Light Microscopy: Methods and Protocols [1 ed.] 1607619490, 9781607619499, 9781607619505

Citation preview

ME T H O D S

IN

MO L E C U L A R BI O L O G Y

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK

For other titles published in this series, go to www.springer.com/series/7651

TM

Light Microscopy Methods and Protocols

Edited by

Hélio Chiarini-Garcia Laboratory of Structural Biology and Reproduction, Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil

Rossana C.N. Melo Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil

Editors Hélio Chiarini-Garcia Department of Morphology Federal University of Minas Gerais Belo Horizonte, MG 31270-901, Brazil [email protected]

Rossana C.N. Melo Department of Biology Federal University of Juiz de Fora Juiz de Fora, MG 36036-900, Brazil [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-949-9 e-ISBN 978-1-60761-950-5 DOI 10.1007/978-1-60761-950-5 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2010936902 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface Of all scientific instruments, probably none has had more applications in the life sciences than the light microscope. Advances in microscope instrumentation, sample preparation and imaging techniques have been producing fundamental insights into the functions of cells and tissues. The protocols in Light Microscopy: Methods and Protocols cover a variety of brightfield and fluorescence microscopy-based approaches central to the study of a range of biological questions. The book provides information on how to prepare cells and tissues for microscopic investigations, including detailed staining procedures and how to analyze images and interpret results accurately. Techniques are presented in a friendly, step-by-step fashion with helpful information and useful tips. Section I covers selected applications of bright-field microscopy to the study of animal and plant biology. Section II covers the fundamental principles of fluorescence microscopy as well as its applications to multiple fields including immunology, ecology, cancer biology and cell signaling. Light Microscopy: Methods and Protocols addresses different needs of researchers, who are exploring the microscopic and intriguing world of the cell. We thank Prof. John M. Walker and the staff at Humana Press for their invitation, editorial guidance, and assistance throughout the preparation of this book for publication. We also would like to express our sincere appreciation and gratitude to the contributors for sharing their precious laboratory expertise with the microscopy research community. Hélio Chiarini-Garcia Rossana C.N. Melo

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

SECTION I 1.

2.

BRIGHT-FIELD MICROSCOPY APPLICATIONS

Glycol Methacrylate Embedding for Improved Morphological, Morphometrical, and Immunohistochemical Investigations Under Light Microscopy: Testes as a Model . . . . . . . . . . . . . . . . . . . . . . . . . . . Hélio Chiarini-Garcia, Gleydes Gambogi Parreira, and Fernanda R.C.L. Almeida

3

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . Gerluza Aparecida Borges Silva, Adriana Moreira, and José Bento Alves

19

3.

Large Plant Samples: How to Process for GMA Embedding? . . . . . . . . . . . Élder Antônio Sousa Paiva, Sheila Zambello de Pinho, and Denise Maria Trombert Oliveira

37

4.

Image Cytometry: Nuclear and Chromosomal DNA Quantification Carlos Roberto Carvalho, Wellington Ronildo Clarindo, and Isabella Santiago Abreu

. . . . . . .

51

5.

Histological Approaches to Study Tissue Parasitism During the Experimental Trypanosoma cruzi Infection . . . . . . . . . . . . . . . . . . . . Daniela L. Fabrino, Grazielle A. Ribeiro, Lívia Teixeira, and Rossana C.N. Melo

6.

Intravital Microscopy to Study Leukocyte Recruitment In Vivo . . . . . . . . . . Vanessa Pinho, Fernanda Matos Coelho, Gustavo Batista Menezes, and Denise Carmona Cara

SECTION II

69

81

FLUORESCENCE MICROSCOPY APPLICATIONS

7.

Introduction to Fluorescence Microscopy . . . . . . . . . . . . . . . . . . . . . Ionita C. Ghiran

8.

Using the Fluorescent Styryl Dye FM1-43 to Visualize Synaptic Vesicles Exocytosis and Endocytosis in Motor Nerve Terminals . . . . . . . . . . . . . . 137 Ernani Amaral, Silvia Guatimosim, and Cristina Guatimosim

9.

Imaging Lipid Bodies Within Leukocytes with Different Light Microscopy Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Rossana C.N. Melo, Heloisa D’Ávila, Patricia T. Bozza, and Peter F. Weller

vii

93

viii

Contents

10.

EicosaCell – An Immunofluorescent-Based Assay to Localize Newly Synthesized Eicosanoid Lipid Mediators at Intracellular Sites . . . . . . . . . . . 163 Christianne Bandeira-Melo, Peter F. Weller, and Patricia T. Bozza

11.

Nestin-Driven Green Fluorescent Protein as an Imaging Marker for Nascent Blood Vessels in Mouse Models of Cancer . . . . . . . . . . . . . . . . 183 Robert M. Hoffman

12.

Imaging Calcium Sparks in Cardiac Myocytes . . . . . . . . . . . . . . . . . . . 205 Silvia Guatimosim, Cristina Guatimosim, and Long-Sheng Song

13.

Light Microscopy in Aquatic Ecology: Methods for Plankton Communities Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Maria Carolina S. Soares, Lúcia M. Lobão, Luciana O. Vidal, Natália P. Noyma, Nathan O. Barros, Simone J. Cardoso, and Fábio Roland

14.

Fluorescence Immunohistochemistry in Combination with Differential Interference Contrast Microscopy for Studies of Semi-ultrathin Specimens of Epoxy Resin-Embedded Samples . . . . . . . . . . . . . . . . . . 229 Shin-ichi Iwasaki and Hidekazu Aoyagi

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Contributors ISABELLA SANTIAGO ABREU • Laboratory of Cytogenetics and Cytometry, Department of General Biology, Federal University of Viçosa, Viçosa, MG, Brazil FERNANDA R.C.L. ALMEIDA • Laboratory of Structural Biology and Reproduction, Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil JOSÉ BENTO ALVES • University of Uberaba, Uberaba, MG, Brazil ERNANI AMARAL • Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil HIDEKAZU AOYAGI • Advanced Research Center, School of Life Dentistry at Niigata, The Nippon Dental University, Niigata, Japan CHRISTIANNE BANDEIRA-MELO • Laboratory of Inflammation, Carlos Chagas Filho Institute of Biophysics, Federal University of Rio de Janeiro, Rio de Janeiro, RJ, Brazil NATHAN O. BARROS • Laboratory of Aquatic Ecology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil PATRICIA T. BOZZA • Laboratory of Immunopharmacology, IOC, Oswaldo Cruz Foundation, Rio de Janeiro, RJ, Brazil SIMONE J. CARDOSO • Laboratory of Aquatic Ecology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil CARLOS ROBERTO CARVALHO • Laboratory of Cytogenetics and Cytometry, Department of General Biology, Federal University of Viçosa, Viçosa, MG, Brazil HÉLIO CHIARINI-GARCIA • Laboratory of Structural Biology and Reproduction, Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil WELLINGTON RONILDO CLARINDO • Laboratory of Cytogenetics and Cytometry, Department of General Biology, Federal University of Viçosa, Viçosa, MG, Brazil FERNANDA MATOS COELHO • Laboratory of Immunopharmacology, Department of Biochemistry and Immunology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil DENISE CARMONA CARA • Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil HELOÍSA D’ÁVILA • Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil DANIELA L. FABRINO • Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil IONITA C. GHIRAN • Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA CRISTINA GUATIMOSIM • Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil

ix

x

Contributors

SILVIA GUATIMOSIM • Department of Physiology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil ROBERT M. HOFFMAN • AntiCancer Inc and Department of Surgery, University of California, San Diego, CA, USA SHIN-ICHI IWASAKI • Advanced Research Center, School of Life Dentistry at Niigata, The Nippon Dental University, Niigata, Japan LÚCIA M. LOBÃO • Laboratory of Aquatic Ecology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil ROSSANA C.N. MELO • Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil GUSTAVO BATISTA MENEZES • Laboratory of Immunopharmacology, Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil ADRIANA MOREIRA • Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil NATÁLIA P. NOYMA • Laboratory of Aquatic Ecology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil DENISE M.T. OLIVEIRA • Department of Botanic, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil ÉLDER ANTÔNIO SOUSA PAIVA • Department of Botanic, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil GLEYDES GAMBOGI PARREIRA • Laboratory of Structural Biology and Reproduction, Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil SHEILA ZAMBELLO DEPINHO • Department of Biostatistics, Institute of Biosciences, UNESP – Universidade Estadual Paulista, Botucatu, SP, Brazil VANESSA PINHO • Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil GRAZIELLE A. RIBEIRO • Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil FÁBIO ROLAND • Laboratory of Aquatic Ecology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil GERLUZA APARECIDA BORGES SILVA • Department of Morphology, ICB, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil MARIA CAROLINA S. SOARES • Laboratory of Aquatic Ecology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil LONG-SHENG SONG • Division of Cardiovascular Medicine, Department of Internal Medicine, University of Iowa Carver College of Medicine, Iowa City, IA, USA LÍVIA TEIXEIRA • Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil LUCIANA O. VIDAL • Laboratory of Cellular Biology, Department of Biology, ICB, Federal University of Juiz de Fora, Juiz de Fora, MG, Brazil PETER F. WELLER • Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA

Section I Bright-Field Microscopy Applications

Chapter 1 Glycol Methacrylate Embedding for Improved Morphological, Morphometrical, and Immunohistochemical Investigations Under Light Microscopy: Testes as a Model Hélio Chiarini-Garcia, Gleydes Gambogi Parreira, and Fernanda R.C.L. Almeida Abstract Glycol methacrylate (GMA), a water and ethanol miscible plastic resin, is a medium handy to use for light microscopy embedding that has a number of advantages than paraffin embedding. The GMA improves the histological, morphometrical, and immunohistochemical evaluations, mainly due to the accurate assessment of cytological details. This chapter focuses on our experience in the GMA processing and describes in detail the fixation, embedding, and staining methods that we have been using for testes evaluations. Key words: Glycol methacrylate, fixation, embedding, light microscopy, testes.

1. Introduction The first studies that described the seminiferous epithelium structure as we know today emerged at the end of 1950 and starting of 1960 (1–3). These studies demonstrated that germ cells in different steps of the spermatogenic process – spermatogonial, spermatocytary, and spermiogenic – are distributed in a wellorganized way along the seminiferous tubules. These initial studies were developed mainly in testes fixed in Zenker or Bouin solutions and embedded in paraffin. Although these histological processing were adequate to describe the development of the acrossomal system (stained in purple by periodic acid-Schiff), to H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_1, © Springer Science+Business Media, LLC 2011

3

4

Chiarini-Garcia, Parreira, and Almeida

differentiate steps of the spermatocyte and able to distinguish some spermatogonia, accurate morphological details of all process, mainly those related to spermatogonial subtypes, were not clearly demonstrated. At the same time, another morphological method was also standardized; this was very important for the spermatogonial biology studies. It was a whole mount method that analyzed the testis in toto and classified the spermatogonial subtypes by their grouping, that is, if they were single, in pair or aligned in 4 up to 32 spermatogonia together in the same clone (4). However, this method was not accurate to distinguish the spermatogonial subtypes by their morphology. More recently, it was demonstrated that a morphological method used in the past for important male reproduction researches (5), applying glutaraldehyde fixation, araldite embedding, and semi-thin sections, could be used for high-resolution light microscopic evaluations of the spermatogonial cell in different morphological and morphometrical approaches (6–13). However, as this method is a pre-preparation for transmission electron microscopy process, fragments have to be very small (2 mm2 ), allowing adequate penetration of fixatives and resins into tissues. A method that combines different advantages of the method exposed above and allows morphological, morphometrical, and immunohistochemical studies in semi-thin or thick sections, in large fragments and with satisfactory morphology, is the one that uses fixation with paraformaldehyde and/or glutaraldehyde and embedding in plastic resin based in glycol methacrylate (GMA). The GMA embedding has been used to present some advantages over the usual methods (14–16), namely (a) fast processing, (b) hydrosoluble, (c) easy handling, (d) infiltration and polymerization at room temperature, (e) possible to obtain semi-thin section (0.5 µm), (f) less distortion and artifacts, and (g) better resolution over light microscopy. We have been using GMA embedding since 1990 for different studies, such as mast cells (17–22), male reproduction (23, 24), equine endometrium (25), and aquatic organisms (26). Here, we present, in detail, our experience in GMA processing, and its tricks, focusing on testis preparation for its high performance studies.

2. Materials 1. Heparin (Liquemine, Roche). 2. Sodium thiopental (Thiopentax, Cristalia) intravenous bottle. 3. Three-way stopcock.

GMA for Improved Investigations Under Light Microscopy

5

4. Catheter (Angiocath, BD). 5. Saline (sodium chloride at 0.9%). 6. Phosphate buffered solution of 0.1 M and pH 7.4: Dissolve 1.38 g NaH2 HPO4 .H2 O (0.1 M) in 100 mL distilled water (solution A) and 1.42 g Na2 H.HPO4 (0.1 M) in 100 mL distilled water (solution B). To prepare the buffer at pH 7.4, mix 19 mL of solution A in 81 mL of solution B. Adjust pH with the same solutions. 7. 8% Paraformaldehyde solution: Heat 70 mL distilled water at 60–70◦ C and add 8 g of paraformaldehyde. Mix well and add drops of NaOH (0.1 M) until a clean solution is obtained. Wait to get to room temperature before using. 8. 4% paraformaldehyde in phosphate buffer 0.05 M pH 7.4: Prepare 100 mL solution by mixing 50 mL of 8% paraformaldehyde, freshly prepared in 50 mL of phosphate buffer at 0.1 M and pH 7.4. 9. 5% glutaraldehyde in phosphate buffer 0.05 M pH 7.4: Mix 10 mL of glutaraldehyde (biological grade at 50%) in 50 mL of 0.1 M phosphate buffer at pH 7.4 and complete the volume to 100 mL with distilled water. 10. Karnovsky’s fixative − 2% paraformaldehyde, 2.5% glutaraldehyde in phosphate buffer 0.05 M pH 7.4: We have used the original formula diluted in a half as follows: mix 50 mL phosphate buffer (0.1 M), 5 mL glutaraldehyde (50%), 20 mL paraformaldehyde (10%) and complete the volume to 100 mL with distilled water. 11. Alcohol (from 70 to 100% in distilled water). 12. GMA kit (Historesin, Leica). 13. Plastic mold for embedding. 14. Wooden pin holder. 15. Dentist acrylic resin kit: Mix the powder and the liquid to prepare a viscous medium. Just after this, pour the mixture into the mold holes and immediately put a wooden pin. The resin takes only a couple of minutes to polymerize. 16. Razor blade. 17. Glass knife: prepared with glass strips (400 × 25 × 6.4 mm) from Leica in an LKB Knifemaker, model 7880B. 18. Toluidine blue-borate: 1 g toluidine blue O (Allied Chemical) and 1 g of sodium borate (Na2 B4 O7 anidrous) dissolved in 100 mL distilled water. 19. Erythrosine-Orange-Toluidine: solution A (0.2 g erythrosine, 1 g orange G in 100 mL distilled water); solution B (toluidine blue-borate, detailed above at item 18).

6

Chiarini-Garcia, Parreira, and Almeida

20. Harris’ hematoxylin: mix 1 g hematoxylin, previously dissolved in 10 mL ethanol, with 20 g aluminum potassium sulfate (AlKO8 S2 12H2 O) previously dissolved in 200 mL of heated distilled water. Immediately, add 0.5 g mercury oxide (HgO). Take out from the heater and cool the solution by immersing it cold water. To increase the nuclear contrast, 4% acetic acid can be added to the solution. 21. Mordent solution: add 2% solution of ammonium iron sulfate (FeH8 N2 O8 S2 .6H2 O) in distilled water. 22. Eosin solution: mix 1 g yellow eosin dissolved in 10 mL absolute ethanol in 0.5 potassium dichromate (K2 Cr2 O7 ) dissolved in 80 mL distilled water, followed by 10 mL of saturated solution of picric acid (for saturation, add 1.4 g picric acid in 100 mL distilled water). 23. Period acid: periodic acid at 0.5% in distilled water. 24. Differentiator solution: mix 6 mL of 10% sodium metabisulfite (Na2 O5 S2 ) and 5 mL of chloride acid 1 N (8.35 mL HCl up to 100 mL distilled water) in distilled water and complete the volume up to 100 mL. 25. Schiff reactive: dissolve 1 g basic fuchsin in hot water but without boiling it. Wait to cool down to 50◦ C and then add 10 mL of chloride acid (1 N). Wait to cool down to 25◦ C and add 1 g of sodium metabisulfite. Mix for 1 h and keep it in a dark place for 24 h at room temperature. Keep the final solution at 4◦ C. 26. 5-Bromo-2-deoxyuridine (Sigma) – diluted 6 mg/mL in phophate buffer solution PBS. 27. Colorfrost Plus Microscope Slides (Fisher-Scientific). 28. 0.6% Hydrogen Peroxide (H2 O2 ): Dilute 1 mL of 30% H2 O2 (Sigma-Aldrich) in 49 mL of distilled water. 29. 0.1% Protease (Sigma-Aldrich) diluted in PBS. 30. 10× phosphate-buffered saline: Dissolve 76.0 g NaCl, 3.6 g NaH2 PO4 and 9.94 g Na2 HPO4 in 1000 mL of distilled water. To make 1× PBS, dilute 10× PBS at a 1:9 ratio in distilled water. Adjust to pH 7.4 with HCl. Store in glass bottles at room temperature. 31. 2 N hydrogen chloride (HCl): Dilute 73 mL of HCl to 1 L of solution in distilled water. Store the solution in a glass bottle at room temperature. 32. 0.1 M sodium borate (Na2 B4 O7 ): Dissolve 38.14 g of Na2 B4 O7. 10H2 O (decahydrate) in distilled water, up to 1 L. Mix it in a hot plate, until the salt is completely dissolved. Store the solution in a glass bottle at room temperature.

GMA for Improved Investigations Under Light Microscopy

7

33. Triton X-100 (Sigma-Aldrich). 34. PBST solution: 0.2% Triton X-100 in 1× PBS. Add 200 µL of Triton X-100 to 1 L of PBS. Store the solution in a glass bottle at room temperature. 35. Normal horse serum (Sigma-Aldrich) diluted in PBST . 36. Primary antibody anti-BrdU B44 (BD Biosciences). 37. ABC kit, Mouse IgG (Vector Labs): This kit contains secondary biotinylated antibody and reagents A and B. 38. Normal goat serum (Sigma-Aldrich), diluted in PBST. 39. DAB (3,3′ -diaminobenzidine) kit (Vector Labs). 40. Xylene. 41. Entelan medium (Merck). 42. Coverslip.

3. Methods To obtain a good tissue preparation, all the processing steps have to be carefully developed. The sum of minimal defect in some steps can impair the final results. Thus, care has to be taken regarding fixation, embedding, sectioning, and staining. Below, details about each of these steps will be described considering different experimental possibilities. 3.1. Fixation and Storage

Glutaraldehyde is a non-coagulant fixative known to cross-link protein, preserving very well cellular structures. As protein is a universal component of cells, found in membranes and in the cytosol, the glutaraldehyde can fix it as a whole, making the cell as a single interlocking structure (27). Besides, this aldehyde reaction is not limited to protein. It can also react in a lower degree with lipids, carbohydrates, and nucleic acids. Formaldehyde is also another aldehyde used for cell fixation. However, it makes less cross-linking, reducing the tissue meshwork. Considering that the rate of penetration of glutaraldehyde into tissues is very low and the formaldehyde penetration is about five times faster than glutaraldehyde, a combination of both is frequently used, mainly in tissues of difficult penetration and/or in fixation by immersion. This method was described by Karnovsky (28). Fixatives using picric acid and alcohol are coagulant. They can denaturate proteins, permanently modify their structure and affect the tissue resolution. Thus, for morphological evaluation of testes under light microscopy, aldehydes are normally chosen. To avoid lower pH reduction during the fixation procedure and the introduction

8

Chiarini-Garcia, Parreira, and Almeida

of artifacts, a buffering system should be used with the fixative. The main buffers used are phosphate and cacodylate buffer at pH 7.2–7.4. 3.1.1. Testes Fixation by Perfusion of Whole Body

When small animals are used, like rodents, marmosets, opossums or cats, the best method to preserve whole testes is fixing them by the injection of the fixative into the whole body through the circulatory system (11). This method of fixation by perfusion is reached by introducing a catheter into the left ventricle, in the direction of the aorta (Note 1). This needle is also connected to two vials, one with saline and the other with fixative, through a triway plastic pipe. These vials are positioned ∼1.2 m above the heart, reaching a liquid pressure of ∼80 mmHg inside the vascular system. Just before the beginning of the perfusion process, the right atrium is cut for draining the blood and solutions (saline and fixatives) that will be injected. First, the circulatory system is perfused with saline for 5–10 min, cleaning of blood cells (Note 2). Immediately after, it initiates the perfusion with the fixative for about 25–30 min (Note 3). Fifteen minutes before perfusion, heparin is injected intraperitonealy in the proportion of 125 IU/kg of body weight (Note 4). After perfusion, testes should be cut in thin slabs and fixed by immersion for 12–24 h at 4◦ C.

3.1.2. Fixation by Perfusion of Isolated Testis

When testes of large animals are used, like bull, boar, and ram, the whole body perfusion method becomes very expensive. In this way, the orchiectomy is made and only the testes are perfused. The perfusion is made by introducing the catheter into the testicular artery, once this artery is easily identified (Note 5). To avoid blood clumps during the perfusion process, heparin should be added to the saline solution in a proportion of 10 IU/L. The time of saline perfusion should be enough to clean the testis from blood cells (∼5 min) and can be verified visually by the clear liquid running from the testicular vein. The fixation time should be between 20 and 30 min (Note 3). Both, saline and fixative, should be perfused at a pressure of ∼80 mmHg. After perfusion, testes should be cut in thin slabs and fixed by immersion for 12–24 h at 4◦ C.

3.1.3. Testis Fixation by Immersion

If for some reason it is not possible to fix the testis by perfusion, they can be fixed by immersion. However, for good morphological preservation, some cares have to be taken. The use of a fixative solution with combined aldehyde (glutaraldehyde and formaldehyde) can be an alternative solution, once formaldehyde penetrates faster and temporarily stabilizes cellular structures. Glutaraldehyde, which penetrates slowly, arrives later and permanently stabilizes cellular components. In case of the use of glutaraldehyde only as a fixative, only thin slabs (∼1 mm thickness)

GMA for Improved Investigations Under Light Microscopy

9

should be cut from the fragment surface (Note 6). The fixative time by immersion should be for 12–24 h at 4◦ C. 3.1.4. Comments

Regardless of the method used, testes fragments can be kept in buffer after fixation at 4◦ C for a long time (Note 7). Testes fixed by perfusion are used to keep the seminiferous tubules together, preserving the interstitial tissues and possibly making more accurate morphometrical evaluation of testes compounds. Otherwise, testes fixed by immersion normally show seminiferous tubules dispersed. The artifactual spaces among them are provoked by the pressure in a soft tissue during the cut processing in small slabs. For immunohistochemistry evaluations, fixatives with glutaraldehyde are avoided once the resulted cross-link into tissues could block the antibody receptors. The most common fixatives used for immunohistochemical is the formaldehyde, which in spite of not preserving very well morphological details, keeps the tissue receptors exposed for antibodies.

3.2. Embedding

Testes fragments should be progressively dehydrated with alcohol at crescent concentrations (see Note 8), infiltrated and embedded in GMA, as described below: Ethanol 70% Ethanol 85% Ethanol 95% Absolute ethanol Infiltration resin Pure resin Embedding (see Note 9, Fig. 1.1)

3.3. Sectioning

30 min 30 min 30 min 30 min (2×) overnight overnight

The tissue embedded in GMA blocks can be cut from 0.25 µm up to 15 µm thickness (see Note 10). For the best sectioning, excessive resin should be trimmed with a razor blade, keeping a border of approximately 1–2 mm around all tissue. To obtain a nice section, glass knife can be used in a microtome. For distention, sections are floated in distilled water just after the collection at room temperature (see Note 11). The section should be picked up with a slide and transferred to a hot plate (70–80◦ C) until the water droplet over the section evaporates (Fig. 1.1).

3.4. Staining 3.4.1. For Morphological and Morphometrical Studies

For different histological evaluations, testes embedded in GMA can be stained by different methods, namely, toluidine blueborate, erythrosine-orange-toluidine, hematoxylin-eosin, and PAS-hematoxylin. As tissues embedding in GMA present weak staining, when compared with those embedded in paraffin

10

Chiarini-Garcia, Parreira, and Almeida

Mi

HP WB A

Re BL

T

GK

B

Mo

7

C

4

1

WP BL

D

Fig. 1.1. In a, microtome (Mi, Reichert Jung, model 1140/Autocut) setup with GMA block (BL) and glass knife (GK) for sectioning, evident in the circle (detail in B). Moreover, a room temperature water bath (WB) for section distension and a hot plate (HP) for drying and section adherence on the slide. In b, detail of the circle in a, showing also the tissue (T) facing the glass knife. In c, a slide with histological sections of 7, 4, and 1 µm thickness, showing the different staining intensity. d shows components used for GMA embedding: plastic mold (Mo), wooden pin (WP), dentist resin (Re), and tissue into the molds under polymerization (black arrows) and block with tissue (BL) after polymerization and with wooden pin adherence.

(13–15), the methods described below were modified from those originally used to stain testes embedded in paraffin, with the purpose of obtaining good staining and contrast and, consequently,

GMA for Improved Investigations Under Light Microscopy 7µm

4µm

1µm

s

AT

s

s

A

11

C

B

s

HE s

s

D

E

s

s

s

EOGAT

F

H

G

I s

s s

PAS

J

K

L

Fig. 1.2. Photomicrographies of the human seminiferous epithelium stained with toluidine blue-borate (a–c), hematoxilin-eosin (d–f), erythrosine-orange G-toluidine (g–i), and PAS-hematoxylin (j–l). These pictures show the relationship between resolution and section thickness. In the first column (a, d, g, j), sections were obtained at 7 µm and the spermatocyte (S) heterochromatin were not easily visible. In the second column (b, e, h, k), sections at 4 µm thickness allowed the observation of details from the spermatocyte (S) heterochromatin. One micrometer is the thickness of sections in the third column (c, f, i, l) and more details can be observed in the spermatocyte (S) nuclei. The PAS stained in purple grains in the cytoplasm of the germ cells, which were differently observed depending on the section thickness. While in J they were compactly observed (arrows), in k some grains can be seen and in L the PAS grains were individually observed. Bar: 10 µm.

best resolution under light microscopy (Fig. 1.2). All the staining methods presented below can be changed depending on the tissue, fixative applied, and the section thickness. Tests should be made to standardize them for each specific experimental approach. 3.4.1.1. Toluidine Blue-Borate

Although it presents just one color, exceptions for metachromatic cells/tissues (mast cells, goblet cells, mucus), it is elected as one of the best staining for tissues embedded in GMA for good contrast and resolution, mainly in black & white pictures (see Note 12, Fig. 1.2). 1. Put several drops of toluidine blue-borate over the sections on a slide for 1 min.

12

Chiarini-Garcia, Parreira, and Almeida

2. Rinse the slide with running water to clean the excessive staining (see Note 13). 3. Remove excess water by gently pressing the slide, with the tissue facing down, over a piece of filter paper. 4. Allow the slide to dry completely at room temperature and mount with a coverslip using any conventional medium. 3.4.1.2. Hematoxylin-Eosin

Even after some methodological modification, the H&E staining does not present very nice contrast in GMA-embedded sections. The contrast can be increased on thicker sections; however, the resolution can be impaired. In order to intensify the H&E staining, the use of a mordant solution (see Note 14) and the increase of the staining time (Fig. 1.2) are recommended. 1. Put several drops of mordent solution over sections on a slide for 10 min. 2. Rinse the slide with running water for 5 min. 3. Put several drops of hematoxilin solution over sections for 15 min. 4. Rinse the slide with running water for 5 min. 5. Put several drops of eosin solution over sections for 30 s. 6. Rinse the slide with running water to clean the excessive staining (see Note 13). 7. Remove excess water by gently pressing the slide, with the tissue facing down, over a piece of filter paper. 8. Allow the slide to dry completely at room temperature and mount with a coverslip using any conventional medium.

3.4.1.3. ErythrosineOrange-Toluidine

In paraffin, the trichrome stains are used for cytoplasmic stains combined with nuclear stains. However, these methods do not work in GMA. We have applied an alternative method which has been used with relative success for tissues embedded in GMA (Fig. 1.2). 1. Put several drops of solution A (erythrosine-orange) over sections on a slide for 10 min. 2. Rinse the slide with running water to clean the excessive staining. 3. Put several drops of solution B (toluidine blue-borate) over sections on a slide for 1 min. 4. Rinse the slide with running water to clean the excessive staining (see Note 13). 5. Remove excess water by gently pressing the slide, with the tissue facing down, over a piece of filter paper. 6. Allow the slide to dry completely at room temperature and mount with a coverslip using any conventional medium.

GMA for Improved Investigations Under Light Microscopy

3.4.1.4. PAS-Hematoxylin

13

Used for neutral amino-glycol localization in tissues. Although this method has been commonly used for acrosomal identification in the testis, here it was used to show PAS-positive granules into the cytoplasm of germ cells (Fig. 1.2). 1. Put several drops of periodic acid solution over sections for 20 min. 2. Rinse the slide with distilled water for 5 min. 3. Put several drops of Schiff reactive over sections for 60 min. 4. Rinse in three baths of differentiator solution in a total of 3 min. 5. Rinse the slide with running water for 30 min. 6. Put several drops of hematoxylin for 10 min. 7. Rinse the slide with running water for 30 min. 8. Remove excess water by gently pressing the slide, with the tissue facing down, over a piece of filter paper. 9. Allow the slide to dry completely at room temperature and mount with a coverslip using any conventional medium.

3.4.2. For Immunohistochemical Studies

Some immunohistochemical studies can be performed using GMA. As an example, 5-bromo-2-deoxyuridine (BrdU) method has been used to study cellular cycle in the testis (29). During cellular division, BrdU incorporates into the DNA chain and can be followed during the cellular cycle using antibody against BrdU. BrdU was injected intraperitonealy in a dose of 60 mg/kg of body weight one hour before killing the mice. The spermatogonia that divided during this time incorporated the BrdU in their nuclei. After one hour, the mice were fixed by perfusion and the testes embedded in GMA as described above. Testes sections of 5 µm thickness were used for the present evaluation. Examples of immunohistological staining of BrdU in spermatogonia are presented in Fig. 1.3.

A

B

Fig. 1.3. a and b: Sections of the seminiferous epithelium of a mouse showing immunohistochemical staining for BrdU in spermatogonia (arrows). In b, note the staining concentrated in the inner border of the envelope nuclear. Bar: 10 µm.

14

Chiarini-Garcia, Parreira, and Almeida

To perform the BrdU immunostaining in GMA, the following steps must be taken: 1. Put slides in distillated water for 1 min. 2. Prepare 0.6% H2 O2 in distilled water and immerse slides in a 50 mL Coplin jar, at room temperature, for 5 min. 3. Wash slides in distilled water for 1 min. 4. In a moisture chamber, incubate sections with 0.1% protease in 1× PBS, for 60 min, at room temperature. 5. Rinse slides two times with 1× PBS for 5 min each time. 6. Denature sections with 2 N HCl for 50 min, in a Coplin jar, at room temperature. 7. Neutralize the acid with 0.1 M Na2 B4 O7 for 2 min. 8. Wash slides two times in PBST , for 5 min each. Prepare the blocking solution while washing slides. 10% NHS (normal horse serum) in PBST is used as blocking solution, before the primary antibody incubation step. 9. Incubate slides with the blocking solution, in a moisture chamber, during 30 min at 37◦ C. While incubating with the NHS, prepare the 1◦ antibody (B44) in 10%NHS in PBST , following 1:200 dilution. 10. Pipet off the blocking solution from sections to be tested, keeping one section as a negative control. Add 100 µL of the antibody prepared to each section on the slide. Incubate slides with the 1◦ antibody for 60 min, at 37◦ C, in a moisture chamber. 11. After incubation with the B44, wash slides twice with PBST , for 5 min each. While washing slides, prepare the secondary biotinylated antibody in 10% NGS (normal goat serum) with PBST , following 1:1000 dilution. 12. Add 100 µL of secondary antibody to each section and keep it at 37◦ C, for 30 min, in a moisture chamber. Prepare the ABC reagent while incubating with the 2◦ antibody. Add 20 µL of Reagent A plus 20 µL of Reagent B in 800 µL 1× PBS. Keep the same proportion for all reagents when preparing more than 1000 µL. 13. Wash slides twice with PBST , for 5 min each. 14. Add 100 µL of the prepared Reagent ABC to sections and incubate at 37◦ C, for 30 min, in a moisture chamber. 15. Wash slides twice with 1× PBS, for 5 min each. Prepare the DAB while washing slides for the last time in 1× PBS. The DAB should be prepared following the manufacter’s instructions. When using Vector Labs DAB kit, add two drops of buffer, plus four drops of DAB, plus two drops of H2 O2 to 5 mL distilled water.

GMA for Improved Investigations Under Light Microscopy

15

16. Add one drop of DAB to each section, for 30 s, then wash in distilled water. Keep slides in distilled water while staining the other slides with DAB. 17. Stain the section with hematoxylin, diluted 1:1 in distilled water, for 1 min. Wash with tap water for 5 min. 18. Dehydrate slides using 95% ethanol and 100% ethanol for 3 min each and emerge slides in xylene for 5 min. 19. Mount slides with conventional mounting media and coverslip. 3.5. Photomicrography

Photomicrographies were obtained using a microscopy BX-51 in which a Q-Color 3 digital camera from Olympus was connected. The obtained images were transferred to a computer through the Image-Pro Express (Media Cybernetics) software and adjusted for resolution (1000 dpi), sharpness, contrast, brightness, and gray levels using Photoshop (Adobe System, Inc., Mountain View, CA). Plates were organized and characters added using the Adobe Illustrator software.

4. Notes 1. The circulatory system should be closed and under pressure for adequate entrance of the fixative in the aorta, reaching testes afterwards and exiting by the right atrium. The most common mistake during the perfusion is the needle placing. If a hole is made in the interventricular septum – a thin wall that divides the two ventricular chambers – the pressure comes down and the fixative will enter the right ventricle reaching the pulmonary circulatory system. This mistake will decrease the pressure inside the general circulatory system and impair testes fixation. 2. The effective time for saline perfusion is the one necessary for cleaning the blood vessels. When the saline that comes out of the right atrium is clean, it is time to stop the saline and start the fixative perfusion. 3. As the fixative penetrates the testis by the blood vessels, it reaches all testicular compartments and cells by diffusion. These processes take a long time and require a fixation time of at least 25 min, even if the animal body is apparently well perfused. 4. The use of heparin is very important for a successful testes perfusion success (30), as it avoids blood clumps into blood vessels during the fixative perfusion.

16

Chiarini-Garcia, Parreira, and Almeida

5. For testes fixation directly through the testicular artery, to avoid reflux of solutions the testicular artery has to be tied with a line, but not too strong to avoid cutting the vessel wall. 6. For testes fixation by immersion, the albuginea capsule has to be taken out once it is a dense connective tissue that avoids fixative penetration by diffusion. If it is not possible to take the albuginea completely out, small holes or cuts should be done on it, around the entire testis. Another alternative is to cut the testis in large slabs and put them in the fixative by immersion. After 12–24 h, only thin slabs from the surface (1–2 mm thickness) of the big slabs should be taken. The rest has to be discharged. Fixation by immersion requires a fixative volume of at least 30 times greater than the tissue volume, for adequate fixation. 7. If it is necessary to keep tissues in phosphate buffer for a long time, even under 4◦ C, add one drop of glutaraldehyde in the flask to avoid fungi growth. 8. If for any reason it is not possible to dehydrate with alcohol, the water can be eliminated by crescent concentration of resin in water, such as 50, 70, 80, 90% and finally pure resin. 9. Testes fragments should be smaller than the cutting surface of the block and centrally positioned (Fig. 1.1). During the cutting procedure, the border of blocks could be damaged, impairing the histological analyses of the whole tissues. After resin polymerization, a support should be attached to each resin block to firmly attach them in the microtome for sectioning. Originally aluminum support has been used. However, in our laboratory we have used wooden pins as support, which are attached to the resin block using dentist resin (Fig. 1.1). 10. The best section thickness depends on the type of the tissue and the researcher’s interest. When the research requires low magnification under microscopy (×2 to ×10 objectives), the tissue should be sectioned at a range thickness of 4 to 8 µm, once the stain intensity of the biological tissues is proportional to the section thickness. Otherwise, sections of 0.5–3 µm are adequate for high magnifications (×40 to ×100 objectives), once there is less over superposition of the cellular structures. As the correct thickness depends on the tissue, if it has more cells or connective tissues, we previously define the correct thickness by cutting sections of the same block from 1 up to 6 µm and collect them in a same slide. Afterwards, two thicknesses are chosen from them to develop the project, one thicker and the other thinner. We have frequently used slides with sections of 2 and 4 or 3 and 5 µm.

GMA for Improved Investigations Under Light Microscopy

17

11. During the sectioning procedure, the section should be taken individually with a forceps and laid down in a water with distilled water at room temperature. We should wait a couple of minutes for the section distension and collect it over a clean slide. 12. The green filter has been used in microscopy to increase the contrast of black and white micrographies. 13. When the staining is excessive or if it is necessary to take the stain out of the tissue, slides can be immersed several times and quickly in acid–water to clean tissues. Acid–water solutions are made with chloride or acetic acid at the proportion of 0.5–2%. Acid–alcohol solution should not be used once the alcohol wrinkles the GMA. 14. The mordant acts by increasing the electrostatic forces of tissue macromolecules, intensifying stains attachment.

Acknowledgments These methods were standardized during the development of different projects that were partially supported by Brazilian financial foundations (CAPES, CNPq, FAPEMIG, PRPq-UFMG). We thank Ana Luiza Drumond for helping in the immunohistochemistry processing. References 1. Clermont, Y., Perey, B. (1957) Quantitative study of the cell population of the seminiferous tubules of immature rats. Am J Anat 100, 241–268. 2. Clermont, Y., Perey, B. (1957) The stages of the cycle of the seminiferous epithelium of the rat: practical definitions in PA-Schiffhematoxylin stained sections. Rev Can Biol 16, 451–462. 3. Clermont, Y. (1962) Quantitative analysis of spermatogenesis of the rat: a revised model for the renewal of spermatogonia. Am J Anat 111, 111–129. 4. Clermont, Y., Bustos-Obregon, E. (1968) Re-examinations spermatogonial renewal in the rat by means of seminiferous tubules mounted ‘in toto’. Am J Anat 122, 237–247. 5. Russell, L. D., Clermont, Y. (1977) Degeneration of germ cells in normal, hypophysec-

6.

7.

8.

9.

tomized and hormone treated hypophysectomized rats. Anat Rec 187, 347–366. Chiarini-Garcia, H., Russell, L. D. (2001) High-resolution light microscopic characterization of mouse spermatogonia. Biol Reprod 65, 1170–1178. Chiarini-Garcia, H., Hornick, J. R., Griswold, M. D., Russell, L. D. (2001) Distribution of type-A spermatogonia in the mouse is not random. Biol Reprod 65, 1179–1185. Russell, L. D., Chiarini-Garcia, H., Korsmeyer, S. J., Knudson, C. M. (2002) Bax-dependent spermatogonia apoptosis is required for testicular development and spermatogenesis. Biol Reprod 66, 950–958. Chiarini-Garcia, H., Raymer, A. M., Russell, L. D. (2003) Non-random distribution of spermatogonia in rats: evidence of niches in

18

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Chiarini-Garcia, Parreira, and Almeida the seminiferous tubules. Reproduction 126, 669–680. Bolden-Tiller, O. U., Chiarini-Garcia, H., Poirier, C., Alves-Freitas, D., Weng, C. C., Shetty, G., Meistrich, M. L. (2007) Genetic factors contributing to defective spermatogonial differentiation in juvenile spermatogonial depletion (Utp14b jsd ) mice. Biol Reprod 77, 237–246. Nascimento, H. F., Drumond, A. L., França, L. R., Chiarini-Garcia, H. (2008) Spermatogonial morphology, kinetics and niches in hamsters exposed to short- and longphotoperiod. Int J Androl, 32, 486–497. Doi:10.1111/j.1365–2605.2008.00884.x. Chiarini-Garcia, H., Meistrich, M. L. (2008) High-resolution light microscopic characterization of spermatogonia, in (Hou, S. X., Singh, S. R., eds.), Germline Stem Cells, vol 450. Humana Press, Totowa, NJ, Methods in molecular biology, pp. 95–107. Chiarini-Garcia, H., Alves-Freitas, D., Barbosa, I. S., Almeida, F. R. L. C. (2009) Evaluation of the seminiferous epithelial cycle, spermatogonial kinetics and niche in donkeys (Equus asinus). Anim Reprod Sci, 116, 139–154. Doi:10.1016/j.anireprosci.2008.12.019. Bennett, H. S., Wyrick, A. D., Lee, S. W., McNeil, J. H. (1976) Science and art in preparing tissues embedded in plastic for light microscopy, with special reference to glycol methacrylate, glass knives and simple stains. Stain Technol 51, 71–97. Cole, M. B., Jr., Sykes, S. M. (1974) Glycol methacrylate in microscopy: a routine method for embedding and sectioning animal tissues. Stain Technol 49, 387–400. Woodruff, J. M., Greenfield, S. A. (1979) Advantages of glycol methacrylate embedding systems for light microscopy. J Histotechnol 2, 164–167. Chiarini-Garcia, H., Machado, C. R. S. (1992) Mast cell types in the lymph nodes of the opossum Didelphis albiventris (Marsupialia, Didelphidae). Cell Tissue Res 268, 571–574. Chiarini-Garcia, H., Ferreira, R. M. A. (1992) Histochemical evidence of heparin in granular cells of Hoplias malabaricus Bloch. J Fish Biol 41, 155–157. Chiarini-Garcia, H., Pereira, F. M. A. (1999) Comparative studies of lymph nodes mast cell populations form five different marsupials species. Tissue Cell 31, 318–326.

20. Chiarini-Garcia, H., Santos, A. A. D., Machado, C. R. S. (2000) Mast cell types and cell-to-cell interactions in lymph nodes of the opossum Didelphis albiventris. Anat Embryol 201, 197–206. 21. Santos, A. A. D., Chiarini-Garcia, H., Oliveira, K. R., Machado, C. R. S. (2003) Development of different mast cell types in the opossum Didelphis albiventris. Anat Embryol 206, 239–245. 22. Rocha, J. S., Chiarini-Garcia, H. (2007) Mast cell heterogeneity between two different species of Hoplias sp (Characiformes: Erythrinidae): response to fixatives, anatomical distribution, histochemical contents and ultrastructural features. Fish Shellfish Immun 22, 218–229. 23. Paula, T. A. R., Chiarini-Garcia, H., França, L. R. (1999) Seminiferous epithelium cycle and its duration in capybaras (Hydrochoerus hydrochaeris). Tissue Cell 31, 327–334. 24. Neves, E. S., Chiarini-Garcia, H., França, L. R. (2002) Comparative testis morphometry and seminiferous epithelium cycle length in donkey and mules. Biol Reprod 67, 247–255. 25. Amaral, D., Chiarini-Garcia, H., Vale Filho, V. R., Allen, W. R. (2004) Effects of formalin and bouin fixation upon the mare’s endometrial biopsies embedded in plastic resin. Braz J Vet Anim Sci 56, 7–12. 26. Melo, R. C. N., Rosa, P. G., Noyma, N. P., Pereira, W. F., Tavares, L. E. R., Parreira, G. G., Chiarini-Garcia, H., Roland, F. (2007) Histological approaches for highquality imaging of zooplanktonic organisms. Micron (Oxford) 38, 714–721. 27. Bozzola, J. J., Russell, L. D. (1999) Electron Microscopy: Principles and Techniques for Biologists, 2nd edn. Jones and Bartlett, Sudbury, MA. 28. Karnovsky, M. J. A. (1965) A formaldehydeglutaraldehyde fixative of high osmolarity for use in electron microscopy. J Cell Biol 27, 137A–138A. 29. Shuttlesworth, G. A., de Rooij, D.G., Huhtaniemi, I., Reissmann, T., Russell, L.D., Shetty, G., Wilson, G., Meistrich, M.L.. (2000) Enhancement of a spermatogonial proliferation and differentiation in irradiated rats by gonadotropin-releasing hormone antagonist administration. Endocrinol 141, 37–49. 30. Russell, L. D., Ettlin, R.A., Sinha Hikim, A. P., Clegg, E. D. (eds.) (1990) Histological and Histopathological Evaluation of the Testis. Cache River Press, Clearwater, IL.

Chapter 2 Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis Gerluza Aparecida Borges Silva, Adriana Moreira, and José Bento Alves Abstract It is possible to obtain histological preparation of teeth and periodontium with satisfactory levels of quality by means of routine histological techniques, since specific cares are implemented during the sample processing. The formation of access ducts for the quick penetration of the fixative solution, the complete removal of the demineralizing agent and the increase of the time of dehydration, clearing, and paraffin embedding are some of these cares. A variety of fixing and demineralizing solutions have been proposed in the literature for teeth and periodontium processing. The author’s’ experience along the years demonstrated the possibility of satisfactory results with 10% buffered neutral formalin as fixative solution and 10% pH 7.3 EDTA as demineralizing solution. Sections of 6 µm in thickness obtained from paraffin-embedded samples, stained with hematoxylin and eosin, comply with the most morphological and morphometric evaluations. Besides, this routine protocol allows the use of serial sectioning for more specific techniques such as histochemical and immunohistochemical analyses, which are suitable for cellular constituent and extracellular matrix evaluation of teeth and periodontium. For the study of mineralized phases of isolated human teeth, ground sections can be obtained by the cutting–grinding technique. Though it is a recognized method of study, there are some technical difficulties involved, which are little exploited in the literature. This chapter presents a detailed cutting–grinding protocol for the histological evaluation of undecalcified isolated teeth and routine histology, which can be easily reproduced in any research or teaching support laboratory. Key words: Teeth, periodontium, histological processing, buffered neutral formalin, EDTA, cutting–grinding technique.

1. Introduction The light microscopy evaluation of teeth and periodontium is often performed by means of two basic histological techniques: 1. Cutting–grinding technique: a method for the study H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_2, © Springer Science+Business Media, LLC 2011

19

20

Silva, Moreira, and Alves

of undemineralized samples and 2. Routine histological technique for the study of demineralized teeth and periodontium. The cutting–grinding technique is an appropriate method in the evaluation of undecalcified isolated teeth, bone, and other hard tissues through macroscopic and light microscopic investigations. The fact that teeth and periodontium have both tissues of very different consistencies, constituted by mineralized connective tissues juxtaposed to the loose connective tissues, makes more complex the histological processing of these samples. Artifacts, as displacements, can be easily produced and may interfere in samples’ evaluation. Some researchers have been using the cutting–grinding technique for simultaneous evaluation of soft and hard tissues of teeth, periodontium, bone, and bone-anchored implants with good histological results (1–5). However, when the study refers only to evaluation of mineralized portion of isolated teeth, the cutting–grinding technique is a simple method that allows results with excellent quality for histological and morphometrical analysis. Though the cutting–grinding technique is a recognized method of study, there are some technical difficulties involved, which are little exploited in the literature. This chapter presents a detailed cutting–grinding protocol for isolated human teeth that can be easily reproduced in any research or teaching support laboratory. The soft tissues and organic matrix present in the teeth and periodontal structures can be evaluated from demineralized sections, prepared according to routine histological technique. A large variety of histological methods for demineralized samples have been suggested in the literature, with use of different associations between fixing and demineralizing solutions (6–14). However, their protocols are poorly described, making its repeatability as well as their comparative analysis difficult. The quality of histological sections is conditioned to the technique selected and the incorporation of some specific cares during histological processing of samples, aiming at the simultaneous preservation of tissues that constitute both the teeth and periodontium. Observations related to several cellular events, such as presence of inflammatory infiltrate and resorption processes, require material of excellent histological quality obtained by standardized and reproducible technique. At least four factors must be considered for the selection of the histological method: (a) the case urgency, (b) the tissue mineralization stage, (c) the purpose of the research, and (d) the staining technique that will be used (14). For example, the more rapid the decalcifier, the more injurious are its effects on subsequent staining likely to be. The effect is most noticeable in nucleic acids and manifests itself chiefly in the failure of nuclear chromatin to take up hematoxylin and basic dyes as readily as undecalcified soft tissues (15).

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

21

Our laboratory experience along the years demonstrated that histological technique considered as routine (use of 10% buffered neutral formalin as fixative solution, 10% EDTA as demineralizing solution, paraffin embedding and hematoxylin–eosin staining) allow to obtain satisfactory histological sections to the most qualitative and morphometric evaluations of teeth and periodontium. The staining with hematoxylin and eosin is the most commonly used for general purpose in the histological laboratory. Hematoxylin can be thought of being a basic dye when combined to aluminum, iron, copper, and tungsten salts, having an affinity to nucleic acids of the cell nucleus (16). It binds to acidic structures, structures yielding a blue-purple color. As such, the nucleus stains blue. Eosin is an acidic dye, which stains basic structures resulting from electrostatic combinations with tissues. The cytoplasm, collagen, and muscle are usually stained in red or pinkish red by eosin (17, 18). The staining with Gomori’s trichrome can be an alternative for the analysis of teeth and periodontium processes of repair, since it shows up the type I collagen – the main component of the organic matrix of these tissues. The expected results are collagen in blue and nuclei in blue to black (19, 20). Besides, the Gomori’s trichrome can be also indicated for the visualization of vascular alterations, such as hemorrhagic areas, hyperemia, and processes of vascular neoformations, since it enhances vessels and erythrocytes (cytoplasm in red), with a higher contrast than that of hematoxylin and eosin staining. This chapter presents, in details, protocols of cutting– grinding technique for undemineralized human teeth and routine histology for the analysis of demineralized samples, both standardized in our laboratory, with notes and orientations, which make possible its reproduction with predictable results.

2. Materials 2.1. Cutting–Grinding Technique: Mineralized Samples

1. Wax blocks for dental carving measuring 1.5 cm × 4 cm × 1.5 cm. 2. Carton cuts, scotch tape and petroleum jelly. 3. Synthetic rubber of white silicone 8001; room temperature vulcanizing (base and HS II catalyser) – Prepare a mixture with 3% of the catalyser. In the pattern, e.g., we used 100 g of silicone and 3 g of catalyzer, mixed with a wooden spatula to form a homogeneous mixture. Silicone 8001 has a mold durability of 2 years.

22

Silva, Moreira, and Alves

4. Crystal polyester resin (3061) prepared with 10% of styrene and one drop of hardener liquid (MEKP) for each 5 mL of solution (see Note 1). As a pattern, for three blocks of resin, we use 27 mL of resin + 3 mL of styrene + 6 drops of hardener liquid. 5. Pins for disposition of teeth during resin embedding. These pins can be made either of wood, plastic or acrylic. Examples: toothpicks cuts or cottontail rods. R 6. Rapid glue – Super Bonder .

7. Apparatus: (a) IsoMet 1000 – Precision saw with diamond wafering blade series 15 LC diamond (6′′ diameter × 0.220 – 152 mm × 0.5 mm)/Buehler-USA. (b) Motored polisher sanding machine. 8. Wet sand papers (600, 1000, and 1200 granulation/mm2 ). 9. Wood/aluminum blocks for supporting of dental sections during the final grinding. 10. Xylene. Inflammable: exhibits neurological effects. Can also cause irritation of the skin, eyes, nose, and throat. Requires fume hood for safe usage. 11. Glass slides and histological cover glass (24 × 50 mm). R 12. Mounting medium: Entellan .

2.2. Histological Processing for Demineralized Samples

1. Double-face diamond disk adapted in a low rotation equipment (microrectifier with flexible axle). 2. Discardable steel blades or surgical knifes. 3. 10% Neutral buffered formalin solution: (a) 36% Formaldehyde solution; Carcinogenic: toxic by inhalation, in contact with skin and if swallowed. Cause burns. May cause sensitization by skin contact. Use gloves to avoid bare hand contact. May cause heritable genetic damage. Use only in well ventilated areas. (b) Disodium hydrogen orthophosphate anhydrous – Na2 HPO4 – 0.65% w/v; (c) Sodium dihydrogen orthophosphate monohydrate – NaH2 PO4 .H2 O – 0.4% w/v. Prepare solution 0.65% w/v disodium hydrogen orthophosphate anhydrous and 0.4% w/v sodium dihydrogen orthophosphate monohydrate in distilled water. Stir up until obtaining a homogeneous solution. Add 36% formaldehyde solution – 10% v/v. Store it in a dark vial (see Note 2).

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

23

4. Ethylenediaminetetraacetic acid disodium salt 2-hydrate (EDTA). Prepare 10% w/v EDTA aqueous solution pH 7.2–7.4. Dissolve EDTA under stir in distilled water at 59◦ C. Wait until the solution reaches room temperature and adjust pH using sodium hydroxide pastilles. Complete with distilled water to final volume of solution. 5. Absolute ethanol – The preparation of alcohol solutions 70, 80, 90, and 95% by dilution in distilled water is performed by using a Gay-Lussac alcoholmeter. 6. Xylene. 7. Embedding agent for histology – pastilles-solidification point 56–58◦ C. 8. Microtome blades PTFE coated – Low profile blades. 9. Glass slides and cover glasses (24 × 50 mm). 10. Mayer’s albumin: Two white eggs, glycerin 87% and thymol. Beat the white eggs to resemble firm snow, then reserve for a 24 h period at 4◦ C, filter through filter paper and add glycerin 1:1 v/v. Add thymol crystals 0.1% w/v (see Note 3). 11. Staining methods: (a) Routine staining with hematoxylin and eosin: i. Harris hematoxylin solution – ready for use. ii. Putt’s eosin: eosin yellowish p.a.; potassium dichromate p.a.; ethanol p.a. and saturated picric acid solution (1.2%). Dissolve 1 g of eosin yellowish in 10 mL of ethanol. Dissolve 0.5 g of potassium dichromate in 80 mL of distilled water and add the eosin solution. Add 10 mL of saturated picric acid solution (see Note 4). (b) Staining for collagen fibers: Gomori’s trichrome i. Harris hematoxylin solution – ready for use. ii. Gomori’s trichrome solution: chromotrope 2R, fast green, phosphotungstic acid p.a., acetic acid p.a. Prepare the solution by dissolving 1.8 g of chromotrope 2R + 0.9 g of fast green + 1.8 g of phosphotungstic acid in 300 mL of distilled water. Heat up to dissolve. Wait until the solution reaches room temperature and add 3 mL of acetic acid. 12. Quantitative filter paper – rapid filtration/white band (15 cm diameter) R 13. Mounting medium: Entellan .

24

Silva, Moreira, and Alves

3. Methods 3.1. Cutting–Grinding Technique: Mineralized Samples

1. Preparation of the mold for silicone (see Note 5): Draw a mold in a carton sheet, in a proportional size to the number of resin blocks (see Fig. 2.1a–c). After drawing the box, the carton is cut with a scissor and walls fitted with adhesive tape (see Fig. 2.1d). 2. Glue one (or more) block of wax for dental carving inside of the carton box (see Note 6). Fix a little support (about 1.0 cm) on one of the surfaces of the block. This fixation is mechanically performed, only with a small pressure of the

A

B

C

D

E

F

G

H

I

J

K

L

Fig. 2.1. Sequence of stages for cutting–grinding technique: a histological method for mineralized dental samples. a–h Preparation of silicone mold for attainment of blocks in resin. i–k Preparation of teeth for embedding. l Dental sample embedded in crystal polyester resin.

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

25

pin on the wax surface. Such block(s) will act as mold for the cavity to be filled with resin in other step. The coupled pin generates an orifice that will work as a guide for the placement of teeth (see Fig. 2.1 h and j). Lubricate the bottom of the carton box with petroleum jelly before applying the silicone. 3. Wearing gloves, handle the silicone and throw it inside of the fitted box (see Fig. 2.1e), covering about 1 cm above the block of wax. Wait 24 h for rubber vulcanization (hardening). Dismount the box of paper (see Fig. 2.1f). Remove blocks of wax. 4. Preparation of teeth for embedding: (a) Glue a support with a measure such like that used in blocks of wax, on the palatine/lingual surface of teeth (see Fig. 2.1i). Such procedure is performed with rapid R glue (Super Bonder ) and aims at standardizing the embedding position (see Fig. 2.1 k). (b) Using a paint brush, overlay the tooth surface with the resin mixture before the embedding so as to reduce the surface tension. 5. Locate the tooth (teeth) on the center of cavity(ies) for embedding (see Fig. 2.1j, k). Prepare the solution of resin– styrene (see Note 1) and pour it into the silicone mold, in cavities molded by blocks of wax. 6. After a 24 h period of resin cure, withdraw the set resin + tooth from the silicone mold (see Fig. 2.1 l). 7. Make a guide for the block cutting, by drawing a dotted line on the resin surface passing along the axis of the embedded tooth (detail in Fig. 2.2c). With the aid of the cutting set IsoMet (slicing machine fitted with a diamondimpregnated cutting disc of 0.5 mm thick (see Fig. 2.2a, c)), regulated to the speed of 300 rpm with a loading of 100 g (see Note 7), one can section the block, under refrigeration (see Note 8). 8. Using a motorized polisher sanding machine (see Fig. 2.2b) – polish the halves attained, with the surface of teeth turned to the wet sand paper sheet (see Fig. 2.2d), using all the sequence of granulation (600, 1000, and 1200 granulation/mm2 ) (see Note 9). 9. Return with the block to IsoMet cutting set for a new cutting, thus obtaining a slice of about 1.0 mm thick (see Fig. 2.2e). 10. Glue the tooth slice onto a piece of aluminum or plane wood (support for gripping), with a mounting medium R Entellan (see Fig. 2.2f). The polished side, already treated, must be turned to the support, leaving free the

26

Silva, Moreira, and Alves

A

B

C

D

E

H

F

G

I

Fig. 2.2. a IsoMet 1000. b APL-04 motorized sand polisher machine. c First section of resin block and detaching of the tooth in two halves. d First finish stage of dental face in refrigerated wet sand paper sheets. e Second section of resin block for obtaining slices of 1 mm thick. f Gluing of slice onto an aluminum support. g Trimming of the slice and finishing in sand paper sheets 600, 1000, 1200. h Image of slice of the tooth imbedded in resin, after trimming in the polisher. i Ground section prepared from half of the tooth, using a cutting–grinding system. Section ready for mounting and analysis at light microscope.

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

27

side not yet submitted to the polishing. Keep inside of a drying kiln at 35–37◦ C during a 24-h period for resin polymerization. 11. Perform the same treatment described in item 8 (see Fig. 2.2 g), until obtaining a smooth surface and a thickness of 30 µm (see Fig. 2.2 h) that allows the microscope light to pass through the specimen (see Note 10). 12. Set free the cut obtained, by immersing in xylene the surface glued onto the wood. 13. Wash dental sections in two baths of absolute ethanol, under stir or ultrasonic cleaning to remove debris. Immediately after, immerse in two baths of xylene (5 min each). 14. Mount the dental cuts on glass slides, with mountR ing medium Entellan , overlapping a cover glass (see Fig. 2.2i). Transfer slides to a drying kiln (35–37◦ C) during a 24–48 h period. 15. Evaluate at light microscope with partial closing of condenser diaphragm. Results are shown in Fig. 2.3. 3.2. Routine Histological Processing of Demineralized Samples 3.2.1. Samples Collection

1. Rats maxillae fragments: After decapitation of animals by guillotine, heads must be rapidly dissected for skin removal, separation of maxillae and brain removal in order to facilitate the access of the fixative solution to the remaining tissues. 2. Isolated human teeth: Immediately after exodontics, the radicular apex (apices) must be removed aiming at enlarging the apical orifice to facilitate the access of the fixative solution to pulpal tissue. Apicectomy must be performed under refrigeration with new double-face diamond disk.

3.2.2. Sample Fixation and Demineralization

1. Keep specimens in 10% buffered neutral formalin, into dark vials, of large mouth, at 4◦ C, for 24 h (see Note 11). After this period of time, replace the fixative solution and keep samples at room temperature for 24 h (see Note 12). The volume of the fixative solution must be at least 20 times larger than the tissue to be fixed (21, 22). 2. Wash in running water (3 × 10 min). 3. Demineralization with 10% EDTA aqueous solution (see Note 13) in a shaker at room temperature. The decalcifying solution is changed every two days. Demineralization is controlled by means of superficial cuttings performed with a

28

Silva, Moreira, and Alves

Fig. 2.3. Light photomicrographs of human permanent teeth sections obtained by the cutting–grinding technique. a Ground section of a canine. b Image of the tooth crown showing the Enamel (E); Mantle Dentin (MD); Dentinal tubules (DT), and Interglobular dentin (arrows). c Dentin (D); Incremental lines (arrows). d Dentin (D); Enamel (E); Dentin-enamel Junction (DEJ). e Pulp chamber (PC); Tertiary dentin (arrow). f Secundary dentin (arrow). g Dentinal tubules (thin arrows); Enamel lamella (thick arrows); h Dentinal canalicules (tubules) in the radicular wall (RC); Sharpey’s fiber (arrows); Acellular cementum (∗ ). i Root dentin (RD); Cellular cementum (∗ ). j Granular layer of Tomes (arrows); Cellular cementum (∗ ). Bars = 400 µm.

steel blade in fragments edges. The attainment of cuts of firm texture, but without resistance to the steel blade indicates the ideal point of decalcification. After 15 days of demineralization, it is possible to perform the trimming of samples – reduction of fragments locating the area of interest, trimming fragments so as to standardize the embedding position

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

29

Fig. 2.4. Schematic drawing showing the preparation of maxillae of rats for histological evaluation of demineralized sections of molars and periodontium in longitudinal plane. a After initial demineralization stage in EDTA, maxillae are reduced by means of two cuts: Section 1. Crosscut with discard of pre-maxilla. Section 2. Reduction of the palatal faces in a plane parallel to the imaginary line traced over molars occlusal surfaces. b The inclusion of fragments, with palatal surface turned to the plane of microtomy (arrows), allows the simultaneous visualization of teeth (crown and roots) and periodontium. Due to tipping of first molar mesial root of rodents, it is recommended to maintain part of the diastema region (∗ ), so as to make possible the visualization of this root and associated periodontal structures.

(Fig. 2.4). Samples of human teeth also can be reduced, cut or divided after 30 or 45 days of demineralization, depending on dental type. After this preparation, it is advisable to keep fragments in the demineralizing solution for a week more, in case of maxillae, and for two more weeks in case of a human teeth, in order to assure the complete demineralization of samples. 3.2.3. Tissue Processing for Paraffin Embedding and Microtomy

1. Removal of EDTA: wash in running tap water (3 × 30 min + 1× 14–18 h). 2. Dehydration. 3. Clearing. 4. Paraffin infiltration: see Table 2.1. 5. Embedding in paraffin. 6. Microtomy: Sections of 6 µm collected in glass slides covered by Mayer’s albumin. Allow it to dry inside a kiln (35–37◦ C) for 24 h before staining.

3.2.4. Staining

It is possible to perform staining in serial sections with hematoxylin and eosin (H&E) and Gomori’s Trichrome, according to protocols described in Table 2.2. Figure 2.5 shows results obtained with histological routine staining technique (H&E) and Gomori’s trichrome staining.

30

Silva, Moreira, and Alves

Table 2.1 Protocol for dehydration, clearing and paraffin infiltration of maxillae of rat and isolated human teeth Maxillae of rat

Isolated human teeth

–

1 × 30 min

70%; 80%; 90% ethanol

2 × 30 min

2 × 30 min

Absolute ethanol

3 × 30 min

3 × 60 min

1 × 20 min

3 × 20 min

2 × 10 min

(see Note 14)

Xylene: paraffin (1:1)

–

1 × 30 min

Paraffin

3 × 30 min

3 × 60 min

Step 2. Dehydration – series of alcohol solutions 50% ethanol

3. Clearing Xylene (In exhaustion hood, with gloves) 4. Infiltration in paraffin (inside kiln 58◦ C)

For evaluation of processing quality, some parameters can be taken as reference: (a) preservation of odontoblasts layers without displacement of predentin (Fig. 2.5a–c); (b) preservation of endothelial wall of blood vessels (Fig. 2.5a); (c) preservation of collagen fibers and extracellular matrix (Fig. 2.5a–c, h); and (d) bone trabeculae with osteocytes and endosteum layer well preserved (Fig. 2.5i).

4. Notes 1. Formation of small blisters in the preparation of resin is common due to its viscosity. The incorporation of styrene to the resin formula induces to lower viscosity and, as a result, smaller quantity of blisters in the mixture. Styrene prolongs the resin cure stage a little, allowing blisters to flow together and reach the surface so as to be eliminated before resin hardening. 2. The formaldehyde decomposes itself easily in formic acid by light action and may negatively interfere in the staining of sections. The use of buffers neutralizes the action of these acids (23). For this reason, 10% neutral buffered formalin solution is preferably used and must be stocked into amber vials, for protection against luminosity that can favor acid formic formation.

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

31

Table 2.2 Protocols for routine staining (hematoxylin and eosin) and for type I collagen fibers (Gomori’s trichrome) Step

H&E

Gomori’s trichrome

1. Deparaffinization Xylene Xylene

1 × 30 min 2 × 15 min

1 × 30 min 2 × 15 min

2. Hydration Absolute ethanol 90% ethanol 80% ethanol 70% ethanol Running tap water

3 × 2 min 1 × 2 min 1 × 2 min 1 × 2 min 1 × 20 min

3 × 2 min 1 × 2 min 1 × 2 min 1 × 2 min 1 × 20 min

Filtered Harris hematoxylin solution 1 × 1 min

Filtered Harris hematoxylin solution 1 × 1 min

4. Running tap water

1 × 20 min

1 × 20 min

5. 2ª staining solution

Putt’s eosin solution 40 s to 1 min (see Note 15)

Gomori’s Trichrome solution 1 × 15 min (see Note 16)

6. Stain washing Running tap water

(10–20 s)

(10–20 s)

7. Dehydration 70% ethanol 80% ethanol 95% ethanol Absolute ethanol

1 × 10 s 1 × 10 s 1 × 30 s 3 × 2 min

– – 1 × 30 s 3 × 2 min

8. Clearing Xylene Xylene

2 × 2 min 1 × 10 min

2 × 2 min 1 × 10 min

R Cover slip using Entellan mounting medium

R Cover slip using Entellan mounting medium

3. 1ª staining solution

9. Mounting (see Note 17)

3. The glycerin prevents the slide from completely drying while the egg albumin, a protein, is denatured when immersed in 70% ethanol and fixes sections onto the glass slide. The addition of thymol crystals allows the solution preservation, without contamination by fungi, at 4◦ C even for 12 months. 4. Putt’s eosin solution intensely stains the acidophilic structures when recently prepared. Dilution 1:1 in distilled water is recommended to reduce the staining intensity. 5. The silicone form can be prepared either for one detached block or can be amplified with several cavities for simultaneous embedding of several teeth. Preparation of the resin

32

Silva, Moreira, and Alves

Fig. 2.5. a–e Dentin-pulp complex of human tooth. a Panoramic view. HE. Bar = 140 µm. b Layers of pulp. HE. Bar = 50 µm. c Dentin-pulp complex stained with Gomori’s trichrome. Bar = 160 µm. d Collagen of dentin bridge (arrowheads) induced by direct pulp capping with calcium hydroxide. Bar = 140 µm. e Presence of extravasated erythrocytes (arrow) in inflammated dental pulp. Gomori’s trichrome. Bar = 160 µm. f Components of periodontium stained with Gomori’s trichrome. Bar = 350 µm. g Panoramic view of the first and second molars region in a longitudinal section of rat maxilla. HE. Bar = 600 µm. h Higher magnification of rat periodontium. HE. Bar = 80 µm. i Resorption area of alveolar bone in higher magnification. Osteoclasts (arrows). HE. Bar = 80 µm. Dentin (D); Predentin (Pd.); Vessel (V); Odontoblast layer (Od); Cell-free zone (CFZ); Cell-rich zone (CRZ); Central layer (CL); Pulp (P); Root dentin (RD); Alveolar bone (AB); Periodontal Membrane (PM) or (∗ ); Interradicular septum (IS).

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

33

for the embedding of several teeth is desirable, since the proportion of the resin components for a larger volume is more reliable. 6. The wax blocks can be replaced by wood blocks prepared in the same dimensions or larger than the wax blocks. In this case, they must be lubricated with petroleum jelly before applying the silicone. 7. Faster speed as well as heavier weight tends to damage the specimen surface. With such careful management (200 or 300 rpm and 100 g), the detailed structure of specimens will be preserved from damage. 8. The cut of resin blocks can be handmade by using a doubleface diamond disc, in a mandrel, adapted either to an electric micromotor or to a microrectifier with flexible axle. 9. Wear and finishing process of dental sections embedded in resin can be either manually performed or by means of other types of electric polisher sand machines (orbital), since it allows the replacement of sand paper sheets. In the case of manual processing, it is important to fix the wet sand paper sheets onto a plane surface and moisten them constantly. 10. The thickness of sections is variable. It is important to obtain a homogeneous, semi-transparent, whitish texture. It has been reported that 100 µm thick sections are satisfactory for light microscopy (24). 11. The buffered neutral formalin is considered a universal fixative, with good results for most tissues (25). Formalin is not significantly harmful to any tissue types and for this reason it allows viewers to observe the tissue in its almost natural state. 12. Samples can be stocked in buffered neutral formalin for several days without significant tissue alterations for the routine histological analyses. However, for both histochemical and immunohistochemical techniques, the fixation of specimens for at the most 72 h is recommended. We have observed that a 48 h period is sufficient for a good fixation and this information is in agreement with the literature (14). 13. A variety of methods has been recommended for decalcifying the hard tissues. Decalcification can be performed through immersion into acids (weak or strong solutions) or into chelant components. All acids, in some way, interfere with the tissue stability, in reliance on the acidity of solution and the time of sample demineralization. In whole tooth specimens, the distortion of the pulp tissue can be easily demonstrated if a stronger decalcifying agent is used.

34

Silva, Moreira, and Alves

The removal of calcium by the acid produces carbon dioxide. The pressure of this gas and its movement through the tissue can be one factor causing the separation of connective tissues observed after decalcification (26). The EDTA is a chelant substance that captures metallic ions, among which is calcium, removing them slowly with excellent preservation of histological details (16). Hard tissues, particularly dental tissues, decalcified by chelators (e.g. 10% EDTA neutralized with sodium hydroxide, pH 7.2), show a minimum of artifact. Staining is of an acceptable standard after using this chelating agent and good results can be obtained (17). Although these chemical agents are the slowest, it preserves cells and matrix constituents, particularly mucopolysaccharides (27), and gives the best antigenic/enzymatic preservation. In case of urgent dental and periodontal analysis, we suggest the utilization of: (a) Planck Rychol’s solution (mixture composed by 7% aluminum chloride (w/v); 8.5% hydrochloric acid (v/v); and 5% formic acid (v/v) or (b) Anna Morse’s solution (mixture composed by 50% formic acid solution (v/v) and 20% sodium citrate (w/v) aqueous solution, used 1:1 (14, 28). Good results can be reached, but samples must be monitored every day, because these acid solutions demineralize in a short time, depending on the size of samples. There are risks of irreparable degradation of the organic matrix, in case the acid solution is kept in a higher time than that required for demineralization of hard tissues of tooth and periodontium. 14. The third bath of xylene, particularly for dental samples, can extend for even 2 h, for total clarifying of samples. Dental tissues clarify very well thus coming to the diaphanous aspect; therefore, any whitish spot inside the sample is a sign of the requirement of extending the clarification period. Nevertheless, the monitoring of the third bath is extremely significant, because the excessive time in xylene can harden the piece and make the microtomy stage difficult. 15. After the eosin staining and rapid washing in running tap water, one can proceed the regressive staining (differentiation), when the tissue presents itself extremely stained. Regressive staining consists in dipping slices rapidly (for some seconds) in 1% chloridric acid solution in 70% ethanol and washing in running tap water for 2–3 min. 16. Formalin-fixed tissues may benefit from secondary fixation of sections in Bouin’s fluid, which enhances the red color of muscle fibers and epithelial cells in staining with trichromics (29). The Bouin’s fluid considerably enhances trichrome

Histological Processing of Teeth and Periodontal Tissues for Light Microscopy Analysis

35

intensity and brilliance due the action of picric acid in the tissue. After hydration step, plunge sections into Bouin’ solution heated at 56◦ C (saturated picric acid 750 mL; formaldehyde 250 mL; glacial acetic acid 50 mL) for 1 h. Wash slides in running tap water for 10 min before application of the advocated stains for Gomori’s technique (20). R 17. The Entellan is a rapid cure resin and polymerizes in a maximum of 24 h. Alternatively, one can opt for synthetic Canada balsam due to the significantly lower cost when R compared to Entellan ; however, this medium requires about 4–5 days for polymerization.

References 1. Donath, K., Breuner, G. (1982) A method for the study of undecalcified bones and teeth with attached soft tissues. The SägeSchliff (sawing and grinding) technique. J Oral Pathol 11(4), 318–326. 2. Lan, W. H., Kwan, H. W., Sunada, I. (1986) Slicing technique for tooth specimens in histological preparation. Bull Tokyo Med Dent Univ 33(4), 129–136. 3. Rohrer, M. D., Schubert, C. C. (1992) The cutting-grinding technique for histologic preparation of undecalcified bone and bone-anchored implants. Improvements in instrumentation and procedures. Oral Surg Oral Med Oral Pathol 74(1), 73–78. 4. Günhan, M., Günhan, O., Celasun, B., Safali, M. (1996) Examination of periodontal tissues by a cutting-grinding technique. Aust Dent J 41(3), 173–175. 5. Cano-Sánchez, J., Campo-Trapero, J., Gonzalo-Lafuente, J. C., Moreno-Lopes, L. A., Bascones-Martínez, A. (2005) Undecalcified bone samples: a description of the technique and its utility based on the literature. Med Oral Pathol Oral Cir Bucal 10 Suppl 1, E74–E87. 6. Abreu, E. M. (1990) Processo de reparação de feridas de extração dentária, após implante de colágeno microcristalino: estudo histológico em ratos. Rev Port de Est e Cir Maxilofac 31(2), 95–102. 7. D’Souza, N. R., Bachman, T., Baumgardner, K. R., Butler, L. M. (1995) Characterization of cellular responses involved in reparative dentinogenesis in rat molars. J Dent Res 74(2), 702–709. 8. Lekic, P., Sodek, J., Mcculloch, C. A. (1996) Osteopontin and bone sialoprotein expression in regenerating rat periodontal ligament and alveolar bone. Anat Rec 244(1), 50–58.

9. Lamano Carvalho, T. L., Bombonato, K. F., Brentegani, L. G. (1997) Histometric analysis of rat alveolar wound healing. Braz Dent J 8(1), 9–12. 10. Terai, K., Takano-Yamamoto, T., Ohba, Y., Hiura, K., Sugimoto, M., Sato, M., Kawahata, H., Inaguma, N., Kitamura, Y., Nomura, S. (1999) Role of osteopontin in bone remodeling caused by mechanical stress. J Bone Miner Res 14(6), 839–849. 11. Gonçalves, E. L., Pavan, A. J., Tavano, O., Guimarães, S. A. C. (2002) Morphogenetic activity of demineralized dentin matrix: a study in dogs. Rev Fac Odontol Bauru 10(1), 51–56. 12. Silva, G. A. B., Lanza, L. D., Lopes-Júnior, N., Moreira, A., Alves, J. B. (2006) Direct pulp capping with a dentin bonding system in human teeth: a clinical and histological evaluation. Oper Dent 31(3), 297–307. 13. Sawada, T., Sugawara, Y., Asai, T., Aida, N., Yanagisawa, T., Ohta, K., Inoue, S. (2006) Immunohistochemical characterization of elastic system fibers in rat molar periodontal ligament. J Histochem Cytochem 54(10), 1095–1103. 14. Fernandes, M. I., Gaio, E. J., Rosing, C. K., Oppermann, R. V., Rado, P. V. (2007) Microscopic qualitative evaluation of fixation time and decalcification media in rat maxillary periodontium. Braz Oral Res 21(2), 134–139. 15. Bancroft, J. D., Stevens, A., Turner, D. R. (1990) Bone, in Theory and Practice of Histological Techniques, 3rd edn. Ed. Churchill Livingstone, New York, NY, pp. 309–342. 16. Tolosa, E. M. C., Rodrigues, C. J., Behmer, O. A., Freitas-Neto, A. G. (2003) Técnica histológica, in Manual de Técnicas para

36

17.

18.

19. 20.

21.

22.

Silva, Moreira, and Alves Histologia Normal e Patológica, 2nd edn. Ed. Manole, Barueri, SP, pp. 19–86. Bancroft, J. D., Cook, H. C. (1994) Principles of tissue demonstration and routine morphological staining, in Manual of Histological Techniques and Their Diagnostic Application. Ed. Churchill Livingstone, New York, NY, pp. 425–428. Burkit, H. G., Stevens, A., Lowe, J. S., Young, B. (1996) Notes on staining techniques, in Wheater’s Basic Histopathology: A Colour Atlas and Text. Ed. Churchill Livingstone, New York, NY, p. 277. Gomori, G. (1950) A rapid one-step trichrome stain. Am J Clin Pathol 20(7), 661–664. Behmer, O. A., Tolosa, E. M. C., FreitasNeto, A. G. (1976) Coloração do tecido conjuntivo, in Manual de T´ecnicas para Histologia Normal e Patol´ogica. Ed. EDART – USP, São Paulo, pp. 109–131. Michalany, J. (1980) Fixação, in T´ecnica Histol´ogica em Anatomia Patol´ogica: com Instruc¸˜oes para o Cirurgi˜ao, Enfermeira e Citot´ecnico. Ed. EPUP, São Paulo, pp. 40– 51. Junqueira, L. C. U., Junqueira, L. M. M. S. (1983) Fixação e descalcificação, in T´ecnicas ´ basicas de Citologia e Histologia, Instituto de

23.

24.

25. 26. 27. 28.

29.

Ciˆencias Biom´edicas e Faculdades de Medicina da USP, Ed. Santos, São Paulo, pp. 13–20. Burck, H. C., Carilla, P. C. (Transl.) (1969) Técnica Histológica: Manual para realizar preparaciones microscópicas en el laboratorio, Ed. Paz Montalvo, Madrid. Souza, E. M. D., Stott, G. G., Alves, J. B. (1999) Determination of age from cemental incremental lines for forensic dentistry. Biotech Histochem 74(4), 185–193. Grimaldi-Filho, G. (1981) Manual de Técnica Histológica. Fiocruz–Centro de Microscopia Eletrônica, Rio de Janeiro, 70p. Brain, B. E. (1966) The Preparation of Decalcified Sections. Charles C Thomas, Springfield, IL, pp. 69–135. Bélanger, L. F., Copp, D. H., Morton, M. A. (1965) Demineralization with EDTA by constant replacement. Anat Rec 153(1), 41–47. Morse, A. (1945) Formic acid-sodium citrate decalcification and butyl alcohol dehydration of teeth and bones for sectioning in paraffin. J Dent Res 24, 143–153. Tolosa, E. M. C., Rodrigues, C. J., Behmer, O. A., Freitas-Neto, A. G. (2003) Coloração para tecido conjuntivo, in Manual de T´ecnicas para Histologia Normal e Patol´ogica, 2nd edn. Ed. Manole, Barueri, SP, pp. 111–142.

Chapter 3 Large Plant Samples: How to Process for GMA Embedding? Élder Antônio Sousa Paiva, Sheila Zambello de Pinho, and Denise Maria Trombert Oliveira Abstract It is often necessary to process large plant samples for light microscopy studies, but due to structural characteristics of plant tissues, especially intercellular spaces, large vacuoles, and phenolic substances, results are often unsatisfactory. When large samples are embedded in glycol methacrylate (GMA), their core may not polymerize, remaining soft and moist and making it difficult to cut microtome sections. This situation has been erroneously interpreted as the result of poor infiltration, when the soft core of these samples is actually the result of incomplete polymerization. While GMA is in fact present inside samples, unsatisfactory polymerization results from rapid external polymerization that does not allow sufficient hardener to reach the sample core, while the relatively large volume of GMA inside the tissue block also dilutes the hardener. In this chapter we propose a new method for processing large plant specimens that avoids these problems by: (1) slowing the polymerization process through cooling in order to permit the penetration of hardener into the sample core and (2) increasing the hardener:GMA ratio to aid polymerization of the sample core. Key words: (2-Hydroxyethyl)-methacrylate, historesin, embedding, GMA, HEMA, plant anatomy, plant tissue, resin polymerization.

1. Introduction Paraffin has traditionally been used in routine histological work with animal and plant tissue samples, but resin embedding media commonly used in electron microscopy has been increasingly employed in light microscopy in the last few decades because it preserves morphological features far better than paraffin mixtures (1). This resin embedding media was developed to resist damage H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_3, © Springer Science+Business Media, LLC 2011

37

38

Paiva, Pinho, and Oliveira

by the electron beam and high vacuum, and to preserve cellular and subcellular details (1). The glycol methacrylate (GMA, synonymous with (2-hydroxyethyl)-methacrylate [HEMA]) (1) is currently employed for light microscopy purposes throughout the world. Among the advantages of the use of GMA as an embedding medium is the ease of preparing heterogeneous tissue sections (1). Glycol methacrylate embedding also has disadvantages, including difficulties encountered in embedding large samples. Instructions provided with the historesin kits are sufficient for standard samples, usually animal tissue that are relatively homogeneous, and when plant tissue samples with similar characteristics are used embedding is largely successful. Plant tissues generally have greater structural diversity; however, with unique physical and chemical properties they demand specific procedures and methodological adaptations to assure good results. Several tissues are found in plant organs, making them a mosaic of cells with different physical–chemical properties. Plant samples used for anatomical studies vary widely in their proportions of intercellular spaces, dimensions, the chemical contents of their cell vacuoles, and the mechanical properties of their cell walls. These characteristics can negatively affect procedures of dehydration, infiltration, and embedding. The most common problems encountered in processing plant samples are principally related to difficulties with GMA infiltration, although these can usually be minimized (or solved) by simply reducing the sample size. Sample size reduction, however, can make the posterior analysis of the material more difficult, and must often be avoided. In the specific case of plant tissues, the use of a vacuum in at least one of the processing steps will contribute to minimizing infiltration problems; the vacuum should preferably be applied during fixing as this procedure removes air from the intercellular spaces and quickly replaces it with fixative. When studies demand whole-organ interpretation (e.g. studies of floral vasculature, fruit and seed structural organization, and nodal anatomy), large samples are essential. These large samples require more time at each dehydration step and according to Feder and O’Brien (2), the specific dehydration time intervals will depend on the size and permeability of the specimen. In the case of infiltration and polymerization of the GMA, however, simple time adjustments alone may not be sufficient to solve the problem. This chapter presents solutions to some of the problems encountered in processing large plant samples that can significantly increase GMA polymerization efficacy.

Large Plant Samples: How to Process for GMA Embedding?

39

2. Materials 1. Ginger rhizomes (Zingiber officinale Roscoe): cubes (0.8 cm on each side) without periderm and cortical portions were prepared using a razor blade. 2. Formalin-aceto-alcohol (FAA): 50 mL of 37% paraformaldehyde, 50 mL acetic acid, and 900 mL of 50% aqueous ethanol. 3. Ethanol. 4. Histomolds (1 cm3 cells). 5. HEMA (LeicaTM Historesin kit). 6. Wood stubs and adhesive (we used epoxy-based glue – AralditeTM ). 7. Steel knife: Stainless Steel C-Profile Knife. 8. 0.1 M pH 6.8 Phosphate buffer (1.38 g NaH2 HPO4 .H2 O [0.1 M] in 100 mL of distilled water [solution A] and 1.42 g Na2 H.HPO4 [0.1 M] in 100 mL of distilled water [solution B]). To prepare the pH 6.8 buffer, 51 mL of solution A was mixed with 49 mL of solution B and the pH was adjusted with the same solutions. 9. Toluidine Blue O: (to prepare a 100 mL solution, add 0.05 g of Toluidine Blue O [Allied Chemical] to 0.1 M pH 6.8 100 mL phosphate buffer). 10. Glass slides. 11. Coverslips.

3. Methods 3.1. Fixation

Samples were obtained from mature ginger rhizomes. The periderm and the cortical portion were removed with a razor blade in order to prepare tissue cubes (0.8 × 0.8 × 0.8 cm) of the central cylinder region. This region presents fairly homogeneous structural characteristics, so sample size is a major obstacle to good embedding. Fifteen samples were employed in this experimental assay and these were fixed by immersion in FAA 50 (3) and submitted to moderate vacuum (180 mmHg) using a vacuum pump for 10 min in order to remove all air from the intercellular spaces and vessel lumens (see Note 1). Samples remained in the fixative for 48 h (see Note 2).

40

Paiva, Pinho, and Oliveira

3.2. Dehydration

3.3. Pre-infiltration, Infiltration, and Embedding Treatments

Samples were dehydrated in a 50–95% graded ethanol series. The dehydration stages were conducted at room temperature, as detailed below. It is important to emphasize that the times required for dehydration vary according to samples’ sizes and their densities. Due to difficulties involved in evaluating parameters related to absorption rates, we recommend at least 2 h for each dehydration step for large samples (∼1 cm3 ). All alcohol changes must be performed with care to avoid exposure to the open air and minimize air penetration into samples. Summary of the dehydration steps: Step

Duration

Conditions

50% ethanol

2h

Room temperature

70% ethanol (see Note 2)

2h

Room temperature

85% ethanol

2h

Room temperature

95% ethanol (see Note 3)

2h

Room temperature

After dehydration, samples were submitted to infiltration with GMA historesin; we only recommend the use of GMA activated by dibenzoylperoxide, as indicated in the historesin embedding kit. The alcohol is first substituted by pre-infiltration of a mixture of 95% ethanol: GMA (1:1) for several hours. This use of the alcohol: GMA solution as an intermediate step with plant samples facilitates infiltration and reduces problems related to resin viscosity (1). During the embedding stage when the hardener (dimethyl sulfoxide) is added to the GMA, resin polymerization time is short and positioning samples into histomolds must be done efficiently and quickly. When several samples are to be embedded, we suggest putting the histomold in an ice bath, which will delay polymerization and provide enough time to adjust positions of samples (see Note 4). Steps referred to here are crucial to processing large samples that frequently show embedding problems due to incomplete resin polymerization. We experimented with different preinfiltration/infiltration/embedding repertoires, altering times, temperatures, and hardener concentrations, with five repetitions (five samples) each: 1. Treatment 1 (T1) – The manufacturer’s instructions that accompany the resin kit (Leica Historesin Embedding Kit) suggested longer infiltration times for plant specimens, but details of each step are not specified. According to the manufacturer, an intermediate infiltration solution (95% ethanol: GMA, 1:1) is recommended to ensure even sample penetra-

Large Plant Samples: How to Process for GMA Embedding?

41

tion. Two hours of intermediate infiltration solution (preinfiltration) were employed here and samples were then transferred to a full-strength infiltration solution (GMA) for 24 h. The embedding solution used the GMA: hardener proportion suggested by the manufacturer (15 mL:1 mL) and it was placed into histomolds together with the tissue samples. All steps were conduced at room temperature. This treatment sought to establish the effectiveness of the standard protocol with large specimens. T1 Protocol:

Step

Duration

Conditions

Pre-infiltration (ethanol: GMA, 1:1)

2h

Room temperature

Infiltration (GMA)

24 h

Room temperature

Embedding (GMAa : hardener, 15:1)

∼30 min

Room temperature

a The resin volume must be sufficient to assure complete sample submersion in the histomold cell.

2. Treatment 2 (T2) – Increasing the pre-infiltration exposure time to 48 h, followed by extended cold infiltration in GMA (24 h in a refrigerator and 48 h in a freezer). Samples were kept for 48 h in a freezer to increase infiltration time under low temperatures. After this time period, embedding was proceeded using the GMA: hardener proportions suggested by the manufacturer (15 mL:1 mL). Samples were placed into histomolds and transferred to room temperature for polymerization. This treatment sought to determine effects of increasing infiltration time under cold conditions. T2 Protocol:

Step

Duration

Conditions

Pre-infiltration (ethanol: GMA, 1:1)

48 h

Room temperature

Infiltration (GMA)

24 h

Refrigerator (±5◦ C)

Infiltration (same solution as in previous step)

48 h

Freezer (±18◦ C)

Embedding (GMAa : hardener, 15:1)

∼30 min

Room temperature

a The resin volume must be sufficient to assure complete sample submersion in the histomold cell.

42

Paiva, Pinho, and Oliveira

3. Treatment 3 (T3) – Increasing the pre-infiltration exposure time to 48 h followed by cold infiltration in GMA (24 h under refrigeration) as in T2, plus cold infiltration of the embedding solution (48 h). The T3 protocol examined effects of prolonged cold exposure (more than 48 h) to the embedding solution (GMA + hardener). After this cold infiltration period, histomolds were transferred to room temperature for polymerization. It is important to carefully arrange samples in histomolds, as polymerization can be started during this stage since the hardener has already been added. This treatment sought to determine effects of correcting hardener volume by considering the amount of resin present in the tissue samples and extending GMA + hardener penetration time. T3 Protocol:

Step

Duration

Conditions

Pre-infiltration (ethanol: GMA, 1:1)

48 h

Room temperature

Infiltration (GMA)

24 h

Refrigerator (±5◦ C)

Infiltration: Embedding (GMAa : hardenerb )

48 h

Freezer (±18◦ C)

Embedding (same solution from previous step)

∼ 30 min

Room temperature

a The resin volume must be sufficient to assure complete sample submersion in the histomold cell. b The proportion of hardener must be calculated considering the resin volume outside the tissue sample plus 80% of the tissue sample volume (assuming that the sample is fully infiltrated with resin). In the present treatment, the sample has 0.512 cm3 (0.8 × 0.8 × 0.8 cm), which corresponds to a volume of 0.410 mL (80% of 0.512 mL) of resin inside each block (i.e. 2.05 mL of resin in the five blocks). This value must be considered in calculating the hardener volume.

3.4. Microtomy and Staining

The following methodology was employed: 1. After 24 h of polymerization in histomolds, the resin blocks were glued to wooden stubs (Fig. 3.1) using commercial epoxy glue (AralditeTM ). These blocks must be kept in a low-humidity environment; we recommend storage under silica gel. Using this procedure, samples can be maintained indefinitely before sectioning (see Note 5). 2. Transverse sections (10 µm thick) were obtained using a LeicaTM rotary microtome equipped with a steel knife (Cprofile). This microtome has a section counter and we discarded the first 400 sections, collecting four sections from the median region of each block. Twenty sections were therefore obtained from each treatment (five blocks, with four sections from each block).

Large Plant Samples: How to Process for GMA Embedding?

43

Fig. 3.1. Blocks obtained from the three treatments; from front to back, T1–T3, each row shows the five repetitions per treatment.

3. Each section was carefully transferred to a droplet of distilled water on a glass slide. 4. Slides were dried on a hot plate (50◦ C–60◦ C) until the water droplet completely evaporated (see Note 6). 5. Sections were stained with 0.05% Toluidine Blue, pH 4.7 (4) and mounted in distilled water under a clean coverslip. 3.5. Analysis

The 60 sections obtained (20 per treatment) were examined with a light microscope coupled to a camera lucida; the external outlines of sections were drawn, as were their internal outlines, when non-polymerized areas were present. Illustrations of the polymerized areas were analyzed using the Planimetry System software package (SPLAN) developed by CINAG, São Paulo State University, Botucatu, São Paulo, Brazil, to quantify the polymerized areas. For statistical analysis, we considered the average polymerized area per treatment. Averages were compared using the Tukey test (p < 0.05).

3.6. Main Results and Discussion

The tissue samples processed according to the manufacturer’s recommendations (T1) demonstrated non-polymerized areas in the core portion of blocks corresponding to 49.36% of their crosssection (Fig. 3.2a). During microtomy, these incompletely polymerized samples tended to fragment, making it very difficult to obtain whole sections. Additionally, these sections did not spread well on the glass slides and sections were only rarely suitable for microscopic analysis. This situation is often erroneously interpreted as being due to poor infiltration. Infiltration proved to be satisfactory and resin could be seen inside samples when using the standard manufacturer’s techniques. On the other hand, although resin penetrated

44

Paiva, Pinho, and Oliveira

Fig. 3.2. Blocks and sections obtained from the three treatments. a–c Treatment 1. a General aspect of the block (arrow indicates the non-polymerized area); this is the first repetition of T1, as indicated on the side of the wood stub (1/1). b General view of the section; note limits of hardened resin around the section (arrowhead). (c) Detail of the section, showing the hollow core (asterisk). d–f Treatment 2. d General aspect of the block (arrow indicates the non-polymerized area); this is the third repetition of T2, as indicated on the side of the wood stub (2/3). e General view of the section; note limits of hardened resin around the section (arrowhead). f Detail of the section, showing the hollow core (asterisk). g–i Treatment 3. g General aspect of the block without any non-polymerized areas; this is the second repetition of T3, as indicated on the side of the wood stub (3/2). h General view of the section; note limits of hardened resin around the section (arrowhead). i Detail of the perfectly extended section. Dotted lines in a, d, and g indicate the sample border.

to the sample core, polymerization was not observed in this region. If infiltration was therefore adequate, why did polymerization not occur in the sample core? Apparently, although the sample was well infiltrated, there was not enough hardener present for polymerization to occur rapidly and homogeneously. This is a crucial point, for once hardener is added to the activated resin and poured into histomolds, the polymerization occurs quite rapidly, completing in about 30 min at room temperature. This period apparently is not sufficient to permit hardener penetration into the tissue sample and polymerization only occurs in the external resin pool and on a small superficial portion of the sample, generating a decreasing gradient of polymerization towards the sample core (Fig. 3.2b, c).

Large Plant Samples: How to Process for GMA Embedding?

45

As explained below, we do not recommend using the manufacturer’s protocols for processing large samples. In T2, the pre-infiltration stage was extended from 2 to 48 h and the infiltration stage itself was conducted at low temperatures and divided into two stages: the first under refrigeration (5◦ C) for 24 h and the second in a freezer (−18◦ C) for 48 h. After infiltration, the embedding and hardening media was prepared following the manufacturer’s recommendations (15 mL GMA:1 mL hardener) and samples were placed into histomolds at room temperature for polymerization. Results of this procedure were similar to those of T1, with polymerization being restricted to the superficial portion of the tissue blocks; the non-polymerized area occupied 44.61% of the sample core (Fig. 3.2d). In addition to poor polymerization, sections demonstrated problems during mounting and staining, resulting in low quality material for microscopic analysis (Fig. 3.2e, f). We therefore conclude that increasing the infiltration stage under cooling improves infiltration but does not facilitate resin polymerization in the sample core upon addition of the hardener. In T3, the times of pre-infiltration and infiltration were identical to T2, but the hardener was added in the beginning of the second infiltration step when samples were kept in freezer, slowing the polymerization reaction. The hardener volume was also modified to compensate for the fact that large samples contain significant resin volumes (∼80% for the ginger samples in the present work). Thus the calculation of the amount of hardener to be added must consider the volume of resin used to compose the block around the sample plus 80% of the volume of the sample. For details of the hardener calculation, see Note 7. The T3 procedures yielded 100% polymerization of the five samples (Fig. 3.2g), resulting in sections with excellent quality for histological analysis (Fig. 3.2h, i). What caused these large differences in polymerization with this treatment? Two factors were critical: the addition of extra hardener in proportions appropriated for resin polymerization inside the sample and the extension of the infiltration/embedding times by maintaining histomolds with complete embedding media in a freezer. The addition of extra hardener would normally produce even faster polymerization than the conventional approach if maintained at room temperature, but maintaining a very low temperature allowed the GMA: hardener solution to penetrate into the tissue material without polymerizing. For polymerization, see Note 8. Statistical analyses presented in the following table confirm observations discussed here, showing the average polymerized area in each treatment, compared by Tukey test (averages followed by same letter do not differ statistically at a 5% level of probability):

46

Paiva, Pinho, and Oliveira

Treatment

Polymerized area (%)

T1

50.64b

T2

55.39b

T3

100.00a

a,b The average values followed by different letters, in the same column,

differ at 5% level of probability in the Tukey test.

It is important to note that this T3 procedure is only necessary when bulky or heavy materials are used, for the GMA: hardener solution penetrates very rapidly into small and porous materials, ensuring good results with this material while using standard procedures.

4. Notes 1. Several authors have noted fixation problems and the occurrence of artifacts due to the air present in the intercellular spaces in plant tissues. Unfortunately, air removal is not a common practice in plant anatomy laboratories, due in large part to the lack of vacuum pumps in plant anatomy laboratories. We were able to develop an alternative, very inexpensive and portable method that can substitute the use of a vacuum pump – a simple disposable syringe. The tissue samples were immersed in fixative solution inside a syringe (we found the 20 mL variety, without a needle, to be most useful) and the plunger was then pushed in to remove any air remaining in the syringe itself. When a small amount of fixative can be seen emerging from the syringe opening, seal that aperture with your forefinger (do not forget to wear gloves!) and then pull the plunger back until the inner syringe volume doubles; maintain the plunger in this position and observe the air bubbles exiting the sample. Repeat this process twice to ensure complete air removal. Samples are now air-free and the fixative is in contact with the innermost cells. Additionally, air removal facilitates resin infiltration. In addition to being extremely economical, air removal using syringes (our portable ‘vacuum pump’) allows adequate fixation of plant samples in the field, immediately after their collection. 2. According to Johansen (3), the minimum fixation time for FAA is 18 h and the material can then be left in this solution almost indefinitely without appreciable damage. This author also remarked that ‘this property of nearly perfect

Large Plant Samples: How to Process for GMA Embedding?

47

preservation makes FAA the ideal fluid to take on collecting trips’. Jensen (5) noted that tissues fixed in FAA for a minimum of 4 h can be stored indefinitely, but considered this solution to be only a moderately good fixative for most plant tissues: ‘In most tissues, it causes considerable shrinkage, hardens cells and occasionally makes sectioning difficult’. Formalin-aceto-alcohol is the most popular fixative in plant anatomy because it readily penetrates most tissues and is very cheap and easy to use in the field. To avoid problems detected by Jensen, samples can be conserved in 70% ethanol after FAA fixation. Experience has shown that fixation times should not be longer than 48 h and that samples should then be washed with 50% ethanol (which has the same concentration as the fixative and helps remove its residues) before storing in 70% ethanol. If samples will be processed quickly, they can be left overnight in 60 or 70% ethanol. 3. As GMA is water soluble, complete sample dehydration is not necessary (as stated in the Leica Historesin Embedding Kit) and the ethanol series was run only to commercial ethanol levels (92–98%). Do not forget to immediately wash (with ethanol) all the glasswares that come into contact with GMA + hardener. 4. Important: all materials employed during embedding (including hardener, histomolds and GMA, and most glassware) must be kept at low temperatures (∼5◦ C). Pipettes must be used at room temperature to avoid imprecision. If the glassware or histomold are warm, polymerization occurs more rapidly, making it difficult for the hardener to penetrate deep inside samples, which will result in unsatisfactory polymerization in cores of large tissue samples. Additionally, rapid polymerization makes it difficult to correctly position samples inside the histomold cells. Maintaining histomolds over ice will guarantee enough time to adjust the tissue sample positions. 5. The sample blocks embedded in GMA must be kept under very low-humidity conditions in a sealed recipient with silica gel crystals because GMA easily hydrates, becoming soft and very difficult to section. 6. We recommend temperatures of between 50 and 60◦ C for drying the glass slides on hot plates. At higher temperature (>60◦ C), air present in the water can form bubbles between the slide and the thin-section, in detriment to staining and microscopic analysis. 7. To calculate the amount of additional hardener needed for blocks with large specimens, assume that about 80% of the sample volume is resin. The sample volume must therefore

48

Paiva, Pinho, and Oliveira

be multiplied by 0.8 to calculate the amount of resin it contains (volume A). Add to volume A the amount of resin required to enclose the tissue sample in the histomold (volume B) and then calculate the quantity of hardener needed at the rate of 15 mL of GMA to 1 mL of hardener (volume C). Quickly mix the volume of resin for the block (volume B) with the hardener (volume C) in a beaker and quickly transfer it to the histomold with the sample; then move the mold immediately to the freezer. This operation must be performed quickly to avoid the beginning of polymerization. Cover histomolds with plastic film in order to protect the freezer from the volatile historesin, especially after adding the hardener. A sample calculation: a tissue sample that is 0.8 × 0.8 × 0.8 cm (i.e. 0.512 cm3 ) that is properly infiltrated with resin will have ∼80% of its volume filled by GMA. Volume A is therefore 0.512 (sample size) times 0.8 (to reach 80%), which equals 0.4096 mL. If histomolds hold 1.0 cm3 , then 0.488 mL of resin (volume of histomold minus the volume of the sample, i.e. 1.0–0.512) will be needed to top-off each cell (this is the volume B). Adding volume A (0.4096) to volume B (0.488) yields the total volume of resin (0.8976 mL), to be used to calculate the hardener volume in a proportion of 15:1. The volume of hardener (volume C) is therefore 0.05984 mL (the total volume of resin [0.8976 mL] divided by 15). Remember that you will use only 0.488 mL of new resin in each histomold because the rest is already inside the tissue sample. Thus for the embedding media you will mix 0.488 mL of GMA to 0.05984 mL of hardener per histomold cell. Remember, also, that these values are for each cell, and must be multiplied by the number of blocks to be made. If it is difficult to estimate the tissue sample volume because it has a very irregular contour, increase the volume of hardener by 40–80% in tests to determine the most efficient proportions. 8. When the embedding mixture becomes viscous at room temperature (25◦ C), the polymerization process should be finished in an oven at 40◦ C until final hardening (∼30 min) as suggested by Igersheim and Cichocki (6).

Acknowledgments Denise M.T. Oliveira and Elder A.S. Paiva thank CNPq for their research grants. This method was standardized during the development of projects that were partially supported by Brazilian foundations (CNPq, CAPES, FAPEMIG, and FAPESP).

Large Plant Samples: How to Process for GMA Embedding?

49

References 1. Gerrits, P. O. (1991) The Application of Glycol Methacrylate in Histotechnology: Some Fundamental Principles. Heidelberg, German: Leica Gmgh, 80p. 2. Feder, N., O’Brien, T. P. (1968) Plant microtechnique: some principles and new methods. Am J Bot 55, 123–142. 3. Johansen, D. A. (1940) Plant Microtechnique. McGraw-Hill, New York, NY, 523p. 4. O’Brien, T. P., Feder, N., McCully, M. E. (1964) Polychromatic staining of plant cell

walls by Toluidine Blue O. Protoplasma 59, 368–373. 5. Jensen, W. (1962) Botanical Histochemistry: Principles and Practice. W.H. Freeman, San Francisco, CA, 408p. 6. Igersheim, A., Cichocki, O. (1996) A simple method for microtome sectioning of prehistoric charcoal specimens, embedded in 2-hydroxyethyl methacrylate (HEMA). Rev Palaeobot Palynol 92, 389–393.

Chapter 4 Image Cytometry: Nuclear and Chromosomal DNA Quantification Carlos Roberto Carvalho, Wellington Ronildo Clarindo, and Isabella Santiago Abreu Abstract Image cytometry (ICM) associates microscopy, digital image and software technologies, and has been particularly useful in spatial and densitometric cytological analyses, such as DNA ploidy and DNA content measurements. Basically, ICM integrates methodologies of optical microscopy calibration, standard density filters, digital CCD camera, and image analysis softwares for quantitative applications. Apart from all system calibration and setup, cytological protocols must provide good slide preparations for efficient and reliable ICM analysis. In this chapter, procedures for ICM applications employed in our laboratory are described. Protocols shown here for human DNA ploidy determination and quantification of nuclear and chromosomal DNA content in plants could be used as described, or adapted for other studies. Key words: Image cytometry, C value, nuclear DNA content, chromosomal DNA content, picograms, quantitative microscopy, integrated optical density, Feulgen reaction, genome size.

1. Introduction Image cytometry (ICM) is an association of microscopy and digital-image-based techniques (1), particularly useful for spatial and densitometric measurements in the cytological area (2, 3). This tool has been widely used in quantitative applications for DNA ploidy studies in cancer pathology (4–6) and in absolute C value determination in plant (1, 7–9) and animal species (10, 11). Considering these approaches, the microscope, camera, and software industries have developed new integrated technologies in order to simplify the ICM laboratorial routine. With H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_4, © Springer Science+Business Media, LLC 2011

51

52

Carvalho, Clarindo, and Abreu

the evolution in digital cameras and system stations for highend laser analysis, ICM can be performed in almost all cytology laboratories using a good digital camera coupled to a light microscope (12). As main advantages in comparison with others methodologies, like flow cytometry (FCM), ICM carries out analysis in samples with relatively small number of nuclei (13), allowing visual recognition, in a semi-interactive way, in order to monitor structures of interest (14) and is cost-accessible for most specialized laboratories. Various examples of ICM studies in qualitative and quantitative DNA analysis have been reported in the following areas: (i) human pathology, mainly in tumorigenesis screening based on aneuploidy and genome size instability, such as diagnosis and monitoring of malignant melanomas and squamous cell carcinoma (5, 15), neoplasia (15, 16), cervical intraepithelial lesions and invasive carcinoma (5), and primary achalasia (17); (ii) animal genome size quantification, especially applied in taxonomic and evolution studies (10, 11); and (iii) plant science, for determination of absolute nuclear DNA values and ploidy level in agronomical (1, 7–9) and ecological species (18). Besides nuclear genome size, chromosomal DNA content has also been quantified by ICM, such as in Zea mays (19) and Capsicum annuum (20), resolving the genome size at the chromosomal level. The nuclear and chromosomal DNA contents, reported in picograms (pg) and/or base pairs (bp) have become useful in genome sequencing projects and genetic mapping (21). For nuclear and chromosomal DNA densitometric measurements by ICM, the system components basically include a computer with image analysis software, linked to a digital video camera coupled to a microscope (2, 14, 22, 23). Digital images are captured from slides subjected to Feulgen reaction (24) and then stored on hard disk. As images are recorded in pixel matrix, the system needs to be calibrated so as to grab images in known spatial and densitometric parameters (2). Since pixels do not have an intrinsic value, which depends on the system resolution, spatial and optical density (OD) values from micrometric and gray reference scales are attributed to them, respectively. To find the measurement of interest, the nucleus or chromosome area (µm2 ) is multiplied by their average OD, resulting in the integrated optical density (IOD). Generally, IOD values are automatically estimated by software algorithms (2, 23). Seeing that the IOD of the sample is equivalent to the DNA content, this value can be converted to pg or bp (1). As this methodology combines the high technology of optical, electronic, and digital instruments, additional technical details may be found in a wide variety of specialized references. Good theoretical references on digital cam-

Image Cytometry: Nuclear and Chromosomal DNA Quantification

53

era technology and optical microscopy have been overviewed (2, 12, 25). In this chapter, we describe our ICM protocols for system calibration and setup and for DNA content quantification in the following materials: (a) human and plant nuclei and (b) plant chromosomes. Protocols presented here are reproducible in our routine and were transcribed to contribute as a useful guideline for cytogenetic and cytometric laboratories. Given the need to assure accurate measurements, mainly in clinical practice, the application of ICM methodology requires important quality control of instrumentation and procedures (26–32).

2. Materials The equipments, reagents, labware trademarks and models cited in protocols correspond to those used in our laboratory routine, but similar ones can also be employed. We strongly recommend observing the hazard classification on the label and the safety data sheet of reagents used, according to the manufacturer’s warnings. All laboratorial precautions should be taken to avoid accidents. 2.1. Calibration and System Setup 2.1.1. Slide Preparation

R 1. A 5 mL tube with sodium heparin (Vacutainer ) for taking chicken red blood cells (CRBCs).

2. CRBCs are obtained from healthy male and female individuals (Gallus domesticus). 3. Micropipette and tips. 4. Glass microscope slides (26 × 76 × 1 mm), pre-cleaned with ethanol to remove dirt and oil. 5. Hot plate (surface temperature of 50◦ C). 6. Coplin jar with screw cap. 7. Freezer (−20◦ C). R 8. Fixative solution 1: 70% ethanol (Merck ).

9. Fixative solution 2: 4% phosphate-buffered formaldehyde, freshly prepared from 37% formaldehyde and stored at −20◦ C. 2.1.2. Feulgen Solution

1. Distilled water (dH2 O). 2. 5 M HCl: 42.5% (v/v) 37% HCl. Caution in the preparation: first add the dH2 O, then the acid. Store at room temperature.

54

Carvalho, Clarindo, and Abreu

3. Schiff’s reagent (see Section 2.2). 4. 0.5% SO2 water: 5% (v/v) 1 N HCl, 5% (v/v) 10% metabisulphite. Store at room temperature. 5. Immersion oils. Use a refractive index oil (nD = 1.525 − 1.540) between the slide and the coverslip, and a refractive index oil (nD = 1.515) between the coverslip and the microscope objective lens. 6. Glass coverslips (24 × 50 mm). 7. Nail polish. 2.1.3. System Setup

1. Trinocular Photomicroscope (OlympusTM BX-60) with: (a) stabilized light, (b) UPlanFI objective magnification ×40 with 0.75 numeric aperture, (c) PlanApo ×60 oil immersion objective with 1.40 numeric aperture, (d) PlanApo ×100 oil immersion objective with 1.40 numeric aperture, (e) Aplanat Achromat condenser with aperture 1.4, and (f) neutral density filter (ND6). 2. Monochromatic charge-coupled device (CCD) digital video camera of 12 bits gray and frame grabber card (Photometrics R CoolSNAP Pro – Roper Scientific, Tucson, AZ, USA) (see Note 1). 3. Computer Pentium 4 HT, CPU 3.2 GHz, 1 GB RAM, R 80 GB ROM, Microsoft Windows XP Professional V 2002 Operational System. R R 4. Image Pro – Plus 6.1 software (Media Cybernetics ).

5. Slide Micrometer (1000 µm, OlympusTM ) 6. Interference filter (green color, 550–570 nm). 7. Neutral density filters (ND6): 0.15, 0.30, 0.40, 0.60, 0.90, R and 2.50 (Edmund Industrial Optics , Barrington, NJ, USA). 8. Linear 11 stepped density filter (Edmund Industrial R , Barrington, NJ, USA). Optics 2.2. Preparation of Schiff’s Reagent

To prepare 100 mL: 1. Dissolve the reagent: in a glass bottle, dissolve 0.5 g basic fuchsin in 85 mL boiling dH2 O, with the help of a magnetic shaker (see Note 2). 2. Decolorize with bisulfite: when the solution temperature reaches 50◦ C, add 15 mL 1 N HCl and 2.23 g K2 S2 O5 and shake until dissolving. 3. Allow fuchsin to decolorize: after it cools down, store at 4◦ C in a dark glass bottle for 24 h, shaking occasionally to dissolve any pink precipitate. 4. Remove organic impurities: add 0.703 g (up to 1.0 g/100 mL is acceptable if demanded) activated charcoal and shake.

Image Cytometry: Nuclear and Chromosomal DNA Quantification

55

Then, pour through filter paper using a Büchner funnel as long as needed. Use only enough charcoal to decolorize the stain solution, which should be colorless or champagne color (23, 33). 5. Storage of Schiff’s reagent: store in a tightly capped dark bottle in a refrigerator (4◦ C), and use exposing as little as possible to light (see Note 3). 2.3. Slide Preparation for Ploidy Determination of Human Nuclear DNA by ICM

Items 3–9 under Section 2.1.1, items 1–7 under Section 2.1.2 and the following additional materials are needed: 1. Endocervical swab and tube kit for collection of cellular material. 2. Cervical cellular material (sample), collected by an authorized professional. 3. Cellular material of female buccal mucosa (standard). 4. Refrigerator (4◦ C). 5. Vortex. 6. Microcentrifuge tubes of 2 mL. 7. Microcentrifuge. 8. Diamond pencil. 9. Cytocentrifuge. 10. Filter paper.

2.4. Slide Preparation for Plant Nuclear DNA Quantification by ICM

Items 4–7, 9 under Section 2.1.1, 1–7 under Section 2.1.2, 6, 8 under Section 2.3, and the following additional materials are needed: 1. Plant nuclei material: root tips from germinated seeds of the standard (see Note 4) and sample species. 2. Petri dishes. 3. B.O.D incubator, with temperature ranging from 28 to 30◦ C. 4. Enzymatic solution (see Note 5). 5. Water bath 34◦ C. 6. Stereoscopic microscope. 7. Scalpel. 8. Fixative solution 3: methanol and glacial acetic acid in proportion of 3:1 (v/v), freshly prepared and stored at −20◦ C. 9. Fixative solution 4: 95% ethanol.

2.5. Slide Preparation for Plant Chromosomal DNA Quantification by ICM

Items 4–7 under Section 2.1.1, 1–4 under Section 2.1.2, 6 under Section 2.3, 2–9 under Section 2.4, and the following additional materials are needed: 1. Plant chromosomal material: root tips from germinated seeds.

56

Carvalho, Clarindo, and Abreu

2. Metaphase blocking solution: a chemical inhibitor for metaphase arrest, such as colchicine, oryzalin, trifluralin, 8-hydroxyquinoline and amiprophos-methyl. 3. Fixative solution 5: methanol, 37% formaldehyde and glacial acetic acid in proportion of 17:5:1 (v/v/v), freshly prepared.

3. Methods 3.1. Calibration and Setup of System 3.1.1. Slide Preparation

1. Prepare the slide using the smear technique: drop 5 µL male or female CRBC near one end of a slide and smear using another slide inclined at an angle of approximately 45◦ (see Note 6). 2. Air-dry the slide through fast arm waving and place it on a hot plate. 3. Store the slide in the fixative solution 1 in a Coplin jar at −20◦ C for at least 12 h, then fix it in solution 2 for 1 h at 25◦ C.

3.1.2. Feulgen Reaction (see Note 7)

1. After fixation, wash the slide in running water for 10 min and air-dry. 2. Hydrolyze in 5 M HCl at 25◦ C. The time should be adjusted from 10 to 60 min, according to cell type. Wash the slide three times (for 3 min each time) in dH2 O. 3. Stain the slide with Schiff’s reagent for 12 h at 4◦ C, in the dark. Subsequently, wash it three times (for 3 min each time) in 0.5% SO2 water and three times (for 1 min each time) in dH2 O and air-dry. 4. Drop 10 µL of immersion oil on the slide, add a coverslip, seal with nail polish and store it in the dark until image analysis, but no longer than two days.

3.1.3. Microscopy and Digital System Setup

1. All optical parts (filters, objectives, condenser and eyepieces) should be extremely clean. 2. Turn on the camera, computer and microscope, and wait for 15 min (see item 3 under Section 3.1.6) for light stabilization. 3. Adjust the microscope prior to each slide capture session by applying the Köhler method to obtain an optimal light path and, consequently, to reduce stray light. Attention! Do the setup for each objective magnification.

Image Cytometry: Nuclear and Chromosomal DNA Quantification

3.1.4. Spatial Calibration – From Pixel to Micrometer

57

1. Check your camera and software performance according to instruction manuals and make sure that particular functions are correct, especially the black and incident (white) level settings. 2. Use the slide micrometer scale and the software spatial calibration tools to establish the unit of measurement of the image from pixel to micrometer and record it in the preferences.

3.1.5. Density Calibration – From OD to IOD

1. Couple interference filter to field diaphragm. 2. Mount a “blank” slide, without cell material, add oil and lay a coverslip. Place it on the stage and focus at slide level. 3. Close field of the iris diaphragm to a slightly larger size than the image size. 4. Determine OD range (minimum–maximum) by opening the histogram live-window of the software and adjusting the microscope light intensity knob, so that the highest gray level (peak moves to the right) is slightly lower than the maximum value on the gray scale (Fig. 4.1). 5. Set 1/125 (8 ms) of exposure time for capture, or determine another value (i.e. 5–30 ms). 6. Use eight accumulated frames and divide by 8, in the accumulate settings command.

Fig. 4.1. Histogram showing the 12 bits gray range (0 to 4095 – coordinate x) and the light intensity peak at the position slightly lower than maximum saturation value on the gray scale.

58

Carvalho, Clarindo, and Abreu

7. Mount a step tablet with the standard neutral density filters set to calibrate the OD scale. For that, capture a series of empty images, interposing each one of these filters (or in combination to generate other OD values) over the interference filter in the light path (Fig. 4.2). 8. Use these step tablets and the density calibration tool of the software to fit the curve of intensity to OD and save it in preferences. This calibration process should be carried out for each objective (Fig. 4.2). 9. Remake the illumination settings before the capture session of every new slide.

Fig. 4.2. Plot (left) showing a linear density before calibration. Plot (right) of the calibrated intensity curve obtained from stepped tablet values mounted with standard neutral density filters (ND6 – 0.15, 0.30, 0.40, 0.60, 0.90, and 2.50, R , Barrington, NJ, USA). Edmund Industrial Optics

3.1.6. Calibration Tests

Calibration and evaluation of the image analysis system is performed based on three tests: stability (34), linearity (2, 23, 34), and uniformity (22). 1. Stability test: use the frame coordinates x, y on the center of the field to measure gray level of one pixel. Capture images every 3 min during 1 h. Plot the data on a spreadsheet to evaluate light source variations. Attention! In capture routine, use the analysis system only after its stabilization time (Fig. 4.3). 2. Linearity test: place the linear 11 stepped density filter on the microscope stage and capture each step. Use the density command to convert the captured image into a calibrated one and save. Apply the appropriate tool to segment each

Image Cytometry: Nuclear and Chromosomal DNA Quantification

59

Fig. 4.3. Stability test obtained from OD mean values (coordinate y) of one pixel (e.g. frame position: x = 150, y = 150) monitored every 3 min during 1 h (coordinate x). Note that this analysis system stabilized after 12 min.

step region and calculate the OD. Use the spreadsheet to compare, by linear regression, the obtained measurements with the known OD certified for each step (see Note 8 and Fig. 4.4). 3. Uniformity test: grab images, using the ×40 objective (×60 or ×100) of one single CRBC nucleus at distinct and well-distributed visual field positions (e.g. 36). Attention! Be careful to maintain the moved nucleus at the same focus. Calculate its OD values and coefficient of variation (CV) (see Note 9). 3.2. Slide Preparation for Ploidy Determination of Human Nuclear DNA by ICM 3.2.1. Protocol Employing Air-Drying Technique

1. Transfer 1 mL of cellular material (sample or standard) in a microcentrifuge tube and fill up to 2 mL with fixative solution 1. 2. Centrifuge the material at 100g for 5 min. Discard the supernatant and add 2 mL of fixative solution 1. Repeat this step three times, with intervals of 10 min and store at −20◦ C for at least 24 h.

60

Carvalho, Clarindo, and Abreu

Fig. 4.4. Linear 11 stepped density filter (above), showing 11 discrete density steps R , Barrington, NJ, USA), used in the with increments of 0.1 (Edmund Industrial Optics linearity test. Stepped tablet (below) exhibiting the 11 square images captured from each step of the filter and showing the 11 OD calibrated. The OD calculated (0.04–0.95) was compared with the certified for each step (0.04–1.00) by linear regression, using the spreadsheet. The R2 and SE obtained were 0.999 and 1.14, respectively.

3. Centrifuge the cellular suspension at 100g for 5 min. Discard the supernatant, add 500 µL of fixative solution 1 and store at 4◦ C. 4. Vortex at medium speed for 40–60 s. 5. Carefully drop 50 µL of sample cellular suspension near the left slide end; air-dry it through fast arm waving and place it on a hot plate. Repeat this step for the standard material near the right slide end (see Note 10). 6. Place the slide in fixative solution 2, in a Coplin jar, at 25◦ C for 1 h. 7. Repeat steps 1–4 of Section 3.1.2. 3.2.2. Protocol Employing Cytocentrifuge Technique

1. Repeat Steps 1–4 of Section 3.2.1. 2. Cytocentrifuge 200 µL of sample cellular suspension, so that the centrifuged material is imprinted centrally on the slide (on the right half), at 1800 rpm for 5 min. 3. Air-dry the slide by fast arm waving and place it on a hot plate. 4. Take the same slide and mount the centrifuge apparatus for standard cellular suspension, in a way that the centrifuged material is imprinted centrally on the slide, on the opposite end (i.e. on the left half). Cytocentrifuge at 1800 rpm for 5 min.

Image Cytometry: Nuclear and Chromosomal DNA Quantification

61

5. Repeat Step 3. 6. Repeat Steps 6 and 7 of Section 3.2.1. 3.3. Slide Preparation for Plant Nuclear DNA Content Quantification by ICM

1. Germinate seeds (sample and standard) in Petri dishes with dH2 O in a B.O.D., at 28–30◦ C. 2. Place the root tips in fixative solution 3, making three changes with 10 min intervals. Store at −20◦ C for 24 h. 3. Replace this solution with fixative solution 4, change three times and store at −20◦ C for 24 h. 4. Put the root tips in a microcentrifuge tubes and macerate them in an enzymatic solution (see Note 5) at 34◦ C in a water bath. 5. Wash the root tips for 10 min in dH2 O, place them in the fixative solution 4, changing it three times and store at −20◦ C for at least 24 h. 6. Using the diamond pencil, trace a soft transversal line across the middle of the microscope slide and mark “a” for sample and “b” for standard material, on the end corners. 7. Place the slide inclined at an angle of 25–30◦ under a stereoscopic microscope, keeping the slide center in focus. 8. Place the root tip (e.g. sample) right below the traced line. 9. Apply the cellular dissociation and air-drying techniques (35): quickly dissociate the meristem with a scalpel, while dripping one to three drops of fixative solution 4. Air-dry the slide through fast arm waving and place on a hot plate to completely dry up (see Note 11 and Fig. 4.5). 10. Incline the same slide at 25–30◦ under the stereoscopic microscope, this time with the other (empty) slide half in focus. 11. Place another root tip (e.g. standard) left below the traced line (see Note 10). 12. Repeat Step 9. 13. Place the slide immediately in fixative solution 2, in a Coplin jar, for 12 h at 25◦ C. 14. Repeat Steps 1–4 of Section 3.1.2.

3.4. Slide Preparation for Plant Chromosomal DNA Quantification by ICM (see Note 12)

1. Germinate the sample seeds in Petri dishes with dH2 O in a B.O.D, at 28–30◦ C. 2. Treat the root tip meristems with a blocking solution, with treatment time and concentration being established for each species. 3. Wash roots for 20 min in dH2 O and place them in fixative solution 3, changing it three times and store at −20◦ C for 24 h.

62

Carvalho, Clarindo, and Abreu

Fig. 4.5. Start position for cellular dissociation procedure of the plant nuclei and chromosomes slide preparation. Note that the root tip (white draw) is placed on the slide inclined at 25–30◦ .

4. Replace it with fixative solution 4, changing it three times and store at −20◦ C for at least 24 h (20). 5. Wash the root tips for 10 min in dH2 O, transfer to microcentrifuge tubes containing enzymatic solution (see Note 5) and incubate at 34◦ C in water bath. 6. Wash roots for 10 min in dH2 O, place them in fixative solution 4 and store at −20◦ C. 7. Prepare slides using the cellular dissociation and air-drying techniques (35): place each root tip on a slide inclined at 25–30◦ under a stereoscopic microscope. Quickly dissociate it with a scalpel, while dripping one to three drops of fixative solution 4. Air-dry the slide by fast arm waving movements and place it on a hot plate (see Note 11 and Fig. 4.5). 8. Place the slide in a fixative solution 5, in a Coplin jar, at 25◦ C for 12 h. 9. Repeat Steps 1–4 of Section 3.1.2. 3.5. Image Cytometry 3.5.1. Human Nuclear ICM

1. Place the slide on a microscope, recheck the microscope setup and recall the spatial and OD calibration settings. 2. Capture (×40 or ×60 objective) 20–40 isolated and wellpreserved nuclei of the standard.

Image Cytometry: Nuclear and Chromosomal DNA Quantification

63

3. Move to the other slide half (without changing the microscope setup) and capture over 200 isolated and wellpreserved target nuclei of the sample. 4. Make sure that the captured images are saved as calibrated ones. When not, open the customized spatial and density commands to convert the captured images to calibrated ones and save them. 5. Mount a frame with a collection of the standard nuclei as a “family portrait”, using copy and paste, macro-customized functions or programming sources. Repeat the procedure for the sample nuclei, placing them below the standard collection. 6. Select the IOD in the measurement parameters list. If the software does not have this parameter, alternatively select AREA and OD and multiply these values to obtain the IOD. 7. Apply the selection tool around the standard nuclei collection, adjust the appropriate segmentation level and apply the count/size tool to measure the IOD. Repeat this step for the sample nuclei collection. 8. Export the data to a spreadsheet for statistical analyses. 9. Find the IOD values MODE for the standard nuclei, divide each one of these values by the MODE value and multiply by two, in order to convert them into C value (36). 10. Open a histogram window and customize its X coordinate for 0, 2, 4, 6, 8 units and so on (2 corresponds to G1 diploid nuclei). 11. Plot these C values on the X coordinate of the histogram and use as reference. 12. Repeat Step 9 for sample nuclei, plot these C values on the X coordinate of the histogram and compare with the reference data (see Note 13 and Fig. 4.6). 3.5.2. Plant Nuclear ICM

1. Place the slide under the microscope, recheck the microscope setup and recall the customized spatial and OD calibration settings. 2. Capture (×40 or ×60 objective) ten isolated and wellpreserved early prophase nuclei of the standard and repeat the session for late telophase. 3. Move to the other slide end (without changing the microscope setup) and repeat the capture sessions for the sample. 4. Repeat Steps 4–8 of Section 3.5.1 (Fig. 4.7). 5. Measure the nuclear DNA content, using the formula:

64

Carvalho, Clarindo, and Abreu

Fig. 4.6. Histogram obtained from DNA ploidy values measurement of cervical nuclei. Note that images of nucleus (above) were selected to show some DNA ploidy level (C = 1.56–7.91).

Fig. 4.7 Collection of nuclei at the early prophase (P) and late telophase (T) captured with ×40 objective. a Nuclei of Pisum sativum L. “Ctirad” used as internal standard (2C = 9.09 pg). b Nuclei of Capsicum annuum L. “Fortuna Super” (2C = 8.10 pg). The CV and 4C/2C ratio obtained were 3.45 and 2.0, respectively. Bar 5 µm.

(I) 1 Cs = (IODs × ICp )/IODp where 1Cs = 1C nuclear DNA content of the sample; 1Cp = 1C nuclear DNA content of the standard; IODs = nuclear IOD value of the sample; IODp = nuclear IOD value of the standard; 3.5.3. Plant Chromosomal ICM

1. Place the slide under the microscope, recheck the microscope setup and recall the customized spatial and OD calibration settings.

Image Cytometry: Nuclear and Chromosomal DNA Quantification

65

2. Capture (×100 objective) the ten metaphases with all non-overlapping and well-preserved chromosomes. 3. Repeat Step 4 of Section 3.5.1. 4. Open a metaphase image file and use the software tools to assemble the karyogram, in the same frame, according to cytogenetic rules. 5. Repeat Step 6 of Section 3.5.1. 6. Select chromosomes, adjust the appropriate segmentation level and apply the count/size tool to measure the IOD of each of them. 7. Export the data to a spreadsheet and do the statistical analyses. 8. Measure the mean DNA content of each chromosome, using formulas: (I) 1Cn = (2Cn /r)/2 (II) IODc = (IODpc × n)/4 (III) IODt = IODc (IV) 1Cc = (1Cn × IODc )/IODt where 1Cn = mean 1C nuclear DNA content; 2Cn = 2C nuclear DNA content; r = number of FCM replications; IODc = mean IOD value of the chromosome with one chromatid (1C); IODpc = IOD of the chromosome with two chromatids (2C); n = number of metaphases; IODt = IOD of all chromosomes; 1Cc = mean 1C chromosomal DNA content.

4. Notes 1. Besides the PhotometricTM camera, other high-performance CCD digital cameras can be used (i.e. SonyTM , OlympusTM , AndorTM , and HamamatsuTM ). A good technical reference for a choice of the most appropriate digital camera for a particular interest can be found at the Scientific Digital Camera solutions page (AndorTM technology catalog – www.andor.com). 2. Alternatively, another reagent can be used: pararosaniline R (prepare the solution as recommended by the Merck microscopy support).

66

Carvalho, Clarindo, and Abreu

3. When a white precipitate begins to form, the stain solution should be discarded. Schiff’s reagent is commercially available, but sometimes it is not as reliable as freshly prepared solutions and, therefore, it is not recommended (23, 33). 4. An ideal DNA reference standard should be known in absolute value, genetically stable, easy to manipulate, available in sufficient quantities for analyses, and should have a genome size close to that of the target species. Different plant standards have been used, including Arabidopsis thaliana, Raphanus sativus, Pisum sativum (37, 38). 5. Carry out several test sessions to find the optimal time and concentration of the enzymatic treatment, according to cell type and species. Test the enzymatic solution prepared with pectinase, hemicellulase, cellulase, driselase, separately or combined. 6. This method is recommended for providing undamaged monolayers of blood cells and greatly preferred over “pushing” the blood, which can damage cells (23). 7. The Feulgen reaction principle basically includes: one hydrolysis step, using strong acid to generate free aldehyde groups in the DNA molecule, specifically by splitting up the purine bases A and G, thus producing “apurinic acid”, and another step, in which the fuchsin molecule decolorized with SO2 can bind and recover its pink color (24, 33). 8. In our laboratory, the software of the image analysis system automatically calculated R2 = 0.999 in the linearity test. The R2 and standard deviation considered adequate in the medical area are above 0.99 and below 5%, respectively (34). 9. Coefficient of variation below 3% is recommended for ICM analysis in cancer diagnosis (23, 34). Our CV value is lower than others of similar tests described in the literature. 10. In the internal standardization procedure, sample and standard nuclei are strictly processed on the same slide (7, 39, 40), in order to avoid random instrument drift and variation in the preparation and staining. 11. Alternatively, slides can be prepared using the squashing technique: meristems are squashed onto glass slides, coverslips removed over a cold plate, and slides air-dried (1, 7–9, 40). However, the cellular dissociation technique provides target material preparations flattened on the slide, well spread and showing little cytoplasmic background, overlaps or structural deformations of the chromatin (20, 35). 12. In this case, there is no direct reference standard. The nuclear DNA content of the same target species is used,

Image Cytometry: Nuclear and Chromosomal DNA Quantification

67

whose value can be established through nuclear ICM or by FCM, which also uses an internal standard (19, 20). 13. Basic performance standards for sampling, measurement, and scaling in ICM analyses must follow guidelines, consensuses, and reports of the European Society for Analytical Cellular Pathology – ESACP (27–32) and algorithms described by (26).

Acknowledgments We thank CNPq – Conselho Nacional de Desenvolvimento Científico e Tecnológico and CBP&D/Café – Consórcio Brasileiro de Pesquisa e Desenvolvimento do Café, Brazil, for their financial support. References 1. Vilhar, B., Greilhuber, J., Koce, J. D., Temsch, E. M., Dermastia, M. (2001) Plant genome size measurement with DNA image cytometry. Ann Bot Fenn 87, 719–728. 2. Chieco, P., Jonker, A., Van Noorden, C. J. F. (2001) Image Cytometry. Microscopy Handbooks 46. Springer, New York, NY, p. 116. 3. Pektas, Z. O., Keskin, A., Ömer, G., Karslio˘glu, Y. (2006) Evaluation of nuclear morphometry and DNA ploidy status for detection of malignant and premalignant oral lesions: quantitative cytologic assessment and review of methods for cytomorphometric measurements. J Oral Maxillofac Surg 64, 628–635. 4. Biesterfeld, S., Reus, K., Bayer-Pietsch, E., Mihalcea, A. M., Böcking, A. (2001) DNA image cytometry in the differential diagnosis of endocervical adenocarcinoma. Cancer Cytopathol 93, 160–164. 5. Böcking, A., Nguyen, V. Q. H. (2003) Diagnostic and prognostic use of DNA image cytometry in cervical squamous intraepithelial lesions and invasive carcinoma. Cancer Cytopathol 102, 41–54. 6. Grote, H. J., Nguyen, H. V. Q., Leick, A. G., Böcking, A. (2004) Identification of progressive cervical epithelial cell abnormalities using DNA image cytometry. Cancer Cytopathol 102, 373–379. 7. Greilhuber, J., Ebert, I. (1994) Genome size variation in Pisum sativum. Genome 37, 646–655.

8. Greilhuber, J., Obermayer, R. (1997) Genome size and maturity group in Glycine max (soybean). Heredity 78, 547–551. 9. Baranyi, M., Greilhuber, J. (1999) Genome size in Allium: in quest of reproducible data. Ann Bot Fenn 83, 687–695. 10. Gregory, T. R. (2001) The bigger the C-value, the larger the cell: genome size and red blood cell size in vertebrates. Blood Cell Mol Dis 27, 830–843. 11. Gregory, T. R. (2003) Genome size estimates for two important freshwater molluscs, the zebra mussel (Dreissena polymorpha) and the schistosomiasis vector snail (Biomphalaria glabrata). Genome 46, 841–844. 12. Russ, J. C., Russ, J. C. (ed.) (2008) Introduction to Image Processing and Analysis. CRC Press – Taylor & Francis Group, Upper Saddle River, NJ, p. 355. 13. Greilhuber, J. (2008) Cytochemistry and C-values: the less-well-known world of nuclear DNA amounts. Ann Bot Fenn 101, 791–804. 14. Rodenacker, K., Bengtsson, E. (2003) A feature set for cytometry on digitized microscopic images. Anal Cell Pathol 25, 1–36. 15. Bollmann, R., Méhes, G., Torka, R., Speich, N., Schmitt, C., Bollmann, M. (2003) Human papillomavirus typing and DNA ploidy determination of squamous intraepithelial lesions in liquid-based cytologic samples. Cancer Cytopathol 99, 57–62.

68

Carvalho, Clarindo, and Abreu

16. Motherby, H., Pomjanski, N., Kube, M., Boros, A., Heiden, T., Tribukait, B., Böcking, A. (2002) Diagnostic DNA-flow vs. image-cytometry in effusion cytology. Anal Cell Pathol 24, 5–15. 17. Gockel, I., Kämmerer, P., Brieger, J., Heinrich, U. R., Mann, W. J., Bittinger, F., Eckardt, V. F., Junginger, T. (2006) Image cytometric DNA analysis of mucosal biopsies in patients with primary achalasia. World J Gastroenterol 12, 3020–3025. 18. Temsch, E. M., Greilhuber, J., Krisai, R. (1998) Genome size in Sphagnum (Peat Moss). Bot Acta 111, 325–330. 19. Rosado, T. B., Carvalho, C. R, Saraiva, L. S. (2005) DNA content of maize metaphasic A and B chromosomes determined by image cytometry. Maize Genet Cooperation Newsl 79, 48–49. 20. Abreu, I. S., Carvalho, C. R., Clarindo, W. R. (2008) Chromosomal DNA content of sweet pepper determined by association of cytogenetic and cytometric tools. Plant Cell Rep 27, 1227–1233. 21. Doležel, J., Bartoš, J. (2005) Plant DNA flow cytometry and estimation of nuclear genome size. Ann Bot Fenn 95, 99–100. 22. Puech, M., Giroud, F. (1999) Standardization of DNA quantitation by image analysis: quality control of instrumentation. Cytometry 36, 11–17. 23. Hardie, D. C., Gregory, T. R., Hebert, P. D. N. (2002) From pixels to picograms: a beginners’ guide to genome quantification by Feulgen image analysis densitometry. J Histochem Cytochem 50, 735–749. 24. Feulgen, R., Rossenbeck, H. (1924) Mikroskopisch-chemischer Nachweis einer Nukleinasäure von Typus der Thymonukleinsäure und die darauf beruhende selective Färbung von Zellkernen in mikroskopischen Präparaten. Hoppe Seylers Z Physiol Chem 135, 203–248. 25. Andor TM Technology (2006) Scientific Digital Camera Solutions Catalog, p. 316. 26. Böcking, A., Adler, C. P., Common, H. H., Hilgarth, M., Granzen, B., Auffermann, W. (1984) Algorithm for a DNAcytophotometric diagnosis and grading of malignancy. Anal Quant Cytol 6, 1–8. 27. Chieco, P., Jonker, A., Melchiorri, C., Vanni, G., Van Noorden, C. J. F. (1994) An user’s guide for avoiding errors in absorbance image cytometry: a review with original experimental observations. Histochem J 26, 1–19. 28. Böcking, A., Giroud, F., Reith, A. (1995) Consensus report of the ESACP task force

29.

30.

31.

32.

33. 34. 35.

36.

37.

38.

39.

40.

on standardization of diagnostic DNA image cytometry. Anal Cell Pathol 8, 67–74. Haroske, G., Dimmer, V., Meyer, W., Kunze, K. D. (1997) DNA histogram interpretation based on statistical approaches. Anal Cell Pathol 15, 157–173. Giroud, F., Haroske, G., Reith, A., Böcking, A. (1998) ESACP consensus report on diagnostic DNA image cytometry. Part II: specific recommendations for quality assurance. Anal Cell Pathol 17, 201–208. Haroske, G., Giroud, F., Reith, A., Böcking, A. (1998) ESACP consensus report on diagnostic DNA image cytometry. Part I: basic considerations and recommendations for preparation, measurements and interpretation. Anal Cell Pathol 17, 189–200. Haroske, G., Baak, J. P. A., Danielsen, H. et al, Giroud, F., Gschwendtner, A., Oberholzer, M., Reith, A., Spieler, P., and Böcking, A. (2001) Fourth updated ESACP consensus report on diagnostic DNA image cytometry. Anal Cell Pathol 23, 89–95. Chieco, P., Derenzini, M. (1999) The Feulgen reaction 75 years on. Hystochem Cell Biol 111, 345–358. Vilhar, B., Dermastia, M. (2002) Standardization of instrumentation in plant DNA image cytometry. Acta Bot Croat 61, 11–26. Carvalho, C. R., Clarindo, W. R., Almeida, P. M. (2007) Plant cytogenetics: still looking for the perfect mitotic chromosomes. Nucleus 50, 453–463. Gonçalves, S., Haas, P., Spada, C., Fontana, C. S., Rangel, L., Mendonça, M. A. C., Carvalho, C. R. (2007) Citometria de imagem do conteúdo de DNA nuclear de células ´ epiteliais do colo uterino. Rev Bras de Anal Cl´ın 39, 71–78. Doležel, J., Sgorbati, S., Lucretti, S. (1992) Comparison of three DNA fluorochromes for flow cytometric estimation of nuclear DNA content in plants. Plant Physiol 85, 625–631. Johnston, J. S., Bennett, M. D., Rayburn, A. L., Galbraith, D. W., Price, H. J. (1999) Reference standards for determination of DNA content of plant nuclei. Am J Bot 86, 609–613. Thunnissen, F. B. J. M., Perdaen, H., Forrest, J. (1992) Influence of different cell extraction methods on cytometric features. Cytometry 13, 485–489. Greilhuber, J., Temsch, E. M. (2001) Feulgen densitometry: some observations relevant to best practice in quantitative nuclear DNA content determination. Acta Bot Croat 60, 285–298.

Chapter 5 Histological Approaches to Study Tissue Parasitism During the Experimental Trypanosoma cruzi Infection Daniela L. Fabrino, Grazielle A. Ribeiro, Lívia Teixeira, and Rossana C.N. Melo Abstract During acute infection with the parasite Trypanosoma cruzi, the causal agent of Chagas’ disease, tissue damage is related to intense tissue parasitism. Here we discuss histological approaches for an optimal visualization and quantification of T. cruzi nests in the heart, the main target organ of the parasite. These analyses are important to evaluate the course of the infection in different experimental models and also can be used to investigate parasite colonization and inflammatory processes in other infected tissues and biopsies. Key words: Chagas’ disease, Trypanosoma cruzi, histology, heart, inflammation, glycol methacrylate, animal models.

1. Introduction The flagellated protozoa T. cruzi is the causal agent of Chagas’ disease (also known as American trypanosomiasis), discovered around a century ago by the Brazilian physician Carlos Chagas (1). This disease remains a major problem with a great impact on public health in Latin America. Acute infections are usually asymptomatic, but the resulting chronic T. cruzi infections lead to high ratios of mortality, associated mainly with heart lesions (2, 3). This parasite has an obligate intracellular, proliferative, nonflagellate form, termed amastigote. Once the infection has established in vertebrate hosts, intracellular parasite replication occurs H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_5, © Springer Science+Business Media, LLC 2011

69

70

Fabrino et al.

as amastigotes, followed by the release of flagellate forms (termed trypomastigotes) that can be carried by the bloodstream to infect different organs, especially the heart (reviewed in (2, 4, 5)). Because of the intense parasitism in the myocardium induced by the acute Chagas’ disease, histological analyses of the heart are frequently performed during experimental studies to evaluate the course of the infection. However, the value of the histological techniques in enabling high-quality imaging of parasite nests can be limited by inadequate fixation, handling and/or processing/analysis of the tissue of interest. Here we report on the application of appropriate histological approaches for improved tissue morphology and optimal visualization and quantification of parasite nests in the heart. These approaches include the right choice and preparation of the fixative, embedding with a plastic resin for improved tissue resolution and adequate analysis of the semi-serial sections. These basic methodologies can also be used to investigate parasitism in other infected organs as well as to evaluate the presence of inflammatory infiltrates, which generally occurs in parallel to the tissue colonization by the parasite.

2. Materials 2.1. For Fixation

1. Phosphate buffer, 0.2 M, pH 7.3 (stock solution). For buffer preparation, dissolve 2.75 g of monobasic sodium phosphate (Na H2 PO4 .H2 O, MW = 137.99) into 50mL of distilled water and complete the volume up to 100 mL (final volume). Prepare another solution with dibasic sodium phosphate (Na2 HPO4 .7H2 O, MW = 268.14): dissolve 5.36 g of this salt into 50 mL of distilled water and complete the volume to 100 mL (final volume). To obtain the buffer at the pH 7.3 mix 20 mL of phosphate monobasic solution with 77 mL of phosphate dibasic solution. Adjust to pH 7.3 (see Note 1). 2. Paraformaldehyde (4 g) is dissolved in 40 mL of distilled water. Dilutions should be made in fume hood (see Note 2) under heating at a hot plate. Add 3–5 drops of 0.1 N sodium hydroxide and heat the solution until 60◦ C (do not allow going over this temperature). When the paraformaldehyde solution is completely mixed, cool to room temperature and complete the volume to 50 mL (final volume). Mix 50 mL of the paraformaldehyde solution with 50 mL of phosphate buffer 0.2 M. Adjust to pH 7.3. The final concentration of the fixative will be 0.1 M. Fresh solutions of paraformaldehyde should be used in each experiment for optimal fixation.

Histological Approaches to Study Tissue Parasitism

2.2. For Processing and Embedding

71

1. Glycol methacrylate (GMA) embedding kit prepared according to the manufacturer’s instructions. The GMA kit is composed of basic resin (solution of GMA monomer: 2-hydroxyethyl methacrylate) and other solutions such as polyethylene glycol and benzoyl peroxide. 2. Ethanol (from 70 to 100% in distilled water). 3. Plastic molds for embedding.

2.3. For Sectioning

1. Microtome. 2. Knife maker. 3. Glass knives − 400 × 25 × 6.4 mm. 4. Histological bath or large beaker. 5. Slides and coverslips.

2.4. For Staining and Mounting

1. Hematoxylin solution (Harris’ hematoxylin) can be purchased as a ready solution or alternatively be prepared. For preparation, dissolve 1 g of hematoxylin powder in 10 mL of 95% ethanol (solution I). Prepare the solution II: in a flask, to 200 mL of distilled water, add 20 g of potassium alum. Place the flask on a heater/stirrer, turn on the heater and allow mixing until the alum dissolves – this takes about 15 min. Mix solutions I and II quickly and allow boiling for 1 min. Add 0.5 g of red mercury oxide. The solution becomes dark purple. Take out the flask from the heater and allow cooling by immersing the flask in a container with cold water. After that, the solution must be stored for at least 48 h before use. The indicative sign that the solution is ready to use is the uprising of a metallic layer on the surface. Hematoxylin solution lasts for years, but eventually can deteriorate. The hematoxylin solution can be used as concentrated or diluted 1% in distilled water. To increase the nuclear contrast, 4% acetic acid can be added to the solution. 2. Two percent ammonium iron sulfate (also known as ferric ammonium sulfate or iron alum) dissolved in distilled water. 3. Yellow eosin solution is prepared by dissolving 1 g yellow eosin in 10 mL absolute ethanol and mixing this solution in 0.5 potassium dichromate dissolved in 80 mL distilled water. Just after, add 10 mL of saturated solution of picric acid (for saturation, add 1.4 g of picric acid in 100 mL distilled water). Eosin preparation should be made in fume hood. Alternatively, eosin solution can be purchased as a ready solution. 4. Acid–alcohol solution is prepared by dissolving 1 mL hydrochloric acid (HCl) (0.5–2%) plus 99 mL of 70%

72

Fabrino et al.

alcohol. Alternatively, an acid–water solution prepared with HCl or acetic acid (0.5–2%) can be used. 5. Toluidine blue-borate solution is prepared by dissolving 1 g sodium tetraborate (Na2 B4 O7 ) and 1 g toluidine blue O in 100 mL of distilled water. 6. Basic fuchsin solution (stock solution) prepared by dissolving 1 g basic fuchsin powder in 100 mL of 50% ethanol. The working solution is prepared by adding 3 mL of this solution to 60 mL of distilled water. 7. Mounting medium for histology.

3. Methods 3.1. Fixation, Processing, and Embedding

Fixation is a crucial step during sample preparation. This process is intended to preserve cell structure by avoiding tissue autolysis. In addition, fixation inhibits bacterial and fungal growth, and makes the tissue resistant to damage during the subsequent processing. Aldehyde solutions fix tissues by introducing cross-links between different tissue components (proteins, nucleic acids, and lipids). Generally, tissue fragments are fixed at least for 12 h. However, it is not recommended to fix more than 24 h. Sample collection for fixation requires a special care. Tissues must be collected as fast as possible, cut into small pieces (to enable optimal fixation) and promptly transferred to the fixative. If necessary, samples may be quickly washed in saline before cutting to clean excessive blood. It is very important to handle samples very gently to prevent mechanic damage that can lead to morphological alterations. For histological studies of the heart, it is important to define the anatomical region of interest (if atrium or ventricle, for example) before organ fragmentation. After this procedure, organ pieces are then carefully transferred to previously and adequately labeled vials containing the fixative. Fixation is followed by sample dehydration, a mandatory step before embedding. Embedding is performed by using a resin (glycol methacrylate) instead of paraffin (6). Glycol methacrylate is a water and ethanol-miscible plastic resin. One advantage of this resin is to avoid tissue damage induced by heating required for the embedding step with paraffin. Another critical advantage of plastic resin is that it permits increased tissue resolution, which is crucial to visualize fine cellular morphological details. Better resolution is due to plastic polymerization that causes less shrinkage and retraction compared to conventional techniques. Embedding with GMA has other important advantages when compared with

Histological Approaches to Study Tissue Parasitism

73

usual methods. This plastic resin is hydro-soluble, easily handled, and enables faster processing compared to paraffin embedding. Yet, it allows microtomy with section thickness varying from 0.5 to 12 µm (7, 8). The use of GMA embedding for improved histological, morphometrical, and immunohistochemical evaluations is also discussed in Chapter 1. Fix samples overnight at 4◦ C (see Note 3). Refer to Section 2.1 for fixative preparation. 1. Transfer samples to vials with cold phosphate buffer 0.1 M and, if necessary, keep at 4◦ C. Refer to Note 4 for sample storage. 2. Dehydrate samples. Dehydration step consists of successive baths in graded ethanol (70, 80, 90% – 20 min in each solution – followed by 100% ethanol (two changes of 20 min). Samples are kept in the same vial and the alcohol solution quickly replaced. Keep vials closed during the dehydration time. 3. Infiltrate samples in a first bath of resin (infiltrating resin) at room temperature, overnight. 4. Discard the resin (since it will contain alcohol) as the manufacturer’s instructions. 5. Infiltrate samples in a second bath of resin (pure resin), at room temperature, overnight. Use an amount of resin just to cover the sample. Refer to Note 5 for resin re-use. 6. After labeling molds with an identification code of your material, carefully transfer samples to within them and fill out with the embedding resin. 7. Keep molds inside a hood on a steady and plane surface for 24 h at room temperature. 8. Unblock and store blocks in a dry place. 3.2. Sectioning

For parasitism analyses, it is strongly recommended preparing semi-serial sections from the organ of interest to enable a reliable evaluation of the parasite nests within the organ. The indicated section thickness is usually 5 µm and it is important to keep a 70 µm-interval between sections to avoid parasite recounting (9, 10). 1. Set the microtome with the appropriate glass knife and knife holder for plastic resin. 2. Carefully cut the sample block. Sections come out as single sections. If you are performing semi-serial sections, collect just sections between intervals of interest. 3. Carefully transfer the selected sections to a recipient with tap water (histological bath or large becker) at room temperature. Use preferentially an antistatic tweezer to transfer

74

Fabrino et al.

sections to the water surface. Sections will stretch on the water and stay floating. 4. Collect sections with a clean slide and transfer it to a hot plate (70–80◦ C) until the water droplet evaporates (approximately 2 min). 3.3. Staining and Mounting

3.3.1. Hematoxylin–Eosin

Tissues embedded in GMA usually stain weakly with conventional methods when compared to paraffin-embedded tissues, and for this reason, some minor staining changes are required (11). Here we describe two useful and basic staining methods for tissue parasitism evaluations: hematoxylin/eosin and toluidine blue/basic fuchsin (12). Staining patterns obtained with these techniques are similar (Fig. 5.1). Both methods stain cell nuclei in dark-blue and cytoplasm and connective tissue fibers in various shades and intensities of pink (Fig. 5.1). However, different degrees of contrast can be obtained when toluidine blue/basic fuchsin is used (13). 1. Wash slides with sections in tap water for 5 min. 2. Immerse slides in the solution of iron–ammonium sulfate for 10 min. This solution acts as a mordant to increase the intensity of the hematoxylin staining. 3. Wash in tap water for 5 min. 4. Stain with Hematoxylin for 15–25 min (Note 6). 5. Rinse slides for 10 min in running tap water. 6. Immerse slides several times and quickly in an acid–alcohol or acid–water solution to enable nuclear differentiation (30 s to 1 min). Refer to Section 2.4, Section 4, for preparation of this solution (Note 7). 7. Rinse in tap water for 5 min. 8. Stain with eosin for 1–2 min. 9. Rinse in tap water for 2–5 min (Note 8). 10. Blot excess water from the slide by gently pressing the slide with a filter paper. Be careful not to touch sections. 11. Dry slides at room temperature (Note 9). 12. Mount using a conventional mounting medium for histology and let dry at room temperature or 37◦ C. (see Fig. 5.2 for the slide-mounting technique). Refer to Note 10 for removing excess mounting medium.

3.3.2. Toluidine Blue-Basic Fuchsin

1. Cover sections with a few drops of toluidine blue solution for 3–5 min on a hot plate at 55–60◦ C (see Note 11). Refer to Section 2.4, Section 5, for toluidine blue preparation. Wash in running tap water to remove excess stain. 2. Quickly dry on a hot plate (see Note 11).

Histological Approaches to Study Tissue Parasitism

75

Fig. 5.1. Light micrographs of the heart stained with hematoxylin-eosin (a, c) and basic fuchsin-toluidine blue (b, d) from uninfected (a, b) and infected (c, d) rats at day 12 of the acute infection with Trypanosoma cruzi. Histopathological analyses show parasite nests (arrows) and inflammatory processes (encircled in d). The boxed area in c shows a nest with amastigote forms of the parasite in higher magnification. Note in d a clear visualization of the cross striations in the cytoplasm of cardiomyocytes. Scale bar, 15 µm (a–c), 10 µm (d).

76

Fabrino et al.

Fig. 5.2. Procedure for slide-mounting. (1) Place the slide on a plain surface and apply a drop of a mounting medium in the center of the slide using an eyedropper. (2) Hold the cover slip at an angle of 45◦ so that it rests on the slide and touches one side of the medium drop. (3) Once the medium spreads along the coverslip edge, carefully lower the cover slip over the section and medium. Make sure there are no bubbles under the cover slip carefully. (4) Wait drying. Full drying at room temperature may require up to 48 h; however, drying can be accelerated at 37◦ C.

3. Cover sections with few drops of basic fuchsin solution (working solution) for 1 min at room temperature. Refer to Section 2.4, Section 6, for basic fuchsin solution preparation. 4. Wash well in running tap water. 5. Blot excess water from the slide by gently pressing the slide with a filter paper. Be careful not to touch sections. 6. Dry slides completely at room temperature. 7. Mount using a conventional mounting medium for histology and let dry at room temperature (see Fig. 5.2 for the slide-mounting technique). Refer to Note 10 for removing excess mounting medium. 3.4. Parasite Analysis and Quantification

Studies of tissue parasitism are dependent on histopathological analyses of a considerable number of sections of the organ of interest so as to have a reliable picture of the infection. In experimental research, it is also crucial to evaluate material obtained from at least three animals per group for optimal sampling. Parasite quantification can be performed by different methods. Our group has been using two methodologies for this purpose: enumeration of parasite nests and/or quantification of tissue areas occupied by the parasite. For both methods, we use 5 µmthickness sections. Moreover, considering that a single T. cruzi nest can reach up to 75 µm depth (9), enumeration of parasite nests must be performed on semi-serial sections with an interval

Histological Approaches to Study Tissue Parasitism

77

of 70 µm between them to avoid parasite recounting. By keeping this interval and using rats as experimental models, we found that 20–40 sections of the heart per animal (around 10–15 obtained from atrium and 15–20 obtained from ventricle) provide a consistent analysis of the organ. Trypanosoma cruzi nests can greatly vary in size. Enumeration of these nests is performed by analyzing successive fields (see Note 12) and counting the numbers of nests on a light microscope at magnification of 400× (Fig. 5.3). To compare different experimental groups, the same number of fields must be analyzed in each group. For example, in our experimental studies in rats, we generally evaluate a total of 1600 fields (800 from atria and 800 from ventricles) for each group of animals (3–4 animals). The tissue area occupied by parasites is obtained by using a microscope equipped with an integrating eyepiece usually with 100 squares at a magnification of 400× (14). The numbers of squares partially or totally occupied by the parasite are then enumerated and the proportion is calculated based on the total number of counted squares (Fig. 5.4). For example, by analyzing 145 heart sections from infected rats we found 1098 squares occupied by the parasite after counting a total of 43,000 squares. This represents an area of 2.20% occupied by the parasite. Using this methodology, it is also possible to determine the area occupied by cardiomyocytes and inflammatory processes. After enumeration, statistical analysis is usually done with a non-parametric test such as the Mann-Whitney Test (also known as U test or Wilcoxon rank-sum test) that is used to compare two independent groups of sampled data (15). Inflammation analyses can also be performed using the same sections prepared for parasitism studies. However, the quantification of inflammatory infiltrates is usually done at the magnification of 200×, which enables the observation of a higher area of tissue and a better identification of the intensity of the infiltrate and if it is focal or diffuse. On the other hand, cell features of the inflammatory and other cells are observed at 400× (Fig. 5.1).

4. Notes 1. To raise pH, add more phosphate dibasic solution. To low down pH, add more phosphate monobasic solution, always under magnetic stirring. 2. Paraformaldehyde is volatile and its fumes are very toxic (causes severe irritation to eyes, skin, and respiratory tract).

78

Fabrino et al.

Fig. 5.3. Schematic representation of the parasitism analysis on a light microscope. Correct enumeration (a) is achieved by counting parasite nests on successive fields at 400×. An incorrect analysis of fields (b) may provide a deficient picture of the infection.

Thus, any manipulation involving this chemical must be performed in a fume hood and wearing gloves. 3. To fix appropriately, aldehyde-based fixatives must be freshly prepared to avoid formic acid formation.

Histological Approaches to Study Tissue Parasitism

79

Fig. 5.4. Schematic representation of the parasitism evaluation using an integrating eyepiece on a light microscope. The numbers of squares partially or totally occupied by the parasite (marked with an “x” in the figure) are enumerated at 400× and the proportion is calculated based on the total number of counted squares.

Alternatively, these fixatives may be stored for a couple of months under minus 20◦ C. 4. After fixation, the tissue samples are transferred to vials containing phosphate buffer 0.1 M and can be stored in this solution at 4◦ C until processing. It is not recommended storing for more than 2 months. 5. The second resin can be re-used. Store at 4◦ C and re-use for the first bath of resin in another embedding procedure. 6. Filter the staining solutions immediately before their use. This procedure will avoid precipitation of sediments on sections. 7. Differentiation with acid solutions requires some practical experience to ascertain the correct end-point, since these solutions are used to bleach or decolor samples and the color of the tissue can be modified to red. The correct endpoint is when, after bluing up, the background is almost colorless. 8. Eosin is highly soluble in water. Over-staining is removed by washing in running water. 9. After staining, dehydration of sections is done just by letting sections dry at room temperature. Alcohol dehy-

80

Fabrino et al.

dration must be avoided since this solvent can damage sections. 10. After drying, any excess medium can be removed with a cotton swab moistened with a bit of xylene. 11. Slide drying on a hot plate can sometimes induce a light blue background on the resin. In this case, drying can be done at room temperature. 12. To avoid an incorrect enumeration of parasite nests, be careful not to overlap fields during the histopathological analysis on the light microscope (Fig. 5.3). References 1. Chagas, C. (1909) New human trypanosomiasis. Morphology and life cycle of Schysotrypanum cruzi, the course of a new human disease. Mem I Oswaldo Cruz 1, 159–218. 2. Melo, R. C. N. (2009) Acute heart inflammation: ultrastructural and functional aspects of macrophages elicited by Trypanosoma cruzi infection. J Cell Mol Med 13, 279–294. 3. Teixeira, A. R., Nitz, N., Guimaro, M. C., Gomes, C., Santos-Buch, C. A. (2006) Chagas’ disease. Postgrad Med J 82, 788–798. 4. Ropert, C., Ferreira, L. R., Campos, M. A., Procopio, D. O., Travassos, L. R., Ferguson, M. A., Reis, L. F., Teixeira, M. M., Almeida, I. C., Gazzinelli, R. T. (2002) Macrophage signaling by glycosylphosphatidylinositolanchored mucin-like glycoproteins derived from Trypanosoma cruzi trypomastigotes. Microbes Infect 4, 1015–1025. 5. Teixeira, A. R., Nascimento, R. J, Sturm, N. R. (2006) Evolution and pathology in Chagas’ disease – a review. Mem I Oswaldo Cruz 101, 463–491. 6. Melo, R. C. N., Rosa, P. G., Noyma, N. P., Pereira, W. F., Tavares, L. E., Parreira, G. G., Chiarini-Garcia, H., Roland, F. (2007) Histological approaches for high-quality imaging of zooplanktonic organisms. Micron 38, 714–721. 7. Bennett, H. S., Wyrick, A. D., Lee, S. W., McNeil, J. H. (1976) Science and art in preparing tissues embedded in plastic for light microscopy, with special reference to glycol methacrylate, glass knives and simple stains. Stain Technol 51, 71–97. 8. Cole, M. B., Jr., Sykes, S. M. (1974) Glycol methacrylate in light microscopy: a routine

9.

10.

11.

12.

13.

14.

15.

method for embedding and sectioning animal tissues. Stain Technol 49, 387–400. Hanson, W. L., Roberson, E. L. (1974) Density of parasites in various organs and the relation to numbers of trypomastigotes in the blood during acute infections of Trypanosoma cruzi in mice. J Protozool 21, 512–517. Melo, R. C. N., Machado, C. R. S. (2001) Trypanosoma cruzi: peripheral blood monocytes and heart macrophages in the resistance to acute experimental infection in rats. Exp Parasitol 97, 15–23. Cerri, P. S., Sasso-Cerri, E. (2003) Staining methods applied to glycol methacrylate embedded tissue sections. Micron 34, 365–372. Fabrino, D. L., Leon, L. L., Parreira, G. G., Genestra, M., Almeida, P. E., Melo, R. C. N. (2004) Peripheral blood monocytes show morphological pattern of activation and decreased nitric oxide production during acute Chagas’ disease in rats. Nitric Oxide 11, 166–174. Abreu, M., Baroza, L., Rossi, M. (1993) Toluidine blue-basic fuchsin stain for glycolmethacrylate embedded tissue. J Histotechnol 16, 139–140. Melo, R. C. N., Machado, C. R. S. (1998) Depletion of radiosensitive leukocytes exacerbates the heart sympathetic denervation and parasitism in experimental Chagas’ disease in rats. J Neuroimmunol 84, 151–157. Mann, H. B., Whitney, D. R. (1947) On a test of whether one of two random variables is stochastically larger than the other. Ann Math Statistics 18, 50–60.

Chapter 6 Intravital Microscopy to Study Leukocyte Recruitment In Vivo Vanessa Pinho, Fernanda Matos Coelho, Gustavo Batista Menezes, and Denise Carmona Cara Abstract The intravital microscopy is a valuable tool to capture images of cells in living organisms and to make studies of molecular determinants of leukocyte trafficking easier. Using this technique, we can directly visualize and measure each step of the leukocyte recruitment paradigm, including leukocyte rolling flux, rolling velocity, adhesion, and emigration. Thus, it is possible to understand the process involved in leukocyte homing as well as the cell recruitment to inflammatory tissues. Nowadays, two types of intravital microscopy are used routinely. The light microscopy is used to assess migration of intravascular cells in thin, tissues which must be sufficiently translucent. Epifluorescence microscopy allows the visualization of the microcirculation while permitting the distinction of leukocyte subpopulations in solid organs. Key words: Intravital microscopy, leukocyte, recruitment, inflammation, light microscopy, epifluorescence microscopy.

1. Introduction Intravital microscopy is an extremely useful tool used as a qualitative and quantitative method to observe leukocyte–endothelial cell interactions in vivo. For many years, the involvement of microcirculation in inflammation has been a fascinating and widely studied subject. The major point of concern at the outset was the now known involvement of the microcirculation in the inflammatory insult, but as time and techniques have progressed, the interest has moved onto leukocyte–endothelial interactions. One of the best techniques that have enabled scientists to understand these processes is intravital microscopy. The study H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_6, © Springer Science+Business Media, LLC 2011

81

82

Pinho et al.

of the microcirculation using intravital microscopy represents a sophisticated research tool to analyze complex biological interactions and disease mechanisms. This technique is also important in studies that aim to develop novel prophylactic and therapeutic approaches intending to modify microvascular disorders and cellular dysfunction associated with inflammatory diseases. Leukocyte recruitment is a multistep recruitment process that is optimally visualized and studied in vivo using intravital microscopy. Making use of this approach, we directly visualize and measure each of the steps of leukocyte recruitment paradigm, including leukocyte rolling flux, rolling velocity, adhesion, and emigration (1). The initial use of intravital microscopy date back century by Dutrochet in 1824 (reviewed in (1)). This researcher observed that leukocytes emigrated across small blood vessel walls. This finding that leukocyte emigration occurred due to tissue injury was finally reported in 1843 by Addison (reviewed in (1)). Several animals can be used for theses studies; however, rodents, especially mice due to genetic manipulations, have been more used. Using light microscopy, it is possible to assess the vasculature and leukocytes within some thin and transparent tissues without contrast-enhancing methods, such as hamster cheek pouch (2), mesentery (3), and cremaster muscle (4). In these preparations we can see leukocyte rolling, adhesion, emigration, and measure blood flux. Also, microcirculation of nontransparent organs (called generically as solid) and tissues can be studied using epifluorescence microscopy, including liver (5), spleen (6), brain (7), lymph node (8), intestinal wall (9), knee joint (10), and others. Epifluorescence microscopy can be useful for studies of leukocyte rolling and adhesion and vascular permeability (11). The intravenous injection of FITC-labeled BSA (fluorescein isothiocyanate-labeled bovine serum albumin) allows the evaluation of vascular albumin leakage from the microcirculation to the extravascular space as a parameter of vascular permeability in several tissues (12, 13). To study leukocyte recruitment, leukocytes can be labeled with dyes such as rhodamine (14) or with fluorochromes-conjugated antibodies (15). These strategies make possible the visualization of microcirculation while permitting the distinction of leukocyte subpopulations. Also, it is possible to visualize transfected cells that express the green fluorescent protein derived from transgenic mice (16). Concerning the intravital studies developed in leukocyte recruitment field, several cell structures and movements were clarified. A recent illustration of this situation is a new step on the leukocyte recruitment cascade, when cells “crawl” into endothelial layer, looking for a more adequate way to emigrate to extravascular spaces. Strikingly, the inhibition of certain subtypes of adhesion molecules can modify the way used by leukocytes to emigrate, especially, increasing the rate of transcellular

Intravital Microscopy to Study Leukocyte Recruitment In Vivo

83

emigration (through the endothelial cell) instead of the conventional intercellular extravasation (between the endothelial cells), forming indeed “dome-like” structures expressed by the endothelial cell that never had been visualized before (17, 18). The most recent advanced observation of microcirculation includes the spinning disk and the multi-photon confocal microscopy. The confocal laser scanning microscopy was an option to get rid of the out-of-focus haze of objects that can emit fluorescence. In order to improve the scanning velocity of image acquisition, a new different type of scanning method, called initially as Nipkow disk was developed and it is now currently known as spinning disk confocal microscopy. As a consequence of this improvement, a fast phenomenon would be more clearly observed as leukocyte movements under a blood flow. Indeed, lasers can have different wavelength, culminating in different fluorescence production by the stained cell or molecule, allowing different colors and co-localization of structures in a merged picture. The optical microscopy is still the only way to examine the four dimensions of a live phenomenon (the three conventional dimensions − x, y, z – allied with time course evaluation), which is very close to the realistic conditions found under physiological states. In addition to the spinning disk confocal technology, it is widely used nowadays in the acquisition of images using more than one laser wavelength simultaneously. Nowadays a twophoton microscopy has been used to study several physiological phenomena, such as for imaging electrical activity in deep tissues, for assessing the rate of blood flow, and for tracking immune-cell motility and morphology (19). The outlook of confocal intravital microscopy is very promising for physiological studies and many important advances are currently underway, like portable twophoton micro-endoscope for tridimensional imaging of mouse brains and lungs. Additional developments have been achieved on fluorescent probes and new fluorophores designed to provide a more intense bright using a lower time and a lower power of laser incidence, and improving the protection of the organ or tissue. In this chapter, we outlined techniques using a cremaster muscle preparation, as an example of light intravital microscopy, and a joint knee preparation, as an example of epifluorescence intravital microscopy.

2. Materials 2.1. Materials for Both Videomicroscopy

1. Ketamine (200 mg/kg). 2. Xylazine hydrochloride (10 mg/kg). 3. Bicarbonate-buffered saline (131.9 mM NaCl; 4.7 mM KCl; 1.2 mM MgSO4 ; 20 mM NaHCO3 ; pH 7.4).

84

Pinho et al.

4. P10 canule (Polyethylene tube). 5. Board with an optically clear viewing pedestal for cremaster (20) or knee joint preparation (21). 6. Surgery tools. 7. Cautery. 8. 4-0 suture thread. 9. Rhodamine 6G chloride and/or coupled antibodies. 10. LPS (lipopolysaccharide; sigma; B4:0111). 2.2. Light Microscopy Set

1. Light microscope with ×25 objective lens and ×10 eyepiece. 2. Video camera. 3. TV monitor. 4. DVD. 5. Optical Doppler velocimeter (Microcirculation Research Institute, Texas A&M University, College Station, TX, USA).

2.3. Epifluorescence Microscopy Set

1. Epifluorescence microscope. 2. Video camera. 3. Filter 45 MM NCP11. 4. Computer 2 DUO 8 GB. 5. Software NIS ELEMENT.

3. Methods Methods described below outline the preparation of an experimental model (mouse) for intravital microscopy (1), the surgery for cremaster preparation (2), and preparation of knee joint (3). The basic set up of intravital equipment for light intravital microscopy (see Fig. 6.1a) and epifluorescence intravital microscopy (see Fig. 6.1c) is described. 3.1. Mouse Preparation for Intravital Microscopy

Male mice are used for intravital observations. The process of leukocyte recruitment in cremaster muscle and knee joint (and also in other tissues) may be induced by administration of different inflammatory agents. Here we use LPS or specific antigens (21). 1. Anesthetize the animal with an intraperitoneal injection of a mixture of xylazine and ketamine hydrochloride. Although other anesthetics may be used, this approach generates very steady hemodynamic parameters, which is absolutely necessary to avoid blood flow effects on leukocyte recruitment.

Intravital Microscopy to Study Leukocyte Recruitment In Vivo

A

C

B

D

85

Fig. 6.1. Schematic representation of an intravital setup. a and c show the setup for light and epifluorescence intravital microscopy, respectively. In b and d is observed a cremaster muscle preparation (b) and the intra-articular synovial tissue (d) of the knee joint. Arrow indicates the synovial microcirculation. Scale bar = 0.5 cm.

2. Cannulate the left jugular vein to administer additional anesthetic or drug if necessary. 3. Administrate the inflammatory agent. For intravital microscopy in cremaster muscle, LPS is administered by subcutaneous injection beneath the left scrotal skin. For intravital microscopy in knee joint, an intra-articular injection with antigen (model of antigen-induced arthritis) is administered (21). 3.2. Intravital Microscopy of Cremaster Muscle

1. Cut the scrotal skin to expose the left cremaster muscle. 2. Carefully dissect this muscle free of the associated fascia. 3. Cut the cremaster muscle longitudinally with a cautery. 4. Separate the right testicle and the epididymis from the underlying muscle. 5. Move them back into the abdominal cavity. The muscle is held flat on an optically clear viewing pedestal and is secured along edges with 4-0 suture (Fig. 6.1b). 6. Superfuse the exposed tissue with 37◦ C warmedbicarbonate-buffered saline (0.15 M, pH 7.4) (see Note 1). 7. Examine the cremasteric microcirculation by using a light microscope equipped with ×10 or ×20 objective and a ×10 eyepiece. A video camera is used to project images onto a

86

Pinho et al.

monitor, and images are recorded for playback analysis using a DVD recorder. Single unbranched cremasteric venules (25–40 µm in diameter) are preferred and to minimize variability, the same section of cremasteric venule should be observed throughout the experiment (Fig. 6.2a).

AA

B

C

D

E

F

Fig. 6.2. Evaluation of the interaction between leukocytes and endothelial cells in the cremaster muscle microvasculation. a, b Images captured from the cremaster muscle microcirculation of mice, 4 h after intrascrotal injection with saline (a) or E. coli lipopolysaccharide (LPS, 0.05 µg/kg) (b). LPS induces a decrease of leukocyte velocity (e) and an increase of the number of rolling cells (c), adherent cells (d), and emigrated cells in the extravascular space (f). ∗ indicates a statistical difference in relation to saline group. p < 0.05, Student T test, n = 5/group. Scale bar = 50 µm.

3.3. Intravital Microscopy of Knee Joint

Unlike the cremaster muscle, the intravital microscopy is performed in the synovial microcirculation of the mouse knee. 1. Anesthetize the mouse and place the hind limb on a stage, with the knee slightly flexed and the patellar tendon mobilized and partly resected (see Fig. 6.1d). The intra-articular synovial tissue of the knee joint is then visualized for the determination of leukocyte rolling and adhesion (see Fig. 6.1d).

Intravital Microscopy to Study Leukocyte Recruitment In Vivo

87

2. Select 2–4 regions of interest (unbranched venules with 25–40 µm in diameter) in each mouse by using an ×20 objective and a ×10 eyepiece. 3. To measure the leukocyte–endothelial cell interactions, inject the fluorescent marker rhodamine 6G intravenously throughout caudal vein as a single bolus of 0.15 mg/kg immediately before measurements. Rhodamine epilumination is achieved with a 103 W/2 variable HBO mercury lamp in conjunction with a filter set (see Note 2). 4. Capture the microscopic images with a video camera and record them on a computer. Data analysis is performed offline using the imaging software NIS ELEMENT. 3.4. Evaluation of Leukocyte Recruitment

The number of rolling, adherent, and emigrated leukocytes is determined offline as cited above. Rolling leukocytes is defined as those cells moving at a velocity less than that of erythrocytes within a given vessel. The flux of rolling cells is measured as the number of rolling leukocytes passing a given point in the venule per minute. Leukocyte rolling velocity is measured for the first 20 leukocytes entering the field of view at the time of recording and calculated from the time required for a leukocyte to roll along a

B

A

C

D

Rolling

Adhesion

10.0

72

*

cells/min

36

*

18

cells/100µm

* 54

7.5 5.0 2.5 0.0

0

saline

antigen

saline

antigen

Fig. 6.3. Evaluation of the interaction between leukocytes and endothelial cells in the synovial microvasculature. a, b Images captured after intra-articular administration of saline (a) and antigen (b) in mice. Rolling cells (c) and adherent leukocytes (d) to the synovial endothelium were assessed after injection of antigen or sterile saline (control) into the knee joint of mice. ∗ indicates a statistical difference in relation to saline group. p < 0.05, Student T test, n = 5/group. Scale bar =50 mm.

88

Pinho et al.

100 µm length of a venule. A leukocyte is considered to be adherent if it remained stationary for at least 30 s, and total leukocyte adhesion is quantified as the number of adherent cells within a 100 µm length of venule in 5 min (see Figs. 6.2 and 6.3). For light intravital microscopy, it is possible also to count leukocyte emigration (see Note 3 and Fig. 6.2). Leukocyte emigration is quantified as the number of cells in the extravascular space on the visible field adjacent to the observed venule. Only cells adjacent to and clearly outside the vessel under study should be counted as an emigrated leucocyte. All these parameters can be altered, for example, 4 h after the intrascrotal injection of LPS, as shown in Fig. 6.2. Mean red blood cell velocity can be measured using an optical Doppler velocimeter, which uses a pair of photodiodes to generate a voltage from an image of moving red cells that is a linear representation of red cell velocity. Wall shear rate is calculated based on the Newtonian definition as (mean red blood cell velocity/diameter) × 8 (s−1 ) (22).

4. Notes 1. The prevention of bleeding is crucial. Excessive bleeding will affect mainly the number of circulating leukocytes, and so it is crucial to assess the total leukocyte count after every experiment. During the whole experimental protocol, the animal should be kept under physiological conditions. So, dehydration, volume lost, hypothermia should always be avoided, especially on those preparations where the peritoneum cavity is exposed (liver, intestine, mesentery, etc.) and an important modification on animal parameters is induced. The cover of exposed organs that will not be observed with a soaked gaze, napkin of a PVC-wrap will minimize the water and temperature loss, and even reduce the movement of organs. Importantly, frequent and strong movement caused by animal breathing or instability of the microscope components will make the correct analyses of the data difficult. A warmer lamp (infrared) is strongly recommended in rooms where the temperature is lower than the mice (about 25–30◦ C) and in peritoneum-opened procedures. 2. Especially related to fluorescence microscopy, the continuous exposure of the same field for many minutes can cause “bleaching” of the fluorescence and incorrect analyses. The stability of the fluorophore should be evaluated

Intravital Microscopy to Study Leukocyte Recruitment In Vivo

89

in accordance with exigencies of the experimental protocol. Finally, as the leukocyte recruitment will begin since the first cells are injured, the improvement of the operatory skills will avoid unexpected leukocyte recruitment and will afford mainly the reproducibly of the data. An exceedingly traumatic operatory approach and even an incorrect administration of the pro-inflammatory mediators (dose and vial) will interfere in the local blood flow, which is a sine qua noncondition to leukocytes to get vessels on the field of view. 3. Light microscopy illuminates a translucent tissue and this facilitates the emigration enumeration. However, in the epifluorescence evaluation of solid organs, it is hard to clearly see cells outside microcirculation. Additionally, the leakage of rhodamine from the microcirculation to the extravascular space makes it difficult to track cells present in extravascular space. References 1. Springer, T. A. (1994) Traffic signals of lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76(2), 301–314. 2. Duling, B. R. (1973) The preparation and use of the hamster cheek pouch for studies of the microcirculation. Microvasc Res 5(3), 423–429. 3. Kvietys, P. R., Perry, M. A., Gaginella, T. S., Granger, D. N. (1990) Ethanol enhances leukocyte–endothelial cell interactions in mesenteric venules. Am J Physiol 259, 578–583. 4. Goldberg, M., Serafin, D., Klitzman, B. (1990) Quantification of neutrophil adhesion to skeletal muscle venules following ischemiareperfusion. J Reconstr Microsurg 6(3), 267–270. 5. Vollmar, B., Glasz, J., Menger, M. D., Messmer, K. (1995) Leukocytes contribute to hepatic ischemia/reperfusion injury via intercellular adhesion molecule-1mediated venular adherence. Surgery 117(2), 195–200. 6. Schmidt, E. E., MacDonald, I. C., Groom, A. C. (1990) Interactions of leukocytes with vessel walls and with other blood cells, studied by high-resolution intravital videomicroscopy of spleen. Microvasc Res 40(1), 99–117. 7. Mooradian, A. D., McCuskey, R. S. (1992) In vivo microscopic studies of age-related changes in the structure and the reactivity of cerebral microvessels. Mech Ageing Dev 64(3), 247–254.

8. Von Andrian, U. H. (1996) Intravital microscopy of the peripheral lymph node microcirculation in mice. Microcirculation 3(3), 287–300. 9. Sekizuka, E., Benoit, J. N., Grisham, M. B., Granger, D. N. (1989) Dimethylsulfoxide prevents chemoattractant-induced leukocyte adherence. Am J Physiol 256, 594–597. 10. Veihelmann, A., Szczesny, G., Nolte, D., Krombach, F., Refior, H. J., Messmer, K. (1998) A novel model for the study of synovial microcirculation in the mouse knee joint in vivo. Res Exp Med 198(1), 43–54. 11. Hulström, D., Svensjö, E. (1979) Intravital and electron microscopic study of bradykinin-induced vascular permeability changes using FITC-dextran as a tracer. J Pathol 129(3), 125–133. 12. Kubes, P., Gaboury, J. P. (1996) Rapid mast cell activation causes leukocyte dependent and -independent permeability alterations. Am J Physiol 271, 2438–2446. 13. Cara, D. C., Ebbert, K. V., McCafferty, D. M. (2004) Mast cell-independent mechanisms of immediate hypersensitivity: a role for platelets. J Immunol 15, 4964–4971. 14. Baatz, H., Steinbauer, M., Harris, A. G., Krombach, F. (1995) Kinetics of white blood cell staining by intravascular administration of rhodamine 6G. Int J Microcirc Clin Exp 15(2), 85–91. 15. McDonald, B., McAvoy, E. F., Lam, F., Gill, V., de la Motte, C., Savani, R. C., Kubes, P. (2008) Interaction of CD44 and hyaluronan is the dominant mechanism for neutrophil

90

Pinho et al.

sequestration in inflamed liver sinusoids. J Exp Med 205(4), 915–927. 16. Stein, J. V., Rot, A., Luo, Y., Narasimhaswamy, M., Nakano, H., Gunn, M. D., Matsuzawa, A., Quackenbush, E. J., Dorf, M. E., von Andrian, U. H. (2000) The CC chemokine thymus-derived chemotactic agent 4 (TCA-4, secondary lymphoid tissue chemokine, 6Ckine, exodus-2) triggers lymphocyte function-associated antigen 1-mediated arrest of rolling T lymphocytes in peripheral lymph node high endothelial venules. J Exp Med 191(1), 61–76. 17. Phillipson, M., Kaur, J., Colarusso, P., Ballantyne, C. M., Kubes, P. (2008) Endothelial domes encapsulate adherent neutrophils and minimize increases in vascular permeability in paracellular and transcellular emigration. PLoS One 3(2), 1649. 18. Phillipson, M., Heit, B., Colarusso, P., Liu, L., Ballantyne, C. M., Kubes, P. (2006) Intraluminal crawling of neutrophils to emigration sites: a molecularly distinct process

19.

20.

21.

22.

from adhesion in the recruitment cascade. J Exp Med 203(12), 2569–2575. Benninger, R. K. P., Hao, M., Piston, D. W. (2008) Multi-photon excitation imaging of dynamic processes in living cells and tissues. Rev Physiol Biochem Pharmacol 160, 71–92. Cara, D. C., Kubes, P. (2004) Intravital microscopy as a tool for studying recruitment and chemotaxis. Methods Mol Biol 239, 123–132. Coelho, F. M., Pinh, V., Amaral, F. A., Sachs, D., Costa, V. V., Rodrigues, D. H., Vieira, A. T., Silva, T. A., Souza, D. G., Bertini, R., Teixeira, A. L., Teixeira, M. M. (2008) The chemokine receptors CXCR1/CXCR2 modulate antigen-induced arthritis by regulating adhesion of neutrophils to the synovial microvasculature. Arthritis Rheum 58(8), 2329–2337. Borders, J. L., Granger, H. J. (1984) An optical doppler intravital velocimeter. Microvasc Res 27(1), 117–127.

Section II Fluorescence Microscopy Applications

Chapter 7 Introduction to Fluorescence Microscopy Ionita C. Ghiran Abstract This chapter is an overview of basic principles of fluorescence microscopy, including a brief history on the invention of this type of microscopy. The chapter highlights important points related to properties of fluorochromes, resolution in fluorescence microscopy, phase contrast and fluorescence, fluorescence filters, construction of a fluorescence microscope, and tips on the correct use of this equipment. Key words: Fluorescence microscopy, fluorochromes, fluorescence filters.

1. A Brief History of Fluorescence Microscopy

The invention of fluorescence microscopy was the result of a series of fortunate but largely unrelated discoveries that culminated at the beginning of the 20th century. Long before, the curious properties of various minerals and especially plant extracts to either emit light when kept in the dark (phosphorescence) or reflect a Disclaimer: When preparing the manuscript I had to use examples that were pertinent to fluorescence microscopy and, therefore, to use actual fluorescence microscopes, objectives, cameras and image acquisition and analysis software. With one exception (the Nikon 63 × 1.49 objective, which was borrowed from Perkin-Elmer), all the other instruments and accessories were already present in the laboratory. The fact that I used those particular instruments to record images present in this chapter does not constitute a quality judgment, either positive or negative, about their optical and mechanical performances. The decision about purchasing a certain brand over another is a complex process that has to balance quality, price, previous experience, technical support and idiosyncratic personal preference.

H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_7, © Springer Science+Business Media, LLC 2011

93

94

Ghiran

certain color and transmit a different one when illuminated with sun light (fluorescence) were well known. A special place in the history of fluorescence is reserved for Nicolas Monardes, a 16th century Spanish botanist who wrote in “La historia medicinal de las cosas que se traen de nuestras Indias Occidentals” (“The medical history of the objects that are brought from our Occidental Indies”) about the peculiar properties of the yellow aqueous extract of the kidney wood (Lignum nephriticum) that had a distinct blue tint (fluorescence) in reflected light. However, the person most responsible for a scientific approach to the fluorescence phenomenon was George Stokes, an Irish physicist who noticed in 1852, while working at Cambridge University that the mineral fluorspar (or fluorite, CaF2 ) when illuminated with blue light, emitted red light. To describe this phenomenon he coined the term “fluorescence,” 1 based on the name of the mineral he used for his experiments. He was also the first to notice that fluorescent substances emitted light at a longer wavelength than the excitation light, describing for the first time the shift in wavelength that now bears his name. 1.1. Invention of Fluorescence Microscopy

At the same time physicists were trying to understand the phenomenon of fluorescence, a race for microscope objectives with improved resolution was underway in England and Germany. Before long (around 1886), the highest practical numerical aperture (NA) of objectives was reached based on Abbe’s (1872) published relationship between resolution and wavelength. It then became clear that one could produce higher resolution images with a given objective if light with a shorter wavelength (i.e. blue, violet and ultraviolet) was used to illuminate the object. Although invisible to the human eye, the existence of UV light had been known at that time for over 70 years due to research conducted in 1801 by the young German physicist, Johann Wilhelm Ritter (1776–1810). He noticed while studying the response of silver chloride (AgCl) to light that the intensity of the reaction increased as the interacting color was closer and even beyond the left end of the visible spectrum (blue–violet–ultraviolet) and decreased towards the right end (red and infra-red). Using blue and UV light appeared to be a good method to increase microscope resolution, but there were several significant drawbacks that made implementation difficult. First, ordinary glass is not very transparent beyond 380 nm; therefore, new quartz lenses had to be designed and employed for UV microscopy. Next, because the eye is not sensitive to UV light, images had to be focused using

1

The name fluorite describes the crystal’s low melting point, from the latin “fluo”, which means to flow that is also the origin of the word fluid.

Introduction to Fluorescence Microscopy

95

visible light. Due to uncorrected axial chromatic aberrations of objectives, out-of-focus images were recorded when structures of interest were photographed using UV light results. Finally, experimenting with UV imaging was potentially hazardous because all microscopes at that time used trans-illumination that exposed the observer directly to dangerous UV light. In 1904, while experimenting with blue and UV light, August Kohler (a German physicist who 10 years earlier, at only 27, described a new illumination method that became the gold standard for all microscopes produced since) noticed that certain tissues fluoresce when illuminated with UV light. He was the first to describe primary fluorescence (auto-fluorescence) of tissues, although he reported it initially as a nuisance rather than a discovery with possible practical applications. In 1913, a dedicated fluorescence microscope was built by Heinrich Lehmann and Stanislaus Josef Mathias von Prowazek that could observe and also measure fluorescence signals. However, the breakthrough that first demonstrated the exceptional potential of fluorescence microscopy in biology had to wait until 1941 when an American immunologist, Albert Hewett Coons, successfully used anthracene–isocyanate and later fluorescein to directly label pneumococcal anti-serum. Results obtained, although exceptional, were not well received by the scientific community due to low specificity (especially when using whole serum) and poor signal amplification when using monoclonal antibodies labeled with the rapidly fading fluorescein (1). A second significant breakthrough happened in 1954 when Thomas Weller (who received in the same year the Nobel prize in Medicine), working together with Albert Coons, developed and used succesfully an indirect method of staining, comprised of a non-labeled primary and a fluoresceinlabeled secondary antibody, to identify viral particles in culture (1). It is difficult to name another method that shaped cell biology, molecular biology and immunology more than the use of fluorescently labeled antibodies. Whether fluorescence microscopy, flow cytometry, real-time PCR or gene array, all these methods rely on the use of a fluorescently labeled probe (antibody, DNA, or RNA fragment) that specifically and quantitatively recognizes an epitope. Even the use of GFP-tagged proteins stems from the same basic idea.

2. What Is Fluorescence? Fluorescence is the property of atoms and molecules to emit light following excitation by an outside source of energy. A molecule capable of fluorescence is called a fluorochrome. Fluorescence

96

Ghiran

Fig. 7.1. Jablonski diagram of a theoretical fluorescent molecule showing most of the possible electronic transitions during excitation and emission (see text for details).

occurs when excess energy released by electrons returning to the ground state is released as photons (2). The illustration describing this process is known as the Jablonski diagram (Fig. 7.1), after the Polish physicist Aleksander Jablonski who used it first in 1935 to describe the transition of electrons during absorption and emission of light. Because fluorescence microscopy is based on fluorescence of organic molecules rather than atoms, we will use molecular fluorescence to discuss the physical basis of fluorescence. In steadystate conditions, electrons are found in the lowest energy level, S0 , or the ground state. When an organic substance absorbs photons with sufficient and appropriate energy, electrons from S0 transit to higher energy states called excited single states, S1 , S2 or Sn . The transition phase from S0 to S1 or S2 is very fast and takes about 10−15 s. The excited single states of organic molecules are not as discreet as those of atoms, so the energy levels become energy bands. The direct consequence of this is the very broad excitation and emission spectra of organic fluorochromes (Fig. 7.2) compared to the sharp peaks of the emission spectrum characteristic of individual atoms (see below). The ground or excited state is called single because all electrons on these bands are spin-paired. When the excited electron returns to the ground state it can take several paths depending on the molecular configuration of the molecule, and the presence of other species of molecules in the media, can result in 1. fluorescence 2. phosphorescence 3. radiation-less conversion (heat)

Introduction to Fluorescence Microscopy

97

Fig. 7.2. Excitation and emission spectra of a hypothetical fluorescent molecule illustrating the red shift (Stokes’s shift) of emission wavelength compared to excitation. The amount of the shift depends on the molecular composition of the molecule. Fluorochromes with larger shifts are desirable for fluorescence microscopy.

When returning from an excited state, electrons usually go from one energy level to the next energy level down, releasing excess energy as heat in a process known as internal conversion. Because the energy levels S2 and S1 are very close, the probability that an electron from S2 will bypass S1 and go directly to S0 is significantly lower than going from S2 to S1 . Therefore, most fluorescence happens when electrons return from S1 to S0 . The difference in energy between S1 and S0 is the energy of the emitted photon (Kasha’s rule). In other cases the energy lost by an electron going from S1 to S0 can be given up as heat. In this situation there will be a fluorescence-less process, where no photons are emitted by the organic molecule. The energy state of electrons depicted on the right side of the Jablonski diagram is labeled “triplet state,” denoting that one set of electron spins is unpaired. This happens when spin of an electron from one of the higher vibrational energy levels (S1 or S2 ) is reversed, switching the electron to the triplet state through a process called intersystem crossing. The triplet state is depicted below the corresponding singlet state with the same electronic configuration because it has a slightly lower energy level. Therefore, the lowest excited state of a molecule is often a triplet state. Although the return from this state to the ground state does not result in fluorescence but rather phosphorescence or heat, the triplet state is important because in this state electrons are quite reactive, leading to intermolecular reactions that cause a decrease in fluorescence efficiency (photobleaching) and to interactions with oxygen that produce cell-toxic, reactive intermediate species (3). 2.1. The Chemical Basis of Fluorescence

Certain organic molecules fluoresce whereas others do not. What makes molecules fluorescent is the presence of conjugated, alternate double bonds that allow electrons to be displaced through

98

Ghiran

the rigid and planar molecule. Therefore, the benzene ring (C6 H6 ) is a very common group among fluorescent molecules. Usually there is a correlation between the size of the molecule (the number of double bonds) and wavelengths of the excitation and emission states, specific for those particular molecules.

3. Properties of Fluorochromes 3.1. Fluorescence Quantum Yield

Fluorescence Quantum Yield (quantum efficiency, QE) is a measure of the efficiency of the fluorescence process. It is defined as the ratio between the number of emitted photons and the number of absorbed photons. This does not mean that in the case of fluorescein that has a quantum yield of 90%, the energy of the emitted photons is 90% of the energy of the absorbed photons. It means that number-wise, there is a 10 to 9 relationship between excitation and emission. The loss in energy is due to non-radiative processes (Stokes’s losses) that render the emitted photon to be red-shifted compared to the excitation photon. However, the QE of fluorescein varies greatly under different conditions, especially pH. At a pH of 8, fluorescein has the highest QE; whereas as the pH drops towards 6, the QE reaches 0.3. This is important because the same amount of fluorescein will fluoresce brightly outside the cell and be very dim once inside the cell in an acidic compartment (i.e. phagosome). Therefore, one should avoid fluorescein alone to track phagocytosed particles (or any biological processes where the pH varies) and choose a pH-insensitive dye. On the other hand, tagging a particle with both fluorescein and a pH-insensitive dye represents a very powerful method that allows one to track particles intracellularly and simultaneously monitor pH changes.

3.2. Quenching

Quenching is defined as a decrease in the QE due to interactions with molecular species nearby such as proteins or fluorochromes. The energy transfer is non-radiative and results in less efficient fluorescence. The emitted light of a quenched fluorochrome displays identical spectral characteristics to the non-quenched fluorochrome. Commonly, fluorochromes lose some of their QE following conjugation to proteins such as BSA (bovine serum albumin) or IgG (immune globulins) if the conjugation density of the fluorochrome is too high. Therefore, it is important to keep the molar ratio fluorochrome-to-protein low rather than high to prevent quenching due to fluorochrome–fluorochrome interactions. (Each company that sells fluorescence labeling kits has an optimal range depending on the size of the fluorochrome and the size of the protein.) Other molecules such as ANS

Introduction to Fluorescence Microscopy

99

(8-anilino-1-naphthalene sulfonate) behave differently, displaying low fluorescence in polar solvents and increased fluorescence upon interaction with various proteins. Quenching is not always detrimental; it may offer useful information about the nearby species as well as the distance between them, which is always at least an order of magnitude smaller than the resolution of a standard fluorescence method. 3.3. Photobleaching

Unlike quenching, which can be reversible, photobleaching is generally irreversible. This process is due to electrons transitioning from single to the triplet state; electrons will interact efficiently with other molecular species with triplet ground states such as the ubiquitously present oxygen. The net result is an irreversible change in the molecular structure of the fluorochrome, which results in a permanent loss of fluorescence. Another side effect of this interaction is the formation of reactive oxygen species in or around the fluorescently labeled cell that will chemically interact and damage cellular structures. In our experience, very low concentrations of ascorbic acid (nanomolar range) offer a cell friendly way to prevent photobleaching and oxygen-mediated cell damage.

3.4. Molar Extinction

Molar extinction (ε) (synonyms: molar absorption or molar absorptivity), as defined by the Beer-Lambert law, measures the efficiency of a fluorochrome (or any substance) to absorb light at a certain wavelength (λ). Importantly, the absorption depends on the molecular structure of the molecule, the wavelength of the absorbed light, pH value and temperature. Formula: A = εcl where A=absorbance of the homogenous, isotropic sample ε = molar extinction c = concentration of fluorescent dye (mol/L) l = thickness of the fluorescent solution, expressed in cm (although theoretically it should be expressed in meters)

3.5. Fluorescence Lifetime

Fluorescence lifetime represents the average time a molecule spends in the excited state before returning to the ground state. In Fig. 7.1, we saw that following excitation, molecules transit from the ground state to excited states (Sn ). The deactivation phase is a two-step process that includes a non-radiative (internal conversion) and a radiative (fluorescence) process that depopulate the excited phase. If there is no energy transfer to the environment (no acceptors), the fluorescence lifetime is constant for

100

Ghiran

a given molecule in a given solvent and is usually in the order of nanoseconds. However, when acceptors such as oxygen, calcium, magnesium or hydrogen are present, the lifetime of fluorochrome is significantly decreased. Measuring fluorescence lifetime (not general fluorescence) can offer precise information about the environment surrounding the fluorochrome that allow researchers to map cells with unsurpassed space and time resolution for parameters such as: calcium, magnesium and pH, as well as for the presence of certain quenching molecules. There is little doubt that fluorescence lifetime imaging will be a widespread fluorescence microscopy method in the years to come. 3.6. Stokes Shift

4. Resolution in Fluorescence Microscopy

Stokes shift represents the shifted emission of a fluorochrome compared to excitation (Fig. 7.2). When the excitation and emission spectra are displayed together the Stokes shift is represented by the distance between the two peaks. The curve on the left side of graph is called the excitation spectrum and in general (but not always) is identical with the absorption spectrum of the fluorochrome. When the spectrum of excitation is very different than that of emission, it means that non-fluorescent species, likely dimers, are present in the solution (4). A better representation of the excitation curve would be against the molar extinction coefficient, which is the wavelength-dependent absorptivity (see above) of that particular fluorochrome in the solvent used to generate the spectra. However, because excitation and emission spectra are generally displayed on a single graph, fluorescence excitation is used against various wavelengths. The emission spectrum is a plot of the fluorescence intensity versus various wavelengths. The shape of the emission plot does not depend on the excitation wavelength, but only on the molecular identity of the fluorochrome due to Kasha’s rule (see above).

Classically, resolution of an optical microscope is defined as the minimal distance (expressed in µm) at which two points in the sample are seen separately. Ernst Abbe showed in 1872 that the minimal distance between two points (d) that are imaged using transmitted light (a regular bright-field microscope) depends on the half-angle of the aperture of the objective (sinα), the wavelength (λ) of the light used to form the image and the refractive index (n) of the media between the frontal lens and the sample (or coverglass). While the aperture (opening) of an objective is measured in degrees, the NA is an adimensional value that depends on the aperture of the objective and the refractive

Introduction to Fluorescence Microscopy

101

index of the media between the sample and the frontal lens. The smallest detail resolved with an objective with a NA of 1.40 (the highest possible number until a few years ago without using special immersion oil and special cover glasses; the value now is 1.49, used with green light (λ = 550 nm) and with an immersion condenser (NA = 1.40), so called double immersion) on a perfectly aligned microscope would be about 220 nm. d=

λ λ = NAcond + NAobj 2NA

Anyone who has ever used a fluorescence microscope knows that when using fluorescence, it is possible to see light originating from structures that are about 10 times smaller than the theoretical resolution of the best immersion lens. Even with a dry objective (40×), it is possible to see quantum dots, which are fluorescent crystals the size of a ribosome (about 20 nm). So what is the explanation? In fluorescence microscopy, the resolution does not measure the size of the smallest self-luminous point it can detect but rather the radius of the image formed by the objective that represents that self-luminous point. If one images three particles having 20, 100 and 180 nm in diameter, the microscope will see them as having the same size. The intensity associated with particles may be different depending on their actual size but as far as the final size the microscope will show, all particles will be the same. In the same way, we can see on a dark night, a lit candle miles away or stars that are hundreds of light years away from us although our eyes do not have the angular resolution needed to resolve neither the flame nor the disk of the stars at that distance. What we see is the light coming from the flame or stars because of the dark background. Notably, the light seen with unaided eyes will contain no details regarding the shape or original size of the luminous points. Similarly, fluorescence microscopy allows us to detect the presence of fluorescent particles that are significantly below the resolution limit of the objective without any indication of structure, shape or actual size. The initial work regarding the resolution of optical instruments imaging self-luminous objects was done using astronomical instruments (stars are perfect selfluminous points). Therefore, not surprisingly, the theory explaining the resolution of fluorescent microscopy is based on the work of a British astronomer George Airy. He was the first to show that when in focus, the image of a star seen in a telescope is not a point but rather a small bright disk that is surrounded by larger and ever fainter bright disks alternating with dark disks. Based on Airy’s work, Lord John Rayleigh (who received the Nobel Prize in 1904) derived an empirical formula that shows that in fluorescence microscopy the smallest fluorescent structure will form a disk with a radius of

102

Ghiran

r=

0.61λ 0.61λ 1.22λ = = NAcond + NAobj NA n sin α

In this formula, λ is the wavelength of the fluorescence (emitted light), α is the half-angle of the objective aperture and n is the refractive index of the media between the frontal lens and the sample. In fluorescence microscopy, the objective is the condenser; therefore, the NA of the condenser will be equal to the NA of the objective. Rayleigh resolution criteria also states that in order for two self-luminous points to be resolved, the center of one object has to be on the first minimum (the first dark ring that surrounds the airy disk) of the adjacent point. Although this depends significantly on one’s ability to resolve different shades of gray, novel digital methods used to record and process microscopy images help standardize and even exceed this criterion (5).

5. The Construction of a Fluorescence Microscope

Current fluorescence microscopes are based on the design used by Johan Ploem that was developed while working together with Leitz in the early 1960s.2 Today, virtually all fluorescence microscopes are based on Ploem’s design, and, therefore, we will present it in the following pages. The simplest way to represent the basic fluorescence microscope design is shown in Fig. 7.3a. The light emitted by the fluorescence light source (excitation) bounces off of a dichromatic mirror, passes through the objective and excites fluorochromes in the sample. Due to the nature of the dichromatic mirrors that behave like true mirrors for shorter wavelengths and are transparent to longer wavelengths∗ , the fluorescent light (emission) originating in the sample, passes unreflected through the dichromatic mirror towards the detector, which can be the eye or an electronic device (i.e. CDD camera). Some spinning disk confocal microscopes have dichromatic mirrors that allow shorter wavelengths to pass through and reflect the fluorescent light to the detector. Figure 7.3b shows the fluorescence path of a commercial fluorescence microscope along with the bright-field in the lower part of the microscope. Although the

2

Due to poor circulation of scientific journals at that time and lack of translation of most scientific articles written in languages other than English, neither Ploem nor researchers from Leitz were aware that more than 5 years before, two Russian scientists, Brumberg and Krylova, did not only described the advantages of employing “dividing mirrors” in fluorescence microscopy but actually used them.

Introduction to Fluorescence Microscopy

103

Fig. 7.3. Diagram illustrating the basic principle of epi-fluorescence microscopy. The light source emits light that is reflected by the dichromatic mirror through the objective into the sample. The fluorochromes present in the sample following excitation will emit light of longer wavelength (fluorescence) that will go back through the objective and dichromatic mirror and form the final image on a detector that can be either the eye or a CCD camera.

presence of a series of lenses, diaphragms and filters in the fluorescence path make the design of the fluorescence microscope fluorescence path seem more complicated than the schematic, the basic principles apply. Therefore, to understand how a fluorescence microscope works, and more importantly be able to get the most out of it and troubleshoot potential problems, we shall start by describing components of a fluorescence microscope following the light path, as shown in Fig. 7.4a. Although it is not obvious, a fluorescence microscope is basically a folded bright-field microscope (see Fig. 7.4a), where the objective lens play the role of both the condenser and the objective. As a rule, in epi-fluorescence microscopy everything involved in image formation is located above the microscope slide. None of the controls that are located under the microscope stage, such as the ones that modify the condenser height, the size of the field and aperture diaphragms, the position of neutral density, or green interference filters affect the fluorescence image. In some particular instances, the height of the condenser may have a negative impact on the image (see below), but in general all of the adjustments for the fluorescence microscope are located on the fluorescence illuminator. On a bright-field microscope, the light that forms the image originates from the halogen light bulb located for upright microscopes in the base of the microscope. Less intuitive, in fluorescence microscopy the light that forms the final image is generated and emitted by fluorochromes present in the sample and not by the light source located at the back-end of the fluorescence illuminator. In a functional epi-fluorescent microscope,

104

Ghiran

Fig. 7.4. a Diagram of the light paths (epi-fluorescence and trans-illumination) of a modern Olympus microscope. All the optical and mechanical components involved in fluorescence image formation are located above the stage. The light source can be either attached directly to the illuminator or be connected to the microscope through a liquid light guide in which case it will be away from the microscope body (see text). b Comparison between the apertures of a 10× and 40× objective at the imaging distance. While a 10× objective has a significantly larger frontal lens than a 40× objective, it is also located further away from the sample, which translates into a smaller actual imaging aperture. An objective with smaller aperture will be less efficient in gathering photons and thus will render images with less resolution.

none of the light that originates from the light source will ever reach the detector. Because a fluorescent structure is formed by numerous self-luminous points, image formation in fluorescence microscopy have, in some respects, more in common with images formed by an astronomical instrument than with images formed by a bright-field microscope. Each fluorescence image is generated by a large number of individual self-luminous points of light or, more correctly, disks of light (see above) that scatter light in all directions around them. Only a fraction of this scattered light will be captured by the objective lens. The amount of light gathered by a lens is directly proportional to the opening (aperture) of the objective at the imaging distance of the lens. This is important because although a 10 × 0.25 objective has a significantly larger frontal lens than a 40 × 0.95 lens, the angle subtended by the lens at the working distance is only about 15◦ for the 10× lens and around 71◦ for the 40× lens (Fig. 7.4b). This is possible because the working distance of the 10 × 0.25 objective is about 6 mm, whereas for the 40 × 0.95 it is about 0.14 mm. In other words, the larger the angle of the objective

Introduction to Fluorescence Microscopy

105

frontal lens at the imaging distance, the more light will enter the objective and the more photons will contribute to the image (see Section 4). 5.1. Light Sources or Illuminators

A light source suitable for fluorescence imaging has to deliver a beam of light that is homogenous (uniform output), high power, constant power output in time (no flickering) and contains all wavelengths that are needed to excite fluorochromes present in a given specimen. Today, most fluorescence microscopes are equipped with one of the three main types of illuminators: mercury, xenon or metal halide lamp. Lamps are made out of fused silica quartz round or elliptical envelopes that are strong enough to sustain high temperature and extreme pressures. Mercury lamps and xenon lamps are very common with most microscopes 5–10 years and older, with their lamp housing physically attached to the microscope (see Fig. 7.3b). Newer microscopes are fitted with metal halide illuminators or LED light sources separated from the microscope stand using liquid light guides (3–5 ft long) to deliver the light to the fluorescence illuminator. Mercury and xenon lamps deliver a very bright and continuous spectrum from UV to infrared (Fig. 7.5a and b). The power of the lamps varies usually between 50 and 200 W for mercury lamps and 75 and 150 W for xenon lamps. A closer inspection of the mercury bulb output graph shows that at certain wavelengths the emission is significantly greater than others, such as 365, 405, 436, 546 and 579 nm. Most of the fluorochromes that were synthesized in the past and are still in use today were centered on these wavelengths to maximize the efficiency of the mercury light source. Additionally, most of the objectives used for wide-field fluorescence or confocal microscopy were also designed to offer maximum correction for these particular wavelengths. Interestingly, the most used fluorochrome, fluorescein, has its excitation maxima at 494 nm where the mercury lamp has no major emission lines. However, due to its high quantum efficiency and broad excitation spectra, fluorescein is bright even when excited with a mercury lamp. Fluorescein has several disadvantages, such as fast photobleaching and high pH sensitivity, which makes it less suitable for fluorescence microscopy but more appropriate for FACS (fluorescence-activated cell sorting), where the interaction of the laser beam with the fluorescein is brief enough such that bleaching is not an issue. Due to its characteristic larger distance between electrodes, mercury bulbs fill up the back focal plane of the objective significantly more compared to xenon sources, allowing a uniform illumination of the specimen even at small magnifications. On the other hand, mercury lamps have short life spans (about 200 h) and exhibit a steady decrease of power output due to carbon deposits that begin to appear after the first hours of use.

106

Ghiran

Fig. 7.5. Spectral characteristics of various light sources. a Mercury lamp, b xenon lamp and c metal halide.

Introduction to Fluorescence Microscopy

107

Xenon lamps have a more homogenous output throughout the spectrum without any obvious emission peaks (Fig. 7.5b). The emission in the visual range is greater than that of the mercury lamp, which is an advantage in certain fluorescence applications or when using particular fluorochromes. However, the output in the red and infra-red part of the spectrum is also considerable (almost half of the entire emission spectrum); therefore, the heat generated by these sources is also significant and creates problems during long time-lapse experiments. The heat produced by the light source (xenon or mercury) is transmitted by the microscope frame to the stage and is reflected as an “x−y” drift of the image. To minimize this problem, certain companies (i.e. Olympus) manufacture microscope stands out of a metal– ceramic alloy that has a minimal thermal expansion coefficient and allows for long time experiments with a minimal heat-dependent drift. Another efficient approach to correct for this problem is to use a liquid light guide that allows a complete separation between the heat source (halogen or mercury/xenon lamp) and the microscope body. A common problem for most fluorescence light sources is the variability in the power output (seen as flickering of the excitation and emission light) due to either variation in the output of the power source or lack of stable connections between electrodes when the lamp is on. While correcting for the latter is more difficult and requires the use of stabilized power sources (more expensive), the former is easier to prevent (never to correct) by simply allowing a new bulb to “burn in” when turned on for the very first time for at least one hour. By doing so, the discharge arc will etch stable contact points (small pits) on the electrodes. On these two points (one on each electrode), the resistance will be lower than on any other area of electrodes. Therefore, the next time the lamp is turned on, the discharge arc will form quickly and stably between the two points generating a steady output of light. Subsequently, if the lamp is turned off before “burn in,” the next time the lamp is turned on (and for the remaining of the 200 h) the arc will “look” for areas with lower resistance on the electrodes, producing quick drops in output as it jumps from one point to another. These drops in output are seen by the user as “flickering” of the fluorescence. The main disadvantage of flickering is the inability to generate any quantitative data using fluorescence microscopy, especially during time-lapse experiments, as flickering of the lamp is completely random. Metal halide lamps are becoming more and more popular because they have all of the advantages of mercury lamps and few, if any, of their disadvantages. The emission spectra is almost identical but appreciably brighter, lamps last significantly longer (2000 has opposed to 200 h) and are almost flicker-free due to

108

Ghiran

a stabilized power supply. Most lamps are sold with a liquid light guide that allows the power supply assembly containing the lamp to be stored safely away from the microscope, preventing any heat transfer to the stand or stage. Also, using a liquid light guide takes away the laborious task of centering the lamp after each replacement and provides a uniform light beam that will fill the back focal plane of each objective. Disadvantages of liquid light guides are the high susceptibility to mechanical strains that renders them inoperable seconds after stress (bending) and to the formation of air bubbles inside the guide after prolonged use that results in non-homogenous illumination. Taping the guide on the wall to prevent any accidental handling and making sure that the liquid guide does not form any sharp angles (especially behind the power source) will prolong its use. Every 2000 h or so (each time the lamp is replaced), unplugging the guide from the fluorescent illuminator and projecting it on the wall is a good practice to check the integrity of the light guide. If the resulting circle of light (very bright white) is not uniformly illuminated, it may be time to purchase a replacement. Meanwhile, performing the flatfiled correction again for each objective is desirable (see below). Never point the liquid light guide toward a person when the lamp is on and never point it toward reflective surfaces, as it is impossible to predict where the light will end up. Always prevent people from entering the room while checking the integrity of the liquid light guide by displaying a clear sign on the door.

LED (light emitting diode) technology was adopted by the microscopy industry only relatively recently, as in the past the low power output of these light sources made them best suited mostly for machine vision and the auto industry. Today, it is estimated that due to advances in LED technology the brightness of these devices doubles every couple of years, which makes LEDs the most promising illumination technology for fluorescence microscopy. The output wavelengths of LEDs available today cover most if not all of the fluorochromes that are currently in use. While the full width at half maximum (see above) is not as narrow as the laser light, ultimately this is an advantage because the use of certain fluorochromes is no longer prevented by the mismatch between the emission of the light source and the excitation of the fluorochrome. There are several advantages of the LED that make this technology superior to current illumination methods: LEDs are small, compact and have low power usage with little heat generated during “on” times. LEDs also require no warm-up time, last for 10,000 h or more with outstanding

Introduction to Fluorescence Microscopy

109

stability (experiencing less than 1% decrease in the power output over their lifetime). Another notable advantage over traditional fluorescence sources is their ability to be turned on and off within milliseconds, circumventing the need for mechanical shutters, a known source of vibration that becomes a serious imaging problem especially during fast time-lapse experiments. The downside of LED technology is the low power output that is equivalent (and that is true only for certain wavelengths) to a 75 W xenon lamp. However, it is just a matter of time before the LED will become bright enough to compete and likely replace most standard mercury and metal halide fluorescence sources. 5.1.1. Operating Rules of Fluorescence Source Power Supplies

When turning on a fluorescence light source, there is a short (less than a couple of seconds) discharge of about 50,000 V that will create a powerful magnetic and electric field that potentially could destroy electronic components found nearby (computer-related equipment). Even though all of the cables and the power sources are (theoretically) shielded, the possibility remains. Therefore, in a laboratory with multiple users, it is always useful to post a warning regarding the sequence in which components should be tuned on and tag each piece of equipment next to its power switch with a label that specify the order and the total number of power switches that have to be on for all the equipment to work. This is necessary because older pieces of equipment have the on/off switches hidden on the side or on the backside and are easy to miss. Therefore labels should read: 1/7, 2/7 etc. In any case, as a rule, the first component that has to be turned on and the last to be turned off is the fluorescence light source.

5.2. Fluorescence Illuminator

Fluorescence illuminator or epi-illuminator classically consists of the lamp housing located at the opposite part of the nosepiece, the collector lens assembly and the intermediate part that contains slots for balancer, neutral density filters, field diaphragm and the aperture diaphragm (see Fig. 7.6). The collector lens assembly is located between the light source and the balancer. The lens system of the assembly (which is located near the light source) can be adjusted i.e. focused, to collect (hence the name of the lens system) most of the light emitted by the source and send it to the filter cube turret. A second lens system (or several on newer microscopes) focuses the light originating from the collector lens into and fills the back focal plane of the condenser (represented by the objective in epi-fluorescence microscopy). High optical quality of the collector lens assembly is a must especially as new devices that provide confocal-like images by creating virtual pinholes such as DSU (disk scanning unit, Olympus) or Zeiss (Apotome) are used with increasing frequency. These devices create images that are formed by using slits that focus light with different wavelengths on the sample. Therefore,

110

Ghiran

Fig. 7.6. Picture of an epi-fluorescence illuminator with its main parts labeled: CL, collector lens assembly; ND, neutral density filters; AD, aperture diaphragm – or sometimes called aperture sto-p; FD, field diaphragm; FC, filter cubes. Before any imaging session, one has to check the status of all these accessories.

apochromatic corrected illuminators are essential. Each time the lamp is replaced it must be centered and adjusted to meet the criteria for Koehler illumination. Most new microscopes have in the lamp house either a concave or parabolic mirror. Therefore, the image of the lamp will be double when imaged without an objective in the optical path. It is important to know that the purpose of centering the lamp is not to overlap the two images of the lamp (one direct and the other one mirrored) but rather to align them beside each other. Once this is accomplished, the image must be slightly defocused by adjusting the collector lamp to even out the illumination of the field of view. Although this step may seem unimportant, it is crucial when using fluorescence images for quantitative purposes. When samples are imaged with low-to-intermediate magnification objectives (10× to 60×) and recorded using medium-to-large format CCD cameras (1.2– 3/4′′ ), an uneven illumination will cause unequal excitation of fluorochromes in the sample, rendering different readouts from virtually identical fluorescent structures. To make the situation worse, because of the nature of fluorescence imaging where most of the field of view is black, even significant variations in illumination are not readily identifiable. Therefore, flat-field correction must be performed before any quantitative analysis for each objective separately and, of course, after each bulb replacement. This operation will correct not only for uneven illumination due to the light source but also for dust particles located on optical surfaces between the objective and the CCD chip. Therefore, in theory,

Introduction to Fluorescence Microscopy

111

each filter cube should be flat-field corrected individually. Even on a homogenous illuminated field, the fluorescence originating from the sample will be absorbed by the dust particles located on the optical path, and the resulting fluorescence will be lower. If one quantifies changes in fluorescence of moving cells for a period of time, an artifactual, biologically irrelevant, increase or decrease in fluorescence could be recorded when cells start in or cross such areas. Only images that are corrected in such a way are suitable for any kind of image analysis when quantitative information is required. If parameters such as length, width, area etc., are of interest, flat-field correction may not be critical, but in any case one should always be aware of it. If flat-field correction cannot be done for various reasons and quantitative analysis must be performed, then comparing images that were acquired using the very same area of the field of view could bypass the need for flat-field correction. This can be done by using an eyepiece reticule and acquire images that were framed in the same quadrant of the reticule. Nonetheless, results generated by this method are an approximation at best and should not be used routinely. To perform flat-field correction one needs to acquire three images: (1) the image of interest, (2) a dark frame and (3) a flat-field image. The first image (I ) will contain structures of interest plus any artifacts induced by imperfect illumination, dust and electronic noise. Importantly, the exposure time should be recorded and used for the flat-field image and the dark frame. The second image, the dark frame (Df ), has to be acquired with no light reaching the detector (Fig. 7.7a) and is needed to subtract the contribution of the electronic noise from both the flatfield image and from the image of interest. The third image, the flat-field (Ff ) (Fig. 7.7b) can be the image of a concentrated solution of a fluorescent dye (fluorescein) that lacks any structures. It may help to slightly defocus the image to obtain a homogenous field although it would be ideal to acquire all images at

Fig. 7.7. Surface representation of the dark frame and uncorrected uniformly fluorescent surface. a The noise levels are less than 5% of the maximum value. b Surface representation of a homogenous fluorescein solution showing significant variation in illumination throughout the recorded field.

112

Ghiran

the same focus. It is also recommended that the exposure time of the flat-field image be long enough so that at least half of the maximum value of the pixel is reached (around 2400 for a 12-bit CCD). Because the exposure time is dictated by the image of interest, using neutral density filters and changing the concentration of the fluorescein may be needed. Also, acquiring several images (10–15) that will be averaged in the end are needed to limit the influence of noise in the flat-field image. Although the image will look rather uniform, any image analysis program will show significant differences in intensity throughout the image due to imperfect illumination and dust on the optical components. Pseudo-coloring the image or displaying it in 3D-mode based on the pixel value would make the unevenness of the field very apparent (Fig. 7.7b). The exposure time should match that of the image of interest as the electronic noise increases with exposure time. Electronic noise is present in every image and has more or less a constant value, which is about 5% (a pixel value of 180– 220) of the maximum pixel value (4095 for a 12-bit camera). Scientific-grade CCD chips found in virtually all cameras suitable for fluorescence microscopy will display, when imaging a perfectly uniform bright surface, a range of values that are within 5% of the mean pixel intensity. Consequently, when acquiring a background image (no illumination), the mean, lowest and highest pixel value, will be almost the same and around 5% of the maximum pixel value (see the “Y” axis in Fig. 7.7b).3 Therefore, when acquiring fluorescence images, one should aim for obtaining a pixel value of structures of interest between 3600 and 3900 to increase the signal-to-noise ratio. Saturated pixels, having a value of 4095, are useless when the image is needed for quantitative analysis because the actual value of the pixel, whether it is 4095 or any other value above that, would still be displayed as 4095, making further analysis inaccurate. Once all three images are available, most if not all of the image analysis programs have built-in capabilities that will use the following formula (or a variation of it) to generate the flat-field corrected image, Ic(x,y) : Ic(x,y) = M (I(x,y) − Df(x,y) )/Ff(x,y) − Df(x,y)

3

This is true for a certain exposure range that depends directly on the temperature of the chip. The lower the temperatures of the CCD chip, the lower the noise levels during longer exposures times. In general, cameras used for astronomical imaging require deep cooling because it is not uncommon to expose an object for hours at the time. Luckily, biological applications that require exposure time longer than several minutes are rather rare. CCD cameras that are cooled at −40 to −65◦ C are needed for applications where the fluorescence associated with biological structures or phenomena are very dim and short-lived. In these instances, it is crucial to lower the noise levels as much as possible so that faint fluorescence signal would become apparent.

Introduction to Fluorescence Microscopy

113

This operation will subtract the value of each pixel of the dark frame (Df(x,y) ) from the corresponding pixel in the actual image (I(x,y) ) and flat-field image (Ff(x,y) ) and divide the results. (M) represents the mean value of the corrected flat-field frame. Because the value of pixels from the original image will be changed irreversibly, this operation should be performed on a duplicate of the original image. The illuminator also houses the fluorescence shutter, as well as several important and often overlooked devices, such as the heat and neutral density filters, the excitation balancer, the aperture and the field diaphragms. Improper settings of any of these accessories could result in low-resolution, low-contrast images that are independent of the sample and the optical components used to generate images. The heat filter is located between the collector lens and the optical component of the epi-illuminator of all microscopes (Fig. 7.6) and is used to block wavelengths above 600 nm that could potentially overheat and shorten the lifespan of the coated optical elements. If near-IR or IR imaging is used, it may be useful to remove the filter to increase the excitation efficiency of IR fluorescent probes. Also, most cameras used for microscopy have a IR cut-off filter in front of the CCD/CMOS that could also decrease the efficiency of near-IR or IR imaging. One should always check with the microscope/camera manufacturer before attempting to remove any built-in optical or mechanical components from the microscope or camera head. It is important to remember that for 10–15 s or longer after the bright-field source (likely a 100 W halogen lamp) is turned off, the bulb will continue to transmit infrared radiation (heat) that could potentially affect the CCD and fog the resulting image. If a brightfield method is used to locate the cell or tissue of interest and the fluorescence imaging requires long exposure times, the IRinduced fogging could become a problem. There are several ways to avoid image fogging: waiting for 20–30 s once the halogen lamp is turned off to image in fluorescence mode, using a heat filter or using a mechanical shutter in the bright-field path. A shutter may be a better choice because all incandescent sources turn-on and die-off slowly. Without a shutter during time-lapse acquisition of alternating fluorescence and bright-field images, such as phase contrast or DIC (see below for analyzer-induced artifacts), there will be a bleed-through of the bright-field light into the next fluorescence image, especially during fast acquisition. Having a mechanical shutter will completely circumvent the problem, although it may transmit significant vibrations in the microscope stand that would blur the next fluorescence image. Most software dedicated to image acquisition have the option to delay the step that follows the closing or opening of a mechanical shutter (either on fluorescence or bright-field path) or change in X−Y or

114

Ghiran

Z-axis of the microscope stage, to prevent blurring of the image due to mechanical vibrations. This may not be a viable option if fast acquisition is required. Having liquid light guides for both light sources (bright-field and fluorescence) completely circumvents this problem, as the mechanical shutters are located away from the microscope stand. The excitation balancer (EB) slider (Fig. 7.6) is usually found in the fluorescence illuminator between the collector lens assembly and the neutral density filters and is useful only when used with a double or triple filter cube in the turret. It functions as a variable excitation filter, shifting the excitation wavelengths by either changing the angle of a dichromatic mirror from 0 to 45◦ (system employed by Olympus) or changing between a long and a short-pass excitation filter (Nikon). The balancer is very useful when an image with two or more fluorochromes is analyzed and certain features are stained very brightly with one fluorochrome whereas others stained with another fluorochrome are very dim. By gradually decreasing the excitation of the bright fluorochrome and increasing the brightness of the dim fluorochrome, the topological relationship between the two structures can easily be recognized without the need to repeat the experiment to correct for the unbalanced signal strength. Also, colocalization (if any) between two colors is easier to record because there is no “pixel shift” artifact, a common occurrence when using two or more single band cubes (see below). The disadvantage is that in order to record images rendered by a balancer, a color camera is needed and most of the scientific-grade cameras use monochrome CCD chips. Using a color filter in front of the camera can easily fix this problem. The downside of having the balancer installed on a microscope that is shared among several users is that more often than not, after they are done using it, most users leave the balancer engaged in the optical path. So when the next user images a sample using a single band filter cube, a significantly dimmer image than normal (or sometime no image at all) is obtained for reasons that have nothing to do with the sample itself. Again, a good practice is to always check the position of all accessories that could be in the optical path before any imaging session. Neutral density (ND) interference filters are used to decrease the intensity (amplitude) of the excitation light (Fig. 7.6) by reducing the intensity uniformly across the spectrum (from 400 to 700 nm). In fluorescence microscopy, filters are used mainly to protect the sample from excessive bleaching and, when imaging live samples, to diminish the generation of reactive oxygen species that depend directly on the amount of UV illumination. Most modern fluorescence microscopes have either two neutral filters in the optical path or a wheel with several ND filters of increasing absorbency from 20 to 90%. Depending on the mechanism employed by the filter to attenuate the incoming light, there are

Introduction to Fluorescence Microscopy

115

two types of ND filters: absorbing ND filters that have a smoked appearance or reflecting ND filters that look like a mirror on the reflecting side (which must face the light source). The aperture diaphragm (Fig. 7.6, AD) can usually be centered on the X−Y axis and is located in a conjugated plane with the rear aperture of the objective and, therefore, controls the intensity of excitation without affecting the size of the area that is illuminated. Closing the aperture diaphragm will reduce the glare of bright fluorescent images, although in our experience, it is less useful than the iris diaphragms that are fitted on some high NA, immersion objectives. A semi-closed or closed aperture diaphragm will render dim images, and, because it does not affect the size of the field of view but only the amount of fluorescence light that illuminates the sample, the problem is not obvious from the beginning. The status of this diaphragm should always be checked before any imaging session. The field diaphragm (Fig. 7.6, FD) is also centerable and is located between the aperture diaphragm and the filter cubes and can be either round or rectangular. It provides very useful means to control the size of the illuminated area to prevent unnecessary bleaching of the sample and to increase the contrast in samples with high background. For maximum contrast, the illuminated field should be as large as the structured imaged even if it means closing the diaphragm down all the way. However, this can also be a means to block-out the excitation of structures deemed “non-representative”; therefore, it should be used appropriately. Filter cubes (Fig. 7.6, FC) are located above the objective in either a manual or motorized turret (newer microscopes). Older microscopes used separate sliders for excitation and emission filters. Very few, if any, such microscopes are found today in research laboratories and, therefore, we will not discuss them any further. A fluorescence filter cube has three main components: an excitation filter, a dichromatic filter (dichroic mirror or beam splitter) and an emission filter (Fig. 7.8). Please note that arrows on the emission and excitation filters point in different directions depending on the manufacturer of filters. There are three main types of interference filters (Fig. 7.9) that are classified according to the range of wavelengths they allow to pass. The blocking range is usually not mentioned in the name of a filter. The three types are short pass (SP) filters that allow light with short wavelengths to pass while blocking everything else, long pass (LP) filters that allow light with long wavelengths to pass while blocking short wavelengths and band-pass (BP) filters. Band-pass filters allow only a range of wavelengths from the spectrum (a band or a window) for which the filter is perfectly transparent. Figure 7.9 shows the measured transmission profiles of the three main filters that are characterized by center wavelength (CWL, band filters), peak (P %T) and average

116

Ghiran

Fig. 7.8. The basic configuration of a fluorescent filter. Parts are shown before assembly for a better representation of the relationships between various components. Note the angled emission filter. The orientation of filters as shown by arrows on the emission and excitation filter are not universal. See text for details.

Fig. 7.9. Representation of the three simple filters used for fluorescence microscopy and their defining parameters. Long pass (LP) and short pass (SP) filters are defined by their cut-on and cut-off wavelengths. The bandpass filters (Bandpass) are defined by their peak transmission, center wavelengths (CWL) and full-width half maximum (FWHM) (Image courtesy of Chroma Inc.).

transmission (Avg. %T), or for bandpass filters, the full-width half maximum (FWHM), which is a measure of the filter slope (the steeper the better). Combinations of these three filter types are found in virtually all filter cubes. Figure 7.10 shows the diagrams of a high-quality simple-band DAPI filter. The novel methods employed to manufacture filters have reached almost the lowest limit for auto-fluorescence of the glass and the highest limit for

Introduction to Fluorescence Microscopy

117

Fig. 7.10. Diagram showing a high-quality single-band DAPI filter. Note the almost rectangular bandpass of the excitation and emission filters that have the 50% cut-on and cut-off values almost identical with the blocking range. The dichromatic mirror has an almost vertical slope and close to 0% transmittance for the excitation range and over 90% transmittance for the emission range (Diagram courtesy of Semrock, Inc.).

transmittance, barrier and slope of filters. Note the almost perfect wavelengths selection of the interference filters that ensures bright signals and black background. Excitation filters: In the past, selecting the desired wavelengths was realized by using colored solutions of specific dyes that were placed in glass containers in the light path. Colored gelatin sheets and colored glass filters later replaced containers. These approaches for selecting the excitation wavelengths worked well with trans-fluorescence methods where samples were illuminated through an immersion dark field condenser, which due to its specific design prevented most of the high-energy excitation light to enter the objective and allowed only the emitted fluorescence to form the image. While these filters were very affordable, they suffered from poor optical performance, low transmittance and mediocre wavelength discrimination. A significant problem was the high level of auto-fluorescence of dyes that were used in these filters, which considerably decreased the signal-to-noise ratio of most samples. In recent years, the majority of filters used in fluorescence microscopy are interference filters. The main technologies used today for constructing an interference filter use either soft or hard coating. Soft coating is based on deposition on several optically flat pieces of glass glued together by adhesive of multiple thin films of several different dielectric substances (metal

118

Ghiran

salts). The alternative, hard coating method, deposits dielectric layers on a glass disc using a special technology (ion-beam sputtering) that creates several layers on just one optically flat piece of glass avoiding the use of any adhesives, which are responsible for most of the auto-fluorescence of the soft-coated filters. Aside from being more resistant to light damage and moisture, the later filters are easier to manufacture with perfectly flat surfaces, a paramount characteristic when co-localization studies between two, three or more colors are performed. 5.2.1. Registration Artifacts When Using Two or More Filters

In Fig. 7.11a is shown the overlap image of a triple-labeled 10 µm-bead imaged by two microscopes, one using four standard DAPI, FITC, TRITC and triple-band filters and the other using filters of the same type but of “no image-shift” quality. In the first image, all images are shifted in respect with the optical axis creating the false impression of no or partial co-localization, although all images originate from the same structure. The “no image-shift” grade filters do not suffer from such an effect (Fig. 7.11c). To better represent the lack of co-localization, images in the left were processed using the Sorbell filter (ImageProPlus, Media Cybernetics) to show only the outline of the imaged bead (Fig. 7.11b). The origin of the problem can be the “wedge” shape of the emission filter but also the “wedge” shape of the dichromatic filter (dichroic mirror) that would create the same effect regardless of the quality of the emission filter. In some situations both filters suffer from the same defect. Therefore, it is not indicated to “mix and match” dichromatic and emission filters of different qualities if co-localization studies are needed. If standard filters are the only ones available, co-localization studies still can be accurately performed if the amount of shifting for two channels is calculated, assuming the third one to be perfectly aligned with the optical axis of the microscope and a back-shift

Fig. 7.11. Filter cube-induced artifact during co-localization. A triple-labeled 10 µm latex bead was imaged either through regular (a) or co-localization-quality (c) filter cubes. The white image of the bead (a, left) was recorded using a triple filter. The raw (a) and the processed image (b) show significant misalignment among filters, although the same structure contained all fluorochromes and there was no shift of the bead during acquisition. Co-localization-quality (c) filter cubes rendered a perfect overlapping image. The difference in bead size between the two images is due to the different pixel size of cameras used for the acquisition.

Introduction to Fluorescence Microscopy

119

correction is applied to the first two channels. All major image acquisition and processing pieces of software, as well as several freeware programs, have built-in modules for this option. Having triple-labeled beads on the slide along with cells of interest during acquisition can considerably simplify the task. The smaller the beads, the more precise the correction. If the amount of shifting between channels changes overtime, it could indicate a loose component (likely the dichroic mirror) in one of the filters. Types of interference filters: Depending on which range of wavelengths a filter allows to pass, there can be band-pass filters that allow only certain range of wavelengths to pass and can be either narrow or broad band-pass, short pass edge filters that allow only short wavelengths to pass while blocking long wavelengths or long pass edge filters that block short wavelengths and only allow long wavelengths to pass (see Fig. 7.8). Understanding the properties of a filter and the way it does or does not match the excitation and emission spectra of a certain fluorochrome is essential when buying a filter or, more importantly, when assembling a new filter cube from components available in the laboratory. Most filters must be mounted in a certain way and the direction is usually shown with an arrow that has to point to the dichromatic mirror. Also, it is important to know that most filters (so called soft-coated) are considered consumables with a limited lifespan that, depending on how much the microscope is used, have to be replaced every 1–2 years, as the transmittance will decrease below 20–30% (so called “burnout” filters). Figure 7.12a shows the picture of a soft-coated filter 9 months after installation followed by average to high usage. Figure 7.12b shows in detail the moisture damage centered on impurities found between the thin layers of coating. Using the microscope in areas with high humidity also favors the growth of fungus on all optical surfaces

Fig. 7.12. Damage of a soft-coated emission filter after less than a year of moderate to heavy use. a Image on the left shows the damage that started at the periphery (discolored areas) and then progressively moved toward the center. The first indication of the damage was the out-of-focus round shapes found mostly at the edges of the field of view noticed first during bright-field imaging. b Image on the right shows a low magnification detail of the affected area (see text for details).

120

Ghiran

(including filters) that drastically alters their optical properties. A visual inspection would show structures that look like hairs or vines branching from a center if there is fungal growth present or round darker areas if there is any delamination due to heat or impurities in the filter coat. All these problems could eventually lead to photo-darkening. Gradual decreases in light efficiency while imaging in fluorescence and the appearance of out-of-focus patterns when the filter is used in bright-field are signs that the filter has to be changed. It may be wise investing in hard-coated filters that, although are more expensive than the regular filters, do not “burn-out.” However, they are not impervious to moisturerelated damage (fungal growth). Most microscope manufactures use an antifungal treatment of all optical components (lenses and prisms) to slow down the growth of fungus that ultimately would etch the glass surface underneath rendering it useless. Dichromatic beam-splitters (dichroic mirrors) are special long or band-pass filters that are mounted at 45◦ degrees in the filter cube. Any variation of this angle will significantly change the intended optical properties of the filter. This property is used to change the range of wavelengths that will pass through the filter (see above, excitation balancers). Once the excitation filter has selected a certain range of wavelengths, the dichromatic filter performs like a mirror for the incoming shorter wavelengths, characterized by a percent reflection (close to 100% for certain wavelength range), reflecting the light through the objective into the sample. The fluorescence that originates from the sample has a longer wavelength than the excitation light (Stokes shift, see above) and, therefore, the dichromatic mirror allows the light to pass un-reflected (therefore characterized by percent transmission close to 100% for a certain wavelength range). Depending on the type of the filter cube that it is used, some dichromatic filters can be long pass filters (single excitation filter cubes) or band-pass filters (double and triple filter cubes). If the dichromatic mirror surfaces are angled (edge-shaped) instead of being parallel, the image formed by that particular filter cube will be shifted from the optical axis of the instrument. The same is true for the emission filters (see above). Figure 7.13 shows the diagram of a triple-band filters that allows simultaneous and specific excitation and emission of several (up to three) fluorochromes. Note that in a triple-band filter the dichromatic mirror and both the emission and excitation filters are band-pass type filters. Emission filters can be long pass (allows light with longer wavelength to pass), increasing the overall brightness of the signal but also increasing the background or band-pass type, in which case the brightness decreases slightly but the background is darker. They are easy to identify in a filter cube because they are mounted under a 5◦ angle to decrease the brightness of the background and prevent back-reflection.

Introduction to Fluorescence Microscopy

121

Fig. 7.13. Diagram of a high-quality triple-band-pass filter. The filter is designed to allow simultaneous imaging of up to three fluorochromes. Note that all the filters found in this cube are band-pass filters (Diagram courtesy of Semrock, Inc.).

5.2.2. Orientation of the Fluorescence Filters in a Filter Cube

The direction a filter has to be mounted in a filter cube is usually indicated with an arrow, but different companies use the arrow to point to different parts of the filter cube. In most cases, the arrow on the excitation filter points toward the dichromatic mirror (center of the cube), whereas the arrow on the emission filter points toward the dichromatic filter (Chroma) or towards eyepieces (Semrock). The situation is more complicated for the dichromatic filter, which has a coated surface that always has to face the light source for optimal performance. Because the dichromatic mirror is very thin, there are no arrows to show the proper orientation of the mirror. One can find the coated surface if one uses a light source (a desk lamp works well) to reflect it off from the dichromatic mirror; if the coated surface is pointing up, there will be only one image of the light source; if the coated surface points down, the image of the light source will be doubled and slightly shifted. If the performance of a filter set is less than expected, especially if the filter was custom built from preexisting parts in the laboratory, it is always a good idea to check the orientation of filters and consult the manufacturer.

5.2.3. Methods to Prevent Stray Light from Exiting the Filter Cube

Because there are no perfect filters, a certain percentage of the light that traverses through the excitation filter, instead of being redirected by the dichromatic mirror into the objective, passes through it and bounces off the back wall of the cube and then is reflected by the dichromatic mirror though the emission filter into the eyepiece or detector, increasing the brightness of the background and decreasing the dynamic range of the actual

122

Ghiran

image. Various methods have been devised over time to prevent this from happening, starting with coating the interior surface of the filter cube with a black, non-reflective paint (black baffled) or mounting the emission filter under an angle of 5◦ . In addition, increasing the efficiency of the filter also prevents reflection of the back-reflected light by the telan (tube) lens. Some microscope manufactures, in addition to an angled emission filter, use a slightly modified filter cube that contains either an absorbent neutral density filter on the opposite side of the excitation filter or an opening (Fig. 7.14) that allows stray photons to completely escape the filter cube (“Light-trap” first employed by Zeiss and Background terminator, employed by Nikon). The next component in the light path is the objective, which in all epi-illumination techniques plays the role of the condenser.

Fig. 7.14. Comparison between a standard fluorescence filter cube and light-trap (Zeiss). a In a standard fluorescence filter cube, the light that passes through the dichromatic filter (thin green arrows) is reflected between the back wall of the filter cube and the upper surface of the dichromatic filter, and crosses the emission filter increasing the background level of the final fluorescence image. b The light-trap filter cube (1) has the back wall removed so after passing through the dichromatic mirror (2), the stray light (3) is directed by the tapered wall of the fluorescence illuminator (4) outside the cube (5) thus improving significantly the signal-to-noise ratio (Diagram courtesy of Carl Zeiss MicroImaging, Inc.).

5.3. Objectives

Maybe the most important and specialized optical component of a microscope is the objective, which points to and forms the image of the examined object. In epi-fluorescence microscopy, the objective is also the condenser in the optical path, focusing the excitation light into the specimen. There is no other microscope component with a greater role in the image formation than the objective. In the diagram shown in Fig. 7.15, a cross-section through a microscope objective illustrates the mechanical and optical complexity of a fully corrected plan-apochromatic objective that can contain between 13 and 15 lenses. Therefore, the quality (or the lack thereof) of the objective will be crucial in determining the amount of information in the final image. That is not to say that a low-quality filter cube or an imaging device that is unsuitable for a certain application (due

Introduction to Fluorescence Microscopy

123

Fig. 7.15. Cross-section through a high-magnification planapochromatic objective. The image shows the mechanical and optical complexity of high-quality objectives that can contain up to 15 carefully polished and aligned lenses. Any mechanical stress (due to hitting or dropping of the objective) could potentially shift lenses from the optical axis of the objective significantly decreasing its optical qualities (Image courtesy of Leica Microsystems).

to speed, sensitivity, dynamic range or resolution) will not have a negative impact on the final image. However, irrespective of the quality of the other components involved in the image formation or acquisition, if the objective does not have the required NA or the appropriate chromatic and spherical aberration corrections for a particular application, no amount of correction or postacquisition processing will add the information that was missed by the objective. Therefore, for demanding specialized applications (calcium imaging etc.), building a new system starts with the objective followed by the camera, filters etc. If the resolution required for an application is known, for instance 0.25 µm, it is always a good idea to buy only that particular objective of the highest quality (such as PlanApo 60 × 1.40, 1.42 or 1.49) and the rest of the objectives (4×, 10×, 20× and maybe 100×) of standard quality (planacromats or planfluorite, see below). The premium one payments for fully corrected lenses are in the range of thousands of dollars and the money could be better used towards other components (camera, filters, image acquisition and

124

Ghiran

analysis software). If for most of the bright-field applications the type of objective (correction-wise) is not critical, for fluorescence microscopy, and especially for application such as co-localization of two or more colors in three dimensions (3D), the choice of objectives is paramount, due to the impact of the chromatic and spherical aberration on image formation (Fig. 7.16).

Fig. 7.16. Diagram showing the two main aberrations of microscope objectives. a Spherical aberrations are due to fact that light passing through the edge of a lens is refracted closer to the lens than the light going through the center. Therefore, a simple lens will not have a single focus but rather the focus will be present on a certain distance (area) along the optical axis of the lens between the two extremes formed by the periphery and central rays. The place along this area where focus is the “best” is called circle of least confusion and it was first described for astronomical refractors and photographic cameras. Its size depends on the focal distance of the lens and its diameter. b Chromatic aberrations are a direct consequence of the variability of the refractive index of a transparent material with the wavelength. Therefore, blue light will be refracted more (closer to the lens) and red light less (farther from the lens) inducing a halo around the imaged objects that changes color with focus.

There are several major optical aberrations that affect, in various degrees, all microscope objectives due to the geometry of lens and the wide variation of the refractive index of a particular media (glass or water) with wavelength and temperature: spherical aberration, chromatic aberration, astigmatism, field curvature and coma. We will address here only the first two as they are directly relevant for fluorescence imaging. Spherical aberrations are due to the fact that the thickness of a lens varies significantly between the center and the periphery. As a consequence, the light is focused in different points along the optical axis depending on the place where the incident light entered the lens. The light that is closer to the optical axis will pass through the lens with minimal refraction focusing farther from the lens, whereas the light entering the lens near the periphery will undergo a more significant change in a direction focusing closer to the lens (Fig. 7.16). The larger the frontal lens (aperture), the worst the aberrations. Images that result from an insufficiently spherically corrected objective are soft, lacking contrast and sharpness. It is important to remember that corrections that are built into any objective are calculated for specific conditions:

Introduction to Fluorescence Microscopy

125

certain refractive index and temperature of the media between the lens and the coverslip, specific distance between the top surface of the coverslip and the imaged objects, and the presence or absence of the coverslip. Any departures from such conditions would significantly degrade the final image. Efforts to decrease spherical aberration started, surprisingly, over 300 million years ago when trilobites developed as the first animals that had crystalline lens made out of two lenses with different refractive indexes. New microscope objectives have lenses manufactured out of special types of glasses that are polished at precisely calculated angles∗ that significantly minimize these aberrations. Chromatic aberration is particularly important in fluorescence microscopy and, therefore, we will discuss it in more detail. This aberration is not due to the geometry of the lens but rather is a direct result of light interacting with glass (or any transparent media). Depending on the wavelength of light, the refractive index of the glass changes, increasing as the energy of the light increases and the wavelength decreases. This means that if white light passes through a lens, its blue component will be refracted closer to the lens, and the red component further from the lens. Green light will be focused in between (Fig. 7.16). On an X−Y plane, this results in images of objects that have colored rings around them. On the X−Z and Y−Z planes this means that in order to focus properly using the three colors one has to change the height of the stage to accommodate the three focal points from the optical axis. The danger of using an insufficiently chromatically corrected objective for 3D co-localization studies becomes obvious. A Z-stack of the same very object recorded in blue, green and red channels will be displayed at three different heights when a 3D rendering of the stack is performed. This would suggest that depending on the size of the object, there are either adjacent or even separated structures when, in fact, there is just one structure. The further away the fluorochromes used for labeling (blue, red or far red), the more likely the objective will generate an artifactual image. Based on the degree of chromatic correction, there are three main types of objectives: achromat, fluorite and apochromatic. Achromatic objectives are usually found in low-end microscopes and are corrected for two wavelengths, blue (490 nm) and red (660 nm), which are brought in a single focus. From a spherical aberration point of view, acromats are corrected for green light. These lenses are not particularly suitable for fluorescence microscopy and should be avoided whenever possible. Fluorite objectives were designed to correct for two or three wavelengths by using a glass containing a special mineral, fluorite (hence the name of the lens). Because the mineral is very transparent for blue and UV light, fluorite objectives were used primarily for fluorescence microscopy. In addition to increased UV transparency, the

126

Ghiran

optical properties of the fluorite lens allows for correction of an objective for spherical aberrations for up to three wavelengths. This translates to objectives with a higher NA for the same magnification compared to acromats that significantly increases their resolution and brightness. While acromats may not have a specific symbol on the barrel, fluorite objectives are usually labeled as FL, Fluo or Fluorite. The third type of lens, the apochromatic objectives, first built by Zeiss following the specification of Ernst Abbe in 1886, offers the highest level of correction for both spherical and chromatic aberrations. However, all these corrections come with a very high price tag. Apochromatic lenses are corrected for four colors (some super-apochromatic objectives even for five wavelengths) for both chromatic and spherical aberration. These lenses have the highest NA and usually provide the brightest image. Due to their NA, apochromats are ideal for fluorescence work, although for some specialized methods, such as calcium imaging with ratiometric dyes excited in the UV range, major microscope manufacturers sell particular fluorite lenses that surpass apochromats. The correction type for apochromatic lenses is written always on the barrel as Apo, or Apochromatic. On the objective barrel, in addition to the extent of spherical and chromatic corrections, there is always additional information such as magnification (both as number and color code), numerical aperture (as a number or, in the case of objectives with a built-in iris diaphragm, as a range or numbers), the type of tube length either finite (160 mm), “160” (Fig. 7.17, right) or infinity tube length microscope systems “∞” (Fig. 7.17, left) application suitability (such as, phase

Fig. 7.17. Comparison between a new (infinity corrected) and old-style (finite corrected) microscope objectives. The type of geometrical (plan) and chromatic (SApo or Apo) corrections are clearly indicated on the barrels. Also, the correction collar range, tube length and the eyepiece field of view (26.5, left) are shown.

Introduction to Fluorescence Microscopy

127

contrast, differential interference contrast, polarization, dark filed), requirement for coverslip (“–” for indifferent, “0” for no coverslip and “0.17” for regular coverslip (170 µm thick), and correction range for coverslips’ thicknesses common for high NA dry objectives. On newer objectives, additional information is added by some manufacturers but not others, such as the maximum field of view of the eyepiece expressed in mm (FN = 26.5), the type of DIC prism needed for the objective and the location with respect to the back focal plane of the objective, and working distance (WD) of the objective, which is measured from the frontal lens to the top of the coverglass (also expressed in mm). Aside for the color codes for linear magnification of objectives and type of immersion required (see below), which are universal across all microscope manufacturers, each major brand has its own set of codes both letters and colors for extra corrections or suitability for a particular method that makes reading (and understanding) the full potential of an objective a bit of a challenge even for an experienced microscopist. 1/2x 1x, 1.25x, 1.5x 2x, 2.5x 4x, 5x 10x, 16x, 20x 25x, 32x 40x, 50x 60x, 63x 100x, 150x, 250x

no color assigned Black Brown Red Yellow Green Turquoise Light Blue Cobalt Blue White

Correcting spherical and chromatic aberrations for oilimmersion objectives is somewhat easier because standard immersion media has a refractive index close or identical with that of the coverglass (1.515) and the frontal lens of the objective. For dry objectives, the air in between the frontal lens and the coverglass along with a variable amount of mounting media between the bottom of the coverglass and the actual imaged plane in the specimen create additional problems that become significant as the NA increases. The vast majority of dry objectives are designed to image samples covered with a 0.17 mm (170 µm) coverglass with a refractive index of n = 1.515 and with minimal amount of media between the coverglass and the specimen. It is important to know that 170 µm is the distance between the top surface of the coverglass and the sample to be analyzed for which the objective was spherically and chromatically corrected. This value includes all the media that is between the coverglass and the object (6). An objective will only form a perfect image of the sample that is adjacent to the coverslip, at 170 µm of distance from the top surface of the coverglass. If the coverglass is thinner than 170 µm,

128

Ghiran

then the combined thickness of the coverglass and the media that is on top of the sample (which is expected to have a refractive index of n = 1.515) has to be 170 µm. For no other depth will the objective perform as expected. Having structures in the image surrounded by a halo is a certain sign of significant spherical aberration, likely due to excessive mounting media (assuming that the frontal lens is not dirty). This effect is less obvious for low NA objectives, which are less sensitive to variations in coverglass thickness. Figure 7.18 shows two images recorded on the same slide that was mounted with optimal amount of mounting media on the left side of the slide and excess in the right. The nucleus showed on the left was right below the coverglass, whereas the nucleus on the right, although next to the one on the right, had a significant amount of mounting media present between its top part and the coverglass. Note the increased background levels and decreased resolution on the right image. Images were recorded under identical conditions using an immersion objective (60 × 1.42 UPlanApo). As a rule of thumb, all specimens must be mounted (not just placing a coverslip over a dry specimen) in media with a coverslip if an objective with an NA higher than 0.3 is used for examination. A few high magnification objectives (90 and 100×) require specimens, usually blood smears, to be examined dry without coverslip. In this case, mounting the specimen with a coverslip would significantly degrade the image because the objective is designed to form an image without coverslip. The lack of black ring (which signifies oil immersion) on a high-magnification objective and the inscription “0” on the barrel (160/0 for finite or ∞/0 for infinity microscope systems) should help identify such objectives. On the positive side, these objectives are rarely found in research laboratories.

Fig. 7.18. The effect of spherical aberration induced by increased distance between the coverslip and the imaging sample. Epithelial cells stained with DAPI, present either on the coverglass or on the microscope slide, were mounted together (coverglass + slide) and imaged under identical conditions. The nucleus of a cell adherent on the coverglass is shown on the left. The nucleus on the right was fewer than 500 µm away from the nucleus on the left but more than 90 µm below. Note the increase background and reduced resolution on the left nucleus. Scale represents 10 µm.

Introduction to Fluorescence Microscopy

129

In general, for objectives with NAs between 0.30 and 0.7, optical aberrations introduced by coverslips with thicknesses other than 0.17 mm does not degrade the image significantly. For higher NAs, dry objectives (0.7–0.95), minimal variations of the coverslip thickness, increased distance (over 10–20 µm) between coverslip and specimen, and differences in the refractive index of the mounting media and coverglass are major sources of spherical and chromatic aberrations that significantly alter the quality of the final image. To compensate for these aberrations, all high NA, dry objectives or immersion objectives that use media other than oil (such as glycerol or water) have a built-in correction collar that will compensate for a range of coverslip thicknesses that is written on the objective barrel. For most of the objectives that are used in upright microscopes, the range is typically between 0.11–0.23, whereas for most objectives that are dedicated for work with tissue culture dishes on inverted microscopes (phase contrast, DIC Nomarski and fluorescence), the range is wider, between 0 and 2 mm. If such objectives are used to examine blood smear preparations, the correction collar has to be set at 0, which signifies no coverslip. Also, the side with the sample should face the objective. Since the introduction of fluorescent tags that allow in vivo tracking of intracellular proteins for long periods of time, imaging cells in physiological conditions (temperature, CO2 and O2 pressure) has become the norm in many laboratories that are geared towards microscopy (7). These new approaches require not just environmental chambers but also newer optics to accommodate imaging in new conditions. Because the refractive index of a media decreases as the temperature increases, newer immersion objectives also incorporate a correction collar to compensate for the aberration induced by variations in the refractive index of the immersion media when the objective is used at room temperature (Fig. 7.19). When adjusting the correction collar, a group of lenses inside the objective barrel moves closer to the specimen when correcting for a coverslip that is thicker than 0.17 mm coverslip or away when correcting for thinner coverslips. The correction collar is a very valuable tool if used correctly but also can render images (especially fluorescence images) almost useless if used improperly. When imaging fluorescent samples, improper adjustment of the correction collar renders hazy images (due to presence of high levels of spherical aberrations, Fig. 7.20, left), which decreases the dynamic range of the final image and makes focusing difficult. Although it may seem a chore to adjust the correction collar each time one changes the slide (or even on the same slide if there is an uneven distribution of mounting media), it is a worthy endeavor (Fig. 7.20, right). Correction collars are misused (or not used at all) to such a degree that some microscope dealers refuse to quote them to avoid having users complaining about poor quality of images when imaging with these pricey objectives. Also, it is

130

Ghiran

Fig. 7.19. The effect of spherical aberration induced by increased distance between the coverslip and the imaging sample. Epithelial cells stained with DAPI, present either on the cover glass or on the microscope slide, were mounted together (coverglass + slide) and imaged under identical conditions. The nucleus of a cell adherent on the coverglass is shown on the left. The nucleus on the right was fewer than 500 µm away from the nucleus on the left but more than 90 µm below. Note the increase background and reduced resolution on the left nucleus. Scale represents 10 µm. The highest NA objective currently available that does not require special oil or coverglass. Only a decade ago the highest NA available for 60× and 10× objectives was 1.40. Note the correction collar for temperature. Temperature variations of the immersion oil induce small variation in its refractive index that, due to high NA of this objective, would significantly impact the image quality (Courtesy of Davis Warren, Perkin-Elmer).

important to keep in mind that when adjusting the correction collar, the parafocality of the objective will change; hence, when switching back to another objective, one will have to refocus the image, although all objectives are theoretically parafocal. 5.3.1. Adjusting the Correction Collar

1. Set the collar so that the vertical mark points to the 0.17 writing on the objective barrel, which is the thickness, expressed in millimeters, of most coverglasses used for biological applications. 2. Focus on a small high contrast detail on the specimen 3. Rotate the correction collar slightly to the left (0.18, 0.19), because most of the samples are mounted with an excess of media that create a thicker than optimal coverslip plus media stack.

Introduction to Fluorescence Microscopy

131

Fig. 7.20. Comparison between images of the same sample recorded with two different correction collars settings. The image on the left was taken with the correction collar set at 0.17, whereas the image on the right was recorded with the collar set at 0.20. Both images were recorded with the same objective (SPlanApo 40 × 0.95), using the same exposure time, dark level, gain and aperture diaphragm setting. The only difference between the two images is the correction collar setting. The image on the right appears blurry and out-of-focus with high background (Images courtesy of Sorina Ghiran, BIDMC.).

4. Refocus; if the image has not improved, continue to rotate the collar to the left while refocusing. 5. If the image keeps deteriorating, reverse the direction, rotating the correction collar to the right slightly and refocus. 6. Repeat the operation until the image becomes crisp.

6. Phase Contrast and Fluorescence Since fluorescence images show only the tagged structures in cells or tissues, it is sometimes useful to show the same image both using fluorescence imaging and a contrast enhancement technique, such as phase contrast or DIC Nomarski, that would show the whole cell or tissue, allowing a better orientation in the sample. Using a phase contrast objective for imaging fluorescent samples will significantly decrease the brightness of the image due to the presence of the phase annulus near or in the back focal plane of the objective. If the analyzed sample is fixed, longer exposure will render the desired brightness levels. If the sample is alive, increasing the binning or the gain of the camera will allow shortening the exposure time enough that even if images are recorded successively, there will be minimal or no shift of the imaged structures between the fluorescence and phase contrast image. Of course, this may not be true in all instances. However, if the same phase contrast objective is used to acquire “z”-stacks of images that are used for deconvolution, the “ghost” of the phase ring will induce artifacts in the final restored stack that have no real correspondent

132

Ghiran

Fig. 7.21. Ghost image of the phase ring during fluorescence imaging. The out-offocus image of a 10 µm fluorescent bead acquired using a regular objective suitable for fluorescence work (40 × 1.30 UPlanFl) is shown on the left and the same image recorded with a 40 × 1.00 UPlanApo phase is shown on the right. Note the dark ring that is overlapped with the actual (out-of-focus) image of the bead. The ring represents the “ghost” of the phase ring, which is located on the back focal plane of the objective. The “ghost” will induce an artifact in the final image if a stack of images acquired with a phase contrast objective is used for deconvolution. When the sample is in focus the ring will induce a significant drop in fluorescence and slight decrease in resolution.

in the original stack (Fig. 7.21). Therefore phase contrast objectives are not suitable for acquiring images used for deconvolution and should be avoided if 3D-acquisition followed by deconvolution is needed.

7. DIC and Fluorescence Objectives that are used to generate DIC Nomarski images do not contain any phase rings that would negatively impact the final image, but in order to generate the final 3D-like image, they do require, among other components, the presence of an analyzer (a polarization filter) in the light path that will decrease the intensity of the image about 60% or more. Because the analyzer is located above the filter cube, it will not decrease the light used to excite the fluorochromes in the sample; thus imaging with the analyzer in the path will lead to unnecessary bleaching of the sample. Therefore, it is important to check the status of the analyzer and remove it if it is engaged to prevent damaging the sample. Along with the analyzer, formation of a DIC Nomarski image also requires the presence of a modified Wollaston (Nomarski) prism above the objective. Depending on the type of DIC employed, each objective can have its own small DIC prism mounted right above (or below in the case of inverted microscopes) the objective (Zeiss and Nikon Fig. 7.22a) or there can be a prism located higher above the

Introduction to Fluorescence Microscopy

133

Fig. 7.22. Various locations of DIC prisms on microscopes. a On Nikon and Zeiss microscopes, Wollaston prisms are located above each objective (arrows) either in or close to their back focal plane (Image courtesy of Scott Berger, BIDMC). b Olympus microscopes have only one slot for the Wollaston prisms (arrow), which is shared among several objectives. The height of the prism can be adjusted, depending on the objective used for imaging using a lever located on the DIC slider.

objective in the nosepiece (Olympus and Leica) (Fig. 7.22b). When using bright-field imaging, if the Wollaston prism is left unintentionally in the path, the splitting effect of the prism in the final image will depend on the magnification of the objective (the higher the magnification the worse the effect) and the type of contrast (high contrast prisms have a more significant effect compared to low contrast ones). In the case of fluorescent imaging, the effect is significant and becomes very obvious when using high magnification, high-resolution objectives to image small structures (Fig. 7.23). To record these images, red cell complement receptor 1, (CD35 or CR1) was indirectly labeled with Alexa 594 and imaged either without (left) or with (right) the Wollaston prism engaged using a 60 × 1.42 UPlanApo objective on an Olympus BX 62 microscope fitted with a high-contrast Wollaston prism (Fig. 7.23a). In the left image, receptors were split in two due to the presence of the Wollaston prism in the optical path (false “dimerization” of the receptors). Importantly, the size and

Fig. 7.23. The effect of Wollaston prism during fluorescence imaging. Red cells labeled for CR1 were imaged without (A) or with (B) a high contrast Wollaston prism in the path. Note the doubling effect of the prism on all fluorescent structures present in the sample. The direction of shear is 11-5.

134

Ghiran

the intensity of each pair were lower than the original and the whole image shifted with respect to the optical axis of the microscope. The “tell-tale” signs of this type of artifact are two-fold: the doubling effect is found in all structures present in the image and the splitting of all structures happens in the same direction. To the best of the author’s knowledge, only motorized Leica microscopes automatically remove the Wollaston prism from the optical pathway when the imaging method is changed from DIC to an other method, because the prism is located on a motorized turret instead of a slider. 7.1. Condensers

Regular sub-stage condensers are not needed for imaging fluorescence samples, as the objective plays the role of condenser in epi-fluorescence microscopy (see above). However, even if it does not participate directly in image formation, it can certainly induce artifacts that can range from increasing the background of the image to duplicating bright fluorescence structures present in the field of view. Most motorized microscopes are equipped with swing-out top lens sub-stage condensers that automatically swing out the top lens when the fluorescence shutter is open. And microscopes are designed to do so for good reasons. Figure 7.24 shows the same image acquired with and without the top lens in the optical path (swung out). Aside from increasing the background throughout the image, the top lens also creates an overlapping image that shows two “ghosts” that are not actually present in the sample but represent the slightly shifted mirrored image of the actual cells (with only a fraction of the original intensity); the brighter the objects in the field of view, the brighter the ghosts. Any automatic script (macro) used for post-acquisition image analysis could potentially count the ghosts as actual cells with different fluorescence characteristics. If a microscope is not equipped with a swing out top-lens condenser, simply lowering the condenser will prevent the top lens of the condenser to act as a

Fig. 7.24. The effect of transmitted light condenser on fluorescence imaging. Image on the left was acquired with the top lens of the condenser in place and at the height required by the Kohler illumination for that particular objective when used for bright-field applications. The image on the right was recorded with the top lens swung out. Note the increased background and the apparition of the reflections (arrow) of the actual cells as dimmer counterparts. Both images were recorded using the same exposure time.

Introduction to Fluorescence Microscopy

135

mirror for the incoming excitation light. In addition, most microscopes are sold with a thin non-reflective black metal plate that is meant to be placed under the microscope slide when imaging in fluorescence to prevent the mirroring effect of the condenser top lens.

8. Trinocular Heads In very few instances and settings, the trinocular head can create problems, interestingly only when imaging and not when observing samples directly. For that to happen, several factors have to come together and set the stage for an artifact that it is not at all easy to identify and it is even harder to reproduce. First, the trinocular head has to have three positions for the splitting prism (100% to eyepieces, 20–80% eyepieces–camera and 100% camera) and not two. Secondly, the room has to be lit during image acquisition and thirdly, the lever has to be pulled only half way (to 20– 80% setting) while recording the fluorescent image. If all these conditions are fulfilled then something similar to images showed in Fig. 7.25 can be obtained. The light that enters the eyepiece will contribute to the final image and form a bright haze that will overlap the actual fluorescent image (Fig. 7.25a). If a light bulb is present in direct optical path of the microscope (across from the microscope on a self or even fluorescent ceiling light), then its shape will be overlapped with the fluorescence (or whatever is

Fig. 7.25. The role of ambient light and trinocular settings for fluorescence imaging. a Red cells labeled with a lipophilic cell tracker and for CR1 expression were imaged with the trinocular head lever half way out allowing a 20–80 split of the optical path (20% of the original intensity of the image to the eyepieces and 80% to the CCD camera). This setting allowed ambient light to reach the detector contaminating the final image, showed on the left. b The image was recorded using the same exposure time but with fluorescence off. Note the presence of the light bulb image in the center of the first two images. The image of the bulb is doubled because the light entered through both eyepieces, as the lamp was located on a shelf across from the microscope. c The image on the right was recorded using the same settings as in (a), but with the ambient light off.

136

Ghiran

left of it) image. Turning the light off and using illuminated keyboards and pulling the lever to the very last setting (100% of the light to the camera) will help one avoid this artifact.

Acknowledgments I acknowledge Olympus America Inc., Leica MicroSystems, Carl Zeiss MicroImaging Inc., and Chroma Technology Corp. for allowing the use of several of their images in this chapter. References 1. Coons, A. H. (1951) Fluorescent antibodies as histochemical tools. Fed Proc 10, 558–559. 2. Weller, T. H., Coons, A. H. (1954) Fluorescent antibody studies with agents of varicella and herpes zoster propagated in vitro. Proc Soc Exp Biol Med 86, 789–794. 3. Herman, B. (1999) in (Slavik, J., ed.) 1998, Fluorescence Microscopy and Fluorescent Probes, vol 2. Plenum Press, New York and London, p 292. 4. Murphy, D. B. (2001) Fundamentals of Light Microscopy and Electronic Imaging. WileyLiss, New York, NY.

5. Wolf, D. E. (2007) Fundamentals of fluorescence and fluorescence microscopy. Methods Cell Biol 81, 63–91. 6. Pawley, J. B. (2006) Handbook of Biological Confocal Microscopy. Springer, New York, NY. 7. Cox, G. (2007) Optical Imaging Techniques in Cell Biology. CRC/Taylor & Francis, Boca Raton, FL, USA 8. Goldman, R. D., Spector, D. L. (2005) Live Cell Imaging: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Chapter 8 Using the Fluorescent Styryl Dye FM1-43 to Visualize Synaptic Vesicles Exocytosis and Endocytosis in Motor Nerve Terminals Ernani Amaral, Silvia Guatimosim, and Cristina Guatimosim Abstract The styryl dye FM1-43 is a powerful tool to track exocytosis, endocytosis and recycling of secretory granules or vesicles. Due to its unique structure, dye molecules reversibly partition into the outer leaflet of surface membrane without permeating due to two cationic charges located in their headgroup. When a secretory cell is stimulated to evoke exocytosis, FM1-43 molecules that were inserted in the membrane are internalized during compensatory endocytosis and newly formed secretory granules or vesicles become stained with dye (staining/endocytosis). If stained secretory granules or vesicles undergo exocytosis in dye-free medium, due to concentration gradient, FM1-43 molecules dissociate from the membrane and loose fluorescence (destaining/exocytosis). Using a fluorescence microscope attached to a CCD camera or a confocal, it is possible to follow secretion in live cell or tissue preparations and in this chapter, we will make a description of FM1-43 staining and destaining protocol using the neuromuscular junction as experimental model. This technique has allowed answering important questions concerning synaptic vesicle recycling, which is a key step for neuronal communication. In addition, FM1-43 has proven to be an excellent tool for investigating membrane internalization and endosome recycling in a variety of cell types. Key words: FM1-43, neuromuscular junction, fluorescence microscopy, exocytosis, endocytosis, synaptic vesicles.

1. Introduction Neurons have an amazing capacity to communicate and this is mediated by the release of chemical messengers (neurotransmitters) during the fusion of synaptic vesicles (exocytosis). Optical monitoring of neuronal activity is an approach that has yielded H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_8, © Springer Science+Business Media, LLC 2011

137

138

Amaral, Guatimosim, and Guatimosim

new information about neuronal communication. In the past, the use of microelectrodes was the only approach with both the spatial and the temporal resolution required to investigate the real-time function of individual neurons or neuronal networks. However, microelectrode recordings have some limitations that motivated researchers to seek for alternative approaches. Styryl dyes are a subgroup of the cyanine dyes that were first designed as voltage-sensitive probes (1) and were successfully used in the past to perform optical detection of neuronal activity. The first optical detection of neuronal activity from a preparation stained with an extrinsic voltage-sensitive dye was reported by Tasaki and collaborators (1968) (2). Voltage-sensitive probe molecules are used to vitally stain a preparation. These dye molecules bind to the external surface of excitable membranes and act as molecular transducers that transform changes in membrane potential into optical signals (reviewed by (2)). For many years, styryl dyes were employed as optical probes for the rapid changes in membrane potential in a wide variety of preparations (3–5). Lichtman and colleagues (1985) (6) visualized the potential of those dyes to stain multiple innervations from snake adult and embryonic neuromuscular junctions and performed pioneering studies of activity-dependent uptake of styryl pyridinium dyes (see also (7)). However, dyes used in their study could not be expanded to nonreptile preparation, which was a major limitation of the technique. This was overcome by William Betz’s group, who introduced a new styryl dye synthesized by Fei Mao, from Molecular Probes at that time, called FM1-43. This dye, which allowed direct visualization of synaptic vesicle recycling in preparations extra-vivo, presented a hydrophobic tail that reversibly binds to membranes; a polar dicationic head that prevents membrane permeation and a body containing two aromatic rings and a double bond that determine spectral fluorescence properties ((8, 9); reviewed by (10)). Therefore, FM1-43 binds to synaptic membrane and when the nerve terminal is submitted to a stimulus that results in exocytosis, and consequently compensatory endocytosis, the fluorescent dye is incorporated, resulting in a typical pattern of staining (9). Another relevant feature is that FM1-43 is weakly fluorescent in water, presenting a quantum yield that increases in two orders of magnitude in lipid environments. So, when a previously labeled terminal is submitted to a new round of stimulation, in the absence of FM1-43 in the external medium, the dye is released to the hydrophilic medium resulting in a decrease of fluorescence intensity, which reflects synaptic vesicles exocytosis ((8, 9, 11)). The hypothesized mechanism of action for FM1-43 was confirmed by ultrastructural evidence that the dye is confined to synaptic vesicles and by fluirometric measurements showing that it is released from the nerve terminals during stimulation (12). Therefore, during the last decades, FM1-43 has been

Use of Styryl Dye FM1-43 to Track Exocytosis and Endocytosis

139

a powerful tool for visualizing exocytosis and endocytosis in a variety of neuronal and non-neuronal cell types, such as endothelial, epithelial, pancreatic beta cells, T-lymphocytes, lung pneumocytes, adipocytes, parathyroid cells and spermatozoa. Considering that original experiments using FM1-43 to visualized synaptic vesicle recycling were performed using nerve-muscle preparations from frog and mice, we will next describe detailed protocols for FM1-43 staining and destaining in both preparations, which were previously described (13, 14).

2. Materials 1. Deionized, distilled water. R R (1 mg) and FM1-43FX (10× 1 µg) purchased 2. FM1-43 TM from Invitrogen .

3. Frog Ringer with the following composition (mM): 115 NaCl, 2.5 KCl, 1.8 CaCl2 , 5 Hepes; pH 7.2 (13). 4. Mouse Ringer with the following composition (mM): 135 NaCl, 5 KCl, 2 CaCl2 , 1 MgCl2 , 12 NaHCO3 , 1 NaH2 PO4 and 11D-glucose. This solution has to be aerated with 95%CO2 −5%O2 and the pH has to be corrected to 7,4 (15). 5. PBS 1× (g/L): 4 NaCl, 2,34 NaH2 PO4 H2 O and 10,32 Na2 HPO4. 6. All salts and buffers used in frog and mouse Ringer or PBS R . can be purchased from Sigma-Aldrich R Silicone Dielectric Gel can be purchased from 7. Sylgard Dow Corning Corporation and mounted in plastic Petri dishes.

8. High potassium solution (60 mM). Qualitatively, it has the same components of frog or mouse Ringer; however, external [Na+ ] should be reduced to compensate the [K+ ] increase. 9. An electrical stimulator which fires tetanic pulses and a suction electrode will be necessary in protocols with electrical stimulation. R 10. D-tubocurarine chloride (Sigma-Aldrich ) diluted in deionized water to make stock solutions of 16 mM stored at −20◦ C. Aliquots will be dissolved in frog or mouse Ringer to a final concentration of 16 µM. R 11. Advasep-7 (Sigma-Aldrich ) diluted in frog or mouse Ringer solution to obtain a final concentration of 1 mM.

140

Amaral, Guatimosim, and Guatimosim R 12. Paraformaldehyde (Sigma-Aldrich ) solution: prepared 4% ◦ in PBS and kept at 4 C. R ) solution: 1.5 mg/mL of PBS. 13. Glycine (Sigma-Aldrich

14. Mounting Media: ProLong Antifade Kit (InvitrogenTM ) or HydramountTM (National Diagnostic, Atlanta, GA) 15. Glass slides and coverslips (VWR International, West Chester, PA) 16. Fluorescence microscope equipped with FITC optics. 17. Confocal microscope equipped with a 488 nm laser for specimen excitation and emission at 510 nm.

3. Methods 3.1. Making Stock Solutions of FM1-43

1. FM1-43 is a water-soluble probe commercialized as lyophilized material that can be dissolved in either deionized water or DMSO (Dimethyl sulfoxide) to make stock solutions. For neuromuscular junctions staining, it is typically used stock solutions of 4 mM (see Note 1). 2. Aliquots should be stored at −20◦ C to preserve the chemical properties of the probe and avoid solvent evaporation that can occur with stock solutions in deionized water, stored at 2–8◦ C for prolonged time. 3. During storage or handling, protect aliquots from light.

3.2. Monitoring Synaptic Vesicle Cycling in Frog Neuromuscular Junctions 3.2.1. Staining Frog Neuromuscular Junctions Synaptic Vesicles with FM1-43

1. Frog cutaneous pectoris nerve-muscle preparation is an excellent model that can be used to stain recycling vesicles with FM1-43. This model provides a very thin muscle associated to a large number of motor terminals. R 2. Dissected muscles should be mounted in a Sylgard -lined Petri dish containing the appropriate saline solution (see Note 2). Entomological pins could be used to fix specimens R to the Sylgard (see Note 3).

3. Remove the excess of connective tissue over the specimen. The preparation has to be free of debris, unwanted or damaged tissue that could retain FM molecules and hinder efficient staining.

Use of Styryl Dye FM1-43 to Track Exocytosis and Endocytosis

141

4. Add working samples of FM1-43 to the Ringer, which will be used to bath the specimen until a final concentration of 4 µM. 5. FM dyes are internalized by recycling vesicles, so a stimuli should be applied to induce exocytosis (vesicle release) and FM uptake during compensatory endocytosis (see Note 4). In frog cutaneous pectoris nerve-muscle preparations, depolarizing stimuli with KCl (60 mM for 10 min) or electrical pulses (20 Hz, 0.5 ms, square wave pulses, 4 V during 10 min), fired by a suction electrode through the nerve fragment, are efficient means to load recycling vesicles with FM1-43 (Fig. 8.1). 6. After stimulation, the dissected muscles should rest in saline solution with FM1-43 (4 µM) for 15 min to assure complete staining of recycling vesicles (see Note 5).

Fig. 8.1. Illustrative destaining of a frog motor nerve terminal labeled with FM1-43. a Representative frog neuromuscular junction nerve terminal stained with FM1-43 during electrical stimulation (20 Hz, 10 min). Note the punctuate pattern of synaptic vesicles clusters labeled with the fluorescent dye (Scale bar: 10 µm). b–f When submitted to a second round of stimulation with high potassium solution (KCl 60 mM), the same terminal presented in panel “a” shows destaining as synaptic vesicles are exocytosed. b, c, d, e and f represent images acquired after 1, 3, 5, 7 and 10 min of stimulation with high potassium, respectively. g Time-course curve of the illustrative destaining presented from “a” to “f” (Seven fluorescence spots were considered for analysis. Error bars: S.E.M).

3.2.2. Removing the Non-internalized FM1-43 from Frog Neuromuscular Junction Preparations

1. The excess of non-internalized FM adhered to the membrane of muscle cells or to the myelin of nerves can be removed during a washing period in saline solution without the probe for at least one hour (see Notes 6 and 7). However, if longer washing is possible (e.g. overnight at 4–8◦ C), it will be more effective to reduce the background fluorescence.

142

Amaral, Guatimosim, and Guatimosim

2. Advasep-7 (1 mM), a sulfonated β-cyclodextrin, which has a higher affinity for FM1-43 than the plasma membrane (16), can be used to remove FM1-43 nonspecifically bound to the outer leaflet of the plasma membrane or extracellular molecules, significantly reducing background staining and the washing time after FM1-43 labeling (see Note 8). 3. After washing, images should be acquired with the appropriate optical parameters described in the following sections. When fixable form of FM1-43 is used, tissue fixation should be carried after the washing time. 3.2.3. Tissue Fixation

1. Immerse the frog cutaneous pectoris muscle in a 4◦ C solution of paraformaldehyde (4% in PBS, pH: 7.2) for 20–40 min. 2. After fixation, wash muscles in a solution of glycine (1.5 mg/mL of PBS) for 15 min in order to quench the fluorescence of paraformaldehyde. 3. Cut insertions of the cutaneous pectoris and remove the skin and sternum fragments used to fix the muscle to the Sylgard. 4. Put the muscle on a slide and then dry the excess of paraformaldehyde using a piece of paper. 5. Cover the specimen with a mounting medium like hydramount or ProLong Antifade (see Note 9). 6. Put a coverslip over the specimen and wait until the mounting medium had dried. The lamina should be maintained at 4–8◦ C and it should be protected from light exposure before image acquisition.

3.2.4. Image Acquisition

1. Images can be collected in “ex-vivo” preparations using fluorescence microscopes equipped with CCD cameras and water immersion objectives (40× or 63×). The excitation light proceeding from Hg lamps passes through filters to select the fluorescein or FITC spectra. 2. In confocal microscope, preparations stained with FM143 should be excited using a 488 nm laser and the emission spectra should be collected from 510 to 626 nm (see Note 10). 3. It is essential to maintain parameters of image acquisition exactly the same when comparing motor terminals in control and test conditions. The ideal image parameters will depend from the quality of dissection and optical resources of microscopes employed on image acquisition. 4. Frog neuromuscular junctions loaded with FM1-43 will appear like a row of fluorescent spots over the muscle fibers length.

Use of Styryl Dye FM1-43 to Track Exocytosis and Endocytosis

3.2.5. Destaining of Non-fixed Preparations of Frog Neuromuscular Junctions

143

1. Non-fixed motor nerve terminals loaded with FM1-43 will destain during a new stimuli round for exocytosis. High potassium solution (60 mM) or tetanic pulses (20 Hz) will stimulate synaptic vesicles release and FM1-43 unloading into saline solution. During destaining, images acquired will reflect the rate of exocytosis (Fig. 8.1) 2. To prevent muscle contractions during destaining, the preparation should be preincubated with curare (16 µM) added to the bath during the washing time. Twitches during destaining can make the experiment impracticable and determine the loss of data.

3.2.6. Image Analysis

1. Softwares like Metamorph Imaging System 7.5 and Image J allow to draw lines around regions of interest and can be used to measure the brightness levels from each fluorescent spot. 2. The fluorescence signal emitted by clusters of synaptic vesicles loaded with FM1-43 represents an estimate of endocytosis. The examiner can compare the brightness of motor terminals stained in a control condition with the brightness of motor terminals stained in a test condition, which might influence the recycling of synaptic vesicle and FM143 uptake. The data obtained can be plotted in histograms using softwares like Microsoft Excel, Sigma Plot 10.0 or Graph Pad Prisma 5. 3. To monitor exocytosis, the brightness of each spot can be quantified during all destaining or at specific time intervals (for example, at every one or five minutes). The data registered can be plotted as time-course curves. For FM143 destaining experiments, it is important to make control curves that indicate the photobleaching levels. The photobleaching curves should be used for comparative analysis with experimental conditions that stimulate or inhibit vesicle release.

3.3. Monitoring Synaptic Vesicle Cycle in Mouse Neuromuscular Junctions 3.3.1. Staining Mouse Neuromuscular Junctions Synaptic Vesicles with FM1-43

1. The protocol used to stain/destain mouse neuromuscular junctions is quite similar to that described for frog neuromuscular junctions. However, some important aspects have to be stood out. 2. In preparations of mouse neuromuscular junctions, the diaphragm can be easily dissected associated to a fragment

144

Amaral, Guatimosim, and Guatimosim

of phrenic nerve. During dissection, remove any excess of connective tissue and avoid accumulation of blood over the muscle (see Note 11). R -lined 3. Dissected muscles should be mounted in a Sylgard Petri dish containing the appropriate mouse Ringer solution. Entomological pins could be used to fix specimens to the R Sylgard (see Note 12).

4. Add working samples of FM1-43 to the mouse Ringer, which will be used to bath the specimen until a final concentration of 4 µM (see Note 13). 5. To promote synaptic vesicle recycling and FM1-43 uptake, stimulate the muscle nerve preparation with high K+ solution (60 mM for 10 min) or electrical pulses (20 Hz, 0.5 ms, square wave pulses, 4 V for 10 min) fired by a suction electrode through the fragment of phrenic nerve (see Note 14). 6. After stimulation, hemidiaphragms should rest in mouse Ringer solution with FM1-43 (4 µM) for 15 min to assure complete staining of recycling vesicles (Fig. 8.2). 3.3.2. Removing the Non-internalized FM1-43 in Preparations of Mouse Neuromuscular Junctions

1. The excess of non-internalized FM adhered to the membrane is removed during a washing period in saline solution without the probe. However, preparations of mouse neuromuscular junctions should be washed between 20 and 40 min in mouse Ringer at room temperature. The Ringer should be aerated with carbogenic mixture during the washing time. 2. Advasep-7 (1 mM) can be used to reduce background fluorescence (see Note 15). 3. After the washing time, images should be acquired with the appropriate optical parameters for FM1-43 fluorescence. When fixable form of FM1-43 is used, tissue fixation should be carried after the washing time.

3.3.3. Tissue Fixation and Mounting of Slides

1. Immerse the mouse hemidiaphragms in a 4◦ C solution of paraformaldehyde (4% in PBS; pH: 7.4) for 20–40 min. 2. After fixation, wash muscles in a solution of glycine (1.5 mg/mL of PBS) for 15 min to quench the fluorescence from paraformaldehyde. 3. Detach hemidiaphragms from its insertions on the ribs. 4. Put the muscle on a slide and then dry the excess of paraformaldehyde using a piece of paper. 5. Cover specimen with mounting medium such as hydramount or ProLong Antifade as was done for frog neuromuscular junction preparation.

Use of Styryl Dye FM1-43 to Track Exocytosis and Endocytosis

145

Fig. 8.2. Destaining of a mouse motor nerve terminal labeled with FM1-43. a Illustrative image of mouse neuromuscular junction stained with FM1-43FX, a fixable analog of FM1-43 (Scale bar: 10 µm). b Mouse neuromuscular junction labeled with FM1-43 during a high potassium stimulus (60 mM, 10 min – Scale bar: 10 µm). c–e The same terminal presented in “b” shows significant destaining during a second stimulus with high potassium solution. Images in c, d and e were acquired after 1, 3 and 5 min of stimulation. f Time-course curve of the destaining observed in images b to e (Six fluorescence spots were considered for analysis. Error bars: S.E.M).

6. Keep slides at low temperature (2–8◦ C) and protect them from light exposure. 3.3.4. Destaining of Non-fixed Mouse Neuromuscular Junctions

1. Non-fixed mouse neuromuscular junctions labeled with FM1-43 can be destained during a second round of stimulation with tetanic pulses (20 Hz) or high potassium solution (KCl 60 mM) (Fig. 8.2). 2. Hemidiaphragms should be preincubated with curare (16 µM) to avoid muscle contractions, which can cause disturbance during image acquisition.

3.3.5. Image Acquisition and Analysis

1. Images can be collected in “ex-vivo” or fixed preparations of mouse neuromuscular junctions according to the same parameters described for frog motor end plates stained with FM1-43. 2. Image analysis can be processed using the same softwares employed on image analysis of frog neuromuscular junctions.

146

Amaral, Guatimosim, and Guatimosim

4. Notes 1. Vials containing FM1-43 should be stored at −20◦ C and prepared preferentially in deionized water. Dye dilution in deionized water can prevent any interference of DMSO in the experimental data. It is important to homogenate solutions to guarantee the correct concentration of the probe in each working aliquot. 2. Frog cutaneous pectoris nerve-muscle preparation can be dissected from animals (e.g. Rana catesbeiana) with approximately 60 g. The muscle should be maintained in frog Ringer during experiment. R 3. When attaching muscles to the Sylgard , do not insert pins over the muscle. Use the skin and sternum fragments associated to frog cutaneous pectoris muscle to insert pins in order to avoid muscle damage and increase in background fluorescence.

4. Before applying the stimulus, it is recommended to loosen muscles in order to prevent damage to the muscular fibers during tetanic contraction. Any lesion to the muscle could result in FM1-43 retention in the sarcolema. Curare (16 µM) should be used to avoid harmful contractions of muscles. 5. Staining with FM1-43 should be conducted at room temperature. Low temperature inhibits endocytosis and compromises FM1-43 uptake. 6. Since FM1-43 partitioning into membranes is reversible, during the washing time, it will spread out to the bath. These FM molecules free on the extracellular medium emit no significant fluorescence when compared to molecules linked to the membrane and they will not interfere with the fluorescent signal from vesicular pools labeled with the dye. Washing period after staining represents a crucial step to reduce background fluorescence. 7. To inhibit spontaneous vesicle release and consequent loss of fluorescent signal during the washing time, preparations should be maintained at low temperature (4–8◦ C). 8. A well-dissected nerve-muscle preparation of frog cutaneous pectoris stained with FM1-43 and then submitted to a sufficient washing time will not present significant background fluorescence. Therefore, the use of Advasep-7 is unnecessary in this case. 9. ProLong Antifade can be more useful since it provides fluorescence stabilization and reduces photobleaching with no quenching of the fluorescence signal.

Use of Styryl Dye FM1-43 to Track Exocytosis and Endocytosis

147

10. The use of a confocal microscope may increase image quality since they can eliminate most of the out-of-focus light and removes background fluorescence. Moreover, confocal microscopes allow optical sectioning and 3D reconstructions. 11. Mice diaphragms can be dissected since fetal until adulthood (e.g. Swiss or C57). The muscle associated to the fragment of phrenic nerve should be maintained in aerated mouse Ringer (95%CO2 –5%O2 ) during the experiment. Blood accumulation over the muscle, excess of connective tissue or damaged areas will turn image acquisition into a quiet difficult task. The diaphragm can be sectioned in hemidiaphragms in order to offer a control and a test specimen for paired experiments. R , insert pins on ribs 12. When attaching muscles to the Sylgard edges associated to mouse diaphragm and in its tendinous centre.

13. 4 µM of FM1-43 is sufficient to label mouse neuromuscular junctions. If it is necessary to improve the staining, the final concentration could be adjusted to 8 µM, which is used to label fetal mouse neuromuscular junctions. 14. Before stimulus application, loosen muscles in order to prevent damage to the muscular fibers during tetanic stimulation and add curare (16 µM) to avoid any harmful contraction of hemidiaphragms. 15. As mentioned previously, a well-dissected preparation of neuromuscular junction usually presents low background fluorescence, making the use of Advasep-7 not essential.

Acknowledgments The authors would like to thank Professor William J. Betz for transforming FM dyes into a powerful tool that allowed the visualization of the synaptic vesicles’ life cycle. This work was supported by CNPq, FAPEMIG and CAPES. References 1. Grinvald, A., Hildesheim, R., Farber, I. C., Anglister, L. (1982) Improved fluorescent probes for the measurement of rapid changes in membrane potential. Biophys J 39, 301–308.

2. Tasaki, I., Watanabe, A., Sandlin, R., Carnay, L. (1968) Changes in fluorescence, turbidity, and birefringence associated with nerve excitation. Proc Natl Acad Sci USA 61, 883–888.

148

Amaral, Guatimosim, and Guatimosim

3. Cohen, L. B., Salzberg, B. M., Grinvald, A. (1978) Optical methods for monitoring neuron activity. Annu Rev Neurosci 1, 171–182. 4. Cohen, L. B., Salzberg, B. M. (1978) Optical measurement of membrane potential. Rev Physiol Biochem Pharmacol 83, 35–88. 5. Waggoner, A. S. (1979) The use of cyanine dyes for the determination of membrane potentials in cells, organelles, and vesicles. Methods Enzymol 55, 689–695. 6. Lichtman, J. W., Wilkinson, R. S., Rich, M. M. (1985) Multiple innervation of tonic endplates revealed by activity-dependent uptake of fluorescent probes. Nature 314, 357–359. 7. Lichtman, J. W., Wilkinson, R. S. (1987) Properties of motor units in the transversus abdominis muscle of the garter snake. J Physiol 393, 355–374. 8. Betz, W. J., Bewick, G. S. (1992) Optical analysis of synaptic vesicle recycling at the frog neuromuscular junction. Science 255, 200–203. 9. Betz, W. J., Mao, F., Bewick, G. S. (1992) Activity-dependent fluorescent staining and destaining of living vertebrate motor nerve terminals. J Neurosci 12, 363–375. 10. Betz, W. J., Mao, F., Smith, C. B. (1996) Imaging exocytosis and endocytosis. Curr Opin Neurobiol 6, 365–371.

11. Gaffield, M. A., Betz, W. J. (2006) Imaging synaptic vesicle exocytosis and endocytosis with FM dyes. Nat Protoc 1, 2916–2921. 12. Henkel, A. W., Lübke, J., Betz, W. J. (1996) FM1-43 dye ultrastructural localization in and release from frog motor nerve terminals. Proc Natl Acad Sci USA 93, 1918–1923. 13. Guatimosim, C., Romano-Silva, M. A., Gomez, M. V., Prado, M. A. (1998) Recycling of synaptic vesicles at the frog neuromuscular junction in the presence of strontium. J Neurochem 70, 2477–2483. 14. Prado, V. F., Martins-Silva, C., de Castro, B. M., et al. (2006) Mice deficient for the vesicular acetylcholine transporter are myasthenic and have deficits in object and social recognition. Neuron 51, 601–612. 15. Xu, Y. F., Atchison, W. D. (1996) Effects of omega-agatoxin-IVA and omega-conotoxinMVIIC on perineurial Ca++ and Ca(++ )activated K+ currents of mouse motor nerve terminals. J Pharmacol Exp Ther 279, 1229–1236. 16. Kay, A. R., Alfonso, A., Alford, S., et al. (1999) Imaging synaptic activity in intact brain and slices with FM1-43 inC elegans, lamprey, and rat. Neuron 24, 809–817.

Chapter 9 Imaging Lipid Bodies Within Leukocytes with Different Light Microscopy Techniques Rossana C.N. Melo, Heloisa D’Ávila, Patricia T. Bozza, and Peter F. Weller Abstract Lipid bodies, also known as lipid droplets, are present in most eukaryotic cells. In leukocytes, lipid bodies are functionally active organelles with central roles in inflammation and are considered structural markers of inflammatory cells in a range of diseases. The identification of lipid bodies has methodological limitations because lipid bodies dissipate upon drying or dissolve upon fixation and staining with alcohol-based reagents. Here we discuss several techniques to detect and visualize lipid bodies within leukocytes by light microscopy. These techniques include staining with osmium or use of different fluorescent probes such as Nile red, BODIPY, Oil red, P96 and immunofluorescence labeling for adipose differentiation-related protein (ADRP). Key words: Lipid bodies, lipid droplets, leukocytes, bright field and fluorescence microscopy, osmium staining, nile red, oil red O, BODIPY, 1-pyrenedodecanoic acid, adipose differentiationrelated protein (ADRP).

1. Introduction Lipid bodies, also named lipid droplets or adiposomes, are now recognized as key organelles involved in lipid storage and metabolism, cell signaling and inflammation (1, 2). Lipid bodies are lipid-rich organelles distributed in the cytoplasm as roughly spherical organelles lacking a delimiting classical bilayer membrane (3–6), but surrounded by an outer monolayer of phospholipids, which at least in some cells may have a unique fatty acid composition (4, 7). The internal core of lipid bodies is rich H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_9, © Springer Science+Business Media, LLC 2011

149

150

Melo et al.

in neutral lipids; and it is likely that more complex membranous domains, often obscured by overlying neutral lipids, are present within lipid bodies (8, 9). Indeed, studies of lipid bodies are providing functional, ultrastructural and protein compositional evidences that lipid bodies are not inert depots of neutral lipid. Rather, it has become evident that lipid bodies are highly regulated, dynamic and functional organelles. Over the past years substantial progresses have been made to demonstrate that lipid body biogenesis is a highly regulated process, which culminates in the compartmentalization of a specific set of proteins and lipids (reviewed in (1, 2)). Lipid body accumulation within cells is observed in both clinical and experimental metabolic, infectious, neoplasic and other inflammatory conditions. Because lipid bodies can be destroyed by drying or fixation and staining with alcohol-based reagents, there are consequently some methodological limitations to their study (2, 10). Indeed, routinely used hematological staining as May–Grunwald–Giemsa staining lead to dissolution of lipid bodies commonly precluding their identification. However, using appropriate fixation procedures followed by methods of identification of lipids and/or of lipid body-specific proteins, lipid bodies can be readily identified within cells. In this chapter, we detail different techniques to visualize lipid bodies in different cell suspensions such as leukocytes isolated from the blood, cell lineages and peritoneal, pleural or bronchoalveolar cells.

2. Materials 2.1. Osmium Staining

1. Sodium cacodylate (cacodylic acid – sodium salt) is dissolved (4.28 g) in 180 mL of distilled water. Adjust pH to 7.4 with HCl and then make up to 200 mL with distilled water for 0.1 M final concentration. 2. Osmium tetroxide (see Note 1): to prepare a stock solution (1.5%), dissolve 1.5 g of osmium tetroxide in 100 mL of 0.1 M sodium cacodylate buffer. Aliquot in small glass tubes (∼2 mL per tube) and store at 4–8◦ C. Protect from light. 3. Paraformaldehyde or formaldehyde solution (formalin) (see Note 2). For paraformaldehyde preparation, dilute paraformaldehyde to 2% in Hanks-buffered salt solution without calcium chloride and magnesium chloride (HBSS−/− ) or phosphate-buffered saline (PBS). Dilutions should be made in fume hood and fresh dilutions of paraformaldehyde should be used in each experiment. Protect from light. For formalin preparation, dilute formalin (saturated solution of formaldehyde 37%) to 3.7% in PBS or HBSS−/− . Adjust to pH 7.4.

Imaging Lipid Bodies Within Leukocytes

151

4. Thiocarbohydrazide solution 1% should be freshly prepared by dissolving 50 mg of thiocarbohydrazide in 5 mL of distilled water. The solution should be heated at a hot plate or microwave (30 s), followed by cooling to room temperature just prior to use. 5. Aqueous mounting medium. 6. Liquid blocker pen. 7. Glass microscope slides and coverslips. 2.2. Nile Red

1. Nile red: 9-diethylamino-5H-benzo [α] phenoxazine-5-one is a phenoxazone dye 1-Acho que é phenoxazine dye, 2incluir espaço entre as palavras poorly soluble in water but it does dissolve in a wide variety of organic solvents (11). Stock solution: dissolve Nile red in acetone (1 mg/mL). Aliquot in small test tubes and store at −20◦ C. Working solution (prepare fresh): dilute at 1:10,000 in HBSS−/− or PBS from the stock solution. Keep protected from light (see Note 3). 2. Paraformaldehyde or formaldehyde solution (see Note 2). Refer to Section 2.1 (Item 3) for fixative preparation. 3. Anti-fading mounting medium for fluorescence microscopy. 4. Glass microscope slides and coverslips.

2.3. Oil Red O

1. Oil Red O: 1-([4-(Xylylazo)xylyl]azo)-2-naphthol, MW 408.49 is prepared at 0.5%: add 5 mL of propylene glycol (100%) to 0.5 g of oil red O with stirring and gradually complete the volume with propylene glycol to 100 mL. Heat the solution until 95◦ C, but do not allow going over 100◦ C. Filter through paper filter. The solution can be stored at room temperature. 2. Hematoxylin solution. 3. Aqueous mounting medium. 4. Paper filter.

2.4. BODIPY

R 1. BODIPY 493/503: 4,4-difluoro-1,3,5,7,8-pentamethyl4-bora-3a,4a-diaza-s-indacene, (molecular weight: 262; Molecular Probes, cat no. D-3922) is stored at−20◦ C, protected from light (see Note 3). Stock solution: dissolve BODIPY in dimethyl sulfoxide (DMSO) at 1 mM. Aliquot in small test tubes (∼10 µL per tube) and store at −20◦ C. Working solution (prepare fresh): dilute 1000× in HBSS−/− . All solutions must be protected from light (see Note 3).

2. Paraformaldehyde or formaldehyde solution (see Note 2). Refer to Section 2.1 (Item 3) for fixative preparation. 3. Anti-fading mounting medium for fluorescence microscopy. 4. Glass microscope slides and coverslips.

152

Melo et al.

2.5. 1-Pyrenedodecanoic Acid

1. 1-pyrenedodecanoic acid (molecular weight 400.56.2; Molecular Probes, cat. no. P-96). It is poorly soluble in water but it does dissolve in a variety of organic solvents including DMSO. Stock solution: dissolve P96 in DMSO (10 mM). Aliquot in small test tubes (∼10 µL per tube) and store at −20◦ C, protected from light. Working solution (prepare fresh): dilute 1000× in HBSS−/− or Hanksbuffered salt solution with calcium chloride and magnesium chloride (HBSS−/− ) (see Note 4). Keep protected from light (see Note 3). 2. Formaldehyde solution (formalin) prepared as Section 2.1 (Item 3). 3. Glass microscope slides and coverslips. 4. Anti-fading mounting medium for fluorescence microscopy.

2.6. Adipose DifferentiationRelated Protein (ADRP, Adipophilin)

1. Monoclonal or polyclonal antibody to ADRP. 2. Fluorescent-labeled secondary antibodies. 3. Formaldehyde solution (formalin) prepared as Section 2.1 (Item 3). R X-100 (t-Octylphenoxypoly-ethoxyethanol). 4. Triton

5. Glass microscope slides and coverslips. 6. Anti-fading mounting medium for fluorescence.

3. Methods 3.1. Sample Preparation onto Slides

After obtaining a cell suspension (∼0.5–1.0 × 106 cells/mL of medium), preparation of samples (cell suspensions) onto slides for lipid body staining can be done in two ways: using a cytocentrifuge or by spreading a mixture of cells with melted agarose matrix onto a slide. For comparison of cell morphology observed with these two techniques, refer to Note 5. For cytospin preparations: label slides and cytocentrifuge a volume of 100 µL (∼0.5–1.0 × 105 cells) of a cell suspension sample, at 18–23 g for 5 min. For agarose preparations: prepare first an agarose matrix. Weigh 0.125 g of agarose (low-melting point agarose; mp 65.5◦ C, gelling point 24◦ C) into a 125-mL Erlenmeyer flask and dilute to 2.5% by adding 5 mL of distilled water. Cover with aluminum foil. Mix well, but avoid swirling to prevent agarose binding to flask wall. Solubilize agarose in 70◦ C water bath for 15 min with gentle agitation. Aliquot in small test tubes (∼1.5 mL) and store at 4◦ C. For slide preparation, resolubilize a tube at 70◦ C,

Imaging Lipid Bodies Within Leukocytes

153

mix cells with the liquid matrix and gently spread the cell–agarose mixture (∼20 µL) onto microscope slides using a microtip. Use covered surface slides for cell adhesion. For spreading the cells, use surface tension to move agarose mixture throughout the slide. Cover gently this thin layer of agarose/cell mixture with a perfusion chamber (CoverWellTM ). 3.2. Osmium Staining

Properties: Osmium tetroxide binds to unsaturated lipids, and is reduced by organic materials to elemental osmium, an easily visible and permanent black substance (Fig. 9.1a). Reduction of osmium by thiocarbohydrazide highly enhances lipid labeling. 1. Prepare slides with samples using a cytocentrifuge (see Section 3.1). 2. Fix samples, while still moist, with paraformaldehyde or formalin for 10 min. Refer to Note 2 for cell fixation and Note 6 for “moist cells”. 3. Rinse slides in distilled water. 4. Circumscribe the adhered cells with liquid blocker pen to facilitate the staining procedure. 5. Stain the adhered cells by adding one drop of 0.1 M cacodylate buffer and one drop of 1.5% osmium tetroxide for 30 min. Refer to Note 1 for osmium manipulation. 6. Rinse slides in distilled water. 7. Immerse in thiocarbohydrazide solution for 5 min at room temperature. 8. Rinse the adhered cells twice with distilled water. 9. Re-stain by adding one drop of 0.1 M cacodylate buffer and one drop of 1.5% osmium tetroxide for 3 min. 10. Rinse slides in distilled water. 11. Let the slides dry. 12. Mount with aqueous mounting medium. Alternatively, osmium staining can be performed on agarose preparations. 1. Spread a mixture of cells/agarose onto a microscope slide and cover with a perfusion chamber (see Section 3.1). 2. Carefully pipet 400 µL of the fixative (2% paraformaldehyde) over sample through the chamber access port, ensuring that chamber area is uniformly saturated and let for 10 min. 3. Rinse slides in distilled water through the chamber access port. 4. Rinse slides in 0.1 M cacodylate buffer through the chamber access port.

154

Melo et al.

Fig. 9.1. Lipid bodies within leukocytes imaged by light microscopy after staining with osmium (a), Nile red (b), BODIPY (c), oil red O (ORO) (d, e) or double labeled with 1-pyrenedodecanoic acid (P 96) and anti-adipose differentiation-related protein (ADRP) (f). Lipid bodies appear as round, dark (a, arrowheads), fluorescent red (b, e) or green (c) organelles distributed throughout the cytoplasm. ORO-stained lipid bodies appear as round red organelles at both bright field (d) and fluorescence (e) microscopy, while the nucleus is imaged in light blue after counterstaining with hematoxylin (d). In f, merged images show P 96-labeled lipid bodies in blue at UV filter and ADRP immunolabeling as a ring in the periphery of the lipid body. In a, cells were counterstained with Hema R kit (Fisher Scientific). a and b show eosinophils isolated from the blood of normal 3 human donors and stimulated with eotaxin (a) or calcium ionophore (b) as before (20, 21). c and d show murine peritoneal leukocytes and macrophages, respectively. Bars, 6 µm (A, B, D, E and F), 10 µm (c).

Imaging Lipid Bodies Within Leukocytes

155

5. Stain by adding 1.5% osmium tetroxide through the chamber access port for 30 min. Refer to Note 1 for osmium manipulation. 6. Rinse slides in 0.1 M cacodylate buffer through the chamber access port. 7. Carefully remove the chamber. 8. Immerse in thiocarbohydrazide for 5 min at room temperature. 9. Rinse slides in 0.1 M cacodylate buffer. 10. Re-stain with 1.5% osmium tetroxide for 3 min. 11. Let the slides dry. 12. Mount with aqueous mounting medium. 3.3. Nile Red Staining

Properties: Nile red is intensely fluorescent, and can serve as a sensitive stain for the detection of cytoplasmic lipid bodies (11, 12). 1. Incubate a cell suspension (1.0 × 106 cells/mL) with a working solution of Nile red (see Section 2.2, Item 1) for 5 min at room temperature and protected from light. Cells are incubated in a test tube. 2. Centrifuge (120 g/5 min) and resuspend in HBSS−/− or PBS to wash cells. Repeat this step once. 3. Cytospin onto slides using 100 µL of cell suspension at 18–23 g for 5 min. 4. Fix with paraformaldehyde or formalin (see Note 2 and Section 2.1, Item 3, for fixative preparation) for 5 min at room temperature. 5. Wash twice in HBSS−/− or PBS. 6. Mount while wet using anti-fading mounting medium. Keep slides in the dark (see Notes 3 and 7). Alternatively, Nile Red staining can be done on agarose preparations. 1. Spread a mixture of cells/agarose onto a microscope slide and cover with a perfusion chamber (see Section 3.1). 2. Carefully pipet 400 µL of the fixative (2% paraformaldehyde) over sample through the chamber access port, ensuring that chamber area is uniformly saturated and let for 5 min. This step can be performed after incubation with Nile Red. Refer to Note 8 for Nile red staining in fixed/unfixed cells. 3. Wash twice with HBSS−/− (2× 400 µL) adding buffer through the chamber access port.

156

Melo et al.

4. Incubate with a working solution of Nile Red (400 µL) for 5–10 min at room temperature. Keep slides protected from light (see Note 3). 5. Wash twice with HBSS−/− . 6. Carefully remove the chamber. 7. Mount while moist with HBSS−/− or with anti-fading mounting medium after drying. Keep slides in the dark (see Note 3). 3.4. Oil Red O Staining

Properties: Oil red O belongs to the polyazo group of dyes which also includes the Sudan series of dyes. The principle of the lipid staining is based on the physical properties of the dye that preferentially divide into lipid-rich compartments. Oil red O staining can be readily visualized in both bright field and fluorescent microscopy (Fig. 9.1d, f) (13). 1. For slide preparation, cytocentrifuge 100 µL of a sample cell suspension, at 18–23 g for 5 min. 2. Fix cells with 3.7% formalin in HBSS−/− (see Section 2.1, Item 3). 3. Wash twice in distilled water. 4. Place slides in absolute propylene glycol for 5 min. 5. Stain in 0.5% oil red O solution (see Section 2.4) for 10 min in the incubator at 60◦ C. 6. Rinse cells in 85% propylene glycol solution for 5 min. 7. Wash twice in distilled water. 8. Counterstain (see Note 9).

with

hematoxylin

solution

for

30

s

9. Wash thoroughly in tap water. 10. Mount with aqueous mounting medium. 3.5. BODIPY Staining

R Properties: BODIPY lipid probe is an effective dye for staining neutral lipids and, for this reason it is very efficient for lipid body staining (Fig. 9.1c). The fluorescence quantum yield of the BODIPY dyes is not diminished in water and this method can be used in conjugation with immunofluorescence (see Note 10 and Chapter 11). 1. Incubate the cell suspension with 1 µM BODIPY for 1 h at 37◦ C. Cells are incubated in a test tube inside a water bath.

2. Pellet the cells (120 g/5 min) and resuspend in HBSS−/− or PBS to wash cells. Repeat this step once. 3. Cytospin onto slides using 100 µL of cell suspension at 18–23 g for 5 min.

Imaging Lipid Bodies Within Leukocytes

157

4. Fix with paraformaldehyde or formalin in HBSS−/− or PBS for 5–10 min (see Section 2.1, Item 3, for fixative preparation). 5. Wash twice in HBSS−/− or PBS. 6. Mount while wet using anti-fading mounting medium. Slides are stored at room temperature in the dark until analysis (see Notes 3 and 7). Alternatively, the BODIPY staining can be done using an agarose preparation. 1. Spread a mixture of cells/agarose onto a microscope slide and cover with a perfusion chamber (see Section 3.1). 2. Wash twice with HBSS−/− (2× 400 µL) adding buffer through the chamber access port. 3. Carefully pipet 400 µL of BODIPY solution over sample through the chamber access port, ensuring that chamber area is uniformly saturated. Place slides on a tray atop hydrated pad. Place tray in humidified incubator (37◦ C, 5% CO2 ) for 1 h. 4. Wash twice with HBSS−/− (2× 400 µL) adding buffer through the chamber access port. 5. Pipet 400 µL of the fixative (2% paraformaldehyde) over sample through the chamber access port, ensuring that chamber area is uniformly saturated and let for 5–10 min. 6. Wash twice in HBSS−/− or PBS. 7. Carefully remove the chamber. 8. Mount while wet with HBSS−/− or anti-fading mounting medium. Keep slides protected from light (see Note 3). 3.6. 1-Pyrenedodecanoic Acid Staining

Properties: 1-Pyrenedodecanoic acid (P-96) is a fluorescent fatty acid analog with the environment sensitive pyrene attached to the terminal carbon atom that is furthest from the carboxylate moiety. P96 is readily incorporated into lipid bodies and P96 fluorescent-labeled lipid bodies are visualized under the UV (excitation/emission 340/376 nm) (Fig. 9.1e) (14, 15). 1. Incubate a cell suspension with a working solution of P96 (see Section 2.5, Item 1). for 1 h at room temperature and protected from light (see Note 3). 2. Pellet the cells (120 g/5 min) and resuspend in HBSS−/− or PBS to wash cells. Repeat this step once. 3. Cytospin onto slides using 100 µL of cell suspension at 18–23 g for 5 min. 4. Fix with 3.7% formalin in HBSS−/− or PBS for 5 min at room temperature (see Section 2.1, Item 3, for fixative preparation).

158

Melo et al.

5. Wash twice in distilled water. 6. Mount while wet using anti-fading medium for fluorescence microscopy (see Note 7). 3.7. ADRP Staining

Properties: Adipose differentiation-related protein (ADRP) is a structural protein of lipid bodies considered essential for lipid storage and metabolism (reviewed in (16)). ADRP is ubiquitously associated with cytoplasmic lipid bodies in different types of cells and is described as a specific protein marker for lipid bodies (Fig. 9.1f) (17–19). This method can be conjugated with immuno-labeling for other proteins (see Note 11). 1. For slide preparation, cytocentrifuge 100 µL of a sample cell suspension, at 18–23 g for 5 min. 2. Fix slides with 3.7% formalin in HBSS−/− or PBS (see Section 2.1, Item 3, for fixative preparation). 3. Wash once in HBSS−/− or PBS. 4. Permeabilize the cells with 0.1% Triton X-100 in HBSS−/− for 10 min. 5. Circumscribe the adhered cells with liquid blocker pen. 6. a. For human cells incubate with mouse anti-human ADRP at dilution of 1:20 (2.5 µg/mL, final concentration) for 1 h at room temperature. b. For mouse, rat, human or bovine cells incubate with guinea pig anti-human ADRP polyclonal antibody at dilution of 1:300 (final dilution) for 1 h at room temperature. 7. Wash three times in HBSS−/− or PBS. 8. Incubate with fluorescent-labeled secondary antibody for 1 h room temperature. 9. Wash three times in HBSS−/− or PBS. 10. Mount in mounting medium for fluorescence microscopy (see Note 7).

3.8. Lipid Body Analysis and Quantification

Lipid body analysis is performed on a bright field (osmium and oil red O staining) or fluorescence microscope (oil red O and fluorescent probes) at 1000×. For example, analyses and image acquisition can be obtained using an Olympus BX-FLA fluorescence microscope equipped with a Plan Apo 100 × 1.4 Ph3 objective (Olympus) and CoolSNAP-Pro CF digital camera in conjunction R with Image Pro Plus software (Media Cybernetics). The detection of lipid bodies using different techniques will appear as round dark (osmium staining, Fig. 9.1a) or fluorescent red (Nile red or Oil red O) (Fig. 9.1b and e, respectively) or green (BODIPY) (Fig. 9.1c) organelles. Of note, Nile red can be observed through both green (fluorescein) and red (rhodamine) channels and Oil red O can be also observed at bright field microscopy (Fig. 9.1d).

Imaging Lipid Bodies Within Leukocytes

159

ADRP immuno-labeling has a characteristic ring-shape appearance as this protein localizes preferentially at the periphery of the lipid body (Fig. 9.1f). Lipid bodies are usually enumerated using a 100× objective lens in 50 consecutively scanned cells (10). Alternatively, lipid bodies can be quantified by the measurement of oil red O (ORO) or BODIPY fluorescent area. The measurement of the area of lipid bodies is obtained with a 60 objective (at least four fields per slide). Images are transformed into black and white pictures and analyzed with the Image 2D software (GE Healthcare). Spots are determined by automatic spot detection and the total area of fluorescent lipid bodies is obtained for each field and divided by the number of cells in the respective field.

4. Notes 1. Osmium tetroxide is volatile and its fumes are very toxic (causes severe irritation to eyes, skin and respiratory tract). Thus, any manipulation involving this chemical must be performed in a fume hood and wearing gloves. 2. Fixation of cells before osmium, Nile Red or BODIPY staining can be done using either 2% paraformaldehyde or 3.7% formalin. 3. When staining with fluorescent probes such as Nile red, P 96 and BODIPY, fluorescence is usually not stable for a long period and fluorescence bleaching will occur after a certain time. Keep the cell preparations in the dark to avoid fluorescence loss. 4. For some type of cells, the P96 staining can have better results using HBSS+/+ . 5. In general, cells kept in agarose show better morphology compared with cells from cytospin preparations because cells are kept in a hydrated system. In addition, shape changes, a feature of activated leukocytes, can be observed when cells are embedded in an agarose matrix. On the other hand, cytospin slides are fast prepared. In both cytospin and agarose preparations, lipid bodies are well preserved and can be easily detected. 6. It is very important to keep cells moist during the staining with osmium. Dried cells will appear with very bad morphology. 7. For fluorescence microscopy, it is important to use a mounting medium that prevents rapid loss of fluorescence

160

Melo et al.

during microscopic examination and retains its anti-fading ability during long-term storage. 8. Staining with Nile red can be carried out on either fixed or unfixed cells with no apparent difference in distribution or intensity of fluorescence. 9. Counterstaining with hematoxylin is important to show nuclear aspects. This step is not mandatory if visualization of nuclei is not necessary. 10. BODIPY staining can be combined with an immunofluorescence protocol (for example, ADRP immuno-labeling). In this case, BODIPY can be mixed with the secondary antibody (Chapter 10). 11. Immuno-labeling for ADRP can be conjugated with immuno-labeling for other proteins by simultaneous incubation with two primary antibodies (raised in distinct hosts) followed by incubation with two secondary antibodies (with distinct ranges of excitation/emission).

Acknowledgments Supported by CNPq and FAPEMIG (Brazil) to RCNM; CNPq, FAPERJ and PRONEX (Brazil) to PTB and NIH grants (USA) AI020241, AI051645 and AI022571 to PFW. We acknowledge Clarissa M. Maya-Monteiro for the Fig. 9.1c used in this chapter. References 1. Martin, S., Parton, R. G. (2006) Lipid droplets: a unified view of a dynamic organelle. Nat Rev Mol Cell Biol 7, 373–378. 2. Bozza, P. T., Melo, R. C. N., BandeiraMelo, C. (2007) Leukocyte lipid bodies regulation and function: contribution to allergy and host defense. Pharmacol Ther 113, 30–49. 3. Dvorak, A. M., Dvorak, H. F., Peters, S. P., Shulman, E. S., MacGlashan, D. W., Jr., Pyne, K., Harvey, V. S., Galli, S. J., Lichtenstein, L. M. (1983) Lipid bodies: cytoplasmic organelles important to arachidonate metabolism in macrophages and mast cells. J Immunol 131, 2965–2976. 4. Tauchi-Sato, K., Ozeki, S., Houjou, T., Taguchi, R., Fujimoto, T. (2002) The surface of lipid droplets is a phospholipid monolayer

with a unique fatty acid composition. J Biol Chem 277, 44507–44512. 5. Murphy, D. J. (2001) The biogenesis and functions of lipid bodies in animals, plants and microorganisms. Prog Lipid Res 40, 325–438. 6. Weller, P. F., Monahan-Earley, R. A., Dvorak, H. F., Dvorak, A. M. (1991) Cytoplasmic lipid bodies of human eosinophils. Subcellular isolation and analysis of arachidonate incorporation. Am J Pathol 138, 141–148. 7. Bartz, R., Li, W. H., Venables, B., Zehmer, J. K., Roth, M. R., Welti, R., Anderson, R. G., Liu, P., Chapman, K. D. (2007) Lipidomics reveals that adiposomes store ether lipids and mediate phospholipid traffic. J Lipid Res 48, 837–847.

Imaging Lipid Bodies Within Leukocytes 8. Wan, H. C., Melo, R. C., Jin, Z., Dvorak, A. M., Weller, P. F. (2007) Roles and origins of leukocyte lipid bodies: proteomic and ultrastructural studies. FASEB J 21, 167–178. 9. Bozza, P. T., Magalhaes, K., Weller, P. F. (2009) Leukocyte lipid bodies–biogenesis and functions in inflammation. Biochim Biophys Acta doi:10.1016/j.bbalip_2009_ 01_005. 10. Melo, R. C. N., Sabban, A., Weller, P. F. (2006) Leukocyte lipid bodies: inflammation-related organelles are rapidly detected by wet scanning electron microscopy. J Lipid Res 47, 2589–2594. 11. Greenspan, P., Mayer, E. P., Fowler, S. D. (1985) Nile red: a selective fluorescent stain for intracellular lipid droplets. J Cell Biol 100, 965–973. 12. Fukumoto, S., Fujimoto, T. (2002) Deformation of lipid droplets in fixed samples. Histochem Cell Biol 118, 423–428. 13. Koopman, R., Schaart, G., Hesselink, M. K. (2001) Optimisation of oil red O staining permits combination with immunofluorescence and automated quantification of lipids. Histochem Cell Biol 116, 63–68. 14. Radom, J., Salvayre, R., Maret, A., Negre, A., Douste-Blazy, L. (1987) Metabolism of 1-pyrenedecanoic acid and accumulation of neutral fluorescent lipids in cultured fibroblasts of multisystemic lipid storage myopathy. Biochim Biophys Acta 920, 131–139. 15. Yu, W., Bozza, P. T., Tzizik, D. M., Gray, J. P., Cassara, J., Dvorak, A. M., Weller, P. F. (1998) Co-compartmentalization of MAP

16.

17.

18.

19.

20.

21.

161

kinases and cytosolic phospholipase A2 at cytoplasmic arachidonate-rich lipid bodies. Am J Pathol 152, 759–769. Brasaemle, D. L. (2007) Thematic review series: adipocyte biology. The perilipin family of structural lipid droplet proteins: Stabilization of lipid droplets and control of lipolysis. J Lipid Res 48, 2547–2559. Brasaemle, D. L., Barber, T., Wolins, N. E., Serrero, G., Blanchette-Mackie, E. J., Londos, C. (1997) Adipose differentiationrelated protein is an ubiquitously expressed lipid storage droplet-associated protein. J Lipid Res 38, 2249–2263. Heid, H. W., Moll, R., Schwetlick, I., Rackwitz, H. R., Keenan, T. W. (1998) Adipophilin is a specific marker of lipid accumulation in diverse cell types and diseases. Cell Tissue Res 294, 309–321. D’Avila, H., Melo, R. C. N., Parreira, G. G., Werneck-Barroso, E., Castro-Faria-Neto, H. C., Bozza, P. T. (2006) Mycobacterium bovis bacillus Calmette-Guerin induces TLR2mediated formation of lipid bodies: intracellular domains for eicosanoid synthesis in vivo. J Immunol 176, 3087–3097. Melo, R. C. N., Perez, S. A. C., Spencer, L. A., Dvorak, A. M., Weller, P. F. (2005) Intragranular vesiculotubular compartments are involved in piecemeal degranulation by activated human eosinophils. Traffic 6, 866–879. Bandeira-Melo, C., Perez, S. A. C., Melo, R. C. N., Ghiran, I., Weller, P. F. (2003) EliCell assay for the detection of released cytokines from eosinophils. J Immunol Methods 276, 227–237.

Chapter 10 EicosaCell – An Immunofluorescent-Based Assay to Localize Newly Synthesized Eicosanoid Lipid Mediators at Intracellular Sites Christianne Bandeira-Melo, Peter F. Weller, and Patricia T. Bozza Abstract Eicosanoids (prostaglandins, leukotrienes and lipoxins) are a family of signaling lipids derived from arachidonic acid that have important roles in physiological and pathological processes. Over the past years, it has been established that successful eicosanoid production is not merely determined by arachidonic acid and eicosanoid-forming enzymes availability, but requires sequential interactions between specific biosynthetic proteins acting in cascade and may involve very unique spatial interactions. Direct assessment of specific subcellular locales of eicosanoid synthesis has been elusive, as those lipid mediators are newly formed, not stored and often rapidly released upon cell stimulation. In this chapter, we discuss the EicosaCell protocol for intracellular detection of eicosanoid-synthesizing compartments by means of a strategy to covalently cross-link and immobilize the lipid mediators at their sites of synthesis followed by immunofluorescent-based localization of the targeted eicosanoid. Key words: Eicosanoids, prostaglandin, leukotriene, biosynthesis, compartmentalization, carbodiimide, EDAC (1-ethyl-3-(3-dimethylamino-propyl) carbodiimide), lipid droplets, phagosomes, perinuclear.

1. Introduction Eicosanoids – including leukotrienes and prostaglandins – are a family of signaling lipids derived from the enzymatic oxygenation of arachidonic acid (AA) that control key processes involving cell–cell communication, including cell activation, proliferation, apoptosis, metabolism and migration (1, 2). Thus, eicosanoids H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_10, © Springer Science+Business Media, LLC 2011

163

164

Bandeira-Melo, Weller, and Bozza

have important roles in physiological and pathological conditions such as tissue homeostasis, host defense, inflammation and cancer. In view of the magnitude of eicosanoid actions, great efforts have been aimed at understanding the biochemical, cellular and molecular aspects of their biosynthetic pathway. In all cells, the highly regulated generation of eicosanoids is dependent on activation of specific phospholipases and specific eicosanoid-synthesizing enzymes and involves small molecules (e.g. Ca2+ ) and activation-dependent localization of enzymes at specific compartments within cells (3–8). Intracellular compartmentalization of eicosanoid synthesis within leukocytes has emerged as a key feature that regulates the amount and may also regulate the eicosanoid produced. Such intracellular sites of eicosanoid formation in any cell have been inferred based on the permanent or temporary localization of specific eicosanoidforming enzymes under proper cell activation, since the direct observation of sites of eicosanoid synthesis has been hard to define as those lipid mediators are newly formed, non-storable and often rapidly released upon cell stimulation. It was recently established that successful eicosanoid production is not merely determined by AA and eicosanoid-forming enzymes availability, but requires sequential interactions between specific biosynthetic proteins acting in cascade, and may involve very unique spatial interactions. Therefore, just by detecting eicosanoid-forming enzymes within discrete subcellular structures, one cannot assure that those sites are indeed accountable for the efficient and enhanced eicosanoid synthesis observed during inflammatory responses. The immunolocalization of eicosanoid-forming proteins does not necessarily ascertain that the localized protein is functional and activated to synthesize a specific eicosanoid lipid at an intracellular site. We previously developed a method to capture and localize the eicosanoid, prostaglandin E2 (PG E2 ), released extracellularly by a nematode parasite (9). By means of a strategy to covalently crosslink, capture and localize newly formed eicosanoids at their sites of synthesis, we developed a more direct approach to detect the intracellular sites of arachidonic acid (AA)-derived lipid mediator formation in leukocytes and other cell types. To develop our new strategy for in situ immuno-localization of newly formed eicosanoids to ascertain the intracellular compartmentalization of their synthesis – the EicosaCell assay – modifications of a prior technique was used (9). The EicosaCell rational relies on the specific features of the heterobifunctional cross-linker 1-ethyl-3-(3-dimethylamino-propyl) carbodiimide (C8 H17 N3 HCl; EDAC) used. EDAC immobilizes newly synthesized eicosanoids by cross-linking the eicosanoid carboxyl groups to the amines of adjacent proteins localized at eicosanoid-synthesizing compartment. Such EDAC-mediated reaction forms a bond without any spacer length between the two molecules, favoring an

EicosaCell

165

accurate positioning of the newly synthesized eicosanoid within the cell. In addition, while other cross-linkers formed bonds that often generate foreign molecules, EDAC-driven eicosanoid-bond is homologous to native eicosanoid that allows immunoassays like EicosaCell. Besides the precise positioned coupling of an immuno-detectable eicosanoid at its sites of formation, EDAC enables: (I) the ending of cell stimulation step; (II) cell fixation; (III) cell permeabilization, allowing the penetration of both anti-eicosanoid and the detecting fluorochrome-conjugated antibodies into cells; and, importantly, (IV) the relative preservation of lipid domains, such as membranes and droplets, which dissipate with air drying or commonly used alcohol fixation.

2. Materials 2.1. Conventional EicosaCell

1. EDAC (1-ethyl-3-(3-dimethylamino-propyl) carbodiimide hydrochloride) is diluted in Hanks-buffered salt solution without calcium chloride and magnesium chloride (HBSS−/− ). Refer to Note 1 for EDAC solution handling. EDAC final concentration with cells varies according to cell type and protocol used (see next subheadings). The working solution should have twice concentration of the final concentration with cells. For instance, specifically regarding purified human eosinophils stimulated as a cell suspension, EDAC final concentration with eosinophils should be 0.1% in HBSS−/− , therefore the EDAC working solution should be diluted to 0.2%. Alternatively, with adherent macrophages stimulated in 6 wells plate, EDAC final concentration should be 0.5% in HBSS−/− , therefore the EDAC working solution should be diluted to 1.0%. 2. Primary antibody to the eicosanoid of interest. 3. Fluorescent-labeled secondary antibodies. 4. Glass microscope slides and coverslips. 5. Anti-fading mounting medium for fluorescence.

2.2. Double-Labeling Purposes

1. DAPI (4′ , 6′ -Diamidino-2-phenylindole dihydrochloride) stock solution is prepared by dissolving 1 mg/mL of powder in distilled water. Aliquots should be stored at −20◦ protected from light. 2. Monoclonal antibody against membrane protein (LAMP) 1.

lysosome-associated

166

Bandeira-Melo, Weller, and Bozza R 3. BODIPY 493/503 (4,4-difluoro-1,3,5,7,8-pentamethyl4-bora-3a,4a-diaza-s-indacene) (Molecular Probes; cat no. D-3922, molecular weight: 262). To prepare BODIPY stock solution, BODIPY should be dissolved in DMSO (1 mM), aliquoted in small Eppendorf tubes (∼10 µL per tube) and stored at −20◦ C protected from light. BODIPY working solution should be diluted fresh 1000× in HBSS−/− and kept from light.

4. Monoclonal or polyclonal antibody differentiation-related protein (ADRP).

to

adipose

3. Methods 3.1. EicosaCell with Cells in Suspension

EicosaCell can be easily performed with a varied of cell types in suspension, such as purified human blood leukocytes, cell lineages, as well as, peritoneal, pleural or bronchoalveolar animal cells. After in vivo or in vitro stimulation of these cell populations, incubation with EDAC should instantaneously guarantee the immobilization of eicosanoids at their synthesizing spot within the cell, just before cytospin slides are prepared to allow microscopic analysis. As schematically illustrated in Fig. 10.1a, after preparing a cell suspension, EDAC working solution should be added to cell suspension and incubated for a period of time to ensure cell fixation, immobilization of eicosanoid and cell permeabilization. 1. Prepare a cell suspension of 2 × 106 /mL. Gently and immediately add an equal volume of EDAC solution, prepared as described in Section 2.1, Step 1 (refer to Notes 1 and 2 for details), to the cell suspension. 2. Incubate the cell suspension with EDAC for 30 min to 1 h at 37◦ C. 3. Cytospin the cells onto slides using 100 µL of the cell suspension at 23 g for 5 min. 4. Wash twice in HBSS−/− . 5. Labeling of newly formed eicosanoids can be done with a variety of already tested antibodies, as already published elsewhere (10–13). Incubate cells with the primary antibody to the eicosanoid of interest for 1 h at room temperature. The non-immune serum from the animal where the secondary antibody was produced may be added to the primary antibody so as to decrease unspecific labeling.

EicosaCell

A

167

1.In 1. vivo stimulation

CO 2

harvested leukocytes

EDAC: - eicosanoid-protein cross-linking - cell permeabilization

2. Fixation

3. Cystospin - anti-eicosanoid - fluorochrome-labeled secundary Ab

Y

Y

*

Y

*

*

4.Detection

Fluorescence Microscopy

B

cells/agarose mixture

o

± stimuli o

1. Induction

A23187 Lipid bodies LTC4 15 min o 37 C

1h 37oC

EDAC: 2. Fixation

- eicosanoid-protein cross-linking - cell permeabilization - anti-eicosanoid - fluorochrome-labeled secundaryAb

*

*

Y

Y

*

Y

3. Detection

Phase-Contrast and Fluorescence Microscopy

Fig. 10.1. Schematic illustration of EicosaAssay method. EicosaCell preparations, which undergo EDAC-dependent capturing and fixation of newly formed-eicosanoids at their sites of synthesis, are analyzed by phase-contrast and fluorescence microscopy and can employ cytospun cells (a), adherent cells or cells embedded in a gel matrix (b).

168

Bandeira-Melo, Weller, and Bozza

6. Wash 2–3 times in HBSS−/− . 7. Incubate with the fluorescent-labeled secondary antibody for 1 h at room temperature. 8. At the end of the staining procedure, cytospun cells should be always extensively washed with HBSS−/− , at least 3 times for 5 min each. 9. Slides should be mounted using an aqueous mounting medium, preferentially with anti-fading. 10. Analysis is performed on phase contrast to observe cell morphology and fluorescence microscope or confocal scanning laser microscope to identify the eicosanoid labeling. For example, analyses and image acquisition can be obtained using an Olympus BX-FLA fluorescence microscope equipped with a Plan Apo 100× 1.4 Ph3 objective and CoolSNAP-Pro CF digital camera in conjunction with R Image Pro Plus software (Media Cybernetics) (see Note 3 for details). 11. The specificity of the eicosanoid immuno-labeling using EicosaCell system should be always ascertained by including some mandatory control conditions as detailed in Note 4. As shown in Fig. 10.2, EicosaCell system was successfully employed on macrophages recovered from pleural cavities of BCG-infected or control mice (11). Briefly, cells obtained 24 h after infection with BCG and controls were recovered from the pleural cavity with 500 µL of HBSS−/− and immediately mixed with 500 µL of EDAC (1% in HBSS−/− ). After 30 min incubation at 37◦ C with EDAC, pleural leukocytes were then washed with HBSS−/− , cytospun onto glass slides and incubated with mouse anti-PGE2 in 0.1% normal goat serum and guinea pig polyclonal anti-mouse ADRP (see Section 2.2) in 0.1% normal donkey serum simultaneously for 1 h at room temperature. Isotyping matching antibodies (murine IgG1) were used as controls (Fig. 10.2). Cells were washed twice and incubated with secondary antibodies, goat antimouse conjugated with AlexaFluor-488 (1/1000, Molecular Probes) and CY3-conjugated donkey anti-guinea pig (1/1000). Slides were washed (three times, 10 min each) and mounted with aqueous mounting medium. Cells were analyzed by both phase-contrast and fluorescence microscopy. As a control for PGE2 specificity of detection, one group of BCG-infected animals was treated with indomethacin (4 mg/kg), 4 h before sacrificing animals for cell recovery (not shown).

EicosaCell

169

Fig. 10.2. EicosaCell for PGE2 immuno-localization within BCG-infected cytospun macrophages. In upper panels, macrophages from BCG-infected animals were labeled for ADRP-associated lipid bodies (red staining) and for newly formed PGE2 (green staining). Merged image showed co-localization of PGE2 in ADRP-associated lipid bodies (yellow staining). In bottom panels, IgG1 irrelevant isotype (MOPC) was used as control for PGE2 labeling. Briefly, pleural macrophages obtained 24 h after infection with BCG were recovered from the thoracic cavity with 500 µL of HBSS, immediately mixed with 500 µL of EDAC (1% in HBSS) and incubated for 30 min at 37◦ C. Cells were then washed with HBSS, cytospun onto glass slides and incubated with mouse anti-PGE2 (1/100) or MOPC 21 in 0.1% normal goat serum and guinea pig polyclonal anti-mouse ADRP (1/1000) in 0.1% normal donkey serum simultaneously for 1 h at RT. Cells were washed twice and incubated with secondary Abs, goat anti-mouse conjugated with AlexaFluor-488 (1/1000) and CY3-conjugated donkey anti-guinea pig (1/1000). The slides were washed (three times, 10 min each) and mounted with aqueous mounting medium.

3.2. EicosaCell with Adherent Cells

To study the intracellular compartmentalization of eicosanoid synthesis by EicosaCell in adherent cells, extra care should be taken to ensure the conservation of cell adherence and morphology during EDAC step. EicosaCell have succeeded to immunolocalize PGE2 within at least three distinct cell types: plated murine macrophages (D’Avila et al., unpublished) and two lineages of intestinal cells, CACO-2 (a human colon adenocarcinoma cell line) (Fig. 10.3) (14) and IEC-6 (a rat epithelial cell line (15)). 1. While adherent on glass coverslips, cells can be incubated for 30 min to 1 h at 37◦ C with EDAC at 0.5% in HBSS−/− to cross-link the lipid mediator of interest to carboxyl groups to amines in adjacent proteins (Fig. 10.3) without affecting

170

Bandeira-Melo, Weller, and Bozza

Fig. 10.3. EicosaCell for PGE2 immuno-localization within adherent CACO-2 cells. The largest panel shows fluorescent microscopy of CACO-2 cells labeled for newly formed PGE2 (red staining). Bottom three images panel showed immunofluorescent PGE2 (red staining), BODIPY-associated lipid bodies (green staining) and a merged image showing co-localization of PGE2 in lipid bodies (yellow staining). Insert panel showed lack of PGE2 immuno-labeling within lipidbody-enriched CACO-2 cells, which were treated with indomethacin (4 mg/kg) 1 h before EDAC. Briefly, CACO-2 cells were fixed and permeabilized during 1 h at 37◦ C with EDAC (0.5% in HBSS–/– ). Then, cells were washed with HBSS and blocked with 2% donkey serum for 15 min before incubation with anti-PGE2 monoclonal antibody (Cayman Chemicals) for 45 min. Cells were washed with HBSS and incubated with fluorescent secondary antibody Cy3-conjugated affiniPure F(ab’) fragment donkey anti-mouse and BODIPY 493/503 (Molecular Probes, CA) for 45 min.

EicosaCell

171

cell morphology (refer to Note 5). Alternatively, cells grown in Lab-Tek chambers can be used. 2. Gently wash cells 2–3 times in HBSS−/− . 3. Incubate cells with the primary antibody to the eicosanoid of interest for 1 h at room temperature. The non-immune serum from the animal where the secondary antibody was produced may be added to the primary antibody so as to decrease unspecific labeling. 4. Gently wash 2–3 times in HBSS−/− . 5. Incubate with the fluorescent-labeled secondary antibody for 1 h at room temperature. 6. At the end of the staining procedure, cells should be gently washed with HBSS−/− , at least 3 times for 5 min each. 7. Cell-containing coverslips should be carefully glued to the slide and mounted using an aqueous mounting medium, preferentially with anti-fading. 8. Analysis is performed on phase contrast to observe cell morphology and fluorescence microscope or confocal scanning laser microscope to identify the eicosanoid labeling. For example, analyses and image acquisition can be obtained using an Olympus BX-FLA fluorescence microscope equipped with a Plan Apo 100 × 1.4 Ph3 objective (Olympus) and CoolSNAP-Pro CF digital camera in conR junction with Image Pro Plus software (Media Cybernetics) (see Note 3 for details). 9. As for cytospun cells, the specificity of the eicosanoid immuno-labeling using EicosaCell system should be ascertained by including mandatory controls listed in Note 4. 3.3. EicosaCell with Cells Embedded in a Gel Matrix

In contrast to analyzing cytospun cells which do not preserve in situ morphology, cells embedded in an agarose matrix, that are kept in a hydrated system with a substrate where they can crawl, display tissue-like cell morphology exhibiting polarization and other characteristics of activated leukocyte, for instance. Therefore, by immuno-localizing eicosanoids at its formation sites within agarose-embedded cells (as schematically illustrated in Fig. 10.1b), generated products may be microscopically localized at cell structures assembled during stimulation and preserved in cells that are not cytospun into slides (10, 16). 1. To prepare the agarose matrix, 2.5% agarose (24 C gelling point) (Promega) in sterile distilled H2 O, is melted at 70◦ C; and while liquid at 37◦ C, 9 volumes of agarose are mixed with 1 volume of 10× concentrated RPMI 1640 medium.

172

Bandeira-Melo, Weller, and Bozza

2. One volume of this medium-supplemented agarose is mixed with one volume of RPMI 1640 medium containing 2% fatty acid free-albumin at 37◦ C and with three volumes of the studied cell, exemplified here as human eosinophils (Fig. 10.1b), which should be at 15 × 106 cells/mL in RPMI 1640 medium containing 1% fatty acid free-human albumin. 3. Stimuli are then added in 0.1 volumes to agarose/eosinophil mixtures. As schematically illustrated in Fig. 10.1b, immediately thereafter, 20 µL samples are gently spread onto microscope slides and covered with perfusion chamber (CoverWellTM ). 4. Each slide is overlaid with RPMI 1640 medium containing 1% albumin and an identical concentration of the stimulus present in the agarose/eosinophil mixture. 5. Slides can be incubated (37◦ C, humidified 5% CO2 ) for varying periods of time. 6. Overlying medium should be removed and replaced with RPMI 1640, 1% albumin medium, that may contain or not 0.1 µM calcium ionophore (A23187) and incubated for extra 15 min (37◦ C; 5% CO2 ). 7. Stimulations are stopped by removing chambers and adding EDAC. Fixation and permeabilization of cells with proper immobilization of newly formed eicosanoids at its intracellular sites of synthesis are achieved by immersing the slides containing stimulated cells in 0.5% EDAC (in HBSS−/− ) for 30 min. 8. After three washes (5 min each) with HBSS−/− , the fluorochrome-labeled anti-eicosanoid, for instance Alexa488-labeled rat anti-cysteinyl leukotriene (LT) detection mAb (Sigma) (AlexaTM 488 protein labeling using a kit from Molecular Probes) should be added (400 µL of 10 µg/mL) for 1 h. 1. Slides need to be extensively washed with HBSS−/− , at least 3 times for 5 min each. 2. Aqueous mounting medium should be applied to each slide before coverslip attachment. 3. Slides can be viewed by both phase-contrast and fluorescence microscopy as detailed above – Section 3.1 (Step 11). 4. Mandatory control conditions, as listed in Note 4, should be always included as for cytospun and adherent cells to ascertain the specificity of the eicosanoid immuno-labeling using EicosaCell system.

EicosaCell

173

3.4. Double-Labeling Procedures to Identify EicosanoidSynthesizing Intracellular Sites 3.4.1. Nuclear Localization

To better visualize perinuclear eicosanoid synthesis by EicosaCell, a double labeling with DAPI is advised. 1. After EDAC and antibody incubation steps, EicosaCell slides preparations should be extensively washed in HBSS−/− and then incubated with DAPI (DAPI working solution 100 ng/mL or 300 nM, see Section 2.2, Step 1) for 5 min before aqueous mounting medium application. 2. The morphology of the cells’ nuclei is observed using a fluorescence microscope at excitation wavelength 350 nm.

3.4.2. Phagosomal Localization

As performed by Balestrieri and coworkers (17), phagosome involvement in eicosanoid synthesis can be ascertained by colocalizing the phagosomal protein marker LAMP-1 in EicosaCell preparations. 1. After incubation at 37◦ C with EDAC, cells should be washed with HBSS−/− , cytospun onto glass slides and incubated with the anti-eicosanoid of interest and the primary antibody against LAMP-1 (2.5 µg/mL) in blocking buffer (5% normal donkey serum) for 2 h at room temperature. 2. Negative control cells are instead incubated for 2 h with appropriate IgG. 3. After 2 h, the cells are washed extensively with HBSS−/− and incubated for 1 h at room temperature with fluorescentlabeled secondary antibody to detect the primary antibody against the eicosanoid of interest and with fluorescentlabeled secondary antibody to detect the primary antibody against LAMP-1(1:200). 4. The cells should be washed five times with HBSS−/− and then mounted in aqueous mounting medium.

3.4.3. Lipid Body Localization

R 3.4.3.1. BODIPY 493/503 Lipid Body-Labeling

To investigate lipid body role in eicosanoid synthesis by EicosaCell assay, two double-labeling strategies can be employed: BODIPY or anti-ADRP immunostaining (for further information on lipid body labeling refer to Chapter 9). Both approaches can be used for adherent, suspension or agarose-embedded cells. R 1. To employ BODIPY 493/503 strategy, incubate EicosaCell preparations (coverslips or slides) with 1 µm BODIPY (working solution) simultaneously to secondary antibody incubation for 45–60 min at 37◦ C (water bath).

174

Bandeira-Melo, Weller, and Bozza

2. To remove free BODIPY after incubation, EicosaCell preparations should be washed at least twice in HBSS−/− before aqueous mounting medium application and coverslip attachment to slides. 3.4.3.2. ADRP Lipid Body-Labeling

Alternatively, to visualize lipid bodies, anti-ADRP immunolabeling may be performed as detailed in Chapter 9. 1. Add anti-ADRP antibodies together with the antieicosanoid antibody of interest for 1 h at room temperature. (a) For human cells incubate with mouse anti-human ADRP at dilution of 1:20 (2.5 µg/mL, final concentration); (b) For mouse, rat, human or bovine cells, incubate with guinea pig anti-human ADRP polyclonal antibody at dilution of 1:300 (final dilution). 2. Wash three times in HBSS−/− or PBS. 3. Incubate with fluorescent-labeled secondary antibody for 1 h room temperature (together with the secondary antibody for the EicosaCell labeling. 4. Wash three times in HBSS−/− or PBS. 5. Mount in mounting medium for fluorescence microscopy. Common problems and non-obvious features found in immunofluorescent-detection of eicosanoids in EDAC preparations by EicosaCell with their possible explanations and potential solutions are described in Note 6.

4. Notes 1. EDAC working solution should be prepared fresh, kept protected from light and discarded after each experiment. 2. Incubation of cells with EDAC can be carried out on either cell in suspension or with the cells already cytospun onto slides by dropping EDAC on top of the cells. Even though the latter method is less costly, some differences in preservation of cell morphology, cell permeabilization and eicosanoid detection may occur and should be analyzed with care. 3. Analysis of EicosaCell preparations should be performed as soon as slides are mounted, inasmuch as immunofluorescent labeling is usually not stable for a long period bleaching after a certain time. Even though freezing may preserve fluorescence overnight, EDAC-treated cells may display altered cell appearance after freezing–thawing cycle.

EicosaCell

175

4. The specificity of the eicosanoid immuno-labeling using EicosaCell system should be always ascertained by including some mandatory control conditions: (i) non-stimulated EDAC-treated cells labeled with the proper anti-eicosanoid antibody; (ii) the incubation (1 h min before EDAC) with the eicosanoid synthesis inhibitors, such as cPLA2 -α inhibitor (e.g. pyrrolidine-2; 1 µM), COX inhibitor (e.g. indomethacin; 1 µg/mL), FLAP inhibitor (e.g. MK886; 50 µg/animal or 10 µM for in vitro incubations) or 5-LO inhibitor (e.g. zileuton; 50 µg/animal or 10 µM for in vitro incubations) to avoid the eicosanoid synthesis and (iii) the use of an irrelevant antibody control. Optionally, other suitable controls to check specificity and performance of EicosaCell are (i) to use, instead of EDAC, paraformaldehyde, which will not immobilize the newly synthesized eicosanoid within cells; (ii) to, in parallel, carry out the EicosaCell in a different cell type that lacks the ability to synthesize the targeted eicosanoid (for instance, to use neutrophils to check specificity of LTC4 immunodetection by EicosaCell) or (iii) to analyze mixed populations of responsive plus unresponsive cells to a specific stimulus, so you can reassure that the targeted eicosanoid is specifically detected only within stimulated cells. 5. While adherent CACO 2 cells can be incubated with EDAC for 1 h, IEC-6 cells can be incubated for at most 30 min (at same concentration; 0.5% in HBSS−/− ) to retain reasonable cell morphology and PGE2 immuno-detection at synthesizing compartments (refer to the original articles (14, 15) for details of blocking and staining conditions with anti-PGE2 monoclonal antibody (Cayman Chemicals) and proper secondary antibodies. 6. Common problems and non-obvious features found in immunofluorescent-detection of eicosanoids in EDAC preparations by EicosaCell: Lack of eicosanoid detection: When few or no eicosanoid specific immunostaining is observed (but expected), the problem usually lies in the improper fixation (e.g. EDAC-driven cross-linking) of targeted eicosanoid at its sites of synthesis. Thus, the newly formed eicosanoid would be washed-out from the EicosaCell preparation turning detection impossible. Resolution of this problem is normally achieved by adjusting (slight increase) concentration and/or time of incubation of EDAC. Alternatively, the lack of immunodetection of newly formed eicosanoids can be due to inefficient stimulation; a positive control with a known agonist should be always included in experiments.

176

Bandeira-Melo, Weller, and Bozza

Eicosanoid detection within non-stimulated cells: Eicosanoids are lipid mediators non-storable in the cell and newly formed upon stimulation, therefore non-stimulated cells should not show any immunostaining for the targeted eicosanoid. Thus, non-stimulated cells should always be included as an important negative control. However, cell activation during procedures including cell incubation at 37◦ C or cell fixation/permeabilization with EDAC can lead to spontaneous, stimulus-independent eicosanoid synthesis. Throughout cell preparation, care is needed to ensure that cells are not mechanically, chemically or immunologically stimulated. Unexpected eicosanoid detection within EicosaCell preparations can also result from non-specific detection (discussed below). Non-specific detection: Fluorescent detection antibodies may non-specifically bind to other lipids found within cells or bind to other cellular structures. The crosslinking properties of EDAC may favor the tendency for cells to be sticky; therefore antibodies could interact through lowaffinity non-antigen binding site. To investigate non-specific binding in EicosaCell preparations, a proper control using host/isotype-matched irrelevant antibodies, must be always included. An additional mandatory control that needs to be always included in the experimental design to rule out non-specific immuno-staining is the condition with a synthesis inhibitor of the targeted eicosanoid. Synthesis inhibitortreated controls should show no immune-labeling confirming specific detection of targeted eicosanoid. If non-specific staining is too high (>10% positive), there are several possible remedies. The detecting antibody may be diluted further, or a different one from a different host may be tried. Also, it is possible to try an adsorbing reagent that effectively blocks out non-specific sites, such as a normal serum (same host of the detecting antibody). Non-specific fluorescence can also be detected when the solution of detecting antibody contains a high degree of aggregated antibody; therefore, it is important to centrifuge the detecting antibody before adding to cell preparations. Poor preservation of cell morphology: During EDAC incubation step of EicosaCell assays, cell appearance may change from unimportant to severe modification of typical cell morphology. This undesirable effect of EDAC on cells can be avoided by adjusting both EDAC concentration and incubation time. Losing cell adherence with EDAC: Similar to unwanted EDAC effect on cell morphology, the ability of cells to stay

EicosaCell

177

adhered to coverslips or other substrates can be affected by EDAC incubation. Again, previous careful setting of EDAC incubation step is obligatory and should be adjusted for each cell type. Lost of cell integrity: Eicosanoid localization within cells by EicosaAssay may be tricky sometimes since some cell types are destroyed during EDAC-driven crosslinking/permeabilization step. For instance, even though lipid bodies of human neutrophils and basophils are sites of 5-LO localization (10, 18), EicosaCell assays with agaroseembedded neutrophils and basophils were not feasible since these cells did not endure to EDAC-driven fixation/permeabilization process, which precedes eicosanoid immuno-detection by EicosaCell, indicating that the combination of gel matrix with EDAC step may be useful to study only a small group of tough cells, like eosinophils. Compartmentalization studies of eicosanoid synthesis within more fragile cells like neutrophils and basophils, however, can be carried out with EicosaCell system in non-gel solutions.

5. Applications Over the past decade, intracellular compartmentalization of eicosanoid-synthetic machinery has emerged as a key component of the regulation of eicosanoid synthesis (reviewed in (4–6, 19). However, the direct evaluation of specific subcellular locales of eicosanoid synthesis has been elusive, as those lipid mediators are newly formed, not stored and often rapidly released upon cell stimulation. Thus, in the majority of studies, intracellular sites of eicosanoid synthesis have been inferred based on the identification of eicosanoid-forming enzymes localization. The EicosaCell technique described herein enables to directly pinpoint the intracellular locales of eicosanoid synthesis by detecting the newly formed lipids and has been successfully able to confirm the dynamic aspect involved in the localization of eicosanoid synthesis, providing direct evidence of compartmentalization within perinuclear envelope (10, 14, 15, 20), phagosomes (17) or lipid bodies in accord to cell type and stimulatory condition (10–12, 15, 21); (Figs. 10.4 and 10.5). So far, the EicosaCell assay has been used to identify the production of leukotriene C4 (LTC4 ) (10, 13, 17, 22), leukotriene B4 (LTB4 ) (12, 23), prostaglandin E2 (PGE2 ) (11, 14, 15)

178

Bandeira-Melo, Weller, and Bozza

Fig. 10.4. EicosaCell for LTC4 immuno-labeling within gel-immersed human eosinophils. Fluorescent microscopy of agarose-embedded eosinophils, fixed with EDAC (0.5% in HBSS–/– ) and stained with Alexa488-labeled anti-cysteinyl LT mAb. To facilitate intracellular localization of anti-LTC4 immunoreactive sites (green staining) within representative eosinophils, blue and white dotted lines were drawn to delineate, respectively, the nuclear and cellular perimeters. As indicated, A23187-, eotaxin- or eotaxin plus A23187-stimulated eosinophils display fluorescent LTC4 immunostaining.

EicosaCell

179

Cellular compartments of eicosanoid synthesis identified by EicosaCell

1

2

3

1. Perinuclear Membrane Bandeira-Melo et al., 2001 Tedla et al., 2003 Accioly et al., 2008 Moreira et al., 2009 2. Lipid Bodies Bandeira-Melo et al., 2001 Bandeira-Melo et al., 2002 Vieira-de-Abreu et al., 2005 Mesquita-Santos et al., 2006 D’Avila et al., 2006 Pacheco et al., 2007 Accioly et al., 2008 Moreira et al., 2009 3.Phagosomes Balestrieri et al., 2006

eicosanoid imunnofluorescence

Fig. 10.5. Schematic summary of EiocosaCell-derived reports identifying three distinct intracellular compartments of eicosanoid synthesis: the nuclear envelope, cytoplasmic lipid bodies and zymozan-driven phagosomes.

and prostaglandin D2 (PGD2 ) (unpublished observations) in different cell types and under different stimulatory conditions. Moreover, it could in principle be adapted to intracellular detection of other lipid mediators as long as specific antibodies are available. Of note, the EicosaCell Assay has high sensitivity, enabling the detection of low levels of intracellular generated eicosanoids even when extracellular released eicosanoids could not be detected by conventional eicosanoid enzyme immune assay (10, 15). Indeed, it has been shown that besides paracrine/autocrine activities, eicosanoids may display intracrine functions (22, 24). For instance, by employing EicosaCell technique, it has been uncovered that a lipid-body-derived LTC4 have intracellular functions in controlling cytokine release from eosinophils (25). Therefore, by identifying compartmentalized levels of eicosanoids, besides providing new insights of regulation of eicosanoid biosynthesis, EicosaCell assay may contribute to identification of likely intracellular functions of newly synthesized eicosanoids.

◭ Fig. 10.4. (continued) While, eosinophils stimulated with A23187 (0.1 µM) for 15 min exhibited exclusively perinuclear (stars) immunoreactive LTC4 , eotaxin (100 ng/mL)stimulated eosinophils showed punctate cytoplasmic lipid body-comprised LTC4 (arrows). Differently, eosinophils stimulated with eotaxin for 1 h and activated for extra 15 min with A23187 exhibit perinuclear (stars) and punctate cytoplasmic (arrows) immunoreactive LTC4 .

180

Bandeira-Melo, Weller, and Bozza

Acknowledgments The work of authors is supported by PRONEX-MCT, Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq, Brazil), PAPES-FIOCRUZ, Fundação de Amparo à Pesquisa do Rio de Janeiro (FAPERJ, Brazil) and NIH grants (AI022571, AI020241, AI051645). Authors are indebted with Dr. Heloisa D’Avila and Dr. Patricia Pacheco for the contributions to figures used in the manuscript.

References 1. Yaqoob, P. (2003) Fatty acids as gatekeepers of immune cell regulation. Trends Immunol 24, 639–645. 2. Wymann, M. P., Schneiter, R. (2008) Lipid signaling in disease. Nat Rev Mol Cell Biol 9, 162–176. 3. Smith, W. L., DeWitt, D. L., Garavito, R. M. (2000) Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem 69, 145–182. 4. Bozza, P. T., Magalhaes, K., Weller, P. F. (2009) Leukocyte lipid bodies – biogenesis and functions in inflammation. Biochim Biophys Acta 1791, 540–51. 5. Mandal, A. K., Skoch, J., Bacskai, B. J., Hyman, B. T., Christmas, P., Miller, D., Yamin, T. T., Xu, S., Wisniewski, D., Evans, J. F., Soberman, R. J. (2004) The membrane organization of leukotriene synthesis. Proc Natl Acad Sci USA 101, 6587–6592. 6. Peters-Golden, M., Brock, T. G. (2001) Intracellular compartmentalization of leukotriene synthesis: unexpected nuclear secrets. FEBS Lett 487, 323–326. 7 Diaz, B. L., Arm, J. P. (2003) Phospholipase A(2). Prostaglandins leukotrienes essent. Fatty Acids 69, 87–97. 8. Bandeira-Melo, C., Weller, P. F. (2003) Eosinophils and cysteinyl leukotrienes. Prostaglandins Leukot Essent Fatty Acids 69, 135–143. 9. Liu, L. X., Buhlmann, J. E., Weller, P. F. (1992) Release of prostaglandin E2 by microfilariae of Wuchereria bancrofti and Brugia malayi. Am J Trop Med Hyg 46, 520–523. 10. Bandeira-Melo, C., Phoofolo, M., Weller, P. F. (2001) Extranuclear lipid bodies, elicited by CCR3-mediated signaling pathways, are

11.

12.

13.

14.

15.

the sites of chemokine-enhanced leukotriene C4 production in eosinophils and basophils. J Biol Chem 276, 22779–22787. D’Avila, H., Melo, R. C. N., Parreira, G. G., Werneck-Barroso, E., Castro-Faria-Neto, H. C., Bozza, P. T. (2006) Mycobacterium bovis bacillus Calmette-Guerin induces TLR2mediated formation of lipid bodies: intracellular domains for eicosanoid synthesis in vivo. J Immunol 176, 3087–3097. Pacheco, P., Vieira-de-Abreu, A., Gomes, R. N., Barbosa-Lima, G., Wermelinger, L. B., Maya-Monteiro, C. M., Silva, A. R., Bozza, M. T., Castro-Faria-Neto, H. C., BandeiraMelo, C., Bozza, P. T. (2007) Monocyte chemoattractant protein-1/CC chemokine ligand 2 controls microtubule-driven biogenesis and leukotriene B4-synthesizing function of macrophage lipid bodies elicited by innate immune response. J Immunol 179, 8500–8508. Vieira-de-Abreu, A., Assis, E. F., Gomes, G. S., Castro-Faria-Neto, H. C., Weller, P. F., Bandeira-Melo, C., Bozza, P. T. (2005) Allergic challenge-elicited lipid bodies compartmentalize in vivo leukotriene C4 synthesis within eosinophils. Am J Respir Cell Mol 33, 254–261. Accioly, M. T., Pacheco, P., Maya-Monteiro, C. M., Carrossini, N., Robbs, B. K., Oliveira, S. S., Kaufmann, C., Morgado-Diaz, J. A., Bozza, P. T., Viola, J. P. (2008) Lipid bodies are reservoirs of cyclooxygenase-2 and sites of prostaglandin-E2 synthesis in colon cancer cells. Cancer Res 68, 1732–1740. Moreira, L. S., Piva, B., Gentile, L. B., Mesquita-Santos, F. P., D’Avila, H., MayaMonteiro, C. M., Bozza, P. T., BandeiraMelo, C., Diaz, B. L. (2009) Cytosolic phos-

EicosaCell

16.

17.

18.

19.

20.

pholipase A2-driven PGE2 synthesis within unsaturated fatty acids-induced lipid bodies of epithelial cells. Biochim Biophys Acta 1791, 156–165. Bandeira-Melo, C., Sugiyama, K., Woods, L. J., Phoofolo, M., Center, D. M., Cruikshank, W. W., Weller, P. F. (2002) IL-16 promotes leukotriene C(4) and IL-4 release from human eosinophils via CD4- and autocrine CCR3-chemokine-mediated signaling. J Immunol 168, 4756–4763. Balestrieri, B., Hsu, V. W., Gilbert, H., Leslie, C. C., Han, W. K., Bonventre, J. V., Arm, J. P. (2006) Group V secretory phospholipase A2 translocates to the phagosome after zymosan stimulation of mouse peritoneal macrophages and regulates phagocytosis. J Biol Chem 281, 6691–6698. Pacheco, P., Bozza, F. A., Gomes, R. N., Bozza, M., Weller, P. F., CastroFaria-Neto, H. C., Bozza, P. T. (2002) Lipopolysaccharide-induced leukocyte lipid body formation in vivo: innate immunity elicited intracellular loci involved in eicosanoid metabolism. J Immunol 169, 6498–6506. Bandeira-Melo, C., Bozza, P. T., Weller, P. F. (2002) The cellular biology of eosinophil eicosanoid formation and function. J Allergy Clin Immunol 109, 393–400. Tedla, N., Bandeira-Melo, C., Tassinari, P., Sloane, D. E., Samplaski, M., Cosman, D., Borges, L., Weller, P. F., Arm, J. P. (2003) Activation of human eosinophils through leukocyte immunoglobulin-like

21.

22.

23.

24.

25.

181

receptor 7. Proc Natl Acad Sci USA 100, 1174–1179. Mesquita-Santos, F. P., Vieira-de-Abreu, A., Calheiros, A. S., Figueiredo, I. H., Castro-Faria-Neto, H. C., Weller, P. F., Bozza, P. T., Diaz, B. L., Bandeira-Melo, C. (2006) Cutting edge: prostaglandin D2 enhances leukotriene C4 synthesis by eosinophils during allergic inflammation: synergistic in vivo role of endogenous eotaxin. J Immunol 176, 1326–1330. Devchand, P. R., Keller, H., Peters, J. M., Vazquez, M., Gonzalez, F. J., Wahli, W. (1996) The PPARalpha-leukotriene B4 pathway to inflammation control. Nature 384, 39–43. Silva, A. R., Pacheco, P., Vieira-De-Abreu, A., Dalegria, B., Bandeira-Melo, C., CastroFaria-Neto, H. C., Bozza, P. T. (2009) Lipid bodies in oxidized LDL-induced foam cells are leukotriene-synthesizing organelles: a MCP-1/CCL2 regulated phenomenon. Biochim Biophys Acta 1791, 1066–75. Kliewer, S. A., Lenhard, J. M., Willson, T. M., Patel, I., Morris, D. C., Lehmann, J. M. (1995) A prostaglandin J2 metabolite binds peroxisome proliferator-activated receptor gamma and promotes adipocyte differentiation. Cell 83, 813–819. Bandeira-Melo, C., Woods, L. J., Phoofolo, M., Weller, P. F. (2002) Intracrine cysteinyl leukotriene receptor-mediated signaling of eosinophil vesicular transportmediated interleukin-4 secretion. J Exp Med 196, 841–850.

Chapter 11 Nestin-Driven Green Fluorescent Protein as an Imaging Marker for Nascent Blood Vessels in Mouse Models of Cancer Robert M. Hoffman Abstract A transgenic mouse, in which the regulatory elements of the stem cell marker, nestin drive green fluorescent protein (ND-GFP), expresses GFP in nascent blood vessels. Red fluorescent protein (RFP)expressing tumors transplanted to nestin-GFP mice enable specific visualization of nascent vessels in the growing tumors. The ND-GFP mouse was also utilized to develop a rapid in vivo/ex vivo fluorescent R , a surgical sponge derived from pigskin, which was rapidly angiogenesis assay by implanting Gelfoam vascularized by fluorescent nascent blood vessels. Angiogenesis could be imaged and quantified when R . These fluorescent models stimulated or inhibited by specific compounds in both tumors and Gelfoam can be used to study the early events of angiogenesis and to quantitatively determine efficacy of antiangiogenesis compounds. Key words: Green fluorescent protein, red fluorescent protein, nude mouse, human tumors, color-coded, imaging, nascent blood vessels.

1. Introduction 1.1. Previous Models Used to Determine Angiogenesis

The discovery and evaluation of antiangiogenic substances initially relied on methods such as the chorioallantoic membrane assay (1, 2), the monkey iris neovascularization model (3), the disk angiogenesis assay (4) and various models that use the cornea to assess blood vessel growth (5–10). Although they are important for understanding mechanisms of blood vessel induction, these models do not represent tumor angiogenesis.

H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_11, © Springer Science+Business Media, LLC 2011

183

184

Hoffman

Subcutaneous tumor xenograft mouse models have been developed to study tumor angiogenesis, but these require cumbersome pathological examination procedures, such as histology and immunohistochemistry. Measurements require animal sacrifice and therefore preclude ongoing angiogenesis studies in individual, live, tumor-bearing animals. Moreover, subcutaneous tumor xenografts are not representative models of metastatic cancer. Tumors transplanted in the cornea of rodents (11–13) and rodent skin-fold window chambers have also been used for angiogenesis studies (14–20). The cornea and skin-fold chamber models provide a means for studying tumor angiogenesis in living animals. However, quantification requires specialized procedures and the sites do not represent natural environments for tumor growth. The cornea and skin-fold window chamber tumor models do not allow metastasis to occur, which may involve mechanisms of angiogenesis (21) that are qualitatively different from those occurring in ectopic models that do not metastasize. We describe here the clinically-relevant imageable mouse models of cancer to visualize and quantify tumor angiogenesis and efficacy of inhibitors. 1.2. Fluorescent Proteins to Image Angiogenesis

For in vivo imaging, a strong signal and high resolution are necessary. The green fluorescent protein (GFP) gene, cloned from the bioluminescent jellyfish Aequorea victoria (22), was chosen to satisfy these conditions because it has great potential for use as a cellular marker (23, 24). Green fluorescent protein cDNA encodes a 283-amino acid monomeric polypeptide with Mr = 27,000 (25, 26) that requires no other A. victoria proteins, substrates or cofactors to fluoresce (27). Gain-of-function bright mutants expressing the GFP gene have been generated by various techniques (28–30) and have been humanized for high expression and signal (31). Red fluorescent proteins (RFP) from the Discosoma coral have similar features as well as the advantage of longer wavelength emission (32–34). Our laboratory has pioneered the use of GFP for in vivo imaging (35) including non-invasive whole-body imaging (36, 37). The Nobel Prize in 2008 was awarded for the discovery and first practical uses of GFP. Fluorescent proteins have been shown by our laboratory to be very useful for imaging tumor angiogenesis. We have developed unique mouse models to image tumor angiogenesis with fluorescent proteins, which are described in this review.

1.3. Orthotopic Tumor Models Expressing Fluorescent Proteins to Visualize Tumor Angiogenesis

For realistic and real-time imageable tumor angiogenesis models, we have developed surgical orthotopic implantation (SOI) metastatic models of human cancer (38). These models place tumors in natural microenvironments and replicate clinical tumor behavior more closely than do ectopic implantation models (38).

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

185

The orthotopically-growing tumors, in contrast to most other models, give rise to spontaneous metastases that resemble, both in target tissues and in frequency of occurrence, the clinical behavior of the original human tumor (38). Tumors implanted in the orthotopic model have been transduced and selected to strongly express GFP or RFP in vivo (37). Orthotopically-implanted GFPor RFP-labeled tumors enable the visualization of the role of angiogenesis in metastasis. As Li et al. (18) point out, angiogenesis initiation in metastatic tumors may be very different from that of primary tumors and require different interventions. Moreover, the extreme detection sensitivity afforded by the strong GFP or RFP fluorescence allows imaging of very early events in blood vessel induction. GFP or RFP expression in primary tumors and in their metastases in the mouse models can be detected by an intense fluorescence seen by intravital or by whole-body imaging. The nonluminous angiogenic blood vessels appear in contrast as sharply defined dark networks against this bright background. The highimage resolution permits quantitative measurements of total vessel length. These genetically fluorescent tumor models thereby allow quantitative optical imaging of angiogenesis in vivo. Tumor growth, vascularization and metastasis could be followed in real time (39). 1.4. Intravital Images of Angiogenesis of Orthotopic Pancreas Cancer

The clarity of angiogenic blood vessel imaging was initially illustrated by intravital examination of the orthotopic growth of the BxPC3-GFP pancreatic tumor. The non-luminous blood vessels were clearly visible in contrast against the GFP fluorescence of the primary tumor. Angiogenesis associated with metastatic growths was also readily imaged by intravital examination (39) (see Note 1). Because angiogenesis could be measured without animal sacrifice, it was possible to determine a time course for individual animals. Sequential intravital images of angiogenesis for the PC-3 human prostate tumor expressing GFP and growing orthotopically in a single nude mouse were acquired. The tumor-associated blood vessels were clearly visible by day 7 and continued to increase at least until day 20 (39).

1.5. Whole-Body Imaging of Angiogenesis in Orthotopic Breast Cancer

We have demonstrated whole-body images and quantitation of the time course of angiogenesis of the MDA-MB-435-GFP human breast cancer growing orthotopically in the mammary fat pad in a nude mouse. The development of the tumor and its angiogenesis could be imaged in a completely noninvasive manner (39). The mouse mammary fat pad is the orthotopic environment for the implanted MDA-MB-435-GFP breast cancer and allows noninvasive, whole-body imaging of tumor angiogenesis. The quantitative angiogenesis data show that microvessel density

186

Hoffman

increased over 20 weeks (39). Similar results were obtained by noninvasive imaging of angiogenesis of the Lewis lung cancerGFP growing in the footpad of a nude mouse (39). 1.5.1. Non-invasive Imaging of Tumor Blood Flow

To noninvasively image cancer cell/stromal cell interaction in the tumor microenvironment and drug response at the cellular level in live animals in real time, we developed a new imageable threecolor animal model. The model consists of GFP-expressing mice transplanted with dual-color cancer cells labeled with GFP in the nucleus and RFP in the cytoplasm. The Olympus IV100 Laser Scanning Microscope, with ultra-narrow microscope objectives (‘stick objectives’), was used for three-color whole-body imaging of the two-color cancer cells interacting with the GFP-expressing stromal cells. In this model, drug response of both cancer and stromal cells in the intact live animal is also imaged in real time. Various in vivo phenomena of tumor-host interaction and cellular dynamics were non-invasively imaged, including tumor vasculature and tumor blood flow (40).

1.6. Skin Flaps Enable Ultra-high Resolution External Imaging of Tumor Angiogenesis

Opening a reversible skin flap in the light path markedly reduced signal attenuation, increasing detection sensitivity many-fold. The observable depth of tissue is thereby greatly increased (41). The brilliance of the tumor GFP fluorescence, facilitated by the lucidity of the skin-flap window, allowed imaging of the induced microvessels by their dark contrast against a bright background. The orthotopically-growing BxPC3-GFP human pancreatic tumor was externally visualized under fluorescence microscopy to be surrounded by its microvessels visible by their dark contrast (41).

1.7. Imaging of Nascent Angiogenesis Using Nestin-Driven GFP Transgenic Mice

We initially reported that in mice in which the gene for the stem cell marker, nestin drives GFP (ND-GFP) (42), that ND-GFP also labels developing skin blood vessels. The ND-GFP labeled vessels appear to originate from hair follicles and form a folliclelinking network. This was seen most clearly by transplanting NDGFP-labeled vibrissa (whisker) hair follicles to unlabeled nude mice. New vessels grew from the transplanted follicle and these vessels increased when the local recipient skin was wounded. The ND-GFP-expressing structures are blood vessels, because they display the characteristic endothelial-cell-specific markers CD31 and von Willebrand factor. This model displays very early events in angiogenesis and can serve for rapid antiangiogenesis drug screening (43).

1.7.1. Dual-Color Imaging of Tumor Angiogenesis

We visualized tumor angiogenesis by dual-color fluorescence imaging in ND-GFP transgenic mice after transplantation of the murine melanoma cell line B16F10 expressing RFP. NDGFP was highly expressed in proliferating endothelial cells and nascent blood vessels in the growing tumor (Fig. 11.1). Results

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

187

ND-GFP-expressing blood vessels growing into tumor mass 100 µm

Fig. 11.1. Visualization of ND-GFP vessels in an RFP-expressing tumor. On day 14 after implantation of RFP-expressing B16 mouse melanoma cells subcutaneously in ND-GFP mice, ND-GFP-expressing blood vessels (white arrows) could be seen in the growing tumor. Nascent ND-GFP blood vessels (white arrows) were forming a network in the growing tumor. Bar, 100 µm (44).

of immunohistochemical staining showed that the blood vesselspecific antigen CD31 was expressed in ND-GFP-expressing nascent blood vessels. ND-GFP expression was diminished in vessels with increased blood flow. Progressive angiogenesis during tumor growth was readily visualized by GFP expression. Doxorubicin inhibited the nascent tumor angiogenesis as well as tumor growth in the ND-GFP mice transplanted with B16F10-RFP (44) (Fig. 11.2) (see Note 2). 1.8. Nestin-Driven GFP Transgenic Nude Mice

The ND-GFP gene was crossed into nude mice on the C57/B6 background to obtain ND-GFP nude mice. ND-GFP was expressed in the brain, spinal cord, pancreas, stomach, esophagus, heart, lung, blood vessels of glomeruli, blood vessels of skeletal muscle, testis, hair follicles and blood vessel network in the skin of ND-GFP nude mice. Human lung cancer, pancreatic cancer, breast and colon cancer cell lines as well as a murine melanoma cell line expressing RFP were implanted orthotopically and an RFP-expressing human fibrosarcoma was implanted s.c. in the ND-GFP nude mice. These tumors grew extensively in the ND-GFP mice. ND-GFP was highly expressed in proliferating endothelial cells and nascent blood vessels in the growing tumors, visualized by dual-color fluorescence imaging (Fig.11.3). The ND-GFP transgenic nude mouse model thus enables the visualization of nascent angiogenesis in human and mouse tumor progression (45).

188

Fig. 11.2.

Hoffman

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

1.9. Imaging of Nascent Angiogenesis in Tumors in ND-GFP Nude Mice

189

Angiogenesis in experimental lung and liver metastases of melanoma was imaged in the ND-GFP transgenic mice. The murine melanoma cell line, B16F10 expressing RFP, was injected i.v. in ND-GFP mice. ND-GFP was highly expressed in proliferating nascent blood vessels in tumors that developed in the lung after tail vein injection and in tumors that developed in the liver after portal vein injection of RFP-expressing melanoma cells. Liver metastasis and angiogenesis were imaged intravitally. Doxorubicin significantly decreased metastatic angiogenesis (Fig. 11.3) (46). Dual-color fluorescence imaging visualized tumor angiogenesis in the ND-GFP transgenic nude mice after orthotopic transplantation of the MIA PaCa-2 human pancreatic cancer line expressing RFP. Mice were treated with gemcitabine at 150 mg/kg/dose on days 3, 6, 10 and 13 after tumor implantation. At day 14, mice were sacrificed and mean nascent blood vessel density and tumor volume were calculated and compared to control mice. The density of nascent blood vessels in the tumor was readily quantitated. Gemcitabine significantly decreased the mean nascent blood vessel density in the tumor as well as decreased tumor volume. The dual-color model of the ND-GFP nude mouse orthotopically implanted with RFP-expressing pancreatic tumor cells enabled the simultaneous visualization and quantitation of tumor angiogenesis and tumor volume. These results demonstrated for the first time that gemcitabine is an inhibitor of angiogenesis as well as tumor growth in pancreatic cancer. The results have important implications for the clinical application of gemcitabine in this disease (47). Nascent angiogenesis was imaged in pancreatic cancer liver metastasis in the ND-GFP transgenic nude mice, formed after intra-splenic injection of XPA-1 human pancreatic cancer cells expressing RFP, using color-coded fluorescence imaging. NDGFP was highly expressed in proliferating endothelial cells and nascent blood vessels in the growing liver metastasis. The density of nascent blood vessels in the tumor was readily quantitated. Gemcitabine significantly decreased the mean nascent blood vessel density in the pancreatic liver metastases (48).

◭ Fig. 11.2. (continued) Effect of doxorubicin on tumor angiogenesis. a On day 10 after implantation of tumor cells, the ND-GFP nascent blood vessels (white arrows) were forming a network in the central tumor. b In the marginal area of the tumor, many newly formed nascent ND-GFP blood vessels were growing. The nascent ND-GFP blood vessels (white arrows) had many branches and were connected to each other. c and d The mice were given daily i.p. injections of 5 µg/g of doxorubicin at days 0, 1 and 2 after implantation of tumor cells. C, by day 10 after implantation of tumor cells, the nascent ND-GFP blood vessels were not seen in the central area of the tumor. d In the marginal area of the tumor, ND-GFP blood vessels (white arrows) were growing slightly. e Number of nascent blood vessels per tumor volume in the doxorubicin-treated mice was less than NaCl solution- injected mice (p < 0.05). Bars, 100 µm (44).

190

Hoffman

Fig. 11.3. Fluorescence imaging of tumor angiogenesis in transgenic ND-GFP nude mice. Human HT1080 fibrosarcoma on day 14 after s.c. injection. Dual-color HT1080 cells expressing GFP in the nucleus and RFP in the cytoplasm are polarized towards ND-GFP-expressing blood vessels (white arrows) growing in the tumor mass. Bar, 100 µm (45).

The antiangiogenic efficacy of CPT-11 was evaluated in a human colon tumor growing in ND-GFP nude mice using colorcoded fluorescence imaging. We orthotopically implanted NDGFP nude mice with the human colon cancer cell line HCT-116 expressing RFP. The mice were treated with CPT-11 at 40 mg/kg on days 7, 10 and 14. Tumor angiogenesis was imaged and visualized by color-coded fluorescence imaging on day 17, three days after the last CPT-11 treatment. Tumor volume and the mean nascent blood vessel density were determined and compared to the control mice. The nascent blood vessels were highly fluorescent and their density was determined. ND-GFP nude mice that were administered CPT-11 showed significant reduction in the mean nascent blood vessel density and tumor volume. The results showed that CPT-11 is an effective inhibitor of angiogenesis and provided strong implications for wider clinical application of CPT-11 for colon cancer (49). Angiogenesis of the HT-1080 human fibrosarcoma cell line, expressing RFP, was imaged in the ND-GFP nude mice. Cancer cells were injected into either the muscle or the bone. Nestin was highly expressed in proliferating endothelial cells and nascent blood vessels in the growing tumors, including the surrounding tissues. CD31 co-localized in ND-GFP-expressing nascent

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

191

blood vessels. The density of nascent blood vessels in the tumor was readily quantitated. The mice were given daily i.p. injections of 5 mg/kg doxorubicin after implantation of cancer cells. Doxorubicin significantly decreased the mean nascent blood vessel density in tumors as well as decreased tumor volume. These data suggest targeting angiogenesis of sarcomas as a promising clinical approach (50). 1.10. Rapid In Vivo/Ex Vivo Nascent Angiogenesis Assay

We developed a very convenient imageable in vivo angiogenesis R (Pharmacia & Upjohn assay after transplantation of Gelfoam Company, Kalamazoo, MI, USA) in the ND-GFP mice (51). R Gelfoam is rapidly vascularized with GFP-expressing vessels in the presence of an angiogenesis stimulator. Anti-angiogenesis agents inhibit this process. Thus, this rapid and simple new in vivo assay can rapidly identify angiogenesis stimulators and R inhibitors. Gelfoam was treated with β-fibroblast growth facR tor (bFGF). The treated Gelfoam was then transplanted into the subcutis on both flanks of the ND-GFP transgenic mice. The mice were given daily intraperitoneal (i.p.) injections of doxorubicin or NaCl solution at day 0, 1 and 2 after transplantation R of Gelfoam . Skin flaps were made at day 7 after transplantaR tion of Gelfoam under anesthesia. Angiogenesis was quantified by measuring the length of ND-GFP-expressing nascent blood R vessels in the Gelfoam in the skin flap by in vivo fluorescence microscopy imaging. The vessels on the surface were counted under fluorescence microscopy. Each experimental group conR sisted of five mice. Gelfoam , treated with bFGF, implanted in the ND-GFP mice was rapidly vascularized with ND-GFPexpressing blood vessels. At day 7 after transplantation, the NDGFP-expressing nascent blood vessels were observed forming R a network on the surface of the bFGF-treated Gelfoam in R  the skin flap (Fig. 11.4). Implanted Gelfoam that was not treated with bFGF was not vascularized. The ND-GFP vessels R in the Gelfoam stained positively for CD31, demonstrating the presence of endothelial cells. Day 7 was chosen as an arbitrary time point to measure the GFP vessels in the implanted R R Gelfoam . The Gelfoam can be analyzed at any time point and an optimal time for measurement would depend on the angiogenesis drug being tested. ND-GFP mice that received i.p. injections of doxorubicin (5 µg/g) at day 0, 1 and 2 after R transplantation of Gelfoam , with or without bFGF, had fewer ND-GFP-expressing nascent blood vessels than NaCl-treated mice (57). Future experiments will address the destruction of R preformed vessels in Gelfoam by vascular disrupting agents (VDAs) (52).

192

Hoffman

R with GFP-expressing vessels. ND-GFP mice were given daily intraperiFig. 11.4. Angiogenesis of implanted Gelfoam R with or without β toneal (i.p.) injections of 0.9% NaCl solution at day 0, 1, and 2 after transplantation of Gelfoam R with bFGF, the ND-GFP-expressing nascent fibroblast growth factor (bFGF). a At day 7 after transplantation of Gelfoam R visualized in a skin flap. The ND-GFP-expressing nascent blood vessels formed a network on the surface of Gelfoam R transplanted ND-GFP mice were blood vessels had many branches that were connected to each other. b The Gelfoam treated with 5 µg/g doxorubicin (DOX) at day 0, 1 and 2 after transplantation. Doxorubicin significantly decreased the blood-vessel density in the presence of bFGF at day 7. Scale bar, 500 µm (51).

1.11. A BrainMetastatic Paralyzing Spinal Cord Glioma That Induces Angiogenesis and Neurogenesis

Cancer of the spinal cord is a highly malignant disease that often leads to paralysis and death. To develop an imageable, patientlike model of this disease, U87 human glioma tumor fragments, expressing RFP, were transplanted by surgical orthotopic implantation (SOI) into the spinal cord in non-transgenic nude mice or ND-GFP transgenic nude mice. In the ND-GFP mice, GFP is expressed in nascent blood vessels and neural stem cells. Animals were treated with temozolomide or vehicle control. The intramedullary spinal cord tumor (IMSCT) grew at the primary site, causing hind-limb paralysis and also metastasized to the brain. Temozolomide inhibited tumor size and prevented metastasis as well as prevented paralysis or delayed paralysis. The tumor stimulated both neurogenesis and angiogenesis (53).

1.12. Growth and Metastasis of Spinal Cord Glioma

Four weeks after transplantation, the spinal cord was exposed by laminectomy. The RFP-expressing glioma in the spinal cord was observed by fluorescence imaging. The skull was opened and the brain was excised en bloc. Brain metastases were detected even though they were small. Metastases were found mainly around the brain stem and in leptomeninges at the basilar sulcus (53).

1.13. Efficacy of Chemotherapy on Spinal Cord Growth and Metastasis

The primary tumor size was significantly reduced by temozolomide compared to untreated controls (0.55 ± 1.1 vs. 9.7 ± 4.1 mm2 , p < 0.01, n = 5 for each group). Brain metastases were found only in the control group (60%). Histological analysis of the control group showed aggressive tumor invasion in

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

193

the spinal cord. In contrast, the temozolomide-treated animals showed mostly scar tissue after tumor transplantation (53). 1.14. Efficacy of Chemotherapy on Hind-Limb Paralysis Due to Spinal Cord Glioma

The untreated mice showed progressive paralysis beginning at 14 days after U87 glioma transplantation. The untreated control group developed complete paralysis (BBB score = 0) between 18 and 31 days after tumor transplantation. Some of the temozolomide-treated mice started to show paralysis at approximately 35 days after transplantation and four mice were still not paralyzed at 60 days. The temozolomide-treated mice survived without complete paralysis for at least 45 days. There was a significant delay to onset or else complete inhibition of paralysis in the treated animals (log rank statistic 8.37, p < 0.005) (53).

1.15. Angiogenesis and Neurogenesis of Spinal Cord Glioma

In frozen sections of the normal spinal cord, ND-GFP-expressing cells were mainly seen around the central canal. Ten days after U87-RFP glioma transplantation, ND-GFP-expressing cells appeared stimulated by the tumor and started to surround it. The main stem of the ND-GFP-expressing cells had many small branches and ND-GFP-expressing cells appeared to originate from cells around the central spinal cord canal. In young (6 weeks) mice, more ND-GFP-expressing cells were observed surrounding the tumor than in old (16 weeks) mice. The mean GFP intensity around the tumor in young mice was significantly higher than in old mice (p < 0.05) (52). Neuronal class III-β-tubulin is a marker of neuronal cells. Some of the ND-GFP-expressing cells surrounding the spinal cord glioma also expressed III-β-tubulin, indicating that some of the ND-GFP-expressing cells accumulating around the tumor are of neural origin. Some of the ND-GFP-expressing cells also appeared to be endothelial cells since they expressed CD31. Frozen sections showing the ND-GFP host cells and RFPexpressing U87 cells under fluorescence microscopy were compared with sister sections immunohistochemically stained for CD31. This comparison demonstrated co-localization of NDGFP and CD-31. These results indicate that the ND-GFPexpressing cells surrounding the tumor contained both neural and endothelial types (53). There have been only few reports of human tumors metastasizing to the brain in mouse models (54–57) making the present model very valuable to study brain metastasis. Temozolomide is effective in this model against primary tumor growth, in the spinal cord, paralysis and brain metastasis. Differential labeling of the tumor and host enabled the observation that the primary tumor stimulates both blood vessel and nerve growth. This novel mouse model will enable a deeper understanding of spinal cord cancer and provide a clinically-relevant system for drug discovery and evaluation.

194

Hoffman

2. Materials 2.1. Reagents

1. Restriction enzymes HindIII and NotI 2. RFP cDNA (pDsRed2; Clontech) 3. Plasmid pLNCX2 4. PT67 packaging cells (Clontech); 3T3 cells for viral titering; human and mouse cell lines to be transfected with genes encoding fluorescent proteins 5. Growth medium (normal and selective) appropriate for cell culture, such as DMEM (Invitrogen; Irvine Scientific) 6. Fetal bovine serum (FBS; Gemini Biological Products) 7. Lipofectamine PLUS (Invitrogen) 8. G418 neomycin (Invitrogen) 9. Polysulfonic filter, 4.5 µm 10. Polybrene 11. Trypsin-EDTA (Fisher Scientific) and trypsin 12. Mice expressing GFP (‘GFP mice’; Jackson Laboratories; Japan SLC, Hamamatsu, Japan) 13. Immunocompetent and immunodeficient mice (Charles River; Taconic; Harlan Teklad). Mice are fed an autoclaved laboratory rodent diet (Tecklad LM-485, Western Research Products). 14. Anesthetic reagents: ketamine mixture (10 µL ketamine HCl, 7.6 µL xylazine and 2.4 µL acepromazine maleate, injected s.c.). 15. Nair (Carter-Wallace) 16. Doxorubicin 17. NaCl, 0.9% 18. Optimum cutting temperature blocks 19. Antibody to rat immunoglobulin (anti-rat immunoglobulin) and anti-mouse immunoglobulin horseradish peroxidase detection kits (BD PharMingen, San Diego, CA) 20. Monoclonal anti-CD31 (CBL1337; Chemicon, Temecula, CA) 21. Monoclonal anti-nestin (rat 401; BD PharMingen, San Diego, CA) 22. Monoclonal anti-III-β-tubulin (Covance Research Products, Berkeley, CA) 23. Substrate-chromogen 3,3′ -diaminobenzidine

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

195

24. Doxorubicin (5 µg/g body weight in a 2 mg/mL solution of 0.9% NaCl) 2.2. Equipment

1. Culture dishes, 60 mm (Fisher Scientific); flask, 25 mm; plates, 96-well 2. Humidified incubator at 37◦ C and 5% CO2 3. Cloning cylinders (Bel-Art Products) 4. 27G2 latex-free syringe, 1 mL (Becton Dickinson) 5. 8-0 surgical suture 6. Leica fluorescence stereo microscope, model LZ12, with a mercury 50-W power supply 7. D425/60 bandpass filter and 470 DCXR dichroic mirror 8. D470/40 emission filter and GG475 emission filter (Chroma Technology) 9. C5810 three-chip cooled color charge-coupled device (CCD) camera (Hamamatsu Photonics Systems) or DP70 CCD camera (Olympus) for high-resolution capture (1024 × 724 pixels) 10. Image-Pro Plus 3.1 or 4.0 software (Media Cybernetics) 11. Personal computer (PC; IBM or Fujitsu-Siemens) 12. VCR (Sony, model SLV-R1000) 13. Blue LED flashlight (LDP LLC) 14. Coolpix camera (Nikon) 15. Fluorescent light box with fiberoptic lighting at 470 nm (Lightools Fluorescent Imaging System; Lightools Research) 16. OV100 Small Animal Imaging System (Olympus) with an M20 light source (Olympus Biosystems) and 470-nm excitation light 17. IV100 Laser Scanning Microscope (Olympus) 18. Paint Shop Pro 8 (Corel) and cellR (Olympus Biosystems) 19. Olympus BH 2-RFCA fluorescence microscope equipped with a mercury 100-W lamp power supply 20. Leica CM1850 cryostat

3. Methods Use one of the following options to establish a mouse tumor model of fluorescent protein–expressing tumor cells: i.v. cell injection or SOI.

196

Hoffman

3.1. Cell Injection to Establish an Experimental Metastasis Model

1. Collect fluorescent protein-expressing cancer cells by trypsinization for 3 min at 37◦ C with 0.25% trypsin. 2. Wash cells three times with cold serum-free medium using a tabletop centrifuge at 500g. 3. Resuspend cells in approximately 0.2 mL serum-free medium. 4. Within 30 min of collecting cells, inject 1 × 106 tumor cells in a total volume of 0.2 mL into 6-week-old C57BL/6 GFP mice or nude (nu/nu) GFP mice, or ND-GFP C57BL/6 immunocompetent or nu/nu mice, in the lateral tail vein or subcutaneously using a 1-mL 27G2 latex-free syringe. Cells in suspension may lose viability over time and therefore should be injected as soon as possible. 5. For liver colonization, inject fluorescent protein-expressing cells directly into the portal vein in anesthetized mice (details on inducing anesthesia are presented below).

3.2. Surgical Orthotopic Implantation to Establish a Spontaneous Metastasis Model (IND)

1. Induce anesthesia with a ketamine mixture 2. Use a microscope with magnification of ×6 to ×40 for all procedures of the operation. 3. Isolate fluorescent protein-expressing tumor fragments (1 mm3 ) from subcutaneously growing tumors, formed by injection of RFP-expressing cancer cells, by mincing tumor tissue into 1-mm3 fragments. After proper exposure of the target organ, implant three tumor fragments per mouse. 4. With 8-0 surgical suture, penetrate the tumor fragments and suture fragments onto the target organ. Orthotopic implantation of tumor fragments results in higher spontaneous metastatic rates than injection of a cell suspension. 5. Keep mice in a barrier facility under high-efficiency particulate air filtration.

3.3. Skin-Flap Windows

Tumor cells on the various internal organs are visualized through the body wall through a skin-flap window over the abdomen (41). 1. Animals are anesthetized with a ketamine mixture. 2. An arc-shaped incision is made in the skin and s.c. connective tissue is separated to free the skin flap. 3. The skin flap can be opened repeatedly to image cancer cells on the internal organs through the nearly transparent body wall. The skin is closed with a 6–0 suture. This procedure greatly reduces the scatter of fluorescent photons.

3.4. Tumor Tissue Sampling

1. Obtain tumor tissue biopsies from 3 days to 4 weeks after inoculation of cancer cells. Biopsies of tumor tissue can be

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

197

obtained from anesthetized mice by removal of a small piece of tumor tissue (1 mm3 or less) with a scalpel. Staunch bleeding by pressing the wound with sterile gauze. Alternatively, the mouse can be killed and the tissue can be collected and processed for analysis. 2. Cut fresh tissue into pieces of about 1 mm3 and gently press onto slides for fluorescence microscopy. This procedure is done manually on normal slides. 3. To analyze tumor angiogenesis, digest tissues with trypsinEDTA for 5 min at 37◦ C before examination. 4. After trypsinization, put tissues on pre-cleaned microscope slides and cover with another microscope slide. 3.5. Fluorescence Microscopy

1. Use a fluorescence microscope equipped with a mercury 100-W lamp power supply. 2. To visualize both GFP and RFP fluorescence at the same time, produce excitation light via a D425/60 filter and a 470 DCXR mirror. 3. Collect emitted fluorescence light through a GG475 filter. 4. Capture high-resolution images and store directly on a PC. 5. Process images for contrast and brightness using Image-Pro software or its equivalent.

3.6. Methods for Color-Coded Imaging of Tumor Blood Vessels of Mice 3.6.1. Microscopy

Use one of the following methods for whole-body imaging of mice: microscopy, flashlight imaging, light-box imaging or chamber imaging. 1. Use a fluorescence stereo microscope equipped with a mercury lamp power supply. 2. Produce selective excitation of GFP via a D425/60 filter and 470 DCXR mirror. 3. Collect emitted fluorescence through a long-pass filter (GG475) on a CCD camera. 4. Process images for contrast and brightness with the use of Image-Pro software or its equivalent. 5. Capture high-resolution images directly on a PC or continuously through video output on a high-resolution VCR. 6. For C57BL/6 mice, remove hair with Nair or by shaving before images are obtained. Hair is highly autofluorescent, so improper removal of hair will result in low-quality images.

3.6.2. Flashlight Imaging

1. Use a blue LED flashlight with an excitation filter (midpoint wavelength peak of 470 nm) and a D470/40 emission filter

198

Hoffman

for whole-body imaging of GFP mice with RFP-expressing tumors growing in or on internal organs. Correct filters are necessary to eliminate tissue autofluorescence (58). 2. Acquire images with a digital camera and store on a PC as described above. 3. For C57BL/6 mice, remove hair with Nair or by shaving before images are obtained. Hair is highly autofluorescent, so improper removal of hair will result in low-quality images. 3.6.3. Light-Box Imaging

1. Do whole-body imaging in a fluorescent light box illuminated by fiberoptic lighting at 470 nm (36). 2. Collect emitted fluorescence through a GG475 filter on a CCD camera. (Use of separate band-pass filters for RFP or GFP emission allows a monochrome camera to be used.) 3. Capture high-resolution images directly on a PC. 4. Process images for contrast and brightness with the use of Image-Pro software or its equivalent. 5. For C57BL/6 mice, remove hair with Nair or by shaving before images are obtained. Hair is highly autofluorescent, so improper removal of hair will result in low-quality images.

3.6.4. Chamber Imaging

1. Do whole-body imaging with an Olympus OV100 imaging system (59). 2. Collect emitted fluorescence through appropriate filters configured on a filter wheel with a CCD camera. Variable magnification imaging can be done with a series of objective lenses. 3. Capture images on a PC and process images for contrast and brightness with Paint Shop Pro. 4. For C57BL/6 mice, remove hair with Nair or by shaving before images are obtained. Hair is highly autofluorescent, so improper removal of hair will result in low-quality images.

3.6.5. Imaging with the Olympus IV-100 Laser Scanning Microscope System

1. The tissue to be imaged using the Olympus IV-100 may be imaged either ex vivo or in a deeply anesthetized animal while secured. It is critical for optimal image resolution that the tissue being imaged does not move with the respiratory and cardiac variation in the animal (40). 2. Ex vivo tissue can be simply placed on a dark surface under the IV-100 objective with frequent application of PBS to keep the tissue moist during imaging. 3. Imaging of blood vessels, lymphatics and tumor tissue in a skin flap requires stabilization of the skin flap itself away from the body of the animal.

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

199

4. Likewise, imaging of tumor blood levels, lymphatics or blood vessels in the leg can be achieved by stabilization of the extremity such that the animal’s respiratory variation does not cause movement artifact in the tissue being imaged. 5. Intravital imaging of deeper organs in living animals requires stabilization of the organ and tumor tissue in question. This can be achieved in some organs, such as the pancreatic tail, which can be moved and stabilized without sacrifice of the animal, provided that the mouse remains deeply anesthetized throughout the duration of the imaging procedure. 6. Variable magnification down to the subcellular level can be imaged using the full range of objectives. Differential excitation of fluorophores can be achieved in this system by the use of three different lasers for excitation at 488, 561 and 633 nm. 3.6.6. Imaging Using Spectral Separation

In general, spectral separation imaging systems can provide greater sensitivity for specific fluorophore emission, although not all systems are equipped for high-resolution imaging (60). 1. The standard fluorescence imaging system previously described is replaced with a cooled monochrome camera and liquid-crystal tunable filter (CRI, Inc., Woburn, MA or equivalent) positioned in front of a conventional macro-lens. 2. A series of images is typically acquired every 10 nm from 500 to 650 nm and assembled into a spectral ‘stack’. 3. Using the predefined GFP and RFP emission spectra, the collected spectral ‘stack’ can be resolved into various images corresponding to specific wavelengths of interest that represent autofluorescence, GFP and RFP signals. 4. This method allows for maximal signal-to-noise ratio acquisition by virtue of its ability to separate out the competing autofluorescence or other fluorescence signals. 5. It is critical for this image acquisition that there be no movement in the tissue image when overlay images of multiple fluorescence signals are to be created (61).

3.7. Methods of Angiogenesis Analysis in GFP Models 3.7.1. Fluorescence Contrast or Color-Coded Imaging

1. Whole-body, skin-flap or intravital imaging can be performed (39, 40, 62). 2. Selective excitation of GFP is produced through a D425/60 filter and a 470 DCXR mirror.

200

Hoffman

3. Emitted fluorescence is collected through a GG475 filter on a CCD camera. 4. Images are processed for contrast and brightness and analyzed with the use of Image-Pro Plus software. 5. High-resolution images are captured directly on a PC or continuously through video output on a high-resolution VCR. 3.7.2. Quantitative Analysis of Angiogenesis

Periodically, the tumor-bearing mice are examined by intravital or whole-body fluorescence imaging (39, 40, 62). 1. The extent of blood vessel development in a tumor is evaluated based on the total length of blood vessels (L) in chosen areas: areas containing the highest number of vessels were identified by scanning tumors using intravital or whole-body imaging. 2. To compare the level of vascularization during tumor growth, the ‘hot’ areas with the maximum development of vessels per unit area are quantitated for L expressed in pixels. Captured images were corrected for unevenness in illumination. 3. Then the total number of pixels derived from the blood vessels is quantified with IMAGE PRO PLUS software.

3.8. Evaluation of Anti-angiogenetic Agents

1. Give mice daily i.p. injections of doxorubicin or other drugs or 0.9% NaCl solution (vehicle controls) on days 0, 1 and 2 after implantation of tumor cells (44, 45). 2. Anesthetize mice with the ketamine mixture and obtain biopsies on days 10, 14, 21 and 28 after implantation. 3. Gently flatten the tumor tissue between the slide and coverslip. 4. Quantify angiogenesis in the tumor tissue by measuring the length of GFP-expressing blood vessels in all fields using fluorescence microscopy. 5. Obtain measurements in all fields at ×40 or ×100 magnification to calculate the total length of GFP-expressing blood vessels. 6. Calculate the vessel density by dividing the total length of GFP-expressing blood vessels (in mm) by the tumor volume (in mm3 ).

3.9. Immunohistochemical Staining

1. ‘Snap-freeze’ fresh tissue with liquid nitrogen, then orient and embed the frozen tissue in optimum cutting temperature blocks and store at −80◦ C. Cut the frozen sections to a thickness of 5 µm with a cryostat (51).

Nestin-Driven Green Fluorescent Protein as an Imaging Marker

201

2. Detect co-localization of GFP fluorescence, CD31, IIIβ-tubulin and nestin in the frozen skin sections of mice transgenic for ND-GFP expression using the antirat immunoglobulin and anti-mouse immunoglobulin horseradish peroxidase detection kits following the manufacturer’s instructions. 3. Use monoclonal anti-CD31 (1:50 dilution), monoclonal anti-III-β-tubulin (1:500 dilution) and monoclonal antinestin (1:80 dilution) as primary antibodies. 4. Use staining with substrate-chromogen 3,3′ diaminobenzidine for antigen detection (see Note 3).

4. Notes 1. Advantages of visualizing blood vessels by their contrast to GFP-expressing tumors are the simplicity of the method and possibility of whole-body or external imaging. The disadvantage is that very small capillaries may not be visible. 2. The advantage of the dual-color model is the great resolution it affords to visualize very fine vessels. The disadvantage is that for highest resolution, tissue preparation may be needed. 3. The fluorescent models of angiogenesis will enable the process to be visualized at unprecedented resolution in real time (63).

Acknowledgments These studies were supported in part by grants CA099258 and CA103563 from the National Cancer Institute. References 1. Auerbach, R., Kubai, L., Knighton, D., Folkman, J. (1974) A simple procedure for the long-term cultivation of chicken embryos. Dev Biol 41, 391–394. 2. Crum, R., Szabo, S., Folkman, J. (1985) A new class of steroids inhibits angiogenesis in the presence of heparin or a heparin fragment. Science 230, 1375–1378.

3. Miller, J. W., Stinson, W. G., Folkman, J. (1993) Regression of experimental iris neovascularization with systemic alphainterferon. Ophthalmology 100, 9–14. 4. Passaniti, A., Taylor, R. M., Pili, R., Guo, Y., Long, P. V., Haney, J. A., et al. (1992) A simple, quantitative method for assessing angiogenesis and antiangiogenic agents using

202

5.

6.

7.

8.

9.

10.

11.

12.

13. 14.

15.

16.

17.

Hoffman reconstituted basement membrane, heparin, and fibroblast growth factor. Lab Invest 67, 519–528. Alessandri, G., Raju, K., Gullino, P. M. (1983) Mobilization of capillary endothelium in vitro induced by effectors of angiogenesis in vivo. Cancer Res 43, 1790–1797. Deutsch, T. A., Hughes, W. F. (1979) Suppressive effects of indomethacin on thermally induced neovascularization of rabbit corneas. Am J Ophthalmol 87, 536–540. Korey, M., Peyman, G. A., Berkowitz, R. (1977) The effect of hypertonic ointments on corneal alkali burns. Ann Ophthalmol 9, 1383–1387. Mahoney, J. M., Waterbury, L. D. (1985) Drug effects on the neovascularization response to silver nitrate cauterization of the rat cornea. Curr Eye Res 4, 531–535. Li, W. W., Grayson, G., Folkman, J., D’Amore, P. A. (1991) Sustained-release endotoxin. A model for inducing corneal neovascularization. Invest Ophthalmol Vis Sci 32, 2906–2911. Epstein, R. J., Hendricks, R. L., Stulting, R. D. (1990) Interleukin-2 induces corneal neovascularization in A/J mice. Cornea 9, 318–323. Gimbrone, M. A., Jr., Cotran, R. S., Leapman, S. B., Folkman, J. (1974) Tumor growth and neovascularization: An experimental model using the rabbit cornea. J Natl Cancer Inst 52, 413–427. Fournier, G. A., Lutty, G. A., Watt, S., Fenselau, A., Patz, A. (1981) A corneal micropocket assay for angiogenesis in the rat eye. Invest Ophthalmol Vis Sci 21, 351–354. Muthukkaruppan, V., Auerbach, R. (1979) Angiogenesis in the mouse cornea. Science 205, 1416–1418. Papenfuss, H. D., Gross, J. F., Intaglietta, M., Treese, F. A. (1979) A transparent access chamber for the rat dorsal skin fold. Microvasc Res 18, 311–318. Shan, S., Robson, N. D., Cai, Y., Qiao, T., Li, C. Y., Kontos, C. D., et al. (2004) Responses of vascular endothelial cells to angiogenic signaling are important for tumor cell survival. FASEB J 18, 326–328. Dewhirst, M., Gross, J., Sim, D., Arnold, P., Boyer, D. (1984) The effect of rate of heating or cooling prior to heating on tumor and normal tissue microcirculatory blood flow. Biorheol 21, 539–558. Fukumura, D., Xavier, R., Sugiura, T., Chen, Y., Park, E. C., Lu, N., et al. (1998) Tumor induction of VEGF promoter activity in stromal cells. Cell 94, 715–725.

18. Li, C. Y., Shan, S., Huang, Q., Braun, R. D., Lanzen, J., Hu, K., et al. (2000) Initial stages of tumor cell-induced angiogenesis: evaluation via skin window chambers in rodent models. J Natl Cancer Inst 92, 143–147. 19. Al-Mehdi, A. B., Tozawa, K., Fisher, A. B., Shientag, L., Lee, A., Muschel, R. J. (2000) Intravascular origin of metastasis from the proliferation of endothelium-attached tumor cells: a new model for metastasis. Nat Med 6, 100–102. 20. Huang, Q., Shan, S., Braun, R. D., Lanzen, J., Anyrhambatla, G., Kong, G., et al. (1999) Noninvasive visualization of tumors in rodent dorsal skin window chambers. Nat Biotechnol 17, 1033–1035. 21. Cowen, S. E., Bibby, M. C., Double, J. A. (1995) Characterisation of the vasculature within a murine adenocarcinoma growing in different sites to evaluate the potential of vascular therapies. Acta Oncol 34, 357–360. 22. Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G., Cormier, M. J. (1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233. 23. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., Prasher, D. C. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805. 24. Cheng, L., Fu, J., Tsukamoto, A., Hawley, R. G. (1996) Use of green fluorescent protein variants to monitor gene transfer and expression in mammalian cells. Nat Biotechnol 14, 606–609. 25. Cody, C. W., Prasher, D. C., Westler, W. M., Prendergast, F. G., Ward, W. W. (1993) Chemical structure of the hexapeptide chromophore of the Aequorea green fluorescent protein. Biochemistry 32, 1212–1218. 26. Yang, F., Moss, L. G., Phillips, G. N., Jr. (1996) The molecular structure of green fluorescent protein. Nat Biotechnol 14, 1246–1251. 27. Morin, J., Hastings, J. (1971) Energy transfer in a bioluminescent system. J Cell Physiol 77, 313–318. 28. Cormack, B., Valdivia, R., Falkow, S. (1996) FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38. 29. Crameri, A., Whitehorn, E. A., Tate, E., Stemmer, W. P. (1996) Improved green fluorescent protein by molecular evolution using DNA shuffling. Nat Biotechnol 14, 315–319.

Nestin-Driven Green Fluorescent Protein as an Imaging Marker 30. Delagrave, S., Hawtin, R. E., Silva, C. M., Yang, M. M., Youvan, D. C. (1995) Red-shifted excitation mutants of the green fluorescent protein. Biotechnology 13, 151–154. 31. Heim, R., Cubitt, A. B., Tsien, R. Y. (1995) Improved green fluorescence. Nature 373, 663–664. 32. Zolotukhin, S., Potter, M., Hauswirth, W. W., Guy, J., Muzyczka, N. (1996) A ‘humanized’ green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J Virol 70, 4646–4654. 33. Gross, L. A., Baird, G. S., Hoffman, R. C., Baldridge, K. K., Tsien, R. Y. (2000) The structure of the chromophore within DsRed, a red fluorescent protein from coral. Proc Natl Acad Sci USA 97, 11990–11995. 34. Fradkov, A. F., Chen, Y., Ding, L., Barsova, E. V., Matz, M. V., Lukyanov, S. A. (2000) Novel fluorescent protein from Discosoma coral and its mutants possesses a unique farred fluorescence. FEBS Lett 479,127–130. 35. Chishima, T., Miyagi, Y., Wang, X., Yamaoka, H., Shimada, H., Moossa, A. R. et al. (1997) Cancer invasion and micrometastasis visualized in live tissue by green fluorescent protein expression. Cancer Res 57,2042–2047. 36. Yang, M., Baranov, E., Jiang, P., Sun, F. X., Li, X. M., Li, L., et al. (2000) Whole-body optical imaging of green fluorescent protein-expressing tumors and metastases. Proc Natl Acad Sci USA 97, 1206–1211. 37. Hoffman, R. M. (2002) Green fluorescent protein imaging of tumour growth, metastasis, and angiogenesis in mouse models. Lancet Oncol 3, 546–556. 38. Hoffman, R. M. (1999) Orthotopic metastatic mouse models for anticancer drug discovery and evaluation: a bridge to the clinic. Invest New Drugs 17, 343–359. 39. Yang, M., Baranov, E., Li, X. M., Wang, J. W., Jiang, P., Li, L., et al. (2001) Whole-body and intravital optical imaging of angiogenesis in orthotopically implanted tumors. Proc Natl Acad Sci USA 98, 2616–2621. 40. Yang, M., Jiang, P., Hoffman, R. M. (2007) Whole-body subcellular multicolor imaging of tumor-host interaction and drug response in real time. Cancer Res 67, 5195–5200. 41. Yang, M., Baranov, E., Wang, J. W., Jiang, P., Wang, X., Sun, F. X. et al. (2002) Direct external imaging of nascent cancer, tumor progression, angiogenesis, and metastasis on internal organs in the fluorescent orthotopic model. Proc Natl Acad Sci USA 99, 3824–3829.

203

42. Li, L., Mignone, J., Yang, M., Matic, M., Penman, S., Enikolopov, G., Hoffman, R. M. (2003) Nestin expression in hair follicle sheath progenitor cells. Proc Natl Acad Sci USA 100, 9958–9961. 43. Amoh, Y., Li, L., Yang, M., Moossa, A. R., Katsuoka, K., Penman, S., et al. (2004) Nascent blood vessels in the skin arise from nestin-expressing hair follicle cells. Proc Natl Acad Sci USA 101, 13291–13295. 44. Amoh, Y., Li, L., Yang, M., Jiang, P., Moossa, A. R., Katsuoka, K., et al. (2005) Hair follicle-derived blood vessels vascularize tumors in skin and are inhibited by doxorubicin. Cancer Res 65, 2337–2343. 45. Amoh, Y., Yang, M., Li, L., Reynoso, J., Bouvet, M., Moossa, A. R., et al. (2005) Nestin-linked green fluorescent protein transgenic nude mouse for imaging human tumor angiogenesis. Cancer Res 65, 5352–5357. 46. Amoh, Y., Bouvet, M., Li, L., Tsuji, K., Moossa, A. R., Katsuoka, K., et al. (2006) Visualization of nascent tumor angiogenesis in lung and liver metastasis by differential dual-color fluorescence imaging in nestin-linked-GFP mice. Clin Exp Metastasis 23, 315–322. 47. Amoh, Y., Li, L., Tsuji, K., Moossa, A. R., Katsuoka, K., Hoffman, R. M., et al. (2006) Dual-color imaging of nascent blood vessels vascularizing pancreatic cancer in an orthotopic model demonstrates antiangiogenesis efficacy of gemcitabine. J Surg Res 132, 164–169. 48. Amoh, Y., Nagakura, C., Maitra, A., Moossa, A. R., Katsuoka, K., Hoffman, R. M., et al. (2006) Dual-color imaging of nascent angiogenesis and its inhibition in liver metastases of pancreatic cancer. Anticancer Res 26, 3237–3242. 49. Ji, Y., Hayashi, K., Amoh, Y., Tsuji, K., Yamauchi, K., Yamamoto, N., et al. (2007) The camptothecin derivative CPT-11 inhibits angiogenesis in a dual-color imageable orthotopic metastatic nude mouse model of human colon cancer. Anticancer Res 27, 713–718. 50. Hayashi, K., Yamauchi, K., Yamamoto, N., Tsuchiya, H., Tomita, K., Amoh, Y., et al. (2007) Dual-color imaging of angiogenesis and its inhibition in bone and soft tissue sarcoma. J Surg Res 140, 165–170. 51. Amoh, Y., Li, L., Katsuoka, K., Bouvet, M., Hoffman, R. M. (2007) GFP-expressing R vascularization of Gelfoam as a rapid in vivo assay of angiogenesis stimulators and inhibitors. Biotechniques 42, 294–298.

204

Hoffman

52. Hayashi, K., Yamauchi, K., Yamamoto, N., Tsuchiya, H., Tomita, K., Bouvet, M., Wessels, J., Hoffman, R. M. (2009) A color-coded orthotopic nude-mouse treatment model of brain-metastatic paralyzing spinal cord cancer that induces angiogenesis and neurogenesis. Cell Proliferat 42, 75–82. 53. Shaked, Y., Ciarrocchi, A., Franco, M., Lee, C. R., Man, S., Cheung, A. M., et al. (2006) Therapy-induced acute recruitment of circulating endothelial progenitor cells to tumors. Science 313, 1785–1787. 54. Yang, M., Jiang, P., Sun, F. X., Hasegawa, S., Baranov, E., Chishima, T., Shimada, H., Moossa, A. R., Hoffman, R. M. (1999) A fluorescent orthotopic bone metastasis model of human prostate cancer. Cancer Res 59, 781–786. 55. Yang, M., Jiang, P., An, Z., Baranov, E., Li, L., Hasegawa, S., Al-Tuwaijri, M., Chishima, T., Shimada, H., Moossa, A. R., Hoffman, R. M. (1999) Genetically fluorescent melanoma bone and organ metastasis models. Clin Cancer Res 5, 3549–3559. 56. Cruz-Munoz, W., Man, S., Xu, P., Kerbel, R. S. (2008) Development of a preclinical model of spontaneous human melanoma central nervous system metastasis. Cancer Res 68, 4500–4505. 57. Hoffman, R. M. (2009) Comment Re: Preclinical model of spontaneous melanoma metastasis. Cancer Res 69, 719.

58. Yang, M., Luiken, G., Baranov, E., Hoffman, R. M. (2005) Facile whole-body imaging of internal fluorescent tumors in mice with an LED flashlight. Biotechniques 39, 170–172. 59. Yamauchi, K., Yang, M., Jiang, P., Xu, M., Yamamoto, N., Tsuchiya, H., Tomita, K., Moossa, A. R., Bouvet, M., Hoffman, R. M. (2006) Development of real-time subcellular dynamic multicolor imaging of cancer cell trafficking in live mice with a variablemagnification whole-mouse imaging system. Cancer Res 66, 4208–4214. 60. Mayes, P. A., Dicker, D. T., Liu, Y., ElDeiry, W. S. (2008) Noninvasive vascular imaging in fluorescent tumors using multispectral unmixing. Biotechniques 45, 459–464. 61. Mansfield, J. R., Gossage, K. W., Hoyt, C. C., Levenson, R. M. (2005) Autofluorescence removal, multiplexing, and automated analysis methods for in-vivo fluorescence imaging. J Biomed Opt 10, 41207. 62. Yang, M., Li, L., Jiang, P., Moossa, A. R., Penman, S., Hoffman, R. M. (2003) Dualcolor fluorescence imaging distinguishes tumor cells from induced host angiogenic vessels and stromal cells. Proc Natl Acad Sci USA 100, 14259–14262. 63. Amoh, Y., Katsuoka, K., Hoffman, R. M. (2008) Color-coded fluorescent protein imaging of angiogenesis: the AngioMouse models. Curr Pharm Des 14, 3810–3819.

Chapter 12 Imaging Calcium Sparks in Cardiac Myocytes Silvia Guatimosim, Cristina Guatimosim, and Long-Sheng Song Abstract Calcium ions play fundamental roles in many cellular processes in virtually all type of cells. The use of Ca2+ sensitive fluorescent indicators has proven to be an indispensable tool for studying the spatiotemporal dynamics of intracellular calcium ([Ca2+ ]i ). With the aid of laser scanning confocal microscopy and new generation of Ca2+ indicators, highly localized, short-lived Ca2+ signals, namely Ca2+ sparks, were revealed as elementary Ca2+ release events during excitation–contraction coupling in cardiomyocytes. Since the discovery of Ca2+ sparks in 1993, the demonstration of dynamic Ca2+ micro-domains in living cardiomyocytes has revolutionized our understanding of Ca2+ -mediated signal transduction in normal and diseased hearts. In this chapter, we have described a commonly used method for recording local and global Ca2+ signals in cardiomyocytes using the fluorescent indicator fluo-4 acetoxymethyl (AM) and laser scanning confocal microscopy. Key words: Calcium sparks, confocal microscopy, ventricular myocytes, fluorescence, calcium indicators.

1. Introduction Cytosolic free Ca2+ ([Ca2+ ]i ) is a versatile second messenger that can simultaneously regulate multiple processes within an individual cell. In the cardiac cell, cytosolic [Ca2+ ] ([Ca2+ ]i ) is actively maintained at a very low level of around 100 nM by Ca2+ homeostatic mechanisms, including the SR Ca2+ -ATPase (SERCA) and the plasmalemmal Na+ /Ca2+ exchanger (NCX) and Ca2+ -ATPase and by a number of Ca2+ buffering molecules (1). During each heart beat, a time-dependent transient increase in intracellular Ca2+ concentration (“[Ca2+ ]i transient”) occurs and is responsible for activating contraction in a process called H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_12, © Springer Science+Business Media, LLC 2011

205

206

Guatimosim, Guatimosim, and Song

excitation–contraction coupling. The [Ca2+ ]i transient is triggered by the cardiac action potential (AP) and spreads through the heart as the AP is propagated (2). [Ca2+ ]i transients during cardiac excitation–contraction (EC) coupling were first described as aequorin luminescence in frog cardiac muscle by David G. Allen (3) and in canine Pukinje fibers by Gil Wier (4) using microinjection of photoprotein into muscle cells. The development of new fluorescent Ca2+ indicators, such as fura-2 and indo-1 by Tsien and co-workers (5), had improved [Ca2+ ] measurements in single cardiac muscle cells. There are two advantages of these dyes. First, the acetoxymethyl (AM) ester derivatives of fura-2 and indo-1 can permeate the cell membrane, which makes the application of these Ca2+ indicators much easier comparing to aequorin microinjection. Second, their unique excitation/emission features allow one to be able to make accurate, ratio measurement of intracellular Ca2+ concentrations. In conjunction with digital imaging technique, Wier et al. were the first to document spontaneous [Ca2+ ] waves and different patterns of subcellular Ca2+ concentration in quiescent, spontaneously active or hyper-contracting cardiomyocytes (6). However, the spatialtemporal resolution was not high enough for them to be able to detect the highly localized Ca2+ release events in cardiac myocytes as we can observe routinely nowadays. Advances in Ca2+ fluorescence technology (still driven by Tsien and his colleagues) combined with the advent of the laser scanning confocal microscope made it possible for the discovery of Ca2+ sparks in cardiac myocytes. Ca2+ sparks were firstly reported in quiescent ventricular myocytes by Cheng et al. (7). Since then Ca2+ sparks or local Ca2+ release events with spark characteristics have been recorded in skeletal muscle (8–10), smooth muscle (11), neurons (12) and more recently in fibroblasts (13). In heart muscle, Ca2+ sparks are now well accepted as the elementary events of SR Ca2+ release underlying EC coupling, originated from the opening of a cluster of sarcoplasmic reticulum (SR) Ca2+ release channels or ryanodine receptors (RyRs). Ca2+ spark observed in unstimulated resting single cardiac myocytes represents a local transient increase in intracellular [Ca2+ ]i . It has a rapid rise (∼10 ms, time to peak) and a moderately quick decay kinetics (∼20 ms, half-time of decay) and is confined to an area of ∼2.0 µm in diameter or ∼8 fl by volume (14). At diastolic [Ca2+ ]i of about 100 nM, Ca2+ sparks are spontaneously firing at a very low rate (∼100 per second per cell). The occurrence of spontaneous Ca2+ sparks does not require Ca2+ entry into the cardiomyocyte through L-type Ca2+ channels (LCCs) or by other Ca2+ pathways across the sarcolemma. However, during cardiac EC coupling, Ca2+ influx through LCCs evokes synchronous activation of tens of thousands of Ca2+ sparks by a mechanism

Imaging Calcium Sparks in Cardiac Myocytes

207

called Ca2+ -induced Ca2+ release (CICR) (15, 16). This process is locally controlled within a 12 nm junctional subspace between t-tubular and SR membrane (17). It has been elegantly shown that Ca2+ sparks can be triggered by adjacent single LCC openings (16, 18, 19). The summation of numerous Ca2+ sparks activated simultaneously all over the myocyte compose a uniform Ca2+ transient (see Fig. 12.1).

Fig. 12.1. a A train of steady-state Ca2+ transients elicited by 1-Hz field stimulation. b Spatially averaged Ca2+ profile showing the dynamic change of Ca2+ signals with time. This panel also depicts the analysis of Ca2+ transient amplitude and kinetics.

Fluo-3 and fluo-4 have been the indicators of choice in Ca2+ spark experiments, because of their unique properties that confer a high signal-to-noise ratio, fast “on” and “off” kinetics and high sensitivity when the indicator responds to [Ca2+ ] gradients. Since its introduction in 1989, fluo-3 confocal Ca2+ imaging has made significant contribution to our understanding of spatial dynamics of many elementary process of Ca2+ signaling in different cell types. Fluo-4, an analog of fluo-3, with higher quantum yield when excited at 488 nm, provides brighter emission signals in response to Ca2+ binding when compared to fluo-3. When estimating [Ca2+ ]i from the observed fluorescence signal (F), a common practice is to express the data as the ratio: R = F/F0 , where F0 refers to the baseline fluorescence at resting [Ca2+ ]i . The major disadvantage of fluo-family of dyes (fluo-3/fluo4), however, is that, upon binding of Ca2+ ions, there is little or no shift in its excitation or emission spectrum, which makes it impossible to perform ratiometric measurements of [Ca2+ ] (20). Most chemical fluorescent indicators are cell impermeant, therefore many of the fluorescent Ca2+ indicators are derivatized with AM ester groups. The AM form of the indicator can diffuse across cell membranes, and once inside the cell, esterases cleave

208

Guatimosim, Guatimosim, and Song

the AM group off the probe leading to a cell-impermeant indicator. Because the AM derivative has low aqueous solubility, some dispersing agents such as Pluronic F-127 are often used to help solubilize large dye molecules in physiological media and facilitate cell loading (20).

2. Materials 1. Fluo-4 AM (10× 50 µg, F14201, Invitrogen). Store at −20◦ C. 2. 20% Pluronic F127 in DMSO solution (P-3000MP, Invitrogen). Store at room temperature. 3. Fluo-4 AM loading stock solution: dissolve 50 µg fluo-4 AM with 50 µL 20% Pluronic F-127 DMSO solution. Store stock solution at −20◦ C. 4. Tyrode’s solution with the following composition (mM): 140 NaCl, 5 KCl, 5 HEPES, 1 NaH2 PO4 , 1 MgCl2 , 1.8 CaCl2 and 10 glucose (pH 7.4) adjusted with NaOH. All salts and buffers used for the preparation of normal Tyrode’s R solution can be purchased from Sigma-Aldrich . Store at ◦ 4 C. 5. Modified Dulbecco’s Modified Eagle Medium (DMEM): The basic medium routinely used to keep the isolated adult myocytes is supplemented DMEM (powder purchased from Sigma, catalog #D1152). To make up 50 mL of media for incubating the cells, add 5 mL of inactivated fetal bovine serum, 5 µL insulin (3.66 mg/mL) and 550 µL NaCl (4 M) to 40 mL DMEM solution (made from 0.87 g DMEM powder in 40 mL Milli-Q grade water). Adjust the pH with NaOH to 7.2 and complete the volume. Then keep at room temperature for use on the same day. 6. Electrical stimulator for field stimulation of myocytes, with the capacity to deliver at least 20 V square pulses. 7. Perfusion chamber with attached platinum wires, mounted on the stage of a confocal microscope (see Note 1). 8. Confocal microscope equipped with an Argon laser of 488 nm line for fluo-4 excitation and appropriate filters for acquiring emission signals at certain wavelength range (for example, long pass filter that passes emission signals of wavelength >505 nm or band-pass filter that passes emission signals of wavelength between 505 and 550 nm).

Imaging Calcium Sparks in Cardiac Myocytes

209

3. Methods 3.1. Isolating Ventricular Myocytes from Adult Rat

Adult rat ventricular cells will be prepared by standard methods as previously described in the literature (21). Briefly, male rats weighing between 200 and 300 g will be sacrificed by lethal intraperitoneal injection of pentobarbital sodium (100 mg/kg). The hearts will be rapidly removed and perfused via the Langendorff apparatus with Ca2+ -free modified Tyrode solution until the blood is washed out. Hearts will then be perfused with Tyrode solution containing 50 µM CaCl2 along with 1.4 mg/mL collagenase (type 2) and 0.04 mg/mL protease (type XIV) until they are soft (approximately 10 min). The hearts will then be removed from the perfusion apparatus, minced into ~1-mm chunks and stirred for 4 min in Tyrode solution containing 50 µM CaCl2 , 0.7 mg/mL collagenase and 0.02 mg/mL protease. Cells will be filtered through a 200 µm mesh to remove tissue chunks and extracellular Ca2+ concentration is raised to 0.5 mM over 10 min through three centrifuge cycles (0.1 mM Ca2+ , 0.2 mM Ca2+ , 0.5 mM Ca2+ ). Finally myocytes will be harvested and stored in modified DMEM until they are used (within 5 h) (22, 23).

3.2. Fluo-4 AM Loading

Add 10 µL fluo-4 AM stock solution to 1 mL of cell suspension (final fluo-4 AM concentration = 10 µM). Cells should remain in the dark at room temperature for 20 min (see Notes 2 and 3). Centrifuge the cells (2 min at 200–300 rpm centrifugation), remove the supernatant and gently re-suspend the pellet in indicator-free Tyrode solution. Wait for 20 min to allow for complete de-esterification of AM esters. Then, cells will be ready for Ca2+ spark imaging with confocal microscope for up to 2 h (see Note 4). The anion-transport inhibitor probenecid (2 mM) may be added to the cell solution to reduce leakage of the de-esterified indicator.

3.3. Ca2+ Imaging in Ventricular Myocytes

Rod-shaped myocytes with clear striations and without actively spontaneous contraction (less than one per minute) are considered healthy Ca2+ tolerant cells and will be chosen for Ca2+ imaging (see Note 5). Ca2+ transients will be elicited by field stimulation through a pair of platinum electrodes, with a 2 ms suprathreshold square voltage pulse delivered by a commercially available electrical stimulator (such as Myopacer 100, IonOptix Inc.). Cells are normally stimulated at 1 Hz for 15 s to let reach a steadystate condition before recording. An LSM 510 scanning system (Zeiss GmbH, Jena, Germany) equipped with a ×63 oil immersion objective (numerical aperture (NA) = 1.4) will be used for confocal imaging of Ca2+ fluorescence (see Note 6). Fluo-4 will

210

Guatimosim, Guatimosim, and Song

be excited by 488 nm line of an Argon laser and emission signals over 505 nm will be collected. The brightness of the fluorescent signals represents the relative level of intracellular [Ca2+ ]i . For recording Ca2+ transients/sparks, a line scan mode is normally utilized. The confocal pinhole is set to render spatial resolutions of 0·4 µm in the horizontal plane and 0·9 µm in the axial direction (see Note 7). Ideally, the detector gain is set at around 700 (no digital gain). Line-scan images are acquired at sampling rate of 1.54 or 1.92 ms per line, along the longitudinal axis of the cell. Each line comprises 512 pixels spaced at 0.14 µm intervals. After a sequential scanning, a two-dimensional (2D) image of 512 × 1000 lines or 512 × 2000 lines will be generated and stored for offline analysis (see below). It is not recommended to scan a cell in the same line region for prolonged time (see Note 8). 3.4. Recording Ca2+ Transients in Ventricular Myocytes

Fluo-4 AM-loaded cells will be allowed to settle on a coated glass coverslip by gravity (see Note 9). Wait for 5–10 min and then turn the perfusion solution on; the cells will be bathed in Tyrode’s solution. Cells can be paced with parallel platinum wires connected to an electrical stimulator (Myopacer 100, IonOptix Inc.). Stimulation settings should be as follows: duration: 2 ms; continuous biphasic pulse stimulation; voltage: adjusted to 120% of the threshold voltage that induces Ca2+ transients. When cells are stimulated at 1 Hz, the spatially averaged [Ca2+ ]i transient obtained by integrating the line-scan image should be similar to that presented in Fig. 12.1 (see Note 10).

3.5. Recording Ca2+ Sparks in Ventricular Myocytes

Spontaneous Ca2+ sparks may be recorded using the same confocal settings used for Ca2+ transient imaging (e.g. line scan mode, laser power, pinhole size, detector gain, etc.). During spark recording, cells are kept in Tyrode’s solution under resting conditions (non-stimulated). For guinea pig, rabbit, canine and other large mammalian heart cells, a 15-s field stimulation (1 Hz) is required to load the SR prior to spark recording (see Note 11). Soon after the halt of field stimulation, a series of line scan images (e.g. 6 sweeps; each sweep image can be 512 pixels × 1000 lines) will be acquired at a rate of 1.54 or 1.92 ms per line. For rat and mouse heart cells, spontaneous Ca2+ sparks may be visualized with or without pre-stimulation. However, it is more accurate to compare Ca2+ sparks recorded at steady-state conditions. Figure 12.2 shows a typical confocal line scan image of a Ca2+ spark recorded in a control rat ventricular myocyte.

3.6. Image Analyses

Digital image processing will be performed by using customdevised routines created with IDL programming language (Research Systems, Boulder, CO) (24). The Ca2+ level is reported as F/F0 (or as F /F0 ), where F0 is the resting Ca2+ fluorescence.

Imaging Calcium Sparks in Cardiac Myocytes

211

Fig. 12.2. Typical Ca2+ spark recorded from a control rat ventricular myocyte at resting condition, loaded with fluo-4 AM. a The analysis of basic Ca2+ spark characteristics (amplitude, FDHM and FWHM). b A surface plot of the Ca2+ spark shown in Panel a.

By using the following equation, we can convert the Ca2+ fluorescence ratio to a Ca2+ concentration (7): [Ca2+ ]i = KR/{K/[Ca2+ ]rest + 1) − R} where K is the dissociation constant of the Ca2+ indicator used, R is the fluorescence ratio (F /F0 ), [Ca2+ ]rest is the resting Ca2+ concentration. Assuming the dissociation constant (K) of fluo4 AM is 400 nM (see Invitrogen Inc.), the resting [Ca2+ ]i of a cardiac myocyte is 100 nM, the amplitude of a typical Ca2+ spark is 2; we can then estimate that the peak Ca2+ concentration of Ca2+ spark is around 270 nM. Figure 12.1 shows a typical train of Ca2+ transients from a normal rat ventricular myocyte. With the aid of computer programming analysis, the rising phase (time to peak, tpeak ), the amplitude (F/F0 ) and the decay kinetics (t50 , t75 , t90 ) may be extracted from original Ca2+ images (25, 26). Figure 12.2 displays a typical Ca2+ spark and illustrates the analyses of key Ca2+ spark parameters: amplitude (F/F0 ), duration (FDHM, full duration at half-maximal amplitude) and spatial width (FWHM, full-width at half-maximal amplitude). These parameters represent the basic gating properties of RyRs Ca2+

212

Guatimosim, Guatimosim, and Song

release channels: the release flux (F/F0 ) and the gating kinetics of RyRs (FDHM). Ca2+ sparks of ventricular myocytes, on average, are about 1.8F/F0 in amplitude, 2 µm wide and 25 ms long (27).

4. Notes 1. The chamber bottom glass has to be no more than 170 µm thick (#1.5 glass coverslip). High numerical aperture lens (e.g. Zeiss Plan-Apochromat 63× oil immersion) have a short working distance of 190 µm. We usually use 120 µm thick #1 glass coverslip). 2. Fluo-4 AM dye loading can vary, for example, rabbit or dog myocytes may require longer loading time. Myocytes with an optimal fluo-4 AM loading shall report a baseline fluorescence of 30–40. Too high or too low baseline fluorescence levels indicate overloading or underloading of the fluorescent indicator into the cell. 3. Loading fluo-4 AM may be done at room temperature, rather than at 37◦ C. Acetoxymethyl ester loading at high temperature may often cause severe subcellular compartmentalization of the indicator and may interfere with the measurement of cytosolic Ca2+ concentration. 4. Cardiomyocytes can last for hours after Fluo-4 AM loading and still provide Ca2+ transient data; however, because of the leakage of esterified indicator (although slow), myocytes loaded with fluo-4 AM will exhibit dim fluorescence with time. Routinely, we examined cells for 2 h following loading without marked deterioration of Ca2+ signals. 5. Cardiomyocytes suitable for loading and measurement should appear rod shaped, with clear striations and without cytoplasmic protrusions or blebs under phase-contrast light microscopy. 6. Water immersion objective lens can be used. These objective lens are designed for use with aqueous specimens and immersion medium and can be “corrected” for the unavoidable refractive index mismatch produced by the use of a glass coverslip (h = 1.51). Practically, oil immersion lens are often used for the much higher expense of water immersion lens. 7. The optical resolution of the confocal microscope is 0·4 µm in the horizontal plane and 0.9 µm in the axial direction, as determined by measuring the point spread function of

Imaging Calcium Sparks in Cardiac Myocytes

213

0·1 µm fluorescent bead (Molecular Probes – Invitrogen Inc.). 8. Prolonged scanning in the same region may cause photobleaching of fluorescent molecules and experimental artifacts induced by photo-damage. 9. Glass coverslip can be coated with laminin. Laminin promotes cell attachment, preventing stimulated contraction out of the focal plane during imaging. Laminin is diluted to a final concentration of between 1 and 5 µg/mL in a volume of phosphate-buffered saline or culture medium, which adequately covers the culture surface. Laminin should be applied to the coverslip at least 30 min before plating out cells. 10. Cells should be field-stimulated at 1.0 Hz for 15 s to reach steady-state Ca2+ dynamics before image acquisition. 11. Myocytes of large mammals tend to unload their SR Ca2+ content, which makes Ca2+ sparks hard to be detected at resting conditions. One way to circumvent this problem is to use a conditioning protocol to upload Ca2+ into the SR before Ca2+ spark recording (28).

References 1. Bers, D. M., (2002) Cardiac excitationcontraction coupling. Nature 415, 198–205. 2. Guatimosim, S., Dilly, K., Santana, L. F., Saleet, J. M., Sobie, E. A., Lederer, W. J. (2002) Local Ca(2+ ) signaling and EC coupling in heart: Ca(2+ ) sparks and the regulation of the [Ca(2+ )](i) transient. J Mol Cell Cardiol 34, 941–950. 3. Allen, D. G., Blinks, J. R. (1978) Calcium transients in aequorin-injected frog cardiac muscle. Nature 273, 509–513. 4. Wier, W. G. (1980) Calcium transients during excitation-contraction coupling in mammalian heart: aequorin signals of canine Purkinje fibers. Science 207, 1085–1087. 5. Minta, A., Kao, J. P., Tsien, R. Y. (1989) Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. J Biol Chem 264, 8171–8178. 6. Wier, W. G., Cannell, M. B., Berlin, J. R, Marban, E., Lederer, W. J. (1987) Cellular and subcellular heterogeneity of [Ca2+ ]i in single heart cells revealed by fura-2. Science 235, 325–328. 7. Cheng, H., Lederer, W. J., Cannell, M. B. (1993) Calcium sparks: elementary events underlying excitation-contraction coupling in heart muscle. Science 262, 740–744.

8. Klein, M. G., Cheng, H., Santana, L. F., Jiang, Y. H., Lederer, W. J., Schneider, M. F. (1996) Two mechanisms of quantized calcium release in skeletal muscle. Nature 379, 455–458. 9. Tsugorka, A., Rios, E., Blatter, L. A. (1995) Imaging elementary events of calcium release in skeletal muscle cells. Science 269, 1723–1726. 10. Wang, X., Weisleder, N., Collet, C., et al. (2005) Uncontrolled calcium sparks act as a dystrophic signal for mammalian skeletal muscle. Nat Cell Biol 7, 525–530. 11. Nelson, M. T., Cheng, H., Rubart, M., et al. (1995) Relaxation of arterial smooth muscle by calcium sparks. Science 270, 633–637. 12. Ouyang, K., Zheng, H., Qin, X., et al. (2005) Ca2+ sparks and secretion in dorsal root ganglion neurons. Proc Natl Acad Sci USA 102, 12259–12264. 13. Wei, C., Wang, X., Chen, M., Ouyang, K., Song, L. S., Cheng, H. (2009) Calcium flickers steer cell migration. Nature 457, 901–905. 14. Cheng, H., Lederer, W. J. (2008) Calcium sparks. Physiol Rev 88, 1491–1545.

214

Guatimosim, Guatimosim, and Song

15. Cannell, M. B., Cheng, H., Lederer, W. J. (1994) Spatial non-uniformities in [Ca2+ ]i during excitation-contraction coupling in cardiac myocytes. Biophys J 67, 1942–1956. 16. Cannell, M. B., Cheng, H., Lederer, W. J. (1995) The control of calcium release in heart muscle. Science 268, 1045–1049. 17. Franzini-Armstrong, C., Protasi, F., Ramesh, V. (1999) Shape, size, and distribution of Ca(2+ ) release units and couplons in skeletal and cardiac muscles. Biophys J 77, 1528–1539. 18. Lopez-Lopez, J. R., Shacklock, P. S., Balke, C. W., Wier, W. G. (1995) Local calcium transients triggered by single L-type calcium channel currents in cardiac cells. Science 268, 1042–1045. 19. Wang, S. Q., Song, L. S., Lakatta, E. G., Cheng, H. (2001) Ca2+ signalling between single L-type Ca2+ channels and ryanodine receptors in heart cells. Nature 410, 592–596. 20. Takahashi, A., Camacho, P., Lechleiter, J. D., Herman, B. (1999) Measurement of intracellular calcium. Physiol Rev 79, 1089–1125. 21. Guatimosim, S., Sobie, E. A., dos Santos, C. J., Martin, L. A., Lederer, W. J. (2001) Molecular identification of a TTX-sensitive Ca(2+ ) current. Am J Physiol Cell Physiol 280, C1327–C1339. 22. Mitcheson, J. S., Hancox, J. C., Levi, A. J. (1998) Cultured adult cardiac myocytes:

23.

24.

25.

26.

27.

28.

future applications, culture methods, morphological and electrophysiological properties. Cardiovasc Res 39, 280–300. Mitra, R., Morad, M. (1985) A uniform enzymatic method for dissociation of myocytes from hearts and stomachs of vertebrates. Am J Physiol 249, H1056–H1060. Cheng, H., Song, L. S., Shirokova, N., et al. (1999) Amplitude distribution of calcium sparks in confocal images: theory and studies with an automatic detection method. Biophys J 76, 606–617. Dias-Peixoto, M. F., Santos, R. A., Gomes, E. R., et al. (2008) Molecular mechanisms involved in the angiotensin-(1–7)/Mas signaling pathway in cardiomyocytes. Hypertension 52, 542–548. Song, L. S., Sobie, E. A., McCulle, S., Lederer, W. J., Balke, C. W., Cheng, H. (2006) Orphaned ryanodine receptors in the failing heart. Proc Natl Acad Sci USA 103, 4305–4310. Song, L. S., Guia, A., Muth, J. N., et al. (2002) Ca(2+ ) signaling in cardiac myocytes overexpressing the alpha(1) subunit of L-type Ca(2+ ) channel. Circ Res 90, 174–181. Song, L. S., Pi, Y., Kim, S. J., et al. (2005) Paradoxical cellular Ca2+ signaling in severe but compensated canine left ventricular hypertrophy. Circ Res 97, 457–464.

Chapter 13 Light Microscopy in Aquatic Ecology: Methods for Plankton Communities Studies Maria Carolina S. Soares, Lúcia M. Lobão, Luciana O. Vidal, Natália P. Noyma, Nathan O. Barros, Simone J. Cardoso, and Fábio Roland Abstract Planktonic organisms dominate waters in ponds, lakes and oceans. Because of their short life cycles, plankters respond quickly to environmental changes and the variability in their density and composition are more likely to indicate the quality of the water mass in which they are found. Planktonic community is formed by numerous organisms from distinct taxonomic position, ranging from 0.2 µm up to 2 mm. Despite others, the light microscopy is the most used apparatus to enumerate these organisms and different techniques are necessary to cover differences in morphology and size. Here we present some of the main light microscopy methods used to quantify different components of planktonic communities, such as virus, bacteria, algae and animals. Key words: Bacterioplankton, phytoplankton, limnology, enumeration, virioplankton, zooplankton.

1. Introduction The general understanding of the word plankton is referred to any drifting organism (virus, bacteria, plants or animals) that inhabit the pelagic zone of oceans, seas or bodies of freshwater. Planktonic community comprises organisms that are taxonomically diverse and more defined by their ecological niche rather than their genetic classification. It includes: • Virioplankton: the most diverse and abundant component of plankton, infecting a wide range of planktonic organisms. H. Chiarini-Garcia, R.C.N. Melo (eds.), Light Microscopy, Methods in Molecular Biology 689, DOI 10.1007/978-1-60761-950-5_13, © Springer Science+Business Media, LLC 2011

215

216

Soares et al.

• Bacterioplankton: morphological diverse group of prokaryotic organisms, which play an important role in remineralizing organic material down the water column. • Phytoplankton: autotrophic, prokaryotic or eukaryotic organisms that live near the water surface where there is sufficient light to support photosynthesis. • Zooplankton: small protozoans or metazoans (e.g. crustaceans and other animals) that feed on other plankton. Some of the eggs and larvae of larger animals, such as fish, crustaceans and annelids, are included here. These organisms are distributed along a wide range of size, varying from