Metabolism of Nutrients by Gut Microbiota (Issn) [1 ed.] 9781788017480, 9781839160950, 9781839160967, 178801748X

Citation preview

Metabolism of Nutrients by Gut Microbiota

Food Chemistry, Function and Analysis Series editors:

Gary Williamson, Monash University, Australia Alejandro G. Marangoni, University of Guelph, Canada Graham A. Bonwick, AgriFoodX Limited, UK Catherine S. Birch, AgriFoodX Limited, UK

Titles in the series:

1: Food Biosensors 2: Sensing Techniques for Food Safety and Quality Control 3: Edible Oil Structuring: Concepts, Methods and Applications 4: Food Irradiation Technologies: Concepts, Applications and Outcomes 5: Non-­extractable Polyphenols and Carotenoids: Importance in Human Nutrition and Health 6: Cereal Grain-­based Functional Foods: Carbohydrate and Phytochemical Components 7: Steviol Glycosides: Cultivation, Processing, Analysis and Applications in Food 8: Legumes: Nutritional Quality, Processing and Potential Health Benefits 9: Tomato Chemistry, Industrial Processing and Product Development 10: Food Contact Materials Analysis: Mass Spectrometry Techniques 11: Vitamin E: Chemistry and Nutritional Benefits 12: Anthocyanins from Natural Sources: Exploiting Targeted Delivery for Improved Health 13: Carotenoid Esters in Foods: Physical, Chemical and Biological Properties 14: Eggs as Functional Foods and Nutraceuticals for Human Health 15: Rapid Antibody-­based Technologies in Food Analysis 16: DNA Techniques to Verify Food Authenticity: Applications in Food Fraud 17: Advanced Gas Chromatography in Food Analysis 18: Handbook of Food Structure Development 19: Mitigating Contamination from Food Processing 20: Biogenic Amines in Food: Analysis, Occurrence and Toxicity 21: Nutrition and Cancer Prevention: From Molecular Mechanisms to Dietary Recommendations 22: Health Claims and Food Labelling 23: Nutraceuticals and Human Health: The Food-­to-­supplement Paradigm 24: Nutritional Signalling Pathway Activities in Obesity and Diabetes 25: The Chemistry and Bioactive Components of Turmeric 26: Foodomics 27: Food Proteins and Peptides: Emerging Biofunctions, Food and Biomaterial Applications 28: Handbook of Antioxidant Methodology: Approaches to Activity Determination

29: Fats and Associated Compounds: Consumption and Human Health 30: Oral Processing and Consumer Perception: Biophysics, Food Microstructures and Health 31: Development of Trans-­free Lipid Systems and their Use in Food Products 32: Advanced Spectroscopic Techniques for Food Quality 33: Berries and Berry Bioactive Compounds in Promoting Health 34: Metabolism of Nutrients by Gut Microbiota

How to obtain future titles on publication:

A standing order plan is available for this series. A standing order will bring delivery of each new volume immediately on publication.

For further information please contact:

Book Sales Department, Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, CB4 0WF, UK Telephone: +44 (0)1223 420066, Fax: +44 (0)1223 420247 Email: [email protected] Visit our website at www.rsc.org/books

     

Metabolism of Nutrients by Gut Microbiota Edited by

Joseph F. Pierre

University of Wisconsin–Madison, USA Email: [email protected]

Food Chemistry, Function and Analysis No. 34 Print ISBN: 978-­1-­78801-­748-­0 PDF ISBN: 978-­1-­83916-­095-­0 EPUB ISBN: 978-­1-­83916-­096-­7 Print ISSN: 2398-­0656 Electronic ISSN: 2398-­0664 A catalogue record for this book is available from the British Library © The Royal Society of Chemistry 2022 All rights reserved Apart from fair dealing for the purposes of research for non-­commercial purposes or for private study, criticism or review, as permitted under the Copyright, Designs and Patents Act 1988 and the Copyright and Related Rights Regulations 2003, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of The Royal Society of Chemistry or the copyright owner, or in the case of reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK, or in accordance with the terms of the licences issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms stated here should be sent to The Royal Society of Chemistry at the address printed on this page. Whilst this material has been produced with all due care, The Royal Society of Chemistry cannot be held responsible or liable for its accuracy and completeness, nor for any consequences arising from any errors or the use of the information contained in this publication. The publication of advertisements does not constitute any endorsement by The Royal Society of Chemistry or Authors of any products advertised. The views and opinions advanced by contributors do not necessarily reflect those of The Royal Society of Chemistry which shall not be liable for any resulting loss or damage arising as a result of reliance upon this material. The Royal Society of Chemistry is a charity, registered in England and Wales, Number 207890, and a company incorporated in England by Royal Charter (Registered No. RC000524), registered office: Burlington House, Piccadilly, London W1J 0BA, UK, Telephone: +44 (0) 20 7437 8656. Visit our website at www.rsc.org/books Printed in the United Kingdom by CPI Group (UK) Ltd, Croydon, CR0 4YY, UK

Preface Metabolism of Nutrients by Gut Microbiota is a timely collection on the state of knowledge and recent advancements in microbiota research related to human nutrition, metabolism, and health. This book was motivated by a desire to produce a comprehensive summary of nutritionally related microbial–host interactions in a single volume. Our overarching goal was to cover the key topical areas relevant for career researchers, academic students, industry professionals, and the lay public alike to gain deeper insights into mechanisms and recent advances in the field of microbiome and metabolism. Over the past two decades, microbiome research has expanded exponentially, shifting from early characterization and description of microbial communities towards more recent emphasis on understanding causality, function, and identification of microbially modified metabolites that hold the key to deciphering host–microbial interaction. Metabolism of Nutrients by Gut Microbiota focuses first on dietary macronutrient intake, which is a dominant driver of human microbiota membership and function that has profound effects on health. In addition to macronutrient influences, we wanted to capture other key elements of how the microbiome is shaped and involved in health that would be either directly relevant to, or of great interest to, many of our readers. We choose to include the role of dietary prebiotics, probiotics, and synbiotics, pharmaceutical drug–microbial interactions, early life variables that affect microbiome assemblage following birth, and, finally, the mechanisms and effects of bariatric surgery in the treatment of obesity. Collectively, each of these microbiome altering factors have unique effects on health – in-­part through microbial interaction – with implications on body composition and metabolic setpoints, immune programming and response, drug metabolism and efficacy, and lifelong chronic and acute disease risk.

  Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

vii

viii

Preface

Metabolism of Nutrients by Gut Microbiota was made possible from the exceptional contributions of twenty-­t wo authors at nine universities who summarized their fields of research with impressive clarity and acumen. Our authors were challenged with summarizing complex findings in rapidly evolving fields of study and highlighting the key mechanistic roles that the microbiome plays in nutrition and metabolism. I would like to personally thank and acknowledge each of our contributing authors for sharing their unique expertise and their time. The result is an outstanding collection that contributes to the impressive book series on Food Chemistry, Function, and Analysis by the Royal Society of Chemistry and author experts from around the globe. We sincerely hope that you will enjoy reading this text as much as we enjoyed putting it together. Joseph F. Pierre

Contents Chapter 1 I ntroduction and Background to Microbiome Research  Joseph F. Pierre

1



1 2 3



1.1 Introduction to the Gut Microbiota  1.2 Approaches in Microbiome Research 1.2.1 Fingerprinting the Microbial Community  1.2.2 Identification and Classification of Microbial Membership and Their Functions  1.3 In Vivo Models for Investigating Microbial Causality in Nutrition and Metabolism  1.4 In Vitro Models to Study the Microbiome 1.4.1 Microbial Culture-­based Technologies to Study Host–Microbe Interactions  1.4.2 Cell Culture Model Systems to Study Host–Microbe Interactions  1.5 Heterogeneity of Mammalian Gut Microbes – Implications for Nutritional Science  1.6 Summary  References 

Chapter 2 M  etabolism of Dietary Carbohydrates by Intestinal Bacteria  Ebru Ece Gulsan, Farrhin Nowshad, Arul Jayaraman and Kyongbum Lee

2.1 Introduction  2.2 Dietary Fiber 2.2.1 Cellulose 

5 8 9 9 10 10 12 12 18

18 21 22

 Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

ix

Contents

x



2.2.2 Hemicellulose  2.2.3 Pectin  2.2.4 Oligosaccharides  2.2.5 Resistant Starch  2.2.6 Lignin  2.3 Polyphenols  2.4 Amino Sugars  2.5 Tools for Identifying Products of Microbiota Metabolism  2.6 Future Directions  Acknowledgements  References 

22 24 25 30 31 31 36 38 40 42 42

Chapter 3 T  he Microbiome and Amino Acid Metabolism  N. E. Diether and B. P. Willing

48



48 49 49 50 51 52 52 52 53 53 54 54 54 55 55 57 58 58 58



3.1 Introduction  3.2 Microbes and Protein in the Gut Compartments 3.2.1 Microbes and Protein in the Small Intestine  3.2.2 Microbes and Protein in the Large Intestine  3.3 Metabolic Pathways of Proteolytic Fermentation 3.3.1 Deamination  3.3.2 Decarboxylation  3.3.3 Stickland Reaction  3.4 Metabolites Produced by Proteolytic Fermentation 3.4.1 Ammonia  3.4.2 Amines  3.4.3 Branched Chain Fatty Acids  3.4.4 Phenols and Indoles  3.5 Fermentation of Aromatic Amino Acids 3.5.1 Tryptophan  3.5.2 Tyrosine  3.5.3 Phenylalanine  3.6 Proteolytic Fermentation and Health 3.6.1 Proteolytic Metabolites and the Gut–Brain Axis  3.6.2 Irritable Bowel Syndrome (IBS) and Inflammatory Bowel Disease (IBD)  3.6.3 Colorectal Cancer  3.6.4 Metabolic Syndrome  3.7 Conclusions  References 

59 61 62 62 63

Chapter 4 F  at Absorption, Metabolism, and Global Regulation  Nayaab Rana, Peymaun Mozaffari, Danial Asim and Kristina Martinez-­Guryn

68



68 70

4.1 Introduction  4.2 Obesity and the Gut Microbiota 

Contents



xi

4.3 Dietary Modulation of the Gut Microbiota 4.3.1 Diet-­mediated Shifts in Gut Microbiota Community Composition  4.3.2 Direct Microbial Metabolism of Dietary Components  4.4 Local Effects of Gut Microbes on the Gastrointestinal Tract 4.4.1 Lipid Digestion and Absorption  4.5 Microbial Regulation of Peripheral Metabolic Organs 4.5.1 Gut Microbiota–Liver Interactions  4.5.2 Gut Microbiota–Adipose Interactions  4.5.3 Gut Microbiota–Muscle Interactions  4.6 Conclusion  Conflicts of Interest  Acknowledgements  References 

72 72 74 76 76 78 78 80 81 81 82 82 82

Chapter 5 P  robiotics, Prebiotics, and Synbiotics in Human Health  Olivia L. Wells, Sidharth Mishra and Hariom Yadav

86



86 87 87 89 90



5.1 Introduction 5.1.1 Probiotics  5.1.2 Prebiotics  5.1.3 Synbiotics  5.2 The Gut Microbiome and Human Health  5.3 Role of Probiotics, Prebiotics, and Synbiotics in Illnesses Related to Gut Dysbiosis 5.3.1 Introduction  5.3.2 Intra-­intestinal Disorders  5.3.3 Extra-­intestinal Disorders  5.4 Conclusion  Abbreviations  References 

92 92 92 99 105 106 107

Chapter 6 M  icrobial Drug Interactions and Human Health  Zehra Esra Ilhan and Melissa M. Herbst-­Kralovetz

120



120



6.1 Introduction  6.2 Drugs Perturb Gut Microbiota Structure, Function, and Host Health  6.3 Gut Microbiome as a Modulator of Pharmokinetics  6.4 Microbial Biochemistry of Drug Metabolism  6.5 Approaches and Model Systems to Study Gut Pharmacomicrobiomics  6.6 Personalized Pharmacomicrobiomics and the Future of Microbiome-­centric Therapies  6.7 Summary  References 

121 125 129 131 135 137 137

Contents

xii

Chapter 7 E  arly Life Microbiome Colonization and Human Health  Tahliyah S. Mims, Jun Miyoshi and Joseph F. Pierre

150



150



7.1 Introduction  7.2 Microbiome Acquisition and Factors Shaping Composition 7.2.1 During Gestation  7.2.2 Following Birth or Delivery  7.2.3 The Effect of Birth Route  7.2.4 The Role of Infant Nutrition  7.2.5 Exposure to Antibiotics  7.3 Early Life Microbiome and Normal Organ System Development 7.3.1 Digestive Organ Development  7.3.2 Brain Growth and Development  7.3.3 Bone Mass  7.4 Early Life Microbiome and Obesity Risk  7.5 Early Life Microbiome and Immunological Disturbance 7.5.1 Inflammatory Bowel Disease (IBD)  7.5.2 Asthma and Food Allergy  7.5.3 Vaccine Efficacy  7.6 Conclusions  References 

151 151 152 153 155 156 158 158 159 160 161 162 162 163 164 164 165

Chapter 8 T  he Gut Microbiome and Metabolic Surgery  Mehdi Chaib, Matthew J. Davis, Stacey Kubovec, Liza Makowski and Joseph F. Pierre

173



173 176 177 177 179



8.1 Introduction and History of Bariatric Surgery  8.2 Types of Bariatric Surgery  8.3 Mechanisms of Metabolic Improvement 8.3.1 Role of the Endocrine System  8.3.2 Role of the Nervous System  8.4 Microbiome and Bile Acid Changes Following Metabolic Surgery 8.4.1 The Microbiome in Obesity  8.4.2 The Microbiome and Metabolic Surgery  8.4.3 Bile Acid Metabolism and Signaling  8.5 Effects of Metabolic Surgery, Microbiome, and Bile Acids on Host Immunity 8.5.1 The Microbiome and Host Immunity  8.5.2 Bile Acids and Host Immunity  8.5.3 Metabolic Surgery and Host Immunity  8.6 Areas of Current and Future Research  8.7 Conclusion  References 

Subject Index 

180 180 182 184 186 186 187 187 188 188 189 196

Chapter 1

Introduction and Background to Microbiome Research Joseph F. Pierre* Department of Nutritional Sciences, College of Agriculture and Life Science, University of Wisconsin–Madison, Wisconsin, USA *E-­mail: [email protected]

1.1  Introduction to the Gut Microbiota Microbial communities colonize the gut and virtually all other body compartments, including the skin, mammary ducts, respiratory tract, and, as recent evidence supports, even the circulation. These dynamic communities are fundamentally involved in homeostasis and disease progression under environmentally and genetically shaped susceptible conditions. Through their assistance with digestion and fermentation, stimulation of nutrient absorption and endocrine regulation, production of vitamins, priming and agonism of immune education and response, and production of many small molecules, a growing number of relevant host–microbial interactions have been uncovered rapidly by microbiome researchers in recent years.1 Despite the enormous effect microbiome research has already had on our understanding of health and many diseases, this field is still young and continues to evolve in both the amount of empirical data collected annually and through the continuous improvements in the methodological approaches and techniques used to explore the microbiome. Scientists had historically long attempted to investigate microorganisms, but for hundreds of years the investigation into microbial life forms   Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

1

2

Chapter 1

was limited by what could be observed under the microscope or by indiscriminately cultivating organisms in artificial broths and solid substrates. While historical descriptions over several centuries more or less accurately described major microbial kingdoms – bacteria, fungi, and protists – the work of Louis Pasteur most notably first tied the importance of microbes to human health with his experimentation and theory of germs. Pasteur posited that certain microbes could cause specific diseases when introduced, setting into motion an unravelling of the superstition surrounding the origins of disease and improved understanding of microbial life. This new framework of empirical testing of microbes led Robert Koch to demonstrate that a single bacterium – bacilli – could cause anthrax in animals.2 Koch subsequently developed criteria, termed Koch's postulates, for the empirical determination for the role of bacteria in disease aetiology. These postulates included the necessity to isolate suspected bacteria in all cases of a specific disease (but not from healthy individuals), successfully initiation of disease when that isolated microbe was introduced to a healthy host, and subsequent resolution of disease when that suspected microbe was eliminated. It is now apparent that the majority of microbes that live within us are not readily grown in traditional laboratory media, instead requiring unique energetic substrates, the presence or absence of specific atmospheric gases, or specific mutualistic or even parasitic interactions with other unique members of the gut community and their metabolites to thrive. These initial challenges led to the rise of culture independent methods, including next-­generation sequencing, that allowed full ecological characterization of isolated microbial DNA. Much of this early research has been around metabolism and nutrition.3–5 In the early 2000s, these techniques first led to the rapid growth in understanding of the complex microbial communities in the gut, where the number of microbial cells is equal to the number of human cells in the human body, but with substantial genetic diversity at 100 to 150 times our own.3,6 This diversity is encapsulated in enormous numbers of bacteria, reaching up to 1012 bacteria per gram in the distal intestine.7 The enormous genetic capacity harboured by gut microbes has led to the concept that the microbiome is a virtual mammalian organ, being shaped by other host homeostatic and immune systems, and one that can subsequently be transplanted in composition and function between hosts with variable degrees of success. Microbiome research over the past two decades has done much to shift the viewpoint of our microbes towards them being helpful participants in normal development, homeostasis, and nutrition, revising an outdated concept of microbes as simply harmful pathogens and disease-­causing organisms.

1.2  Approaches in Microbiome Research In order to determine community membership, along with functional characteristics of the microbial community, a number of technologies can be used either independently or in parallel as multi-­omics-­based platforms. These technologies enabled culture-­independent (next-­generation sequencing)

Introduction and Background to Microbiome Research

3

or culture-­dependent approaches (anaerobic chambers, fermentation cultivar systems) to identify and isolate novel microbial strains that may contribute individually or as keystone community members of broader ecology in response to various perturbations, and in particular, nutrition. The following sections outline tools and techniques used in microbiome research to explore and elucidate the effects of and interplay between nutrition and dietary intake on gut microbes.

1.2.1  Fingerprinting the Microbial Community 1.2.1.1 Microbial Gene Amplicon Sequencing Techniques The identification and use of highly conserved regions of prokaryotic rRNA found in all bacteria and fungi enabled amplification-­based sequencing approaches. Amplicons are generally 150 to 250 nucleic acid base pairs in length. Studies examining humans and animals routinely rely on the 16S rRNA marker gene amplicon sequencing platforms, which include Sanger-­based sequencing, Roches454, PacBio, IonTorrent, and Illumina MiSeq/HiSeq/NexSeq platforms.8 Despite slightly different technological approaches, these platforms can each generate millions of short read sequences (amplicons) along with unique barcodes for identifying the source of each sequence against specific samples. These sequencing platforms were complemented by advanced computation approaches allowing analysis of the millions of sequences generated, including through mothur,9 Quantitative Insights Into Microbial Ecology (QIIME1 and 2),10 and Minimum Entropy Decomposition (MED).11 Each technique enables insight into microbial community composition, community diversity, and numerous methods to determine relatedness and unique signatures of microbial communities. Dependence on 16S rRNA amplicon sequencing inherently targets bacteria, which are the predominant colonizers, making up roughly 99% of microbial cell numbers, but are not the only kingdom of microorganisms, fungi and yeasts, viruses, and protists are excluded. Advancements of rRNA amplicon sequencing approaches have recently been developed to target other kingdoms and domains, including yeasts, and are more commonly being used to study human and animal health and disease. The importance of these communities remains debated, as compared with bacterial communities that remain relatively stable within individuals,12 the fungal membership changes more considerably between timepoints, with only 20% of species found consistently through temporal sampling.13 These analyses are complicated by the number of environmental and dietary ingested yeast and fungal species that may be transient and limit detection of true colonizers in individuals. To address these challenges, additional rRNA regions, such as the 18S rRNA subunit and the internal transcribed spacer 1 and 2 (ITS1 and ITS2) are used for phylogenetic assignment of eukaryotic microorganisms, specifically yeast and fungi.14 Another challenge in eukaryotic microbial research is obtaining

4

Chapter 1

well developed and curated databases for assigning reads generated from next generation platforms. While bacterial and archaeal databases, including SILVA and GreenGenes, are fairly well established, databases specific to yeasts and fungi are still under relatively recent but rapid curation and validation, including UNITE and Targeted Host-­associated Fungi (THF). Additionally, since sequencing of complex eukaryotic communities is a relatively new technique, many of the detected organisms have not been isolated or cultured from the host, leading to challenges in accurate classification and questions about their origins from the gut, diet, and environment. As a counter example of the importance of yeast in human health and disease, the members of the well described genus of Candida have no known reservoir outside of the mammalian gut and are considered true mammalian gut residents.15 The members of this genus of fungi are also well established opportunistic pathogens and are especially problematic in immune compromised individuals. However, the well-­studied yeast genus Saccharomyces, members of which are used in fermentation of food products, remain questionable as core community residents, as their detection in the gut may be the result of dietary intake.16 Functional understanding of fungal organisms has given some insight into the fidelity of eukaryotic colonization. For example, the yeast Malassezia have lost the ability to synthesize lipids and therefore require the host for their metabolic substrate, rendering them likely to be true mammalian colonizers. Other fungi, including Debaryomyces and Penicillium, although commonly found in the gut, cannot readily replicate at mammalian body temperature and their detection is, to date, considered environmental contamination.16 Importantly, host dietary intake based on protein rich or carbohydrate rich diets has been demonstrated to shift the interkingdom dynamics between bacteria and fungi.17 Accordingly, greater resolution of bacterial and other rare microbial members and their role in gut ecological dynamics is needed to understand the interactions between prokaryotes and eukaryotes and their synergistic function in influencing host homeostasis. These advances will probably come through synergy in deeper sequencing, further database curation, and increased sequencing coverage that will enable a more comprehensive snapshot of the complex inter-­kingdom colonizing populations.

1.2.1.2 Transfer RNA (tRNA) Sequencing (seq) Techniques Beyond 16S rRNA amplicon sequencing for microbial community fingerprinting, recent advances in microbially-­derived transfer RNAs (tRNAs) that facilitate translation of messenger RNA protein are also used to distinguish microbial communities with high accuracy. This method was originally developed for isolated microbial cultures but was then coupled with enzyme treatment to analyse demethylated tRNA, allowing insights into both tRNA transcripts and tRNA post-­translational modifications.18 With similar computational approaches to those for 16S rRNA, microbial tRNA reference libraries can produce highly accurate phylogenetic community analysis. In addition to taxonomic information, data on protein expression provides an

Introduction and Background to Microbiome Research

5

opportunity to obtain more advanced functional insights, especially in the context of investigating the role of nutrition and dietary interactions with microbial functions.

1.2.2  I dentification and Classification of Microbial Membership and Their Functions 1.2.2.1 Sequence-­based Approaches 1.2.2.1.1  Shotgun Metagenomics.  Large and rapid amounts of descriptive data can be obtained from amplicon sequencing-­based techniques, however little functional information can be generated from taxonomic description alone. Some efforts to combine taxonomic descriptions with known metagenomes of functional genes has been made, for instance through development of the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt), which is used to predict Kyoto Encyclopaedia of Genes and Genomes (KEGG) ortholog functions and metabolic pathways harboured by microbial communities based on 16S rRNA marker gene amplicon sequences.19 However, as databases are incomplete and variably updated, new and rare members of the community of curated genome databases may be missing or underrepresented and these techniques are limited in capturing the true gene functions and metabolic pathways within microbial communities.20 The limitations of amplicon-­based tools (16S, 18S, ITS rRNA) can be more readily overcome through the use of high-­throughput shotgun metagenomic sequencing since this method provides untargeted collection of all genetic content isolated in a microbial sample. These approaches are also valuable because they include genetic capture of fungal, virus and bacteriophage, in addition to bacterial, genomes. To date, two general approaches are used: mapping to reference databases or de novo assembly of sequenced reads. For sequence mapping purposes, on-­line servers are available, including Metagenomic Rapid Annotations using Subsystems Technology (MG-­RAST)21 and J. Craig Venter Institute (JCVI) Metagenomics Reports (METAREP).22 While computer based programs, such as Human Microbiome Project Unified Metabolic Analysis Network (HUMAnN),23 are utilized locally. De novo assembly is performed after functional annotations are performed, such as with platforms to reveal metagenome assembled genomes (MAGs). Further, several assembly programs are available, including khmer,24 and visualization tools have been developed for these analyses, including analysis and visualization platform for 'omics data (Anvi'o).25 Finally, strategies have been developed to complement metagenomic data with 16S sequencing, including ribosomal flanking region-­sequencing (RiboFR-­Seq), which provides 16S variable region information as well as the immediate protein-­coding genes surround the 16S gene.26 These advanced and functionally informed strategies are enabling greater insights into metabolic capacity by eliminating reliance on 16S based inferences as a stand-­alone.

6

Chapter 1

1.2.2.1.2  Single-­cell Genomics Coupled with Next-­Generation Sequencing Approaches.  In addition to collective microbial community DNA, much greater functional insights have been made using single-­cell sequencing. These strategies are especially useful for understanding the metabolic and functional capacity of rare or low abundance microbes within samples. This strategy is more difficult to employ as it includes initial isolation or enrichment of the microbe of interest, using flow cytometry or other antibody-­based identification and enrichment strategies. Following isolation, microbe identity can be confirmed by 16S rRNA amplification and whole genome-­based sequencing, with the computation analyses strategies described above. For greater review of the single-­cell genomic isolation and sequencing techniques, see the technical review by Qin et al.27 1.2.2.1.3  Metatranscriptomics.  Metagenomic shotgun sequencing and single-­cell genomics used separately or in combination, can aid in identifying gene content and function of gut microbiota communities, yet the activity or abundance of microbial gene expression cannot be discerned from genomic DNA-­based approaches alone. Metatranscriptomic shotgun sequencing (RNAseq) in combination with metagenomics is one strategy employed to identify the genomic potential as well as the active microbial genomes. This is carried out by isolating total RNA from the microbial community followed by enrichment for RNA [mRNA, long intergenic non-­coding RNA (lincRNA), and microRNA] and fragmentation. RNA is then converted to complementary DNA via reverse transcriptase with oligo(dT) primers and/or random hexamers and constructed libraries can then be sequenced.28 Despite providing insight into the activity of the microbial whole-­genome, this technology can be hampered by technical issues that, to date, limit its effectiveness. For instance, integrity of gut microbial RNA can be compromised by sample collection and storage where RNA quality is compromised, leading to insufficient yields of high-­quality microbial RNA, limiting purification efficiency and sequence fidelity. Furthermore, remnant RNA preservation solutions can interfere with downstream library preparation, biasing sequencing results. However, if these limitations are overcome, downstream data analysis for metatranscriptomic sequencing data then rely on similar strategies to shotgun metagenomic analysis, including Anvi'o and HUMAnN,23,25,28 which allows for taxonomic assignment coupled with complementary identification of actively expressed gene functions and greater insight into microbial functions within a given community and environment. 1.2.2.1.4  Long Read Sequencing Technology.  In contrast to amplifying restricted regions of the bacterial 16S or fungal ITS gene, other technologies have been optimized to amplify longer regions of microbial genomes. These platforms provide greater depth resolution of community membership and metabolism, but the current lack of functional annotation can be problematic. One way to overcome annotation gaps is to improve coverage and computationally perform assembly, which benefits from longer sequencing read

Introduction and Background to Microbiome Research

7

lengths (>10 000 bp). Long read technologies include the Oxford Nanopore MinION,29 the Pacific BioSciences single molecule real time (SMRT) sequencer,30 and Illumina Moleculo.31 In depth reviews of these approaches benefits, limitations, and pitfalls are described elsewhere.32,33

1.2.2.2 Non Sequence-­based Approaches to Identify Microbial Functions In addition to genome and transcript analysis, greater emphasis is now being placed on the functional outputs of microbial metabolism, specifically through the quantitation of microbial modified metabolites. Microbial metabolites are abundant in the gut lumen but also throughout the systemic circulation and in host organs, where they are assessed by targeted or untargeted metabolomic, proteomic, and lipidomic approaches. These analyses are providing important insights into microbial community function and roles in mammalian biology. Contributions of microbial metabolites are collectively part of the host's global metabonome. Untargeted and targeted methodologies are used to detect host and microbial inorganic and organic metabonomes through the use of gas chromatography–mass spectrometry (GC-­MS), liquid chromatography–mass spectrometry (LC-­MS), and nuclear magnetic resonance spectrometry (NMR), for assessment of organic lipids, amino acids, and simple and complex carbohydrates. Furthermore, introduction of isotope labelled dietary nutrients, especially those not digestible by host enzymes (such as C13 labelled inulin) enable investigators to delineate microbially produced metabolites generated from dietary components from those metabolites produced by the host. Another approach to understanding the role of microbial generated metabolites is through the employment of germ-­free animals (described in greater detail below) compared with conventional counterparts. Since germ-­free animals lack any microbial colonizers, comparisons can be made of systemic and secreted metabolite profiles in the presence and absence of microbes that allow characterization of microbial generated or influenced metabolites compared with those generated by the host alone. These studies have generated evidence of unique metabolites found in the major organs, such as the kidney, heart, and brain.34–38 In addition to lipids, amino acids, and carbohydrates, many other vastly diverse small molecules and peptides are detected, with their effects on the host remaining largely unknown.39 Despite the classical limitations of culturing microbes in the laboratory, another approach to understand the role of microbial metabolites is culturomics-­based analysis, which complements metabolomic quantification and 16S sequencing on isolated and cultured microorganisms.40,41 These techniques have led two databases, including the Culturable Genome Reference (CGR), which contains over 6000 indigenous isolates.42 One strength of this type of approach is the ability to identify toxins or harmful metabolites generated by microbial isolates, especially compared with highly similar

8

Chapter 1

microbes based on taxonomic assignment. For instance, functional differences have been observed in strains of Clostridium butyricum isolated from neonates with necrotizing enterocolitis (NEC) compared with controls,43 indicating that the taxonomic presence of this bacteria is insufficient for disease without production of their toxins. By identifying microbes from diseases and environments, the complex metabolite signatures can be assessed by metabolomic quantification and comparison with databases, including MetaCyc,44 Human Metabolome Database (HMDB),45 SetupX and BinBase.46

1.3  I n Vivo Models for Investigating Microbial Causality in Nutrition and Metabolism Humans are inseparably associated with their microbiomes – often beginning at or just before birth (see Chapter 7) – and determining causal effects of microbes on the host is not directly possible. To overcome this limitation, experimental models have been developed that allow the study of microbes in microbially naive hosts, most notably through the development of germ-­ free animal colonies. The most common germ-­free animals currently used in medical research are mice, due to their size, cost, short life cycles, similar organ and immune functions to those of humans, and the large number of genetic mutants available. Other germ-­free animals include rats, guinea pigs, swine, poultry, zebrafish, fruit flies, and nematodes. The goal of germ-­free animal research is to begin with an animal devoid of any detectable microbes, including bacteria, archaea, fungi, protists, bacteriophages, or viruses. The first successful germ-­free animal studies were performed by Nuttall and Thierfelder in the early 1900s, where guinea pigs were maintained under sterile conditions for two weeks, providing the first evidence that complex mammals could survive in the absence of microorganisms.47 James Reyniers established the first sustainable germ-­free mouse colony in 1931 at the University of Notre Dame. His approach was to perform caesarean section on timed pregnant mice, transfer pups into sterile isolators, and rear animals for multiple generations, an approach that has remained almost unchanged to this day. Interest in the role of microbes in health and disease has continued to grow over the past century, especially following the widespread acceptance and use of antibiotics. Germ-­free mice are especially useful as they can be moved into dedicated sterile experimental isolators or positive pressure cages where individual microbes (monocolonization) or complex communities of microbes are introduced or transplanted (termed conventionalization or transfaunation), leading to an animal with a controlled microbiome membership (gnotobiotics; known life).48 All cages, bedding, food and water, and other supplies are irradiated or more commonly autoclaved before introduction to the isolators or cage. Animals can be administered microbes before or at birth (by colonizing the pregnant mother) or at any point in the life cycle. Controlling the timing and composition of microbes as well as the hosts genetics,

Introduction and Background to Microbiome Research

9

diet, and environment, is the most robust way to causally determine the role of microbes in mammalian health, immune function, metabolism, and disease risk. Following the advent of next generation sequencing, the utility of germ-­ free mice and piglets has again become popular, largely because of organ, immune, and metabolic similarities between mice, pigs, and humans depending on the focus of the study.49,50 Comparison of host metatranscriptomes of germ-­free vs. conventional piglets revealed that almost 70% of the transcriptome in gut and systemic organs, especially immune specific genes, are influenced by gut and mucosal associated microorganisms.51 It should be noted that certain limitations exist in germ-­free animals, as microbes and their metabolites are paramount to normal neurological, immune and physiological development, so microbial colonization, especially at later timepoints in the animals lifecycle are not fully translational to the effects of these microbes in conventionally housed and microbially competent animals. Despite these limitations, insights into the nutritional requirements of animals, contribution of microbial colonizers to host metabolism, and effects on growth rates have been gleaned from germ-­free animals when compared with conventionally reared control animals.48 In addition to mice and pigs, other vertebrate germ-­free animals have been developed, including zebrafish.52 Fish are useful due to their extremely short life cycle, low cost of housing per animal, modifiable genetics, and transparent organs through development. In regard to nutrition, lipid digestion and absorption has been investigated in great detail using zebrafish.53 Like all animal models, fish also have limitations in the translational relevance to humans, specifically differences in body temperature, organ arrangement and function, and vastly different microbial members that thrive in marine environments. Invertebrate germ-­free animal models include Drosophila melanogaster54 and Caenorhabditis elegans.55 Collectively, the use of germ-­free animals has contributed valuable insights into host–microbial interactions and has much to offer in advancing our understanding of host–microbial interactions in the context of nutrition in the years to come.

1.4  In Vitro Models to Study the Microbiome 1.4.1  M  icrobial Culture-­based Technologies to Study   Host–Microbe Interactions In vitro model systems have been developed to further our understanding of host–microbial interactions. These simplified experimental models allow careful control of the microenvironment including nutrient composition, microbial membership, metabolite formation and release, partial gas pressures, and pH changes. One example is chemostats, which range from simple to complex equipment meant to mimic the various regions the gastrointestinal tract and which are used to explore the role of microbes and their community dynamics in response to available nutrients.56 Perhaps the most

10

Chapter 1

technologically advanced chemostat, the twin simulator of the human intestinal microbial ecosystem (TWINSHIME®) enables investigators to study both the mucosal and luminal microbial populations as available nutrients move throughout the modeled gastrointestinal tract, resulting in a final stool output.57 Chemostats have enabled detailed understanding of microbial dietary utilization and metabolism.

1.4.2  C  ell Culture Model Systems to Study Host–Microbe Interactions Over the last decade, advances in mammalian cell culture have been made with the development of organoids, which can be generated from intestinal, hepatic, and neuronal precursor cells. Within the gut, pluripotent intestinal stem cells (piSCs) are isolated from animals or humans via biopsy or whole tissue sections and grown in culture to form intestinal ‘miniguts’. piSCs obtained from each region of the intestine recapitulate the epithelial biology of their source, including the differentiation patterns, abundances, and functions of epithelial cells for each region of the gut, including the stomach (gastroids), small intestine (enteroids) and large intestine (colonoids).58–60 Coculturing intestinal organoids with immune or isolated neurons (spheroids) and microbes has enabled the modeling of complex cell to cell interactions for focused investigation. While many organoids are grown as three-­dimensional structures within a collagen structure, this approach limits the study of the luminal microbes. This limitation has been overcome in-­part through the use of microinjectors, which transplant small volumes of microorganisms or solutions into the basolateral organoid compartment. However, an alternative approach is to grow organoids on a flat collagen basement membrane, where they form two-­dimensional monolayers. Several laboratories have used this approach to generate ‘gut-­on-­a-­chip’ tools, where careful control of media through a microfluidic chamber allows the study of epithelial cell growth, dynamics, and responses to luminal stimulus.61 The development of organoids has allowed unparalleled advancement into the study of intestinal disease, the regulation of epithelial signals that maintain gut homeostasis, microbial– host interactions, and nutrition.

1.5  H  eterogeneity of Mammalian Gut Microbes – Implications for Nutritional Science As described above, characterization of microbial community composition alone has limited functional value. This limitation is further exacerbated by the fact that humans (and other complex animals) display a large heterogeneity in their microbiome communities. For instance, two individuals with similar health status share only 10 to 30% of gut bacteria with one another. This variability it is driven by diet and environmental exposure, but also

Introduction and Background to Microbiome Research

11

genetics as genetically identical twins do harbor more microbial similarity than dizygotic twins or other family members when microbial membership is compared.62–65 A deeper assessment of microbial community metatranscriptomes has revealed that even fairly dissimilar metagenome communities often contain similar functional core features, such as carbohydrate fermentation, or other pathways for surviving in the mammalian gut. Therefore, a certain degree of microbial community heterogeneity is explained by the functional redundancy in the community that is preserved rather than specific taxonomic membership of a given community. Despite the noted variability, the composition of the microbiome has been linked to diseases in human populations, such as obesity, inflammatory bowel disease (IBD), and numerous other metabolic, immunological, and neurological disorders.64–66 The most advanced survey of the human microbiome was recently published with the second phase of the human microbiome project (HMP2), which investigated the microbiome in IBD, premature birth, and type-­2 diabetes using multi-­omics strategies (metagenomics, metabolomics, metatranscriptomics).67 The results of HMP2 demonstrate that despite variability in the taxonomic community, functional differences exist under each of these common human disease states. Given the large degree of variability in the gut microbiome, investigators need to be aware of limitations when planning, designing, and carrying out nutritional science research in humans and animal models. Specific to experimental animals, it is known that the sources of commercially available research animals contain disparate microbiome compositions that can alter the physiological outcomes of research studies.68,69 For example, a classical example was that mice from Taconic Farms contain segmented filamentous bacteria (SFB), while mice from the Jackson (JAX) laboratory do not. Comparison of mice between these vendors demonstrated that SFB were strong drivers of T helper (Th) 17 cluster of differentiation 4 positive (CD4+) T cells70 and disease progression was initiated by the presence of these bacteria. In addition, each individual animal facility, different rooms within the same facilities, and even individual cages within a given room can contain variable microbiomes.71,72 Subtle differences in the gut microbiota of research animals can alter the immunological and metabolic outcomes of animals, or lead to different microbiome communities and functions on different diets. One approach to these challenges has been to routinely mix cage bedding during experiments to normalize a core microbiome, and bank that bedding for future experimental replicates.73 Furthermore, the source of diets, ventilation systems, and even the investigators handling the animals can have environmental effects on gut microbiomes. These limitations can be overcome through thoughtful experimental design, frequent sterilization of surfaces and gloves used to handle animals, and the use of biological safety cabinets to prevent novel microbial introduction. Finally, appropriately powered and careful data analysis can be employed to successfully carry out studies of host–microbial interactions in nutritional science research.

12

Chapter 1

1.6  Summary The microbiome is now appreciated to influence host homeostasis and metabolism. While many techniques, approaches, and experimental models are available to examine microbial interactions in the setting of nutritional science, tools and methodologies used to explore microbial functional effects on the host continue to evolve. In the setting of nutrition, the timing and composition of diet is one of the most profound influencers of microbial community structure and function.74 Ingested nutrients form the energetic substrates that are utilized by the host and microbes alike, directly or indirectly. In addition, microbes ferment substrates and generate de novo nutrients otherwise unavailable to the host. In return, the host generates secreted metabolites that help stabilize the microbial population within the gut. The role of dietary composition therefore fundamentally orchestrates the host–microbial ecology in meaningful ways. The chapters that follow here will focus on the role of major dietary macronutrient intake, interactions between microbes and drug metabolism, the state of prebiotics, probiotics, and synbiotics, the role of microbes in bariatric surgically induced weight loss, and how initial microbial colonizers following birth shape lifelong chronic disease risk and metabolism.

References 1. I. Montoliu, U. Genick, M. Ledda, S. Collino, F.-­P. Martin, J. le Coutre and S. Rezzi, Genes Nutr., 2013, 8, 19–27. 2. L. S. King, J. Hist. Med. Allied Sci., 1952, VII, 350–361. 3. R. E. Ley, D. A. Peterson and J. I. Gordon, Cell, 2006, 124, 837–848. 4. B. L. Cantarel, V. Lombard and B. Henrissat, PLoS One, 2012, 7, e28742. 5. P. J. Turnbaugh, R. E. Ley, M. A. Mahowald, V. Magrini, E. R. Mardis and J. I. Gordon, Nature, 2006, 444, 1027–1131. 6. R. Sender, S. Fuchs and R. Milo, PLoS Biol., 2016, 14, e1002533. 7. E. T. Hillman, H. Lu, T. Yao and C. H. Nakatsu, Microbes Environ., 2017, 32, 300–313. 8. J. G. Caporaso, C. L. Lauber, W. A. Walters, D. Berg-­Lyons, C. A. Lozupone, P. J. Turnbaugh, N. Fierer and R. Knight, Proc. Natl. Acad. Sci. U. S. A., 2011, 108 Suppl 1, 4516–4522. 9. P. D. Schloss, S. L. Westcott, T. Ryabin, J. R. Hall, M. Hartmann, E. B. Hollister, R. A. Lesniewski, B. B. Oakley, D. H. Parks, C. J. Robinson, J. W. Sahl, B. Stres, G. G. Thallinger, D. J. Van Horn and C. F. Weber, Appl. Environ. Microbiol., 2009, 75, 7537–7541. 10. J. G. Caporaso, J. Kuczynski, J. Stombaugh, K. Bittinger, F. D. Bushman, E. K. Costello, N. Fierer, A. G. Peña, J. K. Goodrich, J. I. Gordon, G. A. Huttley, S. T. Kelley, D. Knights, J. E. Koenig, R. E. Ley, C. A. Lozupone, D. McDonald, B. D. Muegge, M. Pirrung, J. Reeder, J. R. Sevinsky, P. J. Turnbaugh, W. A. Walters, J. Widmann, T. Yatsunenko, J. Zaneveld and R. Knight, Nat. Methods, 2010, 7, 335–336.

Introduction and Background to Microbiome Research

13

11. A. M. Eren, H. G. Morrison, P. J. Lescault, J. Reveillaud, J. H. Vineis and M. L. Sogin, ISME J., 2014, 9, 968–979. 12. C. A. Lozupone, J. I. Stombaugh, J. I. Gordon, J. K. Jansson and R. Knight, Nature, 2012, 489, 220–230. 13. H. E. Hallen-­Adams, S. D. Kachman, J. Kim, R. M. Legge and I. Martínez, Fungal Ecol., 2015, 15, 9–17. 14. E. A. Witherden, S. Shoaie, R. A. Hall and D. L. Moyes, J. Fungi, 2017, 56. 15. T. A. Auchtung, T. Y. Fofanova, C. J. Stewart, A. K. Nash, M. C. Wong, J. R. Gesell, J. M. Auchtung, N. J. Ajami and J. F. Petrosino, mSphere, 2018, e00092-­18. 16. R. A. Hall and M. C. Noverr, Curr. Opin. Microbiol., 2017, 40, 58–64. 17. C. Hoffmann, S. Dollive, S. Grunberg, J. Chen, H. Li, G. D. Wu, J. D. Lewis and F. D. Bushman, PLoS One, 2013, 8, e66019. 18. M. H. Schwartz, H. Wang, J. N. Pan, W. C. Clark, S. Cui, M. J. Eckwahl, D. W. Pan, M. Parisien, S. M. Owens, B. L. Cheng, K. Martinez, J. Xu, E. B. Chang, T. Pan and A. M. Eren, Nat. Commun., 2018, 9, 5353. 19. M. G. I. Langille, J. Zaneveld, J. G. Caporaso, D. McDonald, D. Knights, J. A. Reyes, J. C. Clemente, D. E. Burkepile, R. L. Vega Thurber, R. Knight, R. G. Beiko and C. Huttenhower, Nat. Biotechnol., 2013, 31, 814–821. 20. T. J. Sharpton, Front. Plant Sci., 2014, 5, 209. 21. F. Meyer, D. Paarmann, M. D'Souza, R. Olson, E. Glass, M. Kubal, T. Paczian, A. Rodriguez, R. Stevens, A. Wilke, J. Wilkening and R. Edwards, BMC Bioinf., 2008, 9, 386. 22. J. Goll, D. B. Rusch, D. M. Tanenbaum, M. Thiagarajan, K. Li, B. A. Methé and S. Yooseph, Bioinformatics, 2010, 26, 2631–2632. 23. S. Abubucker, N. Segata, J. Goll, A. M. Schubert, J. Izard, B. L. Cantarel, B. Rodriguez-­Mueller, J. Zucker, M. Thiagarajan, B. Henrissat, O. White, S. T. Kelley, B. Methé, P. D. Schloss, D. Gevers, M. Mitreva and C. Huttenhower, PLoS Comput. Biol., 2012, 8, e1002358. 24. M. R. Crusoe, H. F. Alameldin, S. Awad, E. Boucher, A. Caldwell, R. Cartwright, A. Charbonneau, B. Constantinides, G. Edvenson, S. Fay, J. Fenton, T. Fenzl, J. Fish, L. Garcia-­Gutierrez, P. Garland, J. Gluck, I. González, S. Guermond, J. Guo, A. Gupta, J. R. Herr, A. Howe, A. Hyer, A. Härpfer, L. Irber, R. Kidd, D. Lin, J. Lippi, T. Mansour, P. McA'Nulty, E. McDonald, J. Mizzi, K. D. Murray, J. R. Nahum, K. Nanlohy, A. J. Nederbragt, H. Ortiz-­ Zuazaga, J. Ory, J. Pell, C. Pepe-­Ranney, Z. N. Russ, E. Schwarz, C. Scott, J. Seaman, S. Sievert, J. Simpson, C. T. Skennerton, J. Spencer, R. Srinivasan, D. Standage, J. A. Stapleton, S. R. Steinman, J. Stein, B. Taylor, W. Trimble, H. L. Wiencko, M. Wright, B. Wyss, Q. Zhang, E. Zyme and C. T. Brown, F1000Research, 2015, 900. 25. A. M. Eren, Ö. C. Esen, C. Quince, J. H. Vineis, H. G. Morrison, M. L. Sogin and T. O. Delmont, PeerJ, 2015, 3, e1319. 26. Y. Zhang, P. Ji, J. Wang and F. Zhao, Nucleic Acids Res., 2016, 44, e99. 27. J. Qin, R. Li, J. Raes, M. Arumugam, K. S. Burgdorf, C. Manichanh, T. Nielsen, N. Pons, F. Levenez, T. Yamada, D. R. Mende, J. Li, J. Xu, S. Li, D. Li, J. Cao, B. Wang, H. Liang, H. Zheng, Y. Xie, J. Tap, P. Lepage,

14

Chapter 1

M. Bertalan, J.-­M. Batto, T. Hansen, D. Le Paslier, A. Linneberg, H. B. Nielsen, E. Pelletier, P. Renault, T. Sicheritz-­Ponten, K. Turner, H. Zhu, C. Yu, S. Li, M. Jian, Y. Zhou, Y. Li, X. Zhang, S. Li, N. Qin, H. Yang, J. Wang, S. Brunak, J. Doré, F. Guarner, K. Kristiansen, O. Pedersen, J. Parkhill, J. Weissenbach, B. MetaHIT Consortium, P. Bork, S. D. Ehrlich and J. Wang, Nature, 2010, 464, 59–65. 28. S. Bashiardes, G. Zilberman-­Schapira and E. Elinav, Bioinf. Biol. Insights, 2016, 2016, 19–25. 29. A. S. Mikheyev and M. M. Y. Tin, Mol. Ecol. Resour., 2014, 14, 1097–1102. 30. A. McCarthy, Chem. Biol., 2010, 17, 675–676. 31. R. C. McCoy, R. W. Taylor, T. A. Blauwkamp, J. L. Kelley, M. Kertesz, D. Pushkarev, D. A. Petrov and A.-­S. Fiston-­Lavier, PLoS One, 2014, 9, e106689. 32. S. Koren and A. M. Phillippy, Curr. Opin. Microbiol., 2015, 23, 110–120. 33. M. O. Pollard, D. Gurdasani, A. J. Mentzer, T. Porter and M. S. Sandhu, Hum. Mol. Genet., 2018, 27, R234–R241. 34. F.-­P. J. Martin, M.-­E. Dumas, Y. Wang, C. Legido-­Quigley, I. K. S. Yap, H. Tang, S. Zirah, G. M. Murphy, O. Cloarec, J. C. Lindon, N. Sprenger, L. B. Fay, S. Kochhar, P. van Bladeren, E. Holmes and J. K. Nicholson, Mol. Syst. Biol., 2007, 112. 35. S. P. Claus, T. M. Tsang, Y. Wang, O. Cloarec, E. Skordi, F.-­P. Martin, S. Rezzi, A. Ross, S. Kochhar, E. Holmes and J. K. Nicholson, Mol. Syst. Biol., 2008, 4, 219. 36. W. R. Wikoff, A. T. Anfora, J. Liu, P. G. Schultz, S. A. Lesley, E. C. Peters and G. Siuzdak, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 3698–3703. 37. M. Matsumoto, R. Kibe, T. Ooga, Y. Aiba, S. Kurihara, E. Sawaki, Y. Koga, Y. Benno, J. Qin, G. W. Tannock, P. V. Kirjavainen, T. Arvola, S. J. Salminen, E. Isolauri, J. Penders, E. E. Stobberingh, P. A. van den Brandt, C. Thijs, I. Sobhani, P. J. Turnbaugh, Y. Song, C. Liu, S. M. Finegold, H. A. Gordon, E. Bruckner-­Kardoss, B. S. Wostmann, A. Suau, S. R. Gill, K. Kurokawa, S. P. Claus, S. P. Claus, P. B. Mortensen, M. R. Clausen, M. Matsumoto, K. Kakizoe, Y. Benno, M. Matsumoto, Y. Benno, L. M. Raamsdonk, Y. Ohashi, W. R. Wikoff, M. Li, F. P. Martin, F. P. Martin, B. Vitali, T. Soga, M. R. Monton, T. Soga, W. G. Bergen, G. Wu, C. C. Metges, T. Soga, T. Mitsuoka, K. Hayakawa, J. Neu, M. Matsumoto, S. Kurihara, R. Kibe, H. Ashida, Y. Benno, A. E. Pegg, P. P. McCann, G. M. Murphy, C. Loser, A. Eisel, D. Harms, U. R. Folsch, G. D. Lux, L. J. Marton, S. B. Baylin, M. Zhang, C. H. Clarke, D. M. Shankel, T. Eisenberg, M. Matsumoto, Y. Benno, M. Matsumoto, M. Sakamoto, Y. Benno, M. Burrell, C. C. Hanfrey, E. J. Murray, N. R. Stanley-­Wall, A. J. Michael, T. Chinen, M. A. Wozny, M. P. Bryant, L. V. Holdeman, W. E. Moore, M. H. Floch, T. Pedron, P. Sansonetti, P. J. Sansonetti, R. Medzhitov, Y. Guigoz, J. Dore, E. J. Schiffrin, T. Ooga, T. Matsuki, M. Matsumoto, M. Sakamoto and Y. Benno, Sci. Rep., 2012, 2, 59–65. 38. M. Matsumoto, R. Kibe, T. Ooga, Y. Aiba, E. Sawaki, Y. Koga and Y. Benno, Front. Syst. Neurosci., 2013, 7, 9.

Introduction and Background to Microbiome Research

15

39. M. S. Donia and M. A. Fischbach, Science, 2015, 349, 1254766. 40. J.-­C. Lagier, S. Khelaifia, M. T. Alou, S. Ndongo, N. Dione, P. Hugon, A. Caputo, F. Cadoret, S. I. Traore, E. H. Seck, G. Dubourg, G. Durand, G. Mourembou, E. Guilhot, A. Togo, S. Bellali, D. Bachar, N. Cassir, F. Bittar, J. Delerce, M. Mailhe, D. Ricaboni, M. Bilen, N. P. M. Dangui Nieko, N. M. Dia Badiane, C. Valles, D. Mouelhi, K. Diop, M. Million, D. Musso, J. Abrahão, E. I. Azhar, F. Bibi, M. Yasir, A. Diallo, C. Sokhna, F. Djossou, V. Vitton, C. Robert, J. M. Rolain, B. La Scola, P.-­E. Fournier, A. Levasseur and D. Raoult, Nat. Microbiol., 2016, 1, 16203. 41. N. Singhal, M. Kumar, P. K. Kanaujia and J. S. Virdi, Front. Microbiol., 2015, 6, 791. 42. M. Kogawa, M. Hosokawa, Y. Nishikawa, K. Mori and H. Takeyama, Sci. Rep., 2018, 8, 2059. 43. N. Cassir, S. Benamar, J. B. Khalil, O. Croce, M. Saint-­Faust, A. Jacquot, M. Million, S. Azza, N. Armstrong, M. Henry, P. Jardot, C. Robert, C. Gire, J.-­ C. Lagier, E. Chabrière, E. Ghigo, H. Marchandin, C. Sartor, P. Boutte, G. Cambonie, U. Simeoni, D. Raoult and B. La Scola, Clin. Infect. Dis., 2015, 61, 1107–1115. 44. R. Caspi, H. Foerster, C. A. Fulcher, P. Kaipa, M. Krummenacker, M. Latendresse, S. Paley, S. Y. Rhee, A. G. Shearer, C. Tissier, T. C. Walk, P. Zhang and P. D. Karp, Nucleic Acids Res., 2008, 36, D623–D631. 45. D. S. Wishart, Y. D. Feunang, A. Marcu, A. C. Guo, K. Liang, R. Vázquez-­ Fresno, T. Sajed, D. Johnson, C. Li, N. Karu, Z. Sayeeda, E. Lo, N. Assempour, M. Berjanskii, S. Singhal, D. Arndt, Y. Liang, H. Badran, J. Grant, A. Serra-­Cayuela, Y. Liu, R. Mandal, V. Neveu, A. Pon, C. Knox, M. Wilson, C. Manach and A. Scalbert, Nucleic Acids Res., 2018, 46, D608–D617. 46. D. S. Wishart, Briefings Bioinf., 2007, 8, 279–293. 47. G. H. F. Nuttall and H. Thierfelder, Hoppe-­Seyler's Zeitschrift für Physiol. Chemie, 1897, 22, 62–73. 48. T. Luckey, Germfree Life and Gnotobiology, Elsevier Science, 1963. 49. M. Wang and S. M. Donovan, ILAR J., 2015, 56, 63–73. 50. M. R. Charbonneau, D. O'Donnell, L. V Blanton, S. M. Totten, J. C. C. Davis, M. J. Barratt, J. Cheng, J. Guruge, M. Talcott, J. R. Bain, M. J. Muehlbauer, O. Ilkayeva, C. Wu, T. Struckmeyer, D. Barile, C. Mangani, J. Jorgensen, Y. Fan, K. Maleta, K. G. Dewey, P. Ashorn, C. B. Newgard, C. Lebrilla, D. A. Mills and J. I. Gordon, Cell, 2016, 164, 859–871. 51. J. Sun, H. Zhong, L. Du, X. Li, Y. Ding, H. Cao, Z. Liu and L. Ge, Sci. Rep., 2018, 8, 10745. 52. E. Melancon, S. Gomez De La Torre Canny, S. Sichel, M. Kelly, T. J. Wiles, J. F. Rawls, J. S. Eisen and K. Guillemin, Methods Cell Biol., 2017, 61–100. 53. S. Wong, W. Z. Stephens, A. R. Burns, K. Stagaman, L. A. David, B. J. M. Bohannan, K. Guillemin and J. F. Rawls, mBio, 2015, 6, e00687-­15. 54. C. Kietz, V. Pollari and A. Meinander, Curr. Protoc. Toxicol., 2018, 77, e52. 55. F. Zhang, M. Berg, K. Dierking, M.-­A. Félix, M. Shapira, B. S. Samuel and H. Schulenburg, Front. Microbiol., 2017, 8, 485.

16

Chapter 1

56. M. Guzman-­Rodriguez, J. A. K. McDonald, R. Hyde, E. Allen-­Vercoe, E. C. Claud, P. M. Sheth and E. O. Petrof, Methods, 2018, 149, 31–41. 57. L. Liu, J. Firrman, C. Tanes, K. Bittinger, A. Thomas-­Gahring, G. D. Wu, P. Van den Abbeele and P. M. Tomasula, PLoS One, 2018, 13, e0197692. 58. S. Bartfeld and H. Clevers, J. Mol. Med., 2017, 729–738. 59. N. Sachs, Y. Tsukamoto, P. Kujala, P. J. Peters and H. Clevers, Development, 2017, 144, 1107–1112. 60. H. Yu, N. M. Hasan, J. G. In, M. K. Estes, O. Kovbasnjuk, N. C. Zachos and M. Donowitz, Annu. Rev. Physiol., 2017, 79, 291–312. 61. A. Bein, W. Shin, S. Jalili-­Firoozinezhad, M. H. Park, A. Sontheimer-­ Phelps, A. Tovaglieri, A. Chalkiadaki, H. J. Kim and D. E. Ingber, Cell. Mol. Gastroenterol. Hepatol., 2018, 5, 659–668. 62. T. Yatsunenko, F. E. Rey, M. J. Manary, I. Trehan, M. G. Dominguez-­Bello, M. Contreras, M. Magris, G. Hidalgo, R. N. Baldassano, A. P. Anokhin, A. C. Heath, B. Warner, J. Reeder, J. Kuczynski, J. G. Caporaso, C. A. Lozupone, C. Lauber, J. C. Clemente, D. Knights, R. Knight and J. I. Gordon, Nature, 2012, 486, 222–227. 63. P. J. Turnbaugh, M. Hamady, T. Yatsunenko, B. L. Cantarel, A. Duncan, R. E. Ley, M. L. Sogin, W. J. Jones, B. A. Roe, J. P. Affourtit, M. Egholm, B. Henrissat, A. C. Heath, R. Knight and J. I. Gordon, Nature, 2009, 457, 480–484. 64. P. Lepage, R. Häsler, M. E. Spehlmann, A. Rehman, A. Zvirbliene, A. Begun, S. Ott, L. Kupcinskas, J. Doré, A. Raedler and S. Schreiber, Gastroenterology, 2011, 141, 227–236. 65. J. K. Goodrich, E. R. Davenport, M. Beaumont, M. A. Jackson, R. Knight, C. Ober, T. D. Spector, J. T. Bell, A. G. Clark and R. E. Ley, Cell Host Microbe, 2016, 19, 731–743. 66. V. K. Ridaura, J. J. Faith, F. E. Rey, J. Cheng, A. E. Duncan, A. L. Kau, N. W. Griffin, V. Lombard, B. Henrissat, J. R. Bain, M. J. Muehlbauer, O. Ilkayeva, C. F. Semenkovich, K. Funai, D. K. Hayashi, B. J. Lyle, M. C. Martini, L. K. Ursell, J. C. Clemente, W. Van Treuren, W. A. Walters, R. Knight, C. B. Newgard, A. C. Heath and J. I. Gordon, Science, 2013, 341, 1241214. 67. The Integrative HMP (iHMP) Research Network Consortium, The Integrative Human Microbiome Project, Nature, 2019, 569, 641–648. 68. P. Rausch, M. Basic, A. Batra, S. C. Bischoff, M. Blaut, T. Clavel, J. Gläsner, S. Gopalakrishnan, G. A. Grassl, C. Günther, M. Hirose, S. Ibrahim, G. Loh, J. Mattner, S. Nagel, O. Pabst, F. Schmidt, B. Siegmund, T. Strowig, V. Volynets, S. Wirtz, S. Zeissig, Y. Zeissig, A. Bleich and J. F. Baines, Int. J. Med. Microbiol., 2016, 306, 343–355. 69. K. D. Parker, S. E. Albeke, J. P. Gigley, A. M. Goldstein and N. L. Ward, Front. Microbiol., 2018, 9, 1598. 70. I. I. Ivanov, K. Atarashi, N. Manel, E. L. Brodie, T. Shima, U. Karaoz, D. Wei, K. C. Goldfarb, C. A. Santee, S. V. Lynch, T. Tanoue, A. Imaoka, K. Itoh, K. Takeda, Y. Umesaki, K. Honda and D. R. Littman, Cell, 2009, 139, 485–498.

Introduction and Background to Microbiome Research

17

71. H. E. Jakobsson, A. M. Rodríguez-­Piñeiro, A. Schütte, A. Ermund, P. Boysen, M. Bemark, F. Sommer, F. Bäckhed, G. C. Hansson and M. E. V. Johansson, EMBO Rep., 2015, 16, 164–177. 72. R. N. N. Carmody, G. K. K. Gerber, J. M. M. Luevano, D. M. M. Gatti, L. Somes, K. L. L. Svenson and P. J. J. Turnbaugh, Cell Host Microbe, 2014, 17, 72–84. 73. E. B. Chang, J. F. Pierre, R. Hinterleitner, R. Bouziat, N. Hubert, V. Leone, J. Miyoshi and B. Jabri, Data Brief, 2018, 387–393. 74. L. A. David, C. F. Maurice, R. N. Carmody, D. B. Gootenberg, J. E. Button, B. E. Wolfe, A. V. Ling, A. S. Devlin, Y. Varma, M. A. Fischbach, S. B. Biddinger, R. J. Dutton and P. J. Turnbaugh, Nature, 2014, 505, 559–563.

Chapter 2

Metabolism of Dietary Carbohydrates by Intestinal Bacteria EBRU Ece Gulsan†a, Farrhin Nowshad†b, Arul Jayaraman*b and Kyongbum Lee*a a

Department of Chemical and Biological Engineering, Tufts University, Medford, MA 02155, USA; bArtie McFerrin Department of Chemical Engineering, Texas A&M University, College Station, Texas 77843, USA *E-­mail: [email protected], [email protected]

2.1  Introduction Dietary carbohydrates comprise a diverse class of molecules, ranging from simple sugars to high molecular weight polysaccharides composed of complex monosaccharide chains bound together by glycosidic linkages. Carbohydrates are significant components of both plant-­ and animal-­based human diets. Depending on the type of diet, calories from carbohydrates can account for more than 70% of the total daily energy intake of human adults.1 Complex carbohydrates are found in whole grains, cereals, starchy and non-­starchy vegetables, beans, and legumes, whereas simpler sugars are enriched in raw sugar, fruits, dairy products, and some grains.2 Complex carbohydrates are only partially digested in the stomach and are not fully

†

These authors contributed equally to the work.

  Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

18

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

19

absorbed in the small intestine. Consequently, bacteria residing in the colon have greater access to complex carbohydrates, and microbial metabolism of these dietary components in the colon is increasingly viewed as an important factor in human health and disease. The first step in polysaccharide metabolism in the body is the breakdown of glycosidic bonds to form simple sugars and oligosaccharides. This step occurs in the mouth and the small intestine, where reactions are catalyzed primarily by host enzymes. The breakdown of glycosidic bonds can also occur in the colon, where the reactions are catalyzed by bacterial enzymes. In humans, the dominant digestive enzymes in the mouth and small intestine are glycoside hydrolases (GHs). Seventeen different GHs, including α-­amylase, maltase, sucrase, and lactase, have been identified in the human mouth, stomach, and small intestine. Together, these enzymes can hydrolyze a broad set of substrates. However, many dietary carbohydrates, especially cellulose based complex carbohydrates with high fiber content, cannot be fully degraded by human enzymes. These carbohydrate molecules are quantitatively available to gut bacteria, and have been designated as microbiota-­ accessible carbohydrates (MACs). MACs are broken down by bacterial GHs as well as by polysaccharide lyases (PLs), which use a β-­elimination mechanism rather than hydrolysis to break glycosidic bonds.3 These enzymes act on uronic acid (e.g. glucuronic acid) containing polysaccharides, and generate oligosaccharides having higher degrees of unsaturation.4 Along with simpler sugars formed in the upper digestive tract by GH activity, MACs are fermented in the colon.3 Figure 2.1 shows an overview of various steps in the digestion, transformation, and metabolism of MACs. Fermentation is necessary for absorption of polysaccharide derived products by the host as well as their utilization by colonic bacteria. Major primary degraders of MACs in the colon include Bifidobacterium spp., Bacteroides spp., and Ruminococcus bromii. Metabolism of MACs by these bacteria results in the production of simpler sugars and small organic acids such as propionic, formic, and acetic acid (Figure 2.2).5,6 The sugar monomers are further metabolized by secondary degraders, primarily Firmicutes spp., to generate additional organic acids such as butyric acid.7 Acetate, propionate, and butyrate are collectively known as short-­chain fatty acids (SCFAs) and play essential roles in mediating the metabolic interactions between microbiota and the host. Immune cell signaling, intestinal epithelial barrier integrity, and regulation of whole-­body energy metabolism are well-­known examples of physiological processes that are directly affected by SCFAs.8 In addition to modulating physiological processes by directly engaging host cellular pathways, SCFAs are also used as substrates by colonic bacteria, and thereby indirectly affect host physiology.9 This chapter reviews the metabolism of MACs by intestinal bacteria. We focus on the enzymatic pathways of carbohydrate degradation and fermentation, the bacteria that possess these pathways, and the interaction of fermentation products with host physiological processes. Intestinal and liver pathways are discussed in the context of co-­metabolism, where complete

20

Chapter 2

Figure 2.1  Overview  of dietary carbohydrate metabolism in the intestine. Diges-

tion of dietary carbohydrates begins with hydrolysis in the mouth, followed by further degradation by host enzymes in the small intestine. Digestion-­resistant polysaccharides and their hydrolysis products pass to the colon where they are metabolized by commensal bacteria. In addition to breaking glycosidic bonds, biochemical reactions carried out by colonic bacteria include decarboxylation, deglycosylation, ring fission, and demethylation. The carbohydrate monomers and other reaction products can be fermented into CO2 and small organic acids, notably short chain fatty acids (SCFA). The organic acids are quantitatively absorbed by colonocytes and utilized by the intestinal epithelial cells as substrates for producing energy. The SCFAs that are not oxidized by the colonocytes can reach the liver via portal circulation. In the liver, SCFAs and other fermentation products are utilized by hepatocytes for energy production, as well as de novo synthesis of glucose, cholesterol, and fatty acids. In addition to providing metabolic substrates, gut bacterial fermentation products also generate ligands that engage host cellular receptors. MCT1, monocarboxylate transporter 1; SMCT1, sodium-­dependent monocarboxylate transporter 1.

breakdown and utilization of dietary carbohydrate spans the host and gut microbiota. We begin with an overview of the major classes of carbohydrates found in dietary fiber (DF), with particular emphasis on polysaccharides and oligosaccharides that are resistant to mammalian hydrolysis enzymes. We then highlight key pathways and products of polyphenol and amino sugar metabolism. We conclude the chapter by discussing the outstanding gaps in knowledge as well as methodological advances that are needed to address these gaps.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

21

Figure 2.2  Metabolism  of microbiota accessible carbohydrates (MACs) by colonic

bacteria. Dietary carbohydrates resistant to hydrolysis in the mouth and small intestine reach the colon, where primary bacterial degraders such as Bifidobacterium spp., Bacteroides spp., and R. bromii metabolize them to propionic acid, formic acid, acetic acid, or simple sugar monomers. The sugars can be further fermented to butyric acid by secondary bacterial degraders.

2.2  Dietary Fiber Historically, dietary fiber (DF) referred to plant cell wall components, such as cellulose, hemicellulose, and lignin.10 Later, this definition was expanded to include other plant oligosaccharides and polysaccharides that are resistant to digestion and absorption in the intestine,11 such as inulin and resistant starches.12 In 2009, the Codex Alimentarius Commission finalized a regulatory definition of DF as carbohydrate polymers that have ten or more monomeric units, and are not hydrolyzed by the endogenous enzymes in the small

Chapter 2

22 13,14

intestine of humans. Broadly, DF is divided into three categories: non-­ starch polysaccharides, lignin, and analogous carbohydrates. The largest category, polysaccharides, is further subdivided into cellulose, hemicellulose, pectin, and oligosaccharides. These DF components are partially or completely fermented depending on the type of polysaccharide and composition of the colonic microbiota. The main products are organic acids and gases (hydrogen, methane, and carbon dioxide).10 The organic acids are absorbed and metabolized by colonic epithelial cells, or enter the liver via portal circulation.15 Not all DF polysaccharides reaching the colon are fully fermented, with approximately half of DF leaving the body as waste.15,16 The gases resulting from fermentation are either absorbed into the bloodstream or pass out of body as flatus.16 The following sections describe the major DF polysaccharides, their general structures, food sources, fermentation pathways, and metabolic products.

2.2.1  Cellulose Examples of foods high in cellulose content include bran, legumes, nuts, peas, root vegetables, vegetables of the cabbage family, the outer covering of seeds, and apples. Cellulose is highly crystalline, and composed of linear chains of β-­1,4-­d-­glucopyranose units. The chains are packed in layers and held together by van der Waals forces and intra-­ and inter-­chain hydrogen bonds.17 The intra-­chain hydrogen bonding stabilizes the β-­1,4 glycosidic bonds and results in the cellulose chains' linear configuration.18,19 The main products of cellulose fermentation are SCFAs.19,20 Gut bacteria express endocellulases and exocellulases21 (e.g. β-­1,4-­endoglucanase and exoglucanase), respectively, to expose chain ends and cleave disaccharides and tetrasaccharides from the exposed chains. The conventional view is that aerobes secrete free cellulases, whereas anaerobes arrange these enzymes onto scaffolding proteins of cellulosomes via dockerin binding.22 This view does not capture the full diversity of bacterial cellulase systems, as examples of free-­cellulase expressing anaerobes have been found in soil and intestines of ruminants.23,24 Hydrolysis of the small oligosaccharides produced by exocellulases is catalyzed by β-­glucosidases, which results in the formation of d-­glucose. The dominant cellulase expressing gut bacteria are Bacteroides succinogenes, Ruminococcus. albus, and Ruminococcus flavefaciens.18,25 Cellulase expressing strains of Bacteriodetes have not yet been characterized in human isolates. Several other colonic bacteria, including Clostridium spp., Actinomycetes spp., Butyrivibrio fibrisolvens, and Methanobrevibacter ruminantium show cellulose degrading capability.25

2.2.2  Hemicellulose Foods that are relatively high in hemicellulose content include bran and whole grains as well as nuts, legumes, and some vegetables and fruits. Hemicellulose molecules are smaller than cellulose and typically contain 50–200

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

23

20

monomer units. Unlike cellulose, the polymer backbones of hemicelluloses comprise several different sugars, including xylose, mannose, glucose, and galactose; furthermore, these sugars are often acetylated (Figure 2.3). The side chains are also heterogeneous, and consist of various sugars and sugar derived acids, including arabinose, fucose, and glucuronic acid. The composition, length, and arrangement of side chains are highly variable; while some hemicelluloses are relatively linear, others are highly branched.19 The composition of sugar monomers in the side chains plays an important role in determining the physicochemical properties of hemicelluloses, which, in turn, affects the metabolism of these molecules. For example, side chains of glucuronic acids can add charge and increase solubility, facilitating access to the side chains' sugars and acids for bond cleavage by bacterial enzymes.19 Intestinal bacteria responsible for degradation and fermentation of hemicellulose include strains of B. succinogenes, R. albus, and R. flavefaciens. While these strains can hydrolyze hemicelluloses by expressing hemicellulases (mannanase and galactosidase), they do not appear to effectively utilize the hydrolysis products as substrates for growth.26,27 Table 2.1 summarizes representative hemicelluloses of edible plants, their common food sources, and gut bacteria that have shown the capacity to degrade and metabolize the polysaccharides.

Figure 2.3  Polymer  backbones of representative hemicelluloses in foods. (A) Xylans

have a backbone of β-­1,4-­linked xylose (Xyl) units, with side chains of α-­glucuronic acids (GlcA) and/or α-­arabinofuranose. The glucuronic acid units are often methylated (MeGlcA). (B) Mannans have a backbone of β-­1,4-­linked mannose and glucose units with side chains of galactose (Gal) or glucose (Glc). (C) Xyloglucans have a backbone of β-­1,4-­linked glucose units, with side chains of α-­1,6-­linked xylose, β-­1,2-­linked galactose, and/or α-­1,2-­linked fucose (Fco). (D) Chemical structures of common side chain monomers.

Chapter 2

24

Table 2.1  Hemicelluloses  and gut bacteria expressing degradation enzymes. Hemicellulose

Monomera Foods

Enzyme

Bacteria R. albus

Xylan (glucuXylose ronoxylan, arabinoglucoronoxylan, arabinoxylan)

Sugar cane, corn, bran, legumes

α-­Arabinofura­ nosidase, xylanase β-­Xylosidase103

Mannan (galac- Mannose tomannan, Galactose glucomannan, Glucose galactoglucomannan)

Aloe vera, green coffee beans, ivory nuts, whole grains, legumes Pepper, potato, tomato, basil, plantain

β-­Mannase β-­Glucosidase β-­Mannosidase27

Xyloglucan

Xylose Glucose

Bifidobacterium adolescentis Bacteroides ovatus Cytophaga xylanolytica103 R. albus R. flavefaciens Bacteroides succinogenes Bacillus subtilis27

Endoglucanase104 Clostridium thermocellum104 endo-­Xylanase Bacillus circulans exo-­Xylanase Bacillus polymyxa103

a

Monomer refers to the backbone sugar residues.

2.2.3  Pectin Pectin is a high-­molecular weight carbohydrate polymer present in almost all plants. It forms part of the primary cell wall as well as the middle lamella joining two cells walls.28 Rich sources of pectin include acidic fruits like apples, strawberries, and citrus fruits. Legumes, nuts, and some vegetables also are sources of pectin. There is a large diversity of pectins, which vary in size, chemical structure, and content of neutral sugars.29 The polysaccharides in pectin can be broadly classified into three major groups: homogalacturonan, rhamnogalacturonan-­I, and rhamnogalacturonan-­II. Depending on the plant source, other polymeric components of pectins include arabinan, galactan, arabinogalactan, xylogalacturonan, and apiogalacturonan.30 The polymer backbone of pectins is usually an unbranched chain of α-­1-­4-­d-­ galacturonic acid units. Other sugars found in pectin are galacturonic acid chains, including rhamnose, arabinose, xylose, fucose, and galactose.28 Like cellulose and hemicellulose, pectin can be broken down and fermented by colonic microbiota.28 Primary pectin-­degraders in the intestine are Bacteroides spp., Bifidobacterium spp., and Prevotella spp., which possess carbohydrate-­active lyases, methylesterases, and acetylesterases that facilitate the breakdown of pectins.31,32 Acetate and butyrate are, respectively, the most and least abundant products of intestinal pectin fermentation.33 In vitro studies revealed that pectins stimulate the growth of commensal bacteria associated with health-­promoting effects of DF, including Bifidobacteria, Lactobacilli, Faecalibacterium prausnitzii, Roseburia spp. and Eubacterium rectale.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

25

Table 2.2  Pectic  polysaccharides and gut bacteria expressing degradation enzymes. Polysaccharide Monomera

Foods

Homogalacturonan

d-­Galacturonic Apples, pears, acid guavas, plums, sunflower seeds Rhamnogalac- d-­Galacturonic Orange peel, turonan-­I acid potatoes, sugar beet, soybeans l-­Rhamnose

Enzyme

Bacteria

Pectin lyase endo-­Polygalac­ turonase II

Lactobacillus spp. Bacteroides thetaiotaomicron Escherichia coli105 Bacteroides spp.

Rhamnogalac­ turonan-­ hydrolase106 Rhamnogalac­ turonan lyase107 d-­Galactose Rhamnogalac­ Rhamnogalac- d-­Galacturonic Tomatoes, turonan-­ turonan-­II acid sweet hydrolase106 potatoes, squashes, l-­Rhamnose Rhamnogalac­ green turonan beans, peas lyase107 d-­Galactose l-­Arabinose

Bacillus licheniformis Bacillus subtilis106 Bacteroides spp. Bacillus licheniformis Bacillus subtilis106

a

Monomer refers to the backbone sugar residues.

Other bacterial taxa that increase in abundance upon exposure to pectin or pectin-­containing foods include Bacteroides spp., Prevotella spp., E. rectale/ Clostridium coccoides group and Clostridium spp.31 Table 2.2 summarizes representative pectic polysaccharides, their common food sources, and gut bacteria that have shown the capacity to degrade and metabolize the polysaccharides.

2.2.4  Oligosaccharides Oligosaccharides are composed of two to ten monosaccharide monomers and are classified on the basis of the number and type of constituent monosaccharides. Figure 2.4 shows representative oligosaccharides that are abundant in human diets. Similar to the larger polysaccharides, the metabolism of oligosaccharides depends on the type of monomers and glycosidic bonds linking the monomers. Lactobacillus is a major genus contributing to the fermentation of oligosaccharides in the colon. Metabolism of fructo-­oligosaccharides (FOS) in these gut bacteria occurs in three steps. After an initial extracellular hydrolysis step, the degradation products are taken up by various sugar transporters. The monomers are directly phosphorylated and catabolized, whereas larger degradation products, such as trisaccharides and tetrasaccharides, are further hydrolyzed intracellularly prior to phosphorylation and catabolism. Several Lactobacillus spp. express extracellular fructansucrases, which are bifunctional enzymes that catalyze both hydrolysis and glycosyl transfer reactions. These enzymes couple fructan metabolism with formation of extracellular polysaccharides, or

26

Chapter 2

Figure 2.4  Representative  oligosaccharides in foods. (A) Fructans (e.g. inulin)

consist of a variable number fructosyl units that terminate with a glucose monomer. (B) α-­Galacto-­oligosaccharides (α-­GOS) consist of α-­1,6-­ linked galactosyl (Gal) units. The galactosyl units may be interspersed with other α-­1,6-­linked sugars, including glucose (Glc), fructose (Frc), sucrose, raffinose and/or stachyose. (C) β-­Galacto-­oligosaccharides (β-­GOS) consist of β-­1,4-­linked galactosyl units interspersed with other β-­1,4-­linked sugars, including glucose and lactose. (D) Resistant starch consists of α-­1,4-­linked glucose units.

exopolysaccharides, which are structural components of intestinal biofilms associated with Lactobacillus spp. The rate-­limiting step in the fermentation of FOS-­derived sugars (mostly fructose containing disaccharides) by Lactobacillus spp. is cellular uptake. Transport of the sugars into cells is directly coupled to phosphorylation by a (putative) sucrose-­specific phosphotransferase system (pts1BCA) (Figure 2.5). In the case of trisaccharides and tetrasaccharides, further hydrolysis of the imported sugars is catalyzed by a phospho-­fructo-­furanosidase [encoded by several different genes, depending on the organism, including sacA, sucrose utilization gene B (scrB), and scrP]. Other microorganisms utilizing FOS as a substrate include Actinobacteria, Bacteriodetes, Firmicutes, Klebsiella, and Clostridium species.34–36 Galacto-­oligosaccharides (GOS) are also resistant to hydrolysis by salivary and intestinal digestive enzymes, and are not absorbed in the upper gastrointestinal tract. In the colon, they are utilized as substrates by Lactobacilli, Actinobacteria, Bacteriodetes, and Firmicutes species. Lactose, the most abundant oligosaccharide in breast milk, has the same β-­1,4 glycosidic linkage as GOS.37 The β-­galactosidases expressed by colonic bacteria can hydrolyze lactose as well as GOS. These β-­galactosidases also catalyze the transglycosylation of lactose to allolactose, which has a β-­1,6 glycosidic linkage. Allolactose can induce the lac operon, which is required for lactose transport and metabolism. β-­Galactosidases are active in several

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

27

Figure 2.5  Fructo-­  oligosaccharide (FOS) metabolism in lactic acid bacteria.

(1) Commensal Lactobacillus species can transport a FOS into the cytosol using a four-­unit [multiple sugar metabolism (Msm) F, MsmG and two MsmK] ATP binding cassette (ABC) transporter complex. In the cytosol, FOS is further hydrolyzed to glucose and fructose by strain-­ specific hydrolases [beta-­fructosidase A (BfrA) and sucrase A (SacA)]. (2) In L. plantarum, FOS-­derived sucrose is taken up and phosphorylated by a sucrose specific phosphotransferase system (pts1BCA). In the cytosol, a strain-­specific phospho-­hydrolase [sucrose phosphorylase (ScrP)] degrades the phosphorylated sucrose into glucose and fructose-­6-­phosphate.

Lactobacillus and Bifidobacterium species. Lactobacilli use two pathways to metabolize GOS (Figure 2.6). In the more common pathway, GOS is first transported into the cell by lactose permease (lacS), and then hydrolyzed by β-­galactosidase. In the second pathway, GOS is transported into the cell and phosphorylated by phosphotransferases (lacE and lacF), and then hydrolyzed by β-­phospho-­galactosidase (lacG). Bifidobacterium species use a more specialized pathway for GOS metabolism. These species express hydrolytic enzymes (endo-­α-­N-­acetylogalactosaminidase and lacto-­N-­biosidase) that extracellularly digest β-­GOS into amine-­containing disaccharides. The hydrolysis products (galacto-­N-­biose and lacto-­N-­biose) are taken up into the cell and cleaved by lacto-­N-­biose phosphorylase (lnpA) to form galactose 1-­phosphate and N-­acetylgalactosamine, which then enter glycolysis and amino sugar metabolism pathways.38–40

28

Chapter 2

Figure 2.6  Galacto-­  oligosaccharide (GOS) metabolism in lactic acid bacteria.

(A) Uptake of a GOS is facilitated by lactose permease (lacS). In the cytosol, the GOS is further degraded to glucose and galactose by strain-­ specific hydrolases. (B) In some species (e.g., L. casei), lactose derived from GOS lactose is transported and phosphorylated by lactose-­specific enzyme IIB (lacE) and/or IIA (lacF). The phosphorylate disaccharide is then degraded by a strain-­specific phospho-­hydrolase to glucose and galactose-­6-­phosphate.

The conventional view is that metabolism of oligosaccharides by gut microbiota generally has a beneficial effect on gut health. For example, fermentation of inulin by Bifidobacterium spp. and Lactobacilli results in the production acetate and butyrate. These SCFAs promote the growth of C. coccoides and E. rectale, which in turn increases the abundance of butyrate-­ producing bacteria. The fermentation of GOS generates other small organic acids such as succinate and lactate. These molecules are used as substrates by Bifidobacterium spp. and Faecalibacterium prausnitzii, two bacterial groups associated with enhanced epithelial barrier integrity and amelioration of irritable bowel syndrome and Crohn's disease symptoms.41 The organic acids from GOS fermentation locally decrease the pH by neutralizing basic compounds and acidifying the colonic content. A more acidic colonic luminal

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

29

environment promotes increased microbial metabolic activity, which, in turn, facilitates absorption of nutrients and stimulates proliferation of mucosal cells. Some species of Bifidobacterium ferment GOS into long chain fatty acids, notably β-­palmitate. Consuming milk formula enriched in this fatty acid has been shown to increase the abundance of Bifidobacterium spp. in human infants.42 Not all oligosaccharides consumed by humans are derived from DF, and another major source of oligosaccharides is milk. Exposure to human milk oligosaccharides (HMOs) has been shown to play a crucial role in increasing the diversity of gut microbiota during the early stages of postnatal development. Like other dietary oligosaccharides, HMOs lead to increased abundance of Bifidobacterium spp., and can be fermented to produce SCFAs. Different HMOs bind with different affinities to various epithelial glycan receptors, resulting in the modulation of different functions. For example, the dendritic cell-­specific intercellular adhesion molecule-­3 grabbing nonintegrin (DC-­SIGN) receptor, a C-­t ype lectin receptor on macrophage and dendritic cell surfaces, specifically recognizes α-­linked fucose HMOs, although it is unclear if this directly affects downstream gene expression.43 Selectins [cluster of differentiation 62 (CD62)] are a family of cell adhesion molecules that participate in the immune response by mediating cell-­to-­cell interactions, and are activated by glycan ligands with siali Lewis blood group epitopes. Sialyl-­Lewis X (sLeX) is a well-­known tetrasaccharide selectin ligand that is attached to O-­glycans on cell surfaces. The results of an in vitro study performed on model membranes displaying P-­selectin indicated that HMOs has a similar effect to sLeX on P-­selectins.44,45 Rather than completely blocking the binding site, the HMOs alter the affinity of other ligands. In addition to selectins, other adhesion molecules such as galectins, and siglecs also show selective binding of HMOs,46 see Table 2.3. The beneficial effects of HMOs depend on the intestinal microbiota composition. Preterm infants have lower abundances of anaerobes, such as Bifidobacterium, Bacteroides, and Atopobium, while Proteobacteria are more abundant in the infants' intestines.47,48 The immature microbiota of these infants hinders digestion of HMOs and can lead to intestinal inflammation. Another variable is the type of HMO. Structural characterization studies on human milk using nuclear magnetic resonance (NMR) spectroscopy showed that milk from mothers that actively secrete fucose HMOs was more strongly associated with prebiotic effects.49 Although oligosaccharides are available to the gut microbiota in amounts that are similar to the amounts of monosaccharides and disaccharides, less is known about oligosaccharide metabolism and fermentation. A major challenge is the scarcity of high-­purity oligosaccharides that can be used for characterization studies and biological experimentation. While computational approaches could help address some gaps in knowledge, detailed in silico metabolic models describing specific transport systems, hydrolases, and enzymatic pathways for oligosaccharide metabolism in gut bacterial species are also lacking.

Chapter 2

30

Table 2.3  Oligosaccharides  and gut bacteria expressing degradation enzymes. Oligosac­ charide Fructan

α-­1,6 GOS

β-­1,4 GOS

RS

Monomera

Foods

Enzyme

Fructose– glucose

Bananas, onions, Fructansucrase chicory root, garlic, asparagus, jicama, and leeks Galactosyl-­ Soybean seeds, α-­Galactosidase sucrose sugar beet, raffinose/ human milk stachyose Galactosyl-­ Human milk, soy- β-­Galactosidase lactose bean seeds

Glucose

Plantains, green bananas, beans, peas, lentils

Amylase, GH

Bacteria Lactobacillus spp., Klebsiella spp., Clostridium spp., Bacteroides spp. Bifidobacterium spp., Bacillus coagulans108 Lactobacillus, Bifidobacterium, Actinobacteria, Bacteriodetes, Firmicutes Bifidobacterium spp., Bacteroides spp., Fusobacterium spp., Eubacterium, Clostridium, Streptococcus, Propionibacterium51

Monomer refers to the backbone sugar residues. α-­1,6 GOS, galacto-­oligosaccharide with a polymer backbone of α-­1,6-­linked galactose units. β-­1,4 GO, GOS with a polymer backbone of β-­1,4-­linked galactose units. RS, resistant starch with a polymer backbone of α-­1,4-­linked d-­glucose units. GH, glycoside hydrolase.

a

2.2.5  Resistant Starch Some starches are resistant to digestion in the small intestine, and thus are not absorbed. Foods that are rich in resistant starch (RS) include plantains, green bananas, beans, peas, lentils, and grains such as oats, barley, and rice. Resistant starches behave like soluble fiber due to their water solubility and indigestibility.50 These starches are categorized into several classes, depending on whether resistance to digestive enzymes is due to their native physical structure that blocks access to enzymes or modifications resulting from food processing.20 The fermentation products of RS by gut bacteria are similar to other DF components, and include gases (methane, hydrogen, carbon dioxide) and SCFAs. Smaller amounts of other organic acids (mostly lactic, succinic, and formic acids), branched SCFAs (isobutyrate and isovalerate), and alcohols (methanol and ethanol) are also produced. Starch degradation in the colon is a cooperative process involving multiple species, including Bifidobacterium spp., Bacteroides spp., Fusobacterium spp., and strains from the genera Eubacterium, Clostridium, Streptococcus, and Propionibacterium.51 The first step is degradation of starch polymers into glucose. This is followed by uptake and catabolism of glucose via glycolysis, which forms SCFAs and

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

31

other organic acids. The glucose monomers are also available to other methane and hydrogen producing bacteria as well as colonic Archaea spp.51

2.2.6  Lignin Lignin is found in the stems and seeds of fruits and vegetables and in the bran layer of cereals, especially wheat, mature root vegetables, such as carrots, and fruits with edible seeds, such as strawberries. While lignin is not a saccharide, it is closely associated with other plant cell wall components and affects the physiological effects of DF.20 Lignin is a highly branched polymer of phenolic compounds, which affords strong intramolecular bonding. The main building blocks are trans-­coniferyl, trans-­sinapyl, and trans-­p-­coumaryl alcohols.19 Lignin is insoluble in water, due to its high hydrophobicity, consumed in smaller amounts, and fermented to a lesser degree by colonic bacteria compared with other DF components. The fermenters include Proteobacteria, Actinobacteria, and Firmicutes that express laccase, a ligninolytic enzyme.52 Major metabolic products are SCFAs and lignan enterolactone (ENL), a weak phytoestrogen.19 The latter product is more commonly associated with metabolism of lignans by gut bacteria, but the results of studies in rodents indicate that lignin may be another source of ENL and other mammalian lignans53 (Figure 2.7).

2.3  Polyphenols Polyphenols are a large family of phytochemicals characterized by aromatic rings bearing hydroxyl moieties.54 They are abundant in plant-­based foods as well as wine, extra virgin olive oil, cereals, dry legumes, coffee, and tea. Polyphenols have been associated with a variety of health promoting effects, including defense against ultraviolet radiation, and aging, and investigated as potential treatments for metabolic, cardiovascular and neurological diseases, as well as some types of cancer.55,56 Polyphenol classes are identified by the number of phenol rings and the functional groups attached to the rings. Flavonoids are the largest class of polyphenols in terms of number of structurally distinct molecules occurring in nature, followed by phenolic acids, stilbenes, and lignans. The basic structure of flavonoids consists of a chromone ring with a phenol ring substituted at the C2 or C3 position (Figure 2.8). Hydroxyl groups can be at any position of the basic structure and allow for substitutions of methyl or glycosyl groups. Thus, there is a wide variety of possible flavonoids, with over 6000 unique compounds that are currently known. It is becoming increasingly clear that the health benefits of polyphenols are concentration and context dependent. Dietary flavonoids are usually present as glycosides, which are poorly absorbed in the small intestine. Bioavailability studies in rats on genistein and its glucoside form, genistin, revealed

32

Chapter 2

Figure 2.7  Putative  route of lignin fermentation by gut bacteria. Lignin is first

degraded into lignan glycosides, e.g. secoisolariciresinol diglucoside. Following removal of the O-­linked sugar, the diet-­derived lignan undergoes a series of transformation reactions that results in the formation of enterodiol. This intestinal lignan can be further transformed to enterolactone.

that the former aglycone form exhibited a higher bioavailability, with high amounts of genistein and its derivatives recovered from intestinal luminal contents and feces.57 One reason for poor bioavailability is phase II metabolism (mainly sulfation, methylation, and glucuronidation) in the small intestine and the liver,58,59 which facilitates excretion of flavonoids through bile and urine. In addition to phase II metabolism, the position and structure of the attached sugar can alter polyphenol absorption and bioavailability.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

33

Figure 2.8  Major  subclasses of flavonoids. Flavonoids represent the largest class

of polyphenols. A flavonoid has the basic structure of a chromane (benzodihydropyran) with a phenyl ring attached at the C-­2 or C-­3 position. Flavanones have a keto group at the C-­4 position of this basic structure. Compared with flavanones, flavones have an additional degree of unsaturation between C-­2 and C-­3. Isoflavones are isomers of flavones with the phenyl ring attached to C-­3 rather than C-­2. Flavonols are alcohols of flavones, with a hydroxyl group at C-­3. Flavan-­3-­ols also have a hydroxyl group at C-­3, but lack a keto group at C-­4. Anthocyanins are similar to flavan-­3-­ols, but carry a net positive charge on the pyran oxygen (O-­1) due to an additional degree of unsaturation.

A clinical study comparing the bioavailability of two O-­linked quercetin glycosides revealed that the concentration of orally administered quercetin-­4′-­ O-­glucoside reached a tenfold higher maximal plasma concentration than that of quercetin-­3-­O-­rutinoside in one-­tenth of the time it took the latter to reach maximal concentration.60 Polyphenol metabolism proceeds through biotransformation reactions that occur in the small intestine and liver. The small intestine harbors bacteria that can hydrolyze flavonoids into aglycones, which facilitates absorption and further metabolism in the liver. Unabsorbed polyphenols pass to the colon where they can be converted to aglycones by bacterial glycosidases (e.g. α-­rhamnosidase). The aglycones formed in the colon can be either absorbed into circulation and reach the liver or utilized by the colonic microbiota.61

34

Chapter 2

Bacterial metabolism of aglycones in the colon usually involves an initial phenolic ring fission step. Ring fission typically occurs at the C–C bond between the 1-­and 2-­positions or the 4-­and 5-­positions. Depending on their chemical structure, the ring-­opened aglycons are further metabolized via decarboxylation and dihydroxylation reactions. Figure 2.9 shows a schematic overview of polyphenol metabolism in the gastrointestinal tract. The route of bacterial polyphenol metabolism depends on the polyphenol's scaffold, and thus varies across different classes. Anthocyanins are a class of flavonoids that are conjugated to five-­carbon sugars, such as arabinose or xylose, and resistant to absorption or breakdown by human enzymes. Consequently, anthocyanins are highly abundant in their conjugated form in the colon and available for biotransformation and metabolism by the colonic bacteria. Metabolic products of anthocyanins include protocatechuic acid and phloroglucinaldehyde, which are quantitatively re-­conjugated in the colon.62 Another factor that affects polyphenol metabolism is the diversity of intestinal microbiota. Common biotransformation reactions such as deglycosylation and hydrolysis can be carried out by many species.63,64 However, some

Figure 2.9  Overview  of polyphenol metabolism in the host intestine. Polyphenols

such as flavonoids are ingested in their glycoside (mainly glucoside) forms. In the small intestine, the glucosides are hydrolyzed by host and bacterial enzymes into their aglycon (or genin) forms. These aglycons can either be absorbed and trafficked to the liver or passed to the colon. Colon bacteria transform the aglycons through a variety of reactions, including glucuronidation, sulfation, methylation, oxidation, and ring fission. These reactions depend on the base chemical structure of the aglycon and its functional groups.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

35

transformations such as ring cleavage and enzyme specific reduction reactions are limited to a narrower spectrum of microbiota species. For example, many strains of Clostridium, Eubacterium, and Bacteroides can metabolize flavonoids.65 On the other hand, formation of a particular metabolic product, e.g. equol from daidzein, requires specific bacteria (Lactobacillus mucosae, Enterococcus faecium, Finegoldia magna, and Veillonella sp.). Humans lacking these organisms do not produce equol, illustrating the essential role of gut bacteria in metabolizing the flavonoid.66 Bacterial polyphenol metabolism is initiated by a hydrolysis reaction catalyzed by glycoside hydrolases [e.g. lactase phloridzin hydrolase (LPH)], where the glycosyl group is replaced by a hydrogen atom, resulting in the formation of the polyphenol's aglycone form. Colonic Bifidobacteria and Lactobacilli express a variety of glycoside hydrolases, including β-­glucosidases that have similar substrate specificity to host epithelial hydrolases. Select species also express more specialized glycosidases that can hydrolyze O-­glycosides of flavonoids (Figure 2.10). Flavonoid-­O-­glycosidases have been characterized

Figure 2.10  Comparison  of apigenin-­O-­glucoside and apigenin-­C-­glucoside deg-

radation. An O-­glycoside can be degraded by both bacterial and host enzymes in the small intestine or large intestine, whereas a C-­glycoside can only be degraded by specialized glycosidases expressed by select colon microorganisms. The resulting aglycon (i.e. apigenin) can either be absorbed into the bloodstream or further broken down by bacterial enzymes to smaller molecules and carbon dioxide through ring fission and fermentation reactions.

Chapter 2

36 67

in Eubacterium cellulosolvens and Lachnospiraceae strain CG19-­1. Further metabolism of the aglycones can also require specialized enzymes that are strain specific. For example, conversion of daidzein to genistein is catalyzed by daidzein reductase, dihydrodaidzein racemase, and tetrahydrodaidzein reductase. These enzymes have thus far been found in only a few strains of Lactococcus garvieae (e.g. strain 20-­92). The colonic microbiota also participates in second-­pass metabolism of polyphenol derivatives formed in the liver. These biotransformation products can reenter the small intestine through the bile duct and be further metabolized via ring fission into hydroxycinnamates and other smaller phenolic acids.109 These aromatic acids can be fermented into SCFAs, converted to enterodiols and enterolactones, or trafficked back to the liver for elimination.68–73 The intestinal bacteria play an essential role in metabolizing polyphenol derived phenolic acids, as humans lack the esterases necessary to break the ester bonds in ring-­opened polyphenols. Only about one third of polyphenol derived acids are adsorbed in the small intestine; the majority of phenolic acids are modified by colonic microbiota. The modifications include cleavage of conjugating moieties and addition of functional groups, leading to the formation of hydroxyphenylacetic acids and urolithins. The hydroxphenylacetic acids have antioxidant properties, whereas urolithins have estrogenic or anti-­estrogenic activities. In a rat colitis model, urolithin A was found to induce nitric oxide synthase, cyclooxygenase-­2 (COX-­2), prostaglandin E synthase, and prostaglandin E2 in colonic mucosa.74 Phenolic acid metabolites also enhance the growth of Bifidobacterium spp. and Lactobacillus spp. As carbon sources utilized by these microorganisms.75 In addition to polyphenols, phenolic acids metabolites are also derived from lignans. Studies in germ-­free and humanized gnotobiotic rats have found that formation of enterolactones and enterodiols from lignans requires Bacteroides and Clostridium species.76 However, the enzymatic pathways metabolizing these compounds remain to be elucidated.

2.4  Amino Sugars An amino sugar molecule has at least one of the sugar's hydroxyl groups replaced by an amine group. Amino sugars are components of glycolipids, mucopolysaccharides, and mucoproteins, and essential in the formation of connective tissues, nucleotides, antibiotics, and bacterial capsular polysaccharides.77 Glucosamine, galactosamine and neuraminic acid are the principal naturally occurring hexosamines. N-­Acetylglucosamine is the most common form of glucosamine. It is available in food sources such as shellfish and shiitake mushrooms. Sialic acids, a family of neuraminic acid derivatives, can be found in dairy products, whey protein isolates, and eggs of hens. Human milk contains a large amount of glycoconjugates of sialic acid, in particular N-­acetylneuraminic acid. Unlike DF components, amino sugars are readily absorbed in the small intestine and metabolized in liver. However, a fraction passes to the colon, where the amino sugars are fermented by the microbiota. The primary fermentation products are SCFAs and ammonia.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

37

The metabolism of sialic acids has been widely studied because of its potential health benefits on brain development. Sialic acids have a nine-­carbon backbone with α-­ketones attached to the backbone. The core structure of sialic acids is highly conserved, and contains a carboxylic acid ring, glycerol tail, and acetamido and hydroxylic groups. The most common modifications to this core structure are O-­acetylation, O-­lactylation, O-­sulfation and O-­methylation.78 In humans, the most common sialic acids are N-­acetylneuraminic acid (Neu-­5-­Ac) and deaminated N-­acetylglucolylneuraminic acid (Neu-­5-­Gc). Metabolism of sialic acids in the colon typically involves co-­metabolism between the commensal bacteria and host tissue. For example, Neu-­5-­Ac becomes available for microbiota metabolism in the lumen as a digested diet residue or from host enzymatic (glycoside hydrolase or sialidase) activity.79 Host sialidases thus directly affect sialic acid metabolism by the microbiota. Sialidases comprise a broad family of enzymes and include N-­acetylneuraminic acid lyase, N-­acetylmannosamine kinase, N-­acetylmannosamine 6-­phosphate epimerase, N-­acetylglucosamine-­6-­phosphate deacetylase, glucosmine-­6-­ phosphate deaminase, and sialitransferases. Interestingly, phylogenetic distribution analyses of sialic acid catabolism based on bacterial genomes have identified genes encoding sialidases in several colonic bacteria. In addition to Bifidobacterium and Lactobacilli species, sialidases have been identified in Streptococcus pneumoniae, Ruminococcus gnavus, and Bacteroides fragilis, which are all capable of metabolizing amino sugars.80 The deacetylated sialic acids are utilized by colonic bacteria as energy substrates, with acetate as the major metabolic byproduct. Gram-­negative bacteria require active transport for uptake of Neu-­5-­Ac. The transporters vary between species. For example, Escherichia coli uses the N-­acetylneuraminate transporter (NanT) sialase transporter, whereas Haemophilus ducreyi uses tripartite ATP-­independent periplasmic (TRAP) transporters.81 Following uptake, Neu-­5-­Ac is cleaved by acetylneuraminate lyase (NAL) into pyruvate and N-­acetylmannosamine.82 The latter product is further modified by N-­acetylmannosamine kinase (nanK) and N-­acetly-­ mannosamine-­6-­phosphatase to generate N-­acetlglucosamine-­6-­phosphate (GlcNAc-­6-­P). The phosphorylated metabolite is deacylated and deaminated by N-­acetylglucosamine-­6-­phosphate deacetylase (nagA) and glucosamine-­6-­ phosphate deaminase (nagB), respectively. The major products of Neu-­5-­Ac metabolism are fructose-­6-­phosphate and ammonia. The ammonia is used as a nitrogen source by colonic bacteria, and reduces the burden for cells to actively transport ammonium ions from the environment. Because amino sugar metabolism yields a carbon-­rich glycolysis intermediate, it is not surprising that several commensal bacteria utilize sialic acids as a carbon source. For non-­pathogenic E. coli, Neu-­5-­Ac is a preferred nutrient that is metabolized to a similar extent to that for glucuronate even in the presence of sugars such as mannose, fucose and ribose.83 Enteric pathogens, such as Vibrio cholerae, can also utilize sialic acid as a carbon source. This pathogen uses a special sialidase encoded by neuraminidase H (nanH),

38

Chapter 2

which can cleave sialic acid molecules from siaylated gangliosides of intestinal epithelial cells. Moreover, V. cholerae exploits this reaction to unmask the monosialotetrahexosylganglioside (GM1) ganglioside, which is a receptor for cholera toxin.84 Similarly to sialic acids, other amino sugars can also be used as a carbon and energy source by several bacteria. Chitooligosaccharides are oligomers of common amino sugars N-­acetylglucosamine and d-­glucosamine that appears to be a major carbon source for intestinal bacteria.85 Bacteroides fragilis and Clostridium perfringens are among species that can utilize N-­acetyl chitooligosaccharides for anerobic growth.86

2.5  T  ools for Identifying Products of Microbiota Metabolism One of the major challenges in identifying metabolites produced by the microbiota upon exposure to dietary molecules or derived from dietary molecules is their low abundance in biological specimens that are collected to study the intestinal metabolome. Typically, fecal material, rather than intestinal tissue and luminal contents, are collected to profile microbiota metabolites. The use of fecal material is especially common in human subject studies, where options for invasive sample collection are limited. Analyzing fecal material can paint an incomplete picture, as diet-­derived and diet-­induced microbiota metabolites often do not accumulate in the intestinal tract. Instead, these metabolites are used as substrates by other microorganisms in the intestinal community or absorbed by the host. A majority of published studies on interactions between dietary carbohydrates and the gut microbiota use metabolite measurements [typically obtained from liquid chromatography-­ (LC-­) or gas chromatography-­mass spectrometry (GC-­MS) experiments] to complement metagenomic analyses on microbiota community composition. For example, Yin et al. administered xylo-­oligosaccharides (XOS) to weaned piglets, and inferred from 16S rRNA sequencing based metagenome prediction that the XOS increased the relative abundance of several genera (Lactobacillus, Streptococcus, and Turicibacter) to alter, among other functions, lipid metabolism by the intestinal microbiota.87 The authors used LC-­MS experiments to find that XOS administration reduced the level of pentadecanal in the intestine. The reduction in this long chain fatty acid derived aldehyde, thought to inhibit Staphylococcus biofilms, was consistent with the predicted effects of XOS on the metagenome. Similar approaches have been used for investigating the effect of inulin-­t ype fructans in healthy adults,88 dietary mannan oligosaccharides in mice,89 and polyphenols in rats.90 A second, related, challenge lies in determining the chemical identities of microbiota metabolites. This requires an initial annotation of a compound measured by the analytical experiment, followed by confirmation using a chemical standard. Annotation rates reported from untargeted metabolomics studies on the gut microbiota are typically low, ranging from 2% 91 to

Metabolism of Dietary Carbohydrates by Intestinal Bacteria 92

39

5%. The bulk of compounds detected in these studies remain unannotated data features, or metabolic “dark matter”. One reason for the low annotation rate is insufficient coverage of gut microbiota metabolites in publicly available chemical libraries. Ideally, metabolomics data are annotated by matching several orthogonal measures of detected compounds [e.g. accurate mass, chromatographic retention time, tandem mass spectrometry (MS/MS) spectrum, ion mobility collision cross section, etc.] against experimental databases. Generating a comprehensive experimental database is clearly difficult, and further hampered by limited availability of chemical standards as off-­ the-­shelf products. Metabolite annotation and identification clearly remain bottlenecks in untargeted metabolomics, and further efforts are warranted to continue expanding coverage of metabolites from commensal gut bacteria in chemical libraries, while developing alternative, e.g. simulation-­based, approaches to efficiently annotate metabolomics data. A third consideration in identifying microbiota-­derived products of dietary molecules is the potential for co-­metabolism between the microbiota and the host. The metabolites generated by a microbiota species can not only be consumed by other species in the community, but also absorbed by the host. These metabolites can be subsequently modified by host metabolism in the intestine and/or liver and reintroduced into the intestinal tract through the bile duct. An illustrative example is co-­metabolism of tryptophan, an essential amino acid that humans obtain from dietary sources. Comparing germ-­ free (GF) and conventionally raised (CONV-­R) mice Sridharan et al. showed that tryptophan metabolism in murine intestine quantitatively depends on the gut microbiota.93 Conversion of tryptophan to one of the microbiota-­ dependent metabolic products, indole-­3-­acetate (I3A), proceeds through a three-­step pathway, where the first and third steps are catalyzed by enzymes available to both mice and gut bacteria, whereas the second step is strictly bacterial. Krishnan et al. found that high-­fat diet-­induced dysbiosis depletes I3A not only in murine intestine (cecum), but also in circulation and liver.94 Another example is the production of indole and indoxyl sulfate from tryptophan. Indole is abundantly produced in the colon of both mice and humans by fermentation of tryptophan. This requires tryptophanase (tnA), a strictly bacterial enzyme. Indole is absorbed into circulation to reach the liver, where phase I (hydroxylation) and phase II transformation (sulfonation) generate the uremic metabolite indoxyl-­sulfate.95 These examples illustrate that comprehensive identification of microbiota-­dependent diet-­derived metabolites requires thorough characterization of not only intestinal luminal and fecal metabolomes, but also other host compartments, such as intestinal tissue, blood, and liver. The metabolomics studies should be greatly aided by experiments using GF and gnotobiotic animal models that can resolve if a specific diet-­derived metabolite depends on microbial enzymes or can also be generated by the host.96 An emerging approach in identifying metabolites produced by the microbiota is to use predictive tools and metabolic modeling for identifying possible molecules that can be produced by the intestinal community

40

Chapter 2

from a source molecule. In the aforementioned study on tryptophan metabolism,96 Kuntz et al. used a method for computational pathway construction in conjunction with a community-­level metabolic model to predict exclusively bacterial products of aromatic amino acid metabolism. Using LC-­MS experiments, the authors showed that 26 of the 49 predicted metabolites were present in luminal contents of CONV-­R mice and depleted in GF mice. This approach also has the benefit of linking metabolites with the responsible enzymatic pathways as well as potential source organisms that harbor the pathways. Using a community-­level metabolic model, Lei et al.97 correlated changes in metabolite levels and relative abundance of bacteria in an in vitro culture model of murine cecal microbiota. The authors identified a significant association between a tyrosine metabolite (4-­hydroxyphenylacetic acid) and Clostridium species, which they confirmed by measuring the production of the metabolite in a monoculture of Clostridium bolteae. In addition to facilitating the identification of microbiota metabolites and their source organisms, metabolic models could also provide insights into metabolic interactions in the microbiota. For example, Kumar et al.98 used genome-­scale metabolic models (GEMs) of individual gut bacteria to show that a depletion of essential amino acids in plasma of malnourished children could be explained by a reduction in the production of these amino acids by the gut microbiota of these children compared with healthy controls, where the reduction is due to differences in interactions between microbiota species of malnourished and healthy children. Similarly, Hale et al.99 analyzed targeted metabolomics data on amino acids and SCFAs using GEMs to identify distinct metabolic roles performed by microbiota associated with two different subtypes of colorectal cancer. In recent years, a number of tools have become available for automated reconstruction of GEMs, including models of human gut microbiota organisms,100,101 which should further accelerate the adoption of powerful modeling approaches for identifying diet-­derived gut microbiota metabolites and investigating both microbiota and host pathways that engage these metabolites.

2.6  Future Directions The examples discussed in this chapter as well as many other studies provide robust evidence that the gut microbiota is a critical mediator of interactions between dietary residues, host metabolism, and physiological processes, although many of the molecular aspects underpinning the interactions remain unknown. The implications for human health are broad, extending beyond the intestine to whole body systems and distal organs, including the brain. For example, results from recent studies indicate that GOS utilization by gut bacteria could benefit cognitive functions, including memory. Brain-­derived neurotrophic factor (BDNF) supports learning and memory

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

41

by promoting differentiation of neurons and formation of synapses. Rats fed GOS showed increased BDNF in hippocampus, N-­methyl-­d-­aspartate receptor (NMDAR) subunits in hippocampus and frontal cortex.102 In contrast, BDNF and NMDAR subunits were decreased in GF rats. The authors also reported that the GOS diet increased the level of peptide YY (PYY), a microbiota-­dependent gut hormone, in plasma while expanding the relative abundance of Bifidobacterium spp., and that PYY stimulates BDNF expression in vitro. These findings indicate that the effects of GOS feeding on brain cognitive function are linked to the oligosaccharide's proliferative potency on select gut bacteria. The physiological effects of dietary carbohydrate fermentation by the gut microbiota are likely to be mediated by the metabolic end products or byproducts. A well-­known set of fermentation products comprise SCFAs, which have been unambiguously linked in animal models to improved intestinal epithelial barrier function, resilience of gut health, and regulation of whole-­body energy metabolism. On the other hand, linking diet-­derived metabolites with specific physiological effects in humans remains challenging. This is partially due to the inherent difficulties in conducting gain-­ and loss-­of-­function experiments in human subjects, which necessitates the use of animal or in vitro models. In this vein, there is less data available from controlled studies involving human subjects. In particular, quantitative data on biochemical fluxes describing enzymatic and signaling activity are severely lacking. Studies that have been published to date mostly report on DNA abundance (and to a lesser extent transcript levels) and concentrations of metabolites. While these data are clearly valuable, they do not necessarily reflect metabolic activity. In this regard, an important future direction is to develop more comprehensive analysis approaches that combine other “omics” data, including proteomics and metabolomics data collected from stable isotopic labeling experiments. Ideally, these analyses can be combined with human intervention trials, with the goal of elucidating mechanisms that are directly relevant to human health. Key questions that warrant further investigation include: 1. What are the in vivo fluxes of microbiota metabolite production and host uptake under different diets (e.g. varying ratios of polysaccharides and oligosaccharides) and perturbed microbiota conditions (e.g. disease states)? 2. What are the physiological roles of these metabolites in different tissues (e.g. intestine, liver, brain, etc.) and what are the mechanisms (i.e. receptors and target pathways)? 3. Does the demand by the host for a specific metabolite (e.g. butyrate) drive a change in microbial metabolism and hence community composition? 4. What are the time scales of changes in the microbiota upon alteration of diet? Addressing these questions should yield important insights into the biochemical basis of the clearly profound effects that diet composition exerts on human health, and provide cause-­and-­effect explanations that will facilitate rational design of dietary interventions for both disease prevention and treatment.

42

Chapter 2

Acknowledgements This work was supported in part by a grant from the NIH (R01 AT010282) to AJ and KL.

References 1. L. M. Nilsson, A. Winkvist, M. Eliasson, J. H. Jansson, G. Hallmans, I. Johansson, B. Lindahl, P. Lenner and B. Van Guelpen, Eur. J. Clin. Nutr., 2012, 66, 694–700. 2. F. Ferretti and M. Mariani, Int. J. Environ. Res. Public Health, 2017, 14(10), 1174. 3. V. Lombard, T. Bernard, C. Rancurel, H. Brumer, P. M. Coutinho and B. Henrissat, Biochem. J., 2010, 432, 437–444. 4. L. C. MacDonald and B. W. Berger, J. Biol. Chem., 2014, 289, 312–325. 5. X. Ze, S. H. Duncan, P. Louis and H. J. Flint, ISME J., 2012, 6, 1535–1543. 6. E. C. Leitch, A. W. Walker, S. H. Duncan, G. Holtrop and H. J. Flint, Environ. Microbiol., 2007, 9, 667–679. 7. A. El Kaoutari, F. Armougom, J. I. Gordon, D. Raoult and B. Henrissat, Nat. Rev. Microbiol., 2013, 11, 497–504. 8. B. S. Samuel, A. Shaito, T. Motoike, F. E. Rey, F. Backhed, J. K. Manchester, R. E. Hammer, S. C. Williams, J. Crowley, M. Yanagisawa and J. I. Gordon, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 16767–16772. 9. J. Rafter, M. Bennett, G. Caderni, Y. Clune, R. Hughes, P. C. Karlsson, A. Klinder, M. O'Riordan, G. C. O'Sullivan, B. Pool-­Zobel, G. Rechkemmer, M. Roller, I. Rowland, M. Salvadori, H. Thijs, J. Van Loo, B. Watzl and J. K. Collins, Am. J. Clin. Nutr., 2007, 85, 488–496. 10. J. L. Buttriss and C. S. Stokes, Nutr. Bull., 2008, 33, 186–200. 11. Y. O. Li and A. R. Komarek, Food Qual. Saf., 2017, 1, 47–59. 12. J. W. Anderson, P. Baird and R. Davis, Nutr. Rev., 2009, 67(4), 188–205. 13. E. Capuano, Crit. Rev. Food Sci. Nutr., 2017, 57, 3543–3564. 14. J. M. Jones, Nutr. J., 2014, 13(1), 34. 15. M. M.-­L. Grundy, C. H. Edwards, A. R. Mackie, M. J. Gidley, P. J. Butterworth and P. R. Ellis, Br. J. Nutr., 2016, 116, 816–833. 16. Carbohydrates in human nutrition. Report of a Joint FAO/WHO Expert Consultation, FAO Food Nutr. Pap., 1998, 66, 1–140. 17. S. M. Notley, B. Pettersson and L. Wagberg, J. Am. Chem. Soc., 2004, 126, 13930–13931. 18. L. R. de Souza Moreira, D. Lo Sciuto and E. X. F. Filho, in New and Future Developments in Microbial Biotechnology and Bioengineering, ed. V. K. Gupta, Elsevier, Amsterdam, 2016, pp. 165–175, DOI: 10.1016/ B978-­0-­444-­63507-­5.00014-­9. 19. S. A. S. Gropper, J. L. Smith, and J. L. Groff, Fiber, in Advanced Nutrition and Human Metabolism, Wadsworth/Cengage Learning, Australia, 2009, ch. 4, pp. 107–121.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

43

20. D. Mudgil, in Dietary Fiber for the Prevention of Cardiovascular Disease, 2017. 21. G. Suen, P. J. Weimer, D. M. Stevenson, F. O. Aylward, J. Boyum, J. Deneke, C. Drinkwater, N. N. Ivanova, N. Mikhailova, O. Chertkov, L. A. Goodwin, C. R. Currie, D. Mead and P. J. Brumm, PLoS One, 2011, 6, e18814. 22. A. E. Naas, A. K. MacKenzie, B. Dalhus, V. G. H. Eijsink and P. B. Pope, Sci. Rep., 2015, 5, 11666. 23. Y. Zhou, P. B. Pope, S. Li, B. Wen, F. Tan, S. Cheng, J. Chen, J. Yang, F. Liu, X. Lei, Q. Su, C. Zhou, J. Zhao, X. Dong, T. Jin, X. Zhou, S. Yang, G. Zhang, H. Yang, J. Wang, R. Yang, V. G. Eijsink and J. Wang, Sci. Rep., 2014, 4, 5288. 24. S. Cai, J. Li, F. Z. Hu, K. Zhang, Y. Luo, B. Janto, R. Boissy, G. Ehrlich and X. Dong, Appl. Environ. Microbiol., 2010, 76, 3818–3824. 25. P. Gupta, K. Samant and A. Sahu, Int. J. Microbiol., 2012, 2012, 5. 26. R. Hespell, Microbiol. Sci., 1988, 5, 362–365. 27. L. R. S. Moreira and E. X. F. Filho, Appl. Microbiol. Biotechnol., 2008, 79, 165. 28. S. S. Gropper, J. L. Smith and J. L. Groff, in Advanced Nutrition and Human Metabolism, Wadsworth, Canada, 5th edn, 2005, pp. 107–121. 29. L. Flutto, in Encyclopedia of Food Sciences and Nutrition, Academic Press, 2nd edn, 2003, pp. 4440–4449. 30. E. N. Makarova, E. G. Shakhmatov, a. E. V. Udoratina and A. V. Kutchin, Russ. Chem. Bull. Int. Ed., 2015, 64, 1302–1318. 31. N. Larsen, C. Bussolo de Souza, L. Krych, T. Barbosa Cahú, M. Wiese, W. Kot, K. M. Hansen, A. Blennow, K. Venema and L. Jespersen, Front. Microbiol., 2019, 10, 223. 32. F. Fåk, G. Jakobsdottir, E. Kulcinskaja, N. Marungruang, C. Matziouridou, U. Nilsson, H. Stålbrand and M. Nyman, PLoS One, 2015, 10(5), e0127252. 33. J. H. Cummings and G. T. Macfarlane, Clin. Nutr., 1997, 16, 3–11. 34. S. A. van Hijum, S. Kralj, L. K. Ozimek, L. Dijkhuizen and I. G. van Geel-­ Schutten, Microbiol. Mol. Biol. Rev., 2006, 70, 157–176. 35. M. Tieking, M. A. Ehrmann, R. F. Vogel and M. G. Ganzle, Appl. Microbiol. Biotechnol., 2005, 66, 655–663. 36. S. J. Reid and V. R. Abratt, Appl. Microbiol. Biotechnol., 2005, 67, 312–321. 37. M. Intanon, S. L. Arreola, N. H. Pham, W. Kneifel, D. Haltrich and T. H. Nguyen, FEMS Microbiol. Lett., 2014, 353, 89–97. 38. T. Thongaram, J. L. Hoeflinger, J. Chow and M. J. Miller, J. Agric. Food Chem., 2017, 65, 4184–4192. 39. J. M. Andersen, R. Barrangou, M. Abou Hachem, S. Lahtinen, Y. J. Goh, B. Svensson and T. R. Klaenhammer, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 17785–17790. 40. M. Nishimoto and M. Kitaoka, Appl. Environ. Microbiol., 2007, 73, 6444–6449.

44

Chapter 2

41. M. C. Marin-­Manzano, L. Abecia, O. Hernandez-­Hernandez, M. L. Sanz, A. Montilla, A. Olano, L. A. Rubio, F. J. Moreno and A. Clemente, J. Agric. Food Chem., 2013, 61, 7560–7567. 42. S. Yaron, D. Shachar, L. Abramas, A. Riskin, D. Bader, I. Litmanovitz, F. Bar-­Yoseph, T. Cohen, L. Levi, Y. Lifshitz, R. Shamir and R. Shaoul, J. Pediatr. Gastroenterol. Nutr., 2013, 56, 376–381. 43. A. J. Noll, Y. Yu, Y. Lasanajak, G. Duska-­McEwen, R. H. Buck, D. F. Smith and R. D. Cummings, Biochem. J., 2016, 473, 1343–1353. 44. M. Chichlowski, G. De Lartigue, J. B. German, H. E. Raybould and D. A. Mills, J. Pediatr. Gastroenterol. Nutr., 2012, 55, 321–327. 45. G. Schumacher, G. Bendas, B. Stahl and C. Beermann, Nutrition, 2006, 22, 620–627. 46. K. Shams-­Ud-­Doha, E. N. Kitova, P. I. Kitov, Y. St-­Pierre and J. S. Klassen, Anal. Chem., 2017, 89, 4914–4921. 47. X. Cong, W. Xu, S. Janton, W. A. Henderson, A. Matson, J. M. McGrath, K. Maas and J. Graf, PLoS One, 2016, 11, e0152751. 48. S. Arboleya, G. Solis, N. Fernandez, C. G. de los Reyes-­Gavilan and M. Gueimonde, Gut Microbes, 2012, 3, 583–588. 49. M. A. Underwood, S. Gaerlan, M. L. De Leoz, L. Dimapasoc, K. M. Kalanetra, D. G. Lemay, J. B. German, D. A. Mills and C. B. Lebrilla, Pediatr. Res., 2015, 78, 670–677. 50. M. M. Murphy, J. S. Douglass and A. Birkett, J. Am. Diet. Assoc., 2008, 108, 67–78. 51. D. F. Birt, T. Boylston, S. Hendrich, J.-­L. Jane, J. Hollis, L. Li, J. McClelland, S. Moore, G. J. Phillips, M. Rowling, K. Schalinske, M. P. Scott and E. M. Whitley, Adv. Nutr., 2013, 4, 587–601. 52. G. Janusz, A. Pawlik, J. Sulej, U. Świderska-­Burek, A. Jarosz-­Wilkołazka and A. Paszczyński, FEMS Microbiol. Rev., 2017, 41, 941–962. 53. A. N. Begum, C. Nicolle, I. Mila, C. Lapierre, K. Nagano, K. Fukushima, S. M. Heinonen, H. Adlercreutz, C. Remesy and A. Scalbert, J. Nutr., 2004, 134, 120–127. 54. A. Tresserra-­Rimbau, R. M. Lamuela-­Raventos and J. J. Moreno, Biochem. Pharmacol., 2018, 156, 186–195. 55. H. Cardenas, D. Arango, C. Nicholas, S. Duarte, G. J. Nuovo, W. He, O. H. Voss, M. E. Gonzalez-­Mejia, D. C. Guttridge, E. Grotewold and A. I. Doseff, Int. J. Mol. Sci., 2016, 17, 323. 56. J. Warford, Q. R. Jones, M. Nichols, V. Sullivan, H. P. Rupasinghe and G. S. Robertson, J. Neuroimmunol., 2014, 268, 71–83. 57. A. Steensma, M. A. Faassen-­Peters, H. P. Noteborn and I. M. Rietjens, J. Agric. Food Chem., 2006, 54, 8006–8012. 58. J. H. Kim, D. K. Hwang, J. Y. Moon, Y. Lee, J. S. Yoo, D. H. Shin and H. S. Lee, Molecules, 2017, 22(4), 670. 59. S. C. Liang, Y. L. Xia, J. Hou, G. B. Ge, J. W. Zhang, Y. Q. He, J. Y. Wang, X. Y. Qi and L. Yang, J. Pharm. Sci., 2016, 105, 808–816. 60. E. U. Graefe, J. Wittig, S. Mueller, A. K. Riethling, B. Uehleke, B. Drewelow, H. Pforte, G. Jacobasch, H. Derendorf and M. Veit, J. Clin. Pharmacol., 2001, 41, 492–499.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

45

61. F. Perez-­Vizcaino, J. Duarte and C. Santos-­Buelga, J. Sci. Food Agric., 2012, 92, 1822–1825. 62. H. T. Aboufarrag, P. W. Needs, G. Rimbach and P. A. Kroon, Nutrients, 2019, 11(12), 2872. 63. S. Yokoyama, K. Oshima, I. Nomura, M. Hattori and T. Suzuki, J. Bacteriol., 2011, 193, 5570–5571. 64. A. Hassaninasab, Y. Hashimoto, K. Tomita-­Yokotani and M. Kobayashi, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 6615–6620. 65. H. Diao, Z. Gao, B. Yu, P. Zheng, J. He, J. Yu, Z. Huang, D. Chen and X. Mao, J. Anim. Sci. Biotechnol., 2016, 7, 32. 66. K. Decroos, S. Vanhemmens, S. Cattoir, N. Boon and W. Verstraete, Arch. Microbiol., 2005, 183, 45–55. 67. I. B. Jaganath, W. Mullen, C. A. Edwards and A. Crozier, Free Radical Res., 2006, 40, 1035–1046. 68. C. Schroder, A. Matthies, W. Engst, M. Blaut and A. Braune, Appl. Environ. Microbiol., 2013, 79, 3494–3502. 69. L. Schoefer, R. Mohan, A. Schwiertz, A. Braune and M. Blaut, Appl. Environ. Microbiol., 2003, 69, 5849–5854. 70. A. Braune, M. Gutschow, W. Engst and M. Blaut, Appl. Environ. Microbiol., 2001, 67, 5558–5567. 71. H. Schneider and M. Blaut, Arch. Microbiol., 2000, 173, 71–75. 72. H. Schneider, A. Schwiertz, M. D. Collins and M. Blaut, Arch. Microbiol., 1999, 171, 81–91. 73. L. R. Krumholz, R. L. Crawford, M. E. Hemling and M. P. Bryant, J. Bacteriol., 1987, 169, 1886–1890. 74. M. Larrosa, A. Gonzalez-­Sarrias, M. J. Yanez-­Gascon, M. V. Selma, M. Azorin-­Ortuno, S. Toti, F. Tomas-­Barberan, P. Dolara and J. C. Espin, J. Nutr. Biochem., 2010, 21, 717–725. 75. D. Bialonska, S. G. Kasimsetty, S. I. Khan and D. Ferreira, J. Agric. Food Chem., 2009, 57, 10181–10186. 76. E. Bowey, H. Adlercreutz and I. Rowland, Food Chem. Toxicol., 2003, 41, 631–636. 77. K. Skarbek and M. J. Milewska, Carbohydr. Res., 2016, 434, 44–71. 78. S. M. Muthana, C. T. Campbell and J. C. Gildersleeve, ACS Chem. Biol., 2012, 7, 31–43. 79. W. G. Lewis, L. S. Robinson, N. M. Gilbert, J. C. Perry and A. L. Lewis, J. Biol. Chem., 2013, 288, 12067–12079. 80. L. E. Tailford, C. D. Owen, J. Walshaw, E. H. Crost, J. Hardy-­Goddard, G. Le Gall, W. M. de Vos, G. L. Taylor and N. Juge, Nat. Commun., 2015, 6, 7624. 81. M. Fischer, A. P. Hopkins, E. Severi, J. Hawkhead, D. Bawdon, A. G. Watts, R. E. Hubbard and G. H. Thomas, J. Biol. Chem., 2015, 290, 27113–27123. 82. S. L. Wang, Y. L. Li, Z. Han, X. Chen, Q. J. Chen, Y. Wang and L. S. He, Mar. Drugs, 2018, 16(3), 80. 83. D. E. Chang, D. J. Smalley, D. L. Tucker, M. P. Leatham, W. E. Norris, S. J. Stevenson, A. B. Anderson, J. E. Grissom, D. C. Laux, P. S. Cohen and T. Conway, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 7427–7432.

46

Chapter 2

84. J. Cervin, A. M. Wands, A. Casselbrant, H. Wu, S. Krishnamurthy, A. Cvjetkovic, J. Estelius, B. Dedic, A. Sethi, K. L. Wallom, R. Riise, M. Backstrom, V. Wallenius, F. M. Platt, M. Lebens, S. Teneberg, L. Fandriks, J. J. Kohler and U. Yrlid, PLoS Pathog., 2018, 14, e1006862. 85. B. B. Aam, E. B. Heggset, A. L. Norberg, M. Sorlie, K. M. Varum and V. G. Eijsink, Mar. Drugs, 2010, 8, 1482–1517. 86. H. C. Chen, C. C. Chang, W. J. Mau and L. S. Yen, FEMS Microbiol. Lett., 2002, 209, 53–56. 87. J. Yin, F. Li, X. Kong, C. Wen, Q. Guo, L. Zhang, W. Wang, Y. Duan, T. Li, Z. Tan and Y. Yin, Food Funct., 2019, 10, 2701–2709. 88. D. Vandeputte, G. Falony, S. Vieira-­Silva, J. Wang, M. Sailer, S. Theis, K. Verbeke and J. Raes, Gut, 2017, 66, 1968–1974. 89. L. R. Hoving, S. Katiraei, M. Heijink, A. Pronk, L. van der Wee-­Pals, T. Streefland, M. Giera, K. Willems van Dijk and V. van Harmelen, Mol. Nutr. Food Res., 2018, 62, e1700942. 90. Q. Li, F. Liu, J. Liu, S. Liao and Y. Zou, Nutrients, 2019, 11(5), 1017. 91. S. Fujisaka, J. Avila-­Pacheco, M. Soto, A. Kostic, J. M. Dreyfuss, H. Pan, S. Ussar, E. Altindis, N. Li, L. Bry, C. B. Clish and C. R. Kahn, Cell Rep., 2018, 22, 3072–3086. 92. E. A. Franzosa, A. Sirota-­Madi, J. Avila-­Pacheco, N. Fornelos, H. J. Haiser, S. Reinker, T. Vatanen, A. B. Hall, H. Mallick, L. J. McIver, J. S. Sauk, R. G. Wilson, B. W. Stevens, J. M. Scott, K. Pierce, A. A. Deik, K. Bullock, F. Imhann, J. A. Porter, A. Zhernakova, J. Fu, R. K. Weersma, C. Wijmenga, C. B. Clish, H. Vlamakis, C. Huttenhower and R. J. Xavier, Nat. Microbiol., 2019, 4, 293–305. 93. G. V. Sridharan, K. Choi, C. Klemashevich, C. Wu, D. Prabakaran, L. B. Pan, S. Steinmeyer, C. Mueller, M. Yousofshahi, R. C. Alaniz, K. Lee and A. Jayaraman, Nat. Commun., 2014, 5, 5492. 94. S. Krishnan, Y. Ding, N. Saeidi, M. Choi, G. V. Sridharan, D. H. Sherr, M. L. Yarmush, R. C. Alaniz, A. Jayaraman and K. Lee, Cell Rep., 2019, 28, 3285. 95. W. R. Wikoff, A. T. Anfora, J. Liu, P. G. Schultz, S. A. Lesley, E. C. Peters and G. Siuzdak, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 3698–3703. 96. S. Kuntz, C. Kunz, C. Borsch, E. Vazquez, R. Buck, M. Reutzel, G. P. Eckert and S. Rudloff, Mol. Nutr. Food Res., 2019, e1900035. 97. M. Lei, R. Menon, S. Manteiga, N. Alden, C. Hunt, R. C. Alaniz, K. Lee and A. Jayaraman, mSystems, 2019, 4(6), e00724–e00819. 98. M. Kumar, B. Ji, P. Babaei, P. Das, D. Lappa, G. Ramakrishnan, T. E. Fox, R. Haque, W. A. Petri, F. Backhed and J. Nielsen, Metab. Eng., 2018, 49, 128–142. 99. V. L. Hale, P. Jeraldo, J. Chen, M. Mundy, J. Yao, S. Priya, G. Keeney, K. Lyke, J. Ridlon, B. A. White, A. J. French, S. N. Thibodeau, C. Diener, O. Resendis-­Antonio, J. Gransee, T. Dutta, X. M. Petterson, J. Sung, R. Blekhman, L. Boardman, D. Larson, H. Nelson and N. Chia, Genome Med., 2018, 10, 78.

Metabolism of Dietary Carbohydrates by Intestinal Bacteria

47

100. K. C. H. van der Ark, R. G. A. van Heck, V. A. P. Martins Dos Santos, C. Belzer and W. M. de Vos, Microbiome, 2017, 5, 78. 101. S. Magnusdottir, A. Heinken, L. Kutt, D. A. Ravcheev, E. Bauer, A. Noronha, K. Greenhalgh, C. Jager, J. Baginska, P. Wilmes, R. M. Fleming and I. Thiele, Nat. Biotechnol., 2017, 35, 81–89. 102. H. M. Savignac, G. Corona, H. Mills, L. Chen, J. P. Spencer, G. Tzortzis and P. W. Burnet, Neurochem. Int., 2013, 63, 756–764. 103. B. C. Saha and R. J. Bothast, in Biopolymers, American Chemical Society, 1999, vol. 723, ch. 11, pp. 167–194. 104. J. Herlet, W. H. Schwarz, V. V. Zverlov, W. Liebl and P. Kornberger, Biotechnol. Biofuels, 2018, 11, 238. 105. G. Dongowski, A. Lorenz and H. Anger, Appl. Environ. Microbiol., 2000, 66, 1321. 106. V. Ochoa-­Jiménez and M. Tiznado-­Hernández, Emirates J. Food Agric., 2018, 30, DOI: 10.9755/ejfa.2018.v30.i11.1858. 107. I. R. Silva, C. Jers, A. S. Meyer and J. D. Mikkelsen, New Biotechnol., 2016, 33, 41–54. 108. R. Zhao, R. Zhao, Y. Tu, X. Zhang, L. Deng and X. Chen, PLoS One, 2018, 13, e0197067. 109. L. Hanske, G. Loh, S. Sczesny, M. Blaut and A. Braune, J. Nutr., 2009, 139(6), 1095–1102.

Chapter 3

The Microbiome and Amino Acid Metabolism N. E. Diethera and B. P. Willing*a a

Department of Agriculture, Food & Nutritional Science, University of Alberta, Ag/For Centre, Edmonton, Alberta, T6G 2P5, Canada *E-­mail: [email protected]

3.1  Introduction Host–microbiota metabolic interactions are highly networked processes, where the microbiota provides many enzymatic capabilities not encoded in mammalian genomes.1 One of these enzymatic capacities is the ability to catabolize amino acids into a diverse range of end products. When high levels of dietary protein are consumed, rapid shifts in the gut microbiota occur due to the changes in nutrient flow within the digestive tract.2 While most dietary protein is digested and absorbed by the terminal ileum, some undigested protein reaches the large intestine where it can be rapidly converted into many bioactive metabolites by the microbiota.3 The production of these metabolites can affect host health through alterations to gut barrier function, epithelial cell DNA integrity, or by entering the systemic circulation.4,5 Microbes also sequester amino acids in their own cells for production of microbial proteins.6 For gut microbes, the sources of these amino acids are unabsorbed dietary protein, host secretions, and sloughed epithelial cells. The fate of amino acids is determined by the enzymatic repertoire of the microbial species within the gastrointestinal tract, as well as by dietary factors, such as the availability of fermentable polysaccharides.4,7 While a   Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

48

The Microbiome and Amino Acid Metabolism

49

number of key protein fermenting species and their metabolites have been identified,8,9 the effects of these metabolites on health are still being explored. This chapter will explore the use of protein by microbes in different parts of the gastrointestinal tract, the metabolic pathways used to catabolize amino acids, and the metabolites produced, and discuss emerging research on the effects of these products on health.

3.2  Microbes and Protein in the Gut Compartments 3.2.1  Microbes and Protein in the Small Intestine The small intestine is the site of highly efficient peptide digestion and absorption by the host. However, when dietary peptides and amino acids reach the small intestine, there is also rapid disappearance within the gut. This first pass metabolism is responsible for a great degree of amino acid disappearance, limiting the concentration of amino acids reaching the portal vein and becoming available to the host. To date, the extent to which microbial fermentation and sequestration of protein plays a role in this disappearance has been difficult to quantify. This is in part due to the difficulty in mimicking the conditions of the ileum in vitro. In particular, fast transit time and fluctuating levels of pancreatic enzymes lead to challenges with interpretation of culture-­based findings; where cultures typically occur over longer periods. While the majority of catabolism and absorption occurring is due to host derived peptidases and transporters, as well as demand for amino acids by enterocytes, the utilization of amino acids by microbes should not be overlooked. The microbiota of the small intestine certainly has the capacity for proteolytic fermentation, as evidenced by the results of culture experiments.10 Furthermore, microbes such as Escherichia coli, which are found in the ileum, are known to possess serine proteases.11 When cultured on amino acid containing media and sequentially sub-­ cultured, bacterial communities from all three segments of the small intestine of pigs show high rates of utilization of lysine, arginine, threonine, glutamate, and leucine.10 Species identified as contributing to this effect were Klebsiella spp., Succinivibrio dextrinosolvens, E. coli, Streptococcus spp., Megasphaera elsdenii, Mitsuokella spp., Anaerovibrio lipolytica, and Acidaminococcus fermentans. However, one limitation of this study is that the metabolic fate of these amino acids was not measured, and it is unclear whether their disappearance represents incorporation into microbial protein or fermentation for energy utilization. Moderate dietary protein restriction results in changes to ileal microbiota structure, decreasing Clostridium spp., and amine concentrations, indicating that some portion of the disappearance is due to protein fermentation.12 Regardless of fate, microbial utilization of 50–90% of amino acids provided over a period of 24 hours demonstrates a high degree of competition between host and microbes for amino acids within the small intestine.10 In particular, extensive microbial utilization of lysine, an important amino acid for growth and muscle turnover, indicates that microbial

50

Chapter 3

utilization may be built into the minimum requirements for this indispensable amino acid.10 Understanding microbial requirements is also complex due to their ability to synthesize amino acids and our limited understanding of the role of microbial protein in supplying host requirements. The results of experiments using 15N labelled nitrogen sources indicate that at least a portion of host amino acid requirements are met by microbial synthesized amino acids produced from various sources including endogenous protein (host secretions and sloughed cells) as well as dietary sources.13 These proteins and peptides can be catabolized, and the amino acids converted to those required by the bacteria. Most experimental evidence does not support any biologically relevant absorption of amino acids by the host past the ileocecal junction,14 which is further supported by a lack of PEPT1 peptide transporters in the large intestine.13 However, some uptake cannot be discounted due to the presence of small amounts of PEPT2 transporters, as well as the results of some experiments indicating an improvement in nitrogen balance when amino acids are infused into the large intestine.13 Experimental evidence to date supports this, demonstrating a much larger microbial contribution to host amino acid balance from microbial protein found in the small intestine.13,15 In the small intestine, microbial utilization of dietary protein primarily affects amino acid availability to the host through incorporation into microbial cells, though some fermentation may occur. However, this effect is offset to some degree by the ability of microbial protein to contribute to host amino acid balance. Our understanding of how these effects come together and shape host physiology is still limited. More research in animals colonized with defined communities of microbes may help us understand the role of specific protein fermenting microbes found in the ileum, such as E. coli, or Lactobacillus spp., under normal physiological conditions.16

3.2.2  Microbes and Protein in the Large Intestine Compared with the small intestine, much more is understood about microbial metabolism of protein in the large intestine. Slower digesta transit, and a dense population of microbes in the large intestine facilitates intense fermentation of substrates reaching the large intestine. These substrates originate primarily from the diet, but also include sloughed cells, pancreatic enzymes, and bile acids that may flow past the ileum. In a modern western diet, more than double the amount of dietary protein may reach the large intestine and become available for proteolytic fermentation compared with lower protein diets.17 Proteolytic fermentation increases over the length of the large intestine due to changing fermentation conditions in the gut, while the physiology of the large intestine means that many fermentation products are able to accumulate within the colon. In the proximal colon, very little proteolytic fermentation occurs due to the availability of dietary fibre.18 The fermentation of fiber in the proximal colon produces short-­chain fatty acids which, in turn, create acidic conditions

The Microbiome and Amino Acid Metabolism

51 18

(pH 5.5) that are inhibitory to most extracellular bacterial proteases. The availability of abundant energy sources in this segment of the gastrointestinal tract may also lead to sequestration of amino acids as microbial protein to support growth and division rather than fermentation. In the distal colon these fermentable polysaccharides are depleted, creating conditions which necessitate proteolytic fermentation for energy.3 The fermentation conditions of the distal colon are accompanied by negligible host absorption of free-­amino acids generated by bacterial proteases; this can lead to the generation of much higher concentrations of bioactive microbial metabolites, such as amines, aryl hydrocarbon receptor ligands, and neurotransmitters, which accumulate over the length of the colon.3,19,20 Deamination inherently releases ammonia into the colonic lumen, which can have detrimental effects on the colonic epithelial cell metabolism of butyrate which will be subsequently discussed.21,22 Fermentation of aromatic amino acids releases compounds including phenol, p-­cresol, and indole, which all exert effects on the barrier integrity of the gut epithelium.23,24 The extensive fermentation and accumulation of end products occurring in the large intestine has led to increasing interest in the role that proteolytic fermentation in the large intestine plays in conditions such as colitis and colorectal cancer. More recently, increased understanding of the bioactivity of many other compounds generated, such as amines and indole-­containing compounds, has led to interest in the roles that proteolytic metabolites play outside of the gastrointestinal tract, including in conditions such as multiple sclerosis, Alzheimer's disease, obesity and insulin resistance.

3.3  Metabolic Pathways of Proteolytic Fermentation The interaction between host and microbial metabolic pathways is an intricate network in which microbes contribute essential enzymatic capacities that the host is not capable of. Approximately 50% of all enzymatic reactions occurring within the host–microbiome metabolic network are only found within microbial genomes, and when a broader view of metabolic pathways is taken, 75% of pathways require microbial contributions.23 Because of this highly networked metabolism and the difficulties of assessing contributions of individual microbes in the complex gut community, most of what is known about proteolytic fermentation in the gut is based on culture of microbes in vitro, in particular work conducted using Clostridium spp. which are well known for their ability to ferment amino acids.25,26 Microbial catabolism starts with extracellular proteases and peptidases that are able to hydrolyze protein that has not been digested and absorbed by the host. The resulting peptides and amino acids are available for transport into the microbial cell. Three key transporters used in this process by well-­ known protein fermenting microbes are25    1. ABC transporters of oligopeptides, methionine, and branched-­chain amino acids

52

Chapter 3

2. serine/threonine symporters 3. arginine/ornithine antiporters    Peptides are preferentially imported into the cell due to the energy efficiency of moving one larger molecule instead of several free-­amino acids.18 Once inside the cell, amino acids are incorporated into microbial protein, or catabolized. This catabolism starts with either deamination or decarboxylation and is followed by a series of reactions allowing any of the 20 amino acids to be transformed into one of seven intermediates than can enter into the tricarboxylic acid cycle.27,28

3.3.1  Deamination Deamination is typically the first step in microbial catabolism of amino acids. Deamination frees the carbon skeleton of amino acids by removal of the amine group. In the anaerobic environment of the gastrointestinal tract, this deamination always involves both oxidation and reduction reactions, where the oxidation step is quite similar to those performed aerobically with the exception of oxygenation and fatty-­acid oxidation reactions, which cannot be performed.29 This reaction generates a keto-­acid, but also inherently creates ammonia.6 Non-­specific deamination of both d-­ and l-­amino acids occurs via non-­specific flavoprotein oxidases.6,27 Some amino acids are also deaminated by specific nicotinamide adenine dinucleotide phosphate (NADP+)-­linked dehydrogenases, or transaminated to generate ammonia and pyruvate.27 This step is also the starting point for the production of short-­chain fatty acids from proteolytic fermentation.30

3.3.2  Decarboxylation Decarboxylation of basic amino acids (arginine, histidine, lysine) can be performed by many common constituents of the gut microbiota, generating amines and carbon dioxide.28,30 Decarboxylase enzymes are amino acid specific, and their action is favored under acidic conditions, compared with deaminases.31 Arginine decarboxylation generates putrescine, as well as spermidine and spermine. Histidine is decarboxylated into histamine. Lysine can be decarboxylated into cadaverine, the biogenic amine that is the least characterized with respect to its effects on host physiology. Decarboxylation of aromatic amino acids is also performed by gut microbes, resulting in a more diverse array of metabolites, many of which are now being identified as important in regulating host physiology within the gut and beyond.

3.3.3  Stickland Reaction Stickland reactions are coupled deamination reactions, where one amino acid is oxidatively deaminated and decarboxylated while the other is reduced.31 Clostridia are particularly well studied with respect to their

The Microbiome and Amino Acid Metabolism

53

ability to perform these reactions, which generate keto-­acids, carboxylic acids, carbon dioxide, and ammonia.9,32 As with deamination reactions, the products of Stickland reactions are easily degraded resulting in the production of short-­chain fatty acids, succinate, and lactate.33 When these reactions are performed, alanine, leucine, isoleucine, valine, and histidine are preferentially oxidized while glycine, proline, ornithine, arginine, and tryptophan are preferentially reduced.31 Tyrosine and phenylalanine are the least utilized amino acids for these reactions.34 While microbes use substrate level phosphorylation to generate adenosine triphosphate (ATP) from these reactions, from a host health perspective, the generation of fatty acids, and ammonia are most important. While these reactions provide the general framework for amino acid catabolism by microbes, more research is required examining dietary changes in animals colonized with defined microbial communities to truly understand which reactions are favored under different physiological conditions. Although metabolic predictions are becoming more feasible with the availability of computational modelling tools and complete microbial genomes, this work in vivo will allow better understanding of how metabolic capacities are affected by diet, microbial community, and health for better prediction of available metabolite pools and their effects. This work is of particular interest for aromatic amino acids, as increased structural complexity allows for catabolism into multiple end products with different effects on host physiology. As many of these compounds are implicated in health conditions, a better understanding of how and when they are produced by the gut microbiota is important for future work focused on using microbes to affect host disease outcomes.

3.4  M  etabolites Produced by Proteolytic Fermentation Microbial fermentation of protein produces short-­chain fatty acids, similar to fermentation of carbohydrates. However, proteolytic fermentation also generates other unique fermentation products, such as ammonia, biogenic diamines, and polyamines, branched chain fatty acids, indoles, and phenols. Sulfur may also be released when catabolism of sulfur-­containing amino acids occurs. These unique metabolites underlie many of the effects of proteolytic fermentation on host health outcomes.

3.4.1  Ammonia During microbial deamination of amino acids, ammonia is inherently produced. This ammonia can subsequently enter the host urea cycle for excretion, however excessive quantities may be detrimental to the gastrointestinal epithelium.19 Microbes may also utilize some of the ammonia available in the gut lumen as a nitrogen source for protein synthesis.6 In particular, if abundant fermentable carbohydrate is available, the conditions of the

54

Chapter 3

gut environment may favor incorporation into microbial cells and help to decrease luminal ammonia concentrations.35 However, prolonged exposure to high levels of ammonia can be detrimental to colonocyte health and may contribute to increased colorectal cancer risk.36,37

3.4.2  Amines Amines are produced when amino acids are decarboxylated by one of many different microbial decarboxylases, but may also be produced by colonocytes via ornithine decarboxylase.8 Key amines produced by microbes include spermine, spermidine, histamine, cadaverine, and putrescine. Biogenic amines play important roles in shaping host physiological responses due to their importance to cell physiology and immune signaling.23,38,39 Polyamines are important for microbial physiology and are utilized in RNA, cell membrane, and peptidoglycan synthesis. Amines are also produced during times of physiological stress for protection against reactive oxygen species.40 In the host, these amines are highly bioactive, affecting cell proliferation, DNA and protein integrity, and ion transporters.30,41,42 However, more work is needed to understand when these molecules are beneficial, and under what conditions toxicity occurs.

3.4.3  Branched Chain Fatty Acids Branched chain fatty acids are produced via microbial deamination of branched chain amino acids and are therefore considered good markers of proteolytic fermentation. When protein in the diet is increased, a corresponding increase in branched chain fatty acid production is seen within 24 hours,2,43 indicating a rapid adaptability of the gut microbiota to ferment additional dietary protein. Branched-­chain fatty acids may be oxidized by colonocytes when butyrate is not available,19 however, to date they are not implicated in changes in cell proliferation or barrier function.

3.4.4  Phenols and Indoles Phenols and indoles are derived from microbial fermentation of the aromatic amino acids tyrosine and tryptophan respectively. Phenol can be absorbed and transported to the liver where it is detoxified and conjugated for excretion,8 an important process to prevent detrimental effects on gastrointestinal barrier function.38 Indole may also be converted to indoxyl sulphate in the liver and excreted in urine,44 however, it may be converted to many indole derivatives by the microbiota, many of which have important functions in the gut–brain axis, as well as promoting tight-­junction formation within the gastrointestinal tract.23,45,46 The metabolism of aromatic amino acids generates many complex molecules and derivatives, many of which are aryl hydrocarbon receptor (AhR) ligands, necessitating special

The Microbiome and Amino Acid Metabolism

55

consideration of the roles of aromatic amino acid metabolites in shaping host response.

3.5  Fermentation of Aromatic Amino Acids Fermentation of aromatic amino acids is of particular interest in the context of metabolism and health due to their complex structure which can generate a more diverse set of metabolic end products when fermented (Figure 3.1).23 Furthermore, many of these end products exert important effects on host physiology and metabolism, in particular through effects on aryl hydrocarbon receptors.

3.5.1  Tryptophan In the gastrointestinal tract, free tryptophan is metabolized in one of three pathways:44,47    1. Microbial metabolism, which generates amines and bioactive molecules. 2. Immune and epithelial cell metabolism via the kynurenine pathway. 3. Enterochromaffin cell metabolism via the serotonin production pathway.    Microbial fermentation of tryptophan via the enzyme tryptophanase can release indole as well as indole-­containing compounds and derivatives.44 Some of these indole-containing compounds are now well characterized as AhR ligands including indole-­3-­aldehyde, indole-­3-­acetic acid, indole-­3-­propionic acid, indole-­3-­acetaldehyde, and indoleacrylic acid (Figure 3.2).45 These AhR ligands play an important role in immune competency in the gastrointestinal tract, helping to maintain barrier function and cell turnover, as well as through actions on many immune cell types [T helper 17 (Th17), innate lymphoid cells, macrophages, dendritic cells, neutrophils].44 Indole promotes tight junction formation, and acts

Figure 3.1  The  complex structure of aromatic amino acids allows for many diverse compounds to be made via amino acid catabolism. Shown left to right are tryptophan, tyrosine, and phenylalanine. The cyclic structures of many aromatic amino acid metabolites allow them to function as AhR ligands. This figure was created with BioRender.com

56

Chapter 3

Figure 3.2  Microbial  metabolites of tryptophan with important effects on host

health. AhR ligands are denoted with a star. Microbial species known to produce each metabolite are denoted using the remaining symbols: Lactobacilli – triangles, E. coli – pentagons, Clostridium – squares, Bacteroides – hexagons. This figure was created with BioRender.com

on L-­cells to promote secretion of glucagon-­like peptide-­1.23,46 Furthermore, while microbes are not an enzymatic contributor to the kynurenine pathway, they stimulate the production of indoleamine 2,3-­dioxygenase 1, the enzyme that is responsible for converting tryptophan to kynurine44 by epithelial cells. Another, rarer, enzymatic capacity in gut microbes is tryptophan decarboxylation, which is thought to be possible in approximately 10% or microbes contained within the gut microbiota.47 Decarboxylation of tryptophan produces the indole-­containing amine tryptamine. Tryptamine functions as a neurotransmitter, binding trace amine-­associated receptors and potentiating a response to serotonin.48 In the gastrointestinal tract, tryptamine induces serotonin release from enterochromaffin cells, thereby increasing gut motility.47 Due to the diversity of end products, many different microbial species and enzymes are involved in tryptophan catabolism. Clostridium sporogenes, Ruminococcus gnavus, and Lactobacillus bulgaricus are some of the few microbial species known to decarboxylate tryptophan to tryptamine.47 Of the decarboxylation enzymes, most common are pyridoxal-­5′-­phosphate-­ dependent decarboxylases and pyruvoyl-­dependent decarboxylases.47,49

The Microbiome and Amino Acid Metabolism

57

E. coli and Lactobacilli both possess tryptophanase, the enzyme required for indole production, which is only found in microbes.23,50 While indole is an important signaling molecule in host physiology, its production may also affect other microbes in the community.51 Indole can affect antibiotic resistance responses, sporulation and biofilm formation. It can also inhibit quorum sensing and modulate virulence in non-­indole producing pathogens.51 Peptostreptococcus spp. are also significant contributors to tryptophan disappearance and generation of indoleacrylic acid.52,53

3.5.2  Tyrosine Unlike tryptophan fermentation, which occurs through diverse pathways, microbial tyrosine catabolism has been shown to result in only a few metabolites. Only one pathway, for the production of phenol, is exclusively microbial.23 Phenol production is favored at neutral pH, and when free amino acids are available.33 However, microbes also contribute to shared host–microbiota pathways which generate tyramine, levodopa (l-­DOPA), and p-­cresol.23,33 l-­DOPA functions as a neurotransmitter and is an important precursor for the production of dopamine, norepinephrine, and epinephrine; it is produced by bacterial tyrosinases, such as those found in E. coli.33 Tyramine is produced via tyrosine decarboxylase and aromatic-­l-­amino-­acid decarboxylase by species such as lactic acid bacteria.33,54 Bacterial strains that produce phenols typically possess the enzymatic capacity to produce either phenol or p-­cresol, with only a small subset in the gut that can produce both.55 Phenol producing bacteria include Clostridium spp., Fusobacterium spp., Klebsiella pneumoniae, Citrobacter spp., and Morganella morganii, using the enzymes tyrosine phenol-­lyase, tyrosine aminotransferase, and 4-­hydroxybenzoate decarboxylase.55 Tyrosine phenol-­lyase produces phenol from tyrosine in one step while tyrosine aminotransferase and 4-­hydroxybenzoate decarboxylase are part of a more complex metabolic pathway resulting in phenol production. Bacteria of note producing high levels of p-­cresol include Blautia hyrogenotrophica and Clostridium difficile.55 C. difficile can generate p-­cresol in a single step using tyrosine lyase, while most others use a longer pathway including deamination by tyrosine aminotransferase, followed by dehydrogenase and decarboxylase activity, which results in p-­cresol production.55 Bacteria shown to produce both p-­cresol and phenol include Clostridium spp., Veillonella parvula, Anaerostipes hadrus, and Bacteroides spp.55 While the mechanism is yet to be demonstrated, colonization of mice with Parasuterella reduces concentrations of p-­cresol in the gut lumen.56 This demonstrates the complexity of interactions of core constituents of the microbiota in preventing the production of detrimental metabolites, such as p-­cresol. Both phenol and p-­cresol are metabolites that have been shown to have detrimental effects on host health. In cell culture models, increasing doses of phenol alter tight-­junction formation, as measured by fluorescein

Chapter 3

58 24

isothiocyanate–dextran flux, in a dose dependent manner. p-­Cresol impairs mitochondrial function, resulting in genotoxicity via increased superoxide production and decreased cell proliferation due to lower ATP production, both of which contribute to decreased gut barrier function.57 Due to the damaging effects of phenol and p-­cresol on gut epithelial cells and barrier function, these pathways and the conditions that favor them are of particular interest for mitigating the damaging effects of some proteolytic metabolites.

3.5.3  Phenylalanine Less is known about the physiological effects of fermentation of phenylalanine than those of the other two aromatic amino acids. There are 12 pathways by which microbes can utilize phenylalanine, producing phenylethylamine, catechol, cinnamic acid, phenylacetic acid, and many other metabolites, such as 4-­hydroxybenzoate, that are intermediates for additional pathways.23,28,58 Phenylethylamine is produced by aromatic-­l-­amino-­acid decarboxylase, is typically found in mammalian tissues in low concentrations, and may bind trace amine-­associated receptors along with tyramine and tryptamine.33,48 It is considered to be a neuroactive metabolite, though little has been established about its function. Shikimate and salicylate generated from phenylalanine degradation have also both been demonstrated to bind AhRs,23 representing an unexplored pathway for AhR activation due to proteolytic fermentation. Cinnamic acid is produced by phenylalanine ammonia lyase, generating ammonia as a byproduct.59 Cinnamic acid may be converted to phenylpropionic acid and subsequently metabolized by other microbes.29,60 The biological role of these compounds is less understood, probably in part due to the limited catabolism of phenylalanine by gut microbes.34

3.6  Proteolytic Fermentation and Health 3.6.1  Proteolytic Metabolites and the Gut–Brain Axis The role of the microbiota in influencing neurophysiology and behavior has garnered increased attention in recent years.61,62 Alterations to the gut microbiota have been observed in many neurological and neuropsychiatric conditions including multiple sclerosis (MS), Parkinson's disease, Alzheimer's disease, depression, anxiety, and autism.63 Proteolytic fermentation is important in these conditions as it can generate many metabolites that are known to be neuroactive. The decarboxylation of glutamate, tryptophan, histidine, tyrosine, and phenylalanine all generate compounds known to exert functions on the central nervous system.33 While much interest has been paid to the role of microbes in changing gamma-­aminobutyric acid mediated signaling through the vagus nerve,64 other metabolites, such as indole derivatives, and serotonin, are becoming increasingly understood as important signaling molecules produced during proteolytic fermentation.

The Microbiome and Amino Acid Metabolism

59

Serotonin is an interesting molecule in the gut–brain axis because of its important functions at both ends of this signaling system, modulating many processes in the brain and affecting motility in the gastrointestinal tract.65 Serotonin is synthesized by the host from dietary tryptophan, primarily in the enterochromaffin cells within the gastrointestinal tract.66 Microbes also contribute to the systemic serotonin pool by changing the availability of dietary tryptophan, as well as through their own production of serotonin.65 Microbial production of serotonin appears to be important in affecting colonic concentrations of serotonin, which may, in turn, exert wider host effects, though the role of microbial serotonin outside of the gut is not yet clear.66 Indole derivatives are implicated in neuropsychiatric conditions and MS. Indole overproduction models show decreases in motor activity and behavioral indicators of vagus nerve activation.46 Colonization of germ-­free rats with indole-­producing E. coli resulted in increased anxiety-­like behavior and helplessness during behavioral testing, which was decreased when rats were colonized with E. coli which had had their tryptophanase activity knocked-­ out.46 Indole and its derivatives, indole-­3-­propionic acid, indole-­3-­aldehyde, indole-­3-­carbinol, and indole-­3-­sulfate, are all able to attenuate the severity of experimental autoimmune encephalomyelitis (EAE), a mouse model of MS, through their AhR binding capacities (Figure 3.3).63,67–69 Known functions of these metabolites include promoting regulatory T cells (T-­regs) and inhibiting Th17 cells.69 These results are paired with indoxyl-­3-­sulfate's role in reducing inflammatory gene expression in astrocytes via AhR binding, as well as decreased levels of tryptophan-­derived AhR ligands in the serum of MS patients,67 to indicate that indole derivatives may be important in MS disease etiology. The connection between gastrointestinal disorders and the gut–brain axis is being established with increasing strength; many of the conditions discussed above are accompanied by changes in gastrointestinal motility or co-­morbidities such as inflammatory bowel disease or irritable bowel syndrome.63,70 These changes to nutrient flow in the digestive tract may favor increased proteolytic fermentation, and the products may further exacerbate both gastrointestinal symptoms and disease progression of neurological conditions. Due to this relationship, it is important to consider the role of these metabolites in the gut–brain axis and gastrointestinal tract concurrently.

3.6.2  I rritable Bowel Syndrome (IBS) and Inflammatory Bowel Disease (IBD) Generally, proteolytic fermentation products are thought to play a role in the effect of high protein diets in increasing colitis severity.71 This effect may result from increased abundance of Clostridium spp., and immunostimulatory effects on gut epithelial tissue.72 High protein diets have been shown to cause a microbiota dependent increase in interleukin 6 (IL-­6) and tumor necrosis factor alpha (TNFα) expression alongside an increase in gut

60

Chapter 3

Figure 3.3  Aryl  hydrocarbon receptor ligands produced by the gut microbiota

have many complex effects on host physiology locally and systemically. IL-­10R, interleukin 10 receptor; IL-­22, interleukin 22. This figure was created with Biorender.com.

permeability and dextran sodium sulfate (DSS) induced colitis severity.71 Similar pro-­inflammatory effects of high protein can also be seen when microbial composition is not different,73 indicating that a functional change in microbial metabolism underlies this effect rather than a direct effect of community structure. The production of ammonia from proteolytic fermentation may also contribute to inflammation through decreased butyrate transporter expression and alterations in colonocyte metabolism away from butyrate oxidation, favoring glycolysis instead.22,37 Tryptamine produced from decarboxylation of tryptophan alters colonic ion secretion, which may, in turn, affect gut motility and IBS pathogenesis.47 While all of these effects indicate a detrimental role of proteolytic fermentation in conditions such as IBS and IBD, this effect may not be true for all tryptophan metabolites.

The Microbiome and Amino Acid Metabolism

61

The detrimental effects of proteolytic metabolites described above are not seen for indole and indole derivatives. Indole itself decreases inflammation of the intestinal epithelium and increases tight-­junction protein expression, thereby improving barrier function.23 Indole derivatives may decrease IBD severity through their function as AhR ligands which induce IL-­22 secretion by innate lymphoid cells.68 IBD patients with increased genetic susceptibility and decreased IL-­22 secretion also show fewer tryptophan-­derived AhR ligands produced by their gut microbiota.74 In mouse models of this particular genetic change, a microbiota-­dependent and transferrable increase in colitis severity is shown, and can be attenuated by inoculation with tryptophan metabolizing Lactobacilli.74 Indole derivatives also induce IL-­10R expression in the intestinal epithelium, which is important in barrier function.45 In particular, indole-­3-­propionic acid is decreased in both murine DSS colitis, and human ulcerative colitis patients, indicating its importance in inducing IL-­10R expression.45 These conflicting roles of different tryptophan metabolites demonstrate the importance of both dietary and microbial context in shaping host outcomes. While tryptophan decarboxylases are found in only a small proportion of gut microbes, the enzymes to produce indole or indole derivatives are common in many Lactobacilli.47,74 These differences in enzymatic capacities may be an interesting avenue for microbial modulation of IBD and IBS symptoms in the future.

3.6.3  Colorectal Cancer The microbiota of colorectal cancer patients has been shown to be enriched in protein fermenting species, such as Fusobacterium spp., as well as in putrescine and histidine pathways.75 This enrichment of Fusobacterium nucleatum in particular in colorectal carcinoma (CRC) can result in adhesion to colonocytes and generate a pro-­inflammatory host response.76,77 F. nucleatum produces hydrogen sulfide and ammonia through cysteine catabolism and production of butyrate respectively, both of which may contribute to the tissue damage generated.78,79 Sustained exposure of colonocytes to free ammonia results in cytotoxicity, contributing to CRC development; this is especially true in the absence of adequate quantities of short-­chain fatty acids.36 These results are even more compelling when considered with the enrichment of proteolytic fermenting bacteria in later stages of CRC compared with early stages.75 Phenol and p-­cresol are also likely contributors, as demonstrated by the effects of high casein diets in generating increased phenol and p-­cresol as well as increased colonic DNA damage, an effect that could be mitigated by decreasing the production of both of these metabolites through resistant starch supplementation.80 p-­Cresol in particular can alter the cell cycle, decreasing colonocyte proliferation; this may be due to its effect on colonocyte oxidative metabolism and ATP production.5,80 It is probably due to these effects that p-­cresol generated in culture models of protein fermentation is the best predictor of fermentation supernatant genotoxicity against

Chapter 3

62 81

colonocytes. These effects of p-­cresol must also be considered systemically due to its role in conditions such as chronic kidney disease and associated atherosclerosis.82 The role of dietary polyamines in colorectal cancer risk is more conflicting. Results from some recent studies have indicated a protective effect of dietary polyamines against CRC development,83 while others indicate a role of putrescine, spermidine and spermine, which are derived from arginine decarboxylation, in APC-­dependent colorectal cancer.41 While polyamines are required for many diverse physiological processes, they demonstrate toxic effects when introduced at high concentrations as they disrupt proteins and nucleic acids within cells.84

3.6.4  Metabolic Syndrome When proteolytic fermentation metabolites pass out of the gastrointestinal tract and travel to the liver and peripheral tissues, they can also have an effect on metabolic syndrome.33 Metatrancriptomics conducted on the gut microbiota of twins has shown increased metabolism of amino acids by the microbiota of obese individuals, alongside serum alterations in amino acid profile compared with their corresponding lean twins.85 This altered serum amino acid profile appears to be driven by the microbiota, in particular Clostridium hathewayi.85 In many of the studies examining microbial metabolism and obesity or metabolic disease, alterations in branched chain amino acids (BCAAs) seem to be a shared characteristic.85–87 In type-­2 diabetes mellitus, an increase in circulating branched-­chain and aromatic amino acids is beginning to be seen as an early biomarker of risk.87 Likewise elevated plasma levels of BCAAs, tryptophan, lysine, and glutamate are strongly correlated with non-­alcoholic fatty liver disease in obese adolescents, while baseline plasma valine served as a predictive biomarker of liver fat accumulation in the two year follow up.86 When considered alongside the high proportion of branched-­chain amino acids found in microbial protein, this hallmark change in amino acid profile indicates that changes in intestinal permeability or microbial amino acid synthesis may result from dysbiosis seen in type 2 diabetes and metabolic disease. This is supported by the contribution of microbial produced amino acids to the plasma amino acid pool of adults, even when diets are not deficient in protein.13

3.7  Conclusions The fate of dietary protein in the gut and its overall effect on host amino acid balance is determined by the collaborative effects of host and microbial metabolism.23 However, more work is needed to model metabolic interactions in vivo due to the dynamic nature of nutrient flow and the complexity of the colonized gut. Changes in proteolytic fermentation in response to differing levels and types of fermentable carbohydrate demonstrate

The Microbiome and Amino Acid Metabolism

63

the importance of proteolytic fermentation in diet–microbe–host interactions.3,88 These interactions may contribute to the development of diseases such a colorectal cancer and metabolic disease, as well as neurological and neuropsychiatric conditions. Further work in elucidating the role of proteolytic fermentation in these disease states using mass spectrometry and defined microbial communities will help to further develop our understanding of the contribution of proteolytic metabolites, such as amines, indoles, and phenols, to host health.

References 1. F. Bäckhed, R. E. Ley, J. L. Sonnenburg, D. A. Peterson and J. I. Gordon, Science, 2005, 307, 1915–1920. 2. L. A. David, C. F. Maurice, R. N. Carmody, D. B. Gootenberg, J. E. Button, B. E. Wolfe, A. V. Ling, A. S. Devlin, Y. Varma, M. A. Fischbach, S. B. Biddinger, R. J. Dutton and P. J. Turnbaugh, Nature, 2014, 505, 559–563. 3. K. Korpela, Annu. Rev. Food Sci. Technol., 2018, 9, 4.1–4.20. 4. R. Pieper, W. Vahjen and J. Zentek, Anim. Prod. Sci., 2015, 55, 1367–1375. 5. M. Andriamihaja, A. Lan, M. Beaumont, M. Audebert, X. Wong, K. Yamada, Y. Yin, D. Tomé, C. Carrasco-­Pozo, M. Gotteland, X. Kong and F. Blachier, Free Radical Biol. Med., 2015, 85, 219–227. 6. G. Gottschalk, Bacterial Metabolism, Springer, New York, 1986. 7. E. Neis, C. Dejong and S. Rensen, Nutrients, 2015, 7, 2930–2946. 8. R. Pieper, C. Villodre Tudela, M. Taciak, J. Bindelle, J. F. Pérez and J. Zentek, Anim. Health Res. Rev., 2016, 17, 137–147. 9. E. . Smith and G. . Macfarlane, FEMS Microbiol. Ecol., 1998, 25, 355–368. 10. Z.-­L. Dai, J. Zhang, G. Wu and W.-­Y. Zhu, Amino Acids, 2010, 39, 1201–1215. 11. S. A. W. Gibson, C. McFarlan, S. Hay and G. T. Macfarlane, Appl. Environ. Microbiol., 1989, 55, 679–683. 12. P. Fan, P. Liu, P. Song, X. Chen and X. Ma, Sci. Rep., 2017, 7, 43412. 13. C. C. Metges, J. Nutr., 2000, 130, 1857S–1864S. 14. A. J. Darragh, P. D. Cranwell and P. J. Moughan, Br. J. Nutr., 1994, 71, 739–752. 15. G. Macfarlane, FEMS Microbiol. Lett., 1986, 38, 19–24. 16. T. Ju, Y. Shoblak, Y. Gao, K. Yang, J. Fouhse, B. B. Finlay, Y. W. So, P. Stothard and B. P. Willing, Appl. Environ. Microbiol., 2017, 83, e01107-­17. 17. A. Chacko and J. H. Cummings, Gut, 1988, 29, 809–815. 18. E. A. Smith and G. T. Macfarlane, J. Appl. Bacteriol., 1996, 81, 288–302. 19. F. Blachier, F. Mariotti, J. F. Huneau and D. Tomé, Amino Acids, 2007, 33, 547–562. 20. N. van der Wielen, P. J. Moughan and M. Mensink, J. Nutr., 2017, 147, 1493–1498. 21. H. W. Doelle, in Bacterial Metabolism, Elsevier, 1969, pp. 402–422. 22. B. Darcy-­Vrillon, C. Cherbuy, M. T. Morel, M. Durand and P. H. Duée, Mol. Cell. Biochem., 1996, 156, 145–151.

64

Chapter 3

23. G. V Sridharan, K. Choi, C. Klemashevich, C. Wu, D. Prabakaran, L. Bin Pan, S. Steinmeyer, C. Mueller, M. Yousofshahi, R. C. Alaniz, K. Lee and A. Jayaraman, Nat. Commun., 2014, 5, 5492. 24. I. C. McCall, A. Betanzos, D. A. Weber, P. Nava, G. W. Miller and C. A. Parkos, Toxicol. Appl. Pharmacol., 2009, 241, 61–70. 25. N. Fonknechten, S. Chaussonnerie, S. Tricot, A. Lajus, J. R. Andreesen, N. Perchat, E. Pelletier, M. Gouyvenoux, V. Barbe, M. Salanoubat, D. Le Paslier, J. Weissenbach, G. N. Cohen and A. Kreimeyer, BMC Genomics, 2010, 11, 555. 26. M. Neumann-­Schaal, D. Jahn and K. Schmidt-­Hohagen, Front. Microbiol., 2019, 10, 219. 27. D. White, J. T. Drummond and C. Fuqua, The Physiology and Biochemistry of Prokaryotes, Oxford University Press, 2011. 28. K. Oliphant and E. Allen-­Vercoe, Microbiome, 2019, 7, 91. 29. H. A. Barker, Annu. Rev. Biochem., 1981, 50, 23–40. 30. P. Fan, L. Li, A. Rezaei, S. Eslamfam, D. Che and X. Ma, Curr. Protein Pept. Sci., 2015, 16, 646–654. 31. A.-­M. Davila, F. Blachier, M. Gotteland, M. Andriamihaja, P.-­H. Benetti, Y. Sanz and D. Tomé, Pharmacol. Res., 2013, 69, 114–126. 32. M. L. Britz and R. G. Wilkinson, Can. J. Microbiol., 1982, 28, 291–300. 33. K. J. Portune, M. Beaumont, A.-­M. Davila, D. Tomé, F. Blachier and Y. Sanz, Trends Food Sci. Technol., 2016, 57, 213–232. 34. M. Steglich, J. D. Hofmann, J. Helmecke, J. Sikorski, C. Spröer, T. Riedel, B. Bunk, J. Overmann, M. Neumann-­Schaal and U. Nübel, Front. Microbiol., 2018, 9, 901. 35. J. H. Cummings and G. T. Macfarlane, J. Appl. Bacteriol., 1991, 70, 443–459. 36. K. Fung, C. Ooi, M. Zucker, T. Lockett, D. Williams, L. Cosgrove and D. Topping, Int. J. Mol. Sci., 2013, 14, 13525–13541. 37. C. Villodre Tudela, C. Boudry, F. Stumpff, J. R. Aschenbach, W. Vahjen, J. Zentek and R. Pieper, Br. J. Nutr., 2015, 113, 610–617. 38. R. Hughes, M. J. Kurth, V. McGilligan, H. McGlynn and I. Rowland, Nutr. Cancer, 2008, 60, 259–266. 39. N. E. Flynn, J. G. Bird and A. S. Guthrie, Amino Acids, 2009, 37, 123–129. 40. P. Shah and E. Swiatlo, Mol. Microbiol., 2008, 68, 4–16. 41. E. W. Gerner, Biochem. Soc. Trans., 2007, 35, 322–325. 42. S. Bardócz, T. J. Duguid, D. S. Brown, G. Grant, A. Pusztai, A. White and A. Ralph, Br. J. Nutr., 1995, 73, 819. 43. M. Aguirre, A. Eck, M. E. Koenen, P. H. M. Savelkoul, A. E. Budding and K. Venema, Res. Microbiol., 2016, 167, 114–125. 44. A. Agus, J. Planchais and H. Sokol, Cell Host Microbe, 2018, 23, 716–724. 45. E. E. Alexeev, J. M. Lanis, D. J. Kao, E. L. Campbell, C. J. Kelly, K. D. Battista, M. E. Gerich, B. R. Jenkins, S. T. Walk, D. J. Kominsky and S. P. Colgan, Am. J. Pathol., 2018, 188, 1183–1194. 46. M. Jaglin, M. Rhimi, C. Philippe, N. Pons, A. Bruneau, B. Goustard, V. Daugé, E. Maguin, L. Naudon and S. Rabot, Front. Neurosci., 2018, 12, 216.

The Microbiome and Amino Acid Metabolism

65

47. B. B. Williams, A. H. Van Benschoten, P. Cimermancic, M. S. Donia, M. Zimmermann, M. Taketani, A. Ishihara, P. C. Kashyap, J. S. Fraser and M. A. Fischbach, Cell Host Microbe, 2014, 16, 495–503. 48. R. Zucchi, G. Chiellini, T. S. Scanlan and D. K. Grandy, Br. J. Pharmacol., 2006, 149, 967–978. 49. R. A. John, Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol., 1995, 1248, 81–96. 50. T. D. Hubbard, I. A. Murray, W. H. Bisson, T. S. Lahoti, K. Gowda, S. G. Amin, A. D. Patterson and G. H. Perdew, Sci. Rep., 2015, 5, 12689. 51. J.-­H. Lee, T. K. Wood and J. Lee, Trends Microbiol., 2015, 23, 707–718. 52. M. Wlodarska, C. Luo, R. Kolde, E. d'Hennezel, J. W. Annand, C. E. Heim, P. Krastel, E. K. Schmitt, A. S. Omar, E. A. Creasey, A. L. Garner, S. Mohammadi, D. J. O'Connell, S. Abubucker, T. D. Arthur, E. A. Franzosa, C. Huttenhower, L. O. Murphy, H. J. Haiser, H. Vlamakis, J. A. Porter and R. J. Xavier, Cell Host Microbe, 2017, 3–4. 53. R. Lin, W. Liu, M. Piao and H. Zhu, Amino Acids, 2017, 49, 2083–2090. 54. A. Marcobal, B. De Las Rivas, J. M. Landete, L. Tabera and R. Muñoz, Crit. Rev. Food Sci. Nutr., 2012, 52, 448–467. 55. Y. Saito, T. Sato, K. Nomoto and H. Tsuji, FEMS Microbiol. Ecol., 2018, fiy125. 56. T. Ju, J. Y. Kong, P. Stothard and B. P. Willing, ISME J., 2019, 13, 1520–1534. 57. M. Andriamihaja, A. Lan, M. Beaumont, M. Audebert, X. Wong, K. Yamada, Y. Yin, D. Tomé, C. Carrasco-­Pozo, M. Gotteland, X. Kong and F. Blachier, Free Radical Biol. Med., 2015, 85, 219–227. 58. T. A. Clayton, FEBS Lett., 2012, 586, 956–961. 59. E. Díaz, A. Ferrández, M. A. Prieto and J. L. García, Microbiol. Mol. Biol. Rev., 2001, 65, 523–569, table of contents. 60. R. Burlingame and P. J. Chapman, J. Bacteriol., 1983, 155, 113–121. 61. J. F. Cryan and S. M. O'Mahony, Neurogastroenterol. Motil., 2011, 23, 187–192. 62. T. R. Sampson and S. K. Mazmanian, Cell Host Microbe, 2015, 17, 565–576. 63. T. C. Fung, C. A. Olson and E. Y. Hsiao, Nat. Neurosci., 2017, 20,  145–155. 64. J. A. Bravo, P. Forsythe, M. V. Chew, E. Escaravage, H. M. Savignac, T. G. Dinan, J. Bienenstock and J. F. Cryan, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 16050–16055. 65. S. M. O'Mahony, G. Clarke, Y. E. Borre, T. G. Dinan and J. F. Cryan, Behav. Brain Res., 2015, 277, 32–48. 66. J. M. Yano, K. Yu, G. P. Donaldson, G. G. Shastri, P. Ann, L. Ma, C. R. Nagler, R. F. Ismagilov, S. K. Mazmanian and E. Y. Hsiao, Cell, 2015, 161, 264–276. 67. V. Rothhammer, I. D. Mascanfroni, L. Bunse, M. C. Takenaka, J. E. Kenison, L. Mayo, C. C. Chao, B. Patel, R. Yan, M. Blain, J. I. Alvarez, H. Kébir, N. Anandasabapathy, G. Izquierdo, S. Jung, N. Obholzer, N. Pochet, C. B. Clish, M. Prinz, A. Prat, J. Antel and F. J. Quintana, Nat. Med., 2016, 22, 586–597.

66

Chapter 3

68. T. Zelante, R. G. Iannitti, C. Cunha, A. De Luca, G. Giovannini, G. Pieraccini, R. Zecchi, C. D'Angelo, C. Massi-­Benedetti, F. Fallarino, A. Carvalho, P. Puccetti and L. Romani, Immunity, 2013, 39, 372–385. 69. M. Rouse, N. P. Singh, P. S. Nagarkatti and M. Nagarkatti, Br. J. Pharmacol., 2013, 169, 1305–1321. 70. A. V. Golubeva, S. A. Joyce, G. Moloney, A. Burokas, E. Sherwin, S. Arboleya, I. Flynn, D. Khochanskiy, A. Moya-­Pérez, V. Peterson, K. Rea, K. Murphy, O. Makarova, S. Buravkov, N. P. Hyland, C. Stanton, G. Clarke, C. G. M. Gahan, T. G. Dinan and J. F. Cryan, EBioMedicine, 2017, 24, 166–178. 71. S. R. Llewellyn, G. J. Britton, E. J. Contijoch, O. H. Vennaro, A. Mortha, J.-­F. Colombel, A. Grinspan, J. C. Clemente, M. Merad and J. J. Faith, Gastroenterology, 2018, 154, 1037–1046.e2. 72. R. Pieper, S. Kröger, J. F. Richter, J. Wang, L. Martin, J. Bindelle, J. K. Htoo, D. von Smolinski, W. Vahjen, J. Zentek and A. G. Van Kessel, J. Nutr., 2012, 142, 661–667. 73. J. Zentek, R. Pieper, J. Wang, S. Kröger, A. G. Van Kessel, D. von Smolinski, J. Bindelle, J. K. Htoo, L. Martin, J. F. Richter and W. Vahjen, J. Nutr., 2012, 142, 661–667. 74. B. Lamas, M. L. Richard, V. Leducq, H.-­P. Pham, M.-­L. Michel, G. Da Costa, C. Bridonneau, S. Jegou, T. W. Hoffmann, J. M. Natividad, L. Brot, S. Taleb, A. Couturier-­Maillard, I. Nion-­Larmurier, F. Merabtene, P. Seksik, A. Bourrier, J. Cosnes, B. Ryffel, L. Beaugerie, J.-­M. Launay, P. Langella, R. J. Xavier and H. Sokol, Nat. Med., 2016, 22, 598–605. 75. H. Kaur, C. Das and S. S. Mande, Front. Microbiol., 2017, 8, 2166. 76. A. D. Kostic, D. Gevers, C. S. Pedamallu, M. Michaud, F. Duke, A. M. Earl, A. I. Ojesina, J. Jung, A. J. Bass, J. Tabernero, J. Baselga, C. Liu, R. A. Shivdasani, S. Ogino, B. W. Birren, C. Huttenhower, W. S. Garrett and M. Meyerson, Genome Res., 2012, 22, 292–298. 77. E. Allen-­Vercoe and C. Jobin, Immunol. Lett., 2014, 162, 54–61. 78. F. Carbonero, A. C. Benefiel, A. H. Alizadeh-­Ghamsari and H. R. Gaskins, Front. Physiol., 2012, 3, 448. 79. S. Anand, H. Kaur and S. S. Mande, Front. Microbiol., 2016, 1945. 80. S. Toden, A. R. Bird, D. L. Topping and M. A. Conlon, Nutr. Cancer, 2005, 51, 45–51. 81. E. A. Al Hinai, P. Kullamethee, I. R. Rowland, J. Swann, G. E. Walton and D. M. Commane, Gut Microbes, 2019, 10, 398–411. 82. M.-­C. Chang, H.-­H. Chang, C.-­P. Chan, S.-­Y. Yeung, H.-­C. Hsien, B.-­R. Lin, C.-­Y. Yeh, W.-­Y. Tseng, S.-­K. Tseng and J.-­H. Jeng, PLoS One, 2014, 9, e114446. 83. A. J. Vargas, E. L. Ashbeck, B. C. Wertheim, R. B. Wallace, M. L. Neuhouser, C. A. Thomson and P. A. Thompson, Am. J. Clin. Nutr., 2015, 102, 411–419. 84. A. E. Pegg, Chem. Res. Toxicol., 2013, 26, 1782–1800. 85. V. K. Ridaura, J. J. Faith, F. E. Rey, J. Cheng, A. E. Duncan, A. L. Kau, N. W. Griffin, V. Lombard, B. Henrissat, J. R. Bain, M. J. Muehlbauer, O. Ilkayeva, C. F. Semenkovich, K. Funai, D. K. Hayashi, B. J. Lyle, M. C. Martini, L. K. Ursell, J. C. Clemente, W. Van Treuren and W. A. Walters, Science (80-. ), 2013, 341, 1241214.

The Microbiome and Amino Acid Metabolism

67

86. M. Goffredo, N. Santoro, D. Tricò, C. Giannini, E. D'Adamo, H. Zhao, G. Peng, X. Yu, T. Lam, B. Pierpont, S. Caprio and R. Herzog, Nutrients, 2017, 9, 642. 87. T. J. Wang, M. G. Larson, R. S. Vasan, S. Cheng, E. P. Rhee, E. McCabe, G. D. Lewis, C. S. Fox, P. F. Jacques, C. Fernandez, C. J. O'Donnell, S. A. Carr, V. K. Mootha, J. C. Florez, A. Souza, O. Melander, C. B. Clish and R. E. Gerszten, Nat. Med., 2011, 17, 448–453. 88. A. Salonen, L. Lahti, J. Salojärvi, G. Holtrop, K. Korpela, S. H. Duncan, P. Date, F. Farquharson, A. M. Johnstone, G. E. Lobley, P. Louis, H. J. Flint and W. M. de Vos, ISME J., 2014, 8, 2218–2230.

Chapter 4

Fat Absorption, Metabolism, and Global Regulation Nayaab Ranaa, Peymaun Mozaffaria, Danial Asima and Kristina Martinez-­Guryn*a a

Department of Biomedical Sciences, College of Graduate Studies, Midwestern University, Downers Grove, IL, USA *E-­mail: [email protected]

4.1  Introduction The human host consists of millions of its own cells, each tasked with unique structure and function, but also houses trillions of microorganisms including bacteria, yeast, fungi, protozoa and archaea1 that exist in a mutualistic relationship.2 Thus, the body is gifted with an extremely diverse and complex ecological environment. This collection of microorganisms is referred to as the microbiota and concentrates primarily throughout the gastrointestinal tract.3 Gut microbes are key players in aiding various biochemical and physiological functions of the host, including generation of energy, digestion, absorption, immune development, protection against pathogens, and metabolism.4 A balanced microbiota is therefore of valued importance in maintaining health and preventing disease. The composition of the microbiota is directly affected by the environment, diet, genetics, age, drugs and antibiotics, and host health status.5,6 Deleterious alterations in gut microbiota structure and function may result in inflammatory conditions, cardiovascular stress, vascular abnormality, and metabolic disorders in many tissue   Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

68

Fat Absorption, Metabolism, and Global Regulation

69

Figure 4.1  The  composition of the gut microbiota is dramatically affected by diet, particularly Western diets. The gut microbiota plays a role in influencing host health and disease via complex interactions with the gut, and peripheral organs including the liver, adipose tissue, muscle, and heart.

depots.7 (Figure 4.1) Given the important role of gut microbiota in human health and disease, studies published on gut microbiota research have increased 100-­fold over the last 15 years.4 Although gut microbiota research is becoming more and more popular and advancing at a rapid pace, gut microbiota research is by no means new. Dietary effects on the gut microbiota have been documented as early as the end of the 19th century. For example, in 1891, it was documented that a meat-­based vs. a carbohydrate diet altered the gut microbial community in small intestinal contents. A patient, Magdalene Spycher, underwent surgical removal of necrotic tissue from her ileum and cecum. After consuming a diet that predominated in meats and then carbohydrates, her intestinal contents were collected via the resulting fistula and microbes were isolated and cultured. The meat-­based diet yielded differing types of isolated microorganisms compared with the carbohydrate-­based diet. From the meat-­based diet, eight microorganisms were isolated (four bacilli, an oval bacterium, streptococcus, yeast fungi, and mold fungus reminiscent of Oidium lactis). On the carbohydrate diet, a slender bacillus was detected as well as numerous yeast fungi. A total of seven microorganisms were isolated including a streptococcus, a rod-­shaped bacillus, a chain bacillus, two types of diplococcus, a

70

Chapter 4

bacterium resembling Bacterium coli commune, and yeast forms. Mold fungi were not identified after a carbohydrate diet. Altogether it was concluded that diet elicited dramatic differences in the types of bacteria that predominated the small bowel with an overall lack of putrefactive bacteria.8 Still, over 100 years later, we continue to seek an understanding of the effects of diet on the gut microbiota. For example, in 2014, David et al. reported the dramatic and rapid effects of a meat-­based diet on the human gut microbiota compared with a plant-­based diet, but now using advanced next-­generation sequencing technology. This study revealed a strong association between altered microbial gene profiles and dietary intake, including amino acid and central metabolic pathways. The meat-­based diet significantly increased the abundance of putrefactive bacteria Bilophila, Alistipes, and Bacteroides.9 In addition to describing shifts in microbial community structure, important host–microbe interactions have been identified over the past 20–30 years. An early breakthrough for gut microbiology was that of Barry Marshall's discovery that ulcers were triggered by Helicobacter pylori. During the 1980s many doctors attributed ulcers to stress while Barry Marshall proposed the hypothesis that antibiotics would provide an effective treatment. After conducting multiple biopsies, he found that H. pylori is abundant in patients that have ulcers. Due to ethical concerns with his research, Marshall decided to consume a culture of H. pylori and developed an ulcer. Upon examining biopsies from his own gut, H. pylori was indeed present, thus demonstrating a causal effect of bacteria on ulcer development. Seminal discoveries, such as these, changed the perspective on the role of microbes in health and disease. However, given the technological advancements in gut microbiota research (i.e. microfluidics, culturomic techniques, advanced imaging, 3D organ culture, and gnotobiotics, etc.), thousands of microbial species have been identified and their involvement in host pathology has been defined. Still, there is much to be uncovered. In this chapter, the effects of obesity and diet on the gut microbiota are explored followed by a discussion on the mechanisms by which gut microbes regulate local functions of the gut, including nutrient digestion and absorption, and how they affect metabolism globally in peripheral metabolic tissues.

4.2  Obesity and the Gut Microbiota Obesity, a leading risk factor for type 2 diabetes, has been associated with shifts in the human microbiome.10 Obesity is a multifactorial disorder and results from an excess buildup of adipose tissue in subcutaneous or visceral depots.11 This over-­accumulation of adipose tissue is caused by an imbalance in the energy consumption : expenditure ratio. Obesity is influenced by various factors, such as genetic predisposition, sleep disruption, stress, and alterations in gut microbiota structure. It is generally understood that obesity is related to a decrease in bacterial diversity as well as phylum-­level alterations of the microbiota in terms of relative abundance. Dominant phyla of bacteria in the gut that are shared by most humans include Bacteroidetes,

Fat Absorption, Metabolism, and Global Regulation

71 12

Firmicutes, Actinobacteria, Proteobacteria, and Verrucomicrobia. Shifts at the phyla level were associated with specific diseases, such as obesity, or consumption of a high fat diet, which is typically characterized by an increased abundance of Firmicutes and decreased abundance of Bacteroidetes.13 While these shifts are consistently observed, the changes in the gut microbiota and interactions with the host are extremely complex and members within these phyla shift and play specific functional roles that are either beneficial or deleterious to the host. For example, some members of the Firmicutes such as Clostridium cluster XIV are short chain fatty acid (SCFA) producers that yield important metabolic benefits, while other members of the Firmicutes, such as Clostridium difficile, promote colitis. Therefore, shifts at the phyla level help characterize general features of the microbiota in obesity, but do not fully explain intricate host–microbe interactions underlying the disease. Early work has focused on demonstrating an overall link between microbe and host crosstalk using conventionalization or fecal microbiota transplant (FMT) of germ-­free (GF) animals with complex microbial communities collected from animals fed specialized diets, or with a specific disease, or genetic background.14,15 For instance, gut microbiota transplant from high fat diet (HFD) or obese conditions promotes adiposity and/or lipid absorption in naïve recipient GF mice.14,15 As gut microbiota research is advancing, functions of microbial communities are being examined to determine what the microbes are doing and how exactly they interact with host systems. Although commensal microbes are generally beneficial to the host, the gut microbiota may be easily altered by their environment, including the diet and genetics of the host. Consumption of a diet persistently high in fat and sugar promotes an obesogenic microbiota.13 This process may be influenced by host genetics, as murine loss-­of-­function knockout models display altered microbial composition in the gut.16 Development of disease often involves complex interactions between diet, gut microbiota, and host genetics.17 Recent research has demonstrated the pivotal role of the microbiome in obesity as the microbiota and their metabolic pathways affect the metabolism and adiposity of the host.13,14 Obese and lean phenotypes have been shown to be dissimilar in multiple microbial factors: species composition, functional genes, and metabolic activity.18 The gut microbiota has been demonstrated to have a causal role in obesity in both humans and animals.14,19 For example, it was shown that microbiota from humanized mice fed a Western diet increased adiposity in GF recipient mice fed a low fat diet (LFD).14 Interestingly, a human patient receiving FMT for Clostridium difficile infection gained a significant amount of weight and it was later found that the fecal matter was obtained from an overweight donor.19 Conversely, in a human intervention study, it was found that FMT from lean participants decreased complications of metabolic syndrome in overweight individuals.20 These findings highlight the possibilities of targeting the gut microbiota therapeutically to aide in preventing or resolving metabolic disorders.

72

Chapter 4

Additional evidence of altered gut microbiota in obesity is offered from studies involving gastric bypass surgery. For instance, the gut microbiota is significantly altered following Roux-­en-­Y gastric bypass surgery characterized by increased abundance of Verrucomicrobiales and Bacteroidales and decreased Clostridiales.21 In subsequent studies evaluating the effectiveness of biliary diversion, as an alternative to bariatric surgery in which the bile duct is diverted to distal regions of the small intestine, patients exhibited exhibit not only dramatic weight loss but also shifts in gut microbiota structure.22 Various mechanisms have been proposed to explain the interaction between gut microbiota and obesity, including increased liberation of SCFAs (albeit SCFAs also have many positive metabolic benefits), disruption of epithelial barrier and increased circulating lipopolysaccharide (LPS), enhanced lipid digestion and absorption in the gut, and interactions with the liver and muscle.13 These topics will be discussed in more detail below and in the latter sections of this chapter. Briefly, one mechanism linking gut microbes to obesity is bacterial-­derived LPS, that triggers inflammation in the gut and distal metabolic organs. LPS is found in the cell walls of Gram-­negative bacteria. High fat diets are generally associated with having higher levels of LPS due to the disruption of the epithelial barrier of the intestine. Elevated circulating LPS levels are associated with obesity and reduced insulin sensitivity. This represents a commonly reported mechanism linking gut microbiota to low-­grade inflammation associated with obesity.13

4.3  Dietary Modulation of the Gut Microbiota The dietary effect on the gut microbiota is profound and begins at birth. For example, breast-­fed vs. formula-­fed babies exhibit differences in microbiota composition. By three years of age, the gastrointestinal-­microbial system develops to be like that of an adult's stable system. At this point, anywhere from two thirds to nearly three quarters of the microbiota composition settles and remains constant throughout the host's life.23 However, this means that great potential for changes exists in the microbiota (up to 40%) which can be altered by diet, exercise, lifestyle, illnesses/infections, and treatments such as antibiotics. Overall, diet-­mediated effects on the gut microbiota have been a major area of research focus as they may be beneficial or detrimental to host health.

4.3.1  D  iet-­mediated Shifts in Gut Microbiota Community Composition Diet composition and various dietary constituents have been reported to alter the gut microbiota with regards to differing macronutrient ratios (high fat vs. high carbohydrate),9,15 types of dietary fat (saturated vs. unsaturated),17,24 fiber content,25 and polyphenolic foods and compounds.26,27 The features of

Fat Absorption, Metabolism, and Global Regulation

73

an obesity-­associated microbiota are typically similar to what is observed in humans and animals consuming a Western diet, a diet high in fat and simple sugars. Long-­term, such diets are congruent with the development of obesity, type 2 diabetes, and cardiovascular disease.28 As may be expected, consuming large amounts of fats and sugars in the diet alters microbial composition and contributes to gut dysbiosis, wherein the microbiota undergo changes in their structure, function, and species diversity. For example, a Western diet promotes the expansion of members of the phyla Firmicutes and Proteobacteria, along with a reduction in Actinobacteria and Bacteroidetes.15 David et al. showed that an animal-­based diet consisting of meat, eggs, and cheeses significantly altered gut microbial communities in as little as two days compared with a diet rich in plant-­based foods like grains, legumes, fruits, and vegetables.9 Beta diversity (assessment of microbial community composition differences between samples) was significantly reduced in subjects consuming the animal-­based diet compared to baseline samples but not in subjects consuming the plant-based diet. The conclusions of this study imply that high fat and cholesterol diets may have a detrimental effect on the gut microbiota that occurs in a rapid fashion.9 This research group also investigated whether or not diet or genotype was a greater driver of microbial community structure. Carmody et al. conducted a diet study in which five different strains of mice were fed a low-­fat plant-­based diet, or a high-­fat high-­sugar diet. The data overwhelmingly indicated that the difference in diet altered the microbiota more significantly than the genetic differences across the five strains.16 Dietary shifts impose lasting changes in gut microbiota, as it was also observed that switching from a HFD to a LFD does not completely restore the original community structure, indicating imprinting effects of diet. While these results demonstrate the overall effect of diet and host genetics on microbial communities, information related to distinct host– microbe interactions involving diet, host genetics, and obesity are still lacking. High-­fat diets may lead to increased intestinal permeability, an inflamed gastrointestinal (GI) tract, and a rise in circulating levels of LPS. Disorders involving mucosal immune dysregulation are more common in a dysregulated, dysbiotic gut. In addition to a deleterious restructuring of the gut microbiota, gut dysbiosis can be characterized by an increase in the presence and spread of pathogenic bacteria within the GI tract, which, in a host more susceptible to disease, can result in inflammatory diseases. Dietary changes can cause some members of the commensal microbiota to become pathobionts. Pathobionts demonstrate behavior like pathogens in particular environments, considering host susceptibility and context,29 leading to development of celiac disease and inflammatory bowel disease.30 For example, Devkota et al. demonstrated that feeding interleukin 10 negative (IL10−/−) mice, susceptible to colitis, a diet rich in saturated milk fat increased production of taurocholic acid that selectively promoted the growth of the pathobiont Bilophila wadsworthia. The bloom of this pathobiont ultimately exacerbated the severity and penetrance of colitis.17 Another diet–microbe–host response was presented by Brown and Hazen, in which consumption of choline-­ and

74

Chapter 4

carnitine-­based animal products resulted in the production of trimethylamine (TMA) by bacteria in the gut, that was next converted to trimethylamine N-­oxide (TMAO) in the liver, resulting in cardiovascular disease.31 The Western diet is often associated with the reduced consumption of dietary fiber that would otherwise result in generation of SCFAs that are critical for maintaining epithelial integrity of the gut and providing an additional fuel source for the host.32 Dietary fiber also plays a role in modulating microbial composition. The amount of dietary fiber consumed may affect the growth of particular bacterial species, allowing for greater fermentative metabolism, that concurrently compete with pathogenic bacteria. Dietary fiber and its effects on the host system were the focus of one particular study that compared the fecal microbiota of European children (EU) with that of rural African children in Burkina Faso (BF).33 Using 16S rRNA sequencing it was found that there were marked differences in the gut microbiota of each group. The high-­fiber diet group (BF) demonstrated increased abundance of Bacteroides, Prevotella, and Xylanibacter with a notable decrease in Firmicutes. The larger than expected presence of Prevotella and Xylanibacter highlighted that populations with high-­fiber diets have increased ability to hydrolyze cellulose and xylan due to a different set of bacterial genes present. These bacterial species were not present in the opposing EU group ingesting less dietary fiber. An additional finding was that the amount of Enterobacteriaceae bacteria was significantly less in the BF group. This study highlighted the dramatic relationship between dietary fiber and microbial composition and function. Notably, dietary effects may be dependent on the region of the gut being examined. A large body of evidence exists based on samples that can be acquired noninvasively, such as stools. Yet, the influence of HFD was shown to have a greater effect on the small bowel microbiota compared with more distal regions such as the cecum.15,34,35 The regional site of dietary or host-­ driven shifts in the gut microbiota may be important when understanding the resultant host response. Several examples of specific regional localization and host interactions exist for bacteria, including Streptococcus mutans in the mouth,36 H. pylori in the stomach, and C. difficile in the colon.35 Mechanisms regarding individual bacterial species, their unique metabolic activities and outputs, and host physiological effects are still actively being investigated.

4.3.2  Direct Microbial Metabolism of Dietary Components Not only does diet shift microbial populations in the gut, but microbes also directly metabolize dietary constituents or convert them to more bioavailable and bioactive products. Examples include bacterial fermentation of non-­digestible carbohydrates, and also conversion of the fatty acid linoleic acid into conjugated derivatives. Research is beginning to unravel the cornucopia of information regarding gut microbial composition, proliferation, metabolic activity, and function. As the link between metabolic disease and the microbes of the gut continues to grow, the metabolic byproducts of the microbiome have been a large focus of study.37

Fat Absorption, Metabolism, and Global Regulation

75

The gut microbiota assist in diet digestion and yield several vitamins and SCFAs. SCFAs are volatile fatty acids and principally include butyrate, propionate, and acetate.38 These metabolites can reach high levels in the proximal colon and influence intestinal epithelial health. Butyrate especially plays a known role in this by being a beneficial, preferred metabolic substrate for these early cells in the GI tract and maintaining host mucosal health. Thus, the metabolic byproduct of butyrate is deduced to be of critical importance in host disease resistance and protection. Additionally, it is these epithelial cells that allow for a source and location for the initial interface of host– microbe crosstalk, since epithelial cells provide the host with SCFA metabolic products from luminal microbes. Linoleic acids are polyunsaturated fatty acids derived from plant oils that are essential to the human diet as they play a role in the biosynthesis of prostaglandins and cell membranes. Conjugated linoleic acids (CLA) are derivatives of linoleic acid (positional and geometric) such as cis-­9, trans-­11 and trans-­10, cis-­12 CLA often found naturally in meat and dairy products.39 In a study by Rosberg-­Cody et al.,40 it was demonstrated that a metabolically active strain of Bifidobacterium breve could result in alteration of the fatty acid composition in the host. The bacteria were administered orally and formed a cis-­9, trans-­11 conjugated linoleic acid (CLA). The resulting bacteria–CLA interaction also led to increased concentrations of the same CLA class (c9, t11) as well as omega-­3 fatty acids in both the liver and fat tissue. When Bifidobacterium breve expressing the single gene linoleic acid isomerase was utilized, there was four-­fold greater CLA (t10, c12). In this way, the fatty acid composition in host adipocytes and tissue was found to be capable of modulation via a single gene from one microbial species. Microbe-­mediated alterations in the dietary and host lipid profile are significant as it was also shown that CLA decreases tumor necrosis factor (TNF) and interferon (IFN)-­gamma levels. Furthermore, it was found that orally administered microbial species that produce CLA were associated with lower levels of TNF-­alpha and IFN-­gamma, both types of pro-­inflammatory cytokines.41 Other researchers have found that CLA also provides anti-­ carcinogenic effects to the host, especially c9, t11. The isomer t10, c12 functions to decrease host body fat and thus is considered to be a more powerful agent on blood lipid profiles. Both isomers have the less beneficial effect of insulin resistance in human hosts. CLAs have also been demonstrated to have less marked effects in human hosts vs. animal hosts, particularly with regard to body weight and fat.42 In non-­human trials, the consumption of conjugated linoleic acids was demonstrated to decrease body fat. This finding was not so clearly replicated in human studies. When a comprehensive review of the scientific literature was conducted in order to properly substantiate the claims for CLA with regards to various pathologies (cancer, cardiovascular disease, diabetes, obesity, osteoporosis), there was not strong evidence. Conversely, both CLA isomers c9, t11 and t10, c12 were instead found to encourage diabetic profiles in those at risk for diabetes.43 With regards to the differences seen in human and

76

Chapter 4

non-­human test subjects, the investigators considered that the disparities may in fact be due to variations in concentration of the CLA administered. Altogether, CLAs represent an example of important diet–gut microbe–host relationships. While new data continues to emerge, many mechanisms and pathways remain at present elusive.

4.4  L  ocal Effects of Gut Microbes on the Gastrointestinal Tract 4.4.1  Lipid Digestion and Absorption Gut microbes regulate host digestion and absorption in the small intestine, which involves regulation at multiple levels beginning with fatty acid sensing in the mouth, enteroendocrine signaling, increased expression of fatty acid and cholesterol transport proteins, and lipid droplet formation. These interactions have been largely demonstrated through the use of germ free (GF) and conventionalized mice (ConvD; GF mice conventionalized with microbiota). The process of lipid digestion and absorption is complex and involves the coordinated action of differing regions of the gut, requiring specialized enteroendocrine cells and accessory organs such as the pancreas (i.e. lipase production and secretion) and liver (i.e. production of bile for fat emulsification). Fat digestion and absorption begins in the mouth with fat sensing receptors and secretion of salivary lipase. Intriguingly, GF animals, devoid of all microorganisms have increased expression of the long-­chain fatty acid transporter cluster of differentiation 36 (CD36), which in the mouth acts as a fatty acid sensor.44 GF mice also exhibited reduced enteroendocrine cell numbers in the ileum44 and reduced cholecystokinin (CCK) receptor expression in the pancreas,15 indicating impaired digestive function and regulation by microorganisms. Consistent with these findings, GF mice display elevated fecal lipid content following HFD feeding15,45 indicating impaired fat absorption. In addition, antibiotic-­treated rats displayed reduced lymphatic lipid levels in a rat lymph cannulation model.46 Intriguingly, monoassociation with Bacteroidetes thetaiotamicron increased colipase expression in the ileum of gnotobiotic mice.47 Using fluorescently-­labeled fatty acids, it was clearly demonstrated that microbes regulate lipid absorption in zebrafish. Bodipy-­labeled short-­ and long-­chain fatty acid levels were reduced in the GF zebrafish epithelium but increased following bacterial colonization in both fasted and fed states, with more dramatic effects observed under fed conditions. Individual bacterial strains from different phyla, including Firmicutes and Bacteroidetes, were used to determine the specific effects on fat absorption. Here, it was found that a member of the Firmicutes, Exiguobacterium spp., increased lipid droplet number and other bacterial strains, Chryseobacterium spp. or Pseudomonas spp., increased lipid droplet size.48 These findings were corroborated and expanded upon using GF and ConvD mice. Martinez-­Guryn et al. showed

Fat Absorption, Metabolism, and Global Regulation

77

that GF mice have a decreased rate of lipid absorption which was assessed via delivery of radiolabeled triglyceride and cholesterol. First, specific pathogen-­ free (SPF) or conventional mice fed a HFD for four weeks gained more weight and fat mass compared with LFD-­fed mice, whereas GF mice given the same diets had negligible changes in body weight and body fat percentage. It was also demonstrated in this study that HFDs significantly affected the gut microbiota in the jejunum and ileum, including increasing the abundance of Clostridiaceae. Using radiolabeled lipid absorption assays, it was shown that GF mice have impaired fat absorption compared with SPF mice. GF mice conventionalized with jejunal contents collected from HFD-­fed mice displayed increased lipid absorption compared with mice receiving LFD-­induced jejunal microbiota. It was also shown that Lactobacillus rhamnosus GG selectively induced diacylglycerol O-­acyltransferase 1 (Dgat1) expression while Clostridium bifermentans induced Dgat2 expression in the duodenum and jejunum.15 A separate group demonstrated that Clostridium ramosum may regulate Cd36 expression in the small intestine in mice.49 C. ramosum is a microbe shown to be related to obesity and found in higher numbers in obese patients.50 Further studies were done on this microbe and it was found that C. ramosum was correlated with high fat diets. In another study, mice were divided into four treatment groups: LFD, HFD, LFD + C. ramosum and HFD + C. ramosum. In this experiment, it was observed that mice fed with a HFD with C. ramosum present were more obese compared with controls. While the mice that had the HFD weighed more than the LFD-­fed mice, the mice with C. ramosum had the largest body fat percentage compared with all other groups, as well as the most weight of the fat pads, such as mesenteric fat and epididymal fat.49 Complex interactions between microbes and host circadian rhythm and lipid absorption pathways have been demonstrated by Lora Hooper's group. First, it was shown that lipid absorption is regulated by the circadian transcription factor nuclear factor interleukin-­3 (Nfil3).51 Nfil3 expression is reduced in GF mice restored with the Gram-­negative, flagellated bacterial species Salmonella typhimurium and Escherichia coli. RNA sequencing analysis from Nfil3fl/fl [Nifl3 locus of cross-­over in P1 (loxP) flanked knock in homozygotes] versus Nfil3ΔIEC (intestinal epithelial cell-­specific Nfil3 knockouts) mice revealed that 17 transcripts related to lipid transport and metabolism were altered, including reduced Cd36 and stearoyl-­coenzyme A desaturase 1 (Scd1) which was associated with decreased lipid content in the intestinal epithelium. The same group showed that gut microbiota regulate histone deacetylase 3 (HDAC3) circadian rhythmicity as an additional mechanism for the microbial regulation of lipid absorption. HDAC3 forms a protein complex with proliferator-­activated receptor gamma co-­activator 1 alpha (PGC1α) and estrogen-­related receptor (ERR)-­α in intestinal epithelial cells to induce Cd36 expression, contributing to increased lipid absorption and diet-­induced obesity.52 Taken together, the results of these studies reveal that collective microbial communities affect the overall process of lipid digestion and absorption, wherein different bacterial strains have specialized effects

78

Chapter 4

on lipid absorption machinery (i.e. targeting different genes involved in lipid digestion and absorption or differential regulation of lipid droplet formation, and enteroendocrine signaling).15,34,48,51,52 Additional evidence indicates that microbes also directly affect metabolic activity of the small intestine. For example, it was concluded from microarray analyses that conventionalization of GF animals switched the jejunal gene profile from oxidative to anabolic metabolic pathways as early as one day following conventionalization.53 Similarly to the situation in the proximal gut, the interaction between the microbiota and the distal gut are essential determinants of host health. These interactions have been largely exemplified by the microbial production of SCFAs, particularly butyrate. Epithelial cells of the colonic mucosa are efficient at absorbing and metabolizing butyrate above other SCFAs. There is minimal butyrate found in the portal circulation and it constitutes anywhere from 15% to 20% of the SCFA substrates in the colon with over 10 mM absolute concentrations in human fecal matter.38 In fact, intestinal epithelial cells obtain an estimated 70% of their energy from butyrate. The commensal microbiota species that metabolize the greatest amounts of this SCFA is Clostridia, a type of Firmicutes. Thus, butyrate is a predominant source of energy from microbiota metabolism and provides barrier protection in the distal gut. Additional protective roles of butyrate include having anti-­carcinogenic and anti-­inflammatory properties such as through the regulation of immune responses via expansion of T-­regulatory cells. A study by Kelly et al.38 was performed to further understanding of the influence of butyrate metabolism in regulating O2 consumption by colonic epithelial cells of the distal gut. Butyrate was found to regulate oxygen consumption through stabilizing hypoxia-­inducible factor (HIF), which is a transcription factor that regulates epithelial barrier protection. Herein, GF mice were found to retain fewer oxygen-­sensitive dyes and had less stabilized HIF. HIF expression was reduced after antibiotic administration which was restored with delivery of butyrate. In addition, cells lacking HIF were unresponsive to butyrate. Altogether, these findings provide support for the hypothesis that the production of butyrate by colonic microbiota stabilizes HIF and epithelial barrier function in the host. Thus, butyrate mediates host–microbe crosstalk that facilitates the protection and maintenance of the gut epithelial barrier. SCFAs also affect peripheral metabolic organs such as the liver, discussed in the following section.

4.5  M  icrobial Regulation of Peripheral Metabolic Organs 4.5.1  Gut Microbiota–Liver Interactions While gut microbes have direct contact and influence host gut absorption and metabolism, they also significantly affect host liver function.54 A direct conduit from the gut to the liver is the portal vein through which small molecules generated by bacteria can be transported to the liver.55 Delivery of

Fat Absorption, Metabolism, and Global Regulation

79

gut-­derived compounds to the liver allows for further xenobiotic metabolism of foreign compounds that bypass the gut that could otherwise have detrimental effects on the host system. Sources of energy are also delivered to the liver via the portal vein, such as SCFAs produced from gut microbes. Interestingly, metabolite profiles in portal plasma are dramatically different between conventionalized or Ex-­GF and GF animals.56 For instance, Ex-­GF BALB/c mice were found to have 33 metabolites specifically collected from the portal vein that are significantly different compared with those of GF mice. These metabolites included the neurotransmitter gamma-­aminobutyric acid (GABA), glutamine, glycerol 3-­phosphate, succinic acid, ribulose 5-­phosphate, taurocholic acid, O-­acetylcarnitine, and trimethylamine N-­oxide, among others. As previously mentioned, bacterial metabolites from the gut, such as TMA, can be further metabolized by hepatic xenobiotic pathways, generating byproducts like TMAO that elicit cardiovascular disease.31 Not only do bacterial metabolites transport to the liver, but disruption of the gut epithelium can allow for bacterial translocation or for bacterial products like LPS to be transported to the liver, which may lead to liver pathology. Gut dysbiosis has been linked to liver malfunction and development of non-­alcoholic fatty liver disease (NAFLD) and non-­alcoholic steatohepatitis (NASH).57–59 For example, human patients with NAFLD or NASH were shown to have a significantly altered gut microbiota composition compared with healthy controls. Boursier et al. demonstrated that the severity of NAFLD is associated with a dysbiotic gut.57 Furthermore, transplant of fecal material collected from human NAFLD patients increased fatty liver, histological necrosis scores, and serum inflammatory cytokines in recipient GF mice fed a HFD,60 indicating a direct link between microbes and NAFLD development. Another group implicated the involvement of nucleotide-­binding oligomerization domain, leucine rich repeat and pyrin domain containing (NLRP) inflammasomes in this process. NLRP3 and NLRP6 inflammasomes act as detectors for pathogen-­associated molecular patterns (PAMPs) and damage-­ associated molecular patterns (DAMPs) that cleave pro-­interleukin-­1β and interleukin-­18 into their active forms. Mice deficient in apoptosis-­associated speck-­like protein containing a caspase-­recruitment domain (CARD) (Asc) or pyrin and CARD (Pycard), NLRP6, and NLRP3, fed a methionine-­choline-­ deficient diet (MCDD) developed exacerbated NASH characterized by increased plasma alanine aminotransferase (ALT) and aspartate transaminase (AST) levels and greater NAFLD activity scores. In addition, Asc−/− mice displayed altered gut microbiota composition, particularly an increase in members of the family Porphyromonadaceae, that promoted the development of NAFLD in wild-­t ype mice upon co-­housing.58 Conversely, Zhou et al. demonstrated that FMT improved markers of NASH in HFD-­fed C57Bl6 mice, including decreased hepatic oil red O levels accompanied by an increased abundance of the beneficial bacteria Christensenellaceae and Lactobacillus.61 Taken together, these findings reveal gut microbe interactions that directly affect liver function and may have grave consequences for host metabolic health.

80

Chapter 4

4.5.2  Gut Microbiota–Adipose Interactions Direct links have also been established between the gut microbiota and adipose tissue. An early and highly-­cited report is that from Jeffrey Gordon's group, in which it was demonstrated that conventionalization of GF mice with a typical microbiome increased adiposity compared with controls and also mediated insulin resistance.62 This occurred within two weeks and despite a lowered dietary intake. It was found that conventionalized mice exhibited decreased production of fasting-­induced adipocyte factor (Fiaf), an angiopoietin-­like protein. Fiaf is selectively suppressed in the intestinal epithelium of conventional mice compared with that of GF mice. Fiaf is a circulating protein that inhibits lipoprotein lipase in peripheral adipose tissue that would otherwise allow for cleavage of fatty acids from circulating lipoproteins and increase fat storage in adipose depots. Therefore, the suppression of Fiaf was identified as a mechanism by which the gut microbiota mediated triglyceride deposition in adipocytes. Another explanation of this effect is that SCFA production increases accessible energy to the host. SCFAs also influence insulin signaling, incretin production, and inflammation.2 First, studies have revealed that SCFAs function as natural ligands for several G-­protein coupled receptors (GPRs). In the intestine, adipose tissue, immune cells, and pancreas, free fatty acid receptors 2 and 3 have demonstrated an increased preference for the SCFAs propionate and acetate. In the intestine, more than anywhere else in the host body, GPR109A expresses a marked affinity for butyrate (which has already been determined to influence gut inflammation). Additionally, SCFAs have been shown to act as histone deacetylase inhibitors. By maintaining histone acetylation, SCFAs help maintain the regulation of transcription factor activation. This also influences the regulation of chromatin structure in the DNA and thus downstream gene expression. The use of probiotics and prebiotics have provided important insights into microbe–host interactions related to liver and adipose metabolism. For example, supplementation with Saccharomyces boulardii in diabetic db/db mice revealed that S. boulardii significantly reduced liver weight. This study highlights the use of leptin-­resistant obese and type 2 diabetic mice. S. boulardii is a yeast strain, part of the microbial flora in humans and other animals. In the study conducted, mice were divided into two groups with administration of sterile saline or S. boulardii in equal amounts over a span of four weeks. The mice exhibited a modest decrease in body weight gain which was about 15% lower in S. boulardii-­treated mice versus the controls.63 This effect was accompanied by a significantly reduced whole-­body fat mass, which was assessed by weighing the main fat depots: visceral, epididymal, and subcutaneous tissues. Plasma cytokine levels were also measured, showing changes in low-­grade inflammation. The administration of the yeast strain significantly decreased monocytic and cytokine activity by half. This indicates that microbes can enhance the activities of peripheral organs that are muted due to hormonal insufficiencies.

Fat Absorption, Metabolism, and Global Regulation

81

4.5.3  Gut Microbiota–Muscle Interactions The results of recent research demonstrate that gut microbes influence both muscle mass and glucose metabolism. For example, GF mice exhibited reduced skeletal muscle weight compared with SPF mice. Transplanting the gut microbiota of SPF mice into GF mice repleted muscle mass.64 Genes that regulate skeletal muscle differentiation, including the myogenic differentiation (MyoD) and Mygonin genes, were also reduced in GF skeletal muscle compared with the levels of expression in SPF mice. Interestingly, treatment with a cocktail of SCFAs increased gastrocnemius muscle mass and increased muscle strength.64 In another study, mice treated with a 21-­day course of broad spectrum antibiotics displayed decreased running endurance compared with untreated controls.65 Antibiotic treatment decreased gastrocnemius and quadriceps muscle weight and ex vivo muscle fatigue index and depleted glycogen stores in the gastrocnemius muscle. This was associated with reduced G protein coupled receptor 41 (Gpr41) and sodium-­dependent glucose cotransporter1 (Sglt1) expression in the ileum. Each of these indices were restored in mice that were given a ten-­day recovery following antibiotic treatment. Collectively, these findings support a role for gut microbes in regulating muscle mass and function that may be related to SCFA signaling and glucose transport from the gut to the muscle.65 Use of probiotics has provided additional insight into gut microbiota– muscle interactions. For example, a collection of Bacillus species protected mice from high fat diet-­induced obesity and insulin resistance.66 Specifically, the probiotic mixture decreased subcutaneous and mesenteric fat accumulation and also enhanced glucose tolerance, as well as reduced inflammatory cytokine expression in skeletal muscle.66 In addition to live bacteria, microbial metabolites have been demonstrated to regulate glucose uptake in skeletal muscle. For instance, the microbe-­derived metabolite isovanillic acid 3-­O-­sulfate (IVAS), increased glucose transport through glucose transporter 4 (GLUT4). IVAS up-­regulated GLUT1, GLUT4, and phosphatidylinositol-­3 kinase (PI3K) p85α protein, and increased phosphorylation of AKR mouse thyoma protein (Akt) in differentiated human muscle cells.67 These findings support a link between gut microbiota and stimulation of increased muscle mass, along with improved glucose uptake and metabolism.

4.6  Conclusion Taken together, the results of the studies presented herein reveal that host diet, gut microbiota, and subsequent host responses are intricately intertwined, leading to major physiological outcomes, including development of metabolic disease. These interactions begin at the site of the gastrointestinal tract but influence major peripheral metabolic organs, altogether affecting the health of the host. Moving forward, additional research is needed to further unravel these interactions and also to identify therapeutic strategies targeting the gut microbiota or harnessing microbe-­derived products for the betterment of human health.

82

Chapter 4

Conflicts of Interest There are no conflicts to declare.

Acknowledgements We would like to extend a special thank you to Sarah Quinlan and Sudeep Poludasu for their critical review of this chapter.

References 1. S. Sirisinha, Asian Pac. J. Allergy Immunol., 2016, 34, 249–264. 2. O. A. Baothman, M. A. Zamzami, I. Taher, J. Abubaker and M. Abu-­Farha, Lipids Health Dis., 2016, 15, 1–8. 3. B. Chassaing, M. Vijay-­Kumar and A. T. Gewirtz, Curr. Opin. Gastroenterol., 2017, 33, 417–421. 4. P. D. Cani, Gut, 2018, 67, 1716–1725. 5. A. F. Pihl, C. E. Fonvig, T. Stjernholm, T. Hansen, O. Pedersen and J. C. Holm, Child. Obes., 2016, 12, 292–299. 6. J. A. Oteo, M. J. Villanueva-­Millán and P. Pérez-­Matute, J. Physiol. Biochem., 2015, 71, 509–525. 7. C. Manichanh, S. Panda and F. Guarner, Endocr., Metab. Immune Disord.: Drug Targets, 2014, 14, 290–299. 8. A. MacFayden, M. Nencki and N. Sieber, J. Anat. Physiol., 1891, 25, 390–427. 9. L. A. David, C. F. Maurice, R. N. Carmody, D. B. Gootenberg, J. E. Button, B. E. Wolfe, A. V. A. Ling, S. Devlin, Y. Varma, M. A. Fischbach, S. B. Biddinger, R. J. Dutton and P. J. Turnbaugh, Nature, 2014, 505, 559–563. 10. L. K. Brahe, A. Astrup and L. H. Larsen, Adv. Nutr., 2016, 7, 90–101. 11. C. Sanmiguel, A. Gupta and E. A. Mayer, Curr. Obes. Rep., 2015, 4, 250–261. 12. S. Pettersson, B. Bjorkholm, C. M. Bok, A. Lundin, J. Rafter and M. L. Hibberd, PLoS One, 2009, e6958. 13. P. Ojeda, A. Bobe, K. Dolan, V. Leone and K. Martinez, J. Nutr. Biochem., 2016, 28, 191–200. 14. P. J. Turnbaugh, V. K. Ridaura, J. J. Faith, F. E. Rey and J. I. Gordon, Sci. Transl. Med., 2009, 1, 6rA14. 15. K. Martinez-­Guryn, N. Hubert, K. Frazier, S. Urlass, M. W. Musch, P. Ojeda, J. F. Pierre, J. Miyoshi, T. J. Sontag, C. M. Cham, C. A. Reardon, V. Leone and E. B. Chang, Cell Host Microbe, 2018, 23, 458–469.e5. 16. R. N. Carmody, G. K. Gerber, J. M. Luevano, D. M. Gatti, L. Somes, K. L. Svenson and P. J. Turnbaugh, Cell Host Microbe, 2015, 17, 72–84. 17. S. Devkota, Y. Wang, M. W. Musch, V. Leone, H. Fehlner-­peach, A. Nadimpalli, D. A. Antonopoulos, B. Jabri and E. B. Chang, Nature, 2012, 487, 104–109. 18. K. B. Martinez, V. Leone and E. B. Chang, Gut Microbes, 2017, 8, 130–142.

Fat Absorption, Metabolism, and Global Regulation

83

19. N. Alang and C. Kelly, Open Forum Infect. Dis., 2015, 2, 1–2. 20. A. Vrieze, E. Van Nood, F. Holleman, J. Salojärvi, R. S. Kootte, J. F. W. M. Bartelsman, G. M. Dallinga-­Thie, M. T. Ackermans, M. J. Serlie, R. Oozeer, M. Derrien, A. Druesne, J. E. T. Van Hylckama Vlieg, V. W. Bloks, A. K. Groen, H. G. H. J. Heilig, E. G. Zoetendal, E. S. Stroes, W. M. De Vos, J. B. L. Hoekstra and M. Nieuwdorp, Gastroenterology, 2012, 143, 913–916.e7. 21. A. Liou, M. Paziuk, J.-­M. Luevano Jr, S. Machineni, P. J. Turnbaugh and L. M. Kaplan, Sci. Transl. Med., 2013, 5, 178ra41. 22. J. F. Pierre, K. B. Martinez, H. Ye, A. Nadimpalli, T. C. Morton, J. Yang, Q. Wang, N. Patno, E. B. Chang and D. P. Yin, Am. J. Physiol.: Gastrointest. Liver Physiol., 2016, 311, G286–G304. 23. S. A. Boytsov, D. A. Kashtanova, A. S. Popenko, O. N. Tkacheva, A. B. Tyakht and D. G. Alexeev, Nutrition, 2016, 32, 620–627. 24. E. Y. Huang, V. A. Leone, S. Devkota, Y. Wang, M. J. Brady and E. B. Chang, JPEN, J. Parenter. Enteral Nutr., 2013, 746–754. 25. A. Everard, V. Lazarevic, N. Gaïa, M. Johansson, M. Ståhlman, F. Backhed, N. M. Delzenne, J. Schrenzel, P. François and P. D. Cani, ISME J., 2014, 8, 2116–2130. 26. B. Collins, J. Hoffman, K. Martinez, M. Grace, M. A. Lila, C. Cockrell, A. Nadimpalli, E. Chang, C. C. Chuang, W. Zhong, J. Mackert, W. Shen, P. Cooney, R. Hopkins and M. McIntosh, J. Nutr. Biochem., 2016, 31, 150–165. 27. J. Baldwin, B. Collins, P. G. Wolf, K. Martinez, W. Shen, C. C. Chuang, W. Zhong, P. Cooney, C. Cockrell, E. Chang, H. R. Gaskins and M. K. McIntosh, J. Nutr. Biochem., 2016, 27, 123–135. 28. S. C. Chattipakorn, C. Proctor, P. Thiennimitr and N. Chattipakorn, Metab. Brain Dis., 2017, 32, 1–17. 29. F. Shanahan, Curr. Opin. Gastroenterol., 2013, 29, 49–54. 30. C. Moran, D. Sheehan and F. Shanahan, Curr. Opin. Gastroenterol., 2015, 31, 130–136. 31. J. M. Brown and S. L. Hazen, Annu. Rev. Med., 2015, 66, 343–359. 32. T. D. Shen, Nestle Nutr. Inst. Workshop Ser., 2017, 88, 117–126. 33. C. De Filippo, D. Cavalieri, M. Di Paola, M. Ramazzotti, J. B. Poullet, S. Massart, S. Collini, G. Pieraccini and P. Lionetti, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 14691–14696. 34. E. B. Chang and K. Martinez-­Guryn, Gut Microbes, 2018, 1–6. 35. K. Martinez-­guryn, V. Leone and E. B. Chang, Cell Host Microbe, 2019, 26, 314–324. 36. A. A. Ribeiro, M. A. Azcarate-­Peril, M. B. Cadenas, N. Butz, B. J. Paster, T. Chen, E. Bair and R. R. Arnold, PLoS One, 2017, 12, 1–16. 37. V. Tremaroli and F. Bäckhed, Nature, 2012, 489, 242–249. 38. C. J. Kelly, L. Zheng, E. L. Campbell, B. Saeedi, C. C. Scholz, A. J. Bayless, K. E. Wilson, L. E. Glove, D. J. Kominsky, A. Magnuson, T. L. Weir, S. F. Ehrentraut, C. Pickel, K. A. Kuhn, J. M. Lanis, V. Nguyen, C. T. Taylor and S. P. Colgan, Cell Host Microbe, 2015, 17, 662–671.

84

Chapter 4

39. I. Churruca, A. Fernández-­Quintela and M. P. Portillo, BioFactors, 2009, 35, 105–111. 40. E. Rosberg-­Cody, C. Stanton, L. O'Mahony, R. Wall, F. Shanahan, E. M. Quigley, G. F. Fitzgerald and R. P. Ross, Microbiology, 2011, 157, 609–615. 41. R. Wall, R. P. Ross, F. Shanahan, L. O'Mahony, C. O'Mahony, M. Coakley, O. Hart, P. Lawlor, E. M. Quigley, B. Kiely, G. F. Fitzgerald and C. Stanton, Am. J. Clin. Nutr., 2009, 89, 1393–1401. 42. P. J. Jones, M. Plourde, S. Jew and S. C. Cunnane, Nutr. Rev., 2008, 66, 415–421. 43. T. A. McCrorie, E. M. Keaveney, J. M. W. Wallace, N. Binns and M. B. E. Livingstone, Nutr. Res. Rev., 2011, 24, 206–227. 44. F. A. Duca, T. D. Swartz, Y. Sakar and M. Covasa, PLoS One, 2012, 7, e39748. 45. S. Rabot, M. Membrez, P. Gérard, T. Harach, M. Moser, F. Raymond, R. Mansourian, C. J. Chou, FASEB J., 2010, 24, 4948–4959. 46. H. Sato, L. Zhang, K. Martinez, E. B. Chang, Q. Yang, F. Wang, P. N. Howles, R. Hokari, S. Miura and P. Tso, Gastroenterology, 2016, 151, 923–932. 47. L. V Hooper, M. H. Wong, A. Thelin, L. Hansson, P. G. Falk and J. I. Gordon, Science, 2001, 291, 881–885. 48. I. Semova, J. D. Carten, J. Stombaugh, L. C. Mackey, R. Knight and S. A. Farber, Cell Host Microbe, 2012, 12, 277–288. 49. A. Woting and M. Blaut, Nutrients, 2016, 202. 50. E. Le Chatelier, T. Nielsen, J. Qin, E. Prifti, F. Hildebrand, G. Falony, M. Almeida, M. Arumugam, J. M. Batto, S. Kennedy, P. Leonard, J. Li, K. Burgdorf, N. Grarup, T. Jørgensen, I. Brandslund, H. B. Nielsen, A. S. Juncker, M. Bertalan, F. Levenez, N. Pons, S. Rasmussen, S. Sunagawa, J. Tap, S. Tims, E. G. Zoetendal, S. Brunak, K. Clément, J. Doré, M. Kleerebezem, K. Kristiansen, P. Renault, T. Sicheritz-­Ponten, W. M. De Vos, J. D. Zucker, J. Raes, T. Hansen, P. Bork, J. Wang, S. D. Ehrlich, O. Pedersen, E. Guedon, C. Delorme, S. Layec, G. Khaci, M. Van De Guchte, G. Vandemeulebrouck, A. Jamet, R. Dervyn, N. Sanchez, E. Maguin, F. Haimet, Y. Winogradski, A. Cultrone, M. Leclerc, C. Juste, H. Blottière, E. Pelletier, D. Lepaslier, F. Artiguenave, T. Bruls, J. Weissenbach, K. Turner, J. Parkhill, M. Antolin, C. Manichanh, F. Casellas, N. Boruel, E. Varela, A. Torrejon, F. Guarner, G. Denariaz, M. Derrien, J. E. T. Van Hylckama Vlieg, P. Veiga, R. Oozeer, J. Knol, M. Rescigno, C. Brechot, C. M'Rini, A. Mérieux and T. Yamada, Nature, 2013, 500, 541–546. 51. Y. Wang, Z. Kuang, X. Yu, K. A. Ruhn, M. Kubo and L. V Hooper, Science, 2017, 357, 912–916. 52. Z. Kuang, Y. Wang, L. Yun, C. Ye, K. A. Ruhn, C. L. Behrendt, E. N. Olson and L. V. Hooper, Science, 2019, 365, 1428–1434. 53. S. El Aidy, C. A. Merrifield, M. Derrien, P. Van Baarlen, G. Hooiveld, F. Levenez, J. Doré, J. Dekker, E. Holmes, S. P. Claus, D. J. Reijngoud and M. Kleerebezem, Gut, 2013, 62, 1306–1314. 54. A. Wahlström, Biochim. Biophys. Acta, Mol. Basis Dis., 2019, 1865, 912–919.

Fat Absorption, Metabolism, and Global Regulation

85

55. H. Nakao, C. Kindberg, J. W. Suttie, K. Uchida and K. Hirauchi, J. Nutr., 1987, 117, 1032–1035. 56. M. Matsumoto, T. Ooga, R. Kibe, Y. Aiba, Y. Koga and Y. Benno, PLoS One, 2017, 12, 1–15. 57. J. Boursier, O. Mueller, M. Barret, M. Machado, L. Fizanne, F. Araujo-­ perez, C. D. Guy, P. C. Seed, J. F. Rawls, A. Lawrence, G. Hunault, F. Oberti, P. Calès and A. M. Diehl, Hepatology, 2017, 63, 764–775. 58. J. Henao-­Mejia, E. Elinav, C. Jin, L. Hao, W. Z. Mehal, T. Strowig, C. A. Thaiss, A. L. Kau, S. C. Eisenbarth, M. J. Jurczak, J. P. Camporez, G. I. Shulman, J. I. Gordon, H. M. Hoffman and R. A. Flavell, Nature, 2012, 482, 179–185. 59. L. Zhu, S. S. Baker, C. Gill, W. Liu, R. Alkhouri, R. D. Baker and S. R. Gill, Hepatology, 2013, 57, 601–609. 60. C. C. Chiu, Y. H. Ching, Y. P. Li, J. Y. Liu, Y. Te Huang, Y. W. Huang, S. S. Yang, W. C. Huang and H. L. Chuang, Nutrients, 2017, E1220. 61. D. Zhou, Q. Pan, F. Shen, H. X. Cao, W. J. Ding, Y. W. Chen and J. G. Fan, Sci. Rep., 2017, 7, 1–11. 62. F. Bäckhed, H. Ding, T. Wang, L. V. Hooper, Y. K. Gou, A. Nagy, C. F. Semenkovich and J. I. Gordon, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 15718–15723. 63. A. Everard, S. Matamoros, L. Geurts, N. M. Delzenne and P. D. Cani, mBio, 2014, 5, 1–9. 64. S. Lahiri, H. Kim, I. Garcia-­perez, M. M. Reza, K. A. Martin, P. Kundu, L. M. Cox, J. Selkrig, J. M. Posma, H. Zhang, P. Padmanabhan, C. Moret, B. Gulyas, M. J. Blaser, J. Auwerx, E. Holmes, J. Nicholson, W. Wahli and S. Pettersson, Sci. Transl. Med., 2019, 11, eaan5662. 65. K. Nay, M. Jollet, B. Goustard, N. Baati, B. Vernus, M. Pontones, L. Lefeuvre-­Orfila, C. Bendavid, O. Rué, M. Mariadassou, A. Bonnieu, V. Ollendorff, P. Lepage, F. Derbré and C. Koechlin-­Ramonatxo, Am. J. Physiol., 2019, 317, E158–E171. 66. B. Kim, J. Kwon, M. S. Kim, H. Park, Y. Ji, W. Holzapfel and C. K. Hyun, PLoS One, 2018, 13, 1–17. 67. M. J. Houghton, A. Kerimi, V. Mouly, S. Tumova and G. Williamson, FASEB J., 2019, 33, 1887–1898.

Chapter 5

Probiotics, Prebiotics, and Synbiotics in Human Health Olivia L. Wellsa, Sidharth Mishrab and Hariom Yadav*c a

Department of Internal Medicine, Wake Forest Baptist Medical Center, Winston Salem, NC, USA; USF Center for Microbiome Research, Microbiomes Institute, Department of Neurosurgery and Brain Repair, University of South Florida, Tampa, FL, USA; USF Center for Microbiome Research, Microbiomes Institute, Center of Excellence for Aging and Brain Repair, Byrd Alzheimer’s Center, Department of Neurosurgery and Brain Repair, University of South Florida, Tampa, FL, USA *E-­mail: [email protected], [email protected], [email protected]

5.1  Introduction This chapter will explore the definitions of probiotics, prebiotics and synbiotics as well as the role that these entities play in the maintenance of the delicate ecosystem of the human gut microbiota. The body of research reviewed primarily focuses on the relationship between the composition of gut flora as it pertains to general host health and pathology as studied in both human and animal trials. The gastrointestinal tract (GIT) harbors one of the highest concentrations of bacteria in the world making this microbial ecosystem an exceedingly unusual and challenging area of study. The human intestinal microbiome of a healthy individual is estimated to contain approximately 100 trillion bacteria.1 The large majority of these species originate from four   Food Chemistry, Function and Analysis No. 34 Metabolism of Nutrients by Gut Microbiota Edited by Joseph F. Pierre © The Royal Society of Chemistry 2022 Published by the Royal Society of Chemistry, www.rsc.org

86

Probiotics, Prebiotics, and Synbiotics in Human Health

87

main phyla: Bacteroidetes, Firmicutes, Actinobacteria, and Proteobacteria.1 In addition to these bacteria, the gastrointestinal tract is home to viruses, fungi, yeasts, helminths, protists and archaea.1 The human microbiome is an ecological organization of symbiotic, pathogenic and commensal microorganisms that occupy space in the host environment where they reside.2 Our gut microbiome is unique to the individual. The microorganismal constituents vary from person to person due to dietary habits, geography, environment, age, genetics, and many other factors.2 As more information on the gut microbiota is unfolding, the understanding of the roles of probiotics, prebiotics, and synbiotics and their therapeutic potential is even more important. This chapter will explore the role of these bioactive substances in the GIT, specifically with regard to host metabolism, specific disease processes, and overall health.

5.1.1  Probiotics The World Health Organization as well as the International Scientific Association for Probiotics and Prebiotics (ISAPP) define probiotics as “live microorganisms which, when administered in adequate amounts, confer a health benefit on the host.”3 Experts have delineated this definition by stating that probiotics must have a large enough number of viable cells to be capable of surviving and metabolizing in the gut to exert beneficial health effects.3,4 This notion has been an area of study for decades; however, considerable strides been made since the human intestinal bacterial genome sequences became a part of the public database. The United States National Institutes of Health Human Microbiome Project (HMP) was completed in 2012 and has served as a launching pad for a multitude of studies pertaining to the intestinal microbiome, laying the foundations for research projects worldwide.5 This project studied intestinal samples from over 250 volunteers in an effort to provide a standardized data resource and to take advantage of new technologies to characterize the human microbiome. The objective of the HMP was to provide the scientific community with a resource that could differentiate changes in the microbiome and their association with health and disease. The HMP validated the potential for therapeutic opportunities to better human health with observation and understanding of the gut microbiome.5 Similarly, the European Metagenomics of the Human Intestinal Tract (MetaHIT) program demonstrated the beneficial properties of the gut flora at a genetic level.6 In doing so, the role of probiotics and their therapeutic capacity became enhanced with this newfound breadth of information regarding the profile of the microbiota of the gut. This chapter will explore some of the health benefits of probiotics in specific pathologies as they are utilized in clinical practice today.

5.1.2  Prebiotics Prebiotics are a distinct entity from probiotics. Prebiotics are defined as host non-­digestible polysaccharides and related oligosaccharides from either plant or food sources that serve as a beneficial energy source for selective

88

Chapter 5

microbes in the GIT or other microbially colonized sites such as the skin or vaginal tract.7 Prebiotics can selectively modulate the growth of bacteria in the large intestine because they function as the nutrient source for probiotic supplementation and, in many cases, for the native gut flora as well. Deliberate supplementation of prebiotics in the human diet can enhance bacterial fermentation and have a selective beneficial effect once metabolized in the intestinal tract. The health benefits must be documented to define a substance as a prebiotic.8 In humans, ingestion of prebiotics leads to an increase in production of some fermented metabolites such as short-­chain fatty acids (SCFAs; i.e. acetate, propionate, and butyrate) that have significant health benefits.9 Prebiotics are divided into two categories: fermentable and non-­fermentable. Non-­fermentable polysaccharides travel through the large intestine and are excreted as waste. On the other hand, indigestible but fermentable polysaccharides are metabolized by the microflora of the large intestine and are fermented in the colon to produce SCFAs that serve as an energy source for the host as represented in Figure 5.1.7 Acetate (two carbon, C2), propionate (C3), and butyrate (C4) are all taken up by the colonic mucosa and can influence sites beyond the gut in different capacities and are also commonly referred to as “postbiotics.”9 Nagpal et al. utilized human derived probiotic strains (Lactobacillus and Enterococcus), administered them to mice, and noted an increase in SCFA production (specifically butyrate and propionate) in the GITs of the mice. Their findings demonstrate how inoculation of the GIT with probiotics can directly modulate the microbiome and its metabolites, which can influence future health outcomes.9,10

Figure 5.1  Fermentation  of indigestible polysaccharides by gut bacteria. Indigest-

ible but fermentable polysaccharides are metabolized by the microflora of the large intestine and are fermented in the colon to produce SCFAs that serve as an energy source for the host. Acetate (two carbon, C2), propionate (C3), and butyrate (C4) are displayed as “postbiotics” that can influence sites beyond the gut in different capacities.

Probiotics, Prebiotics, and Synbiotics in Human Health

89

Further distinction categorizes prebiotics into three major groups: fructooligosaccharides (FOS) or galactooligosaccharides (GOS), lactulose, and other non-­digestible carbohydrates. The food ingredients considered non digestible carbohydrates include large polysaccharides, such as inulin, starches, gums, cellulose, and pectins, as well as some oligosaccharides that are not digested, and unabsorbed sugars and alcohols.11 Not all of the non-­digestible carbohydrates, however, are considered prebiotics according to the ISAPP.8 A distinction is made between certain dietary fibers and prebiotics in that fibers such as cellulose, pectins and xylans foster growth of a variety of different gut organisms, whereas prebiotics elicit growth from health-­promoting microorganisms within the gut of the host.8 The two most commonly used prebiotics are FOS and GOS. Fermentation that generates a prebiotic effect of a polysaccharide depends upon its chemical composition, chain length, branching, solubility, porosity and the presence of other proteins and/or lipids.12 Historically, initial studies of prebiotics (FOS and GOS in particular) demonstrated promotion of growth of Bifidobacterium or Lactobacillus but not members of Clostridia class. It is now recognized that the effect of prebiotics is not selective to these two taxa alone, as the results of many studies in both humans and animals have indicated.8 A recent study in humans demonstrated a response of Bifidobacteria to prebiotic use, as well as an increase in abundance of Faecalibacterium prausnitzii, a species known for its potential anti-­inflammatory characteristics.13 Another study in humans detected an increase in Anaerostipes spp. and a decrease in Bilophila spp.14 One study looked at the effects of different types of prebiotics used. The effects of prebiotics such as acorn-­ and sago-­derived prebiotics were compared with those of inulin prebiotics in mice. It was found that Bacteroidetes abundance was significantly increased in the acorn prebiotic treated healthy fecal microbiomes of mice and humans, while both acorn and sago prebiotics increased Bacteroidetes and decreased Firmicutes in diseased microbiomes.9 The novel prebiotics and inulin alike increased SCFA levels in the mice, reduced fecal pH, and reduced mucosal inflammatory markers.9 These studies, among many others, demonstrate the potential for targeted gut microbiome modulation and multidimensional therapeutic benefits.

5.1.3  Synbiotics The term synbiotics refers to the administration of prebiotic substrates with live bacteria that enables the enhancement of that energy source. There are two types of synergism between prebiotics and probiotics. The first mechanism of synbiotics involves a combination of probiotics and prebiotics that has a synergistic effect in which the prebiotic promotes the growth of the probiotic. However, there is also a belief that the prebiotics and the probiotics act separately in the gut where they stimulate host microbiome development independently.15 There are fewer studies on synbiotics singularly, as they are often grouped together with the studies that are looking at either prebiotics or probiotics.

90

Chapter 5

5.2  The Gut Microbiome and Human Health The microbiota of the human gut has coevolved from an evolutionary standpoint with the human host to confer a symbiotic relationship that plays a vital role in our overall health. During this evolution, the bacteria have cultivated multiple mechanisms that influence the immune, inflammatory and metabolic functions of the human host. Given the unique metabolic and immunologic integration between the gut microbiota and the human host, humans are considered by some to be “meta-­organisms,” by virtue of the tenfold greater number of bacterial cells than animal cells in the human body.16 Generally speaking, the bacteria residing in the human gut are predominantly anaerobes. Anaerobes are more prevalent in the bacterial community, and the majority of the population is represented by Bacteroidetes and Firmicutes, two anaerobic Gram-­positive phyla, which constitute approximately 90% of all the microbes.17 The gut associated immune system and the gut microbiome have evolved together over time to maintain an interface between the host and the external world. The majority of this chapter will focus on this bacterial community; however, it bears mentioning that the mycobiota (the yeast and fungal community) are commensal microorganisms that are integral to host health in the GIT as well. Currently the mycobiota is not as thoroughly researched as the bacterial organisms. Yeasts are a part of the microbiota that constitute approximately