 935, 173–201. Amelio, M., Rizzo, R., Varazni, F., 1992. Determination of sterols, erythrodiol, uvaol and alkanols in olive oils using combined solidphase extraction, high performance liquid chromatographic and high resolution gas chromatographic techniques. J. Chromatogr. A 606, 179–185. Ardrey, R.E., 2003. Liquid Chromatography-Mass Spectrometry: An Introduction. John-Wiley and Sons Ltd., Chichester, pp. 98–126. Azadmard-Damirchi, S., Dutta, P.C., 2006. Novel solid-phase extraction method to separate 4-desmethyl-, 4-monomethyl-, and 4,4⬘-dimethylsterols in vegetable oils. J. Chromatogr. A 1108, 183–187. Ballesteros, E., Gallego, M., Valcárcel, M., 1995. Simultaneous determination of sterols in edible oils by use of a continuous separation module coupled to a gas chromatograph. Anal. Chim. Acta 308, 253–260. Cañabate-Díaz, B., Segura-Carretero, A., Fernández-Gutiérrez, A., Belmonte-Vega, A., Garrido-Frenich, A., Martínez-Vidal, J.L., DuránMartos, J., 2007. Separation and determination of sterols in olive oil by HPLC-MS. Food Chem. 102, 593–598.
Careri, M., Elviri, L., Mangia, A., 2001. Liquid chromatography-UV determination and liquid chromatography-atmospheric pressure chemical ionization mass spectrometric characterization of sitosterol and stigmasterol in soybean oil. J. Chromatogr. A 935, 249–257. Cercaci, L., Rodríguez-Estrada, M.T., Lercker, G., 2003. Solid-phase extraction-thin-layer chromatography-gas-chromatography method for the detection of hazelnut oil in olive oils by determination of esterified sterols. J. Chromatogr. A 985, 211–220. Cortés, J.M., Sánchez, R., Villén, J., Vázquez, A., 2006. Analysis of unsaponifiable compounds of edible oils by automated on-line coupling reversed-phase liquid chromatography-gas chromatography using the through oven transfer adsorption desorption interface. J. Agric. Food Chem. 54, 6963–6968. Cunha, S.S., Fernandes, J.O., Oliveira, M.B.P.P., 2006. Quantification of free and esterified sterols in Portuguese olive oils by solid-phase extraction and gas chromatography-mass spectrometry. J. Chromatogr. A 1128, 220–227. Domeño, C., Ruiz, B., Nerín, C., 2005. Determination of sterols in biological samples by SPME with on-fiber derivatization and GC/FID. Anal. Bional. Chem. 381, 1576–1583. EC. 1991. Commission Regulation EC 2568/91, July 1991, on the Characteristic methods of olive oils and their analytical methods. EC Journal L 8121.10.91. Ferrari, R.A., Esteves, W., Mukherjee, K.D., Schulte, E., 1997. Alteration of sterols and steryl esters in vegetable oils during industrial refining. J. Agric. Food Chem. 45, 4753–4757. Giacometti, J., 2001. Determination of aliphatic alcohols, squalene, α-tocopherol and sterols in olive oils: direct method involving gas chromatography of the unsaponifiable fraction following silylation. Analyst 126, 472–475. Gül, M.K., Seker, M., 2006. Comparative analysis of phytosterol components from rapeseed (Brassica napus L.) and olive (Olea europaea L.) varieties. Eur. J. Lipid Sci. Technol. 108, 759–765. Gutiérrez, F., Jiménez, B., Ruiz, A., Albi, M.A., 1999. Effect of olive ripeness on the oxidative stability of virgin olive oil extracted from the varieties Picual and Hojiblanca and on the different components involved. J. Agric. Food Chem. 47, 121–127. Hurtado-Benavides, A.M., Señoráns, F.J., Ibáñez, E., Reglero, G., 2004. Countercurrent packed column supercritical CO2 extraction of olive oil. Mass transfer evaluation. J. Supercrit. Fluids 28, 29–35. Kamm, W., Dionisi, F., Hischenhuber, C., Engel, K., 2001. Authenticity assessment of fats and oils. Food Rev. Int. 71, 249–290. Lechner, M., Reiter, B., Lorbeer, E., 1999. Determination of tocopherols and sterols in vegetable oils by solid-phase extraction and subsequent capillary gas chromatography analysis. J. Chromatogr. A 857, 231–238. Lembcke, J., Ceglarek, U., Fiedler, G.M., Baumann, S., Leichtle, A., Thiery, J., 2008. Rapid quantification of free and esterified phytosterols in human serum using APPI-LC-MS/MS. J. Lipid Res. 46, 21–26. Li, J., Ho, C., Li, H., Tao, H., Tao, L., 2000. Separation of sterols and triterpene alcohols from unsaponifiable fractions of three plant seed oils. J. Food Lipids 7, 11–20. López-López, A., Montaño, A., Ruíz-Méndez, M.V., Garrido-Fernández, A., 2008. Sterols, fatty alcohols, and triterpenic alcohols in commercial table olives. J. Am. Oil Chem. Soc. 85, 253–263. Lu, B., Zhang, Y., Wu, X., Shi, J., 2007. Separation and determination of diversiform phytosterols in food materials using supercritical carbon dioxide extraction and ultraperformance liquid chromatographyatmospheric pressure chemical ionization-mass spectrometry. Anal. Chim. Acta 588, 50–63.
CHAPTER | 64 Liquid Chromatography-Mass Spectrometry Determination of Sterols in Olive Oil
Mariani, C., Bellan, G., Lestini, E., Aparicio, R., 2006. The detection of the presence of hazelnut oil in olive oil by free and esterified sterols. Eur. Food Res. Technol. 223, 655–661. Martínez-Vidal, J.L., Garrido-Frenich, A., Escobar-García, M.A., Romero-González, R., 2007. LC-MS determination of sterols in olive oil. Chromatographia 65, 695–699. Medvedovici, A., David, F., Sandra, P., 1997. Analysis of sterols in vegetable oils using off-line SFC/capillary GC-MS. Chromatographia 44, 37–42. Mezine, I., Zhang, H., Macku, C., Lijana, R., 2003. Analysis of plant sterol and stanol esters in cholesterol-lowering spreads and beverages using high-performance liquid chromatography-atmospheric pressure chemical ionization-mass spectroscopy. J. Agric. Food Chem. 51, 5639–5646. Nagy, K., Bongiorno, D., Avellone, G., Agozzino, P., Ceraulo, L., Vékey, K., 2005. High performance liquid chromatography-mass spectrometry based chemometric characterization of olive oils. J. Chromatogr. A 1078, 90–97. Nagy, K., Jakab, A., Pollreisz, F., Bongiorno, D., Ceraulo, L., Averna, M.R., Noto, D., Vékey, K., 2006. Analysis of sterols by high-performance liquid chromatography/mass spectrometry combined with chemometrics. Rapid Commun. Mass Spectrom. 20, 2433–2440. Niessen, W.M.A., 2006. Liquid Chromatography-Mass Spectrometry. Taylor and Francis Group, Boca Raton, pp. 23–50 and pp. 359–380. Palmgrén, J.J., Töyräs, A., Mauriala, T., Mönkkönen, J., Auriola, S., 2005. Quantitative determination of cholesterol, sitosterol, and sitostanol in cultured Caco-2 cells by liquid chromatography-atmospheric pressure chemical ionization mass spectrometry. J. Chromatogr. B 821, 144–152. Parcerisa, J., Casals, I., Boatella, J., Codony, R., Rafecas, M., 2000. Analysis of olive and hazelnut oil mixtures by high-performance liquid chromatography-atmospheric pressure chemical ionisation mass spectrometry of non-saponifiable compounds (tocopherols and sterols). J. Chromatogr. A 881, 148–149. Phillips, K.M., Ruggio, D.M., Toivo, J.I., Swank, M.A., Simpkins, A.H., 2002. Free and esterified sterol composition of edible oils and fats. J. Food Compos. Anal. 15, 123–142. Piironen, V., Lindsay, D.G., Miettinen, T.A., Toivo, J., Lampi, A.M., 2000. Plant sterols: biosynthesis, biological function and their importance to human nutrition. J. Sci. Food Agric. 80, 939–966. Rozenberg, R., Ruibal-Mendieta, N.L., Petitjean, G., Cani, P., Delacroix, D.L., Delzenne, N.M., Meurens, M., Queting-Leclerq, J.Q., HabibJiwan, H., 2003. Phytosterol analysis and characterization in spelt
601
(Triticum aestivum ssp. spelta L.) and wheat (T. aestivum L.) lipids by LC/APCI-MS. J. Cereal Sci. 38, 189–197. Rui-Alves, M., Cunha, S.C., Amaral, J.S., Pereira, J.A., Oliveira, M.B., 2005. Classification of PDO olive oils on the basis of their sterol composition by multivariate analysis. Anal. Chim. Acta 549, 166–178. Ruibal-Mendieta, N.L., Rozenberg, R., Delacroix, D.L., Petitjean, G., Dekeyser, A., Baccelli, C., Marques, C., Delzenne, N.M., Meurens, M., Habib-Jiwan, J.L., Quetin-Leclerq, J.Q., 2004. Spelt (Triticum spelta L.) and Winter Wheat (Triticum aestivum L.) wholemeal have similar sterol profiles, as determined by quantitative liquid chromatography and mass spectrometry analysis. J. Agric. Food Chem. 52, 4802–4807. Ruiz-Gutiérrez, V., Pérez-Camino, M.C., 2000. Update on solid-phase extraction for the analysis of lipid classes and related compounds. J. Chromatogr. A 885, 321–341. Sánchez-Casas, J., Osorio-Bueno, E., Montaño-García, A.M., MartínezCano, M., 2004. Sterol and erythrodiol ⫹ uvaol content of virgin olive oils from cultivars of Extremadura (Spain). Food Chem. 87, 225–230. Sánchez-Machado, D.I., López-Hernández, J.L., Paseiro-Losada, P., López-Cervantes, J., 2004. An HPLC method for the quantification of sterols in edible seaweeds. Biomed. Chromatogr. 18, 183–190. Señoráns, F.J., Tabera, J., Herraiz, M., 1996. Rapid separation of free sterols in edible oils by on-line coupled reversed phase liquid chromatography-gas chromatography. J. Agric. Food Chem. 44, 3189–3192. Thanh, T.T., Vergnes, M.F., Kaloustian, J., El-Moselhy, T.F., AmiotCarlin, M.J., Portugal, H., 2006. Effect of storage and heating on phytosterol concentrations in vegetable oils determined by GC/MS. J. Sci. Food Agric. 86, 220–225. Toivo, J., Piironen, V., Kalo, P., Varo, P., 1998. Gas chromatographic determination of major sterols in edible oils and fats using solid phase extraction in sample preparation. Chromatographia 48, 745–750. Trösken, E.R., Straube, E., Lutz, W.K., Völkel, W., Patten, C., 2004. Quantitation of lanosterol and its major metabolite FF-MAS in an inhibition assay of CYP51 by azoles with atmospheric pressure photoionization based LC-MS/MS. J. Am. Soc. Mass Spectrom. 15, 1216–1221. Villén, J., Blanch, G.P., Ruiz del Castillo, M.L., Herraiz, M., 1998. Rapid and simultaneous analysis of free sterols, tocopherols, and squalene in edible oils by coupled reversed-phase liquid chromatography-gas chromatography. J. Agric. Food Chem. 46, 1419–1422.
This page intentionally left blank
Chapter 65
13
C Nuclear Magnetic Resonance Spectroscopy as a New Quantitative Method for Determining Fatty Acid Positional Distribution in Olive Oil Triacylglycerols: Applications to Olive Oil Authenticity Giovanna Vlahov CRA – OLI Centro per l’Olivicoltura e l’Industria Olearia, Sede Scientifica Città S. Angelo, Angelo (PE), Italy
65.1 INTRODUCTION Triacylglycerols are triesters of glycerol alcohol esterified with three long-chain carboxylic acids. The stereospecific numbering system (sn), which is used to describe without ambiguity the asymmetrical glycerides, states that C-1 of glycerol is the one that appears on top in the Fisher projection of L-glycerol ester showing a vertical carbon chain with the secondary hydroxyl group to the left. Triglyceride chirality is determined by introducing different fatty acids at the two primary hydroxyl groups of the glycerol molecule which by itself is achiral due to the plane of symmetry at C-2 (Gunstone, 1967). The aim of a regiospecific analysis of triacylglycerols is to determine the fatty acid composition of the sn-1(3)and sn-2-positions where the sn-1- and sn-3-positions are not differentiated. The chromatographic resolution of regioisomers of glycerolipids is based on differences in polarity between the primary (X-1,2-isomer) and the less polar secondary (X-1,3-isomer) alcohols (Kuksis, 1996). The stereospecific analysis of triacylglycerols determines how the fatty acids of triacylglycerols are distributed over the three different positions of glycerol. Stereoanalysis can involve either the resolution by HPLC of triacylglycerol diastereoisomers derived from enantiomers by reaction with a chiral reagent, or the resolution of triacylglycerol enantiomers on a chiral stationary phase (Kuksis, 1996). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Stereoanalyses of triacylglycerols proved to be useful to ascertain the accuracy of theories of fatty acid distribution. Theory of 1,3-random-2-random distribution has now been accepted. The theory states that glycerol is fully esterified at 2-positions with a mixture of fatty acids, and with another mixture of fatty acids at 1- and 3-positions, these positions assumed to be identical. However, 13C nuclear magnetic resonance spectroscopy is the emerging analytical technique for determining the positional distribution of fatty acids among the 1,3- and 2-positions of triacylglycerols where any chemical reaction is eliminated.
65.2 HIGH-RESOLUTION 13C NMR OF OLIVE OIL TRIACYLGLYCEROLS 13
C NMR spectroscopy of an olive oil sample registers the resonances of carbon-13 nuclei of the fatty acid and glycerol moieties of triglycerides which represent not less than 98% of olive oil. Resonance assignments of the whole 13C NMR spectrum (10–174 ppm) of an olive oil sample are reported in Table 65.1 along with the chemical shifts of the corresponding 1H resonances. 13 C resonances are grouped in four sets of signals in correspondence of different functional groups, namely carboxy (172.6–174.3 ppm), double bonds (126–132 ppm),
603
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
604
SECTION | I Natural Components
glycerol backbone (60–72 ppm), methylene and methyl (10–35 ppm) groups. It is worth stressing that 13C NMR spectroscopy assures all the advantages of working with triglyceride resonances spread over a 180 ppm frequency range as compared to a 6 ppm range of 1H NMR spectroscopy. The wide gap in frequency resolution between 13C and 1H nuclei is emphasized in Table 65.1. Wollemberg evidenced that carboxy and olefinic carbon resonances of acyl chains enable the determination of triacylglycerol structures (Wollemberg, 1990). Carboxy carbon region of 13C spectrum of an olive oil sample, in the frequency range from 172.6 to 174.3 ppm,
detects saturated (C16–22 n ⫽ 0, 173.063 ppm), vaccenate (C18:1 11cis, 173.052 ppm), oleate (C18:1 9cis, 173.034 ppm) and linoleate (C18:2 9,12cis, 173.025 ppm) chains at 1,3-positions and oleate (172.635 ppm) and linoleate (172.625 ppm) chains at 2-positions of glycerol, where the 1,3-position chains resonate at a higher frequency than the 2-position chains by 0.40 ppm (Howarth et al., 1995). However, olefinic carbon resonances spreading over a wider range of frequency (from 126 to 132 ppm) provide a higher shift resolution. As a consequence, olefinic carbons are the signals of choice to determine the full pattern of triglyceride acyl chains (with the exception of saturated
TABLE 65.1 Assignments of 1H and 13C resonances of olive oil triacylglycerols. 1
H
13
1
C
H
13
C
Functional groups
Chemical shifts (ppm)
Chemical shifts (ppm)
Functional groups
Chemical shifts (ppm)
Chemical shifts (ppm)
COOR
–
173.063 S1 α 173.052 V1 α 173.034 O1α 173.025 L1 α 172.635 O1 β 172.625 L1 β
CH2 ⫺ CH ⫽ CH
2.03 (m)
CH2CH2COOR
1.63 (m)
27.131 O11 α,β 27.112 L14 α,β 27.097 L8 α,β 27.079 O8 α,β 24.791 O,L3 β 24.776 S3 α 24.754 O,L3 α
(CH2)n
1.28
⫺CH ⫽ CH⫺
5.344 (m)
130.045 L13 β 130.077 L13 α 129.879 O10 β 129.864 O10 α 129.837 L9 α 129.812 L9 β 129.782 V12 α,β 129.688 V11 α,β 129.572 O9 α 129.545 O9 β 127.981 L10 β 127.963 L10 α 127.801 L12 α 127.789 L12 β
29.685 O12 α,β 29.586 S7 α 29.452 O14 α,β 29.398 S6 α 29.246 O13,15 α,β 29.192 S5 α 29.111 O,L5 β 29.089 O,L5 α 29.000 O,L4 α 29.962 O,L4 β 31.852 S16 α 31.832 O16 α,β 31.709 V16 α,β 31.444 L16 α,β
CHOH Gl β
5.262 (m)
68.818
22.609 S17α
CH2OH Gl α, α‘
4.300 (dd, J⫽ 4.5, 12 Hz)/4.143 (dd, J⫽ 6, 12 Hz).
61.985
22.599 O17α,β
CH ⫽ CHCH2-CH ⫽ CH
2.77 (m)
25.541 L11 α,β
22.491 L17 α,β
CH2–COOR
2.32
34.078 O,L 2 β 33.936 S2 33.914 O,L2 α
14.008 S18α 14.0003 O18α,β 13.963 L18 α,β
CH3
0.88 (t)
The chemical shift assignments of proton (1H) and carbon-13 (13C) resonances of triglyceride acyl chains were reported along with the 1H coupling constant (J) patterns where m ⫽ multiplet, dd ⫽ doublet of doublet, t ⫽ triplet. Saturated (S), oleate (O), vaccenate (V) and linoleate (L) chains were also indicated according to their positions on the glycerol backbone, where 1,3- and 2-positions were labeled by Greek letters α and β, respectively.
CHAPTER | 65
13
605
C NMR Spectroscopy: Applications to Olive Oil Authenticity
chains). The olefinic carbon regions of olive and soybean oils are reported in Figure 65.1. Because α-linolenic acid is a minor component (0.9%) of olive oil, soybean oil 13C spectrum was measured and the olefinic carbon region is shown along with expansions in Figure 65.2. The assignments of unsaturated carbons of oleate, linoleate and α-linolenate chains of soybean oil triglycerides are reported in Table 65.2.
65.3 13C NMR FOR CARRYING OUT FATTY ACID POSITIONAL ANALYSIS OF OLIVE OIL TRIACYLGLYCEROLS 65.3.1 Conventional 13C NMR Spectroscopy Conventional 13C NMR spectroscopy, which applies the basic one-pulse sequence to carry out signal averaging, can be used to measure carboxy carbon resonances and determine the positional acyl chain composition of triglycerides. Signal averaging represents a crucial point because of a low sensitivity of 13C nucleus due to its low natural abundance (1.1%) and to a gyromagnetic ratio which is four times lower than that of 1H. When conventional 13C NMR spectroscopy is used for quantitative measurements, resonance intensity distortions have to be eliminated by using
relaxation delays long enough to avoid signal saturation, and by suppressing NOE occurring during proton decoupling (the inverse-gated pulse sequence must be applied). Unfortunately, a dramatic increase in signal averaging time is registered. Nevertheless, the spectrum sensitivity can be improved by measuring proton-decoupled spectra with full NOE enhancement after demonstrating that proton decoupling affects carboxy carbon intensities to the same extent (Vlahov, 1998). Evidence was also given to the decrease of spin-lattice relaxation times T1 and NOE factors of carboxy carbons when operating at a higher magnetic field. According to the results reported in Table 65.3, the high field dependence of T1 and NOE values, which were constantly lower at a higher field (125.7 MHz), supported a chemical shift anisotropy mechanism (CSA) of relaxation rather than a dipolar mechanism (D) in agreement with relaxation rates R1CSA measured at a higher field (125.7 MHz) which were markedly higher (T1 were lower) than R1D rates, where R1CSA ⫽ 1/T1CSA and R1D ⫽ 1/T1D (Vlahov, 2006b). Long signal averaging times have also to be used when unsaturated carbons are detected by conventional 13C NMR spectroscopy, in correspondence of unsaturated carbon T1 ranging from 1.38 to 3.01 s (Table 65.4). Since a higher resolution degree can be achieved for unsaturated carbons,
O10
O9
L13
130.5
130.0
L9
L10
129.5
129.0
128.5
128.0
Ln9 Ln13
130.5 13
130.0
129.5
129.0
128.5
Ln12
128.0
L12
127.5
Ln10
127.5
127.0
ppm
Ln15
127.0
ppm 13
FIGURE 65.1 C spectrum of olive and soybean oils: olefinic carbons. Olefinic carbon region 126.5–130.5 ppm of the 500 MHz C spectrum of olive oil (upper trace), a ‘high oleic acid oil’, and of soybean oil (bottom trace) a ‘high linoleic acid oil’. Oleate (O), linoleate (L) and linolenate (Ln) chains are shown.
606
130.1
130.0
129.9
129.8
131.5
129.7
131.0
129.6
130.5
130.0
128.3
129.5
129.0
128.1
128.5
127.9
128.0
127.7
127.5
127.015 127.010
127.669 127.651
128.163 128.155 128.111 128.099
ppm
127.798 127.786
127.961 127.979
129.841 129.817
129.574 129.548
130.062
129.883 129.867
130.051
130.043
SECTION | I Natural Components
127.5
127.3
127.1 ppm
127.0 ppm
13
FIGURE 65.2 C spectrum of soybean oil: olefinic carbons. Olefinic carbon region 126.5–132.0 ppm of the 500 MHz 13C spectrum of soybean oil: expansions of C-9 and C-13 range (left upper trace) and of C-10, C-12, and C-15 range (right upper trace) are shown.
Vlahov proposed the use of DEPT pulse sequence to measure unsaturated carbon resonances and determine quantitative positional profiles of triglyceride acyl chains. (Vlahov, 1997).
65.3.2 Non-Conventional 13C NMR Spectroscopy DEPT pulse sequence applies polarization transfer from 1 H concentrate spins to 13C dilute spins with which they are coupled. Signal obtained from protons is 64 times that of carbons because, in general, the signal available from a nucleus with gyromagnetic ratio γ is proportional to γ3 (gyromagnetic ratio of 13C nucleus is four times lower than that of 1H nucleus). Polarization transfer makes 13C signal intensity increase by a factor of 4 (I ⫽ I0 ⫻ γ1H/γ13C 艑 I0 ⫻ 4) unlike the increase by a factor of 3 obtained for protonated carbons (I ⫽ I0 ⫻ (1 ⫹ γ1H/2 γ13C) 艑 I0 ⫻ 3) detected under proton decoupling in presence of full NOE from protons (I represents the enhanced signal intensity and I0 the intensity without enhancement) (Derome, 1991). But the real benefit of using polarization transfer is that repetition rate is determined by T1 of 1H nuclei, which are considerably shorter than those of 13C nuclei (Table 65.4).
Therefore, the intensity distortions due to signal saturation occurring when an insufficient relaxation delay D1 between pulses is allowed are avoided by using a delay D1 ⫽ 4 ⫻ T1 where the longest T1 (2.4 s) was measured in correspondence of methyl protons. As a result, the spectrum sensitivity was considerably improved in an experiment time reduced by 50% as compared to that of conventional 13C NMR.
65.3.3 Fatty Acid Positional Analysis In order to determine the positional acyl chain composition of olive oil triacylglycerols, the use of 13C NMR DEPT methodology is strongly recommended to obtain, in a reasonably short experiment time, a spectrum sensitivity high enough to enable an accurate measurement of resonance intensities. Carboxy carbons which detect both saturated and unsaturated chain resonances and their positions (1,3- and 2-) on the glycerol backbone, are not detected by DEPT sequence (Derome, 1991). As a consequence, the full composition (mol %) of saturated, oleate and linoleate chains of the whole triglyceride was determined in the frequency region from 31.90 to 31.40 ppm (Figure 65.3), where ω-3
CHAPTER | 65
13
607
C NMR Spectroscopy: Applications to Olive Oil Authenticity
TABLE 65.2 Assignments of 13C resonances of unsaturated carbons of soybean oil triacylglycerols. Carbon
Oleate (O) C 18:1 9cis
Linoleate (L) C 18:2 9,12cis
α-Linolenate(Ln) C 18:3 9,12,15cis
C-9 1,3-pos
129.574
129.841
130.062
C-9 2-pos
129.548
129.817
130.051
C-10 1,3-pos
129.867
127.961
127.651
C-10 2-pos
129.883
127.979
127.669
C-12 1,3-pos
127.798
128.111
C-12 2-pos
127.786
128.099
C-13 1,3-pos
130.043
128.155
C-13 2-pos
130.051
128.163
C-15 1,3-pos
127.015
C-15 2-pos
127.010
C-16 1,3-pos
131.780
C-16 2-pos
131.780
Unsaturated carbons were assigned according to the chain double bond number and to the chain positions on glycerol backbone of triacylglycerols, i.e., 1,3- and 2-positions, where 1- and 3-positions cannot be differentiated.
TABLE 65.3 T1 and NOE of carboxy carbons of standard triglycerides. T1-NOE R1D-R1CSA
Tripalmitin C 16:0
Triolein C 18:1 9cis
Trilinolein C 18:2 9,12cis
75.4 MHz
125.7 MHz
75.4 MHz
125.7 MHz
75.4 MHz
125.7 MHz
1,3-positions T1(s) NOE (η ⫽ I–I0/I0) R1D ⫽ 1/T1D (sec) R1CSA ⫽ 1/T1CSA (sec)
6.2 0.8 0.068 0.060
3.7 0.5 0.061 0.165
6.0 1.0 0.082 0.048
3.9 0.4 0.048 0.134
6.4 0.9 0.074 0.056
3.8 0.4 0.055 0.156
2-position T1 NOE R1D R1CSA
5.4 0.7 0.068 0.085
3.0 0.4 0.067 0.236
4.9 0.9 0.094 0.047
3.4 0.5 0.076 0.132
5.4 0.9 0.095 0.054
3.3 0.4 0.055 0.150
Longitudinal relaxation times (T1) and nuclear Overhauser enhancement factors (NOE, η was calculated on the basis of I, enhanced signal intensity, and I0, intensity without enhancement) of carboxy carbons decreased when operating at a higher magnetic field (125.7 MHz) in agreement with a predominant chemical shift anisotropy mechanism of relaxation (CSA) over a dipolar mechanism (D).
carbons of saturated (31.852 ppm), oleate (31.832 ppm), vaccenate (31.709 ppm) and linoleate (31.444 ppm) chains, resonate. Because ω-3 carbons do not detect the unsaturated chain position on the glycerol backbone, the chain distribution between 1,3- and 2-positions was calculated
by using unsaturated carbon resonances. The data, since saturated chains esterify only 1,3-positions of triacylglycerols (Gunstone, 1967), were corrected for the saturated chain percentage determined from the ω-3 carbon resonance.
608
SECTION | I Natural Components
TABLE 65.4 T1 of 1H and 13C nuclei of olive oil triglycerides. 1
13
Glycerol moiety –CH2OH 1(3)-positions
0.55
0.28
–CHOH 2-position
0.85
0.40
Acyl chain moiety –CH2–COOR
0.82
0.69
–CH2CH2COOR
0.74
0.74
–(CH2)n
0.88
0.9÷2.15
–CH2–CH ⫽ CH
0.95
1.20
–CH ⫽ CH–
1.50
1.38÷3.01
–CH3
2.44
4.20
Resonances
H T1 s
C T1 s
31.832
Longitudinal relaxation times T1 have to be measured when 1H and 13C NMR spectroscopy is used for quantitative measurements. The use of relaxation delays four times longer than the longest T1 measured in correspondence of spectrum signals, prevents any resonance intensity distortion due to signal saturation.
Olive oils sampled in the Abruzzo, Puglia and Sicilia Italian regions were used to calculate the acyl chain composition of triglycerides, their percentages at 1,3- (sn-1- and sn-3-positions cannot be separately detected) and 2-glycerol positions (Table 65.5). The percentages of oleate (Otheory) and linoleate (Ltheory) chains in the total of oleate and linoleate chains in the triglyceride, were compared to the percentages of oleate (Ofound) and linoleate (Lfound) chains in the total of these chains at 2position (Vlahov, 2005). The data showed that oleate chain at 2-position deviated from a random distribution pattern. Oleate chain would have been randomly distributed at 2-position (the random pattern is represented by a dotted line) if its percentages given by Otheory were equal to Ofound percentages. Deviation rate increased upon decreasing of oleate percentage in triglyceride (Figure 65.4). As far as the distribution of linoleate chain at 2-position is concerned, the chain deviated from a random distribution pattern (Lfound values being higher) at an increasing rate for increasing values of linoleate percentage in the total of oleate and linoleate chains in the triglyceride (Ltheory) (Figure 65.5). Since the composition of oleate and linoleate
31.709
31.816
S
31.444
31.852
O
L V
31.85 13
31.80
31.75
31.70
31.65
31.60
31.55
31.50
ppm 13
FIGURE 65.3 C spectrum of olive oil: omega-3 carbons. Omega-3 carbon region 31.4–31.9 ppm of the 500 MHz C spectrum of olive oil where saturated (S), oleate (O), vaccenate (V) and linoleate (L) chains are detected.
CHAPTER | 65
13
chains depends on one another because they were measured in percentage, the gradient for both lines was defined by the same slope value (Table 65.6). Structural data of triacylglycerols were completed by determining the distribution of saturated, oleate and
TABLE 65.5 Composition and positional distribution of acyl chains in triacylglycerols of olive oils from the Abruzzo, Puglia and Sicilia Italian regions. Triacylglycerol Abruzzo structure
Puglia
Sicilia
Composition (%) Saturated 19.8 ⫾ 1.5 Oleate 70.5 ⫾ 4.6 Linoleate 9.7 ⫾ 3.3
17.3 ⫾ 1.7 73.9 ⫾ 3.4 8.8 ⫾ 2.1
18.3 ⫾ 1.9 71.5 ⫾ 3.3 10.1 ⫾ 2.0
1,3-Distribution (%) Saturated 29.7 ⫾ 2.2 Oleate 63.3 ⫾ 4.1 Linoleate 7.0 ⫾ 2.2
26.0 ⫾ 2.5 67.3 ⫾ 3.7 6.7 ⫾ 1.6
27.5 ⫾ 2.9 64.5 ⫾ 3.9 8.0 ⫾ 1.6
2-Distribution (%) Oleate 84.9 ⫾ 5.6 Linoleate 15.1 ⫾ 5.6
86.9 ⫾ 3.1 13.0 ⫾ 3.1
85.6 ⫾ 2.8 14.4 ⫾ 2.8
2-Specificity (%) Oleate 40.1 ⫾ 0.6 Linoleate 51.2 ⫾ 2.0
39.2 ⫾ 0.9 50.2 ⫾ 2.3
39.9 ⫾ 1.3 47.5 ⫾ 1.7
The composition of saturated, oleate and linoleate chains in the whole triglyceride, their distribution between 1,3- and 2- positions and their 2-position specificity were calculated by using the intensities of 13C resonances of omega-3 and unsaturated carbons of triglyceride acyl chains. The precision of data was quoted as standard deviation of the mean.
linoleate chains in the combined 1,3-positions. Percentages of saturated (Stheory), oleate (Otheory) and linoleate (Ltheory) chains in their total percentages in the triacylglycerols were compared with percentages of saturated (Sfound), oleate (Ofound), and linoleate (Lfound) chains in their total percentages at the 1,3-positions. 1,3-position chains moved away from a random distribution pattern, where saturated chains, unlike oleate and linoleate chains, entered the 1,3-positions more than would have been expected for a random distribution. Moreover, saturated and linoleate chain, unlike oleate chain, deviated from a random distribution pattern at an increasing rate in correspondence to the increase in chain concentration in triacylglycerol. Statistics of the linear models which were calculated for the chain distribution at 1,3- and 2-positions (Chainfound) as compared to the chain concentration in the triglyceride (Chaintheory), are reported in Table 65.6 (‘Chain’ can be saturated, oleate, linoleate chains). The chain distribution patterns at 1,3- and 2-positions were in agreement with the linear models calculated by using the 13C NMR data measured on an olive oil set made up of 691 samples (Vlahov, 2006a). The linear model equations based on this set of olive oil samples, were applied to predict the distribution of oleate and linoleate chains at 2-position for 25 olive oil samples (Vlahov, 2006a) and the results were compared to the data predicted by the Computer method. The Computer method is based on the 1,3-random– 2-random pattern of acyl chain distribution in triglycerides, where glycerol is randomly esterified at 2-positions with a mixture of fatty acids, and 1- and 3-positions assumed to be identical, are esterified at random with another mixture of fatty acids. However, the Computer method is an indirect method because it determines the fatty acid composition at
100 95 90 85 80 75 70 70
75
80
85
90
95
100
Otheory = Otg x 100 / Otg + Ltg FIGURE 65.4 Positional analysis of olive oil triglycerides: oleic acid. Percentages of oleic acid in the total of oleic and linoleic acids in triglyceride (Otheory) are compared to its percentages in the total of oleic and linoleic acids at 2-position (Ofound).
Lfound = L2dis x 100 / O2dis + L2dis
Ofound = O2dis x 100 / O2dis + L2dis
609
C NMR Spectroscopy: Applications to Olive Oil Authenticity
40 35 30 25 20 15 10 5 0
0
5
10 15 20 25 Ltheory = Ltg x 100 / Otg + Ltg
30
35
FIGURE 65.5 Positional analysis of olive oil triglycerides: linoleic acid. Percentages of linoleic acid in the total of oleic and linoleic acids in triglyceride (Ltheory) are compared to its percentages in the total of oleic and linoleic acids at 2-position (Lfound).
610
SECTION | I Natural Components
TABLE 65.6 Statistics of linear models calculated for the chain distribution at 1,3- and 2positions (Chainfound) as compared to the chain concentration in triglycerides (Chaintheory) of olive oils from the Abruzzo, Puglia and Sicilia Italian regions. Triglyceride acyl chains
Slope
Intercept
r
R2, %
Sy/x
F
1,3 –positions C n:0 (S) C 18:1 9c (O) C 18:2 9,12c (L)
1.50 ⫾ 0.007 0.99 ⫾ 0.02 0.69 ⫾ 0.014
0.047 ⫾ 0.13 ⫺6.55 ⫾ 1.45 0.51 ⫾ 0.14
0.99 0.98 0.98
99.81 96.19 95.97
0.13 0.85 0.40
50999 2428 2286
2-position C 18:1 9c (O) C 18:2 9,12c (L)
1.24 ⫾ 0.021 1.24 ⫾ 0.021
⫺23.63 ⫾ 1.87 ⫺0.17 ⫾ 0.26
0.99 0.99
97.28 97.28
0.76 0.76
3430 3430
The correlation coefficients r ⱖ 0.98 indicated a strong relationship between Chainfound and Chaintheory in correspondence of saturated, oleate and linoleate chains at 1,3- and 2-positions, where the coefficient of determination R2 showed that the linear models explained ⱖ95.97 % of the variability in the Chainfound values. The observed values for F statistic higher than critical F for p ⫽ 0.05 significance level, confirmed that linear correlations were not random. Residual standard deviation Sy/x was also indicated.
25
100 95
20
90 15 85 10
80
5
75 70
1
3
5
7
9
11
13
15
17
19
21
23
25
0
13
FIGURE 65.6 Computer method versus C NMR method: oleate chain. Distribution data of oleate chain at triglyceride 2-position (Ofd2) obtained for a set of 25 olive oil samples by using the computer method (CM, 䊏) and the 13C NMR method (NMR, 䊊).
2-position by repeated subtractions starting from the molar percentage of saturated chains at 2-position which is calculated by using the coefficient of 0.06 (Pallotta, 1994). As expected, oleate chain percentages at 2-position calculated by the Computer method, were higher than those predicted by the 13C NMR linear models (Figure 65.6), however, 13C NMR data of linoleate chain at 2-position were found to be 0.84 times higher than those calculated by the Computer method (Figure 65.7). The results confirmed that the 1,3-random-2-random distribution theory cannot adequately predict the structure of triacylglycerols. They also suggested that 13C NMR spectroscopy can provide the equations, calculated on the basis of a large data set, for determining the fatty
1
5
9
13
17
21
25
13
FIGURE 65.7 Computer method versus C NMR method: linoleate chain. Distribution data of linoleate chain at triglyceride 2-position (Lfd2) obtained for a set of 25 olive oil samples by using the computer method (CM, 䊏) and the 13C NMR method (NMR, 䊊).
acid composition of the two pools of fatty acids esterifying 1,3- and 2-positions of triacylglycerols, respectively.
65.4 13C NMR OF TRIACYLGLYCEROLS FOR DETERMINING OLIVE OIL AUTHENTICITY 65.4.1 Triacylglycerols as Tracer Substances for Determining Olive Oil Authenticity The new positional data of triacylglycerols provided by C NMR spectroscopy played a main role in determining olive oil authenticity in terms of either geographical origin, or adulteration, or grades of olive oils. In particular, the 13
13
611
C NMR Spectroscopy: Applications to Olive Oil Authenticity
deviation of unsaturated chains from a 2-random distribution, at different extents according to the fatty acid concentration of triglycerides, introduced a new discrimination factor of olive oil triglycerides – the positional specificity factor. Application of 13C NMR spectroscopy to measure the quantitative profile of olive oil triacylglycerols is based on the assumption that all olive oil samples are composed of a common set of structural units – the long-chain acids of triacylglycerols (Table 65.1). Because olive oil samples differ in relative amounts of these components, their resonance intensities are the quantitative parameters for finding correlations among the oils in correspondence of the same characteristic factors, namely geographical origin, adulteration and grades. Triacylglycerols, which are the most abundant fraction (98%) of olive oil, are selected as tracer substances of olive oil authenticity because they assure a high signal intensity in spite of the low sensitivity of 13CNMR spectroscopy. Moreover, the 13C NMR spectrum can be considered almost a fingerprint of an olive oil sample because sample preparation is zeroed and consequently, any chemical reaction artefact is eliminated. Repeatability of the 13C NMR method, i.e., within-run (a short term) precision, and reproducibility, i.e., betweenrun (a very long term) precision, are also considerably improved as compared with the procedures based on chemical and physical constants. In particular, coefficients of variation for resonance intensities were lower than 4% thus making 13C NMR DEPT methodology a rigorous quantitative methodology (Vlahov et al., 2001). With the aim of finding differences among olive oil group means with the same authenticity factor, statistical methods of multivariate analysis were applied to the resonance intensities from different olive oil spectra. Multivariate calibration methods derive their power from the simultaneous use of multiple variables, they are represented by the resonance intensities of triacylglycerol 13 C nuclei.
65.4.2 Geographical Origin of Olive Oil Triacylglycerols were selected as tracer substances to determine the authenticity of olive oils in terms of their geographical origin. Fatty acid profiles of triglycerides are highly influenced by the genetic makeup of the cultivar and by the latitude which determines temperatures where olive trees grow. In particular, lower temperatures support the synthesis of fatty acids with a higher unsaturation level. Thus both cultivar and latitude factors appear to influence the fatty acid profiles of olive oil triglycerides (Lotti et al., 1982). The determination of geographical origin of olive oils by using the 13C NMR DEPT method was carried out on
4.8
Function 2
CHAPTER | 65
2.8 0.8 −1.2 −3.2 −3.3
−1.3
0.7
2.7
4.7
Function 1 FIGURE 65.8 Determination of geographical origin of olive oil. Plot of 92 olive oil samples from different Italian regions against their values for two discriminant functions: Abruzzo (䊐; 1.48, 0.053), Puglia (⫻; ⫺1.23, ⫺1.22) and Sicilia (䊊; ⫺1.41, 1.81) regions. The centroids (⫹) for each olive oil group are indicated in parentheses.
olive oils samples from 13 Italian areas with Denomination of Protected Origin (DOP). Multivariate structure of 13C NMR data was explored by using an unsupervised method of pattern recognition, namely principal component analysis (PCA). The two principal components 1 and 2 explaining 68.1% and 15.9% of variance, respectively, showed that the oil samples from the same geographical area grouped tightly provided that oil composition is monovarietal (Vlahov et al., 2001). The main role of cultivar factor was confirmed by applying linear discriminant analysis (LDA) to the oils sampled in the three DOP areas of the Puglia region. LDA is a supervised pattern recognition method which, unlike PCA, aims to define linear combinations of the original variables which maximize the ratio of variance between groups to variance within groups. Two discriminating functions correctly classified in the true group 90% of olive oils from the Colline di Brindisi and Terra di Bari DOPs (they were monovarietal oils based on Ogliarola and Coratina cultivars, respectively), but only 74% of Dauno DOP oils, they are based on Peranzana, Ogliarola and Coratina cultivars, was assigned to the true group (Vlahov et al., 2003). Classification of multivarietal olive oils on a regional scale, evidenced that 95.4, 93.3 and 84.2% of olive oils sampled in the Abruzzo, Puglia and Sicilia Italian regions, respectively, were correctly classified in the true group by the discriminating model based on two functions. The score plot of oil samples on discriminant functions 1 and 2 is reported in Figure 65.8 along with the centroid values (they represent average values of each function within the oil groups). The first discriminant dimension separated the Abruzzo oil group from the Sicilia and Puglia oil groups, where the highest positive standardized coefficient of Function 1 was measured for the saturated chain variable in agreement with the highest content of saturated chains of the Abruzzo oils (Table 65.5).
612
2.3 1.3
Function 2
In terms of the second dimension, the Sicilia oil group was separated from the Puglia and Abruzzo oil groups which appeared indistinguishable. Discriminant dimension 2 was most highly weighted (higher absolute value for standardized coefficient of Function 2) in the negative direction with C-9 of linoleate chain at 2-position. This result was in agreement with the higher 2-position specificity values measured for linoleate chain in the Abruzzo and Puglia oils (51.2 and 50.2, respectively) as compared to the value of 47.5 determined in the Sicilia oil group (2-position specificity measures the percentage of a chain at 2-position relative to the total of the chain in the 1,3- and 2-positions) (Table 65.5).
SECTION | I Natural Components
0.3 −0.7 −1.7 −2.7 −3.7 −10
−6
−2
2
6
10
14
Function 1 FIGURE 65.9 Determination of olive oil adulteration. Plot of eight olive oil samples adulterated by mixture with hazelnut oil, against their values for two discriminant functions: extra virgin oils (䊐; ⫺ 7.93, 0.92), 10% hazelnut (⫻; ⫺ 4.13, ⫺1.31), 20% hazelnut (䊊; 1.26, 0.26), and 40% hazelnut (†; 10.81, 0.14) oil. The centroids (⫹) for each olive oil group are indicated in parentheses.
65.4.3 Adulteration of Olive Oil Adulteration of extra virgin olive oil with cheap, highly refined hazelnut oil imported from Turkey is one of the most common frauds. The presence of a refined oil may be potentially dangerous to consumers’ health because of chemical changes of oil compounds caused by the refining process. Once more 13C NMR methodology based on triacylglycerol resonances, coupled with multivariate method of LDA, proved that high discrimination can be achieved for extra virgin olive oil samples from eight olive fruit cultivars, adulterated at different concentrations of hazelnut oil. In particular, 100% of the oil groups corresponding to 10, 20, and 40%, and 94% of the oil group at 50% hazelnut oil, were correctly classified in the true group. However, less than 70% of extra virgin olive oils and of oils at 5% hazelnut were correctly classified (Vlahov, 2006c). After demonstrating that different hazelnut oil concentrations can be detected in olive oils from different cultivars, the influence of growing area factor was explored on discrimination of hazelnut adulteration levels. 13C NMR spectra of olive oils sampled in the Abruzzo, Puglia and Sicilia regions and adulterated by mixture with hazelnut oil at 10, 20 and 40% were measured. The score plot of the values of discriminant functions 1 and 2 in correspondence of the oil samples (Figure 65.9), evidenced that the highest discrimination was achieved on the discriminant function 1 where the oils adulterated by 40 and 20% hazelnut oil formed two groups well separated among themselves and from the olive oil samples at 0 and 10% hazelnut oil. The discriminating model correctly classified 100% of the oils in the true group.
65.4.4 Grades of Olive Oil 13
C NMR was also applied to verify the reliability of triacylglycerols in differentiating four olive oil grades: extra virgin olive oils, olive oils, olive pomace oils and cold-pressed olive oils (Vlahov, 2006d).
TABLE 65.7 2-specificity of oleate and linoleate chains in triacylglycerols of extra virgin and cold-pressed olive oils. Triacylglycerol structure
Extra virgin olive oil
Cold-pressed olive oil
Composition (%) Saturated Oleate Linoleate
20.7 72.0 7.3
21.9 70.9 7.2
2-specificity Oleate Linoleate
37.4 50.0
32.8 42.7
The lower 2-specificity values of oleate and linoleate chains in coldpressed olive oils support the discrimination of these oils from extra virgin olive oils.
LDA of resonance intensities of triacylglycerols from four olive oil grades showed that discriminant function 1 enabled the classification in the true classes of 80.8% of virgin olive oils, 100% of cold-pressed olive oils and olive pomace oils, whereas only 73.3% of olive oils were correctly classified. Acyl chain positional data confirmed their main role in discriminating olive oil grades. 2-specificity values of oleate and linoleate chains in cold-pressed olive oils were lower (the 1,3-position values were correspondingly higher) than those determined for extra virgin olive oils (the data are reported in Table 65.7) and matched the separation of the cold-pressed olive oils at lower values for discriminant function 1. They were the higher values (absolute values) of the standardized coefficients of the variables C-10 of oleate chain and C-12 of linoleate chain at 1,3-positions
CHAPTER | 65
13
C NMR Spectroscopy: Applications to Olive Oil Authenticity
which determined the lower values for the discriminant function 1 in correspondence with the lower 2-specificity values for the chains.
SUMMARY POINTS ●
●
●
●
●
●
13
C and 1H NMR spectra of olive oil triacylglycerols were assigned. Conventional 13C NMR spectroscopy for quantitative measurements of carboxy and olefinic carbons of triacylglycerols was discussed. Non-conventional 13C NMR spectroscopy which applies DEPT pulse sequence for quantitative measurements of olefinic carbons, was examined. Positional analysis of fatty acids in triacylglycerols of olive oils was carried out by 13C NMR spectroscopy. Fatty acid compositions of 1,3- and 2-triacylglycerol positions as measured by 13C NMR spectroscopy, evidenced that 1,3-random–2-random distribution theory cannot adequately predict the structure of triacylglycerols. 13 C NMR spectroscopy of olive oil triacylglycerols can be successfully used to determine olive oil authenticity in terms of geographical origin, adulteration and grades of olive oil.
REFERENCES Derome, A.E., 1991. Modern NMR Techniques for Chemistry Research. Pergamon Press Limited, Oxford. Kuksis, A., 1996. Analysis of positional isomers of glycerolipids by non-enzymatic methods. In: Christie, W.W. (Ed.), Advances in Lipid Methodology – Three. The Oily Press, Dundee, pp. 1–36. Gunstone, F.D., 1967. An Introduction to the Chemistry and Biochemistry of Fatty Acids and Their Glycerides. Chapman and Hall Ltd, Great Britain.
613
Howarth, O.W., Samuel, C.J., Vlahov, G., 1995. The σ-inductive effects of C⫽C and C⬅C bonds: predictability of NMR shifts at sp2 carbon in non-conjugated polyenoic acids, esters and glycerides. J. Chem. Soc. Perkin Trans. 2, 2307–2310. Lotti, G., Izzo, R., Riu, R., 1982. Influenza del clima sulla composizione acidica e sterolica degli oli di oliva. Riv. Soc. It. Sci. Alim. 2, 115–126. Pallotta, U., 1994. A review of Italian research on the genuineness and quality of extra virgin olive oil. Ital. J. Food Sci. 3, 259–274. Vlahov, G., 1997. Quantitative 13C NMR method using the DEPT pulse sequence for the detection of olive oil adulteration with soybean oil. Magn. Reson. Chem. 35, S8–S12. Vlahov, G., 1998. Regiospecific analysis of natural mixtures of triglycerides using quantitative 13C nuclear magnetic resonance of acyl chain carbonyl carbons. Magn. Reson. Chem. 36, 359–362. Vlahov, G., Schiavone, C., Simone, N., 2001. Quantitative 13C NMR method using the DEPT pulse sequence for the determination of the geographical origin (DOP) of olive oils. Magn. Reson. Chem. 39, 689–695. Vlahov, G., Del Re, P., Simone, N., 2003. Determination of geographical origin of olive oils using 13C nuclear magnetic resonance spectroscopy. I – Classification of olive oils of the Puglia region with Denomination of Protected Origin. J. Agric. Food Chem. 51, 5612–5615. Vlahov, G., 2005. 13C nuclear magnetic resonance spectroscopy to check 1,3-random, 2-random pattern of fatty acid distribution in olive oil triacylglycerols. Spectroscopy 19, 109–117. Vlahov, G., 2006a. Determination of the 1,3- and 2-positional distribution of fatty acids in olive oil triacylglycerols by 13C nuclear magnetic resonance spectroscopy. J. AOAC Int. 89, 1071–1076. Vlahov, G., 2006b. 13C nuclear magnetic resonance studies of mono-, diand tri-acylglycerols: NOE factors and spin-lattice relaxation of acyl chain carboxy carbons. In: Welson, L.T. (Ed.), Triglycerides and Cholesterol Research. Nova Science Publishers, Inc., Hauppauge, NY, pp. 251–270. Vlahov, G., 2006c. 13C NMR determination of olive oil adulteration by hazelnut oil. INFORM 17, 535–536. Vlahov, G., 2006d. 13C nuclear magnetic resonance spectroscopy to determine olive oil grades. Anal. Chim. Acta 577, 281–287. Wollemberg, K.F., 1990. Quantitative high resolution 13C NMR of the olefinic and carbonyl carbons of edible vegetable oils. J. Am. Oil Chem. Soc. 67, 487–494.
This page intentionally left blank
Chapter 66
Extraction Techniques for the Analysis of Virgin Olive Oil Aroma Stefania Vichi Departament de Nutrició i Bromatologia, Facultat de Farmàcia, Universitat de Barcelona, Spain
66.1 INTRODUCTION Sensory characteristics are used to define virgin olive oil (VOO) quality. This oil has a characteristic flavor that distinguishes it from other edible vegetable oils. After its extraction from the fruit of Olea europaea, VOO can be consumed without refining and it preserves its typical aroma. In recent years, the need for analytical procedures to evaluate the quality of VOO has led to several studies addressing its volatile fraction. Various analytical methods have been developed to examine these volatile compounds. In this way, a large number of components that contribute to the aroma of olive oil have been identified.
The volatile profile of VOO closely depends upon the method of extraction used (Cavalli et al., 2003; Vichi et al., 2007).
66.2 FEATURES OF VIRGIN OLIVE OIL AROMA The characteristic aroma of VOO, and in particular, the green and fruity attributes, depend on many volatile compounds derived from the degradation of polyunsaturated fatty acids through a chain of enzymatic reactions known as the lipoxygenase (LOX) pathway taking place during the oil extraction process (Figure 66.1).
Linoleic acid (LA)
Linolenic acid (LNA)
LOX
LOX
13-LA Hydroperoxide
13-LNA Hydroperoxide
HPL Hexanal
HPL (Z)-3-Hexenal
13-Alcoxy radical
ADH Hexanol
Isomerase ADH
Pentene radical
(E)-2-Hexenal
AAT Hexyl acetate
ADH Pentene dimers
(Z)-3-Hexenol
(E)-2-Hexenol
AAT 2-Penten-1-ol 1-Penten-3-ol 2-Pentenal 1-Penten-3-one
(Z)-3-Hexenyl acetate
FIGURE 66.1 VOO volatile compounds derived from the LOX action on LA and LNA. The figure resumes the pathways of formation of C6 and C5 volatile compounds of virgin olive oil, following the attack of LOX on polyunsaturated fatty acids. LA: linoleic acid; LNA: linoleic acid; LOX: lipoxygenase; HPL: hydroperoxide lyase; ADH: alcohol dehydrogenase; AAT: alcohol acyl transferase. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
615
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
616
Variable amounts of hexanal, hexanol, and hexyl acetate derive from the degradation of linoleic acid, whereas (Z)-3-hexenal, (E)-2-hexenal, (E)-2-hexenol, (Z)-3-hexenol, and (Z)-3-hexenyl acetate result from the enzymatic degradation of linolenic acid (Olías et al., 1993). Moreover, pentene dimers, pentenols, and C5 carbonyl compounds are thought to originate from β-scission of alkoxy radicals formed from 13-hydroperoxides by a LOX-mediated mechanism (Angerosa et al., 1998). Other linear alcohols, acids, esters and ketones, together with mono- and sesquiterpenes, deriving from the fruit metabolism, have also been found as constituents of VOO volatile fraction (Cavalli et al., 2003; Vichi et al., 2003a, 2006; Zunin et al., 2004). The volatile profiles of VOOs are believed to be influenced by factors such as the cultivar of the olives, climate, soil quality, degree of ripeness of the fruit, and oil extraction process (Montedoro et al., 1978; Angerosa, 2002). The preservation of olive fruits or olive oil under inadequate conditions, as well as unsuitable harvesting methods or technological processes, may decrease the sensory quality of olive oil by altering its typical volatile fraction composition. Basically fermentations, exogenous enzymatic processes and oxidation reactions give rise to undesirable volatile compounds affecting the VOO aromatic profile (Angerosa, 2002).
66.3 HEADSPACE EXTRACTION TECHNIQUES 66.3.1 Static Headspace (SHS) Headspace extraction techniques are selective techniques exploiting the volatility of some compounds with respect to the sample matrix, and their capacity to move from the sample to the headspace of a sealed container, from where they are collected. Static headspace (SHS) sampling is the simplest of these techniques, and it consists in removing a volume of the sample headspace after the equilibrium has been reached and introducing it into the analytical instrument. The concentration of the analyte in the headspace depends on its amount in the sample, its volatility and solubility in the sample matrix, the volume of the headspace and the temperature. Therefore, the sample can be heated to increase the concentration of analytes in the headspace. Because of its limited sensitivity, SHS is rarely employed for olive oil analysis. This extraction technique was inefficient for the analysis of VOO volatiles when used prior to gas chromatographic techniques (Cavalli et al., 2003). However, SHS directly coupled to a mass spectrometer (SHS-MS) has been recently applied for the characterization of olive oil (Cerrato Oliveros et al., 2005), for the quantification of VOO sensory attributes (López-Feria et al., 2007), and for the detection of adulterants (Lorenzo et al., 2002) and contaminants (Peña et al., 2004) in olive oil. In these works, the
SECTION | I Natural Components
total ion current signal was considered a spectral fingerprint of the oil sample, and statistical analyses were applied to relate the independent variables with the dependent variables represented by each m/z ratio obtained by the MS.
66.3.2 Dynamic Headspace (DHS) In order to overcome some of the sensitivity limitations of static headspace sampling, a dynamic headspace (DHS) technique, also known as purge and trap sampling, has been introduced. DHS involves the passing of carrier gas through a liquid sample, sweeping out the volatiles, and their continuous transferring through a trap which retains them. Heating the sample can increase the concentration of the compounds in the headspace and so increase the amount of sample on the trap. Nevertheless, in the case of VOO, the oxidative degradation must be taken into account when fixing the extraction temperature. In addition to temperature, also the flow rate, the stripping time and the volume of sample have to be optimized to obtain the highest uptakes and to avoid losses of analytes through the trap. One of the critical points of DHS is the loading capacity of traps. Over a certain volume of gaseous phase, the trap can be overloaded, with the resulting loss of analytes. This breakthrough can be prevented by carefully choosing the most adequate trapping material and working conditions. The use of several different trapping materials has been reported in the literature: solid or liquid sorbents, cold-traps or solvents (Pillonel et al., 2002). Carbon-based trapping materials and Tenax are the most widely used trapping media for the analysis of olive oil aroma. Tenax is characterized by a high thermal stability, a low bleed, and a good adsorption capacity for medium and high boiling compounds, in particular aldehydes, esters and hydrocarbons. Carbon-based traps show a strong adsorbent capacity for both low and medium molecular weight and in particular for alcohols and ketones (Angerosa, 2002; Pillonel et al., 2002). The high affinity of these traps against water is not a relevant inconvenience for olive oil analysis because of the low amount of water of this matrix. A comparative study showed that carbon-based traps are more efficient than Tenax for the analysis of VOO volatiles, independently from the applied conditions (Kanavouras et al., 2005). Desorption of thermally stable analytes from Tenax is generally performed by heating the trap and transferring the released volatiles to the gas chromatographic system by a carrier gas, while carbon traps require the elution at room temperature with a suitable solvent. Table 66.1 reports some applications of DHS in the analysis of VOO volatile composition, using Tenax and carbon-based traps.
66.3.2.1 Closed-Loop Stripping Apparatus (CLSA) A variation on dynamic headspace sampling is the closedloop stripping apparatus (CLSA). In this configuration the
CHAPTER | 66 Extraction Techniques for the Analysis of Virgin Olive Oil Aroma
617
TABLE 66.1 Applications of principal sorbents used in DHS for the analysis of VOO aroma. Adsorbent
Application
Reference
Tenax
Identification of volatile compounds in VOO
Montedoro et al., 1978; Dobarganes-García et al., 1980; Morales et al., 1994a
Analysis of volatile compounds in relation with olive ripeness
Morales et al., 1996
Analysis of volatile compounds related with VOO sensory attributes
Morales et al., 1994b; Aparicio et al., 1996
Analysis of volatiles related with off-flavors
Morales et al., 1997, 2005
Identification and evaluation of VOO biogenic compounds
Olías et al., 1993; Angerosa et al., 1998, 1999
Volatile determination for cultivar characterization
Dhifi et al., 2005
Analysis of volatile compounds related with VOO sensory attributes
Servili et al., 1995; Angerosa et al., 2000
Analysis of volatiles related with off-flavors
Solinas et al., 1987; Angerosa et al., 1992, 1996
Determination of monoaromatic hydrocarbons as possible products of olive metabolism
Biedermann et al., 1995
Charcoal
Tenax and charcoal are the principal sorbents used in DHS for the analysis of VOO aroma. As can be seen in the table, both of them have been largely used for similar types of applications.
gaseous phase flows through the sample and the trap in a closed circuit. Volatiles are purged from the sample and concentrated in the trap, and since this is a cyclic system, compounds eluting from the end of the trap will not be lost but will be carried back through the system until reaching equilibrium. CLSA has been applied for the analysis of VOO aroma and compared with other extraction techniques (Vichi et al., 2007). CLSA showed a great efficiency in extracting esters and hydrocarbons, including sesquiterpene compounds. Esters principally consisted of hexyl acetate and hexenyl acetate, while hydrocarbons mainly included compounds such as pentene dimers and alkylated benzenes. The latter are environmental pollutants previously documented in VOO volatile fraction (Biedermann et al., 1995; Vichi et al., 2005).
66.3.3 Solid-Phase Microextraction (SPME) Solid-phase microextraction (SPME) technique has been introduced as an alternative to the dynamic headspace technique as a sample preconcentration method prior to chromatographic analysis. SPME is a rapid, sensitive and solvent-free sampling technique first developed by Pawliszyn and co-workers (Arthur and Pawliszyn, 1990) for the analysis of pollutants in water. In recent years,
SPME has extended its applications to numerous other fields, in particular food flavor analysis. SPME is based on the partitioning of organic components between a liquid sample or its vapor phase and a thin sorbent phase coated onto fused silica fibers. In general, the trapped compounds are thermally desorbed by introducing the fiber into the injection port of a gas chromatograph. The amount of analytes adsorbed or absorbed by the fiber is regulated by their concentration in the sample and their partition coefficient. The partition of analytes between the sample and the fiber is influenced by the type of sorbent and the extraction conditions. The uptake of the analytes and therefore the sensitivity of the method can be improved by optimizing parameters such as time and temperature of extraction, stirring, volume of sample and headspace, and in the case of aqueous samples, by regulating the pH and the ionic strength (Pawliszyn, 1999). Due to the nature and composition of olive oil, only headspace-SPME (HSSPME) has been used for the analysis of its volatile fraction. The principal difficulty of the SPME analysis of lipid samples is the matrix effect, which causes the decrease in SPME efficiency. Indeed, the lipid sample participates in the distribution equilibrium of volatiles as well as fiber coatings, having a high affinity with organic compounds. In order to optimize the efficiency and sensitivity of the
618
SECTION | I Natural Components
TABLE 66.2 Commercial SPME fibers commonly used for the analysis of VOO aroma, and characteristics of sorbent phases. Fibers
Type of phase
Principal sorption process
Polarity
Length
References
PDMS
Polymeric
AB
Apolar
1 cm
Vichi et al., 2003a; Cavalli et al., 2003; Flamini et al., 2003; Tura et al., 2004; Kanavouras et al., 2005; Jimenez et al., 2006; Temime et al., 2006
CAR-PDMS
Porous particles/polymeric
AD
Semipolar
1 cm
Vichi et al., 2003a; Cavalli et al., 2003
PDMS-DVB
Porous particles/polymeric
AD
Semipolar
1 cm
Vichi et al., 2003a; Kanavouras et al., 2005; Jimenez et al., 2006
DVB-CAR-PDMS
Porous particles/polymeric
AD
Semipolar
2 cm
Vichi et al., 2003a
1 cm
Cavalli et al., 2003
2 cm
Runcio et al., 2008
2 cm
Manai et al., 2008
1 cm
Cavalli et al., 2003; Benincasa et al., 2003; Servili et al., 2003; Jimenez et al., 2006
CW-DVB
Porous particles/polymeric
AD
Polar
Mostly porous particles phases imbedded into polymeric phases have been used for the analysis of VOO aroma. AB: absorption; AD: adsorption.
method it is necessary to identify the most suitable SPME conditions. Basically, the type of fiber coating, temperature and time of extraction are the parameters to be taken into account in the case of olive oil since, in the case of lipids, the amount of sample does not affect the mass of analyte absorbed by the SPME coating (Page and Lacroix, 2000). Several sorbent phases for SPME with different characteristics are commercially available and have been used for the analysis of VOO aroma (Table 66.2). The efficiency of SPME for the qualitative and quantitative analysis of VOO volatile compounds was evaluated for the first time by comparing the behavior of four fiber coatings and by identifying the most suitable SPME sampling conditions (Vichi et al., 2003a). Polydimethylsiloxane (PDMS), carboxen-polydimethylsiloxane (CAR-PDMS), polydimethylsiloxane-divinylbenzene (PDMS-DVB) and divinylbenzene-carboxen-polydimethylsiloxane (DVB-CARPDMS) were compared for sensitivity, repeatability and linearity of response, and DVB-CAR-PDMS coating was found to be the most suitable for the analysis of VOO volatiles. Later, this result was also confirmed by comparing PDMS, CAR-PDMS, carbowax-DVB (CW-DVB) and DVB-CAR-PDMS (Cavalli et al., 2003). Sampling at 40 °C during 30 min allowed detection and identification of more than 100 compounds, some of them had not previously been reported in virgin olive oil (Vichi et al.,
2003a). Figure 66.2 reports the chromatographic profile of the same VOO extracted by HS-SPME and CLSA.
66.3.3.1 Specific Applications of SPME in the Analysis of VOO Aroma Among other applications, SPME allowed the characterization of VOOs from different olive varieties and geographical areas (Benincasa et al., 2003; Vichi et al., 2003b; Temime et al., 2006; Manai et al., 2008; Runcio et al., 2008) and the evaluation of processing and storage effects (Servili et al., 2003; Vichi et al., 2003c; Tura et al., 2004; Jiménez et al., 2006). SPME was also applied for the determination of semivolatile compounds of VOO by optimizing the extraction conditions, in particular testing several extraction temperatures (Vichi et al., 2005, 2006). The uptake of less volatile compounds increased at high temperatures because of the improvement of the mass-transfer process from the sample to the headspace. On the other hand, it has to be taken into account that the adsorption of analytes by the fiber coating is an exothermic process, and the partition coefficient decreases by increasing temperature, negatively affecting the adsorption of more volatile analytes. Optimized extraction conditions, with DVB-CAR-PDMS as sorbent during 60 min at 100 °C, allowed the simultaneous determination in VOOs of volatile and semivolatile aromatic hydrocarbons.
619
CHAPTER | 66 Extraction Techniques for the Analysis of Virgin Olive Oil Aroma
2.0 E6
(E)-2-hexenol (Z)-3-hexenol (Z)-3-hexenyl acetate SPME
1-hexanol (E)-2-hexenal
(E,E)-α-farnesene
2.6 E6 (Z)-3-hexenyl acetate
(E,E)-α-farnesene
DHS (CLSA) hexyl acetate (E)-2-hexenal
1-hexanol (Z)-3-hexenol (E)-2-hexenol
8.00
12.00
16.00
20.00
24.00
FIGURE 66.2 Chromatograms of VOO volatile profile obtained by SPME (DVB-CAR-PDMS, 40 °C) and DHS (CLSA mode). The figure shows that SPME sampling allows a higher uptake of LOX-derived alcohols to be obtained while DHS led to higher amounts of esters and semivolatile hydrocarbons.
These environmental contaminants can be especially abundant in oils and fats due to their lipophilic nature, especially in VOOs which lack refining processes (Vichi et al., 2005). Also, an SPME method was developed for the determination of semivolatile sesquiterpene hydrocarbons in VOOs, useful to distinguish samples from different cultivar and geographical origin (Vichi et al., 2006). The 30 sesquiterpenes extracted by DVB-CAR-PDMS at 70 °C during 60 min comprised hydrocarbons not previously documented as present in VOO (Figure 66.3).
66.3.4 Headspace Sorptive Extraction (HSSE) The main limit of SPME is the relatively low capacity given by the small amount of sorbent present on the fiber (0.5– 1 μL). To overcome this problem, Baltussen et al. (1999) introduced the stir bar sorptive extraction (SBSE), a magnetic stir bar coated with a thick film of PDMS the volume of which ranges from 25 to 250 μL. As an extension of SBSE, headspace sorptive extraction (HSSE) was introduced for the extraction of headspace components, and performed by suspending the PDMS stir bar in the vapor phase, in equilibrium or not with the matrix (Bicchi et al., 2000). Only HSSE configuration was applied for the analysis of VOO. After sampling, the stir bar is placed in a glass tube and transferred to a thermo-desorption system from where the analytes are conveyed to the gas chromatographic system. The principles of HSSE are similar to those
of SPME, but with an increased capacity due to the higher amount of sorbent phase. So, the extraction conditions have to be optimized in the same way to allow the highest uptakes of compounds of interest. HSSE was applied for the analysis of VOOs, and compared with SPME, SHS and direct thermal desorption (DTD) (Cavalli et al., 2003). HSSE allowed the determination of all the compounds detected by SPME using a DVB-CAR-PDMS fiber, and a larger number of sesquiterpenes, showing a higher concentration capacity than SPME. Nevertheless, as the introduction of the stir bar into the vial breaks the equilibrium, the best performances are obtained after more than one hour. The optimum extraction time was fixed at 120 min.
66.4 DIRECT THERMAL DESORPTION (DTD) Thermal desorption systems were initially developed for the extraction of volatiles from SBSE stir-bars, and later proposed for direct thermal desorption (DTD). In this technique, the sample is directly placed in the thermal desorption unit without the presence of any adsorbent. The analytes are volatilized by increasing the temperature and stripped by the carrier gas into a cold trap placed in the injector. Differently from the techniques based on the use of sorbents, the recovery of the analytes in DTD is only based on their volatility, avoiding their partition between the sample and the sorbent.
620
SECTION | I Natural Components
1 α-pinene 2 β-pinene 3 sabinene 4 δ3-carene 5 myrcene 6 α-terpinene 7 dl-limonene 8 p-mentha-1,5,8-triene 9 (Z)-b-ocimene 10 γ-terpinene 11 (E)-b-ocimene 12 p-cymene 13 4,8-dimethyl-1,3,7-nonatriene 14 (Z)-alloocimene 15 (E)-alloocimene 16 α-cubebene 17 cyclosativene 18 α-copaene 19 sativene 20 α-cedrene 21 β-cubebene 22 (E)-α-bergamotene 23 β-gurjunene (calarene) 24 β-caryophyllene 25 (Z)-β-farnesene 26 (E)-β-farnesene 27 γ-gurjunene 28 γ-muurolene 29 γ-curcumene 30 β-acoradiene 31 α-selinene 32 eremophyllene 33 α-zingiberene 34 α-muurolene 35 δ-guaiene 36 (E,E)-α-farnesene 37 δ-cadinene 38 β-sesquiphellandrene 39 ar-curcumene 40 n.i. sesquiterpene 41 calamenene 42 n.i. m/z 93, 107, 135, 204 43 n.i. m/z 93, 107, 135, 204 44 β-calacorene 45 n.i. m/z 93, 107, 135, 204 FIGURE 66.3 Extracted ion chromatogram of mono- and sesquiterpene hydrocarbons in VOO, by HS-SPME coupled to gas chromatography/mass spectrometry. Reprinted from Vichi S., et al. J. Chromatogr. A 2006; 1125:117–123, with permission. The figure allows appreciating the sensibility of the SPME method coupled to gas chromatography/mass spectrometry (single ion monitoring) for the analysis of terpene and sesquiterpene hydrocarbons, which are present at trace levels in VOO.
DTD is a rapid, solvent-free technique requiring a very small amount of sample, usually a few microliters, which does not need any manipulation prior to the analysis. DTD has been applied for the characterization of French, Spanish and Italian VOOs (Cavalli et al., 2003; Zunin et al., 2004). Desorption conditions evaluated to develop the analytical methods were related both to the extraction step and
the injection step. They were: final temperature, temperature ramp rate, time, cooling temperature and carrier gas flow. Desorptions at 40 °C and 80 °C during 20 min were chosen for the analysis of VOO volatiles, allowing the detection of a large number of compounds characteristic of VOO (Cavalli et al., 2003; Zunin et al., 2004). As expected, the number and amount of semivolatile sesquiterpenes were proportional
621
CHAPTER | 66 Extraction Techniques for the Analysis of Virgin Olive Oil Aroma
50 SPME
SDE
CLSA
Percentage of areas
40
30
20
10
0 Alcohols
Aldehydes
Ketones
Hydrocarbons
Esters
Terpenoids
FIGURE 66.4 Percentages of VOO’s major families of volatiles extracted by SPME, SDE and CLSA. Reprinted from Vichi S., et al. Food Chem. 1007; 105: 1171–1178, with permission. Better affinities for alcohols and aldehydes are shown by SPME and SDE, respectively, while CLSA gives higher uptakes of esters and hydrocarbons. The best recoveries of terpenoids are obtained by SDE and DHS (CLSA).
to the temperature applied in these studies. A further study was conducted for the determination of these semivolatile terpenoid hydrocarbons in VOO, and the subsequent discrimination of the samples according to the geographical origin (Zunin et al., 2005). When compared with SPME, DTD was more efficient in the extraction of semivolatile sesquiterpenes and esters of fatty acids, while no relevant differences were observed in the recovery of other volatiles (Cavalli et al., 2003). Anyway, it must be taken into account that the extraction temperatures applied in this study were not the same for both methods. In fact, SPME was performed at 25°C, while DTD was performed at 80°C, which is likely to facilitate the volatilization of semivolatile compounds. Recently, a multi-step direct thermal desorption method coupled to comprehensive gas chromatography-time-offlight-mass spectrometry was developed and applied for the characterization of fresh and aged olive oils at typical frying temperatures (de Koning et al., 2008). Different temperatures of desorption were tested to evaluate the compounds present in native VOOs (70°C), in olive oil during frying or baking (175 and 250°C), evidencing differences between fresh and aged olive oil.
66.5 DISTILLATION AND FLUID-BASED EXTRACTION TECHNIQUES 66.5.1 Simultaneous Distillation/Extraction (SDE) and Hydrodistillation Distillation methods have traditionally been applied in the analysis of plant materials. Hydrodistillation and simultaneous distillation/extraction (SDE) were used for this purpose, and SDE appeared to afford the most favorable uptake for mono- and sesquiterpenes, as well as their oxygenated
analogues (Marriott et al., 2001). Hydrodistillation has been applied for the analysis of leaf, fruit and virgin oil volatiles of an Italian olive cultivar (Flamini et al., 2003). The uptake of olive oil volatiles was comparable to that obtained by SPME using a PDMS fiber. The most remarkable difference was the higher amount of α-farnesene obtained by hydrodistillation. However, with hydrodistillation the volatiles in the steam distillate are heavily diluted by water when collected in cold traps. This is overcome in SDE via solvent extraction of the distillate. SDE was evaluated for the analysis of VOO aroma using a modified Likens and Nickerson apparatus, and using pentane and dichloromethane as solvents (Vichi et al., 2007). In comparison with the extraction of the same oil by SPME and CLSA, SDE gave higher percentages of aldehydes correlated with the oxidative degradation of VOO, indicating that the extraction conditions induced the thermal alteration of the oil sample (Figure 66.4). Regarding the extraction of semivolatile compounds, α-zingiberene and α-farnesene were the only compounds detected at a higher percentage by SDE, while other sesquiterpenes were more abundant by CLSA extraction.
66.5.2 Supercritical Fluid Extraction (SFE) Supercritical fluid extraction (SFE) has been widely used for extracting flavor and fragrance compounds from complex matrices without the use of organic solvents. This method is based on the use of gases at determined conditions of pressure and temperature, known as supercritical region. At these conditions gases can acquire solvent properties that are superior to those of liquid solvents. Supercritical CO2 is among the most used fluids due to its good proven extractive capacity and safety, and it was also tested for the extraction of volatile and semivolatile compounds from VOO (Morales et al., 1998). After SFE, a concentration of volatiles on a Tenax trap was required
622
SECTION | I Natural Components
prior to the chromatographic analysis in order to enhance the sensitivity of the method. Different profiles of volatile and semivolatile compounds were found in the SFE extracts, depending on experimental parameters. All the major VOO volatiles were identified in the extract. Soft extraction conditions allowed higher uptakes of the most volatile compounds to be obtained, while drastic conditions led to a higher extraction of semivolatile compounds and induced the oxidation of the sample.
SUMMARY POINTS ●
●
●
●
●
VOO has a unique flavor that distinguishes it from other vegetable oils. Various analytical methods have been developed to examine these volatile compounds. The volatile profile of VOO closely depends upon the method of extraction used. Headspace extraction techniques are the most widely used for the analysis of olive oil aroma, and SHS, DHS, SPME, HSSE and DTD have been applied to VOO, depending on the kind of determination required. Among distillation and fluid-based extraction techniques, only hydrodistillation, SDE and SFE have been tentatively applied for the analysis of VOO aroma.
REFERENCES Angerosa, F., Di Giacinto, L., Solinas, M., 1992. Influence of Dacus oleae infestation on flavor of oils, extracted from attacked olive fruits, by HPLC and HRGC analyses of volatile compounds. Grasas Aceites 43, 134–142. Angerosa, F., Lanza, B., Marsilio, V., 1996. Biogenesis of “fusty” defect in virgin olive oils. Grasas Aceites 47, 142–150. Angerosa, F., Camera, L., d’Alessandro, N., Mellerio, G., 1998. Characterization of seven new hydrocarbon compounds present in the aroma of virgin olive oil. J. Agric. Food Chem. 46, 648–653. Angerosa, F., Basti, C., Vito, R., 1999. Virgin olive oil volatile compounds from lipoxygenase pathway and characterization of some italian cultivars. J. Agric. Food Chem. 47, 836–839. Angerosa, F., Mostallino, R., Basti, C., Vito, R., 2000. Virgin olive oil odour notes: their relationship with volatile compounds from the lipoxygenase pathway and secoiridoid compounds. Food Chem. 68, 283–287. Angerosa, F., 2002. Influence of volatile compounds on virgin olive oil quality evaluated by analytical approaches and sensor panels. Eur. J. Lipid Sci. Technol. 104, 639–660. Aparicio, R., Morales, M.T., Alonso, M.V., 1996. Relationship between volatile compounds and sensory attributes of olive oils by the sensory wheel. J. Am. Oil Chem. Soc. 73, 1253–1264. Arthur, C.L., Pawliszyn, J., 1990. Solid phase microextraction with thermal desorption using fused silica optical fibers. Anal. Chem. 62, 2145–2148. Baltussen, E., Sandra, P., David, F., Cramers, C.A., 1999. Stir bar sorptive extraction (SBSE), a novel extraction technique for aqueous samples: theory and principles. J. Microcol. Sep. 11, 737–747.
Benincasa, C., De Nino, A., Lombardo, N., Perri, E., Sindona, G., Tagarelli, A., 2003. Assay of aroma active components of virgin olive oils from southern Italian regions by SPME-GC/ion trap mass spectrometry. J. Agric. Food Chem. 51, 733–741. Bicchi, C., D’Amato, A., David, F., Sandra, P., 2000. Headspace sorptive extraction (HSSE) in the headspace analysis of aromatic and medicinal plants. J. High Res. Chromatogr. 23, 539–546. Biedermann, M., Grob, K., Morchio, G., 1995. On the origin of benzene, toluene, ethylbenzene and xylene in extra virgin olive oil. Z. Lebensm. Unters. Forsch. 200, 266–272. Cavalli, J.F., Fernandez, X., Lizzani-Cuvelier, L., Loiseau, A.M., 2003. Comparative study of different extraction techniques for the analysis of virgin olive oil aroma. J. Agric. Food Chem. 51, 7709–7716. Cerrato Oliveros, C., Boggia, R., Casale, M., Armanino, C., Forina, M., 2005. Optimisation of a new headspace mass spectrometry instrument. Discrimination of different geographical origin olive oils. J. Chromatogr. A 1076, 7–15. de Koning, S., Kaal, E., Janssen, H.G., van Platerink, C., Brinkman, U.A.T., 2008. Characterization of olive oil volatiles by multistep direct thermal desorption-comprehensive gas chromatography-time-of-flightmass spectrometry using a programmed temperature vaporizing injector. J. Chromatogr. A 1186, 228–235. Dhifi, W., Angerosa, F., Serraiocco, A., Oumar, I., Hamrouni, I., Marzouk, B., 2005. Virgin olive oil aroma: characterization of some Tunisian cultivars. Food Chem. 93, 697–701. Dobarganes-García, M.C., Olías, J.M., González-Quijano, R.G., 1980. Componentes volátiles en el aroma del aceite de oliva virgen. III. Reproducibilidad del método utilizado para su aislamiento, concentración y separación. Grasas Aceites 31, 317–321. Flamini, G., Cioni, P.L., Morelli, I., 2003. Volatiles from leaves, fruits, and virgin oil from Olea europaea Cv. Olivastra Seggianese from Italy. J. Agric. Food Chem. 51, 1382–1386. Jimenez, A., Aguilera, M.P., Beltran, G., Uceda, M., 2006. Application of solid-phase microextraction to virgin olive oil quality control. J. Chromatogr. A 1121, 140–144. Kanavouras, A., Kiritsakis, A., Hernandez, R.J., 2005. Comparative study on volatile analysis of extra virgin olive oil by dynamic headspace and solid phase microextraction. Food Chem. 90, 69–79. López-Feria, S., Cárdenas, S., García-Mesa, J.A., Fernández-Hernández, A., Valcárcel, M., 2007. Quantification of the intensity of virgin olive oil sensory attributes by direct coupling headspace-mass spectrometry and multivariate calibration techniques. J. Chromatogr. A 1147, 144–152. Lorenzo, M.I., Pavón, J.L.P., Laespada, M.E.F., Pinto, C.G., Cordero, B.M., 2002. Detection of adulterants in olive oil by headspace-mass spectrometry. J. Chromatogr. A 945, 221–230. Manai, H., Mahjoub-Haddada, F., Oueslati, I., Daoud, D., Zarrouk, M., 2008. Characterization of monovarietal virgin olive oils from six crossing varieties. Sci. Hortic. 115, 252–260. Marriott, P.J., Shellie, R., Cornwell, C., 2001. Gas chromatographic technologies for the analysis of essential oils. J. Chromatogr. A 936, 1–22. Montedoro, G.F., Bertuccioli, M., Anichini, F., 1978. Aroma analysis of virgin olive oil by head space (volatiles) and extraction (polyphenols) techniques. In: Charalambous, G., Inglett, G. (Eds.), Flavor of Foods and Beverages. Academic Press, New York, pp. 247–281. Morales, M.T., Aparicio, R., Rios, J.J., 1994a. Dynamic headspace gas chromatographic method for determining volatiles in virgin olive oil. J. Chromatogr. A 668, 455–462.
CHAPTER | 66 Extraction Techniques for the Analysis of Virgin Olive Oil Aroma
Morales, M.T., Alonso, M.V., Rios, J.J., Aparicio, R., 1994b. Virgin olive oil aroma: relationship between volatile compounds and sensory attributes by chemometrics. J. Agric. Food Chem. 43, 2925–2931. Morales, M.T., Aparicio, R., Calvente, J.J., 1996. Influence of olive ripeness on the concentration of aroma compounds in virgin olive oil. Flavour Fragrance J. 11, 171–178. Morales, M.T., Rios, J.J., Aparicio, R., 1997. Changes in the volatile composition of virgin olive oil during oxidation: flavours and off-flavours. J. Agric. Food Chem. 45, 2666–2673. Morales, M.T., Berry, A.J., McIntyre, P.S., Aparicio, R., 1998. Tentative analysis of virgin olive oil aroma by supercritical fluid extraction–high-resolution gas chromatography–mass spectrometry. J. Chromatogr. A 819, 267–275. Morales, M.T., Luna, G., Aparicio, R., 2005. Comparative study of virgin olive oil sensory defects. Food Chem. 91, 293–301. Olías, J.M., Pérez, A.G., Ríos, J.J., Sanz, L.C., 1993. Aroma of virgin olive oil: biogenesis of the “green” odor notes. J. Agric. Food Chem. 41, 2368–2373. Pawliszyn, J., 1999. Quantitative aspects of SPME. In: Pawliszyn, J. (Ed.), Applications of Solid Phase Microextraction. Royal Society of Chemistry, Cambridge (UK), pp. 3–21. Page, B.D., Lacroix, G., 2000. Analysis of volatile contaminants in vegetable oils by headspace solid-phase microextraction with carboxenbased fibres. J. Chromatogr. A 873, 79–94. Peña, F., Cárdenas, S., Gallego, M., Valcárcel, M., 2004. Direct screening of olive oil samples for residual benzene hydrocarbon compounds by headspace-mass spectrometry. Anal. Chim. Acta 526, 77–82. Pillonel, L., Bosset, J.O., Tabacchi, R., 2002. Rapid preconcentration and enrichment techniques for the analysis of food volatile. A review. LWT-Lebensm.-Wiss. Technol. 35, 1–14. Runcio, A., Sorgona, L., Mincione, A., Santacaterina, S., Poiana, M., 2008. Volatile compounds of virgin olive oil obtained from Italian cultivars grown in Calabria. Food Chem. 106, 735–740. Servili, M., Conner, J.M., Piggott, J.R., Withers, S.J., Paterson, A., 1995. Sensory characterisation of virgin olive oil and relationship with headspace composition. J. Sci. Food Agric. 67, 61–70. Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G., 2003. Volatile compounds and phenolic composition of virgin olive oil: optimization of temperature and time of exposure of olive pastes to air contact during the mechanical extraction process. J. Agric. Food Chem. 51, 7980–7988. Solinas, M., Angerosa, F., Cucurachi, A., 1987. Connessione tra i prodotti di neoformazione ossidativa delle sostanze grasse e insorgenza del
623
difetto di rancidità all’esame organolettico. Nota II. Determinazione quantitativa. Riv. Ital. Sost. Grasse. 64, 137–145. Temime, S.B., Campeol, E., Cioni, P.L., Daoud, D., Zarrouk, M., 2006. Volatile compounds from Chétoui olive oil and variations induced by growing area. Food Chem. 99, 315–325. Tura, D., Prenzler, P.D., Bedgood, D.R., Antolovich, M., Robards, K., 2004. Varietal and processing effects on the volatile profile of Australian olive oils. Food Chem. 84, 341–349. Vichi, S., Castellote, A.I., Pizzale, L., Conte, L.S., Buxaderas, S., LopezTamames, E., 2003a. Analysis of virgin olive oil volatile compounds by headspace solid-phase microextraction coupled to gas chromatography with mass spectrometric and flame ionization detection. J. Chromatogr. A 983, 19–23. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., Lopez-Tamames, E., 2003b. Solid-phase microextraction in the analysis of virgin olive oil volatile fraction: characterization of virgin olive oils from two distinct geographical areas of northern Italy. J. Agric. Food Chem. 51, 6572–6577. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., Lopez-Tamames, E., 2003c. Solid-phase microextraction in the analysis of virgin olive oil volatile fraction: modifications induced by oxidation and suitable markers of oxidative status. J. Agric. Food Chem. 51, 6564–6571. Vichi, S., Pizzale, L., Conte, L.S., Buxaderas, S., Lopez-Tamames, E., 2005. Simultaneous determination of volatile and semi-volatile aromatic hydrocarbons in virgin olive oil by headspace solid-phase microextraction coupled to gas chromatography/mass spectrometry. J. Chromatogr. A 1090, 146–154. Vichi, S., Guadayol, J.M., Caixach, J., López-Tamames, E., Buxaderas, S., 2006. Monoterpene and sesquiterpene hydrocarbons of virgin olive oil by headspace solid-phase microextraction coupled to gas chromatography/mass spectrometry. J. Chromatogr. A 1125, 117–123. Vichi, S., Guadayol, J.M., Caixach, J., López-Tamames, E., Buxaderas, S., 2007. Comparative study of different extraction techniques for the analysis of virgin olive oil aroma. Food Chem. 105, 1171–1178. Zunin, P., Boggia, R., Lanteri, S., Leardi, R., De Andreis, R., Evangelisti, F., 2004. Direct thermal extraction and gas chromatographic-mass spectrometric determination of volatile compounds of extra-virgin olive oils. J. Chromatogr. A 1023, 271–276. Zunin, P., Boggia, R., Salvadeo, P., Evangelisti, F., 2005. Geographical traceability of West Liguria extravirgin olive oils by the analysis of volatile terpenoid hydrocarbons. J. Chromatogr. A 1089, 243–249.
This page intentionally left blank
Chapter 67
Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil Lorenzo Cerretani and Alessandra Bendini Università di Bologna, Alma Mater Studiorum, Dipartimento di Scienze degli Alimenti, Cesena (FC), Italy
67.1 GENERAL CONCEPTS RELATING TO ANTIOXIDANT ACTIVITY OF PHENOLIC MOLECULES Phenols (PH) can slow lipid oxidation, inhibiting initiation and propagation steps, by inactivating or scavenging free radicals. By acting as free radical scavengers or chainbreaking antioxidants, PH in virgin olive oil (VOO) are able to accept a radical from oxidizing lipid species such as peroxyl (LOO•) and alkoxyl (LO•) radicals through the following reaction:
capable to react, donating a hydrogen to it, with a feasible reaction from a kinetic point of view (Decker, 2002). PH can lose antioxidant activity if present in a lipid substrate (LH) at concentrations that are too high, and also behave as pro-oxidants by involvement in initiation reactions schematized as the following reactions (Shahidi and Naczk, 2004): P• ⫹ O2 → POO• POO• ⫹ LH → POOH ⫹ L•
LOO• / LO• ⫹ PH → LOOH/LOH ⫹ P• Hydrogen donation generally occurs through the hydroxyl group of PH, and the subsequently formed radical is stabilized by resonance delocalization throughout the phenolic ring structure. A second hydroxy group, especially at the ortho position, stabilizes the phenoxyl radical through an intramolecular hydrogen bond; thus phenols with a catechol moiety, known as o-diphenols, such as hydroxytyrosol, decarboxymethyl oleuropein aglycon and oleuropein aglycon, are the most effective antioxidants in VOO (Bendini et al., 2007a). An o-quinol group present at the aromatic ring of PH also has metal-chelating abilities, and consequently the chelation of metal ions (as copper or iron) renders them catalytically inactive against lipid oxidation. The antioxidant efficiency of PH of VOO is dependent on the ability of the antioxidant molecule to donate hydrogen to the free radical, such that the transfer of the hydrogen to the free radical is more rapid and energetically favorable as the hydrogen bond energy of the PH decreases. A PH (for example catechol E°⬘ ⫽ 530 mV) having a reduction potential lower than the reduction potential of a radical (for example peroxyl radical E°⬘ ⫽ 1000 mV) is Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
P• ⫹ LH → PH ⫹ L• These reactions do not occur in VOO since phenols are micro-components present in variable quantities, generally between 0.05% and 0.5%, according to the product. These reactions could however take place when a natural or synthetic antioxidant is introduced, in quantities not adequately controlled, into a lipid substrate to prevent oxidation. Another general concept to keep in mind, which has important implications for the antioxidant activity, is related to the distribution in lipidic substrates of different molecules. In fact, natural antioxidants exhibit complex properties between air–oil and oil–water interfaces that significantly affect their relative activities in different lipidic systems. The presence of hydrophilic phenolic compounds in VOO, and their high antioxidant activity, can be explained by the so-called ‘polar paradox’ (Porter et al., 1989; Frankel, 1996). According to this, ‘polar antioxidants are more effective in non-polar lipids, whereas non-polar antioxidants are more active in polar lipid emulsions’. The polar paradox is based on the fact that in a bulk oil system hydrophilic antioxidants, such as polar phenols, are oriented in the air–oil interface (a low quantity of air is always
625
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
626
trapped in the oil) and become more protective against oxidation than the lipophilic antioxidants, like tocopherols, which remain in solution in the oil.
SECTION | I Natural Components
metabisulfite and sulfite) interfere with the Folin–Ciocalteu reactive to give apparently elevated phenolic concentrations.
67.2.1.2 ABTS and DPPH Tests
67.2 GENERAL CONCEPTS ON DIRECT AND INDIRECT METHODS TO DETERMINE THE ANTIOXIDANT ACTIVITY OF PHENOLS IN VIRGIN OLIVE OIL There are many approaches to determine the antioxidant activity of the PH present in VOO. For simplicity, these can be divided into two groups: indirect and direct methods (Roginsky and Lissi, 2005). Whereas the former measure the ability of PH to block free radicals that are not linked to oxidative degradation, the latter are closely associated with the study of the effect of PH with respect of lipid peroxidation. Generally, indirect methods are used more frequently than direct methods, and each type of method has advantages and disadvantages. Direct methods are more satisfactory in principle and are generally more sensitive; on the other hand, they are more time-consuming and more difficult experimentally.
67.2.1 Indirect Methods Indirect methods commonly provide information on the ability of PH to scavenge stable free radicals such as DPPH and ABTS or to reduce a metal in trace such as FRAP and CUPRAC. One widely applied indirect method is the Folin–Ciocalteu test, which allows estimation of the reductive capacity of phenolic molecules (previously extracted by VOO). The Folin–Ciocalteu test is a well-standardized indirect method that is useful for routine estimation of antioxidant activity of PH in VOO. It is also possible to use methods based on chemiluminescence to assess the antioxidant capacity of PH.
67.2.1.1 Folin-Ciocalteu Assay The reactive is a mixture of tungstate and molybdate in a basic medium that is able to oxidize PH through formation of molybdenum oxide, which has an intense absorbance around 750 nm. PH determined by the Folin–Ciocalteu test are most frequently expressed in gallic acid equivalents, even if for VOO the expression with respect to other standards such as tyrosol, 3,4-dihydroxyphenylacetic acid or caffeic acid is also utilized. Similar to the DPPH or ABTS assays, the reaction with Folin–Ciocalteu is not selective due to contemporary determination of all kinds of phenolic molecules (phenols and o-diphenols) in VOO extract. A wide variety of substances (particularly sugars, proteins, aromatic amines, sulfur dioxide, ascorbic acid and other enediols and reductones, organic acids, Fe2⫹, sodium
These two radicals (ABTS•⫹ and DPPH•) may be neutralized either by direct reduction via electron transfers or by radical quenching via hydrogen atom transfer (an acid medium facilitates the mechanism by electron transfer) (Prior et al., 2005). The ABTS method is based on spectrophotometric monitoring of the decay of the radical-cation ABTS•⫹ produced by the oxidation of 2,2⬘-azinobis(3ethylbenzothiaziline-6-sulfonate) (ABTS) caused by the addition of PH. In fact, ABTS•⫹ has strong absorption at longer wavelengths (with absorption maxima at 415, 645, 734, and 815 nm), but 734 nm is adopted by most investigators. In the absence of PH, ABTS• is rather stable, but reacts energetically with molecules able to donate hydrogen atoms or electrons, such as phenolic compounds, leading to a disappearance of the blue/green color of this radical. Thermodynamically, a compound can reduce ABTS•⫹ if it has a redox potential lower than that of ABTS (680 mV). Trolox (hydrosoluble analogue of α-tocopherol) equivalents are most frequently used to express the results (TEAC, Trolox Equivalent Antioxidant Capacity). Several modifications of ABTS protocols are used depending on how ABTS•⫹ is generated. Classically, it was reacted with the ferrymyoglobin radical produced, followed by the monitoring of metamyoglobin and hydrogen peroxide in the presence of peroxidase. Later, AAPH or ABAP was used, while more recently potassium persulfate has been adopted. Generally, whereas enzyme generation is faster and the reaction conditions are milder, the chemical generation requires a long time (e.g., up to 16 h for potassium persulfate) or high temperatures (e.g., 60 °C for ABAP). ABTS•⫹ can be solubilized in both aqueous and organic solvents, and thus multiple media can be used to determine both hydrophilic and lipophilic antioxidants. Most applications of this assay on VOO samples are carried out to test the antioxidant activity of the phenolic extracts, and so ABTS is solubilized in polar solvents. The 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay measures the scavenging capacity of this radical dissolved in different solvent mixtures (i.e., methanol, methanol/water), induced by PH previously extracted from VOO using spectrophotometric determination at 515–517 nm. The DPPH radical is one of the few stable organic nitrogen radicals, which produces solutions colored deep purple. It is commercially available, and thus does not have to be previously generated as for ABTS. With many single PH or phenolic extracts of VOO, the reaction with ABTS•⫹ or DPPH• occurs rather slowly. Therefore, the results of these tests depend on the time of incubation to reach a steady state and on the ratio of sample quantity to radical concentrations. Poor selectivity of
CHAPTER | 67 Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil
ABTS•⫹ in the reaction with hydrogen-atom donors is one limitation of this method, and is also reduced by aromatic OH-groups that do not significantly contribute to antioxidation. In contrast, DPPH• is likely to be more selective than ABTS•⫹ in the reaction with hydrogen-donors, and, for example, does not react with aromatic acids containing only one OH-group. One of the major disadvantages is steric accessibility, since small molecules have better access to the radical site and can show higher apparent antioxidant activity. The hydrogen-donating potential of a sample evaluated by the DPPH test is most frequently expressed as 1/EC50, where the concentration that causes a decrease in the initial DPPH concentration by 50% is defined as EC50. Recent automated versions combine the DPPH test with an HLPC assay (Bandoniene and Murkovic, 2002; Polasek et al., 2004). In this case, HPLC analysis of a phenolic extract with detection at 280 nm is immediately followed by monitoring the decay of DPPH at 515 nm within a single compound corresponding to each HPLC peak.
67.2.1.3 FRAP and CUPRAC Tests The ferric reducing antioxidant power (FRAP) mechanism is based on electron transfer rather than hydrogen atom transfer (Prior et al., 2005). The FRAP assay is based on the ability of PH to reduce Fe3⫹ to Fe2⫹. The FRAP reaction is conducted at acidic pH 3.6 to maintain iron solubility, so the reaction at low pH decreases the ionization potential that drives hydrogen atom transfer and increases the redox potential, which is the dominant reaction mechanism. When the reduction of Fe3⫹ to Fe2⫹ occurs in the presence of 2,4,6-trypyridyl-s-triazine, the reaction is accompanied by the formation of a colored complex with Fe2⫹ (absorption at 593 nm). The reducing power appears to be related to the degree of hydroxylation and extent of conjugation in PH. Because the reaction detects compounds with redox potentials of ⬍700 mV, which is comparable with that of ABTS•⫹ (680 mV), similar compounds react in both the TEAC and FRAP assays. FRAP cannot detect compounds that act by radical quenching (hydrogen transfer), particularly thiols (as glutathione) and proteins. However, FRAP is simple, rapid (generally 4–6 min), inexpensive and can be performed using semiautomatic or automated protocols. A variant of the FRAP assay using Cu instead of Fe is known as CUPRAC (copper reduction assay), which is based on the reduction of Cu2⫹ to Cu⫹ by the combined action of reducing agents in the sample. Bathocuproine (2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline) or neocuproine (2,9-dimethyl-1,10-phenanthroline) are used to form chromophores with Cu⫹ that absorb at 490 nm or 450 nm, respectively. CUPRAC values are generally comparable to TEAC values for PH. The low redox potential of copper both in the free and complexed form makes it more selective in reactions than iron, and can also indicate the potential pro-oxidant activity of PH.
627
67.2.1.4 Chemiluminescence Methods Chemiluminescence (CL) is defined as the production of ultraviolet, visible or infrared radiation when a chemical reaction yields an electronically excited intermediate or product that either luminesces or donates its energy to another molecule, which then luminesces. Analytical methods based on chemiluminescence have several advantages: (1) high sensitivity, as same analyte in CL has lower detection limits than absorption or fluorescence; (2) good selectivity, based on the fact that the analyte of interest often produces its signal even in the presence of interfering compounds; and (3) a wide linear detection range, permitting the analysis of samples with a large concentration range (Navas and Jimenez, 2007). The general principle of antioxidant evaluation by chemiluminescence methods is based on the ability of chemiluminescent probes (such as luminol or lucigenin) that react with reactive oxygen species producing a very bright green emission that decays to greenish blue and finally blue. The addition of a free radical scavenger such as phenols extracted by VOO results in quenching of chemiluminescence that corresponds with a pronounced induction period. The results are expressed as Trolox equivalents. CL protocols proposed by several researchers differ in exciting with lucigenin or luminol (i.e., by hydrogen peroxide, AAPH, the xanthine-hypoxantine-oxidase system or by a photosensitized reaction). Chemiluminescence shows promise for routine testing due to its high productivity and relative feasibility, although to date only a limited number of publications using the CL test have been applied to the evaluation of antioxidant activity of PH of VOO.
67.2.2 Direct Methods The first important choice concerns the lipidic substrate in which the antioxidant activity of PH is tested. Preference should be given to individual lipids such as methyl linoleate, linoleic acid or trilinolein because their commercial accessibility provides reproducibility (CarrascoPancorbo et al., 2005). Moreover, these compounds are relatively inexpensive, and their oxidation is representative of the most essential features of lipid peroxidation. Another possibility is the use of a VOO in which its antioxidants and pro-oxidant components were eliminated through washing with appropriate solvent mixtures. The initiation of lipid peroxidation can be started by using an initiator, such as a thermolabile azo-compound (i.e., water soluble AAPH or lipid-soluble AMVN), or a transition metal (i.e., Fe3⫹ or Cu2⫹). Monitoring of lipid peroxidation is possible through two means: discontinuous, via collection of several aliquots, or continuous. This latter is preferable because it allows both a higher throughput and more detailed observation of the process.
628
The most popular determinations are based on the quantification of the primary products of oxidation (i.e., through the measure of peroxides and of conjugated dienes), of secondary products (i.e., through the TBARS assay or of specific volatile aldehydes as hexanal or nonanal) or of oxygen consumption. Measurement via oxygen consumption seems to be more representative of the oxidation development and is also quite sensitive. Studies on lipid matrices enriched in PH or VOO can be performed under normal storage conditions or under accelerated oxidation such as active oxygen method (AOM), Schaal oven test, oxygen uptake/absorption, or by using a fully automated oxidative stability instrument (Oxidative Stability Instrument or Rancimat apparatus). Automatic assays of lipid oxidation in bulk by using enzymatic sensors to record the consumption of phenolic components have been demonstrated (Campanella et al., 1999; Capannesi et al., 2000).
67.2.2.1 Methods Based on Competitive Bleaching Along with the direct methods based on the studies of chain lipid peroxidation, there are several methods based on competition between PH and a reference free radical scavenger for the peroxyl radical. The more widely used are the oxygen radical absorbance capacity (ORAC) test and the total radical-trapping antioxidant parameter (TRAP) assay: in early studies, a fluorescent natural protein β-phycoerythrin (PE) was applied as a target of free radical attack, whereas newer fluorescent probes such as fluorescein or dichlorofluorescein are preferred due to their high stability and low reactivity. The intensity of fluorescence of a probe decreases with time under the flux of the peroxyl radical. The peroxyl radicals are formed by thermolysis of AAPH in aqueous buffer, and when in the presence of a tested sample containing chain-breaking antioxidants, the decay of fluorescence is retarded. The differences between these two methods lie in the shapes of the observed kinetic curves and in the way used to calculate antioxidant activity from the kinetic curves. The reason for this discrepancy is the difference in the initial concentrations of the fluorescent probe. Both TRAP and ORAC are expressed in Trolox equivalents. The ORAC and TRAP assay can be adapted to detect both hydrophilic and hydrophobic antioxidants by altering the radical source and solvent. The major criticisms of these tests are the need for a fluorometer, which may not be routinely available in analytical laboratories, and the long analysis time (1 h). However, some researchers (Guo et al., 1997; Caldwell, 2000) have carried out the ORAC test using an automated device that allows the contemporary analysis of large number of samples. ORAC has recently been proposed as a quality index of VOO since it measures the efficiency of phenolic compounds previously extracted by samples in the protection against peroxyl radicals.
SECTION | I Natural Components
Crocin is a natural compound with extremely strong absorbance in the visible range that undergoes bleaching under attack of the peroxyl radical. The addition of a sample containing molecules as PH that donate hydrogen atoms to quench radicals results in a decrease in the rate of crocin decay. Crocin is preferred to β-carotene because it bleaches only by the radical oxidation pathway. Color loss is followed optically at 443 nm, so that the reaction requires no special instrumentation, although the control of temperature is a critical aspect. Because of the need to calculate the IC50, multiple dilutions of the same sample need to be run. A modified microplate-based version suitable for routine determinations was described by Lussignoli et al. (1999).
67.2.2.2 Methods Based on Automated Oxidation Under Accelerated Conditions: Rancimat or OSI Test The extension of the induction period of the oxidation by addition of a PH or a phenolic extract to a lipid matrix is related to its antioxidant efficacy, which is sometimes expressed as antioxidant index or protection factor. Due to the high oxidative stability of VOO (that usually ranges from 9 to more than 18 months), several researchers have employed accelerated methods to estimate the induction period of the oxidation reaction in a relatively short period of time (from hours to few days). It is well known that a good relationship between the oxidative stability of VOO exists, determined using accelerated oxidation tests (i.e., AOM or Rancimat or OSI), and its initial content of natural antioxidants, especially phenolic compounds. Rancimat or OSI (Oxidative Stability Instrument) tests working under temperatures higher than room temperature (tested from 90 °C to 140 °C) and under a forced air flow (from 7 L h⫺1 to 24 L h⫺1) are widely employed to shorten the analysis time and have gained acceptance because of their ease of use and reproducibility (Mateos et al., 2006). In these tests, an oil sample (from 3.0 to 5.0 g ⫾ 0.1 g) is heated at a selected temperature under atmospheric pressure, and air is bubbled through the oil so that short-chain volatile acids (secondary oxidation products) are produced, recovered and measured conductimetrically in distilled water. The time required (i.e., known as OSI time) to produce a sudden increase in the conductivity due to volatile acid formation determines an induction period (IP), which can be defined as a measure of the stability of the oil. Several critical aspects of these tests have been highlighted: (1) the high temperature applied has more influence on secondary reactions of polymerization that normally are negligible in the process of oxidation at room temperature; (2) the achievement of the endpoint of the period of induction is at a level of oxidation higher than the room temperature (also when the increase of temperature a decrease of oxygen solubility in oil occurs); (3) it underestimates antioxidant activity from thermolabile molecules (Frankel, 1993). On
CHAPTER | 67 Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil
629
the basis of these observations, several researchers recognize the validity of these accelerated tests for comparisons of the oxidative stability of various VOO or to determine the antioxidant effectiveness of individual molecules or phenolic extracts added to a system lipid model, but not for extrapolating indicative values of the VOO shelf-life stored at room temperature (Gomez-Alonso et al., 2007).
processes: first, the polymer film provides extraction and pre-concentration of the antioxidant, and secondly the immobilized radical reacts with the antioxidant. The major advantages of this method are the possibility to avoid pretreatment of samples, and that it can be used with colored or turbid samples.
67.2.2.3 Methods Based on Electrochemical or Enzymatic Sensors
67.3. RECENT APPLICATIONS OF DIFFERENT METHODS FOR THE EVALUATION OF ANTIOXIDANT ACTIVITY IN VIRGIN OLIVE OIL
Cyclic voltammetry (CV) appears to be rather informative and promising. The electrochemical oxidation of compounds on an inert carbon glassy electrode is accompanied by the appearance of the current at a certain potential. While the potential at which a CV peak appears is determined by the redox properties of the compound tested, the value of the current reflects the quantity of the compound. In addition, the compound may be characterized by the reversibility of the CV traces recorded when the potential is altered in both directions. Since the oxidation potential of a compound depends on the energy required to donate an electron, the lower the oxidization potential, the more easily the compound will donate an electron, and the higher its expected antioxidant activity. Sensors represent a promising tool to complement various analytical methods. Among these, biosensors are one class of liquid sensors. The major advantages that make them useful in the rapid screening of a large number of samples are low cost, short analysis time per sample, small size and high selectivity. In particular, for the estimation of PH several authors proposed the use of biosensors based on the enzyme tyrosinase. Tyrosinase (phenol oxidase) is a bi-functional enzyme that has two types of activities: a hydroxylase activity that converts monophenols to o-diphenols, and an oxidase activity through which o-diphenols are oxidized to o-quinones. The o-quinones are electrochemically active and can be reduced back to the catechol form by applying a low potential, so that the resultant cathodic current is related to the amount of PH present in the solution. Enzymes with lower specificity for phenolic compounds, such as peroxidases, are also used. To overcome some of the problems connected with analysis of PH in VOO, the immobilization of tyrosinase through physical entrapment in a protective biocompatible polymer matrix onto screen-printed electrodes has been proposed. Optical sensors based on immobilized chemical compounds are particularly suited to rapid and low-cost screening applications since they have good sensitivity and, at the same time, can be provided in the form of inexpensive test-kits and strips. In these applications, polymer films containing immobilized radicals respond to antioxidants by changing color irreversibly, and the membrane absorption maxima can be followed with a spectrophotometer. The kinetics of the reaction between the immobilized radical and an antioxidant is the result of two consequential
It has been recently demonstrated (Servili and Montedoro, 2002) that the value of the antiradical properties of VOO, evaluated with different ORAC methods, is significantly correlated with its total phenolic and o-diphenolic components as well as with the induction time obtained with the Rancimat test. In particular, the same group (Ninfali et al., 2002) analyzed 33 VOO for phenolics, o-diphenolics, tocopherols, β-carotene and lutein determining both the ORAC assay and the induction period (IP) by the Rancimat test. The ORAC assay was initiated by adding AAPH to β-phycoerythrin in buffer at pH 7, whereas the Rancimat test was developed at 120 °C and under an air flow of 20 L h⫺1. The authors established that the main difference among oils was related mainly to the phenolic content, and in particular, to o-diphenols. Both ORAC and IP values depended markedly on the phenolic pool, even though the Rancimat value takes into account the entire oil, whereas the ORAC value considers any of the hydrophilic phenolic extracts. It was concluded that both methods may be considered useful for evaluating the stability of VOO against oxidation. Gorinstein et al. (2003) found very high correlations between total phenols of five VOO by Folin–Ciocalteu and different antiradical assays (TRAP, DPPH•, ABTS•⫹, β-carotenelinoleate model system). According to these authors, the best method for determination of the antioxidant capacity of VOO is the β-carotene test, which showed the highest correlation with total phenols (R2 ⫽ 0.9958). In 2007, Samaniego Sánchez and co-workers confirmed a good correlation between the total phenols by the Folin– Ciocalteu method and the antioxidant capacities of 39 VOO measured by four different methods (ABTS•⫹, DPPH•, ORAC, β-carotene-linoleate model system). All the methods tested were found to be suitable for determining the antioxidant capacity of VOO, but the highest correlation was seen with the ABTS•⫹ test (R2 ⫽ 0.9905) (Table 67.1). Paiva-Martins and co-authors (2006) investigated the antioxidant or pro-oxidant activity of individual phenolic compounds in VOO (hydroxytyrosol, decarboxymethyl oleuropein aglycon, oleuropein aglycon and oleuropein), in olive oil and in oil-in-water emulsions oxidized in the dark at 60 °C. In particular, the ability of these phenolic compounds
630
TABLE 67.1 Most diffuse methods for evaluation of antioxidant activity in virgin olive oil and range of measures in applicative works. Principle of method
Sample preparation
Observations
Range of measure in VOO
References
Folin-Ciocalteu
Reduction of phenol compounds by a mixture of tungstate and molibdate in basic medium
Extraction of phenolic compounds in polar solution methanol/water
A number of substances interfere with the assay, and may also react with the reagent solution to give apparently elevated phenolic concentrations
50–1000 mg of gallic acid/kg of oil
Montedoro et al., 1992
o-diphenols
Colorimetric quantification of o-diphenols by reaction with molybdate
Method selective for the main phenols responsible for oxidative stability of VOO, such as hydroxytyrosol and its oleosidic forms. Proposed as official method to International Olive Oil Council
10–350 mg of gallic acid/kg of oil
Bendini et al., 2007a,b
DPPH•⫹
Estimates the radical scavenging activity of phenolic extract toward the synthetic 2,2-diphenyl-1picrylhydrazyl radical (DPPH). The absorbance decreasing at 515–517 nm is measured following the addition of antioxidant extracts to the DPPH solution
DPPH• and ABTS•⫹ radicals are among the most popular spectrophotometric methods for determination of the antioxidant capacity of food, beverages and vegetable extracts. The absorbance reduction shows a different kinetic trend for each phenolic compound, and this information is lost in the final measurement
200–590 as μmol of Trolox/L of oil (TEAC)
Samaniego Sánchez et al., 2007
70–700 as μmol of Trolox/kg of oil (TEAC)
Bendini et al., 2006a
This test is based on the decrease in absorbance at 734 nm following the addition of antioxidant extracts to a solution of the radical of 2,2’-azinobis-(3ethylbensothiazoline)-6-sulfonic acid (ABTS)
560–1000 as μmol of Trolox/L of oil (TEAC)
Samaniego Sánchez et al., 2007
Scavenging activity is correlated with reference to the Trolox
110–1170 as μmol of Trolox/kg of oil (TEAC)
Bendini et al., 2006b
200–1550 as μmol of Trolox/L of oil (TEAC)
Manna et al., 2002
ABTS•⫹
FRAP
The Ferric Reducing Antioxidant Power assay (FRAP) is based on the ability of phenols to reduce Fe(III) to Fe(II). The test is conducted in acidic medium to maintain iron solubility
In contrast to other tests of total antioxidant power, the FRAP assay is simple, rapid, inexpensive, robust and does not require specialized equipment. The FRAP assay can be performed using automated, semiautomatic, or manual methods
SECTION | I Natural Components
Type of analysis
The general principle of chemiluminescence methods is based on the ability of chemiluminescent reagents (such as luminol or lucigenin) to react with reactive oxygen species producing a radiation emission. The addition of a free radical scavenger as phenols results in chemiluminescence quenching
Direct analysis of VOO or analysis of phenolic extracts
Analytical methods based on chemiluminescence have interesting advantages: high sensitivity, selectivity and a wide linear detection range
0.06–5.60 μmol of Trolox/L of extract
Ferri et al., 2006
ORAC
The oxygen radical absorbance capacity (ORAC) assay is based on the intensity of fluorescence of a probe (β-phycoerythrin, fluorescein or dichlorofluorescein) and decreases with time under the flux of the peroxyl radical formed by thermolysis of APPH in aqueous buffer. In presence of chain-breaking antioxidants, the decay of fluorescence is retarded
Extraction of phenolic compounds in polar solution methanol/water
The major criticisms of these tests are the need for a fluorometer, which may not be routinely available in analytical laboratories and the long analysis time (1 h), but some researchers have automated the ORAC test
1000–18 000 μmol of Trolox/kg of oil
Ninfali et al., 2002, 2008
OSI and RANCIMAT
Oxidative Stability Instrument (OSI) or Rancimat tests necessitate accelerated oxidation. In these tests, an oil sample is heated at selected temperature under atmospheric pressure and the air is bubbled through the oil, so that short-chain volatile acids (secondary oxidation products) are produced, recovered and measured conductimetrically in distilled water. The time required to produce a sudden increase of the conductivity determines an induction period (IP for Rancimat or OSI time), which can be defined as a measure of the stability of the oil
Direct analysis of VOO
These methods are widely employed to shorten the analysis time and have gained acceptance because of their ease of use and reproducibility. Several critical aspects of these tests are based on the employ of high temperature and air flow for acceleration of lipid oxidation
7.9–65.0 h (OSI at 110 °C and air flow of 9 L/h)
Bendini et al., 2006b; Cerretani et al., 2008; Aparicio et al., 1999
10.8–78.7 h (Rancimat at 100 °C and air flow of 10 L/h)
Electrochemical
The electrochemical oxidation of a certain compound Direct analysis of on an inert carbon glassy electrode is accompanied VOO or analysis of by the appearance of the current at a certain potential. phenolic extracts While the potential at which a cyclic voltammetry peak appears to be determined by the redox properties of the tested compound, the value of the current shows the quantity of this compound
Sensors are a promising tool to complement various analytical methods. The electrochemical procedures working directly on virgin olive oil diluted in organic solvent are easy and quick.
5–220 mg of quercetin/ kg of oil
Del Carlo et al., 2004; Bendini et al., 2007a
Enzymatic sensors
A biosensor (tyrosinase- or peroxidase-based) constructed using disposable screen-printed carbon electrodes with the enzyme as biorecognition element. The sensor is coupled with a simple extraction procedure and optimized for use in flow injection analysis
The major advantages of biosensor assays are the rapid screening of a large number of samples, low costs and short analysis time per sample, small size and selectivity. To analyze phenols directly in VOO, it is possible to immobilize tyrosinase in a protective biocompatible polymer matrix onto screen-printed electrodes
10–230 μmol of catechol /L of oil
Busch et al., 2006
Analysis of aqueous phenolic extracts
CHAPTER | 67 Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil
Chemiluminescence
631
632
other purported antioxidants do not contribute significantly to VOO stability in the presence of phenols. In 2003, Mateos and co-authors spiked a purified olive oil with several single phenols and tocopherols (with concentration ranges similar to those found in VOO) to evaluate their antioxidant or pro-oxidant activities under accelerated oxidation in a Rancimat apparatus at 100 °C. The data were in agreement with the fact that antioxidant activity is correlated with the number of phenolic hydroxyls in the molecule and mainly the o-disubstitution. In fact, hydroxytyrosol, hydroxytyrosyl acetate and oleuropein aglycon showed similar antioxidant activities per mmol of substance, whereas the activity of α-tocopherol was significantly lower than that of similar concentrations of these phenols. There was only a slight contribution of tyrosol to oil stability. The flavone luteolin showed an antioxidant activity similar to that of hydroxytyrosol, whereas apigenin did not show any effects. In investigations carried out by Carrasco-Pancorbo and co-authors (2005), the antioxidant activity of several single phenolic compounds of VOO (hydroxytyrosol, tyrosol, elenolic acid, decarboxymethyl oleuropein aglycon, (⫹)pinoresinol, (⫹)-1-acetoxypinoresinol, oleuropein aglycon and ligstroside aglycon) was evaluated by three different chemical approaches: radical assay (DPPH) reported in Figure 67.1, accelerated oxidation in a lipid model system by OSI in Figure 67.2 and an electrochemical method (flow injection analysis FIA amperometry and cyclic voltammetry). These authors verified that, as is generally assumed, the presence of a single hydroxyl group on the benzene ring conferred only limited antioxidant activity. On the other hand, the presence of a catechol moiety enhances the ability of the phenolic compounds to act as antioxidants. In
30
218 ppm 50 ppm
25 20 15 μmol TEACg−1
to inhibit lipid oxidation in bulk oil and emulsions in the presence and absence of copper were studied using the copper reduction assay (CUPRAC) and the DPPH test. In 2004, Triantis et al. described a sensitive and simple procedure for measuring the antioxidant activity of aqueous methanolic extracts of VOO. This assay was based on the use of the chemiluminescence of lucigenin and alkaline hydrogen peroxide. It was demonstrated that this method was able to determine the total phenols in VOO, because the phenol contents of extracts were similar to those measured by the Folin–Ciocalteu assay, and the reduction in light signals of lucigenin was proportional to antioxidant activity and linked to phenolic content. In 2006, Ferri and co-authors presented the preliminary results comparing various luminescent-based assays to VOO. These different tests, which work in an aqueous or lipophilic environment, permitted the collection of data on VOO stability and qualitative properties in terms of antioxidant capacity. Samples of VOO, produced from different cultivars, were directly analyzed or separated, by extraction, in their hydrophilic and lipophilic fractions that were analyzed independently. The total antioxidant capacity (TAC) of the hydrophilic fraction containing PH was determined by applying the luminol/H2O2/HRP inhibition assay. Hydrophilic PH were also determined spectrophotometrically using the Folin– Ciocalteu reagent. On the lipophilic fraction, as well as on the whole VOO, the evaluation of the spontaneously formed hydroperoxides was attempted by measuring the luminol emission in the presence of cytochrome c, which acts as a heme catalyst of hydroperoxide degradation to radical species. Extensive peroxidation of samples was induced by the addition of potassium superoxide and evaluated by measuring the low level of chemiluminescence of oxidation products. The H2O2 scavenging ability of VOO and its fractions was determined by applying a peroxyoxalate assay, using 9,10 diphenylantracene as a fluorophore. The contribution of spontaneous light emission of VOO samples to the luminescence measured in various assays was always determined, and in most cases was not negligible. Good correspondence between the total phenols content and the TAC values measured by luminol was not confirmed. According to their results, the H2O2 scavenging ability could be mainly ascribed to the lipophilic fraction, being negligible for the hydrophilic components. Various researchers have studied the importance of the total or individual phenol contents with regards to VOO stability. The active phenols in VOO are mainly o-diphenols such as hydroxytyrosol and its oleosidic forms. However, the monophenol tyrosol and its oleosidic and derivative forms show less antioxidant activity. In 1999, Aparicio et al. used Rancimat (set to 100 °C) to study the relative contribution of several chemical compounds (phenols, carotenoids, tocopherols, chlorophylls) to the oxidative stability of 79 VOO. It was observed that the main components protecting VOO against oxidation are o-diphenols, and the
SECTION | I Natural Components
10 5 4 3 2 1 0 a
b
c
d e Compounds
f
g
h
FIGURE 67.1 DPPH test for eight isolated phenolic compounds at two different concentrations (218 and 50 mg kg⫺1 of each compound), from Carrasco-Pancorbo et al. (2005). Compounds: a, hydroxytyrosol; b, tyrosol; c, elenolic acid; d, deacetoxy oleuropein aglycon; e, (⫹)-pinoresinol; f, (⫹)-1-acetoxypinoresinol; g, oleuropein aglycon; h, ligstroside aglycon.
CHAPTER | 67 Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil
fact, the results obtained with all three different approaches showed that o-diphenols of VOO (hydroxytyrosol, decarboxymethyl oleuropein aglycon and oleuropein aglycon) were the strongest in terms of antioxidant power. Moreover, by OSI the authors demonstrated a pro-oxidant effect of elenolic acid, (⫹)-pinoresinol, tyrosol, ligstroside aglycon and (⫹)-1-acetoxypinoresinol (Figure 67.2). Figure 67.3 shows an increase in oxidative stability (⫹2.19 Δh) with the phenolic compound mixture with respect to that obtained with the theoretical sum (⫺1.63 Δh) given by each phenolic compound. So, the result demonstrates a contrary effect with an antioxidant activity prevailing (synergistic effect). In 1999, Mannino et al. proposed a procedure to evaluate the antioxidant capacity of VOO based on electrochemical properties by direct injection of the samples in
FIGURE 67.2 Increase or decrease in values evaluated by oxidative stability index (OSI) of eight isolated phenolic compounds at two different concentrations (218 and 50 mg kg⫺1 of each compound), from CarrascoPancorbo et al. (2005). Compounds: a, hydroxytyrosol; b, tyrosol; c, elenolic acid; d, deacetoxy oleuropein aglycon; e, (⫹)-pinoresinol; f, (⫹)-1-acetoxypinoresinol; g, oleuropein aglycon; h, ligstroside aglycon.
633
an FIA system with an electrochemical detector operating at a potential of ⫹500 mV (vs Ag/AgCl). The results from this procedure were compared with those by Rancimat and the ABTS•⫹ assay (after dilution of samples with n-hexane). In a previous work by the same authors (Mannino et al., 1993) it was shown that molecules with an oxidation potential lower than ⫹600 mV (vs Ag/AgCl) possess good antioxidant activity whereas molecules with higher values show moderate or no antioxidant activity. Through the study of the hydrodynamic voltammetric profiles of VOO compounds, they demonstrated that only a few compounds of VOO tested are oxidizable at potentials lower than ⫹600 mV (i.e., 3,4-dihydroxyphenylacetic acid, caffeic acid, gallic acid, α-tocopherol, oleuropein, protocatechuic acid), and no compounds can be detected at potentials lower than ⫹400 mV. There was good agreement between the results of the proposed electrochemical method on several VOO and those of the ABTS•⫹ decoloration assay (r ⫽ 0.988), whereas a lower linear correlation was observed with data by Rancimat (r ⫽ 0.893). In 2004, Del Carlo and co-workers applied the hydrodynamic voltammetry in flow injection analysis (FIA) for the antioxidant power evaluation of 22 samples of VOO. They also analyzed the total phenol content and the antioxidant activity by Folin–Ciocalteu and DPPH, respectively. The antioxidant activity of PH extracts determined by DPPH and hydrodynamic voltammetry in flow injection analysis was positively correlated with the oxidative stability measured with Rancimat (r ⫽ 0.810 and r ⫽ 0.808, respectively). In 2006, Busch et al. tested two amperometric enzymebased biosensors employing tyrosinase or peroxidase for rapid measurement of polar phenolics in 48 VOO samples. In particular, the authors found that coupling an off-line extraction of VOO to flow injection analysis of the biosensor led to a simplification of the method and a considerable reduction of the analysis time (3–5 min per sample). In 2007, Murkovic Steinberg and Milardovic used solid-state optical sensor membranes based on an immobilized chromogenic DPPH radical for the assessment of antioxidant activity of several phenolic standards and olive oil. DPPH• was chosen because of it widespread use, solubility in plasticizer and availability as stable neutral radical. They demonstrated that the reactivity of immobilized DPPH• is comparable to standard solution-based DPPH assays and that a rapid and simple qualitative screening test of untreated samples is possible using this test strip based on immobilized DPPH radical.
SUMMARY POINTS ●
FIGURE 67.3 Synergistic effect evaluated by OSI under accelerated oxidative conditions.
General concepts on the role of phenolic compounds in contrastiny the lipid oxidation process that naturally occurs in virgin olive oil have been explained.
634
●
●
●
SECTION | I Natural Components
Several methods exist in the literature to evaluate the antioxidant capacity of the phenolic fraction of virgin olive oil. A special emphasis has been directed to the main direct and indirect procedures. The most widespread methods have been compared and their main advantages or criticisms have been discussed. The results of recent applications of different methods for the evaluation of antioxidant activity in virgin olive oil have been summarized.
REFERENCES Aparicio, R., Roda, L., Albi, M.A., Gutierrez, F., 1999. Effect of various compounds on virgin olive oil stability measured by Rancimat. J. Agric. Food Chem. 47, 4150–4155. Bandoniene, D., Murkovic, M., 2002. On-line HPLC-DPPH screening method for evaluation of radical scavenging phenols extracted from apples (Malus domestica L.). J. Agric. Food Chem. 50, 2482–2487. Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gómez-Caravaca, A.M., Segura-Carretero, A., Fernández-Gutiérrez, A., Lercker, G., 2007a. Phenolic molecules in virgin olive oils: a survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules 12, 1679–1719. Bendini, A., Gómez-Caravaca, A.M., Cerretani, L., Del Carlo, M., SeguraCarretero, A., Compagnone, D., Cichelli, A., Lercker, G., 2007b. Evaluation of contribution of micro and macro components to oxidative stability on virgin oils obtained from olives characterized by different health quality. Progr. Nutr. 9, 210–215. Bendini, A., Cerretani, L., Poerio, A., Bonoli-Carbognin, M., Gallina Toschi, T., Lercker, G., 2006a. Oxidative stability of virgin olive oils, produced by organic, integrated or conventional agricultural methods. Progr. Nutr. 8 (2), 104–115. Bendini, A., Cerretani, L., Vecchi, S., Carrasco-Pancorbo, A., Lercker, G., 2006b. Protective effects of extra virgin olive oil phenolics on oxidative stability in the presence or absence of copper ions. J. Agric. Food Chem. 54, 4880–4887. Busch, J.L.H.C., Hrncirik, K., Bulukin, E., Boucon, C., Mascini, M., 2006. Biosensor measurements of polar phenolics for the assessment of the bitterness and pungency of virgin olive oil. J. Agric. Food Chem. 54, 4371–4377. Caldwell, C.R., 2000. A device for the semiautomatic determination of oxygen-radical absorbance capacity. Anal. Biochem. 287, 226–233. Campanella, L., Favero, G., Pastorino, M., Tomassetti, M., 1999. Monitoring the rancidification process in olive oils using a biosensor operating in organic solvents. Biosens. Bioelectron. 14, 179–186. Capannesi, C., Palchetti, I., Mascini, M., Parenti, A., 2000. Electrochemical sensor and biosensor for polyphenols detection in olive oils. Food Chem. 71, 553–562. Carrasco-Pancorbo, A., Cerretani, L., Bendini, A., Segura-Carretero, A., Del Carlo, M., Gallina-Toschi, T., Lercker, G., Compagnone, D., Fernandez-Gutierrez, A., 2005. Evaluation of the antioxidant capacity of individual phenolic compounds in virgin olive oil. J. Agric. Food Chem. 53, 8918–8925. Cerretani, L., Baccouri, O., and Bendini, A., 2008. Improving of oxidative stability and nutritional properties of virgin olive oils by fruit destoning. Agro Food Ind Hi Tec. 19, 21–23. Decker, E.A., 2002. Antioxidant mechanisms. In: Akoh, C., Min, D.B. (Eds.) Food Lipids: Chemistry, Nutrition, and Biotechnology, 2nd edn. Marcel Dekker Inc., New York.
Del Carlo, M., Sacchetti, G., Di Mattia, C., Compagnone, D., Mastrocola, D., Liberatore, L., Cichelli, A., 2004. Contribution of the phenolic fraction to the antioxidant activity and oxidative stability of olive oil. J. Agric. Food Chem. 52, 4072–4079. Ferri, E., Girotti, S., Cerretani, L., Bendini, A., 2006. Various luminescent methods applied to evaluate olive oil Total Antioxidant Capacity. Poster presented at the XIIth ISLS International Symposium Luminescence Spectrometry, Lugo Spain, July 2006. Frankel, E.N., 1993. In search of better methods to evaluate natural antioxidants and oxidative stability in food lipids. Trends Food Sci. Technol. 4, 220–225. Frankel, E.N., 1996. Antioxidants in lipid foods and their impact on food quality. Food Chem. 57, 51–55. Gomez-Alonso, S., Mancebo-Campos, V., Salvador, D., Fregapane, G., 2007. Evolution of major and minor components and oxidation indices of virgin olive oil during 21 months storage at room temperature. Food Chem. 100, 36–42. Gorinstein, S., Martin-Belloso, O., Katrich, E., Lojek, A., Cız, M., Gligelmo-Miguel, N., Haruenkit, R., Park, Y.-S., Jung, S.-T., Trakhtenberg, S., 2003. Comparison of the contents of the main biochemical compounds and the antioxidant activity of some Spanish olive oils as determined by four different radical scavenging tests. J. Nutr. Biochem. 14, 154–159. Guo, C., Cao, G., Sofic, E., Prior, R.L., 1997. High-performance liquid chromatography coupled with colometric array detection of electroactive components in fruits and vegetables: relationship to oxygen radical absorbance capacity. J. Agric. Food Chem. 45, 1787–1796. Lussignoli, S., Fraccaroli, M., Andrioli, G., Brocco, G., Bellavite, P., 1999. A microplate-based colorimetric assay of the total peroxyl radical trapping capability of human plasma. Anal. Biochem. 269, 38–44. Manna, C., D’Angelo, S., Migliardi, V., Loffredi, E., Mazzoni, O., Morrica, P., Galletti, P., Zappia, V., 2002. Protective effect of the phenolic fraction from virgin olive oils against oxidative stress in human cells. J. Agric. Food Chem. 50, 6521–6526. Mannino, S., Buratti, S., Cosio, M.S., Pellegrini, N., 1999. Evaluation of the ‘antioxidant power’ of olive oils based on a FIA system with amperometric detection. Analyst 124, 1115–1118. Mannino, S., Cosio, M.S., Bertuccioli, M., 1993. High performance liquid chromatography of phenolic compounds in virgin olive oils using amperometric detection. Ital. J. Food Sci. 4, 363–370. Mateos, R., Dominguez, M.M., Espartero, J.L., Cert, A., 2003. Antioxidant effect of phenolic compounds, α-tocopherol, and other minor components in virgin olive oil. J. Agric. Food Chem. 51, 7170–7175. Mateos, R., Uceda, M., Aguilera, M.P., Escuderos, M.E., Beltran Maza, G., 2006. Relationship of Rancimat method values at varying temperatures for virgin olive oils. Eur. Food Res. Technol. 223, 246–252. Montedoro, G.F., Servili, M., Baldioli, M., Miniati, E., 1992. Simple and hydrolyzable phenolic compounds in virgin olive oil. 1. Their extraction, separation, and quantitative and semiquantitative evaluation by HPLC. J. Agric. Food Chem. 40, 1571–1576. Murkovic Steinberg, I., Milardovic, S., 2007. Chromogenic radical based optical sensor membrane for screening of antioxidant activity. Talanta 71, 1782–1787. Navas, M.J., Jimenez, A.M., 2007. Chemiluminescent methods in olive oil analysis. J. Am. Oil Chem. Soc. 84, 405–411. Ninfali, P., Bacchiocca, M., Biagiotti, E., Servili, M., Montedoro, G.F., 2002. Validation of the Oxygen Radical Absorbance Capacity (ORAC) parameter as a new index of quality and stability of virgin olive oil. J. Am. Oil Chem. Soc. 79, 977–982. Ninfali, P., Bacchiocca, M., Biagiotti, E., Esposto, S., Servili, M., Rosati, A., Montedoro, G., 2008. A 3-year study on quality, nutritional and
CHAPTER | 67 Rapid Assays to Evaluate the Antioxidant Capacity of Phenols in Virgin Olive Oil
organoleptic evaluation of organic and conventional extra-virgin olive oils. J. Am. Oil Chem. Soc. 85, 151–158. Paiva-Martins, F., Santos, V., Mangericao, H., Gordon, M.H., 2006. Effects of copper on the antioxidant activity of olive polyphenols in bulk oil and oil-in-water emulsions. J. Agric. Food Chem. 54, 3738–3743. Polášek, M., Skála, P., Opletal, L., Jahodárˇ, L., 2004. Rapid automated assay of anti-oxidation/radical-scavenging activity of natural substances by sequential injection technique (SIA) using spectrophotometric detection. Anal. Bioanal. Chem. 379, 754–758. Porter, W.L., Black, E.D., Drolet, A.M., 1989. Use of polyamide oxidative fluorescence test on lipid emulsions: contrast in relative effectiveness of antioxidants in bulk versus dispersed systems. J. Agric. Food Chem. 37, 615–624. Prior, R.L., Wu, X., Schaich, K., 2005. Standardized methods for the determination of antioxidant capacity and phenolics in foods and dietary supplements. J. Agric. Food Chem. 53, 4290–4302.
635
Roginsky, V., Lissi, E.A., 2005. Review of methods to determine chainbreaking antioxidant activity in food. Food Chem. 92, 235–254. Samaniego Sánchez, C., Troncoso González, A.M., García-Parrilla, M.C., Quesada Granados, J.J., López García de la Serrana, H., López Martínez, M.C., 2007. Different radical scavenging tests in virgin olive oil and their relation to the total phenol content. Anal. Chim. Acta. 593, 103–107. Servili, M., Montedoro, G., 2002. Contribution of phenolic compounds to virgin olive oil quality. Eur. J. Lipid Sci. Technol. 104, 602–613. Shahidi, F., Naczk, M., 2004. Antioxidant properties of food phenolics. In: Shahidi, F., Naczk, M. (Eds.) Phenolics in Food and Nutraceuticals. CRC Press, Boca Raton. Triantis, T., Papadopoulos, K., Stellakis, A., Dimotikali, D., 2004. Studies on the antioxidant activity of aqueous extracts of olive oils and seed oils using chemiluminescence. Chem. Phys. Lipids 130, 57.
This page intentionally left blank
Chapter 68
Polycyclic Aromatic Hydrocarbons (PAHs) in Olive Oil: Methodological Aspects of Analysis Martin Rose Food and Environment Research Agency (Fera), Sand Hutton, York, UK
68.1 INTRODUCTION: OCCURRENCE AND FORMATION OF PAHS Polycyclic aromatic hydrocarbons (also called arenes) consist of a large family of aromatic compounds containing three or more fused aromatic rings made up of carbon and hydrogen. The term is most often used for unsubstituted parent compounds and their alkyl substituted derivatives, but can also be used for functionalized derivatives, such as chloro- or nitroPAHs and heterocyclic analogues such as indole, quinoline, benzothiophene, 9-cyanoanthracene, and dibenzothiophene. There are about 250 compounds generally included in the term PAH and the structures of some important unsubstituted PAHs are shown in Figure 68.1 (Cano-Lerida et al., 2008). PAHs and their derivatives are widespread in the environment and are found in the atmosphere, surface water, sediments and soil, food and lipid tissues. They arise from both natural and anthropogenic sources, such as burning fossil fuels and other processes involving incomplete combustion (European Commission, 2002). Temperatures above 350 °C and the combustion of fats produce greatest concentrations of PAHs and certainly account for the major source of most of these compounds in food. PAHs are formed when organic compounds are partially ‘cracked’ to produce smaller unstable fragments, mostly radicals that recombine to give relatively stable PAHs (pyrosynthesis). Aromatization that occurs at lower temperatures (100–150 °C) can also produce PAHs, but this requires much more time and produces large quantities of alkylated PAHs.
68.2 TOXICOLOGY Some PAHs are known or suspected genotoxic carcinogens. As such and according to current toxicological Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
(1)
(2)
(5)
(6)
(9)
(10)
(3)
(4)
(7)
(8)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
FIGURE 68.1 Structures of some PAHs. The chemical structures of some common PAHs are shown: (1) phenanthrene, (2) anthracene, (3) benz[a]anthrecene, (4) pyrene, (5) chrysene, (6) naphthacene, (7) benzo[c]phenanthrene, (8) benzo[ghi]fluoranthene, (9) dibenzo[c,g]phenanthrene, (10) benzo[ghi]perylene, (11) triphenylene, (12) o-tephenyl, (13) o-terphenyl, (14) p-terphenyl, (15) benzo[a]pyrene, (16) tetrabenzonaphthalene, (17) phenanthro[3,4-c]phenanthrene, (18) coronene. Adapted from Cano-Lerida et al. (2008).
understanding, there is no safe level of exposure and intake from food should be as low as is reasonably achievable. This is known as the ALARA principle. They have been known for a long time to occur at low levels in food, especially where food is smoked or dried during the production process. Legislation has been introduced within the
637
Crown Copyright © 2009. Published by Elsevier Inc. All rights reserved.
638
SECTION | I Natural Components
TABLE 68.1 Estimated dietary intakes by adults of PAHs in different countries. Type of samples analyzed
Country
Year
Estimated dietary exposure (ng kg⫺1 bw day⫺1) BaP
Total PAHs*
Reference
Total diet study
UK
2000
1.6
69
FSA, 2002
Total diet study
Milan, Italy
ns
ns
50
Lodovici, 1995
Market basket
Netherlands
1984–1986
4.8
284
De Vos, 1990
Retail foods
USA
ns
0.3–1.3**
ns
Kazerouni, 2001
*
The sets of PAHs analyzed were not the same in all the surveys. Range of dietary intakes by 90% of the consumers. ns Not stated. Intakes of PAHs are estimated from total diet surveys, market basket surveys and retail surveys, and combined with estimates of food consumption to give estimates of intake via the diet. **
EU that establishes maximum limits for benzo[a]pyrene (BaP) legally allowed in food, with an instruction to monitor other PAHs with a view to including these in future legislation (European Union, 2005). Most work done on PAHs monitors BaP as an indicator substance. The choice of BaP as an indicator substance was based on its prevalence and the fact that it is one of the most toxic PAHs. The maximum limit for BaP in oils and fats is 2 μg kg⫺1 (that is, 2 parts-per-billion) and there are other food types included in the legislation with limits ranging from 1 μg kg⫺1 for baby foods and infant formulae to 10 μg kg⫺1 for smoked bivalve molluscs. Foodstuffs represent a major source of exposure of humans to PAHs although there are a few incidents of direct occupational exposure due to combustion processes that have been recorded. Previous work on the UK Total Diet Study Survey (from 2000) revealed that the fats and oils group contained one of the highest concentrations of PAHs at 11.05 μg kg⫺1 (sum of 19 PAHs) (Food Standards Agency, 2002). Estimates of exposure to PAHs through the diet are summarized in Table 68.1 and limits in force within the EU are shown in Table 68.2. The PAHs that are recommended for further investigation include the 15 EU priority PAHs together with one PAH (benzo[c]fluorine) highlighted by the JECFA committee on Food Additives in 2005. Because of the likelihood of further development of European legislation in this area, analytical methodology for multiple PAHs in food is required in order to get a clearer picture of human exposure to this class of compound (Wenzl et al., 2006). Because of the uncertainties surrounding levels of PAHs in foods, especially those with both genotoxic and carcinogenic properties, EFSA has established an online analytical database in collaboration with EU Member States in order to collect occurrence data. This information should also show whether or not BaP is a reliable marker
TABLE 68.2 EU maximum levels for benzo[a]pyrene in foods. Product
Maximum level (μg kg⫺1 wet weight)
Oils and fats intended for direct human consumption or use as an ingredient in foods
2
Foods for infants and young children: ● Baby foods and processed cereal-based foods for infants and young children ● Infant formulae and follow-on formulae, including infant milk and follow-on milk Dietary foods for special medical ● purposes intended specifically for infants
1
Smoked meats and smoked meat products
5
Muscle meat of smoked fish and smoked fishery products
5
Muscle meat of fish, other than smoked fish
2
Bivalve molluscs, crustaceans, cephalopods, other than smoked
5
Smoked bivalve molluscs
10
These limits are in force within the EU. When exceeded, food should be removed from sale and should not reach the consumer. They are set such that if good manufacturing processes are used, they should be easily achievable.
for more general PAH exposure (Wenzl et al., 2006) or if, to the contrary, the level of BaP relative to the other PAHs in different foods is too variable and introduces too much uncertainty.
CHAPTER | 68 Polycyclic Aromatic Hydrocarbons (PAHs) in Olive Oil: Methodological Aspects of Analysis
68.3 SOURCES OF PAHS IN FOODS 68.3.1 Environmental Sources PAHs of lower molecular mass (generally those with three aromatic rings) are prevalent in the atmosphere whereas those of higher molecular mass (the majority) enter the environment adsorbed onto particulate matter. The hydrosphere and geosphere are affected by dry and wet deposition of PAHs. Solubility in water affects the distribution patterns of PAHs in food (Lerario et al., 2003). Phenanthrene for example has a relatively high solubility when compared to other PAHs, and is therefore likely to have a higher environmental mobility; consequently, this PAH is found in highest concentrations in aquatic samples. Since PAHs are lipophilic, uptake resulting from absorption by plants from contaminated soils is low, but the waxy surfaces of vegetables and fruits can concentrate low-molecular-mass PAHs through surface adsorption (European Commission, 2002). Most animals will metabolize the PAHs and so exposure from animal-derived food products (milk, eggs, meat, etc.) eaten by humans is low unless further formed by cooking. Although PAHs are present in some foodstuffs as a result of environmental exposure, the processing of food by smoking, drying, barbecuing and other cooking methods, is the largest source of PAHs in food (Mottier et al., 2000). Charcoal grilling in particular can generate high levels of PAHs, and quantities are related to factors such as fat content, temperature and cooking time.
68.3.2 Formation of PAHs in Olive Oil Production Significant quantities of PAHs have been found in a range of vegetable oils and fats, including sunflower, grapeseed and most notably in pomace oil derived from olives. Olive oil produced by cold pressing either traditionally or by an industrial process will not produce significant quantities of PAHs. But there have been cases where pomace oil has been used to dilute the more expensive olive oil, resulting in more significant amounts of PAHs in the product. This is both a fraud and food safety issue. Contamination occurs when heat is used to dry the olive oil pomace prior to the use of an extraction solvent to remove residual oil. The dry cake that is left has a high calorific value and is used to fuel the heating process for subsequent batches. The amount of PAH produced is dependent upon several variables including: (i) contaminants in the ingredients; (ii) the production process used; (iii) the temperature of combustion; (iv) whether or not there is direct contact with the fumes of combustion; and (v) the water level of the pomace, which can effect the heating time. There are two main commercial processes used for continuous extraction of olive oil. The three-stage process
639
involves grinding olives to a paste. The paste is placed in a ‘Thermobeater’ with heated water to produce a liquid and solid pomace waste. The immiscible oil and water can be separated by gravity or centrifuge to yield pomace solids (containing 30–40% water) and ‘Alpechin’, a waste product consisting of water and organic matter. The two-stage process does not use water and thus the process is simpler. The paste is centrifuged continuously and the difference in viscosity allows the production of olive oil and pomace oil. There is not much heating of the paste required in this process and the resulting pomace has a water content of 70–80%. The three-stage process is declining in popularity because the Alpechin is difficult to dispose of in a sustainable or ‘green’ way. The pomace still contains 1–7% oil and this can be used to produce olive pomace oil using an extraction solvent (typically hexane) after as much water as possible is removed by heating. The pomace from the two-stage process contains more water and therefore needs more heating; hence the potential for increased PAH production. The fuel for heating is usually the spent pomace cake. Similar equivalent extraction and refining procedures are used for the production of oils from other vegetable and seeds. This particular problem was behind the incident in July 2001 when Spanish, Italian and Greek olive pomace oils were found to be contaminated with PAHs (European Union, 2002c) resulting in products being withdrawn from sale. As a result of the variability within results reported for the same products associated with this incident, the European Commission made inspection visits of laboratories and production facilities in all olive-oil-producing countries. The report following the inspection highlighted the need for standardized methods of analysis (European Union, 2002c).
68.4 METHODS OF ANALYSIS: CURRENT STATUS Due to the ubiquity of PAHs and their potentially deleterious effects on human health, these compounds are often analyzed in a variety of matrices such as foods, oils, waters, soils, etc. The following section does not provide an exhaustive list of analytical methods but aims to identify well-established methods that are used as standard methods of analysis and that are applicable to olive oil. There are several well-established analytical procedures for the analysis of PAHs. Most of them involve pretreatment of the sample, consisting of an initial extraction of the PAHs from the sample using liquid/liquid, solid-phase and/or ultrasonic extractions. This pretreatment stage serves the purpose of pre-cleaning some interfering compounds and also acts as a pre-concentration step (Mastral et al., 2004). Extracts are analyzed by either high-performance liquid chromatography (HPLC) or by gas chromatography (GC)
640
depending on the nature of the sample and its volatility. The detectors used to measure the concentration are with ultraviolet (UV) or fluorescence (FLD) spectrophotometers (for the HPLC separation) and mass spectrometers (MS) for the GC separation (Li et al., 2003; Nemcik, 2005). Thin-layer chromatography (TLC) is an inexpensive, quick analytical technique but has low separation efficiency and is commonly used only for identification purposes of individual compounds or for very simple sample types. With the exception of TLC, these techniques are accurate and can handle complicated mixtures, but at the same time they are expensive and require qualified and experienced personnel to operate them. Their widespread use is justified by the high complexity of ‘real’ samples mostly containing a large number of PAHs. As an example a standard capillary GC column containing 3000 theoretical separating plates per meter, allows good separation of mixtures of about 100 PAHs (IPCS, 1998). The limits of detection of these analytical methods for PAHs need to be low if they are used to measure concentrations of PAHs that are normally present at trace amounts (⬍1 ppb), which can nevertheless be significant due to the high associated toxicity (Manoli, 1999). Immunoassays have been applied extensively to evaluate environmental contamination, and analysis of biological fluids. Molecules detected by immunoassays vary widely in size, chemical and physical properties, and biological activity. PAHs are enzymatically converted to highly reactive metabolites that bind covalently to macromolecules such as DNA, thereby causing mutagenesis and carcinogenesis in experimental animals. For example, BaP is activated by microsomal enzymes to 7β, 8α-dihydroxy-(9α,10α)epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene and binds covalently to DNA, resulting in formation of BPDE-DNA adducts (Pavanello et al., 1999; Wani et al., 2002; Smith and Hurtubise, 2004). Immunoassays are sensitive methods and able to detect PAH–DNA adducts in the blood and tissues of humans and animals and include, for example, enzyme-linked immunosorbent assays (ELISA) (Chuang et al., 1998), radioimmunoassay (Schneider et al., 1995; Baran et al., 2003), dissociation-enhanced lanthanide fluoroimmunoassay (DELFIA) (Divi et al., 2002), and ultrasensitive enzyme radioimmunoassay (USERIA) (Ovrebo et al., 1992, 1994); 32P- and 35S-postlabeling with radioactivity counting (Gorelick and Reeder, 1993); surface-enhanced Raman spectroscopy (Olson et al., 2004); and synchronous luminescence spectroscopy (SLS) (Matuszewska, 2000). The US EPA has developed a standard immunoassay method (number 4035) for screening of soils for PAHs. This method is a semi-quantitative enzyme immunoassay and correctly identifies 95% of samples that are PAH-free and those containing 1 ppm total PAH (as phenanthrene). The immunoassay is based on the use of antibodies immobilized on the walls of the test tubes that bind either PAH or PAH– enzyme conjugate. When PAH is present in the sample,
SECTION | I Natural Components
it competes with the PAH–enzyme conjugate for a limited number of binding sites on the immobilized antibodies. Great caution should be exercised before transferring methods designed for soil or environmental samples to use with food or in particular olive oil in this instance. Solid phase extraction (Biernoth and Rost, 1968) and solid-phase microextraction (Paschke et al., 1999; Guillen and Sopelana, 2005; Hawthorne et al., 2005; Hu et al., 2005; Ter Laak et al., 2005) are also widely used. Supported liquid membrane extraction (SLM) (Zabiegala et al., 2000), supercritical fluid extraction (SFE) (Rigou et al., 2004; Chiu et al., 2005; Librando et al., 2005), online capillary microextraction (Bigham et al., 2002) and membrane extraction with sorbent interface (MESI) have also been used although less frequently (Roy et al., 2005). Recent work by Roy et al. demonstrates the coupling of stir bar sorptive extraction to a new generation of gas chromatography mass spectrometry (GC-MS), using the field apparatus EM 640 S from Bruker (Roy et al., 2005). Many laboratories undertaking the analysis of foods for PAHs use an extraction method, sometimes including solidphase extraction (e.g., C18 cartridges), followed by HPLC with FLD (European Union, 2004). This methodology has prevailed for many years due to the inherent fluorescence of PAHs giving rise to reasonably high selectivity and sensitivity. However, final analysis using low-resolution gas chromatography-mass spectrometry (GC-MS) with stable isotope dilution GC-MS offers an alternative procedure with greater accuracy, precision and selectivity than that offered by HPLC. There are several reference methods for PAHs. These include ISO 15302 (1998) ‘Methods of analysis of fats and fatty oils’. Other methods include ‘Determination of benzo[a]pyrene content by reverse-phase HPLC’ which is an international standard for the determination of BAP in crude or refined edible oils and fats by reverse-phase HPLC using fluorimetric detection in the range from 0.1 μg kg⫺1 to 50 μg kg⫺1. ISO/FDIS 15753 (2006) ‘Animal and vegetable fats and oils – Determination of polycyclic aromatic hydrocarbons’ is an international standard describing two methods for the determination of 15 polycyclic aromatic hydrocarbons (PAHs) in animal and vegetable fats and oils: (i) a general method, and (ii) a method specific for coconut oil and short-chain vegetable oils. These ISO methods are not quantitative for the more volatile compounds such as naphthalene, acenaphthene and fluorene. Due to interferences provided by the matrix itself, palm oil and olive pomace oil cannot be analyzed using these methods. The quantification limit is 0.2 μg kg⫺1 for almost all compounds analyzed (for fluoranthene and benzo(g,h,i)perylene the quantification limit is 0.3 μg kg⫺1, and for indeno(1,2,3-c,d)pyrene the quantification limit is 1 μg kg⫺1). The PAHs are extracted with an acetonitrile/ acetone mixture followed by purification on C18 reversed-phase and then florisil bonded-phase cartridges.
641
CHAPTER | 68 Polycyclic Aromatic Hydrocarbons (PAHs) in Olive Oil: Methodological Aspects of Analysis
Determination is achieved by HPLC and fluorescence detection. ISO TC 34/SC 11 N863 Working Draft ‘Animal and vegetable fats and oils – Determination of polycyclic aromatic hydrocarbons in edible fats and oils by on-line donor acceptor complex chromatography and HPLC with fluorescence detection’ describes an HPLC procedure which has been validated for coconut, olive, sunflower and bean oils. The limit of determination for the PAHs is 0.1 μg kg⫺1. The validated concentration range of the method is 0.1 μg kg⫺1 to 3.5 μg kg⫺1 for each individual PAH. If it is expected that the level of the (light) PAHs in samples to be analyzed will be ⬎3.5 μg kg⫺1, these samples have to be diluted prior to analysis. Seventeen PAHs can be determined by this method: anthracene, phenanthrene, fluoranthene, pyrene, chrysene, 1,2-benzanthracene, benz(e)pyrene, benz(a)pyrene, perylene, 1,12-benzperylene, anthanthrene, 1,2,5,6-dibenzanthracene, coronene, indeno(123-cd)pyrene, benz(a)fluoranthene, benz(b)fluoranthene and benz(k)fluoranthene. The PAHs in edible oils can be determined by online coupling of donor acceptor complex chromatography and reversed-phase HPLC with fluorescence detection. The oil samples are eluted through a modified stationary phase XC15589 100
Benzo(e)pyrene
(DACC column) which acts as an electron acceptor. This column retains the PAHs (electron donors) by p–p interactions. The PAHs are transferred online to the analytical reversed-phase column. The individual PAHs are detected at different wavelengths and are identified according to retention time and quantified using external calibration. More recent methods have been proposed for use within the oils and fats sector in response to the need for a fully validated GC-MS PAH method using isotopically labeled internal standards as raised by various international standards committees including CEN/TC307, ISO/TC34/ SC2 and ISO/TC34/SC11. Internal standardization using 13 C-labeled isotopes of the PAH analytes provides an automatic correction for analytical recovery. Moreover, the inclusion of 13C isotopes in every sample also corrects for any variation in recovery due to matrix effects. Mass spectrometry also enables the monitoring of fragment ions which are consistent and can act as confirmation of the presence of an analyte. A chromatogram showing benzo[a]pyrene in olive oil along with the 13C-labeled isotope added to assist quantification is shown in Figure 68.2. The high accuracy of data obtained using stable isotope dilution is well recognized and it is, for example, the
60.25
Scan ES+ 250.1 3.14e5
Benzo(a)pyrene
60.66
59.91
61.44
%
−32 XC15589 100
60.25
Scan ES+ 252.1 8.12e5
60.66
% 61.42 59.91 −12 XC15589 100
Scan ES+ 254.1 9.01e5
60.66
13-C Benzo(a)pyrene
%
−11 XC15589 100
Scan ES+ 256.1 2.88e6
60.66
% 59.88 −3
Time 59.00
59.25
59.50
59.75
60.00
60.25
60.50
60.75
61.00
61.25
61.50
61.75
62.00
62.25
62.50
FIGURE 68.2 Chromatogram showing benzo(a)pyrene in olive oil. GC-MS chromatogram showing the two most abundant ions for benzo[a]pyrene (top two traces) and 13C internal standard corresponding ions (bottom two traces). The internal standard is added in a precise known quantity and the amount of benzo[a]pyrene in the olive oil sample can be calculated accurately by comparing the ratio with the corresponding peak in the top traces. Benzo[e]pyrene is also shown in the oil but no internal standard is used for this compound, so it is also quantified against the internal standard for benzo[a]pyrene.
642
SECTION | I Natural Components
recommended confirmatory method for the highly exacting measurement of dioxins and other substances in food (European Union, 2002a, b; Rose et al., 2007). In summary, PAHs can be formed during the production of olive oil. Some of these compounds are known or suspected genotoxic carcinogens, and as such human exposure via the diet should be kept as low as possible. There is legislation governing PAHs in olive oil (and other foods) in some parts of the world and such legislation is likely to increase in the future. Robust and reliable analytical methodology is available to enforce this legislation. The most widely used methods are based on HPLC, usually with fluorescence detection making use of the fact that most PAHs are themselves fluorescent compounds. There is growing use of GC-MS methods for the determination of PAHs; GC offers better resolution of the different PAH compounds and the use of MS offers high specificity and the possibility of the use of stable isotope dilution methods, which give high quantitative accuracy and confirmation of identity of the analyte. There is still however a demand for lowcost high-throughput methods to enable producers to better monitor PAHs in oil production.
SUMMARY POINTS ●
●
●
●
●
PAHs can be formed during the production of olive oil. They are formed as products of incomplete combustion. Some PAHs are genotoxic carcinogens and human exposure should be as low as reasonably achievable. Some countries have regulations setting maximum limits for PAHs in olive oil. Good methods are needed to enforce regulatory limits and to monitor for the presence of PAHs in olive oil from a food safety perspective. Most methods used are based on HPLC or GCMS.
REFERENCES Baran, S., et al., 2003. Pesticides food contaminants and agricultural wastes J. Environ. Sci. Health B 38, 799–812. Biernoth, G., Rost, H.E., 1968. Arch. Hyg. Bakteriol 152, 238–250. Bigham, S., et al., 2002. Anal. Chem. 74, 752–761. Cano-Lerida, L., Rose, M., Walton, P., 2008. Polycyclic aromatic hydrocarbons. In: Gilbert, J., Senyuva, H. (Eds.) Bioactive Compounds in Foods – Natural Toxicants and Process Contaminants. Blackwell Publishing. Chiu, K.H., Yak, H.K., Wai, C.A., Lang, Q.Y., 2005. Talanta 65, 149–154. Chuang, J.C., Pollard, M.A., Chou, Y.L., Menton, R.G., Wilson, N.K., 1998. Sci. Total Environ. 224, 189–199. De Vos, R.H., et al., 1990. Food Chem. Toxicol. 28, 263–268. Divi, R.L., Beland, F.A., Fu, P.P., et al., 2002. Carcinogenesis 23, 2043–2049. European Commission. 2002. Opinion of the Scientific Committee on Food on the risks to human health of Polycyclic Aromatic Hydrocarbons in food, SCF/CS/CNTM/PAH/29.
European Union. 2002a. Commission Directive 2002/69/EC. (26 July 2002). Laying down the sampling methods and the methods of analysis for the official control of dioxins and the determination of dioxin-like PCBs in foodstuffs. http://europa.eu.int/eur-lex/pri/en/oj/ dat/2002/l_209/l_20920020806en00050014.pdf. European Union. 2002b. Commission Directive 2002/70/EC. (26 July 2002). Establishing requirements for the determination of levels of dioxins and dioxin-like PCBs in feedingstuffs. http://europa.eu.int/ eurlex/pri/en/oj/dat/2002/l_209/l_20920020806en00150021.pdf. European Union. 2002c. Final report of a mission carried out in Spain from 8th to 12th April 2002 in order to assess the control measures in place for vegetable oil production and in particular for the assessment of controls on PAH contamination of such oils, DG(SANCO)/8600/2002MR final. European Union. 2004. Directorate-General Health and Consumer Protection. Reports on tasks for scientific cooperation. Report of experts participating in Task 3.2.12 (October 2004) Collection of occurrence data on polycyclic aromatic hydrocarbons in food. http:// europa.eu.int/comm/food/food/chemicalsafety/contaminants/scoop_ 3-2-12_final_report_pah_en.pdf. European Union. 2005. Commission Regulation (EC) No 208/2005 of 4 February 2005 amending Regulation (EC) No 466/2001 as regards polycyclic aromatic hydrocarbons. Food Standards Agency. 2002. Information sheet 31/02. PAHs in the UK Diet: 2000 Total Diet Study Samples. December 2002. http://www. food.gov.uk/science/surveillance/fsis-2002/31pah. Gorelick, N.J., Reeder, N.L., 1993. Environ. Health Perspect. 99, 207–211. Guillen, M.D., Sopelana, P., 2005. J. Dairy Sci. 88, 13–20. Hawthorne, S.B., Grabanski, C.B., Miller, D.J., Kreitinger, J.P., 2005. Environ. Sci. Technol. 39, 2795–2803. Hu, Y.L., Yang, Y.Y., Huang, J.X., Li, G.K., 2005. Anal. Chim. Acta 543, 17–24. IPCS. 1998. Selected non-heterocyclic polycyclic aromatic hydrocarbons. Environmental Health Criteria 202. Kazerouni, N., et al., 2001. Food Chem. Toxicol. 39. Lerario, V.L., Giandomenico, S., Lopez, L., Cardellicchio, N., 2003. Ann. Chim. 93, 397–406. Li, S., Olegario, R.M., Banyasz, J.L., Shafer, K.H., 2003. J. Anal. Appl. Pyrol. 66, 155–163. Librando, V., Tomaselli, G., Tringali, G., 2005. Ann. Chim. 95, 211–216. Lodovici, M., et al., 1995. Food Addit. Contam. 12, 703–713. Manoli, E.S.C., 1999. Trac-Trends Anal. Chem. 18, 417–428. Mastral, A.M., Garcia, T., Lopez, J.M., Murillo, R., Callen, M.S., Navarro, M.V., 2004. Polycyclic Aromatic Compounds 24, 325–332. Matuszewska, A.C.M., 2000. Talanta 52, 457–464. Mottier, P., Parisod, V., Turesky, R.J., 2000. J. Agric. Food Chem. 48, 1160–1166. Olson, L.G., Uibel, R.H., Harris, J.M., 2004. Appl. Spectrosc. 58, 1394–1400. Ovrebo, S.H.A., Fjeldstad, P.E., et al., 1994. J. Occup. Med. 36, 303–310. Ovrebo, S.H.A., Phillips, D.H., et al., 1992. Cancer Res. 52, 1510–1514. Paschke, A., Popp, P., Schuurmann, G., 1999. Fresenius J. Anal. Chem. 363, 426–428. Pavanello, S., Favretto, D., Brugnone, F., Mastrangelo, G., Dal Pra, G., Clonfero, E., 1999. Carcinogenesis 20, 431–435. Rigou, P., Saini, S., Setford, S.J., 2004. Intl. J. Environ. Anal. Chem. 84, 979–994. Rose, M., White, S., Macarthur, R., Petch, R., Holland, J., Damant, A., 2007. Food Addit. Contam. 24 (6), 635–651.
CHAPTER | 68 Polycyclic Aromatic Hydrocarbons (PAHs) in Olive Oil: Methodological Aspects of Analysis
Roy, G., Vuillemin, R., Guyomarch, J., 2005. Talanta 66, 540–546. Schneider, U.A., Brown, M.M., Logan, R.A., 1995. Environ. Sci. Technol. 29, 2595–2602. Smith, B.W., Hurtubise, R.J., 2004. Anal. Chim. Acta 502, 149–159. Ter Laak, T.L., Mayer, P., Busser, F.J.M., Klamer, H.J.C., Hermens, J.L.M., 2005. Environ. Sci. Technol. 39, 4220–4225. Wani, M.A., El-Mahdy, M.A., Hamada, F.M., Wani, G., Zhu, Q., Wang, Q., Wani, A.A., 2002. Intl. J. Cancer 334–334.
643
Wenzl, T., Simon, R., Kleiner, J., Anklam, E., 2006. Analytical methods for polycyclic aromatic hydrocarbons (PAHs) in food and the environment needed for new food legislation in the European Union. Trends Anal. Chem. 25 (7), 716–725. Zabiegala, B., Kot, A., Namiesnik, J., 2000. Chemia Analityczna 45, 645–657.
This page intentionally left blank
Chapter 69
Determination of Aflatoxins and Ochratoxin A in Olive Oil Chiara Cavaliere, Patrizia Foglia, Roberto Samperi and Aldo Laganà Dipartimento di Chimica, ‘Sapienza’ Università di Roma, Italy
69.1 INTRODUCTION Mycotoxins are secondary metabolites produced by various species of filamentous fungi growing under a wide range of climatic conditions on agricultural commodities (grains, spices, fruit, coffee, nuts, etc.) in the field and during storage (Scudamore and Livesey, 1998; Zöllner and MayerHelm, 2006). Their occurrence in food, beverages and feed is caused by direct contamination of plant materials or their products (Ominski et al., 1994), or by a ‘carry over’ of mycotoxins and their metabolites into animal tissues, milk and eggs after intake of contaminated feed (Zöllner and Mayer-Helm, 2006). The mycotoxins differ widely in their chemistry and toxicology (Ominski et al., 1994). In terms of structural complexity, in fact, these toxic compounds vary from simple C4-compounds to complex substances, and range in mostly low molecular weight from about 200–500 Da. Consequently, also the mycotoxicoses that these mold metabolites can cause are very diverse (ICMSF, 1996). About 300–400 compounds are now recognized as mycotoxins, and approximately a dozen groups receive regular attention as they are considered a threat to human and animal health (Diener et al., 1987). Some fungi produce a single toxin, while others may produce many toxic compounds, which may be shared across fungal genera. Nevertheless, there are mycotoxins related to a specific genus. The main five genera of fungi, Aspergillus, Penicillium, Fusarium, Alternaria, and Claviceps, produce mycotoxins belonging to eight groups that are relevant to the food industry: aflatoxins, citrinin, fumonisins, ochratoxins, patulin and other small lactones, trichothecenes, resorcyclic lactones and ergot alkaloids.
A. parasiticus, and the rare A. nomius, found in a wide variety of important agricultural products (peanuts, plant oleaginous seeds, maize and its by products, etc.) as a result of the currently unavoidable invasion by molds before or during harvest or because of inappropriate storage (Rustom, 1997; Sweeney and Dobson, 1998; Moss, 2002). Although approximately 20 AFs have been identified, only four of them, aflatoxins B1 (AFB1), B2 (AFB2), G1 (AFG1), and G2 (AFG2), occur naturally (Weidenbörner, 2001) (Figure 69.1). The letters ‘B’ and ‘G’ refer to the blue and green fluorescent colors produced by these compounds under UV light, while the numbers 1 and 2 indicate major and minor compounds, respectively (Sweeney and Dobson, 1998; Hussein and Brasel, 2001; Weidenbörner, 2001). A. flavus only produces B aflatoxins, while A. parasiticus and A. nomius also produce G aflatoxins (JEFCA, 2002). The AFs are difurancoumarin compounds and are soluble in polar organic and chlorinated solvents, and in
O
O
O O
O
OCH3 AFB2
O
O
O
O O
O
O
O O
OCH3 AFG1
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
O
OCH3 AFB1
69.2 AFLATOXINS
O
O
O
O
Aflatoxins (AFs) are a class of mycotoxins produced mainly by the fungal species Aspergillus flavus and
O
O
O
OCH3 AFG2
FIGURE 69.1 The chemical difurancoumarinic structures of the aflatoxins B1 (AFB1), B2 (AFB2), G1 (AFG1), and G2 (AFG2).
645
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
646
SECTION | I Adverse Components
TABLE 69.1 Features of aflatoxins. 1. AFs are secondary metabolites produced by the fungal species Aspergillus flavus and A. parasiticus, and the rare A. nomius 2. AFs are found in a wide variety of important agricultural products, such as peanuts, plant oleaginous seeds, maize, and byproducts 3. The four naturally occurring AFs are B1 (AFB1), B2 (AFB2), G1 (AFG1), and G2 (AFG2) 4. AFs ‘B’ and ‘G’ under UV light produce blue and green fluorescent colors, respectively 5. AFs have been classified as potential carcinogenic agents to humans
aromatic hydrocarbons. These mycotoxins are biologically active, and have powerful hepatotoxic, immunosuppressive, mutagenic, carcinogenic, and teratogenic effects even in small amounts (Peraica et al., 1999; Hussein and Brasel, 2001). Following epidemiological evidence, AFs have been classified as potential carcinogenic agents to humans (group 1) by the International Agency for Research on Cancer (IARC) (IARC, 2002). AFB1 is the most widespread and potent of the four major AFs. Although AFB1, AFB2 and AFG1 are common in the same food samples, AFB1 predominates, representing 60–80% of the total AF content. Generally, AFB2, AFG1, and AFG2 do not occur in the absence of AFB1; in most cases, AFG1 is found in higher concentration than AFB2 and AFG2 (Weidenbörner, 2001). The main features of AFs are reported in Table 69.1.
69.3 OCHRATOXIN A Ochratoxins are a group of mycotoxins produced by secondary metabolism of several Aspergillus and Penicillium species in semitropical and temperate climates, mainly by P. verrucosum, A. ochraceus (Visconti et al., 1999; Jornet et al., 2000; Lau et al., 2000), and A. carbonarius together with a low percentage of the closely related A. niger (JEFCA, 2002). These fungi differ in their ecological niches, in the commodities affected, and in the frequency of their occurrence in different geographic regions. They can infect various food commodities of which cereals and cereal products, spices, dried fruits, coffee, cocoa, grapes are the most important (JEFCA, 2002; Zöllner and MayerHelm, 2006). Ochratoxins include nine different molecules, but the most widespread and toxic one is ochratoxin A (OTA, Figure 69.2). OTA has a dihydroisocoumarin moiety linked over a 7-carboxy group to L-phenylalanine via a peptide bond (Zöllner and Mayer-Helm, 2006); it is soluble in organic polar solvents and in weakly basic water solutions. OTA has been shown to be nephrotoxic, hepatotoxic, teratogenic, and immunotoxic for several animal species and to cause kidney and liver tumors in mice and rats
HO O
OH
O
O
O
NH
H3C Cl
FIGURE 69.2 The chemical dihydroisocoumarinic structure of ochratoxin A (OTA).
(Becker et al., 1998; Visconti et al., 1999). For these reasons, in 1993 the IARC classified OTA as a possible carcinogen to humans (group 2B) (IARC, 1993). The main features of OTA are reported in Table 69.2.
69.4 AFLATOXIN AND OCHRATOXIN A CONTAMINATION IN OLIVE OIL The Joint FAO/WHO Expert Committee on Food Additives has set a provisional tolerable weekly intake for OTA of 100 ng kg⫺1 of body weight (JEFCA, 2002), whereas the European Commission suggested an even lower level of 5 ng kg⫺1 body weight per day (Commission Regulation (EC) 1881/2006). Regarding the genotoxic aflatoxins, no tolerable daily intake can be set. Based on these recommendations, national and international guideline levels and maximum levels have been set for AFs and OTA in various foods (limits established by the European Union are reported in Table 69.3). However, no maximum level has been set for the two mycotoxin classes in olive oil. AF and OTA contamination is favored by particular climatic conditions, therefore it mainly occurs in regions where temperature and humidity are optimal for the growth of molds and for production of these toxins (Hussein and Brasel, 2001; JEFCA, 2002; Zöllner and Mayer-Helm,
647
CHAPTER | 69 Determination of Aflatoxins and Ochratoxin A in Olive Oil
TABLE 69.2 Features of ochratoxin A. 1. OTA is produced by secondary metabolism of several Aspergillus and Penicillium species, mainly P. verrucosum, A. ochraceus, and A. carbonarius 2. OTA can infect various food commodities of which cereals and cereal products, spices, dried fruits, coffee, cocoa, and grapes are the most important 3. OTA has been shown to be nephrotoxic, hepatotoxic, teratogenic, and immunotoxic in animals and is classified as a possible carcinogen to humans 4. The World Health Organization has set a provisional tolerable weekly intake for OTA of 100 ng kg–1 of body weight
TABLE 69.3 Maximum levels for aflatoxins and ochratoxin A in foodstuffs in the European Union. Maximum level (μg kg⫺1)
Foodstuffs
AFs
OTA
AFB1
AFB1, AFB2, AFG1, AFG2 (sum)
Groundnuts to be subjected to sorting, or other physical treatment, before human consumption or use as an ingredient in foodstuffsa
8.0
15.0
n.l.c
Nutsa, dried fruit and maize to be subjected to sorting, or other physical treatment before human consumption or use as an ingredient in foodstuffs
5.0
10.0
n.l.c
Groundnutsa, nutsa, and dried fruit and processed products thereof, intended for direct human consumption or use as an ingredient in foodstuffs
2.0
4.0
n.l.c
All cereals and all products derived from cereals, including processed cereal products, with the exception of foodstuffs listed elsewhere
2.0
4.0
3.0
Unprocessed cereals; roasted coffee beans and ground roasted coffee, excluding soluble coffee
n.l.c
n.l.c
5.0
Spices: Capsicum spp. (dried fruits thereof, whole or ground, including chillies, chilli powder, cayenne and paprika); Piper spp. (fruits thereof, including white and black pepper); Myristica fragrans (nutmeg); Zingiber officinale (ginger); Curcuma longa (turmeric)
5.0
10.0
n.l.c
Dried vine fruit (currants, raisins and sultanas); soluble coffee (instant coffee)
n.l.c
n.l.c
10.0
Wine (including sparkling wine, excluding liqueur wine and wine with an alcoholic strength of not less than 15% vol) and fruit wine; aromatized wine, aromatized wine-based drinks and aromatized wine-product cocktails; grape juice, concentrated grape juice as reconstituted, grape nectar, grape must and concentrated grape must as reconstituted, intended for direct human consumption
n.l.c
n.l.c
2.0
Green coffee, dried fruit other than dried vine fruit, beer, cocoa and cocoa products, liqueur wines, meat products, spices and liquorice
n.l.c
n.l.c
-
Processed cereal-based foods and baby foods for infants and young childrenb
0.10
–
0.50
Dietary foods for special medical purposes intended specifically for infants
0.10
–
0.50
a
The maximum levels refer to the edible part of groundnuts and nuts. If groundnuts and nuts ‘in shell’ are analyzed, it is assumed when calculating the aflatoxin content all the contamination is on the edible part. b The maximum level refers to the dry matter. c Not legislated (n.l.): these foods are not included in Regulation for the specific mycotoxin class. This table lists the maximum levels established for aflatoxins B1 (AFB1), B2 (AFB2), G1 (AFG1), and G2 (AFG2), and ochratoxin A (OTA) in foodstuffs in the European Union, according to Commission Regulation (EC) No 1881/2006 and successive amendments and corrections. As can be seen, no maximum level has been set for these mycotoxins in olive oil.
648
2006). The incidence of mycotoxins is also incremented by different factors, such as stress or damage to the crop due to drought before harvest, insect activity, soil type, and inadequate storage conditions. Olive (Olea europaea L.) and its derivatives, in particular olive oil, represent one of the most significant products in the Mediterranean and Middle East regions (Bircan, 2006). Compared with other agricultural commodities, the studies concerning mycotoxin production on olives and consequently their presence in olive oil are few. However, several authors investigated the possibility that olives could support both AFs and OTA production, arriving at contradictory conclusions (Tantaoui-Elaraki et al., 1983; Tantaoui-Elaraki and Le Tutour, 1985; Gourama and Bullerman, 1988; Mahjoub and Bullerman, 1990; Yassa et al., 1994; Eltem, 1996; Leontopoulos et al., 2003; Bircan, 2006; Ghitakou et al., 2006; Roussos et al., 2006). In fact, sometimes olives remain on the soil for a long time after ripening, and are often stored for weeks in conditions that promote mold growth (Daradimos et al., 2000; Papachristou and Markaki, 2004); therefore, the consequent possible presence of mycotoxins in olives can lead to their transfer in oil (Le Tutour et al., 1983; Magnarini et al., 1990; Daradimos et al., 2000; Papachristou and Markaki, 2004; Finoli et al., 2005; Cavaliere et al., 2007; Ferracane et al., 2007). Mycotoxin contamination mainly affects virgin olive oils, since oil refining, and in particular sodium hydroxide treatment, could partially remove AFs (Parker and Melnick, 1966). Crude olive-residue oil obtained by extraction also seems to contain AFs in smaller amounts than virgin oil obtained by pressing (Tantaoui-Elaraki and Le Tutour, 1985).
69.5 CONVENTIONAL ANALYTICAL METHODS FOR AFLATOXINS AND OCHRATOXIN A DETERMINATION IN FOOD In general, fast and easy-to-use enzyme-linked immunosorbent assay (ELISA)-based screening kits are commercially available for all major types of AFs and OTA (Zöllner and Mayer-Helm, 2006). However, ELISA methods are not reliable when used for quantitation, and for legal purposes positive results require confirmation by an accepted reference method (JEFCA, 2002). Several analytical confirmation methods have been developed for determining AFs and OTA in various foodstuffs (cereals, spices, beer, nuts, etc.), animal tissues, and feeds (Zöllner and Mayer-Helm, 2006). Generally, for the extraction of AFs and OTA from these matrices liquid–liquid partitioning or, more recently, solid-phase extraction (SPE) cartridges are used. The most modern clean-up tool for mycotoxin analysis is immunoaffinity chromatography
SECTION | I Adverse Components
(IAC) that allows a selective enrichment and isolation of the target mycotoxin class, thus improving the specificity and sensitivity of the analytical method (Zöllner and Mayer-Helm, 2006; Sforza et al., 2006; Lattanzio et al., 2007; Trucksess et al., 2007). The IAC minicolumns are packed with sepharose into which monoclonal antibodies specific for a single class of compounds are immobilized. Reversed-phase liquid chromatography (RPLC) is currently the technique of choice for mycotoxin analysis. Due to the excellent native fluorescence activity of both the AFs and OTA, very frequently RPLC is coupled to fluorescence detection (FLD). Moreover, AF detection limits can be lowered by post-column derivatization (e.g., by adding iodine or bromine) or pre-column derivatization (by adding a strong organic acid) (Zöllner and Mayer-Helm, 2006; Sforza et al., 2006; Trucksess et al., 2007). Although the official methods for AFs and OTA detection in food are based on IAC RPLC-FLD, nevertheless several mass spectrometry (MS)-based confirmation methods for AFs and OTA determination in different foodstuffs have been published (Lau et al., 2000; Zöllner and MayerHelm, 2006; Sforza et al., 2006; Cavaliere et al., 2007; Lattanzio et al., 2007; Bacaloni et al., 2008). In the mycotoxin field, RPLC-MS seems to be just a minor alternative to the already well-established, reliable and robust RPLCFLD methodology. On the other hand, according to the European Union statement about confirmation analysis for organic residues or contaminants ‘methods based only on chromatographic analysis without the use of spectrometric detection are not suitable on their own for use as confirmatory methods’ (Commission Decision 2002/657/EC).
69.6 ANALYTICAL METHODS FOR AFLATOXINS AND OCHRATOXIN A DETERMINATION IN OLIVE OIL 69.6.1 Sample Preparation Olive oil is constituted for 98–99% by triglycerides; some minor components are free fatty acids, sterols and pigments. Therefore, sample preparation represents a critical step. The most common procedure for extracting AFs and OTA from olive oil is based on sequential liquid extraction and partition with different solvents or solvent mixtures. Le Tutour (Le Tutour et al., 1983) dissolved an oil aliquot in hexane and sequentially extracted AFB1 and OTA with methanol/water containing NaCl and aqueous NaCl. Thereafter, the combined acqueous/methanolic and aqueous extracts were washed with hexane, and cleaned-up using lead acetate. Finally, both mycotoxins were partitioned in chloroform. The same protocol was employed by other authors for AFs and OTA (Finoli et al., 2005; Ferracane et al., 2007) or only AFB1 (Daradimos et al., 2000) isolation
649
CHAPTER | 69 Determination of Aflatoxins and Ochratoxin A in Olive Oil
TIC of + MRM (Turbo Spray) 500
AFM1
400
300
200
AFG1 100
AFB1
AFB2
AFG2
0 11.4
11.6
11.8
12.0
12.2
12.4
12.6
12.8
13.0
13.2
13.4
13.6
13.8
14.0
14.2
14.4
14.6
14.8
Time (min) FIGURE 69.3 The liquid chromatography tandem mass spectrometry (LC/ESI-MS/MS) total ion current (TIC) in multiple reaction monitoring (MRM) acquisition mode obtained by injecting 50 μL of an aflatoxin-free olive oil sample of 320 mg fortified with 0.1 ng g⫺1 of aflatoxins B1 (AFB1), B2 (AFB2), G1 (AFG1), and G2 (AFG2). Aflatoxin M1 (AFM1, the hepatic metabolite of AFB1) was used as internal standard and 0.2 ng were added to the extract. Reprinted from Cavaliere, C., Foglia, P., Guarino, C., Nazzari, M., Samperi, R., and Laganà, A. Determination of aflatoxins in olive oil by liquid chromatography-tandem mass spectrometry. Analytica Chimica Acta, 596, 141–148. Copyright (2007), with permission from Elsevier.
from olive oil. Mixing oil with a hydro-alcoholic mixture and collecting supernatant after complete sedimentation was an alternative extraction procedure (Magnarini et al., 1990; Daradimos et al., 2000; Papachristou and Markaki, 2004). Cavaliere et al. (2007) employed a non-retentive technique, i.e., matrix solid-phase dispersion extraction (MSPDE), for extracting AFB1, AFB2, AFG1, and AFG2 from olive oil. In this case, a very small aliquot of oil was finely dispersed on a low-retention octadecyl (C18) silica bonded adsorbent material. Thereafter, the four naturally occurring AFs were selectively eluted from C18 with a methanol/water mixture in order to avoid the simultaneous elution of non-polar lipids.
69.6.2 Sample Clean-Up For mycotoxin determination in olive oil IAC is the technique chosen by several authors for sample clean-up. It was adopted both for AF and OTA analysis (Magnarini et al., 1990; Daradimos et al., 2000; Papachristou and Markaki, 2004; Finoli et al., 2005). The main drawbacks of this technique are the high cost and the necessity of having a specific antibody for each target mycotoxin class. The more versatile SPE on silica cartridges was also proposed (Daradimos et al., 2000; Ferracane et al., 2007). Other authors did not employ any clean-up step (Cavaliere et al., 2007).
69.6.3 Instrumental Analysis Le Tutour and co-workers (1983) employed thin-layer chromatography (TLC) for determining AFB1 and OTA in olive oil. However, most of the authors dealing with mycotoxin analysis in olive oil chose FLD after RPLC
separation (Magnarini et al., 1990; Daradimos et al., 2000; Papachristou and Markaki, 2004; Finoli et al., 2005; Ferracane et al., 2007). A confirmation method based on RPLC separation coupled to tandem mass spectrometry with electrospray ionization (ESI-MS/MS) allowed determination of AFB1, AFB2, AFG1, and AFG2 even with detection limits 10 times higher than those obtained with FLD (Cavaliere et al., 2007). Figure 69.3 shows the total ion current obtained by analyzing an aflatoxin-free olive oil sample of 0.320 g fortified with 0.1 ng g⫺1 of each of the four AFs by RPLC/ESIMS/MS in multiple reaction monitoring acquisition mode. Aflatoxin M1, the hepatic metabolite of AFB1, was used as internal standard. A mass spectrometric method was also utilized for an unambiguous confirmation of the occurrence of AFB1 and OTA in some olive oils previously analyzed by RPLC-FLD (Ferracane et al., 2007). An overview of the analytical methodologies employed for AFs and OTA determination in olive oil is reported in Table 69.4.
69.6.4 Data on Aflatoxins and Ochratoxin A Contamination in Olive Oil In the literature, results on mycotoxin contamination in olive oil are quite contradictory (Table 69.5), since mycotoxin diffusion depends on the geographic origin of the olives employed for oil production, on the particular climatic conditions of the crop year, on storage conditions, etc. On the other hand, also the capacity of olives to support mold growth is an argument under dispute (Tantaoui-Elaraki et al., 1983; Tantaoui-Elaraki and Le Tutour, 1985; Gourama and Bullerman, 1988; Mahjoub and Bullerman, 1990; Yassa et al.,
650
SECTION | I Adverse Components
TABLE 69.4 Analytical procedures utilized for the determination of aflatoxins and ochratoxin A in olive oil. MDLb (ng kg⫺1)
Reference
90 90–100
4000 40 000
Le Tutour et al., 1983
RPLC-FLD
81 (7) 84 (7) 82 (5) 86 (3)
100–150 100–150 100–150 100–150
Magnarini et al., 1990
Silica SPE
RPLC-FLD
87 (7)
2.8
Daradimos et al., 2000
Magnarini et al., 1990
IAC
RPLC-FLD
85 (18)
56
Daradimos et al., 2000
10
Daradimos et al., 2000
IAC
RPLC-FLD
85 (18)
56
Papachristou and Markaki, 2004
50
Le Tutour et al., 1983
IAC
RPLC-FLD
MSPDE
-
RPLC-MS/MS
Le Tutour et al., 1983
Silica SPE
RPLC-FLD (RPLC-MS/MS)
Mycotoxins
Sample size (g)
Extraction
Clean-up
Instrumental analysis
AFB1 OTA
50
LLE (CH3OH, H2O, C6H6, CHCl3)
–
TLC
AFB1 AFB2 AFG1 AFG2
5
LLE (CH3OH, H2O)
IAC
AFB1
20
Le Tutour et al., 1983
AFB1
10
AFB1 OTA AFB1 AFB2 AFG1 AFG2 OTA
AFB1 AFB2 AFG1 AFG2 AFB1 OTA
0.32
20
Recovery % (RSDa)
108 (5) 82 (3) 84 (4) 85 (2) 82 (6) 87 (5)
4.6 5 2.5 5 2.5 10
92 (11) 102 (7) 98 (8) 96 (13)
50 60 90 20
Cavaliere et al., 2007
250 100
Ferracane et al., 2007
74.6 71.3
Finoli et al., 2005
a
Relative standard deviation. Method detection limit. In this table an overview of the analytical procedures utilized for the determination of aflatoxins B1 (AFB1), B2 (AFB2), G1 (AFG1), G2 (AFG2), and ochratoxin A (OTA) in olive oil is presented. Liquid–liquid extraction (LLE) and matrix solid-phase dispersion extraction (MSPDE) are generally employed for extracting mycotoxins from olive oil; sample clean-up is based on immuno-affinity chromatography (IAC) or solid-phase extraction (SPE). For instrumental analysis, reversed-phase liquid chromatography (RPLC) with fluorescence detection (FLD) or mass spectrometry (MS) detection are the techniques of choice; however, in the past thin-layer chromatography (TLC) was also employed.
b
1994; Eltem, 1996; Leontopoulos et al., 2003; Bircan, 2006; Ghitakou et al., 2006; Roussos et al., 2006). The analysis of 60 samples of crude olive oil from Morocco revealed the absence of AFB1 contamination, while OTA was detected in only three samples, but at concentration levels close to the limit of detection (Le Tutour et al., 1983). Nevertheless, the limits of detection of the method employed were quite high compared to the limits achieved nowadays. More recently, 72% of 50 Greek olive oil samples produced during the period 1995–1998 were found to contain AFB1, even though only one sample showed an appreciable contamination (Daradimos et al., 2000). The same method
applied also to OTA determination on olive oil samples produced in the period 1998–2001, confirmed the low AFB1 contamination level and the widespread OTA contamination in larger amounts (Papachristou and Markaki, 2004). Small traces of AFB1 together with even smaller amounts of AFB2, AFG1, and AFG2 were detected in some olive oils from southern Italy (Finoli et al., 2005). OTA contamination was present too, but in different samples and at a higher concentration than that of AFs. After a survey on commercial oils, Ferracane et al. (2007) concluded that commercial class evaluation of olive oil and its safety level were not correlated. Only three
651
CHAPTER | 69 Determination of Aflatoxins and Ochratoxin A in Olive Oil
TABLE 69.5 Aflatoxin B1 and ochratoxin A presence in different olive oils. AFB1
OTA ⫺1
Incidence (pos/tot)
Range (μg kg )
Incidence (pos/tot)
Range (μg kg⫺1)
Reference
0/60
–
3/60
⬃40a
Le Tutour et al., 1983
36/50
0.003–0.046
–
–
Daradimos et al., 2000
1/50
0.060
44/50
0.005–1.030
Papachristou and Markaki, 2004
7/28
0.005–0.021
7/28
0.052–0.244
Finoli et al., 2005
3/35
⬍0.04b
–
–
Cavaliere et al., 2007
3/30
0.5–2.4
24/30
0.1–17.0
Ferracane et al., 2007
a
Method detection limit. Method quantification limit. The table shows published data relative to surveys on aflatoxin B1 (AFB1) and ochratoxin A (OTA) in olive oils of different origin. Results indicate that there is a great variability in sample contamination.
b
samples were found contaminated by AFB1 and OTA as well. Consequently, it appeared that olives were a poor substrate for AFB1 biosynthesis (Ferracane et al., 2007), and the AF contamination in olive oil was much lower than that reported in the literature (Cavaliere et al., 2007).
SUMMARY POINTS ● ● ●
●
69.7 CONCLUDING REMARKS
●
Scientific data on mycotoxin contamination in olive oil have shown that the presence of OTA is more diffuse and more appreciable than that of AFs. Nevertheless, the contamination levels found in olive oil do not seem to represent any considerable risk to public health. In fact, in the three countries with the largest olive oil consumption, i.e, Greece, Italy, and Spain, this ranges from 11–16 kg per capita per year (source: Faostat 2003, http:// faostat.fao.org), whereas in the other European and North African countries it is less than 5 kg per capita per year. In the same regions, the total consumption of cereals (i.e., the crops most affected by mycotoxin contamination) ranges from 100–250 kg per capita per year. Consequently, it is evident that the intake of AFs and OTA via contaminated olive oil is negligible compared to the intake via other contaminated widespread foods such as wheat and maize. However, AFs and OTA have a high toxicity, and carcinogenic effect, and their simultaneous presence on food could lead to adverse additive or synergic effects. Therefore, as for other agricultural commodities, it is fundamental to prevent molds and then mycotoxin contamination in olives by ensuring optimal cultivation practices and storage conditions.
●
●
Introduction to mycotoxins. Properties and toxicology of aflatoxins and ochratoxin A. International guidelines and European regulations on maximum levels of aflatoxins and ochratoxin A in food. Presence of aflatoxins and ochratoxin A in olive oil. Conventional aflatoxins and ochratoxin A analysis in foods. Analytical methods for determination of aflatoxins and ochratoxin A analysis in olive oil. Data on surveys of aflatoxins and ochratoxin A in olive oils of different origin.
REFERENCES Bacaloni, A., Cavaliere, C., Cucci, F., Foglia, P., Samperi, R., Laganà, A., 2008. Determination of aflatoxins in hazelnuts by various sample preparation methods and liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 1179, 182–189. Becker, M., Degelmann, P., Herderich, M., Schreier, P., Humpf, H.U., 1998. Column liquid chromatography-electrospray ionisation tandem mass spectrometry for the analysis of ochratoxin. J. Chromatogr. A 818, 260–264. Bircan, C., 2006. Determination of aflatoxin contamination in olives by immunoaffinity column using high-performance liquid chromatography. J. Food Quality 29, 126–138. Cavaliere, C., Foglia, P., Guarino, C., Nazzari, M., Samperi, R., Laganà, A., 2007. Determination of aflatoxins in olive oil by liquid chromatography-tandem mass spectrometry. Anal. Chim. Acta 596, 141–148. Commission Decision. 2002/657/EC of 12 August 2002 implementing Council Directive 96/23/EC concerning the performance of analytical
652
methods and the interpretation of results. 2002. Off. J. Eur. Commun. L 221 (August 17), pp. 8–36. Commission Regulation (EC). No 1881/2006 of 19 December 2006 setting maximum levels for certain contaminants in foodstuffs, 2006. Off. J. Eur. Union L 364 (December 20), pp. 5–24. Daradimos, E., Marcaki, P., Koupparis, M., 2000. Evaluation and validation of two fluorimetric HPLC methods for the detemination of aflatoxin B1 in olive oil. Food Addit. Contam. 17, 65–73. Diener, U.L., Cole, R.J., Sanders, T.H., Payne, A., Lee, L.S., Klich, M.A., 1987. Epidemiology of aflatoxin formation by Aspergillus flavus. Annu. Rev. Phytopathol. 25, 249–270. Eltem, R., 1996. Growth and aflatoxin B1 production on olives and olive paste by molds isolated from Turkish-style natural black olives in brine. Int. J. Food Microbiol. 32, 217–223. Ferracane, R., Tafuri, A., Logieco, A., Galvano, F., Balzano, D., Ritieni, A., 2007. Simultaneous determination of aflatoxin B1 and ochratoxin A and their natural occurrence in Mediterranean virgin olive oil. Food Addit. Contam. 24, 173–180. Finoli, C., Vecchio, A., Planeta, D., 2005. Presenza di micotossine in oli extra vergini d’oliva ed olive da mensa (Mycotoxin occurrence in extra virgin olive oils and in olives). Industrie Alimentari 44, 506–514. Ghitakou, S., Koutras, K., Kanellou, E., Markaki, P., 2006. Study of aflatoxin B1 and ochratoxin A production by natural microflora and Aspergillus parasiticus in black and green olives of Greek origin. Food Microbiol. 23, 612–621. Gourama, H., Bullerman, L.B., 1988. Mycotoxin production by molds isolated from ‘Greek-style’ black olives. Int. J. Food Microbiol. 6, 81–90. Hussein, H.S., Brasel, J.M., 2001. Toxicity, metabolism, and impact of mycotoxins on humans and animals. Toxicology 167, 101–134. IARC (International Agency for Research on Cancer). 1993. Some naturally occurring substances: food items and constituents, heterocyclic aromatic amines and mycotoxins. Monograph on the evaluation of carcinogenic risks to humans Vol. 56 IARC WHO, Lyon, France, pp. 489–521. IARC (International Agency for Research on Cancer). 2002. Some traditional herbal medicines, some mycotoxins, naphthalene and styrene. Monograph on the Evaluation of Carcinogenic Risk to Humans, Vol. 82. IARC WHO, Lyon, France, pp. 171–192. ICMSF (International Commission on Microbiological Specifications for Foods), 1996. Toxigenic fungi: Aspergillus. In: Roberts, T.A., BairdParker, A.C., Tompkin, R.B. (Eds.), Microorganisms in Foods 5: Microbiological Specifications of Food Pathogens. Blackie Academic and Professional, an imprint of Chapman & Hall, London, UK, pp. 347–381. JEFCA (Joint FAO/WHO Expert Committee on Food Additives). 2002. Evaluation of certain mycotoxins in food. Fifty-sixth Report of the JEFCA. WHO Technical Report Series 906. WHO, Geneva, Switzerland. Lattanzio, V.M.T., Solfrizzo, M., Powers, S., Visconti, A., 2007. Simultaneous determination of aflatoxins, ochratoxin A and Fusarium toxins in maize by liquid chromatography/tandem mass spectrometry after multitoxin immunoaffinity cleanup. Rapid Commun. Mass Spectrom. 21, 3253–3261. Lau, B.P.Y., Scott, P.M., Lewis, D.A., Kanhere, S.R., 2000. Quantitative determination of ochratoxin A by liquid chromatography/electrospray tandem mass spectrometry. J. Mass Spectrom. 35, 23–32. Le Tutour, B., Tantaoui-Elaraki, A., Ihlal, L., 1983. Simultaneous detection of aflatoxin B1 and ochratoxin A in olive oil. J. Am. Oil Chem. Soc. 60, 835–837. Leontopoulos, D., Siafaka, A., Markaki, P., 2003. Black olives as substrate for Aspergillus parasiticus growth and aflatoxin B1 production. Food Microbiol. 20, 119–126.
SECTION | I Adverse Components
Magnarini, C., Mezzetti, T., Cossignani, L., Cecchetti, V., Gubbiotti, C., Santinelli, F., Burini, G., Damiani, P., 1990. Determinazione RPHPLC di aflatossine in oli alimentari (RP-HPLC determination of aflatoxins in edible oils). Rassegna Chimica 42, 273–276. Mahjoub, A., Bullerman, L.B., 1990. A method for aflatoxin B1 determination in olives. Revue Francaise des Corps. Gras. 37, 245–246. Moss, O.M., 2002. Risk assessment for aflatoxins in foodstuffs. Int. Biodeter. Biodegr. 50, 137–142. Ominski, K.H., Marquardt, R.R., Sinha, R.N., Abramson, D., 1994. Ecological aspects of growth and mycotoxin production by storage fungi. In: Miller, J.D., Trenholm, H.L. (Eds.), Mycotoxins in Grains. Compounds other than Aflatoxin. Eagen Press, St. Paul, MN, USA, pp. 287–305. Papachristou, A., Markaki, P., 2004. Determination of ochratoxin A in virgin oils of Greek origin by immunoaffinity column clean-up and highperformance liquid chromatography. Food Addit. Contam. 21, 85–92. Parker, W.A., Melnick, D., 1966. Absence of aflatoxins from refined vegetable oils. J. Am. Oil Chem. Soc. 43, 635–638. Peraica, M., Radic´, B., Lucic´, A., Pavlovic´, M.B., 1999. Toxic effects of mycotoxins in humans. Bull. World Health Organ. 77, 754–766. Roussos, S., Zaouia, N., Salih, G., Tantaoui-Elaraki, A., Lamrani, K., Cheheb, M., Hassouni, H., Verhé, F., Perraud-Gaime, I., Augur, C., Ismaili-Alaoui, M., 2006. Characterization of filamentous fungi isolated from Moroccan olive and olive cake: toxinogenic potential of Aspergillus strains. Mol. Nutr. Food Res. 50, 500–506. Rustom, I.Y.S., 1997. Aflatoxins in food and feed: occurrence, legislation and inactivation by physical methods. Food Chem. 59, 57–67. Scudamore, K.A., Livesey, C.T., 1998. Occurrence and significance of mycotoxins in forage crops and silage: a review. J. Sci. Food Agr. 77, 1–17. Sforza, S., Dall’Asta, C., Marchelli, R., 2006. Recent advances in mycotoxin determination in food and feed by hyphenated chromatographic techniques/mass spectrometry. Mass Spectrom. Rev. 25, 54–76. Sweeney, M.J., Dobson, A.D.W., 1998. Mycotoxin production by Aspergillus, Fusarium and Penicillium species. Int. J. Food Microbiol. 43, 141–158. Tantaoui-Elaraki, B., Le Tutour, M., Bouzid, M., Keddani, M.J., 1983. Contamination des olives noirs “façon Grèce” par les spores d’Aspergillus toxinogènes et leurs toxines. Industries Alimentaires et Agricoles 100, 997–1000. Tantaoui-Elaraki, A., Le Tutour, B., 1985. Contamination éventuelle des olives et dérivés par les mycotoxines: le point (possible mycotoxin contamination of olives and olive products: latest developments). Oléagineux 40, 451–454. Trucksess, M.W., Weaver, C.M., Oles, C.J., Rump, L.V., White, K.D., Betz, J.M., Rader, J.I., 2007. Use of multitoxin immunoaffinity columns for determination of aflatoxins and ochratoxin A in ginseng and ginger. J. AOAC Int. 90, 1042–1049. Visconti, A., Pascale, M., Centonze, G., 1999. Determination of ochratoxin A in wine by means of immunoaffinity column cleanup and highperformance liquid chromatography. J. Chromatogr. A 864, 89–101. Weidenbörner, M., 2001. Encyclopedia of Food Mycotoxins. Springer, Berlin. Yassa, I.A., Abdalla, E.A.M., Aziz, S.Y., 1994. Aflatoxin B1 production by molds isolated from black table olives. Annal. Agric. Sci. 39, 525–537. Zöllner, P., Mayer-Helm, B., 2006. Trace mycotoxin analysis in complex biological and food matrices by liquid chromatography–atmospheric pressure ionisation mass spectrometry. J. Chromatogr. A 1136, 123–169.
Chapter 70
Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils Sara C. Cunha1, Steven J. Lehotay2, Katerina Mastovska2, José O. Fernandes1 and M. Beatriz P.P. Oliveira1 1 2
REQUIMTE/Serviço de Bromatologia, Faculdade de Farmácia, Universidade do Porto, Portugal U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center, Wyndmoor, PA, USA
70.1 INTRODUCTION The widespread use of pesticides for improving agricultural productivity has raised public concern about the possible presence of residues in food. Laws have been established in each country and internationally in an attempt to protect the health of humans, wildlife, and the environment from harmful effects of pesticides. Monitoring of pesticide residues is done to help ensure that pesticide applications are made in accordance with national and international guidelines. Maximum legal admissible levels, known as the maximum residue limits (MRLs) (or tolerances in the USA), are established depending on each registered pesticide and commodity pair. The trend in regulations regarding pesticide residues in food is that they are becoming more stringent, and the number of active substances for crop treatment is increasing. Olive oil is the principal source of lipids in the Mediterranean diet, and its consumption in the world is increasing due to related potential health benefits, such as a lower incidence of cardiovascular diseases, neurological disorders, breast and colon cancers, as well as its hypolipidemic and antioxidant properties (Gimeno et al., 2002). Olive oil quality is linked to the quality of the raw material, which must be free of defects such as surface blemishes, scars, punctures, or other damages caused by pests. Therefore control of parasites and diseases through pesticide application is important, and pesticide residues can occur in olives, and consequently in olive oil. Thus, MRLs in olives and olive oil have been established domestically in different countries and internationally by Codex Alimentarius (www.mrldatabase.com/query.cfm). In the European Union (EU), the national MRLs are in the process Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
of being harmonized for use within the EU. Thus far, EU MRLs in the range of 0.01–5 mg kg⫺1 have been established for 94 pesticide residues in olives for olive oil production and for 15 residues in table olives (http://ec.europa. eu/food/plant/protection/pesticides/index_en.htm). The large number of pesticides to be monitored, coupled with the typically low concentrations of the MRLs and non-registered residues in food, require sensitive and selective methods for their identification and quantification. Due to its high degree of separation power, gas chromatography (GC) has long been the method of choice, either with selective detectors such as the electron capture detector (ECD), flame photometric detector (FPD), nitrogen phosphorus detector (NPD), or preferentially with mass spectrometric (MS) detection (Lehotay and Hajšlova, 2002). More recently, liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS) or time-of-flight (TOF) MS have gained popularity (Soler and Picó, 2007). The low separation power that traditionally limits LC is overcome by the high selectivity of the MS instruments. The introduction of sub-2-μm particle LC columns, which enable very fast analysis with no or minimal sacrifice in chromatographic resolution, has also improved the ability to perform LC separations, approaching the capabilities of GC (Leandro et al., 2007; Frenich et al., 2008). Regardless of the selectivity and sensitivity in the determinative step, sample preparation (extraction and cleanup) often remains the main limiting step in the analysis of pesticide residues. Thus, several factors need to be taken into consideration when designing a sample preparation method, such as: (i) physicochemical properties of the pesticide residue(s); (ii) detection and quantification limits required; (iii) analytical instrumentation available for use; and (iv) composition of the sample matrices (Lehotay, 2000).
653
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
654
Carbohydrates, proteins, fat and water are the four major components in foods (mineral composition is rarely ⬎5%). Carbohydrates, proteins, and water (and pH) make the biggest differences in the analysis of hydrophilic pesticides from foods, and lipid and water contents typically make the greatest difference in the analysis of lipophilic pesticides. Water nearly always needs to be added prior to extraction in the case of dry foods in order to swell the matrix and allow better solvent access to the residues. Olives contain high levels of lipid substances which can cause problems in pesticide analysis because they are soluble in many organic solvents used for extraction. The lipids must be removed from the extracts (via clean-up) prior to analysis or the chromatographic columns and instruments can be damaged. Particularly in the case of GC, lipid compounds tend to coat the column and injector and slowly degrade into volatile products that adversely affect the detector (Seiber, 1999). The aim of this chapter is to critically review the analysis of pesticide residues in olives and olive oils, with an emphasis on the extraction and clean-up procedures employed before the chromatographic determination. Also, we discuss pesticides used for olive protection and their residue levels found in olive food products and relate it to consumer exposure.
70.2 OLIVE PESTS AND DISEASES AND PESTICIDES USED FOR TREATMENT To obtain high-production and good-quality oils, the olives must be protected from pest attacks and diseases. The major insects attacking olive trees are the olive fruit fly (Bactrocera oleae), the olive-kernel borer or olive moth (Prays oleae), and the black scale (Saissetia oleae). Although B. oleae is considered the most serious insect threat, all three are widely distributed in the Mediterranean region (http://www.tdcolive.net/). Other insect pests can occur in particular areas or conditions that can also cause serious damage, e.g., Euphyllura olivina, Zeuzera pyrina, Aspidiotus nerii, Resseliella oleisuga. Others, e.g., Parlatoria oleae, Leucaspis riccae, Philippia follicularis, although occurring only occasionally, can disrupt the biological balance of the ecosystem. Among olive tree diseases, the most important to address are verticillium wilt, olive knot, leaf spot and fruit mummification. Insecticides and fungicides are used to control these insects and fungal diseases, respectively. Common insecticides used on olives include fenthion, phosmet, dimethoate, methidation, carbaryl, malathion, deltamethrin, which belong to organophosphorus, carbamate, organochlorine, pyrethroid and other chemical classes. Fungicides (e.g., fosetyl-Al, benomyl) often include phthalimides, triazoles, imidazoles, sulfamides and others chemical classes (Tomlin, 2003). Another type of pesticide with large applications in olive groves is herbicides (e.g., sulfonylurea, diphenyl ethers).
SECTION | I Adverse Components
70.3 ANALYTICAL METHODS FOR PESTICIDE RESIDUE DETERMINATION Consumer concerns associated to the stringent regulations and legislation have been responsible, in large part, for the development of new analytical methods to determine pesticide residues in food. Due to the wide scope of pesticide analytes and commodities, and the low concentrations involved, analysis of pesticide residues is one of the most challenging and complex areas of analytical chemistry. In complex matrices such as olives (10–30% lipids) and olive oils (95–100% lipids), sample preparation steps are of critical importance in the overall process. Oftentimes, the analytes of interest are associated with the lipids, which requires extracting the lipids as well. The small difference in the polarity of lipids and many common pesticides makes the choice of a selective extraction solvent difficult. A traditional approach to extract and isolate pesticide residues in olives and olive oils involves the use of a watermiscible solvent (e.g., methanol, acetonitrile) followed by liquid–liquid partitioning with an organic solvent (e.g., n-hexane saturated with acetonitrile) (Cabras et al., 1997; Hiskia et al., 1998; Tsatsakis et al., 2003; Botitsi et al., 2004; Dugo et al., 2005). These extracts are usually subjected to further clean-up, such as solid-phase extraction (SPE) (Rotunno and Argenti, 1997; Rastrelli et al., 2002; Yagüe et al., 2005; Amvrazi and Albanis, 2006), gelpermeation chromatography (GPC) (Barrek et al., 2003; Ballesteros et al., 2006; Guardia-Rubio et al., 2006), or other chromatographic clean-up approach (Díaz-Plaza et al., 2008) before analysis. Despite the quality of the separation of pesticide residues from the matrix, these procedures are time-consuming, laborious, and often require large amounts of potentially hazardous solvents. To overcome these drawbacks, techniques or methods, such as solid-phase microextraction (SPME), matrix solidphase dispersion (MSPD), and QuEChERS (quick, easy, cheap, effective, rugged, and safe), have been employed in the analysis of pesticide residues in olives and olive oil. These approaches require no or very small solvent volumes and often combine extraction and clean-up into a single step.
70.3.1 SPME SPME involves a sorbent coating placed on a glass fiber device, and chemicals are adsorbed onto or absorbed into the sorbent through an equilibrium process (Arthur et al., 1992; Pawliszyn, 1997). The fiber can be placed into a liquid sample or into the headspace above it (closed system). The headspace approach is often superior because pesticides are often more volatile than the matrix and thus the sorbent does not extract so many matrix interferences. The headspace approach also extends the limited lifespan of the SPME device. The extraction step is influenced by
CHAPTER | 70 Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils
parameters that affect the distribution between the chemicals and the sorbent coating, such as physicochemical properties (e.g., partitioning coefficients), volumes (both of the sample and coating), temperature, time, sample agitation, pH, ionic strength, and other matrix effects. After extraction, a desorption step is needed for analysis by GC or LC. In GC, desorption is done quickly at high temperature and rapid gas flow rate, and in LC, a strong solvent is typically employed for rapid desorption (O’Reilly et al., 2005). Tsoutsi and Albanis (2004) studied and compared the SPME extraction capacity of four fiber coatings, polydimethylsiloxane (PDMS), polyacrylate, PDMS-divinylbenzene, and Carbowax-PDMS, in the extraction of seven organophosphorus pesticides (ethion, bromophos, bromophosmethyl, diazinon, fenthion, parathion, and fenitrothion) from olive oils. PDMS by itself showed the better extraction efficiency for the analytes. This approach can be automated and has advantages of generating no waste, being easy to perform, and using little space or labware. However, SPME is not suitable for solid samples, is not necessarily fast or sensitive (depending on the pesticide and matrix), is not easily quantitative (large matrix effects), requires relatively costly devices, and has a narrow scope of pesticides for which it is effective.
70.3.2 MSPD MSPD is based on thorough matrix homogenization with a solid sorbent powder. The mixed sample-sorbent material is then packed into a column and a solvent is used to elute chemicals (e.g., pesticide analytes) from the mixture. Ideally, chemical interferences are retained on the sorbent. Ferrer et al. (2005) used an aminopropyl sorbent in MSPD followed by additional clean-up with Florisil during elution of 12 pesticides at trace levels in olives and olive oils. Recently, a similar approach was published for the same matrices by Cunha et al. (2007a) for extraction of fenthion and its metabolites. The same authors also evaluated C18 as the sorbent in the MSPD extraction of phosmet and its metabolites in olives and olive oils (Cunha et al., 2007b).
which already contains water, or water is added, followed by addition of anhydrous MgSO4 and NaCl to induce partitioning of the extract (upper phase) from the water (lower phase). The extraction is typically conducted by shaking, but blending is also an option. The use of buffering with either acetic acid/acetate (Lehotay et al., 2005c) or citric acid/citrate salts (Payá et al., 2007; www.quechers.com) during extraction is also an option depending on the analytes and matrices. After centrifugation, an aliquot of the extract is mixed with sorbent(s) in a centrifuge tube in an approach known as dispersive-SPE clean-up. The final extract is analyzed for pesticides by chromatographic methods, with detection ideally using MS techniques. As outlined in Figure 70.1, a QuEChERS method was applied for residue analysis in olives and olive oils (Cunha et al., 2007c; Hernando et al., 2007). The main difference from the original method was the use of the different sorbents [C18, primary secondary amine (PSA), and graphitized carbon black (GCB)] instead of PSA alone. Frenich et al. (2008) also successfully used an extraction based on the buffered QuEChERS approach (Lehotay et al., 2005c) in food commodities such as olives. Lehotay et al. (2005a) made a comparison of MSPD with QuEChERS in the analysis of many representative pesticides in semi-fatty foods (milk, eggs, and avocado). Each approach has advantages and disadvantages with respect to each other, depending on the pesticide and matrix. Both MSPD and QuEChERS have the advantages of being simple, rapid, and inexpensive procedures requiring little labor and few materials, space, and solvents. Both give recoveries that are affected by lipophilicity and hydrophilicity of the pesticide analytes in the fatty/watery matrix. A key advantage is that QuEChERS has wider scope of analysis and proven quality of results. Two similar QuEChERS methods have achieved the status of Official
QuEChERS method Weigh out 3 g of olive oil (add 7 mL water) or 10 g olives in a 50 mL centrifuge tube
Add 4 g anh. MgSO4 + 1 g NaCl Shake vigorously for 1 min and centrifuge
Clean-up
QuEChERS was introduced by Anastassiades et al. (2003) for the multiclass, multiresidue analysis of pesticides in fruits and vegetables. It is a very flexible approach that has been adapted for the analysis of many pesticide residues in food and animal tissues (Lehotay et al., 2005a, b; Lehotay, 2007; Nguyen et al., 2007; Payá et al., 2007; Cajka et al., 2008) as well as in other applications (Posyniak et al., 2005; Mastovska and Lehotay, 2006; Plössl et al., 2006). The main concept entails the use of an organic solvent, typically acetonitrile but alternatively ethyl acetate or acetone, for extraction of the pre-homogenized sample,
Extraction
Add 10 mL of acetonitrile
70.3.3 QuEChERS
655
Take 1 mL of the upper layer and mix it with 150 mg MgSO4 and 50 mg each of PSA, C18 (and GCB) Shake vigorously for 20 s and centrifuge
GC-MS and LC-MS/MS analysis
FIGURE 70.1 QuEChERS sample preparation procedure for pesticide residues in olives or olive oil.
656
Method of AOAC International (#2007.01) and European Committee for Standardization (CEN) (#EN 15662). This makes QuEChERS a ‘gold standard’ method for pesticide residue analysis of fruits and vegetables, with applicability to other foods with proper validation.
70.3.3 Analysis The determination of pesticide residues in olives and olive oils has traditionally been performed by GC using ECD and NPD (Cabras et al., 1997; Rastrelli et al., 2002; Tsatsakis et al., 2003; Amvrazi and Albanis, 2006; Yagüe et al., 2006). Lately, usage of GC-MS has increased, which permits: (i) the simultaneous quantification and identification of detected analytes; (ii) the detection of a wide range of analytes independently of elemental composition; (iii) spectrometric resolution of co-eluting peaks; and (iv) potentially faster analysis times (Barrek et al., 2003; Ferrer et al., 2005; Cunha et al., 2007b, c). Pesticide residues in olives and olive oil have also been analyzed by GC-MS/MS (Guardia-Rubio et al., 2006, 2007; Frenich et al., 2007), which further improves the selectivity, sensitivity, and the ability to make identifications of targeted analytes. Despite the recent advances in GC-MS systems, the analysis of polar, non-volatile and/or thermally labile pesticides by this technique is limited, usually requiring chemical derivatization. LC-MS/MS has become a standard approach in developed countries to expand on the range of pesticides quantified and identified in complex matrices, and examples in the case of olives and olive oils have been published (Barrek et al., 2003; Ferrer et al., 2005; Cunha et al., 2007c; Hernando et al., 2007; Frenich et al., 2008).
70.4 PESTICIDE LEVELS AND HUMAN EXPOSURE Several experiments have studied the persistence of pesticides applied in olive groves and the subsequent residue levels in the olives and olive oil produced. Table 70.1 shows the concentrations of pesticide residues in olives and olive oils reported by different authors in the last decade. Most of these studies pertain to olive oils, but a few concern olives only. Cabras et al. (1997, 2002) showed that azinphos-methyl, diazinon, methidation, parathion-methyl, rotenone and quinalphos had higher residue levels in olive oils than in the olives, which can be explained by the need for ⬇5 kg of olives to obtain ⬇1 L of oil. The lipophilic pesticides have higher affinity to the oil than the pulp. In contrast, the more polar pesticide, dimethoate, does not occur in the oil during processing (Cabras et al., 1997). Table 70.1 also demonstrates that the residue levels of the same pesticides can vary greatly. Sample conditions, the type and number of treatments made, environmental
SECTION | I Adverse Components
factors, and the type of processing have a large effect on the results. In any event, nearly all the samples presented residue concentrations below the EU or Codex MRLs, with the exceptions of six violations for dimethoate (Hiskia et al., 1998; Rastrelli et al., 2002; Tsatsakis et al., 2003) and two for endosulfan sulfate (Amvrazi and Albanis, 2006). Another important aspect is that the presence of metabolites can be more toxic than the parent pesticides. Cunha et al. (2007a) performed the identification and quantification of fenthion and its metabolites in olives and olive oils from an olive grove where treatment was carried out at the recommended dose. All samples (collected during the preharvest interval) had a common qualitative pattern, showing four identifiable substances (fenthion, fenthion sulfoxide, fenoxon and fenoxon sulfoxide). Fenthion and the three metabolites were still found in olives at harvest time. The same authors have studied the presence of metabolites of phosmet in olives obtained in similar condition as described for fenthion. They found phosmet, phosmet-oxon, phthalimide, N-hydroxymethylphthalimide and phthalic acid in olives and olive oil (Cunha et al., 2007b). In order to measure the heath risks from olive oil ingestion, the estimated daily intakes (EDI) for pesticides are generally very low compared to their corresponding acceptable daily intake (ADI), as shown in Table 70.1. Therefore, the risk associated with the pesticide exposure via olives and olive oil consumption is very low.
70.5 CONCLUSION The use of pesticides to control pests and diseases has contributed to the increase in the production of olive oil in Mediterranean countries. However, the use of pesticides is also associated with certain risks. In order to meet regulatory requirements and protect the consumer and the environment, many analytical methods have been reported to monitor and control residue levels in olives and olive oils. The use of classical extraction techniques requiring large volumes of harmful solvent has been reduced by new techniques such as SPME, MSPD and QuEChERS. These methods are ideally suited for the powerful features of GC-MS and LC-MS/MS for analysis and can provide rapid, reliable and sensitive analysis of pesticides in food matrices, such as olives and olive oils. Based on the reported levels, human exposure to pesticides through the consumption of olives and olive oil is very low.
SUMMARY POINTS ●
Analytical methods are reviewed for the determinations of pesticide residues in olives and olive oil, with emphasis on the modern approaches for sample preparation, including QuEChERS.
Pesticide
Acephate
Chemical group
OP
Product
VOO
No treat.
-c
Azinphos-ethyl
OP
VOO
-
Azinphos-methyl
OP
O and OO
1
Azadirachtin
Carbaryl
Chlorpyrifos
TT
CR
OP
O
Days after last treat.
No samples analyzed
-c
19
-
No samples positive 1
Residues mean value (μg kg⫺1) Olives -c
Olive oils 2.9
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
30
0.0052
Spain
Yagüe et al., 2005
Italy
Rastrelli et al., 2002
Italy
Cabras et al., 1997
Italy
Caboni et al., 2002
Spain
Ballesteros et al., 2006
c
-
-
c
65
2
-
1
-c
-c
1820
4570.0
8
-
-
1030
3100
14
-
-
690
0
-
-
350
-
1
-
-
280
-
2
-
-
100
-
3
-
-
30
-
-
7
-
-
⬍20
-
-
4
90.0
ADIa (μg kg⫺1 b.w.)
8.226 5
1620.0
5.580 2.916
-
-
-
-
OPO
-
-
10
8
-
3.9
0.007
ROO
-
-
15
5
-
2.9
VOO
-
-
15
15
-
4.5
0.0081
VOO from individual growers
-
-
48
4
-
42.7
0.0769
Greece
Hiskia et al., 1998
VOO
-
-
20
7
-
10.4
0.0187
Greece
Amvrazi and Albinis, 2006
OO
3
14
22
-
-
1017.4
OPO
-
-
10
3
-
8.8
0.0158
ROO
-
-
15
1
-
6.9
0.0124
VOO
-
-
15
4
-
10.7
0.0193
8
10
0.0052
1.8313
Sanchez et al., 2006 Spain
Ballesteros et al., 2006
CHAPTER | 70 Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils
TABLE 70.1 Pesticide residues in olives and olive oil: Number of treatments (no treat.), days after the last treatment, number of samples analyzed, and the number of positives.
(Continued)
657
658
TABLE 70.1 (Continued) Pesticide
Chlorpyrifos-methyl
α–Cyhalothrin
α–Cypermethrin
Chemical group
OP
PYR
PYR
Product
No treat.
Days after last treat.
No samples analyzed
No samples positive
Residues mean value (μg kg⫺1) Olives
Olive oils
ADIa (μg kg⫺1 b.w.)
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
10
0.1440
Italy
Rastrelli et al., 2002
VOO
-
-
65
4
-
80.0
OPO
-
-
10
5
-
1.5
0.0027
ROO
-
-
15
3
-
1.7
0.0031
VOO
-
-
15
8
-
3.4
0.0061
VOO
-
-
20
3
-
19.5
OPO
-
-
10
8
-
2.6
0.0047
ROO
-
-
15
11
-
1.2
0.0022
VOO
-
-
15
14
-
1.4
0.0025
VOO
-
-
20
1
-
48.9
OPO
-
-
10
9
-
ROO
-
-
15
10
VOO
-
-
15
2
Spain
Grece
Amvrazi and Albinis, 2006
Spain
Ballesteros et al., 2006
0.0880
Grece
Amvrazi and Albinis, 2006
4.1
0.0074
Spain
Ballesteros et al., 2006
-
2
0.0036
12
-
2.3
0.0041 0.0032
Spain
Yagüe et al., 2005
2
0.0351
Ballesteros et al., 2006
OC
VOO
-
-
19
4
-
1.8
ρ,ρ- DDT
OC
VOO
-
-
19
8
-
5.2
10
0.0094
Spain
Yagüe et al., 2005
Deltamethrin
PYR
VOO
-
-
20
3
-
45.2
10
0.814
Grece
Amvrazi and Albinis, 2006
OPO
-
-
10
3
-
1.6
0.0029
Spain
Ballesteros et al., 2006
O and OO
1
1
-
-
1340
4430
7.974
Italy
Cabras et al., 1997
8
-
-
1110
3780
6.804
13
-
-
680
2150
3.87
20
-
-
350
1950
3.51
-
-
65
3
-
-
-
20
2
-
Diazinon
OP
VOO
83 3.3
2
0.1494 0.0059
Rastrelli et al., 2002 Grece
Amvrazi and Albinis, 2006
SECTION | I Adverse Components
ρ,ρ’- DDE
TABLE 70.1 (Continued) Pesticide
Dimethoate
Diuron
PYC
OP
U
Product
No treat.
Days after last treat.
No samples analyzed
No samples positive
Residues mean value (μg kg⫺1) Olives
Olive oils 1.9
ADIa (μg kg⫺1 b.w.)
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
-
-
Spain
Ballesteros et al., 2006
Italy
Cabras et al., 1997
Greece
Hiskia et al., 1998
OPO
-
-
10
7
-
ROO
-
-
15
12
-
3.2
-
VOO
-
-
15
13
-
4.5
-
O and OO
1
1
-
-
1600
530
8
-
-
1080
240
14
-
-
170
2
0.954 0.432
-
VOO from individual growers
-
-
48
29
-
65
0.117
VOO
-
-
14
6
-
10.2
-
-
-
65
29
-
61
0.1098
Italy
Rastrelli et al., 2002
OO from organic cultivation
-
-
53
-
-
4.9
0.088
Greece
Tsatsakis et al., 2003
OO from conventional cultivation
-
-
50
-
-
25.4
0.0457
VOO
-
-
20
16
-
6.6
0.0119
OPO
-
-
10
2
-
1.3
0.0023
VOO
-
-
15
2
-
1.3
0.0023
OO from soil olives
-
-
10
9
-
68.8
OO from not separated olives
-
-
5
3
-
22.6
0.0407
OO from flight olives
-
-
9
4
-
14
0.0252
OPO
-
-
10
4
-
2.3
0.0041
ROO
-
-
15
9
-
1.4
0.0025
VOO
-
-
15
15
-
20.1
0.0362
2
0.1238
Amvrazi and Albinis, 2006 Spain
Ballesteros et al., 2006
Spain
Guardia-Rubio et al., 2006
Ballesteros et al., 2006
CHAPTER | 70 Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils
Diflufenican
Chemical group
(Continued)
659
660
TABLE 70.1 (Continued) Pesticide
Endosulfan sulfate
α–Endosulfan
β–Endosulfan
Chemical group
OC
OC
OC
ADIa (μg kg⫺1 b.w.)
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
6
0.0056
Spain
Yagüe et al., 2005
28.1
0.0506
Greece
Amvrazi and Albinis, 2006
-
13.3
0.0239
Spain
Guardia-Rubio et al., 2006
1
-
4.6
0.0083
9
5
-
24.8
0.0446
-
10
9
-
7.9
0.0142
-
-
15
12
-
7
0.0126
VOO
-
-
15
15
-
18.7
0.0337
OO
-
-
-
1
-
30
-
VOO
-
-
20
4
-
6.7
-
-
15
1
-
6.3
OO
-
-
8
1
-
VOO
-
-
20
4
-
7.8
-
Greece
Amvrazi and Albinis, 2006
-
-
15
1
-
8.7
-
Spain
Ballesteros et al., 2006
2
-
24.1
2
0.0434
Greece
Amvrazi and Albinis, 2006
No samples positive
Days after last treat.
No samples analyzed
-
-
19
2
-
3.1
-
-
20
6
-
OO from soil olives
-
-
10
3
OO from not separated olives
-
-
5
OO from flight olives
-
-
OPO
-
ROO
Product
VOO
No treat.
Residues mean value (μg kg⫺1) Olives
Olive oils
-
30
Ballesteros et al., 2006
Fontcuberta et al., 2008
-
Greece
Amvrazi and Albinis, 2006
-
Spain
Ballesteros et al., 2006
-
Frenich et al., 2007
OP
VOO
-
-
20
Fenitrothion
OP
OO
3
14
22
-
-
2349.1
5
4.2284
Spain
Sanchez et al., 2006
Fenthion
OP
VOO from individual growers
-
-
48
17
-
47.3
7
0.0851
Greece
Hiskia et al., 1998
VOO
-
-
14
11
-
44.3
0.0797
-
-
65
18
-
73
0.1314
Italy
Rastrelli et al., 2002
SECTION | I Adverse Components
Ethion
TABLE 70.1 (Continued)
Fenthion sulfoxide
Chemical group
OP
Product
No treat.
Days after last treat.
No samples analyzed
No samples positive
Residues mean value (μg kg⫺1) Olives
Olive oils 11.6
ADIa (μg kg⫺1 b.w.)
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
0.0209
Greece
Tsatsakis et al., 2003
OO from organic cultivation
-
-
53
-
-
OO from conventional cultivation
-
-
50
-
-
VOO
-
-
20
18
-
OO from individual growers of Apulian
-
-
28
6
-
350
-
Italy
Dugo et al., 2005
OO from individual growers of Silician
-
-
51
7
-
200
-
Italy
Dugo et al., 2005
OO
-
-
48
36
-
0.0841
Greece
Botitsi et al., 2004
O
1
7
-
-
20
Portugal
Cunha et al., 2007a
14
-
-
10
21
-
-
20
35
-
-
20
42
-
-
20 Greece
Amvrazi and Albinis, 2006
146
0.2628
17.6
0.0317
46.7
VOO
-
-
20
17
-
23.3
-
0.0419
OO
-
-
48
33
-
126.7
-
0.2281
O
1
7
-
-
30
-
-
14
-
-
20
-
-
21
-
-
30
-
-
35
-
-
20
-
-
42
-
-
10
-
-
Amvrazi and Albinis, 2006
Botitsi et al., 2004 Portugal
Cunha et al., 2007a
CHAPTER | 70 Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils
Pesticide
(Continued)
661
662
TABLE 70.1 (Continued) Pesticide
Chemical group
Product
No treat.
Days after last treat.
No samples analyzed
No samples positive
Residues mean value (μg kg⫺1) Olives
Olive oils
ADIa (μg kg⫺1 b.w.)
Fenthion sulfone
OP
OO
-
-
48
21
-
36.7
-
Fenoxon
OP
O
1
14
-
-
20
-
-
21
-
-
10
-
-
35
-
-
10
-
-
42
-
-
10
-
-
14
-
-
30
-
-
21
-
-
50
-
-
35
-
-
50
-
-
42
-
-
30
-
-
Fenoxon sulfoxide
OP
O
1
EDIb Olive oil (μg kg⫺1 b.w.) 0.0661
Country
Reference
-
Botitsi et al., 2004
Portugal
Cunha et al., 2007a
Portugal
Cunha et al., 2007a
PYR
VOO
-
-
20
2
-
-
20
-
Amvrazi and Albinis, 2006
Fenvalerate II
PYR
VOO
-
-
20
2
-
-
-
-
Amvrazi and Albinis, 2006
Formothion
OP
VOO
-
-
65
1
-
82
-
-
Italy
Rastrelli et al., 2002
α ⫺ Hexachlorocyclohexane
OC
VOO
-
-
19
5
-
1.5
-
-
Spain
Yagüe et al., 2005
β ⫺ Hexachlorocyclohexane
OC
VOO
-
-
19
1
-
2.3
-
-
Spain
Yagüe et al., 2005
Lindane
OC
VOO
-
-
19
9
-
3.1
1
0.0056
Spain
Yagüe et al., 2005
Malathion
OP
VOO from individual growers
-
-
48
1
-
19.7
300
0.0355
Greece
Hiskia et al., 1998
Methidathion
OP
O and OO
1
1
-
-
3010
6780
1
Italy
Cabras et al., 1997
8
-
-
1680
5690
10.242
14
-
-
1280
3370
6.066 Greece
Hiskia et al., 1998
-
Rastrelli et al., 2002
12.204
VOO from individual growers
-
-
48
2
-
63
0.1134
VOO
-
-
14
1
-
3
0.0054
-
-
65
3
-
63
0.1134
SECTION | I Adverse Components
Fenvalerate I
TABLE 70.1 (Continued) Pesticide
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
6.6
0.0119
Greece
Amvrazi and Albinis, 2006
-
3064.7
5.5165
Spain
Sanchez et al., 2006
0.0414
Greece
Hiskia et al., 1998
0.0061
Spain
Ballesteros et al., 2006
Greece
Hiskia et al., 1998
Days after last treat.
No samples analyzed
No samples positive
Residues mean value (μg kg⫺1)
-
-
20
4
-
OO
3
14
22
-
Product
No treat.
Olives
Olive oils
ADIa (μg kg⫺1 b.w.)
Omethoate
OP
VOO from individual growers
-
-
48
5
-
23.0
Oxyfluorfen
DE
OPO
-
-
10
5
-
3.4
ROO
-
-
15
7
-
2.1
0.0038
VOO
-
-
15
14
-
8.8
0.0158
VOO from individual growers
-
-
48
3
-
69.6
VOO
-
-
14
3
-
7
0.0126
-
-
65
2
-
80
0.144
Italy
Rastrelli et al., 2002
1
-
-
1400
4000
7.2
Italy
Cabras et al., 1997
8
-
-
610
2910
5.238
13
-
-
350
1770
3.186
20
-
-
190
1330
2.394 Greece
Hiskia et al., 1998
Parathion
Parathion-methyl
Phosmet
Pirimiphos-methyl
OP
OP
OP
OP
O and OO
1
-
3
4
3
0.1253
VOO from individual growers
-
-
48
3
33.5
0.0603
VOO
-
-
14
2
-
5.5
0.0099
-
-
65
1
-
56
0.1008
Italy
Rastrelli et al., 2002
-
-
20
4
-
10
0.018
Greece
Amvrazi and Albinis, 2006
OPO
-
-
10
3
-
1.5
0.0027
Spain
Ballesteros et al., 2006
ROO
-
-
15
1
-
1.0
O
1
7
-
Portugal
Cunha et al., 2007b
ROO
-
-
15
2
-
1.2
0.0022
Spain
Ballesteros et al., 2006
VOO
-
-
15
1
-
2.6
10
0.0018
210 30
CHAPTER | 70 Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils
Chemical group
0.0047
(Continued) 663
664
TABLE 70.1 (Continued) Pesticide
Rotenone
Quinalphos
Terbuthylazine
Chemical group
PYR
OP
T
Product
O and OO
O and OO
No treat.
Days after last treat.
Residues mean value (μg kg⫺1)
ADIa (μg kg⫺1 b.w.)
EDIb Olive oil (μg kg⫺1 b.w.)
Country
Reference
-
-
Italy
Cabras et al., 2002
Italy
Cabras et al., 1997
Spain
Guardia-Rubio et al., 2006
No samples analyzed
No samples positive
0
-
-
990
2
-
-
520
1890
-
5
-
-
440
1050
-
9
-
-
190
510
-
12
-
-
110
530
-
1
-
-
1840
2630
8
-
-
680
2130
-
13
-
-
360
500
-
20
-
-
200
800
2
1
Olives
Olive oils
-
-
OO from soil olives
-
-
10
9
-
127.5
2.2
0.2295
OO from not separated olives
-
-
5
5
-
83.4
0.1501
OO from flight olives
-
-
9
7
-
45.5
0.0819
OPO
-
-
10
5
-
7.5
0.0135
ROO
-
-
15
6
-
4.4
0.0079
VOO
-
-
15
15
-
56.5
0.1017
Ballesteros et al., 2006
T
OPO
-
-
10
2
-
1.7
27
0.0031
Spain
Ballesteros et al., 2006
Trichlorfon
OP
VOO
-
-
15
1
-
5.2
20
0.0094
Spain
Ballesteros et al., 2006
OP-Organophosphate, T-Triazine, CR-Carbamate, PYR-Pyrethroid, OC-Organochlorine, PYC-Pyridinecarboxamide, U-Urea, DE-Diphenyl Ether, T-Triazine, TT-Tetranortriterpenoid; VOO-Virgin olive oil; O-Olives, OO-Olive oil, ROO-Refined olive oil, OPO-Olive-pomace oil a –Acceptable Daily Intake (g/kg body weight). Source: Tomlin, C.D.S. (2003). b – Estimated Daily Intake ⫽ Olive oil consumption in Europe (g kg⫺1 body weight) x residue (μg kg⫺1). Source: http://www.internationaloliveoil.org c – not referenced
SECTION | I Adverse Components
Terbutryn
CHAPTER | 70 Sample Preparation Approaches for the Analysis of Pesticide Residues in Olives and Olive Oils
●
●
The major olive pests and diseases and the pesticides used for their treatment are described. Pesticide residues levels found in olives and olive oils are listed from the literature, and the health risks estimated from pesticide residues in olive oil.
REFERENCES Amvrazi, E.G., Albanis, T.A., 2006. Multiresidue method for determination of 35 pesticides in virgin olive oil by using liquid-liquid extraction techniques coupled with solid-phase extraction clean up and gas chromatography with nitrogen phosphorus detection and electron capture detection. J. Agric. Food Chem. 54, 9642–9651. Anastassiades, M., Lehotay, S.J., Štajnbaher, D., Schenck, F.J., 2003. Fast and easy multiresidue method employing acetonitrile extraction/partitioning and “dispersive solid-phase extraction” for the determination of pesticide residues in produce. J. AOAC Int. 86, 412–431. Arthur, C.L., Killam, L.M., Buchholz, K.D., Pawliszyn, J., Berg, J.R., 1992. Automation and optimization of solid-phase microextraction. Anal. Chem. 64, 1960–1966. Ballesteros, E., Sánchez, G.A., Martos, N.R., 2006. Simultaneous multidetermination of residues of pesticides and polycyclic aromatic hydrocarbons in olive and olive-pomace oils by gas chromatography/ tandem mass spectrometry. J. Chromatogr. A 1111, 89–96. Barrek, S., Paisse, O., Grenier-Loustalot, M.-F., 2003. Determination of residual pesticides in olive oil by GC-MS and HPLC-MS after extraction by size-exclusion chromatography. Anal. Bioanal. Chem. 376, 355–359. Botitsi, E., Kormali, P., Kontou, S., Mourkojanni, A., Stavrakaki, E., Tsipi, D., 2004. Monitoring of pesticide residues in olive oil samples: results and remarks between 1999 and 2002. Int. J. Environ. Anal. Chem. 84, 231–239. Caboni, P., Cabras, M., Angioni, A., Russo, M., Cabras, P., 2002. Persistence of azadirachtin residues on olives after field treatment. J. Agric. Food Chem. 50, 3491–3494. Cabras, P., Angioni, A., Garau, V.L., Melis, M., Pirisi, F.M., Karim, M., Minelli, E.V., 1997. Persistence of insecticide residues in olives and olive oil. J. Agric. Food Chem. 45, 2244–2247. Cabras, P., Caboni, P., Cabras, M., Angioni, A., Russo, M., 2002. Rotenone residues on olives and in olive oil. J. Agric. Food Chem. 50, 2576–2580. Cajka, T., Hajslova, J., Lacina, O., Mastovska, K., Lehotay, S.J., 2008. Rapid analysis of multiple pesticide residues in fruit-based baby food using programmed temperature vaporiser injection-low-pressure gas chromatography-high-resolution time-of-flight mass spectrometry. J. Chromatogr. A 1186, 281–294. Cunha, S.C., Fernandes, J.O., Oliveira, M.B.P.P., 2007b. Determination of phosmet and its metabolites in olives by matrix solid-phase dispersion and gas-chromatography mass spectrometry. Talanta 73, 514–522. Cunha, S.C., Fernandes, J.O., Oliveira, M.B.P.P., 2007a. Comparison of matrix-solid phase dispersion and liquid-liquid extraction for the chromatographic determination of fenthion and its metabolites in olives and olive oils. Food Add. Cont. 24, 156–164. Cunha, S.C., Lehotay, S.J., Mastovska, K., Fernandes, J.O., Oliveira, M. B.P.P., 2007c. Evaluation of the QuEChERS sample preparation approach for the analysis of pesticide residues in olives. J. Sep. Sci. 30, 620–632. Díaz-Plaza, E.M., Cortés, J.M., Vázquez, A., Villén, J., 2008. Automated determination of pesticide residues in olive oil by on-line reversed-phase
665
liquid chromatography using the through oven transfer adsorption desorption interface with electron-capture and nitrogen-phosphorus detectors operating simultaneously. J. Chromatogr. A 1174, 145–150. Dugo, G., Di Bella, G., Torre, L., Saitta, M., 2005. Rapid GC-FPD determination of organophosphorus pesticide residues in Sicilian and Apulian olive oil. Food Control 16, 435–438. Ferrer, C., Gomez,, M.J., Garcia-Reyes,, J.F., Ferrer, I., Thurman, E.M., Fernandez-Alba,, A.R., 2005. Determination of pesticide residues in olives and olive oil by matrix solid-phase dispersion followed by gas chromatography/mass spectrometry and liquid chromatography/tandem mass spectrometry. J. Chromatogr. A 1069, 183–194. Fontcuberta, M., Arqués, J.F., Villalbí, J.R., Martínez, M., Centrich, F., Serrahima, E., Pineda, L., Duran, J., Casas, C., 2008. Chlorinated organic pesticides in marketed food: Barcelona, 2001–06. Sci. Total Environ. 389, 52–57. Frenich, A.G., Moreno, J.L.F., Vidal, J.L.M., Liébanas, F.J.A., 2007. Application of gas chromatography coupled to triple quadrupole mass spectrometry for the multiresidue analysis of pesticides in olive oil. J. Agric. Food Chem. 55, 8346–8352. Frenich, A.G., Vidal, J.L.M., Pastor-Montoro, E., Romero-González, R., 2008. High-throughput determination of pesticide residues in food commodities by use of ultra-performance liquid chromatography-tandem mass spectrometry. Anal. Bioanal. Chem. 390, 947–959. Gimeno, E., Fitó, M., Lamuela-Raventós, R.M., Castellote, A.I., Covas, M., Farré, M., 2002. Effect of ingestion of virgin olive oil on human low-density lipoprotein composition. Europ. J. Clin. Nut. 56, 114–120. Guardia-Rubio, M., Fernandez-De Cordova,, M.L., Ayora-Canada, M.J., Ruiz-Medina, A., 2006. Simplified pesticide multiresidue analysis in virgin olive oil by gas chromatography with thermoionic specific, electron-capture and mass spectrometric detection. J. Chromatogr. A 1108, 231–239. Guardia-Rubio, M., Marchal-Lopez,, R.M., Ayora-Canada,, M.J., RuizMedina, A., 2007. Determination of pesticides in olives by gas chromatography using different detection systems. J. Chromatogr. A 1145, 195–203. Hernando, M.D., Ferrer, C., Ulaszewska, M., Garcia-Reyes,, J.F., MolinaDiaz, A., Fernandez-Alba,, A.R., 2007. Application of high-performance liquid chromatography-tandem mass spectrometry with a quadrupole/ linear ion trap instrument for the analysis of pesticide residues in olive oil. Anal. Bioanal. Chem. 389, 1815–1831. Hiskia, A.E., Atmajidou, M.E., Tsipi, D.F., 1998. Determination of organophosphorus pesticide residues in Greek virgin olive oil by capillary gas chromatography. J. Agric. Food Chem. 46, 570–574. Leandro, C.C., Hancock, P., Fussell, R.J., Keely, B.J., 2007. Ultraperformance liquid chromatography for the determination of pesticide residues in foods by tandem quadrupole mass spectrometry with polarity switching. J. Chromatogr. A 1144, 161–169. Lehotay, S.J., 2007. Determination of pesticide residues in foods by acetonitrile extraction and partitioning with magnesium sulfate: collaborative study. J AOAC Int. 90, 485–520. Lehotay, S.J., 2000. Multiclass, multiresidue analysis of pesticides, strategies for. In: Meyers, R.A. (Ed.), Encyclopedia of Analytical Chemistry. John Wiley and Sons, Chichester, UK, pp. 6344–6884. Lehotay, S.J., Hajšlova , J., 2002. Application of gas chromatography in food analysis. Trends Anal. Chem. 21, 686–697. Lehotay, S.J., Mastovska, K., Yun, S.J., 2005a. Evaluation of two fast and easy methods for pesticide residue analysis in fatty food matrixes. J. AOAC Int. 88, 630–638. Lehotay, S.J., de Kok, A., Hiemstra, M., Van Bodegraven, P., 2005b. Validation of a fast and easy method for the determination of residues
666
from 229 pesticides in fruits and vegetables using gas and liquid chromatography and mass spectrometric detection. J. AOAC Int. 88, 595–614. Lehotay, S.J., Mastovska, K., Lightfield, A.R., 2005c. Use of buffering and other means to improve results of problematic pesticides in a fast and easy method for residue analysis of fruits and vegetables. J. AOAC Int. 88, 615–629. Mastovska, K., Lehotay, S.J., 2006. Rapid sample preparation method for LC-MS/MS or GC-MS analysis of acrylamide in various food matrices. J. Agric. Food Chem. 54, 7001–7008. Nguyen, T.D., Lee, B.S., Lee, B.R., Lee, D.M., Lee, G.H., 2007. A multiresidue method for the determination of 109 pesticides in rice using the quick easy cheap effective rugged and safe (QuEChERS) sample preparation method and gas chromatography/mass spectrometry with temperature control and vacuum concentration. Rap. Commun. Mass Spectrom. 21, 3115–3122. O’Reilly, J., Wang, Q., Setkova, L., Hutchinson, J.P., Chen, Y., Lord, H.L., Linton, C.M., Pawliszyn, J., 2005. Automation of solid-phase microextraction. J. Sep. Sci. 28, 2010–2022. Pawliszyn, J., 1997. Solid Phase Microextraction Theory and Practice. Wiley-VCH, New York. Payá, P., Anastassiades, M., Mack, D., Sigalova, I., Tasdelen, B., Oliva, J., Barba, A., 2007. Analysis of pesticide residues using the quick easy cheap effective rugged and safe (QuEChERS) pesticide multiresidue method in combination with gas and liquid chromatography and tandem mass spectrometric detection. Anal. Bioanal. Chem. 389, 1697–1714. Plössl, F., Giera, M., Bracher, F., 2006. Multiresidue analytical method using dispersive solid-phase extraction and gas chromatography/ion trap mass spectrometry to determine pharmaceuticals in whole blood. J. Chromatogr. A 1135, 19–26. Posyniak, A., Zmudzki, J., Mitrowska, K., 2005. Dispersive solid-phase extraction for the determination of sulfonamides in chicken muscle by liquid chromatography. J. Chromatogr. A 1087, 259–264.
SECTION | I Adverse Components
Rastrelli, L., Totaro, K., De Simone, F., 2002. Determination of organophosphorus pesticide residues in Cilento (Campania, Italy) virgin olive oil by capillary gas chromatography. Food Chem. 79, 303–305. Rotunno, T.C., Argenti, R.L., 1997. Decay of fenthion in green table olives. J. Agric. Food Chem. 45, 3957–3960. Sanchez, R., Vazquez, A., Villén-Altamirano, J., Villén, J., 2006. Analysis of pesticide residues by on-line reversed-phase liquid chromatography– gas chromatography in the oil from olives grown in an experimental plot. J. Sci. Food Agric. 86, 129–134. Seiber, J.N., 1999. Extraction, cleanup, and fractionation methods. In: Fong, W.G., Moye, H.A., Seiber, J.N., Toth, J.P. (Eds.) Pesticide Residues in Foods: Methods, Technique and Regulations. John Wiley and Sons, New York, pp. 17–63. Soler, C., Picó, Y., 2007. Recent trends in liquid chromatography-tandem mass spectrometry to determine pesticides and their metabolites in food. Trends Anal. Chem. 26, 103–115. Tomlin, C.D.S., 2003. The Pesticide Manual, 14th edn. British Crop Protection Council, Surrey, UK. Tsatsakis, A.M., Tsakiris, I.N., Tzatzarakis, M.N., Agourakis, Z.B., Tutudaki, M., Alegakis, A.K., 2003. Three-year study of fenthion and dimethoate pesticides in olive oil from organic and conventional cultivation. Food Addit. Cont. 20, 553–559. Tsoutsi, C.S., Albanis, T.A., 2004. Optimization of headspace solidphase microextraction conditions for the determination of organophosphorus insecticides in olive oil. Int. J. Environ. Anal. Chem. 84, 3–13. Yagüe, C., Bayarri, S., Conchello, P., Lazaro, R., Perez-Arquillué, C., Herrera, A., Ariño, A., 2005. Determination of pesticides and PCBs in virgin olive oil by multicolumn solid-phase extraction cleanup followed by GC-NPD/ECD and confirmation by ion-trap GC-MS. J. Agric. Food Chem. 53, 5105–5109.
Chapter 71
Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS Alberto Marinas1, Fernando Lafont2, María A. Aramendía1, I.M. García2, José M. Marinas1 and Francisco J. Urbano1 1 2
Department of Organic Chemistry, University of Córdoba, Spain Servicio Central de Apoyo a la Investigación (SCAI), Unidad de Espectrometría de Masas, Universidad de Córdoba, Spain
71.1 INTRODUCTION Olive oil, which is a major ingredient of the Mediterranean diet, is said to have significant protective effects against some types of cancer and coronary heart disease. This property has been ascribed to the presence of phenols in addition to squalene or oleic acid and has prompted a demand for this product worldwide (Owen et al., 2000; Wahrburg et al., 2002). In order to control diseases and pests, some pesticides are applied to olive groves. The residues of such pesticides could persist to the harvest stage, thus contaminating olives and possibly olive oil (Guardia-Rubio et al., 2007). Therefore, there is a need for analytical methods to monitor such residues which ensure consumer protection. The European Union produces 80% of the global output of olive oil (more than 2 million tonnes each year), with Spain as the leading country (34% of the output). Spanish olive groves are located mainly in Andalusia, a strongly agriculture-dependent region where the competent council is continuously striving to preserve the high-quality features of the oil. There is currently no specific legislation in Spain as regards maximum residue levels (MRLs, see Table 71.1) for pesticides in olive oil, but only in olives. However, levels 5 times higher than those set for olives are commonly accepted for oil. Moreover, the presence of pesticide residues at low or even undetectable levels can be expected to become a major quality criterion for rejection of olive oil in the future. This has raised the need to develop new methods capable of detecting herbicides at levels below those accepted by current regulations. Until such methods are developed, some producers are offering organic olive Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
oil, which is produced from olives treated with no pesticide or man-made fertilizer. The new methods should rely on sensitive enough techniques to allow herbicides to be detected at very low levels in order to avoid fraud and/or alert farm owners to the presence of contamination in nearby holdings. Analytical methods for determining pesticides in fats and oils almost invariably require extraction of pesticide residues and fat from the matrix, followed by clean-up of the extract prior to chromatographic analysis. The most widely used technique in this context is liquid–liquid partitioning; however, other, more environmentally friendly methods such as supercritical fluid extraction with carbon dioxide (Hopper, 1999; Gonçalves et al., 2006) are gradually gaining ground for this purpose. Extracts are usually cleaned up by passage through Florisil, alumina or silica columns; matrix solid-phase dispersion; solidphase extraction or gel-permeation chromatography, GPC (Balinova, 1993, 1998; Sabik et al., 2000; Albero et al., 2004; Aramendia et al., 2007). Finally, in a similar way as done for pesticides in other matrixes, identification and determination are performed with gas [GC-ECD (GuardiaRubio et al., 2007; Fontcuberta et al., 2008), GC-NPD (Gómez de Barreda et al., 1998; Sánchez-Brunete et al., 1998), GC-FPD (Fontcuberta et al., 2008), GC-TSD (Guardia-Rubio et al., 2007), GC-MS (Hernández et al., 1998; Sánchez-Brunete et al., 1998; Frías et al., 2003)] or liquid chromatography [LC-LC-UV (Hidalgo et al., 1997; Martínez Galera et al., 1997; Hernández et al., 1998), HPLC-DAD (Martínez Galera et al., 1997; Gómez de Barreda et al., 1998; Sabik et al., 2000), HPLC-APcI-MS (Thomas, 1998; Asperger et al., 2002), HPLC-ES-MS (Lacassie et al., 1999; Huang et al., 2006)]. For a good
667
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
668
SECTION | I Adverse Components
TABLE 71.1 Key features of maximum residue levels (MRLs). 1. Plant protection products are widely used throughout the world to reduce the loss in crop production caused by harmful organisms and weeds 2. However, pesticide residues could persist to the harvest stage. In this sense, MRLs represent the maximum amount of pesticide residue (expressed as milligrams per kilogram or ppm) that might be expected in/on a food commodity when a product is used according to good agricultural practices (GAP) 3. According to Codex Alimentarius, GAP is defined as the ‘…nationally authorized safe uses of pesticides under actual conditions necessary for effective and reliable pest control’ 4. Establishing MRLs implies the following steps: a. Estimation of the residue level in or on an agricultural crop treated with the pesticide under conditions of good agricultural practice (GAP) in a supervised trial b. Estimation of the total daily intake of the specific pesticide using appropriate consumer intake models and the established residue level c. Estimation of an ‘acceptable daily intake’ (ADI) using data from toxicological tests. This involves finding the highest dose that would produce no adverse effects over a lifetime (chronic) exposure period and then applying appropriate safety factors d. Setting the maximum residue level (MRL) from (a) under the condition that: daily consumer intake (b) ⬍ acceptable daily intake (c) 5. Some time ago each country had its own MRL legislation. However, there is a tendency to harmonization. In this sense, EU MRL Regulation (396/2005/EC) will ensure free trade throughout the EU and the European Free Trade Area (EFTA) because of harmonization; will replace the variety of national MRLs with a unique MRL setting on the EU-level; and will give a chance for great progress towards more protection of consumers, especially children 6. For all pesticide/commodity combinations without an established MRL, an MRL of 0.01 mg kg⫺1 is to be applied 7. Similarly to the EU, the USA, Japan and other countries have their own legislations though there is certain harmonization through the Codex Alimentarius Commission, created by FAO and WHO 8. MRLs are based on the best data available at the time and can be changed based on more current information and higher standards 9. Finally, in order to ensure that MRLs are not exceeded, sensitive analytical methods, normally implying the use of mass spectrometry (tandem mass spectrometry when possible), are needed. Moreover, there is a current tendency to the development of multiresidue methods which allow simultaneous determination of a large number of pesticides This table lists the key facts of maximum residue limits (MRLs) including the definition, steps to establish the values, legislation and need for analytical methods sensitive enough as to ensure that such values are not exceeded (Siebers and Hänel, 2003; Smolka, 2006).
review on the methods for determination of pesticide residues in olive oil and olives see García-Reyes et al. (2007). On some occasions both liquid and gas chromatographies are coupled using the through oven transfer adsorption desorption (TOTAD) interface, the pesticides being detected by electron-capture and nitrogen-phosphorus detectors (Díaz-Plaza et al., 2007). Because of the large variety of pesticides used, the trace analysis of these compounds requires the development of multiresidue methods for the determination of the greatest number of compounds possible, with the fewest number of extraction and clean-up steps. The US Environmental Protection Agency (EPA) recommends the use of analytical methods where identification is confirmed by mass spectrometry (MS). The US Food and Drug Administration endorses the use of this technique with three ions for confirmatory purposes (Makovi and Mahon, 1999). Moreover, selected-reaction monitoring with triple quadrupole instruments (Dagnac et al., 2005; Gonçalves et al., 2006; Sauret-Szczepanski et al., 2006) or ion-trap equipment has become the preferred tool for quantitation in this context. What follows is the description of a multiresidue method for analysis of residues of ca. 220 low- and mediumpolarity pesticides in olive oil using GC-MS/MS.
71.2 DESCRIPTION OF THE MULTIRESIDUE METHOD 71.2.1 Reagents and Chemicals Certified pesticide standards were purchased from Riedelde-Haën; purities were always ⬎97%. Pesticide-quality solvents (acetonitrile, n-hexane, ethyl acetate and cyclohexane) were supplied by Supelco (Madrid, Spain). Stock standard solutions of individual compounds (concentration around 250 mg L⫺1) were prepared by exact weighing of 25 mg of the powder or liquid and dissolution in 100 mL of acetone, which were then stored in a freezer. A multicompound working standard solution (2 μg mL⫺1 concentration of each compound) was prepared by appropriate dilutions of the stock solutions with acetone and stored under refrigeration (4 °C).
71.2.2 Instrumentation A vibromatic from Selecta was employed for liquid–liquid extraction of pesticides from oil. A GPC system equipped with an autosampler (Perkin Elmer), HPLC LKB pump and a VARIAN fraction collector was employed for oil elimination from olive oil extracts. The column was an Envirogel 19 ⫻ 300 mm from Waters Inst.
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
A GC system Varian 3800 (Varian Instruments) equipped with electronic flow control was used throughout the study. Samples were injected with an 8400 Autosampler into a split/ splitless programmed-temperature injector using a 100 μL syringe. A GC fused-silica capillary column (25 m ⫻ 0.25 mm i.d. ⫻ 0.25 μm film thickness) VF-5MS Factor Four from Varian Instruments was utilized. Helium (99.9999%) at a constant flow rate of 1 mL min⫺1 was used as carrier gas. The GC was interfaced with a Varian 1200 L triple quadrupole mass spectrometer operated with EI ionization mode at 70 eV. Source was maintained at 285 °C. Argon (99.999%) was used as collision gas.
71.2.3 Sample Extraction and Clean-Up Figure 71.1 illustrates the main steps of the method for determination of pesticides in olive oil. Initially, an aliquot of 2.0 g olive oil is exactly weighed into a 20 mL glass flask, and mixed with 2 mL of n-hexane. A 25 μL solution containing different internal standards (deuterated chlorpyriphos, terbuthylazine, bromophos ethyl and endosulfan sulfate) is then added. The concentration of each internal standard in
1. Extraction from matrix
the olive oil is 0.125 mg kg⫺1. Finally, acetonitrile (10 mL) and sodium sulfate (0.2 g) are added. The flask is adapted in the vibromatic and maintained under intense agitation for 20 min. After 60 min allowed for phase separation, 9 mL of acetonitrile phase is rotavaporated (40 °C under strong vacuum) and reconstituted in 4 mL of the GPC mobile phase (cyclohexane/ethyl acetate 50/50 (v/v). Two mL (previously filtered through 0.2 μm) are injected onto the GPC system with the autosampler. The GPC mobile phase is set at 5 mL min⫺1. Pesticides elute between 12.5 and 22.5 min (50 mL fraction collected). Pesticide fraction is evaporated to dryness and reconstituted in 1 mL of cyclohexane/ethyl acetate 9/1 (v/v) prior to GC-MS analysis.
71.2.4 Chromatographic GC-MS/MS Conditions Ten μL of samples are injected into the GC. Injector programmed-temperature is as follows: 70 °C for 0.5 min, then increased at 250 °C min⫺1 until 300 °C for 10 min. The injector split ratio is initially 20:1. At 0.01 min, splitless mode is switched on until minute 3, then the split ratio
2. Clean-up
GPC column -2 g oil -2 mL n-hexane -25 μL Internal standard solution -10 mL acetonitrile -0.2 g sodium sulfate
3.GC-MS/MS Analysis
10 μL of samples are injected into the GC through an autosampler
agitation 20 min Determination of 225 pesticides within 38 min
60 min phase separation Rotavaporation of acetonitrile phase + reconstitutionin GPC mobile phase (cyclohexane/ethyl acetate 50/50 (v/v)
669
Pesticides elute between 12.5 and 22.5 min
Evaporation to dryness and reconstitutionin 1 mL of cyclohexane/ethyl acetate 9/1 (v/v)
FIGURE 71.1 Schematic representation of the analytical method developed for analysis of low- and medium-polarity pesticides in olive oil.
670
is 50:1. Column temperature is set at 70 °C, with 2-min hold, and then increased at a rate of 15 °C min⫺1 to 175 °C (0.1 min) and increased at 5 °C min⫺1 until 310 °C (5 min). The QqQ mass spectrometer is operated in MRM mode. The temperatures of the transfer line, manifold, and source of ionization are set at 285 °C, 45 °C and 285 °C, respectively. The specific MRM parameters used are shown in Table 71.2.
71.2.5 Qualitative/Quantitative Analysis Ten μL of oil extracts are injected into the GC-MS system set up for the confirmation/quantification method, monitoring two or three selected product ions for each of the compounds. These compounds are confirmed if the tolerance of 25% for the relative intensities of the selected daughter ions in the samples and in the standards has been met. In this case, analytes are quantified by internal standard method.
71.2.6 Quality Control To ensure the quality of results when the proposed methodology is applied to routine analysis, various quality criteria are established. The set of samples analyzed each day is processed together with the following: (i) the analysis of a blank extract; (ii) the analysis of a blank sample spiked immediately prior to the instrumental determination with standards of the pesticides at the pre-established concentration level of each compound to state the response to this level; and (iii) the analysis of a blank oil spiked with the standards of the pesticides at the pre-established concentration level and then extracted to check the recovery of the extraction step. The efficiency of the extraction procedure is checked by comparison with the responses of the standard. Analysis of samples within the sequence is performed if recoveries are between 60 and 120% for the spiked matrix.
71.3 METHODOLOGICAL CONSIDERATIONS 71.3.1 GC-QqQ-MS Parameters The gas chromatographic separation of the target analytes in QqQ-based methods is not very critical because this MS/MS detection mode is able to determine up to 40 compounds at the same time in the selected experimental conditions. Therefore, the developed gas chromatographic program allows elution and so determination of 225 pesticides within 38 minutes. To program the isolation of precursor ions for every compound along the chromatographic run, the overall analysis time is split into 12 segments. Within each time segment, precursor ions are isolated in the first quadrupole, and then fragmented by collision induced
SECTION | I Adverse Components
in the second quadrupole with the goal of generating spectra at different voltages, and in the third quadrupole, the product ions are separated before detection.
71.3.2 Validation of the Methodology In most of the cases, an accurate determination of pesticide residues in food commodities becomes unnecessary in routine quality-control laboratories since the only information required is whether the concentration of a specific compound in a particular matrix is under or over a regulated value, named maximum residue level (MRL). Only accurate quantification is needed when the detected quantity is expected to exceed the MRL. MRL values are both compound- and matrix-dependent, and in addition, these levels can differ from one country to another. In general, MRLs are established at trace levels, and 10 μg kg⫺1 is the lowest value established. This chapter shows the validation of the approach developed at this concentration level in olive oil matrix. Table 71.3 shows the main validation parameters obtained: mean recovery obtained after fortification of blank olive oil at 0.1 and 0.01 mg kg⫺1 (N ⫽ 10 under conditions of reproducibility and repetitivity) and relative expanded uncertainty (%). Our results evidence the efficiency of the developed method which allows detection of 225 pesticides at the 10 μg kg⫺1 level with average recoveries in the specified range and acceptable uncertainties in all the cases. In all cases linear ranges for calibration curves are between 0.01 and 0.25 mg kg⫺1, with r2 ⬎ 0.98 (samples spiked at 0.01, 0.05, 0.1 and 0.25 mg kg⫺1) and standard deviation of residuals less than 20%. The method was developed one year ago and has already been applied to 650 samples of Andalucian olive oil. The most frequently found pesticides have been terbuthylazine, diuron, oxyfluorfen, diflufenican, dimethoate, chlorpyrifos ethyl, diazinon, fenitrothion and endosulfan sulfate though in most of the cases they were present at very low concentrations. Only 0.56% of analyzed samples showed pesticide levels over the MRLs. Moreover, it is important to point out that detected levels are significantly lower than those found in previous years, especially as far as terbuthylazine, diuron and endosulfan sulfate are concerned which evidences the success of the public campaigns oriented to disseminate good practices among farmers.
SUMMARY POINTS ●
●
There is a need for multiresidue methods for simultaneous analysis of as many pesticides as possible in food. Methods should be sensitive enough as to ensure that if there are residues, all levels are below the maximum residue limits (MRLs).
671
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
TABLE 71.2 Description of optimized MRM parameters: ions are expressed as m/z and collision energy values are given in eV. The first daughter ion is the one selected for quantitation. PESTICIDE PARENT ION ION 1-ENERGY 1 ION 2-ENERGY 2 ION 3-ENERGY 3 ACEPHATE 136 42 -10 94 -15 112 -10
CARBOPHENOTHION 342 171 -5 157 -10
ACETOCHLOR 162 144 -10 120 -15 147 -10
CARBOSULFAN 160 118 -10 62 -10 160 -15
ACINBENZOLAR-S-METHYL 182 181 -10 135 -20 153 -20
CARBOXIN 235 143 -10 87 -20 235 -5
ACLONIFEN 264 194 -15 182 -25 264 -15
CHLORFENAPYR 247 227 -15 199 -30 247 -20
ACRINATHRIN 181 152 -20 127 -30
CHLORFENVINPHOS 267 159 -20 203 -10 123 -40
ALACHLOR 188 160 -10 130 -40
CHLORMEFOS 234 121 -15 154 -5 188 -5
ALDRIN 263 228 -20 193 -30
CHLOROTHALONIL 266 231 -18 168 -25 133 -30
AMITRAZ 162 132 -10 121 -15 117 -25
CHLORPROPHAM 213 127 -20 171 -10
ATRAZINE 215 200 -10 172 -15 138 -10
CHLORPYRIFOS METHYL 286 241 -25 208 -10 136 -20
AZINPHOS ETHYL 132 77 -15 132 -15
CHLORTHAL DIMETHYL 301 273 -18 223 -25 167 -40
AZINPHOS METHYL 160 132 -5 105 -15 102 -25
CIHALOFOP-BUTYL 357 256 -10 120 -25 357 -5
AZOXYSTROBIN 344 329 -10 156 -40
CIMOXANIL 111 110 -5 64 -20 82 -10
BENALAXYL 206 132 -18 117 -30 162 -10
CLODINAFOP-PROPARGYL 349 266 -20 91 -35 239 -40
BENDIOCARB 166 151 -10 126 -25 166 -15
CLOFENTEZIN 137 75 -45 102 -15 137 -5
BENFLURALIN 292 264 -10 206 -15 160 -20
CLOMAZONE 204 107 -15 106 -15 77 -30
BENFURASATE 256 163 -10 121 -30 256 -5
CLOQUINTOCET- MEXYL 192 162 -25 190 -15 127 -30
BIFENTHRIN 181 166 -10 165 -18 115 -40
CYFLUTHRIN 206 151 -15 177 -20 179 -20
BIPHENOX 341 189 -20 281 -20
CYPERMETHRIN 181 152 -20 127 -30
BIPHENYL 154 152 -25 153 -10 154 -15
CYPROCONAZOLE 222 125 -20 153 -10
BITERTANOL 170 170 -5 115 -30 141 -15
CYPRODINIL 224 118 -40 208 -18 104 -20
BROMAZIL 205 188 -18 132 -25 162 -15
o,p-DDE 318 246 -25 176 -40 318 -5
BROMOPROPILATE 341 185 -15 183 -15 157 -35
p,p´-DDE 318 246 -25 176 -40 318 -15
BROMUCONAZOLE 295 173 -15 175 -15 295 -5
o,p-DDT 235 165 -20 199 -15 235 -15
BUPIRIMATE 273 193 -10 150 -10 108 -18
p,p´-DDT 235 165 -20 199 -15 235 -15
BUPROFEZIN 249 193 -10 106 -25
DELTAMETHRIN 253 172 -10 119 -30 199 -25
BUTRALIN 266 190 -20 174 -25 266 -15
DIAFENTIURON 254 220 -15 165 -20 127 -35
CADUSAFOS 159 97 -15 131 -10 115 -15
DIAZINON 304 179 -10 137 -35 164 -35
CAPTAN 149 70 -18 77 -25 105 -5
DICHLOBENIL 171 136 -15 100 -30 171 -15
CARBARYL 144 115 -25 116 -15 144 -15
DICHLOFLUANID 224 123 -10 224 -5 (Continued)
672
SECTION | I Adverse Components
TABLE 71.2 (Continued) PESTICIDE PARENT ION ION 1-ENERGY 1 ION 2-ENERGY 2 ION 3-ENERGY 3 DICHLORAN 176 148 -10 114 -20 85 -35
ETHOXYQUIN 202 174 -10 202 -15
DICHLORMID 172 172 -15 108 -10 91 -10
ETOXAZOLE 300 285 -10 270 -15 300 -15
DICHLORVOS 185 93 -10 109 -20 63 -15
ETRIDIAZOLE 211 183 -10 108 -40 140 -25
DICOFOL 250 215 -10 139 -18 111 -35
ETRIMFOS 292 181 -10 153 -20 292 -5
o,p’-DICOFOL 251 139 -10 111 -35
FAMOXADONE 330 237 -10 196 -20 329 -10
p,p’-DICOFOL 251 139 -10 111 -35
FENAMIDONE 268 180 -15 92 -15 77 -25
DIELDRIN 277 241 -10 206 -15 170 -40
FENAMIPHOS 303 154 -18 139 -30 180 -18
DIETHOFENCARB 267 225 -10 168 -20 197 -15
FENARIMOL 330 139 -10 251 -10 111 -30
DIFENCONAZOLE 323 265 -15 323 -5
FENAZAQUIN 145 116 -15 117 -15 91 -25
DIFLUFENICAN 394 266 -10 238 -35
FENBUCONAZOLE 198 129 -10 102 -30
DIMETHENAMIDE 230 154 -10 111 -25 119.5 -25
FENITROTHION 260 125 -15 109 -10 151 -20
DIMETHOATE 125 79 -10 125 -15 93 -40
FENOXAPROP-ETHYL 288 288 -5 91 -10 119 -10
DINICONAZOLE 268 232 -10 204 -15 136 -30
FENPROPATRIN 265 210 -10 181 -20 153 -25
DINOBUTON 211 163 -10 147 -10 117 -15
FENPROPIMORPH 128 128 -15 70 -10
DINOCAP 69 39 -20 41 -10 69 -15
FENTHION 278 245 -10 125 -18 278 -15
DIPHENYLAMINE 169 168 -5 167 -30 169 -15
FENVALERATE 225 147 -15 119 -20 225 -15
DIURON 187 159 -15 124 -25
FIPRONIL 367 213 -30 178 -40 255 -15
DODEMORPH 154 97 -10 136 -10 81 -10
FLUAZIFOP-BUTYL 383 282 -10 254 -30
ENDOSULFAN ALPHA 241 206 -15 170 -15 172 -15
FLUCYTHRINATE 199 157 -10 107 -30 199 -5
ENDOSULFAN BETA 241 206 -15 170 -18 172 -18
FLUDIOXINIL 248 127 -25 154 -15 182 -10
ENDOSULFAN SULFATE 272 236 -30 117 -30 234 -20
FLUMETRALIN 143 107 -20 143 -30 117 -20
ENDRIN 281 173 -40 209 -25 245 -10
FLUMIOXAZIN 354 326 -5 176 -25 354 -15
EPOXICONAZOLE 192 138 -15 111 -25 102 -30
FLUQUINCONAZOLE 342 300 -15 315 -18 288 -20
ESFENVALERATE 225 147 -10 119 -18 91 -25
FLUROCHLORIDONE 311 174 -15 103 -20 187 -10
ETHALFLURALIN 316 276 -10 202 -30 232 -10
FLUSILAZOLE 233 165 -15 152 -10
ETHION 231 175 -10 129 -25 203 -10
FLUTALONIL 323 173 -15 281 -5 323 -5
ETHOFENPROX 163 107 -20 135 -20 77 -45
FLUTRIAFOL 164 95 -35 122 -25
ETHOFUMESATE 286 207 -10 161 -20
FOLPET 260 232 -10 130 -15 102 -35
ETHOPROPHOS 158 114 -10 97 -18 81 -15
FONOPHOS 246 109 -18 137 -10 202 -5
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
FORMOTHION 224 125 -20 155 -10 196 -5
METHOLACHLOR 238 162 -10 133 -30 238 -5
HALOXIFOP-R-METHYL 375 288 -5 316 -10 375 -5
METHOXYCHLOR 227 169 -25 141 -35 212 -15
HCH-ALPHA 219 183 -10 145 -20 109 -35
METRIBUZIN 198 82 -18 198 -5 -15
HCH-BETA 219 183 -10 145 -20 109 -35
MEVINPHOS 127 109 -10 95 -20 79 -18
HCH-DELTA 219 183 -10 145 -20 109 -35
MOLINATE 126 55 -10 126 -15
HEPTACHLOR 272 237 -18 141 -30 117 -5
MONOCROTOPHOS 127 109 -15 95 -20 79 -18
HEPTACHLOR EPOXIDE 353 282 -15 217 -20 353 -15
MYCLOBUTANIL 179 125 -15 129 -20 152 -10
HEPTENOPHOS 215 89 -15 200 -10 78 -30
NALED 185 109 -20 93 -15 185 -15
HEXACHLOROBENZENE 284 249 -20 214 -25 284 -15
NAPROPAMIDE 271 72 -10 100 -15 128 -5
HEXACONAZOLE 214 152 -25 124 -25 -172 -20
NUARIMOL 235 139 -18 123 -18 235 -15
HEXAZINONE 171 71 -15 85 -15 114 -10
OFURACE 232 186 -10 158 -18
IMAZAMETHABENZ 187 144 -10 89 -30 116 -25
OMETHOATE 156 141 -5 110 -10 79 -25
INDOXACARB 218 203 -10 134 -30 218 -5
OXADIARGYL 213 150 -10 146 -20 114 -15
IPRODIONE 314 245 -10 271 -10 162 -20
OXADIAZON 258 175 -10 147 -15 112 -25
IPROVALICARB 158 116 -5 -98 -10
OXADIXYL 233 118 -30 102 -35 143 -10
ISOCARBOPHOS 230 212 -10 136 -10 155 -25
OXYFLUORFEN 300 223 -20 167 -25 132 -40
ISOFENPHOS ETHYL 213 185 -15 121 -18
PARATHION ETHYL 291 109 -15 142 -5 263 -5
ISOFENPHOS METHYL 231 121 -15 137 -10 155 -20
PARATHION METHYL 263 109 -15 246 -5 127 -10
ISOXATHION 177 102 -20 116 -15 130 -10
PENCONAZOLE 248 157 -20 192 -15 206 -10
KRESOXIM-METHYL 206 116 -10 130 -18
PENDIMETHALIN 252 161 -15 191 -10 118 -30
LAMBDA-CYHALOTHRIN 181 152 -20 127 -30
PERMETHRIN 183 168 -10 152 -20 128 -20
LENACIL 153 136 -15 82 -15 110 -15
PHENTHOATE 274 121 -15 146 -5
LINDANE 219 145 -20 183 -10 109
o-PHENYLPHENOL 168 139 -45 114 -40 167 -10
MALATHION 173 99 -18 117 -10 127 -10
PHORATE 260 231 -5 75 -10 260 -5
MECARBAM 159 131 -10 102 -18
PHOSALONE 367 182 -10 138 -30 111 -30
MEFENPYR DIETHYL 299 253 -20 190 -30 271 -10
PHOSMET 160 133 -10 105 -18
MEPANIPYRIM 222 207 -15 118 -30 221 -10
PHOSPHAMIDON 127 109 -15 95 -25 127 -15
METALAXYL 206 132 -20 162 -10 206 -15
PIRIMICARB 166 96 -15 123 -10 83 -18
METAZACHLOR 209 132 -15 117 -30
PIRIMIPHOS-METHYL 290 151 -18 125 -35 233 -10
METHAMIDOPHOS 141 95 -10 64 -20 79 -15
PROCYMIDONE 283 96 -10 177 -30 255 -10
METHIDATHION 145 85 -10 58 -15
PROFENOFOS 339 269 -10 251 -25 188 -20
673
(Continued)
674
SECTION | I Adverse Components
TABLE 71.2 (Continued) PESTICIDE PARENT ION ION 1-ENERGY 1 ION 2-ENERGY 2 ION 3-ENERGY 3 PROMETRYN 242 184 -10 199 -10 111 -30
TAU-FLUVALINATE 250 250 -5 200 -18 208 -18
PROPACHLOR 176 57 -10 120 -10 92 -15
TEBUCONAZOLE 250 125 -15 153 -10 163 -10
PROPANIL 161 126 -20 99 -25 90 -20
TEBUFENPYRAD 318 131 -10 145 -15 117 -35
PROPARGITE 173 135 -15 105 -10 173 -20
TEFLUTHRIN 177 107 -30 87 -25 127 -15
PROPICONAZOL 259 173 -20 145 -40 191 -10
TERBACIL 161 144 -20 88 -30
PROPUXUR 152 109 -10 92 -30
TERBUMETON 226 154 -15 169 -5
PROPYZAMIDE 254 226 -10 191 -15 176 -25
TERBUTHYLAZINE 214 132 -10 119 -10 104 -15
PROSULFOCARB 128 86 -5 128 -15 251 -5
TERBUTRYN 241 185 -5 170 -15 110 -25
PROTHIOPHOS 309 239 -15 205 -30 221 -25
TETRACONAZOLE 336 156 -25 204 -40 218 -15
PYPERONYL BUTOXIDE 176 131 -10 117 -18 103 -25
TETRADIFON 229 201 -15 131 -25 166 -20
PYRAZOPHOS 265 210 -10 182 -18 138 -25
TETRAMETHRIN 164 107 -25 77 -35 135 -15
PYRIDABEN 309 147 -15 132 -25 119 -25
THIOMETON 125 79 -10 93 -40 125 -15
PYRIDAPHENTHION 340 199 -10 109 -20 203 -25
TOLCLOFOS METHYL 265 220 -20 250 -10 109 -40
PYRIFENOX 262 164 -40 192 -15 200 -15
TOLYLFLUANID 238 137 -20 136 -15
PYRIMETHANIL 198 182 -18 118 -35 156 -25
TRIADIMEFON 208 181 -10 111 -25 127 -15
PYRIPROXYFEN 226 186 -18 105 -10 157 -30
TRIADIMENOL 168 70 -10 168 -15
QUINALPHOS 298 156 -10 190 -10 129 -30
TRI-ALLATE 268 226 -10 184 -15 268 -15
QUINOMETHIONATE 234 206 -10 148 -25 116 -30
TRIAZOPHOS 257 162 -10 134 -20 119 -25
QUINOXIFEN 307 237 -20 272 -10
TRICHLORFON 145 109 -10 113 -20 145 -15
QUINTOZENE 295 237 -18 265 -10 119 -35
TRIFLOXYSTROBIN 222 162 -10 130 -10 190 -5
SIMAZINE 201 138 -10 172 -15
TRIFLUMIZOLE 278 73 -10 206 -15 278 -5
SPIROMESIFEN 272 254 -5 209 -15 272 -15
TRIFLURALIN 306 264 -10 206 -15 159 -30
SULPROFOS 322 156 -10 198 -5 280 -5
VINCLOZOLIN 212 172 -15 109 -40 145 -20
This table gives for each pesticide the parent ion as well as the so-called daughter ions which result from fragmentation of parent ion. All daughter ions are accompanied by the optimized collision energy at which they were obtained.
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
TABLE 71.3 Validation parameters for all pesticides analyzed. Pesticide
ACEPHATE
Mean recovery % (0.01 mg kg⫺1) N ⫽ 10
Mean recovery % (0.1 mg kg⫺1) N ⫽ 10
U expanded % (k ⫽ 2)
75.6
79.4
21.3
103.3
74.6
35.6
73.3
71.2
15.7
106.3
107.6
15.2
ACRINATHRIN
83.2
71.3
16.8
ALACHLOR
70.0
71.0
10.0
107.0
108.7
5.3
AMITRAZ
98.8
98.0
22.3
ATRAZINE
100.3
96.3
15.8
AZINPHOS ETHYL
84.2
76.9
17.7
AZINPHOS METHYL
96.5
102.1
10.9
AZOXYSTROBIN
97.5
109.0
12.2
107.4
107.7
11.9
BENDIOCARB
96.8
90.3
13.8
BENFLURALIN
91.1
104.3
19.8
BENFURASATE
104.4
110.7
18.9
CHLORFENAPYR
90.9
72.4
29.6
CHLORFENVINPHOS
76.4
79.5
19.7
CHLORMEFOS
98.2
93.2
18.8
CHLOROTHALONIL
81.8
84.1
16.7
CHLORPROPHAM
96.6
102.2
40.3
105.1
96.7
19.5
CHLORPYRIFOS METHYL
72.4
86.8
6.3
CHLORTHAL DIMETHYL
83.1
84.1
9.9
CIHALOFOP-BUTYL
77.6
78.9
9.8
CIMOXANIL
101.6
107.1
27.9
CLODINAFOP-PROPARGYL
117.2
118.4
19.4
CLOFENTEZIN
61.5
72.8
22.4
CLOMAZONE
97.0
98.5
28.7
ACETOCHLOR ACINBENZOLAR-S-METHYL ACLONIFEN
ALDRIN
BENALAXYL
CHLORPYRIFOS ETHYL
(Continued)
675
676
SECTION | I Adverse Components
TABLE 71.3 (Continued) Pesticide
Mean recovery % (0.01 mg kg⫺1) N ⫽ 10
Mean recovery % (0.1 mg kg⫺1) N ⫽ 10
U expanded % (k ⫽ 2)
CLOQUINTOCET- MEXYL
74.9
74.9
19.4
CYFLUTHRIN
75.3
88.9
15.6
CYPERMETHRIN
91.7
88.8
16.1
104.3
108.8
11.5
CYPRODINIL
91.8
83.7
15.8
o,p-DDE
96.8
105.4
16.8
p,p´-DDE
76.6
84.9
33.2
106.7
90.7
14.7
74.1
81.6
15.6
DIPHENYLAMINE
102.5
104.2
15.4
DIURON
103.3
99.6
37.2
73.3
89.6
23.4
106.3
97.7
25.9
ENDOSULFAN BETA
73.2
99.7
24.1
ENDOSULFAN SULFATE
70.0
89.5
25.2
107.0
96.0
20.0
EPOXICONAZOLE
98.8
85.2
18.6
ESFENVALERATE
80.3
86.2
21.5
ETHALFLURALIN
96.5
91.1
16.0
ETHION
86.5
100.2
23.7
ETHOFENPROX
107.4
106.7
17.2
ETHOFUMESATE
96.8
84.1
9.7
ETHOPROPHOS
101.1
81.4
21.6
ETHOXYQUIN
101.6
88.7
22.0
ETOXAZOLE
104.4
78.9
21.1
ETRIDIAZOLE
115.1
86.4
16.0
ETRIMFOS
82.4
78.1
11.7
FLUSILAZOLE
71.5
92.0
13.5
FLUTALONIL
97.0
76.9
19.3
FLUTRIAFOL
74.9
108.5
21.4
CYPROCONAZOLE
DINOBUTON DINOCAP
DODEMORPH ENDOSULFAN ALPHA
ENDRIN
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
FOLPET
90.9
74.6
19.3
FONOPHOS
76.4
89.9
12.7
105.1
94.3
15.4
81.8
71.1
21.1
HCH-ALPHA
101.9
105.3
22.8
HCH-BETA
107.4
83.7
36.5
HCH-DELTA
74.7
90.8
17.8
100.7
90.6
15.7
HEPTACHLOR EPOXIDE
82.7
74.2
18.8
HEPTENOPHOS
95.5
88.4
25.2
108.3
126.8
21.9
99.4
91.0
6.0
102.0
94.4
16.7
89.7
102.0
19.9
INDOXACARB
111.0
114.8
18.4
IPRODIONE
102.5
100.5
32.1
98.4
93.3
20.2
MOLINATE
101.6
107.3
18.3
MONOCROTOPHOS
115.1
101.6
20.3
MYCLOBUTANIL
104.4
102.6
11.8
NALED
82.4
89.1
13.2
NAPROPAMIDE
98.6
101.6
14.6
NUARIMOL
92.8
76.6
20.5
OFURACE
89.1
91.5
18.0
OMETHOATE
101.6
97.0
17.8
OXADIARGYL
85.9
91.7
14.6
OXADIAZON
102.6
104.3
13.6
FORMOTHION HALOXIFOP-R-METHYL
HEPTACHLOR
HEXACHLOROBENZENE HEXACONAZOLE HEXAZINONE IMAZAMETHABENZ
IPROVALICARB
OXADIXYL OXYFLUORFEN PARATHION ETHYL PARATHION METHYL PENCONAZOLE
97.3
91.8
17.2
101.6
107.2
13.2
78.0
98.2
15.6
101.6
105.1
32.9
91.7
92.4
20.1 (Continued)
677
678
SECTION | I Adverse Components
TABLE 71.3 (Continued) Mean recovery % (0.1 mg kg⫺1) N ⫽ 10
U expanded % (k ⫽ 2)
Pesticide
Mean recovery % (0.01 mg kg⫺1) N ⫽ 10
PENDIMETHALIN
104.3
96.6
23.2
PERMETHRIN
91.8
83.1
17.5
PHENTHOATE
101.6
77.1
15.9
o-PHENYLPHENOL
102.4
101.6
13.5
PHORATE
98.2
96.7
15.3
PYRIFENOX
74.9
74.7
39.3
PYRIMETHANIL
90.9
100.7
35.3
PYRIPROXYFEN
105.1
108.3
12.6
QUINALPHOS
99.4
98.0
34.6
102.0
106.3
17.7
QUINOXIFEN
89.7
86.9
17.2
QUINTOZENE
111.0
102.1
10.7
SIMAZINE
102.5
109.0
14.2
SPIROMESIFEN
84.2
86.2
25.8
SULPROFOS
98.4
107.7
16.3
TAU-FLUVALINATE
98.5
90.3
13.9
TEBUCONAZOLE
76.6
94.3
31.1
TEBUFENPYRAD
85.4
110.7
17.9
TEFLUTHRIN
102.3
111.1
16.1
TERBACIL
106.7
108.9
19.3
TERBUMETON
74.1
99.6
25.6
TERBUTHYLAZINE
93.3
100.8
26.5
TERBUTRYN
83.3
85.8
18.2
106.3
97.1
19.6
TETRADIFON
93.2
103.5
25.6
TETRAMETHRIN
60.0
95.4
33.2
BIFENTHRIN
82.4
99.6
11.5
BIPHENOX
115.1
100.9
11.1
BIPHENYL
101.6
111.1
16.2
98.6
100.8
12.0
QUINOMETHIONATE
TETRACONAZOLE
BITERTANOL
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
BROMAZIL
92.8
85.8
24.4
BROMOPROPILATE
96.8
97.1
17.4
BROMUCONAZOLE
111.8
113.5
13.9
BUPIRIMATE
108.1
95.4
21.2
BUPROFEZIN
102.4
81.1
15.1
BUTRALIN
102.6
94.6
13.4
87.3
88.5
27.0
101.6
101.7
18.2
CARBARYL
89.1
95.1
20.2
CARBOPHENOTHION
76.6
87.1
17.3
CARBOSULFAN
101.6
80.2
17.0
CARBOXIN
105.4
95.7
16.8
o,p-DDT
111.8
108.3
30.7
p,p´-DDT
105.4
90.9
16.7
DELTAMETHRIN
101.9
100.0
13.0
DIAFENTIURON
107.4
80.4
9.4
74.7
81.5
20.2
100.7
94.6
20.0
DICHLOFLUANID
82.7
95.6
17.6
DICHLORAN
95.5
73.2
24.0
DICHLORMID
108.3
91.6
19.5
DICHLORVOS
99.4
88.4
35.5
DICOFOL
102.0
96.2
17.1
o,p’-DICOFOL
118.1
102.3
25.3
p,p’-DICOFOL
76.3
76.4
10.2
DIELDRIN
89.7
110.1
18.5
DIETHOFENCARB
111.0
94.8
23.5
DIFENCONAZOLE
98.4
92.7
11.9
DIFLUFENICAN
98.5
96.3
16.3
DIMETHENAMIDE
76.6
86.5
25.5
DIMETHOATE
85.4
83.9
23.8
102.3
98.9
9.0
CADUSAFOS CAPTAN
DIAZINON DICHLOBENIL
DINICONAZOLE
(Continued)
679
680
SECTION | I Adverse Components
TABLE 71.3 (Continued) Pesticide
Mean recovery % (0.01 mg kg⫺1) N ⫽ 10
Mean recovery % (0.1 mg kg⫺1) N ⫽ 10
U expanded % (k ⫽ 2)
FAMOXADONE
98.6
108.3
20.1
FENAMIDONE
92.8
98.5
30.3
FENAMIPHOS
96.8
83.0
13.1
FENARIMOL
111.8
99.3
17.2
FENAZAQUIN
118.1
116.9
20.3
FENBUCONAZOLE
102.4
91.5
17.5
85.9
78.3
13.3
102.6
92.9
25.4
97.3
94.3
18.0
101.6
92.6
16.7
FENTHION
89.1
83.9
18.0
FENVALERATE
76.6
81.1
14.1
105.4
98.3
19.6
FLUAZIFOP-BUTYL
69.9
78.9
25.6
FLUCYTHRINATE
89.5
98.5
15.9
FLUDIOXINIL
101.6
96.9
19.1
FLUMETRALIN
91.7
100.7
21.2
FLUMIOXAZIN
104.3
107.3
18.2
FLUQUINCONAZOLE
91.8
82.5
18.7
FLUROCHLORIDONE
117.2
95.7
12.8
ISOCARBOPHOS
98.5
91.9
19.3
ISOFENPHOS ETHYL
76.6
82.6
16.8
ISOFENPHOS METHYL
85.4
84.6
17.2
ISOXATHION
102.3
95.4
15.7
KRESOXIM-METHYL
106.7
89.9
11.6
74.1
85.9
47.1
LENACIL
103.3
100.3
21.7
LINDANE
73.3
89.9
16.8
MALATHION
106.3
99.7
23.3
MECARBAM
73.2
104.4
16.8
FENITROTHION FENOXAPROP-ETHYL FENPROPATRIN FENPROPIMORPH
FIPRONIL
LAMBDA-CYHALOTHRIN
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
MEFENPYR DIETHYL
70.0
101.6
35.4
107.0
175.1
23.2
METALAXYL
98.8
82.4
22.0
METAZACHLOR
84.2
92.8
22.5
110.3
98.6
32.7
METHIDATHION
96.5
96.8
29.0
METHOLACHLOR
89.6
111.8
15.2
METHOXYCHLOR
107.4
108.1
36.4
METRIBUZIN
96.8
102.4
14.8
MEVINPHOS
71.1
85.9
32.7
PHOSALONE
72.4
88.4
26.3
PHOSMET
83.1
79.8
19.4
PHOSPHAMIDON
96.6
96.9
22.5
PIRIMICARB
76.4
82.7
11.1
PIRIMIPHOS-METHYL
98.2
95.5
11.5
PROCYMIDONE
72.4
99.4
22.6
PROFENOFOS
96.6
102.0
33.8
PROMETRYN
83.1
89.7
21.5
PROPACHLOR
81.8
111.0
22.8
PROPANIL
101.9
102.5
17.7
PROPARGITE
107.4
98.4
20.4
PROPICONAZOL
74.7
71.2
18.9
PROPUXUR
82.7
71.3
19.7
100.7
107.6
15.7
PROSULFOCARB
95.5
91.0
24.9
PROTHIOPHOS
108.3
108.7
18.8
85.9
86.8
20.9
117.2
91.8
8.6
PYRIDABEN
61.5
101.9
30.4
PYRIDAPHENTHION
97.0
107.4
7.3
107.0
81.1
28.9
98.8
86.8
29.6
MEPANIPYRIM
METHAMIDOPHOS
PROPYZAMIDE
PYPERONYL BUTOXIDE PYRAZOPHOS
THIOMETON TOLCLOFOS METHYL
(Continued)
681
682
SECTION | I Adverse Components
TABLE 71.3 (Continued) Pesticide
Mean recovery % (0.01 mg kg⫺1) N ⫽ 10
Mean recovery % (0.1 mg kg⫺1) N ⫽ 10
U expanded % (k ⫽ 2)
TOLYLFLUANID
80.3
94.6
22.1
TRIADIMEFON
84.2
98.5
15.8
TRIADIMENOL
96.5
99.6
9.8
TRI-ALLATE
85.6
89.4
25.6
107.4
100.3
29.6
96.8
99.6
33.2
TRIFLOXYSTROBIN
101.1
89.9
24.8
TRIFLUMIZOLE
104.4
100.3
28.9
TRIFLURALIN
101.6
98.5
28.9
VINCLOZOLIN
115.1
112.3
33.2
TRIAZOPHOS TRICHLORFON
This table shows the main validation parameters obtained: mean recovery obtained after fortification of blank olive oil at 0.1 and 0.01 mg kg⫺1 (N ⫽ 10 under conditions of reproducibility and repetitivity) and relative expanded uncertainty (%).
●
●
●
Whatever the method, it is recommended to resort to mass spectrometry (MS/MS when possible) for confirmation. A rapid multiresidue method for determination of up to 225 pesticides in olive oil which has already been successfully applied for one year in Andalucia has been described. The analytical strategy developed can be summarized as follows: (a) simple semiautomated extraction of pesticides from oils; (b) direct qualitative and quantitative analysis by GC-QqQ-MS. The approach has been validated obtaining good results for every parameter. Also, it was applied to routine sample analysis obtaining very good results.
ACKNOWLEDGMENTS The authors gratefully acknowledge funding from the Consejería de Agricultura, Pesca y Alimentación de la Junta de Andalucía (Project CAO00-005), the Consejería de Innovación, Ciencia y Empresa de la Junta de Andalucía (Projects FQM-191, FQM-3931 and P07-FQM-2695) and Spanish Ministerio de Educación y Ciencia (Projects CTQ2007-65754 and CTQ2008-01330, co-financed with FEDER funds).
REFERENCES Albero, B., Sánchez-Brunete, C., Donoso, A., Tadeo, J.L., 2004. Determination of herbicide residues in juice by matrix solid-phase
dispersion and gas chromatography-mass spectrometry. J. Chromatogr. A 1043, 127–133. Aramendia, M.A., Borau, V., Lafont, F., Marinas, A., Marinas, J.M., Moreno, J.M., Urbano, F.J., 2007. Determination of herbicide residues in olive oil by gas chromatography-tandem mass spectrometry. Food Chem. 105, 855–861. Asperger, A., Efer, J., Koal, T., Engewald, W., 2002. Trace determination of priority pesticides in water by means of high-speed on-line solidphase extraction–liquid chromatography–tandem mass spectrometry using turbulent-flow chromatography columns for enrichment and a short monolithic column for fast liquid chromatographic separation. J. Chromatogr. A 960, 109–119. Balinova, A., 1993. Solid-phase extraction followed by high-performance liquid chromatographic analysis for monitoring herbicides in drinking water. J. Chromatogr. A 643, 203–207. Balinova, A., 1998. Multiresidue determination of pesticides in plants by high-performance liquid chromatography following gel permeation chromatographic clean-up. J. Chromatogr. A 823, 11–16. Dagnac, T., Bristeau, S., Jeannot, R., Mouvet, C., Baran, N., 2005. Determination of chloroacetanilides, triazines and phenylureas and some of their metabolites in soils by pressurized liquid extraction, GC-MS/MS, LC-MS and LC-MS/MS. J. Chromatogr. A 1067, 225–233. Díaz-Plaza, E.M., Cortés, J.M., Vázquez, A., Villén, J., 2007. Automated determination of pesticide residues in olive oil by on-line reversedphase liquid chromatography-gas chromatography using the through oven transfer adsorption desorption interface with electron-capture and nitrogen-phosphorus detectors operating simultaneously. J. Chromatogr. A 1174, 145–150. Fontcuberta, M., Arqués, J.F., Villalbí, J.R., Martínez, M., Centrich, F., Serrahima, E., Pineda, L., Duran, J., Casas, C., 2008. Chlorinated
CHAPTER | 71 Multiresidue Analysis of Low- and Medium-polarity Pesticides in Olive Oil by GC-MS/MS
organic pesticides in marketed food: Barcelona, 2001-06. Sci. Total Environ. 389, 52–57. Frías, S., Rodríguez, M.A., Conde, J.E., Pérez-Trujillo, J.P., 2003. Optimisation of a solid-phase microextraction procedure for the determination of triazines in water with gas chromatography–mass spectrometry detection. J. Chromatogr. A 1007, 127–135. García-Reyes, J.F., Ferrer, C., Gómez-Ramos, M.J., Molina-Díaz, A., Fernández-Alba, A.R., 2007. Determination of pesticide residues in olive oil and olives. Trends Anal. Chem. 26, 239–251. Gómez de Barreda Jr., D., Gamón Vila, M., Lorenzo Rueda, E., Saez Olmo, A., Gómez de Barreda, D., García de la Cuadra, J., Ten, A., Peris, C., 1998. Dissipation of some citrus selective residual herbicides in an irrigation well. J. Chromatogr. A 795, 125–131. Gonçalves, C., Carvalho, J.J., Azenha, M.A., Alpendurada, M.F., 2006. Optimization of supercritical fluid extraction of pesticide residues in soil by means of central composite design and analysis by gas chromatography–tandem mass spectrometry. J. Chromatogr. A 1110, 6–14. Guardia-Rubio, M., Marchal-López, R.M., Ayora-Cañada, M.J., RuizMedina, A., 2007. Determination of pesticides in olives by gas chromatography using different detection systems. J. Chromatogr. A 1145, 195–203. Hernández, F., Hidalgo, C., Sancho, J.V., López, F.J., 1998. Coupledcolumn liquid chromatography applied to the trace-level determination of triazine herbicides and some of their metabolites in water samples. Anal. Chem. 70, 3322–3328. Hidalgo, C., Sancho, J.V., Hernández, F., 1997. Trace determination of triazine herbicides by means of coupled-column liquid chromatography and large volume injection. Anal. Chim. Acta 338, 223–229. Hopper, M.L., 1999. Automated one-step supercritical fluid extraction and clean-up system for the analysis of pesticide residues in fatty matrices. J. Chromatogr. A 840, 93–105. Huang, S.-B., Mayer, T.J., Yokley, R.A., Perez, R., 2006. Direct aqueous injection liquid chromatography/electrospray ionization–mass spectrometry/mass spectrometry analysis of water for atrazine, simazine, and their chlorotriazine metabolites. J. Agr. Food Chem. 54 (3), 713–719. Lacassie, E., Dreyfuss, M.F., Daguet, J.L., Vignaud, M., Marquet, P., Lachâtre, G., 1999. Liquid chromatography–electrospray mass spectrometry multi-residue determination of pesticides in apples and pears. J. Chromatogr. A 830, 135–143.
683
Makovi, C.M., Mahon, B.M., Rev October 1999. Multiresidue methods, US Department of Health and Human Services, Public Health Service Food and Drug Administration, (3rd edn.), Vol. 1. Pesticide Analytical Manual. Martínez Galera, M., Martínez Vidal, J.L., Garrido Frenich, A., Gil García, M.D., 1997. Evaluation of multiwavelength chromatograms for the quantification of mixtures of pesticides by high-performance liquid chromatography–diode array detection with multivariate calibration. J. Chromatogr. A 778, 139–149. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Würtele, G., Spiegelhalder, B., Bartsch, H., 2000. Olive-oil consumption and health: the possible role of antioxidants. Lancet Oncol. 1, 107–112. Sabik, H., Jeannot, R., Rondeau, B., 2000. Multiresidue methods using solid-phase extraction techniques for monitoring pesticides, including triazines and degradation products, in ground and surface waters. J. Chromatogr. A 885, 217–236. Sánchez-Brunete, C., Pérez, R.A., Miguel, E., Tadeo, J.L., 1998. Multiresidue herbicide analysis in soil samples by means of extraction in small columns and gas chromatography with nitrogen–phosphorus and mass spectrometric detection. J. Chromatogr. A 823, 17–24. Sauret-Szczepanski, N., Mirabel, P., Wortham, H., 2006. Development of an SPME-GC-MS/MS method for the determination of pesticides in rainwater: laboratory and field experiments. Environ. Pollut. 139 (1), 133–142. Siebers, J., Hänel, R., 2003. Assessment of residue analytical methods for crops, food, feed, and environmental samples: the approach of the European Union. In: Lee, P.W., Aizawa, H., Barefoot, A.C., Murphy, J.J. (Eds.) Handbook of Residue Analytical Methods for Agrochemicals, Vol. 1. John Wiley & Sons Ltd, Chichester, West Sussex, England, pp. 13–37. Smolka, S., 2006, seminar entitled “Pesticide Residues in Food-regulation, monitoring, policy” available http://www.pan-germany.org/download/ proceedings/01_Smolka_EU_MRL_in_food.pdf Thomas, K.V., 1998. Determination of selected antifouling booster biocides by high-performance liquid chromatography–atmospheric pressure chemical ionisation mass spectrometry. J. Chromatogr. A 825, 29–35. Wahrburg, U., Kratz, M., Cullen, P., 2002. Mediterranean diet, olive oil and health. Eur. J. Lipid Sci. Technol. 104, 698–705.
This page intentionally left blank
Section 2
Nutritional, Pharmacological and Metabolic Properties of Olives and Olive Oil
This page intentionally left blank
2.1
General Nutrition General Aspects and Changes in Food Processing
This page intentionally left blank
Chapter 72
Trends in Olive Oil Production, Supply and Consumption in Mediterranean Countries from 1961 to the Present Day Genevieve Buckland and Carlos A. González Unit of Nutrition, Environment and Cancer, Cancer Epidemiology Program, Catalan Institute of Oncology (ICO), Barcelona, Spain
72.1 INTRODUCTION The Mediterranean diet was originally defined as the dietary pattern found in some of the olive-growing regions in the Mediterranean Basin at the beginning of the 1960s, which in itself emphasizes the key role of olive oil within the traditional Mediterranean diet. It is accepted as a healthy dietary pattern as well as an expression of lifestyle and a cultural model (Trichopoulou and Lagiou, 1997). Although there has been considerable debate within the scientific community on the accepted definition, the significant features have been defined by experts in the field (SerraMajem et al., 2004). The common pattern is a diet rich in plant foods and the principal source of fat is from olive oil, resulting in a high monounsaturated to saturated fat ratio (Gonzalez et al., 2002). Olive oil is placed as an integral feature of the Mediterranean diet, and it is probably the most globally traded and consumed product that is connected to the traditional Mediterranean diet (Anania et al., 2007). Despite this there are enormous geographical differences in olive oil supply among Mediterranean countries, with 30-fold differences found during the 1960s (Garcia-Closas et al., 2006). Thus describing olive oil supply at a regional level can provide useful information on the regional variations in dietary patterns compared to describing the Mediterranean diet as a uniform entity. In addition to important geographical differences, several studies have shown that considerable temporal changes have taken place in the dietary patterns of Mediterranean countries over the last 3–4 decades (Serra-Majem et al., 1993; Moreno et al., 2002; Garcia-Closas et al., 2006; Balanza et al., 2007; Zeghichi-Hamri and Kallithraka, 2007), with important changes in some, but not all, of the components. In terms of Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
olive oil, there are differences in olive oil supply and consumption across time, but these changes are not uniform across Mediterranean countries (Garcia-Closas et al., 2006). This chapter gives an overview of the temporal trends in olive oil production, supply and consumption in 15 Mediterranean countries from the 1960s up until the present day. The countries have been defined as Mediterranean using the broad definition that they are in geographical contact with the Mediterranean Sea. They include five European Mediterranean countries – Spain, France, Italy, Greece, Cyprus, and ten African or Asiatic Mediterranean countries – Morocco, Algeria, Tunisia, Libya, Egypt, Lebanon, Syria, Israel, Jordan and Turkey. As non-Mediterranean countries account for less than 3.5% of world olive oil production, we have focused on production in Mediterranean countries only, using production data provided by the Food and Agriculture Organization (FAO) of the United Nations (Food and Agriculture Organization, 2008). Dietary estimates of olive oil supply and consumption are presented at three different levels: (i) at a population level, from the FAO’s food balance sheets (FBS) that utilize ecological information to estimate olive oil supply; (ii) at a household level, from household food budget surveys (HBS) carried out in nationally representative samples of households that provide an estimate of olive oil supply, compiled and standardized by the Data Food Networking (DAFNE) project (DAFNE Group, 2008); and (iii) at an individual level, from individual dietary surveys (IDS) using regional or national nutritional surveys and from the European Prospective Investigation into Cancer and Nutrition (EPIC) study, that provides an estimate of olive oil consumption in free-living individuals. Finally, a comparison has been made between the data on olive oil supply
689
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
690
SECTION | II General Aspects and Changes in Food Processing
Syria 6%
Other France 2% 0.2% Algeria Turkey Morocco 1% 3% 4% Cyprus 0.1% Spain 36%
Tunisia 6%
Greece 15% Italy 27%
year averages were calculated for each study period in order to show overall production tendencies and minimize yearly fluctuations, because in some countries the quantity of olive oil produced is extremely unstable from year to year due to crop and weather conditions. Olive oil production in Libya, Lebanon, Israel and Jordan is relatively low and so they have been grouped into a single category ‘Other’.
3000 2500 2000 1500 1000
72.2.1 Current Olive Oil Production in Mediterranean Countries
500
Tu rk ey or oc co Al ge ria Fr an c C e yp ru s O th er M
a
ria Sy
e
si
Tu ni
ly
ec
re
G
ai
Ita
Sp
To t
n
0
al
Production quantity, 2004–06 average, per 1000 tonnes
FIGURE 72.1 Production of virgin olive oil in Mediterranean countries, 2004–2006 percentage average of tonnes produced. Other Mediterranean countries: Libya, Lebanon, Israel and Jordan. Source: Food and Agriculture Organization.
Mediterranean country
FIGURE 72.2 Virgin olive oil production quantity in Mediterranean countries in 2004–2006 (3-year average). Other Mediterranean countries: Libya, Lebanon, Israel and Jordan. Source: Food and Agriculture Organization.
and consumption in relation to the sources of information and their respective limitations.
72.2 OLIVE OIL PRODUCTION IN MEDITERRANEAN COUNTRIES Data on olive oil production was obtained from the FAO within the agriculture production domain. Production data are given in tonnes per year and represent total production from olive crops harvested in the indicated year, expressed as oil equivalent, and assume the entire crop was processed in the same year as harvested. The total production of olives harvested is not entirely processed into olive oil, as important quantities may be used for food. Therefore an extraction rate is calculated, although this is the same for all countries despite inter-country variability. Trends in olive oil production are presented from 1961– 2006, the period with available information. Three consecutive
Spain is currently the leading producer of olive oil, and has been consistently since the early 1990s. In 2004– 2006 Spain accounted for 36% of all olive oil produced in Mediterranean countries (Figure 72.1) and produced a 3-year average of 927 000 tonnes (Figure 72.2). The second and third largest olive oil producers within Mediterranean countries in 2004–2006 are Italy and Greece, accounting for 27% (3-year average of 697 000 tonnes) and 15% (3-year average of 382 000 tonnes) of olive oil, respectively. Together, Spain, Italy and Greece account for over three-quarters of all olive oil produced in Mediterranean countries. The share of olive oil produced by the African or Asiatic Mediterranean countries is therefore relatively small and the primary producers are Tunisia (6%), Syria (6%), Turkey (4%) and Morocco (3%).
72.2.2 Trends in Olive Oil Production in Mediterranean Countries Over the last decade there has been an important expansion in global olive oil production, which has increased by 35% from 1990–1991 to 2004–2005 (Figure 72.2), with a 3-year average production of 2.7 million tonnes in 2004–2006 (Figure 72.2). The greatest increases in production have occurred in the largest producing countries; Spain, Italy and Greece. Until the end of the 1980s Italy was the largest producer of
CHAPTER | 72 Trends in Olive Oil Production, Supply and Consumption in Mediterranean Countries from 1961
Spain (SP) Syria (SY) Algeria (AL)
1400
Italy (IT) Turkey (TR) Cyprus (CY)
Greece (GR) Morocco (MO) France (FR)
691
Tunisia (TN) Others (OTH)
Production quantity, per 1000 tonnes per year (3-year averages)
1200
1000 SP 800 IT 600
400
GR
200
TN SY TR MO OTH FR AL CY
6 00
3 00 20
04
–2
0 20
01
–2
7
00 –2
98
19
19
95
–1
99
4
1 19
92
–1
99
99
8 98 19
89
–1
5 98 86 19
–1
19
83
98
–1
2
9 19
80
–1
–1
97
6 19
77
97
3 19
74
–1
0
97
97 19
71
–1
7 96
–1 68
19
–1
–1 19
61 19
65
96
4
0
Year
FIGURE 72.3 Virgin olive oil production in Mediterranean countries from 1961–2006 (3-year averages). Other Mediterranean countries: Libya, Lebanon, Israel and Jordan. Source: Food and Agriculture Organization.
olive oil, until it was overtaken by Spain, whose production has soared during the last decade, although there has also been some production decline since the turn of the century. There has been a steady increase in olive oil production in Greece, which produces just under half the quantity of olive oil of Spain. Comparing 10-yearly production averages of 1967–1976 with 1997–2006, production has increased by 156% in Spain, by 91% in Greece and by 32% in Italy (data not shown), demonstrating that changes in production are not uniform. The temporal changes from 1961–2006 have been minimal in most of the African or Asiatic Mediterranean countries, although there have been slight upward trends in Tunisia, Syria and Morocco, whose 3-yearly averages of production in 2003–2006 were 165 000 tonnes, 148 000 tonnes and 78 000 tonnes, respectively.
72.3 OLIVE OIL SUPPLY IN MEDITERRANEAN COUNTRIES 72.3.1 Olive Oil Supply at a Population Level Data on nationwide olive oil supply (the amount available for human consumption) were obtained from the FAO-assembled FBS, which show agriculture supply and
utilization data. FBS are a valid tool to make geographical and temporal comparisons (Sasaki and Kesteloot, 1992a; Serra-Majem, 2001) and are the only source of standardized information on olive oil supply that is available yearly, over the study period (1961–2003) for all the Mediterranean countries. Data on olive oil supply from the FAO’s FBS were only available up until 2003, thus the study period could not be extended up until the present day. Total olive oil supply in each country is estimated by adding the total quantity of olive oil produced by a country to the total quantity imported, and then subtracting the total quantity exported, fed to livestock, used for seed in agriculture, put to manufacture for non-food use, and losses during transportation and storage. This figure is then adjusted to any stock changes during that year. Per capita olive oil supply is estimated by dividing the respective annual quantity, from the above calculations, by the country’s population size in the corresponding year. Olive oil supply data are presented by the FAO in kg of olive oil per person per year. However, for ease of understanding and comparison with other information, olive oil supply from FAO data is expressed as grams per person per day. Throughout the study period (1961–2003) three consecutive year averages have been calculated in order to minimize inter-annual variability in the data. Olive oil supply per capita is relatively low in Lebanon, Israel and Egypt and therefore they have been grouped into a single category ‘Other’.
692
SECTION | II General Aspects and Changes in Food Processing
France 5% Syria 7%
Algeria Jordan 1% Tunisia 1% Libya 2% Other Cyprus 1% Morocco Turkey 1% 3% 0.1% 3% Italy 41%
Greece 9% Spain 26%
FIGURE 72.4 Olive oil availability by country in Mediterranean countries, 2001–2003 percentage average of tonnes available. Other Mediterranean countries: Lebanon, Israel and Egypt. Source: Food and Agriculture Organization.
Availability of olive oil, 2001–03 average (g/person/day)
50 45 40 35 30 25 20 15 10 5
e
ai n Sy ria Tu ni si a Li by a Jo rd a C n yp r M us or oc c Fr o an c Al e ge ria O th e Tu r rk e Av y er ag e
Sp
Ita
ec re G
ly
0
Mediterranean country
FIGURE 72.5 Olive oil availability per capita in Mediterranean countries in 2001–2003 (3-year average). Other Mediterranean countries: Lebanon, Israel and Egypt. Source: Food and Agriculture Organization.
72.3.2 Current Olive Oil Supply by Country in Mediterranean Countries Although Italy is the second largest producer of olive oil, it had the highest olive supply in 2001–2003 (using 3-year average), accounting for 41% of all available olive oil in these Mediterranean countries (Figure 72.4). Spain, Greece and Syria then follow, accounting for 26%, 9% and 7% respectively. France accounts for only 5%. The remaining Mediterranean countries make up a relatively small proportion of total olive oil supply.
72.3.3 Current Olive Oil Supply per Capita in Mediterranean Countries A country’s per capita supply is a result of the total amount of olive oil available in each Mediterranean country
and its population size. The highest daily per capita olive oil supply in 2001–2003 (using 3-year averages) was in Greece (44.7 g) followed by Italy (36.0 g), Spain (32.5 g), Syria (22.0 g) and Tunisia (17.5 g) (Figure 72.5), all of which were above the average daily per capita olive oil supply of 15.2 g. There were large variations in supply between the Mediterranean countries, for instance Greece’s per capita supply was 6–7 times greater than that of Cyprus and Morocco. In general, per capita olive oil supply was much higher in European Mediterranean countries than African and Asiatic Mediterranean countries, with the exception of France which has a very low olive oil supply. Olive oil supply was also very low in many of the Asiatic Mediterranean countries; Algeria, Turkey and those grouped under Other (Jordan, Lebanon and Israel) which had an average per capita supply of less than 3 g day⫺1.
72.3.4 Trends in Olive Oil Supply in Mediterranean Countries During the 1960s almost 30-fold differences in olive oil supply were found between Mediterranean countries (Figure 72.6). Greece had the highest per capita olive oil supply, which was around double of that found in Spain and Italy. France and many of the African and Asiatic Mediterranean countries had a very low olive oil supply, of less than 5 g person⫺1 day⫺1. Over the last four decades there have been small but steady increases in olive oil supply in Spain and Italy and to a lesser extent in Greece; from 1961–1964 to 2001–2003 per capita daily olive oil supply increased in Spain from 23.6 g to 32.5 g, in Italy from 26.6 g to 36.0 g and in Greece from 40.6 g to 44.7 g. Olive oil supply was highest in Greece at the end of the 1970s, peaking at 58.6 g person⫺1 day⫺1, and has since begun to gradually decline. Differences between countries have remained relatively constant over time, apart from
CHAPTER | 72 Trends in Olive Oil Production, Supply and Consumption in Mediterranean Countries from 1961
Spain (SP) Morocco (MO) Cyprus (CY) Other (OTH)
70
France (FR) Algeria (AL) Syria (SY)
Italy (IT) Tunisia (TN) Libya (LI)
693
Greece (GR) Turkey (TR) Jordan (JO)
Olive oil avaliability, g person−1 day−1
60
50 GR
40 IT SP
30 SY
20 TN LI JO CY MO FR OTH AL TR
10
3 00
0 20
01
–2
00
7 19
98
–2
99
4 19
95
–1
99
1 19
92
–1
99
8 19
89
–1
98
5 19
86
–1
98
2 19
83
–1
98
9 19
80
–1
97
6 19
77
–1
97
3 19
74
–1
97
0 19
71
–1
97
7 –1
96 68 19
–1 65 19
19
61
–1
96
4
0
Year
FIGURE 72.6 Per capita daily olive oil availability in Mediterranean countries from 1961–2003 (3-year averages). Other Mediterranean countries: Lebanon, Israel and Egypt. Source: Food and Agriculture Organization.
in Tunisia and Libya, which have experienced distinctive temporal changes. Olive oil supply has also increased steadily in Syria which now has the fourth highest per capita supply. Syria, Tunisia, Libya and Jordan now have an intermediate olive oil supply of above 10 g person⫺1 day⫺1. However, the rest of the African and Asiatic Mediterranean countries (Morocco, Algeria, Turkey, Cyprus, Lebanon, Israel and Egypt) have a low per capita olive oil supply, which has remained constantly low throughout this time period.
72.3.5 Olive Oil Supply at a Household Level Data on olive oil supply was also obtained from DAFNE’s DafneSoft online databank which compiles, standardizes and harmonizes dietary data from nationally representative European household budget surveys (HBS) (Trichopoulou and Naska, 2003). Although the databank contains food supply in 24 European countries, only five of these are Mediterranean countries (Italy, Spain, France, Greece and Cyprus). In addition, data are only available for a limited number of years (and often in different years between countries) therefore making comparisons between countries and conclusions regarding temporal trends is slightly restricted. However, supply of olive oil and total added lipid in milliliters per person per day are presented for these five Mediterranean countries
(Figure 72.7) for the years available (ranging from 1980–2004).
72.3.6 Trends in Olive Oil Supply at a Household Level Household data on supply of added lipids (excluding olive oil) and of olive oil in Italy, Spain, Greece and Cyprus reveal a number of differences between countries and over time (Figure 72.7). Olive oil supply in the 1980s was highest in Greece, Spain and Italy (in descending order) with very low amounts available in France and Cyprus (in 1996–1997). In contrast, added lipids excluding olive oil were highest in France and Cyprus, where olive oil supply throughout the studied time period represented a very small proportion of total added lipids. In Italy, Spain and Greece, however, almost all added lipids supply was from olive oil. In terms of total added lipids, supply was highest in Greece, then Italy, Spain, Cyprus and France (descending order). Differences between countries have remained fairly constant across time and therefore a similar pattern was still observed in the late 1990s or early 2000s (depending on latest data available). Despite this there have been temporal changes within countries, with a general tendency for decreases over time in total added lipids and oils, mainly due to a decrease in supply of olive oil as supply of other added lipids excluding olive oil has remained reasonably stable across time.
694
SECTION | II General Aspects and Changes in Food Processing
Added lipids excluding olive oil Olive oil
90
Olive oil availability, ml person−1 day−1
80 70 60 50 40 30 20 10 0 1980– 1990– 1998- 1990 1981 1991 1999 Spain
1993
1996
Italy
1981
1987
1998
Greece
2004
1985
1991 1996– 2003 1997
France
Cyprus
Country, year
FIGURE 72.7 Olive oil availability in five European countries from 1980–2004. Data from household budget surveys compiled in the Data Food Networking (DAFNE) project of the European Union.
72.4 OLIVE OIL CONSUMPTION IN MEDITERRANEAN COUNTRIES
72.4.1 Olive Oil Consumption at an Individual Level
Data on olive oil consumption at an individual level was obtained from individual dietary surveys and from the EPIC study whose methodological details have been published previously (Riboli and Kaaks, 1997). EPIC is a multicenter European prospective cohort study which recruited approximately half a million participants, of both sexes and mostly from 35–69 years, between 1992–1998 from ten European countries, four of which are Mediterranean (Greece, Spain, Italy and France). Although the majority of the cohort was recruited from the general population, some exceptions include France (health insurance members and only female), Italy–Florence (breast cancer screening programs) and Spain and Italy (included blood donors). Data obtained from a detailed 24-hour diet recall undertaken between 1995–2000 on a random sub-sample (7%) of the cohort, has been used to compare sex-specific intakes of total added fats and oils and specifically olive oil between the four EPIC Mediterranean countries and between different regions within these countries. Consumption of added fats and oils across all ten EPIC countries has been described previously (Linseisen et al., 2002). Olive oil intake data from other individual dietary surveys were obtained from surveys carried out within the study period in Italy (Turrini et al., 2001), Spain (RibasBarba et al., 2007) and Greece (Ferro-Luzzi et al., 2002), the three Mediterranean countries with the highest per capita olive oil consumption. Finally, these data have been compared to olive oil supply from FBS and HBS.
Olive oil data from EPIC (estimating intake between 1995– 2000) shows that the average daily intake of added fats and oils (excluding olive oil) and of olive oil is fairly heterogeneous across and within the European Mediterranean countries included in the study; Greece, Spain, Italy and France (Figures 72.8 and 72.9). It also clearly shows that olive oil is the primary source of added oils in all these countries, except for France. In men and women, olive oil consumption was highest in Greece, with a daily intake of 40.6 g and 29.4 g, respectively (men consuming around a third more than women). Spain had the second highest olive oil consumption, although there were large regional differences; the highest consumption in women in Navarra in Northern Spain (26.9 g day⫺1) was more than double of that consumed by women in Asturias in Southern Spain (12.6 g day⫺1). With the exception of Navarra, the consumption was higher in Southern Spain (Murcia, Granada) than in the Northern Spain (San Sebastian, Asturias). Olive oil consumption in Italy was on average around half the amount of that found in Greece; in Italy the average for women was 15.2 g day⫺1 and for men was 20.7 g day⫺1. As with Spain, there were also reasonably large regional differences in consumption in Italy, with the highest consumption found in Florence (even higher than in Ragusa, Sicily) and lowest in Varese. The lowest consumption of olive oil out of the EPIC Mediterranean countries was in France (data only available for women), ranging from 1.7 g day⫺1 in the North-West to 3.5 g day⫺1 in the South. In contrast France had the highest
CHAPTER | 72 Trends in Olive Oil Production, Supply and Consumption in Mediterranean Countries from 1961
Added fats and oils excluding olive oil Olive oil
40
30
20
Spain
Greece
Italy Country, center
North-east
North-west
South
South coast
Varese
Turin
Florence
Naples
Ragusa
Asturias
San Sebastian
Navarra
Murcia
0
Granada
10
Total
Quantity of added fat and oil intake (adjusted means in g day−1)
50
695
France
FIGURE 72.8 Daily intake of total added fats and oils and olive oil in women from 15 centers across the four Mediterranean countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) calibrated study (24-hour recall).
Added fats and oils exluding olive oil Olive oil
40
30
20
Greece
Spain
Varese
Turin
Florence
Ragusa
Asturias
San Sebastian
Navarra
Murcia
0
Granada
10
Total
Quantity of added fat and oil intake (adjusted mean, g day−1)
50
Italy Country, center
FIGURE 72.9 Daily intake of total added fats and oils and olive oil in men from 10 centers across the three Mediterranean countries participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) calibrated study (24-hour recall).
consumption of other types of added fats and oil (excluding olive oil), reaching 23.8 g day⫺1 in the North-East of France, more than double of the amount consumed in Greece, Spain and Italy. In terms of total added fats and oil, including olive oil, consumption was highest in Spain in Murcia (36.7 g day⫺1 in women) and Navarra (36.2 g day⫺1 in women) and then Greece (35.7 g day⫺1 in women). Data from country-specific nutritional surveys in Greece, Spain and Italy in the mid 1990s have reported similar olive oil intakes to those in EPIC (which estimated a mean daily
intake of 35 g in Greece, 25.5 g in Spain and 18.0 g in Italy). For instance, a cross-sectional nutrition survey carried out in Italy in 1994–1996 estimated a mean olive oil intake of 19.8 g person⫺1 day⫺1 and reported that 96.6% of the surveyed population consumed olive oil (Turrini et al., 2001), comparable to EPIC results. A cross-sectional nutritional survey carried out in Catalonia, Spain in 1992–1993 estimated that the average olive oil consumption in males was 33 g person⫺1 day⫺1 and in females 27 g person⫺1 day⫺1 (Ribas-Barba et al., 2007), which are comparable to the
696
SECTION | II General Aspects and Changes in Food Processing
results from EPIC in the region of Navarra. A comprehensive overview of the dietary fat content in Greece, showed that in IDS of adults the amount of energy derived from fat in the 1990s ranged from 35–44% (Ferro-Luzzi et al., 2002), the principal source of fat being olive oil. Although these figures cannot be directly compared to those of EPIC, they confirm the high olive oil intake found in Greece.
72.5 TECHNICAL CONSIDERATIONS Although data on olive oil supply at a population and household level are useful for identifying inter-country patterns and intra-country temporal trends, there are inherent limitations in both sources of data, which should be taken into account when interpreting the results. Household food data do not take into account olive oil purchased and consumed outside the home, which can contribute substantially to food supply. In Mediterranean countries it is estimated that between 16–24% of total expenditure on food is spent away from home (Byrd-Bredbenner et al., 2000), which will have a substantial impact on the estimate of olive oil supply using household data. In addition, HBS may also be inaccurate due to loss during storage, consumption of food by visitors or under-recording of intakes. A limitation that applies to olive oil data from HBS and FBS is that they estimate average household and nationwide consumption, respectively, and therefore it is not feasible to extrapolate to individual olive oil intake. Only dietary data from IDS and epidemiological studies can be used to estimate dietary consumption at an individual level, and even these surveys have their inherent limitations depending on the methods used (dietary history questionnaire, food frequency questionnaire and 24-hour dietary recall or food records). The FAO’s FBS are limited by the inaccuracy of the underlying data sources (the basic statistics of population, supply and utilization of foods) which is heterogeneous between countries and within countries over time. In addition, FBS do not provide information on probable differences in olive oil consumption between different population groups (by age group, gender, socio-economical levels, geographical area, etc.). A key limitation is that FBS provide data on olive oil available for human consumption in terms of the quantity that reaches the consumer, but not the amount of food actually consumed. Although FBS data take into account the waste during manufacture, processing and distribution, it does not take into consideration the losses of olive oil at the household level that happen during storage, in preparation and cooking, as plate waste and when olive oil is thrown away. Additionally, oil waste when cooking in food establishments can create important amounts of waste (Ash et al., 2008). Consequently, the amount of food actually consumed tends to be lower than the amount available
(in developing countries), and the reliability of data on fats and oils from FBS for making inferences on temporal changes has been questioned by some authors (SerraMajem, 2001). However, a study comparing dietary fat intake from the FAO with 52 individual dietary surveys carried out in 19 countries reported significant correlations between the two types of data sources for monounsaturated fatty acids (Sasaki and Kesteloot, 1992b). Therefore despite these limitations, the FAO’s FBS are recognized as a valid source of food data for comparison between countries and years and are also an appropriate tool for evaluating food policies (Serra-Majem, 2001). When comparing different dietary data sources, FBS data tend to overestimate food consumption compared to HBS and IDS. HBS have also been found to overestimate food intake compared to IDS in a comparative analysis of nutritional data from national, household and individual levels across countries (Serra-Majem et al., 2003). In line with this, the olive oil estimates from EPIC and other IDS mentioned here were lower than those from the FBS data and the HBS data in a similar time period. The results on olive oil supply between the FBS and HBS were fairly similar apart from in Greece, where olive oil supply was overestimated by the HBS. Interestingly, the FBS and HBS showed different temporal trends with respect to Spain and Italy, as the FBS indicate the olive oil supply has increased gradually since the 1980s, whereas the HBS indicate a decrease over this time period, with a fall of more than 10 ml person⫺1 day⫺1 in Spain from 1981–1990. In Greece the FBS and HBS both show overall decreases in olive oil supply since the 1980s.
72.6 CONCLUSION In summary, there has been an overall increase in production of olive oil in Mediterranean countries since the 1960s and Spain is now the largest producer and showed the greatest increases over time. In contrast, per capita daily olive oil supply is highest in Greece and Italy. Although there is evidence that there have been considerable deviations from the traditional Mediterranean dietary pattern in relation to some food groups (Moreno et al., 2002; Tessier and Gerber, 2005; Garcia-Closas et al., 2006; Balanza et al., 2007; Ribas-Barba et al., 2007; ZeghichiHamri and Kallithraka, 2007), the evidence provided here by FBS shows that the typical high intake of olive oil has been retained or is increasing in many of the Mediterranean countries. These findings are substantiated by results from previous studies, also analyzing temporal trends in food supply from FBS data (Garcia-Closas et al., 2006; Balanza et al., 2007), which reported that in Mediterranean Europe there was greater energy supply from olive oil over time (Balanza et al., 2007). The evidence also shows that many of the Mediterranean countries (especially African and
CHAPTER | 72 Trends in Olive Oil Production, Supply and Consumption in Mediterranean Countries from 1961
Asiatic Mediterranean countries) have consistently had very low olive oil supply, demonstrating the heterogeneity of the dietary pattern between Mediterranean countries. Olive oil data from different sources and at different levels – at a population level (from FBS), a household level (from HBS) and an individual level (from IDS) – show relatively similar overall patterns of olive oil consumption between countries. An exception is the opposing trend in Italy and Spain regarding the data from FBS and from HBS. However, an epidemiological study in Sardinia, Italy reported increases in olive oil intake over time (between two generations) (Tessier and Gerber, 2005), which is consistent with the FBS data. The highest supply and intake of olive oil is found in Greece, which is in line with other studies confirming that, during the 1990s, it was the country with a food pattern closest to the defined ‘Mediterranean diet’, followed by Italy and Spain (Garcia-Closas et al., 2006; Balanza et al., 2007). A diet rich in olive oil has many health benefits, especially in relation to cardiovascular disease and therefore the fact that this distinguishing dietary trait is still found in many Mediterranean countries is encouraging considering the general westernization of dietary habits in Mediterranean regions.
SUMMARY POINTS ●
●
●
●
●
●
Olive oil production in Mediterranean countries has increased by 35% from 1990–1991 to 2004–2005. Spain is currently the leading producer of olive oil (36%), followed by Italy (27%) and Greece (15%). The highest daily per capita olive supply in 2001–2003 was observed in Greece (44.7 g), followed by Italy (36.0 g), Spain (32.5 g) Syria (22.0 g) and Tunisia (17.5 g). Olive oil supply is very low in France and other African and Asiatic Mediterranean countries. Over the last four decades there have been small but steady increases in olive supply in Spain and Italy and to a lesser extent in Greece. Data from individual dietary surveys and individual consumption from epidemiological studies in Mediterranean countries show relatively similar overall patterns to data on food supply obtained from food balance sheets and food budget surveys.
REFERENCES Anania, G., D’Andrea, P., and Rosaria, M., 2007. The Global Market for Olive Oil: Actors, Trends, Policies, Prospects and Research Needs. TRADEAG – Agricultural Trade Agreements – Working Papers. Available at http://purl.umn.edu/6109 (accessed May 2008). Ash, M., Baker, A., Blayney, D., Childs, N., Dohlman, E., Haley, S. Food availability: documentation [online] Washington, DC: USDA
697
Economic Research Service. Available at http://www.ers.usda.gov/ Data/FoodConsumption/FoodAvailDoc.htm#fats (accessed July 2008). Balanza, R., Garcia-Lorda, P., Perez-Rodrigo, C., Aranceta, J., Bonet, M.B., Salas-Salvado, J., 2007. Trends in food availability determined by the Food and Agriculture Organization’s food balance sheets in Mediterranean Europe in comparison with other European areas. Public Health Nutr. 10, 168–176. Byrd-Bredbenner, C., Lagiou, P., Trichopoulou, A., 2000. A comparison of household food availability in 11 countries. J. Hum. Nutr. Diet. 13, 197–204. DAFNE Group. The Data Food Networking (DAFNE) project. Available at http://www.nut.uoa.gr (accessed July 2008). Ferro-Luzzi, A., James, W.P., Kafatos, A., 2002. The high-fat Greek diet: a recipe for all? Eur. J. Clin. Nutr. 56, 796–809. Food and Agriculture Organization. Food Balance Sheets. Available at http://www.fao.org (accessed May 2008). Garcia-Closas, R., Berenguer, A., Gonzalez, C.A., 2006. Changes in food supply in Mediterranean countries from 1961 to 2001. Public Health Nutr. 9, 53–60. Gonzalez, C.A., Argilaga, S., Agudo, A., Amiano, P., Barricarte, A., Beguiristain, J.M., Chirlaque, M.D., Dorronsoro, M., Martinez, C., Navarro, C., Quiros, J.R., Rodriguez, M., Tormo, M.J., 2002. [Sociodemographic differences in adherence to the Mediterranean dietary pattern in Spanish populations]. Gac. Sanit. 16, 214–221. Linseisen, J., Bergstrom, E., Gafa, L., Gonzalez, C.A., Thiebaut, A., Trichopoulou, A., Tumino, R., Navarro, S.C., Martinez, G.C., Mattisson, I., Nilsson, S., Welch, A., Spencer, E.A., Overvad, K., Tjonneland, A., Clavel-Chapelon, F., Kesse, E., Miller, A.B., Schulz, M., Botsi, K., Naska, A., Sieri, S., Sacerdote, C., Ocke, M.C., Peeters, P.H., Skeie, G., Engeset, D., Charrondiere, U.R., Slimani, N., 2002. Consumption of added fats and oils in the European Prospective Investigation into Cancer and Nutrition (EPIC) centres across 10 European countries as assessed by 24-hour dietary recalls. Public Health Nutr. 5, 1227–1242. Moreno, L.A., Sarria, A., Popkin, B.M., 2002. The nutrition transition in Spain: a European Mediterranean country. Eur. J. Clin. Nutr. 56, 992–1003. Ribas-Barba, L., Serra-Majem, L., Salvador, G., Castell, C., Cabezas, C., Salleras, L., Plasencia, A., 2007. Trends in dietary habits and food consumption in Catalonia, Spain (1992–2003). Public Health Nutr. 10, 1340–1353. Riboli, E., Kaaks, R., 1997. The EPIC project: rationale and study design. Int. J. Epidemiol. 26, S6–S14. Sasaki, S., Kesteloot, H., 1992. Value of Food and Agriculture Organization data on food-balance sheets as a data source for dietary fat intake in epidemiologic studies. Am. J. Clin. Nutr. 56, 716–723. Serra-Majem, L., 2001. Food availability and consumption at national, household and individual levels: implications for food-based dietary guidelines development. Public Health Nutr. 4, 673–676. Serra-Majem, L., MacLean, D., Ribas, L., Brule, D., Sekula, W., Prattala, R., Garcia-Closas, R., Yngve, A., Lalonde, M., Petrasovits, A., 2003. Comparative analysis of nutrition data from national, household, and individual levels: results from a WHO-CINDI collaborative project in Canada, Finland, Poland, and Spain. J. Epidemiol. Commun. Health 57, 74–80. Serra-Majem, L., Ribas, L., Lloveras, G., Salleras, L., 1993. Changing patterns of fat consumption in Spain. Eur. J. Clin. Nutr. 47, S13–S20. Serra-Majem, L., Trichopoulou, A., de la Cruz, J.N., Cervera, P., Garcia Alvarez, A., la Vecchia, C., Lemtouni, A., Trichopoulos, D. and on behalf of the International Task Force on the Mediterranean Diet,
698
SECTION | II General Aspects and Changes in Food Processing
2004. Foreword. Does the definition of the Mediterranean diet need to be updated? Public Health Nutr. 7, 927–929. Tessier, S., Gerber, M., 2005. Factors determining the nutrition transition in two Mediterranean islands: Sardinia and Malta. Public Health Nutr. 8, 1286–1292. Trichopoulou, A., Lagiou, P., 1997. Healthy traditional Mediterranean diet: an expression of culture, history, and lifestyle. Nutr. Rev. 55, 383–389.
Trichopoulou, A., Naska, A., 2003. European food availability databank based on household budget surveys: the Data Food Networking initiative. Eur. J. Public Health 13 (3 Suppl), 24–28. Turrini, A., Saba, A., Perrone, D., Cialfa, E., D’Amicis, A., 2001. Food consumption patterns in Italy: the INN-CA Study 1994–1996. Eur. J. Clin. Nutr. 55, 571–588. Zeghichi-Hamri, S., Kallithraka, S., 2007. Mediterranean diet in the Maghreb: an update. World. Rev. Nutr. Diet. 97, 139–161.
Chapter 73
The Bioavailability of Olive Oil Phenolic Compounds María-Isabel Covas, Montserrat Fitó, Olha Khymenets and Rafael de la Torre Cardiovascular Risk and Nutrition and Human Pharmacology and Clinical Neurosciences Research Groups, Institut Municipal d´Investigació Mèdica (IMIM-Hospital del Mar), Barcelona, Spain. CIBER de Fisiopatología de la Obesidad y Nutrición (CIBEROBN)
73.1 INTRODUCTION The beneficial effects of olive oil on cardiovascular risk factors are now recognized and often only attributed to its high levels of monounsaturated fatty acids (MUFA). On November 2004, the Federal Drug Administration of the USA permitted a claim on olive oil labels concerning: ‘the benefits on the risk of coronary heart disease of eating about 2 tablespoons (23 grams) of olive oil daily, due to the MUFA in olive oil’ (US FDA Press Release P04100. November 1, 2004. http://www.fda.gov/bbs/topics/ news/2004/NEW01129.htlm; accessed on May 2, 2008). Olive oil is, however, more than a MUFA fat. Olive oil is a functional food which besides having a high level of MUFA contains other minor components with biological properties (Covas et al., 2006c). The content of the minor components of an olive oil varies, depending on several conditions such as the cultivar, climate, ripeness of the olives at harvesting, and the processing system employed to produce the olive oil. Three types of olive oil are currently present on the market: virgin, ordinary, or pomace (Gimeno, 2002). Virgin olive oils are produced by direct pressing or centrifugation of the olives, among them those with an acidity greater than or equal to 3.3 degrees (2 degrees in the European Union) are submitted to a refining process in which some components, mainly phenolic compounds, and to a lesser degree squalene, are lost (Owen et al., 2000a). By mixing virgin and refined olive oil, an ordinary olive oil (olive oil, UE 1991) is produced and marketed. After virgin olive oil production, the rest of the olive drupe and seed is processed and submitted to a refining process, resulting in pomace olive oil, to which a certain quantity of virgin olive oil is added before marketing. The minor components of virgin olive oil are classified into two types: the unsaponificable fraction, defined as the fraction extracted with solvents after the saponification of the oil, and the soluble fraction which includes the phenolic compounds (Covas et al., 2006c). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
The major phenolic compounds in olive oil are: (1) simple phenols (e.g., hydroxytyrosol, tyrosol, vanillic acid); (2) secoiridoids: oleuropein glucoside, and SIDs which are the dialdehydic form of oleuropein (SID-1) and ligstroside (SID-2) lacking a carboxymethyl group, and the aglycone form of oleuropein glucoside (SID-3) and ligstroside (SID4); and (3) polyphenols: lignans (e.g., (⫹)-pinoresinol and (⫹)-1-acetoxypinoresinol) and flavonols (Owen, 2000; Covas et al., 2006c) (Figure 73.1). Tyrosol, hydroxytyrosol, and their secoiridoid derivatives make up around 90% of the total phenolic content of a virgin olive oil.
73.2 BIOAVAILABILITY OF OLIVE OIL PHENOLIC COMPOUNDS 73.2.1 Absorption and Disposition After olive oil ingestion, tyrosol (Tyr) and hydroxytyrosol (OH-Tyr), as well as their glucosides and aglycones, such as oleuropein, undergo rapid hydrolysis under gastric conditions, resulting in significant increases in the amount of Tyr and OH-Tyr free forms which enter the small intestine (Corona et al., 2006). In in vitro models, both OH-Tyr and Tyr are able to cross human Caco-2 cell monolayers and rat segments of jejunum and ileum (Manna et al., 2000; Corona et al., 2006). Data from experiments performed in Caco-2 cell monolayers using 14C–OH-Tyr, showed the transport of the phenolic compound occurs via a bidirectional passive diffusion mechanism (Manna et al., 2000). In animal models, when Tyr and OH-Tyr were orally administered in oil or water solutions, the orally administered oil dosing promoted a recovery of the phenolics in 24-h urine greater (25%) than that obtained with the oral aqueous dose (Tuck et al., 2001). When Tyr and OH-Tyr were administered intravenously in saline solution no significant differences were observed
699
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
700
SECTION | II General Aspects and Changes in Food Processing
A
D
C
OH
HO
O
HO
O OH
OH
O
HO
COOCH3
E
O
COOCH3
HO
O O
COOCH3
OH O
O
O O
B
HO
O
O
O
OH
HO
HO
CH2OH OH
F
G
O
HO
O
COOCH3
HO
O O
COOCH3
OH
O
O OH
OH
FIGURE 73.1 Structures of tyrosol, hydroxytyrosol and derivatives: (A): hydroxytyrosol; (B): tyrosol; (C): oleuropein glucoside; (D): SID-1; (E): SID-2; (F): SID-3; (G): SID-4.
in the amount of phenolic compounds recovered in urine between the intravenous and the oral oil doses for either Tyr or OH-Tyr. Similar results were obtained in humans when the bioavailability of OH-Tyr was compared by administering this compound in different matrices (olive oil, spiked refined oil, or yoghurt) (Visioli et al., 2003). Urinary OHTyr recovery was higher after virgin olive oil administration (44.2% of the OH-Tyr administered) than after addition of OH-Tyr to a refined olive oil (23% of the OH-Tyr administered), or to a yoghurt (5.8% of dose or approximately, 13% of that recorded after virgin olive oil intake) (Visioli et al., 2003). Concerning oleuropein, data from animal models show it can be absorbed, albeit poorly, from isolated perfused rat intestine. Thus, the most plausible way for oleuropein to exert its biological activities seems to be through its conversion to OH-Tyr (Edgecombe et al., 2000). This idea is supported by the results of bioavailability studies in rats, in which peak plasma concentrations reached after ingestion of high doses of oleuropein (100 mg kg⫺1) were in the nanogram range, whereas those of OH-Tyr were highly increased (Del Boccio et al., 2003; Bazoti et al., 2005). These observations have been further confirmed in humans (Vissers et al., 2002; Visioli et al., 2003). In the process of crossing epithelial cells of the gastrointestinal tract, phenolic compounds from olive oil are subject to a biotransformation phase and, therefore, subjected to an important first-pass metabolism. According to data of in vitro studies, about 10% of OH-Tyr is converted in
homovanyl alcohol by the catechol-O-methyltransferase (Manna et al., 2000). In addition to the O-methylated derivative of OH-Tyr, the glucuronides of OH-Tyr and Tyr have also been described (Corona et al., 2006). In contrast, there was no absorption of oleuropein as it was rapidly degraded by the colonic microflora resulting in OH-Tyr formation (Corona et al., 2006). The hepatic metabolism of the olive oil phenols has been studied in human hepatoma HepG2 cells. After incubation, culture media and cell lysates were hydrolyzed with β-glucuronidase and sulfatase. Methylated and glucuronidated forms of OH-Tyr were detected after 18 h of incubation, together with methyl-glucuronidated metabolites. Hydroxytyrosyl acetate was largely converted into free OH-Tyr and subsequently metabolized, although small amounts of glucuronidated hydroxytyrosyl acetate were detected. Tyrosol was poorly metabolized, with ⬍10% of the phenol glucuronidated after 18 h. Minor amounts of free or conjugated phenols were detected in cell lysates. No sulfated metabolites were found (Mateos et al., 2005). The pharmacokinetics of OH-Tyr intravenously administered to rats indicates a fast and extensive uptake of the molecule by the organs and tissues, with a preferential renal uptake (D’Angelo et al., 2001). Hydroxytyrosol was recovered mainly in the sulfo-conjugated forms. The recovery of OH-Tyr in urine was about 6% of the dose administered: 0.3% recovered as 3-methyl-4-hydroxy-phenylethanol (HVAL or MOPET), 12.3% as 3,4-dihydroxy-phenylacetic acid (DOPAC), 23.6%
701
CHAPTER | 73 The Bioavailability of Olive Oil Phenolic Compounds
as homovanillic acid (3-methyl-4-hydroxy-phenylacetic acid, HVA), and 26% as 3,4-dihydroxy-phenylacetaldehyde (DOPAL) (D’Angelo et al., 2001). It has been suggested that non-absorbable phenolic compounds can exert local antioxidant activities in the gastrointestinal tract (Ursini et al., 1998), an idea supported by the capacity of phenolic compounds isolated from olive oil to scavenge the free radicals generated by the fecal matrix (Owen et al., 2000b) and those induced in intestinal epithelium cells (Manna et al., 1996). However, one of the prerequisites to assess their in vivo physiological significance is to determine their absorbability and presence in human plasma. The first report on the bioavailability and disposition of olive oil phenolic compounds in humans was provided by Visioli et al. (2000). In this experiment, Tyr and OH-Tyr were spiked to a refined olive oil (very low phenolic content) and administered to healthy volunteers. Phenolic compounds were dose-dependently absorbed in humans, most phenolic compounds being recovered in biological fluids as conjugates in a dose-dependent manner with the phenolic content of the olive oil administered (Visioli et al., 2000). Also, in human studies it was demonstrated that tyrosol, hydroxytyrosol and oleuropein were absorbed at the small intestine level. Oleuropein was not quantified in plasma nor in urine but it was shown that it was metabolized in the body and recovered in urine mainly in the form of hydroxytyrosol (Vissers, 2000). Phenolic compounds, particularly those bearing a catechol group, are typically biotransformed by three enzymatic systems: catechol-O-methyl-transferase, sulfatases, and glucuronosyltransferases. Depending on the dose and the availability of co-factors the proportion of methyl, sulfate, and glucuronide conjugates varies among subjects. Further studies on the olive oil phenolic compound bioavailability (Miró-Casas et al., 2001, 2003a, b) were performed with virgin olive oil in its
natural form. After administering 25 mL of virgin oil (with an estimated content 1.2 mg of OH-Tyr), OH-Tyr plasma concentrations peaked at 30 minutes and those of its methylated metabolite, HVAL at 50 min. Plasma peak concentrations were around 25 ng mL⫺1 for OH-Tyr and 4 ng mL⫺1 for HVAL. The estimated half-life for OH-Tyr was 3 hours, reaching baseline concentrations after 8 hours of the virgin olive oil ingestion (Figure 73.2). More than 98% of both OH-Tyr and HVAL were in their conjugated forms, mainly glucuronates, confirming previous findings. In urine, OH-Tyr and HVAL concentrations peaked in the collection period 0–2 h (MiróCasas et al., 2003a). Despite the short half-life of Tyr and OH-Tyr, sustained consumption promotes an increase of olive oil phenolic compounds in biological fluids (Figure 73.2). Plasma and urinary levels of OH-Tyr and Tyr increase in a dose-dependent manner with the phenolic content of the olive oil administered (Miró-Casas et al., 2003b; Marrugat et al., 2004; Weinbrenner et al., 2004; Covas et al., 2006a, b) (Figure 73.3). Table 73.1 shows the urinary recoveries of OH-Tyr, Tyr, and HVAL after olive oil of medium (164 mg kg⫺1) and high (466 mg kg⫺1) phenolic content. The fact that a dose-dependent increase of Tyr and OH-Tyr with the phenolic content of the olive oil administered exists, at real-life olive oil doses, confirms the usefulness of these compounds as biomarkers of compliance in clinical trials (Covas et al., 2006b). With regard to the dose–effect relationship, 24-h urinary Tyr seems to be a better biomarker of sustained and moderate doses of virgin olive oil consumption than OH-Tyr (Miró-Casas et al., 2003b). Both OH-Tyr and Tyr urinary concentrations have been used, and are currently in use, in nutritional intervention studies as biomarkers of virgin olive oil ingestion Covas et al., 2006a, b; Fitó et al., 2007, 2008). Most bioavailability studies on olive oil phenols have measured total Tyr and OH-Tyr concentrations in blood or
Conc(ng/mL−1)
9 8
LPC
7
MPC HPC
6 5 4 3 2 1 0 0 Without supplementation
24
48 Hours
72
96 After supplementation
FIGURE 73.2 Plasma hydroxytyrosol concentrations after ingestion of olive oil with low (LPC, 10 mg kg⫺1), medium (133 mg kg⫺1), and high (486 mg kg⫺1) phenolic content. A single 25 mL dose was ingested before and after 4 days of olive oil supplementation (25 mL day⫺1). Adapted from Weinbrenner, 2004.
702
SECTION | II General Aspects and Changes in Food Processing
in some physiopathological states, such as inflammation or cancer, with a concomitant in situ deconjugation of phenol metabolites. Data are still very scarce and further studies are needed to elucidate the role and bioactivity of the human biological metabolites of olive oil phenolic compounds.
Percentage (%) of change
250
200
Tyrosol Hydroxytyrosol
150
100
73.2.2 Binding of Olive Oil Phenolic Compounds and their Metabolites to Human Lipoproteins
50
0 Low Medium High Olive oil phenolic content FIGURE 73.3 Changes in urinary tyrosol and hydroxytyrosol after 3 weeks of sustained olive oil (range of phenolic compounds from 0 to 150 mg kg⫺1) ingestion. Adapted from Marrugat et al., 2004.
TABLE 73.1 Urinary recoveries (μmol) of olive oil phenolic compounds. Olive oil
Hydroxytyrosol
Tyrosol
HVAL
MPC
1.21 ⫾ 0.07
1.33 ⫾ 0.11
0.38 ⫾ 0.04
HPC
3.12 ⫾ 0.21
2.69 ⫾ 0.19
0.78 ⫾ 0.06
Values are expressed as Mean ⫾ SD. HPC, high phenolic content olive oil (466 mg kg⫺1); MPC, medium phenolic content olive oil (164 mg kg⫺1); HVAL,3-methyl-4-hydroxy-phenylethanol, the methylated biological metabolite of hydroxytyrosol.
urine after acidic or enzymatic treatment of the samples. There is a lack of studies in which glucuronide and sulfate conjugates of Tyr and OH-Tyr in biological samples have been measured. It could be hypothesized that some conjugates of phenolic compounds may behave as carriers of the free forms of the phenolic compounds to target tissues, the ‘depot hypothesis’. Within this hypothesis, the determination of the bioavailability of phenol metabolites in plasma or tissues may be more relevant than that in urine. The proportion of free aglycones in some tissues can differ from that observed in blood (D’Angelo et al., 2001); this may be explained by a specific uptake of the aglycone or intracellular deconjugation. This last hypothesis implies that anionic conjugates could be transported across plasma membranes via carrier systems, as has been shown for acyl-glucuronides (Sallustio et al., 2000). Furthermore, β-glucuronidase is located in the lumen of the endoplasmic reticulum in various organs, which could be reached by phenol glucuronides. β-Glucuronidase is also present in the lysosomes of several cells, from which the enzyme can be released under some particular conditions such as an oxidative stress situation. β-Glucuronidase activity can increase
Phenolic compounds which can bind low-density lipoproteins (LDL) are likely to perform their peroxyl scavenging activity in the arterial intima, where full LDL oxidation occurs in microdomains sequestered from the richness of antioxidants present in plasma (Witzum, 1994). Olive oil phenolic compounds can bind lipoproteins in vivo in humans. Tyrosol and hydroxytyrosol were recovered in all human lipoprotein fractions after virgin olive oil ingestion, except in very low-density lipoproteins, with concentrations peaking between 1 and 2 hours after olive oil ingestion (Bonanome et al., 2000). Tyrosol and hydroxytyrosol, as well as their biological metabolites hydroxytyrosol monoglucuronide, hydroxytyrosol monosulfate, tyrosol glucuronide, tyrosol sulfate, and homovanillic acid sulfate, have been identified in LDL after virgin olive oil ingestion (De la Torre-Carbot et al., 2006, 2007). In addition, the concentration of total phenolic compounds in LDL has been shown to be directly correlated with the phenolic concentration of the olive oils ingested and with the resistance of LDL to their in vitro oxidation (Gimeno et al., 2007). After ingestion of 40 mL of virgin olive oil, the phenolic content of the LDL directly correlated with the plasma concentrations of Tyr and OH-Tyr (Covas et al., 2006a). The susceptibility of the LDL to oxidation depends not only on their fatty acid content, but also on the LDL antioxidant content (i.e., vitamin E and polyphenols) bound to the LDL (Fuller and Jialal, 1994). Further studies are required to establish the nature of the bond between the LDL and the phenolic compounds, including olive oil phenolic compounds and their metabolites, due to the physiopathological implications involved.
REFERENCES Bazoti, F.N., Gikas, E., Puel, C., Coxam, V., Tsarbopoulos, A., 2005. Development of a sensitive and specific solid phase extraction-gas chromatography-tandem mass spectrometry method for the determination of elenolic acid, hydroxytyrosol, and tyrosol in rat urine. J. Agric. Food Chem. 53, 6213–6221. Bonanome, A., Pagnan, A., Caruso, D., Toia, A., Xamin, A., Fedeli, E., Berra, B., Zamburlini, A., Ursini, F., Galli, G., 2000. Evidence of postprandial absorption of olive oil in humans. Nutr. Metab. Cardiovas. Dis. 10, 111–120. Corona, G., Tzounis, X., Assunta-Dessa, M., Deiana, M., Debnam, E.S., Visoli, F., Spencer, J.P., 2006. The fate of olive oil polyphenols in the
CHAPTER | 73 The Bioavailability of Olive Oil Phenolic Compounds
gastrointestinal tract: implications of gastric and colonic microfloradependent biotransformation. Free Radic. Res. 40, 647–658. Covas, M.I., de la Torre, K., Farre-Albaladejo, M., Kaikkonen, J., Fitó, M., López-Sabater, C., Pujadas-Bastardes, M.A., Joglar, J., Weinbrenner, T., Lamuela-Raventós, R.M., de la Torre, R., 2006a. Postprandial LDL phenolic content and LDL oxidation are modulated by olive oil phenolic compounds in humans. Free Rad. Biol. Med. 40, 608–616. Covas, M.I., Nyyssonen, K., Poulsen, H.E., Kaikkonen, J., Zunft, H.J., Kiesewetter, H., Gaddi, A., de la Torre, R., Mursu, J., Baümler, H., Nascetti, S., Salonen, J.T., Fitó, M., Virtanen, J., Marrugat, J. EUROLIVE Study Group, , 2006b. The effect of polyphenols in olive oil on heart disease risk factors: a randomized trial. Ann. Intern. Med. 145, 333–341. Covas, M.I., Ruiz-Gutiérrez, V., de la Torre, R., Kafatos, A., LamuelaRaventós, R.M., Osada, J., Owen, R.W., Visioli, F., 2006c. Minor components of olive oil: evidence to date of health benefits in humans. Nutr. Rev. 64 (Suppl. 1), 20–30. D’Angelo, S., Manna, C., Migliardi, V., Mazzoni, O., Morrica, P., Capasso, G., Pontoni, G., Galletti, P., Zappia, V., 2001. Pharmacokinetics and metabolism of hydroxytyrosol, a natural antioxidant from olive oil. Drug Metab. Dispos. 11, 1492–1498. Del Boccio, P., Di Deo, A., De Curtis, A., Celli, N., Lacoviello, L., Rotilio, D., 2003. Liquid chromatography-tandem mass spectrometry analysis of oleuropein and its metabolite hydroxytyrosol in rat plasma and urine after oral administration. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 785, 47–56. De la Torre-Carbot, K., Jauregui, O., Castellote, A.I., Lamuela-Raventós, R.M., Covas, M.I., Casals, I., López-Sabater, M.C., 2006. Rapid highperformance liquid chromatography-electrospray ionization tandem mass spectrometry method for qualitative and quantitative analysis of virgin olive oil phenolic metabolites in human low-density lipoproteins. J. Chromatogr. A 1116, 69–75. De la Torre-Carbot, K., Chávez-Servin, J.L., Jáuregui, O., Castellote, A.I., Lamuela-Raventós, R.M., Fitó, M., Covas, M.I., Muñoz-Aguayo, D., López-Sabater, M.C., 2007. Presence of virgin olive oil phenolic metabolites in human low density lipoprotein fraction: determination by high-performance liquid chromatography-electrospray ionization tandem mass spectrometry. Anal. Chim. Acta 583, 402–410. Edgecombe, S.C., Stretch, G.L., Hayball, P.J., 2000. Oleuropein, an antioxidant polyphenol from olive oil, is poorly absorbed from isolated perfused rat intestine. J. Nutr. 130, 2996–3002. Fitó, M., Guxens, M., Corella, D., Sáez, G., Estruch, R., de la Torre, R., Francés, F., Cabezas, C., López-Sabater, M.C., Marrugat, J., GarcíaArellano, A., Arós, F., Ruiz-Gutiérrez, V., Ros, E., Salas-Salvadó, J., Fiol, M., Solá, R., Covas, M.I. for the PREDIMED Study Investigators, 2007. Effect of a traditional Mediterranean diet on lipoprotein oxidation: a randomized controlled trial. Arch. Intern. Med. 67, 1195–1203. Fitó, M., Cladellas, M., de la Torre, R., Martí, J., Muñoz, D., Schröder, H., Alcántara, M., Pujadas-Bastardes, M., Marrugat, J., López-Sabater, M.C., Bruguera, J., Covas, M.I. for the SOLOS Investigators, 2008. Antiinflammatory effect of virgin olive oil in stable coronary disease patients: a randomized, crossover, controlled trial. Eur. J. Clin. Nutr. 62, 570–574. Fuller, C.J., Jialal, I., 1994. Effects of antioxidants and fatty acids on low density lipoprotein oxidation. Am. J. Clin. Nutr. 60, 1010–1013. Gimeno, E., de la Torre-Carbot, K., Lamuela-Raventos, R.M., Castellote, A.I., Fitó, M., de la Torre, R., Covas, M.I., López-Sabater, M.C., 2007. Changes in the phenolic content of low density lipoprotein after olive oil consumption in men. A randomized crossover controlled trial. Br. J. Nutr. 98, 1243–1250.
703
Manna, C., Galletti, P., Cucciolla, V., 1996. The protective effect of the olive oil polyphenol (3,4-dihydroxyphenyl)-ethanol counteracts reactive oxygen metabolite-induced cytotoxicity in Caco-2 cells. J. Nutr. 127, 286–292. Manna, C., Galletti, P., Maisto, G., Cucciolla, V., D’Angelo, S., Zappia, V., 2000. Transport mechanism and metabolism of olive oil hydroxytyrosol in Caco-2 cells. FEBS Lett. 470, 341–344. Marrugat, J., Covas, M.I., Fitó, M., Schröder, H., Miró-Casas, E., Gimeno, E., López-Sabater, M.C., de la Torre, R., Farré, M. and the SOLOS Investigators, 2004. Effects of differing phenolic content in dietary olive oils on lipids and LDL oxidation. A randomized controlled trial. Eur. J. Nutr. 43, 140–147. Mateos, R., Goya, L., Bravo, L., 2005. Metabolism of the olive oil phenols hydroxytyrosol, tyrosol, and hydroxytyrosyl acetate by human hepatoma HepG2 cells. J. Agric. Food Chem. 53, 9897–9905. Miró-Casas, E., Farré-Albadalejo, M., Covas Planells, M.I., Fitó Colomer, M., Lamuela Raventós, R.M., de la Torre, R., 2001. Tyrosol bioavailability in humans after ingestion of virgin olive oil. Clin. Chem. 47, 341–343. Miró-Casas, E., Covas, M.-I., Farré, M., Fitó, M., Ortuño, J., Weinbrenner, T., Roset, P., de la Torre, R., et al., 2003a. Hydroxytyrosol disposition in humans. Clin. Chem. 49, 945–952. Miró-Casas, E., Covas, M.I., Fitó, M., Farré-Albaladejo, M., Marrugat, J., de la Torre, R., 2003b. Tyrosol and hydroxytyrosol are absorbed from moderate and sustained doses of virgin olive oil in humans. Eur. J. Clin. Nutr. 57, 186–190. Owen, R.W., Mier, W., Giacosa, A., Hule, W.E., Spiegelhalder, B., Bartsch, H., 2000a. Phenolic compounds and squalene in olive oils: the concentration and antioxidant potential of total phenols, simple phenols, secoroids, lignans and squalene. Food Chem. Toxicol. 38, 647–659. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Spigelhalder, B., Bartsch, H., 2000b. The antioxidant/anticancer potential of phenolic compounds from olive oil. Eur. J. Cancer 36, 1235–1247. Sallustio, B.C., Sabordo, L., Evans, A.M., Nation, R.L., 2000. Hepatic disposition of electrophilic acyl glucuronide conjugates. Curr. Drug. Metab. 1, 163–180. Tuck, K.L., Freeman, M.P., Hayball, P.J., Stretch, G.L., Stupans, I., 2001. The in vivo fate of hydroxytyrosol and tyrosol, antioxidant phenolic constituents of olive oil, after intravenous and oral dosing of labeled compounds to rats. J. Nutr. 131, 1993–1996. Ursini, F., Zamburlini, A., Cazzolato, G., Maiorino, M., Bon, G.B., Sevanian, A., 1998. Postprandial plasma lipid hydroperoxides: a possible link between diet and atherosclerosis. Free Radic. Biol. Med. 25, 250–252. Vissers, M.N., Zock, P.L., Roodenburg, A.J., Leenen, R., Katan, M.B., 2002. Olive oil phenols are absorbed in humans. J. Nutr. 132, 409–417. Visioli, F., Galli, C., Bornet, F., Mattei, A., Patelli, R., Galli, G., Caruso, D., 2000. Olive oil phenolics are dose-dependently absorbed in humans. FEBS Lett. 468, 159–160. Visioli, F., Galli, C., Grande, S., Colonelli, K., Patelli, C., Galli, G., Caruso, D., 2003. Hydroxytyrosol excretion differs between rats and humans and depends on the vehicle of administration. J. Nutr. 133, 2612–2615. Weinbrenner, T., Fitó, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Farré-Albaladejo, M., Abanades, S., Schröder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321. Witzum, J.L., 1994. The oxidation hypothesis of atherosclerosis. Lancet 344, 793–795.
This page intentionally left blank
Chapter 74
Nutrient Profiles of Commercial Table Olives: Proteins and Vitamins Antonio López-López, Alfredo Montaño and Antonio Garrido-Fernández Departamento de Biotecnología de Alimentos, Instituto de la Grasa (CSIC), Sevilla, Spain
80
For information about table olive production and processing styles the reader is referred to Chapter 75 or other specialized sources (Garrido Fernández et al., 1997; IOOC, 2004).
70 Percentage observations
74.1 INTRODUCTION
74.2 PROTEIN AND AMINO ACID CONTENTS IN TABLE OLIVES 74.2.1 Protein
The distribution (Figure 74.1) ranges mainly from 0.5 to 1.5 g 100 g⫺1 edible portion (e.p.) and shows a tail to the right due to commercial presentations stuffed with fish or animal material. The effect of the processing style is significant and green (1.21 ⫾ 0.06 g 100 g⫺1 e.p.) and directly brined (1.23 ⫾ 0.04 g 100 g⫺1 e.p.) olives show (p ⬍ 0.05) higher concentrations than ripe olives (0.89 ⫾ 0.003). Proteins are compounds strongly fixed to the flesh structure and only strong treatments, such as those applied in the elaboration of ripe olives, can partially remove them. Regardless of the processing style (Table 74.1), Hojiblanca, Carrasqueña, and Cacereña (devoted almost exclusively to the preparation of ripe olives) cvs. have the lowest contents. The stuffing materials can increase the protein content (Figure 74.2). AVGA and AVMA had high contents, 3.4 and 3.7 g protein 100 g⫺1 e.p, respectively, due to the high Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
50 40 30 20 10
Proteins of fresh olive fruit mesocarp constitute 1.3–1.8% of its dry weight, and cultivar and fruit ripening do not produce important changes (Zamora et al., 2001).
74.2.2 Protein Distribution According to Styles and Cultivars
60
0
0.5
1.0 1.5 2.0 2.5 3.0 3.5 4.0 Proteins (mg 100g−1 edible portion)
4.5
FIGURE 74.1 Distribution of protein (g 100 g⫺1 edible portion) in table olives. The figure shows the distribution of proteins in table olives. Most commercial presentations have from 0.5–1.5 g 100 g⫺1 edible portion but some presentations have higher contents due to the inclusion of some stuffing materials rich in proteins.
proportion of the stuffing material, together with the additional high protein level in almonds (21.3 g 100 g⫺1 e.p.) (USDA, 2007). AVMPAL, 2.1 g 100 g⫺1 e.p., followed by green Manzanilla stuffed with hazelnut (1.8 g 100 g⫺1 e.p.) and those with stuffing materials of animal origin (green Manzanilla stuffed with ham, 1.8 g 100 g⫺1 e.p., and green Manzanilla stuffed with tuna, 1.7 g 100 g⫺1 e.p.) also show high concentrations.
74.2.3 Amino Acids in Table Olive Proteins Fermented green table olives have a high content of total essential amino acids (EAA) (Montaño et al., 2005) when compared with the FAO/WHO/UNU (1985) reference protein
705
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
706
SECTION | II General Aspects and Changes in Food Processing
TABLE 74.1 Weighted average content of proteins in Spanish commercial table olives according to styles and cultivars. Processing style
Cultivar
Protein content (g 100 g⫺1 e.p.)
Standard errora
nb
Green
Gordal
1.32
0.14
24
Manzanilla
1.24
0.07
74
Carrasqueña
0.90
0.07
6
Hojiblanca
0.84
0.03
8
Gordal
1.31
0.03
2
Manzanilla
1.31
0.07
6
Hojiblanca
0.92
0.02
2
Arbequina
1.45
0.05
2
Aloreña
1.25
0.05
5
Verdial
1.16
0.02
2
Gordal
1.00
0.07
2
Manzanilla
1.14
0.02
2
Carrasqueña
0.90
0.01
4
Hojiblanca
0.93
0.04
6
Cacereña
0.72
0.05
6
Directly brined
Ripe
The protein content in table olives depends on elaboration styles and cultivars and can be related to the energy of processing. a Standard error estimated from GLM analysis; b n, number of determinations; e.p.: edible portion.
pattern for adults. The range of the individual amino acid content according to processing styles is shown in Table 74.2. A more detailed discussion can be found elsewhere (López et al., 2007). In Spanish-style green olives, the major amino acids are Glx, Asx and leucine, as previously reported (Fernández et al., 1985; Montaño et al., 2005). The range for Glx/Asx ratio is 0.9–5.2. Since this ratio is quite close to 1.0 in table olives (Montaño et al., 2005; USDA, 2007), higher values could indicate the addition of MSG. According to USDA (2007), almonds and hazelnuts present Glx/Asx ratios of about 2, which could explain these relatively high values for AVGA, AVMA and AVMPAL (Figure 74.2). Values for total EAA range from 41 g 16 g⫺1 N in green Manzanilla cv. stuffed with natural anchovy to 57 g 16 g⫺1 N in green Gordal stuffed with jalapeño pepper. They are higher than the reference protein values for adults (12.7 g 16 g⫺1 N) or a 2–5-year-old child (33.9 g 16 g⫺1 N) proposed by the FAO/WHO/UNU (1985). Then, the amino acid pro-
file of all presentations within this processing style showed a good balance of total EAA. In this style, according to the FAO pattern for adults, the first limiting amino acid is lysine in most presentations, but in 14 of 47 commercial presentations the most limiting amino acids are cysteine and methionine. In green olive presentations, the chemical score average is 101% but it is lower for Manzanilla and Hojiblanca cultivars (49–56%) based on olives alone; however, the protein chemical score is greater than 100% for commercial presentations from Gordal and Carrasqueña cvs., respectively. Previously reported chemical scores for green Manzanilla cultivar range from 45 to 60% (Montaño et al., 2005). In olives with stuffing materials, the highest scores correspond to a mixture of green Manzanilla olives and capers (155%) and seasoned green Gordal olives (160%). In directly brined olives, with the exception of the limiting amino acid (lysine), differences in amino acids among presentations are reduced (Table 74.2). The Glx/Asx ratio values are quite close to 1, indicating that MSG is
CHAPTER | 74 Nutrient Profiles of Commercial Table Olives: Proteins and Vitamins
707
4 AVMA
Proteins (g 100 g−1 edible portion)
AVGA
AVMPAL 2
Gordal;
0
Manzanilla;
Arbequina;
AVGH AVGA
Aloreña;
Carrasqueña; Verdial,
Hojiblanca Cacereña
AVGJP AVMRS AVMALC AVMJP ANCE ASA AVMALE AVMA AVMPPP ASME AVHPP ANCCP
Processing styles (selection) FIGURE 74.2 Weighted average concentration of proteins (g 100 g⫺1 edible portion) in table olives, according to processing styles and cultivars. Commercial presentations from Gordal and Manzanilla varieties, which include almond and hazelnut as stuffing material, have significantly higher proportions of proteins. AVGA: Green Gordal olives stuffed with whole almonds. AVMA: Green Manzanilla olives stuffed with whole almonds. AVMPAL: Green Manzanilla olives stuffed with ‘piquillo’ pepper and almond.
not added in this style. Average total EAA content is 54 g 16 g⫺1 N, slightly higher (p ⬍ 0.05) than for commercial Spanish-style presentations (51 g 16 g⫺1 N). The best protein quality, based on the chemical score, is for seasoned Gordal, Verdial, and Manzanilla while the worst is for the organically grown Manzanilla presentation. On average, protein quality, as indicated by the mean value of the chemical score, is lower than in the case of Spanishstyle olives. This is an indication that lactic fermentation may improve the quality of table olive protein. In ripe olives, the mean values of each amino acid are generally lower than those for directly brined olives, but the amino acid profiles are quite similar. Total EAA content average in ripe olives is 56 g 16 g⫺1 N (Table 74.2), which is slightly higher (p ⬍ 0.05) than the mean value for directly brined olives. However, the lysine content is lower than in any other style. The lye treatments and water washes used for processing this style may favor lysine loss due to lysinoalanine formation (Meade et al., 2005). The chemical score is low, averaging 21%. Thus, the protein quality in this product is the poorest of all the processing styles.
74.3 VITAMIN E PROFILE IN TABLE OLIVES The only two factors related to vitamin E compounds found in table olives have been α-tocopherol and γ-tocopherol, in
a markedly lower proportion. The α-tocotrienol was found in naturally ripe black olives (0.5–1.0 mg 100 g⫺1 flesh) but almost disappeared after olive processing (Hassapidou et al., 1994). No peaks corresponding to other tocotrienols (Eitenmiller and Landen, 1999) have ever been observed. The α-tocopherol in table olives ranges from 0.13– 5.73 mg 100 g⫺1 e.p. with an overall average of 3.04 mg 100 g⫺1 e.p. There is a good correlation (p ⬍ 0.05) with fat content, but the total variance explained is fairly low (⬇23%). The GLM analysis shows no significant differences among styles in α-tocopherol levels because all treatments applied for processing are aqueous solutions. The mean α-tocopherol content in table olives is similar to those in fresh vegetables such as broccoli (Ching and Mohamed, 2001) but higher than the values (⬍0.1 mg 100 g⫺1 e.p.) given for other pickles like garlic, cucumbers, sauerkraut, and jalapeño pepper (Casado et al., 2004; USDA, 2007). In green olives, the content of α-tocopherol ranges from a minimum of 0.95 to 5.73 mg 100 g⫺1 e.p. with an average of 2.96 (⫾1.00) mg 100 g⫺1 e.p. (Figure 74.3). Most of the green olives have a concentration between 3.5 and 4.0 mg 100 g⫺1 e.p. but higher concentrations are also found in numerous commercial presentations. In directly brined olives, the minimum is 0.94 mg 100 g⫺1 e.p. while the maximum is 5.5 mg 100 g⫺1 e.p. with an average of 3.15 (⫾1.62) mg 100 g⫺1 e.p. This concentration is slightly
708
SECTION | II General Aspects and Changes in Food Processing
22
Amino acid
Green (47)a
Directly brined (9)
Ripe (10)
Asxb
60–299
70–97
46–95
Serine
29–120
34–54
21–48
c
Glx
62–651
75–103
50–99
Glycine
29–144
32–45
21–50
Histidine
14–69
16–23
8–25
Arginine
29–267
40–67
20–56
Threonine
26–94
33–46
20–48
Alanine
29–128
35–54
24–47
Proline
28–128
31–52
21–46
Tyrosine
22–93
25–38
14–48
Valine
32–136
39–56
25–53
Lysine
4–53
1–13
1–6
Isoleucine
30–127
35–49
23–48
Leucine
46–209
54–76
36–75
Phenylalanine
29–156
33–50
20–64
Tryptophan
11–30
13–23
7–17
Cysteine
2–23
5 –9
3–6
Methionine
5–23
8–10
5–12
Total
493–2746
584–817
364–816
Glx/Asx
0.9–5.2
1.0–1.1
1.0–1.1
Total EAAd
41–57
52–56
54–58
Chemical scoree
48–160
12–122
9–54
20 18
Percentage observations
TABLE 74.2 Range of individual amino acids, expressed as mg 100 g⫺1 edible portion, in table olives, according to processing styles.
Mean=2.96, N =95, SD=0.99, Max=5.73, Min=0.95
16 14 12 10 8 6 4 2
Table olives have all the essential amino acids. Their contents are fairly close regardless of the elaboration style but chemical score is markedly lower in ripe olives. a Number of samples included in the range in parenthesis; b Aspartic acid plus asparagine; c Glutamic acid plus glutamine; d Essential amino acid expressed as g 16 g⫺1 N, corrected to 100% nitrogen recovery; e Using the FAO pattern for adults (FAO/WHO/UNU, 1985) as reference and lysine or methionine ⫹ cysteine as the first limiting amino acid.
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
α-Tocopherol (mg 100 g−1 edible portion) FIGURE 74.3 Distribution of α-tocopherol (mg 100 g⫺1 edible portion) in green table olives. Overall, the vitamin E (α-tocopherol) content ranges from 0.5 to 6.0 mg 100 g⫺1 edible portion). In general, table olives may be considered a reasonable source of vitamin E.
higher than that in green olives but without significant differences. In directly ripe olives, the minimum was 0.14 and the maximum 4.44 mg 100 g⫺1 e.p. with an average of 3.27 (⫾1.26) mg 100 g⫺1 e.p. There were differences in α-tocopherols among cultivars within processing styles (Table 74.3). In green olives, the lowest content is found in Gordal cv. (⬇2.0 mg 100 g⫺1 e.p.) while the highest is in Hojiblanca cv. (⬇3.7 mg 100 g⫺1 e.p.). Up to five statistically different groups were established by López et al. (2005a) within this style. The highest concentrations are found in green Gordal or Manzanilla cvs. stuffed with almond and hazelnut, respectively. The presence of this stuffing material, in a high proportion with respect to the eatable portion, increases the concentration of vitamin E because of the usual high content of vitamin E in nuts (Maguire et al., 2004). Presentations with stuffing material from vegetable origin (peppers, garlic, etc.) or fish or animal origin have lower than average vitamin E content due to the low fat content in these products. The pitting and slicing processes tend to produce presentations with the lowest content of vitamin E due to the loss of lipids during these operations. When comparing the same commercial presentations, Hojiblanca has the highest concentrations of vitamin E. The levels found in the green presentations are of the same order as those reported by the USDA, 1.7 mg 100 g⫺1 e.p. (USDA, 2007). In directly brined olives, a similar trend to that observed in green olives is also followed. Hojiblanca and Arbequina (Table 74.3) have the highest vitamin E levels while Gordal cv. has the lowest. The concentration of vitamin E in ripe olives is unexpectedly high (Table 74.3) and similar to those found in most green olive presentations. Apparently, the strong
709
CHAPTER | 74 Nutrient Profiles of Commercial Table Olives: Proteins and Vitamins
TABLE 74.3 Weighted average content of α-tocopherol in Spanish commercial table olives, according to styles and cultivars. Processing style
Cultivar
α-tocopherol (mg 100 g⫺1 e.p.)
Standard errora
nb
Green
Gordal
1.96
0.16
22
Manzanilla
3.29
0.11
59
Carrasqueña
2.63
0.34
6
Hojiblanca
3.65
0.16
8
Gordal
0.72
0.08
2
Manzanilla
3.62
0.09
6
Hojiblanca
5.30
0.19
2
Arbequina
5.25
0.10
2
Aloreña
1.35
0.15
4
Verdial
3.48
0.03
2
Gordal
0.17
0.04
2
Manzanilla
2.26
0.16
2
Carrasqueña
4.08
0.04
4
Hojiblanca
3.96
0.15
6
Cacereña
3.41
0.28
6
Directly brined
Ripe
Processing styles have no effect on the content of α-tocopherol in table olives. In green olives, the highest content is in Hojiblanca and the lowest in Gordal. In directly brined olives, the lowest is in Gordal and the highest in Hojiblanca. In ripe olives, the highest content is found in Carrasqueña and the lowest in Gordal. a Standard error estimated from GLM analysis; b n, number of determinations; e.p.: edible portion.
processing conditions to which this style is subjected do not affect the content of α-tocopherol. The lowest content is in Gordal, as in green or directly brined olives and the highest in Carrasqueña and Hojiblanca cvs. (Table 74.3).
74.4 PROVITAMIN A CAROTENOIDS PROFILE IN TABLE OLIVES Plant carotenoids are the precursors of vitamin A found in the animal kingdom. The carotenoids with provitamin A activity are α-carotene, β-carotene, γ-carotene, and β-cryptoxanthin (Eitenmiller and Landen, 1999). The only provitamin A carotenoid found in all commercial presentations in table olives is β-carotene, while α-carotene is present in only a reduced number of them.
Results will be provided as μg of edible portion. β-Carotene transformation into IU, RE or RAE can be achieved by dividing the data by 0.6, 6, and 12, respectively. In the case of α-carotene the conversion is obtained similarly but the results must also by divided again by two (IOM, 2001). The total activity would then be that resulting from the addition of both compounds. However, only the content of β-carotene will be provided in this work. The content of α-carotene can be obtained independently, according to its source (USDA, 2007). The carotenes present in olive fruits were studied by Mínguez Mosquera and Garrido Fernández (1986), De Castro Ramos et al. (1979), Nosti Vega et al. (1979) and Vázquez Ladrón et al. (1979). Most carotenoids are fairly stable during table olive processing and β-carotene and lutein remains unaltered (Mínguez-Mosquera and
710
SECTION | II General Aspects and Changes in Food Processing
Hornero-Méndez, 1993). Then, differences among styles should be more related to the diverse maturation degree of the raw material than to the effect of processing. In general, higher contents must be expected in green and ripe than in directly brined olives (Mínguez Mosquera and Garrido Fernández, 1986). Green Spanish-style olives are the best source of β-carotene among table olives (average 305 μg β-carotene 100 g⫺1 e.p.) but without significant difference with respect to the levels found in ripe olives (228 μg β-carotene 100 g⫺1 e.p.). The lowest average proportion was found in directly brined olives (less than 200 μg β-carotene 100 g⫺1 e.p.). Within processing styles (Table 74.4) in green table olives, the β-carotene distribution (Figure 74.4) ranges mainly from 0 to 600 μg 100 g⫺1 e.p., although the greatest proportion was from 200–400 μg 100 g⫺1 e.p. A limited number of specialties have also markedly high proportions.
Gordal cultivar shows the lowest β-carotene content (mean of 259 μg β-carotene 100 g⫺1 e.p.). The average concentration of Manzanilla, Carrasqueña and Hojiblanca is fairly similar and ranges from 311 to 328 μg β-carotene 100 g⫺1 e.p. Stuffing material rich in carotenoids leads to presentations with outstanding provitamin A contents (e.g., Manzanilla stuffed with hot pepper has 1386 μg 100 g⫺1 e.p. average). López et al. (2005b) made up to 19 different groups within this style. Within Gordal cultivar, the highest contents are observed in commercial presentations made from olives stuffed with natural pepper, which are rich in carotenoids (Mínguez Mosquera et al., 1988) followed by green Gordal olives prepared as salads which include red pepper streams, carrots and cucumbers, respectively, as ingredients (Scherz and Senser, 1994), whereas the lowest levels are in green Gordal alone (plain and pitted) or stuffed with pepper strips, almond, onions, garlic and
TABLE 74.4 Weighted average content of β-carotene in Spanish commercial table olives according to styles and cultivars. Processing style
Cultivar
β-carotene (μg 100 g⫺1 e.p.)
Standard errora
nb
Green
Gordal
259.3
16.2
22
Manzanilla
311.1
30.4
60
Carrasqueña
327.5
17.8
6
Hojiblanca
313.8
18.4
8
Gordal
152.1
0.3
2
Manzanilla
137.6
27.2
6
Hojiblanca
37.9
0.5
2
Arbequina
113.0
0.6
2
Aloreña
138.0
19.5
4
Verdial
726.7
4.4
2
Gordal
39.7
0.7
2
Manzanilla
151.9
5.1
2
Carrasqueña
234.0
11.6
4
Hojiblanca
332.0
11.4
6
Cacereña
209.6
18.7
6
Directly brined
Ripe
The β-carotene contents in green and ripe olives are fairly similar but directly brined olives have lower proportion. In general, green and ripe olives may be a reasonable source of vitamin A. a Standard error estimated from GLM analysis; b n, number of determinations; e.p.: edible portion.
CHAPTER | 74 Nutrient Profiles of Commercial Table Olives: Proteins and Vitamins
‘jalapeño’, whose ingredients have low or no provitamin A carotenoids at all. Similarly, within the Manzanilla cv. it was possible to distinguish seven groups (López et al., 2005b). The highest concentration is observed in Manzanilla stuffed with hot pepper due to the high concentration of β-carotene in hot pepper and the high proportion of this material with respect to the olives themselves. Other commercial presentations from Manzanilla cv. like the so-called ‘gazpachas’, stuffed with natural peppers, stuffed with ‘piquillo’ pepper and hot pepper strips, etc., which include ingredients with high carotene contents, also have high concentrations of β-carotene. The lowest β-carotene levels were observed in products with materials that do not have or are poor in provitamin A (stuffed with pepper streams, mixture with capers or ‘alcaparrado’, almond, etc.). Commercial presentations from Carrasqueña and Hojiblanca show markedly lower differences with respect to those from the Gordal and Manzanilla cvs. However, commercial presentations from Hojiblanca cv. stuffed with natural pepper (which increases the provitamin A content in this product) are also higher in β-carotene than those which include plain, pitted and sliced olives. Results reported by De Castro Ramos et al. (1979) and Vázquez Ladrón et al. (1979) were usually lower than those given here but similar to those found by the USDA Nutrient Database for Standard Reference for green, canned or bottled olives (USDA, 2007). Directly brined olives have lower provitamin A content than green olives due to their advanced degree of maturation (Mínguez Mosquera and Garrido Fernández, 1986). Many of the commercial presentations in this style are ‘seasoned’ olives and their β-carotene content is due not only to their
Mean =300.5, N =96, SD=191.3, Max=1433.5, Min =143.8
50
Percentage observations
olive content but also to the contribution of the ingredients such as sliced carrots, pieces of pepper strips, etc. (Marx et al., 2000) which constitute an important part of the product. Values included in Table 74.4 are markedly higher than those reported by Nosti Vega et al. (1979) for Manzanilla and Hojiblanca, elaborated according to this style. Ripe olives have relatively high contents of β-carotene (Table 74.4). β-Carotene is not only stable during the brine or acidic storage (Minguez-Mosquera and Gallardo Guerrero, 1995) to which olives are subjected previously to oxidation but also to the diverse alkaline treatments applied for the oxidation (darkening) process. The lowest content of β-carotene in ripe olive commercial presentations is found in Gordal, possibly due to an excessive strong processing energy. The highest content (332 μg β-carotene 100 g⫺1 e.p.) is observed in Hojiblanca regardless of the presentation form (plain, pitted or slices) and is similar to levels found for the same cultivar when prepared as green but higher than Hojiblanca directly brined olives. Carrasqueña, Cacereña and Manzanilla had progressively decreasing concentrations of β-carotene (Table 74.4). The values found in Spanish ripe olives are of the same order as those provided by the USDA Nutrient Database for Standard Reference (USDA, 2007). The average content of β-carotene in table olives is lower than in carrot or carrot juices (Marx et al., 2000), similar to fresh cucumber (Scherz and Senser, 1994) and higher than in pickles, sour cucumber (USDA, 2007), sauerkraut or onions (Scherz and Senser, 1994) and in most of the vegetables analyzed by Hart and Scott (1995). Thus, table olives, in general, and some commercial presentations, in particular, can be considered a reasonable source of provitamin A carotenids.
74.5 VITAMIN B6 IN TABLE OLIVES
60
40 30 20 10 0
711
0
200 400 600 800 1000 1200 1400 1600 β-Carotene (μg 100 g−1 edible portion)
FIGURE 74.4 Distribution of β-carotene (μg 100 g⫺1 edible portion) in green table olives. Most table olive presentations show values of βcarotene from 200 to 400 μg 100 g⫺1 edible portion but those including carrots, red pepper or other products rich in β-carotene may have higher proportions (400–600 μg 100 g⫺1 edible portion).
There are different forms of vitamin B6 but all of them have the same vitamin activity (Driskell, 1984) and are simultaneously determined by HPLC (Bergaentzlé et al., 1995). Vitamin B6 content in table olives ranges from 0 to 72 μg 100 g⫺1 e.p. and its mean value is 15.6 (⫾13.6) μg 100 g⫺1 e.p. Most of the commercial presentations (about 54%) have between 0 and 10 μg 100 g⫺1 e.p. The content of this vitamin is strongly affected by the processing styles (p ⬍ 0.05) (Figure 74.5). The highest content is found in directly brined olives (mean 35.4 ⫾ 3.2 μg 100 g⫺1 e.p.), which is similar to that found in many other canned vegetables (Martín-Belloso and Lanos-Barriobero, 2001). Green olives (mean 14.0 ⫾ 1.1 μg 100 g⫺1 e.p.) and ripe olives (5.0 ⫾ 0.9 μg 100 g⫺1 e.p.) have progressively lower levels. This behavior can be attributed to losses due to the progressive severity of processing treatments (Garrido Fernández et al. 1997). Values reported by the USDA are 31, 12 and 9 μg 100 g⫺1 e.p., for green, ripe jumbo-super colossal,
712
SECTION | II General Aspects and Changes in Food Processing
olives stuffed with natural red pepper (pimento or cherry pepper variety), with a content of 215 μg 100 g⫺1 e.p, according to USDA (2007). In directly brined olive commercial presentations, Manzanilla and Gordal ‘seasoned’ turning color show the highest content of vitamin B6 (Table 74.5) and presentations from Arbequina and, especially, Aloreña cultivars shows the lowest contents. In commercial presentations of ripe olives, Cacereña olives show the highest content of vitamin B6 (plain, 13.3 μg 100 g⫺1 e.p.; pitted, 9.1 μg 100 g⫺1 e.p.), whereas Hojiblanca shows a fairly low content and the vitamin was absent from Gordal olives.
Vitamin B6 (μg 100 g−1 edible portion)
45 40 35 30 25 20 15 10 5 0 -5
Green
Directly brined
Ripe
Processing styles
FIGURE 74.5 Comparison of vitamin B6 content (μg 100 g⫺1 edible portion) among olive processing styles. Vertical bars denote 95% confidence intervals. The processing style has a significant effect on the content of vitamin B6, as the stronger the processing treatment the lower the content in vitamin B6.
45
Percentage observations
40
Mean =14.1, N =96, SD=11.1, Max =71.7, Min =1.3
35 30 25 20 15 10 5 0
0
10
20
30
40
50
60
70
80
Vitamin B6 (μg 100 g−1 edible portion) FIGURE 74.6 Histogram for the distribution of vitamin B6 (μg 100 g⫺1 edible portion) in green commercial table olives. Most commercial presentations have from 0 to 20 μg 100 g⫺1 edible portion. However, more than 15% of them have values between 20 and 30 μg 100 g⫺1 edible portion.
and ripe small–extra large commercial presentations, respectively (USDA, 2007). The distribution of vitamin B6 in green olives (Figure 74.6) shows a high proportion of samples between 10 and 20 μg 100 g⫺1 e.p., the presentation with the greatest proportion being green Manzanilla cv. stuffed with natural hot pepper (about 70 μg 100 g⫺1 e.p.) due to the contribution of the red pepper, with a vitamin B6 content of 120 μg 100 g⫺1 e.p. according to USDA (2007). Within processing styles (Table 74.5), in green style, Gordal cv. shows the highest average (⬇22 μg 100 g⫺1 e.p.), followed by Carrasqueña. Presentations from Manzanilla cv. show relatively low concentrations in spite of including
74.6 VITAMIN C IN TABLE OLIVES Vitamin C is made internally by almost all organisms, humans being the most well-known exception. More than 90% of the vitamin C in human diets is supplied by fruits and vegetables (Lee and Kader, 2000). There are two biologically active forms of vitamin C, AA and DHAA. Apart from its natural origin, vitamin C is often added to green table olives as a pH corrector and as an antioxidant to prevent browning. Due to its relative instability, it is destroyed during ripe olive processing and it is absent from all their presentations. The average content of vitamin C in green fresh olives is 8.9 ⫾ 0.1 mg 100 g⫺1 e.p. (López et al., 2004). With respect to the final product, no statistical considerations can be made since contents depend on the particular processing practices of each industry. When ascorbic acid is not added, the vitamin content may range from not detected (⬍0.01 mg 100 g⫺1 e.p.) to 0.6 mg 100 g⫺1 e.p. In these cases, the proportion of DHAA is higher than that of AA and may even reach 100% of the total AA. The relatively high values (1.3–3.1 mg AA 100 g⫺1 e.p.) found by Vázquez Ladrón et al. (1979) in plain Spanish-style green olives (Manzanilla and Gordal cultivars) without added AA, could be attributed to the presence of interfering compounds in the AOAC visual titration method used by these authors. When AA can be added, there are no uniform practices (some industries may add it and others not) and therefore, on average, diverse concentrations of total AA can be found. Typical results found in Manzanilla cv. stuffed with capers is 2.9 mg 100 g⫺1 e.p. whereas green ‘seasoned’ Manzanilla has about 13 mg 100 g⫺1 e.p. In the case of cultivars more prone to browning, the content may be higher, e.g., pitted Hojiblanca (⬇24 mg 100 g⫺1 e.p.) or green Hojiblanca stuffed with pepper strips (⬇7 mg 100 g⫺1 e.p.). The highest vitamin C content can be found in the presentations in which this ingredient is mentioned on the label. In these, levels of vitamin C may range from 3 to 36 mg 100 g⫺1 e.p. The wide range of total AA found may indicate that this compound is progressively destroyed during shelf life. The AA can be oxidized to DHAA which may, in
713
CHAPTER | 74 Nutrient Profiles of Commercial Table Olives: Proteins and Vitamins
TABLE 74.5 Weighted average content of vitamin B6 in Spanish commercial table olives, according to styles and cultivars. Standard errora
nb
Processing style
Cultivar
Vitamin B6 (μg 100 g⫺1 e.p.)
Green
Gordal
22.3
1.3
22
Manzanilla
11.8
1.5
60
Carrasqueña
15.7
2.1
6
Hojiblanca
7.7
1.6
8
Gordal
50.1
1.8
2
Manzanilla
40.8
3.4
6
Hojiblanca
31.7
0.4
2
Arbequina
27.0
1.5
2
Aloreña
24.7
10.9
4
Verdial
38.3
0.1
2
Gordal
Nd
Directly brined
Ripe
----
2
Manzanilla
4.7
0.3
2
Carrasqueña
4.0
0.8
4
Hojiblanca
3.8
1.3
6
Cacereña
8.8
1.7
6
Vitamin B6 is hydrosoluble. The overall higher proportion was in directly brined olives, followed by green and ripe olives. a Standard error estimated from GLM analysis; b n, number of determinations; e.p.: edible portion.
later steps, be transformed into other degradation products like 2,3-diketogulonic acid, furfural or 5-hydroxymethylfurfural) (Smoot and Nagy, 1980). On the contrary, the application of pasteurization (treatment extensively used by the industry) can stabilize the added AA. In commercial presentations of directly brined olives, regardless of cultivar, the levels of total AA found are always low (⬍ 0.4 mg 100 g⫺1 e.p.), the major fraction being DHAA, with the only exception of turning color ‘seasoned’ Hojiblanca olives (8.69 mg 100 g⫺1 e.p.). From a nutritional (vitamin C content) standpoint, presentations with a total AA content of ⬃10 mg 100 g⫺1 e.p. appear similar to some fresh vegetables (e.g., tomatoes, potatoes), whereas presentations with contents of 20–40 mg 100 g⫺1 e.p. are similar to some fruits (e.g., blackberries, grapefruit, raspberries, mandarins) (Lee and Kader, 2000).
SUMMARY POINTS ●
●
●
●
●
Proteins in table olives range mainly from 0.5 to 1.5 g 100 g⫺1 e.p. Green and directly brined are the best sources of this compound and their essential amino acids range from 41 to 57 g 16 g⫺1 N. Regardless of processing styles, table olives are a good source of vitamin E, with an average of 3.04 mg 100 g⫺1 e.p. The mean content of pro-vitamin A (β-carotene) ranges from 305 in green olives to 200 μg 100 g⫺1 e.p., in directly brined olives. The mean concentration of vitamin B6 ranges from 35.4, directly brined olives to 5.0 μg 100 g⫺1 e.p., ripe olives. Vitamin C content in green and directly brined olives is diverse and depends on its addition in packing.
714
SECTION | II General Aspects and Changes in Food Processing
REFERENCES Bergaentzlé, M., Arella, F., Bourguignon, J.B., Hasselmann, C., 1995. Determination of vitamin B6 in foods by HPLC: a collaborative study. Food Chem. 52, 81–86. Casado, F.J., López, A., Rejano, L., Sánchez, A.H., Montaño, A., 2004. Nutritional composition of commercial pickled garlic. Eur. Food Res. Technol. 219, 355–359. Ching, L.S., Mohamed, S., 2001. α-tocopherols in 62 edible tropical plants. J. Agric. Food Chem. 49, 3101–3105. De Castro Ramos, R., Nosti Vega, M., Vázquez Ladrón, R., 1979. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. I. Aceitunas verdes aderezadas al estilo sevillano. Grasas Aceites 30, 83–91. Driskell, J.A., 1984. Vitamin B6. In: Machlin, H.J. (Ed.), Handbook of Vitamins – Nutritional, Biochemical and Clinical Aspects. Marcel Dekker, New York, pp. 379–401. Eitenmiller, R.R., Landen, W.O., 1999. Vitamin Analysis for the Health and Food Science. CRC Press, Boca Ratón, FL, pp. 109–148. FAO/WHO/UNU expert consultation. 1985. Energy and Protein Requirements. FAO/WHO Nutrition Meetings, Report Series 724. Food and Agriculture Organization/World Health Organization, Geneva. Fernández, M.J., Castro, R., Garrido, A., Cancho, F.G., Pellissó, F.G., Vega, M.N., Moreno, A.H., Mosquera, I.M., Rejano, L., Quintana, M.C.D., Roldán, F.S., García, P., Castro, A., 1985. Biotecnología de la aceituna de mesa. CSIC (Ed), Madrid. Garrido Fernández, A., Fernández Díez, M.J., Adams, M.R., 1997. Table Olives. Production and Processing. Chapman & Hall, London, UK. Hart, D.J., Scott, K.J., 1995. Development and evaluation of an HPLC method for the analysis of carotenoids in foods, and the measurements of the carotenoid content of vegetables and fruits commonly consumed in the UK. Food Chem. 54, 101–111. Hassapidou, M.N., Balatsouras, G.D., Manoukas, G.D., 1994. Effects of processing upon the tocopherol and tocotrienol composition of table olives. Food Chem. 50, 111–114. IOM (Institute of Medicine), 2001. Dietary Reference Intakes for Vitamin A. Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. National Academic Press, Washington, D.C. IOOC (International Olive Oil Council). 2004. International Trade Standards Applying to Table Olives. Document COI/NC no 1. December 2004. Madrid, Spain. Lee, S.K., Kader, A.A., 2000. Pre-harvest and post-harvest factors influencing vitamin C content of horticultural crops. Postharvest Biol. Tec. 20, 207–220. López, A., Garrido, A., Montaño, A., 2007. Proteins and amino acids in table olives: relationship to processing and commercial presentation. Ital. J. Food Sci. 19, 217–228. López, A., Montaño, A., Garrido, A., 2005a. Evaluation of vitamin E by HPLC in a variety of olive based foodstuffs. J. Am. Oil Chem. Soc. 82, 129–133. López, A., Montaño, A., Garrido, A., 2005b. Provitamin A carotenoids in table olives according to processing styles, cultivars and commercial presentations. Eur. Food Res. Technol. 221, 406–411.
López, A., Montaño, A., García, P., Garrido, A., 2004. Quantification of ascorbic acid in fresh olives and in commercial presentations of table olives. Food Sci. Technol. Int. 11, 199–204. Maguire, L.S., O’Sullivan, S.M., Galvin, K., O’Connor, T.P., O’Brien, N.M., 2004. Fatty acid profile, tocopherol, squalene, and phytosterol content of walnuts, almonds, peanuts, hazelnuts and the Macadamia nut. Int. J. Food Sci. Nutr. 55, 171–178. Martín-Belloso, O., Lanos-Barriobero, E., 2001. Proximate composition, minerals and vitamins in selected canned vegetables. Eur. Food Res. Technol. 212, 182–187. Marx, M., Schieber, A., Carle, R., 2000. Quantitative determination of carotene stereoisomers in carrot juices and vitamin supplement (ATBC) drinks. Food Chem. 70, 403–408. Meade, S.J., Reid, E.A., Gerrard, J.A., 2005. The impact of processing on the nutritional quality of food proteins. J. AOAC Int. 88, 904–922. Mínguez Mosquera, M.I., Gallardo Guerrero, L., 1995. Anomalous transformation of chloroplastic pigments in Gordal variety olives during processing for tables olives. J. Food Protect. 58, 1241–1248. Mínguez Mosquera, M.I., Garrido Fernández, J., 1986. Identificación de pigmentos carotenoides en frutos de distintas variedades de olivo Olea europaea L. Grasas Aceites 37, 272–276. Mínguez Mosquera, M.I., Montaño Asquerino, A., Garrido Fernández, J., Gandul Rojas, B., 1988. Separation and identification by HPLC of chlorophylls and carotenoids from olives. Grasas Aceites 39, 363–366. Mínguez-Mosquera, M.I., Hornero-Méndez, D., 1993. Separation and quantification of the carotenoid pigments in reed peppers (Capsicum Nahum, L.) paprika, and oleoresin by reversed-phase HPLC. J. Agric. Food Chem. 41, 1616–1620. Montaño, A., Casado, F.J., Castro, A., Sánchez, A.H., Rejano, L., 2005. Influence of processing, storage time, and pasteurisation upon the tocopherol and amino acid contents of treated green table olives. Eur. Food Res. Technol. 220, 255–260. Nosti Vega, M., Vázquez Ladrón, R., De Castro Ramos, R., 1979. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. II. Aceitunas verdes en salmuera. Grasas Aceites 30, 93–100. Scherz, H., Senser, F., 1994. Food Composition and Nutrition Tables (Deusche Forschungsanstalt fur Lebersnmittelchemie Garching). Medpharm, Scientific Publishers, Stuttgart and CRC Press, Boca Raton. Smoot, J.M., Nagy, S., 1980. Effects of storage temperature and duration on total vitamin C content of canned single-strength grapefruit juice. J. Agric. Food Chem. 28, 417–421. USDA, Agricultural Research Service. 2007. USDA Nutrient database for standard reference, release 20. Nutrient data laboratory home page, http://www.nal.usda.gov/fnic/foodcomp. Vázquez Ladrón, R.V., Castro, R., Vega, M.N., 1979. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. III. Aceitunas verdes aderezadas envasadas. Grasas Aceites 30, 221–226. Zamora, R., Alaiz, M., Hidalgo, F.J., 2001. Influence of cultivar and fruit ripening on olive (Olea europaea) fruit protein content, composition, and antioxidant activity. J. Agric. Food Chem. 49, 4267–4270.
Chapter 75
Nutrient Profiles of Commercial Table Olives: Fatty Acids, Sterols, and Fatty Alcohols Antonio López-López, Alfredo Montaño and Antonio Garrido-Fernández Departamento de Biotecnología de Alimentos, Instituto de la Grasa (CSIC), Sevilla, Spain
75.1 INTRODUCTION Table olives constitute an important part of the Mediterranean diet. Their world production reached a total of 1 762 000 tons in the 2005/2006 season (IOOC, 2007). Table olives are also becoming more and more popular in many non-producing countries. The most common are green olives (Spanish-style), directly brined olives, and ripe olives (Californian style). The procedure for preparing green Spanish-Seville (green from now on)-style olives consists of treating the fruits with a dilute lye (NaOH) solution, followed by washing and brining. The commercial presentations of green olives are numerous and include the use of many stuffing materials (Garrido Fernández et al., 1997). Directly brined, untreated, olives (green, turning color, or naturally black) are just immersed in brine, where they lose some of their natural bitterness. Then, olives are sorted, graded and packed. Occasionally, they can be cracked or cut along their higher longitudinal diameter and packed with some seasoning material (natural or essential oils). Ripe olives (by alkaline oxidation) are previously preserved in an aqueous solution (brine or acidic water) and darkened throughout the year. Darkening consists of several treatments with lye solutions and water washings with aeration between them. Their most common commercial presentations are plain (whole), pitted, sliced, and olive paste. Any cultivar can be used for table olives but only a reduced number of them have reached general acceptance. Among them, the Spanish Gordal and Manzanilla are predominantly used for green olives, Cacereña and Hojiblanca for ripe olives while these or other more local cultivars are used for directly brined olives. Conservolea from Greece is also very appreciated for preparing natural black olives and Kalamata cv. is the name of a specialty prepared from it. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Table olive processing implies several treatments with diverse water solutions. The alkaline treatment in green olives hydrolyzes the natural polyphenols and strongly modifies the flesh and skin structure. In ripe olives, the successive lye treatments contribute to the polymerization of polyphenols and to fruit darkening. During the immersion in brine, olives usually undergo a fermentation process, with a predominance of lactic acid bacteria (green) or yeast (directly brined olives), which also contributes to the sweetening of the untreated fruits. During processing, washing, and brining of olives, most of the water-soluble compounds present in the olive flesh are lost. In addition, the fermentation/storage process also produces important modifications in the olive composition by using nutrients (sugars, vitamins, etc.) and producing new substances (lactic or acetic acids, volatiles, etc.) (Garrido Fernández et al., 1997). There are only a few studies about the fat composition of commercial table olives or the changes that these compounds suffer during processing. This text is related to the composition of table olives based mainly on the data available from the Spanish cultivars. The results presented here include 67 of the most usual commercial presentations.
75.2 PROXIMATE COMPOSITION OF TABLE OLIVES To disclose the role played by lipids, sterols and fatty and triterpenic alcohols in table olives, it is important to examine the proximate composition of their diverse styles (Fernández Díez et al., 1985). Values are provided in Table 75.1. Apart from moisture, lipids are the major component of table olives with an overall range of 9 to 28 g 100 g⫺1 edible portion (e.p.) Minerals are also in marked proportions (2.0–6.9 g
715
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
716
SECTION | II General Aspects and Changes in Food Processing
24
TABLE 75.1 Proximate composition (g 100 g⫺1 edible portion) of the diverse styles of table olives (Fernández Díez et al., 1985). Component
Green olivesa Range
Ripe olivesb Typical value
Green
20
Directly brined olivesc Typical value
16 12 8
Moisture
61–81
69
60
Lipids
9–28
21
23
Protein
1.0–1.5
1.1
1.2
Fiber
1.5–2.5
2.4
2.2
Ash
4.2–5.5
2.0
6.9
4 0
8
Typically, apart from moisture, lipids are the most abundant components, followed by ash, fiber and proteins. a Ranges include Gordal, Manzanilla, Hojiblanca and Verdial cvs; b Hojiblanca cv; c Naturally black Hojiblanca cv.
100 g⫺1 e.p.), followed by fiber (1.5–2.5 g 100 g⫺1 e.p.). Sugars are practically absent in the commercial product due to their transformation/removal during the fermentation/storage processes. Thus, lipids play an outstanding role in the nutritional value of table olives. There are marked differences within processing styles due to cultivars (Table 75.1). So, a nutritional evaluation of lipids in table olives must consider the possible effects of processing styles and cultivars.
Percentage observations
8
10 12 14 16 18 20 22 24 26 Directly brined
6
4
2
0
8
10 12 14 16 18 20 22 24 26
8 7 Ripe
6 5 4 3
75.3 LIPID CONTENT IN TABLE OLIVES 75.3.1 Fat Distribution According to Processing Styles Different styles of table olives use diverse degrees of maturation. Usually, the fat content increases as the maturation degree progresses. An index was established to fix the appropriate maturation stage for picking olives for olive oil extraction (Hermoso et al., 2001) but no similar method is used for table olives. An initial classification of them is related to the surface color of fruits (IOOC, 2004). The standard distinguishes among green, turning color and naturally black olives but ripe olives are picked in the green or turning color maturation stage. In addition, harvesting takes a period of 1–2 months. As a result, the diverse styles present a range of fat content. The overall average of lipids in table olives is 16.2 g 100 g⫺1 e.p. The range of fat concentration distribution in green olives is fairly wide (Figure 75.1), from 6 to 24 g 100 g⫺1 e.p. due to the diverse fat concentrations in the
2 1 0
8
10 12 14 16 18 20 22 24 26 Total lipids (g 100 g−1 edible portion)
FIGURE 75.1 Distribution of the total lipids (g 100 g⫺1 edible portion) according to processing styles. The figure shows the distribution of lipids in the three main table olive styles. Usually, the more mature the raw material the higher the lipid content in the final product.
cultivars devoted to this style and in the stuffing materials. The range of contents in ripe olives also has a wide distribution range from about 8 to 24 g 100 g⫺1 e.p. and, in practice, overlaps with that of green olives. The distribution of fat in directly brined olives is displaced to higher levels (18–28 g 100 g⫺1 e.p.). These values are in agreement with the fat concentration found by Vázquez Ladrón et al. (1979) in green olives, ⬇9 g 100 g⫺1 flesh in Gordal and about 21 g 100 g⫺1 flesh for Manzanilla cv. In directly brined (untreated) green
717
CHAPTER | 75 Nutrient Profiles of Commercial Table Olives: Fatty Acids, Sterols, and Fatty Alcohols
Total lipids (g 100 g−1 edible portion)
25
TABLE 75.2 Weighted average content of lipids in Spanish commercial table olives, according to styles and cultivars.
24 23 22 21 20 19 18
Processing Cultivar style
Total fat (g 100 g⫺1 e.p.)
Standard errora
nb
Green
Gordal
11.5
0.5
22
Manzanilla
16.2
0.5
60
Carrasqueña
13.5
1.6
6
Hojiblanca
14.7
0.4
8
Gordal
19.7
0.2
2
Manzanilla
21.3
0.8
6
Hojiblanca
18.7
0.1
2
Arbequina
30.7
0.4
2
Aloreña
22.0
0.4
4
Verdial
18.9
0.6
2
Gordal
9.2
0.5
2
Manzanilla
18.2
0.7
2
Carrasqueña
20.6
0.8
4
Hojiblanca
14.9
0.6
6
Cacereña
15.6
0.8
6
17 16 15 14 13
Green
Directly brined
Ripe
Processing styles FIGURE 75.2 Total lipid (g 100 g⫺1 edible portion) contents in commercial presentations of table olives, according to processing styles. The concentrations of lipids in green and ripe olives are similar but the level in directly brined olives is significantly higher.
olives, the range reported by Nosti Vega et al. (1979) was 21 (Hojiblanca) and 29 (Manzanilla) g 100 g⫺1 flesh while in Hojiblanca processed as ripe olives the values were closer (20–22 g 100 g⫺1 flesh) (Nosti Vega and de Castro Ramos, 1985). The total fat reported by Ünal and Nergiz (2003) in green olives was about 15 g 100 g⫺1 e.p. while the concentrations found for Kalamata and naturally black olives were ⬇22 and ⬇25 g 100 g⫺1 e.p., respectively. Total lipids for jumbo-super colossal and small-extra large ripe olives and green olives reported in the USDA National Database for Standard Reference are ⬇6.9, 10.7, and 15.3 g 100 g⫺1 e.p. (US Department of Agriculture, 2007). The fat content of marinated green olives in the Souci-Fachmann-Kraut database is about 14 g 100 g⫺1 e.p. (Scherz and Senser, 2002). The average value found by Balatsouras (1980) in Conservolea and Kalamata olives processed as naturally ripe olives was from 24 to 36 g 100 g⫺1 e.p. while green Conservolea was from 15 to 18 g 100 g⫺1 e.p. (Vamvoukas et al., 1980). The average content of lipids in the Spanish commercial presentations of directly brined olives is higher (p ⬍ 0.05) than in green or ripe olives, which have approximately the same concentrations (Figure 75.2). Their average values were 21.8 ⫾ 0.9, 14.8 ⫾ 0.4, and 16.1 ⫾ 0.8 g 100 g⫺1 e.p., respectively. However, within styles, there are also marked differences due to the diverse lipid content in different cultivars (Table 75.2). In green olives, there are no marked differences among cultivars and their averages ranged from 11.5 to 16.2 g 100 g⫺1 e.p., which is in agreement with the perception of consumers. Some commercial presentations can be especially low in fat, e.g., Gordal stuffed with cucumber, due to the low content of the stuffing material. However, the effect of the stuffing material is limited in most cases because the proportion of stuffing material is
Directly brined
Ripe
Fat content depends on maturation degree of the raw material. It is higher in directly brined, followed by green and ripe olives. Gordal is the cultivar with the lowest lipid content. a Standard error estimated from GLM analysis; b n, number of determinations; e.p.: edible portion.
relatively low (about 5–6%) with respect to the total weight of the stuffed olive. Greater differences can be found in directly brined olives due to the natural differences in fat of the raw material. Their averages range from 18.7 to 30.7 g 100 g⫺1 e.p. Arbequina cv. is the richest in fat, and has a fairly small fruit and a low flesh/pit relationship, but differences among the other cvs. are not marked. When cultivars used for preparing green olives (Gordal, Manzanilla, and Hojiblanca) are used in this style, their fat content is usually higher than that in green because the maturation stage for this style is more advanced. In ripe olives, differences in fat content are wide, ranging from 9.2 to 20.6 g 100 g⫺1 e.p., with the level in Carrasqueña (quite similar in characteristics to Manzanilla) particularly high, possibly due to
718
SECTION | II General Aspects and Changes in Food Processing
its more advanced maturation stage. On the contrary, differences among Manzanilla, Hojiblanca and Cacereña (almost entirely devoted to preparing ripe olives) are scarce.
75.3.2 Fatty Acid Content in Table Olives The most abundant fatty acids in table olives are (Figure 75.3): C18:1, C16:0, C18:2 n-6, and C18:0, with averages of 10.0, 2.5, 1.0, and 0.5 g 100 g⫺1 e.p., respectively. Other acids in lower proportions are C18:1t, C16:1, C18:3 n-3, and C20:0, with averages ranging from approximately 0.07 to 0.17 g 100 g⫺1 e.p. The rest of the acids are in proportions below 0.05 g 100 g⫺1 e.p. The average content, according to elaboration styles, and their standard error are shown in Table 75.3. It is worthwhile to emphasize the high proportion in directly brined olives of C18:1 (13.5 g fatty acid 100 g⫺1 e.p.). The effect of cultivar was also marked. The level of C18:1 ranged from 6.9 (Gordal) to 10.7 (Manzanilla) g fatty acids 100 g⫺1 e.p., in green olives; from 11.4 (Verdial) to 19.2 (Arbequina) g fatty acids 100 g⫺1 e.p., in directly brined; and from 5.5 (Gordal) to 13.8 (Carrasqueña) g fatty acids 100 g⫺1 e.p.,
in ripe olives. The second most abundant fatty acid, C16:0 ranged from 1.2 (Gordal) to 2.5 g fatty acids 100 g⫺1 e.p. (Manzanilla) in green olives, from 2.1 (Hojiblanca) to 5.1 (Arbequina) g fatty acids 100 g⫺1 e.p. in directly brined olives, and from 1.2 (Gordal) to 2.9 (Carrasqueña) in ripe olives. More detailed information about the diverse fatty acid composition according to cultivars can be found elsewhere (López et al., 2006). Ünal and Nergiz (2003) found that in Memecik cv., the most abundant fatty acid in green olives was also C18:1 (67.5%), followed by C16:0 (16.4%), and C18:2 (11.9%) after 12 months’ storage while in Kalamata or directly brined natural black olives the proportions were slightly higher. Nosti Vega and de Castro Ramos (1985) reported proportions of C18:1 in the range 74–81%, C16:0 from 10.6 to 13.7%, C18:2 from 4.6 to 7.7%, and C18:0 from 2.4 to 3.0%. The fatty acid composition for jumbosuper colossal and small-extra large ripe olives and green olives reported in the USDA National Database for Standard Reference were also in agreement to values reported here (US Department of Agriculture, 2007). The Souci-FachmannKraut database reports only a limited number of fatty acids for table olives, but their contents were also included in those shown for directly brined olives (Scherz and Senser, 2002).
18 16
Mean
Mean±SD
Mean±1,96*SD
14
Fatty acid content (g 100 g−1 edible portion)
12 10 8 6 4 2 0 C16:0
C18:0
C18:1
C18:2 n-6
0.30 0.25
Mean
Mean±SD
Mean±1,96*SD
0.20 0.15 0.10 0.05
C14:0 C15:0 C17:0 C20:0 C21:0 C22:0 C23:0 C24:0 C16:1 C17:1 C20:1 C24:1 C18:3 n-3 C20:2 C20:3 n6 C22:2 n-6 C18:1t C18:2t C18:3t
0.00
Fatty acid (acronym) FIGURE 75.3 Graph Box-Whisker of the fatty acid content (g 100 g⫺1 edible portion) in table olives. The concentration of the diverse fatty acids in table olives is not homogeneous. The most abundant are oleic, palmitic, linoleic, and stearic acids.
75.3.3 Nutritional Fat Fractions in Table Olives The highest content of triglycerides (total fat for nutritional labeling purposes), was found in directly brined olives (Table 75.4). Arbequina cv. (30 g 100 g⫺1 e.p.) has the highest proportion, followed by Manzanilla, Hojiblanca and Gordal with levels of about 20 g 100 g⫺1 e.p. In green table olives, the concentrations are, in general, lower and range from about 11 g 100 g⫺1 e.p. (Gordal) to 16 g 100 g⫺1 e.p. (Manzanilla). Ripe olives have concentrations of triglycerides fairly similar to green olives; the highest content is observed in Carrasqueña, which has about 20 g 100 g⫺1 e.p., followed by Manzanilla (18 g 100 g⫺1 e.p.). These values are comparable to those reported by Ünal and Nergiz (2003), who also found a gradation from green to naturally black olives, or Borzillo et al. (2000) for Oinotria table olives. The averages of the separate fat labeling fractions are higher in directly brined olives because of their greater fat content in general (Table 75.4). The MUFA fraction is the main component, due to the high content of oleic acid, and ranges from 5.7 to 19.4 g fatty acid 100 g⫺1 flesh. As in olive oil, it accounts for about 60–80% of the fat content in table olives. The proportion of SFA, 1.5–6.0 g fatty acid 100 g⫺1 flesh, is comparatively low. The content of PUFA is also marked and ranged from 0.5 to 3.9 g fatty acid 100 g⫺1 flesh. PUFA is about 1 g 100 g⫺1 e.p. in most green and ripe olives, except for Carrasqueña (green), Gordal (ripe), and Cacereña (ripe) which are poorer in such fat fraction. TFA had a very limited presence (0.1–0.4 g fatty acid 100 g⫺1
719
CHAPTER | 75 Nutrient Profiles of Commercial Table Olives: Fatty Acids, Sterols, and Fatty Alcohols
TABLE 75.3 Average values (⫾standard error) of fatty acid composition, expressed as g 100 g⫺1 edible portion, according to elaboration styles. Green olives (n ⫽ 96)b
Fatty acid
Directly brined olives (n ⫽ 18)
Ripe olives (n ⫽ 20)
Average
Standard errora
Average
Standard error
Average
Standard error
C14:0
0.0049
0.0006
0.0049
0.0002
0.0015
0.0003
C15:0
0.0007
0.0001
0.0004
0.0002
Nd
–
C16:0
2.28
0.05
3.16
0.19
2.16
0.13
C17:0
0.0243
0.0005
0.0263
0.0025
0.0205
0.0020
C18:0
0.3458
0.0095
0.5110
0.0252
0.338
0.0256
C20:0
0.0726
0.0017
0.1078
0.0063
0.0706
0.0048
C21:0
0.0029
0.0003
0.0025
0.0008
0.0047
0.0007
C22:0
0.0209
0.0006
0.0312
0.0029
0.0205
0.0016
C23:0
0.0024
0.0002
0.0033
0.0009
0.0018
0.0005
C24:0
0.0083
0.0002
0.0118
0.0010
0.0077
0.0005
C16:1
0.1372
0.0051
0.1433
0.0195
0.1228
0.0102
C17:1
0.0457
0.0010
0.0496
0.0049
0.0402
0.0039
C18:1
9.64
0.21
13.47
0.57
10.55
0.57
C20:1
0.0420
0.0009
0.0669
0.0037
0.0456
0.0020
C24:1
0.0043
0.0012
0.0056
0.0026
0.0085
0.0019
C18:2 n-6
0.9610
0.0459
2.0954
0.1847
0.6877
0.0491
C18:3 n-3
0.1287
0.0021
0.1777
0.0063
0.1242
0.0039
C20:2
0.0010
0.0002
Nd
–
0.0002
0.0002
C20:3 n-6
0.0017
0.0007
0.0010
0.0010
Nd
–
C22:2 n-6
0.0432
0.0019
0.0627
0.0048
0.0092
0.0010
C18:1t
0.1531
0.0049
0.2353
0.0152
0.1411
0.0087
C18:2t
0.0140
0.0007
0.0277
0.0028
0.0067
0.0007
C18:3t
0.0131
0.0013
0.0068
0.0017
0.0124
0.0027
The most abundant fatty acids in table olives are oleic, palmitic, linoleic, and stearic acids. a Standard error estimated from GLM analysis; n, number of determinations.
b
flesh) and their value per USA and Canada serving size (about 15 g e.p.) could always be expressed as 0. There are several indexes for the evaluation of the nutritional value of fat composition. The most commonly used are the PUFA/SFA ratio and the (PUFA⫹MUFA)/(SFA⫹TFA) ratio (Alonso et al., 2002; Serrano et al., 2005). Nutritional
guidelines recommend a PUFA/SFA ratio above 0.4 (Wood et al., 2003). No specific recommendation concerning the second ratio has yet been published, but for keeping plasma and liver cholesterol low, the (PUFA⫹MUFA)/(SFA⫹TFA) ratio has been suggested not to exceed 3 (Chang and Huang, 1999). A significantly higher PUFA/SFA ratio is found in
720
SECTION | II General Aspects and Changes in Food Processing
TABLE 75.4 Average values (⫾ standard error) of saturated (SFA), monounsaturated (MUFA), polyunsaturated (PUFA) and trans (TFA) fat, expressed as fatty acid g 100 g⫺1 edible portion, according to elaboration styles. Green olives (n ⫽ 96)b
Fatty acid
Directly brined olives (n ⫽ 18)
Ripe olives (n ⫽ 20)
Average
Standard errora
Average
Standard error
Average
Standard error
SFA
2.77
0.06
3.86
0.21
2.63
0.16
MUFA
9.87
0.22
13.73
0.57
10.77
0.59
PUFA
1.14
0.05
2.34
0.19
0.82
0.05
TFA
0.1801
0.0058
0.2698
0.0179
0.1602
0.0113
PUFA/SFA
0.41
0.02
0.61
0.04
0.32
0.02
(MUFA⫹PUFA)/(SFA⫹TFA)
3.75
0.04
3.94
0.08
4.19
0.06
In table olives, the main lipid fraction considered in nutritional labeling is monounsaturated, followed by saturated and polyunsaturated fat. The proportion of trans fat (TFA) is very low. a Standard error estimated from GLM analysis; b n, number of determinations.
directly brined olives (mean ⫾ standard error, 0.61 ⫾ 0.04) than in green (0.41 ⫾ 0.02) or ripe olives (0.32 ⫾ 0.02). However, the second index is higher in ripe olives (4.19 ⫾ 0.06) than in directly brined (3.94 ⫾ 0.08) or green olives (3.75 ⫾ 0.04), which are fairly close. Apparently, it appears that the directly brined olive is a slightly healthier product (López et al., 2006).
75.4 STEROLS AND FATTY AND TRITERPENIC ALCOHOLS IN TABLE OLIVES These compounds are in the unsaponifiable matter of the olive oil and in the lipids from table olives. At least parts of the physiological benefits of olive oils are probably determined by the large amount of minor components in the unsaponifiable fraction of olive oil. Usually, the proportion of unsaponifiable matter in olive oil is about 1–2% (Aparicio and Harwood, 2003). Declaration of cholesterol content in food labeling is compulsory. Other plant sterols (phytosterols) have important biological activities to lower or control the levels of cholesterol in blood (Plat and Mensink, 2005). Clinical studies have demonstrated that the dietary intake of plant sterols as part of a normal diet, or as a supplement, may decrease blood cholesterol levels inhibiting its absorption from the small intestine (Richelle et al., 2004). Also, phytosterols have been recognized as a cancer-preventive biologically active substance, although not yet confirmed in epidemiological studies (Normén et al., 2001).
75.4.1 Unsaponifiable Matter The analysis of unsaponifiable components requires a collection via saponification, followed by thin-layer chromatography fractionation of the different classes present in the extract (hydrocarbons, carotenes, tocopherols, linear alcohols, triterpenic alcohols, methyl sterols, sterols and triterpenic dialcohols) (Lercker and Rodriguez-Estrada, 2000). The distribution of the unsaponifiable matter in the lipids from table olives (Figure 75.4) shows that the highest percentage of commercial presentations (about 55%) is in the range of 2–5 g 100 g⫺1 lipid. The mean value is 4.5 g 100 g⫺1 lipid, with a minimum of 0.51 g 100 g⫺1 lipid in seasoned Arbequina turning-color olives, and a maximum of 12.17 g 100 g⫺1 lipid in pitted Carrasqueña ripe olives. The distribution is not normal and shows a marked tail to the right. These proportions are higher than those found in olive oil (Aparicio and Harwood, 2003). Differences among elaboration styles (Figure 75.5) are significant (p ⬍ 0.05). The highest proportions are found in lipids from ripe olives while the lowest levels are in those from directly brined olives. Apparently, the level of unsaponifiable matter may be inversely related to the energy of the treatment applied in each case, which is lower in the last style. However, there is no correlation with lipid content. The proportion of unsaponifiable matter can also be referred to the edible portion (Table 75.5). In this case, ripe olives show the highest contents, especially Carrasqueña ripe olives (2.09 g 100 g⫺1 e.p) while the lowest are in directly brined olives, except for Manzanilla
CHAPTER | 75 Nutrient Profiles of Commercial Table Olives: Fatty Acids, Sterols, and Fatty Alcohols
Percentage observations
35 30 25 20 15 10 5 0
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
Unsaponifiable matter (%lipids) FIGURE 75.4 Distribution of unsaponifiable matter (% lipids) in the main commercial presentations of table olives. The distribution of unsaponifiable matter in table olives has a marked bias towards high values. It contains phytosterols and other minor components of marked biological value.
10.0 Unsaponifiable matter (% lipids)
9.0 8.0
Vertical bars denote 0.95 confidence intervals
7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0
Green
Directly brined
Ripe
Processing styles
FIGURE 75.5 Unsaponifiable matter (% lipid) in commercial presentations of table olives, according to processing styles. The unsaponifiable matter is strongly related to the energy of the treatments used for processing and increases in the order directly brined, green and ripe olives.
(0.58 g 100 g⫺1 e.p.). Green olives have intermediate values (0.50–0.66 g 100 g⫺1 e.p.).
75.4.2 Concentration of Sterols in Table Olives Most of the total sterol contents range from 20 to 30 mg 100 g⫺1 e.p. but some directly brined olives have higher concentrations (Arbequina and Aloreña with 54.8 and 52.6 mg 100 g⫺1 e.p., respectively). Total fatty alcohols are clearly higher in directly brined olives and slightly below 10 mg 100 g⫺1 e.p. in green and ripe olives. The overall average concentration of sterols is 28.7 mg 100 g⫺1 e.p.
721
with a minimum of 17.10 mg 100 g⫺1 e.p. in Cacereña sliced olives, and a maximum of 57.06 mg 100 g⫺1 e.p. in seasoned Arbequina turning-color olives. The major sterols were β-sitosterol, Δ5-avenasterol and campesterol, with overall mean contents of 23.5, 1.5, and 0.9 mg 100 g⫺1 e.p., respectively. Brassicasterol, Δ7-stigmastenol, and Δ7-avenasterol are usually absent from table olives. Mean values of each sterol were significantly (p ⬍ 0.05) higher in samples of directly brined olives compared to the other processing types, with the exception of cholesterol and stigmasterol. In green olives, the mean concentrations of β-sitosterol, campesterol, and campestanol are slightly lower than those reported in the literature by Normén et al. (1999), 34, 1.1, and 0.09 mg 100 g⫺1 e.p., respectively, but the concentration of stigmasterol is higher (0.51 vs. 0.29 mg 100 g⫺1 e.p.). A detailed study of sterols in table olives can be found elsewhere (López-López et al., 2008). Special mention should be devoted to cholesterol due to its compulsory declaration in nutritional labeling. Cholesterol is absent or its content fairly low in olive oil (Brescia et al., 2003). However, in table olives, cholesterol has always been present and, in some commercial presentations, the levels are significant. In general, the stuffing materials from animal or fish origins in some commercial presentations lead to a slight increase in cholesterol concentrations. The cholesterol distribution in table olives is shown in Figure 75.6. In general, its content is below 0.5 mg 100 g⫺1 e.p. and most of the commercial presentations have less than 0.3 mg 100 g⫺1 e.p. with a minimum of 0.08 mg 100 g⫺1 e.p. in Manzanilla olives stuffed with ‘piquillo’ pepper, and a maximum of 4.9 mg 100 g⫺1 e.p. in Manzanilla olives stuffed with marinated anchovy strips. Cholesterol content is also relatively high in Manzanilla olives stuffed with anchovy strips (3.4 mg 100 g⫺1 e.p.) (the most popular commercial presentation in Spain), in Manzanilla olives stuffed with salmon strips (2.6 mg 100 g⫺1 e.p.), in Manzanilla stuffed with ham paste (1.7 mg 100 g⫺1 e.p.), and in Manzanilla olives stuffed with tuna strips (1.1 mg 100 g⫺1 e.p.). However, a relatively low level is observed in more than 90% of cases. The origin of this small amount of cholesterol is unknown and has never been studied but a hypothesis could be that it is formed during the lyses of the microorganisms which grow during the fermentation/storage processes. The contents found are negligible with respect to the nutritional labeling in the USA or Canada because the serving size in both cases is of about 15 g and the cholesterol content for this amount of edible portion can always be declared as 0 in both countries. The most abundant sterol in table olives is β-sitosterol. In olive oil its content, together with five other adjacent sterols, must be above certain limits, expressed as a percentage (Official Diary of the European Community, 1991). Its distribution in table olives is shown in Figure 75.7 and
722
SECTION | II General Aspects and Changes in Food Processing
TABLE 75.5 Weighted average content and standard error of unsaponifiable matter, total sterols and total fatty and triterpenic alcohols in Spanish commercial table olives, according to styles and cultivars. Processing style
Green
Directly brined
Ripe
Cultivar
Unsaponifiable matter (g 100 g⫺1 e.p.)
Total sterols (mg 100 g⫺1 e.p.)
Total fatty alcohols (mg 100 g⫺1 e.p.)
Total triterpenic nb alcohols (mg/100 g e.p.)
Average
Standard Average errora
Standard error
Average
Standard Average error
Standard error
Gordal
0.56
0.03
26.9
0.9
6.7
0.3
2.19
0.12
22
Manzanilla
0.66
0.03
27.5
0.7
9.1
0.3
3.09
0.07
60
Carrasqueña
0.60
0.08
25.8
2.4
8.5
1.1
3.89
0.38
6
Hojiblanca
0.50
0.05
29.2
1.9
9.3
1.2
4.75
0.43
8
Gordal
0.31
0.01
32.8
0.7
14.3
0.4
2.63
0.16
2
Manzanilla
0.58
0.08
34.6
2.5
22.0
1.1
4.30
0.30
6
Hojiblanca
0.24
0.03
27.8
0.5
12.9
0.5
2.29
0.19
2
Arbequina
0.30
0.01
54.8
2.3
36.0
0.3
5.81
0.32
2
Aloreña
0.34
0.02
52.6
1.2
18.0
1.0
6.93
0.31
4
Verdial
0.34
0.03
33.1
4.6
9.2
0.2
11.59
1.08
2
Gordal
0.91
0.06
22.6
0.4
4.6
0.1
2.61
0.19
2
Manzanilla
0.98
0.01
30.1
0.5
8.9
0.7
3.05
0.16
2
Carrasqueña
2.09
0.09
28.0
2.3
7.6
0.6
5.54
1.75
4
Hojiblanca
1.10
0.18
26.0
1.1
6.2
0.3
4.19
1.02
6
Cacereña
0.96
0.08
20.2
0.7
7.6
0.3
3.58
0.25
6
Unsaponifiable matter in table olives is related to the strength of the treatments suffered during processing each style. Average sterol contents ranged from 20–55 mg 100 g⫺1 edible portion, β-sitosterol being the most representative. Standard error estimated from GLM analysis; b n, number of determinations; e.p.: edible portion.
a
ranged from 15 to 35 mg 100 g⫺1 e.p., although some reduced numbers of presentations have values slightly higher. When added with the rest of the adjacent sterols, the percentage is similar to the limits (93%) permitted in olive oil. The presence of β-sitosterol in table olives is important because of the biological anticancer effect of this and similar compounds (Awad and Fink, 2000).
75.4.3 Content of Fatty and Triterpenic Alcohols The mean concentration of total alcohols (fatty and triterpenic alcohols) was 13.3 mg 100 g⫺1 e.p. with a minimum of 6.49 mg 100 g⫺1 e.p. in Gordal olive ‘salads’, and maximum
41.80 mg 100 g⫺1 e.p. in seasoned Arbequina turning-color olives. The most abundant are octacosanol, hexacosanol, and erythrodiol, with global averages of 4.7, 3.5, and 2.6 mg 100 g⫺1 e.p., respectively. Levels of triterpenic dialcohols are higher in olive-pomace oils (extracted from olive pomace with solvents) than in virgin and refined olive oils (CañabateDíaz et al., 2007). The alcohols in table olives, expressed as mg 100 g⫺1 total lipid, are always in higher concentrations than in virgin olive oil but lower than in pomace olive oil. The overall fatty alcohols are always below 10 mg 100 g⫺1 e.p. in green and ripe olives while they reach values from 12.9 to 36 mg 100 g⫺1 e.p. in directly brined olives except in Verdial with 9.2 mg 100 g⫺1 e.p. (Table 75.5). Triterpenic alcohol concentrations in table olives are markedly lower than fatty alcohols. In green and ripe olives,
CHAPTER | 75 Nutrient Profiles of Commercial Table Olives: Fatty Acids, Sterols, and Fatty Alcohols
104% Mean: 0.45; SD= 0.75; N =134 Maximum:4.87; Minimum: 0.08
Percentage observations
90%
●
75% 60%
●
45% 30% ●
15% 0%
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Cholesterol
(mg 100 g−1
edible portion)
FIGURE 75.6 Distribution of cholesterol (mg 100 g⫺1 edible portion) in the main commercial presentations of table olives. Table olives have very low concentrations of cholesterol. Most of the commercial presentations have from 0.0–0.5 mg 100 g⫺1 edible portion.
Percentage observations
50 40 30 20 10
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
β-sitosterol (mg 100 g−1 edible portion) FIGURE 75.7 Distribution of β-sitosterol (mg 100 g⫺1 edible portion) in the main commercial presentations of table olives. The most abundant phytosterol in table olives is β-sitosterol, which content ranges from 15 to 30 mg 100 g⫺1 edible portion.
the level is always equal to or below 5.54 mg 100 g⫺1 e.p. but is higher in many commercial presentations of directly brined olives, with the highest content found in the Verdial cultivar (11.59 g 100 g⫺1 e.p.) (Table 75.5). Detailed information on the fatty and triterpenic alcohol levels in table olives can be found elsewhere (López-López et al., 2008).
SUMMARY POINTS ●
●
polyunsaturated, 0.5–3.9 g 100 g⫺1 e.p.; and trans, 0.1–0.4 g 100 g⫺1 e.p. The averages of the main fatty acids are: oleic, 10.0 g 100 g⫺1 e.p; palmitic, 2.5 g 100 g⫺1 e.p; linoleic, 1.0 g 100 g⫺1 e.p; and stearic, 0.5 g 100 g⫺1 e.p. The average sterol content is 28.7 mg 100 g⫺1 e.p, with β-sitosterol (23.5 mg 100 g⫺1 e.p), Δ5-avenasterol (1.5 mg 100 g⫺1 e.p) and campesterol (0.9 mg 100 g⫺1 e.p) the most abundant among them. The overall mean content of cholesterol is 0.5 mg 100 g⫺1 e.p. For a service size of 15 g⫺1 e.p., its content can always be expressed as 0.0 mg service⫺1. The overall mean contents of fatty alcohols in greater concentrations are: octacosanol (4.7 mg 100 g⫺1 e.p.) and hexacosanol (3.5 mg 100 g⫺1 e.p.). The overall mean content of erythrodiol, the most abundant triterpenic alcohol, is 2.6 mg 100 g⫺1 e.p.
REFERENCES
60
0
●
723
The overall content of lipids in table olives is approximately 16.2 g 100 g⫺1 e.p. Their distribution in the diverse nutritional fractions is: saturated, 1.5–6.0 g 100 g⫺1 e.p; monounsaturated, 5.7–19.4 g 100 g⫺1 e.p;
Alonso, L., Fraga, M.J., Juárez, M., Carmona, P., 2002. Fatty acid composition of Spanish shortenings with special emphasis on trans unsaturation content as determined by Fourier transform infrared spectroscopy and gas chromatography. J. Am. Oil Chem. Soc. 79, 1–6. Aparicio, R., Harwood, J., 2003. Manual del Aceite de Oliva. AVM Ediciones y Mundi Prensa, Madrid. Awad, A.B., Fink, C.S., 2000. Phytosterols as anticancer dietary components: evidence and mechanism of action. J. Nutr. 130, 2127–2130. Balatsouras, G., 1980. Nutritive and nutritive value of the Greek table olives. The biological value of table olives. II World Congress on the biological value of olive oil. Chania (Crete), Greece. Borzillo, A., Iannotta, N., Uccella, N., 2000. Oinotria table olives: quality evaluation during ripening and processing by biomolecular components. Eur. Food Res. Technol. 212, 113–121. Brescia, M.A., Alviti, G., Liuzzi, V., Sacco, A., 2003. Chemometric classification of olive cultivars based on compositional data of oils. J. Am. Oil Chem. 80, 945–950. Cañabate-Díaz, B., Segura Carretero, A., Fernández-Gutiérrez, A., Belmonte Vega, A., Garrido Frenich, A., Martínez Vidal, J.L., Duran Martos, J., 2007. Separation and determination of sterols in olive oil by HPLC-MS. Food Chem. 102, 593–598. Chang, N.W., Huang, P.C., 1999. Comparative effects of polyunsaturatedto saturated fatty acid ratio versus polyunsaturated- and monounsaturated fatty acids to saturated fatty acid ratio on lipid metabolism in rats. Atherosclerosis 142, 185–191. Fernández Díez, M.J., de Castro Ramos, R., Garrido Fernández, A., González Cancho, F., González Pellissó, F., Nosti Vega, M., Heredia Moreno, A., Mínguez Mosquera, M.I., Rejano Navarro, L., Durán Quintana, M.C., Sánchez Roldán, F., García García, P., de Castro Gómez-Millán, A., 1985. Biotecnología de la Aceituna de Mesa. CSIC (Ed), Madrid. Garrido Fernández, A., Fernández Díez, M.J., Adams, M.R., 1997. Table Olives. Production and Processing. Chapman & Hall, London, UK. Hermoso, M., Uceda, M., Frias, L., Beltrán, G., 2001. Maturation. In: Barranco, D., Fernández-Escobar, R., Rallo, L. (Eds.), Cultivo del Olivo. Junta de Andalucía-Ediciones Mundi-Prensa, Madrid-Sevilla, pp. 153–169.
724
SECTION | II General Aspects and Changes in Food Processing
IOOC (International Olive Oil Council), 2004. International Trade Standards Applying to Table Olives. Document COI/NC nº 1. December 2004. Madrid, Spain. IOOC (International Olive Oil Council), 2007. Key figures on the world market for table olives. 95th session of the IOOC. Madrid, Spain. Lercker, G., Rodriguez-Estrada, M.T., 2000. Chromatographic analysis of unsaponifiable compounds of olive oils and fat-containing foods. J. Chromatogr. A 881, 105–129. López, A., Montaño, A., García, P., Garrido, A., 2006. Fatty acid profile of table olives and its multivariate characterization using unsupervised (PCA) and supervised (DA) chemometrics. J. Agric. Food Chem. 54, 6747–6753. López-López, A., Montaño, A., Ruiz-Méndez, M.V., Garrido-Fernández, A., 2008. Sterols, fatty alcohols, and triterpenic alcohols in commercial table olives. J. Am. Oil Chem. Soc. 85, 253–262. Normén, A.L., Brants, H.A., Voorrips, L.E., Andersson, H.A., Van Den Brandt, P.A., Goldbohm, R.A., 2001. Plant sterol intakes and colorectal cancer risk in the Netherlands Cohort Study on Diet and Cancer. Am. J. Clin. Nutr. 74, 141–148. Normén, L., Jonson, M., Andersson, H., van Gameren, Y., Dutta, P., 1999. Plant sterols in vegetables and fruits commonly consumed in Sweden. Eur. J. Nutr. 38, 84–89. Nosti Vega, M., de Castro Ramos, R., 1985. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. VII Aceitunas negras oxidadas. Grasas Aceites 36, 203–206. Nosti Vega, M., Vázquez Ladrón, R., de Castro Ramos, R., 1979. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. II Aceitunas verdes en salmuera. Grasas Aceites 30, 93–100. Official Diary of the European Community Reglament (CEE) Nº 2458/91 of the meeting of 11 of June of 1991 relative to the characteristics of olive oils and its methods of analysis, and its modifications in the Reglament 183/93 (1991).
Plat, J., Mensink, R.P., 2005. Plant stanols and sterol esters in the control of blood cholesterol levels: mechanism and safety aspects. Am. J. Cardiol. 96 (Issue 1, supplement 1), 15–22. Richelle, M., Enslen, M., Hager, C., Groux, M., Tavazzi, I., Godin, J.P., Berger, A., Métairon, S., Quaile, S., Piguet-Welsch, C., Sagalowicz, L., Green, H., Fay, L.B., 2004. Both free and esterified plant sterols reduce cholesterol absorption and the bioavailability of ß-carotene and α-tocopherol in normocholesterolemic humans. Am. J. Clin. Nutr. 80, 171–177. Scherz, H., Senser, F., 2002. Souci-Fachmann-Kraut Food Composition and Nutrition Tables, 6th edition. Medpharm Scienfic Publishers, Stuttgart, Germany. Serrano, A., Cofrades, S., Ruiz-Capillas, C., Olmedilla-Alonso, B., Herrero-Barbudo, C., Jiménez-Colmenero, F., 2005. Nutritional profile of restructured beef steak with added walnuts. Meat Sci. 70, 647–654. US Department of Agriculture, 2007. USDA National Nutrient Database for Standard Reference, release 20. Agriculture Research Service. Nutrient Data Laboratory Home Page: http://www.ars.usda. gov/ba/bhnrc/ndl. Ünal, K., Nergiz, C., 2003. The effect of table olive preparing methods and storage on the composition and nutritive value of olives. Grasas Aceites 54, 71–76. Vamvoukas, D., Stefanoudakis-Katzourakis, E., Loupasakis-Androulakis, M., Kiritsakis, A., 1980. Results from the chemical analysis and determinations on the main cultivars and styles of Greek table olives. II World Congress on the biological value of olive oil. Chania (Crete), Greece. Vázquez Ladrón, R., de Castro Ramos, R., Nosti Vega, M., 1979. Composición y valor nutritivo de algunas variedades españolas de aceitunas de mesa. III. Aceitunas verdes aderezadas envasadas. Grasas Aceites 30, 221–226. Wood, J.D., Richardson, R.I., Nute, G.R., Fisher, A.V., Campo, M.M., Kasapidou, E., Sheard, P.R., Enser, M., 2003. Effects of fatty acids on meat quality: a review. Meat Sci. 66, 21–32.
Chapter 76
Nutritional Properties of Extra Virgin Olive Oils from the Emilia-Romagna Region: Profiles of Phenols, Vitamins and Fatty Acids Annalisa Rotondi1 and Chiara Lapucci2,3 1
Institute of Biometeorology, National Research Council, Bologna, Italy Institute of Biometeorology, National Research Council, Florence, Italy 3 LaMMa (Laboratory of Monitoring and Environmental Modelling for the sustainable development), Sesto Fiorentino (Florence), Italy
2
76.1 INTRODUCTION Olives in Emilia-Romagna are grown mainly in the three provinces located in the south-eastern part of the Romagna region: Rimini, Forlì-Cesena and Ravenna. Climatic conditions in the olive orchards in the hill areas are made milder by the Adriatic Sea and chains of gullies, characteristic of the Ravenna province. Because of these microclimatic characteristics as well as the presence of some autochthonous cultivars found only in Emilia-Romagna, extra virgin olive oils (EVOO) produced in the region are of a high standard. Emilia-Romagna boasts two PDO labels for oil production: Brisighello and Colline di Romagna (Rotondi et al., 2003). In the region there are five widely grown cultivars: Correggiolo, Moraiolo, Leccino, Nostrana di Brisighella and Ghiacciolo. As Correggiolo, Moraiolo and Leccino are also found in other Italian regions, they are ecotypes of Emilia-Romagna, with a genotype similar to that of the same cultivars found in the other regions, they adapted with time to the local microclimate of Emilia-Romagna. The Nostrana di Brisighella and Ghiacciolo cultivars, identified as autochthonous, only found in the Emilia-Romagna region, are characterized by a genotype which differs greatly from those of other Italian cultivars. In this region there are other autochthonous cultivars, which are only found in small areas; their fruits also produce high-quality EVOO: these are Colombina, Selvatico, Rossina, Capolga and Orfana. Emilia-Romagna, along with Apulia, Abruzzo and Tuscany, is one of the four Italian regions where national Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
genetic and sanitary certification processes have started. This process is a definite point of strength for the region, which is now able to benefit from a very solid foundation for a maximum production qualification (Rotondi et al., 2006).
76.2 METHODOLOGICAL CONSIDERATIONS The nutritional properties of EVOO produced in the Emilia-Romagna region have been determined in relation to many principal factors such as the genetic matrix; EVOO nutritional properties will, moreover, be related to the quality of raw materials, particularly considering the influence of both the olive ripening process and the olive preservation time before crushing.
76.2.1 Chemical and Sensory Analyses of Extra Virgin Olive Oil In order to obtain information about the nutritional properties of the EVOO produced in Emilia-Romagna, it is important to take into consideration the data regarding fatty acid composition, total phenols and total ortho-diphenols, single minor polar compounds and their antioxidant activity, vitamins and pigments such as tocopherols, lutein and β-carotene. As there is a strong correlation (Andrews et al., 2003; Gutièrrez et al., 2003) between total minor polar compounds and bitter and pungent sensory characteristics in EVOO, the organoleptic characteristics have also been described.
725
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
726
SECTION | II General Aspects and Changes in Food Processing
Fatty acid composition was determined by GC as fatty methyl esters according to the methods stated in Regulation n. 1513/2001 of the EU. The OSI expressed in hours was evaluated by the approach of accelerated oxidation in a lipid model system and shows the stability of olive oil (Jebe et al., 1999). Total phenolic and o-diphenolic contents were determined employing a spectrophotometric method using caffeic or gallic acid as internal standard (Pirisi et al., 2001). Quantitative analyses of tocopherols, lutein and β-carotene were carried out using HPLC-DAD following methodologies reported in Rotondi et al. (2004). Individual phenols were analyzed by HPLC-DAD, using retention time and UV-visible detection, and the quantitative evaluation was performed through the use of authentic standards (Romani et al., 2001). Sensory analysis was performed by a fully trained analytical taste panel recognized by the IOOC. A panel test was established using a standard profile sheet (IOOC/T20) modified by IBIMET-CNR in order to obtain a complete description of the organoleptic properties of the oils sampled (Cerretani et al., 2004).
76.2.2 Processing of Monocultivar Olive Productions Many factors can affect the chemical and organoleptic characteristics of EVOO, and the method of crushing is one of the most significant. Therefore a mill able to process small quantities of olives was chosen (TEM Compact 50, Florence, Italy), and the many technological variables of the transformation process have been thus standardized (Cerretani et al., 2005).
76.3 GENETIC MATRIX AND NUTRITIONAL PROPERTIES OF MONOVARIETAL OLIVE OILS The genetic matrix, the cultivar, is very important to the nutritional properties of EVOO. Quality and quantity of minor polar compounds depend on many factors, such as the method of crushing, fruit conditions, environment and cultural techniques: the most important factor, even in this case, is the cultivar (Gimeno et al., 2002; Tura et al., 2004). Minor polar compounds include many different chemical classes, such as aliphatic and triterpenic alcohols, sterols, hydrocarbons, volatile compounds, carotenoids and phenols. Some of those substances are defined as cultivar-dependent because of their strong dependence on genetic matrix (Brenes et al., 2002). Some of them, such as carotenes, tocopherols and hydrophilic phenol compounds, possess antioxidant properties: this is why they are related to the health properties of extra virgin olive oil, as they have a role in the prevention of cardiovascular diseases and some forms of cancer (Covas et al., 2006).
It is important to underline that, while tocopherols and carotenoids may also be found in other vegetable or animal fats, the most important hydrophilic phenolic classes are only found in EVOO. These compounds are formed from the phenols in the fruits during mechanical extraction of the oil. The most important phenolic classes found in the olive are: the phenyl acids, the phenyl-ethanols, flavonoids and secoiridoids. The 3,4-DHPEA and the p-HPEA are phenylalcohols mostly present in the fruit. There are two classes of flavonoids: glucoside such as luteolin-7-glucoside and rutin and the others such as anthocyans and cyanidin. Secoiridoids, such as oleuripein, demethyloleuropein, ligustroside and verbascoside are the most concentrated phenolic compounds in the fruit (Morellò et al., 2004). It is important to emphasize that hydrophilic phenols are strongly connected with the organoleptic quality of EVOO, and are responsible for pungent and bitter notes. There is also a connection between pungent and bitter minor polar compounds and their biological activity. In fact the compounds involved in the prevention of cardiovascular disease and cancer are those which give pungent and bitter notes. The specific molecules involved are oleuropein derivatives, demethyloleuropein and ligustroside, and 3,4 DHPEA and p-HPEA (Gutièrrez et al., 1989). In the clonal selection process of Emilia-Romagna olive tree cultivars, a strongly selective factor has been a high content of minor polar compounds in the EVOO, in order to obtain oils of a high nutritional standard. In this process fatty acid composition in monocultivar EVOO has also been considered, evaluating in particular a high oleic acid content, as it is an important monounsaturated fatty acid involved in the prevention of cardiovascular and chronic degenerative diseases (Stark and Madar, 2002). The nutritional and organoleptic properties of monocultivar EVOO of Emilia-Romagna are described below: in order to minimize the effect of annual variations on the chemical and sensory quality of extra virgin olive oil, experimentation was carried out over 3 years. Data refer to mean values, with the total phenol amounts expressed in caffeic acid (ppm), and as regards fatty acid content only oleic acid data are reported, due to its nutritional properties. ●
●
The EVOO obtained from the Capolga cultivar shows 70–73% oleic acid content. TP content is of 400 and 500 ppm, and α-tocopherol levels range between 120 and 150 mg kg⫺1 of oil. Stability expressed as oil resistance to accelerated oxidation is 40–43 hours. The sensory profile is characterized by intense olive fruity, high bitter and pungent notes, and grass, artichoke, thistle and bitter almond flavor. The EVOO obtained from the Colombina cultivar shows a high oleic acid content, 77–79%. Content of natural antioxidant compounds is in the medium range: 184 ppm of TP and 161 mg kg⫺1 of oil for α-tocopherol. Stability to oxidation is 20–22 hours. The sensory
CHAPTER | 76 Nutritional Properties of Extra Virgin Olive Oils from the Emilia-Romagna Region
●
●
●
●
profile is characterized by an olive fruit aroma of medium intensity, a touch of grass, slightly bitter and a strong almond flavor. The single phenol composition shows a high amount – 161.76 mgL⫺1 (unpublished data) – of lignans, known as powerful antioxidant compounds. The Correggiolo is the most widespread cultivar in Emilia-Romagna: the varietal composition of a PDO Colline di Romagna label EVOO must contain over 60% Correggiolo. Oleic acid content is between 73 and 76%. TP values range between 200 and 250 ppm, α-tocopherol ranges between 120 and 150 mg kg⫺1 of oil, lutein content is 1.03 mg kg⫺1 of oil, while β-carotene is 0.67 mg kg⫺1 of oil. Oxidative stability index is 20–22 hours. The sensory profile is characterized by average intensity of olive fruit aroma, slight intensity of grassy, bitter, pungent notes and a strong almond flavor. The single phenol profile is characterized by high levels of lignans (124.16 mg L⫺1), elenolic acid (69.99 mg L⫺1) and derivatives (87.49 mg L⫺1) (unpublished data). The Ghiacciolo cultivar is exclusively used to obtain the extra virgin olive oil trademark ‘Nobil Drupa’. The oleic acid content ranges between 75 and 78%. Antioxidant compound levels are very high: total phenols range from 400 to 500 ppm, α-tocopherol ranges between 170 and 220 mg kg⫺1 of oil, lutein is 2.10 mg kg⫺1 of oil and β-carotene is 1.42 mg kg⫺1 of oil. Because of the high level of antioxidant compounds, this oil shows a high OSI, ranging between 35 and 40 hours. The sensory profile is peculiar, and it is characterized by a bitter note and strong pungent hints. Olive fruit aroma is intense, with artichoke and green tomato notes. The single phenol composition shows a very high level of secoiridoid derivatives (77.13 mg L⫺1, unpublished data), which is in accordance with the high stability to oxidization, as antioxidant activity is high in secoiridoid derivatives. Very high levels of elenolic acid (76.79 mg L⫺1) and derivatives (85.54 mg L⫺1) were also found (unpublished data). The EVOO obtained from the Leccino cultivar shows an oleic acid content of 72–77%. TP content ranges between 120 and 200 mg kg⫺1 of oil, luteolin is 0.73 mg kg⫺1 of oil and β-carotene is 0.38 mg kg⫺1 of oil. Stability to oxidation is 16–20 hours. The sensory profile is characterized by a smooth olive fruit aroma, predominantly sweet, and a strong almond flavor. The Nostrana di Brisighella cultivar is known for its use in the PDO ‘Brisighello’ label, which requires at least 90% of this cultivar. Oleic acid content ranges between 75 and 80%. This oil shows a high amount of antioxidant compounds: TP are in the 300–400 ppm range, α-tocopherol ranges from 140 to 180 mg kg⫺1 of oil, lutein is 1.65 mg kg⫺1 of oil and β-carotene is 0.74 mg kg⫺1 of oil. The high phenol and tocopherol amounts are likely to play a synergic antioxidant role,
●
●
●
727
with an OSI of 30–39 hours. The sensory profile shows a good balance of bitter and pungent, an intense green olive fruit flavor with grass notes, together with artichoke and green tomato aromas. Oleic acid content in EVOO obtained from the Rossina cultivar is in the 76–78% range. TP content ranges between 350 and 430 ppm, α-tocopherol ranges from 150 to 160 mg kg⫺1 of oil. OSI varies between 32 and 46 hours. The sensory profile is characterized by an average olive fruit flavor, green, bitter and pungent aromas and a strong almond fragrance. The EVOO obtained from the Selvatico cultivar has an oleic acid content in the 78–80% range. TP content ranges between 250 and 300 ppm. Levels of lutein, 3.09 mg kg⫺1 of oil, β-carotene, 2.06 mg kg⫺1 of oil, and α-tocopherol, ranging between 170 and 200 mg kg⫺1 of oil, are high: the nutritional value of this oil is thus very high. The stability to oxidation is 34 hours. The sensory profile is quite peculiar: a bitter note peculiarity and pungent notes, intense olive fruit flavor of grass, such as artichoke and mint, and bitter almond aroma. The single phenol composition shows a quite high content of lignans (61.86 mg L⫺1) and elenolic acid (43.60 mg L⫺1) (unpublished data). The Orfana cultivar is known for production of extra virgin olive oil with the ‘Orfanello’ trademark. Oleic acid content ranges between 73 and 76%. TP content ranges between 200 and 270 ppm, and α-tocopherol levels range between 110 and 140 mg kg⫺1 of oil. Stability to oxidation is 20 to 22 hours. The sensory profile is characterized by a light green olive fruit flavor, predominantly sweet with field grass, fresh peas, artichoke and thistle hints.
As there are so many cultivars in Emilia-Romagna, some extra virgin olive oil boasts PDO labels (cvv. Nostrana di Brisighella and Correggiolo) and others are registered trademarks (cvv. Ghiacciolo and Orfana). A deep knowledge of monocultivar nutritional and organoleptic properties, together with the genetic and sanitary certification of the breeding ground plants, allows the producer to ‘construct’ oils with particular chemical and sensory features at the time of olive planting by choosing specific cultivars. Cultivar clonal selection, together with the particular microclimatic characteristics of Northern Italy, turned out to be a valuable link in obtaining EVOO with high levels of oleic acid and antioxidant compounds. As previously mentioned, the total phenol content and the bitter and pungent attributes are positively correlated; in fact, the oils examined are notable for the high bitter and pungent attributes in sensory profiles. Nowadays many consumers are aware that in extra virgin olive oil, strong bitter and pungent attributes are markers of high phenol contents, and as a consequence, of a higher nutritional value. The average consumer is not used to bitter and
728
SECTION | II General Aspects and Changes in Food Processing
pungent EVOO, whether used for cooking or for dressing (Bendini et al., 2007). The monovarietal oils obtained from these cultivars may be consumed separately, or be used in blends giving sweeter oils, such as the ones obtained from Leccino, offering a good balance between bitter and sweet, while improving the nutritional value.
76.4 INFLUENCE OF OLIVE RIPENING STAGE ON EXTRA VIRGIN OLIVE OIL NUTRITIONAL PROPERTIES There is increasing interest in the phenols found in olive oil both because of their intrinsic biological properties as well as their contribution to the flavor and shelf life of the finished product. Some of the most representative phenolic compounds in olive oils can be grouped in three classes: the oleuropein and ligstroside aglycons and their derivatives, the simple phenol derivatives from phenylethyl alcohol, cinnamic and benzoic acids, and other phenolic compounds recently identified as lignans. Studies conducted on the change of the phenolic substances have indicated that during olive ripening, the concentration of phenols progressively increases to a maximum level at the ‘half pigmentation’ stage, then decreases sharply as ripening progresses. An appropriate index of fruit ripening must be established specifically for each individual olive cultivar (Rotondi et al., 2004). The olive ripening stage, as previously mentioned, plays a crucial role in the oil extraction process, influencing its chemical and sensory qualities. The setting up of a quick and easy-to-perform method to determine the olive ripening stage is helpful for choosing the best harvesting period, and also for the following phases of oil transformation and preservation. In Emilia-Romagna the Jaèn index (Uceda and Hermoso, 1998) was used on the different cultivars during the different cultivation environments. The olive ripening trend is deeply affected by the genetic matrix, and at the same time by seasonality. The study of the optimal olive ripening stage for production of extra virgin olive oil with high nutritional value allowed for the maximal exploitation of each cultivar’s potential. From multi-year studies analyzing the nutritional qualities of oils produced from olives harvested at different RI, a precise ripening range for each cultivar was identified: within those ranges high chemical and sensory qualities are guaranteed. The cultivar Nostrana di Brisighella is reported here as an example.
76.4.1 Case Study: cv. Nostrana di Brisighella Olives of the Nostrana di Brisighella cv. grown in EmiliaRomagna were picked at four different RI and immediately
processed in the low-scale mill. The polar extracts of oil samples were submitted to spectrophotometric analysis of TP and o-diphenols, and to HPLC-DAD/MS determination of their quali-quantitative profile. In order to attain a complete description of oil samples, fatty acid composition and organoleptic properties were also determined. The TP content decreased significantly during olive ripening, reducing its content to 47.4% at stage four of ripening. The reduction in TP content during ripening was also demonstrated by a negative correlation (r2 ⫽ ⫺0.88) between TP and RI (Tables 76.1, 76.2). Nostrana di Brisighella olive oils showed high initial levels of o-diphenols which, at stage three of olive ripening, saw a significant decrease of 60% compared to the initial content observed at the first stage of ripening. The SID levels present in the oils sampled decreased as ripening progressed and moreover, the trend in SIDs showed a high positive correlation with the OSI time values (r2 ⫽ 0.95) while it showed an inverse correlation with RI (r2 ⫽ ⫺0.92) (Tables 76.1, 76.2). The Ls levels did not show significant differences during the ripening stage. Regarding the amount of SPs, the highest value was found on samples taken at the second harvest time. Nostrana di Brisighella oils obtained from olives harvested at the first ripeness index, corresponding to an RII ⫽ 2.3, exhibited the highest stability of any of the oils tested in this study, with an OSI value of 47.09 hours. This value underwent a gradual decrease in subsequent ripeness stages, where we determined a 75% decrease at the RIIV, the highest level of olive pigmentation (RI ⫽ 5.11). Even at this high RI, the corresponding OSI ⫽ 35.30 h was still very high when compared to the average OSI values of commercial extra virgin olive oil. Although the ripening process of Nostrana di Brisighella olives is slow and gradual, it was important to describe the marked decrease of TP content, and consequently the decrease of OSI, as ripening progressed. The fatty acid composition of olive oil is an important parameter in the length of shelf life, which is quantitatively affected by two main factors: the olive variety used in the production of the oil and the ripening stage at which the olives are harvested (Aparicio and Luna, 2002). Changes observed from first harvest to the last harvest in the oleic/ linoleic acid ratio show a decreasing trend during ripening, which was also confirmed by a high inverse correlation (r2 ⫽ ⫺0.99) between this ratio and the RI (Table 76.2). It is clear that the oil produced from olives at the initial ripeness stage (RII), thanks to the higher absolute level of phenolic compounds and the oleic acid/linoleic acid ratio, ensures the best quality from the oxidative stability point of view. This was also demonstrated by the high correlations (r2 ⫽ 0.98 and r2 ⫽ 0.89) observed between the OSI and TP levels and C18:1/C18:2 ratios, respectively (Table 76.2). Sensory QDA and triangular tests were performed to establish the influence of olive ripening degree on the
729
CHAPTER | 76 Nutritional Properties of Extra Virgin Olive Oils from the Emilia-Romagna Region
TABLE 76.1 Chemical parameters determined in oils of cv. Nostrana di Brisighella extracted from olives at different ripeness indices. I
II
III
IV
harvest date
Oct 22, 2002
Nov 7, 2002
Nov 19, 2002
Dec 3, 2002
ripeness index
2.38
4.21
4.86
5.11
free acidity (%)
0.19 b
0.26 a
0.26 a
0.27 a
POV (mequiv of O2/kg)
6.03 b
7.39 ab
7.94 a
6.22 b
K232
1.33 a
1.60 a
1.39 a
1.54 a
K270
0.08 b
0.11 a
0.09 b
0.09 b
ΔK
⫺0.004 a
⫺0.003 b
⫺0.003 b
⫺0.003 b
TP (mg of gallic acid/kg)
441.43 a
379.51b
277.43 c
209.57 d
o-diphenols (mg of gallic acid/kg)
212.19 a
228.06 a
153.50 b
127.47 c
SPs (area in mAU)
735.96 c
1682.33 a
1212.65 b
1260.19 b
SIDs (area in mAU)
7796.28 a
5435.00 b
4697.80 b
2778.97 c
Ls (area in mAU)
2512.63 a
2501.90 a
2369.92 a
2579.93 a
ARP
4.02
3.99
2.44
2.05
OSI time (h)
47.09 a
43.10 b
38.43 c
35.30 d
C18:1/C18:2
19.66 a
14.54 b
13.45 c
13.33 c
I-IV different harvest dates; POV peroxide value; K232, K270, and ΔK UV-spectrophotometric indices; TP total phenols; SP simple phenols; SID secoiridoid derivates; Ls lignan derivates; ARP antiradical power; C18:1/C18:2 ratio to oleic acid to linoleic acid; OSI oxidative stability index. Significant differences in the same row are shown by different letters (a–d). Reprinted from Rotondi et al. J. Agric. Food Chem. 2004; 52: 3649–3654, with permission.
resulting oil’s organoleptic properties. QDA indicated a clear decreasing trend of the positive olive oil descriptors as olives ripened. The decrease of bitterness and pungency is also related to the reduction in TP and o-diphenols levels (Figure 76.1). In the present study, we observed a weak positive correlation between the SID content and the bitter (r2 ⫽ 0.57) and pungent (r2 ⫽ 0.65) sensory attributes. It is important to underline that the results of the triangular test confirmed the statistical results of sensory QDA: in fact, when compared with the triangular test results, oils obtained from the first and fourth olive ripening stage, 100% of the assessors correctly identified all of the samples provided during the panel tests. Data from the present study clearly emphasize the importance of identifying the optimal RI for Nostrana di Brisighella olives. In fact, the earliest harvest date, corresponding to an RII of 2.38, gave the best results in terms of all the parameters considered. In oil extracted from olives picked 15 days later at RIII of 4.21, a noticeable increase in
linoleic acid levels, a decrease in TP content, and a lower resistance to forced oxidation were observed. Also, the marked decrease in some positive attributes such as bitterness, pungency, grassiness and green-leaf observed in oils obtained from olives with RIs above 3.5 confirmed the overall deterioration in the oils’ sensory profile. In conclusion, we suggest that for the production of optimal extra virgin olive oil, Nostrana di Brisighella olives growing in the Emilia-Romagna region should be harvested at an RI between 2.5 and 3.5.
76.5 INFLUENCE OF RAW MATERIAL STATUS ON EXTRA VIRGIN OLIVE OIL NUTRITIONAL PROPERTIES The main purpose of this section was to describe the effects on antioxidant compounds of EVOO of some parameters referring to olive quality. Several representative industrial
730
SECTION | II General Aspects and Changes in Food Processing
TABLE 76.2 Correlation between Ripeness Index of olive fruits and some analytical parameters(second column) and between Oxidative Stability Index and the same analytical parameters (third column) of cv. Nostrana di Brisighella extravirgin olive oils. ripeness index
OSI time (h)
ripeness index
1.00
⫺0.94*
free acidity (%)
0.95*
⫺0.82*
POV (mequiv of O2/kg)
0.45
⫺0.19
K232
0.38
⫺0.26
K270
0.12
0.14
ΔK
0.92*
⫺0.88*
TP (mg of gallic acid/kg)
⫺0.88*
0.98*
o-diphenols (mg of gallic acid/kg)
⫺0.67*
0.85*
SPs (area in mAU)
0.63*
⫺0.38
SIDs (area in mAU)
⫺0.92*
0.95*
Ls (area in mAU)
⫺0.08
0.01
ARP
⫺0.80*
0.95*
OSI time (h)
⫺0.94*
1.00
C18:1/C18:2
⫺0.99*
0.89*
POV peroxide value; K232, K270, and ΔK UV-spectrophotometric indices; TP total phenols; SP simple phenols; SID secoiridoid derivates; Ls lignan derivates; ARP antiradical power; C18:1/C18:2 ratio to oleic acid to linoleic acid; OSI oxidative stability index. Values marked with an asterisk are correlated by Pearson test (p ⬍ 0.05). Reprinted from Rotondi et al. J. Agric. Food Chem. 2004; 52: 3649-3654, with permission.
olive mills in Emilia-Romagna region were analyzed for three crop seasons. Olive and oil production were analyzed in order to gauge correlation with the quality of raw material, such as cultivar ripening index and time of olive storage before their transformation. High quality of olives depends on ideal ripening, as well as short storage times before processing. These factors, accompanied by specific genetic traits of the olive, play a key role in the overall olive oil quality. The separated processing of each cultivar represents 55% of the Emilia-Romagna extra virgin olive oil production, while the remaining 45% is composed of oils produced from mixed varieties. In the latter case it has been difficult to estimate the influence of status of raw materials on the nutritional properties of EVOO, as it is not easy to
attribute an RI to olives belonging to different cultivars. As previously mentioned, the RI, expressed as Jaèn index, is specific to each cultivar. The monocultivar oils were produced from the cultivars Correggiolo, Leccino, Nostrana di Brisighella, Ghiacciolo and Selvatico. The simultaneous effect of the olive ripening stage and olive storage factors on oil nutritional properties was estimated on the monocultivar EVOO. The cultivars Nostrana di Brisighella and Correggiolo are reported here as two different examples of cultivars distinguished by deeply different ripening trends.
76.5.1 Case Study: cv. Correggiolo The Correggiolo variety cultivated in Emilia-Romagna, has a gradual and rather late ripening process, taking the value of 2.5 as the optimal olive ripening degree, expressed as a Jaèn index. As shown in Table 76.3, lutein was affected by the length of olive storage times at both ripening degrees, and its levels significantly increased as olive storage times increased with a correlation coefficient of r2 ⫽ 0.70. The correlation between TP and time of olive storage (r2 ⫽ 0.63) indicated that TP decreased as olive storage times increased. Tocopherols and β-carotene contents were not correlated with olive storage times.
76.5.2 Case Study: cv. Nostrana di Brisighella In the Nostrana di Brisighella variety we considered the Jaèn index of 3.5 as an optimal olive ripening degree. Olives belonging to this variety are characterized by a gradual and very late ripening process. The slow pigmentation of olives, accompanied by the maintenance of firmness of the pulp, facilitates harvest and post-harvest operations. In spite of this, it is important to underline the high correlation (r2 ⫽ 0.90) between TP and post harvest times of olives prior to their processing. Table 76.4 describes the marked decrease of TP when olives were stored for more than 48 hours at both values of Jaèn index. The initial high content of phenolic substances, due to genetic traits of this variety, decreased significantly in olives stored for over 48 hours. The α-tocopherol showed the greatest correlation (r2 ⫽ 0.96); their contents decreased as time of olive storage progressed. Lutein, β-carotene and α⫹γ-tocopherols contents did not show any correlation with different times of olive storage. Several authors analyzed the influence of olive storage times before their processing on the chemical composition of olive oils (Gutièrrez et al., 2002). As the percentage of lutein and total pigment content may be used to distinguish different monovarietal virgin olive oils (Gandui-Rojas et al., 2000), it is important to know how the concentration of these antioxidant substances can alter in relation to
731
CHAPTER | 76 Nutritional Properties of Extra Virgin Olive Oils from the Emilia-Romagna Region
Olfactive intensity 100 90 80 70
Pleasantness
Olive fruity
60 50 40
RI
30
RII
20
RIII
10
RIV
0
Bitterness
Pleasant flavors
Green-leaf
Pungency
FIGURE 76.1 Sensory profiles of Nostrana di Brisighella extra virgin olive oils obtained from olives at different ripening indices. The axes of the figure report the intensity of the different perceived attributes. RI-RIV indicate the four different ripening indices. Reprinted from Rotondi et al. J. Agric. Food Chem. 2004; 52: 3649–3654, with permission.
TABLE 76.3 Influence of olive ripeness index and olive storage times on the content of some antioxidants substances determined in Correggiolo olive oil. Cv. Correggiolo
Lutein (mg/100 mL)
βⴙγ-tocopherols (mg/100 mL)
α-tocopherol (mg/100 mL)
β-carotene (mg/100 mL)
Total phenols (ppm caffeic acid)
Ripeness index
Time of olive storage
R.I. ⬍ 2.5
⬍48 hours ⬎48 hours
0.080 0.141
0.496 0.458
13.030 12.690
0.072 0.112
274.132 162.562
R.I. ⬎ 2.5
⬍48 hours ⬎48 hours
0.085 0.104
0.636 0.739
13.076 14.586
0.056 0.060
221.665 159.978
RI olive ripening index; values reported in the table represent the mean value of three crop seasons and are expressed in mg 100 mL⫺1 of oil. Reprinted from Rotondi et al. Progress in Nutrition 2004; 6: 139–145, with permission.
olive quality. The correlation found between lutein content and olive storage times was observed only in Correggiolo oils. Lutein represents a degradation product of carotenoid pigment; whereas Correggiolo olives are characterized by earlier ripening than that of the Nostrana di Brisighella variety. Correlation between TP contents and olive storage times were observed for both cultivars, confirming the importance of olive processing within 24 hours. Even if olives were harvested at an ideal RI and stored correctly, extending olive storage times to greater than 48 hours saw the degradation of phenolic substances taking place. The α-tocopherol concentration of Nostrana di Brisighella oils
diminished significantly in olives stored for over 48 hours, thus reducing the nutritional quality of the oil.
SUMMARY POINTS ●
●
Eight autochthonous cultivars from Emilia-Romagna are the object of this study. The cultivar is crucial to the nutritional properties of extra virgin olive oil: in fact it is the most important factor affecting quality and quantity of phenols.
732
SECTION | II General Aspects and Changes in Food Processing
TABLE 76.4 Influence of olive ripeness index and olive storage times on the content of some antioxidant substances determined in Nostrana di Brisighella olive oil. Cv. Nostrana di B.
Lutein (mg/100 mL)
βⴙγ-tocopherols (mg/100 mL)
α-tocopherol (mg/100 mL)
β-carotene (mg/100 mL)
Total phenols (ppm caffeic acid)
Ripeness index
Time of olive storage
R.I. ⬍3.5
⬍ 48 hours ⬎ 48 hours
0.167 0.169
0.784 0.732
17.744 15.248
0.084 0.106
303.977 172.516
R.I. ⬎ 3.5
⬍ 48 hours ⬎ 48 hours
0.168 0.158
0.716 0.880
15.658 15.428
0.063 0.056
251.942 192.158
RI olive ripening index; values reported in the table represent the mean value of three crop seasons and are expressed in mg 100 mL⫺1 of oil. Reprinted from Rotondi et al. Progress in Nutrition 2004; 6: 139–145, with permission.
●
●
●
All the oils studied show important nutritional properties. The influence of the olive ripening stage on the nutritional properties of extra virgin olive oil has allowed the identification of a precise ripening range for each cultivar: within those ranges high quality – both chemical and sensory – is guaranteed. Raw material status – cultivar ripening index and olive storage times prior to processing – turned out to be crucial to the nutritional properties of extra virgin olive oil.
REFERENCES Andrews, P., Busch, J.L.HC., De Joode, T., Groenewegen, A., Alexandre, H., 2003. Sensory properties of virgin olive oil polyphenols: identification of deacetoxy-ligstroside aglicon as a key contributor to pungency. J. Agric. Food Chem. 51, 1415–1420. Aparicio, R., Luna, G., 2002. Characterisation of monovarietal virgin olive oils. Eur. J. Lipid Sci. Technol. 104, 614–627. Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gomez-Caravaca, A.M., Segura-Carettero, A., Fernàndez-Gutierrez, A., Lercker, G., 2007. Phenolic molecules in virgin olive oil: a survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules 12, 1679–1719. Brenes, M., Garcia, A., Rios, J.J., Garcia, P., Garrido, A., 2002. Use of acetoxypinoresinol to authenticate Picual olive oils. Int. J. Food Sci. Technol. 37, 615–625. Cerretani, L., Bendini, A., Rotondi, A., Lercker, G., Gallina Toschi, T., 2005. Analytical comparison of monovarietal extravirgin olive oils, obtained by both a continuous industrial plant and a low-scale mill. Eur. J. Lipid Sci. Technol. 107, 93–100. Cerretani, L., Bendini, A., Rotondi, A., Mari, M., Lercker, G., Gallina Toschi, T., 2004. Evaluation of the oxidative stability and organoleptic properties of extra-virgin olive oils in relation to olive ripening degree. Prog. Nutr. 6 (1), 50–56.
Covas, M.I., Ruiz-Gutierrez, V., De la Torre, R., Kafatos, A., LamuelaRaventòs, R.M., Osada, J., Owen, R.W., Visioli, F., 2006. Minor components of olive oil: evidence to date of health benefits in humans. Nutr. Rev. 64, 20–30. Gimeno, E., Castellote, A.I., Lamuela-Raventòs, R.M., De la Torre, M.C., Lòpez-Sabater, M.C., 2002. The effects of harvest and extraction methods on the antioxidant content in virgin olive oil. Food Chem. 78, 207–211. Gandui-Rojas, B.L., Cepero, M.R., Minguez-Mosquera, I., 2000. Use of chlorophyll and carotenoid pigments composition to determine authenticity of virgin olive oil. JAOCS 77, 853–858. Gutièrrez, F., Albi, M.A., Palma, R., Rios, J.J., Olias, J.M., 1989. Bitter taste of virgin olive oil: correlation of sensory evaluation and instrumental HPLC analysis. J. Food Sci. 54, 68–79. Jebe, T.A., Matlock, M.G., Sleeter, R.T., 1999. Collaborative study of the oil stability index analysis. J Am. Oil Chem. Soc. 70, 1055–1061. Morellò, J.R., Vuorela, S., Romero, M.P., Motiva, M.J., Heinonen, M., 2004. Antioxidant activity of olive pulp and olive oil phenolic compounds of the Arbequina cultivar. J. Agric. Food Chem. 53, 2002–2008. Pirisi, F.M., Cabras, P., Falqui Cao, C., Migliorini, M., Mugelli, M., 2001. Phenolic compounds in virgin olive oil. 2. Reappraisal of the extraction, HPLC separation and quantification procedures. J. Agric. Food Chem. 48, 1191–1196. Romani, A., Pinelli, P., Mulinacci, N., Galardi, C., Vincieri, F.F., Liberatore, L., Cichelli, A., 2001. HPLC and HRGC analyses of polyphenols and secoiridoids in olive oil. Chromatographia 53, 279–284. Rotondi, A., Bendini, A., Cerretani, L., Mari, M., Lercker, G., Gallina Toschi, T., 2004. Effect of olive ripening degree on the oxidative stability and organoleptic properties of Nostrana di Brisighella extravirgin olive oil. J. Agric. Food Chem. 52, 3649–3654. Rotondi, A., Bertazza, G., Magli, M., 2004. Effect of olive fruits quality on the natural antioxidant compounds in extravirgin olive oil of Emilia-Romagna region. Prog. Nutr. 6, 139–145. Rotondi, A., Magli, M., Ricciolini, C., Baldoni, L., 2003. Morphological and molecular analyses for the characterization of a group of Italian olive cultivars. Euphytica 132, 129–137.
CHAPTER | 76 Nutritional Properties of Extra Virgin Olive Oils from the Emilia-Romagna Region
Rotondi, A., Rossi, F., Ratti, C., Bianchi, L., Babini, A.R., Rubies Autonell, C., 2006. Genetic and sanitary selection of olive germplasm in Emilia-Romagna region. Olivebioteq- Second International SeminarBiotechnology and quality of olive tree products around the basin. 1, 169–172. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food. Epidemiology and nutritional approaches. Nutr. Rev. 69, 170–176.
733
Tura Jr., D., Prenzler, P.D., Bedgood, D.R., Antonlovich, M., Robards, K., 2004. Varietal and processing effect on the volatile profile of Australian olive oils. Food Chem. 84, 341–349. Uceda, M., Hermoso, M., 1998. La calidad del aceite de oliva. In: Barranco, D., Fernàndez-Escobar, R., Rallo, L., de Andalucía, J. (Eds.), El cultivo del olivo. Mundi-Prensa, Madrid, pp. 547–572.
This page intentionally left blank
Chapter 77
Table Olives: A Carrier for Delivering Probiotic Bacteria to Humans Paola Lavermicocca1, Mauro Rossi2, Francesco Russo3 and Rajaventhan Srirajaskanthan4 1
Institute of Sciences of Food Production, National Research Council, Bari, Italy Institute of Food Sciences, National Research Council, Avellino, Italy 3 Laboratory of Experimental Biochemistry I.R.C.C.S. “Saverio de Bellis”, National Institute of Digestive Diseases, Castellana Grotte (Ba), Italy 4 Centre of Gastroenterology, Royal Free Hospital, London, UK
2
77.1 INTRODUCTION
TABLE 77.1 Key facts of probiotic bacteria.
A deeper understanding that foods can have beneficial health effects beyond their basic nutritional value has recently led to a worldwide increase in consumer interest for functional products, and to an increase in their commercial value. Probiotic products contain selected microbial strains which – if they reach the gastrointestinal (GI) tract in viable form – can improve GI function, and thus, the health of the host (Table 77.1). Both clinical and animal experimental studies have provided evidence of the potential use of probiotics in several GI diseases, such as different forms of diarrhea, inflammatory bowel disease, irritable bowel syndrome as well as lactose intolerance, hypercholesterolemia, and food allergy. In recent years, probiotics have been the focus of considerable interest, due to their immune-modulatory properties and perceived low adverse side effect profile. In particular, probiotics are a very promising instrument for the efficient induction of peripheral tolerance in humans suffering from various antigen-driven immune diseases. While the above beneficial effects have been attributed to probiotic lactic acid bacteria (LAB), perhaps their most controversial and debated action remains that of anticancer activity in GI neoplasms, especially colon cancer. Currently probiotic products marketed for human consumption are mainly incorporated in fermented milk products, such as yoghurt and milk derivatives, or concentrated probiotic preparations available such as capsules, powders or liquids. There is a new shift by the pharmaceutical and commercial food industries to develop new non-dairy foods. The aim of this is to ensure regular consumption of probiotics by using other foods and beverages which are part of a normal diet and may be better tolerated by the population Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
●
Probiotic bacteria are defined as ‘Live microorganisms which when consumed in adequate amounts as part of food confer a health benefit on the host’ (FAO/WHO Report, 2001)
●
The mechanisms by which consumption of lactobacilli and bifidobacteria with probiotic characteristics positively affect human health include the increase of the relative numbers of ‘beneficial bacteria’ of gut microflora
●
Increases in lactobacilli and bifidobacteria may result in acidification of the gut, in improvements to the nutritional status of gut epithelium, and in a decrease in intestinal permeability to toxic molecules
●
Probiotic bacteria may play a key role in strengthening intestinal barrier function, antagonizing pathogens, stimulating a protective immune response and maintaining host health
●
They transiently colonize the intestinal tract; large populations need to be ingested on a daily basis for positive effects to be maintained
●
Therapeutic application of probiotics are demonstrated in host defense and gastroenterology
●
Probiotics are a very promising instrument for treatment of immune diseases.
as a whole. Technology for preparing foods which enables people to consume beneficial microorganisms in vegetable products without affecting perceived food quality is under development. Research teams and agricultural industrial companies are looking at processed vegetables e.g., table
735
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
736
SECTION | II General Aspects and Changes in Food Processing
olives, artichokes, salads, etc. as food environments for solutions that will improve the stability and viability of bacterial populations for innovative probiotic foods (Lavermicocca et al., 2005a; Valerio et al., 2006).
77.2 THERAPEUTIC APPLICATION OF PROBIOTICS IN HOST DEFENSE AND GASTROENTEROLOGY Probiotics are live microorganisms that alter the enteric microflora and can confer beneficial effect on health when consumed in adequate amounts (Table 77.1). Bacteria associated with probiotic activity have frequently been LAB or bifidobacteria, while Escherichia coli strains have also been used. The mechanisms by which probiotics positively affect human health include strengthening intestinal barrier function, modulating immune response and antagonizing pathogens, either by producing antimicrobial compounds or by competing for mucosal binding sites (Sartor, 2005; Jonkers and Stockbrügger, 2007). Probiotics are administered orally and need to pass through the complex physical–chemical environment of the GI tract to reach the target region of the large bowel. Moreover, adhesion to human intestinal mucosa has been considered essential for efficient gut colonization and is used as an important criterion for selecting new probiotic strains (Lavermicocca et al., 2008). Microbiota carry out many important functions for the host, such as transforming (pro)carcinogenic substances, producing vitamins, breaking down bile acids and digesting nutrients (Jonkers and Stockbrügger, 2007). Through the fermentation process of complex carbohydrates not digested in the upper gut, beneficial microflora produce lactic acid, hydrogen, carbon dioxide and the short-chain fatty acids (SCFAs) acetate, propionate and butyrate (Perez Chaia and Oliver, 2003). SCFAs are then rapidly adsorbed and have important effects on gut physiology. The primary effect is energy recovery from undigested components such as dietary fibers, resistant starches and oligosaccharides, at least 20 g of which reach the human gut per day. SCFAs improve the nutritional status of gut epithelium and stimulate salt and water adsorption. The trophic effect of SCFAs, in particular butyrate, is therefore important in maintaining gut integrity and may play a role in preventing the bacterial translocation occurring in several GI diseases (Perez Chaia and Oliver, 2003).
77.3 EFFECTS OF PROBIOTICS ON THE IMMUNE RESPONSE In recent years, probiotics have raised considerable interest due to their immune-stimulating properties. In particular, LAB are reported to have a stimulatory effect on cells of the innate immune system in vitro (Haller et al., 2000),
in animals (Perdigon et al., 1988) and in humans (Schiffrin et al., 1997). These findings highlighted the usefulness of LAB as adjuvants of innate immune responses to increase early defense mechanisms in response to GI infections. Innate immune responses represent not only the first line of defense against pathogens but are also fundamental for the development of adaptive immune responses driven by subpopulations of T cells (known as CD4⫹ Th1 and Th2 cells), which secrete cytokines. Th1-derived cytokines are principally responsible for the cell-mediated immune response, used to eliminate intracellular parasites, while Th2 cytokines mostly favor the generation of humoral responses, characterized by the production of antibodies involved in eradicating extracellular pathogens such as helminth infections. The balance between Th1 and Th2 responses is important in maintaining immune homeostasis, as an exaggerated Th1 or Th2 response is associated with a series of immune-mediated diseases. This balance is driven by other subsets of CD4⫹ cells that produce down-regulatory cytokines, such as Th3 cells and Treg cells secreting transforming growth factor- (TGF-)β and IL-10, respectively. It is noteworthy that the antigen-presenting cells (APCs) of the innate immune system, mainly dendritic cells (DCs) and macrophages, interact with naive CD4⫹ T cells and prime them toward distinct effector (Th1, Th2) or regulatory phenotypes (Th3, Treg) (Figure 77.1). Accordingly, the differential modulatory activity exerted by both pathogenic and probiotic microorganisms on APCs is instrumental in defining the phenotype of the subsequent adaptive response. In fact, orally administered Lactobacillus strains can induce different cytokine profiles in the gut of mice (Maassen et al., 2000). In particular, L. reuteri and L. brevis were found to induce Th1 [interleukin- (IL-)1β, IL-2 and tumor necrosis factor- (TNF-)α] but not Th2 cytokines (IL-10 and IL-4), whereas increased production of IL-10 was associated with L. casei administration. In vitro results with L. acidophilus showed increased production of pro-inflammatory cytokines [IL-1 and TNF-α or interferon- (IFN-)γ] by macrophages and enhanced phagocytosis (Rangavajhyala et al., 1997). Notably, the known tumor-suppressive activity of lactobacilli is believed to be related to the cytotoxic activity of TNF-α and IL-1 produced by stimulated macrophages (Pool-Zobel et al., 1996). The molecular mechanisms involved in the interaction between probiotics and APCs are still under debate. Cell activation is probably mediated by components in the cell wall of lactobacilli. Over the last few years, a series of receptors that recognize bacterial molecules have been identified on APCs, e.g., the Toll-like receptors (TLRs), which are widely expressed on macrophages and other immune cells within the intestine. Upon stimulation, TLRs recruit various protein kinases via several adaptor molecules, leading to the activation of nuclear factor (NFκB), a crucial intracellular signaling molecule, and subsequent induction of cytokine transcription (Figure 77.1). By contrast, another study has shown that the protective effect of VSL#3, a probiotic mixture, is essentially mediated by its own DNA (Rachmilewitz et al., 2004).
737
CHAPTER | 77 Table Olives: A Carrier for Delivering Probiotic Bacteria to Humans
Bacteria
APC
Naive T cell
Effector/regulatory T cell
IL-12 IL-10 TLR
MHC-TCR
Th1 Th2 Th3 Treg
Costimulatory signals
Innate immunity
Adaptive immunity
FIGURE 77.1 Bacteria-mediated innate immunity is critical to adaptive immune responses. Antigen-presenting cells (APCs) are activated through recognition of bacteria by TLR receptors. This activation leads to the production of cytokines and the expression of co-stimulatory molecules on the cell surface. Bacterial antigens will be presented by MHC molecules on APCs to naive T lymphocytes. This is not sufficient to activate T lymphocytes and they need an additional signal from co-stimulatory molecules. Activated T lymphocytes become differentiated to effector or regulatory T lymphocytes by stimulation with cytokines such as IL-12 and IL-10, respectively. IL-: interleukin-; TLR: Toll-like receptor; MHC: major hystompatibility; TCR: T cell receptor.
In this study, interaction with TLR9 was found to be essential in mediating the anti-inflammatory effect of probiotics, whereas live micro-organisms were not required to attenuate experimental colitis.
ACTIVE IMMUNITY (L. reuteri, L. brevis, L. acidophilus)
ACTIVE IMMUNITY
VACCINATION
ALLERGY
(L. rhamnosus GG)
77.4 IMMUNE-THERAPEUTIC STRATEGIES The previously reported works paved the way for potential therapeutic strategies in humans (Figure 77.2). Recently, in vitro studies on human DCs incubated with lactobacilli showed induction of the cytokine IL-12 but not of IL-10. Cytokine IL-12 is mainly produced by DCs and is a critical factor in switching T cells in Th1 responses leading to vigorous immunity against infection and other diseases (Mohamadzadeh et al., 2005). Moreover, it is well known that the Th1-stimulating activity of some Lactobacillus strains is instrumental in inhibiting Th2-driven allergic responses. Notably, the use of probiotic therapy in preventing allergic diseases and reducing the incidence of atopic eczema and cow’s milk allergy has been shown in studies using Lactobacillus rhamnosus GG in neonates (Furrie, 2005). The ability of ingested probiotic strains to modulate both intestinal and systemic immunity also highlighted potential applications in autoimmune diseases, characterized by the lack of natural tolerance to self protein components. Strategies aimed at recovering antigen-specific tolerance have been elaborated in various experimental models of autoimmunity. In non-obese diabetic mice that spontaneously develop a pathology featuring human insulin-dependent diabetes mellitus, administration of Lactobacillus casei strain Shirota significantly reduced the incidence of diabetes. The potential of L. casei strain Shirota has been also underscored in a mouse model of type II collagen (CII)-induced arthritis (Kato et al., 1998). Notably, oral administration of L. casei strain Shirota was effective in reducing both CII-induced antibodies in serum and IFN-γ secretion from splenocytes ex vivo (Kato et al., 1998).
TOLERANCE (L. lactis, L.casei Shirota)
AUTOIMMUNITY
FIGURE 77.2 Probiotic exploitation for modulating immune responses is strain-dependent. On the basis of their intrinsic properties, different strains may potentially be used for enhancing immunity (vaccination) or for inhibiting allergic and autoimmune diseases.
In conclusion the reported data underline probiotics as a very promising instrument for efficient induction of antigenspecific peripheral tolerance in humans suffering from different antigen-driven immune diseases.
77.5 ANTI-CANCER ACTIVITIES OF PROBIOTICS IN COLORECTAL NEOPLASMS Colorectal cancer (CRC) represents an important cause of cancer morbidity and mortality in Western countries. Approximately 70% of CRC is associated with environmental factors, including diet (Jass et al., 2002). In an unbalanced diet, ‘unfriendly’ bacteria allow cancer-causing substances to grow in the colon. For this reason, increasing attention has been paid to particular LAB strains, since these bacteria have been shown to possess antimutagenic and anticarcinogenic properties. Data from experiments, mainly on laboratory animals, indicate that LAB ingestion may reduce the risk of certain types of cancer and inhibits tumor growth. A study has shown that lactobacilli and bifidobacteria strains prevent DNA damage induced by the carcinogen 1,2-dimethylhydrazine in the rat colonic mucosa. It is believed that probiotic potential protection
738
SECTION | II General Aspects and Changes in Food Processing
against CRC could be associated with modifications in cell proliferation and growth (Oberreuther-Moschner et al., 2004). Unfortunately, much is still unknown about the precise mechanisms of antineoplastic action of probiotics, although data suggest that their consumption may decrease cancer risk by different ways (Table 77.2).
TABLE 77.2 The proposed antineoplastic mechanisms of action of probiotics. ●
A reduction in the incidence of chemically induced tumors in rats
●
A reduction in the activity of fecal enzymes (betaglucuronidase, azoreductase, nitroreductase, and 7-alphadehydrogenase) thought to play a role in colon cancer in human and animal subjects
●
Degradation of nitrosamines
●
A weakening of the mutagenic activity of substances tested in the laboratory
●
In vitro prevention of damage to DNA in colonic cell lines
●
In vitro binding of mutagens by cell wall components of probiotic bacteria
●
Enhancement of immune system functioning
Several early or intermediate cancer biomarkers have been proposed for investigating the LAB mechanisms of action: these biomarkers may be biochemical, molecular, or cellular. In this framework, research has emphasized a relationship between polyamine biosynthesis and probiotics during carcinogenesis and tumor growth. Polyamines (putrescine, spermidine, and spermine) are organic polycations with multiple functions in cell proliferation and differentiation. Their functions lie essentially in the regulation of gene expression, both by altering DNA structure and modulating signal transduction pathways (Peng and Jackson, 2000). For optimal functioning of the cell, intracellular polyamine content needs to be strictly controlled, and this occurs at the levels of biosynthesis, catabolism, uptake, and efflux (Figure 77.3). Recently, evidence has been provided that a Lactobacillus brevis strain induces apoptosis of Jurkat cells and it has been hypothesized that their apoptotic death-inducing ability could be associated with polyamine synthesis (Di Marzio et al., 2001). In mice, it has been observed that the administration of Bifidobacterium longum significantly suppresses ornithine decarboxylase activity and the rate of cell proliferation in the azoxymethane induced tumor. These results thus indicate that the modulation of the above-mentioned intermediate biomarkers could be used to monitor colon cancer inhibition by B. longum and to predict colon tumor outcome. The ability of probiotics to affect cell proliferation and polyamine metabolism has further been supported by Intracellular
S-adenosylmethionine
Arginine uptake
Arginase
SAMDC CO2
Extracellular
L-ornithine ODC
Decarboxylated Sadenosylmethionine
AZ
CO2 Putrescine PAO
Spermidine synthase
export N1-acetylspermidine SSAT
Methylthioadenosine
Spermidine PAO
Spermine synthase
SMO
N1-acetylspermine
export
SSAT Spermine
Methylthioadenosine
FIGURE 77.3 The metabolic pathway of polyamines. The figure shows the polyamine biosynthesis in mammals. The process in cells begins with the production of putrescine by the decarboxylation of the amino acid ornithine by ornithine decarboxylase (ODC). Subsequent addition of an aminopropyl group to putrescine, by the decarboxylation of S-adenosylmethionine through the action of S-adenosylmethionine decarboxylase (SAMDC), leads to the synthesis of spermidine and further addition of another aminopropyl group leads to the formation of spermine. AZ: antizyme; PAO: polyamine oxidase. SMO: spermine oxidase; SSAT: spermidine/spermine N1-acetyltransferase.
739
CHAPTER | 77 Table Olives: A Carrier for Delivering Probiotic Bacteria to Humans
another in vivo study. It has been observed that 4 weeks’ administration of high doses of a cocktail of different bacterial strains (VSL#3) reduces polyamine biosynthesis and cell proliferation also in the normal colonic mucosa of rats (Linsalata et al., 2005). However, the link between cell proliferation, polyamines and probiotics could be regulated by different factors: the particular metabolic features of bacterial strains, different survival times in the lumen, period of administration, as well as the proliferative behavior of different segments along the GI tract.
77.6 THERAPEUTIC USE OF PROBIOTICS IN PATIENTS WITH CONSTIPATION Constipation is defined as fewer than three bowel movements per week. There are many causes of constipation: medications, low-fiber diets, abuse of laxatives, hormonal disorders, etc. More than 95% of the time, however, the slowing occurs in the colon: i.e., colonic inertia and pelvic floor dysfunction. Many treatments for constipation have been proposed, including modification of lifestyle, use of dietary fibers, laxatives, biofeedback training, and even surgery. A new therapeutic approach for constipation could be based on modulating intestinal microflora. Colonic microflora influences peristalsis of the colon (Picard et al., 2005). Therefore, imbalance in the colonic microflora has also been suggested to play a role in constipation. LAB strains produce lactic, acetic and other acids, thus lowering pH in the colon. A lower pH enhances peristalsis of the colon and subsequently decreases colonic transit time which is beneficial in the treatment of constipation (Bekkali et al.,
(A)
2007). However, the identification of effective probiotics in constipation still requires appropriate in vitro and in vivo studies, which would also help to identify their modulatory activities on bowel movements.
77.7 BENEFICIAL BACTERIA AND TABLE OLIVES: DEVELOPING A NEW PROBIOTIC FOOD In order to display a therapeutic role, probiotic strains have to be viable during process preparation and to survive passage through the GI tract and reach the intestine in viable form. Cellular stress is elicited by low pH values (⬍4.0) of the stomach. Contact with bile secreted in the small intestine further affects cell integrity by destroying cell membranes. Although survival is a strain-related ability, a concrete improvement can be obtained through food technology, since the performance of probiotic strains is influenced by the protective action of the carrier food. Table olives represent an especially good source of nutrients which can protect microbial strains and sustain their growth by acting as probiotic molecules. The protective action exerted by table olives is demonstrated by the performance of selected lactobacilli – a food isolate of Lactobacillus plantarum and a human isolate of Lactobacillus paracasei – carried by olives during a simulated GI digestion (Figure 77.4). Besides, a protective action can also be performed by the micro-architecture of the olive surface which allows bacteria to survive by adhering to the olive skin and improves their performance in the GI tract (Lavermicocca et al., 2005b; Valerio et al., 2006). This is demonstrated by the electron microscopic observation of the surface of probiotic olives (Figure 77.5).
(B) 12 Survival log10 CFU g−1 or CFU ml−1
Survival log10 CFU g−1 or CFU ml−1
12 10 8 6 4 2
10 8 6 4 2 0
0 0 Gastric digestion
180
180
0
Intestinal digestion
Gastric digestion
0
Time (hours)
180
0
180 Intestinal digestion
Time (hours)
FIGURE 77.4 Survival of L. plantarum ITM21B (A) and L. paracasei IMPC2.1 (B) during simulated gastric (pH 2.0) and intestinal digestion (pH 8.0) in the presence of saline solution (䊉), skimmed milk (䊐), olives (䊊) and artichokes (䊏). (Source: Valerio et al., 2006, with permission). The figure shows that more than 95% of the total bacterial population anchored to olives survive during simulated digestion while most of unprotected cells suspended in saline buffer die (36% survival). Data, expressed as means ⫾ standard errors, are from three independent experiments with two replicates each (n ⫽ 6).
740
SECTION | II General Aspects and Changes in Food Processing
9
9
8
8
7
7 Log CFU ml−1
Log CFU ml−1
FIGURE 77.5 Colonization of Lactobacillus paracasei (A), L. plantarum (B), Bifidobacterium longum (C) and B. bifidum (D) on the olive surface. (Scanning electron microscopy observations by Maria Luisa Callegari). Images disclose the spectacular attachment of bacterial cells, firmly colonizing the vegetable surface; besides probiotic lactobacilli, strains belonging to the bifidobacteria group are able to colonize the olive surface. As reported in the text, strains of this group are widely sought after in probiotic products for their health-promoting properties.
Strain selection among the various potential probiotic bacteria should take colonization ability into account in order to ensure a daily intake that will provide beneficial effects. In fact, since probiotic bacteria are only transient in the intestinal tract and do not become part of the host’s gut microflora, their regular consumption is required for the maintenance of positive effects. Therefore, probiotic strains must be ingested in large quantities and on a daily basis. One billion viable cells is the recommended daily dose to obtain beneficial effects through a probiotic food (FAO/WHO Report, 2001; Maukonen et al., 2006). The high survival rate observed for probiotic strains on olives implies that a portion of olives (10–15 olives corresponding to about 80 g) allows the ingestion of more than one billion live selected L. paracasei or L. plantarum strains. Human feeding studies demonstrated that olives act as a food vector by ensuring vitality of bacterial populations during transfer through the GI tract until their release. The LAB population in the human gut was enriched when the subjects ate a dozen probiotic table olives as part of a normal daily diet for 10–15 days. The selected strains L. paracasei or L. plantarum carried by olives temporarily become resident in the gut and their colonization alters composition of the gut microflora which switches towards beneficial populations (Figure 77.6).
6 5 4 3
6 5 4 3
2
2
1
1
0
0 1
2 Subject
3
1
2
3 Subject
4
5
Subject 3
Subject 2
Before consumption After consumption
Before consumption After consumption
Before consumption
End of consumption
Wash out
L. plantarum or L. paracasei recovery FIGURE 77.6 Lactic populations recovered from fecal samples from healthy volunteers fed for 10 days with 10–15 table olives per day containing more than one billion live cells of L. paracasei (right) and L. plantarum (left). The recovery of each strain was molecularly assessed as previously reported (Lavermicocca et al., 2005b; Lavermicocca, 2006). Note microbial modulation in subjects 3 and 2 fed with olives carrying L. plantarum and L. paracasei strain, respectively.
741
CHAPTER | 77 Table Olives: A Carrier for Delivering Probiotic Bacteria to Humans
The presence of large populations of metabolically active microorganisms during food processing and in the final product is required for the efficacy of probiotic foods. To achieve this, an appropriate formulation of table olives is required to enhance the viability of probiotic populations during processing and storage. Technological efforts have been made to pilot a fermentation process of table olives versus a mild lactic acid fermentation: the selected human isolate L. paracasei was successfully applied in the manufacturing process of debittered green olive cultivar ‘Bella di Cerignola’ in an industrial plant (Figure 77.7). This technology has enabled a previously uncontrolled process to be overcome, thus obtaining a product of high quality which can be variously seasoned (olive oil, spices, etc.), packed in a protective atmosphere and stored at refrigerated temperature. The final product carries one billion live cells of the probiotic strain per portion throughout its shelf life of several months (Figure 77.7). The major microbial groups occurring during table olive processing survive during the process in a similar way to the probiotic strain. The dynamics of the microbial populations belonging to the different groups indicate that neither olives nor brines host enterobacteria and that the behavior of aerobic mesophilic populations reflects the survival performance observed for LAB
population, which was mainly represented by the selected L. paracasei. Yeast populations are present in very low numbers during the whole process (Figure 77.7). Therefore the selected L. paracasei strain can be considered as an example of a probiotic strain that is suitable for processing table olives, since it combines the features of fermentative and probiotic bacteria. Strains of L. paracasei have been isolated from olive fermentations (Van den Berg et al., 1993), and this species has a close taxonomical relationship with other bacterial species, e.g., Lactobacillus casei, involved in the natural fermentation of table olives (Randazzo et al., 2004).
77.8 CONCLUSIONS Olive-based probiotic products are under development in the gastronomy sector, since fruit and vegetable products provide a concrete opportunity to convey probiotic benefits already appreciated by consumers in other market sectors (yoghurt and dairy). Current information regarding benefits derived from good eating habits is motivating consumers to become more aware of their diet without sacrificing the desires of the palate. Italian research institutions in collaboration with
9
10
8
9
8
6
7
5 4
pH
Log CFU gr−1
7
6
3 5
2
4
1 02
9
30
60
90
Days 9 days
90 days
FIGURE 77.7 Behavior of the probiotic Lactobacillus paracasei strain and of the main microbial group during the processing of table olive (cultivar ‘Bella di Cerignola’). Symbols: lactic acid bacteria (䊊); aerobic mesophilic bacteria(䉱); L. paracasei (䊏); yeasts (䊉); pH (䉬). The behavior of the probiotic strain was molecularly assessed: the figure shows the REP-PCR profiles obtained from L. paracasei colonies isolated at sampling days 9 and 90. The arrow indicates the type strain.
742
SECTION | II General Aspects and Changes in Food Processing
food companies have translated scientific advances in the field of probiotic vegetable products developing a new probiotic vegetable line bringing the organoleptic excellence of gastronomic products to new customers (Lavermicocca et al., 2005a). The probiotic benefits come in a range of traditional foods, e.g., seasoned table olives, artichokes, salads, with high levels of consumer satisfaction. In addition, probiotic vegetable preparations will have the potential to attract more consumers to functional products since vegetables provide those who are intolerant to milk and its derivatives or on low-cholesterol diets access to new probiotic products. The development of new matrices for probiotic delivery represents an important sector in ongoing international food research. Accordingly, novel in vitro and in vivo protocols have been undertaken to compare the efficacy of several vegetable probiotic foods in sustaining well-being or in reducing the risk of GI diseases.
SUMMARY POINTS ●
●
●
●
●
Probiotics and probiotic foods can prevent gastrointestinal disturbances and improve the well-being of the organism by reinforcing its natural defenses. The main functions carried out by the gut microbiota are involved in the mechanism of action of probiotics; the understanding of these functions helps explain the implication of probiotics on host health. Scientific progress has been made on (i) the effects of probiotics on immune response; (ii) the anticancer activities of probiotics in colorectal neoplasms; and (iii) the therapeutic use of probiotics in patients with constipation. A protective role is played by table olives on probiotic strains since the vegetable carrier helps probiotics to display their abilities in the GI tract. Technological aspects have been solved for the development of probiotic olives by cooperative actions between research institutions and food industries.
REFERENCES Bekkali, N., Bongers, M.E.J., Van den Berg, M.M., Liem, O., Benninga, M.A., 2007. The role of a probiotics mixture in the treatment of childhood constipation: a pilot study. Nutr. J. 6, 17. Di Marzio, L., Russo, F.P., D’Alò, S., Biordi, L., Ulisse, S., Amicosante, G., De Simone, C., Cifone, M.G., 2001. Apoptotic effects of selected strains of lactic acid bacteria on a human T leukaemia cell line are associated with bacterial arginine deiminase and/or sphigomyelinase activities. Nutr. Cancer 40, 185–196. FAO/WHO. Report on Joint FAO/WHO Expert Consultation on Evaluation of Health and Nutritional Properties of Probiotics in Food Including Powder Milk with Live Lactic Acid Bacteria. 2001. ftp:// ftp.fao.org/es/esn/food/probio_report_en.pdf. Furrie, E., 2005. Probiotics and allergy. Pro. Nutr. Soc. 64, 465–469.
Haller, D., Blum, S., Bode, C., Hammes, W.P., Schiffrin, E.J., 2000. Activation of human peripheral blood mononuclear cells by nonpathogenic bacteria in vitro: evidence of NK cells as primary targets. Infect. Immun. 68, 752–759. Kato, I., Endo-Tanaka, K., Yokokura, T., 1998. Suppressive effects of the oral administration of Lactobacillus casei one type II collageninduced arthritis in DBA/1 mice. Life Sci. 63, 635–644. Jass, J.R., Whitehall, V.L., Young, J., Leggett, B.A., 2002. Emerging concepts in colorectal neoplasia. Gastroenterology 123, 862–876. Jonkers, D., Stockbrügger, R., 2007. Review article: probiotics in gastrointestinal and liver diseases. Aliment. Pharmacol. Ther. 26 (Suppl 2), 133–148. Lavermicocca, P., Valerio, F., Lonigro, S.L., Di Leo, A., Visconti, A., 2008. Antagonistic activity of potential probiotic lactobacilli against the ureolytic pathogen Yersinia enterocolitica. Curr. Microbiol. 56, 175–181. Lavermicocca, P., 2006. Highlights on new food research. Dig. Liver Dis. 38, S295–S299. Lavermicocca, P., Lonigro, S.L., Visconti, A., De Angelis, M., Valerio, F., Morelli, L., 2005a. Table olives containing probiotic micro-organisms. Patent application WO 2005/053430A1. Lavermicocca, P., Valerio, F., Lonigro, S.L., De Angelis, M., Morelli, L., Callegari, M.L., Rizzello, C.G., Visconti, A., 2005b. Adhesion and survival of Lactobacilli and Bifidobacteria on table olives with the aim of formulating a new probiotic food. Appl. Environ. Microbiol. 71, 4233–4240. Linsalata, M., Russo, F., Berloco, P., Valentini, A.M., Caruso, M.L., De Simone, C., Barone, M., Polimeno, L., Di Leo, A., 2005. Effects of probiotic bacteria (VSL#3) on the polyamine biosynthesis and cell proliferation of normal colonic mucosa of rats. In vivo 19, 989–996. Maassen, C.B., van Holten-Neelen, C., Balk, F., den Bak-Glashouwer, M.J., Leer, R.J., Laman, J.D., Boersma, W.J., Claassen, E., 2000. Strain-dependent induction of cytokine profiles in the gut by orally administered Lactobacillus strains. Vaccine 18, 2615–2623. Maukonen, J., Alakomi, H.L., Nohynek, L., Hallamaa, K., Leppämäki, S., Mättö, J., Saarela, M., 2006. Suitability of the fluorescent techniques for the enumeration of probiotic bacteria in commercial non-dairy drinks and in pharmaceutical products. Food Res. Int. 39, 22–32. Mohamadzadeh, M., Olson, S., Kalina, W.V., Ruthel, G., Demmin, G.L., Warfield, K.L., Bavari, S., Klaenhammer, T.R., 2005. Lactobacilli activate human dendritic cells that skew T cells toward T helper 1 polarization. Proc. Natl. Acad. Sci. USA 102, 2880–2885. Oberreuther-Moschner, D.L., Jahreis, G., Rechkemmer, G., Pool-Zobel, B.L., 2004. Dietary intervention with the probiotics Lactobacillus acidophilus 145 and Bifidobacterium longum 913 modulates the potential of human faecal water to induce damage in HT29clone 19A cells. Br. J. Nutr. 91, 925–932. Peng, H.F., Jackson, V., 2000. In vitro studies on the maintenance of transcription-induced stress by histones and polyamines. J. Biol. Chem. 275, 657–668. Perdigon, G., de Macias, M.E., Alvarez, S., Oliver, G., de Ruiz, H., 1988. Systemic augmentation of the immune response in mice by feeding fermented milks with Lactobacillus casei and Lactobacillus acidophilus. Immunology 63, 17–23. Perez Chaia, A., Oliver, G., 2003. Intestinal microflora and metabolic activity. In: Fuller, R., Perdigón, G. (Eds.) Gut Flora, Nutrition, Immunity and Health. Blackwell Publishing, Oxford, UK, pp. 77–98. Picard, C., Fioramonti, J., Francois, A., Robinson, T., Neant, F., Matuchansky, C., 2005. Review article: bifidobacteria as probiotic agents – physiological effects and clinical benefits. Aliment. Pharmacol. Ther. 22, 495–512.
CHAPTER | 77 Table Olives: A Carrier for Delivering Probiotic Bacteria to Humans
Pool-Zobel, B.L., Neudecker, C., Domizla, I., Ji, S., Schillinger, U., Rumney, C., Moretti, M., Vilarini, I., Scassellati-Sforzolini, R., Rowland, I., 1996. Lactobacillus- and Bifidobacterium-mediated antigenotoxicity in the colon of rats. Nutr. Cancer 26, 365–380. Rachmilewitz, D., Katakura, K., Karmeli, F., Hayashi, T., Reinus, C., Rudensky, B., Akira, S., Takeda, K., Lee, J., Takabayashi, K., Raz, E., 2004. Toll-like receptor 9 signaling mediates the anti-inflammatory effects of probiotics in murine experimental colitis. Gastroenterology 126, 520–528. Randazzo, C.L., Restuccia, C., Romano, A.D., Caggia, C., 2004. Lactobacillus casei, dominant species in naturally fermented Sicilian green olives. Int. J. Food Microbiol. 90, 9–14. Rangavajhyala, N., Shahani, K.M., Sridevi, G., Srikumaran, S., 1997. Nonlipopolysaccharide component(s) of Lactobacillus acidophilus stimulate(s) the production of interleukin-1 alpha and tumor
743
necrosis factor-alpha by murine macrophages. Nutr. Cancer 28, 130–134. Sartor, R.B., 2005. Probiotic therapy of intestinal inflammation and infections. Curr. Opin. Gastroenterol. 21, 44–50. Schiffrin, E.J., Brassart, D., Servin, A.L., Rochat, F., Donnet-Hughes, A., 1997. Immune modulation of blood leukocytes in humans by lactic acid bacteria: criteria for strain selection. Am. J. Clin. Nutr. 66, 515S–520S. Valerio, F., De Bellis, P., Lonigro, S.L., Morelli, L., Visconti, A., Lavermicocca, P., 2006. In vitro and in vivo survival and transit tolerance of potentially probiotic strains carried by artichokes in the gastrointestinal tract. Appl. Environ. Microbiol. 72, 3042–3045. Van den Berg, D.J.C., Smits, A., Pot, B., Ledeboer, A.M., Kesters, K., Verbakel, J.M.A., Verrips, C.T., 1993. Isolation, screening and identification of lacto acid bacteria from traditional food fermentation processes and culture collections. Food Biotechnol. 7, 189–205.
This page intentionally left blank
Chapter 78
The Oxidative State of Olive Oil Used in Bakery Products with Special Reference to Focaccia Tommaso Gomes, Debora Delcuratolo, Vito Michele Paradiso and Raffaella Nasti Dipartimento di Progettazione e Gestione dei Sistemi Agro-zootecnici e Forestali, University of Bari, Italy
78.1 INTRODUCTION Lipids are a basic component of the daily human diet, as they play many important roles: they are a concentrated source of energy (1 g supplies about 9 kcal), even in the form of storage fat, main constituents of cell membranes in all tissues, precursors of the regulating substances of the cardiovascular system, of blood coagulation, of renal function, of the immune system such as prostaglandin, thromboxanes, prostacycline and leukotrienes (Lee et al., 1989; Kelley et al., 1991), source and carrier of liposoluble vitamins (A, D, K and E), they meet the essential fatty acids requirement (EFA ⫽ essential fatty acids). The bio-nutritional importance of food lipids is closely related to the quantitative and qualitative supply of fatty acids, to the saturated/monounsaturated/polyunsaturated fatty acids ratio and to the intake of the micronutrients contained in the unsaponifiable fraction, especially liposoluble vitamins (notably tocopherols) and phenolic compounds, whose antioxidant properties protect food lipids from oxidative processes (http://www.sinu.it/larn/lipidi.asp). The latter are responsible for undesired sensory characteristics, the decline in the bio-nutritional quality and the formation of harmful substances for human health (Bortolomeazzi, 2003; Parpinel, 2003). The recommended lipid allowance calculated for the Italian population (daily intake levels of nutrients, SINU— Italian Society of Human Nutrition, 1996) is set at 30% of total calories till adolescence, and at 25% in adults. In terms of quality, saturated fatty acids should not exceed 10% of the diet energy (WHO, 1990). As to polyunsaturated fatty acids, the recommended total allowance of essential fatty acids, in adults, should not exceed 2.5% of total calories: 1–2% of calories in the form of linoleic acid (ω-6) and 0.2–0.5% as polyunsaturated fatty acids of Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
the ω-3 series (Commission of the European Communities, 1993) (http://www.sinu.it/larn/lipidi.asp).
78.2 LIPIDS IN FOOD The dietary lipids may include seasonings, natural constituents of food, as well as ingredients used for the preparation of foods. A survey conducted in Italy (Pizzoferrato et al., 1999) within the EC-AIR concerted action Transfair: ‘Trans fatty acid intake and risk factors for cardiovascular disease in Europe’, revealed that lipids supply a considerable percentage, i.e., 30–32% of the total diet energy of the Italian population. Although it is one of the lowest in Europe (Hulshof et al., 1999), this value is markedly higher than the level recommended (25%) for adults (LARN, 1996). The energy supplied by saturated fatty acids is on average 10.6% of the total, thus exceeding the recommended limit of 10% (LARN,1996); this value seems closer to that of Greece (10.5%), but sensibly lower than the levels observed in other European countries (18% in Iceland) (Hulshof et al.,1999). The cis-monounsaturated acids, mostly found in the diet in Central and Southern Italy (respectively 14.2% and 13.7% of total energy), have shown a mean national value of 13.2% of total energy, less than in Greece (17.9%) and Spain (16.3%) (Hulshof et al., 1999). For cis-polyunsaturated acids the intake level observed, equal to 5.1% of the total diet calories, seemed, at the European level, intermediate between the lowest in Sweden (3.3%) and the maximum in Spain (7.6%) (Hulshof et al., 1999). As to the latter group, the comparison between these values and the recommended levels (LARN, 1996) reveals a conspicuous intake for linoleic acid, with 4.7%, (2% recommended), and a poor intake
745
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
746
SECTION | II General Aspects and Changes in Food Processing
for linolenic acid, i.e. 0.3% (0.5% recommended) of the total energy in the national diet. The amount of trans fatty acids (features of trans isomers of fatty acids in Table 78.1), expressed in g capita⫺1 day⫺1 , was poor and among the lowest in Europe (Hulshof et al., 1999): the intake levels were far from the maximum level indicated of 5 g day⫺1, with values ranging from 1.3 g day⫺1 to 1.9 g day⫺1 (mean national value: 1.6 g day⫺1 equal to 0.5% of the total diet energy). In the Italian diet the foods that supply the greatest amount of lipids are oils and fats, followed by milk and its derivatives, meat and charcuterie. A lower but equally important share is provided by bakery products, which contribute to a varying extent depending on the dietary habits of the geographical areas involved (North Western, North Eastern, Central and Southern Italy), by 9.2% to 12.7% of total lipids. Bakery products supply 12.1% of saturated fatty acids (on the total SFA dietary intake), 8.2% of cis-monounsaturated (on the total intake), 11.2% of cis-polyunsaturated (on the total intake) and 15.5% of trans fatty acids (on the total intake) (Pizzoferrato et al., 1999).
78.3 LIPIDS IN BAKERY PRODUCTS: PROPERTIES The fatty substances used in bakery products vary a lot in nature and content. The most largely used lipids are butter, lard, hydrogenated vegetable oils, margarines, olive oil and olive-pomace oil; their content may range from 5 to 15% for some bread substitutes, such as focaccia, rusks, crackers, breadsticks, and up to 20–30% in the case of pastry products, such as biscuits and cakes. The choice of the most suitable kind of lipid and the optimal amount needed in the formulation is closely related
to the desired bakery product, and is based on different parameters such as the dough workability, the product’s rheological and sensory properties, the shelf-life and the consumers’ needs (Anese and Manzocco, 2003). Already during the first processing step the ‘lipid binding’ starts up: chemical and physical interactions are established between the lipids and other ingredients, such as starch and proteins (gluten). This process also involves free endogenous lipids (naturally contained in flours), i.e., lipids not chemically bound to cell structures or to other meal components. The formation of gliadine–lipid–glutenine complexes allows the dough to incorporate the gases, which develop during fermentation, and to retain them in the gluten mesh, thus giving volume and softness to the leavened product. The complexes that are formed between the lipids and the amylose enable the delay of starch retrogradation, which leads to bakery products’ staleness. The complex formation is influenced by the acid composition of the lipid, in particular its degree of unsaturation. Saturated acids, for their ‘concrete’ nature, are less suitable for this function. Unsaturated lipids, instead, help slow down the staling process. By controlling the migration of moisture, responsible for the change in consistency, lipids further influence the shelf-life of the bakery product. The product stability, in terms of conservability, is also related to the use of saturated fatty acids, which slow down the oxidative degradation thanks to their lower oxidation sensitivity. Fats influence dramatically the organoleptic properties of bakery products, notably the flavor (the result of both taste and smell), the aspect (the oils sprayed on the surface of crackers and/or focaccia make it glossy, which is highly appreciated by the consumer) and the texture (Anese and Manzocco. 2003).
TABLE 78.1 Features of trans isomers of fatty acids. ●
Trans fatty acids are unsaturated fatty acids with at least one double bond in the trans configuration
●
They do not occur in vegetable foods (in which double bonds have the cis configuration)
●
Hydrogenated vegetable fats (e.g., margarine) are the main source of trans fatty acids in our diet
●
The metabolic effects of trans isomers are today a matter of controversy generating diverse extreme positions in light of biochemical, nutritional, and epidemiological studies
●
The most recent research pointed out that the consumption of trans fatty acids would increase the risk of cardiovascular disease
●
The food industries have developed several strategies to reduce the trans content of hydrogenated fats, and now products containing low or virtually zero trans are available in the retail market
This table deals with the features of trans fatty acids including chemical structure, the main food source, the metabolic effects, the industrial interest to reduce the trans content in the foods.
CHAPTER | 78 The Oxidative State of Olive Oil used in Bakery Products with Special Reference to Focaccia
The consumers’ request for low-calorie foods has led, over the last few years, to the marketing of bakery products of low lipid content, commonly called ‘light’. In order not to jeopardize the product quality in terms of stability and sensory properties, the industry has adopted a strategy that involves the partial or complete substitution of lipids by substances that have similar functional properties but less calories (polysaccharides, proteins and modified lipids) (Nicoli, 2003b). Acting on the product formulation, i.e., substituting the normally used fats by non-conventional ones, one can meet even specific nutritional and health requirements. For instance, it is possible to obtain bakery products low in cholesterol content using vegetable fats, notably olive oil, for the higher content of unsaturated fatty acids (Anese and Manzocco, 2003).
78.4 FOCACCIA 78.4.1 Short Historical Outline The word focaccia, derived from the Latin focus, meaning fireplace, has very ancient origins. It was already known to the Phoenicians and the Carthaginians, and later to the Greeks. The ‘De Agricoltura’ by Marco Porcio Catone (234–149 B.C.) is the first Latin written source that mentions the Libum farrem, the very ancient Roman focaccia, offered in sacrifices and during the celebration of weddings. Known in the Middle Ages as a poor food, it was re-valued in the Renaissance (http://www.taccuinistorici.it; http://www.prodottitipici.com). Today it is a niche bakery product, a typical food of some Italian regions. It is prepared ‘by well-established processing methods, which are homogeneous all over the areas concerned and which follow traditional rules,’ so it has been officially recognized as a Traditional Agri-food Product (according to art. 8 of D.Lgs. April 30 1998, no 173) (http://it.wikipedia.org/wiki/Focaccia_ genovese) for Piedmont and Liguria regions, included in a special list, produced by the Ministry for Agricultural, Food and Forestry Policies with the collaboration of the Regions, and added to the list of Slow Food products (Traditional Focaccia made in Genoa) (http://www.fondazioneslowfood. it/pdf/Italiani2007.pdf). Highly appreciated for its organoleptic properties, it is usually consumed as a snack, an appetizer, an alternative to bread, and is also available in sweet versions. Traditional focaccia is obtained from few and simple ingredients: flour, yeast, water, salt and oil. Some variations are obtained by topping it, prior to cooking, with various ingredients (fresh tomato, onions, potatoes, olives, cheese, etc.) or flavoring it with herbs (rosemary, sage, oregano, etc.). After leavening, the dough is broken in pieces that are hand-shaped and adapted to the baking pans first by an ‘ironing’ process, then by using fingertips so as to make dimples on the surface. It is oven-cooked in flat iron pans.
747
Olive oil, notably extra virgin olive oil, is an essential ingredient in focaccia preparation. The partial or total substitution of extra virgin olive oil by lard and/or pomaceoil, as unfortunately happens today, affects negatively the focaccia’s sensory properties, digestibility, shelf life and nutritional value. Virgin olive oil makes focaccia pleasant and palatable, giving a characteristic flavor (smell and taste); this is partly due to fatty acids, which are maximum when the oleic/linoleic acid ratio is ⭓7, but above all to some components of the unsaponifiable fraction, notably phenolic compounds, alcohols and aldehydes with 6–8 carbon atoms (Conte, 2004; Cappelli and Vannucchi, 2005). Focaccia digestibility is related to the virgin olive oil and its high content of oleic acid (the optimal value being not less than 73%) (Conte, 2004), which is the most digestible monounsaturated fatty acid for the human organism. Thanks to its organoleptic properties and its melting point that is below the human body temperature, it stimulates the pancreatic lipase secretion, by conditioned reflexes, and favors its hydrolytic activity. Moreover, the high absorption coefficient of oleic acid, due to the stimulation on biliary secretion (cholecystokinetic effect), also favors the absorption of the liposoluble vitamins contained in food (Cappelli and Vannucchi, 2005). From a bio-nutritional and health point of view, olive oil – by its chemical composition – supplies nutrients that pomace or lard cannot supply. Besides the high oleic acid content and the supply of essential fatty acids, it supplements the product with a wide range of substances, components of the non-saponifiable fraction (sterols, aliphatic and triterpene alcohols, polyphenols, tocopherols…), which are quantitatively poor (0.5–1.4%), but which play an important role in the organism as they are involved in many biochemical and physiological processes (Cappelli and Vannucchi, 2005).
78.5 EFFECTS OF TEMPERATURE ON THE LIPID FRACTION OF OVEN-COOKED FOCACCIAS 78.5.1 Introduction Little research has focused on the impact of baking on the oxidative state of the lipid fraction of baked products, with the exception of some investigations regarding trans fatty acids (van Erp-baart et al., 1998; Priego-Capote et al., 2004). Oxidative phenomena regarding lipids have been studied essentially for other cooking methods, such as frying (Coni et al., 2004; Gil et al., 2004; Zhang et al., 2004; Naz et al., 2005). Our investigation was directed to evaluate the impact of technology on focaccias baked in non-industrial ovens focusing, in particular, on the hydrolytic and oxidative changes of the lipid fraction.
748
SECTION | II General Aspects and Changes in Food Processing
Among the numberless variations supplied on the market, we selected four types of focaccia, which were particularly appreciated by consumers: tomato-topped focaccia, potato-topped focaccia, onion-topped focaccia, and rosemary-topped focaccia. The focaccias, made from the same dough (about 400 g), were all seasoned with 200– 250 g of topping, except for the last type topped with 2 g of dried rosemary, flavored with the same extra virgin olive oil and then baked in a thermo-ventilated gas oven with four baking chambers at 220 °C for 20 minutes.
78.5.2 Conventional Analytical Parameters The acid composition of the extra virgin olive oil used to make the focaccias is reported in Figure 78.1 (Official Journal of the European Communities, 1991). Table 78.2 contains the mean percent amounts of trans isomers of fatty acids and the results of the statistical analyses. The uncooked oil had 0.02% of C18:1 trans, which was considerably lower than the allowed limit (0.05%) and contained trace levels of trans isomers of linoleic and linolenic acids. The baked oils extracted from the focaccias had amounts of C18:1 trans that were 4 to 5 times greater than the levels in the uncooked oil and the differences were statistically significant. The levels of trans isomers of linoleic and linolenic acids in the baked oils ranged from 0.01 to 0.02% without exceeding the allowed limit for virgin olive oils (0.05%) (Official Journal of the European Communities, 1992). The oil sampled from the rosemary-topped focaccia had lower levels of both C18:1 trans and trans isomers of linoleic and linolenic acids. The amounts of trans isomers of fatty acids found in the present investigation were smaller than the amounts generally reported for other baked products or breads (van Erpbaart et al., 1998). The results of the routine analyses (FFA, PV, p-AV, K232, K270 and ΔK) performed on the raw oil and on the baked oil samples (Table 78.2) provided initial information on the hydrolytic and oxidative degradation that occurred during the baking of the focaccias. The percent free fatty acids increased substantially in the tomato-topped focaccia with values that were three-fold greater than in the raw oil. This was probably ascribable to the acidic nature of the ingredient used for seasoning. The levels of hydrolytic degradation remained below 2%, the threshold value prescribed by the European legislation for virgin olive oils (Official Journal of the European Communities, 2003). By contrast, the indices of oxidative degradation pointed to a quality decline of the oils extracted from the focaccias after baking with values that were beyond those of virgin olive oils. The peroxide value of the raw extra virgin olive oil used in the investigation was 16.4 meq kg⫺1 – hence, within the allowed limit of 20 meq kg⫺1 – the oils
FIGURE 78.1 Percent acidic composition of the extra virgin olive oil employed in the study. This figure shows the most characteristic fatty acids of the extra virgin olive oil: C18:1 ⫽ oleic acid; C16:0 ⫽ palmitic acid; C18:2 ⫽ linoleic acid; C18:0 ⫽ stearic acid; C16:1 ⫽ palmitoleic acid; C18:3 ⫽ linolenic acid.
extracted after baking had substantially greater values ranging from 26.3 to 57.6 meq kg⫺1. The p-AV test results indicated that extensive secondary oxidative degradation had occurred since the p-AV determinations of the baked oils were almost two-fold greater than the values of the uncooked oil. K270 increased dramatically after baking from 0.136 (allowed limit: 0.22) in the unbaked oil to values ranging from 0.655 to 1.041, whereas K232 and ΔK showed only slight increases. Total oxidation of the baked oils, evaluated as TOTOX (2PV ⫹ p-AV) was twoto three-fold greater than in the unbaked oil with particularly higher values in the oils extracted from the focaccias topped with onions and rosemary (Table 78.2).
78.5.3 Unconventional Analytical Parameters High-performance size exclusion chromatography (HPSEC) of the polar compounds (PC) provided more detailed information on the oxidative and hydrolytic degradation of the baked oil samples via the determination of the following classes of compounds: triglyceride oligopolymers (TGP); oxidized triglycerides (ox-TG) and diglycerides (DG) (Gomes and Caponio, 1999). The HPSEC chromatograms of the polar compounds of the unbaked oil and of the same oil sampled after baking from the onion-topped focaccia are shown in Figure 78.2. As already shown by the routine analyses, the starting uncooked oil already contained detectable amounts of TGP that were indicative of the inception of the oxidative process. The same oil, extracted from the baked focaccias, contained substantially higher TGP and ox-TG levels. The PC data and the results of the HPSEC analyses of the PC of the oils examined are reported in Table 78.2
CHAPTER | 78 The Oxidative State of Olive Oil used in Bakery Products with Special Reference to Focaccia
749
TABLE 78.2 Mean results of analyses for each oil sample examineda. Oil extracted from Samples
Extra virgin olive oil
Potato-topped focaccia
Tomato-topped focaccia
Onion-topped focaccia
Rosemarytopped focaccia
FFA (%)a
0.6
1.0
1.9
0.7
1.1
PV (meq kg⫺1)a
16.4
30.7
26.3
51.7
57.6
p-Ava
5.90
16.60
32.77
22.37
12.03
K232a
2.110
2.386
2.396
2.121
2.217
K270a
0.136
0.734
1.041
0.655
0.680
ΔKa
⫺0.021
0.004
0.025
0.011
0.011
TOTOXa
38.7
78.0
85.4
125.8
127.2
PCb
3.34 ⫾ 0.10a
3.68 ⫾ 0.06b
5.44 ⫾ 0.10c
4.78 ⫾ 0.12d
5.55 ⫾ 0.12c
TGPb
0.08 ⫾ 0.01a
0.13 ⫾ 0.03b
0.22 ⫾ 0.03c
0.27 ⫾ 0.02d
0.25 ⫾ 0.03c
ox-TGb
0.75 ⫾ 0.04a
0.73 ⫾ 0.05a
1.06 ⫾ 0.13b
1.69 ⫾ 0.19c
2.08 ⫾ 0.32d
DGb
1.63 ⫾ 0.16a
1.55 ⫾ 0.01a
1.95 ⫾ 0.05b
1.83 ⫾ 0.06b
1.85 ⫾ 0.05b
2TGP ⫹ ox-TGb
0.91 ⫾ 0.04a
0.99 ⫾ 0.06b
1.50 ⫾ 0.12c
2.23 ⫾ 0.15d
2.58 ⫾ 0.31d
C18:1tb
0.022 ⫾ 0.002a
0.110 ⫾ 0.018bc
0.105 ⫾ 0.006bc
0.125 ⫾ 0.007b
0.085 ⫾ 0.008c
C18:2t ⫹ C18:3tb
tr.a
0.023 ⫾ 0.003b
0.012 ⫾ 0.001c
0.010 ⫾ 0.001c
0.008 ⫾ 0.004c
This table shows the results of the analyses performed on the raw oil and on the baked oil samples. Abbreviations used in the table. a Mean values of two independent repetitions. FFA, free fatty acids; PV, peroxide value; p-AV, p-anisidine value; K232, specific absorption at 232 nm; K270, specific absorption at 270 nm; ΔK ⫽ K270 ⫺ (K266 ⫹ K274)/2; TOTOX ⫽ 2PV ⫹ p-AV; b Results of statistical analysis at p ⬍ 0.05. Mean values of three independent repetitions ⫾ SD; one common letter following an entry indicates no significance. PC, polar compounds; TGP, triglyceride polymers; ox-TG, oxidized triglycerides; DG, diglycerides; C18:1t ⫽ trans oleic acid; C18:2t ⫽ trans linoleic acid; C18:3t ⫽ trans linolenic acid; tr., traces (not integrated). Reprinted with minimal modifications from Delcuratolo D., et al. Food Chem. 2008; 106: 222–226, with permission.
together with the statistical data. PC defines the extent of the overall degradation of an oil, as they include classes of substances of triglyceride oxidation, polymerization and hydrolysis. The amount of PC in the baked oils was significantly greater in all samples than in the uncooked extra virgin olive oil. The smallest difference was observed in the potato-topped focaccia (10% increase over the uncooked oil). In the oil samples from the three other types of focaccia, substantial increases in the PC ranged from 43% to 66%. The percent amount of ox-TG measured in the uncooked oil was 0.75%, which was not statistically different from the amount measured in the oil sampled from the potato-topped focaccia. By contrast, statistically significant differences were registered in the oil samples from the other types of focaccia with ox-TG values ranging from 1.4 to 2.7 times the amount measured in the uncooked oil. These findings confirm that the oils sampled from the
focaccias topped with tomatoes, onions and rosemary had undergone more intense oxidation, as already highlighted by the TOTOX values in Table 78.2. Baking led to a significant increase in the percent amount of oligopolymers in all the oils sampled from the baked focaccias. The percent amount of TGP in the uncooked oil was 0.08%. In the potato-topped focaccia it was 0.13%, namely over 1.5 times that of the uncooked oil, while in the three other sampled oils, the amount of TGP was almost three-fold that of the raw oil. Evaluation of the oligopolymers provided further evidence that the oil extracted from the potato-topped focaccia presented a less intense degradation than did the other baked oils. The percent amount of diglycerides in the uncooked oil was 1.63%. Significantly higher values were found in the baked oils, except for the oil sampled from the potatotopped focaccia.
750
SECTION | II General Aspects and Changes in Food Processing
Finally, Table 78.2 shows the mean values and SD of the sum of 2TGP% ⫹ ox-TG%, a parameter which provides a better evaluation of the overall oxidation (Gomes et al., 2003). Substantial increases in overall oxidative degradation were found after baking in oils from focaccias topped with tomato, onion and rosemary. Once again, the oil from the potato-topped focaccia seemed to be less affected by the baking process with an overall oxidation index (2TGP% ⫹ ox-TG%) of 0.99 as compared to 0.91, which had been measured in the uncooked oil. As already shown with the total oxidation index (TOTOX), the oils from the focaccias topped with onions and rosemary proved to be the most oxidized. TOTOX and (2TGP% ⫹ ox-TG%) for all samples showed a substantial agreement and a positive correlation (p ⬍0.05).
78.5.4 Influence of the Toppings The different levels of oxidation found in the oils sampled from the different types of focaccias seem to be ascribable to the amounts of the toppings used, their properties and their percent moisture. The toppings that covered the whole surface (potato, onion and tomato) had the effect of mitigating the rise in temperature in the focaccia, partly as a consequence of their high water content (78.5–94% range), which evaporated during cooking, thus exposing oil to a less severe heat stress. Seasoning with diced potatoes seemed to have better protected the lipid fraction from oxidation. This result could be explained by the onset of Maillard’s reaction on the potato submitted to heat treatment (Mottram et al., 2002). Previous investigations have shown that the MRPs have a strong antioxidant activity effectively slowing down the lipids’ oxidative degradation (Dalla Rosa et al., 1992; Munari et al., 1995; Severini and Lerici, 1995; Wijewickreme and Kitts, 1997; Nicoli, 2003a). The highest oxidative degradation level observed in the seasoning oil extracted from rosemary-topped focaccia could be due to the use of 2 g only of dry rosemary scattered over the whole focaccia surface; hence, without the protective effect of water vapor during heating, the focaccia would have been more exposed to the heat action. Furthermore, other factors may have affected the rosemary antioxidant activity, at pre- and post-harvest (storage, drying process), and the heat stress this herb has been exposed to during cooking. Actually, the stability of the antioxidant power is little known in relation to cooking time and temperatures (Nicoli, 2003a), so that these variables could have also influenced the rosemary’s protective activity, due especially to carnoxic acid and carnosol (Offord et al., 1997), and the degradation of the applied oil. The results of the analyses performed on the baked focaccia oils, compared with those obtained from refined oils, proved that the overall level of oxidation, expressed in terms of (2TGP% ⫹ ox-TG%), of the extra virgin olive oil used for focaccia baking was lower than the levels found in refined oils (Gomes and Caponio, 1997; Gomes et al., 2003) and considerably lower than the levels found in oils cooked by different methods, especially fried oils (Sebedio et al., 1987; Arroyo et al., 1992).
78.5.5 Statistical Comparison with Refined Seed Oils
FIGURE 78.2 HPSEC chromatograms of the polar compounds of the unbaked oil (A) and of the same oil sampled after baking from the oniontopped focaccia (B). This figure shows the trend of the polar compounds classes: 1, triglyceride oligopolymers; 2, oxidized triglycerides; and 3, diglycerides. Reprinted from Delcuratolo D., et al. Food Chem. 2008; 106: 222–226, with permission.
The analytical data were compared with those obtained in previous works (Gomes, 1992; Gomes et al., 2003) that used routine analyses and HPSEC to investigate the oxidative state of extra virgin olive oils and refined seed oils. Principal components analysis (PCA) was performed on the dataset to allow the comparison. Figures 78.3 and 78.4 show respectively the correlation circle and the score plot of the first two principal components
CHAPTER | 78 The Oxidative State of Olive Oil used in Bakery Products with Special Reference to Focaccia
FIGURE 78.3 Correlation circle of the first two principal components obtained by PCA of the analytical data. Correlation circle shows a projection of the initial variables in the factors space. When variables are far from the center, their variability can be satisfactorily explained by the factors; when the variables are close to the center, it means that some information is carried on other axes. If variables are close to each other, they are positively correlated; if they are orthogonal, they are not correlated; if variables are on opposite sides of the center, they are negatively correlated. This graph is useful in interpreting the meaning of the axes. DG, diglycerides; FFA, free fatty acids; K232, specific absorption at 232 nm; ox-TG, oxidized triglycerides; PC, polar compounds; PV, peroxide value; TGP, triglyceride oligopolymers; PC1, first principal component; PC2, second principal component (in parentheses the variance is given).
751
that account for 78% of the total variability, obtained by PCA of the analytical data. PC1 is strongly correlated to TGP, K232 and ox-TG and shows negative correlations with FFA and PV. PC2 is positively correlated with FFA, PV, ox-TG, PC and DG, though the variability of the latter is partially expressed by the two first principal components. It can be observed that the oils extracted from focaccias are closer to the extra virgin olive oils than the refined seed oils. The oxidative and hydrolytic processes that occur during baking cause a remarkable increase in PV and FFA, but the increase in the amount of oxidized triglycerides and triglyceride oligopolymers is moderate, if compared with the levels observed in refined oils. It should be considered that peroxides are degraded during the refinement, so that the assessment of the peroxide value is not reliable to assess the oxidative state of refined oils. The amounts of TGP and ox-TG are more reliable indexes for this purpose and show that refined oils are characterized by more pronounced oxidative phenomena as compared to an extra virgin olive oil submitted to a baking process. Thus the oxidation involving seasoning oil during baking proved to be moderate when employing an extra virgin olive oil, which shows lower oxidation levels than uncooked refined seed oils.
78.6 CONCLUSIONS In conclusion, regardless of the toppings used, the modifications induced by technology on the fat extracted from
FIGURE 78.4 Score plot of the first two principal components obtained by PCA of the analytical data of the focaccias. This score plot represents extra virgin olive oils, oils extracted from focaccias and refined seed oils on a two-dimensional map. It enables description of the samples with only two variables (principal components), explaining most of the variability expressed by the analytical data. PC1, first principal component; PC2, second principal component (in parentheses the variance is given); EVOO, extra virgin olive oil employed in focaccias; OTFO, oil extracted from onion-topped focaccia; PTFO, oil extracted from potato-topped focaccia; RTFO, oil extracted from rosemary-topped focaccia; TTFO, oil extracted from tomato-topped focaccia.
752
SECTION | II General Aspects and Changes in Food Processing
investigated focaccias, are not so important and do not jeopardize the final product quality. Hence, in focaccia preparation, it is better to use a good-quality olive oil like extra virgin oil because, thanks to the presence of highly antioxidant micronutrients and its particular acid composition, with a polyunsaturated/monounsaturated/saturated ratio equal to 0.5:5:1 (Conte, 2004), it has resulted in being particularly resistant to thermal oxidation ensuring a ‘healthy’ product.
SUMMARY POINTS ●
●
●
●
●
●
●
●
●
In the Italian diet, bakery products supply a varying amount of lipids, depending on the dietary habits of the geographical areas involved from 9.2% to 12.7% of total lipids. The fatty substances used in bakery products vary a lot in nature. The most commonly used lipids are butter, lard, hydrogenated vegetable oils, margarines, olive oil and olive-pomace oil. In technological processes involving the use of heat, the virgin olive oil, among all edible fats, is the most stable to oxidative and hydrolytic degradation, thanks to the high oleic acid content and the antioxidant micronutrients contained in the unsaponifiable fraction. Olive oil, notably extra virgin olive oil, is an essential ingredient in focaccia preparation. Focaccia is a niche bakery product, a typical food of some Italian regions. The partial or total substitution of extra virgin olive oil by lard and/or pomace-oil affects negatively the focaccia’s sensory properties, digestibility, shelf life and nutritional value. Our investigation showed that the level of degradation of the seasoning oil is influenced by the amounts of the toppings used, their properties and their percent moisture. The level of degradation found after baking of the focaccias proved to be rather little and lower than that generally found in refined oils. Regardless of the toppings used, the modifications induced by technology on the fat extracted from investigated focaccias, are not so important and do not jeopardize the final product quality.
REFERENCES Anese, M., Manzocco, L., 2003. Le sostanze grasse nei prodotti da fornoLe funzioni tecnologiche dei grassi. In: Impiego di oli e grassi nella formulazione dei prodotti da forno. AREA Science Park – Progetto Novimpresa, pp. 59–70. Arroyo, R., Cuesta, C., Garrido-Polonio, C., López-Varela, S., SánchezMuniz, F.J., 1992. High-performance size-exclusion chromatographic studies on polar components formeds in sunflower oil used for frying. J. Am. Oil Chem. Soc. 69, 557–563. Bortolomeazzi, R., 2003. Le sostanze grasse alimentari-Stabilità e modalità di magazzinaggio. In: Impiego di oli e grassi nella formulazione
dei prodotti da forno. AREA Science Park – Progetto Novimpresa, pp. 49–58. Cappelli, P., Vannucchi, V., 2005. Oli e grassi vegetali: olio di oliva. In: Zanichelli, X. (Ed.), Chimica degli alimenti. Conservazione e trasformazioni, 3rd edn. Bologna, pp. 538–560. Coni, E., Podestà, E., Catone, T., 2004. Oxidizability of different vegetables oils evaluated by thermogravimetric analysis. Thermochim. Acta 418, 11–15. Conte, L., 2004. Olio di oliva. In Chimica degli alimenti. Piccin Nuova Libraria, Padova, pp. 209–228. Dalla Rosa, M., Bressa, F., Barbanti, D., 1992. Maillard reaction products and lipid oxidation: studies on model systems and foods. In: Dalla Rosa, M., Sensidoni, A., Stecchini, M. (Eds.) Proceedings of the Workshop Azioni combinate nella stabilizzazione dei prodotti alimentary. CNR-RAISA Flair- Flow Europe, Udine, Italy. Gil, B., Cho, Y.J., Yoon, S.H., 2004. Rapid determination of polar compounds in frying fats and oils using image analysis. Lebensm.-Wiss. u.-Technol. 37, 657–661. Gomes, T., 1992. Oligopolymer, diglyceride and oxidized triglyceride contents as measure of olive oil quality. J. Am. Oil Chem. Soc. 69, 1219–1223. Gomes, T., Caponio, F., 1997. Investigation on the degree 261 of oxidation and hydrolysis of refined olive oils. An approach for better product characterisation. Ital. J. Food Sci. 4, 277–285. Gomes, T., Caponio, F., 1999. Effort to improve the quantitative determination of oxidation and hydrolysis compound classes in edible vegetable oils. J. Chromatogr. A 844, 77–86. Gomes, T., Caponio, F., Delcuratolo, D., 2003. Fate of oxidized triglycerides during refining of seed oils. J. Agric. Food Chem. 51, 4647–4651. Hulshof, K.F.A.M., van Erp-Baart, M.A., Anttolainen, M., Becker, W., Church, S.M., Couet, C., Hermann-Kunz, E., Kesteloot, H., Leth, T., Martins, I., Moreiras, O., Moschandreas, J., Pizzoferrato, L., Rimestad, A.H., Thorgeirsdottir, H., van Amelsvoort, J.M.M., Aro, A., Kafatos, A.G., Lanzmann-Petithory, D., van Poppel, G., 1999. Intake of fatty acids in Western Europe with emphasis on trans fatty acids: the TRANSFAIR study. Eur. J. Clin. Nutr. 53, 143–157. Kelley, D.S., Branch, L.B., Love, J.E., Taylor, P.C., Rivera, Y.M., Iacono, J.M., 1991. Dietary α-linolenic acid and immunocompetence in humans. Am. J. Clin. Nutr. 53, 40–46. LARN, 1996. Livelli di Assunzione Raccomandati di Energia e Nutrienti per la Popolazione Italiana. Società Italiana di Nutrizione Umana, Roma, pp. 63–68. Lee, T.H., Hoover, R.L., Williams, J.D., Sperling, R.J., Ravalese, J., Spur, B.W., et al., 1989. Effect of dietary enrichment with eicosapentaenoic and docosahexaenoic acids on in vitro neutrophil and monocyte leukotriene generation and neutrophil function. N. Engl. J. Med. 312, 1217–1224. Mottram, D.S., Wedzicha, B.L., Dodson, A.T., 2002. Acrylamide is formed in the Maillard reaction. Nature 419, 448–449. Munari, M., Mastrocola, D., Nicoli, M.C., Lerici, C.R., 1995. Interazione tra prodotti della reazione di Maillard (MRP) e ossidazione dei lipidi in sistemi modello ad umidità intermedia. Riv. Ital. Sostanze Grasse. 72, 351–354. Naz, S., Siddiqi, R., Sheikh, H., Sayeed, S.A., 2005. Deterioration of olive, corn and soybean oils due to air, light, heat and deep-frying. Food Res. Int. 38, 127–134. Nicoli, M.C., 2003a. Le sostanze grasse nei prodotti da forno-Interazioni tra grassi e altri ingredienti in fase di formulazione e cottura. In: Impiego di oli e grassi nella formulazione dei prodotti da forno. AREA Science Park – Progetto Novimpresa, pp. 71–75.
CHAPTER | 78 The Oxidative State of Olive Oil used in Bakery Products with Special Reference to Focaccia
Nicoli, M.C., 2003b. Le sostanze grasse nei prodotti da forno-I sostituti dei grassi. In: Impiego di oli e grassi nella formulazione dei prodotti da forno. AREA Science Park – Progetto Novimpresa, pp. 85–94. Official Journal of the European Communities. 1991. 248 (Sept 5, 1991), EC Regulation 2568/91. Official Journal of the European Communities. 1992. 248 (May 26, 1992), EC Regulation 1429/92. Official Journal of the European Communities. 2003. 295 (Nov 6, 2003), EC Regulation 284 1989/2003. Offord, E.A., Guillot, F., Aeschbach, R., Loliger, J., Pfeifer, A.M.A., 1997. Antioxidant and Biological Propierties of Rosemary Components: Implications for Food and Helth in Natural Antioxidants. AOCS press Champaign, Illinois. Parpinel, M., 2003. Lipidi e nutrizione – Aspetti nutrizionali di oli e grassi. In: Impiego di oli e grassi nella formulazione dei prodotti da forno. AREA Science Park – Progetto Novimpresa, pp. 95–106. Pizzoferrato, L., Leclercq, C., Turrini, A., Van Erp-Baart, M.A., Hulshof, K., 1999. Livelli di ingestione di lipidi ed acidi grassi in Italia: I risultati dell’azione concertata CE “Transfair”. La Rivista di Scienza dell’Alimentazione 3, 259–270. Priego-Capote, F., Ruiz-Jiménez, J., García-Olmo, J., Luque De Castro, M.D., 2004. Fast method for the determination of total fat and trans fatty-acids content in bakery products based on microwave-assisted Soxhlet extraction and medium infrared spectroscopy detection. Anal. Chim. Acta. 517,13-20.
753
Sebedio, J.L., Grandgirard, A., Septier, C., Prevost, J., 1987. 307 Etat d’altération de quelques huiles de friture prélevées en restauration. Revue. Française des Corps. Gras. 34, 15–18. Severini, C., Lerici, C.R., 1995. Interaction between the Maillard reaction and lipid oxidation in model systems during high temperature treatment. Ital. J. Food Sci. 2, 189–196. van Erp-Baart, M.A., Couet, C., Cuadrado, C., Kafatos, A., Stanley, J., van Poppel, G., 1998. Trans fatty acids in bakery products from 14 European countries: the TRANSFAIR study. J. Food Compos. Anal. 11, 161–169. Wijewickreme, A.N., Kitts, D.D., 1997. Influence of reaction conditions on the oxidative behavior of model Maillard reaction products. J. Agric. Food Chem. 45, 4571–4576. WHO, 1990. Diet, nutrition and the prevention of chronic diseases. Technical Report Series 797. Geneva, WHO. Zhang, C.X., Wu, H., Weng, X.C., 2004. Two novel synthetic antioxidants for deep frying oils. Food Chem 84, 219–222.
WEBSITES http://www.sinu.it/larn/lipidi.asp http://www.fondazioneslowfood.it/pdf/Italiani2007.pdf http://www.taccuinistorici.it http://www.it.wikipedia.org/wiki/Focacciagenovese http://www.prodottitipici.com
This page intentionally left blank
Chapter 79
Recovery and Distribution of Macro- and Selected Microconstituents after Panfrying of Mediterranean Fish in Virgin Olive Oil Nick Kalogeropoulos and Antonia Chiou Laboratory of Chemistry–Biochemistry–Physical Chemistry of Foods, Department of Science of Dietetics-Nutrition, Harokopio University, Athens, Greece
79.1 INTRODUCTION Frying is a very old cooking technique, used as early as 1600 BC by the ancient Egyptians and later by the Greeks and the Romans (Banks, 1996). It involves heat and mass transfer and includes complex interactions between the food and the frying medium, some of which are dependent on the process itself and others on the food and the oil type used (Saguy and Dana, 2003). Frying is a fast-cooking and efficient method that improves food sensory quality by the formation of flavor, attractive color, crust, and texture, together with food sterilization. As a result of their unique and delicious sensory characteristics fried foods are consumed worldwide with sustainable popularity, despite their considerable fat content and the consumers’ awareness of the relationship between food, nutrition and health (Saguy and Dana, 2003). Olive oil together with vegetables, fruits, grains, legumes and fish are central elements of the Mediterranean diet. Unlike other fat-rich diets such as the Western-type diet, some 85% of the fat content in the Mediterranean diet comes from olive oil (Simopoulos, 2001). Furthermore, owing to the special gastronomic characteristics possessed by olive oil, its addition into certain dishes facilitates the consumption of products containing high proportions of low-glycemic-index carbohydrates with health-promoting potential such as fruits and vegetables. The same is also true for fish, valued as food of high nutritional quality and frequently consumed in the Mediterranean diet (Simopoulos, 2001). Pan-frying of fish in olive oil is common practice Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
in the Mediterranean as well as other European countries (Sioen et al., 2006). In Greece, pan-frying is actually the only cooking method for small, relatively lean fish like Atherina boyeri, Mullus barbatus and Spicara smaris, while the same is also true for small Boops boops. The average fish consumption in Greece, Italy, Spain and Portugal is estimated to be 40, 30, 48 and 71 g person⫺1 day⫺1, respectively (DAFNE, 2008). During frying, a wide spectrum of physical and chemical changes occurs. In the presence of oxygen, food moisture, and high temperatures, the oil is deteriorated through hydrolysis, oxidation and thermal alteration, leading to a variety of byproducts that reduce its organoleptic characteristics and its nutritional value. Among the commodities used for frying, fish are highly susceptible to oxidation as a consequence of their high content of polyunsaturated fatty acids (PUFA). These changes can be kept at low levels if pan-frying is conducted in oils of good quality, preferably with low unsaturation, which are used only for 1–2 fryings. Despite common belief, research has indicated that frying has the same or even less effect on nutrient and vitamin losses compared to other cooking methods (Fillion and Henry, 1988; Bognár, 1998), while furthermore, the nutritive value of food may increase due to the absorption of frying oils, which are usually rich in unsaturated fatty acids and vitamin E. Virgin olive oil (VOO) is unique among cooking oils and fats, being very rich in monounsaturated fatty acids (MUFA) and containing significant amounts of health-promoting
755
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
756
SECTION | II General Aspects and Changes in Food Processing
microconstituents, i.e., tocopherols, polyphenols, squalene, phytosterols, and terpenic acids. Several observational and clinical studies conducted on humans have shown that increased MUFA intake may be associated with cardiovascular heart disease risk reduction and may be protective against age-related cognitive decline and Alzheimer’s disease (Covas, 2007; Perez-Jimenez et al., 2007). Oleic acid has also been associated with in vitro chemoprotection (Waterman and Lockwood, 2007). Polyphenols are phytochemicals with strong antioxidant potential that demonstrate high ability for free radical scavenging (Visioli et al., 1998). They are important preventive agents against several degenerative diseases, protecting body tissues against oxidative stress. Numerous studies have shown that polyphenols may be protective in vitro against several types of cancer, such as breast, prostate, skin and colon cancer, and associated with low incidence of cardiovascular diseases (Covas, 2007; Perez-Jimenez et al., 2007). Tocopherols have a fundamental role as natural lipidphase antioxidants, scavenging radicals in cellular and subcellular membranes and lipoprotein particles (Traber and Atkinson, 2007). Although mainly acting as an antioxidant, vitamin E can also have non-antioxidant functions, i.e., as a signaling molecule, as a regulator of gene expression, and, possibly, in the prevention of cancer and atherosclerosis (Schneider, 2005). Phytosterols are natural dietary components with serum cholesterol-lowering properties (Law, 2000), while in a recent critical review it was proposed/confirmed that additionally they possess anticancer, anti-inflammatory, antiatherogenic and antioxidative activities (Berger et al., 2004). Squalene is a triterpene involved in the biosynthesis of sterols in plants and animals (Kelly, 1999). It is an inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A reductase activity and increases the activity of the acyl coenzyme A cholesterol acyltransferase. It has been suggested that the former activity is responsible for the squalene tumorinhibitory activity observed in animal models (Covas et al., 2006). It is widely distributed in nature and exists in reasonable amounts in VOO (Smith, 2000). Squalene, together with phenolic compounds and oleic acid is considered to contribute to the olive oil anti-inflammatory properties (Kelly, 1999). Squalene anticarcinogenic activity, especially for colon cancer in experimental animals (Smith, 2000) and humans (Rao et al., 1998), has also been reported. VOO furthermore contains hydroxy pentacyclic triterpene acids (HPTA), particularly oleanolic, maslinic and ursolic acids (Pérez-Camino and Cert, 1999). During the last two decades, pharmacological studies of oleanolic and ursolic acids indicated that they exhibit beneficial effects, notably hepato-protection, anti-inflammation, antitumorpromotion and antihyperlipidemia (Liu, 1995), while maslinic and ursolic acids exhibit anti-HIV activity (Xu et al., 1996).
79.2 COMPOSITIONAL CHANGES OF FISH PAN-FRIED IN VOO 79.2.1 Macronutrients and energy content The proximate composition and energy content of the edible portion of fresh and pan-fried fish are presented in Table 79.1. The major compositional changes observed during fish frying are the significant decrement of moisture due to water evaporation together with an increase of fat content due to the absorption of frying oil. The oil uptake of fried fish is negatively correlated with the size and the initial fat of the fish (Mai et al., 1978; Gall et al., 1983; Bognár, 1998; Kalogeropoulos et al., 2004). Water loss and oil uptake result also in a net increase of protein and energy content of pan-fried relatively lean Mediterranean fish (Kalogeropoulos et al., 2004, 2006) (Table 79.1). The concentrating effect of water loss and the diluting effect of oil absorption normally result in an increment of protein content in fried fish (Table 79.1). Although the energy content of fried fish on a fresh weight basis significantly increases (Table 79.1), it should be mentioned here that, in the Mediterranean cuisines pan-fried fish are traditionally accompanied by fresh salads or boiled greens, i.e., food of low fat content. Therefore, a typical meal containing fried fish is not expected to supply excess fat and energy.
79.2.2 Fatty Acids Fish lipids, and especially the n-3 polyunsaturated fatty acids (n-3 PUFA) such as eicosapentaenoic acid (EPA, 20:5 n-3) and docosahexaenoic acid (DHA, 22:6 n-3), exhibit anti-inflammatory activity (Kulas and Ackman, 2001; Calder, 2005) and are considered to be associated with a lower incidence of fatal coronary heart disease (Brouwer et al., 2006). According to their lipid content, finfish can be classified as lean (⬍2%), low fat (2–4%), medium fat (4–8%), and high fat (⬎8%) (Kołakowska et al., 2003). Fish lipids usually vary, affected by environmental factors, reproductive season, fishery period, etc. (Varela and Ruiz-Roso, 1992); for example, the annual variation of sardine lipids was reported to be between 1.2–18% w/w (Bandarra et al., 1997). During fish frying, moisture loss, exchange of lipids between frying fat and food and oxidation reactions generated by free radicals in the hot culinary fat take place. On the other hand, the initial fat content of the fish and the presence of coating (flour or batter) affect the direction and the extent of fat exchange between the frying medium and food. Batter coating appears to protect fish fillets against moisture loss, oil absorption, and flavor volatiles dilution or loss (Nawar et al., 1990). Pan-frying of fatty fish like salmon and sardines has been shown to result in a slight migration of fatty acids from the fish to the frying oils and fats, which is less
CHAPTER | 79 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying
757
TABLE 79.1 Proximate composition and energy content in the edible portion of raw and pan-fried fish on a fresh weight basis. Finfish
Moisture (%)
Protein (%)
Fat (%)
Energy (kcal 100 g⫺1)
A. boyeri
Raw Pan-fried
76.6c, 82.8a,b 43.0a,b, 57.1c
13.0a,b, 17.2c 20.9c, 25.8a,b
1.4a,b, 2.1c 14.2c, 20.4a,b
71.6a,b, 128.2c 296.0c, 369.6a,b
S. smaris
Raw Pan-fried
71.2c, 75.9a,b 45.4c, 48.5a,b
18.0a,b, 19.0c 21.5a,b, 66.9c
2.6a,b, 3.9c 15.7c, 23.4a,b
108.4a,b, 166.7c 327.8a,b, 389.2c
E. encrasicholus
Raw Pan-fried
74.5c, 77.3a,b 31.2a,b, 49.3c
17.2a,b, 20.5c 26.8c, 32.0a,b
2.1a,b, 2.7c 18.3c, 28.6a,b
112.2a,b, 147.3c 345.8c, 428.7a,b
M. barbatus
Raw Pan-fried
73.0a,b, 79.7c 39.1a,b, 59.8c
14.7c, 17.5a,b 24.4c, 28.7a,b
3.3a,b, 3.5c 12.1c, 21.2a,b
123.1c, 163.5a,b 269.3c, 398.1a,b
B. boops
Raw Pan-fried
78.8c, 80.2a,b 48.7a,b, 64.8c
16.0a,b, 18.2c 22.0a,b, 24.7c
1.3c, 1.6a,b 8.6c, 21.7a,b
91.7a,b, 120.7c 226.6c, 361.8a,b
T. trachurus
Raw Pan-fried
77.5c, 77.9a,b 45.9a,b, 61.5c
16.0c, 17.3a,b 25.7a,b, 72.6c
2.7a,b, 3.8c 11.1c, 19.9a,b
105.5a,b, 138.9c 259.8c, 345.4a,b
M. merluccius
Raw Pan-fried
84.1a,b 58.0a,b
12.7a,b 16.3a,b
2.0a,b 19.1a,b
80.2a,b 263.6a,b
S. pilchardus
Raw Pan-fried
81.0a,b 53.1a,b
13.7a,b 22.7a,b
3.8a,b 19.2a,b
95.5a,b 306.7a,b
S. salar
Raw Pan-fried
64.6d, 65.5e 57.7d, 57.7e
nr nr
9.8e, 15.6d 18.0d, 19.5e
nr nr
a
Kalogeropoulos et al. (2007); Kalogeropoulos et al. (2006); c Kalogeropoulos et al. (2004); d Al-Saghir et al. (2004); e Echarte et al. (2001); nr: not reported. b
extended when using olive oil, with the latter additionally enriching fried fish with MUFA (Mai et al., 1978; SanchezMuniz et al., 1992; Candela et al., 1998; Al Saghir et al., 2004). On the contrary, during frying of relatively lean fish, fatty acid absorption is more dominant than fat migration out of the fish (Mai et al., 1978; Gall et al., 1983; Agren and Hanninen, 1993; Candela et al., 1997), resulting in significant changes of the fried fish fatty acid composition. This was the case of pan-frying lean Mediterranean finfish in VOO (Kalogeropoulos et al., 2004, 2006) which resulted in a 2–5.8 times increment of monounsaturated fatty acids (MUFA) content which became predominant in all fried fish, followed by a respective decrease in SFA and PUFA content (Table 79.2). During recent decades, MUFA have received increasing attention as being potentially beneficial for the reduction of cardiovascular heart disease risk, based on studies in the olive-oil-consuming populations of the Mediterranean basin (Keys et al., 1986). The recent Dietary Guidelines for Americans (2005), published by the U.S. Department of Health and Human Services and the U.S. Department of Agriculture, recommend that most dietary fat should be
derived from sources of PUFA and MUFA, such as fish, nuts, and vegetable oils. The n-6/n-3 ratios of fresh and fried fish samples, also presented in Table 79.2, indicate that the absorption of olive oil with an n-6/n-3 ratio equal to 11.8 (Kalogeropoulos et al., 2006) results in an increment of the ratios in almost all pan-fried fish (Table 79.2). These ratios are in the majority of cases closer to one (Sanchez-Muniz et al., 1992; Echarte et al., 2001; Kalogeropoulos et al., 2006), which is suggested as the desirable value for n-6/n-3 ratios by Simopoulos (2001), as being close to the ratios of essential dietary fatty acids on which human beings initially evolved. In modern Western diets these ratios range between 15 and 16.7 (Simopoulos, 2001) and such high n6/n-3 ratios are considered to promote the pathogenesis of many diseases, including cardiovascular disease (Hu et al., 2002) and diabetes (Hu et al., 2003).
79.2.3 Sterols The cholesterol content of raw and cooked finfish is presented in Table 79.3. Compared with the raw fish, all the
758
SECTION | II General Aspects and Changes in Food Processing
TABLE 79.2 Fatty acid classes (% of total fatty acids) of raw and pan-fried fish. Finfish
SFA
MUFA
PUFA
(n-6)/(n-3)
A. boyeri
Raw Fried
39.1c, 42.1a,b 14.8a,b, 22.1c
32.9a,b, 39.8c 60.2c, 70.4a,b
18.9a,b, 19.2c 13.5,a,b, 16.2c
0.3a,b, 0.5c 1.2a,b, 1.4c
S. smaris
Raw Fried
25.2a,b, 46.9c 14.4a,b, 23.6c
12.5a,b, 30.5c 56.3c, 70.1a,b
20.4c, 52.1a,b 14.9a,b, 15.9c
0.1a,b, 0.2c 0.8c, 0.9a,b
E. encrasicholus
Raw Fried
37.7a,b, 40.4c 15.2a,b, 19.7c
27.2a,b, 28.4c 61.7a,b, 64.1c
29.3c, 30.1a,b 14.4c, 20,5a,b
0.2a,b,c 0.5a,b, 1.0c
M. barbatus
Raw Fried
32.0a,b, 39.8c 14.2a,b, 24.5c
36.5a,b, 42.3c 58.6c, 71.8a,b
15.7c, 26.0a,b 13.0a,b, 14.9c
0.2a,b, 0.4c 1.2c, 1.3a,b
B. boops
Raw Fried
31.6a,b, 45.2c 15.5a,b, 21.5c
16.9a,b, 29.5c 58.4c, 62.2a,b
22.7c, 45,5a,b 18.8c, 21,4a,b
0.1a,b, 0.2 0.4a,b, 0.9c
T. trachurus
Raw Fried
30.4a,b, 34.8c 13.9a,b, 23.2c
24.9a,b, 38.6c 57.0c, 70.5a,b
25.1c, 40,1a,b 14,5a,b, 18.0c
0.1a,b,c 0.6c, 0.9a,b
M. merluccius
Raw Fried
24.4a,b 16.3a,b
17.6a,b 61.7a,b
53,4a,b 20,9a,b
0.1a,b 0.5a,b
S. pilchardus
Raw Fried
38.1a,b, 42.0d 17.9a,b, 20.6d
16.2a,b, 27.3d 59.1a,b, 68.1d
42,4a,b, 30.7d 21,2a,b, 11.3d
0.2a,b,h 0.4a,b, 2.5d
S. salar
Raw Fried
15.9e, 18.0f, 20.9g 16.9e, 19.6f, 21.1g
22.4f, 53.7e, 53.9g 18.8f, 54.6e, 55.0g
30.2c, 58.8f, 25.2g 24.0g, 28.4e, 60.3f
0.2f, 0.3g, 0.6e 0.2f, 0.3g, 0.6e
G. morhua
Raw Fried
23.9g 18.6g
22.9g 59.8g
53.2g 21.6g
0.1g 0.4g
SFA: saturated fatty acids; MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids; a Kalogeropoulos et al. (2007); bKalogeropoulos et al. (2006); cKalogeropoulos et al. (2004); dSanchez-Muniz et al. (1992), deep fried fish; e Al-Saghir et al. (2004); fEcharte et al. (2001); gSioen et al. (2006).
fried samples contain higher cholesterol concentrations on a fresh weight basis. Differences observed in the cholesterol content among different studies of the same fish (Table 79.3) could be attributed to regional and seasonal variations as well as to differences in moisture content and the culinary practice followed. Frying has also been reported to raise the cholesterol content in 11 fish species from the Arabian Gulf (Ewaida, 1993). On a dry weight basis a net decrease of cholesterol in fried fish was observed in the studies of Kalogeropoulos et al. (2004, 2006), in agreement with the findings of Sanchez-Muniz et al. (1992) for sardines fried in olive oil, and it was attributed to the ‘dilution’ effect of the absorbed oil and/or to leaching of cholesterol into the frying oil (Dobarganes et al., 2000). Kalogeropoulos et al. (2006) calculated that 6–13% of the cholesterol initially present in fish migrated to the frying oil. Likewise, the decrease of cholesterol in fried salmon was attributed to the absorption of vegetable oil and/or the oxidation process taking place during frying (Echarte et al., 2001; Al-Saghir et al., 2004).
The sum of the phytosterols, namely β-sitosterol, campesterol and stigmasterol, content of eight Mediterranean finfish after pan frying in VOO (Kalogeropoulos et al., 2006) is also presented in Table 79.3. β-Sitosterol, being predominant among olive oil phytosterols, comprised the 79–91% of phytosterols determined, while no plant sterols were detected in fresh fish. Frying resulted in phytosterol enrichment of fried fish, their concentrations being proportional to the amount of olive oil absorbed. Based on the phytosterol retentions in the frying oil and their overall retention in fried oils and fish it was proposed that the fate of phytosterols absorbed by the food or remaining in the frying oil is similar.
79.2.4 Squalene In the Mediterranean countries squalene intake is ten times higher than in northern European countries or the United States, a fact attributed to olive oil consumption (Smith, 2000; Covas et al., 2006).
CHAPTER | 79 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying
759
TABLE 79.3 Cholesterol, phytosterols and squalene content in the edible portion of raw and pan-fried fish, on a fresh weight basis (mg 100 g⫺1). Fish species
Cholesterol
Phytosterolse
Squalene
A. boyeri
Raw Pan-fried
60.9a,b, 45.8c 70.8c, 99.6a,b
nda,b 29.7a,b
0.6a,b, 1.4c 36.4c, 49.6a,b
S. smaris
Raw Pan-fried
45.1a,b, 41.3c 55.5c, 97.8a,b
nda,b 24.0a,b
0.7a,b, 1.3c 49.6c, 82.5a,b
E. encrasicholus
Raw Pan-fried
34.5c, 44.5a,b 50.1c, 59.5a,b
nda,b 17.3a,b
0.2a,b, 1.2c 85.5a,b, 94.1c
M. barbatus
Raw Pan-fried
27.4c, 36.1a,b 42.9c, 64.1a,b
nda,b 23.2a,b
1.0c, 2.1a,b 38.9c, 68.8a,b
B. boops
Raw Pan-fried
25.3c, 37.9a,b 32.9c, 72.5a,b
nda,b 19.3a,b
0.5a,b, 1.0c 27.3c, 62.1a,b
T. trachurus
Raw Pan-fried
34.2c, 41.9a,b 39.7c, 82.8a,b
nda,b 18.7a,b
0.8a,b, 1.1c 22.0c, 66.1a,b
M. merluccius
Raw Pan-fried
37.1a,b 51.4a,b
nda,b 16.4a,b
0.4a,b 64.2
S. pilchardus
Raw Pan-fried
30.6a,b 44.5a,b
nda,b 13.3a,b
1.3a,b 81.6a,b
S. salar
Raw Pan-fried
53.3d 70.0d
nr nr
nr nr
a
Kalogeropoulos et al. (2007); bKalogeropoulos et al. (2006); Kalogeropoulos et al. (2004); dEcharte et al. (2001); e sum of β-sitosterol, campesterol and stigmasterol content; nd: not detected; nr: not reported. c
Squalene content of finfish before and after pan frying in VOO has been reported in two studies of Kalogeropoulos et al. (2004, 2006). In these studies fresh VOO squalene content was reported as 495 and 616 mg 100 g⫺1, respectively. As raw finfish contained only minute amounts of squalene (0.2–2.1 mg 100 g⫺1 fw), the significant increase observed in pan-fried samples (Table 79.3) was obviously the result of olive oil absorption. The overall squalene retentions based on the squalene content of both oil and food before and after frying ranged from 40 to 69% (Kalogeropoulos et al., 2006) being similar to squalene retentions in frying oils. Therefore the fate of squalene absorbed by the food must be similar to that of squalene remaining in the frying oil.
79.2.5 Vitamin E Vegetable oils, fish, and nuts are among the major dietary sources of vitamin E. The presence of relatively high amounts of tocopherols in foods which – like fish – are rich in polyunsaturated fatty acids is beneficial for human consumption, as such unsaturated diets increase the
peroxidizability of the lipids and reduce the time required to develop symptoms of vitamin E deficiency, leading to increased requirements for vitamin E in order to prevent tissue PUFA oxidation. As a result, a ratio of at least 0.6 mg α-tocopherol equivalent per g PUFA intake is suggested (Valk and Hornstra, 2000). In the study of Al-Saghir et al. (2004) a non-significant decrease of α-, γ-, and δ-tocopherol was reported during pan frying of salmon fillets in olive oil. However, in the case of lean and low fat fish a 1–2 orders of magnitude enrichment of fried fish with α-tocopherol during pan-frying in VOO was reported (Table 79.4), with α-tocopherol recoveries in frying oils being almost identical with the respective overall α-tocopherol recovery (Kalogeropoulos et al., 2007). In the latter case, when the α-tocopherol remaining in the frying oil was compared with that absorbed by the fried food (quotient A/F, Table 79.4) results indicated that, perhaps with the exception of E. encrasicholus, the composition of the oil remaining in the frying pan was similar to that of the oil absorbed by the food. Thus, no specific absorption of tocopherols towards fried food seems to occur. This finding is consistent
760
SECTION | II General Aspects and Changes in Food Processing
TABLE 79.4 α-Tocopherol contenta of raw and pan-fried finfish in virgin olive oil (mg/100 g) on a fresh weight basis; Quotient A/Fb and recoveries (%) of α-tocopherol. αTocopherol
A. boyeri
E. encrasicholus
S. smaris
Raw fish
0.06
0.08
0.03
Pan-fried fish
1.93
2.78
2.79
A/F
0.8
1.8
Recovery in fried oil
28.8
Overall recoveryc
28.4
M. barbatus
nd
B. boops
T. trachurus
M. merluccius
S. pilchardus
0.10
0.03
0.01
0.03
2.97
2.92
2.29
2.36
2.61
1.0
1.0
0.8
1.0
0.8
1.3
39.0
53.5
80.8
76.2
51.2
77.9
65.9
43.8
53.3
80.5
74.0
51.1
74.8
67.7
a
Kalogeropoulos et al. (2007); A/F: calculated as the quotient of α-tocopherol concentrations in the oil absorbed by finfish (A, mg/100 g absorbed oil) and the respective concentration in the fried oil (F, mg 100 g⫺1 fried oil). Absorbed oil was calculated from the weights of oil and food before and after frying and its α-tocopherol content was corrected by subtracting the α-tocopherol present in fresh food; c Calculated from α-tocopherol present in both food and oil, before and after pan-frying; nd: not detected. b
with findings regarding commercial frying of potatoes (Dobarganes et al., 2000). Given the amounts of n-3 PUFA (0.8–10.3 g) and α-tocopherol (2.5–3.9 mg) provided by consuming a serving of pan-fried fish (Table 79.5), α-tocopherol appears to be sufficient to protect 4–6.5 g of n-3 PUFA, surpassing in most cases the respective amount of PUFA.
79.2.6 Polyphenols and Terpenic Acids Polyphenols are widely distributed phytochemicals. In the case of olive oil, a range of phenolic compounds provides some of its health benefits. Extra VOO has higher phenolic content than refined olive oil (Waterman and Lockwood, 2007). The most abundant simple triterpenes in olive oil are oleanolic and maslinic acids, together with erythrodiol and uvaol alcohol (Covas et al., 2006). Kalogeropoulos et al. (2007) studied the recovery and distribution of polyphenols and other natural antioxidants during the pan-frying of eight Mediterranean finfish in VOO. Nine individual polyphenols were determined in VOO, namely tyrosol, hydroxytyrosol, homovanillic alcohol, phydroxy-benzoic acid, p-hydroxy-phenylacetic acid, vanillin, vannilic acid, ferulic acid and p-coumaric acid, with tyrosol comprising more than 70% of the determined polyphenols. Additionally, significant amounts of three hydroxy pentacyclic triterpene acids (HPTA), i.e., oleanolic, maslinic, and ursolic acids were also present in VOO. As expected, raw fish did not contain any detectable amounts of polyphenols and HPTA. Pan frying resulted in an enrichment of fried
fish with VOO-originating polyphenols, with tyrosol predominating, as in the case of olive oil (Figure 79.1). On the other hand, pan frying resulted in a significant loss of oil polyphenols (48–64%), due to antioxidant deterioration and/or migration to the food tissue. The overall polyphenol retention (average 73%), calculated from the amounts present in the oil and the food before and after frying, was in all cases higher than the respective retention in the frying oils (average 46%). This better survival of polyphenols when absorbed by finfish than when remaining in the frying pan, was clearly demonstrated when polyphenol amounts in fish were expressed as mg 100 g⫺1 of absorbed oil and compared with their respective concentrations in the frying oils (Figure 79.2). In the absorbed oil, polyphenol concentration was 3.5–8 times higher than the respective concentration in the frying oil (Figure 79.2), a distribution attributed either to a better survival of polyphenols from the frying procedure when absorbed by the food or to a diffusion of the relatively polar polyphenols towards the water-rich food tissue. The latter possible explanation is supported by several studies (Sacchi et al., 2002; Rodis et al., 2002, Chiou et al., 2007). Pan-frying resulted also in a partial loss of HPTA from oils and an enrichment of fried fish with HPTA (Figure 79.3). HPTA concentrations in pan-fried fish were proportional to the amount of oil absorbed, as in the case of polyphenols, while the overall retentions of HPTA (average 71%) were slightly lower than their recoveries in fried olive oils (average 76%), opposite to the case of polyphenols. Moreover, HPTA concentrations in fried
Fish species
Protein (g)
Fat (g)
Energy (kcal)
MUFA (g)
A. boyeri
33.5
26.5
481
17.3
1.4
S. smaris
28.0
30.4
426
17.4
E. encrasicholus
41.6
37.2
557
M. barbatus
37.3
27.6
B. boops
28.6
T. trachurus
Cholesterol (mg)
Plant sterols (mg)
α-Tocopherol (mg)
Polyphenols (mg)
Terpenic acids (mg)
65
130
38.9
2.5
2.8
16.4
3.8
107
127
35.6
3.6
2.7
6.5
9.6
10.3
111
77
23.9
3.6
1.3
4.4
518
18.3
1.4
89
83
30.2
3.9
2.5
9.2
28.2
470
18.9
0.8
81
94
29.5
3.8
2.4
7.9
33.4
25.9
449
16.9
1.8
86
108
26.1
3.0
2.3
8.6
M. merluccius
21.2
24.8
343
14.2
3.1
84
67
23.1
3.1
0.9
3.8
S. pilchardus
29.5
25.0
399
13.7
3.5
106
58
19.6
3.4
1.6
8.8
VOO: virgin olive oil; a one serving ⫽ 130 g fried fish; based on data from Kalogeropoulos et al. (2006, 2007).
b
n-3 PUFA (g)
Squalene (mg)
CHAPTER | 79 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying
TABLE 79.5 Dietary intake of macro- and microconstituents after consuming a servinga of finfish pan-fried in VOOb.
761
762
SECTION | II General Aspects and Changes in Food Processing
2.5
I
II
mg 100 g−1
2.0
1.5
1.0
0.5
S. pilchardus
M. merluccius
T. trachurus
B. boops
M. barbatus
S. smaris
A. boyeri
E.encrasicholus
0.0
FIGURE 79.1 Total polyphenols and tyrosol content (mg 100 g⫺1) on a fresh weight basis, in the edible portion of finfish after pan-frying in VOO. I: sum of nine polyphenols in fried fish; II: tyrosol in fried fish.
Polyphenol(s) (g 100 g−1 oil)
10
I
II
8 6 4 2
S. pilchardus
M. merluccius
T. trachurus
B. boops
M. barbatus
S. smaris
E. encrasicholus
A. boyeri
0
FIGURE 79.2 Distribution of total polyphenols and tyrosol in the fried olive oil and in the oil absorbed by the fried fish. Absorbed oil was calculated by the weights of oil and food before and after frying. I: Sum of nine polyphenols in fried olive oil; II: Sum of nine polyphenols in absorbed olive oil.
oils were higher than that in the oils absorbed by the food (Figure 79.4). Without excluding the possibility of a higher HPTA loss in the tissue of fried fish, the observed distribution was attributed to the lipophilic nature of HPTA, keeping them in the oil phase rather than migrating towards the water-rich fish tissue. This theory was further supported by the relative
HPTA abundances in the fried oils and in fried fish, i.e., lower concentrations of the more lipophilic maslinic acid were observed than those expected from its abundance in frying oils. Among the HPTA studied, maslinic acid is the only one containing two adjacent hydroxyl groups, participating in endomolecular hydrogen bond formation, and thus reducing its hydrophilic potency (Kalogeropoulos et al., 2007).
763
CHAPTER | 79 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying
I
12
II
mg 100 g−1
10 8 6 4 2
S. pilchardus
M. merluccius
T. trachurus
B. boops
M. barbatus
A. boyeri
E. encrasicholus
S. smaris
0
FIGURE 79.3 Sum of the hydroxy pentacyclic triterpene acids (HPTA) and oleanolic acid content (mg 100 g⫺1) on a fresh weight basis, in the edible portion of finfish after pan-frying in VOO. I: Oleanolic acid; II: Total HPTA.
Sum of three HPTA (g 100 g−1 oil)
60
I
II
50 40 30 20 10
S. pilchardus
M. merluccius
T. trachurus
B. boops
M. barbatus
S. smaris
E. encrasicholus
A. boyeri
0
FIGURE 79.4 Distribution of total HPTA in the fried olive oil and in the oil absorbed by the fried fish. Absorbed oil was calculated by the weights of oil and food before and after frying. I: Total HPTA in fried olive oil; II: Total HPTA in absorbed olive oil.
79.3 NUTRITIONAL EVALUATION OF PAN-FRIED FISH The dietary intake of several macro- and microconstituents provided after the consumption of one serving – 130 g – of fish pan-fried in VOO is presented in Table 79.5. Kalogeropoulos et al. (2006, 2007) calculated that the consumption of one serving of fried fish could cover a significant fraction of the daily intake of protein, MUFA, squalene and natural antioxidants like α-tocopherol, polyphenols and teprenic acids. It could also provide 8–15% of the typical daily phytosterols intake, and energy covering
17–26% of a 2000 kcal day⫺1 diet. Keeping in mind that in the Mediterranean cuisines fried fish are usually served together with fresh salads or boiled greens, i.e., food of low fat content, a typical meal containing fried fish is not expected to supply excess fat and energy. In regard to cholesterol, the consumption of one serving of fried fish is expected to contribute 19–43% (average 31%) to the recommended maximum intake of 300 mg per day (Dietary Guidelines for Americans, 2005). The intake of n-3 PUFA by consuming a serving of pan-fried fish, presented in Table 79.5, surpasses the recommended 650 mg day⫺1 of PUFA for healthy adults, which corresponds to 4–5 fish servings per week (Simopoulos et al., 1999).
764
SECTION | II General Aspects and Changes in Food Processing
Given the seasonal and regional variability of fish composition – mainly fat content – these values should be considered as indicative only.
SUMMARY POINTS ●
●
●
●
●
●
●
Domestic pan-frying of Mediterranean fish in VOO causes significant changes in their crude composition, i.e., water loss and increased fat, protein and energy content. The use of VOO for frying fish is beneficial for the fatty acid profile of fried products, as they contain less SFA and more MUFA, while still providing significant amounts of n-3 PUFA, having n-6/n-3 ratios close to 1. Virgin olive oil absorption by fried fish enriches them with health-promoting microconstituents like squalene and phytosterols, together with natural antioxidants (tocopherols, polyphenols and terpenic acids), a significant fraction of which survives frying, thus becoming part of our diet. Differences in the antioxidants polarity seem to govern – to some extent – their distribution between the frying oil and the water-rich fish. Pan-fried fish do not overload the consumer with cholesterol, while part of the fish cholesterol seems to leach into the frying oil, remaining in the fryer. A serving of fish pan-fried in VOO provides a significant portion of the daily intakes of protein, oleic acid, n-3 PUFA, vitamin E, polyphenols, terpenic acids, squalene and plant sterols. Based on the above, pan-fried fish – as part of a balanced diet – have a place in our diets, assuming that frying is performed in unused VOO.
REFERENCES Agren, J.J., Hanninen, O., 1993. Effects of cooking on the fatty acids of three fresh-water fish species. Food Chem. 46, 377–382. Al-Saghir, S., Thurner, K., Wagner, K.H., Frisch, G., Luf, W., RazzaziFazeli, E., Elmadfa, I., 2004. Effects of different cooking procedures on lipid quality and cholesterol oxidation of farmed salmon fish (Salmo salar). J. Agric. Food Chem. 52, 5290–5296. Bandarra, N.M., Batista, I., Nunes, M.L., Empis, J.M., Christie, W.W., 1997. Seasonal changes in lipid composition of sardine (Sardina pilchardus). J. Food Sci. 62, 40–42. Banks, D., 1996. Introduction. In: Perkins, E.G., Erickson, M.D. (Eds.), Deep Frying: Chemistry, Nutrition, and Practical Applications. AOCS Press, Champaign, Ill, pp. 1–3. Berger, A., Jones, P.J.H., Abumweis, S.S., 2004. Plant sterols: factors affecting their efficacy and safety as functional food ingredients. Lipids Health Dis. 3:5, accessed on-line at http://www.lipidworld. com/content/3/1/5.
Bognár, A., 1998. Comparative study of frying to other cooking techniques influence on the nutritive value. Grasas Aceites 49, 250–260. Brouwer, I.A., Geelen, A., Katan, M.B., 2006. n-3 Fatty acids, cardiac arrhythmia and fatal coronary heart disease. Prog. Lipid Res. 45, 357–367. Calder, P.C., 2005. Polyunsaturated fatty acids and inflammation. Prostag. Leukotr. Ess. 75, 197–202. Candela, M., Astiasaran, I., Bello, J., 1997. Effects of frying and warm holding on fatty acids and cholesterol of sole (Solea solea), codfish (Gadus morhua) and hake (Merluccius merluccius). Food Chem. 58, 227–231. Candela, M., Astiasaran, I., Bello, J., 1998. Deep-fat frying modifies highfat fish lipid fraction. J. Agric. Food Chem. 46, 2793–2796. Chiou, A., Salta, F.N., Kalogeropoulos, N., Mylona, A., Ntalla, I., Andrikopoulos, N.K., 2007. Retention and distribution of polyphenols after pan-frying of French-fries in oils enriched with olive leaf extract. J. Food Sci. 72, S574–S584. Covas, M.I., 2007. Olive oil and the cardiovascular system. Pharmacol. Res. 55, 175–186. Covas, M.I., Ruiz-Gutierrez, V., de la Torre, R., Kafatos, A., LamuelaRaventos, R.M., Osada, J., Owen, R.W., Visioli, F., 2006. Minor components of olive oil: evidence to date of health benefits in humans. Nutr. Rev. 64, S20–S30. DAFNE, Data Food Networking, http://www.nut.uoa.gr/Dafnesoftweb/ Main.aspx (Last accessed 19 April 2008). Dietary Guidelines for Americans, 2005. U.S. Department of Health and Human Services and U.S. Department of Agriculture. Published on the World Wide Web: http://www.healthierus.gov/dietaryguidelines (Last accessed 19 April 2008). Dobarganes, C., Márquez-Ruiz, G., Velasco, J., 2000. Interactions between fat and food during deep-frying. Eur. J. Lipid Sci. Tech. 102, 521–528. Ewaida, E.H., 1993. Cholesterol, fat and food energy content of selected raw and cooked commercial fish species from the Arabian Gulf. Ecol. Food Nutr. 30, 283–292. Fillion, L., Henry, C.J.K., 1988. Nutrient losses and gains during frying: a review. Int. J. Food Sci. Nutr. 49, 157–168. Gall, K.L., Otwell, W.S., Koburger, J.A., Appledorf, H., 1983. Effects of four cooking methods on the proximate, mineral and fatty acid composition of fish fillets. J. Food Sci. 48, 1068–1074. Hu, F.B., Bronner, L., Willett, W.C., Stampfer, M.J., Rexrode, K.M., Albert, C.M., Hunter, D., Mansonm, J.E., 2002. Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. J. Am. Med. Assoc. 287, 1815–1821. Hu, F.B., Cho, E., Rexrode, K.M., Albert, C.M., Manson, J.E., 2003. Fish and long-chain omega-3 fatty acid intake and risk of coronary heart disease and total mortality in diabetic women. Circulation 107, 1852–1857. Kalogeropoulos, N., Chiou, A., Mylona, A., Ioannou, M.S., Andrikopoulos, N.K., 2007. Recovery and distribution of natural antioxidants (αtocopherol, polyphenols and terpenic acids) after pan-frying of Mediterranean finfish in virgin olive oil. Food Chem. 100, 509–517. Kalogeropoulos, N., Hassapidou, M., Andrikopoulos, N.K., 2004. Dietary evaluation of Mediterranean fish and mollusks pan-fried in virgin olive oil. J. Sci. Food Agric. 84, 1750–1758. Kalogeropoulos, N., Kotsiopoulou, C., Mylona, A., Christea, M., Andrikopoulos, N.K., 2006. Dietary evaluation of vegetables panfried in virgin olive oil following the Greek traditional culinary practice. Ecol. Food Nutr. 45, 171–188.
CHAPTER | 79 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying
Keys, A., Menotti, A., Karvonen, M.J., Aravanis, C., Blackburn, H., Buzina, R., Djordjevic, B.S., Dontas, A.S., Fidanza, F., Keys, M.H., Kromhout, D., Nedeljkovic, S., Punsar, S., Seccareccia, F., Toshima, H., 1986. The diet and 15-year death rate in the Seven Countries Study. Am. J. Epidemiol. 124, 903–915. Kelly, G.S., 1999. Squalene and its potential clinical use. Alt. Med. Rev. 4, 29–36. Kołakowska, A., Olley, J., Dunstan, G.A., 2003. Fish Lipids, Chapt. 12. In: Sikorski, Z.E., Kołakowska, A. (Eds.), Chemical and Functional Properties of Food Lipids. CRC Press, Boca Raton, pp. 221–264. Kulas, E., Ackman, R.G., 2001. Different tocopherols and the relationship between two methods for determination of primary oxidation products in fish oil. J. Agric. Food Chem. 49, 1724–1729. Law, M.R., 2000. Plant sterol and stanol margarines and health. Br. Med. J. 320, 861–864. Liu, J., 1995. Pharmacology of oleanolic acid and ursolic acid. J. Ethnopharmacol. 49, 57–68. Mai, J., Shimp, J., Weihrauch, J., Kinsella, J.E., 1978. Lipids of fish fillets – changes following cooking by different methods. J. Food Sci. 43, 1669–1674. Nawar, W., Hultin, H., Li, Y.J., Xing, Y.H., Kelleher, S., Wilhem, C., 1990. Lipid oxidation in seafoods under conventional conditions. Food Rev. Int. 6, 647–660. Pérez-Camino, M.C., Cert, A., 1999. Quantitative determination of hydroxy pentacyclic triterpene acids in vegetable oils. J. Agric. Food Chem. 47, 1558–1562. Perez-Jimenez, F., Ruano, H., Perez-Martinez, P., Lopez-Segura, F., Lopez-Miranda, J., 2007. The influence of olive oil on human health: not a question of fat alone. Mol. Nutr. Food Res. 51, 1199–1208. Rao, C.V., Newmark, H.L., Reddy, B.S., 1998. Chemopreventive effect of squalene on colon cancer. Carcinogenesis 19, 287–290. Rodis, P.S., Karathanos, V.T., Mantzavinou, A., 2002. Partitioning of olive oil antioxidants between oil and water phases. J. Agric. Food Chem. 50, 596–601. Sacchi, R., Paduano, A., Fiore, F., Della Medaglia, D., Ambrosino, M.L., Medina, I., 2002. Partition behavior of virgin olive oil phenolic com-
765
pounds in oil-brine mixtures during thermal processing of fish canning. J. Agric. Food Chem. 50, 2830–2835. Saguy, I.S., Dana, D., 2003. Integrated approach to deep fat frying: engineering, nutrition, health and consumer aspects. J. Food Eng. 56, 143–152. Sanchez-Muniz, F.J., Viejo, J.M., Medina, R., 1992. Deep-frying of sardines in different culinary fats. Changes in the fatty acid composition of sardines and frying fats. J. Agric. Food Chem. 40, 2252–2256. Schneider, C., 2005. Chemistry and biology of vitamin E. Mol. Nutr. Food Res. 49, 7–30. Simopoulos, A.P., 2001. The Mediterranean diets: what is so special about the diet of Greece? The scientific evidence. J. Nutr. 131, 3065S–3073S. Simopoulos, A.P., Leaf, A., Salem, N., 1999. Essentiality of and recommended dietary intakes for omega-6 and omega-3 fatty acids. Ann. Nutr. Metab. 43, 127–130. Sioen, I., Haak, L., Raes, K., Hermans, C., De Henauw, S., De Smet, S., Van Camp, J., 2006. Effects of pan-frying in margarine and olive oil on the fatty acid composition of cod and salmon. Food Chem. 98, 609–617. Smith, T.J., 2000. Squalene: potential chemopreventive agent. Expert Opin. Inv. Drug. 9, 1841–1848. Traber, M.G., Atkinson, J., 2007. Vitamin E, antioxidant and nothing more. Free Rad. Biol. Med. 43, 4–15. Valk, E.E., Hornstra, G., 2000. Relationship between vitamin E requirement and polyunsaturated fatty acid intake in man: a review. Int. J. Vitam. Nutr. Res. 70, 31–42. Varela, G., Ruiz-Roso, B., 1992. Some effects of deep-frying on dietary fat intake. Nutr. Rev. 50, 256–262. Visioli, F., Bellomo, G., Galli, C., 1998. Free radical scavenging properties of olive oil polyphenols. Biochem. Biophys. Res. Commun. 247, 60–64. Waterman, E., Lockwood, B., 2007. Active components and clinical applications of olive oil. Alt. Med. 12, 331–341. Xu, H., Zeng, F., Wan, M., Sim, K., 1996. Anti-HIV triterpene acids from Geum japonicum. J. Nat. Prod. 59, 643–645.
This page intentionally left blank
Chapter 80
Recovery and Distribution of Macro- and Selected Microconstituents after Panfrying of Vegetables in Virgin Olive Oil Nick Kalogeropoulos Laboratory of Chemistry-Biochemistry-Physical Chemistry of Foods, Department of Science of Dietetics-Nutrition, Harokopio University, Athens, Greece
80.1 INTRODUCTION 80.1.1 Frying Frying is an ancient cooking technique, used as early as 1600 BC by the ancient Egyptians and later by the Greeks and the Romans (Morton, 1998; Banks, 1996). Nowadays, frying is considered as an inexpensive, fast and efficient method for cooking and food surface sterilization, which additionally improves food’s sensory quality by the formation of aroma compounds, attractive color, crust, and texture (Pinthus et al., 1995). As a result of their good palatability, fried foods are consumed with sustainable popularity, despite their considerable fat content and the consumers’ awareness of the relationships between food, nutrition and health (Saguy and Dana, 2003). Frying is usually performed as deep (or immersed) frying, and pan (or shallow) frying and requires a cooking oil or fat as a heat transfer medium. Research has indicated that, contrary to common belief, frying appears to have the same or even less effect on nutrient and vitamin losses compared with other cooking methods (Fillion and Henry, 1988; Bognár, 1998; Saguy and Dana, 2003). When virgin olive oil is used as frying oil, absorption of oleic acid and olive oil bioactive microconstituents is expected to further enhance the nutritional and health benefits of fried food.
80.1.2 Mediterranean Diet The central elements of Mediterranean diet are variety, high fiber content, n-3 fatty acids, as well as phytochemicals from olive oil, legumes, whole grains, fruits, and vegetables. This dietary pattern, besides its well-established preventive impact on cardiovascular health, can also help Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
fight diseases related to chronic inflammation, including visceral obesity, type II diabetes, and the metabolic syndrome (Esposito et al., 2007; Kontogianni and Panagiotakos, 2007; Giugliano and Esposito, 2008), and have a preventive role against several dietary-related forms of cancer like large bowel, breast, endometrium, and prostate cancer (Trichopoulou et al., 2000; Hashim et al., 2005; StacewiczSapuntzakis et al., 2008). The average daily intake of vegetables in Mediterranean European countries is estimated to range between 121–284 g person–1 (DAFNE, 2008).
80.1.3 Virgin Olive Oil Virgin olive oil (VOO) is produced by direct pressing or centrifugation of the olive fruit, without any further refining. The use of olive oil as the primary source of fat in the Mediterranean diet provides benefits extending beyond a mere reduction in the low-density lipoprotein cholesterol, and affects lipoprotein metabolism, oxidative damage, inflammation, endothelial dysfunction, blood pressure, thrombosis, and carbohydrate metabolism. Increased olive oil intake could explain the low rate of cardiovascular mortality found in Southern European Mediterranean countries, in comparison with other Western countries, despite a high prevalence of coronary heart disease risk factors (Covas, 2007). The nutritional and health importance of VOO is attributed to its high content of the peroxidation-resistant oleic acid, to a balanced contribution quantity of polyunsaturated fatty acids, to its richness in phenolic compounds which act as natural antioxidants and may contribute to the prevention of several human diseases (Bendini et al., 2007), and to the presence of α-tocopherol and substantial amounts of other
767
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
768
SECTION | II General Aspects and Changes in Food Processing
compounds deemed to be anticancer agents like squalene and terpenoids (Owen et al., 2004).
80.1.4 Cooking with Olive Oil As regards lipid intake, the Mediterranean diet is characterized by: (a) lesser intake of saturated fatty acids and higher intake of unsaturated fatty acids – mainly monounsaturated and n-3 polyunsaturated fatty acids, due to the consumption of olive oil and fish – and (b) the fact that only a small fraction of the fat is eaten raw (as dressings), while the greatest proportion is used in thermal culinary processes, mainly frying, and also canning and stewing in oil (Varela and Ruiz-Roso, 1992). In the Mediterranean diet about 50% of total fat intake is derived not from the food itself but rather from the cooking fat (Varela and Ruiz-Roso, 1992). Pan-frying of vegetables (like potatoes, green peppers, zucchinis, eggplants) in olive oil is a common practice in the Mediterranean olive-oil-producing countries. Vegetables are normally fried uncoated or coated with flour or batter and the fried products are either served as starters or used as ingredients of other Mediterranean recipes. The major process contributing to the instability of olive oil when heated is lipid oxidation, which can lead to significant changes in the oil composition that affect its biological properties, as lipid peroxidation products have been linked to cancer and cardiovascular disease (Harwood and Yaqoob, 2002). Although antioxidants protect olive oil from thermal degradation, frying reduces the oil antioxidative capacity (Quiles et al., 2002), a particularly important fact when the same oil is used repeatedly. Comparing frying oils and frying techniques, the combination of virgin olive oil and deep-frying has been shown to be the better choice, as olive oil has a relatively long deep-fat frying ‘shelf life’ and is comparatively more stable than other oils for repeated frying (Bastida and Sánchez-Muniz, 2001; Andrikopoulos et al., 2002a, b; Harwood and Yaqoob, 2002). Deep frying is considered less stressful for the oil and food, as contact with atmospheric oxygen is minimized, and it is preferably used for commercial frying. However, as the oil is re-used several times, accumulation of polymeric compounds occurs as the oil antioxidant capacity is reduced (Harwood and Yaqoob, 2002; Galeone et al., 2007). Therefore, it seems that the traditional Mediterranean practice of pan-frying in virgin olive oil that has not been used for more than 1 or 2 times, appears more reasonable for domestic cooking.
80.2 COMPOSITIONAL CHANGES IN VEGETABLES AFTER FRYING 80.2.1 Macroconstituents During frying, physical and chemical interactions between frying fats or oils and fried foods take place, which result
in changes of frying oils and modifications of fried food (Dobarganes et al., 2000). Oil absorption, water loss and lipid exchange between frying oil and food are the main physical changes involved, while chemical reactions include interactions between food constituents and oxidized lipids, hydrolysis of frying fats due to food moisture and oil thermal and oxidative decomposition (Dobarganes et al., 2000; Warner, 2002). These chemical reactions and their products can be kept minimal if frying is performed under household or household-like conditions, avoiding overheating, and employing good-quality oils which have not been used for more than 2–3 frying sessions (Andrikopoulos et al., 2002a, b; Salta et al., 2008).
80.2.2 Oil Uptake and Water Loss Frying is considered as a dehydration process with the following characteristics: (a) high oil temperature (160– 180 °C) that enables rapid heat transfer and a short cooking time, (b) product temperature (except for the crust region) does not exceed 100 °C, (c) water-soluble compound leaching is minimal (Saguy and Dana, 2003). The main compositional changes caused during frying of vegetables are water loss due to evaporation and oil uptake, accompanied by the corresponding increase of energy content. Oil uptake by fried vegetables is governed by many factors including oil quality, frying temperature and duration, product shape and texture, initial moisture and fat content, existence of coating, and crust physical properties (Pinthus et al., 1995; Dobarganes et al., 2000). Oil uptake can be described as a replacement between oil and the evaporated water. This can be seen in Figure 80.1, showing the positive (exponential) correlation between water loss and oil uptake during the pan-frying of potatoes, green peppers, zucchinis and eggplants in VOO (data from Kalogeropoulos et al., 2006); after frying, oil content increased from 50 times in zucchini and green pepper to 360 times in eggplant and differences in water loss and oil uptake were attributed to the texture of the vegetables and the presence of coating (Kalogeropoulos et al., 2006).
80.2.3 Energy Content Oil uptake and water loss are accompanied by an increase of energy content. Kalogeropoulos et al. (2006) reported an increment of energy content ranging from 3 to 18 times after pan-frying of vegetables in VOO. Flour and batter cover increased the energy content in the case of zucchini, while the opposite was true for eggplants. Despite the awareness of the Western consumer towards limiting the intake of calories originating from fat, it should be kept in mind that, in some developing countries, the fried food contribution is of great importance due to the possible deficiency in nutritional components as well as calorie intake (Saguy and Dana, 2003).
CHAPTER | 80 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying of Vegetables
769
Oil content (g 100 g−1 fw)
40 y = 2.3372e0.0483x R2 = 0.7836
30
eggplant eggplant floured
20
eggplant battered zucchini floured
10
zucchini battered potato
zucchini green pepper 0 0
10
20
30 40 Water loss (g 100 g−1 fw)
50
60
FIGURE 80.1 Water loss and oil content of vegetables pan-fried in VOO.
80.2.4 Fatty Acids As vegetables contain only minute amounts of fat, not exceeding 0.1–0.2% fw, they tend to absorb oil during frying, therefore the lipid profiles of fried products are essentially those of the frying oils or fats. When VOO is used as frying oil, the fried vegetables become rich in monounsaturated fatty acids which have been proved to be potentially beneficial for cardiovascular heart disease risk reduction (Kris-Etherton et al., 1999). The oil content of fried vegetables ranges normally from 7–14% in French fries to 34.6% in potato chips (Saguy and Dana, 2003). The effect of a single pan-frying session on olive oil’s fatty acid composition is reflected in changes of the n-6/n-3 ratios in fried oils, which were slightly elevated in oils used for frying vegetables (12.1-13.7) compared to unused VOO (n-6/n-3 5 11.8) as a result of the higher susceptibility to oxidation of the more unsaturated n-3 fatty acids (Kalogeropoulos et al., 2006).
80.2.5 Microconstituents 80.2.5.1 Vitamins Frying has significant advantages over other cooking methods: the temperature within the product is below 100°C, short frying time is achieved, and insolubility of watersoluble vitamins results in less deterioration of heat-sensitive vitamins, as compared with baking or boiling (Bognár, 1998; Saguy and Dana, 2003). Retentions of vitamin C as high as 75–96% in fried potatoes have been reported (Fillion and Henry, 1998). Additionally, as all vegetable oils used for frying contain vitamin E, fried foods are enriched with considerable amounts of the vitamin due to oil uptake. Tocopherols are considered as the most important lipidphase natural antioxidants, which prevent lipid peroxidation in membranes and lipoprotein particles in mammalian cells (Schneider, 2005; Traber and Atkinson, 2007). Fresh
vegetables contain minute amounts of tocopherols, not exceeding 1 mg 100 g⫺1 (Table 80.1), while after frying in VOO a 6–43 times increase of α-tocopherol in fried food was observed, as a result of VOO uptake (Table 80.1). α-Tocopherol retentions in fried oils have been reported to range from 23.7–65.1% (average 43.3%), and the respective overall retentions ranged from 32.3–63.6% (average 45.7%), being higher when frying eggplants, regardless of the culinary practice employed (Figure 80.2).
80.2.5.2 Polyphenols Polyphenols are known to possess antioxidant activity with respect to oxidative alterations and have been associated with lower risk of coronary heart disease, some types of cancer and inflammation. Recently, Covas et al. (2006) reviewed the antioxidant effect of olive oil phenolic compounds in humans, while in a crossover controlled trial Gimeno et al. (2007) reported that dietary-supplied olive oil phenolic compounds enriched LDL in human subjects, confirming their role as antioxidants in vivo. In traditional Mediterranean cooking, olive oil is conventionally boiled or heated. During frying, the virgin olive oil phenolic compounds deteriorate at different rates, protecting both oil and food from oxidative damage. For conventional heating, a time-dependent effect is observed, with the phenolic content being reduced as heating time increases (Andrikopoulos et al., 2002a; Brenes et al., 2002). Individual phenols behave differently to conventional heating. Brenes et al. (2002), for example, reported that hydroxytyrosol levels decrease rapidly, as lignans do, though at slower rate, while Andrikopoulos et al. (2002a) reported that during eight successive deep- and pan-frying of potatoes, tannic acid, oleuropein and hydroxytyrosolelenolic acid dialdeydic form showed remarkable resistance, with hydroxytyrosol and hydroxytyrosol-elenolic acid being eliminated faster. Also, Gómez-Alonso et al. (2003)
770
SECTION | II General Aspects and Changes in Food Processing
.
TABLE 80.1 Selected microconstituents content (mg 100 g⫺1) of fresh VOO and raw and pan-fried vegetables in VOO (data from Kalogeropoulos et al., 2006, 2007). Vegetable
Squalene
Plant sterols
α-Tocopherol
Polyphenols
Terpenic acids
Fresh VOO
615.6
104.0
17.2
2.67
65.7
Potato
raw
0.04
pan-fried Green pepper
26.8
raw
0.02
pan-fried Zucchini
Eggplant
15.6
1.4
0.28
0.26
–
11.2
1.72
0.53
2.09
1.2
0.16
0.16
–
6.0
1.21
0.86
3.94
raw
0.03
0.9
0.09
0.10
0.12
pan-fried
6.8
3.9
1.5
0.68
4.32
floured
35.5
5.5
1.94
0.74
5.92
battered
42.2
17.9
3.89
0.70
6.28
1.2
0.18
36.9
0.05
raw
0.04
pan-fried
97.6
38.6
5.61
33.5
6.22
floured
73.6
27.1
2.57
27.9
5.88
battered
67.8
20.4
1.75
20.6
5.35
70 in frying oil
overall
60
Retention (%)
50 40 30 20 10
Eggplant battered
Eggplant floured
Eggplant
Zucchini battered
Zucchini floured
Zucchini
Green pepper
Potato
0
FIGURE 80.2 α-Tocopherol retention in frying oil and overall retentions based on the α-tocopherol content of both oil and food before and after panfrying of vegetables in VOO.
reported that hydroxytyrosol (3,4-DHPEA) and its secoiridoid derivatives (3,4-DHPEA-EDA and 3,4-DHPEA-EA) decreased rapidly during deep-frying of potatoes in VOO, while tyrosol (p-HPEA) and its derivatives (p-HPEA-EDA and p-HPEA-EA) were more stable.
The frying technique also affects polyphenol degradation, with deep-frying resulting in better antioxidant recoveries as compared with pan-frying (Andrikopoulos et al., 2002a, b). Ioku et al. (2001) compared various cooking methods of onions and concluded that frying did not
CHAPTER | 80 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying of Vegetables
771
Abudance 70 000 10 65 000 60 000 55 000 50 000 45 000
2
40 000 35 000 30 000 25 000 20 000 15 000 10 000
1
11
3 4 5
6 7
12
8 9
5000
Time-->
20.00
40.00
FIGURE 80.3 GC–MS SIM chromatogram of the phenolic extract obtained from floured eggplants pan-fried in VOO. Peaks correspond to the following single polyphenols and terpenic acids: 1: vanillin; 2: tyrosol; 3: p-hydroxybenzoic acid; 4: p-hydroxyphenylacetic acid; 5: internal standard, 3-(4-hydroxyphenyl)-1-propanol; 6: vanillic acid; 7: hydroxytyrosol; 8 caffeic acid; 9: ferrulic acid; 10: chlorogenic acid; 11: oleanolic acid; 12: maslinic acid.
affect flavonoids, while Sahlin et al. (2004) reported that frying significantly reduced ascorbic acid, total phenols and lycopene content of tomatoes, while shallow frying of several vegetables resulted in a reduction of polyphenols in the case of amaranth, spinach and potato, and in an increment in the cases of carrot, tomato and brinjal (Devi and Shlvaprakash, 2004). The fate of individual polyphenols during the domestic pan-frying of vegetables in VOO was studied by Kalogeropoulos et al. (2007). Eleven polyphenols – namely tyrosol, hydroxytyrosol, homovanillic alcohol, p-hydroxybenzoic acid, p-hydroxy-phenylacetic acid, vanillin, vannilic acid, ferulic acid, chlorogenic acid, p-coumaric acid, and caffeic acid – were found in fresh VOO at a concentration of 2.67 mg 100 g⫺1, with tyrosol and hydroxytyrosol comprising 71 and 18% of the polyphenols determined. The polyphenol content of fresh vegetables ranged from 0.1 mg 100 g⫺1 in zucchinis to 36.9 mg 100 g⫺1 in eggplants, with chlorogenic acid predominating in potatoes, green peppers and eggplants, and vanillic acid predominating in fresh zucchinis. Most of these polyphenols were also present in panfried vegetables (Figure 80.3). The respective polyphenol content of pan-fried vegetables ranged from 0.53 mg 100 g⫺1 in fried potatoes to 33.5 mg 100 g⫺1 in fried eggplants.
The amount of polyphenols present in fried samples is the net result of their enrichment due to compensation and VOO polyphenol uptake, and their loss as they are oxidized protecting oil and food from oxidative deterioration. In fried potatoes, zucchinis and green peppers a net increase of polyphenols was observed, as they contained 2–7.4 times more polyphenols than the fresh samples (Table 80.1). On the contrary, fried eggplants contained less polyphenols than the fresh, obviously as the result of partial loss of the eggplants’ endogenous polyphenols which did not counteract the enrichment caused by water loss and VOO uptake. The retention of polyphenols in fried oils ranged from 20.6–42.7% (average 28.2%), and was affected both by vegetable species and by the culinary practice, as lower polyphenol retentions were observed in oils used for frying floured or battered vegetables (Figure 80.4). The overall retention of polyphenols, calculated by the respective amounts present in oil and food before and after frying, ranged from 24.7–69.5% (average 39.8%).
80.2.5.3 Terpenic Acids Virgin olive oil contains significant amounts of hydroxy pentacyclic triterpene acids (HPTA) particularly oleanolic,
772
SECTION | II General Aspects and Changes in Food Processing
80 70
in frying oil
overall
Retention (%)
60 50 40 30 20 10
Eggplant battered
Eggplant floured
Eggplant
Zucchini battered
Zucchini floured
Zucchini
Green pepper
Potato
0
FIGURE 80.4 Polyphenol retention in frying oil and overall retentions based on the polyphenol content of both oil and food before and after pan-frying of vegetables in VOO.
100 90 80 Retention (%)
70 60 50 40 30 20 10 Eggplant battered
Eggplant floured
Eggplant
Zucchini battered
Zucchini floured
Zucchini
Green pepper
Potato
0
FIGURE 80.5 Terpenic acid retention in frying oil and overall retentions based on the terpenic acid content of both oil and food before and after panfrying of vegetables in VOO.
maslinic and ursolic acids (Pérez-Camino and Cert, 1999). Oleanolic and ursolic acid intake has been reported to have hepato-protective, anti-inflammatory, antitumor and antihyperlipidemic properties (Liu, 1995) while ursolic and maslinic acids have been reported to exhibit anti-HIV activity (Xu et al., 1996). Raw potatoes and peppers were not found to contain any detectable amounts of terpenic acids, while small amounts of oleanolic acid were present in fresh zucchinis and eggplants (Table 80.1). Pan-frying caused partial loss of terpenic acids from fried oils and the enrichment of fried vegetables with HPTA, which ranged from 2.1 mg 100 g⫺1 in potatoes to 6.28 mg 100 g⫺1 in battered zucchini (Table 80.1). The terpenic acids overall retention, ranged from 37.3–87.3% (average 60.2%) (Figure 80.5). The comparison of HPTA concentrations in the fried oils
and in the oils absorbed by the fried vegetables indicated that the fate of HPTA is – to some extent – affected by vegetable species and texture.
80.2.5.4 Squalene Squalene, an isoprenoid compound structurally similar to beta-carotene, is a major intermediate in the biosynthesis of cholesterol and is widely distributed in nature – found both in plants and animals. Squalene is distributed ubiquitously in human tissues, with the greatest concentrations in the skin, where it appears to function as a quencher of singlet oxygen, protecting the skin surface from lipid peroxidation due to exposure to UV radiation. Additionally it is considered to protect the retina from oxidative damage (Aguilera
CHAPTER | 80 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying of Vegetables
773
120 in frying oil
overall
Retention (%)
100 80 60 40 20
Eggplant battered
Eggplant floured
Eggplant
Zucchini battered
Zucchini floured
Zucchini
Green pepper
Potato
0
FIGURE 80.6 Squalene retention in frying oil and overall retentions based on the squalene content of both oil and food before and after pan-frying of vegetables in VOO.
et al., 2005). During the past decades, research has indicated that squalene may contribute to the reported anticarcinogenic activity of olive oil, especially for colon cancer (Newmark, 1997; Smith, 2000) and that together with phenolic compounds and oleic acid it is responsible for the antiinflammatory properties of olive oil (Owen et al., 2004). Virgin olive oil is a good source of squalene, containing 200–700 mg 100 g⫺1 oil (Smith, 2000), while other cooking oils and fats have squalene levels in the range of 10–49 mg 100 g⫺1 (Kalogeropoulos and Andrikopoulos, 2004). In domestic pan-frying of vegetables in VOO, Kalogeropoulos et al. (2006) showed that the almost squalene-free fresh vegetables – with squalene content of 0.02–0.04 mg 100 g⫺1 – were enriched with squalene after frying in VOO which contained 615.6 mg 100 g⫺1. Due to VOO uptake, the squalene content of fried vegetables reached 6.8–97.6 mg 100 g⫺1 fw which represents an increase of 2–3 orders of magnitude (Table 80.1). The recoveries of squalene – both overall and in frying oils – ranged between 52–92% and 55–97% respectively, being to some extent affected by the culinary practice, as squalene was found to survive better after the pan-frying of floured or battered vegetables (Figure 80.6). In successive pan- and deep-frying experiments of prefried potatoes in VOO, squalene was found to be distributed uniformly between frying oil and fried potatoes (Kalogeropoulos, unpublished results).
Kalogeropoulos et al. (2006) reported that vegetables pan-fried in VOO were enriched in phytosterols – mainly β-sitosterol – by a factor of 3–30 (Table 80.1). The increment was obviously a result of frying oil uptake, as the phytosterol content of fried vegetables was linearly correlated with the amount of VOO absorbed. The recoveries of phytosterols in fried oil and their overall recoveries ranged from 50% up to 87% (Figure 80.7). Phytosterols are generally uniformly distributed between fried food and frying oil (Kalogeropoulos et al., 2006, Salta et al., 2008).
80.3 NUTRITIONAL EVALUATION OF PAN-FRIED VEGETABLES In Table 80.2, the dietary intake of several macro- and microconstituents provided after the consumption of one serving (140 g) of vegetables pan-fried in VOO is presented. According to the data of Table 80.2, a serving of fried vegetables is expected to provide 5.4–38.5 g of monounsaturated fatty acids, with oleic acid and its isomers comprising more than 98.5%. It could also cover the 5.4–32.4% of a 2000 kcal diet, 3–47% of the daily squalene intake in the Mediterranean countries, 2–30% of typical daily phytosterols intake, 17–54% of the RDI value for α-tocopherol, and 3–49% of the daily average flavonoid intake from a traditional Greek plant-based diet (Kalogeropoulos et al., 2006, 2007).
80.2.5.5 Phytosterols Phytosterols – or plant sterols – are bioactive plant constituents considered as important dietary components for lowering LDL cholesterol and maintaining good heart health, while furthermore they possess anticancer, anti-inflammatory and antioxidant activities (Law, 2000; Berger et al., 2004).
SUMMARY POINTS ●
Pan-frying of vegetables in VOO under household conditions resulted in fried products which were enriched with squalene, tocopherol, polyphenols, terpenic acids
774
SECTION | II General Aspects and Changes in Food Processing
100 90
in frying oil
overall
Retention (%)
80 70 60 50 40 30 20 10 Eggplant battered
Eggplant floured
Eggplant
Zucchini battered
Zucchini floured
Zucchini
Green pepper
Potato
0
FIGURE 80.7 Phytosterol retention in frying oil and overall retentions based on the phytosterol content of both oil and food before and after pan-frying of vegetables in VOO.
TABLE 80.2 Dietary intake of macro- and microconstituents expressed per servinga of vegetables pan-fried in VOO. Pan-fried vegetable
Energy (kcal)
SFA (g)
MUFAb (g)
n-6 PUFA (g)
n-3 PUFA (g)
Tocopherol (mg)
Potato
349
1.9
11.8
1.1
0.1
2.5
0.8
3
38.9
16.2
Green pepper
107
1.0
5.6
0.6
0.1
1.8
1.2
5.7
22.5
8.7
Zucchini
123
1.0
5.4
0.5
0.2
2.2
1.0
6.3
9.9
5.4
Zucchini floured
206
2.3
13.0
1.2
0.3
2.8
1.1
8.6
51.4
8
Zucchini battered
393
3.1
17.5
1.6
0.2
5.6
1.0
9.1
61.2
26
Eggplant
648
6.4
39.4
3.3
0.3
8.1
48.6
9
141.4
56
Eggplant floured
510
6.6
38.5
3.3
0.3
3.7
40.4
8.5
134.2
39.3
Eggplant battered
280
3.1
17.9
1.6
0.1
2.5
29.8
7.8
98
29.6
Polyphenols Terpenic (mg) acids (mg)
Squalene Plant (mg) sterols (mg)
a
one serving ⫽145 g; oleic acid and its isomers more than 98.5% of MUFA; VOO: virgin olive oil; SFA: saturated fatty acids; MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids.
b
●
●
and phytosterols originating in VOO, in amounts proportional to that of the oil absorbed. The VOO-originating microconstituents showed high retentions after one pan-frying session, which were to some extent affected by the culinary practice followed. Besides energy, the consumption of vegetables panfried in VOO is expected to provide the beneficial constituents of both the vegetables and VOO, as the
●
microconstituents provided by consuming a serving of vegetables pan-fried in VOO represent a significant fraction of the respective daily intakes, especially in the case of eggplants. Within the context of the Mediterranean diet, fried vegetables have a place in our diet, assuming that frying is performed in virgin olive oil which has been used no more than one or two times.
CHAPTER | 80 Recovery and Distribution of Macro- and Selected Microconstituents after Pan-frying of Vegetables
REFERENCES Aguilera, Y., Dorado, M.E., Prada, F.A., Martínez, J.J., Quesada, A., RuizGutiérrez, V., 2005. The protective role of squalene in alcohol damage in the chick embryo retina. Exp. Eye Res. 80, 535–543. Andrikopoulos, N.K., Dedoussis, G.V.Z., Falirea, A., Kalogeropoulos, N., Hatzinikola, H.S., 2002a. Deterioration of natural antioxidant species of vegetable edible oils during the domestic deep-frying and pan-frying of potatoes. Int. J. Food Sci. Nutr. 53, 351–363. Andrikopoulos, N.K., Kalogeropoulos, N., Falirea, A., Barbagianni, M.N., 2002b. Performance of virgin olive oil and vegetable shortening during domestic deep-frying and pan-frying of potatoes. Int. J. Food Sci. Tech. 37, 177–190. Banks, D., 1996. Introduction. In: Perkins, E.G., Erickson, M.D. (Eds.) Deep Frying: Chemistry, Nutrition, and Practical Applications. AOCS Press, Champaign, Ill, pp. 1–3. Bastida, S., Sánchez-Muniz, F.J., 2001. Thermal oxidation of olive oil, sunflower oil and a mix of both oils during forty discontinuous domestic fryings of different foods. Food. Sci. Technol. Int. 7, 15–21. Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gómez-Caravaca, A. M., Segura-Carretero, A., Fernández-Gutiérrez, A., Lercker, G., 2007. Phenolic molecules in virgin olive oils: a survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules 12, 1679–1719. Berger, A., Jones, P.J.H., Abumweis, S.S., 2004. Plant sterols: factors affecting their efficacy and safety as functional food ingredients. Lipids in Health and Disease 3:5. (http://www.lipidworld.com/content/3/1/5). Bognár, A., 1998. Comparative study of frying to other cooking techniques influence on the nutritive value. Grasas Aceites 49, 250–260. Brenes, M., Garcia, A., Dobarganes, M.C., Velasco, J., Romero, C., 2002. Influence of thermal treatments simulating cooking processes on the polyphenol content in virgin olive oil. J. Agric. Food Chem. 50, 5962–5967. Covas, M.-I., 2007. Olive oil and the cardiovascular system. Pharmacol. Res. 55, 175–186. Covas, M.-I., Ruiz-Gutiérrez, V., de la Torre, R., Kafatos, A., LamuelaRaventós, R.M., Osada, J., Owen, R.W., Visioli, F., 2006. Minor components of olive oil: evidence to date of health benefits in humans. Nutr. Rev. 64, S20–S30. DAFNE, Data Food Networking, http://www.nut.uoa.gr/Dafnesoftweb/ Main.aspx (Last accessed 19 April 2008). Devi, P.Y., Shlvaprakash, M., 2004. Effect of shallow frying on total phenolic content and antioxidant activity in selected vegetables. J. Food Sci. Tech. 41, 666–668. Dobarganes, C., Márquez-Ruiz, G., Velasco, J., 2000. Interactions between fat and food during deep-frying. Eur. J. Lipid Sci. Technol. 102, 521–528. Esposito, K., Ceriello, A., Giugliano, D., 2007. Diet and the metabolic syndrome. Metab. Syn. Relat. Disord. 5, 291–295. Fillion, L., Henry, C.J.K., 1988. Nutrient losses and gains during frying: a review. Int. J. Food Sci. Nutr. 49, 157–168. Galeone, C., Talamini, R., Levi, F., Pelucchi, C., Negri, E., Giacosa, A., Montellas, M., Franceschi, S., La Vecchia, C., 2007. Fried foods, olive oil and colorectal cancer. Ann. Oncol. 18, 36–39. Gimeno, E., de la Torre-Carbot, K., Lamuela-Raventós, R.M., Castellote, A.I., Fitó, M., de la Torre, R., Covas, M.-I., López-Sabater, M.C., 2007. Changes in the phenolic content of low density lipoprotein after olive oil consumption in men. A randomized crossover controlled trial. Brit. J. Nutr. 98, 1243–1250.
775
Giugliano, D., Esposito, K., 2008. Mediterranean diet and metabolic diseases. Curr. Opin. Lipidol. 19, 63–68. Gómez-Alonso, S., Fregapane, G., Salvador, M.D., Gordon, M.H., 2003. Changes in phenolic composition and antioxidant activity of virgin olive oil during frying. J. Agric. Food Chem. 51, 667–672. Harwood, J.L., Yaqoob, P., 2002. Nutritional and health aspects of olive oil. Eur. J. Lipid. Sci. Technol. 104, 685–697. Hashim, Y.Z.H-Y., Gill, C.I.R., McGlynn, H., Rowland, I.R., 2005. Components of olive oil and chemoprevention of colorectal cancer. Nutr. Rev. 63, 374–386. Ioku, K., Aoyama, Y., Tokuno, A., Terao, J., Nakatani, N., Takei, Y., 2001. Various cooking methods and the flavonoid content in onion. J. Nutr. Sci. Vitaminol. 47, 78–83. Kalogeropoulos, N., Andrikopoulos, N.K., 2004. Squalene in oils and fats from domestic and commercial fryings of potatoes. Int. J. Food Sci. Nutr. 55, 125–129. Kalogeropoulos, N., Grigorakis, D., Mylona, A., Falirea, A., Andrikopoulos, N.K., 2006. Dietary evaluation of vegetables panfried in virgin olive oil following the Greek traditional culinary practice. Ecol. Food Nutr. 45, 105–123. Kalogeropoulos, N., Mylona, A., Chiou, A., Ioannou, M.S., Andrikopoulos, N. K., 2007. Retention and distribution of natural antioxidants (α-tocopherol, polyphenols and terpenic acids) after shallow frying of vegetables in virgin olive oil. LWT-Food Sci. Technol. 40, 1008–1017. Kontogianni, M.D., Panagiotakos, D.B., 2007. Current epidemiological and clinical evidence on the relationship between Mediterranean diet and the metabolic syndrome. Curr. Nutr. Food Sci. 3, 296–299. Kris-Etherton, P.M., Pearson, T.A., Wan, Y., Hargrove, R.L., Moriarty, K., Fishell, V., Etherton, T.D., 1999. High-monounsaturated fatty acid diets lower both plasma cholesterol and triacylglycerol concentrations. Am. J. Clin. Nutr. 70, 1009–1015. Law, M.R., 2000. Plant sterol and stanol margarines and health. Brit. Med. J. 320, 861–864. Liu, J., 1995. Pharmacology of oleanolic acid and ursolic acid. J. Ethnopharmacol. 49, 57–68. Morton, I.D., 1998. Geography and history of the frying process. Grasas Aceites 49, 247–249. Newmark, H.L., 1997. Squalene, olive oil, and cancer risk: a review and hypothesis, Cancer Epidemiol. Biomarkers Prev. 6, 1101–1103. Owen, R.W., Haubner, R., Würtele, G., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2004. Olives and olive oil in cancer prevention. Eur. J. Cancer Prev. 13, 319–326. Pérez-Camino, M.C., Cert, A., 1999. Quantitative determination of hydroxy pentacyclic triterpene acids in vegetable oils. J. Agric. Food Chem. 47, 1558–1562. Pinthus, E.J., Weinberg, P., Saguy, I.S., 1995. Deep-fat fried potato product oil uptake as affected by crust physical properties. J. Food Sci. 60, 770–772. Quiles, J.L., Ramírez-Tortosa, M.C., Gómez, J.A., Huertas, J.R., Mataix, J., 2002. Role of vitamin E and phenolic compounds in the antioxidant capacity, measured by ESR, of virgin olive, olive and sunflower oils after frying. Food Chem. 76, 461–468. Saguy, I.S., Dana, D., 2003. Integrated approach to deep fat frying: engineering, nutrition, health and consumer aspects. J. Food Eng. 56, 143–152. Sahlin, E., Savage, G.P., Lister, C.E., 2004. Investigation of the antioxidant properties of tomatoes after processing. J. Food Compos. Anal. 17, 635–647. Salta, F.N., Kalogeropoulos, N., Karavanou, N., Andrikopoulos, N.K., 2008. Distribution and retention of phytosterols in frying oils and
776
SECTION | II General Aspects and Changes in Food Processing
fried potatoes during repeated deep and pan frying. Eur Food Res Technol. 227, 391–400. Schneider, C., 2005. Chemistry and biology of vitamin E. Mol. Nutr. Food Res. 49, 7–30. Smith, T.J., 2000. Squalene: potential chemopreventive agent. Expert Opin. Inv. Drug. 9, 1841–1848. Stacewicz-Sapuntzakis, M., Borthakur, G., Burns, J.L., Bowen, P.E., 2008. Correlations of dietary patterns with prostate health. Mol. Nutr. Food Res. 52, 114–130. Traber, M.G., Atkinson, J., 2007. Vitamin E, antioxidant and nothing more. Free Rad. Biol. Med. 43, 4–15.
Trichopoulou, A., Lagiou, P., Kuper, H., Trichopoulos, D., 2000. Cancer and Mediterranean dietary traditions. Canc. Epidem. Biomark. 9, 869–873. Varela, G., Ruiz-Roso, B., 1992. Some effects of deep frying on dietary fat intake. Nutr. Rev. 50, 256–262. Warner, K., 2002. Chemistry of frying oils. In: Akoh, C.C., Min, D.B. (Eds.) Food Lipids. Chemistry, Nutrition and Biotechnology. Marcel Dekker, New York, pp. 205–221. Xu, H., Zeng, F., Wan, M., Sim, K., 1996. Anti-HIV triterpene acids from Geum japonicum. J. Nat. Prod. 59, 643–645.
2.2
Cardiovascular Cardiac Aspects Vascular Aspects Including Hypertension Lipid Aspects
This page intentionally left blank
Chapter 81
Myocardial Infarction and Protection with Olive Oil Miguel A. Martínez-González1, Moises Rodríguez-Manero2 and Félix Valencia-Serrano3 1
Department of Preventive Medicine and Public Health, Medical School-Clinica Universitaria, University of Navarra, Pamplona, Spain Department of Cardiology, Medical School-Clinica Universitaria, University of Navarra, Pamplona, Spain 3 Cardiology Department, Hospital Virgen de la Victoria, University of Malaga, Spain
2
81.1 INTRODUCTION Globally, non-communicable diseases will cause over threequarters of all deaths in 2030. The major non-communicable conditions are cardiovascular diseases. Worldwide deaths from cardiovascular diseases are projected to rise from 17.1 million in 2004 to 23.4 million in 2030. Ischemic heart disease (IHD) is ranked as the first global cause of death in 2004 and it is unfortunately forecast to remain the first cause of death in 2030 (WHO, 2008). In addition to remaining as the major cause of death in industrialized countries, the incidence and mortality from IHD are also rapidly growing in less developed countries. Acute myocardial infarction (AMI) constitutes a catastrophic manifestation of IHD. Coronary atherosclerosis and a superimposed thrombosis play pivotal parts as the underlying substrates of AMI. Despite the development of new therapeutics and patient care strategies in the last 20 years that have substantially improved the morbidity and mortality associated with AMI, the implementation of preventive actions that may benefit the whole population continue to be a great challenge. Unacceptable high absolute rates of mortality from AMI still exist in many developed countries. But, a surprisingly low incidence is found in several Southern European countries such as France, Spain, Greece, Italy and Portugal, leading to a higher life expectancy in Mediterranean areas as compared with Northern European countries or the USA. Diet and lifestyle-related factors may be responsible for this advantage. The Mediterranean diet (Med-Diet) has been proposed as the major protective factor. However, some inconsistencies still persist (Martínez-González and Sánchez-Villegas, 2004). In spite of its relatively high-fat Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
content, or precisely because of it, there is an important advantage of the Med-Diet pattern in the promotion of vegetable consumption, because fat-free or low-fat dressings make vegetables much less palatable than the use of olive oil. The sautéing or stir-frying of vegetables with olive oil instead of using low-fat spreads or steaming enhances flavor and favors the long-term maintenance of a vegetablerich diet. These culinary techniques for the preparation of vegetables are customary in Mediterranean countries. Hence, in health promotion and nutritional education, good compliance with the Med-Diet can be expected by people with a cultural preference for the relatively strong taste of olive oil. In fact, some trials of weight loss with hypocaloric diets (McManus et al., 2001) reported a better adherence to a low-calorie Med-Diet than to a standard low-fat diet. Despite these potential benefits of the Med-Diet, an undesirable departure from the traditional Med-Diet has been reported to have occurred in Southern European countries (Sánchez-Villegas et al., 2003). The classical Med-Diet is identified as the traditional dietary pattern found in olive-growing areas of Crete, Greece, and Southern Italy in the late 1950s and early 1960s (Trichopoulou et al., 1995). One of the most important characteristics of this diet is the presence of virgin olive oil as the principal source of energy from fat, which is a good source of monounsaturated fatty acids (MUFA) and hundreds of micronutrients. Virgin olive oil retains all the lipophilic components of the fruit, alfa-tocopherol, and phenolic compounds with strong antioxidant and antiinflamatory properties. A strong protection by olive oil against AMI is expected to be found in analytical epidemiological studies assessing this relationship.
779
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
780
81.2 BIOLOGICAL MECHANISMS In comparison with saturated fatty acids, olive oil reduces low-density lipoprotein (LDL) cholesterol, and compared with carbohydrates, it maintains or even increases the levels of high-density lipoprotein (HDL) cholesterol. In addition, it is relatively resistant to oxidation and contains a large amount of antioxidants relative to its polyunsaturated fat content. Some polyphenol constituents of virgin olive oil (hydroxytyrosol and oleuropein) are potent scavengers of superoxide radicals and inhibit LDL oxidation (Fitó et al., 2000, 2007). Olive oil has induced a regression of atherosclerosis in animal models and may slow the development of coronary atherosclerosis, being associated with a reduced DNA synthesis in human coronary smooth muscle cells (Mata et al., 1997). Olive oil also favorably affects postprandial factor VII activity, avoiding a prolonged thrombotic response to a high-fat diet. A beneficial effect of MUFA on von Willebrand factor, as well as other benefits of olive oil on the hemostatic system have also been suggested (Pérez-Jiménez et al., 2002). The preservation of endothelial function is a key mechanism for the prevention of atherosclerosis. The available information about the effects of olive oil on endothelial function also suggests a benefit. The vascular endothelium plays a key role in local vascular tone regulation and can be modulated by dietary fat (Vogel et al., 1997). In a randomized trial, a MUFA-rich Med-Diet improved the endotheliumdependent vasodilatory response suggesting that a Med-Diet rich in olive oil may be able to avoid the postprandial deterioration of endothelial function (Fuentes et al., 2008). It is well known that the metabolic syndrome increases the risk of AMI. In a Spanish cohort following a sample of 3947 initially healthy participants, a higher adherence to the classical Med-Diet was found to be associated with a lower risk of developing the metabolic syndrome after 74month follow-up (Tortosa et al., 2007). In diabetic patients, olive oil improves the lipid profile and glycemic control (Garg, 1999; Ros, 2003). Moreover, recently two large cohort studies have reported a strong protection of MedDiets, rich in olive oil, against type-2 diabetes (Mozaffarian et al., 2007; Martinez-Gonzalez et al., 2008).
81.3 AVAILABLE EPIDEMIOLOGICAL EVIDENCE No primary prevention trial or prospective cohort study has ever assessed the association between adherence to a Med-Diet and the incidence (and not only mortality) of a first IHD event. A very important cohort study found that a Mediterranean food pattern was protective against mortality from IHD (Trichopoulou et al., 2003). However, they included only fatal IHD cases (54 coronary deaths) as the outcome. Subsequently other studies confirmed the protection of Mediterranean-type
SECTION | II
Cardiac Aspects
diets against overall mortality (Knoops et al., 2004; Mitrou et al., 2007). In any case, mortality from IHD is not only related to its incidence but also to the quality and timeliness of medical care. If the quality of medical care is associated with the adherence to a Med-Diet pattern, the use of mortality as outcome would lead to confounding (Martínez-González and Sánchez-Villegas, 2004). A randomized secondary prevention trial conducted in France (de Lorgeril et al., 1999) showed an impressive protection provided by an experimental Mediterranean diet on the risk of death and re-infarction among survivors of a first AMI. Nevertheless, as the major element of the assigned diet was an experimental canola-oil-based margarine and the diet simultaneously included a high intake of alpha-linolenic acid, fruit and vegetables, it was not possible to attribute its benefit to a single factor. In addition, no special consideration was given to olive oil, which is the major source of MUFA in Mediterranean countries. The fat composition of the experimental group in the Lyon Diet Heart Study was 30.5% of energy intake as total fat (12.9% MUFA). These values are far from the characteristic 35–40% total fat and 15–20% MUFA content present in the traditional Med-Diet. Some methodological caveats (Robertson and Smaha, 2001) have been raised on the Lyon Diet Heart Study, including the small number of observed primary events (14 vs. 44). A few studies have assessed the specific role of olive oil on the risk of clinical coronary events. A protective role for olive oil on mortality among patients with a previous AMI has been reported by the investigators of the large cohort of patients who participated in a previous trial. This study included 11 246 survivors of a myocardial infarction. The authors assessed with a brief, non-validated questionnaire, the consumption of five food items, at baseline, and after 6, 18 and 42 months of follow-up. After 6.5-year followup the odds ratio (OR) for the categories of olive oil consumption ‘often’ and ‘regularly’ compared with the ‘never or sometimes’ category were 0.77 (95% CI 0.62–0.94) and 0.71 (95% CI 0.60–0.84) respectively (Barzi et al., 2003). In a recent case-control study (748 cases and 1048 controls) conducted in Greece, protection was also found for the exclusive use of olive oil (relative risk ⫽ 0.53; 95% CI: 0.34–0.71) against acute coronary syndromes (Kontogianni et al., 2007). However, the authors grouped the participants in three categories (no use; exclusive use; olive plus other oils or fats), but they apparently did not further quantify the amount of olive oil. Nevertheless, conflicting results have been also reported. Specifically for olive oil, a small randomized trial of corn oil and olive oil carried out more than 40 years ago found no benefit for olive oil and even an adverse significant effect for corn oil in 80 coronary patients after 2 years of follow-up (Rose et al., 1965). Two Italian case-control studies reported no significant benefit for olive oil consumption (Gramenzi et al., 1990; Bertuzzi et al., 2002).
781
CHAPTER | 81 Myocardial Infarction and Protection with Olive Oil
81.4 THE CASE-CONTROL STUDY IN THE UNIVERSITY OF NAVARRA With the aim of assessing the potential role of olive oil for the primary prevention of CHD and to quantify the reduction in the risk of a first acute myocardial infarction that can be provided by a high olive oil intake, a hospitalbased case-control study was conducted at the University of Navarra, Spain (Fernández-Jarne et al., 2002). The case series was comprised by 171 patients (81% males, age ⬍80 years) who suffered their first acute myocardial infarction. The control series included 171 age-, gender- and hospitalmatched controls with a wide variety of conditions believed to be unrelated to diet. A previously validated semi-quantitative food-frequency questionnaire (136 items) was used to appraise previous long-term dietary exposures. The same physician conducted the face-to-face interview for each case patient and his/her matched control. Conditional logistic regression modeling was used to take into account potential dietary and non-dietary confounders. The description of cases and controls is shown in Tables 81.1 and 81.2. Crude olive oil intake (unadjusted for total energy intake) categorized in quintiles was used as the independent variable (the first quintile was considered as the reference category). Then, energy-adjusted quintiles of olive oil were used as exposure. No distinction between virgin olive oil and refined (common) olive oil was done. The selection of dietary and non-dietary confounders was done by taking into account previous published literature about coronary risk factors and avoided the reliance on p values or stepwise approaches (Table 81.3). When the quintiles of olive oil intake without energy-adjustment were used as the exposure variable (Table 81.4), the point-estimates for the OR were lower than 1 in the three upper quintiles of olive oil intake. Exposure to the upper quintile of olive oil was associated with a relative risk reduction of 64% (OR ⫽ 0.36, 95% CI: 0.12–1.08) with respect to the first quintile (median intake: 7 g day⫺1). The linear trend test was in the limit of statistical significance when the following potential confounders were also taken into account: smoking (four categories), BMI (continuous variable, adding a quadratic term to account for non-linearity), high blood pressure, high blood cholesterol, diabetes, leisure-time physical activity (METS-h week⫺1, continuous variable, adding a quadratic term), marital status, occupation and educational level (four categories). Further adjustment for other nutrients led to statistically significant results with OR ⫽ 0.26 (0.08–0.85) for the upper quintile and p ⫽ 0.02 for the linear trend test (Table 81.5). Conditional logistic regression models were also fitted using quintiles of energy-adjusted intake of olive oil as the exposure variable (Table 81.6). The risk reduction was then more apparent. Point-estimates lower than 1 for the OR were found in the four upper quintiles of energy-adjusted olive oil intake and a significant linear trend test either when
TABLE 81.1 Non-nutritional characteristics of case and control participants. Cases
Controls
Agea (years, mean)
61.7
61.4
Gendera (% men)
81
81
⬍Primary
28
29
Primary
44
45
Secondary
12
16
University
16
10
White collar
21
25
Blue collar
18
16
Retired
44
46
Housewife
12
10
Other
5
3
79
76
Never
32
44
Currently
40
23
Ex-smoker (⬍3 years)
12
6
Ex-smoker (ⱖ3 years)
17
26
Body mass index (kg/m2, mean)
27.7
27.3
History of hypertension (%)
42
30
History of diabetes (%)
16
8
High blood cholesterol in last 5 years (%)
19
11
Leisure-time physical activity (METSb-h/week, mean)
31.5
34.5
Educational level (%)
Occupational level (%)
Marital status (% married) Smoking (%)
a
Age and gender were matching variables. Metabolic equivalents.
b
non-dietary confounders (p ⫽ 0.03) or also some relevant nutrients (p ⫽ 0.03) were adjusted for. The OR of a first myocardial infarction for the upper quintile was OR ⫽ 0.22 (0.07–0.67) after adjusting for non-dietary confounders and OR ⫽ 0.18 (0.05–0.63) after adjustment for dietary and non-dietary confounders (Table 81.6).
782
SECTION | II
Cardiac Aspects
81.5 DISCUSSION
TABLE 81.2 Nutritional characteristics of case and control participants. Cases
Controls
Total energy intake (kcal day⫺1, mean)
2631
2578
% energy from fat (mean)
31.2
31.3
MUFA/SFAa intake (g day⫺1, mean)
1.51
1.54
% energy from trans-fatty acids (mean)
0.17
0.19
Total ethanol intake (g day⫺1, mean)
19.0
18.2
Glycemic load (g day⫺1, mean)
233
225
Vitamin B6 intake (mg day⫺1, mean)
2.8
2.8
Vitamin C intake (mg day⫺1, mean)
268
257
Vitamin E intake (mg day⫺1, mean)
8.0
7.7
Folic acid intake (μg day⫺1, mean)
419
428
Quintiles of olive oil intake defined according to the distribution among controls were compared regarding several potential nutritional and nonnutritional confounding variables (Tables 81.3 and 81.4). a
Monounsaturated fatty acids/saturated fatty acids.
These results suggest that olive oil consumption may reduce the risk of coronary disease. Interestingly, a previous diagnosis of angina pectoris, a previous history of IHD or other prior diagnosis of major cardiovascular disease were exclusion criteria in this case-control study. This is important because an inverse association between olive oil and IHD has long been suggested. This belief is also held by the general public in Mediterranean countries. Therefore, it is likely that subjects who perceive themselves at higher risk of IHD may think that they should increase their consumption of olive oil to get a better protection. This possibility would introduce a selection bias in the case series. This could be a plausible explanation of the discordant findings of this case-control study (Fernández-Jarne et al., 2002) with respect to two other case-control studies conducted in Italy (Bertuzzi et al., 2002; Gramenzi et al., 1990), which did not find any significant association. Recall bias is a potential concern when the casecontrol design is used. But differential over-reporting would be more probable to exist among cases than among controls because cases are more likely to be aware of the role of nutrition as a determinant of IHD. Therefore, recall bias does not seem to be a likely alternative explanation of
TABLE 81.3 Distribution of potential non-nutritional confounding variables across quintiles of energy-adjusted olive oil intake among control subjects (n ⴝ 171). Quintiles of energy-adjusted olive oil intake 1
2–4
5
Energy adjusted olive-oil (g day⫺1, mean)
3.6
22.5
54.1
Body mass index (kg/m2, mean)
26.8
27.3
27.8
0.30
% white collar
18
28
24
0.91
% educational level higher than primary
18
30
21
0.75
% married
56
80
85
0.03
% smokers
18
21
35
0.06
% high blood cholesterol
15
12
6
0.24
% high blood pressure
29
33
21
0.25
% diabetes
6
5
18
0.02
Leisure-time physical activity (METSa-h week⫺1, mean)
30.5
36.3
33.8
0.90
a
Metabolic equivalents.
p for trend
783
CHAPTER | 81 Myocardial Infarction and Protection with Olive Oil
TABLE 81.4 Distribution of potential nutritional confounding variables across quintiles of energy-adjusted olive oil intake among control subjects (n ⴝ 171). Quintiles of energy-adjusted olive oil intake 1
2–4
5
p for trend
Energy adjusted olive-oil (g day⫺1, mean)
3.6
22.5
54.1
Total energy intake (kcal day⫺1, mean)
2882
2417
2778
0.49
Ethanol intake (g day⫺1, mean)
16
16
27
0.02
% energy from fat (mean)
29
31
36
⬍0.001
% energy from saturated fat (mean)
11
10
10
0.58
% energy from monounsaturated fat (mean)
12
15
20
⬍0.001
MUFA/SFAa intake (mean)
1.18
1.50
2.02
⬍0.001
% energy from trans-fatty acids (mean)
0.23
0.19
0.14
0.11
Glycemic load (g day⫺1, mean)
235
231
207
0.34
Total fiber intake (g day⫺1, mean)
38
30
29
0.04
Folic acid intake (μg day⫺1, mean)
513
409
395
0.03
Vitamin B6 intake (mg day⫺1, mean)
3.3
2.6
2.5
0.02
Vitamin C intake (mg day⫺1, mean)
320
262
233
0.06
Vitamin E intake (mg day⫺1, mean)
8.7
7.4
7.5
0.54
a
Monounsaturated fatty acids/saturated fatty acids.
TABLE 81.5 Odds ratio (OR) (95% CI) of a first myocardial infarction according to olive oil intake (unadjusted for total energy intake). Quintiles of olive oil intake 1
2
3
4
5
P
Controls/case (n)
32/36
35/37
36/30
31/39
37/29
Median intake (g day⫺1)
7.2
12.0
25.0
29.3
54.3
Multivariate adjusted ORa (95% CI)
1 (Ref.)
1.17 (0.46–3.02)
0.69 (0.28–1.67)
0.91 (0.38–2.18)
0.36 (0.12–1.08)
0.05
Multivariate adjusted ORb (95% CI)
1 (Ref.)
1.16 (0.46–2.95)
0.60 (0.24–1.49)
0.83 (0.34–2.01)
0.26 (0.08–0.85)
0.02
P (p value for linear trend). a conditional logistic regression (age-, hospital- and gender-matched pairs), adjusted for smoking, body mass index, high blood pressure, high blood cholesterol, diabetes, leisure time physical activity (METS-hours week⫺1), marital status, occupation and study level. b additionally adjusted for saturated fat, trans fat and total fiber intake.
784
SECTION | II
Cardiac Aspects
TABLE 81.6 Odds ratio (OR) (95% CI) of a first myocardial infarction according to energy-adjusted olive oil intake (unadjusted for total energy intake). Quintiles of energy adjusted olive oil intake 1
2
3
4
5
P
Controls/case (n)
28/40
38/31
38/30
29/40
38/30
Median intake (g day⫺1)
6.1
13.6
21.0
30.9
52.2
Multivariate adjusted ORa (95% CI)
1 (Ref.)
0.39 (0.15–1.00)
0.40 (0.17–0.93)
0.59 (0.23–1.52)
0.22 (0.07–0.67)
0.03
Multivariate adjusted ORb (95% CI)
1 (Ref.)
0.45 (0.16–1.25)
0.44 (0.18–1.07)
0.70 (0.24–2.02)
0.18 (0.05–0.63)
0.03
P (p value for linear trend). a Conditional logistic regression (age-, hospital- and gender-matched pairs), adjusted for smoking, body mass index, high blood pressure, high blood cholesterol, diabetes, leisure time physical activity (METS-hours/week), marital status, occupation and study level. b Additionally adjusted for % energy derived from saturated fat, % energy derived from trans fat, total fiber consumption, folic acid intake, vitamin C intake, glycemic load and ethanol intake (adding a quadratic term to account for non-linearity).
these findings. Moreover, when the assessment of exposure was done through different items in a comprehensive questionnaire, such as in this case, it would be more difficult that patients might consistently underestimate their exposure to olive oil. Although the in-hospital selection of controls facilitates a higher participation, it also imposes some caution in the interpretation of findings because the exposure may be related to the diseases causing the hospital admission of controls. However, olive oil has not been found to induce any trauma or genitourinary disease or any common disease needing minor surgery (the most frequent diseases in the control series). Thus, a control selection bias does not seem very likely. Furthermore, subsequent studies have confirmed the findings of the Spanish case-control study regarding clinical coronary events (Barzi et al., 2003; Kontogianni et al., 2007), hypertension (Alonso and Martínez-González, 2004), cardiovascular risk factors (Esposito et al., 2004; Estruch et al., 2006) and total mortality (Trichopoulou et al., 2003; Knoops et al., 2004; Mitrou et al., 2007). The on-going SUN study is a dynamic prospective cohort (Martinez-Gonzalez et al., 2002) currently including more than 20 000 healthy university graduates followedup every 2 years. The recruitment is permanently open and the first participants were included in 1999. However most participants have been admitted to the cohort during the period 2003–2007 and some additional follow-up time (2–3 years more) will be needed before being able to prospectively ascertain the association between the consumption of olive oil and the incidence of new clinical events of IHD. The PREDIMED study (Estruch et al., 2006) is a primary cardiovascular prevention trial that has already randomized about 7000 older high-risk participants. They have been
randomly allocated to three different food patterns, one of them is a Med-Diet rich in virgin olive oil (Zazpe et al., 2008). The recruitment started in 2003 and the final results regarding clinical events are expected in 2011. The results of the SUN cohort and the PREDIMED trial will provide in the next few years the best evidence on this topic.
SUMMARY POINTS ●
●
●
●
●
A wide variety of mechanistic reasons support that olive oil may exert a beneficial effect on the risk of coronary disease. International comparisons, with the pioneering results of the Seven Countries Study, also support this hypothesis. In a case-control study, conducted at the University of Navarra (Spain), the relative risk reduction for myocardial infarction was above 75% for participants in the highest quintile of olive oil consumption (median ⫽ 54 g day⫺1) versus the first quintile (median ⫽ 7 g day⫺1) after multivariate adjustment for a wide array of dietary and non-dietary confounders. Subsequent studies (an Italian cohort and a recent Greek case-control study) confirmed this inverse association. Some inconsistencies remain: previous case-control studies conducted in Italy and Greece did not find any association, and a small randomized trial on coronary patients conducted by Rose more than 40 years ago was also negative for olive oil, and even harmful for corn oil. However, the Greek and Italian case-control studies
CHAPTER | 81 Myocardial Infarction and Protection with Olive Oil
●
had some methodologic flaws, and the trial did not have sufficient statistical power. A large, Spanish prospective cohort study (the SUN cohort, n ⬎ 20 000) and a primary prevention trial in Spain (the PREDIMED trial, n ⫽ 7000) are currently assessing the role of olive oil not only on IHD mortality but also on incidence of coronary clinical events. Results are expected in the next 2–3 years. Preliminary results on hypertension, diabetes, insulin resistance, and other risk factors are supportive of a strong cardiovascular protection by olive oil.
REFERENCES Alonso, A., Martínez-González, M.A., 2004. Olive oil consumption and reduced incidence of hypertension: the SUN study. Lipids 39, 1233–1238. Barzi, F., Woodward, M., Marfisi, R.M., Tavazzi, L., Valagussa, L., Marchioli, R., 2003. Mediterranean diet and all causes mortality after myocardial infartion: results from the GISSI-Preventiozione trial. Eur. J. Nutr. 57, 604–611. Bertuzzi, M., Tavani, A., Negri, E., La Vecchia, C., 2002. Olive oil consumption and risk of non-fatal myocardial infarction in Italy. Int. J. Epidemiol. 31, 1274–1277. De Lorgeril, M., Salen, P., Martin, J.L., Monjaud, I., Delaye, J., Mamelle, N., 1999. Mediterranean diet, traditional risk factors and the rate of cardiovascular complications after myocardial infarction; final report of the Lyon Diet Heart Study. Circulation 99, 779–785. Esposito, K., Marfella, R., Ciotola, M., Di Palo, C., Giugliano, F., Giugliano, G., D’Armiento, M., D’Andrea, F., Giugliano, D., 2004. Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA 292, 1440–1446. Estruch, R., Martínez-González, M.A., Corella, D., Salas-Salvadó, J., RuizGutiérrez, V., Covas, M.I., Fiol, M., Gómez-Gracia, E., López-Sabater, M.C., Vinyoles, E., Arós, F., Conde, M., Lahoz, C., Lapetra, J., Sáez, G., Ros, E. For the PREDIMED Study Investigators, 2006. Effects of a Mediterranean-style diet on cardiovascular risk factors: a randomized trial. Ann. Intern. Med. 145, 1–11. Fernández-Jarne, E., Martínez-Losa, E., Prado-Santamaría, M., Brugarolas-Brufau, C., Serrano-Martínez, M., Martínez-González, M.A., 2002. Risk of first non-fatal myocardial infarction negatively associated with olive oil consumption: a case-control study in Spain. Int. J. Epidemiol. 31, 474–480. Fitó, M., Guxens, M., Corella, D., Sáez, G., Estruch, R., de-la-Torre, R., Francés, F., Cabezas, C., López-Sabater, M.C., Marrugat, J., GarcíaArellano, A., Arós, F., Ruiz-Gutierrez, V., Ros, E., Salas-Salvadó, J., Fiol, M., Solá, R., Covas, M.I. For the PREDIMED Study Investigators, 2007. Effect of a traditional Mediterranean diet on lipoprotein oxidation: a randomized controlled trial. Arch. Intern. Med. 167, 1195–1203. Fitó, M., Covas, M.I., Lamuela-Raventós, R.M., Vila, J., Torrents, L., de-la-Torre, C., Marrugat, J., 2000. Protective effect of olive oil and its phenolic compounds against low density lipoprotein oxidation. Lipids 35, 633–638. Fuentes, F., López-Miranda, J., Pérez-Martínez, P., Jiménez, Y., Marín, C., Gómez, P., Fernández, J.M., Caballero, J., Delgado-Lista, J.
785
Pérez-Jiménez, F., 2008. Chronic effects of a high-fat diet enriched with virgin olive oil and a low-fat diet enriched with alpha-linolenic acid on postprandial endothelial function in healthy men. Br. J. Nutr. Feb 14:1-7 [Epub ahead of print]. Garg, A., 1999. High-monounsaturated-fat diets for patients with diabetes mellitus: a meta-analysis. Am. J. Clin. Nutr. 67, 577S–582S. Gramenzi, A., Gentile, A., Fasoli, M., Negri, E., Parazzini, F., La Vecchia, C., 1990. Association between certain foods and risk of acute myocardial infarction in women. BMJ 300, 771–773. Knoops, K.T., de Groot, L.C., Kromhout, D., Perrin, A.E., MoreirasVarela, O., Menotti, A., van Staveren, W.A., 2004. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women. The HALE project. JAMA 292, 1433–1439. Kontogianni, M.D., Panagiotakos, D.B., Chrysohoou, C., Pitsavos, C., Zampelas, A., Stefanadis, C., 2007. The impact of olive oil consumption pattern on the risk of acute coronary syndromes: the CARDIO2000 case-control study. Clin. Cardiol. 30, 125–129. Martínez-González, M.A., Sánchez-Villegas, A., De Irala, J., Marti, A., Martínez, J.A., 2002. Mediterranean diet and stroke: objectives and design of the SUN project. Seguimiento Universidad de Navarra. Nutr. Neurosci. 5, 65–73. Martínez-González, M.A., Sánchez-Villegas, A., 2004. The emerging role of Mediterranean diets in cardiovascular epidemiology: monounsaturated fats, olive oil, red wine or the whole pattern? Eur. J. Epidemiol. 19, 9–13. Martínez-González, M.A., de-la-Fuente-Arrillaga, C., Nuñez-Córdoba, J.M., Basterra-Gortari, F.J., Beunza, J.J., Vázquez, Z., Benito, S., Tortosa, A., Bes-Rastrollo, M., 2008. Adherence to the Mediterranean diet inversely associated with the risk of developing diabetes: the SUN prospective cohort. BMJ 336, 348–351. Mata, P., Varela, O., Alonso, R., Lahoz, C., de Oya, M., Badimon, L., 1997. Monounsaturated and polyunsaturated n-6 fatty acid-enriched diets modify LDL oxidation and decrease human coronary smooth muscle cell DNA synthesis. Arterioscler. Thromb. Vasc. Biol. 17, 2088–2095. McManus, K., Antinoro, L., Sacks, F., 2001. A randomized controlled trial of a moderate-fat, low-energy diet compared with a low fat, low-energy diet for weight loss in overweight adults. Int. J. Obes. 25, 1503–1511. Mitrou, P.N., Kipnis, V., Thiébaut, A.C., Reedy, J., Subar, A.F., Wirfält, E., Flood, A., Mouw, T., Hollenbeck, A.R., Leitzmann, M.F., Schatzkin, A., 2007. Mediterranean dietary pattern and prediction of all-cause mortality in a US population: results from the NIH-AARP diet and health study. Arch. Intern. Med. 167, 2461–2468. Mozaffarian, D., Marfisi, R., Levantesi, G., Silletta, M.G., Tavazzi, L., Tognoni, G., Valagussa, F., Marchioli, R., 2007. Incidence of newonset diabetes and impaired fasting glucose in patients with recent myocardial infarction and the effect of clinical and lifestyle risk factors. Lancet 370, 667–675. Pérez-Jiménez, F., López-Miranda, J., Mata, P., 2002. Protective effect of dietary monounsaturated fat on arteriosclerosis: beyond cholesterol. Atherosclerosis 163, 385–398. Robertson, R.M., Smaha, L., 2001. Can a Mediterranean-style diet reduce heart disease? Circulation 103, 1821–1822. Ros, E., 2003. Dietary cis-monounsaturated fatty acids and metabolic control in type 2 diabetes. Am. J. Clin. Nutr. 78 (suppl), 617–625S. Rose, G.A., Thomson, W.B., Williams, R.T., 1965. Corn oil in treatment of ischaemic heart disease. BMJ 1, 1531–1533. Sánchez-Villegas, A., Delgado-Rodríguez, M., Martínez-González, M.A., De Irala, J., 2003. Gender, age, socio-demographic and lifestyle factors associated with major dietary patterns in the Spanish Project
786
SUN (Seguimiento Universidad de Navarra). Eur. J. Clin. Nutr. 57, 285–292. Tortosa, A., Bes-Rastrollo, M., Sánchez-Villegas, A., Basterra-Gortari, F.J., Nuñez-Córdoba, J.M., Martínez-González, M.A., 2007. Mediterranean diet inversely associated with the incidence of metabolic syndrome: the SUN prospective cohort. Diabetes Care 30, 2957–2959. Trichopoulou, A., Kouris-Blazos, A., Wahlqvist, M.L., Gnardellis, C., Lagiou, P., Polychronopoulos, E., Vassilakou, T., Lipworth, L., Trichopoulos, D., 1995. Diet and overall survival in elderly people. BMJ 311, 1457–1460. Trichopoulou, A., Costacou, T., Bamia, C., Trichopoulos, D., 2003. Adherence to a Mediterranean diet and survival in a Greek population. N. Engl. J. Med. 348, 2599–2608.
SECTION | II
Cardiac Aspects
Vogel, R.A., Coretti, M.C., Plottnick, G.D., 1997. Effects of a single highfat meal on endothelial function in healthy subjects. Am. J. Cardiol. 79, 350–354. World Health Organization, 2008. World health statistics 2008. WHO, Geneve. Zazpe, I., Sánchez-Tainta, A., Estruch, R., Lamuela-Raventos, R.M., Schröder, H., Salas-Salvado, J., Corella, D., Fiol, M., Gomez-Gracia, E., Aros, F., Ros, E., Ruiz-Gutierrez, V., Iglesias, P., Conde-Herrera, M., Martínez-González, M.A., 2008. A large randomized individual and groupal intervention conducted by registered dietitians increased adherence to Mediterranean-type diets: the PREDIMED study. J. Am. Diet. Assoc. 108, 1134–1144.
Chapter 82
Beneficial Effects of Olive Oil Compared with Fish, Canola, Palm and Soybean Oils on Cardiovascular and Renal Adverse Remodeling due to Hypertension and Diabetes in Rat Marcia Barbosa Aguila and Carlos Alberto Mandarim-de-Lacerda Laboratory of Morphometry and Cardiovascular Morphology, Biomedical Centre, Institute of Biology, State University of Rio de Janeiro, Brazil
82.1 INTRODUCTION A cardiac event is the major cause of mortality in people with diabetes and many factors, including hypertension, contribute to this high prevalence of cardiovascular diseases (CVD). Other important risk factors for CVD include obesity, atherosclerosis, dyslipidemia, microalbuminuria, endothelial dysfunction, platelet hyperaggregability and coagulation abnormalities (Sowers et al., 2001). Chronic hypertension and diabetes, isolated or combined, lead to microvascular alteration in target organs like heart and kidney, causing numerous structural damages including cardiomyocyte and glomeruli losses, which resulted in function impairment of these organs and, finally, heart and renal failure (Keller et al., 2003; Bezerra et al., 2008). The long-term intake of different edible oils of common use in human nutrition (olive oil – monounsaturated lipid source; fish oil – polyunsaturated n-3 lipid oil; canola oil – monounsaturated ⫹ polyunsaturated n-3 lipid source; palm oil – saturated lipid source; soybean oil – polyunsaturated n-6 lipid source) has distinct effects on target organs of diabetic and/or hypertensive animal models. Table 82.1 shows the edible oils’ fatty acid composition and Table 82.2 shows the main edible oil effects on health.
82.2 BLOOD PRESSURE Experimental evidences from our laboratory (www.lmmc. uerj.br) demonstrated beneficial effects of edible oil supplementation to hypertensive or to diabetic and hypertensive Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
animals. In spontaneously hypertensive rats (SHR), fish oil supplementation led to a significant reduction in blood pressure (BP) (Figure 82.1) (Aguila et al., 2004). In hypertensive and streptozotocin-diabetic rats (SHR-Stz), fish oil attenuated the usual cardiac adverse remodeling in these animals, an effect that was less marked with both the palm oil and olive oil supplementation (Medeiros et al., 2005). The olive oil extract administered intravenously reduced systolic, diastolic, and mean arterial BP in normotensive rats, mimicking the effects of the calcium channel blocker drug verapamil (Gilani et al., 2005). Additionally, phenols and oleic acid may contribute to improved endothelial function by reducing reactive oxygen species (ROS) (Acin et al., 2007). Palm oil has the potential of elevating the aortic cyclic guanosine monophosphate (cGMP), which is a biologically active nitric oxide (NO) signaling mediator (Ganafa et al., 2002). NO diffuses out of endothelial cells, where it is synthesized, and stimulates guanylate cyclase in vascular smooth muscle cells (SMC), causing vascular relaxation (Carr and Frei, 2000). This effect of palm oil administration could explain the reduction of BP in SHR and consequent reduction in cardiomyocyte hypertrophy and intramyocardial vessel alterations (Aguila et al., 2004). Palm oil supplementation during 4 weeks increased the aortic prostacyclin (PGI2) and reduced the thromboxane A2 (TXA2) in Sprague-Dawley rats (Sugano and Imaizumi, 1991). Moreover, palm oil reduced buthionine sulfoximineinduced oxidative stress and attenuated hypertension by mechanisms involving changes in endothelium-derived factors (Ganafa et al., 2002). It is noteworthy to say that
787
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
788
SECTION | II
Cardiac Aspects
TABLE 82.1 Fatty acid composition of soybean, canola, olive, palm and fish oil (g fatty acid per 100 g total fatty acids). Fatty acids
Edible oils Soybean
Canola
Caprylic (8:0)
–
–
Capric (10:0)
–
Lauric (12:0) Myristic (14:0)
Palm
Fish
–
–
–
–
–
–
–
–
–
–
0.2
–
–
–
–
0.9
2.0
10.3
5.4
9.1
43.4
16.0
Palmitoleic (16:1)
0.1
0.3
0.8
0.4
6.0
Stearic (18:0)
3.9
1.6
2.5
4.9
4.0
Oleic (18:1)
22.1
56.3
81.6
39.6
18.0
Linoleic (18:2)
54.8
25.0
5.3
9.8
4.0
Linolenic (18:3)
7.5
8.4
0.7
0.3
–
Arachidic (20:0)
0.4
0.6
-
0.5
–
Eicosenoic (20:1)
0.2
1.3
–
–
10.0
Arachidonic (20:4)
–
–
–
–
1.1
Eicosapentaenoic (20:5)
–
–
–
–
9.9
0.4
0.4
–
–
–
Erucic*, Cetoleic* (22:1)
–
0.5
–
–
8.0
Docosapentaenoic (22:5)
–
–
–
–
3.9
Docosahexaenoic (22:6)
–
–
–
–
11.1
Lignoceric (24:0)
–
0.2
–
–
Palmic (16:0)
Behenic (22:0)
Olive
–
⌺ Saturated fatty acid
15.0
8.2
11.6
49.4
22.0
⌺ Monounsaturated fatty acid
22.4
58.4
82.4
40.0
42.0
⌺ Polyunsaturated fatty acid
62.3
33.4
6.0
10.1
30.0
7.3
3.0
7.6
32.7
0.3
n-6 Fatty acid/n-3 fatty acid ratio
Lipid numbers take the form C:D, where C is the number of carbon atoms in the fatty acid and D is the number of double bonds in the fatty acid. * Erucic in plant lipids, cetoleic in fish oil.
palm oil has a relatively higher proportion of saturated and antioxidant vitamins than most other major oils. In addition to its antioxidant potential and effects on BP, beneficial effects of palm oil on arterial thrombosis have been reported (Osim et al., 1996).
n-3 Polyunsaturated fatty acid provides protection due to its ability to suppress inflammation or coagulation by interfering with the proinflammatory, procoagulation prostanoids, thromboxanes, and/or leukotrienes production (Spurney et al., 1994). Studies in animal models demonstrated that
CHAPTER | 82 Beneficial Effects of Olive Oil Compared with Fish, Canola, Palm and Soybean Oils
789
TABLE 82.2 Edible oils effects on the health. Edible oil
Contents-Effects
Soybean
Precursor formation from arachidonic acid: Thromboxane A2 (platelet aggregator and vasoconstricitve) and Leucotriene B4 (pro-inflammatory, cell adhesive and chemotactic); competition between n-6 and n-3 fatty acid by Δ6 desaturase
Canola
Linolenic acid. Eicosapentaenoic and docosahexaenoic acids precursor. Linolenic acid competes with linoleic acid for desaturation and chain-elongation. Improves experimental myocardial vascularization
Olive
Decreases membrane lipids peroxidation. Does not compete with linolenic acid for the same enzymes. Typical Mediterranean diet (rich in oleic acid). Monounsaturated and polyunsaturated fatty acids: reduces cardiovascular disease risk
Palm
Vitamins A and E. Experimental findings: increases prostacyclins I2 and decreases Thromboxane A2. Lauric acid (12:0) and miristic acid (14:0) have more effects on blood lipids
Fish
DHA (22:6n-3) and EPA (20:5n-3) source. TG hepatic synthesis inhibition. Thromboxane A3: biologically almost inactive. Inhibition Thromboxane A2: platelet aggregator and vasoconstriction. PGI3: platelet anti-aggregator and vasodilatation. Inhibition of smooth muscle cells proliferation. NO production: vasodilatation. Increases cell membrane fluidity
180 Control a Soybean
160
Olive
Blood pressure (mmHg)
a
abce a
canola oil, which contains approximately 10% alpha-linolenic n-3 fatty acid, reduced the severity of experimentally induced cardiac arrhythmia and ventricular fibrillation (McLennan and Dallimore, 1995). Since vascular endothelium is central to the proper maintenance of cardiovascular homeostasis, the reported cardiovascular benefits of polyunsaturated n-3 fatty acids of plant origin may also be evident at the vascular endothelial cell level (Abeywardena and Head, 2001a) and have potential myocardium and vasoprotective actions (Aguila and Mandarim-de-Lacerda, 2001, 2003b).
Palm
140
Canola abc
abcd
abc
a
120
abcde
Fish abcde
100
1
4
7
10
abcde
13
Week FIGURE 82.1 Blood pressure variation in spontaneously hypertensive rats supplemented with different edible oils (mean and standard error of the mean). ANOVA and post-hoc test of Newmans-Keuls: in signaled cases, when compared, p ⱕ 0.05, if: (a) when compared with Control group, (b) with Soybean group, (c) with Olive oil group, (d) with Palm oil group, and (e) with Canola oil group (according to Aguila et al., 2004).
82.3 CARDIAC STRUCTURE As for cardiac structure, hypertension and diabetes treatment targets to improve microvasculature. In this way, fish oil supplementation showed the best indicators of myocardial microvasculature in diabetic SHR, with palm and olive oil groups less beneficial than fish oil in this treatment. The vascular efficiency can be evaluated by stereology: the length density of intramyocardial arteries and the artery-tocardiomyocyte ratio (Mandarim-de-Lacerda, 2003). These findings suggest that any edible oil treatment is better than no treatment. The long-chain polyenoics of eicosapentaenoic acid and docosahexaenoic acid are reported to have several beneficial actions on the vasculature, including incorporation of polyunsaturated fatty acids into vascular membranes and the resulting changes in membrane fluidity that may inhibit the expression of adhesion molecules (De Caterina et al., 2000), proliferation of SMC and changing
790
82.4 ARTERIAL WALL STRUCTURE Different edible oils can also have distinct effects in the arterial wall: wall thickness, elastic lamellae composition and number, and tunica media SMC. The elastic laminae and SMC constitute, approximately, equal major fractions of the aortic wall, and it is particularly important to study the lamellae surface density on the grounds that it reflects the tunica media alterations. The lamellae surface density is more affected by changes in length and sinuosity of lamellae than its thickness (Aguila and Mandarimde-Lacerda, 2003a). SHR supplemented with canola oil showed higher lamellae surface density and smaller aorta wall thickness. Fish oil, and more specifically, eicosapentaenoic acid and docosahexaenoic acid, can inhibit the proliferation of vascular SMC (Nakayama et al., 1999). Furthermore, the docosahexaenoic acid triggers vascular
Cardiac Aspects
50 abcde 45 Number of cardiomyocyte nuclei (million)
the activity of several ion channels (Abeywardena and Head, 2001b). In addition, it has been proposed that long-chain polyunsaturated fatty acids serve as endogenous regulators of angiotensin-converting enzyme activity, superoxide ion, endogenous NO generation, and TGF-beta expression (Das, 2004). Cardiomyocyte hypertrophy, observed as a consequence of arterial hypertension, leads to potential cardiac ischemia due to the increased artery-to-cardiomyocyte distance, which is a sufficient stimulus to cardiomyocyte apoptosis and necrosis (Mandarim-de-Lacerda and Pereira, 2000; Bezerra et al., 2008). NO plays a role in attenuating cardiac remodeling and apoptosis via suppression of oxidative stress-mediated signaling pathways (Smith et al., 2005) and the endogenous NO generation is enhanced by long-chain polyunsaturated fatty acids (Das, 2004). Cardiomyocyte apoptosis occurs in diverse conditions and this event may be a contributing cause to the loss and functional abnormalities of cardiomyocytes with important pathophysiological consequences. The programmed cell death may be triggered in the stressed myocardium independently of the etiology of the overload (Anversa et al., 1997). The number of cardiomyocyte nuclei (N[cmyn]) had a negative correlation with BP (R ⫽ ⫺0.98, p ⬍ 0.01) and was maintained higher in hypertensive rats submitted to edible oil supplementation than in untreated SHR. The N[cmyn] was over 130% greater in the fish oil group, and more than 25% greater in the canola, palm, and olive oils groups compared to the untreated group (Figure 82.2). Therefore, nutritional management of hypertension and subsequent prevention/attenuation of left ventricular cardiomyocyte loss suggest a novel therapeutic strategy to be seriously considered in long-term hypertension treatment. The goal is to postpone the consequences of cardiomyocyte number decrease and heart failure in hypertension (Aguila et al., 2005b; Bezerra et al., 2008).
SECTION | II
40 abcd 35 abc 30 a
a
25
20
15
Control Soybean
Olive
Palm
Canola
Fish
FIGURE 82.2 Number of cardiomyocyte nuclei in the left ventricle of spontaneously hypertensive rats supplemented with different edible oils (mean and standard error of the mean). ANOVA and post-hoc test of Newman-Keuls: in signaled cases, when compared, p ⱕ 0.05, if: (a) when compared with Control group, (b) with Soybean group, (c) with Olive group, (d) with Palm group, and (e) with Canola group (according to Aguila et al., 2005b).
smooth muscle cell apoptosis, implicating a role in the vascular remodeling (Diep et al., 2000). Although palm oil supplementation reduced the BP in SHR, palm oil intake favored the increase in aorta wall thickness. Palm oil has almost 50% of saturated fatty acids with an unsaturatedto-saturated fatty acid ratio close to one. Studies show that saturated-fatty-acid-rich diets reduce the formation of vasodilator prostaglandins, and favor the elevation of BP (Grimsgaard et al., 1999), but palm oil is also rich in antioxidant vitamins, which can favor the reduction of buthionine sulfoximine-induced oxidative stress and attenuate hypertension by mechanisms involving changes in endothelium-derived factors (Ganafa et al., 2002). In SHR supplemented with edible oils, the number of aortic lamellae was smaller in the fish oil group but greater in the soybean oil, canola oil, and olive oil groups. Canola oil reduced aortic wall thickness, but palm oil did not. In the groups that were given fish, canola and olive oils, the tunica media smooth muscle cells were less hypertrophied (Figure 82.3) (Aguila et al., 2004).
82.5 KIDNEY STRUCTURE The common renal injury of diabetes and hypertension is a complex dynamic process involving several players such as inflammatory agents, cytokines, vasoactive agents and
791
CHAPTER | 82 Beneficial Effects of Olive Oil Compared with Fish, Canola, Palm and Soybean Oils
FIGURE 82.3 Photomicrographs of aortic wall sections from 12-month-old SHR (spontaneously hypertensive rat) supplemented with different edible oils. Sections were stained with Masson’s trichrome stain and taken with the same magnification (bar ⫽ 30 μm). All pictures have the endothelium at the right side: (A) SHR with fish oil supplementation, (B) SHR with soybean oil supplementation, (C) SHR with canola oil supplementation, (D) SHR with olive oil supplementation, and (E) SHR with palm oil supplementation. The number of aortic lamellae is smaller in the fish oil group but greater in the soybean oil, canola oil and olive oil groups. Canola oil reduces aortic wall thickness, but palm oil does not. The number of smooth muscle cells in the tunica media is smaller in the groups given fish oil, canola oil and olive oil (according to Aguila et al., 2004).
25
ab Number of glomeruli per kidney (thousand)
enzymes participating in extracellular matrix assembly, anchoring or degradation (Boffa et al., 2003) and renal physiology. This is partially enzyme cyclooxygenasedependent and prostaglandins are active in the regulation of vascular tone and salt and water homeostasis by modulating the glomerular hemodynamics and also by adjusting the distal nephron function. The cyclooxygenase enzyme system is the major pathway for arachidonic acid in the kidney and these products modulate the action of other hormones or autacoids, particularly their physiologic actions on renal vascular tone, mesangial and glomerular functions and salt and water handling (Dunn, 1987). Vasoactive eicosanoids contribute to hemodynamic adaptations to kidney disease in rats (Forpomes et al., 1997). Rich n-3 polyunsaturated fatty acid fat and n-6 polyunsaturated fatty acid sources tested in renal-insufficient dogs showed that animals receiving fish oil showed lower serum cholesterol concentration and tended to have lower urinary PGE2 and TXA2 excretion than untreated dogs. Conversely, animals receiving safflower oil had higher mean glomerular capillary pressure and larger glomeruli, and tended to have higher eicosanoid excretion rates than untreated dogs; this could partly explain the useful effects of n-3 polyunsaturated fatty acid and the detrimental effects of n-6 polyunsaturated fatty acid when administered on a long-term basis in renal insufficiency (Brown et al., 2000) (Figure 82.4). For example, better effects on renal cortex and decreased BP levels were seen in SHR that received long-term fish oil, palm oil and canola oil supplementation (Aguila et al., 2005a). In different experimental models that lead to BP elevation, the fish oil supplementation showed beneficial effects by decreasing BP: in SHR (Aguila et al., 2004, 2005a, b), in diabetic SHR (Medeiros et al., 2005, 2006) and in rat offspring of low-protein pregnancies (Catta-Preta et al., 2006). Furthermore, fish oil also decreased the response of perfused mesenteric resistance to exogenous norepinephrine stimulation, suggesting that the antihypertensive effects are due to a reduction in the vascular response (Chu et al., 1992).
20
b b
15
a
10
5
0
SHR
Untreated
Olive
Palm
Fish
SHR-Stz FIGURE 82.4 Number of glomeruli per kidney in SHR (spontaneously hypertensive rat) and SHR-Stz (streptozotocin-diabetic hypertensive rat) supplemented with different edible oils (mean and standard error of the mean). ANOVA and post-hoc test of Newman-Keuls: in signaled cases, when compared, p ⱕ 0.05, if: (a) when compared with SHR group, (b) with untreated SHR-Stz group (according to Medeiros et al., 2006).
This hypotensive effect of dietary n-3 fatty acids is associated with enhanced endothelium-dependent relaxation by a replacement of endogenous arachidonic acid and suppression of the concomitant release of vasoconstrictor prostaglandins from the endothelium (Kim et al., 1992). Oxidative stress could be involved in different inflammatory glomerular lesions caused by a series of mediators, including cytokines and chemokines that lead to leukocyte
792
activation, production of reactive oxygen species and increased glomerular damage. Additionally, different aspects should be considered concerning loss of glomeruli in both hypertension and diabetes. The estimation of glomeruli number and glomerular volume gives insight into the renal response to hypertension progression and/or antihypertensive treatment efficiency (Bezerra and Mandarim-de-Lacerda, 2005). Thus, diabetic patients with severe diabetic glomerulopathy had significantly less glomeruli compared with diabetic patients with mild or no glomerulopathy (Bendtsen and Nyengaard, 1992) and rat offspring of either hyperglycemic or diabetic mothers have fewer nephrons (Gross et al., 2005). Usual BP increases glomerulosclerosis, glomerular enlargement, and glomeruli loss in SHR, all of which have been prevented by fish, canola and palm oils or attenuated by long-term intake of olive and soybean oils. The most favorable effect has been seen with fish oil, followed by both canola and palm oils, and finally both olive and soybean oils (Aguila et al., 2005a). Another study confirms the idea of renal beneficial effects of fish oil, palm oil and olive oil intake supplementation (in sequence from the more to the less beneficial oil supplementation) to diabetic SHR. The long-term administration of these edible oils has beneficial effects on renal cortex, retarding the usual loss of glomeruli and attenuating the renal cortex adverse remodeling in diabetic SHR (Medeiros et al., 2006).
82.6 LIPID METABOLISM Eicosanoids and fatty acids can regulate gene transcription through peroxisome proliferator-actived receptors (PPARs) (Kersten et al., 2000). The best-characterized action of PPAR-alpha is mediation of the uptake and betaoxidation of fatty acids in various tissues including the liver and heart (Francis et al., 2003). Eicosapentaenoic acid and docosahexaenoic acid have been shown to bind and activate PPAR-gamma and a growing body of evidence suggests that PPAR-gamma ligands can inhibit tumor necrosis factor (TNF)-alpha, interleukin (IL)-6 and IL-1-beta expression in monocytes, inducible NO synthase, matrix metalloprotease9 and scavenger receptor-A expression in macrophages among others (Krey et al., 1997). Thus, n-3 polyunsaturated fatty acids may also mediate some of their effects on lipid metabolism by mechanisms involving PPAR-gamma (Lindi et al., 2003). PPAR-alpha and PPAR-gamma appear to have protective effects on the activity of vascular SMC and seem to be able to respond to a vast number of ligands, from fatty acids to eicosanoids and linoleic acid metabolites (Bishop-Bailey, 2000). Perhaps, other fatty acids and their metabolites such as eicosapentaenoic acid and docosahexaenoic acid, alpha-linolenic, monounsaturated fatty acid and saturated fatty acids can also affect PPARs. Leukotriene B4 (LTB4), a pro-inflammatory eicosanoid, binds to PPARalpha and induces the transcription of genes involved in
SECTION | II
Cardiac Aspects
omega- and beta-oxidation, which lead to induction of its own catabolism (Devchand et al., 1996). PPAR-gamma regulates the expression of adipocyte-secreted proteins, adipocytokines, which include leptin, TNF-alpha, and adiponectin (Combs et al., 2002). Adiponectin has been postulated to have anti-inflammatory effects, especially in endothelial cells and in macrophages (Ouchi et al., 2000). It underlines the importance of the quality of lipids in the diet when associations with the PPARs are made. The different fatty acid composition of the different lipid supplements probably activates the PPARs in a different manner. The role of PPARs and fatty acid metabolites in myocardium and arterial wall is an exciting question for future research.
82.7 INFLAMMATION Moreover, novel mediators generated from eicosapentaenoic acid and docosahexaenoic acid were first identified as potent compounds in resolving inflammatory exudates in tissues enriched with docosahexaenoic acid. Resolvin (resolution-phase interaction products) and docosatrienes were proposed as these novel series since they possess potent anti-inflammatory and immunoregulatory actions. Compounds derived from eicosapentaenoic acid carrying potent biological actions are designated E series and denoted resolvins of the E series (resolvin E1 or RvE1), and those biosynthesized from the precursor docosahexaenoic acid are denoted resolvins of the D series (resolvin D1 or RvD1) (Serhan et al., 2004). The endogenous lipid mediator RvE1 counter-regulates leukocyte-mediated tissue injury and proinflammatory gene expression, demonstrating an endogenous mechanism that may underlie the beneficial actions of n-3 eicosapentaenoic acid and provide targeted approaches for the treatment of inflammation (Arita et al., 2005).
SUMMARY POINTS ●
●
●
The long-term intake of different edible oils of common use in human nutrition has shown distinct effects on target organs of diabetic and/or hypertensive animal models. Fish, canola and palm oils have controlled the blood pressure alteration more than olive and soybean oils. Fish oil has also treated usual cardiac adverse remodeling better than both the palm and olive oils. Fish, canola and olive oils have shown beneficial effects in controlling the hypertrophy of the tunica media smooth muscle cells. Fish, palm and olive oil supplementation (in sequence from the more to the less beneficial oil supplementation) have shown beneficial effects on renal cortex, retarding the usual loss of glomeruli and attenuating the renal cortex adverse remodeling.
CHAPTER | 82 Beneficial Effects of Olive Oil Compared with Fish, Canola, Palm and Soybean Oils
●
Hypertension and diabetes nutritional management and the subsequent beneficial effects on target organs suggest a novel therapeutic strategy to be seriously considered in long-term treatment. The goal is to postpone the consequences of cardiomyocyte and glomeruli number decrease and heart and renal failure in hypertension and diabetes.
REFERENCES Abeywardena, M.Y., Head, R.J., 2001a. Dietary polyunsaturated fatty acid and antioxidant modulation of vascular dysfunction in the spontaneously hypertensive rat. Prostaglandins Leukot. Essent. Fatty Acids 65, 91–97. Abeywardena, M.Y., Head, R.J., 2001b. Longchain n-3 polyunsaturated fatty acids and blood vessel function. Cardiovasc. Res. 52, 361–371. Acin, S., Navarro, M.A., Perona, J.S., Arbones-Mainar, J.M., Surra, J.C., Guzman, M.A., Carnicer, R., Arnal, C., Orman, I., Segovia, J.C., Osada, J., Ruiz-Gutierrez, V., 2007. Olive oil preparation determines the atherosclerotic protection in apolipoprotein E knockout mice. J. Nutr. Biochem. 18, 418–424. Aguila, M.B., Mandarim-de-Lacerda, C.A., 2001. Blood pressure, ventricular volume and number of cardiomyocyte nuclei in rats fed for 12 months on diets differing in fat composition. Mech. Ageing. Dev. 122, 77–88. Aguila, M.B., Mandarim-de-Lacerda, C.A., 2003a. Aorta wall quantitative alterations due to different long-term high-fat diet in rats. Food Chem. Toxicol. 41, 1391–1397. Aguila, M.B., Mandarim-de-Lacerda, C.A., 2003b. Heart and blood pressure adaptations in Wistar rats fed with different high-fat diets for 18 months. Nutrition 19, 347–352. Aguila, M.B., Pinheiro, A.R., Aquino, J.C., Gomes, A.P., Mandarimde-Lacerda, C.A., 2005a. Different edible oil beneficial effects (canola oil, fish oil, palm oil, olive oil, and soybean oil) on spontaneously hypertensive rat glomerular enlargement and glomeruli number. Prostaglandins Other Lipid Mediat. 76, 74–85. Aguila, M.B., Pinheiro, A.R., Mandarim-de-Lacerda, C.A., 2005b. Spontaneously hypertensive rats left ventricular cardiomyocyte loss attenuation through different edible oils long-term intake. Int. J. Cardiol. 100, 461–466. Aguila, M.B., Sa Silva, S.P., Pinheiro, A.R., Mandarim-de-Lacerda, C.A., 2004. Effects of long-term intake of edible oils on hypertension and myocardial and aortic remodelling in spontaneously hypertensive rats. J. Hypertens. 22, 921–929. Anversa, P., Olivetti, G., Leri, A., Liu, Y., Kajstura, J., 1997. Myocyte cell death and ventricular remodeling. Curr. Opin. Nephrol. Hypertens. 6, 169–176. Arita, M., Bianchini, F., Aliberti, J., Sher, A., Chiang, N., Hong, S., Yang, R., Petasis, N.A., Serhan, C.N., 2005. Stereochemical assignment, antiinflammatory properties, and receptor for the omega-3 lipid mediator resolvin E1. J. Exp. Med. 201, 713–722. Bendtsen, T.F., Nyengaard, J.R., 1992. The number of glomeruli in type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetic patients. Diabetologia 35, 844–850. Bezerra, D.G., Andrade, L.M., Cruz, F.O., Mandarim-de-Lacerda, C.A., 2008. Atorvastatin attenuates cardiomyocyte loss in adult rats from protein-restricted dams. J. Card. Fail. 14, 151–160. Bezerra, D.G., Mandarim-de-Lacerda, C.A., 2005. Beneficial effect of simvastatin and pravastatin treatment on adverse cardiac remodelling
793
and glomeruli loss in spontaneously hypertensive rats. Clin. Sci. (Lond.) 108, 349–355. Bishop-Bailey, D., 2000. Peroxisome proliferator-activated receptors in the cardiovascular system. Br. J. Pharmacol. 129, 823–834. Boffa, J.J., Lu, Y., Placier, S., Stefanski, A., Dussaule, J.C., Chatziantoniou, C., 2003. Regression of renal vascular and glomerular fibrosis: role of angiotensin II receptor antagonism and matrix metalloproteinases. J. Am. Soc. Nephrol. 14, 1132–1144. Brown, S.A., Brown, C.A., Crowell, W.A., Barsanti, J.A., Kang, C.W., Allen, T., Cowell, C., Finco, D.R., 2000. Effects of dietary polyunsaturated fatty acid supplementation in early renal insufficiency in dogs. J. Lab. Clin. Med. 135, 275–286. Carr, A., Frei, B., 2000. The role of natural antioxidants in preserving the biological activity of endothelium-derived nitric oxide. Free. Radic. Biol. Med. 28, 1806–1814. Catta-Preta, M., Oliveira, D.A., Mandarim-de-Lacerda, C.A., Aguila, M.B., 2006. Adult cardiorenal benefits from postnatal fish oil supplement in rat offspring of low-protein pregnancies. Life Sci. 80, 219–229. Chu, Z.M., Yin, K., Beilin, L.J., 1992. Fish oil feeding selectively attenuates contractile responses to noradrenaline and electrical stimulation in the perfused mesenteric resistance vessels of spontaneously hypertensive rats. Clin. Exp. Pharmacol. Physiol. 19, 177–181. Combs, T.P., Wagner, J.A., Berger, J., Doebber, T., Wang, W.J., Zhang, B.B., Tanen, M., Berg, A.H., O’Rahilly, S., Savage, D.B., Chatterjee, K., Weiss, S., Larson, P.J., Gottesdiener, K.M., Gertz, B.J., Charron, M.J., Scherer, P.E., Moller, D.E., 2002. Induction of adipocyte complementrelated protein of 30 kilodaltons by PPARgamma agonists: a potential mechanism of insulin sensitization. Endocrinology 143, 998–1007. Das, U.N., 2004. Long-chain polyunsaturated fatty acids interact with nitric oxide, superoxide anion, and transforming growth factor-beta to prevent human essential hypertension. Eur. J. Clin. Nutr. 58, 195–203. De Caterina, R., Liao, J.K., Libby, P., 2000. Fatty acid modulation of endothelial activation. Am. J. Clin. Nutr. 71, 213S–223S. Devchand, P.R., Keller, H., Peters, J.M., Vazquez, M., Gonzalez, F.J., Wahli, W., 1996. The PPARalpha-leukotriene B4 pathway to inflammation control. Nature 384, 39–43. Diep, Q.N., Touyz, R.M., Schiffrin, E.L., 2000. Docosahexaenoic acid, a peroxisome proliferator-activated receptor-alpha ligand, induces apoptosis in vascular smooth muscle cells by stimulation of p38 mitogenactivated protein kinase. Hypertension 36, 851–855. Dunn, M., 1987. The role of arachidonic acid metabolites in renal homeostasis. Non-steroidal anti-inflammatory drugs renal function and biochemical, histological and clinical effects and drug interactions. Drugs 33 (suppl 1), 56–66. Forpomes, O., Combe, C., Cambar, J., Aparicio, M., 1997. CyclosporinA- and angiotensin-II-induced vasoreactivity in isolated glomeruli and cultured mesangial cells, 4 and 24 h after renal mass reduction: role of vasodilatory prostaglandins. Nephron 76, 337–344. Francis, G.A., Annicotte, J.S., Auwerx, J., 2003. PPAR agonists in the treatment of atherosclerosis. Curr. Opin. Pharmacol. 3, 186–191. Ganafa, A.A., Socci, R.R., Eatman, D., Silvestrov, N., Abukhalaf, I. K., Bayorh, M.A., 2002. Effect of palm oil on oxidative stressinduced hypertension in Sprague-Dawley rats. Am. J. Hypertens. 15, 725–731. Gilani, A.H., Khan, A.U., Shah, A.J., Connor, J., Jabeen, Q., 2005. Blood pressure lowering effect of olive is mediated through calcium channel blockade. Int. J. Food Sci. Nutr. 56, 613–620. Grimsgaard, S., Bonaa, K.H., Jacobsen, B.K., Bjerve, K.S., 1999. Plasma saturated and linoleic fatty acids are independently associated with blood pressure. Hypertension 34, 478–483.
794
Gross, M.L., Amann, K., Ritz, E., 2005. Nephron number and renal risk in hypertension and diabetes. J. Am. Soc. Nephrol. 16 (suppl 1), S27–S29. Keller, G., Zimmer, G., Mall, G., Ritz, E., Amann, K., 2003. Nephron number in patients with primary hypertension. N. Engl. J. Med. 348, 101–108. Kersten, S., Desvergne, B., Wahli, W., 2000. Roles of PPARs in health and disease. Nature 405, 421–424. Kim, P., Shimokawa, H., Vanhoutte, P.M., 1992. Dietary omega-3 fatty acids and endothelium-dependent responses in porcine cerebral arteries. Stroke 23, 407–413. Krey, G., Braissant, O., L’Horset, F., Kalkhoven, E., Perroud, M., Parker, M.G., Wahli, W., 1997. Fatty acids, eicosanoids, and hypolipidemic agents identified as ligands of peroxisome proliferator-activated receptors by coactivator-dependent receptor ligand assay. Mol. Endocrinol. 11, 779–791. Lindi, V., Schwab, U., Louheranta, A., Laakso, M., Vessby, B., Hermansen, K., Storlien, L., Riccardi, G., A, A.R., 2003. Impact of the Pro12Ala polymorphism of the PPAR-gamma2 gene on serum triacylglycerol response to n-3 fatty acid supplementation. Mol. Genet. Metab. 79, 52–60. Mandarim-de-Lacerda, C.A., 2003. Stereological tools in biomedical research. An. Acad. Bras. Cienc. 75, 469–486. Mandarim-de-Lacerda, C.A., Pereira, L.M., 2000. Numerical density of cardiomyocytes in chronic nitric oxide synthesis inhibition. Pathobiology 68, 36–42. McLennan, P.L., Dallimore, J.A., 1995. Dietary canola oil modifies myocardial fatty acids and inhibits cardiac arrhythmias in rats. J. Nutr. 125, 1003–1009. Medeiros, F.J., Aguila, M.B., Mandarim-de-Lacerda, C.A., 2006. Renal cortex remodeling in streptozotocin-induced diabetic spontaneously hypertensive rats treated with olive oil, palm oil and fish oil from Menhaden. Prostaglandins Leukot. Essent. Fatty Acids 75, 357–365. Medeiros, F.J., Mothe, C.G., Aguila, M.B., Mandarim-de-Lacerda, C.A., 2005. Long-term intake of edible oils benefits blood pressure and myocardial structure in spontaneously hypertensive rat (SHR) and
SECTION | II
Cardiac Aspects
streptozotocin diabetic SHR. Prostaglandins Other Lipid Mediat. 78, 231–248. Nakayama, M., Fukuda, N., Watanabe, Y., Soma, M., Hu, W.Y., Kishioka, H., Satoh, C., Kubo, A., Kanmatsuse, K., 1999. Low dose of eicosapentaenoic acid inhibits the exaggerated growth of vascular smooth muscle cells from spontaneously hypertensive rats through suppression of transforming growth factor-beta. J. Hypertens. 17, 1421–1430. Osim, E.E., Owu, D.U., Etta, K.M., 1996. Arterial pressure and lipid profile in rats following chronic ingestion of palm oil diets. Afr. J. Med. Med. Sci. 25, 335–340. Ouchi, N., Kihara, S., Arita, Y., Okamoto, Y., Maeda, K., Kuriyama, H., Hotta, K., Nishida, M., Takahashi, M., Muraguchi, M., Ohmoto, Y., Nakamura, T., Yamashita, S., Funahashi, T., Matsuzawa, Y., 2000. Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-kappaB signaling through a cAMP-dependent pathway. Circulation 102, 1296–1301. Serhan, C.N., Arita, M., Hong, S., Gotlinger, K., 2004. Resolvins, docosatrienes, and neuroprotectins, novel omega-3-derived mediators, and their endogenous aspirin-triggered epimers. Lipids 39, 1125–1132. Smith Jr., R.S., Agata, J., Xia, C.F., Chao, L., Chao, J., 2005. Human endothelial nitric oxide synthase gene delivery protects against cardiac remodeling and reduces oxidative stress after myocardial infarction. Life Sci. 76, 2457–2471. Sowers, J.R., Epstein, M., Frohlich, E.D., 2001. Diabetes, hypertension, and cardiovascular disease: an update. Hypertension 37, 1053–1059. Spurney, R.F., Ibrahim, S., Butterly, D., Klotman, P.E., Sanfilippo, F., Coffman, T.M., 1994. Leukotrienes in renal transplant rejection in rats. Distinct roles for leukotriene B4 and peptidoleukotrienes in the pathogenesis of allograft injury. J. Immunol. 152, 867–876. Sugano, M., Imaizumi, K., 1991. Effect of palm oil on lipid and lipoprotein metabolism and eicosanoid production in rats. Am. J. Clin. Nutr. 53, 1034S–1038S.
Chapter 83
Olive Oil and Acute Coronary Syndromes: The CARDIO2000 Case-control Study Demosthenes Panagiotakos1 and Rena Kosti2 1 2
Department of Nutrition–Dietetics, Harokopio University, Athens, Greece Unit of Human Nutrition, Department of Food Science and Technology, Agricultural University of Athens, Greece
83.1 INTRODUCTION: MEDITERRANEAN DIET AND CORONARY HEART DISEASE Over the years, the Mediterranean diet has been promoted as a model for healthy eating (Hu, 2003). The diet, however, has been surrounded by as much myth as scientific evidence (Hu, 2003). Actually, there is no single ‘Mediterranean diet’. More than 15 countries border the Mediterranean Sea, and their dietary habits, the types of food produced, and their cultures vary considerably (Hu, 2003). Cardiovascular disease is the main cause of death in industrialized countries, but incidence rates have marked geographic differences (Estruch et al., 2006). The low incidence of coronary heart disease (CHD) in Mediterranean countries has been partly ascribed to dietary habits (Tunstall-Pedoe et al., 1999). Moreover, research showed that adherence to the Mediterranean diet is associated with reduced markers of vascular inflammation (Chrysohoou et al., 2004). These beneficial effects on surrogate markers of cardiovascular risk add biological plausibility to the epidemiologic evidence that supports a protective effect of the Mediterranean diet (Estruch et al., 2006). Therefore, the hallmark of the traditional Mediterranean diet is considered to be the higher levels of consumption of olive oil (Hu, 2003).
stress (Owen et al., 2000a). The physical methods used to produce olive oil preserve many of its antioxidant compounds (Waterman and Lockwood, 2007). Olive oil is approximately 72% oleic acid, a monounsaturated fatty acid which has one double bond, making it much less susceptible to oxidation and contributing to the antioxidant action, high stability, and long shelf life of olive oil (Newmark, 1997; Owen et al., 2000b). In addition, a range of phenols in olive oil provides some of its health benefits. The total phenolic content has been reported to be in the range of 196–500 mg kg⫺1 (Owen et al., 2000a). Olive oil phenols can be divided into three categories: simple phenols, secoiridoids, and lignans, all of which inhibit auto-oxidation. Major phenols include hydroxytyrosol, tyrosol, oleuropein, (Perona et al., 2006) and ligstroside (Owen et al., 2000a). Hydroxytyrosol and oleuropein scavenge free radicals and inhibit low-density lipoprotein (LDL) oxidation (Owen et al., 2000a; Visioli et al., 2002). Using hydroxyl radical scavenging as a measure of antioxidant capacity, research concluded that olive oil has a higher antioxidant capacity than seed oils and extra virgin olive oil is more potent than refined virgin olive oil due to its higher concentration of antioxidants (Owen et al., 2000b).
83.2 ACTIVE COMPONENTS OF OLIVE OIL
83.3 HEALTH BENEFITS OF OLIVE OIL CONSUMPTION
The olive tree, Olea europaea, produces the olive fruit. Olive oil is believed to exert its biological benefits mainly via constituent antioxidants (Waterman and Lockwood, 2007). Although the composition of olive oil is complex, the major groups of compounds thought to contribute to its observed health benefits include oleic acid, phenolics, and squalene, all of which have been found to inhibit oxidative
Indeed, for centuries, olive oil has been treasured in Greece and other Mediterranean countries for its healing and nutritional properties. Cumulative evidence suggests that olive oil may have a role in the prevention of coronary disease and several types of cancer because of its high levels of monounsaturated fatty acids and polyphenolic compounds (Hu, 2003).
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
795
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
796
Virgin olive oil retains all the lipophilic components of the fruit, α-tocopherol, and phenolic compounds with strong antioxidant and anti-inflammatory properties (Visioli and Galli, 2001; Beauchamp et al., 2005). Antioxidant and anti-inflammatory properties and improvements in endothelial dysfunction and the lipid profile have been reported for dietary polyphenols (Zern and Fernandez, 2005). Studies have recently suggested that Mediterranean health benefits may be due to synergistic combination of phytochemicals and fatty acids (Fortes, 2005). To date, most of the protective effect of olive oil within the Mediterranean diet has been attributed to its high monounsaturated fatty acid content (FDA, 2005). However, if the effect of olive oil can be attributed solely to its monounsaturated fatty acid content, any type of olive oil, rapeseed or canola oil, or monounsaturated fatty acid-enriched fat would provide similar health benefits (Covas et al., 2006). Whether the beneficial effects of olive oil on the cardiovascular system are exclusively due to oleic acid remains to be elucidated (Covas et al., 2006). The minor components, particularly the phenolic compounds, in olive oil may contribute to the health benefits derived from the Mediterranean diet (Covas et al., 2006).
83.4 DIETARY LIPIDS AND CORONARY HEART DISEASE (CHD) The role of dietary lipids in the etiology of coronary heart disease (CHD) continues to evolve as we gain a better understanding of the metabolic effects of individual fatty acids and their impact on surrogate markers of risk. Saturated fatty acids have been associated with a higher risk of CHD, whereas monounsaturated fatty acids (MUFA) and n-3 polyunsaturated fatty acids have protective effects against CHD (Wahrburg, 2004). The Seven Countries Study yielded the first convincing epidemiologic evidence for a negative correlation between the dietary intake of MUFA and mortality from CHD (Keys et al., 1986). Moreover, regression analysis of data of the Nurses Health study confirmed a protective effect of MUFA with regard to CHD risk (Hu et al., 1997). A protective role for olive oil on mortality among patients with a previous myocardial infarction has been found in the GISSI trial (Barzi et al., 2003), while casecontrol studies have shown conflicting results, indicating no association (Gramenzi et al., 1990; Bertuzzi et al., 2002) or inverse association (Fernandez-Jarne et al., 2002) between olive oil and risk of CHD. A Greek case–control study (Tzonou et al., 1993), focused on the type of cooking oil, reported no significant protection from MUFA intake against CHD, while investigators from the EPIC study showed that olive oil intake, per se, was inversely associated with both systolic and diastolic blood pressure, surrogate markers of CHD (Psaltopoulou et al., 2004).
SECTION | II
Cardiac Aspects
In this chapter, the main hypothesis tested was whether the prevalence of a first, non-fatal event of an acute coronary syndrome (ACS), myocardial infarction (MI) or unstable angina (UA), in a Greek sample, was associated with the pattern of olive oil and other oils and fats consumption in daily cooking or preparation of food.
83.5 DESIGN OF CARDIO2000 STUDY The CARDIO2000 is a multicenter case-control study that explores the association between several socio-demographic, nutritional, lifestyle and clinical factors with the risk of developing non-fatal ACS. Eight hundred and forty-eight patients participated in the study, randomly selected from various Greek hospitals, with a first symptom of CHD in their life. The number of participants was determined through power analysis, in order to evaluate a minimum difference of 7% in odds ratios per one unit increase in the explanatory variables (statistical power 0.80, p ⬍ 0.05). The inclusion criteria for the patients were: ●
●
First event of acute MI diagnosed by two or more of the following features: typical electrocardiographic changes, compatible clinical symptoms, specific diagnostic enzyme elevations, or First diagnosed unstable angina corresponding to class III of the Braunwald classification (The Joint European Society of Cardiology, 2000).
In addition, the study investigators randomly selected 1078 subjects (830 males and 248 females) without any clinical symptoms or suspicions of cardiovascular disease, in their medical history, matched to the patients by age (⫾3 years), sex, and region (i.e., the controls). Information regarding medical factors was retrieved from medical records and lifestyle characteristics information through a detailed questionnaire. Nutritional habits were evaluated with a semi-quantitative food-frequency questionnaire (Trichopoulou et al., 1995). In addition, consumption of olive oil, vegetable seed oils, butter and margarines was assessed by asking all participants about the use of oils in daily cooking and/or preparation of food (e.g., addition to salads, etc.). Data concerning alcohol consumption, smoking status, physical activity and education level, as well as presence of premature CHD among firstdegree relatives, weight and height, were also recorded. Hypertension was defined as systolic/diastolic blood pressure ⬎140/90 mmHg or use of special treatment, hypercholesterolemia as total cholesterol ⬎200 mg dl⫺1 or use of lipid-lowering agents and diabetes as fasting blood glucose ⬎125 mg dl⫺1 or use of antidiabetic medication. The CARDIO2000 study was approved by the Ethics Committee of the Department of Cardiology, Athens Medical School. Further details regarding the CARDIO2000 study have been previously published (Panagiotakos et al., 2002).
CHAPTER | 83 Olive Oil and Acute Coronary Syndromes: The CARDIO2000 Case-control Study
83.6 RESEARCH OBSERVATIONS OF THE CARDIO2000 STUDY Results showed that exclusive use of olive oil was reported by 65.2% of controls and 58.6% of patients (p ⫽ 0.002), whereas no use of olive oil was reported by 16.3% of controls and 22.4% of patients (p ⫽ 0.002). Olive oil plus other oils or fats were consumed by 18.5% of the controls and 19% of patients (p ⫽ 0.002) (Table 83.1). Particularly among patients, no significant associations between oil consumption pattern and several demographic and clinical characteristics were observed. However, in controls, oil consumption pattern was associated with age, sex, BMI, smoking, prevalence of hypertension and hypercholesterolemia and these factors were taken into account during the regression analysis. Moreover, in both patients and controls, it was observed that those who used olive oil had lower consumption of white meat (p ⬍ 0.001) and higher consumption of fish (p ⬍ 0.001), legumes (p ⫽ 0.002), fruits and vegetables (p ⬍ 0.001), compared to non-olive consumers. The age- and sex-adjusted odds ratio of having ACS was 0.66 (95% CI 0.52 to 0.84) for those consuming olive oil exclusively, compared to those not using olive oil, whereas consumption of both olive oil and other oils/fats did not show any significant relationship. Following the adjustments for BMI, smoking, physical activity level, educational status, the presence of family history of CHD, as well as hypertension, hypercholesterolemia and diabetes, multiple logistic regression analysis revealed that the exclusive use of olive oil was associated with 47% lower likelihood of having ACS (95% CI: 0.4–0.71), after controlling for the factors mentioned above. It should be also mentioned that consumption of olive oil in combination with other oils/fats was not significantly associated with lower odds of ACS compared to no olive oil consumption (p ⫽ 0.14). The association between olive oil intake and likelihood of ACS was not altered following the adjustment for the consumption of various food groups or alcohol (Table 83.2, Figure 83.1).
TABLE 83.1 Frequency of consumption of oils (%). Controls
Patients p
No use of olive oil (reference category %)
16.3
22.4
0.002
Exclusive use of olive oil (%)
62.5
58.6
0.002
Olive plus other oils or fats (%)
18.5
19.0
0.002
Oil category
Exclusive olive oil intake was more common in controls compared to patients.
797
Finally, residual confounding may still exist, thus the analysis was stratified by hypercholesterolemia, diabetes and hypertension status. It was found that in people who had hypercholesterolemia before the acute cardiac event, exclusive olive oil consumption was associated with 0.55 times lower likelihood of having ACS (95% CI: 0.35–0.81), while when diabetes status was taken into account no significant associations were observed. Finally, when the analysis was stratified by hypertension status, it was observed that exclusive olive oil consumption was associated with a 48% lower likelihood of ACS (odds ratio ⫽ 0.52, 95% CI: 0.34–0.81) among hypertensive subjects, as well as among people who had normal blood pressure levels (odds ratio ⫽ 0.59, 95% CI: 0.42–0.85). No other significant interactions were observed between pattern of olive oil intake and various socio-demographic, lifestyle or other clinical characteristics and the likelihood of having ACS. The key facts of CARDIO2000 study are presented in Table 83.3.
83.7 AN ANALYSIS OF THE ROLE OF OLIVE OIL ON ACUTE CORONARY SYNDROME This chapter, based on the results of the CARDIO2000 study, reveals that exclusive use of olive oil during cooking and as added oil is significantly associated with lower odds of having acute coronary events, even after controlling for several potential confounding risk factors. Greece is a country that traditionally uses olive oil during the preperation of food. In the EPIC study, the highest olive oil consumption was reported in Greece (Linseisen et al., 2002) and, according to the DAFNE project, Mediterranean countries had the least availability of added animal and vegetable fat and olive oil was the primary vegetable oil available in Greece and Spain (Byrd-Bredbenner et al., 2000). In the present study, 65% of controls and 59% of patients reported exclusive use of olive oil, whereas no use of olive oil was reported only by 16% of controls and 22% of patients. Moreover, people who reported exclusive use of olive oil also reported higher consumption of fish, legumes, fruits and vegetables, reflecting a dietary pattern close to the Mediterranean diet. However, the olive oil effect on the reduction of CHD risk was independent of these food groups. An inverse association between olive oil consumption and the risk of acute myocardial infraction (AMI) was also found in a case-control study from Spain (Fernandez-Jarne et al., 2002) in which the upper quintile of energy-adjusted olive oil was associated with a statistically significant 82% relative reduction in the risk of AMI, after adjustment for several confounders. Unfortunately, the studied data did not allow either the quantification of the oils and fats intake or the estimation of the amount of MUFA consumed, however, the recording of type of added oils and fats during food preparation (i.e., cooking, frying, salad dressing)
798
SECTION | II
Cardiac Aspects
TABLE 83.2 Odds ratios (95% confidence interval) for non-fatal acute coronary syndromes (acute myocardial infarction and unstable angina) associated with type of cooking oil among men and women of the study sample. Odds ratio
95% confidence interval
p
Oil category No use of olive oil (reference category)
1.00
–
–
Exclusive use of olive oil
0.53
0.34
0.71
⬍0.001
Olive plus other oils or fats
0.77
0.54
1.09
0.14
Age (per year)
1.03
1.01
1.04
⬍0.001
Male vs. female sex
2.73
2.10
3.63
⬍0.001
Family history of cardiovascular disease (yes vs. no)
4.07
3.20
5.19
⬍0.001
Current smoking habits (yes vs. no)
1.79
1.41
2.27
⬍0.001
Hypertension (yes vs. no)
1.94
1.54
2.45
⬍0.001
Hypercholesterolemia (yes vs. no)
3.76
3.01
4.69
⬍0.001
Diabetes mellitus (yes vs. no)
2.39
1.76
3.25
⬍0.001
Body mass index (per 1 kg/m2)
0.97
0.94
1.00
0.07
Physical inactivity (yes vs. no)
0.81
0.65
1.00
0.05
Higher vs. medium or lower education
0.94
0.70
1.20
0.68
Red meat intake (servings week⫺1)
1.68
1.40
2.03
⬍0.001
Fruits and vegetables intake (servings week⫺1)
0.66
0.57
0.76
0.001
Cereals intake (servings week⫺1)
1.15
0.98
1.33
0.06
Alcohol intake (wine glasses day⫺1)
1.09
0.97
1.22
0.14
Exclusive olive oil consumption was associated with lower odds of having ACS; however, use of olive oil together with other fats was not associated with the prevalence of the disease.
1,2 1
Age-sex adjusted
Odds ratio
0,8
Multi adjusted
0,6 0,4 0,2 0 Non consumers
Olive oil (exclusive)
Olive oil (exclusive)
FIGURE 83.1 Exclusive olive oil consumption and odds of having ACS.
could include less recall bias than a quantitative approach and provided important information that allow messages for public health to be carried out. The observed unique effect that olive oil use seems to exert on CHD risk can be attributed to several effects of olive oil, e.g., in the reduction of total and LDL cholesterol and triglyceride levels (Mata et al., 1992; Gardner and Kraemer 1995; Pieke et al., 2000). In addition, a diet rich in olive oil reduces the thrombotic propensity (Frost Larsen et al., 1999) and may slow the development of coronary atherosclerosis (Mata et al., 1997). MUFA also exert a protective effect against LDL oxidation (Mata et al., 1997) and oleic acid appears to interfere directly with the inflammatory response that characterizes early atherogenesis (Massaro et al., 1999).
CHAPTER | 83 Olive Oil and Acute Coronary Syndromes: The CARDIO2000 Case-control Study
TABLE 83.3 Key facts of CARDIO2000. ●
Exclusive use of olive oil was associated with 47% (95% CI: 0.4–0.71) lower likelihood of having ACS, compared to non-use, after adjusting for BMI, smoking, physical activity level, educational status, the presence of family history of coronary heart disease, as well as hypertension, hypercholesterolemia and diabetes
●
Consumption of olive oil in combination with other oils or fats was not significantly associated with lower odds of ACS compared to no olive oil consumption (p ⫽ 0.14)
●
Exclusive use of olive oil during food preparation seems to offer significant protection against CHD, irrespective of various clinical, lifestyle and other characteristics of the participants
●
The exclusive use of olive oil during cooking and food preparation should be promoted through cardiovascular disease prevention programs
799
when they were evaluated by a cardiologist. Another limitation is that studied data did not allow either the quantification of the oils and fats intake or the estimation of the amount of MUFA consumed, however, the recording of type of added oils and fats during food preparation (i.e., cooking, frying, salad dressing) could include less recall bias than a quantitative approach and provide important information that allow messages for public health to be carried out. Concerning information bias in this study, an attempt was made to avoid it through accurate and detailed data from subjects’ medical records. Moreover, the coronary patients who died at entry or the day after were not included in the study. This bias could influence the results, but, since this proportion of deaths was estimated between 2–4%, it is believed that it did not alter, significantly, the research findings. Furthermore, regarding the potential effect of uncontrolled, unknown confounders, an attempt was made to reduce it by using the same study base, both for patients and controls.
SUMMARY POINTS In the hypercholesterolemic subjects of the studied sample exclusive olive oil consumption was associated with 0.55 times likelihood of ACS and this is in line with the results of several metabolic studies (Mata et al., 1992; Gardner and Kraemer, 1995). Additionally, the Third Report of the NCEP Adult Treatment Panel (NCEP, 2001) recommended a program of therapeutic lifestyle changes which emphasized more on fat quality, promoting the ingestion of monounsaturated fatty acids, than on the quantity of dietary fat. Although metabolic studies have shown that olive oil improves lipid profile and glycemic control in diabetic patients (Garg, 1998), this study failed to show any significant association in this group of patients, probably due to the small number of diabetic people included in the studied sample. Finally, it was observed that exclusive olive oil consumption was associated with 48% lower likelihood of ACS among hypertensive subjects. Olive oil has been inversely associated with blood pressure in epidemiological studies (Psaltopoulou et al., 2004) and has been shown to lower blood pressure and reduce daily antihypertensive dosage among hypertensives (Ferrara et al., 2000).
83.8 MAIN METHODOLOGICAL LIMITATIONS It is worth mentioning that in case-control studies two main sources of systematic errors may exist, the selection and the recall bias. In order to eliminate selection bias in this study, an attempt was made to set objective criteria, both for patients and controls. However, insignificant misclassification may exist, since a small percentage of asymptomatic coronary patients may be wrongly assigned to controls, even
●
●
●
●
Exclusive use of olive oil was associated with 47% (95% CI: 0.4–0.71) lower likelihood of having ACS, compared to non-use, after adjusting for BMI, smoking, physical activity level, educational status, the presence of family history of coronary heart disease, as well as hypertension, hypercholesterolemia and diabetes. Consumption of olive oil in combination with other oils or fats was not significantly associated with lower odds of ACS compared to no olive oil consumption (p ⫽ 0.14). Exclusive use of olive oil during food preparation seems to offer significant protection against CHD, irrespective of various clinical, lifestyle and other characteristics of the participants. The exclusive use of olive oil during cooking and food preparation should be promoted through cardiovascular disease prevention programs.
REFERENCES Barzi, F., Woodward, M., Marsi, R.M., Tavazzi, L., Valagussa, F., Marchioli, R. GISSI-Prevenzione Investigators, et al., 2003. Mediterranean diet and all-causes mortality after myocardial infarction: results from the GISSI Prevenzione trial. Eur. J. Clin. Nutr. 57, 604–611. Beauchamp, G.K., Keast, R.S., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A., 2005. Phytochemistry: ibuprofen-like activity in extra-virgin olive oil. Nature. 437, 45–46. Bertuzzi, M., Tavani, A., Negri, E., La Vecchia, C., 2002. Olive oil consumption and risk of non-fatal myocardial infarction in Italy. Int. J. Epidemiol. 31, 1274–1275. Byrd-Bredbenner, C., Lagiou, P., Trichopoulou, A., 2000. A comparison of household food availability in 11 countries. J. Hum. Nutr. Diet. 13, 197–204. Chrysohoou, C., Panagiotakos, D.B., Pitsavos, C., Das, U.N., Stefanadis, C., 2004. Adherence to Mediterranean diet attenuates inflammation and
800
coagulation process in healthy adults: the ATTICA Study. J. Am. Coll. Cardiol. 44, 152–158. Covas, M.-I., Nyyssönen, K., Poulsen, H.E., Kaikkonen, J., Zunft, H.-J.F., Kiesewetter, H., Gaddi, A., de la Tore, R., Mursu, J., Bäumler, H., Nascetti, S., Salonen, J.T., Fitó, M., Virtanen, J., Marrugat, J. Eurolive study group, 2006. The effect of polyphenols in olive oil on heart disease risk factors. Ann. Intern. Med. 145, 333–341. Estruch, R., Martínez-Conzález, M.A., Corella, D., Salas-Salvadó, J., RuizGutiérrez, V., Covas, M.I., Fiol, M., Gómez-Garcia, E., López-Sabater, M.C., Vinyoles, E., Arós, F., Conde, M., Lahoz, C., Lapetra, J., Sáez, G., Ros, E. For the PREDIMED Study Investigators, 2006. Effects of a Mediterraneanstyle diet on cardiovascular risk factors. Ann. Intern. Med. 145, 1–11. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. 2001. Executive summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. JAMA 285, 2486–2497. Fernandez-Jarne, E., Martinez-Losa, E., Prado-Santamaria, M., BrugarolasBrufau, C., Serrano-Martinez, M., Martinez-Gonzalez, M.A., 2002. Risk of first non-fatal myocardial infarction negatively associated with olive oil consumption: a case–control study in Spain. Int. J. Epidemiol. 31, 474–480. Ferrara, L.A., Raimondi, A.S., d’Episcopo, L., Guida, L., Dello Russo, A., Marotta, T., 2000. Olive oil and reduced need for antihypertensive medications. Arch. Intern. Med. 160, 837–842. FDA allows qualified health claim to decrease risk of coronary heart disease [Press Release].Press release P04-100.Rockville, MD:U.S. Food and Drug Administration;1 November 2004.Accessed at www.fda. gov/bbs/topics/news/2004/NEW01129.html on 8 December 2005. Fortes, C., 2005. Mediterranean diet health benefits may be due to a synergistic combination of phytochemicals and fatty-acids. BMJ 331, E366. Frost Larsen, L., Jespersen, J., Marckmann, P., 1999. Are olive oil diets antithrombotic? Diets enriched with olive, rapeseed, or sunflower oil affect postprandial factor VII differently1–3. Am. J. Clin. Nutr. 70, 976–982. Gardner, C.D., Kraemer, H.C., 1995. Monounsaturated versus polyunsaturated dietary fat and serum lipids: a metaanalysis. Arterioscler. Throm. Vasc. Biol. 15, 1918–1927. Garg, A., 1998. High-monounsaturated fat diets for patients with diabetes mellitus: a meta-analysis. Am. J. Clin. Nutr. 67, 577S–582S. Gramenzi, A., Gentile, A., Fasoli, M., Negri, E., Parazzini, F., La Vecchia, C., 1990. Association between certain foods and risk of acute myocardial infarction in women. BMJ 300, 771–773. Hu, F.B., 2003. The Mediterranean diet and mortality – olive oil and beyond. N. Engl. J. Med. 348, 2595–2596. Hu, F.B., Stampfer, M.J., Manson, J.E., Rimm, E., Colditz, G.A., Rosner, B.A., Hennekens, C.H., Willett, W.C., 1997. Dietary fat intake and the risk of coronary heart disease in women. N. Engl. J. Med. 337, 1491–1499. Keys, A., Menotti, A., Karvonen, M.J., Aravanis, C., Blackburn, H., Buzina, R., Djordjevic, B.S., Dontas, A.S., Fidanza, F., Keys, M.A., et al., 1986. The diet and 15-year death rate in the Seven Countries Study. Am. J. Epidemiol. 124, 903–915. Linseisen, J., Bergstrom, E., Gafa, L., González, C.A., Thiébaut, A., Trichopoulou, A., Tumino, R., Navarro Sánchez, C., Martínez Garcia, C., Mattisson, A., Nilson, S., Welch, A., Spencer, E.A., Overvad, K., Tjønneland, A., Clavel-Chapelon, F., Kesse, E., Miller, A.B., Schulz, M., Botsi, K., Naska, A., Sieri, S., Sacerdote, C., Ocké, M.C., Peeters, P.H., Skeie, G., Engeset, D., Charrondière, U.R., Slimani, N., 2002. Consumption of added fats and oils in the European Prospective Investigation into Cancer and Nutrition (EPIC) centres across 10 European countries as assessed by 24-hour dietary recalls. Public Health Nutr. 5, 1227–1242.
SECTION | II
Cardiac Aspects
Massaro, M., Carluccio, M.A., De Caterina, R., 1999. Direct vascular antiatherogenic effects of oleic acid: a clue to the cardioprotective effects of the Mediterranean diet. Cardiologia 44, 507–513. Mata, P., Alvarez-Sala, L.A., Ubio, M.J., Nuño, J., De Oya, M., 1992. Effects of long-term monounsaturated vs polyunsaturated-enriched diets on lipoproteins in healthy men and women. Am. J. Clin. Nutr. 55, 846–850. Mata, P., Varela, O., Alonso, R., Lahoz, C., de Oya, M., Badimon, L., 1997. Monounsaturated and polyunsaturated n-6 fatty acid-enriched diets modify LDL oxidation and decrease human coronary smooth muscle cell DNA synthesis. Arterioscler. Thromb. Vasc. Biol. 17, 2088–2095. Myocardial infarction redefined – a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction, 2002. Eur. Heart J. 21, 1502–1513. Newmark, H.L., 1997. Squalene, olive oil, and cancer risk: a review and hypothesis. Cancer Epidemiol. Biomarkers Prev. 6, 1101–1103. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Wurtele, G., Spiegelhalder, B., Bartsch, H., 2000a. Olive-oil consumption and health: the possible role of antioxidants. Lancet Oncol. 1, 107–112. Owen, R.W., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2000b. Phenolic compounds and squalene in olive oils: the concentration and antioxidant potential of total phenols, simple plenols, secoiridoids, lignans and squalene. Food Chem. Toxicol. 38, 647–659. Panagiotakos, D.B., Pitsavos, C., Chrysohoou, C., Stefanadis, C., Toutouzas, P.K., 2002. Risk stratification of coronary heart disease, in Greece: final results from CARDIO2000 epidemiological study. Prev. Med. 35, 548–556. Perona, J.S., Cabello-Moruno, R., Ruiz-Gutierrez, V., 2006. The role of virgin olive oil components in the modulation of endothelial function. J. Nutr. Biochem. 17, 429–445. Pieke, B., von Eckardstein, A., Gülbahce, E., Chirazi, A., Schulte, H., Assmann, G., Wahrburg, U., 2000. Treatment of hypertriglyceridemia by two diets rich either in unsaturated fatty acids or in carbohydrates: effects on lipoprotein subclasses, lipolytic enzymes, lipid transfer proteins, insulin and leptin. Int. J. Obes. Rel. Metab. Dis. 24, 1286–1296. Psaltopoulou, T., Naska, A., Orfanos, P., Trichopoulos, D., Mountokalakis, T., Trichopoulou, A., 2004. Olive oil, the Mediterranean diet, and arterial blood pressure: the Greek European Prospective Investigation into Cancer and Nutrition (EPIC) study. Am. J. Clin. Nutr. 80, 1012–1018. Trichopoulou, A., Kouris-Blazos, A., Wahlqvist, M., Gnardellis, C., Lagiou, P., Polychronopoulos, E., Vassilakou, T., Lipworth, L., Trichopoulos, D., 1995. Diet and overall survival in elderly people. Br. Med. J. 311, 1457–1460. Tunstall-Pedoe, H., Kuulasmaa, K., Mähönen, M., Tolonen, H., Ruokokoski, E., Amouyel, P., 1999. Contribution of trends in survival and coronaryevent rates to changes in coronary heart disease mortality:10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet 353, 1547–1557. Tzonou, A., Kalandidi, A., Trichopoulou, A., Hsieh, C.C., Toupadaki, N., Willett, W., Trichopoulos, D., 1993. Diet and coronary heart disease: a case-control study in Athens, Greece. Epidemiology 4, 511–516. Visioli, F., Galli, C., 2001. Antiatherogenic components of olive oil. Curr. Atheroscler. Rep. 3, 64–67. Visioli, F., Poli, A., Gall, C., 2002. Antioxidant and other biological activities of phenols from olives and olive oil. Med. Res. Rev. 22, 65–75. Wahrburg, U., 2004. What are the health effects of fat? Eur. J. Nutr. 43 (suppl 1), I/6. Waterman, E., Lockwood, B., 2007. Active components and clinical applications of olive oil. Altern. Med. Rev. 12, 331–342. Zern, T.L., Fernandez, M.L., 2005. Cardioprotective effects of dietary phenols. J. Nutr. 135, 2291–2294.
Chapter 84
Olive Oil Consumption and Reduced Incidence of Hypertension: The SUN Study Alvaro Alonso1, Javier S. Perona2, Valentina Ruiz-Gutiérrez2 and Miguel A. Martínez-González3 1
Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, MN, USA Instituto de la Grasa, Seville, Spain 3 Department of Preventive Medicine, School of Medicine, University of Navarra, Pamplona, Spain
2
84.1 INTRODUCTION
84.2 OLIVE OIL AND HYPERTENSION
Hypertension (high blood pressure) is a major risk factor for cardiovascular disease. Hypertensive individuals have a higher risk of developing coronary heart disease, heart failure, stroke and kidney disease (Kannel, 2000). In addition to its harmful effect on cardiovascular disease and mortality, hypertension is highly prevalent. In Western Europe, more than 40% of the adult population can be considered hypertensive (Wolf-Maier et al., 2003). Both factors make hypertension a major public health problem: a recent estimate suggests that, each year, more than 7 million deaths worldwide are attributable to hypertension (Lopez et al., 2006) (Table 84.1). Different studies have shown that, in addition to genetic factors, both dietary and non-dietary factors can influence the risk of developing hypertension. Thus, excessive weight, lack of physical activity, excessive consumption of alcohol and sodium, and a low intake of potassium, are factors that have been consistently found to increase the risk of hypertension (Chobanian et al., 2003). Similarly, interventions aimed to tackle these risk factors have been shown to be effective in the prevention and treatment of hypertension (Appel et al., 2006). Despite our considerable knowledge on the association between diet and hypertension, some unanswered questions remain. Most studies evaluating the association between diet and risk of hypertension have been conducted in the US and northern Europe, regions with specific dietary patterns. The role of foods less represented in these populations in the prevention of hypertension, such as olive oil, is less clear.
Studies comparing the rates of cardiovascular disease and risk factors across countries have shown consistently that Mediterranean populations had a lower burden of these diseases (Keys et al., 1986). The food patterns followed in these areas, the so-called Mediterranean diets, have been proposed as one of the explanations for their lower incidence of cardiovascular disease. The traditional Mediterranean food pattern, in general, is characterized by a high consumption of fruits, vegetables, legumes and nuts,
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
801
TABLE 84.1 Key facts of hypertension. ●
●
●
●
●
Hypertension (or high blood pressure) is defined as levels of systolic blood pressure equal or greater than 140 mmHg or diastolic blood pressure equal or greater than 90 mmHg Hypertension is extremely prevalent worldwide. More than 40% of adults in Western countries and 25% worldwide can be considered hypertensives. This figure is higher in older individuals Individuals with hypertension have a higher risk of developing cardiovascular disease (heart attacks and strokes) and kidney disease Approximately 54% of all strokes, 47% of ischemic heart disease and 7.6 million deaths annually worldwide can be attributed to hypertension Obesity, lack of physical activity, high intake of sodium and alcohol, and low intake of potassium, fruits and vegetables are associated with a higher risk of developing hypertension. Other dietary factors might be associated but evidence is wanting
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
802
cereals, a low consumption of meat and meat products, and a moderate consumption of alcohol, particularly red wine. A major attribute of the Mediterranean diet is the important presence of monounsaturated fatty acids (MUFA), obtained from a high consumption of olive oil. The supported cardiovascular benefit of this traditional Mediterranean food pattern has spurred interest in the identification of its components that could be beneficial, including olive oil. Until recently, however, data on the possible benefits of olive oil in the prevention or treatment of hypertension were scarce. Only a few controlled experiments conducted in small groups of individuals and a trial suggesting that consumption of olive oil could reduce the need for antihypertensive medications on hypertensives supported a role for olive oil in the prevention of hypertension (Ferrara et al., 2000; Alonso et al., 2006).
84.3 OLIVE OIL AND HYPERTENSION IN THE SUN STUDY: METHODS The Seguimiento Universidad de Navarra (SUN, University of Navarra Follow-up) Study is a prospective dynamic cohort study started in 1999. A major objective of the SUN study was to determine the role of the traditional Mediterranean food pattern and its components on the prevention of obesity, hypertension, diabetes and cardiovascular disease. Details of the SUN study design have been published elsewhere (Seguí-Gómez et al., 2006). Below, we provide a brief description of its methods. Beginning in December of 1999, university graduates from the University of Navarra, registered nurses in some Spanish provinces, and members of some Spanish professional associations (all of them with college degrees), received a mailed questionnaire inviting them to participate in the SUN study and asking for personal information on lifestyle, diet and medical variables. In April 2008, more than 20 000 individuals had answered the baseline questionnaire. Subsequently, study participants received biennial mailed questionnaires, gathering information on changes in lifestyle and in the occurrence of some medical outcomes, including a diagnosis of hypertension. Overall, nearly 90% of participants have responded to the follow-up questionnaires. Diet was assessed using a semi-quantitative foodfrequency questionnaire, previously validated in a Spanish population (Martín-Moreno et al., 1993). This questionnaire requested information on the usual intake of 136 food items, including a specific question for olive oil consumption and questions about type of oils and fats used for cooking. The questionnaire gathered data on time spent doing exercise (17 different activities) and on sedentary behaviors. Other relevant information ascertained through the baseline questionnaire was smoking habits, self-reported weight and height, use of antihypertensive medications, history of any cardiovascular disease, cancer, diabetes,
SECTION | II Vascular Aspects Including Hypertension
hypercholesterolemia or hypertension, and family history of hypertension. Different validation studies conducted in the SUN population point to a high validity of self-reported weight and physical activity (Bes-Rastrollo et al., 2005; Martínez-González et al., 2005). New cases of hypertension in the SUN cohort were identified through follow-up questionnaires. Specifically, the biennial mailed questionnaires asked: ‘Has a physician diagnosed you with hypertension since you answered the previous questionnaire?’ Data from other populations show that self-reported information on diagnosis of hypertension, particularly among educated individuals such as our population, has enough validity for epidemiologic studies (Colditz et al., 1986; Vargas et al., 1997; Tormo et al., 2000). Nonetheless, to confirm the validity of our outcome ascertainment, we measured blood pressure in a random sample of 127 SUN participants residing in the metropolitan area of Pamplona to determine the specific validity of our follow-up questionnaires. The positive and negative predictive values for self-reported hypertension were, respectively, 82% and 85%. Details of this validation study have been previously published (Alonso et al., 2005).
84.4 OLIVE OIL AND HYPERTENSION IN THE SUN STUDY: RESULTS Table 84.2 shows the characteristics of SUN participants who answered the baseline questionnaire before January 2005 and met the inclusion criteria for our study on the association of dietary and lifestyle factors with the risk of hypertension, detailed in a previous publication (Beunza et al., 2007). The SUN cohort is relatively young, with a higher representation of women, and with a fair intake of fruits, vegetables and olive oil. We estimated the association between consumption of olive oil and the incidence of hypertension among the first 7000 SUN participants with 2 years of follow-up in 2004 (Alonso and Martínez-González, 2004). We excluded from this analysis individuals with presence of hypertension in the baseline questionnaire and those with extremely low or high caloric intakes, leaving 5573 participants available for analysis. Olive oil consumption was categorized in quintiles, after adjusting for energy intake using the residual methods (Willett and Stampfer, 1998). We identified 161 incident cases of hypertension among our participants (102 among men and 59 among women). Overall, a higher consumption of olive oil was associated with a lower risk of hypertension among men but not among women. Men in the highest category of olive oil intake had a significant 50% reduction in their risk of developing hypertension compared to those in the category with the lowest intake (Figure 84.1). There was a significant trend in the association between olive oil consumption and risk of hypertension in men, i.e., the higher the consumption of olive oil, the lower the risk of being diagnosed with hypertension.
803
CHAPTER | 84 Olive Oil Consumption and Reduced Incidence of Hypertension: The SUN Study
Women
Men
N
5880
3676
Age (years)
34.1 (9.5)
39.3 (11.1)
Men 1.5 Odds ratio
TABLE 84.2 Characteristics of SUN study participants at baseline, 1999–2005. These values correspond to participants included in the paper by Beunza et al. (2007).
1 0.5 0
Q1
Q2
Q3
Q4
Q5
Quintiles of olive oil consumption Women
BMI (kg/m2)
21.9 (3.7)
24.9 (3.1)
Smoking (% current)
25.3
21.6
Physical activity (METs-hour week⫺1)
22.0 (18.5)
28.5 (25.6)
Odds ratio
3 2 1 0 Q1
Family history of hypertension (%)
40.1
35.2
Hypercholesterolemia (%)
10.4
19.2
Total energy intake (kcal day⫺1)
2321 (568)
2522 (687)
Sodium intake (g day⫺1)
3.7 (2.8)
3.5 (2.6)
Alcohol intake (g day⫺1)
3.9 (5.8)
10.2 (12.3)
Fruit intake (g day⫺1)
351.8 (280.0)
293.6 (266.9)
Vegetable intake (g day⫺1)
541.9 (313.6)
448.3 (290.0)
Olive oil intake (g day⫺1)
13.2 (15.9)
11.5 (15.4)
Whole grain intake (g day⫺1)
14.1 (35.0)
14.4 (36.9)
Low-fat dairy intake (g day⫺1)
251.8 (256.0)
157.8 (217.5)
BMI: body mass index. Values represent means (standard deviations) unless otherwise stated.
The lack of association in women could be explained due to the lower number of hypertension cases among them. A high consumption of fruit and vegetables is an important characteristic of the traditional Mediterranean food pattern, and olive oil is usually consumed jointly with vegetables (as a dressing in salads, for example). A high intake of fruit and vegetables, also, could reduce levels of blood pressure (Appel et al., 1997). Therefore, it would be reasonable to study the joint association of fruit, vegetable and olive oil with the risk of hypertension. In a cross-sectional analysis of the SUN study, we observed that the consumption of fruit and vegetables was associated with a lower prevalence of undiagnosed hypertension particularly among individuals with low monounsaturated fatty acids intake, but not so among those
Q2
Q3
Q4
Q5
Quintiles of olive oil consumption
FIGURE 84.1 Odds ratio and 95% confidence intervals of hypertension according to olive oil consumption in men and women separately; SUN Study, 1999–2004. Analysis adjusted for age, body mass index, energy intake, alcohol consumption, calcium intake, and physical activity during leisure time. Adapted from Alonso and Martínez-González (2004). The figure represents the odds ratio of hypertension according to quintiles of olive oil intake, using individuals in the first quintile as reference category. The analysis is adjusted for different potential confounders. An odds ratio is a measure of association between a particular exposure (olive oil intake, in this case) and an outcome (incidence of hypertension here). It can be interpreted as how many times the outcome is more frequent between a particular level of the exposure compared to a reference category. In this case, in men the odds ratio of hypertension in the fifth quintile is lower than 1 suggesting that individuals with the highest intake of olive oil had a lower risk of hypertension. This is not apparent in women.
with high monounsaturated fatty acids intake (Alonso et al., 2004). More recently, we examined the same hypothesis prospectively in the SUN study. Again, fruit and vegetable intake was associated with a lower risk of developing hypertension among those with low olive oil intake but not among those with high olive oil intake (Figure 84.2) (Nuñez-Córdoba et al., 2008). The hazard ratio (95% confidence interval) of hypertension among individuals having 5 or more servings per day of fruit and vegetables compared to those having 2 or less was 0.56 (0.35, 0.89) and 1.12 (0.64, 1.96) among those with low and high olive oil consumption respectively. A significant inverse trend between fruit and vegetable consumption and risk of hypertension was observed only among individuals with low olive oil consumption. These results suggest that a saturation effect could exist for the joint beneficial effect of fruits, vegetables and olive oil on the risk of developing hypertension, with a reduction of hypertension risk if any of the exposures is present, but lack of further risk reduction with additional high consumption of the other foods.
84.5 OLIVE OIL AND HYPERTENSION IN OTHER STUDIES In addition to the results from the SUN cohort, other observational and experimental studies have provided additional
804
SECTION | II Vascular Aspects Including Hypertension
84.6 BIOLOGICAL MECHANISMS
Hazard ratio
1.2
0.8
0.4 High olive oil Low olive oil
0
>2 − 4
>4 − =5
Fruit & vegetables (servings/day) Low olive oil
The mechanisms through which olive oil can reduce the risk of hypertension are diverse, including modification of membrane phospholipids, reductions in the production of polar compounds during cooking, antioxidant and anti-inflammatory effects of polyphenols present in virgin olive oil, and potential protection of the endothelial function. A review of these mechanisms can be found in Alonso et al. (2006).
High olive oil
FIGURE 84.2 Hazard ratios of hypertension by consumption of fruit and vegetables, stratified by levels of olive oil consumption. Analysis adjusted for age, gender, total energy intake, body mass index, physical activity, alcohol consumption, family history of hypertension, sodium intake, lowfat dairy consumption, whole grains consumption, fish consumption, and smoking. Adapted from Núñez-Córdoba et al. (2008). This figure shows the association between fruit and vegetable intake and the risk of hypertension according to levels of olive oil intake. Association is measured using hazard ratios, considering those with the lowest intake of fruit and vegetables as reference category. A hazard ratio lower than 1 means that the risk of hypertension is lower in that category compared to the reference one. In this case, those with high intake of fruit and vegetables have a lower hazard ratio of hypertension only among the group with low olive oil intake, suggesting that fruit and vegetables could interact with intake of olive oil to protect against hypertension.
evidence of the beneficial effect of olive oil on blood pressure and the risk of hypertension. In a cross-sectional analysis of 20 343 participants of the Greek arm of the European Prospective Investigation into Cancer and Nutrition study, consumption of olive oil was associated with lower levels of both systolic and diastolic blood pressure, determined through direct measurement (Psaltopoulou et al., 2004). Numerous experimental studies have shown that olive oil supplementation could reduce levels of blood pressure. In a feeding trial conducted in a group of 31 hypertensives and 31 normotensives in southern Spain, 4-week supplementation of the diet with virgin olive oil reduced significantly blood pressure levels in hypertensive individuals compared to sunflower oil supplementation (Perona et al., 2004). More recently, supplementation with olive oil rich in phenolic compounds was effective in reducing blood pressure levels in a group of individuals from non-Mediterranean countries (Bondia-Pons et al., 2007). These results highlight the potential role of olive oil as a non-pharmacologic treatment of hypertension, and complement the evidence coming from observational studies, such as the SUN cohort. In the PREDIMED trial, a large randomized primary prevention trial of cardiovascular disease conducted in Spain, after a 3-month follow-up, mean systolic and diastolic blood pressure were significantly reduced in the group allocated to a Mediterranean-type diet supplemented with virgin olive oil (Estruch et al., 2006). These results are consistent with previous smaller trials (Esposito et al., 2004; Fito et al., 2005).
84.7 CONCLUSION Observational studies, which examine populations for long periods of time, and randomized experiments, conducted in a controlled setting for a limited time, point together to olive oil consumption as a beneficial component in any approach to prevent and treat hypertension in the general population. Further research has to determine potential interactions between olive oil intake and other diet components, and identify subgroups that will benefit the most from increasing their olive oil intake. Finally, more laboratory research has to focus on the mechanisms through which olive oil could produce this beneficial effect on blood pressure. New treatments for hypertension could result from this effort.
SUMMARY POINTS ●
●
●
●
●
●
Hypertension is a major public health problem, accounting for more than 7 million deaths each year worldwide. Diet has a major impact in the development of hypertension. A high intake of alcohol and sodium, as well as low intake of potassium, are factors consistently related with a higher risk of hypertension. Olive oil intake, as a major component of the Mediterranean dietary pattern, has been considered as a protective factor for developing hypertension. Until very recently, however, the evidence linking olive oil consumption and a lower risk of hypertension was scarce. The SUN Study is a large prospective cohort aimed to unveil the link between lifestyles and chronic diseases. A major objective is to clarify the role of olive oil in health and disease outcomes. In the SUN Study, men with a higher intake of olive oil had a lower risk of developing hypertension during the follow-up. This association was not present among women. Consumption of fruit and vegetables, also associated with a lower risk of hypertension, might interact with olive oil in the diet. Only among individuals with low olive oil consumption, did fruit and vegetable consumption seem to have a protective effect.
CHAPTER | 84 Olive Oil Consumption and Reduced Incidence of Hypertension: The SUN Study
●
Results from other observational and experimental studies offer additional support for a potential beneficial effect of olive oil in the prevention and treatment of hypertension.
REFERENCES Alonso, A., Beunza, J.J., Delgado-Rodríguez, M., Martínez-González, M.A., 2005. Validation of self reported diagnosis of hypertension in a cohort of university graduates in Spain. BMC Public Health 5, 94. Alonso, A., de la Fuente, C., Martín-Arnau, A.M., de Irala, J., Martínez, J.A., Martínez-González, M.A., 2004. Fruit and vegetable consumption is inversely associated with blood pressure in a Mediterranean population with a high vegetable-fat intake: the Seguimiento Universidad de Navarra (SUN) Study. Br. J. Nutr. 92, 311–319. Alonso, A., Martínez-González, M.A., 2004. Olive oil consumption and reduced incidence of hypertension: the SUN Study. Lipids 39, 1233–1238. Alonso, A., Ruiz-Gutiérrez, V., Martínez-González, M.A., 2006. Monounsaturated fatty acids, olive oil and blood pressure: epidemiological, clinical and experimental studies. Public Health Nutr. 9, 251–257. Appel, L.J., Brands, M.W., Daniels, S.R., Karanja, N., Elmer, P.J., Sacks, F.M., 2006. Dietary approaches to prevent and treat hypertension: a scientific statement from the American Heart Association. Hypertension 47, 296–308. Appel, L.J., Moore, T.J., Obarzanek, E., Vollmer, W.M., Svetkey, L.P., Sacks, F.M., Bray, G.A., Vogt, T.M., Cutler, J.A., Windhauser, M.M., Lin, P.H., Karanja, N., 1997. A clinical trial of the effects of dietary patterns on blood pressure. N. Engl. J. Med. 336, 1117–1124. Bes-Rastrollo, M., Pérez Valdivieso, J.R., Sánchez-Villegas, A., Alonso, A., Martínez-González, M.A., 2005. Validación del peso e índice de masa corporal auto-declarados de los participantes de una cohorte de graduados universitarios. Rev. Esp. Obes. 3, 352–358. Beunza, J.J., Martínez-González, M.A., Ebrahim, S., Bes-Rastrollo, M., Núñez-Córdoba, J., Martínez, J.A., Alonso, A., 2007. Sedentary behaviors and the risk of incident hypertension: the SUN cohort. Am. J. Hypertens. 20, 1156–1162. Bondia-Pons, I., Schroder, H., Covas, M.-I., Castellote, A.I., Kaikkonen, J., Poulsen, H.E., Gaddi, A.V., Machowetz, A., Kiesewetter, H., LopezSabater, M.C., 2007. Moderate consumption of olive oil by healthy European men reduces systolic blood pressure in non-Mediterranean participants. J. Nutr. 137, 84–87. Chobanian Jr., A.V., Bakris, G.L., Black, H.R., Cushman, W.C., Green, L.A., Izzo, J.L., Jones Jr., D.W., Materson, B.J., Oparil, S., Wright, J.T., Roccella, E.J.and National High Blood Pressure Education Program Coordinating Committee, 2003. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: The JNC 7 Report. JAMA 289, 2560–2571. Colditz, G.A., Martin, P., Stampfer, M.J., Willett, W.C., Sampson, L., Rosner, B., Hennekens, C.H., Speizer, F.E., 1986. Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women. Am. J. Epidemiol. 123, 894–900.
805
Ferrara, L.A., Raimondi, A.S., d’Episcopo, L., Guida, L., Dello Russo, A., Marotta, T., 2000. Olive oil and reduced need for antihypertensive medications. Arch. Intern. Med. 160, 837–842. Kannel, W.B., 2000. Fifty years of Framingham Study contributions to understanding hypertension. J. Hum. Hypertens. 14, 83–90. Keys, A., Menotti, A., Karvonen, M.J., Aravanis, C., Blackburn, H., Buzina, R., Djordevic, B.S., Dontas, A.S., Fidanza, F., Keys, M.H., Kromhout, D., Nedeljkovic, S., Punsar, S., Seccareccia, F., Toshima, H., 1986. The diet and 15-year death rate in the Seven Countries Study. Am. J. Epidemiol. 124, 903–915. Lopez, A.D., Mathers, C.D., Ezzati, M., Jamison, D.T., Murray, C.J.L., 2006. Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet 367, 1747–1757. Martín-Moreno, J.M., Boyle, P., Gorgojo, L., Maisonneuve, P., FernándezRodríguez, J.C., Salvini, S., Willett, W.C., 1993. Development and validation of a food frequency questionnaire in Spain. Int. J. Epidemiol. 22, 512–519. Martínez-González, M.A., López-Fontana, C., Varo, J.J., SánchezVillegas, A., Martínez, J.A., 2005. Validation of the Spanish version of the physical activity questionnaire used in the Nurses’ Health Study and Health Professionals’ Follow-up Study. Public Health Nutr 8, 920–927. Nuñez-Córdoba, J.M., Alonso, A., Beunza, J.J., Palma, S., GómezGracias, E., Martínez-González, M.A., 2009. Role of vegetables and fruits in Mediterranean diets to prevent hypertension. Eur. J. Clin. Nutr. 63, 605–612. Perona, J.S., Cañizares, J., Montero, E., Sánchez-Dóminguez, J.M., Catalá, A., Ruiz-Gutiérrez, V., 2004. Virgin olive oil reduces blood pressure in hypertensive elderly subjects. Clin. Nutr. 23, 1113–1121. Psaltopoulou, T., Naska, A., Orfanos, P., Trichopoulos, D., Mountokalakis, T., Trichopoulou, A., 2004. Olive oil, the Mediterranean diet, and arterial blood pressure: the Greek European Prospective Investigation into Cancer and Nutrition (EPIC) study. Am. J. Clin. Nutr. 80, 1012–1018. Seguí-Gómez, M., de la Fuente, C., Vazquez, Z., Irala, J.D., MartínezGonzález, M.A., 2006. Cohort profile: The ‘Seguimiento Universidad de Navarra’ (SUN) study. Int. J. Epidemiol. 35, 1417–1422. Tormo, M.J., Navarro, C., Chirlaque, M.D., Barber, X., 2000. Validation of self diagnosis of high blood pressure in a sample of the Spanish EPIC cohort: overall agreement and predictive values. J. Epidemiol. Community Health 54, 221–226. Vargas, C.M., Burt, V.L., Gillum, R.F., Pamuk, E.R., 1997. Validity of self-reported hypertension in the National Health and Nutrition Examination Survey III, 1988–1991. Prev. Med. 26, 678–685. Willett, W.C., Stampfer, M.J., 1998. Implications of total energy intake for epidemiologic analysis. In: Willett, W.C. (Ed.), Nutritional Epidemiology. Oxford University Press, New York, pp. 273–301. Wolf-Maier, K., Cooper, R.S., Banegas, J.R., Giampaoli, S., Hense, H.-W., Joffres, M., Kastarinen, M., Poulter, N.R., Primatesta, P., RodríguezArtalejo, F., Stegmayr, B., Thamm, M., Tuomilehto, J., Vanuzzo, D., Vescio, F., 2003. Hypertension prevalence and blood pressure levels in 6 European countries, Canada, and the United States. JAMA 289, 2363–2369.
This page intentionally left blank
Chapter 85
Virgin Olive Oil and Blood Pressure in Hypertensive Elderly Subjects Javier S. Perona1, Alvaro Alonso2, Miguel A. Martínez-González3 and Valentina Ruiz-Gutiérrez4 1
Nutrition and Lipid Metabolism. Instituto de la Grasa, CSIC, Spain Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, MN, USA 3 Department of Preventive Medicine, School of Medicine, University of Navarra, Spain 4 Nutrition and Lipid Metabolism. Instituto de la Grasa, CSIC, Spain
2
85.1 INTRODUCTION
TABLE 85.1 Key features of hypertension.
Elderly people are an increasing group of the population, which has come to the realization that they can enjoy an active and productive life beyond the retirement age. In elderly people, plasma cholesterol levels decline after age 70, leading to a reduction of the total cholesterol to highdensity-lipoprotein cholesterol (HDL)–cholesterol ratio (Newschaffer et al., 1992) but a concomitant decrease of the absolute risk of coronary heart disease is not observed (Benfante et al., 1992). Both isolated systolic hypertension and combined systolic/diastolic hypertension are considered the most important risk factors for coronary heart disease morbidity and mortality in elderly people (Moser, 1999; Forette et al., 2000). In fact, hypertension affects more than one half of the elderly and its prevalence increases with age (Basile, 2002). Aging of the cardiovascular system is accompanied by endothelial dysfunction, activation of the reninangiotensin system and, consequently, vascular remodeling (Table 85.1). This process leads to an increase in large artery stiffness and an increase in arterial wave reflections to the heart, which is clinically translated to an increase in systolic blood pressure (Duprez, 2008). Additionally, multiple metabolic abnormalities have been associated with hypertension, including changes in lipoprotein levels, hypertriglyceridemia, hypercholesterolemia and insulin resistance (Ruiz-Gutierrez et al., 1996). When hypertension is treated, the cardiovascular morbidity and mortality is reduced to a greater extent than could be expected from the results of trials in middle-aged subjects (Whelton, 1994). Evidence-based medicine recommendations to treat systolic hypertension in the elderly are based on landmark Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
●
●
●
●
●
●
Hypertension, referred to as high blood pressure, is a medical condition in which the blood pressure is chronically elevated Hypertension is considered to be present when a person’s systolic blood pressure is consistently 140 mmHg or greater, and/or their diastolic blood pressure is consistently 90 mmHg or greater With age, the number of collagen fibers in artery and arteriole walls increases, making blood vessels less elastic and so raising mean arterial blood pressure Treatment usually involves lifestyle changes to reduce weight and salt, tobacco and alcohol consumption Antihypertensive drugs are necessary in individuals with a high blood pressure If not treated, hypertension can lead to cardiovascular diseases
and recent clinical trials, which clearly demonstrate that treatment of isolated systolic hypertension is associated with significant decreases in cardiovascular morbidity and mortality. Dietary measures and lifestyle modifications are important components of the treatment plan in elderly patients. Weight reduction and reduced salt intake are particularly important in this patient group, but these objectives are often difficult to achieve (Neutel and Gilderman, 2008). Nutritional supplementation has acquired an important relevance, partially because many antihypertensive agents elevate low-density-lipoprotein cholesterol (LDL) or triacylglycerol (TG) concentrations and/or lower HDL concentrations, paradoxically increasing the cardiovascular
807
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
808
risk (Brook, 2000). However, for the moment, data on the effect of diet on blood pressure in very old subjects (⬎85 years) are scarce and conflicting. Despite fish oil or n-3 fatty acid supplementations being employed as adjuvants in the treatment of hypertension, several studies did not find a reduction in blood pressure of hypertensive patients when compared with virgin olive oil (VOO) supplementation (Prisco et al., 1998).
85.2 VIRGIN OLIVE OIL AND HYPERTENSION The traditional Mediterranean dietary pattern has been proposed as a healthy standard for the prevention of cardiovascular disease, which at least in part can be ascribed to a favorable effect on blood pressure (Alonso et al., 2006). Olive oil is the major source of fat in the Mediterranean diet, which has been firmly associated with improvements in plasma lipid and lipoprotein levels and prevention of cardiovascular disease (Mata et al., 1992; Perez-Jimenez et al., 1995). There is growing evidence suggesting that dietary VOO consumption reduces blood pressure in normotensive (Lahoz et al., 1997) and in hypertensive individuals (Ruiz-Gutierrez et al., 1996). The Greek European Prospective Investigation into Cancer and Nutrition (EPIC) study, including participants that had never received a diagnosis of hypertension, concluded that the Mediterranean diet is inversely associated with arterial blood pressure and that olive oil intake, per se, is inversely associated with both systolic and diastolic blood pressure (Psaltopoulou et al., 2004). Ferrara et al. (2000) reported that VOO reduces the need for medication in hypertensive subjects, which was attributed to enhanced nitric oxide levels by polyphenols. In our laboratory, we demonstrated that in normotensive and in hypertensive normocholesterolemic and hypercholesterolemic subjects dietary VOO lowered blood pressure when compared with another oleic acid-rich oil, such as with high-oleic sunflower oil (HOSO) (Ruiz-Gutierrez et al., 1996). Subsequently, other authors have corroborated our findings (Ferrara et al., 2000). The Seguimiento Universidad de Navarra (SUN, University of Navarra Follow-up) Study, a prospective dynamic cohort study, estimated the association between consumption of olive oil and the incidence of hypertension among the first 7000 participants with 2 years of follow-up in 2004 (Alonso and Martínez-González, 2004). Overall, a higher consumption of olive oil was associated with a lower risk of hypertension among men but not among women. More recently, a large, multicenter, randomized, controlled study has shown that a Mediterranean-like diet enriched in VOO can efficiently reduce blood pressure in subjects with a high risk for cardiovascular disease (Estruch et al., 2006). Despite these evidences, very few studies have focused on the effect of virgin olive oil on blood pressure in hypertensive elderly individuals.
SECTION | II Vascular Aspects Including Hypertension
85.3 VIRGIN OLIVE OIL AND HYPERTENSION IN ELDERLY SUBJECTS In order to study the normotensive effect of VOO in the elderly, and the possible influence of non-oleic acid components of the oil, we employed two different VOO of the same variety, season and geographic localization. The oils presented a similar composition in minor components and very little differences in oleic and linoleic acids, resulting in significant differences in TG molecular species composition. These oils were administered within the diet to 81 normotensive and hypertensive very old subjects (mean age 84 years) at a home for the elderly for 4 weeks each, including a 4-week washout period with sunflower oil before and in between the VOO. The results corroborated previous studies, demonstrating that VOO consumption can reduce systolic blood pressure in hypertensive subjects (Perona et al., 2004) (Figure 85.1). However, it was interesting to note that the concentrations of total and LDL-cholesterol were reduced only after consumption of the VOO with the highest triolein content. The cholesterol reduction was related with the TG molecular species composition of very-low-density lipoproteins (VLDL). In these particles, fatty acids were selectively distributed to form TG after the consumption of either VOO. Since LDL are synthesised in plasma from VLDL remnants, LDL formation and composition are, in part, determined by the TG composition of VLDL and the activity of lipoprotein lipase, the enzyme responsible for their hydrolysis. Therefore, we concluded that the differences in TG molecular species in VOO, and not oleic acid itself, were responsible for the effects on the concentration of plasma cholesterol observed (Perona and RuizGutierrez, 2005). In contrast, as both VOO studied reduced systolic and diastolic blood pressure, the effect could not be ascribed to TG. Previous studies had suggested that non-oleic acid components can also cause the normotensive effect of VOO (Ruiz-Gutierrez et al., 1996). Therefore, these results pointed towards a key role of minor components in this regard (Perona et al., 2004).
85.4 BIOLOGICAL MECHANISMS FOR THE HYPOTENSIVE EFFECTS OF VIRGIN OLIVE OIL Additionally, we deepened in the knowledge of the mechanisms by which VOO can reduce blood pressure in hypertensive elderly individuals. We found differences in the composition and structure, both lipid and protein, of cell membranes after VOO consumption, improving membrane fluidity and the expression of proteins involved in cell signal transduction, which are involved in the regulation of blood pressure (Escriba et al., 2003).
809
CHAPTER | 85 Virgin Olive Oil and Blood Pressure in Hypertensive Elderly Subjects
170
NT
100
HT
NT
b
HT
a
160
a
a
a
150
a a
a
140
130
120
Diastolic pressure (mmHg)
Systolic pressure (mmHg)
80
60
40
20 110
100
0 SO
VOO
SO
VOO
FIGURE 85.1 Systolic and diastolic pressures of normotensive (NT) and hypertensive (HT) elderly subjects after consuming sunflower (SO) or virgin olive oil diets (VOO). Subjects received diets with VOO or SO as dietary fat for 4 weeks. Blood pressure was measured before and after consumption of the diets. Mean values within a row sharing the same letter are not significantly different (p ⬎ 0.01). Adapted from Perona et al. (2004).
In elderly hypertensive subjects, long-term VOO consumption reduced the cholesterol/phospholipid ratio in erythrocyte membranes, normalizing the values to those of normotensives. Such reductions in the cholesterol concentrations in membranes after olive oil consumption have been associated with increased membrane fluidity (North and Fleischer, 1983). Likewise, a reduction in membrane fluidity (high cholesterol) has been associated with the development of hypertension (Tsuda et al., 2000) and with an age-related impaired β-adrenergic-mediated vasorelaxation and Gαs coupling (Noble et al., 1999). VOO consumption also induced significant changes of specific fatty acid moiety concentrations in phospholipids and cholesterol esters. In both lipid classes, monounsaturated fatty acids (MUFA) increased significantly in elderly humans after long-term VOO consumption, mainly due to a rise in the proportion of oleic acid (18:1, n-9). This fact was reflected in the significant increases of the MUFA:saturated fatty acid (SFA) (from 0.57 to 0.64 in the normotensive group and from 0.57 to 0.65 in the hypertensive subjects) and MUFA:polyunsaturated fatty acids (PUFA) ratios (from 0.62 to 0.71 in normotensive subjects and from 0.57 to 0.73 in hypertensive subjects) in membrane phospholipids. In contrast, the PUFA:SFA ratio did not change markedly. Since phospholipids contribute approximately 75% of the total fatty acid content in membranes, these changes are relevant to the membrane structure. In line with these results, some membrane lipids regulate membrane structure, influencing the localization and activity of several membraneassociated proteins (e.g., G proteins and PKC) (Martínez et al., 2005). Therefore, the normotensive effects of VOO could be originated by modulation of the interaction
of G proteins and PKC, in addition or alternatively to membrane fluidity. In this sense, long-term VOO consumption significantly reduced the membrane levels of Gαi1/2, Gαs, Gβ and PKCα in elderly hypertensive subjects. These effects were accompanied by a reduction in blood pressure (mean systolic and diastolic blood pressure values were 162.4 and 81.0 mmHg, respectively before and 138.0 and 72.3 mmHg after VOO consumption, p ⬍ 0.05). The reduction of blood pressure by the Mediterranean diet and VOO has been also linked to improvements in the endothelial function. Esposito et al. (2004) carried out a randomized trial among 180 subjects with metabolic syndrome, who were instructed to follow a Mediterraneanstyle diet, including VOO. After a two-year follow-up, they observed improved endothelial function as a measure of blood pressure and platelet aggregation response to L-arginine, the natural precursor of nitric oxide. They also reported a significant reduction of markers of systemic vascular inflammation, such as C-reactive protein and interleukins 6 (IL-6), 7 (IL-7) and 18 (IL-18).
85.5 VIRGIN OLIVE OIL MINOR COMPONENTS AND HYPERTENSION Besides its high MUFA content, VOO has important amounts of antioxidants and other phytochemicals. However, when refined or heated to frying temperature in air, VOO loses most of these non-lipidic natural compounds (Ros, 2003). The normotensive effects of VOO have been attributed to some of these minor components, which account for approximately 1–2% of the whole oil and include very
810
(A) 100
%Effect
***
Recent investigations are reinforcing the current knowledge on the healthy properties of VOO and that it can help
Control
0 −7
−6 −5 Log [Olean olic acid]
−4
(B) 100 *** 50 Control L-Name 0 −7
−6 −5 Log [Erythrodio]
−4
FIGURE 85.2 Relaxant effect of oleanolic acid (A) and erythrodiol (B) in phenylephrine (10-6 M) precontracted rat aortic rings. Addition of the NO-synthase inhibitor L-NAME (3 ⫻ 10⫺4 M) produced a significant reduction of the endothelium-dependent relaxation to both triterpenic compounds (circles) compared to control (squares). ***: p ⬍ 0.001, vs. Control. Adapted from Rodriguez-Rodriguez et al. (2004).
to prevent and treat hypertension in the general population, thus protecting against cardiovascular disease. It is also interesting to note that these effects can also be applicable to elderly subjects, a population group that is growing in Western societies. It is also becoming clear that the content of oleic acid alone cannot fully explain the impact on health of olive oil and that VOO being a unique fruitderived oil, is rich in a number of minor compounds with relevant physiological and pharmacological functions. Some of these compounds may improve the endothelial function by reducing levels of ROS, but their roles in the endothelium and the putative consequences in the normalization of blood pressure need to be further studied.
SUMMARY POINTS ●
85.6 CONCLUSIONS
50 L-Name
%Effect
highly bioactive compounds like sterols, tocopherols, phenolics and terpenoids. Reactive oxygen species (ROS) cause endothelial dysfunction, a process that has been implicated in the pathophysiology of hypertension (Cai and Harrison, 2000). Phenols present in VOO, such as oleuropein, hydroxytyrosol, tyrosol and caffeic acid are strong antioxidants and radical scavengers (Tuck and Hayball, 2002), which can help to revert the imbalance between increased oxidative stress and impaired antioxidant defense that affects endothelial function (Visioli and Galli, 1998). Besides the antioxidant properties of the phenolic compounds from extra VOO, antiinflammatory effects have been demonstrated in several cell types. These compounds inhibit leukotriene B4 production at the 5-lipoxygenase (5-LOX) level and reduce the generation of ROS in rat leukocytes (de la Puerta et al., 1999). α-Tocopherol is another antioxidant compound present in VOO that has been shown to improve endothelial function, by modulating eicosanoid metabolism in endothelial cells. Actually, α-tocopherol can restore the reduced PGI2 synthesis in endothelial cells (Kunisaki et al., 1992) and there are data indicating that it inhibits 5-LOX (Jialal et al., 2001) and cyclooxygenase-2 (COX-2) (Wu et al., 2001). The potential therapeutic importance of olive oil triterpenoids, encompassed of acids and alcohols, has not been extensively studied. Although their concentration in VOO is quite low, they are present in relevant concentrations in pomace olive oil, which is obtained from the residues of VOO extraction. Oleanolic acid has been identified in a multitude of medicinal plants (Liu, 1995) and has been reported to possess a number of biological pharmacological activities, including some affecting inflammation (Martínez-González et al., 2008). Oleanolic acid inhibits LOX and COX-2 activities (Simon et al., 1992; Ringbom et al., 1998), therefore reducing the production of PGE2 and LTB4. We recently developed a study to evaluate the properties of oleanolic acid as a vasodilator agent and to determine its mechanism of action (Rodriguez-Rodriguez et al., 2004). The vasorelaxant effect induced by oleanolic acid and erythrodiol was studied in isolated thoracic rat aorta. Results from this work introduced the first in vitro evidence that these triterpenoids evoke an endothelium-dependent vasorelaxation in rat aorta, and suggested that the mechanism of relaxation is mainly mediated by the endothelial production of NO (Figure 85.2). According to the pharmacological effects obtained, it was concluded that oleanolic acid and erythrodiol may have interesting therapeutic potential as new vasodilators present in pomace olive oil and VOO.
SECTION | II Vascular Aspects Including Hypertension
●
Hypertension affects more than one half of the elderly and its prevalence increases with age. Treatment of isolated systolic hypertension is associated with significant decreases in cardiovascular morbidity and mortality, also among elderly individuals.
CHAPTER | 85 Virgin Olive Oil and Blood Pressure in Hypertensive Elderly Subjects
●
●
●
●
●
●
The traditional Mediterranean diet, including virgin olive oil (VOO) as its main dietary fat, has been proposed as a healthy pattern for the prevention of cardiovascular disease. Growing evidence suggests that dietary VOO consumption reduces blood pressure in hypertensive individuals. Non-oleic acid components of VOO may be responsible for its normotensive effect. VOO consumption normalizes membrane parameters that are impaired in hypertensives, improving membrane fluidity and the expression of proteins involved in cell signal transduction, which are involved in the regulation of blood pressure. Minor components from VOO may improve the endothelial function by reducing levels of ROS. The roles of minor components on the endothelium and the putative consequences in the normalization of blood pressure need to be further studied.
REFERENCES Alonso, A., Martínez-González, M.A., 2004. Olive oil consumption and reduced incidence of hypertension: the SUN study. Lipids 39, 1233–1238. Alonso, A., Ruiz-Gutierrez, V., Martínez-González, M.A., 2006. Monounsaturated fatty acids, olive oil and blood pressure: epidemiological, clinical and experimental evidence. Public Health Nutr. 9, 251–257. Basile, J., 2002. Hypertension in the elderly: a review of the importance of systolic blood pressure elevation. Clin. Hypertens. (Greenwich) 4, 108–112. Benfante, R., Reed, D., Frank, J., 1992. Do coronary heart disease risk factors measured in the elderly have the same predictive roles as in the middle aged. Comparisons of relative and attributable risks. Ann. Epidemiol. 2, 273–282. Brook, RD., 2000. Mechanism of differential effects of antihypertensive agents on serum lipids. Curr. Hypertens. Rep. 2, 370–377. Cai, H., Harrison, D.G., 2000. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ. Res. 87, 840–844. de la Puerta, R., Ruiz-Gutierrez, V., Hoult, J.R.S., 1999. Inhibition of leukocyte 5-lipoxygenase by phenolics from virgin olive oil. Biochem. Pharmacol. 57, 445–449. Duprez, D.A., 2008. Systolic hypertension in the elderly: addressing an unmet need. Am. J. Med. 121, 179–184. Escribá, P.V., Sánchez-Dominguez, J.M., Alemany, R., Perona, J.S., RuizGutiérrez, V., 2003. Alteration of lipids, G proteins, and PKC in cell membranes of elderly hypertensives. Hypertension 41, 176–182. Esposito, K., Marfella, R., Ciotola, M., Di Palo, C., Giugliano, F., Giugliano, G., D’Armiento, M., D’Andrea, F., Giugliano, D., 2004. Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA 292, 1440–1446. Estruch, R., Martínez-González, M.A., Corella, D., Salas-Salvadó, J., RuizGutiérrez, V., Covas, M.I., Fiol, M., Gómez-Gracia, E., López-Sabater, M.C., Vinyoles, E., Arós, F., Conde, M., Lahoz, C., Lapetra, J., Sáez, G., Ros, E. PREDIMED Study Investigators, 2006. Effects of a Mediterranean-style diet on cardiovascular risk factors: a randomized trial. Ann. Intern. Med. 145, 1–11.
811
Ferrara, L.A., Raimondi, A.S., d’Episcopo, L., Guida, L., Dello Russo, A., Marotta, T., 2000. Olive oil and reduced need for antihypertensive medications. Arch. Intern. Med. 160, 837–842. Forette, F., Lechowski, L., Rigaud, A.S., Seux, M.L., Dessi, F., Forette, B., 2000. Does the benefit of antihypertensive treatment outweigh the risk in very elderly hypertensive patients? J. Hypertens. Suppl. 18, S9–S12. Jialal, I., Devaraj, S., Kaul, N., 2001. The effect of alpha-tocopherol on monocyte proatherogenic activity. J. Nutr. 131, 389S–394S. Kunisaki, M., Umeda, F., Inoguchi, T., Nawata, H., 1992. Vitamin E binds to specific binding sites and enhances prostacyclin production by cultured aortic endothelial cells. Thromb. Haemost. 68, 744–751. Lahoz, C., Alonso, R., Ordovas, J.M., Lopez-Farre, A., de Oya, M., Mata, P., 1997. Effects of dietary fat saturation on eicosanoid production, platelet aggregation and blood pressure. Eur. J. Clin. Invest. 27, 780–787. Liu, J., 1995. Pharmacology of oleanolic acid and ursolic acid. J. Ethnopharmacol. 49, 57–68. Martínez, J., Vögler, O., Casas, J., Barceló, F., Alemany, R., Prades, J., Nagy, T., Baamonde, C., Kasprzyk, P.G., Terés, S., Saus, C., Escribá, P.V., 2005. Membrane structure modulation, protein kinase C alpha activation, and anticancer activity of minerval. Mol. Pharmacol. 67, 531–540. Martínez-González, J., Rodríguez-Rodríguez, R., González-Díez, M., Rodríguez, C., Herrera, M.D., Ruiz-Gutierrez, V., Badimon, L., 2008. Oleanolic acid induces prostacyclin release in human vascular smooth muscle cells through a cyclooxygenase-2-dependent mechanism. J. Nutr. 138, 443–448. Mata, P., Alvarez-Sala, L.A., Rubio, M.J., Nuño, J., De Oya, M., 1992. Effects of long-term monounsaturated- vs polyunsaturated-enriched diets on lipoproteins in healthy men and women. Am. J. Clin. Nutr. 55, 846–850. Moser, M., 1999. Hypertension treatment and the prevention of coronary heart disease in the elderly. Am. Fam. Physician 59, 1248–1256. Neutel, J.M., Gilderman, L.I., 2008. Hypertension control in the elderly. J. Clin. Hypertens. (Greenwich) 10, 33–39. Newschaffer, C.J., Bush, T.L., Hale, W.E., 1992. Aging and total cholesterol levels: cohort, period, and survivorship effects. Am. J. Epidemiol. 136, 23–34. Noble, J.M., Thomas, T.H., Ford, G.A., 1999. Effect of age on plasma membrane asymmetry and membrane fluidity in human leukocytes and platelets. J. Gerontol. A Biol. Sci. Med. Sci. 54, M601–M606. North, P., Fleischer, S., 1983. Alteration of synaptic membrane cholesterol/ phospholipid ratio using a lipid transfer protein. Effect on gammaaminobutyric acid uptake. J. Biol. Chem. 258, 1242–1253. Perez-Jimenez, F., Espino, A., Lopez-Segura, F., Blanco, J., RuizGutierrez, V., Prada, J.L., Lopez-Miranda, J., Jimenez-Pereperez, J., Ordovas, J.M., 1995. Lipoprotein concentrations in normolipidemic males consuming oleic acid-rich diets from two different sources: olive oil and oleic acid-rich sunflower oil. Am. J. Clin. Nutr. 62, 769–775. Perona, J.S., Cañizares, J., Montero, E., Sánchez-Domínguez, J.M., Catalá, A., Ruiz-Gutiérrez, V., 2004. Virgin olive oil reduces blood pressure in hypertensive elderly subjects. Clin. Nutr. 23, 1113–1121. Perona, J.S., Ruiz-Gutierrez, V., 2005. Triacylglycerol molecular species are depleted to different extents in the myocardium of spontaneously hypertensive rats fed two oleic acid-rich oils. Am. J. Hypertens. 18, 72–80. Prisco, D., Paniccia, R., Bandinelli, B., Filippini, M., Francalanci, I., Giusti, B., Giurlani, L., Gensini, G.F., Abbate, R., Neri Serneri, G.G., 1998. Effect of medium-term supplementation with a moderate dose of (n-3) polyunsaturated fatty acids on blood pressure in mild hypertensive patients. Thromb. Res. 91, 105–112.
812
Psaltopoulou, T., Naska, A., Orfanos, P., Trichopoulos, D., Mountokalakis, T., Trichopoulou, A., 2004. Olive oil, the Mediterranean diet, and arterial blood pressure: the Greek European Prospective Investigation into Cancer and Nutrition (EPIC) study. Am. J. Clin. Nutr. 80, 1012–1018. Ringbom, T., Segura, L., Noreen, Y., Perera, P., Bohlin, L., 1998. Ursolic acid from Plantago major, a selective inhibitor of cyclooxygenase-2 catalyzed prostaglandin biosynthesis. J. Nat. Prod. 61, 1212–1215. Rodriguez-Rodriguez, R., Herrera, M.D., Perona, J.S., Ruiz-Gutierrez, V., 2004. Potential vasorelaxant effects of oleanolic acid and erythrodiol, two triterpenoids contained in ‘orujo’ olive oil, on rat aorta. Br. J. Nutr. 92, 635–642. Ros, E., 2003. Dietary cis-monounsaturated fatty acids and metabolic control in type 2 diabetes. Am. J. Clin. Nutr. 78, 617S–625S. Ruiz-Gutierrez, V., Muriana, F.J., Guerrero, A., Cert, A.M., Villar, J., 1996. Plasma lipids, erythrocyte membrane lipids and blood pressure of hypertensive women after ingestion of dietary oleic acid from two different sources. J. Hypertens. 14, 1483–1490.
SECTION | II Vascular Aspects Including Hypertension
Simon, A., Najid, A., Chulia, A.J., Delage, C., Rigaud, M., 1992. Inhibition of lipoxygenase activity and HL60 leukemic cell proliferation by ursolic acid isolated from heather flowers (Calluna vulgaris). Biochim. Biophys. Acta 1125, 68–72. Tsuda, K., Kinoshita, Y., Nishio, I., Masuyama, Y., 2000. Role of insulin in the regulation of membrane fluidity of erythrocytes in essential hypertension: an electron paramagnetic resonance investigation. Am. J. Hypertens. 13, 376–382. Tuck, K.L., Hayball, P.J., 2002. Major phenolic compounds in olive oil: metabolism and health effects. J. Nutr. Biochem. 13, 636–644. Visioli, F., Galli, C., 1998. The effect of minor constituents of olive oil on cardiovascular disease: new findings. Nutr. Rev. 56, 142–147. Whelton, P.K., 1994. Epidemiology of hypertension. Lancet 344, 101–106. Wu, D., Hayek, M.G., Meydani, S.N., 2001. Vitamin E and macrophage cyclooxygenase regulation in the aged. J. Nutr. 131, 382S–388S.
Chapter 86
Vasorelaxant Effects of Oleanolic Acid and Erythrodiol in Pomace Olive Oil Rosalia Rodriguez-Rodriguez and Valentina Ruiz-Gutiérrez Instituto de la Grasa (CSIC), Seville, Spain
86.1 EVIDENCE OF MEDITERRANEANSTYLE DIET EFFECT ON CARDIOVASCULAR DISEASE Over the years, a number of epidemiological and clinical studies developed in different countries constitute a firm and reliable experimental base supporting the beneficial effect of the Mediterranean diet pattern, containing olive oil as the main source of fat, with regards to the reduction of cardiovascular diseases. Much interest has been directed towards the constituents from olive oil and their molecular mechanisms contributing to its beneficial effects on health. The healthy properties associated with olive oil consumption on cardiovascular risk factors and endothelial function have been partly attributed to its high content of monounsaturated fatty acids such as oleic acid. Nevertheless, an increasing number of studies propose that the content of oleic acid alone cannot fully explain the healthful properties of olive oil. In this regard, minor components from olive oil such as tocopherols and triterpenoids have been related to antiinflammatory and antioxidant activities. Therefore, despite the fact that they are in low proportion, they confer important biological activities to olive oil and its derivatives.
86.2 TRITERPENIC FRACTION IN OLIVE OIL AND POMACE OLIVE OIL Amongst minor components from olive oil, interest on its triterpenic fraction has increased in recent years. The triterpenic fraction in olive oil and olive skin is mainly constituted by pentacyclic terpenic acids (oleanolic and maslinic acid) and diols (erythrodiol and uvaol). The concentration range of these triterpenoids reaches values up to 400 mg kg⫺1 in the skin of the olive fruit whereas concentrations in olive oil are 25–50 mg kg⫺1 for triterpenic acids and 6–18 mg kg⫺1 for erythrodiol and uvaol, depending on acidity and olive variety (Pérez-Camino and Cert, 1999). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
The procedure applied for extraction of olive oil is also important for the content of minor constituents. In this regard, pomace olive oil (called ‘orujo’ olive oil in Spain) is obtained from the residue that remains after virgin olive oil mechanical extraction. As a consequence of the extraction processes needed to obtain both refined and pomace olive oils, the hydrosoluble fraction, including polyphenols, is lost. In spite of the lack of polyphenols, pomace olive oil, extracted using a new procedure (patent number 200400755) contains higher concentrations of triterpenic acids and alcohols than virgin olive oil exceeding values of 120 mg kg⫺1. Among olive and pomace olive oil triterpenoids, oleanolic acid has been the most widely studied and distributed in the vegetable kingdom. In fact, a wide range of biological activities has been attributed to this triterpenoid (Herrera et al., 2006). Nevertheless, there are not many studies focused on the cardiovascular properties ascribed to oleanolic acid and structure-related triterpenoids such as erythrodiol. Accordingly, parenteral administration of this triterpenic acid lowers blood pressure in salt-sensitive rats, an effect mainly associated with antioxidative actions (Somova et al., 2003). In addition, it has been recently reported that ingestion of pomace olive oil with a high proportion of oleanolic acid and erythrodiol offers a delay in the progression of lipid peroxidation in rat liver microsomes (Perona et al., 2005). Despite these recent findings, the effects of oleanolic acid and pomace olive oil on vascular and endothelial function have remained unknown until now. In this chapter, we will summarize the recent and novel advances in the vasoprotective profile provided by olive and pomace olive oil and their triterpenoids in terms of vasodilatation.
86.3 IMPORTANCE OF THE ENDOTHELIUM ON VASCULAR FUNCTION Due to its strategic position between the circulating blood and the smooth muscle cells (SMC), vascular endothelium
813
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
814
SECTION | II Vascular Aspects Including Hypertension
TABLE 86.1 Key features of the vasoactive effects of oleanolic acid. ●
Oleanolic acid is the major triterpenoid found in olive and pomace olive oil. It is widely distributed in plants and numerous pharmacological activities have been attributed to this triterpenic acid
●
Cardiovascular properties of oleanolic acid include vasodilatation, antihypertensive actions and antioxidant effects
●
In terms of vascular protection, oleanolic acid induces the release of endothelial-derived vasorelaxing and atheroprotective substances, mainly nitric oxide and prostacyclin
●
The mechanisms underlying this vasoprotective effect involve activation of endothelial nitric oxide synthase by phosphorylation and up-regulation of vascular cyclooxygenase through mitogen-activated protein kinase-dependent pathways
●
Chronic intake of pomace olive oil rich in oleanolic acid restores endothelial function
plays a key role in a number of homeostatic functions including regulation of vascular tone, antithrombotic and antiadhesive properties, permeability and integrity of the vessel wall. In response to various physical or chemical stimuli, endothelial cells exert these regulatory functions on vascular wall through the synthesis and release of vasoactive factors promoting vasodilatation, such as nitric oxide (NO), prostacyclin (PGI2), and endothelial-derived hyperpolarizing factor (EDHF), or vasoconstriction, such as thromboxane A2 (TXA2), isoprostanes, superoxide anion and endothelin-1 (Table 86.1). As a major regulator of local vascular homeostasis, the endothelium maintains the balance between the production of relaxing and contracting factors. Upsetting this tightly regulated balance leads to endothelial dysfunction. In fact, the dysfunction of the endothelial layer is considered one of the earliest events in the development of atherosclerosis. Moreover, cardiovascular risk factors such as hypertension and hypercholesterolemia are frequently associated with abnormalities in vascular function, characterized by an increased response to specific vasoconstrictor agents and a pronounced attenuation of endothelium-dependent vasorelaxation (Mombouli and Vanhoutte, 1999).
86.3.1 Nitric Oxide NO is synthesized from L-arginine by the endothelial NO synthase (eNOS), which requires Ca2⫹/calmodulin, FAD, FMN and BH4 as cofactors. The activation of eNOS occurs upon increases in cytosolic Ca2⫹ in response to various stimuli such as shear stress, hormones and several drugs. NO diffuses to the adjacent SMC where it stimulates soluble guanylyl cyclase and induces arterial vasodilatation. Therefore, NO plays a crucial role in controlling blood pressure and tissue flow. In addition to its effects on vascular tone, NO also exerts positive effects in platelet aggregation, endothelial regeneration, SMC proliferation
and adhesion molecules. Decreased bioavailability of NO has been identified as a major cause for endothelial dysfunction. The mechanisms involved in the endothelialderived NO responses have been extensively studied and affect NO breakdown as well as NO synthesis. The former is mainly related to changes in the vascular levels of superoxide anion inactivating NO. Reduced NO synthesis associated with endothelial dysfunction may be due to decreased expression of eNOS, post-transcriptional modifications of the enzyme (e.g., by phosphorylation), protein interactions, lack of the substrate or cofactors, or the presence of endogenous NOS inhibitors (Fleming and Busse, 2003).
86.3.2 Prostanoids Derived from Cyclooxygenase Under normal circumstances, PGI2 is the major metabolite derived from cyclooxygenase (COX) in endothelial cells. PGI2 evokes a potent vasodilator activity by acting on specific receptors on vascular SMC that are coupled to adenylyl cyclase. Other metabolites derived from COX, such as TXA2, act on endoperoxide/thromboxane receptors in vascular smooth muscle to induce vasoconstriction. However, in physiological situations, the influence of the small amounts of vasoconstrictor prostanoids released by endothelial cells is masked by the production of the main endothelial-derived relaxing substances NO, PGI2 and EDHF. The use of COX inhibitors like indomethacin revealed an increased participation of COX-derived vasoconstrictor products in conductance and resistance arteries from animal models of cardiovascular disease and humans (Taddei et al., 1997). Changes in the traditional COX isoforms expression (COX-1 and COX-2, constitutive and inducible isoforms, respectively) have been suggested to cause the release of vasoconstrictor products, but it is not clear which isoform is responsible for this event.
CHAPTER | 86 Vasorelaxant Effects of Oleanolic Acid and Erythrodiol in Pomace Olive Oil
86.3.3 EDHF EDHF is another important endothelium-derived relaxing factor, especially in medium- to resistance-sized arteries. The identity of EDHF is still controversial and varies depending on the arterial type. EDHF has been proposed to be epoxyeicosatrienoic acid (EET), K⫹, anandamide and H2O2 (Busse et al., 2002). Responses mediated by EDHF are resistant to the combined inhibition of both NO and COX but are sensitive to the combination of the smalland intermediate-conductance Ca2⫹-activated K⫹ channel inhibitors apamin and charibdotoxin (Busse et al., 2002). The participation of EDHF is not uniform among different vascular beds and species.
86.3.4 Olive Oil Triterpenoids and Endothelial Function One of the main targets against a dysfunctional endothelium consists of improving endothelium-dependent vasodilatation by increasing protecting factors released from the endothelium such as NO and PGI2 or decreasing synthesis/ release of pathogenic and contracting factors such as free radicals. In recent years, there has been an emerging interest in nutritional supplements or the so-called nutraceuticals focused on the therapy of the main cardiovascular risk factors. Although the association between a diet rich in olive oil and the incidence of cardiovascular pathologies is strongly supported by numerous investigations, only a few studies have addressed the direct effects of long-term consumption of olive oil on endothelial function. The Mediterranean diet has been shown to attenuate endothelial dysfunction in diabetic and hypercholesterolemic patients, and in subjects with metabolic syndrome (Esposito et al., 2004). Nevertheless, the exact mechanisms underlying the effects of olive oil in endothelial function and the bioactive compounds from olive oil responsible for these actions are poorly understood. In recent years, particular interest has been directed towards minor components from olive oil, especially phenols and triterpenoids. The important antioxidant and anti-inflammatory activities showed by pentacyclic triterpenoids, particularly oleanolic acid, support their potential as cardioprotective molecules and probably their participation in the beneficial effect provided by olive oil consumption on cardiovascular risk factors. Mostly, our attention has been focused on the vasoactive actions of triterpenoids from olive oil.
86.4 VASODILATOR EFFECTS OF OLIVE OIL TRITERPENOIDS IN ISOLATED RAT AORTA The vasodilator effects evoked by olive oil triterpenoids (oleanolic acid, erythrodiol, uvaol and maslinic acid) were evidenced for the first time in conductance arteries from
815
normotensive and spontaneously hypertensive rats (SHR) (Rodriguez-Rodriguez et al., 2004, 2006) (Figure 86.1). These four triterpenoids were able to elicit concentrationdependent relaxation in precontracted aortic rings from Wistar and SHR, albeit to a different extent. These responses, except that evoked by uvaol, were significantly attenuated after endothelium removal. Hypertension is associated with functional changes in the arteries involving alterations in the responsiveness of vascular smooth muscle and disturbances in the activity of the endothelium (Shimokawa, 1998). Under these conditions, the endothelium-dependent vasodilatation to acetylcholine is impaired in different models of hypertension, such as SHR, when compared to normotensive rats due to an imbalance in the activity of relaxing and contracting factors derived from the endothelium. We found that the endothelial dependence and the vasodilatation induced by olive oil triterpenoids in the aorta from normotensive rats were maintained in arterial preparations from SHR and no significant differences in the rate of relaxation between both strains were appreciated (Rodriguez-Rodriguez et al., 2004, 2006). While the tested compounds were structurally similar, we could observe differences in the potency of relaxation and in the detailed mechanisms through which they induce vasodilatation. Accordingly, in aortic rings with functional endothelium the order of maximal dilatation could be summarized as oleanolic acid ⫽ maslinic acid ⬎ erythrodiol ⫽ uvaol. The endothelium-dependent relaxation to oleanolic and maslinic acids, and erythrodiol clearly involves the release of NO from the vascular endothelium because vasodilatation was considerably attenuated in the presence of the NOS inhibitor, L-NAME, reaching the same level of attenuation as that observed in aortic rings lacking endothelium. In contrast, the vasorelaxation caused by uvaol was NO- and endotheliumindependent (Rodriguez-Rodriguez et al., 2004, 2006). The exhibited differences regarding endothelial dependence, NO involvement and potency in the vasodilatation activity evoked by olive oil triterpenoids led us to consider the key role of functional groups (acid or alcohol) and the existence of an additional hydroxyl or methyl group in the molecule. These structural differences could explain the disparity in the antioxidant and anti-inflammatory responses evidenced between oleanolic acid, maslinic acid, erythrodiol and uvaol (Marquez-Martin et al., 2006). The use of COX and endoperoxide/thromboxane receptor inhibitors discarded the involvement of prostanoids derived from COX in the vasodilatation promoted by most of the tested triterpenoids in rat aorta. Nevertheless, erythrodiol induced a persisting relaxation in the presence of the simultaneous blockage of the NO synthesis and the COX pathway, as well as the higher relaxation observed in the presence of the TP receptor antagonist, suggesting a potential participation of COX-derived factors in the vasoactive response to triterpenic diol. Other mediators involved in the vascular homeostasis such as superoxide anion seem not to
816
SECTION | II Vascular Aspects Including Hypertension
B 0
0
25
25
** ***
50
% Relaxation
% Relaxation
A
75
* 75
100
100 −7
−6 −5 −4 Log [Oleanolic acid]
−7
−6
−5
−4
Log [Maslinic acid]
C
D 0
0
25
25
*** ***
50 75
% Relaxation
% Relaxation
*
50
50 75 100
100 −7
−6 −5 −4 Log [Erythrodiol] Control E+
−7
L-NAME
−6 −5 Log [Uvaol]
−4
Control E−
FIGURE 86.1 Vasodilator effects of olive oil triterpenoids in isolated rat aorta. Endothelium- and nitric oxide (NO)-dependent vasorelaxation induced by olive oil triterpenoids in precontracted aortic segments from spontaneously hypertensive rats. In most cases (A–C), incubation with the NO synthesis inhibitor N-nitro-L-arginine methyl ester (L-NAME) significantly attenuated concentration–response relaxation curves to triterpenoids compared with control arterial preparations (Control E⫹). Endothelium removal (Control E-) also abolished vasorelaxation. ***p ⬍ 0.001 vs control E⫹. Adapted from Rodriguez-Rodriguez et al. (2006).
be involved in the vasodilatation to the tested triterpenoids (Rodriguez-Rodriguez et al., 2004, 2006).
86.5 VASODILATOR EFFECTS OF OLIVE OIL TRITERPENOIDS IN ISOLATED RAT MESENTERIC ARTERIES Previous reports found variations in vasorelaxant responses to different chemical or physical stimuli between vascular beds, or indeed between vessels within the same bed (O’Sullivan et al., 2004). Accordingly, the relative contribution of EDHF varies among species, vascular bed, and vessel size. Particularly in arterioles and small arteries such as resistance mesenteric arteries, EDHF appears to be of major importance, whereas in larger arteries such as superior mesenteric artery and aorta, the role of NO is more pronounced (Shimokawa et al., 1996). The vasorelaxant effect of olive oil triterpenoids has been evaluated in two differentsized arteries: rat superior and low-resistance mesenteric
arteries. According to previous results in conductance arteries, oleanolic acid evoked endothelium-dependent vasodilatation in superior and mesenteric small rat arteries (Rodriguez-Rodriguez et al., 2008) (Figure 86.2). This response was strongly attenuated by endothelial removal and after NOS inhibition whereas blocking of COX did not alter the vasodilatation. Moreover, the use of calciumactivated K⫹ channel inhibitors in combination with NOS and COX inhibitors indirectly suggest that rather than EDHF, NO mediates oleanolic acid relaxation in rat superior and small mesenteric arteries. The importance of NO mediating vasodilatation induced by oleanolic acid in different-sized arteries was further sustained by simultaneous measurements of NO and relaxation to oleanolic acid by using a selective NO microelectrode inserted into the lumen of the superior mesenteric arterial segment. Interestingly, endothelial cell calcium determinations revealed that oleanolic acid induces endothelial NO release involving mechanisms independent of both calcium influx and release of calcium from endoplasmic reticulum.
817
CHAPTER | 86 Vasorelaxant Effects of Oleanolic Acid and Erythrodiol in Pomace Olive Oil
0.01
A
0.1
1
Oleanolic acid
Ca2+
0.5 g
Acetylcholine
10 min
M
10
PLC
100
Gq
IP3 PI3K
NA Ca e.r.
B
0.01
0.1
2+
P Ser473
AMPK
Akt
1177 P Ser
1 NOS
10
100 L-arginine
NA FIGURE 86.2 Vasodilator effects of olive oil triterpenoids in isolated rat mesenteric arteries. Original trace recordings showing vasodilatation evoked by cumulative addition of oleanolic acid (0.01–100 μM) in rat superior (A) or small resistance mesenteric artery (B) with intact vascular endothelium contracted by norepinephrine (NA, 0.5 μM). Adapted from Rodriguez-Rodriguez et al. (2008).
86.6 MOLECULAR BASIS OF THE VASOPROTECTIVE EFFECTS OF OLIVE OIL TRITERPENOIDS 86.6.1 Effect of Oleanolic Acid on NO Release and eNOS Activation Regulation of eNOS activity is pursued as a strategy for the prevention of cardiovascular diseases. Regulatory mechanisms of eNOS include phosphorylation at key sites in response to several stimuli such as eNOS-Ser1177 (Fleming and Busse, 2003). A number of kinases phosphorylate eNOS-Ser1177 increasing eNOS activity and NO production, even via Ca2⫹-independent mechanisms (Fleming and Busse, 2003). Our previous findings of increase in NO concentration without associated increase in intracellular calcium to oleanolic acid, led us to investigate whether this triterpenoid phosphorylates eNOS at this site. We found for the first time that oleanolic acid elicited a rapid phosphorylation of eNOS-Ser1177 in arterial segments and cultured endothelial cells (Rodriguez-Rodriguez et al., 2008). We also reported that oleanolic acid phosphorylates Akt-Ser473 in endothelial cells leading to the downstream eNOS-Ser1177 stimulation (Figure 86.3). Other experimental studies have provided evidence on the effects of minor compounds present in olive oil and pomace oil in the regulation of eNOS and NO release. The vasoprotective effects of alpha-tocopherol in vivo have been related to its NO-dependent relaxation mediated by eNOS phosphorylation at Ser1177 in endothelial cell culture (Heller et al., 2004). Although the antiatherogenic activity of the olive oil
NO
FIGURE 86.3 Effect of oleanolic acid on nitric oxide (NO) release and endothelial NO synthase (eNOS) activation. Model for oleanolic acidevoked release of NO in vascular endothelial cells. Conventional agonists such as acetylcholine increase intracellular calcium through activation of G-protein coupled receptors, while oleanolic acid leads to phosphorylation of eNOS at Ser1177 residues through mechanisms sensitive to inhibition of phosphoinositide 3-kinase (PI3K)/Akt and AMP-activated protein kinase (AMPK). e.r.: endoplasmic reticulum; IP3: inositol triphosphate; PLC: phospholipase C; M: muscarinic receptor.
polyphenol hydroxytyrosol has been firmly evidenced, this phenol is unable to evoke any effect on eNOS expression and NO release in non-stimulated human endothelial cells (Schmitt et al., 2007). Accordingly, the authors suggest that hydroxytyrosol may exert beneficial effects on the endothelium only in pathological conditions, but not necessarily in a healthy environment.
86.6.2 Effect of Oleanolic Acid on PGI2 Release and COX-2 Regulation Although COX-2 has traditionally been associated with proinflammatory and proatherogenic states, this enzyme may contribute to vascular PGI2 formation in healthy humans and data from animal models also establish that PGI2 released from COX-2 prevents local thrombosis and neointima formation (Martínez-González and Badimon, 2007). According to these positive vascular effects of COX-2, it has been shown that the vasoprotective actions of high-density lipoproteins (HDL) are partly mediated by their ability to induce PGI2 release in a COX-2-dependent manner in vascular SMC and endothelial cells (MartínezGonzález and Badimon, 2007). These facts suggest an antiatherogenic role of the isoenzyme. Our recent data demonstrate that one of the mechanisms by which oleanolic acid evokes vasoprotection is via induction of PGI2 release through COX-2-dependent mechanisms in vascular cells (Martínez-González et al., 2008). In this study, oleanolic acid is a strong inducer of PGI2 synthesis and COX-2 up-regulation in human coronary SMC,
818
SECTION | II Vascular Aspects Including Hypertension
Oleanolic acid
COOH
HO
MEK1/2
p38 MAPK
ERK1/2
AA
CREB COX-2
COX-2
PGI2
FIGURE 86.4 Effect of oleanolic acid on vascular prostacyclin (PGI2) release and cyclooxygenase-2 (COX-2) regulation. This figure summarizes the signaling pathways involved in oleanolic-acid-induced COX-2 up-regulation and PGI2 release in human coronary smooth muscle cells. This response involves the activation of the transcription factor CREB (cAMP response element binding protein) by the upstream stimulation of p38-mitogen-activated protein kinase (MAPK) and extracellular signalregulated kinase (ERK). AA: arachidonic acid. Adapted from MartinezGonzalez et al. (2008).
without any effect on the synthesis of the vasoncostrictor eicosanoid TXA2. The pattern of modulation of vascular eicosanoid production and COX-2 up-regulation evoked by oleanolic acid is similar to that activated by HDL, involving p38 MAPK and ERK1/2 activation (Figure 86.4). This study also evidenced that erythrodiol also up-regulates COX-2 expression and PGI2 release in coronary SMC but at a lower extent than that provided by oleanolic acid. On the contrary, the olive oil phenol hydroxytyrosol was unable to modify PGI2 levels and COX-2 expression. Therefore, we established that up-regulation of the COX-2 pathway could be a property of triterpenoids not shared by other bioactive compounds of olive oil.
86.7 EFFECTS OF LONG-TERM INTAKE OF OLIVE OIL RICH IN TRITERPENOIDS ON ENDOTHELIAL FUNCTION Epidemiological studies and feeding trials indicate that olive oil could favorably affect blood pressure control (Martínez-González, 2006; Bondia-Pons et al., 2007). Although the precise molecular mechanisms and contribution of different components from olive oil to blood pressure lowering are not fully understood, several authors suggest that this action is mediated by improving endothelial function (Herrera et al., 2001; Alonso et al., 2006; Perona et al., 2006). We recently evaluated the effects
of pomace olive oil oral intake on endothelial dysfunction associated with hypertension (Rodriguez-Rodriguez et al., 2007). The major finding of this study evidenced that long-term intake of a diet enriched in olive pomace oil, rich in minor constituents such as oleanolic acid, restores the impaired endothelial NO-mediated vasodilatation of aortic rings from SHR as measured by the relaxant response to acetylcholine (Figure 86.5). The main mechanism mediating the improvement in endothelial function related to hypertension was an increased NO bioavailability since animals fed with pomace olive-oil-enriched diets showed an enhanced aortic eNOS expression (Figure 86.5) and increased plasma NOx (nitrates ⫹ nitrites, main NO metabolites) levels. Furthermore, oral intake of a pomace olive oil diet supplemented in oleanolic acid (up to 800 ppm) also increased vasodilatation to acetylcholine and attenuated phenylephrine-induced contractions in aorta from normotensive and SHR by NO-dependent mechanisms (Rodriguez-Rodriguez et al., 2007). Because blood pressure was not modified by any of the diets assayed, we cannot explain the endothelial function improvement associated with pomace olive oil rich in oleanolic acid in diet by a secondary event to antihypertensive effect but by their protective action in the vascular endothelium. In contrast to this, previous investigations evidenced that in vivo administration of oleanolic acid prevented hypertension in another rat model of hypertension (Dahl salt-sensitive rats) (Somova et al., 2003). However, this antihypertensive effect has been shown with a higher dose of oleanolic acid and after intraperitoneal administration. The authors established the involvement of hypolipidemic, antioxidant and diuretic actions in the antihypertensive response induced by this olive oil triterpenoid. Overall, the benefits of pomace olive oil, olive oil, and their minor constituents on blood pressure regulation could be mediated through their protective effect on vascular endothelial function.
86.8 CONCLUSION Numerous epidemiological studies have evidenced an inverse relationship between olive oil intake and cardiovascular diseases. However, the main olive oil constituents responsible for this cardiovascular protection and the exact mechanisms involved are not entirely understood. Evaluation of cardioand vasculoprotective activities exerted by triterpenoids from olive and pomace olive oil is currently in progress. In vitro data reveal that olive oil triterpenoids (oleanolic acid, maslinic acid, erythrodiol, uvaol) induce vasodilatation in isolated conductance and resistance arteries from normotensive and hypertensive rats. The response is mainly mediated by endothelial NO. Differences observed in potency of dilatation and endothelial dependence are attributed to structural differences. The most prominent endothelium-dependent vasodilatation is elicited by oleanolic acid which evokes
819
CHAPTER | 86 Vasorelaxant Effects of Oleanolic Acid and Erythrodiol in Pomace Olive Oil
B
0
0
% Relaxation
% Relaxation
A
25 50
25 50
75 100
75 *
WKY −9
C BD
100 −8
−7 −6 Log [ACh]
−5
−9
WKY OL POM POMO
* *
SHR
D BD
−8
−7 −6 Log [ACh]
−5
SHR OL POM POMO
eNOS
−140 KDa
eNOS
−140 KDa
α-actin
−42 KDa
α-actin
−42 KDa
BD
OL
POM
POMO
FIGURE 86.5 Effects of long-term intake of olive oil rich in oleanolic acid on endothelial function. Chronic intake of pomace olive-oil-enriched diets reduced the endothelial dysfunction associated with hypertension by an improvement in endothelium- and NO-dependent relaxation and vascular endothelial nitric oxide synthase (eNOS) expression. Panels (A) and (B) show the relaxation induced by acetylcholine (ACh) on intact aortic segments from either normotensive (WKY) or spontaneously hypertensive rats (SHR) fed with BD (䉲, low fat, basal diet), OL (䉭, olive oil), POM (䊊, pomace olive oil) or POMO (䊉, pomace olive oil enriched in oleanolic acid) for 12 weeks. Values are mean ⫾ SEM. *p ⬍ 0.05 vs BD group. (C,D): Western blots showing eNOS protein expression in aorta homogenates from WKY and SHR from different dietary groups. Adapted from Rodriguez-Rodriguez et al. (2007).
NO release by calcium-independent mechanisms via eNOS phosphorylation. Moreover, we recently determined that the vasorelaxant effects of oleanolic acid are also associated with up-regulation of COX-2 and PGI2 release in human coronary SMC. In vivo studies have reported the antihypertensive effect of olive oil triterpenoids involving antioxidant and antihyperlipidemic activities. Additionally, long-term intake of pomace olive oil enriched in oleanolic acid restores the endothelial dysfunction associated with hypertension by an enhancement of endothelial NO and vascular eNOS expression. Altogether, oleanolic acid and structuralrelated triterpenoids could be regarded as bioactive molecules contributing to the beneficial effects of olive oil and the Mediterranean diet. However, further investigations are needed to confirm the relevance in humans regularly consuming olive oil, especially olive oils such as pomace olive oil with a high content of oleanolic acid.
●
●
●
●
●
Olive oil triterpenoids (oleanolic acid, maslinic acid, erythrodiol, uvaol) induce vasodilatation in isolated conductance and resistance arteries from normotensive and hypertensive rats. The most prominent endothelium-dependent vasodilatation is elicited by oleanolic acid, which evokes nitric oxide release by eNOS phosphorylation via calciumindependent mechanisms. Oleanolic acid induces the release of the vasodilator and atheroprotector prostanoid prostacyclin from human vascular endothelial cells. Long-term intake of pomace olive oil enriched in oleanolic acid reduces the endothelial dysfunction associated to hypertension. Altogether we propose that the intake of pomace olive oil and oleanolic acid could protect the vascular wall against cardiovascular pathologies such as hypertension and atherosclerosis.
SUMMARY POINTS ●
●
The olive subproduct pomace olive oil is an important source of pentacyclic triterpenoids, mainly oleanolic acid, with concentrations exceeding the values found in olive oil. Despite the numerous biological activities attributed to oleanolic acid, the effects of oleanolic acid and pomace olive oil on vascular and endothelial function have remained unknown until now.
REFERENCES Alonso, A., Ruiz-Gutierrez, V., Martinez-Gonzalez, M.A., 2006. Monounsaturated fatty acids, olive oil and blood pressure: epidemiological, clinical and experimental evidence. Public Health Nutr. 9, 251–257. Bondia-Pons, I., Schröder, H., Covas, M.I., Castellote, A.I., Kaikkonen, J., Poulsen, H.E., Gaddi, A.V., Machowetz, A., Kiesewetter, H., LópezSabater, M.C., 2007. Moderate consumption of olive oil by healthy
820
European men reduces systolic blood pressure in non-Mediterranean participants. J. Nutr. 137, 84–87. Busse, R., Edwards, G., Feletou, M., Fleming, I., Vanhoutte, P.M., Weston, A.H., 2002. EDHF: bringing the concepts together. Trends Pharmacol. Sci. 23, 374–380. Esposito, K., Marfella, R., Ciotola, M., Di Palo, C., Giugliano, F., Giugliano, G., D’Armiento, M., D’Andrea, F., Giugliano, D., 2004. Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA 292, 1440–1446. Fleming, I., Busse, R., 2003. Molecular mechanisms involved in the regulation of the endothelial nitric oxide synthase. Am. J. Physiol. Regul. Integr. Comp. Physiol. 284, R1–R12. Heller, R., Hecker, M., Stahmann, N., Thiele, J.J., Werner-Felmayer, G., Werner, E.R., 2004. Alpha-tocopherol amplifies phosphorylation of endothelial nitric oxide synthase at serine 1177 and its short-chain derivative trolox stabilizes tetrahydrobiopterin. Free Radic. Biol. Med. 37, 620–631. Herrera, M.D., Pérez-Guerrero, C., Marhuenda, E., Ruiz-Gutiérrez, V., 2001. Effects of dietary oleic-rich oils (virgin olive and high-oleicacid sunflower) on vascular reactivity in Wistar-Kyoto and spontaneously hypertensive rats. Br. J. Nutr. 86, 349–357. Herrera, M.D., Rodríguez-Rodríguez, R., Ruiz-Gutiérrez, V., 2006. Functional properties of pentacyclic triterpenes contained in “orujo” olive oil. Curr. Nutr. Food Sci. 2, 45–50. Marquez-Martin, A., de la Puerta-Vazquez, R., Fernandez-Arche, A., Ruiz-Gutierrez, V., 2006. Supressive effect of maslinic acid from pomace olive oil on oxidative stress and cytokine production in stimulated murine macrophages. Free Radic. Res. 40, 295–302. Martínez-González, J., Badimon, L., 2007. Mechanisms underlying the cardiovascular effects of COX-inhibition: benefits and risks. Curr. Pharm. Des. 13, 2215–2227. Martínez-González, J., Rodríguez-Rodríguez, R., González-Díez, M., Rodríguez, C., Herrera, M.D., Ruiz-Gutierrez, V., Badimon, L., 2008. Oleanolic acid induces prostacyclin release in human vascular smooth muscle cells through a cyclooxygenase-2-dependent mechanism. J. Nutr. 138, 443–448. Martínez-González, M.A., 2006. The SUN cohort study (Seguimiento University of Navarra). Public Health. Nutr. 9, 127–131. Mombouli, J.V., Vanhoutte, P.M., 1999. Endothelial dysfunction: from physiology to therapy. J. Mol. Cell. Cardiol. 31, 61–74. O’Sullivan, S.E., Kendall, D.A., Randall, M.D., 2004. Heterogeneity in the mechanisms of vasorelaxation to anandamide in resistance and conduit rat mesenteric arteries. Br. J. Pharmacol. 142, 435–442.
SECTION | II Vascular Aspects Including Hypertension
Pérez-Camino, M.C., Cert, A., 1999. Quantitative determination of hydroxy pentacyclic triterpene acids in vegetable oils. J. Agric. Food Chem. 47, 1558–1562. Perona, J.S., Arcemis, C., Ruiz-Gutierrez, V., Catala, A., 2005. Effect of dietary high-oleic-acid oils that are rich in antioxidants on microsomal lipid peroxidation in rats. J. Agric. Food Chem. 53, 730–735. Perona, J.S., Cabello-Moruno, R., Ruiz-Gutierrez, V., 2006. The role of virgin olive oil components in the modulation of endothelial function. J. Nutr. Biochem. 17, 429–445. Rodriguez-Rodriguez, R., Herrera, M.D., de Sotomayor, M.A., RuizGutierrez, V., 2007. Pomace olive oil improves endothelial function in spontaneously hypertensive rats by increasing endothelial nitric oxide synthase expression. Am. J. Hypertens. 20, 728–734. Rodriguez-Rodriguez, R., Herrera, M.D., Perona, J.S., Ruiz-Gutierrez, V., 2004. Potential vasorelaxant effects of oleanolic acid and erythrodiol, two triterpenoids contained in ‘orujo’ olive oil, on rat aorta. Br. J. Nutr. 92, 635–642. Rodriguez-Rodriguez, R., Perona, J.S., Herrera, M.D., Ruiz-Gutierrez, V., 2006. Triterpenic compounds from “orujo” olive oil elicit vasorelaxation in aorta from Spontaneously Hypertensive rats. J. Agric. Food Chem. 54, 2096–2102. Rodriguez-Rodriguez, R., Stankevicius, E., Herrera, M.D., Petersen, L.O., Andersen, M.R., Ruiz-Gutierrez, V., Simonsen, U., 2008. Oleanolic acid, a component of olive oil, evokes calcium-independent release of endothelium-derived nitric oxide. Br. J. Pharmacol. In press. Schmitt, C.A., Handler, N., Heiss, E.H., Erker, T., Dirsch, V.M., 2007. No evidence for modulation of endothelial nitric oxide synthase by the olive oil polyphenol hydroxytyrosol in human endothelial cells. Atherosclerosis 195, 58–64. Shimokawa, H., Yasutake, H., Fujii, K., Owada, M.K., Nakaike, R., Fukumoto, Y., Takayanagi, T., Nagao, T., Egashira, K., Fujishima, M., Takeshita, A., 1996. The importance of the hyperpolarizing mechanism increases as the vessel size decreases in endothelium-dependent relaxations in rat mesenteric circulation. J. Cardiovasc. Pharmacol. 28, 703–711. Shimokawa, H., 1998. Endothelial dysfunction in hypertension. J. Atheroscler. Thromb. 4, 118–127. Somova, L.I., Nadar, A., Rammanan, P., Shode, F.O., 2003. Cardiovascular, antihyperlipidemic and antioxidant effects of oleanolic and ursolic acids in experimental hypertension. Phytomedicine 10, 115–121. Taddei, S., Virdis, A., Ghiadoni, L., Magagna, A., Salvetti, A., 1997. Cyclooxygenase inhibition restores nitric oxide activity in essential hypertension. Hypertension 29, 274–279.
Chapter 87
Endothelial Activation and Olive Oil Maria Annunziata Carluccio1,2, Marika Massaro1,2, Egeria Scoditti1,2 and Raffaele De Caterina2,3 1
C.N.R. Institute of Clinical Physiology, Lecce Section, Italy C.N.R. Institute of Clinical Physiology, Pisa, Italy 3 Institute of Cardiology and Center of Excellence on Aging, ‘G. d’Annunzio’ University, Chieti, Italy
2
87.1 INTRODUCTION Atherosclerosis remains the leading cause of death in the developed countries and is expected to become the most common cause of disease-related disability and mortality worldwide in the near future due to its close association with the alarming pandemics of obesity and diabetes, two major cardiovascular risk factors. Nutritional factors have probably the most important role in the development of atherosclerosis. A consensus about their role has gradually emerged (Willet, 1994). On the one hand, diets high in saturated and trans-fatty acids are pro-atherogenic. On the other hand, a high intake of unsaturated fatty acid and/or antioxidant compounds (vitamins and non-vitamins) can reduce both pro-atherogenic risk factors and the onset and progression of the plaque.
87.2 MEDITERRANEAN DIETS AND CARDIOVASCULAR DISEASE One of the emerging strategies that has gained particular credit at this present time is the dietary preventive approach termed ‘Mediterranean diet’ or, better, ‘Mediterranean diets’, which share a high consumption, normally with meals, of plant-based foods (fruits, vegetables, whole grain cereals, nuts and legumes), a low consumption of red meat, a moderate consumption of fish and wine, and olive oil as the principal source of fat. The concept of the Mediterranean diet originated from the Seven Countries Study initiated by Keys in the 1950s. This study showed a low prevalence of coronary heart disease (CHD) in residents of Mediterranean countries, despite a high fat intake. The results of a 25-year follow-up study also have indicated that the incidence of cardiovascular disease in general was much lower in southern than in northern European countries (Menotti et al., 2000). A potential explanation for the lower cardiovascular mortality rates among the Mediterranean population is their traditional diet. Accordingly, the favorable Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
mortality statistics were confirmed by the World Health Organization database, covering the period from 1960 to 1990, suggesting that something unusual has been positively affecting cardiovascular health and the adult life expectancy in the Mediterranean populations for centuries, despite the fact that all socio-economic indicators explored were actually much worse and the prevalence of smoking unusually high in countries of the Mediterranean region compared with the more industrialized ones (Willet, 1994). In a recent prospective investigation involving a large Greek cohort of the European Prospective Investigation into Cancer and Nutrition (EPIC) study, an inverse correlation was shown between a greater adherence to the Mediterranean diet and total mortality, in particular mortality from CHD, independent of sex, smoking status, level of education, body mass index, and the level of physical activity (Trichopoulou et al., 2003). The relation becomes stronger with increasing age, thus reflecting increasing cumulative exposure to dietary factors. In other studies, the adherence to similar healthful lifestyle practices was associated with an 83% reduction in the rate of CHD (Stampfer et al., 2000), a 91% reduction of diabetes in women (Hu et al., 2001), and a 71% reduction in colon cancer in men (Platz et al., 2000). Therefore, it has become clear that the Mediterranean diet is a healthy-eating pattern with protective effects against a variety of chronic diseases, especially cardiovascular diseases.
87.3 ATHEROSCLEROSIS AS AN INFLAMMATORY DISEASE Atherosclerosis is by far the most frequent underlying cause of cardiac, cerebrovascular and peripheral arterial disease. Atherosclerosis is a chronic, progressive disease of the arterial wall with relevant inflammatory components in its inception, progression and complications (De Caterina and Libby, 1997; Ross, 1999). Atherosclerosis begins with the accumulation of monocytederived macrophages in the arterial intima. These cells, by
821
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
822
taking up lipid droplets (mostly oxidized or otherwise modified low-density lipoproteins – LDL), become foam cells, which are typical cellular elements of the ‘fatty streak’, the earliest detectable atherosclerotic lesion (Gimbrone et al., 1993). Animal observations have also shown that fatty streaks precede the development of ‘intermediate lesions’ (Masuda and Ross, 1990a, b), which are composed of macrophages and smooth muscle cells. The activation of these cell types, in the context of changes broadly described as ‘inflammatory’, leads to the release of hydrolytic enzymes (metalloproteinases and other proteases), cytokines, chemokines, and growth factors, which can induce focal necrosis or apoptosis. Cycles of mononuclear cell accumulation, migration and proliferation of smooth muscle cells, as well as the formation of fibrous tissue, lead to the enlargement and restructuring of the lesion, with the formation of a fibrous cap and other morphological changes. Plaques range from those with a prevailing lipid component (fatty lesions) to those with prevailing fibrous tissue (fibrous plaques). It is now believed that a physical rupture of inflamed plaques – and not lumen stenosis – is responsible for the majority of acute clinical manifestations of atherosclerosis, including acute myocardial infarction, stroke and sudden cardiac death (Libby et al., 2002).
87.4 PATHOGENESIS OF ATHEROSCLEROSIS: ROLE OF ENDOTHELIAL ACTIVATION Cardiovascular risk factors include elevated plasma cholesterol levels, hypertension, diabetes, smoking, male gender and new inflammatory markers (e.g., C-reactive protein, cytokines, hyperhomocysteinemia). Qualitatively and quantitatively abnormal stimuli associated with these risk factors, likely through the induction of heightened oxidative stress, alter the normal homeostatic function of the vascular endothelium (Table 87.1). A complex of functional phenotypic changes, in such conditions, renders the endothelium adhesive to circulating leukocytes, more permeable to solutes, and capable of the enhanced synthesis of mitogenic, chemoattractive, and inflammatory mediators. This concerted transcriptional activation of genes involved in leukocyte recruitment into the intima is termed ‘endothelial activation’ (Gimbrone et al., 1993), which is a subset of a more complex perturbation of normal functional properties of the endothelium known as ‘endothelial dysfunction’. Endothelial activation participates importantly in the initiation, progression, and clinical emergence of atherosclerotic vascular disease. While leukocytes do not adhere to a normal endothelium, they do adhere to an activated endothelium. Specific adhesion molecules expressed on the surface of activated endothelial cells mediate the processes of leukocyte adhesion and transmigration into the intima. Endothelial
SECTION | II Vascular Aspects Including Hypertension
TABLE 87.1 Functional properties of vascular endothelium. Properties affecting thrombosis Antiplatelet effects Anticoagulant effects Profibrinolytic effects Antifibrinolytic effects Modulation of vascular tone Promotion of vasodilatation Promotion of vasoconstriction Modulation of smooth muscle cell proliferation and migration Promotion Inhibition Selective permeability Inhibition of leukocyte adhesion and migration
leukocyte adhesion molecules specifically involved in atherogenesis belong to the selectin and the immunoglobulin superfamily (Table 87.2). Selectins, by binding to sialyl Lewis X (sLex) or closely related carbohydrate structures as their ligands, likely mediate a transient, labile contact, or ‘rolling’, of leucocytes on the endothelial surface. Intercellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1, two members of the immunoglobulin superfamily, mediate the stable adhesion of leukocytes to the activated endothelium (Figure 87.1). In particular, VCAM-1 mediates the steady adhesion of leukocytes to the endothelium by binding the cognate integrin ligand very late antigen (VLA)-4, which is expressed specifically on the surface of circulating monocytes and T lymphocytes, thus mediating the selective mononuclear leukocyte recruitment occurring in early atherogenesis (Cybulsky and Gimbrone, 1991). While ICAM-1 is also abundantly expressed at lesion-prone areas in wild-type mice with normal cholesterol levels, VCAM-1 is specifically increased in arterial endothelial cells at lesion-prone areas in hypercholesterolemic mice and rabbits (Li et al., 1993; Iiyama et al., 1999). A recent study has shown that hypomorphic variants expressing less VCAM-1 in atherosclerosis-prone LDL receptor (LDLR)⫺/⫺ mice feature reduced lesion formation (Dansky et al., 2001). Proinflammatory cytokines such as interleukin (IL)-1, tumor necrosis-factor (TNF)-α and CD40 ligand (CD40L) expressed in atheromata induce the expression of VCAM-1, ICAM-1, E-selectin, and soluble endothelial products, including monocyte chemoattractant protein (MCP)-1, macrophage-colony stimulating factor (M-CSF), IL-6, and IL-8. Components of oxidized lipoproteins, such as oxidized phospholipids and short-chain aldehydes, as well as the
823
CHAPTER | 87 Endothelial Activation and Olive Oil
TABLE 87.2 Endothelial–leukocyte adhesion molecules, chemoattractants and their cognate ligands implicated in atherosclerosis. Gene family
Molecules/CD nomenclature
Cognate ligands
Selectins
L-selectin (CD62L) E-selectin (CD62E) P-selectin (CD62P)
Sialyl Lewisx Sialyl Lewisx Sialyl Lewisx
Immunoglobulins
ICAM-1(CD54)
LFA-1 (CD11a/CD18) Mac-1 (CD11b/CD18) (CD11c/CD18) LFA-1, Mac-1, CD11c/CD18 LFA-1, Mac-1, CD11c/CD18 VLA4 (CD49/CD29) PECAM-1(CD31
Surface associated
ICAM-2 (CD102) ICAM-3 (CD50) VCAM-1 (CD106) PECAM-1(CD31)
Secreted Cytokines Lipid
MCP-1 M-CSF PAF
MCP-1 receptor M-CSF-recepto PAF receptor
ICAM-1,-2,-3: intercellular adhesion molecule-1,-2,-3; VCAM-1: vascular adhesion molecule-1 PECAM-1: platelet endothelial cells adhesion molecule-1; LFA-1: leukocyte function associated antigen1; Mac-1: macrophage antigen-1; VLA4: very late antigen-4; MCP-1: monocyte chemoattractant protein-1; M-CSF: macrophage-colony stimulating factor-1; PAF: platelet activating factor.
advanced glycation end-products, also induce VCAM-1 expression and endothelial activation, thus promoting early atherogenesis (Kume et al., 1992; Basta et al., 2002).
87.5 NUCLEAR FACTOR-κB: A COMMON DENOMINATOR IN VASCULAR DISEASE Since endothelial activation involves de novo or increased expression of specific leukocyte adhesion molecules and other soluble products, it has been suggested that one or a few transcription factors may be responsible for the concerted induction of various endothelial genes involved in endothelial activation. The nuclear factor-kappa B (NF-κB) system is now recognized as the common denominator of endothelial activation (De Martin et al., 2000). NF-κB comprises a family of transcription factors that form homo- and heterodimers, the most prominent being the p65/p50 heterodimer. Interaction with the inhibitory subunit IκB retains NF-κB in the cytoplasm in an inactive state under basal conditions. NF-κB activates rapidly in response to a variety of stimuli that always lead to IκB degradation. Important stimuli include TNF-α, IL-1, bacterial lipopolysaccharide
(LPS), hyperglycemia, platelet-activating factor, shear stress, oxidized lipids, oxidant stress, and hypoxia/reperfusion (Pahl, 1999). Independent of the type of stimuli, co-treatment with antioxidants or metal chelators can inhibit NF-κB activation. Therefore, changes in the cellular redox balance may alter NF-κB activation (Collins et al., 1995). NF-κB exhibits very rapid and transient activation, making it well-suited for the expression of many immuneand ‘stress’-response genes that require action on demand for only a limited period of time.
87.6 EFFECTS OF OLIVE OIL ON ENDOTHELIAL ACTIVATION Although the healthful properties of Mediterranean diets as a whole have gained recognition, underlying mechanisms are less known and most likely quite varied and complex. Basic research is nowadays concentrating efforts in clarifying the mechanism of action of individual food items, e.g., cereals, fruits, vegetable, olive oil, and of their components, including fibers, vitamins and polyphenols. For centuries, olive oil, largely produced in Mediterranean countries, has been thought not only to contribute nutrients to the diet, but also to have a positive impact on health, so that it was traditionally treasured as a functional food. The healthful properties of olive oil have been attributed to its high content in monounsaturated fatty acids, namely in the form of oleic acid. Mata et al. assessed the effect of an olive-oil-rich diet on inflammation. LDL induction of monocyte-adhesion to endothelial cells was lower after monounsaturated fatty acid consumption than after consumption of either saturated or polyunsaturated fatty acids (Mata et al., 1996). Isolated human LDL enriched in oleic acid promoted less monocyte chemotaxis and monocyte adhesion as compared with linoleic-acid-enriched LDL, when exposed to oxidative stress (Tsimikas et al., 1999). These data are in agreement with the previously demonstrated capacity of oleic acid of reducing the susceptibility of LDL to oxidation compared with linoleic acid, in addition to reducing plasma concentrations of LDL cholesterol (Parthasarathy et al., 1990). However, recent data also support the concept that oleic acid is not solely responsible for all antiatherogenic properties of olive oil. Indeed, virgin and extra virgin olive oils have a peculiar composition based on hundreds of non-fat components with great biological potential, including a number of polyphenolic compounds, such as the simple phenols (tyrosol and hydroxytyrosol) and the secoiridoids (oleuropein and its conjugated forms) (Visioli and Galli, 2000). Several recent lines of evidence indicate that olive oil and its isolated components, oleic acid and polyphenols, besides improving lipid profile (Covas et al., 2006), blood pressure (Fito et al., 2005) and endothelial dysfunction (Ruano et al., 2005), also interfere with atherogenesis, by a direct effect on vascular cells, as discussed here below.
824
SECTION | II Vascular Aspects Including Hypertension
MONOCYTE CD11a/b VLA
4
ROLLING
SLex FIRM ADHESION
ACTIVATED ENDOTHELIUM VCAM-1
DIAPEDESIS
E-selectin ICAM-1
PECAM-1 1 Extracellular matrix MACROPHAGE INTIMA
FOAM CELL
FIGURE 87.1 The different phases of monocyte adhesion to the activated endothelium and molecular ligands implicated. The first phase, denoted as ‘rolling’, consists of a slowing-down of monocytes when activated endothelial cells, expressing E-selectin, tether flowing monocytes through a labile binding of E-selectin to carbohydrate ligands (syalyl Lewisx, sLex) already constitutively expressed on the monocyte surface. By considerably decreasing monocyte speed, rolling allows monocytes to ‘sense’ the atmosphere of specific chemoattractants, namely the chemokine MCP-1. Monocyte integrintype receptors (LFA-1, Mac-1, and VLA4) recognize immunoglobulin ligands expressed de novo on the surface of activated endothelial cells, namely ICAM-1 and VCAM-1. This brings to a labile and then firm attachment of monocytes on the endothelial surface (‘arrest’ and ‘spreading’) and their subsequent ‘diapedesis’, this last through other types of molecular interactions likely involving platelet–endothelial cell adhesion molecule-1 (PECAM-1) on the endothelial surface.
87.7 OLEIC ACID AND ENDOTHELIAL ACTIVATION Studies on endothelial cells in vitro have shown that oleic acid, as well as omega-3 polyunsaturated fatty acids, can decrease endothelial dysfunction and activation (Christon, 2003). In an in vitro model of early atherogenesis based on cultured endothelial cells challenged with inflammatory stimuli, we have observed that supplementation of oleic acid is able to reduce the stimulated surface expression of endothelial leukocyte adhesion molecules, such as VCAM-1, ICAM-1 and E-selectin, in a concentration(10–100 μmol L⫺1) and time-dependent (6–72 h) fashion, also decreasing the stimulated production of endothelial chemoattractants, such as M-CSF (Carluccio et al., 1999; Massaro et al., 1999, 2002). The effect appears to be a generalized ‘quenching’ of endothelial activation, and is accompanied by a functional counterpart in terms of inhibition of monocytoid cell adhesion to the endothelium. The inhibition of protein expression was paralleled by a decrease in the corresponding messenger RNA (mRNA), suggesting possible transcriptional inhibition. Indeed oleic acid was shown to be able to inhibit the nuclear translocation and consequent activation of NF-κB, thus providing a possible general mechanism by which it might reduce the expression of pro-inflammatory genes involved in atherogenesis (Figure 87.2). NF-κB is a redox-sensitive transcription factor, which can be activated by an overproduction of reactive oxygen species (ROS) (Collins et al., 1995). Two potential mechanisms might explain the capacity of oleic acid to inhibit ROS-mediated NF-κB activation: reduced enzymatic production or increased scavenging of ROS. Both mechanisms seem to be in operation with oleic acid: the incubation
of cytokine-stimulated endothelial cells with oleic acid prevented the depletion of glutathione, and partially prevented the stimuli-induced increase of intracellular ROS (Massaro et al., 2002) (Figure 87.2). This occurred without any change in the activity of glutathione-related antioxidant enzymes, superoxide dismutase and catalase. Therefore, oleic acid may exert direct atheroprotective effects by inhibiting endothelial activation through a quenching of ROS (Massaro et al., 2002). The inhibitory potency of unsaturated fatty acids appears to be directly proportional to their number of double bonds (Massaro et al., 1999). Oleic acid would therefore act if it is able to selectively displace saturated fatty acid, but not polyunsaturated fatty acids, in cell membrane phospholipids, thus modulating NF-κB activation and the expression of genes encoding molecules involved in monocyte recruitment (Massaro et al., 2002). The plausibility of an in vivo relevance of these findings is in the relative abundance of oleic acid in the diet (oleate being the most abundant dietary unsaturated fatty acid), and in the fact that plasma concentrations of oleic acid under conditions of high olive oil consumption are likely to be fully in the range of concentrations exerting biological effects in most studies, between 10 and 100 μmol L⫺1.
87.8 OLIVE OIL POLYPHENOLIC COMPOUNDS AND ENDOTHELIAL ACTIVATION Olive oil obtained from the whole fruit of Olea europaea by physical pressure contains several characteristic minor compounds, which certainly contribute to its healthful
825
CHAPTER | 87 Endothelial Activation and Olive Oil
Cytokines
Phorbol ester
LPS
(−) Monocyte adhesion
ox-LDL
PKC NF-κB
IKK ROS
VCAM-1
p50 p65
(−)
IκB
p50 p65
E-Selectin ICAM-1
P
IL-8
p50 p65 mRNA
IL-6 M-CSF
Plasma membrane
MCP-1, etc.
P (−)
5
0
p5
Nucleus
p6
NF-κB target gene
FIGURE 87.2 Possible molecular sites of interference by olive oil components with endothelial activation. The NF-κB system comprises a series of transcription factors that form homo- and heterodimers, the most prominent one being the p65/p50 heterodimer, normally sequestered in the cytoplasm and bound to inhibitory-κB (IκB). In the classical pathway of NF-κB activation, upon the influence of inflammatory cytokines, lipopolysaccharide (LPS) and other stimuli able to activate endothelial cells, the generation of intracellular reactive oxygen species (ROS), mostly H2O2, occurs. This activates the IκB kinase (IKK) complex, resulting in the phosphorylation of IκBα. This, in turn, leads to its subsequent ubiquitination and degradation by the proteasome. Once released, NF-κB translocates to the nucleus and binds specific sequences in the regulatory regions of target genes, resulting in the increased expression of monocyte adhesion molecules (VCAM-1, ICAM-1, E-selectin) and chemoattractants (M-CSF, MCP-1, IL-6, IL-8, etc.). The proposed molecular model for oleic acid and olive oil polyphenol interference with endothelial activation is illustrated (minus sign). Both oleic acid and olive oil polyphenols, possibly by a quenching of ROS, would prevent the activation of the NF-κB signaling pathway and the subsequent induction of adhesion molecules and chemoattractants, thus inhibiting monocyte recruitment to vascular endothelium.
properties. Minor compounds account for 1–2% of the oil. Among these, the best studied are the polyphenol hydroxytyrosol and its derivative oleuropein, both endowed with potent antioxidant capacity. Indeed, hydroxytyrosol and oleuropein are able to scavenge free radicals, due to the hydrogen-donating capacity of the hydroxyl group in the ortho-diphenolic structure (Visioli et al., 1995) (Figure 87.3). These compounds inhibit the oxidation of LDL induced by copper sulfate and by metal-independent oxidative systems (Visioli et al., 1995). Saija et al. have hypothesized that, while hydroxytyrosol can serve as a scavenger of aqueous peroxyl radicals near the membrane surface, oleuropein also acts as a scavenger of chain-propagating lipid peroxyl radicals within the membranes (Saija et al., 1998). The antioxidant capacity of olive oil polyphenols has also been confirmed in vivo. Visioli et al. found that passive-smoking-induced stress in human volunteers is inhibited by olive oil polyphenols, as demonstrated by a reduction in the urinary excretion of F2 isoprostanes, known markers of enhanced oxidative stress (Visioli et al., 2000b). Recent clinical studies in healthy male volunteers have also shown that markers of oxidative stress, including cholesterol-conjugated dienes, hydroxyfatty acids and products of DNA oxidative damage, all decreased linearly with the increase in the phenolic content of olive oil (Weinbrenner et al., 2004; Covas et al., 2006).
Name
Structure
Antioxidant properties
HO
Hydroxytyrosol
Radical scavenger
OH
HO HO HO
COOCH3
O H
Oleuropein aglycone
Radical scavenger
O OH
FIGURE 87.3 Structure, name and physico-chemical characteristics of hydroxytyrosol and oleuropein aglycone used in our experiments.
Interventional studies administering doses of phenolic compounds in virgin olive oil have also reported a reduced oxidative status in dyslipidemic subjects (Visioli et al., 2005). Since oxidative stress is critically involved in endothelial activation, we devised a series of experiments to investigate whether olive oil polyphenols influence endothelial responses to proinflammatory stimuli triggering endothelial activation. In accordance with the widely recognized antioxidant capacity of olive oil polyphenols and the involvement of enhanced oxidative stress in endothelial activation, we found that oleuropein and hydroxytyrosol significantly inhibited LPS or cytokine-stimulated expression
of VCAM-1, as well as the adhesion of monocytes to activated endothelium in a concentration-dependent fashion. The IC50 for these effects was around 30 μmol L⫺1 for hydroxytyrosol and 15 μmol L⫺1 for oleuropein aglycone, this last being more effective than its corresponding glycoside analogue (Carluccio et al., 2003). No significant effects on adhesion molecule expression were obtained with the two other tested phytochemicals from olive oil, elenolic acid and tyrosol, devoid of antioxidant activity (Carluccio et al., 2003). Similarly to VCAM-1, hydroxytyrosol and oleuropein reduced the stimulated expression of E-selectin and ICAM-1, indicating a generalized effect on endothelial activation (Carluccio et al., 2003). Moreover, the combination of low concentration of hydroxytyrosol and oleic acid, which better reproduces olive oil composition, resulted in an additive inhibitory effect on monocyte adhesion to the endothelium (Figure 87.4). Our results are in agreement with a more recent study of Dell’Agli et al., who found that the phenolic extract from extra virgin olive oil, as well as its individual components oleuropein aglycone and hydroxytyrosol, reduce the endothelial surface expression of the adhesion molecules VCAM-1, ICAM-1 and E-selectin and the corresponding mRNA levels at physiologically relevant concentrations (IC50 ⬍ 1 μmol L⫺1) (Dell’Agli et al., 2006). Indeed, olive oil polyphenols are dose-dependently absorbed and metabolized in humans (Visioli et al., 2000a). Vissers et al. reported ⬎55% absorption over a wide dose range of extra virgin olive-oil-derived phenolic compounds by healthy volunteers, with plasma concentration not exceeding 1 μmol L⫺1, similarly to other phenolic compounds (Vissers et al., 2002). It is likely that beneficial effects on endothelial activation would be amplified in vivo because of the continuous exposure of vascular endothelium to these compounds and a possible additive or synergistic effect of other co-administrated bioactive compounds, including other polyphenols. Indeed, co-treatment of endothelial cells with hydroxytyrosol and resveratrol, a typical red wine polyphenol, produced a more-than-additive inhibitory effect on monocyte adhesion to endothelial cells (Carluccio et al., 2007). These effects occurred independent of the stimuli used to elicit endothelial activation, arguing that antioxidant polyphenols act downstream of any membrane receptor, at a step common to all agents and all genes of endothelial activation. Accordingly, we have recently shown that hydroxytyrosol, at micromolar concentrations, inhibits endothelial VCAM-1 expression induced by elevated levels of homocysteine, a pro-oxidant and proatherogenic trigger (Carluccio et al., 2007). Being the gene of VCAM-1 transcriptionally regulated by redox-sensitive transcription factors, such as NF-κB and activator protein (AP)-1 (Collins et al., 1995), we evaluated the potential interference with VCAM-1 gene expression by olive oil polyphenols. Transfection studies using various VCAM-1 gene promoter constructs showed that oleuropein aglycone and hydroxytyrosol repressed VCAM-1 gene
SECTION | II Vascular Aspects Including Hypertension
20 TNFα 10 ng/mL U937 cell adhesion to HUVEC (fold induction vs. control) (mean ±SD)
826
+ Oleate 10 μmol/L 15 + HT 1 μmol/L
10
+ HT 1 μmol/L + oleate 10 μmol/L
*
** **
5
0 TNFα10 ng/mL
FIGURE 87.4 Effects of hydroxytyrosol, oleic acid, or both on cytokineinduced monocytoid cell adhesion to HUVEC. The olive oil components hydroxytyrosol (1 μmol L–1), oleate (10 μmol L⫺1), or both significantly decrease U937 monocytoid cell adhesion to human umbilical vein endothelial cells (HUVEC) stimulated with 10 ng mL⫺1 TNFα for 20 hours. HUVEC were pre-treated with 1 μmol L⫺1 hydroxytyrosol for 1 hour, or 10 μmol L⫺1 oleate for 48 hours, or their combination, before the stimulation with 10 ng mL⫺1 TNFα for 20 hours, after which U937 cell adhesion to the endothelium was evaluated as described (Carluccio et al., 1999). Monocyte adhesion was expressed as fold-induction, making the value of unstimulated and untreated cells (control) ⫽ 1. Results represent the mean ⫾S.D. of adhering cells from three experiments, each consisting of 8 counts per condition. *p ⬍ 0.05 and **p ⬍ 0.01 vs. stimulated cells.
transcription through a mechanism involving an interference with activation of AP-1 and, mostly, NF-κB, as confirmed by electrophoretic mobility shift assays (Carluccio et al., 2003) (Figure 87.2). The above-reported in vitro data may explain the antiatherosclerotic effect of the extra virgin olive oil and hydroxytyrosol recently shown in animal models of atherosclerosis (Gonzalez-Santiago et al., 2006; Arbones-Mainar et al., 2007). Extra virgin olive oil decreased atherosclerotic lesions, reduced plaque size, and decreased macrophage recruitment in apolipoprotein E-deficient mice. Similarly, hydroxytyrosol improved the antioxidant status and reduced the size of atherosclerotic lesions in hyperlipidemic rabbits. Conflicting data have been recently reported for hydroxytyrosol in apoE-deficient mice, showing enhanced atherosclerotic lesion development (Acin et al., 2006). The discrepancy between such experimental outcomes may be explained by differences in ways of administration (i.e., hydroxytyrosol administered in isolation or in the original food matrix), experimental design and possible species-specific effect.
87.9 CONCLUSIONS Epidemiological studies, intervention studies and biochemical and metabolic results provide compelling evidence for the existence of cardiovascular benefits of olive oil, mostly
CHAPTER | 87 Endothelial Activation and Olive Oil
when synergistic and interactive effects among nutrients or foods potentially occur. The vasculoprotective effects of olive oil may be ascribed to its peculiar fat composition and its high content of bioactive compounds, in particular polyphenols, absent in seed oils. Olive oil, as a whole, can indeed be regarded as a functional food, with protective effects against several cardiovascular risk factors, such as oxidative stress, LDL cholesterol, hypertension, diabetes. In addition, olive oil directly modulates the responsiveness of the vascular wall to pro-atherogenic triggers by inhibiting the expression of genes involved in endothelial activation.
SUMMARY POINTS ●
●
●
●
●
●
Atherosclerotic vascular disease is the main cause of death and illness worldwide. Endothelial activation represents a crucial pathogenetic event in atherosclerosis. Mediterranean diets can be a cornerstone of cardiovascular disease prevention. Evidence from epidemiological and experimental studies has suggested that olive oil, the main fat of Mediterranean diets, features several antiatherogenic activities, both by reducing some pro-atherogenic risk factors and directly affecting the vascular responsiveness to inflammatory/pro-atherogenic triggers. The vasculoprotective effects of olive oil are attributable to its high content of oleic acid and, in addition, of the polyphenolic compounds, including hydroxytyrosol and its derivative oleuropein. Oleic acid and the polyphenolic compounds hydroxytyrosol and oleuropein inhibit endothelial activation, thus partially explaining atheroprotection from Mediterranean diets.
REFERENCES Acin, S., Navarro, M.A., Arbones-Mainar, J.M., Guillen, N., Sarria, A.J., Carnicer, R., Surra, J.C., Orman, I., Segovia, J.C., de la Torre, R., Covas, M.I., Fernandez-Bolanos, J., Ruiz-Gutierrez, V., Osada, J., 2006. Hydroxytyrosol administration enhances atherosclerotic lesion development in apo E deficient mice. J. Biochem. 140, 383–391. Arbones-Mainar, J.M., Navarro, M.A., Carnicer, R., Guillen, N., Surra, J.C., Acin, S., Guzman, M.A., Sarria, A.J., Arnal, C., Aguilera, M.P., Jimenez, A., Beltran, G., Uceda, M., Osada, J., 2007. Accelerated atherosclerosis in apolipoprotein E-deficient mice fed Western diets containing palm oil compared with extra virgin olive oils: a role for small, dense high-density lipoproteins. Atherosclerosis 194, 372–382. Basta, G., Lazzerini, G., Massaro, M., Simoncini, T., Tanganelli, P., Fu, C., Kislinger, T., Stern, D.M., Schmidt, A.M., De Caterina, R., 2002. Advanced glycation end products activate endothelium through signaltransduction receptor RAGE: a mechanism for amplification of inflammatory responses. Circulation 105, 816–822. Carluccio, M.A., Ancora, M.A., Massaro, M., Carluccio, M., Scoditti, E., Distante, A., Storelli, C., De Caterina, R., 2007. Homocysteine
827
induces VCAM-1 gene expression through NF-kappaB and NAD(P)H oxidase activation: protective role of Mediterranean diet polyphenolic antioxidants. Am. J. Physiol. Heart Circ. Physiol. 293, H2344–H2354. Carluccio, M.A., Massaro, M., Bonfrate, C., Siculella, L., Maffia, M., Nicolardi, G., Distante, A., Storelli, C., De Caterina, R., 1999. Oleic acid inhibits endothelial cell activation. Arterioscler. Thromb. Vasc. Biol. 19, 220–228. Carluccio, M.A., Siculella, L., Ancora, M.A., Massaro, M., Scoditti, E., Storelli, C., Visioli, F., Distante, A., De Caterina, R., 2003. Olive oil and red wine antioxidant polyphenols inhibit endothelial activation: antiatherogenic properties of Mediterranean diet phytochemicals. Arterioscler. Thromb. Vasc. Biol. 23, 622–629. Christon, R.A., 2003. Mechanisms of action of dietary fatty acids in regulating the activation of vascular endothelial cells during atherogenesis. Nutr. Rev. 61, 272–279. Collins, T., Read, M.A., Neish, A.S., Whitley, M.Z., Thanos, D., Maniatis, T., 1995. Transcriptional regulation of endothelial cell adhesion molecule: NF-κB and cytokine-inducible enhancer. FASEB J. 9, 899–909. Covas, M.I., Nyyssonen, K., Poulsen, H.E., Kaikkonen, J., Zunft, H.J., Kiesewetter, H., Gaddi, A., de la Torre, R., Mursu, J., Baumler, H., Nascetti, S., Salonen, J.T., Fito, M., Virtanen, J., Marrugat, J., Group, E.S., 2006. The effect of polyphenols in olive oil on heart disease risk factors: a randomized trial. Ann. Intern. Med. 145, 333–341. Cybulsky, M.I., Gimbrone, M.A., 1991. Endothelial expression of a mononuclear leukocyte adhesion molecule during atherogenesis. Science 251, 788–791. Dansky, H., Barlow, C., Lominska, C., Sikes, J., Kao, C., Weinsaft, J., Cybulsky, M., Smith, J., 2001. Adhesion of monocytes to arterial endothelium and initiation of atherosclerosis are critically dependent on vascular cell adhesion molecule-1 gene dosage. Arterioscler. Thromb. Vasc. Biol. 21, 1662–1667. De Caterina, R., Libby, P., 1997. Towards an understanding of the molecular pathogenesis of acute coronary syndromes. Cardiologia 42, 359–374. De Martin, R., Hoeth, M., Hofer-Warbinek, R., Schmid, J.A., 2000. The transcription factor NF-kappa B and the regulation of vascular cell function. Arterioscler. Thromb. Vasc. Biol. 20, E83–E88. Dell’Agli, M., Fagnani, R., Mitro, N., Scurati, S., Masciadri, M., Mussoni, L., Galli, G.V., Bosisio, E., Crestani, M., De Fabiani, E., Tremoli, E., Caruso, D., 2006. Minor components of olive oil modulate proatherogenic adhesion molecules involved in endothelial activation. J. Agric. Food Chem. 54, 3259–3264. Fito, M., Cladellas, M., de la Torre, R., Marti, J., Alcantara, M., PujadasBastardes, M., Marrugat, J., Bruguera, J., Lopez-Sabater, M.C., Vila, J., Covas, M.I., 2005. Antioxidant effect of virgin olive oil in patients with stable coronary heart disease: a randomized, crossover, controlled, clinical trial. Atherosclerosis 181, 149–158. Gimbrone, M.A., Kume, N., Cybulsky, M.I., 1993. Vascular endothelium dysfunction and the pathogenesis of atherosclerosis, Atherosclerosis Reviews. Reaven Press, New York. Gonzalez-Santiago, M., Martin-Bautista, E., Carrero, J.J., Fonolla, J., Baro, L., Bartolome, M.V., Gil-Loyzaga, P., Lopez-Huertas, E., 2006. One-month administration of hydroxytyrosol, a phenolic antioxidant present in olive oil, to hyperlipemic rabbits improves blood lipid profile, antioxidant status and reduces atherosclerosis development. Atherosclerosis 188, 35–42. Hu, F.B., Manson, J.E., Stampfer, M.J., Colditz, G., Liu, S., Solomon, C.G., Willett, W.C., 2001. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N. Engl. J. Med. 345, 790–797.
828
Iiyama, K., Hajra, L., Iiyama, M., Li, H., DiChiara, M., Medoff, B.D., Cybulsky, M.I., 1999. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ. Res. 85, 199–207. Kume, N., Cybulsky, M., Gimbrone, M.J., 1992. Lysophosphatidylcholine, a component of atherogenic lipoproteins, induces mononuclear leukocyte adhesion molecules in cultured human and rabbit arterial endothelial cells. J. Clin. Invest. 90, 1138–1144. Li Jr., H., Cybulsky, M.I., Gimbrone, M.A., Libby, P., 1993. An atherogenic diet rapidly induces VCAM-1, a cytokine-regulatable mononuclear leukocyte adhesion molecule, in rabbit aortic endothelium. Arterioscler. Thromb. 13, 197–204. Libby, P., Ridker, P., Maseri, A., 2002. Inflammation and atherosclerosis. Circulation 105, 1135–1143. Massaro, M., Basta, G., Lazzerini, G., Carluccio, M., Bosetti, F., Solaini, G., Visioli, F., Paolicchi, A., De Caterina, R., 2002. Quenching of intracellular ROS generation as a mechanism for oleate-induced reduction of endothelial activation and early atherogenesis. Thromb. Haemost. 88, 176–375. Massaro, M., Carluccio, M.A., Bonfrate, C., Siculella, L., Lazzerini, G., Bernini, W., Basta, G., De Caterina, R., 1999. The double bond in unsaturated fatty acids is the necessary and sufficient requirement for the inhibition of expression of endothelial leukocyte adhesion molecules through interference with nuclear factor-kappaB activation. Lipids 34 (Suppl), S213–S214. Masuda, J., Ross, R., 1990a. Atherogenesis during low level hypercholesterolemia in the nonhuman primate. I Fatty streak formation. Arteriosclerosis. 10, 178–187. Masuda, J., Ross, R., 1990b. Atherogenesis during low level hypercholesterolemia in the nonhuman primate. II Fatty streak conversion to plaque fibrous. Arteriosclerosis 10, 187–197. Mata, P., Alonso, R., Lopez-Farre, A., Ordovas, J.M., Lahoz, C., Garces, C., Caramelo, C., Codoceo, R., Blazquez, E., de Oya, M., 1996. Effect of dietary fat saturation on LDL oxidation and monocyte adhesion to human endothelial cells in vitro. Arterioscler. Thromb. Vasc. Biol. 16, 1347–1355. Menotti, A., Puddu, P.E., Lanti, M., 2000. Comparison of the Framingham risk function-based coronary chart with risk function from an Italian population study. Eur. Heart. J. 21, 365–370. Pahl, H.L., 1999. Activators and target genes of Rel/NF-kappaB transcription factors. Oncogene 18, 6853–6866. Parthasarathy, S., Khoo, J.C., Miller, E., Barnett, J., Witztum, J.L., Steinberg, D., 1990. Low density lipprotein rich in oleic acid is protected against oxidative modification: implications for dietary prevention of atherosclerosis. Proc. Natl. Acad Sci. 87, 3894–3898. Platz, E.A., Willett, W.C., Colditz, G.A., Rimm, E.B., Spiegelman, D., Giovannucci, E., 2000. Proportion of colon cancer risk that might
SECTION | II Vascular Aspects Including Hypertension
be preventable in a cohort of middle-aged US men. Cancer Causes Control 11, 579–588. Ross, R., 1999. Atherosclerosis – an inflammatory disease. N. Engl. J. Med. 340, 115–126. Ruano, J., Lopez-Miranda, J., Fuentes, F., Moreno, J.A., Bellido, C., Perez-Martinez, P., Lozano, A., Gomez, P., Jimenez, Y., Perez Jimenez, F., 2005. Phenolic content of virgin olive oil improves ischemic reactive hyperemia in hypercholesterolemic patients. J. Am. Coll. Cardiol. 46, 1864–1868. Saija, A., Trombetta, D., Tomaino, A., 1998. In vitro evaluation of the antioxidant activity and biomembrane interaction of the plant phenols oleuropein and hydroxytyrosol. Int. J. Pharm. 166, 123–1333. Stampfer, M.J., Hu, F.B., Manson, J.E., Rimm, E.B., Willett, W.C., 2000. Primary prevention of coronary heart disease in women through diet and lifestyle. N. Engl. J. Med. 343, 16–22. Trichopoulou, A., Costacou, T., Bamia, C., Trichopoulos, D., 2003. Adherence to a Mediterranean diet and survival in a Greek population. N. Engl. J. Med. 348, 2599–2608. Tsimikas, S., Philis-Tsimikas, A., Alexopoulos, S., Sigari, F., Lee, C., Reaven, P.D., 1999. LDL isolated from Greek subjects on a typical diet or from American subjects on an oleate-supplemented diet induces less monocyte chemotaxis and adhesion when exposed to oxidative stress. Arterioscler. Thromb. Vasc. Biol. 19, 122–130. Visioli, F., Bellomo, G., Montedoro, G.F., Galli, C., 1995. Low density lipoprotein oxidation is inhibited in vitro by olive oil constituents. Atherosclerosis 117, 25–32. Visioli, F., Caruso, D., Grande, S., Bosisio, R., Villa, M., Galli, G., Sirtori, C., Galli, C., 2005. Virgin Olive Oil Study (VOLOS): vasoprotective potential of extra virgin olive oil in mildly dyslipidemic patients. Eur. J. Nutr. 44, 121–127. Visioli, F., Galli, C., 2000. Olive oil: more than just oleic acid. Am. J. Clin. Nutr. 72, 853. Visioli, F., Galli, C., Bornet, F., Mattei, A., Patelli, R., Galli, G., Caruso, D., 2000a. Olive oil phenolics are dose-dependently absorbed in humans. FEBS Lett. 468, 159–160. Visioli, F., Galli, C., Plasmati, E., Viappiani, S., Hernandez, A., Colombo, C., Sala, A., 2000b. Olive phenol hydroxytyrosol prevents passive smokinginduced oxidative stress. Circulation 102, 2169–2171. Vissers, M.N., Zock, P.L., Roodenburg, A.J., Leenen, R., Katan, M.B., 2002. Olive oil phenols are absorbed in humans. J. Nutr. 132, 409–417. Weinbrenner, T., Fito, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Albaladejo, M.F., Abanades, S., Schroder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321. Willet, W.C., 1994. Diet and health: what should we eat? Science 264, 532–537.
Chapter 88
Pomace Olive Oil and Endothelial Function Javier S. Perona, Rosana Cabello-Moruno and Valentina Ruiz-Gutiérrez Nutrition and Lipid Metabolism, Instituto de la Grasa, CSIC, Seville, Spain
88.1 INTRODUCTION Atherosclerosis remains as the leading cause of mortality in Western countries and is on track to become the most common cause of disease-related disability and death by the year 2020 (Fuster, 1999). One of the first events occurring during the development of atherosclerosis is endothelial dysfunction, which initiates inflammatory responses by releasing proinflammatory cytokines and chemokines. The activation of the endothelium is therefore manifested as an increase in the expression of specific cytokines and adhesion molecules, such as intracellular adhesion molecule (ICAM-1), vascular cell adhesion molecule (VCAM-1) and E-selectin (Hennig and Toborek, 2001). These molecules cause the attraction and attachment of circulating monocytes to the endothelium, from which they can transmigrate to the subendothelial space. Once within the vascular intima, monocytes can be differentiated into macrophages, which scavenge altered LDL and chylomicron remnants (CMr), becoming foam cells and contributing to the formation of the atherosclerotic plaque (Ross, 1999). The vascular endothelium is an active and dynamic monolayer of cells, which serves as a semipermeable barrier between blood and tissue. Due to this strategic location it is involved in maintaining homeostasis, by sensing changes in hemodynamic forces and signals and responding to them by releasing bioactive substances (Duvall, 2005). Among these substances, heparan sulfate, nitric oxide (NO) and prostacyclin are vasodilators, whereas thromboxane A2 (TBA2), prostaglandin H2 (PGH2) and endothelin 1 are vasoconstrictors (Ross, 1999). Ample evidence is indicating that high concentrations of vascular free oxygen levels are the main mechanism for endothelial dysfunction (Duvall, 2005). Reactive oxygen species (ROS) are produced during normal metabolism or after oxidative processes (Voetsch et al., 2004). The formation of ROS is balanced by a range of antioxidant defenses, but the excess, as in hypercholesterolemia, hypertriacylglycerolemia or Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
high postprandial triacylglycerolemia, can overwhelm these systems, leading to oxidative stress. For this reason, there is increasing interest in investigations regarding the putative effect of dietary components, including antioxidants, anti-inflammatory and hypocholesterolemic substances on the reduction of these risk factors in order to prevent the initiation of the early stages of atherogenesis. Cumulative epidemiological and experimental studies, developed in different countries, constitute a firm and reliable base in support of the modulation of the development of atherosclerosis by dietary components. Among dietary patterns, the so-called ‘Mediterranean diet’ has been revealed as one of the healthiest. The Seven Countries Study (Keys et al., 1986) was the first to show that a diet poor in saturated fatty acids and rich in monounsaturated fatty acids, was related to the lower incidence of cardiovascular disease and cancer in the Mediterranean countries. Recently, the PREDIMED (Prevencion con Dieta Mediterranea (Prevention with Mediterranean Diet)) study has demonstrated that risk factors for cardiovascular disease, including LDL cholesterol and triacylglycerol levels and blood pressure, can be reduced by a Mediterranean-style diet rich in virgin olive oil or nuts (Estruch et al., 2006). Despite the interest on the influence of nutrients and dietary components on endothelial function, the effects of dietary oils have not been sufficiently addressed (Perona et al., 2006). In vitro studies support favorable effects of fish oil on endothelial function, but results from in vivo studies are less consistent and all effects are attributed to n-3 fatty acids (Brown and Hu, 2001). Virgin olive oil has been shown to improve endothelial function in diabetic (Ryan et al., 2000) and hypercholesterolemic patients (Fuentes et al., 2001). We recently incubated endothelial cells with postprandial triacylglycerol-rich lipoproteins (TRL) derived from the intake of meals containing virgin olive oil and virgin olive oil enriched in its minor components (Perona et al., 2004). We found a reduction in the production of prostaglandin E2 (PGE2) and thromboxane
829
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
830
100 a
a a
50 b 25
88.2 CHEMICAL COMPOSITION OF OLIVE OIL
0
Control
400
HOSO
VOD
EVO B
a a
TXB2 (ng/L)
300
200
b b
100
0
Control
HOSO
VOD
EVO
2000
6-keto PGF1 α (ng/L)
Unlike most dietary oils, which are obtained from the plant seeds by means of solvent extraction, and need to be refined before being edible, virgin olive oil is obtained from the whole fruit of Olea europaea L, and only by physical mechanisms. This procedure makes virgin olive oil unique, since some compounds are transferred from the leaves and skin of the fruit to the oil. At the present time, virgin olive oil is obtained after pressure or centrifugation from the olive fruit. The resulting oil is called ‘virgin’ and if the quality is sufficient is commercialized as ‘extra virgin olive oil’. If not, it is refined and blended with extra virgin olive oil to be commercialized as ‘olive oil’, which is poorer in terms of organoleptic characteristics and minor component content. Virgin olive oil can be classified into two fractions from a quantitative point of view. The major fraction constitutes 98–99% of the oil and it is mainly composed of saponifiable glyceridic compounds, largely triacylglycerols. The main fatty acid in this fraction is oleic acid (18:1, n-9), ranging from 60–84% of total fatty acids, while linoleic acid (18:2, n-6), is present in concentrations between 3–21%. Minor components account for the remaining 1–2% of the oil, and despite being a small proportion, they confer important biologic activities. The minor components of virgin olive oil are hydrocarbons, tocopherols, fatty alcohols, 4-methylesterols, sterols, triterpenic dialcohols, polar-colored pigments and phenolic compounds (Ruiz-Gutiérrez et al., 2000). Squalene is a polyunsaturated hydrocarbon that appears at a high concentration, which makes up 60–75% of the unsaponifiable fraction of virgin olive oil (Tiscornia and Evangelisti, 1982), and it is a precursor in the biosynthesis of cholesterol and steroid hormones (Figure 88.1). β-Carotene is also a triterpenic polyunsaturated hydrocarbon that plays an important role as a precursor of vitamin A and, along with lycopene, conferring the yellowish color to the oil. Another important group of compounds within the unsaponifiable fraction is sterols. The main sterol found in virgin olive oil is β-sitosterol (95%) but campesterol, Δ-7 stigmastenol, stigmasterol, spinasterol and avenasterol are also present. Virgin olive oil also contains α, β, γ and Δ-tocopherols, but α-tocopherol typically accounts for more than 85% of total tocopherols. Phenolic compounds constitute the ‘polar fraction’ in virgin olive oil and contribute to its characteristic flavor and taste, and provide
A
75 PGE2 (ng/L)
B2 (TXB2) after the incubation with TRL obtained after the intake of the enriched oil, compared with virgin olive oil. The results suggested that minor components from virgin olive oil that are transported postprandially in TRL may have favorable effects on endothelial function by improving the balance between vasoprotective and prothrombotic factors released by endothelial cells.
SECTION | II Vascular Aspects Including Hypertension
C a
ab
ab
1000
0
b
Control
HOSO
VOD
EVO
FIGURE 88.1 Effect of the unsaponifiable fraction of olive oil in triacylglycerol-rich lipoproteins (TRL) on eicosanoid production. Human umbilical vein endothelial cells (HUVEC) were incubated for (24 h) with TRL obtained 2 h after the ingestion of high-oleic sunflower (HOSO), virgin olive (VOO) or enriched-virgin olive (EVO) oils. Eicosanoids released to the medium were determined by enzyme-linked immunoassays. Results corresponding to PGE2, TXB2 and 6-keto-PGF1α are shown in panels A, B and C respectively. a: p ⬍ 0.05, vs. Control; b: p ⬍ 0.05, vs. HOSO; c: p ⬍ 0.05, vs. VOO.
the resistance of virgin olive oil to oxidative rancidity (Boskov, 1996). Among these compounds, oleuropein itself and its derivatives, tyrosol and hydroxytyrosol, have been reported to have a protective role against LDL oxidation in vitro, equivalent to vitamin E (Visioli and Galli, 1998).
831
CHAPTER | 88 Pomace Olive Oil and Endothelial Function
Recent interventional studies have also shown that administration of increasing doses of phenolic compounds in virgin olive oil have reported reduced oxidative status in healthy (Weinbrenner et al., 2004) and dyslipemic subjects (Visioli et al., 2005). However, these compounds are only present in virgin olive oil, as they are lost in other olive oils, like pomace olive oil, during the refining process due to their hydrosolubility.
88.3 POMACE OLIVE OIL Traditionally, after molturation of the olive fruit, olive oil was extracted by simple pressure, but today, the centrifugation or continuous method prevails. There are two types of centrifugation systems: three and two phases. The threephase system yields a liquid fatty phase (oil), a liquid aqueous phase and a solid phase (pomace). In contrast, after the two-phase system only the oil and a semi-aqueous phase are obtained. Pomace olive oil is obtained from this latter phase by means of solvent extraction or centrifugation, in which case it is called second-centrifugation pomace olive oil (Alba et al., 1996). Before commercialization, the raw pomace olive oil needs to be refined and blended with virgin olive oil to improve its organoleptic characteristics. This extraction system allows the enrichment of the oil with components from the skin of the fruit and leaves of the tree, which are of great interest due to their biological activity, and are found only in traces in virgin olive oil (Perez-Camino and Cert, 1999). In contrast it does not modify the glyceridic composition, thus the fatty acid and triacylglycerol composition of the resulting oil is similar to that of virgin olive oil. Due to this extraction method, second-centrifugation pomace olive oil is enriched in components from the unsaponifiable fraction, which are lipophilic, but are devoid of the hydrophilic phenolic compounds. Among the unsaponifiable components, hydrocarbons, sterols and tocopherols are also present in pomace olive oil, although the concentration is usually higher than in virgin olive oil (Table 88.1). However, some other minor components are absent in virgin olive oil, whilst they are present in relevant concentrations in pomace olive oil, like fatty alcohols and the triterpenic compounds oleanolic and maslinic acids (Figure 88.2). Recent studies in cell and animal experimental models have shown the benefits of pomace olive oil consumption, as well as the properties of some of its minor components. It has been demonstrated that regular phytosterol consumption has a hypocholesterolemic effect as these substances compete with cholesterol for the same absorption mechanisms, resulting in reduced absorption of cholesterol, and eventually in LDL-cholesterol plasmatic concentrations (Ellegård et al., 2007). Additionally it has been suggested that β-sitosterol may have an anti-inflammatory effect (de la Puerta et al., 2000), as well as certain protective activity against LDL oxidation in vitro (Andrikopoulos et al., 2002). Pomace
TABLE 88.1 Minor component composition of secondcentrifugation pomace olive oil. Sub-fraction
Component
Concentration (mg kg⫺1)
Hydrocarbons
Squalene β-Carotene Polycyclic aromatic hydrocarbons
2539 0.5 Traces
Sterols
β-Sytosterol Campesterol Δ7-Stigmasterol Brassicasterol
2137 54 20 7
Tocopherols
α-Tocopherol β-Tocopherol γ-Tocopherol
792 88 102
Triterpenic compounds
Erythrodiol ⫹ uvaol Oleanolic acid Maslinic acid
500 60–400 100–300
Waxes
Fatty alcohols
3484
Others
Flavor components
Traces
The minor components of second-centrifugation pomace olive oil are similar to those of virgin olive oil. An exception to this is the presence of triterpenic compounds. The concentration of these components is always higher in second-centrifugation pomace olive oil except for squalene.
olive oil also has important tocopherol concentrations with vitamin E activity. This vitamin is the main lipophilic antioxidant in the human body and the incorporation of tocopherols into membranes and lipoproteins prevents their oxidation (Andrikopoulos et al., 2002). Triterpenic compounds, which are characteristic of pomace olive oil, have been observed to present ant-inflammatory activity in models of acute and chronic inflammation (de la Puerta et al., 2000). Additionally, they present certain antioxidant activity as shown in LDL (Andrikopoulos et al., 2002) and hepatic microsomes (Perona et al., 2005). Finally, clinical studies have described that waxes, which are esters of long-chain fatty alcohols and acids, reduce the concentration of LDL cholesterol in humans (Hargrove et al., 2004). It has been suggested that this effect might also be related to competitive absorption with cholesterol in the intestine.
88.4 EFFECTS OF THE COMPONENTS OF POMACE OLIVE OIL ON ENDOTHELIAL FUNCTION Apart from these striking observations, some minor components of pomace olive oil have been shown to exert direct
832
IL-6 (Pg/mL)
SECTION | II Vascular Aspects Including Hypertension
50000
TABLE 88.2 Key features of endothelial dysfunction.
40000
1. The vascular endothelium is an active and dynamic monolayer of cells, which serves as a semipermeable barrier between blood and tissue
30000
2. The endothelium has a major function in thrombotic and coagulant activities, synthesizing several molecules that are released in response to different stimuli
20000 **
**
10000
10 μm
100 μm
(A)
50 μm
0
25 μm
**
TNF-a (Pg/mL)
2000
1000 **
10 μm
100 μm
50 μm
25 μm
**
(B)
4. Risk factors for atherosclerosis, such as smoking, hypercholesterolemia, hypertension, diabetes and hyperhomocysteinemia, predispose to endothelial dysfunction 5. When dysfunctional, the endothelium increases flow disturbances due to improper vasoreactivity and initiates inflammatory responses
*
0
3. Endothelial dysfunction occurs early in the development of atherosclerosis, even before the formation of the plaque
Unstimulated cells
LPS + (10 μg/mL)
LPS + Maslinic acid
LPS + Dexamethasone
FIGURE 88.2 Effect of maslinic acid on IL-6 (A) and TNFα (B) production by peripheral blood mononuclear cells (2 ⫻ 106 cell mL⫺1). Cells were stimulated with lipopolysaccharide (10 μg mL⫺1) in the presence of different concentrations of the compounds for 24 h. Each value represents means ⫾ SEM for six experiments. Results are expressed as percentage of cytokine liberation versus control using dexamethasone (10 μg mL⫺1). Statistically significant differences *: p ⬍ 0.05; *: p ⬍ 0.01 vs. control.
effects on endothelial function. However, the number of studies focused on the effect of this dietary oil derived from olive oil extraction and/or its components is still very scarce. A study carried out by Rodriguez-Rodriguez et al. (2007) has demonstrated that chronic consumption of diets rich in second-centrifugation pomace olive oil improves endothelial dysfunction in the aorta of spontaneously hypertensive rats (SHR) (Table 88.2). The results provided the first evidence of the effects of pomace olive oil on endothelial function in hypertensive animals. These authors indicated that the effect was mediated by the enhancement of the expression of endothelial nitric oxide synthase (eNOS). Actually, previous studies had already pointed out an important vasorelaxant activity of triterpenic compounds contained in pomace olive oil. Oleanolic and maslinic acids, as well as erythrodiol and uvaol induced vasorelaxation in vitro in aortic rings from normotensive (Rodríguez-Rodríguez et al., 2004) and hypertensive
6. The determination of endothelial markers in blood is the most common method for the assessment of endothelial function. However, the most promising non-invasive method is the measurement of the flow-mediated dilation in the brachial artery by Doppler ultrasonography 7. Minor components from virgin olive have favorable effects on endothelial function
(Rodriguez-Rodriguez et al., 2006) rats. These data evidenced the potential therapeutic effect of these compounds as protectors of the cardiovascular system. In these studies, the authors observed that the vasorelaxant effect was lost when the aortic rings were incubated with the eNOS inhibitor L-NAME, which pointed towards the involvement of nitric oxide in the mechanism. More recently, Martínez-González et al. (2008) incubated oleanolic acid and erythrodiol with human coronary smooth muscle cells (SMC). Both triterponids induced prostaglandin I2 (PGI2) release, an effect that was prevented by celecoxib, specific inhibitor of cyclooxygenase-2 (COX-2). Oleanolic acid also induced an early phosphorylation of p38 MAPK and p42/44 MAPK but not c-Jun N-terminal kinase1 (JNK-1). SB203580 (p38MAPK inhibitor) and U0126 (MAPK kinase1/2 inhibitor) abrogated the up-regulation of COX-2 and PGI2 release induced by oleanolic acid. The results showed that oleanolic acid contributes to vascular homeostasis by inducing PGI2 release in a COX-2-dependent manner. Maslinic acid is another triterpenic compound present in pomace olive oil, which has been associated with anti-inflammatory properties and, therefore, might be helpful in the restoration of endothelial function. In peritoneal murine macrophages, maslinic acid significantly inhibited the enhanced production of nitric oxide induced by lipopolysaccharide and reduced the secretion of the inflammatory cytokines interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) (Marquez-Martín et al., 2006).
833
CHAPTER | 88 Pomace Olive Oil and Endothelial Function
For the moment no clinical studies have used pomace olive oil as the dietary fat source. The only interventional study in which pomace olive oil has been included was carried out in our laboratory and was recently published (Cabello-Moruno et al., 2007). We administered pomace olive oil and refined olive oil to nine healthy young males and we determined the composition of TRL postprandially. The main difference between the oils was the minor component composition. Serum triacylglycerols were not affected by the administration of either oil but the triacylglycerol concentration in the postprandial lipoproteins and the size of the particles was higher 2 hours after pomace olive oil ingestion. However, lipoproteins originated after the intake of pomace olive oil were cleared from plasma at a higher rate. In view of these results, there is a clear need for long-term interventional studies to assess the potential benefits of pomace olive oil.
88.5 CONCLUSIONS Virgin olive oil, as the main fat source of the Mediterranean diet has been proposed as one of the healthiest dietary oils, since there is increasing evidence demonstrating beneficial effects on the pathophysiology of atherosclerosis. Certainly, a number of experimental studies have revealed that virgin olive oil consumption can have favorable effects on hypercholesterolemia and hypertension. Traditionally the beneficial effect of virgin olive oil was ascribed to its fatty acid composition and the high monounsaturated fatty acid content. However, it has been demonstrated that minor components also have a very relevant role. In this regard, second-centrifugation pomace olive oil is particularly rich in some minor components with an elevated biological activity. Despite being devoid of polyphenols, the concentration of sterols and tocopherols is higher in pomace olive oil. In addition, pomace olive oil contains relevant amounts of triterpenic compounds, like the acids oleanolic and maslinic and the alcohols erythrodiol and uvaol, which have been related with the improvement in endothelial dysfunction in hypertensive animals. Although there is still a lack of clinical studies that should be carried out, pomace olive oil is emerging as a healthy dietary oil, with important implications in endothelial dysfunction, one of the first events occurring during the development of atherosclerosis.
SUMMARY POINTS ●
●
●
Endothelial dysfunction is one of the first events involved in the development of atherosclerosis. Minor components from dietary oils can modulate or even improve endothelial function. Some of the beneficial effects on the pathophysiology of atherosclerosis of virgin olive oil have been attributed to its minor components.
●
●
●
●
●
Minor components of virgin olive oil are comprised of hydrocarbons, polyphenols, tocopherols, sterols, triterpenoids and trace compounds, which have antioxidant, anti-inflammatory or hypocholesterolemic effects. Second-centrifugation pomace olive oil is derived from the extraction process of virgin olive oil, and is particularly rich in some of these minor components. Chronic consumption of diets rich in second-centrifugation pomace olive oil improves endothelial dysfunction in the aorta of spontaneously hypertensive rats (SHR). Triterepenic acids from pomace olive oil can induce vasorelaxation in vitro in aortic rings from normotensive and hypertensive rats. Atherogenic postprandial lipoproteins, originated after the intake of pomace olive oil, are cleared from plasma at a higher rate.
REFERENCES Alba, J., Hidolgo, F., Ruiz-Gómez, M.A., Martinez, F., Moyano, M.J., Cert, A., Pérez-Camino, M.C, Ruiz-Méndez, M.V., 1996. Características de los aceites de oliva de primera y segunda centrifugación. Grasas Aceites 47, 163–181. Andrikopoulos, N.K., Kaliora, A.C., Assimopoulou, A.N., Papageorgiou, V.P., 2002. Inhibitory activity of minor polyphenolic and nonpolyphenolic constituents of olive oil against in vitro low-density lipoprotein oxidation. J. Med. Food. 5, 1–7. Boskov, D., 1996. Olive oil chemistry and technology. AOCS Press, Illinois, pp. 115–117. Brown, A.A., Hu, F.B., 2001. Dietary modulation of endothelial function: implications for cardiovascular disease. Am. J. Clin Nutr. 73, 673–686. Cabello-Moruno, R., Perona, J.S., Osada, J., Garcia, M., Ruiz-Gutierrez, V., 2007. Modifications in postprandial triacylglycerol-rich lipoprotein composition and size after the intake of pomace olive oil. J. Am. Coll. Nutr. 26, 24–31. de la Puerta, R., Martinez-Dominguez, E., Ruiz-Gutierrez, V., 2000. Effect of minor components of virgin olive oil on topical antiinflammatory assays. Z. Naturforsch. [C] 55, 814–819. Duvall, W.L., 2005. Endothelial dysfunction and antioxidants. Mt. Sinai. J. Med. 72, 71–80. Ellegård, L.H., Andersson, S.W., Normén, A.L., Andersson, H.A., 2007. Dietary plant sterols and cholesterol metabolism. Nutr. Rev. 65, 39–45. Estruch, R., Martínez-González, M.A., Corella, D., Salas-Salvadó, J., Ruiz-Gutiérrez, V., Covas, M.I., Fiol, M., Gómez-Gracia, E., LópezSabater, M.C., Vinyoles, E., Arós, F., Conde, M., Lahoz, C., Lapetra, J., Sáez, G., Ros, E. PREDIMED Study Investigators, 2006. Effects of a Mediterranean-style diet on cardiovascular risk factors: a randomized trial. Ann. Intern. Med. 145, 1–11. Fuentes, F., López-Miranda, J., Sánchez, E., Sánchez, F., Paez, J., PazRojas, E., Marín, C., Gómez, P., Jimenez-Perepérez, J., Ordovás, J.M., Pérez-Jiménez, F., 2001. Mediterranean and low-fat diets improve endothelial function in hypercholesterolemic men. Ann. Intern. Med. 134, 1115–1119. Fuster, V., 1999. Epidemic of cardiovascular disease and stroke: the three main challenges. Presented at the 71st scientific sessions of the American Heart Association, Dallas, Texas. Circulation 99, 1132–1137.
834
Hargrove, J.L., Greenspan, P., Hartle, D.K., 2004. Nutritional significance and metabolism of very long chain fatty alcohols and acids from dietary waxes. Exp. Biol. Med. (Maywood) 229, 215–226. Hennig, B., Toborek, M., 2001. Nutrition and endothelial cell function: implications in atherosclerosis. Nutr. Res. 21, 279–293. Keys, A., Menotti, A., Karvonen, M.J., et al., 1986. The diet and 15-year death rate in the seven countries study. Am. J. Epidemiol. 124, 903–915. Marquez-Martín, A., de la Puerta Vázquez, R., Fernández-Arche, A., Ruiz-Gutiérrez, V., 2006. Supressive effect of maslinic acid from pomace olive oil on oxidative stress and cytokine production in stimulated murine macrophages. Free Radic. Res. 40, 295–302. Martínez-González, J., Rodríguez-Rodríguez, R., González-Díez, M., Rodríguez, C., Herrera, M.D., Ruiz-Gutierrez, V., Badimon, L., 2008. Oleanolic acid induces prostacyclin release in human vascular smooth muscle cells through a cyclooxygenase-2-dependent mechanism. J. Nutr. 138, 443–448. Perez-Camino, M.C., Cert, A., 1999. Quantitative determination of hydroxy pentacyclic triterpene acids in vegetable oils. J. Agric. Food. Chem. 47, 1558–1562. Perona, J.S., Martínez-González, J., Sanchez-Domínguez, J.M., Badimon, L., Ruiz-Gutierrez, V., 2004. The unsaponifiable fraction of virgin olive oil in chylomicrons from men improves the balance between vasoprotective and prothrombotic factors released by endothelial cells. J. Nutr. 134, 3284–3289. Perona, J.S., Arcemis, C., Ruiz-Gutierrez, V., Catalá, A., 2005. Effect of dietary high-oleic-acid oils that are rich in antioxidants on microsomal lipid peroxidation in rats. J. Agric. Food Chem. 53, 730–735. Perona, J.S., Cabello-Moruno, R., Ruiz-Gutierrez, V., 2006. The role of virgin olive oil components in the modulation of endothelial function. J. Nutr. Biochem. 17, 429–445. Rodriguez-Rodriguez, R., Herrera, M.D., de Sotomayor, M.A., RuizGutierrez, V., 2007. Pomace olive oil improves endothelial function in spontaneously hypertensive rats by increasing endothelial nitric oxide synthase expression. Am. J. Hypertens. 20, 728–734.
SECTION | II Vascular Aspects Including Hypertension
Rodríguez-Rodríguez, R., Herrera, M.D., Perona, J.S., Ruiz-Gutiérrez, V., 2004. Potential vasorelaxant effects of oleanolic acid and erythrodiol, two triterpenoids contained in ‘orujo’ olive oil, on rat aorta. Br. J. Nutr. 92, 635–642. Rodriguez-Rodriguez, R., Perona, J.S., Herrera, M.D., Ruiz-Gutierrez, V., 2006. Triterpenic compounds from “orujo” olive oil elicit vasorelaxation in aorta from spontaneously hypertensive rats. J. Agric. Food. Chem. 54, 2096–2102. Ross, R., 1999. Atherosclerosis – an inflammatory disease. N. Engl. J. Med. 14, 115–126. Ruiz-Gutiérrez, V., De la Puerta, R., Perona, J.S., 2000. Beneficial effects of olive oil on health. In: Recent Research and Development in Nutrition. Researchsignpost, Trivandrum, India, pp. 173–197. Ryan, M., McInerney, D., Owens, D., Collins, P., Johnson, A., Tomkin, G.H., 2000. Diabetes and the Mediterranean diet: a beneficial effect of oleic acid on insulin sensitivity, adipocyte glucose transport and endotheliumdependent vasoreactivity. QJM 93, 85–91. Tiscornia, E.F.M., Evangelisti, F., 1982. Chemical composition of olive oil and its variations induced by refining. Riv. Ital. Sostanza. Grasse. 59, 519–556. Visioli, F., Galli, C., 1998. The effect of minor constituents of olive oil on cardiovascular disease: new findings. Nutr. Rev. 56, 142–147. Visioli, F., Caruso, D., Grande, S., Bosisio, R., Villa, M., Galli, G., Sirtori, C., Galli, C., 2005. Virgin olive oil study (VOLOS): vasoprotective potential of extra virgin olive oil in mildly dyslipidemic patients. Eur. J. Nutr. 44, 121–127. Voetsch, B., Jin, R.C., Loscalzo, J., 2004. Nitric oxide insufficiency and atherothrombosis. Histochem. Cell. Biol. 122, 353–367. Weinbrenner, T., Fito, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Albaladejo, M.F., Abanades, S., Schroder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321.
Chapter 89
Medicinal Value of Combination of Cholinergic and Calcium Antagonist Constituents in Olives Anwarul Hassan Gilani1 and Arif-ullah Khan2 1 2
Natural Product Research Division, Department of Biological and Biomedical Sciences, Aga Khan University, Karachi, Pakistan Institute of Pharmaceutical Sciences, Kohat University of Science and Technology, Kohat, Pakistan
89.1 INTRODUCTION Olea europea commonly known as ‘olive’ and locally as ‘zaitoon’ belongs to the family Oleaceae. It is a slowgrowing evergreen tree, up to 18 meters tall with thornless branches, elliptical leaves and fragrant flowers. The plant is native to parts of Asia Minor, Europe and cultivated largely on the shores of the Mediterranean as well as in California, Australia and other parts of the world. In Pakistan it is cultivated in Baluchistan and Punjab. Its parts used for therapeutic purposes are fruit, oil, bark and gum (Baquar, 1989). Olives have multiple medicinal uses including disorders of gastrointestinal and cardiovascular systems. Olive fruit and leaves are indicated in arrhythmia, atherosclerosis, cardiopathy, colic (spasm), diarrhea, fever, gout, headache, hepatosis, hypercholesterolemia, hyperglycemia and hypertension. Olive oil is used in traditional medicine as a cardioprotective, gastroprotective, enteroprotective, effective in cancer, constipation, diabetes, rheumatism and is also used for edible purposes (Duke et al., 2002). Phytochemical studies revealed that olives contain multiple chemicals, such as aescultin, alpha-tocopherol, apigenin, arabinose, betacarotene, caffeic acid, catechin, choline, cinchonidine, cinchonine, elenolide, erythrodiol, esculin, estrone, fat, fiber, glucoside, iron, linoleic acid, luteolin, mannitol, myristic acid, oleaniolic acid, oleoside, olivine, oleuropeic acid, oleuropein, pectin, palmitic acid, quercetin, quinone, rhamnose, rutin, squalene, tyrosol, verbascoside, tannins, saponins and secoiridoids (Gilani et al., 2006a). Olive oil has been reported to reduce the incidence of coronary heart diseases and acts as an antioxidant (Keys, 1987). It is considered as one of the dietary constituents that contribute to the cardioprotective effect. Clinical intervention reported that olive oil decreases cardiovascular risk factors. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
Diets rich in olive oil have been shown to be more effective in lowering total cholesterol and low-density-lipoprotein cholesterol. The therapeutic properties of olive oil are often attributed to its high levels of monosaturated fatty acids (Stark and Madar, 2002). It has been shown to have beneficial effects on blood pressure and to reduce the systolic blood pressure of hypertensive patients (Covas, 2007). Glycero-alcoholic extracts of its shoots, leaves and flowers showed hypocholesterolemic effects (De Pasquale et al., 1991). The aqueous extracts of olive leaves inhibited angiotensin-converting enzyme (Hansen et al., 1996). Its leaves are also reported as antiarrhythmic (Occhiuto et al., 1990), antihypertensive (Lasserre et al., 1983) and vasodilatory (Zarzuelo et al., 1991). Oleuropein, an active principle of olive enhances nitric oxide production by mouse macrophages (Visioli et al., 1998). We have recently reported the presence of cholinergic and Ca⫹⫹ antagonist constituents in olive fruit, which provides a pharmacological basis for its medicinal use in abdominal spasm, diarrhea, constipation and hypertension (Gilani et al., 2005a, 2006a) and this chapter focuses on how the presence of these constituents can help to explain the medicinal uses of olives in gastrointestinal and cardiovascular disorders.
89.2 GASTROINTESTINAL AND CARDIOVASCULAR DISEASES The gastrointestinal tract is in a state of continuous contractile and secretory activity. Coordinated gastrointestinal motility depends on neural and hormonal control of sphincter, longitudinal and circular smooth muscle contraction and relaxation. Changes in tension of smooth muscle wall control regional intraluminal pressure, which in turn
835
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
836
regulates the movement of intraluminal contents through the bowel. Myoelectric activity, including slow waves and spike potentials, coordinates regional intestinal smooth muscle contraction by controlling their frequency and by electric linking of neighboring smooth muscle cells. The slow wave is a cyclic change in membrane potential that occurs in gut from stomach to small intestine and colon. Ca⫹⫹ influx during spike potentials, superimposed upon the slow waves results in gut contraction. Enteric neurons contain different excitatory neurotransmitters, such as acetylcholine, serotonin and cholecystokinin, which are responsible for regulating bowel motility. Peptides, such as gastrin and motilin, which are released from the gut mucosa into the blood after eating, act as hormones and cause gut activation (Meyer, 1987). Gut motility can be influenced by multiple factors including psychic, emotional, individual general health and quality or bulk of the intestinal contents. The bulk of the intestinal contents stimulates the sensory receptors of intestinal mucosa and promotes the propulsive peristaltic activity of intestine. Intestinal motility is also altered by amyloidosis, hyperthyroidism, primary intestinal pseudo-obstruction, carcinomas, muscular dystrophy, jejunal diverticulosis, irritable bowel syndrome, diabetes mellitus, hemorrhoids, fissures, paraplegia, inflammations, infections and certain drugs, e.g., antacids, iron preparations, tricyclic antidepressants, opiates and laxatives (William and Snape, 1992). The abnormality caused in the gastrointestinal motility may result in various major disorders, such as abdominal spasm, diarrhea and constipation. Abdominal colic (spasm) is an involuntary movement of gut smooth muscle, accompanied by severe pain and interference with function. It is usually caused by inflammation (appendicitis, diverticulitis, colitis, diarrhea), by stretching or distention of an organ (obstruction of the intestine, blockage of a bile duct by gallstones, swelling of liver) or by ischemic colitis. Acute abdominal pain represents 5–10% of all emergency department visits and requires careful history taking, thorough evaluation of symptoms, complete physical examination and judicious use of laboratory tests to simplify the evaluation of this complaint (Thomas et al., 1996). Diarrhea is an increase in stool mass, frequency and fluidity. It results from more rapid transit of the intestinal contents through either the small intestine or the colon. The mechanisms and mediators involved in diarrhea are multiple, including extracellular mediators (hormones, neurotransmitters, prostaglandins, enterotoxins), intracellular mediators (Ca⫹⫹, cyclic adenosine monophosphate, cyclic guanosine monophosphate, calmodulin, phospholipids), intramural blood flow and intestinal motility (Fine et al., 1989). Diarrhea is a common problem in primary care and emergency department settings especially for children younger than five years. Every year, approximately 500 million children younger than five years have at least one diarrheal illness. Children younger than three years
SECTION | II Vascular Aspects Including Hypertension
have about two episodes of diarrhea per year. Older children and adults have about one episode of diarrhea a year. Approximately 400 children die from diarrhea in the USA each year (Liebelt, 1998). Many developing countries do not have the resources to properly treat diarrhea, leading to a disproportionately high mortality rate. About 2 million children worldwide are estimated to die each year due to diarrheal illness (Liebelt, 1998). The diarrhea may result in serious complications. Fluid and electrolyte (Na⫹, K⫹, Mg⫹⫹, Cl⫺) loss with consequent dehydration and even vascular collapse may occur. Collapse may develop rapidly in patients who are very young or old, debilitated or have severe diarrhea. HCO3⫺ loss may cause metabolic acidosis. Hypokalemia may occur in severe or chronic diarrhea or if the stools contain excess mucus. Hypomagnesemia after prolonged diarrhea may cause tetany (Fine et al., 1989). Constipation is a common symptom that may be idiopathic or secondary to a disease. It is defined as the infrequent passage of hard stools, resulting from slow transit of the intestinal contents through either the small intestine or the colon. Constipation is divided with considerable overlap into issues of stool consistency (hard, painful stools) and issues of defecatory behavior (infrequency, difficulty in evacuation, straining during defecation). Constipation is the end effect of several factors such as diabetes mellitus, hypothyroidism, poor diet, lack of exercise, motility abnormalities, anatomic defects along with the patient expectations and psychological factors, recto-anal problems and due to certain gut-relaxing drugs. Chronic constipation is a growing problem in the modern age and self-medication is common, partly because of a public obsession with once daily bowel movements without realizing that defecation even after two days is normal so long as it is free from strain (Devroede, 1989). The use of chemical drugs in chronic constipation is discouraged; rather the best approach is to manage with lifestyle changes and appropriate diet including fruits and vegetables (Gilani, 1992). The Mediterranean diet rich in olives or olive oil is not only considered best for cardiovascular disorders but also useful for normal functioning of gut. The beauty of nature is that fruits like olives are considered useful not only for constipation, but also for diarrhea, which is the opposite disease state of the gut function to that of constipation. Hypertension is one of the most common and rapidly spreading cardiovascular diseases, which is a major cause of morbidity and mortality in mankind and an important risk factor for the development of other cardiovascular diseases (MacMahon et al., 1990). Hypertension is an important public health challenge worldwide. Prevention, detection, treatment, and control of this condition should receive high priority. The estimated total number of adults with hypertension in 2000 was 972 million; 333 million in economically developed countries and 639 million in economically developing countries. The number of adults with hypertension in 2025 was predicted to increase by about 60% to a
CHAPTER | 89 Medicinal Value of Combination of Cholinergic and Calcium Antagonist Constituents in Olives
total of 1.56 billion. It is generally anticipated that the prevalence of hypertension will increase if no specific measures with respect to prevention or improved treatment are taken. The pathophysiologic basis for over 90% of the cases of the disease remains unexplained, and therefore, the condition is called primary or essential hypertension. The heritability of essential hypertension is estimated to be about 30%. The remaining 5–10% may be due to a number of known causes, including renal artery stenosis, aortic coarctation, Cushing’s syndrome and pheochromocytoma, known as secondary hypertension (DiPalma, 1990). Hypertension is a serious affliction, and although often asymptomatic, over time it leads to a variety of health problems including premature sickness, stroke, congestive heart failure, myocardial infarction, peripheral vascular disease, retinopathy, dementia, renal dysfunction, cerebrovascular damage, disability and death in the adult population (Flack et al., 2003). The hypertension-related morbidity and mortality are related directly to the level of blood pressure and the incidence significantly decreases when hypertension is diagnosed early and is properly treated, whereas untreated hypertension is known as the ‘silent killer’. The risk of developing a cardiovascular complication is higher when the individual combines hypertension with other risk factors, such as hypercholesterolemia, smoking, diabetes and a family history of cardiovascular diseases (Mancia et al., 2004).
89.3 PHARMACOLOGY OF CHOLINERGIC AND CALCIUM CHANNEL BLOCKERS Acetylcholine is an important physiological neurotransmitter released by the parasympathetic nervous system, which achieves its effect in different organ systems through the activation of muscarinic receptors. The drugs with actions like acetylcholine are called cholinomimetics or cholinergics. Muscarinic receptors are responsible for an array of responses, linked to the parasympathetic system. There are five major sub-classes of muscarinic receptors, i.e., M1, M2, M3, M4 and M5 (Caulfield, 1993). The M4 and M5 subtypes were only recently discovered and are located in the brain. Muscarinic receptors are linked with the super family of G-protein-coupled receptors, whose overall mechanism of activation is as follows: (i) the receptor oscillates spontaneously between an active and an inactive conformation; (ii) agonist binding strongly favors the active conformation of the receptor leading to the exchange of guanosine triphosphate (GTP) with the guanosine diphosphate (GDP) on the α subunit of the G-protein; (iii) binding of GTP causes dissociation of the G-protein from the activated receptor; (iv) heterotrimeric G-protein further dissociates, the α subunit being responsible for the activation of a particular second messenger system, and the βγ dimer often exhibiting independent actions; (v) hydrolysis of GTP on the α subunit to GDP by an intrinsic GTPase activity leads to its inactivation;
837
(vi) the inactivated α subunit then recombines with the βγ subunits, which in turn re-associates with the receptor. The M1, M3 and M5 receptors preferentially couple via Gq guanine nucleotide-binding proteins leading to phosphatidyl inositol breakdown, while the M2 and M4 subtypes act via Gi and Gk to inhibit adenylyl cyclase. The cholinergic drugs activate the muscarinic receptors in different systems producing various effects. For example, in the gastrointestinal system they increase gut motility, in airways they mediate bronchial constriction, in the heart they cause bradycardia, in blood vessels they lead to vasodilatation and in exocrine glands they stimulate gastric acid secretion, salivation and lacrimation (Brown and Taylor, 2001). Calcium channel blockers also known as calcium antagonists are a structurally and functionally heterogeneous group of compounds that interfere with inward movement of Ca⫹⫹ across the cell membranes through voltage-dependent calcium channels. Ca⫹⫹ antagonists are classified as dihydropyridines or non-dihydropyridines. Dihydropyridines include amlodipine, felodipine, nicardipine, and nifedipine, whereas non-dihydropyridines comprise agents, such as diltiazem and verapamil (Farre et al., 1991). Ca⫹⫹ channels are membrane-spanning, funnel-shaped glycoproteins that function like ion-selective valves. They form a water-filled pore that opens or closes to permit Ca⫹⫹ to move in the direction of its electrochemical concentration gradient. Until now, four subtypes of calcium channels have been identified: L (being large in conductance, present in smooth muscle); T (being transient in duration of opening; i.e., the channel opens only briefly during the cardiac cycle); N (being neuronal in distribution); and P (being particularly prominent in cerebellar Perkinje cells). Most excitable cells have many types of Ca⫹⫹ channels, but the L-type is prominent. These channels are distinguished by their electrophysiological, pharmacological and ligand-binding characteristics. Ca⫹⫹ channels also exhibit distinct activation, inactivation and conductance properties. The isolated calcium channel appears to consist of five subunits (α1, α2, β, γ and δ), although the precise stoichiometry and obligate function of all the subunits remain to be established (Wray et al., 2005). Voltage-dependent calcium channels play a specific role in the excitation–contraction– relaxation cycle. Under resting conditions, when the intracellular Ca⫹⫹ concentration is low (⬍10⫺7 M), the regulatory proteins (troponin I, T and C and acto-myosin) prevent interaction of actin and myosin filaments with each other and the muscle is relaxed. When the intracellular Ca⫹⫹ concentration increases (⬎10⫺7 M) from the influx of Ca⫹⫹ through calcium channels and the release of Ca⫹⫹ from internal stores, Ca⫹⫹ occupies specific Ca⫹⫹-binding sites on troponin C (cardiac and skeletal muscle) and calmodulin (vascular smooth muscle). These then interact with other regulatory proteins and enzymes (e.g., troponin I) in cardiac and skeletal muscles and myosin light-chain kinase in smooth muscles, facilitating cross-bridge formation
838
SECTION | II Vascular Aspects Including Hypertension
between actin and myosin and resulting in muscle contraction. When calcium channels close and calcium ATPase system pumps Ca⫹⫹ out of the cytosol into intracellular stores and to the extracellular space, the cytosolic Ca⫹⫹ concentration decreases to its resting level, Ca⫹⫹ dissociates from Ca⫹⫹-binding proteins, activation of contractile protein is reversed, actin dissociates from myosin, and muscle relaxation occurs. Thus the agents affecting calcium channels also affect muscle contraction and relaxation. Hence, Ca⫹⫹ antagonists, which decrease the intracellular free Ca⫹⫹ by virtue of blocking calcium channels induces muscle relaxation (Wingard et al., 1991). Ca⫹⫹ channel blockers are widely used for cardiovascular disorders, such as hypertension, arrhythmias, angina pectoris, myocardial infarction and congestive cardiac failure. The potential therapeutic spectrum of these drugs is rapidly increasing. More recently, Ca⫹⫹ antagonists have been found to be effective in disorders of gut hypermotility, asthma, epilepsy, migraine, Alzheimer’s disease, depression and urinary incontinence (Stoelting, 1995).
89.4 NATURALLY OCCURRING CHOLINERGIC AND CAⴙⴙ ANTAGONIST COMBINATION IN OLIVES Based on the medicinal use of olives in gastrointestinal motility disorders, the aqueous-methanolic extract of olive fuit (OeF.Cr) was studied in spontaneously contracting isolated rabbit jejunum, where it inhibited the spontaneous contractions (Figures 89.1 and 89.2), thus showing an antispasmodic effect. In order to elucidate the possible mode of spasmolytic effect, a high dose of K⫹ (80 mM) was used to obtain a sustained contraction, which was relaxed by the olive extract (Figure 89.2). High K⫹ (⬎30 mM) is known to cause smooth muscle contractions through opening of voltage-dependent Ca⫹⫹ channels, thus allowing
influx of extracellular Ca⫹⫹ causing a contractile effect (Bolton, 1979) and a substance causing inhibition of high K⫹-induced contraction is considered a Ca⫹⫹ influx inhibitor (Godfraind et al., 1986). The presence of Ca⫹⫹ antagonist constituent(s) was further confirmed when the olive extract, like verapamil, a standard calcium channel blocker (Fleckenstein, 1977) shifted the Ca⫹⫹ concentration– response curves, constructed in Ca⫹⫹ free medium, to the right, accompanied with suppression of the maximum contractile effect (Figure 89.3). The presence of Ca⫹⫹ antagonist(s) explains the olive folkloric use in abdominal colic and diarrhea, as discussed earlier; such agents are well known to be effective in hyperactive gut disorders (Findling et al., 1996). Olives have also been used to relieve constipation, which means that they also contain some stimulant constituent(s), masked by the dominant spasmolytic (Ca⫹⫹ antagonist) component. Pretreatment of the tissue with atropine potentiated the inhibitory effect of the olive extract both against spontaneous and high K⫹-induced contractions (Figure 89.4). Atropine is a pharmacological antagonist of acetylcholine at muscarinic receptors (Brown and Taylor, 2001) and the potentiation of spasmolytic effect in the presence of atropine points towards the presence of a cholinergic component in olives, which was perhaps interfering with the spasmolytic component. To confirm the presence of cholinergic constituent(s), activity-guided fractionation of the parent extract was carried out to obtain the aqueous and organic fractions. The results showed that in the absence of any intervention, the aqueous fraction produced a moderate spasmogenic effect at lower concentrations, followed by inhibitory effect at higher concentrations. In atropinized tissues, the spasmogenic effect was blocked and the spasmolytic effect was potentiated (Figures 89.5 and 89.6A),
100
75 Control
% of control
1 min
50 Spontaneous (n=7) 25
K+ (80 mM) (n=8)
OeF.Cr 0 0.03 0.1
0.3
1.0
3.0mg mL
⫺1
FIGURE 89.1 Tracing showing concentration-dependent inhibitory effect of the crude extract of Olea europea fruit (OeF.Cr) on spontaneously contracting isolated rabbit jejunum preparation (from Gilani et al., 2006a).
0.30 [OeF.Cr]
3.00
mg mL⫺1
FIGURE 89.2 Concentration-dependent inhibitory effect of the crude extract of Olea europea fruit (OeF.Cr) on spontaneous and K⫹-induced contractions in isolated rabbit jejunum preparations. The symbols represent mean ⫾ SEM, n ⫽ 7–8 (from Gilani et al., 2006a).
839
CHAPTER | 89 Medicinal Value of Combination of Cholinergic and Calcium Antagonist Constituents in Olives
which confirmed that the spasmogenic component was of cholinergic type, which was initially rendering the spasmolytic effect less effective. The spasmolytic effect of the aqueous fraction was also found to be mediated through Ca⫹⫹ channel blockade (Figure 89.6B). The petroleum spirit fraction showed a similar pattern of activity producing a spasmogenic effect at lower concentrations followed by a spasmolytic effect. The chloroform and ethyl acetate fractions produced only a relaxant effect (Figure 89.7). The Ca⫹⫹ antagonist and cholinergic effects of olives may be due to the presence of flavonoids (quercetin, rutin, catechin) and saponins, as we observed that such phytochemicals exhibit Ca⫹⫹ channel blocking (Gilani et al., 2006b; Ghayur et al., 2007) and cholinergic (Gilani et al., 1994) mechanisms, respectively. Choline, a known constituent of olive and precursor for acetylcholine synthesis
Control OeF.Cr 0.3 mg mL⫺1 OeF.Cr 1 mg mL⫺1
A 100 % of spontaneous contraction
% of control maximum
A 100
is also reported as a direct muscarinic agonist (Pomara et al., 1983), which might be contributing to the cholinergic action of olives. The co-existence of spasmolytic (Ca⫹⫹ antagonist) and spasmogenic (cholinergic) constituents in olives rationalizes the medicinal use of olives in gastrointestinal disorders, such as abdominal colic, diarrhea and constipation. Interestingly, the presence of a novel combination of Ca⫹⫹ antagonist and cholinergic components, which exhibit opposite effects in gut, makes olives a useful remedy as reflected from its wide and common use. The inhibitory component (Ca⫹⫹ antagonist) does not allow the stimulatory one to go beyond a certain limit, above which the cholinergic(s) is likely to produce abdominal cramps usually seen with conventional chemical laxative drugs (Gilani and Atta-ur-Rahman, 2005), hence olives exhibit side-effect-neutralizing potential.
75
50
25
0
75
50
25 Without atropine (n=7) With atropine (n=4)
0 −4.5
−4.0
−3.5
−3.0
−2.5
−2.0
−1.5
0.03
Log [Ca++] M
75
50
25
75
50
25 Without atropine (n=8) With atropine (n=4)
0
0 −4.5
−4.0
−3.5
3.00
B 100
Control Verapamil 0.1 μM Verapamil 0.3 μM Verapamil 1 μM
% of K+ induced contraction
% of control maximum
B 100
0.30 [OeF.Cr] mg mL⫺1
−3.0
−2.5
−2.0
−1.5
Log [Ca++] M FIGURE 89.3 Concentration–response curves of Ca⫹⫹ in the absence and presence of the increasing doses of (A) crude extract of Olea europea fruit (OeF.Cr) and (B) verapamil in isolated rabbit jejunum preparations. The symbols represent mean ⫾ SEM, n ⫽ 3–5 (from Gilani et al., 2006a).
0.03
0.30
3.00
[OeF.Cr] mg mL⫺1 FIGURE 89.4 Concentration-dependent inhibitory effect of the crude extract of Olea europea fruit (OeF.Cr) on (A) spontaneous and (B) K⫹ (80 mM)-induced contractions in the absence and presence of atropine in isolated rabbit jejunum preparations. The symbols represent mean ⫾ SEM, n ⫽ 4–8 (from Gilani et al., 2006a).
840
SECTION | II Vascular Aspects Including Hypertension
1 min
A 150 % of ACh max.
Control
OeF.Aq
Atropine (0.03 μM)
ACh (1 μM)
0.1
0.3
100
3.0 mg mL⫺1 % of control
1.0
0.3
50
With atropine (n=4)
1.0 mg mL⫺1 0
OeF.Aq
FIGURE 89.5 Tracing showing effect of aqueous fraction of Olea europea fruit (OeF.Aq) on spontaneous contractions, in comparison to acetylcholine (ACh), in the absence and presence of atropine in isolated rabbit jejunum preparation (from Gilani et al., 2006a).
We have previously observed the presence of such a novel combination of cholinergic and Ca⫹⫹ antagonist components in some other medicinal plants, which are widely used in the hypo- and hyperactive status of the gut, namely: cardamom (Gilani et al., 2008), ginger (Ghayur and Gilani, 2005), ispaghul (Gilani and Atta-ur-Rahman, 2005), Calendula officinalis (Bashir et al., 2006), Carum copticum (Gilani et al., 2005b), Fumeria indica (Gilani et al., 2005c), Hibiscus rosasinensis (Gilani et al., 2005d), Piper betle (Gilani et al., 2000) and Saussurea lappa (Gilani et al., 2007). The Ca⫹⫹ antagonists and cholinergic drugs are known to exhibit opposite effects (excitatory and inhibitory) in the gastrointestinal tract, but similar effects (both are inhibitory) in the cardiovascular system and effectively reduce the blood pressure (Gilani and Atta-ur-Rahman, 2005). The intravenous administration of olive extract caused a fall in systolic, diastolic and mean arterial blood pressure of normotensive anesthetized rats (Figure 89.8). As the blood pressure is considered the product of cardiac output and peripheral resistance (Johansen, 1992), the olive extract was further studied in isolated cardiac and vascular smooth muscle preparations (Gilani et al., 2005a). In spontaneously beating isolated guinea-pig right atria, the olive extract caused negative inotropic and chronotropic effects (Figure 89.9). When tested on rabbit thoracic aorta preparations, the olive extract relaxed the aortic rings precontracted with high K⫹ and/or phenylephrine, indicating a non-specific vasodilator effect (Figure 89.10) mediated
0.03
0.30 [OeF.Aq] mg mL⫺1
3.00
B 100 % of K+ induced contraction
ACh 0.1 (1 μM)
Without atropine (n=6)
75
50 Without atropine (n=4) With atropine (n=4)
25
0 0.03
0.30 [OeF.Aq] mg mL⫺1
3.00
FIGURE 89.6 Effect of aqueous fraction of Olea europea fruit (OeF. Aq) on (A) spontaneous and (B) K⫹ (80 mM)-induced contractions in the absence and presence of atropine in isolated rabbit jejunum preparations. The spasmogenic responses are expressed as percent of acetylcholine maximum (ACh Max.). The symbols represent mean ⫾ SEM, n ⫽ 4–6 (from Gilani et al., 2006a).
through calcium channel blockade though additional mechanisms cannot be ruled out. Olives are well known for their medicinal value in cardiovascular disorders and the presence of various constituents possessing hypocholesterolemic (De Pasquale et al., 1991) and angiotensin-converting enzyme inhibitory (Hansen et al., 1996) activities along with monosaturated fatty acids (Stark and Madar, 2002) have been considered the contributing factors. Our recent reports for the presence
841
CHAPTER | 89 Medicinal Value of Combination of Cholinergic and Calcium Antagonist Constituents in Olives
150
100
75 % of control
% of control % of ACh max.
OeF.Pet
100
50
Atrial force Atrial rate
25
50
0 OeF.CHCl3 OeF.EtAc
0.03
0.30
3.00
[OeF.Cr] mg mL⫺1 FIGURE 89.9 Inhibitory effect of the crude extract of Olea europea fruit (OeF.Cr) on force of contraction and rate of spontaneously beating isolated guinea-pig right atria preparations. The symbols represent mean ⫾ SEM, n ⫽ 4 (from Gilani et al., 2005a).
0 0.01
0.10
1.00
[Dose] mg mL⫺1 FIGURE 89.7 Concentration–response curves showing the effect of organic fractions of the crude extract of Olea europea fruit viz. petroleum spirit (OeF. Pet), chloroform (OeF.CHCl3) and ethyl acetate (OeF.EtAc) on spontaneous contractions in isolated rabbit jejunum preparations. The spasmogenic responses are expressed as percent of acetylcholine maximum (ACh Max.), n ⫽ 1 (from Gilani et al., 2006a).
% of induced contraction
100
40
% fall in MABP
30
75
50 K+ (80 mM) 25
PE (1 μM)
0 20
0.03
0.30
3.00
[OeF.Cr] mg mL⫺1 FIGURE 89.10 Concentration-dependent inhibitory effect of the crude extract of Olea europea fruit (OeF.Cr) on K⫹ and phenylepherine (PE)induced contractions in isolated rabbit aorta preparations. The symbols represent mean ⫾ SEM, n ⫽ 4 (from Gilani et al., 2005a).
10
0 30.0
60.0 [OeF.Cr] mg kg⫺1
100.0
FIGURE 89.8 Dose-dependent effects of crude extract of Olea europea fruit (OeF.Cr) on mean arterial blood pressure (MABP) in anesthetized rats. Values presented as mean ⫾SEM, n ⫽ 6 (from Gilani et al., 2005a).
SUMMARY POINTS ●
●
of calcium antagonist and cholinergic constituents (Gilani et al., 2005a, 2006a) provide additional evidence for the usefulness of olives in cardiovascular disorders, though further studies are required in support of the endothelial-dependent and atropine-sensitive vasodilator effects of olives.
●
Olives possess a combination of cholinergic and calcium-channel-blocking mechanisms, which provides a pharmacological basis for its medicinal use in gut motility disorders, like colic, diarrhea and constipation. The spasmolytic component (Ca⫹⫹ antagonist), being dominant, masks the stimulant effect. The cholinergic component was concentrated in the aqueous fraction of olive extract, while the choloroform and ethylacetate fractions were found devoid of any spasmogenic effect.
842
●
●
SECTION | II Vascular Aspects Including Hypertension
The presence of flavonoids, choline and saponins might account for the observed Ca⫹⫹ antagonist and cholinergic actions of olives. Olives exhibit blood-pressure-lowering, cardiodepressant and vasodilatory activities, which explains their application in hypertension.
REFERENCES Baquar, S.R., 1989. Medicinal and Poisonous Plants of Pakistan. Printas, Karachi pp. 304–305. Bashir, S., Janbaz, K.H., Jabeen, Q., Gilani, A.H., 2006. Studies on spasmogenic and spasmolytic activities of Calendula officinalis flowers. Phytother. Res. 20, 906–910. Bolton, T.B., 1979. Mechanism of action of transmitter and other substances on smooth muscles. Physiol. Rev. 59, 606–718. Brown, J.H., Taylor, P., 2001. Muscarinic receptors agonist and antagonists. In: Hardman, J.G., Limbird, L.E., Gilman, A.G. (Eds.), Goodman and Gilman’s the Pharmacological Basis of Therapeutics, 10th edn. McGraw-Hill, New York, pp. 141–160. Caulfield, M.P., 1993. Muscarinic receptors: characterization, coupling and function. Pharmacol. Ther. 58, 319–379. Covas, M.I., 2007. Olive oil and the cardiovascular system. Pharmacol. Res., 175–186. De Pasquale, R., Monforte, M.T., Trozzi, A., Raccuia, A., Tommasini, S., Ragusa, S., 1991. Effects of leaves and shoots of Olea europea and oleuropein on experimental hypercholesterolemia in rat. Plant Med. Phytother. 25, 134–140. Devroede, G.J., 1989. Constipation: mechanism and management. In: Sleisenger, M.H., Fordtran, J.S. (Eds.), Gastrointestinal Disease, 4th edn. W.B. Saunders, Philadelphia, pp. 331–368. DiPalma, J.R., 1990. Antihypertensive drugs. In: DiPalma, J.R., DiGregorio, G.J. (Eds.), Basic Pharmacology in Medicine, 3rd edn. McGraw-Hill, New York, pp. 421–435. Duke, J.A., Bogenschutz-Godwin, M.J., Ducelliar, J., Duke, P.A.K., 2002. Hand Book of Medicinal Herbs, 2nd edn. CRC Press, Boca Raton pp. 535–536. Farre, A.J., Columbo, M., Fort, M., Gutierrez, B., 1991. Differential effects of various Ca⫹⫹ antagonists. Gen. Pharmacol. 22, 177–181. Findling, R., Frishman, W., Javed, M.T., Heffer, S., Brandt, L., 1996. Calcium channel blockers and the gastrointestinal tract. Am. J. Ther. 3, 383–408. Fine, K.D., Krejs, G.J., Fordtran, J.S., 1989. Diarrhea. In: Sleisenger, M.H., Fordtran, J.S. (Eds.), Gastrointestinal Disease, 4th edn. W.B. Saunders, Philadelphia, pp. 290–315. Flack, J.M., Peters, R., Shafi, T., Alrefai, H., Nasser, S.A., Crook, E., 2003. Prevention of hypertension and its complications: theoretical basis and guidelines for treatment. J. Am. Soc. Nephrol. 14, 92–98. Fleckenstein, A., 1977. Specific pharmacology of Ca⫹⫹ in myocardium, cardiac pacemakers and vascular smooth muscle. Rev. Pharmacol. Toxicol. 17, 149–166. Ghayur, M.N., Gilani, A.H., 2005. Pharmacological basis for the medicinal use of ginger in gastrointestinal disorders. Dig. Dis. Sci. 50, 1889–1897. Ghayur, M.N., Khan, H., Gilani, A.H., 2007. Antispasmodic, bronchodilator and vasodilator activities of (⫹)-catechin, a naturally occurring flavonoid. Arch. Pharm. Res. 30, 970–975.
Gilani, A.H., 1992. Constipation and its treatment with natural drugs used in Pakistan. In: Cappaso, F., Mascolo, N. (Eds.), Natural Drugs and the Digestive Tract. EMSI, Roma, pp. 117–121. Gilani, A.H., Aftab, K., Ahmed, S., 1994. Cholinergic actions of crude saponins from Castanospermum australe. Int. J. Pharmacog. 32, 209–216. Gilani, A.H., Aziz, N., Khurram, I.M., Rao, Z.A., Ali, N.K., 2000. The presence of cholinomimetic and calcium antagonist consituents in piper betle. Phytother. Res. 14, 436–442. Gilani, A.H., Atta-ur-Rahman, 2005. Trends in ethnopharmacology. J. Ethnopharmacol. 100, 43–49. Gilani, A.H., Khan, A., Shah, A.J., Cornor, J., Jabeen, Q., 2005a. Blood pressure lowering effect of olives is mediated through calcium channel blockade. Int. J. Food Sci. Nutr. 56, 613–620. Gilani, A.H., Jabeen, Q., Ghayur, M.N., Janbaz, K.H., Akhtar, M.S., 2005b. Studies on the antihypertensive, antispasmodic, bronchodilator and hepatoprotective activities of the Carum copticum seed extract. J. Ethnopharmacol. 98, 127–135. Gilani, A.H., Bashir, S., Janbaz, K.H., Khan, A., 2005c. Pharmacological basis for the use of Fumaria indica in constipation and diarrhoea. J. Ethnopharmacol. 96, 585–589. Gilani, A.H., Bashir, S., Janbaz, K.H., Shah, A.J., 2005d. Presence of cholinergic and calcium channel blocking activities explains the traditional use of Hibiscus rosasinensis in constipation and diarrhea. J. Ethnopharmacol. 102, 289–294. Gilani, A.H., Khan, A., Ghayur, M.N., 2006a. Ca⫹2 antagonist and cholinergic activities explain the medicinal use of olive in gut disorders. Nutr. Res. 26, 277–283. Gilani, A.H., Khan, A., Ghayur, M.N., Ali, S.F., Herzig, J.W., 2006b. Antispasmodic effect of Rooibos tea (Aspalathus linearis) is mediated predominantly through K⫹ channel activation. Basic Clin. Pharmacol. Toxicol. 99, 365–373. Gilani, A.H., Shah, A.J., Yaeesh, S., 2007. Presence of cholinergic and calcium antagonist constituents in Saussurea lappa explains its use in constipation and spasm. Phytother. Res. 21, 541–544. Gilani, A.H., Jabeen, Q., Khan, A., Shah, A.J., 2008. Gut modulatory, blood pressure lowering, diuretic and sedative activities of cardamom. J. Ethnopharmacol. 115, 463–472. Godfraind, T., Miller, R., Wibo, M., 1986. Calcium antagonism and calcium entry blockade. Pharmacol. Rev. 38, 312–416. Hansen, K., Adsersen, A., Christensen, S.B., Jensen, S.R., Nyman, U., Smitt, U.W., 1996. Isolation of an angiotensin converting enzyme inhibitor from Olea europea and Olea lancea. Phytomedicine 2, 319–325. Johansen, P.L., 1992. Hemodynamic effects of calcium antagonists in hypertension. In: Epstein, M. (Ed.), Calcium Antagonists in Clinical Medicine. Hanley and Belfus Inc., Philadelphia, pp. 69–88. Keys, A., 1987. Olive oil and coronary heart diseases. Lancet 1, 938–984. Lasserre, B., Kaiser, R., Pham, H.C., Ifansyah, N., Gleye, J., Moulis, C., 1983. Effect on rats of aqueous extracts of plants used in folk medicine as antihypertensive agents. Naturwissenschaften 70, 95–96. Liebelt, E.L., 1998. Clinical and laboratory evaluation and management of children with vomiting and dehydration. Curr. Opin. Pediatr. 10, 461–469. MacMahon, S., Peto, R., Cutler, J., Collins, R., Sorlie, P., Neaton, J., Abbott, R., Godwin, J., Dyer, A., Stamler, J., 1990. Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 335, 765–774.
CHAPTER | 89 Medicinal Value of Combination of Cholinergic and Calcium Antagonist Constituents in Olives
Mancia, G., Volpe, R., Boros, S., Ilardi, M., Giannattsio, C., 2004. Cardiovascular risk profile and blood pressure control in Italian hypertensive patients under specialist care. J. Hypertens. 22, 51–57. Meyer, J.H., 1987. Motility of the stomach and gastroduodenal junction. In: Johnsin, L.R. (Ed.), Physiology of the Gastrointestinal Tract, 2nd edn. Raven Press, New York, pp. 613–630. Occhiuto, F., Circosta, C., Gregorio, A., Busa, G., 1990. Olea europea and oleuropein: effects on excito-conduction and on monophasic action potential in anaesthetized dogs. Phytother. Res. 4, 140–143. Pomara, N., Block, R., Demetriou, S., Fucek, F., Stanley, M., Gershon, S., 1983. Attenuation of pilocarpine-induced hypothermia in response to chronic adminstration of choline. Psychopharmacology 80, 129–130. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food: epidemiology and nutritional approaches. Nutr. Rev. 60, 170–176. Stoelting, R.K., 1995. Handbook of Pharmacology & Physiology in Anesthetic Practice. Lippincott-Raven Publishers, New York pp. 278-287. Thomas, R., Hendrix, M.D., Gregory, B., Bulkely, M.D., Marvin, M., Schuster, M., 1996. Abdominal pain. In: Stobo, J.D., Hellman, D.B.,
843
Landenson, P.W., Petty, B.G., Traill, T.A. (Eds.) The Principles and Practice of Medicine, 23rd edn. Prentice International Inc, London, pp. 439–451. Visioli, F., Bellosta, S., Galli, C., 1998. Oleuropein, the bitter principle of olives, enhances nitric oxide production by mouse macrophages. Life Sci. 62, 541–546. William, J., Snape, J.R., 1992. Disorders of gastrointestinal motility. In: Wyngaarden, J.B., Smith, L.H., Bennet, J.C. (Eds.), Cecil Textbook of Medicine, 20th edn. W.B. Saunders, Philadelphia, pp. 671–680. Wingard, L.B., Brody, T.M., Larner, J., Schwartz, A., 1991. Human Pharmacology: Molecular to Clinical. Mosby-Year Book, St. Louis, pp. 167–222. Wray, S., Burdyga, T., Noble, K., 2005. Calcium signalling in smooth muscle. Cell Calcium 38, 397–407. Zarzuelo, A., Duarte, J., Jimenez, J., Gonzalez, M., Utrilla, M.P., 1991. Vasodilator effect of olive leaf. Planta Med. 57, 417–419.
This page intentionally left blank
Chapter 90
Olive Oil Cultivars and Atherosclerotic Protection in Apolipoprotein E-knockout Mice José Miguel Arbonés-Mainar1 and Jesús Osada1,2 1 2
Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Veterinaria, Universidad de Zaragoza, Spain CIBER de Fisiopatología de la Obesidad y Nutrición, Instituto de Salud Carlos III, Spain
90.1 INTRODUCTION The Mediterranean diet is associated with a reduction in all causes of mortality (Trichopoulou et al., 2003; Knoops et al., 2004) despite the high intake of fat, mainly derived from olive oil (Keys et al., 1986). Extra virgin olive oil, as a fruit juice, is a complex mixture where triglycerides are combined with other biologically active substances such as tocopherols, phenolic compounds, phytosterols and triterpenoids, some of which have antioxidant and anti-inflammatory activities (Visioli et al., 2000, 2003; de la Puerta et al., 2001; de la Puerta-Vazquez et al., 2004; Perona et al., 2006). The current view proposes that these components might be responsible for the benefits of virgin olive oil in animal models (Calleja et al., 1999; Rodriguez-Rodriguez et al., 2004; Herrera et al., 2006; Acin et al., 2007) and in human studies (Kris-Etherton et al., 1999; Abia et al., 2001; Perona et al., 2003, 2006; Perez-Jimenez, 2005). Extra virgin olive oil (EVOO) from different cultivars has subtle differences in fatty acids and minor components (Arbonés-Mainar et al., 2008) which therefore are a source of variability worthy of being considered and tested regarding the anti-atherogenic properties of EVOO. Interactions among its components and diets should be also examined using experimental models. The apoE-deficient mouse develops spontaneous atherosclerosis on a regular low-fat/low-cholesterol diet. The progression and histopathology of lesions in this model show similar features to those observed in humans and other species (Sarria et al., 2006). The aim of this study was to investigate the effects of a Western diet (WD, high fat, high cholesterol) enriched either with an EVOO of different cultivars or with palm oil on serum lipids, lipoproteins and the development of atherosclerosis in apoE-deficient mice. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
90.2 EFFECTS OF DIFFERENT WD ON PLASMA LIPIDS The study was carried out in 54 female apoE-deficient mice bred at the Unidad Mixta de Investigación, Zaragoza (Spain) with ad libitum access to food and water. Mice with similar baseline plasma cholesterol concentrations were randomly assigned to one of five groups and, for 10 weeks, fed a diet of standard mouse chow diet (B & K Universal Ltd., Humberside, UK) supplemented with 0.15% (w/w) of cholesterol and 20% (w/w) of fat. In four of the groups, the fat source was extra virgin olive oil from the following Spanish cultivars: Arbequina, Picual, Cornicabra or Empeltre. In the fifth group, the fat source was palm oil. All of the diets were isocaloric and isonitrogenous, and provided similar amounts of fat, carbohydrates and cholesterol (Table 90.1). All of the diets that contained olive oil had less saturated fatty acids and a higher P/S ratio than did the diet supplemented with palm oil. The diets that contained EVOO had high proportions of phenolic compounds, although there was substantial variation among diets; thus, the content in Picual was six times higher than in the diet enriched with Cornicabra olive oil. The experimental protocols were approved by the Ethics Committee for Animal Research of the University of Zaragoza. At the end of the experiment, the animals were sacrificed by suffocation with CO2 and blood samples were taken. Plasma total triglyceride and cholesterol concentrations were measured. Mice fed a diet enriched with either Picual or Cornicabra EVOO exhibited plasma triglyceride levels that were lower than those of mice fed a diet enriched with Arbequina EVOO or palm oil. Mice fed the palm-oil-enriched diet had levels of total plasma
845
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
846
SECTION | II Vascular Aspects Including Hypertension
cholesterol that were significantly lower than those of mice fed one of the four diets enriched with olive oil (Table 90.2), and the distribution of cholesterol corroborated a lower content of cholesterol in VLDL and LDL (Table 90.2).
Mice fed Picual had a higher proportion of cholesterol in the HDL fraction than did mice fed Arbequina or palm oil, no statistical differences were observed among the other groups.
TABLE 90.1 Chemical composition of experimental diets fed to five groups of apoE-deficient mice. Extra virgin olive oil varieties Arbequina
Picual
Palm oil
Cornicabra
Empeltre
Energetic content (kJ g⫺1)
18
18
18
18
18
Carbohydrate
46
46
46
46
46
Protein
13
13
13
13
13
Fat
22
22
22
22
22
Cholesterol (mg %) Monounsaturated Polyunsaturated Saturated P/S ratio
175 71.18 12.82 15.96 0.80
Total phenols (mg kg⫺1)
175 78.54 6.54 15.45 0.42
25
175 79.57 8.08 12.34 0.65
90
175 75.19 10.74 14.06 0.76
15
175 39.25 15.42 45.32 0.34
45
2
This table summarizes the dietary components expressed as g % (w/w). Other components of the chow diet include crude fiber 4.5% and minerals 6.8%. A total dry matter of 87.5%. P/S ratio ⫽ polyunsaturated/saturated fatty acid ratio. Reprinted from Arbonés-Mainar et al, Atherosclerosis 194: 372–382, Copyright (2007), with permission from Elsevier.
TABLE 90.2 Effects of experimental diets on apoE-deficient mice. Varieties of extra virgin olive oil
Palm oil (n ⫽ 10)
Arbequina (n ⫽ 11)
Picual (n ⫽ 11)
Cornicabra (n ⫽ 11)
Empeltre (n ⫽ 11)
TG (mmol L⫺1)
2.5 ⫾ 0.7
1.6 ⫾ 0.6a
1.6 ⫾ 0.5a
2.1 ⫾ 0.5
2.2 ⫾ 0.4b,c
TC (mmol L⫺1)
37 ⫾ 4
41 ⫾ 2
37 ⫾ 5
38 ⫾ 6
30 ⫾ 4 a,b,c,d
VLDLc (mmol L⫺1)
23 ⫾ 2
25 ⫾ 1
24 ⫾ 3
23 ⫾ 4
19 ⫾ 2 a,b,c,d
LDLc (mmol L⫺1)
12 ⫾ 1
15 ⫾ 1
14 ⫾ 2
13 ⫾ 2
11 ⫾ 1b,c,d
HDLc (mmol L⫺1)
1.2 ⫾ 0.1
1.4 ⫾ 0.1a
1.4 ⫾ 0.2
1.4 ⫾ 0.2
1.2 ⫾ 0.2b
ApoA-I (AU L⫺1)
30 ⫾ 7
36 ⫾ 9
35 ⫾ 6
31 ⫾ 9
36 ⫾ 6
ApoA-IV (AU L⫺1)
22 ⫾ 6
26 ⫾ 6
22 ⫾ 3
23 ⫾ 3
19 ⫾ 2b,d
PON Arylesterase (UI L⫺1)
42557 ⫾ 4542
44968 ⫾ 15180
43468 ⫾ 10004
38497 ⫾ 5034
49014 ⫾ 5517
Data are given as mean ⫾ SD. Statistical analyses to assess differences between diets were done using one-way ANOVA, either Bonferroni (with similar variances) or Tamhane (if variances were different) were performed as post-hoc tests. Different superscripts (a vs. Arbequina, b vs. Picual, c vs. Cornicabra and d vs. Empeltre) are significantly different from each other at p ⬍ 0.05. HDL, high-density lipoproteins; LDL, low-density lipoproteins; PON, paraoxonase; TC, total cholesterol; TG, triglycerides; VLDL, very-low-density lipoproteins. Adapted from Arbonés-Mainar et al, Atherosclerosis 194: 372–382, Copyright (2007), with permission from Elsevier.
CHAPTER | 90 Olive Oil Cultivars and Atherosclerotic Protection in Apolipoprotein E-knockout Mice
90.3 EFFECTS OF DIFFERENT WD ON ATHEROSCLEROTIC LESIONS At the end of the experiments, hearts from the sacrificed animals were perfused with a buffered saline phosphate solution, removed and frozen until analysis of aortic atherosclerosis. Serial cryosections of the aortic base of hearts were carried out, stained, and lesion areas estimated using Scion Image software (Scion Corporation, Frederick, Maryland, USA) (Calleja et al., 1999). Individual values of area of atherosclerotic lesion at the aortic root are shown in Figure 90.1. Aortic cross-sectional areas of atherosclerotic lesion were significantly lower in mice supplemented with EVOO from any cultivar than in mice fed the diet enriched with palm oil. All of the EVOO cultivars had similar efficacies.
90.4 EFFECTS OF DIETS ON PARAOXONASE AND PLASMA APOLIPOPROTEIN A-I AND A-IV CONCENTRATIONS Paraoxonase was estimated using arylesterase activity (Acin et al., 2005) in sera obtained from the animals of the study. Plasma apolipoproteins A-I and A-IV were quantified using ELISA and specific polyclonal antibodies (Navarro et al., 2005). Levels of arylesterase activity and apoA-I plasma concentrations did not differ significantly between mice fed palm oil and mice fed EVOO (Table 90.2), but concentrations of apoA-IV were higher in
0.25
mm2
0.20 0.15
847
mice fed an EVOO-enriched diet than palm oil diet, and the differences were significant in the Picual and Empeltre EVOO dietary groups.
90.5 EXPERIMENTAL OVERVIEW AND KEYNOTES This study examined the potential effects of four extra virgin olive oils versus palm oil as sources of fat on the development of atherosclerosis in apoE-deficient mice fed high-fat, high-cholesterol diets. Compared to the diet containing palm oil, a Western diet supplemented with any extra virgin olive oil of different cultivars led to a reduction in atherosclerotic lesion area. Here, we prove that the reduction in atherosclerosis after the nutritional intervention using EVOO was not associated with plasma total paraoxonase activity, apoA-I, plasma triglycerides and cholesterol. In fact, despite an augmented aortic lesion, the diet supplemented with palm oil induced a more positive lipid profile compared to all groups fed olive oil. Similar lipid changes were also observed in rabbits (Nielsen et al., 1995). In addition, mice fed a diet supplemented with palm oil exhibited triglyceride levels that were higher than the levels observed in mice fed Picual or Cornicabra EVOO, but similar to the levels observed in mice fed Arbequina or Empeltre EVOO supplements, which in turn had significantly less lesion surface area. Similarly, when we measured HDL cholesterol, although mice fed Picual EVOO did not differ from mice fed diets enriched with Cornicabra or Empeltre EVOO, they had HDLc levels higher than those in mice fed diets enriched with palm oil or Arbequina, and this parameter did not correlate with lesion surface area. These disparities underscored that even in a WD the effects of EVOO on atherosclerosis development were more favorable than those of palm oil, and independent of plasma cholesterol levels. In addition, our results demonstrate the discrepancies among secondary markers and the development of atherosclerosis, and the need for more reliable markers that correlate with the development of atherosclerosis in studies of olive oil intervention.
0.10 0.05 0 arbequina
picual
cornicabra
empeltre
palm oil
extra virgin olive oils
FIGURE 90.1 Effects of diets on the atherosclerotic lesion areas in apoE-deficient mice. Cross-sections of aortic regions from 51 animals were considered (three samples were not properly cut and, therefore, not included). Each symbol represents the lesion area of each animal; bars represent means. * p ⬍ 0.001 vs. all olive oil groups according to Kruskal-Wallis test followed of Mann-Whitney between pairs. Reprinted from Arbonés-Mainar et al., Atherosclerosis 194: 372–382, Copyright (2007), with permission from Elsevier.
90.6 THE SEARCH FOR NEW PLASMA BIOMARKERS TO EXPLAIN THE ATHEROSCLEROTIC CHANGES To accomplish such an endeavor, we focused on protein components of HDL, particularly paraoxonase, which protects LDL against oxidation (Tward et al., 2002), and apoA-I, which can remove LDL lipid hydroperoxides (Navab et al., 2000), and both considered as key keepers of the redox status (Barter et al., 2004). No effect of the experimental diets on plasma PON arylesterase activity or apoA-I plasma concentration were noted, which suggests that the
848
SECTION | II Vascular Aspects Including Hypertension
90.7 APOLIPOPROTEIN DISTRIBUTION AND ARYLESTERASE ACTIVITY AMONG LIPOPROTEIN SUBCLASSES
beneficial effects of EVOO could be hidden by the complex interactions of HDL particles. ApoA-IV has been associated with the antioxidant properties of HDL (Ostos et al., 2001), and its abundance depends on dietary fat (Kratz et al., 2003; Navarro et al., 2004). An increase in plasma apoA-IV concentrations in mice fed EVOO-enriched diets compared to mice fed a palm-oil supplement was found. Furthermore, a negative correlation between apoA-IV concentration and aortic lesion surface area was observed (see Figure 90.4A). This relationship has been observed in humans with coronary artery disease (CAD) (Kronenberg et al., 2000), and mice overexpressing human (Duverger et al., 1996) or murine (Cohen et al., 1997) apoA-IV showed reduced atherosclerosis. Our results support the notion that apoA-IV can protect against atherosclerosis through mechanisms that go beyond plasma lipids because the protection observed in our experiments was in the presence of an impaired lipid profile.
Plasma lipoprotein separation was carried out by fast protein liquid chromatography gel filtration (FPLC) as previously described (Calleja et al., 1999) and lipoproteins obtained are shown in Figure 90.2. Fraction numbers 5–10, 11–18, 19–29 and 30–40 corresponded to VLDL, LDL, HDL and lipid-free serum (LFS), respectively. The HDL fraction could be also subdivided into two: cholesterol-rich HDL particles (cr-HDL, fractions 19–24) and cholesterol-poor HDL particles, which have cholesterol concentrations ⬍100 μmol L⫺1 (cp-HDL, fractions 25–29) (ArbonesMainar et al., 2006). To further explore the role of HDL proteins, the distribution of apoA-I, apoA-IV and arylesterase activity among the HDL subclasses isolated by FPLC was undertaken and the results of cr-HDL, cp-HDL and
VLDL LDL cr-HDL cp-HDL LFS 100
100
80
80
60
μg cholesterol/fraction
VLDL LDL cr-HDL cp-HDL LFS
60
arbequina
40
40
20
20
0
0
100
100
80
80
60
picual
60
cornicabra
empeltre
40
40
20
20
0
0 1
100
10
20 30 Fraction number
40
80 60
palm
40 20 0
1
10
20
30
40
Fraction number FIGURE 90.2 Plasma lipoprotein profile of apoE-deficient mice following the experimental diets. Lipoproteins were fractionated by fast-performance liquid chromatography and results are presented as μg of cholesterol per fraction. Representative pattern for each diet is shown. Fraction numbers 5–10 corresponded to VLDL, 11–18 to LDL, 19–24 to cholesterol-rich HDL (cr-HDL), 25–29 to cholesterol-poor HDL (cp-HDL) and 30–40 to lipid free serum (LFS). HDL, high-density lipoproteins; LDL, low-density lipoproteins; VLDL, very-low-density lipoproteins. Reprinted from Arbonés-Mainar et al., Atherosclerosis 194: 372–382, Copyright (2007), with permission from Elsevier.
CHAPTER | 90 Olive Oil Cultivars and Atherosclerotic Protection in Apolipoprotein E-knockout Mice
100
arbequina 80
ApoA-I
849
picual
ApoA-IV PON-AE
60 % 40 20 0 100 80
cornicabra
empeltre
60 % 40 20 0
cr-HDL
cp-HDL
LFS
100 80
palm
60 % 40 20 0
cr-HDL
cp-HDL
LFS
FIGURE 90.3 Effect of different diets on the distribution of apoA-I, apoA-IV, and arylesterase activity in different lipoprotein fractions. Obtained by fast-performance liquid chromatography, fractions corresponding to cholesterol-rich HDL (cr-HDL), cholesterol-poor HDL (cp-HDL) and lipoproteinfree serum (LFS), as in Figure 90.2, were concentrated. ApoA-I, apoA-IV, and arylesterase activity concentrations were expressed as the proportion (%) of the total activity within each fraction. Results are given as mean ⫾ SD. HDL, high-density lipoproteins; LDL, low-density lipoproteins; VLDL, verylow-density lipoproteins. Reprinted from Arbonés-Mainar et al., Atherosclerosis 194: 372–382, Copyright (2007), with permission from Elsevier.
LFS are shown in Figure 90.3. Apolipoprotein A-I was mainly in the cholesterol-rich HDL subfraction, and its distribution profile was similar in all of the virgin olive oil and palm oil diets. In contrast, the apoA-IV exhibited different distributions. Thus, in mice fed palm oil, apoA-IV appeared mostly in the cr-HDL, whereas in all of the EVOO dietary groups, apoA-IV appeared elevated in the cp-HDL fraction. Most of the studies in humans (von Eckardstein et al., 1995; Kronenberg et al., 2000; Ezeh et al., 2003), but not all (Lagrost et al., 1989; Malmendier et al., 1991), describe a bimodal distribution in which apoA-IV normally exists both as a free apolipoprotein and in association with HDL particles. An observation initially considered an artefact of HDL preparation (ultracentrifugation vs. FPLC) due to the low affinity of apoA-IV for HDL. Using the mild procedure FPLC for isolating HDL, we observed that a bimodal distribution of apoA-IV in pigs was a consequence of a nutritional intervention with EVOO (Navarro et al., 2004). In our experiment, in mice fed a diet enriched
with EVOO or palm oil, apoA-IV remained invariant in the cholesterol-rich HDL, whereas its concentration in the cholesterol-poor HDL fraction was increased in mice fed an EVOO-enriched diet, and therefore contributing to the increase in total apoA-IV (Figure 90.3). Total plasma apoA-IV and aortic lesion area were inversely correlated (r ⫽ –0.292, p ⫽ 0.046). However, when only the apoA-IV present in the cholesterol-poor HDL fraction was considered (Figure 90.4A), the negative correlation with aortic lesion surface area was stronger (r ⫽ ⫺0.667, p ⬍ 0.001). Thus, all of the EVOO produced small cholesterolpoor, apoA-IV-enriched lipoparticles with high anti-atherosclerotic association. Lipid-poor, protein-rich discoid and spherical HDL particles that have low molecular mass possess greater efflux capabilities because they can acquire oxidized lipids more efficiently than large HDL (Asztalos et al., 1997) and apoA-IV has a great capacity to increase cholesterol efflux (Thorngate et al., 2003). In addition, this small, dense HDL is endowed with a potent antioxidative
850
SECTION | II Vascular Aspects Including Hypertension
apo A-IV (A.U)
0.15
r = −0.667, P400 °C) by pyrolysis. Three heterocyclic amines generated at moderate temperature (180–200 °C) by the Maillard reaction: 3-amino-1,4-dimethyl-5H-pyrido(4,3-b)-indole (Trp-P-1), 2-amino-3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx), and 2-amino-l-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP).
activity generated in foods with high protein content. They are called heterocyclic amines (HCAs) and are isolated from the basic extracts of cooked meat and fish (Sugimura, 1997). According to their structure and to the mechanism of formation, they are classified into two groups: those formed at moderate temperatures – imidazoquinolones, imidazoquinoxalines and imidazopyridines – which are called IQ-type mutagens, and those generated at high pyrolytic temperatures – pyrido-imidazoles and pyrido-indoles – which are non-IQtype mutagens (Figure 106.1). More than 20 different HCA compounds have been isolated from cooked meat, identified, and quantified by modern LC-MS techniques.
106.2 FORMATION OF MUTAGENS AT MODERATE TEMPERATURES AND TOXICITY IQ-type HCAs are generated from the Maillard reaction between free amino acids and sugars, with the participation of creatin(in)e (Figure 106.2). The heterocyclic pyridines and pyrazines are formed in the Maillard reaction between hexose and amino acids. The creatine undergoes dehydration and cyclization, thereby forming creatinine. Then precursors undergo further transformation with the participation of Strecker aldehydes and creatinine, producing imidazoquinoxalines, perhaps through free-radical reactions. Different combinations of creatinine, amino acids and sugars have generated MeIQx, 7,8-DiMeIQx, 4,8-DiMeIQx and PhIP in model systems (Jägerstad et al., 1983). HCAs are mutagenic in bacteria and can produce chromosomal aberrations and sister chromatid exchanges in
106.3 THE AMES TEST IS USED TO DETERMINE THE MUTAGENIC ACTIVITY The mutagenic activity of HCAs and PAHs has been principally evaluated by means of the Salmonella typhimurium reverse mutation assay or Ames test (Maron and Ames, 1983). This is an ‘in vitro’ procedure that uses strains of Salmonella typhimurium which do not synthesize histidine because they have a point mutation in genes coding for enzymes that participate in the synthesis of this amino acid. Point mutations are base substitutions or small deletions (frameshift mutations) that can be reversed spontaneously or by a mutagenic compound. Therefore, only retromutated bacteria are able to grow in a medium lacking histidine and, consequently, they form a bacterial clone that can be counted. Each Salmonella typhimurium strain has a different mutation, and a set of four or five strains are generally included in a complete assay to test the genotoxic potential of a chemical compound or mixture. It has been demonstrated that the HCAs are very mutagenic in strains that detect frameshift mutations, in particular the strains TA1538 and TA98 (Wakabayasi et al., 1992). For this reason, these strains are the ones that have been more frequently used for determining the mutagenic activity of food extracts or isolated compounds. A compound is considered mutagenic when the number of spontaneous revertants is at least double with respect to the untreated control and when a relationship between the dose of the compound and the number of revertants has been demonstrated. At least five concentrations of the compound are required in the experiment. In the case of cooked meat samples, the external crust that contains the mutagens is generally separated. Extraction procedures under basic conditions and purification methods have been developed to obtain extracts that can be tested in bacteria cultures (Bjeldanes et al., 1982a). The Ames assay must be performed in the presence of metabolic activation because the HCAs are converted to a DNA reactive compound through phase I and phase II reactions. All of the HCAs require metabolic activation but the reactions are different depending on the compound. Several isoforms of cytochrome P450 (CYP 1A1, CYP 1A2) are always implicated and various phase II enzymes
CHAPTER | 106 Mutagenic Activity in Meat Samples after Deep-frying in Olive Oil: Comparison with other Oils
991
H HN
C
R
3
HOOC
C6H12O6
N
COO− Glucose
CH3
aminoacid
Y
creatine
Z O
OHC
+ X
N
+
CH3
R
NH2 N
CH3 creatinine
aldehyde
pyridine or pyrazine
N
NH2 N Y
Z
N
X
N
R
CH3
Compound type IQ
FIGURE 106.2 A possible route of production of heterocyclic amines. The Maillard reaction between an amino acid and glucose gives rise to a pyridine or pyrazine nucleus. The creatine present in meat is converted into creatinine. Both precursors undergo further transformation with the participation of Strecker aldehydes, and yield type IQ compounds.
(N-acetyltransferase, sulfotransferase) can also be involved. To reproduce the in vivo metabolism in vitro, a mixture prepared with the microsomal or S9 fraction of rat liver homogenates and a NADPH-generating system are added to the cultures.
106.4 FACTORS THAT DETERMINE THE PRESENCE OF MUTAGENS IN MEAT 106.4.1 Influence of Time and Temperature The formation of mutagens in meat depends on different factors. Among them, temperature and cooking time are the most important because they have a profound effect on the quantities of HCAs (Knize et al., 1994a). Meat products which have not been cooked do not present any level of mutagenic activity (Davies et al., 1993). In cooking procedures, temperature can be transmitted by conduction, by convection or by radiation. Heating by conduction is the most aggressive method because food is in direct contact with the calorific surface, and the temperature increases very rapidly. Frying in a pan or on a grill plate is a very frequently
used cooking method throughout the world. Temperatures ranging between 175 °C and 250 °C are easily attained and maintained for 5–20 minutes, depending on the consumer preference or habit. Under these conditions, it has been well demonstrated that a considerable quantity of HCAs are produced (Pariza et al., 1979; Jägerstad and Skog, 1991). In pan-fried meat at temperatures below 150 °C, mutagenic activity is very low but with temperatures ranging from 190 °C to 210 °C, the mutagenic activity increases considerably (Pariza et al., 1979; Knize et al., 1994b). It is also known that mutagens are concentrated on the surface of the meat because in the inside, temperature rarely exceeded 100 °C (Felton et al., 1994). On the other hand, it has also been demonstrated in a great number of studies that the mutagenic activity of meat products increases with time, being practically undetectable during the first 5–10 minutes (Spingarn and Weisburger, 1979; Miller and Buchanan, 1983; Pérez et al., 2002).
106.4.2 Influence of the Cooking Method Due to the fact that temperature is a critical factor, cooking methods such as boiling or stewing do not generate significant
992
mutagenic activity (Bjeldanes et al., 1982b; Barrington et al., 1990). Frying and grilling are the cooking methods that generate the greatest amount of mutagenic compounds, even more than roasting. This is because heat transfer in an oven is less effective by convection (Holtz et al., 1985; Reuterswärd et al., 1987). Moreover, the external brown crust that is produced on the surface of the meat, and which contains the mutagens, represents a greater percentage with respect to the total weight in fried meat (5–25%) than in roasted meat (1–2%). A great number of studies have related the browning degree of the external crust in cooked meat with mutagenic activity (Layton et al., 1995; Augustsson et al., 1997; Sinha et al., 1998; Pérez et al., 2002).
106.4.3 Influence of Fat and Oil The influence of fat on the generation of mutagens can be seen from two different points of view: the content of fat in the meat and the use of fat or vegetable oils to cook the meat. Lipids present in the meat may participate in the Maillard reaction, increasing the production of pyridines, pyrazines and aldehydes (Jägerstad et al., 1983; Arnoldi et al., 1987). In addition, lipids may increase the mutagenic activity by the generation of free radicals (Skog et al., 1998). On the other hand, vegetable oils used in cooking procedures may modulate mutagenic activity by participating in the reactions that give rise to the mutagenic compounds and by affecting heat transmission during the cooking procedure. In model systems, with the components of the Maillard reaction (glucose and glycine) and creatin(in)e in adequate proportions, it was observed that the fat content significantly affected mutagen formation (Johansson and Jägerstad, 1993) and that the addition of corn oil or olive oil at 180 °C during 30 minutes doubled the yield of MeIQx (Johansson et al., 1993).
106.5 MUTAGENIC ACTIVITY IN FRIED MEAT Frying meat products in fat or oil is a very common practice in many countries worldwide. In northern European countries and in the United States, it is frequently done using small quantities (20–50 g) of butter, fats or oils that are added to the pan. Deep-frying, which uses a large quantity of oil in which meat products are submerged, is a common practice in Spain and other Mediterranean countries. The reports that have been published regarding the influence of oil or fat on mutagenic activity of fried meat have been reviewed. Nilsson et al. (1986) studied the mutagenic activity in lean pork meat fried at two different pan temperatures, 200 °C and 250 °C, with and without the addition of fat. Nine different fats with varying chemical compositions were tested: coconut oil, sunflower seed oil, olive oil, lard, butter or four margarines. The Ames test on Salmonella
SECTION | II
Cancer
strain TA98 with metabolic activation was performed. All fried meat samples were shown to be mutagenic. The meat samples fried at 250 °C were considerably more mutagenic than the samples fried at 200 °C. At 200 °C, the differences between meat fried in different fats or without fats were small. At 250 °C, the addition of fat caused a significant rise in mutagenic activity. The authors attributed this difference to a more efficient heat transfer from the bottom of the frying pan to the meat samples, although they were unable to discard other factors. Barrington et al. (1990) studied the mutagenic activity in meat samples of lamb and beef that were cooked using common household methods in Australia. They used the Salmonella typhimurium TA1538 strain in the presence of S9 homogenate to assess the mutagenicity of the basic fraction of cooked meat; the mutagenic potential was estimated by extrapolation from the slope of the TA1538 reverse mutants versus volume of meat extract (revertants 100 g⫺1 cooked meat). One of the cooking methods evaluated was frying rump or fillet steak in butter, margarine or sunflower oil (20 g), which was added to a preheated (3 min) flat iron pan. Meat samples were fried for 5 minutes (2.5 min each side) at different temperatures depending on the fat and on the degree of cooking. The temperature range was between 90 °C and 140 °C when using margarine/butter and between 105 °C and 200 °C with sunflower oil, for slightly or well-done meat, respectively. They found a great deal of mutagenic activity in well-done steak samples, and this activity was higher in samples fried in butter or margarine than in sunflower oil. When comparing the aforementioned with other cooking methods, frying and grilling gave the highest mutagenic activity. Johansson et al. (1995) studied the influence of six frying fats (butter, margarine, margarine fat phase, liquid margarine, rapeseed oil and sunflower seed oil) on the formation of HCAs during the frying of beefburgers. Four beefburgers (340 g) were fried at a time, using 50 g margarine or butter, or 40 g margarine fat phase or oil, respectively. The fat was placed in a preheated teflon-coated pan. After 2 minutes, the beefburgers were added and fried for 5 minutes on one side and 3 minutes on the other. The total heating time of the fat was 10 minutes and the temperature was 165 °C or 200 °C. The fried beefburgers and their corresponding pan residues were purified using solid-phase extraction and then analyzed for HCAs using HPLC with photodiode array UV and fluorescence detection. Five HCAs were recovered: 2-amino-3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx), 2amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline (DiMeIQx), 2-amino-l-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), 9H-pyrido[3,4-b]indole (norharman) and l-methyl9Hpyrido[3,4-b]indole (harman). The quantity increased with the rise in temperature, and the content of HCAs in the pan residue was much higher than in the corresponding beefburger. The total number of HCAs in meat and pan residue combined was significantly lower after frying in sunflower seed oil or margarine than after frying with other fats.
CHAPTER | 106 Mutagenic Activity in Meat Samples after Deep-frying in Olive Oil: Comparison with other Oils
106.5.1 Mutagenic Activity in Deep-fried Meat in Olive Oil As previously mentioned, deep-frying in oil is a cooking procedure frequently used in Mediterranean countries. Vegetables oils are used for this purpose, with olive oil or sunflower oil being the most frequently used. Pérez et al. (2002) studied the mutagenic activity found in meat samples that were deep-fried in a large volume of olive oil (200 mL). Olive oil was compared with other vegetable oils – sunflower oil, marc olive oil (‘orujo’) and soya bean oil, with butter, and with cooking without any fat, cooking practices which are more frequently carried out in northern countries. Soya oil is not normally used in cooking but it was selected for its different lipid profile. Beefburgers (100 g) were fried in 200 mL of oil placed in a teflon-coated pan for 10, 20 or 30 minutes. The initial frying temperature was 170–180 °C and it was controlled with a thermocouple probe throughout the entire process. In the meat samples, the initial temperature was 4 °C and it increased gradually; the outside reached the same temperature value as the oil but the inside part never surpassed 100 °C. After 10 minutes, the beefburgers were well-done but edible; longer times gave very dark brown products. The mutagenic activity was evaluated in Salmonella TA98, with and without metabolic activation by testing the basic extract prepared from 50 g of the external crust of cooked meat. In each assay, the pure extract and four dilutions (1/2, 1/4, 1/8 and 1/16) in DMSO prepared from the meat extract were tested (50 μL). Two independent assays were performed with extracts obtained from samples cooked in the same conditions. As the initial quantity was identical, comparisons between the different conditions could be made by comparing the dose–response trend lines (revertants versus extract dilution) induced by each extract. The results obtained with olive oil, sunflower oil, butter or without any oil at all are presented in Figure 106.3. The external crust of all the meat samples induced His⫹ revertants in a dose-dependent way, with metabolic activation. Longer cooking times generated samples with a higher mutagenic potency. In order to compare the mutagenic potential of each extract, the number of revertants per 10 μL of extract was obtained from the equations of the dose– response tendency lines (Table 106.2). At 10 minutes, the order of potency was olive oil > without oil > butter > sunflower oil. It has to be taken into account that the cooking temperature in olive oil was maintained around 180 °C as this is the smoke temperature. With sunflower oil or butter, the temperature was around 170 °C, and without any oil the temperature was approximately 180 °C. Since temperature is one of the most important factors in the generation of mutagens, this difference of approximately 10 °C may account for the greater potency of extracts in olive oil or without any fat. Moreover, comparing these two conditions at the same range of around 180 °C, the presence of olive
993
oil generated a more mutagenic extract probably due to the most efficient heat transfer. At 20 minutes, there was a significant increase in the mutagenic potency of all the extracts except in olive oil. The most potent extracts were obtained after frying in sunflower oil for a very long time (Figure 106.3, Table 106.2). The fact that longer cooking times produced a larger number of revertants in sunflower oil than in olive oil could be explained by the higher stability of olive oil due to its lower content of polyunsaturated fatty acids (PUFAs); sunflower oil has 50% of PUFA and olive oil has 11.2% (Table 106.2). Oxidative reactions are actively produced during frying, and they may generate free radicals that can potentiate the Maillard reaction. PUFAs are more susceptible to these reactions; therefore, the high potency of extracts obtained in sunflower oil may be explained by the intensive oxidation of this oil after long cooking times. When frankfurter samples were fried for 10 minutes under the same conditions as beefburgers, the potency of the extracts was considerably lower in all of the conditions tested (Figure 106.4). These results are due to the different compositions of the samples because beefburgers have a higher protein content and a lower fat proportion than frankfurters: 22.10 ⫾ 2.30 versus 12.25 ⫾ 0.62 and 8.53 ⫾ 0.74 versus 19.51 ⫾ 1.26, respectively. Although it was not measured, it can be assumed that there was a lower content of creatinine in the frankfurters. Protein and creatinine are components of the Maillard reaction; therefore, a lower level of both would give rise to a small number of mutagens. With respect to the fat present in the meat, two opposite effects are possible. The fat may contribute to better transmission of heat and, therefore, facilitate the formation of mutagens. However, when fat is present in high proportions it can also dilute the precursors of the Maillard reaction, thereby decreasing mutagenic activity (Pérez et al., 2002).
106.6 CONCLUSIONS Meat composition and temperature are the most important factors that determine the generation of mutagens in fried meat. At moderate temperatures (below 200 °C), a high content of proteins and creatine is necessary for generating, by Maillard reaction, the HCAs that are responsible for mutagenic activity. For this reason, the browner the external crust of fried meat, the more potent the mutagenic activity. The presence of fat or oil has two opposite effects: it can facilitate heat transfer and thus increase the generation of mutagens or it may act as a dilutant for the mutagenic compounds. Deep-frying of meat in vegetable oils for long times (approximately 10 min) gives rise to a significant quantity of mutagens; olive oil yields slightly more mutagens than sunflower oil does, probably because the smoking temperature for olive oil is higher (180 °C versus 170 °C). Very long frying times (20–30 min) significantly increase
994
SECTION | II
Olive oil
His+ revertants
250 200
Sunflower oil 300
0% S9 10 min
250
20 min
His+ revertants
300
30 min
150 100
200
30 min
0 1/16
1/8
1/4
1/2
1
Extract dilution
0
1/8 1/4 Extract dilution
Butter
Without oil 300
0% S9 10 min
250
20 min
His+ revertants
His+ revertants
20 min
50
0
200
10 min
100
0
250
0% S9
150
50
300
Cancer
30 min
150 100
200
1/16
1/2
1
1/2
1
0% S9 10 min 20 min 30 min
150 100
50
50
0
0 0
1/16
1/8
1/4
1/2
1
0
1/16
Extract dilution
1/8
1/4
Extract dilution
FIGURE 106.3 Mutagenic activity of beefburger meat extracts. Basic meat extracts were obtained from 50 g of the external crust of beefburgers (100 g) deep-fried in olive oil or in sunflower oil, with butter or without oil. Four dilutions in DMSO were prepared. 50 μL of the pure extract and the four dilutions (1/2, 1/4, 1/8 and 1/16) were assayed in the Ames test with metabolic activation (10% S9). The number of His⫹ revertants at the different concentrations is presented together with the dose–response curves. A control without metabolic activation (0% S9) is included. All the extracts were mutagenic in a dose–response way. The longer the time of cooking, the higher the number of revertants. This effect is less pronounced in olive oil.
TABLE 106.2 Fatty acid profiles of the oils and fats used to cook the meat, temperature range during cooking, and number of His⫹ revertants per 10 μL of extracts obtained from the crust of fried frankfurters or beefburgers at different times. Oil/Fat
Temp.(°C)
Fatty acid profiles
Beefburgers cooking time
SFA
MUFA
PUFA
10 min
20 min
30 min
Olive
163–187°C
14.1%
69.7%
11.2%
243
263
342
Sunflower
150–174°C
13.1%
31.8%
50.0%
107
349
433
Butter
152–175°C
38.4%
24.2%
2.3%
160
254
266
Without
163–188°C
–
–
172
330
351
–
The number of His⫹ revertants was obtained from the equations of the dose–response curves. At 10 min., deep-frying in olive oil gives extracts with a high mutagenic activity, which correlates with the higher temperature reached with this oil. At 20 and 30 min., deep-frying in sunflower oil gives the highest mutagenic potency, which correlates with the highest content on polyunsaturated fatty acids (PUFA) of this oil.
the mutagenic activity in meat fried in sunflower oil, probably due to the degradation of PUFA. Thus, a general recommendation for reducing the presence of mutagens in fried meat would be to cook during short times and at moderate temperatures (below 170 °C).
SUMMARY POINTS ●
High temperatures applied to meat yield a variety of compounds that produce mutations in the DNA and are carcinogenic in rodents.
60
olive oil
50
sunflower
His+ revertants / 10 μL of extract
CHAPTER | 106 Mutagenic Activity in Meat Samples after Deep-frying in Olive Oil: Comparison with other Oils
His+ revertants
butter 40
without oil
30 20 10 0 0
1/16
1/8
1/4
1/2
300
Frankfurters Beefburgers
250 200 150 100 50 0 Olive oil
1
995
Sunflower oil
Butter
Without oil
Extract dilution A
B
FIGURE 106.4 Mutagenic activity of frankfurter meat extracts and comparison with beefburgers. (A) Number of His⫹ revertants induced by 50 μL of the basic meat extract (1) and four dilutions in DMSO (1/2, 1/4, 1/8 and 1/16), obtained from 50 g of the external crust of frankfurters deep-fried in different conditions. All the extracts were mutagenic in a dose–response way, without significant differences between cooking conditions. (B) Mutagenic potential (revertants His⫹/10 μL of extract) of beefburgers and frankfurters fried in different conditions. Beefburger extracts are significantly more mutagenic than frankfurters.
●
●
●
●
●
●
Mutation is a molecular event that is related to cancer. Some known carcinogens are also mutagens. Therefore, the generation of mutagens in food poses a problem to human health. Frying in fat or oil is a cooking method that may generate mutagenic compounds because temperatures higher than 170–180 °C may be reached. Vegetable oils used for frying may modulate mutagenic activity by participating in the reactions that give rise to the mutagenic compounds and by affecting heat transmission during the cooking process. Deep-frying in olive oil for short times produces a more mutagenic crust extract than frying in sunflower oil because the smoke temperature is higher (180 °C versus 170 °C). Very long frying times produce more mutagenic crust extracts in sunflower oil than in olive oil probably due to the increased degradation of the former which has a higher content of polyunsaturated fatty acids (50% versus 11.2%). It is recommended to fry meat for short times at temperatures below 170 °C in order to reduce the generation of mutagens.
REFERENCES Alaejos, M.S., González, V., Afonso, A.M., 2008. Exposure to heterocyclic aromatic amines from the consumption of cooked red meat and its effect on human cancer risk: A review. Food Addit. Contam. 25, 2–24. Arnoldi, A., Arnoldi, C., Baldi, O., Griffini, A., 1987. Strecker degradation of leucine and valine in a lipidic model system. J. Agric. Food Chem. 35, 1035–1038. Augustsson, K., Skog, K., Jägerstad, M., Steineck, G., 1997. Assessment of the human exposure to heterocyclic amines. Carcinogenesis 18, 1931–1935.
Barrington, P.J., Baker, R.S., Truswell, A.S., Bonin, A.M., Ryan, A.J., Paulin, A.P., 1990. Mutagenicity of basic fractions derived from lamb and beef cooked by common household methods. Food Chem. Toxicol. 28, 141–146. Bjeldanes, L.F., Grose, K.R., Davis, P.H., Stuermer, D.H., Healy, S.K., Felton, J.S., 1982a. An XAD-2 resin method for efficient extraction of mutagens from fried ground beef. Mutat. Res. 105, 43–49. Bjeldanes, L.F., Morris, M.M., Felton, J.S., Healy, S., Stuermer, D., Berry, P., Timourian, H., Hatch, F.T., 1982b. Mutagens from the cooking of food. III. Survey by Ames/”Salmonella” test of mutagen formation in secondary sources of cooked dietary protein. Food Chem. Toxicol. 20, 365–369. Davies, J.E., Chipman, J.K., Cooke, M.A., 1993. Mutagen formation in beefburgers processed by frying or microwave with use of flavoring and browning agents. J. Food Sci. 58, 1216–1218. Felton, J.S., Fultz, E., Dolbeare, F.A., Wu, R., 1994. Effect of microwave pretreatment on heterocyclic aromatic amine mutagens/carcinogens in fried beef patties. Food Chem. Toxicol. 32, 897–903. Holtz, E., Skjöldebrand, C., Jägerstad, M., Reuterswärd, A.L., Isberg, P.E., 1985. Effect of recipes on crust formation and mutagenicity in meat products during baking. Int. J. Food Sci. Tech. 20, 57–66. IARC, 1993. IARC Monographs on the Evaluation of the Carcinogenic Risk to Humans, Vol. 56: Some Naturally Occurring Substances: Food Items and Constituents, Heterocyclic Aromatic Amines and Mycotoxins, pp. 13–18. Jägerstad, M., Reuterswärd, A.L., Öste, R., Dahlqvist, A., Grivas, S., Olsson, K., and Nyhammar, T., 1983. Creatinine and Maillard reaction products as precursors of mutagenic compounds formed in fried beef. In: Waller, G.R., Feather, M.S. (eds) The Maillard Reaction in Foods and Nutrition, ACS Symposium Series N° 215, pp. 507–519. Jägerstad, M., Skog, K., 1991. Formation of meat mutagens. Adv. Exp. Med. Biol. 289, 83–105. Johansson, M., Jägerstad, M., 1993. Influence of oxidized deep-fryng fat and iron on the formation of food mutagens in a model system. Food Chem. Toxicol. 31, 971–979. Johansson, M., Skog, K., Jägerstad, M., 1993. Effects of edible oils and fatty acids on the formation of mutagenic heterocyclic amines in a model system. Carcinogenesis 14, 89–94.
996
Johansson, M.A., Fredholm, L., Bjerne, I., Jägerstad, M., 1995. Influence of frying fat on the formation of heterocyclic amines in fried beefburgers and pan residues. Food Chem. Toxicol. 33, 993–1004. Knize, M.G., Cunningham, P.L., Avila, J.R., Jones, A.L., Griffin, E.A., Felton, J.S., 1994a. Formation of mutagenic activity from amino acids heated at cooking temperatures. Food Chem. Toxicol. 32, 55–60. Knize, M.G., Dolbeare, F.A., Carroll, K.L., Moore, D.H., Felton, J.S., 1994b. Effect of cooking time and temperature on the heterocyclic amine content of fried beef patties. Food Chem. Toxicol. 32, 595–603. Larsson, S.C., Wolk, A., 2006. Meat consumption and risk of colorectal cancer: a meta-analysis of prospective studies. Int. J. Cancer. 119, 2657–2664. Layton, D.W., Bogen, K.T., Knize, M.G., Hatch, F.T., Johnson, V.M., Felton, J.S., 1995. Cancer risk of heterocyclic amines in cooked foods: an analysis and implications for research. Carcinogenesis 16, 39–52. Maron, M., Ames, B., 1983. Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173–215. Miller, A.J., Buchanan, R.L., 1983. Reduction of mutagen formation in cooked nitrite-free bacon by selected cooking treatments. J. Food Sci. 48, 1772–1775. Nilsson, L., Övervik, E., Fredholm, L., Levin, O., Nord, C., Gustafsson, JA., 1986. Influence of frying fat on mutagenic activity in lean pork meat. Mutat. Res. 171, 115–121. Norat, T., Lukanova, A., Ferrari, P., Riboli, E., 2002. Meat consumption and colorectal cancer risk: dose-response meta-analysis of epidemiological studies. Int. J. Cancer. 98, 241–256. NTP report on carcinogens, 11th edition. US Department of Health and Human Services, Public Health Service, National Toxicology Program. pp. 135–138.
SECTION | II
Cancer
Pariza, M.W., Ashoor, S.H., Chu, F.S., Lund, D.B., 1979. Effects of temperature and time on mutagen formation in pan-fried hamburger. Cancer Lett. 7, 63–69. Pérez, C., López de Cerain, A., Bello, J., 2002. Modulation of mutagenic activity in meat samples after deep-frying in vegetable oils. Mutagenesis 17, 63–66. Reuterswärd, A.L., Skog, K., Jägerstad, M., 1987. Mutagenicity of pan-fried bovine tissue in relation to their content of creatine, creatinine, monosaccharides and free amino acids. Food Chem. Toxicol. 25, 755–762. Shibamoto, T., Bjeldanes, L.F., 1993. Toxicants formed during food processing. In: Taylor, S.L. (ed.) Introduction to Food Toxicology. Academic Press Inc., pp. 183–198. Sinha, R., Rothman, N., Salmon, C.P., Knize, M.G., Brown, E.D., Swanson, C.A., Rhodes, D., Rossi, S., Felton, J.S., Levander, O.A., 1998. Heterocyclic amine content in beef cooked by different methods to varying degrees of doneness and gravy made from meat dripping. Food Chem. Toxicol. 36, 279–287. Spingarn, N.E., Weisburger, J.H., 1979. Formation of mutagens in cooked foods. I. Beef. Cancer Lett. 7, 259–264. Sugimura, T., Nagao, M., Honda, M., Yahagi, T., Seino, Y., Sato, S., Matsukura, N., Matsushima, T., Shirai, A., Sawamura, M., Matsumoto, H., 1977. Mutagen-carcinogens in food, with special reference to highly mutagenic pyrolytic products in broiled foods. In: Hiatt, H.H., Watson, J.D., Winsten, J.A. (eds) Origins of Human Cancer, pp. 1561–1577. Sugimura, T., 1997. Overview of carcinogenic heterocyclic amines. Mutat. Res. 376, 211–219. Wakabayasi, K., Nagao, M., Esumi, H., Sugimura, T., 1992. Food-derived mutagens and carcinogens. Cancer Res. 52, 2092–2098.
Chapter 107
Azoxymethane-induced Colon Carcinogenesis through Wnt/beta-catenin Signaling and the Effects of Olive Oil Takehiro Fujise, Ryuichi Iwakiri, Ryosuke Shiraishi, Bin Wu and Kazuma Fujimoto Department of Internal Medicine and Gastrointestinal Endoscopy, Saga Medical School, Japan
107.1 INTRODUCTION Colon cancer is one of the leading causes of cancer death in both men and women in Western countries, and in recent years, it has increasingly become a major cause of cancer mortality in Oriental countries, including Japan. These changes are associated with the westernization of Japanese dietary habits, which involves high consumption of meat and fat, together with low consumption of fruit, vegetables, vitamins and fibers, compared with classical Japanese diets. In fact, many epidemiological studies have shown a positive relationship between dietary fat consumption and colorectal cancer. Cancer results from the accumulation of multiple independent genetic alterations. In colon cancer, these genetic changes affect colon epithelial cell proliferation, differentiation and apoptosis. In familial adenomatous polyposis (FAP), which is one of the hereditary forms of colon cancer, a gene has been identified and its protein is a key molecule in the Wnt signaling pathway. Many experiments have shown that the Wnt signaling pathway plays a crucial role in the etiology of colon cancer, including hereditary and sporadic forms. In this chapter, we focus on dysregulation of colonic mucosal homeostasis in carcinogen-induced colon cancer models and the effects of dietary fatty acid.
107.2 WNT SIGNALING PATHWAY AND COLON CANCER The intestinal tract is characterized by rapid epithelial cell turnover, which continues throughout its life. This process involves the crypts and their associated villi, and is maintained and strictly regulated by stem cells, which give rise Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
to all the intestinal epithelial cell lineages: enterocytes and absorptive and secretory cells. The position along the crypt– villous axis defines the various stages in the life of an intestinal epithelial cell. The Wnt signaling pathway regulates a wide variety of processes in embryonic development and adult homeostasis, including cell proliferation, morphology, motility, and cell fate at the cellular level (Reya and Clevers, 2005), and plays a critical role in the development of the gastrointestinal tract. The key molecules in this pathway are a multiprotein scaffold that consists of β-catenin, glycogen synthase kinase-3β (GSK3β), axin, and adenomatous polyposis coli (APC). The Wnt signaling pathway begins with the Wnt ligand, one of a family of at least 16 members in mammals (Cadigan and Nusse, 1997). Wnts are secreted as growth factors that act through the cell surface Frizzled (Fz) receptor to initiate the signaling cascade (Bhanot et al., 1996; He et al., 1997). Disheveled is activated upon ligand binding to Fz, which causes inhibition of GSK3β. Inhibition of GSK3β activity prevents phosphorylation of β-catenin, thus blocking APC and axin-mediated degradation of β-catenin. β-Catenin is stabilized and translocated to the nucleus to bind members of the lymphoid enhancer factor/T cell factor (LEF/TCF) family of transcription factors and induce target gene expression (Barker et al., 2000; Giles et al., 2003). Targets of Wnt-signaling-regulated transcription include the protooncogene myc (He et al., 1998), cyclooxgenase 2 (Howe et al., 1999), matrilysin (Crawford et al., 1999), cyclin D1 (Tetsu and McCormick, 1999), and a member of the LEF/TCF family (Roose et al., 1999). APC was originally identified as an intestinal tumor suppressor gene by genetic analysis in patients with FAP (Groden et al., 1991). Hereditary forms of colorectal cancer and ⬃85% of all sporadic colorectal cancers have loss of APC function (Kinzler and Vogelstein, 1996). It is commonly accepted
997
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
998
that the crucial tumor suppressor role of APC lies in its ability to destabilize cytoplasmic free β-catenin (von Kries et al., 2000). Therefore, disrupted regulation of the Wnt signaling pathway plays a central role in the etiology of colon carcinogenesis (Liu et al., 2000; von Kries et al., 2000).
107.3 CARCINOGEN-INDUCED COLON CANCER MODEL AND ALTERATION IN SIGNALING PATHWAY ASSOCIATED WITH COLON MUCOSAL HOMEOSTASIS It is important that reliable animal models are available that demonstrate similarity to human disease in studies of various types of cancer. Elucidation of the mechanisms of colon carcinogenesis and evaluation of the effect of chemopreventive agents have been carried out using several animal models. Azoxymethane (AOM)-induced colon cancer is one of the major models. AOM is a very potent carcinogen that induces colorectal cancer with a high incidence in rats and mice. Aberrant crypt foci (ACF) induced by treatment with AOM in rodents can be used as biomarkers in shortterm experiments. ACF were first identified in the colon mucosa of rodents exposed to carcinogens (Bird, 1987), and have also been confirmed to be present in the human colon (Noda et al., 1998). ACF are regarded as preneoplastic or precancerous lesions in the colorectum of humans and rodents (Bird, 1987; Noda et al., 1998). The number of crypt/foci has been shown to increase with time following carcinogen treatment, and ACF demonstrate increased cell proliferation in rodents (Pretow et al., 1992, 1994). Thus, the formation and growth of these putative preneoplastic lesions are thought to be valuable indices of the effects of carcinogens and agents, promoting or preventing carcinogenesis in the colon cancer. Colon carcinogenesis is known to be a multistep process that involves multiple genetic alterations to K-ras. APC, DCC and P53. In AOM-induced colon cancer model of rats and mice, K-ras mutation has frequently been seen (70%) in hyperplastic ACF. β-Catenin mutation has also been found frequently (66%) in dysplastic ACF and alteration of the cellular localization of β-catenin has been observed in all dysplastic ACF (Boone et al., 1992). These studies have shown that there is a genetic alteration associated with Wnt/β-catenin signaling and these changes may play an important role in AOM-induced colon carcinogenesis. One previous study has evaluated expression of Wnts, Bone morphogenic protein (BMP), and Fz in human tissues and colon cancer cell lines, by using in situ hybridization (Holcombe et al., 2002). Wnt 2 has been detected in colon cancer but was undetectable in normal colonic mucosa. Differential expression of Wnt 5a with increased expression at the bottom of crypts compared to luminal villi is seen in normal colonic mucosa, whereas, in colon cancer tissue, Wnt 5a expression is increased. The expression of other Wnts (1, 4, 5b, 6, 7b and 10b) does not show any marked
SECTION | II
Cancer
difference between normal colonic mucosa and colon cancer tissue. In poorly differentiated adenocarcinoma, a high degree of Fz receptor expression was seen compared with no expression in normal colonic mucosa and well-differentiated adenocarcinoma tissue. These results indicate that Wnt ligands and Fz receptor, distinct from alterations of key molecules such as APC and β-catenin in Wnt/β-catenin signaling pathways, are integrally associated with colon carcinogenesis. They also suggest that Wnt 2 and 5a play an important role in the progression from normal to neoplastic colonic mucosa, and that Fz receptor is involved in processes associated with cancer invasion. Another study has shown that Wnt 2 expression and cytoplasmic/nuclear β-catenin accumulation are seen in most gastric cancers, irrespective of morphological phenotype, and that Wnt 2 up-regulation and β-catenin nuclear translocation signaling may play an important role in cancer formation, invasion and dissemination (Cheng et al., 2005). These studies indicate that disruption of Wnt/β-catenin signaling is one of the mechanisms of carcinogenesis in AOM-induced colon cancer models.
107.4 DIETARY FATTY ACID COMPOSITION AND COLON CANCER Apart from these genetic factors, environmental factors are thought to be involved in colon carcinogenesis, among which, dietary habits play a pivotal role. High consumption of meat and fat, together with low consumption of fruit, vegetables, vitamins and fiber, has been suggested to increase risk of colorectal cancer (Willett et al., 1990; Bostick et al., 2003; Giles et al., 2003; Riboli and Norat, 2003). Many epidemiological studies have revealed a positive relationship between dietary fat intake and colorectal cancer (Dai et al., 2002). Experimental studies have demonstrated that a highfat diet rich in n-6 polyunsaturated fatty acid (PUFA) and saturated fatty acids (SFAs) promotes colon carcinogenesis, particularly in post-initiation or promotional phases, and/or both (Rao et al., 2001; Dai et al., 2002; Wu et al., 2004); whereas diets rich in n-3 PUFAs and n-9 monounsaturated fatty acids (MUFAs) have been reported to inhibit colon carcinogenesis in both initiation and post-initiation phases (Bartoli et al., 2000a, b; Rao et al., 2001; Reddy, 2004). This supports epidemiological reports that show that an n-3 PUFA-rich diet suppresses the risk of colon cancer in humans (Caygill and Hill, 1995; Byers, 1996). These previous studies indicate that amount of fat intake and composition of ingested dietary fatty acids are important factors for colon carcinogenesis. Recently, we have investigated the effects of dietary intake of diverse fatty acids on colon carcinogenesis in an AOMinduced rat colon cancer model, and effects of various fatty acids on the Wnt signaling pathway (Fujise et al., 2007). Male Sprague-Dawley rats were given intraperitoneal injections of AOM once weekly for 2 weeks at a dose of 15 mg kg⫺1 body
CHAPTER | 107 Azoxymethane-induced Colon Carcinogenesis
weight, whereas control rats were given an equal volume of physiological saline. One day after the first AOM or saline treatment, rats designed for corn oil, olive oil, beef tallow, and fish oil diets, began to be fed with diets high in n-6 PUFAs, n-9 MUFAs, SFAs and n-3 PUFAs; whereas, one group continued to be fed with standard chow for 44 weeks. Twelve weeks after the last injection of AOM or saline, some rats were sacrificed, and colon ACF formation was analyzed. The remaining rats were sacrificed at 44 weeks, and tumors were examined. Colon mucosa was collected for further analysis. Normal-appearing colon mucosal proliferation and apoptosis were evaluated by Bromodeoxyuridine (BrdU) incorporation and percentage fragmented DNA, respectively. Expression of β-catenin, cyclin D1, Wnt2, Wnt3 and Wnt5a in normal-appearing colon mucosa was analyzed by Western blotting. Twelve weeks from the start of the experiment, colon ACF developed in all rats treated with AOM. Among AOM-treated rats, those fed with 10% corn oil and 10% beef tallow had significantly greater numbers of ACF per colon compared to rats fed with standard chow (standard chow group: 129 ⫾ 2.3/colon; corn oil group: 188 ⫾ 7.4/colon; beef tallow group: 206 ⫾ 4.0/colon; p ⬍ 0.05), whereas numbers of ACF per colon significantly decreased in AOMtreated rats fed with 10% olive oil and 10% fish oil compared to standard chow-fed rats (standard chow group: 129 ⫾ 2.3/ colon; olive oil group: 94 ⫾ 3.1/colon; fish oil group: 96 ⫾ 4.9/colon; p ⬍ 0.05). In contrast, few ACF were found in rats with saline treatment only. The number of multicrypt ACF, which were defined as those containing four or more aberrant crypts per focus, was significantly higher in rats fed with 10% corn oil and 10% beef tallow than in those fed with standard chow (p ⬍ 0.05). On the other hand, rats fed with 10% olive oil and 10% fish oil showed a significant decrease in multicrypt ACF compared to standard chow-fed rats (p ⬍ 0.05). At 44 weeks, no rat without AOM treatment developed colon adenoma or carcinoma. On the other hand, all rats fed with standard chow, 10% corn oil and 10% beef tallow diet developed colon cancer 44 weeks after the last injection of AOM (Figure 107.1). Two of six rats fed with 10% olive oil did not develop colon cancer, and four of six rats fed with 10% fish oil did not develop colon cancer. The number of colon tumors per rat was significantly higher in rats fed with 10% corn oil and 10% beef tallow than in those fed with standard chow (standard chow group: 2.3 ⫾ 0.5/ colon; corn oil group: 4.3 ⫾ 0.5/colon; beef tallow group: 5.2 ⫾ 0.5/colon; p ⬍ 0.05), whereas tumor multiplicity in the colon of rats fed with 10% fish oil was significantly lower than in those fed with standard chow (standard chow group: 2.3 ⫾ 0.5/colon; fish oil group: 0.5 ⫾ 0.2/colon; p ⬍ 0.05). The number of tumors in rats fed with 10% olive oil was smaller than that in standard chow-fed rats, but there was no significant difference between these two groups (olive oil group: 1.5 ⫾ 0.5/colon) (Figure 107.2). The number of BrdU-incorporated epithelial cells per crypt significantly increased in AOM-treated rats compared to saline-treated rats
999
FIGURE 107.1 Colon tumors were not observed in rats without AOM treatment throughout the study period. At 44 weeks, almost all tumors showed well-differentiated adenocarcinomas with polypoid growth.
at 44 weeks. In rats fed with 10% corn oil and 10% beef tallow, BrdU-incorporated cell number significantly increased compared to that in standard chow-fed rats regardless of AOM treatment (standard chow group: 2.7 ⫾ 0.15/crypt; corn oil group: 3.2 ⫾ 0.21/crypt; beef tallow group: 3.9 ⫾ 0.21/ crypt; p ⬍ 0.05; standard chow ⫹ AOM group: 4.5 ⫾ 0.22/ crypt; corn oil ⫹ AOM group: 6.4 ⫾ 0.15/crypt, beef tallow ⫹ AOM group: 6.9 ⫾ 0.23/crypt; p ⬍ 0.05). The number of BrdU-incorporated cells in rats fed with 10% olive oil and 10% fish oil significantly decreased (standard chow group: 2.7 ⫾ 0.15/crypt; olive oil group: 1.3 ⫾ 0.09/crypt; fish oil group: 1.8 ⫾ 0.21/crypt; p ⬍ 0.05; standard chow ⫹ AOM group: 4.5 ⫾ 0.22/crypt; olive oil ⫹ AOM group: 1.9 ⫾ 0.15/crypt; fish oil group: 3.7 ⫾ 0.10/crypt; p ⬍ 0.05) (Figure 107.3A). BrdU-incorporated epithelial cells were normally observed at the base of crypts. Among rats fed with 10% corn oil and 10% beef tallow, BrdU-positive cells spread to the upper portion of crypts even without AOM treatment. BrdU-incorporated cells were often observed at the top of crypts in rats fed with 10% beef tallow. Changes in distribution of BrdU-positive cells were more evident in rats treated with AOM. Many epidemiological and experimental studies have demonstrated that the amount and type of dietary fat play an important role in colon carcinogenesis (Reddy, 2004; Roynette et al., 2004), while controversy still exists regarding the influence of dietary fat on colon tumorigenesis. Our study showed that any type of high-fat diet did not result in the development of ACF or colon tumors per se. With AOM treatment, both 10% corn oil and 10% beef tallow diets significantly enhanced numbers of ACF and crypt multiplicity of foci 12 weeks after AOM treatment. Furthermore, all rats fed with standard chow, 10% corn oil or 10% beef tallow developed colon tumors at 44 weeks, and the number of tumors significantly increased in rats fed with 10% corn oil and 10% beef tallow compared to rats fed
1000
SECTION | II
6
* * * BrdU incorporated cell number
7 6
AOM(–) AOM(+)
* * *
* * *
**
**
5
*
4
**
*
3 2
*
**
*
1
A
Standard chow
Corn oil
Olive oil
Beef tallow
Fish oil
* * *
* * *
* * *
* * *
* * *
10 9 8
*
AOM(–) AOM(+)
7
*
6 5
**
**
4 3
**
2
**
1 0 B
4 3 2 1
*
0 Corn oil
* * *
0
*
Standard chow
* * *
8
*
5 Tumor number/colon
at the base of crypts in the colon mucosa. In colon carcinogenesis, two main hypotheses have been proposed for morphogenesis of colon tumor: bottom-up and top-down morphogenesis. Development of human adenomatous polyps is believed to proceed through the top-down mechanism, in which genetically altered cells in the superficial portions of the mucosa spread laterally, and downward to form new
% Fragmented DNA
with standard chow. These results indicated that dietary intake of corn oil rich in n-6 PUFAs and beef tallow rich in SFAs promoted colon carcinogenesis in AOM-treated rats. In contrast to corn oil and beef tallow, olive oil rich in n-9 MUFAs and fish oil rich in n-3 PUFAs ameliorated AOMinduced ACF formation and colon carcinogenesis, compared to rats fed with standard chow. Our results were supported by previous studies that have shown that a high-fat diet rich in n-6 PUFA and SFA promotes colon tumorigenesis, particularly in post-initiation and/or promotional phases in rodents (Dai et al., 2002; Wu et al., 2004), whereas a high-fat diet rich in n-9 MUFA and n-3 PUFA inhibits colon carcinogenesis in both post-initiation and/or promotional phases (Bartoli et al., 2000; Rao et al., 2001; Reddy, 2004). Many previous studies have evaluated cancerous tissues to analyze the effects of dietary fat (Bartoli et al., 2000; Rao et al., 2001; Dai et al., 2002; Reddy, 2004; Wu et al., 2004). Our study focused on background characteristics of the colonic mucosa after long-term feeding of different types of dietary fat, and assessed colonic epithelial proliferation in normal-appearing colon mucosa in colon tumorigenesis. Few studies have investigated the effects of dietary fatty acid composition on normal-appearing colon mucosa. BrdU incorporation in normal-appearing colon mucosa surrounding colon tumors significantly increased in rats fed with 10% corn oil and 10% beef tallow, but decreased in those fed with 10% olive oil and 10% fish oil diets. These results showed that long-term feeding of 10% corn oil and 10% beef tallow accelerated the proliferation potential of the colonic mucosal epithelium, which might have promoted colon carcinogenesis after AOM treatment. Our study also showed that the range of BrdU-positive cells spread to the upper portion of crypts in rats fed with corn oil and beef tallow. Commonly, stem cells are located
Cancer
Olive oil
Beef tallow
Fish oil
FIGURE 107.2 Tumor number in rats treated with AOM. No rat without AOM treatment showed colon tumors, including adenoma and carcinoma. Numbers of colon tumors per rat treated with AOM were significantly higher in rats fed with 10% corn oil and 10% beef tallow compared to rats fed with standard chow, whereas tumor multiplicity of the colon in rats fed with 10% fish oil diet was significantly lower than that of rats fed with standard chow. *p ⬍ 0.05.
Standard chow
Corn oil
Olive oil
Beef tallow
Fish oil
FIGURE 107.3 Effect of dietary fatty acid components on BrdU incorporation and DNA fragmentation in the rat colon mucosa. (A) The number of BrdU-incorporated epithelial cells per crypt significantly increased in AOM-treated rats compared to rats treated with saline 44 weeks after initial treatment. Among with or without AOM-treated rats, BrdU-incorporated cell number significantly increased in rats fed with 10% corn oil and 10% beef tallow diets compared to standard chow-fed rats, and significantly decreased in rats fed with 10% olive oil and 10% fish oil. *p ⬍ 0.05 compared to saline-treated rats fed with standard chow. **p ⬍ 0.05 compared to AOM-treated rats fed with standard chow. ***p ⬍ 0.05 compared to rats treated with saline. (B) Following AOM treatment, percentage of fragmented DNA was significantly inhibited in the colon mucosa among all dietary groups. In rats fed with 10% olive oil and 10% fish oil, % fragmented DNA was significantly higher compared to that of rats fed with standard chow regardless of AOM treatment. *p ⬍ 0.05 compared to saline-treated rats fed with standard chow. **p ⬍ 0.05 compared to AOMtreated rats fed with standard chow. ***p ⬍ 0.05 compared to rats treated with saline.
CHAPTER | 107 Azoxymethane-induced Colon Carcinogenesis
crypts that connect to pre-existing normal crypts and eventually replace them (Shih et al., 2001; Wright and Poulson, 2002). Our study of BrdU-positive cells in upper portions of crypts indicated alterations in the distribution of proliferating cells in rats fed with 10% corn oil and 10% beef tallow. This suggests the involvement of top-down morphogenesis in colon carcinogenesis, which results from a high n-6 PUFA and SFA diet. In view of homeostasis of the crypt–villous axis in the colon mucosa, BMP and Hedgehog signaling without Wnt/β-catenin signaling play a crucial role, including cell proliferation, differentiation and apoptosis, which is generally regulated strictly by these signal transduction pathways (Radtke and Clevers, 2005; Zhang and Li, 2005). One previous study has demonstrated altered expression of BMP in cancerous tissue compared to normal colonic mucosa (Holcombe et al., 2002). Fatty acid composition may influence the crucial signal transduction pathways, which maintain homeostasis of the crypt–villous axis in the colon mucosa, including BMP signaling and Hedgehog signaling. Apoptosis in normal-appearing colon mucosa at 44 weeks after the last injection of AOM or saline was evaluated using a DNA fragmentation assay (Figure 107.3B). Following AOM treatment, the percentage of fragmented DNA was significantly inhibited in the colon mucosa of all dietary groups. In rats fed with 10% olive oil and 10% fish oil, percentage fragmented DNA was significantly higher than that in rats fed with standard chow, with or without AOM treatment (standard chow group: 3.4 ⫾ 0.30%; olive oil group: 6.0 ⫾ 0.57%; fish oil group: 8.7 ⫾ 0.53%; p ⬍ 0.05; standard chow ⫹ AOM group: 2.6 ⫾ 0.33%; olive oil ⫹ AOM group: 3.78 ⫾ 0.29%; fish oil group: 4.08 ⫾ 0.12%; p ⬍ 0.05). In our previous study, we have revealed that a higher risk for colon cancer with dietary corn oil rich in n-6 PUFA is attributed to reduced apoptosis in the colon mucosa, and this was associated with inhibition of the tumor suppressor gene p53-mediated mitochondria-dependent apoptotic pathway (Wu et al., 2004). This study demonstrated a significant decrease in apoptosis in rats fed with 10% corn oil and 10% beef tallow, and reduction of colon mucosal apoptosis by dietary corn oil and beef tallow might be the result of accelerated cell proliferation during colon tumorigenesis. Many studies have suggested that the chemopreventive effect of n-3 PUFA was partly due to increased mucosal apoptosis (Davidson et al., 2000). This study showed that there was an apparent suppression of colon carcinogenesis, together with increased mucosal apoptosis in rats treated with AOM, by dietary intake of olive oil rich in n-9 MUFA and fish oil rich in n-3 PUFA. Increased apoptosis might be one of the crucial factors involved in tumor inhibition. Other factors, including ornithine decarboxylase activity (Rao and Reddy, 1993), diacylglycerol (Jiang et al., 1996), COX-2 and iNOS (Narayanan et al., 2003), may contribute to reduced carcinogenesis, together with n-3 PUFAs and n-9 MUFAs.
1001
Expression of cyclin D1 increased in AOM-treated rats. This increase was more significant in rats fed with 10% corn oil and 10% beef tallow, with or without AOM treatment, with a similar pattern as the BrdU incorporation assay. Results of the β-catenin assay were similar to those with cyclin D1. Namely, accumulation of cytoplasmic βcatenin was significantly elevated in rats fed with 10% corn oil and 10% beef tallow, with or without AOM treatment. Thus, we investigated Wnt proteins that existed as upstream signals in the Wnt/β-catenin signaling pathway. Expression of Wnt 2 and Wnt 3 was significantly higher in rats fed with 10% beef tallow, with or without AOM treatment, than that in other dietary groups. In rats fed with 10% corn oil, expression of Wnt 5a significantly increased compared to that in other rats fed with other diets. Resent studies have indicated that the Wnt signaling pathway is required during stem cell homeostasis for normal progression of intestinal epithelial cells through the crypt–villous axis (Radtke and Clevers, 2005; Reya and Clevers, 2005), and dysregulation of the Wnt signaling pathway is observed in many cancer tissues (Radtke and Clevers, 2005; Reya and Clevers, 2005). This study evaluated expression of cyclin D1, a product gene in Wnt/β-catenin signaling, which is activated through accumulation of β-catenin in the cytosol. Expression of cyclin D1 increased in rats fed with 10% corn oil and 10% beef tallow, and significant accumulation of β-catenin in the cytosol was observed, which suggested that Wnt/β-catenin signaling was activated by dietary consumption of corn oil and beef tallow. Wnt expression, an upstream signal in the Wnt/β-catenin signaling pathway, increased in rats fed with corn oil; and Wnt 2 and 3 increased in rats fed with beef tallow in rat colon mucosa, regardless of AOM treatment. These results indicated that increased colon mucosal proliferation potential was, at least in part, attributed to activation of Wnt/β-catenin signaling. Several studies in humans have indicated that expression of Wnt genes is accelerated in colon carcinoma tissues compared to surrounding normalappearing mucosa, using in situ hybridization (Mohammed et al., 2001; Holcombe et al., 2002). In another study of human gastric cancer, co-existence of Wnt 2 up-regulation and β-catenin translocation was positively associated with lymph node metastasis (Cheng et al., 2005). In this study, it is noteworthy that up-regulation of Wnt signaling in the normal-appearing surrounding colon mucosa was observed in rats with long-term consumption of corn oil and beef tallow. Our study indicated that long-term intake of dietary SFA and n-6 PUFA accelerated colon carcinogenesis by increasing cell proliferation through up-regulating the Wnt/β-catenin signaling pathway in AOM-treated rats. In contrast, n-9 MUFA and n-3 PUFA had a suppressive effect on colon carcinogenesis through decreasing cell proliferation and increasing cell apoptosis in colon mucosa. These results indicate that dietary fatty acid composition might be an important factor for modulation of mucosal proliferation potential and apoptosis.
1002
Many studies have investigated the inhibitory effect of olive oil on colon carcinogenesis. Dietary intake of olive oil inhibited the formation of ACF in an AOM-induced rat colon cancer model, and decreased colon mucosal arachidonate concentration compared with dietary intake of corn oil. Furthermore, AOM treatment induced a significant increase in prostaglandin E2 (PGE2) formation, but no change was found in rats fed with olive oil (Bartoli et al., 2000). The results indicated that these inhibitory effects may be due to modulation of arachidonic acid metabolism and mucosal PGE2 synthesis (Bartoli et al., 2000). Other studies using IL-10 knockout mice have shown that an olive oil diet inhibits COX-2 immunostaining in colon mucosa and decreases the risk of neoplasia-associated chronic colitis. However, a fish oil diet, which generally shows a chemopreventive effect on colon carcinogenesis, encourages chronic colitis and colitis-associated premalignant changes (Hegazi et al., 2006). An olive oil diet has an additive inhibitory effect with sulindac. These effects were mediated by regulating COX-2 expression, which is involved in prostaglandin synthesis and caspase-3, which plays a critical role in apoptosis (Schwautz et al., 2004). Recently, olive oil polyphenols have shown a strong inhibitory effect on cancer cell proliferation, which has been linked to the induction of a G2/M phase cell cycle block in cancer cell lines. These cell cycle blocks are mediated by inhibition of p38 and cyclic AMP response element binding protein (CREB) phosphorylation, which leads to a downstream reduction in COX-2 expression (Corona et al., 2007). Other studies have indicated that olive oil extract induces cell apoptosis and cell cycle arrest (Fabiani et al., 2002; Fini et al., 2008). In conclusion, many studies have indicated that olive oil has chemopreventive effects in colon carcinogenesis, through inhibition of cell proliferation and induction of apoptosis and/or cell cycle arrest (Figure 107.4). Our studies have investigated the role of the Wnt/β-catenin
SECTION | II
Cancer
signaling pathway in the colon-cancer-promoting effect of dietary fatty acid in an AOM-induced rat colon cancer model. However, the detailed mechanism associated with the inhibition of cell proliferation has not been elucidated. Further investigations are warranted to fully evaluate the mechanism of inhibitory and promoting effects of dietary fatty acid composition in colon carcinogenesis.
SUMMARY POINTS ●
●
●
●
●
●
●
●
Many epidemiological studies have revealed a positive relationship between dietary fat intake and colorectal cancer. Many experiments have shown that the Wnt signaling pathway plays a crucial role in the etiology of colon cancer, including hereditary and sporadic forms. Disruption of Wnt/β-catenin signaling is one of the mechanisms of carcinogenesis in AOM-induced colon cancer models. Amount of fat intake and composition of ingested dietary fatty acids are important factors for colon carcinogenesis. Diets rich in n-3 PUFAs and n-9 monounsaturated fatty acids (MUFAs) have been reported to inhibit colon carcinogenesis in both initiation and post-initiation phases. Olive oil rich in n-9 MUFAs and fish oil rich in n-3 PUFAs ameliorated AOM-induced ACF formation and colon carcinogenesis, compared to rats fed with standard chow. There was an apparent suppression of colon carcinogenesis, together with increased mucosal apoptosis in rats treated with AOM, by dietary intake of olive oil rich in n-9 MUFA and fish oil rich in n-3 PUFA. Many studies have indicated that olive oil has chemopreventive effects in colon carcinogenesis, through inhibition of cell proliferation and induction of apoptosis and/or cell cycle arrest.
REFERENCES
FIGURE 107.4 Schematic design of chemopreventive effect of olive oil in colon carcinogenesis.
Barker, N., Morin, P.J., Clevers, H., 2000. The Yin-Yang of TCF/beta-catenin signaling. Adv. Cancer Res. 77, 1–24. Bartoli, R., Fernandez-Banares, F., Navarro, E., Castella, E., Mane, J., Alvarez, M., Pastor, C., Cabre, E., Gassull, M.A., 2000. Effect of olive oil on early and late events of colon carcinogenesis in rats: modulation of arachidonic acid metabolism and local prostaglandin E2 synthesis. Gut 46, 191–199. Bartoli, R., Femandes-Banares, F., Navarro, E., Castella, E., Mane, J., Alvarez, M., Pastor, C., Cabre, E., Gassull, M.A., 2000. Effect of olive oil on early and late events of colon carcinogenesis in rats: modulation of arachidonic acid metabolism and local prostaglandin E2 synthesis. Gut 46, 191–199. Bhanot, P., Brink, M., Samos, C.H., Hsieh, J.C., Wang, Y., Macke, J.P., Andrew, D., Nathans, J., Nusse, R., 1996. A new member of the frizzled
CHAPTER | 107 Azoxymethane-induced Colon Carcinogenesis
family from drosophila functions as a Wingless receptor. Nature 282, 225–230. Bird, R.P., 1987. Observation and quantification of aberrant crypts in the murine colon treated with a colon carcinogen: preliminary findings. Cancer Lett. 37, 147–151. Boone, C.W., Steele, V.E., Kelloff, G.J., 1992. Screening for chenopreventive (anticarcinogenic) compoumds in rodents. Mutat Res. 267, 251–255. Bostick, R.M., Potter, J.D., McKenzie, D.R., Sellers, T.A., Kushi, L.H., Steinmetz, K.A., Folson, A.R., 1993. Reduced risk of colon cancer with high intake of vitamin E: The Iowa Women’s Health Study. Cancer Res. 53, 4230–4237. Byers, T., 1996. Nutrition and cancer among American Indians and Alaska Natives. Cancer 78, 1612–1616. Cadigan, K.M., Nusse, R., 1997. Wnt signaling: a common theme in animal development. Genes. Dev. 11, 3286–3305. Caygill, C.P., Hill, M.J., 1995. Fish n-3 fatty acids and human colorectal and breast cancer mortality. Eur. J. Cancer. Prev. 4, 329–332. Cheng, X.X., Wang, Z.C., Chen, X.Y., Sun, Y., Kong, Q.Y., Liu, J., Li, H., 2005. Correlation of Wnt-2 expression and beta-catenin intracellular accumulation in Chinese gastric cancers: relevance with tumor dissemination. Cancer Lett. 223, 339–347. Corona, G., Deiana, M., Incani, A., Vauzour, D., Dessi, M.A., Spencer, J.P., 2007. Inhibition of p38/CREB phosphorylation and COX-2 expression by olive oil polyphenols underlies their anti-proliferative effects. Biochem. Biophys. Res. Commun. 362, 606–611. Crawford, H.C., Fingleton, B.M., Rudolph-Owen, L.A., Goss, K.J., Rubinfeld, B., Polakis, P., Matrisian, L.M., 1999. The metalloproteinase matrilysin is a target of beta-catenin transactivation in intestinal tumors. Oncogene 18, 2883–2891. Dai, W., Liu, T., Wang, Q., Rao, C.V., Reddy, B.S., 2002. Down-regulation of PLK3 gene expression by types and amount of dietary fat in rat colon tumors. Int. J. Oncol. 20, 121–126. Davidson, L., Brown, R., Chang, W., Morris, J., Wang, N., Carroll, R., Turner, N., Lupton, J., Chapkin, R., 2000. Morphodensitometric analysis of protein kinase C beta (II) expression in rat colon: modulation by diet and relation to in situ cell proliferation and apoptosis. Carcinogenesis 21, 1513–1519. Fabiani, R., De Bartolomeo, A., Rosignoli, P., Servili, M., Montedoro, G.F., Morozzi, G., 2002. Cancer chemoprevention by hydroxytyrosol isolated from virgin olive oil through G1 cell cycle and apoptosis. Eur. J. Cancer Prev. 11, 351–358. Fini, L., Hotchkiss, E., Fogliano, V., Graziani, G., Romano, M., De Vol, E.B., Qin, H., Selgrad, M., Boland, C.R., Ricciardiello, L., 2008. Chemopreventive properties of pinoresinol-rich olive oil involve a selective activation of ATM-p53 cascade in colon cancer cell lines. Carcinogenesis 29, 139–146. Fujise, T., Iwakiri, R., Kakimoto, T., Shiraishi, R., Sakata, Y., Wu, B., Tsunada, S., Ootani, A., Fujimoto, K., 2007. Long-term feeding of various fat diets modulates azoxymethane-induced colon carcinogenesis through Wnt/beta-catenin signalimg in rats. Am. J. Physiol. Gastrointest. Liver Physiol. 292, G1150–G1156. Giles, R.H., van Es, J.H., Clevers, H., 2003. Caught up in a Wnt storm; Wnt signaling in cancer. Biochim. Biophys. Acta 1653, 1–24. Groden, J., Thliveris, A., Samowitz, W., Carlson, M., Gelgert, L., Albertsen, H., Joslyn, G., Stevens, J., Spilio, L., Robertson, M., 1991. Identification and characterization of the familial adenomatous polyposis coli gene. Cell 66, 589–600. He, T.C., Sparks, A.B., Rago, C., Hermeking, H., Zawel, L., da Costa, L.T., Morin, P.J., Vogelstein, B., Kinzler, K.W., 1998. Identification of c-MYC as a target of the APC pathway. Science 281, 1438–1439.
1003
He, X., Saint-Jeannet, J.-P., Wang, Y., Nathans, J., Dawid, I., Varmus, H., 1997. A member of the Frizzled protein family mediating axis induction by Wnt-5a. Science 275, 1652–1654. Hegazi, R.A., Saad, R.S., Mady, H., Matarese, L.E., O’Keefe, S., Kandil, H.M., 2006. Dietary fatty acids modulate chronic colitis, colitis-associated colon neoplasia and COX-2 exoression in IL-10 knock out mice. Nutrition 22, 275–282. Holcombe, R.F., Marsh, J.L., Waterman, M.L., Lin, F., Milovanovic, T., Truong, T., 2002. Expression of Wnt ligands and frizzled receptors in colonic mucosa and in colon carcinoma. J. Clin. Pathol.: Mol. Pathol. 55, 220–226. Howe, L.R., Subbaramaiah, K., Chung, W.J., Dannenberg, A.J., Brown, A.M., 1999. Transcriptional activation of cyclooxygenase-2 in Wnt1-transformed mouse mammary epithelial cells. Cancer Res. 59, 1572–1577. Jiang, Y., Lupton, J., Chapkin, R., 1996. Dietary fat and fiber differentially alter intracellular second messengers during tumor development in rat colon. Carcinogenesis 17, 1227–1233. Kinzler, K.W., Vogelstein, B., 1996. Lessons from hereditary colorectal cancer. Cell 87, 159–170. Liu, W., Dong, X., Mai, M., Seelan, R.S., Taniguchi, K., Krishnandath, K.K., Halling, K.C., Cunningham, J.M., Boardman, L.A., Qian, C., Christensen, E., Schmidt, S.S., Roche, P.C., Smith, D. I., Thibodeau, S.N., 2000. Mutations in Axin2 cause colorectal cancer with detective mismatch repair by activating beta-catenin–Tcf signaling. Nat. Genet. 26, 146–147. Mohammed, I.M., Roczo, N., Phillips, W.A., Baindur-Hudson, S., 2001. Expression of Wnt genes in human colon cancers. Cancer Lett. 166, 185–191. Narayanan, B., Narayanan, N., Simi, B., Reddy, B.S., 2003. Modulation of inducible nitric oxide synthase and related proinflammatory genes by the omega-3 fatty acid docosahexaenoic acid in human colon cancer cell. Cancer Res. 63, 972–979. Noda, T., Iwakiri, R., Fujimoto, K., Matsuo, S., Aw, T.Y., 1998. Programmed cell death induced by ischemia-reperfusion in rat intestinal mucosa. Am. J. Physiol. 274, G270–G276. Pretow, T.P., Cheyer, C., O’Riordan, M.A., 1994. Aberrant crypt foci and colon tumors in F344 rats have similar increases in proliferative activity. Int. J. Cancer 56, 599–602. Pretow, T.P., O’Riordan, M.A., Somich, G.A., Amini, S.B., Pretlow, T.G., 1992. Aberrant crypts correlate with tumor incidence in F344 rats treated with azoxymethane and phytate. Carcinogenesis 13, 1509–1512. Radtke, F., Clevers, H., 2005. Self-renewal and cancer of the gut: Two sides of a coin. Science 307, 1904–1909. Rao, C., Reddy, B.S., 1993. Modulating effect of amount and types of dietary fat on ornithine decarboxylase, tyrosine protein kinase and prostaglandins production during colon carcinogenesis in F344 rats. Carcinogenesis 14, 1327–1333. Rao, C.V., Hirose, Y., Indranie, C., Reddy, B.S., 2001. Moduration of experimental colon tumorigenesis by types and amounts of dietary fatty acids. Cancer Res. 61, 1927–1933. Reddy, B.S., 2004. Omega-3 fatty acids in colorectal cancer prevention. Int. J. Cancer 112, 1–7. Reya, T., Clevers, H., 2005. Wnt signaling in stem cells and cancer. Nature 434, 843–850. Reya, T., Clevers, H., 2005. Wnt signaling in stem cells and cancer. Nature 434, 843–850. Riboli, E., Norat, T., 2003. Epidemiologic evidence of the protective effect of fruit and vegetables on cancer risk. Am. J. Clin. Nutr. 78, 559S–569S.
1004
Roose, J., Huls, G., van Beest, M., Moerer, P., van der Horn, K., Goldschmeding, R., Logtenberg, T., Clevers, H., 1999. Synergy between tumor suppressor APC and the beta-catenin-Tcf4 target Tcf1. Science 285, 1923–1926. Roynette, C.E., Calder, P.C., Dupertuis, Y.M., Pichard, C., 2004. n-3 polyunsaturated fatty acids and colon cancer prevention. Clin. Nutr. 23, 139–151. Schwautz, B., Birk, Y., Raz, A., Madar, Z., 2004. Nutritional-pharmacological combinations. Eur. Nutr. 43, 221–229. Shih, I.M., Wang, T.L., Traverso, G., Romans, K., Hamilton, S.R., BenSasson, S., Kinzler, K.W., Vogelstein, B., 2001. Top-down morphogenesis of colorectal tumors. PNAS 27, 2640–2645. Tetsu, O., McCormick, F., 1999. Beta-catenin regulates expression of Cyclin D1 in colon carcinoma cells. Nature 398, 422–426. von kries, J.P., Winbeck, G., Asbrand, C., Schwarz-Romond, T., Sochnikowa, N., Dell’Oro, A., Behrens, J., Birchmeier, W., 2000.
SECTION | II
Cancer
Hot spots in beta-catenin for interactions with LEF-1, conduction and APC. Nat. Struct. Biol. 7, 800–807. Willett, W.C., Stampfer, M.J., Colditz, G.A., Rosner, B.A., Speizer, F.E., 1990. Relation of meat, fat, and fiber intake to the risk of colon cancer in a prospective study among women. N. Eng. J. Med. 323, 1664–1672. Wright, N.A., Poulson, R., 2002. Top down or bottom up? Competing management structures in the morphogenesis of colorectal neoplasm. Gut 51, 306–308. Wu, B., Iwakiri, R., Ootani, A., Tsunada, S., Fujise, T., Sakata, Y., Sakata, H., Toda, S., Fujimoto, K., 2004. Dietary corn oil promotes colon cancer by inhibiting mitochondria-dependent apoptosis in azoxymethane-treated rats. Exp. Biol. Med. 229, 1017–1025. Zhang, J., Li, L., 2005. BMP signaling and stem cell regulation. Dev. Biol. 284, 1–11.
Chapter 108
Olive Oil and its Phenolic Components and their Effects on Early- and Late-stage Events in Carcinogenesis Chris I.R. Gill 1, Yumi Z.H.-Y. Hashim2, Maurizio Servili3 and Ian R. Rowland4 1
Northern Ireland Centre for Food and Health (NICHE), University of Ulster (Coleraine), Coleraine, Northern Ireland, UK Department of Biotechnology Engineering, Kulliyyah of Engineering, International Islamic University Malaysia, Kuala Lumpur, Malaysia 3 Dipartmento di Scienze degli Alimenti, Sezione di Tecnologie e Biotecnologie degli Alimenti, Perugia, Italy 4 Department of Food and Nutritional Sciences, University of Reading, Whiteknights, Reading, UK
2
108.1 INTRODUCTION Olive oil is a key component in the Mediterranean-style diet (Stark and Madar, 2002). It has been recognized as having great potential to prevent the onset of oxidativedamage-associated diseases such as cancer, aging and cardiovascular problems. The Mediterranean diet, for example, is associated with lower incidence of colorectal cancer and it has been estimated that the incidence of colorectal cancer among the developed Western countries’ population could be reduced by 25% if they were to consume the Mediterranean-style diet (Trichopoulou et al., 2000). Apart from the high monounsaturated fatty acid content, squalene, vitamin E and phenolic compounds are also present in olive oil and have been suggested to have roles in modulating cancer risk; this area was reviewed in detail by Hashim et al. with respect to colorectal cancer who indicated a number of potential mechanisms through which olive oil and its components may exert an effect (Hashim et al., 2005). The hydrophilic phenols are the most abundant natural antioxidants of virgin olive oil (VOO), while the phenolic alcohols and acids, the major classes of hydrophilic phenols found in VOO, include secoiridoids, flavonoids and lignans. Secoiridoids, like the aglycon derivatives of oleuropein, demethyloleuropein and ligstroside, are the most abundant VOO phenolic antioxidants present in olive fruit. These phenolic compounds, in particular, may act as anticarcinogens through several mechanisms such as quenching or preventing the formation of reactive oxygen species, inhibiting arachidonic acid metabolism leading to reduced proinflammatory or mitogenic metabolites and modulating Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
cancer-related genes in favor of inhibition of carcinogenesis (Yang et al., 2001) (Figure 108.1).
108.2 CARCINOGENESIS, DNA DAMAGE AND METASTASIS Cancer is without question an important global public health problem. Each year approximately 10.1 million new cancer cases are diagnosed with a further 6.2 million people losing their lives worldwide. This disease accounts for a quarter of all deaths in countries with a westernized lifestyle. There is convincing evidence that genetic events are involved in the initiation, promotion and progression phases of carcinogenesis. One of the best examples is colorectal cancer, which is induced by a series of mutations and deletions in a number of critical oncogenes and tumor suppressor genes such as Apc, K-ras and p53 (Vogelstein et al., 1988). Genetic events in cancer encompass a range of genotoxic effects including DNA adducts, DNA damage, gene mutations, and cytogenetic alterations. Deficiencies in DNA repair processes can lead to genetic instability and subsequently to increased incidence of genetic lesions and elevated cancer risk. Arguably the most devastating stage of the carcinogenic process for most cancers is the metastatic phase of the disease progression with the spread of secondary tumors to distant organs which results in greatest mortality. For example, surgical resection in colorectal cancer is an effective treatment for localized disease, achieving a 5-year survival rate of 90% but treatments for metastatic disease remain ineffective (Saha et al., 2002).
1005
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1006
SECTION | II
Secoiridoids 2'
HO HO 4'
7'
1'
3'
6' O
9
10
7
8'
6' O
O COOH 3
7
5'
4
6
6' CH2OH O 5' OH 4' 1' O OH 2' 3' HO
6' CH2OH O 5' OH 4' 1' O OH 2' 3' HO 1 O
8
Demethyloleuropein
3
5
9
10
1 O
8
Oleuropein
HO 4'
3
5
9
10
6' O 5' CH2OH 4' OH 1' O 2' 3' OH HO
COOH
7'
1'
3'
4
6
1 O
8
O
5'
3
5
2' 8'
6' O
HO 4'
4
6
7'
1'
3'
O COOCH 3
7
5'
2'
HO
8'
Ligstroside
3
OH 4
COOCH3
2 1
8
5''
1''
O 7 OH 3'' 2'' OH
O
5
6
4 6'' O 7 5 3 OH O 9 1 O 4' 8
HO
O
10
3'
3
5' 6' OH
O
OH 2' OH
4 HO
6' 4' CH2OH O 1 7 8 9 O 7' 5' O O 2' 1' 2 3' 8' 1' 6' 5' OH O 1' H3C 6' HO 2' 4' HO 3' OH 6
OH 4'
O
Nüzhenide
2'
OH 3'
5'
4' OH
Verbascoside
Phenolic compounds 2' 2' 3' HO 4'
7'
1'
5'
7' 4
6 9
10
7'
8' 6' O O 4' HO 5' 7' 4 6 5 9 10 1 8
8' O COOCH3
6' O
1'
3'
3
5 1
O
8 OH
2' 3'
HO 4' 5'
1'
7' 9
8
2'
HO
8'
6
3'
O COOCH3
HO 4' 5'
4 3
5 1
8' 6'
HO 4'
O
Dialdialdehydic form of decarboxymethyl elenolic acid linked to p-HPEA (p-HPEA-EDA) = Oleochantal
7'
6' O
7'
1'
OH
5'
OH
Ligstroside aglycon (p-HPEA-EA)
HO
2' 3'
3
1'
7' 3'
O 7' 9
10
HO 4'
4
O
Lignans OH 5'
4' 3'
6'
4 H 6'' 5'' 4'' HO
O
5 6 1'' 2'' 3'' O
1'
2'
O
CH3
(+)-1-Acetoxypinoresinol
OH 4' 3'
5' 6'
O CH3
2 1 8
7'
6'
8' OH
5'
(3,4-Dihydroxyphenyl) ethanol (3,4-HPEA)
1 O
Dialdialdehydic from of decarboxymethyl enolic acid linked to 3,4-HPEA (3,4 DHPEA-EDA)
Oleuropein aglycon (3,4 DHPEA-EA)
1'
3
5
8
O
2'
HO
8'
6' O 6
O
(p-HYDROXYPHENYL) ETHANOL (p-HPEA)
4 OCCH3
H
O 6'' 5'' 4'' HO
2'
2
5 6 1'' 2'' 3'' O
1'
O 1 8
H
O
CH3
(+)-1-Pinoresinol
FIGURE 108.1 Chemical structure of secoiridoids, phenolics and lignans present in olive oil.
O CH3
Cancer
CHAPTER | 108 Olive Oil and its Phenolic Components and their Effects
108.3 OLIVE OIL, PHENOLICS AND DNA DAMAGE 108.3.1 DNA Damage and the Initiation of Cancer With regard to the initiation of the carcinogenesis process the interaction of an electrophile or free radical with DNA usually results in the formation of DNA adducts. Depending on the type of chemical involved, alkylated bases, oxidized bases (e.g. 8-oxo-deoxy-guanosine [8-oxo-dG]), bulky adducts, etheno- and propano-adducts can result (Veglia et al., 2003). One of the consequences of adduct formation is the breaking of a base into small residues or its removal from DNA creating an abasic site resulting in strand breaks. Strand breaks can also result from the enzymatic activity of endonucleases in apoptosis or during DNA repair (see below). DNA strand breaks can be assessed by the singlecell gel electrophoresis or Comet assay and have been detected in biopsies of tissue derived from laboratory animals treated with carcinogens and human subjects (PoolZobel et al., 1994, 1996). A number of specific repair pathways exist to correct damaged DNA including base excision repair, nucleotide excision repair and mismatch repair. Damage caused by methylation of guanine can be removed by O6-methylguanine methyltransferase, and cross-links and double-strand breaks can be repaired by homologous recombination (Hoeijmakers, 2001). The important role that these repair processes play in carcinogenesis is apparent from the elevated cancer risk in people with inherited defects in these pathways (Hoeijmakers, 2001).
108.3.2 Lymphocyte DNA Damage – a Biomarker of Anticancer Risk Mediterranean countries have lower rates of colorectal cancer compared to other Western countries (Trichopoulou et al., 2000) and this has been attributed to a number of factors, one of which is olive oil (Serra-Majem et al., 2003). While epidemiological associations provide indications that olive oil may decrease cancer risk, they have their limitations in that they do not establish direct causation, as there could be other non-dietary (health and lifestyle) factors linked to such observations. In order to provide evidence for a reduction of cancer risk, it is essential to carry out human intervention studies with foods or their components, thought to decrease cancer risk, to substantiate whether these dietary factors can indeed reduce cancer incidence. However, using cancer as the endpoint in dietary interventions is time-consuming, costly and impractical. Hence, biomarkers or intermediate endpoints of cancer risk are used instead in order to establish any causal role(s) of foods/nutrients/dietary components in cancer prevention. Several diet-related biomarkers are currently being used to
1007
determine cancer risk in humans (Rafter et al., 2004). These mainly included the use of non-invasive, surrogate markers that are more readily accessible and cause least discomfort in volunteers taking part in human nutrition studies. In particular, measurement of lymphocyte DNA damage has been extensively used in dietary intervention trials as a surrogate marker of cancer risk with studies reporting the ability of single dietary components containing various mixtures of phytochemicals (e.g. simple phenols, flavonols, catechins, tannins) to decrease DNA damage in lymphocytes or to increase their ability to resist damage (Riso et al., 2005; Gill et al., 2007). This surrogate marker has largely been used as a biomarker of exposure to mutagenic and antimutagenic dietary components. However, whether changes in surrogate markers truly reflect the events in the target tissues is still a matter of debate.
108.3.3 Olive Oil Phenolics Decrease DNA Damage: Evidence in Different Experimental Systems The ability for olive oil or its components to decrease the amount of DNA damage lends itself to the concept of risk reduction and a reduced lifetime risk. It is a seductively simple approach to underpinning epidemiological observations made regarding olive oil and cancer incidence. But the scope of the literature on this topic is limited, a mixture of in vitro models using olive oil phenolics and small olive oil biomarker-based studies in humans (Table 108.1). Although these studies in general show antigenotoxic effects, they should be considered as preliminary and indicative only. The amount of dietary olive oil intake has been reported to reach 50 g day⫺1, which corresponds to approximately 25 mg of total phenolics (Ferro-Luzzi and Sette, 1989). It has been shown from ileostomy studies that 55–73% of ingested olive oil phenols intake are absorbed (Vissers et al., 2002). Another study reported that 100% hydroxytyrosol is absorbed in small intestine (Manna et al., 2000) while hydroxytyrosol and tyrosol were shown to be dosedependently absorbed in humans and excreted in urine as glucuronide conjugates (Visioli et al., 2000). Consequently it is probable that phenolic concentrations of up to 50 μM and possibly 100 μM may be physiologically relevant amounts to test within in vitro systems. We have reported an antigenotoxic activity (significant linear trend) for a virgin olive oil phenolic extract on the colonic cell line HT29 using the Comet assay (Gill et al., 2005). The extract was tested at a range 0–100 μg mL⫺1 and cells were pre-treated for 24 h before exposure to a hydrogen peroxide challenge (75 μM). A reduction of approximately 25% in induced DNA damage was reported at the maximal concentration of 100 μg mL⫺1 olive oil extract, which contained approximately 5.2 μM hydroxytyrosol, 3 μM tyrosol and 20 μM pinoresinol. Similarly another
1008
SECTION | II
Cancer
TABLE 108.1 Effect of olive oil and olive oil phenolics on DNA damage. Study type
Treatment/exposure conditions
DNA damage marker
Effects
Reference
In vitro – PC3 cells (prostate)
Hydroxytyrosol, tyrosol, caffeic acid, 24 hour (0–250 μM)
Comet assay with H2O2 challenge (60 μM)
Dose–response decrease in DNA damage
Quiles et al., 2002
In vitro – HT29 cells (colonic)
Olive oil phenolic extract 24 hour (0–100 μg mL⫺1)
Comet assay with H2O2 challenge (75 μM)
Decrease in DNA damage
Gill et al., 2005
In vitro – Jurkat cells (T-lymphocytic)
Olive oil extract 30 min (0–500 μg mL⫺1)
Comet assay with glucose oxidase generated H2O2 challenge (12 μM min⫺1)
Dose–response decrease in DNA damage
Nousis et al., 2005
Ex vivo – human whole blood cells
Hydroxytyrosol, lipophilic derivatives, 20 min (50 μM)
Atypical comet assay with H2O2 challenge (200 μM)
Decrease in DNA damage
Grasso et al., 2007
Intervention – 12 healthy men
Olive oil high, medium, low phenol content, 25 mL for 4 days
8-oxo-dG adducts in lymphocyte, 8-oxo-dG urine excretion
Decrease adducts with increased olive oil phenolics
Weinbrenner et al., 2004
Intervention – 10 postmenopausal women
Olive oil high vs low phenolic content, 50 g day⫺1, 8 weeks
Lymphocyte comet assay
Decrease DNA damage with increased phenolics
Salvini et al., 2006
Intervention –28 healthy men
Olive oil high, medium, low phenol content, 25 mL day⫺1, 3 weeks
Etheno-DNA adduct excretion in urine
No effect
Hillestrom et al., 2006
Intervention –182 healthy males
Olive oil high, medium, low phenol content, 25 mL day⫺1, 3 weeks
8-oxo-dG adducts excretion in urine
Decreased DNA oxidation for olive oil irrespective of phenol content
Machowetz et al., 2007
This table summarizes the outcomes of cell-based and human intervention studies examining the effect of olive oil and/or its phenolic components on DNA damage. (H2O2: Hydrogen peroxide; 8-oxo-dG: 8-oxo-deoxy-guanosine).
Comet assay study (Nousis et al., 2005) reported that an extract from olive oil caused a significant dose–response reduction (0–500 μg mL⫺1) in DNA damage in Jurkat cells (a T-lymphocyte cell line), while an ex vivo study by Grasso and colleagues using normal human cells (whole blood cells) also reported effects similar to those observed in the aforementioned in vitro studies (Grasso et al., 2007). Currently only one human intervention study has been undertaken which has assessed the effects of olive oil consumption on DNA damage in human subjects using the Comet assay. The study by Salivini et al. in ten postmenopausal women (47–67 years) was a randomized crossover trial; the subjects consumed 50 g day⫺1 of either an extra virgin olive oil with a high phenolic (591.8 mg total phenols kg⫺1 of which total hydroxytyrosol accounted for 157.3 mg kg⫺1) or low phenolic content (147.3 mg total phenols kg⫺1 of which total hydroxytyrosol accounted for
15.4. mg kg⫺1) for an 8-week period with 8-week washout periods. The average level of DNA damage (% tail DNA) at baseline was 11.9 ⫹/⫺ 7.67 compared to 5.6 ⫹/⫺ 5.1 and 6.5 ⫹/⫺ 5.1 for the high and low treatments respectively. The results indicated that average lymphocyte DNA damage (four timepoints) was 30% lower in the high phenolic olive oil treatment as compared to the low phenolic olive oil treatment (p ⫽ 0.02) (Salvini et al., 2006). Although the sample size is small, subjects received a reasonable duration of intervention with a relevant quantity of oil, and the study demonstrated a reduction in DNA damage in lymphocytes, more noticeable in the olive oil with a high total phenolic and hydroxytyrosol content. The remaining studies assessing the effect of olive oil on human health assess DNA damage by DNA adduct excretion or incorporation. Hydroxylation of guanine in the 8-position to 8-oxo-deoxy-guanosine (8-oxo-dG) is the most
CHAPTER | 108 Olive Oil and its Phenolic Components and their Effects
common form of DNA modification (Kasai, 1997) and the use of this adduct is considered a biomarker for DNA oxidative damage. The protective effect associated with the phenolic content of olive oil was reported in a short-term doubleblind, randomized, cross-over feeding study with 12 healthy males. The subjects consumed olive oils 25 mL day⫺1 with varying phenolics content (10 mg kg⫺1, 133 mg kg⫺1, 486 mg kg⫺1) for 4 days, preceded by 10-day washout periods. The maximum effect was observed for consumption of the high phenol content oil with a reduction of approximately 8-oxo-dG in mitochondrial DNA (⫺49.2%, p ⬍ 0.01) and in urine (⫺51.67%, p ⬍ 0.01). In general, consumption of olive oils reduced 8-oxo-dG in urine and mitochondrial DNA in a dose-dependent manner in relation to the phenolic content of the olive oil administered (Weinbrenner et al., 2004). The most recent study on olive oil and DNA damage, EUROLIVE, utilized three virgin olive oils of varying total phenolic contents (366 mg kg⫺1, 164 mg kg⫺1, 2.7 mg kg⫺1): in a double-blind randomized cross-over trial, 25 mL day⫺1 of the olive oil was administered to 182 healthy males for a 3-week period with 2-week washout periods. A 13% reduction in urinary 8-oxo-dG (p ⫽ 0.008) was observed in relation to olive oil consumption, but there was no evidence of an enhanced effect with olive oil of a higher phenolic content (Machowetz et al., 2007). A subset analysis study on 28 subjects from the EUROLIVE study reported that olive oil consumption did not alter the excretion in urine of etheno-DNA adducts formed as a result of oxidative stress and lipid peroxidation (Hillestrom et al., 2006). It is clear that more studies need to be undertaken examining the relationship between olive oil consumption and DNA damage, but given the data from the existing human interventions they generally show an antigenotoxic activity which, when taken in the context of the existing experimental data is suggestive of a protective effect, given the apparent consistency of effect from in vitro to in vivo studies. But again it must be stressed that these few studies in humans use biomarkers and not cancer as an endpoint and can therefore only be suggestive of a potential anticancer activity through a mechanism associated with reduction in DNA damage.
108.4 OLIVE OIL, PHENOLICS AND METASTASIS-RELATED EVENTS 108.4.1 Invasion and Metastasis: A Sequence of Events Metastasis can be considered as a cascade of interrelated sequential steps. Cells must be able to disseminate from the primary tumor, invade the surrounding tissue, enter the circulatory system, evade immune responses, arrest at and colonize a distant site (MacDonald et al., 1993). A three-step hypothesis has been proposed to describe invasion of tumor cells: attachment to basement membrane or extracellular matrices
1009
FIGURE 108.2 Effects of an olive oil phenolic extract (OVP) and its corresponding isolated phenolics on HT115 cell invasion. The graph describes the effects of OVP treatment (24 h) and its corresponding isolated phenolics (DHPEA: hydroxytyrosol; HPEA: tyrosol, PINO: pinoresinol) on HT115 cell invasion and total cell number in Matrigel invasion assay. Results are expressed as the mean of three independent experiments performed in duplicate. Values ⫽ mean ⫾ SEM. *p ⬍ 0.05. (Adapted from Hashim et al., 2008.)
(ECM), protease activity that induces local degradation of the matrix, and migration of tumor cells through the modified matrix (MacDonald et al., 1993). Integrins mediate tumor cell attachment to ECM components such as laminin, fibronectin, vitronectin and collagens as well as modulating proteolytic enzymes, intracellular signaling pathways governing the cytoskeletal organization and gene expression (Hood and Cheresh, 2002). Inhibition of invasion constitutes a new class of targets for chemoprevention in which intervention could commence between the period of tumor proliferation and the onset of invasion, preventing the series of events leading to metastasis (Woodhouse et al., 1997).
108.4.2 Olive Oil Phenolics Inhibit Invasion and Adhesion of Colon Cancer Cells Olive oil phenolics have been shown to exert potential anticancer effects in a number of studies (reviewed in Wahle et al., 2004; Hashim et al., 2005). Our previous study demonstrated that olive oil phenolics have the abilities to inhibit initiation, promotion and invasion events in in vitro models of colorectal carcinogenesis (Gill et al., 2005). A further paper focused on processes related to metastasis, namely invasion and adhesion (Hashim et al., 2008). In this study, we reported different dose-related anti-invasive effects of olive oil phenolics on HT115 human colon carcinoma cells using the Matrigel invasion assay. The compounds tested included a phenolic extract from virgin olive oil (OVP) and its main constituents: hydroxytyrosol, tyrosol, pinoresinol and caffeic acid. At 25 μg mL⫺1 OVP and equivalent doses of the individual compounds inhibited the invasive capacity of HT115 cells in a range between 45–55% of control. It was reported that OVP, but not the isolated phenolics, significantly reduced total cell number in the Matrigel invasion assay without affecting cell viability; this indicated the reduction of cell number in the Matrigel invasion assay was not due to cytotoxicity (Figure 108.2). Rather, OVP when
1010
SECTION | II
Cancer
FIGURE 108.3 Effects of an olive oil phenolic extract (OVP) and its corresponding isolated phenolics on HT115 cell adhesion to extracellular matrix constituents. The graph describes the effects of hydroxytyrosol (DHPEA) and OVP co-treatment on HT115 cell adhesion to specific extracellular matrix protein. Results are expressed as the mean of six independent experiments for co-treatment (only). Values ⫽ mean ⫾ SEM. *p ⬍ 0.05 (GLM Univariate and Kruskal-Wallis with Mann-Whitney post-hoc test for co-treatment). Abs: absorbance. (Adapted from Hashim et al., 2008.)
FIGURE 108.4 Effects of an olive oil phenolic extract (OVP) and its corresponding isolated phenolics on HT115 cell adhesion under two exposure regimens. The graph describes the effects of hydroxytyrosol (DHPEA) and OVP pre-treatment (pre-tx) and pre-treatment with additional co-treatment (co-tx) on HT115 cell adhesion to specific extracellular matrix protein. Results are expressed as the mean of three independent experiments for pretreatment and co-treatment study. Values ⫽ mean ⫾ SEM. *p ⬍ 0.05 (one-tailed independent samples t-test against control in each treatment group for pre-treatment and co-treatment study). Abs ⫽ absorbance. (Adapted from Hashim et al., 2008.)
administered as a co-treatment with cells in a Matrigelcoated well altered the adhesion properties of the cells in relation to the ECM, as no significant effects on HT115 cell attachment were observed when the ECM coating was removed and a plastic substrate used in its place. These
results indicate the importance of ECM in modulating the anti-invasive effects of OVP. When OVP treatment was applied to HT115 cells that had already been allowed to adhere to the Matrigel layer (pre-treatment with OVP) no significant reduction in total cell number occurred unlike
CHAPTER | 108 Olive Oil and its Phenolic Components and their Effects
the case for OVP added to adhering cells (co-treatment). This further suggests that the OVP may affect the integrin– ECM interaction of adhering cells but not causing detachment of adherent cells. Treatment of adhering cells with OVP extract showed significant decrease of adhesion towards collagen type IV both in co-treatment (p ⫽ 0.004) study (Figure 108.3) and in pre-treatment with co-treatment (p ⫽ 0.0465) study while no significant effects was observed in pre-treatment (only) study (Figure 108.4). Hydroxytyrosol showed no significant effects on adhesion towards specific ECM, confirming the absence of detachment shown in invasion assay. Other isolated phenolics (tyrosol, pinoresinol and caffeic acid) tested at 0–25 μM (co-treatment) showed no effects on cell adhesion towards ECM. Following this observation, a possible mechanism by which OVP exerts its anti-invasion and anti-attachment effect may be via interrupting integrin–ECM interaction. In this context, caffeic acid phenethyl ester was shown to regulate integrin-mediated signaling in human colon carcinoma cells (Weyant et al., 2000).
phenolic compounds from green tea have reported just such an inhibition of MMPs (Demeule et al., 2000; Annabi et al., 2002; Adhami et al., 2003).
108.5 CONCLUSIONS Epidemiological studies suggest that there is a link between olive oil consumption and decreased cancer risk. Experimental studies provide some evidence to support the suggestion that olive oil and its phenolics may beneficially modulate earlystage and late-stage events in the carcinogenic pathway.
SUMMARY POINTS ●
●
●
108.4.3 Anti-invasive Effects of Olive Oil Phenolics: Other Mechanisms In general, phenolic antioxidant actions have mainly been associated with free radical scavenging activity; modulation of cell signaling pathways and metal chelating properties (Soobrattee et al., 2005). From an invasion and metastasis perspective, reactive oxygen species are associated with induction of invasion-related genes and may also function as intracellular signaling molecules to favor a pro-invasive state (Nonaka et al., 1993; Brenneisen et al., 1997). Antioxidative activity from tea compounds (Zhang et al., 2000) and lipophilic ascorbic acid derivatives (Liu et al., 2006) have been shown to suppress invasion. It is possible that other compounds present in the OVP conferred the anti-attachment effects. Baldioli and colleagues argued that oxidative stability of virgin olive oil is due to dialdehydic form of elenolic acid linked to hydroxytyrosol (3,4-DHPEA-EDA), which present at a higher amount as compared to hydroxytyrosol in its free form (Baldioli et al., 1996). In fact, OVP used in this study does contain a higher amount of hydroxytyrosol and tyrosol in their linked forms (Gill et al., 2005). 3,4-DHPEA-EDA and p-HPEA-EDA (dialdehydic form of elenolic acid linked to tyrosol) were shown to partly account for the antiproliferative and apoptotic effects of a virgin olive oil phenol extract in human promyelocytic leukemia cells (Fabiani et al., 2006). 3,4dihydroxyphenylethanol-elenolic acid (3,4-DHPEA-EA) has also been demonstrated to have a strong antioxidant activity (Masella et al., 1999). The metal chelating properties of phenolic compounds may potentially interfere with matrix metalloproteinases (MMPs) disrupting their proteolytic activity by binding the requisite metal ions, thus inhibiting overall invasion (Maeda-Yamamoto et al., 1999). Several studies of
1011
●
●
Epidemiological evidence suggests a link between olive oil consumption and decreased cancer risk. In vitro studies indicate antigenotoxic activity for olive oil phenolics used at physiologically relevant amounts in a range of cell models. Human intervention studies have reported antigenotoxic activity for olive oil consumption and some studies have related that to the phenolics content. Although limited by the amount of scientific literature the in vitro data to date are supported by in vivo data, which collectively are suggestive of an antigenotoxic effect for olive oil phenolics. Olive oil phenolics are able to inhibit invasion in in vitro model systems.
REFERENCES Adhami, V.M., Ahmad, N., Mukhtar, H., 2003. Molecular targets for green tea in prostate cancer prevention. J. Nutr. 133, 2417S–2424S. Annabi, B., Lachambre, M.P., Bousquet-Gagnon, N., Page, M., Gingras, D., Beliveau, R., 2002. Green tea polyphenol (-)-epigallocatechin 3-gallate inhibits MMP-2 secretion and MT1-MMP-driven migration in glioblastoma cells. Biochim. Biophys. Acta. 1542, 209–220. Baldioli, M., Servilli, M., Pevretti, G., Montedoro, G.F., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oil. J. Am. Oil Chem. Soc. 73, 1589–1593. Brenneisen, P., Briviba, K., Wlaschek, M., Wenk, J., ScharffetterKochanek, K., 1997. Hydrogen peroxide (H2O2) increases the steady-state mRNA levels of collagenase/MMP-1 in human dermal fibroblasts. Free Radic. Biol. Med. 22, 515–524. Demeule, M., Brossard, M., Page, M., Gingras, D., Beliveau, R., 2000. Matrix metalloproteinase inhibition by green tea catechins. Biochim. Biophys. Acta. 1478, 51–60. Fabiani, R., De Bartolomeo, A., Rosignoli, P., Servili, M., Selvaggini, R., Montedoro, G.F., Di Saverio, C., Morozzi, G., 2006. Virgin olive oil phenols inhibit proliferation of human promyelocytic leukemia cells (HL60) by inducing apoptosis and differentiation. J. Nutr. 136, 614–619. Ferro-Luzzi, A., Sette, S., 1989. The Mediterranean Diet: an attempt to define its present and past composition. Eur. J. Clin. Nutr. 43, 13–29. Gill, C.I., Boyd, A., McDermott, E., McCann, M., Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G., McGlynn, H., Rowland, I.,
1012
2005. Potential anti-cancer effects of virgin olive oil phenols on colorectal carcinogenesis models in vitro. Int. J. Cancer. 117, 1–7. Gill, C.I., Haldar, S., Boyd, L.A., Bennett, R., Whiteford, J., Butler, M., Pearson, J.R., Bradbury, I., Rowland, I.R., 2007. Watercress supplementation in diet reduces lymphocyte DNA damage and alters blood antioxidant status in healthy adults. Am. J. Clin. Nutr. 85, 504–510. Grasso, S., Siracusa, L., Spatafora, C., Renis, M., Tringali, C., 2007. Hydroxytyrosol lipophilic analogues: Enzymatic synthesis, radical scavenging activity and DNA oxidative damage protection. Bioorg. Chem. 35, 137–152. Hashim, Y.Z., Eng, M., Gill, C.I., McGlynn, H., Rowland, I.R., 2005. Components of olive oil and chemoprevention of colorectal cancer. Nutr. Rev. 63, 374–386. Hashim, Y.Z., Rowland, I.R., McGlynn, H., Servili, M., Selvaggini, R., Taticchi, A., Esposto, S., Montedoro, G., Kaisalo, L., Wahala, K., Gill, C.I., 2008. Inhibitory effects of olive oil phenolics on invasion in human colon adenocarcinoma cells in vitro. Int. J. Cancer 122, 495–500. Hillestrom, P.R., Covas, M.I., Poulsen, H.E., 2006. Effect of dietary virgin olive oil on urinary excretion of etheno-DNA adducts. Free Radic. Biol. Med. 41, 1133–1138. Hoeijmakers, J.H., 2001. Genome maintenance mechanisms for preventing cancer. Nature 411, 366–374. Hood, J.D., Cheresh, D.A., 2002. Role of integrins in cell invasion and migration. Nat. Rev. Cancer 2, 91–100. Kasai, H., 1997. Analysis of a form of oxidative DNA damage, 8hydroxy-2’-deoxyguanosine, as a marker of cellular oxidative stress during carcinogenesis. Mutat. Res. 387, 147–163. Liu, J., Zhang, X., Yang, F., Li, T., Wei, D., Ren, Y., 2006. Antimetastatic effect of a lipophilic ascorbic acid derivative with antioxidation through inhibition of tumor invasion. Cancer Chemother. Pharmacol. 57, 584–590. MacDonald, I.A., Bokkenheuser, V.D., Winter, J., 1993. Degradation of steriods in the human gut. J. Lipid Res. 24, 675–700. Machowetz, A., Poulsen, H.E., Gruendel, S., Weimann, A., Fito, M., Marrugat, J., de la Torre, R., Salonen, J.T., Nyyssonen, K., Mursu, J., Nascetti, S., Gaddi, A., Kiesewetter, H., Baumler, H., Selmi, H., Kaikkonen, J., Zunft, H.J., Covas, M.I., Koebnick, C., 2007. Effect of olive oils on biomarkers of oxidative DNA stress in Northern and Southern Europeans. FASEB J. 21, 45–52. Maeda-Yamamoto, M., Kawahara, H., Tahara, N., Tsuji, K., Hara, Y., Isemura, M., 1999. Effects of tea polyphenols on the invasion and matrix metalloproteinases activities of human fibrosarcoma HT1080 cells. J. Agric. Food Chem. 47, 2350–2354. Manna, C., Galletti, P., Maisto, G., Cucciolla, V., D’Angelo, S., Zappia, V., 2000. Transport mechanism and metabolism of olive oil hydroxytyrosol in Caco-2 cells. FEBS. Lett. 470, 341–344. Masella, R., Cantafora, A., Modesti, D., Cardilli, A., Gennaro, L., Bocca, A., Coni, E., 1999. Antioxidant activity of 3,4-DHPEA-EA and protocatechuic acid: A comparative assessment with other olive oil biophenols. Redox. Rep. 4, 113–121. Nonaka, Y., Iwagaki, H., Kimura, T., Fuchimoto, S., Orita, K., 1993. Effect of reactive oxygen intermediates on the in vitro invasive capacity of tumor cells and liver metastasis in mice. Int. J. Cancer. 54, 983–986. Nousis, L., Doulias, P.T., Aligiannis, N., Bazios, D., Agalias, A., Galaris, D., Mitakou, S., 2005. DNA protecting and genotoxic effects of olive oil related components in cells exposed to hydrogen peroxide. Free Radic. Res. 39, 787–795. Pool-Zobel, B.L., Lotzmann, N., Knoll, M., Kuchenmeister, F., Lambertz, R., Leucht, U., Schroder, H.G., Schmezer, P., 1994. Detection of genotoxic effects in human gastric and nasal mucosa cells isolated from biopsy samples. Environ. Mol. Mutagen. 24, 23–45.
SECTION | II
Cancer
Pool-Zobel, B.L., Neudecker, C., Domizlaff, I., Ji, S., Schillinger, U., Rumney, C., Moretti, M., Vilarini, I., Scassellati-Sforzolini, R., Rowland, I., 1996. Lactobacillus- and bifidobacterium-mediated antigenotoxicity in the colon of rats. Nutr. Cancer 26, 365–380. Quiles, J.L., Farquharson, A.J., Simpson, D.K., Grant, I., Wahle, K.W., 2002. Olive oil phenolics: effects on DNA oxidation and redox enzyme mRNA in prostate cells. Br. J. Nutr. 88, 225–234 discussion 223–4. Rafter, J., Govers, M., Martel, P., Pannemans, D., Pool-Zobel, B., Rechkemmer, G., Rowland, I., Tuijtelaars, S., van Loo, J., 2004. PASSCLAIM--diet-related cancer. Eur J. Nutr. 43 (Suppl 2), II47–II84. Riso, P., Visioli, F., Gardana, C., Grande, S., Brusamolino, A., Galvano, F., Galvano, G., Porrini, M., 2005. Effects of blood orange juice intake on antioxidant bioavailability and on different markers related to oxidative stress. J. Agric. Food Chem. 53, 941–947. Saha, D., Roman, C., Beauchamp, R.D., 2002. New strategies for colorectal cancer prevention and treatment. World J. Surg. 26, 762–766. Salvini, S., Sera, F., Caruso, D., Giovannelli, L., Visioli, F., Saieva, C., Masala, G., Ceroti, M., Giovacchini, V., Pitozzi, V., Galli, C., Romani, A., Mulinacci, N., Bortolomeazzi, R., Dolara, P., Palli, D., 2006. Daily consumption of a high-phenol extra-virgin olive oil reduces oxidative DNA damage in postmenopausal women. Br. J. Nutr. 95, 742–751. Serra-Majem, L., Ngo de la Cruz, J., Ribas, L., Tur, J.A., 2003. Olive oil and the Mediterranean diet: beyond the rhetoric. Eur. J. Clin. Nutr. 57 (Suppl 1), S2–S7. Soobrattee, M.A., Neergheen, V.S., Luximon-Ramma, A., Aruoma, O.I., Bahorun, T., 2005. Phenolics as potential antioxidant therapeutic agents: mechanism and actions. Mutat. Res. 579, 200–213. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food: epidemiology and nutritional approaches. Nutr. Rev. 60, 170–176. Trichopoulou, A., Lagiou, P., Kuper, H., Trichopoulos, D., 2000. Cancer and Mediterranean dietary traditions. Cancer Epidemiol. Biomarkers Prev. 9, 869–873. Veglia, F., Matullo, G., Vineis, P., 2003. Bulky DNA adducts and risk of cancer: a meta-analysis. Cancer Epidemiol. Biomarkers Prev. 12, 157–160. Visioli, F., Galli, C., Bornet, F., Mattei, A., Patelli, R., Galli, G., Caruso, D., 2000. Olive oil phenolics are dose-dependently absorbed in humans. FEBS Lett. 468, 159–160. Vissers, M.N., Zock, P.L., Roodenburg, A.J., Leenen, R., Katan, M.B., 2002. Olive oil phenols are absorbed in humans. J. Nutr. 132, 409–417. Vogelstein, B., Fearon, E.R., Hamilton, S.R., Kern, S.E., Preisinger, A.C., Leppert, M., Nakamura, Y., White, R., Smits, A.M., Bos, J.L., 1988. Genetic alterations during colorectal-tumor development. N. Engl. J. Med. 319, 525–532. Wahle, K.W., Caruso, D., Ochoa, J.J., Quiles, J.L., 2004. Olive oil and modulation of cell signaling in disease prevention. Lipids 39, 1223–1231. Weinbrenner, T., Fito, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Albaladejo, M.F., Abanades, S., Schroder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321. Weyant, M.J., Carothers, A.M., Bertagnolli, M.E., Bertagnolli, M.M., 2000. Colon cancer chemopreventive drugs modulate integrin-mediated signaling pathways. Clin. Cancer Res. 6, 949–956. Woodhouse, E.C., Chuaqui, R.F., Liotta, L.A., 1997. General mechanisms of metastasis. Cancer 80, 1529–1537. Yang, C.S., Landau, J.M., Huang, M.T., Newmark, H.L., 2001. Inhibition of carcinogenesis by dietary polyphenolic compounds. Annu. Rev. Nutr. 21, 381–406. Zhang, G., Miura, Y., Yagasaki, K., 2000. Suppression of adhesion and invasion of hepatoma cells in culture by tea compounds through antioxidative activity. Cancer Lett. 159, 169–173.
Chapter 109
Olives and Olive Oil Compounds Active Against Pathogenic Microorganisms Manuel Brenes, Eduardo Medina, Aranzazu García, Concepción Romero and Antonio de Castro Food Biotechnology Department, Instituto de la Grasa (CSIC), Seville, Spain
109.1 INTRODUCTION Olive oil and olive leaf extracts have been used in folk medicine since ancient times. Romans and Greeks employed olive extracts to treat many diseases and an extract of boiled olive leaves was administered as a drink to malaria patients during the 19th century. Hence, the Mediterranean countries have cultivated the olive tree (Olea europaea L.) to produce olive oil, table olives and olive leaf extracts for centuries. At present, both olive oil and table olives are important components of the Mediterranean diet and are largely consumed throughout the world. In addition, there are many enterprises
that commercialize olive leaf extracts to treat a myriad of diseases, many of them caused by microorganisms. Features of the main food-related bacteria pathogens investigated in relation to olive antimicrobials are presented in Table 109.1. Nevertheless, it was not until the middle of the 20th century that researchers started to look for antimicrobial compounds in olive products, especially in table olives. Juven et al. (1968) stated that oleuropein was the inhibitory substance of lactic acid bacteria growth during the fermentation of the Spanish-style green olives. Later, Fleming et al. (1973) and Kubo et al. (1985) demonstrated that the aglycon of oleuropein was more inhibitory than oleuropein
TABLE 109.1 Main food-related bacterial pathogens which have been investigated in relation to olive antimicrobials. Species
Important traits
Salmonella enterica sv. Enteritidis
Very common food-poisoning organism. Salmonellosis mortality is low, but symptoms may be severe. Poultry, eggs and meat are the products usually involved
Escherichia coli 0157:H7
This strain causes hemorrhagic colitis, particularly in vulnerable groups. Associated with raw or undercooked meat, improperly processed dairy products and vegetables
Shigella sonnei
Normally transmitted by food-handlers. Shigellosis varies from asymptomatic infection to severe illness. Salads, and contaminated water or food are the main sources of infection
Yersinia enterocolitica
Gastroenteritis caused by this organism can be confused with appendicitis. Milk and milk products are implicated foods, but person-to-person transmission has also been reported
Helicobacter pylori
Linked to a majority of peptic ulcers and to some types of gastric cancer. Its water- or food-transmission is still under research
Staphylococcus aureus
Forms heat-stable enterotoxin which causes common food poisoning worldwide. Present in recontaminated cooked meat and poultry, cream-filled pastries, milk, and other foods
Listeria monocytogenes
Listeriosis may cause abortion in pregnant women, and mortality in immunocompromised individuals may reach 40%. Ubiquitous organism: salads, raw milk, cheese, meat, and other products
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1013
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1014
against several bacteria, including certain lactic acid bacteria. More recently, it was demonstrated that the main phenolic compound in olive brines is hydroxytyrosol, a component of the oleuropein moiety, which inhibited the growth of Lactobacillus plantarum in a model system (Ruiz-Barba et al., 1993), but its presence in the brines of both olives non-treated and treated with NaOH did not lead to an explanation for the lack of lactic acid fermentation in the former olive brines. There is controversy over the antimicrobial activity of hydroxytyrosol. Fleming et al. (1973) did not find any inhibitory activity in this substance against several bacteria, Capasso et al. (1995) detected a very limited activity against Pseudomonas savastanoi and Corynebacterium michiganensis, and Obied et al. (2007) reported that pure hydroxytyrosol was less active against Staphylococcus aureus, Bacillus subtilis, Escherichia coli, and Pseudomonas aeruginosa than a phenolic olive extract containing this substance. By contrast, Bisignano et al. (1999) studied the in vitro susceptibility of several human intestinal or respiratory track pathogens to hydroxytyrosol and oleuropein, and the results of this study suggested a high antimicrobial activity of hydroxytyrosol against these microorganisms. Thus, this substance was proposed as useful for the treatment of intestinal or respiratory tract infections in humans. Both oleuropein and hydroxytyrosol have also been identified as effective against HIV viral fusion and integration (Lee-Huang et al., 2007), and the secoiridoid glucoside oleuropein has shown antiviral activity against several pathogens in vitro (Ma et al., 2001). Indeed, many enterprises that commercialize olive leaf extracts claim antiviral activity for their products because of their content in oleuropein and elenolic acid, a component of the oleuropein moiety. The antiviral activity of synthetic calcium elenolate was extensively studied in vitro by pharmaceutical companies during the 1960s and 1970s (Renis, 1970) but experiments in vivo failed. More recently, olive leaf extract enterprises have defended the antiviral properties of their products on the basis of a supposed different isomeric form of elenolic acid in these natural products than that of the synthetic calcium elenolate, but a clear scientific demonstration of this statement is not available. However, researchers have been reporting antifungal and antibacterial activity in olive leaf extracts for years. Micol et al. (2005) referred these extracts as being very effective in inhibiting the viral infection exerted by the Viral Haemorrhagie Septicaemia Rhabdovirus in fish, Pereira et al. (2007) and Markin et al. (2003) found antibacterial (Gram-positive and -negative) and antifungal (Candida albicans and Cryptococcus neoformans) activity in olive leaf extracts, and Korukluoglu et al. (2008) have recently reported the inhibition in vitro of the growth of ten molds by these extracts. All these studies have pointed out the phenolic compounds, in particular oleuropein, as the active agent in the extracts.
SECTION | II Immunology and Inflammation
Furthermore, some reports have shown that phenolic compounds can also be active against phytopathogenic microorganisms (Capasso et al., 1995; Baidez et al., 2006), and several simple phenols were tested in vitro against fungi and bacteria. Capasso et al. (1995) found that the most active phenolic compound in the olive oil mill wastewaters against Pseudomonas savastanoi and Clavibacter michiganensis was methylcatechol, whereas Baidez et al. (2006) proposed quercetin and luteolin as the most antifungal phenolic compounds in the olive plant. Therefore, results are contradictory in many cases and researchers have studied the antimicrobial activity of olives and their derived products by using commercial phenolic compounds (oleuropein, hydroxytyrosol, luteolin, quercetin and others) because of the difficulty to isolate pure substances. Besides, some of these substances, such as oleuropein, luteolin and quercetin, are absent or found in a very low concentration in olive oil and table olives.
109.2 MAIN ANTIMICROBIAL COMPOUNDS IN OLIVE OIL In recent years, several studies have been focused on the antimicrobial activity of olive oil and the identification of its active compounds (Medina et al., 2006, 2007a). A comparison of this activity was tested in vitro among different edible vegetable oils (Figure 109.1). Virgin olive oil had a strong bactericidal activity against a broad spectrum of microorganisms, this effect being higher in general against Gram-positive than Gram-negative bacteria. Moreover, this activity was higher in virgin olive oils, followed by that of olive oils and pomace olive oils, which is in accordance with their content in phenolic compounds. By contrast, this effect was not observed for other edible vegetable oils (corn, sunflower, soybean, rapeseed and cotton). Olive oil exerted bactericidal activity in vitro against the enteric microorganisms Escherichia coli and Clostridium perfringens but also against the beneficial bacteria Lactobacillus acidophilus and Bifidobacterium bifidum (Medina et al., 2006). Zampa et al. (2006) have tested in vitro the antimicrobial activity of hydroxytyrosol and an olive phenolic extract against a variety of microorganisms of the gut microbiota. Strains of Lactobacillus salivarius, Bifidobacterium adolescentis and Bacteroides vulgatus were highly sensitive to both hydroxytyrosol and the phenolic extract, while the strains of C. perfringens, Clostridium clostridiforme and Enterococcus faecalis showed moderate or no susceptibility to them. Hence, it seems that the consumption of olive oil could modulate the growth of intestinal microbiota, although it would depend on the amount of oil ingested and the absorption of polyphenols before they reach the colon (Vissers et al., 2004). A strong bactericidal effect of olive oil against several foodborne pathogens such as Listeria monocytogenes, S. aureus, Yersinia sp., Salmonella Enteritidis and Shigella
CHAPTER | 109
Olives and Olive Oil Compounds Active Against Pathogenic Microorganisms
1015
C. perfringens E. faecium E. faecalis Bacteroides sp. L. acidophylus B. bifidubacterium S. mutans S. aureus Yersinia sp. S. Enteritidis L. monocytogenes E. coli
S. sonnei
Other vegetable oils
Virgin olive oil −6
−5
−4
−3
−2
C. albicans −1
0
Log N1/N0 FIGURE 109.1 Comparison of the antimicrobial activity of virgin olive oil and other edible vegetable oils (sunflower, corn, rapeseed, soybean and cotton). N0, CFU mL⫺1 inoculated; N1, CFU mL⫺1 after 1 h. Virgin olive oil exerted stronger bactericidal activity than other edible vegetable oils.
Log CFU g–1 3.5
Mayonnaise
3
Lettuce
2.5 2 1.5 1 0.5 Not detected
0 Sunflower oil
Not detected
Virgin olive oil
FIGURE 109.2 Survival of S. Enteritidis in egg mayonnaises and L. monocytogenes in lettuce salad elaborated with different oils and initially inoculated with 2 ⫻ 103 CFU g⫺1. This figure shows that pathogenic microorganisms did not survive in mayonnaises made with virgin olive oil.
sonnei (Medina et al., 2006) has also been demonstrated in vitro. These findings confirmed the previous results obtained by Radford et al. (1991). They made egg mayonnaise with virgin olive oil and inoculated it with S. Enteritidis, and the number of microorganisms reduced to an undetectable level after 48 h. Medina et al. (2007a) have confirmed these results inoculating egg and milk mayonnaises made with different types of vegetable oils with S. Enteritidis, and lettuce salad with L. monocytogenes (Figure 109.2). Consequently, these results open up the possibility of using olive oil as a food preservative to prevent the growth of foodborne pathogens or to delay the onset of food spoilage. Several food commodities have been attributed with antimicrobial activity (Puupponen-Pimiä et al., 2001; Koo and Cho, 2004). Among them, wine and tea are the two most reported products with bactericidal activity. Survival of Salmonella Typhimurium, S. sonnei and E. coli in common beverages (cola, beer, milk and wine) has been evaluated (Sheth et al., 1988), and wine showed the highest bactericidal effect. Many other researchers have confirmed the inactivation
of bacteria in wine and its killing effect when consumed with a meal, the activity being assigned to its content in ethanol, organic acids, the polymeric phenolic fraction, simple phenols and low pH (Weisse et al., 1995; Møretrø and Daeschel, 2004; Rhodes et al., 2006). Likewise, catechins purified from green and black tea inhibited the growth of many bacterial species (Friedman et al., 2006). Coffee extracts also exhibited strong bactericidal action against a broad spectrum of microorganisms (Okabe et al., 2003), and vinegar is a wellrecognized bactericidal foodstuff because of its high content of acetic acid (Entani et al., 1998). Taking into account all these previous data, Medina et al. (2007a) ran some experiments to compare the survival of pathogenic bacteria (S. aureus, L. monocytogenes, S. Enteritidis, E. coli, S. sonnei and Yersinia sp.) in olive oil extracts and several common beverages after 5 min of contact. The main conclusion of this work is reflected in Figure 109.3. All tested pathogens survived in peach, pineapple and orange juices, cow’s milk, yogurt drink, coffee extracts, beer with and without alcohol, and Coca-Cola. By contrast, green and black tea showed slight bactericidal activity, followed by red and white wine. Vinegar exerted the strongest bactericidal effect, followed by the aqueous extracts of virgin olive oil. It was, therefore, confirmed that compounds in olive oil possess a strong killer effect against pathogens, which is higher than reported for many other foodstuffs such as coffee, wine or tea. The question is: which are these bactericidal compounds? Very few studies have tried to correlate the antimicrobial activity of olive oil with specific compounds. Radford et al. (1991) and Keceli and Robinson (2002) attributed this activity to the high acidity of the oil as well as simple phenols such as hydroxytyrosol and tyrosol. However, the main phenolic compounds in olive oil are the secoiridoid aglycons of oleuropein and ligstroside, and the lignans (Montedoro et al., 1993; Brenes et al., 1999). Finally, Medina et al. (2006) have correlated the antimicrobial activity of olive oil with the following phenolic
1016
SECTION | II Immunology and Inflammation
Vinegar Increase bactericidal activity
Virgin olive oil
Wine Coffee Coca Cola Beer Fruit juices Others
Tea
FIGURE 109.3 Comparison of the antimicrobial activity of virgin olive oil extracts and other foodstuffs against S. aureus, L. monocytogenes, S. Enteritidis, E. coli, S. sonnei, and Yersinia sp. This figure summarizes the antibacterial efficacy of olive oil in comparison with other liquid foodstuffs.
O
R
O O Dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol HyEDA (R = hydroxytyrosol) Dialdehydic form of decarboxymethyl elenolic acid linked to tyrosol, TyEDA (R = tyrosol) Dialdehydic form of decarboxymethyl elenolic acid, EDA (R = OH) FIGURE 109.4 Structures of the main antimicrobial compounds in olive oil and table olives.
compounds: the dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol (HyEDA) and tyrosol (TyEDA), and free hydroxytyrosol and tyrosol. They isolated each phenolic compound from virgin olive oil by HPLC, and tested their antibacterial activity against L. monocytogenes. Each compound was tested at its concentration found in oil and at a similar constant concentration for all of them. It was confirmed that the compounds with strong bactericidal activity were TyEDA followed by HyEDA (Figure 109.4). No significant effect was observed for hydroxytyrosol and tyrosol and the oleuropein and ligstroside aglycons. It seems that the oleosidic part of the active molecule is very important to exert the killer action. Helicobacter pylori is responsible for most peptic ulcers and gastric cancer, and the infection is currently eradicated with antibiotics although this therapy fails in 10–30% of patients (Cavallaro et al., 2006). Many herbal extracts, essential oils and foodstuffs have exhibited inhibitory activity
against H. pylori in vitro (Chum et al., 2005; Ho et al., 2006; Nohynek et al., 2006) although experiments in vivo failed (McNulty et al., 2001; Zhang et al., 2005). The in vitro activity of olive oil polyphenols against H. pylori has also been studied (Romero et al., 2007). These researchers have discovered a strong anti-H. pylori activity exerted by olive oil extracts rich in phenolic compounds. This activity was even effective against some antibiotic-resistant strains and, more importantly, a very low concentration of the olive oil extract was necessary. Again, the compounds that showed a stronger bactericidal activity against H. pylori were TyEDA and HyEDA. Thus, it has been confirmed for many microorganisms that these two substances are responsible for most of the bactericidal activity in olive oil. Indeed, TyEDA account for most of the anti-H. pylori activity and exerted its bioactivity at a very low concentration in vitro (⬍1.5 μg mL⫺1). TyEDA and HyEDA are complex phenolic compounds present in most virgin olive oils in concentrations up to 250 mg kg⫺1 oil (García et al., 2003) that can be hydrolyzed during olive oil storage (Brenes et al., 2001). The antibacterial treatment of H. pylori is difficult because of the habitat occupied by the organism below the layer of mucus adherent to gastric mucosa. Thus, it is necessary for the antibacterial compounds to diffuse from the oil to the gastric juice and be stable during digestion. Corona et al. (2006) have suggested that these secoiridoids are hydrolyzed at the low pH of the gastric juice but Romero et al. (2007) have demonstrated under simulated gastric conditions that both HyEDA and TyEDA are stable for hours at a pH as low as 2. The latter researchers have also reported that almost half of the phenolic compounds diffused from the oil into the simulated gastric juice; the more polar the compound the more complete the diffusion reached. Hydroxytyrosol and tyrosol totally diffused into the acidic water whereas the diffusion yield of lignans, flavones, and oleuropein and ligstroside aglycons was lower. It was concluded that approximately half of the HyEDA and TyEDA present in the oil diffused into the simulated gastric juice. Therefore, it was demonstrated that these antiH. pylori substances can diffuse from the oil to the gastric juice during digestion, are stable for hours in the acidic environment of the gastric juice and could exert their bioactivity against the microorganism. However, these promising results must be confirmed in vivo and human trials are in progress. Thus, these results could open the possibility of considering virgin olive oil a chemopreventive agent for peptic ulcers or gastric cancer.
109.3 MAIN ANTIMICROBIAL COMPOUNDS IN TABLE OLIVES Table olives were the first olive product that investigators started for the search for antimicrobials (Etchells et al., 1966; Juven and Henis, 1970; Federici and Bongi, 1983).
CHAPTER | 109
1017
Olives and Olive Oil Compounds Active Against Pathogenic Microorganisms
TABLE 109.2 Effect of isolated phenolic and oleosidic compounds from olive brines on the viability of L. pentosus. Log CFU mL⫺1 after 48 h incubation of Gordal brines enriched with the isolated compounds, and inoculated with 6 Log CFU mL⫺1 cells. Compound
Log CFU mL⫺1
TABLE 109.3 Effect of oleuropein (5.8 mM), hydroxytyrosol (5.2 mM) and the dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol (0.5 mM) on the viability of several microorganisms present in table olives. Initial inoculations were 6 Log CFU mL⫺1 and 3.5 Log CFU mL⫺1 for bacteria and yeasts respectively.
Hydroxytyrosol
⬎8
Hydroxytyrosol 1-glucoside
⬎8
Microorganism
Hydroxytyrosol 4-glucoside
⬎8
E. aerogenes
5.2
2.9
1.6
Oleoside
⬎8
E. coli
4.4
1.3
2.1
Tyrosol
⬎8
E. faecium
5.6
5.5
1.3
Secoxyloganin
⬎8
E. faecalis
5.3
4.0
⬍1.0
Secologanoside
⬎8
L. mesenteroides
⬎8.0
5.0
2.0
Log CFU mL⫺1 after 48 h incubation Oleuropein Hydroxytyrosol HyEDA
Oleoside 11-methyl ester
3
L. pentosus
⬎8.0
⬎8.0
⬍1.0
EDA
3.5
L. plantarum
⬎8.0
⬎8.0
3.0
⬍1.2
S. cerevisiae
⬎6.0
⬎6.0
⬎6.0
P. membranaefaciens
⬎6.0
⬎6.0
⬎6.0
HyEDA
This table shows that oleoside 11-methyl ester, HyEDA and EDA were the strongest anti-lactic acid bacteria compounds in table olive brines. EDA, didaldehydic form of decarboxymethyl elenolic acid; HyEDA, dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol.
Many researchers have tried to explain the inhibition of the lactic acid fermentation in olive brines because of their content in oleuropein and hydroxytyrosol (Juven et al., 1968; Ruiz-Barba et al., 1993; Landete et al., 2008). Other researchers have reported a potential antimicrobial activity of table olive extracts but they have not assigned this activity to any compounds (Pereira et al., 2006). However, Medina et al. (2007b) have recently demonstrated that the main antimicrobial compounds in table olives are the dialdehydic form of decarboxymethyl elenolic acid (EDA) (Figure 109.4), HyEDA and an isomer of oleoside 11-methyl ester. In fact, both EDA and HyEDA can explain most of the antimicrobial activity found in olive brines. It must be noted that TyEDA was not detected in the olive brines. These investigators stored olives of the Manzanilla and Gordal variety in an acidified brine under aseptic conditions for 2 months and inoculated the brines with a strain of Lactobacillus pentosus. The microorganism died in the Manzanilla brines and survived in the Gordal. Subsequently, the content of the brines in phenolic and oleosidic compounds was analyzed, and all these substances were isolated by HPLC and their bactericidal activity against L. pentosus was tested. It can be observed in
This table shows that HyEDA exerted a stronger bactericidal activity than oleuropein and hydroxytyrosol against several microorganisms found in table olive brines. HyEDA, dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol.
Table 109.2 that only the three above-mentioned compounds significantly reduced the number of inoculated cells, although HyEDA was the strongest bactericidal compound detected in the olive brines and it could alone explain the inhibition of lactobacillus growth in Manzanilla brines. However, an additive effect was also observed in the three antimicrobial compounds identified, and depending on their concentration, a synergistic effect of all inhibitors must not be ruled out to explain lactic acid bacteria growth inhibition in olive brines. Additionally, the antimicrobial activity of oleuropein, hydroxytyrosol, and HyEDA against several microorganisms found in table olives was studied (Table 109.3). It must be noted that the concentration of HyEDA was ten times lower than that of the other two substances. Oleuropein was ineffective against most microorganisms and hydroxytyrosol allowed for the growth of Lactobacillus plantarum and L. pentosus. It was again confirmed that HyEDA exerted the highest activity against all bacteria tested. These findings have been recently confirmed at a pilot plant scale with olives processed according to the Spanish-style green olive method (Medina et al., 2008). It is well-known that these olives do not properly ferment by
1018
SECTION | II Immunology and Inflammation
H
ent
aO th N
wi
atm Tre Pu
t di
Olives
rec
tly
Insufficient → Formation of HyEDA and EDA Sufficient → No formation of antimicrobials
in b
rine
Formation of HyEDA and EDA FIGURE 109.5 Formation of antimicrobial compounds during table olive processing. This figure reflects that EDA and HyEDA are formed during processing of olives non-treated with NaOH or with an insufficient alkaline treatment. EDA, didaldehydic form of decarboxymethyl elenolic acid; HyEDA, dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol.
lactic acid bacteria when an insufficient NaOH treatment is undertaken (Rodríguez-Borbolla et al., 1969), and there was no explanation for this phenomenon until now. Medina et al. (2008) have demonstrated that an alkaline treatment with a low NaOH strength and insufficient alkali penetration allowed for the presence of EDA and HyEDA in brines (Figure 109.5), and they inhibited the growth of L. pentosus. By contrast, a more intense alkaline treatment gave rise to an abundant growth of the microorganism without any antimicrobials in brines. Likewise, both HyEDA and EDA were always found in Manzanilla brines of olives nontreated with NaOH. Therefore, olive brines can be a food source of antimicrobial compounds such as HyEDA and EDA, in particular those brines with olives non-treated with NaOH.
SUMMARY POINTS ●
●
●
●
●
●
Olive oil and table olives contain compounds with a strong antimicrobial activity. The main antimicrobial compounds in olive oil are the dialdehydic form of decarboxymethyl elenolic acid linked to hydroxytyrosol or tyrosol. The main antimicrobial compounds in table olives are the dialdehydic form of decarboxymethyl elenolic acid either free or linked to hydroxytyrosol. The bactericidal action of virgin olive oil is higher than that of other foodstuffs such as wine, tea, coffee, beer and others. Olive oil could be a hurdle component in certain processed foods and exert a protective effect against foodborne pathogens when contaminated foods are ingested. The antimicrobial compounds of olive oil possess a strong anti-H. pylori activity that could be useful for the prevention of peptic ulcers and cancers in the future.
ACKNOWLEDGMENT This work was supported by the projects AGL2003-00823 and AGL2006-01552.
REFERENCES Baidez, A.G., Gómez, P., del Río, J.A., Ortuño, A., 2006. Antifungal capacity of major phenolic compounds of Olea europaea L. against Phytophtora megasperma Drechsler and Cylindrocarpon destructans (Zinssm.) Scholten. Phyisiol. Mol. Plant. Pathol. 69, 224–229. Bisignano, G., Tomaino, A., lo Cascio, R., Crisafi, G., Uccella, N., Saija, A., 1999. On the “in vitro” antimicrobial activity of oleuropein and hydroxytyrosol. J. Pharm. Pharmacol. 51, 971–974. Brenes, M., García, A., García, P., Rios, J.L., Garrido, A., 1999. Phenolic compounds in Spanish olive oils. J. Agric. Food Chem. 47, 3535–3540. Brenes, M., García, A., García, P., Garrido, A., 2001. Acid hydrolysis of secoiridoid aglycons during storage of virgin olive oils. J. Agric. Food Chem. 49, 5609–5614. Capasso, R., Evidente, A., Schivo, L., Orru, G., Marcialis, M., Cristinzo, G., 1995. Antibacterial polyphenols from olive oil mill waste water. J. Appl. Bact. 79, 393–398. Cavallaro, L., Egan, B., O´Morain, C., Di Mario, F., 2006. Treatment of Helicobacter pylori infection. Helicobacter 11, 36–39. Chum, S., Vattem, D.A., Lin, Y., Shetty, K., 2005. Phenolic antioxidants from clonal oregano (Origanum vulgare) with antimicrobial activity against Helicobacter pylori. Process Biochem. 40, 809–816. Corona, G., Tzounis, X., Dessi, M.A., Deiana, M., Debnam, E.S., Visioli, F., Spencer, J.P.E., 2006. The fate of olive oil polyphenols in the gastrointestinal tract: Implications of gastric and colonic microflora-dependent biotransformation. Free Radical Res. 40, 647–658. Entani, E., Asai, M., Tsujihata, S., Tsukamoto, Y., Ohta, M., 1998. Antibacterial action of vinegar against foodborne pathogenic bacteria including Escherichia coli O157:H7. J. Food Prot. 61, 953–959. Etchells, J.L., Borj, A.F., Kittel, I.D., Bell, T.A., Fleming, H.P., 1966. Pure culture fermentation of green olives. Appl. Microbiol. 14, 1027–1041. Federici, F., Bongi, G., 1983. Improved method for the isolation of bacterial inhibitors from oleuropein hydrolysis. Appl. Environ. Microbiol. 46, 509–510. Fleming, H.P., Walter, W.M., Etchells, J.L., 1973. Antimicrobial properties of oleuropein and products of its hydrolysis. Appl. Microbiol. 26, 777–782. Friedman, M., Henika, P.R., Levin, C.E., Mandrell, R.E., Kozukue, N., 2006. Antimicrobial activities of tea catechins and teaflavins and tea extracts against Bacillus cereus. J. Food Prot. 69, 354–361. García, A., Brenes, M., García, P., Romero, C., Garrido, A., 2003. Phenolic compounds in commercial olive oils. Eur. Food Res. Technol. 216, 520–525. Ho, C.Y., Lin, Y.T., Labbe, R.G., Shetty, K., 2006. Inhibition of Helicobacter pylori by phenolic extracts of sprouted peas (Pisum sativum L.). J. Food Biochem. 30, 21–34.
CHAPTER | 109
Olives and Olive Oil Compounds Active Against Pathogenic Microorganisms
Juven, B., Samish, Z., Henis, Y., Jacoby, B., 1968. Identification of oleuropein as a natural inhibitor of lactic acid fermentation of green olives. J. Agric. Res. 18, 137–138. Juven, B., Henis, Y., 1970. Studies on the antimicrobial activity of olive phenolic compounds. J. Appl. Bact. 33, 721–732. Keceli, T., Robinson, R.K., 2002. Antimicrobial activity of phenolic extracts from virgin olive oil. Milchwissenschaft 57, 436–440. Koo, M.W.L., Cho, C.H., 2004. Pharmacological effects of green tea on the gastrointestinal system. Eur. J. Pharmacol. 500, 177–185. Korukluoglu, M., Sahan, Y., Yigit, A., 2008. Antifungal properties of olive leaf extracts and their phenolic compounds. J. Food Safety 28, 76–87. Kubo, I., Matsumoto, A., Takase, I., 1985. A multichemical defense mechanism of bitter olive Olea europeaea (Oleaceae). Is oleuropein a phytoalexin precursor? J. Chem. Ecol. 11, 251–263. Landete, J.M., Curiel, J.A., Rodríguez, H., de las Rivas, B., Muñoz, R., 2008. Study of the inhibitory activity of phenolic compounds found in olive products and their degradation by Lactobacillus plantarum strains. Food Chem. 107, 320–326. Lee-Huang, S., Huang, P.L., Zhang, D., Lee, J.W., Bao, J., Sun, Y., Chang, Y., Zhang, J., Huamg, P.L., 2007. Discovery of small-molecule HIV1 fusion and integrase inhibitors oleuropein and hydroxytyrosol: Part I. Integrase inhibition. Biochem. Biophys. Res. Commun. 354, 872–878. Ma, S.C., He, Z.D., Deng, X.L., Burt, P.P., Ooi, V.E., Xu, H.X., Lee, S.H., Lee, S.F., 2001. In vitro evaluation of secoiridoid glucosides from the fruits of Ligustrum lucidum as antiviral agents. Chem. Pharm. Bull. 49, 1471–1473. Markin, D., Duek, L., Berdicevsky, I., 2003. In vitro antimicrobial activity of olive leaves. Mycoses 46, 132–136. McNulty, C.A.M., Wilson, M.P., Havinga, W., Johnston, B., O´Gara, E.A., Maslin, D.J., 2001. A pilot study to determine the effectiveness of garlic oil capsules in the treatment of dyspeptic patients with Helicobacter pylori. Helicobacter 6, 249–253. Medina, E., de Castro, A., Romero, C., Brenes, M., 2006. Comparison of the concentration of phenolic compunds in olive oils and other plant oils: correlation with antimicrobial activity. J. Agric. Food Chem. 54, 4954–4961. Medina, E., Romero, C., Brenes, M., de Castro, A., 2007a. Antimicrobial activity of olive oil, vinegar, and various beverages against foodborne pathogenes. J. Food Prot. 70, 1194–1199. Medina, E., Brenes, M., Romero, C., García, A., de Castro, A., 2007b. Main antimicrobials in table olives. J. Agric. Food Chem. 55, 9817–9823. Medina, E., Romero, C., de Castro, A., Brenes, M., García, A., 2008. Inhibitors of lactic acid fermentation in Spanish-style green olive brines of the Manzanilla variety. Food Chem. 110, 932–937. Micol, V., Caturla, N., Pérez-Fons, L., Mas, V., Pérez, L., Estepa, A., 2005. The olive leaf extract exhibits antiviral activity against viral haemorrhagic septicemia rhabdovirus (VHSV). Antiviral Res. 66, 129–136. Montedoro, G.F., Servili, M., Baldioli, M., Selvaggini, R., Miniati, E., Macchioni, A., 1993. Simple and hydrolyzable compounds in virgin olive oil. 3. Spectroscopic characterization of the secoiridoid derivatives. J. Agric. Food Chem. 41, 2228–2234. Møretrø, T., Daeschel, M.A., 2004. Wine is bactericidal to foodborne pathogens. J. Food Sci. 69, 251–257.
1019
Nohynek, L.J., Alakomi, H., Kähkönen, M.P., Heinonen, M., Helander, I.M., Oksman-Caldentey, K., Puupponen-Pimiä, R.H., 2006. Berry phenolics: Antimicrobial properties and mechanisms of action against sever human pathogens. Nutr. Cancer 54, 18–32. Obied, H.K., Bedgood, D.R., Prenzler, P.D., Robards, K., 2007. Bioscreening of Australian olive mill waste extracts: biophenol content, antioxidant, antimicrobial and molluscicidal activities. Food Chem. Toxicol. 47, 1238–1248. Okabe, Y., Yamamoto, Y., Yasuda, K., Hochito, K., Ishii, N., 2003. The antibacterial effects of coffee on Escherichia coli and Helicobacter pylori. J. Clin. Biochem. Nutr. 34, 85–87. Pereira, A.P., Ferreira, I.C.F.R., Marcelino, F., Valentão, P., Andrade, P.B., Reabra, R., Estevinho, L., Bento, A., Pereira, J.A., 2007. Phenolic compounds and antimicrobial activity of olive (Olea europaea L. Cv. Cobrançosa) leaves. Molecules 12, 1153–1162. Pereira, J.A., Pereira, A.P.G., Ferreira, I.C.F.R., Valentao, P., Andrade, P.B., Seabra, R., Estenvinho, L., Bento, L., 2006. Table olives from Portugal: phenolic compounds, antioxidant potential, and antimicrobial activity. J. Agric. Food Chem. 52, 8425–8431. Puupponen-Pimiä, R., Nohynek, L., Meier, C., Kähkönen, M., Heinonen, M., Hopia, A., Oksman-Caldentey., 2001. Antimicrobial properties of phenolic compounds from berries. J. Appl. Microbiol. 90, 494–507. Radford, S.A., Tassou, C.C., Nychas, G.J.E., Board, R.G., 1991. The influence of different oils on the death rate of Salmonella enteritidis in homemade mayonnaise. Lett. Appl. Microbiol. 12, 125–128. Renis, H.E., 1970. In vitro antiviral activity of calcium elenolate. Antimicrob. Agents Chemother., 167–172. Rhodes, P.L., Mitchell, J.W., Wilson, M.W., Melton, L.D., 2006. Antilisterial activity of grape juice and grape extracts derived from Vitis vinifera variety Ribier. Int. J. Food Microbiol. 107, 281–286. Rodríguez-Borbolla, J.M., Fernández-Díaz, M.J., González-Cancho, F., 1969. Influence of pasteurization and lye treatment on the fermentation of Spanish-style Manzanilla olives. Appl. Microbiol. 17, 734–736. Romero, C., Medina, E., de Castro, A., Brenes, M., 2007. In vitro activity of olive oil polyphenols against Helicobacter pylori. J. Agric. Food Chem. 55, 680–686. Ruiz-Barba, J.L., Brenes, M., Jiménez, R., García, P., Garrido, A., 1993. Inhibition of Lactobacillus plantarum by polyphenols extracted from two different kinds of olive brines. J. Appl. Bact. 74, 15–19. Sheth, N.K., Wisniewski, T.R., Frason, T.R., 1988. Survival of enteric pathogens in common beverages: an in vitro study. Am. J. Gastroenterol. 83, 658–660. Vissers, M.N., Zock, P.L., Katan, M.B., 2004. Bioavailability and antioxidant effects of olive oil phenols in humans: a review. Eur. J. Clin. Nutr. 58, 955–965. Weisse, M.E., Eberly, B., Person, D.A., 1995. Wine as a digestive aid: Comparative antimicrobial effects of bismuth salicylate and red and white wine. Br. Med. 311, 1657–1660. Zampa, A., Silvi, S., Servili, M., Montedoro, G.F., Orpianese, C., Cresci, A., 2006. In vitro modulatory effects of colonic microflora by olive oil iridoids. Microb. Ecol. Health Dis. 18, 147–153. Zhang, L., Ma, J., Pan, K., Go, V.L., Chen, J., You, W., 2005. Efficacy of cranberry juice on Helicobacter pylori infection: A double-blind randomized placebo-controlled trial. Helicobacter 10, 139–145.
This page intentionally left blank
Chapter 110
Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy Eva Batanero, Rosalía Rodríguez and Mayte Villalba Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Químicas, Universidad Complutense, Madrid, Spain
110.1 INTRODUCTION Type I allergy is a major public health problem that affects the quality of life of millions of children and adults, and its prevalence has dramatically increased in many countries throughout the last few decades, particularly in those that are industrialized, where it affects more than 25% of the population. This disorder is characterized by raised IgE antibody levels to otherwise harmless environmental antigens (the so-called allergens), which are responsible for the clinical symptoms such as asthma, eczema, rhinoconjunctivitis or anaphylactic shock, due to the activation of cells in latter encounters between the allergen and the body (Figure 110.1). In Mediterranean countries and some areas of America, South Africa, Japan and Australia, olive (Olea europaea) pollen constitutes one of the most important causes of pollinosis (Liccardi et al., 1996). This species sheds its pollen in high concentrations (reaching a weekly average of 500 grains/m3 and exceptional daily peaks higher than 5000 grains/m3 in some areas of southern Spain) during the pollination season, leading to allergic symptoms from seasonal rhinoconjunctivitis to asthma in susceptible individuals. The onset of allergic symptoms in pollensensitive patients is often related to the number of airborne pollen grains. For olive, the threshold level of airborne pollen required to elicit clinical symptoms in sensitized patients is extremely high, around 400 grains/m3, compared to the 50 grains/m3 for patients allergic to grass pollens (Quiralte et al., 2007). However, the threshold level appears to vary among sensitive patients and during the pollination season. Over the last few years, intense efforts have been made to define the allergenic components (allergogram) of olive pollen (Rodríguez et al., 2007a), after pioneering studies in the 1980s (Blanca et al., 1983). The knowledge of the complete olive pollen allergogram would enable us to Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
understand the mechanisms involved in the development of pollinosis as well as to design novel strategies for an accurate diagnosis and a safer and more effective immunotherapy. Standard laboratory techniques have shown that olive pollen extract has a very complex and heterogeneous allergogram, containing at least 20 protein bands with allergenic activity (Figure 110.2). To date, ten allergens have already been identified, isolated and characterized – named Ole e 1 to Ole e 10 according to the recommendations published by the International Union of Immunological Societies. However, new allergens are waiting to be detected and identified. In this sense, the advances in proteomic tools – e.g. two-dimensional electrophoresis in combination with mass spectrometry – have allowed the detection of two minor components of the extract and their identification as allergens with clinical significance (Rodríguez et al., 2007b), and their studies are being carried out at present: Ole e 11 and Ole e 12 (unpublished data). Several olive allergens have been reported as major allergens, because they exhibit prevalence higher than 50%, such as Ole e 1 and Ole e 4. However, the prevalence of olive pollen allergens (major versus minor allergens) are related with the levels of airborne pollen, and therefore, with the geographical area of the sensitized population (Rodríguez et al., 2001; Barber et al., 2007; Quiralte et al., 2007). While Ole e 1 seems to be the only relevant allergen involved in olive sensitization in areas of low/intermediate exposure, minor allergens (prevalence ⬍50%) that rarely sensitized allergic patients in normally exposed areas, become major allergens (e.g. Ole e 6, Ole e 7, Ole e 9 and Ole e 10) in locations with high levels of exposure. According to this fact, it has been suggested that allergic patients from areas with extremely high levels of exposure exhibit different and more complex allergograms, as determined by molecular diagnosis, when compared with
1021
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1022
SECTION | II Immunology and Inflammation
FIGURE 110.1 Mechanism of type I allergy. Allergy involves two temporally different processes: sensitization and provocation. In an initial exposure to the allergen of susceptible individuals, the uptake of allergen by professional antigen-presenting cells (APCs) leads to the activation of allergen-specific T helper 2 (Th2) cells which produce key cytokines. These cytokines are involved in the class-switching of B cells to IgE synthesis. These antibodies specifically bind to a high-affinity receptor on effector cells (mast cells and basophils), resulting in allergic sensitization. Subsequent encounters with the allergen cause cross-linking of effector cell-bound IgE (provocation stage), leading to the cell activation and the rapid secretion of a wide array of mediators responsible for the allergic symptoms.
FIGURE 110.2 Allergogram of olive pollen. Olive pollen proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the extract and stained with Coomassie blue (CB) or detected with individual sera from patient allergic to olive pollen (IgE). The IgE-binding patterns were selected as representative for variability and complexity of responses in individuals suffering from olive pollinosis. Ribbon diagram of 3-D structures of Ole e 6 (Treviño et al., 2004) and C-terminal domain of Ole e 9 (Treviño et al., 2008) are shown.
patients living in areas with lower pollen count. Regarding allergogram heterogeneity, 45 different allergograms were observed when eight olive pollen allergens were tested in 156 patients from Jaén with olive pollinosis (Quiralte et al., 2007). The main properties of olive pollen allergens are summarized in Table 110.1. Olive pollen allergens have been
classified into seven of the 29 pollen allergen families, according to their physicochemical properties. Olive pollen allergens reveal to be restricted to a small number of taxonomically diverse plant families such as Ole e 1, or are ubiquitous (pan-allergens) such as Ole e 2 (profilin) and Ole e 3 (polcalcin). Pan-allergens constitute families of homologous and structurally related proteins from different species
CHAPTER | 110 Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy
1023
TABLE 110.1 Olive pollen allergens and their main features. Olive pollen allergens Allergen
MW (kDa)1
Pl2
Prevalence (%) Cross-reactivity
Family
Recombinant expression
Ole e 1
16.3/18.5*
5.5–6.5*
55–90
Oleaceae
Ole e 1-like
E. coli, P. pastoris
Ole e 2
14-16**
5.1**
24
Pollens, foods and latex
Profilin
E. coli
Ole e 3
9.2*
4.3*
20–30
Pollens
Polcalcin
E. coli, A. thaliana
Ole e 4
32**
4.6-5.1*
80
ND
Unknown
–
Ole e 5
16**
4.2*
35
ND
Cu/Zn-superoxide dismutase
E. coli
Ole e 6
5.8*
14.6**
10–55
Oleaceae (PD)
Unknown
P. pastoris
Ole e 7
10*
ⱖ9 *
473
Pollens (low) (PD)
Lipid transfer protein
–
Ole e 8
18.8*
4.5**
5
Oleaceae (PD)
Ca 2⫹-binding protein
E. coli, A. thaliana
Ole e 9
46.4*
4.8–5.4*
653
Pollens, foods and latex
1,3-β-glucanase
P. pastoris (CtD and NtD)
Ole e 10
10.8*
5.8*
553
Pollens, foods and latex
Carbohydrate-binding module 43
P. pastoris, S. frugiperda
1 Molecular mass determined by mass spectrometry (*) or sodium dodecyl sulfate-polyacrylamide gel electrophoresis (**). For Ole e 1, molecular masses of non-glycosylated/glycosylated forms are shown. 2 Isoelectric point (pl) determined experimentally (*) or deduced from the amino acid sequence (**). 3 Data from allergic patients living in a region with high exposure to olive pollen. ND, not determined; PD, preliminary data.
responsible for extensive IgE cross-reactivity among a variety of allergic sources. Interestingly, the prevalence of panallergens Ole e 2 and Ole e 3 is usually low (around 20%), indicating that they might not be relevant olive allergens and that sensitization to them might be caused by other sources. Biochemical and molecular studies to characterize olive pollen allergens have shown that polymorphism is a general feature. Ole e 1, Ole e 5 and Ole e 7 present a high degree of polymorphism. For Ole e 9, a relatively low, although still significant, degree of polymorphism has been detected. Allergen polymorphism is closely related to the cultivar origin of olive pollen (Alché et al., 2007), as it has been described for other sources of plant allergens such as date palm (Phoenix dactylifera) or birch (Betula pendula) pollens and apple (Malus domesticus). In this context, it has been speculated that broad polymorphism could be involved in the physiology of the olive reproductive system, including the adaptation of the plant to different environmental conditions, the establishment of the compatibility system, and pollen performance (Alché
et al., 2007). Regarding the clinical implications of allergen polymorphism, the differences in allergen composition in cultivars, particularly in Ole e 1, are responsible for important differences in allergenic activity (Alché et al., 2007). The concentration of Ole e 9 has also been reported to vary several hundred times between different pollen batches. Preliminary studies regarding the levels of expression for Ole e 2, Ole e 3, Ole e 5 and Ole e 6 in the major olive cultivars indicate the presence of significant differences (Alché et al., 2007). This is a major concern for clinicians since reliability of pollen extracts used for clinical purposes is required for an accurate diagnosis and effective and safe immunotherapy. Pollen extracts should imitate as much as possible the composition of the panel of allergens to which patients are normally exposed and are reactive. Therefore, the quantification of both major and minor allergens must be an integral part of the standardization of olive pollen extracts. Furthermore, it has been reported that olive pollen allergens are quickly released in different rates from pollen, having an impact on sensitization and the elicitation
1024
of allergic symptoms. High yields of Ole e 1, Ole e 6 and Ole e 7 are obtained after 15 minutes of pollen hydration in mild saline buffers, and others such as Ole e 3, Ole e 9 and Ole e 10 are completely extracted after 3 hours. This property facilitates the accessibility of the proteins once the pollen grains come in contact with respiratory mucosa, explaining the fast induction of allergic symptoms. Olive pollen allergy is a complex disease which could result from the interaction between environmental and genetic factors (Cárdaba et al., 2007; Quiralte et al., 2007). A strong association between HLA class II antigens DR7 and DQB2 and the IgE response to Ole e 1, Ole e 2, and Ole e 3 has been reported in unrelated populations. HLADR2 antigen is associated with the IgE response to Ole e 10. Other factors brought about by human civilization, including atmospheric pollution, exposure to tobacco, lifestyle related to diet and hygiene habits, may also have considerable effects on pollen allergy. Thus, olive pollen allergy represents an interesting model to study the allergic response; therefore, the identification of allergens of olive pollen will be required for the development of rational strategies for standardization, patient diagnosis and therapy which may increase the quality of life of allergic patients.
110.2 OLE E 1 AS A MARKER FOR SENSITIZATION TO OLEACEAE POLLENS Ole e 1 is the main allergen of olive pollen with a prevalence ranging from 55 to 90%, depending on the geographical areas. It is the most abundant protein of olive pollen, representing up to 20% of the total protein content of pollen in the most profuse varieties. Ole e 1 contributes significantly to the total allergenicity of the olive pollen extracts, and its concentration correlates well with the total allergenic potency of the extracts. Moreover, it also represents the main sensitizing agent within the protein family of Ole e 1-like proteins, and it is a diagnosis marker for sensitization to Oleaceae pollens (Palomares et al., 2006a). Ole e 1 is a polymorphic and acidic (isoelectric point, pI, 5.5–6.5) glycoprotein of 145 amino acid (aa) residues with a glycan at position asparagine-111 (N-glycan) (Villalba et al., 1993; Batanero et al., 1994). Because of the presence of the glycan, Ole e 1 exhibits in sodium dodecyl sulfate-polyacrylamide gel electrophoresis a pattern of multiple bands with two main components, a glycosylated form of 20 kDa (85% of the whole allergen) and a non-glycosylated variant of 18.5 kDa (10%). Two minor variants corresponding to the hyperglycosylated (22 kDa) and the 20 kDa-dimer (40 kDa) are frequently present in Ole e 1 preparations. Its single polypeptide chain contains six cysteine residues which are involved in three disulfide bridges: Cys19-Cys90, Cys22-Cys131, and Cys43-Cys78 (González et al., 2000).
SECTION | II Immunology and Inflammation
Ole e 1 belongs to a large family of homologous proteins (Ole e 1-like), which are specifically expressed in pollen tissue, and it has been suggested to be involved in fertilization events: pollen hydration and/or pollen germination (Villalba et al., 1994; Alché et al., 2007). This is in agreement with Ole e 1 localization in the endoplasmic reticulum, pollen wall and tapetum, and in the outer region of the pollen exine. This family comprises both allergenic members such as Ole e 1, Fra e 1 (Fraxinus excelsior), Lig v 1 (Ligustrum vulgare), Syr v 1 (Syringa vulgaris) – members of Oleaceae family, Pla l 1(Plantago lanceolata), Che a 1 (Chenopodium album), Lol p 11 (Lolium perenne) and Phl p 11 (Phleum pratense), as well as non-allergenic members such as BB18 from Betula verrucosa. Other members of the family are known through their corresponding nucleotide sequences and their derived mature proteins have not been isolated or characterized; therefore, their potential allergenicity has not been explored: LAT52 (Lycopersicon esculentum), Zmc13 (Zea mays), OSPG (Oryza sativa), and putative proteins from Phalaris coerulenscens, Sambucus nigra and Arabidopsis thaliana. The position of the six cysteine residues is conserved in all members, suggesting similar three-dimensional (3-D) structures. In addition, all these proteins exhibit a putative N-glycosylation site, which seems to be conserved among members of the same taxonomic family. This fact could be related to a potential role for the N-glycan in modulation of the biological function and/or discrimination between species. A high degree of cross-reactivity has been described among Oleaceae pollens, and Ole e 1-like proteins are one of the molecules involved in such process (Rodríguez et al., 2001, 2007b). Recently, it has been demonstrated that epitopes of Ole e 1 are only present in Oleaceae pollens but not in unrelated ones (Palomares et al., 2006a). This could be explained by comparing the amino acid sequence of the members of this family: identity values ranging from 86% to 89% are obtained for the Oleaceae counterparts, whereas low but significant identities (32% to 39%) are displayed in the remaining members. Moreover, isoforms of Ole e 1 and Ole e 1-like allergens from Oleaceae, which differ only in a few amino acids, show differences in both IgE and IgG reactivity. Finally, ash pollen is an important cause of allergy in Central Europe, whereas privet and lilac (two ornamental plants) have been described as elicitors of allergic symptoms in conditions of local exposure (Liccardi et al., 1996). Thus, the study of Ole e 1-like allergens will provide knowledge of the molecular organization of allergenic epitopes which is invaluable in terms of allergen standardization and diagnostics as well as in the designing of novel allergen-specific immunotherapy. T-cell (Cárdaba et al., 2007) and B-cell (González et al., 2006) epitopes of Ole e 1 have been analyzed. Cárdaba et al. have defined the regions 91–102 and 109–130 of Ole e 1 as immunodominant T-cell epitopes, but display no IgE-binding capacity. At least four B-cell epitopes have
CHAPTER | 110 Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy
been defined by means of 12-aa overlapping synthetic peptides and recombinant large fragments, being the C-terminal an immunodominant region and the tyrosine at position 141 a critical residue for IgE binding (González et al., 2006).
110.3 OLE E 2 AND OLE E 10, TWO ALLERGENS ASSOCIATED WITH ASTHMA Quiralte et al. (2007) reported that Ole e 2 and Ole e 10 show a statistically significant association with asthma. Ole e 2, an acidic (pI 5.1) 16 kDa protein, belongs to the well-known pan-allergen family of profilins (Ledesma et al., 1998a). Its molecular and immunological properties did not differ from those of other profilins. Ole e 2 exhibits polymorphism, with important implications for the 3-D structure of the molecule (Alché et al., 2007). Profilins contain a large number of members from plant and animal tissues which are involved in the assembly of actin filaments, being recognized as pan-allergens from fruits, vegetables, pollens and latex. Interestingly, the comparison of plant profilins with homologous from fungi and vertebrates revealed that they form a highly conserved family with sequence identities between 70% and 85% among each other, but low identities (30% to 40%) compared with non-allergenic profilins from other eukaryotes, including human beings. This explains their implication in cross-reactivity between profilins of different vegetable sources, and their low clinical relevance (around 20% of prevalence). However, because of sensitivity to thermal denaturalization and gastric proteolysis, they are involved in the oral allergy syndrome (OAS), whose symptoms, limited to the oropharyngeal mucosa, are elicited after eating raw foods. Ole e 10, a Cys-rich small (10.8 kDa) and acidic (pI 5.8) protein, has been identified as a major allergen from olive pollen in high-exposure areas (Barral et al., 2004a). Ole e 10 shows identity with deduced sequences from Arabidopsis thaliana genes (42–46% identity), with the non-catalytic C-terminal domain of plant 1,3-β-glucanases (27–53% identity) such as Ole e 9, and with Cys-box domains from three families of 1,3-β-glucanosyltransferases involved in yeast development: Epd1, Gas-1p and Phr2 families (23% identity) (Barral et al., 2004a). However, it is important to remark that Ole e 10 is an independent protein that defines a novel family of carbohydrate-binding modules (CBMs), so-called CBM43 (Barral et al., 2005a). The ability of Ole e 10 to bind specifically 1,3-β-glucans, its localization within the mature pollen grain inside Golgiderived vesicles and its co-localization with callose (1,3-βglucans) in the growing pollen tube suggest a role for this protein in the pollen tube wall re-formation during germination (Barral et al., 2005a). Regarding its allergenic activity, Ole e 10 is an allergen per se that can act as a primary
1025
sensitizer: it defines a new family of pan-allergens that shows notable intra- and interspecies cross-reactivity, and is a powerful candidate for pollen-latex-fruit syndrome (Barral et al., 2004a). Although it is a relevant allergen, Ole e 10 exhibits low intrinsic antigenicity because no specific IgG antibodies were elicited after different attempts to immunize animals with pure antigen (Barral et al., 2004a). This fact, added to its biological activity and retardation release from pollen after hydration allow us to speculate that this allergen could be expelled from the pollen linked to particles such as 1,3-β-glucans – it has been shown that this glycan promotes an airway allergic response in humans. In this way, the allergenic potential of Ole e 10 could be increased, contributing to eliciting asthma in allergic patients.
110.4 OLE E 3 AND OLE E 8: CA2⫹-BINDING ALLERGENS Ole e 3 and Ole e 8 are two Ca2⫹-binding proteins (CaBPs) of the EF-hand family. Ole e 3 is a small (9.2 kDa) and acidic protein (pI 4.3) with a single polypeptide chain which does not contain cysteines (Batanero et al., 1996). Ole e 3 belongs to the widespread polcalcin family, which is characterized by its specific expression in pollen and the presence of two EFhand motifs (Ledesma et al., 1998b). Polcalcins belong to the buffering type CaBP subfamily and may have a role as inhibitors of cytoplasmic streaming of Ca2⫹ in growing pollen tubes (Ledesma et al., 1998b). The reported prevalence for this family of pan-allergens varies between 5–46% and it is around 20–30% for Ole e 3. Members of this family show low or no polymorphism and sequence identities ranging from 64% to 92%, which explains their strong cross-reactivity. Thus, diagnosis of polcalcinsensitized patients could be achieved whatever polcalcin used, whereas for immunotherapy, the polcalcin that acts as the primary sensitization agent should be used. Ole e 8 (18.75 kDa and pI 4.5) seems to be the first member of a new family of CaBPs with four EF-hand motifs present in the pollen tissue: it is a calcium-sensor protein which should display a regulatory function and perhaps may be involved in signal transduction pathways (Ledesma et al., 2000). This allergen is present at very low levels in the pollen (0.02–0.05% of total protein), and shows low prevalence (5%) (Ledesma et al., 2002a). In addition, all sera reactive to Ole e 8 also recognize Ole e 3 (Ledesma et al., 2000). Ole e 8 shows a low sequence identity with Ole e 3 and other CaBPs out of the EF-hand sites. Moreover, this allergen seems not to have counterparts with conserved amino acid sequence in other taxonomically non-related allergenic pollens, as significant crossreactivity was observed only with Oleaceae (Ledesma et al., 2002a).
1026
It is important to note that EF-hand motifs of the Ca2⫹binding allergens have a non-significant role in their IgE and IgG epitopes, and cross-reactivity; however, their IgEbinding capacity is affected by the conformational change induced by the binding/releasing of Ca2⫹ ions (Ledesma et al., 2002b). This property would allow designing hypoallergenic derivatives to be used in immunotherapy.
110.5 OLE E 7, A NON-SPECIFIC LIPID TRANSFER PROTEIN, AND ITS CLINICAL SIGNIFICANCE Ole e 7 is a basic protein (pI ⱖ 9) with molecular mass around 10 kDa (Tejera et al., 1999). Although it is a minor allergen, its prevalence increases up to 47% in populations exposed to high levels of olive pollen. Interestingly, a large number of adverse reactions are recorded in patients sensitized to Ole e 7 as well as to Ole e 9; these patients are less tolerant to immunotherapy at the recommended allergen doses (Barber et al., 2007). Ole e 7 belongs to the non-specific lipid transfer proteins (nsLTP) family of around 9 kDa ubiquitously distributed through the plant kingdom, and whose members show sequence identities from 47% to 92% (Salcedo et al., 2007). All of them conserve a pattern of eight cysteines, forming four disulfide bridges which are essential for the compact fold of nsLTPs in the characteristic all-α-type structure and, therefore, the lipid-binding and allergenic properties of these proteins. Interestingly, other major groups of plant food allergens, the 2S albumins and cereal α-amylase/trypsin inhibitors, show 3-D structures close to nsLTP fold; these three groups constitute the prolamin superfamily. nsLTPs bind different types of lipids and are involved in plant defense mechanisms against phytopathogenic bacteria and fungi – becoming members of the pathogenesis-related 14 (PR-14) protein family – and probably in the assembly of hydrophobic protective layers of surface polymers such as cutin. Several members of the nsLTP family have been identified as relevant allergens in plant foods and even latex and pollens (Salcedo et al., 2007). Because of their ubiquitous distribution in different plant species and tissues, nsLTPs have been proposed as a novel pan-allergen with a potential spectrum of cross-reactivity comparable to that reported for profilins. Interestingly, nsLTP sensitization shows an unexpected pattern throughout Europe, with high prevalence in the Mediterranean area, but a low incidence in Northern and Central Europe. Genetic factors, differences in dietary habits and level and composition of pollen exposure could account for such differences. Food nsLTPs are considered to be ‘true’ food allergens that sensitize directly via the oral route and are responsible for the induction of severe symptoms in many patients. These features seem to be related to their high resistance
SECTION | II Immunology and Inflammation
to both heat treatment and digestive proteolyis. Moreover, cross-reactivity among food nsLTPs allergens from botanically related and unrelated species has been described (Salcedo et al., 2007). However, concerns exist about whether pollen nsLTP allergens can act as a primary sensitization agent itself via the respiratory tract leading to food allergy because of cross-reactivity to food nsLTPs. Three different types of pollen nsLTP allergens can be defined according to their cross-reactivity with those from foods and their prevalence among pollinic patients (Salcedo et al., 2007). Regarding Ole e 7, our preliminary data indicate that it does not cross-react with foods in spite of the association of this allergen with plant-derived food anaphylaxis reported by Florido et al. (2002): only a very limited cross-reactivity with related and unrelated pollens has been observed (unpublished data). Future studies will help to clarify the different routes of sensitization and geographic patterns of sensitization to nsLTPs.
110.6 OLE E 9 AND POLLEN-LATEX-FRUIT SYNDROME Ole e 9 is a 1,3-β-glucanase belonging to the PR-2 protein family whose enzymatic activity has been shown (Huecas et al., 2001). It exhibits low sequence identity (32–39%) to long 1,3-β-glucanases from plants. It is an acidic (pI 4.8– 5.4) and polymorphic glycoprotein (46.4 kDa) composed of two structurally and immunologically well-defined domains which are connected by a segment of 10–15 aa (Palomares et al., 2003, 2005). The N-terminal domain (NtD) of 334 aa contains the 1,3-β-glucanase activity, and its 3-D modeling fits well to a triose-phosphate isomerase (TIM)-barrel structure common to all known 1,3-β-glucanases. The C-terminal domain (CtD), with around 100 aa, is a CBM43 which shows sequence identity with 1,3-β-glucanases from plant tissues, the Epd1/Gas-1p/Phr2 protein families and Ole e 10 (Palomares et al., 2003). Its capacity to bind 1,3β-glucans suggests a role in the binding of the substrate (Rodríguez et al., 2007b). Disulfide bridges of the molecule have been determined at positions Cys14-Cys76, Cys33-Cys94, and Cys39-Cys48 (Palomares et al., 2003). Recently, its 3-D structure has been resolved, representing a novel type of allergen folding which consists of two parallel α-helices, a small antiparallel β-sheet and a 3–10 helix turn, all connected by long coil segments (Figure 110.2) (Treviño et al., 2008). Moreover, the CtD-epitope mapping shows that B-cell epitopes are mainly located on the loops (Treviño et al., 2008). Ole e 9 is a major allergen in populations living in highly exposed areas with a prevalence of 65% (Huecas et al., 2001). The ubiquity of 1,3-β-glucanases in higher plants suggests that they could be involved in the pollenlatex-fruit syndrome. This is supported by a study showing the involvement of NtD in cross-reactivity among
CHAPTER | 110 Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy
pollens, vegetable foods and latex (Palomares et al., 2005). Moreover, it has been shown that NtD may be a useful marker of disease in diagnosis of pollen-allergic patients at risk of developing allergic symptoms to other vegetable sources (Palomares et al., 2006b). CtD has been described as a marker for patients who could develop asthma (Palomares et al., 2006b). Recently, Ole e 9 has been identified as the causative agent of occupational rhinitis in a researcher (Palomares et al., 2008a).
110.7 OTHER ALLERGENS FROM OLIVE POLLEN: OLE E 4, OLE E 5 AND OLE E 6 Ole e 4 is a partially characterized polymorphic allergen with an apparent molecular mass of 32 kDa and an acidic pI (4.6–5.1) (Boluda et al., 1998). However, it is not clear whether Ole e 4 is a genuine allergen or a proteolytic degradation product of Ole e 9 since all of the peptides obtained for the protein aligned with segments of Ole e 9. Ole e 5 (16 kDa) is the Cu/Zn-superoxide dismutase (SOD) of olive pollen, with an 80–90% sequence identity with Cu/Zn-SODs from other plants (Boluda et al., 1998; Butteroni et al., 2005). SODs catalyze the dismutation of superoxide anions into molecular oxygen and hydrogen peroxide. It has been suggested that pollen antioxidant systems such as Ole e 5 could play a role in pollen–stigma interactions and defense because of the constitutive accumulation of reactive oxygen species/H2O2 in angiosperm stigmas (Alché et al., 2007). Ole e 5 is a minor allergen with prevalence around 35%; however, it could be involved as a putative cross-reactive allergen in the pollen-latex-fruit syndrome due to the ubiquity of this enzyme family and its sequence identity with the allergenic Cu/Zn-SOD from tomato fruit. Moreover, it has been suggested that the allergenic Mn-SOD from latex (Hev b 10) could be involved in cross-reactivity with SODs from related and unrelated species. Ole e 6 is a small (5.83 kDa), acidic (pI 5.8) and Cysrich allergen whose prevalence is very dependent on the degree of pollen exposure, ranging from 10% to 55% (Batanero et al., 1997). It displays a peculiar twice-repeated cysteine motif (Cys-X3-Cys-X3-Cys) which is also present in the amino acid sequence deduced (107 aa) from Tap1, a stamen-specific gene from snapdragon (Antirrhinum majus) (Batanero et al., 1997). The 3-D structure of Ole e 6 has been resolved and consists of two parallel α-helices joined together by three disulfide bridges (Figure 110.2) (Treviño et al., 2004). Ole e 6 shows 59% sequence identity with a Cys-rich pollen surface molecule (Ntp-CysR, 63 aa) from Nicotiana tabacum expressed almost exclusively in developing anthers and highly abundant in pollen. It has been suggested that Ntp-CysR could interact with pistil factors. Concerning cross-reactivity, an Ole e 6-like protein has been reported in ash pollen (Rodríguez et al., 2007b),
1027
and it is expected that homologous allergens exist in other Oleaceae members.
110.8 THE ROLE OF N-GLYCANS IN OLIVE POLLEN ALLERGY Even though allergologists are sceptical about the clinical significance of glycan-related IgE reactivity, increasing numbers of reports have demonstrated that glycans, as part of an allergen, can elicit IgE antibodies in susceptible subjects. Many of these glycoepitopes behave as pan-epitopes, leading to extensive cross-reactivity among pollens, plant foods and insect venoms because of the presence of common monosaccharide components at specific positions along the chain (Altmann, 2007). Thus, they are so-called ‘crossreactive carbohydrate determinants’. The N-glycan of Ole e 1 is involved in antigenic and allergenic activities, being responsible for cross-reactivity between Ole e 1 and non-related glycoproteins (Batanero et al., 1994, 1999; van Ree et al., 2000). The primary structure of the N-glycan of Ole e 1 has been determined and the presence of glycoforms (different but close related glycan structures at a single N-glycosylation site) has been described (Figure 110.3) (van Ree et al., 2000). Ole e 1 has one major ‘complex’ N-glycan (GlcNAcMan3XylGlcNAc2) and one major ‘high mannose’ N-glycan (Man7GlcNAc2). A minor ‘complex’ N-glycan having an α(1,3)-fucose residue attached to the proximal glucosamine residue has also been detected (Batanero et al., 1999; van Ree et al., 2000).
FIGURE 110.3 N-glycans of Ole e 1. Structures of the major N-glycans isolated from Ole e 1 were determined by 1H NMR and MALDI-TOF. Minor α(1,3)-fucosylated N-glycans were also detected. αn and βn indicate α (1,n) and β(1,n) linkages (n ⫽ 2–6).
1028
It was demonstrated that both β(1,2)-xylose and α(1,3)fucose are involved in IgE binding, making the involvement of ‘high mannose’ structures in IgE binding unlikely (Batanero et al., 1999; van Ree et al., 2000). It has been suggested that both monosaccharides could account for the immunogenicity of plant glycans in humans since they are absent in human N-glycans. In addition, the free N-glycan of Ole e 1 is able to induce in vitro histamine release from basophils of olive-sensitized patients, confirming the allergenic character of this glycan (Batanero et al., 1999). The role of the N-glycan of Ole e 9 in the allergenicity of the molecule has not been studied so far.
110.9 RECOMBINANT OLIVE POLLEN ALLERGENS AS DIAGNOSTIC AND THERAPEUTIC REAGENTS Many of the disadvantages associated with the use of allergen extracts from biological sources for diagnosis and treatment of allergy – e.g. difficult allergen standardization, risk of new sensitization and anaphylactic side-effects and endotoxin contamination – can be overcome with the use of recombinant allergens. Various heterologous systems are currently used for the production of recombinant allergens equalling the natural ones as defined molecules in consistent quality and high amount. Most of the olive pollen allergens have been obtained using these methods. Ole e 2, Ole e 3, Ole e 5 (as a fusion protein with glutation S-transferase, GST) and Ole e 8 allergens have been produced in the bacteria Escherichia coli as soluble highyield recombinant proteins (Asturias et al., 1997; Ledesma et al., 1998b, 2000; Butteroni et al., 2005), whereas the production of recombinant Ole e 1 as a fusion protein with GST in this system rendered a poor quality product: it was mainly obtained as insoluble inclusion bodies (Villalba et al., 1994). The yeast Pichia pastoris has been used as a successful expression system for olive pollen allergens with posttranslational modifications (such as glycosylation) and complex folding (including disulfide bridges): Ole e 1 (Huecas et al., 1999), Ole e 6 (Barral et al., 2004b), and the NtD and CtD of Ole e 9 (Palomares et al., 2003, 2005) have been obtained with high yields and they maintain their intrinsic properties. For Ole e 10, better yields and lower degradation of the soluble and functional recombinant protein have been achieved using baculovirus in host insect cells (Spodoptera frugiperda) system (Barral et al., 2006) than with the yeast P. pastoris (Barral et al., 2005b). Finally, Ole e 3 and Ole e 8 have also been produced in stable transgenic plants of Arabidopsis thaliana (Ledesma et al., 2006). These transgenic plants could constitute important tools to design edible vaccines for allergy. Recombinant allergens allow us for the first time to determine the precise sensitization profile (allergogram) of patients, which is a prerequisite to select the allergens for
SECTION | II Immunology and Inflammation
patient-tailored immunotherapy. It is well established that a panel of a few recombinant allergens is sufficient to diagnose most pollen-allergic patients because of the extensive cross-reactivity. Thus, the first step for this procedure is the selection of the most relevant allergens which contain most of the important B- and T-cell epitopes and represent the originally sensitizing agents within a cross-reacting group, e.g. for Oleaceae pollinosis Ole e 1, the main sensitizing agent of Ole e 1-like family, could be use as diagnostic marker (Palomares et al., 2006a). Moreover, recombinant allergens can be engineered to produce hypoallergens or hypoallergenic derivatives that exhibit reduced or no allergenic activity but preserved T-cell epitopes and immunogenicity for safer forms of immunotherapy. Several recombinant DNA strategies, as well as synthetic peptide chemistry, have been employed to convert allergens into hypoallergenic derivatives: fragments, oligomers, point or deletion mutants, hybrids and T- or Bpeptides. In vitro and in vivo evaluation of hypoallergens is then required to identify the best vaccines for immunotherapy before clinical application. Clinical trials performed with hypoallergenic derivatives as well as recombinant wildtype allergens, have shown that these molecules can be used in immunotherapy in the near future. Hypoallergenic derivatives of Ole e 1 have been engineered based on the disruption of the immunodominant IgE epitope of the C-terminal of the molecule by producing one point and two deletion mutants (Marazuela et al., 2007). In addition, a peptide T of Ole e 1 has been designed based on epitope mapping studies (Marazuela et al., 2008a). To select the most suitable derivatives for immunotherapy they have been tested in vitro and in vivo (Marazuela et al., 2007, 2008a). Finally, recombinant allergens can be used to study the properties of olive pollen allergens, e.g. determination of the 3-D structures of Ole e 6 (Treviño et al., 2004) and CtD of Ole e 9 (Treviño et al., 2008) (Figure 110.2), assignment of disulfide bridges of Ole e 1 (González et al., 2000) and Ole e 9 CtD (Palomares et al., 2003), epitope mapping of Ole e 1 (González et al., 2006) and Ole e 9 (Palomares et al., 2003, 2005; Treviño et al., 2008), or epidemiological studies (Palomares et al., 2006b; Barber et al., 2007; Quiralte et al., 2007). So far, Ole e 1 is the only olive pollen recombinant allergen used in prick-test with patients in a study performed a few years ago, since current Spanish legislation does not allow the use of recombinant molecules on humans, although they may improve diagnosis and treatment of olive pollinosis.
110.10 NEW CONCEPTS FOR SPECIFIC IMMUNOTHERAPY USING OLE E 1 AS A MODEL Currently, allergen-specific immunotherapy is the only curative treatment available for allergy. Even though this treatment can offer protection, it has several disadvantages including
CHAPTER | 110 Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy
long duration, anaphylactic side-effects and limited efficacy. Mucosal tolerance induction with nasal vaccines based on free or encapsulated hypoallergenic derivatives is a promising alternative strategy to conventional immunotherapy. A mouse model of IgE sensitization to Ole e 1 mimicking the human B- and T-cell responses has been established for preclinical testing of new vaccines against allergy (Marazuela et al., 2008a). Four prophylactic approaches were conducted to investigate whether nasal tolerance induction with vaccines based on Ole e 1 or derivatives could prevent sensitization in mice. In a first approach, low doses of a nasal vaccine based on a deletion mutant of Ole e 1 were able to protect mice against sensitization to the allergen (Rodríguez et al., 2007a). Recently, Marazuela et al. (2008a) have demonstrated that prophylactic i.n. administration of a peptide T of Ole e 1 may substitute for the whole protein in protecting mice against subsequent sensitization to the allergen. Moreover, specific protection for the long term was maintained. In a third study, it was shown that i.n. administration of micrograms of the peptide T of Ole e 1 encapsulated in poly
1029
(lactide-co-glycolide) (PLG) microparticles as carrier vaccines prevented subsequent sensitization to the allergen (Marazuela et al., 2008b). In a previous work, PLG microparticles were described as a suitable vehicle vaccine for Ole e 1 that elicits a specific Th1-type response in mice, thus becoming a promising concept for allergy vaccine. During the last few years, exosome-based vaccines have been proposed as a novel strategy for treatment of human diseases including allergy. Exosomes are nanovesicles which are released in the extracellular environment by a variety of cell types and showed immunomodulatory properties. Our group has observed that intranasal administration of tolerogenic exosomes protects mice against sensitization to Ole e 1 (Prado et al., 2008) (Figure 110.4). In this respect, although the four mentioned approaches (using the same prophylactic protocol) suppress the most important clinical features of allergy – specific-IgE antibodies in serum, Th2-response and airway inflammation – exosomes have advantages over the previous reported vaccines. They are acellular and stable structures containing a wide array of cellular proteins, some
FIGURE 110.4 Intranasal pretreatment with tolerogenic exosomes protects mice against allergic sensitization. Exosomes were isolated from bronchoalveolar lavage fluid (BALF) from mice that were tolerized by respiratory exposure to Ole e 1. Exosomes isolated from naïve mice were used as controls. These exosomes were assayed as a preventive vaccine in a mouse model of allergy induced by intraperitoneal (i.p.) sensitization to Ole e 1 followed by airway allergen challenge. Mice were intranasal (i.n.) pretreated for 3 consecutive days with tolerogenic exosomes one week prior to sensitization/challenge with the allergen, and the allergic response was analyzed. Pretreatment with tolerogenic exosomes inhibit both airway inflammation and specific-IgE production. (A) Representative lung section stained with hematoxylin-eosin of tolerogenic exosomes-pretreated mice shows a reduced inflammatory cell infiltration compared to sham-pretreated mice that received control exosomes. Magnifications, ⫻ 20 (ExoTol and ExoCon), ⫻ 10 (naïve). (B) Serum IgE levels were determined by ELISA. Data are expressed as means ⫾ standard error (n ⫽ 15 mice/group) from three independent experiments. *p ⬍ 0.001. (Based on data from Prado et al., 2008). ExoTol, mice pretreated with tolerogenic exosomes; ExoCon, animals pretreated with control exosomes; Naïve, no-treated mice.
1030
SECTION | II Immunology and Inflammation
of which modulate immune responses. Since exosomes are natural antigen-transferring units between immune cells, they allow cross-presentation and contribute to amplify immune responses reducing the dose of antigen required to induce an immune response. Finally, ‘exosome display technology’ permits manipulation of their protein composition and tailoring for different functions. These studies emphasize the high potential of nasal vaccines against allergy. Despite the clinical relevance to test the therapeutic effects of these vaccines, the possibility of using prophylactic vaccines for early prevention in atopic individuals or children at risk has been proposed.
110.11 OLIVE FRUIT: A NEW SOURCE OF OLIVE ALLERGENS Olive fruit is consumed either directly or through processed products as olive oil. Allergic reactions to olive fruit and its derived products have rarely been documented in literature: few cases of contact dermatitis on workers caused by olive oil, olive-induced urticaria and anaphylactic shock to olive fruits have been reported. Recently Palomares et al. (2008b) have described an olive oil mill worker suffering from occupational asthma (OA) who was sensitized to olive fruit particles by mucosal exposure. Respiratory symptoms were improved when he was away from the workplace, but relapsed at work. A 23-kDa protein, which shows homology to thaumatin-like proteins (TLPs) from plant foods and pollen, has been isolated and identified as the major causative allergen of OA to olive fruits. TLPs belong to the PR-5 family, which play a role in the plant defense system against pathogens or adverse environmental factors. In recent years, TLPs have been recognized as a new class of pan-allergens in food and pollens which could be involved in the OAS.
SUMMARY POINTS ●
●
●
●
●
Olive pollen is one of the main causes of allergy in Mediterranean countries and some areas of America, South Africa, Japan and Australia. Ten allergens – named Ole e 1 to Ole e 10 – have been isolated, characterized, and many of them obtained as recombinant molecules. Recombinant proteins can be used to study allergen properties and the mechanism of allergy and immunotherapy. Moreover, recombinant allergens may improve diagnosis of allergic patients. Finally, hypoallergenic allergen derivatives have been in vivo and in vitro evaluated as vaccines for improved immunotherapy of allergy. A mouse model of allergy to olive pollen that mimics the features of allergic patients has been established. Preclinical studies in mice suggest the potential use of nasal vaccines for the treatment of allergy.
●
Olive fruit has been identified as a new source of allergens, causing occupational asthma.
REFERENCES Alché, J.D., Castro, A.J., Jiménez-López, J.C., Morales, S., Zafra, A., Hamman-Khalifa, A.M., Rodríguez-García, M.I., 2007. Differential characterization of olive pollen from different cultivars: biological and clinical implications. J. Investig. Allergol. Clin. Immunol. 17 (Suppl. 1), 69–75. Altmann, F., 2007. The role of protein glycosylation in allergy. Int. Arch. Allergy Immunol. 142, 99–115. Asturias, J.A., Arilla, M.C., Gómez-Bayón, N., Martínez, J., Martínez, A., Palacios, R., 1997. Cloning and expression of the panallergen profilin and the major allergen (Ole e 1) from olive pollen. J. Allergy Clin. Immunol. 100, 365–372. Barber, D., Moreno, C., Ledesma, A., Serrano, P., Galán, A., Villalba, M., Guerrra, F., Lombardero, M., Rodríguez, R., 2007. Degree of olive pollen exposure and sensitization patterns. Clinical implications. J. Investig. Allergol. Clin. Immunol. 17 (Suppl. 1), 63–68. Barral, P., Batanero, E., Palomares, O., Quiralte, J., Villalba, M., Rodríguez, R., 2004a. A major allergen from olive pollen defines a novel family of plant proteins and show intra- and interspecies crossreactivity. J. Immunol. 172, 3644–3651. Barral, P., Tejera, M.L., Treviño, M.A., Batanero, E., Villalba, M., Bruix, M., Rodríguez, R., 2004b. Recombinant expression of Ole e 6, a Cysenriched pollen allergen, in Pichia pastoris yeast. Detection of partial oxidation of methionine by NMR. Protein Expr. Purif. 37, 336–343. Barral, P., Suárez, C., Batanero, E., Alfonso, C., Alché, J.D., RodríguezGarcía, M.I., Villalba, M., Rivas, G., Rodríguez, R., 2005a. An olive pollen protein with allergenic activity, Ole e 10, defines a novel family of carbohydrate-binding modules and is potentially implicated in pollen germination. Biochem. J. 390, 77–84. Barral, P., Batanero, E., Villalba, M., Rodríguez, R., 2005b. Expression of the major olive pollen allergen Ole e 10 in the yeast Pichia pastoris: evidence of post-translational modifications. Protein Expr. Purif. 44, 147–154. Barral, P., Serrano, A.G., Batanero, E., Pérez-Gil, J., Villalba, M., Rodríguez, R., 2006. A recombinant functional variant of the olive pollen allergen Ole e 10 expressed in baculovirus system. J. Biotechnol. 121, 402–409. Batanero, E., Villalba, M., Rodríguez, R., 1994. Glycosylation site of the major allergen from olive tree pollen. Mol. Immunol. 31, 31–37. Batanero, E., Villalba, M., Ledesma, A., Puente, X.S., Rodríguez, R., 1996. Ole e 3, an olive-tree allergen, belongs to a widespread family of pollen proteins. Eur. J. Biochem. 241, 772–778. Batanero, E., Ledesma, A., Villalba, M., Rodríguez, R., 1997. Purification, amino acid sequence and immunological characterization of Ole e 6, a cysteine-enriched allergen from olive tree pollen. FEBS Lett. 410, 293–296. Batanero, E., Crespo, J.F., Monsalve, R.I., Martín-Esteban, M., Villalba, M., Rodríguez, R., 1999. IgE-binding and histamine release capabilities of the main carbohydrate component isolated from the major allergen of olive tree pollen, Ole e 1. J. Allergy Clin. Immunol. 103, 147–153. Blanca, M., Boulton, P., Brostoff, J., González-Reguera, I., 1983. Studies of the allergens of Olea europea pollen. Clin. Allergy 66, 473–478. Boluda, L., Alonso, C., Fernández-Caldas, E., 1998. Purification, characterization, and partial sequencing of two new allergens of Olea europaea. J. Allergy Clin. Immunol. 101, 210–216. Butteroni, C., Afferni, C., Barletta, B., Iacovacci, P., Corinti, S., Brunetto, B., Tinghino, R., Ariano, R., Panzani, R.C., Pini, C., Di Felice, G., 2005.
CHAPTER | 110 Olive Pollen Allergens: An Insight into Clinical, Diagnostic and Therapeutic Concepts of Allergy
Cloning and expression of the Olea europaea allergen Ole e 5, the pollen Cu/Zn superoxide dismutase. Int. Arch. Allergy Immunol. 137, 9–17. Cárdaba, B., Llanes, E., Chacártegui, M., Sastre, B., López, E., Mollá, R., del Pozo, V., Florido, F., Quiralte, J., Palomino, P., Lahoz, C., 2007. Modulation of allergic response by gene-environment interaction: olive pollen allergy. J. Investig. Allergol. Clin. Immunol. 17 (Suppl. 1), 83–87. Florido, J.F., Quiralte, J., Arias de Saavedra, J.M., Sáenz, B., Martín, E., 2002. An allergen from Olea europea pollen (Ole e 7) is associated with plant-derived food anaphylaxis. Allergy 57 (Suppl. 71), 53–59. González, E.M., Monsalve, R.I., Puente, X.S., Villalba, M., Rodríguez, R., 2000. Assignment of the disulphide bonds of Ole e 1, a major allergen of olive tree pollen involved in fertilization. J. Pept. Res. 55, 18–23. González, E.M., Villalba, M., Quiralte, J., Batanero, E., Roncal, F., Albar, J.P., Rodríguez, R., 2006. Analysis of IgE and IgG B-cell immunodominant regions of Ole e 1, the main allergen from olive pollen. Mol. Immunol. 43, 570–578. Huecas, S., Villalba, M., González, E.M., Martínez-Ruiz, A., Rodríguez, R., 1999. Production and detailed characterization of biologically active olive pollen allergen Ole e 1 secreted by the yeast Pichia pastoris. Eur. J. Biochem. 261, 539–546. Huecas, S., Villalba, M., Rodríguez, R., 2001. Ole e 9, a major olive pollen allergen is a 1,3-β-glucanase. Isolation, characterization, amino acid sequence, and tissue specificity. J. Biol. Chem. 276, 27959–27966. Ledesma, A., Rodríguez, R., Villalba, M., 1998a. Olive profilin. Molecular and immunological properties. Allergy 53, 520–526. Ledesma, A., Villalba, M., Batanero, E., Rodríguez, R., 1998b. Molecular cloning and expression of active Ole e 3, a major allergen from olivetree pollen and member of a novel family of Ca2⫹-binding proteins (polcalcins) involved in allergy. Eur. J. Biochem. 258, 454–459. Ledesma, A., Villalba, M., Rodríguez, R., 2000. Cloning, expression and characterization of a novel four EF-hand Ca2⫹-binding protein from olive pollen with allergenic activity. FEBS Lett. 466, 192–196. Ledesma, A., Villalba, M., Vivanco, F., Rodríguez, R., 2002a. Olive pollen allergen Ole e 8: identification in mature pollen and presence of Ole e 8-like proteins in different pollens. Allergy 57, 40–43. Ledesma, A., González, E., Pascual, C.Y., Quiralte, J., Villalba, M., Rodríguez, R., 2002b. Are Ca2⫹-binding motifs involved in the immunoglobuling E-binding of allergens? Olive pollen allergens as model of study. Clin. Exp. Allergy 32, 1476–1483. Ledesma, A., Moral, V., Villalba, M., Salinas, J., Rodríguez, R., 2006. Ca2⫹binding allergens from olive pollen exhibit biochemical and immunological activity when expressed in stable transgenic Arabidopsis. FEBS J. 273, 4425–4434. Liccardi, G., D’Amato, M., D’Amato, G., 1996. Oleaceae pollinosis: a review. Int. Arch. Allergy Immunol. 111, 210–217. Marazuela, E.G., Rodríguez, R., Barber, D., Villalba, M., Batanero, E., 2007. Hypoallergenic mutants of Ole e 1, the major olive pollen allergen, as candidates for allergy vaccines. Clin. Exp. Allergy 37, 251–260. Marazuela, E.G., Rodríguez, R., Fernández-García, H., García, M.S., Villalba, M., Batanero, E., 2008a. Intranasal immunization with a dominant T-cell epitope peptide of a major allergen of olive pollen prevents mice from sensitization to the whole allergen. Mol. Immunol. 45, 438–445. Marazuela, E.G., Prado, N., Moro, E., Fernández-García, H., Villalba, M., Rodríguez, R., Batanero, E., 2008b. Intranasal vaccination with poly(lactide-co-glycolide) microparticles containing a peptide T of Ole e 1 prevents mice against sensitization. Clin. Exp. Allergy 38, 520–528. Palomares, O., Villalba, M., Rodríguez, R., 2003. The C-terminal segment of the 1,3-β-glucanase Ole e 9 from olive (Olea europaea) pollen is
1031
an independent domain with allergenic activity: expression in Pichia pastoris and characterization. Biochem. J. 369, 593–601. Palomares, O., Villalba, M., Quiralte, J., Polo, F., Rodríguez, R., 2005. 1,3-βglucanases as candidates in latex-pollen-vegetable food cross-reactivity. Clin. Exp. Allergy 35, 345–351. Palomares, O., Swoboda, I., Villalba, M., Balic, N., Spitzauer, S., Rodríguez, R., Valenta, R., 2006a. The major allergen of olive pollen Ole e 1 is a diagnostic marker for sensitization to Oleaceae. Int. Arch. Allergy Immunol. 141, 110–118. Palomares, O., Villalba, M., Quiralte, J., Rodríguez, R., 2006b. Allergenic contribution of the IgE-reactive domains of the 1,3-β-glucanase Ole e 9: diagnostic value in olive pollen allergy. Ann. Allergy Asthma Immunol. 97, 61–65. Palomares, O., Fernández-Nieto, M., Villalba, M., Rodríguez, R., CuestaHerranz, J., 2008a. Occupational allergy in a researcher due to Ole e 9, an allergenic 1,3-β-glucanase from olive pollen. Allergy 63, 784–785. Palomares, O., Alcántara, M., Quiralte, J., Villalba, M., Garzón, F., Rodríguez, R., 2008b. Airway disease and thaumatin-like protein in an olive-oil mill worker. N. Engl. J. Med. 358, 1306–1308. Prado, N., Marazuela, E.G., Segura, E., Fernández-García, H., Villalba, M., Théry, C., Rodríguez, R., Batanero, E., 2008. Exosomes from bronchoalveolar fluid of tolerized mice prevent allergic reaction. J. Immunol. 181, 1519–1525. Quiralte, J., Palacios, L., Rodríguez, R., Cárdaba, B., Arias de Saavedra, J.M., Villalba, M., Florido, J.F., Lahoz, C., 2007. Modelling diseases: the allergens of Olea europaea pollen. J. Investig. Clin. Immunol. 17 (Suppl. 1), 76–82. Rodríguez, R., Villalba, M., Monsalve, R.I., Batanero, E., 2001. The spectrum of olive pollen allergens. Int. Arch. Allergy Immunol. 125, 185–195. Rodríguez, R., Villalba, M., Batanero, E., Palomares, O., Quiralte, J., Salamanca, G., Sirvent, S., Castro, L., Prado, N., 2007a. Olive pollen recombinant allergens: value in diagnosis and immunotherapy. J. Investig. Allergol. Clin. Immunol. 17 (Suppl. 1), 56–62. Rodríguez, R., Villalba, M., Batanero, E., Palomares, O., Salamanca, G., 2007b. Emerging pollen allergens. Biomed. Pharmacother. 61, 1–7. Salcedo, G., Sánchez-Monge, R., Barber, D., Díaz-Perales, A., 2007. Plant non-specific lipid transfer proteins: an interface between plant defence and human allergy. Biochim. Biophys. Acta 1771, 781–791. Tejera, M.L., Villalba, M., Batanero, E., Rodríguez, R., 1999. Identification, isolation, and characterization of Ole e 7, a new allergen of olive tree pollen. J. Allergy Clin. Immunol. 104, 797–802. Treviño, M.A., García-Mayoral, M.F., Barral, P., Villalba, M., Santoro, J., Rico, M., Rodríguez, R., Bruix, M., 2004. NMR solution structure of Ole e 6, a major allergen from olive tree pollen. J. Biol. Chem. 279, 39035–39041. Treviño, M.A., Palomares, O., Castrillo, I., Villalba, M., Rodríguez, R., Rico, M., Santoro, J., Bruix, M., 2008. Solution structure of the C-terminal domain of Ole e 9, a major allergen of olive pollen. Protein Sci. 17, 371–376. Van Ree, R., Cabanes-Macheteau, M., Akkerdaas, J., Milazzo, J.-P., Loutelier-Bourhis, C., Rayon, C., Villalba, M., Koppelman, S., Aalberse, R., Rodríguez, R., Faye, L., Lerouge, P., 2000. β(1,2)-xylose and α(1,3)-fucose residues have a strong contribution in IgE binding to plant glycoallergens. J. Biol. Chem. 275, 11451–11458. Villalba, M., Batanero, E., López-Otín, C., Sánchez, L.M., Monsalve, R.I., González de la Peña, M.A., Lahoz, C., Rodríguez, R., 1993. The amino acid sequence of Ole I, the major allergen from olive tree (Olea europaea) pollen. Eur. J. Biochem. 216, 863–869. Villalba, M., Batanero, E., Monsalve, R.I., González de la Peña, M.A., Lahoz, C., Rodríguez, R., 1994. Cloning and expression of Ole I, the major allergen from olive tree pollen. Polymorphism analysis and tissue specificity. J. Biol. Chem. 269, 15217–15222.
This page intentionally left blank
Chapter 111
Olive Oil and Septic Pulmonary Dysfunctions J. Glatzle Department of General and Transplantation Surgery, University Hospital of Tübingen, University of Tübingen, Germany
111.1 INTRODUCTION The gastrointestinal tract contains the largest lymphatic system in the body. There is some evidence that the gastrointestinal tract plays an important role in multiple organ dysfunction during peritonitis or sepsis. This would indicate its significance in producing inflammatory mediators which are drained into the systemic circulation causing systemic inflammatory responses and acute lung injury (Deitch et al., 2004; Glatzle et al., 2004, 2007a, b). The hypothesis that multiple organ dysfunction is caused by bacterial translocation from the gut into the blood circulation has been expanded. Many studies have thus far demonstrated that gut-derived non-bacterial inflammatory agents are responsible for acute lung injury during an acute insult to the gastrointestinal tract (Deitch, 2001; Deitch et al., 2001). Via the thoracic duct, the lung is the first organ exposed to gutderived inflammatory mediators drained into the mesenteric lymph. It has been demonstrated by Deitch et al. that diversion of the thoracic duct prevents acute lung injury, neutrophil activation, endothelial cell apoptosis and red blood cell dysfunction during an acute insult to the gastrointestinal tract (Deitch et al., 2001). Enteral immune-modulating diets, also called immunonutritions, are able to interact with the immune system in order to modulate the response of immune cells within the gastrointestinal tract. This might be beneficial during an acute insult to the gastrointestinal tract, such as in the case of hemorrhagic shock, peritonitis or sepsis. Recently it has been demonstrated that enteral nutrition containing long-chain fatty acids in the form of olive oil increases the likelihood of survival during sepsis in mice (Leite et al., 2005). The regulative pathway by which enteral nutrition with long-chain fatty acids mediates its immunemodulating effect remains unclear. However, it is likely that the so-called cholinergic anti-inflammatory pathway Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
discovered by the research team under Kevin Tracey plays a central role (Borovikova et al., 2000; Wang et al., 2003; Pavlov and Tracey, 2005).
111.2 DEFINITION OF IMMUNONUTRITION Nutrients which are able to modulate the immune system are called immunonutritions. Commercially available immunemodulating diets usually contain a variety of substances such as vitamins, trace elements, amino acids (arginin or glutamine), and all kinds of lipids (ω-3 fatty acids (fish oil), ω-6 fatty acids (soybean oil) or ω-9 fatty acids (olive oil)). Some randomized, controlled trials in either surgical or critically ill patients have so far indicated certain benefits for patients treated with commercially available immunonutritions. In general, a notable reduction in infections and in length of hospital stay was observed. However, no investigations have been carried out to evaluate the immune-modulating effect of a single component of commercially available immunonutritions. In the present chapter, we investigate the role of olive oil and its immune-modulating effect on the gastrointestinal tract during sepsis.
111.3 LIPID ABSORPTION AND ACTIVATION OF THE CHOLINERGIC ANTIINFLAMMATORY PATHWAY Lipids in the form of triglycerides are digested intraluminally and mainly into free fatty acids and monoglycerides. The digestion products with a chain length equal to, or greater than, 12 carbon atoms (C12) are absorbed into the
1033
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1034
FIGURE 111.1 Intestinal lipid absorption and chylomicron formation. Triglycerides are absorbed from the enterocytes, packed in chylomicrons and released contralaterally into the lymph lacteals. Apolipoproteins on the surface of the chylomicrons interact with endocrine cells and release cholecystokinin (CCK), which activates afferent nerve fibers via the CCK1 receptor. Short-chain fatty acids diffuse out of the enterocytes and are transported via the portal vein.
enterocytes, where they are resynthesized into triglycerides, packed into chylomicrons, removed from the enterocyte by exocytosis on the contralateral site and transported via the mesenteric lymph into the systemic circulation (Tso and Balint, 1986). Fatty acids with a chain length less than C10 diffuse out of the enterocyte and are transported via the portal blood to the liver, bound on albumin. In order to make the chylomicrons water-soluble, proteins on the surface of the chylomicrons, such as apolipoproteins, are needed. There is some evidence that apolipoprotein A-IV interacts with the endocrine cells within the gut mucosa, stimulating the release of cholecystokinin (CCK) (Figure 111.1). Cholecystokinin (CCK) activates vagal afferents through the CCK1 receptors on vagal afferent nerve terminals initiating a vago-vagal reflex pathway. Activation of the vagus nerve leads to the release of acetylcholine (Ach) on the efferent nerve terminals that binds to, among others, the α-7 nicotinic acetylcholine receptors (α-7 nAch receptor) located on the surface of cytokine-producing cells such as macrophages. Ligation of the nicotinic receptor by acetylcholine inhibits cytokine synthesis and release by preventing the activation and nuclear translocation of NF-KB, and by stimulating the anti-inflammatory JAK3-SOCS3 pathway (Wang et al., 2003, 2004) (Figure 111.2). This vago-vagal reflex pathway is also called the cholinergic anti-inflammatory pathway.
SECTION | II Immunology and Inflammation
FIGURE 111.2 The ‘cholinergic anti-inflammatory pathway’. Triglyceride absorption initiates the release of cholecystokinin (CCK) within the gut mucosa, which activates, via CCK1 receptors, the afferent part of the vago-vagal reflex pathway, connected through the nucleus tractus solitarius (NTS). Activation of the efferent part of this vago-vagal reflex pathway results in the release of acetylcholine (Ach) from the efferent vagal nerve terminals, activating macrophages via the α-7nAch receptor. Activation of the α-7nAch leads to a reduced release of inflammatory mediators such as tumor necrosis factor alpha (TNFα) or interleukin 6 (IL-6), whereas activation of the toll-like receptor (TLR) by bacterial products leads to an increased release of inflammatory cytokines such as TNFα and IL-6.
111.4 EXPERIMENTAL SETUP The efferent side of this cholinergic anti-inflammatory pathway has been well examined and confirmed. However, it is hypothetical and uncertain whether enteral immunonutrition with long-chain fatty acids in the form of olive oil is able to activate the afferent part of this reflex pathway, thereby reducing the inflammatory mediator release from the gastrointestinal tract during sepsis, and whether enteral immunonutrition with olive oil would be able to reduce septic pulmonary dysfunction. In order to address this issue, we developed an animal model with donor and recipient rats (Figure 111.3). Lipopolysaccharide (LPS, 5 mg kg⫺1), a glycolipid component of the Gram-negative bacteria membrane, commonly used in experimental sepsis models (since recognition of LPS by the host immune system initiates an inflammatory cascade similar to a Gram-negative infection) was injected intraperitoneally in donor rats (healthy male Sprague-Dawley rats) in order to induce an abdominal sepsis (Deitch, 1998). The LPS dose used in the present study was a sub-lethal dose, since none of the animals died during sepsis. However, the dose
1035
CHAPTER | 111 Olive Oil and Septic Pulmonary Dysfunctions
was sufficient enough to induce an inflammatory response in the gut wall of the donor rats and resulted in the release of significantly increased amounts of pro-inflammatory mediators such as TNFα (Figure 111.4A). Sepsis lymph likely consists of a ‘cocktail’ of gastrointestinal inflammatory and antiinflammatory mediators released from the gut during sepsis. Therefore, TNFα seems to represent one of several inflammatory cytokines whose level increases in sepsis lymph. Continuous enteral immunonutrition with olive oil (ClinOleic 1%, Baxter) during sepsis significantly reduced the release of the pro-inflammatory cytokine TNFα into the mesenteric lymph of donor rats (Figure 111.4B). The greatest effects were seen within the first 2 hours after septic shock was induced. The reduction of TNFα was more than 75% as compared to animals treated without the immunemodulating diet. Mesenteric lymph was collected from donor rats during sepsis (sepsis lymph, SL) and during sepsis with immunonutrition of 1% olive oil (SL-OO). Sepsis lymph from
FIGURE 111.3 Animal model of sepsis. Animal model of donor rats and recipient rats. Sepsis was induced in donor rats in order to investigate the inflammatory mediator release from the gastrointestinal tract during sepsis and the effect of immune modulation diet on inflammatory mediator release from the gastrointestinal tract. The effect of mesenteric lymph, collected during control conditions and during sepsis with or without enteral immunonutrition on pulmonary dysfunction was investigated in separate healthy recipient rats. A
these donor rats was then re-infused into the jugular vein of separate healthy recipient rats for 90 minutes with an infusion rate of 2.5 mL h⫺1, corresponding to the flow-rate of mesenteric lymph of the donor rats. The lung was harvested immediately after the termination of lymph infusion, and fixed in paraformaldehyde for histomorphological analysis. Three parameters were observed for septic pulmonary dysfunction: (1) the thickness of the alveolar walls as a parameter for the distance of oxygen transport; (2) myeloperoxidase (MPO) positive cells as a marker for the inflammatory response, and (3) TUNEL-positive cells as a marker for cell apoptosis causing permanent lung injury. Sepsis-lymph infusion into healthy recipient rats induced a more than two-fold increase in alveolar wall thickness as compared to control animals receiving NaCl as an intravenous infusion (Figures 111.5 and 111.6). This indicates that the distance for oxygen diffusion is more than twice as long as under healthy conditions. In contrast, infusion of sepsis lymph obtained during enteral immunonutrition with olive oil (SL-OO) did not cause thickening of the alveolar walls in recipient rats (Figure 111.5). Further, infusion of sepsis lymph (SL) induced a significant inflammatory response in the lungs of healthy recipient rats as compared to the control group. The number of myeloperoxidase (MPO)-positive cells increased more than nine-fold in rats receiving sepsis lymph as compared to rats receiving physiological saline (control). In contrast, infusion of sepsis lymph obtained during immunonutrition with olive oil (SL-OO) induced a clearly ameliorated inflammatory reaction, about 60% less compared to sepsis lymph infusion obtained without immunonutrition (Figure 111.7). Infusion of sepsis lymph (SL) induced a significant increase of permanent lung injury in the lungs of recipient rats, indicated by the increase in apoptotic cells (positive for TUNEL immunoreactivity) as compared to the control group. This increase was more than 3.5-fold as compared B 15 000
*
12 500
TNFα in ClinOleic (1%) sepsis lymph (pg mL–1)
TNFα in sepsis lymph (pg mL–1)
15 000
10 000 **
7500 5000 2500 0
Basal
1+2
3+4
5+6
Time (h)
12 500 10 000 7500 5000
*
2500 0
Basal
1+2
3+4
5+6
Time (h) * p < 0.0005 vs. basal ** p < 0.05 vs. basal
* p < 0.05 vs. basal
FIGURE 111.4 Inflammatory mediator release into mesenteric lymph during sepsis (A) and during sepsis with simultaneous enteral immunonutrition (B). The histograms show the tumor necrosis factor alpha (TNFα) release from the gastrointestinal tract into mesenteric lymph. Lymph was collected during basal condition without sepsis and after the induction of sepsis with LPS. Sepsis lymph was collected thereafter in 2-hour intervals for TNFα measurements.
1036
SECTION | II Immunology and Inflammation
15.0 TUNEL positive cells in lung tissue (number/optical section)
17.5 Thickness alveolar walls (μm)
* 15.0 12.5 10.0 7.5 5.0 2.5 0.0
12.5 10.0 7.5 5.0 2.5 0.0
Control
SL
SL-OO
* p < 0.01 vs. control or SL-OO FIGURE 111.5 Increase in alveolar wall thickness after infusion of sepsis lymph. The histogram shows the thickness of the alveolar walls (μm) after infusion of physiological saline (white bar), infusion of sepsis lymph (SL, black bar) and infusion of sepsis lymph obtained during enteral immunonutrition with olive oil (SL-OO, gray bar).
* Control
SL
SL-OO
* p < 0.01 vs. control or SL-OO FIGURE 111.8 Increase of TUNEL positive cells in the lung of recipient rats after infusion of sepsis lymph, indicating programmed cell death. The histogram shows the number of TUNEL positive cells per optical section after infusion of physiological saline (white bar), infusion of sepsis lymph (SL, black bar) and infusion of sepsis lymph obtained during enteral immunonutrition with olive oil (SL-OO, gray bar).
to the control animals receiving physiological saline. In contrast, infusion of sepsis lymph obtained during enteral immunonutrition with olive oil (SL-OO) did not induce apoptosis in the lungs of recipient rats (Figure 111.8).
111.5 CONCLUSION
FIGURE 111.6 Specimens of the lung after infusion of physiological saline (NaCl) or sepsis lymph. The picture shows hematoxylin and eosin stainings of the lung after infusion of physiological saline (left picture) and infusion of sepsis lymph (right picture).
MPO positive cells in lung tissue (number/optical section)
60 50 40 30
*
20 10 0
Control
SL
SL-OO *p < 0.001 vs. SL
FIGURE 111.7 Increase of myeloperoxidase (MPO) positive cells in the lung of recipient rats after infusion of sepsis lymph. The histogram shows the number of myeloperoxidase (MPO) positive cells per optical section after infusion of physiological saline (white bar), infusion of sepsis lymph (SL, black bar) and infusion of sepsis lymph obtained during enteral immunonutrition with olive oil (SL-OO, gray bar).
The present study demonstrates that inflammatory mediators are released into mesenteric lymph during sepsis and that these mediators are associated with septic pulmonary dysfunctions. The release of inflammatory mediators into mesenteric lymph during sepsis could be significantly reduced by enteral immunonutrition containing olive oil. It could also be demonstrated that sepsis lymph causes pulmonary dysfunction, such as an increased distance for oxygen transport, inflammation, and apoptosis. In contrast, sepsis lymph collected during an enteral immune-modulating diet in the form of olive oil attenuated septic pulmonary dysfunction. According to the present findings, the cholinergic antiinflammatory pathway may explain to some extent the reduced TNFα release into mesenteric lymph during sepsis, when the rats were intestinally infused with olive oil. Therefore, sepsis lymph obtained during enteral immunonutrition with olive oil might have a reduced inflammatory potency in causing septic pulmonary dysfunction. However, apolipoprotein A-IV also has a distinct anti-inflammatory potency, since experimental colitis could be successfully treated by exogenous apo-A-IV administration in mice (Vowinkel et al., 2004). The reduced inflammatory potency of sepsis lymph obtained during enteral immunonutrition with olive oil might indicate the combined effect of activation of the cholinergic anti-inflammatory pathway and anti-inflammatory components of lipid digestion, such as the increased apolipoprotein A-IV secretion.
CHAPTER | 111 Olive Oil and Septic Pulmonary Dysfunctions
SUMMARY POINTS ●
●
●
This recently discovered cholinergic anti-inflammatory pathway opens up several possibilities for experimental and clinical studies in critically ill patients. An enteral nutritional diet with long-chain fatty acids in the source of olive oil might be an inexpensive and useful treatment in reducing inflammatory mediator release from the gastrointestinal tract during an acute insult to the gut such as through abdominal operations, peritonitis or sepsis. However, the optimal timing and use of an immunemodulating diet, such as the enteral application of olive oil in critically ill patients, remains to be addressed.
ACKNOWLEDGMENTS I would like to thank my mentors Helen Raybould and Alfred Königsrainer for their constant support and training. I would also like to thank Hannes Schramm and Karin Glatzle who printed Figures 111.1 and 111.2, and Figure 111.3, respectively.
REFERENCES Borovikova, L.V., Ivanova, S., Zhang, M., Yang, H., Botchkina, G.I., Watkins, L.R., Wang, H., Abumrad, N., Eaton, J.W., Tracey, K.J., 2000. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 405, 458–462. Deitch, E.A., 1998. Animal models of sepsis and shock: a review and lessons learned. Shock 9, 1–11. Deitch, E.A., 2001. Role of the gut lymphatic system in multiple organ failure. Curr. Opin. Crit. Care 7, 92–98. Deitch, E.A., Adams, C., Lu, Q., Xu, D.Z., 2001. A time course study of the protective effect of mesenteric lymph duct ligation on hemorrhagic shock-induced pulmonary injury and the toxic effects of lymph from shocked rats on endothelial cell monolayer permeability. Surgery 129, 39–47.
1037
Deitch, E.A., Forsythe, R., Anjaria, D., Livingston, D.H., Lu, Q., Xu, D.Z., Redl, H., 2004. The role of lymph factors in lung injury, bone marrow suppression, and endothelial cell dysfunction in a primate model of trauma-hemorrhagic shock. Shock 22, 221–228. Glatzle, J., Beckert, S., Kasparek, M.S., Mueller, M.H., Mayer, P., Meile, T., Konigsrainer, A., Steurer, W., 2007a. Olive oil is more potent than fish oil to reduce septic pulmonary dysfunctions in rats. Langenbecks Arch. Surg. 392, 323–329. Glatzle, J., Kasparek, M.S., Mueller, M.H., Binder, F., Meile, T., Kreis, M.E., Konigsrainer, A., Steurer, W., 2007b. Enteral immunonutrition during sepsis prevents pulmonary dysfunction in a rat model. J. Gastrointest. Surg. 11, 719–724. Glatzle, J., Leutenegger, C.M., Mueller, M.H., Kreis, M.E., Raybould, H.E., Zittel, T.T., 2004. Mesenteric lymph collected during peritonitis or sepsis potently inhibits gastric motility in rats. J. Gastrointest. Surg. 8, 645–652. Leite, M.S., Pacheco, P., Gomes, R.N., Guedes, A.T., Castro-FariaNeto, H.C., Bozza, P.T., Koatz, V.L., 2005. Mechanisms of increased survival after lipopolysaccharide-induced endotoxic shock in mice consuming olive oil-enriched diet. Shock 23, 173–178. Pavlov, V.A., Tracey, K.J., 2005. The cholinergic anti-inflammatory pathway. Brain Behav. Immun. 19, 493–499. Tso, P., Balint, J.A., 1986. Formation and transport of chylomicrons by enterocytes to the lymphatics. Am. J. Physiol. 250, G715–G726. Vowinkel, T., Mori, M., Krieglstein, C.F., Russell, J., Saijo, F., Bharwani, S., Turnage, R.H., Davidson, W.S., Tso, P., Granger, D.N., Kalogeris, T.J., 2004. Apolipoprotein A-IV inhibits experimental colitis. J. Clin. Invest. 114, 260–269. Wang, H., Liao, H., Ochani, M., Justiniani, M., Lin, X., Yang, L., Al Abed, Y., Wang, H., Metz, C., Miller, E.J., Tracey, K.J., Ulloa, L., 2004. Cholinergic agonists inhibit HMGB1 release and improve survival in experimental sepsis. Nat. Med. 10, 1216–1221. Wang, H., Yu, M., Ochani, M., Amella, C.A., Tanovic, M., Susarla, S., Li, J.H., Wang, H., Yang, H., Ulloa, L., Al Abed, Y., Czura, C.J., Tracey, K.J., 2003. Nicotinic acetylcholine receptor alpha7 subunit is an essential regulator of inflammation. Nature 421, 384–388.
This page intentionally left blank
Chapter 112
Olive Oil and Immune Resistance to Infectious Microorganisms María A. Puertollano, Elena Puertollano, Gerardo Álvarez de Cienfuegos and Manuel A. de Pablo Unit of Microbiology, Department of Health Sciences, Faculty of Experimental Sciences, University of Jaén, Spain
112.1 INTRODUCTION The interdependency among the disciplines of nutrition, immunology and infection was suitably recognized in the 1970s, when immunological determinations were incorporated as a critical component of assessing nutritional status (Scrimshaw and SanGiovanni, 1997; Klasing and Leshchinsky, 2000). The evolution has provided both humans and animals with a complex network of cells and factors responsible for the host protection from infectious agents, which exist in the environment (bacteria, viruses, fungi, parasites). It was clearly established that different nutrients are capable of modulating the function of these cells and factors that constitute immune system. Therefore these nutrients are strongly associated with increased or reduced susceptibility to infectious diseases (Field et al., 2002). Different epidemiological, experimental or clinical studies have intensely examined the fundamental role of different dietary lipids in the regulation of immune system functions. Biochemical or metabolic activities have classically been attributed to fatty acids, which are important for the cell biological functions, because they serve as a main source of metabolic energy; as the substrates for cell membrane biogenesis, contributing to the physical and biological properties; as covalent modifiers of protein structure; as precursors for synthesis of many intracellular signalling molecules (biological lipids mediators like prostaglandins (PG), leukotrienes (LT), resolvins and lipoxins); and as regulators of gene expression (reviewed in Mills et al., 2005). However, fatty acids are also important mediators of immune system functions. Here, we review the studies reporting the impact of dietary lipids on the immune response, which have allowed a better understanding of the influence of olive oil on the host natural resistance to pathogenic agents. Although the Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
effects of olive oil on the immune functions are less known, currently, there is enough evidence for corroborating the protective and beneficial effects of olives on human health and particularly on the immune response.
112.2 IMMUNE SYSTEM: A BRIEF DESCRIPTION The concept of immunity may be defined as a state of resistance and protection against environmental pathogens (such as viruses, bacteria, fungi or parasites) that allows the elimination and the destruction of these infectious agents. Therefore, the immune system constitutes an integrated and coordinated defense network that comprises two separate but interacting and interdependent types: (i) the innate, natural or non-specific immune system or (ii) the acquired, adaptive or specific immune system. Both intrinsic components of immunity involve the production and release of different factors and the recruitment of various types of cells (leukocytes), which are originated in bone marrow. In addition, both innate and acquired immune responses are integrated through the interactions of cells and cytokine production generated as a result of a specific stimulus. Thus, the infectious agent is eliminated and immunological memory remains; therefore, the response of the immune system against a second exposure by a similar antigen is more rapid (Deveraux, 2002) (Table 112.1). These cells circulate in the bloodstream or are organized into lymphoid organs. They are divided into two broad groups: phagocytes (macrophages/monocytes, eosinophils, neutrophils and basophils) and lymphocytes (T lymphocytes, B lymphocytes and natural killer cells (NK)). Similarly, T lymphocytes may be divided into two fundamental groups: on the one hand, Th1 lymphocytes, which are regulated by interleukin-12 (IL-12) and interferon-γ (IFN-γ), produce
1039
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1040
SECTION | II Immunology and Inflammation
TABLE 112.1 Key features of the immune system. 1. 2. 3. 4. 5.
Immunity is defined as a state of resistance or protection against pathogens The main function of the immune system is the elimination and destruction of infectious agents The immune system comprises two types: the innate and acquired immune response The immune system is characterized by an immunological memory after the exposition to an antigen Different types of immune cells circulate in the bloodstream: phagocytes and lymphocytes. These cells are differentiated with specialized functions 6. The immune system should recognize the self from non-self and eliminate non-self (foreign agents) 7. The loss of this discriminatory function leads to the generation of autoimmune diseases characterized by an overactivation of immune system (inflammatory disorders)
Macrophages, Cytotoxic T lymphocytes, B lymphocytes, Natural killer cells
B lymphocytes
Chronic inflammation Cell-mediated immunity
Th0
Inflammation Antibodies
+
IL-4 IL-5
–
IL-12
IL-10
–
+
+
IL-4
+ IFN-γ TNF-α
Th1
Th2
IL-4 IL-5
IFN-γ
–
Eosinophils
IL-4 IL-2
– IL-10
FIGURE 112.1 Interaction between Th1 and Th2 lymphocyte response. IL, interleukin; IFN, interferon; NK, natural killer; TNF, tumor necrosis factor; ⫹, activation; ⫺, inhibition.
pro-inflammatory cytokines (such as IL-2 or IFN-γ) and activate macrophages, NK cells and cytotoxic T lymphocytes; on the other hand, Th2 lymphocytes, which are regulated by IL-4, and are specialized in the production of cytokines with an anti-inflammatory activity (such as IL-4, IL-5 or IL-10). Hence, infection with intracellular pathogens (several parasites, certain bacteria and viruses) will induce the differentiation along the Th1 pathway, whereas infection with extracellular pathogens (the most part of bacteria, fungi and several parasites) will promote the differentiation along the Th2 pathway (Figure 112.1).
112.3 FATTY ACIDS AND IMMUNE SYSTEM FUNCTIONS Initially, epidemiological studies with Greenland Eskimos in the early 1980s determined the crucial role of long-chain
n-3 polyunsaturated fatty acids as responsible for a low incidence of inflammatory disorders in this population (Kromann and Green, 1980). On the basis of this line of evidence, different studies have suggested that an excess in the administration of diets containing long-chain n-3 polyunsaturated fatty acids may exert a detrimental effect and they may be harmful to human health due mainly to a generalized immunosuppression and to a reduced inflammatory activity. Nevertheless, a low incidence of inflammatory disorders has been associated with Mediterranean diet intake, in which the first and most important source of fatty acids is supplied by olive oil (mainly constituted by n-9 monounsaturated fatty acids and other minor components with antioxidant function) (Stark and Madar, 2002). Fatty acids are divided into two great families: n-3 and n-6. These are considered as essential fatty acids to the mammalian cells and should be administered in the diet. Essential fatty acids are constituted by n-3 series derived
1041
CHAPTER | 112 Olive Oil and Immune Resistance to Infectious Microorganisms
18:0 (stearic acid) Δ9-desaturase
18:3 (n-3) (α-linolenic acid)
Δ12-desaturase
Δ15-desaturase
18:2 (n-6) (Linoleic acid)
18:1 (n-9) (Oleic acid) Desaturase
Desaturase
18:2 (n-9)
18:3 (n-6) (γ-linolenic acid)
18:4 (n-3) (Stearidonic acid)
Elongase 20:2 (n-9)
Elongase
Desaturase
20:4 (n-3) 20:3 (n-6) (dihomo γ-linolenic acid)
20:3 (n-9) (eicosatrienoic acid) Elongase
3-series prostaglandins and thromboxanes
5-series leukotrienes
Desaturase
Cyclooxygenase
Cyclooxygenase 20:5 (n-3)
20:4 (n-6)
22:3 (n-9)
2-series prostaglandins and thromboxanes
(Arachidonic acid)
(Eicosapentaenoic acid)
Lipooxygenase
Lipooxygenase
4-series leukotrienes, lipoxins, HPETE, 12 and 15 HETEs
22:6 (n-3) (Docosahexaenoic acid) Supplied from the diet or biosynthesis
Supplied from the diet (essential fatty acids)
FIGURE 112.2 Different pathways of n-3, n-6 and n-9 fatty acid synthesis. HETEs, hydroxyl-eicosatetraenoic acid; HPETEs, hydroperoxyeicosatetraenoic acids.
from linolenic acid and n-6 series derived from linoleic acid. Both n-3 and n-6 polyunsaturated fatty acids are not interconvertible in animals. Different biochemical processes lead to the production of eicosapentaenoic acid or docosahexaenoic acid (from linolenic acid), as well as arachidonic acid (from linoleic acid). Likewise, another family of fatty acids (not essential fatty acids) such as n-9 series derived from oleic acid (monounsaturated fatty acid) also seems to play an important role in the immunomodulatory process (Yaqoob, 2002; Puertollano et al., 2007) (Figure 112.2).
112.4 MECHANISMS WHEREBY FATTY ACIDS MODULATE IMMUNE SYSTEM FUNCTIONS In recent years, numerous studies have attempted to elucidate the mechanisms whereby several dietary lipids produce a potential effect on immune system functions (Clamp et al., 1997; Mills et al., 2005). Current investigations examining the mechanisms of action responsible for the modulation of immune system functions have proposed various hypotheses, which are summarized in Table 112.2.
TABLE 112.2 Hypothetical mechanisms of dietary lipids on immune functions: factors determining the modulation of immune system. • • • • • • • •
Eicosanoid production Membrane fluidity and lipid rafts Oxidative stress Dietary lipids and signaling transduction Dietary lipids and gene expression Apoptosis Ability to antigen presentation Modulation of gastrointestinal microbiota
112.5 OLIVE OIL AND ALTERATION OF IMMUNE FUNCTIONS Oleic acid is the main fatty acid contained in olive oil; however, other additional different chemical components such as sterols, alcohols, antioxidants, and other fatty acids (stearic, palmitic, linoleic and α-linolenic acids) of minor relevance participate in the constitution of this fat. Olive oil has traditionally been used as placebo treatment in the studies
1042
SECTION | II Immunology and Inflammation
TABLE 112.3 Summary of data demonstrating the effect of olive oil on both animals’ and humans’ immune system functions. Study properties
Species
Duration (weeks)
Results
References
Administration of olive oil (6.8 g oleic acid day⫺1) to patients suffering from rheumatoid arthritis
Human
24
Suppressed immune functions, which are associated with beneficial effects on the progression of rheumatoid arthritis
Kremer et al., 1990
Elderly subjects suffering from rheumatoid arthritis
Human
27
Risk decreased by high consumption of olive oil
Linos et al., 1991
Diet containing 200 g kg⫺1 of olive oil
Rat
10
Inhibited spleen and lymph node proliferation
Yaqoob et al., 1994a
Diet containing 200 g kg⫺1 of olive oil
Rat
10
Decreased spleen NK cell activity
Yaqoob et al., 1994b
Diets containing 200 g kg⫺1 olive oil, sunflower oil or high oleic sunflower oil
Rat
6
The immunosuppressive effects attributed to an olive oil diet are mainly due to oleic acid
Jeffery et al., 1997a
Administration of a diet containing olive oil (18.4% of energy) to healthy middleaged men
Human
8
Unchanged PBMC proliferative response. Suppressed NK cell activity. Reduced ICAM-1 expression
Yaqoob et al., 1998
Diet containing 150 g kg⫺1 of olive oil
Mouse
12
Increased IL-2 and TNF-α production compared with a low fat diet
De Pablo et al., 1998a,b
In vitro exposure of human PBMC to an olive oil-based lipid emulsion
Human
i.v.
Decreased production of pro-inflammatory cytokines. Maintenance of immunity and reduced inflammatory response
Granato et al., 2000; Reimund et al., 2004
Diet containing 200 g kg⫺1 of olive oil and challenged with Listeria monocytogenes
Mouse
4
Higher survival rates and decreased recovery of viable bacteria from spleen compared with a fish oil diet
De Pablo et al., 2000
Olive oil-based lipid emulsion
Rat
4
Increased expression of the IL-2 receptor (CD25)
Moussa et al., 2000
Diets containing 200 g kg⫺1 olive oil, fish oil or hydrogenated coconut oil and infected with L. monocytogenes
Mouse
4
Increased invasion and adherence of L. monocytogenes to splenic cells
Puertollano et al., 2002
Diets containing 200 g kg⫺1 olive oil, fish oil or hydrogenated coconut oil, challenged with L. monocytogenes and treated with NAC
Mouse
4
Administration of NAC exerts a moderate detrimental effect after challenge with L. monocytogenes
Puertollano et al., 2003
Diets containing 200 g kg⫺1 olive oil, fish oil or hydrogenated coconut oil and challenged with L. monocytogenes
Mouse
4
Production of IL-12 is not modified, but increased IL-4 production after L. monocytogenes infection
Puertollano et al., 2004
Supplementation of olive oil and fish oil in patients suffering from rheumatoid arthritis
Human
24
Patients showed a significant improvement when treated with a combination of both olive oil and fish oil
Berbert et al., 2005
1043
CHAPTER | 112 Olive Oil and Immune Resistance to Infectious Microorganisms
Diets containing 7% (w/w) olive oil diet for 6 weeks and injected with LPS
Mouse
6
Decreased inflammatory response and improved survival
Leite et al., 2005
Diet containing 200 g kg⫺1 of olive oil and challenged with Listeria monocytogenes in cyclophosphamidetreated mice
Mouse
4
Increased resistance to infection in immunosuppressed animals
Cruz-Chamorro et al., 2007
This table summarizes different studies investigating the administration of olive oil and its action on immune system or on host resistance after exposition to an infectious agent. i.v., in vitro; ICAM-1, intercellular adhesion molecule-1; IL, interleukin; LPS, lipopolysaccharide; NAC, N-acetylcysteine; NK, natural killer; PBMC, peripheral blood mononuclear cells; TNF, tumor necrosis factor.
investigating the potential action of other dietary lipids (particularly fish oil) on the modulation of immune functions. Thus, monounsaturated fatty acids that constitute olive oil were initially considered as neutral fatty acids. Nevertheless, different studies demonstrated that olive oil is clearly involved in anti-inflammatory activities, in the modulation of immune system functions and in the alteration of host defense against infectious microorganisms (Table 112.3). Evidence from intervention studies have indicated that either fish oil or olive oil diets improve significantly the symptoms of inflammatory disorders due to a suppression of the immune system by altering the mediators that participate in the immune response. An important study determined the effects of diets containing fish oil on the progression and severity of rheumatoid arthritis (Kremer et al., 1990). Although the group fed a diet containing olive oil was used as control, a reduction of symptoms was detected after olive oil administration (Kremer et al., 1990). Epidemiological studies suggest that olive oil improves the symptoms of rheumatoid arthritis by alteration of the production of inflammatory response mediators (Linos et al., 1991, 1999). The anti-inflammatory activity of oleic acid appears to be associated with the production of its metabolite eicosatrienoic acid (20:3 n-9), which is a potent inhibitor of the LTB4 synthesis. Therefore, the anti-inflammatory effects of oleic acid are exerted through a mechanism similar to that of fish oil, which contains eicosapentaenoic acid (James et al., 1993). It is probable that the beneficial effects of olive oil may be in part attributed to the presence of natural antioxidants, contributing to increase the stability of the oil (Linos et al., 1999). In spite of the fact that the effects of olive oil on the immune system are related to oleic acid rather than other minor components of this fat (Jeffery et al., 1997a), a recent study has demonstrated that the polyphenolic substances contained in olive oil possess anti-inflammatory properties. This feature underlines the importance of these compounds with beneficial and protective antioxidant effects (Beauchamp et al., 2005).
112.5.1 Lymphocyte Proliferation Studies investigating the action of fatty acids on the immune functions have demonstrated that a diet containing olive oil
supplied to animals is capable of promoting a significant reduction of lymphocyte proliferation in response to the mitogen concanavalin A (Con A) (Yaqoob et al., 1994a). Although some animal investigations have reported a reduction of lymphocyte proliferation in response to Con A, investigations from our laboratory did not find a significant effect of olive oil on the proliferation of murine lymphocytes stimulated with either Con A or lipopolysaccharide (LPS) (Puertollano et al., 2002). In contrast, human studies have produced conflicting results, because the administration of a diet containing olive oil did not affect proliferation of mitogen-stimulated lymphocytes (Yaqoob et al., 1998). It is probable that the observed differences between animal and human studies may be attributed to the amount of monounsaturated fatty acids contained in the diet. Thus, monounsaturated fatty acids contributed approximately 25–30% of total energy in animal studies, whereas diets constituted by monounsaturated fatty acids supplied approximately 18% of total energy in human studies (Yaqoob, 2002).
112.5.2 Cytokine Production Cytokine production is modified after the administration of diets containing olive oil. Cytokines are proteins that regulate the growth and differentiation of lymphocyte subsets triggering and regulating cells that participate in the inflammatory response. Interleukin-2 (IL-2) is an important cytokine responsible for the proliferation of T lymphocytes. Administration of diets containing olive oil for 8 or 12 weeks produced an increase in IL-2 production in mice (Yaqoob and Calder, 1995; de Pablo et al., 1998a). IL-4 production, a cytokine with antiinflammatory functions, was increased in mice after the administration of a diet containing olive oil for 4 weeks (Puertollano et al., 2004). In contrast, the production of IL-10, another anti-inflammatory cytokine, was decreased after the administration of a diet containing olive oil, although the production of IL-10 was not different from that seen after feeding mice a low-fat diet (Yaqoob et al., 1994a). Finally, the concentration of IL-12, a cytokine that participates as a pro-inflammatory protein, was reduced in mice fed with diets containing olive oil (Puertollano
1044
et al., 2004). Other cytokines are mainly produced by monocytes/macrophages. The synthesis of these factors is also modulated by the administration of a diet containing olive oil. Thus, IL-1 production was diminished, whereas both TNF-α and IL-6 production were increased or not modified in mice (Yaqoob et al., 1995; de Pablo et al., 1998b). Overall, diets containing olive oil are also related to the suppression of cytokine production, but this effect is not as severe as that produced by the administration of a fish oil-rich diet. Although several investigations have reported that olive oil exerts an immunosuppressive activity as well as an anti-inflammatory action, a nonsignificant reduction in TNF-α production was observed in peritoneal macrophages of rats fed a diet containing olive oil for 8 weeks in response to LPS.
112.5.3 Activity of Natural Killer Cells Natural killer (NK) cell activity is modulated by the action of certain dietary lipids. NK cells constitute a central lymphocyte subset found in blood and spleen that destroy virusinvaded cells or transformed cells. In animal studies, the administration of an olive oil diet reduced NK cell activity (Yaqoob et al., 1994b). Comparison of data among animals fed different amounts of oleic acid showed a linear negative relationship, indicating that oleic acid is responsible for a reduction of NK cell activity in rats (Jeffery et al., 1997b). Once again, human studies have not confirmed the animal results; the measurement of NK cell activity after the administration for 1 or 2 months of a diet containing olive oil did not show significant differences from the control group, although the activity of these cells declined after 2 months of consuming this diet (Yaqoob et al., 1998).
112.5.4 Adhesion Molecules An interesting line of investigation is based on the potential effect of dietary lipids and the effects promoted in the expression of adhesion molecules. Thus, there was a decrease in the levels of expression of CD-2, lymphocyte function antigen-1 (LFA-1) and intercellular adhesion molecule-1 (ICAM-1) on spleen lymphocytes from rats fed with diets containing olive oil or fish oil (Sanderson et al., 1995). Similarly, an olive oil diet reduced the levels of ICAM-1 after 2 months of dietary administration in humans (Yaqoob et al., 1998). This molecule participates in leukocyte–leukocyte adhesion as well as in leukocyte–endothelial cell adhesion, and it is also expressed on mononuclear cells isolated from inflamed synovium in individuals suffering from rheumatoid arthritis, and it also participates in the recruitment of mononuclear cells responsible for the formation and development of atherosclerotic plaque.
SECTION | II Immunology and Inflammation
112.6 OLIVE OIL AND INFECTIOUS DISEASE RESISTANCE: AN INTRODUCTION Previous experimental studies reported the involvement of long-chain n-3 polyunsaturated fatty acids in the reduction of natural resistance against different infectious agents such as Salmonella typhimurium serovar Typhimurium, Mycobacterium tuberculosis, Listeria monocytogenes, influenza virus or parasites (reviewed in Anderson and Fritsche, 2002; de Pablo et al., 2002; Puertollano et al., 2007). Obviously, available data are only referred to experimental animals and the effect of dietary lipids on infectious resistance in healthy humans has not been elucidated yet. Nevertheless, early epidemiological studies have revealed that the administration of diets containing fish oil increases the risk of infection and the incidence of tuberculosis in native Eskimos (Kaplan et al., 1972), and a recent study has determined that higher oleic acid intake may reduce the risk of community-acquired pneumonia in women (Alperovich et al., 2007).
112.6.1 Olive Oil and Host Immune Resistance Results from our laboratory have established that experimental infection with an intracellular pathogen such as L. monocytogenes after administration of diets containing olive oil does not reduce the percentage of survival to an extent similar to a fish oil diet. However, this percentage was lesser than the control (animals fed a low-fat diet) (de Pablo et al., 2000; Puertollano et al., 2003). In addition, the recovery of bacteria from liver (CruzChamorro et al., 2007) and spleen (de Pablo et al., 2000; Puertollano et al., 2005) was reduced at an early stage of infection in mice fed with an olive oil diet in comparison to values from a fish oil diet, suggesting that a reduction of immune system functions originates an immune state unable to efficiently eliminate L. monocytogenes. As a result, survival of mice after challenge with a lethal dose of L. monocytogenes was increased with respect to the group fed a fish oil diet or a low-fat diet (Figures 112.3 and 112.4). Finally, bactericidal activity of peritoneal cells from mice fed a diet containing olive is more efficient than that from mice fed a diet containing fish oil (Puertollano et al., 2001). In spite of the fact that the effects on the immune response appear to be more effective in mice fed with an olive oil diet, the number of adhering bacteria as well as the number of invading bacteria was substantially larger (Puertollano et al., 2002), although the mechanisms of bacteria destruction are more efficient, conferring a broader protection. IL-12 is an important cytokine involved in the activation of Th1 response that plays a relevant role in the elimination of intracellular bacteria. Thus, IL-12 production was not initially reduced in mice fed a diet containing olive
1045
CHAPTER | 112 Olive Oil and Immune Resistance to Infectious Microorganisms
120
120
Low fat
Olive oil
80 60 40 20
80 60 40 20
0
0
24
48
72
0
96 120
Time post-infection (h)
Fish oil
100 Survival (%)
100 Survival (%)
Survival (%)
100
120
80 60 40 20
0
24
48
72
96 120
0
Time post-infection (h)
0
24
48
72
96 120
Time post-infection (h)
FIGURE 112.3 Percentage of mice survival after Listeria monocytogenes infection. Mice were fed their respective diets for 4 weeks (n ⫽ 15 in each dietary group). Adapted from Puertollano et al., 2003 (with permission).
7 * Viable bacteria (log10)
6 * 5 4
*
*
*
3 2 1 0
24
48 72 96 Time post-infection (h)
Low fat
Olive oil
It is possible that the immunosuppression exerted by diets containing olive oil and particularly by fish oil diets is not modified by the administration of other antioxidants not contained in olive oil. On the other hand, the administration of cyclophosphamide, an immunosuppressive agent, to mice fed a diet containing olive oil indicated that this fat is not involved in a reduction in host defense to L. monocytogenes infection. In fact, mice fed an olive oil diet and treated with cyclophosphamide were more resistant than mice fed a diet containing fish oil, which aggravates the immunosuppressive state (Cruz-Chamorro et al., 2007).
SUMMARY POINTS
Fish oil
FIGURE 112.4 Recovery of viable Listeria monocytogenes from spleens of mice (n ⫽ 5 in each dietary group) fed dietary lipids (means ⫾ SEM of two independent determinations in duplicate after logarithmic transformation of these variables). *p ⬍ 0.05 compared with the low-fat group within the same time period.
oil, suggesting that Th1 response was not altered after the administration of this diet (Puertollano et al., 2004). In addition, both TNF-α and IFN-γ production were increased in sera from animals fed a diet containing olive oil (Puertollano et al., 2004, 2007). This event acquires a particular interest because it has been proposed that the production of both IFN-γ and TNF-α mediate the endothelial cell activation in vivo, allowing the recruitment of T cells to the site of infection and contributing to the protective acquired immune response against L. monocytogenes (Xiong et al., 1994). As mentioned previously, the impact of olive oil on the immune system is due to oleic acid rather than other components of this fat. The administration of a diet containing olive oil in mice infected with L. monocytogenes and intraperitoneally treated with N-acetyl-L-cysteine (a precursor of the antioxidant glutathione) had no effects on the survival of mice after infection or on the recovery of viable bacteria from mice spleen (Puertollano et al., 2003).
●
●
●
●
●
●
Different hypotheses have been proposed to explain the intrinsic mechanisms involved in the modulation of immune system functions. The impact of olive oil is mainly due to oleic acid, although the anti-inflammatory properties of olive oil are also attributed to the activity of natural phenolic compounds. The administration of olive oil may modulate immune system functions (lymphocyte proliferation, cytokine production, etc.). This action has been applied in the resolution and prevention of diseases characterized by inflammatory disorders. The modulation of immune functions exhibited by monounsaturated fatty acids could lead to a moderate reduction in host defense to pathogens. The diminution of host infectious disease resistance reported by diets containing olive oil is less significant and more protective than that promoted by other fats (for instance fish oil diets). It would be important to determine the main mechanisms by which olive oil affects host responses to a wider diversity of important human pathogens. This fact may particularly have a major significance when olive oil diets are administered to patients at risk of sepsis.
1046
REFERENCES Alperovich, M., Neuman, M.I., Willett, W.C., Curhan, G.C., 2007. Fatty acid intake and the risk of community-acquired pneumonia in US women. Nutrition 23, 196–202. Anderson, M., Fritsche, K.L., 2002. (n-3) Fatty acids and infectious disease resistance. J. Nutr. 132, 3566–3576. Beauchamp, G.K., Keast, R.S., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A., 2005. Phytochemistry: ibuprofenlike activity in extra-virgin olive oil. Nature 437, 45–46. Berbert, A.A., Kondo, C.R., Almendra, C.L., Matsuo, T., Dichi, I., 2005. Supplementation of fish oil and olive oil in patients with rheumatoid arthritis. Nutrition 21, 131–136. Clamp, A.G., Ladha, S., Clark, D.C., Grimble, R.F., Lund, E.K., 1997. The influence of dietary lipids on the composition and membrane fluidity of rat hepatocyte plasma membrane. Lipids 32, 179–184. Cruz-Chamorro, L., Puertollano, M.A., Puertollano, E., Alvarez de Cienfuegos, G., de Pablo, M.A., 2007. Examination of host immune resistance against Listeria monocytogenes infection in cyclophosphamidetreated mice after dietary lipid administration. Clin. Nutr. 26, 631–639. de Pablo, M.A., Ortega, E., Gallego, A.M., Pancorbo, P.L., Alvarez de Cienfuegos, G., 1998a. Influence of diets containing olive oil, sunflower oil or hydrogenated coconut oil on the immune response of mice. J. Clin. Biochem. Nutr. 25, 11–23. de Pablo, M.A., Ortega, E., Gallego, A.M., Pancorbo, P.L., Alvarez de Cienfuegos, G., 1998b. The effect of dietary fatty acid manipulation on phagocytic activity and cytokine production by peritoneal cells from Balb/c mice. J. Nutr. Sci. Vitaminol. 44, 57–67. de Pablo, M.A., Puertollano, M.A., Gálvez, A., Ortega, E., Gaforio, J.J., Alvarez de Cienfuegos, G., 2000. Determination of natural resistance of mice fed dietary lipids to experimental infection induced by Listeria monocytogenes. FEMS Immunol. Med. Microbiol. 27, 127–133. de Pablo, M.A., Puertollano, M.A., Alvarez de Cienfuegos, G., 2002. Biological and clinical significance of lipids as modulators of immune system functions. Clin. Diagn. Lab. Immunol. 9, 945–950. Deveraux, G., 2002. The immune system: an overview. In: Calder, P. C., Field, C.J., Gill, H.S. (eds) Nutrition and immune function. Cabi Publishing, Oxon, pp. 1–20. Field, C.J., Johnson, I.R., Schley, P.D., 2002. Nutrients and their role in host resistance to infection. J. Leukoc. Biol. 71, 16–32. Granato, D., Blum, S., Rössle, C., Le Boucher, J., Malnoe, A., Dutot, G., 2000. Effects of parenteral lipid emulsions with different fatty acid composition on immune cell functions in vitro. J. Parent. Ent. Nutr. 24, 113–118. James, M.J., Gibson, R.A., Neumann, M.A., Cleland, L.G., 1993. Effect of dietary supplementation with n-9 eicosatrienoic acid on leukotriene B4 synthesis in rats: a novel approach to inhibition of eicosanoid synthesis. J. Exp. Med. 178, 2261–2265. Jeffery, N.M., Yaqoob, P., Newsholme, E.A., Calder, P.C., 1997a. The effects of olive oil upon rat serum lipid levels and lymphocyte functions are due to oleic acid. Ann. Nutr. Metab. 40, 71–80. Jeffery, N.M., Cortina, M., Newsholme, E.A., Calder, P.C., 1997b. Effects of variations in the proportions of saturated, monounsaturated and polyunsaturated fatty acids in the rat diet on spleen lymphocyte functions. Br. J. Nutr. 77, 805–823. Kaplan, G.J., Fraser, R.I., Comstock, G.W., 1972. Tuberculosis in Alaska, 1970. The continued decline of the tuberculosis epidemic. Am. Rev. Respir. Dis. 105, 920–926. Klasing, K.C., Leshchinsky, T.V., 2000. Interactions between nutrition and immunity. In: Gershwin, M.E., German, J.B., Keen, C.L. (eds)
SECTION | II Immunology and Inflammation
Nutrition and Immunity: Principles and Practice. Humana Press Inc., New Jersey, pp. 363–373. Kremer, J.M., Lawrence, D.A., Jubiz, W., DiGiacomo, R., Rynes, R., Bartholomew, L.E., Sherman, M., 1990. Dietary fish oil and olive oil supplementation in patients with rheumatoid arthritis. Clinical and immunologic effects. Arthritis Rheum. 33, 810–820. Kromann, N., Green, A., 1980. Epidemiological studies in the Upernavik district, Greenland. Incidence of some chronic diseases 1950–74. Acta Med. 208, 401–406. Leite, M.S., Pacheco, P., Gomes, R.N., Guedes, A.T., Castro-Faria-Neto, H.C., Bozza, P.T., Koatz, V.L., 2005. Mechanisms of increased survival after lipopolysaccharide-induced endotoxic shock in mice consuming olive oil-enriched diet. Shock 23, 173–178. Linos, A., Kaklamanis, E., Kontomerkos, A., Koumantaki, Y., Gazi, S., Vaiopoulos, G., Tsokos, G.C., Kaklamanis, P., 1991. The effect of olive oil and fish consumption on rheumatoid arthritis-a case control study. Scand. J. Rheumatol. 20, 419–426. Linos, A., Kaklamani, V.G., Kaklamani, E., Koumantaki, Y., Giziaki, E., Papazoglou, S., Mantzoros, C.S., 1999. Dietary factors in relation to rheumatoid arthritis: a role for olive oil and cooked vegetables? Am. J. Clin. Nutr. 70, 1077–1082. Mills, S.C., Windsor, A.C., Knight, S.C., 2005. The potential interactions between polyunsaturated fatty acids and colonic inflammatory processes. Clin. Exp. Immunol. 142, 216–228. Moussa, M., Le Boucher, J., García, J., Thaczok, J., Ragab, J., Dutot, G., Ohayon, E., Ghisolfi, J., Thouvenot, J.P., 2000. In vivo effects of olive oil-based lipid emulsion on lymphocyte activation in rats. Clin. Nutr. 19, 49–54. Puertollano, M.A., de Pablo, M.A., Alvarez de Cienfuegos, G., 2001. Immunomodulatory effects of dietary lipids alter host natural resistance of mice to Listeria monocytogenes infection. FEMS Immunol. Med. Microbiol. 32, 47–52. Puertollano, M.A., de Pablo, M.A., Alvarez de Cienfuegos, G., 2002. Relevance of dietary lipids as modulators of immune functions in cells infected with Listeria monocytogenes. Clin. Diagn. Lab. Immunol. 9, 352–357. Puertollano, M.A., de Pablo, M.A., Alvarez de Cienfuegos, G., 2003. Anti-oxidant properties of N-acetyl-L-cysteine do not improve the immune resistance of mice fed dietary lipids to Listeria monocytogenes infection. Clin. Nutr. 22, 313–319. Puertollano, M.A., Puertollano, E., Ruiz-Bravo, A., Jimenez-Valera, M., de Pablo, M.A., Alvarez de Cienfuegos, G., 2004. Changes in the immune functions and susceptibility to Listeria monocytogenes infection in mice fed dietary lipids. Immunol. Cell Biol. 82, 370–376. Puertollano, M.A., Cruz-Chamorro, L., Puertollano, E., Pérez-Toscano, M.T., Alvarez de Cienfuegos, G., de Pablo, M.A., 2005. Assessment of interleukin-12, gamma interferon, and tumor necrosis factor alpha secretion in sera from mice fed with dietary lipids during different stages of Listeria monocytogenes infection. Clin. Diagn. Lab. Immunol. 12, 1098–1103. Puertollano, M.A., Puertollano, E., Alvarez de Cienfuegos, G., de Pablo, M.A., 2007. Significance of olive oil in the host immune resistance to infection. Br. J. Nutr. 98, S54–S58. Reimund, J.M., Scheer, O., Muller, C.D., Pinna, G., Duclos, B., Baumann, R., 2004. In vitro modulation of inflammatory cytokine production by three lipid emulsions with different fatty acid compositions. Clin. Nutr. 23, 1324–1332. Sanderson, P., Yaqoob, P., Calder, P.C., 1995. Effects of dietary lipid manipulation upon graft vs host and host vs graft responses in the rat. Cell. Immunol. 164, 240–247.
CHAPTER | 112 Olive Oil and Immune Resistance to Infectious Microorganisms
Scrimshaw, N.S., SanGiovanni, J.P., 1997. Synergism of nutrition, infection, and immunity: an overview. Am. J. Clin. Nutr. 66, 464S–477S. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food: epidemiology and nutritional approaches. Nutr. Rev. 60, 170–176. Xiong, H., Kawamura, I., Nishibori, T., Mitsuyama, M., 1994. Cytokine gene expression in mice at an early stage of infection with various strains of Listeria spp. differing in virulence. Infect. Immun. 62, 3649–3654. Yaqoob, P., Newsholme, E.A., Calder, P.C., 1994a. The effect of dietary lipid manipulation on rat lymphocyte subsets and proliferation. Immunology 82, 603–610.
1047
Yaqoob, P., Newsholme, E.A., Calder, P.C., 1994b. Inhibition of natural killer cell activity by dietary lipids. Immunol. Lett. 41, 241–247. Yaqoob, P., Calder, P.C., 1995. The effects of dietary lipid manipulation on the production of murine T cell-derived cytokines. Cytokine 7, 548–553. Yaqoob, P., Knapper, J.A., Webb, D.A., Williams, C.M., Newsholme, E.A., Calder, P.C., 1998. Effect of olive oil on immune function in middle-aged men. Am. J. Clin. Nutr. 67, 129–135. Yaqoob, P., 2002. Monounsaturated fatty acids and immune function. Eur. J. Clin. Nutr. 56, S9–S13.
This page intentionally left blank
Chapter 113
Intestinal Anti-inflammatory Activity of Dietary Olive Oil Julio Gálvez, Desiree Camuesco, Maria Elena Rodríguez-Cabezas and Antonio Zarzuelo Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Department of Pharmacology, School of Pharmacy, University of Granada, Spain
113.1 INTRODUCTION Inflammatory bowel disease (IBD) is a chronic disease of the digestive tract, and usually refers to two related conditions, namely ulcerative colitis and Crohn’s disease, characterized by chronic and spontaneously relapsing inflammation (Table 113.1). Although the etiology of IBD remains unknown, there is increasing experimental evidence to support that it is probably related to an abnormal exacerbated immune response to otherwise innocuous stimuli, which is not properly counteracted by the feedback system. As in other inflammatory conditions, there is an up-regulation of the synthesis and release of different proinflammatory mediators, including reactive oxygen and
nitrogen metabolites, eicosanoids, platelet activating factor (PAF) and cytokines, thus influencing mucosal integrity and leading to excessive tissue injury (Figure 113.1). Moreover, most of these mediators can induce the biosynthesis and release of some others, generating a ‘vicious cycle’ that may result in the propagation and perpetuation of the inflammatory response. Nowadays, a specific causal treatment of IBD is not available yet, since the predisposing and trigger factors of this exacerbated immune response have not been clearly identified. Therefore, therapeutic and preventive strategies for these disorders must rely on interrupting or inhibiting the immunopathogenic mechanisms that are involved, preferably with a single drug treatment. In fact, the drugs currently used for the
TABLE 113.1 Key features of inflammatory bowel diseases. 1. The human intestinal tract mucosa is continuously exposed to an enormous number of luminal bacteria. The innate immune system (mainly constituted by resident cells in the intestine like epithelial cells, dendritic cells, macrophages…) contributes to the protection of the host from invasion by these bacteria and provides a rapid response to pathogens, while it recognizes the commensal bacteria as non-pathogenic 2. In genetically predisposed people, a dysregulated immunologic response to commensal bacteria from the intestinal lumen occurs, and they may develop inflammatory bowel diseases (mainly ulcerative colitis and Crohn’s disease), characterized by chronic and spontaneously relapsing inflammation. The primary symptoms are abdominal cramping, persistent diarrhea associated to weight loss and malnutrition 3. Inflammatory bowel diseases are characterized by the up-regulation of the synthesis and release of a variety of pro-inflammatory mediators, such as eicosanoids, platelet activating factor, reactive oxygen and nitrogen metabolites and cytokines, thus influencing mucosal integrity and leading to excessive tissue injury 4. The objectives in the therapy of inflammatory bowel disease are to induce the remission of the flare-ups and maintain this state. Among the drug treatments used, 5-aminosalicylic acid derivatives, systemic or local glucocorticoids, and immunosuppressants like azathioprine/6-mercaptopurine or cyclosporine try to interfere with multiple stages of the inflammatory cascade, while others like the chimeric anti-TNFα antibody infliximab target specific key mediators of the disease 5. The dietary manipulation can help in the management of these diseases. It improves the malnutrition condition as well as it has additional benefits, since some dietary components display anti-inflammatory and antioxidant properties, like olive oil
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1049
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1050
SECTION | II Immunology and Inflammation
Genetic factors
Environmental factors
Soluble antigen
Bacteria
M cell
Intraepithelial lymphocyte
Dendritic cell
Lymphocyte NO
Epithelial cell
Neutrophil Macrophage IL-1β
TNFα ROS
IL-6
Proteases PAF
LTB4 IL-8
FIGURE 113.1 Immunopathogenesis of inflammatory bowel disease. The bacterial antigens gain access to the intestinal epithelium whose permeability is compromised. The antigen presenting cells, like dendritic cells, macrophage, epithelial cells, present these antigens to the T lymphocytes, thus promoting the synthesis and release of pro-inflammatory cytokines (IL-1β, TNFα, IL-6, IL-8…). The recruitment of circulating cells, including neutrophils, occurs, and many more inflammatory mediators are released, such as eicosanoids (LTB4), platelet activating factor (PAF), reactive oxygen species (ROS) and nitric oxide (NO).
management of human IBD, including 5-aminosalicylic acid (5-ASA) derivatives, systemic or local glucocorticoids and other immunosuppressants exert their beneficial effects through a combination of different mechanisms (Travis and Jewel, 1994). Unfortunately, these drugs are not devoid of potentially serious side effects, thus limiting their use. Consequently, dietary management of IBD may result in an interesting alternative to drug therapy if it provides effectiveness without adverse effects or, at least it may collaborate in achieving a higher efficacy when combined with a given treatment, thus allowing the reduction of the doses of the drug with the concomitant lower risk or severity of adverse effects. This is of special relevance when considering that malnutrition is a common feature in IBD patients, and it becomes worse with disease progression. The malnutrition is the result of different pathogenic factors that converges in these pathologies, including anorexia, malabsorption, altered metabolism, and fluid and electrolyte loss, as well as side effects of the medications used in the treatment. In fact, an adequate enteral nutrition has been shown to maintain the disease in remission (Akobeng and Thomas, 2007). It has been hypothesized that the clinical improvement is clearly mediated by restitution of the patient’s nutritional status, which helps to promote the reparative process in the damaged intestine, or by supplying
compounds that modulate the immune system toward a lower level of inflammation.
113.2 INTESTINAL ANTI-INFLAMMATORY ACTIVITY OF DIETARY OLIVE OIL Different epidemiological studies have revealed that both ulcerative colitis and Crohn’s disease appear to be more frequent in the northern parts of the USA than in the south. Similarly, previous studies in Northern and Western Europe in the 1970s and 1980s suggested that the incidence is decreasing from north to south (Lakatos, 2006), but in the early 1990s, the European IBD Study Group found comparable rates between Southern and Northern Europe (Shivananda et al., 1996). Although it has not been established yet, several factors may be involved including climate, diet, economic wealth, and development or genetic susceptibility. In consequence, a factor that may contribute to the reported increases in the incidence of IBD, at least in the Mediterranean regions, is the change of nutritional habits termed as the Mediterranean diet, in which olive oil plays a key role. Olive oil has been described to have health benefits that include modification of the immune and inflammatory responses (Stark and Madar, 2002), and it has been suggested that these effects may be derived from the antioxidant properties attributed to its components, including the monounsaturated fatty acid oleic acid, as well as tocopherols and phenolic acids, which exhibit antioxidant properties (Owen et al., 2000) (Table 113.2). It is important to note that a situation of intense oxidative insult is a common feature in human IBD, and free radical generation has been proposed to play an important role early on in the pathogenesis of IBD, since it contributes to the initial neutrophil infiltration in the inflamed colonic mucosa. The recruitment and activation of these cells results in a dramatic increase in free radical production that overwhelms the tissue antioxidant protective mechanisms, resulting in a situation of oxidative stress, which definitively perpetuates colonic inflammation. As a consequence, a rapid inhibition of free radical generation could contribute to a lower level of leukocyte infiltration into the inflamed tissue with a positive impact on intestinal inflammation. In fact, previous studies have shown that the antioxidant and/or radical scavenging properties of drugs currently used in the treatment of human IBD, like salicylates, play a key role in their beneficial effects (Travis and Jewel, 1994). Considering all the above, it is plausible that dietary manipulation with olive oil may exert beneficial effects in these intestinal conditions. However, little is known about the role of olive oil in intestinal inflammation, although it has been reported that the ability of its main constituent, oleic acid, to protect human intestinal smooth muscle cells isolated from patients with
CHAPTER | 113 Intestinal Anti-inflammatory Activity of Dietary Olive Oil
TABLE 113.2 Biological activities of the different olive oil components. Olive oil component
Biological activity
Oleic acid
Antioxidant activity PPAR activation Apoptosis and necrosis of lymphocytes
Tyrosol
Antioxidant activity Eicosanoid synthesis downregulation
Hydroxytyrosol
Antioxidant activity Eicosanoid synthesis downregulation Inhibition of NF-κB activation
Oleuropein
Antioxidant activity Eicosanoid synthesis downregulation Inhibition of NF-κB activation Reduction of IL-1β concentration
Crohn’s disease from the oxidative stress that characterizes this intestinal condition (Alzoghaibi et al., 2003), thus supporting its potential use in the treatment of human IBD. Most of the data related with the beneficial effects of olive oil in IBD come from experimental models of rat colitis, including the trinitrobenzenesulfonic acid (TNBS) and dextran sodium sulfate (DSS) models of rat colitis, which shows morphological and biochemical alterations similar to those exhibited in human IBD. Thus, Nieto et al. (2002) reported that rats fed diet enriched in olive oil after TNBS colitis induction showed a reduction in colonic myeloperoxidase (MPO) activity, which is considered as a biochemical marker of neutrophil infiltration and it has been widely used to detect and follow intestinal inflammation, since a reduction in the activity of this enzyme can be interpreted as a manifestation of the anti-inflammatory activity of a given compound. These beneficial effects exerted by an olive oil supplemented diet on intestinal inflammation were also reported in the DSS model of rat colitis (Camuesco et al., 2005), by using a preventative protocol, rather than the curative one used in the TNBS model commented above. Thus, rats fed with an enriched olive oil diet showed an overall lower impact of DSS-induced colonic damage compared to the untreated control group, as evidenced by a decreased incidence of diarrhea and presence of blood in feces, as well as from a reduced weight loss in treated colitic rats, all typical features of this experimental model of rat colitis as well as in human IBD. The macroscopic evaluation of the colonic segments from colitic rats receiving the olive oil-based diet showed a significant reduction in weight/length ratio, thus
1051
revealing attenuation in the edema present in the colonic tissue as a consequence of the inflammatory process. The intestinal anti-inflammatory effects achieved after olive oil incorporation to diet of colitic rats were confirmed histologically by examination of the colonic specimens since they showed an almost complete restoration of the damaged colon displaying a moderate inflammatory infiltrate with a focal distribution, in comparison with untreated rats which revealed the presence of an intense mixed cell infiltration, mainly composed of mononuclear cells (macrophages, lymphocytes and plasma cells) and granulocytes. The lower granulocyte infiltration was confirmed biochemically by a reduction in colonic MPO activity, similarly to that reported in the TNBS model of rat colitis (Nieto et al., 2002). However, the reduction in the granulocyte infiltration was not due to an inhibition in the production of the chemotactic eicosanoid leukotriene B4 (LTB4) (Camuesco et al., 2005). However, this effect could be ascribed to an improvement in the colonic oxidative status, derived from the antioxidant properties attributed to the different components of olive oil (Owen et al., 2000; Stark and Madar, 2002), with the subsequent attenuation of the deleterious effects evoked after free radical overproduction in intestinal inflammation. This was evidenced by a complete restoration in the colonic glutathione content (Camuesco et al., 2005), whose depletion is a feature of the colonic inflammatory status induced by DSS in rats (Camuesco et al., 2004). Similarly, decreased antioxidant defenses in intestinal mucosa have been reported to occur in human IBD (Lih-Brody et al., 1996). In addition to its effect on the oxidative status, other mechanisms may be involved in the beneficial effects exerted after incorporation of olive oil into the diet of DSS colitic rats. Thus, in the same study, Camuesco et al. (2005) reported that the intestinal anti-inflammatory effects of olive oil-enriched diet were associated to a reduced expression of inducible nitric oxide synthase (iNOS). During the last decade it has become increasingly clear that nitric oxide (NO) overproduction by iNOS is deleterious to intestinal function, thus contributing significantly to gastrointestinal immunopathology during the chronic inflammatory events that take place in IBD. This probably results from the intense activation of macrophages, which takes place as a consequence of the inflammatory insult (Camuesco et al., 2004), and which infiltration is attenuated in colitic rats fed enriched olive oil diet (Camuesco et al., 2005). In fact, macrophages are considered an important source of proinflammatory mediators, such as NO and tumor necrosis factor α (TNFα), playing a key role in the pathophysiology of IBD. Furthermore, another effect that can participate involved in the intestinal anti-inflammatory activity of olive oil is the ability of its main component, oleic acid, to promote apoptosis and necrosis of lymphocytes, which can be relevant because a defect in T cell apoptosis is involved in the pathogenesis of inflammatory bowel diseases.
1052
Although the potential beneficial effects of dietary olive oil in intestinal inflammation are promising, it is important to note that, generally, individual dietary compounds have only led to improvement in markers of inflammation, without a clear improvement in the clinical course of the disease. In consequence, an interesting approach in the treatment of IBD may be the dietary manipulation through the combination of different compounds with immunomodulatory properties, which has been shown to display beneficial effects in these intestinal conditions. Supporting this, in a study conducted by Seidner et al. (2005), which combined several dietary compounds, a total of 121 patients with moderately active ulcerative colitis were randomly assigned to receive an oral supplement enriched with fish oil, soluble fiber, and antioxidants or a placebo formula in addition to their normal diet for 6 months; they also received conventional medications (corticosteroids or salicylates, or both) to manage their disease. They observed an improvement in clinical response combined with a decreased requirement for corticosteroids suggesting that this enriched oral supplement can be a useful adjuvant therapy in patients with UC. The role of fish oil in the management of inflammatory bowel diseases is supported by several epidemiological studies performed in Eskimos, which have revealed the low incidence of IBD when compared with Western countries, thus suggesting the protective role of the dietary intake of (n-3) polyunsaturated fatty acids (PUFA) (Belluzzi et al., 2000). In addition, patients with chronic intestinal disorders, including IBD, have been shown to present lower plasma levels of (n-3) PUFA in comparison with normal subjects (Siguel and Lerman, 1996). In fact, the intake of (n-3) PUFA has been suggested to exert a beneficial effect in these intestinal conditions by competing with the (n-6) PUFA linoleic acid for the production of lipid inflammatory mediators, such as prostaglandin E2 (PGE2) or LTB4. Thus, PGE2 alters the absorptive and secretory functions of the intestine as well as cellular immunology, and leads to cell damage (Gil, 2002); whereas LTB4 induces neutrophil adherence to the vascular wall and facilitates the effects of other mediators, like platelet activating factor (PAF), thus promoting neutrophil accumulation within the mucosa with the subsequent tissue damage as a result of cellular activity of this type of leukocytes. In fact (n-3) PUFAs, including eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are metabolized to 3-series prostaglandins (PGE3) and 5-series leukotrienes (LTB5), which generally lead to less inflammation and, in some aspects, with anti-inflammatory effects when compared with their (n-6) PUFA counterparts (Simopoulos, 2002). Of note, the ability to inhibit the synthesis and/or release on LTB4 has been reported to be an important mechanism in the intestinal anti-inflammatory activity of different drugs used in the treatment of IBD like sulfasalazine and 5-ASA (Travis and Jewel, 1994). This mechanism has been also proposed to contribute to the beneficial effects shown
SECTION | II Immunology and Inflammation
by (n-3) PUFA administration in different experimental models of rat intestinal inflammation, induced either by TNBS (Nieto et al., 2002) or DSS (Shimizu et al., 2001). Based on this anti-inflammatory activity, different trials have been performed to evaluate the potential beneficial effects of (n-3) PUFA dietary intake on inflammatory bowel diseases. Although the clinical studies dealing with the use of (n-3) PUFA in IBD have provided controversial results, probably related to discrepancies in the patient selection, formulations and dosages used in the different protocols (Belluzzi et al., 2000), most of them indicate a potential effectiveness of these fatty acids in the therapy of UC and CD. More recently, a systematic review evaluating the efficacy and safety of (n-3) PUFA for maintaining remission in Crohn’s disease have been reported (Turner et al., 2007), indicating that the dietary incorporation of these fatty acids to IBD patients is safe and they may be effective for maintenance of remission in human IBD. However, the authors conclude that, at present, there are not sufficient data to recommend their routine because the small number of patients in the included studies warrants further larger randomized controlled trials (Turner et al., 2007). Considering the outcome of the different dietary interventions, fish oil may be considered suitable to be combined with olive oil to improve its intestinal anti-inflammatory properties. This has been demonstrated in the experimental model of DSS rat colitis. In this sense, the intestinal antiinflammatory effects of dietary administration of olive oil, when supplemented with fish oil-derived (n-3) PUFA, was clearly improved in comparison with the incorporation of olive oil alone to the diet of colitic rats (Camuesco et al., 2005). In fact, in addition to the previously mentioned effects of both types of oil (olive and fish oil) on biochemical markers of inflammation, such as the increased levels of glutathione or reduced iNOS expression due to the antioxidant properties of olive oil or the ability to down-regulate the production of chemotactic eicosanoids by fish oil, a clear reduction in the pro-inflammatory mediator TNFα was observed after the association of both types of oil in the diet of colitic rats. TNFα is a cytokine that is considered to play a key role in human IBD; in fact, different drugs capable of interfering with the activity of this mediator, including infliximab or adalimumab, have been successfully developed for IBD therapy (Kaser and Tilg, 2008). Of note, the synthesis and release of this pro-inflammatory cytokine is up-regulated in different cell types residing in the inflamed mucosa, especially macrophages. This is of relevance since immunofluorescence studies performed in intestinal specimens from DSS colitic rats have revealed that macrophages constitute the predominant cell type in the inflamed areas of the intestine (Camuesco et al., 2004), which infiltration and activity is clearly down-regulated after dietary incorporation of olive and fish oils to colitic rats (Camuesco et al., 2005). Furthermore, in vitro studies have consistently demonstrated that the anti-inflammatory properties attributed
1053
CHAPTER | 113 Intestinal Anti-inflammatory Activity of Dietary Olive Oil
to fish oil emulsions rich in (n-3) PUFA can be exerted through their effects on the macrophage component of the inflammatory response, by inhibiting LPS-mediated TNFα expression (Novak et al., 2003). Also, Yamada et al. (2008) have reported the ability of olive oil in inhibiting TNFα production in human basophilic (KU812) cells. This effect is most probably related to the inhibition of the nuclear factor-κB (NF-κB) signal transduction cascade that plays an important role in controlling inflammatory gene activation. Thus, virgin olive oil has been described to inhibit NF-κB activation in human monocyte/macrophages (Brunelleschi et al., 2007). This effect is most probably due to the presence of antioxidant compounds like hydroxytyrosol or oleuropein, given the redox-sensitive nature of this transcription factor (Li and Verma, 2002), which can be activated by various stimuli, including reactive oxygen species, and whose production is elevated in IBD. Since a relationship has been established between NF-κB activation and up-regulation of iNOS expression, the inhibition of this or other signaling cascades may also justify the inhibitory effect on colonic iNOS expression observed in DSS colitic rats fed a diet supplemented with both olive oil and fish oil (Camuesco et al., 2005). Finally, it has been shown that the expression of other transcription factors, like peroxisome proliferatoractivated receptor (PPAR), are not inhibited by the fatty acids (n-3) PUFA or oleic acid (Brunelleschi et al., 2007), but they can even be activated (Xu et al., 1999). This may be relevant since PPAR-γ agonists have been proposed to exert beneficial effects in human IBD (Dubuquoy et al., 2006). One of the most interesting features of this dietary manipulation is that it provides the possibility of incorporating a lower amount of fish oil (5% of the total fat) in the diet than that used in other studies (Nieto et al., 2002), which may have important consequences. Firstly, it is an affordable supplementation by dietary manipulation which allows the effective dietary fatty acid incorporation in colonic tissue (Camuesco et al., 2005). In fact, an adequate fatty acid incorporation into membrane phospholipid pools has been suggested to influence not only the production of eicosanoids, with the commented implications in the inflamed response, but also to modulate lipid-related intracellular signaling events including actions above different second messengers or transduction pathways such as PPAR and modifying gene expression, thus modulating the inflammatory response (Calder, 2003) (Figure 113.2). Secondly, it is enough to change the (n-6)/(n-3) ratio in the diet to a more balanced ratio of about 4/1, as recommended by nutritional authorities, rather than the ratio of 20–15/1 provided by the current Western diet. And, finally, the lower amount of supplementation of fish oil, together with the incorporation of olive oil that confers antioxidant properties, would also prevent the decrease in the colonic antioxidant defense system that the (n-3) PUFA could induce at high concentrations as it has been described in laboratory animals (Nieto et al., 2002).
The latter may be also achieved by the incorporation of dietary antioxidants, including α-tocopherol or flavonoids, which have an additional advantage, since they have shown intestinal anti-inflammatory activity in experimental models of colitis (Gonzalez et al., 2001; Camuesco et al., 2004). Supporting this, studies performed in the experimental model of DSS rat colitis have demonstrated the synergic effect that occurs when the dietary olive oil supplemented with fish oil, rich in EPA and DHA (n-3) PUFA is associated with the flavonoid quercitrin (Camuesco et al., 2006). The results obtained in this study revealed that the administration of this flavonoid increased the beneficial effects of the dietary manipulation inducing a higher inhibition in the colonic levels of interleukin (IL)-1β in colitic rats than that obtained with enriched diet in olive and fish oils. IL-1β has also been described to play a key role in IBD, since high concentrations of this cytokine are found in the mononuclear cells isolated from intestinal mucosa from IBD patients. Previous studies have shown the ability of quercitrin to decrease IL-1β levels in the inflamed Fatty acid availability Fatty acid composition of immune and inflammatory cell membrane phospholipids
Eicosanoid synthesis
Cell signaling
Membrane fluidity
Gene expression
Nuclei
Lipid raft Eicosanoids
FIGURE 113.2 Mechanisms by which fatty acids can affect immune cell function. The possible intestinal anti-inflammatory activity of dietary fatty acids when they are incorporated to the phospholipid membrane pool after dietary supplementation includes modulation of eicosanoids synthesis, cell signaling, membrane fluidity and gene expression.
Olive oil +
− −
Glutathione +
Quercitrin
MPO iNOS
−
LTB4
Macrophage
−
Neutrophil
− −
−
−
IL1β
TNFα −
Fish oil
Lymphocyte
FIGURE 113.3 Putative effects of olive oil, fish oil and flavonoids responsible for the intestinal anti-inflammatory activity. Olive oil, fish oil and flavonoids may exert their anti-inflammatory activity through the modulation of different inflammatory markers acting in an additive and complementary manner.
1054
SECTION | II Immunology and Inflammation
TABLE 113.3 Anti-inflammatory activity of olive oil-enriched diet and supplementation with fish oil and quercitrin in the DSS model of rat colitis.
Dietary supplementation
Inflammatory marker
Myeloperoxidase activity
Glutathione depletion recovery
LTB4 synthesis
Alkaline phosphatase activity
TNFα production
IL-1β production
Olive oil
13
57
26
42
38
(data not available)
Olive ⫹ fish oil
25
52
74
85
67
48
Olive ⫹ fish oil ⫹ quercitrin
49
100
69
79
68
98
Less neutrophil infiltration
Improvement of the oxidative status
Diminished chemotactic activity
Reduction of the inflammation
Biological significance
This table summarizes the effects of olive oil-supplemented diet on different inflammatory markers that justify the intestinal anti-inflammatory activity shown in the DSS experimental model of rat colitis. These beneficial effects are increased when other active dietary components (fish oil and the flavonoid quercitrin) are added to this diet. Data are expressed as percentage of improvement, obtained from the results reported in Camuesco et al., 2004, 2005.
colonic tissue in colitic rats, thus contributing to its intestinal anti-inflammatory effect (Comalada et al., 2005). It is interesting to note that in vitro studies performed in cell cultures have demonstrated that main constituents of fish oil, EPA and DHA, can also inhibit the production of IL-1β by monocytes (Chu et al., 1999). Furthermore, fish oil can decrease the production of this cytokine in laboratory animals both in healthy and in colitic conditions (Whiting et al., 2005). However, relatively high amounts of PUFA are required to display these activities, so the association with other compounds with inhibitory effects on cytokine production, like the flavonoid quercitrin, may be valuable. In consequence, the dietary enrichment of olive oil, fish oil (containing EPA and DHA) and flavonoids, like quercitrin, can be an interesting approach in IBD therapy to be studied in clinical trials in humans. This dietary manipulation would down-regulate the exacerbated immune response that characterizes IBD through the interference with multiple stages of the inflammatory cascade (Figure 113.3; Table 113.3).
SUMMARY POINTS ●
●
●
Olive oil is one of the key features of Mediterranean diet and is associated to health benefits, including modulation of immune and inflammatory responses. Olive oil can be considered suitable for the dietary management of different immune conditions like human inflammatory bowel diseases. Olive oil-enriched diet attenuates inflammatory response in experimental models of rodent colitis. This beneficial effect may be related to the antioxidant properties ascribed to some olive oil components.
●
The efficacy of olive oil treatment can be improved by combining it with other nutrients like (n-3) polyunsaturated fatty acids containing fish oil or flavonoids, also showing intestinal anti-inflammatory properties.
REFERENCES Akobeng, A.K., Thomas, A.G., 2007. Enteral nutrition for maintenance of remission in Crohn’s disease. Cochrane Database Syst. Rev. Jul 18 (3), CD005984. Alzoghaibi, M.A., Walsh, S.W., Willey, A., Fowler, A.A., Graham, M.F., 2003. Linoleic acid, but not oleic acid, upregulates the production of interleukin-8 by human intestinal smooth muscle cells isolated from patients with Crohn’s disease. Clin. Nutr. 22, 529–535. Belluzzi, A., Boschi, S., Brignola, C., Munarini, A., Cariani, G., Miglio, F., 2000. Polyunsaturated fatty acids and inflammatory bowel disease. Am. J. Clin. Nutr. 71, 339S–342S. Calder, P.C., 2003. N-3 polyunsaturated fatty acids and inflammation: from molecular biology to the clinic. Lipids 38, 343–352. Brunelleschi, S., Bardelli, C., Amoruso, A., Gunella, G., Ieri, F., Romani, A., Malorni, W., Franconi, F., 2007. Minor polar compounds extra-virgin olive oil extract (MPC-OOE) inhibits NF-kappa B translocation in human monocyte/macrophages. Pharmacol. Res. 56, 542–549. Camuesco, D., Comalada, M., Rodriguez-Cabezas, M.E., Nieto, A., Lorente, M.D., Concha, A., Zarzuelo, A., Galvez, J., 2004. The intestinal anti-inflammatory effect of quercitrin is associated with an inhibition in iNOS expression. Br. J. Pharmacol. 143, 908–918. Camuesco, D., Galvez, J., Nieto, A., Comalada, M., Rodriguez-Cabezas, M.E., Concha, A., Xaus, J., Zarzuelo, A., 2005. Dietary olive oil supplemented with fish oil, rich in EPA and DHA (n-3) polyunsaturated fatty acids, attenuates colonic inflammation in rats with DSS-induced colitis. J. Nutr. 135, 687–694. Camuesco, D., Comalada, M., Concha, A., Nieto, A., Sierra, S., Xaus, J., Zarzuelo, A., Galvez, J., 2006. Intestinal anti-inflammatory activity of combined quercitrin and dietary olive oil supplemented with fish oil,
CHAPTER | 113 Intestinal Anti-inflammatory Activity of Dietary Olive Oil
rich in EPA and DHA (n-3) polyunsaturated fatty acids, in rats with DSS-induced colitis. Clin. Nutr. 25, 466–476. Chu, A.J., Walton, M.A., Prasad, J.K., Seto, A., 1999. Blockade by polyunsaturated n-3 fatty acids of endotoxin-induced monocytic tissue factor activation is mediated by the depressed receptor expression in THP-1 cells. J. Surg. Res. 87, 217–224. Comalada, M., Camuesco, D., Sierra, S., Ballester, I., Xaus, J., Gálvez, J., Zarzuelo, A., 2005. In vivo quercitrin anti-inflammatory effect involves release of quercetin, which inhibits inflammation through downregulation of the NF-kappaB pathway. Eur. J. Immunol. 35, 584–592. Dubuquoy, L., Rousseaux, C., Thuru, X., Peyrin-Biroulet, L., Romano, O., Chavatte, P., Chamaillard, M., Desreumaux, P., 2006. PPARgamma as a new therapeutic target in inflammatory bowel diseases. Gut 55, 1341–1349. Fiocchi, C., 1998. Inflammatory bowel disease: aetiology and pathogenesis. Gastroenterology 115, 182–205. Gil, A., 2002. Polyunsaturated fatty acids and inflammatory diseases. Biomed. Pharmacother. 56, 388–396. Gonzalez, R., Sanchez de Medina, F., Galvez, J., Rodriguez-Cabezas, M.E., Duarte, J., Zarzuelo, A., 2001. Dietary vitamin E supplementation protects the rat large intestine from experimental inflammation. Int. J. Vitam. Nutr. Res. 71, 243–250. Kaser, A., Tilg, H., 2008. Novel therapeutic targets in the treatment of IBD. Expert Opin. Ther. Targets 12, 553–563. Lakatos, P.L., 2006. Recent trends in the epidemiology of inflammatory bowel diseases: Up or down? World J. Gastroenterol. 12, 6102–6108. Lih-Brody, L., Powell, S.R., Collier, K.P., Reddy, G.M., Cerchia, R., Kahn, E., Weissman, G.S., Katz, S., Floyd, R.A., McKinley, M.J., Fisher, S.E., Mullin, G.E., 1996. Increased oxidative stress and decreased antioxidant defences in mucosa of inflammatory bowel disease. Dig. Dis. Sci. 41, 2078–2086. Nieto, N., Torres, M.I., Rios, A., Gil, A., 2002. Dietary polyunsaturated fatty acids improve histological and biochemical alterations in rats with experimental ulcerative colitis. J. Nutr. 132, 11–19. Novak, T.E., Babcock, T.A., Jho, D.H., Helton, W.S., Espat, N.J., 2003. NF-kappa B inhibition by omega-3 fatty acids modulates LPS-stimulated macrophage TNF-alpha transcription. Am. J. Physiol. Lung Cell Mol. Physiol. 284, L84–L89. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Wurtele, G., Spiegelhalder, B., Bartsch, H., 2000. Olive-oil consumption and health: the possible role of antioxidants. Lancet Oncol. 1, 107–112.
1055
Seidner, D.L., Lashner, B.A., Brzezinski, A., Banks, P.L., Goldblum, J., Fiocchi, C., Katz, J., Lichtenstein, G.R., Anton, P.A., Kam, L.Y., Garleb, K.A., Demichele, S.J., 2005. An oral supplement enriched with fish oil, soluble fiber, and antioxidants for corticosteroid sparing in ulcerative colitis: a randomized, controlled trial. Clin. Gastroenterol. Hepatol. 3, 358–369. Shimizu, T., Igarashi, J., Ohtuka, Y., Oguchi, S., Kaneko, K., Yamashiro, Y., 2001. Effects of n-3 polyunsaturated fatty acids and vitamin E on colonic mucosal leukotriene generation, lipid peroxidation, and microcirculation in rats with experimental colitis. Digestion 63, 49–54. Shivananda, S., Lennard-Jones, J., Logan, R., Fear, N., Price, A., Carpenter, L., van Blankenstein, M., 1996. Incidence of inflammatory bowel disease across Europe: is there a difference between north and south? Results of the European Collaborative Study on Inflammatory Bowel Disease (EC-IBD). Gut 39, 690–697. Siguel, E.N., Lerman, R.H., 1996. Prevalence of essential fatty acid deficiency in patients with chronic gastrointestinal disorders. Metabolism 45, 12–23. Simopoulos, A.P., 2002. Omega-3 fatty acids in inflammation and autoimmune diseases. J. Am. Coll. Nutr. 21, 495–505. Stark, A.H., Madar, Z., 2002. Olive oil as a functional food: epidemiology and nutritional approaches. Nutr. Rev. 60, 170–176. Travis, S.P.L., Jewel, D.P., 1994. Salicylates for ulcerative colitis – their mode of action. Pharmac. Ther. 63, 135–161. Turner, D., Zlotkin, S.H., Shah, P.S., Griffiths, A.M., 2007. Omega 3 fatty acids (fish oil) for maintenance of remission in Crohn’s disease. Cochrane Database Syst. Rev. Apr 18 (2), CD006320. Whiting, C.V., Bland, P.W., Tarlton, J.F., 2005. Dietary n-3 polyunsaturated fatty acids reduce disease and colonic proinflammatory cytokines in a mouse model of colitis. Inflamm. Bowel Dis. 11, 340–349. Yamada, P., Zarrouk, M., Kawasaki, K., Isoda, H., 2008. Inhibitory effect of various Tunisian olive oils on chemical mediator release and cytokine production by basophilic cells. J. Ethnopharmacol. 116, 279–287. Xu, H.E., Lambert, M.H., Montana, V.G., Parks, D.J., Blanchard, S.G., Brownm, P.J., Sternbach, D.D., Lehmann, J.M., Wisely, G.B., Willson, T.M., Kliewer, S.A., Milburn, M.V., 1999. Molecular recognition of fatty acids by peroxisome proliferator-activated receptors. Mol. Cell 3, 397–403.
This page intentionally left blank
Chapter 114
Use of Olive Oil in Patients with Rheumatoid Arthritis Décio Sabbatini Barbosa1, Andréa Colado Simão1 and Isaias Dichi2 1 2
Department of Pathology, Clinical Analysis and Toxicology-University of Londrina Paraná, Brazil Department of Internal Medicine, University of Londrina, Paraná, Brazil
114.1 INTRODUCTION Rheumatoid arthritis (RA) is a systemic inflammatory disorder that mainly affects the diarthrodial joint. It is the most common form of inflammatory arthritis, and has a substantial societal effect in terms of cost, disability, and lost productivity. This disease affects about 1% of the population, in a female/male ratio of 2.5/1, and can occur at any age, but it is most common among those aged 40–70 years, its incidence increasing with age (Lee and Weinblatt, 2001). RA is characterized in the early stages by persistent inflammation of the synovial lining of the joints, which can lead to the destruction of bone and cartilage and result in joint deformity. The range of presentations of rheumatoid arthritis is broad, but disease onset is insidious in most cases, and diagnosis can be delayed for several months before it can be ascertained. The predominant symptoms are pain, morning stiffness, and swelling of peripheral joints. The clinical course of the disorder is extremely variable, ranging from mild, self-limiting arthritis to rapidly progressive multisystem inflammation with profound morbidity and mortality. The systemic disturbances are most commonly fatigue, stiffness, anemia, weight loss, and extra-articular features such as rheumatoid nodules (Lee and Weinblatt, 2001). Two main characteristics of the condition are the presence of rheumatoid factor (RF) measured in blood and typical RA erosions seen on radiological examination of hands and feet. RF is used both in diagnosis and prognosis, but lacks sensitivity because it is found in only 70–80% of RA cases. It is also seen in approximately 5% of normal populations.
114.2 ETIOLOGY The etiology of RA, a chronic inflammatory disease, remains largely unknown, although microbiological, immune, genetic, hormonal, and dietary factors have been implicated in its pathogenesis (Linos et al., 1991). The cause of RA remains Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
uncertain, but genetic factors are known to be involved and the locus has been identified. It has been suggested that genetic susceptibility explains ⱕ40% of the risk of developing RA (Silman et al., 1993). The suggested causes of RA include: viral infections, immunization, hormonal factors (prolactin production during lactation), smoking, previous blood transfusion, obesity, stress and physical trauma (Symmons and Harrison, 2000). RA is characterized by infiltration of T lymphocytes, macrophages and plasma cells into the synovium, and the initiation of a chronic inflammatory state that involves overproduction of pro-inflammatory cytokines and a deregulated T-helper-1-type response. Products of free radical oxidation have been identified in synovial fluid, thus lending further support to the theory that RA itself may be, in part, mediated by free radical activity (Pattison et al., 2004). Increased lipid peroxidation and depletion of ascorbate is a result of oxidation during its antioxidant activity and antioxidant micronutrients may have an important role in preventing tissue damage caused by free radicals (Halliwell et al., 1998).
114.3 PATHOPHYSIOLOGY OF INFLAMMATION Inflammation is the body’s reaction to invasion by an infectious agent, antigen or physical damage. Antigen exposure triggers the immune response, resulting in a cascade of cellular activity and an inflammatory response in the end organ. In RA, the inflammatory response continues in articular tissue, as though in response to a persistent stimulus, leading in time to irreversible damage to tendons and joints. Although a greater understanding of efferent mechanisms of inflammation and tissue destruction in RA has evolved over the past 10 years, there is still little understanding as to why inflammation persists in RA (Pattison et al., 2004).
1057
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1058
SECTION | II Immunology and Inflammation
An inflamed synovium is central to the pathophysiology of RA. It is histologically striking, showing pronounced angiogenesis, cellular hyperplasia, influx of inflammatory leukocytes, and changes in the expression of cell-surface adhesion molecules, proteinases, proteinase inhibitors and many pro-inflammatory cytokines (Lee and Weinblatt, 2001). Following initiation of the immune response, activated macrophages, monocytes and granulocytes generate free oxygen radicals (Halliwell et al., 1998). The rapidly proliferating cells of the immune system are uniquely prone to oxidative damage by free radicals and pro-inflammatory prostaglandin E2 (PGE2), leukotriene B4 (LTB4), and cytokines, in particular TNF-α and IL-1. The increased production of cytokines and the subsequent increase in reactive oxygen and nitrogen species are recognized hallmarks of inflammation. This process is regulated by a negative feedback mechanism and is closely followed by the secretion of anti-inflammatory cytokines to reduce the accumulation of reactive species. The cellular antioxidant defense system is also activated to limit the development of chronic inflammation (Bulló et al., 2007). The binding of pro-inflammatory cytokines to their receptors triggers the mitogen-activated protein kinase (MAPK) pathway that ultimately results in the activation of two redox-sensitive transcription factors: nuclear factor kappa B (NFκB) and the c-Jun part of activating protein-1 (AP-1). These transcription factors activate the expression of a wide variety of genes including cytokines, chemokines, adhesion molecules and inducible effector enzymes such as iNOS and cyclooxygenase-2 (COX-2) (Bulló et al., 2007). The main features of rheumatoid arthritis are shown in Table 114.1.
less unsaturated than EPA, it may have greater chemical stability, which would be an advantage for use as a dietary constituent or supplement (James et al., 1993). Virgin olive oil is a rich source of MUFA and retains all the lipophilic components of the olive fruit, especially the phenolic compounds with strong anti-oxidant and anti-inflammatory properties. The administration of OO with a high phenolic content has been shown to protect against inflammation. In addition, phenolic compounds derived from extra virgin oil were recently shown to decrease the production of inflammatory mediators in human whole-blood cultures and to inhibit endothelial adhesion molecule expression in vitro (Bulló et al., 2007). It has been attributed to extra virgin olive oil (EVOO) preventive properties with regard to chronic diseases, particularly those with an inflammatory etiology such as heart disease, cancer and RA (Wahle et al., 2004; Pacheco et al., 2007). The beneficial effects can be explained not only to the high monounsaturated content of OO, but also to the antioxidant property of its minor elements with high activity. The phenolic compounds are both lipophilic and hydrophilic. The lipophilics include tocopherols, while the hydrophilics include flavonoids, phenolic alcohols and acids, secoiridoids (oleuropein and ligstroside) and lignans (Tripoli et al., 2005). Oleuropein is the main polyphenol found in OO (Tripoli et al., 2005).
TABLE 114.1 Key features of rheumatoid arthritis. ●
●
114.4 OLIVE OIL AND INFLAMMATION Since the 1970s, a number of epidemiological studies have suggested that certain dietary fatty acids affect the immune response in both animals and humans, and more recently that they have anti-inflammatory effects. Olive oil (OO) exerts its beneficial effect modulating immune function, particularly the inflammatory process associated with the immune system, as seems to be the case in RA (Wale et al., 2004). Olive oil is mainly composed of oleic acid (18:1 n-9), a monounsaturated fatty acid (MUFA) that is converted to 8,9,11 eicosatrienoic acid (20:3 n-9; ETA) under restriction of n-6 fatty acids. ETA is converted to LTA3, which is a potent inhibitor of LTB4 synthesis (Figure 114.1). Therefore, oleic acid and its metabolite ETA may exert inflammatory effects through a mechanism similar to that of fish oil, which contains EPA. Because ETA is substantially Restriction of OLEIC ACID
n-6 fatty acids
●
●
●
●
●
Rheumatoid arthritis (RA) is a systemic inflammatory disorder that mainly affects the joints It is the most common form of inflammatory arthritis and can lead to destruction of bone and cartilage resulting in profound morbidity and mortality The predominant symptoms are pain, morning stiffness, and swelling of peripheral joints, but systemic disturbances can also occur Two main characteristics of the condition are the presence of RF and typical RA erosions seen on radiological examination of hands and feet The etiology of RA remains largely unknown, although immune, and genetic factors are likely the most implicated in its pathogenesis RA is characterized by infiltration of T lymphocytes, macrophages and plasma cells into the synovium, and the initiation of a chronic inflammatory state that involves overproduction of pro-inflammatory cytokines, reactive oxygen and nitrogen species, and a deregulated T-helper1-type response The binding of pro-inflammatory cytokines to their receptors triggers the mitogen-activated protein kinase pathway that ultimately results in the activation of nuclear factor kappa B
5-lipoxygenase ETA
LTA synthase
inhibit LTA3
LTA hydrolase
FIGURE 114.1 Anti-inflammatory action of oleic acid. ETA – 8,9,11 eicosatrienoic acid; LT – leukotriene.
LTB4
1059
CHAPTER | 114 Use of Olive Oil in Patients with Rheumatoid Arthritis
Lipid radicals are produced during reactions involved in the metabolism of arachidonic acid, during the synthesis of the eicosanoids by the action of the lipoxygenase (LOX) and COX. The reactive oxygen species (ROS) and reactive nitrogen species involved in RA include superoxide anions, hydrogen peroxide, hydroxyl and peroxynitrite radicals, and nitric oxide. The first three of these are produced by xanthine oxidase and are also generated by activated macrophages and neutrophils as a result of respiratory chain activity known as the oxidative burst. This is primarily due to NADPH oxidase activity leading to the formation of hypochlorous acid (HOCl) as a bactericidal agent. Hydrogen peroxide is formed partly by superoxide dismutase (SOD), by the reaction between superoxide radicals and protons. Hydrogen peroxide is metabolized by catalase and peroxidase enzymes, chiefly glutathione peroxidase (Darlington and Stone, 2001). Indeed, the evidence that oxidative damage occurs in RA is very strong. Thus, antioxidant micronutrients may have an important role in preventing tissue damage caused by ROS (Pattison et al., 2004). The biological activity of phenolic compounds of OO is not limited to their antioxidant ability, but extends to their interaction with important enzymic systems. In particular, it has been found that olive oil phenols inhibit platelet aggregation, reduce pro-inflammatory molecule formation such as thromboxane B2 (TBX2) and LTB4, and inhibit the use of oxygen in human neutrophils (Tripoli et al., 2005) (Figure 114.2). Olive oil is a non-oxidative dietary component, and the attenuation of the inflammatory process it elicits could explain the beneficial effects on disease risk since oxidative and inflammatory stresses appear to be underlying factors in the etiology of inflammatory diseases in humans. The antioxidant effects of olive oil are probably due to a combination of its high oleic acid content (low oxidation potential compared with linoleic acid) and its content of a variety of plant antioxidants, particularly oleuropein, hydroxytyrosol, and tyrosol (Walde et al., 2004).
114.5 ARE THERE DIFFERENCES BETWEEN OLIVE OIL AND EXTRA VIRGIN OLIVE OIL USE IN THE INFLAMMATORY PROCESS? The concentration of phenolic compounds in OO is the result of a complex interaction of various factors and is also affected by the extraction process (Visioli and Galli, 1998; Tripoli et al., 2005). It is necessary to point out that refined oils do not have a significant content of polyphenols. Extra virgin olive oil (EVOO) is obtained from the first physical cold pressure of the olive paste and is rich in phenolic compounds. Virgin olive oil, obtained through percolation (first extraction), has a higher content in phenols, 0-diphenols, hydroxytyrosol and tyrosol aglycones, and tochopherols than oils obtained through centrifugation (second extraction) (Visioli and Galli, 1998). The effect of a virgin olive oil-enriched diet in acute and chronic inflammation models in rats was analyzed and determined the effect of supplementing this oil with a higher content of its polyphenolic fraction (MartinezDomínguez et al., 2001). This study demonstrated that virgin olive oil with a high content of polyphenolic compounds, similar to those of extra virgin olive oil, shows protective effects in both models of inflammation (Martinez-Domínguez et al., 2001). Another study compared the effects of two diets enriched in olive oils, having the same fatty acid composition but with EVOO and without (refined olive oil, ROO) minor compounds, on postprandial levels of triacylglycerol and on the accumulation of soluble intercellular adhesion molecule (sICAM-1) and soluble vascular cell adhesion molecule (sVCAM-1) in healthy and dyslipidemic humans (Pacheco et al., 2007). The results of this study indicated that the consumption of EVOO may help in reducing postprandial levels of adhesion molecules, which suggests a protective postprandial antiinflammatory effect in healthy and hypertriacylglycerolemic subjects (Pacheco et al., 2007). It was also investigated
Bone Venule Synovial fibroblasts TNF-Alfa IL-6: IL-8: PGE2
ROS*
, OH, HOCL, ONOO−) (O− 2 *TXB /LTB* 2 4
T cells and macrophages
IL-1Beta
Neutrophils Synovial membrane Cartilage
FIGURE 114.2 Synovial inflammatory process in the knee joint. Asterisks indicate the most plausible mechanisms of olive oil action. ROS – reactive oxygen species; TXB2 – thromboxane B2; LTB4 – leukotriene B4; IL – interleukin; PGE2 – prostaglandin E2.
1060
whether a bolus ingestion of phenol-rich extra virgin olive oil affects the postprandial lipid profile, as well as selected surrogate markers of cardiovascular risk, of healthy volunteers. Moreover, the effects of EVOO with those of OO were compared (Bogani et al., 2007). A significant decrease in inflammatory markers, namely TXB2 and LTB4, 2 and 6 h after EVOO (but not OO) consumption and a concomitant increase of serum antioxidant capacity were recorded. The authors of this study concluded that these data reinforce the notion that the Mediterranean diet reduces the incidence of coronary heart disease (an inflammatory disease) partially due to the protective role of its phenolic components, including those of EVOO (Bogani et al., 2007). Although OO contains a relatively low concentration of α-tocopherol, it is known to be highly resistant to oxidative degradation. This is due, in part, to the relatively low content of polyunsaturated fatty acids and also to the high concentration of polyphenolic antioxidants, particularly in EVOO. The antioxidant activity of OO phenolic compounds, and in particular of oleuropein and its byproduct hydroxytyrosol, has been studied in many experimental models (Wahle et al., 2004; Tripoli et al., 2005). The antioxidant activity of oleuropein and hydroxytyrosol has also been demonstrated in cellular models and animals (Tripoli et al., 2005). Some studies with human volunteers do not show the same attenuating effects of MUFA/OO on immune function and could reflect the high content of these components in the animals’ diets (Wahle et al., 2004). However, according to the discussion above, it is reasonable to imagine a possible mechanism of the use of extra virgin olive oil (with the characteristic high concentration of MUFA and the presence of minor components) in inflammatory diseases: it has inhibitory action on COX and LOX; reduces pro-inflammatory molecule formation such as TXB2 and LTB4, reducing the adhesion molecules and free radicals formation (Tripoli et al., 2005). Maybe future large trials and other studies can confirm this assumption.
114.6 CAN OLIVE OIL PREDICT THE RISK OF RHEUMATOID ARTHRITIS? Olive oil, as well as fish consumption, was shown to be an important predictor of risk of RA in a case-control study in Greece (Linos et al., 1991). The Greek diet is based mainly on fruit and vegetables, either raw or cooked with olive oil, and contains less meat and more fish and pulses than the Western diet, food items that may influence risk of RA. Some years later, the same group (Linos et al., 1999) performed a case-control study in Greece, where the 145 patients and 188 controls were paired by sex, age, and health care facility. Persons in the lowest category of OO consumption had a 2.5 times higher risk of developing RA than did persons in the highest category of consumption. The excess daily OO consumption was 43 g day⫺1 approximately.
SECTION | II Immunology and Inflammation
Consumption of OO was inversely and independently associated with risk of RA in this population.
114.7 STUDIES USING OLIVE OIL AS PLACEBO IN RHEUMATOID ARTHRITIS PATIENTS Many patients look for complementary and alternative medicine options in coping with this debilitating disease and an estimated 60–90% of persons with RA use them (Rao et al., 1999). Olive oil certainly is one of these options. In the beginning of the 1980s, clinical studies which evaluated RA patients used OO as placebo. Olive oil was regularly used as control supplementation in experiments with fish oil (n-3 polyunsaturated fatty acids). At that time, MUFAs were typically regarded as being neutral fatty acids without interfering in the inflammatory process of RA (Firestein, 2003).
114.7.1 Studies Using n-3 Fatty Acids in Rheumatoid Arthritis Patients and Olive Oil as Placebo Darlington and Ramsey (1987) reported significant decreases in pain intensity, duration of morning stiffness, time taken to walk 18 m, and fibrinogen levels, and improved trends in erythrocyte sedimentation rate, C3, and right grip strength after 12 weeks of OO ingestion. Cleland et al. (1988) investigated clinical and biochemical effects of dietary fish oil supplements in RA and used OO as placebo. They demonstrated that 18 g day⫺1 of OO reduced morning stiffness and pain score. They suggested that these findings could account for the lack of statistical significance between the groups. Kremer et al. (1990) performed a prospective, randomized, double-blind, parallel study with 49 patients with active arthritis during 24 weeks. Three groups were studied: two groups received different doses of fish oil dietary supplements, and one group received olive oil supplements as placebo. Patients who received fish oil demonstrated improvement on several clinical manifestations of RA. However, the improvements from baseline in the patients ingesting fish oil were usually not statistically significant compared with the patients taking olive supplements. Although the large and small doses of fish oil n-3 fatty acids produced better overall results, the OO group was the only one to show improvements in the patients’ global assessment. It should be noted that the clinical outcomes in the OO group may have been biased toward more favorable results because of the withdrawal of 11 of the original 23 patients in this study group. Macrophage IL-1 production decreased by 38.5% in the OO group. The authors concluded that dietary supplementation with OO is associated with certain changes in immune function. Olive oil
1061
CHAPTER | 114 Use of Olive Oil in Patients with Rheumatoid Arthritis
itself could have beneficial effects in RA patients due to cell membrane changes in lymphocytes resulting in altered immune function through a variety of mechanisms. Sköldstam et al. (1992) performed a controlled, randomized, double-blind study to assess the effect of n-3 fatty acids in patients with RA. Forty-three patients were allocated in two groups: 22 patients received fish oil (1 g) whereas 21 received a mixture of oils (corn, olive and peppermint) during 6 months. The group of patients which took fish oil showed a decrease in arthritis activity with significant reduction in non-steroidal anti-inflammatory drugs after 3 and 6 months. The control group patients (mixture oils) presented an increase in arthritis activity after 6 months. The groups did not show any improvement in joint pain, duration of morning stiffness, functional capacity, or in biochemical markers of inflammation. Geusens et al. (1994) found significant decreases in Ritchie’s articular index of pain and the number of painful
joints after 12 months of olive oil (6 g) and also after combined use of fish oil and olive oil. Therefore, although OO has classically been used as a placebo in studies investigating the effects of other oils, mainly fish oil, in patients with RA, without a definite purpose of being part of the therapeutic arsenal of RA, several studies above demonstrated its beneficial effects.
114.7.2 Studies Using n-3 Fatty Acids in Rheumatoid Arthritis Patients and Olive Oil as Adjuvant Therapy Berbert et al. (2005) evaluated whether supplementation with OO during 6 months could improve clinical and laboratory parameters of disease activity in patients with RA who were using fish oil supplements (Tables 114.2 and 114.3). Forty-three patients were assigned to one of three groups.
TABLE 114.2 Clinical indicators of disease activity in patients with rheumatoid arthritis receiving placebo (G1), n-3 fatty acids (G2) or n-3 fatty acids and oleic acid (G3). Baseline
12 weeks
24 weeks
Morning stiffness (minutes)
G1 G2 G3
38 ⫾ 42 44 ⫾ 68 60 ⫾ 65
46 ⫾ 47a 21 ⫾ 49 20 ⫾ 39
51 ⫾ 50a,b 5⫾8 11 ⫾ 26
Joint pain intensity
G1 G2 G3*
1.77 ⫾ 0.93 2.31 ⫾ 0.86 2.18 ⫾ 0.73
1.77 ⫾ 1.17a,b 1.46 ⫾ 0.66 1.00 ⫾ 0.71
1.85 ⫾ 1.21a,b 1.23 ⫾ 0.83 0.53 ⫾ 0.80
Time of onset of fatigue (min)
G1 G2 G3
19.5 ⫾ 3.5 20.9 ⫾ 12.1 23.4 ⫾ 14.1
19.3 ⫾ 3.5 17.1 ⫾ 10.2 21.7 ⫾ 13.0
21.4 ⫾ 5.2a,b 16.3 ⫾ 10.3 19.1 ⫾ 10.6
Ritchie articular index
G1 G2 G3
6.9 ⫾ 5.4 15.8 ⫾ 9.9 15.9 ⫾ 12.6
5.5 ⫾ 7.5 7.6 ⫾ 6.7 5.8 ⫾ 8.2
5.2 ⫾ 4.4a,b 3.6 ⫾ 2.4 1.2 ⫾ 2.3
Grip strength (right hand)
G1 G2 G3
62 ⫾ 37 54 ⫾ 35 63 ⫾ 58
60 ⫾ 29a,b 91 ⫾ 62 101 ⫾ 69
68 ⫾ 31a,b 105 ⫾ 78 114 ⫾ 84
Grip strength (left hand)
G1 G2 G3
74 ⫾ 43 57 ⫾ 35 67 ⫾ 52
68 ⫾ 35a,b 78 ⫾ 62 94 ⫾ 62
70 ⫾ 28a,b 108 ⫾ 75 110 ⫾ 82
Patient global assessment
G1 G2 G3
1.25 ⫾ 0.75 1.54 ⫾ 0.88 1.82 ⫾ 0.53
1.42 ⫾ 0.67a 1.62 ⫾ 0.87c 1.12 ⫾ 0.60
1.31 ⫾ 0.95a 1.23 ⫾ 0.60 0.88 ⫾ 0.70
Classification of functional status
G1 G2 G3*
1.85 ⫾ 0.80 2.54 ⫾ 0.78 2.65 ⫾ 1.22
2.00 ⫾ 1.08 2.23 ⫾ 1.01 2.53 ⫾ 1.28
2.00 ⫾ 0.91 2.15 ⫾ 1.07 2.00 ⫾ 1.28
Percentage change from baseline: G1 versus G3 (p ⬍ 0.05) b G1 versus G2 (p ⬍ 0.05) c G2 versus G3 (p ⬍ 0.05) * -within-group changes (p ⬍ 0.05). a
1062
SECTION | II Immunology and Inflammation
TABLE 114.3 Health Assessment Questionnaire in patients with rheumatoid arthritis receiving placebo (G1), n-3 fatty acids (G2) or n-3 fatty acids and oleic acid (G3). Baseline
12 weeks
24 weeks
Dressing, including tying shoelaces
G1 G2 G3
0.15 ⫾ 0.56 1.46 ⫾ 1.27 1.24 ⫾ 1.30
0.15 ⫾ 0.38 0.69 ⫾ 1.32 0.77 ⫾ 1.15
0.23 ⫾ 0.60 0.77 ⫾ 1.17 0.53 ⫾ 1.18
Get in and out of bed
G1 G2 G3
0.15 ⫾ 0.38 0.54 ⫾ 1.13 0.77 ⫾ 0.90
0.23 ⫾ 0.60 0.39 ⫾ 0.96 0.53 ⫾ 1.01
0.46 ⫾ 0.88 0.39 ⫾ 0.87 0.65 ⫾ 1.17
Lift a cup or glass to your mouth
G1 G2 G3
0.0 ⫾ 0.0 0.46 ⫾ 0.66 0.53 ⫾ 0.87
0.08 ⫾ 0.28 0.31 ⫾ 0.48 0.18 ⫾ 0.73
0.00 ⫾ 0.0 0.15 ⫾ 0.56 0.24 ⫾ 0.07
Walk on flat ground
G1 G2 G3
0.39 ⫾ 0.77 0.62 ⫾ 1.04 1.18 ⫾ 1.13
0.39 ⫾ 0.77 0.54 ⫾ 0.78 0.88 ⫾ 1.17
0.54 ⫾ 1.05 0.54 ⫾ 0.78 0.77 ⫾ 1.20
Wash and dry your entire body
G1 G2 G3
0.08 ⫾ 0.28 0.77 ⫾ 1.30 0.82 ⫾ 1.02
0.23 ⫾ 0.60 0.54 ⫾ 0.97 0.41 ⫾ 0.87
0.31 ⫾ 0.75 0.39 ⫾ 0.77 0.29 ⫾ 0.85
Bend down to pick up clothing from the floor
G1 G2 G3
0.69 ⫾ 1.03 1.85 ⫾ 1.21 1.65 ⫾ 1.32
1.0 ⫾ 1.08b 1.15 ⫾ 1.41 1.35 ⫾ 1.41
1.31 ⫾ 1.38a,b 1.23 ⫾ 1.36 1.06 ⫾ 1.48
Turn faucets on and off
G1 G2 G3
0.92 ⫾ 1.04 1.23 ⫾ 1.10 1.59 ⫾ 1.06
1.08 ⫾ 1.04a 0.85 ⫾ 0.80 1.00 ⫾ 1.17
1.23 ⫾ 1.09a 0.92 ⫾ 0.86 0.77 ⫾ 1.09
Get in and out of a car
G1 G2 G3
0.46 ⫾ 0.66 1.46 ⫾ 1.27 1.53 ⫾ 1.23
0.31 ⫾ 0.63 0.69 ⫾ 1.18 1.18 ⫾ 1.33
0.69 ⫾ 1.03a,b 0.69 ⫾ 1.11 0.94 ⫾ 1.39
Percentage change from baseline: a G1 versus G3 (p ⬍ 0.05) b G1 versus G2 (p ⬍ 0.05).
The first group (G1) received soy oil (placebo), the second group (G2) received fish oil n-3 fatty acids (3 g day⫺1), and the third group (G3) received fish oil n-3 fatty acids (3 g day) and 6.8 g day⫺1 of oleic acid (9.6 mL of OO). There was a statistically significant improvement in G2 and G3 in relation to G1 with respect to joint pain intensity, right and left handgrip strength, duration of morning stiffness, onset of fatigue, Ritchie’s articular index for pain joints, ability to bend down to pick up clothing from the floor, and getting in and out of a car. G3, but not G2, in relation to G1 showed additional improvements with respect to duration of morning stiffness, patient global assessment after, ability to turn faucets on and off, and rheumatoid factor. In addition, G3 showed a significant improvement in patient global assessment in relation to G2. The rheumatoid factor decrease in G3 has a huge clinical significance because patients who have high titers tend to have a more aggressive, destructive course. The authors concluded that ingestion of fish oil n-3 fatty acids relieved several clinical parameters used in the
present study. However, patients showed a more precocious and accentuated improvement when fish oil supplements were used in combination with OO. Table 114.4 resumes the trials in which olive oil was used in patients with RA.
114.8 FUTURE PERSPECTIVES As discussed above there is some limitation in the studies which used OO in patients with rheumatoid arthritis. First of all, OO was designed as placebo in almost all studies which verified the action of n-3 fatty acids to decrease inflammatory activity in RA. Although the results obtained through these studies are indirect, comparison of fish and olive oils with similar results in Kremer et al.’s (1990) work make a possible role of olive oil in RA patients very likely. Furthermore, a synergistic effect of OO on fish oil supplementation was clearly demonstrated in Berbert et al.’s
1063
CHAPTER | 114 Use of Olive Oil in Patients with Rheumatoid Arthritis
TABLE 114.4 Trials in which olive oil was used in patients with rheumatoid arthritis. Authors
n
Darlington and Ramsey (1987)
Cleland et al. (1988)
60
Design
Dose of n-3 fatty acids or oleic acid
Time of treatment
Clinical improvement with Olive oil
Laboratorial improvement with Olive oil
fish oil X olive oil
18 g
12 weeks
Yes
Yes
fish oil X olive oil
18 g d⫺1
12 weeks
Yes
No
18 g
18 g d⫺1
Kremer et al. (1990)
49
fish oil X olive oil
45 or 90 mg kg⫺1 d⫺1 X 6.8 g d⫺1 (9.6 ml d⫺1)
24 weeks
Yes
Yes
Sköldstam et al. (1992)
43
Fish oil X Olive ⫹ maize ⫹ peppermint oils
10 g d⫺1
6 months
No
No
10 g d⫺1
Geusens et al. (1994)
90
fish oil X olive oil
3 g d ⫺1 6 g d⫺1
12 months
No
No
Berbert et al. (2005)
43
fish oil X fish oil ⫹olive oil X soy oil
3 g d ⫺1
6 months
Yes
Yes
3 g d⫺1⫹ 6.8 g d⫺1
n – number of patients; d – day.
(2005) study. A future study designed to have four groups of RA patients being evaluated concomitantly constituted by a first group receiving OO, a second receiving n-3 fatty acids, the third receiving both oils, and the fourth receiving placebo, certainly would help to identify more accurately the action of OO in RA. Another issue of concern when studying oils in RA patients is to choose the ideal placebo. There is no such ideal placebo oil in inflammatory disease studies. The ideal placebo oil should have the n-6/n-3 fatty acid ratio of 4:1 or less, similar to the ratio to reach a healthy diet (Crawford et al., 2000). Soy oil has a ratio near 8 (54 n-6 : 7 n-3), and corn, sunflower, and safflower oils have even higher ratios than soy oil (Alexander, 1998), and therefore could have a pro-inflammatory influence. On the other hand, oils rich in both n-3 and n-9, like canola oil, could have an antiinflammatory effect, and should not be used as placebo in inflammatory studies as well. However, it has been verified that soy oil may amplify the stress response in severe stress, but not in moderate stress (Furukawa et al., 2002).
The data obtained in our studies with patients with ulcerative colitis (Dichi et al., 2000; Barbosa et al., 2003) and RA (Berbert et al., 2005) in mild or moderate inflammatory activity using soy oil as placebo seem to confirm that this concern is related basically to severe stress conditions. In summary, more studies specifically designed to study the effects of OO alone or in combination with fish oil in RA patients are warranted to help to decrease the doses of the classical anti-inflammatory drugs and to proportionate a better quality of life to patients with this important disability disease.
REFERENCES Alexander, J.W., 1998. Immunonutrition: the role of ω-3 fatty acids. Nutrition 14, 627–633. Barbosa, D.S., Cecchini, R., El Kadri, M.Z., Rodrigues, M.A.M., Burini, R.C., Dichi, I., 2003. Decreased oxidative stress in patients with ulcerative colitis supplemented with fish oil ω-3 fatty acids. Nutrition 19, 837–842.
1064
Berbert, A.A., Kondo, C.R.M., Almendra, C.L., Matsuo, T., Dichi, I., 2005. Supplementation of fish oil and olive oil in patients with rheumatoid arthritis. Nutrition 21, 131–136. Bogani, P., Galli, C., Villa, M., Visioli, F., 2007. Postprandial anti-inflammatory and antioxidant effects of extra virgin olive oil. Atherosclerosis 190, 181–186. Bulló, M., Casas-Agustench, P., Amigo-Correig, P., Aranceta, J., SalasSalvadó, J., 2007. Inflammation, obesity and comorbidities: the role of diet. Public Health Nutr 10 (10A), 1164–1172. Cleland, L.G., French, J.K., Betts, W.H. et al., 1988. Clinical and biochemical effects of dietary fish oil supplements in rheumatoid arthritis. J. Rheum. 15, 1471–1475. Crawford, M., Galli, C., Visioli, F., Renaud, S., Simopoulos, A.P., Spector, A.A., 2000. Role of plant-derived omega-3 fatty acids in human nutrition. Ann. Nutr. Metab. 44, 263–265. Darlington, L.G., Ramsey, N.W., 1987. Olive oil for rheumatoid arthritis? Br. J. Rheumatol. 26 (Suppl), 215. Darlington, L.G., Stone, T.W., 2001. Antioxidants and fatty acids in the amelioration of rheumatoid arthritis and related disorders. Br. J. Nut. 85, 251–269. Dichi, I., Frenhane, P., Dichi, J.B., Correa, C.R., Angeleli, A.Y.O., Bicudo, M.H., Rodrigues, M.A.M., Victoria, C.R., Burini, R.C., 2000. Comparison of ω-3 fatty acids and sulfasalazine in ulcerative colitis. Nutrition 16, 87–90. Firestein, G.S., 2003. Evolving concepts of rheumatoid arthritis. Nature 423, 356. Furukawa, K., Yamamori, H., Takagi, K., Suzuki, R., Nakagima, N., Tashiro, T., 2002. Influences of soybean oil emulsion on stress response and cell-mediated immune function in moderately or severely stressed patients. Nutrition 18, 235–240. Geusens, P., Wouters, C., Nijs, J. et al., 1994. Long-term effects of omega-3 fatty acid supplementation in active rheumatoid arthritis. A 12-month, double-blind, controlled study. Arth. Rheum. 37 (6), 824–829. Halliwell, B., Hoult, J.R., Blake, D.R., 1998. Oxidants, inflammation and anti-inflammatory drugs. FASEB J. 2, 2867–2873. James, M.J., Gibson, R.A., Neumann, M.A., Cleland, L.S., 1993. Effects of dietary supplementation with n-9 eicosatrienoic acid on leukotriene B4 synthesis in rats: a novel approach to inhibition of eicosanoid synthesis. J. Exp. Med. 178, 2261–2265.
SECTION | II Immunology and Inflammation
Kremer, J.M., Lawrence, D.A., Jubiz, W. et al., 1990. Dietary fish oil and olive oil supplementation in patients with rheumatoid arthritis. Arthritis Rheum. 33 (6), 810–819. Lee, D.M., Weinblatt, M.E., 2001. Rheumatoid arthritis. Lancet 358, 903–911. Linos, A., Kaklamani, V.G., Kaklamania, E., Koumantaki, Y., Giziaki, E., Papazoglou, S., Mantzoro, C.S., 1999. Dietary factors in relation to rheumatoid arthritis: a role for olive oil and cooked vegetables? Am. J. Clin. Nutr. 70, 1077–1082. Linos, A., Kaklamania, E., Kontomerkos, A. et al., 1991. The effect of olive oil and fish consumption on rheumatoid arthritis: a case control study. Scand. J. Rheumatol. 20, 419–426. Martinez-Domínguez, E., de la Puerta, R., Ruiz-Gutiérrez, V., 2001. Protective effects upon experimental inflammation models of a polyphenolsupplemented virgin olive oil diet. Inflamm. Res. 50, 102–106. Pacheco, Y.M., Bermúdez, B., López, S., Abia, R., Villar, J., Muriana, J.G., 2007. Minor compounds of olive oil have postprandial antiinflammatory effects. Br. J. Nutr. 98, 260–263. Pattison, D.J., Symmons, D.P.M., Young, A., 2004. Does diet have a role in the aetiology of rheumatoid arthritis? Proc. Nutr. Soc. 63, 137–143. Rao, J., Mihaliak, K., Kroenke, K., Bradley, J., Tierney, W., Weinberger, M., 1999. Use of complementary therapies for arthritis among patients of rheumatologists. Ann. Intern. Med. 131, 409–416. Silman, A.J., MacGregor, A.J., Thomson, W. et al., 1993. Twin concordance rates for rheumatoid arthritis: results from a nationwide study. Br. J. Rheum. 32, 903–907. Sköldstam, L., Börjesson, O., Kjällman, A. et al., 1992. Effects of six months of fish oil supplementation in stable rheumatoid arthritis. A double-blind, controlled study. Scand. J. Rheum. 21, 178–185. Symmons, D., Harrison, B., 2000. Early inflammatory polyarthritis: results from the Norfolk Arthritis Register with a review of the literature. I. Risk factors for the development of inflammatory polyarthritis and rheumatoid arthritis. Rheumatology 39, 835–843. Tripoli, E., Giammanco, M., Tabacchi, G., Di Majo, D., Giammanco, S., La Guardia, M., 2005. The phenolic compounds of olive oil: structure, biological activity and beneficial effects on human health. Nutr. Res. Rev. 18, 98–112. Visioli, F., Galli, C., 1998. Olive oil phenols and their potential effects on human health. J. Agric. Food Chem. 46, 4292–4296. Wahle, K.W., Caruso, D., Ochoa, J.J., Quiles, J.L., 2004. Olive oil and modulation of cell signaling in disease prevention. Lipids 39 (12), 1223–1231.
2.5
Other Effects, Uses and Diseases Cells and Cellular Effects Skin and Cosmeceuticals Major Organ Systems Including Liver and Metabolism
This page intentionally left blank
Chapter 115
The Beneficial Effects of Virgin Olive Oil on Nuclear Transcription Factor kappaB and Other Inflammatory Markers 1
Pablo Perez-Martinez, Francisco Perez-Jimenez and Jose Lopez-Miranda
Reina Sofía University Hospital, Lipids and Atherosclerosis Research Unit, University of Córdoba, CIBER Fisiopatologia de la Obesidad y Nutricion (CIBEROBN), Instituto Maimonides de Investigacion Biomedica de Córdoba (IMIBIC), Spain
115.1 INTRODUCTION Epidemiologic evidence indicates that the Mediterranean diet (MD), in which olive oil is the principal source of fat, reduces the risk of coronary heart disease (CHD) (KrisEtherton et al., 2001). Data from controlled clinical studies have shown that monounsaturated fatty acid (MUFA) intake favorably affects many risk factors related to the development of CHD. Compared with saturated fatty acids (SFAs), MUFAs lower plasma total and low-density lipoprotein (LDL cholesterol) concentrations, increase highdensity lipoprotein (HDL cholesterol) concentrations, and decrease total plasma triacylglycerol concentrations. In addition, virgin olive oil, besides high levels of MUFA, contains several minor components with biological properties. However, and particularly in the course of the past decade, a new paradigm has emerged, with the demonstration that the effects of this diet go much further than cholesterol and even traditional risk factors, as we and other groups have described in previous studies (Perez-Jimenez et al., 2002; Estruch et al., 2006). Atherosclerosis is considered an inflammatory disease, characterized by the accumulation of macrophage-derived foam cells in the vessel wall and accompanied by the production of cytokines, chemokines and growth factors. The cellular mechanism that mediates the expression of the genes involved in the inflammatory response, both in the endothelium and the other cells that participate in the inflammation of the vascular wall, depends on transcription factors, of which nuclear factor κB (NF-κB) is particularly well known, Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
since it is a nuclear transcription factor that is sensitive to oxidative changes. In this field, an interesting aspect is the demonstration that supplementing an endothelial cell culture with oleic acid reduces the transcriptional activation of this factor in these cells, similar to what is done by α-linolenic acid, an omega-3 fatty acid, and the opposite of the inflammatory effect of linoleic acid (Toborek et al., 2002). This is in agreement with an earlier study of Carluccio et al. (1999), who observed, also in an endothelial cell culture model, that the incorporation of oleic acid into cellular membrane lipids reduced the expression of vascular cell adhesion molecule 1 (VCAM-1). Furthermore, we ourselves have observed that the expression of VCAM-1 and E-selectin in human umbilical vascular endothelial cells (HUVECs), following the addition of minimally oxidized LDL, was less with LDL obtained from persons who had followed a diet rich in olive oil than from persons whose diet was rich in saturated fat (Bellido et al., 2006). This anti-inflammatory action of MUFA also explains the fact that the enrichment of LDL particles with oleic acid, during the consumption of different types of diet, reduces their capacity to induce monocyte chemotaxis and adhesion (Tsimikas et al., 1999). In accordance with our results, a previous study has shown that LDL obtained from a MUFA-rich diet induced a lower rate of monocyte adhesion to endothelial cells (Mata et al., 1996). The mechanism by which LDL from carbohydrate (CHO) and Mediterranean diets induces a lower expression of VCAM-1 and E-selectin is unknown; however, several hypotheses have been suggested.
1067
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1068
C
A
NF-κB
Free probe
B 3.5 3 Arbitrary units
For instance, the interaction of mononuclear leukocytes with vascular endothelial cells is most likely mediated by a complex amalgam of interacting regulatory signals in the inflammatory response characteristic of early atherogenesis. A common feature underlying these complex regulatory interactions may involve redox-sensitive nuclear transcription factors that modulate the expression of gene products (e.g., endothelial adhesion molecules) (Crossin, 2002). NF-κB is a redox-sensitive transcription factor; therefore, agents that diminish oxidant stress may stabilize the NFκB system, which would attenuate atherosclerotic lesion formation. Hennig et al. (2000) exposed porcine endothelial cells to 18-carbon fatty acids. Both linoleic and stearic fatty acids activated endothelial cells more markedly than did either oleic or linolenic fatty acids. Also, compared with control cultures, treatment with stearic and linoleic acids decreased glutathione concentrations, which suggested an increase in cellular oxidative stress. This increase in oxidative stress with the subsequent activation of NF-κB could be one of the mechanisms of the inflammatory properties of 18:0 and 18:2. Thus, in a further step, we have observed that the consumption of olive oil reduces the expression of NF-κB in mononuclear cells obtained from healthy subjects during the postprandial phase, similar to the effect that has also been observed following the ingestion of linolenic acid, and the opposite of the pro-inflammatory effect of saturated fats. Furthermore, in the same study, olive oil reduced plasma levels of intercellular adhesion molecule 1 (ICAM-1), another adhesion molecule (Bellido et al., 2004). This anti-inflammatory effect has been observed in metabolic syndrome patients who modified their diets for 2 years. In the group that followed an MD model, the prevalence of this syndrome was reduced, improved insulin sensitivity and lowered the levels of C-reactive protein (CRP) and interleukin 6 (IL-6), 7 and 18 (Esposito et al., 2004). More recently, in a sample of more than 700 persons at high risk of cardiovascular disease, it was observed that a diet rich in virgin olive oil lowered CRP levels, as opposed to a fat-poor diet and another that was rich in walnuts. Olive oil consumption also reduced levels of IL-6, VCAM-1 and ICAM-1 (Estruch et al., 2006). Nevertheless, no data exist about the chronic effect of dietary models on the activation of NF-κB in a healthy population, which is more reliable for potential recommended intervention of dietary habits. Then, we have recently examined the effect of three dietary models, MD, typical Western and high CHO enriched in n-3 diets on the activation of NFκB in mononuclear cells from healthy volunteers (Figure 115.1) (Perez-Martinez et al., 2007). Interestingly, we have observed that consumption of the MD decreases the activation of NF-κB compared with a typical Western diet. These findings suggest that chronic
on tro l
115.2 DIETARY FAT AND NF-κB IN MONONUCLEAR LEUKOCYTES
W es te rn di et M ed ite rra ne CH an O di an et d n3 di et
SECTION | II Cells and Cellular Effects
2.5 2 1.5 1 0.5 0 Western diet
Mediterranean diet
CHO and n-3 diet
FIGURE 115.1 NF-κB activation in peripheral mononuclear cells obtained after the intake of the dietary models: Western, Mediterranean and high CHO diet enriched in n-3 fatty acid diets. (A) Representative electrophoretic activity shift assay (EMSA) of nuclear proteins from peripheral mononuclear cells showed a retarded band that was increased 2.7-fold with Western diet compared with Mediterranean diet and 1.79fold than the intake of a high CHO enriched in n-3. No differences were found in NF-κB activation between the Mediterranean and high CHO enriched in n-3 diets. (B) Densitometer quantification of NF-κB activity in peripheral mononuclear cells. Two-factor analysis of variance (ANOVA) for repeated measures. Results are expressed in arbitrary units (generated by densitometer). Results are mean for all of the volunteers (n ⫽ 16) ⫾ S.E.M. *p , 0.05 vs. Western diet. Reprinted from PerezMartinez et al., Atherosclerosis 2007; 194:141–146, with permission.
consumption of olive oil could prevent the activation of the NF-κB system, and this effect could be associated with either the intake of MUFAs or the protective influence of the antioxidant components contained in virgin olive oil. In this sense, it is known that antioxidants reduce NF-κB activity. These protective effects of the olive oil on the NF-κB activity, partly attributed to its antioxidant compounds, have also been suggested for red wine, indicating that foods
CHAPTER | 115 The Beneficial Effects of Virgin Olive Oil on Nuclear Transcription Factor
being particularly effective on the p50 subunit. Interestingly, this effect occurred at concentrations found in human plasma after nutritional ingestion of virgin olive oil and was quantitatively similar to the effect exerted by glitazone, a PPAR-γ ligand. However, MPC-OOE did not affect PPAR-γ expression in monocytes and MDM (Brunelleschi et al., 2007). These data provide further evidence of the beneficial effects of extra virgin olive oil by indicating its ability to inhibit NFκB activation in human monocyte/macrophages.
800 750 700 VCAM-1 (ng mL–1)
1069
650 600 550 500 450 400 350 300 Western diet
Mediterranean diet
CHO and n-3 diet
FIGURE 115.2 Plasma concentrations of VCAM-1 after the intake of the three dietary models: Western, Mediterranean and high CHO diet enriched in n-3 fatty acid diets. Results are means of all volunteers (n ⫽ 16) ⫾ S.E.M. *p , 0.05 vs. Western diet. Western diet increases VCAM-1 plasma concentrations compared with a Mediterranean and a high CHO enriched in n-3 diets (663 ⫾ 97 versus 495 ⫾ 35 and 486 ⫾ 49, respectively, p ⬍ 0.05). However, no significant differences were found in VCAM1 concentrations between the ingestion of a Mediterranean diet and a high CHO diet enriched in n-3. Reprinted from Perez-Martinez et al., Atherosclerosis 2007; 194:141–146, with permission
with higher antioxidant content could have a major inhibitory effect on NF-κB. Besides olive oil, the MD contains other sources of potentially cardioprotective nutrients from fruits and vegetables which could also enhance this beneficial effect (Wahle et al., 2004). In contrast, the opposite effect was observed after the chronic intake of a Western diet rich in saturated fatty acids, corroborating our previous data after the acute intake of a butter meal. The effect of a high CHO diet enriched in n-3 fatty acids on the NF-κB activation was intermediate. In this sense, previous studies have suggested that n-3 α-linolenic acid found mainly in plants and walnuts may reduce cardiovascular risk through a variety of biological mechanisms, including inhibiting vascular inflammation (De Caterina et al., 2004). In our study, the intake of a high CHO diet enriched in n-3 fatty acids showed a tendency to decrease NF-κB activation, indicating that α-linolenic acid exerts an anti-inflammatory effect through the NF-κB system. We have studied VCAM-1 plasma concentrations after the consumption of the three diets, because this protein is also regulated by the NF-κB system. Consequently, the intake of a Western diet increases VCAM-1 plasma concentrations compared with Mediterranean and high CHO enriched in n-3 diets, indicating that the last two diets may provide additional benefits compared with the typical Western diet (Figure 115.2). In the same line, but at functional level, Brunelleschi et al. (2007) investigated the effects of an extra virgin olive oil extract (OOE), particularly rich in minor polar compounds (MPC), on NF-κB translocation in monocytes and monocyte-derived macrophages (MDM) isolated from healthy volunteers. In a concentration-dependent manner, MPC-OOE inhibited p50 and p65 NF-κB translocation in both unstimulated and phorbol-myristate acetate (PMA)-challenged cells,
115.3 CONCLUSION Although more data are mandatory, the anti-inflammatory effect of olive oil needs to be considered in the context of the inflammation that is produced during the ingestion of highenergy diets that can promote an overproduction of reactive oxygen species and inducing changes in complement fraction 3, with the resulting endothelial activation, leukocyte adhesion to blood vessel walls and their subsequent emigration to the subendothelial space (Charo and Ransohoff, 2006; van Oostrom et al., 2007). These data, together with previous studies, suggest that the beneficial effects of MD and olive oil in particular, are not limited to their effects on plasma cholesterol levels. These evidences indicate that other atherosclerosis pathogenic mechanisms, especially endothelium, may be influenced by these dietary interventions. However, despite the significant advances of recent years, the final proof about the specific mechanisms and contributing role of the different dietary models and nutrients to its beneficial effects requires further investigations.
SUMMARY POINTS ●
●
●
●
●
Previous evidences indicate that atherosclerosis pathogenic mechanisms, especially endothelium, may be influenced by dietary interventions. Nuclear factor κB (NF-κB) is a redox-sensitive transcription factor; therefore, agents that diminish oxidant stress may stabilize the NF-κB system. Consumption of virgin olive oil reduces the expression of NF-κB in mononuclear cells obtained from healthy subjects during the postprandial phase. LDL obtained from a MUFA-rich diet induced a lower rate of monocyte adhesion to endothelial cells. The intake of a Western diet increases VCAM-1 plasma concentrations compared with a Mediterranean and a high CHO enriched in n-3 diets.
ACKNOWLEDGMENTS This work was supported by research grants from CIBER (CBO/6/03), Instituto de Salud Carlos III; CICYT (SAF 01/2466-C05 04 to F P-J, SAF 01/0366 to J L-M, AGL 2004-07907 to J L-M, AGL 2006-01979 to JL-M),
1070
the Spanish Ministry of Health (FIS 01/0449, FIS PI041619 to CM); Fundación Cultural “Hospital Reina Sofía-Cajasur”; Consejería de Salud, Servicio Andaluz de Salud (00/212, 00/39, 01/239, 01/243, 02/64, 02/65, 02/78, 03/73, 03/75, 04/237, 04/191, 04/238, 05/396); Consejería de Educación, Plan Andaluz de Investigación, Universidad de Córdoba.
REFERENCES Bellido, C., Lopez-Miranda, J., Blanco-Colio, L.M., Perez-Martinez, P., Muriana, F.J., Martin-Ventura, J.L., Marin, C., Gomez, P., Fuentes, F., Egido, J., Perez-Jimenez, F., 2004. Butter and walnuts, but not olive oil, elicit postprandial activation of nuclear transcription factor kappaB in peripheral blood mononuclear cells from healthy men. Am. J. Clin. Nutr. 80, 1487–1491. Bellido, C., Lopez-Miranda, J., Perez-Martinez, P., Paz, E., Marin, C., Gomez, P., Moreno, J.A., Moreno, R., Perez-Jimenez, F., 2006. The Mediterranean and CHO diets decrease VCAM-1 and E-selectin expression induced by modified low-density lipoprotein in HUVECs. Nutr. Metab. Cardiovasc. Dis. 16, 524–530. Brunelleschi, S., Bardelli, C., Amoruso, A., Gunella, G., Ieri, F., Romani, A., Malorni, W., Franconi, F., 2007. Minor polar compounds extra-virgin olive oil extract (MPC-OOE) inhibits NF-kappa B translocation in human monocyte/macrophages. Pharmacol. Res. 56, 542–549. Carluccio, M.A., Massaro, M., Bonfrate, C., Siculella, L., Maffia, M., Nicolardi, G., Distante, A., Storelli, C., De Caterina, R., 1999. Oleic acid inhibits endothelial activation: a direct vascular antiatherogenic mechanism of a nutritional component in the Mediterranean diet. Arterioscler. Thromb. Vasc. Biol. 19, 220–228. Charo, I.F., Ransohoff, R.M., 2006. The many roles of chemokines and chemokine receptors in inflammation. N. Engl. J. Med. 354, 610–621. Crossin, K.L., 2002. Cell adhesion molecules activate signaling networks that influence proliferation, gene expression, and differentiation. Ann. N. Y. Acad. Sci. 961, 159–160. De Caterina, R., Madonna, R., Massaro, M., 2004. Effects of omega-3 fatty acids on cytokines and adhesion molecules. Curr. Atheroscler. Rep. 6, 485–491. Esposito, K., Marfella, R., Ciotola, M., Di Palo, C., Giugliano, F., Giugliano, G., D’Armiento, M., D’Andrea, F., Giugliano, D., 2004. Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA 292, 1440–1446.
SECTION | II Cells and Cellular Effects
Estruch, R., Martinez-Gonzalez, M.A., Corella, D., Salas-Salvado, J., Ruiz-Gutierrez, V., Covas, M.I., Fiol, M., Gomez-Gracia, E., Lopez-Sabater, M.C., Vinyoles, E., Aros, F., Conde, M., Lahoz, C., Lapetra, J., Saez, G., Ros, E., PREDIMED Study Investigators, 2006. Effects of a Mediterranean-style diet on cardiovascular risk factors: a randomized trial. Ann. Intern. Med. 145, 1–11. Hennig, B., Meerarani, P., Ramadass, P., Watkins, B.A., Toborek, M., 2000. Fatty acid-mediated activation of vascular endothelial cells. Metabolism 49, 1006–1013. Kris-Etherton, P., Eckel, R.H., Howard, B.V., St Jeor, S., Bazzarre, T.L., 2001. Benefits of a Mediterranean-style, national cholesterol education program/American heart association Step I dietary pattern on cardiovascular disease. Circulation 103, 1823–1825. Mata, P., Alonso, R., Lopez-Farre, A., Ordovas, J.M., Lahoz, C., Garces, C., Caramelo, C., Codoceo, R., Blazquez, E., de Oya, M., 1996. Effect of dietary fat saturation on LDL oxidation and monocyte adhesion to human endothelial cells in vitro. Arterioscler. Thromb. Vasc. Biol. 16, 1347–1355. Perez-Jimenez, F., Lopez-Miranda, J., Mata, P., 2002. Protective effect of dietary monounsaturated fat on arteriosclerosis: beyond cholesterol. Atherosclerosis 163, 385–398. Perez-Martinez, P., Lopez-Miranda, J., Blanco-Colio, L., Bellido, C., Jiménez, Y., Moreno, J.A., Delgado-Lista, J., Egido, J., PerezJimenez, F., 2007. The chronic intake of a Mediterranean diet enriched in virgin olive oil, decreases nuclear transcription factor kappaB activation in peripheral blood mononuclear cells from healthy men. Atherosclerosis 194, 141–146. Toborek, M., Lee, Y.W., Garrido, R., Kaiser, S., Hennig, B., 2002. Unsaturated fatty acids selectively induce an inflammatory environment in human endothelial cells. Am. J. Clin. Nutr. 75, 119–125. Tsimikas, S., Philis-Tsimikas, A., Alexopoulos, S., Sigari, F., Lee, C., Reaven, P.D., 1999. LDL isolated from Greek subjects on a typical diet or from American subjects on an oleate-supplemented diet induces less monocyte chemotaxis and adhesion when exposed to oxidative stress. Arterioscler. Thromb. Vasc. Biol. 19, 122–130. van Oostrom, A.J., Alipour, A., Plokker, T.W., Sniderman, A.D., Cabezas, M.C., 2007. The metabolic syndrome in relation to complement component 3 and postprandial lipemia in patients from an outpatient lipid clinic and healthy volunteers. Atherosclerosis 190, 167–173. Wahle, K.W., Caruso, D., Ochoa, J.J., Quiles, J.L., 2004. Olive oil and modulation of cell signaling in disease prevention. Lipids 39, 1223–1231.
Chapter 116
In vivo Cytogenetic Effects of Multiple Doses of Dietary Vegetable Oils: Position of Olive Oils Lusânia Maria Greggi Antunes and Maria de Lourdes Pires Bianchi Departamento Análises Clínicas, Toxicológicas e Bromatológicas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto – USP, São Paulo, Brasil
116.1 INTRODUCTION Recent attention has focused on the growing body of experimental evidence and experimental studies supporting the hypothesis that high consumption of fruits and vegetables is capable of inhibiting, retarding or reversing the multiple steps of mutagenic and carcinogenic processes. Experimental studies have demonstrated that dietary compounds, such as carotenoids, vitamins E and A, folic acid and phenolics could contribute to increasing the antioxidant defense and, as a consequence, inhibit mutagenesis and carcinogenesis. The Mediterranean diet, rich in fruits, vegetables, fish, red wine and olive oil can contribute to a lower incidence of coronary heart disease and cancer (Tuck and Hayball, 2002). Dietary constituents that are mutagenesis and carcinogenesis inhibitors are of particular importance, because they may be useful in preventing human cancer due to their non-toxic effects. Olive oil is the predominant source of fat in the Mediterranean diet, which has been associated with a lower incidence of colorectal and breast cancers. Many endogenous substances, usually present in fruits and vegetables, possess some inhibitory activity towards biological, chemical or physical mutagenic substances that often increase cancer incidence. Recently, much attention has been given to the antimutagenic effects of natural dietary compounds. The principal focus is the identification and characterization of dietary antimutagenic agents that are very likely to offer chemoprotection against the damage induced by environmental mutagens, such as chemical carcinogens and ionizing radiations. Dietary antioxidants would likely exert their protective action by counteracting the oxidative damaging effect on cellular components. According to Fenech (2008), DNA is constantly exposed to damaging agents that act by different Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
mechanisms, including chromosome aberrations and chromosome rearrangement. Chromosome rearrangements arise from structural change by translocation or inversion. Chromosome or chromatid breaks and other rearrangements can contribute to loss of DNA nucleotides or genes and genome instability, with subsequent transformation of a normal cell into cancerous cells.
116.2 CHROMOSOME STUDIES IN GENETIC TOXICOLOGY The field of mutation research is mainly based on investigations and original observations that mutations are caused by DNA replication errors during cell proliferation or by exposure to specific physical agents, chemical mutagens or viruses that are present in our environment as natural or anthropogenic products. These mutations can occur deliberately under cellular control, during processes like meiosis or mitosis and can be subdivided into germline mutations, which can be passed on to descendants, and somatic mutations, which cannot. These events can produce significant critical alterations to the genome and are a cause of cancer, neuronal diseases and developmental defects (Antunes and Bianchi, 2008). Among the parameters applied in genetic toxicology, chromosome studies are an important laboratorial diagnostic procedure and these tests are increasingly used to identify mutagenic and antimutagenic agents in different cells and tissues. Cytogenetics is a research field that studies the changes in chromosome numbers or structure. Many cytogenetic tests exist that analyze animal and human chromosomes, micronuclei or sister chromatid exchange. The most frequently observed types of chromosome aberrations are chromatid or chromosome breaks. Micronuclei formation is
1071
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1072
the result of chromosome fragment or whole chromosomes that lag behind at anaphase during nuclear division (Fenech, 2008). The formation of micronuclei originating from chromosome fragments or chromosome loss events requires mitotic or meiotic division. Sister chromatid exchange is the result of breaks in both DNA strands followed by an exchange of whole DNA strands during mitosis or cell division. Potent mutagenic agents are able to increase the rate of sister chromatid exchange. Direct and indirect evidence suggests that DNA is the main target of mutagenic agents responsible for the induction of chromosome aberrations. These alterations are disruptions in the normal chromosome of a cell. Chromosome aberrations in lymphocytes and bone marrow cells are thought to represent a surrogate endpoint for more specific chromosome alterations in target tissues of carcinogenesis. Clastogenic agents, for example, those that break chromosomes, can be classified as S-dependent agents that produce chromatid-type aberrations, or as S-independent agents that are able to induce chromosome-type and various others types of damage (Antunes and Bianchi, 2008). Cytogenetic biomarkers, such as the frequency of chromosomal aberrations and micronuclei, have been applied as intermediate endpoints to evaluate the DNA-damaging effects resulting from a wide range of exposures in the diet and dietary supplementation. These biomarkers are the most frequently used endpoint in human population studies and the most sensitive for measuring exposure to mutagenic agents and in their role as early predictors of cancer risk (Bonassi et al., 2005). The cytogenetic characterization and classification of different types of chromosome aberrations have an important role in human genetics and many cancers are associated with specific types of aberrations. Identification of the principal chromosome damage is of broad utility for research into cancer and mutagenesis (Heng et al., 2006). Since few studies are available on the mutagenic and antimutagenic effects of olive oils, the purpose of this chapter is to retrieve data from literature papers that discuss the effects of olive oils regarding cytogenetic analysis and summarize the possible mechanisms of olive oils in preventing chromosomal damage in experimental models in rodents.
116.3 OLIVE OIL Olive oil, apart from being a principal source of fatty acids in the Mediterranean diet, contains a number of phenolic compounds that have demonstrated antioxidant effects in vivo and in vitro. Phenolic compounds found in olive oil and virgin olive oil could provide protective effects by inhibiting oxidative damage. Investigation of the associations between freeradical-induced DNA damage and the antioxidant properties of dietary oils is a field of great interest for elucidating mechanisms of mutagenesis and carcinogenesis (Owen et al., 2000).
SECTION | II Cells and Cellular Effects
In the last few years, a significant increase in the number of publications regarding the determination of the composition and biological profiles of phenolics found in olive oils has been observed (Nousis et al., 2005). Ever since El-Nahas et al. (1993) reported that irradiated albino rats pretreated for 6 months with olive oil presented fewer chromosome aberrations than those pretreated with vitamin E and, thus, that olive oil presented a radioprotective effect in rats subjected to whole-body irradiation, our group has been interested in evaluating the effects of dietary vegetable oils on DNA-induced damage by quantifying chromosome aberrations as cytogenetic endpoints of chemoprevention. Early on, the mutagenicity of olive oil was investigated in rat bone marrow cells in vivo due to concerns regarding the recommendation by research guidelines that suggested that in in vivo chromosome aberration tests, if the substance test was insoluble in water or saline, it should be dissolved or homogeneously suspended in vegetable oil (Tice et al., 1994). However, it is possible that the antimutagenic effect of the oils themselves could interfere with the results obtained. A single dose of olive oil of 10 mL kg⫺1 body weight administered by gavage simultaneously with an intraperitoneal (i.p.) injection with antitumoral doxorubicin in Wistar rats, a known inducer of chromosome damage, resulted in a significant reduction in the total of chromosome aberrations by 48.6% when compared with the group that receive the antitumoral agent alone (Antunes and Takahashi, 1999). In these investigations, the antimutagenic effects of olive oil were evident and ascribed to the antioxidants found in olive oil (Figure 116.1). This prompted our group to undertake other cytogenetic tests with dietary oils in rodents. Bone marrow cells were chosen because they present high mitogenic activity, are highly susceptible and provide great sensitivity for detecting whether or not a chemical or natural product under study can induce chromosome aberrations at a frequency significantly higher than that found in unexposed control animals. The most appropriate way to predict effects in humans is to use a route of exposure that most resembles that anticipated or known to be the route of human exposure. After the mutagen or the agent tested exposure, DNA repair mechanisms are responsible for both the maintenance of chromosome integrity and the persistence of cytogenetic damage. The conclusions that can be drawn from in vivo cytogenetic bone marrow cells are that a compound is clearly mutagenic or that it is non-mutagenic within the restrictions of the protocol (Preston et al., 1987).
116.4 OLIVE OIL ANTIMUTAGENICITY Previous studies from our laboratory assessed the cytotoxicity of dietary oils in bone marrow cells of male Wistar rats. The mitotic index is an important endpoint in the cytogenetic tests. It can be used to determine the maximum dose that
CHAPTER | 116 In vivo Cytogenetic Effects of Multiple Doses of Dietary Vegetable Oils: Position of Olive Oils
1073
Without antioxidant
Free radicals Chromosome break
With antioxidant No break
Olive oil
Free radicals
Antioxidants
Scavenging of free radicals by antioxidants
FIGURE 116.1 Possible mechanism of olive oil actions in the inhibition of free-radical-induced chromosome aberrations. The protective effects of olive oil were evident and ascribed to the antioxidants found in olive oil.
will be investigated in the chromosomal aberrations tests. If significant cytotoxicity or slowed cell cycle progression occurs, a reduction in the mitotic index can be observed. The results of the cytotoxic assay involving different dietary oils, such as olive, virgin olive, canola or corn oils at doses of 5 mL kg⫺1 body weight administered by gavage in an acute or subacute treatments, showed mitotic indices similar to the untreated control groups (Evangelista et al., 2004, 2006). Another objective of our group was the evaluation of the mutagenic effects of four vegetal dietary oils in rat bone marrow cells. Olive oil is very high in monounsaturated fats, which are low-risk fats, a fact determined in animal experiments. Major phenolic compounds present in olive oils, such as tyrosol, hydroxytyrosol, oleuropein, caffeic acid and others, are regarded as strong antioxidants and radical scavengers (Owen et al., 2000). Canola oil was chosen because some reports suggest that this oil also possesses antioxidant compounds (Amarowiczs et al., 2000). Raj and Katz (1984) observed a reduction in 7,12-dimethylbenz[a]anthraceneinduced chromosome aberrations in mouse bone cells pretreated with corn oil. Olive, virgin olive, canola and corn oils are commonly consumed by humans and frequently used as drug vehicles. The experiments were performed with 6–7-week-old healthy male Wistar rats, weighing 100–110 g. The rats were maintained under laboratory conditions on a 12 h light/dark cycle at 23 ⫾ 2°C, were housed in polycarbonate cages with steel wire tops, with free access to standard rat chow and fresh water. The study was approved by the Animal Ethics Committee of São Paulo University, Ribeirão Preto Campus.
Based on the analyses of preliminary experiments, the dose of the oils selected was 5 mg kg⫺1 body weight for the acute treatment (24 h) and for the subacute treatment (72, 48 and 24 h) prior to euthanasia. The results showed that acute and subacute treatments with the chosen oils did not induce any increase in total chromosomal aberrations or in the number of metaphases with aberrations in rat bone marrow cells, when compared with the untreated controls. Thus, these dietary oils were not clastogenic under the test conditions (Evangelista et al., 2004, 2006). These data are in agreement with other published results which showed that olive and corn oil were not mutagenic in Wistar rat bone marrow cells (Antunes and Takahashi, 1999; Sendão et al., 2006). Only two other papers were found in the literature concerning the mutagenicity of olive oil evaluated by cytogenetic biomarkers, the results from El-Nahas et al. (2003) already discussed above and a study by Elmadfa and Park (1999). These authors compared the effects of two dietary oils, γ-tocopherol in corn oil and α-tocopherol in an olive/ sunflower oil mixture, considering their different tocopherol and fatty acid patterns, in order to assess DNA damage by the sister chromatid exchange assay in healthy young men. The effect of two different dietary oils on the frequency of sister chromatid exchange in human lymphocytes of healthy young men was studied. After 2 weeks of this diet, the results showed that the corn oil diet significantly reduced sister chromatid exchange, compared to values at baseline. The mean rate of sister chromatid exchange after the olive/ sunflower oil diet was not significantly different compared
1074
to the corn oil diet. The sister chromatid exchange is a sensitive indicator of DNA damage in mammalian cells and the dietary oil investigated presented no mutagenic effects in human lymphocytes. Five weeks after the diet intervention, the mean sister chromatid exchange frequencies returned to basal levels (Elmadfa and Park, 1999). Previous investigation evaluated the mutagenicity of dietary vegetable oils. Seven vegetable oils consumed by humans were tested for mutagenic activity in the Drosophila somatic mutation and recombinant test. These one-generation tests have proven to be very efficient and sensitive. This assay can detect different mutagenic events, both mitotic recombination and mutation. Drosophila larvae were fed with 6% or 12% of each of the oil tested. Sesame and soy oils showed some mutagenic activity, sunflower and lowgrade olive oil presented inconclusive results and virgin olive oil was clearly non-mutagenic (Rojas-Molina et al., 2005). At concentrations up to 75 μg mL⫺1, olive oil extracts showed a marked protection of around a 60% decrease in H2O2-induced DNA damage in Jurkat cells (immortalized line of T lymphocytes) in culture medium by the single cell gel electrophoresis methodology. On the other hand, above a concentration of 100 μg mL⫺1, olive oil exerted DNAdamaging effects, inducing the formation of DNA singlestrand breaks (Nousis et al., 2005). It should be highlighted that in vitro study does not necessarily reproduce in vivo effects. Vegetable oils may show different effects on specific hepatic CYP isoforms (cytochrome P450 enzymes) and may contribute to metabolism variability when xenobiotics are administered using dietary oil as drug vehicles (Brunner and Bai, 2000). The absence of any mutagenic effects of olive and virgin olive oils in rodent bone marrow cells and human lymphocytes are of great importance, since these dietary oils are widely used by humans. These results led to other experimental investigations to determine whether olive oil could be an antimutagenic agent when tested in association with chemical agents.
116.5 POSSIBLE MECHANISMS OF OLIVE OIL ANTIMUTAGENICITY Many dietary compounds described in numerous foods and beverages from natural sources and used in human consumption are active as antimutagenic or anticarcinogenic agents. A wide variety of antioxidant compounds, such as carotenoids, vitamins C and E and many phenolics, found mainly in vegetables and fruits, have been extensively investigated and attracted attention as promising chemopreventive agents. Phenolic antioxidants from herbal, medicinal or edible plants have recently received much attention as promising agents for reducing the risk of free-radical-induced neurodegenerative and cardiovascular diseases and cancer. The use of naturally occurring antioxidant agents that are found in foods and beverages for chemoprevention
SECTION | II Cells and Cellular Effects
purposes provides a strategy for inhibiting cancer that should exhibit limited toxicity and mutagenicity. However, knowledge regarding the mode of action of constituents in olive oils on rodent and human cells exposed to DNAdamaging agents, remains limited. Potent mutagenic activity was identified with mutagenic heterocyclic amines that are formed during the cooking of meat or fish, and some of these are considered to be possible human carcinogens. The presence of virgin olive oil in an in vitro test inhibited the formation of mutagenic heterocyclic amines by 50% compared with controls (Monti et al., 2001). The antioxidant capacity and the possible effects on DNA double-strand breaks of two different fat diets, one involving virgin oil and the other involving sunflower oil, were evaluated in Wistar male rats. DNA double-strand breaks were higher in the rats fed on a sunflower oil diet. DNA damage to peripheral lymphocytes was lower after the consumption of virgin olive oil (Quiles et al., 2004). The identification of additional and more effective antimutagenic compounds could contribute to improving the use of dietary compounds that can be useful during chemotherapy. Since the use of complementary medicines in cancer patients in combination with their conventional therapy may increase the risks of unwanted interactions, our group was encouraged to investigate the possible antimutagenicity of dietary oils against the mutagenicity of the antitumoral agents doxorubicin and cisplatin using cytogenetic biomarkers, such as chromosomal aberrations. Elucidating the cellular mechanisms by which olive oil could reduce cancer risks is of some interest. Many compounds that are well-known mutagenic agents in mammalian systems or substances used as cancer chemotherapeutic drugs that are capable of inducing chromosomal aberrations have been tested in short-term animal experiments in association with possible antimutagenic agents, to reduce or inhibit their mutagenic or clastogenic activity. To investigate the possible antimutagenic effect of pretreatment with olive, virgin olive, canola or corn oil on cisplatin-induced chromosome aberrations in Wistar bone marrow cells, the rats received pretreatment with a single dose of dietary oils (5 mL kg⫺1 body weight) by gavage 30 min before cisplatin i.p. (5 mg kg⫺1 body weight). Untreated groups were similarly treated with water by gavage and saline i.p. Euthanasia occurred 24 h after cisplatin or saline administration. Subacute treatments with these dietary oils were also performed. In this case, the rats receive three doses of the dietary oils, 72, 48 and 30 min prior to cisplatin i.p. The dietary oils tested in these experiments did not alter the mitotic index. The mitotic index, evaluated as the percentage of dividing cells, showed no significant variation between treatments with olive and virgin olive oil when compared with untreated controls. Cisplatin showed a slight inhibition in the mitotic index, but the values were recovered when associated with olive and virgin oils (Figure 116.2). Acute and subacute treatments with the chosen dietary oils reduced the chromosomal damage induced by the
1075
CHAPTER | 116 In vivo Cytogenetic Effects of Multiple Doses of Dietary Vegetable Oils: Position of Olive Oils
chemotherapeutic drug cisplatin. In the acute treatment, a single dose of olive, virgin olive and canola oil caused a statistically significant decrease in the total of chromosome aberrations (Figure 116.3). The administration of olive, virgin olive and canola oils resulted in a reduction in cisplatin-induced chromosome damage of 45.7, 41.5, and 36.9% respectively, in comparison with the cisplatin groups (Evangelista et al., 2004). The subacute administration of dietary oils prior to the antitumoral agent seemed most effective at protecting bone marrow cells against cisplatin mutagenicity. The results indicated a 47.9 and 55.5% decrease in cisplatin-induced
chromosome damage in rats that received canola and olive oil, respectively. Virgin olive oil was also effective, resulting in a decrease of 56.5% in induced damage when compared to the cisplatin group (Evangelista et al., 2006). Figure 116.4 shows the comparative percentage reduction in bone marrow cells showing cisplatin-induced chromosome aberrations by the dietary oils. The protective effects of olive and virgin oils on DNA damage measured as single-, double-strand breaks or chromosome aberrations are frequently associated with antimutagenicity. The exact mechanisms of the antimutagenic effects of olive oil are not well understood. The most
Mean mitotic index (%)
4
3
2
1
0 Untreated control
OLO 5.0
VOO 5.0
cDDP 5.0 cDDP + OLO cDDP + VOO
% Reduction of chromosome aberrations
FIGURE 116.2 Mitotic index after olive and virgin olive oil treatments. Histogram showing the reduction in mitotic indexes in bone marrow cells of male Wistar rats treated with cisplatin i.p. (cDDP 5.0 mg kg⫺1 body weight) alone or in association with olive oil (OLO) or virgin olive oil (VOO) at 5.0 mL kg⫺1 body weight by gavage, and respective untreated group, euthanized 24 h after the treatments. One thousand cells were analyzed per rat, for a total of 6000 cells per treatment (Evangelista et al., 2006).
60
Acute
50 40 30 20 10 0 OLO
FIGURE 116.3 Photomicrographs showing chromosomal aberrations in the bone marrow cells of Rattus norvegicus. (A) Untreated control with normal metaphase; (B) treatment with 5.0 mg kg⫺1 bw of the antitumor drug cisplatin i.p.; (C) treatment with 5.0 mg kg⫺1 bw of olive oil; (D) treatment with cisplatin and olive oil. F ⫽ fragment, B ⫽ chromatid break.
Subacute
VOO
CAO
COO
FIGURE 116.4 Histogram showing the percentage of reduction in the total of chromosome aberrations induced by cisplatin in bone marrow cells of male Wistar rats under acute or subacute treatment with 5 mL kg⫺1 bw of olive oil (OLO), virgin olive oil (VOO), canola oil (CAO) or corn oil (COO), euthanized 24 h after the treatments. One hundred cells were analyzed per rat, for a total of 600 cells per treatment (Evangelista et al., 2004, 2006).
1076
SECTION | II Cells and Cellular Effects
likely mechanism of olive oil activity in these antimutagenicity studies is that this dietary oil acts as an antioxidant, inhibiting the damages that would be induced by reactive oxygen species released by mutagenic agents. Many phenolic compounds have demonstrated clear antioxidant properties attributable to their free-radical-scavenging and metal-chelating properties. It should be noted that phenolics might exert other biological activities, such as effects on cell-signaling pathways and on gene expression (Soobrattee et al., 2005). The antioxidant effects of olive oils probably ascribe to a combination of its high oleic acid content and a variety of plant antioxidants, particularly oleuropein, hydroxytyrosol, tyrosol, caffeic acid and some minor components, such as rutin, luteolin and squalene (Visioli and Galli, 2002). Caffeic acid showed significant protection against the mutagenicity of 7,12-dimethylbenz[a]anthracene-induced micronuclei in mouse bone marrow cells (Raj et al., 1983). Rutin, isolated from a polyphenolic extract of Olea europaea L. leaves, was antimutagenic and induced a reduction in the frequency of micronucleated polychromatic erythrocytes in mouse bone marrow after X-ray irradiation. The antimutagenic activity of the total extract was ascribed to the presence of oleuropein, rutin and hydroxytyrosol (Benavente-García et al., 2002). A recent and interesting review by Fitó et al. (2007) summarizes the evidence concerning the protective role of the major and minor components of olive oil and oxidative stress. In conclusion, the results presented in this chapter should help elucidate that olive oil is not mutagenic and that it presents antimutagenic effects in cytogenetic shorttest in vivo. The mechanisms of olive oil protection appear to be very complex, depending on the doses applied and the parameters observed. More investigations are required to more clearly understand the activity of olive oil and its antioxidant constituents in foods, so that the beneficial and protective properties of olive and virgin olive oils on human health can be characterized before suggesting its use in future chemoprevention trials.
●
●
●
●
●
●
●
●
●
●
SUMMARY POINTS ●
●
●
High consumption of fruits and vegetables is capable of inhibiting, retarding or reversing the multiple steps of mutagenic and carcinogenic processes. Experimental studies have demonstrated that dietary compounds, such as carotenoids, vitamins E and A, folic acid and phenolics could contribute to increasing the antioxidant defense and, as a consequence, inhibit mutagenesis and carcinogenesis. Dietary constituents that are mutagenesis and carcinogenesis inhibitors are of particular importance, because they may be useful in preventing human cancer due to their non-toxic effects.
The principal focus is the identification and characterization of dietary antimutagenic agents that are very likely to offer chemoprotection against the damage induced by environmental mutagens, such as chemical carcinogens and ionizing radiations. Among the parameters applied in genetic toxicology, chromosome studies are an important laboratorial diagnostic procedure and these tests are increasingly used to identify mutagenic and antimutagenic agents in different cells and tissues. Chromosome aberrations in lymphocytes and bone marrow cells are thought to represent a surrogate endpoint for more specific chromosome alterations in target tissues of carcinogenesis. The cytogenetic characterization and classification of different types of chromosome aberrations have an important role in human genetics and many cancers are associated with specific types of aberrations. Investigation of the associations between free-radicalinduced DNA damage and the antioxidant properties of dietary oils is a field of great interest for elucidating mechanisms of mutagenesis and carcinogenesis. The absence of any mutagenic effects of olive and virgin olive oils in rodent bone marrow cells and human lymphocytes are of great importance, since these dietary oils are widely used by humans. Acute and subacute treatments with the chosen dietary oils reduced the chromosomal damage induced by the chemotherapeutic drug cisplatin in rats. The protective effects of olive and virgin oils on DNA damage measured as single-, double-strand breaks or chromosome aberrations are frequently associated with antimutagenicity. The exact mechanisms of the antimutagenic effects of olive oil are not well understood. The most likely mechanism of olive oil activity in these antimutagenicity studies is that this dietary oil acts as an antioxidant. More investigations are required to more clearly understand the activity of olive oil and its antioxidant constituents in foods, so that the beneficial and protective properties of olive and virgin olive oils on human health can be characterized before suggesting its use in future chemoprevention trials.
ACKNOWLEDGMENTS The authors are grateful to Dr. Cristina Márcia Wolf Evangelista at the Universidade de Ribeirão Preto – UNAERP, and their co-workers at the Faculdade de Ciências Farmacêuticas de Ribeirão Preto–USP, Mrs. Joana D’Arc C. Darin, MsC. Alexandre Aissa Ferro and Dr. Regislaine Valéria Burim, for their constant support and valuable technical assistance.
CHAPTER | 116 In vivo Cytogenetic Effects of Multiple Doses of Dietary Vegetable Oils: Position of Olive Oils
REFERENCES Amarowiczs, R., Naczk, M., Shahidi, F., 2000. Antioxidant activity of various fractions of non-tannin phenolics of canola hulls. J. Agr. Food Chem. 48, 2755–2759. Antunes, L.M.G., Takahashi, C.S., 1999. Olive oil protects against chromosomal aberrations induced by doxorubicin in Wistar rat bone marrow cells. Genet. Mol. Biol. 22, 225–227. Antunes, L.M.G., Bianchi, M.L.P., 2008. Lycopene and chromosome aberrations. In: Preedy, V.P., Watson, R.R. (eds), Lycopene: nutritional, medicinal and therapeutic properties. Science Publishers, New Hampshire, USA, pp. 183–200. Benavente-García, O., Castillo, J., Lorente, J., Alcaraz, M., 2002. Radioprotective effects in vivo of phenolics extracted from Olea europaea L. leaves against X-ray-induced chromosomal damage: comparative study versus several flavonoids and sulfur-containing compounds. J. Med. Food 5, 125–135. Bonassi, S., Ugolini, D., Kirsch-Volders, M., Stromberg, U., Vermeulen, R., Tucker, J.D., 2005. Human population studies with cytogenetic biomarkers: review of the literature and future prospectives. Environ. Mol. Mutagen. 45, 258–270. Brunner, L.J., Bai, S., 2000. Effect of dietary oil intake on hepatic cytochrome P450 activity in rat. Pharm. Sci. 89, 1022–1027. Elmadfa, I., Park, E., 1999. Impact of diets with corn oil or olive/sunflower oils on DNA damage in healthy young men. Eur. J. Nutr. 38, 286–292. El-Nahas, S.M., Mattar, F.E., Mohamed, A.A., 1993. Radioprotective effect of vitamins C and E. Mutat. Res. 301, 143–147. Evangelista, C.M.W., Antunes, L.M.G., Francescato, H.D.C., Bianchi, M.L.P., 2004. Effects of the olive, extra virgin olive and canola oils on cisplatin-induced clastogenesis in Wistar rats. Food Chem. Toxicol. 42, 1291–1297. Evangelista, C.M.W., Antunes, L.M.G., Bianchi, M.L.P., 2006. In vivo cytogenetic effects of multiple doses of dietary vegetable oils. Genet. Mol. Biol. 29, 730–734. Visioli, F., Galli, C., 2002. Biological properties of olive oil phytochemicals. Crit. Rev. Food Sci. Nutr. 42, 209–221. Fenech, M., 2008. Genome health nutrigenomics and nutrigeneticsdiagnosis and nutritional treatment of genome damage on an individual basis. Food Chem. Toxicol. 46, 1365–1370. Fitó, M., de la Torre, R., Covas, M.I., 2007. Olive oil and oxidative stress. Mol. Nutr. Food Res. 51, 1215–1224. Heng, H.H., Bremer, S.W., Stevens, J., Ye, K.J., Miller, F., Liu, G., Ye, C.J., 2006. Cancer progression by non-clonal chromosome aberrations. J. Cell Biochem. 98, 1424–1435.
1077
Monti, S.M., Ritieni, A., Sacchi, R., Skog, K., Borgen, E., Fogliano, V., 2001. Characterization of phenolics compounds in virgin olive oil and their effect on the formation of carcinogenic/mutagenic heterocyclic amines in a model system. J. Agric. Food Chem. 49, 3969–3975. Nousis, L., Doulias, P.T., Aligiannis, N., Bazios, D., Agalias, A., Galaris, D., Mitakou, S., 2005. DNA protecting and genotoxic effects of olive oil related components in cells exposed to hydrogen peroxide. Free Rad. Res. 39, 787–795. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Spiegelhalder, B., Bartsch, H., 2000. The antioxidant/anticancer potential of phenolics compounds isolated from olive oil. Eur. J. Cancer 36, 1235–1247. Preston, R.J., Dean, B.J., Galloway, S., Holden, H., McFee, A.F., Shelby, M., 1987. Analysis of chromosome aberrations in bone marrow cells. Mutat. Res. 189, 157–165. Quiles, J.L., Ochoa, J.J., Ramirez-Tortosa, C., Battino, M., Huertas, J.R., Martin, Y., Mataix, J., 2004. Dietary fat type (virgin olive vs sunflower oils) affects age-related changes in DNA double-strandbreaks, antioxidant capacity and blood lipids in rats. Exp. Gerontol. 39, 1189–1198. Raj, A.S., Heddle, J.A., Newmark, H.L., Katz, M., 1983. Caffeic acid as an inhibitor of DMBA-induced chromosomal breakage in mice assessed by bone-marrow micronucleus test. Mutat. Res. 124, 247–253. Raj, A.S., Katz, M., 1984. Corn oil and its minor constituents as inhibitors of DMBA-induced chromosomal breaks in vivo. Mutat. Res. 136, 247–253. Rojas-Molina, M., Campos-Sánchez, J., Analla, M., Munõz-Serrano, A., Alonso-Moraga, A., 2005. Genotoxicity of vegetable cooking oils in the Drosophila wing spot test. Environ. Mol. Mutagen. 45, 90–95. Sendão, M.C., Behling, E.B., Santos, R.A., Antunes, L.M.G., Bianchi, M.L.P., 2006. Comparative effects of acute and subacute lycopene administration on chromosomal aberrations induced by cisplatin in male rats. Food Chem. Toxicol. 44, 1334–1339. Soobrattee, M.A., Neergheen, V.S., Luximon-Ramma, A., Aruoma, O.I., Bahorun, T., 2005. Phenolics as potential antioxidant therapeutic agents: Mechanism and actions. Mutat. Res. 579, 200–213. Tice, R.R., Hayashi, M., MacGregor, J.T., Anderson, D., Blakey, D.H., Holden, H.E., Kirsh-Volders, M., Oleson, F.B., Pacchierotti, F., Preston, R.J., 1994. Report from the working group on the in vivo mammalian bone marrow chromosomal aberrations test. Mutat. Res. 312, 305–312. Tuck, K.L., Hayball, P.J., 2002. Major phenolics compounds in olive oil: Metabolism and heath effects. J. Nutr. Biochem. 13, 636–644.
This page intentionally left blank
Chapter 117
Minor Polar Compounds in Olive Oil and NF-κB Translocation Sandra Brunelleschi1,2, Angela Amoruso1, Claudio Bardelli1, Annalisa Romani3, Francesca Ieri3 and Flavia Franconi4 1
Department of Medical Sciences, University of Piemonte Orientale ‘A. Avogadro’, Novara, Italy IRCAD, University of Piemonte Orientale ‘A. Avogadro’, Novara, Italy 3 Department of Pharmaceutical Sciences, University of Florence, Firenze, Italy 4 Department of Pharmacology, University of Sassari, Sassari, Italy
2
117.1 INTRODUCTION Diet is a cornerstone of disease prevention and epidemiological studies demonstrate that Mediterranean populations have a lower incidence of cardiovascular diseases and cancer (Kris-Etherton et al., 2001; Hu, 2003; Owen et al., 2004). The healthy effect of the Mediterranean diet has been attributed to the consumption of large amounts of fiber, fruits, vegetables and unsaturated fatty acid provided by olive oil. Olive oil, the main source of fat of the Mediterranean diet, consists primarily of triacylglycerols rich in the monounsaturated fatty acid, oleic acid, and also contains relatively high amounts of at least 30 phenolic compounds (Miles et al., 2005). The amount of these ‘non-nutrients’ or ‘minor components’ in the olive oil is variable, depending on several factors, such as olive cultivar, location, climate, degree of maturation, agronomic and technological aspects of production (Oliveras-Lopez et al., 2007). Phenols largely contribute to oil flavor and taste, the phenolic content representing a main parameter for the evaluation of olive oil quality (Oliveras-Lopez et al., 2007). Virgin olive oils contain phenyl alcohols (e.g., tyrosol and hydroxytyrosol), phenolic acids, flavonoids (e.g., luteolin and apigenin), more complex secoiridoid derivatives from oleuropein and, at least in some olive oils, lignans, such as (⫹)-pinoresinol and (⫹)-1-acetoxypinerol (Oliveras-Lopez et al., 2007). The concentration of olive oil phenolic compounds such as oleuropein and hydroxytyrosol is maximal (up to 800 mg kg⫺1) in the first pressed extra virgin olive oil (Visioli et al., 2000a) and is also relevant in olives (about 2 g 100 g⫺1 of dry weight) (Marsilio et al., 2001). Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
The cardio-protective role of olive oil has been highlighted in several studies (Ferro-Luzzi and Branca, 1995; Kris-Etherton et al., 2001; Hu, 2003; Trichopoulou et al., 2003; Chrysohoou et al., 2004; Covas, 2007) and beneficial health effects have been attributed, at least in part, to the presence of non-nutrient minor polar compounds (MPC) (Covas et al., 2006b; Franconi et al., 2006) (Table 117.1). As an example, phenolic compounds in virgin olive oil improve endothelial function in hypercholesterolemic patients (Ruano et al., 2005), inhibit low-density lipoprotein (LDL) oxidation in humans and animals, in vivo and in vitro (Visioli et al., 1995, 2000c; Fito et al., 2000; Covas et al., 2006b), hydroxytyrosol and oleuropein being also potent scavengers of several free radicals (Rietjens et al., 2007). After consumption of 25 mL extra virgin olive oil, plasma concentrations of hydroxytyrosol range from 50 to 160 nM (Miro Casas et al., 2003; Weinbrenner et al., 2004). Hydroxytyrosol affects oxidative stress as well as arachidonic acid mobilization and metabolism by macrophage RAW 264.7 cells (Moreno, 2003). In the same cell line, it also reduces iNOS (inducible nitric oxide synthase)
1079
TABLE 117.1 Principal classes of minor polar compounds (MPC). ● ●
● ●
Simple phenols (e.g., tyrosol, 5-hydroxytyrosol) Secoiridoids (e.g., elenolic acid, elenolic acid derivatives, oleuropein aglycones, deacetoxy-oleuropein aglycone, oleocanthal, secoiridoid derivatives) Lignan derivatives (e.g., acetoxypinoresinol) Flavones (e.g., luteolin)
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1080
and COX-2 (cyclo-oxygenase 2) gene expression, mainly by preventing the activation of NF-κB (nuclear factor-kappa B), STAT-1α (signal transducer and activator of transcription1α) and IRF-1 (interferon regulatory factor-1) (Maiuri et al., 2005). At nutritionally relevant concentrations, hydroxytyrosol reduces monocyte adhesion to human endothelial cells, as well as the expression of adhesion molecules such as VCAM-1 (vascular cell adhesion molecule-1), an effect that seems to be mediated by NF-κB and activator protein-1 (AP-1) (Carluccio et al., 2003). The redox-sensitive transcription factor NF-κB can be activated by various stimuli, including reactive oxygen species, oxidized LDL, hypoxia/anoxia, cytokines, bacterial and viral products, and is largely recognized as a therapeutic target in inflammation and atherosclerosis (Kutuk and Basaga, 2003; de Winther et al., 2005). Ex vivo observations in healthy volunteers demonstrate that, at variance from butter- and walnut-rich meals, consumption of an olive oil-rich meal does not induce postprandial activation of the NF-κB pathway in monocytes (Bellido et al., 2004). We recently demonstrated that an extra virgin olive oil extract, particularly rich in MPC (MPC olive oil extract), efficiently inhibits NF-κB translocation in monocytes/macrophages from healthy volunteers (Brunelleschi et al., 2007). By using monocytes/macrophages from healthy volunteers, in the present work, we have carefully evaluated the ability of increasing concentrations of MPC olive oil extract to modulate NF-κB activity and cytokine release.
SECTION | II Cells and Cellular Effects
secoiridoids (elenolic acid, elenolic acid derivatives, oleuropein aglycones, deacetoxy-oleuropein aglycone, oleocanthal and secoiridoid derivatives), lignan derivatives (acetoxypinoresinol) and flavones (luteolin) (Table 117.2 and Figure 117.1). Oleocanthal, deacetoxy-oleuropein aglycone and tyrosol are the main components (all ⬎6 mM), followed by secoiridoid derivatives (5.98 mM), elenolic acid (4.94 mM) and hydroxytyrosol (4.41 mM); total polyphenols are about 40 mM
TABLE 117.2 Composition of MPC olive oil extract (in mM). Class
Compound
mM
Simple phenols
Tyrosol 5-hydroxytyrosol
6.03 4.41
Secoiridoids
Elenolic acid Elenolic acid derivatives Oleuropein aglycones Deacetoxy-oleuropein aglycone Oleocanthal Secoiridoid derivatives
4.94 2.62 1.29 6.04 6.10 5.98
Lignan derivatives
Acetoxypinoresinol, mainly
2.65
Flavones
Luteolin
0.04
Total polyphenols in this MPC olive oil extract were 40.09 mM. Data reported are the mean of 3 determinations, each performed in triplicate. S.E.M. was in the range 1–3%.
117.2 FEATURES OF MINOR POLAR COMPOUND (MPC) OLIVE OIL EXTRACT
2400 2200 2000 1800 Composition mg L–1
We used an extra virgin olive oil kindly supplied by a Tuscan enterprise; the preparation, characterization and quantification of MPC olive oil extract were carried out as previously reported (Brunelleschi et al., 2007). The MPC were identified based on their retention times and spectroscopic and spectrometric data, using 5-hydroxytyrosol, tyrosol, luteolin and oleuropein as reference compounds. Lignan was identified and analyzed as described (Mulinacci et al., 2006); oleocanthal was identified according to Beauchamp et al. (2005). The single minor compounds were quantified with HPLC-DAD using a four-point regression curve constructed with the available standards; MPC concentrations were calculated after applying corrections for changes in molecular weight. Given the molecular weight of each compound (PMx), their actual concentration was obtained by applying a multiplication factor of PMx/ PMy, where PMy is the molecular weight of the specific reference compound. As reported in Table 117.2 and Figure 117.1, the MPC identified and quantified in this olive oil extract belong to four classes: simple phenols (tyrosol and 5-hydroxytyrosol),
2600
1600 1400 1200 1000 800 600 400 200 0 1
2
3
4
5
6
7
8
9
10
FIGURE 117.1 Composition of MPC olive oil extract (in mg L⫺1). Total polyphenols in this MPC olive oil extract were 11 015 mg L⫺1. 䊏 Simple phenols (6 ⫽ 5-hydroxytyrosol; 8 ⫽ tyrosol); Secoiridoids (1 ⫽ secoiridoid derivatives; 2 ⫽ deacetoxy-oleuropein aglycone; 3 ⫽ oleocanthal; 4 ⫽ elenolic acid; 7 ⫽ elenolic acid derivatives; 9 ⫽ oleuropein aglycone); Lignan derivatives (5 ⫽ acetoxypinoresinol, mainly); Flavones (10 ⫽ luteolin). Data reported are the mean of 3 determinations, each performed in triplicate. S.E. was in the range 1–3%.
1081
CHAPTER | 117 Minor Polar Compounds in Olive Oil and NF-κB Translocation
(Table 117.2). As previously demonstrated, oleocanthal has a chemical structure similar to ibuprofen and inhibits prostaglandin biosynthesis pathway (Beauchamp et al., 2005).
Monocytes
p65 Ab
–
+
–
–
–
–
–
p50 Ab
–
–
+
+
+
+
+
1
2
3
4
5
6
7
p65 Ab
–
+
–
–
–
–
–
p50 Ab
–
–
+
+
+
+
+
1
2
3
4
5
6
7
p50/p65 supershift p50/p50 supershift
117.3 MPC-OLIVE OIL EXTRACT INHIBITS NF-κB ACTIVATION IN HUMAN MONOCYTES/MACROPHAGES This MPC olive oil extract was used for experiments in human cells from healthy non-smoker volunteers. Purified monocyte populations were obtained by adhesion (90 min, 37°C, 5% CO2); cell viability (trypan blue dye exclusion) was usually ⬎98% (Brunelleschi et al., 2007). Experiments were initiated on the day of blood collection; all manipulations were carried out under endotoxin-free conditions. Monocyte-derived macrophages (MDM) were prepared from monocytes by cell culture (8–10 days, 37°C, 5% CO2) in RPMI 1640 medium containing 20% fetal bovine serum (FBS) and antibiotics, as described (Brunelleschi et al., 2007). By this procedure, monocytes acquire a morphological macrophage-like profile, which is accompanied by an increase in CD68⫹ cells and a decrease in CD14⫹ cells, as compared to monocytes (Amoruso et al., 2008). To evaluate NF-κB activation, cells were challenged with MPC olive oil extract (10 nM–10 μM) for 3 hours and then stimulated by phorbol 12-myristate 13-acetate (PMA) 10⫺6 M for 1 hour. As a positive control, we used the PPAR-γ (peroxisome proliferator-activated receptor-gamma) agonist ciglitazone (50 μM), which is known to inhibit NF-κB activation (Zingarelli et al., 2003). The activation of NF-κB was evaluated by measuring the nuclear migration (by electrophoretic mobility shift assay; EMSA) as well as the nuclear content of p50 and p65 subunits (by ELISA). For EMSA assays, nuclear extracts (5 μg) were incubated with 2 μg poly (dI-dC) and the γ [32P]ATP-labeled oligonucleotide probe in binding buffer for 30 min at room temperature; the nucleotide– protein complex was separated on a 5% polyacrylamide gel and radioactive bands were detected by autoradiography (Brunelleschi et al., 2007). As shown in Figure 117.2, NF-κB is constitutively low activated in both monocytes and MDM from healthy volunteers; PMA, at 10⫺6 M, potently stimulates NF-κB nuclear translocation (Figure 117.2). Supershift assays also reveal that, in human monocytes/ macrophages, NF-κB is present as p50/p65 heterodimer or p50/p50 homodimer (Figure 117.2). At 10 μM, MPC olive oil extract effectively inhibits NF-κB nuclear translocation in unstimulated and PMA-stimulated cells; the PPAR-γ agonist ciglitazone, a known inhibitor of NF-κB activation, has been used as positive control. For the purpose of clarity and brevity, Figure 117.2 deals with supershifts only, except for PMA; please note that the p50 antibody reveals the p50/p65 heterodimer too.
NF-κB
A
MDM
p50/p65 supershift p50/p50 supershift
NF-κB
B
FIGURE 117.2 NF-κB activation, p50 and p65 supershifts in human monocytes and MDM. In (A): monocytes, in (B): MDM. For brevity, only supershifts are demonstrated in all cases, except PMA. Lane 1: PMA 10⫺6 M, total effect; Lane 2: PMA, supershift with p65 antibody; Lane 3: PMA, supershift with p50 antibody (which also reveals the p50/p65 heterodimer); Lane 4: control, unstimulated cells, supershift; Lane 5: ciglitazone 50 μM, supershift; Lane 6: MPC olive oil extract 10 μM, supershift; Lane 7: PMA ⫹ MPC-olive oil extract 10 μM, supershift. Reprinted from Brunelleschi et al., 2007.
To get a better quantitative evaluation, we assessed the translocation of NF-κB subunits by means of commercially available ELISA kits for p50 and p65 subunits. Nuclear and cytosolic fractions from human monocytes and MDM (about 5 ⫻ 106 cells) were prepared as described (Brunelleschi et al., 2007). Both fractions were evaluated for the presence of p50 and p65/RelA subunits using Trans AM™ NF-κB p50 and p65 Chemi Transcription Factor Assay kits (Active Motif Europe, Belgium), according to the manufacturer’s instructions. These assay kits specifically detected bound NF-κB p65 or p50 subunits in human extracts; the amount of translocated p50 and p65 subunits was evaluated as the nuclear/cytosol (N/C) ratio (Brunelleschi et al., 2007). A low basal activation of NF-κB is detected in both unstimulated monocytes and MDM, and, as previously indicated (Bardelli et al., 2005), the p50 subunit is the most abundant and efficiently translocated in both cell types (Figure 117.3). At 10⫺6 M, PMA potently stimulates p50 (Figure 117.3A) and p65 (Figure 117.3B) nuclear translocation, its effect being significantly decreased in the presence of the PPAR-γ agonist ciglitazone, which has been used as a positive control (Figure 117.3). In the concentration range
1082
SECTION | II Cells and Cellular Effects
p50 Subunit MDM
Monocyte
A
6 N/C ratio
N/C ratio
6
4
2
4
2
0
0 C −9
−7 −5 PMA −9
+ MPC-OOE log (M)
−7
−5 Cigl
C −9
+ MPC-OOE log (M)
−7 −5 PMA −9
+ MPC-OOE log (M)
−7
−5 Cigl
+ MPC-OOE log (M)
p65 Subunit Monocyte
MDM
12
12
0.8
0.8
N/C ratio
N/C ratio
B
0.4
0
0.4
0 C −9
−7 −5 PMA −9
+ MPC-OOE log (M)
−7
−5 Cigl
+ MPC-OOE log (M)
C −9
−7 −5 PMA −9
+ MPC-OOE log (M)
−7
−5 Cigl
+ MPC-OOE log (M)
FIGURE 117.3 MPC-olive oil extract inhibits NF-κB translocation in human monocytes and MDM. MPC-olive oil extract (MPC-OOE) inhibits, in a concentration-dependent manner, the translocation of the activated p50 subunit (A) and p65 subunit (B) in cells stimulated by PMA 10⫺6 M, but has minor effects in unstimulated, control (C) cells. The effects of ciglitazone (Cigl: 50 μM) are shown for comparison. Results are expressed as nuclear/cytosol (N/C) ratio and are mean ⫾ S.E.M. of five experiments. °p ⬍ 0.05, °°p ⬍ 0.01, °°°p ⬍ 0.001 vs control cells; *p ⬍ 0.05, **p ⬍ 0.01, ***p ⬍ 0.001 vs PMA-stimulated cells. Reprinted from Brunelleschi et al., 2007.
1 nM–10 μM, MPC olive oil extract inhibits the nuclear translocation of the NF-κB p50 subunit, being about as effective as ciglitazone; at the highest 10 μM concentration, PMA-induced p50 translocation is inhibited by about 70% in both monocytes and MDM (Figure 117.3A). Interestingly, at 10 μM, MPC olive oil extract also significantly decreases p50 translocation in unstimulated monocytes (p ⬍ 0.05 vs control; Figure 117.3A). MPC olive oil extract has no effect on p65 translocation in unstimulated cells, but it inhibits the PMA-induced translocation in a concentration-dependent manner, being even more effective than ciglitazone, at the maximal 10 μM concentration (Figure 117.3B).
117.4 MPC OLIVE OIL EXTRACT INHIBITS CYTOKINE SECRETION IN HUMAN MONOCYTES To get more insight into the beneficial effects of extra virgin olive oil, we also evaluated MPC olive oil extract’s ability to reduce TNF-α secretion from PMA-challenged monocytes. In these experiments, monocytes (1 ⫻ 106 cells) were challenged
with MPC olive oil extract (10 nM–10 μM) for 30 min and then stimulated by PMA 10⫺7 M for 2 hours or 24 hours; the PPAR-γ agonist ciglitazone (50 μM) or the NF-κB inhibitor Bay 11–7082 (10 μM) were used as positive controls. The 24-hour stimulation time was chosen to ensure maximal cytokine release, as observed previously (Bardelli et al., 2005), and the amount of TNF-α in monocyte supernatants was estimated by ELISA (Pelikine Compact™ human ELISA kit), following the manufacturer’s instructions (CLB/Sanquin, Netherlands). In these experiments, monocytes challenged with PMA 10⫺7 M released 160 ⫾ 12 pg mL⫺1 (n ⫽ 4) of TNF-α after 2-hour stimulation and 246 ⫾ 15 pg mL⫺1 (n ⫽ 4) after 24-hour stimulation (p ⬍ 0.01 vs 2 hours), so confirming previous observations (Bardelli et al., 2005). As depicted in Figure 117.4, MPC olive oil extract (10 nM–10 μM) inhibited, in a concentration-dependent manner, PMA-induced TNF-α release: maximal inhibition was observed at 10 μM concentration and was higher following the short 2-hour stimulation period with PMA (about 95% inhibition; p ⬍ 0.05 vs 24-hour stimulation period). The calculated IC50 values for MPC olive oil extract were
1083
CHAPTER | 117 Minor Polar Compounds in Olive Oil and NF-κB Translocation
100
∗
% inhibition
80
∗
∗
60 ∗ 40 20 0
−8
−7 −6 log [MPC-OOE] (M)
−5
Cigl
Bay
FIGURE 117.4 MPC-olive oil extract inhibits PMA-induced TNF-α release in human monocytes. Monocytes were treated for 30 min with MPC-olive oil extract (MPC-OOE; 10 nM–10 μM), ciglitazone (Cigl) 50 μM, or the NF-κB inhibitor Bay 11-7082 (Bay) 10 μM, and then stimulated by PMA 10⫺7 M for 2 hours (䊏) or 24 hours (䊐). Means ⫾ S.E.M.; n ⫽ 4. *p ⬍ 0.05.
in the same range: 0.2 μM (2 hours) and 0.38 μM (24 hours). Interestingly, MPC olive oil extract was about as effective as ciglitazone and Bay 11-7082.
TABLE 117.3 Factors known to activate NF-κB. ● ●
117.5 AN ANALYSIS OF THE ROLE OF MPC IN CARDIOVASCULAR DISEASES
● ● ● ●
The lower incidence of cardiovascular diseases in the Mediterranean area has led to the suggestion that high consumption of olive oil protects against the development of atherosclerosis (Ferro-Luzzi and Branca, 1995; Kris-Etherton et al., 2001; Hu, 2003; Trichopoulou et al., 2003; Chrysohoou et al., 2004; Covas, 2007). Recently, interest has been focused on the phenolic compounds that are present in olive oil, as they have strong antioxidant properties and can protect LDL from oxidation (Visioli et al., 1995, 2000c; Fito et al., 2000). This latter property has been recognized as a major mechanism of cardio-protection, since LDL oxidation represents an initial step to atherosclerotic plaque formation. However, the development of atherosclerotic plaque also involves an inflammatory component, infiltration of the vessel walls by monocytes representing a pivotal feature (Ross, 1999). As known, monocytes secrete inflammatory cytokines that play a relevant role in the disease progression (Frostegard et al., 1999; Ross, 1999; Plutzky, 2001). However, only little information is available about the ability of olive oil phenolics to affect cytokine secretion. In LPS-stimulated human whole blood, oleuropein glycoside (but not tyrosol) inhibited IL-1β secretion in a concentration-dependent manner (Miles et al., 2005). Accordingly, in human endothelial cells, oleuropein aglycone was more potent than tyrosol, hydroxytyrosol and oleuropein glycoside, in reducing adhesion molecule expression (Carluccio et al., 2003). We now demonstrate that MPC olive oil extract potently inhibits TNF-α release by PMA-challenged human monocytes, the IC50 value being 0.2 μM. The MPC olive oil extract contains a greater amount of total polyphenols than
● ●
Inflammatory cytokines Oxidized lipids Bacterial and viral products Hypoxia/anoxia Oxidants and reactive oxygen species Activators of protein kinase C Substance P Macrophage stimulating protein
a previously evaluated Tuscan olive oil extract (Franconi et al., 2006), is abundant in MPC (and especially in the anti-inflammatory component oleocanthal), and is deprived in other bio-active compounds such as fatty acids, tocopherol and other lipophilic components. The anti-inflammatory potential of MPC olive oil extract is strongly supported by its ability to inhibit, at nutritional concentrations, PMA-induced NF-κB activation in monocytes and MDM from healthy volunteers, thus extending the idea of the cardio-protective effect of olive oil-enriched diets (Bellido et al., 2004; Covas et al., 2006a; Perez-Martinez et al., 2007). In fact, MPC olive oil extract potently inhibits NF-κB nuclear translocation in monocytes/macrophages, being as active as the PPAR-γ agonist ciglitazone is. As known, NF-κB is a redox-sensitive transcription factor that comprises five members, RelA (p65), NF-κB1 (p50 and p105), NF-κB2 (p52 and p100), c-Rel and RelB, all sharing the Rel homology domain, which allows their dimerization and translocation to the nucleus. NF-κB dimers are bound to inhibitory proteins of the IκB family, which retain NF-κB in the cytoplasm (Li and Verma, 2002; de Winther et al., 2005). Different stimuli, including inflammatory cytokines, oxidized lipids, bacterial and viral products, hypoxia/anoxia and reactive oxygen species (Table 117.3), activate NFκB through the phosphorylation of IκB and its subsequent
1084
SECTION | II Cells and Cellular Effects
Activation signals Inflammatory proteins
I κB kinase(s) IκB-α p50 p65 P
P I κ B- α
Proteasome
-Ub-Ub-Ub
p50 p65
IκB-α Degradation
Cell membrane
p50 p65
Translocation
p50 p65
Cytoplasm mRNA
Nucleus
κB inflammatory gene FIGURE 117.5 Schematic representation of the events leading to the activation of the transcription factor NF-κB. See text for details.
release from the complex; this results in the translocation of NF-κB subunits into the nucleus, where they bind to target genes involved in the inflammatory and immune response and induce their transcription (Li and Verma, 2002; de Winther et al., 2005) (Figure 117.5). Although different homo- and heterodimeric forms of this factor have been described, NF-κB is usually composed of the p50/p65 heterodimer, p50/p50 homodimers being suggested as transcriptional repressors (Li and Verma, 2002). Since the optimal binding sequences for p65 and p50 are similar but not identical (Perkins et al., 1992), it is possible that NF-κB-induced responses might be fine-tuned by minor sequence variations that alter the relative binding of subunits to different regulatory elements. Therefore, it is important to elucidate the role of individual NF-κB subunits and to measure their relative amounts. In keeping with previous observations (Bardelli et al., 2005; Brunelleschi et al., 2007), we confirm that, in human monocytes and MDM, p50 protein is about six-fold more abundant than p65 protein, and also demonstrates that MPC olive oil extract effectively inhibits its translocation. Interestingly, the p50 subunit has been shown to play a crucial role in atherosclerosis (Cha-Molstad et al., 2000; Kanters et al., 2004; Frantz et al., 2006). In fact, over-expression of p50 protein has been shown to induce the transcription of C-reactive protein (a major marker of cardiovascular inflammation), whereas p65 over-expression inhibits it (Cha-Molstad et al., 2000). Kanters et al. (2004) observed that p50-deficient mice have a lower rate
of atherosclerosis than control mice, p50-deficient macrophages also presenting an altered cytokine secretion and a reduced uptake of oxidized LDL. Recently, mice with a targeted deletion of the p50 NF-κB subunit have been demonstrated to undergo a reduced early mortality after myocardial infarction (as compared to wild-type), which is associated with lower collagen content and matrix metalloproteinase-9 expression (Frantz et al., 2006). In our experiments, MPC olive oil extract inhibits TNF-α release and NF-κB activation at nutritional concentrations. As previously reported (Visioli et al., 2000b), a Mediterranean diet rich in olive oil supplies 10–20 mg of phenols per day and ensures MPC plasma levels of about 0.6 μM, well within the in vitro concentrations we used. Interestingly, at the highest concentration evaluated, MPC olive oil extract also significantly reduces p50 translocation in unstimulated monocytes, in good agreement with recent ex vivo observations (Perez-Martinez et al., 2007). In healthy volunteers submitted to three diet intervention periods of 4 weeks’ duration, Perez-Martinez et al. (2007) clearly evidenced that 1 month’s consumption of a Mediterranean diet enriched in olive oil reduces NF-κB activation in monocytes and VCAM-1 plasma concentrations. Thus, we suggest that the ability of MPC olive oil extract to inhibit p50 translocation can exert a relevant antiatherosclerotic role and might, therefore, largely contribute to the cardio-protective effect of virgin olive oil. We also suggest that inhibition of NF-κB activation and TNF-α release (as documented in our in vitro experiments
CHAPTER | 117 Minor Polar Compounds in Olive Oil and NF-κB Translocation
with MPC olive oil extract) might represent useful targets for reducing the risk of coronary heart diseases. On this regard, it is worth reminding that the transcription factor NF-κB regulates the expression of many pro-inflammatory genes, including that of TNF-α, and in turn TNF-α is a potent inducer of NF-κB activation (Li and Verma, 2002). As known, the regulation of TNF-α production is largely NF-κB-dependent, although evidence exists that TNF-α and other cytokines can also be induced through NF-κBindependent pathways (Andreakos et al., 2004). In this study, TNF-α release is almost completely abolished by the NF-κB inhibitor Bay 11-7082, so indicating that, in our model, TNF-α production is largely NF-κB-dependent. In conclusion, we demonstrate that MPC olive oil extract reduces NF-κB activation at the concentration range required for inhibition of TNF-α production.
SUMMARY POINTS ●
●
●
●
●
●
●
●
●
Extra virgin olive oils contain different amounts of minor polar compounds (MPC). MPC modulate the responsiveness of human monocytes/macrophages. Evaluation of an extra virgin olive oil extract, particularly rich in MPC (MPC olive oil extract). MPC olive oil extract acts dose-dependently (1 nM–10 μM). MPC olive oil extract inhibits NF-κB translocation in human monocytes/macrophages. MPC olive oil extract is particularly effective on the p50 subunit of NF-κB. MPC olive oil extract inhibits TNF-α release in human monocytes/macrophages. Inhibitory effects of MPC olive oil occur at nutritional concentrations. MPC contribute to beneficial effects of olive oil and could represent relevant molecules for atherosclerosis prevention.
REFERENCES Amoruso, A., Bardelli, C., Gunella, G., Ribichini, F., Brunelleschi, S., 2008. A novel activity for substance P: stimulation of peroxisome proliferator-activated receptor-gamma protein expression in human monocytes and macrophages. Br. J. Pharmacol. 154, 144–152. Andreakos, E., Sacre, S.M., Smith, C., Lundberg, A., Kiriakidis, S., Stonehause, T., Monaco, C., Feldmann, M., Foxwell, B.M., 2004. Distinct pathways of LPS-induced NF-κB activation and cytokine production in human myeloid and nonmyeloid cells defined by selective utilization of MyD88 and Mal/TIRAP. Blood 103, 2229–2237. Bardelli, C., Gunella, G., Varsaldi, F., Balbo, P., Del Boca, E., Seren Bernardone, I., Amoruso, A., Brunelleschi, S., 2005. Expression of functional NK1 receptors in human alveolar macrophages: superoxide anion production, cytokine release and involvement of NF-κB pathway. Br. J. Pharmacol. 145, 385–396.
1085
Beauchamp, G.K., Keast, R.S.J., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A.S., 2005. Ibuprofen-like activity in extra-virgin olive oil. Nature 437, 45–46. Bellido, C., Lopez-Miranda, J., Blanco-Colio, L.M., Perez-Martinez, P., Muriana, F.J., Martin-Ventura, J.L., Marin, C., Gomez, P., Fuentes, F., Egido, J., Perez-Jimenez, F., 2004. Butter and walnuts, but not olive oil, elicit postprandial activation of nuclear transcription factor κB in peripheral blood mononuclear cells from healthy men. Am. J. Clin. Nutr. 80, 1487–1491. Brunelleschi, S., Bardelli, C., Amoruso, A., Gunella, G., Ieri, F., Romani, A., Malorni, W., Franconi, F., 2007. Minor polar compounds extra-virgin olive oil extract (MPC-OOE) inhibits NF-κB translocation in human monocyte/macrophages. Pharmacol. Res. 56, 542–549. Carluccio, M.A., Siculella, L., Ancora, M.A., Massaro, M., Scoditti, E., Storelli, C., Visioli, F., Distante, A., De Caterina, R., 2003. Olive oil and red wine antioxidant polyphenols inhibit endothelial activation: anti-atherogenic properties of Mediterranean diet phytochemicals. Arterioscler. Thromb. Vasc. Biol. 23, 622–629. Cha-Molstad, H., Agrawal, A., Zhang, D., Samols, D., Kushner, I., 2000. The Rel family member p50 mediates cytokine-induced C-reactive protein expression by a novel mechanism. J. Immun. 165, 4592–4597. Chrysohoou, C., Panagiotakos, D.B., Pitsavos, C., Das, U.N., Stefanadis, C., 2004. Adherence to the Mediterranean diet attenuates inflammation and coagulation process in healthy adults: the ATTICA Study. J. Am. Coll. Cardiol. 44, 152–158. Covas, M.I., 2007. Olive oil and the cardiovascular system. Pharmacol. Res. 55, 175–186. Covas, M.I., Nyyssonen, K., Poulsen, H.E., Kaikkonen, J., Zunft, H.J.F., Kiesewetter, H., Gaddi, A., de la Torre, R., Mursu, J., Baumler, H., Nascetti, S., Salonen, J.T., Fito, M., Virtanen, J., Marrugat, J., 2006a. The effects of polyphenols in olive oil on heart disease risk factors. A randomized trial. Ann. Int. Med. 145, 333–341. Covas, M.I., Ruiz-Gutiérrez, V., de la Torre, R., Kafatos, A., LamuelaRaventos, R.M., Osada, J., Owen, R.W., Visioli, F., 2006b. Minor components of olive oil: evidence to date of health benefits in humans. Nutr. Rev. 64, S20–S30. de Winther, M.P.J., Kanters, E., Kraal, G., Hofker, M.H., 2005. Nuclear Factor κB signaling in atherogenesis. Arterioscler. Thromb. Vasc. Biol. 25, 904–914. Ferro-Luzzi, A., Branca, F., 1995. Mediterranean diet, Italian-style: prototype of a healthy diet. Am. J. Clin. Nutr. 61, 1338S–1345S. Fito, M., Covas, M.I., Lamuela-Raventos, R.M., Vila, J., Torrents, L., de la Torre, C., Marrugat, J., 2000. Protective effect of olive oil and its phenolic compounds against low density lipoprotein oxidation. Lipids 35, 633–638. Franconi, F., Coinu, R., Carta, S., Urgeghe, P.P., Ieri, F., Mulinacci, N., Romani, A., 2006. Antioxidant effect of two virgin olive oils depends on the concentration and composition of minor polar compounds. J. Agric. Food Chem. 54, 3121–3125. Frantz, S., Hu, K., Bayer, B., Gerondakis, S., Strotmann, J., Adamek, A., Ertl, G., Bauersachs, J., 2006. Absence of NF-κB subunit p50 improves heart failure after myocardial infarction. FASEB J. 20, E1309–E1314. Frostegard, J., Ulfgren, A.K., Nyberg, P., Hedin, U., Swedenborg, J., Andersson, U., Hansson, G.K., 1999. Cytokine expression in advanced human atherosclerotic plaques: dominance of pro-inflammatory (Th1) and macrophage-stimulating cytokines. Atherosclerosis 145, 33–43. Hu, F.B., 2003. The Mediterranean diet and mortality—olive oil and beyond. N. Engl. J. Med. 348, 2595–2596.
1086
Kanters, E., Gijbels, M.J.J., van der Made, I., Vergouwe, M.N., Heeringa, P., Kraal, G., Hofker, M.H., de Winther, M.P.J., 2004. Hematopoietic NF-κB1 deficiency results in small atherosclerotic lesions with an inflammatory phenotype. Blood 103, 934–940. Kris-Etherton, P., Eckel, R.H., Howard, B.V., St Jeor, S., Bazzarre, T.L., 2001. AHA Science Advisory: Lyon Diet Heart Study. Benefits of a Mediterranean-style diet. NCEP/AHA. Step I. Dietary pattern on cardiovascular diseases. Circulation 103, 1823–1825. Kutuk, O., Basaga, H., 2003. Inflammation meets oxidation: NF-κB as a mediator of initial lesion development in atherosclerosis. Trends Mol. Med. 9, 549–557. Li, Q., Verma, I.M., 2002. NF-κB regulation in the immune system. Nat. Rev. Immun. 2, 725–734. Maiuri, M.C., Di Stefano, D., Di Meglio, P., Irace, C., Savarese, M., Sacchi, R., Cinelli, M.P., Carnuccio, R., 2005. Hydroxytyrosol, a phenolic compound from virgin olive oil, prevents macrophage activation. Naunyn-Schmied. Arch. Pharmacol. 371, 457–465. Marsilio, M., Campestre, C., Lanza, B., 2001. Phenolic compounds change during California-style ripe olive processing. Food Chem 74, 55–60. Miles, E.A., Zoubouli, P., Calder, P.C., 2005. Differential anti-inflammatory effects of phenolic compounds from extra virgin olive oil identified in human whole blood cultures. Nutrition 21, 389–394. Miro Casas, E., Covas, M.I., Farre, M., Fito, M., Ortuno, J., Weinbrenner, T., Roset, P., de la Torre, R., 2003. Hydroxytyrosol disposition in humans. Clin. Chem. 49, 945–952. Moreno, J.J., 2003. Effect of olive oil minor components on oxidative stress and arachidonic acid mobilization and metabolism by macrophages RAW 264.7. Free Radic. Biol. Med. 35, 1073–1081. Mulinacci, N., Giaccherini, C., Ieri, F., Romani, A., Vincieri, F.F., 2006. Evaluation of lignans and free and linked hydroxy-tyrosol and tyrosol in extra virgin olive oil after hydrolysis processes. J. Sci. Food Agric. 86, 757–764. Oliveras-Lopez, M.J., Innocenti, M., Giaccherini, C., Ieri, F., Romani, A., Mulinacci, N., 2007. Study of the phenolic composition of Spanish and Italian monocultivar extra virgin olive oils: Distribution of lignans, secoiridoidic, simple phenols and flavonoids. Talanta 73, 726–732. Owen, R.W., Haubner, R., Wuertele, G., Hull, E., Spiegelhalder, B., Bartsch, H., 2004. Olives and olive oil in cancer prevention. Eur. J. Cancer Prev. 13, 319–326. Perez-Martinez, P., Lopez-da, J., Blanco-Colio, L., Bellido, C., Jimenez, Y., Moreno, J.A., Delgado-Lista, J., Egido, J., Perez-Jimenez, F., 2007. The chronic intake of a Mediterranean diet enriched in virgin olive oil,
SECTION | II Cells and Cellular Effects
decreases nuclear transcription factor κB activation in peripheral blood mononuclear cells from healthy men. Atherosclerosis 194, e141–e146. Perkins, N.D., Schmidt, R.M., Duckett, C.S., Leung, K., Rice, N.R., Nabel, G.J., 1992. Distinct combinations of NF-kB subunits determine the specificity of transcriptional activation. Proc. Natl. Acad. Sci. USA 89, 1529–1533. Plutzky, J., 2001. Inflammatory pathways in atherosclerosis and acute coronary syndromes. Am. J. Cardiol. 88, 10K–15K. Rietjens, S.J., Bast, A., Haenen, G.R., 2007. New insights into controversies on the antioxidant potential of the olive oil antioxidant hydroxytyrosol. J. Agric. Food Chem. 55, 7609–7614. Ross, R., 1999. Atherosclerosis—an inflammatory disease. New Engl. J. Med. 340, 115–126. Ruano, J., Lopez-Miranda, J., Fuentes, F., Moreno, J.A., Bellido, C., Perez-Martinez, P., Lozano, A., Gomez, P., Jimenez, Y., PerezJimenez, F., 2005. Phenolic content of virgin olive oil improves ischemic reactive hyperemia in hypercholesterolemic patients. J. Am. Coll. Cardiol. 46, 1864–1868. Trichopoulou, A., Costacou, T., Bamia, C., Trichopoulos, D., 2003. Adherence to a Mediterranean diet and survival in a Greek population. N. Engl. J. Med. 348, 2599–2608. Visioli, F., Bellomo, G., Montedoro, G., Galli, C., 1995. Low density lipoprotein oxidation is inhibited in vitro by olive oil constituents. Atherosclerosis 117, 25–32. Visioli, F., Borsani, L., Galli, C., 2000a. Diet and prevention of coronary heart disease: the potential role of phytochemicals. Cardiovasc. Res. 47, 419–425. Visioli, F., Galli, C., Bornet, F., Mattei, A., Patelli, R., Galli, G., Caruso, D., 2000b. Olive oil phenolics are dose-dependently absorbed in humans. FEBS Lett. 468, 159–160. Visioli, F., Galli, C., Plasmati, E., Viappiani, S., Hernandez, A., Colombo, C., Sala, A., 2000c. Olive oil hydroxytyrosol prevents passive smokinginduced oxidative stress. Circulation 102, 2169–2171. Weinbrenner, T., Fito, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Albaladejo, M.F., Abadenes, S., Schroeder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321. Zingarelli, B., Sheehan, M., Hake, P.W., O’Connor, M., Denenberg, A., Cook, J.A., 2003. Peroxisome proliferator activator receptor-γ ligands, 15-deoxy-prostaglandin J2 and ciglitazone, reduce systemic inflammation in polymicrobial sepsis by modulation of signal transduction pathways. J. Immunol. 171, 6827–6837.
Chapter 118
Olive Oil and Uncoupling Proteins Alfredo Fernández-Quintela, Itziar Churruca and María P. Portillo Department of Nutrition and Food Science, University of País Vasco, Paseo de la Universidad, Vitoria, Spain
118.1 INTRODUCTION The literature currently indicates that the physiological roles of the uncoupling proteins (UCPs) in mammals extend into energy substrate partitioning (thermogenesis), fatty acid oxidation, glucose disposal rates, insulin secretion, reactive oxygen species (ROS) production, apoptosis, and aging. At the moment, five UCP homologues (UCP1– UCP5) have been described (Table 118.1).
118.1.1 UCP1 UCP1 was first observed in brown adipose tissue mitochondria in 1976 (Ricquier and Kader, 1976). Its molecular weight varies between 32 and 33 kDa among different species. UCP1 is encoded by a single nuclear gene, which is located on chromosome 4 in humans (Cassard et al., 1990). UCP1 has classically been associated to brown adipose tissue of mammals and human infants, although it has also recently been discovered in other rat and mouse tissues (Carroll et al., 2005). Sympathetic innervation of brown adipose tissue, usually as a result of cold adaptation/acclimation, increases proton flux through fatty acid oxidation and electron transport chain activity resulting in heat being produced by that tissue (Nicholls and Locke, 1984). UCP1 gene regulation is mainly transcriptional, and therefore UCP1 mRNA levels reflect protein levels. The best stimulus of brown adipose tissue thermogenesis is cold exposure. Moreover, leptin, insulin, thyroid hormones and retinoic acid also increase UCP1 mRNA level (Lanni et al., 2003). Thus, the hormones regulating UCP1 are all involved in energy balance and obesity.
118.1.3 UCP3
118.1.2 UCP2 UCP2 has an amino acid sequence that is 59% identical to UCP1 and consists of 309 amino acids with a molecular weight of 33 kDa. The UCP2 gene is located on chromosome Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
7 in mice and on the chromosome region 11q13 in humans, in a region linked to energy homeostasis and obesity (Fleury et al., 1997). UCP2, unlike UCP1, is widely expressed in organs such as skeletal muscle, kidney, intestine, pancreatic islets, lung, heart, white and brown adipose tissue (Gimeno et al., 1997; Samec et al., 1998; Chan et al., 2001). However, the amount of UCP2 is smaller when compared to UCP1 abundance in brown adipose tissue, and consequently, contribution of UCP2 to thermogenesis is likely to be much more modest. Furthermore, UCP2 expression is particularly high in immune system tissues (i.e. spleen, thymus, macrophages) (Schrauwen et al., 1999; Mozo et al., 2005). Even though the protonophoric activity of UCP2 has been demonstrated, the biological functions of UCP2 have not been made fully clear. It has been suggested that UCP2 plays a role in lipid metabolism as well as in mitochondrial bioenergetics, oxidative stress, apoptosis, and even carcinogenesis (Adams, 2000). There is a reciprocal relationship between UCPs and mitochondrial ROS. UCP2 is part of a negative feedback mechanism that is able to respond to oxidative stress by controlling mitochondrial superoxide production. Thus, strategies providing cell protection from oxidative stress by enhanced UCP2 expression have been successful. Similarly, in genetically induced obese mice, hyperphagia triggers a substrate overload and a higher ROS production. This is compensated by an adaptative response, up-regulating UCP2 expression, which is an energetic spendthrift system. UCP2 expression is also up-regulated after cold exposure and fasting and subsequently decreased in response to refeeding (Baffy, 2005; Mozo et al., 2005).
UCP3 is predominantly expressed in skeletal muscle and brown adipose tissue of humans (Boss et al., 1997). It shows an amino acid sequence similar to that of UCP1 and UCP2, being 73% identical to UCP2. Both UCP2 and UCP3
1087
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1088
SECTION | II Cells and Cellular Effects
TABLE 118.1 Features of uncoupling proteins. Function
Expression
References
UCP1
Thermogenesis Decreased ROS formation
Brown adipose tissue Thymus
Ricquier and Kader, 1976 Nicholls and Locke, 1984 Nishikawa et al., 2000 Carroll et al., 2005
UCP2
Regulation of free fatty acid metabolism and transport Inhibition of insulin secretion
Immune system Brain Pancreas Heart Skeletal muscle White adipose tissue Liver
Fleury et al., 1997 Gimeno et al., 1997 Samec et al., 1998 Chan et al., 2001 Schrauwen and Hesselink, 2002
UCP3
Regulation of free fatty acid metabolism and transport Decreased ROS formation
Skeletal muscle Brown adipose tissue Heart
Boss et al., 1997 Fleury et al., 1997 Dulloo et al., 2001 Nishikawa et al., 2000 Schrauwen and Hesselink, 2002
UCP4
Thermogenesis Decreased ROS formation
Brain
Mao et al., 1999
UCP5
Thermogenesis
Brain Liver Skeletal muscle
Sanchis et al., 1998 Adams, 2000 Yu et al., 2000
UCP: uncoupling protein; ROS: reactive oxygen species.
were mapped in close proximity (75–150 kb) to each other (Lentes et al., 1999; Schrauwen and Hesselink, 2002). Physiologically, UCP3 gene expression is up-regulated in response to starvation, a situation in which energy expenditure in muscle is reduced rather than increased. Skeletal muscle and particularly glycolytic muscles alter their fuel substrate from glucose to lipids during starvation. It has been proposed that the biological role of UCP3 is to regulate lipid metabolism. Furthermore, physiological situations when free fatty acid availability increases are able to up-regulate UCP3 mRNA, which has led to the suggestion that UCP3 might be involved in fatty acid metabolism (Dulloo et al., 2001). Like UCP2, UCP3 controls mitochondrial ROS production through mild uncoupling. This underlies a putative function in protection against oxidative damage (Nishikawa et al., 2000).
118.1.4 UCP4 and UCP5 Two homologues with abundant expression in brain have been described recently. UCP4 shares 29, 33 and 34% homology with UCP1, UCP2 and UCP3, respectively, in humans. UCP5, termed brain mitochondrial carrier protein1 (BMCP1), is a protein with 34, 38 and 39% identity to UCP1, UCP2, and UCP3, respectively. UCP4 and UCP5 homologues may mediate metabolic changes induced by cold exposure, high-fat feeding or fasting, possibly playing
an important role in local or whole-body adaptational thermoregulatory processes (Sanchis et al., 1998, Mao et al., 1999; Adams, 2000; Yu et al., 2000).
118.2 DIETARY LIPID SOURCE AND UCPs Olive oil is the food that most represents the traditional Mediterranean diet. It is a very rich source of monounsaturated fatty acids, mainly oleic acid (18:1 n-9). It has been shown that dietary fats differing in fatty acid composition influence diet-induced non-shivering thermogenesis in different ways (Sadurskis et al., 1995; Matsuo et al., 1996; Portillo et al., 1998; Clarke, 2000; Rodríguez et al., 2002). With regard to effects of fatty acids on UCPs, in vivo, two processes have to be considered: (a) the alteration of protein amount due to changes in its expression and (b) the alteration of protein activity.
118.2.1 UCP expression Several studies have analyzed the effects of fatty acids on UCP expression, by using either isolated cells or different animal models. In vitro studies usually target the effects of isolated fatty acids whereas in vivo studies consider the effect of different lipid sources. Armstrong and Towle (2001) addressed a study on isolated hepatocytes in order to determine the effects of
CHAPTER | 118 Olive Oil and Uncoupling Proteins
different fatty acids on the expression of UCP2. They observed that UCP2 mRNA was up-regulated by eicosapentaenoic acid (20:5 n-3). A modest effect was seen in response to treatment with oleic acid (18:1 n-9) and no effect with saturated fatty acids. The authors also analyzed potential mediators for the fatty acid effects. In hepatocytes, UCP2 appeared to be activated through a PPARα-mediated pathway. Treatment of hepatocytes with a Wy-14643, a PPARα agonist, resulted in significant increase in UCP2 expression. This effect has also been observed in vivo. Mice fed a diet supplemented with Wy-14643 showed a four-fold increase in UCP2 expression in the liver (Kelly et al., 1998). Other members of the PPAR family have also been involved in the regulation of UCP2. In white and brown adipose tissues, activators of PPARγ increased expression of UCP2 (Camirand et al., 1996; Aubert et al., 1997). With regard to this regulatory pathway, it could be pointed out that PPARα binds polyunsaturated fatty acids with a higher affinity than saturated fatty acids (Kliewer et al., 1997; Latruffe and Vamecq, 1997). This may explain, at least in part, the results published by Armstrong and Towle. Several studies in the literature have compared the effects of oleic-acid-rich oils on UCP expression with those of other fat sources differing in fatty acid profile. Tsuboyama-Kasaoka et al. (1999), in a study performed on mice, analyzed the effects of high-oleic safflower oil, an excellent source of oleic acid (46%) and linoleic acid (45%), and fish oil, a source of n-3 fatty acids, on UCP expression in several tissues. Fish-oil-fed animals showed up-regulation of UCP2 in liver and brown adipose tissue, and UCP3 in skeletal muscle, but down-regulation of UCP2 in white adipose tissue and UCP3 in brown adipose tissue, with regard to high-oleic safflower oil-fed animals. Data on these two studies suggest that n-3 polyunsaturated fatty acids are more effective than n-6 polyunsaturated and monounsaturated fatty acids at stimulating hepatic UCP2 expression. As far as brown and white adipose tissues are concerned, there is no net effect, because this depends on tissue and UCP homologue. In a previous study from our laboratory (Rodríguez et al., 2002), we found that mRNAs of UCP1, UCP2 and UCP3 in interscapular brown adipose tissue, as well as that of UCP3 in gastrocnemius muscle, were higher in rats fed an olive oil (rich in monounsaturated fatty acids)-based diet than those in rats fed diets whose lipid sources were sunflower oil (rich in n-6 polyunsaturated fatty acids), palm oil or beef tallow (rich in saturated fatty acids). However UCP2 expression remained unchanged in white adipose tissue. Changes in mRNAs were not accompanied by close changes in UCPs’ protein content in interscapular brown adipose tissue. With regard to gastrocnemius muscle, rats fed olive oil showed significantly higher levels of both UCP3 mRNA and protein than rats fed the other three diets.
1089
In order to gain more insight into the mechanisms underlying these effects, the potential involvement of several hormones was considered. Some studies demonstrated that triiodothyronine increases the expression of UCP1, UCP2 and UCP3 (Gong et al., 1997; Masaki et al., 1997; Ricquier et al., 2000). The involvement of T3 in the up-regulating effect of olive oil was not evident because no significant differences in their serum concentrations were found among the four dietary groups. Circulating glucocorticoid and insulin concentrations are known to be potent regulators of fuel metabolism and gene transcription (Gong et al., 1997). In this study, insulin and corticosterone, the major glucocorticoid found in the plasma of rats, were not affected by dietary fats, thus giving no indication of any involvement of this hormone in the UCP up-regulating effect of olive oil. Several authors have proposed that fatty acids represent an important signal for the regulation of UCP activity (González-Barroso et al., 1996). In this study, the effects of dietary fat on UCP gene expression were not related to modifications in circulating free fatty acids because no differences were found among our experimental groups. Thus, the explanation of the effects of olive oil is not clear. It seems that these effects are not mediated by systemic metabolic changes, but rather related to a local effect on interscapular brown adipose tissue and gastrocnemius muscle, produced by oleic acid. The results found by Tsuboyama-Kasaoka et al. (1999) and Rodríguez et al. (2002) show that fatty acids can regulate each UCP differently, and also the same UCP in different tissues. Takeuchi et al. (1995) carried out an experiment in rats in order to address the effects of animal fat (lard, rich in saturated fatty acids), high-oleic safflower oil (rich in monounsaturated fatty acids), safflower oil (rich in n-6 polyunsaturated fatty acids) and linseed oil (rich in n-3 polyunsaturated fatty acids) on diet-induced thermogenesis and sympathetic activity in brown adipose tissue. In order to determine diet-induced thermogenesis, body oxygen consumption was measured 1 hour before and after meals. After 4–5 weeks oxygen consumption was found to be similar among all diet groups, but the postprandial level was significantly lower in rats fed the lard diet than in those fed the other diets. Also norepinephrine turnover rates were lower in the lard-fed group than in other groups. Thus, these results lead to the conclusion that saturated fatty acids reduce diet-induced thermogenesis. No differences were observed between monounsaturated and polyunsaturated fatty acids. Although the present work measured neither UCP mRNA nor UCP protein, taking into account that UCPs play an important role in diet-induced thermogenesis, the potential involvement of changes in UCP expression among experimental groups in the observed effects on oxygen consumption could be argued. Samec et al. (1999) carried out a study to examine poststarvation gene transcription of UCP2 and UCP3 in skeletal
1090
SECTION | II Cells and Cellular Effects
muscle during refeeding on diets varying in fat types. For this purpose, male Sprague-Dawley rats were fed restricted diets (50% energy restriction from the spontaneous food intake of ad libitum fed rats) for 2 weeks. At the end of this food restriction period, the animals were re-fed diets in isocaloric quantities, whose dietary fat sources were coconut oil, lard, olive oil, safflower oil, and fish oil, for periods of either 5 or 14 days. Dietary fat type did not affect UCP2 or UCP3 expression in skeletal muscle. In contrast, feeding the safflower diet (rich in n-6 polyunsaturated fatty acids) resulted in lower energy efficiency, in other words higher rates of energy expenditure, than feeding either the lard or olive oil diet. An intermediate situation between safflower oil and olive oil was induced by fish oil. This dissociation between energy expenditure values and skeletal muscle expression was also observed in the lack of a significant correlation between muscle UCP expression and energy expenditure, in a regression analysis of all diet groups pooled together. This discrepancy could be due to potential changes in other important UCPs, such as UCP1 in brown adipose tissue and UCP2 in white adipose tissue, not being considered.
The oxidative phosphorylation of ADP into ATP takes place in the mitochondrial inner membrane respiratory chain, which is made up of four complexes (I–IV) and the F0/F1 ATP-synthase. Complexes I and II reoxidize the reduced cofactors (NADH, FADH2) produced by cellular metabolic processes, starting a set of redox reactions leading the electrons to the complex IV, which reduces oxygen to water. At the same time, complexes I, III and IV pump protons from the mitochondrial matrix to the intermembrane space, generating a proton gradient on either side of the inner membrane (Mitchell’s chemiosmotic theory), that gives the F0/F1 ATP-synthase the energy that it requires to phosphorylate the ADP into ATP. Thus, ATP synthesis is directly coupled to mitochondrial respiration (oxygen consumption). However, in several tissues there is a proton leak through the inner mitochondrial membrane that is not associated with ATP synthesis. One possible explanation is a slippage in proton pumping by the respiratory chain complexes. Another possibility is a proton leak through the inner membrane due to the presence of UCPs. These proteins act as H⫹ or OH⫺ ions translocators in a carrier-like fashion, providing an alternative route for protons to re-enter the mitochondrial matrix, uncoupling the processes of electron transport/proton-gradient generation on the one hand, and ATP synthesis on the other, since fewer protons go through the F0/F1 ATP synthase. Therefore, the electrochemical gradient is lowered and the activity of the respiratory chain is enhanced to maintain the mitochondrial membrane potential, thereby the energy that is derived from the oxidized substrates is released as heat, affecting energy metabolism efficiency (Schrauwen et al., 1999; Mozo et al., 2005) (Figure 118.1).
118.2.2 UCP Activity It has been shown that uncoupling activity of UCPs exhibits an absolute requirement for fatty acids (Winkler and Klingenberg, 1994). There are currently two hypotheses to explain this issue. In order to better understand these two hypotheses, a brief explanation of the UCPs’ mechanism of action is provided beneath.
H+
H+ H+ H+
H+ H+
H+
H+
H+
H+ H+
H+
Intermembrane space
H+
H+
H+
H+
H+
H+
H+ H+ H+
H+
e−
H+
H+ H+ H+
H+
e−
e− IV
III
H+
H+
NADH
H+
F0 F1
II H+
H+ H+
UCP I
H+
H+
Inner membrane
e−
H+
H+
H+
H+ H+
H+ H+
H+
H+
½ O2
FADH2 Mitochondrial matrix
H+
H2O
ADP + Pi
H+
ATP
F0/F1 ATP synthase
FIGURE 118.1 Uncoupling proteins and mitochondrial oxidative phosphorylation. Proton leak through the inner mitochondrial membrane not associated with ATP synthesis. Uncoupling proteins provide an alternative route for protons to re-enter the mitochondrial matrix. H⫹: proton; UCP: uncoupling protein; NADH: nicotine adenine dinucleotide (reduced form); FADH2: flavin adenine dinucleotide (reduced form); ADP: adenosine diphosphate; ATP: adenosine triphosphate; Pi: inorganic phosphorous. Modified with permission from Mozo et al., 2005.
1091
CHAPTER | 118 Olive Oil and Uncoupling Proteins
118.2.3 Hypotheses Explaining the Role of Fatty Acids in the UCP Proton Flux
118.3 MINOR CONSTITUENTS OF OLIVE OIL AND UCPs
On the one hand, Klingenberg and Huang (1999) proposed that fatty acids provide a free carboxy group that serves as a catalyst of H⫹ translocation. Thus, free fatty acids may be H⫹ capturer and injector. On the other hand, Garlid et al. (1998) proposed the fatty acid cycling model. The fatty acid carboxylate group moves freely by a flip-flop mechanism to the other side of the membrane, delivering protons electroneutrally to the mitochondrial matrix. Then, anion diffuses laterally in the bilayer to bind on UCP, which facilitates the backward transport of the deprotonated anionic fatty acid across the membrane (Figure 118.2). This latter mechanism is influenced by dietary fatty acid profile. Fatty acid comprising membrane phospholipids can be more saturated or unsaturated depending on the dietary fat composition. Saturated fatty acids have a straight conformation, whereas unsaturated fatty acids have kinks in the tails because of cis double bonds. Due to these conformations, saturated fatty acids appear more packed together in the cell membrane, and in turn, more arranged than the unsaturated fatty acids do, affecting some physicochemical membrane properties such as fluidity. Based on the fatty acid cycling model, Beck et al. (2007) demonstrated that polyunsaturated fatty acids are the most potent of the UCP activators, due to the fact that double bonds lead to high membrane fluidity promoting flip-flop of fatty acids. As a result, the basal proton leak in mitochondria from different tissues depends on the fatty acid composition of inner membrane phospholipids, and the proton and membrane conductance increases in the range palmitic ⬍ oleic ⬍ eicosatrienoic ⬍ linoleic ⬍ retinoic⬍ arachidonic acids.
In addition to the high amount of monounsaturated fatty acids, virgin olive oil contains at least 30 phenolic compounds (Visioli and Galli, 1994, 1998; Tuck and Hayball, 2002; Visioli et al., 2002; Vissers et al., 2004), e.g., hydroxytyrosol and oleuropein, which are strong antioxidants and radical scavengers, and have been demonstrated to be absorbed by humans (Vissers et al., 2002). The effects of these compounds on UCPs have not been widely studied. It has been demonstrated that interscapular brown adipose tissue UCP1 content in rats fed an extra virgin olive oil diet is significantly higher than that in rats fed either corn oil or refined olive oil diets (Oi-Kano et al., 2007). More precisely, phenols in extra virgin olive oil stimulate norepinephrine and epinephrine secretions and suppress body fat accumulation, by increasing triglyceride catabolism through the enhancement of thermogenesis in interscapular brown adipose tissue via an increase in UCP1 content. Furthermore, Oi-Kano et al. (2007) demonstrated that among the extra virgin olive oil phenolic compounds, the oleuropein fraction, containing mainly oleuropein glycoside and oleuropein aglycone, is responsible for this effect, while the fraction containing mainly hydroxytyrosol, tyrosol, vanillic acid and ferulic acid does not affect the thermogenesis.
SUMMARY POINTS ●
●
COOH
H+
COO−
COO−
Intermembrane space
●
●
UCP
●
COOH
COO−
COO−
H+ Mitochondrial matrix FIGURE 118.2 Garlid’s fatty acid cycling model. The fatty acid carboxylate group moves freely by a flip-flop mechanism to the other side of the membrane, delivering protons electroneutrally to the mitochondrial matrix. H⫹: proton; UCP: uncoupling protein. Modified with permission from Garlid et al., 1998.
Dietary fatty acids can influence both expression and activity of uncoupling proteins. As far as UCP expression is concerned, n-3 polyunsaturated fatty acids are more effective than n-6 polyunsaturated and monounsaturated fatty acids at stimulating hepatic UCP2 expression. Fatty acids can regulate each UCP differently and also the same UCP in different tissues. The activation of UCPs by fatty acids increases in the range palmitic ⬍ oleic ⬍ eicosatrienoic ⬍ linoleic ⬍ retinoic ⬍ arachidonic acids. Oleuropein glycoside and oleuropein aglycone present in extra virgin olive oil have been shown to increase UCP1 content in interscapular brown adipose tissue.
REFERENCES Adams, S.H., 2000. Uncoupling protein homologs: emerging views of physiological function. J. Nutr. 130, 711–714. Armstrong, M.B., Towle, H.C., 2001. Polyunsaturated fatty acids stimulate UCP-2 expression via a PPARα-mediated pathway. Am. J. Physiol. 281, E1197–E1204. Aubert, J., Champagny, O., Saint-Mare, P., Negrel, R., Collins, S., Ricquier, D., Aihlaud, G., 1997. Up-regulation of UCP-2 gene expression by PPAR
1092
agonists in preadipose and adipose cells. Biochem. Biophys. Res. Commun. 238, 606–611. Baffy, G., 2005. Uncoupling protein-2 and non-alcoholic fatty liver disease. Front. Biosci. 10, 2082–2096. Beck, V., Jaburek, M., Demina, T., Rupprecht, A., Porter, R.K., Jezek, P., Pohl, E.E., 2007. Polyunsaturated fatty acids activate human uncoupling proteins 1 and 2 in planar lipid bilayers. FASEB J. 21, 1137–1144. Boss, O., Samec, S., Paoloni-Giacobino, A., Rossier, C., Dulloo, A.G., Seydoux, J., Muzzin, P., Giacobino, J.P., 1997. Uncoupling protein-3: a new member of the mitochondrial carrier family with tissue-specific expression. FEBS Lett. 408, 39–42. Camirand, A., Marie, V., Rabelo, R., Silva, J.E., 1996. Thiazolidinediones stimulate uncoupling protein-2 expression in cells lines representing white and brown adipose tissue and skeletal muscle. Endocrinology 139, 428–431. Carroll, A.M., Haines, L.R., Pearson, T.W., Fallon, P., Walsch, C., Brennan, C.M., Breen, E.P., Porter, R.K., 2005. Identification of a functioning mitochondrial uncoupling protein 1 in thymus. J. Biol. Chem. 280, 15534–15543. Cassard, S.D., Bouillaud, F., Mattei, M.G., Hentz, E., Raimbault, S., Thomas, M., Ricquier, D., 1990. Human uncoupling protein gene: structure, comparison with rat gene, and assignment to the long arm of chromosome 4. J. Cell Biochem. 43, 255–264. Chan, C.B., De Leo, D., Joseph, J.W., McQuaid, T.S., Ha, X.F., Tsushima, R.G., Pennefather, P.S., Salapatek, A.M., Wheeler, M.B., 2001. Increased uncoupling protein-2 levels in β-cells are associated with impaired glucose-stimulated insulin secretion. Diabetes 50, 1302–1310. Clarke, S.D., 2000. Polyunsaturated fatty acid regulation of gene transcription: a mechanism to improve energy balance and insulin resistance. Br. J. Nutr. 83 (Suppl 1), S59–S66. Dulloo, A.G., Samec, S., Seydoux, J., 2001. Uncoupling protein 3 and fatty acid metabolism. Biochem. Soc. Transact. 29, 785–791. Fleury, C., Neverova, M., Collins, S., Raimbault, S., Champigny, O., LeviMeyrueis, C., Bouillaud, F., Seldin, M.F., Surwitt, R.S., Ricquier, D., Warden, C.H., 1997. Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat. Genet. 15, 269–723. Garlid, K.D., Jaburek, M., Jezek, P., 1998. The mechanism of proton transport mediated by mitochondrial uncoupling proteins. FEBS Lett. 438, 10–14. Gimeno, R.E., Dembski, M., Weng, X., Deng, N., Shyjan, A.W., Gimeno, C.J., Iris, F., Ellis, S.J., Woolf, E.A., Tartaglia, L.A., 1997. Cloning and characterization of an uncoupling protein homolog: a potential molecular mediator of human thermogenesis. Diabetes 46, 900–906. Gong, D.W., He, Y., Karas, M., Reitman, M., 1997. Uncoupling protein3 is a mediator of thermogenesis regulated by thyroid hormone, β3adrenergic agonists and leptin. J. Biol. Chem. 272, 24129–24132. González-Barroso, M.M., Fleury, C., Arechaga, I., Zaragoza, P., Levi-Meyrueis, C., Raimbault, S., Ricquier, D., Bouillaud, F., Rial, E., 1996. Activation of the uncoupling protein by fatty acids is modulated by mutations in the c-terminal region of the protein. Eur. J. Biochem. 239, 445–450. Kelly, L.J., Vicario, P.P., Thompson, G.M., Candelore, M.R., Doebber, T.W., Ventre, J., Wu, M.S., Meurer, R., Forrest, M.J., Conner, M.W., Cascieri, M.A., Moller, D.E., 1998. Peroxisome proliferator-activated receptor γ and α in vivo regulation of uncoupling protein (UCP1, UCP2, UCP3) gene expression. Endocrinology 139, 4921–4927. Kliewer, S.A., Sundseth, S.S., Jones, S.A., Brown, P.J., Wisely, G.B., Koble, C.S., Devechand, P., Whali, W., Wilson, T.M., Lenhard, J.M., Lehmann, J.M., 1997. Fatty acids and eicosanoids regulate gene expression through direct interactions with peroxisome proliferator-activated receptor alpha and gamma. Proc. Natl. Acad. Sci. USA 94, 4318–4323.
SECTION | II Cells and Cellular Effects
Klingenberg, M., Huang, S.G., 1999. Structure and function of the uncoupling protein from brown adipose tissue. Biochim. Biophys. Acta 1415, 271–296. Lanni, A., Moreno, M., Lombardi, A., Goglia, F., 2003. Thyroid hormone and uncoupling proteins. FEBS Lett. 543, 5–10. Latruffe, N., Vamecq, J., 1997. Peroxisome proliferator activated receptors (PPARs) are regulators of lipid metabolism. Biochimie 79, 81–94. Lentes, K.U., Tu, N., Chen, H., Winnikes, U., Reinert, I., Marmann, G., Pirke, K.M., 1999. Genomic organization and mutational analysis of the human UCP2 gene, a prime candidate gene for human obesity. J. Recept. Signal. Transduct. Res. 19, 229–244. Mao, W., Yu, X.X., Zhong, A., Li, W., Brush, J., Sherwood, S.W., Adams, S.H., Pan, G., 1999. UCP4, a novel brain-specific mitochondrial protein that reduces membrane potential in mammalian cells. FEBS Lett. 443, 326–330. Masaki, T., Yashimatsu, H., Kakuma, T., Hidaka, S., Kurakowa, M., Sakata, T., 1997. Enhanced expression of uncoupling protein 2 gene in rat white adipose tissue and skeletal muscle following chronic treatment with thyroid hormone. FEBS Lett. 418, 323–326. Matsuo, T., Komuro, M., Suzuki, M., 1996. Beef tallow diet decreases uncoupling protein content in the brown adipose tissue of rats. J. Nutr. Sci. Vitaminol. 42, 595–601. Mozo, J., Emre, Y., Bouillaud, F., Ricquier, D., Criscuolo, F., 2005. Thermoregulation: what role for UCPs in mammals and birds? Biosci. Rep. 25, 227–249. Nicholls, D.G., Locke, R.M., 1984. Thermogenic mechanisms in brown fat. Physiol. Rev. 64, 1–64. Nishikawa, T., Edelstein, D., Du, X.L., Yamagishi, S., Matsumura, T., Kaneda, Y., Yorek, M.A., Beebe, D., Oates, P.J., Hammes, H.P., Giardino, I., Brownlee, M., 2000. Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage. Nature 404, 787–790. Oi-Kano, Y., Kawada, T., Watanabe, T., Koyama, F., Watanabe, K., Senbongi, R., Iwai, K., 2007. Extra virgin olive oil increases uncoupling protein 1 content in brown adipose tissue and enhances noradrenaline and adrenaline secretions in rats. J. Nutr. Biochem. 18, 685–692. Portillo, M., Serra, F., Simón, E., del Barrio, A.S., Palou, A., 1998. Energy restriction with high-fat diet enriched with coconut oil gives higher UCP1 and lower fat in rats. Int. J. Obes. 23, 974–979. Ricquier, D., Kader, J.C., 1976. Mitochondrial protein alteration in active brown fat: a sodium dodecyl sulfate-polyacrylamide gel electrophoretic study. Biochem. Biophys. Res. Commun. 73, 577–583. Ricquier, D., Miroux, B., Larose, M., Cassard-Doulcier, A.M., Bouillaud, F., 2000. Endocrine regulation of uncoupling proteins and energy expenditure. Int. J. Obes. 24, S86–S88. Rodríguez, V.M., Portillo, M.P., Picó, C., Macarulla, M.T., Palou, A., 2002. Olive oil feeding up-regulates uncoupling genes in rat brown adipose tissue and skeletal muscle. Am. J. Clin. Nutr. 75, 213–220. Sadurskis, A., Dicker, A., Cannon, B., Nedergaard, J., 1995. Polyunsaturated fatty acids recruit brown adipose tissue: increased UCP content and NST capacity. Am. J. Physiol. 269, E351–E360. Samec, S., Seydoux, J., Dulloo, A.G., 1998. Role of UCP homologues in skeletal muscles and brown adipose tissue: mediators of thermogenesis or regulators of lipids as fuel substrate? FASEB J. 12, 715–724. Samec, S., Seydoux, J., Dulloo, A.G., 1999. Post-starvation gene expression of skeletal muscle uncoupling protein 2 and uncoupling protein 3 in response to dietary fat levels and fatty acid composition. Diabetes 48, 436–441. Sanchis, D., Fleury, D., Chomiki, N., Goubern, M., Huang, Q., Neverova, M., Grégoire, F., Easlick, J., Raimbault, S., Lévi-Meyrueis, C.,
CHAPTER | 118 Olive Oil and Uncoupling Proteins
Miroux, B., Collins, S., Seldin, M., Richard, D., Warden, C., Bouillaud, F., Ricquier, D., 1998. BMCP1, a novel mitochondrial carrier with high expression in the central nervous system of humans and rodents, and respiration uncoupling activity in recombinant yeast. J. Biol. Chem. 273, 34611–34615. Schrauwen, P., Hesselink, M., 2002. UCP2 and UCP3 in muscle controlling body metabolism. J. Exp. Biol. 205, 2275–2285. Schrauwen, P., Walder, K., Ravussin, E., 1999. Human uncoupling proteins and obesity. Obes. Res. 7, 97–105. Takeuchi, H., Matsuo, T., Tokuyama, K., Shimomura, Y., Suzuki, M., 1995. Diet-induced thermogenesis is lower in rats fed a lard diet than in those fed a high oleic acid safflower oil diet, a safflower oil diet or a linseeds oil diet. J. Nutr. 125, 920–925. Tsuboyama-Kasaoka, N., Takahashi, M., Kim, H., Ezaki, O., 1999. Upregulation of liver uncoupling protein-2 mRNA by either fish oil feeding or fibrate administration in mice. Biochem. Biophys. Res. Commun. 257, 879–885. Tuck, K.L., Hayball, P.J., 2002. Major phenolic compounds in olive oil: metabolism and health. J. Nutr. Biochem. 13, 636–644.
1093
Visioli, F., Galli, C., 1994. Oleuropein protects low density lipoprotein from oxidation. Life Sci. 55, 1965–1971. Visioli, F., Galli, C., 1998. Olive oil phenols and their potential effects on human health. J. Agric. Food Chem. 46, 4292–4296. Visioli, F., Poli, A., Galli, C., 2002. Antioxidant and other biological activities of phenols from olive and olive oil. Med. Res. Rev. 22, 65–75. Vissers, M.N., Zock, P.L., Katan, M.B., 2004. Bioavailability and antioxidant effects of olive oil phenols in humans: a review. Eur. J. Clin. Nutr. 58, 955–965. Vissers, M.N., Zock, P.L., Roodenburg, A.J.C., Leenen, R., Katan, M.B., 2002. Olive oil phenols are absorbed in humans. J. Nutr. 132, 409–417. Winkler, E., Klingenberg, M., 1994. Effect of fatty acids on H⫹ transport activity of the reconstituted uncoupling protein. J. Biol. Chem. 269, 2508–2515. Yu, X.X., Mao, W., Zhong, A., Scow, P., Brush, J., Sherwood, S.W., Adams, S.H., Pan, G., 2000. Characterization of novel UCP5/BMCP1 isoforms and differential regulation of UCP4 and UCP5 expression through dietary or temperature manipulation. FASEB J. 14, 1611–1618.
This page intentionally left blank
Chapter 119
Extra Virgin Olive Oil Biophenols and mRNA Transcription of Glutathionerelated Enzymes Rosaria Varì, Beatrice Scazzocchio, Claudio Giovannini and Roberta Masella Nutrition Unit, Department of Veterinary Public Health and Food Safety, Istituto Superiore di Sanità, Rome, Italy
119.1 INTRODUCTION 119.1.1 Reactive Oxygen Species, Oxidative Stress and Degenerative Diseases Pro-oxidant species, in particular oxygen free radicals, the so-called reactive oxygen species (ROS), are normally generated during cell metabolism, either as bio-products of several enzymes or as a result of the intracellular metabolism of foreign compounds, and by ionizing radiation. Besides playing different positive roles in vivo, ROS may be also highly damaging as they can attack biomolecules, namely proteins, lipids and nucleic acids, and propagate the oxidative damage by generating organic radical species. As a result, ROS can cause enzyme inactivation, membrane and DNA damage, gene and signaling pathway modulation. When the level of ROS exceeds the antioxidant capacity of the cell, the intracellular redox homeostasis is altered and oxidative stress takes place. Oxidative stress is a feature common to aging and several degenerative diseases. To cope with an excess of free radicals, humans have developed sophisticated mechanisms to maintain redox homeostasis. These protective mechanisms either scavenge or detoxify ROS or block their production and include enzymatic and non-enzymatic antioxidant defenses produced in the body, the endogenous antioxidant defenses. Other antioxidants, exogenous ones, are supplied with the diet.
119.1.2 Health Benefits of Extra Virgin Olive Oil Biophenols Epidemiological studies have demonstrated that the incidence of coronary heart disease and cancer is lower in the Mediterranean area than in other Western countries. This Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
finding has been attributed to the Mediterranean diet (Hu, 2003) characterized by a high consumption of vegetables, fruits, legumes, grains, nuts and seeds, a moderate to high consumption of fish, a low to moderate consumption of dairy products and ethanol (mainly wine) and a low consumption of red meats. In this dietary pattern, olive oil is the principal source of fat. The health benefit of olive oil is surely due to the high amount of monounsaturated fatty acids, but it is likely also the result of the presence of several non-fat microcomponents with relevant biological properties. In particular, among the minor components of extra virgin olive oil (EVOO), biophenols have long attracted the attention of researchers, especially because of their strong antioxidant power, which has been attributed a beneficial role in the prevention of cardiovascular disease, type 2 diabetes and cancer, by counteracting the occurrence of oxidative processes which are strictly linked to their pathogenesis (Masella et al., 1999; Tuck and Hayball, 2002; Covas, 2007; Giovannini et al., 2007). Specifically, a great body of evidence has supported lipid peroxidation as one of the main pathogenic mechanisms of atherosclerosis. This hypothesis arises from the evidence of oxidized lipoproteins in atherosclerotic plaques, by the greater oxidizability of low-density lipoproteins isolated from patients with cardiovascular diseases, and by the protective effects of some antioxidants. The daily consumption of EVOO has been demonstrated to increase biophenols in plasma and resistance against oxidative damages to lipoproteins and DNA in humans (Masella et al., 2001; Fito et al., 2007) and in animal models (Coni et al., 2000). However, the concept that the beneficial effects of biophenols depend on their antioxidant capacity appears now a simplistic way to conceive their activity. In particular,
1095
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1096
SECTION | II Cells and Cellular Effects
in complex biological systems, EVOO biophenols can influence the intracellular redox balance by affecting specific signaling pathways and nuclear factors (Beauchamp et al., 2005; Masella et al., 2005; Santangelo et al., 2007). The aim of this chapter is to describe the effects that EVOO biophenols exert on endogenous antioxidant defenses, namely, glutathione and related enzymes.
119.2 GLUTATHIONE AND GSH-RELATED ENZYMES 119.2.1 Glutathione A central role in the regulation of intracellular redox balance is played by the tripeptide γ-glutamylcysteinylglycine, i.e., glutathione (GSH) (Figure 119.1). This cysteine-containing tripeptide is the major non-enzymatic regulator of intracellular redox homeostasis; ubiquitous in millimolar concentration in all cell types, it participates in redox reactions by the reversible oxidation of its active thiol. In fact, glutathione – either in reduced (GSH) or oxidized (GSSG) form – is better referred to as glutathione disulfide (Figure 119.1). The GSH buffer system modulates cell response to redox changes, and is important in the regulation of most cellular metabolic processes and in maintaining cellular viability. Furthermore, protein thiol status is a critical factor in mediating biological processes. Finally, GSH can exert a regulatory role by covalently binding to proteins through a process called glutathionylation (Filomeni et al., 2005). Under normal cellular redox conditions, the major portion of glutathione is in its reduced form and is distributed in nucleus, endoplasmic reticulum and mitochondria. In the presence of oxidative stress, GSH concentration rapidly decreases while GSSG – potentially highly cytotoxic – increases because of the reduction of peroxides or as a result of free radical scavenging. This mechanism has two important consequences: (1) the thiol redox status of the cell will shift and activate transcriptional elements, and (2) GSSG may be preferentially secreted by the cell and degraded extracellularly, increasing the cellular depletion of GSH. Being both a nucleophile and a reductant, GSH can react with electrophilic toxicants (to produce non-toxic, watersoluble compounds readily excreted in urine), and oxidizing
O
Glutathione (GSH)
+ NH3
O
C CH CH2 CH2 C O−
NH
H
O
C
C
O NH CH2 C
CH2
O−
SH Glutamate
Cysteine
Glycine
2 GSH + H2O2 2H2O + GSSG FIGURE 119.1 Chemical structure of glutathione. Reduced glutathione (GSH) is a tripeptide of γ-glutamate, cysteine and glycine.
species (to avoid damages to the biological macromolecules), respectively. GSH can directly scavenge free radicals or act as a substrate of enzymes, including glutathione peroxidases (GPx) and glutathione S-transferases (GST), involved in the detoxification/reduction of hydrogen peroxide, lipid hydroperoxides and electrophilic compounds (Figure 119.2).
119.2.2 Glutathione-related Enzymes and GSH Cycle GPx constitute a family of enzymes, which are capable of reducing a variety of organic and inorganic hydroperoxides to the corresponding hydroxy compounds by utilizing GSH and/or other reducing equivalents (Ursini et al., 1995). All GPx enzymes are seleno-proteins and their primary function is to counteract oxidative attack. When GSH is used as susbstrate, a seleno-disulfide is formed, that is cleaved by a second GSH molecule to yield the reduced GPx. GST comprise several enzymes which detoxify noxious electrophilic metabolites of xenobiotics produced intracellularly after the exposure to pollutants, consumption of overcooked or mycotoxin-contaminated food, or polluted water (Hayes et al., 2005). They also protect against reactive compounds produced in vivo during oxidative stress by catalyzing the conjugation of GSH with oxidation end-products. During the reaction catalyzed by GPx, the exaggerated production of GSSG can lead to the formation of mixed Glutamate cysteine glycine
Toxic compounds
γ-GCS
s NADP+
H2O2 Hydroperoxides
GSH
GST GRed NADPH
GPx GSSG
H2O Hydroxy compounds
GSH-adducts in out
FIGURE 119.2 Glutathione and related enzymes. Reduced glutathione (GSH) can directly scavenge free radicals or act as a substrate for glutathione peroxidases (GPx) and glutathione transferases (GST) during the detoxification of hydrogen peroxide, lipid hydroperoxides and electrophilic compounds. During GST-mediated reactions, GSH is conjugated with various electrophiles, and the GSH-adducts thus formed are actively secreted by the cell. The production of oxidized glutathione (GSSG) by GPx can lead to: (i) the release of GSSG by the cell to maintain the intracellular GSH/GSSG ratio, or (ii) the back-reduction to GSH by glutathione reductase (GR) utilizing NADPH as reductant. The resulting depletion of cellular GSH can be replaced by de novo synthesis through two sequential ATP-dependent reactions catalyzed by γ-glutamylcysteine synthetase (γ-GCS) – the rate-limiting enzyme – and glutathione synthetase. Reprinted with permission from Masella et al., 2005.
CHAPTER | 119 Extra Virgin Olive Oil Biophenols and mRNA Transcription of Glutathione-related Enzymes
disulfides in cellular proteins, or to the release of excess GSSG by the cell. During the GST-mediated reactions, GSH is conjugated with electrophiles, and the GSH adducts are actively secreted by the cell. Mixed disulfide formation together with GSSG or GS-conjugated efflux can result in the depletion of cellular GSH. To meet the increased requirement for GSH, cells can react by a de novo synthesis or by reducing the formed GSSG. GSH is de novo synthesized in two sequential ATPdependent reactions catalyzed by γ-glutamylcysteine synthetase (γ-GCS), the rate-limiting enzyme, and glutathione synthetase, respectively (Lu, 2000). On the other hand, GSSG can be reduced back to GSH by the action of glutathione reductase (GR), a flavoenzyme which utilizes NADPH as reductant (Argyrou and Blanchard, 2004). In conclusion, the presence of GSH is essential but not in itself sufficient to prevent the cytotoxicity of ROS, being of fundamental importance to the efficiency of the entire GSH cycle which is determined by the functionality of glutathione-related enzymes (Table 119.1).
119.3 EVOO BIOPHENOLS AND GSH CYCLE 119.3.1 Antioxidant Activity of Biophenols Different classes of biophenols have been demonstrated to strongly increase GSH concentration in different cell types
TABLE 119.1 Features of glutathione-related enzymes. ●
Cell production of reactive oxygen species (ROS) can exceed the endogenous antioxidant defenses, leading to oxidative stress and consequently to oxidation of cellular macromolecules
●
A central role in the regulation of intracellular redox balance is played by the tripeptide glutathione which is able to scavenge free radicals by its active thiol group
●
The maintenance of the normal balance between the reduced (GSH) and oxidized (GSSG) form of glutathione is due to the efficiency of glutathione-related enzymes
●
Glutathione peroxidases (GPx) reduce a variety of organic and inorganic hydroperoxides to the corresponding hydroxy compounds by oxidizing GSH to GSSG
●
Glutathione S-transferases (GST) are involved in the detoxification/reduction of hydrogen peroxide, lipid hydroperoxides and electrophilic compounds by forming GS-adducts
●
GPx and GST activities lead to depletion of cellular GSH
●
Glutathione reductase (GR) and γ-glutamylcysteine synthetase (γ-GCS) restore the intracellular GSH content
●
GR reduces GSSG back to GSH
●
γ-GCS is the key enzyme in the de novo synthesis of GSH
1097
thanks to an improved activity of the enzymes involved in GSH synthesis (Moskaug et al., 2005) and metabolism (Hofmann et al., 2006). Even though several studies have demonstrated that EVOO biophenol consumption improves the redox balance in humans (Weinbrenner et al., 2004), no studies have been addressed to specifically evaluate the effects of EVOO biophenols on glutathione levels and the efficiency of the glutathione cycle. Recently, an increase in GPx has been reported in humans following the consumption of olive oil rich in biophenols (Fito et al., 2005). We have demonstrated that specific EVOO biophenols, namely tyrosol, hydroxytyrosol, oleuropein and protocatechuic acid, spare GSH and increase GSH content in cells undergoing oxidative stress (Giovannini et al., 1999, 2002, 2003, 2008; Masella et al., 2004; Di Benedetto et al., 2007). Although these effects can be associated with the antioxidant properties of the tested biophenols, we have found evidence that other mechanisms of action are involved in strengthening the antioxidant defense system. This hypothesis has been first suggested by the experiments carried out in the human colon adenocarcinoma cell line Caco2 under oxidative stress. These cells, treated with oxidized lipids, show signs of cytotoxicity and apoptosis which are completely overcome by tyrosol, the most abundant EVOO biophenol which has, however, weak antioxidant properties (Giovannini et al., 1999). Its protective effects are likely to be linked to the increased intracellular GSH level, that exerts powerful protective effects against oxidative injuries in cell systems (Giovannini et al., 2002, 2003, 2008). These results are consistent with those obtained in the murine macrophage-like cell line J774 A.1. These cells, like human macrophages, smooth muscle cells and endothelial cells in the arterial wall, are able to oxidize low-density lipoproteins (LDL), giving origin to oxidized lipoproteins (oxLDL), which are believed to play a pivotal role in the development of atherosclerotic lesions. Specifically, the oxidative process depends on the intracellular balance between pro-oxidants, such as ROS produced by the cells, and antioxidants including the GSH system.
119.3.2 Biophenols and the GSH System In the same cell model we have compared the effects of the weak antioxidant tyrosol and the strong antioxidant hydroxytyrosol on the redox balance during cell-induced LDL oxidation (Di Benedetto et al., 2007). Unlike hydroxytyrosol, tyrosol has been demonstrated to counteract intracellular ROS production only after 24 hours, or when the cells are pre-incubated with the biophenol for 2 h before starting the oxidation of LDL (Figure 119.3). Conversely, both biophenols preserve GSH from consumption and GSH-related enzymes from oxidative inactivation (Figure 119.4). It may be hypothesized that tyrosol acts in different ways. Its antioxidant ability, in the strict sense of the word,
1098
SECTION | II Cells and Cellular Effects
A 250 *
LDL+Tyr
LDL+OHTyr
A
*
140 120
*
* *
% of control
% of control
200
LDL
150 100
Control
LDL
LDL + Tyr
LDL + OHtyr
100 *
80
*
60 40
50
20 6h
12h
B 200
*
* *
% of control
0
24h
150
100
B
5
mU mg–1 cell protein
0
4
6h
12h
24h
3 2 1 * 0
50
0
6h
12h
24h
FIGURE 119.3 H2O2 production in J774 A.1 cells during cell-mediated oxidation of low-density lipoproteins (LDL). Cytofluorimetric analysis of H2O2 production after 6, 12, and 24 h incubation with LDL alone (LDL) and in the presence of 0.5 mmol L⫺1 tyrosol (LDL ⫹ Tyr) or 3 μmol L⫺1 hydroxytyrosol (LDL ⫹ OHtyr). Biophenols were added to the medium (A) together with LDL for the entire oxidation process, or (B) for 2 h before LDL addition and then removed from the medium. Values (mean ⫾ S.E.M., n ⫽ 4) represent the fluorescence intensity of treated cells compared with that of control cells considered as 100%. *p ⬍ 0.001 vs. control.
appears of little intensity, but when its capacity to spare GSH and reinforce intracellular antioxidant defenses is considered, tyrosol shows a high protective effect that is likely linked to its ability to penetrate and accumulate in cells. Further and clear confirmations have been acquired by testing the efficacy of other EVOO biophenols, namely oleuropein and protocatechuic acid, in the same cell model (Masella et al., 2004). Both biophenols have shown a strong power to inhibit cell-mediated oxidation of LDL by completely counteracting the onset of cellular oxidative stress. In fact, when these biophenols are put together with the lipoproteins in the culture medium, they counteract the peroxidation of the lipid components. These findings could be ascribed to the ability of protocatechuic acid and oleuropein to act as scavengers for radicals and chain-breaking antioxidants, both actions that are related to the ortho- and para-hydroxyl groups present on the phenol ring. On the other hand, similar results have been obtained when the cells were pre-incubated with the biophenols for 2 h and then treated with LDL in the absence of external antioxidants (Masella et al., 2004). The inhibition of lipid peroxidation seems to depend, thus, on the intracellular accumulation of these biophenols and on their capability to somehow block some of the early intracellular events
mU mg–1 cell protein
C
Control
LDL
LDL + Tyr
LDL + OHtyr
LDL + Tyr
LDL + OHtyr
12 10 8 6 *
4 2 0
Control
LDL
FIGURE 119.4 Effects of biophenols on glutathione (GSH) content and GSH-related enzyme activities in J774 A.1 cells. (A) Cytofluorimetric analysis of intracellular GSH after 6, 12, and 24 h incubation with lowdensity lipoproteins alone (LDL) and in the presence of 0.5 mmol L⫺1 tyrosol (LDL ⫹ Tyr), or 3 μmol L⫺1 hydroxytyrosol (LDL ⫹ OHTyr). Values (mean ⫾S.E.M., n ⫽ 4) represent the fluorescence intensity obtained from treated cells and compared with those of control cells considered as 100%. (B) Glutathione reductase (GR), and (C) Glutathione peroxidase (GPx) activities in J774 A.1 cells after 24 h incubation performed as described above. Values (mean ⫾S.E.M., n ⫽ 3) are expressed as mU mg⫺1 of protein. *p ⬍ 0.05 vs. control, LDL ⫹ Tyr, LDL ⫹ OHtyr. Reprinted with permission from Di Benedetto et al., 2007.
necessary to begin the oxidative process in lipoproteins. These protective effects exerted by EVOO biophenols could be due to the improvement of the GSH redox cycle. In fact, it is worth noting that, when pre-incubated with the cells, protocatechuic acid and oleuropein exerted a potent inhibitory effect on the increase in ROS production that was already evident after 6 h of incubation with LDL, and on the decrease in GSH content after 12 h, with respect to control cells. The mechanisms underlying this specific activity of the biophenols are associated with the preservation of the functional efficiency of the redox enzymes GPx and GR. In particular, GPx activity is significantly increased not only with respect to the cells treated with low-density lipoproteins, but also with respect to control cells (Figure 119.5A), suggesting that the biophenols might exert an effect on the expression of the enzymes involved in the GSH cycle.
1099
CHAPTER | 119 Extra Virgin Olive Oil Biophenols and mRNA Transcription of Glutathione-related Enzymes
mU mg–1 protein
10
Control LDL LDL+Prot LDL+Oleu
*
GR expression
A 160
* 140 % respect to the control
A 12
8 6 4
*
Control Cells + protocatechuic acid Cells + oleuropein
*
120 100 80 60 40 20
2
0
0
2h B
B 150
6h
12
12h GR activity ***
# #
100 75
10 mU mg–1 protein
% of control
125
24h
** 8
*
*
6 4
50 2
25
0 2h
6h
12h
24h
0 C 180 160 % respect to the control
FIGURE 119.5 Biophenols restore glutathione peroxidase (GPx) activity and gene expression in J774 A.1 cells. The biophenols were added to the cells for 2 h and then removed by renewing the medium before the addition of low-density lipoproteins (LDL) (0.2 g protein L⫺1). GPx (A) activity, and (B) mRNA expression evaluated by RT-PCR, were measured after 24 h incubation with LDL. Values are the mean ⫾S.E.M. (n ⫽ 4). Control ⫽ untreated cells; LDL ⫽ cells treated with LDL alone; LDL ⫹ Prot and LDL ⫹ Oleu ⫽ cells pre-treated with 25 μM protocatechuic acid or oleuropein, respectively. *p ⬍ 0.05 vs. control and LDL; # p ⬍ 0.05 vs. control.
**
*
GR protein ** *
140 120 100 80 60 40 20
119.4 EVOO BIOPHENOLS AND GENE EXPRESSION 119.4.1 mRNA Transcription Interesting results have been obtained by evaluating the action exerted by protocatechuic acid and oleuropein on the expression of GPx gene, demonstrating the overproduction of mRNA transcription in cells pre-treated with the biophenol as compared to controls (Masella et al., 2004) (Figure 119.5B). These findings have been confirmed in macrophages incubated with the specific biophenols alone. The results obtained clearly demonstrated a direct effect of the two biophenols on GR (Figure 119.6A, B, C) and in particular on GPx (Figure 119.7A, B, C), since they significantly increased gene and protein expressions within 2 h and the related enzyme activity within 6 h after exposure. These findings indicate a direct effect of the biophenols on the DNA transcription of GSH-related enzymes, and suggest that this is one of the mechanisms that improves antioxidant cellular defenses.
0 2h
6h
12h
FIGURE 119.6 Direct effect of protocatechuic acid and oleuropein on activity, gene expression and protein content of glutathione reductase (GR) in J774 A.1. Time-course evaluation of GR (A) mRNA by RT-PCR, (B) activity, (C) protein by Western blot. Cells were exposed for 2 h to 25 μM protocatechuic acid or oleuropein, then removed from the medium. Values (mean ⫾ S.E.M., n ⫽ 3) were compared to control ones for each time point considered. Control values were considered as 100% for mRNA and protein evaluation (A and C). *p ⬍ 0.05, **p ⬍ 0.01 and *** p ⬍ 0.001 vs. control.
Furthermore, these results are consistent with a bulk of published data which demonstrate the capability of different classes of biophenols to improve the endogenous antioxidant system through the direct activation of gene expression for detoxifying/antioxidant enzymes (Alia et al., 2006). Moskaug et al. have recently published interesting data demonstrating that the increase in GSH levels found in COS-1 and HepG2 cells treated with quercitin or onion extract depends on the increased transcription of γ-GCS gene (Moskaug et al., 2005).
1100
SECTION | II Cells and Cellular Effects
A 200
GPx expression
*** ***
Polyphenols ?
Control Cells + protocatechuic acid Cells + oleuropein
160 140 120
Polyphenols
% respect to the control
180
100 80 60 40
Nrf 2 Keap1
Nrf 2 ?
Cytoskeleton
MAPK cascade
Keap1
P Nrf 2
20 0 2h
6h
12h
24h
*** GPx activity
B 18
Maf c-Jun
mU mg–1 protein
16 14
**
**
12
*** **
P Nrf 2 *
10
ARE/EpRE
8 6
Nucleus
4 2 0 2h
6h
C 180
**
**
140 120
24h
GPx protein
160 % respect to the control
12h
*
*
100 80 60 40 20 0 2h
6h
12h
FIGURE 119.7 Direct effect of protocatechuic acid and oleuropein on activity, gene expression and protein content of glutathione peroxidase (GPx) in J774 A.1. Time-course evaluation of GPx (A) mRNA by RTPCR, (B) activity, (C) protein by Western blot. Cells were exposed to 25 μM protocatechuic acid or oleuropein for 2 h, then removed from the medium. Values (mean ⫾S.E.M., n ⫽ 3) were compared to control ones for each time point considered. Control values were considered as 100% for mRNA and protein evaluation (A and C). *p ⬍ 0.05, **p ⬍ 0.01 and *** p ⬍ 0.001 vs. control.
119.4.2 Antioxidant Responsive Elements and the Nrf2 Pathway Exogenous and endogenous antioxidants appear to act in a coordinated fashion and it is reasonable to hypothesize that this coordination is achieved, at least in part, through the antioxidant responsive elements (ARE), also called electrophile responsive elements (EpRE). They are cis-acting elements present in the promoter region of several genes mostly related to antioxidant/detoxifying activities. There is a fair amount of evidence that ARE/EpRE sequences play a pivotal role in the regulation of the cellular defense system,
FIGURE 119.8 Antioxidant/electrophile responsive elements (ARE/ EpRE) and NF E2-related factor 2 (Nrf2) pathway: interaction with biophenols. ARE/EpRE sequences are present in the promoter region of genes related to antioxidant/detoxifying activity. They are regulated by the cytoplasmatic Nrf2, which translocates to the nucleus and activates the ARE/EpRE sequences by strongly binding to them. Nrf2 is, in turn, regulated by Kelch-like ECH-associated protein1 (Keap1), which sequesters Nrf2 in the cytoplasm by forming heterodimers and inhibiting its translocation to the nucleus. Several mechanisms can be involved in the modulation of Nrf2-Keap1 binding, but phosphorylation induced by signaling kinases seems to be the major mechanism leading to free Nrf2 stabilization. Biophenols could influence the pathways that regulate ARE/EpRE activation at different stages. Biophenols may modify the capacity of Keap1 to sequester Nrf2, and/or activate mitogen activated protein kinase (MAPK) proteins (ERK, JNK and p38), and/or facilitate Nrf2 binding to ARE/EpRE sequences. Reprinted with permission from Masella et al., 2005.
being in turn strictly regulated by transcriptional factors, such as the NF E2-related factor 1 (Nrf1) and mainly NF E2-related factor 2 (Nrf2), ubiquitously expressed and belonging to the basic region leucine zipper superfamily (Nguyen et al., 2003). Nrf2 protein binds strongly to the ARE/EpRE sequences and positively regulates their activity (Figure 119.8). The interaction between Nrf2 and ARE/EpRE also involves several inhibiting or activating cofactors. It was demonstrated that Kelch-like ECH-associated protein1 (Keap1) – bound to actin protein and localized in the perinuclear space – sequesters Nrf2 in the cytoplasm by forming heterodimers and, inhibiting its translocation to the nucleus, renders it incapable of activating the ARE/EpRE sequences (Nguyen et al., 2005). The modulation of Keap-Nrf2 binding seems to be a central event in the cellular response to oxidative stress, even though the exact mechanism of dissociation of Nrf2 from its inhibitor, as well as the signal transduction pathway from oxidants to Nrf2-Keap1, remain largely
CHAPTER | 119 Extra Virgin Olive Oil Biophenols and mRNA Transcription of Glutathione-related Enzymes
unknown (Figure 119.8). Even though the involvement of superoxides and electrophiles as possible messengers in the oxidative stress pathway has not been demonstrated, there is some evidence that they, presumably passing through unknown cytosolic factors, may help in regulating Nrf2 release from Keap1 and its subsequent translocation into the nucleus. There are Nrf2 heterodimerizes with c-Jun or Small Maf or other unknown partners inducing ARE activation (Rahman, 2005).
1101
It is noteworthy that, as a consequence of Nrf2 knockdown, GPx and GR proteins significantly decreased in protocatechuic acid-treated macrophages, clearly indicating that the enzyme over-expression in these cells is, at least partially, Nrf2-dependent. (Varì et al., Central European Congress on Obesity. From Nutrition to Metabolic Syndrome. Karlovy Vary, Czech Republic, 2008; manuscript in preparation).
SUMMARY POINTS
119.4.3 EVOO Biophenols, the Nrf2 Pathway and ARE/EpRE
●
Phosphorylation seems to be a major mechanism in Nrf2 stabilization and involves several signaling kinases among which are mitogen-activated protein kinases (MAPKs), protein kinase C (PKC), and phosphatidylinositol-3-kinase (PI3K) (Sherratt et al., 2004). As a consequence of kinase activation, phosphorylated Nrf2 dissociates from Keap1 and translocates to the nucleus where Nrf2 protein accumulates (Figure 119.8). However, different signaling systems exist which modulate the upstream kinase signaling pathways. Biophenols influence the pathways that regulate ARE activation by modulating Nrf2 activity. This can be achieved by acting on different points of the sequence of events leading to the transcription of detoxifying enzymes. In particular, biophenols can influence the phosphorylation and stabilization of Nrf2 and/or its binding to the ARE sequences (Patel and Maru, 2008) (Figure 119.8). We have obtained interesting data about the effect exerted by protocatechuic acid on Nrf2 activation in J774 A.1 macrophages. A time-course treatment of J774 A.1 with the biophenol resulted in the transient up-regulation of Nrf2 mRNA level detected at 30’ (⫹30% with respect to the untreated control cells) which rapidly returned to the basal level. The biophenol induced a parallel, significant increase in the nuclear Nrf2 translocation that is required to regulate specific gene expression. Nuclear extracts from cells incubated with protocatechuic acid showed, in fact, a strong, transient up-regulation of Nrf2 level in the nucleus, reaching its maximum expression (⫹62%) at 2 h (Masella, R. et al., 4th International Conference on Polyphenols, Malta, 2007; manuscript in preparation). Nrf2 activation is likely responsible for the observed up-regulation of GPx and GR through the activation of ARE/EpRE. It is indeed worthy of note that the increase in Nrf2 level occurred before the significant increase in GPx and GR protein. To determine whether the activation of Nrf2 was responsible for the increase in GPx expression induced by protocatechuic acid, we knocked down Nrf2 by short interfering RNA (siRNA). Protocatechuic acid-treated cells transfected with the anti-Nrf2 siRNA showed a remarkable reduction of Nrf2 nuclear protein peak at 2 h (⫺70%; p ⱕ 0.05) with respect to the untransfected protocatechuic acid-treated cells.
●
●
●
●
●
Although most biophenols have antioxidant properties, these properties alone may not account for all their beneficial effects. Emerging findings suggest that biophenols have a variety of potential mechanisms of action in cytoprotection against oxidative stress, which may be independent of conventional antioxidant-reducing activities. Such mechanisms might entail the interaction of biophenols with cell signaling, and influence gene expression, and hence modulate specific enzymatic activities that drive the intracellular response against oxidative stress. We have demonstrated that biophenols contained in extra virgin olive oil induce an increased mRNA transcription of glutathione peroxidase and glutathione reductase genes in murine macrophages. As a consequence, glutathione is preserved from consumption and the endogenous antioxidant defenses are strengthened. The molecular mechanism responsible for the modulation of gene expression by EVOO biophenols is likely to involve the activation of ARE/EpRE obtained through the activation of the Nrf2 pathway.
ACKNOWLEDGMENTS This work was partially supported by grant 530/F18 ISS/NIH. The authors thank Ms. Monica Brocco for the linguistic revision of the manuscript.
REFERENCES Alia, M., Mateos, R., Ramos, S., Lecumberri, E., Bravo, L., Goya, L., 2006. Influence of quercetin and rutin on growth and antioxidant defense system of a human hepatoma cell line (HepG2). Eur. J. Nutr. 45, 19–28. Argyrou, A., Blanchard, J.S., 2004. Flavoprotein disulfide reductases: advances in chemistry and function. Prog. Nucleic. Acid Res. Mol. Biol. 78, 89–142. Beauchamp, G.K., Keast, R.S., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A., 2005. Phytochemistry: ibuprofen-like activity in extra-virgin olive oil. Nature 437, 45–46. Coni, E., Di Benedetto, R., Di Pasquale, M., Masella, R., Modesti, D., Mattei, R., Carlini, E.A., 2000. Protective effect of oleuropein, an olive oil biophenol, on low density lipoprotein oxidizability in rabbits. Lipids 35, 45–54.
1102
Covas, M.I., 2007. Olive oil and the cardiovascular system. Pharmacol. Res. 55, 175–186. Di Benedetto, R., Vari, R., Scazzocchio, B., Filesi, C., Santangelo, C., Giovannini, C., Matarrese, P., D’Archivio, M., Masella, R., 2007. Tyrosol, the major extra virgin olive oil compound, restored intracellular antioxidant defences in spite of its weak antioxidative effectiveness. Nutr. Metab. Cardiovasc. Dis. 17, 535–545. Filomeni, G., Rotilio, G., Ciriolo, M.R., 2005. Disulfide relays and phosphorylative cascades: partners in redox-mediated signaling pathways. Cell Death Differ. 12, 1555–1563. Fito, M., Cladellas, M., de la Torre, R., Marti, J., Alcantara, M., PujadasBastardes, M., Marrugat, J., Bruguera, J., Lopez-Sabater, M.C., Vila, J., Covas, M.I., 2005. Antioxidant effect of virgin olive oil in patients with stable coronary heart disease: a randomized, crossover, controlled, clinical trial. Atherosclerosis 181, 149–158. Fito, M., de la Torre, R., Covas, M.I., 2007. Olive oil and oxidative stress. Mol. Nutr. Food Res. 51, 1215–1224. Giovannini, C., Matarrese, P., Scazzocchio, B., Sanchez, M., Masella, R., Malorni, W., 2002. Mitochondria hyperpolarization is an early event in oxidized low-density lipoprotein-induced apoptosis in Caco-2 intestinal cells. FEBS Lett. 523, 200–206. Giovannini, C., Matarrese, P., Scazzocchio, B., Vari, R., D’Archivio, M., Straface, E., Masella, R., Malorni, W., De Vincenzi, M., 2003. Wheat gliadin induces apoptosis of intestinal cells via an autocrine mechanism involving Fas-Fas ligand pathway. FEBS Lett. 540, 117–124. Giovannini, C., Scazzocchio, B., Matarrese, P., Vari, R., D’Archivio, M., Di Benedetto, R., Casciani, S., Dessi, M.R., Straface, E., Malorni, W., Masella, R., 2008. Apoptosis induced by oxidized lipids is associated with up-regulation of p66Shc in intestinal Caco-2 cells: protective effects of phenolic compounds. J. Nutr. Biochem. 19, 118–128. Giovannini, C., Scazzocchio, B., Vari, R., Santangelo, C., D’Archivio, M., Masella, R., 2007. Apoptosis in cancer and atherosclerosis: polyphenol activities. Ann. Ist Super Sanita 43, 406–416. Giovannini, C., Straface, E., Modesti, D., Coni, E., Cantafora, A., De Vincenzi, M., Malorni, W., Masella, R., 1999. Tyrosol, the major olive oil biophenol, protects against oxidized-LDL-induced injury in Caco2 cells. J. Nutr. 129, 1269–1277. Hayes, J.D., Flanagan, J.U., Jowsey, I.R., 2005. Glutathione transferases. Annu. Rev. Pharmacol. Toxicol. 45, 51–88. Hofmann, T., Liegibel, U., Winterhalter, P., Bub, A., Rechkemmer, G., Pool-Zobel, B.L., 2006. Intervention with polyphenol-rich fruit juices results in an elevation of glutathione S-transferase P1 (hGSTP1) protein expression in human leucocytes of healthy volunteers. Mol. Nutr. Food Res. 50, 1191–1200. Hu, F.B., 2003. The Mediterranean diet and mortality – olive oil and beyond. N. Engl. J. Med. 348, 2595–2596. Lu, S.C., 2000. Regulation of glutathione synthesis. Curr. Top Cell Regul. 36, 95–116.
SECTION | II Cells and Cellular Effects
Masella, R., Cantafora, A., Modesti, D., Cardilli, A., Gennaro, L., Bocca, A., Coni, E., 1999. Antioxidant activity of 3,4-DHPEA-EA and protocatechuic acid: a comparative assessment with other olive oil biophenols. Redox Rep. 4, 113–121. Masella, R., Di Benedetto, R., Vari, R., Filesi, C., Giovannini, C., 2005. Novel mechanisms of natural antioxidant compounds in biological systems: involvement of glutathione and glutathione-related enzymes. J. Nutr. Biochem. 16, 577–586. Masella, R., Giovannini, C., Vari, R., Di Benedetto, R., Coni, E., Volpe, R., Fraone, N., Bucci, A., 2001. Effects of dietary virgin olive oil phenols on low density lipoprotein oxidation in hyperlipidemic patients. Lipids 36, 1195–1202. Masella, R., Vari, R., D’Archivio, M., Di Benedetto, R., Matarrese, P., Malorni, W., Scazzocchio, B., Giovannini, C., 2004. Extra virgin olive oil biophenols inhibit cell-mediated oxidation of LDL by increasing the mRNA transcription of glutathione-related enzymes. J. Nutr. 134, 785–791. Moskaug, J.O., Carlsen, H., Myhrstad, M.C., Blomhoff, R., 2005. Polyphenols and glutathione synthesis regulation. Am. J. Clin. Nutr. 81, 277S–283S. Nguyen, T., Sherratt, P.J., Nioi, P., Yang, C.S., Pickett, C.B., 2005. Nrf2 controls constitutive and inducible expression of ARE-driven genes through a dynamic pathway involving nucleocytoplasmic shuttling by Keap1. J. Biol. Chem. 280, 32485–32492. Nguyen, T., Sherratt, P.J., Pickett, C.B., 2003. Regulatory mechanisms controlling gene expression mediated by the antioxidant response element. Annu. Rev. Pharmacol. Toxicol. 43, 233–260. Patel, R., Maru, G., 2008. Polymeric black tea polyphenols induce phase II enzymes via Nrf2 in mouse liver and lungs. Free Radic. Biol. Med. 44, 1897–1911. Rahman, I., 2005. Regulation of glutathione in inflammation and chronic lung diseases. Mutat. Res. 579, 58–80. Santangelo, C., Vari, R., Scazzocchio, B., Di Benedetto, R., Filesi, C., Masella, R., 2007. Polyphenols, intracellular signalling and inflammation. Ann. Ist Super Sanita 43, 394–405. Sherratt, P.J., Huang, H.C., Nguyen, T., Pickett, C.B., 2004. Role of protein phosphorylation in the regulation of NF-E2-related factor 2 activity. Methods Enzymol. 378, 286–301. Tuck, K.L., Hayball, P.J., 2002. Major phenolic compounds in olive oil: metabolism and health effects. J. Nutr. Biochem. 13, 636–644. Ursini, F., Maiorino, M., Brigelius-Flohe, R., Aumann, K.D., Roveri, A., Schomburg, D., Flohe, L., 1995. Diversity of glutathione peroxidases. Methods Enzymol. 252, 38–53. Weinbrenner, T., Fito, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Albaladejo, M.F., Abanades, S., Schroder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321.
Chapter 120
Protective Effects of Olive Oil Components Against Hydrogen Peroxideinduced DNA Damage: The Potential Role of Iron Chelation Alexandra Barbouti1, Evangelos Briasoulis2 and Dimitrios Galaris1 1 2
Laboratory of Biological Chemistry, University of Ioannina Medical School, Ioannina, Greece Department of Oncology, University of Ioannina Medical School, Ioannina, Greece
120.1 INTRODUCTION Reactive oxygen species (ROS) are continuously generated and removed in all kinds of aerobic cells, thus creating an intracellular dynamic equilibrium which varies among different types of cells or even in the same cell type under different conditions (Halliwell and Cutteridge, 1999). Increased rates of generation or decreased capacity to remove ROS leads to elevated steady-state levels of these species, a phenomenon usually assigned as ‘oxidative stress’ (Halliwell and Cutteridge, 1999). Depending on the intensity of the oxidative stress applied, basic cellular functions such as cell proliferation and differentiation are modulated, while higher levels of oxidative stress can induce transient or permanent cell arrest or even cell death either by apoptosis or necrosis (Davies, 1999; Barbouti et al., 2002). Consequently, oxidative stress has been proposed to be implicated in pathogenic molecular mechanisms of a variety of diseases, including cardiovascular disease, ischemia–reperfusion syndrome, neurodegenerative diseases, cancer, and the physiological process of normal aging (Ames et al., 1993; Griendling and Harrison, 1999; Galaris and Pantopoulos, 2008; Galaris et al., 2008). Intensive research work has been performed in order to elucidate these mechanisms and to discover agents able to prevent or modulate the deleterious effects of oxidative stress that contribute to the development of these pathological complications. An unlimited number of potentially protective agents are believed to be present in diet and especially in the so-called ‘Mediterranean diet’, which has been shown to be able to prevent or impede the development of Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
diseases associated with oxidative stress (Keys et al., 1986; Trichopoulou and Lagiou, 1997; de Lorgeril and Salen, 2007). The prevailing general idea is that antioxidants and free radical scavengers present predominantly in this type of diet are mainly responsible for the protection of cells and tissues against the deleterious effects of oxidative stress. It has to be acknowledged though, that scientific support for this proposal is weak and the exact molecular base for the action of the bioactive compounds in the Mediterranean diet remains elusive and needs further investigation. In this presentation, we provide evidence that natural phenolic compounds present ubiquitously in Mediterranean diet and in olive oil exert their cytoprotective action mainly through chelation of intracellular redox-active iron, thus preventing the catalysis of the formation of deleterious free radicals near basic cell components, like DNA, proteins, and lipids.
120.2 MEDITERRANEAN DIET AND OLIVE OIL The observation in the middle of the previous century that rates of coronary heart disease were extremely low in Greece and other Southern Europe countries (as indicated in the famous Seven Countries Study) was proposed to be strongly connected to differences in dietary habits of these populations (Aravanis et al., 1970; Gerber, 1994). Subsequent epidemiological studies further supported the notion that the Cretan diet, which subsequently was used as a basis to form the worldwide known ‘Mediterranean diet’, contributed to a drastic reduction of diseases like coronary heart disease,
1103
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1104
stroke, type-2 diabetes, and cancer (Trichopoulou et al., 2003; Knoops et al., 2004; Willett, 2006). Key components of this diet are olive oil and olives (Visioli et al., 2005). The healthful properties of olive oil have been often attributed to the high levels of monounsaturated fatty acids (especially in the form of oleic acid) it contains. However, it is unlikely that oleic acid is exclusively responsible for the healthful effects of olive oil. Apart from being the source of fatty acids, olive oil also contains a remarkable amount of phenolic compounds which are regarded to be responsible for its resistance to oxidative rancidity (Blekas et al., 2002; Boskou et al., 2006). It has been repeatedly proposed that uptake of such compounds through consumption of olive oil may provide resistance toward oxidative stress, which, as noted above, is regarded as a major contributor to the development of a number of chronic diseases (Keys, 1987; Tuck and Hayball, 2002; Owen et al., 2004). It is generally believed that the above-mentioned phenolic compounds exert their beneficial
SECTION | II Cells and Cellular Effects
effects by acting as strong antioxidants and free radical scavengers (Visioli et al., 1998). However, despite the initial optimistic expectations, it was never proved that classical antioxidants were effective in preventing the development of diseases associated with the oxidative stress.
120.3 EVALUATION OF THE PROTECTION OFFERED BY EXOGENOUSLY ADDED COMPOUNDS AGAINST HYDROGEN PEROXIDE-INDUCED DNA DAMAGE Over several years, our research group has been involved in investigations aimed to elucidate the molecular mechanisms by which exposure of cells to oxidants, like H2O2, ONOO⫺, and HOCl, which are known to be generated in vivo, can induce oxidative damage to basic cellular constituents such
FIGURE 120.1 Induction of DNA damage in cells exposed to H2O2 and protection offered by iron-chelating compounds. (A) Jurkat cells in culture (1.5 ⫻ 106 cells mL⫺1) were exposed to increasing rates of H2O2 generated by the action of the enzyme glucose oxidase (GO) which was added directly into the growth medium. The final concentrations of GO ranged from 1 to 1000 ng mL⫺1 (able to generate between 0.02 to 20.0 μM H2O2 min⫺1). The DNA damage of individual cells was estimated 10 min after the exposure to H2O2 by using the so-called ‘comet assay’ methodology. The evaluation of DNA damage for the entire cell population was estimated by analyzing at least 100 nuclei as described in Panagiotidis et al., 1999. Each value represents the mean of duplicate measurements which differed less than 10%. (B) Conditions were identical as in (A), except that Jurkat cells were exposed to 0.6 μg mL⫺1 glucose oxidase (able to generate about 12.0 μM H2O2 min⫺1) in the presence (filled bars) or in the absence (empty bars) of 1.0 mM of each individual antioxidant added into culture medium 30 min before the exposure to H2O2. Ten minutes after the addition of glucose oxidase, DNA damage was determined by the comet assay as described above. One hundred percent DNA damage represents the degree of damage induced by H2O2 under the same conditions in the absence of any antioxidant pre-treatment. Each value represents the mean of triplicate measurements in two different experiments which differed less than 10%. (C) Conditions were as in (B) except that cells were pre-incubated with the indicated concentrations of 1,10-phenanthroline (empty bars) or 1,7-phenanthroline (filled bars) 30 min before their exposure to H2O2. (D) Conditions were as in (B) except that cells were pre-incubated with the indicated concentrations of salicylaldehyde isonicotinoyl hydrazone (SIH) (empty bars) or 2,2’-dipyridyl (filled bars) 30 min before their exposure to H2O2. AA: ascorbic acid, α-toc: α-tocopherol, α-LA: α-lipoic acid, GO: glucose oxidase; NAC: N-acetyl-cystein, SIH: salicylaldehyde isonicotinoyl hydrazone.
CHAPTER | 120 Protective Effects of Olive Oil Components Against Hydrogen Peroxide-induced DNA Damage
as proteins, lipids and DNA. It was soon realized that DNA was more sensitive compared to other cell components and for this reason more suitable as a marker for oxidantinduced effects on cells (Panagiotidis et al., 1999; Barbouti et al., 2001; Doulias et al., 2001). In these studies, we used a cell-culture-based experimental system in which human mononuclear cells or Jurkat cells (a T-lymphocytic cell line) were exposed to a continuously generated flow of H2O2 by the action of the enzyme ‘glucose oxidase’ which was added directly into the growth medium. Single-strand breaks were rapidly apparent in the cellular DNA as indicated by formation of the characteristic comet-like tails after gel electrophoresis of DNA from individual nuclei (Panagiotidis et al., 1999). DNA damage was not homogeneous in individual cells as indicated by the percentage of DNA in the tail and the distance moved from the central DNA head. The overall DNA damage of the cell population was expressed in arbitrary units based on evaluation of 100 single cells as described in Panagiotidis et al. As shown in Figure 120.1A, DNA damage increased along with the rate of H2O2 generation in which cells were exposed. We estimated that a flow of H2O2 equal to 20.0 ⫾ 2.0 μM min⫺1 generated by the action of 1.0 μg mL⫺1 glucose oxidase was able to induce the maximum of damage that could be detected by this particular method. So, an amount of 0.6 μg mL⫺1 of glucose oxidase (able to generate 12.0 ⫾ 1.2 μM H2O2 min⫺1) was used in the rest of the experiments.
120.4 THE ROLE OF IRON IN HYDROGENPEROXIDE-INDUCED DNA DAMAGE After establishing the methodology, we used this system in experiments intending to evaluate the potential of different compounds to protect cellular DNA. Traditional antioxidants, like ascorbic acid, α-tocopherol, trolox, N-acetyl-cystein and α-lipoic acid, when added 30 min before the addition of the H2O2 at 1.0 mM concentrations offered slight or no protection at all, indicating the inability
1105
of these compounds to effectively scavenge reactive free radicals generated inside the cells (Figure 120.1B). On the contrary, all iron chelators tested offered strong protection, on the condition that they could penetrate plasma membrane (Figure 120.1C, D, and results not shown). A characteristic experiment is presented in Figure 120.1C, where 1,10-phenanthroline offered significant protection, while 1,7-phenanthroline was completely ineffective. The chemical structures of the two compounds differ only in the position of one nitrogen atom. This difference, however, makes 1,7-phenantroline unable to chelate iron, in contrast to 1,10phenanthroline which is a strong iron-chelator. It became apparent from these results that the prevention of cellular DNA oxidation is due to chelation of intracellular labile iron. This conclusion was further supported by experiments with several other iron-chelating compounds, two of which (2,2-dipyridyl and salicylaldehyde isonicotinoyl hydrazone (SIH)) are presented in Figure 120.1D (for more information about iron and iron-chelators, see Table 120.1). Taken together, these experiments indicate that H2O2 induces DNA damage through iron-mediated formation of hydroxyl radicals in Fenton-type reactions that take place on or very close to DNA. This postulation is based on the fact that hydroxyl radicals are extremely reactive species and interact exclusively in the vicinity of their generation. Thus, exogenously added antioxidants are impossible to be present at the right place and in such a high concentration in order to compete with all potential cellular reactants for interaction with hydroxyl radicals. On the other hand, it is plausible to assume that chelation of intracellular redoxactive iron and inhibition of hydroxyl radical generation represents an effective mechanism for protection of cell constituents in conditions of oxidative stress. It has to be noted here, that apart from DNA, other cellular components like proteins, lipids and carbohydrates may also bind iron in its redox-active form and in this way become vulnerable for iron-mediated oxidations. Although not proved directly, it is plausible to imagine that removal of iron from these positions can prevent their oxidation.
TABLE 120.1 Iron and iron chelators. • Iron represents the most abundant transition metal in the human body and plays crucial roles in several key functions, like transport of oxygen and electrons, synthesis of DNA, detoxication of xenobiotics, etc. • Apart from its beneficial roles, iron can exert injurious effects by catalyzing the formation of extremely reactive free radicals, which have been proposed to be implicated in numerous pathological conditions • Iron chelators are chemical compounds that are capable of binding ‘free’ (redox-active) iron in biological systems, thus forming relatively stable complexes and in this way preventing its deleterious actions under conditions of oxidative stress This table lists the main facts regarding the role of iron in pathological and physiological conditions. Decreasing the concentration of labile iron inside cells by using specific iron-chelating compounds can make tissues resistant to oxidative stress-induced damage.
1106
A
DNA damage (arbitrary units)
In the next set of experiments, we used the above-described system in order to search if extracts derived from natural sources contained compounds able to protect cellular DNA from damage. Cells were treated with the tested extracts from natural sources (or isolated pure compounds) for 30 min before exposed to H2O2 as described above, and the protective effects were evaluated from their ability to decrease the DNA damage. Results from such typical experiments are presented in Figure 120.2. It was observed that polar extracts from olive oil offered considerable protection, indicating that they contain components that can penetrate plasma membranes and interrupt the reaction(s) that lead to DNA damage (Figure 120.2A). It has to be noted, however, that apart from offering a variable degree of protection, the same extracts at higher concentrations induced DNA damage by themselves in the absence of H2O2, indicating that some of their components were genotoxic at higher concentrations (results not shown). Surprising enough, extracts from olive oil mill wastewater also offered protection against H2O2-induced DNA damage, although at a lesser extent, as indicated from the fact that higher amounts were needed in order to exert the same degree of protection compared to olive oil extracts (Figure 120.2B). It has to be noted that a number of protective compounds present in olive fruit are amphiphilic in nature and more soluble in the water than in the oil phase. Consequently, the main part of them will be lost with the wastewater during processing, while only about 1–2% of the available pool remains in olive oil. In an attempt to identify individual compounds in olive oil that contributed to DNA protection, the main components of olive oil extracts were isolated and tested separately in the above-described experimental system. Several compounds contained in olive oil and olive mill wastewater offered considerable protection (Nousis et al., 2005). Prominent among them were hydroxytyrosol (2(3,4-dihydroxyphenyl)ethanol), caffeic acid and some specific flavonoids. On the other hand, tyrosol (2-(4hydroxyphenyl)ethanol) that lacks only one extra hydroxyl group at the ortho-position compared to hydroxytyrosol was shown to be ineffective, while oleuropein induced DNA damage in the absence of H2O2 (Nousis et al., 2005). These observations could not be explained on the basis of the antioxidant capacities of the tested compounds. Based on the results described in the previous section, we decided to test the possibility that the ability of phenolic constituents of olive oil exert their protective effects through binding intracellular ‘labile iron’ and in this way preventing the generation of extremely reactive hydroxyl radicals. So, a structure–activity relationship study was performed by using a number of flavonoids with closely related chemical structures.
DNA damage (arbitrary units)
120.5 PROTECTION OFFERED BY COMPOUNDS PRESENT IN OLIVE OIL
SECTION | II Cells and Cellular Effects
300
200
100
0
0
1 10 50 100 Olive oil (μg mL–1)
B
300
200
100
0
0 0.1 1 2 5 Olive mill wastewater (mg mL–1)
FIGURE 120.2 Protection offered by polar extracts from olive oil (A) and olive oil mill wastewater (B). Conditions were as in Figure 120.1B, except that cells were pre-incubated for 30 min with the indicated concentrations of extracts from olive oil (A) or olive mill wastewater (B) before exposure to H2O2. Empty bars represent the basal level of singlestrand breaks observed in the cells in the absence of H2O2 while filled bars indicate the DNA damage after exposure to H2O2. Each value represents the mean of triplicate measurements in two different experiments which differed less than 10%.
120.6 FLAVONOIDS PROTECT CELLS BY CHELATING INTRACELLULAR IRON Flavonoids represent a group of plant-derived phenolic compounds that are ubiquitously distributed in the main foods consumed in the Mediterranean area, including olive oil. In order to investigate in detail the molecular mode of action of this type of phenolic compounds, we estimated the capacity of a number of flavonoids with closely related chemical structures (Figure 120.3) to protect nuclear DNA from damage induced when Jurkat cells were exposed to H2O2 (Melidou et al., 2005). By repeating the same experiment in the presence or absence of increasing concentrations of each individual flavonoid we were able to estimate the concentration of each one that offered 50% protection to DNA (IC50 value). The capacity of individual flavonoids to protect DNA was highly variable, with minimal structural changes exerting profound effects on their effectiveness. It was concluded from this kind of experiments that the chemical characteristics that contributed toward increasing the protective capacity were: (a) the presence of an ortho-dihydroxy group, (b) the presence of a hydroxyl group next to an alcoxy group, and (c) the presence of an extended network of conjugated double bonds inside the molecule. Previous studies have shown that all these chemical characteristics contributed also toward increasing the free radical scavenging capacity of flavonoids (Rice-Evans et al., 1996). A first indication that intracellular chelation of loosely bound iron was responsible for the observed protection of DNA came from experiments in which the flavonoids used were gradually saturated with iron before their addition in
CHAPTER | 120 Protective Effects of Olive Oil Components Against Hydrogen Peroxide-induced DNA Damage
Flavones
3' 2' 8 7 A
4' B
O
5' 6'
C
6
3 5 O
Flavone 5-hydroxyflavone 7-hydroxyflavone Chrysin 7,8-dihydroxyflavone Baicalein Apigenin Luteolin
1107
OH-substitution 5 7 5,7 7,8 5,6,7 5,7,4´ 5,7,3´,4´
Flavonoles O
OH O
3-hydroxyflavone Galangin Kaempferol Fisetin Morin Quercetin Myricetin
3 3,5,7 3,5,7,4´ 3,7,3´,4´ 3,5,7,2´,4´ 3,5,7,3´,4´ 3,5,7,3´,4´,5´
Flavanones
O
6-hydroxyflavanone Narigenin Eriodictyol Taxifolin
6 5,7,4´ 5,7,3´,4´ 3,5,7,3´,4´
O
O
OH
Flavan-3-oles Catechin
3,5,7,3',4'
FIGURE 120.3 Flavonoids. Chemical formulas of the various classes of flavonoids and the positions of OH-substitutions in the compounds used in this study.
the culture medium. It was shown that the protective ability of flavonoids was decreased in parallel with their saturation with iron, indicating that their action was dependent on their capacity to bind iron inside the cells (Melidou et al., 2005). In subsequent experiments we estimated the ironbinding capacity of individual flavonoids in situ by using the so-called ‘calcein method’. Cells were loaded with calcein and their fluorescence was detected in a fluorometer. Part of calcein fluorescence in this system was quenched due to its binding to intracellular ‘chelatable’ iron. The amount of such an iron pool was subsequently estimated from the increase in the fluorescence following the addition of a strong and specific iron chelator, namely salicylaldehyde isonicotinoyl hydrazone (SIH), which effectively removed iron from calcein. The potential of individual flavonoids to remove iron from calcein (indicated as EC50) was characteristic for each compound, but more importantly it was positively correlated with their ability to protect cells from H2O2-induced DNA damage (Figure 120.4A). On the
other hand, no correlation was observed between the values of antioxidant capacity of the flavonoids used (as estimated by a common commercial method) and their ability to protect cellular DNA (Figure 120.4B). Taken together, these observations clearly indicate that H2O2-induced DNA damage is mediated by redox-active iron and the protection offered by individual flavonoids was due to the capacity of flavonoids to chelate this iron.
120.7 CONCLUDING REMARKS Oxidative stress has been proposed to be involved in the molecular mechanisms of almost any pathological condition in animals and humans (Ames et al., 1993; Griendling and Harrison, 1999; Galaris and Pantopoulos, 2008; Galaris et al., 2008). Consequently, intensive research work has been performed in order to discover agents able to prevent or cure the deleterious effects of oxidative stress.
1108
SECTION | II Cells and Cellular Effects
500
500 400 ρ = 0.9
300
IC50 (μM)
IC50 (μM)
400
200 100
300 200 100
0
0 1.5
2.5 3.5 4.5 Antioxidant capacity A B (mM Trolox) FIGURE 120.4 Correlation of the iron-binding capacity of flavonoids with their ability to protect cellular DNA. Correlation of the ability of individual flavonoids to protect cellular DNA from H2O2-induced damage (IC50) with their capacity to chelate intracellular labile iron in intact cells (EC50) (A) but not with their antioxidant capacity (B). EC50: the concentration of the compound that chelated 50% of the intracellular iron; IC50: the concentration of the compound that offered 50% protection. 0
100
200
300
400
EC50 (μM)
Numerous such potential protective agents are believed to be present in the Mediterranean diet which is associated with reduced incidence of a number of pathological complications, especially coronary heart diseases, diabetes, certain tumors, and the aging process (Keys et al., 1987; Trichopoulou and Lagiou, 1997). The prevailing idea today is that antioxidants and free radical scavengers present predominantly in the Mediterranean diet are mainly responsible for the observed beneficial effects. It has to be stressed, however, that the experimental support for this postulation is still weak and the exact molecular base for the action of bioactive compounds contained in the Mediterranean diet remains elusive. Based on experiments performed in our as well as other laboratories, we propose that the main mode of the cytoprotective action of natural bioactive agents, such as flavonoids and other phenolic compounds present in the Mediterranean diet, is due mainly to their capacity to penetrate through biological membranes and to chelate intracellular redox-active iron (labile iron), thus preventing the formation of extremely reactive free radicals. The importance of redox-active iron in mediating oxidative stressinduced cell and tissue injuries and its contribution to development of common diseases has been highly appreciated in recent years (see reviews, Hentze et al., 2004; Richardson, 2004; Papanikolaou and Pantopoulos, 2005; Galaris and Pantopoulos., 2008). It has to be noted here, that apart from DNA iron is also loosely bound on a variety of other cellular ligands, including proteins, lipids, and carbohydrates. Iron-mediated oxidation of such ligands is thought to play important roles in a number of molecular signaling pathways that regulate basic cellular functions (Galaris and Pantopoulos, 2008; Takeda et al., 2008). Thus, it is tempting to speculate that flavonoids and other related phenolic natural compounds are able to modulate the above pathways just by binding and relocating intracellular iron ions. It is plausible to imagine that such actions should have profound effects on basic cellular functions that are dependent on these signaling pathways.
SUMMARY POINTS ●
●
●
●
Oxidative stress is implicated in the molecular mechanisms of numerous diseases, including cardiovascular disease, ischemia–reperfusion syndrome, neurodegenerative diseases, cancer, and the physiological process of normal aging. The Mediterranean diet contains an unlimited number of compounds that act in a beneficial way regarding the development of several chronic diseases. Polar extract preparations from olive oil, which represents an important component of the Mediterranean diet, contain compounds able to protect cellular DNA in conditions of oxidative stress. The capacity of natural phenolic compounds contained in olive oil to chelate iron inside the cells (but not their antioxidant capacity) is responsible for their cytoprotective properties.
REFERENCES Ames, B.N., Shigenaga, M.K., Hagen, T.M., 1993. Oxidants, antioxidants, and the degenerative diseases of aging. Proc. Natl. Acad. Sci. USA 90, 7915–7922. Aravanis, C., Corcondilas, A., Dontas, A.S., Lekos, D., Keys, A., 1970. Coronary heart disease in seven countries. IX. The Greek islands of Crete and Corfu. Circulation 41, 100–188. Barbouti, A., Doulias, P.T., Nousis, L., Tenopoulou, M., Galaris, D., 2002. DNA damage and apoptosis in hydrogen peroxide-exposed Jurkat cells, bolus addition versus continuous generation of H2O2. Free Radic. Biol. Med. 33, 691–702. Barbouti, A., Doulias, P.T., Zhu, B.Z., Frei, B., Galaris, D., 2001. Intracellular iron, but not copper, plays a critical role in hydrogen peroxide-induced DNA damage. Free Radic. Biol. Med. 31, 490–498. Blekas, G., Vassilakis, C., Harizanis, C., Tsimidou, M., Boskou, D.G., 2002. Biophenols in table olives. J. Agric. Food Chem. 50, 3688–3692. Boskou, G., Salta, F.N., Chrysostomou, S., Mylona, A., Chiou, A., Andrikopoulos, N.K., 2006. Antioxidant capacity and phenolic profile of table olives from the Greek market. Food Chem. 94, 558–564.
CHAPTER | 120 Protective Effects of Olive Oil Components Against Hydrogen Peroxide-induced DNA Damage
Davies, K.J., 1999. The broad spectrum of responses to oxidants in proliferating cells, a new paradigm for oxidative stress. IUBMB Life 48, 41–47. de Lorgeril, M., Salen, P., 2007. Modified cretan Mediterranean diet in the prevention of coronary heart disease and cancer. An update. World Rev. Nutr. Diet. 97, 1–32. Doulias, P.T., Barbouti, A., Galaris, D., Ischiropoulos, H., 2001. SIN1-induced DNA damage in isolated human peripheral blood lymphocytes as assessed by single cell gel electrophoresis (comet assay). Free Radic. Biol. Med. 30, 679–685. Galaris, D., Pantopoulos, K., 2008. Oxidative stress and iron homeostasis, mechanistic and health aspects. Crit. Rev. Clin. Lab. Sci. 45, 1–23. Galaris, D., Mantzaris, M., Amorgianiotis, C., 2008. Oxidative stress and aging, the potential role of iron. Hormones (Athens) 7, 114–122. Gerber, M., 1994. Olive oil and cancer. In: Hill, M.J., Giacosa, A., Caygill, C.P.G. (eds), Epidemiology of Diet and Cancer. Ellis Horwood, Chichester, pp. 267–275. Griendling, K.K., Harrison, D.G., 1999. Dual role of reactive oxygen species in vascular growth. Circ. Res. 85, 562–563. Halliwell, B., Cutteridge, J.C. (eds), 1999. Free Radicals in Biology and Medicine. Oxford University Press, Oxford, UK. Hentze, M., Muckenthaler, M., Andrews, N., 2004. Balancing acts: molecular control of mammalian iron metabolism. Cell 117, 285– 297. Keys, A., 1987. Olive oil and coronary heart disease. Lancet 1, 983–984. Keys, A., Menotti, A., Karvonen, M.J., Aravanis, C., Blackburn, H., Buzina, R., Djordjevic, B.S., Dontas, A.S., Fidanza, F., Keys, M.H., et al., 1986. The diet and 15-year death rate in the seven countries study. Am. J. Epidemiol. 124, 903–915. Knoops, K.T., de Groot, L.C., Kromhout, D., Perrin, A.E., MoreirasVarela, O., Menotti, A., van Staveren, W.A., 2004. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women, the HALE project. JAMA 292, 1433–1439. Melidou, M., Riganakos, K., Galaris, D., 2005. Protection against nuclear DNA damage offered by flavonoids in cells exposed to hydrogen
1109
peroxide, the role of iron chelation. Free Radic. Biol. Med. 39, 1591–1600. Nousis, L., Doulias, P.T., Aligiannis, N., Bazios, D., Agalias, A., Galaris, D., Mitakou, S., 2005. DNA protecting and genotoxic effects of olive oil related components in cells exposed to hydrogen peroxide. Free Radic. Res. 39, 787–795. Owen, R.W., Haubner, R., Wurtele, G., Hull, E., Spiegelhalder, B., Bartsch, H., 2004. Olives and olive oil in cancer prevention. Eur. J. Cancer Prev. 13, 319–326. Panagiotidis, M., Tsolas, O., Galaris, D., 1999. Glucose oxidase-produced H2O2 induces Ca2⫹-dependent DNA damage in human peripheral blood lymphocytes. Free Radic. Biol. Med. 26, 548–556. Papanikolaou, G., Pantopoulos, K., 2005. Iron metabolism and toxicity. Toxicol. Appl. Pharmacol. 202, 199–211. Rice-Evans, C.A., Miller, N.J., Paganga, G., 1996. Structure-antioxidant activity relationships of flavonoids and phenolic acids. Free Radic. Biol. Med. 20, 933–956. Richardson, D.R., 2004. Mysteries of the transferrin-transferrin receptor 1 interaction uncovered. Cell 116, 483–485. Takeda, K., Noguchi, T., Naguro, I., Ichijo, H., 2008. Apoptosis signalregulating kinase 1 in stress and immune response. Annu. Rev. Pharmacol. Toxicol. 48, 199–225. Trichopoulou, A., Lagiou, P., 1997. Healthy traditional Mediterranean diet, an expression of culture, history, and lifestyle. Nutr. Rev. 55, 383–389. Trichopoulou, A., Costacou, T., Bamia, C., Trichopoulos, D., 2003. Adherence to a Mediterranean diet and survival in a Greek population. N. Engl. J. Med. 348, 2599–2608. Tuck, K.L., Hayball, P.J., 2002. Major phenolic compounds in olive oil, metabolism and health effects. J. Nutr. Biochem. 13, 636–644. Visioli, F., Bellomo, G., Galli, C., 1998. Free radical-scavenging properties of olive oil polyphenols. Biochem. Biophys. Res. Commun. 247, 60–64. Visioli, F., Bogani, P., Grande, S., Galli, C., 2005. Mediterranean food and health, building human evidence. J. Physiol. Pharmacol. 56, 37–49. Willett, W.C., 2006. The Mediterranean diet, science and practice. Public Health Nutr. 9, 105–110.
This page intentionally left blank
Chapter 121
Olive Oil Phenols and Nitric Oxide Affect Lymphomonocyte Cytosolic Calcium Giuseppe Arienti, Michela Mazzoni and Carlo A. Palmerini Dipartimento di Medicina Interna, Perugia 06127, Italy
121.1 INTRODUCTION The nutritional value of any foods is not limited to their content in essential nutrients or in nutrients generally speaking. Of course, food is a source of energy and essential nutrients, but it is also a mixture of many molecules, some in high, low, or very low amounts. These may interact in various ways and may influence human health in yet unpredictable manners either in the short or long term. For this reason, efforts are being made towards the understanding of the physiological and medical significance of food. The chief constituent of alimentary fat is triacylglycerol. Various triacylglycerol molecules may differ in the proportion and position of the fatty acids they contain. Much work has been done on this subject, especially with reference to polyunsaturated essential fatty acid. However, olive oil contains a number of molecules besides triacylglycerols. This review is dedicated to some of these, namely to phenols.
121.3 CYTOSOLIC CALCIUM
121.2 PHENOLS Typical components of the Mediterranean diet, such as olive oil and red wines, contain high concentrations of phenols that may have important antioxidant roles. The main phenols identified in extra virgin olive oil belong to different classes: simple phenols such as 3,4-(dihydroxyphenyl)ethanol (or hydroxythyrosol, HT) or p-(hydroxyphenyl)ethanol (or thyrosol, T) and secoiridoids (oleuropein, the aglycone of ligstroside and their respective decarboxylated dialdehyde derivatives) (Servili et al., 1999; Owen et al., 2000). Epidemiological studies suggest that phenols reduce the incidence of coronary heart disease (Visioli et al., 2002). Phenols help confer its peculiar taste to olive oil. Besides, they may exert a number of effects on some important parameters of living cells which may account for the above-reported beneficial effects. We think that the study of the connections between epidemiological effects and the Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
biochemical action of phenols is important for several reasons. Technically speaking, oils can be prepared to contain higher or lower amounts of phenols; moreover, these substances could be used to enrich food or as supplements for the prevention of important illnesses, such as coronary heart disease. The mechanism(s) responsible for the effects of phenols have not yet been fully clarified. Antioxidant activity (Visioli et al., 2001) and free radical scavenging (Visioli et al., 1998) may be relevant in this connection. Yet, other hypotheses have been put forward. For instance, it has been reported that red wine phenols increase cytosolic calcium ([Ca2⫹]c) in bovine endothelial cells, similarly to bradikinin and to ATP (Andriantsitohaina et al., 1999). In this short review, we report the effect of extra virgin olive oil phenols on cytosolic calcium [Ca2⫹]c variations due to nitric oxide (NO).
The variation of cytosolic Ca2⫹ concentration ([Ca2⫹]c) is an important signal in either excitable or non-excitable cells, including immune cells, and regulates some fundamental processes such as cell activation, growth and differentiation (Gardner, 1989; Clapham, 1995). The homeostasis of [Ca2⫹]c in mammalian cells is a complex phenomenon requiring careful regulation by many mechanisms and entailing the participation of several cellular and extracellular compartments. The endoplasmic reticulum (ER) is the major site for intracellular calcium storage and it is involved in calcium signaling and in the maintenance of intracellular calcium homeostasis (Rao et al., 2001). Calcium is pumped from the cytoplasm into the ER by the action of a Ca2⫹-dependent ATPase (SERCA); on the other hand, inositol 1,4,5trisphosphate (IP3) channels release calcium from the ER (and other cellular compartments) to cytosol. Therefore,
1111
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1112
the movement of calcium between ER and cytosol is hormonally regulated. In turn, a decrease of ER calcium would stimulate the entry of the ion from the extracellular milieu. This process, also known as capacitative entry, is permitted by receptors located on the cellular plasma membrane. In excitable cells, these receptors are voltage-dependent (voltage-operated calcium channels; VOCCs) (Catterall et al., 1992; Berger et al., 1994; Guillermet et al., 2006) and in non-excitable cells they are operated by the amount of stored calcium (store-operated calcium channels; SOCCs) (Nusse et al., 1997). The ER is highly sensitive to alterations of calcium concentrations and a number of pharmacological agents can induce ER stress (Denmeade et al., 2005). Among these, some drugs disrupt ER calcium stores through the inhibition of SERCA, such as thapsigargin (TG) (Treiman et al., 1998). The entry of calcium external to the cell is essential for the maintenance of [Ca2⫹]c homeostasis because the spontaneous loss of calcium from the ER is but partially restored by the reuptake from cytosol (Leung et al., 1999). In addition, the depletion of calcium stores can trigger the uptake of extracellular calcium via store-operated Ca2⫹ channels (SOCCs) (Nusse et al., 1997) that can also be activated by TG (Treiman et al., 1998). Some dihydropyridine derivatives specifically bind with high affinity to voltage-operated Ca2⫹ channels (VOCCs) in excitable cells and (with less affinity) to store-operated calcium channels (SOCCs) in non-excitable cells (Catterall et al., 1992; Berger et al., 1994; Guillermet et al., 2006). Among these, nifedipine (NIF) inhibits the opening of the channel, so impairing the calcium influx through it (Berger et al., 1994; Kochegarov, 2003). Therefore, TG and NIF have been employed to study the movements of Ca2⫹ into and from the ER (Leung et al., 1996) and to better understand the effects of phenols and NO on this phenomenon (Palmerini et al., 2005a, b).
121.4 NITRIC OXIDE Nitric oxide (NO) has received much attention because of its widespread biological effects. Its short half-life, the opposite effects seen at low and at high concentrations (Dedkova and Blatter, 2002) and the possibility of regulating cellular functions through protein nitrosylation/denitrosylation cycles make this compound interesting, albeit difficult to study (Beckman and Koppenol, 1996; Patel et al., 1999). NO influences [Ca2⫹]c in several cell types, but its mechanism(s) of action is (are) still a matter for debate (Xu et al., 1994; Clementi, 1998; Nagy et al., 2003). The complexity of this topic is increased further by the biphasic behavior of NO that may show opposite effects on some biological systems, depending on its concentration (Dedkova and Blatter, 2002; Palmerini et al., 2005b). Various NO donors have been used to release NO in cellular preparations; S-nitrosocysteine (CysNO) rapidly
SECTION | II Cells and Cellular Effects
donates NO, where other donors have slower kinetics of NO release (Arnelle et al., 1995; Williams, 1999). This property of CysNO is necessary to follow the rapid variations of [Ca2⫹]c. CysNO concentrations around 16 μM are comparable to a NO signal and concentrations about 160 μM would produce toxic effects as a result of peroxynitrite formation following the reaction of NO with superoxide (Beckman et al., 1996; Boccini et al., 2004; Palmerini et al., 2008). NO decreases [Ca2⫹]c levels in lymphomonocytes (Figure 121.1). The decrease of [Ca2⫹]c is dose-dependent and induces the entry of calcium from the extracellular milieu to maintain cellular homeostasis. SERCA pump inhibition by TG increases [Ca2⫹]c showing an effect opposite to that of CysNO (Figure 121.2). As indicated by Figures 121.1 and 121.2, CysNO improves the exit of calcium from cytosol in agreement with findings on other cellular systems (Chen et al., 2000), whereas TG blocks the uptake (reuptake) of the ion in the ER (Treiman et al., 1998; Palmerini et al., 2008). As expected, CysNO inhibits the effects of TG when both substances are present (Figure 121.2). Upon the addition of Ca2⫹ to the external milieu, TG markedly raises [Ca2⫹]c as a consequence of the depletion of ER calcium following the inhibition by SERCA. This effect of TG is similar to that of CysNO, even for low CysNO concentration (Chen et al., 2000) (Figure 121.2). Although the mechanism of action of NO opposes that of TG, both substances cooperate to maintain the calcium homeostasis by activating the capacitative entry. NIF has been reported to decrease the movement of Ca2⫹ from the cytosol to ER by inhibiting the transport of calcium through the SOCCs channels from the extracellular milieu in non-excitable cells (Kochegarov, 2003; Guillermet et al., 2006). On the other hand, NIF decreases the entry of calcium induced by NO (16–160 μM CysNO) (Palmerini et al., 2008).
121.5 MOLECULAR EFFECTS OF PHENOLS Red wine phenols increase [Ca2⫹]c, so inducing NO production in endothelial cells because endothelial NO synthase is Ca2⫹-dependent (Visioli et al., 2002); the inhibition of calcium entry into cytosol is related to NOS inhibition in acinar rat pancreatic cells (Xu et al., 1994). In human lymphocytes, the exposure to SIN-1, which generates superoxide and NO, results in DNA damage; on the other hand, calcium-chelating agents decrease the damage, so indicating a role of calcium in the process (Doulias et al., 2001). The molecular effects of phenols are scarcely known. Few data about the effects on [Ca2⫹]c have been reported in vitro on human lymphomonocyte cell preparations (De la Rosa et al., 2001). Although these cells are used as a model for non-excitable cells, the homeostasis of [Ca2⫹]c in the preparation is very complex and has been studied with
CHAPTER | 121 Olive Oil Phenols and Nitric Oxide Affect Lymphomonocyte Cytosolic Calcium
No [Ca2+]
140
1113
1 mM [Ca2+]
120
[Ca2+], (nM)
100 80
CysNO
60 40 20
CysNO 2,0
50
100
150
200
250
300
350
400
450
500
Time (s) FIGURE 121.1 Representative example of the variations of lymphomonocyte cytosolic calcium concentration ([Ca2⫹]c) in the presence of either 16 (dotted line) or 160 (continuous line) μM S-nitrosocysteine (CysNO). No calcium was present in the external medium. At the indicated point external calcium was brought to 1 mM. [Ca2⫹]c was monitored as described by Grynkiewicz et al. (1985) using the fluorescent probe FURA-2AM. [Ca2⫹]c ⫽ concentration of cytosolic calcium; CysNO: S-nitrosocysteine. Redrawn from Palmerini et al., 2008, with kind permission of Wiley & Sons, Inc. 200
No [Ca2+]
1 mM [Ca2+]
180 160
[Ca2+], (nM)
140 120 100 80
A
60 40 20 0 50
100
150
200
250
300
350
400
450
Time (s) FIGURE 121.2 Representative example of the variations of lymphomonocyte cytosolic calcium concentration ([Ca2⫹]c) in the presence of 1 μM thapsigargin (TG) and S-nitrosocysteine (CysNO). Continuous line no CysNO, dashed line 16 μM CysNO, dotted line 160 μM CysNO. No calcium was present in the external medium. At the indicated point external calcium was brought to 1 mM. [Ca2⫹]c was monitored as described by Grynkiewicz et al. (1985) using the fluorescent probe FURA-2AM. [Ca2⫹]c ⫽ concentration of cytosolic calcium; CysNO: S-nitrosocysteine; TG: thapsigargin. Redrawn from Palmerini et al., 2008, with kind permission of Wiley & Sons, Inc.
the help of drugs, such as TG and NIF, enable to inhibit selectively some parts of the mechanism of calcium control. Olive oil phenols (HT and T) exert effects on [Ca2⫹]i in human lymphomonocyte populations (Partiseti et al., 1994; de la Rosa et al., 2001). HT is more efficient than T at increasing the level of [Ca2⫹]i in human lymphomonocytes (Palmerini et al., 2005a) (Figure 121.3). Phenols increase [Ca2⫹]c by stimulating calcium exit from the RE. The so-induced depletion of RE calcium activates capacitative calcium entry. This effect is antagonized by NIF that
inhibits the entry of the ion from the extracellular milieu (inhibition of SOCCs) (Guillermet et al., 2006).
121.6 PHENOL–NITRIC OXIDE INTERACTION Both T and HT increase [Ca2⫹]c in lymphomonocytes in a dose-dependent way (Palmerini et al., 2005a). The concentrations of phenolics used to study the biological effects of these substances is (60–240 μmol L–1) (Palmerini et al., 2005a)
1114
SECTION | II Cells and Cellular Effects
80 HT HT+ 200 μM CysNO HT+ 500 μM CysNO
70
Δ Ca2+ (nM)
60 50 40 30 20 10 0 –10 40
FIGURE 121.3 Effects of hydroxythyrosol (HT) and of nifedipine (NIF) on lymphomonocyte cytosolic calcium concentration ([Ca2⫹]c). [Ca2⫹]c was monitored as described by Grynkiewicz et al. (1985) using the fluorescent probe FURA-2AM. Cytosolic calcium in controls was 70 ⫾ 10 nM (10 determinations); results are expressed as the difference from controls (no HT, no NIF). The results are the average of 10 experiments. Vertical bars represent the S.E.M. ANOVA: Effects of nifedipine and of HT, p ⬍ 0.001. [Ca2⫹]c ⫽ concentration of cytosolic calcium; HT: 3,4-(dihydroxyphenyl)ethanol; NIF: nifedipine. Redrawn from Palmerini et al. 2005a, with kind permission of Elsevier Limited.
or (10⫺6–10⫺4 mol L–1) (Visioli et al., 2001). The relationship between NO and phenols is complex: phenols increase [Ca2⫹]c which, in its turn, would increase the formation of NO through the stimulation of NO synthase (Cruz et al., 1998; Visioli et al., 1998; Andriantsitohaina et al., 1999). NO decreases [Ca2⫹]c in lymphomonocyte preparations (Palmerini et al., 2008) and in vascular endothelial cells (Chen et al., 2000). In the presence of both CysNO and HT, the opposite effects of the two agents add to each other (Figure 121.4). This suggests a different mechanism of action on [Ca2⫹]c modifications. CysNO acts by facilitating the exit of calcium from cytosol, whereas phenols stimulate the exit of the ion from RE. HT shows its effects even in the presence of high concentrations (200–500 μM) of CysNO (Figure 121.4), in agreement with the hypothesis that olive oil phenols act though their scavenger properties towards nitrogen reactive species similarly to what has been reported for oxygenreactive species (Visioli et al., 2001). Toxic effects related to peroxynitrite formation are noticed with high concentrations of NO and in the presence of superoxide radical. Therefore, peroxinitrite formation is possible in some pathological situations (Annane et al., 2000; Kingwell, 2000).
60
80
100
120
140
160
180
200
220
240
FIGURE 121.4 Effects of S-nitrosocysteine (CysNO) and of hydroxythyrosol (HT) on lymphomonocyte cytosolic calcium concentration ([Ca2⫹]c). [Ca2⫹]c was monitored as described by Grynkiewicz et al. (1985) using the fluorescent probe FURA-2AM. Lymphomonocyte [Ca2⫹]c in controls (no CysNO, no HT) was 85 ⫾ 5 nM (10 determinations). Results were expressed as the difference from control (no CysNO, no HT) ⫾S.E.M. ANOVA: effects of HT and of CysNO, p ⬍ 0.001. [Ca2⫹]c ⫽ concentration of cytosolic calcium; CysNO: S-nitrosocysteine; HT: 3,4-(dihydroxyphenyl)ethanol. Redrawn from Palmerini et al. 2005b, with kind permission of Springer Science and Business Media.
Ca2+
NIF
NO Decreased cytosolic Calcium
Ca
2+
- + NO, HT
SOCCs
2+
[Ca ]
Cell
Ca2+
Ca2+
+HT
ER
TG Ca2+
- SERCA
+
Depleted Calcium stores
FIGURE 121.5 Schematic representation of hydroxythyrosol (HT) and NO effects on calcium homeostasis control in human lymphomonocytes. NO decreases [Ca2⫹]c by allowing the exit of the ion from the cell; HT increases [Ca2⫹]c by potentiating the exit of the ion from ER. The depletion of ER calcium induced by TG – inhibition of SERCA and activation of calcium flow towards cytosol – stimulates SOCCs (store-operated calcium channels). The depletion of [Ca2⫹]c due to NO enhances the exit of calcium from ER and indirectly increases the capacitative entry of calcium into the cell. [Ca2⫹]c ⫽ concentration of cytosolic calcium; CysNO: S-nitrosocysteine; TG: thapsigargin; NIF: nifedipine; HT: 3,4(dihydroxyphenyl)ethanol; ER: endoplasmic reticulum; SERCA: sarcoendoplasmic reticulum Ca2⫹-dependent ATPase; SOCCs: store-operated Ca2⫹ channels.
SUMMARY POINTS ●
NO participates in calcium homeostasis in lymphomonocyte cell preparations by stimulating the exit of the ion from the cytosolic compartment and indirectly activating the capacitative entry of calcium (Figure 121.5).
260
HT, μM
●
The extent of the decrease depends on the NO concentration. High NO concentrations disrupt the homeostatic mechanisms of calcium and deplete [Ca2⫹]c (Dedkova and Blatter, 2002).
CHAPTER | 121 Olive Oil Phenols and Nitric Oxide Affect Lymphomonocyte Cytosolic Calcium
●
●
●
Phenols (and especially HT) produce a dose-dependent increase of [Ca2⫹]c, so opposing the effects of NO (Figure 121.5). HT compensates the decrease of [Ca2⫹]c produced by NO, by favoring the exit of the ion from the ER and contributes to the maintenance of calcium homeostasis in lymphomonocytes. Phenols, therefore, oppose the excessive losses of calcium from the cell due to high concentrations of NO. The consumption of extra virgin olive oil phenols in a typical Mediterranean diet ranges from a few hundred mg to 1.5 g (Hertog et al., 1993); therefore the effects of these substances of human health can be relevant.
REFERENCES Andriantsitohaina, R., Andriambeloson, E., Stoclet, J.C., 1999. Pharmacological approaches of endothelial nitroc oxide-dependent vasorelaxation induced by polyphenols from plant extracts. Methods Enzymol. 301, 522–531. Annane, D., Sanquer, S., Sebille, V., Faye, A., Djuranovic, D., Raphael, J.C., Gajdos, P., Bellissant, E., 2000. Compartmentalised inducible nitric-oxide synthase activity in septic shock. Lancet 355, 1143–1148. Arnelle, D.R., Stamler, J.S., 1995. NO⫹, NO and NO⫺ donation by S-nitrosothiols: implications for regulation of physiological functions by S-nitrosylation and acceleration of disulfide formation. Arch. Biochem. Biophys. 318, 279–285. Beckman, J.S., Koppenol, W.H., 1996. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am. J. Physiol. 271, C1424–C1437. Berger, W., Prinz, H., Striessnig, J., Kang, H.C., Haugland, R., Glossmann, H., 1994. Complex molecular mechanism for dihydropyridine binding to L-type Ca(2⫹)-channels as revealed by fluorescence resonance energy transfer. Biochemistry 33, 11875–11883. Boccini, F., Herold, S., 2004. Mechanistic studies of the oxidation of oxyhemoglobin by peroxynitrite. Biochemistry 43, 16393–16404. Catterall, W.A., Striessnig, J., 1992. Receptor sites for Ca2⫹ channel antagonists. Trends Pharmacol. Sci. 13, 256–262. Chen, J., Wang, Y., Nakajima, T., Iwasawa, K., Hikiji, H., Sunamoto, M., Choi, D.K., Yoshida, Y., Sakaki, Y., Toyo-Oka, T., 2000. Autocrine action and its underlying mechanism of nitric oxide on intracellular Ca2⫹ homeostasis in vascular endothelial cells. J. Biol. Chem. 275, 28739–28749. Clapham, D.E., 1995. Calcium signaling. Cell 80, 259–268. Clementi, E., 1998. Role of nitric oxide and its intracellular signalling pathways in the control of Ca2⫹ homeostasis. Biochem. Pharmacol. 55, 713–718. Cruz, M.T., Carmo, A., Carvalho, A.P., Lopes, M.C., 1998. Calciumdependent nitric oxide synthase activity in rat thymocytes. Biochem. Biophys. Res. Commun. 248, 98–103. de la Rosa, L.A., Alfonso, A., Vilarino, N., Vieytes, M.R., Botana, L.M., 2001. Modulation of cytosolic calcium levels of human lymphocytes by yessotoxin, a novel marine phycotoxin. Biochem. Pharmacol. 61, 827–833. Dedkova, E.N., Blatter, L.A., 2002. Nitric oxide inhibits capacitative Ca2⫹ entry and enhances endoplasmic reticulum Ca2⫹ uptake in bovine vascular endothelial cells. J. Physiol. 539, 77–91. Denmeade, S.R., Isaacs, J.T., 2005. The SERCA pump as a therapeutic target: making a “smart bomb” for prostate cancer. Cancer Biol. Ther. 4, 14–22.
1115
Doulias, P.T., Barbouti, A., Galaris, D., Ischiropoulos, H., 2001. SIN-1 induced DNA damage in isolated human peripheral blood lymphocytes as assessed by single cell gel electrophoresis (comet assay). Free Radic. Biol. Med. 30, 679–685. Gardner, P., 1989. Calcium and T lymphocyte activation. Cell 59, 15–20. Grynkiewicz, G., Poenie, M., Tsien, R.Y., 1985. A new generation of Ca2⫹ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260, 3440–3450. Guillermet, S., Vuillez, J.P., Fontaine, E., Caravel, J.P., Marti-Batlle, D., Pasqualini, R., Fagret, D., 2006. Cellular uptake of (99 m)TcN-NOET in human leukaemic HL-60 cells is related to calcium channel activation and cell proliferation. Eur. J. Nucl. Med. Mol. Imaging 33, 66–72. Hertog, M.G., Hollman, P.C., Katan, M.B., Kromhout, D., 1993. Intake of potentially anticarcinogenic flavonoids and their determinants in adults in The Netherlands. Nutr. Cancer 20, 21–29. Kingwell, B.A., 2000. Nitric oxide-mediated metabolic regulation during exercise: effects of training in health and cardiovascular disease. FASEB J. 14, 1685–1696. Kochegarov, A.A., 2003. Pharmacological modulators of voltage-gated calcium channels and their therapeutical application. Cell Calcium 33, 145–162. Leung, Y.M., Kwan, C.Y., 1999. Current perspectives in the pharmacological studies of store-operated Ca2⫹ entry blockers. Jpn. J. Pharmacol. 81, 253–258. Leung, Y.M., Kwan, C.Y., Loh, T.T., 1996. Dual effects of SK&F 96365 in human leukemic HL-60 cells. Inhibition of calcium entry and activation of a novel cation influx pathway. Biochem. Pharmacol. 51, 605–612. Nagy, G., Koncz, A., Perl, A., 2003. T cell activation-induced mitochondrial hyperpolarization is mediated by Ca2⫹- and redox-dependent production of nitric oxide. J. Immunol. 171, 5188–5197. Nusse, O., Serrander, L., Foyouzi-Youssefi, R., Monod, A., Lew, D.P., Krause, K.H., 1997. Store-operated Ca2⫹ influx and stimulation of exocytosis in HL-60 granulocytes. J. Biol. Chem. 272, 28360–28367. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Wurtele, G., Spiegelhalder, B., Bartsch, H., 2000. Olive-oil consumption and health: the possible role of antioxidants. Lancet Oncol. 1, 107–112. Palmerini, C.A., Carlini, E., Saccardi, C., Servili, M., Montedoro, G., Arienti, G., 2005a. Activity of olive oil phenols on lymphomonocyte cytosolic calcium. J. Nutr. Biochem. 16, 109–113. Palmerini, C.A., Carlini, E., Saccardi, C., Servili, M., Montedoro, G., Arienti, G., 2005b. Antagonism between olive oil phenolics and nitric oxide on lymphomonocyte cytosolic calcium. Mol. Cell. Biochem. 280, 181–184. Palmerini, C.A., Mazzoni, M., Saccardi, C., Arienti, G., 2008. The cytosolic calcium concentration is affected by S-nitrosocysteine in human lymphomonocytes. J. Biochem. Mol. Toxicol. 22, 35–40. Partiseti, M., Le Deist, F., Hivroz, C., Fischer, A., Korn, H., Choquet, D., 1994. The calcium current activated by T cell receptor and store depletion in human lymphocytes is absent in a primary immunodeficiency. J. Biol. Chem. 269, 32327–32335. Patel, R.P., McAndrew, J., Sellak, H., White, C.R., Jo, H., Freeman, B.A., Darley-Usmar, V.M., 1999. Biological aspects of reactive nitrogen species. Biochim. Biophys. Acta 1411, 385–400. Rao, R.V., Hermel, E., Castro-Obregon, S., del, R.G., Ellerby, L.M., Ellerby, H.M., Bredesen, D.E., 2001. Coupling endoplasmic reticulum stress to the cell death program. Mechanism of caspase activation. J. Biol. Chem. 276, 33869–33874. Servili, M., Baldioli, M., Selvaggini, R., Macchioni, A., Montedoro, G., 1999. Phenolic compounds of olive fruit: one- and two-dimensional
1116
nuclear magnetic resonance characterization of Nuzhenide and its distribution in the constitutive parts of fruit 16. J. Agric. Food Chem. 47, 12–18. Treiman, M., Caspersen, C., Christensen, S.B., 1998. A tool coming of age: thapsigargin as an inhibitor of sarco-endoplasmic reticulum Ca(2⫹)-ATPases. Trends Pharmacol. Sci. 19, 131–135. Visioli, F., Bellosta, S., Galli, C., 1998. Oleuropein, the bitter principle of olives, enhances nitric oxide production by mouse macrophages. Life Sci. 62, 541–546.
SECTION | II Cells and Cellular Effects
Visioli, F., Galli, C., 2002. Biological properties of olive oil phytochemicals. Crit. Rev. Food Sci. Nutr. 42, 209–221. Visioli, F., Galli, C., 2001. The role of antioxidants in the Mediterranean diet. Lipids 36 (Suppl), S49–S52. Williams, D.L.H., 1999. The chemistry of S-nitrosothiols. Acc. Chem. Res. 32, 869–876. Xu, X., Star, R.A., Tortorici, G., Muallem, S., 1994. Depletion of intracellular Ca2⫹ stores activates nitric-oxide synthase to generate cGMP and regulate Ca2⫹ influx. J. Biol. Chem. 269, 12645–12653.
Chapter 122
Olive Oil in Botanical Cosmeceuticals Leslie Baumann1 and Edmund Weisberg2 1 2
University of Miami Cosmetic Group, Miami Beach, FL, USA Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, USA
122.1 INTRODUCTION The primary source of fat in the Mediterranean diet, known to be one of the healthiest around the world, olive oil (Olea europaea) has long been considered one of the most important of the natural essential oils. For millennia, as long as it has been part of the human diet, people have also used olive oil for various non-culinary purposes, including for its beneficial effects on the skin (see Table 122.1). The ancient Egyptians and Romans used olive oil in food, cosmetics, massage oil for athletes, anointing oil, and salve for soothing wounds, and the ancient Greeks bathed with olive oil (Aburjai and Natsheh, 2003). In recent years, the topical application of olive oil has been reported to be an effective option in treating xerosis, rosacea, psoriasis, atopic dermatitis, contact dermatitis (particularly diaper dermatitis), eczema (including severe hand and foot eczema), seborrhea, pruritus, and various inflammations, burns and other cutaneous damage (Perricone, 2001; Aburjai and Natsheh, 2003) (see Table 122.2). In addition, olive oil is showing promise as a potential photoprotective agent. This chapter TABLE 122.1 Key facts about botanical cosmeceuticals. ●
Uses of many herbs in modern products date back centuries or millennia to ancient or traditional cultures
●
Natural products pose fewer risks than chemical-based products in terms of toxicity and adverse effects
●
Technological advances have enabled us to more carefully study the biochemistry of botanical extracts and thus verify traditional uses and identify novel therapeutic applications or potential
●
●
will focus on the chemical constituents of olive oil believed to confer beneficial health effects before concentrating on the dermatologic implications of olive oil consumption and topical application, including a brief discussion of some of the available formulations.
122.2 CHEMISTRY 122.2.1 Constituents Olive oil contains several active ingredients, including polyphenols, squalene, fatty acids (particularly the monounsaturated fat oleic acid), triglycerides, tocopherols, carotenoids, sterols, and chlorophylls (Aburjai and Natsheh, 2003) (see Table 122.3). The phenols in virgin olive oil have been found to scavenge reactive oxygen (Aburjai and Natsheh, 2003) and nitrogen species implicated in human disease, but it has not yet been established whether these olive oil compounds impart an effect beyond those on extracellular sources of nitric oxide (de la Puerta et al., 2001). Squalene, which acts as an antioxidant in olive oil, is present in unusually high concentrations in olive oil in comparison to other fats and oils typically found in the human diet (Kohno et al., 1995; Newmark, 1997; Budiyanto, 2000). TABLE 122.2 Cutaneous indications for olive oil.
Careful screening for a substance’s inflammatory potential is necessary before including a natural ingredient in skin care formulations Before using a natural ingredient as a therapeutic tool, laboratory benefits must also be translated to human benefits through extensive clinical testing
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1117
●
Atopic dermatitis
●
Burns
●
Contact dermatitis (especially diaper dermatitis)
●
Eczema (particularly severe cases in hands and feet)
●
Pruritus
●
Psoriasis
●
Rosacea
●
Seborrhea
●
Various inflammations
●
Xerosis
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1118
SECTION | II
Skin and Cosmeceuticals
TABLE 122.3 Active ingredients of olive oil.
TABLE 122.4 Polyphenolic constituents in olive oil.
Primary active ingredients
Unsaponifiable Polar fraction fraction
Simple phenols
Secoiridoids
Lignans
Carotenoids
Beta-sitosterol
Hydroxytyrosol Tyrosol
Oleuropein Ligstroside 10-hydroxyoleuropein 10-hydroxyligstroside
Acetoxypinoresinol Pinoresinol
Chlorophylls Fatty acids (especially oleic acid) Polyphenols Squalene Sterols Tocopherols Triglycerides
Erythrodiol Squalene
The polyphenols caffeic acid, hydroxytyrosol, oleuropein, and tyrosol
The major constituents of the unsaponifiable fraction of virgin olive oil include erythrodiol, beta-sitosterol, and squalene; the key components of the polar fraction include the polyphenols oleuropein, tyrosol, hydroxytyrosol, and caffeic acid, of which antioxidant properties have been clearly identified (de la Puerta et al., 2000) (Table 122.3). In fact, beta-sitosterol and tyrosol have been found to modulate reactive oxygen species and nitric oxide synthesis, arachidonic acid (AA) release, and the production of AA metabolites (Moreno, 2003). Antioxidants act to neutralize free radicals, also known as reactive oxygen species. Free radicals develop when oxygen molecules combine with other molecules, leaving an odd number of electrons. An oxygen molecule with paired electrons is stable, but oxygen with an unpaired electron is considered ‘reactive’, because it searches for and takes electrons from surrounding vital components rendering them damaged (Werninghaus, 1995).
122.2.2 Properties A study of the topical application of virgin olive oil on edema in mice induced by AA or 12-O-tetradecanoylphorbol acetate (TPA) was designed to assess the anti-inflammatory effects of the unsaponifiable and polar fractions. Assays with both classes of compounds revealed anti-inflammatory effects exhibited by both groups, with beta-sitosterol and erythrodiol demonstrating a potent anti-edematous effect in the TPA model and oleuropein, tyrosol, hydroxytyrosol, and caffeic acid showing a significant inhibitory effect. The unsaponifiable fraction more strongly inhibited AA, and oleuropein was a potent inhibitor among the polar components. The investigators concluded that the anti-inflammatory activity ascribed to both groups of compounds may be influential in the salutary effects attributed to virgin olive oil (de la Puerta et al., 2000). Other studies have also shown that the polyphenolic
constituents of olive oil exhibit protective activity against inflammation (Martínez-Domínguez et al., 2001; Aburjai and Natsheh, 2003), which plays a significant contributory role in the majority of dermatologic disorders. Importantly, while olive oil is gaining traction as an anti-inflammatory agent, it is also generally considered as very weakly irritant, with adverse side effects to its topical use rarely reported, though it is deemed unsuitable or contraindicated in patients with venous insufficiency and related eczema on the lower extremities (Kränke et al., 1997). The main phenolic compounds contained in olive oil, all of which display significant antioxidant activity, include simple phenols (hydroxytyrosol and tyrosol), secoiridoids (oleuropein, the aglycone of ligstroside, and their respective decarboxylated dialdehyde derivatives), and the lignans [(⫹)-]-acetoxypinoresinol and pinoresinol (Owen et al., 2000a) (see Table 122.4). Lignans, specifically, have been found to be potent antioxidants that inhibit cellular growth in skin, breast, colon, and lung cancers as well as inhibit in vivo lipid peroxidation (Owen et al., 2000b). Notably, the high consumption of extra virgin olive oil, which is rife with antioxidant characteristics derived from these three polyphenol groups as well as squalene and oleic acid, is thought to confer protection against oxidative stress and some of its manifestations, such as aging, as well as skin and other cancers (Owen et al., 2000a). Consistent and heavy consumption of olive oil, along with vegetables and legumes, has also been found to impart protection against actinic damage (Purba et al., 2001). While oleic acid is clearly an important constituent of olive oil, the fact that it is contained in widely eaten animal foods such as poultry and pork suggests that it is not the main component responsible for the salutary effects ascribed to olive oil (Visioli, 2002).
122.3 DIETARY PROTECTION The notion that one’s diet can impact the health and appearance of the skin might seem intuitive, but it was somewhat undermined by a couple of studies around 1970 related to acne and various potential dietary triggers. While convincingly arguing for the abandonment of the traditional belief expressed by the dermatology community, promulgated by
1119
CHAPTER | 122 Olive Oil in Botanical Cosmeceuticals
these studies, that diet does not contribute to acne etiology, Cordain specifically claims that the dogma claiming that diet and acne are unrelated was based on two fundamentally flawed studies from 1969 by Fulton et al. and 1971 by Anderson. He contends that these studies lacked control groups, statistical data treatment, as well as blinding and/ or placebos and were characterized by inadequate sample sizes and insufficient or absent baseline diet data, among other deficiencies (Cordain, 2005). Such studies likely contributed to diverting the dermatology community from a deeper appreciation of the role that nutrition can play in skin health, other than offering the healthy suggestions of eating more fruits and vegetables and specifically trying to identify food allergens with cutaneous manifestations or food triggers of skin conditions (particularly rosacea). A plethora of recent research and, likely, current research underway, on the direct effects on health from the consumption or supplementation of various nutrients points us in a different direction, however. Much of this work focuses specifically on the potential benefits imparted to the skin as a result of the intake of certain foods or supplements. For example, in 2003, investigators conducted a cross-sectional study of 302 healthy men and women, collecting data on serum concentrations of nutrients, dietary consumption of nutrients and various cutaneous measurements (including hydration, sebum content, and surface pH). The study demonstrated statistically significant relationships between serum vitamin A and cutaneous sebum content and surface pH as well as between skin hydration and dietary consumption of total fat, saturated fat, and monounsaturated fat. The authors concluded that such findings suggest that changes in an individual’s baseline nutritional status can affect the condition of the skin (Boelsma et al., 2003). This is but one example to buttress the argument that oral delivery of nutrients must be taken into account as well as systemic administration and topical application of medication and cosmetics when considering skin health.
122.3.1 Monounsaturated Fats For optimum health, the chief dietary fat should be derived from foods and oils rich in monounsaturated fat. When monounsaturated fats predominate, saturated fats, trans fatty acids, and omega-6 polyunsaturated fatty acids are counterbalanced, and the ratio of omega-6 to omega-3 fatty acids improves as the proportion of omega-3 acids increases. This is notable particularly in the West, where the ratio of omega-6 to omega-3 fatty acid consumption has grown from 10:1 (Sugano, 1996) during the mid-1990s to approximately 15:1 to 16.7:1 (Simopoulos, 2006) now, far from what is considered the healthy ratio of closer to 4:1 (Sugano, 1996). Both of these omega groups are essential for healthy human growth and development. A higher risk
TABLE 122.5 Dietary ingredients high in omega-3 monounsaturated fats, which confer positive effects on the skin. ● ● ● ● ● ● ●
Avocados Canola oil Nuts (except walnuts and butternuts) Olive oil and olives Peanuts Safflower oil Sunflower oil
for depression and various inflammatory conditions has been associated with a high ratio of omega-6 to omega-3 fatty acids (Kiecolt-Glaser et al., 2007). Olive oil and olives, along with nuts (except for walnuts and butternuts), peanuts (a legume), avocados, canola oil, high-oleic sunflower oil, and high-oleic safflower oil all contain significant levels of monounsaturated fats and are thought to be influential in ameliorating dry skin (see Table 122.5). An interesting study of a potential link between dietary intake and skin wrinkling in sun-exposed areas was conducted by Purba et al. in 2001. They used food frequency questionnaires and cutaneous microtopographic measurements to evaluate diet and skin wrinkling in 177 Greek-born individuals living in Melbourne, Australia, 69 Greek subjects residing in rural Greece, 48 AngloCeltic Australian elderly individuals living in Melbourne, Australia, and 159 Swedish elderly participants living in Sweden. The Swedish elderly were found to manifest the least wrinkling in sun-exposed areas, followed by the Greek-born in Melbourne, rural Greek elderly, and then Anglo-Celtic Australians. Correlation and regression analyses yielded data that prompted the investigators to conclude that diet may very well have an impact on skin wrinkling. Generally, they found that individuals that consumed more vegetables (especially green leafy vegetables, spinach specifically, as well as asparagus, celery, eggplant, garlic, and onions/leeks), olive oil, monounsaturated fat, and legumes and lower levels of milk and milk products, butter, margarine, and sugar products exhibited fewer wrinkles in sun-exposed skin. A high intake of olive oil, legumes, vegetables, fish, and cereal, were found to be particularly protective against photodamage. Significantly, the authors suggested that diets high in monounsaturated acids may increase the levels of monounsaturated fatty acids in the epidermis, which resist oxidative damage, as opposed to epidermal polyunsaturated fatty acids, which are more susceptible to oxidation. Further, they speculated that such a phenomenon may explain their findings of a correlation between monounsaturated olive oil and less wrinkling as well as the higher level of wrinkling associated with the consumption of polyunsaturated margarine (Purba et al., 2001). A previous study found that olive oil appeared to be important
1120
in potentiating the antioxidant activity of plasma lycopene after the consumption of tomato products along with olive oil (Lee et al., 2000).
122.3.2 An Olive Oil-Rich Diet In 2001, Moreno et al. performed a study to ascertain the effect of a diet rich in olive oil on important inflammation mediators, namely oxidative stress and prostaglandin synthesis. In the process, they compared the effects on rats of an olive oil-rich diet to those of corn oil-rich and fish oilrich diets. Both olive oil and fish oil were found to decrease AA release and the ensuing synthesis of AA metabolites, but olive oil was more efficient in reducing oxidative stress. Prostaglandin E2 levels were measured to be lower in the rats fed the olive oil or fish oil diets as compared to the corn oil diet (Moreno et al., 2001).
SECTION | II
Skin and Cosmeceuticals
oil. The authors speculated that the mechanism of action of extra virgin olive oil in this context may work by reducing 8-OHdG formation, which is known to be involved in gene mutation. They concluded that the daily topical use of extra virgin olive oil following sun exposure may have the same impact on human skin, potentially retarding the development of UV-induced skin cancer (Ichihashi et al., 2000). Recently, researchers conducted an in vivo investigation in 32 volunteers to ascertain the effects of various topical agents, including olive oil, on the transmission of UVB radiation phototherapy. The minimal erythema dose (MED) necessary to cause skin reddening for each volunteer was identified through a phototest and then was repeated using each test formulation. The effects on MED of each product were assessed after 24 hours. Of the agents tested, olive oil and glycerine had no impact on MED and were thus deemed suitable for use prior to phototherapy (Fetil et al., 2006).
122.4 PHOTOPROTECTION
122.4.2 UVA
122.4.1 UVB
Hydroxytyrosol, one of the key polyphenolic components of the polar fraction of olive oil, has been studied recently for its effects on UVA-induced cell damage in the human melanoma cell line M14. Investigators identified a dose-dependent protective effect imparted by the phenol, which prevented an increase in the usual markers of oxidative stress. The arrest of melanoma cell proliferation was observed at higher hydroxytyrosol concentrations. The olive oil constituent also displayed a dose-dependent apoptotic effect on melanoma cells. In addition, it was metabolized by catechol-O-methyltransferase into methylated derivatives, which bear more responsibility than the original molecule in manifesting in vivo antioxidant activity in cosmetics and functional foods containing olive oil. Further, such derivatives have the potential to interfere with the activity of certain methyltransferases involved in the catabolism of various pharmacological compounds. The investigators concluded that hydroxytyrosol prevents protein damage engendered by UVA in melanoma cells, but does not completely counteract the effects of irradiation, and that the dose-dependent differential impact on melanoma cells of hydroxytyrosol should be considered in relation to the selection of cosmetic and food products containing olive oil and potential interactions (D’Angelo et al., 2005). Previously, hydroxytyrosol was studied for its effects on the proliferation, apoptosis, and cell cycle of human leukemia (HL60) and colon adenocarcinoma cells (HT29 and HT29 clone 19A). The olive oil antioxidant displayed a capacity to inhibit cell proliferation in both cancer types, more successfully in HL60. It was also found to induce apoptosis in HL60 cells after 24 hours of incubation and arrest the HL60 cells in the G0/G1 phase of the cell cycle. Investigators concluded that hydroxytyrosol appears to exhibit protective properties against cancer, and may be
In a relatively recent study, investigators assessed the capacity of extra virgin olive oil to combat reactive oxygen species and skin tumors induced by ultraviolet (UV) light exposure. Topical application of the oil to hairless mice before or after repeated exposure to UVB radiation led to delays in the onset of skin tumors as compared to exposures in control mice. Differences between control mice and the mice pre-treated with olive oil declined with additional exposures to UVB. Interestingly, mice treated with olive oil following UVB exposure exhibited significantly fewer tumors than mice in the control group. Researchers concluded that topical application of olive oil after UVB exposure is effective in retarding the onset and suppressing the number of murine skin tumors caused by UVB exposure (Budiyanto et al., 2000). Based on previous evidence showing that topically applied vitamin E and epigallocatechin gallate extracted from green tea had delayed the onset of UV-induced skin cancer in mice, as well as the reportedly potent in vitro antioxidant activity of olive oil, several of the same investigators sought to examine the effects of topically applied olive oil, particularly in delaying the onset and decreasing the number of UV-induced skin cancers in mice. Results comparable to the Budiyanto et al. study were obtained insofar as extra virgin olive oil topically applied immediately following UVB radiation exposure did, in fact, significantly lower the number and delay the onset of skin cancer lesions as well as reduce the formation in murine epidermis of the free radicalinduced 8-hydroxy-deoxyguanosine (8-OHdG). However, no effects on skin cancer formation were associated with extra virgin olive oil topically applied prior to UV exposure or the pre- or post-exposure application of regular olive
1121
CHAPTER | 122 Olive Oil in Botanical Cosmeceuticals
a major factor in the reputed anticancer activity of virgin olive oil (Fabiani et al., 2002).
122.5 TOPICAL APPLICATIONS FOR DERMATOLOGIC CONDITIONS In contemporary times, olive oil is not generally known as a first-line treatment for cutaneous disorders, but it has come to be considered an effective therapeutic option for several conditions. A brief review follows of the literature on such applications.
122.5.1 Olive Oil and Dry Skin In a recent study, investigators compared the cutaneous effects and stability against oxidation of olive oil and hempseed oil. Both of these unsaturated fatty acids are thought to ameliorate xerosis and related manifestations of aging. The researchers observed that olive oil was more stable than hemp-seed oil against peroxidation, and that the chlorophyll found in extra virgin olive oil exhibited greater photostability than that included in hemp-seed oil, which they speculated might be the result of a higher concentration of antioxidants in olive oil. After preparing emulsions with the two oils, they concluded that some of the gel-emulsion preparations were appropriate for spraying on the skin (Sapino et al., 2005). It is worth noting that the unusual stability of extra virgin olive oil, as well as its pungent taste, has been attributed to the polyphenolic antioxidant constituents hydroxytyrosol and oleuropein (Visioli et al., 2002). In addition, these olive oil components have been demonstrated to exhibit more potent antioxidant properties than vitamin E and the food preservative butylhydroxytoluene (Visioli et al., 2002).
122.5.2 Olive Oil and Dermatitis In a randomized controlled trial to test the cutaneous effects of two different topical ointments on the skin of premature infants conducted from October 2004 to November 2006, investigators prospectively enrolled 173 infants between 25 and 36 weeks of gestation admitted into a neonatal intensive care unit, and randomly scheduled them for daily therapy with a waterin-oil emollient cream, an olive oil cream (70% lanolin, 30% olive oil), or a control ointment. After a maximum of four weeks of treatment, statistically less dermatitis was observed in the neonates treated with olive oil cream as compared to the emollient cream, with beneficial effects enduring throughout the trial and both test groups displaying better outcomes than the control group (Kiechl-Kohlendorfer et al., 2008).
122.5.3 Antifungal Properties of Olive Oil A recent study has also suggested an antimicrobial function for olive fruit and olive oil. Specifically, researchers studied
the antifungal activity of the aliphatic aldehydes in olives – hexanal, nonanal, (E)-2-hexenal, (E)-2-heptenal, (E)-2-octenal, and (E)-2-nonenal – against six strains of Tricophyton mentagrophytes, one strain of Microsporum canis, and seven strains of Candida spp. The tested aldehydes inhibited the growth of T. mentagrophytes and M. canis but had no impact on any of the Candida spp. strains. In a concentration-dependent fashion, elastase activity was inhibited by (E)-2-octenal and (E)-2-nonenal, which is significant insofar as elastase is an enzyme that breaks down elastin, one of the integral building blocks of the dermis. Investigators suggested that their findings support the use of olives and olive oil in the topical treatment of cutaneous diseases, particularly fungal skin infections (Battinelli et al., 2006).
122.5.4 Olive Oil in Combination 122.5.4.1 Atopic Dermatitis and Psoriasis Olive oil has also shown promise in combination with other ingredients for the treatment of dermatoses. In 2003, Al-Waili conducted a partially controlled, single-blind study to evaluate the effects of a honey, olive oil, and beeswax mixture (1:1:1) on 21 patients with atopic dermatitis (AD) and 18 patients with psoriasis. Eleven of 21 AD patients were instructed to use topical betamethasone esters and ten of 18 psoriasis patients used clobetasol propionate. Eight of the ten AD patients in the honey/olive oil/beeswax treatment group displayed significant improvement in the evaluated symptoms (i.e., erythema, scaling, lichenification, excoriation, indurations, oozing, and pruritus) after 2 weeks. Over the same time period, five of the eight psoriasis patients in the honey/olive oil/beeswax group also demonstrated significant improvement in the assessed symptoms (i.e., redness, scaling, thickening and pruritus). In addition, the symptoms in five of the 11 AD patients treated with betamethasone and five of the ten psoriasis patients treated with clobetasol propionate remained stable after their treatments were replaced with the honey/olive oil/beeswax mixture combined with corticosteroid representing a 75% reduction in their original dose. Al-Waili concluded that the honey/olive oil/beeswax mixture appears effective in the management of these skin disorders (Al-Waili, 2003).
122.5.4.2 Fungal and Bacterial Infections The following year, Al-Waili tested this same mixture in 37 patients as a treatment for the cutaneous fungal infections pityriasis versicolor, tinea cruris, tinea corporis, and tinea faciei. The honey/olive oil/beeswax mixture was applied to the various lesions three times daily for up to 4 weeks. The author observed a clinical response in terms of erythema, scaling, and pruritus in 86% of pityriasis versicolor patients, 78% of tinea cruris patients, and 75% of tinea corporis patients, with mycological cure achieved in a significant percentage
1122
of patients (75%, 71% and 62% of patients with pityriasis versicolor, tinea cruris, and tinea corporis, respectively). In addition, after 3 weeks of therapy, the lone patient with tinea faciei exhibited clinical as well as mycological resolution. Al-Waili suggested that these findings indicate that the honey/olive oil/beeswax mixture appears effective for treating these cutaneous fungi and that future controlled trials evaluating this compound are warranted (Al-Waili, 2004). In 2005, Al-Waili assessed the effects of the same honey/ olive oil/beeswax mixture on the growth of Staphylococcus aureus and Candida albicans isolated from humans and found that both the honey mixture as well as honey alone were effective in suppressing bacterial growth, whereas mild to moderate growth was observed on media containing olive oil or beeswax (Al-Waili, 2005a). That same year, Al-Waili tested the honey, olive oil, and beeswax mixture on 12 infants experiencing diaper dermatitis. The infants were treated four times daily for 7 days, and erythema was assessed on a five-point scale. A significant reduction in the mean lesion score was observed from baseline (2.91 ⫾ 0.79) to day 7 (0.66 ⫾ 0.98). In addition, Candida albicans was isolated in four patients prior to the treatment regimen, but only two patients after the 1-week of treatment. Overall, the honey/olive oil/beeswax compound was deemed to be a safe as well as clinically and mycologically effective therapy for diaper dermatitis (Al-Waili, 2005b).
122.5.4.3 Anal Fissure and Hemorrhoids More recently, Al-Waili et al. conducted a prospective pilot study to assess the therapeutic effects of topical application of the mixture on patients with anal fissure or hemorrhoids. Fifteen consecutive patients who presented with anal fissure or first- to third-degree hemorrhoids were treated with a 12-hour application of the honey/olive oil/beeswax ointment. The symptoms were scored at baseline and on a weekly basis up to 4 weeks. The natural mixture allayed itching and significantly decreased bleeding in hemorrhoid patients and similar improvements were noted in the anal fissure patients, with no side effects reported in any patients. The authors concluded that the honey/olive oil/beeswax compound is clinically effective and safe for the treatment of hemorrhoids and anal fissure and warrants additional testing in randomized double-blind trials (Al-Waili et al., 2006).
122.6 COSMECEUTICALS The designation ‘cosmeceutical’ has still not been codified or officially recognized by regulatory bodies such as the United States Food and Drug Administration (FDA), although it has been debated for decades. In other words, it has no legal meaning. Importantly, though, the term is increasingly part of the public lexicon yet might not be
SECTION | II
Skin and Cosmeceuticals
clearly understood. The word ‘cosmeceuticals’ refers to products that are known to have a biologic action but are regulated as cosmetics (e.g., products containing retinol) rather than as drugs. This means that manufacturers, by not listing certain ingredients as active or the main ingredients, can more readily market products. In particular, this allows for the introduction of products touted for their inclusion of specific ingredients that are not listed as active or as the main ingredients. While the FDA does not regulate such products, consumers purchase cosmeceuticals with the hope that some tangible differences will be achieved through their use.
122.6.1 Products Despite promising results regarding its photoprotective potential, olive oil has not yet been featured in after-sun products. However, olive oil is now found in most other, more frequently used types of over-the-counter skin care products, including bar and liquid soaps, bath oils, soaks for nails, lip balms, massage oils, shampoos, and moisturizers. In addition, there are a few product lines based on olive oil as the featured ingredient. The Macrovita Face Products with Olive Oil line includes Olive Oil and Calendula Cleansing Milk, Olive Oil and Calendula Tonic Lotion, Olive Oil and Propolis Deep Cleansing Liquid Soap, Olive Oil and White Tea Beauty Peel-Off Mask, and various other products such as hydrating cream, eye contour cream, skin reinforcing oil complex, and shampoo for dry scalps. The 7 Wonders Miracle Oil and Lotion line also contains olive oil as the primary touted ingredient in their various oil (body, hot, bath, baby, tanning, massage, and cuticle) formulations. N.V. Perricone, M.D. Cosmeceuticals offers a range of products featuring olive oil polyphenols, including Body Hydrator, Gentle Face Hydrator, Hydrating Mask, Nutrient Face Fortifier, and Moist Lips. La Nature Hartmann Cosmetic is a natural bath and body care manufacturer that sells a large range of home fragrances, accessories and spa-inspired gifts, including an olive oil line of products. The abundance of glycerides and fatty acids in olive oil render it gentle enough to use on sensitive skin, according to the manufacturers of the Jardin de l’Olivier products. This diverse line includes After Sun Oil, Bath & Shower Crème, Bath Bar, Bath Oil, Body Lotion, Dry Oil, Hand Cream, Crème Shampoo, Leaf Soap, Tonic Lotion, Cleansing Cream, Dry Skin Cream, Sun Protection Oil, and Tonic Lotion each with olive oil as the chief component. MedAssist Therapeutic Skin Cremes is a line of products made from all natural ingredients. Their line of olive oil formulations is many and varied: Olive Branch Moisturizing Crème, Olive Branch Hydrating Lotion, Olive Branch Soothing Oil, Olive Branch Revitalizing Cleanser, Olive Branch Hair and Scalp Treatment, Olive Essence Hand and
1123
CHAPTER | 122 Olive Oil in Botanical Cosmeceuticals
Body Lotion, Olive Essence Silky Moisturizing Body Spray, and Olive Essence Gentle Cleansing Herbal Shampoo. Countryrose Soap Company uses blends of olive, coconut, palm, castor, and jojoba oils with all of the products containing a high percentage of olive oil (100% Olive Oil Castile Soap). Besides these specific lines, there are numerous companies featuring olive oil as an active ingredient in select products. It is important to note that without regulation, any testing conducted on such products is likely to remain proprietary. As is often the case with herbal-based products, there is a paucity of randomized, double-blind, controlled clinical trials to establish efficacy. Such research is the best way to examine the efficacy of these topical formulations, though, as well as any other products, and are clearly justified given the promising results on olive oil thus far seen.
122.7 CONCLUSION The Mediterranean diet, thought to be one of the healthiest around the world, has garnered significant attention in recent years as a result of a lower incidence of coronary heart disease as well as breast and colon cancers in the peoples living in that region. Olive oil, as the primary source of healthy fat in this diet, is considered one of the chief components of the Mediterranean diet, along with plant foods and fish. The wider benefits of olive oil, like many other botanically derived products, were appreciated in the ancient world and are now actively in the process of being re-discovered and more deeply elucidated. In terms of contemporary dermatologic medicine, olive oil appears to be an effective therapeutic option for several conditions, shows promise for future inclusion in photoprotective products, and has been incorporated into a wide array of over-the-counter topical products. The fact that several constituents in olive oil are known to exhibit significant antioxidant activity along with the results of recent studies yielding evidence of anti-inflammatory and anticarcinogenic effects conferred by olive oil provide reasons for future research and optimism regarding the expansion of medical and dermatologic applications. If olive oil is found to be even remotely as well regarded an antioxidant, antiinflammatory, and photoprotective agent in topical formulations as it is a healthy food and cooking oil, it might be bumped to the head of the growing list of potent herbal ingredients used and studied in dermatology.
SUMMARY POINTS ●
●
Olive oil has been used for dermatologic purposes for thousands of years, since the times of the ancient Egyptians, Greeks, and Romans. Recent epidemiologic evidence suggests an association between olive oil consumption and a lower incidence of cardiovascular disease and certain cancers.
●
●
●
The anti-inflammatory and antioxidant activity exhibited by olive oil is also thought to have implications for dermatologic applications. Use as a photoprotective agent is an important potential dermatologic application of olive oil. Olive oil is currently used in topical applications for the treatment of several skin conditions, including dry skin, itch, and inflammation as well as disorders such as rosacea.
REFERENCES Aburjai, T., Natsheh, F.M., 2003. Plants used in cosmetics. Phytother. Res. 17, 987–1000. Al-Waili, N.S., 2003. Topical application of natural honey, beeswax and olive oil mixture for atopic dermatitis or psoriasis: partially controlled, single-blinded study. Complement. Ther. Med. 11, 226–234. Al-Waili, N.S., 2004. An alternative treatment for pityriasis versicolor, tinea cruris, tinea corporis and tinea faciei with topical application of honey, olive oil and beeswax mixture: an open pilot study. Complement. Ther. Med. 12, 45–47. Al-Waili, N.S., 2005a. Mixture of honey, beeswax and olive oil inhibits growth of Staphylococcus aureus and Candida albicans. Arch. Med. Res. 36, 10–13. Al-Waili, N.S., 2005b. Clinical and mycological benefits of topical application of honey, olive oil and beeswax in diaper dermatitis. Clin. Microbiol. Infect. 11, 160–163. Al-Waili, N.S., Saloom, K.S., Al-Waili, T.N., Al-Waili, A.N., 2006. The safety and efficacy of a mixture of honey, olive oil, and beeswax for the management of hemorrhoids and anal fissure: a pilot study. Sci. World J. 6, 1998–2005. Anderson, P.C., 1971. Foods as the cause of acne. Am. Fam. Physician. 3, 102–103. Battinelli, L., Daniele, C., Cristiani, M., Bisignano, G., Saija, A., Mazzanti, G., 2006. In vitro antifungal and anti-elastase activity of some aliphatic aldehydes from Olea europaea L. fruit. Phytomedicine 13, 558–563. Boelsma, E., van de Vijver, L.P., Goldbohm, R.A., Klöpping-Ketelaars, I.A., Hendriks, H.F., Roza, L., 2003. Human skin condition and its associations with nutrient concentrations in serum and diet. Am. J. Clin. Nutr. 77, 348–355. Budiyanto, A., Ahmed, N.U., Wu, A., Bito, T., Nikaido, O., Osawa, T., Ueda, M., Ichihashi, M., 2000. Protective effect of topically applied olive oil against photocarcinogenesis following UVB exposure of mice. Carcinogenesis 21, 2085–2090. Cordain, L., 2005. Implications for the role of diet in acne. Semin. Cutan. Med. Surg. 24, 84–91. D’Angelo, S., Ingrosso, D., Migliardi, V., Sorrentino, A., Donnarumma, G., Baroni, A., Masella, L., Tufano, M.A., Zappia, M., Galletti, P., 2005. Hydroxytyrosol, a natural antioxidant from olive oil, prevents protein damage induced by long-wave ultraviolet radiation in melanoma cells. Free Radic. Biol. Med. 38, 908–919. de la Puerta, R., Martínez Domínguez, M.E., Ruíz-Gutíerrez, V., Flavill, J.A., Hoult, J.R., 2001. Effects of virgin olive oil phenolics on scavenging of reactive nitrogen species and upon nitrergic neurotransmission. Life Sci. 69, 1213–1222. de la Puerta, R., Martínez Domínguez, M.E., Ruíz-Gutíerrez, V., 2000. Effect of minor components of virgin olive oil on topical antiinflammatory assays. Z. Naturforsch. [C] 55, 814–819.
1124
Fabiani, R., De Bartolomeo, A., Rosignoli, P., Servili, M., Montedoro, G.F., Morozzi, G., 2002. Cancer chemoprevention by hydroxytyrosol isolated from virgin olive oil through G1 cell cycle arrest and apoptosis. Eur. J. Cancer Prev. 11, 351–358. Fetil, E., Akarsu, S., Ilknur, T., Kusku, E., Günes, A.T., 2006. Effects of some emollients on the transmission of ultraviolet. Photodermatol. Photoimmunol. Photomed. 22, 137–140. Fulton, J.E., Plewig, G., Kligman, A.M., 1969. Effect of chocolate on acne vulgaris. JAMA 210, 2071–2074. Ichihashi, M., Ahmed, N.U., Budiyanto, A., Wu, A., Bito, T., Ueda, M., Osawa, T., 2000. Preventive effect of an antioxidant on ultraviolet-induced skin cancer in mice. J. Dermatol. Sci. 23, S45–S50. Kiechl-Kohlendorfer, U., Berger, C., Inzinger, R., 2008. The effect of daily treatment with an olive oil/lanolin emollient on skin integrity in preterm infants: a randomized controlled trial. Pediatr. Dermatol. 25, 174–178. Kiecolt-Glaser, J.K., Belury, M.A., Porter, K., Beversdorf, D.Q., Glaser, R., 2007. Depressive symptoms, omega-6:omega-3 fatty acids, and inflammation in older adults. Psychosom. Med. 69, 217–224. Kohno, Y., Egawa, Y., Itoh, S., Nagaoka, S., Takahashi, M., Mukai, K., 1995. Kinetic study of quenching reaction of singlet oxygen and scavenging reaction of free radical by squalene in n-butanol. Biochim. Biophys. Acta 1256, 52–56. Kränke, B., Komericki, P., Aberer, W., 1997. Olive oil—contact sensitizer or irritant? Contact Dermatitis 36, 5–10. Lee, A., Thurnham, D.I., Chopra, M., 2000. Consumption of tomato products with olive oil but not sunflower oil increases the antioxidant activity of plasma. Free Radic. Biol. Med. 29, 1051–1055. Martínez-Domínguez, E., de la Puerta, R., Ruíz-Gutíerrez, V., 2001. Protective effects upon experimental inflammation models of a polyphenolsupplemented virgin olive oil diet. Inflamm. Res. 50, 102–106. Moreno, J.J., 2003. Effect of olive oil minor components on oxidative stress and arachidonic acid mobilization and metabolism by macrophages RAW 264.7. Free Radic. Biol. Med. 35, 1073–1081.
SECTION | II
Skin and Cosmeceuticals
Moreno, J.J., Carbonell, T., Sánchez, T., Miret, S., Mitjavila, M.T., 2001. Olive oil decreases both oxidative stress and the production of arachidonic acid metabolites by the prostaglandin G/H synthase pathway in rat macrophages. J. Nutr. 131, 2145–2149. Newmark, H.L., 1997. Squalene, olive oil, and cancer risk: a review and hypothesis. Cancer Epidemiol. Biomarkers Prev. 6, 1101–1103. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Würtele, G., Spiegelhalder, B., Bartsch, H., 2000a. Olive-oil consumption and health: the possible role of antioxidants. Lancet Oncol. 1, 107–112. Owen, R.W., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2000b. Identification of lignans as major components in the phenolic fraction of olive oil. Clin. Chem. 46, 976–988. Perricone, N.V., 2001. Treatment of skin disorders with olive oil polyphenols. PCT Int. Appl. 16, WO0176579. Purba, M.B., Kouris-Blazos, A., Wattanapenpaiboon, N., Lukito, W., Rothenberg, E.M., Steen, B.C., Wahlqvist, M.L., 2001. Skin wrinkling: can food make a difference? J. Am. Coll. Nutr. 20, 71–80. Sapino, S., Carlotti, M.E., Peira, E., Gallarate, M., 2005. Hemp-seed and olive oils: their stability against oxidation and use in O/W emulsions. J. Cosmet. Sci. 56, 227–251. Simopoulos, A.P., 2006. Evolutionary aspects of diet, the omega-6/omega-3 ratio and genetic variation: nutritional implications for chronic diseases. Biomed. Pharmacother. 60, 502–507. Sugano, M., 1996. Characteristics of fats in Japanese diets and current recommendations. Lipids 31, S283–S286. Visioli, F., Poli, A., Galli, C., 2002. Antioxidant and other biological activities of phenols from olives and olive oil. Med. Res. Rev. 22, 65–75. Werninghaus, K., 1995. The role of antioxidants in reducing photodamage. In: Gilchrest, B. (ed.), Photodamage. Blackwell Science Inc., London, p. 249.
Chapter 123
Effect of Olive Oil on the Skin Diana Badiu1, Rafael Luque2 and Rajkumar Rajendram3 1
Department of Biochemistry, Ovidius University of Constanza, Constanza, Romania Departamento de Quimica Orgánica, Universidad de Córdoba, Córdoba, Spain 3 Nutritional Sciences Research Division, School of Life Sciences, King’s College London, London, UK
2
123.1 PHYSIOLOGICAL PROPERTIES OF THE SKIN Human skin is a complex organ that regulates heat and water loss from the body whilst preventing the entry of toxic substances and microorganisms. It is large, accounting for 10% of body mass and has an average area of 1.7 m2. Theoretically the skin should provide multiple potential sites for the administration of medication for both local and systemic diseases. However, it is surprisingly tough and much more like a barrier than a membrane. The skin is a highly efficient self-repairing barrier designed to keep ‘the insides in and the outside out’. It consists of several layers of cells that are constantly being turned over. The cycle of shedding and regeneration occurs approximately once every 30 days in vivo. Although often regarded simply as a physical barrier, the skin is metabolically active and has immunological and histological responses to assault. The structure of human skin can be divided into four layers (Figure 123.1): ●
● ● ●
the stratum corneum (outermost non-viable epidermal layer) the viable epidermis the underlying dermis the innermost subcutaneous fat layer (hypodermis).
123.2 STRATUM CORNEUM The outermost and toughest layer of skin, the stratum corneum is also known as the ‘horny layer’. Stratum corneum consists of only 10–15 layers of non-viable corneocytes (cornified keratinocytes). These are the same dead cells that fingernails are made from. When dry the Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
stratum corneum is approximately ten micrometers (μm) thick; a significant depth in dermatological terms. When wet it swells to several times this. This thin membrane regulates water loss from the body, preventing desiccation whilst simultaneously restricting the entry of toxins and microorganisms. The stratum corneum can be described in terms of a bricks-and-mortar construction. The large, dead, anucleate, keratinized cells (bricks) of the epidermis are tightly packed together and the small spaces between the cells are filled with lipids (mortar). The corneocytes prevent the passage of foreign material while the lipids ensure that water is retained.
123.3 DERMIS AND EPIDERMIS The viable epidermis is about 80 μm thick. The second layer (dermis) contains capillaries and nerve endings and is several millimeters thick (Figure 123.2). Most skin-care products moisturize only the ‘dead’ layer of the skin; the 10 μm thick stratum corneum. Hydration of this layer prevents the sensation of dry skin. However this is insufficient and in most cases deleterious as the stratum corneum is easily waterlogged. The transfer of foreign material across the skin is restricted by both the ‘dead’ stratum corneum and the ‘live’ epidermis. Generally, only small water-soluble molecules can diffuse slowly into the skin. However, physical exfoliation of approximately 10 μm of skin increases the penetration of the active ingredients of topical skin-care products across the stratum corneum, especially if rough scar tissue is present. In addition to increasing the transport of molecules into the epidermis and dermis, exfoliation also levels and softens the surface of the skin, making it look and feel smoother (Figure 123.3).
1125
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1126
SECTION | II
Skin and Cosmeceuticals
FIGURE 123.2 Low-power (left) and medium-power (right) photomicrographs of thick skin on the sole of the foot. Left: The skin is composed of the epidermis (A) and dermis (B). The uneven boundary between the epidermis and dermis is the dermo–epidermal junction. The epidermis consists chiefly of keratin (stratum corneum). Beneath the dermis is the hypodermis. This contains adipose tissue, sweat glands (arrows) and a Vater-Pacini corpuscle (P). H and E stain. Low power (⫻40). Right: This figure shows the Malpighian layer (strata germinativum and spinosum), the stratum granulosum (arrows) and the stratum corneum of the epidermis. The duct of a sweat gland crosses the epidermis. The stratum lucidum is not visible in this section. The papillary layer of the dermis is seen in the lower left. H and E stain. Medium power (⫻100). (Adapted from Leeson et al., 2003.)
FIGURE 123.1 Diagram of the pilosebaceous unit and an eccrine sweat gland. The pilosebaceous unit consists of a hair follicle, an associated sebaceous gland, and erector pili muscle. (Adapted from Leeson et al., 2003.)
123.4 PATHOPHYSIOLOGY OF WRINKLES, PRURITUS AND XEROSIS 123.4.1 Wrinkles Wrinkles result from major cellular changes which occur within the skin. Cell turnover and regeneration gradually slows between the ages of 35 and 45 years; the sub-clinical phase of aging. The process of aging and the appearance of wrinkles generally increase after the age of 45; the clinical phase of aging. Very few people over the age of 50 have escaped wrinkles. The difference is only in the degree of the blemish. With increasing age the skin thins as cell turnover and regeneration slows. The normally undulating, ridge-like dermal–epidermal interface (DEI) flattens. This reduces the surface area available for the exchange of nutrients between the dermis below and the epidermis above. Thus aging reduces the supply of nutrients to the epidermis, accelerating
FIGURE 123.3 High-power (left) and oil immersion (right) photomicrographs of thick skin. Left: The stratum germinativum is a single layer of columnar cells. Each columnar cell has short cytoplasmic processes on its basal surface (arrowed). The stratum spinosum is several layers thick and composed of irregular, polyhedral cells, slightly separated from each other. The surface of the cells is covered with short cytoplasmic spines that meet similar projections from adjacent cells, forming intercellular bridges. The stratum granulosum consists of four to five layers of flattened cells that contain basophilic keratohyalin granules. The stratum lucidum (L) and the stratum corneum (C) are just visible. Plastic section. H and E stain. High power (⫻100). Right: Epidermis, stratum spinosum. The prickle cells have large nuclei with distinct nucleoli and show intercellular cytoplasmatic bridges. In two regions (arrowed), the cytoplasmic processes are shown in cross-section. Masson stain. Oil immersion (⫻200). (Adapted from Leeson et al., 2003.)
the degeneration of epidermal cells. Although the metabolism of epidermal cells slows, the elimination of the waste products of cellular metabolism (e.g. free radicals) is also reduced. Intracellular accumulation of free radicals can induce genetic mutations which may ultimately cause cancer.
1127
CHAPTER | 123 Effect of Olive Oil on the Skin
123.4.2 Collagen The dermis and epidermis are usually bound by collagen at the DEI. Collagen is a very important protein. It is the main constituent of the extracellular matrix (ECM) and has a critical role in the support structure of the skin. Simplistically, wrinkles are caused by a reduction in the collagen content of the skin (Varani et al., 2001). There are 19 forms of collagen which account for almost 20% of total body protein (Cawston, 1991). The most abundant is type I collagen (Cawston, 1991). Types IV and VII are particularly important at the dermal–epidermal junction (DEJ) (Reymermier et al., 2003). Type VII collagen is anchored to sheets of type IV collagen which forms a basal layer. The progressive loss of nutrients at the DEI reduces the circulation of the various factors that stimulate and support collagen synthesis. A vicious cycle ensues. Without sufficient collagen, the skin sags further, exacerbating the deficiency of nutrients (Burgeson, 1996). In addition to the loss of skin thickness due to lack of collagen, aging skin is looser and lacks elasticity. These are hallmarks of wrinkles. Matured aging skin contains more elastin than younger skin (Waller and Maibach, 2006). The elastin fills the space left by the deficiency of collagen. Unfortunately the elastin in aging skin is fragmented, calcified and contains excessive amounts of lipids (Waller and Maibach, 2006). The spatial arrangement of the collagen network also depends on the presence of supporting macromolecules known as proteoglycans and glycosaminoglycans (GAGs).
123.4.3 Glycosaminoglycans GAGs form a water-saturated gel in which hydrophilic molecules, hormones, peptides and ions circulate. In the process of aging, collagen is gradually replaced by the weaker GAGs (Bradbury and Parish, 1991). Collagen provides a much better support structure than GAGs. The reduction in the amount of collagen and replacement with weaker macromolecules results in thicker skin that is less elastic (Bradbury and Parish, 1991). With age the GAG gel tends to sag, further compromising cellular metabolism and mitosis.
123.4.4 Fibroblasts Fibroblasts synthesize and maintain the ECM. Fibroblasts provide a structural framework for the skin and play a critical role in wound healing. The main function of fibroblasts is to maintain the structural integrity of connective tissues by continuous secretion of collagen, the main constituent of the ECM. Fibroblasts in aged tissue produce less collagen (Navsaria, 2007). However, fibroblasts isolated from aged tissue can still produce significant amounts of collagen when stimulated by
endogenous factors such as transforming growth factor-beta (TGF-β) (Werner et al., 2007). Type I collagen is the predominant collagen in reconstructed healthy skin which is based on a good DEI with abundant types IV and VII collagen (Garrone et al., 1997). Successful stimulation of fibroblasts to produce more collagen and GAGs in vivo could potentially rejuvenate aging skin. Wrinkles will disappear and skin thickness will increase. Unsurprisingly, there is significant interest in the dermo-restitutive effects of olive oil in this area of skin care research. There is also considerable interest in the effects of olive oil on diseases of the skin.
123.4.5 Xerosis and Pruritus Xerosis (dry skin) and pruritus (itchiness) are the two most common dermatological conditions that affect the elderly. Xerosis, which has several potential causes, can result in complications such as pruritus and infection of the skin, particularly if left untreated. Pruritus may be caused by any of a number of infectious, hepatic, metabolic and hematological diseases. In addition, there are many dermatological causes of pruritus, the most common of which is xerosis (Simon et al., 2001). ‘Dead’ corneocytes on the surface of healthy skin detach from neighboring cells and are shed into the environment. These are replaced by younger cells from the deeper layers. This orderly process of desquamation is controlled by corneodesmosomes and lipids (Harding et al., 2000). Tissue thickness is maintained by the intercellular actions of these two components. Corneodesmosomes bind the corneocytes together maintaining intercellular cohesion and tissue integrity. Effective desquamation requires breakdown of corneodesmosomes by corneodesmolysis. In healthy skin, corneodesmolysis effectively eliminates the corneodesmosomes. In xerotic skin, corneodesmolysis is inefficient, so corneodesmosomes persist, disturbing the orderly process of desquamation (Harding et al., 2000). Free water is required for corneodesmolysis. Lipids retain water in the skin, and so are required for effective corneodesmolysis (Coderch et al., 2003). Dehydrated skin cannot provide this free water. Therefore, dehydration and/or a deficiency in lipid content can cause xerosis (Woodward et al., 1995; De Paepe et al., 2000; Yamaguchi et al., 2001).
123.5 BENEFICIAL PROPERTIES OF OLIVE OIL Olive oil can be used to treat both xerosis and pruritus (Herting, 1997). Administration of olive oil improved the
1128
symptoms of pruritus in patients with renal failure treated with hemodialysis (Herting, 1997). Pruritus in patients on maintenance hemodialysis is often due to dehydration of the stratum corneum. Treatment with simple emollients may relieve this (Morton et al., 1996). The improvement of pruritus in patients on hemodialysis was also noted with topical administration of fish oil and safflower oil on the skin (Herting, 1997). However, it is not impossible that olive oil may simply act as an emollient in this case. Peroxides are the main products of olive oil oxidation. However, the food peroxide content is often either disregarded or simply assumed to be low in studies of unsaturated dietary oils. Peroxides are not known to have any therapeutic value, it is therefore important to demonstrate that oils used to treat skin diseases have low peroxide contents. Antioxidants protect dietary oils from oxidation in situ and also prevent lipid peroxidation in vivo (Tsoureli-Nikita et al., 2002; Briganti and Picardo, 2003). Dietary ingestion of olive oil by atopic patients resulted in significant changes in plasma fatty acid (FA) profiles (Callaway et al., 2005). All lipid fractions were affected. The changes in plasma linoleic acid (LA), alpha-linolenic acid (ALA) and gamma-linolenic acid (GLA) were particularly significant (Callaway et al., 2005). Stearidonic acid (SDA) was detected in plasma, but could not be quantified because the amount present was below the threshold for quantification (Callaway et al., 2005). Administration of SDA, a rare omega-3 fatty acid, increases production of eicosapentaenoic acid (EPA) in vivo more than ALA, (James et al., 2003). Dietary supplementation with SDA (0.75–1.50 g) increased levels of EPA in both erythrocytes and plasma phospholipids (James et al., 2003). However, daily intake of 0.60 g SDA was not sufficient to significantly increase production of EPA or phospholipids (James et al., 2003). Ceramides may have an important role in the barrier function of skin and reduced levels of ceramides in the stratum corneum may be relevant in the etiology of atopic dermatitis. LA is esterified to ceramide 1 but oleic acid (OA) is not. OA could function as a molecular rivet in the stabilization of lipid lamellar sheets. Stabilization of lipid lamellar sheets in the skin could reduce the loss of water from the skin, especially in elderly people (Rogers et al., 1996). In view of the significant amount of OA present in olive oil, it is surprising that consumption of olive oil has little effect on the plasma profile of OA. OA is not an essential fatty acid (EFA) and so is not required for health. OA is not taken up as aggressively into plasma lipids as polyunsaturated fatty acids (PUFAs). Overall changes in fatty cholesterol esters were less robust than those for triglycerides or phospholipids. However, if symptoms of atopic dermatitis are more dependent on membrane function than eicosanoid production, then such an increase in PUFAs in phospholipid
SECTION | II
Skin and Cosmeceuticals
bilayers could effectively increase membrane fluidity and function (Callaway et al., 2005). Individuals who have used olive oil for several years report that the strength of their fingernails increased within months and the thickness of their hair increased within a few years (Janowiak and Ham, 2004). Improvements in skin conditions occurred within weeks of starting to use olive oil (Janowiak and Ham, 2004). The times taken for these effects to manifest correspond to the time required for newly formed cells to become physically apparent. These cell lines of the skin, fingernails and hair are developed from dermal stem cells using the FAs available at the time of their formation (Janowiak and Ham, 2004). Thus the formation of these tissues from dermal stem cells is complex and dependent on dietary FAs. The beneficial effects of dietary olive oil on skin, hair and nails, could be due, at least in part, to the large amounts of PUFAs present (⬎80%) (Kankaanpaa et al., 1999). These PUFAs have a metabolically favorable n-6/n-3 ratio of approximately 2:1 (Simopoulos, 2002). These FAs have important roles in the immune response (Harbridge, 1998). The FA profile of olive oil is very similar to blackcurrant seed oil, which has been reported to improve immune function (Barre, 2001). The presence of both GLA and SDA in olive oil (the metabolic products of the EFAs, LA and ALA respectively) bypasses delta-6-desaturase, the ratelimiting enzyme in production of EPA from ALA (James et al., 2003). This could explain the improvement of atopic symptoms associated with the use of olive oil. Healing of the skin is a highly ordered systemic process that requires coordination of cellular proliferation and migration as well as tissue remodeling (Burr and Burr, 1929). The skin reacts to messenger molecules (e.g. cytokines and chemokines) by producing biological response modifiers (e.g. monoclonal antibodies, recombinant cytokines or fusion proteins) which activate enzymes (e.g. collagenase and gellatinase) that recognize and degrade damaged cells and non-functional tissues (Villadsen et al., 2003). This process supports the immune system and repair mechanisms to treat injuries (cuts, bruises, abrasions, acne lesions, scars, wounds, burns), dysfunction (keratosis, keloid and hypertrophic scars, excess dryness) and prevent reactions such as immune-mediated inflammatory skin diseases (inflammatory acne, rosacea, eczema and most dermatitis) that damage the skin.
123.6 THE EFFECTS OF OLIVE OIL ON THE SKIN Although olive oil has only recently been included in modern cosmetics, this pleiotrophic oil has been used on the skin for thousands of years. Egyptian pharaohs used olive oil to moisturize their skin and hair and the Romans used the healing properties of olive oil to treat wounds.
CHAPTER | 123 Effect of Olive Oil on the Skin
1129
123.6.1 Antioxidant Properties of Olive Oil Extracts from olive oil can scavenge hydroxyl radicals better than other oils (Litridou et al., 1997). In addition to the direct antioxidant effects of olive oil, extracts are also potent inhibitors of xanthine oxidase activity. Olive oil contains significantly more squalene than seed oils. Squalene is largely transferred to the skin, probably by scavenging singlet oxygen generated by ultraviolet light (Kohno et al., 1995). Virgin olive oil applied to the skin after sunbathing could protect against skin cancer by slowing the growth of tumors (Abe et al., 1998).
123.6.2 Wound Healing We investigated the effects of ingestion of two fish oils from the Black Sea Coast on wound healing in vivo (Badiu et al., 2008). The experimental model of thermal burns induced on Wistar rats was used. Blood and tissue samples were extracted from the wounds at various intervals up to 22 days after the wound was inflicted. The fish oils reduced the time taken for the wounds to heal from 20–22 days in untreated animals to 12–16 days in treated animals. The full blood count parameters of the treated rats were normal (Badiu et al., 2008) including the white blood cell (WBC), lymphocyte, granulocyte and red blood cell (RBC) counts, hemoglobin (Hb), hematocrit (Hct) and mean corpuscle volume (MCV). However, the blood of untreated (control) rats had increased numbers of lymphocytes, eosinophils and monocytes (Badiu et al., 2008). Histological examination of the tissues of treated rats revealed moderate to complete re-epithelialization and minimal intercellular or subepithelial edema, without crusting (Badiu et al., 2008). In general, granulation tissue was characterized by dense collagen matrix deposition, slight edema and few scattered inflammatory infiltrates mainly confined to the deep dermis and the wound margins. The capillaries were significantly bigger than those found in untreated rats. Furthermore there was no evidence of fibrin deposition, hemorrhage or vascular congestion. Skeletal muscle cells formed a border between normal and wounded tissue (Figure 123.4). In contrast, the skin of the untreated rats was not completely healed (Badiu et al., 2008). Scar tissue and edematous, inflamed granulation tissue with little epithelial cover were seen. The collagenous connective tissue stroma was undeveloped with a moderate accumulation of granulation tissue and minimal substitution with adipose tissue. We recently investigated the anti-inflammatory and dermo-restitutive effects of olive oil in Wistar rats in vivo using the same model of induced thermal burn injury (unpublished data). The same histological and hematological analyses as described above were performed. The olive oil reduced the time taken for the wounds to heal to 14–16 days (unpublished data). Of the seven hematological parameters assessed, three (RBC, Hb and
FIGURE 123.4 High-power photomicrographs of the skin of Wistar rats treated with fish oil. H and E stain. Medium power (⫻100).
FIGURE 123.5 High-power photomicrographs of the skin of Wistar rats treated with olive oil. H and E stain. Medium power (⫻100).
Hct) were normal in the animals treated with olive oil (Figure 123.5, unpublished data). Histological analysis demonstrated partial reconstruction of the basal stratum from the epidermis, the blood vessels, collagen and fibroblasts of the dermis and hypodermis (unpublished data). The results of this study suggest that the anti-inflammatory and dermo-regenerative properties of olive oil were almost the same as those of the two fish oils described in our previous study. The mechanisms of these effects remain unclear but it is possible that the effects of olive oil on capillary blood flow and endothelial function are relevant.
123.6.3 Olive Oil and Endothelial Function Very few studies have addressed the effects of longterm olive oil consumption on endothelial function. The
1130
Mediterranean diet, rich in olive oil, improved endothelial function in patients with diabetes or hypercholesterolemia (Coulston et al., 1989), as assessed by measuring endothelium-dependent vasoreactivity. In a randomized cross-over trial, Perona et al. (2003) demonstrated an improvement in endothelial function in 22 hypercholesterolemic subjects fed a Mediterranean diet. However, the results were more significant when the dietary olive oil was partially replaced with walnuts. Perona et al. (2004) observed a greater improvement in flow-mediated dilatation (FMD) in patients on fluvastatin who were also advised to eat a Mediterranean-style diet. Esposito et al. (2004) conducted a randomized controlled trial in 180 subjects with the metabolic syndrome who were instructed to eat a Mediterranean-style diet, including olive oil. After 2 years of follow-up, they observed improved endothelial function as measured by reduction in blood pressure and platelet aggregation response to L-arginine, the precursor of nitric oxide (NO). They also reported a significant reduction of markers of systemic vascular inflammation, such as C-reactive protein and the interleukins (IL), IL-6, IL-7 and IL-18 as well as reduction in total and low-density lipoprotein (LDL) cholesterol. However, the mechanisms by which dietary olive oil produces these effects and the actual components of the oil responsible for these effects are unclear. The current understanding of the effects of the components of olive oil on endothelial activation is described below. It has been suggested that lipolytic enzymes are up-regulated as energy requirements increase during acute inflammation (Aller et al., 2004). This releases free FAs from circulating lipoproteins for use by tissues including the endothelium. The increase in endothelial free FA concentrations decreases endothelial NO bioactivity due to both superoxide generation and reduced endothelial NO synthase (eNOS) activity (Inoguchi et al., 2000). Studies in vitro suggest that PUFAs are more pro-inflammatory than monounsaturated fatty acids (MUFAs) and saturated fatty acids (SFA; Williams et al., 2004). In fact, LA (18:2, n-6) has greater capacity to induce oxidative and inflammatory stress than other FAs. Incubation of LA with endothelial cells promotes nuclear factor kappa B (NF-κB) activation and transcriptional activity (Toborek et al., 1996). This effect is attenuated by vitamin E (Toborek et al., 1996). In addition, exposure of endothelial cells to LA induces production of cytokines such as IL-6 and IL-8 which are involved in the initiation and progression of atherosclerosis (Yudkin et al., 2000). Conversely, n-3 PUFAs are believed to exert a protective effect on the endothelium. Particularly, docosahexaenoic acid (DHA; 22:6, n-3) which decreases the expression of vascular cell adhesion molecule 1 (VCAM-1) on the vascular endothelium, reducing monocyte adhesion (Khalfoun et al., 1996) and EPA (20:4, n-3) which increases NO production. Although the results from in vitro studies are promising, the results of in vivo studies are more
SECTION | II
Skin and Cosmeceuticals
controversial. There was no reduction in soluble adhesion molecules in patients receiving n-3 FAs after 6 weeks of treatment (Abe et al., 1998). However, intercellular adhesion molecule 1 (ICAM-1) and E-selectin were reduced after 7 months’ treatment n-3 FAs (Abe et al., 1998). Male smokers treated with n-3 FAs for 6 weeks had reduced prothrombogenic von Willebrand factor but increased VCAM1 and E-selectin (Seljeflot et al., 1998). Their results were corroborated by Johansen et al. (1999). The key inflammatory mediators released by the endothelium include the eicosanoids which are derived from the n-6 PUFA, arachidonic acid (AA; 20:4, n-6). Prostaglandin E2 (PGE2) can cause pain and vasodilation and leukotriene B4 (LTB4) is a chemo-attractant and activates neutrophils. PGE2 and LTB4 are formed from AA via cyclo-oxygenase (COX) and lipoxygenase (LOX) pathways, respectively. However, EPA is also a potential COX substrate and can compete with AA, leading to formation of PGE3. However, the synthesis of PGE3 from EPA is very inefficient (Hawkes et al., 1992). EPA is also a substrate for LOX, forming LTB5, with less inflammatory activity than LTB4 (James et al., 2003). Thus, increasing the n-3 FA content in the diet can shift the eicosanoids produced to those with less inflammatory effects. The skin is at high risk of oxidative damage from reactive oxygen species (ROS) for at least two reasons. Firstly, it is exposed to oxygen from a rich blood flow and the air at the surface of the skin. Secondly, a number of the physiological processes which occur in the skin including cellular metabolism and differentiation are photosensitive (Darr and Fridivich, 1994). The compounds in the skin which absorb light and act as photosensitizers generate ROS in the presence of ultraviolet light. Oxidative damage to epidermal cells (Moysan et al., 1993) and underlying connective tissue is recognized cosmetically as aging of the skin. The question is whether food components can modulate this actinic damage. Candidate nutrients that might prevent or reduce actinic damage include those which are present in sufficient quantity in skin and are either oxidizable, antioxidant or indirectly influence these activities. The FA composition of the epidermis is 25% unsaturated, relatively chemically unstable and susceptible to ROS (Bergfeld, 1999). There is considerable interest in the use of natural compounds in skin protection. Topical application of antioxidants, experimentally, indicates that they may usefully decrease photo-damage and associated inflammation (Moysan et al., 1993). The use of various antioxidants such as vitamin C (Darr et al., 1992; Simon et al., 2001), vitamin E (Ricciarelli et al., 1999) alone and in combination (Griffiths, 1999) as topical photo-protectants has been investigated. It has also been shown that hydroxytyrosol (HT) in extra virgin olive oil is highly protective against the DNA damage caused by peroxynitrite, which is formed by superoxide radicals and NO (Deiana et al., 1999). However, the effect of dietary intake on actinic skin damage has not been studied to date (Ricciarelli et al., 1999).
CHAPTER | 123 Effect of Olive Oil on the Skin
123.7 CONCLUSIONS AND FUTURE RESEARCH Cosmetic products are used to improve appearance and well-being. Cosmetics can be easily formulated from natural substances that are readily available, for example olive oil and olive extract. The popularity of cosmetics derived from natural sources is increasing. Such products are ecologically ‘ethical’ and are effective and safe to use. Vitamin E is the main lipophilic antioxidant that inhibits peroxidation, especially if associated with ‘natural’ moisturizers such as the lipids in olive oil and olive extract. Vegetable oils containing EFAs have proven to be of great use in the production of cosmetics as either active incipient or raw materials for the synthesis of novel compounds. EFAs are easily integrated into the skin’s hydro-lipid film and are nourishing, moisturizing and protective. Some of these substances have been used for centuries yet can still meet the needs of today’s consumers. Apart from their moisturizing and soothing effects, anti-inflammatory and immune-modulatory function these products reduce aging of the skin with their antioxidant, stabilizing action on the cellular membranes. Treatment with olive oil has no side effects. Olive oil does not burn or traumatize the skin. However, several synthetic compounds have been used in the development of cosmetics to adulterate the many essential oils derived from natural sources. However, lessons must be learnt from the previous failures of organic chemistry. Environmentally polluting chemicals approved by the Department of Transportation (DOT) were initially claimed to be safe. Several pharmaceutical drugs have been licensed only to be subsequently withdrawn due to their serious side effects. These failures demonstrate that although chemical principles may be useful in the development of synthetic cosmetics the potential for serious side effects must be considered.
SUMMARY POINTS ●
●
●
●
Human skin is a complex organ that regulates heat and water loss from the body whilst preventing the entry of toxic substances and microorganisms. Although olive oil has only recently been included in modern cosmetics, olive oil has been used on the skin for thousands of years. Xerosis (dry skin) and pruritus (itchiness) are the two most common dermatological conditions that affect the elderly. Both xerosis and pruritus conditions can be treated with olive oil.
REFERENCES Abe, Y., El-Masri, B., Kimball, K.T., Pownall, H., Reilly, C.F., Osmundsen, K., 1998. Soluble cell adhesion molecules in
1131
hypertriglyceridemia and potential significance on monocyte adhesion. Arterioscler. Thromb. Vas. Biol. 18, 723–731. Aller, M.A., Arias, J.L., Nava, M.P., Arias, J., 2004. Posttraumatic inflammation is a complex response based on the pathological expression of the nervous immune and endocrine functional systems. Exper. Biol. Med. 229, 170–181. Badiu, D.L., Balu, A.M., Barbes¸, L., Luque, R., Nit¸a˘, R., Radu, M., Tanase, E., Ros¸oiu, N., 2008. Physico-chemical characterisation of lipids from Mytilus galloprovincialis Lmk. and Rapana venosa and their healing preperties on skin burns. J. Lipids Res. 43, 829–841. Barre, D.E., 2001. Potential of evening primrose, borage, black currant, and fungal oils in human health. Ann. Nutr. Metab. 45, 47–57. Bergfeld, W.F., 1999. A lifetime of healthy skin: implications for women. Int. J. Fertil. Womens Med. 44, 83–95. Bradbury, M.G., Parish, C.R., 1991. Characterization of lymphocyte receptors for glycosaminoglycans. Immunology 72, 231–238. Briganti, S., Picardo, M., 2003. Antioxidant activity, lipid peroxidation and skin diseases. What’s new. J. Acad. Dermatol. Venereol. 17, 663–669. Burgeson, R.E., 1996. Laminins in epidermal structures. In: The laminins, edited by Peter Ekblom and Rupert Timpl, part of the book series: Cell Adhesion and Communication, 321. Burr, G.O., Burr, M.M., 1929. New deficiency disease produced by the rigid exclusion of fat from the diet. J. Biol. Chem. 82, 345–367. Callaway, J., Schwab, U., Harvima, I., Halonen, P., Mykkanen, O., Hyvonen, P., Jarvinen, T., 2005. Efficacy of dietary hempseed oil in patients with atopic dermatitis. J. Dermatol. Treat. 16, 87–94. Cawston, T., 1991. Arthritis and collagen connection. Collagen, the body’s most abundant protein, breaks down in arthritis. A better understanding of the cocktail of enzymes at work may lead to more effective drug. New Sci. Mag. 1772, 39. Available at: www.newscientist. com/article/mg13017725.600-arthritis-and-the-collagen-connectioncollagen-the-bodysmost-abundant-protein-breaks-down-in-arthritis-abetter-understandingofthe-cocktail-of-enzymes-at-work-may-lead-tomore-effective-drugs-.html. Coderch, L., Lopez, O., De La Maza, A., Parra, J.L., 2003. Ceramides and skin function (Review Article). Am. J. Clin. Dermatol. 4 (2), 107–129. Coulston, A.M., Hollenbeck, C.B., Swislocki, A.L., Reaven, G.M., 1989. Persistence of hypertriglyceridemic effect of low-fat high-carbohydrate diets in NIDDM patients. Diabetes Care 12 (2), 94–101. Darr, D., Combs, S., Dunston, S., Manning, T., Pinnell, S., 1992. Topical vitamin C protects swine skin from ultraviolet radiation-induced damage. Br. J. Dermatol. 127, 247–253. Darr, D., Fridivich, I., 1994. Free radicals in cutaneous biology. J. Invest. Dermatol. 102, 671–675. Deiana, M., Aruoma, O.L., Bianchi, M.L., Spencer, J.P., Kaur, H., Halliwell, B., Aeschbach, R., Banni, S., Dessi, M.A., Corongiu, F.P., 1999. Inhibition of peroxynitrite dependent DNA base modification and tyrosine nitration by the extra virgin olive oil-derived antioxidant hydroxytyrosol. Free Radic. Biol. Med. 26, 762–769. De Paepe, K., Derde, M., Roseeuw, D., Rogiers, V., 2000. Incorporation of ceramide 3B in dermatocosmetic emulsions : Effect on the transepidermal water loss of sodium lauryl sulphate-damaged skin. J. Eur. Acad. Dermatol. Venereol. 14 (4), 272–279. Esposito, K., Marfella, R., Ciotola, M., Di-Palo, C., Giugliano, F., Giugliano, G., 2004. Effect of Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome. J.A.M.A. 292, 1440–1446. Garrone, R., Lethias, C., Le Guellec, D., 1997. Distribution of minor collagens during skin development. Microsc. Res. Tech. 15; 38 (4), 407–412.
1132
Griffiths, C.E., 1999. Drug treatment of photoaged skin. Drug Aging 14, 289–301. Harbridge, L., 1998. Dietary n-6 and n-3 fatty acids in immunity and autoimmune disease. Proc. Nutr. Soc. 57, 555–562. Harding, C.R., Watkinson, A., Rawlings, A.V., Scott, I.R., 2000. Dry skin, moisturization and corneodesmolysis. Internat. J. Cosmet. Sci. 22 (1), 21–52. Hawkes, J.S., James, M.J., Cleland, L.G., 1992. Biological activity of prostaglandin E3 with regard to oedema formation in mice. Agents Actions 35, 85–87. Herting, Jr., R., 1997. Dermatology: pruritus. In: University of Iowa, Family Practice Handbook, 3rd edn. University of Iowa, Iowa City, pp. 1–13. Inoguchi, T., Li, P., Umeda, F., Yu, H.Y., Kakimoto, M., Imamura, M., 2000. High glucose level and free fatty acid stimulate reactive oxygen species production through protein kinase C-dependent activation of NAD(P)H oxidase in cultured vascular cells. Diabetes 49 (11), 1939–1945. James, M.J., Ursin, V.M., Cleland, L.G., 2003. Metabolism of stearidonic acid in human subjects: comparison with the metabolism of other n-3 fatty acids. Am. J. Clin. Nutr. 77, 1140–1145. Janowiak, J.J., Ham, C., 2004. A practitioner’s guide to hair loss, Part I: History, Biology, Genetics, prevention, conventional treatments and herbals. Altern. Compl. Ther. 10 (3), 135–143. Johansen, O., Seljeflot, I., Hostmark, A.T., Arnesen, H., 1999. The effect of supplementation with omega-3 fatty acids on soluble markers of endothelial function in patients with coronary heart disease. Arterioscler. Thromb. Vasc. Biol. 19, 1681–1686. Kankaanpaa, P., Sutas, Y., Salminen, S., Lichtenstein, A., Isolauri, E., 1999. Dietary fatty acids and allergy. Am. Med. 31, 182–287. Khalfoun, B., Thibault, G., Bardos, P., Lebranchu, Y., 1996. Docosahexaenoic and eicosapentaenoic acids inhibit in vitro human lymphocyte-endothelial adhesion. Transplantation 62, 1649–1657. Kohno, Y., Egawa, Y., Itoh, S., Nagaoka, S., Takahashi, M., Mukai, K., 1995. Kinetic study of quercing reaction of singlet oxygen and scavenging reaction of free radical by squalene in n-butanol. Biochim. Biophys. Acta 1256, 52–56. Leeson, T.S., Leeson, C.R., Paparo, A.A., 2003. The sin and its appendages (The Integument), In: Text atlas of histology, 456. Litridou, M., Linssen, J., Schols, H., Bergmans, M., Posthumus, M., Tsimidou, M., Boskou, D., 1997. Phenolic compounds in virgin olive oils: fractionation by solid phase extraction and antioxidant activity assessment. J. Sci. Food. Agric. 74, 169–174. Morton, C.A., Laafferty, M., Hau, C., 1996. Pruritus and skin hydration during dialysis. Nephrol. Dial. Transplant. 11 (10), 2031–2036. Moysan, A., Marquis, I., Gaboriau, F., Santus, F., Dubertet, L., Morliere, P., 1993. Ultraviolet A-induced lipid peroxidation and antioxidant defense system in cultured human skin fibroblast. J. Invest. Dermatol. 100, 692–698. Navsaria, H., 2007. Epithelial-mesenchymal interactions and tissue engineering. Br. J. Dermatol. 156 (6), 1149–1155. Perona, J.S., Canizares, J., Montero, E., Sanchez-Dominiguez, J.M., RuizGutierrez, V., 2003. Plasma lipid modifications in elderly people after administration of two virgin olive oils of the same variety (Olea europaea var. hojiblanca) with different triacylglycerol composition. Br. J. Nutr. 89 (6), 819–826. Perona, J.S., Canizares, J., Montero, E., Sanchez-Dominiguez, J.M., Catala, V., Ruiz-Gutierrez, V., 2004. Virgin olive oil reduces blood pressure in hypertensive elderly subjects. Clin. Nutr. 23 (5), 1113–1121.
SECTION | II
Skin and Cosmeceuticals
Reymermier, C., Guezennec, A., Branica, J.E., Guesnet, J., Pierrier, E., 2003. In vitro stimulation of synthesis of key DEJ constituents in a reconstructed skin model : a quantitative study. Intern. J. Cosmet. Sci. 25 (1–2), 55–62. Ricciarelli, R., Maroni, P., Ozer, N., Zingg, J.M., Azii, A., 1999. Agedependent increase of collagenase expression can be reduced by alphatocopherol via protein kinase C inhibition. Free Radic. Biol. Med. 27, 729–737. Rogers, J., Harding, C., Mayo, A., Banks, J., Rawlings, A., 1996. Stratum corneum lipids: the effect of ageing and the season. Arch. Dermatol. Res. 288, 765–770. Seljeflot, I., Arnesen, H., Brude, I.R., Nenseter, M.S., Drevon, C.A., Hjermann, I., 1998. Effects of omega-3 fatty acids and/or antioxidants on endothelial cell markers. Eur. J. Clin. Invest. 28, 629–635. Simon, M., Bernard, D., Minondo, A.M., 2001. Persistence of both peripheral and non-peripheral corneodesmosomes in the upper stratum corneum of wimter xerosis skin versus only peripheral in normal skin. J. Invest. Dermatol. 116 (1), 23–30. Simopoulos, A.P., 2002. The importance of the omega-6/omega-3 essential fatty acids. Biomed. Pharmacother. 56, 365–379. Swift, M.E., Burns, A.L., Gray, K.L., DiPietro, L.A., 2001. Age-related alterations in the inflammatory response to dermal injury. J. Invest. Dermatol. 117 (5), 1027–1035. Toborek, M., Barger, S.W., Mattson, M.P., Barve, S., McClain, C.J., Henning, B., 1996. Linoleic acid and TNF-alpha cross-amplify oxidative injury and dysfunction of endothelial cells. J. Lipid Res. 37 (1), 123–135. Tsoureli-Nikita, E., Hercogova, J., Lotti, T., Menchini, G., 2002. Evaluation of dietary intake of vitamin E in the treatment of atopic dermatitis: a study of the clinical course and evaluation of the immunoglobulin E serum levels. Int. J. Dermatol. 41, 146–150. Varani, J., Dara Spearman, D., Patricia Perone, P., Suzanne, E.G., Fligiel, S.E.G., Datta, S.C., Wang, Z.Q., Shao, Y., Kang, S., Fisher, G.J., Voorhees, J.J., 2001. Inhibition of type I procollagen synthesis by damagen collagen in photoaged skin and by collagenasedegraded collagen, in vitro. Am. J. Path. 158, 931–942. Villadsen, L.S., Skov, L., Baadsgaard, O., 2003. Biological response modifiers and their potential use in the treatment of inflammatory skin diseases. Exper. Dermatol. 12 (1), 1–10. Waller, J.M., Maibach, H.I., 2006. Age and skin structure and function, a quantitative approach (II): protein, glycosaminoglycan, water and lipid content and structure. Skin Res. Technol. 12 (3), 145–154. Werner, S., Krieg, T., Smola, H., 2007. Keratinocyte-fibroblast interactions in wound healing. J. Inv. Dermatol. 127, 998–1008. Williams, C.M., Maitin, V., Jackson, K.G., 2004. Triacylglycerol-rich lipoprotein-gene interactions in endothelial cells. Biochem. Soc. Trans. 32, 994–998. Woodward, D.F., Neives, A.L., Spada, C.S., 1995. Characterisation of a behavioral model for peripherally evoked itch suggests platelet-activating factor as a potent pruritogen. J. Pharmacol. Exp. Ther. 272 (2), 758–765. Yamaguchi, T., Maekawa, T., Nishikawa, Y., 2001. Characterisation of itch-associated responses of NC mice with mite-induced chronic dermatitis. J. Dermatol. Sci. 25 (1), 20–28. Yudkin, J.S., Kumari, M., Humphrie, S.E., Mohamed-Ali, V., 2000. Inflammation, obesity, stress and coronary heart disease: is interleukin-6 the link? Atherosclerosis 148, 209–214.
Chapter 124
Skin Creams Made with Olive Oil M. Adolfina Ruiz, José L. Arias and Visitación Gallardo Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, University of Granada, Spain
124.1 INTRODUCTION 124.1.1 The Skin The human skin is an important target for drug delivery and, despite the protective function of the skin against external attacks, many formulations are administered to induce drug adsorption, penetration or absorption process (Flynn, 1996). There are two important layers in the human skin: the epidermis and the dermis (Figure 124.1). The epidermis is the main barrier for drug absorption and does not contain blood vessels. It is divided into several sublayers where cells are formed through mitosis at the innermost structures. They
move up the strata changing shape and composition as they differentiate and become filled with keratin, finally reaching the stratum corneum. The stratum corneum is an unstructured layer that has no key role in drug absorption. The dermis is the layer of the skin beneath the epidermis, tightly connected to it by a basement membrane, and consists of connective tissue which protects the skin from stress and strain. It also harbors many mechanoreceptor/nerve endings that provide the sense of touch and heat, and contains hair follicles, sweat glands, sebaceous glands, apocrine glands, lymphatic vessels and blood vessels. The blood vessels in the dermis provide nourishment and waste removal to the epidermis.
Adsorption Absorption
Permeation
Stratum corneum Epidermis
Dermis Sebaceous glands
Hair follicles Subcutaneous tissue (hypodermis) Sweat glands
Blood vessels FIGURE 124.1 Structure of the skin, and drug absorption, adsorption and permeation processes. Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1133
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1134
SECTION | II
124.1.2 Absorption Routes The first step that follows a pharmaceutical formulation applied on the skin is the dissolution of the active agent in the stratum corneum. Once it is dissolved, the drug can (Figure 124.1): (i) develop its therapeutic action at the surface of the skin (drug adsorption by physicochemical bonds with cutaneous components of high molecular weight, e.g., keratin and melanin); (ii) continue to deeper histological layers by passive diffusion (drug permeation); or (iii) gain access to the blood (drug absorption: systemic effect) (Hsieh, 1994). Drug absorption to the blood capillaries can occur by means of a transepidermic or a transpedicular mechanism (Smith and Maibach, 1995). In the first case, polar drugs are absorbed through the lipids of the stratum corneum (transcellular route) and apolar drugs are absorbed through the lipids of the intercellular space (intercellular route). The transpedicular mechanism is followed by lipophilic molecules of great size and some electrolytes; however, this mechanism is less important due to the smaller surface available for absorption. Finally, the active agent can also be absorbed through the hair follicles or the sebaceous glands (transfollicular route), or through the sweat glands. The desired action in the skin and the permeation properties of the drug molecules through the skin structures are given in Table 124.1.
124.1.3 Factors Determining the Dermic Absorption The stratum corneum is the main barrier against dermic permeation and, therefore, the cutaneous absorption is limited by the permeation rate (Morimoto et al., 1996). The factors that affect this process are related to the drug (physicochemical properties and physic state), the vehicle (drug–excipient interaction) and the skin (hydration and health state).
Skin and Cosmeceuticals
The absorption of lipophilic compounds is favored by the lipid composition of the membranes and the intercellular spaces. However, the main limitation is the molecular weight and the molecular size of these lipophilic drugs. Thus, lipid molecules with short chain lengths are extremely volatile and will evaporate at the skin surface; nevertheless, lipophilic compounds with longer chain lengths will accumulate at the lipid zone of the skin without achieving deeper structures. The vehicle used in the formulation of the drug can modify the penetration of these molecules. Therefore, the solubility of the active agent and its concentration in the formulation will determine the permeation process. The permeation of lipophilic drugs is conditioned by the physicochemical properties of the vehicle: if it is hydrophilic, the permeation is favored. On the other hand, if a hydrophilic drug is formulated in a lipidic vehicle (e.g., olive oil) the absorption process will also be favored. The presence of hydrophilic and lipophilic domains in the stratum corneum determines this behavior. Thus, active agents with both characteristics (lipophilicity and hydrophilicity), a small molecular weight and a suitable solubility will undergo a fast permeation. As an example, Figure 124.2 shows the penetration capacity (permeability) of different alcohols as a function of the hydrophilic/lipophilic characteristics of the vehicle. As observed, the permeation capacity of these alcohols from a hydrophilic vehicle is higher as their aliphatic chain is longer, due to their higher lipophilicity. The best vehicles for the absorption of hydrophilic drugs are butanol, the mixture water:ether (1:1) and the mixture water:ethanol (1:1) as hydrophilic excipients, and olive oil and isopropyl palmitate as lipophilic vehicles.
124.2 OLIVE OIL (OLEUM OLIVAE ) Vegetable oils are widely used with cosmetic and dermocosmetic purposes (Juliano et al., 2005). One of their main applications is the use of the unsaponifiable fraction to
TABLE 124.1 Activity and permeation properties of drug molecules in the skin. Skin layer
Expected effect
Activity
Permeation
Stratum corneum
Protection, hydration, protection against sun damage
Cosmetic
⫺
Epidermis
Emollient, exfoliant, antiseptic
Cosmetic
⫹
Dermis, including sebaceous glands and sweat glands
Antiseptic, anesthetic, antiinflammatory, anti-histaminic
Local
⫹⫹
Hypodermis
Miorelaxing, analgesic, antiinflammatory
Local
⫹⫹⫹
Dermis: blood vessels
Hormonal, cardiovascular
Dermic or systemic
⫹⫹ ⫹ ⫹
1135
CHAPTER | 124 Skin Creams Made with Olive Oil
Oleic acid can act as a percutaneous absorption enhancer of several drugs, as it induces the disruption of the lipid structure of the stratum corneum, allowing drug permeation. This organic acid is able to increase the bioavailability of the corticoids betamethasone 17-benzoate and betamethasone 17-valerate (Table 124.2), ketorolac (Figure 124.3), metronidazole, hydrocortisone, progesterone and estradiol, to cite just a few (Bennet et al., 1985; Brian and Adrian, 1988; Yu et al., 1988). It is also used in the preparation of the benzyl benzoate lotion for the treatment of scabies. Oleic and linoleic acid can induce the in vitro suppression of the skin pigmentation.
Permeation (× 103 cm h–1) 50
Water
40 Isopropyl palmitate
30 20
Ethanol Propanol
10
Pentanol 1
2
3
4
5
Octanol 8
Number of carbons
FIGURE 124.2 Permeation (⫻103 cm h⫺1) of different alcohols from a hydrophilic (water: dotted line) and a lipophilic vehicle (isopropyl palmitate: solid line) as a function of their aliphatic chain length.
activate the cutaneous metabolism and, therefore, to induce emollience, hydration, dermoprotection or photoprotection. The most commonly used oils are the ones with the highest unsaponifiable content and the less expensive. The unsaponifiable content of some vegetable oils included in topical formulations are: karité lard (3.5–12%), avocado oil (3–10%), corn oil (1.5–3%), rice oil (1–3%), sesame oil (1–1.5%), soya oil (0.5–1.5%), olive oil (0.6–1.2%), peanut oil (0.2–0.9%), castor oil (0.3–0.8%) and coconut oil (0.1–0.3%). Olive oil, due to its high unsaponifiable fraction, is one of the most appreciated and promising vegetable oils in pharmaceutical technology, and can be found in many dermatological preparations, as well as in the form of ozonated olive oil. This vegetable oil is considered officinal in the majority of the international pharmacopeias. Olive oil is obtained mainly by cold expression of the ripe fruits of Olea europaea L. (Oleaceae). It occurs as a pale yellow or light greenishyellow oily liquid (density: 0.910–0.916) with a slight, characteristic odor and taste, and it also has a faint acrid aftertaste. It is water-insoluble, slightly soluble in ethanol, miscible with ether and slightly miscible with acetone. The acids included in its composition are oleic (65–80%), palmitic (7–20%), linoleic (4%), estearic (2–4%), miristic (1%) and, occasionally, lauric and arachidonic. It must be preserved from light and kept at less than 40°C (Ruiz et al., 1999; Loyd, 2000).
124.2.1 Oleic Acid Oleic acid is obtained by hydrolysis of olive oil. It is insoluble in water and very soluble in alcohol. This acid is used in the elaboration of topic formulations as excipient, e.g., of emulsions, due to its capacity to react with alkalis forming soaps with emulgent properties.
124.2.2 The Unsaponifiable Fraction The unsaponifiable fraction of olive oil can be considered as an active agent by itself and it is used in the elaboration of gels, creams and body milks for fragile, sensitive and dry skins. It is a regenerative and protective agent extensively used in the elaboration of anti-age formulations and sun-care products that are also used as soft conditioner for hair care. Because of the content in sterols, triterpenic alcohols and β-sitosterol (chemical structures close to corticoids), the dose-dependent anti-inflammatory action of these compounds is important. Moreover, oleuropein, tyrosol, squalene and the fraction of sterols and triterpenoid dialcohols of the unsaponifiable fraction obtained from extra virgin olive oil had been tested as cytostatic agents; however, only sterols and triterpenic dialcohols showed a promising activity (Saenz et al., 1998). Oleuropein is one of the components of the unsaponifiable fraction of olive oil used in the cosmetic treatment of devitalized and senile skins, and in the formulation of sunscreen compounds, because of its vessel-distending and antioxidant effects (Saija et al., 1998).
124.2.3 Other Components of Olive Oil Vitamins A, D and E, polyphenols, mineral salts, oleuropeoside, phytosterols, squalene and linoleic acid are other components of olive oil of interest, because of their good properties for health care. The great content in monounsaturated fatty acids and in antioxidative substances (e.g., vitamin E) justifies the indication of olive oil in childhood and old age. Moreover, olive oil contains oleocantal, a cyclooxygenase inhibitor with anti-inflammatory and analgesic properties very close to the non-steroidal anti-inflammatory drug ibuprofen. Despite the content in oleocantal of a daily ration of extra virgin olive oil (50 g) being equivalent to less than 10% of a normal dose of ibuprofen (9 mg), it can be responsible for the benefits associated to the diets based on olive oil, e.g., the reduction in cardiovascular events. Furthermore, the association of olive oil to other anti-inflammatory molecules should allow the use of lower doses of the latter, inducing
1136
SECTION | II
Skin and Cosmeceuticals
TABLE 124.2 Blanching responses to: a) betamethasone 17-benzoate solutions; and b) betamethasone 17-valerate creams in rank order of area under the curve (AUC) values expressed in terms of bioavailability. Preparation
Initial bioavailabilitya
First occlusion bioavailabilitya
(i)
(ii)
(i)
(ii)
2P
4.4
3.4
9.9
7.6
NMP
3.6
3.3
5.3
4.9
PG ⫹ OA
2.3
2.1
1.7
1.4
PG ⫹ A
2.2
1.9
0.88
1.3
DMF
2.1
1.9
1.4
1.9
PG
1.6
1.4
1.8
1.8
DMI ⫹ OA
1.0
0.92
1.7
1.5
DMI
1.0
1.0
1.0
1.0
DMI ⫹ A
0.90
0.83
0.56
0.69
BC ⫹ OA
1.0
1.1
2.7
2.3
BC*
1.0
1.0
1.0
1.0
BC
0.99
0.98
0.98
0.75
BC ⫹ A
0.94
1.1
4.0
Solutions
Creams
As observed, excepting DMI, the solvents tested are penetration enhancers. Oleic acid clearly increases the steroid bioavailability in comparison to DMI. Abbreviations used in the table: 2P: 2-pyrrolidone, NMP: N-methyl-2-pyrrolidone; DMF: dimethyl formamide; A: azone; OA: oleic acid; BC: betnovate cream; BC*: betnovate cream, control preparation; PG: propylene glycol; DMI: dimethylisosorbide. aFor steroid solutions, defined by the relations: (i) AUC for steroid solution/AUC for steroid in DMI; (ii) summed total possible score for steroid solution (%)/ summed total possible score for steroid in DMI (%). For betamethasone 17-valerate creams, similar relations are applied but with the denominator being the control BC. (Reproduced from Bennet et al., 1985, with permission from Pharmaceutical Press, Copyright 1985.)
a reduction in the incidence of adverse effects associated to these molecules (Briante et al., 2002a, b).
124.3 OLIVE OIL DERIVATIVES Novel olive oil derivatives had been developed for dermatological formulations (Amari et al., 1998; Rigano et al., 1999a, b). PEG-4 olivate is a non-ionic o/w emulgent characterized by its stabilizing action and by its capacity to induce the formation of very stable creams without the need for other emulsifiers. In addition, this compound is able to protect lipids from oxidation, and has a strong moisturizing effect due to its capability to reduce the transepidermal
water loss. The best results are achieved by using a PEG-4 olivate concentration of 10% and lipids of middle-polarity (i.e., yoyoba oil) or apolar (i.e., paraffin). Sorbitan olivate is a non-ionic w/o emulgent which is biodegradable, non-toxic and non-irritant. It is used at a concentration of 5–10% with oil phases of middle or low polarity (hydrocarbon esters or ethers) to obtain a stable emulsion. PEG-7 olive oil is an o/w co-emulgent and emollient, biodegradable and non-irritant, which is water-soluble and oil-soluble. It is widely used in the formulation of soaps and shampoos. Its emollient action reduces the irritation associated to surfactants. Sodium PEG-7 olive oil carboxilate is an anionic surfactant that can be used under acid or basic conditions. The
1137
CHAPTER | 124 Skin Creams Made with Olive Oil
8500 8000 7500
Plasma concentration (ng mL–1)
7000 6500
Patch removed
6000 5500 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50
Time (hours) FIGURE 124.3 Plasma levels of ketorolac and ketorolac tromethamine following a transdermal administration of ketorolac in propylene glycol/water (䊊), propylene glycol/linoleic acid (䊐), propylene glycol/oleic acid (Δ), propylene glycol/water/tween® 20 (ⵜ) and ethanol/water (䊉), and ketorolac tromethamine in propylene glycol/LA (䊏) and propylene glycol/oleic acid (䉱). (Reproduced from Yu et al., 1988, with kind permission of Springer Science and Business Media, Copyright 1988.)
surfactant effect is stable and similar to the one induced by sodium dodecyl sulfate (similar cleaning properties). Olivem® 700, 800, 900 and 1000 is a polyethylene glycol free o/w emulsifier that is used as a primary emulsifier in the preparation of o/w creams of high viscosity (3–5%) and o/w lotions of low viscosity (1.5–3%). It should also be mentioned that the molecule oleuropein and the unsaponifiable fraction of olive oil a very stable product with great affinity for the skin and very interesting moisturizing properties, as previously commented.
124.4 SKIN PATHOLOGIES AND THE TREATMENT WITH OLIVE OIL Olive oil is used in the preparation of many pharmaceutical formulations and cosmetics. Taking into account the zone in which the preparation is applied, formulations of diverse viscosity can be prepared by changing the gelificant agent. The use of olive oil in these preparations is justified by its protective or emollient properties and, therefore, it can act as an active agent by itself or as an excipient. Olive oil is used in the: (i) formulation of creams, gels and body lotions, because of its emollient and softener properties; (ii) treatment of eczemas, psoriasis, ulcers, burns and skin irritations, and in the formulation of after-suns, due to its regenerator, lenitive, soothing and healing properties; (iii) preparation of sun screen creams, because of its UV protective and anti-aging properties; and (iv) elaboration of bath
products, foam creams and hair conditioners because of its conditioner properties.
124.4.1 Therapeutic Action on the Skin The incorporation of drugs to semi-solid formulations is intended to achieve a local or transdermic effect. Several investigations had highlighted the utility of olive oil and oleic acid in topic formulations, as mere excipients or because of their direct action on the skin, e.g., skin oil formulations such as salicylic oil and camphorated oil. Due to its emollient action on the skin, olive oil is employed in the treatment of inflammatory processes associated to eczemas, scabs and psoriasis (Le Tutour and Guedon, 1992; Linos, 1999; Ruiz et al., 2003; Zamora et al., 2004). Very promising results had been obtained with olive oil microemulsions containing the anti-inflammatory drug aceclofenac (Shakeel et al., 2007), with triple emulsions (Figure 124.4A) (Ruiz et al., 2000; Huailiang et al., 2001) and with lipid vesicles as antigen delivery systems (Alcon et al., 2005). The microemulsions are formulated by using olive oil (2%), polyoxyl 40 hydrogenated castor oil (2%), methylparaben (0.15%), propylparaben (0.05%) and a phospholipid phase (soya phosphatidylcholine: Phospholipon® 90 H, Phospholipia® GmbH). Due to their influence on drug permeation processes, olive oil and oleic acid are used in the preparation of nanoparticles and microparticles as drug delivery systems (Figure 124.4B) (Cuevas et al., 2006). Gelatin-acacia
1138
SECTION | II
Skin and Cosmeceuticals
Colegio Oficial de Farmacéuticos de Vizcaya, 1997; Ruiz et al., 2003). Olive oil had been associated to vitamin E in the formulation of antiaging preparations (Muñoz et al., 2004). Moreover, the association of olive oil to well-known moisturizing agents, such as aloe vera, induces a higher moisturizing effect on skins. For example, the composition of an olive oil latex made of aloe vera can be: 5% aloe vera gel, 23% olive oil, 5% Olivem® 700, 0.3% n-decane, and up to 100% with distilled water (Ruiz et al., 2004).
A
B FIGURE 124.4 Triple emulsion made of olive oil (A) and olive oil microparticles (B) transmission electron microscopy photographs.
microcapsules containing olive oil have been prepared by phase separation. The effect of capsule size on the permeability of each capsule fraction was estimated from the change in electrical conductance with time of the microcapsule suspension and sodium chloride solution. Results showed that permeability decreased with decreasing capsule size (Jalsenjak and Kondo, 1981). Oleic acid is also used as solvent of several drugs, such as ketoprofen, genistein or daidzein, in order to achieve an enhancement in the permeation of these active agents (Kim et al., 1993; Minghetti et al., 2006). The use of oleic acid can also improve the skin permeation of melathonin and shorten its lag time more effectively than vehicles of various compositions (Han-Joon et al., 2001). The permeation profile of the anticancer drug tegafur across excised hairless mouse skin had been investigated (Figure 124.5), showing a positive effect of oleic acid as an excipient of topic formulations of this antineoplastic prodrug (Lee et al., 1993).
124.4.2 Cosmetic Applications Olive oil is a promising excipient widely formulated in cosmetics, such as soaps, creams, pomades, body milks, liniments, skin oils, and sunscreens preparations, where it acts as a penetration enhancer. It is also used to soften earwax, as massage oil and in the preparation of buccal deodorants (Le Tutour and Guedon, 1992; Castillo et al., 1994;
124.5 TOPICAL PHARMACEUTICAL FORMULATIONS MADE WITH OLIVE OIL The physician Galen (129–199 BC) designed the first moisturizing cream made of olive oil, using also water and vegetal wax. This cream demonstrated great moisturizing properties and increased the elasticity of skins. Olive oil has also been classically added to bath water. The formulation of olive oil with vaseline and menthol allows to obtain cosmetic products with good moisturizing properties. Olive oil is also used as massage lubricant due to its capability to induce the relaxation of muscles and nerves. It is also recommended by dermatologists in the treatment of hemorrhoids in pregnant women, ulcers, mycotic infections of the scalp (mixed with ketoconazole) and in the pre-treatment of scalp psoriasis (as a daily massage during 3 days before the instauration of the treatment with corticoids) (Chadwick, 1998; Loyd, 2003). This vegetable oil is also taken orally, inducing smoothness and brightness to the skin, hair and cuticles of the nails. One of the main applications of olive oil is the preparation of skin formulations with a protective and moisturizing action on the epidermis. There are several cosmetic products prepared with olive oil, such as baby foams (composed of lime tree extract, olive oil, panthenol and cleaning agents) or nutritive bath gels (composed of olive oil, vitamin E, conditioners, and natural and smooth cleaning agents) (Asin, 2008). Due to its moisturizing, keratolytic and keratoplastic properties, olive oil has been included in the formulation of salicylic oil as an excipient (salicylic concentration: 1, 2, 3, 5, 10 or 20%). This pharmaceutical preparation is indicated in the treatment of hyperkeratotic lesions such as scalp psoriasis or eczemas, and should be applied daily (at night, removing it in the morning) (Del Arco, 1994). Olive oil is also formulated with testosterone in topic preparations (testosterone concentration: 2%) for the treatment of vulvar hypertrophy and other postmenopausal disorders. Furthermore, the combination of olive oil and menthol (5– 10%) is extensively used in muscular pain relief. Finally, the use of olive oil as excipient of tretinoin creams increases their stability; for example, a commonly used composition of a lanolin excipient for these creams is: 65% lanolin, 15%
1139
Cumulative amount (μg cm2)
CHAPTER | 124 Skin Creams Made with Olive Oil
2500
2500
2000
2000
1500
1500
1000
1000
500
500
0
0
2
A
4 6 8 Time (hours)
10
0
12
0
2
B
4 6 8 Time (hours)
10
12
FIGURE 124.5 Permeation profiles of tegafur across excised hairless mouse skin from (A) ethanol/panasate 800 (40/60) and (B) ethanol/water (60/40) binary vehicle containing various fatty acids. Key: containing no fatty acids: (䊉) ethanol/panasate 800 (40/60), (䊊) ethanol/water (60/40); containing fatty acids: (䊐) C6, (䊏) C8, (Δ) C10, (䉱) C12, (䊐) C14, (Δ) C16, (⫻) C18, () C18:19, (⫹) C18:29,12. (Reprinted from Lee et al., 1993, with permission of John Wiley & Sons, Inc., Copyright 1993.)
olive oil and 20% distilled water (Brisaert and PlaizierVercammen, 2007). Table 124.3 shows commonly used formulations containing olive oil or oleic acid. Olive oil had been successfully investigated in the formulation of lipogels with controlled rheological properties for the delivery of anti-inflammatory drugs, such as meloxicam (Ruiz et al., 2007). Moreover, olive oil lipogels containing vitamin E had shown better release profiles than vitamin E hydrogels (Gallardo et al., 2005). The composition of both kinds of lipogels is depicted in Tables 124.4 and 124.5, respectively. Oleic acid is used in the preparation of scabies lotions and creams. The addition of a viscosity agent allows to control the viscosity of the formulation (Consejo General de Colegios Oficiales de Farmacéuticos de España, 1995).
124.6 OZONATED OLIVE OIL Ozone was discovered in 1840 by Christian Friedrich and has been known as a powerful oxidant agent capable of producing adverse effects in humans when inhaled in significant quantities. This damage is correlated with a rise in the infiltration of neutrophils, inflammatory mediators and cytokines. This molecule had been classically used as an antiseptic, in fistulas, lipomatosis, acne, eczemas, mycosis, scars, psoriasis, scalds, etc. (Ruiz et al., 1999). Ozonated olive oil is obtained when an ozone flow passed through olive oil for several days. It has a semisolid consistency, close to vaseline, with the odor of ozone. Ozonated olive oil is stable when kept under refrigeration, maintaining all the potency without stabilizers or preservatives. It can also be obtained as ‘oil sticks’ (small cylinders) for implants. The germicidal effect of ozonated olive oil against viruses, fungi or bacteria, and the tissue regeneration and blood stimulation properties is well known.
TABLE 124.3 Composition of pharmaceutical formulations containing olive oil or oleic acid. Sunscreen formulation Giv-Tan® F
4%
Liquid paraffin
50%
Olive oil
6%
Isopropyl myristate
40%
Salicylic oil Salicylic acid
5–10%
Castor oil
50%
Olive oil
up to 100 g
Anti-scabies lotion Benzyl benzoate
25%
Diethanolamine
0.5%
Oleic acid
2%
Distilled water
up to 100 g
In these preparations, olive oil and oleic acid are used as mere excipients or because of their direct action on the skin, e.g., as emollient.
Currently, it is used in the treatment of both contact and atopic dermatitis, such as seborrhea (Ruiz et al., 1999). The main indications of ozonated olive oil are herpes simplex (genital or labial, 2 or 3 times/day reduce the
1140
SECTION | II
Skin and Cosmeceuticals
TABLE 124.4 Composition of meloxicam lipogels for topical application. Formulation 1
Formulation 2
Formulation 3
Meloxicam
0.3%
0.3%
0.3%
Ethylcellulose
3–5%
3–5%
3–5%
Span® 80
5%
–
–
Olivem® 900
–
5%
–
Olivem® 700
–
–
5%
Olive oil
up to 100%
up to 100%
up to 100%
Each formulation allows to control the rheological properties for the delivery of this anti-inflammatory molecule.
TABLE 124.5 Composition of vitamin E lipogels for topical application.
observed), in gingivoestomatitis (a reduction in pain and fever is observed by two applications/day) and in diffuse external otitis caused by Pseudomonas aeruginosa, Enterobacter spp., Candida albicans, Klebsiella spp., Aspergillus spp. or Proteus spp., to mention just a few. This ozonated vegetable oil can be administered by the rectal and vaginal routes (Ruiz et al., 1999).
Formulation 1
Formulation 2
Vitamin E
2%
2%
Ethlylcellulose
3–5%
3–5%
Olivem® 900
5%
–
Olivem® 700
–
5%
SUMMARY POINTS
Olive oil
up to 100%
up to 100%
●
Both formulations allow controlled release profiles to be obtained in contrast to vitamin E hydrogels.
●
frequency of recidives and their intensity), acne (15–30 days of treatment, twice a day; even if acne is complicated with infections, fistula or quistic lesions) and skin infections caused by Staphylococcus spp., as well as cellulitis, impetigo, folliculitis, furuncles, carbuncles, infections of the sweat glands and of the nail bed. Ozonated olive oil is applied in several mycoses such as ‘athlete’s foot’ (due to Epidermophyton spp., Candida spp. or Pityrosporum spp.), epidermophytosis (twice a day assures the disappearance of pain and pruritus), onicomycosis and external mycosis. It is also used in the treatment of other skin-related problems (psoriasis and inflammations caused by drug intoxication), gastroduodenal ulcers and gastritis (capsules of ozonated olive oil), and to prevent secondary postoperative infections (Ruiz et al., 1999). The effectiveness of ozonated olive oil had also been probed in vulvo-vaginal infections (one application/day during 5 days assures the eradication of the infection and, only after 24 hours, an improvement of the symptoms is
●
●
●
●
Olive oil and its derivatives are extensively used in the preparation of many pharmaceutical formulations for therapeutic or cosmetic applications. When a drug is included in these preparations, a local or systemic effect is achieved. Because of its properties, olive oil can act as an active agent by itself or as an excipient of several formulations. Oleic acid and the unsaponifiable fraction are two components of olive oil very important in health care. The unsaponifiable fraction of olive oil can be considered as an active agent by itself, and oleic acid is used in the elaboration of topic formulations as excipient but can also act as a percutaneous absorption enhancer of several drugs. Olive oil also contains oleocantal, a cyclooxygenase inhibitor with anti-inflammatory and analgesic properties similar to the non-steroidal anti-inflammatory drug ibuprofen. Furthermore, olive oil and oleic acid have been tested in the preparation of drug delivery systems with very promising results. Ozonated olive oil is widely used in the treatment of diverse skin diseases such as dermatitis, herpes simplex, acne or cellulitis, to cite just a few.
CHAPTER | 124 Skin Creams Made with Olive Oil
REFERENCES Alcon, V., Baca-Estrada, M., Vega-Lopez, M., Willson, P., Babiuk, L.A., Kumar, P., Hecker, R., Foldvari, M., 2005. Mucosal delivery of bacterial antigens and CpG oligonucleotides formulated in biphasic lipid vesicles in pigs. AAPS Pharm. Sci. Tech. 7, 57–63. Amari, S., Orta, V., Roig, N., 1998. Olive oil derivatives: new components for cosmetic formulations (in Spanish). N.C.P. 236, 11–13. Asin, M., 2008. Procedures in cosmetic dermatology (in Spanish). Elsevier, Madrid, p. 81. Bennet, S.L., Barry, B.W., Woodford, R., 1985. Optimization of bioavailability of topical steroids: non occluded penetration enhancers under thermodynamic control. J. Pharm. Pharmacol. 37, 298–304. Brian, W.B., Adrian, C.W., 1988. Permeation enhancement through skin. In: Swarbrick, J., Boylan, J.C. (eds) Encyclopedia of Pharmaceutical Technology, Vol. 11. Marcel Dekker Inc., New York, p. 449. Briante, R., Patumi, N., Terenziani, S., Bismuto, E., Febbraio, F., Nucci, R., 2002a. Olea europaea L. leaf extract and derivatives: antioxidant properties. J. Agric. Food Chem. 14, 4934–4940. Briante, R., La Cara, F., Febbraio, F., Patumi, N., Nucci, R., 2002b. Bioactive derivatives from oleuropein by a biotransformation on Olea europaea leaf extracts. J. Biotechnol. 14, 109–119. Brisaert, M., Plaizier-Vercammen, J.A., 2007. Investigation on the photostability of tretinoin in creams. Int. J. Pharm. 334, 56–61. Castillo, A., Ibanez, F., Silva, J., 1994. Hirsutism and virilization caused by the administration of topical testosterone (in Spanish). Med. Clin. 102, 78–79. Chadwick, D., 1998. Rewards of treating decubitus ulcers. Int. J. Pharm. Compd. 4, 282–283. Colegio Oficial de Farmacéuticos de Vizcaya, 1997. Pharmaceutical compounding in primary health care (in Spanish). Colegio Oficial de Farmacéuticos de Vizcaya, Spain, p. 313. Consejo General de Colegios Oficiales de Farmacéuticos de España, 1995. Pharmaceutical formulary (in Spanish). Consejo General de Colegios Oficiales de Farmacéuticos, Madrid, p. 169. Cuevas, C., Gómez, A., Morales, M.E., Ruiz, M.A., 2006. Release of salicylic acid from an olive oil latex and a glucidic gel. Proceedings of the 5th World Meeting on Pharmaceutics, Biopharmaceutics and Pharmaceutical Technology, Geneve (Switzerland). Del Arco, J., 1994. Pharmaceutical compounding of medicines (in Spanish). Colegio Oficial de Farmacéuticos de Vizcaya, Spain, p. 50. Flynn, G.L., 1996. Cutaneous and transdermal delivery: processes and systems of delivery. In: Banker, G.S., Rhodes, C.T. (eds) Modern Pharmaceutics. Marcel Dekker Inc., New York, pp. 239–299. Gallardo, V., Muñoz, M., Ruiz, M.A., 2005. Formulations of hydrogels and lipogels with vitamin E. J. Cosmet. Dermat. 4, 187–192. Han-Joon, O., Yu-Kyoung, O., Chong-Kook, K., 2001. Effects of vehicles and enhancers on transdermal delivery of melatonin. Int. J. Pharm. 212, 63–71. Hsieh, D.S., 1994. Drug penetration enhacement. Marcel Dekker Inc., New York. Huailiang, W., Chandrasekharan, R., Weiner, N.D., Roessler, B.J., 2001. Topical transport of hydrophilic compounds using water-in-oil nanoemulsions. Int. J. Pharm. 220, 63–75. Jalsenjak, I., Kondo, T., 1981. Effect of capsule size on the permeability of gelatine acacia microcapsules toward sodium chloride. J. Pharm. Sci. 70, 456–457. Juliano, C., Cossu, M., Alamanni, M.C., Piu, L., 2005. Antioxidant activity of gamma-oryzanol: mechanism of action and its effect on oxidative stability of pharmaceutical oils. Int. J. Pharm. 299, 146–154.
1141
Kim, C.K., Kim, J.J., Chi, S.C., Shim, C.K., 1993. Effect of fatty acids and urea on the penetration of ketoprofen through rat skin. Int. J. Pharm. 99, 109–118. Le Tutour, B., Guedon, D., 1992. Antioxidative activities of Olea europaea leaves and related fenolic compounds. Phytochemistry 31, 1173–1178. Lee, C.K., Uchida, T., Noguchi, E., Kim, N.S., Goto, S., 1993. Skin permeation enhancement of tegafur by ethanol/panasate 800 or ethanol/ water binary vehicle and combined effect of fatty acids and fatty alcohols. J. Pharm. Sci. 82, 1155–1159. Linos, A., 1999. Dietary factors in relation to rheumatoid arthritis: a role for olive oil and cooked vegetables? Am. J. Clin. Nutr. 70, 1077–1082. Loyd, A., 2000. Oleaginous vehicles, featured excipient. Int. J. Pharm. Compd. 4, 470–471. Loyd, A., 2003. Natural emollient oil. Int. J. Pharm. Compd. 7, 462. Minghetti, P., Cilurzo, F., Casiraghi, A., Montanari, L., 2006. Evaluation of ex vivo human skin permeation of genistein and daidzein. Drug Deliv. 13, 411–415. Morimoto, K., Tojima, H., Haruta, T., Suzuki, M., Kakemi, M., 1996. Enhancing effects of unsaturated fatty acids with various structures on the permeation of indomethacin through rat skin. J. Pharm. Pharmacol. 48, 1133–1137. Muñoz, M., Lopez-Viota, M., Ruiz, M.A., 2004. New formulations of olive oil and vitamin E (in Spanish). Rev. Fitoter. 4, 173–174. Rigano, L., Trenti, R., Maramaldi, G., 1999a. Ottimizzazione applicativa di un emulsionanti derivato d⬘all olio di oliva. Cosmet. News 124, 42–46. Rigano, L., Leporatti, R., Maramaldi, G., 1999b. Olivem 900: Make up e skin care. Cosmet. News XXII (126), 211–215. Ruiz, M.A., Navarro, J.D., Gallardo, V., 1999. Dermatological applications of olive oil. J. Applied Cosmetol. 17, 19–22. Ruiz, M.A., Navarro, J.D., Ramos, M.M., Gallardo, V., 2000. Triple emulsions as drug carriers: a viscosimetry study (in Spanish). Proceedings of the 4th Meeting of the Colloids and Interfaces Specialized, Barcelona (Spain). Ruiz, M.A., Muñoz, M., Morales, M.E., Gallardo, V., 2003. Influence of the concentration of a gelling agent and the type of surfactant on the rheological characteristics of oleogels. Il Farmaco. 58, 1289–1294. Ruiz, M.A., Pleguezuelos, M., Muñoz, M., Gallardo, V., 2004. Moisturizing capacity of aloe vera gel in skin creams made with silicone-based and olive oil-based latex preparations. J. Appl. Cosmetol. 22, 25–33. Ruiz, M.A., López-Viota, J., Muñoz, M., García, J.D., Gallardo, V., 2007. Rheological behavior of gels and meloxicam release. Int. J. Pharm. 33, 17–23. Saija, A., Trombetta, D., Tomaino, A., Lo Cascio, R., Princi, P., Uccella, N., Bonina, F., Castelli, F., 1998. In vitro evaluation of the antioxidant activity and biomembrane interaction of the plant phenols oleuropein and hydroxytyrosol. Int. J. Pharm. 166, 123–133. Saenz, M.T., Garcia, M.D., Ahumada, M.C., Ruiz, V., 1998. Cytostatic activity of some compounds from the unsaponifiable fraction obtained from virgin olive oil. Il Farmaco. 53, 448–449. Shakeel, F., Baboota, S., Ahuja, A., Ali, J., Agil, M., Shafiq, S., 2007. Nanoemulsions as vehicles for transdermal delivery of aceclofenac. AAPS Pharm. Sci. Tech. 8, 104–113. Smith, E.W., Maibach, H.I., 1995. Percutaneous penetration enhancers. CRC Press, Boca Raton, Florida. Yu, D., Sanders, L.M., Davidson, G.W., Marvin, M.J., Ling, T., 1988. Percutaneous absorption of nicardipine and ketorolac in rhesus monkeys. Pharm. Res. 5, 457–462. Zamora, M.A., Bañez, F., Báñez, C., Alaminos, P., 2004. Olive oil: influence and benefits on some pathologies (in Spanish). Ann. Med. Interne 21, 138–142.
This page intentionally left blank
Chapter 125
Microarray Analysis of Hepatic Genes Altered in Response to Olive Oil Fractions María Victoria Martínez-Gracia1,2 and Jesús Osada1,2 1
Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Veterinaria, Universidad de Zaragoza CIBER de Fisiopatología de la Obesidad y Nutrición, Instituto de Salud Carlos III, Spain
2
125.1 INTRODUCTION The Mediterranean diet is associated with a reduced risk of cardiovascular mortality despite the high intake of fat, mainly derived from olive oil (Keys, 1995). Olive oil, as a fruit juice, is a complex mixture where triglycerides are combined with other biologically active substances such as tocopherols, phenolic compounds, phytosterols and triterpenoids, some of which have antioxidant and anti-inflammatory activities (Visioli et al., 2000; de la Puerta et al., 2001; Visioli et al., 2003; de la Puerta-Vazquez et al., 2004; Perona et al., 2006). The current view proposes that these components might be responsible for the benefits of virgin olive oil in animal models ( Calleja et al., 1999; Herrera et al., 2001; Rodriguez-Rodriguez et al., 2004; Acin et al., 2007) and in human studies (Ruiz-Gutierrez et al., 1996; Kris-Etherton et al., 1999; Abia et al., 2001; Perona et al., 2003, 2006). But still a lot of research needs to be done to understand their biological action.
125.2 THE LIVER, AN ORGAN SENSITIVE TO THE DIET COMPOSITION The liver secretes phospholipids, cholesterol and triglycerides into plasma as lipoprotein complexes (VLDL and HDL) which allow the transport of those lipids into the aqueous medium of blood. Apolipoproteins such as ApoB-100, ApoA-I, ApoA-II and ApoE are the main protein constituents of lipoproteins. Furthermore, this organ also secretes the enzymes (hepatic lipase, lecithin cholesterol acyl transferase and phospholipid transfer protein) involved in the plasma transformation of lipoproteins (den Boer et al., 2004). ApoE-deficient mice lack apolipoprotein E and as a consequence the elimination of lipoproteins from blood is impaired. Due to this fact, lipoproteins accumulate in walls Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
of vessels, contributing to the development of spontaneous atherosclerosis (Osada et al., 2000). When fed high-fat diets these mice responded, inducing changes in plasma apolipoproteins (Arbonés-Mainar et al., 2006), hepatic gene expression being responsible for these variations (Arbones-Mainar et al., 2006). The hepatic expression of apolipoprotein A-I was also modified in response to the type and amount of olive oil provided (Calleja et al., 1999; Acín et al., 2005). Thus, the liver may undergo important metabolic changes under the influence of different olive oils. Therefore, the liver is a central organ in the management of dietary lipids and an ideal model to verify subtle changes in diet composition due to its rapid adaptation response. To test the hypothesis that the unsaponifiable fraction of olive oil significantly influences hepatic gene expression, apoE-deficient mice of different genetic backgrounds (one from pure C57BL/6J mice and the other hybrid C57BL/6J x Ola129) were fed diets supplemented with either 10% (w/w) olive oil or 10% unsaponifiable fraction-enriched olive oil. Gene expression was then determined by microarray analysis that allows the global analysis of all mRNA present in a tissue and later confirmed by another procedure (quantitative RT-PCR) to reinforce the validity of results.
125.3 METHODOLOGICAL CONSIDERATION Two-month-old, male, homozygous apoE-deficient mice with different genetic backgrounds were used, and two study groups of equal plasma cholesterol were established: (a) one received a chow diet supplemented with 10% (w/w) olive oil (diet OO) (n ⫽ 9), and (b) the other received the same chow diet but supplemented with 10% (w/w) unsaponifiable-fraction enriched olive oil (diet UEOO; n ⫽ 8). This unsaponifiable-fraction enriched olive oil had greater
1143
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1144
SECTION | II Major Organ Systems Including Liver and Metabolism
Olive (9 mice)
UEOO (8 mice)
• RNA isolation from liver • Removal of contaminant DNA
1 pool RNA (9 mice/pool)
1 pool RNA (8 mice/pool)
Affymetrix array
Selection of genes: • Detection call, absorbance > 3 times noise • Signal log2 ratio > 3 or < −3
Primer design for qRT-PCR of selected genes
Olive individual mRNA expression by qRT-PCR
UEOO individual mRNA expression by qRT-PCR
FIGURE 125.1 Schematic workflow of the protocol. Diagram displaying the analytical steps used to prepare the RNA, its use to hybridize microarrays, critical criteria used to select meaningful genes and a final confirmation of individual samples by (qRT-PCR) quantitative reverse transcriptase-polymerase chain reaction. OO, olive oil; UEOO, unsaponifiable fraction-enriched olive oil.
quantities of phytosterols, waxes, triterpenes (erythrodiol, uvaol and maslinic) and tocopherols (Acin et al., 2007). For 11 weeks, the animals were fed the experimental diets which were well-tolerated. After this time period, animals were sacrificed and the liver removed, frozen in liquid nitrogen and used to extract its total RNA. The changes in hepatic gene expression induced by the unsaponifiable fraction of olive oil were tested by the expression of 22 690 transcripts represented on the Affymetrix GeneChip Murine Genome MOE430A array and quantified in pooled liver samples of nine animals that received the OO diet and another eight that received the UEOO diet. A summary of the approach is reflected in Figure 125.1. The huge amounts of information provided by microarrays requires further action be undertaken if meaningful and
manageable data are to be obtained, such as selecting only the genes with the highest expression changes (Dutta et al., 2003; Vergnes et al., 2003; Artieda et al., 2005; Calpe-Berdiel et al., 2005) or those involved in a certain metabolic pathway (Horton et al., 2003; Kreeft et al., 2005). In the present work, the first approach has been adopted and only those genes whose expression was strongly (signal log2 ratio higher or lower than 3 or ᎑3) modified were considered to be potential markers of the intake of the unsaponifiable fraction.
125.4 GLOBAL CHANGES IN GENE EXPRESSION IN LIVERS FED THE DIFFERENT DIETS USING MICROARRAYS The livers of OO animals expressed 10 455 transcripts, while those of the UEOO animals expressed 10 675 (identified as ‘present’ by Affymetrix software). Using the Mann–Whitney ranking feature of the Affymetrix software to determine significant differences in gene expression (p ⬍ 0.01), the increased expression of 660 genes plus the reduced expression of 324 genes was identified in samples from the animals on the UEOO diet compared to those on the OO diet. The complete datasets were deposited in the GEO database (accession number GSE2261). Differentially regulated genes with a signal log2 ratio higher than 3 (for those genes up-regulated) or lower than – 3 (for those repressed) were taken into account. Tables 125.1 and 125.2 list the genes whose mRNAs reflected these expressions. Six genes fulfilled the criterion of showing increased expression as a response to the unsaponifiable fraction of olive oil (Table 125.1). Two of these genes coded for acute phase proteins (Orm2 and Saa2), one was involved in signal transduction (Lepr), one coded for ion-binding proteins (Mt2), one for metabolite transport proteins (Fabp5), and finally one was an enzyme involved in nicotinamide metabolism (Nnmt). Six genes met the criterion of showing a reduced expression as a response to the presence of the unsaponifiable fraction of olive oil (Table 125.2). Of these, three were involved in proteolysis (Chym, Ela2 and Try4), one in lipid metabolism (Pnlip), one coded for an enzyme involved in carbohydrate metabolism (Gck) and one cell surface protein (Sycn). To validate the results obtained with the microarray, the expressions of these 11 genes – Chym, Ela2, Fabp5, Gck, Lepr, Mt2, Nnmt, Orm2, Pnlip, Saa2, Try4 – were individually studied by specific qRT-PCR assays previously described (Acín et al., 2007) and the results are shown in Table 125.3. The six up-regulated genes included in the validation analysis – Fabp5, Lepr, Mt2, Nnmt, Orm2, Saa2 – and the five down-regulated genes selected – Chym, Ela2, Gck, Pnlip, Try4 – were confirmed by these approaches. Good agreement was observed between the Affymetrix chip and qRT-PCR data (r ⫽ 0.9382, p ⬍ 0.0001). The present
1145
CHAPTER | 125 Microarray Analysis of Hepatic Genes Altered in Response to Olive Oil Fractions
TABLE 125.1 Hepatic genes up-regulated at the level of signal log2 ratio ⱖ 3 by the unsaponifiable fraction of olive oil. Biological process
GenBank
Affymetrix ID
Name
Gene Symbol
Signal transduction
U42467
1425644_at
Leptin receptor
Lepr
Acute phase response
NM_011016
1420438_at
Orosomucoid 2
Methyl transferase
AK006371
1432517_a_at
Acute phase response
NM_011314
Metal binding protein Transport (fatty acids)
Olive
UE olive
Signal log2 ratio
3
166
5.8
Orm2
71
1497
3.9
Nicotinamide Nmethyltransferase
Nnmt
107
1158
3.4
1449326_x_at
Serum amyloid A 2
Saa2
858
4292
3.3
AA796766
1428942_at
Metallothionein 2
Mt2
421
4657
3.1
BC002008
1416022_at
Fatty acid binding protein 5
Fabp5
33
238
3
Data represent intensity of signal for each condition with the Affymetrix chip. ID, identification; UE, unsaponifiable fraction-enriched. Adapted from Acín et al., Br J Nutr 97: 628–638, Copyright (2007), with permission from the authors.
TABLE 125.2 Hepatic genes down-regulated at the level of signal log2 ratio ⱕ ⫺ 3 by the unsaponifiable fraction of olive oil. Biological process
GenBank
Affymetrix ID
Name
Gene symbol
Olive
UE olive
Signal log2 ratio
Lipid metabolism
AI326372
1433431_at
Pancreatic lipase
Pnlip
858
79
⫺3.3
Proteolysis
AK003088
1428062_at
Carboxypeptidase A1 Cpa1
800
71
⫺3.3
Proteolysis
NM_025583
1448220_at
Chymotrypsinogen
Chym
778
59
⫺3.3
Proteolysis
NM_007919
1448281_a_at
Elastase 2
Ela2
666
80
⫺3.3
Carbohydrate metabolism
BC011139
1419146_a_at
Glucokinase
Gck
64
8
⫺3.2
Surface protein
BC019567
1451228_a_at
Syncollin
Sycn
243
14
⫺3
Data represent intensity of signal for each condition with the Affymetrix chip. ID, identification; UE, unsaponifiable fraction-enriched. Adapted from Acín et al., Br J Nutr 97: 628–638, copyright (2007), with permission from the authors.
data clearly show that pooling RNA from different animals and using this pool in microarray analysis is a reliable screening method for the search of biological effects in terms of saving samples, time and economic resources, as other authors have found (Napoli et al., 2002; Dutta et al., 2003; Artieda et al., 2005; Calpe-Berdiel et al., 2005; Kreeft et al., 2005). However, the main drawback of this approach is the lack of information on biological variability of individual samples. This limitation may be particularly important in the nutrition field in order to distinguish dietary responders and non-responders. Therefore, the experimenter should be aware of this caveat before deciding to pool samples.
125.5 PLASMA PRESENCE OF UNSAPONIFIABLE-ACTIVATED GENE PRODUCTS Since two of the overexpressed genes (Orm2 and Saa2) coded for circulating proteins that may influence general homeostasis, their plasma levels were determined to test the relevance of these mRNA changes at the protein level. Figure 125.2 shows the results of Western analysis of serum amyloid protein concentration. The observed increase of Saa mRNA expression (Table 125.1) was not reflected in plasma in either the OO or UEOO mice, suggesting the absence of an acute phase reaction. However, the increased
1146
SECTION | II Major Organ Systems Including Liver and Metabolism
TABLE 125.3 Hepatic genes regulated by the unsaponifiable fraction of olive oil in mice with C57BL/6J x Ola129 genetic background. Olive (n ⫽ 9)
UE olive (n ⫽ 8)
Fold change
Signal log2 ratio
Genes up-regulated Fabp5
1.0 ⫾ 0.8
5.5 ⫾ 0.4a
5.3
2.4
Lepr
0.4 ⫾ 0.5
4.1 ⫾ 0.2a
10.9
3.4
Mt2
0.5 ⫾ 0.6
13.6 ⫾ 1.5a
29.1
4.8
Nnmt
0.1 ⫾ 0.3
2.4 ⫾ 0.4a
16.8
4.1
Orm2
2.4 ⫾ 0.3
75.2 ⫾ 2a
31.8
5.0
Saa2
3.4 ⫾ 0.3
108.5 ⫾ 1a
32.2
5.0
Genes down-regulated Chym
35.1 ⫾ 0.1
5.7 ⫾ 0.3a
0.16
⫺2.6
Ela2
16.2 ⫾ 0.1
4.2 ⫾ 0.3a
0.26
⫺1.9
Gck
3.7 ⫾ 0.8
0.8 ⫾ 0.6a
0.21
⫺2.2
Pnlip
12.2 ⫾ 3
2.0 ⫾ 0.2a
0.16
⫺2.6
Try4
15.2 ⫾ 0.1
4.6 ⫾ 0.6a
0.30
⫺1.7
Data represent arbitrary units normalized to the β-actin expression for each condition with the quantitative reverse transcriptase–polymerase chain reaction procedure and are mean and standard deviation. a p ⬍ 0.001 according to Mann–Whitney U-test. UE, unsaponifiable fraction-enriched. Adapted from Acín et al., Br J Nutr 97: 628–638, copyright (2007), with permission from the authors.
1
2
3
4
SAA
14 kDa
Light Ig
22 kDa
FIGURE 125.2 Plasma SAA levels evidenced by Western blot analysis. Lane 1 shows a positive control corresponding to rat plasma from an animal treated with turpentine to induce plasma expression. Lanes 2–4, plasma from mice consuming the different diets: chow, OO diet and the UEOO diet respectively. Light chain immunoglobulin detection was used as a loading control. OO-olive oil; UEOO-unsaponifiable fractionenriched olive oil. Adapted from Acín et al, Br J Nutr 97: 628–638, copyright (2007), with permission from the authors.
hepatic Orm2 mRNA levels observed in the chip analysis and confirmed by qRT-PCR (Table 125.3) were reflected in the plasma concentration of animals consuming the UEOO diet (Figure 125.3A). The induction of orosomucoids has to date been attributed to acute-phase reactions (Hochepied et al., 2003). In this regard, the absence of any hepatic steatosis or inflammation described for this model (Acín et al., 2007), plus a lack of change in SAA after the administration of the unsaponifiable fraction, suggest that increased orosomucoid plasma levels are a unique response
elicited by these compounds via the induction of Orm2 expression. Recent studies have found that subjects with increased plasma concentrations of this protein have higher levels of vitamin A (Thurnham et al., 2003). In addition, in transgenic mice overexpressing Srebp1 and Srebp2, transcriptional factors involved in lipid metabolism, increased expression of this gene has also been described (Horton et al., 2003) although at a lower intensity than in mice consuming the UEOO diet. Together, these data suggest an unknown role for orosomucoid in the handling of unsaponifiable compounds.
125.6 EFFECTS OF MOUSE GENETIC BACKGROUND ON THE RESPONSE TO UNSAPONIFIABLE FRACTION-ENRICHED OLIVE OIL To investigate which of the selected genes were influenced by the genetic background, another study was carried out. In the latter, 12 homozygous apoE-deficient mice with a
1147
CHAPTER | 125 Microarray Analysis of Hepatic Genes Altered in Response to Olive Oil Fractions
* μg mL–1
800
*
*
*
Hybrid (C57BL/6J × Ola 129)
*
Pure (C57BL/6J)
6
600 400
4
200 0
A
8
2 Olive
UE Olive
0
AU
10
−2 −4
5
0
B
* Nnmt
Olive
UE Olive
FIGURE 125.3 Influence of genetic background on orosomucoid expression in apoE-deficient mice consuming the different diets. (A), plasma orosomucoid 2 levels in apoE-deficient mice with C57BL/6J x Ola129 genetic background. (B), orosomucoid 2 mRNA expression in livers of apoE-deficient mice with C57BL/6J genetic background (determined by qRT-PCR). Data represent mean and SEM for each group. Means were compared by the Mann–Whitney U-test. *p ⬍ 0.001 compared to results for OO diet. OO, olive oil; UEOO, unsaponifiable fraction-enriched olive oil. Adapted from Acín et al., Br J Nutr 97: 628–638, copyright (2007), with permission from the authors.
C57BL6J genetic background received the same diets used in the previous experiment and their livers were studied for the expression of these genes by qRT-PCR. Although the UEOO diets led to a reduction in hepatic Orm2 mRNA levels (Figure 125.3B) surprisingly this was not reflected at the plasma level (data not shown). The increase in hepatic Orm2 mRNA expression noted in the first strain of mice is probably a specific response elicited by the unsaponifiable fraction of olive oil, and is not related to an acute-phase response but conditioned by the genetic background of the mice. The results of the other genes, expressed as signal log2 ratios in both genetic backgrounds, are presented in Figure 125.4. Interestingly, lesser variability of response was observed in C57BL/6J background and no Pnlip expression was detected in the livers of these mice (data not shown). For the genes Lepr and Saa2, a significant opposite response was seen in mice of different genetic background. The expression levels of Lepr, Orm2 and Saa2 as markers of the presence of the unsaponifiable fraction of olive oil showed an interaction with genetic background. In contrast, seven of these genes – Chym, Ela2, Fabp5, Gck, Mt2, Nnmt and Try4 – were representative markers of the presence of the unsaponifiable fraction of olive oil in the diet, although the magnitude of response significantly differed among them (more pronounced for the Fabp5 gene in the C57BL/6J background, and less so in genes Gck, Nnmt and Try4 in the C57BL/6J x Ola129 background) and for other genes the influence of the UEOO diet was similar in both types of mice (Mt2, Chym and Ela2 expressions), so the genetic background did not influence the response.
Saa2
Mt2
Lepr
Fabp5
Gck
* Chym
Ela2
Try4
FIGURE 125.4 Influence of genetic background of apoE-deficient mice on the pattern of gene expression in response to the unsaponifiable fraction-enriched olive oil. Data as mean and SD for each group are expressed as signal log2 ratios of hepatic mRNA expression for each gene in apoEdeficient mice with C57BL/6J x Ola129 and C57BL/6J genetic backgrounds consuming either the OO or UEOO diet. Animals receiving the OO diet were used as the reference against which to compare the effects of the UEOO diet. *p ⬍ 0.001 between genetic backgrounds according to the Mann–Whitney U-test. OO, olive oil; UEOO, unsaponifiable fraction-enriched olive oil. Reprinted from Acín et al., Br J Nutr 97: 628–638, Copyright (2007), with permission from the authors.
The biological change produced by these variations in gene expression is complex. Thus, three of these genes – Chym, Ela2 and Try4 – are involved in proteolysis and showed reduced expression in the UEOO animals. Gck, an enzyme involved in glucose metabolism and also repressed in animals receiving high-fat diets (Dutta et al., 2003), showed similar behavior. The opposite (up-regulation) was observed for the expressions of Fabp5, Mt2 and Nnmt. Fabp5 (mal1) is considered to be an epidermal protein although it is also expressed in adipocytes (Maeda et al., 2003) and the liver (see GenBank accession AK167389 for a clone isolated from a liver cDNA library, and the present data). The exact role of this protein is not yet completely known, although it has been proposed to bind leukotriene A4 (Zimmer et al., 2004) and to play a role in systemic insulin sensitivity (Maeda et al., 2003). The change in its expression induced by the UEOO diet was particularly dramatic in the C57BL/6J animals. Mt2 is thought to be associated with obesity since knock-out mice lacking this gene develop this problem (Miura and Koizumi, 2005). In both studied substrates, the expression of this gene was upregulated (Figure 125.4). Nnmt has been recently associated with plasma homocysteine levels (Souto et al., 2005). Its genetic background-dependent response might explain the variation in homocysteine levels in different strains of mice. Taken as a whole, these results suggest that the unsaponifiable components of olive oil play an important role in controlling the expression of genes with participation in obesity, insulin sensitivity and cardiovascular risk factors, and that it deserves further attention.
1148
SECTION | II
SUMMARY POINTS ●
●
●
●
●
This nutrigenomic approach clearly illustrates the important effects of the unsaponifiable fraction of olive oil influencing the expression of hepatic genes, and provides further support for the idea that not all MUFAcontaining oils behave in the same way (Kritchevsky et al., 1984; Kris-Etherton et al., 1999). The results suggest that it is no longer appropriate to speak of monounsaturated fatty-acid-enriched oils (avocado, oleic acid-enriched safflower, oleic acid-enriched sunflower, olive and peanut oils) as though all had the same effects. Future studies should be aware of this to avoid confusion – both to researchers and consumers. This approach also shows new connections between nutrition and gene expression. A gene product with unknown biological function, orosomucoid, was up-regulated to an extent depending on the genetic background of the mice. Fabp5 and Mt2 were strongly up-regulated while the expression of several proteases was repressed by the UEOO diet. The modifications in gene expression could be used as markers of the intake of the unsaponifiable fraction of olive oil. The present results also show the usefulness of Affymetrix chip technology for characterizing gene expression levels in response to nutritional components in intact animal systems.
ACKNOWLEDGMENTS This research was supported by grant FEDER-CICYT (SAF2007-60173) and Redes DGA (B-69). We thank Drs. C. Junquera and L. Osaba of Progenika Biopharma for performing the microarray analyses. Thanks are also owed to Angel Beltrán, Jesús Cazo, Jesús Navarro, Carmen Navarro and Clara Tapia of the Unidad Mixta de Investigación for their invaluable help in maintaining the experimental animals.
REFERENCES Abia, R., Pacheco, Y.M., Perona, J.S., Montero, E., Muriana, F.J., RuizGutierrez, V., 2001. The metabolic availability of dietary triacylglycerols from two high oleic oils during the postprandial period does not depend on the amount of oleic acid ingested by healthy men. J. Nutr. 131, 59–65. Acín, S., Navarro, M.A., Carnicer, R., Arbonés, J.M., Guzmán, M.A., Arnal, C., Beltrán, G., Uceda, M., Maeda, N., Osada, J., 2005. Dietary cholesterol suppresses the ability of olive oil to delay the development of atherosclerotic lesions in apolipoprotein E knockout mice. Atherosclerosis 182, 17–28. Acin, S., Navarro, M.A., Perona, J.S., Arbones-Mainar, J.M., Surra, J.C., Guzman, M.A., Carnicer, R., Arnal, C., Orman, I., Segovia, J.C., Osada, J., Ruiz-Gutierrez, V., 2007. Olive oil preparation determines the atherosclerotic protection in apolipoprotein E knockout mice. J. Nutr. Biochem. 18, 418–424. Acín, S., Navarro, M.A., Perona, J.S., Surra, J.C., Guillen, N., Arnal, C., Sarría, A.J., Arbonés-Mainar, J.M., Carnicer, R., Ruiz-Gutiérrez, V.,
Major Organ Systems Including Liver and Metabolism
Osada, J., 2007. Microarray analysis of hepatic genes differentially expressed in the presence of the unsaponifiable fraction of olive oil in apolipoprotein E-deficient mice. Br. J. Nutr. 97, 628–638. Arbonés-Mainar, J.M., Navarro, M.A., Acín, S., Guzmán, M.A., Arnal, C., Surra, J.C., Carnicer, R., Roche, H.M., Osada, J., 2006. trans-10, cis-12and cis-9, trans-11-conjugated linoleic acid isomers selectively modify HDL-apolipoprotein composition in apolipoprotein E knockout mice. J. Nutr. 136, 353–359. Arbones-Mainar, J.M., Navarro, M.A., Guzman, M.A., Arnal, C., Surra, J.C., Acin, S., Carnicer, R., Osada, J., Roche, H.M., 2006. Selective effect of conjugated linoleic acid isomers on atherosclerotic lesion development in apolipoprotein E knockout mice. Atherosclerosis 189, 318–327. Artieda, M., Cenarro, A., Junquera, C., Lasierra, P., Martinez-Lorenzo, M.J., Pocovi, M., Civeira, F., 2005. Tendon xanthomas in familial hypercholesterolemia are associated with a differential inflammatory response of macrophages to oxidized LDL. FEBS Lett. 579, 4503–4512. Calpe-Berdiel, L., Escola-Gil, J.C., Ribas, V., Navarro-Sastre, A., GarcesGarces, J., Blanco-Vaca, F., 2005. Changes in intestinal and liver global gene expression in response to a phytosterol-enriched diet. Atherosclerosis 181, 75–85. Calleja, L., Paris, M.A., Paul, A., Vilella, E., Joven, J., Jimenez, A., Beltran, G., Uceda, M., Maeda, N., Osada, J., 1999. Low-cholesterol and high-fat diets reduce atherosclerotic lesion development in ApoEknockout mice. Arterioscler. Thromb. Vasc. Biol. 19, 2368–2375. de la Puerta-Vazquez, R., Martinez-Dominguez, E., Sanchez Perona, J., Ruiz-Gutierrez, V., 2004. Effects of different dietary oils on inflammatory mediator generation and fatty acid composition in rat neutrophils. Metabolism 53, 59–65. de la Puerta, R., Martinez Dominguez, M.E., Ruiz-Gutierrez, V., Flavill, J.A., Hoult, J.R., 2001. Effects of virgin olive oil phenolics on scavenging of reactive nitrogen species and upon nitrergic neurotransmission. Life Sci. 69, 1213–1222. den Boer, M., Voshol, P.J., Kuipers, F., Havekes, L.M., Romijn, J.A., 2004. Hepatic steatosis: a mediator of the metabolic syndrome. Lessons from animal models. Arterioscler. Thromb. Vasc. Biol. 24, 644–649. Dutta, R., Singh, U., Li, T.B., Fornage, M., Teng, B.B., 2003. Hepatic gene expression profiling reveals perturbed calcium signaling in a mouse model lacking both LDL receptor and Apobec1 genes. Atherosclerosis 169, 51–62. Herrera, M.D., Perez-Guerrero, C., Marhuenda, E., Ruiz-Gutierrez, V., 2001. Effects of dietary oleic-rich oils (virgin olive and high-oleicacid sunflower) on vascular reactivity in Wistar-Kyoto and spontaneously hypertensive rats. Br. J. Nutr. 86, 349–357. Hochepied, T., Berger, F.G., Baumann, H., Libert, C., 2003. Alpha(1)-acid glycoprotein: an acute phase protein with inflammatory and immunomodulating properties. Cytokine Growth Factor Rev. 14, 25–34. Horton, J.D., Shah, N.A., Warrington, J.A., Anderson, N.N., Park, S.W., Brown, M.S., Goldstein, J.L., 2003. Combined analysis of oligonucleotide microarray data from transgenic and knockout mice identifies direct SREBP target genes. Proc. Natl. Acad. Sci. U S A 100, 12027–12032. Keys, A., 1995. Mediterranean diet and public health: personal reflections. Am. J. Clin. Nutr. 61 (Suppl), 1321S–1323S. Kreeft, A.J., Moen, C.J., Porter, G., Kasanmoentalib, S., Sverdlov, R., van Gorp, P.J., Havekes, L.M., Frants, R.R., Hofker, M.H., 2005. Genomic analysis of the response of mouse models to high-fat feeding shows a major role of nuclear receptors in the simultaneous regulation of lipid and inflammatory genes. Atherosclerosis 182, 249–257. Kris-Etherton, P.M., Pearson, T.A., Wan, Y., Hargrove, R.L., Moriarty, K., Fishell, V., Etherton, T.D., 1999. High-monounsaturated fatty acid
CHAPTER | 125 Microarray Analysis of Hepatic Genes Altered in Response to Olive Oil Fractions
diets lower both plasma cholesterol and triacylglycerol concentrations. Am. J. Clin. Nutr. 70, 1009–1015. Kritchevsky, D., Tepper, S.A., Klurfeld, D.M., Vesselinovitch, D., Wissler, R.W., 1984. Experimental atherosclerosis in rabbits fed cholesterol-free diets. Part 12. Comparison of peanut and olive oils. Atherosclerosis 50, 253–259. Maeda, K., Uysal, K.T., Makowski, L., Gorgun, C.Z., Atsumi, G., Parker, R.A., Bruning, J., Hertzel, A.V., Bernlohr, D.A., Hotamisligil, G.S., 2003. Role of the fatty acid binding protein mal1 in obesity and insulin resistance. Diabetes 52, 300–307. Miura, N., Koizumi, S., 2005. Gene expression profiles in the liver and kidney of metallothionein-null mice. Biochem. Biophys. Res. Commun. 332, 949–955. Napoli, C., de Nigris, F., Welch, J.S., Calara, F.B., Stuart, R.O., Glass, C.K., Palinski, W., 2002. Maternal hypercholesterolemia during pregnancy promotes early atherogenesis in LDL receptor-deficient mice and alters aortic gene expression determined by microarray. Circulation 105, 1360–1367. Osada, J., Joven, J., Maeda, N., 2000. The value of apolipoprotein E knockout mice for studying the effects of dietary fat and cholesterol on atherogenesis. Curr. Opin. Lipidol. 11, 25–29. Perona, J.S., Cabello-Moruno, R., Ruiz-Gutiérrez, V., 2006. The role of virgin olive oil components in the modulation of endothelial function. J. Nutr. Biochem. 17, 429–445. Perona, J.S., Canizares, J., Montero, E., Sanchez-Dominguez, J.M., RuizGutierrez, V., 2003. Plasma lipid modifications in elderly people after administration of two virgin olive oils of the same variety (Olea europaea var. hojiblanca) with different triacylglycerol composition. Br. J. Nutr. 89, 819–826. Rodriguez-Rodriguez, R., Herrera, M.D., Perona, J.S., Ruiz-Gutierrez, V., 2004. Potential vasorelaxant effects of oleanolic acid and erythrodiol,
1149
two triterpenoids contained in ‘orujo’ olive oil, on rat aorta. Br. J. Nutr. 92, 635–642. Ruiz-Gutierrez, V., Muriana, F.J., Guerrero, A., Cert, A.M., Villar, J., 1996. Plasma lipids, erythrocyte membrane lipids and blood pressure of hypertensive women after ingestion of dietary oleic acid from two different sources. J. Hypertens. 14, 1483–1490. Souto, J.C., Blanco-Vaca, F., Soria, J.M., Buil, A., Almasy, L., OrdonezLlanos, J., Martin-Campos, J.M., Lathrop, M., Stone, W., Blangero, J., Fontcuberta, J., 2005. A genomewide exploration suggests a new candidate gene at chromosome 11q23 as the major determinant of plasma homocysteine levels: results from the GAIT project. Am. J. Hum. Genet. 76, 925–933. Thurnham, D.I., McCabe, G.P., Northrop-Clewes, C.A., Nestel, P., 2003. Effects of subclinical infection on plasma retinol concentrations and assessment of prevalence of vitamin A deficiency: meta-analysis. Lancet 362, 2052–2058. Vergnes, L., Phan, J., Strauss, M., Tafuri, S., Reue, K., 2003. Cholesterol and cholate components of an atherogenic diet induce distinct stages of hepatic inflammatory gene expression. J. Biol. Chem. 278, 42774–42784. Visioli, F., Galli, C., Grande, S., Colonnelli, K., Patelli, C., Galli, G., Caruso, D., 2003. Hydroxytyrosol excretion differs between rats and humans and depends on the vehicle of administration. J. Nutr. 133, 2612–2615. Visioli, F., Galli, C., Plasmati, E., Viappiani, S., Hernandez, A., Colombo, C., Sala, A., 2000. Olive phenol hydroxytyrosol prevents passive smokinginduced oxidative stress. Circulation 102, 2169–2171. Zimmer, J.S., Dyckes, D.F., Bernlohr, D.A., Murphy, R.C., 2004. Fatty acid binding proteins stabilize leukotriene A4: competition with arachidonic acid but not other lipoxygenase products. J. Lipid Res. 45, 2138–2144.
This page intentionally left blank
Chapter 126
Monounsaturated Fat Enriched with Olive Oil in Non-alcoholic Fatty Liver Disease Nimmer Assy1,4, Faris Nassar2,4 and Maria Grosovski3 1
Liver Unit, Ziv Medical Center, Safed, Israel Department of Medicine, Western Galilee Hospital, Nahariya, Israel 3 Department of Biotechnology, Ort-Braude College, Karmiel, Israel 4 Technion Institute of Technology, Rappaport Faculty of Medicine, Haifa, Israel
2
126.1 INTRODUCTION Normally ⬎5% of the liver mass is fat by weight, but in patients with non-alcoholic fatty liver disease (NAFLD) as much as 80% of the liver may be made up of fat, mostly in the form of triglycerides (Willner et al., 2001). NAFLD and NASH (non-alcoholic steatohepatitis) occur in 10–24% of the general population. The potential to progress to fibrosis (20–40%), cirrhosis (30%) and hepatocellular carcinoma (Bugianesi et al., 2002) makes these conditions clinically significant. ‘Fatty liver’ is the most common cause of cryptogenic cirrhosis and a common indication for liver transplantation. Obesity, diabetes and hyperlipidemia are conditions frequently associated with NAFLD (Assy et al., 2000). The pathogenesis of NAFLD includes insulin resistance, lipotoxicity, increased exposure of hepatocytes to TNF-α and increased oxidative stress (Postic and Girard, 2008). Etiologic mechanism of NAFLD includes increased influx of free fatty acids to the liver from dietary triglycerides and from free fatty acids that are released from adipocytes during fasting, reduced free fatty acid β-oxidation, reduced hepatic secretion of triglycerides-rich lipoprotein, and increased lipid per oxidation (Postic and Girard, 2008). An impaired postprandial triglyceride response has been recently reported in patients with NASH and may play a role by favoring triglyceride accumulation in the liver (Cassader et al., 2001). Diet and nutrition, in particular the amount and type of fat intake, were recently linked to insulin resistance, increased risk of developing type 2 diabetes and impaired postprandial lipid metabolism (Thomsen et al., 1999; Hu et al., 2001). Moreover, animal and human models suggest Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
that dietary factors can affect fatty infiltration and lipid peroxidation in different types of liver disease including NAFLD (Fernandez et al., 1997). Although few studies of the effects of different diets on NAFLD have been performed in humans, the Mediterranean diet has been proposed for the prevention of metabolic syndrome, hypertension, and cardiovascular disease (Martinez-Gonzalez and Sanchez-Villegas, 2004). The major part of its beneficial effect is a high supply of energy coming from monounsaturated fatty acids (MUFA), mainly from olive oil. In this review we describe dietary sources of MUFA, dietary habits and their relation to insulin resistance and postprandial glucose and triglyceride levels in NASH, the mechanism by which olive oil ameliorates fatty liver, experimental and clinical studies of olive oil and NAFLD and future perspectives.
126.2 COMPOSITION OF MONOUNSATURATED FATTY ACID (MUFA), OLIVE OIL Monounsaturated fatty acids (MUFA) include palmitic (C16:1), oleic (C18:1), elaidic (C18:1) and vacentic acids (C18:1). The most abundant MUFA in the diet is oleic acid (C18:1 n-9) (Table 126.1). In Mediterranean countries, the main source of MUFA is olive oil (74 g 100 g⫺1). Other oil sources of MUFA are canola (59 g 100 g⫺1), peanut (46 g 100 g⫺1), sunflower (32 g 100 g⫺1), corn (29 g 100 g⫺1), soybean (24 g 100 g⫺1) and safflower oil (14 g 100 g⫺1) (Nicklas et al., 2004). In addition to a high MUFA content,
1151
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1152
SECTION | II
TABLE 126.1 Fatty acid composition of 100 gram olive oil (gram/100 gram). Total MUFA monounsaturated fatty acids: 73.7 n-9 Oleic 18:1: 72.5
Major Organ Systems Including Liver and Metabolism
concentration increases. It is unclear how impairment in lipid export via VLDL secretion, β-oxidation of free fatty acid, or other metabolic pathway results in an inability to maintain fat balance, which leads to the development of fatty liver (Salmenniemi et al., 2004).
n-9 Palmitoleic 16:1: 1.2 Total SFA saturated fatty acids: 13.5 Palmitic acid 16:0: 11 Stearic 18:0: 2.2 Total PUFA: polyunsaturated fatty acids: 8.4 n-6 linoleic acid 18:2: 7.9 n-3 alpha-linoleic acid 18:3: 0.6
virgin (unrefined) olive oil contains a significant amount of antioxidants and α-tocopherol and phytochemicals. However, when refined or heated, olive oil loses this natural compound (Ros, 2003).
126.3 PATHOPHYSIOLOGY OF NAFLD, DIETARY FAT AND HEPATIC LIPIDS 126.3.1 Fat Metabolism in Fatty Liver Excessive inappropriate dietary fat intake combined with peripheral insulin resistance, continued triglyceride hydrolysis via lipoprotein lipase and other genetic alterations in key lipid metabolic pathways results in increased blood free fatty acid concentration (Sanyal et al., 2001) leading to excessive muscle fat accumulation and increased liver concentration of triglyceride and cholesterol esters. High blood triglyceride concentration in the form of VLDL tends to accompany this condition and induce cholesterol ester transfer protein (CETP) activity, resulting in an increased transfer of TG from VLDL to HDL and a subsequent increase in HDL clearance and decreased HDL concentration (Sanyal et al., 2001).
126.3.2 Insulin Resistance in Fatty Liver Peripheral insulin resistance affects carbohydrate and fat metabolism, causing triglyceride accumulation in the liver. Resistance to insulin stimulation of glucose uptake via glucose transporter-4 (GLUT-4) by skeletal muscle and adipose tissue in conjunction with insulin’s inhibition of lipolysis in adipose tissue diverts glucose to the liver where the insulin continues to stimulate de novo lipogenesis and increase the flux of fatty acids from adipose tissue to the liver (Hu et al., 2001; Salmenniemi et al., 2004). As a result, liver triglyceride
126.3.3 Fat Induces Hepatic Insulin Resistance The mechanism underlying fat-induced hepatic insulin resistance is not understood. Recent evidence points to an accumulation of fat metabolites that activate various serine/threonine kinases, (i.e., protein kinase-c , c-JUN NH2-terminal kinase-1 and IkBa kinase) as a key event in the pathway of fat-induced hepatic insulin resistance (Hu et al., 2001; Park and Giacca, 2007). Under conditions of insulin resistance, excess lipid metabolites such as diacylglycerol can cause insulin resistance by activating PKC which binds to an insulin receptor and inhibits its tyrosine kinase activity. The activation of PKC may also interfere with the ability of insulin to phosphorylate IRS-2 (Samuel et al., 2007). The principal triglyceride fatty acid esters present in normal liver are palmitate (16:0) and oleate (18:1 n-9). In patients with alcoholic fatty liver, the proportion of linoleate (C18:2 n-6) and linoleate (C18:3 n-3) decrease and the proportion of oleate (C18:1 n-9) increases compared with diabetics with fatty liver and control subjects who underwent liver biopsies (Cairns and Peters, 1983).
126.3.4 Dietary Habits and Relation to Insulin Resistance and Postprandial Lipemia Impaired postprandial triglyceride response has been reported in patients with NASH. This may promote the infiltration of fat in the liver by increasing triglyceride uptake in the postprandial period (Cooper, 1997). Enhanced lipogenesis appears as a prominent abnormality of hepatic fatty metabolism in subjects with NASH; the contribution of hepatic lipogenesis to triglyceride secretion was three times higher in patients with NAFLD as compared to healthy controls (Diraison et al., 2003). NASH patients had significantly higher overnight fasting glucose or free fatty acids than controls, as well as higher saturated and monounsaturated levels in both studied lipid fractions, mainly due to an increase in palmitate, palmitoleate and oleic acid (Araya et al., 2004). NASH patients showed depletion of polyunsaturated (n-3 and n-6) in liver triglycerols. This results from defective PUFA desaturation or from a higher lipid peroxidation (Araya et al., 2004). Diet from NAFLD patients who were free of hyperlipidemia, diabetes and obesity was richer in saturated fat and poor in PUFA (Musso et al., 2003) (Table 126.2). Finally, a MUFA-rich diet improves
CHAPTER | 126 Monounsaturated Fat Enriched with Olive Oil in Non-alcoholic Fatty Liver Disease
TABLE 126.2 Dietary intake in non-alcoholic steatohepatitis. Dietary component
NASH n ⴝ 25
Controls n ⴝ 25
Total kcal kg
33 ⫾ 5
32 ⫾ 6
Fiber (g)
13 ⫾ 4
23 ⫾ 8
102 ⫾ 31
92 ⫾ 35
SFA
40 ⫾ 13
29 ⫾ 11
MUFA
52 ⫾ 17
48 ⫾ 17
PUFA
10 ⫾ 5 (10% TF)
13 ⫾ 13 (15% TD)
Total fat (g)
0.24 ⫾ 0.1
0.45 ⫾ 0.1
C
84 ⫾ 43
114 ⫾ 63
E
5.4 ⫾ 2
8.7 ⫾ 3
Iron (mg)
12 ⫾ 2
1.5 ⫾ 4
P/S Vitamins (mg)
MUFA, monounsaturated fatty acids; P/S, polyunsaturated fatty acid ratio; PUFA, polyunsaturated fatty acids; SFA, saturated fatty acids. Data from Williams (2001).
postprandial glucose, lipid and glyp-1 responses in insulin-resistant subjects. Ingestion of olive-oil-based breakfast decreased postprandial glucose and insulin levels (Paniagua et al., 2007a, b).
126.4 MONOUNSATURATED FATTY ACIDS (MUFA) AND NAFLD The beneficial effect of MUFA such as those found in olive oil, nuts and avocados on cardiovascular diseases risk and lipid profile has been studied (Erkkilä et al., 2006). Dietary MUFA (oleic acid) decreased oxidized LDL (Lapointe et al., 2006), LDL cholesterol and triglyceride concentration without the concomitant decrease in HDL (Williams, 2001; Sacks, 2002). Additionally, the replacement of carbohydrate and saturated fat with MUFA leads to a reduction in glucose and blood pressure and to an increase in HDL in patients with diabetes (Julius, 2003). A MUFA-rich diet (40% of energy as fat), also decreases VLDL cholesterol and VLDL triglycerol and was acceptable to patients with NIDDM than was a higher carbohydrate diet (28% of energy as fat) (Rodríguez-Villar et al., 2004). A meta-analysis of studies with individuals with diabetes showed that a high-fat diet with 22–33% energy from MUFA resulted in lower plasma total cholesterol, VLDL, and triglyceride levels than did
1153
low-fat, high-carbohydrate (49% to 60% energy) diets (Garg, 1998). Therefore, an increase in intake of MUFA, particularly as a replacement for SFA and as a higher proportion in the diet instead of carbohydrate, may be beneficial for NAFLD patients. NAFLD, hypertension and hypertriglyceridemia are major components of the metabolic syndrome. Four clinical studies have documented the beneficial effect of monounsaturated fatty acid in decreasing blood pressure levels (Williams et al., 1987; Trevisan et al., 1990; Alonso and Martínez-González, 2004; Psaltopoulou et al., 2004). Moreover, another six feeding trials assessing the effect of monounsaturated fatty acid intake on blood pressure showed beneficial effect (Alonso et al., 2006). Although there are some inconsistencies in these studies, MUFA from olive oil in the context of Mediterranean diet plays a role in the primary prevention of NAFLD. Recently, we evaluated the effect of different types of dietary fats on the hepatic lipid content and oxidative stress parameters in rat liver with experimental NAFLD (Hussein et al., 2007). We demonstrated that olive oil decreases the accumulation of triglycerides in the livers of rats. Severe fatty liver was seen in methionine-choline-deficient diet (MCDD), MCDD ⫹ fish oil and in MCDD ⫹ butter fat groups, but not in MCDD ⫹ olive oil group (Table 126.3). Hepatic TG increase in MCDD ⫹ olive oil group was blunted by 30% compared with MCDD group. Serum triglyceride increase was lower by 10% in the MCDD ⫹ olive oil group compared with the MCDD group. In comparison with the control group, long-chain polyunsaturated fatty acid (PUFA) n6:n3 ratio increased in MCDD ⫹ olive oil, MCDD ⫹ fish oil and MCDD ⫹ butter fat groups by 345-, 30- and 397-fold, respectively. Hepatic MDA contents in MCDD ⫹ olive oil, MCDD ⫹ fish oil, and MCDD ⫹ butter group were higher by 108%, 91% and 87%, respectively, in comparison to the MCDD group. Olive oil improved insulin resistance, increased the release of triglyceride from the liver and decreased leak of free fatty acids from peripheral adipose tissue back to the liver (Hussein et al., 2007). A study from Spain showed that treatment with a balanced diet rich in olive oil contributes to the recovery of the liver from hepatic steatosis (Hernández et al., 2005). This was achieved by decreasing activation of hepatic stellate cell (decrease hepatic collagen) by monounsaturated fatty acid, which is less susceptible to lipid peroxidation as compared to polyunsaturated fatty acids. Moreover, previous studies carried out in fibrotic rats showed that olive oil, in contrast to polyunsaturated oils, could protect against the development of fibrosis (Szende et al., 1994). In animal studies, saturated fat significantly increased insulin resistance, long- and short-chain omega-3 fatty acids improved it, and the effect of monounsaturated fatty acids and omega-6 polyunsaturated fatty acids ranged somewhere between the two (Hussein et al., 2007). In humans, shifting from a diet rich in saturated fatty acids to one rich in monounsaturated
1154
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 126.3 Effect of methionine-choline-deficient diet (MCDD) enriched with different fats on fatty acids percentage in the rat liver. Fatty acids
Control
MCDD
MCDD ⴙ olive oil
MCDD ⴙ fish oil
MCDD ⴙ butter
P1 P2 (Friedman Test)
Components 0.5 ⫾ 0.1
0.3 ⫾ 0.2
0.4 ⫾ 0.2
0.4 ⫾ 0.2
0.003 0.002
17.0 ⫾ 1.1
15.3 ⫾ 1.4
15.5 ⫾ 1.1
15.7 ⫾ 2.3
0.001 0.006
0.1 ⫾ 0.2
0.7 ⫾ 0.1
0.4 ⫾ 0.2
0.7 ⫾ 0.1
0.4 ⫾ 0.2
0.007 0.02
21.8 ⫾ 1.6
5.8 ⫾ 0.6
5.7 ⫾ 0.7
6.2 ⫾ 0.5
6.1 ⫾ 0.9
0.002 0.01
C18:1 n9t Elaidic
2.1 ⫾ 0.1
1.5 ⫾ 1.3
22.8 ⫾ 1.3
17.8 ⫾ 1.1
21.3 ⫾ 1.1
0.001 0.006
C18:1 n9c Oleic1,2
4.9 ⫾ 0.8
20.7 ⫾ 0.7
25.9 ⫾ 3.3
2.0 ⫾ 0.1
5.3 ⫾ 8.7
0.001 0.006
18.6 ⫾ 1.8
32.4 ⫾ 0.7
30.5 ⫾ 1.7
29.5 ⫾ 2.1
31.5 ⫾ 2.1
0.001 0.006
C18:3 n3 Linolenic2
0.5 ⫾ 0.3
0.3 ⫾ 0.1
0.2 ⫾ 0.2
0.2 ⫾ 0.2
0.3 ⫾ 0.2
0.007 0.02
C23:0 Tricosanoic1,2
22.0 ⫾ 2.4
10.0 ⫾ 1.1
7.8 ⫾ 2.1
C14:0 Myristic C16:0 Palmitic1,2 C16:1 Palmitoleic C18:0 Stearic1,2
C18:2 n6c Linolelaidic1,2
0.2 ⫾ 0.2 19.0 ⫾ 1
C20:4 n6 Arachidonic1,2
0.1 ⫾ 0.2
C22:6 n3 Docosahexaenoic
4.5 ⫾ 0.9
C20:5 n3 Eicosapentaenoic
0
0 1.3 ⫾ 0.5 0.1 ⫾ 0.1
0
0
0.003 0.02
9.2 ⫾ 0.6
6.5 ⫾ 0.6
9.7 ⫾ 1.2
0.002 0.006
1.2 ⫾ 0.4
9.0 ⫾ 1.2
1.1 ⫾ 0.6
0.006 0.01
0
0.9 ⫾ 0.1
0
0.003 0.002
MCDD enrichment by olive oil increases the oleic acid, long-chain PUFA n6:n3 ratio, and arachidonic acid percentages in the rat livers. P1: difference between MCDD group and treatment groups; P2: difference between three treatment groups. 1 P ⬍ 0.02, MCDD ⫹ olive oil vs MCDD ⫹ fish oil. 2 P ⬍ 0.02 MCDD ⫹ olive oil vs MCDD ⫹ butter.
fatty acids improved insulin sensitivity in healthy people (Rivellese et al., 2002). A monounsaturated fat-rich diet prevented central body fat distribution, improved insulin sensitivity, and increased postprandial adiponectine expression compared to a carbohydrate-rich diet (with similar caloric intake) in insulin-resistant subjects (Paniagua et al., 2007). Furthermore, fasting plasma leptin fell during monounsaturated-rich diet subjects and has been associated with improved insulin action (Ruige et al., 1999). Weight maintenance with a MUFA-rich diet improves HOMA and fasting pro-insulin levels in insulin-resistant subjects. Ingestion of a virgin olive-oil-based breakfast decreased postprandial glucose and insulin concentrations, and increased HDL-C and GLP-1 concentrations as compared with a CHO-rich diet (Paniagua et al., 2007).
126.5 THE SPECIAL MECHANISM OF OLIVE OIL Olive oil has traditionally been the principal oil of the Mediterranean diet. A monounsaturated fat-rich diet prevents central body fat distribution and decreases postprandial
adiponectine expression induced by a carbohydrate-rich diet in insulin-resistant subjects (Paniagua et al., 2007). Mechanistic studies show a direct beneficial role for olive oil in improving plasma lipids and the treatment of metabolic syndrome (Alonso et al., 2006). Unrefined or virgin olive oil has bioactive compounds with beneficial actions (Kris-Etherton et al., 2002). The exact mechanism through which MUFA and olive oil could modify hepatic triglyceride content is not clear. Oleic acid from cooking oil was associated with lower insulin resistance in the general population (Soriguer et al., 2004). An olive oil-enriched diet contributes to redistribution of body fat and modifies lipolytic efficiency of fat cells (Soriguer et al., 2003). Furthermore, n-9 fatty acids may regulate gene expression related to peripheral insulin sensitivity (Clark et al., 2001), increased endothelial flow vasoreactivity (Ryan et al., 2000), induction of an up-regulating effect on uncoupling protein mRNA in adipose tissue and muscle (Rodríguez et al., 2002) and expression of GLUT-2 in the liver (Berry, 1997). Oleic acid decreases the expression of genes involved in hepatic gluconeogenesis and lipogenesis; sterol regulatory element binding protein (SREP) in Zucker fatty rats (Sato et al., 2007). Additional effects of olive oil beyond its MUFA composition include polyphenols.
CHAPTER | 126 Monounsaturated Fat Enriched with Olive Oil in Non-alcoholic Fatty Liver Disease
Polyphenols present in olive oil, such as oleuropein, hydroxytyrosol, tyrosol, and caffeic acid, have an important antioxidant and anti-inflammatory effect (Lampe, 1999). In rat leukocytes, these molecules have been shown to inhibit leukotriene B4 generation at the 5-lipoxygenase level and to reduce the generation of reactive oxygen species (de la Puerta et al., 1999). Moreover, a diet rich in olive oil improves endothelial function compared with a high-carbohydrate diet or a high-linoleic acid diet (Fuentes et al., 2001).
126.6 PERSPECTIVE Dietary fat content modified liver fat in overweight nondiabetic subjects. NAFLD patients have a higher postprandial TG response and an increased production of a large VLDL detected by an oral fat load compared with controls, despite normal fasting blood lipid concentration, which suggests that the metabolism of dietary fat is impaired in these individuals. Decreasing total fat consumption and shifting to monounsaturated fat found in olive oil (20–40% of total energy) or n-3 polyunsaturated fatty acids found in fish oil (2 g day⫺1) could lead to a decrease in postprandial lipidemia and steatosis. Further studies in humans are needed to ascertain whether the consumption of olive oil may be helpful in NAFLD patients.
REFERENCES Alonso, A., Martínez-González, M.A., 2004. Olive oil consumption and reduced incidence of hypertension, the SUN study. Lipids 39, 1233–1238. Alonso, A., Ruiz-Gutierrez, V., Martínez-González, M.A., 2006. Monounsaturated fatty acids, olive oil and blood pressure, epidemiological, clinical and experimental evidence. Public Health Nutr. 9, 251–257. Araya, J., Rodrigo, R., Videla, L.A., Thielemann, L., Orellana, M., Pettinelli, P., Poniachick, J., 2004. Increase in long-chain polyunsaturated fatty acid n-6/n-3 ratio in relation to hepatic steatosis in nonalcoholic fatty liver disease. Clin. Sci. 106, 635–643. Assy, N., Kaita, K., Mymin, D., Levy, C., Rosser, B., Minuk, G., 2000. Fatty infiltration of liver in hyperlipidemic patients. Dig. Dis. Sci. 45, 1929–1934. Berry, E.M., 1997. Dietary fatty acids in the management of diabetes mellitus. Am. J. Clin. Nutr. 66 (Suppl 4), 991S–997S. Bugianesi, E., Leone, N., Vanni, E., Marchesini, G., Brunello, F., Carucci, P., Musso, A., de Paolis, P., Capussotti, L., Salizzoni, M., Rizzetto, M., 2002. Expanding the natural history of nonalcoholic steatohepatitis, from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology 123, 134–140. Cairns, S.R., Peters, T.J., 1983. Biochemical analysis of hepatic lipid in alcoholic and diabetic and control subjects. Clin. Sci. (Lond.) 65, 645–652. Cassader, M., Gambino, R., Musso, G., Depetris, N., Mecca1, F., CavalloPerin, P., Pacini, G., Rizzetto, M., Pagano, G., 2001. Postprandial triglyceride-rich lipoprotein metabolism and insulin sensitivity in non alcoholic steatohepatitis patients. Lipids 36, 1117–1124.
1155
Clark, S.J., Shojaee-Moradie, F., Croos, P., Seed, P.T., Umpleby, A.M., Wendon, J.A., Miell, J., 2001. Temporal changes in insulin sensitivity following the development of acute liver failure secondary to acetaminophen. Hepatology 34, 109–115. Cooper, A.D., 1997. Hepatic uptake of chylomicron remnants. J. Lipid Res. 38, 2173–2192. de la Puerta, R., Ruiz Gutierrez, V., Hoult, J.R., 1999. Inhibition of leukocyte 5-lipoxygenase by phenolics from virgin olive oil. Biochem. Pharmacol. 57, 445–449. Diraison, F., Moulin, P., Beylot, M., 2003. Contribution of hepatic de novo lipogenesis and reesterification of plasma non esterified fatty acids to plasma triglyceride synthesis during non-alcoholic fatty liver disease. Diabetes Metab. 29, 478–485. Erkkilä, A.T., Matthan, N.R., Herrington, D.M., Lichtenstein, A.H., 2006. Higher plasma docosahexaenoic acid is associated with reduced progression of coronary atherosclerosis in women with CAD. J. Lipid. Res. 47, 2814–2819. Fernandez, M.I., Torres, M., Gil, A., Rios, A., 1997. Steatosis and collagen content in experimental liver cirrhosis are affected by dietary monounsaturated and polyunsaturated fatty acids. Scand. J. Gastroenterology 32, 350–356. Fuentes, F., López-Miranda, J., Sánchez, E., Sánchez, F., Paez, J., PazRojas, E., Marín, C., Gómez, P., Jimenez-Perepérez, J., Ordovás, J.M., Pérez-Jiménez, F., 2001. Mediterranean and low-fat diets improve endothelial function in hypercholesterolemic men. Ann. Intern. Med. 134, 1115–1119. Garg, A., 1998. High-monounsaturated-fat diets for patients with diabetes mellitus, a meta-analysis. Am. J. Clin. Nutr. 67 (Suppl 3), 577S–582S. Hernández, R., Martínez-Lara, E., Cañuelo, A., del Moral, M.L., Blanco, S., Siles, E., Jiménez, A., Pedrosa, J.A., Peinado, M.A., 2005. Steatosis recovery after treatment with a balanced sunflower or olive oil-based diet, involvement of perisinusoidal stellate cells. World J. Gastroenterol. 11, 7480–7485. Hu, F.B., Van Dam, R.M., Liu, S., 2001. Diet and risk of type II diabetes, the role of type of fat and carbohydrate. Diabetologia 44, 805–817. Hussein, O., Grosovski, M., Lasri, E., Szvalb, S., Ravid, U., Assy, N., 2007. Monounsaturated fat decreases hepatic lipid content in non-alcoholic fatty liver disease in rats. World J. Gastroenterol. 13, 361–368. Julius, U., 2003. Influence of plasma free fatty acids on lipoprotein synthesis and diabetic dyslipidemia. Exp. Clin. Endocrinol. Diabetes 111, 246–250. Kris-Etherton, P.M., Hecker, K.D., Bonanome, A., Coval, S.M., Binkoski, A.E., Hilpert, K.F., Griel, A.E., Etherton, T.D., 2002. Bioactive compounds in foods, their role in the prevention of cardiovascular disease and cancer. Am. J. Med. 113 (Suppl 9B), 71S–88S. Lampe, J.W., 1999. Health effects of vegetables and fruit, assessing mechanisms of action in human experimental studies. Am. J. Clin. Nutr. 70 (Suppl 3), 475S–490S. Lapointe, A., Couillard, C., Lemieux, S., 2006. Effects of dietary factors on oxidation of low-density lipoprotein particles. J. Nutr. Biochem. 17, 645–658. Martinez-Gonzalez, M.A., Sanchez-Villegas, A., 2004. The emerging role of Meditarranean diets in cardiovascular epidemiology, monounsaturated fat, olive oil, red wine, or the whole pattern. Eur. J. Epidemiol. 19, 9–13. Musso, G., Gambino, R., De Michieli, F., Cassader, M., Rizzetto, M., Durazzo, M., Fagà, E., Silli, B., Pagano, G., 2003. Dietary habits and their relations to insulin resistance and postprandial lipemia in nonalcoholic steatohepatitis. Hepatology 37, 909–916.
1156
SECTION | II
Nicklas, T.A., Hampl, J.S., Taylor, C.A., Thompson, V.J., Heird, W.C., 2004. Monounsaturated fatty acid intake by childrens and adults, temporal trends and demographic differences. Nutr. Rev. 62, 132–141. Paniagua, J.A., de la Sacristana, A.G., Sánchez, E., Romero, I., VidalPuig, A., Berral, F.J., Escribano, A., Moyano, M.J., Peréz-Martinez, P., López-Miranda, J., Pérez-Jiménez, F., 2007a. A MUFA-rich diet improves posprandial glucose, lipid and GLP-1 responses in insulinresistant subjects. J. Am. Coll. Nutr. 26, 434–444. Paniagua, J.A., Gallego de la Sacristana, A., Romero, I., Vidal-Puig, A., Latre, J.M., Sanchez, E., Perez-Martinez, P., Lopez-Miranda, J., Perez-Jimenez, F., 2007b. Monounsaturated fat-rich diet prevents central body fat distribution and decreases postprandial adiponectin expression induced by a carbohydrate-rich diet in insulin-resistant subjects. Diabetes Care 30, 1717–1723. Park, E., Giacca, A., 2007. Mechanism underlying fat induced hepatic insulin resistance. Future Lipiol. 2, 503–512. Postic, C., Girard, J., 2008. Contribution of de novo fatty acid synthesis to hepatic steatosis and insulin resistance, lessons from genetically engineered mice. J. Clin. Invest. 118, 829–838. Psaltopoulou, T., Naska, A., Orfanos, P., Trichopoulos, D., Mountokalakis, T., Trichopoulou, A., 2004. Olive oil, the Mediterranean diet, and arterial blood pressure, the Greek European Prospective Investigation into Cancer and Nutrition (EPIC) study. Am. J. Clin. Nutr. 80, 1012–1018. Rivellese, A.A., De Natale, C., Lilli, S., 2002. Type of dietary fat and insulin resistance. Ann. N. Y. Acad. Sci. 967, 329–335. Rodríguez, V.M., Portillo, M.P., Picó, C., Macarulla, M.T., Palou, A., 2002. Olive oil feeding up-regulates uncoupling protein genes in rat brown adipose tissue and skeletal muscle. Am. J. Clin. Nutr. 75, 213–220. Rodríguez-Villar, C., Pérez-Heras, A., Mercadé, I., Casals, E., Ros, E., 2004. Comparison of a high-carbohydrate and a highmonounsaturated fat, olive oil-rich diet on the susceptibility of LDL to oxidative modification in subjects with Type 2 diabetes mellitus. Diabet. Med. 21, 142–149. Ros, E., 2003. Dietary cis-monounsaturated fatty acids and metabolic control in type 2 diabetes. Am. J. Clin. Nutr. 78 (Suppl), 617S–625S. Ruige, J.B., Dekker, J.M., Blum, W.F., Stehouwer, C.D., Nijpels, G., Mooy, J., Kostense, P.J., Bouter, L.M., Heine, R.J., 1999. Leptin and variables of body adiposity, energy balance, and insulin resistance in a population-based study. The Hoorn Study. Diabetes Care 22, 1097–1104. Ryan, M., McInerney, D., Owens, D., Collins, P., Johnson, A., Tomkin, G.H., 2000. Diabetes and the Mediterranean diet, a beneficial effect of oleic acid on insulin sensitivity, adipocyte glucose transport and endothelium-dependent vasoreactivity. QJM 93, 85–91. Sacks, F.M., 2002. Dietary fat, the Mediterranean diet, and health, reports from scientific exchanges, 1998 and 2000. Introduction. Am. J. Med. 113 (Suppl 9B), 1S–4S. Salmenniemi, U., Ruotsalainen, E., Pihlajamäki, J., Vauhkonen, I., Kainulainen, S., Punnonen, K., Vanninen, E., Laakso, M., 2004.
Major Organ Systems Including Liver and Metabolism
Multiple abnormalities in glucose and energy metabolism and coordinated changes in levels of adiponectin, cytokines, and adhesion molecules in subjects with metabolic syndrome. Circulation 110, 3842–3848. Samuel, V.T., Liu, Z.X., Wang, A., Beddow, S.A., Geisler, J.G., Kahn, M., Zhang, X.M., Monia, B.P., Bhanot, S., Shulman, G.I., 2007. Inhibition of protein kinase Cepsilon prevents hepatic insulin resistance in nonalcoholic fatty liver disease. Clin. Invest. 117, 739–745. Sanyal, A.J., Campbell-Sargent, C., Mirshahi, F., Rizzo, W.B., Contos, M.J., Sterling, R.K., Luketic, V.A., Shiffman, M.L., Clore, J.N., 2001. Nonalcoholic steatohepatitis, association of insulin resistance and mitochondrial abnormalities. Gastroenterology 120, 1183–1192. Sato, K., Arai, H., Mizuno, A., Fukaya, M., Sato, T., Koganei, M., Sasaki, H., Yamamoto, H., Taketani, Y., Doi, T., Takeda, E., 2007. Dietary palatinose and oleic acid ameliorate disorders of glucose and lipid metabolism in Zucker fatty rats. J. Nutr. 137, 1908–1915. Soriguer, F., Esteva, I., Rojo-Martinez, G., Ruiz de Adana, M.S., Dobarganes, M.C., García-Almeida, J.M., Tinahones, F., Beltrán, M., González-Romero, S., Olveira, G., Gómez-Zumaquero, J.M., 2004. Oleic acid from cooking oils is associated with lower insulin resistance in the general population (Pizarra study). Eur. J. Endocrinol. 150, 33–39. Soriguer, F., Moreno, F., Rojo-Martínez, G., García-Fuentes, E., Tinahones, F., Gómez-Zumaquero, J.M., Cuesta-Muñoz, A.L., Cardona, F., Morcillo, S., 2003. Monounsaturated n-9 fatty acids and adipocyte lipolysis in rats. Br. J. Nutr. 90, 1015–1022. Szende, B., Timár, F., Hargitai, B., 1994. Olive oil decreases liver damage in rats caused by carbon tetrachloride (CCl4). Exp. Toxicol. Pathol. 46, 355–359. Thomsen, C., Rasmussen, O., Lousen, T., Holst, J.J., Fenselau, S., Schrezenmeir, J., Hermansen, K., 1999. Differential effects of saturated and monounsaturated fatty acids on postprandial lipemia and incretin responses in healthy subjects. Am. J. Clin. Nutr. 69, 1135–1143. Trevisan, M., Krogh, V., Freudenheim, J., Blake, A., Muti, P., Panico, S., Farinaro, E., Mancini, M., Menotti, A., Ricci, G., 1990. Consumption of olive oil, butter, and vegetable oils and coronary heart disease risk factors. The Research Group ATS-RF2 of the Italian National Research Council. JAMA 263 (5), 688–692. Williams, C.M., 2001. Beneficial nutritional properties of olive oil, implications for postprandial lipoproteins and factor VII. Nutr. Metab. Cardiovasc. Dis. 11 (Suppl), 51–56. Williams, P.T., Fortmann, S.P., Terry, R.B., Garay, S.C., Vranizan, K.M., Ellsworth, N., Wood, P.D., 1987. Associations of dietary fat, regional adiposity, and blood pressure in men. JAMA 257, 3251–3256. Willner, I.R., Waters, B., Patil, S.R., Reuben, A., Morelli, J., Riely, C.A., 2001. Ninety patients with nonalcoholic steatohepatitis, insulin resistance, familial tendency, and severity of disease. Am. J. Gastroenterol. 96, 2957–2961.
Chapter 127
Uptake, Metabolism and Biological Effect of the Olive Oil Phenol Hydroxytyrosol in Human HepG2 Cells Luis Goya, Raquel Mateos, M. Angeles Martín, Sonia Ramos and Laura Bravo Departamento de Metabolismo y Nutrición, Instituto del Frío – ICTAN (CSIC), Madrid, Spain
127.1 INTRODUCTION 127.1.1 Olive Oil Phenolics Epidemiological studies have brought about interest in the Mediterranean diet, showing a relationship between this diet, rich in fruit, vegetables and legumes, and a reduced incidence of pathologies such as coronary heart disease and cancer (Lipworth et al., 1997). A central hallmark of this diet is the high consumption of virgin olive oil as the main source of fat. Converging evidence suggests that olive oil’s beneficial effects are related not only to its elevated oleic acid content, but also to the high level of antioxidants in the non-saponifiable fraction, including phenolic compounds absent in seed oils (Bravo, 1998; Visioli and Galli 1998). The main phenolic compounds in virgin olive oil are secoiridoid derivatives of 2-(3,4-dihydroxyphenyl)ethanol (hydroxytyrosol, HTy) (Figure 127.1) and 2-(4-hydroxyphenyl)ethanol (tyrosol) that occur either as simple phenols or esterified with elenolic acid to form oleuropein and ligstroside aglycones (Mateos et al., 2001). In addition,
hydroxytyrosyl acetate (2-(3,4-dihydroxyphenyl)ethyl acetate) has also been identified in virgin olive oil (Brenes et al., 1999). These phenols, together with some flavonoids and lignans, constitute the antioxidant fraction of olive oil (Mateos et al., 2001, 2003). The phenolic fraction of virgin olive oil has proved to have antioxidant activity in vitro, scavenging peroxyl, hydroxyl and other free radicals, reactive nitrogen species, and superoxide anions, breaking peroxidative chain reactions and preventing metal ion catalyzed production of reactive oxygen species (ROS) (Mateos et al., 2003; Tripoli et al., 2005). Also, the inhibitory action of olive oil constituents on LDL oxidation shown in vitro (Caruso et al., 1999) and in vivo (Coni et al., 2000), would contribute to the protective effect of olive oil against cardiovascular disease. Human and animal studies have shown that olive oil phenolics are bioavailable. The main urinary metabolites identified were methylated derivatives, as well as sulfate and glucuronide conjugates (Vissers et al., 2002; Visioli et al., 2003). Using differentiated Caco-2 cell monolayers
R1
R3
R2 (R1 = OH) (R1 = OH) (R1 = OH) (R1 = OCH3)
(R2 = OH) (R2 = OCH3) (R2 = OGlucur) (R2 = OGlucur)
(R3= O) (R3= OH) (R3= OH) (R3= OH)
Hydroxytyrosol (HTy) Homovanillic alcohol Monoglucuronide conjugate of HTy Methylglucuronide conjugate of HTy
FIGURE 127.1 Chemical structures of hydroxytyrosol and its metabolites. The figure represents the basic structure common to all compounds with the different substitutions (R) regarding each specific compound.
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1157
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1158
SECTION | II
as a model system of the human intestinal epithelium, it has been shown that 14C-hydroxytyrosol is transported via passive diffusion, bidirectionally and in a dose-dependent manner (Manna et al., 2000). The only metabolite identified was homovanillic alcohol (Figure 127.1), the methylated derivative of HTy. Further conjugation of olive oil phenols with sulfate or glucuronic acid (Figure 127.1) is expected to take place in the liver, which is the main organ involved in the metabolism of xenobiotics (Brandon et al., 2006), along with the intestinal mucosa and the kidney.
127.1.2 Experimental Model The liver is the first tissue where the absorbed polyphenols exert their biological effects. Therefore, studies dealing with the effect of antioxidant dietary phenolics at a physiological level (liver in live animals) and at a cellular level (cultured cells from liver origin) should be prioritized. Both live animal (Alía et al., 2003) and cell culture (Valls-Bellés et al., 2004; El-Sayed et al., 2007) experimental approaches have been used by researchers to carry out such studies. On the one hand, the simultaneous fluctuations of fuels and hormones and the complexity of tissue and organ composition in experiments in live animal models make it very difficult to demonstrate molecular mechanisms of specific regulation (Alía et al., 2003). On the other hand, the cell model should be able to stay in culture for a long period both to allow long treatments with high stability and to ensure a constant and reliable response to a repeated condition. All these handicaps frequently preclude the use of cultured hepatocytes as a model for oxidative stress in cell culture (Valls-Bellés et al., 2004). The study of the regulation of antioxidant defense mechanisms at the cellular level may benefit from the use of an established cell culture line. Human HepG2 (Table 127.1), a well-differentiated transformed cell line, is a reliable model, easy to culture, well-characterized and widely used for biochemical and nutritional studies (Alía et al., 2006a; Brandon et al., 2006). The response of the antioxidant defense system to a range of doses of tert-butyl hydroperoxide (t-BOOH) or hydrogen peroxide (H2O2) has been tested in cultured HepG2 cells, and the results demonstrate that a condition of cellular stress is evoked when HepG2 cells in culture are treated with t-BOOH but not with H2O2 (Alía et al., 2005). Therefore, treatment of human HepG2 with t-BOOH yields an excellent model of oxidative stress in cell culture (Table 127.1). The aims of this overview were: (a) to show the uptake and metabolism of olive oil phenolic HTy in cell cultures of HepG2 as a cell model system for human hepatocytes, and (b) to give evidence for the potential protective effect of HTy against an oxidative stress induced by t-BOOH in the same experimental model.
Major Organ Systems Including Liver and Metabolism
127.2 UPTAKE AND METABOLISM OF HTy IN CELL CULTURE 127.2.1 Uptake and Metabolism of HTy by Caco-2 Olive oil, the main dietary fat in Mediterranean countries, is rich in phenolic compounds with proven antioxidant activity, thus potentially contributing to the beneficial effects attributed to the Mediterranean diet. Such contribution would only be plausible if olive oil phenols are bioavailable. An in vitro study of the uptake and metabolism of HTy using Caco-2 cells as a model system of the human intestinal epithelium (Manna et al., 2000), showed that this phenol passively diffused into the enterocyte, with significant transepithelial flux bidirectionally (from the apical to the basolateral sides of the differentiated cells and vice versa). The high apparent permeability calculated for the apical to basolateral transport led these authors to conclude that HTy could be quantitatively absorbed in the intestine. Several human (Miró-Casas et al., 1998; Vissers et al., 2002; Visioli et al., 2003) and animal (Tuck et al., 2001; Visioli et al., 2003) studies have shown that oleuropein, HTy and tyrosol are bioavailable. In a human ileostomy study it was shown that up to 66% of the ingested olive oil phenols were absorbed in the small intestine (Vissers et al., 2002). Urinary recoveries as high as 80% of the ingested amounts of HTy have been reported in humans (Miró-Casas et al., 1998). Over 90% of the urinary metabolites were conjugates (Miró-Casas et al. 1998; Vissers et al., 2002), mainly glucuronidated metabolites, yet free and methylconjugates, with or without glucuronidation, were also excreted in human urine. Methylglucuronides (Figure 127.1), as well as sulfoconjugates of HTy, tyrosol or their metabolites (methyl or glucuronide conjugates) have only been observed in animal experiments (Tuck et al., 2001). Passive diffusion has been suggested as the mechanism of transport responsible for HTy uptake in Caco-2 cells (Manna et al., 2000). In that study, most HTy (90%) was transported as the free molecule, the only metabolite detected in the culture medium being its methylated derivative (Figure 127.1), homovanillic alcohol (Manna et al., 2000). The Caco-2 cell line, which exhibits enterocyte-like characteristics, has shown its capacity to glucuronidate and sulfate different compounds, including flavonoids (Liu and Hu, 2002). The fact that only limited methylation of HTy takes place in this model of the human intestinal epithelium is surprising since the small intestine seems to be the main organ responsible for glucuronidation, having also an important role in the methylation of phenols (Liu and Hu, 2002). After absorption, further methylation, glucuronidation and sulfation can take place in the liver (Bravo, 1998; Donovan et al., 2001).
CHAPTER | 127 Uptake, Metabolism and Biological Effect of the Olive Oil Phenol Hydroxytyrosol
1159
TABLE 127.1 Key features of HepG2 cell culture. Positive aspects 1. The liver is the first tissue where the absorbed polyphenols exert their biological effects. Studies on the effect of antioxidant dietary phenolics at hepatic level should be prioritized 2. The simultaneous fluctuations of fuels and hormones and the complexity of tissue and organ composition in live animals prevents study of molecular regulatory mechanisms 3. HepG2, a well-differentiated human cell line, is a reliable model, easy to culture, well-characterized and widely used for biochemical and nutritional studies where many conditions can be assayed with minor interassay variations 4. Genotyping of phase I and phase II enzymes, and drug transporter polymorphisms in these cells confirmed HepG2 as a suitable model for metabolic studies 5. Metabolites and glucuronide conjugates formed by HepG2 resembled the metabolic profile of polyphenols observed in studies in vivo 6. HepG2 cells are able to stay unvarying in culture to allow long treatments with high stability and to ensure a constant and reliable response to a repeated condition 7. HepG2 cells are strong enough to react to a process of oxidative stress by activating defence mechanisms to face and endure it 8. It has been proved that a condition of cellular stress is evoked when HepG2 are treated with t-BOOH yielding an excellent model of oxidative stress in cell culture
Negative aspects 1. HepG2 is a transformed cell line with an unrestrained cell cycle 2. All cultured cells grow in artificial conditions; (a) monolayer, (b) seeded on plastic substrate, (c) static culture medium vs. dynamic blood flow in tissues 3. Several steps in order to extrapolate results to humans; (a) to plain hepatocytes, (b) to liver of live animals, (c) to humans This table shows the positive and negative aspects of the use of HepG2 cell culture in studies on the metabolic fate and biological effect of diet antioxidants.
127.2.2 Uptake and Metabolism of HTy by HepG2 Considering the limited metabolism of HTy by colonderived Caco-2 cells reported by Manna et al. (2000), it follows that biotransformation of absorbed HTy should take place mostly in the liver. To test this hypothesis, we studied the metabolism of HTy in an in vitro hepatic model system using human HepG2 cells (Table 127.1). HepG2 cells were grown in DMEM F-12 medium, supplemented with 2.5% fetal bovine serum and 50 mg L⫺1 each of gentamicin, penicillin and streptomycin. The same medium deprived of serum but containing the antibiotic mixture was used in all experiments to prevent any potential interference from serum components. Cells were grown in a humidified incubator containing 5% CO2 and 95% air at 37°C. The cells were incubated with 100 μM HTy for 2 and 18 hours, and the products formed as a consequence of cell metabolism
were analyzed in the extracellular culture medium and in cytoplasmic contents after cell lysis (Mateos et al., 2005). A typical profile of the chromatographic separation of different molecular species detectable in medium from HepG2 in culture after incubation with HTy is shown in Figure 127.2. The results of identification of the molecular species are summarized in Table 127.2.
127.2.2.1 Identification of HTy Metabolites In order to identify the different metabolites formed after incubation of human HepG2 cells with HTy, several steps were followed: in vitro conjugation of pure standards, enzymatic hydrolysis of metabolites formed by HepG2 cells, and confirmation of structures by liquid chromatography-mass spectrometry (LC-MS). In vitro conjugation of the olive oil phenolic was performed by treatment of pure standards with catechol-O-methyl transferase or a microsomal
1160
SECTION | II
fraction obtained from rat hepatocytes in the presence of the corresponding methyl (S-adenosyl-L-methionine), sulfate (3⬘-phosphoadenosine-5⬘-phosphosulfate) or UDP-glucuronic acid donors (Mateos et al., 2005). Metabolites detected in the culture medium of cells incubated with HTy disappeared after treatment with
Absorbance (mAU)
8
Major Organ Systems Including Liver and Metabolism
β-glucuronidase. In turn, HTy, formerly absent in the medium, could be detected. This indicates that the disappeared peaks were glucuronides and methylglucuronides of HTy (Table 127.2), since removal of the glucuronic acid moiety resulted in the formation of the parent HTy and methyl-HTy (Mateos et al., 2005). The same metabolites
HTy
M4
a
4 M2
0
M5
M3
8 M4 M1+ M2
4
M3
b M5
P1
0 0
5
10
15
20 Time (min)
25
30
35
FIGURE 127.2 Chromatographic separation of species in medium from HepG2 in culture after incubation with HTy. The figure shows a typical profile of the chromatographic separation of different molecular species detectable in medium from HepG2 in culture after incubation with 100 μM HTy for (a) 2 hours and (b) 18 hours. Peaks M1 to M5 correspond to different metabolites of hydroxytyrosol formed after incubation with the hepatic cells; peak P1 is accumulated in the medium after 18 h incubation in the absence of test compounds. See Table 127.2 for peak identification. Reproduced with permission from J. Agric. Food Chem. 2005, 53, 9897–9905, Copyright 2005 American Chemical Society.
TABLE 127.2 Chromatographic and spectral characteristics of hydroxytyrosol and the metabolites formed after incubation with HepG2 cells. Compound
MW
RT
λmax
[M-H]⫺
Fragment ions
Proposed structure
HTy
154
15.0
280
153.0
123.1
HTy
M1
330
16.2
277
329.0
153.0
Monoglucuronide of HTy
M2
330
16.4
278
329.0
153.0
Monoglucuronide of HTy
M3
344
18.7
276
343.1
166.9-153.0
Methylglucuronide of HTy
M4
344
20.7
277
343.1
166.8-153.0
Methylglucuronide of HTy
M5
168
21.8
280
–
–
Homovanillic acid (methyl HTy)
HTy: hydroxytyrosol; MW: molecular weight; RT: retention time; λmax: wavelength of maximal absorption. This table shows the characteristics of the peaks identified after chromatographic separation of medium from HepG2 in culture after incubation with 100 μM HTy (see Figure 127.2). Peaks M1 to M5 correspond to different metabolites of hydroxytyrosol formed after incubation with the hepatic cells. Reproduced with permission from J. Agric. Food Chem. 2005, 53, 9897–9905, Copyright 2005 American Chemical Society.
1161
CHAPTER | 127 Uptake, Metabolism and Biological Effect of the Olive Oil Phenol Hydroxytyrosol
as in non-hydrolyzed samples were observed after incubation with sulfatase, suggesting that no sulfate-conjugated metabolites of olive oil phenols are formed by HepG2 cells in culture.
127.2.2.2 Quantification of HTy Metabolites Table 127.3 shows the percentage of HTy and its metabolites in the cytoplasmic content and culture medium after 2 and 18 hours of incubation of 100 μM of the olive oil phenolic with HepG2 cells. It can be seen that differences occurred in the metabolism of the phenol studied, depending on the length of incubation with the hepatic cells.
After 2 hours, most HTy was present in the culture medium as the free molecule, with metabolites representing less than 15% of the total phenols in the extracellular medium. Half of these metabolites were methyl conjugates, the rest being monoglucuronides and methylglucuronides (Table 127.3). At 18 hours, however, HTy metabolites exceeded 75% of the analyzed phenols, with about 25% of free, non-metabolized HTy being detected in the culture medium. The extent of glucuronidation was comparable to that of methylation (32% vs. 26% of the total amount of phenols in the culture medium, respectively), with up to 18% of methylglucuronides (Table 127.3). The results obtained for cytoplasmic contents are in agreement with
TABLE 127.3 Concentrations and percentages of compounds after incubation of HepG2 cells with hydroxytyrosol. Extracellular culture medium 2 h, 100 μM
18 h, 100 μM %
μM
%
μM
HTy
85.56 ⫾ 1.75
86.43
22.30 ⫾ 2.41
25.28
M1
–
–
7.03 ⫾ 0.14
7.97
M2
5.31 ⫾ 0.18
5.36
21.92 ⫾ 1.15
24.86
M3
0.97 ⫾ 0.20
0.98
12.43 ⫾ 1.67
14.10
M4
1.04 ⫾ 0.17
1.05
2.93 ⫾ 0.22
3.32
M5
6.11 ⫾ 0.19
6.17
21.58 ⫾ 0.40
24.47
Total
98.99 ⫾ 1.39
100
88.19 ⫾ 0.85
100
Cytoplasmic content 2 h, 100 μM μM
18 h, 100 μM %
μM
%
HTy
0.072
46.86
0.087
28.43
M1
–
–
–
–
M2
0.049
31.93
0.079
25.93
M3
–
–
0.049
15.91
M4
–
–
0.015
4.89
M5
0.033
21.21
0.076
24.84
Total
0.155
100
0.306
100
HTy: hydroxytyrosol. This table shows the concentrations expressed in μM and relative percentages of hydroxytyrosol and its identified metabolites accumulated in the extracellular culture medium and in cytoplasmic contents after 2 and 18 hours of incubation of HepG2 cells in the presence of 100 μM hydroxytyrosol. Mean values ⫾ SD (n ⫽ 3) for culture medium samples. Three different samples of cytoplasmic content were pooled. Reproduced with permission from J. Agric. Food Chem. 2005, 53, 9897–9905, Copyright 2005 American Chemical Society.
1162
SECTION | II
those reported for extracellular culture medium: limited metabolism at 2 hours in comparison to the observed at a longer time of incubation, 18 hours. Sulfate metabolites were only detected in urine from animals after intravenous administration of HTy, indicating that rat hepatocytes are capable of sulfation of this phenol. However, when we attempted in vitro sulfation of olive oil phenolics using microsomes obtained from rat liver, no metabolites were detected after 2 hours of incubation (Mateos et al., 2005). Perhaps, longer incubation times are required to achieve conjugation in vitro, although rat liver microsomes were able to sulfate quercetin and isorhamnetin after 1 hour of incubation (Mateos et al., 2006). These results would suggest that phase II metabolism of olive oil phenols takes place primarily in the liver, yet further studies either with Caco-2 cells or with perfused intestinal segments are needed to ascertain the lack of major metabolism of these phenolic compounds by the intestinal epithelium. Therefore, extensive uptake and metabolism of the olive oil phenolic HTy was observed using HepG2 cells as a model system of the human liver. The main metabolites were glucuronides, methylglucuronides and methyl conjugates. This is in agreement with data from human studies where methylated and glucuronidated metabolites were the only conjugates observed in plasma and urine samples (Vissers et al., 2002).
127.3 ANTIOXIDANT EFFECT OF HTy IN CELL CULTURE 127.3.1 Effect of HTy on the Redox Status and Antioxidant Defenses of HepG2 127.3.1.1 Effect of HTy on Cell Viability and Redox Status Although plant phenolics may have potent antioxidant effects in vitro and in vivo, both in cell culture and live animals, elevated doses of these dietary compounds can also be toxic and mutagenic in cell culture systems and excess consumption by mammals could cause adverse metabolic reactions (Rodgers and Grant, 1998). The range of doses used in the study of the antioxidant effect of HTy was selected after an extensive search in the literature showing that only concentrations of or above 10 μM HTy were effective in most conditions (Goya et al., 2007a). In addition, preliminary experiments in our laboratory proved that concentrations of HTy below 50 μM could be safely used in order to test its potential protective effect against a condition of oxidative stress. Therefore, cell toxicity and cellular redox status were determined in cells treated long term with different concentrations of HTy in the range between
Major Organ Systems Including Liver and Metabolism
10 and 40 μM. None of the HTy doses tested evoked any cytotoxicity in cultured HepG2 (Goya et al., 2007a). Direct fluorescent evaluation of the ROS by the intracellular dichlorofluorescin assay can be used as an index to quantify the overall oxidative stress in cells (LeBel et al., 1992). A pro-oxidant such as t-BOOH can directly oxidize dichlorofluorescin to fluorescent dichlorofluorescein, and it can also decompose to peroxyl radicals and enhance generation of lipid peroxides and ROS, thus increasing fluorescence over control values. Treatment of HepG2 cells with micromolar concentrations of HTy did not evoke an increase in ROS generation, maintaining fluorescent values within the range of control untreated cells (Goya et al., 2007a). An important step in the degradation of cell membranes is the reaction of ROS with the double bonds of polyunsaturated fatty acids (PUFAs) to yield lipid hydroperoxides. On breakdown of such hydroperoxides malondialdehyde (MDA), a three-carbon compound formed by scission of peroxidized PUFAs, is one of the main products of lipid peroxidation (Suttnar et al., 2001). Since MDA is elevated in various diseases thought to be related to free radical damage, it has been widely used as an index of lipoperoxidation in biological and medical sciences (Suttnar et al., 2001). HepG2 cell treatment with HTy up to 24 hours did not induce changes in the intracellular MDA concentration (Goya et al., 2007a).
127.3.1.2 Effect of HTy on the Cell Antioxidant Defences Reduced glutathione (GSH) is the main non-enzymatic antioxidant defense within the cell, reducing different peroxides, hydroperoxides and radicals such as alkyl, alkoxyl, peroxyl, etc. (Scharf et al., 2003). It is usually assumed that GSH depletion reflects intracellular oxidation whereas an increase in GSH concentration could be expected to prepare the cell against a potential oxidative insult (Scharf et al., 2003). Plain treatment of HepG2 cells with HTy did not significantly affect GSH concentration (Goya et al., 2007a). In the defense against oxidative stress, the cellular antioxidant enzyme system plays a crucial role. Among others, the system includes superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GPx) and glutathione reductase (GR). There are three forms of SOD in mammals that catalyze the dismutation of the superoxide radical anion, i.e., Mn SOD located in mitochondria, Cu/Zn SOD found mainly in cytosol and an extracellular SOD (Liochev and Fridovich, 2007). CAT converts H2O2 to H2O (Alía et al., 2005); GPx catalyzes GSH oxidation to oxidized glutathione (GSSG) at the expense of H2O2 or other organic peroxides (Ursini et al., 1995) and GR recycles GSSG back to reduced glutathione using NADPH (Röhrdanz et al., 2002).
1163
CHAPTER | 127 Uptake, Metabolism and Biological Effect of the Olive Oil Phenol Hydroxytyrosol
50
14 000
Fluorescence units
%LDH in medium
30 20 10
b b
a
Ctrl tBOOH 10 μM HTy
12 000
a
40
10 000
ab
a
8000 6000
a
4000
ab
ab b
2000
b b
0
0 Ctrl
10 μM HTy
t-BOOH
0
20
50
90
Time (min) 1.6
100 a
1.2 b b
0.8
0.4
nmol GSH/mg protein
nmol MDA/mg protein
a
0
b
75
50 c 25
0 Ctrl
10 μM HTy
t-BOOH
Ctrl
t-BOOH
10 μM HTy
100 mU GPx/mg protein
a 75 b 50
b
25
0 Ctrl
t-BOOH
10 μM HTy
FIGURE 127.3 Protective effect of HTy against t-BOOH-induced oxidative stress. HepG2 cells were treated with 10 μM HTy for 20 hours (10) or left untreated (Ctrl and t-BOOH), then they were washed with PBS and 200 μM t-BOOH was added to all the cells except controls for 3 hours. Results of lactate dehydrogenase (LDH) leakage are expressed as percent of total LDH activity in the culture medium. Results of ROS, malondialdehyde (MDA), reduced glutathione (GSH) and GPx activity are expressed as indicated in the y axis. Different letters upon bars or line points indicate statistically significant differences (p ⬍ 0.05) among different data, except in the ROS results where letters a and b indicate statistically significant differences (p ⬍ 0.05) when the data are compared to their time-mate control (a) or t-BOOH (b). In all assays, values are means ⫾ SD of 7–8 different samples per condition. Data used with kind permission of Springer Science ⫹ Business Media.
Oxidative challenges lead to an increase in activity of antioxidant enzymes, particularly GPx and GR, in cultured cells such as mouse skeletal muscle cells (Zhou et al., 2001), primary rat hepatocytes (Valls-Bellés et al., 2004), and human HepG2 (Yoo et al., 1999; Alía et al., 2006a, b). Enzyme defenses activate in order to face the increasing generation of ROS induced by the potent pro-oxidant
t-BOOH (Rodgers and Grant 1998; Alía et al., 2006a, b). HTy treatment of HepG2 cells evoked no changes in the activity of GPx (Goya et al., 2007a), whereas a significant increase in this activity was observed after a 3-hour treatment with 200 μM t-BOOH, indicating a positive response of the cell defense system to face an oxidative insult (Alía et al., 2006a, b).
1164
SECTION | II
127.3.2 Protective Effect of HTy on HepG2 Against Oxidative Stress
SUMMARY POINTS ●
127.3.2.1 Protective Effect of HTy on Cell Viability and Redox Status HTy has been previously reported to counteract cytotoxicity induced by ROS in some human cellular models such as Caco-2 (Manna et al., 1997) and erythrocytes (Manna et al., 1999). When human HepG2 cells were pretreated with 10 μM HTy for 20 hours prior to being submitted to an oxidative stress by 200 μM t-BOOH for 3 hours, cell toxicity was completely prevented (Figure 127.3), indicating that the antioxidant-treated cells were totally protected against the oxidative insult. Moreover, the results of our study clearly show that pre-treatment of cultured HepG2 cells with micromolar concentrations of HTy reduces about a 50% the generation of ROS induced by t-BOOH (Figure 127.3), thus preventing or delaying conditions which favor oxidative stress in the cell (Goya et al., 2007a). Furthermore, the increase of MDA, as a marker for lipid peroxidation, induced by t-BOOH was completely averted by a pre-treatment of cells with micromolar doses of HTy (Figure 127.3). The protection against lipid peroxidation by HTy in a cell culture, reported for the first time by our laboratory, is in concert with other studies that showed a similar protection by tea catechins (Chen et al., 2002; Murakami et al., 2002), beta-carotene or lutein (Chen et al., 2002), quercetin (Alía et al., 2006b), a high-molecular-weight coffee melanoidin (Goya et al., 2007b) and Se-methylselenocysteine (Cuello et al., 2007) in the same cell line, human HepG2.
127.3.2.2 Protective Effect of HTy on the Cell Antioxidant Defenses In our experimental conditions, pretreatment with 10 μM HTy for 20 hours completely prevented the marked decrease in the concentration of reduced glutathione induced by t-BOOH (Figure 127.3). Maintaining GSH concentration above a critical threshold while facing a stressful situation represents an advantage for cell survival (Scharf et al., 2003; Alía et al., 2006a, b). Figure 127.3 also shows that treatment of human HepG2 cells for 20 hours with 10 μM HTy prevents the increase in the activity of GPx induced by oxidative stress either by direct action or via the decrease in ROS production or the increase in GSH (Goya et al., 2007a). The results indicate that at the end of a chemically induced stress period, the antioxidant defense system of cells that had been pretreated with the olive oil phenolic HTy has more efficiently returned to a steadystate activity diminishing, therefore, cell damage and enabling the cell to cope in better conditions with further oxidative insults.
Major Organ Systems Including Liver and Metabolism
●
●
●
●
Extensive uptake and metabolism of the olive oil phenolic hydroxytyrosol was observed using HepG2 cells as a model system of the human liver. The main metabolites were glucuronides, methylglucuronides and methyl conjugates, in agreement with data from human studies where methylated and glucuronidated metabolites were the only conjugates observed in plasma and urine samples. The antioxidant effect of hydroxytyrosol in cultured liver cells extends the protective effect reported for other dietary bioactive compounds to one of the most common of them in the Mediterranean diet. Hydroxytyrosol may prepare the antioxidant defense system of the cell to successfully face a condition of oxidative stress by molecular mechanisms that are currently being studied in our laboratory. Hydroxytyrosol may contribute to the protection afforded by fruit- and vegetable-rich diets against diseases, such as cardiovascular disease, for which excess production of ROS has been implicated as a causal or contributory factor.
ACKNOWLEDGMENTS This work was supported by grants AGL2004-302, AGL2007-64042/ALI and project CSD2007-00063 from Programa Consolider-Ingenio from the Spanish Ministry of Science and Innovation.
REFERENCES Alía, M., Horcajo, C., Bravo, L., Goya, L., 2003. Effect of grape antioxidant dietary fiber on the total antioxidant capacity and the activity of liver antioxidant enzymes in rats. Nutr. Res. 23, 1251–1267. Alía, M., Ramos, S., Mateos, R., Bravo, L., Goya, L., 2005. Response of the antioxidant defense system to t-butyl hydroperoxide and hydrogen peroxide in a human hepatoma cell line (HepG2). J. Biochem. Mol. Toxicol. 19, 119–128. Alía, M., Mateos, R., Ramos, S., Lecumberri, E., Bravo, L., Goya, L., 2006a. Influence of quercetin and rutin on growth and the antioxidant defense system in a human hepatoma cell line (HepG2). Eur. J. Nutr. 45, 19–28. Alía, M., Ramos, S., Mateos, R., Bravo, L., Goya, L., 2006b. Quercetin protects human hepatoma cell line (HepG2) against oxidative stress induced by tertbutyl hydroperoxide. Toxicol. Appl. Pharm. 212, 110–118. Brandon, E.F.A., Bosch, T.M., Deenen, M.J., Levink, R., Van der Wal, E., Van Meerveld, J.B.M., Bijl, M., Beijnen, J.H., Schellens, J.H.M., Meijerman, I., 2006. Validation of in vitro cell models used in drug metabolism and transport studies: genotyping of cytochrome P450, phase II enzymes and drug transporter polymorphisms in the human hepatoma (HepG2), ovarian carcinoma (IGROV-1) and colon carcinoma (CaCo-2, LS180) cell lines. Toxicol. Appl. Pharmacol. 211, 1–10.
CHAPTER | 127 Uptake, Metabolism and Biological Effect of the Olive Oil Phenol Hydroxytyrosol
Bravo, L., 1998. Polyphenols: chemistry, dietary sources, metabolism, and nutritional significance. Nutr. Rev. 56, 317–333. Brenes, M., García, A., García, P., Ríos, J.J., Garrido, A., 1999. Phenolic compounds in Spanish olive oils. J. Agric. Food Chem. 47, 3535–3540. Caruso, D., Berra, B., Giavarini, F., Cortesi, N., Fedeli, E., Galli, G., 1999. Effect of virgin olive oil phenolic compounds on in vitro oxidation of human low density lipoproteins. Nutr. Metab. Cardiovasc. Dis. 9, 102–107. Chen, L., Yang, X., Jiao, H., Zhao, B., 2002. Tea catechins protect against lead-induced cytotoxicity, lipid peroxidation, and membrane fluidity in HepG2 cells. Toxicol. Sci. 69, 149–156. Coni, E., Di Benedetto, R., Di Pasquale, M., Masella, R., Modesti, D., Mattei, R., Carlini, E.A., 2000. Protective effect of oleuropein, an olive oil biophenol, on low density lipoprotein oxidizability in rabbits. Lipids 35, 45–54. Cuello, S., Ramos, S., Mateos, R., Martín, M.A., Madrid, Y., Cámara, C., Bravo, L., Goya, L., 2007. Selenium methylselenocysteine protects human hepatoma HepG2 cells against oxidative stress induced by tertbutyl hydroperoxide. Anal. Bioanal. Chem. 389, 2167–2178. Donovan, J.L., Crespy, V., Manach, C., Morand, C., Besson, C., Scalbert, A., Remesy, C., 2001. Catechin is metabolized by both the small intestine and liver of rats. J. Nutr. 131, 1753–1757. El-Sayed, W.M., Aboul-Fadl, T., Roberts, J.C., Lamb, J.G., Franklin, M.R., 2007. Murine hepatoma (Hepa1c1c7) cells: A responsive in vitro system for chemoprotective enzyme induction by organoselenium compounds. Toxicol. In Vitro 21, 157–164. Goya, L., Mateos, R., Bravo, L., 2007a. Effect of the olive oil phenol hydroxytyrosol on human hepatoma HepG2 cells. Protection against oxidative stress induced by tert-butylhydroperoxide. Eur. J. Nutr. 46, 70–78. Goya, L., Delgado-Andrade, C., Rufián-Henares, J.A., Bravo, L., Morales, F.J., 2007b. Effect of coffee Melanoidin on human hepatoma HepG2 cells. Protection against oxidative stress induced by tertbutyl hydroperoxide. Mol. Nutr. Food Res. 51, 536–545. LeBel, C.P., Ishiropoulos, H., Bondy, S.C., 1992. Evaluation of the probe 2⬘, 7⬘-dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem. Res. Toxicol. 5, 227–231. Liochev, S.I., Fridovich, I., 2007. The effects of superoxide dismutase on H2O2 formation. Free Rad. Biol. Med. 42, 1465–1469. Lipworth, L., Martínez, M.E., Angell, J., Hsieh, C.C., Trichopoulos, D., 1997. Olive oil and human cancer: an assessment of evidence. Prev. Med. 26, 181–190. Liu, Y., Hu, M., 2002. Absorption and metabolism of flavonoids in the Caco-2 cell culture model and a perused rat intestinal model. Drug. Metab. Disp. 30, 370–377. Manna, C., Galletti, P., Cucciolla, V., Moltedo, O., Leone, A., Zappia, V., 1997. The protective effect of the olive oil polyphenols (3,4-dihydroxyphenyl)-ethanol counteracts reactive oxygen metabolite-induced cytotoxicity in Caco-2 cells. J. Nutr. 127, 286–292. Manna, C., Galletti, P., Cucciolla, V., Montedoro, G., Zappia, V., 1999. Olive oil hydroxytyrosol protects human erythrocytes against oxidative damages. J. Nutr. Biochem. 10, 159–165. Manna, C., Galletti, P., Maisto, G., Cucciolla, V., D’Angelo, S., Zappia, V., 2000. Transport mechanism and metabolism of olive oil hydroxytyrosol in Caco-2 cells. FEBS Lett. 470, 341–344. Mateos, R., Espartero, J.L., Trujillo, M., Ríos, J.J., León-Camacho, M., Alcudia, F., Cert, A., 2001. Determination of phenols, flavones, and lignans in virgin olive oils by solid-phase extraction and highperformance liquid chromatography with diode array ultraviolet detection. J. Agric. Food Chem. 49, 2185–2192.
1165
Mateos, R., Domínguez, M.M., Espartero, J.L., Cert, A., 2003. Antioxidant effect of phenolic compounds, α-tocopherol, and other minor components in virgin olive oil. J. Agric. Food Chem. 51, 7170–7175. Mateos, R., Goya, L., Bravo, L., 2005. Metabolism of the olive oil phenols hydroxytyrosol, tyrosol and hydroxytyrosyl acetate by human hepatoma HepG2 cells. J. Agric. Food. Chem. 53, 9897–9905. Mateos, R., Goya, L., Bravo, L., 2006. Uptake and metabolism of hydroxycinnamic acids (chlorogenic, caffeic and ferulic acids) by HepG2 cells as a model of human liver. J. Agric. Food Chem. 54, 8724–8732. Miró-Casas, E., Covas, M.-I., Farré, M., Fito, M., Ortuño, J., Weinbrenner, T., Roset, P., de la Torre, R., 1998. Hydroxytyrosol disposition in humans. Clin. Chem. 49, 945–952. Murakami, C., Hirakawa, Y., Inui, H., Nakano, Y., Yoshida, H., 2002. Effect of tea catechins on cellular lipid peroxidation and cytotoxicity in HepG2 cells. Biosci. Biotechnol. Biochem. 66, 1559–1562. Rodgers, E.H., Grant, M.H., 1998. The effect of the flavonoids, quercetin, myricetin and epicatechin on the growth and enzyme activities of MCF7 human breast cancer cells. Chem. Biol. Interac. 116, 213–228. Röhrdanz, E., Ohler, S., Tran-thi, Q.-H., Kahl, R., 2002. The phytoestrogen daidzein affects the antioxidant enzyme system of rat hepatoma H4IIE cells. J. Nutr. 123, 370–375. Scharf, G., Prustomersky, S., Knasmuller, S., Schulte-Hermann, R., Huber, W.W., 2003. Enhancement of glutathione and g-glutamylcysteine synthetase, the rate limiting enzyme of glutathione synthesis, by chemoprotective plant-derived food and beverage components in the human hepatoma cell line HepG2. Nutr. Cancer 45, 74–83. Suttnar, J., Masova, L., Dyr, E., 2001. Influence of citrate and EDTA anticoagulants on plasma malondialdehyde concentrations estimated by high-performance liquid chromatography. J. Chrom. B. 751, 193–199. Tripoli, E., Giammanco, M., Tabacchi, G., Di Majo, D., Giammanco, S., La Guardia, M., 2005. The phenolic compounds of olive oil: structure, biological activity and beneficial effects on human health. Nutr. Res. Rev. 18, 98–112. Tuck, K.L., Freeman, M.P., Hayball, P.J., Stretch, G.L., Stupans, I., 2001. The in vivo fate of hydroxytyrosol and tyrosol, antioxidant phenolic constituents of olive oil, following intravenous and oral dosing of labelled compounds to rats. J. Nutr. 131, 1993–1996. Ursini, F., Maiorino, M., Brigelius-Flohe, R., Aumann, K.D., Roveri, A., Schomburg, D., Flohe, L., 1995. Diversity of glutathione peroxidases. Meth. Enzymol. 252, 38–114. Valls-Bellés, V., Torres, M.C., Muñiz, P., Boix, L., González-Sanjosé, M.L., Codoner-Franch, P., 2004. The protective effects of melanoidins in adriamycin-induced oxidative stress in isolated rat hepatocytes. J. Sci. Food. Agric. 84, 1701–1707. Visioli, F., Galli, C., 1998. The effect of minor constituents of olive oil on cardiovascular disease: new findings. Nutr. Rev. 56, 142–147. Visioli, F., Galli, C., Grande, S., Colonnelli, K., Patelli, C., Galli, G., Caruso, D., 2003. Hydroxytyrosol excretion differs between rats and humans and depends on the vehicle of administration. J. Nutr. 133, 2612–2615. Vissers, M.H., Zock, P.L., Roodenburg, A.J., Leenen, R., Katan, M.B., 2002. Olive oil phenols are absorbed in humans. J. Nutr. 132, 409–417. Yoo, H.Y., Chang, M.S., Rho, H.M., 1999. Xenobiotic-responsive element for the transcriptional activation of the rat Cu/Zn superoxide dismutase gene. Biochem. Biophys. Res. Comm. 256, 133–137. Zhou, L.Z., Johnson, A.P., Rando, T.A., 2001. NF kappa B and AP-1 mediate transcriptional responses to oxidative stress in skeletal muscle cells. Free Rad. Biol. Med. 31, 1405–1416.
This page intentionally left blank
Chapter 128
Modulation of Hepatic Apoptotic Pathways by Dietary Olive and Sunflower Oil María I. Burón, Mónica Santos-González and José M. Villalba Departamento de Biología Celular, Fisiología e Inmunología, Facultad de Ciencias, Universidad de Córdoba, Spain
128.1 INTRODUCTION 128.1.1 Antioxidant-dependent and -independent Actions of Olive Oil It is well accepted that the Mediterranean diet (with a high content of olive oil, fiber, fruits, vegetables, and fish) has many beneficial effects on health – preventing cancer, coronary heart disease, and cognitive impairment – and many of the health effects of this diet have been attributed to olive oil (Owen et al., 2000; Perona et al., 2006; Yang et al., 2007). Beneficial consequences of olive oil consumption can be explained partially on the basis of the antioxidant action of molecules contained in this oil (Owen et al., 2000). The monounsaturated oleic acid, which accounts for over 70% of fat content in olive oil, is much less susceptible to oxidation than the polyunsaturated fatty acid, linoleic acid, which predominates in sunflower oil (Owen et al., 2000; Quiles et al., 2004, 2006). Minor constituents of the olive oil include hydrocarbons (as squalene), sterols (as β-sitosterol), polyphenols (as tyrosol, hydroxytyrosol and oleuropein, among many others), tocopherols, terpenoids (as erythrodiol), and other components usually found as traces (Perona et al., 2006). Some of the minor constituents of olive oil, such as β-sitosterol and tyrosol, also behave as antioxidants and are believed to play major roles in determining its overall antioxidant effect (Papadopoulos and Boskau, 1991; Owen et al., 2000; Moreno, 2003). However, the plasma concentration of antioxidant phenols resulting from dietary intake of olive oil might be too low to exert effective antioxidant actions in vivo (Vissers et al., 2004). Recently, it has been recognized that olive oil phenols display some pharmacological effects (Yang et al., 2007). Minor components of olive oils show antioxidant, antiinflammatory and/or hypolipidemic properties (Perona Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
et al. 2006). Oleocanthal inhibits cyclooxygenases-1 and -2 (Beauchamp et al., 2005); hydroxytyrosol and hydroxyisochromans are inhibitors of platelet aggregation (Togna et al., 2003), and oleuropein can form a non-covalent complex with amyloid-b peptide (Bazoti et al., 2006). It is thus clear that, besides its antioxidant role, dietary olive oil shows other biological activities that may contribute to a nutritional state that prevents disease by reduction of inflammatory and autoimmune disorders or by diminution of cancer incidence (De la Puerta et al., 2004; Yang et al., 2007).
128.1.2 Basic Aspects of Apoptotic Cell Death Apoptosis is a tightly regulated form of cell death that plays a critical role in the maintenance of tissue homeostasis. Apoptotic cell death can be initiated by two basic pathways (Lawen, 2003). The extrinsic pathway is initiated by ligand binding (such as Fas ligand) to receptors of the tumor-necrosis factor receptor (TNFR) family at the plasma membrane (such as Fas), leading to the exposure of death domains at the cytosolic surface, and the subsequent recruitment of initiator caspases -8 and -10. Once activated, they activate effector (executioner) caspases, such as caspase-3. Executioner caspases are responsible for many of the degradative events taking place in apoptotic cell death. On the other hand, the mitochondrial or intrinsic pathway is activated by a variety of extra- and intracellular stresses. A critical step in the initiation of the intrinsic pathway is the relocalization of Bax protein from the cytosol to the outer mitochondrial membrane. Cytochrome c is then released from the intermembrane space to the cytosol, where it binds to apoptotic protease activating factor-1 (Apaf-1) and activates initiator caspase-9 and then executioner caspase-3.
1167
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1168
SECTION | II
During the final phases of the process, apoptotic cells are engulfed and cleared away by phagocytic or viable neighboring cells without inflammatory reaction (Table 128.1).
128.1.3 Apoptosis and Aging Accumulating evidence strongly suggests that dysregulation of apoptosis is associated with the aging process. However, whether aging suppresses or enhances apoptosis in vivo is still controversial (Higami and Shimokawa, 2000; Stoka et al., 2006). Furthermore, apoptosis can be considered as a two-edged sword with respect to aging. On the one hand, increased rates of apoptosis in postmitotic cells (as neurons) could contribute to excessive cell loss and the functional decline of the organs (Higami and Shimokawa, 2000). On the other hand, since apoptosis serves to purge organs of transformed cells, exaggerated inhibition of this process in aged proliferative tissues (such as liver) would lead to the accumulation of damaged cells that could be a significant factor in the development of cancer, whose incidence is significantly increased with advanced age (Suh et al., 2002; Youssef et al., 2003). A recent study has reported that deletion of caspase-2, an important component of the intrinsic pathway of apoptosis, enhanced a number of aging-related traits in mice and evidenced a significant role for apoptosis in aging (Zhang et al., 2007). Due to its central position in the regulation of mammalian metabolism, liver is one of the best models to study the effects of environmental or endogenous variables on growth
Major Organ Systems Including Liver and Metabolism
control, apoptosis, and carcinogenesis in vivo. Since dietary changes can have a major impact on aging liver (LópezTorres et al., 2002; Quiles et al., 2006), we have reviewed in this chapter how dietary fat (sunflower versus virgin olive oil) modulates several components that regulate the apoptotic response in the rat. Susceptibility of liver cells to different apoptotic stimuli varies according to the aging process. Thus, we have considered this factor in order to better understand how dietary oils modulate liver apoptosis.
128.2 OLIVE OIL AND APOPTOSIS IN AGING 128.2.1 Caspase Activity and Dietary Oil in Aging Early studies have indicated that the extrinsic pathway of apoptosis could play a role in liver cell homeostasis, and this pathway could be up-regulated in aged animals. In this way, the Fas system regulates liver homeostasis (Adachi et al., 1995), and levels of Fas mRNA increase with age in rat liver (Higami et al., 1997). In addition, previous reports have also indicated that the endogenous rate of hepatic apoptosis is increased during aging (Muskhelishvili et al., 1995; Suh et al., 2002; Zhang et al., 2002), with significant increases in the activities of executioner caspase-3, -6, and -7, and in caspases-2 and -9, markers of the mitochondrial pathway of apoptosis, in the liver of aged rats (23–27 months) compared to their young counterparts. Furthermore, hepatocytes
TABLE 128.1 Key features of apoptosis. 1. Apoptosis is a physiological form of cell death, characterized by a well-established set of morphological and biochemical features (cell shrinkage, DNA fragmentation and chromatin condensation, cleavage of many protein targets and cell rupture into apoptotic bodies, etc.). Apoptotic cells are engulfed and degraded by phagocytic or viable neighboring cells without a trace 2. A balance between proliferation and apoptosis is required for health. Exacerbated apoptosis in postmitotic cells can contribute to degenerative diseases, but exaggerated inhibition of apoptosis in aged proliferative tissues (such as liver) can contribute to the development of cancer 3. Apoptotic cell death plays a crucial role during development and in tissue homeostasis. Apoptosis may be also induced by a variety of stimuli (such as oxidative stress, anti-tumor drugs, etc.) in cells that are not programmed to die 4. Two major general pathways of induction of apoptosis exist: the plasma membrane receptor-mediated (extrinsic) pathway and the mitochondrial (intrinsic) pathway Caspases are Cysteinyl-requiring Aspartate proteases. Initiator caspases (such as caspase-8/10 and caspase-9 for the extrinsic and intrinsic pathways respectively) activate executioner caspases (as caspase-3) which are responsible for most of the degradative events of apoptosis Many proteins are involved in the regulation of apoptosis. Members of the Bcl-2 family are either antiapoptotic (as Bcl-2 and Bcl-xL) or proapoptotic (as Bax, Bad and Bak) This table lists the key facts of apoptosis including morphological and biochemical features, functions, the results of apoptosis dysregulation, the main pathways of apoptosis induction, and the main proteins involved in its regulation and execution.
1169
CHAPTER | 128 Modulation of Hepatic Apoptotic Pathways by Dietary Olive and Sunflower Oil
isolated from old animals were more sensitive to oxidative stress that targets the mitochondria than hepatocytes isolated from young animals (Zhang et al., 2002). Also, apoptosis induced by cycloheximide inhibition of protein synthesis was also enhanced in hepatocytes of old rats (Higami et al., 1996). However, the fact that aging might enhance liver apoptosis in vivo is still controversial (Higami and Shimokawa, 2000; Bello et al., 2006; Stoka et al., 2006). In this way, susceptibility to apoptosis induced by either genotoxic stress (Suh et al., 2002) or the PPARα agonists clofibrate and Wy-14643 (Youssef et al., 2003) is significantly decreased in hepatocytes from old rats, and the activity of caspase-8, a key mediator of Fas-induced apoptosis, does not increase in aged rat liver (Zhang et al., 2002). These apparently contradictory results could indicate that liver apoptotic response is modulated by environmental variables. The diet is a factor that could influence tissue susceptibility to cell death, and liver appears to be particularly sensitive to changes in diet (López-Torres et al., 2002; Bello et al., 2006; Quiles et al., 2006). We have measured the activities of caspases representative of extrinsic or intrinsic pathways, and the activity of an executioner caspase in the liver of rats fed lifelong with two experimental diets based on different lipid sources frequently used in Europe: virgin olive oil and sunflower oil (Bello et al., 2006). Our results agree with those of Zhang et al. (2002), indicating that the extrinsic pathway of apoptosis does not experience major changes with respect to either aging or dietary fat, because similar levels of caspases-8/10 activity (representative of this pathway) were observed in liver from young (6 months) and old rats (24 months) for both dietary oil groups (Figure 128.1A). In addition, our data also indicate that the intrinsic pathway
25
is strongly affected by dietary oils. In this way, the activity of caspase-9 (representative of this pathway) was significantly decreased as a function of age in animals fed a sunflower oil-based diet, but this aging-related decrease was completely abolished in rats that were fed an olive oilbased diet (Figure 128.1B). Changes in the activity of the executioner caspase-3 paralleled those of caspase-9. Thus, caspase-3 was significantly depressed during aging in animals fed the sunflower oil diet and, similar to what was found for caspase-9, activity of caspase-3 in old animals from the olive oil group was significantly higher than in the sunflower oil group (Figure 128.1C). Caspase-3 activity was also significantly increased in thymocytes from young (2 months) mice fed for 4 weeks with a diet containing olive oil (Puertollano et al., 2004).
128.2.2 Bcl-2 and Bax Levels in Liver Membranes from Rats Fed on Different Lipid Sources Increase of Bax levels bound to membranes is a hallmark in the activation of the intrinsic pathway of apoptosis (Lawen, 2003). Measurement of the levels of Bcl-2 and Bax polypeptides in liver membranes of the two dietary groups (olive and sunflower oil) confirms that dietary oils modulate the intrinsic pathway of apoptosis during aging (Bello et al., 2006). Bcl-2 levels in membranes obtained from aged animals (24 months) of the olive oil group were lower than in membranes from age-matched animals of the sunflower oil group and the reverse was found for membrane-bound Bax, whose levels were significantly higher in the olive oil
20
9 8 7
15 10
15 ***
6 5 4
a
3
Caspase-3
Caspase-9
Caspase-8/10
20
a 5
2
5
*** 10
1 0 24
6
Sunflower A
0
0 6
Diet
24
6
Olive
24
6
Sunflower B
Diet
24
6
Olive
24
6
Sunflower C
24
Olive
Diet
FIGURE 128.1 Effect of aging and dietary fat on the activity of caspases representative of the extrinsic (caspases-8/10, A) and the intrinsic (caspase-9, B) pathways of apoptosis, and the executioner caspase-3 (C). Caspase activity was assessed by using a fluorogenic assay in cytosol samples obtained from rats fed for 6 or 24 months with a basal semisynthetic diet containing either sunflower or virgin olive oil as dietary fat (Bello et al., 2006). Values (expressed as arbitrary units min⫺1 mg⫺1) are means with standard errors (n ⫽ 8). Significant differences between diets for a given feeding period are indicated with asterisks (***, p ⬍ 0.001). Statistically significant differences between feeding periods for a given diet are also indicated with letters (a, statistically significant differences with p ⬍ 0.05). Depicted data are adapted from Bello et al. (2006).
1170
SECTION | II
group (Figure 128.2A). As a result, Bcl-2/Bax ratio, which determines cellular fate towards survival or death, was significantly lower in liver from rats fed the olive oil-based diet (Figure 128.2B). Olive oil also reduced expression of Bcl-2 in two human adenocarcinoma cell lines, HT-29 and Caco2, and increased apoptosis in these cells (Llor et al., 2003). Furthermore, consistent with our results, olive oil down-regulated the expression of Bcl-2 and stimulated caspase-3 in colonic mucosa (Schwartz et al., 2004) and decreased Bcl-2, and increased Bak and caspase-3 in rat mammary cancer and normal tissue (Stark et al., 2003). Thus, alterations of Bcl-2/ Bax ratio may be a common mechanism of apoptosis induction by olive oil in mitotic cells.
128.2.3 Sphingomyelinase Activity and Oxidative Stress Sphingomyelinases (SMases) are a group of phospholipase enzymes that are involved in a variety of cellular responses, including those related with a number of apoptotic stimuli. Upon activation, SMases hydrolyze sphingomyelin to phosphorylcoline and ceramide, the latter acting as a second messenger in the propagation of the apoptotic response. Among the different types of sphingomyelinases that have been described in mammal cells, a plasma membrane-bound Mg2⫹-dependent neutral SMase (nSMase) has emerged as one of the more important SMases in the
Major Organ Systems Including Liver and Metabolism
regulation of cell growth and apoptosis (Liu et al., 1997). Liver plasma membrane contains nSMase activity whose levels increase with aging, which suggests a positive correlation between the expression of the nSMase and the aging process (Liu et al., 1997; Lighte et al., 2000). We have studied how dietary sunflower or olive oil affect nSMase activity in young (6 months) and old (24 months) rats (Bello et al., 2006). Interestingly, when animals were fed on the sunflower oil-based diet, levels of nSMase activity showed a dramatic increase at 24 months. Feeding with a diet enriched with virgin olive oil abolished this aging-related increase of the nSMase (Figure 128.3). Using a rodent model of nutritional programming focused upon the liver, evidence of increased oxidative processes related to aging has been recently reported (Langley-Evans and Sculley, 2008). Since oxidative stress is one condition that leads to activation of the nSMase (Verheij et al., 1996), the aging-related increase of nSMase (observed in the sunflower but not in the olive oil group) is in accordance with a better antioxidant status of animals fed with an olive oilbased diet. In accordance with this interpretation, previous studies have demonstrated that, compared with a diet based on sunflower oil, a diet based on virgin olive oil results in increased antioxidant capacity (Ochoa et al., 2003; Quiles et al., 2004, 2006). In addition, we have found that membranes from the olive oil group displayed much less sensitivity to oxidation induced by thermal decomposition
Diet Sunflower Bcl-2
Olive
Bcl-2/Bax ratio
2
1 *
Bax 0 Sunflower A
B
Olive
Diet
FIGURE 128.2 Immunostaining of Bcl-2 and Bax polypeptides (A, showing representative results of two samples from each diet) and Bcl-2/ Bax ratio (B) in liver membranes from aged (24 months) rats fed lifelong on either sunflower or virgin olive oil-based diets. Membrane proteins were separated by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose sheets, and then stained with specific commercial antibodies (A). Quantification of polypeptide levels was accomplished by densitometry of the protein band obtained after immunostaining. Values so obtained were referred to the corresponding lane stained for total protein with Ponceau-S to correct any difference in protein loading between samples. Quantification data of Bcl-2 and Bax polypeptides were then used to calculate Bcl-2/Bax ratio (B). Values of Bcl-2/Bax ratio are means with standard errors (n ⫽ 8). Significant differences between diets were found with p ⬍ 0.05. Depicted data are adapted from Bello et al. (2006).
nSMase activity (cpm min−1 mg−1)
1400 a
1200 1000 800 600 400 200
***
0 6
24
Sunflower Diet
6
24
Olive
FIGURE 128.3 Effect of aging and dietary oil on plasma membrane Mg2⫹-dependent neutral sphingomyelinase (nSMase) activity. Liver plasma membranes were isolated from young (6 months) or old (24 months) rats by aqueous two-phase partition as described (Bello et al., 2006). Values (expressed as cpm min⫺1 mg⫺1) are means with standard errors (n ⫽ 8). Significant differences between diets for a given feeding period are indicated with asterisks (***, p ⬍ 0.001). Statistically significant differences between feeding periods for a given diet are also indicated (a, statistically significant differences with 6 months, p ⬍ 0.001). Depicted data are adapted from Bello et al. (2006).
1171
CHAPTER | 128 Modulation of Hepatic Apoptotic Pathways by Dietary Olive and Sunflower Oil
128.3 CONCLUSIONS AND PERSPECTIVES Current evidence points out the importance of dietary fat in aging-related regulation of the mitochondrial pathway of apoptosis. Lifelong intake of sunflower oil has been reported to cause a worse preservation of mitochondrial structure in livers of aged rats, and these changes were attenuated in animals fed with olive oil (Quiles et al., 2006). In addition to the animal models in vivo, stimulation of apoptosis by olive oil or by some of its constituents has also been demonstrated in cellular models in vitro and this effect has been attributed mainly to minor constituents contained in this oil, such as the polyphenols (Table 128.2). Squalene was inactive in the induction of hepatic apoptosis (Scolastici et al., 2004). Alterations of Bcl-2/Bax ratio may be a common mechanism of apoptosis induction by olive oil in mitotic liver and gastrointestinal cells. The effect of virgin olive oil, which increases the apoptotic response of old liver in comparison with sunflower oil, might resemble that of calorie restriction, the best-defined dietary intervention that increases lifespan. Many reports have demonstrated that calorie restriction increases hepatic apoptosis during aging. Since decreased apoptotic potential
900
60
800
A
0
CoQ10 (pmoles mg−1 protein)
70 CoQ10 (nmoles g−1 fresh weight)
of the azo initiator 2,2⬘ azobis amidinopropane (AAPH) (Bello et al., 2006). Better antioxidant status of animals fed olive oil compared with animals fed sunflower oil has been explained on the basis of the lower autoxidability of monounsaturated oleic acid compared with polyunsaturated linoleic acid (Quiles et al., 2004, 2006). Minor antioxidant constituents of olive oil could also play a role in modulating this oxidative stress-related apoptotic pathway. We have previously documented that the nSMase is efficiently inhibited by the antioxidant coenzyme Q (CoQ) at the plasma membrane, both in cellular systems and in enzyme assays carried out in vitro (Navas and Villalba, 2004). Furthermore, our previous data have shown that dietary supplementation with the antioxidant CoQ10 fully abolished aging-related increase of liver plasma membrane nSMase in rats fed a sunflower oil-rich diet (Bello et al., 2005). Interestingly, consumption of the olive oil diet (but not sunflower oil) resulted in a better preservation of both CoQ9 and CoQ10 at the plasma membrane of aged rats (Figure 128.4). Higher CoQ levels found in membranes from the olive oil group, particularly in aged rats, may have also contributed to a better antioxidant protection and thus, to decreased nSMase activity.
50 40 30 20 10
B
0
600 500 400 300 200 0 2000
200 150
a
100 50 6
24
6
Sunflower oil Diet
24
Olive oil
CoQ9
250
(pmoles mg−1 protein)
CoQ9 (nmoles g−1 fresh weight) C
700
100
350 300
*** a
D
*** a
1500
1000
500
0
6
24
6
Sunflower oil
24
Olive oil
Diet
FIGURE 128.4 Levels of CoQ10 and CoQ9 in liver homogenate (A and B) and purified plasma membrane (C and D). Rats were fed for 6 or 24 months with the sunflower or virgin olive oil. Lipids were then extracted from liver homogenates or from isolated plasma membranes. CoQ10 and CoQ9 were quantified by HPLC with electrochemical detection. Values are means with standard errors (n ⫽ 8). Significant differences between diets for a given feeding period are indicated with asterisks (***, p ⬍ 0.001). Statistically significant differences between feeding periods for a given diet are also indicated (a, statistically significant differences with 6 months, p ⬍ 0.001). Depicted data are adapted from Bello et al. (2006).
1172
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 128.2 Some examples of apoptosis stimulation by olive oil or its minor constituents in cellular and animal models. Cellular/animal model
Active constituent
Parameters measured/ mechanism of action
References
HT-29 cells
Erythrodiol
Increased caspase-3-like activity
Juan et al., 2008
HT-29 cells
Hydroxytyrosol
Activation of protein Ser/Thr phosphatase (PP2A)
Guichard et al., 2006
Caco-2 and HT-29 cells
Not determined
Decreased Bcl-2
Llor et al., 2003
HL-60 cells
Hydroxytyrosol
Increased DNA fragmentation, caspase activation, degradation of targeted proteins, and cytochrome c release
Della Ragione et al., 2000; Fabiani et al., 2006, 2008
RKO, HT-116, SW-480 cells
Pinoresinol
Increased DNA fragmentation and Bax transcripts
Fini et al., 2008
Mice thymocytes
Not determined
Increased caspase-3 activity
Puertollano et al., 2004
Rat mammary cancer and normal tissue
Not determined
Decreased Bc-2, increased Bak and caspase-3
Stark et al., 2003
Rat colonic mucosa
Not determined
Decreased Bcl-2, stimulated caspase-3
Schwartz et al., 2004
Rat mammary cancer
Not determined
Decreased Bcl-2, increased Bak and caspase-3
Stark et al., 2003
Rat normal liver in aging
Not determined
Decreased Bcl-2, increased Bax, caspase-9 and caspase-3
Bello et al., 2006
This table summarizes some examples of apoptosis induction by olive oil or by its minor constituents in animal models and in vitro systems, including the model of study, the active principle (when determined) and the effect of olive oil (or active principle) in the regulatory machinery of apoptosis.
likely contributes to the increased incidence of liver cancer in old rodents (Suh et al., 2002), stimulation of apoptosis by calorie restriction could be related with its beneficial effects on aging by eliminating preneoplastic cells and antagonizing carcinogenesis in rat liver (Muskhelishvili et al., 1995). Mitochondrial pathway of apoptosis is also increased in αMUPA mice, a transgenic model for increased longevity induced by calorie restriction (Tirosh et al., 2003). The use of olive oil as dietary lipid resource also plays a clear beneficial role in gastric defense mechanisms favoring mucosal barrier efficiency, through mucus secretion and apoptotic effects, and reducing the inflammatory responses (Motilva et al., 2008). Epidemiological studies show that in the countries where the populations fulfilled a typical Mediterranean diet, such as Spain, Greece and Italy, where virgin olive oil is the principal source of fat, cancer incidence rates are lower than in Northern European countries. The fact that, compared to sunflower oil, virgin olive oil activates the intrinsic pathway of liver apoptosis under conditions
of decreased oxidative stress is consistent with the growing evidence about the protective effect of virgin olive oil against cancer in experimental animals. Additional studies including other organs, particularly those representing postmitotic tissues such as skeletal muscle, heart and brain, are needed to clarify how virgin olive oil modulates apoptosis in the organism as a whole.
SUMMARY POINTS ●
●
●
●
Olive oil exerts many beneficial effects on health that can not be accounted for merely by its antioxidant action. Dietary oil modulates the rate of hepatic apoptosis during aging. The extrinsic pathway of hepatic apoptosis is not affected by either aging or diet (sunflower versus olive oil). The intrinsic pathway of hepatic apoptosis is significantly depressed in old rats fed a sunflower oil-based
CHAPTER | 128 Modulation of Hepatic Apoptotic Pathways by Dietary Olive and Sunflower Oil
●
●
●
diet. Compared with sunflower oil, a diet based on olive oil produces a significant increase in the intrinsic pathway of hepatic apoptosis. A decrease of Bcl-2 expression may be a common mechanism of apoptosis induction by olive oil. A diet based on olive oil results in a better antioxidant status and the preservation of antioxidant coenzyme Q in the liver. Olive oil abolishes the aging-related increase in liver plasma membrane Mg2⫹-dependent neutral sphingomyelinase that is observed on old animals fed with sunflower oil. Increase in the intrinsic pathway of hepatic apoptosis during aging under conditions of decreased oxidative stress may explain, at least partially, the protective effect of virgin olive oil against cancer in experimental animals.
REFERENCES Adachi, M., Suematsu, S., Kondo, T., Ogasawara, J., Tanaka, T., Yoshida, N., Nagata, S., 1995. Targeted mutation in the Fas gene causes hyperplasia in peripheral lymphoid organs and liver. Nat. Genet. 11, 294–300. Bazoti, F.N., Bergquist, J., Markides, K.E., Tsarbopoulos, A., 2006. Noncovalent interaction between amyloid-beta-peptide (1–40) and oleuropein studied by electrospray ionization mass spectrometry. J. Amer. Soc. Mass Spectr. 17, 568–575. Beauchamp, G.K., Keast, R.S.J., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A., 2005. Phytochemistry: ibuprofen-like activity in extra-virgin olive oil. Nature 437, 45–46. Bello, R.I., Gómez-Díaz, C., Burón, M.I., Alcaín, F.J., Navas, P., Villalba, J.M., 2005. Enhanced anti-oxidant protection of liver membranes in long-lived rats fed on a coenzyme Q10-supplemented diet. Exp. Gerontol. 40, 694–706. Bello, R.I., Gómez-Díaz, C., Burón, M.I., Navas, P., Villalba, J.M., 2006. Differential regulation of hepatic apoptotic pathways by dietary olive and sunflower oils in the aging rat. Exp. Gerontol. 41, 1174–1184. De la Puerta, R., Vázquez, R., Martínez-Domínguez, E., Sánchez-Perona, J., Ruíz-Gutiérrez, V., 2004. Effects of different dietary oils on inflammation models of a polyphenol-supplemented virgin olive oil diet. Metabolism 53, 59–65. Della Ragione, F., Cucciolla, V., Borriello, A., Della Pietra, V., Pontoni, G., Racioppi, L., Manna, C., Galletti, P., Zappia, V., 2000. Hydroxytyrosol, a natural molecule occurring in olive oil, induces cytochrome c-dependent apoptosis. Biochem. Biophys. Res. Commun. 278, 733–739. Fabiani, R., de Bartolomeo, A., Rosignoli, P., Servili, M., Selvaggini, R., Montedoro, G.F., di Saverio, C., Morozzi, G., 2006. Virgin olive oil phenols inhibit proliferation of human promyelicytic leukaemia cells (HL60) by inducing apoptosis and differentiation. J. Nutr. 136, 614–619. Fabiani, R., Rosignoli, P., de Bartolomeo, A., Fuccelli, R., Morozzi, G., 2008. Inhibition of cell cycle progression by hydroxytyrosol is associated with upregulation of cyclin-dependent protein kinase inhibitors p21 (WAF1/Cip1) and p27(Kip1) and with induction of differentiation in HL60 cells. J. Nutr. 138, 42–48. Fini, L., Hotchkiss, E., Fogliano, V., Graziani, G., Romano, M., De Vol, E.B., Qin, H., Selgrad, M., Boland, C.R., Ricciardello, L.,
1173
2008. Chemopreventive properties of pinoresinol-rich olive oil involve a selective activation of the ATM-p53 cascade in colon cancer cell lines. Carcinogenesis 29, 139–146. Guichard, C., Pedruzzi, E., Fay, M., Marie, J.C., Braut-Boucher, F., Daniel, F., Grodet, A., Gougerot-Pocidalo, M.A., Chastre, E., Kotelevets, L., Lizard, G., Vandewalle, A., Driss, F., Ogier-Denis, E., 2006. Dihydroxyphenylethanol induces apoptosis by activating serine/threonine protein phosphatase PP2A and promotes the endoplasmic reticulum stress response in human colon carcinoma cells. Carcinogenesis 27, 1812–1827. Higami, Y., Shimokawa, I., Okimoto, T., Tomita, M., Yuo, T., Ikeda, T., 1996. Susceptibility of hepatocytes to cell death induced by single administration of cycloheximide in young and old F344 rats. Effect of dietary restriction. Mutat. Res. 357, 225–230. Higami, Y., Shimokawa, I., Tomita, M., Okimoto, T., Koji, T., Kobayashi, N., Ikeda, T., 1997. Aging accelerates but life-long dietary restriction suppresses apoptosis-related Fas expression on hepatocytes. Am. J. Pathol. 151, 659–663. Higami, Y., Shimokawa, I., 2000. Apoptosis in the aging process. Cell Tissue Res. 301, 125–132. Juan, M.E., Wenzel, U., Daniel, H., Planas, J.M., 2008. Erythrodiol, a natural triterpenoid from olives, has antiproliferative and apoptotic activity in HT-29 human adenocarcinoma cells. Mol. Nutr. Food Res. 52, 595–599. Langley-Evans, S. C., and Sculley, D. V. (2008). Programming of hepatic antioxidant capacity and oxidative injury in the ageing. Mech. Ageing Dev. 126, 804–812. Lawen, A., 2003. Apoptosis—An introduction. BioEssays 25, 888–896. Lighte, S.A., Oakley, J.I., Nikolova-Karakashi, M.N., 2000. Activation of sphingolipid turnover and chronic generation of ceramide and sphingosine in liver during aging. Mech. Ageing Dev. 120, 111–125. Liu, B., Obeid, L.M., Hannum, Y.A., 1997. Sphingomyelinases in cell regulation. Semin. Cell Dev. Biol. 8, 311–377. Llor, X., Pons, E., Roca, A., Álvarez, M., Mañé, J., Fernández-Bañares, F., Gassull, M.A., 2003. The effects of fish oil, olive oil, oleic acid and linoleic acid on colorectal neoplastic processes. Clin. Nutr. 22, 71–79. López-Torres, M., Gredilla, R., Sanz, A., Barja, G., 2002. Influence of aging and long-term caloric restriction on oxygen radical generation and oxidative DNA damage in rat liver mitochondria. Free Radic. Biol. Med. 32, 882–889. Moreno, J.J., 2003. Effect of olive oil minor components on oxidative stress and arachidonic acid mobilization and metabolism by macrophages Raw 264.7. Free Radic. Biol. Med. 35, 1073–1081. Motilva, V., Sanchez-Fidalgo, S., Barranco, M.D., Herrerias, J.M., Alarcón de la Lastra, C., 2008. Mechanisms of increased gastric protection after NSAID-administration in rat consuming virgin olive oil diets. e- SPEN 3, 9–16. Navas, P., Villalba, J.M., 2004. Regulation of ceramide signaling by plasma membrane coenzyme Q reductases. Methods Enzymol. 378, 200–206. Muskhelishvili, L., Hart, R.W., Turturro, A., James, S.J., 1995. Agerelated changes in the intrinsic rate of apoptosis in livers of dietrestricted and ad-libitum-fed B6C3F1 mice. Am. J. Pathol. 147, 20–24. Ochoa, J.J., Quiles, J.L., Ibáñez, S., Martínez, E., López-Frías, M., Huertas, J.R., Mataix, J., 2003. Aging-related oxidative stress depends on dietary lipid source in rat postmitotic tissues. J. Bioenerg. Biomembr. 35, 267–275. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Würtele, G., Spiegelhalder, B., Bartsch, H., 2000. Olive-oil consumption and health: the possible role of antioxidants. Lancet Oncol. 1, 107–112. Papadopoulos, G., Boskau, D., 1991. Antioxidant effect of natural phenols on olive oil. J. Am. Oil Chem. 68, 669–671.
1174
SECTION | II
Perona, J.S., Cabello-Moruno, R., Ruíz-Gutiérrez, V., 2006. The role of virgin olive oil components in the modulation of endothelial function. J. Nutr. Biochem. 17, 429–445. Puertollano, M.A., Puertollano, E., Jiménez-Valera, M., Ruiz-Bravo, A., de Pablo, M.A., Álvarez de Cienfuegos, G., 2004. Lack of apoptosis in Listeria monocytogenes-infected thymocytes from mice fed with dietary lipids. Curr. Microbiol. 48, 373–378. Quiles, J.L., Ochoa, J.J., Ramírez-Tortosa, C., Battino, M., Huertas, J.R., Martín, Y., Mataix, J., 2004. Dietary fat type (virgin olive vs. sunflower oils) affects age-related changes in DNA double-strand-breaks, antioxidant capacity and blood lipids in rats. Exp. Gerontol. 39, 1189–1198. Quiles, J.L., Ochoa, J.J., Ramírez-Tortosa, M.C., Huertas, J.R., Mataix, J., 2006. Age-related mitochondrial DNA deletion in rat liver depends on dietary fat unsaturation. J. Gerontol. A Biol. Sci. Med. Sci. 61, 107–114. Schwartz, B., Birk, Y., Raz, A., Madar, Z., 2004. Nutritional-pharmacological combinations—a novel approach to reducing colon cancer incidence. Eur. J. Nutr. 43, 221–229. Scolastici, C., Ong, T.P., Moreno, F.S., 2004. Squalene does not exhibit a chemopreventive activity and increases plasma cholesterol in a Wistar rat hepatocarcinogenesis model. Nutr. Cancer 50, 101–109. Stark, A.H., Kossoy, G., Zusman, I., Yarden, G., Madar, Z., 2003. Olive oil consumption during pregnancy and lactation in rats influences mammary cancer development in female offspring. Nutr. Cancer 46, 59–65. Stoka, V., Turk, V., Bredesen, D.E., 2006. Differential regulation of the intrinsic pathway of apoptosis. FEBS Lett. 580, 3739–3745. Suh, Y., Lee, K.-A., Kim, W.-H., Han, B.-G., Vijg, J., Park, S.C., 2002. Aging alters the apoptotic response to genotoxic stress. Nature Med. 8, 3–4.
Major Organ Systems Including Liver and Metabolism
Tirosh, O., Aronis, A., Zusman, I., Kossoy, G., Yahav, S., Shinder, D., Abramovitz, R., Miskin, R., 2003. Mitochondrion-mediated apoptosis is enhanced in long-lived αMUPA transgenic mice and calorically restricted wild-type mice. Exp. Gerontol. 38, 955–963. Togna, G.I., Togna, A.R., Franconi, M., Marra, C., Guiso, M., 2003. Olive oil isochromans inhibit human platelet reactivity. J. Nutr. 133, 2532–2536. Verheij, M., Bose, R., Lin, X.H., Yao, B., Jarvis, W.D., Grant, S., Birrer, M.J., Szabo, E., Zon, L.I., Kyriakis, J.M., Haimovitz-Friedman, A., Fuks, Z., Kolesnick, R.N., 1996. Requirement for ceramide-initiated SAPK/JNK signaling in stress-induced apoptosis. Nature 380, 75–79. Vissers, M.N., Zock, P.L., Katan, M.B., 2004. Bioavailability and antioxidant effects of olive oil phenols in humans: A review. Eur. J. Clin. Nutr. 58, 955–965. Yang, D.P., Kong, D.X., Zhang, H.Y., 2007. Multiple pharmacological effects of olive oil phenols. Food Chem. 104, 1269–1271. Youssef, J.A., Bouziane, M., Badr, M.Z., 2003. Age-dependent effects of nongenotoxic hepatocarcinogens on liver apoptosis in vivo. Mech. Ageing Dev. 124, 333–340. Zhang, Y., Chong, E., Herman, B., 2002. Age-increases in the activity of multiple caspases in Fisher 344 rat organs. Exp. Gerontol. 37, 777–789. Zhang, Y., Padalecki, S.S., Chaudhuri, A.R., De Waal, E., Goins, B.A., Grubbs, B., Ikeno, Y., Richardson, A., Mundy, G.R., Herman, B., 2007. Caspase-2 deficiency enhances aging-related traits in mice. Mech. Ageing Dev. 128, 213–221.
Chapter 129
Influence of Olive Oil on Pancreatic, Biliary and Gastric Secretion: Role of Gastrointestinal Peptides María Dolores Yago, Nama’a Audi, Mariano Mañas and Emilio Martínez-Victoria Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain
129.1 INTRODUCTION At the invitation of the FAO/WHO, an international group of experts in nutrition, public health and food science gathered in Rome in 1993. At this meeting, that represented a milestone in the role of dietary fats and oils in human nutrition, it was concluded that doing a prudent, serious and impartial research is a challenge for the scientific community, because changing views about the effects of dietary fats and oils can profoundly influence the consumption of various foods and, ultimately, health and nutritional status and education, agricultural production, and food processing technologies, among others. Both at the level of the popular media and the scientific literature, it has been the relationship between the intake of fat and cardiovascular disease that has received most attention. In contrast, the effects of the amount or type of dietary fat on gastrointestinal function are poorly known. It was for this reason that we decided to concentrate our efforts on the relation between the prolonged intake of a specific type of dietary fat and changes in digestive function, particularly secretory function. This is what we call ‘adaptation to dietary fat type’. Since 1985 we have studied in different species the adaptation of biliary, gastric and, mainly, exocrine pancreatic secretion to dietary fat type. Moreover, with the purpose of elucidating the mechanisms involved in adaptation, we have used experimental models ranging from the whole subject (either conscious or anesthetized) to ex vivo preparations (viable pancreatic acinar cells) and continuous cell lines. In our in vivo studies, most of them described in this chapter, we have tested the effects of two dietary fats that are used preferentially in our geographic area: virgin olive oil (VOO) and sunflower oil (SO). While the health and therapeutic benefits of olive oil on the gastrointestinal Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
system were mentioned already by Hippocrates in ancient times, we sought to provide scientific support for the actions of this major component of the Mediterranean diet, now that its role in human health is being actively debated.
129.2 OLIVE OIL AND DIGESTIVE SECRETION IN DOGS 129.2.1 Exocrine Pancreatic Secretion Since the findings of Pavlov in the early 1900s, the adaptation of the exocrine pancreas to dietary changes has been studied in various species, especially the rat. The consensus of these studies is that, given an adequate dietary protein supply, the major digestive enzymes, proteases, amylase and lipase, change in both pancreatic tissue and its secretion proportionally to the amount of their respective substrates in the diet. The physiological significance of this adaptation would be to optimize the digestion and utilization of that substrate. In contrast to the general agreement about the influence of the amount of dietary fat, there is considerable controversy over the effects of the type of fat. On the other hand, the greatest number of studies on the pancreatic adaptation to diet has centred the attention on the changes in tissue enzyme content, but very few reports exist concerning the overall secretory pancreatic response to endogenous or exogenous stimulation after medium- or long-term intake of a specific diet. These reasons moved us to undertake the following investigations in dogs. Weaning mongrel dogs were randomly assigned to one of two isoenergetic and isoproteic diets (33% energy as fat) that differed only in the nature of the fat source (Table 129.1): virgin olive oil (VOO) or sunflower oil (SO). After 8 months on the diets, animals were fitted with permanent pancreatic
1175
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1176
SECTION | II
TABLE 129.1 Fatty acid composition of the experimental diets used in the studies on long-term (8 months) adaptation of exocrine pancreatic secretion to dietary fat type in dogs. Fatty acid
VOO group
16:0 (palmitic acid)
SO group
11.7
9.1
2.4
3.5
18:1 n-9 (oleic acid)
60.9
25.5
18:2 n-6 (linoleic acid)
15.3
56.3
18:3 n-3 (linolenic acid)
0.8
2.3
18:0 (stearic acid)
Weaning dogs were fed for 8 months on diets containing virgin olive oil (VOO) or sunflower oil (SO) as the source of dietary fat (33% energy). Results are expressed as percentage of total fatty acid content. Reprinted from Ballesta MC, et al. Br J Nutr 1992; 68: 175–182, with permission of Cambridge University Press.
Pancreatic and duodenal cannula Weaning
Surgery 8 months (adaptation period) 33% energy as VOO or SO
F
F
P
P
P
P
P
–1
0
1
2
3
4
5
20 h fast
Time (h) Food
FIGURE 129.1 Experimental design of the studies on long-term (8 months) adaptation of exocrine pancreatic secretion to dietary fat type in dogs. Dogs were allowed 1 week of recovery before the experiments were begun. Fasting (F) and postprandial (P) samples of pure pancreatic juice were collected at hourly intervals. The dotted area represents the postprandial period. VOO: virgin olive oil; SO: sunflower oil.
cannulae in order to collect fasting and postprandial samples of pure pancreatic juice (see Figure 129.1). The most striking results in this study (Ballesta et al., 1990a) were observed after food ingestion. Dogs fed on the SO diet responded to food with marked increases in pancreatic flow rate, whereas an attenuated response was observed in VOO animals. The secretion of bicarbonate, total protein, amylase, lipase and chymotrypsin followed the same pattern as the pancreatic flow rate. Parallel experiments conducted in
Major Organ Systems Including Liver and Metabolism
these very same animals during the adaptation period revealed that, while there were no differences between groups VOO and SO in fat digestibility (Ballesta et al., 1990b), an improved protein digestive and metabolic utilization was actually evidenced in VOO animals (Ballesta et al., 1991). Hence, pancreatic adaptation to diets containing olive oil as dietary fat source seems to represent a key advantage in terms of exocrine pancreatic economy, at least in this species. Our findings raised the suggestion that prolonged intake of VOO or SO might result in different circulating levels of some of the gastrointestinal hormones that control exocrine pancreatic secretion (Table 129.2). Oleic acid is a strong stimulus for the release of cholecystokinin (CCK), a major pancreatic secretagogue, but it is also a very effective releaser of peptide YY (PYY) and pancreatic polypeptide (PP), two gastrointestinal peptides that inhibit exocrine pancreatic secretion. We sought confirmation of this hypothesis by submitting dogs to similar diets (35% energy as either VOO or SO) and feeding periods (6 months from weaning). Our data (Yago et al., 1997a) revealed that fasting and postprandial values for plasma PYY (Figure 129.2) and fasting values for plasma PP were significantly higher in animals fed on the VOO diet, which explains the attenuated secretory activity found in this group in our previous study. We also examined if these changes were related not only to the type but also to the amount of fat. For this purpose we fed weaning dogs for 6 months with diets containing 20% energy as fat (VOO or SO). Once more, we found a reduced pancreatic response to food in the VOO group (Yago et al., 1997b). Interestingly, comparison of secretory parameters obtained after high-olive-oil feeding (33% energy, Ballesta et al., 1990a) and medium-olive-oil feeding (20% energy, Yago et al., 1997b) indicated that the higher the content of VOO in the diet, the more attenuated the postprandial pancreatic response was. Considering the ability of oleic acid to elevate in the long term the blood levels of such inhibitory factors as PYY and PP, our results suggest a close inverse relationship between the content of oleic acid in the habitual diet and exocrine pancreatic response to a meal.
129.2.2 Bile Secretion We used the same diets and dietary protocol as in the above study of exocrine pancreatic secretion (Ballesta et al., 1990a). Following the adaptation period to diets, a permanent cannula was implanted in the common bile duct to allow collection of bile samples. No differences in bile flow rate were appreciated between VOO and SO animals in fasting conditions (Ballesta et al., 1992). This parameter increased immediately in response to food intake in both dietary groups, but the temporal pattern was somehow different. In VOO dogs this rise lasted until the second postprandial hour whereas at this time bile flow
CHAPTER | 129 Influence of Olive Oil on Pancreatic, Biliary and Gastric Secretion: Role of Gastrointestinal Peptides
1177
TABLE 129.2 Key facts of gastrointestinal peptide hormones. Gastrin
●
Secretin
Cholecystokinin (CCK)
Stimulates gastric acid secretion and gastric mucosa growth
●
Stimulates pancreatic and biliary secretion of fluid and bicarbonate
●
Inhibits gastrin release, gastric acid secretion and motility
●
Stimulates exocrine pancreatic secretion, gallbladder contraction and pancreatic growth
●
Stimulates gastric somatostatin release
●
Inhibits gastric emptying and reduces food intake
Motilin
●
Stimulates gastrointestinal motility in fasting conditions (migrating motor complex)
Peptide YY (PYY)
●
Inhibits gastric acid secretion, exocrine pancreatic secretion and intestinal motor and secretory functions
●
Inhibits gastric emptying and food intake
Pancreatic polypeptide (PP)
Somatostatin
Ghrelin
●
Inhibits exocrine pancreatic secretion and gallbladder contraction
●
Reduces gut motility and food intake
●
Important paracrine mediator
●
Inhibits gastrin and histamine release and gastric acid secretion, pepsinogen secretion, exocrine pancreatic secretion and biliary secretion of fluid and bicarbonate
●
General inhibitory action on secretion, absorption and release of gastrointestinal hormones
●
Stimulates gastric contractions and emptying, and food intake
This table lists the main peptide hormones of the gastrointestinal tract involved in the regulation of digestive secretions and/or motility and their major functions. Note that gastrointestinal peptides can participate not only in endocrine, but also in paracrine or neurocrine communication. This is the case of some of the above peptides, such as somatostatin, CCK, PYY and, possibly, PP.
1000
Plasma PYY (pmol L–1)
# 800
# #
#
600
400
200
VOO
SO
0 F
1
2
3
4
5
Time (h) FIGURE 129.2 Plasma peptide YY (PYY) concentration in dogs fed for 6 months on diets containing either virgin olive oil (VOO) or sunflower oil (SO) as the fat source (35% energy). F represents the fasting situation. The arrow denotes the time of food ingestion. Values are means ⫾ SEM (n ⫽ 4 for both groups). # p ⬍ 0.05 between the two groups at specific time points. Reprinted from Yago MD, et al. J Nutr Biochem 1997; 8: 502–507, with permission of Elsevier Ltd.
rate in the SO group had returned to premeal values and even a further, significant, decrease was observed. Bile acid (BA) concentration in bile secreted by fasting dogs was significantly higher in those animals long-term fed the SO diet. In this group, food intake did not modify initially the values for this parameter, but a significant decrease was evident from the second postprandial hour onwards. Opposite, the meal evoked in VOO dogs a rapid, significant and transient increase in the concentration of BA. In both groups, changes in BA secretion (output, Figure 129.3) were parallel to those in BA concentration. The marked elevation in BA concentration and secretion after eating suggests a greater involvement of the gallbladder in animals fed with the diet rich in VOO. This enhanced postprandial response in BA would increase the BA circulating pool, the efficiency of the enterohepatic circulation of these anions and hepatic choleresis which, in turn, would account for the prolonged bile flow found in VOO animals as compared to SO ones. Enhanced gallbladder emptying in animals fed on olive oil diets is consistent with the ability of oleic acid to strongly stimulate the release of CCK. This peptide causes gallbladder contraction and relaxation of the sphincter of Oddi, thus allowing concentrated bile (rich in BA) to flow into the duodenum.
1178
SECTION | II
Bile acid secretion (μmol min–1)
150
# *
VOO
SO
120
VOO group
SO group
18:1 n-9 (oleic acid)
61.89 ⫾ 2.00*
29.03 ⫾ 0.66
5.06 ⫾ 0.12*
42.16 ⫾ 1.70
MUFA
63.08 ⫾ 1.95*
29.67 ⫾ 0.66
+
PUFA
8.19 ⫾ 0.14*
44.62 ⫾ 1.55
5
SFA
29.03 ⫾ 1.86
25.93 ⫾ 1.93
2.51 ⫾ 0.23
2.99 ⫾ 0.30
#
18:2 n-6 (linoleic acid)
*
60
30 +
0 F
1
2
3
TABLE 129.3 Fatty acid composition of the liquid test meals administered to cholecystectomized patients. Fatty acid
*
90
Major Organ Systems Including Liver and Metabolism
4
Time (h) FIGURE 129.3 Bile acid secretion in dogs fed for 8 months on diets containing either virgin olive oil (VOO) or sunflower oil (SO) as the fat source (33% energy). F represents the fasting situation. The arrow denotes the time of food ingestion. Values are means ⫾ SEM (n ⫽ 4 for both groups). For the VOO group, * p ⬍ 0.05 as compared with the respective fasting value; for the SO group, ⫹ p ⬍ 0.05 as compared with the respective fasting value; # p ⬍ 0.05 between the two groups at specific time points. Reprinted from Ballesta MC, et al. Br J Nutr 1992; 68: 175–182, with permission of Cambridge University Press.
129.3 OLIVE OIL AND DIGESTIVE SECRETION IN HUMANS Eighteen subjects with gallstones in the gallbladder and scheduled for elective surgery (cholecystectomy) were assigned to a virgin olive oil group (VOO) or a sunflower oil group (SO). They were instructed to consume their habitual diets (at home) for the 30-day period before surgery, with these requirements: (1) the only cooking fat had to be VOO or SO; (2) avoidance of eating food items high in other types of dietary fat (for example, saturated fat). Compliance with the diets was checked by completion of four 7-day dietary records. Dietary assessment showed that intake of energy, protein, carbohydrates and total fat was similar in both groups, with the main difference concerning the intake of MUFA (mean value of 40.1% of total fatty acids in VOO group versus 26.2% in SO group) and PUFA (8.3% in VOO group versus 19.8% in SO group). Samples of gallbladder bile were obtained during cholecystectomy. In addition, a post-cholecystectomy study was conducted 48 h after surgery, once it had been confirmed that subjects could tolerate oral feeding. The test meals were isoenergetic and isonitrogenous, contained no measurable amounts of cholesterol and phospholipids and were composed of 17% energy as protein, 30% as fat, and 53% as carbohydrate, in addition to vitamins and minerals. Virgin olive oil was added to the meal given to the VOO group and sunflower oil to the SO group (Table 129.3). For the post-cholecystectomy study, the participants were intubated with a nasoduodenal tube that enabled separate
UFA/SFA
Subjects were kept on diets that contained virgin olive oil (VOO) or sunflower oil (SO) as the main source of dietary fat for 30 days before cholecystectomy and thereafter were given meals that included the corresponding oil. Results are expressed as percentage of total fatty acid content. Values are means ⫾ SEM (n ⫽ 6 for both groups). * p ⬍ 0.05 between the two meals. MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids; SFA: saturated fatty acids; UFA: unsaturated fatty acids. Reprinted from Yago MD, et al. Nutrition 2005; 21: 339–347, with permission of Elsevier Ltd.
aspiration of gastric and duodenal contents. Each subject was studied on 2 consecutive days. On the first one, after a minimum 8-h fast, samples were taken to establish basal values and then the subjects were given the test meals (200 mL ingested over 30 min). Postprandial sampling took place at 30, 60, 120, and 180 min after beginning the ingestion of the meal (see Figure 129.4). At every point, samples of blood and duodenal and gastric contents were collected. The complete procedure was repeated on the second experimental day. The following sections describe the main findings of this study.
129.3.1 Plasma Profile of Gastrointestinal Peptides The ingestion of the liquid meal (Yago et al., 1997c) led to significantly higher levels of plasma CCK in VOO subjects compared with SO subjects throughout the 30–120 min postprandial period (Figure 129.5A), which seems reasonable given the potency of oleic acid as a CCK releaser. We also determined the circulating levels of the inhibitory peptides PYY (Serrano et al., 1997) and PP (Serrano et al., 1998). In the fasting state, plasma PYY concentration in group VOO was significantly greater than in group SO. Following the test meal, PYY levels in VOO subjects remained consistently higher (Figure 129.5B). Although feeding did not evoke in either group a marked elevation in plasma PYY, values tended to increase, slowly, during the postprandial period (Figure 129.5B). Investigations in
1179
CHAPTER | 129 Influence of Olive Oil on Pancreatic, Biliary and Gastric Secretion: Role of Gastrointestinal Peptides
Cholecystectomy
0
P
P
P
120
180
F
30 60
Time (min)
0
P
P
P
P
120
180
Meal
P
Fast
Fast
30 days (adaptation period) VOO or SO as culinary fat
F
Meal
Surgery
30 60
Time (min)
1st experimental day
2nd experimental day
FIGURE 129.4 Experimental design of the studies on adaptation to dietary fat type in humans. Subjects with gallstones in the gallbladder were kept on diets that contained either virgin olive oil (VOO) or sunflower oil (SO) as the main source of fat for 30 days before elective cholecystectomy. Samples of gallbladder bile were obtained at surgery. Forty-eight hours after cholecystectomy, fasted subjects were given mixed liquid meals with 30% energy of the corresponding oil. Samples of blood and gastrointestinal contents were taken at the indicated times. The complete procedure was repeated next day. F and P denote, respectively, fasting and postprandial samples. Meals were ingested over 30 min (black area). The dotted area represents the rest of the postprandial period.
VOO
SO
125
50
#
400
#
30 20 10
100
75
50
25
0 F A
Plasma PP (pmol L–1)
#
40
*
# Plasma PYY (pmol L–1)
Plasma CCK (pmol L–1)
#
30 60
120
Time (min)
180
# *
300
# *
# *
200 +
+
100
0 F
30 60
B
120
Time (min)
180
F C
30 60
120
180
Time (min)
FIGURE 129.5 Time-course evolution of plasma concentrations of cholecystokinin (CCK, A), peptide YY (PYY, B), and pancreatic polypeptide (PP, C) in cholecystectomized subjects after the administration of a liquid meal containing either virgin olive oil (VOO) or sunflower oil (SO). Subjects were kept on diets that contained VOO or SO as the main source of dietary fat for 30 days before surgery and thereafter were given meals that included the corresponding oil. F represents the fasting situation. The black bar denotes the duration of meal ingestion (30 min). Values are means ⫾ SEM (n ⫽ 18 for both groups). For the VOO group, * p ⬍ 0.05 as compared with the respective fasting value; for the SO group, ⫹ p ⬍ 0.05 as compared with the respective fasting value; # p ⬍ 0.05 between the two groups at specific time points. A: Reprinted from Yago MD, et al. Br J Nutr 1997; 78: 27–39, with permission of Cambridge University Press. B: Reprinted from Serrano P, et al. Dig Dis Sci 1997; 42: 626–633, with permission of Springer. C: Reprinted from Serrano P, et al. Biogenic Amines 1998; 14: 313–330, with permission of Koninklijke Brill NV.
humans have shown a delayed PYY response to food that depends on the size of the meal and its fat content. Fatevoked PYY release is in part due to a direct action of unabsorbed fatty acids reaching the endocrine L cells at the ileo-colonic mucosa. In our conditions, the size, fat content and administration rate of the meal, together with the absorptive processes along the tube may have resulted in a delayed arrival of a small amount of fatty acids to the distal intestine. Postprandial release of PYY is also stimulated by a number of gastrointestinal peptides, including CCK. Again, this is coherent with our results, since postprandial levels of CCK and PYY were higher in group VOO, where
we found a significant positive correlation between both parameters (Yago et al., 1997d). As for pancreatic polypeptide (PP), our data (Serrano et al., 1998) evidence in the two dietary groups was the typical response to a meal, i.e., an early peak followed by a sustained elevation for several hours (Figure 129.5C). The peak is derived from vagal cholinergic activity and the second, intestinal, phase from a complex interaction between nerves and hormones. Our experiments revealed that circulating PP levels were significantly greater in VOO group for the entire postprandial period (Figure 129.5C), in agreement with intestinal perfusion studies showing MUFA as
1180
SECTION | II
strong stimulants for human PP secretion (Scarpello et al., 1982). Moreover, the fact that both CCK and PP plasma levels were higher in VOO subjects is consistent with the role of CCK as a hormonal mediator in the intestinal phase of PP release. Of note, the higher concentrations of the inhibitory peptides PYY and PP associated with the habitual intake of olive oil confirm our previous results in dogs. This has a special relevance if we consider that the 30-day period before surgery cannot be strictly considered as an adaptation period to the diet, at least with the meaning that this word receives in the animal studies, where the experimental subjects can be fed on chemically defined, controlled diets for long periods of time.
129.3.2 Exocrine Pancreatic Secretion In the same subjects we determined the activity of pancreatic amylase, trypsin, chymotrypsin, lipase and colipase in samples of duodenal content (Yago et al., 1997c). No differences became apparent between the groups, with the exception of postprandial lipase activity, which was significantly higher in VOO subjects. Overall, our results do not reveal a great influence of dietary fat type on human pancreatic enzyme secretion, except for lipase. One explanation is that longer periods of time may be needed for the human pancreas to be affected by a modification in dietary fat composition. Yet, our observation of clear differences in the plasma profile of gastrointestinal peptides in these subjects (CCK, PYY, PP) together with the fact that we do not have data of true
Major Organ Systems Including Liver and Metabolism
secretory rates (output) due to the limitations associated with human research, obliges us not to exclude the possibility that the type of fat can affect human pancreatic enzyme secretion to a greater extent than is evident from this study.
129.3.3 Gastric Secretion Since the 19th century, many authors have demonstrated that the presence of fat at different segments of the gastrointestinal tract inhibits acid secretion. In our study (Serrano et al., 1997), gastric pH in fasted subjects did not differ between the dietary groups (Figure 129.6A). Using physiological stimulation by test liquid meals (pH 6.33) that were permitted to seek its natural pH, we found a prompt pH increase after food ingestion in both groups, followed by a decline due to the continuous secretion of acid by the stomach. In group VOO, pH remained elevated until the first postprandial hour and then this parameter showed a slow decrease towards the fasting value, whereas in group SO, the return to baseline was completed within 1 h and a further decline was recorded, so values significantly lower than the fasting ones were reached at the end of the experimental period (Figure 129.6A). Accordingly, H⫹ concentration (mEq L⫺1) was significantly greater in SO volunteers at 60 min postprandially (data not shown). These results provide evidence in humans that a 30-day period of VOO diets results in attenuated intragastric acidity in response to a meal when compared with diets containing SO. The effect of olive oil on gastric acid secretion involved the suppression of serum gastrin (Serrano et al., 1997; Figure 129.6B), the main mediator of the postprandial
40
5 + *
Gastric pH
4
SO
VOO
Serum gastrin (pmol L–1)
VOO
*
3
2
SO
35 +
30
25
20
+ 15
1 F A
30
60
120
Time (min)
F
180 B
30
60
120
180
Time (min)
FIGURE 129.6 Time-course evolution of gastric pH (A) and serum gastrin concentration (B) in cholecystectomized subjects after the administration of a liquid meal containing either virgin olive oil (VOO) or sunflower oil (SO). Subjects were kept on diets that contained VOO or SO as the main source of dietary fat for 30 days before surgery and thereafter were given meals that included the corresponding oil. F represents the fasting situation. The black bar denotes the duration of meal ingestion (30 min). Values are means ⫾ SEM (n ⫽ 18 for both groups). For the VOO group, * p ⬍ 0.05 as compared with the respective fasting value; for the SO group, ⫹ p ⬍ 0.05 as compared with the respective fasting value. Reprinted from Serrano P, et al. Dig Dis Sci 1997; 42: 626–633, with permission of Springer.
CHAPTER | 129 Influence of Olive Oil on Pancreatic, Biliary and Gastric Secretion: Role of Gastrointestinal Peptides
increases in gastric acid secretion in humans. Considering the stimuli for gastrin release, a similar increase in serum levels should have been expected in both dietary groups after ingestion of two test meals that were identical except for the type of fat added. We suggest that gastrin suppression in VOO subjects is associated with some hormonal factor released by oleic acid, with the ability to reduce gastrin secretion from G cells. A candidate is CCK, which displayed higher postprandial levels in VOO subjects. CCK has been described as a typical enterogastrone, i.e. a circulating factor released from the intestine that inhibits gastric acid secretion. The enterogastrone effect of CCK is not direct, but mediated through the release of somatostatin from gastric D cells. Somatostatin, in turn, inhibits not only gastrin release, but also histamine release from enterochromaffin-like cells and acid secretion from parietal cells. It should be mentioned here that, as part of our study, we measured plasma somatostatin and found similar values in groups VOO and SO (Serrano et al., 1997). Nonetheless, this does not rule out our hypothesis, since most somatostatin action on gastric acid secretion is exerted through paracrine pathways. We also propose PYY, higher again in the VOO group, as a mediator for the action of olive oil on acid secretion, since this peptide is a well-established inhibitor of gastric acid secretion. Reduction of gastric acidity by VOO may contribute, together with other mechanisms (Alarcón de la Lastra et al., 2002), to the beneficial effects of virgin olive oil in the prevention and healing of peptic ulcers.
129.3.4 Biliary Lipid Composition and Bile Lithogenicity The difficulty and risks of sample collection and a big disparity in experimental protocols has probably meant that the very few studies on the influence of dietary fat on bile composition and lithogenicity in humans have yielded equivocal results. Our experimental design (see the beginning of section 129.3) made possible the collection of fasting gallbladder bile at surgery. In addition, the post-cholecystectomy experiments allowed us to study the composition of fasting and postprandial hepatic bile. This was important, given that the analysis of only gallbladder bile could have masked the influence of the dietary intervention because of the occurrence of pre-existing gallstones. We collected hepatic bile samples by duodenal intubation. Although duodenal sampling does not permit the measurement of biliary lipid secretion (output), it provides a mechanism by which the proportion of biliary lipid classes (relative to total lipids) in hepatic bile can be quantified, offering data that would otherwise be difficult to obtain in humans in vivo. As noted in Table 129.4, intake of VOO or SO diets for 30 days before cholecystectomy did not affect the cholesterol saturation index (CSI) in gallbladder bile (Yago et al., 2005),
1181
a somehow expected finding considering the presence of established gallstones in the gallbladder. However, when gallbladder bile composition was expressed in terms of absolute concentrations of lipids, the concentration of cholesterol and total bile acids (BA) was shown to be significantly diminished in SO subjects. Gallbladder bile in this group also had a markedly lower total lipid content. A decreased total lipid concentration in gallbladder bile is common in patients with cholesterol gallstones (Chijiiwa et al., 1993), but we found this feature only in the SO group. Differences
TABLE 129.4 Biliary lipid composition of gallbladder bile sampled at the time of surgery. VOO group
SO group
142.23 ⫾ 15.05*
87.95 ⫾ 20.42
66.47 ⫾ 0.56
60.37 ⫾ 3.11
17.60 ⫾ 1.78*
9.92 ⫾ 1.94
8.33 ⫾ 0.52
7.54 ⫾ 1.20
mmol L⫺1
54.06 ⫾ 6.00
42.55 ⫾ 7.94
mol %
25.20 ⫾ 0.25*
32.09 ⫾ 2.43
Cholesterol/bile acids
0.126 ⫾ 0.009
0.132 ⫾ 0.027
Phospholipids/bile acids
0.379 ⫾ 0.006
0.553 ⫾ 0.072
Cholesterol/ phospholipids
0.331 ⫾ 0.021*
0.236 ⫾ 0.034
CSIa
0.835 ⫾ 0.052
0.857 ⫾ 0.167
CSIb
1.056 ⫾ 0.069
1.013 ⫾ 0.176
11.48 ⫾ 1.22†
7.77 ⫾ 1.59
Total bile acids mmol L⫺1 mol % Cholesterol mmol L⫺1 mol % Phospholipids
Linkage coefficients
Total lipids (g dL⫺1)
Subjects were kept on diets that contained virgin olive oil (VOO) or sunflower oil (SO) as the main source of dietary fat for 30 days before surgery (cholecystectomy). Values are means ⫾ SEM (n ⫽ 9 for both groups). *p ⬍ 0.05 between the two groups. †p ⫽ 0.063 between the two groups. a Cholesterol saturation index (CSI) calculated according to Metzger et al. (1972). b CSI calculated according to Carey (1978). Reprinted from Yago MD, et al. Nutrition 2005; 21: 339–347, with permission of Elsevier Ltd.
1182
SECTION | II
in gallbladder bile composition did not reflect those in fasting hepatic bile, which makes us speculate that our results are the consequence of differential concentration capacity of gallbladder epithelium. In the post-cholecystectomy study, we found no differences between our groups in molar percentages of lipids or cholesterol saturation at fasting. CSI was indicative of cholesterol supersaturation in all of them, a physiological phenomenon after overnight fasting, which interrupts the enterohepatic circulation of BA causing bile to supersaturate with cholesterol. Afterwards, meal-evoked recirculation of BA increases BA output and hepatic bile reverts to a less saturated state, a phenomenon that has been observed in healthy, lithiasic, and cholecystectomized subjects. Surprisingly, we found this typical postprandial response only in subjects given the VOO diet whereas hepatic bile remained supersaturated until the end of the experiment in those given SO (Figure 129.7). It is difficult to interpret these results without data of secretory rates, but two mechanisms could explain these differences. On the one hand, the ingestion (chronic, acute, or both) of the two types of fat might have produced different feeding motility patterns through differential stimulation of humoral agents, in turn affecting choleresis. A second possibility is that patients in the SO group have normal BA secretion in response to food but that this is coupled with a supranormal cholesterol secretion rate, which is in line with the presumed hypocholesterolemic effect of n-6 PUFA.
SUMMARY POINTS Compared with sunflower oil (SO), habitual intake of virgin olive oil (VOO) may be beneficial for the gastrointestinal system. The following points summarize our main conclusions: ●
●
●
●
3 VOO
SO
2.5 #
CSI
2 1.5
●
1 * 0.5
*
0 F
30 60
120
Major Organ Systems Including Liver and Metabolism
Long-term adaptation to diets high in VOO attenuates postprandial exocrine pancreatic secretory activity in dogs in a manner proportional to its amount in the diet. This effect is accompanied by elevation in the circulating levels of peptide YY and pancreatic polypeptide, and is exerted without derangement of nutrient utilization, hence representing a key advantage in terms of pancreatic economy. Patterns of biliary response to food in dogs fed for 8 months with diets containing VOO are indicative of a greater involvement of the gallbladder. A 30-day adaptation period to a diet containing VOO results in reduced gastric acidity following a liquid meal with the same oil in humans. Olive oil, then, may be a useful component in the nutritional therapy of those gastrointestinal diseases requiring a limitation of acid secretion. The type of dietary fat habitually consumed can influence bile composition in humans, including those with established cholelithiasis. In our study, this influence was noted in the presence of more concentrated gallbladder bile in subjects given a VOO diet without a parallel increase in cholesterol saturation index. In cholecystectomized subjects, the physiological postprandial decrease in hepatic bile lithogenicity occurred in those given VOO but not SO. Provided they are confirmed in healthy subjects, these results, together with a very efficient gallbladder emptying as suggested by our investigations in dogs, would indicate a beneficial role of olive oil in the pathogenesis of gallstone disease. The above effects in humans are mediated, at least in part, by changes in the fasting and/or postprandial release of several gastrointestinal hormones: intake of VOO results in higher plasma levels of cholecystokinin, peptide YY and pancreatic polypeptide and lower levels of gastrin compared with SO diets.
180
Time (min) FIGURE 129.7 Time-course evolution of the cholesterol saturation index (CSI) in duodenal samples from cholecystectomized subjects after the administration of a liquid meal containing either virgin olive oil (VOO) or sunflower oil (SO). Subjects were kept on diets that contained VOO or SO as the main source of dietary fat for 30 days before surgery and thereafter were given meals that included the corresponding oil. F represents the fasting situation. The black bar denotes the duration of meal ingestion (30 min). Values are means ⫾ SEM (n ⫽ 18 for both groups). CSI was calculated in these samples according to the method of Metzger et al., 1972. For the VOO group, * p ⬍ 0.05 as compared with the respective fasting value; # p ⬍ 0.05 between the two groups at specific time points. Reprinted from Yago MD, et al. Nutrition 2005; 21: 339–347, with permission of Elsevier Ltd.
REFERENCES Alarcón de la Lastra, C., Barranco, M.D., Martín, M.J., Herrerías, J., Motilva, V., 2002. Extra-virgin olive oil-enriched diets reduce indomethacin-induced gastric oxidative damage in rats. Dig. Dis. Sci. 47, 2783–2790. Ballesta, M.C., Mañas, M., Mataix, F.J., Martínez-Victoria, E., Seiquer, I., 1990a. Long-term adaptation of pancreatic response by dogs to dietary fats of different degrees of saturation: olive and sunflower oil. Br. J. Nutr. 64, 487–496. Ballesta, M.C., Mañas, M., Martínez-Victoria, E., Seiquer, I., Huertas, J.R., Mataix, F.J., 1992. Adaptation of biliary response to dietary olive oil and sunflower-seed oil in dogs. Br. J. Nutr. 68, 175–182.
CHAPTER | 129 Influence of Olive Oil on Pancreatic, Biliary and Gastric Secretion: Role of Gastrointestinal Peptides
Ballesta, M.C., Martínez-Victoria, E., Mañas, M., Mataix, F.J., Seiquer, I., Huertas, J.R., 1990b. Fat digestibility in dogs after long-term adaptation to diets varying in the degree of lipid saturation (virgin olive oil and sunflower oil). Med. Sci. Res. 18, 517–519. Ballesta, M.C., Martínez-Victoria, E., Mañas, M., Mataix, F.J., Seiquer, I., Huertas, J.R., 1991. Protein digestibility in dog. Effect of the quantity and quality of dietary fat (virgin olive oil and sunflower oil). Die Nahrung-food 35, 161–167. Carey, M.C., 1978. Critical tables for calculating the cholesterol saturation of native bile. J. Lipid Res. 19, 945–955. Chijiiwa, K., Hirota, I., Noshiro, H., 1993. High vesicular cholesterol and protein in bile are associated with formation of cholesterol but not pigment gallstones. Dig. Dis. Sci. 38, 161–166. Metzger, A.L., Heymsfield, S., Grundy, S.M., 1972. The lithogenic index – a numerical expression for the relative lithogenicity of bile. Gastroenterology 62, 499–501. Scarpello, J.H., Vinik, A.I., Owyang, C., 1982. The intestinal phase of pancreatic polypeptide release. Gastroenterology 82, 406–412. Serrano, P., Yago, M.D., Mañas, M., Calpena, R., Mataix, J., MartínezVictoria, E., 1997. Influence of type of dietary fat (olive and sunflower oil) upon gastric acid secretion and release of gastrin, somatostatin, and peptide YY in man. Dig. Dis. Sci. 42, 626–633. Serrano, P., Yago, M.D., Martínez-Victoria, E., Medrano, J., Mataix, J., Mañas, M., 1998. Influence of the type of dietary fat upon the plasma
1183
levels of secretin and pancreatic polypeptide in cholecystectomized humans. Biog. Amines 14, 313–330. Yago, M.D., Martínez-Victoria, E., Mañas, M., Martínez, M.A., Mataix, J., 1997a. Plasma peptide YY and pancreatic polypeptide in dogs after long-term adaptation to dietary fats of different degrees of saturation: olive and sunflower oil. J. Nutr. Biochem. 8, 502–507. Yago, M.D., Martínez-Victoria, E., Huertas, J.R., Mañas, M., 1997b. Effects of the amount and type of dietary fat on exocrine pancreatic secretion in dogs after different periods of adaptation. Arch. Physiol. Biochem. 105, 78–85. Yago, M.D., González, M.V., Martínez-Victoria, E., Mataix, J., Medrano, J., Calpena, R., Pérez, M.T., Mañas, M., 1997c. Pancreatic enzyme secretion in response to test meals differing in the quality of dietary fat (olive and sunflowerseed oils) in human subjects. Br. J. Nutr. 78, 27–39. Yago, M.D., Mañas, M., González, M.V., Martínez-Victoria, E., Pérez, M.T., Mataix, J., 1997d. Plasma levels of cholecystokinin and peptide YY in humans: response to dietary fats of different degrees of unsaturation (olive and sunflower oil). Biog. Amines 13, 319–331. Yago, M.D., González, V., Serrano, P., Calpena, R., Martínez, M.A., Martínez-Victoria, E., Mañas, M., 2005. Effect of the type of dietary fat on biliary lipid composition and bile lithogenicity in humans with cholesterol gallstone disease. Nutrition 21, 339–347.
This page intentionally left blank
Chapter 130
Effects of Olive Oil on Fatty Acid Composition of Pancreatic Cell Membranes: Modulation of Acinar Cell Function and Signaling María Dolores Yago1, María Alba Martínez1, José Antonio Pariente2, Emilio Martínez-Victoria1 and Mariano Mañas1 1
Departamento de Fisiología, Instituto de Nutrición y Tecnología de Alimentos, Centro de Investigaciones Biomédicas, Universidad de Granada, Spain 2 Departamento de Fisiología, Facultad de Ciencias, Universidad de Extremadura, Badajoz, Spain
130.1 INTRODUCTION The involvement of diet in human health and in the prevention and treatment of disease is a well-known topic, recognized and accepted by international institutions for a long time. Over recent years, many functional foods and ingredients have been developed and characterized. Beyond adequate nutritional effects, their biological actions are relevant to an improved state of health and well-being and/or reduction of risk of chronic disease. It is known that Mediterranean countries have a lower prevalence of cardiovascular diseases, compared to Northern Europe and other developed countries, and life expectancies that are among the highest. Consequently, the Mediterranean diet has been described as a cultural model for dietary improvement. There is a need for sound nutritional and scientific information to clarify the many claims for the health-enhancing effects of the Mediterranean diet and the generation of new data indicating the health benefits of the components of this diet is important. Used as the major culinary fat, olive oil is an integral ingredient of the Mediterranean diet (which contains up to 40% of calories as fat), and accumulating evidence suggests that it may have health benefits. In fact, olive oil appears to be an example of a functional food, with varied components that may contribute to its overall therapeutic characteristics, especially its high levels of monounsaturated fatty acids (MUFA). The type of dietary fat is more closely related to the incidence of some chronic diseases than the level of dietary Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
fat (Keys, 1970). Thus, habitual consumption of dietary fats with different fatty acid profiles may have different physiological consequences and effects on health. It is generally accepted that biological membranes do not have a constant composition. Some factors, such as age, physiologic state, cell type, antioxidant capacity, metabolic activity or the diet can modulate the structure and function of cell membranes (Clandinin et al., 1991). Many authors have investigated the adaptation of human and animal cells to dietary fat, and the results indicate that the response is tissue-specific. After a change in dietary fat composition, membranes of the intestinal epithelial cells seem to be the first to adapt (Stenson et al., 1989). This adaptation involves a modification in membrane fluidity and cell function, particularly the transport capacity and metabolic activity of the membranes (Clandinin et al., 1991). The liver is also very sensitive to dietary-induced changes (Quiles et al., 1999). In contrast, other tissues such as the brain (Abedin et al., 1999) or skeletal muscle (Quiles et al., 1999) seem to show relatively minor adaptive alterations in the lipid composition of their membranes. The biological effects of dietary fatty acids are partly due to their incorporation into the cellular structures of organs and tissues (see Table 130.1). Changes in membrane fatty acid composition may, in turn, influence cell function (Quiles et al., 2001). This is not an unexpected finding, since there is growing evidence that fatty acids, in addition to their role in determining membrane structure and fluidity, can participate in intracellular processes as diverse as signal transduction or the regulation of gene expression.
1185
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1186
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 130.1 Key facts of membrane fatty acid composition and cell function. 1. Mammalian cellular and subcellular membranes consist of a lipid bilayer, with functional proteins either bound to the surface or embedded into the bilayer 2. The lipid bilayer is composed of phospholipids and cholesterol. Polar headgroups of these lipids occupy the outer surfaces of the bilayer while the inner hydrophobic core consists primarily of long fatty acyl chains 3. Although membrane composition is in part determined by endogenous control mechanisms, the diet constitutes a major exogenous determinant. Thus, dietary differences in fatty acid intake alter the fatty acid composition of membrane phospholipids 4. Tissues differ in their relative responsiveness to alterations in the fatty acid profile of the diet 5. Dietary-induced modifications in membrane fatty acids can be extensive enough to affect a number of cellular functions. This can be achieved by different mechanisms: a. Many of the functional responses are likely caused directly by the membrane lipid structural changes, which affect either bulk lipid fluidity or specific lipid domains. The conformation of transporters, receptors, and membrane-linked enzymes is probably sensitive to changes in the structure of their lipid microenvironment, leading to changes in activity b. Phospholipid fatty acids participate themselves in signaling pathways. Examples are found in the hydrolysis of phosphatidylinositol phosphates by phospholipases or eicosanoid production. Consequently, these pathways may be strongly influenced by dietary fat c. Last, a direct link between dietary fatty acid consumption, gene expression and tissue proliferation is provided by the fact that many transcription factors are regulated by fatty acids This table lists the key facts of membrane fatty acids including the basics of membrane composition, influence of dietary fatty acid profile and effects on cell function.
130.2 DIETARY LIPIDS AND PANCREATIC SECRETION Since the findings of Pavlov in the early 1900s many authors have investigated in various species the adaptation of the exocrine pancreas to the type of food available. The general consensus of these studies is that, for dietary components such as carbohydrates, protein, and lipid, there is a positive relationship between their level in the diet and the tissue content of pancreatic enzymes necessary for their breakdown. Concerning lipids, both the amount and type seem to have an influence (Yago et al., 1997c), although a controversy over the effects of the fatty acid composition of dietary fat on the adaptive process of pancreatic enzymes is unresolved. Previous investigations of the in vivo pancreatic response after medium- or long-term intake of diets differing in the fat source indicated that pancreatic adaptation to dietary fat type is mediated, at least in part, by changes in the circulating levels of some gastrointestinal hormones (Yago et al., 1997a, 1997b). So, a first option to explain the mechanisms of the pancreatic adaptation to dietary fat involves the existence of hormonal mediators. However, a second possibility to explain the mechanisms of pancreatic adaptation is that dietary fat composition is affecting directly the secretory activity of pancreatic
acinar cells by changing their responsiveness to circulating secretagogues and/or altering Ca2⫹ signaling as a consequence of the modification of the fatty acid composition of the pancreatic membranes. In different tissues, there is evidence that the lipid profile of the diet can influence the fatty acid composition of cell membranes, this being associated with a modification of cell function (Quiles et al., 2001). Regarding the exocrine pancreas, information on this topic is very limited (Soriguer et al., 2000) but supports the above view. Besides, it should be mentioned that the most frequent methodological approach has consisted of the analysis of the enzyme content of the gland. A different approach is to examine in the conscious animal or in isolated pancreatic acini the secretory response after long-term intake of specific diets, studying the influence of the modification of the lipid profile of pancreatic membranes. In this chapter we will describe the effect of dietary fat type on the fatty acid composition of rat pancreatic membranes and the interaction with the exocrine pancreatic secretion in anesthetized animals, in preparation of isolated rat pancreatic acini and in the AR42J cell line. The two dietary fats chosen were virgin olive oil, a typical component of the Mediterranean diet and a good source of MUFA, and sunflower oil, rich in polyunsaturated fatty acids (PUFA). Both oils compete in Southern European markets for
1187
CHAPTER | 130 Effects of Olive Oil on Fatty Acid Composition of Pancreatic Cell Membranes
consumers’ preference. Furthermore, olive oil is a major component of the Mediterranean diet, and its role in human health is actively debated at present.
130.3 PANCREATIC SECRETION IN ANESTHETIZED RATS This work (Díaz et al., 2003) was carried out in male weaning Wistar rats. The rats were fed over an 8-week period with two semi-purified, isoenergetic and isonitrogenous diets that were essentially AIN-93G diets except that total fat content was increased from 7 to 10 wt% at the expense of carbohydrate. The two diets differed only in the nature of the fat source: virgin olive oil or sunflower oil (Table 130.2). At the end of the 8-week feeding period we examined the exocrine pancreatic secretion, both at rest and following stimulation with cholecystokinin-octapeptide (CCK-8). The protein and amylase content of the pancreas was also analyzed. Furthermore, to confirm a direct effect of the type of dietary fat, the fatty acid composition of pancreatic membranes was determined after feeding the respective diets. This investigation showed that the diets used in the current study did not affect food intake and body weight gain during the 8-week feeding period. We also failed to find any difference between our groups in the weight of pancreases. The analysis of pancreatic cell membranes showed that our dietary protocol was satisfactory, since a direct effect of the type of dietary fat on the pancreatic gland was confirmed. The fatty acid profile of pancreatic cell membranes reflected the composition of the lipid component of the experimental diets. After 8 weeks on the diets, pancreatic membranes in the olive oil group had significantly higher levels of 18:1 n-9 and total MUFA, whereas a higher level of PUFA, particularly n-6 PUFA such as 18:2 n-6 and 20:4 n-6, was found in the sunflower oil group (Table 130.3). These data are consistent with those obtained by others (Begin et al., 1990; Soriguer et al., 2000). The proportion of total SFA was similar in the membranes of the two groups. In fact, except for 18:0 values, no further differences could be detected. This resistance of the SFA fraction to dietary-induced alterations is a common feature in different tissues (Soriguer et al., 2000). In addition, feeding diets rich in virgin olive oil or sunflower oil did not alter significantly the SFA/unsaturated fatty acid (UFA) ratio or the unsaturation index (UI). This suggests that membranes display a considerable degree of homeostasis with respect to these parameters, and that changes in the proportion of several major fatty acids are associated with some metabolic compensation in order to keep plasma membrane fluidity within a certain range of values. By using the method of direct cannulation of the pancreatic duct we were able to confirm in this study a modification of the secretory output as a function of the type of dietary fat. Under resting conditions pancreatic
TABLE 130.2 Fatty acid composition of the experimental diets. Fatty acid
Olive oil group
Sunflower oil group
16:0
11.44
7.31
16:1 n-7
0.85
0.19
18:0
4.38
4.59
18:1 n-9
74.88
32.62
18:2 n-6
7.72
55.17
18:3 n-3
0.62
0.10
SFA
15.82
11.90
MUFA
75.84
32.83
PUFA
8.34
55.27
Two groups of male weaning Wistar rats were fed for 8 weeks with semi-purified diets that differed only in the nature of the fat source (10 wt%): virgin olive oil or sunflower oil. Data are given in percentages of total fatty acid content (mean values of four replicates). SFA: saturated fatty acids; MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids. Reprinted from Díaz et al., Lipids 2003; 38: 1119–1126, with permission of Springer.
flow-rate and amylase output were significantly higher in rats on the sunflower oil diet than in those on the olive oil one (0.68 ⫾ 0.054 versus 0.37 ⫾ 0.021 μL min⫺1 and 61.79 ⫾ 6.03 versus 35.84 ⫾ 3.55 mU min⫺1, respectively). The mechanism underlying this effect is, however, far from clear. In the whole animal, exocrine pancreatic secretion is regulated by a complex integration of hormonal and neural mechanisms, with stimulatory factors being balanced by a variety of inhibitory regulators, including peptide YY (PYY). In the pancreas, PYY has been shown to inhibit not only the secretion stimulated by secretin and/or CCK, but also basal secretion. In previous studies conducted in dogs and humans we found that medium-/long-term intake of diets containing olive oil as the major fat source evoked a significant elevation of plasma PYY in resting conditions as compared to the levels measured after sunflower oil feeding (Yago et al., 1997b). Thus, it is tempting to speculate that the diminished exocrine secretion observed in resting conditions in the olive oil group of the present study might reflect an augmentation of the resting PYY levels. Changes in exocrine pancreatic secretion following intravenous infusion of CCK-8 suggest that the two groups responded differently to this secretagogue, with the time course of changes in the secretory parameters during and after the infusion of CCK-8 showing a different pattern in each group (Figures 130.1 and 130.2). The existence in the
1188
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 130.3 Fatty acid composition of pancreatic cell membranes in rats fed diets containing different dietary fats. Fatty acid
Olive oil group
Sunflower oil group
16:0
25.04 ⫾ 0.50
26.39 ⫾ 1.22
16:1 n-7
5.09 ⫾ 0.45
4.17 ⫾ 0.43
18:0
6.87 ⫾ 0.65*
10.21 ⫾ 0.92
18:1 n-9
43.09 ⫾ 2.19***
24.90 ⫾ 1.08
18:2 n-6
4.58 ⫾ 0.65***
18.58 ⫾ 1.20
18:3 n-3
0.40 ⫾ 0.05
0.32 ⫾ 0.08
SFA
39.12 ⫾ 2.49
39.66 ⫾ 2.04
MUFA
48.30 ⫾ 2.52***
29.20 ⫾ 1.41
PUFA
12.58 ⫾ 1.49***
31.14 ⫾ 2.04
n-6 PUFA
10.89 ⫾ 1.34***
29.29 ⫾ 1.99
2.43 ⫾ 0.22
3.15 ⫾ 0.30
UI
Two groups of male weaning Wistar rats were fed for 8 weeks with semi-purified diets that differed only in the nature of the fat source (10 wt%): virgin olive oil or sunflower oil. Results are expressed as percentage of total fatty acid content. SFA: saturated fatty acids; MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids; UI: unsaturation index. UI was calculated according to: UI ⫽ [sum (fatty acid percent)*(number of double bonds)]/ SFA percent. Values are means ⫾ SEM (n ⫽ 12 for both groups). By row, values with asterisks are significantly different versus sunflower oil group: * p ⬍ 0.05; ***p ⬍ 0.001. Reprinted from Díaz et al., Lipids 2003; 38: 1119–1126, with permission of Springer.
2
250
Olive oil
+
Olive oil
Sunflower oil
Sunflower oil
200 Amylase activity (U mL−1)
Flow rate (μL min−1)
1.6 + 1.2 * *
0.8
# *
0.4
*
*
*
*
* * *
*
150
+ 100
50
CCK-8
CCK-8 0
0 0
40
60
80
100
140
160
Time (min) FIGURE 130.1 Time-course changes evoked by cholecystokininoctapeptide (CCK-8) on pancreatic flow-rate in anesthetized rats fed for 8 weeks with diets containing either virgin olive oil or sunflower oil as the fat source (10 wt%). Time zero represents the basal (resting) situation, which is immediately followed by the start of CCK-8 infusion (dark bar). All data are means ⫾ SEM of n ⫽ 12, except for basals, where n ⫽ 24. For the olive oil group, * p ⬍ 0.05 as compared with the respective basal value; for the sunflower oil group, ⫹ p ⬍ 0.05 as compared with the respective basal value; # p ⬍ 0.05 between the two groups at specific time points. Reprinted from Díaz et al., Lipids 2003; 38: 1119–1126, with permission of Springer.
0
40
80
120
160
Time (min) FIGURE 130.2 Time-course changes evoked by cholecystokininoctapeptide (CCK-8) on amylase activity in pancreatic juice of anesthetized rats fed for 8 weeks with diets containing either virgin olive oil or sunflower oil as the fat source (10 wt%). Time zero represents the basal (resting) situation, which is immediately followed by the start of CCK-8 infusion (dark bar). All data are means ⫾ SEM of n ⫽ 12, except for basals, where n ⫽ 24. For the olive oil group, *p ⬍ 0.05 as compared with the respective basal value; for the sunflower oil group, ⫹ p ⬍ 0.05 as compared with the respective basal value. Reprinted from Díaz et al., Lipids 2003; 38: 1119–1126, with permission of Springer.
1189
CHAPTER | 130 Effects of Olive Oil on Fatty Acid Composition of Pancreatic Cell Membranes
olive oil group, but not in the sunflower oil one, of a positive direct correlation between flow-rate and both protein concentration and amylase activity lends further support for an influence of the type of dietary lipids on the secretory activity of the exocrine pancreas. In conclusion, the present investigation shows that chronic intake of diets differing only in the type of fat added (olive oil or sunflower oil) influences the exocrine pancreatic secretion in anesthetized rats. The major differences between the dietary treatments concerned the magnitude of the secretion of fluid and amylase in resting conditions as well as the time course of changes in all major secretory parameters evoked by a continuous intravenous infusion of CCK-8 (Díaz et al., 2003). At the present time, there is no definite explanation for this behavior of the gland. Clearly, the observed differences in pancreatic secretion from anesthetized rats are not the consequence of a variation in tissue content of protein or amylase because we could not find a clear effect of the type of dietary lipids on protein content and amylase activity in pancreatic homogenates after feeding the experimental diets for 8 weeks. Considering the dietary-induced changes in fatty acid composition of pancreatic membranes, the possibility exists that in vivo data are reflecting a distinct modulatory effect of the type of dietary fat on the secretory activity at the cellular level.
conditions, the strongest secretory effect was observed at 0.1 nM CCK-8, and a characteristic decrease occurred after the addition of higher concentrations of the secretagogue. This pattern was followed in cells from both groups, but quantitatively marked differences were revealed between them. Thus, values for basal amylase release in acini from rats fed the olive oil diet were similar to those reported by most authors (Lajas et al., 1998), whereas release in cells from the sunflower oil-fed rats was markedly higher. In contrast to the observations in basal conditions, net amylase secretion in response to all concentrations of CCK-8 was drastically reduced after sunflower oil feeding (Figure 130.3). This diminished secretory activity may be explained in part by the attenuation of CCK-8-evoked Ca2⫹ responses in the sunflower oil group (Figure 130.4). Moreover, the fact that not only the absolute value of the cytosolic Ca2⫹ concentration ([Ca2⫹]c) peak but also the peak increase over basal were lower in cells from the sunflower oil-fed rats suggests a reduction in the filling state of CCK-8-releasable Ca2⫹ pools and/or a limitation in the production or effectiveness of the mediators that participate in the Ca2⫹-signaling pathways. Parallel experiments undertaken by our group (Martínez et al., 2004) have confirmed the effects of olive oil by using
130.4 EXPERIMENTS IN ISOLATED PANCREATIC ACINI The above idea was reinforced by our results (Yago et al., 2004) in viable pancreatic acinar cells isolated from rats kept on identical dietary protocol as in the previous study. Diets containing either virgin olive oil or sunflower oil were given to separate groups of rats for 8 weeks. Acinar cell function was assessed by determining basal and CCK-8stimulated amylase release in suspensions of viable acini. We also examined the effect of these diets on the mobilization of intracellular free Ca2⫹, a key mediator of CCK-8evoked enzyme secretion. That our dietary protocol was appropriate for our purposes was also confirmed by the membrane fatty acid analyses showing, like in previous work (Díaz et al., 2003), the sensitivity of the pancreas to dietary fat changes. After 8 weeks on the diets we found again that the membrane fatty acids were profoundly affected by the diets; the rats fed the olive oil diet had higher levels of 18:1 n-9 and total MUFA compared with the animals fed the sunflower oil diet. Reciprocally, the sunflower oil diet resulted in greater levels of total and n-6 PUFA than the olive oil diet. Adaptation to the diet did not modify the concentration– response curve for CCK-8-induced amylase release in pancreatic acinar cells. Our results are consistent with those in the literature (Lajas et al., 1998) since, in our experimental
Net amylase release (% total content)
20
** ***
15
***
10
5 Olive oil Sunflower oil
0 0.1
1
10
CCK-8 (nM) FIGURE 130.3 Net amylase release stimulated by cholecystokininoctapeptide (CCK-8) in viable pancreatic acini isolated from rats fed for 8 weeks with diets containing either virgin olive oil or sunflower oil as the fat source (10 wt%). Amylase released during the incubation (30 min) of viable acini with CCK-8 is expressed as a percentage of the initial total cell content. Acini exposed to the incubation medium alone served as unstimulated controls (basal release). The effect of CCK-8 is shown as net amylase release, i.e. increase above basal. Results are mean ⫾ SEM (n ⫽ 15–37 separate experiments). Mean values were significantly different between the dietary groups: ** p ⬍ 0.005, *** p ⬍ 0.001. Reprinted from Yago et al., Br J Nutr 2004; 91: 227–234, with permission of Cambridge University Press.
1190
SECTION | II
Major Organ Systems Including Liver and Metabolism
700
Olive oil
0.1 nM CCK-8
Sunflower oil
600
[Ca2+]c (nM)
500 400 300 200 100 0 0
36
72
108
144
181
217
253
289
325
Time (s) FIGURE 130.4 Time-course changes in cytosolic Ca2⫹ concentration ([Ca2⫹]c) evoked by 0.1 nM cholecystokinin-octapeptide (CCK-8) in suspensions of fura-2-loaded pancreatic acinar cells of rats fed for 8 weeks with diets containing either virgin olive oil or sunflower oil as the fat source (10 wt%). The arrow denotes the time point of addition of CCK-8. Results are mean ⫾ SEM (n ⫽ 5–7 independent experiments). The shaded area represents the existence of significant differences between the groups at individual time points (p ⬍ 0.05). Reprinted from Yago et al., Br J Nutr 2004; 91: 227–234, with permission of Cambridge University Press.
an inverted fluorescence microscope attached to a continuous perfusion system to study cellular Ca2⫹ homeostasis in single cells. A group fed a commercial chow was used as control. Feeding diets rich in virgin olive oil did not significantly alter the resting [Ca2⫹]c values or basal amylase secretion. However, both the Ca2⫹ oscillations and the large Ca2⫹ transients in response, respectively, to low (physiological) and high concentrations of CCK-8 were significantly enhanced by the olive oil diet (Figure 130.5) compared with the control one (Figure 130.6). These effects on Ca2⫹ mobilization correlated, to a great extent, with CCK-8-evoked amylase secretory activity. The differences in acinar secretory activity and Ca2⫹ mobilization in our studies are most probably related to the dietary-induced changes in cell membrane composition. Many steps of the stimulus–secretion coupling process are membrane-dependent; differential enrichment in certain fatty acids may influence the accessibility of the CCK receptor, the interaction with G proteins, or the functionality of enzymes such as phospholipases and protein kinase C, which interact with cell membranes during their activation. Occupation of CCK receptors on pancreatic acinar cells is linked to phospholipase C activation and subsequent production of inositol trisphosphate (IP3), which initiates the Ca2⫹ signal, and diacylglycerol (DAG). The membrane modifications could reasonably involve an alteration in the phosphoinositide turnover and a change in the supply of inositol lipid precursors of IP3. Increased production of IP3 in acini from rats fed with olive oil may explain the enhancement of intracellular Ca2⫹ mobilization in response to CCK-8, because the initial increase in Ca2⫹ transients is due mainly to Ca2⫹ released from IP3-sensitive internal stores.
Alternatively, it is tempting to speculate that DAG, abundantly generated by phospholipase C and possibly with different acyl moieties as a consequence of changes in the membrane may have resulted in our study in differential activation of protein kinase C, a crucial modulator of the secretory machinery of acinar cells. This is strongly supported by the finding in guinea-pig epidermis that DAG with a 18:2 n-6 metabolite at the 2-position inhibited protein kinase C isozymes compared with 1,2-dioleylglycerol (Cho and Ziboh, 1994). In conclusion, the type of dietary fat influences significantly the fatty acid profile of rat pancreatic membranes and this fact is associated with modulation of pancreatic cell function as assessed by intracellular Ca2⫹ mobilization and amylase release in response to CCK-8.
130.5 AR42J STUDIES Recent research in nutritional science is trying to elucidate the effect of dietary lipids on membrane composition and function. Cell culture appears to be a good approach for investigating the molecular aspects of this problem. A much wider array of modifications is possible in cultured cells than in intact animals, and the environmental conditions can be controlled better. Another valuable advantage, in terms of budget and time economy, is that animals require long periods of adaptation to dietary fats, whereas adaptation of cultured cells is achieved in a few days. The AR42J cell line is the only currently available cell line that maintains many characteristics of normal pancreatic acinar cells, such as the synthesis and secretion of
1191
CHAPTER | 130 Effects of Olive Oil on Fatty Acid Composition of Pancreatic Cell Membranes
CCK-8 10 nM
1000
1000
800
800
600 400 200
400
0 0
250
500
750
Time (s)
A
0
250
750
CCK-8 20 pM
1000
750
500 Time (s)
A
CCK-8 20 pM
1000
750 [Ca2+]c (nM)
[Ca2+]c (nM)
600
200
0
500
250
500
250
0
0 0
B
CCK-8 10 nM
1200
[Ca2+]c (nM)
[Ca2+]c (nM)
1200
250
500 750 Time (s)
1000
1250
0 B
250
500 750 Time (s)
1000
1250
FIGURE 130.5 Ca2⫹ mobilization in response to cholecystokininoctapeptide (CCK-8) in fura-2-loaded single pancreatic acinar cells of rats fed for 8 weeks with diets containing virgin olive oil as the fat source. Original chart recordings showing CCK-8-evoked changes in cytosolic Ca2⫹ concentration ([Ca2⫹]c) in fura-2-loaded single cells, studied by fluorescence microscopy. Cells were perfused (black line) with 10 nM CCK-8 (A) or 20 pM CCK-8 (B). Traces are representative of 120 acinar cells taken from 6–10 rats. Reprinted from Martínez et al., Nutrition 2004; 20: 536–541, with permission of Elsevier Ltd.
FIGURE 130.6 Ca2⫹ mobilization in response to cholecystokininoctapeptide (CCK-8) in fura-2-loaded single pancreatic acinar cells of rats fed for 8 weeks with a standard chow. Original chart recordings showing CCK-8-evoked changes in cytosolic Ca2⫹ concentration ([Ca2⫹]c) in fura-2-loaded single cells, studied by fluorescence microscopy. Cells were perfused (black line) with 10 nM CCK-8 (A) or 20 pM CCK-8 (B). Traces are representative of 100 acinar cells taken from 5–9 rats. Reprinted from Martínez et al., Nutrition 2004; 20: 536–541, with permission of Elsevier Ltd.
digestive enzymes. AR42J cells show receptor expression and signal transduction mechanisms parallel to those of normal pancreatic acinar cells, and they are being widely used to study secretion, cell signaling, cytoskeleton function, apoptosis, and pancreatitis responses of the exocrine pancreas (Ikeda and Fukuoka, 2003; Yu et al., 2005). Thus, we thought that this cell line could be a suitable in vitro model for our purpose of examining the molecular mechanisms of the effect of dietary lipids on membrane composition and cell function. Surprisingly, we found that, in spite of the wide usage in recent years, no information was available on the baseline lipid composition of AR42J cell membranes, and there were no attempts made in recorded scientific literature to modify the fatty acid composition of AR42J cells in culture. Our aim was then to determine the membrane fatty acid composition of AR42J cells, to investigate whether these cells adapt their membranes after exposure to different
fatty acids in the culture medium, to confirm if this process is similar to the adaptation of the rat exocrine pancreas that occurs when dietary fat intake is modified (Yago et al., 2006) and if this adaptation is accompanied by a similar modulation of the secretory activity and Ca2⫹ signaling. In this study (Audi et al., 2007), the addition of 18:1 n-9 (oleic-acid group) or 18:2 n-6 (linoleic-acid group) to the culture medium for 72 h profoundly influenced the fatty acid composition of AR42J cell membranes (Table 130.4). The pattern and direction of changes was parallel to that found in rats fed virgin olive oil or sunflower oil. Both virgin olive oil in rats and 18:1 n-9 in cells evoked a significant increase in membrane MUFA (due to 18:1 n-9) at the expense of SFA and PUFA and we also observed that both sunflower oil in rats and 18:2 n-6 in AR42J cells produced significant increases in total and n-6 PUFA at the expense of SFA and MUFA. Variations in the other fatty acid indices, including the SFA/UFA ratio, followed the
1192
SECTION | II
15
AR42J-O
AR42J-L
14:0
3.40 ⫾ 0.20a
4.58 ⫾ 0.08b
4.43 ⫾ 0.09b
16:0
28.05 ⫾ 0.21a
27.39 ⫾ 0.10a
28.14 ⫾ 0.38a
3.47 ⫾ 0.20a
3.85 ⫾ 0.19a
3.70 ⫾ 0.16a
18:0
16.41 ⫾ 0.26a
12.35 ⫾ 0.33b
13.04 ⫾ 0.30b
18:1 n-9
25.69 ⫾ 0.41a
32.40 ⫾ 1.11b
15.95 ⫾ 0.22c
18:2 n-6
3.33 ⫾ 0.12a
2.39 ⫾ 0.03a
17.03 ⫾ 0.21b
18:3 n-3
3.83 ⫾ 0.20a
5.11 ⫾ 0.36b
3.78 ⫾ 0.24a
20:4 n-6
3.80 ⫾ 0.11a
2.01 ⫾ 0.11b
1.86 ⫾ 0.06b
20:5 n-3
0.58 ⫾ 0.02a
0.36 ⫾ 0.02b
0.39 ⫾ 0.06b
22:6 n-3
1.87 ⫾ 0.10a
1.08 ⫾ 0.06b
0.88 ⫾ 0.07b
16:1 n-7
Net amylase release (% total content)
TABLE 130.4 Selected fatty acids in membranes from differentiated AR42J cells cultured for 72 h in medium containing unmodified fetal calf serum (AR42J-C), serum enriched in oleic acid (AR42J-O) or serum enriched in linoleic acid (AR42J-L). Fatty acid AR42J-C
Membrane fatty acid modifications in AR42J cells were evoked during the 72-h differentiation period towards an exocrine phenotype (100 nM dexamethasone) by addition of 18:1 n-9 (AR42J-O cells), 18:2 n-6 (AR42J-L cells) or vehicle (AR42J-C cells). Results are expressed as percentage of total fatty acid content. Values are means ⫾ SEM (AR42J-C: n ⫽ 18, from six batches of cells; AR42J-O: n ⫽ 15, from five batches; AR42J-L: n ⫽ 15, from five batches). For a particular row, values with different superscript letters are significantly different at p ⬍ 0.05. Reprinted from Audi et al., Exp Biol Med 2007; 232: 532–541, with permission of the Society for Experimental Biology and Medicine.
same trend after feeding oils in vivo or growing the cells with the respective major fatty acid. Simultaneously, we analyzed in AR42J cells if, like in pancreatic acini after in vivo feeding, the changes in membrane profile were accompanied by a similar modulation of the secretory activity and Ca2⫹ signaling. This investigation showed (unpublished data) that, at any particular concentration of secretagogue, net amylase secretion in the linoleic-acid group of AR42J cells was lower than in the oleic-acid group in response to CCK-8 (Figure 130.7). These results are similar to those described earlier in pancreatic acinar cells from rats adapted to diets containing either olive oil or sunflower oil. In relation to Ca2⫹ signaling, the Ca2⫹ response expressed as peak increases above basal after the addition of 1 nM CCK-8 was similar in both groups (Figure 130.8). However if the [Ca2⫹] change was expressed as an integrated response (Figure 130.9), the values obtained in the linoleic-acid group of AR42J cells were significantly higher than in the oleic-acid group.
Major Organ Systems Including Liver and Metabolism
12
9 * * 6
3 AR42J-O
AR42J-L
0 10−11
10−10 10−9 10−8 CCK-8 (M)
10−7
FIGURE 130.7 Net amylase release stimulated by cholecystokininoctapeptide (CCK-8) in AR42J cells cultured in medium enriched with different fatty acids (oleic acid: AR42J-O; linoleic acid: AR42J-L). Fetal calf serum enriched with the corresponding fatty acid was added to the basal medium during the 72-h differentiation period. Amylase released during the incubation (50 min) of cells with CCK-8 is expressed as a percentage of the initial total cell content. AR42J cells exposed to the incubation medium alone served as unstimulated controls (basal release). The effect of CCK-8 is shown as net amylase release, i.e. increase above basal. Results are means ⫾ SEM. For a given concentration of CCK-8, n ⫽ 24 observations obtained from three different batches of cells. * p ⬍ 0.05 between the two groups at specific CCK-8 concentrations.
In conclusion, the enrichment of culture medium with oleic acid (18:1 n-9) or linoleic acid (18:2 n-6) modifies the fatty acid spectrum of AR42J cell membranes in a manner that resembles the pattern found in pancreatic membranes of rats fed diets rich in olive oil or sunflower oil, respectively. The changes in AR42J membrane lipid profile induced by enrichment of the culture medium with specific fatty acids are associated with a modulation of the secretory activity that involves Ca2⫹ signaling. Therefore, this rat pancreatoma cell line adapted for 72 h to different fatty acids abundant in dietary fats (oleic acid in olive oil and linoleic acid in sunflower oil) constitutes a suitable model to conduct physiological and pathophysiological studies aiming to investigate the modulatory influence of membrane compositional changes on acinar cell function.
SUMMARY POINTS ●
Chronic (8 weeks) intake of diets containing olive oil (10 wt%) as fat source diminishes the secretion of fluid and amylase in resting conditions compared with diets containing sunflower oil. According to previous studies of our group in other species, this effect could be related to the ability of olive oil to elevate the resting blood levels of certain inhibitory gastrointestinal hormones such as peptide YY.
1193
CHAPTER | 130 Effects of Olive Oil on Fatty Acid Composition of Pancreatic Cell Membranes
Fura-2 fluorescence ratio (350 nm/385 nm)
2.5
AR42J-O
AR42J-L
2
1.5
1
1 nM CCK-8
0.5
0 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Time (min) FIGURE 130.8 Time-course changes in cytosolic Ca2⫹ concentration ([Ca2⫹]c) evoked by 1 nM cholecystokinin-octapeptide (CCK-8) in fura-2-loaded single AR42J cells previously cultured in medium enriched with different fatty acids (oleic acid: AR42J-O; linoleic acid: AR42J-L). Fetal calf serum enriched with the corresponding fatty acid was added to the basal medium during the 72-h differentiation period. The arrow indicates the start of the CCK-8 perfusion. Results are expressed as 350 nm/385 nm fura-2 fluorescence ratio. Values are means of 44 (AR42J-O) and 42 (AR42J-L) individual cells from at least three different batches.
Integrated Ca2+ response (nM)
50
AR42J-O
AR42J-L
40
30
20
a
b
10
0 2⫹
FIGURE 130.9 Integrated Ca response stimulated by 1 nM cholecystokinin-octapeptide (CCK-8) in fura-2-loaded single AR42J cells previously cultured in medium enriched with different fatty acids (oleic acid: AR42J-O; linoleic acid: AR42J-L). For the calculation of the integrated response in each individual Ca2⫹ experiment, basal Ca2⫹ concentration values before the addition of CCK-8 were averaged to estimate the baseline, and this was subtracted from each of the stimulation values. The numbers obtained in this way were then summed. The same number of data points was used for all experiments. Values are means of 44 (AR42J-O) and 42 (AR42J-L) individual cells from at least three different batches. Different letters indicate significant differences (p ⬍ 0.05) between the groups.
●
Adaptation to olive oil or sunflower oil diets also influences the time-course changes of all pancreatic secretory parameters in the anesthetized rat preparation. The differences between dietary groups can not be accounted for by changes in the gland enzyme content. However, the finding that pancreatic cell membranes enrich in those fatty acids most abundant in the fat ingested suggests that in vivo data may reflect a direct modulatory effect of the type of dietary fat on the secretory activity at the cellular level.
●
Indeed, in subsequent studies with viable pancreatic acinar cells of rats kept on identical dietary protocol (i.e. 8 weeks with diets containing 10 wt% of either virgin olive oil or sunflower oil), we could confirm that dietary-induced changes in the fatty acid profile of pancreatic membranes is associated with modulation of pancreatic cell function as assessed by intracellular Ca2⫹ mobilization and amylase release in response to CCK-8. The effects on the Ca2⫹ response were noted not only in the magnitude of large Ca2⫹ transients
1194
●
SECTION | II
evoked by high concentrations of CCK-8 but also in the amplitude and frequency of Ca2⫹ oscillations induced by low (physiological) concentrations of the secretagogue. Our recent work with the AR42J pancreatoma cell line shows that supplementation of culture media with 18:1 n-9 or 18:2 n-6 modifies in 72 h the membrane fatty acid spectrum of these cells, with the pattern and direction of changes being parallel to those found in rats fed diets enriched in virgin olive oil or sunflower oil. In addition, these changes are accompanied by a modulation of the secretory activity that involves Ca2⫹ signaling. Therefore, AR42J cells seem to be a suitable model in physiological and pathophysiological studies aiming to assess the effect of membrane compositional changes on acinar cell function.
REFERENCES Abedin, L., Lien, E.L., Vingrys, A.J., Sinclair, A.J., 1999. The effects of dietary alpha-linolenic acid compared with docosahexaenoic acid on brain, retina, liver, and heart in the guinea pig. Lipids 34, 475–482. Audi, N., Mesa, M.D., Martínez, M.A., Martínez-Victoria, E., Mañas, M., Yago, M.D., 2007. Membrane lipid composition of pancreatic AR42J cells: modification by exposure to different fatty acids. Exp. Biol. Med. 232, 532–541. Begin, M.E., Ells, G., St. Jean, P., Vachereau, A., Beaudoin, A.R., 1990. Fatty acid and enzymatic compositional changes in the pancreas of rats fed dietary n-3 and n-6 polyunsaturated fatty acids. Int. J. Pancreatol. 6, 151–160. Cho, Y., Ziboh, V.A., 1994. Expression of protein kinase C isozymes in guinea pig epidermis: selective inhibition of PKC-beta activity by 13hydroxyoctadecadienoic acid-containing diacylglycerol. J. Lipid. Res. 35, 913–921. Clandinin, M.T., Cheema, S., Field, C.J., Garg, M.L., Venkatraman, J., Clandinin, T.R., 1991. Dietary fat: exogenous determination of membrane structure and cell function. FASEB J. 5, 2761–2769. Díaz, R.J., Yago, M.D., Martínez-Victoria, E., Naranjo, J.A., Martínez, M.A., Mañas, M., 2003. Comparison of the effects of dietary sunflower oil and virgin olive oil on rat exocrine pancreatic secretion in vivo. Lipids 38, 1119–1126. Ikeda, Y., Fukuoka, S., 2003. Phosphatidic acid production, required for cholecystokinin octapeptide-stimulated amylase secretion from pancreatic acinar AR42J cells, is regulated by a wortmannin-sensitive process. Biochem. Biophys. Res. Commun. 306, 943–947.
Major Organ Systems Including Liver and Metabolism
Keys, A., 1970. Coronary heart disease in seven countries. Circulation 41, 1–211. Lajas, A.I., Pozo, M.J., Salido, G.M., Pariente, J.A., 1998. Effect of basic fibroblast growth factor on cholecystokinin-induced amylase release and intracellular calcium increase in male rat pancreatic acinar cells. Biochem. Pharmacol. 55, 903–908. Martínez, M.A., Lajas, A.I., Yago, M.D., Redondo, P.C., Granados, M.P., González, A., Rosado, J.A., Martínez-Victoria, E., Mañas, M., Pariente, J.A., 2004. Dietary virgin olive oil enhances secretagogueevoked signalling in rat pancreatic acinar cells. Nutrition 20, 536–541. Quiles, J.L., Huertas, J.R., Mañas, M., Battino, M., Mataix, J., 1999. Physical exercise affects the lipid profile of mitochondrial membranes in rats fed with virgin oil or sunflower oil. Br. J. Nutr. 81, 21–24. Quiles, J.L., Huertas, J.R., Mañas, M., Ochoa, J.J., Battino, M., Mataix, J., 2001. Dietary fat type and regular exercise affect mitochondrial composition and function depending on specific tissue in the rat. J. Bioenerg. Biomembr. 33, 127–134. Soriguer, F.J., Tinahones, F.J., Monzàn, A., Pareja, A., Rojo-Martínez, G., Moreno, F., Esteva, I., Gómez-Zumaquero, J.M., 2000. Varying incorporation of fatty acids into phospholipids from muscle, adipose and pancreatic exocrine tissues and thymocytes in adult rats fed with diets rich in different fatty acids. Eur. J. Epidemiol. 16, 585–594. Stenson, W.F., Seetharam, B., Talkad, V., Pickett, W., Dudeja, P., Brasitus, T.A., 1989. Effects of dietary fish oil supplementation on membrane fluidity and enzyme activity in rat small intestine. Biochem. J. 263, 41–45. Yago, M.D., Díaz, R.J., Martínez, M.A., Audi, N., Naranjo, J.A., MartínezVictoria, E., Mañas, M., 2006. Effects of the type of dietary fat on acetylcholine-evoked amylase secretion and calcium mobilization in isolated rat pancreatic acinar cells. J. Nutr. Biochem. 17, 242–249. Yago, M.D., Díaz, R.J., Ramirez, R., Martínez, M.A., Mañas, M., Martínez-Victoria, E., 2004. Dietary-induced changes in the fatty acid profile of rat pancreatic membranes are associated with modifications in acinar cell function and signalling. Br. J. Nutr. 91, 227–234. Yago, M.D., González, M.V., Martínez-Victoria, E., Mataix, J., Medrano, J., Calpena, R., Pérez, M.T., Mañas, M., 1997a. Pancreatic enzyme secretion in response to test meals differing in the quality of dietary fat (olive oil and sunflowerseed oils) in human subjects. Br. J. Nutr. 78, 27–39. Yago, M.D., Martínez-Victoria, E., Mañas, M., Martínez, M.A., Mataix, J., 1997b. Plasma peptide YY and pancreatic polypeptide in dogs after long-term adaptation to dietary fats of different degrees of saturation: olive and sunflower oil. J. Nutr. Biochem. 8, 502–507. Yago, M.D., Martínez-Victoria, E., Huertas, J.R., Mañas, M., 1997c. Effects of the amount and type of dietary fat on exocrine pancreatic secretion in dogs after different periods of adaptation. Arch. Physiol. Biochem. 105, 78–85. Yu, J.H., Lim, J.W., Kim, H., Kim, K.H., 2005. NADPH oxidase mediates interleukin-6 expression in cerulein-stimulated pancreatic acinar cells. Int. J. Biochem. Cell. Biol. 37, 1458–1469.
Chapter 131
Olives and Olive Oil in the Prevention of Osteoporosis Véronique Coxam, Caroline Puel and Marie-Jeanne Davicco Unité de Nutrition humaine, UMR1019 (INRA/Université), INRA Theix, Saint Genès, Champanelle, France
131.1 INTRODUCTION The increase in life expectancy since 1850 forms a straight line, and there are no indications that this phenomenon is levelling off. Indeed, since the industrial revolution in the middle of the 19th century, average female life expectancy has increased in Western societies from about 45 years to ⬎80 years currently, which corresponds to an increase of 2.3 years per decade. It has also risen for men, although more slowly. Consequently, in the coming years, addressing the cumulative damage associated with aging will be one of the biggest challenges faced by industrialized countries. The expectation is thus that scientific advances will prevent disease from occurring or, if disease does strike, will protect us from permanent damage. In this light, because bone mass declines progressively with advancing age, under current conditions the projected increase in longevity will be accompanied inevitably by an increase in the prevalence of osteoporosis and its associated complications. This is why osteoporosis, internationally described as ‘a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue with a consequent increase in bone fragility and susceptibility to fracture’, has become a major cause of morbidity, disability and a contributor to medical care costs in many regions of the world. Indeed, 75 million people in the United States, Europe, and Japan combined are affected, including 1 in 3 postmenopausal women and most of the elderly, as well as a substantial number of men. As a matter of fact, based on incidence rates in North America approaches, the estimated remaining lifetime risk for fragility fractures of any type (hip, spine and forearm) in Caucasian women at age 50 years is 40%, and about 13% in men (Cummings et al., 1985). Hip fractures are associated with considerable morbidity, and even lead to an overall mortality of 15–30%. Vertebral fractures also Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
contribute to reduced survival, probably due to comorbidity/clustering. Unfortunately, the prognosis suggests that, unless suitable preventive action can be found, the number of hip fractures occurring in the world each year will rise from 1.7 million in 1990 to 6.3 million by 2050. It is true that currently available curative treatments can slow down or prevent bone loss, but it is uncertain whether the biochemical competence of the skeleton can be restored once increased fragility has occurred, because bone architecture cannot be restored to the premorbid state. Consequently, as bone mineral status in later life is the net outcome of lifelong influences on skeletal mineral accretion and loss, there is now an emerging rationale for early intervention, which may actually be best achieved by initiating sound health behavior early in life and continuing them indefinitely. This is why an open issue in aging research is the extent to which responses to the environment during development can influence variability in the life span and the health profile of elderly populations. Research in nutrition over the past 30 years has led to exciting progress supporting the hypothesis that by modulating specific target functions in the body, diet can help to achieve optimal health by reducing the risk of disease. Specifically, it has been recognized that the human diet contains, in addition to essential macronutrients, a complex array of naturally occurring bioactive molecules. Consequently prevention through dietary means is especially challenging in technologically advanced societies. So far functional foods designed to prevent osteoporosis have focused mainly on calcium because it is the most likely among the bone-building nutrients to be inadequate in terms of dietary intake. However, a large number of nutrients can be considered as possible determinants of bone health. This is why there is an increasing move to focus on primary prevention, with the goal of finding other agents in addition to calcium and vitamin D that (1) can optimize
1195
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1196
SECTION | II
peak bone mass, thereby maximizing the buffer available before bone demineralization compromises structural integrity, (2) can slow the natural rate of bone demineralization, so that existing structural integrity is conserved as much as possible, and (3) can minimize those factors responsible for accelerated demineralization.
131.2 THE RATIONALE TO FOCUS ON THE MEDITERRANEAN DIET There is an increasing rationale to focus on the Mediterranean diet to find a way of improving the prognosis in osteoporosis, due to the observation that adults who live near the Mediterranean Sea have one of the lowest incidences in various chronic diseases in the world and one of the highest life expectancies, in spite of being an alimentary model with a high fat content, in contrast to the diets recommended for many decades by nutritionists (Keys et al., 1986). The traditional Mediterranean diet in the South of Europe is characterized by moderate energy intake, low animal fat, high olive oil, high cereals, high legumes, nuts and vegetables and regular but moderate wine consumption. Olive oil contributes almost 20% of the total energy intake in Mediterranean menus. There has been a remarkable increment in scientific knowledge dealing with the beneficial role of olive-derived products. Basically, olive oil is unique with respect to its high oleic acid content because the majority of seed oils are composed
Major Organ Systems Including Liver and Metabolism
primarily of polyunsaturated fatty acids, including the essential omega-6 fatty acid, linoleic acid. In fact, only virgin oil is a real juice and its composition brings not only fat, but also an elevated intake of other nutritional components responsible for its unique nutritional value. These are not present in any other edible oil because they have to be obtained through complex physical and chemical treatments that result in partial or total loss of many micronutrients. In other words, unlike olive oil, oils derived from seeds need to be refined for human consumption, meaning they lose the vast majority of their original microcomponents, leaving them almost exclusively mere sources of fat (Perez-Jimenez et al., 2007).
131.3 NUTRITIONAL VALUE OF OLIVES AND OLIVE OIL (Table 131.1) The healthful properties of olive oil have been often attributed to its high content of monounsaturated fatty acids, namely in the form of oleic acid (18:1 n-9). Indeed olive oil consists of approximately 72% oleic acid, making it much less susceptible to oxidation and contributing to the antioxidant action. In addition, it should be underlined that olive oil’s beneficial properties lie in its high levels of several micronutrient constituents, including a number of polyphenolic compounds (100–1000 mg kg⫺1), vitamin E, carotenes, squalene, and chlorophyll (Owen et al., 2000), that exert an antioxidant effect in unison with oleic acid to provide protection from oxidative
TABLE 131.1 The concentration of total and individual phenolic compounds in olive oil (Owen et al., 2000). Phenolic compound (mg kg⫺1)
Olive oil All (n ⫽ 23)
VOQ (n ⫽ 18)
RVO (n ⫽ 5)
P value*
196 ⫾ 19
232 ⫾ 15
62 ⫾ 12
⬍0.0001
Hydroxytyrosol
11.66 ⫾ 2.60
14.42 ⫾ 3.01
1.74 ⫾ 0.84
⬍0.05
Tyrosol
22.13 ⫾ 3.82
27.45 ⫾ 4.05
2.98 ⫾ 1.33
⬍0.01
Total simple phenols (TSP) 33.79 ⫾ 4.48
41.87 ⫾ 6.17
4.72 ⫾ 2.15
⬍0.01
Secoiridoid-1
7.97 ⫾ 2.57
9.62 ⫾ 3.18
2.00 ⫾ 0.87
Ns
Secoiridoid-2
15.75 ⫾ 3.54
18.09 ⫾ 4.31
7.30 ⫾ 3.01
Ns
Total secoiridoids (SID)
23.71 ⫾ 5.61
27.72 ⫾ 6.84
9.30 ⫾ 3.81
Ns
Lignans
34.09 ⫾ 4.42
41.53 ⫾ 3.93
7.29 ⫾ 2.56
⬍0.001
91.59 ⫾ 10.57
111.12 ⫾ 9.99
21.31 ⫾ 8.03
⬍0.001
Total
TSP ⫹ SID ⫹ lignans ⫺1
Data expressed in mg kg ⫾ SEM: VOQ: extra virgin oil; RVO: refined virgin oil. * VOQ vs. RVO; Ns ⫽ not significant.
1197
CHAPTER | 131 Olives and Olive Oil in the Prevention of Osteoporosis
stress, which is believed to be the key mechanism behind many diseases (Brenes et al., 1999; Owen et al., 2000; Visioli et al., 2000). The phenolic composition of olive oils is the result of a complex interaction between several factors including cultivar, soil, ripeness of the olives at harvesting, degree of maturation and climate, as well as crushing machine conditions during malaxation, storage, and the processing system employed to produce the types of olive oil currently present on the market: extra virgin, virgin, olive oil, or pomace. The phenolic fraction provides the typical pungent taste and aroma of extra virgin olive oil, which is the kind of oil that is obtained by simple physical separation of the oil from the olive paste. Polyphenolic compounds are important to plant physiology, contributing to resistance to microorganisms and insects, pigmentation and organoleptic characteristics. They display relevant antioxidant properties preserving plant integrity against environmental stress (Visioli et al., 2000). The majority of the phenolic compounds identified and quantified in olive oil belong to three groups; simple phenols (tyrosol and hydroxytyrosol), secoiridoids (oleuropein, the aglycone ligstroside and their respective decarboxylated dialdehyde derivatives) and the lignans ((⫹)-1-acetoxypinoresinol and pinoresinol) (Owen et al., 2000). The content of total phenols in extra virgin olive oil, expressed as gallic acid equivalents, ranges from 14.80 to 121.20 mg 100 g⫺1 with a mean value of 53.72 mg 100 g⫺1. Free forms of tyrosol and hydroxytyrosol and their secoiridoid derivatives represent around 30%, and other conjugated forms such as oleuropein and ligstroside aglycones, almost half of the total phenolic content of a virgin olive oil (Brenes et al., 1999). Black olive pericarp extract has a higher concentration of phenolic compounds and a higher antioxidant capacity than that of green olives (Brenes et al., 1999; Owen et al., 2000). In addition, olive oil is composed of approximately 0.7% squalene, while other foods and oils typically have
squalene levels in the range of 0.002–0.03% (Newmark, 1997). Squalene is a major intermediate in the biosynthesis of cholesterol and due to its structure, it is more likely to scavenge singlet oxygen species than hydroxyl radicals.
131.4 OLIVES, OLIVE OIL AND BONE HEALTH (Figure 131.1, Table 131.2) A higher consumption of olive oil is regarded as the hallmark of the traditional Mediterranean diet, so, because olive oil is a source of at least 30 phenolic compounds, there has been a surge in the number of publications that have investigated their biological properties. However, despite the myriad of potential health benefits of olive oil polyphenolics, there are only very few data relating their possible preventive effect on osteoporosis (Puel et al., 2007a, b).
131.4.1 Olive Consumption and Bone Health Puel et al. (2007a, b) evaluated the possible bone-sparing effect of table olive consumption on bone metabolism in an experimental model for senile osteoporosis. Whereas green olives did not elicit any bone-sparing effect, consumption of a black variety prevented bone loss in the whole femur and at the cortical site, as shown by a higher bone mineral density. This could be accounted for by a potent effect on the diaphyseal subregion, while no change was exhibited on the metaphysis. They speculated that this prevention might result from a reduced inflammatory state, as shown by lower plasma fibrinogen and α1-acid glycoprotein levels, even though splenomegaly and the rise of granulocyte levels were not modulated. In addition, an improvement was observed in the oxidative stress, as measured by isoprostane excretion. The inefficiency of green olives could be explained by the lack of hydroxytyrosol and to the lower amount of tyrosol, and a higher salt content (0.832 g)
Total femoral mineral density (%)
110 bc 108
b
b
bc
106
c
104 102
a
a
100 98 96
Control
Green olives
Black olives
Olive oil
OMWW Oleuropein OHtyrosol
Tyrosol
FIGURE 131.1 Effect of a daily intake for 3 months of olives, olive oil, olive mill wastewater (OMWW), or olive phenolic compounds (oleuropein, hydroxytyrosol and tyrosol) on bone mineral density of ovariectomized rats with inflammation. Green Lucques olives (50% in the diet), black Lucques olives (30%), extra virgin oil (5%), OMWW (0.17%), oleuropein (0.015%), hydroxytyrosol (0.017%) or tyrosol (0.017%). Values not sharing a superscript letter differ significantly (p ⬍ 0.05).
1198
SECTION | II
TABLE 131.2 Benefits of table olive and olive oilrich diet on mechanisms related to bone loss in an experimental model of senile osteoporosis (Puel et al., 2004, 2007). Level of evidence
Type of effect
Demonstrated
Prevented inflammatory state (granulocyte level, acute inflammatory protein) Prevented oxidative stress (urinary isoprostane excretion, ROS production)
Possible
Improved signaling events implicated in osteoclast differentiation, activation and apoptosis
in green olives compared to the amounts for black olives (0.480 g). A high NaCl intake could induce acidosis (blood pH ⬍ 7.4) which enhances osteoclastic resorption and then bone loss. Ostrowska et al. (2006) have shown that growing pigs fed extra virgin olive oil have a high daily increase in bone mineral density. As a matter of fact, MUFA-rich diets along with the phenolic properties of olive oil may have protective effects on BMD and therefore osteoporosis. Indeed, in Greek men and women aged between 25 and 69 years of age, a positive association between the MUFA content of the diet (mostly derived from olive oil in this population) and bone mineral density has been also reported, even after adjustment for non-nutritional variables and energy intake (Trichopoulou et al., 1997).
131.4.2 Olive Oil Consumption and Bone Health Puel et al. (2004) evaluated the effect of olive oil in ovariectomized rats with or without inflammation. They observed a protective effect with a diet providing olive oil (at a dose equivalent in magnitude to current consumption in humans) in inflammatory conditions, as shown by increased femoral failure load and diaphyseal bone mineral density. However, the antioxidant capacity of plasma was not improved.
131.4.3 Polyphenols from Olive Oil and Bone Health Puel et al. (2006) assessed the dose-dependent bone-sparing effect of oleuropein and other olive oil phenolic compounds
Major Organ Systems Including Liver and Metabolism
(tyrosol and hydroxytyrosol) on bone loss induced by talc granulomatosis in estrogen-deficient rat. They demonstrated that the four doses of oleuropein they investigated (2.5, 5, 10 or 15 mg kg⫺1 body weight per day for 100 days) reduced bone loss and improved inflammatory biomarkers, except for the 5 mg kg⫺1 dose. A modulation of inflammatory parameters was observed. As far as tyrosol and hydroxytyrosol are concerned, they were able at a dose of 0.017% in the diet to prevent bone loss in the same model for senile osteoporosis, probably by improving osteoblastic activity as shown by osteoclacin plasma levels. This bone-sparing effect did not seem to be ascribable to an improvement of the inflammatory state, given that the increase of inflammatory parameters tested in this investigation was not prevented, but rather to their antioxidant properties (personal data). The same amounts of olive wastewater were devoid of any effect. To summarize, those data suggest the possible relevance of dietary intake of olives and olive oil in lowering the risk of inflammation-induced osteopenia, based on the capabilities of the major polyphenols. Moreover, there is a growing scientific rationale for the use of olive oil as an adjunct in the treatment of other inflammatory disorders such as rheumatoid arthritis and osteoarthritis as well (Darlington and Stone, 2001). The question remains, however, as to what the basic mechanisms involved in such a health effect might be.
131.5 BIOLOGICAL EFFECTS OF OLIVES AND OLIVE OIL COMPONENTS (Figure 131.2)
131.5.1 Anti-inflammatory Properties Olive oil has been reported to be a molecular modulator of the inflammatory/immune response. Indeed in an animal model, an 8-week regimen of olive oil consumption elicited a decrease in TNF-α production by rat peritoneal macrophages (Tappia and Grimble, 1994). Similar results have been published by Wallace et al. (2000) who investigated the effects of dietary fats on macrophage-mediated cytotoxicity. Moreover, an in vitro study provided evidence of an increase in NO (an important bactericidal and cytostatic factor with vaso-relaxing and anti-aggregating properties) and also documented a decrease in arachidonic acid mobilization and prostaglandin E2 production in macrophages from animals fed olive oil (Moreno et al., 2001). These data could in fact be explained by the fact that the phenol fraction of olive oil has been found to modulate inflammation. Indeed, oleuropein has been shown to increase NO production in macrophages challenged with lipopolysaccharide, through stimulation of the inducible form of enzyme nitric oxide synthase, thus increasing the functional activity
1199
CHAPTER | 131 Olives and Olive Oil in the Prevention of Osteoporosis
Macrophages Activated macrophages
NF-κB
Inflammation
INFγ
Lymphocytes
NF-κB
NF-κB
Oxidative stress
COX2, LOX2 Eicosanoids (PGE2, LKT4) Superoxide anion O2− Hydrogen peroxide HOOH Hydroxyl radical OH− Peroxynitrite radical NOO− NO
NO
iNOs
ROS
Osteoblasts Osteoblast recruitment Osteoblastic activity
Cytokines (IL-1, IL-6, TNF-α)
RANKl/OPG
NF-κB
Osteoclasts NF-κB
Bone resorption
FIGURE 131.2 Schematic model of bone loss. During the remodeling cycle, bone resorption is coupled to bone formation, and this interaction between the osteoclasts and osteoblasts is coordinated primarily through the receptor activator of nuclear factor-κB (RANK)/RANK ligand (RANKL)/osteoprotegerin (OPG) system. Multiple cytokines [e.g. interleukin-1 (IL-1), IL-6, tumor necrosis factor-α (TNF-α)] and oxidative stress not only regulate the coupling between these two systems but also the differentiation and activity of osteoclasts and osteoblasts. A systemic or local imbalance in the factors regulating these systems may result in bone loss (modified according to Seifert and Watkins, 1997).
of these immunocompetent cells (Visioli et al., 1998). Moreover, Chun et al. (2008) have provided evidence that serum C-reactive protein (CRP), a biomarker for chronic inflammation, is inversely related to dietary flavonoid intake in adults in the USA, supporting the idea that consumption of flavonoid-rich foods may reduce inflammation-mediated chronic diseases.
131.5.2 Anti-oxidant Properties (Table 131.3) Among the theories of aging, free radicals produced by the mitochondria are responsible for the damage that affects all biological tissues, leading to the aging phenotype and the manifestation of several degenerative diseases. High olive oil intake has been related to lower mitochondrial oxidative stress. Indeed, over the past 15 years, we have accumulated a good deal of evidence on the effectiveness of dietary virgin olive oil, which appears to be a key component of successful dietary manipulations aimed at strengthening membranes by partially modifying their structure and consequently their features in the daily struggle against oxidative stress-induced damage. Thus olive oil leads to lower levels of polyunsaturated biological membranes, the degree of fatty acid unsaturation of mammalian tissues being negatively correlated with greater longevity (Perez-Jimenez et al., 2007).
These data could be explained by the phenolic compounds present in olive oil, which are strong antioxidants and radical scavengers (Oliveras-López et al., 2008). Indeed, clinical studies in healthy volunteers have shown that oxidative stress markers decreased linearly with the increase in the phenolic content of olive oil (Weinbrenner et al., 2004). Olive oil compounds counteract ROS-mediated cytotoxicity in human erythrocytes and inhibit passive smoking-induced oxidative stress in human volunteers (Visioli et al., 2000). Moreover, recent findings suggest that LDL oxidation is inhibited by polyphenolic compounds from olive oil, such as the complex phenol oleuropein and its derivative hydroxytyrosol. Both inhibit copper sulfate-induced oxidation of LDL (Visioli et al., 1995). They are able to scavenge free radicals due to the hydrogen-donating capacity of the hydroxyl group in the ortho-diphenolic structure. Furthermore, a recently identified secoiridoid derivative, oleocanthal, has demonstrated inhibition of cyclooxygenase enzymes and anti-inflammatory activity (Beauchamp et al., 2005). However, many other effects of polyphenols such as anti-inflammatory, antitumor, antiatherogenic abilities cannot be explained solely on the basis of their antioxidant properties. Evans et al. (2006) expanded the scope of the effect of polyphenolics from slip modulation of enzymes and receptors to gene expression.
1200
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 131.3 Compounds with antioxidant activity reported to be present in olive oil (Trichopoulou and Dilis, 2007). Chemical class
Subclass
Compound
Phenolic compounds
Phenolic acids
4-Hydroxybenzoic acid, protocatechuic acid, gallic acid, vanillic acid, syringic acid, 4hydroxyphenylacetic acid, homovanillic acid, o-coumaric acid, p-coumaric acid, caffeic acid, ferulic acid, sinapic acid
Tyrosol, hydroxytyrosol and derivatives
Tyrosol, hydroxytyrosol, oleuropein, oleuropein aglycon, dialdehyde form of oleuropein aglycon, decarboxymethyl form of oleuropein aglycon, ligstroside aglycon
Lignans
(⫹)-Pinoresinol, (⫹)-1-acetoxypinoresinol
Flavonoids
Apigenin, luteolin, quercetin
Closely related non-phenolic compounds
Elenolic acid, cinnamic acid
Hydrocarbons
Triterpenes
Squalene
Chlorophylls
Chlorophyll and derivatives
Pheophytin α, pheophytin b, chlorophyll α, chlorophyll b, pyropheophytin α
Carotenoids
Carotenes (hydrocarbons)
β-Carotene
Xanthophylls
Lutein, neoxanthin, violaxanthin, luteoxanthin, antheraxanthin, mutatoxanthin, β -cryptoxanthin
–
α, β, γ, and δ-Tocopherol
Tocopherols
131.6 APPLICATION OF BIOLOGICAL EFFECTS OF OLIVES AND OLIVE OIL COMPONENTS TO BONE PHYSIOLOGY (Figure 131.3)
131.6.1 Inflammation and Bone Status With the recent dramatic increase in life expectancy, the immune system must now cope with chronic exposure to antigens, lasting several decades more than in our recent evolutionary past. This continual antigenic stress means the immune system can become over-stimulated over time and inefficient with age, representing a process with the proposed name of ‘inflammaging’. As a result chronic, low-grade inflammation develops (Franceschi, 2007) and is involved in the pathogenesis of several age-associated diseases (Hansson, 2005), such as Alzheimer’s disease, atherosclerosis, diabetes mellitus, sarcopenia. Like other tissues, the skeleton is regulated by hormonal or local factors and the growing understanding of the bone remodeling process supports the theory that inflammation significantly contributes to the etiopathogenesis of osteoporosis. Thus, in the Health ABC study (2985 volunteers), subjects with the greatest number of inflammatory markers
had the highest risk of fracture. Results also showed that baseline markers of inflammation were higher among subjects who subsequently experienced an incident fracture (Cauley et al., 2007). In fact, the term osteoimmunology has been proposed to explain the cross-talk between bone and the immune system. Numerous pro-inflammatory cytokines have been implicated in the regulation of bone cells, and a shift towards an activated immune profile has been hypothesized as an important risk factor. A critical element is the inducible transcription factor κB (NF-κB), which regulates gene expression during inflammatory and immune responses, including many pro-inflammatory cytokines, chemokines and adhesion molecules (Jimi and Ghosh, 2005). This NF-κB signaling pathway is also important for bone homeostasis, in particular for osteoclast differentiation (the primary cells for bone resorption). The integrity of the skeletal tissues is thus altered by changes in the balance between old bone resorption by osteoclasts and new bone formation by osteoblasts. Consequently the key to successful aging of the skeleton is to decrease chronic inflammation without compromising an acute response when exposed to pathogens, i.e. without impairing the inflammatory responses towards those pathogens to which people are exposed in everyday life. In this
1201
CHAPTER | 131 Olives and Olive Oil in the Prevention of Osteoporosis
Polar fraction = Natural antioxidants Flavonoïds Phytosterols
Phenolic compounds Luteolin Lignan
Apigenin
Vitamin E
Oleuropein Hydroxytyrosol Tyrosol Menopause Estrogen deficiency
Skeleton
Calcium Oxidative and inflammatory status
Resorption
Calcium balance Intestinal absorption Tubular reabsorption
Formation
Inflammation, oxidative stress Aging Omega 3 Oleic acid Polyunsaturated fatty acids
Monounsaturated fatty acids Unsaponifiable fraction
FIGURE 131.3 Different pathways of action of olive oil compounds on osteoporosis etiology.
light, due to its anti-inflammatory properties, olive oil provides an interesting tool to prevent bone loss during aging. Again, the fact that hip fractures are substantially less frequent in an olive oil-consuming country, Greece, compared with North America and northern Europe, supports the hypothesis that olive oil could beneficially affect bone mineral density (Trichopoulou et al., 1997).
131.6.2 Oxidative Stress and Bone Status Production of oxide-derived free radicals is known to enhance bone resorption and to inhibit osteoblastic recruitment and the activity of mature cells (Mody et al., 2001). Consequently antioxidant nutrients may stimulate bone formation and reduce the production of free radicals that contribute to bone resorption (Keys et al., 1986). This is why there is an increasing rationale for general health reasons to focus on the Mediterranean diet, known for its antioxidant nutrients, as shown by the results of the EUROLIVE study providing evidence of the in vivo protective role of phenolic compounds from olive oil on lipid oxidative damage in humans, at real-life olive oil doses
(Covas et al., 2006), or the Hale cohort conduct (Knoops et al., 2004), all of which is relevant also for bone health (Fito et al., 2007).
131.7 CONCLUSION Increasing life expectancy is matched by an increase in the prevalence of a number of age-related chronic diseases, including osteoporosis and its associated complications. To prevent this disease, nutritional strategies for optimizing bone health are now being considered, since a dietary approach is more popular amongst osteoporosis sufferers than drug intervention, and long-term drug treatment compliance is relatively poor. Indeed, an increasing body of scientific evidence has demonstrated that compounds derived from food alter the expression of genes in the human body. By turning genes on or off, bioactives in food alter the concentration of specific proteins directly or indirectly associated with human diseases. This may explain why adherence to a Mediterranean-style diet affords protection from degenerative diseases. This diet provides high amounts of olive oil, which is considered as a functional
1202
SECTION | II
food, and which besides having a high level of MUFA, as oleic acid, contains multiple minor components with biological properties. There is accumulating evidence that antioxidant and anti-inflammatory micronutrients could be beneficial to bone health as well. In this light, olives or olive oil consumption remain sources for putative new and innovative dietary health intervention in the nutritional prevention of osteoporosis.
SUMMARY POINTS ●
●
●
●
A lower incidence of osteoporosis was observed in the Mediterranean area which is characterized by a high consumption of olive oil. Olive oil and table olives are a source of phenolic compounds with antioxidant and anti-inflammatory properties. Black olive and olive oil consumption can prevent bone loss by different mechanisms of action: anti-inflammatory and/or antioxidant activities. A better knowledge of signaling pathways of olive oil polyphenols and other compounds might provide a novel therapeutic approach for osteoporosis in the aging population.
REFERENCES Beauchamp, G.K., Keast, R.S., Morel, D., Lin, J., Pika, J., Han, Q., Lee, C.H., Smith, A.B., Breslin, P.A., 2005. Phytochemistry: Ibuprofenlike activity in extravirgin olive oil. Nature 437, 45–46. Brenes, M., García, A., García, P., Rios, J.J., Garrido, A., 1999. Phenolic compounds in Spanish olive oils. J. Agric. Food Chem. 47, 3535–3540. Cauley, J.A., Danielson, M.E., Boudreau, R.M., Forrest, K.Y., Zmuda, J.M., Pahor, M., Tylavsky, F.A., Cummings, S.R., Harris, T.B., Newman, A.B., 2007. Inflammatory markers and incident fracture risk in older men and women: the Health Aging and Body Composition Study. J. Bone Miner. Res. 22, 1088–1095. Covas, M.I., Nyyssönen, K., Poulsen, H.E., Kaikkonen, J., et al., 2006. The effect of polyphenols in olive oil on heart disease risk factors. Ann. Intern. Med. 145, 333–341. Cummings, S.R., Kelsey, J.L., Nevitt, M.C., O’Dowd, K.J., 1985. Epidemiology of osteoporosis and osteoporotic fractures. Epidemiol. Rev. 7, 178–208. Darlington, L.G., Stone, T.W., 2001. Antioxidants and fatty acids in the amelioration of rheumatoid arthritis and related disorders. Br. J. Nutr. 85, 251–269. Evans, D.A., Hiesch, J.B., Dushenkov, S., 2006. Phenolics, inflammation and nutrigenomics. J. Sci. Food Agric. 86, 2503–2509. Fito, M., de la Torre, R., Covas, M.I., 2007. Olive oil and oxidative stress. Mol. Nutr. Food Res. 51, 1215–1224. Franceschi, C., 2007. Inflammation as a major characteristic of old people: Can it be prevented or cured? Nutr. Rev. 65, S173–S176.
Major Organ Systems Including Liver and Metabolism
Hansson, G.K., 2005. Inflammation, atherosclerosis, and coronary artery disease. N. Engl. J. Med. 352, 1685–1695. Jimi, E., Ghosh, S., 2005. Role of nuclear factor-kappaB in the immune system and bone. Immunol. Rev. 208, 80–87. Keys, A., Menotti, A., Karvonen, M.J., Aravanis, C., Blackburn, H., Buzina, R., Djordjevic, B.S., Dontas, A.S., Fidanza, F., Keys, M.H., et al., 1986. The diet and 15-year death rate in the seven countries study. Am. J. Epidemiol. 124, 903–915. Knoops, K.T., de Groot, L.C., Kromhout, D., Perrin, A.E., Perrin, A.E., Moreiras-Varela, O., Menotti, A., van Staveren, W.A., 2004. Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women. The HALE Project. JAMA 292, 1433–1439. Chun, O.K., Chung, S.J., Claycombe, K.J., Song, W.O., 2008. Serum C-reactive protein concentrations are inversely associated with dietary flavonoid intake in US adults. J. Nutr. 138, 753–760. Mody, N., Parhami, F., Sarafian, T.A., Demer, L., 2001. Oxidative stress modulates osteoblastic differentiation of vascular and bone cells. Free Rad. Biol. Med. 31, 509–519. Moreno, J.J., Carbonell, T., Sánchez, T., Miret, S., Mitjavila, M.T., 2001. Olive oil decreases both oxidative stress and the production of arachidonic acid metabolites by the prostaglandin G/H synthase pathway in rat macrophages. J. Nutr. 131, 2145–2149. Newmark, H.L., 1997. Squalene, olive oil, and cancer risk: A review and hypothesis. Cancer Epidemiol. Biomarkers Prev. 6, 1101–1103. Oliveras-López, M.J., Berná, G., Carneiro, E.M., López-García de la Serrana, H., Martín, F., López, M.C., 2008. An extra-virgin olive oil rich in polyphenolic compounds has antioxidant effects in OF1 mice. J. Nutr. 138, 1074–1078. Ostrowska, E., Gabler, N.K., Ridley, D., Suster, D., Eagling, D.R., Dunshea, D.R., 2006. Extra-virgin and refined olive oils decrease plasma triglyceride, moderately affect lipoprotein oxidation susceptibility and increase bone density in growing pigs. J. Sci. Food Agric. 86, 1955–1963. Owen, R.W., Mier, W., Giacosa, A., Hull, W.E., Spiegelhalder, B., Bartsch, H., 2000. Phenolic compounds and squalene in olive oils: The concentration and antioxidant potential of total phenols, simple phenols, secoiridoids, lignans and squalene. Food Chem. Toxicol. 38, 647–659. Perez-Jimenez, F., Ruano, J., Perez-Martinez, P., Lopez-Segura, F., LopezMiranda, J., 2007. The influence of olive oil on human health: Not a question of fat alone. Mol. Nutr. Food Res. 51, 1199–1208. Puel, C., Coxam, V., Davicco, M.J., 2007a. Mediterranean diet and osteoporosis prevention. Med. Sci. (Paris) 23, 756–760. Puel, C., Mardon, J., Kati-Coulibaly, S., Davicco, M.J., Lebecque, P., Obled, C., Rock, E., Horcajada, M.N., Agalias, A., Skaltsounis, L.A., Coxam, V., 2007b. Black Lucques olives prevented bone loss caused by ovariectomy and talc granulomatosis in rats. Br. J. Nutr. 97, 1012–1020. Puel, C., Mathey, J., Agalias, A., Kati-Coulibaly, S., Mardon, J., Obled, C., Davicco, M.J., Lebecque, P., Horcajada, M.N., Skaltsounis, A. L., Coxam, V., 2006. Dose-response study of effect of oleuropein, an olive oil polyphenol, in an ovariectomy/inflammation experimental model of bone loss in the rat. Clin. Nutr. 25, 859–868. Puel, C., Quintin, A., Agalias, A., Mathey, J., Obled, C., Mazur, A., Davicco, M.J., Lebecque, P., Skaltsounis, A.L., Coxam, V., 2004. Olive oil and its main phenolic micronutrient (oleuropein) prevent inflammation-induced bone loss in the ovariectomised rat. Br. J. Nutr. 92, 119–127. Seifert, M.F., Watkins, B.A., 1997. Role of dietary lipid and antioxidants in bone metabolism. Nutr. Res. 17, 1209–1228.
CHAPTER | 131 Olives and Olive Oil in the Prevention of Osteoporosis
Tappia, P.S., Grimble, R.F., 1994. Complex modulation of cytokine induction by endotoxin and tumour necrosis factor from peritoneal macrophages of rats by diets containing fats of different saturated, monounsaturated and polyunsaturated fatty acid composition. Clin. Sci. 87, 173–178. Trichopoulou, A., Georgiou, E., Bassiakos, Y., Lipworth, L., Lagiou, P., Proukakis, C., Trichopoulos, D., 1997. Energy intake and monounsaturated fat in relation to bone mineral density among women and men in Greece. Prev. Med. 26, 395–400. Trichopoulou, A., Dilis, V., 2007. Olive oil and longevity. Mol. Nutr. Food Res. 51, 1275–1278. Visioli, F., Bellomo, G., Galli, C., 1998. Free radical-scavenging properties of olive oil polyphenols. Biochem. Biophys. Res. Commun. 247, 60–64.
1203
Visioli, F., Bellomo, G., Montedoro, G., Galli, C., 1995. Low density lipoprotein oxidation is inhibited in vitro by olive oil constituents. Atherosclerosis 117, 25–32. Visioli, F., Galli, C., Plasmati, E., Viappiani, S., Hernandez, A., Colombo, C., Sala, A., 2000. Olive phenol hydroxytyrosol prevents passive smoking-induced oxidative stress. Circulation 102, 2169–2171. Wallace, F.A., Neely, S.J., Miles, E.A., Calder, P.C., 2000. Dietary fats affect macrophage-mediated cytotoxicity towards tumour cells. Immunol. Cell. Biol. 78, 40–48. Weinbrenner, T., Fitó, M., de la Torre, R., Saez, G.T., Rijken, P., Tormos, C., Coolen, S., Albaladejo, M.F., Abanades, S., Schroder, H., Marrugat, J., Covas, M.I., 2004. Olive oils high in phenolic compounds modulate oxidative/antioxidative status in men. J. Nutr. 134, 2314–2321.
This page intentionally left blank
Chapter 132
Effects of Olive Oil and Guar on Fructoseinduced Insulin Resistance María L. Villanueva-Peñacarrillo1, Pablo G. Prieto1, Jésus Cancelas1, Verónica Sancho1, Paola Moreno1, Willy J. Malaisse2 and Isabel Valverde1 1 2
Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain Laboratory of Experimental Hormonology, Brussels Free University, Brussels, Belgium
132.1 INTRODUCTION Insulin resistance (Table 132.1) is a metabolic disorder consistent in a diminished response to insulin in insulinsensitive tissues, and it is associated with some common diseases such as type 2 diabetes, hypertension, obesity and myocardial infarction (Bloomgarden, 2007). Pharmacological intervention, moderate exercise and diets are accepted as positive factors to ameliorate the insulin resistance (Soriguer et al., 2004; Staels, 2006; Bailey, 2007; Corcoran et al., 2007; Gill, 2007; Tierney and Roche, 2007).
Fructose, as a free hexose or as sucrose, given in an intragastric bolus, favors D-glucose homeostasis rather than augmenting the hyperglycemic response to oral D-glucose intake (Prieto et al., 2004). In spite of that, the long-term administration of D-fructose (Table 132.2) incorporated in either diet or drinking water is currently used as a model to induce insulin resistance in experimental animals (Kamata and Yamashita, 1999; Catena et al., 2003; Nandhini et al., 2005; Chang et al., 2005; Lee et al., 2007). We compared these two modalities of D-fructose administration (Table 132.3) and assessed the reversibility of the fructose-induced insulin
TABLE 132.1 Features of insulin resistance. Symptoms
Associated diseases
Defects on
Treatment
Hyperglycemia
Type 2 diabetes
Insulin action
Insulin-sensitizing agents: biguanides (metformin) and thiazolidinediones (pioglitazone and rosiglitazone)
Hyperinsulinemia
Hypertension
Insulin receptor
Agents that enhance phosphorylation and prolong the tyrosine kinase activity of the insulin receptor
Dyslipidemia
Obesity
Post-receptor signaling
Inhibitors of phosphatases and serine kinases; increasing the activity of phosphatidylinositol 3-kinase
Atherosclerosis
Myocardial infarction
Elevated intramyocellular triacylglycerol
Insulinomimetic agents (GLP-1)
Endothelial dysfunction
Diet (olive oil, guar)
Overconsumption of calories
Exercise
Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
1205
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1206
SECTION | II
resistance in normal rats, and found that the enrichment of the diet with either powdered D-fructose (50%, w/w) or drinking water with the ketohexose (20%, w/v) leads to comparable metabolic changes (Cancelas et al., 2008). In both cases, a lesser gain in body weight was noticed in the fructose rats, even when these rats and the control animals displayed comparable body weight at the onset of the period under consideration; this coincided with a lower food intake in the fructose rats than in the control animals (Table 132.3). The plasma D-glucose and insulin concentrations increased in the rats during exposure to the fructose-enriched diet or drinking water, whilst such was not the case in the control animals, and those increases were observed both in the fed state and after overnight fasting (Table 132.3).
132.2 HOMA, CLAMP AND INSULINOGENIC INDEX In both experimental models, insulin resistance became evident as judged by the following criteria. In the rats exposed to the fructose-enriched diet and examined in fed state, an increase in HOMA was observed, whilst such was not the case
TABLE 132.2 Effects of long-term administration of fructose in normal rats.
Major Organ Systems Including Liver and Metabolism
in the control animals; likewise, after overnight starvation, the HOMA increased (p ⬍ 0.01) in the fructose-fed rats from 9.30 ⫾ 2.11 pM mM⫺1 at day 19 to 27.16 ⫾ 5.07 pM mM⫺1 at day 36 (n ⫽ 8 in both cases), whilst remaining stable (p ⬎ 0.5) in the control rats with mean values of 11.55 ⫾ 3.23 pM mM⫺1 (n ⫽ 6) at day 19 and 8.36 ⫾ 3.24 pM mM⫺1 (n ⫽ 5) at day 36. In the rats given access to drinking water containing D-fructose, an increase in HOMA was also observed between day 0 and day 65 in overnight fasted rats, and between day ⫺5 and day 60 in fed animals. Moreover, in these rats, the rate of D-glucose infusion during a hyperinsulinemiceuglycemic clamp was lower than in control animals, despite a virtually identical mean glycemia and plasma insulin concentration (Table 132.3). In addition, the insulinogenic index, which refers to the secretory responsiveness to D-glucose of insulin-producing cells in the pancreas, also underwent comparable changes in the two experimental models of exposure to D-fructose. In the rats exposed to the fructoseenriched diet, and examined after overnight starvation, such an index increased (p ⬍ 0.025) from 10.4 ⫾ 2.7 pMmM⫺1 at day 19 to 19.0 ⫾ 1.9 pM mM⫺1 at day 36 (n ⫽ 8 in both cases), whilst remaining unchanged (p ⬎ 0.95) in the control animals, with mean values of 10.0 ⫾ 3.6 pM mM⫺1 (n ⫽ 6) on day 19 and 9.7 ⫾ 4.0 pM mM⫺1 (n ⫽ 5) on day 36.
TABLE 132.3 Effects of long-term administration of fructose in the diet or in the drinking water, compared to standard diet, in normal rats (Cancelas et al., 2008).
Kamata and Yamashita, 1999
Catena et al., 2003
Fructose
10% in water
66% in diet
Fructose
Length
12 weeks
2 weeks
20% in water
50% in diet
Weight increment
–
Length in days
65
50
Systolic blood pressure
↑
↑
Weight increment
↓
↓
Fasting glycemia
↑
↑
Plasma glucose
↑
↑
Plasma insulin
↑
↑
Triglyceride
↑
↑
HDL cholesterol
↑ ↑
↑
Glucose-induced insulin secretion
↑
LDL cholesterol Free fatty acids
↑
Insulin sensitivity (HOMA)
↓
↓
Insulin-induced glucose uptake (CLAMP)
↑
Number of insulin receptors
↑
This table shows the results of two studies performed in normal rats, one given the fructose in the drinking water and the other including it in the diet. –: no change; ↑: significantly higher.
Fasting insulinemia
↑
Postprandial glycemia
↑
Glucose tolerance
↓
This table shows the results of a study performed in normal rats, given the fructose in the drinking water or including it in the diet. : trend to higher; ↑ significantly higher; ↓ significantly lower.
1207
CHAPTER | 132 Effects of Olive Oil and Guar on Fructose-induced Insulin Resistance
A comparable situation prevailed when considering the measurements made in the same animals when examined in fed state. Likewise, in the rats given access to the fructose-enriched drinking water and examined in the fed state, the insulinogenic index was significantly higher at day 60 than at day ⫺5. These changes suggest that the exposure to D-fructose did not alter the functional capacity of insulin-producing cells but, instead, enhanced their secretory responsiveness to D-glucose. The latter phenomenon probably reflects an attempt of the endocrine pancreas to compensate for the fructose-induced insulin resistance. It coincided, in the rats fed the fructose-enriched diet, with a decrease in the insulin content of the pancreas, whether the latter content was expressed in absolute terms (μg/pancreas) or relative to either pancreatic wet weight (μg g⫺1 wet wt.) or pancreatic protein content ((μg mg⫺1 protein). In considering the latter change, it should be kept in mind that D-fructose, whilst displaying a modest positive insulinotropic capacity (Prieto et al., 2004) fails to stimulate proinsulin biosynthesis (Viñambres et al., 1997).
132.3 STARVED/FED RATIO OF PLASMA GLUCOSE Fructose-fed rats (Cancelas et al., 2008) reveal a further diabetes-like feature, namely a progressive increase of the starved/fed ratio in plasma D-glucose concentration as a function of the length of exposure to either the diet enriched with D-fructose or the drinking water containing the ketohexose (Table 132.4 and Table 132.5). In the first instance, the starved/fed ratio at day 36 averaged 118.7 ⫾ 12.2% (n ⫽ 8) of the paired value found in the same rat(s) at day 19. Also, the starved/fed ratio at day 60/65 averaged 108.5 ⫾ 5.1% (n ⫽ 13) of the paired value recorded in the same rat(s) at day ⫺5/zero. Pooling all available data, the
starved/fed ratio in plasma D-glucose concentration thus averaged, after prolonged exposure to fructose-enriched food or water, 112.4 ⫾ 5.6% (n ⫽ 21; p ⬍ 0.04) of that recorded either at the onset of the experiments or after a shorter exposure to the fructose-enriched diet. Moreover, either a close-to-significant difference was reached when comparing fructose-fed rats to control animals, or a significant difference was reached when taking into account other variables of glucose homeostasis to assess the relative extent of changes in the starved/fed ratio. Even when considering only the results concerning the plasma D-glucose concentration recorded at day 60/65 versus day ⫺5/zero in the second instance, a highly significant difference in the starved/fed ratio was reached (114.4 ⫾ 3.6%; n ⫽ 11; p ⬍ 0.005) when ignoring two of the 13 individual values, both well below the lower limit of the 95% confidence interval. As a matter of fact, already after 19 days of exposure to a fructose-enriched diet, the starvation-induced impairment of glucose tolerance was less pronounced than that otherwise observed in control animals. In the same perspective, it should be underlined that, in the rats fed the fructose-enriched diet, the fasted/fed ratio for the insulinogenic index increased significantly between day 19 and 36, whilst such was not the case in the control animals. In other words, the secretory responsiveness of insulin-producing cells to D-glucose was less severely affected by starvation after prolonged exposure of the rats to the fructoseenriched diet.
132.4 REVERSIBILITY OF FRUCTOSEINDUCED INSULIN RESISTANCE In the fructose-fed rats, the insulin resistance was reversed within 15 days after 2 months of exposure to fructoseenriched drinking water and within 30 days after 4 months’
TABLE 132.4 Starved/fed ratio (percent) of metabolic variables in control animals and rats fed a fructose-enriched diet (Cancelas et al., 2008). Rats
Control
Fructose-fed
Day
19
36
19
36
Body weight
74.5 ⫾ 2.2 (6)
93.2 ⫾ 0.5 (5)
87.9 ⫾ 0.8 (8)
100.5 ⫾ 1.6 (8)
Plasma D-glucose
70.0 ⫾ 5.1 (6)
55.0 ⫾ 2.4 (5)
60.1 ⫾ 2.8 (8)
70.4 ⫾ 6.8 (8)
Plasma insulin
27.5 ⫾ 5.8 (6)
18.6 ⫾ 6.3 (5)
31.7 ⫾ 9.4 (8)
58.0 ⫾ 7.2 (8)
Insulinogenic index
39.4 ⫾ 9.9 (6)
41.2 ⫾ 10.6 (5)
53.8 ⫾ 17.9 (8)
88.7 ⫾ 11.8 (8)
Data, in percent, are expressed as Mean ⫾ SEM together with the number of observations in parenthesis. From Cancelas, J., Prieto P.G., García-Arévalo, M., Sancho, V., Villanueva-Peñacarrillo, M.L., Malaisse W.J. and Valverde, I. Induction and reversibility of insulin resistance in rats exposed to exogenous D-fructose, Horm. Metab. Res. 2008; 40: 459–466. Reprinted by permission.
1208
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 132.5 Effect of overnight fasting upon metabolic variables before and after 65 days’ exposure of rats to fructose-enriched drinking water (Cancelas et al. 2008). Plasma D-glucose (mM)
Plasma insulin (pM)
Insulinogenic index (pM mM⫺1)
HOMA (pM mM⫺1)
Fed (day ⫺5)
7.4 ⫾ 0.1 (13)
260 ⫾ 10 (13)
35.2 ⫾ 1.5 (13)
85.6 ⫾ 3.5 (13)
Fasted (day 0)
4.6 ⫾ 0.1 (13)
125 ⫾ 10 (13)
27.3 ⫾ 2.1 (13)
25.7 ⫾ 2.3 (13)
62.0 ⫾ 1.5 (13)
46.8 ⫾ 4.8 (13)
75.6 ⫾ 7.7 (13)
29.0 ⫾ 3.1 (13)
Fed (day 60)
8.0 ⫾ 0.2 (13)
292 ⫾ 23 (13)
36.7 ⫾ 2.8 (13)
103.8 ⫾ 8.3 (13)
Fasted (day 65)
5.3 ⫾ 0.2 (13)
170 ⫾ 18 (13)
31.9 ⫾ 3.0 (13)
40.8 ⫾ 4.8 (13)
66.3 ⫾ 3.0 (13)
56.4 ⫾ 7.5 (13)
85.2 ⫾ 10.5 (13)
37.4 ⫾ 5.9 (13)
Fasted/fed ratio (%)
Fasted/fed ratio (%)
Data expressed in Mean ⫾ SEM together with the number of observations in parenthesis. From Cancelas, J., Prieto P.G., García-Arévalo, M., Sancho, V., Villanueva-Peñacarrillo, M.L., Malaisse W.J. and Valverde, I. Induction and reversibility of insulin resistance in rats exposed to exogenous D-fructose, Horm. Metab. Res. 2008; 40: 459–466. Reprinted by permission.
exposure to the same drinking water (Cancelas et al., 2008). If anything, the trend was towards somewhat lower insulin sensitivity, as judged from the rate of D-glucose infusion during the hyperinsulinemic-euglycemic clamp, in the latter rats than in either the former ones or the control animals. In summary, these results (Cancelas et al., 2008) suggest that the two experimental models of exposure to exogenous D-fructose used in our study, i.e. the incorporation of the ketohexose in either the solid diet or drinking water, achieved closely comparable metabolic alterations. While also documenting the reversibility of the fructoseinduced insulin resistance, the work draws attention to the increased secretory responsiveness to D-glucose of insulinproducing cells, considered as an attempt to compensate for the fructose-induced insulin resistance. It also reveals a novel feature of the diabetogenic effect resulting from the long-term exposure to exogenous D-fructose, namely the attenuation of the changes in both glucose homeostasis and secretory sensitivity to D-glucose of insulin-producing cells otherwise attributable to overnight starvation.
TABLE 132.6 Effects of long-term administration of olive oil and guar gum, compared to standard diet, in normal rats. Olive oil (Prieto et al., 2005) Length in days
Guar (Prieto et al., 2006) 65
Weight increment
↑
–
Fasted glycemia
–
–
Postprandial glycemia
–
Glucose tolerance
↑
↑
Glucose-induced insulin secretion
↑
↑
Insulin sensitivity (HOMA)
↑
–: no change;
: trend to lower; ↑: significantly higher.
132.5 EFFECT OF OLIVE OIL AND GUAR ON GLUCOSE HOMEOSTASIS To improve glucose homeostasis, exposure to diets supplemented with several specific nutrients such as olive oil (Garg, 1994; Riccardi and Rivellese, 2000; Kotake et al., 2004), long-chain polyunsaturated ω3 fatty acids (Vessby, 1993; Delarue et al., 2004; Nettleton and Katz, 2005; West et al., 2005) or guar (Slavin and Greenberg, 2003; Russo et al., 2003; Suzuki and Hara, 2004; Prieto et al., 2006), amongst others, is under current investigation.
It was proposed that diets containing an increased proportion of monounsaturated fatty acids, compared to carbohydrates, improve glycemic control and lipid metabolism (Parillo et al., 1992; Riccardi and Parrillo, 1993; Wright, 1998). Recently, it has been reported that (Table 132.6) an olive oil-enriched diet improves insulin response to oral glucose and glucose tolerance in normal rats (Prieto et al., 2005).
CHAPTER | 132 Effects of Olive Oil and Guar on Fructose-induced Insulin Resistance
Guar gum intake immediately decreases postprandial plasma glucose and insulin concentrations in either normal human subjects (Torsdottir et al., 1989; Morgan et al., 1990; Braaten et al., 1991; Kirsten et al., 1991; Fairchild et al., 1996; Sierra et al., 2001) or patients with type 2 diabetes (Russo et al., 2003); it was also reported that in normal subjects (Landin et al., 1992) or patients with type 1 or type 2 diabetes (Ebeling et al., 1988; Vuorinen-Markkola et al., 1992; Groop et al., 1993; Lafrance et al., 1998), longterm ingestion of guar-supplemented diet, for 4–48 weeks, lowers fasting blood glucose, hemoglobin A1C, cholesterol and triglycerides, decreases systolic and diastolic blood pressure, increases insulin sensitivity and improves postprandial glucose tolerance. In pigs, adding guar gum to the fodder reduces glucose absorption and insulin secretion in both short-term (Ellis et al., 1995) and long-term (Nunes and Malmlof, 1992) experiments. Addition of guar gum to the diet for 10–30 days was also reported to improve glucose tolerance and insulin sensitivity in normal (Vachon et al., 1988; Begin et al., 1989) and streptozotocin-induced diabetic (Cameron-Smith et al., 1997) rats, or in rats rendered glucose intolerant and hypertriglyceridemic by fructose feeding (Suzuki and Hara, 2004).
132.6 EFFECT OF GUAR GUM IN NORMAL RATS In normal rats, we have studied (Prieto et al., 2006) the immediate effect of guar gum on plasma glucose and insulin response to either intragastric administration of glucose or food intake by meal-conditioned rats, and the long-term effect of guar supplementation on plasma glucose and insulin response to an intragastric glucose tolerance test performed after overnight fasting. In the short-term experiments, the concomitant intragastric administration of glucose (1.2 mg g⫺1 body weight) and guar (40 μg g⫺1 body weight) reduced the early increment in plasma glucose and insulin. In a comparable manner, the intake of food for 15 minutes by meal-trained rats resulted in a lesser peak increment in plasma glucose and insulin concentrations when guar (20%, w/w) was added to the food. In the long term, the changes in body weight over 62 days of observation were comparable in the control and guar rats. According to intragastric glucose tolerance tests conducted in overnight fasted rats maintained for 19–36 days on either the control or guar-enriched diet, the glycemic response was similar in both groups of rats, while a higher insulin response was observed in the guar group than in the controls. Thus, in this study it was found that, in healthy rats, long-term intake of guar improves glucose tolerance and insulin response to glucose absorption (Table 132.6). This higher secretory response of insulin-producing cells to the glucose tolerance test, which did not include the presence of guar in the solution delivered intragastrically, is likely to
1209
reflect their sensitivity to changes in extracellular glucose concentration. If so, these results would suggest enhanced responsiveness of islet B-cells to hexose in rats otherwise fed the guar-enriched diet. This situation could in turn be attributable to a lesser peak increase of glycemia at the time of food intake, as documented in the meal-trained rats. Secretory responsiveness to glucose, which is modulated in a bell-shaped manner by preabsortive glycemia, may also be influenced by rapid changes in glycemia. For instance the decreased secretory response to glucose of insulinproducing cells prevailing during starvation is suppressed by the pulse intravenous injection of glucose causing shortlived waves of hyperglycemia despite representing a negligible caloric supply (Grey et al., 1970).
132.7 EFFECT OF OLIVE OIL AND GUAR ON INSULIN RESISTANCE RATS We have also assessed the effect of olive oil- (15% v/w: 13% saturated fatty acids, 79% monounsaturated fatty acids and 8% polyunsaturated fatty acids) or gum-guar(20% w/w) enriched diet upon the metabolic consequences of exposure to exogenous fructose (20% w/v) dissolved in tap water, for 65 days (Prieto et al., 2007). It was found that diet supplementation with olive oil or guar failed to affect the increase in plasma insulin concentration and insulinogenic index and the decrease in insulin sensitivity observed in these rats at day 65 after overnight starvation, when compared to those fed in a standard diet (control). Likewise, the glucose infusion rate during the hyperinsulinemic-euglycemic clamp was not significantly different in control rats, and in rats fed with olive oil or guar. Moreover, the decrease in the fasted/fed ratio for either the plasma insulin concentration or insulinogenic index found at day 65 in the rats exposed to exogenous Dfructose, also failed to differ significantly in rats fed with olive oil- or guar-enriched diets, when compared with those fed on the standard diet. In the rats exposed to exogenous fructose, the olive oilfed rats differed from the other two groups of animals in that the daily food intake and gain in body weight were higher, while the guar-fed rats failed to differ significantly from the control rats (Table 132.7). A similar situation (Table 132.6) was observed when comparing normal rats exposed to fructose-free tap water and fed either a control diet or the olive oil-enriched diet (Prieto et al., 2005). The guar-fed rats differed from the control and oliveoil-fed rats in two respects. First, on day 60, the plasma D-glucose concentration of fed rats was significantly lower in guar-fed rats than in either control or olive oil-fed rats (Table 132.7), a finding which is consistent with the lowering of postprandial glycemia by guar, as attributable to decreased glucose absorption (Prieto et al., 2006). Second, in the guar-fed rats, the plasma D-glucose concentration
1210
SECTION | II
TABLE 132.7 Effect of olive oil and guar on fructoseinduced insulin resistant rats, compared to standard diet (Prieto et al., 2007). Fructose-induced insulin resistance Olive oil
●
●
●
Guar ●
Major Organ Systems Including Liver and Metabolism
Fructose-induced insulin resistance is a reversible phenomenon. Long-term ingestion of guar gum improves glucose tolerance and insulin response to glucose absorption. Diet supplementation with olive oil in rats exposed to exogenous fructose does not provoke a decrease in food intake and body weight gain. Diet supplementation with guar gum in rats exposed to exogenous fructose lowers plasma D-glucose concentration in fasting and fed state.
Length in days
65
Weight increment
↑
–
Caloric ingestion
↑
–
Fasting glycemia
–
↓
ACKNOWLEDGMENT
Postprandial glycemia
–
↓
We thank the Institute of Health Carlos III (PI 060076 and RD0600150004), Spain, for financial support.
↓–: no change; ↑: significantly higher; ↓ significantly lower.
REFERENCES found at day 65 after overnight starvation failed to differ from that recorded, in the same animals, before the exposure to exogenous D-fructose, whilst in both the control and olive oil-fed rats, the plasma D-glucose concentration in overnight-fasted rats was significantly higher at day 65 than at day 0. Such a difference could also be somehow related to the lowering of postprandial glycemia by guar, such an antidiabetic effect opposing the diabetogenic action of exogenous D-fructose supplementation. In conclusion, these results indicate, on one hand, that in a rat model of fructose-induced insulin resistance, the long-term administration of an olive oil-enriched diet opposes the lowering effect of the ketohexose on food intake and body weight gain (Table 132.7); and on the other hand, that the long-term administration of a guar-enriched diet prevents the fructose-induced increase in glycemia otherwise recorded after overnight starvation (Table 132.7), whilst maintaining a low-food intake. These findings argue in favor of guar, rather than olive oil, to oppose the effect of exogenous fructose on glucose homeostasis.
SUMMARY POINTS ●
●
Fructose-enriched diet or drinking water is being used as models for experimental insulin resistance. Long-term exposure to exogenous fructose induces: a decrease in both food intake and body weight gain intolerance to D-glucose an increased insulin secretion in response to D-glucose attenuation of the changes in glucose homeostasis and secretory sensitivity of islet B-cells to D-glucose, attributable to overnight starvation. ●
●
●
●
Bailey, C.J., 2007. Treating insulin resistance: future prospects. Diab. Vasc. Dis. Res. 4, 20–31. Begin, F., Vachon, C., Jones, J.D., Wood, P.J., Savoie, L., 1989. Effect of dietary fibers on glycemia and insulinemia and on gastrointestinal function in rats. Can. J. Physiol. Pharmacol. 67, 1265–1271. Bloomgarden, Z.T., 2007. Insulin resistance, dyslipidemia, and cardiovascular disease. Diabetes Care 30, 2164–2170. Braaten, J.T., Wood, P.J., Scott, F.W., Riedel, K.D., Poste, L.M., Collins, M.W., 1991. Oat gum lowers glucose and insulin after an oral glucose load. Am. J. Clin. Nutr. 53, 1425–1430. Cameron-Smith, D., Habito, R., Barnett, M., Collier, G.R., 1997. Dietary guar gum improves insulin sensitivity in streptozotocin-induced diabetic rats. J. Nutr. 127, 359–364. Cancelas, J., Prieto, P.G., García-Arévalo, M., Sancho, V., VillanuevaPeñacarrillo, M.L., Malaisse, W.J., and Valverde, I., 2008. Induction and reversibility of insulin resistance in rats exposed to exogenous D-fructose. Horm. Metab. Res. Apr 7 [Epub ahead of print]. Catena, C., Giacchetti, G., Novello, M., Colussi, G., Cavarape, A., Sechi, L.A., 2003. Cellular mechanisms of insulin resistance in rats with fructose-induced hypertension. Am. J. Hypertens. 16, 973–978. Chang, J.C., Wu, M.C., Liu, I.M., Cheng, J.T., 2005. Increase of insulin sensitivity by stevioside in fructose-rich chow-fed rats. Horm. Metab. Res. 37, 610–616. Corcoran, M.P., Lamon-Fava, S., Fielding, R.A., 2007. Skeletal muscle lipid deposition and insulin resistance: effect of dietary fatty acids and exercise. Am. J. Clin. Nutr. 85, 662–877. Delarue, J., LeFoll, C., Corporeau, C., Lucas, D., 2004. N-3 long chain polyunsaturated fatty acids: a nutritional tool to prevent insulin resistance associated to type 2 diabetes and obesity? Reprod. Nutr. Dev. 44, 289–299. Ebeling, P., Yki-Jarvinen, H., Aro, A., Helve, E., Sinisalo, M., Koivisto, V.A., 1988. Glucose and lipid metabolism and insulin sensitivity in type 1 diabetes: the effect of guar gum. Am. J. Clin. Nutr. 4, 98–103. Ellis, P.R., Roberts, F.G., Low, A.G., Morgan, L.M., 1995. The effect of high-molecularweight guar gum on net apparent glucose absorption and net apparent insulin and gastric inhibitory polypeptide production in the growing pig: relationship to rheological changes in jejunal digesta. Br. J. Nutr. 74, 539–556.
CHAPTER | 132 Effects of Olive Oil and Guar on Fructose-induced Insulin Resistance
Fairchild, R.M., Ellis, P.R., Byrne, A.J., Luzio, S.D., Mir, M.A., 1996. A new breakfast cereal containing guar gum reduces post-prandium plasma glucose and insulin concentrations in normal-weight human subjects. Br. J. Nutr. 76, 63–73. Garg, A., 1994. High monounsaturated fat diet for diabetic patients. Diabetes Care 17, 242–246. Gill, J.M., 2007. Physical activity, cardiorespiratory fitness and insulin resistance: a short update. Curr. Opin. Lipidol. 18, 47–52. Grey, N.J., Goldring, S., Kipnis, D.M., 1970. The effect of fasting, diet, and actinomycin D on insulin secretion in the rat. J. Clin. Invest. 49, 881–885. Groop, P.H., Aro, A., Stenman, S., Groop, L., 1993. Long-term effects of guar gum in subjects with non-insulin-dependent diabetes mellitus. Am. J. Clin. Nutr. 58, 513–518. Kamata, K., Yamashita, K., 1999. Insulin resistance and impaired endothelium-dependent renal vasodilatation in fructose-fed hypertensive rats. Res. Común. Mol. Pathol. Pharmacol. 103, 195–210. Kirsten, R., Nelson, K., Storck, J., Hubner-Steiner, U., Speck, U., 1991. Influence of two guar preparations on glucose and insulin levels during a glucose tolerance test in healthy volunteers. Int. J. Clin. Pharmacol. Ther. Toxicol. 29, 19–22. Kotake, J., Tanaka, Y., Umehara, N., Miyashita, A., Tsuru, T., Hikida, S., Mizote, H., 2004. Effects of a high-monounsaturated fat diet on glucose and lipid metabolisms in normal and diabetic mice. Nutr. Sci. Vitaminol. 50, 106–113. Lafrance, L., Rabasa-Lhoret, R., Poisson, D., Ducros, F., Chiasson, J.L., 1998. Effects of different glycaemic index foods and dietary fibre intake on glycaemic control in type 1 diabetic patients on intensive insulin therapy. Diabet. Med. 15, 972–978. Landin, K., Holm, G., Tengborn, L., Smith, U., 1992. Guar gum improves insulin sensitivity, blood lipids, blood pressure, and fibrinolysis in healthy men. Am. J. Clin. Nutr. 56, 1061–1065. Lee, W.K., Kao, S.T., Liu, I.M., Cheng, J.T., 2007. Ginsenoside Rh2 is one of the active principles of Panax ginseng root to improve insulin sensitivity in fructose-rich chow-fed rats. Horm. Metab. Res. 39, 347–354. Morgan, L.M., Tredger, J.A., Wright, J., Marks, V., 1990. The effect of soluble- and insoluble-fibre supplementation on post--prandium glucose tolerance, insulin and gastric inhibitory polypeptide secretion in healthy subjects. Br. J. Nutr. 64, 103–110. Nandhini, A.T., Thirunavukkarasu, V., Anuradha, C.V., 2005. Taurine modifies insulin signaling enzymes in the fructose-fed insulin resistant rats. Diabetes Metab. 31, 337–344. Nettleton, J.A., Katz, R., 2005. n-3 long-chain polyunsaturated fatty acids in type 2 diabetes: a review. J. Am. Diet. Assoc. 105, 428–440. Nunes, C.S., Malmlof, K., 1992. Effects of guar gum and cellulose on glucose absorption, hormonal release and hepatic metabolism in the pig. Br. J. Nutr. 68, 693–700. Parillo, M., Rivellese, A.A., Ciardullo, A.V., Giasso, A., Genovese, S., Ricardo, G., 1992. A high-monounsaturated-fat/low-carbohydrate diet improves peripheral insulin sensitivity in non-insulin-dependent diabetic patients. Metabolism 41, 1373–1378. Prieto, P.G., Cancelas, J., Moreno, P., Villanueva-Peñacarrillo, M.L., Malaisse, W.J., Valverde, I., 2007. Effects of diet supplementation with olive oil and guar upon fructose-induced insulin resistance in normal rats. Endocrine 31, 294–299. Prieto, P.G., Cancelas, J., Villanueva-Peñacarrillo, M.L., Valverde, I., Malaisse, W.J., 2004. Plasma D-glucose, D-fructose and insulin responses after oral administration of D-glucose, D-fructose and sucrose to normal rats. J. Am. Coll. Nutr. 23, 414–419.
1211
Prieto, P.G., Cancelas, J., Villanueva-Peñacarrillo, M.L., Valverde, I., Malaisse, W.J., 2006. Short-term and long-term effects of guar on postprandial plasma glucose, insulin and glucagon-like peptide 1 concentration in healthy rats. Horm. Metab. Res. 38, 397–404. Prieto, P.G., Cancelas, J., Villanueva-Peñacarrillo, M.L., Valverde, I., Malaisse, W.J., 2005. Effects of an olive oil-enriched diet on plasma GLP-1 concentration and intestinal content, plasma insulin concentration, and glucose tolerance in normal rats. Endocrine 26, 107–115. Riccardi, G., Parrillo, M., 1993. Comparison of the metabolic effects of fat-modified vs low fat diets. Ann. N.Y. Acad. Sci. 683, 192–198. Riccardi, G., Rivellese, A.A., 2000. Dietary treatment of the metabolic syndrome – the optimal diet. Br. J. Nutr. 83, S143–S148. Russo, A., Stevens, J.E., Wilson, T., Wells, F., Tonkin, A., Horowitz, M., Jones, K.L., 2003. Guar attenuates fall in post-prandium blood pressure and slows gastric emptying of oral glucose in type 2 diabetes. Dig. Dis. Sci. 48, 1221–1229. Sierra, M., García, J.J., Fernández, N., Díez, M.J., Calle, A.P., Sahagun, A.M., 2001. Effects of ispaghula husk and guar gum on postprandium glucose and insulin concentrations in healthy subjects. Eur. J. Clin. Nutr. 55, 235–243. Slavin, J.L., Greenberg, N.A., 2003. Partially hydrolyzed guar gum: clinical nutrition uses. Nutrition 19, 549–552. Soriguer, F., Esteva, I., Rojo-Martinez, G., Ruiz de Adana, M.S., Dobarganes, M.C., García-Almeida, J.M., Tinahones, F., Beltrán, M., González-Romero, S., Olveira, G., Gómez-Zumaquero, J.M., 2004. Oleic acid from cooking oils is associated with lower insulin resistance in the general population (Pizarra study). Eur. J. Endocrinol. 150, 33–39. Staels, B., 2006. Metformin and pioglitazone: Effectively treating insulin resistance. Curr. Med. Res. Opin. 22 (Suppl 2), S27–S37. Suzuki, T., Hara, H., 2004. Ingestion of guar gum hydrolysate, a soluble and fermentable nondigestible saccharide, improves glucose intolerance and prevents hypertriglyceridemia in rats fed fructose. J. Nutr. 134, 1942–1947. Tierney, A.C., Roche, H.M., 2007. The potential role of olive oil-derived MUFA in insulin sensitivity. Mol. Nutr. Food Res. 51, 1235–1248. Torsdottir, I., Alpsten, M., Andersson, H., Einarsson, S., 1989. Dietary guar gum effects on post-prandium blood glucose, insulin and hydroxyproline in humans. J. Nutr. 119, 1925–1931. Vachon, C., Jones, J.D., Wood, P.J., Savoie, L., 1988. Concentration effect of soluble dietary fibers on post-prandium glucose and insulin in the rat. Can. J. Physiol. Pharmacol. 66, 801–806. Vessby, B., 1993. Dietary supplementation with N-3 polyunsaturated fatty acids in type 2 diabetes. Effects on glucose homeostasis. Ann. N.Y. Acad. Sci. 683, 244–249. Viñambres, C., Villanueva-Peñacarrillo, M.L., Valverde, I., Malaisse, W.J., 1997. Failure of D-fructose to stimulate protein biosynthesis in pancreatic islets. Biochem. Mol. Biol. Int. 41, 571–574. Vuorinen-Markkola, H., Sinisalo, M., Koivisto, V.A., 1992. Guar gum in insulin-dependent diabetes: effects on glycemic control and serum lipoproteins. Am. J. Clin. Nutr. 56, 1056–1060. West, S.G., Hecker, K.D., Mustad, V.A., Nicholson, S., Schoemer, S.L., Wagner, P., Hinderliter, A.L., Ulbrecht, J., Ruey, P., Kris-Etherton, P.M., 2005. Acute effects of monounsaturated fatty acids with and without omega-3 fatty acids on vascular reactivity in individuals with type 2 diabetes. Diabetologia 48, 113–122. Wright, J., 1998. Effect of high-carbohydrate versus high-monounsaturated fatty acids diet on metabolic control in diabetes and hyperglycaemic patiens. Clin. Nutr. 17, 35–45.
This page intentionally left blank
Chapter 133
Effects of an Olive Oil-enriched Diet on Glucagon-like Peptide-1 Isabel Valverde1, Paola Moreno1, Jesús Cancelas1, Pablo G. Prieto1, María L. Villanueva-Peñacarrillo1 and Willy J. Malaisse2 1 2
Department of Metabolism, Nutrition and Hormones, Fundacíón Jiménez Díaz, Madrid, Spain Laboratory of Experimental Hormonology, Brussels Free University, Brussels, Belgium
133.1 INTRODUCTION Several studies have demonstrated that diets with an increased proportion of monounsaturated fatty acids, compared to high-carbohydrate diets, improve glycemic control and also benefit lipid profiles in patients with non-insulindependent (type 2) diabetes mellitus (see Wright, 1998, for review). For instance, Campbell et al. (1994), by studying in ten type 2 diabetic patients the metabolic effects of a home-prepared high-monounsaturated-fat diet, and comparing it with the recommended high-carbohydrate diet for 2 weeks, found that it resulted in a significantly lower 24-h urinary glucose excretion, fasting triglyceride, and mean profile glucose levels (Table 133.1). Also, Garg et al. (1994) reported that, in a group of 42 type 2 diabetic patients, a high-monounsaturated-fat diet, provided as the sole nutrient for 6–14 weeks, lowers plasma D-glucose values when compared to a high-carbohydrate diet (Table 133.1). Likewise, Low et al. (1996) observed that, in obese patients with type 2 diabetes dieting for 6 weeks on a formula with 50% caloric deficit but enriched in monounsaturated fatty acids, both the fasting glucose level and 24-h glycemia decrease significantly more than in patients dieting with the same caloric deficit but enriched with carbohydrates; moreover, in this study, C-peptide levels in fasting, over a 24-h period and during an oral glucose tolerance test, indicated that the diet enriched in monounsaturated fatty acids increases carbohydrate-induced insulin secretion (Table 133.1). Parrillo et al. (1992) documented that, in ten type 2 diabetic patients, a high-monounsaturated fat diet during 2 weeks induced a decrease in postprandial glucose, plasma insulin concentrations, and also fasting plasma triglyceride levels, when compared to a high-carbohydrate diet, and that the insulin-mediated glucose disposal, evaluated with the euglycemic-hyperinsulinemic clamp, was significantly higher Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
with the high-monounsaturated diet (Table 133.1). More recently, Rodríguez-Villar et al. (2004) reported that in 22 type 2 diabetic patients a high-monounsaturated-fat diet for 6 weeks compared with a high-carbohydrate diet, lowered VLDL cholesterol and VLDL triglyceride, indicating that a high-monounsaturated-fat diet is a good alternative to highcarbohydrate diets for nutrition therapy of diabetes, due to a beneficial effect on the lipid profile and superior patient acceptance (Table 133.1). Even in 78 type 2 diabetic patients under insulin therapy, who required enteral tube feeding due to neurological dysphagia (Pohl et al., 2005), it was found that a low-carbohydrate–high-monounsaturated-fat formula was more effective at controlling glycemia than the standard diet. In patients with type 1 diabetes, it has also been suggested that there is a benefit of a high-monounsaturated-fat diet for controlling lipid profiles (Garg, 1998; Strychar et al., 2003). In 162 normal healthy individuals given either a high-saturated-fat or a high-monounsaturated-fat diet for 3 months, it was shown that the latter significantly improves insulin sensitivity compared to the former diet (Riccardi and Rivellese, 2000).
133.2 GLUCAGON-LIKE PEPTIDE-1 The favorable effect of diets rich in monounsaturated fatty acids on glucose homeostasis apparently involves increased secretion of the intestinal and insulinotropic hormone glucagon-like peptide 1 (GLP-1). GLP-1 is produced (Figure 133.1) by post-translational modification of the glucagon precursor, proglucagon, in L cells located in the small and large intestine (Holst, 1997; Drucker, 2002). It is secreted in response to the presence, in the gut lumen, of nutrients, mainly carbohydrates and fats, and communicates this information to the pancreas where it acts as a potent insulin secretagogue,
1213
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1214
SECTION | II
Major Organ Systems Including Liver and Metabolism
TABLE 133.1 Effects of high-monounsaturated-fat diet upon glucose and lipid metabolism in type 2 diabetic patients compared to high-carbohydrate diet. T2D (Campbell et al., 1994)
T2D (Garg et al., 1994)
Obese T2D (Low et al., 1996)
T2D (Parrillo et al., 1992)
T2D (RodríguezVillar et al., 2004)
Number of patients
10
42
17
10
22
Length of diet (weeks)
2
6–14
6
2
6
24 h urinary glucose
↓ ⫽
Plasma glucose
Fasting triglycerides
↓
↓
↓
↓
Glucose-induced insulin secretion
↓
↓
↓
↑
Insulin-induced glucose disposal
↓
⫽
↓
↑
↑
VLDL cholesterol
↓
↓
VLDL triglyceride
↓
This table shows qualitative changes of plasma glucose and lipids, as well as those of the glucose-induced insulin secretion and insulin-induced glucose disposal in type 2 diabetic patients (TD2), without and with obesity, fed with a monounsaturated-fat diet when compared to a standard diet. ⫽: no change, ↑ significantly higher: ↓ significantly lower.
GRPP
KR
1
30
KR
Glucagon 33
61
KR 64–69
GLP-1 72 78
RR 108
111–123
RK
GLP-2
RR 126
158
160
Pancreas (α cells) Intestine (L cells) GRPP
Glucagon
MPGF
PG 1-61 Glicentin GRPP
GLP-1
GLP-2
Oxyntomodulin
FIGURE 133.1 Post-translational process of proglucagon. The tissue-specific post-translational process of proglucagon yields in the pancreatic α cells mainly glucagon, glicentin-related pancreatic peptide (GRPP), proglucagon1–61 and major proglucagon fragment (MPGF), and in the intestinal L cells yields mainly glucagon-like peptide 1 (GLP-1), glucagon-like peptide 2 (GLP-2), glicentin, GRP and oxyntomodulin.
accounting for as much as 50% of postprandial insulin secretion (Creutzfeldt, 2001). GLP-1, has several pleiotropic effects (Figure 133.2) related to assimilation of nutrients and maintenance of normal blood glucose levels, including inhibition of food intake (by enhancing satiety), gastric emptying and glucagon secretion, as well as stimulation of glucosedependent insulin biosynthesis and insulin-independent antidiabetic character (Creutzfeldt, 2001), apart from proven insulinomimetic properties per se (Valverde et al., 2002).
Other studies indicate that GLP-1, or its agonist, exendin-4 (Figure 133.3), promote β-cell proliferation and islet neogenesis (Buteau et al., 2003; Egan et al., 2003). In type 1 and type 2 diabetic patients, the secretion of GLP-1 is diminished (Vilsbøll et al., 2001), but its insulinotropic effect is maintained in type 2 patients (Nauck et al., 1993). In morbidly obese patients, a lower GLP-1 secretion, both basal and oral glucose- or meal-stimulated, is reversed after bariatric surgery (Lugari et al., 2004; Valverde et al.,
1215
CHAPTER | 133 Effects of an Olive Oil-enriched Diet on Glucagon-like Peptide-1
LIVER PANCREAS
MUSCLE
IPG cAMP
secretion transcription proliferation B-cell neogenesis Somatostatin Glucagon Insulin
IPG Glycogen synthasea Glycogen synthesis CO2 & H2O Kinase activity Adenylate cyclase Modulates GLUT-2 expression.
Glycogen synthasea Glycogen synthesis Glucose transport CO2 & H2O Kinase activity Adenylate cyclase Modulates GLUT-4 expression
FAT
GLP-1 cAMP IPG
STOMACH Glucose transport Glycogen synthasea Glycogen synthesis CO2 & H2O Kinase activity Modulates GLUT-4 expression Lipogenesis Lipolysis
BRAIN
Ischemia protector
cAMP cAMP
Anorexic Neurotrophic
Acid secretion Gastric emptying
FIGURE 133.2 Pleiotropic effect of GLP-1. Effects of GLP-1 in pancreas and in extrapancreatic tissues. ….: pancreatic GLP-1 receptor, extrapancreatic GLP-1 receptor; IPG: inositol; : phosphoglycan; cAMP: cyclic adenosine; : monophosphate; ⫽: no effect; ↑: increase; ↓: decrease.
2005). In these two pathological situations, type 2 diabetes and obesity (Table 133.2), the therapeutic use of GLP-1 has been proposed (Meier et al., 2002). The possible beneficial action of GLP-1 was recently extended to Alzheimer patients and other neurodegenerative conditions, as this peptide plays an important role in the regulation of neuronal plasticity and cell survival (Greig et al., 2004; Perry and Greig, 2005). Also, the infusion of GLP-1 after angioplasty, in patients with acute myocardial infarction, ameliorates left ventricular function (Nikolaidis et al., 2004). A single specific GLP-1 receptor, that is a member of the G-protein coupled seven transmembrane domain superfamily, mediates the biological actions of GLP-1 in the pancreatic islets (Thorens, 1992). But the possible existence of other GLP-1 receptors in extrapancreatic tissues has been documented (Yang et al., 1998; Valverde et al., 2002).
133.3 MONOUNSATURATED FATTY ACIDS AND GLP-1 Monounsaturated fatty acids possessing a free carboxyl group indeed stimulate the secretion of intestinal proglucagonderived peptides, including GLP-1, from fetal rat intestinal
GLP-1
HAEGTFTSDVSSYLEGQAALEFIAWLVKGR.NH2
Ex-4
HGEGTFTSDLSKQMEEEAVRLFIEWLKNGGPSSGAPPPS.NH2
Ex-9
DLSKQMEEEAVRLFIEWLKNGGPSSGAPPPS.NH2
FIGURE 133.3 Amino acid structure of GLP-1 and exendins. The homology in amino acid structure of glucagon-like peptide (GLP-1) and exendin-4 (Ex-4), peptide from salivary gland of Gila Monster, and its truncated form exendine9–39 (Ex-9) is shown by bold letters.
TABLE 133.2 Possible therapeutic use of GLP-1. Human disease
References
Diabetes type 2
Maier et al., 2002
Obesity
Maier et al., 2002
Alzheimer
Greig et al., 2004; Perry and Greig, 2005
Myocardial infarction
Nikolaidis et al., 2004
1216
SECTION | II
cultures, this effect being lost upon full saturation of the concerned fatty acids (Rocca and Brubaker, 1995). Various studies deal specifically with the effects of monounsaturated fatty acid diets on GLP-1 secretion and glycemic tolerance. Thomsen et al. (1999) first compared the postprandial responses of glucose, insulin, fatty acids, triacylglycerol, gastric inhibitory peptide (GIP), and also that of GLP-1, to test meals rich in saturated and monounsaturated fatty acids, in young, lean, healthy subjects. GLP-1 and GIP responses were higher after ingestion of a meal containing 80 g olive oil than those after either 50 g carbohydrate (control meal) or 100 g butter (saturated fatty acid meal); this coincided with a lower average of blood D-glucose concentrations after the olive oil meal than after the control meal; however, the early peak of blood D-glucose and plasma insulin concentrations were highest after ingestion of the control meal, and no significant differences in glucose, insulin, or fatty acid response to the two fat-rich meals were seen (Table 133.3). In a comparable study later conducted by the same investigators in overweight patients with type 2 diabetes (Table 133.3), the GLP-1 response was again highest after the olive oil meal, while no significant difference was seen in the glucose, insulin and fatty acid responses to the two fat-rich meals (Thomsen et al., 2003). In lean Zucker rats (Rocca et al., 2001), pair-fed for 2 weeks with a synthetic diet containing 5% fat derived from either olive oil (74% monounsaturated fatty acids) or coconut oil (87% saturated fatty acids), the olive oil-fed rats showed improved glucose tolerance, compared with the CO group, to both oral and duodenal glucose. Despite such a difference, the plasma insulin pattern was comparable in the two groups, but the secretion of gut glucagonlike immunoreactive material during the duodenal glucose tolerance test was higher (min 10) in the olive oil-fed rats than in coconut oil-fed rats. The benefit in glycemia tolerance conferred by feeding was abolished when the GLP-1 antagonist exendin9–39 (Figure 133.3) was infused 3 min before the duodenal glucose administration.
133.4 OLIVE OIL DIET IN NORMAL RATS We have reported (Prieto et al., 2005) that in meal-trained normal rats, exposure for 15 min to an olive oil-enriched diet (13% saturated fatty acids, 79% monounsaturated fatty acids, and 8% polyunsaturated fatty acids), provoked a marked increase in plasma GLP-1 concentration, a phenomenon not observed in animals given access to the control standard diet (Table 133.4). Over a period of 50 days (Table 133.4), the plasma GLP-1 concentration was also higher in olive oil-fed rats, as distinct from control diet. No significant difference between the two groups of rats was observed, however, in the GLP-1 content of the jejunum, ileum, colon and cecum (Table 133.4). Also, in both male and female rats, the gain in body weight, over the same
Major Organ Systems Including Liver and Metabolism
period of 50 days, was higher in the rats exposed to the olive oil rather than control diet (Table 133.4); and when an oral glucose tolerance test was performed at 19 and 36 days in overnight fasted rats, the increment in plasma insulin concentration was also higher in the rats fed the olive oil (Figure 133.4). This coincided with a higher value, during the glucose tolerance test, for the paired ratio between the increments in plasma insulin and glucose concentration in the rats fed the olive oil diet (Figure 133.4). At day 36, these rats also displayed a higher GLP-1 response (Table 133.4), and improved glucose tolerance, when compared to the animals fed the control diet. Thus, although long-term exposure to a food rich in monounsaturated fatty acids may offer the disadvantage of a higher increase in body weight, it results in a better tolerance to D-glucose. The later phenomenon appears attributable, in part at least, to a higher insulin response to D-glucose enteral intake, which may itself result, among other factors, from a higher level of circulating GLP-1 in the rats fed the olive oil diet. On the sole basis of these results, it cannot be decided whether the improved tolerance to D-glucose eventually present in the rats fed the olive oil diet also entails increased sensitivity to insulin in its target cells; but taken as a whole, the findings argue in support of the proposal that a diet enriched in triglycerides containing a high percentage of monounsaturated fatty acids may favor glucose homeostasis (Wright, 1998).
TABLE 133.3 Responses to a test meal rich in monounsaturated fatty acids compared to a saturated fatty acid rich meal. Healthy subjects (Thomsen et al., 1999)
Obese T2D (Thomsen et al., 2003)
Number of subjects
10
12
Plasma glucose
⫽
⫽
Plasma insulin
⫽
⫽
Fatty acids
⫽
⫽
Triacylglycerol
↓
↓
HDL cholesterol
↑
↑
GLP-1 secretion
↑
↑
GIP secretion
↑
This table shows the qualitative changes of plasma glucose and lipids, and of the secretion of insulin, GLP-1 and GIP (gastric inhibitory peptide) in healthy and obese type 2 diabetic subjects fed on a monounsaturated fatty acid rich meal (80 g olive oil) compared to a saturated fatty acid rich meal (100 g butter). ⫽: no change; ↑: significantly higher; ↓: significantly lower.
1217
CHAPTER | 133 Effects of an Olive Oil-enriched Diet on Glucagon-like Peptide-1
133.5 OLIVE OIL DIET IN TYPE 2 DIABETIC RATS In a rat model of type 2 diabetes, obtained by streptozotocin injection during the neonatal period (STZ rats), we observed (Cancelas et al., 2006) that the olive oil diet, when compared to a standard diet, increased the immediate GLP-1 response in meal-trained rats, but decreased after 50 days
the GLP-1 content of the intestinal tract (Table 133.5). Over a period of 50 days, the body weight gain was lower in the rats fed the olive oil diet, as compared to standard diet. In the former animals, however (Table 133.5), no improvement of glucose tolerance or insulin response during an oral glucose tolerance test was observed (Figure 133.5). Thus, in both the rats fed the standard and olive oil diet, a paradoxical lowering of the insulinogenic index, i.e. the paired ratio
TABLE 133.4 Effect of olive oil diet, compared to standard diet, in normal rats (Prieto et al., 2005). Short term
TABLE 133.5 Effect of olive oil diet, compared to standard diet, in streptozotocin-induced type 2 diabetic rats (Cancelas et al., 2006).
Long term
Short term Weight increment
Long term
↑
Plasma glucose
⫽
⫽
Plasma insulin
⫽
⫽
Insulinogenic index
↑
Plasma GLP-1
↑
GLP-1 content in intestine
⫽
Weight increment
Increments of paired basal (% of control)
Plasma glucose
⫽
⫽
Plasma insulin
↑
⫽
Insulinogenic index
↑
⫽
Plasma GLP-1
↑
↓
GLP-1 content in intestine
This table shows the qualitative changes in normal rats of plasma glucose, insulin and GLP-1 in response to an olive oil-enriched meal (short term) compared to a standard meal; it also shows the corresponding changes, in fed state, along the 50 days of olive oil diet (long term) including those of weight increment and GLP-1 intestinal content, compared to rats fed on standard diet. ⫽: no change, : trend to higher; ↑: significantly higher; : much higher.
↓
This table shows the qualitative changes in STZ rats of plasma glucose, insulin and GLP-1, in response to an olive oil-enriched meal (short term) compared to a standard meal; it also shows the corresponding changes, in the fed state, along the 50 days of olive oil diet (long term), including those of weight increment and GLP-1 intestinal content, compared to diabetic rats fed on standard diet. ⫽: no change, ↑: significantly higher; ↓ significantly lower.
Normal rats
250
200
↓
*#
Control diet Olive oil diet
* *
150
*
*
100
#
50
0 Day
19
36 Glucose
19
36 Insulin
19
36
Insulin/glucose
FIGURE 133.4 Effect of long-term administration of olive oil compared to control diet upon the response to intragastric glucose in normal rats. Effect of long-term administration of olive oil ( ), compared with standard diet control ( ), upon the response to intragastric glucose (1.2 mg per g body wt.) in normal rats, expressed as percent of the increments above paired basal value relative to the mean corresponding value of the control diet. Mean ⫾ SEM, n ⫽ 39–45, *: p ⬍ 0.05 vs. control; p ⬍ 0.05 vs. day 19.
1218
SECTION | II
Major Organ Systems Including Liver and Metabolism
Control diet
Absolute values (% of control)
150
Olive oil diet STZ rats
* 100
50
0 Day
19
36
19
Glucose
36 Insulin
19
36
Insulin/glucose
FIGURE 133.5 Effect of long-term administration of olive oil compared to control diet upon the response to intragastric glucose in type 2 diabetic rats. Effect of long-term administration of olive oil ( ), compared with standard diet control ( ), upon the response to intragastric glucose (1.2 mg per g body wt.) in type 2 diabetic rats, expressed as percent of the absolute values relative to the mean corresponding one measured at the same time of the test in animals fed the control diet. Mean ⫾ SEM, n ⫽ 65–72, *: p ⬍ 0.05.
between plasma insulin and glucose concentration, was recorded during the oral glucose tolerance test. Moreover, the insulin content of the pancreas was equally low in the STZ rats fed either the standard or olive oil diet. These findings are discussed in the framework of possible differences in the pathophysiology of β-cell dysfunction in most patients with type-2 diabetes and the present animal model of non-insulin-dependent diabetes.
133.6 COMPARISON BETWEEN NORMAL AND DIABETIC RATS Several of the observations found in the STZ rats (Cancelas et al., 2006) are similar to those collected, with the same experimental device, in normal rats (Prieto et al., 2005). This applies for instance to the changes in body weight during the meal-training period, as well as the higher GLP-1 increase and higher insulin response (relative to the changes in glycemia) when the meal-trained rats were exposed for 15 min to the olive oil diet, as compared to standard diet. Likewise, in the STZ rats, the absence of any obvious difference between the rats given access to the standard or olive oil diet in terms of the measurements of plasma D-glucose, insulin and GLP-1 concentrations made between day 4 and day 50 in fed animals, duplicates, to a large extent, our prior findings in normal rats (Prieto et al., 2005). The most obvious differences between the normal and STZ rats consisted, apart from the perturbation of glucose homeostasis and low insulin content of the pancreas, in a lesser gain in body weight, and a lowering of the GLP-1 content
of the intestinal tract in the STZ rats fed the olive oil, as distinct from standard diet. Indeed, in normal rats, the opposite situation prevails in terms of the changes in body weight and, except in the colon of female animals, no significant reduction in the GLP-1 content of the intestinal tract was observed. When considering the results obtained at the occasion of the oral glucose tolerance tests performed on day 19 and day 36, another analogy was encountered between normal and STZ rats. Thus, in both cases, the tolerance to glucose was lower in the animals fed the olive oil diet, rather than standard diet, on day 19, whilst such was no more the case on day 36. Obvious differences, however, between normal and STZ rats were also present. First, at variance with the situation found in normal rats, the STZ rats failed to display a higher insulin response to the oral glucose load when fed the olive oil, rather than standard, diet. Second, in sharp contrast to the observation made in normal rats, the insulinogenic index displayed a paradoxical decrease during the oral glucose tolerance in the STZ rats. Such a paradoxical change further illustrates the perturbation of the β-cell secretory responsiveness to glucose in the STZ rats (Malaisse, 1991) and the absence of any sizeable improvement of such a perturbation in response to the long-term intake of the olive oil-enriched diet. In the previous study conducted in normal rats (Prieto et al., 2005), emphasis was placed on the known action of monounsaturated fatty acids as GLP-1 secretagogues as a link between exposure to the olive oil diet and changes in insulin secretion observed either during the oral glucose tolerance test performed after overnight starvation or in animals examined in the fed state. The results of the latter study (Cancelas et al., 2006), however, indicate that the same dietary manipulation fails to improve insulin secretion
CHAPTER | 133 Effects of an Olive Oil-enriched Diet on Glucagon-like Peptide-1
and glucose tolerance in STZ rats, considered as an animal model of type 2 diabetes. In conclusion, therefore, and within the limits of the present investigations, no obvious benefit, apart from a reduction in body weight gain, apparently resulted from the long-term administration of an olive oil-enriched diet to STZ rats. Such a conclusion contrasts from that of prior studies conducted in patients with non-insulin-dependent diabetes (Garg et al., 1994; Low et al., 1996). The question comes inevitably to mind of whether such opposite results may be linked to differences in the pathophysiology of type 2 diabetes in most human subjects affected by this disease and the present animal model of β-cell dysfunction.
133.7 OTHER STUDIES IN HUMANS Gastric emptying (GE) is a major determinant of postprandial glycemia. Because the presence of fat in the small intestine inhibits GE, ingestion of fat may attenuate the glycemic response to carbohydrate. A recent study (Gentilcore et al., 2006), performed in six patients with type 2 diabetes, indicated that ingestion of fat before a carbohydrate meal markedly slows GE and attenuates the postprandial rises in glucose, insulin and GIP, but stimulates GLP-1. Most recently (Paniagua et al., 2007), through a study about the effects of different macronutrient composition diets in 11 insulin-resistant subjects (obese and with type 2 diabetes), it was reported that weight maintenance with a monounsaturated-fat-rich diet (olive oil) improved HOMA-ir (homeostasis model analysis-insulin resistance), decreased plasma fasting proinsulin, and postprandial glucose and insulin levels, and increased HDL-C and GLP-1 concentrations as compared with a carbohydrate-rich diet.
SUMMARY POINTS ●
●
●
●
Glucagon-like peptide 1 (GLP-1) may participate in the beneficial effect on glucose homeostasis of diets rich in monounsaturated fatty acids. In meal-trained rats, the increment in GLP-1 plasma concentration is higher in rats given access to an olive oil-enriched diet, rather than control diet. Long-term exposure of normal rats to the olive oil-rich diet increases body weight gain and the secretory response of insulin-producing cells to oral glucose administration, eventually leading to improved glucose tolerance, without affecting the GLP-1 content of the intestinal tract. In a streptozotocin-induced animal model of type 2 diabetes, food intake, plasma insulin concentration and initial increment in plasma GLP-1 concentration are higher when meal-trained animals are exposed for 15 min to the olive oil-enriched, as compared to the control, diet.
●
●
1219
Long-term exposure of these diabetic rats to the olive oil-rich diet decreases body weight gain and lowers the GLP-1 content of the intestinal tract, whilst failing to affect significantly glucose tolerance, plasma insulin concentration and pancreatic insulin content. In diabetic and obese human subjects, however, diets with an increased content of monounsaturated fatty acids improve glycemic and triglyceridemic control.
ACKNOWLEDGMENT We thank the Institute of Health Carlos III (PI 060076 and RD0600150004), Spain, for financial support.
REFERENCES Buteau, J., Foisy, S., Joly, E., Prentki, M., 2003. Glucagon-like peptide 1 induces pancreatic beta-cell proliferation via transactivation of the epidermal growth factor receptor. Diabetes 52, 124–132. Campbell, L.V., Marmot, P.E., Dyer, J.A., Borkman, M., Storlien, L.H., 1994. The high-monounsaturated fat diet as a practical alternative for NIDDM. Diabetes Care 17, 177–182. Cancelas, J., Prieto, P.G., Villanueva-Peñacarrillo, M.L., Valverde, I., Malaisse, W.J., 2006. Effects of an olive oil-enriched diet on glucagonlike peptide 1 release and intestinal content, plasma insulin concentration, glucose tolerance and pancreatic insulin content in an animal model of type 2 diabetes. Horm. Metab. Res. 38, 98–105. Creutzfeldt, W., 2001. The entero-insular axis in type 2 diabetes. Incretins as therapeutic agents. Exp. Clin. Endocr. Diab. 109, S288–S303. Drucker, D.J., 2002. Biological actions and therapeutic potential of the glucagon-like peptides. Gastroenterology 122, 531–544. Egan, J.M., Bulotta, A., Hui, H., Perfetti, R., 2003. GLP-1 receptor agonists are growth and differentiation factors for pancreatic islet beta cells. Diabetes Metab. Res. Rev. 19, 115–123. Garg, A., 1998. High-monounsaturated-fat diets for patients with diabetes mellitus: a meta-analysis. Am. J. Clin. Nutr. 67 (Suppl. 3), 577S–582S. Garg, A., Bantle, J.P., Henry, R.R., Coulston, A.M., Griver, K.A., Raatz, S.K., Brinkley, L., Chen, Y.D., Grundy, S.M., Huet, B.A., 1994. Effects of varying carbohydrate content of diet in patients with non-insulindependent diabetes mellitus. J. Am. Med. Assoc. 271, 1421–1428. Gentilcore, D., Chaikomin, R., Jones, K.L., Russo, A., Feinle-Bisset, C., Wishart, J.M., Rayner, C.K., Horowitz, M., 2006. Effects of fat on gastric emptying of and the glycemic, insulin, and incretin responses to a carbohydrate meal in type 2 diabetes. J. Clin. Endocrinol. Metab. 91, 2062–2067. Greig, N.H., Mattson, M.P., Perry, T., Chan, S.L., Giordano, T., Sambamurti, K., Rogers, J.T., Ovadia, H., Lahiri, D.K., 2004. New therapeutic strategies and drug candidates for neurodegenerative diseases: p53 and TNF-{alpha} inhibitors, and GLP-1 receptor agonists. Ann. N. Y. Acad. Sci. 1035, 290–315. Holst, J.J., 1997. Enteroglucagon. Annu. Rev. Physiol. 59, 257–271. Low, C.C., Grossman, E.B., Gumbiner, B., 1996. Potentiation of effects of weight loss by monounsaturated fatty acids in obese NIDDM patients. Diabetes 45, 569–575. Lugari, R., Dei Cas, A., Ugolotti, D., Barilli, A.L., Camellini, C., Ganzerla, G.C., Luciani, A., Salerni, B., Mittenperger, F., Nodari, S., Gnudi, A., Zandomeneghi, R., 2004. Glucagon-like peptide 1 (GLP-1) secretion
1220
SECTION | II
and plasma dipeptidyl peptidase IV (DPP-IV) activity in morbidly obese patients undergoing biliopancreatic diversion. Horm. Metab. Res. 36, 111–115. Malaisse, W.J., 1991. Physiology of insulin secretion and its alteration in diabetes: the concept of glucotoxicity. In: Adreani, D., Gueriguian, J.L., Striker, G.E. (eds), Diabetic Complications: Epidemiology and Pathogenic Mechanisms. Raven Press, New York, pp. 73–92. Meier, J.J., Gallwitz, B., Schmidt, W.E., Nauck, M.A., 2002. Glucagonlike peptide 1 as a regulator of food intake and body weight: therapeutic perspectives. Eur. J. Pharmacol. 440, 269–279. Nauck, M.A., Heimesaat, M.M., Orskov, C., Holst, J.J., Ebert, R., Creutzfeldt, W., 1993. Preserved incretin activity of glucagon-like peptide 1 [7–36 amide] but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. J. Clin. Invest. 91, 301–307. Nikolaidis, L.A., Mankad, S., Sokos, G.G., Miske, G., Shah, A., Elahi, D., Shannon, R.P., 2004. Effects of glucagon-like peptide-1 in patients with acute myocardial infarction and left ventricular dysfunction after successful reperfusion. Circulation 2, 962–965. Paniagua, J.A., de la Sacristana, A.G., Sánchez, E., Romero, I., VidalPuig, A., Berral, F.J., Escribano, A., Moyano, M.J., Peréz-Martinez, P., López-Miranda, J., Pérez-Jiménez, F., 2007. A MUFA-rich diet improves posprandial glucose, lipid and GLP-1 responses in insulinresistant subjects. J. Am. Coll. Nutr. 26, 434–444. Parrillo, M., Rivellese, A.A., Ciardullo, B., Giasso, A., Genovese, S., Ricardo, G., 1992. A high-monounsaturated-fat/low-carbohydrate diet improves peripheral insulin sensitivity in non-insulin-dependent diabetic patients. Metabolism 41, 1373–1378. Perry, T., Greig, N.H., 2005. Enhancing central nervous system endogenous GLP-1 receptor pathways for intervention in Alzheimer’s disease. Curr. Alzheimer Res. 2, 377–385. Pohl, M., Mayr, P., Mertl-Roetzer, M., Lauster, F., Lerch, M., Eriksen, J., Haslbeck, M., Rahlfs, V.W., 2005. Glycaemic control in type II diabetic tube-fed patients with a new enteral formula low in carbohydrates and high in monounsaturated fatty acids: a randomised controlled trial. Eur. J. Clin. Nutr. 59, 1221–1232. Prieto, P.G., Cancelas, J., Villanueva-Peñacarrillo, M.L., Valverde, I., Malaisse, W.J., 2005. Effects of an olive oil-enriched diet on plasma GLP-1 concentration and intestinal content, plasma insulin concentration, and glucose tolerance in normal rats. Endocrine 26, 107–115. Riccardi, G., Rivellese, A.A., 2000. Dietary treatment of the metabolic syndrome – the optimal diet. Br. J. Nutr. 83, S143–S148.
Major Organ Systems Including Liver and Metabolism
Rocca, A.S., Brubaker, P.L., 1995. Stereospecific effects of fatty acids on proglucagon-derived peptide secretion in fetal rat intestinal cultures. Endocrinology 136, 5593–5599. Rocca, A.S., La Greca, J., Kalitsky, J., Brubaker, P.L., 2001. Monounsaturated fatty acid diets improve glycaemic tolerance through increased secretion of glucagon-like peptide 1. Endocrinology 142, 1148–1155. Rodríguez-Villar, C., Pérez-Heras, A., Mercadé, I., Casals, E., Ros, E., 2004. Comparison of a high-carbohydrate and a high-monounsaturated fat, olive oil-rich diet on the susceptibility of LDL to oxidative modification in subjects with Type 2 diabetes mellitus. Diabet. Med. 21, 142–149. Strychar, I., Ishac, A., Rivard, M., Lussier-Cacan, S., Beauregard, H., ArisJilwan, N., Radwan, F., Yale, J.F., 2003. Impact of a high-monounsaturated-fat diet on lipid profile in subjects with type 1 diabetes. J. Am. Diet. Assoc. 103, 467–474. Thomsen, C., Rasmussen, O., Lousen, T., Holst, J.J., Fenselaou, S., Schrezenmeir, J., Hermansen, K., 1999. Differential effects of saturated and monounsaturated fats on post-prandium lipemia and incretin responses in healthy subjects. Am. J. Clin. Nutr. 69, 1135–1143. Thomsen, C., Storm, H., Holst, J.J., Hermansen, K., 2003. Differential effects of saturated and monounsaturated fats on post-prandium lipemia and glucagon-like peptide 1 responses in patients with type 2 diabetes. Am. J. Clin. Nutr. 77, 605–611. Thorens, B., 1992. Expression cloning of the pancreatic beta cell receptor for the gluco-incretin hormone glucagon-like peptide 1. Proc. Natl. Acad. Sci. U.S.A. 89, 8641–8645. Valverde, I., Puente, J., Martín-Duce, A., Molina, L., Lozano, O., Sancho, V., Malaisse, W.J., Villanueva-Peñacarrillo, M.L., 2005. Changes in glucagonlike peptide-1 (GLP-1) secretion after biliopancreatic diversion or vertical banded gastroplasty in obese subjects. Obes. Surg. 15, 387–397. Valverde, I., Villanueva-Peñacarrillo, M.L., Malaisse, W.J., 2002. Pancreatic and extrapancreatic effects of GLP-1. Diabetes Metab. 28 3S85–3S89. Vilsbøll, T., Krarup, T., Deacon, C.F., Madsbad, S., Holst, J.J., 2001. Reduced postprandial concentrations of intact biologically active glucagon-like peptide 1 in type 2 diabetic patients. Diabetes 50, 609–613. Wright, J., 1998. Effect of high-carbohydrate versus high-monounsaturated fatty acids diet on metabolic control in diabetes and hyperglycaemic patiens. Clin. Nutr. 17 (Suppl. 2), 35–45. Yang, H., Egan, J.M., Wang, Y., Moyes, C.D., Roth, J., Montrose, M.H., Montrose-Rafizadeh, C., 1998. GLP-1 action in L6 myotubes is via a receptor different from the pancreatic GLP-1 receptor. Am. J. Physiol. 275, C675–C683.
Section 3
Specific Components of Olive Oil and Their Effects on Tissue and Body Systems
This page intentionally left blank
3.1
Tyrosol and Hydroxytyrosol
This page intentionally left blank
Chapter 134
The Chemistry of Tyrosol and Hydroxytyrosol: Implications for Oxidative Stress Alessandra Napolitano, Maria De Lucia, Lucia Panzella and Marco d’Ischia Department of Organic Chemistry and Biochemistry, University of Naples Federico II, Naples, Italy
134.1 INTRODUCTION
OH
Tyrosol (or (2-hydroxyethyl)phenol, p-hydroxyphenethyl alcohol, or 2-(4-hydroxyphenyl)ethanol) (TYR) and hydroxytyrosol (or 4-(2-hydroxyethyl)-1,2-benzenediol, or 3hydroxytyrosol, or 2-(3,4-dihydroxyphenyl)ethanol) (HTYR) are the most representative phenolics of olive fruits and olive oil where they occur as such or in the form of esters of the secoiridoid elenolic acid. TYR is a colorless solid at room temperature, melting at 91–92°C, boiling at 158°C at 4 Torr, and slightly soluble in water. HTYR appears as a clear colorless liquid exhibiting a solubility in water of 5 g 100 mL⫺1 (25°C). The higher solubility in organic solvents of TYR with respect to HTYR is shown by the partitioning coefficients between oil and water phases determined as 0.077 and 0.010 for TYR and HTYR, respectively (Rodis et al., 2002). From the structural viewpoint, these compounds share a phenolic functionality substituted at the para position with a hydroxyethyl chain. HTYR has an additional OH group on the benzene moiety at the position next to the other OH group and is therefore an ortho diphenol or catechol. Such structural modification results in dramatic differences in the susceptibility to oxidation and antioxidant power as well as in the potency of their chemopreventive efficacy under oxidative stress conditions (Figures 134.1 and 134.2). Table 134.1 summarizes the main facts characterizing such processes associated to the onset and development of a number of diseases.
134.2 PROPERTIES The scavenging potential of TYR and HTYR towards oxygen and nitrogen reactive species, including hydroxyl Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
HO
HO
OH
HO TYR
HTYR
FIGURE 134.1 Structures of TYR and HTYR: shown are the molecular structures of the main virgin olive oil phenolics TYR and HTYR. TYR: tyrosol; HTYR: hydroxytyrosol.
O
OH
O
HO
OH
O HTYR quinone
HTYR quinone methide
FIGURE 134.2 HTYR o-quinone and quinone methide tautomers. The tautomeric equilibrium of the o-quinone of HTYR with the quinone methide form is shown. HTYR: hydroxytyrosol.
radical, peroxynitrite, superoxide radical, hydrogen peroxide and hypochlorous acid, has recently been determined (Rietjens et al., 2007). In most cases TYR does not exhibit any activity, whereas HTYR scavenging ability is comparable to that of oleuropein and catechol. Table 134.2 reports literature data for the antioxidant properties as determined by various assays (Stupans et al., 2002; Briante et al., 2003; Valavanidis et al., 2004; Roche et al., 2005). Although the values obtained by different groups are not in fairly good agreement, 2,2-diphenyl-1-pycrylhydrazyl (DPPH) and N,N-dimethyl-p-phenylenediamine (DMPD) radical scavenging tests carried out in different media concur to indicate a marked H-donor ability of HTYR even higher than that of alpha-tocopherol, while TYR is far less effective. This trend is maintained in lipid peroxidation assays although in this case HTYR activity is one order of magnitude lower than the reference compound alpha-tocopherol. Data of superoxide and hydroxyl radical
1225
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1226
SECTION | III
Tyrosol and Hydroxytyrosol
TABLE 134.1 Key features of oxidative stress. ●
The term ‘oxidative stress’ indicates an overproduction of ROS and RNS accompanied by a deficiency of enzymatic and nonenzymatic antioxidants that causes a cellular redox imbalance
●
The excess ROS can damage cellular lipids, proteins or DNA inhibiting their normal function. Because of this, oxidative stress has been implicated in a number of human pathologies like cancer, cardiovascular and neurodegenerative diseases as well as in the aging process
●
ROS/RNS-induced DNA damage involves single- or double-stranded DNA breaks, purine, pyrimidine, or deoxyribose modifications, and DNA cross-links
●
In addition to ROS, redox metals, due to their ability to generate free radicals, or non-redox metals, due to their ability to bind to critical thiols, have been implicated in the oxidative stress
●
Radical-mediated damage to cellular biomembranes results in lipid peroxidation, a process that generates reactive electrophiles such as epoxides and aldehydes
●
The major aldehyde products of lipid peroxidation malondialdehyde and 4-hydroxynonenal are strongly mutagenic
●
It has been demonstrated that ROS interfere with the expression of a number of genes and signal transduction pathways. The activation of transcription factors including MAPK/AP-1 by ROS has a direct effect on cell proliferation and apoptosis
This table lists the key facts of oxidative stress including causes and main biological effects. ROS: reactive oxygen species; RNS: reactive nitrogen species; MAPK/AP-1: mitogen-activated protein kinase associated protein-1.
TABLE 134.2 Antioxidant properties of TYR and HTYR. Cmpd
DPPH scavenging
[Inhibition (%)] (Stupans et al., 2002) TYR
0
HTYR
47 ⫾ 3
alpha-tocopherol
n.a.
Linoleic acid peroxidation inhibition (Roche et al., 2005)
DMPD scavenging (Briante et al., 2003)
Superoxide anion scavenging (Valavanidis et al., 2004)
Hydroxyl radical scavenging (Valavanidis et al., 2004)
IC50 (μM)
IC50 (μM)
IC50 (μM)
IC50 (μM)
n.a.
0
5.4 ⫾ 0.5
15 ⫾ 2.5
0.25 ⫾ 0.02
2.53 ⫾ 0.40
1.56 ⫾ 0.16
1.4 ⫾ 0.2
3.5 ⫾ 0.8
4.8 ⫾ 0.3
0.18 ⫾ 0.02
n.a.
n.a.
n.a.
IC50 (μM) (Valavanidis et al., 2004) 5.8 ⫾ 0.5
Literature data for the antioxidant properties of TYR and HTYR as determined by various assays in comparison with alpha-tocopherol are summarized. Percent inhibition was determined on test compounds at 10.0 μM. n.a., not available; TYR: tyrosol; HTYR: hydroxytyrosol; DPPH: 2,2-diphenyl-1-pycrylhydrazyl; DMPD: N,N-dimethyl-p-phenylenediamine.
scavenging abilities as determined by electron paramagnetic resonance (EPR) methodologies (based on the inhibition of the DMPO (5,5-dimethyl-1-pyrroline-N-oxide)-OOH/-OH spin adduct formation) are more comparable for the two phenols. This trend is expected considering that in this case the reactivity reflects not only the ease to oxidation via H-atom abstraction but also the activation of the benzene ring towards the addition of oxygen radical species.
The marked antioxidant properties of HTYR are also exemplified by its ability to inhibit the copper-sulfateinduced oxidation of low-density lipoproteins as shown by the reduced short-chain aldehydes formation, sparing of vitamin E and decrease in the levels of malondialdehydelysine and 4-hydroxynonenal-lysine adducts indicating protection of the apoprotein layer (Visioli et al., 1995). Transition metals such as copper and iron are implicated in
CHAPTER | 134 The Chemistry of Tyrosol and Hydroxytyrosol: Implications for Oxidative Stress
the production of highly reactive hydroxyl radicals by the Fenton reaction as well as in the reductive decomposition of lipid hydroperoxides to yield alkoxyl or peroxyl radicals as chain propagators. Thus, the protective effects of HTYR may be due to the copper-chelating and -reducing properties, since metal chelation is an effective prevention means of the lipid peroxidation either by sequestering metal ions into inert complexes unable to decompose hydrogen peroxide or by restricting the access of metal ions toward lipid hydroperoxides. Although HTYR is a more potent metal cation scavenger than TYR (Briante et al., 2003), it displays a lower reducing capacity toward Cu(II); both compounds are poor reductants relative to other vegetable phenols as flavonols and isoflavones. Since Cu(I) ions are more reactive in decomposing hydroperoxides than cupric ions, this entails that olive oil phenolics could exert lower pro-oxidant effects in promoting Fenton or Haber-Weiss reactions.
134.3 COMPUTATIONAL STUDIES The structural characteristics accounting for the marked differences in the antioxidant properties have been addressed at the theoretical level (Leopoldini et al., 2004; Li et al., 2007). The bond dissociation enthalpies and ionization potentials were calculated by different authors using density functional theory (DFT) methods to evaluate the H-atom and electron-donating abilities. Lower bond dissociation enthalpy (BDE) values were found for HTYR with respect to TYR with an average value of the difference of 8 kcal mol⫺1. This suggests that the catechol functionality is a major determinant of the weakening of the OH bond. Less marked were the differences in the ionization potentials between the phenol and o-diphenol systems. Spin density determination of the phenoxy radicals from HTYR showed an extensive delocalization of the odd electron over the whole molecule with stabilization afforded by hydrogen bond interactions.
134.4 PREPARATION Very few syntheses of TYR (Zheng et al., 2002) are reported in the literature since its isolation from natural sources and particularly agricultural wastewaters is considered a more convenient access route. By contrast HTYR is more difficult to obtain from plant waste (De Marco et al., 2007) and several synthetic procedures have been developed. High yields (⬎80%) are obtained by reduction of 3,4-dihydroxyphenylacetic acid with LiAlH4 (Capasso et al., 1999). TYR is used as the starting material in several procedures including the hydrogen peroxide photocatalytic oxidation in the presence of iron-containing heterogeneous catalyst (Azabou et al., 2007) (64% yield) or the regioselective hydroxylation by
1227
o-iodoxybenzoic acid followed by reduction of the oquinone (De Lucia et al., 2006). The latter represents a straightforward and easy-to-do method but affords HTYR in rather low yields (average 30%). A 5-step conversion of TYR to HTYR has recently been reported (Bovicelli et al., 2007) which appears a versatile yet laborious method. This also allows access to long-chain aliphatic esters of HTYR, a group of compounds which have attracted considerable attention for their superior antioxidant properties (Torres de Pinedo et al., 2007). Other procedures have been evaluated for industrial applications. Tyrosinase oxidation of TYR in the presence of ascorbic acid is a clean reaction which may be employed in industrial bioreactors (Espìn et al., 2001). Pseudomonas aeruginosa immobilized resting cells have also been exploited for the bioconversion of TYR affording HTYR in about 86% yields (Bouallagui and Sayadi, 2006). Halophiles bacteria Halomonas sp. strain HTB24 isolated from olive mill wastewater proved also able to convert TYR into HTYR and dihydroxyphenyl acetic acid but the actual synthetic potential of such a procedure remains to be assessed (Liebgott et al., 2007).
134.5 OXIDATION REACTIVITY Despite extensive studies on the antioxidant and free radical scavenging properties of olive oil phenolics, the nature of the products formed during such reactions has been relatively poorly investigated. TYR is difficult to oxidize even when exposed to strong oxidants such as periodate or the Fenton system (Fe(II)/ EDTA/H2O2), under which conditions other o-diphenols including oleuropein are rapidly consumed (Antolovich et al., 2004). No product resulting from C–C or C–O radical coupling, as would be expected typically for a phenol compound, has so far been reported. HTYR is more prone to oxidation. HPLC-MS analysis of the periodate oxidation mixture (Roche et al., 2005) provided evidence for the formation of a dimer (2 ⫻ M-4 H, m.w. 304 Da), another product resulting from dimerization with addition of a molecule of water (2 ⫻ M ⫹ H2O-2 H, m.w. 320 Da) and a monomer on which the addition of the elements of water had occurred (M ⫹ H2O-2 H, m.w. 168 Da). For all these compounds speculative structures were proposed, based on the electrophilic reactivity of the quinone methide of HTYR at tautomeric equilibrium with the o-quinone. Spectral characterization of HTYR oxidation products obtained under different but related conditions led to unambiguous structural identification of these products. Oxidation of hydroxytyrosol by tyrosinase, in air or by ferricyanide in phosphate buffer at pH 7.4 led to a couple of products at 1:1 ratio (HPLC analysis) with a significant
1228
SECTION | III
conversion of the starting catechol (⬎80%). These were identified as regioisomeric dimers possessing the unusual 1,2-dihydro-4,12a-methanooxocino[4,5-b]-[1,4]benzodioxin-5-one skeleton shown in Figure 134.3 (Vogna et al., 2003). Interestingly, these isomers have the same molecular weight of the dimer identified in the periodate oxidation mixture of HTYR (Roche et al., 2005). The methanooxocinobenzodioxinone system is sufficiently stable also under the oxidative conditions of the reaction. It appears therefore to be an end-product of the oxidation reaction of hydroxytyrosol and it can be expected that it is not further converted under those conditions such as exposure to air that can be operative in processing and storing of food added with HTYR. A much more complex reaction behavior of HTYR is observed with oxidizing systems of higher efficiency such as the peroxidase/H2O2 (De Lucia et al., 2006). A redbrown mixture of species is obtained with a complete consumption of the substrate. Analysis of the main components of these complex mixtures was possible only after reduction and acetylation treatment. This allowed identification of a hydroxylated HTYR derivative and a dimer formulated as a pentahydroxybiphenyl derivative (Figure 134.4).
O
R1
O OH
R2
O
O
a: R1 = H, R2 = − CH2CH2OH b: R1 = − CH2CH2OH, R2 = H Benzodioxinones FIGURE 134.3 Structures of the isomeric dimers formed by oxidation of HTYR under mild conditions. Molecular structures of the regioisomeric methanooxocinobenzodioxinone compounds obtained by the oxidation of HTYR by tyrosinase, in air or by ferricyanide in phosphate buffer at pH 7.4. HTYR: hydroxytyrosol.
HO HO
OH OH
OH
HO OH
HO
Hydroxylated HTYR
OH HO OH Pentahydroxybiphenyl
FIGURE 134.4 Structures of the products obtained by peroxidase/ H2O2 oxidation of HTYR. Molecular structures of the hydroxylated HTYR derivative and of the pentahydroxybiphenyl compound obtained by the oxidation of HTYR by peroxidase/H2O2 in phosphate buffer at pH 7.4. Compounds were isolated as O-acetyl derivatives. HTYR: hydroxytyrosol.
Tyrosol and Hydroxytyrosol
134.6 MECHANISMS OF THE OXIDATION REACTIONS Overall, the oxidation chemistry of HTYR appears to be dictated by the reactivity of its o-quinone (Figure 134.5). This is a rather unstable species which, however, is persistent enough in organic solvents as under the conditions of the o-iodoxybenzoic-acid-mediated oxidation of TYR. Its formation may be followed spectrophotometrically by monitoring the characteristic yellow-orange chromophore at 390 nm. Depending on the reaction conditions this electrophilic quinone can undergo self coupling by sequential addition of the hydroxyl groups of HTYR leading to a benzodioxin which further evolves with cyclization of the 2-hydroxyethyl side chain (route a in Figure 134.5). Alternatively, in the presence of nucleophilic agents, such as hydrogen peroxide, the 2-hydroxy-p-quinone is formed by carbonyl forming decomposition of the hydroperoxide adduct as shown in route b of Figure 134.5. The actual involvement of hydrogen peroxide in the oxygenation process was shown by ad hoc experiments in which a model quinone was allowed to react with hydrogen peroxide added to the reaction medium leading to a hydroxylated p-quinone in high yields. Formation of a 2-hydroxy-p-quinone in the oxidation reaction of HTYR is of particular interest as this species is fairly stable, highly acidic (pKa 2.90 for structurally related compounds (De Lucia et al., 2006)) and features a red chromophore with a maximum at 490 nm. Differently from the majority of ortho and para quinones, the 2hydroxy-p-quinones exhibit a marked nucleophilicity and may attack the electrophilic sites of other quinones such as HTYR quinone. Such reactivity was confirmed by isolation of a dimer made up of a HTYR and a hydroxylated HTYR as a major component of the oxidation mixture. In separate experiments it was possible to confirm that the pentahydroxybiphenyl compound resulted from reaction between two quinone species involving the 3-position of hydroxy-pquinone, possessing enolate character, and the electrophilic 2-position of HTYR quinone. Addition of other nucleophiles is also likely and in the case of histidine a product isolated from Manduca sexta (tobacco hornworm) pupal cuticle was identified by mass spectrometry and was formulated as the HTYR–histidine adduct 6-(N-1⬘)-histidyl-2-(3,4-dihydroxyphenyl)ethanol (Kerwin et al., 1999). A conjugate of HTYR with glutathione was described as one of the biotransformation products of dietary HTYR formed in the gastrointestinal tract; yet this product, which was also obtained by mushroom tyrosinase oxidation of HTYR in the presence of glutathione, was not fully characterized from a structural viewpoint (Corona et al., 2006). A glutathione-hydroxytyrosol adduct was also described as phase I metabolite found in the urine of women assuming dietary supplements containing Cimicifuga racemosa (Actaea racemosa; black cohosh) (Johnson and van Breemen, 2003).
1229
CHAPTER | 134 The Chemistry of Tyrosol and Hydroxytyrosol: Implications for Oxidative Stress
OH
HO HO OH
HO HO
N
HO
HTYR HRP/H2O2 (Route b)
N COOH
HO
Tyrosinase/O2 (Route a)
His
NH2 HTYR histidine adduct
OH
SG HTYR-glutathione adduct
Glutathione (GSH) OH
O
R1, R2 = H, –CH2CH2OH
O HTYR quinone b OH
O –O
HTYR
OOH H
HO
–H2O
HO
a
HO2–
HO
R1
OH
R2
O
OH OH
[O]
O HO
OH NaBH4
O –O
O
R1
OH
O
O
O
Hydroxylated HTYR R2
OH O O
HO R1
O
O
O
O
OH
O –O
O
NaBH4
R2 Pentahydroxybiphenyl
OH HO OH
Benzodioxinones
FIGURE 134.5 Main reaction routes of the o-quinone from HTYR. Shown are the possible reaction pathways from the o-quinone of HTYR involving dimerization with the starting catechol, addition of different physiologically relevant nucleophiles and the failure to undergo intramolecular cyclization. HTYR: hydroxytyrosol.
Self cyclization of the HTYR quinone would also be a conceivable reaction path but formation of a benzodihydrofuran product has not so far been reported in the oxidation of HTYR.
134.7 REACTIVITY TOWARDS REACTIVE NITROGEN SPECIES Another interesting aspect of the chemistry of TYR and HTYR concerns their ability to react with nitric-oxidederived species. These species are important mediators and contributory factors in the inflammatory response and in carcinogenesis. The major physiologic metabolite of nitric oxide is nitrite ( NO− 2 ) which is present at high levels
(30–210 μM) in saliva and is also found in polluted drinking waters, vegetables, fertilizers and preserved/pickled meats. Within the stomach and other acidic compartments, supporting nitrous acid (HNO2) formation NO− 2 may cause nucleobase deamination and interstrand cross-link formation and production of mutagenic N-nitrosamines (Mirvish, 1995). Therefore characterization of this reactivity is of central relevance to the mechanisms of the chemopreventive action of these compounds. HTYR was shown to be active as a scavenger of NO generated spontaneously by the decomposition of sodium nitroprusside and of chemically generated peroxynitrite, as detected by α1-antiproteinase inactivation assays (67.2– 92.4% decrease at 1 mM) (De la Puerta et al., 2001). HTYR is also highly protective against the peroxynitrite-dependent
1230
SECTION | III
nitration of tyrosine and DNA damage (Deiana et al., 1999). By contrast TYR is less active in most of the above tests. Exposure of extra virgin olive oil to acidic nitrite under mild conditions close to those occurring during digestion, viz. 25–500 μM nitrite at pH 3 at 37°C for 1 h, results in an almost complete consumption of HTYR and the related oleuropein (aglycon decarboxymethyl dialdehydic form) (Napolitano et al., 2004). In the case of TYR and the structurally related ligstroside (aglycon decarboxymethyl dialdehydic form), a decay around 40% is obtained with 500 μM nitrite. The outcome of these reactions is usually conversion to nitrated derivatives (Figure 134.6). A plausible reaction mechanism involves radical coupling of NO2 with phenoxy radicals generated by interaction of NO2 or HNO2 with the phenol system. Alternatively, disproportionation of the semiquinone radicals to the corresponding o-quinones followed by nucleophilic attack of nitrite ions may be a feasible reaction path at those pH values, e.g. 3, at which the concentration of NO− 2 is sufficiently high. The mechanisms depicted in Figure 134.6 represent an oversimplification of the many possible reaction routes, since other non-polar species derived from nitrous acid decomposition, e.g. N2O3, which would preferentially partition into the organic phase, are also likely to be involved. This entails that varying oil-to-water ratios may lead to changes in product patterns, a situation which is relevant to the physiological setting of digestion. The presence of phenolic and o-diphenolic components in virgin olive and their reactivity toward species generated by acidic nitrite account also for the marked properties of
Tyrosol and Hydroxytyrosol
extra virgin olive oil as antinitrosating agent. These were evaluated by the fluorimetric diaminonaphthalene assay at pH 4.0 with a 70% inhibition being observed at 1:2 v/v ratio of the oil phase with respect to the aqueous phase (Napolitano et al., 2004).
134.8 OVERVIEW OF CONCEPTS AND IMPLICATIONS FOR OXIDATIVE STRESS Extra virgin olive oil phenolics HTYR and TYR may be active as antioxidants by a number of potential pathways relating to those seen in the preceding sections. Radical scavenging involving hydrogen atom donation to reactive oxygen species or lipid peroxyl radicals represents a main route available to both TYR and HTYR. However, significant differences in the H donor properties of the two compounds have been determined by experimental measurements and BDE calculations with the o-diphenol HTYR exhibiting a significantly higher facility to H-atom donation. Another mechanism by which antioxidants may exert their effects is metal chelation, particularly Cu2⫹ and Fe2⫹ ions which are able to promote peroxyl/hydroxyl formation by decomposition of hydroperoxides by the Fenton and Haber-Weiss reactions. The ability of HTYR to chelate ferric ions accounts for the marked inhibition of hydroperoxide formation promoted by enzymatic and non-enzymatic process including lactoferrin and iron in liposomes and oil-in-water emulsions (Medina et al., 2002) or hemoglobin,
R3
HO
R2
R1 [O] path b: R1, R3 = H, R2 = CH2CH2OH
path a: R1 = OH, R2 = H, R3 = CH2CH2OH R3
.O
... R1
O O2N
b H
OH
R2
x2
a
NO2
NO2
OH
O O NO–2
OH
O HO O2N
HO
H NO2
OH
HO HO
NO2
OH
FIGURE 134.6 Reaction routes of TYR and HTYR with RNS generated by acidic nitrite. Alternative reaction pathways leading to nitration of the phenolic ring of TYR and HTYR involving one-electron oxidation and radical coupling with NO2 or two electron oxidation to the o-quinone and nucleophilic nitrite ion addition. TYR: tyrosol; HTYR: hydroxytyrosol; RNS: reactive nitrogen species.
CHAPTER | 134 The Chemistry of Tyrosol and Hydroxytyrosol: Implications for Oxidative Stress
enzymatic NADH-iron and non-enzymatic ascorbate-iron (Pazos et al., 2006). Free iron sequestering by HTYR may therefore represent an important route by which this compound exerts the documented beneficial effects towards those pathological conditions and chronic diseases associated with oxidative stress (Hashimoto et al., 2004). In this connection, reaction of HTYR quinone with hydrogen peroxide to give the stable 2-hydroxy-p-quinone may also be ranked among the protective mechanisms of HTYR against hydrogen-peroxide-dependent toxicity mechanisms. In addition to the antioxidant properties, the efficient scavenging effects towards reactive nitrogen species should be taken into account to interpret the protective effects of HTYR both in vitro and in vivo systems. Elevated oxidative and nitrosative stress both impair the integrity and functioning of brain tissue, especially in aging. HTYR rich extracts were shown to attenuate Fe2⫹- and nitric-oxide-induced cytotoxicity in murine-dissociated brain cells (Schaffer et al., 2007). Oral HTYR intake in mice produced enhanced resistance of dissociated brain cells to oxidative stress, as shown by reduced basal and stress-induced lipid peroxidation and moderate hyperpolarization of basal mitochondrial membrane potential, an effect suggestive of cytoprotection (Schaffer et al., 2007). The elevated antinitrosating effects shown by extra virgin olive oil phenolics would suggest that inclusion of such compounds in a nitrite-rich diet may represent a strategy to lower the impact of nitrosation reactions on DNA base deamination and carcinogenic N-nitrosamine formation (Mirvish 1995).
134.8 PERSPECTIVES Overall data on the antioxidant and antinitrosating properties of HTYR concur in indicating the high potential of this phenol in chemopreventive dietary strategies like those aimed at controlling tumor-initiating events. The relative stability of the end-products of oxidation endows HTYR with one of the fundamental prerequisites of valuable antioxidants. The quest for novel HTYR derivatives with improved antioxidant and antinitrosating properties for possible exploitation as food additives or for the preparation of functional foods is an active research issue. One of the leading strategies relies on preparation of more lipophylic compounds by esterification of the alcoholic function with long-chain acids which may expectedly partition more favorably into oil and lipid matrices or exhibit higher affinity for the cellular membrane phospholipids competing efficiently with alpha-tocopherol (Torres de Pinedo et al., 2007). Alternative approaches based on substantial modifications of the catechol ring system in order to enhance the H-donor properties and stabilize the oxidation products seem promising as well. The design of these compounds
1231
may likely be guided by the information deriving from theoretical computations of structural effects on BDE values in model phenols.
SUMMARY POINTS ● ●
●
● ● ●
●
●
●
HTYR is more efficient as antioxidant than TYR. HTYR is able to chelate but not to reduce iron and copper ions. H-donor ability of HTYR is higher than TYR and alpha-tocopherol. TYR is not oxidized even under forcing conditions. HTYR oxidation leads to a reactive o-quinone. Under mild conditions HTYR is oxidized to stable benzodioxinone dimers. In the presence of hydrogen peroxide HTYR is oxidized to 2-hydroxy-p-benzoquinone and pentahydroxybiphenyl derivatives. HTYR and to a lesser extent TYR react with reactive nitrogen species. In the presence of acidic nitrite HTYR and TYR as well as the related oleuropein and ligstroside lead to nitrated derivatives.
REFERENCES Antolovich, M., Bedgood, D.R., Bishop, A.G., Jardine, D., Prenzler, P.D., Robards, K., 2004. LC-MS investigation of oxidation products of phenolics antioxidant. J. Agric. Food Chem. 52, 962–971. Azabou, S., Najjar, W., Ghorbel, A., Sayadi, S., 2007. Mild photochemical synthesis of the antioxidant hydroxytyrosol via conversion of tyrosol. J. Agric. Food Chem. 55, 4877–4882. Bouallagui, Z., Sayadi, S., 2006. Production of high hydroxytyrosol yields via tyrosol conversion by Pseudomonas aeruginosa immobilized resting cells. J. Agric. Food Chem. 54, 9906–9911. Bovicelli, P., Antonioletti, R., Mancini, S., Causio, S., Borioni, G., Ammendola, S., Barontini, M., 2007. Expedient synthesis of hydroxytyrosol and its esters. Synthetic Commun 37, 4245–4252. Briante, R., Febbraio, F., Nucci, R., 2003. Antioxidant properties of low molecular weight phenols present in the Mediterranean diet. J. Agric. Food Chem. 51, 6975–6981. Capasso, R., Evidente, A., Avolio, S., Solla, F., 1999. A highly convenient synthesis of hydroxytyrosol and its recovery from agricultural waste waters. J. Agric. Food Chem. 47, 1745–1748. Corona, G., Tzounis, X., Dessi, A., Deiana, M., Debnam, E.S., Visioli, F., Spencer, J.P.E., 2006. The fate of olive oil polyphenols in the gastrointestinal tract: implications of gastric and colonic microflora-dependent biotransformation. Free Radic. Res. 40, 647–658. Deiana, M., Aruoma, O.I., Bianchi, M.P., Spencer, J.P.E., Kaur, H., Halliwell, B., Aeschbach, R., Banni, S., Dessi, M.A., Corongiu, F.P., 1999. Inhibition of peroxynitrite dependent DNA base modification and tyrosine nitration by the extra virgin olive oil-derived antioxidant hydroxytyrosol. Free Radic. Biol. Med. 26, 762. De la Puerta, R., Dominguez, M.E.M., Ruiz-Gutierrez, V., Flavill, J.A., Hoult, J.R.S., 2001. Effects of virgin olive oil phenolics on scavenging
1232
of reactive nitrogen species and upon nitrergic neurotransmission. Life Sci. 69, 1213–1222. De Lucia, M., Panzella, L., Pezzella, A., Napolitano, A., d’Ischia, M., 2006. Oxidative chemistry of the natural antioxidant hydroxytyrosol: hydrogen peroxide-dependent hydroxylation and hydroxyquinone/ o-quinone coupling pathways. Tetrahedron 62, 1273. De Marco, E., Savarese, M., Paduano, A., Sacchi, R., 2007. Characterization and fractionation of phenolic compounds extracted from olive oil mill wastewaters. Food Chem. 104, 858–867. Espìn, J.C., Soler-Rivas, C., Cantos, E., Tomás-Barberán, F.A., Wichers, H.J., 2001. Synthesis of the antioxidant hydroxytyrosol using tyrosinase as biocatalyst. J. Agric. Food Chem. 49, 1187–1193. Hashimoto, T., Ibi, M., Matsuno, K., Nakashima, S., Tanigawa, T., Yoshikawa, T., Yabe-Nishimura, C., 2004. An endogenous metabolite of dopamine, 3,4-dihydroxyphenylethanol, acts as a unique cytoprotective agent against oxidative stress-induced injury. Free Radic. Biol. Med. 36, 555–564. Kerwin, J.L., Turecek, F., Xu, R., Kramer, K.J., Hopkins, T.L., Gatlin, C.L., Yates, J.R., 1999. Mass spectrometric analysis of catechol-histidine adducts from insect cuticle. Anal. Biochem. 268, 229–237. Johnson, B.M., van Breemen, R.B., 2003. In vitro formation of quinoid metabolites of the dietary supplement Cimicifuga racemosa (black cohosh). Chem. Res.Toxicol. 16, 838–846. Leopoldini, M., Marino, T., Russo, N., Toscano, M., 2004. Antioxidant properties of phenolics compounds: H-atom versus electron transfer mechanism. J. Phys. Chem. A 108, 4916–4922. Li, M.-J., Liu, L., Fu, Y., Guo, Q.-X., 2007. Accurate bond dissociation enthalpies of popular antioxidants predicted by the ONIOM-G3B3 method. Theochem 815, 1–9. Liebgott, P.-P., Labat, M., Casalot, L., Amouric, A., Lorquin, J., 2007. Bioconversion of tyrosol into hydroxytyrosol and 3,4-dihydroxyphenylacetic acid under hypersaline conditions by the new Halomonas sp. strain HTB24. FEMS Microbiol. Lett. 276, 26–33. Medina, I., Tombo, I., Satuè-Gracia, M.T., German, J.B., Frankel, E.N., 2002. Effects of natural phenolic compounds on the antioxidant activity of lactoferrin in liposomes and oil-in-water emulsions. J. Agric. Food Chem. 50, 2392–2399. Mirvish, S., 1995. Role of N-nitroso compound (NOC) and N-nitrosation in etiology of gastric, esophageal, nasopharyngeal and bladder cancer and contribution to cancer of known exposures to NOC. Cancer Lett. 93, 17–48.
SECTION | III
Tyrosol and Hydroxytyrosol
Napolitano, A., Panzella, L., Savarese, M., Sacchi, R., Giudicianni, I., Paolillo, L., d’Ischia, M., 2004. Acid-induced structural modifications of unsaturated fatty acids and phenolic olive constituents by nitrite ions: a chemical assessment. Chem. Res. Toxicol. 17, 1329–1337. Pazos, M., Lois, S., Torres, J.L., Medina, I., 2006. Inhibition of hemoglobin- and iron-promoted oxidation in fish microsomes by natural phenolics. J. Agric. Food Chem. 54, 4417–4423. Rietjens, S.J., Bast, A., Haenen, G.R.M.M., 2007. New insights into controversies on the antioxidant potential of the olive oil antioxidant hydroxytyrosol. J. Agric. Food Chem. 55, 7609–7614. Roche, M., Dufour, C., Mora, N., Dangles, O., 2005. Antioxidant activity of olive phenols: mechanistic investigation and characterization of oxidation products by mass spectrometry. Org. Biomol. Chem. 3, 423–430. Rodis, P.S., Karathanos, V.T., Mantzavinou, A., 2002. Partitioning of olive oil antioxidants between oil and water phases. J. Agric. Food Chem. 50, 596–601. Schaffer, S., Podstawa, M., Visioli, F., Bogani, P., Müller, W.E., Eckert, G.P., 2007. Hydroxytyrosol-rich olive mill wastewater extract protects brain cells in vitro and ex vivo. J. Agric. Food Chem. 55, 5043–5049. Stupans, I., Kirlich, A., Tuck, K.L., Hayball, P.J., 2002. Comparison of radical scavenging effect, inhibition of microsomal oxygen free radical generation, and serum lipoprotein oxidation of several natural antioxidants. J. Agric. Food Chem. 50, 2464–2469. Torres de Pinedo, A., Penalver, P., Perez-Victoria, I., Rondon, D., Morales, J.C., 2007. Synthesis of new phenolic fatty acid esters and their evaluation as lipophilic antioxidants in an oil matrix. Food Chem. 105, 657–666. Valavanidis, A., Nisiotou, C., Papageorgiou, Y., Kremli, I., Satravelas, N., Zinieris, N., Zygalaki, H., 2004. Comparison of the radical scavenging potential of polar and lipidic fractions of olive oil and other vegetable oils under normal conditions and after thermal treatment. J. Agric. Food Chem. 52, 2358–2365. Visioli, F., Bellomo, G., Montedoro, G.F., Galli, C., 1995. Low density lipoprotein oxidation is inhibited in vitro by olive oil constituents. Atherosclerosis 117, 25–32. Vogna, D., Pezzella, A., Panzella, L., Napolitano, A., d’Ischia, M., 2003. Oxidative chemistry of hydroxytyrosol: isolation and characterisation of novel methanooxocinobenzodioxinone derivatives. Tetrahedron Lett. 44, 8289–8292. Zheng, H., Gao, W., Ji, X., Zhang, S., 2002. Improved method for synthesis of tyrosol. Zhongguo Yaowu Huaxue Zazhi 12, 166–167.
Chapter 135
Hydroxytyrosol Lipophilic Analogues: Synthesis, Radical Scavenging Activity and Human Cell Oxidative Damage Protection Rosa Chillemi, Sebastiano Sciuto, Carmela Spatafora and Corrado Tringali Dipartimento di Scienze Chimiche, Università di Catania, Viale A. Doria 6, 95125 Catania, Italy
135.1 INTRODUCTION As detailed in other parts of this volume, a variety of studies indicate that the regular dietary consumption of extra virgin olive oil has a protective effect towards coronary heart diseases (Keys, 1995; Tripoli et al., 2005) and breast cancer (Martin-Moreno et al., 1994). Many researches, oriented at identifying the main protective agents of olive oil, have pointed to the antioxidative phenolic constituents (Owen et al., 2000; Leenen et al., 2002; Kok and Kromhout, 2004). The olive oil phenols include tyrosol (1, 4-hydroxyphenethyl alcohol), hydroxytyrosol (2, reported also as 3,4dihydroxyphenethyl alcohol, 3,4-DHPEA; in the following indicated with the acronym HT) (Tuck and Hayball, 2002) and their secoiridoids and conjugate forms: oleuropein (3), ligstroside (4), verbascoside (5). Further conjugates of HT identified in Olea europaea are demethyloleuropein (6), oleuropein aglycon (7), 2-(3,4-dihydroxyphenyl)ethyl ester of elenolic acid dialdehyde (3,4-DHPEA-EDA, 8), hydroxytyrosol 4-β-glucoside (9) and oleuroside (10) (Fernandez-Bolanos et al., 2008). Hydroxytyrosol is incorporated in the aglycon of oleuropein and other conjugates and is released by hydrolysis during olive storage and pressing (Brenes et al., 2001). Among these phenols, hydroxytyrosol is recognized as the main antioxidant and protective principle of virgin olive oil (Fernandez-Bolanos et al., 2008) and is also considered an important anticancer component of this Mediterranean condiment (Manna et al., 1997; Fabiani et al., 2002). In fact, many literature data indicate the potent in vitro antioxidant activity of HT (Saija et al., 1998; Stupans et al., 2002; Roche et al., 2005) in agreement with theoretical predictions on ortho-diphenols (Goupy et al., 2003). In addition, 2 has been proven to prevent oxidative damage in human erythrocytes (Manna et al., 1999). Further studies Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
suggest that 2 has a high oral bioavailability and is easily absorbed, unlike oleuropein (Manna et al., 2000; Tuck et al., 2001; Tuck and Hayball, 2002). Some natural lipophilic analogues of HT are known: hydroxytyrosol acetate (11) is present in olive oil in a percentage around 10% of HT, which has been reported to be in the range 1.5–15 mg kg⫺1 oil (Tuck and Hayball, 2002). The homovanillic alcohol (12) has been identified as a lipophilic human metabolite of HT and is commercially available; it has been reported as a radical scavenger comparable to HT and reputed by some authors to contribute to the beneficial properties exerted by olive oil (Tuck et al., 2002). Some studies on olive phenols (Manna et al., 1997; Paiva-Martins et al., 2003; Morellò et al., 2005) have highlighted the importance of the lipophilic character of the antioxidant with reference to the dispersion medium (bulk oil, emulsions), to the cell uptake and membrane crossing, and to the substrate to be protected (LDL or other cellular constituents). In addition to these data which are of particular interest in the biomedical field, the agro-industry is undergoing new pressures due to the widespread concern about the use of synthetic additives and antioxidants (such as BHT, 13) and to a growing demand for safer foods and beverages. This has led to a drive, by the industries of the agro-alimentary sector, to search for new antioxidants and other nutritional supplements of natural origin or obtained by simple modification of natural products. In fact, oxidation is one of the main causes of food deterioration, especially in oils and fats. Moreover, oxidized lipids, when absorbed in mammals, are incorporated into lipoproteins and may contribute to atherosclerosis (Parthasarathy et al., 1992; Regnstrom et al., 1992). Thus, the lipophilic HT analogues may be useful in the formulation of ‘beneficial foods’ for the prevention and treatment of chronic pathologies associated with reactive radical damage or as additives
1233
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1234
SECTION | III Tyrosol and Hydroxytyrosol
O
MeO OH
O O
HO R
O
R
OH
OH O O
HO HO
OH 1 R=H 2 R = OH
3 R = OH 4 R=H
OH O OH
O
O
O
HO O
OH
O
O
O
HO OH
O
HO
O
OH
HO
HO HO
OH
HO
OH O O OH
OH 5
6 MeO
O O
O O
HO
HO
O
O O
O
OH
OH OH 7
8 MeO
O
O OH O
HO HO
O
O
HO OH
OH
O
OH
HO
OH O O
HO HO
OH 9
10
FIGURE 135.1
for preserving foods from oxidation processes. On this basis, we compiled this chapter focusing on hydroxytyrosol lipophilic analogues, and in particular on their synthesis, their radical scavenging activities and other protective properties. Due to space constraints, we reviewed only the recent literature and have included almost exclusively those synthetic analogues which have been submitted to antioxidant activity assays.
O OH
O
OH OH 11
OMe OH
12
OH
13
135.2 LIPOPHILIC HYDROXYTYROSOL ANALOGUES
FIGURE 135.2
The most common methodology to convert HT (2) into more lipophilic analogues is the esterification of the primary alcoholic group without affecting the catechol moiety, which is known to be essential for the antioxidant and protective effects of 2. The chemical methods for selective
esterification of alcoholic groups have been paralleled, in recent years, by enzymatic methods, which avoid the use of toxic reagents and allow mild reaction conditions. Thus, Buisman et al. (1998) screened seven lipases in view of enzymatic esterification of 2. In a preliminary test, the
1235
CHAPTER | 135 Hydroxytyrosol Lipophilic Analogues
O
OH
O
octanoic acid CAL-B, cyclohexane, 81 °C OH
OH
14
15 O OH
O octanoic acid CAL-B, cyclohexane, 50 °C
OH
OH
OH
OH 16
2
SCHEME 135.1
O HO
OMe
OH
O O
O HO
O O 17
18
FIGURE 135.3
authors observed low reaction rates except for lipase B from Candida antarctica (CAL-B). In a subsequent series of experiments 2 and 3,5-di-t-butyl-4-hydroxybenzyl alcohol (3,5-DB-4-HBA, 14), whose antioxidant properties have been previously reported ( Papadopoulos and Boskou, 1991; Baldioli et al., 1996), were esterified according to Scheme 135.1. The 3,5-DB-4-HBA octanoate (15) was obtained with 93% yield within 5 h, whereas the HT octanoate (16) was obtained with 65% yield in 20 h. Both esters were evaluated for antioxidant performance by measurement of OSI in refined sunflower oil, in comparison with BHT (13) and 14. The authors observed that the addition of the octanoate esters increases the oxidation induction time of sunflower oil. Nevertheless, the highest antioxidant activity is observed for 14 followed by BHT, whereas the lowest antioxidant activity is obtained for 15. The analogue 16 resulted a less effective antioxidant than HT. As cited above, hydroxytyrosol acetate (11) is a natural constituent of olive oil, but it has also been obtained by synthesis. Gordon et al. (2001) prepared 11 from 2 by a series of chemical conversions, namely protection of the catechol moiety with benzyl bromide, followed by acetylation with acetic acid and subsequent deprotection through
catalytic hydrogenation. The antioxidant activity of 11 was assessed in comparison with that of other olive oil components, namely hydroxytyrosol (2), oleuropein (3), 3,4-DHPEA-EA (17) and α-tocopherol (18) by scavenging of DPPH• radicals as well as by measurement of the oxidative stability of bulk olive oil and oil-in-water emulsions. In the DPPH• scavenging test, the concentration required for 50% reduction in DPPH• radical concentration in 15 min was determined (EC50, reported as ratio moles of antioxidant/moles of DPPH•). At 15 min, the scavenging activity decreased in the order: 2 (EC50 ⫽ 0.19) ⬎ 18 (EC50 ⫽ 0.25), 11 (EC50 ⫽ 0.26). The antioxidant effect of 2 and 11 in bulk oil was evaluated in comparison with 18. The parameters examined were the peroxide value (PV) for primary oxidation products, and the p-anisidine value (AV) for secondary oxidation products; both values showed that the antioxidant activity decreased in the order: 2 ⬎ 11 ⬎⬎ 18 ⬎⬎ control. These results are in agreement with the polar paradox: that is, more polar antioxidants are more effective in less polar media (Porter, 1993; Frankel et al., 1994). When the study was carried out in oil-in-water emulsions (olive oil stripped of natural phenolic compounds and tocopherols), oxidation was faster than in bulk oil, and the order
1236
SECTION | III Tyrosol and Hydroxytyrosol
of antioxidant activity was different, following the order: 18 ⬎ 11 ⬎ 2 ⬎ control. In conclusion, 2 and 11 have a comparable antioxidant activity with 11 being less effective in oil, but slightly more effective in emulsion. In a further paper by the same group (Paiva-Martins et al., 2003), the antioxidant activity of 2, 3, 8, 11, and 17 in a soybean phospholipid liposome system was studied, in comparison with 18 and the water-soluble tocopherol analogue, Trolox. Lipid peroxidation, initiated by AAPH, was determined spectrophotometrically. The end of the lag phase was defined as the point where the slope rapidly increased. The radical scavenging activity of the phenols was evaluated by the ABTS assay (Re et al., 1999), that is, measuring their ability to trap the stable free radical ABTS•⫹. The abovecited phenols showed comparable lipid antioxidant activity with a duration of the lag phase almost twice that of 18 and better than Trolox. Synergistic effects (11–20% increase in lag phase) were observed in the antioxidant activity of combinations of 18 with olive oil phenols both with and without ascorbic acid. Localization of compounds in liposomes was studied by fluorescence anisotropy of probes and fluorescence quenching: these showed that the olive oil phenols did not penetrate into the membrane, but their effectiveness as antioxidants showed they were associated with the surface of the phospholipid bilayer. Torres de Pinedo et al. (2005) reported an efficient enzymatic synthesis of a series of long-chain phenolic esters with the antioxidant ortho-dihydroxy moiety, although they did not carry out any biological or chemical assay on their products. Two of these products were HT esters bearing polyunsaturated acyl chains, namely HT all-Z eicosapentaenoate (19) and HT all-Z docosahexaenoate (20). We consider these products worthy of citation here, in view of the well-known antioxidative properties of polyunsaturated fatty acids which in these esters are coupled to the properties of the catechol moiety. According to Scheme 135.2, they carried out the reaction in the presence of Candida antarctica lipase (CAL) and obtained the esters in good yields and short reaction times. The same authors (Torres de Pinedo et al., 2007) have more recently reported the synthesis of a series of phenolic
fatty acid esters and their evaluation as lipophilic antioxidants in an oil matrix. The HT analogues 21⫺24, the latter being a methylated HT derivative, were prepared by enzymatic acylation of the corresponding phenolic alcohols, using lipase from CAL-B. The HT palmitate 21, the HT stearate 22, the HT oleate 23 and the 3-hydroxy-4-methoxyphenethyl palmitate 24 were obtained as pure compounds, in good to excellent yields. The two oleoyl derivatives 25 and 26 were instead prepared by chemical acylation of the phenolic alcohol with oleic acid, using EDCI as activating reagent. These two compounds were obtained as a 1:1 unseparated mixture in lower yield (65%). The radical scavenging activity of all compounds was evaluated through the above-cited ABTS assay. Their efficacy as food antioxidants was evaluated in refined olive oil using the Rancimat method (Matthaus, 1996), by measuring the IT, that is the time taken until a rapid oxidation of the oil spiked with each antioxidant is observed. The palmitates 21 and 24 showed better scavenging capacities than the positive controls, α-tocopherol and ascorbyl palmitate. Actually, the authors report as ‘surprising’ the result obtained for compound 24, lacking the ortho-phenolic function. Moreover, all the catecholic analogues 21–23 showed higher IT in the Rancimat test than did the control antioxidants and compound 24. The type of fatty acid (palmitate, stearate or oleate) acylating the phenolic alcohols determined negligible differences in ITs. As expected, the mixture of acylated phenols 25 and 26 was a less effective antioxidant than the analogues bearing the ortho-dihydroxy moiety. The protective effect of HT lipophilic analogues against oxidative stress in human cells was recently studied by Manna et al. (2005). Two chemically stable HT analogues were prepared by standard chemical conversions, namely the triacetate 27 and the diacetate 28. As expected, the acetylated analogues were devoid of any chemical antioxidative property, determined by FRAP assay (Benzie and Strain, 1999). Conversely, both acetyl derivatives, at micromolar concentrations, equally protected against oxidative damages induced by t-BHP in human colon carcinoma Caco-2 cells (see Figures 135.5A and B) and human red blood cells (RBC) (see Figures 135.6A and B). O O
ethyl eicosapentaenoate CAL, no solvent, 37 °C OH
OH OH
19 O
OH OH 2
O
ethyl docosahexaenoate CAL, no solvent, 37 °C
OH OH
SCHEME 135.2
20
1237
CHAPTER | 135 Hydroxytyrosol Lipophilic Analogues
The HT triacetate 27 has been the subject of a further paper by the same authors (Capasso et al., 2006). The preparation of 27 (overall yield of 35.6%) from olive mill wastewater (OMWW) organic extract is reported, based on a O (
O
R1
)n
R2 OH
R3
R
O
(
)7
21 R = OH
n = 14
23 R1 = O
22 R = OH
n = 16
25 R1 = OH, R2 = OH, R3 = O
(
)7
R2 = OH, R3 = OH O
(
)7
(
)7
O
FIGURE 135.4
120
Control
Hdrx (1a)
t-BHP
12
Hdrx (1a)
Hemolysis %
100 80 60
100
Concentration (μM) Control
Hdrx (1a)
t-BHP
10 A
Control t-BHP
Diachdrx (1c)
3 2.5 2 1.5 1
50
Concentration (μM) 0.25
Triachdrx (1b)
TBARS (nmol mg–1 protein)
3.5
6
0 50
4
Diachdrx (1c)
8
2
0 A
Triachdrx (1b)
10
4
40 20
TBARS (nmol mg–1 protein)
Control
14
Triachdrx (1b) Diachdrx (1c)
t-BHP
Caco-2 cell viability (%)
mild treatment with an acetylating mixture (HClO4-SiO2 and acetic anhydride). This stable form of HT may be useful as a bioantioxidant, to be converted into the active form by intracellular esterases. More recently, Trujillo et al. (2006) evaluated some lipophilic hydroxytyrosyl esters as antioxidants in lipid matrices and biological systems. The esters were obtained by a procedure under patent (Alcudia et al., 2004): to a solution of HT in ethyl or methyl ester (acetate, butyrate, palmitate, stearate, oleate or linoleate) p-TSA was added, and the mixture was stirred for 24 h. After standard workup, the HT esters 11, 16, 21, 23, 29⫺31 were obtained in 62–86% yields. The authors evaluated the antioxidant activities of these lipophilic HT analogues in comparison with those of HT, 13 and 18 in both glyceridic matrix and biological systems. A lipid glyceridic matrix obtained from virgin olive oil was spiked with the antioxidant under test and subjected to
Hdrx (1a)
0.2
Triachdrx (1b) Diachdrx (1c)
0.15
0.1
0.05
0.5 0
0 50
B
100
Concentration (μM)
FIGURE 135.5 Effect of hydroxytyrosol (hdrx, 1a) and its acetyl derivatives (triachdrx, 1b; and diachdrx, 1c) on t-BHP-induced oxidative stress in Caco-2 cells. Cell viability (A) and TBARS concentration (B) were measured as reported by Manna et al. (2005). Values are means ⫾ SD; n ⫽ 4. Note the correspondence with our text: 1a ⫽ 2; 1b ⫽ 27; 1c ⫽ 28. Reproduced with permission from J. Agric. Food Chem. 2005, 53, 9602– 9607. Copyright © 2005 American Chemical Society.
10 B
50
Concentration (μM)
FIGURE 135.6 Effect of hydroxytyrosol (hdrx, 1a) and its acetyl derivatives (triachdrx, 1b; and diachdrx, 1c) on t-BHP-induced oxidative stress in RBC. Hemolysis percentage (A) and TBARS concentration (B) were measured as reported by Manna et al. (2005). Values are means⫾ SD; n ⫽ 4. Note the correspondence with our text: 1a ⫽ 2; 1b ⫽ 27; 1c ⫽ 28. Reproduced with permission from J. Agric. Food Chem. 2005, 53, 9602– 9607. Copyright © 2005 American Chemical Society.
1238
SECTION | III Tyrosol and Hydroxytyrosol
accelerated oxidation in a Rancimat apparatus operating at 90 °C. The IT was measured as the time required to have a sharp increase of a conductivity measurement, and was reported versus the concentration C (mmol kg⫺1) of each antioxidant (see Figure 135.8). HT and its analogues 11, 16, 21, 23, 29⫺31 showed comparable antioxidant activities, which resulted higher than those of 13 or 18. In a second experiment, the capacity of esters of hydroxytyrosol to protect proteins and lipids against oxidation caused by peroxyl radicals was measured. A brain homogenate model was used, because brain tissue is very vulnerable to oxidative damage due to its relative lack of antioxidant enzymes. The protective effect of the above-cited esters against the damage caused by CH has been evaluated by measuring the MDA, the most abundant aldehyde produced as product of lipid peroxidation when lipid hydroperoxides break down in biological systems (Esterbauer and Cheeseman, 1990), and carbonyl group content, because of the introduction of carbonyl groups into proteins by oxidative mechanisms. The results
O
O
O
OH O
O
R
O
O
O
O
O
OH OH
O
O
27
28
29 R = C3H7 30 R = C11H23 31 R = C17H31
FIGURE 135.7
100 80
IT (h)
60 40
have been compared with those for 2, 13 and 18. The capacity of HT and HT acetate (11), palmitate (21), and oleate (23) to prevent lipid peroxidation was similar to that of 18, but lower than that of 13. Interestingly, the hydroxytyrosyl linoleate (31) showed greater activity than the other esters. A novel methodology of derivatization has been applied by Bernini et al. (2007) to a series of phenolic alcohols, including HT. These authors prepared the carboxymethylated HT analogue 32 by an eco-friendly, chemoselective and efficient method, employing dimethyl carbonate in the presence of DBU or sulfuric acid as catalysts (see Scheme 135.3). After work-up, the product was obtained in quantitative yield. The antioxidant activity of 32 was investigated using the DPPH• radical scavenging test. The results showed that this new compound has an antioxidant activity (EC50 ⫽ 0.11 mol L⫺1 antioxidant mol⫺1 L DPPH•) similar to that reported for HT (Tuck et al., 2002). A paper of Rietjens et al. (2007) is focused on the controversies concerning the antioxidant potential of the olive oil antioxidants, and in particular HT; this study includes tyrosol (1) and homovanillic alcohol (12): the authors observed that both these compounds, lacking a free catechol moiety, resulted less potent radical scavengers with respect to HT, although 12 was a relatively good scavenger of ONOOH and OH•. Recently, some of us (Grasso et al., 2007) have carried out a study on a series of HT lipophilic analogues, to evaluate the possible effects of an enhanced lipophilicity on the antioxidative properties of HT and 12. An environmentally friendly and chemoselective enzymatic method was preferred to the previously reported chemical conversions. Tyrosol, more stable than hydroxytyrosol, was used as a model for a preliminary acetylation screening with 12 different lipases, employing vinyl acetate as reagent and t-butyl methyl ether as solvent, on the basis of previous satisfactory results (Nicolosi et al., 2002). The best results were obtained with Mucor miehei and Candida antarctica (CAL) lipases, this latter being less expensive and consequently employed on a preparative scale. Ten lipophilic esters were prepared by acylation at the alcoholic function of 2 (11, 22, 29, 33, 34) and 12 (35–39) employing vinyl acetate, propionate, butyrate, decanoate and stearate. CAL and the acyl donor were added to a
20 O
0 0.0
0.2
0.4 0.6 C(mmol kg–1)
0.8
1.0
FIGURE 135.8 Induction times (ITs) of lipid matrices spiked with hydroxytyrosol (䊏), hydroxytyrosyl acetate (·), hydroxytyrosyl palmitate (䉱), hydroxytyrosyl oleate ( ), hydroxytyrosyl linoleate (sideways triangle pointing to the left), α-tocopherol (䉲), or BHT ([insert sideways triangle pointing to the right]). Reproduced with permission from J. Agric. Food Chem. 2006, 54, 3779–3785. Copyright © 2006 American Chemical Society.
O
OH DMC/DBU or DMC/H 2SO4
OH
OH OH 2
SCHEME 135.3
OH 32
OMe
1239
CHAPTER | 135 Hydroxytyrosol Lipophilic Analogues
solution of the substrate (2 or 12) in tert-butyl methyl ether and the mixture was shaken at 40 °C: HT acetate (11), HT propionate (33) and HT butyrate (29) were obtained after 35 min with 95–96% yield; HT decanoate (34) was obtained in 93% yield after 75 min, whereas HT stearate (22) was obtained in 92% yield after 3 h. Analogously, 12 afforded with comparable yields the corresponding acetate (35), propionate (36), butyrate (37), decanoate (38) and stearate (39) with reaction times in the range 60 min–4 h. Compounds 11, 22, 29, 33, 34, 35–39 were submitted to the DPPH assay and compared with 1, 2 and 12 (see Table 135.1). As expected, 2 proved a much more effective scavenger than 1. Compound 12 resulted clearly less active than 2, in contrast with previous literature data (Tuck et al., 2002). The SC50 values of the two groups of analogues related to 2 and 12 are respectively in the range 20.5–24.8 and 40.5–44.6 μM. These data confirmed that the anti-radical activity of HT is not notably influenced by the presence and length of an acyl chain at the alcoholic function, and that the homovanillic system has lower radical scavenging activity with respect to the HT system. Compounds 2, 12 and their lipophilic analogues were also subjected to the atypical Comet test (Collins, 2004) on whole blood cells in order to evaluate both their possible basal DNA damaging properties and their capacity to counteract the H2O2 caused oxidative stress. The results were expressed as TDNA or tail intensity, considering this value indicative of the number of DNA breaks. Figure 135.10A reports the TDNA values obtained by treatment of whole cells with compounds 1, 2 and 12 and the HT analogues 11, 22, 29, 33 and 34, compared with the values obtained for untreated control cells (C). Figure 135.10B reports the TDNA values obtained when these compounds were added before the H2O2 insult in comparison with the data obtained for solely H2O2 treated cells (H2O2). The data in Figure 135.10A show that 1 and 12 have a significant basal DNA-damaging effect with respect to the control values, whereas no significant damage is caused by 2. Interestingly, HT acetate (11) and HT propionate (33) show
O
O O
R
O
OH OH 33 R = C2H5 34 R = C9H19
FIGURE 135.9
R
OMe OH 35 36 37 38 39
R = CH3 R = C2H5 R = C3H7 R = C9H19 R = C17H35
a minimal DNA-damaging activity, while the damage significantly increases for compounds 22, 29 and 34 bearing respectively stearoyl, butyryl and decanoyl acyl chain. Figure 135.10B clearly shows that 2 is highly protective towards the oxidative induced DNA damage in whole blood cells, differently from 1 and 12. Among the lipophilic analogues, 11 and 33 are comparable to 2 in counteracting oxidative stress, but the protective effect progressively decreases in the order 29 ⬎ 22 ≈ 34. In separate experiments, the reference compounds 1, 2 and 12 and the homovanillic alcohol analogues 35–39 were examined and the TDNA values thus obtained are reported in Figures 135.11A and B. The experiment on basal DNA damage confirms that 2 does not cause any significant damage at basal level, whereas 12 and its lipophilic analogues 35–39 are comparable in causing a basal damage. The experiment reported in Figure 135.11B confirms the good
TABLE 135.1 DPPH• scavenging activity. Compounds
SC50 (μM)a ⫾ SD
2
33.2 ⫾ 5.6
3
24.6 ⫾ 6.5
4
46.1 ⫾ 3.4
5
21.9 ⫾ 1.4
6
22.9 ⫾ 5.1
7
24.7 ⫾ 1.0
8
24.8 ⫾ 10.2
9
20.5 ⫾ 6.6
10
42.9 ⫾ 5.8
11
44.6 ⫾ 1.8
12
41.8 ⫾ 2.2
13
44.5 ⫾ 4.2
14
40.5 ⫾ 1.8
Note the correspondence with our text: 2 ⫽ 1; 3 ⫽ 2; 4 ⫽ 12; 5 ⫽ 11; 6 ⫽ 33; 7 ⫽ 29; 8 ⫽ 34, 9 ⫽ 22; 10 ⫽ 35; 11 ⫽ 36; 12 ⫽ 37; 13 ⫽ 38 and 14 ⫽ 39. Reprinted from, S. Grasso, L. Siracusa, C. Spatafora, M. Renis, C. Tringali. Hydroxytyrosol lipophilic analogues: enzymatic synthesis, radical scavenging activity and DNA oxidative damage protection. Bioorganic Chemistry 35: 137–152. Copyright (2007), with permission from Elsevier. a SC50 (μM), Scavenging capacity: phenol concentration, expressed in μM, able to quench 50% of DPPH radicals in a 92 μM solution (mAU ⫽ 1, solvent: cyclohexane). Each reported value is the mean of three separate measurements.
TDNA = % of fragmented DNA
A
SECTION | III Tyrosol and Hydroxytyrosol
100
90
90
80 70
* *
*
*
50 40 30 20 10 0
C
2
3
4
5
6
7
8
9
A
80 70 60
50 40 30 20 10 H 2O 2
3
4
5
6
7
8
9
B
*
*
10 0
C
2
3
4
10 11 12 13 14
H2O2
2
3
4
10 11 12 13 14
80 70 60 50 40 30 20 10 0
2
*
20
90
60
*
30
100
70
*
40
90 80
*
*
50
100
0 B
*
60
TDNA = % of fragmented DNA
100
TDNA = % of fragmented DNA
TDNA = % of fragmented DNA
1240
FIGURE 135.10 Atypical alkaline COMET assay. *Significantly different from control untreated whole blood cells (p ⬎ 0.001). °Significantly different from H2O2 alone treated cells (p ⬎ 0.001). The results are reported as TDNA ⫽ % of fragmented DNA. (A) Basal DNA damage: whole blood cells were treated for 20 min at 37 °C with the tested compounds at the concentration 50 μM. (B) Oxidative DNA damage protection: whole blood cells were firstly treated for 20 min at 37 °C like (A) and second, after a wash with PBS 1X, for 20 min at 37 °C with H2O2 (200 μM). Each experiment, performed in duplicate, was repeated three times and the mean ± SEM for each value was calculated. Note the correspondence with our text: 2 ⫽ 1; 3 ⫽ 2; 4 ⫽ 12; 5 ⫽ 11; 6 ⫽ 33; 7 ⫽ 29; 8 ⫽ 34 and 9 ⫽ 22. Reprinted from, S. Grasso, L. Siracusa, C. Spatafora, M. Renis, C. Tringali. Hydroxytyrosol lipophilic analogues: enzymatic synthesis, radical scavenging activity and DNA oxidative damage protection. Bioorganic Chemistry 35: 137–152. Copyright (2007), with permission from Elsevier.
FIGURE 135.11 Atypical alkaline COMET. *Significantly different from control untreated whole blood cells (p ⬎ 0.001). °Significantly different from H2O2 alone treated cells (p ⬎ 0.001). The results are reported as TDNA ⫽ % of fragmented DNA. (A) Basal DNA damage: whole blood cells were treated for 20 min at 37 °C with the tested compounds at the concentration 50 μM. (B) Oxidative DNA damage protection: blood cells were firstly treated for 20 min at 37 °C like (A) and second, after a wash with PBS 1X, for 20 min at 37 °C with H2O2 (200 μM). Each experiment, performed in duplicate, was repeated three times and the mean ± SEM for each value was calculated. Note the correspondence with our text: 2 ⫽ 1; 3 ⫽ 2; 4 ⫽ 12; 10 ⫽ 35; 11 ⫽ 36; 12 ⫽ 37; 13 ⫽ 38 and 14 ⫽ 39. Reprinted from S. Grasso, L. Siracusa, C. Spatafora, M. Renis, C. Tringali. Hydroxytyrosol lipophilic analogues: enzymatic synthesis, radical scavenging activity and DNA oxidative damage protection. Bioorganic Chemistry 35: 137–152. Copyright (2007), with permission from Elsevier.
protective properties elicited by 2 versus H2O2-induced DNA damage, as well as the poor protection given by 12. The lipophilic analogues of the latter show slightly higher TDNA values than that of 12. In this work, we determined the log P value for all compounds. The results of the Comet test show a good protective effect for 2 and its analogues 11 and 33, with log P ⱕ 1.20, a moderate effect for 29 (log P ⫽ 1.77) and negligible protection for 34 and 22 (log P ⬎ 5), these latter exerting a significant DNA damage in basal conditions. Thus, it appears that a longer acyl chain than C4 and a higher log P than 2 cause the loss of the protective properties of HT. In conclusion, the lipophilic analogues 11, 22, 29, 33
and 34 may be profitably used, in principle, as antioxidants in bulk lipids or emulsions; 11 and 33, in particular, could be further evaluated for possible applications in pharmaceutical, nutritional or cosmetic fields. The experiments carried out on compound 12 and its analogues 35–39 gave a different result, indicating that the homovanillic system has poor antioxidant activity, without correlation with increased lipophilicity. A very recent study of Fabiani et al. (2008) also deals with the preventive properties of HT and other related olive phenolics against the oxidative DNA damage in human peripheral blood mononuclear cells (PBMC) and includes the effects towards leukaemia HL60 cells. HT and a complex
1241
CHAPTER | 135 Hydroxytyrosol Lipophilic Analogues
HO
OH
O O
O
OH
40
HO
OH
O O
N H
OH
HO
OH
O
41
N H
OH
42
FIGURE 135.12
mixture of phenols obtained from virgin olive oil and olive mill wastewater reduced the H2O2-induced DNA damage at concentrations as low as 1 μM, as determined by Comet assay. The protection by HT was 93% in HL60 and 89% in PBMC at 10 μM. A comparable protective effect was observed for 8 on both cultured cells. Other phenolic purified compounds, among them 1, 3 and 5, showed a lower protective effect (range of protection 25–75%). In an interesting and recent study by Palozza et al. (2008) a group of novel α-tocopherol analogues are reported, some of them being also HT lipophilic analogues, namely compounds 40–42. The authors designed novel chromanyl derivatives sharing a chromanyl head with α-tocopherol but differing in the side chain. In particular, compounds 40–42 included the further antioxidant moiety of HT or its nitrogenated analogue, dopamine. The protective effect of these compounds towards freeradical-induced oxidative stress was investigated in isolated membranes as well as in intact cells. Tocopherol analogues were added to rat liver microsomes and to RAT-1 fibroblasts and were exposed to the pro-oxidant action of free radical sources including AAPH, t-BOOH, and H2O2. The antioxidant efficiency was evaluated by measuring several parameters of oxidative stress, including MDA and conjugated diene formation, ROS production, cell viability, and expression of heat-shock proteins (hsp70, hsp90) and was compared with that of natural α-tocopherol and α-tocotrienol. The results clearly demonstrate that all synthesized compounds were active in: (i) inhibiting lipid peroxidation in microsomes and (ii) preventing H2O2-induced ROS production, cell damage, and heat-shock protein expression in immortalized RAT-1 fibroblasts. The concomitant presence of a chromanyl head and an additional catechol moiety markedly increased the antioxidant potency of the molecule. In particular, 40 and 41, resulting from the molecular combination of Trolox with HT and dopamine, respectively, were much more potent than α-tocopherol, α-tocotrienol, and the other synthetic compounds. A very recent addition to the above-cited literature reports the isolation and characterization of a new HT
O
HO
O OMe
HO O 43
OMe
FIGURE 135.13
secoiridoid derivative from Olea europaea leaves, namely the bis methylacetal of oleuropein aglycone 43 (PaivaMartins and Pinto, 2008). The radical scavenging activity of this new compound, evaluated by DPPH• assay, resulted much higher than those of 8 or 18. The authors observed an easy conversion of 43 into 8 in acidic aqueous media, and consequently call to attention the need for a careful identification of compounds by HPLC-MS, usually performed in acidic conditions.
SUMMARY POINTS ●
●
●
●
●
●
More than 40 hydroxytyrosol (HT) lipophilic analogues are reported. Selective chemical and enzymatic methods for the preparation of HT lipophilic analogues are described. Analogues including an ortho-diphenol moiety are normally more effective antioxidants than those with a single free phenolic function. The HT metabolite homovanillic alcohol and its lipophilic analogues are poor antioxidants and protective agents with respect to HT and the majority of its lipophilic analogues. The majority of the lipophilic HT analogues showed very good radical scavenging activities and/or effectively inhibited lipid oxidations. In some cases, the HT analogues were more effective than other widely employed antioxidants, such as α-tocopherol or BHT. Some of the reported lipophilic HT analogues were also demonstrated to be protective towards oxidative
1242
●
●
SECTION | III Tyrosol and Hydroxytyrosol
damage to human cells, and in particular towards H2O2induced DNA damage. In a study of H2O2-induced DNA damage, the HT esters with a longer acyl chain than C4 and a higher log P than 2 showed low protective properties and caused significant DNA basal damage. The conjugation with a chromanyl head markedly increased the antioxidant potency of HT.
REFERENCES Alcudia, F., Cert, A., Espartero, J. L., Mateos, R., and Trujillo, M., 2004. Method of preparing hydroxytyrosol esters, esters thus obtained and use of same. PCT WO 2004/005237. Baldioli, M., Servili, M., Perretti, G., Montedoro, G.F., 1996. Antioxidant activity of tocopherols and phenolic compounds of virgin olive oil. J. Am. Oil Chem. Soc. 73, 1589–1593. Benzie, I.F.F., Strain, J.J., 1999. Ferric reducing/antioxidant power assay: direct measure of total antioxidant activity of biological fluids and modified version for simultaneous measurement of total antioxidant power and ascorbic acid concentration. Methods Enzymol. 299, 15–27. Bernini, R., Mincione, E., Crisante, F., Barontini, M., Fabrizi, G., Gentili, P., 2007. Chemoselective and efficient carbomethoxylation of the alcoholic chain of phenols by dimethyl carbonate (DMC). Tetrahedron Lett. 48, 7000–7003. Brenes, M., Garcia, A., Garcia, P., Garrido, A., 2001. Acid hydrolysis of secoiridoid aglycons during storage of virgin olive oil. J. Agric. Food Chem. 49, 5609–5614. Buisman, G.J.H., Van Helteren, C.T.W., Kramer, G.F.H., Veldsink, J.W., Derksen, J.T.P., Cuperus, F.P., 1998. Enzymatic esterifications of functionalized phenols for the synthesis of lipophilic antioxidants. Biotechnol. Lett. 20, 131–136. Capasso, R., Sannino, F., De Martino, A., Manna, C., 2006. Production of triacetylhydroxytyrosol from olive mill waste waters for use as stabilized bioantioxidant. J. Agric. Food Chem. 54, 9063–9070. Collins, A.R., 2004. The comet assay for DNA damage and repair: principles, applications, and limitations. Molec. Biotech. 26, 249–261. Esterbauer, H., Cheeseman, K.H., 1990. Determination of aldehydic lipid peroxidation products: malonaldehyde and 4-hydroxynonenal. Methods Enzymol 186, 407–421. Fabiani, R., De Bartolomeo, A., Rosignoli, P., Servili, M., Montedoro, G.F., Morozzi, G., 2002. Cancer chemoprevention by hydroxytyrosol isolated from virgin olive oil through G1 cell cycle arrest and apoptosis. Eur. J. Cancer Prev. 11, 351–358. Fabiani, R., Rosignoli, P., De Bartolomeo, A., Fuccelli, R., Servili, M., Montedoro, G.F., Morozzi, G., 2008. Oxidative DNA damage is prevented by extracts of olive oil, hydroxytyrosol, and other olive phenolic compounds in human blood mononuclear cells and HL60 cells. J. Nutr. 138, 1411–1416. Fernandez-Bolanos, J.G., Lopez, O., Fernandez-Bolanos, J., RodriguezGutierrez, G., 2008. Hydroxytyrosol and derivatives: isolation, synthesis, and biological properties. Curr. Org. Chem. 12, 442–463. Frankel, E.N., Huang, S.W., Kanner, J., German, J.B., 1994. Interfacial phenomena in the evaluation of antioxidants in bulk oils versus emulsions. J. Agric. Food Chem. 42, 1054–1059. Gordon, M.H., Paiva-Martins, F., Almeida, M., 2001. Antioxidant activity of hydroxytyrosol acetate compared with that of other olive oil polyphenols. J. Agric. Food Chem. 49, 2480–2485.
Goupy, P., Dufour, C., Loonis, M., Dangles, O., 2003. Quantitative kinetic analysis of hydrogen transfer reactions from dietary polyphenols to the DPPH radical. J. Agric. Food Chem. 51, 615–622. Grasso, S., Siracusa, L., Spatafora, C., Renis, M., Tringali, C., 2007. Hydroxytyrosol lipophilic analogues: Enzymatic synthesis, radical scavenging activity and DNA oxidative damage protection. Bioorg. Chem. 35, 137–152. Keys, A., 1995. Mediterranean diet and public health: personal reflections. Am. J. Clin. Nutr. 41, 1321S–1323S. Kok, F.J., Kromhout, D., 2004. Atherosclerosis-epidemiological studies on the health effects of a Mediterranean diet. Eur. J. Nutr. 43 (S. 1), 2–5. Leenen, R., Roodenburg, A.J.C., Vissers, M.N., Schuurbiers, J.A.E., Van Putte, K.P.A.M., Wiseman, S.A., Van de Put, F.H.M., 2002. Supplementation of plasma with olive oil phenols and extracts: influence on LDL oxidation. J. Agric. Food Chem. 50, 1290–1297. Manna, C., Galletti, P., Cucciola, V., Moltedo, O., Leone, A., Zappia, V., 1997. The protective effect of the olive oil polyphenol (3,4dihydroxyphenyl) ethanol counteracts reactive oxygen metaboliteinduced cytotoxicity in Caco-2 cells. J. Nutr. 127, 286–292. Manna, C., Galletti, P., Cucciola, V., Montedoro, G., Zappia, V., 1999. Olive oil hydroxytyrosol protects human erythrocytes against oxidative damages. J. Nutr. Biochem. 10, 159–165. Manna, C., Galletti, P., Misto, G., Cucciola, V., D’Angelo, S., and Zappia, V., 2000. Transport mechanism and metabolism of olive oil hydroxytyrosol in Caco-2 cells. FEBS Lett. 470, 341–344. Manna, C., Migliardi, V., Sannino, F., De Artino, A., Capasso, R., 2005. Protective effects of synthetic hydroxytyrosol acetyl derivatives against oxidative stress in human cells. J. Agric. Food Chem. 53, 9602–9607. Martin-Moreno, J.M., Willett, W.C., Gorgojo, L., Banegas, J.R., Rodriguez-Artalejo, F., Fernandez-Rodriguez, J.C., Maisonneuve, P., Boyle, P., 1994. Dietary fat, olive oil intake and breast cancer risk. Int. J. Cancer 58, 774–780. Matthaus, B.W., 1996. Determination of the oxidative stability of vegetable oils by Rancimat and conductivity and chemiluminescence measurements. J. Am. Oil Chem. Soc. 73, 1039–1043. Morellò, J.-R., Vuorela, S., Romero, M.-P., Motiva, M.-J., Heinonen, M., 2005. Antioxidant activity of olive pulp and olive oil phenolic compounds of the arbequina cultivar. J. Agric. Food Chem. 53, 2002–2008. Nicolosi, G., Spatafora, C., Tringali, C., 2002. Chemo-enzymatic preparation of resveratrol derivatives. J. Mol. Catal. B: Enzym. 16, 223–229. Owen, R.W., Giacosa, A., Hull, W.E., Haubner, R., Spiegelhalder, B., Bartsch, H., 2000. The antioxidant/anticancer potential of phenolic compounds isolated from olive oil. Eur. J. Cancer 36, 1235–1247. Paiva-Martins, F., Gordon, M.H., Gameiro, P., 2003. Activity and location of olive oil phenolic antioxidants in liposomes. Chem. Phys. Lipids 124, 23–36. Paiva-Martins, F., Pinto, M., 2008. Isolation and characterization of a new hydroxytyrosol derivative from olive (Olea europaea) leaves. J. Agric. Food Chem. 56, 5582–5588. Palozza, P., Simone, R., Picci, N., Buzzoni, L., Ciliberti, N., Natangelo, A., Manfredini, S., Vertuani, S., 2008. Design, synthesis, and antioxidant potency of novel α-tocopherol analogues in isolated membranes and intact cells. Free Radical Biol. Med. 44, 1452–1464. Papadopoulos, G., Boskou, D., 1991. Antioxidant effect of natural phenols on olive oil. J. Am. Oil Chem. Soc. 68, 669–671. Parthasarathy, S., Steinberg, D., Witztum, J.L., 1992. The role of oxidized low-density lipoproteins in the pathogenesis of atherosclerosis. Ann. Rev. Med. 43, 219–225. Porter, W.L., 1993. Paradoxical behaviour of antioxidants in food and biological systems. Toxicol. Ind. Health 9, 93–122.
CHAPTER | 135 Hydroxytyrosol Lipophilic Analogues
Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., Rice-Evans, C., 1999. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biol. Med. 26, 1231–1237. Regnstrom, J., Nilsson, J., Tornvall, P., Landou, C., Hamsten, A., 1992. Susceptibility to low-density lipoprotein oxidation and coronary atherosclerosis in man. Lancet 339, 1183–1186. Rietjens, S.J., Bast, A., Haenen, G.R.M.M., 2007. New insights into controversies on the antioxidant potential of the olive oil antioxidant hydroxytyrosol. J. Agric. Food Chem. 55, 7609–7614. Roche, M., Dufour, C., Mora, N., Dangles, O., 2005. Antioxidant activity of olive phenols: mechanistic investigation and characterization of oxidation products by mass spectrometry. Org. Biomol. Chem. 3, 423–430. Saija, A., Trombetta, D., Tomaino, A., Lo Cascio, R., Trinci, P., Uccella, N., Bonina, F., Castelli, F., 1998. In vitro evaluation of the antioxidant activity and biomembrane interaction of the plant phenols oleuropein and hydroxytyrosol. Int. J. Pharm. 166, 123–133. Stupans, I., Kirlich, A., Tuck, K.L., Hayball, P.J., 2002. Comparison of radical scavenging effect, inhibition of microsomal oxygen free radical generation, and serum lipoprotein oxidation of several natural antioxidants. J. Agric. Food Chem. 50, 2464–2469. Torres de Pinedo, A., Penalver, P., Rondon, D., Morales, J.C., 2005. Efficient lipase-catalyzed synthesis of new lipid antioxidants based on a catechol structure. Tetrahedron 61, 7654–7660.
1243
Torres de Pinedo, A., Penalver, P., Perez-Victoria, I., Rondon, D., Morales, J.C., 2007. Synthesis of new phenolic fatty acid esters and their evaluation as lipophilic antioxidants in an oil matrix. Food Chem. 105, 657–665. Tripoli, E., Giammanco, M., Tabacchi, G., Di Majo, D., Giammanco, S., La Guardia, M., 2005. The phenolic compounds of olive oil: structure, biological activity and beneficial effects on human health. Nutr. Res. Rev. 18, 98–112. Trujillo, M., Mateos, R., Collantes de Teran, L., Espartero, J.L., Cert, R., Jover, M., Alcudia, F., Bautista, J., Cert, A., Parrado, J., 2006. Lipophilic hydroxytyrosol esters. Antioxidant activity in lipid matrices and biological systems. J. Agric. Food Chem. 54, 3779–3785. Tuck, K.L., Freeman, M.P., Hayball, P.J., Stretch, G.L., Stupans, I., 2001. The in vivo fate of hydroxytyrosol and tyrosol, antioxidant phenolic constituents of olive oil, after intravenous and oral dosing of labelled compounds to rats. J. Nutr. 131, 1993–1996. Tuck, K.L., Hayball, P.J., 2002. Major phenolic compounds in olive oil: metabolism and health effects. J. Nutr. Biochem. 13, 636–644. Tuck, K.L., Hayball, P.J., Stupans, I., 2002. Structural characterization of the metabolites of hydroxytyrosol, the principal phenolic component in olive oil, in rats. J. Agric. Food Chem. 50, 2404–2409.
This page intentionally left blank
Chapter 136
Effects of Hydroxytyrosol on Cyclosporine Nephrotoxicity Vincenzo Zappia1, Patrizia Galletti1, Caterina Manna1, Stefania D’Angelo1,2, Daniela Napoli1, Maria Luigia De Bonis1 and Giovambattista Capasso3 1
Department of Biochemistry and Biophysics, School of Medicine, Second University of Naples, Italy Faculty of Motor Sciences, Parthenope University, Naples, Italy 3 Chair of Nephrology, Department of Internal Medicine, Second University of Naples, Italy
2
136.1 INTRODUCTION Olive oil represents the typical lipidic source of the Mediterranean diet that has been associated with a low incidence of several pathologies, including cardiovascular diseases and neurological disorders (Bendini et al., 2007; Trichopoulou and Dilis, 2007). The beneficial properties of olive oil have been mainly attributed to its high content of monounsaturated oleic acid, however, in recent years converging evidence indicates that the non-glyceride fraction of olive oil, rich in vitamin and non-vitamin antioxidants including polyphenols, significantly contributes to its benefits on human health (Visioli and Galli, 2002). Hydroxytyrosol (3,4-dihydroxyphenylethanol; DOPET) is the main ortho-diphenolic compound of olive oil and is mainly responsible for the antioxidant properties of this nutrient. Indeed, it has been shown to function as an efficient scavenger of peroxyl radicals in several biological systems and contributes to increase the shelf-life of the oil, preventing its auto-oxidation (Visioli and Galli, 2002).
136.2 BIOLOGICAL EFFECTS OF HYDROXYTYROSOL The biological activities of DOPET have been explored by several groups and are summarized in Table 136.1 (for a comprehensive review see Bendini et al., 2007). Even though the majority of them can be directly ascribed to its antioxidant activity (Manna et al., 1999; Visioli and Galli 2002), emerging evidence (Della Ragione et al., 2002) supports the view that some effects of this molecule are independent from its scavenging properties. DOPET inhibits in vitro low-density lipoprotein oxidation and modulates the Olives and Olive Oil in Health and Disease Prevention. ISBN: 978-0-12-374420-3
oxidative/antioxidative balance in plasma (Covas et al., 2006). Experiments from our group demonstrated that DOPET, which effectively permeates cell membranes via passive diffusion (Manna et al., 2000), counteracts the cytotoxic effects of reactive oxygen species (ROS) in various human systems. Preincubation of intestinal Caco-2 cells with DOPET prevents the typical damages of oxidative stress (Manna et al., 1999). Similarly, this polyphenol exerts a protective effect against the H2O2-induced oxidative hemolysis and lipid peroxidation in red blood cells (Manna et al., 1999). Moreover, it has been demonstrated that the molecule counteracts the up-rise of specific markers of oxidative stress in UVA-irradiated melanoma cells. In fact it prevents the formation of abnormal L-isoaspartyl residues in proteins and reduces lipid peroxidation in irradiated M14 cells (D’Angelo et al., 2005). Furthermore, pretreatment of human hepatocarcinoma HepG2 cells with micromolar concentrations of DOPET completely prevents the decrease of reduced glutathione and the rise of malondialdehyde induced by tert-butyl hydroperoxide in this cell line (Goya et al., 2007). As already mentioned, it has been proposed that the antioxidant effect of DOPET probably contributes to the prevention of some degenerative diseases and supports brain cell survival after oxidative injuries (Bendini et al., 2007; Trichopoulou and Dilis, 2007). The effects of DOPET on inflammation/atherogenesis have also been thoroughly investigated. It has been demonstrated that this antioxidant inhibits the expression of adhesion molecules in a human endothelial cell line (HUVEC) exposed to proinflammatory cytokines (Carluccio et al., 2003). Moreover, DOPET administration in hyperlipidemic rabbits is able to reduce the size of atherosclerotic lesions of the aortic arch (González-Santiago et al., 2006). The ability of DOPET to counteract inflammation is also
1245
Copyright © 2010 Elsevier Inc. All rights of reproduction in any form reserved.
1246
SECTION | III Tyrosol and Hydroxytyrosol
TABLE 136.1 Biological effects of hydroxytyrosol. Affected biological systems
Experimental models
References
LDL oxidation in vitro and in vivo*
Human
Covas et al., 2006
Cytokine-induced endothelial activation and macrophage activation*
Human umbilical vein endothelial cells (HUVEC)
Carluccio et al., 2003
Bovine aortic endothelial cells (BAEC)
Maiuri et al., 2005
Murine macrophages (J774) Platelet and leukocytes arachidonate lipoxygenases*
Rat
de la Puerta et al., 1999
PMA-induced respiratory burst*
Human neutrophils
Visioli and Galli, 2002
Peroxynitrite-dependent tyrosine nitration*
Neuronal hybridoma cell line N-18-RE-105
Deiana et al., 1999
Platelet aggregation*
Human
Dell’Agli et al., 2008
Cell proliferation* and accumulation of cells in the G0/G1 phase**
Human myeloid leukemia cells (HL-60)
Fabiani et al., 2008
Apoptosis**
Human myeloid leukemia cells (HL-60)
Della Ragione et al., 2000
Human carcinoma colon cell line (HT-29)
Fabiani et al., 2008
Atherosclerotic lesion development (*)
Rabbit
González-Santiago et al., 2006
ROS-mediated citotoxicity (*)
Human carcinoma colon cell line (Caco-2)
Manna et al., 1999
Human erythrocytes
Manna et al., 1999
Human melanoma cell line (M14)
D’Angelo et al., 2005
Human hepatocarcinoma cell line (HepG2)
Goya et al., 2007
*
Inhibition. Induction.
**
supported by the evidence that extra virgin olive oil administration significantly decreases the levels of known markers of this process, such as thromboxane B2 (Bendini et al., 2007) and 5- and 12-lipoxygenases (de la Puerta et al., 1999). On the other hand, the anti-inflammatory properties of the phenolic fraction of olive oil are widely documented in the literature, DOPET being a powerful inhibitor of neutrophil respiratory burst (Visioli and Galli, 2002). The prevention of abnormal cell proliferation and the induction of apoptosis, both events involved in carcinogenesis, have been described as effects of DOPET not directly attributable to its antioxidant properties (Bendini et al., 2007). Indeed, it has been recently demonstrated that this compound, in a micromolar concentration range, inhibits the proliferation and induces apoptosis in different human cell lines (Della Ragione et al., 2000; Fabiani et al., 2008). In detail, the polyphenol alters HL60 cell cycle progression, inducing
an accumulation of cells in the G0/G1 phase (Manna et al., 1999; Fabiani et al., 2008). In the same cell line, it has been recently demonstrated that this effect is associated with an up-regulation of cyclin-dependent protein kinase inhibitors and with an induction of cell differentiation (Fabiani et al., 2008). In this respect, it has been proposed that DOPET may affect the expression of genes involved in the regulation of tumor cell proliferation and differentiation, such as p27Kip1 and p21WAF/Cip1 (Fabiani et al., 2008). It is worthwhile mentioning that tyrosol (4-hydroxyphenylethanol), the DOPET analogue which lacks the ortho-diphenolic moiety, does not show any antioxidant activity and fails to exert any inhibitory effect on cell growth, indicating that the two ortho-hydroxyl groups are critical for both antioxidant and antiproliferative effects (Manna et al., 1999). The prevention of DNA damage, responsible for mutagenesis and carcinogenesis, can be envisioned as another
CHAPTER | 136 Effects of Hydroxytyrosol on Cyclosporine Nephrotoxicity
mechanism associated with the DOPET anticancer effect. This phenol, indeed, exerts an inhibitory effect on peroxynitrite-dependent DNA base modifications as well as on tyrosine nitration (Deiana et al., 1999).
136.3 CYCLOSPORINE AND ROS Cyclosporine A (CsA), a cyclic undecapeptide of fungal origin, is the most widely used immunosuppressive drug in organ transplantation and in the treatment of autoimmune disorders (Table 136.2). However, its clinical use is often limited by the frequent and dose-dependent severe side effects, including nephrotoxicity, hypertension, hepatotoxicity, lymphoproliferative disorders, atherosclerosis of the graft and interstitial myocardial fibrosis (Rezzani, 2006). Nephrotoxicity is the main untoward consequence of CsA treatment; moderate to severe renal dysfunction has been, indeed, documented in ⬃30% of CsA-treated patients. On the basis of the length of treatment, two forms of CsA nephrotoxicity have been described: acute CsA treatment induces reversible reduction of both glomerular filtration rate (GFR) and renal blood flow, related to the afferent arteriolar vasoconstriction. This latter effect is associated with an increase in vasoconstrictor factors such as endothelin, thromboxane A2, angiotensin II and/or a decrease in vasodilators, such as prostacyclin and nitric oxide. Thus, an imbalance in the release of vasoactive substances may account for the reported renal vasoconstriction. On the other hand, when chronically administered, CsA can lead to irreversible renal failure due to the synergical effect of several factors, such as renal vasoconstriction, tubulointerstitial fibrosis, tubular atrophy and glomerular sclerosis (Rezzani, 2006). Even though the mechanisms of CsA toxicity are not fully understood, in recent years several experimental evidences have been collected on the involvement of oxidative stress in the toxic effects of CsA treatment. It has been reported that this drug induces membrane lipid peroxidation in several in vitro and in vivo experimental models, as well as in transplanted patients. In vivo, CsA increases lipoperoxidation in the rat kidney and liver, depletes the hepatic and renal pool of glutathione and impairs antioxidant defenses. Moreover, it has been reported that mRNA and protein levels of heme oxygenase-1, an enzyme responsive to changes in the redox status, vary after treatment with CsA (Rezzani, 2006). Several hypotheses have been proposed in order to correlate CsA treatment and oxidative stress, such as up-regulation of the kidney cytochrome P450-dependent ROSproducing system, perturbation of the balance between vasodilation–vasoconstriction, in turn responsible for tubular hypoxia–reoxygenation, increased formation of renal thromboxane A2, induction of nitric oxide production. In addition, a possible direct interference of CsA with the intracellular homeostasis of glutathione has been suggested (Rezzani, 2006).
1247
TABLE 136.2 Key features of cyclosporine nephrotoxicity. 1. Cyclosporine A (CsA), is a cyclic undecapeptide produced by the fungus Tolypocladium inflatum Gams 2. CsA is an immunosuppressant drug primarily used in humans to prevent organ transplant rejection 3. CsA inhibits interleukin-2 gene transcription and the transition of T-lymphocytes from the G0 to G1 phase of the cell cycle 4. CsA, although extensively used in kidney transplantation, causes important renal adverse effects, including acute and chronic renal dysfunction, hemolytic-uremic syndrome, hypertension, electrolyte disturbances and defects in urinary concentrating ability 5. Prolonged CsA administration may lead to structural changes which are no longer dose-dependent and reversible and cause end-stage renal failure 6. Although the mechanisms of nephrotoxicity have not been well defined, some evidence suggests that reactive oxygen species play a causative role
The hypothesis that CsA toxicity is mediated by ROS led investigators to use antioxidant molecules such as taurine, lipoic acid, melatonin and N-acetylcysteine to prevent or ameliorate its adverse effects (Rezzani, 2006). Since many plant-derived ‘phytonutrients’ are becoming increasingly known for their antioxidant activity, the use of plant-derived antioxidants against nephrotoxicity induced by treatment with CsA has been thoroughly explored in animal models. The administration of shallot (Allium ascalonicum) extract along with CsA counteracts its deleterious effects on renal dysfunction, oxidative stress markers and morphological changes (Wongmekiat et al., 2008). Lycopene, the carotenoid pigment found in tomatoes and other red fruits, ameliorates the CsA-induced pathological alterations including: tubular necrosis, degeneration, thickened basement membranes and intertubular fibrosis (Atessahin et al., 2007). Resveratrol, a naturally occurring phenolic compound abundantly present in grapes and red wine, protects against CsA-induced nephrotoxicity through a nitric oxide-dependent mechanism (Rezzani, 2006). Comparable results have been obtained using curcumin, the principal component of the Indian curry spice turmeric (Rezzani, 2006). Similarly, provinol, a polyphenolic extract obtained from red wine, prevents the increase of systolic blood pressure and nephrotoxicity in rat, through a mechanism that involves reduction of oxidative stress and iNOS expression, via nuclear factor-κB pathway (Rezzani, 2006).
1248
6
Fluorescence intensity/106 cells
Both dried black grape and aqueous garlic counteract CsA nephrotoxicity, reducing the malondialdehyde level in the kidney tissue, possibly by preventing oxidative reactions (Durak et al., 2007). Vitamin E protects renal function and structure when administered in vivo to CsA-treated rats (Rezzani, 2006). Moreover, it has been demonstrated that the combination of quercetin and vitamin E plays a protective role against the imbalance elicited by CsA between the production of free radicals and the antioxidant defense systems, suggesting that a combination of these two antioxidants may find clinical application where cellular damage is a consequence of ROS (Mustafi-Pour et al., 2008).
SECTION | III Tyrosol and Hydroxytyrosol
5 4 3
*
2 1 0
A
Control
DOPET
CsA
CsA + DOPET
Control
DOPET
CsA
CsA + DOPET
The aim of the in vitro study (Galletti et al., 2005) was to investigate the possible protective effect of DOPET on CsA-induced nephrotoxicity using immortalized renal proximal tubule cells (RPTc) from normotensive Wistar–Kyoto rats. This cell line has been frequently selected as a model system, since tubular cells represent in vivo the major target of CsA-induced nephrotoxicity both in humans and in animal models (Galletti et al., 2005 and references therein). In the RPT cellular model system, CsA toxicity is a relatively late event, since 95% of RPTc are still largely viable up to 8 h in the presence of 25 μM CsA. Conversely, prolonged incubation time leads to a dramatic decrease in cell viability, with 30% cell death after 20 h exposure to CsA. When RPTc have been incubated with CsA in the presence of 10 μM DOPET, ROS formation was measured using the dichlorofluorescein (DCF) assay. As shown in Figure 136.1A, CsA treatment results in a significant increase in ROS formation. Moreover, the CsA-induced fluorescence signal is totally quenched by DOPET treatment. However, this phenol appears completely ineffective in preventing the toxic effect of the drug and in restoring cell viability (Figure 136.1B). The effect of DOPET on CsA-induced membrane lipoperoxidation has also been evaluated (Figure 136.2). RPTc incubation in the presence of CsA significantly increases the lipoperoxidation products (TBARS) (30% above basal), confirming that CsA treatment exposes cells to an oxidative microenvironment. Again, DOPET effectively counteracts the increase in TBARS formation. The pivotal role played by glutathione (GSH) in cellular protection against free radical damage is well known. The effect of DOPET on CsA-induced imbalance of the GSH redox state is shown in Table 136.3. CsA treatment induces in RPTc a significant increase in oxidized glutathione (GSSG), resulting in a 50% reduction in the [GSH]/ [GSSG] ratio. This result supports a direct interference of
100 80 60 40 20 0
B
FIGURE 136.1 Effect of hydroxytyrosol (DOPET) on cyclosporine A (CsA)-induced reactive oxygen species (ROS) production (A) and cytotoxicity (B) in renal proximal tubule cells (RPTc). RPTc were exposed for 20 h to 25 μM CsA in the presence or absence of 10 μM DOPET. ROS production was detected by means of the fluorescent indicator dichlorofluorescein, as reported by Galletti et al. (2005). Cell viability was evaluated by the trypan blue exclusion method. Data are expressed as means ⫾ SD (n ⫽ 3) *p ⬍ 0.001 vs CsA. 3
TBARS nmol−1 μg protein
136.4 EFFECTS OF HYDROXYTYROSOL ON CYCLOSPORINE CYTOTOXICITY IN RAT RENAL TUBULAR CELLS
Cell viability (% of control)
120
2
* 1
0 Control
DOPET
CsA
CsA + DOPET
FIGURE 136.2 Effect of hydroxytyrosol (DOPET) on cyclosporine A (CsA)-induced lipid peroxidation in renal proximal tubule cells (RPTc). RPTc were incubated for 20 h with 25 μM CsA in the presence or absence of 10 μM DOPET. Data are expressed as means ⫾ SD (n ⫽ 3) *p ⬍ 0.05 vs CsA.
the drug with the intracellular GSH homeostasis. However, DOPET fails to provide any appreciable protection (Table 136.3) against the CsA-induced alterations in glutathione metabolism.
CHAPTER | 136 Effects of Hydroxytyrosol on Cyclosporine Nephrotoxicity
TABLE 136.3 Effect of hydroxytyrosol (DOPET) on glutathione redox state in CsA-treated RPTc. GSSG nmol mg⫺1 protein
GSH nmol mg⫺1 protein
GSH/GSSG molar ratio
Control
1.32 ⫾ 0.65
15.9 ⫾ 4.2
12 ⫾ 3.8
CsA 25 μM
2.85 ⫾ 1.14*
17 ⫾ 3.8
6.0 ⫾ 3.4*
CsA 25 μM ⫹ DOPET 10 μM
2.8 ⫾ 0.15*
18.3 ⫾ 0.54 6.5 ⫾ 0.15*
Effect of cyclosporine-A (CsA) on the glutathione redox state in RPTc. Control samples received CsA vehicle. The glutathione (GSH), the oxidized glutathione (GSSG) and the GSH/GSSG ratio values are reported as mean ⫾SD. Significance (t-test)/ANOVA: *p ⬍ 0.01 vs control.
The reported data indicate that DOPET effectively counteracts both CsA-induced ROS production and membrane lipoperoxidation in RPTc; however, the protection against the CsA-induced oxidative stress is not paralleled by an equivalent decrease in CsA cytotoxicity. On the basis of these findings, ROS generation induced by CsA does not appear strictly related to its nephrotoxicity; therefore, the generalization that antioxidants might exert a protective effect against the adverse effects of CsA, proposed by several authors and shared by nephrologists, is called into question by the reported in vitro approach.
136.5 CYCLOSPORINE NEPHROTOXICITY IN RATS: EFFECT OF HYDROXYTYROSOL IN VIVO To further elucidate whether oxidative stress is responsible for CsA toxicity, a recent study assayed the protective effect of DOPET on oxidative stress, renal histology and hemodynamic alterations induced in rats by chronic CsA treatment (Capasso et al., 2008). In order to evaluate CsA-induced superoxide production within cells of the abdominal aorta and renal artery, the changes in fluorescence resulting from the oxidation of dihydroethidium (DHE) were monitored. The red fluorescence generated by the binding of ethidium–DNA complex is considered an appropriate indicator of superoxide production within the cells. As reported in Figure 136.3, the abdominal aorta of CsA-treated animals presents a red fluorescent signal significantly brighter than controls. When the rats were treated with CsA plus DOPET the fluorescence intensity was similar to controls, indicating that the polyphenol is able to completely prevent the production of superoxide.
1249
This experimental approach indicates that also in vivo DOPET is able to quench CsA-induced ROS production both in aorta and renal artery. These results are fully in agreement with the data of the in vitro study previously reported. TBARS were measured in the rat kidney to evaluate the CsA-induced oxidative alteration of cellular components. A significant increase (p ⬍ 0.01) of TBARS was observable after CsA treatment (Table 136.4); the simultaneous administration of DOPET and CsA restored TBARS to control level, thus demonstrating that, also when administered in vivo, the polyphenol completely prevents the effect of CsA on lipid peroxidation. Moreover, chronic CsA treatment significantly increased the GSSG level (p ⬍ 0.01 versus control). Also in this case, the co-administration of DOPET completely reversed this effect. Accordingly, CsA treatment lowered the cellular [GSH]/[GSSG] ratio by 44%, while DOPET completely prevented this alteration. As already mentioned, clinical observations and experimental evidence (Capasso et al., 2008 and references therein) indicate that CsA administration is associated with major side effects, including hypertension and renal failure related with fibrosis and vasoconstriction. It has been proposed that ROS overproduction induced by CsA may lead to the inhibition of NO synthesis with the consequent appearance of hypertension. Consistently, as shown in Figure 136.4, starting from the second week of CsA treatment, an increase of about 15 mmHg in both systolic blood pressure (BP) and diastolic BP can be observed. According to literature data, CsA increased both systolic BP and diastolic BP by 16–17 mmHg at the end of the treatment period. However, the administration of DOPET did not yield any observable effect on CsA-induced hypertension. A severe impairment of renal hemodynamics is another side effect of chronic administration of CsA. Among the implicated mechanisms, there is general agreement that the major factor of the CsA effect on GFR is mediated by its action on afferent arteriolar resistance. In this study, GFR was measured by means of inulin clearance. The effect of DOPET on chronic CsA-induced renal failure is reported in Figure 136.5. DOPET per se did not exert any change on glomerular function (0.93 ⫾ 0.05 mL min⫺1 100 g⫺1 b.w.). CsA treatment significantly decreased GFR compared to control animals (0.51 ⫾ 0.03 versus 0.94 ⫾ 0.05 mL min⫺1 100 g⫺1 b.w., respectively) (p ⬍ 0.01), while DOPET, in combination with CsA, did not exert any protection (0.46 ⫾ 0.02 mL min⫺1 100 g⫺1 b.w.). Therefore, in contrast with its ROS-quenching effect, DOPET is unable to prevent both the increase in BP and the decrease in GFR. It is worth underlining that DOPET alone had almost no detrimental effect, confirming the low toxicity of this antioxidant phenol. The lack of a protective effect of DOPET on BP and kidney hemodynamics fits very nicely with the histological data, indicating that the antioxidant was unable to act on the arteriolopathy.
1250
SECTION | III Tyrosol and Hydroxytyrosol
Control
DOPET
CsA
CsA + DOPET
FIGURE 136.3 Effect of cyclosporine A (15 mg kg⫺1 day⫺1) and hydroxytyrosol (20 mg kg⫺1 twice daily) on the production of superoxide in the abdominal aorta. Superoxide concentrations within cells of the abdominal aorta were monitored by measuring the changes in fluorescence resulting from the oxidation of dichlorofluorescein. The figure has been reproduced with permission from Capasso et al., (2008).
TABLE 136.4 Oxidative stress markers evaluation after cyclosporine A (CsA) treatment. Control
DOPET (20 mg kg⫺1)
CsA (15 mg kg⫺1)
CsA (15 mg kg⫺1) ⫹ DOPET (20 mg kg⫺1)
TBARS (nmol/g tissue)
20.8 ⫾ 3.4
23.8 ⫾ 2.5
37.2 ⫾ 7.4a**
28.0 ⫾ 2.3b*
GSSG (nmol GSH/g tissue)
1.66 ⫾ 0.11
1.50 ⫾ 0.10
2.17 ⫾ 0.29a**
1.60 ⫾ 0.19b*
GSH/GSSG ratio
31.09 ⫾ 3.4
29.62 ⫾ 2.3
17.08 ⫾ 2.1a**
33.57 ⫾ 2.4b*
Lipid peroxidation was measured in kidney tissues as thiobarbituric acid reacting substances (TBARS); oxidized glutathione (GSSG) was assessed by enzymatic assay. Control group received CsA vehicle, CsA group received CsA 15 mg kg⫺1 day⫺1 and DOPET group received DOPET 20 mg kg⫺1 twice daily. Significance (t-test)/ANOVA: a ⫽ vs control; b ⫽ vs CsA; *p ⬍ 0.05; **p ⬍ 0.01.
136.6 CONCLUSIONS All together these observations lead to the conclusion that CsA-induced kidney injury is only partially due to oxidative stress. Moreover, the in vivo results are in agreement with in vitro data indicating that DOPET is able to completely prevent CsA-induced oxidative stress in rat tubular cells, but is ineffective in ameliorating the associated reduction of cell viability. The most reasonable interpretation of the data obtained in vivo is that the process leading to BP increase and GFR decrease is not necessarily related to ROS-dependent mechanism(s). Thus, even when CsA-induced oxidative stress is completely reverted by DOPET, renal failure and hypertension cannot be prevented, probably because
of other underlying mechanisms, such as artheriolopathy. Such a hypothesis seems to be in contrast with literature data showing that the administration of ‘antioxidant drugs’ like vitamin E (Rezzani, 2006) and lycopene (Atessahin et al., 2007) is able to reduce oxidative stress and ameliorate renal function after CsA treatment. It is highly likely that these two effects are mutually independent. Indeed, it should be underlined that both compounds, beside their antioxidant activity, exert key functions such as modulation of enzymatic activities and alteration of gene expression, which could account for their protective effect against CsA treatment. In this respect, it has been demonstrated that the overexpression of superoxide dismutase 1 by gene delivery, 3 days prior to the in vivo CsA administration, can partially reduce CsA-induced pathological alterations and inhibition of renal function (Rezzani, 2006).
mmHg
CHAPTER | 136 Effects of Hydroxytyrosol on Cyclosporine Nephrotoxicity
1251
Systolic blood pressure
140 135 130 125 120 115 110 105 100 0
5
10
15
20
25
Time (days)
mmHg
A
CsA+DOPET
CsA
DOPET
Control
Diastolic blood pressure
125 120 115 110 105 100 95 90 85 0
5
10
15
20
25
Time (days)
B
CsA+DOPET
CsA
DOPET
Control
FIGURE 136.4 Effect of cyclosporine A (15 mg kg⫺1 day⫺1) and hydroxytyrosol (20 mg kg⫺1 twice daily) on systolic and diastolic pressure in rats. The rats were treated as described in the text. Diastolic and systolic blood pressure were measured on conscious rats using the tail method. Panel A shows the systolic pressure trend during the treatment period. Panel B shows the diastolic pressure results. †p ⬍ 0.05, ††p ⬍ 0.01, †††p ⬍ 0.001 versus day 0; *** p ⬍ 0.001 versus control. The figure has been reproduced with permission from Capasso et al., (2008).
thus supporting the view that kidney injury by CsA is mainly related to pathogenetic mechanisms independent from oxidative stress.
GFR (mL−1min−1g 100−1g B.W.)
0.2 1 0.8 0.6
SUMMARY POINTS
**
0.4
●
0.2 0 Control
DOPET
CsA
CsA + DOPET
FIGURE 136.5 Effect of hydroxytyrosol on the glomerular filtration rate (GFR) measured in rats treated with cyclosporin A. GFR was measured by inulin clearance. Inulin concentration in plasma and urine was calculated by the colorimetric method. **p ⬍ 0.001 vs control. The figure has been reproduced with permission from Capasso et al., (2008).
In conclusion, the in vitro and in vivo effective DOPET protection from CsA-induced oxidative stress is only associated with mild effects on histological damage and does not affect the altered glomerular function and the hypertension,
●
●
●
Cyclosporine A (CsA), although widely used as an immunosuppressive drug, exerts frequent and dosedependent cytotoxic effects, probably related to ROS overproduction. Hydroxytyrosol (DOPET), the powerful olive oil antioxidant, could counteract CsA cytotoxicity through its scavenging properties. In vitro studies with rat renal tubular cells, the main target of CsA cytotoxic activity, have demonstrated that DOPET effectively counteracts ROS production and lipoperoxidation. However, these effects are not paralleled by an equivalent decrease in CsA cytotoxicity. The in vivo protective effect of DOPET has been assayed in rats. While exerting a significant protection toward CsA-induced oxidative stress, this polyphenol
1252
●
●
SECTION | III Tyrosol and Hydroxytyrosol
does not influence renal histological and hemodynamic alterations. In conclusion, DOPET protection is only associated with mild effects on histological damage and does not affect the altered glomerular function and the hypertension. The data support the view that kidney injury by CsA is mainly related to pathogenetic mechanisms independent from oxidative stress.
ACKNOWLEDGMENTS The authors gratefully acknowledge the Oxford University Press for the kind permission to use Figures 1, 6 and 7 from Capasso G. et al., Nephrol. Dial. Transplant. 23, 1186–1195.
REFERENCES Atessahin, A., Ceribasi, A., Yilmaz, S., 2007. Lycopene, a carotenoid, attenuates cyclosporine-induced renal dysfunction and oxidative stress in rats. Basic Clin. Pharm. Toxicol. 100, 372–376. Bendini, A., Cerretani, L., Carrasco-Pancorbo, A., Gómez-Caravaca, A.M., Segura-Carretero, A., Fernández-Gutiérrez, A., Lercker, G., 2007. Phenolic molecules in virgin olive oils: a survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules 12, 1679–1719. Carluccio, M.A., Siculella, L., Ancora, M.A., Massaro, M., Scoditti, E., Storelli, C., Visioli, F., Distante, A., De Caterina, R., 2003. Olive oil and red wine antioxidant polyphenols inhibit endothelial activation: antiatherogenic properties of Mediterranean diet phytochemicals. Arterioscler. Thromb. Vasc. Biol. 23, 622–629. Capasso, G., Di Gennaro, C.I., Della Ragione, F., Manna, C., Ciarcia, R., Florio, S., Perna, A., Pollastro, R.M., Damiano, S., Mazzoni, O., Galletti, P., Zappia, V., 2008. In vivo effect of the natural antioxidant hydroxytyrosol on cyclosporine nephrotoxicity in rats. Nephrol. Dial. Transplant. 23, 1186–1195. Covas, M.I., de la Torre, K., Farré-Albaladejo, M., Kaikkonen, J., Fitó, M., López-Sabater, C., Pujadas-Bastardes, M.A., Joglar, J., Weinbrenner, T., Lamuela-Raventós, R.M., de la Torre, R., 2006. Postprandial LDL phenolic content and LDL oxidation are modulated by olive oil phenolic compounds in human. Free Radic. Biol. Med. 40, 608–616. D’Angelo, S., Ingrosso, D., Migliardi, V., Sorrentino, A., Donnarumma, G., Baroni, A., Masella, L., Tufano, M.A., Zappia, M., Galletti, P., 2005. Hydroxytyrosol, a natural antioxidant from olive oil, prevents protein damage induced by long-wave ultraviolet radiation in melanoma cells. Free Radic. Biol. Med. 38, 908–919. Deiana, M., Aruoma, O.I., Bianchi, M.P., Spencer, J.P.E., Kaur, H., Halliwell, B., Banni, S., Dessi, M.A., Corongiu, F.P., 1999. Inhibition of peroxynitrite dependent DNA base modification and tyrosine nitration by the extra virgin olive oil-derived antioxidant hydroxytyrosol. Free Radic. Biol. Med. 26, 762–769. de la Puerta, R., Ruiz Gutierrez, V., Hoult, R.S., 1999. Inhibition of leukocyte 5-lipoxygenase by phenolics from virgin olive oil. Biochem. Pharmacol. 57, 445–449.
Dell’Agli, M., Maschi, O., Galli, G.V., Fagnani, R., Dal Cero, E., Caruso, D., Bosisio, E., 2008. Inhibition of platelet aggregation by olive oil phenols via cAMP-phosphodiesterase. Br. J. Nutr. 99, 945–951. Della Ragione, F., Cucciolla, V., Borriello, A., Della Pietra, V., Pontoni, G., Racioppi, L., Manna, C., Galletti, P., Zappia, V., 2000. Hydroxytyrosol, a natural molecule occurring in olive oil, induces cytochrome c-dependent apoptosis. Biochem. Biophys. Res. Commun. 278, 733–739. Della Ragione, F., Cucciolla, V., Criniti, V., Indaco, S., Borriello, A., Zappia, V., 2002. Antioxidants induce different phenotypes by a distinct modulation of signal transduction. FEBS Lett. 532, 289–294. Durak, I., Cetin, R., Candir, O., Devrim, E., Kiliçoglù, B., Avci, A., 2007. Black grape and garlic extracts protect against cyclosporine A nephrotoxicity. Immunol. Invest. 36, 105–114. Fabiani, R., Rosignoli, P., De Bartolomeo, A., Fuccelli, A., Morozzi, G., 2008. Inhibition of cell cycle progression by hydroxytyrosol is associated with upregulation of cyclin-dependent protein kinase inhibitors p21WAF1/Cip1 and p27Kip1 and with induction of differentiation in HL60 cells. J. Nutr. 138, 42–48. Galletti, P., Di Gennaro, C.I., Migliardi, V., Indaco, S., Della Ragione, F., Manna, C., Chiodini, P., Capasso, G., Zappia, V., 2005. Diverse effects of natural antioxidants on cyclosporin cytotoxicity in rat renal tubular cells. Nephrol. Dial. Transplant. 20, 1551–1558. González-Santiago, M., Martín-Bautista, E., Carrero, J.J., Fonollá, J., Baró, L., Bartolomé, M.V., Gil-Loyzaga, P., López-Huertas, E., 2006. Onemonth administration of hydroxytyrosol, a phenolic antioxidant present in olive oil, to hyperlipidemic rabbits improves blood lipid profile, antioxidant status and reduces atherosclerosis development. Atherosclerosis 188, 35–42. Goya, L., Mateos, R., Bravo, L., 2007. Effect of the olive oil phenol hydroxytyrosol on human hepatoma HepG2 cells. Protection against oxidative stress induced by tert-butylhydroperoxide. Eur. J. Nutr. 46, 70–78. Manna, C., Della Ragione, F., Cucciolla, V., Borriello, A., D’Angelo, S., Galletti, P., Zappia, V., 1999. Biological effects of hydroxytyrosol, a polyphenol from olive oil endowed with antioxidant activity. Adv. Exp. Med. Biol. 472, 115–130. Manna, C., Galletti, P., Maisto, G., Cucciolla, V., D’Angelo, S., Zappia, V., 2000. Transport mechanism and metabolism of olive oil hydroxytyrosol in Caco-2 cells. FEBS Lett. 470, 341–344. Maiuri, M.C., De Stefano, D., Di Meglio, P., Irace, C., Savarese, M., Sacchi, R., Cinelli, M.P., Carnuccio, R., 2005. Hydroxytyrosol, a phenolic compound from virgin olive oil, prevents macrophage activation. Naunyn Schmiedebergs Arch. Pharmacol. 371, 457–465. Mustafi-Pour, Z., Zal, F., Monabati, A., Vessal, M., 2008. Protective effects of a combination of quercetin and vitamin E against cyclosporine A-induced oxidative stress and hepatotoxicity in rats. Hepatol. Res. 38, 385–392. Rezzani, R., 2006. Exploring cyclosporine A-side effects and the protective role-played by antioxidants: the morphological and immunohistochemical studies. Histol. Histopathol. 21, 301–316. Trichopoulou, A., Dilis, V., 2007. Olive oil and longevity. Mol. Nutr. Food Res. 51, 1275–1278. Visioli, F., Galli, C., 2002. Biological properties of olive oil phytochemicals. Crit. Rev. Food Sci. Nutr. 42, 209–211. Wongmekiat, O., Leelarugrayub, N., Thamprasert, K., 2008. Beneficial effect of shallot (Allium ascalonicum L.) extract on cyclosporine nephrotoxicity in rats. Food Chem. Toxicol. 46, 1844–1850.
Chapter 137
Investigation of the Inhibition of Platelet Activation and Antithrombotic Action of a Hydroxytyrosol-rich Olive Oil Wastewater Extract in Diabetic Subjects1 Claude Louis Léger2 EA ‘Nutrition Humaine et Athérogénèse’, Faculté de Médecine,