Gold and Silver Staining: Techniques in Molecular Morphology (Advances in Pathology, Microscopy, & Molecular Morphology)

GOLD and SILVER STAINING:

Techniques in Molecular Morphology

Advances in Pathology, Microscopy & Molecular Morphology Series Editors Jiang Gu and Gerhard W. Hacker PUBLISHED TITLE Gold and Silver Staining: Techniques in Molecular Morphology Gerhard W. Hacker and Jiang Gu

Advances in Pathology, Microscopy & Molecular Morphology Series Editors Jiang Gu and Gerhard W. Hacker

GOLD and SILVER STAINING:

Techniques in Molecular Morphology

Edited by

GERHARD W. HACKER Forschungsinstitut fuer Grund- und Grenzfragen der Medizin und Biotechnologie St. Johanns Spital, Landeskliniken Salzburg Salzburg, Austria, EU

JIANG GU

Department of Biomedical Sciences University of South Alabama Mobile, Alabama

CRC PR E S S Boca Raton London New York Washington, D.C.

Library of Congress Cataloging-in-Publication Data Gold and silver staining : techniques in molecular morphology / Gerhard W. Hacker, Jiang Gu. p. cm. Includes bibliographical references and index. ISBN 0-8493-1392-9 (alk. paper) 1. Immunogold labeling 2. Silver staining (Microscopy) 3. Histochemistry. I. Hacker, Gerhard W. II. Gu, Jiang. QR187.I482 G65 2002 611′.018—dc21

2002017541 CIP

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. All rights reserved. Authorization to photocopy items for internal or personal use, or the personal or internal use of specific clients, may be granted by CRC Press LLC, provided that $1.50 per page photocopied is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA The fee code for users of the Transactional Reporting Service is ISBN 0-8493-1392-9/02/$0.00+$1.50. The fee is subject to change without notice. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com © 2002 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0-8493-1392-9 Library of Congress Card Number 2002017541 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acid-free paper

DEDICATION I would like to dedicate this book to the flower of my life, my wife Ursula Demarmels-Hacker. I thank her for her patience and encouragement during the months of preparation of this publication, and for her unceasing interest in my work—her smart questions are always a great source of inspiration to me. G.W.H.

v

FOREWORD Since time immemorial, gold and silver have fascinated humankind, whether in the form of costly objects of art or as a means of payment. In the last century, silver has found an application in photography and has been used in biology to visualize various tissue phenomena in a variety of staining processes. In some of these, gold has been used to enhance the silver reaction by heightening the contrast between the marker and the surrounding tissue. Until the late 1960s, most of these silver tissue staining processes were developed empirically. Unfortunately, the different factors in the staining processes were not always under complete control, which led to capricious staining results. However, advances in our knowledge of the chemical properties of silver ions have improved the reproducibility of silver stains. Until 1983, scientists used silver ions almost exclusively in histochemical staining, occasionally with gold chloride as an enhancing agent. Since then, the immunogold–silver staining technique has rendered increasing sensitivity. Silver staining techniques have also been used successfully to identify metallic elements in biological tissues. Applications for the colloidal labels with silver intensification have extended continuously to new fields and include not only cytochemistry and histochemistry but also in situ hybridization and immunoblotting techniques. The advantages of immunogold–silver staining include its superb sensitivity and excellent contrast. However, increasing sensitivity brings with it the risk of unspecific precipitations, which is why these processes need to be carried out with great precision. Consequently, a handy presentation of the various staining methods and up-to-date accounts of their individual advantages and drawbacks will be a boon to researchers. The editors of this book have made significant improvements to the immunogold–silver technique and helped popularize it for routine staining purposes. They must, therefore, be acknowledged for assembling in book form their own and other well-known researchers’ expertise in the field. This will facilitate both research planning and practical laboratory work. Lars Grimelius, MD, Ph.D. Departments of Genetics and Pathology Uppsala University, Uppsala, Sweden

vii

PREFACE The search for highly sensitive in situ antigen and nucleotide detecting techniques recently achieved a new milestone with the microscopic detection of single molecules. In 1981 and 1983, two genius scientists, Gorm Danscher (Denmark) and Clive Holgate (U.K.), successfully developed silver enhancement (autometallography [AMG]) and immunogold–silver staining. The latter method uses the very small particles of colloidal gold as the labeling marker to visualize the location of antibodies and nucleotide probes that have been bonded to specific antigens and nucleotide sequences in tissue sections or cytologic preparations. The initially tiny gold particles are highly uniform in size and are enlarged further by AMG, thereby precipitating metallic silver onto the gold label surface. Under the light microscope (LM), the enlarged particles are black and provide a highly contrasting colored marker, standing out distinctly against the unstained background structures. Under the electron microscope (EM), the gold or gold–silver particles are highly electron-dense and provide a clear marker for target detection at the ultrastructural level. Gold–silver staining is one of the few procedures that can demonstrate the same target molecules at both the LM and EM levels. It is one of the most sensitive techniques available today for low copy number antigen and nucleotide detection. This unique technique has tremendously enhanced the scientific field of molecular morphology. The detecting sensitivity of any immunostaining or in situ hybridization technique depends to a large degree on the signal-to-noise ratio. The larger and more visible the labeling signals are against the same background staining, the more sensitive the technique is. The ultimate achievement of these methods is to detect a single copy of a nucleotide sequence or a single molecule of an antigenic epitope on routinely processed tissue sections. In combination with tyramide signal amplification and comparable setups of label multiplication, gold–silver staining has more or less achieved this goal. In spite of the tremendous power and potential of AMG and gold–silver staining, its merits and applications have not been widely recognized. Different morphological labeling methods are being used by many laboratories; however, gold–silver staining is exceptional in many ways. The small unified sizes of the gold particle can be controlled and can facilitate different degrees of tissue penetration. The degree of particle enlargement by silver or gold salt is also controllable so that to some degree the detecting sensitivity can be adjusted. Gold–silver staining is easy to perform, and the reagents required are readily available. The black color of the end product provides distinct advantages for fast screening of the preparations; the particularity of the stain provides a very sharp and brilliant image. These and other merits of this technique make it a favorite choice of many scientists. The frontier of pathology is effectively moving into the age of molecular morphology. Simple morphologic patterns, provided by routine histochemical stains alone, are no longer adequate for depicting the subtle distinction between normal and diseased states, that can make significant differences in prognosis and therapy. Against this background, our goal is to provide the readers with the latest information about this extremely sensitive approach to microscopically localizing minute quantities of proteins or peptides and genetic material. This volume gives a timely overview and detailed description of different approaches and applications in chapters written by the leading authorities in this emerging field. The authors include such scientists as those who made the landmark contributions to the development of this technique. LM and EM silver staining, AMG, and heavy metal detection ix

are described by two living legends and their collaborators, Lars Grimelius and Gorm Danscher. The inventors of clustered gold covalently bound to macromolecules, Jim Hainfeld and collaborators, describe alternative techniques of silver and gold enhancement as well as their outstanding Nanogold® reagent and its many uses. Immunohistochemical and molecular morphological immunogold–silver staining are reviewed by Gerhard Hacker and Raymond R. Tubbs, both of whom are long-time advocates of the technique. Its original inventor, Osamu Fujimori, discusses his protein A–gold–silver staining method. Rick Powell presents a very promising reagent that combines a fluorescent label with gold–silver technology. Peter Jackson, one of the original inventors of immunogold–silver staining, introduces a new in situ hybridization method using a thermocycler designed for in situ polymerase chain reaction (PCR). Adalberto Merighi and his outstanding group of transmission electron microscopists contribute a chapter on ultrastructural applications of immunogold–silver staining. Hajime Sawada, John M. Robinson, and coauthors describe instructive protocols on pre- and post-embedding immuno-EM using Nanogold. Christian Schöfer and Klara Weipoltshammer, pioneers of EM tyramide signal-amplified gold–silver staining, instruct the reader about supersensitive EM immunocytochemistry and in situ hybridization. Finally, Bao-Le Wang and Jozef Sˇ amaj and coworkers discuss applications of gold–silver scanning electron microscopy in cancer research and other fields. The chapters in this book offer a balanced view of this emerging field. At the same time, they enable the readers to follow step-by-step protocols to make these procedures routine methods in their laboratories. This method allows the visualization of molecules that have never been localized before. With it, new discoveries follow naturally. The editors feel that far too many animals are being used for experimentation, and very often only to fulfill legal requirements, to retest already evaluated drugs, or simply to reach academic titles and degrees. It is the explicit wish of the editors to set an example with this book. Responsible researchers should learn to avoid such experiments whenever they can. Most contributors responded to our urge and used human tissue or cell cultures in conducting experiments whenever possible. We wish to thank them for their kindness towards our friends in the animal kingdom. We would like to thank the many distinguished contributors to this book for sharing their experience with the readers. Without their devotion to molecular morphology, this title, and indeed the advancement of this exciting field, would not have been possible. Gerhard W. Hacker Jiang Gu

x

ABOUT THE EDITORS Gerhard W. Hacker, Ph.D., is full professor of histochemistry, histology, and endocrinology at Salzburg University. He received his Ph.D. in biology and biochemistry at the Faculty of Natural Sciences of the University of Salzburg, Austria, and studied scientific medicine at the University of London, where he received the Diploma in Endocrinology and Pathology of the Royal Postgraduate Medical School. After further postdoctoral training and research periods in Uppsala and Stockholm and several institutions in the U.S., he was appointed to launch a special techniques unit for diagnostics and research at the Institute of Pathology of the Federal Hospital of Salzburg (now Landeskliniken Salzburg). He now heads a research institute on basic and borderline questions of medicine and biotechnology at the St. Johanns Hospital of the Landeskliniken Salzburg and of the University of Salzburg. His main research areas include the development of improved laboratory technologies, such as gold-silver staining, in situ single virus copy detection (in situ PCR), supersensitive in situ hybridization, neuropeptide research, neuroendocrine tumors, medical ethics, and palliative care. Jiang Gu, M.D., Ph.D., is full professor of biomedical sciences and full professor of cell biology and neuroscience at the University of South Alabama, Mobile, AL. He obtained his M.D. in China and was trained as a pathologist at the Beijing Medical University, Beijing, China. He then obtained a Ph.D. at the Department of Pathology, Royal Postgraduate Medical School, University of London. He received postdoctoral training in molecular biology at the National Institutes of Health, Bethesda, MD. He is now co-editor-in-chief of the journal Applied Immunohistochemistry and Molecular Morphology and has published numerous research articles and a number of books in the field of molecular morphology.

xi

CONTRIBUTORS Patrizia Aimar Department of Veterinary Morphophysiology Neurobiology Research Group Università degli Studi di Torino Torino, Italy Wilhelm Barthlott Botanisches Institut University of Bonn Bonn, Germany Elena Beltramo Department of Internal Medicine Università degli Studi di Torino Torino, Italy Annie L.M. Cheung Department of Anatomy University of Hong Kong Hong Kong, China Gorm Danscher Department of Neurobiology Institute of Anatomy University of Aarhus Aarhus C, Denmark

Frederic R. Furuya Nanoprobes Yaphank, NY, USA Lars Grimelius Department of Genetics and Pathology University of Uppsala Uppsala, Sweden Thomas Grogan Department of Pathology University of Arizona School of Medicine Tucson, AZ, USA Gerhard W. Hacker Forschungsinstitut fuer Grund– und Grenzfragen der Medizin und Biotechnologie St. Johanns Spital, Landeskliniken, Salzburg Salzburg, Austria, EU James F. Hainfeld Brookhaven National Laboratory Upton, NY, USA

Hans-Jürgen Ensikat Botanisches Institut University of Bonn Bonn, Germany

Cornelia Hauser-Kronberger Institute of Pathology Landeskliniken Salzburg Salzburg, Austria

Michiyo Esaki Department of Anatomy Yokohama City University School of Medicine Yokohama, Japan

Peter Jackson Department of Histopathology and Molecular Pathology The General Infirmary at Leeds Leeds, England, UK

Nina Flay Electron Microscope Center Finch University of Health Sciences The Chicago Medical School North Chicago, IL, USA

Fraser A. Lewis Department of Histopathology and Molecular Pathology The General Infirmary at Leeds Leeds, England, UK

Osamu Fujimori Department of Anatomy Nagoya City University Medical School Nagoya, Japan

Laura Lossi Department of Veterinary Morphophysiology Neurobiology Research Group Università degli Studi di Torino Torino, Italy

xiii

Adalberto Merighi Department of Veterinary Morphophysiology Neurobiology Research Group Università degli Studi di Torino Torino, Italy James Pettay Department of Clinical Pathology Cleveland Clinic Foundation Cleveland, OH, USA

Meredin Stoltenberg Department of Neurobiology Institute of Anatomy University of Aarhus Aarhus C, Denmark Toshihiro Takizawa Department of Anatomy Jichi Medical School Tochigi, Japan

Richard D. Powell Nanoprobes Yaphank, NY, USA

Raymond R. Tubbs Department of Clinical Pathology The Cleveland Clinic Foundation Cleveland, OH, USA

John M. Robinson Department of Physiology and Cell Biology Ohio State University Columbus, OH, USA

Dale D. Vandré Department of Physiology and Cell Biology Ohio State University Columbus, OH, USA

∨

Jozef Samaj Institute of Plant Genetics and Biotechnology Slovak Academy of Sciences Nitra, Slovak Republic Hajime Sawada Department of Anatomy Yokohama City University School of Medicine Yokohama, Japan Christian Schöfer Institute for Histology and Embryology University of Vienna Vienna, Austria

xiv

Dieter Volkmann Botanisches Institut University of Bonn Bonn, Germany Bao-Le Wang Abbott Laboratories Abbott Park, IL, USA Klara Weipoltshammer Institute for Histology and Embryology University of Vienna Vienna, Austria

EDITORIAL NOTE Inventors of Gold and Silver Staining Gorm Danscher (Denmark) and Clive Holgate (UK) can be regarded as the two original inventors of gold and silver staining technologies. With autometallography, also termed silver enhancement, Danscher laid the foundation for the incredible high sensitivity of the technique. Detection of colloidal gold (and other catalytic metals in tissue sections) was performed first by him, whereas Holgate and colleagues had the superb idea of combining the high amplifying potential of Danscher’s silver precipitation reaction with immunohistochemistry. Together, they initiated a large spectrum of the microscopic and other molecular techniques that are available today and are based on the very same detection principle.

Prof. Gorm Danscher

Prof. Clive Holgate

xv

CONTENTS 1

Silver Stains for the Identification of Neuroendocrine Cell Types . . . . . . . . . .1 L. Grimelius

2

Autometallographic Tracing of Gold, Silver, Bismuth, Mercury, and Zinc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 G. Danscher, G.W. Hacker, and M. Stoltenberg

3

Silver- and Gold-Based Autometallography of Nanogold® . . . . . . . . . . . . . . .29 J.F. Hainfeld and R.D. Powell

4

Immunogold–Silver Staining for Light Microscopy Using Colloidal or Clustered Gold (Nanogold®) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 G.W. Hacker, A.L.M. Cheung, R.R. Tubbs, L. Grimelius, G. Danscher, and C. Hauser-Kronberger

5

The Protein A–Gold–Silver Staining Method . . . . . . . . . . . . . . . . . . . . . . . . .71 O. Fujimori

6

Microscopic Uses of Nanogold® . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 J.F. Hainfeld, R.D. Powell, and F.R. Furuya

7

Combined Fluorescent and Gold Probes for Microscopic and Morphological Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .107 R.D. Powell and J.F. Hainfeld

8

Detection of Human Papillomavirus by In Situ Hybridization Using an In Situ PCR Thermal Cycler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119 P. Jackson and F.A. Lewis

9

Supersensitive In Situ Hybridization by Tyramide Signal Amplification and Nanogold® Silver Staining: The Contribution of Autometallography and Catalyzed Reporter Deposition to the Rejuvenation of In Situ Hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127 R.R. Tubbs, J. Pettay, T. Grogan, A.L.M. Cheung, R.D. Powell, J. Hainfeld, C. Hauser-Kronberger, and G.W. Hacker

10 Immunogold Labeling Techniques for Transmission Electron Microscopy: Applications in Cell and Molecular Biology . . . . . . . . . . . . . . . . . . . . . . . . .145 L. Lossi, P. Aimar, E. Beltramo, and A. Merighi 11 Pre-Embedding Immunoelectron Microscopy with Nanogold® Immunolabeling, Silver Enhancement, and Its Stabilization by Gold . . . . . .169 H. Sawada and M. Esaki 12 Gold Cluster Immunoprobes: Light and Electron Microscopy . . . . . . . . . . .177 J.M. Robinson, T. Takizawa, and D.D. Vandré 13 Highly Sensitive Ultrastructural Immunogold Detection Using Tyramide and Gold Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .189 C. Schöfer and K. Weipoltshammer

xvii

14 In Situ Hybridization at the Electron Microscopic Level . . . . . . . . . . . . . . .199 C. Schöfer and K. Weipoltshammer 15 Immunogold–Silver Staining for Scanning Electron Microscopy in Cancer Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .211 B.-L. Wang, N. Flay, and G.W. Hacker 16 Immunogold–Silver Scanning Electron Microscopy Using Glycerol Liquid Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .223 ∨ J. Samaj, H.-J. Ensikat, W. Barthlott, and D. Volkmann Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .235

xviii

1

Silver Stains for the Identification of Neuroendocrine Cell Types Lars Grimelius

INTRODUCTION Silver stains have been widely applied to identify various cell types and tissue components in routinely processed tissue sections, though they have declined somewhat in importance since the introduction of immunohistochemical techniques. However, some silver techniques are still useful for both histology and histopathology. In this chapter, three silver staining methods— Masson, Grimelius, and Sevier–Munger, useful for identification of neuroendocrine cells—are commented on, their achievements reported, and their chemical backgrounds, when known, are briefly reviewed. For the Masson and Grimelius techniques, microwave irradiation procedures, which allow the completion of the staining in only a few minutes, are outlined. Silver stains have been widely used in both histology and histopathology for the past century to identify various cell types and other tissue components. They have also been applied to identify not only various chemical compounds, including heavy metals and calcium deposits, but also mitotic figures, as well as bacteria, fungi, and spirochetes, among others.1,4,7 Almost all these silver stains have been empirically 0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

developed, but the chemical background is known only for some. Until late 1950s, most of the silver methods had, at least for the demonstration of endocrine cells, a less than good reputation due to capricious staining results and unwanted silver precipitation, which resulted from uncontrolled staining factors, such as the pH of the solutions and the concentrations of ionized silver. Silver stains were therefore continually modified in order to improve their reproducibility. During the 1960s, new silver stains had been developed, giving reproducible results. Some of them, intended to visualize endocrine cells, are still in use. During the last two decades, silver has attracted increasing interest as a label enhancer in immunohistochemistry, in situ hybridization, and immunoblotting,15 in which silver staining processes are used to increase the sensitivity of these techniques. In Chapter 2, Gorm Danscher, the father of autometallography (AMG), describes the use of his silver staining techniques to detect catalytic tissue metals; AMG is also the basis of silver-enhancing colloidal or clustered gold label in immunohistochemistry. The silver staining methods used in conventional histochemistry can be divided in

1

Gold and Silver Staining two categories, argentaffin and argyrophil, but there are others, which fall between these two. Argentaffin methods are based on a physical developer, i.e., a staining solution containing a silver salt, which is a reducing agent, and often also an organic colloid to keep the silver ions and reducing molecules apart. The argyrophil type of silver stains uses a chemical reducer, i.e., a reducing solution that is applied to tissue sections after incubation in the silver solution. The end product in both staining processes is reduced silver. As the different methods for argentaffin and argyrophil staining visualize different tissue structures, one should always name the exact technique applied. Most silver stains were initially developed to demonstrate various nerve structures, but it was found that some were also useful for identifying different endocrine cell types. This relationship between the nervous system and several members of the endocrine cell system has become more obvious during the last three decades, as the nervous system shares histochemical and biochemical properties with several members of the endocrine system. The endocrine cell types displaying nervous properties are now called neuroendocrine (NE). The NE cell system includes the endocrine cells of the gastrointestinal and respiratory tract, in the pancreas, pituitary gland (anterior lobe), thyroid (C cells), parathyroid, adrenal gland (medulla), and paraganglia. Skin (Merkel cells), thymus, kidney, cervix uteri, ovary, prostate gland, and testis also contain NE cells. Most of these cell types are of endodermal origin, most likely derived from multipotent stem cells, but the endocrine cells of the adrenal medulla and paraganglia are of neuroectodermal origin, as also may be the C cells of the thyroid gland. Since the introduction of immunohistochemical techniques, the use of silver stains 2

has declined. However, three silver stains are still useful for demonstrating normal and neoplastic NE cells; namely the Masson,20 the Grimelius,8,9 and the Sevier–Munger.23 The Masson stain belongs to the argentaffin category, the Grimelius to the argyrophil, and the Sevier–Munger falls in between. All these stains are based on commercial, readily available, and well-known chemical substances, and the results are reproducible. The Grimelius stain is a broad spectrum NE marker, the Masson silver stain visualizes the enterochromaffin (EC; serotonin) cells, and the Sevier–Munger technique visualizes the EC and EC-like (histamine) cells. All three staining techniques work well on formaldehyde-fixed paraffin-embedded tissue sections and also fixed cytological specimens, but they do not work well on unfixed frozen sections. Fixatives containing heavy metals cause unspecific silver precipitation, and these fixatives should therefore be avoided here. In these three silver methods, the staining is caused by deposition of pure silver within secretory granules.10,26,27 Frequency and distribution of the silver grains vary to some extent in the various endocrine cell types. The staining procedures take from 1 to 3 h, but by applying microwave irradiation these can be shortened to a few minutes without adversely affecting the staining quality. In the present chapter, these three silver stain methods are described and commented on. THE MASSON TECHNIQUE The Masson stain20 has been modified several times, with the best-known variants being those described by Hamperl14 and Singh.24 In 1986, Portela-Gomes and Grimelius21 described a simple modification, whereby the sections are stained for

Silver Stains for the Identification of Neuroendocrine Cell Types 60 min in a preheated (60°C) 5% ammoniacal silver solution, followed by rinsing and mounting. A similar staining reaction can be obtained by using a 1% ammoniacal silver solution and microwave irradiation;12 the outcome is, however, to some extent dependent on the type of microwave oven used. (For details, see section below on Staining Protocols.) Barter and Pearse2,3 found firm evidence that it was the reaction product of serotonin with paraformaldehyde that caused the silver reaction. Lundqvist et al.18 demonstrated by a dot-blot technique that dopa, dopamine, noradrenalin, adrenalin, and also 5-hydroxytryptamine (serotonin), gave rise to the argentaffin reaction. The Masson stain visualizes the EC cells in the gastrointestinal mucosa (Figure 1.1), as well as in their related tumors (so-called

classic carcinoids) located mainly in the midgut (Figure 1.2). This technique also stains melanin pigment. The presence of serotonin can be demonstrated in a more specific way by immunohistochemical methods. Comparison of the occurrence and distribution of argentaffin and serotonin-immunoreactive cells in midgut carcinoids revealed, however, more silver-positive than serotoninimmunoreactive cells.19 An explanation for this difference could be that the argentaffin serotonin-negative cells contain chemical substances other than serotonin, thus causing the silver reaction in these cells too. Serotonin immunoreactivity has been demonstrated in other NE cells than the EC cells, e.g., secretin5 and polypeptide YY (PYY)17 cells. The argentaffin reaction in these two NE cell types may depend on the presence of some biogenic amine.

Figure 1.1. Human intestinal mucosa, showing argentaffin (serotonin) cells. Masson stain. (Original magnification, ×180.)

3

Gold and Silver Staining THE GRIMELIUS TECHNIQUE The Grimelius technique8 was initially devised to demonstrate a noninsulin cell fraction of human pancreatic islets, later identified as glucagon and pancreatic polypeptide (PP) cells. The sections were incubated in a 0.03% silver nitrate solution (pH 5.6) at 37°C for 24 h or at 60°C for 3 h, followed by reduction in an aqueous hydroquinone-sodium sulfite solution at 45°C for 1 min. When the staining technique was applied to gastrointestinal mucosa to demonstrate NE cells, a more intense argyrophil reaction was obtained when the silver nitrate concentration was increased to 0.07%. A further intensification of staining was seen when the temperature of the reducing solution was increased to 55° to 58°C.11 By using microwave irradiation in both incubation

steps, the staining procedure can be shortened to 10 min.12,16 (For staining details, see section below on Staining Protocols.) The argyrophilic reaction occurs in most NE cell types (Figure 1.3) except cholecystokinin (CCK), insulin, and somatostatin cells.11 The silver reaction also occurs in most NE tumors (Figure 1.4). Some insulinomas display an argyrophil reaction,6 whereas the somatostatinomas generally are nonargyrophil, as are the hindgut carcinoid tumors, i.e., NE tumors located in distal colon and rectum. Rindi et al.22 demonstrated, with a dotblot technique, that it is chromogranin A, a glycoprotein related to the secretory granules, that gives rise to the Grimelius argyrophilic reaction. This finding was confirmed by Lundqvist et al.,18 who also observed using the same technique that dopamine, noradrenalin, and 5-hydroxy-

Figure 1.2. Photomicrograph of midgut (ileal) carcinoid. Most neoplastic cells display argentaffin reaction strongest in the periphery of the tumor cell groups. Masson stain. (Original magnification, ×90.)

4

Silver Stains for the Identification of Neuroendocrine Cell Types tryptamine (serotonin) also elicited the Grimelius silver-positive reaction. By sequential staining of pancreatic islets and intestinal mucosa with an immunohistochemical technique and silver staining of the same tissue section, it was found that chromogranin A immunoreactivity and the Grimelius argyrophilic reaction occur in the same cells.18 THE SEVIER–MUNGER TECHNIQUE The Sevier–Munger technique23 was initially developed to demonstrate neural tissue, but the staining method also visualizes the enterochromaffin-like (ECL) and D1 cells of the human gastric mucosa and the EC cells of the gastrointestinal mucosa and also the gastric inhibitory peptide

(GIP) cells of the intestinal mucosa and the C cells of the thyroid.11 The sections were initially incubated in a 20% aqueous silver nitrate solution, while the subsequent staining procedure takes part in a physical developer, i.e., a solution containing both silver ions and a reducing agent. (For details, see section below on Staining Protocols.) The ECL cells (Figure 1.5) and tumors derived from this cell type (ECLomas) (Figure 1.6) contain the biogenic amine histamine. This amine can be demonstrated with immunohistochemistry, but the histamine antibodies available work best in tissue fixed in formaldehyde-carbodiimide fixative or in freeze-dried tissue, vapor-fixed in diethylpyrocarbonate (DEPC).13,25 In sections from formalin-fixed paraffin-embedded tissue, the histamine immunoreactivity is weak or negative. The ECL cells can, from a practical point of view, be identified indirectly by

Figure 1.3. Human pancreatic islet stained with the Grimelius method. The argyrophil cells are PP and glucagon cells. (Original magnification, ×300.)

5

Gold and Silver Staining combining the Sevier–Munger and Masson methods on sequential sections. The ECL cells take the Sevier–Munger stain, but fail to show argentaffin (Masson) reaction. Weak argyrophilic reaction with the Sevier–Munger method has also been reported in pancreatic PP cells. The chemical background to the Sevier–Munger stain is still not known. STAINING PROTOCOLS Protocol 1: Masson Staining: Modified by Portela-Gomes and Grimelius21 Ammoniacal silver solution is prepared by dissolving 5 g silver nitrate in 100 mL glass-redistilled water and then adding am-

monium hydroxide drop by drop until the brown-black precipitation disappears. Then, add a few more drops of 5% silver nitrate solution until the solution turns slightly cloudy. 1. Rinse the sections thoroughly in deionized water and then in double-distilled water. 2. Preheat the ammoniacal silver solution to 60°C and immerse the sections in this solution for 5 to 20 min (check the staining in the light microscope and interrupt at optimal intensity). 3. Rinse the sections in water. 4. If necessary, sections can be counterstained with nuclear fast red or methyl green. 5. Mount sections in DPX (BDH Bio-

Figure 1.4. Midgut carcinoid (same as in Figure 1,2) showing argyrophil tumor cells. Some argyrophil cells are seen in the adjacent mucosa. Grimelius stain. (Original magnification, ×90.)

6

Silver Stains for the Identification of Neuroendocrine Cell Types chemicals, Poole, UK) or Cytoseal mounting medium (Curtin Matheson Scientific, Wayne, NJ, USA) or Canada balsam. Protocol 2: Masson Staining—Microwave Procedure: Grimelius et al.12 In the Masson staining procedure, prepare the ammoniacal silver solution as described above in Protocol 1, but use 1 g silver instead of 5 g. The solution should be slightly cloudy. Filter the solution into a thoroughly cleansed plastic Coplin jar. 1. Rinse the sections thoroughly in deionized water and then in double-distilled water. 2. Immerse the deparaffinized sections in the silver solution and irradiate in a microwave oven (e.g., M696; Miele

Electronic, Bünde, Germany) at 450 W for 1 min. Do not put more than 8 to 10 slides into 100 mL of silver solution. 3. Allow the sections to remain in the hot silver solution for a few minutes until they turn a light golden color. 4. Rinse the sections well, dehydrate, clear in xylene, and mount. Protocol 3: Grimelius Staining: Modified by Grimelius and Wilander, 19808,11 Silver Solution Dissolve 70 mg of silver nitrate in 100 mL of Walpole’s acetate buffer solution, pH 5.6, diluted 1:10 (buffer preparation: 48 mL of 0.2 mol/L acetic acid plus 452 mL of 0.2 mol/L sodium acetate in 500 mL glass double-distilled water). The

Figure 1.5. Human gastric (oxyntic) mucosa showing argyrophil ECL cells. Sevier–Munger stain. (Original magnification, ×180.)

7

Gold and Silver Staining buffer alone can be stored for months in the refrigerator at (4°–8°C). Reducing Solution Dissolve 1 g hydroquinone and 5 g sodium sulfite (anhydrous) in 100 mL of distilled water. The silver and reducing solutions should both be freshly prepared. Formaldehyde and Bouin’s fluid are the best fixatives for the Grimelius stain. Glutar(di)aldehyde, ethanol, or fixatives containing ethanol cause a weak argyrophil reaction, or the staining may fail. Heavy metals in the fixative cause silver precipitation on the sections. 1. Rinse the sections thoroughly in deionized water and then in double-distilled water. 2. Immerse the sections in the silver solution at room temperature and transfer

them later to an oven maintaining a temperature of 60°C for 3 h. Do not stain more than 6 to 8 glass slides per 100 mL of silver solution. 3. Wipe dry the glass slides around the sections (the sections themselves should not be allowed to dry out). 4. Transfer sections to a prewarmed (57° to 58°C) reducing solution for a minimum of 1 min. 5. Rinse the sections, dehydrate, and mount in Canada balsam, Cytoseal, or DPX (avoid Eukitt [BDH Biochemicals], Entellan [Merck, Darmstadt, Germany], or glycerin/gelatin as mounting media, as they could remove the silver grains). If the argyrophilic reaction appears too weak, it can be enhanced in the following way: transfer the well-rinsed sections to a

Figure 1.6. Solitary gastric carcinoid (ECLoma) localized in the oxyntic mucosa displaying argyrophil cells of varying intensity. Sevier–Munger stain. (Original magnification, ×180.)

8

Silver Stains for the Identification of Neuroendocrine Cell Types freshly prepared silver solution (see above) at room temperature for 10 min. Wipe the glass slides around the sections as above and immerse them in a freshly prepared prewarmed (57° to 58°C) reducing solution (see above) for 1 min. Rinse the sections, dehydrate, clear in xylene, and mount. Counterstain, if necessary, with methyl green or nuclear fast red. Protocol 4: Grimelius Staining—Microwave Procedure: Kok and Boon16 Silver Solution Dissolve 1 g silver nitrate in 90 mL glass redistilled water and 10 mL of Walpole’s acetate buffer solution, pH 5.6. Reducing Solution As in Protocol 3. 1. Rinse the sections thoroughly in deionized water and then in double-distilled water. 2. Immerse the sections in 16 mL of silver solution and microwave at 450 W (95°C) for about 60 sec. 3. Rinse the sections in distilled water for a few seconds. 4. Transfer sections to 16 mL of the reducing solution (described in Protocol 3). 5. Microwave sections in reducing solution at 450 W (95°C) for 60 sec. 6. Rinse sections carefully in double-distilled water. The sections should retain their yellow-brown color. 7. If the staining reaction is too weak, repeat the process (steps 2–5) after rinsing the sections in redistilled water. (The same solutions can be reused after filtration.) 8. Rinse carefully in water and, if necessary, counterstain with 0.2% nuclear fast red for 3 min. 9. Dehydrate, clear in xylene, and mount

in Canada balsam, DPX, or Cytoseal. Protocol 5: Sevier–Munger Staining: Modified by Grimelius and Wilander11 Silver Solution Dissolve 10 g silver nitrate in 50 mL glass double-distilled water and filter into a Coplin jar. Ammonium Hydroxide-Silver Solution Add 28% to 30% (concentrated) ammonium hydroxide drop by drop to 50 mL of a 10% silver nitrate solution until the dark brown precipitate has almost disappeared. Add 0.5 mL of a sodium carbonate solution [prepared by dissolving 8 g Na2CO3.(10 H2O) in 30 mL glass double-distilled water] and shake well. Add a further 25 drops of concentrated ammonium hydroxide during vigorous shaking. The solution should be clear. Filter the solution into a Coplin jar. Add 10 drops of 0.8% formaldehyde (2 mL concentrated formalin plus 98 mL distilled water) immediately before use. The best fixatives for this staining are formaldehyde and Bouin’s fluid. 1. Preheat the 20% silver solution to 60°C and leave the deparaffinized sections at that temperature for 15 min. 2. Rinse the sections thoroughly in tap and deionized water and then in double-distilled water. 3. Transfer them to the Coplin jar containing the ammoniacal silver formalin solution. 4. The Coplin jar should then be placed in a hot oven (60°C) for 15 to 30 min. Check the staining under the light microscope and interrupt the process when the cells appear black. 5. Treat the sections in 5% sodium thiosulfate for 2 min. 6. Rinse the sections well in distilled water. 9

Gold and Silver Staining Counterstaining is usually not necessary. 7. Dehydrate, clear in xylene, and mount in DPX, Cytoseal, or Canada balsam. TECHNICAL HINTS FOR SILVER STAINING PROCEDURES Glass double-redistilled water should be used to prepare silver solutions to avoid unspecific silver precipitation on the sections. Fresh preparations of silver and reducing solutions are recommended. Glassware should be thoroughly cleaned. The choice of mounting medium is important, as some media withdraw silver from already well-stained sections. DPX, Cytoseal, Canada balsam, Permount, and Euparal are all suitable mounting media, whereas Eukitt, Pertex (Histolab Products AB, Göteborg, Sweden), Entellan, and glycerine/gelatin remove silver. The background staining with the present silver methods is usually sufficient for orientation in the tissue sections. The Masson method gives a yellow-tan background; the Grimelius method gives a yellow to greenish-yellow background; and the Sevier–Munger method gives a golden-brown coloration. For the Masson and Grimelius methods, nuclear fast red or light green (0.5% aqueous) staining can, if necessary, be used. REFERENCES 1.Bancroft, J.D. and A. Stevens. 1990. Theory and Practice of Histological Techniques. Churchill Livingstone, New York. 2.Barter, R. and A.G.E. Pearse. 1953. Detection of 5hydroxytryptamine in mammalian enterochromaffin cells. Nature 171:810. 3.Barter, R. and A.G.E. Pearse. 1955. Mammalian enterochromaffin cells as the source of serotonin (5-hydroxytryptamine). J. Pathol. Bacteriol. 69:25-31. 4.Busch, C. and J. Vasko. 1988. Differential staining of mitoses in tissue sections and cultured cells by a modified methenamine-silver method. Lab. Invest. 59:876-

10

878. 5.Cetin, Y. 1990. Secretin cells of the mammalian intestine contain serotonin. Histochemistry 93:601-606. 6.Creutzfeldt, W. 1975. Pancreatic endocrine tumors—the riddle of their origin and hormone secretion. Isr. J. Med. Sci. 11:762-776. 7.Danscher, G. 1984. Autometallography. A new technique for light and electron microscopic visualization of metals in biological tissue (gold, silver, metal sulphides, and metal selenides). Histochemistry 81:331-335. 8.Grimelius, L. 1968. A silver nitrate stain for A2 cells of human pancreatic islets. Acta Soc. Med. Ups. 73:243270. 9.Grimelius, L. 1968. The argyrophil reaction in islet cells of adult human pancreas studied with a new silver nitrate procedure. Acta Soc. Med. Ups. 73:271294. 10.Grimelius, L. 1969. An electron microscopic study of silver stained adult human pancreatic islet cells, with reference to a new silver nitrate procedure. Acta Soc. Med. Ups. 74:28-48. 11.Grimelius, L. and E. Wilander. 1980. Silver stains in the study of endocrine cells of the gut and pancreas. Invest. Cell Pathol. 3:3-12. 12.Grimelius, L., H. Su, and G.W. Hacker. 1994. The use of silver stains in the identification of neuroendocrine cell types, p. 1-8. In J. Gu and G.W. Hacker (Eds.), Modern Methods in Analytical Morphology. Plenum Press, New York. 13.Håkansson, R., G. Böttcher, E. Ekblad, P. Panula, M. Simonsson, M. Dahlsten, T. Hallberg, and F. Sundler. 1986. Histamine in endocrine cells in the stomach. A survey of several species using a panel of histamine antibodies. Histochemistry 86:5-17. 14.Hamperl, H. 1952. Über argyrophile Zellen. Virchows Arch. Path. Anat. 321:482-507. 15.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird. 1983. Immunogold–silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31:938-944. 16.Kok, L.P. and M.E. Boon. 1992. Microwave Cookbook for Microscopists; Art and Science of Visualization, p. 204. Coulomb Press Leyden, Leiden. 17.Lukinius, A.I.C., J.L.E. Ericsson, M.K. Lundqvist, and E.M.O. Wilander. 1986. Ultrastructural localization of serotonin and polypeptide YY (PYY) in endocrine cells of the human rectum. J. Histochem. Cytochem. 34:719-726. 18.Lundqvist, M., H. Arnberg, J. Candell, M. Malmgren, E. Wilander, L. Grimelius, and K. Öberg. 1990. Silver stains for identification of neuroendocrine cells. A study of the chemical background. Histochem. J. 22:615623. 19.Lundqvist, M. and E. Wilander. 1984. Small intestinal chromaffin cells and carcinoid tumours: a study with silver stains, formalin-induced fluorescence and monoclonal antibodies to serotonin. Histochem. J. 16:12471256. 20.Masson, P. 1914. La glande endocrine de l’intestin chez l’homme. C.R. Acad. Sci. Paris 15:59-61. 21.Portela-Gomes, G.M. and L. Grimelius. 1986. Identi-

Silver Stains for the Identification of Neuroendocrine Cell Types fication and characterization of enterochromaffin cells with different staining techniques. Acta Histochem. 79:161-174. 22.Rindi, G., R. Buffa, F. Sessa, O. Tortora, and E. Solcia. 1986. Chromogranin A, B, and C immunoreactivities of mammalian endocrine cells. Distribution, distinction from co-stored hormones/pro-hormones and relationship with the argyrophil component of secretory granules. Histochemistry 85:19-28. 23.Sevier, A.C. and B.L. Munger. 1965. A silver method for paraffin sections of neural tissue. J. Neuropathol. Exp. Neurol. 24:130-135. 24.Singh, I. 1964. A modification of the Masson-Hamperl

method for staining argentaffin cells. Anat. Anz. 115:81-82. 25.Solcia, E., C. Capella, G. Vassallo, and R. Buffa. 1975. Endocrine cells of the gastric mucosa. Int. Rev. Cytol. 42:223-286. 26.Vassallo, G., C. Capella, and E. Solcia. 1971. Endocrine cells of the human gastric mucosa. Z. Zellforsch. 118:49-67. 27.Vassallo, G., C. Capella, and E. Solcia. 1971. Grimelius silver stain for endocrine cell granules, as shown by electron microscopy. Stain Technol. 46:7-13.

11

2

Autometallographic Tracing of Gold, Silver, Bismuth, Mercury, and Zinc Gorm Danscher, Gerhard W. Hacker, and Meredin Stoltenberg

INTRODUCTION Catalytic clusters, consisting of only a few atoms of gold or silver or less than 10 formula units of silver–, mercury–, bismuth–, or zinc sulfide/selenide, can be silver-amplified by an autometallographic (AMG) developer to sizes at which they can be detected in the microscope. With the AMG technique, colloidal or clustered gold, used as labels of macromolecules such as enzymes or antibodies and streptavidin [in this context called immunogold–silver staining (IGSS)], are made visible at electron microscopic (EM) and light microscopic (LM) levels. A specific AMG recipe has been worked out for each metal, and if tissue sections are believed or known to contain a mixture of two or more of the above AMG catalysts, it is feasible to distinguish one from the other. In such cases it is recommended to include a multi-element analysis, e.g., proton-induced X-ray emission (PIXE), in AMG studies. The AMG technique is rooted in the photographic world of the 19th century. It was found at that time that exposed photographic plates, from which the silver bromide crystals had been removed, could not be developed by an ordinary photographic 0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

developer—a so-called chemical developer. It was, however, later observed that such plates still contained a latent image that could be silver-amplified if silver ions were added to the chemical developer. Somewhat misleadingly, this kind of photographic development was termed physical development because it was believed that the process differed from the normal photographic development, in that the silver ions were reduced to silver atoms close to, but not on the surface of, the catalyst and therefore had to move physically after the reduction. The technique was introduced into histology by Liesegang28 primarily as a new silver staining technique for impregnation of tissue sections, and he used the photographic term physical development for the process. Later, the term autometallography was suggested.12 Liesegang also tried to trace silver in exposed animals by silver enhancement, but failed.29 Roberts was more successful in observing a difference in the number of silver grains among tissue sections from animals treated with gold salts and from control animals.44 His work was followed by several important attempts to trace gold in exposed tissues,43 but it was not until Timm’s introduction of his

13

Gold and Silver Staining famous sulfide silver method in 1958 and his mercury method in 196259,60 that the field really gained scientific momentum and began to expand. Timm’s two methods have since proved very valuable and successful. They have resulted in a multitude of publications and have been subjected to a wealth of technical modifications. In 1969, Umeda et al. conducted an autoradiographic study using 203Hg2O.62 In their control sections, i.e., tissue sections from animals exposed to nonradioactive mercury, they thought that they could observe silver grains in the autoradiographic emulsion that covered the sections.62 The authors suggested that some unknown radiation had caused the autoradiographic pattern or, alternatively, that mercury had moved from the tissue section into the emulsion. In 1985, it could be shown that the observed staining was caused by AMG silver enhancement of catalytic clusters in the mercury-containing sections, i.e., the silver grains were located not in the autoradiographic emulsion, but in the underlying tissue section.15 The explanation of the phenomenon was that, while traveling through the autoradiographic emulsion, the chemical developer became enriched with silver ions released from dissolved silver bromide crystallites of the autoradiographic emulsion. In this way, the chemical developer had been transformed into an AMG developer, containing both reducing molecules and silver ions. When penetrating into the underlying tissue section, the AMG developer caused silver enhancement of catalytic HgS clusters. On the other hand, if the tissue sections contain radioactive mercury (203Hg), the above technique will result in slides containing both autoradiographic and AMG information, i.e., silver grains in the emulsion resulting from the radiation and AMG silver grains in the sections resulting from the mercury sulfide/selenide clusters.12,15 14

If differences appear between the autoradiographic pattern and the AMG pattern, they reflect the presence of mercury not bound in clusters by sulfide/selenide ions. At least in principle, such differences could be detected by subtraction of the pattern before and after removal of the autoradiographic emulsion. However, in the real world, the autoradiographic pattern is always diffuse and of little or no analytic help. The informative autoradiographic studies on mercury have always been AMG studies. The only other way to obtain information about the distribution of radioactive mercury in tissue sections is by placing an X-ray film on top of the sections and after a period of exposure removing and developing the film.61 Today, specific histochemical AMG methods exist for tracing atomic clusters of metallic gold and silver and clusters of sulfides and selenides of silver, mercury, and zinc. An AMG technique for detecting bismuth is presently being worked out in full. Copper sulfide/selenide clusters are not created in vivo. Neither does transcardial perfusion with sulphide or selenide ions result in such clusters. Copper sulfide clusters created as the end-product of the cuprolinic blue method can, however, be AMG enhanced.30 Gold ions chemically bound in tissue sections, from, for instance, goldaurothiomalate-exposed humans, have to be reduced to metallic gold, e.g., by UV light, before they can act as AMG catalysts. Apparently, gold sulfide/selenide is not formed by the organism as is the case with silver, mercury, and bismuth, which are all metabolized to sulfide/selenide clusters.9 As shown in the numerous applications reviewed and proposed in the present book, colloidal or clustered gold particles, e.g., used as markers of RNase,16,23 or other macromolecules, such as immunoglobulins or streptavidin, can be directly AMG silver enhanced.

Autometallographic Tracing The AMG selenium method was worked out for localizing in vivo bound zinc ions at LM and EM levels. In animals injected intraperitoneally or intravenously with selenium-containing molecules like sodium selenite, the selenite ions will be reduced to selenide ions in the organism and bound to zinc ions present in, e.g., synaptic vesicles of zinc-enriched (ZEN) neurons.11 The method originates from the observations that mercury, sulfur, and selenium were colocalized in the lysosomes of cerebellar Purkinje cells of a human brain from Minamata,51 and that Purkinje cells in experimental animals contained AMG mercury. Furthermore, silver selenide was shown to be present in the glomerular basement membrane of human argyrotic kidneys,1 and simultaneously, AMG grains were demonstrated in the same basement membrane in kidneys from silver-exposed rats,10 suggesting that certain metal selenide clusters might have the same capacity as the analogous sulfides to catalyze the AMG silver enhancement process. The selenium method demonstrates zinc ions in the terminals of ZEN neurons and secretory ZEN vesicles in several endocrine and exocrine glands following in vivo injections of sodium selenite. Compared to Timm’s sulfide silver method, the selenium method has the extra advantage that zinc selenide is more stable, i.e., less soluble, than zinc sulfide. This chemical quality has been exploited to trace somata of ZEN neurons. After in vivo formation of catalytic clusters of zinc selenide in ZEN terminals, either by intracerebral or intraperitoneal injection of sodium selenite or sodium selenide, some of the zinc selenide clusters formed in synaptic vesicles are transported retrogradely and accumulate in the lysosomes of ZEN neurons. After AMG silver enhancement of these zinc clusters, ZEN neuronal somata can be recognized. ZEN nuclei that have been traced in this way include the terete hypo-

thalamic nucleus and the strio hypothalamic nucleus.11,17,39,41,53 The latest improvements of Timm’s original sulfide silver method are the in vivo sulfide method13 and the biopsy H2S gas method.18 Chemical Background of AMG When the AMG developer permeates a section containing catalytic spikes, both the reducing molecules, hydroquinone (Hq) and silver ions (Ag+) are adsorbed to the surface of these metal-containing clusters. The transfer of electrons from Hq molecules through the clusters to adhering silver ions causes a reduction of the latter to metallic silver atoms (Ago). In this way, the catalytic clusters are encapsulated in metallic silver. As metallic silver is catalytic to the process, this will continue until stopped, e.g., by a stop bath of thiosulfate, whereby all the still free silver ions are removed from the section (Figure 2.1). Localization of Gold by AMG The well-known fact that patients in gold therapy are advised to avoid direct sunlight gave rise to the AMG gold technique. By applying UV light to tissue sections from a rat treated with aurothiomalate, gold ions were reduced to metallic gold atoms and subsequently AMG developed.9,12 The AMG gold method, being a highly sensitive technique for demonstrating nanometer sized gold particles created by radiating the tissue from gold-exposed organisms with UV light, was soon applied to AMG enhancement of colloidal gold particles used as markers of macromolecules.16,20,21,23 AMG-developed brain sections from rats treated with aurothiomalate compounds have demonstrated that in some cases chemically bound gold can penetrate the blood–brain barrier. Tanycytes of the 15

Gold and Silver Staining third ventricle and hypothalamic glia cells seem to be the first cells in the CNS in which gold has been demonstrated.9 Recently, gold has been detected in the lysosomes of sensory neurons and in the satellite cells of spinal ganglia in aurothiomalate-treated rats.50 Gold also accumulates in endocrine organs like the anterior pituitary, adrenals, 9,42 thyroid gland, and ovaries. Goldexposed rats accumulate gold in macrophages, e.g., located in liver, testis, or meninges, and so do peritoneally harvested mouse macrophages exposed in vivo or in vitro to aurothioglucose or aurothiomalate. Ultrastructurally, the gold-containing AMG grains are localized in lysosome-like organelles and in secretory granules.9 Transplacental transport of gold has been demonstrated by AMG in rats33 as well as in man. Furthermore, gold has been found in the ovaries of rats exposed to aurothiomalate, in synovial biopsies from patients in chemotherapy,19 and in synovial fluid cells.26

AMG Silver Enhancement of Gold Particles Used as Markers of Macromolecules As it had been proved that metallic gold clusters were the catalytic centers for AMG development,9 it was obvious to apply AMG to silver enhance colloidal gold particles used as markers of macromolecules introduced a few years before by Bendayan.6 In 1983, Holgate et al.23 and Danscher and Nørgaard16 simultaneously introduced AMG silver enhancement of colloidal gold particles bound to antibodies and enzymes, respectively. Both groups used the silver lactate developer, and Holgate et al. called their approach the immunogold–silver staining (IGSS) technique. Danscher and Nørgaard used the developer to magnify colloidal gold bound to RNase and included procedures for EM studies (Figure 2.2). As the silver lactate developer is somewhat light sensitive and works best if the AMG setup is covered most of the time

Figure 2.1. Schematic representation of autometallograpy. A camera lucida view of the AMG process at an early stage, showing how the AMG catalytic Hg2Se cluster (Se- selenide ions, and Hg++ mercury ions) is immured in a shell of metallic silver. Both the reducing agent hydroquinone, Hq, and silver ions, Ag+, are adsorbed to the surface of the catalytic clusters. The transfer of electrons from Hq through the clusters to adhering silver ions causes a reduction of Ag+ to metallic silver atoms, Ag°.

16

Autometallographic Tracing under a hood, we designed an alternative AMG developer based on silver acetate to be used when development has to take place all the way in daylight or under the microscope.20,21 Since then, many different AMG developers have been worked out,55 including several commercial enhancement kits. Lately, an AMG procedure that is based on enhancement with gold ions instead of silver ions has been introduced22 (GoldEnhance; Nanoprobes, Yaphank, NY, USA). According to our tests, gold AMG leads to a perfect enhancement of gold clusters and colloidal gold particles. It is fair, therefore, to talk about silver AMG and gold AMG depend-

ing on what noble metal is used to amplify the metal markers. Autometallography has become a worldwide applied cotechnique for enhancement of colloidal or clustered gold particles used as markers of various macromolecules. In addition to colloidal gold, a new type of gold label, Nanogold, is now commercially available (Nanoprobes; Web site: http://www. nanoprobes.com/). This label permeates more easily because of its small diameter and seems to be more catalytic compared to, e.g., 1-nm colloidal gold particles. Also, these organic mantle molecules allow a covalent binding to macromolecules, as opposed to the electrochemical

Figure 2.2. Detection of gold by AMG. Electron micrograph of a hepatocyte exposed to RNase gold complexes. The labeled enzyme is primarily located over the heterochromatin of the nucleus and the patchy accumulation of rough endoplasmatic reticulum. (From Danscher, G. and J. Nørgaard. 1983. J. Histochem. Cytochem., 31:1394-1398.) (Original magnification, ×24 ,000.)

17

Gold and Silver Staining forces that keep colloidal gold bound to its target macromolecules (see Chapter 6). The pH-neutral isoelectric point of Nanogold greatly improves the AMG detectability of intranuclear substances such as proliferation markers or even DNA. Bearing this fact in mind, and combining the new marker with tyramide signal amplification (TSA; Perkin Elmer Life Sciences, Boston, MA, USA; Web site: http://lifesciences.perkinelmer.com/) of in situ hybridization, AMG even leads to reproducible detection of single virus copies in the LM.65 Localization of Silver by AMG As mentioned, Liesegang failed to demonstrate exogenous silver in tissues by silver enhancement,29 and a technique was not published until 1981 when silver was traced by AMG at LM and EM levels in the central nervous system (CNS) neurons, glia cells, and most other cells in organisms exposed to silver salts.10 Silver has been a widely used remedy in medical history, for example as a nonspecific prophylactic agent for rich people in ancient times, and is still in use, e.g., as weak solutions of silver nitrate or as folia of metallic silver, because of its bacteria killing and astringent qualities. Recently, colloidal silver has become rather popular in USA as a mostly unauthorized treatment of a wealth of different conditions from sore throat to cancer. Organisms heavily contaminated with silver will show black-colored mucosae because of accumulations of silver sulfide/selenide, and gradually the skin will be stained as well—a condition called classical argyrosis. AMG reveals silver to be located both intracellularly and extracellularly. It is not known at present whether significant amounts of silver in exposed organisms end up bound to chemical species other 18

than sulfide/selenide, but exposure to UV light that would reduce silver ions to metallic silver has not resulted in any observable increase in the AMG staining. This suggests that metabolized silver will finally be bound as either silver sulfide or silver selenide clusters. Metallic silver can exist in the living organism only for a short period of time before it is oxidized. Ag2Se and Ag2S, on the contrary, are stable in the organism and may persist extracellularly or intracellularly for long periods of time, maybe even as long as the contaminated cells, basement membranes, or connective tissue elements exist. The AMG technique has demonstrated silver in all analyzed tissues from exposed animals. The spread and intensity of silver sulfide/selenide accumulations are dose and time dependent.10 In connective tissue, traces of silver are seen in fibroblasts and macrophages, and bound to reticular, collagen, and elastic fibers.10 In the kidneys, silver has been detected by means of AMG in the basal lamina of the glomerulus, the descending limb of Henle’s loop, the renal papilla interstitial tissue, and the proximal convoluted tubular and mesangial cells. In the liver, silver is confined to both Kupffer cells and bile canaliculi and is observed at ultrastructural levels in vesicles of the juxta bile canaliculi parts of the hepatocytes. In the ovaries, silver is in particular present in the zona pellucida and in the theca interna. In adrenals, silver is present in the chromaffin cells, and in the pancreas, silver is found both in the exocrine and endocrine part including beta cells.10 Finally, a detailed description of the presence of silver in structures of the eye after experimentally induced argyrosis is available.47 Silver penetrates the blood–brain barrier and can be demonstrated in neurons from experimentally exposed animals10 and in

Autometallographic Tracing autopsy material. In the CNS, neurons and glia cells accumulate silver depending upon the source of the silver. Motor neurons in the spinal cord and brain stem, neurons of the cerebellar relay nuclei, and neurons in certain parts of the basal ganglia are the neuron types most disposed to contain silver deposits in their lysosomes. Sensory neurons of the brain stem, on the other hand, are void of silver accumulations. Recently, silver has been demonstrated in hypothalamic nuclei of silver-exposed rats.56 In the peripheral nervous system, silver accumulates in Schwann cells and in satellite cells and neurons of dorsal root ganglia as well as neurons of the peripheral ganglia.46 Silver has also been shown to traverse rat placenta and to enter the fetal brain in utero. To our knowledge, AMG has not yet been published to trace colloidal silver. Colloidal silver can of course be AMG enhanced, and studies with the purpose of finding appropriate ways to use it in histological techniques are under way. Instead of colloidal gold, colloidal silver or clustered forms of silver comparable to the construction of Nanogold can be used as markers of macromolecules such as immunoglobulins (unpublished observations). This fact may lead to a new class of detection techniques. Localization of Mercury by AMG Timm worked out the AMG mercury method in 1962.60 A few years earlier, he had demonstrated that tissue sections from a brain fixed in sulfide-containing alcohol revealed a complex pattern of silver grains if the sections were AMG developed. He suggested that most of the silver pattern seen in brains subjected to his sulfide silver method was caused by amplification of zinc sulfide accumulations.59 The information that led to Timm’s famous method was the finding by the photographic scientist Zeiger66 that silver sulfide had the same AMG catalytic

capacity as metallic silver. Being trained in forensic medicine, Timm knew that mercury in the organism is metabolized to mercury sulfide. He therefore deduced that mercury could have the same AMG catalytic capacity as silver sulfide. The fact that the mercury method did not get the impact it deserved at that time was due to misunderstandings of Timm’s method at the user level. The most unfortunate one was the error of including sulfide ions in the presilver amplification process because: (1) zinc ions in the tissue will be bound in zinc sulfide clusters that, when AMG enhanced, will cause a false staining; and (2) mercury sulfide can be dissolved in a surplus of sulfide ions. Adding sulfide ions to the fixative, therefore, obviously caused some of the mercury sulfide to be dissolved or displaced, and at the same time resulted in fake AMG grains in the tissues.15 If, however, the method is correctly performed, it is a highly specific and extremely sensitive tool for tracing metabolized mercury in the organism.15,35 In kidney sections from rats exposed to mercury, it was found that the core of the AMG grains consisted of mercury sulfide. The sections were developed floating in the AMG developer, and after removal of the tissue with enzymes, the AMG grains were isolated and carefully rinsed several times in distilled water before being placed on a micropore filter and subjected to PIXE. It was found that mercury was present in amounts of 0.1% to 0.5% relative to silver, and that treatment with Hg and Se alternately resulted in a three to four times higher content of selenium in the AMG grains.35 The increase in Se was paralleled by an increase in the amount of AMG grains, indicating that mercury combines with selenium to create clusters of mercury selenide that can act as catalytic initiation sites for AMG grains, or alternatively, that catalytic clusters are formed containing 19

Gold and Silver Staining mercury, sulfide, and selenide concurrently. These observations support Komsta–Szumska and Chmielnicka’s finding that the soluble mercury fraction in kidneys from exposed animals diminished by more than 15 times if the animals received mercury and selenium simultaneously.27 Most organs analyzed with the AMG mercury method contain cells with traces of mercury sulfide/selenide in their lysosomes.2,5,7,35 Sometimes, mercury-caused silver grains are also found in secretory vesicles.58 Following selenium treatment 6 h before sacrifice, mercury has been detected by AMG in the nuclei of epithelial cells, in the proximal tubule of nephrons, and in macrophages.15 The distribution of mercury in the brain and spinal cord is rather complicated, and it is dependent on the source of the mercury, whether organic or inorganic, or whether administered intraperitoneally or orally. The AMG pattern is also dependent on dose and time of exposure, on the presence of selenium, and whether the latter is given simultaneously with mercury or a few hours before sacrifice. Møller–Madsen has worked out detailed maps of the distribution of AMG demonstrable mercury in brains and spinal cords from rats exposed to inorganic, organic, and vaporous mercury,32,34,49 and other investigators have studied retrograde axonal transport of mercury and its presence in spinal ganglia.3,48 It is to be expected that brains from a population whose nourishment contains large amounts of selenium, e.g., fishermen, Eskimos, and citizens in some areas of the USA, will demonstrate an AMG–Hg pattern that differs substantially from persons, from, say, Sweden and Norway, who have lived on selenium-poor food. These expectations are supported by findings in sled dogs from Thule, Greenland. Many data have accumulated on the presence of mercury in the brains of 20

humans who have been exposed to the metal, but almost all of these data are gained by quantitative methods involving lumps of brain tissue. The only way of obtaining an accurate knowledge about the cytological localization of mercury is by AMG. Experiments have suggested that approximately 30% of the tissue mercury is bound as mercury sulfide one month after start of exposure, and this pool is expected to grow until all the mercury in the organism has eventually been captured as HgS/HgSe crystallites.36 The AMG technique enables the demonstration of mercury at LM and EM levels and can act as a tool for semiquantitative evaluation of the local level of AMG demonstrable mercury. Detection of Bismuth by AMG The presence of AMG grains in brains from mice exposed to bismuth subnitrate was demonstrated by Ross et al.45 The authors found that the AMG technique revealed a highly organized pattern of staining that was not present in control animals. Detailed protocols for detection of bismuth in tissues have now been worked out.17a We have found that AMG grains harvested from tissue sections of organs from bismuth-intoxicated rats contain silver, bismuth, and sulfur. Animals treated concomitantly with bismuth and selenium contained Ag, Bi, S, and Se atoms. The distribution of bismuth sulfide/ selenide clusters in the CNS, the only organ analyzed in more detail until now, looks very much like the pattern seen in mercury-exposed mice. There are, however, significant differences. Bismuth AMG grains are not only found intracellularly in certain specific types of neurons and glia cells, but also extracellularly associated to blood vessels in some areas of the brain such as the cerebellum, 45 and future studies will almost

Autometallographic Tracing certainly reveal more dissimilarities. Concurrent administration of selenite and bismuth to rats leads to a more intense AMG staining in the CNS, suggesting a similar pathway of metabolism for the two neurotoxic metals.38 In mice subjected to bismuth gunshot pellets, pronounced accumulations of AMG bismuth were found in spinal cord motor neurons.38 In the above studies, we have found bismuth to be located ultrastructurally in lysosomes like organelles in the brain and in the basement membranes of vessels in the cerebellum. Ross et al.45 found bismuthcaused AMG grains in several different tissues of the mouse, and our studies have supported the notion that almost all tissues contain cells that concentrate bismuth in their lysosomes including testis.57 Localization of Zinc by AMG The following methods are used to demonstrate the zinc content of ZEN neurons. Sulfide Silver Methods The classical method of Timm has been modified and improved in different ways and can now be used also for high quality electron microscopical studies.8 A protocol for in vivo exposure to sulfide ions13 and a technique for demonstrating zinc ions in tissue sections from man and experimental animals18 have lately been worked out. Selenium Methods These methods include the in vivo selenium method for demonstrating ZEN terminals,11 the retrograde axonal transport method for tracing individual fiber tracts11,24 (Figure 2.3), and the global retrograde axonal transport method allowing all ZEN somata to be visualized.53 The latter approach can reveal both terminals and

somata of ZEN neurons, depending on the period of time the animals survive after having been exposed to selenium. In the mammalian brain, the densest populations of ZEN terminals are found in telencephalic structures. However, zinc-containing terminals are also present in the cerebellum, hypothalamus, thalamus, mesencephalon, putamen, medulla oblongata, and spinal cord. Until the beginning of the 1980s, most studies on the differentiated and highly laminated patterns of ZEN terminals in telencephalic structures were conducted with Timm’s modifications which were not specific for zinc. Because the brains were exposed to excess sulfide, the sections contained false AMG grains. These false silver grains are organized in an orderly fashion and cannot be distinguished from AMG grains caused by zinc sulfide crystallites. Species that have been studied with the zinc-specific Timm and selenium methods include man, rat, hedgehog, rabbit, mouse, birds, reptiles, and fish.52 All these species have been found to contain ZEN neurons that demonstrate a complicated pattern of zinc-containing terminals. As described initially, in vivo injections of sodium selenite or sodium selenide will result in accumulations of zinc selenide in the lysosomes of the involved ZEN neurons 24 h later.11 The retrograde axonal transport of the zinc selenide clusters makes it feasible to trace ZEN fiber tracts. By using this technique, it is now possible to detect all ZEN pathways, and this work is presently being conducted in several different laboratories. The significance of the synaptic vesicular zinc in ZEN terminals is not known, but recently it has been found that the vesicles contain zinc ion pumps in their membranes.37 The high zinc ion level in the vesicles could serve as a stabilizer of macromolecules as in the proinsulin-containing vesicles of the β-cells of the islets of Langerhans in the pancreas. It is believed that 21

Gold and Silver Staining proteins or peptides will be aggregated by binding to zinc ions and thereby be made osmotically invisible until the vesicles open to the surface. The rather widespread presence of ZEN vesicles in different secretory cells, e.g., acinar excretory cells in the pancreas, granular convoluted tubule cells of murine salivary glands, somatotrope cells of the pituitary, prostatic cells, mast cells, and Paneth cells, may reflect a general function of zinc ions in ZEN vesicles similar to that in the βcells, i.e., reversible aggregation of molecules in order to make them osmotically inert for transport and/or a store for later exocytosis.

Figure 2.3. Detection of selenide by AMG. Light micrograph of a 3µm thick Epon section from the neocortex of a rat that was treated intraperitoneally with 8 mg sodium selenite per kg body weight 24 h before being sacrificed. Retrogradely loaded neuronal somata from layer III demonstrate AMG silver amplified zinc selenide accumulation. (From J. Gu and G.W. Hacker (Eds.). 1994. Modern Methods in Analytical Morphology. Plenum Press, New York.) (Original magnification, ×1,298.)

22

In the brain and spinal cord, the function of ZEN vesicles could be to transport one or more important molecules to the synapses of ZEN boutons (e.g., molecules involved in growth or maintenance of synapses). Approximately 10% of the synaptic vesicles in ZEN neurons contain zinc and release it by exocytosis.4,25,40 In locations like the pancreas and pituitary gland, zinc ions are believed to be released together with the hormone, and probably thereby have fulfilled their function. In the CNS, release of zinc ions into the synaptic clefts may have an additional modulatory effect on postsynaptic receptors as has been suggested by in vitro experiments on slices of

Autometallographic Tracing the hippocampus.63,64 The demonstration of a modulatory binding site for zinc on the GABA receptor complex in cultured rat neurons54 and the demonstration of an antagonizing effect of zinc on NMDA and GABA responses on hippocampal neurons31 support this hypothesis. Other Applications of AMG The possibility of AMG amplification of silver and gold clusters has led to applications like the silver lactate method, which results in a perfect counterstaining of semithin sections of plastic-embedded osmium-fixed tissue11 (Figure 2.4). Another practical application of the technique is infusion of a 40°C 8% gelatin solution containing colloidal

gold particles in the vascular system. The ensuing AMG development reveals the vascular system as a dark web14 (Figure 2.5). PERSPECTIVES OF AMG The gold cluster technique used particularly in immunohistochemistry improves constantly, and its application has become an important engine for the development of new AMG approaches.20,21 Apart from the possibility of using AMG to detect exogenous silver in tissue, colloidal silver might be proved to be valuable as a marker of certain biological molecules. The HgSAMG technique will, when fully recognized, be a most useful tool for increasing our knowledge of the subtle effects of this

Figure 2.4. Counterstaining of semithin section with AMG. Testis fixed in glutaraldehyde, showing part of a seminiferous tubule in cross section. Different stages of spermatogenesis are visible. The cytoplasmic recesses of the Sertoli cells are especially well demonstrated in tissue postfixed in osmium. (From Danscher, G. 1983. Stain Technol. 58:365-372.) (Original magnification, ×519.)

23

Gold and Silver Staining

Figure 2.5. Detection of vessels by AMG. Photomicrograph from rat neocortex showing the local arrangement of capillaries. The animal was perfused with a gold–gelatine solution, and the section was AMG developed for 60 min at 26°C. (From Danscher, G. and A. Andreasen. 1997. J. Neurosci. Methods 77:175-181.) (Original magnification, ×127.)

metal on living organisms. Silver enhancement of bismuth clusters of exposed tissues is in its childhood, but BiSAMG in concert with other techniques will give interesting insight in the toxicity of this metal in the future. The biopsy-H 2 S gas method developed to reveal zinc ions in tissue sections will link the obtained experimental results to the significance of zinc ions in humans and be useful in the ongoing studies of the pathological role of zinc in brain lesions and Alzheimer’s disease. Future advances will also take place in the AMG developer and emulsion field leading to higher specificity including 24

lack of background, and on the marker level new sensitive AMG markers are to be expected.

REFERENCES 1.Aaseth, J., A. Olsen, J. Halse, and T. Hovig. 1981. Argyria-tissue deposition of silver as selenide. Scand. J. Clin. Lab. Invest. 41:247-251. 2.Arenholt-Bindslev, D. and G. Danscher. 1989. Effect of organic and inorganic selenium on mercury accumulation in cultures of normal human epithelial cells. ATLA (Alternative to Laboratory Animals) 16:253256. 3.Arvidson, B. 1987. Retrograde axonal transport of mercury. Exp. Neurol. 98:198-203. 4.Assaf, S.Y. and S.-H. Chung. 1984. Release of endogenous Zn++ from brain tissue during activity. Nature 308:734-736.

Autometallographic Tracing 5.Baatrup, E. and G. Danscher. 1987. Cytochemical demonstration of mercury deposits in trout liver and kidney following methyl mercury intoxication. Differentiation of two mercury pools by selenium. Ecotoxicol. Environ. Safety 14:129-141. 6.Bendayan, M. 1981. Ultrastructural localization of nucleic acids by the use of enzyme-gold complexes. J. Histochem. Cytochem. 29:531-541. 7.Bolewska, J., P. Holmstrup, B. Møller-Madsen, B. Kenrad, and G. Danscher. 1990. Amalgam associated mercury accumulations in normal oral mucosa, oral mucosal lesions of lichen planus and contact lesions associated with amalgam. J. Oral Pathol. Med. 19:3942. 8.Danscher, G. 1981a. Histochemical demonstration of heavy metals. A revised version of the sulfide silver method suitable for both light and electron microscopy. Histochemistry 71:1-16. 9.Danscher, G. 1981b. Localization of gold in biological tissue. A photochemical method for light and electronmicroscopy. Histochemistry 71:81-88. 10.Danscher, G. 1981c. Light and electron microscopic localisation of silver in biological tissue. Histochemistry 71:177-186. 11.Danscher, G. 1982. Exogenous selenium in the brain. A histochemical light and electron microscopical localization of catalytic selenium bonds. Histochemistry 76:281-293. 12.Danscher, G. 1984. Autometallography. A new technique for light and electron microscopic visualization of metals in biological tissues (gold, silver, metal sulfides, and metal selenides). Histochemistry 81:331335. 13.Danscher, G. 1996. The autometallographic zinc-sulfide method. A new approach involving in vivo creation of nanometer-sized zinc sulfide crystal lattices in zincenriched synaptic and secretory vesicles. Histochem. J. 28:361-373. 14.Danscher, G. and A. Andreasen. 1997. Demonstration of vessels in CNS and other organs by AMG silver enhancement of colloidal gold particles dispersed in gelatine. J. Neurosci. Methods 77:175-181. 15.Danscher, G. and B. Møller-Madsen. 1985. Silver amplification of mercury sulfide and selenide. A histochemical method for light and electron microscopic localization of mercury in tissue. J. Histochem. Cytochem. 33:219-228. 16.Danscher, G. and J.O.R. Nørgaard. 1983. Light microscopic visualization of colloidal gold on resinembedded tissue. J. Histochem. Cytochem. 31:13941398. 17.Danscher, G., M. Stoltenberg, and S. Juhl. 1994. How to detect gold, silver, and mercury in human brain and other tissues by autometallographic silver amplification. Neuropathol. Appl. Neurobiol. 20:454-467. 17a.Danscher, G., M. Stoltenberg, K. Kemp, and R. Pamphlett. 2000. Bismuth autometallography: protocol, specificity, and differentiation. J. Histochem. Cytochem. 48:1503–1510. 18.Danscher, G., S. Juhl, M. Stoltenberg, B. Krunderup, H.D. Schrøder, and A. Andreasen. 1997. Autometallographic silver amplification of zinc sulphide crystals

created in cryostat sections from human brain biopsies–AMGZnS-H2S—a new technique that makes it feasible to demonstrate zinc ions in zinc enriched vesicles contained in biopsy material. J. Histochem. Cytochem. 45:1503-1510. 19.Graudal, H., B. Møller-Madsen, and G. Danscher. 1988. Autometallographic demonstration of gold in rheumatoid synovial tissue. Z. Rheumatol. 47:347350. 20.Hacker, G.W., L. Grimelius, G. Danscher, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc in tissues. J. Histotechnol. 11:213-221. 21.Hacker, G.W., G. Danscher, L. Grimelius, C. HauserKronberger, W.H. Muss, A. Schiechl, J. Gu, and O. Dietze. 1995. Silver staining techniques, with special reference to the use of different silver salts in light and electron microscopical immunogold–silver staining, p. 19-45. In M.A. Hayat (Ed.), Immunogold–Silver Staining. Principles, Methods, and Applications. CRC Press, Boca Raton. 22.Hainfeld, J.F., R.D. Powell, J.K. Stein, G.W. Hacker, C. Hauser-Kronberger, A.L.M. Cheung, and C. Schöfer. 1999. Gold-based autometallography, p. 486487. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mansfield, and R.L. Price (Eds.), Proceedings of the Fifty-Seventh Annual Meeting, Microscopy Society of America. Springer-Verlag, New York. 23.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird. 1983. Immunogold–silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31:938-944. 24.Howell, G.A. and C.J. Frederickson. 1989. A retrograde transport method for mapping zinc-containing fiber systems in the brain. Brain Res. 515:277-286. 25.Howell, G.A., M.G. Welch, and C.J. Frederickson. 1984. Stimulation-induced uptake and release of zinc in hippocampal slices. Nature 308:736-738. 26.Jacobsen, E., A. Andreasen, H. Graudal, and G. Danscher. 1989. Autometallographic demonstration of gold uptake in cultured synovial fluid cells from patients with rheumatoid arthritis. Scand. J. Rheumatol. 18:161-164. 27.Komsta-Szumska, E. and J. Chmielnicka. 1981. Organ and subcellular distribution of mercury in rats in the presence of cadmium, zinc, copper, and sodium selenite. Clin. Toxicol. 18:1327-1334. 28.Liesegang, R.E. 1911. Die Kolloidchemie der histologischen Silberfärbungen, p. 1-44. In W. Ostwald (Ed.), Kolloidchemische Beihefte (Ergänzungshefte zur Kolloid-Zeitschrift). Monographien zur reinen und angewandten Kolloidchemie. Verlag Theodor Steinkopff, Dresden-Leipzig, FRG. 29.Liesegang, R.E. and W. Rieder. 1921. Versuche mit einer “Keimmethode” zum Nachweis von Silber in Gewebsschnitten. Z. Wiss. Mikrosk. 38:334-338. 30.Lormée, P., S. Lécolle, D. Septier, D. le Denmat, and M. Goldberg. 1989. Autometallography for histochemical visualization of rat incisor polyanions with cuprolinic blue. J. Histochem. Cytochem. 37:203-208.

25

Gold and Silver Staining 31.Mayer, M.L. and L. Vyklicky, Jr. 1989. The action of zinc on synaptic transmission and neuronal excitability in cultures of mouse hippocampus. J. Physiol. (Lond). 415:351-365. 32.Møller-Madsen, B. 1990. Localization of mercury in CNS of the rat. II. Intraperitoneal injection of methylmercuric chloride (CH3HgCl) and mercuric chloride (HgCl2). Toxicol. Appl. Pharmacol. 103:303323. 33.Møller-Madsen, B. and G. Danscher. 1983. Transplacental transport of gold in rats exposed to sodium aurothiomalate. Exp. Mol. Pathol. 39:327-331. 34.Møller-Madsen, B. and G. Danscher. 1991. Localization of mercury in CNS of the rat. IV. The effect of selenium on orally administered organic and inorganic mercury. Toxicol. Appl. Pharm. 108:457-473. 35.Nørgaard, J.O.R., B. Møller-Madsen, N. Hertel, and G. Danscher. 1989. Silver enhancement of tissue mercury: demonstration of mercury in autometallographic silver grains from rat kidneys. J. Histochem. Cytochem. 37:1545-1547. 36.Nørgaard, J.O.R., E. Ernst, and S. Juhl. 1994. Efficiency of autometallographic detection of mercury in the rat kidney. Histochem. J. 26:100-102. 37.Palmiter, R.D., T.B. Cole, C.J. Quaife, and S.D. Findley. 1996. ZnT-3, a putative transporter of zinc into synaptic vesicles. Proc. Natl. Acad. Sci. USA 93:1493414939. 38.Pamphlett, R., G. Danscher, J. Rungby, and M. Stoltenberg. 2000. Tissue uptake of bismuth from shotgun pellets. Environ. Res. 82:258-262. 39.Paxinos, G. and C. Watson (Eds.). 1987. The Rat Brain in Stereotaxic Coordinates, 2nd ed. Academic Press, Sydney. 40.Pérez-Clausell, J. and G. Danscher. 1986. Release of zinc sulphide accumulations into synaptic clefts after in vivo injection of sodium sulphide. Brain Res. 362:358361. 41.Pérez-Clausell, J., C.J. Frederickson, and G. Danscher. 1989. Amygdaloid efferents through the stria terminalis in the rat give origin to zinc-containing boutons. J. Comp. Neurol. 290:201-212. 42.Poulsen, E.H. and B. Møller-Madsen. 1990. Autometallographic demonstration of gold in the adrenal gland of rats exposed to sodium aurothiomalate. A light and electron microscopic study. Virchows Arch. B, Cell Pathol. 59:48-53. 43.Querido, A. 1947. Gold intoxication of nervous elements. On the permeability of the blood–brain-barrier. Acta Psychiatr. 22-23:97-151. 44.Roberts, W.J. 1935. A new procedure for detection of gold in animal tissue. Proc. R. Acad. (Amsterdam) 38:540-544. 45.Ross, J.F., R.C. Switzer, M.R. Poston, and G.T. Lawhorn. 1996. Distribution of bismuth in the brain after intraperitoneal dosing of bismuth subnitrate in mice: implications for routes of entry of xenobiotic metals into the brain. Brain Res. 725:137-154. 46.Rungby, J. 1986a. Exogenous silver in dorsal root ganglia, peripheral nerve, enteric ganglia, and adrenal

26

medulla. Acta Neuropathol. 69:45-53. 47 Rungby, J. 1986b. Experimental argyrosis: ultrastructural localization of silver in rat eye. Exp. Molec. Pathol. 45:22-30. 48.Schiønning, J.D. 1993. Retrograde axonal transport of mercury in rat sciatic nerve. Toxicol. Appl. Pharmacol. 121:43-49. 49.Schiønning, J.D. and B. Møller-Madsen. 1992. Autometallographic detection of mercury in rat spinal cord after treatment with organic mercury. Virchows Arch. B 61:307-313. 50.Schiønning J.D., E.H. Poulsen, B. Møller-Madsen, and G. Danscher. 1992. Autometallographic detection of gold in dorsal root ganglia of rats treated with sodium aurothiomalate. Exp. Mol. Pathol. 56:239-247. 51.Shirabe, T. 1978. Electron microscopic X-ray microanalysis of the nervous system after mercury intoxication. Folia Psychiatr. Neurol. Jpn. 32:278-283. 52.Slomianka, L. 1992. Neurons of origin of zinc-containing pathways and the distribution of zinc-containing boutons in the hippocampal region of the rat. Neuroscience 48:325-352. 53.Slomianka L., G. Danscher, and C.J. Frederickson. 1990. Labeling of the neurons of origin of zinc-containing pathways by intraperitoneal injections of sodium selenite. Neuroscience 38:843-854. 54.Smart, T.G. 1992. A novel modulatory binding site for zinc on the GABAA receptor complex in cultured rat neurones. J. Physiol. (Lond). 447:587-625. 55.Stierhof, Y.-D., B.M. Humbel, and H. Schwarz. 1991. Suitability of different silver enhancement methods applied to 1 nm colloidal gold particles: an immunoelectron microscopic study. J. Electron Microsc. Tech. 17:336-343. 56.Stoltenberg, M., S. Juhl, E.H. Poulsen, and E. Ernst. 1994. Autometallographic detection of silver in hypothalamic neurons of rats exposed to silver nitrate. J. Appl. Toxicol. 14:275-280. 57.Stoltenberg, M., G. Danscher, R. Pamphlett, M.M. Christensen, and J. Rungby. 2000. Histochemical tracing of bismuth in testis from rats exposed intraperitoneally to bismuth subnitrate. Reprod. Toxicol. 14:6571. 58.Thorlacius-Ussing, O., B. Møller-Madsen, and G. Danscher. 1985. Intracellular accumulation of mercury in the anterior pituitary of rats exposed to mercuric chloride. Exp. Mol. Pathol. 42:278-286. 59.Timm, F. 1958. Zur Histochemie der Schwermetalle. Das Sulfid-Silberverfahren. Dtsch. Z. Gerichtl. Med. 46:706-711. 60.Timm, F. 1962. Histochemische Lokalisation und Nachweis der Schwermetalle. Acta Histochem. Suppl. 3:142-158. 61.Ullberg, S., B. Larsson, and H. Tjälve. 1982. Autoradiography, p. 55-108. In H.J. Glenn and L.G. Colombetti (Eds.), Biological Applications of Radiotracers. CRC Press, Boca Raton. 62.Umeda, M., K. Saito, K. Hirose, and M. Saito. 1969. Cytotoxic effect of inorganic, phenyl, and alkyl mercuric compounds on HeLa cells. Jpn. J. Exp. Med. 39:47-58.

Autometallographic Tracing 63.Westbrook, G.L. and M.L. Mayer. 1987. Micromolar concentrations of Zn2+ antagonize NMDA and GABA responses of hippocampal neurons. Nature 328:640643. 64.Xie, X. and T.G. Smart. 1993. Properties of GABAmediated synaptic potentials induced by zinc in adult rat hippocampal pyramidal neurones. J. Physiol. (Lond). 460:503-523.

65.Zehbe, I., G.W. Hacker, H. Su, C. Hauser-Kronberger, J.F. Hainfeld, and R. Tubbs. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, streptavidin-nanogold, and silver acetate autometallography. Am. J. Pathol. 150:1553-1561. 66.Zeiger, K. 1938. Physikoschemische Grundlagen der histologischen Methodik. Wiss. Forschungsber. 48:55105.

27

3

Silver- and Gold-Based Autometallography of Nanogold® James F. Hainfeld and Richard D. Powell

INTRODUCTION For many applications, silver salt-based autometallography (often also called silver enhancement or silver development)5,8 is required to visualize colloidal gold (1–5 nm in diameter) or the small 1.4 nm Nanogold® particles (Nanoprobes, Yaphank, NY, USA).11 Although even Nanogold may be seen directly by scanning–transmission electron microscopy (STEM), by transmission EM (TEM; in thin sections without stain or ice-embedded cryo-EM samples), energy filtered TEM, and scanning EM (SEM), silver enhancement makes viewing in the EM more facile since the particles are enlarged to approximately 10 to 20 nm, convenient for most specimens. Autometallographic (AMG) enhancement is required in order to visualize smaller gold particles such as Nanogold for light microscopy (LM) or in blots or gels. This chapter includes the following protocols: • Protocol for HQ silver enhancement of Nanogold. • Protocols for use of silver-enhanced Nanogold with osmium tetroxide. A: Procedure using reduced concentration of OsO4. B: Procedures for gold toning.

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

• Protocol for HQ silver enhancement of Nanogold in pre-embedding immunocytochemistry for cell cultures. • Protocol for gold enhancement of Nanogold for EM. • Protocol for gold enhancement of Nanogold for LM. • Protocol for staining blots with Nanogold and silver enhancement. • Protocol for staining gels with Nanogold and silver enhancement. Commonly used heavy metal stains such as osmium tetroxide and lead citrate usually obscure the 1.4 nm gold particles, unless they have been so enhanced. The enhancement process generally follows immunolabeling with Nanogold-labeled Fab′ fragments, Nanogold-labeled IgG, or Nanogold-labeled streptavidin, and can be applied to pre-embedding, postembedding, or ultrathin cryosection protocols. Examples of the development of Nanogold for EM are shown in Figures 3.1 and 3.2. Enhancement is essentially a simple procedure in which the EM grid is simply floated on a drop of developer for several min For LM, silver enhancement is generally always required, and slides may be covered with the developer after immunolabeling with the gold antibody. Development times are

29

Gold and Silver Staining generally 5 to 10 min longer than those required for EM. A new procedure that deposits gold instead of silver is now available.12 This has the advantages of lower background in some cases, higher electron density, which gives higher contrast for EM viewing, a much better backscatter signal for SEM, and full compatibility with OsO4, which can dissolve or etch silver.3 Protocols are given for these enhancement procedures. Silver or gold enhancement can also be used to enhance the signal from Nanogold probes to the point where they are visible with the naked eye. This renders gold labeling visible on gels and blots. This is useful in molecular biology where gels are run, and where it can be used to distinguish bands containing gold-labeled proteins from those that do not; for example,

one lane can be stained with Coomassie blue for protein, and the another with AMG, which will show only those bands that are gold labeled.7,24 Dot blots are very useful for checking the metal enhancement process and can be used to determine development times for EM.4 They are also used to quickly assay or troubleshoot an antigen labeling experiment. In a typical dot blot, the target antigen is placed in dilutions on nitrocellulose; subsequent incubations with primary and secondary (Nanogold-labeled) antibodies, followed by AMG, reveal the sensitivity of antigen detection and provide a format in which dilutions of primary and secondary antibodies or other parameters can be varied to optimize antigen labeling.10,16 Therefore, we include protocols for use with gels and blots.

Figure 3.1. Silver enhancement of Nanogold clusters. (A) TEM photomicrograph of Nanogold clusters without enhancement. Arrow points to a 1.4 nm gold particle. (B) Nanogold clusters after 30-sec development (IntenSE M; Amersham Pharmacia Biotech, Little Chalfont, Bucks, England, UK) giving 1.7 to 3.3 nm particles. Arrow shows one that is 2.9 nm. (C) Nanogold (more dilute) after 3-min silver development, showing 11 to 40 nm particles. Arrow points to a 19 nm silver grain. (D) Control with no Nanogold but exposure to 3 min of development, showing minimal background spots (arrow). Bar = 0.040 µm. (Reprinted with permission from Hainfield, J.E. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40: 177–184.)

30

Silver- and Gold-Based Autometallography STAINING PROTOCOLS Protocol 1. HQ Silver Enhancement of Nanogold HQ Silver (Nanoprobes) is a commercial silver enhancement kit which is optimized for high ultrastructural preservation and uniform particle size in EM. Materials and Reagents • Phosphate-buffered saline (PBS) buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.4. • PBS-BSA (bovine serum albumin) buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, 2 mmol/L sodium azide, and 1.0% BSA, fraction V by heat shock (Sigma, St. Louis, MO, USA), pH adjusted to 7.4. • HQ Silver reagent. • Deionized or distilled water. Procedure 1. Rinse with deionized water (2 times for 5 min).

2. Float grid with specimen on freshly mixed developer for 1 to 8 min or as directed in the instructions for the silver reagent. More or less time can be used to control particle size. A series of different development times should be tried, to find the optimum time for your experiment. With HQ Silver, a development time of 4 min gives 15 to 40 nm round particles. Since HQ Silver is light sensitive, it should be handled in a darkened room, using a safelight, or inside a covering box to avoid the generation of nonspecific background. 3. Rinse with deionized water (3 times for 1 min). 4. Mount and stain as usual. Protocols 2 A–C. Silver Enhancement of Nanogold with Osmium Tetroxide In some cases, OsO4 will oxidize the deposited silver back into solution, resulting in loss of signal. One of three procedures is recommended in such cases: (A) use of lower concentrations of OsO4; (B) gold toning using either procedure 2B or procedure 2C; or (C) use of gold enhancement (discussed later). Investigators therefore have a choice of procedures.

Figure 3.2. Time course for silver enhancement of Nanogold. (A) Gold particles (1.4 nm) adhered to poly-L-lysine-treated formvar-coated EM grid but not incubated with silver enhancement solution. The gold was not visualized by standard transmission EM at this magnification. (B–F) Nanogold particles adhered to grids as in panel A and then incubated with the silver enhancement solution for (B) 1 min, (C) 2 min, (D) 3 min, (E) 4 min, and (F) 5 min. The silver-enhanced gold particles were evident as early as 1 min and continued to increase in size with longer enhancement times. The results of this preparation are typical; however, slight variations in development time were observed with different batches of silver enhancement solution. Bar = 0.1 µm. (Reprinted with permission from Takizawa, T. and J.M. Robinson. 1994. Use of 1.4–nm immunogold particles for immunocytochemistry on ultra-thin cryosections, J. Histochem. Cytochem. 42:1615–1623.)

31

Gold and Silver Staining Protocol 2A. Procedure Using Reduced Concentration of OsO4 This procedure is reported for cells,23 but may be adapted to tissues. Materials and Reagents • PHEM buffer, prepared as follows: 60 mmol/L PIPES, 25 mmol/L HEPES, 10 mmol/L EGTA, 2 mmol/L MgSO4, pH 6.9. Abbreviations used in this buffer system are: PIPES = piperazine-N,N′-bis[2-ethanesulfonic acid], can also be written as 1,4-piperazinediethanesulfonic acid HEPES = N-[2-hydroxyethyl]piperazine-N′-[4-nutanesulfonic acid] EGTA = ethyleneglycol-bis (betaaminoethyl ether) N,N,N′,N′-tetraacetic acid • PBS buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.4. • PBS+ buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, with 1% normal goat serum, 0.1% saponin, 50 mmol/L glycine, 0.1% fish skin gelatin, 1 mg/mL BSA, and 0.02% NaN3. • Glutaraldehyde. • 50 mmol/L HEPES with 200 mmol/L sucrose, pH 5.8. • Fixer: 250 mmol/L sodium thiosulfate and 20 mmol/L HEPES, pH 7.4. Procedure 1. Rinse cells with PHEM buffer, pH 6.9, for 30 sec. 2. Fix cells in 0.7% glutardialdehyde for 15 min in PHEM buffer (use a nonamine containing buffer, i.e., do not use Tris-buffer). 32

3. Lyse cells for 15 min in PHEM buffer containing 0.5% Triton X-100. 4. Rinse cells in 3 changes of PBS, pH 7.4, over 15 min. 5. Quench glutaraldehyde with 2 changes of NaBH4 (1 mg/mL in Tris-buffered saline, pH 7.4) over 15 min. 6. Wash cells with 3 changes of PBS with 1% normal goat serum, 0.1% saponin, 50 mmol/L glycine, 0.1% fish skin gelatin, 1 mg/mL BSA, and 0.02% NaN3 (PBS+). 7. Incubate cells with primary antibody (usually 1:250 dilution or 1:500 dilution of ascites fluid) for 60 min at 37°C. 8. Rinse 3 times in PBS+. 9. Incubate with Nanogold antimouse Fab′ (or IgG) (1:50 dilution) for 60 min at 37°C. 10. Wash 3 times with PBS+. 11. Postfix with 1.6% glutaraldehyde in PBS for 15 min. 12. Wash 4 times with 50 mmol/L HEPES with 200 mmol/L sucrose, pH 5.8, over 30 min. 13. Silver enhance for 5 to 20 min, shielding from light. 14. Rinse 3 times over 5 min in fixer (250 mmol/L sodium thiosulfate and 20 mmol/L HEPES, pH 7.4). 15. Wash 3 times over 15 min with 0.1 mol/L phosphate, pH 7.4, with 0.1 mol/L sucrose. 16. Osmicate with 0.1% OsO4 for 30 min. 17. Dehydrate and embed; section. 18. Stain thin sections with uranyl acetate and lead citrate. Note that since silver ions in the silver enhancer precipitate with chloride ions, all PBS and other chloride buffers must first be removed. This is generally done with

Silver- and Gold-Based Autometallography water washes, but in the above procedure, a more physiological wash buffer is used (Step 12, HEPES-sucrose).

methods.21 An example of the results is shown in Figure 3.3. Materials and Reagents

Protocol 2 B–C. Procedures for Gold Toning Note: Treatment with osmium tetroxide followed by uranyl acetate staining can lead to much more drastic loss of the silver-enhanced Nanogold particles. This may be prevented by gold toning. Procedure 2B2,3 1. After silver enhancement, wash thoroughly with deionized water. 2. 0.05% gold chloride: 10 min at 4°C. 3. Wash with deionized water. 4. 0.5% oxalic acid: 2 min at room temperature. 5. 1% sodium thiosulfate (freshly made) for 1 h. 6. Wash thoroughly with deionized water and embed according to usual procedure. 7. Now osmium staining may be performed. Procedure 2C18 1. Rinse twice quickly in distilled water. 2. 0.05 mol/L sodium acetate (1 min) then rinse again quickly. 3. 0.05% tetrachloroauric acid (2 min). 4. Rinse thoroughly in distilled water for 10 min, then osmicate. Protocol 3. HQ Silver Enhancement of Nanogold in Pre-Embedding Immunocytochemistry for Cell Cultures This procedure has been described by Tanner and coworkers and is reported to give significantly higher densities of silver-enhanced gold particles than other

• Sodium phosphate buffer: 0.1 mol/L sodium phosphate, pH adjusted to 7.4. • PBS buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.4. • Glutaraldehyde and paraformaldehyde. • HQ Silver reagent. • Deionized or distilled water. Procedure 1. Fix for approximately 45 min (for monolayer cultures) with one of the following: (1) 4% paraformaldehyde in 0.1 mol/L sodium phosphate buffer, pH 7.4, or (2) 2% paraformaldehyde with 0.05% to 0.1% glutaraldehyde in 0.1 mol/L sodium phosphate buffer, pH 7.4. 2. Wash with 0.1 mol/L sodium phosphate buffer, pH 7.4, 3 times for 5 min each. 3. Blocking and permeabilize the cells with PBS with 5% goat serum, 0.1% sodium azide, and 0.1% saponin for 1 h. 4. Incubate with primary antibody made in PBS with 5% normal goat serum, 0.1% saponin, and 0.1% sodium azide for 1 h at room temperature. 5. Wash with PBS with 1% goat serum and 0.1% sodium azide for 3 to 4 times for 5 min. 6. Incubate with Nanogold-labeled Fab′ antirabbit or mouse (depending on the primary antibody) secondary antibody conjugate (4 µL) in 1 mL of PBS with 1% goat serum and 0.1% sodium azide for 1 h at room temperature. 7. Wash with PBS containing 1% goat serum with 0.1% sodium azide once, then with PBS twice. 33

Gold and Silver Staining 8. Fix with 2% glutaraldehyde in PBS for 30 min. 9. Wash 3 times in PBS. Store overnight. Next day: 10. Wash with water thoroughly. 11. Perform silver enhancement (HQ Silver enhancement kit). 13. Wash in water. Check under LM carefully; only process the promising specimens for EM. 14. Wash in 0.1 mol/L phosphate buffer, pH 7.4. 15. 0.2% OsO4 in 0.1 mol/L phosphate buffer for 30 min. 16. Wash, stain with uranyl acetate, dehy-

drate in ethanol, and embed. Protocol 4. Gold Enhancement of Nanogold for EM Materials and Reagents • PBS buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.4. • PBS-BSA buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, 2 mmol/L sodium azide, and 1.0% BSA, fraction V by heat shock, pH adjusted to 7.4. • GoldEnhance EM reagent (Nanoprobes).

Figure 3.3 EM immunocytochemistry of the K+ channel, Kv2.1, in brain neurons. The silver-enhanced (HQ Silver) gold grains (Nanogold-anti-mouse Fab′) are distinct on the plasma membrane of the neuronal soma and large dendrites. The plasma membranes facing astrocytic processes shows the heaviest staining, with many more immunograins facing astrocytes than facing synaptic terminals. Intracellularly, the Golgi apparatus is positively stained. Full width, 6.15 µm. (Reprinted with permission from Du, J., et al., 1998. Neuroscience, 84:37–48.)

34

Silver- and Gold-Based Autometallography Procedure 1. Incubate with the immunogold or Nanogold conjugate according to your usual or recommended protocol. 2. Optional: Postfix with 1% glutaraldehyde in PBS. 3. Wash 3 times for 5 min with PBS with 50 mmol/L glycine (after glutaraldehyde postfix only—to remove aldehydes). 4. Wash 3 times for 5 min in PBS-BSA. 5. Wash 3 times for 5 min in distilled water. 6. Gold enhancement (GoldEnhance kit): use equal amounts of the four components (Solutions A, B, C, and D); prepare about 40 µL of reagent per grid. A convenient method is to use one drop (approximately 10 µL) from each bottle. After mixing, a drop may be placed on a sheet of parafilm and a grid floated on it for the required time. a. First mix Solution A (enhancer: green cap) and Solution B (activator: yellow cap).

b. Wait 5 min. c. Add Solution C (initiator: purple cap), then Solution D (white cap) and mix. d. Develop for the optimal particle size (usually between 3–20 min). 7. Rinse with distilled water. Figure 3.4a shows results obtained using GoldEnhance to enlarge 5 nm cells in tissue sections.12 Protocol 5. Gold Enhancement of Nanogold for LM The following procedure was developed for gold enhancement of in situ hybridization (ISH) specimens by Cheung, HauserKronberger, and Hacker, in collaboration with the authors,12 as a modification of the Nanogold-silver staining procedure;9 an example of the results obtained using this method is shown in Figure 3.4b. It has been found to be effective for enhancement of tissue sections for LM observation. We have found enhancement duration

Figure 3.4a Electron micrograph of human testis. (Full width, 1.45 µm). DNA in spermatids was labeled with mouse anti-DNA primary (Roche Molecular Biochemicals, Indianapolis, IN, USA), then biotinylated antimouse antibody (Amersham Pharmacia Biotech), followed by Nanogold–streptavidin, followed by gold autometallography (8 min). (Reprinted with permission from Hainfeld, J.F. et al., 1997. Proc. 57th Ann. Mtg., Micros. Soc. Amer., Springer-Verlag, New York.)

35

Gold and Silver Staining times of 10 to 20 min give optimal results; however, this reagent is intended to function in a wide range of conditions, and different washes and development times may give better results in your application. A similar procedure may be used for blotting applications; a comparison of silver enhancement and GoldEnhance development is shown in Figure 3.4c. You should

follow your normal procedure up to the application of the gold conjugate; the protocol below describes the steps after this: Materials and Reagents • PBS buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.6.

Figure 3.4b. Human papillomavirus (HPV) 16/18 in cervical carcinoma. LM photomicrographs of formalin-fixed serial paraffin sections of cervical squamous cell carcinoma, in situ hybridized for HPV-16/18 using a biotinylated probe (Pathogene-HPV kit; Enzo Diagnostics, Farmingdale, NY, USA) (Bar = 10 µm). (A) Direct detection using streptavidin-peroxidase. (B) Direct detection using Nanogold–streptavidin followed by gold autometallography for 18 min. (Reprinted with permission from Du, J. et al., Neuroscience 84:37–48.)

Figure 3.4c. Immunoblot detection of mouse IgG on nitrocellulose. Gold-goat antimouse IgG (15 nm) is used and amplified with (A) silver AMG (LI Silver) and (B) gold AMG. (C) Key showing the amounts of mouse IgG in each spot for the corresponding divisions of the blots. (Reprinted with permission from Hainfeld, J.F. et al., 1997. Proc. 57th Ann. Mtg., Micros. Soc. Amer., Springer-Verlag, New York.)

36

Silver- and Gold-Based Autometallography • PBS-gelatin buffer: 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, 2 mmol/L sodium azide, and 0.1% gelatin (high purity), pH adjusted to 7.6. Optional: Background may be reduced by using 0.5 mol/L NaCl and 0.05% Tween 20 in this buffer. • GoldEnhance LM reagent (Nanoprobes). Procedure 1. Incubate the sections with Nanogold or colloidal gold conjugate according to current protocols or using the buffers, concentrations, and protocols recommended for the conjugate. 2. Wash in PBS, pH 7.6, 2 times for 5 min each. 3. Wash in PBS-gelatin, pH 7.6, for 5 min. 4. Repeatedly wash in distilled water for at least 10 min altogether, the last 2 rinses in ultrapure water (EM-grade). 5. Prepare GoldEnhance using equal amounts of the four components (Solutions A, B, C, and D); prepare about 80

µL per slide. a. Dispense Solution A (enhancer: green cap) into a clean tube or dish, add Solution B (activator: yellow cap), and mix thoroughly. b. Wait 5 min. c. Add Solution C (initiator: purple cap) and Solution D and mix thoroughly. d. Apply 1 to 2 drops (approximately 80 µL, sufficient to cover the specimen) to the slide. e. Develop specimen for 10 to 20 min. More or less time can be used to control particle size and intensity of signal. 6. When optimum staining is reached, immediately stop by rinsing carefully with deionized water. Protocol 6. Staining of Blots with Nanogold and Silver Enhancement The basic procedure for gold immunoblotting has been described by Moeremans et al.,15 which may be followed. For best results, the membrane should be hydrated before use by simmering in gently boiling water for 15 min. Best results are obtained when the antigen is applied using a 1-µL

Figure 3.5. Immunoblot of serial dilutions of Mouse IgG. Spotted onto a hydrated nitrocellulose membrane, detected using Nanogold-labeled Fab′ goat antimouse IgG, then developed using LI Silver. The last visible spot (arrow) contains 0.1 pg of the target IgG.

37

Gold and Silver Staining capillary tube (Figure 3.5). The procedure for immunoblots is as follows:10,16 Materials and Reagents • Buffer 1 (Blocking): 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, 2 mmol/L sodium azide, and 4.0% BSA (fraction V by heat shock), pH adjusted to 7.4. • Buffer 2 (Incubation): 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, 2 mmol/L sodium azide, 0.8% BSA (fraction V by heat shock), and 1.0% normal serum from the host animal of the Nanogold conjugate antibody, pH adjusted to 7.4. Optional: Even lower backgrounds may be obtained with 0.5 mol/L NaCl and 0.05% Tween 20. • Buffer 3 (Wash): 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, 2 mmol/L sodium azide, and 0.8% BSA (fraction V by heat shock), pH adjusted to 7.4. • Buffer 4 (PBS): PBS buffer: 0.02 mol/ L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.4. • Glutaraldehyde. • 0.05 mol/L disodium EDTA, pH 4.5. • Silver enhancement reagents, e.g., according to Danscher5 or to Hacker et al.8,9 Procedure 1. Spot 1-µL dilutions of the antigen in Buffer 4 onto hydrated nitrocellulose membrane. Use an antigen concentration range from 100 ng to 0.01 pg/µL. 2. Block with Buffer 1 for 30 min at 37°C. 3. Incubate with primary antibody according to usual procedure (1 or 2 h).

38

4. Rinse with Buffer 1 (3 times for 10 min). 5. Incubate with a 1/100 to 1/200 dilution of the Nanogold reagent in Buffer 2 for 2 h at room temperature. 6. Rinse with Buffer 3 (3 times for 5 min), then Buffer 4 (2 times for 5 min). 7. Optional (may improve sensitivity): Postfix with glutaraldehyde, 1% in Buffer 4 (10 min). 8. Rinse with deionized water (2 times for 5 min). 9. Optional (may reduce background): Rinse with 0.05 mol/L EDTA at pH 4.5 (5 min). 10. Develop with freshly mixed silver developer for 5 to 25 min as directed in the instructions for the silver enhancement protocol used. Repeating the process for a second time may be beneficial. If performed twice, between the developments, thorough rinsing with deionized or better distilled water is required. Note: If silver lactate AMG5 is used, it is advisable to shield preparations from daylight, e.g., within a cupboard. Silver acetate AMG8,9 is less sensitive to daylight, and development usually can take place under normal laboratory light conditions if not performed for a longer time. If precipitation takes place (solution turns to gray or black), this may be understood as a sign of too much light intensity (in this case, place a dark dustbin on the vials to shield them from daylight). If the solution turns whitish, the quality of the distilled or deionized water is too low, and chloride ions may be present. 11. Rinse several times and thoroughly with deionized water. Caution: Nanogold particles degrade upon exposure to concentrated thiols

Silver- and Gold-Based Autometallography such as beta-mercaptoethanol or dithiothreitol. If such reagents must be used, concentrations should be kept below 1 mmol/L and exposure restricted to 10 min or less. Protocol 7. Staining Gels with Nanogold and Silver Enhancement Procedure7,24 1. After labeling with Nanogold, remove unbound gold particles by column chromatography, sucrose gradient or other purification means. Leaving excess free Nanogold in the sample will interfere with the intended gel staining. 2. Run gel as usual; however, Nanogold is degraded by beta-mercaptoethanol [or dithiothreitol (DTT)], so the sample must not be mixed with a reducing agent, i.e., a nonreducing gel must be run. Normal concentrations of other ingredients [sodium dodecyl sulfate (SDS), etc.] are acceptable. 3. Gel may be electrotransferred to nitrocellulose if desired, although this is not necessary. 4. Rinse gel with several changes of deionized water. Since the silver developer is precipitated by halides, traces of NaCl must be removed. 5. Place the gel or blot in a suitable dish and apply enough freshly prepared LI Silver (Cat. No. 2013; Nanoprobes) to cover the gel. LI Silver is prepared by mixing equal amounts of a and b components. Do not use the usual gel silver stains, which are quite different from LI Silver and do not develop the Nanogold effectively. 6. Watch development of band(s) which should appear brown-black. Aggregates with gold that did not enter the gel or small amounts of free gold may give background staining. Usual development time is 1 to 5 min. Extensive

development time (>30 min) will lead to some nonspecific background selfnucleation staining by the developer alone. 7. When optimal staining is reached, stop development by rinsing in deionized water. The final stained gel is now a permanent record. 8. For comparison and visualization of all bands, run a duplicate gel and stain with Coomassie blue or gel silver stain. 9. A Nanogold-labeled molecule may run approximately 15 ,000 MW higher on the gel due to the added weight of the Nanogold particle (approximately 15 ,000). However, due to the small hydrodynamic size of the gold cluster, some labeled proteins run close to their native position. Some results from different gel staining experiments run using different conditions are shown in Figures 3.6a7 and 3.6b.11 TECHNICAL HINTS AND DISCUSSION AMG is a versatile method with an increasing variety of refinements, which may be applied to a wide variety of specimens. When correctly optimized, Nanogold labeling with silver or gold enhancement can give higher detection sensitivities than competing technologies, such as enzyme-linked detection.9,25 The results are affected by many factors, and a variety of modifications to these protocols are available that can be used to optimize them for specific systems or experiments or correct problems that may be encountered with the general protocols. Silver enhancers tend to be divided into two types. The first is often based on silver lactate, which includes a thickening agent or protective colloid, usually gum Arabic, although gelatin and polyethylene glycol (carbowax) have also been used, and is 39

Gold and Silver Staining light sensitive. Examples include the Danscher formulation7 and the N-propyl-gallate formulation developed by Burry.4 These may consist of three or more components, and are usually preferred for EM because they produce enhanced particles of a more uniform size and shape and allow improved preservation of ultrastructural morphology. The second type is usually not highly light sensitive, although strong illumination does have an effect, and the formulation is often based on silver acetate, although other silver salts have been used.

Examples include the silver acetate AMG solution suggested by Hacker et al.8 These are simpler to use, usually consisting of two components that are mixed immediately before development, and are preferred for LM and blotting because development can be visually monitored. Krenács and Krenács have reported excellent results with a light insensitive silver acetate developer for postembedding, which gave very uniform 10 nm spheres from Nanogold at the EM level.13 Use of a safelight is recommended for these developers, but development

Figure 3.6a. Electrophoretic analysis of proteasome-amyloid β protein (Aβ)-Nanogold complex. (A) Covalent and select conjugation of monomaleimido-Nanogold to proteins requires the presence of a cysteine residue on the protein. Because Aβ lacks cysteines, we used a peptide variant in which the last amino acid (Val40) was substituted with a cysteine residue (Aβ1-39C40). Aβ1Au 39C40 was coupled to Nanogold as described in Reference 7 to form Aβ in which each labeled Aβ molecule was linked to a single gold particle. The product (0.1 µg) was analyzed by 14% Tris-Tricine polyacrylamide gel electrophoresis (PAGE) (lane 2), Aβ139C40 (lane 1), and Nanogold (lane 3) were used as controls. Proteins were transferred onto the polyvinylidine fluoride (PVDF) membrane for 30 min at 150 mA at 4°C, and AβAu was immunostained with anti-Aβ antibodies (left panel) or stained with the silver enhancement method (right panel). Both staining methods reacted with the same band indicating that AβAu migrates as a complex of 17 kDa. Molecular size markers are shown are shown on the right. Note that because gel electrophoresis was performed under denaturing, but not reducing, conditions to prevent thiol degradation of the gold particle, the control lane with the peptide alone shows both the monomer and dimer forms of Aβ1-39C40 (lane 1). (B and C) Electrophoretic characterization of the proteasome-AβAu complex. For STEM analysis, the complexes were cross-linked as described in Reference 7. Cross-linked proteasomes (panel B, lane 2) and cross-linked proteasome-AβAu complexes (panel B, lane 3) migrated faster than noncross-linked proteasomes (panel B, lane 1). AβAu was incubated with proteasome to form proteasome-AβAu complex. The complex was detected by Coomassie blue (B) and silver enhancement staining (C). Both staining methods identified the same band confirming the formation of the proteasome-AβAu complex. (B) Lane 1, 3 µg of noncross-linked proteasome; lane 2, 3 µg of cross-linked proteasome; lane 3, cross-linked proteasome-AβAu complex. (C) Lane 1, cross-linked proteasome-AβAu complex; lane 2, 1 µg of cross-linked AβAu; lane 3, 3 µg of cross-linked proteasome to Nanogold. (Reprinted with permission from Gregori, L., et al., 1997. J. Biol. Chem., 272: 58–62.)

40

Silver- and Gold-Based Autometallography under a box to exclude direct light in a normally lit room is acceptable. Gold salt-based enhancement is a new procedure, developed by Nanoprobes, in which gold rather than silver is deposited onto gold seed particles.1,12 This procedure has a number of advantages over silver enhancement. In addition to higher contrast in the electron microscope, greatly increased backscatter signal (for SEM), and resistance to osmium etching, gold enhancement gives a longer time between full development and autonucleation. This means that gold enhancement is more suited to systems requiring extensive washing,

or automated processes with longer wait times between steps. Unlike silver, gold is not precipitated by chloride, and therefore gold enhancement can be conducted in the presence of physiological buffers containing saline. Compared with silver enhancement, lower backgrounds have been reported for ISH experiments using Nanogold with gold enhancement as the detection system.12 The biggest challenge with AMG is to select the right development time for the desired particle size or staining level. In the light microscope, a slide can be periodically monitored; but for a light sensitive

Figure 3.6b. SDS polyacrylamide Phast gels of native and Nanogold-labeled proteins, with development by Coomassie blue or silver-enhancement. Lane 1 is a protein molecular weight standard (values listed on left are in kDa), lane 2 is a native Fab′, lane 3 is a Nanogold-Fab′, and lane 4 is F(ab′)2. Gels A and C are developed with Coomassie blue and gels B and D are developed with a silver enhancer (LI Silver). A and B are gels of samples that were not heated before running, and C and D are gels of samples heated to 100°C in 1.3% SDS for 5 min before running. Gels A and B were identical except for staining, as were gels C and D. The unheated samples show native and Nanogold-labeled Fab′ to run anomalously, showing bands greater than 50 kDa, whereas F(ab′)2 runs at approximately 100 kDa as expected. After heating (gels C and D), the Fab′ runs as expected showing bands at 50 kDa and the single light or heavy chains at 25 kDa. The Nanogold-labeled Fab′ bands are nearly indistinguishable from the native Fab′ bands in this case (gel C, lanes 2 and 3). In all cases, the silver enhancement specifically developed the Nanogold labeled proteins selectively (gels B and D), and unlabeled proteins did not develop (gels B and D, lanes 1, 2, and 4). In addition, Nanogold bands with silver enhancement were intense in less than 5 min, whereas Coomassie staining took 1 h (followed by 1 h of destaining). (Reprinted with permission from Hainfeld, J.F. and F. R. Furuya, 1995. Immunogold–Silver Staining: Principles Methods, and Applications, CRC Press, Boca Raton, pp 71–96.)

41

Gold and Silver Staining developer for EM, this is more difficult. Burry has devised a simple test strip method for Nanogold to standardize results from week to week.4 Nanogold-Fab′ was spotted (approximately 0.5 µL) onto a strip of nitrocellulose at 1:10, 1:50, 1:100, and 1:500 dilutions. The strip was run at the same time as the tissue, and the spots turned faint and then dark brown during development. Particles (15–20 nm) in the TEM corresponded to a medium brown spot at the 1:50 dilution; this time point also was just before silver staining could be perceived in the light microscope. Several size distribution studies have been reported for silver-enhanced Nanogold. Burry et al.4a used N-propyl gallate (NPG) developer over a 1 to 15 min time period to study the enhancement of Nanogold and 1 nm colloidal gold. A linear increase in particle density was found for 1 nm colloidal gold, whereas a sigmoidal curve was observed for Nanogold. However, the size distribution variation (standard deviation) at any particular time point was significantly less for Nanogold.4 Cultured cell immunolabeling with Nanogold and silver amplification produced good results at 15 min intensification time for LM, but labeling was optimal for EM after a 6 min development, giving an average size of 20 nm particles (10 min gave usable 35 nm particles). Fixed tissue sections required longer silver amplification times (20–25 min) than cultured cells to produce good results, presumably due to the increased time required for the developer to diffuse into the specimens. Another study documented the size of Nanogold particles adsorbed to poly-Llysine coated formvar grids, enlarged using the same NPG developer.20 Particles (10 nm) were obtained after about 3 min, and 25 nm particles were obtained after 5 min. These authors also used this as a quick test (using the EM) to determine optimal development time for each batch of their 42

silver enhancement solution. Nanogold was compared with undecagold and colloidal gold in a third study.11 Silverenhanced Nanogold was found to be more sensitive for visual detection of a target antigen than either undecagold or 1 or 3 nm colloidal gold. We typically find silveramplified immunodot blots using Nanogold conjugates to be 10 to 100 times more sensitive than colloidal gold conjugates (e.g., 10 nm). Components which improve the performance of silver enhancement reagents include natural products such as gum Arabic, which can vary in composition from lot to lot. Therefore, when using such reagents, it is advisable to test them before using a new batch to ensure that results are reproducible. Tanner and coworkers have used such reagents extensively and, for optimum and consistent performance, recommend the following procedures:21 1. Prepare or order sufficient reagent for several experiments (for consistency). Freeze the component solutions in small lots and thaw when needed. 2. Test on grid before use to obtain an approximate reaction time for the required silver particle size. Make up a 1:10 dilution of the Nanogold, place a formvar-coated grid on a drop of this solution, remove excess, and let dry. Then silver enhance the grid. This provides a test of both the potency of the Nanogold (i.e., the proportion of particles which nucleate enhancement), as well as the reaction time and quality of the silver enhancer. 3. The silver enhancement solutions should not be freeze-thawed more than once. Also, storage in the refrigerator is not recommended, since the properties can change with storage time. 4. When making up the silver enhancement solution, if using HQ Silver, pour the most viscous solution (moderator,

Silver- and Gold-Based Autometallography Solution B) first into a tube with volume markings. Then add equal volume of Solution A (initiator). Mix the two very well, then add Solution C (activator). Mixing should be both very thorough and very quick. The performance of the HQ Silver can change if it is not used immediately after mixing. Best results are obtained when the reagents are mixed and used quickly. 5. The silver reaction can still change even after thorough water wash. Therefore, strong light should be avoided after silver enhancement. 6. Use a low concentration of OsO4 (0.2%). The susceptibility of the deposited silver to osmium etching can vary from batch to batch of silver enhancement reagent. In some experimental systems, background staining — the presence of silverenhanced particles in areas of the specimen known not to contain the target — can be a problem. This can arise from a number of sources: (1) from unbound Nanogold particles, (2) from unbound primary antibody or probe, or (3) from autonucleation of the silver enhancer solution in the absence of gold particles. Reducing the concentration of the primary antibody or probe or the Nanogold conjugate can reduce or eliminate this problem, as can more extensive washing procedures. Incorporation of a detergent such as Tween 20 or saponin into the procedure can also act to facilitate removal of unbound probe. We have found that background signal may be reduced or avoided by washing thoroughly with sodium citrate buffer before enhancement.16 Where HQ Silver is used, 0.02 mol/L sodium citrate buffer at pH 7.0 has been found to be most effective. In preparations utilizing the Danscher silver enhancement protocol,6 0.02 mol/L sodium citrate buffer, adjusted to pH 3.5,

was most effective. In blots, we find that rinsing with 0.05 mol/L disodium EDTA, pH 4.5, immediately before silver enhancement can reduce background. We attribute this effect to the chelation and removal by the EDTA of transition metal ions, which can act as nucleation sites for silver enhancement. In addition to the sodium citrate buffer and using a lower concentration of the Nanogold probe, a number of methods have been described for stopping the silver enhancement reaction, or for “back-developing,” to remove extraneous deposited silver. These prevent the continuation of the reaction in the specimens after development is complete (for example, if the silver is only slowly removed from the tissue), and may help reduce background signal. Sodium thiosulfate (1% aqueous solution, freshly made) is a good stop reagent for both silver and gold enhancement and may be used to stop gold or silver development in situations where repeated water washes are insufficient. Washing with deionized water, then incubation with sodium thiosulfate for 1 to 2 min, followed by rinsing thoroughly again with deionized water is usually sufficient to stop development.22 However, caution should be exercised with this procedure when using gold enhancement. In some experiments, treatment with sodium thiosulfate has been found to reduce signal. Note: In our experience, it is advisable to avoid stopping the enhancement process by sodium thiosulfate or photographic fixer when using Nanogold for supersensitive DNA or RNA detection. We have often observed a strong reduction of staining when using the stop-bath for more than 1 sec, and one had to be very fast. Instead, but with the risk of obtaining some degree of background staining, thorough washing in distilled water can replace the immediate interruption of the 43

Gold and Silver Staining enhancement process with sodium thiosulfate–photographic fixer by this slower, but less invasive water wash. Other methods for stopping the AMG reaction include: 1. 1% acetic acid.19 2. 1% acetic acid followed by photographic fixer (Agefix; Agfa-Gevaert, or Ilfospeed 200; Ilford Photo, Paramus, NJ, USA).19 3. Direct photo fix, using the same photographic fixers listed above.4 4. Brief rinse in 2.5% sodium chloride.19 5. 15% to 25% aqueous sodium thiosulfate plus 15% sodium sulfite.5 6. 1% acetic acid, washes in acetate buffer, toning in 0.05% HAuCl4 for 3 to 10 min, with excess silver removed with 3% sodium thiosulfate.20 We found that Nanogold-labeled proteins run on a polyacrylamide gel kept low backgrounds when stopped with 10% acetic acid with 10% glucose in water, as opposed to just a water stop. 7. Although not reported for Nanogold labeling, silver overdevelopment of immunogold probes has been used, followed by reversal, to lower the background.5 A modified Farmer’s solution was used for the reversal (0.3 mL 7.5% potassium ferricyanide, 1.2 mL of 20% sodium thiosulfate, 60 mL water) [Reference 4; already reported by Hacker in Springall et al. (19a)]. If the higher concentrations of probe required for fluorescence microscopy continue to result in nonspecific signals after AMG, treatment with this solution after AMG may help to reduce it. Conversely, in some procedures, little or no development has been found upon AMG. Results may be improved in these systems by changing from commercial silver enhancement reagents to freshly-pre-

44

pared Danscher and Hacker formulations5,8,9 or by substituting formaldehyde for glutaraldehyde in postfixation. Nanogold with silver enhancement may be followed by standard immunocolloidal gold to a different antigen for double labeling. This was achieved by Takizawa and Robinson,20 who showed that the labels were very distinctly recognizable and that the silver enhancement was gentle enough to preserve antigenicity when the next immunolabel (a 10 nm colloidal gold) was applied. This is useful when one antigen is sparse, since Nanogold generally gives much more dense labeling than colloidal gold. Nanogold with AMG can also be used in conjunction with other staining procedures for multiple antigen staining. In the electron microscope, the particles are easily distinguished from other stains, and in the light microscope, the black staining is also readily distinguished from other commonly used stains. Two studies have described the use of AMG-enhanced Nanogold in conjunction with enzymatic labeling to distinguish different antigens.14,17 Nanogold and silver enhancement should be completed before the application of the enzymatic probe. If the enzymatic probe is applied first, the substrate can act as a nucleating agent during AMG enhancement and give nonspecific background staining. Further optimization of both the formulation and applications of silver and gold enhancement with Nanogold are planned. AMG-enhanced Nanogold offers a unique combination of high spatial resolution and punctate staining for the electron microscope, and the highest sensitivity for LM and blotting. The development of gold enhancement and related technologies makes this process readily applicable to automated staining instruments and molecular diagnostics.

Silver- and Gold-Based Autometallography ACKNOWLEDGMENTS We are grateful to Drs. Frederic Furuya, John M. Robinson, and Susan Cheng for their help in preparing this manuscript and for the figures. We also wish to thank Drs. Gerhard W. Hacker, Christian Schöfer, and Cameron Ackerley for their help with the gold enhancement procedures. This work was supported by the Office of Biological and Environmental Research of the U.S. Department of Energy under Prime Contract No. DE-AC02-98CH10886 with Brookhaven National Laboratory, by National Institutes of Health Grant 2 P41 RR01777 and by NIH Small Business Innovation Research grants GM 48328, GM 49564, and GM 56090. REFERENCES 1.Ackerley, C.A., A. Tilups, C.E. Bear, and L.E. Becker. 1999. Correlative LM/TEM studies are essential in evaluating the effectiveness of liposome mediated delivery of the cystic fibrosis transmembrane regulator (CFTR) as a corrective therapy in a CFTR knockout mouse that develops lung disease, p. 484-485. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mansfield, and R.L. Price (Eds.), Proc. 57th Ann. Mtg., Micros. Soc. Amer., Springer-Verlag, New York. 2.Arai, R., M. Geffard, and A. Calas. 1992. Intensification of labelings of the immunogold–silver staining method by gold toning. Brain Res. Bull. 28:343-345. 3.Arai, R. and I. Nagatsu. 1995. Application of gold toning to immunogold-silver staining, p. 209-216. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 4.Burry, R.W. 1995. Pre-embedding immunocytochemistry with silver-enhanced small gold particles, p. 217230. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 4a.Burry, R.W., D.D. Vandré, and D.M. Hayes. 1992. Silver enhancement of gold antibody probes in preembedding electron microscopic immunocytochemistry. J. Histochem. Cytochem. 40:1849-1856. 5.Danscher, G. 1981. Histochemical demonstration of heavy metals. A revised version of the silver sulphide method suitable for both light and EM. Histochemistry 71:1-16. 6.Danscher, G. 1981. Localization of gold in biological tissue. A photochemical method for light and EM. Histochemistry 71:81-88.

6a.Du, J., J.-H. Tao-Cheng, P. Zerfas, and C.J. McBain. 1998. The K+ channel, Kv2.1, is opposed to astrocytic processes and is associated with inhibitory postsynaptic membranes in hippocampal and cortical principal neurons and inhibitory interneurons. Neuronscience 84:37-48. 7.Gregori, L., J.F. Hainfeld, M.N. Simon, and D. Goldgaber. 1997. Binding of amyloid beta protein to the 20S proteasome. J. Biol. Chem. 272:58-62. 8. Hacker, G.W., L. Grimelius, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc in tissues. J. Histotechnol. 11:213-221. 9.Hacker, G.W., C. Hauser-Kronberger, I. Zehbe, H. Su, A. Schiechl, O. Dietze, and R. Tubbs. 1997. In situ localization of DNA and RNA sequences: supersensitive in situ hybridization using streptavidinNanogold-silver staining: minireview, protocols, and possible applications. Cell Vision 4:54-65. 10.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40:177-184. 11.Hainfeld, J.F. and F.R. Furuya. 1995. Silver enhancement of Nanogold and undecagold, p. 71-96. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 12.Hainfeld, J.F., R.D. Powell, J.K. Stein, G.W. Hacker, C. Hauser-Kronberger, A.L.M. Cheung, and C. Schoefer. 1999. Gold-based autometallography, p. 486-487. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mansfield, and R.L. Price (Eds.), Proc. 57th Ann. Mtg., Micros. Soc. Amer., Springer-Verlag, New York. 13.Krenács T. and L. Krenács. 1995. Comparison of embedding media for immunogold–silver staining, p. 57-70. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 14.Li, H., H. Ohishi, A. Kinoshita, R. Shigemoto, S. Nomura, and N. Mizuno. 1997. Localization of a metabotropic glutamate receptor, mGluR7, in axon terminals of presumed nociceptive, primary afferent fibers in the superficial layers of the spinal dorsal horn: an electron microscope study in the rat. Neurosci. Lett. 223:153-156. 15.Moeremans, M., G. Daneels, A. Van Dijck, G. Langanger, and J. De Mey. 1984. Sensitive visualization of antigen-antibody reactions in dot and blot immune overlay assays with immunogold and immunogold–silver staining. J. Immunol. Meth. 74:353-360. 16.Powell, R.D., C.M.R. Halsey, D.L. Spector, S.L. Kaurin, J. McCann, and J.F. Hainfeld. 1997. A covalent fluorescent-gold immunoprobe: simultaneous detection of a pre-mRNA splicing factor by light and EM. J. Histochem. Cytochem. 45:947-956. 17.Salas, P.J.I. 1999. Insoluble gamma-tubulin-containing structures are anchored to the apical network of intermediate filaments in polarized CACO-2 epithelial cells. J. Cell Biol. 146:645-657.

45

Gold and Silver Staining 18.Sawada, H. and H. Esaki. 1994. Use of Nanogold followed by silver enhancement and gold toning for preembedding immunolocalization. J. Electron Microsc. 43:361-366. 19.Scopsi, L. 1989. Silver-enhanced colloidal gold method, p. 260. In M.A. Hayat (Ed.), In Colloidal Gold: Principles, Methods, and Applications, Vol. 1. Academic Press, San Diego. 19a.Springall, D.R., G.W. Hacker, L. Grimelius, and J.M. Polak. 1984. The potential of the immunogold–silver staining method for paraffin sections. Histochemistry 81:603-608. 20.Takizawa, T. and J.M. Robinson. 1994. Use of 1.4nm immunogold particles for immunocytochemistry on ultra-thin cryosections. J. Histochem. Cytochem. 42:1615-1623. 21.Tanner, V.A., T. Ploug, and J.-H. Tao-Cheng. 1996. Subcellular localization of SV2 and other secretory vesicle components in PC12 cells by an efficient

46

method of preembedding EM immunocytochemistry for cell cultures. J. Histochem. Cytochem. 44:14811488. 22.Van Driel, D. 1997. Gold toning for silver enhanced immunogold reacted tissue. Micros. Today 97-7:28. 23.Vandré, D.D. and R.W. Burry. 1992. Immunoelectron microscopic localization of phosphoproteins associated with the mitotic spindle. J. Histochem. Cytochem. 40:1837-1847. 24.Wilkens, S. and R.A. Capaldi. 1992. Monomaleimidogold labeling of the g subunit of the E. coli F1 ATPase examined by cryoEM. Arch. Biochem. Biophys. 229:105-109. 25.Zehbe, I., G.W. Hacker, H. Su, C. Hauser-Kronberger, J.F. Hainfeld, and R. Tubbs. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, streptavidin-Nanogold, and silver acetate autometallography. Detection of single-copy human papillomavirus. Am. J. Pathol. 150:1553-1561.

4

Immunogold–Silver Staining for Light Microscopy Using Colloidal or Clustered Gold (Nanogold®) Gerhard W. Hacker, Annie L.M. Cheung, Raymond R. Tubbs, Lars Grimelius, Gorm Danscher, and Cornelia Hauser-Kronberger

INTRODUCTION Nearly three decades have passed since Faulk and Taylor first introduced colloidal gold as a label for immunoelectron microscopy (immuno EM) in 1971.23,55, 78 Since then, immunogold-staining (IGS) has gained widespread application by virtue of its various advantages over other, nonparticulate immunostaining techniques. As accumulations of colloidal gold in a concentration high enough to be seen as red color in the light microscope (LM) require the presence of very large amounts of antigen,28 introduction of silver enhancement (autometallography; AMG) by Danscher in 1981 became the prerequisite for obtaining a higher sensitivity in immunogold applications.13,14,16 Combination of this reaction with colloidal gold0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

labeled enzyme histochemistry and immunohistochemistry (IHC) were introduced simultaneously in 1983 by Danscher et al.15 and Holgate et al.,53,54 respectively, and were considered real breakthroughs in immunogold technology. The new IHC technique evolving from this was named immunogold–silver staining (IGSS) and had a dramatically improved sensitivity and detection efficiency compared to the immunostaining techniques available at that time.37,88 As described in Chapters 2 and 3, the high sensitivity of IGSS is largely based on the high intensification potential of AMG (earlier misleadingly called “physical development”), applied to amplify the size of the gold particle(s) used as label of antibodies or streptavidin. Under the LM, conglomerations of gold particles embedded in silver appear as black precipitates with a

47

Gold and Silver Staining distinctly sharper appearance than the reaction products of most enzyme-labeled preparations. Silver lactate was originally used as the ion source in AMG,13,14 and size amplification of gold label is achieved by the formation of shells of metallic silver around each gold particle by the AMG process.18,44,46,61 The process is catalyzed by hydroquinone in a citrate buffer of low pH. Thereby, two surplus electrons present on each molecule of hydroquinone function as electron donors for the transformation of silver ions into metallic silver atoms. By precipitation of these metallic silver atoms onto (similarly sized) gold atoms on the surface of the gold particles, the latter ones are enlarged in size. If gold particles are sufficiently near to each other, which is the case when good immunolabeling is obtained, conglomeration of gold and silver takes place. Under the LM, gold–silver accumulations are clearly visu-

alized as grayish-black dots or areas and stand out well against the nonreacting background (Figures 4.1 through 4.5). In their first reports, Holgate and his team clearly showed that IGSS was extremely sensitive and detection efficient.53, 54,57 Compared to other IHC techniques, including peroxidase antiperoxidase (PAP), which was the standard IHC technique in the seventies and early eighties,91 antigens could often be detected with IGSS but not with PAP in parallel paraffin sections.53,54,57 Also, higher dilutions of primary antibodies and shorter incubation times (reduced from overnight to 60–90 min) could be used to give better results than with the PAP technique. With great enthusiasm, our unforgettable friend and memorable scientist Dr. David R. Springall (1945–1997) and Gerhard Hacker read the original IGSS manuscripts.53,54 We were looking for a good

Figure 4.1. Carcinoma-associated antigen MAM-6 in pleural biopsy. Using indirect IGSS with colloidal gold, intense black membrane-bound and partly cytoplasmic staining is obtained, easing diagnosis of epithelial origin of the tumor. Formalin-fixed paraffin section, counterstained with eosin and nuclear fast red.

48

Immunogold–Silver Staining for Light Microscopy method that would allow reliable detection of neuropeptides in formalin-fixed paraffin sections and felt that IGSS could be the solution. (Usually, one needed to do immunofluorescence on frozen sections to obtain good immunostaining for neuropeptides, which are present only in small accessible quantities.) We started a large series of immunostaining tests, therein comparing IGSS with various immunostaining methods on consecutive paraffin and frozen sections, in combination with antibodies most commonly used in immunohistopathology and neuroscience.37,38 However, our first attempt at the original IGSS method resulted in incredibly high unwanted background staining. We were not successful in frozen sections at all, but we found a great potential for IGSS in material readily available in most pathology institutes: sections from routine paraffin block files that had been fixed conventionally in formalin. We then attempted to modify IGSS to facilitate a highly sensitive demonstration of various

kinds of substances in routine paraffin sections.37–39,88 One of our goals was to successfully access the large pathology institute paraffin block files to retrospectively study regulatory peptide-related disorders such as neuroendocrine tumors. Holgate and his colleagues had proposed an indirect IGSS method. They used antisera to immunoglobulins and membrane surface antigens as the primary layer in human tonsil specimens and non-Hodgkin lymphomas, while the second antiserum was adsorbed to 20 nm diameter gold particles. With time, our extensive testing led to various modifications, which also included the use of smaller gold particle sizes (1–5 nm in diameter), the use of fresh water fish gelatin, and a number of new versions of AMG developers.88 Later on, the indirect method was replaced by a three-layer IGSS technique, utilizing the streptavidin–biotin-complex (S–ABC) principle.7,10,35,44 Antigen retrieval methods, originally proposed for standard enzyme-labeled S–ABC techniques

Figure 4.2. (See color insert Figure 4.2 following page 78.) Desmin in striated muscle fibers. Muscle striation is demonstrated in clear detail using the indirect IGSS method, here performed without heat antigen retrieval or proteinase predigestion. Formalin-fixed paraffin section, counterstained with H & E.

49

Gold and Silver Staining proposed by Shi et al. were also introduced for IGSS.43,60,71,85,86 Numerous improvements were reported by many groups, and the IGSS technique became more widely used. However, it never really reached high popularity. One reason for this can be found in the strict policy of the major IHC reagent-producing companies to rely mainly on enzymes as the label and thus invest substantially into improving and promoting enzyme-label techniques. Along with this, huge advertising campaigns made peroxidase and alkaline phosphatase the most commonly used labeling systems in IHC and in situ hybridization (ISH). This trend has greatly undermined the distinct advantage of IGSS over other immunostaining techniques in histopathology, i.e., an intense black staining product that allows the pathologist to rapidly screen sections under low magnifi-

cation in the microscope and which can be combined with hematoxilin and eosin (H & E) counterstain—the darling tool of pathologists (Figures 4.1 through 4.5). Numerous applications of IGSS in immunohistopathology have been described.9, 21,37,41,44,53,54,89,98 Although in IGSS, colloidal gold most often produced surprisingly good results even when compared to today’s standard IHC techniques (enzymelabeled S–ABC methods). One major and rather annoying drawback was that it could not well demonstrate intra-nuclear substances such as steroid receptors (used in breast carcinoma diagnosis), proliferation markers (such as Ki-67 and related antigens), or some tumor suppressor gene proteins (such as p53). Also, not yet understandable to us, a few other antigens could not be reliably detected with IGSS at all. For instance, one antibody to chromoganin A,

Figure 4.3. (See color insert Figure 4.3 following page 78.) Astrocytes in human brain. The indirect IGSS method with colloidal gold and monoclonal mouse antibodies to glial fibrillary acidic protein (GFAP) can detect glial cells and their processes in excellent detail even without using antigen retrieval or proteinase predigestion. Specimen taken from astrocytoma, formalin-fixed paraffin section, counterstained with H & E.

50

Immunogold–Silver Staining for Light Microscopy which performed well in peroxidase-based IHC, did not give any staining with IGSS. All our earlier attempts to solve these problems did not yield success. However, over the years a new type of label named Nanogold® (Nanoprobes, Yaphank, NY, USA), invented by James Hainfeld and colleagues, evolved.50,51,75,93 As discussed in Chapter 6, Nanogold is clustered gold and is very different from colloidal gold particles. It is not a pure accumulation of gold atoms adsorbed to macromolecules by electrochemical forces, but a gold crystal (1.4 nm in diameter) surrounded by organic molecules, which in sum, give different chemical and physical properties. The whole complex can be covalently bound to macromolecules, and unlike colloidal gold, which has a pH optimum of about 8.4, Nanogold shows a nearneutral isoelectric point. Also, due to this “more electropositive” isoelectric position,

Nanogold covalently bound to streptavidin gives much better intra-nuclear labeling than colloidal gold. In a large test series comparing optimum colloidal gold IGSS with Nanogold-based IGSS and the S–ABC peroxidase method in several applications, we could clearly show that Nanogold-based IGSS is a real alternative to today’s standard IHC methods (Figure 4.5) (Cheung, unpublished). A number of very interesting facts were noted in these experiments: First, Nanogold performed well only when used in a (three-layer) S–ABC method, i.e., bound to streptavidin, whereas indirect (twolayer) methods utilizing Nanogoldlabeled immunoglobulins or Fab fragments did not yield acceptable immunostaining. Earlier, we found that IGSS using streptavidin-conjugated colloidal gold did not have any advantage over the classical indirect IGSS method

Figure 4.4. Small ganglion in urinary bladder. A small group of nerve cells and axons is demonstrated here using monoclonal rat antibodies against neurofilament protein triplet (NFP) with the indirect colloidal gold IGSS method. Bouin’s fixed paraffin section, counterstained with eosin only.

51

Gold and Silver Staining using colloidal gold-bound immunoglobulin.41 Therefore, a fair comparison of colloidal versus clustered gold could only be made between indirect colloidal gold IGSS and a three-step Nanogold-labeled streptavidin IGSS technique. Immunostaining of parallel sections showed that in nearly all cases, Nanogold silver staining performed better than colloidal gold IGSS. For Ki-67 antigen for instance, a much larger number of immunostained nuclei in sections of invasive breast carcinoma was obtained with Nanogold, and this number was in most cases well comparable to the picture revealed by S–ABC peroxidase. An incredibely fine resolution has often been obtained when Nanogold–silver staining was applied, e.g., showing nuclear DNA structures in great detail (Figure 4.5). When Nanogold–silver staining was directly compared to adjacent sections immunostained with S–ABC peroxidase, nearly all

systems tested showed comparable outcome. However, we also noted that there were some cases in which Nanogold–silver gave better staining, and a few other cases in which S–ABC peroxidase was superior. A conclusion which could be drawn from this is that even at the beginning of the new millennium, we still do not have the optimum IHC procedure which could be a solution for every kind of immunostaining problem. In this chapter, protocols and guidelines for use of IGSS in LM IHC are given. A number of chapters contained in this book also deal with EM applications both of colloidal and clustered gold. The principle of IGSS can also be applied to detect substances by means other than IHC. Examples are lectin histochemistry,58,62,82,87,99 enzyme histochemistry,58 or in situ transcription.19 An interesting approach is the new possibiliy to use one-and-the-same

Figure 4.5. Nuclear DNA figures in mammary carcinoma. Using NGSS, it is possible to observe in very fine resolution subnuclear structures, as demonstrated here with monoclonal mouse antibodies to Ki-67. Formalin-fixed paraffin section, heat antigen retrieval in citrate buffer, pH 6.0, using a microwave oven, and lightly counterstained with eosin only.

52

Immunogold–Silver Staining for Light Microscopy reagent, FuoroNanogold-labeled streptavidin, for fluorescent observation (Figure 4.6) and, silver-enhanced, for transmitted light LM.74,76 One of the most promising applications is the detection of DNA and RNA with gold and silver by ISH and in situ polymerase chain reaction (PCR).42, 45–49,63,66,94,97,104,107–110 The combination of IGSS with tyramide signal amplification for ISH4,5,101 and in IHC (not yet tested by us) will undoubtedly expand the scope of IGSS applications to hitherto unknown, single molecule sensitivity. IGSS is especially well suited for computer-based microscopic image analysis.92 The following Protocols describe the use of IGSS with colloidal gold and Nanogold in IHC. As a supporting protocol, silver acetate AMG40 is suggested. Technical hints to critical steps are given

in the discussion, which also contains information on alternative enhancement procedures. PROTOCOLS Protocol 1. Indirect IGSS with Colloidal Gold Materials and Reagents Antigen retrieval reagent: citrate buffer, pH 6.0: 10× concentrate citrate buffer: 10.51 g citrate monohydrate in 450 mL distilled water. Add concentrated NaOH to adjust pH to 5.68. Add 5 mL Tween 20 and mix well. Make up to 500 mL. Store at 4°C. Dilute 10 times and adjust pH to 6.0

Figure 4.6. FluoroNanogold detection of Ki-67 in mammary carcinoma. Using one and the same reagent, it is possible first to examine a section in the fluorescent microscope (as seen here using cyline 3 [Cy3] as the label), then silver or gold enhance the section and watch it in the conventional transmitted LM (not shown here). It would also be possible to finally transfer such preparations to transmission electron microscopy (TEM) via postfixation and resin-embedding using a Popp-Off technique (W.H. Muss, Salzburg) (not shown here; for technical details, see Reference 46).

53

Gold and Silver Staining Table 4.1. Examples of Manufacturers of Clustered and Colloidal Gold Labeled Macromolecules Company Name, Location, and Type of Gold Web Site Nanoprobes, Yaphank, NY, USA Clustered gold

http://www.nanoprobes.com/index.html

Aurion, Wageningen, The Netherlands Colloidal gold

http://www.aurion.nl/

Affinity Research, Exeter, England, UK Colloidal gold

http://www.affiniti-res.com/

Chemicon, Temecula, CA, USA Colloidal gold

http://chemicon.com/

Jackson Immunoresearch—Affinipure Antibodies, West Grove, PA, USA

http://www.jacksonimmuno.com

Colloidal gold Ted Pella, Redding, CA, USA Colloidal gold

with citric acid for use. Phosphate-buffered saline (PBS), pH 7.6: 10× PBS stock (0.1 mol/L PBS): dissolve 11.36 g Na 2HPO 4, 2.72 g KH2PO, and 87.0 g NaCl in 1 L distilled water. Further reagents: • Primary antibodies, e.g., raised in rabbit (polyclonal), mouse, or rat (monoclonal). • Primary antibody diluting buffer (e.g., Cat. No. M35; Biomeda, Foster City, CA, USA). • Secondary antibodies: colloidal gold (1–5 nm)-adsorbed goat, antirabbit, mouse, or rat immunoglobulins (for manufacturers, see Table 1). • DAKO pen (Code No. S2002; DAKO, Carpinteria, CA, USA). • Lugol’s iodine solution (Code No. 54

http://www.tedpella.com/

109261; Merck, West Point, PA, USA). • 2.5% aqueous sodium thiosulfate solution (store at room temperature). • Detergent can be added to the washing buffers if needed, such as 0.1% Tween 20 or 80 or Triton X-100, to further reduce background staining and improve penetration through the cell membranes. • PBS-BSA: PBS, pH 7.6, containing 1% bovine serum albumin (BSA; stored as 1-mL frozen aliquots). • PBS-gelatin: PBS containing 0.1% fish gelatin (Teleostean gelatin from cold water fish skin; Code No G7765; Sigma-Aldrich, Steinheim, Germany). • Nuclear fast red counterstain (“Kernechtrot”). Warm up to about 50°C in microwave oven before use. Reusable (Code No. 115939; Merck). • Plastic forceps (avoid contamination with metal shortly before or during silver enhancement). • Mounting medium (e.g., DPX; Elec-

Immunogold–Silver Staining for Light Microscopy tron Microscopy Sciences (EMS), Fort Washington, PA, USA). For choices of mounting media compatible with silver stains, see also Chapter 1. • Silver enhancement solutions (see Protocol 3). Staining Procedure 1. Deparaffinize sections and bring them to water through graded alcohols. 2. Distilled water (>3 min). 3. Apply antigen retrieval as appropriate. Let slides cool down and take to water. 4. Oxidize in Lugol’s iodine solution (5 min). 5. Rinse in tap water (>10 s), then distilled water (>10 s). 6. Reduce in 2.5% aqueous sodium thiosulfate solution till colorless (a few s). 7. Wash thoroughly in tap water (>10 s), then in distilled water (2 min). 8. Drain off, wipe around section, and surround it with PAP pen. 9. Immerse in PBS containing detergent (>5 min). 10. Apply 1:20 normal serum of the species providing the secondary antibody (5 min). 11. Drain off. 12. Incubate with primary antibody (appropriately diluted with primary antibody diluent) (overnight at 4°C). 13. Wash in PBS-gelatin, also containing detergent (3 times for 5 min). 14. Incubate with goat antirabbit or antimouse immunoglobulins adsorbed to 1 or 5 nm gold (Table 1), appropriately diluted (most often, 1:50–1:100) in PBS-BSA (1 h at room temperature). 15. Wash in PBS-gelatin, also containing detergent (3 times for 5 min).

16. Wash repeatedly in ultrapure water (altogether for at least 15 min). 17. Apply silver acetate AMG (5–10 min); see Protocol 3. 18. Rinse carefully in distilled water (>3 min). 19. Counterstain with H & E or nuclear fast red, dehydrate, and mount. Protocol 2. IGSS with Streptavidin– Nanogold Materials and Reagents The same reagents and materials as in Protocol 1 are required, with the following exceptions: • Biotinylated secondary antibody (e.g., Cat. No. K0492; StreptABComplex/ HRP Duet Kit; DAKO). • Nanogold-labeled streptavidin (Cat. No. 2016; Nanoprobes). Staining Procedure 1. Deparaffinize sections and bring to water through graded alcohols. 2. Distilled water (>3 min). 3. Apply antigen retrieval as appropriate. Let slides cool down and take to water. 4. Oxidize in Lugol’s iodine solution (5 min). 5. Rinse in tap water (>10 s), then distilled water (>10 s). 6. Reduce in 2.5% aqueous sodium thiosulfate solution till colorless (a few s). 7. Wash thoroughly in tap water (>10 s), then in distilled water (2 min). 8. Drain off, wipe around section, and surround it with PAP pen. 9. Immerse in PBS containing detergent (>5 min). 10. Apply 1:20 normal serum of the species providing the secondary antibody (5 min). 55

Gold and Silver Staining 11. Drain off. 12. Incubate with primary antibody (appropriately diluted with primary antibody diluent) (overnight at 4°C). 13. Wash in PBS-gelatin, also containing detergent (3 times for 5 min). 14. Incubate with biotinylated antirabbit or antimouse immunoglobulins (e.g., from DAKO, diluted 1:200 in PBS) (30 min at room temperature). 15. Wash in PBS-gelatin, also containing detergent (3 times for 5 min). 16. Incubate with streptavidin-Nanogold diluted 1:750 in PBS-BSA (60 min at room temperature). 17. Wash in PBS-gelatin, also containing detergent (3 times for 5 min) 18. Wash repeatedly in ultrapure water (altogether for at least 15 min). 19. Apply silver acetate AMG (5–10 min); see Protocol 3. 20. Rinse carefully in distilled water (>3 min). 21. Counterstain with H & E or nuclear fast red, dehydrate, and mount. Protocol 3. Supporting Protocol: Silver Acetate Autometallography Materials and Reagents (According to Reference 40, slightly modified.) The glass containers used for the silver enhancement step must be thoroughly clean and free of all salts and metals. The following solutions A and B should be freshly prepared for every run. • Solution A: 80 mg silver acetate (Code No. 85140; Fluka, Buchs, Switzerland) are dissolved in 40 mL of ultrapure water. Continuous stirring for about 15 min is necessary to dissolve silver acetate crystals adequately. 56

• Citrate buffer: 23.5 g of trisodium citrate dihydrate and 25.5 g citric acid monohydrate are dissolved in 850 mL of ultrapure water. This buffer can be kept at 4°C for at least several weeks. It is important to adjust to pH 3.8 with citric acid solution not more than 60 min before use. • Solution B: 200 mg hydroquinone are dissolved in 40 mL of citrate buffer. • Enhancement solution: Just before use, solution A is mixed with solution B. Procedure 1. Silver enhancement: Slides are placed vertically in a glass container (preferably with about 80 mL volume and up to 19 slides; Schiefferdecker-type) and then covered with the mixture of solutions A and B. During the amplification process, staining intensity can be checked in a light microscope adjusted to low light intensity. Development usually takes about 5 to 20 min, depending on primary antibody or nucleic acid probe concentration, incubation conditions, and the amount of accessible antigen or nucleic acid sequence in question. 2. Stopping bath: Photographic fixer (e.g., Agefix; Agfa Gevaert, Leverkusen, Germany; diluted 1:20) has been used in most IGSS protocols in combination with colloidal gold to stop the enhancement process immediately. Alternatively, a 2.5% aqueous solution of sodium thiosulfate had been suggested. Important Note: It has been shown that both these stopping–fixing treatments can be harmful when working

Immunogold–Silver Staining for Light Microscopy with labeled tyramides or with Nanogold. These fixers may remove some of the black staining signals and should therefore be handled with great care. We now recommend to interrupt the enhancement process simply by washing in distilled water (several changes), but this may lead to a slight increase in background staining. Alternatively, a very quick rinse (no more than 1 s) in sodium thiosulfate solution, followed by immediate washing in distilled water, may be done after development to stop the silver enhancement process.32 3. Slides can then be examined in a light microscope more carefully. If staining intensity is still too low, slides should be washed again in ultrapure water and then can be further developed in enhancement solution. 4. After final washing in deionized or distilled water, sections can be counterstained with nuclear fast red (Kernechtrot), and/or hematoxilin and/or eosin. For demonstrating nerve fibers or nerve cells, a light nuclear fast red or eosin counterstain alone is a good choice, or also just a light hematoxilin counterstain. For good overall counterstaining, a combination of the three stains appears adequate. 5. After washing, dehydrate sections or cell spots in graded alcohols and clear in xylene. DPX (BDH Chemicals, Poole, England, UK or EMS) or Permount (Fisher Scientific, Pittsburgh, PA, USA) are the preferred mounting media. Avoid using Eukitt (SigmaAldrich or Fluka) for silver stains. In our experience, the latter changed the intensity of silver stains after optimal development, sometimes even after weeks and months.

TECHNICAL HINTS AND DISCUSSION Choice of Methods: Indirect IGSS VS. Gold-Labeled S–ABC versus. Protein A–Gold–Silver Staining versus. Bridge Methods Indirect IGSS Method The original IGSS method as described by Holgate et al.53 and modified by Springall et al.88 was an indirect method. A secondary antibody adsorbed to colloidal gold and directed against immunoglobulins of the species in which the first layer antibodies were produced, binds to the unlabeled primary antibody, which reacts with the tissue-bound antigen. The sites of colloidal gold accumulation are subsequently visualized using AMG. The working scheme of indirect IGSS used in our laboratory for over a decade is given in Protocol 1. This protocol usually gives satisfactory immunostaining of many antibodies commonly applied in routine histopathology (Figures 4.1 through 4.4) and often proved excellent in detecting traces of substances within nerve or glial fibers (Figures 4.3 and 4.4) or neuroendocrine cells in formalin-fixed or Bouin’s fluid-fixed, paraffin-embedded specimens. However, immunostaining of intranuclear substances, such as estrogen receptors, progesterone receptors, and mitotic antigens (e.g., Ki-67, protein p53) may not work successfully with the indirect technique employing colloidal gold particles. Labeled Bridge IGSS Method Several methods offering increasing sensitivity and detection efficiency in IHC have been discussed (e.g., see review in Reference 103). One of the successful approaches is the introduction of labeled or unlabeled antibody(ies) between the primary layer and the labeled final antibody layer. Labeled and 57

Gold and Silver Staining unlabeled bridge methods for non-AMGamplified immunogold staining have been proposed.30,33 We also applied the very same approach for IGSS a few years later, using a secondary and a tertiary antibody both labeled with 5 nm colloidal gold particles (Hacker, unpublished data, 1984). Our results with a labeled bridge IGSS method were comparable with parallel sections stained with indirect IGSS, and the number of immunostained structures seemed identical in both methods. However, the labeled bridge IGSS tended to give higher unwanted background staining. Streptavidin–Biotin-IGSS Methods Avidin–biotin methods employing colloidal gold-adsorbed streptavidin were combined with silver enhancement by a number of authors.10,20,39 After application of primary antibodies and a biotinylated second layer, streptavidin absorbed to colloidal gold (5 nm) and subsequent AMG were used. The staining results with colloidal gold were comparable to those achieved with indirect IGSS. A great potential was already seen with the use of streptavidin–colloidal gold in applications such as nonradioactive ISH.12,40,66,104 Dramatic improvement began with the introduction of streptavidin covalently bound to Nanogold50,51 (see Chapter 6). Whereas differences were not so striking for membrane-bound or intracytoplasmic antigens including neuropeptides, we obtained surprisingly stronger staining for intranuclear substances, both by IHC and by ISH. Although a systematic evaluation of the potential of Nanogold–silver staining (NGSS) was not completed until recently (Cheung, unpublished), applications in nonradioactive ISH and related techniques (in situ PCR, in situ 3SR [in situ self-sustained sequence replication-based amplification]) were thoroughly tested.42,45–49,63, 58

66,94,97,101,104,107–110

Chapter 9 contains the best protocols obtained from these tests. The results of our comparisons for IHC showed that the Nanogold-labeled S–ABC IHC method was comparable, but not superior, to S–ABC peroxidase method for staining a variety of histopathological and tumor markers in routinely processed paraffin-embedded tissue. Nanogold S–ABC gave superior results for certain antigens, while the S–ABC peroxidase system worked better for others. The results also confirmed that both Nanogold-labeled and peroxidase-labeled S–ABC methods were more sensitive than indirect colloidal gold IGSS, especially for staining steroid receptors (Cheung, unpublished).

Protein A–Gold–Silver Staining Roth in 1982 introduced protein A–gold into LM applications.56,77 Its combination with AMG can give a highly efficient alternative to IGSS methods employing secondary antibody-gold probes. This special method of IGSS was developed by Fujimori and Nakamura and named “protein A gold–silver staining”25–27 (see Chapter 5). Iodine pretreatment was not used in their original protocol, and in the silver enhancement step, hydroquinone was replaced by bromohydroquinone. The staining results seem to be comparable with those obtained with the indirect IGSS working scheme given in Protocol 1. Black granular immunostaining is seen together with minimal nonspecific background. Similar techniques were reported by other authors too.29,67 Essential Steps in IGSS/NGSS Treatment with Lugol’s Iodine The original authors of IGSS had already reported that pretreating the sections with Lugol’s iodine was essential for

Immunogold–Silver Staining for Light Microscopy obtaining a high detection efficiency and sensitivity. Since then, most authors have confirmed this finding, although the exact process is still not clearly understood(27, 37,57,84,88 (see Chapter 5). Holgate et al.53 proposed that the weak oxidizing activity of the halogen was a necessary step for demonstrating most antigens. However, in later publications staining protocols were used that avoided iodine pretreatment and still resulted in successful immunostaining.61,102 Following our and others’ experience, iodine treatment with subsequent decolorization using sodium thiosulfate often increases the staining efficiency, at least for most antigens in paraffin sections. Without iodine pretreatment, the AMG process has to be prolonged, and staining quality is inferior when only small amounts of antigen are present. The effect of iodine pretreatment tends to be variable for cryostat or resin sections, so that the pretreatment may sometimes be detrimental to the staining quality. It should also be kept in mind that this treatment could also have adverse effects on the antigenicity of some substances. This may explain why certain antigens, easily detectable with immunoenzyme methods (e.g., chromogranin-A and estrogen receptors), sometimes present problems with IGSS and even NGSS. In our laboratories, we generally use Lugol’s iodine pretreatment followed by sodium thiosulfate in paraffin sections, and occasionally for resin or cryostat sections. As opinions on the mechanisms of iodine pretreatment still appear to be diverse, and further studies are needed for clarification, we would like to suggest that the reader test for need of iodine pretreatment in their own experimental systems. Washing Buffers, Detergents, and Gelatin Various washing buffer systems have been suggested in the literature. Tris-

buffered saline (TBS) was most commonly used.53,88 Experiments have shown that high salt concentrations in the buffer used before applying the primary antibody, as well as adjustment of the pH to 8.2 before immunogold incubation, reduce nonspecific background staining.88 The addition of detergents, such as Tween 20 or 80 or Triton X-100, may help to improve the detection of intracellular antigens by washing out the remaining lipid contents of cell membranes. Today, most commercial immunogold reagents available are of better quality than in the early beginnings of the IGSS technique, and modifications easing the buffer system setup can now be used without considerable loss of staining quality. From the use of two different buffer systems (TBS, pH 7.2 and then pH 8.2), we had first switched to one Tris buffer system for the whole procedure (TBS, pH 7.6), and now we simply use PBS (pH 7.6) throughout. Addition of 1% BSA (stored as 1-mL frozen aliquots) is advisable to reduce nonspecific staining, and the addition of 0.1% fish gelatin does strongly help to reduce unspecific precipitation of silver grains onto areas of the cells–section not containing the substance to be detected. Primary Antibodies Immunogold–silver methods have been successfully used for the demonstration of a variety of substances including regulatory peptides, immunohistopathology markers, such as intermediate filament proteins, enzymes, and cell surface antigens, plant antigens, receptors, bacteria, and viruses9,21,37–39,53,54,88,95,105 (Figures 4.1 through 4.5). The use of polyclonal rabbit and guinea pig antibodies, as well as monoclonal mouse or rat antibodies, has been reported. Sections in most protocols were either incubated overnight at 4°C or for 60 to 90 min at room temperature with 59

Gold and Silver Staining primary antibodies. Antibody dilutions in many cases can be the same or even higher than in other IHC techniques, and titration tests are necessary when setting up a new system. Repeated application of the primary antibody allows even higher dilutions and may further increase the detection efficiency.83 As with other highly efficient staining methods, incubation time may also be considerably shortened to give a rapid staining (30 min for the whole procedure), and microwave-mediated incubation may be beneficial (Gu and Hacker, unpublished). For such setups, antibody concentrations need to be increased, and the system needs to be carefully optimized. Specificity Controls As in other IHC techniques, it is crucial to perform specificity controls, at least at the beginning of a new system (new antibodies, new tissues, and new laboratory setting) and from time to time, to check if the specificity is still given. As IGSS methods are highly sensitive, such controls are particularly valuable for validating the staining obtained. As discussed in literature,72,73,103 such controls should at least contain: (1) omission or replacement of primary and/or secondary antibody; (2) pre-absorption controls for the primary antibody; and (3) the use of tissue sections known to contain the antigen in well-characterized structures. With these controls, it is easy to find out if there is argyrophilic– argentaffin staining caused by the AMG system, fixation, or the embedding protocol, and if the primary and secondary antibodies label their antigens specifically. To optimize the appearance of the staining, it is necessary to perform titration tests to find out the optimum dilution of primary and secondary immunoreagents. 60

Gold-Labeled Antibodies For indirect colloidal gold IGSS (Figures 4.1 through 4.4), affinity-purified secondary antibodies, adsorbed to colloidal gold (often referred to as immunogold reagents), can be obtained from a number of manufacturers. These include Amersham Pharmacia Biotech (Uppsala, Sweden), Affinity Research (Exeter, England, UK), Cambridge Research Biochemicals (CRB; Cambridge, England, UK), Jackson Immunoresearch Laboratories (West Grove, PA, USA), Aurion (Wageningen, The Netherlands), and BioClin and BioCell (both: Cardiff, Wales, UK) and others (Table 4.1). Dilutions of immunogold reagents should be carefully optimized and are usually between 1/25 and 1/200. The buffer system used to dilute immunogold should contain 1% BSA to prevent aggregation of gold particles and to reduce overall background staining.88 In our experience, the diluted immunogold reagent may be kept at 4°C for at least several days without significant loss of light microscopical labeling properties. EM grade immunogold reagents (1–5 nm in diameter) should be used even for LM applications—our results obtained with EM grade reagents were usually better than those obtained with LM grade reagents. For LM application, there is no need to centrifuge the immunogold reagents, as in EM use, to remove microaggregates of gold. As shown by Lackie et al.,61 immunolabeling is greatest with particles of a small size, which can penetrate sections better and achieve higher labeling densities than larger ones. Chemical permeabilization of cell membranes by detergents and embedding medium (if resin sections are used) also aids penetration of colloidal gold. For NGSS (Figures 4.5 and 4.6), Nanogold covalently bound to streptavidin is the best choice. These particles have a

Immunogold–Silver Staining for Light Microscopy 1.4 nm core or crystalline gold, surrounded by organic molecules allowing the covalent binding as opposed to adsorption by electrochemical forces used to attach covalent gold to macromolecules. Two-label streptavidin, bound to Nanogold and a fluoresceinating substance such as Cy3 (Amersham Pharmacia Biotech) or fluorescein isothiocyanate (FITC) contained in one complex, is available under the term FluoroNanogold (Nanoprobes) (Figure 4.6). As Nanogold is extensively described in Chapter 6 by its original inventor, we do not go into detail in the present chapter. Glutaraldehyde Postfixation Postfixation in 2% glutaraldehyde may help to prevent release of immunogold reagent from its binding sites in the low pH environment of most AMG solutions. This step is necessary for only some batches of colloidal gold-labeled macromolecules, but not for the covalently-bound Nanogold. If glutaraldehyde is used in PBS (pH 7.2), this solution can be reused for a few times and kept at 4ºC for 2 weeks. Autometallography AMG was worked out in the nineteenth century as a photographic technique.2 With his successful adaptation of the technique to microscopy,13–18 Danscher started a real novelty in microscopy. Not only did he reliably detect, for the first time, various catalytic heavy metals within tissue sections at the LM and EM, he also provided the basis for a great majority of highly amplifying silver enhancement procedures now commonly used in IHC, ISH, and related techniques. In the very same year, 1983, Danscher15 and Holgate and colleagues53,54 simultaneously described the first use of AMG for colloidal gold detection in enzyme histochemistry and

IHC, respectively. In Chapter 2, Danscher and colleagues describe AMG and its use for detection of tissue metals in detail. Chapter 3, by Hainfeld and Powell, describes the use of silver salt-based or gold salt-based AMG for IHC and related techniques. Therefore, in the present chapter, we would like to provide some additional hints based on our experience. The following development process is essentially the same for all available AMG solutions, regardless of whether silver lactate, silver acetate, silver bromide, silver nitrate, or other silver and gold salts are used. However, the light sensitivity of the different AMG compositions is quite different, depending on the silver or gold salt used. Also, the speed of development, the optimum pH values, and the degree of unspecific reactions vary greatly. In general, the AMG process results in the creation of shells of metallic silver around the gold particles.15,17,18 For detecting catalytic heavy metals in tissues, it is important to include a protective colloid in the AMG solution. A crude form of gum Arabic (obtainable inexpensively from Schnappsproducing industries) may be used to slow down the formation of metallic silver for better tolerance to diffuse daylight.15,40 In IGSS, the grains of gold particles bound to second-layer antibodies or to streptavidin catalyze the reduction of silver ions to metallic silver in the presence of a reducing substance. With time, the gold particles are thus encapsulated in growing shells of metallic silver, which gradually become more and more visible at LM level. During development, the initially unstained sections first become macroscopically light brownish or grayish and then relatively rapidly become darker and darker. Usually, when some notable degree of grayness is reached, one should have the staining in the LM adjusted to relatively low-intensity illumination. With time, the 61

Gold and Silver Staining laboratory technician will gain skill to recognize the optimum development simply by the degree of grayness seen with naked eye. In most cases, it is not sensible to start checking in the LM before any gray color is visible. As the development proceeds, the AMG solution turns grayish and finally black, and it is essential to stop the enhancement process early enough before the AMG solution itself is markedly gray or even black. If the blackening goes rather fast, or if the solution turns white, it is probably caused by impurities in the distilled water or glassware. The formation of a white precipitate may also occur and is related to the presence of chloride ions in the solution. On the whole, AMG is essentially the same process that is used in photographic physical development.2 A variety of different AMG procedures may be used for IGSS/NGSS. Most of them are based on Danscher´s AMG developer employing silver lactate as the source.13,14 However, other silver salts may be used, such as silver nitrate,100 silver bromide,44 or silver acetate.40,87 While silver nitrate may have some advantages in certain systems,25–27,59,83 its major disadvantage is the unspecific argyrophilic reactions often present, e.g., in collagenous structures.44 Due to this fact, we suggest to avoid silver nitrate as a silver ion donor for AMG in IGSS/NGSS. Less markedly, silver lactate also leads to some unspecific argyrophilic-type reactions when combined with IGSS, but these appear after a much longer enhancement time than with silver nitrate, and late enough to cause no problem in IGSS. However, silver lactate AMG was found to be more sensitive to daylight than the silver acetate AMG modification established jointly by us.40 The advantage of using silver acetate for AMG is that one can visually control the development process using a normal bright-field LM.40,87 Inclusion of 62

crude gum Arabic as colloid, as also described in the original AMG solutions of Danscher,13,14 prolongs the process and makes it even more controllable, as do low concentrations of the reducing substance, hydroquinone. Comparisons between sections exposed to silver lactate and silver acetate developers showed no significant difference in the number of structures stained; however, the degree of background staining was often lower when silver acetate was used as ion donator, due to visual control of the intensification process.40 Silver acetate itself is not as light sensitive as silver lactate. However, in solution and in the presence of hydroquinone as the reducing substance, the process is in effect comparable with the original silver lactate developers. The important parameters that define the light sensitivity and the speed of development seem to be mainly the concentrations of silver ion donor and reducing agent used, temperature, and the type of colloid and other chemicals used in the enhancement solutions. Whenever highly efficient AMG procedures are used, it is advisable to shield the preparations from very bright daylight during the enhancement process. With silver acetate AMG, sections may usually stay in daylight if the substances to be detected are present in high quantities. If prolonged enhancement duration is needed to detect antigens of low antigen content, the development jar should be covered by a dark box until the sections start to turn grayish. This can be checked macroscopically in daylight, and the degree of immunostaining can be monitored by using a LM adjusted to low light intensity. Compared to some commercially available silver enhancement kits, silver acetate AMG has the advantage of being less expensive, while giving the same degree of highly efficient immunostaining also offered by silver lactate AMG.

Immunogold–Silver Staining for Light Microscopy The purity of water is crucial to obtain optimal staining results. For all washes before the enhancement process and in the AMG setup, it is recommended to use glassdistilled water. Even minimal impurities (e.g., chloride) will induce a precipitate and render the enhancement solutions useless. For the same reasons, it is highly important to use very clean glassware and to avoid contact between metallic objects (use plastic forceps) and the silver solutions. Silver deposits of IGSS sections have been analyzed by LM and EM in combination with computer image analysis by Lackie et al.,61 and important data were also given by other authors.8,22,34,79,100 After 7 min of enhancement using silver lactate AMG without gum Arabic, the volume of silver deposited on the surface of 5 nm gold particles was found to be on average approximately 5000 times that of the original gold particle. If the gold particles are sufficiently close, fusion of silver particles occurs. Silver enhancement usually leads to grayish or brownish-black stained signals. Color development of silver-intensified immunogold markers can offer additional advantages, such as for the use in multiple immunostaining. Many pathologists prefer nonblack immunostains, simply because they are used to the brown color of 3,3′-diaminobenzidine (DAB)-peroxidase. Enzyme-labeled IHC methods, however, give some kind of washy and diffuse signal, which is not at all comparable to the distinct particulate signal resulting from AMG development. However, there is one organ that may present problems for IGSS-mediated immunohistopathological diagnosis. In the lung, the black AMG deposits may not be distinguishable from black deposits such as tar in smoker’s lungs. For such applications, it would be a good idea to modify the black silver color to a colorful but still particulate version,

such as red, brown, blue, or green. In color photography, this problem has already been solved, and it appears a matter of time before we can adapt such reactions for LM. For this, some chemical reactions used in color photography may be applied in a method analogous to that used in color autoradiography.36 Techniques resulting in either a magenta-red or a cyan-blue final reaction product have been proposed.24,64 In certain applications, gold-toning cannot only add specificity and improve staining performance, but also change the black color of the silver precipitate into blue.1 Tissue Fixatives Tissues fixed in numerous ways had been applied for IGSS successfully, in combination with paraffin or resin embedding or cryostat sectioning.37,38,53,54,61,88 Depending on the substance to be demonstrated, excellent results have been obtained by using buffered 4% formaldehyde solutions, Bouin’s fluid, Stefanini’s–Zamboni’s fixative,90,106 glutar(di)aldehyde, paraformaldehyde, and others. Silver enhancement does not work acceptably if parabenzoquinone3 is used as a fixative, as it may function as catalyzer of the reaction and therefore lead to high levels of unspecific staining (Hacker, unpublished). The best combination is buffered formaldehyde or Bouin’s–Stefanini’s fixative, followed by paraffin embedding. Frozen sections are not well suitable for IGSS methods. If one has to use cryosections, some form of pretreatment is necessary to give successful staining. We found that, after the sections are dried for 1 h after cutting, sequential immersion first in absolute ethanol, followed by ethanol–xylene (1:1), pure xylene, ethanol–xylene (1:1), absolute ethanol, 70% ethanol (each step for 5–10 min), and finally distilled water may improve IGSS 63

Gold and Silver Staining quality of cryosections. This trick, however, has not been studied in detail yet. Antigen Retrieval Aldehyde-based fixatives such as formaldehyde and glutardialdehyde are known to cross-link amino acids in peptides and proteins, respectively. The chemical reactions behind this denaturation lead to the formation of new chemical bindings, e.g., between amino groups and between hydroxyl groups, going in parallel with hydrolysis. It is for this reason that in IHC, antigen binding sites are often masked, i.e., not exactly the same as they were on the natural molecule. The degree of these changes depends on how thorough the specimen fixation has been performed (concentration of fixative, temperature, duration, etc.). In IHC, it is necessary to find a compromise between good preservation of morphology and good accessibility of antigenic determinants. In histopathology, in particular, it is not always easy to reach this optimum, as specimens are often already overfixed at their arrival at the pathology department. This was the main problem in the beginning of diagnostic IHC, especially when most antibodies used for diagnostic work at this time (mainly in the 1990s) were not optimized for such working conditions. Today, the quality of antibodies is greatly improved, and there are also reliable techniques which allow detection of a wide variety of useful antigens. The quality of IHC has also improved. Instead of enzyme predigestion, which was often used earlier,103 a new technique of making masked antigens more accessible for the antibodies, termed “antigen retrieval,” was introduced in 1991 by Shan-Rong Shi.85,86 It is based on high-temperature heating of the sections in a bath of buffer in a microwave oven or pressure cooker. Although to be performed with caution, heat antigen retrieval techniques were a real break64

through in IHC, yielding high-quality immunostaining in those kinds of preparations usually available for routine diagnostics, i.e., formalin-fixed paraffin sections. Before the advent of heat antigen retrieval techniques, IGSS had already been shown to be an improvement over the standard IHC techniques of the early 1980s for most routine antigens. The superior sensitivity of IGSS allowed successful immunostaining of antigens not detectable on consecutive sections stained with other IHC techniques. This often applied even without enzyme pretreatment. However, heat antigen retrieval further improved staining quality of IGSS and can—according to our experience—be used in the very same way as in other IHC techniques. Further Applications of IGSS-Related Techniques Silver-Enhanced Nerve Tracing A tracer complex of wheat germ agglutinin and horseradish peroxidase conjugate coupled to colloidal gold for retrograde tracing of neuronal pathways at the LM level was successfully used by Menétrey.68 AMG gives reliable, highly specific, and sensitive labeling even if fixatives with a high percentage of formaldehyde are used. With this introduction, it is now possible to examine sections–tissues in both LM and EM using the same tracer. Silver-Enhanced Lectin–Gold Labeling Lectins are sugar-binding proteins or glycoproteins of nonimmune nature used to demonstrate tissue carbohydrates; they can also be labeled with gold particles.58,62,82,87,99 A method of AMG-enhanced lectin–gold has been described by King et al.,58 who have used wheat germ agglutinin (WGA) adsorbed to colloidal gold to demonstrate the presence of chitin on the surface of nematode eggs.

Immunogold–Silver Staining for Light Microscopy AMG considerably improved the LM and the scanning electron microscopy (SEM) detection of WGA-gold. The combination of avidin–biotin–IGSS and lectin histochemistry allowed the highly accurate localization of tissue glycoconjugates,87 and lectin–gold with subsequent AMG was also used to localize glycoconjugates at the tegument of tapeworms at LM and EM level.82

the AMG steps; (2) the complete alkaline phosphatase or beta-galactosidase methods; (3) the silver enhancement step; and (4) the peroxidase-based IHS method. Numerous other techniques applying IGSS in double or multiple immunostaining have been suggested.26,31,53,59,60,80,81

Multiple Immunostaining Using IGSS Methods

IGSS and related techniques can be used in a number of nonmicroscopic applications such as blotting and immunoassays.65,69,96,52 For most of these applications, IGSS gives higher sensitivity in comparison to other conventional methods.

IGSS can be successfully combined with immunoenzyme20,24,39,53 or immunofluorescence methods83 to detect two or more antigens in one section. Due to a masking effect of the black silver product, the main use of these combinations is the detection of antigens present in different compartments. Since IGSS methods often give optimal results using concentrations of primary antisera well beyond the range of sensitivity of the PAP method, it is possible to use primary antisera raised in the same species for double immunolabeling, provided that IGSS is applied first.39,57 This is also advisable, as AMG may negatively affect peroxidase-diamonobenzidineH2O2-stained preparations. In general, AMG does not seem to destroy the antigenicity of other substances present in the same section.84 However, when IGSS is combined with alkaline phosphatase,11 beta-galactosidase,6 and peroxidasebased70 methods for multiple immunostaining, the silver precipitate may be affected by the substrates of nonperoxidase immunoenzyme methods. In our own experiments, the silver deposit appeared to be completely removed while the gold remained adsorbed to the second layer antibodies. This could be avoided by carrying out the following sequence for quadruple immunostaining: (1) IGSS but without

Nonmicroscopic Applications of IGSS Methods

ACKNOWLEDGMENTS We would like to dedicate this chapter to the memory of Dr. David Springall. He died at a young age on January 27, 1997 and will be remembered as a dear friend, a true gentleman, and a forerunner in the development of IGSS technology. Much of our work related to using AMG/IGSS for DNA or RNA detection was carried out in close collaboration with Assoc. Prof. Dr. Ingeborg Zehbe (Heidelberg, Germany). For her help in typing and referencing in this manuscript, I am grateful to Mrs. Konstanze Hötzer (Salzburg, Austria). We also would like to thank Univ. Prof. Dr. Otto Dietze for his continuous support of our research interests. REFERENCES 1.Arai, R. and I. Nagatsu. 1995. Application of gold toning to immunogold–silver staining, p. 209-216. In M.A. Hayat (Ed.), Immunogold–Silver Staining. Principles, Methods, and Applications. CRC Press, Boca Raton. 2.Arens, H. and J. Eggert. 1929. Das Wachstum des kolloidalen Silbers in Gelatineschichten. Z. Elektrochem. 35:728-733.

65

Gold and Silver Staining 3.Bishop, A.E., J.M. Polak, S.R. Bloom, and A.G.E. Pearse. 1978. A new universal technique for the immunocytochemical localisation of peptidergic innervation. J. Endocrinol. 77:25-26. 4.Bobrow, M.N., D. Thomas, D. Harris, K.J. Shaughnessy, and G.J. Litt. 1989. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Meth. 124:279-285. 5.Bobrow, M.N., K.J. Shaugnessy, and G.J. Litt. 1991. Catalyzed reporter deposition, a novel method of signal amplification. II. Application to membrane immunoassays. J. Immunol. Meth. 137:103-112. 6.Bondi, A., G. Chieregatti, V. Eusebi, E. Fulcheri, and G. Bussolati. 1982. The use of β-galactosidase as a tracer in immunohistochemistry. Histochemistry 76:153-158. 7.Bonnard, C., D.S. Papermaster, and J.P. Kraehenbuhl. 1984. The streptavidin-biotin bridge technique: application in light and electron microscope immunocytochemistry, p. 95-111. In J.M. Polak and I.M. Varndell (Eds.), Immunolabelling for Electron Microscopy. Elsevier, Amsterdam. 8.Burry, R.W., D.D. Vandre, and D.M. Hayes. 1992. Silver enhancement of gold antibody probes in preembedding electron microscopic immunocytochemistry. J. Histochem. Cytochem. 40:1849-1856. 9.Cailliez, J.C., A. Boudrissa, D.W. Mackenzie, and D. Poulain. 1990. Evaluation of a gold–silver staining method for detection and identification of Candida species by light microscopy. Eur. J. Clin. Microbiol. Infect. Dis. 9:886-891. 10.Coggi, G., P. Dell’Orto, and G. Viale. 1986. Avidinbiotin methods, p. 54-70. In J.M. Polak and S. Van Norden (Eds.), Immunocytochemistry—Modern Methods and Applications. Wright, Bristol. 11.Cordell, J.L., B. Falini, W.N. Erber, A.K. Ghosh, Z. Abdulaziz, S. MacDonald, K.A.F. Pulford, H. Stein, and D.Y. Mason. 1984. Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J. Histochem. Cytochem. 32:219-229. 12.Cremers, A.F.M., N. Jansen in de Wal, J. Wiegant, R.W. Dirks, P. Weisbeek, M. van der Ploeg, and J.E. Landegent. 1987. Non-radioactive in situ hybridisation. A comparison of several immunocytochemical detection systems using reflection-contrast and electron microscopy. Histochemistry 86:609-615. 13.Danscher, G. 1981. Histochemical demonstration of heavy metals. A revised version of the sulphide silver method suitable for both light and electron microscopy. Histochemistry 71:1-16. 14.Danscher, G. 1981. Localization of gold biological tissue. A photochemical method for light and electron microscopy. Histochemistry 71:81-88. 15.Danscher, G. and J.O. Rytter Nörgaard. 1983. Light microscopic visualisation of colloidal gold on resinembedded. J. Histochem. Cytochem. 31:1394-1398. 16.Danscher, G. 1984. Autometallography. A new technique for light and electron microscopical visualization of metals in biological tissue (gold, silver, metal sulphides, and metal selenides). Histochemistry 81:331-335.

66

17.Danscher, G., G.W. Hacker, J.O. Norgaard, and L. Grimelius. 1993. Autometallographic silver amplification of colloid gold. J. Histotechnol. 16:201-207. 18.Danscher, G., G.W. Hacker, C. Hauser-Kronberger, and L. Grimelius. 1995. Trends in autometallographic silver amplification of colloidal gold particles, p. 1118. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Methods and Applications. CRC-Press, Boca Raton. 19.De Bault, L.E. and B.-L. Wang. 1994. Localization of mRNA by in situ transcription and immunogold–silver staining. Cell Vision 1:67-70. 20.De Mey, J., G.W. Hacker, M. De Waele, and D. Springall. 1986. Gold probes in light microscopy, p. 71-88. In J.M. Polak and S. Van Norden (Eds.), Immunocytochemistry—Modern Methods and Applications. Wright, Bristol. 21.De Waele, M., J. De Mey, W. Renmans, C. Labeur, K. Jochmans, and B. Van Camp. 1986b. Potential of immunogold–silver staining for the study of leukocyte subpopulations as defined by monoclonal antibodies. J. Histochem. Cytochem. 34:1257-1263. 22.Ellis, I.O, J. Bell, and J.D. Bancroft. 1988. An investigation of optimal particle size for immunohistological immunogold and immunogold–silver staining to be viewed by polarized incident light (EPI polarization) microscopy. J. Histochem. Cytochem. 36:121124. 23.Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemisty 8:1081-1083. 24.Fritz, P., J. Hoenes, J. Schenk, A. Mischlinsky, A. Grau, J.G. Saal, H.V. Tuczek, H. Multhaupt, and G. Pfleiderer. 1986. Color development of immunogoldlabelled antibodies for light microscopy. Histochemistry 85:209-452. 25.Fujimori, O. and M. Nakamura. 1985. Protein A gold–silver staining method for light microscopic immunohistochemistry. Arch. Histol. Jpn. 48:449452. 26.Fujimori, O. 1992. A double protein A–gold–silver staining method for tissue antigens in light microscopy. Histochemical J. 24:61-66. 27.Fujimori, O., T. Ueda, and K. Yamada. 1994. Effects of iodine pretreatment of sections upon immunogold–siver staining in light microscopic immunohistochemistry. Okajimas Folia Anat. Jpn. 71:319-323. 28.Geoghegan, W.D., J.J. Scillian, and G.A. Ackerman. 1978. The detection of human B-lymphocytes by both light and electron microscopy utilizing colloidal gold-labelled anti-immunogobulin. Immunol. Commun. 7:1-12. 29.Geuze, H.J., J.W. Slot, K. Yanagibashi, J.-A. McCracken, L.A. Schwartz, and P.-F. Hall. 1987. Immunogold cytochemistry of cytochromes P-450 in porcine adrenal cortex. Histochemistry 87:551557. 30.Giffin, B.F., K. Gao, R.E. Morris, and R.R. Cardell. 1993. Enhancement of antigenic site detection with gold labeled secondary and tertiary antibodies using the immunogold–silver staining method. Biotech. Histochem. 68:309-315.

Immunogold–Silver Staining for Light Microscopy 31.Gillitzer, R., R. Berger, and H. Moll. 1990. A reliable method for simultaneous demonstration of two antigens using a novel combination of immunogold–silver staining and immunoenzymatic labeling. J. Histochem. Cytochem. 38:307-313. 32.Graf, A.-H., A.L.M. Cheung, C. Hauser-Kronberger, N. Dandachi, R.R. Tubbs, O. Dietze, and G.W. Hacker. 2000. Clinical relevance of HPV 16/18 testing methods in cervical squamous cell carcinoma. Appl. Immunohist. Mol. Morphol. (AIMM) 8:300-309. 33.Gu, J., J. De Mey, M. Moeremans, and J.M. Polak. 1981. Sequential use of the PAP and immunogoldstaining methods for the light microscopical double staining of tissue antigens. Its application to the study of regulatory peptides in the gut. Regul. Pept. 1:365374. 34.Gu, J., M. Forte, N. Carson, G.W. Hacker, M. D’Andrea, and R. Rufner. 1995. Quantitative evaluation of immunogold–silver staining, p. 119-136. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Methods and Applications. CRC Press, Boca Raton. 35.Guesdon, J.L., T. Ternynck, and S. Avrameas. 1979. The use of avidin-biotin interaction in immunoenzymatic techniques. J. Histochem. Cytochem. 27:11311139. 36.Haase, A.T., D. Walker, L. Strowring, P. Ventura, A. Geballe, H. Blum, M. Brahic, R. Goldberg, and K. O’Brien. 1985. Detection of two viral genomes in single cells by double-label hybridisation in situ and colour autoradiography. Science 227:189-192. 37.Hacker, G.W., D.R. Springall, S. Van Noorden, A.E. Bishop, L. Grimelius, and J.M. Polak. 1985. The immunogold–silver staining method. A powerful tool in histopathology. Virchows Arch. A 406:449-461. 38.Hacker, G.W., J.M. Polak, D.R. Springall, J. Ballesta, A. Cadieux, J. Gu, J.Q. Trojanowski, D.Dahl, and P. Marangos. 1985. Antibodies to neurofilament proteins and other brain proteins reveal the innervation of peripheral organs. Histochemistry 82:581-593. 39.Hacker, G.W., D.R. Springall, A. Cheung, S. Van Noorden, and J.M. Polak. 1986. Immunogold–silver staining and its potential use in multiple staining and histopathology [abstract]. J. Histochem. Cytochem. 34:118. 40.Hacker, G.W., L. Grimelius, G. Danscher, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc tissues. J. Histotechnol. 11:213. 41.Hacker, G.W. 1989. Silver enhanced colloidal gold for light microscopy, p. 279-321. In M.A. Hayat (Ed.), Colloidal Gold: Principles, Methods, and Applications, Vol. 1., Academic Press, Orlando. 42.Hacker, G.W., A.-H. Graf, C. Hauser-Kronberger, G. Wirnsberger, A. Schiechl, G. Bernatzky, U. Wittauer, H. Su, H. Adam, J. Thurner et al. 1993. Application of silver acetate autometallography and gold–silver staining methods for in situ DNA hybridisation. Chin. Med. J. 106:83-92.

43.Hacker, G.W. and G. Danscher. 1994. Recent advances in immunogold–silver staining—autometallography. Cell Vision 1:102-109. 44.Hacker, G.W., G. Danscher, L. Grimelius, C. Hauser-Kronberger, W.H. Muss, J. Gu, and O. Dietze. 1995. Silver staining techniques, with special reference to the use of different silver salts in lightand electron microscopical immunogold–silver staining, p. 20-45. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Methods and Applications. CRC Press, Boca Raton. 45.Hacker, G.W., I. Zehbe, J. Hainfeld, J. Sällström, C. Hauser-Kronberger, A.-H. Graf, H. Su, O. Dietze, and O. Bagasra. 1996. High-performance Nanogold in situ hybridisation and in situ PCR. Cell Vision 3:209-214. 46.Hacker, G.W., W.H. Muss, C. Hauser-Kronberger, G. Danscher, R. Rufner, J. Gu, H. Su, A. Andreasen, M. Stoltenberg, and O. Dietze. 1996. Electron microscopical autometallography: immunogold–silver staining (IGSS) and heavy-metal histochemistry. Methods (Companion Methods Enzymol.) 10:257269. 47.Hacker, G.W. and G. Danscher. 1997. Immunogold–silver staining-autometallography: recent developments and protocols, p. 41-54. In J. Gu (Ed.), Analytical Morphology—Theory, Protocols, and Applications. Eaton Publishing, Natick, MA. 48.Hacker, G.W., C. Hauser-Kronberger, I. Zehbe, H. Su, A. Schiechl, O. Dietze, and R. Tubbs. 1997. In situ localization of DNA and RNA sequences: supersensitive in situ hybridisation using streptavidinNanogold-silver staining: minireview, protocols, and possible applications. Cell Vision 4:54-67. 49.Hacker, G.W. 1998. High-performance Nanogold-silver in situ hybridisation. Eur. J. Histochem. 42:111-120. 50.Hainfeld, J.F. 1987. A small gold-conjugated antibody label: improved resolution for electron microscopy. Science 236:450-453. 51.Hainfeld, J.F. and F.R. Furuja. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 30:177-184. 52.Hauser-Kronberger, C., G.W. Hacker, E. Arrer, and G. Danscher. 1995. Nonmicroscopical colloidal gold autometallography (AMGAu): use of immunogold–silver staining in blot staining and immuonassay, p. 289-297. In M.A. Hayat (Ed.), Immunogold–Silver Staining. Principles, Methods, and Applications. CRC Press, Boca Raton. 53.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird. 1983. Immunogold–silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31:938-994. 54.Holgate, C.S., P. Jackson, I. Lauder, P. Cowen, and C. Bird. 1983. Surface membrane staining of immunoglobulins in paraffin sections of nonHodgkin’s lymphomas using immunogold–silver staining technique. J. Clin. Pathol. 36:742-746. 55.Horisberger, M. 1981. Colloidal gold: a cytochemical marker for light and fluorescent microscopy and for transmission and scanning electron microscopy. Scanning Electron Microsc. II:9-31.

67

Gold and Silver Staining 56.Horisberger, M. and M.F. Clerc. 1985. Labelling of colloidal gold with protein A. A quantitative study. Histochemistry 82:219-223. 57.Jackson, P. 1985. Staining with silver and gold. Med. Lab. World 10:16-26. 58.King, T.P., L. Brydon, G.W. Gooday, and L.H. Chappell. 1987. Silver enhancement of lectin-gold and enzym-gold cytochemical labelling of eggs of the nematode Onchocerca gibsoni. Histochem. J. 19:281286. 59.Krenács, T., L. Krenács, B. Bozoky, and B. Ivanyi. 1990. Double and triple immunocytochemical labelling at the light microscope level in histopathology. Histochem. J. 22:530-536. 60.Krenács, T. and L. Krenács. 1997. Immunogold–silver staining (IGSS) for single and multiple antigen detection in archived tissues following antigen retrieval. Cell Vision 4:387-393. 61.Lackie, P.M., R.J. Hennessy, G.W. Hacker, and J.M. Polak. 1985. Investigation of immunogold–silver staining by electron microscopy. Histochemistry 83:545-550. 62.Leathem, A. 1986. Lectin histochemistry, p. 167-187. In J.M. Polak and S. Van Noorden (Eds.), Immunocytochemistry—Modern Methods and Applications. Wright, Bristol. 63.Liesi, P., J.-P. Julien, P. Vilja, F. Grosveld, and L. Rechardt. 1986. Specific detection of neuronal cell bodies: in situ hybridisation with a biotin-labeled neurofilament cDNA probe. J. Histochem. Cytochem. 34:923-926. 64.Liu, Y.F. 1987. An experimental investigation of the colored-IGSS method. Chung Hua Ping Li Hsueh Tsa Chih. 16:248-250. 65.Liu, Y.S., W.P. Du, and Z.X. Wu. 1996. Dotimmunogold–silver staining in the diagnosis of cysticercosis. Int. J. Parasitol. 26:127-129. 66.Löning, T., R.-P. Henke, P. Reichart, and J. Becker. 1987. In situ hybridisation to detect Epstein-Barr virus DNA in oral tissues of HIV-infected patients. Virchows Arch. A. 412:127. 67.Marchetti, A., M. Bistocchi, and A.R. Tognetti. 1987. Silver enhancement of protein A-gold probes on resin-embedded ultrathin sections. Histochemistry 86:371-373. 68.Menetrey, D. 1985. Retrograde tracing of neural pathways with a protein-gold complex. I. Light microscopic detection after silver intensification. Histochemistry 83:391-395. 69.Moeremans, M., G. Daneels, A. Van Dijk, G. Langanger, and J. De Mey. 1984. Sensitive visualisation of antigen antibody reaction in dot and blot immuneoverlay assays with immunogold and immunogold–silver staining. J. Immunol. Methods 74:353-360. 70.Nakane, P.K. and G.B. Pierce. 1966. Enzyme-labelled antibodies: preparation and application for the localization of antigens. J. Histochem. Cytochem. 14:929931. 71.Nyhlin, N., M. el-Salhy, O. Sandstrom, and O. Suhr. 1997. Evaluation of immunohistochemical staining of human duodenal endocrine cells after microwave anti-

68

gen retrieval. Histochem. J. 29:177-181. 72.Petrusz, P., P. DiMeo, P. Ordronneau, C. Weaver, and D.A. Keefer. 1975. Improved immunoglobulinenzyme bridge method for light microscopic demonstration of hormone-containing cells of the rat adenohypophysis. Histochemistry 46:9-15. 73.Petrusz, P., P. Ordronneau, and J.C.W. Finley. 1980. Criteria of reliability for light microscopic immunocytochemical staining. Histochem. J. 12:333-348. 74.Powell, R.D., C.M.R. Halsey, D.L. Spector, S.L. Kaurin, J. McCann, and J.F. Hainfeld. 1997. A covalent fluorescent-gold immunoprobe: simultaneous detection of a pre-mRNA splicing factor by light and electron microscopy. J. Histochem. Cytochem. 45:947-956. 75.Robinson, J.M. and T. Takizawa. 1996. Novel labeling methods for EM analysis of ultrathin cryosections, p. 894-895. In G.W. Bailey, J.M. Corbett, R.V.W. Dimlich, J.R. Michael, and N.J. Zaluzec (Eds.), Proc. Micros. Microanal. San Francisco Press, San Francisco. 76.Robinson, J.M. and D.D. Vandré. 1997. Efficient immunocytochemical labeling of leukocyte microtubules with FluoroNanogold: an important tool for correlative microscopy. J. Histochem. Cytochem. 45:631-642. 77.Roth, J. 1982. Applications of immunocolloids in light microscopy. Preparation of protein A-silver and protein A-gold complexes and their application for the localisation of single and multiple antigens in paraffin sections. J. Histochem. Cytochem. 30:691696. 78.Roth, J. 1983. The colloid gold marker system for light and electron microscopic cytochemistry. Immunocytochemistry 2:217-284. 79.Rufner, R., N.E. Carson, M. Forte, G. Danscher, J.Gu, and G.W. Hacker. 1995. Autometallography for immunogold–silver staining. in light and electron microscopy. Cell Vision 2:327-333. 80.Sako, H., Y. Nakane, K. Okino, K. Nishihara, M. Kodama, M. Kawata, and H. Yamada. 1986. Simultaneous detection of B-cells and T-cells by a double immunohistochemical technique using immunogold–silver staining and the avidin-biotin-peroxidase complex. Histochemistry 86:1-4. 81.Schaumburg-Lever, G. and M. Tronnier. 1989. Combination of the APAAP and the IGSS techniques for double-labeling with two different monoclonal antibodies. Am. J. Dermatopathol. 11:2932. 82.Schmidt, J. and W. Peters. 1987. Localization of glycoconjugates at the tegument of the tapeworms Ymenolepis nana and H. microstoma with gold labelled lectins. Parasitol. Res. 73:80. 83.Scopsi, L. and L.I. Larsson. 1985. Increased sensitivity in immunocytochemistry. Effects of double application of antibodies and silver intensification on immunogold and peroxidase-antiperoxidase staining techniques. Histochemistry 82:321-329. 84.Scopsi, L., L.I. Larsson, L. Bastholm, and M. Hartvig Nielsen. 1986. Silver-enhanced-colloidal gold probes as markers for scanning electron

Immunogold–Silver Staining for Light Microscopy microscopy. Histochemistry 86:35-41. 85.Shi, S.-R., M.E. Key, and K.L. Kalra. 1991. Antigen retrieval in formalin fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections. J. Histochem. Cytochem. 39:741-748. 86.Shi, S.-R., J. Gu, K.L. Kalra, T. Chen, R.J. Cote, and C.R. Taylor. 1995. Antigen retrieval technique: a novel approach to immunohistochemistry on routinely processed tissue sections. Cell Vision 2:6-22. 87.Skutelsky, E., V. Goyal, and J. Alroy. 1987. The use of avidin-gold complex for light microscopic localization of lectin receptors. Histochemistry 86:291-295. 88.Springall, D.R., G.W. Hacker, L. Grimelius, and J.M. Polak. 1984. The potential of the immunogold–silver staining method for paraffin sections. Histochemistry 81:603-608. 89.Springall, D.R. 1986. Immunocytochemistry in diagnostic cytology, p.547-567. In J.M. Polak and S. Van Noorden (Eds.), Immunocytochemistry—Modern Methods and Applications. Wright, Bristol. 90.Stefanini, M., C. DeMartino, and L. Zamboni. 1967. Fixation of ejaculated spermatozoa for electron microscopy. Nature 216:173-174. 91.Sternberger, L.A., P.H. Hardy, J.J. Cuculis, and H.G. Meyer. 1970. The unlabeled antibody-enzyme method for immunohistochemistry. Preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidase) and its use for identification of spirochetes. J. Histochem. Cytochem. 18:315-333. 92.Tajima, Y., K. Kato, N. Utsumi, and K. Hosoi. 1994. Computer-aided image analysis applied to immunogold–silver staining: evaluation of proliferating cell nuclear antigen (PCNA)-reactive sites in paraffin sections. Histochemistry 102:177-181. 93.Takizawa, T. and J.M. Robinson. 1994. Use of 1.4nm immunogold particles for immunocytochemistry on ultra-thin cryosections. J. Histochem. Cytochem. 43:1615-1623. 94.Tbakhi, A., G. Totos, C. Hauser-Kronberger, J. Pettay, D. Baunnoch, G.W. Hacker, and R.R. Tubbs. 1998. Fixation conditions for and RNA in situ hybridization: a reassessment of molecular morphology dogma. Am. J. Pathol. 152:35-41. 95.Teasdale, J., P. Jackson, C.S. Holgate, and P.N. Cowen. 1987. Identification of oestrogen receptors in cells of paraffin-processed breast cancers by IGSS. Histochemistry 87:185-187. 96.Terazawa, K., G.D. Griffiths, A.G. Leith, and M.A. Green. 1989. Electrophoretic determination of ricin using immunogold–silver staining-comparison with simple “protein dot” method. Nippon Hoigaku Zasshi. 43:303-309. 97.Totos, G., A. Tbakhi, C. Hauser-Kronberger, and R.R. Tubbs. 1997. Catalyzed reporter deposition: a new era in molecular immunomorphology. Nanogold-silver staining and colorimetric detection and protocols. Cell Vision 4:433-442. 98.Tronnier, M., G. Schaumburg-Lever, and G. Rassner. 1989. A comparison of immunohistologic technics in dermatopathology. Hautarzt 40:259-265.

99.Ueda, T., S. Kawamura, O. Hattori, and O. Fujimori. 1994. Dual staining methods for carbohydrates using lectin-gold–silver techniques. Okajimas Folia Anat. Jpn. 70:301-311. 100.Van Den Pol, A.N. 1985. Silver-intensified gold and peroxidase as dual immunolabels for preand postsynaptic neuromitters. Science 228:332-335. 101.Van Gijlswijk, R.P., H.J. Zijlmans, J. Wiegant, M.N. Bobrow, T.J. Erickson, K.E. Adler, H.J. Tanke, and A.K. Raap. 1997. Fluorochromelabelled tyramides: use in immunocytochemistry and fluorescence in situ hybridisation. J. Histochem. Cytochem. 45:375-382. 102.Van Laere, O., L. De Wael, and J. De Mey. 1985. Immuno gold staining (IGS) and immuno gold– silver staining (IGSS) for the identification of the plant pathogenetic bacterium Erwinia amylovora (Burill) Winslow et al. Histochemistry 83:397399. 103.Van Noorden, S. 1986. Tissue preparation and immunostaining techniques for light microscopy, p. 26-53. In J.M. Polak and S. Van Noorden (Eds.), Immuncytochemistry—Modern Methods and Applications. Wright, Bristol. 104.Varndell, I.M., J.M. Polak, K.L. Sikri, C.D. Minth, S.R. Bloom, and J.E. Dixon. 1984. Visualisation of messenger RNA directing peptide synthesis by in situ hybridisation using a novel single-stranded cDNA probe. Histochemistry 81:597-601. 105.Westermark, K., M. Lundqvist, G.W. Hacker, A. Karlsson, and B. Westermark. 1987. Growth factor receptors in thyroid follicle cells. Acta Endocrinol. Suppl. (Copenh) 281:252-255. 106.Zamboni, L. and C. De Martino. 1967. Buffered picric acid-formaldehyde: a new fixative for electron microscopy. J. Cell Biol. 35:148A. 107.Zehbe, I., G.W. Hacker, J. Sällström, E. Rylander, and E. Wilander. 1992. In situ polymerase chain reaction (in situ PCR) combined with immunoperoxidase staining and immunogold–silver staining (IGSS) techniques. Detection of single copies of HPV in SiHa cells. Anticancer Res. 12:2165-2168. 108.Zehbe, I., G.W. Hacker, W.H. Muss, J. Sällström, E. Rylander, A.-H. Graf, H. Prömer, and E. Wilander. 1993. An improved protocol of in situ polymerase chain reaction (PCR) for the detection of human papillomavirus (HPV). J. Cancer Res. Clin. Oncol. 119 (Suppl):22S. 109.Zehbe, I., G.W. Hacker, J.F. Sällström, E. Rylander, and E. Wilander. 1994. Self-sustained sequence replication-based amplification (3SR) for the in situ detection of mRNA in cultured cells. Cell Vision 1:20-24. 110.Zehbe, I., G.W. Hacker, H. Su, C. Hauser-Kronberger, J.F. Hainfeld, and R. Tubbs. 1997. Single human papillomavirus copy detection in formalinfixed samples: sensitive in situ hybridization with catalyzed reporter deposition, streptavidin-Nanogold, and silver acetate autometallography. Am. J. Pathol. 150:1553-1561.

69

5

The Protein A–Gold–Silver Staining Method Osamu Fujimori

INTRODUCTION Protein A–gold (PAG) complex techniques were originally used for postembedding immunoelectron microscopy,17 as was the immunoglobulin G (IgG)-gold complex (IgG-gold) method, also referred to as immunogold staining (IGS).2 PAG is a complex of colloidal gold particles and protein A, a protein that is isolated from the cell wall of Staphylococcus aureus. Since protein A binds to the Fc fragment of IgG among different animal species, such as human, rabbit, guinea pig, mouse, horse, bovine, and others,3,11 PAG has proven useful as a secondary immunoreagent in immunohistochemistry (IHC) and immunocytochemistry (ICC). The preparation of PAG has been described in detail by Roth,20 and PAG labeled with various sizes of gold particles is commercially available. The natural color of colloidal gold sol in transmitted light is pale to dark red. The shade depends on the particle size and preparative conditions, and the color is not abolished after complex formation with protein A (or IgG). Under brightfield light microscopy (LM), gold particles of sufficient size and number labeling antigen0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

bound antibody sites in tissue sections also visualize as red, without any revealing agents. This phenomenon has resulted in the development of a PAG staining method for LM.19 Specimens subjected to the PAG method can be dehydrated in ethanol, mounted in stable embedding media, and stored for long periods of time. However, the method has not been commonly used, because the procedure requires antibody and PAG at relatively high concentrations, and because its reaction products are lower in contrast than most other IHC techniques. In LM immunohistochemstry, the PAG staining technique has been efficiently applied in combination with silver enhancement (autometallography, AMG); this combined technique is named the protein A–gold–silver (PAGS) staining method.4 In this method, gold particles are enlarged by deposition of metallic silver on the surface of gold particles, appearing black in the LM (Figures 5.1 and 5.2). In addition to the inherent merits of the PAG staining method, PAGS staining offers a considerably higher detection efficiency, and gives a very high contrast and fine visibility of the reaction product. Because of the high detection sensitivity, the method

71

Gold and Silver Staining only utilizes small amounts of antibody and PAG,6,8 and therefore is more economical. In this chapter, detailed protocols of single and double PAGS methods for IHC at the LM level are presented. PROTOCOLS Protocol 1. Preparation of Tissue Specimens As with other immunogold–silver staining (IGSS) methods,13,14,23 the PAGS staining method is compatible with paraffin- or resin-embedded tissues preserved with a variety of commonly used fixatives. For electron microscopy (EM), resins tested include Epon, LR-White, and Lowicryl K4M. The method also can be used on cryostat sections.8 Protocol 2. Autometallography Several types of silver enhancement processes (AMG) have been successfully combined with the colloidal gold label system.1,4,12,16,17,22 In the PAGS staining method, a “physical” developer containing silver nitrate as silver ion donator was originally employed. Other types of developers can also be utilized. However, for the double PAGS staining methods7,10 described in this chapter, AMG solutions suited for slow development are recommended. In the present protocol, the silver nitrate-based AMG developer is described. Note: It is important to know that silver nitrate used in AMG may sometimes lead to unspecific argyrophilic–argentaffin reactions. We therefore would like to suggest to thoroughly test which developer is best suited for the system. For any AMG developer, the possible presence of unspecific argyrophilic–argentaffin reactions should be checked, e.g., according as specificity controls contained in this chapter. As 72

described in Chapters 2 and 3, ways to slow down the enhancement speed include addition of a protecting colloid, such as gum Arabic. A silver salt with very low amounts of unspecific reactions is silver acetate.12 The developing solution used at the silver enhancement step in the original PAGS method4 is a mixture of a silver nitrate used as a silver ion donator, gum Arabic as a protecting colloid, and bromohydroquinone as a reducing agent dissolved in low pH citrate buffer. It is basically very similar in its constituents to Danscher’s AMG developer,1 however, a different silver salt and a stronger reducer are used. Glass-distilled water is desirable for diluent and buffer. Two solutions are necessary, here referred to as solution A and solution B. Solution A consists of 45 mL 20% gum Arabic aqueous solution and 1 mL 10% silver nitrate aqueous solution. Solution B contains 15 mL distilled water, 200 mg bromohydroquinone, and 300 mg citric acid. The working solution is prepared by mixing solutions A and B immediately prior to use. Twenty percent gum Arabic solution is subjected to ultracentrifugation at 18 ,000 rpm at 0° to 4°C for 30 to 60 min, and the supernatant fluid is used. Centrifugation of gum Arabic solution is important and recommended, even if at a speed lower than 18 ,000 rpm. The supernatant can be stored for several months at 4°C and for at least several years at -20°C. Bromohydroquinone is a potent derivative of hydroquinone and exhibits a significantly higher developing capacity than hydroquinone.15 Bromohydroquinone in the above formula can be replaced with 200 mg hydroquinone and is also utilized commonly in AMG,17 although the developing time needed for sufficient reaction intensity is relatively prolonged. Alternatively, Danscher’s original AMG formulation with silver lactate, or a silver acetate-based developer can be used, both, if necessary, in

Protein A–Gold–Silver Staining Method combination with a protecting colloid.12,17 Silver enhancement is performed in a Coplin jar at 20°C. Light should be excluded by covering the setup with a red plastic shade, placing it in a dark box in an open laboratory, or carrying out the procedure in a dark room under a photographic safe lamp. Protocol 3. Protein A–Gold–Silver Staining Method The PAG technique for IHC was originally developed for EM,18 and Roth19 was the first to employ this nonsilver-enhanced PAG technique for LM. The PAGS staining methods were developed to overcome some drawbacks of the PAG method in LM.4,6 The following protein A–gold–silver staining method is based on those described by Fujimori and Nakamura,4 and Fujimori,6 and subsequently modified by Fujimori.8 All steps are performed at room temperature (approximately 20°C). Paraffin sections are deparaffinized in xylene, dehydrated in a graded ethanol series, and subjected to the following staining procedure. The sequence is started at step 1 for cryostat sections, and at step 2 when using resin sections. Incubation with special reagents such as ovalbumin, antibody, and PAG is carried out by applying 25 to 50 µL of reagent on a tissue section with a micropipet. Throughout the procedure, steps 2, 3, 6, and 7 are performed in a moist chamber. Staining Procedure 1. Hydrate sections in three changes of phosphate-buffered saline (PBS) (0.01 mol/L pH 7.4) for 5 min each. 2. Pretreat with 5% ovalbumin in PBS for 30 min. 3. Incubate with primary antibody dilut-

ed appropriately with PBS (0.01 mol/L, pH 7.4) containing 0.5% to 1% bovine serum albumin (BSA) (BSA-PBS) for 60 min. 4. Drain the antibody off with PBS. 5. Rinse in four changes of PBS for a total of 30 min. 6. Treat with 5% ovalbumin in PBS for 15 min. 7. Incubate with PAG diluted with PBS or BSA-PBS for 30 to 60 min. 8. Drain PAG off with PBS. 9. Rinse in four changes of PBS for a total of 30 min and then two changes of distilled water for a total of 5 min. 10. Place in silver enhancement working solution until a sufficient intensity of reaction products is obtained. Reaction intensity can be checked in the LM after rinsing in running water (preferable deionized or distilled water) for 30 sec to 1 min. If the staining intensity is insufficient, sections may be returned to the AMG development solution, following brief rinsing in distilled water. 11. Immerse in photographic fixer diluted 1:5 with distilled water, or 5% sodium thiosulfate aqueous solution, for 1 to 3 min. 12. Wash in running tap water for 10 min. 13. Counterstain with nuclear fast red (Kernechtrot; Chroma, Stuttgart, Germany) or 1% Evans blue if necessary. 14. Dehydrate in graded ethanols, clear in xylene, and mount in suitable embedding medium (See Chapter 1). Under optimal conditions, the reaction product seen in the LM first appears pale brown, then changes gradually to brown, purplish black, and finally strong black in color, without nonspecific background staining (Figures 5.1 through 5.3). In the PAGS staining, PAG with small gold particles (5 nm or less in diameter) 73

Gold and Silver Staining permits the use of antibody and PAG reagent at higher dilutions than that with larger particles (10 nm or bigger in diameter).6 The author routinely uses PAG with 5 nm colloidal gold particles (EY Laboratories, San Mateo, CA, USA) at 1:320 dilution. When 10 or 15 nm particles (EY Laboratories) are applied, these have to be used at higher concentrations (1:100 or 1:80). However, it takes a slightly longer time to silver enhance the small gold particles compared to those with higher gold particle diameters. Generally, a 5% concentration of ovalbumin is appropriate to inhibit nonspecific immunostaining. This concentration of ovalbumin does not interfere with specific binding between antigen and antibody, or between antibody and PAG, but effectively prevents nonspecific binding between antibody or PAG with tissue constituents other than the antigen to be immunostained. Higher concentrations of ovalbumin (or also of normal serum in the original IGSS method) may inhibit not only nonspecific antibody– or immunogold–tissue binding, but also specific antigen–antibody binding. However, a concentration lower than 5% is not sufficient to effectively prevent nonspecific staining.9 Lugol’s Iodine Pretreatment Pretreatment of paraffin sections with Lugol’s iodine (1% iodine in 2% potassium iodide aqueous solution)23 shortens the duration necessary for development, and this iodination followed by a short reducing step in 2.5% sodium thiosulphate aqueous solution can be interposed between steps 1 and 2 in the protocol. In our experiments, however, iodine treatment often induced nonspecific silver precipitate, and its effects were so pronounced, that the staining intensity of reactive histologic sites was difficult to control. The result was often an 74

overdevelopment of both the specific and nonspecific reaction products. Note: This phenomenon is known also for conventional IGSS methods. Careful testing and optimization of reagent types and their concentrations can often easily help to reduce these side effects, thereby however, earning the great benefit of obtaining a much higher sensitivity of the technique than without iodination. Points to look at here are the type of silver salt used, as well as the concentrations of antibody and gold reagent applied. The longer the washing in running tap water following iodine–sodium thiosulfate sequence, the longer the duration of development needs to be for the visualization of reaction products. If sections are washed overnight in running tap water, the duration needed for development is nearly comparable in length to that in the procedure without iodine treatment.9 Iodine is likely to act as an accelerator, promoting the interaction between silver ions and the reduction reagent. It appears possible that iodine remaining in sections shortens the duration for development in the PAGS and IGSS methods. Therefore, the overdevelopment of both the specific and nonspecific reaction products may be caused by this action of the iodine. In AMG development, silver grains have been postulated to grow asymmetrically by quick reactions of reducing substances with silver ions.1 It is likely that iodine could induce such reactions of the reducing reagents to result not only in enhancement of specific reaction products, but also in a strong nonspecific reaction. According to Hacker,13 it cannot be denied that iodine pretreatment may occasionally cause negative effects on the antigenicity of some substances. Thus, pretreatment with iodine should be carefully used in these procedures. Note: Simple specificity controls and comparison of the PAGS/IGSS results with

Protein A–Gold–Silver Staining Method

75

Figure 5.1. Schematic illustration of the principle of the PAGS staining procedure. Reaction products (silver grains) precipitates on the surface of gold particle during the course of AMG. These are first visualized as a pale brown color under LM. This color changes gradually to brown, then to reddish brown or reddish purple, and finally to black.

Gold and Silver Staining those obtained using other methods of immunostaining can easily give an answer here and help to decide if iodine is helpful in ones staining system or counterproductive. A silver reduction procedure following silver enhancement, using a specially adapted Farmer’s reducer from photography as suggested by Hacker,13 is effective to diminish overdevelopment and background silver deposition. It should be kept in mind that carrying out this process for too long could abolish the reaction products not only in the background, but in specific staining.9 Since Farmer’s reduction decreases the intensity of the essential staining, it should be used only to diminish overdevelopment of specific staining. It is better in every case to avoid over-development from the beginning.

Specificity Controls The specificity of immunolabeling with the PAGS technique should be evaluated under several controls as follows, especially when immunostaining involves a new antibody, a different tissue type, or a different specimen processing protocol. Tissue sections should be (1) incubated with protein A–gold only (omission of specific antibody), showing if any nonspecific absorption of the PAG to tissue sections is present; (2) incubated with antibody absorbed by its specific corresponding antigen (replacement of antibody with preabsorbed antibody), thus revealing specificity of the interaction between antibody and antigenic site on tissue sections; (3) incubated with nonlabeled protein A after incubation with antibody, then treated with protein A–gold, thereby elucidating the specificity of the interaction between protein A and immunoglobulins; (4) incubated with preimmune serum instead of antibody, thereby verifying the nonspecific absorption of antibody to tissue sections; and (5) treated with the developing solution only (omission of immunostaining), which reveals the possible presence of argyrophilic–argentaffin reactions by endogenous development in tissue sections in any AMG techniques. Protocol 4. Double Protein A–Gold–Silver Staining Method Using a Single Type of PAG and Two-Step Development

Figure 5.2. Rat pancreatic B cells stained by the PAGS method. Pancreatic B cells are demonstrated in black shades with antibody to porcine insulin raised in guinea pig and PAG (15 nm). (Original magnification, ×300.) Fixed in Bouin’s fluid and counterstained with nuclear fast red.

76

In double staining methods for tissue antigens, the two antigenic sites need to be clearly differentiated from each other, and the methods for their detection should be simple and reliable. For this purpose, a double PAGS staining method using PAG with a single size of gold particles has been established.7

Protein A–Gold–Silver Staining Method In the PAGS staining procedure using original PAGS AMG developer containing silver nitrate, reaction products are seen first as having a pale brown color, which changes gradually to shades of black during the course of the development. This special feature of the development in the PAGS procedure has been applied to establish a double staining method for the simultaneous visualization of two different antigens in the same tissue section by LM.7 The method consists of a sequence of two PAGS techniques, in which reaction products of the first and second staining procedures are colored black and brown, respectively (Figures 5.4 and 5.5). PAG labeled with 10 nm gold particles was originally used in this double PAGS staining method.

Figure 5.3. Rat pancreatic B cells stained by the PAGS method. In large magnification, B cell granules are shown as black granular reaction products. (Original magnification, ×1300.) Fixed in Bouin’s fluid and counterstained with nuclear fast red.

Staining Procedure The following double PAGS staining protocol has been described by Fujimori.7 All steps are performed at room temperature (approximately 20°C), and steps 3, 4, 7, and 8 are carried out in a moist chamber. In this procedure, the appropriate gold particle diameter is 10 or 15 nm. 1. Deparaffinize sections in xylene and hydrate in a graded ethanol series. 2. Rinse in three changes of PBS (0.01 mol/L, pH 7.4, containing 2.9% sodium chloride) for 5 min each. 3. Pretreat with 5% ovalbumin in PBS for 60 min. 4. Incubate with the first antibody diluted appropriately with 0.5% to 1% BSA-PBS for 60 min. 5. Drain the antibody off with PBS. 6. Rinse in four changes of PBS for a total of 30 min. 7. Treat with 5% ovalbumin in PBS for 15 min. 8. Incubate with PAG diluted 1:80 to 1:100 with PBS or BSA-PBS for 60 min. 9. Drain PAG off with PBS. 10. Rinse in four changes of phosphate buffer (PB) (0.01 mol/L, pH 7.4, without sodium chloride) for a total of 30 min. 11. Wash briefly in two changes of distilled water for a total of 5 min. 12. Place in the silver nitrate AMG developer until brown reaction product is visible. Reaction intensity can be checked under the LM after rinsing in deionized or distilled water for 30 sec to 1 min. If the staining intensity is insufficient, sections can be returned to the AMG development solution, following thorough rinsing in distilled water. 77

Gold and Silver Staining 13. Rinse in photographic fixer diluted 1:5 with distilled water or aqueous 5% sodium thiosulfate solution for 1 min. 14. Wash in running tap water for 10 min. 15. Repeat steps 2 to 14, replacing the first antibody in step 4 with the second antibody. AMG enhancement in the second staining round is performed until reaction products of the second staining appear brown under a light microscope. At the same time, the staining product of the first staining round is automatically becoming black—see below. 16. Counterstain with 1% Evans blue if necessary. Nuclear fast red is not so well suited here, as it will not contrast well with the brown AMG product of the second staining run. 17. Dehydrate in graded ethanols, clear in xylene, and mount in stable embedding media. According to Roth,19 the intermediate removal of the first layer of antibodies is unnecessary in this double PAGS staining method, because PAG methods can be used sequentially without interfering with each other. In this double PAGS staining method, the reaction product for the first antigen is visualized brown or purplish black as a result of the development of the first staining (Figure 5.4). The colors of these reaction products gradually change to black during the subsequent development of the second staining, in which the reaction products against the second antigen are revealed brown (Figure 5.4). Eventually, the reaction product of the first staining is black, whereas those of the second staining remain brown (Figures 5.4 and 5.5). According to the original study7 using PAG with 15 nm gold particles, the appropriate duration of the first and second developments is 30 to 40 min each, although it depends on dilution of antibodies and PAG and other immunological 78

conditions. Lugol’s iodine pretreatment should not be applied in this double staining procedure. Protocol 5. Double Protein A–Gold–Silver Staining Method Using Two Types of PAG and One-Step Development In the PAGS staining method, the time required to visualize reaction products also depends upon the size of the gold particles used. Here, the use of small gold particles6 requires a longer duration of development than larger ones.4 If gold particles of different sizes are developed simultaneously, reaction products obtained exhibit varying colors during the same development time (Figure 5.6). Utilizing such characteristics of the silver enhancement method, we have established a convenient dual PAGS staining method for LM, using two types of PAG, which can detect a couple of antigenic sites by only one development step.10 In this double PAGS staining method, two types of PAG labeled with different sizes of gold particles are used. In our initial report,10 one PAG was labeled with 15 nm gold particles for the first staining round, and the second PAG with 5 nm gold particles. Staining Procedure 1. Deparaffinize sections in xylene and hydrate in a graded ethanol series. 2. Rinse in three changes of PBS (0.01 mol/L, pH 7.4, containing 2.9% sodium chloride) for 5 min each. 3. Treat with 5% ovalbumin in PBS for 30 min. 4. Incubate with the first antibody diluted in 0.5% to 1% BSA-PBS for 60 min. 5. Drain the antibody off and rinse in four changes of PBS for a total of 30 min.

Protein A–Gold–Silver Staining Method

79

Figure 5.4. Schematic illustration of the principle of the double PAGS staining method with a single type of PAG and two-step development. In LM, reaction products for the first antigens are obtained in shades of brown or reddish purple by the first development. These reaction products change their colors to black during the subsequent development in the second staining in which reaction products for the second antigens are visualized brown or reddish purple.

Gold and Silver Staining 6. Treat with 5% ovalbumin in PBS for 15 min. 7. Incubate with PAG solution using 15 nm particles (15 nm PAG) diluted 1:80 with PBS or BSA-PBS for 60 min. 8. Drain 15 nm PAG off and rinse in four changes of PBS for a total of 30 min. 9. Treat with 5% ovalbumin in PBS for 30 min. 10. Incubate for 60 min with the second antibody diluted with BSA-PBS. 11. Drain the second antibody off and rinse in four changes of PBS for a total of 30 min. 12. Treat with 5% ovalbumin in PBS for 15 min.

Figure 5.5. Rat adenohypophysis subjected to the double PAGS staining method with a single type of PAG (15 nm) and two-step development. Immunoreactive adrenocorticotrophic hormone (ACTH) cells for the first staining are irregular shaped and stained black, whereas immunoreactive growth hormone (GH) cells for the second staining are round or oval shaped and revealed pale color (brown). Immunolabeled with rabbit antibodies for synthetic ACTH and rat GH. Bouin’s-fixed and slightly counterstained with Evans blue. (Original magnification, ×480.)

80

13. Incubate with PAG solution using 5 nm particles (5 nm PAG) for 60 min. Diluted 1:320 with PBS or BSA-PBS. 14. Drain 5 nm PAG off and rinse in four changes of PB (0.01 mol/L, pH 7.4, without sodium chloride) for a total of 30 min. 15. Rinse briefly in distilled water. 16. Place in the AMG developer described above until reaction products of both black and brown or reddish purple shades are obtained. 17. Rinse in photographic fixer diluted 1:5 with distilled water or in aqueous 5% sodium thiosulfate solution for 1 min. 18. Wash in running tap water for 10 min. 19. Counterstain with 1% Evans blue aqueous solution if necessary. 20. Dehydrate in a graded ethanol series, clear in xylene, and mount in suitable mounting medium. All steps were performed at room temperature. Incubations of ovalbumin, antibodies, and PAG were carried out in a moist chamber. In double PAGS staining, use of PBS containing 2.9% (0.5 mol/L) sodium chloride tends to result in fine immunostaining without nonspecific labeling as compared with using of PBS containing 0.9% sodium chloride. A sufficient contrast is obtained between the two shades of reaction products. The black reaction product for the first antigen can be easily differentiated from the brown or reddish purple reaction product for the second (Figure 5.7). Such a difference in shade between the two reaction products is believed to be due to both the diameter of gold particles and the amount of metallic silver grains precipitated upon the catalytic specks (gold particles). Fifteen nanometer gold particles are larger in diameter and surface area than 5 nm particles. Therefore,

Protein A–Gold–Silver Staining Method

81

Figure 5.6. Schematic illustration of the principle of the double PAGS staining method with two types of PAG and one-step development. The first antibody binds to the first antigen and then the 15 nm PAG reacts with the Fc fragments of the first antibody. Subsequently, the second antibody binds to the second antigen, and Fc fragments of excess second antibodies occupy all available protein A reactive sites of the 15 nm PAG on the first antigenic site. Therefore, the 5 nm PAG used in the second staining reacts only with the Fc fragment of the second antibody, which is bound to the second antigenic site, and the first antibody need not be removed.19 During AMG, metallic silver precipitates on the surface of all of the gold particles concerned. Due to differences in the volume of the reaction products, a large gold particle is recognized as black in shade, whereas a small one is revealed as brown to reddish purple under LM.10

Gold and Silver Staining silver grains precipitated on the surface of the larger particle would be greater in number than those on 5 nm particles at any defined point during development. Thus, the reaction products obtained by the first staining on the first antigenic sites are larger in volume than those by the second staining on the second antigenic sites. This results in shade differences between the two reaction products; larger ones are black, whereas smaller ones are brown or reddish purple (Figures 5.6 and 5.7). Higher or lower concentration of both sizes of PAG may decrease the contrast of the two reaction products or eliminate the differences altogether.10 The AMG developer used in this double staining needs a relatively longer duration of development for the visualization of the reaction products as compared to developers containing silver lactate1,16 or

silver acetate12 In this double PAGS method using slow development AMG, the shade of reaction products can easily be controlled, and the process of development can be stopped at any time. Iodine pretreatment should not be applied to double PAGS staining procedures. According to literature on double PAGS staining, removal of the first layer of antibodies following the first staining is not needed here, since PAG can be sequentially applied to sections without interfering with each other.19 Fc fragments of the first antibodies on the first antigenic sites are occupied by 15 nm PAG, and, therefore, the first antibody layer would not prevent the second staining (Figure 5.6). This double PAGS staining method is simple and economical, as the reaction products of two different shades can be obtained by using a single AMG visualization procedure.

Figure 5.7. Rat pancreatic insular cells stained by the double PAGS staining method with two types of PAG and one-step development. Peripheral pancreatic A cells are stained black with antiporcine glucagon antibody raised in rabbit and PAG (15 nm) (first staining), whereas a large number of B cells are stained pale (brown) with antiporcine insulin antibody raised in guinea pig and PAG (5 nm) (second staining). Perfusion-fixed with Bouin’s fluid and counterstained with Evans blue. (Original magnification, ×240.)

82

Protein A–Gold–Silver Staining Method Protocol 6. Double Immunostaining by Combined Protein A–Gold–Silver Staining and Immunoperoxidase Immunohistochemistry Double immunostaining by combined PAGS staining and immunoperoxidase IHC was described by Fujimori. 5 All steps are performed at room temperature (approximately 20°C). Throughout this procedure, steps 3, 4, 7, 8, 17, 18, and 21 are performed in a moist chamber. Staining Procedure 1. Deparaffinize sections in xylene and hydrate in a graded ethanol series. 2. Rinse in three changes of PBS (0.01 mol/L, pH 7.4, containing 0.9% sodium chloride) for 5 min each. 3. Pretreat with 5% ovalbumin in PBS for 30 min. 4. Incubate with the first antibody for the first antigen diluted highly with 0.5% to 1% BSA-PBS for 60 min or more at room temperature or overnight at 4°C. 5. Drain the first antibody off with PBS. 6. Rinse in four changes of PBS for a total of 30 min. 7. Treat with 5% ovalbumin in PBS for 15 min. 8. Incubate with PAG appropriately diluted with PBS or BSA-PBS for 60 min. 9. Drain PAG off with PBS. 10. Rinse in four changes of PB (0.01 mol/L, pH 7.4, without sodium chloride) for a total of 30 min. 11. Wash briefly in two changes of distilled water for 5 min. 12. Place in the AMG developer until reaction product appears black.

13. Rinse in photographic fixer diluted 1:5 with distilled water or 5% aqueous sodium thiosulphate solution for 1 min. 14. Wash in running tap water for 10 min. 15. Place in methanol containing 0.03% hydrogen peroxide to suppress the activity of endogenous peroxidase. 16. Rinse in three changes of PBS for 5 min each. 17. Treat with 5% normal serum of the animal species of the horseradish peroxidase (HRP)-labeled secondary antibody (used at step 21) in PBS for 15 min. 18. Incubate with the second antibody against the second antigen diluted with BSA-PBS for 60 min. 19. Drain the second antibody off with PBS. 20. Rinse in four changes of PBS for a total of 30 min. 21. Incubate with HRP-labeled secondary antibody diluted with BSA-PBS for 30 min. 22. Immerse in 100 mL of freshly prepared PB containing 20 mg of 3,3′ diaminobenzidine hydrochloride (DAB) and 17 µL of concentrated hydrogen peroxide until the DAB-hydrogen peroxide reaction products for the second antigen appear brown. 23. Wash in running tap water for 10 min. 24. Counterstain with 1% Evans blue. 25. Dehydrate, clear, and mount. Reaction products obtained by the PAGS method (first staining) are black and those obtained by the immunoperoxidase method (second staining) are brown. For counterstaining, nuclear fast red therefore is not a good choice here, whereas Evans blue is well suited. Since IGSS systems can detect tissue antigens at higher dilutions of specific antibodies beyond the usual range

83

Gold and Silver Staining of sensitivity of the immunoperoxidase and horseradish peroxidase-antihorseradish peroxidase (PAP) methods,21 this double staining procedure may employ two types of specific antibodies without the intermediate removal of the first specific antibody, even if both antibodies are raised in the same animal species. The PAGS staining method is a useful antigen-detecting technique suitable for light and electron microscopic IHC. The method is a highly sensitive, reliable, economical, and straightforward approach with considerable versatility. Some of its advantages in detecting low concentrations of tissue antigen and double immunostaining are superior to the more commonly used enzyme, fluorescence, or radioisotope-labeled procedures. Increased appreciation and development of this technology would benefit significantly the fields of IHC, molecular morphology, and clinical–surgical pathology. REFERENCES 1.Danscher, G. 1981. Histochemical demonstration of heavy metals. A revised version of the sulphide silver method suitable for both light and electron microscopy. Histochemistry 71:1-16. 2. Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemistry 8:1081-1083. 3.Forsgren, A. and J. Sjöquist. 1968. “Protein A” from S. aureus. I. Pseudoimmune reaction with human gamma-globulin. J. Immunol. 97:822-827. 4.Fujimori, O. and M. Nakamura. 1985. Protein A gold–silver staining method for light microscopic immunohistochemistry. Arch. Histol. Jpn. 48:449-452. 5.Fujimori, O., S. Nakamura, S. Komori, and K. Yamada. 1987. A double staining of two tissue antigens by means of protein A gold–silver and immunoperoxidase methods. Nagoya Med. J. 32:121-126. 6.Fujimori, O. 1988. An efficient protein A gold–silver staining method for light microscopic immunohistochemistry. Z. Mikrosk. Anat. Forsch. 102:858-864. 7.Fujimori, O. 1992. A double protein A–gold–silver staining method for tissue antigens in light microscopy. Histochem. J. 24:61-66. 8.Fujimori, O. 1992. Methods of light microscopic immunostaining, p. 49-112. In S. Yokota and O. Fujimori (Eds.), Methods of Immunogold Staining. Soft Sci. Publ., Tokyo (in Japanese).

84

9.Fujimori, O., T. Ueda, and K. Yamada. 1994. Effects of iodine pretreatment of sections upon immunogold–silver staining in light microscopic immunohistochemistry. Okajimas Folia Anat. Jpn. 71:319-324. 10.Fujimori, O., T. Ueda, and K. Yamada. 1996. A double protein A–gold–silver staining method using one step development for light microscopy. Acta Histochem. Cytochem. 29:1-6. 11.Goding, J.W. 1978. Use of staphylococcal protein A as an immunological reagent. J. Immunol. Methods 20:241-253. 12.Hacker, G.W., L. Grimelius, G. Danscher, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc in tissues. J. Histotechnol. 11:213-221. 13.Hacker, G.W. 1989. Silver-enhanced colloidal gold for light microscopy, p. 297-321. In M.A. Hayat (Ed.), Colloidal Gold. Principles, Methods, and Applications, Vol.1. Academic Press, San Diego. 14.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird. 1983. Immunogold–silver staining. New method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31:938-944. 15.Lee, W.E. and E.R. Brown. 1977. The developing agents and their reactions, p. 291-334. In T.H. James (Ed.), Theory of the Photographic Process. MacMillan. New York. 16.Moeremans, M., G. Daneels, A. Van Dijck, G. Langanger, and D.J. Mey. 1984. Sensitive visualization of antigen-antibody reactions in dot and blot immune overlay assays with immunogold and immunogold–silver staining. J. Immunol. Methods 74:353-360. 17.Nakamura, M., H. Kitamura, and K. Yamada. 1985. A sensitive method for the histochemical demonstration of vicinal diols of carbohydrates. Histochem. J. 17:477-485. 18.Romano, E.L. and M. Romano. 1977. Staphylococcal protein A bound to colloidal gold: a useful reagent to label antigen-antibody sites in electron microscopy. Immunochemistry 14:711-715. 19.Roth, J. 1982. Application of immunocolloids in light microscopy. Preparation of protein A-silver and protein A–gold complexes and their application for localization of single and multiple antigens in paraffin sections. J. Histochem. Cytochem. 30:691-696. 20.Roth, J. 1983. The colloidal gold marker system for light microscopy. In G.W. Bullock and P. Petrusz (Eds.), Techniques in Immunocytochemistry, Vol. 2. Academic Press, London. 21.Scopsi, L. and L.-I. Larsson. 1985. Increased sensitivity in immunocytochemistry. Histochemistry 82:321329. 22.Skutelsky, E., V. Goyal, and J. Alroy. 1987. The use of avidin-gold complex for light microscopic localization of lectin receptors. Histochemistry 86:291-295. 23.Springall, D.R., G.W. Hacker, L. Grimelius, and J.M. Polak. 1984. The potential of the immunogold–silver staining method for paraffin sections. Histochemistry 81:603-608.

6

Microscopic Uses of Nanogold® James F. Hainfeld, Richard D. Powell, and Frederic R. Furuya

INTRODUCTION Gold has been used for immunocytochemistry since 1971 when Faulk and Taylor discovered adsorption of antibodies to colloidal gold.5 It is an ideal label for electron microscopy (EM) due to its high atomic number, which scatters electrons efficiently, and the fact that preparative methods have been developed to make uniform particles in the appropriate size range of 5 to 30 nm.16 Use in light microscopy (LM) generally requires silver enhancement (autometallography; AMG) of these small gold particles. Significant advances in this field since that time have included a better understanding of the conditions for best antibody adsorption, more regular gold size production, adsorption of other useful molecules, like protein A,28 and advances in silver enhancement.4,8,18 Many studies have also been accomplished showing the usefulness of these techniques to cell biology and biomedical research. A further advance in this field was the development of Nanogold, a 1.4 nm gold cluster.12 A significant difference from colloidal gold is that Nanogold is actually a coordination compound containing a gold 0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

core covalently linked to surface organic groups (Figures 6.1 and 6.2). These in turn may be covalently attached to antibodies (Figures 6.3 and 6.4). This approach to immunolabeling has several advantages compared to colloidal gold such as vastly better penetration into tissues, generally greater sensitivity, and higher density of labeling.31,32 Since Nanogold is covalently coupled to antibodies, it may also be directly coupled to almost any protein, peptide, carbohydrate, or molecule of interest, including molecules which do not adsorb to colloidal gold. This increases the range of probes possible, and expands the applications of gold labeling. Below are presented some established protocols for use of Nanogold in the LM and EM, and also how to use it as a general labeling reagent. More specifically, this chapter covers: • General comments. • Materials and reagents needed. • Nanogold labeling for LM. • Nanogold labeling of cells in suspension. • Nanogold labeling for transmission EM (TEM) pre-embedding. • Nanogold labeling for transmission EM postembedding.

85

Gold and Silver Staining • Double labeling for EM (Nanogold and 10 nm colloidal gold). • Using negative stains with Nanogold. • Nanogold labeling using monomaleimido–Nanogold. • Labeling Fab′ fragments with Nanogold. • Labeling IgG molecules with Nanogold. • Labeling other proteins with monomaleimido–Nanogold. • Nanogold labeling using mono-NHSNanogold. • Labeling proteins (MW 15 000 or greater) with mono-NHS-Nanogold (NHS = N-hydroxysuccinimide). • Labeling peptides (MW 6000 or less) with mono-NHS-Nanogold. • Labeling liposomes with dipalmitoylphosphatidylethanolamine (DPPE)Nanogold. • Forming heavily gold-labeled liposomes and micelles. • Forming liposomes “spiked” with Nanogold-DPPE. • Forming very large liposomes (>0.5 µm) spiked with Nanogold-DPPE. • Direct viewing of Nanogold in the EM.

Figure 6.1. Schematic diagram of Nanogold. Shown are the 1.4 nm gold core, the 0.6 nm organic shell, and the monofunctional linking group for covalent attachment to other molecules.

86

Since Nanogold is available commercially, additional useful information, applications, and references are accessible at the manufacturer’s web site at www. nanoprobes.com (Nanoprobes, Yaphank, NY, USA). PROTOCOLS General Comments about Using Nanogold Nanogold conjugates can be stored in 0.02 mol/L sodium phosphate buffer with 150 mmol/L sodium chloride, or other buffer solutions usually used with the protein under study. If they are to be stored longer than three days, add 0.1% bovine serum albumin (BSA) and 0.05% sodium azide to prevent bacterial contamination and to prevent the protein from adhering to the surfaces of the storage vessel. Thiol caution: Nanogold particles degrade upon exposure to concentrated

Figure 6.2. Darkfield scanning transmission electron micrograph of Nanogold clusters on a thin carbon film. Each bright dot is a 1.4 nm Nanogold particle. Full width 128 nm. (Reproduced with permission from Hainfeld, J.F., 1996. Labeling with Nanogold and undecagold: techniques and results, Scanning Microsc. Suppl. 10:309–322.)

Microscopic Uses of Nanogold® thiols such as β-mercaptoethanol (BME) or dithiothreitol (DTT). If such reagents must be used, concentrations should be kept below 1 mmol/L and exposure restricted to 10 min or less, but preferably they should be avoided. If aldehyde-containing reagents have been used for fixation, it is recommended that these be quenched before labeling. This may be achieved by incubating the specimens for 5 min in 50 mM glycine solution in phosphate-buffered saline (PBS) (pH 7.4). Ammonium chloride (50 mM) or sodium borohydride (0.5–1 mg/mL) in PBS may be used instead of glycine. Silver enhancement: LI Silver (Nanoprobes) is a procedure recommended by the authors for enhancing Nanogold for LM observation; it is light insensitive enough for this application. HQ Silver (Nanoprobes) is recommended for EM, but must be used under safelight or low light conditions. Alternative developers include those described by Danscher and Norgaard,4 Hacker et al.,8 Burry et al.,3and others available commercially. Most recently, light-insensitive gold saltbased AMG developers have been introduced (see Chapter 3).

Materials and Reagents • PBS Buffer: 20 mmol/L phosphate with 150 mmol/L NaCl, pH 7.4. • PBS-BSA Buffer: 20 mmol/L phosphate with 150 mmol/L NaCl, pH 7.4, 0.5% BSA, 0.1% gelatin (high purity). • Optional, may reduce background: 0.5 mol/L NaCl, 0.05% Tween 20. • Buffer 1: 20 mmol/L phosphate, 150 mmol/L NaCl, pH 7.4, 4% BSA, 2 mmol/L sodium azide (NaN3). • PBS++: PBS containing 1% BSA, 0.1% sodium azide, and 1% normal goat serum. • PBS*: PBS containing 1% non-fat dry milk and 1 mg/mL bovine IgG. • Nanogold goat anti-mouse Fab′ and other Nanogold-Fab′ or Nanogold-IgG secondary antibodies (Nanoprobes). • LI Silver: Silver developer for LM, or • HQ Silver: Silver developer for EM, or • Silver acetate AMG developer:8 see Chapter 4. • Monomaleimido–Nanogold (Nanoprobes). • Mono-NHS-Nanogold (Nanoprobes). • NanoVan (negative stain from Nanoprobes).

Figure 6.3. Diagram showing reaction of maleimido gold cluster with the hinge sulfhydryl of a Fab′′ antibody fragment.

87

Gold and Silver Staining • Centricon centrifuge filtration products (Amicon, subsidiary of Millipore, Bedford, MA, USA). • Amicon gel filtration media (Amicon). • Some Amersham Pharmacia Biotech products (Amersham Pharmacia Biotech, Piscataway, NJ, USA). • Some Bio-Rad products (Bio-Rad, Hercules, CA, USA). Protocol 1. Nanogold Labeling for LM Features labeled with Nanogold will be stained black in the LM upon AMG enhancement. Different development times should be tried to determine which is best for your experiment. The procedure for immunolabeling is similar to that for EM; one suitable procedure adapted for cytology is given below. Optimum protocols for immunohistopathology are given in Chapter 4. Samples must be rinsed with deionized or better redistilled water before silver enhancement. This is because the reagent contains silver ions in solution, which react to form a precipitate with chloride, phosphate, and other anions that are components of buffer solutions.

from the same species as the Nanogold reagent, for 1 h at room temperature. 6. Rinse with PBS (3 times for 5 min). 7. Postfix with 1% glutardialdehyde in PBS at room temperature (3 min). 8. Rinse with deionized water (3 times for 1 min). 9. Develop specimen with freshly mixed developer for 5 to 20 min, or as directed in the instructions for the silver reagent. More or less time can be used to control intensity of signal. A series of different development times may be used, to find the optimum enhancement for your experiment. Generally, a shorter antibody incubation time will require a longer silver development time. Development with light-insensitive silver enhancers may be monitored under the LM (see Chapters 2 and 3). 10. Rinse with deionized water (2 times for 5 min). 11. The specimen may now be counterstained if desired before examination, with usual reagents.

Staining Procedure 1. Spin cells onto slides using Cytospin (Shandon, Pittsburgh, PA, USA), or use paraffin section. 2. Incubate with 1% solution of PBS-BSA for 10 min to block nonspecific protein binding sites. 3. Incubate with primary antibody, diluted at usual working concentration in PBS-BSA (1 h or usual time). 4. Rinse with PBS-BSA (3 times for 2 min). 5. Incubate with Nanogold reagent (e.g., Nanogold-labeled secondary layer antibodies, or Fab′ fragments) diluted 1/40 to 1/200 in PBS-BSA with 1% normal serum 88

Figure 6.4. Darkfield scanning transmission electron micrograph of Nanogold–Fab′′ fragments. Bright dots are the Nanogold clusters (arrowhead), and grey masses (arrow) are the Fab′ fragments. Full width 128 nm.

Microscopic Uses of Nanogold® Protocol 2. Nanogold Labeling of Cells in Suspension Staining Procedure 1. Optional fixing of cells: e.g., with glutardialdehyde (0.05%–1% for 15 min) in PBS. Do not use Tris buffer for fixation since this contains an amine. After fixation, centrifuge cells (e.g., 1 mL at 10000000 cells/mL) at 300× g for 5 min, discard supernatant, and resuspend in 1 mL buffer. Repeat this washing (centrifugation and resuspension) 2 times. 2. Incubate cells with 0.02 mol/L glycine in PBS (5 min). Centrifuge, then resuspend cells in PBS-BSA buffer for 5 min. 3. Place 50 to 200 µL of cells into Eppendorf tube and add 5 to 10 µL of primary antibody (or antiserum). Incubate 30 min with occasional shaking (do not create bubbles which will denature proteins). 4. Wash cells using PBS-BSA as described in step 1 (2 times for 5 min). Resuspend in 1 mL Buffer 1. 5. Dilute Nanogold approximately 50 times in PBS-BSA buffer and add 30 µL to cells. Incubate for 30 min with occasional shaking. 6. Wash cells in PBS-BSA as described in step 1 (2 times for 5 min). 7. Fix cells and antibodies using a final concentration of 1% glutardialdehyde in PBS for 15 min. Then remove fixative by washing with Buffer 1 (3 times for 5 min). Protocol 3. Nanogold Labeling for TEM Pre-embedding Staining Procedure 1. Float on a drop of water for 5 to 10 min. 2. Incubate cells with 1% PBS-BSA buffer at pH 7.4 for 5 min. This blocks

any nonspecific protein binding sites and minimizes nonspecific antibody binding. 3. Incubate with primary antibody, diluted at usual working concentration in PBSBSA (30 min to 1 h or usual time). 4. Rinse with PBS-BSA (3 times for 1 min). 5. Incubate with Nanogold reagent diluted 1/40 to 1/200 in PBS-BSA with 1% normal serum from the same species as the Nanogold reagent, for 10 min to 1 h at room temperature. 6. Rinse with PBS-BSA (3 times for 1 min), then PBS (3 times for 1 min). 7. Postfix with 1% glutaraldehyde in PBS (10 min). 8. Rinse in deionized water (2 times for 5 min). 9. Silver enhance (see appropriate chapter). 10. Dehydrate and embed according to usual procedure. 11. Stain (uranyl acetate, lead citrate, or other positive staining reagent) as usual before examination. Protocol 4. Nanogold Labeling for TEM Postembedding Staining Procedure 1 1. Prepare sections on plastic or carboncoated nickel grid. Float on a drop of water for 5 to 10 min. 2. Incubate with 1% solution of PBSBSA buffer at pH 7.4 for 5 min to block nonspecific protein binding sites. 3. Incubate with primary antibody, diluted at usual working concentration in PBS-BSA (1 h or usual time). 4. Rinse with PBS-BSA (3 times for 1 min). 5. Incubate with Nanogold reagent diluted 1/40 to 1/200 in PBS-BSA with 1% 89

Gold and Silver Staining normal serum from the same species as the Nanogold reagent, for 10 min to 1 h at room temperature. 6. Rinse with PBS (3 times for 1 min). 7. Postfix with 1% glutardialdehyde in PBS at room temperature (3 min). 8. Rinse in deionized water for (2 times for 5 min). 9. Silver enhance (see appropriate chapter in this volume). 10. If desired, contrast sections with uranyl acetate and/or lead citrate before examination. Note: Thin sections mounted on grids are floated on drops of solutions on parafilm or in well plates. Hydrophobic resins usually require pre-etching. AMG silver enhancement renders the small Nanogold particles more easily visible. This is usually recommended and especially if stains such as uranyl acetate or lead citrate are applied. Silver enhancement should be completed before these stains are applied. Staining Procedure 2 This more specific procedure for postembedding labeling is taken from Krenács and Dux:20 1. Fix tissue in 4% formaldehyde with 0.1% glutaraldehyde. 2. Embed from 96% ethanol in LR White (Electron Microscopy Sciences, Fort Washington, PA, USA) according to manufacturer’s instructions. 3. Section. 4. Incubate section with primary antibody (diluted in PBS++ overnight at 4°C) and wash with buffer. 5. Incubate with Nanogold-Fab′ for 40 min (diluted 1:30 in PBS++) and wash. 6. Postfix in 1% glutardialdehyde PBS for 5 min. 90

7. Wash 6 times for 1 min each with bidistilled water. 8. Silver enhance on drops of developer for 2 to 4 min. 9. Wash and counterstain with uranyl acetate and lead citrate for 5 min each. Protocol 5. Double Labeling for EM: Nanogold and 10 nm Colloidal Gold (32) Staining Procedure 1. Incubate cryosections (or standard sections) overnight in PBS with 1% nonfat dry milk. 2. Incubate sections with primary antibody (usually diluted approximately 1:50) in PBS* containing 0.02% sodium azide for 3 h at 22°C. 3. Wash in 5 changes of PBS. 4. Incubate with secondary NanogoldFab′ (diluted 1:50 or 1:100 in PBS*) for 2 h at 22°C. 5. Wash 5 times with PBS. 6. Wash 5 times in 50 mmol/L MES [1(N-morpholino)-ethanesulfonate] buffer, pH 6.1. 7. Silver enhance sections and wash with water, then PBS. 8. Incubate with primary to second antigen to be localized (diluted in PBS* containing 0.02% sodium azide) for 3 h at 22°C. 9. Wash in 5 changes of PBS. 10. Incubate with 10 nm colloidal gold secondary antibody diluted in PBS* for 2 h at 22°C. 11. Wash and poststain as usual. Protocol 6. Negative Stains With Nanogold: EM Because the 1.4 nm Nanogold particles are so small, overstaining with OsO4,

Microscopic Uses of Nanogold® uranyl acetate, or lead citrate may tend to obscure direct visualization of individual (unenhanced) Nanogold particles. Three recommendations for improved visibility of Nanogold are: (1) use of reduced amounts or concentrations of usual stains; (2) use of lower atomic number stains such as NanoVan, a vanadium based stain; or (3) enhancement of Nanogold with AMG silver developers prior to staining. Staining Procedure 1. Wash carbon film on EM grid several times with buffer, wicking with filter paper; final time leave 5 µL. 2. Inject 5 µL of sample (containing, e.g., isolated molecules in solution labeled with Nanogold) into drop on grid. 3. Wait 1 min for attachment of protein molecules to carbon film on grid. 4. Wash 6 times total with buffer, then water (if compatible), wicking off excess. 5. Wash 3 times with NanoVan. 6. After 1 min, wick to thin layer of fluid and allow to air dry. Protocol 7. Nanogold Labeling Using Monomaleimido–Nanogold: Labeling Fab′′ Fragments with Nanogold If F(ab′)2 fragments are available, they should be used for this purpose. If these are not available, they should be prepared from IgG molecules by digestion with ficin or pepsin,23,26 It should be noted that IgG molecules from different host animals vary slightly in structure, and therefore will differ in the ease with which they are digested in this way. The procedure given below has been found to be effective in the preparation of F(ab′)2 fragments from IgG developed in goat. Pepsin becomes more active as the pH is lowered; at pH 7 it is inactivat-

ed. Some monoclonal antibody subclasses may be unsuitable for this process (for example, IgG3). A comprehensive review has been published which describes the use of pepsin digestion to prepare F(ab′)2 fragments from mouse monoclonal IgG molecules and details which classes may be digested in this manner.26 Note: Procedures given for labeling with Monomaleimido–Nanogold refer to labeling with 30 nmol of the Nanogold labeling reagent. If you are using the smaller 6 nmol size, quantities should be reduced by a factor of five. Labeling Procedure 1. Dissolve IgG in 0.1 mol/L sodium citrate buffer at pH 4.5; add a solution of pepsin in 0.5 mL of the same buffer. Use an amount of pepsin equal to 2% of the mass of IgG. Incubate at 37°C for 20 h using a water bath or incubator. 2. Isolate the F(ab′)2 fragments by highperformance liquid chromatography (HPLC), using a column such as Amersham Pharmacia Biotech Superose 6 or 12, TSK 3000, or Toyosoda MacMod GF-250 (which have wide molecular weight fractionation ranges), or a superfine exclusion gel such as Amicon GCL300 (which excludes compounds with molecular weights above 60 000). Dialysis does not provide acceptable purification in this application. Elute with 0.02 mol/L sodium phosphate at pH 7.4 with 150 mmol/L sodium chloride and 1 mmol/L EDTA. The F(ab′)2 fragments will elute in the void volume if a GCL-300 column is used as the first band. Combine the fractions containing F(ab′)2 fragments and calculate the amount using the optical density. Concentrate to 0.5 mL or less using membrane centrifugation. 91

Gold and Silver Staining 3. Dilute the antibody in 0.1 mol/L sodium phosphate buffer, pH 6.0, containing 5 mmol/L EDTA (make up to 1 mL), and dissolve mercaptoethylamine hydrochloride (MEA: 6 mg) in this solution. Incubate at room temperature for 1 h. This step reduces the hinge region cysteine bond(s), separating the F(ab′)2 into two Fab′ molecules. 4. Isolate Fab′ fragments by gel filtration chromatography. Use a desalting gel, such as Amicon GH-25, which has an exclusion cutoff at molecular weight 3000. As the eluent, use 0.02 mol/L sodium phosphate at pH 6.5, with 150 mmol/L sodium chloride and 1 mmol/L EDTA. The reduced antibody will be eluted in the void volume as the first sharp peak in the trace. Combine the fractions containing reduced antibody; the total amount of antibody should be calculated from the optical density (usually for Fab′, E1% at 280 nm = 15.3; concentration in mg/mL = OD280nm × 10/ E1% = OD280nm × 0.65). 5. Dissolve the Nanogold reagent in 0.1 mL isopropanol or DMSO (dimethyl sulfoxide), then dilute to 1 mL with deionized water. Sufficient reagent is supplied to label 0.2 mg of Fab′. If you are using a smaller amount, use a proportionately smaller amount of the Nanogold reagent. Once activated, Nanogold is reconstituted with water, and it should be used immediately. The maleimide group is hydrolyzed in aqueous solution. 6. Add the activated Nanogold solution to the Fab′ fragments, and either: a. Incubate at room temperature for 1 h; or b. Incubate for 12 to 18 h at 4°C. 7. Separate the unbound gold particles from the antibody conjugates using gel exclusion chromatography. The Nano92

gold conjugate may be effectively isolated using a medium such as Superose 6 or 12 (which fractionate a wide range of molecular weights) or Amicon GCL-90 (which excludes molecules of mass 30 000 or greater). Concentrate the reaction mixture to a suitably small volume using membrane centrifugation (e.g., Centricon-30 system). Elute with 0.02 mol/L sodium phosphate at pH 7.4 with 150 mmol/L sodium chloride. The first, faintly colored peak is the conjugate, while the second, darker band is unbound Nanogold particles. For even higher purity, repeat the this process one time. The extent of labeling may be determined from the UV visible spectrum of the conjugate.11 Nanogold has an extinction coefficient at 280 nm of 2.25 × 105 and at 420 nm of 1.12 × 105. This means that for Fab′ labeled 100% (one Nanogold particle to one Fab′ molecule), the ratio of optical densities at 280 and 420 nm is close to 2.7 (extinction coefficient at 280 nm for Fab′ is 7.5 × 104). Nanogold conjugates should be stored in 0.02 mol/L sodium phosphate buffer with 150 mmol/L sodium chloride. If they are to be stored longer than three days, add 0.1% BSA and 0.05% sodium azide to prevent bacterial contamination and to prevent the protein from adhering to the surfaces of the storage vessel. Protocol 8. Labeling IgG Molecules with Monomaleimido–Nanogold IgG molecules contain disulfide bonds which connect the chains in the hinge region. These are reduced with a mild reducing agent, such as MEA, and then reacted with the Nanogold reagent in buffer solution, either for 1 h at room temperature or overnight at 4°C. The coupling reaction should be performed at pH 6.5, since at pH

Microscopic Uses of Nanogold® values greater than 7.5 the maleimido group becomes slightly reactive towards primary amines as well as sulfhydryls and may give nonspecific labeling. The reduced antibody must be isolated from the MEA before reaction. This may be achieved by gel filtration, using a gel such as GH-25. The Nanogoldconjugated product may be isolated by gel filtration, using a fine gel such as Superose 6 or 12, Superdex 75 (Amersham Pharmacia Biotech), or GCL-90. Labeling Procedure 1. Dissolve the antibody (0.6 mg IgG) in 0.1 mol/L sodium phosphate buffer, pH 6.0, containing 5 mmol/L EDTA (1 mL), and dissolve MEA (8 mg) in this solution. Incubate at room temperature for 1 h. 2. Isolate reduced antibody by gel filtration chromatography. Use a gel such as GH-25. Dialysis does not provide acceptable purification in this application. As the eluent, use 0.02 mol/L sodium phosphate at pH 6.5, with 150 mmol/L sodium chloride and 1 mmol/L EDTA. The reduced antibody will be eluted in the void volume as the first sharp peak in the trace. Combine the fractions containing reduced antibody. The total amount of antibody should be calculated from the optical density (usually for IgG, E1% at 280 nm = 14.5; concentration in mg/mL = OD280nm × 10/E1% = OD280nm × 0.69). 3. Dissolve the Nanogold reagent in 0.1 mL isopropanol or DMSO, then dilute to 1 mL with deionized water. Sufficient reagent is supplied to label 0.6 mg of IgG. If you are using a smaller amount, use a proportionately smaller amount of the Nanogold reagent. Once activated Nanogold is reconstituted with water, it should be used immediately. The maleimide group is hydrolyzed in aqueous solution.

4. Add the activated Nanogold solution to the reduced antibody. Incubate for 12 to 18 h at 4°C. 5. Separate the unbound gold particles from the antibody conjugates using gel exclusion chromatography. The Nanogold conjugate may be effectively isolated using a medium such as Superose 6 or 12 or GCL-90. Concentrate the reaction mixture to a suitably small volume using membrane centrifugation (e.g., Centricon-30 system). Elute with 0.02 mol/L sodium phosphate at pH 7.4 with 150 mmol/L sodium chloride. The first, faintly colored peak is the conjugate, while the second, darker band is unbound Nanogold particles. For even higher purity, repeat this process one time. The extent of labeling may be determined from the UV visible spectrum of the conjugate. Nanogold has an extinction coefficient at 280 nm of 2.25 × 105 and at 420 nm of 1.12 × 105. This means that for IgG labeled 100% (one Nanogold particle to one IgG molecule), the ratio of optical densities at 280 and 420 nm is close to 4.0 (extinction coefficient at 280 nm for IgG is 2.2 × 105). Protocol 9. Labeling other Proteins with Monomaleimido–Nanogold Monomaleimido–Nanogold may be used to label any protein with an accessible sulfhydryl group, such as a cysteine residue, in the same manner as described above for antibodies. In some proteins, the sulfhydryl functionality is in the form of a disulfide group. This must be reduced with a mild reducing agent, such as MEA or DTT, before it can be labeled. If you are unsure of the structure of your protein and have sufficient quantity available, it is recommended that the suitability of the sulfhydryl for labeling be 93

Gold and Silver Staining determined first; some sulfhydryl sites may be buried within the protein structure, and therefore inaccessible to the Nanogold reagent. The suitability of a particular protein for Nanogold labeling may be determined using 14C iodoacetic acid before gold labeling is tried. Alternatively, sensitive colorimetric procedures exists for sulfhydryl determination, including use of Ellman’s reagent [5,5′-dithio-bis-(2-nitrobenzoic acid) (DTNB)], or maleimido–fluorescein. Sufficient Nanogold reagent is supplied to label 4 nmol of sulfhydryl groups (e.g., 0.4 mg of a 100 000 molecular weight compound with one sulfhydryl). Once activated Nanogold is reconstituted with water, it should be used immediately. The maleimide group is hydrolyzed in aqueous solution. The reaction should be performed in the same manner as for antibodies and antibody fragments. For purification and isolation steps, alternative buffers may be substituted for those given. However, the labeling reaction itself should be performed using the buffers and conditions specified. Labeling Procedure 1. If the labeling site is in the form of a disulfide group, it may be reduced to the free sulfhydryl with MEA or DTT. Dissolve the protein (0.6 mg IgG) in 0.1 mol/L sodium phosphate buffer, pH 6.0, containing 5 mmol/L EDTA (1 mL), and dissolve MEA (8 mg; 70 mmol/L final) in this solution. Incubate at room temperature for 1 h. Lower amounts of reducing agent (e.g., 10 mmol/L final) may be more gentle and adequate for specific proteins. 2. Isolate the reduced protein by gel filtration chromatography. Use a gel, such as GH-25. Dialysis does not provide acceptable purification in this application. As the eluent, use 0.02 mol/L sodium phosphate at pH 6.5, with 150 mmol/L sodium chloride and 1 mmol/L 94

EDTA. The reduced protein will be eluted in the void volume as the first sharp peak in the trace. Combine the fractions containing reduced protein, and the total amount of protein should be calculated from the optical density. 3. Dissolve the Nanogold reagent in 0.1 mL isopropanol or DMSO, then dilute to 1 mL with deionized water. Smaller volumes are not recommended since the reagent contains buffers. Sufficient reagent is supplied to label 0.2 mg of Fab′. If you are using a smaller amount, use a proportionately smaller amount of the Nanogold reagent. Once activated Nanogold is reconstituted with water, it should be used immediately. The maleimide group is hydrolyzed in aqueous solution. 4. Add the activated Nanogold solution to the reduced protein. Incubate for 12 to 18 h at 4°C. 5. Separate the unbound gold particles from the protein conjugates using gel exclusion chromatography. The Nanogold conjugate may be effectively isolated using a medium such as Superose 6 or 12 or GCL-90. Concentrate the reaction mixture to a suitably small volume using membrane centrifugation (e.g., Centricon-30 system). Elute with 0.02 mol/L sodium phosphate at pH 7.4 with 150 mmol/L sodium chloride. The first, faintly colored peak is the conjugate, while the second, darker band is unbound Nanogold particles. For even higher purity, repeat this process one time. Caution: Nanogold particles degrade upon exposure to thiols such as BME or DTT. Protocol 10. Labeling Proteins (MW 15 000 or Greater) with Mono-NHSNanogold Mono-NHS-Nanogold (here the sulfo

Microscopic Uses of Nanogold® form is used for greater water solubility) is a reagent that is specific for labeling primary amines, including lysines and the alphaamino terminus of proteins. For most proteins, no pretreatment is necessary, since amines will be present. The procedure described below is suggested for labeling larger proteins (MW 15,000 or greater); these are significantly larger than the Nanogold particle (molecular weight approximately 15,000) and may be separated from excess unbound gold by gel exclusion chromatography. The protein is reacted with the Nanogold in buffer solution at pH 7.5 to 8.0, either for 1 h at room temperature or overnight at 4°C. The pH should be near 7.5 after mixing. The Nanogold-conjugated product may be isolated by gel filtration, using a fine gel such as Superose 6 or 12, Superdex 75, or GCL-90. The recommended procedure is given below: Note: The procedures given for the use of mono-sulfo-NHS-Nanogold refers to the larger 30 nmol size. If you are using the smaller 6 nmol size, quantities should be reduced by a factor of five. Labeling Procedure 1. Dissolve the protein in 0.02 mol/L 4,(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer, adjusted to pH 7.5 with sodium hydroxide. Do not use Tris, as it contains reactive amines that will compete with the protein amines for Nanogold labeling. 2. Dissolve the mono-sulfo-NHS-Nanogold reagent in 1 mL deionized water. Sufficient reagent is supplied to label 6 nmol of amine sites; if you are using a smaller amount, use a proportionately smaller amount of the Nanogold reagent. Once activated Nanogold is reconstituted with water, it should be used immediately. The succinimide

ester is hydrolyzed in aqueous solution. 3. Add the activated Nanogold solution to the dissolved protein. Incubate for either 1 h at room temperature or 12 to 18 h at 4°C. 4. Separate the unbound gold particles from the antibody conjugates using gel exclusion chromatography. Filter the mixture before concentration or before injection on the column to remove suspended solids. The Nanogold conjugate may be effectively isolated by HPLC using a gel such as Superose 6 or 12 or GCL-90. Concentrate the reaction mixture to a suitably small volume using membrane centrifugation (e.g., Centricon-30 system). Elute with 0.02 mol/L sodium phosphate at pH 7.4 with 150 mmol/L sodium chloride. The first, pale yellow peak or shoulder is the conjugate, while the second, darker band is unbound Nanogold particles. For even higher purity, repeat this process one more time. The extent of labeling may be calculated from the UV visible spectrum of the conjugate. Mono-NHS-Nanogold has extinction coefficients at 280 nm of 2.3 × 105, and at 420 nm of 1.1 × 105. Protocol 11. Labeling Peptides (MW 6000 or Less) with Mono-NHS-Nanogold The procedure described below is suggested for labeling smaller peptides (MW 6000 or less); these are significantly smaller than the Nanogold particle (molecular weight approximately 15,000). In this case, separation of excess unreacted peptide from gold–peptide conjugates is usually easier than separating conjugates from unbound gold, and therefore, it is advisable to use an excess of the peptide to be labeled. The protein is reacted with the Nanogold in buffer solution at pH 7.5 to

95

Gold and Silver Staining 8.0, either for 1 h at room temperature or overnight at 4°C. The pH should be near 8.0 after mixing. The Nanogold conjugated product may be isolated by gel filtration, using gel which can separate compounds of MW 15,000 and below, such as Superose 12, Superdex 75 or Superdex Peptide, and Bio-Gel P10 or P6 (Bio-Rad). Conjugates of very small peptides (MW 2000 or less) may be separated using a desalting gel such as GH25. Separation may also be achieved by membrane filtration using a microconcentrator such as Centricon-10 (which retains compounds with MW 10,000 and over). Repeated concentrations are necessary to remove all the unbound peptide. Labeling Procedure 1. Dissolve the peptide in 0.02 mol/L HEPES buffer, adjusted to pH 7.5 with dilute sodium hydroxide (1 mL). 2. Dissolve the mono-sulfo-NHS-Nanogold reagent in 1 mL deionized water. 30 nmol of reagent is supplied. Use a 5to 20-fold excess of the peptide to be labeled. Once activated Nanogold is reconstituted with water, it should be used immediately. The succinimide ester is hydrolyzed in aqueous solution. 3. Add the activated Nanogold solution to the dissolved protein. Incubate for either 1 h at room temperature or 12 to 18 h at 4°C. 4. Separate the gold–peptide conjugates from excess unbound peptide using gel exclusion chromatography. Filter the mixture (e.g., through 0.2-µm cellulose acetate spin filter) before concentration or before injection on the column to remove suspended solids. The Nanogold conjugate may be effectively isolated by HPLC using a gel such as Superdex 75, Superdex Peptide, Bio-Gel P10 or P6 96

(MW of Nanogold is near 15,000). Concentrate the reaction mixture to a suitably small volume using membrane centrifugation (e.g., Centricon-10 system). Elute with 0.02 mol/L sodium phosphate at pH 7.4 with 150 mmol/L sodium chloride. The first, pale yellowbrown peak or shoulder is the conjugate, while the second, colorless band is excess peptide. For even higher purity, repeat this process one more time. If the extinction coefficient of the peptide is known, the labeling may be calculated. Sulfo-succinimido-Nanogold has extinction coefficients at 280 nm of 2.3 × 105 and at 420 nm of 1.1 × 105. However, smaller peptides often have much lower extinction coefficients, which may lead to inaccuracies in calculations. Note: Proteins with MW close to that of the Nanogold particle (15,000) may be labeled using NHS-Nanogold, but the products cannot be separated by size exclusion chromatography. Other chromatographic techniques, such as reverse-phase, hydrophobic interaction, or ion exchange chromatography may be used. Nanogold is more hydrophobic than most proteins and elutes differently. A Nanogold–peptide conjugate has intermediate characteristics. PROTOCOLS FOR LABELING LIPOSOMES WITH DPPENANOGOLD DPPE-Nanogold consists of the 1.4 nm Nanogold particle covalently linked to a single molecule of dipalmitoyl phosphatidyl ethanolamine. DPPE-Nanogold is hydrophobic and can insert into organic phases in systems such as micelles and liposomes. It is soluble in methanol and in methanol-trichlroromethane and methanol-dichloromethane mixtures. Extinc-

Microscopic Uses of Nanogold® tion coefficients at specific wavelengths are given below for methanol solution: Extinction Wavelength (nm): Coefficienta 280 2.25 × 105 420 1.12 × 105 for 5 × 10-6 mol/L solution in methanol. As is common with many other Nanogold procedures, visualization of the gold is greatly enhanced in the EM by using AMG silver or gold enhancement. LM applications generally require such enhancement. aMeasured

Protocol 12. Forming Heavily GoldLabeled Liposomes and Micelles Labeling Procedure 1. Dissolve Nanogold-DPPE in methanol, transfer (e.g., 0.5 nmol) to 1.5-mL polyethylene Eppendorf tube. 2. Evaporate solvent. 3. Add 0.5 mL water. 4. Sonicate with a microtip unit for 10 to 15 sec (e.g., using Dismembranator 550 [Fisher Scientific, Pittsburgh, PA, USA] with microtip; power level set at 3). Ranges of 5 sec to 10 min may also be used, but for longer times, a duty cycle should be used to prevent heating (e.g., 20 sec on, 40 sec off ). Protocol 13. Forming Liposomes Spiked with Nanogold-DPPE Labeling Procedure 1. Dissolve Nanogold-DPPE in methanol. 2. Aliquot an amount of Nanogold-DPPE into a solution of unlabeled lipids normally used for vesicle formation. Incorporation of 0.1% to 1% Nanogold-

DPPE with an unlabeled lipid is usually appropriate for preparing Nanogoldlabeled liposomes with the same properties and morphology as those prepared without the gold label. Frequently a 3:1 chloroform:methanol solvent is used for unlabeled lipids, but a 1:1 chloroform:methanol solvent mixture is recommended for better solubility of the Nanogold-DPPE. 3. Add water and sonicate mixture for 15 min or use other standard methods of liposome preparation. A hazy suspension formed at first indicates multilamellar vesicles (MLV); upon further sonication, the suspension becomes transparent, indicating conversion to small unilamellar vesicles (SUVs). Sonication is usually done in a beaker ice bath, or by duty cycle sonication, to avoid heating and degradation of normal lipids. Protocol 14. Forming Very Large Liposomes (>0.5 µM) Spiked with Nanogold-DPPE Labeling Procedure 1. Dissolve Nanogold-DPPE in methanol. 2. Mix 1 nmol Nanogold-DPPE with 10 nmol lecithin in 0.5 mL of a 1:4 solution of methanol:diethyl ether. 3. Inject solution into 0.1 mL of 75°C water using a 25-gauge needle. Protocol 15. Direct Viewing of Nanogold in the EM For most work, autometallographic silver enhancement is recommended to give a good signal in the EM (see companion chapters in this volume). For particular applications, visualization of the Nanogold directly may be desirable. Generally this 97

Gold and Silver Staining requires very thin samples and precludes the use of other high Z stains. Nanogold provides a much improved resolution and smaller probe size over most other gold antibody products. However, because Nanogold is only 1.4 nm in diameter, it will not only be smaller, but will appear less intense than, for example, a 5 nm colloidal gold particle. With careful work, however, Nanogold may be seen directly through the binoculars of a standard EM even in 80-nm thin sections. However, achieving the high resolution necessary for this work may require new demands on your equipment and technique. Several suggestions follow: Procedure 1. Before you start a project with Nanogold, it is helpful to see it so you know what to look for. Dilute the Nanogold stock 1:5 and apply 4 µL to a grid for 1 min. Wick the drop and wash with deionized water 4 times. 2. View Nanogold at 100,000× magnification with 10× binoculars for a final magnification of 1,000,000×. Turn the emission up full and adjust the condenser for maximum illumination. Nanogold is not beam sensitive, so will withstand this dose. 3. The alignment of the microscope should be in order to give 0.3 nm resolution. Although the scope should be well aligned, you may be able to skip this step if you do step 4. 4. Objective stigmators must be optimally set at 100,000×. Even if the rest of the microscope optics are not perfectly aligned, adjustment of the objective stigmators may compensate and give the required resolution. You may want to follow your local protocol for this alignment, but since it is important, a brief protocol is given here: 98

5. At 100,000× (1 × 106 with binoculars), over focus, under focus, then set the objective lens to in focus. This is where there is the least amount of detail seen. 6. Adjust each objective stigmator to give the least amount of detail in the image. 7. Repeat steps 5 and 6 until the in-focus image contains virtually no contrast, no wormy details, and gives a flat featureless image. 8. Now underfocus slightly, move to a fresh area, and you should see small black dots of 1.4 nm size. This is the Nanogold. For the 1:5 dilution suggested, there should be about 5 to 10 gold spots on the small viewing screen used with the binoculars. Contrast and visibility of the gold clusters is best at 0.2 to 0.5 µm defocus and is much worse at typical defocus values of 1.5 to 2.0 µm commonly used for protein molecular imaging. 9. In order to operate at high magnification with high beam current, thin carbon film over fenestrated holey film is recommended. Alternatively, thin carbon or 0.2% Formvar over a 1000 mesh grid is acceptable. Many plastic supports are unstable under these conditions of high magnification–high beam current, and carbon is therefore preferred. Contrast is best using thinner films and thinner sections. 10. Once you have seen Nanogold, you may now be able to reduce the beam current and obtain better images on film. For direct viewing with the binoculars, reduction in magnification from 1,000,000× to 50,000× makes the Nanogold much more difficult to observe and not all of the golds are discernable. At 30,000× (300,000× with 10× binoculars) Nanogold particles are not visible. It is recommended to view at 1,000,000×, with maximum beam current, align the objective stigmators, and then move to a fresh area, reduce

Microscopic Uses of Nanogold® the beam, and record on film. 11. If the demands of high resolution are too taxing, or your sample has an interfering stain, a very good result may be obtained using silver enhancement to give particles easily seen at lower magnification. FURTHER TECHNICAL HINTS AND DISCUSSION Nanogold has two very distinguishing features from typical colloidal gold. It is small and is covalently linkable to other molecules. These chemical differences lead to a number of important advantages over colloidal gold technology: (1) stability of the covalently bound antibodies is excellent compared to adsorbed antibodies;19 (2) penetration is better, reported up to 40 µm in tissues,31 since the gold is smaller and also a Fab′ antibody fragment may be used, which is one-third the size of an IgG (Figure 6.5); (3) higher density

of antigen labeling is achieved using smaller gold particles;32 (4) many other molecules that do not adsorb to colloidal gold may be easily attached to Nanogold, such as peptides, DNA, lipids, etc.; (5) available as a reagent to chemically link to and label a specific site (e.g., a cysteine or lysine residue); (6) uniform size;12 and (7) may be run on gels to follow labeling.7 Some potential disadvantages are: (1) its small size generally requires AMG silver enhancement; (2) the sizes after silver enhancement are generally more irregular than a colloidal gold prep; and (3) it is more difficult to synthesize Nanogold and couple it to proteins. The disadvantages are usually outweighed by the significant improvement in immunolabeling sensitivity and density generally observed, unique applications now made possible, and the availability of off-theshelf and easy to use Nanogold reagents (Figure 6.6). Nanogold conjugates are currently commercially available and include secondary

Figure 6.5. Size comparison diagram of colloidal gold IgG probe versus Nanogold-Fab′′. The significantly smaller Nanogold probe results in much greater penetration into tissues and gels.

99

Gold and Silver Staining antibodies (antimouse, rabbit, goat, etc.), anti-biotin, and Nanogold–streptavidin. Nanogold conjugates are stable to a wide range of buffers and at pH values from 3.0 to 10.0, and therefore, they may be used in a similar manner to the usual protocols for other labels. However, specific experience with Nanogold has led to certain protocols that work well, as described above. Nanogold immunolabeling has now been successfully applied in numerous immunocytochemistry (ICC) studies.6,20,24,32,34 Frequently, the labeling is much denser than with any colloidal gold probes (Figure 6.7), and in some cases, Nanogold labeling is obtained in experiments where labeling with colloidal gold probes produces no useful signal, especially for sparse or poorly accessible antigens.34 It has been pointed out by many investigators that the smaller the gold, the higher the density of labeling.32 It has also been shown, however, that “ultrasmall” colloidal golds in the 0.8 to 3 nm range aggregate

Figure 6.6. Electron micrograph after silver enhancement of Nanogold-Fab′′ labeling, resulting in convenient, approximately 10 nm, particles. Specimen is from the thoracic spinal cord of the rat and shows GABA-containing terminals (asterisks) forming symmetrical synaptic specializations (arrows) with dendrites (D), with excellent preservation of morphology. Postembedding staining was done using GABA antiserum (Incstar 1:2000, 4°C, 18 h [Incstar, Stillwater, MN, USA]), Nanogold goat antirabbit Fab′ (1:40 room temperature, 90 min), and intensified with HQ Silver (Nanoprobes) for 6 min. Tissue was counterstained with lead citrate, and embedding resin was Durcupan (Fluka). (Micrograph provided by S. Bacon, Oxford University, Dept. of Pharmacology, Oxford, University, Dept. of Pharmacology, Oxford, U.K. and reproduced with permission from Hainfeld, J.E. and F.R. Furuya, 1995. Immunogold–Silver Staining: Principles, Methods, and Applications, CRC Press, Boca Raton, pp. 71–96.)

Figure 6.7. Light micrograph comparing ultrasmall colloidal gold labeling (left) with Nanogold-Fab′′ labeling (right). Spindle microtubules were first labeled with a mouse antitubulin primary followed either with an ultrasmall colloidal gold antimouse probe (left) or Nanogold antimouse Fab′ (right). Each sample was treated identically with silver enhancer. (Original magnification, ×1300.) (Micrographs reproduced with permission from Vandre, D.P. and R.W. Burry, 1992. J. Histochem. Cytochem., 40:1837–1847.)

100

Microscopic Uses of Nanogold® and do not appear to perform well compared to the 1.4 nm Nanogold9,34 (Figure 6.7). Immuno-Nanogold labeling has been successfully used both in pre- and postembedding procedures (Figure 6.8). Although

AMG silver enhancement leads to some particle size heterogeneity, it can actually be quite regular (Figure 6.9) and is very useable. The higher density of labeling easily outweighs the slight variation in

Figure 6.8. Pre-embedding (A) and postembedding (B) with Nanogold. Immunoreactivity for the a1 (A) and b2/3 (B) subunits in the molecular layer of cat cerebellum is demonstrated by either pre-embedding (A, epoxy resin) or postembedding techniques (B, freeze-substituted Lowicryl-embedded). (A) Particles are not present in synapses (double triangles) including the type 2 synapse (e.g., open arrow) between a presumed stellate cell terminal (b) and an interneuron dendrite (d). Immunoparticles are often located at the edge of type 2 synapses (double arrowheads) as well as at the extrasynaptic membrane (arrowheads). (B) The postembedding reaction produces a row of silver-intensified gold particles (arrows) on the postsynaptic plasma membrane at a synaptic junction established by a bouton (b) with a purkinje cell dendrite (d). Some immunoparticles are also present at the extrasynaptic plasma membrane (arrowhead). Bar = 0.2 µm. (Reproduced with permission from Nusser, Z. et al., 1995. J. Neurosci., 15:2948–2960.)

101

Gold and Silver Staining enhanced particle size in most applications, compared with using colloidal gold. Nanogold with AMG silver enhancement may be followed by standard immunolabeling of a different antigen with colloidal gold for double labeling. This was achieved by Takizawa and Robinson.32 They showed that the labels were readily distinguished and that the silver enhancement was gentle enough to preserve antigenicity when the next immunolabel (a 10 nm colloidal gold) was applied. This is especially useful when one antigen is sparse, since Nanogold usually gives a much stronger signal than colloidal gold immunoconjugates.27,34 Nanogold is also available as a reagent that can react selectively with thiols or

amines (Figure 6.10).10 An investigator using these reagents and protocols can prepare primary antibodies or Fab′ antibody fragments labeled with Nanogold or covalently couple Nanogold to other molecules. One striking example was the labeling of substance P (SP), an 11 amino acid peptide, with mono-NHS-Nanogold.30 Nanogoldlabeled SP was found to behave identically to an 125I-labeled SP on gels and in tissues. For example, binding of this peptide could be competed out with unlabeled SP or with specific inhibitors. Therefore, the radioactive labeling used in previous studies could be replaced with nontoxic gold. Furthermore, for structural studies, localizations of SP in spinal column sections could be viewed the same day as opposed to several

Figure 6.9. Regular sized silver-enhanced Nanogold (B, 8–10 nm) compared to silver-enhanced 5 nm colloidal gold (A, 15–20 nm). Ws-keratin was detected with a secondary 5 nm gold-IgG goat antimouse IgG (GAM-G5) (A) and Nanogold (B), both followed by a silver enhancement for 3 min. Tissue was a mammary carcinoma, embedded in Lowicryl K4M resin. The quite homogeneous particles are about 8 to 10 nm with Nanogold (B), and 15 to 20 nm using GAM-G5 (A). Bar = 200 nm. (Reproduced with permission from Krenács, T. and L. Krenács, 1995, in Immunogold–Silver Staining: Principles, Methods and Applications, CRC Press, Boca Raton, pp.57–70.)

102

Microscopic Uses of Nanogold® months of film exposure for the autoradiographic approach previously used. Another recent example of the use of Nanogold reagents was the labeling of insulin to study the insulin receptor.22 Nanogold–insulin was found to retain its activity and, when bound to the insulin receptor, was then used to help determine the high resolution structure of this important protein. The high resolution scanning transmission electron microscope (STEM) is excellent for

Figure 6.10. Diagram of Nanogold reagents. (A) Monomaleimido–Nanogold has a maleimide group that reacts with free sulfhydryls, and (B) mono-NHS-Nanogold has a sulfoN-hydroxysuccinimide ester that reacts with amines. Both reactions are very specific and occur rapidly under mild physiological conditions.

Figure 6.11. TEM brightfield micrograph of red blood cells labeled with Nanogold antihuman red blood cell (RBC)Fab′′, then fixed, embedded in Lowicryl K4M, and thin sectioned. No additional staining was used, and in this 80-nm section, individual Nanogold particles are visible (arrow) along the membrane surfaces. Full width 0.18 µm.

Figure 6.12. Nanogold-DPPE schematic diagram, showing covalent attachment of Nanogold to the head group of a 16 carbontailed phospholipid, DPPE.

103

Gold and Silver Staining

Figure 6.13. Nanogold-DPPE labeling of a drug-delivery liposome. (Top panel) Schematic showing liposome containing the antifungal drug, amphotericin B (AmBisome), and Nanogold–lipid-labeled. (Middle panel) Schematic of experiments showing targeting of liposomes (small circles) to a fungal cell, followed by silver enhancement. If drug was absent (a), liposomes targeted the cell, but had no further effect. With drug, (b), the cell wall was penetrated, and drug and lipids entered the cell. Micrographs at bottom show effect after a 14-h incubation of liposomes with the fungal cells Aspergillus fumigatas. Without the drug in the liposome, liposomes remained exterior to cells (a); with the drug, dark internal staining was observed (b), indicating breach of cell wall and membrane and entry into cells, eventually resulting in cell death. Nu = nucleus, L = Nanogold-labeled lipid or liposomes, M = mitochondria, CW = cell wall, and CM = cell membrane. Bar = 200 nm. (Micrographs reproduced with permission from Adler-Moore, J., 1994, Bone Marrow Transplant., 14(Suppl) 5.S3–7.)

104

Microscopic Uses of Nanogold® visualizing Nanogold and even undecagold, tetrairidium, and single heavy atoms.29 However, microscopes are improving all the time, and smaller and smaller gold particles have become visible with more commonly available instruments. Nanogold is visible by both TEM and scanning electron microscopy (SEM).17 In TEM, the specimen must be kept thin, so as not to swamp out the small gold signal. For example, Nanogold has been seen with a Philips 300 (FEI, Hillsboro, OR, USA) in approximately 70 nm thin sections of cells12 but without the use of stains (Figure 6.11). It has also been seen routinely in cryoEM work, where unstained molecules are embedded in a thin (approximately 100 nm) layer of ice.2 For single molecule studies, negative staining has been popular for enhancing the contrast and subunit boundaries, but Nanogold is typically lost in the dense layer of uranyl acetate or lead citrate. However, several lower atomic number stains may be used instead. These stains still enhance the molecular images, but scatter less, and permit visualization of each Nanogold particle. Some stains in this category are methylamine vanadate (NanoVan),7 ammonium molybdate, glucose mixed with molybdate, and aurothioglucose–glucose. A protocol for using the vanadate stain is given. Use of the gold–phospholipid, NanogoldDPPE (Figure 6.12) to visualize liposomes in the EM and LM has now been reported by several groups.14,33 Nanogold-DPPE-spiked liposomes have even been used to follow and document liposome-based antifungal drug delivery1 (Figure 6.13). For many applications, autometallography (silver enhancement) is required to visualize the small gold. Protocols for its use with Nanogold are given in companion chapters of this volume. An in-depth review comparing a number of investigators’ experiences with Nanogold labeling covering additional pro-

tocols and comparison of such things as fixation methods and tissue handling techniques is found in the literature.15 Also, applications of new reagents based on the covalently linkable Nanogold are described in another chapter in this volume. In conclusion, Nanogold molecular and immunolabeling have become important new tools for cell biologists and biomedical researchers and continue to expand the usefulness of gold labeling. ACKNOWLEDGMENTS The authors wish to thank Dr. Martha Simon, Ms. Beth Lin, and Mr. Frank Kito for STEM operation, and Dr. Joseph Wall for helpful discussions. This work was supported by the Office of Biological and Environmental Research of the U.S. Department of Energy under Prime Contract No. DE-AC02-98CH10886 with Brookhaven National Laboratory, by National Institutes of Health (NIH) Grant No. 2P41 RR01777 and by NIH Small Business Innovation Research Grant Nos. GM49564, and GM56090. REFERENCES 1.Adler-Moore, J. 1994. AmBisome targeting to fungal infections. Bone Marrow Transplant. 14(Suppl 5):S3-7. 2.Boisset, N., R. Grassucci, P. Penczek, E. Delain, F. Pochon, J. Frank, and J.N. Lamy. 1992. Three-dimensional reconstruction of a complex of human alpha 2macroglobulin with monomaleimido Nanogold (Au1.4nm) embedded in ice. J. Struct. Biol. 109:3945. 3.Burry, R.W., D.D. Vandre, and D.M. Hayes. 1992. Silver enhancement of gold antibody probes in preembedding electron microscopic immunocytochemistry. J. Histochem. Cytochem. 40:1849-1856. 4.Danscher, G. and J.O. Norgaard. 1983. Light microscopic visualization of colloidal gold on resin-embedded tissue. J. Histochem. Cytochem. 31:1394-1398. 5.Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemistry 8:1081-1083. 6.Gilerovitch, H.G., G.A. Bishop, J.S. King, and R.W. Burry. 1995. The use of electron microscopic

105

Gold and Silver Staining immunocytochemistry with silver-enhanced 1.4-nm gold particles to localize GAD in the cerebellar nuclei. J. Histochem. Cytochem. 43:337-343. 7.Gregori, L., J.F. Hainfeld, M.N. Simon, and D. Goldgaber. 1997. Binding of amyloid beta protein to the 20 S proteasome. J. Biol. Chem. 272:58-62. 8.Hacker, G.W., L. Grimelius, G. Danscher, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc tissues. J. Histotechnol. 11:213. 9.Hainfeld, J. 1990. STEM analysis of Janssen Aurooprobe One, p. 954-955. In G. Bailey (Ed.), Proc. XIIth Int. Cong. Elec. Micros. San Francisco Press, San Francisco. 10.Hainfeld, J.F. 1996. Labeling with nanogold and undecagold: techniques and results. Scanning Microsc. Suppl. 10:309-322. 11.Hainfeld, J.F. 1989. Undecagold-antibody method, p. 413-429. In M.A. Hayat (Ed.), In Colloidal Gold: Principles, Methods, and Applications, Vol. 2. Academic Press, San Diego. 12.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40:177-184. 13.Hainfeld, J.F. and F.R. Furuya. 1995. Silver enhancement of Nanogold and undecagold, p. 71-96. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Principles, Methods and Applications. CRC Press, Boca Raton. 14.Hainfeld, J.F., F.R. Furuya, and R.D. Powell. 1999. Metallosomes. J. Struct. Biol. 127:152-160. 15.Hainfeld, J.F. and R.D. Powell. 1997. Nanogold technology: new frontiers in gold labeling. Cell Vision 4:408-432. 16.Handley, D.A. 1989. Methods for synthesis of colloidal gold p. 13-32. In M.A. Hayat (Ed.), In Colloidal Gold: Principles, Methods, and Applications, Vol. 1. Academic Press, San Diego. 17.Hermann, R., P. Walther, and M. Muller. 1996. Immunogold labeling in scanning electron microscopy [published erratum appears in Histochem. Cell Biol. 1996 Sep;106(3):356. Histochem. Cell Biol. 106:31-39. 18.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird. 1983. Immunogold–silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31:938-944. 19.Kramarcy, N.R. and R. Sealock. 1991. Commercial preparations of colloidal gold-antibody complexes frequently contain free active antibody. J. Histochem. Cytochem. 39:37-39. 20.Krenács, T. and L. Dux. 1994. Silver-enhanced immunogold labeling of calcium-ATPase in sarcoplasmic reticulum of skeletal muscle [letter; comment]. J. Histochem. Cytochem. 42:967-968. 21.Krenács, T. and L. Krenács. 1995. Comparison of embedding media for immunogold–silver staining, p. 57-70. In M.A. Hayat (Ed.), Immunogold–Silver

106

Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 22.Luo, R.Z., D.R. Beniac, A. Fernandes, C.C. Yip, and F.P. Ottensmeyer. 1999. Quaternary structure of the insulin-insulin receptor complex. Science 285:10771080. 23.Mariani, M., M. Camagna, L. Tarditi, and E. Seccamani. 1991. A new enzymatic method to obtain highyield F(ab)2 suitable for clinical use from mouse IgGl. Mol. Immunol. 28:69-77. 24.Nusser, Z., E. Mulvihill, P. Streit, and P. Somogyi. 1994. Subsynaptic segregation of metabotropic and ionotropic glutamate receptors as revealed by immunogold localization. Neuroscience 61:421-427. 25.Nusser, Z., J.D. Roberts, A. Baude, J.G. Richards, and P. Somogyi. 1995. Relative densities of synaptic and extrasynaptic GABAA receptors on cerebellar granule cells as determined by a quantitative immunogold method. J. Neurosci. 15:2948-2960. 26.Parham, P. 1983. On the fragmentation of monoclonal IgG1, IgG2a, and IgG2b from BALB/c mice. J. Immunol. 131:2895-2902. 27.Robinson, J.M., T. Takizawa, D.D. Vandre, and R.W. Burry. 1998. Ultrasmall immunogold particles: important probes for immunocytochemistry. Microsc. Res. Tech. 42:13-23. 28.Roth, J., M. Bendayan, and L. Orci. 1978. Ultrastructural localization of intracellular antigens by the use of protein A-gold complex. J. Histochem. Cytochem. 26:1074-1081. 29.Safer, D., J. Hainfeld, J.S. Wall, and J.E. Reardon. 1982. Biospecific labeling with undecagold: visualization of the biotin-binding site on avidin. Science 218:290-291. 30.Segond von Banchet, G. and B. Heppelmann. 1995. Non-radioactive localization of substance P binding sites in rat brain and spinal cord using peptides labeled with 1.4-nm gold particles. J. Histochem. Cytochem. 43:821-827. 31.Sun, X.J., L.P. Tolbert, and J.G. Hildebrand. 1995. Using laser scanning confocal microscopy as a guide for electron microscopic study: a simple method for correlation of light and electron microscopy. J. Histochem. Cytochem. 43:329-335. 32.Takizawa, T. and J.M. Robinson. 1994. Use of 1.4nm immunogold particles for immunocytochemistry on ultra-thin cryosections. J. Histochem. Cytochem. 42:1615-1623. 33.Thurston, G., J.W. McLean, M. Rizen, P. Baluk, A. Haskell, T.J. Murphy, D. Hanahan, and D.M. McDonald. 1998. Cationic liposomes target angiogenic endothelial cells in tumors and chronic inflammation in mice. J. Clin. Invest. 101:1401-1413. 34.Vandre, D.D. and R.W. Burry. 1992. Immunoelectron microscopic localization of phosphoproteins associated with the mitotic spindle. J. Histochem. Cytochem. 40:1837-1847.

7

Combined Fluorescent and Gold Probes for Microscopic and Morphological Investigations Richard D. Powell and James F. Hainfeld

INTRODUCTION Nanogold®, a gold cluster with a core of gold atoms 1.4 nm in diameter, has proven to be a superior probe label for electron microscopy (EM),14,17 giving both higher labeling density28,31 and improved access to previously hindered or restricted antigens.29,31 It may be visualized by autometallography (AMG) for use in light microscopy (LM): silver-and gold-amplified Nanogold detection12,18 has proven to be one of the most sensitive methods available for the detection of low copy number targets such as viral DNA in cells and tissue specimens.32 AMG enhancement has also made Nanogold an effective detection label in blots15 and gels.10 The following protocols will be described: • Labeling of nuclear components in cells. • Protocol for in situ hybridization and detection with fluorescein–Nanogoldor Cy3–Nanogold-labeled streptavidin. Nanogold is an inert molecule, and generally does not interact with biological

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

molecules unless a specific chemical reactivity is introduced into the molecule. Conjugates are prepared using site-specific chemical conjugation through reactive chemical functionalities introduced during Nanogold preparation, which allows the gold label to be attached to a specific site on the conjugate biomolecule. For example, a maleimido–Nanogold derivative, which is specific for thiol binding, is frequently attached to the hinge region of an antibody at a unique thiol site generated by selective reduction of a hinge disulfide.14 This site is remote from the antigen combining region, and the Nanogold, therefore, does not compromise target binding. Nanogold may also be prepared with specific reactivity towards amines27 or other unique chemical groups. This mode of attachment enables the preparation of probes labeled with both Nanogold and fluorescent labels. Different chemical reactivities are used to attach the Nanogold and the fluorescent groups to different sites in the conjugate biomolecule, as shown in Figure 7.1. In this manner, the two labels are spaced sufficiently far apart that fluorescent resonance energy

107

Gold and Silver Staining transfer8,21 does not quench the fluorescent signal, and the probes may be used to label specimens for fluorescent and EM observation in a single staining procedure.23,26 This reduces the complexity of the staining procedure, allowing less specimen perturbation, and also enables a higher degree of correlation between the fluorescence and EM localization of the target,30 thus increasing the usefulness of the complementary data sets. Since gold and fluorescent-labeled probes are often used at different concentrations under different conditions, optimum procedures for the use of fluorescent and gold probes may entail some degree of compromise between the most appropriate conditions for the two types of probes. However, the chemical stability of the Nanogold label means that it is generally stable to a wide range of use conditions, and the following protocols have been found to be effective for labeling specimens with combined fluorescein and Nanogold-labeled antibody Fab′ probes and with combined Cy3 and Nanogold-labeled streptavidin.

PROTOCOLS Protocol 1. Immunoabeling of Cellular Components, after Spector23 Materials and Reagents • Phosphate-buffered saline (PBS): 0.02 mol/L sodium phosphate buffer with 0.15 mol/L sodium chloride, pH adjusted to 7.6. • Normal goat serum (NGS). • Fluorescein and Nanogold-labeled Fab′ (FluoroNanogold) conjugate (Nanoprobes, Yaphank, NY, USA). • Glutaraldehyde. • 0.02 mol/L sodium citrate buffer, pH 7.0. • HQ Silver enhancement kit (Nanoprobes). • Hydrofluoric acid (37% or 40% aqueous). • Neutralizing solutions for hydrofluoric acid. Dissolve one heaped spoonful of each into water in numbered 100 mL plastic tripour beakers:

Figure 7.1. Structure of combined fluorescein and Nanogold probe. This novel reagent is prepared by sequential conjugation of Nanogold followed by fluorescein.

108

Combined Fluorescent and Gold Probes 1. Converter for hydrofluoric acid; magnesium sulfate. 2. Neutralizer for acids; sodium carbonate. • Uranyl acetate. • Reynolds lead citrate: place lead nitrate, Pb(NO3)2 (1.33 g), trisodium citrate dihydrate, Na3(C6H5O7)•2H2O (1.76 g), and 30 mL of distilled water in a 50 mL volumetric flask, shake vigorously for 1 min, and allow to stand for 30 min with occasional shaking to ensure complete conversion of lead nitrate to lead citrate before use.24 Staining Procedure 1. Grow HeLa cells on coverslips for two days. Fix, wash, and permeabilize.9 2. Incubate cells with primary antibody at appropriate dilution for 1 h (diluted in PBS containing 0.1% NGS). 3. Wash in PBS containing 1% NGS. 4. Incubate with a 1:10 dilution of fluorescein–Nanogold-conjugated Fab′ secondary antibody probe for 1 h at room temperature. At this point, fluorescence photomicrographs may be acquired. 5. Wash cells with PBS. 6. Fix in 1% glutaraldehyde in PBS for 15 min. 7.Wash with PBS (3 time for 10 min). Prior to silver enhancement, change the buffer to 0.02 M sodium citrate buffer, pH 7.0. Wash cells extensively in this buffer (chlorides must be removed before silver enhancement). 8 .Perform the silver enhancement procedure in the darkroom using a Thomas Duplex sodium vapor or equivalent safelight. HQ Silver may be used to enhance the gold probe in the cells. Prepare HQ Silver by vortex mixing a 1:1:1 mixture of initiator, moderator, and activator. Dry the backs of the

coverslips using filter paper, and apply 200 µL of the silver enhancement solution to the cell side of the coverslip. After approximately 15 min, or when the silver changes from clear to gray, wash it off the coverslip using citrate buffer. 9. Once silver enhancement is complete, wash the cells extensively with citrate buffer to remove any nonspecific silver deposits and to prevent any further silver enhancement. 10. Dehydrate cells through a graded series of ethanol. Infiltrate with a 50:50 solution of ethanol and Epon-Araldite (Electron Microscopy Sciences, Fort Washington, PA, USA) for 18 h, followed by 100% Epon-Araldite for 8 h. Embed coverslips in Epon-Araldite and place in an oven at 60°C for 48 h to polymerize. 11. Remove the glass coverslips using hydrofluoric acid (see Addendum below). 12. Section the embedded cells and pick up on 200 mesh copper grids. 13. Counterstain with 5% uranyl acetate for 5 min. 14. Stain with Reynolds lead citrate24 for 1 min; sections may also be regressed by floating on 0.2 M EDTA in distilled water for 30 min between the uranyl acetate and lead citrate staining.4 15. View with transmission EM operated at 75 kV. Protocol Addendum: Removing Coverslips with Hydrofluoric Acid Caution: Hydrofluoric acid is extremely dangerous. Always handle in a fume hood. Keep neutralizing solutions (see below) handy. Use laboratory coat, goggles, and double gloves (nitrile or rubber if at all possible).

109

Gold and Silver Staining Procedure 1. Remove excess resin from bottom of coverslip by scraping with a razor blade. Wipe with an acetone-soaked cotton swab to remove all the resin. The glass has to be completely clean or glass slivers will be left on the face of your blocks after removal. 2. Use half of a 35 mm tissue culture dish for each coverslip. Place these on the Teflon sheet in the fume hood and put the blocks into the 35 mm dish halves, coverslip side up. 3. Set up two more 100 mL plastic beakers with deionized water and put on laboratory coat, goggles, and gloves. You need to double glove; use nitrile or rubber gloves if at all possible. Also have kimwipes, water bottle, forceps, and plastic transfer pipet in the hood. 4. Use a plastic transfer pipet to add hydrofluoric acid dropwise to the coverslips. Fifteen to 20 drops per coverslip is usually plenty. Let sit for 10 min. If the hydrofluoric acid runs over the edge, turn the coverslip over and float it on pool of hydrofluoric acid in the dish. 5. After 10 min, pick up coverslip with forceps and dip in beaker 1, then in beaker 2, then rinse with water, and dry with kimwipes. The glass will probably only be etched after the first exposure to hydrofluoric acid. Repeat steps 4 and 5 until the glass is completely removed. Gridded coverslips tend to take longer to remove than do regular coverslips. 6. Put the plastic sample blocks in a plastic beaker and wash for 2 h with at least 3 changes of water. 7. Dishes, forceps, and pipets exposed to hydrofluoric acid are rinsed through 1, then 2, and then in another beaker of water to rinse extensively before being disposed of as solid waste. Make sure you have not splashed acid on the teflon 110

sheet. If you have, neutralize with a few drops of solutions 1, then 2, then wash with water. Dump solutions 1 and 2 down the sink and wash with running water at least 15 min apart so that they do not react. Wipe the Teflon sheet before storing. An example of labeling using this method is shown in Figures 7.2 and 7.3, in which a monoclonal primary antibody followed by combined fluorescein and Nanogoldlabeled Fab′ secondary have been used to label the pre-mRNA splicing factor SC35 in HeLa cells. Protocol 2. In Situ Hybridization and Detection with Fluorescein–Nanogoldor Cy3–Nanogold-Labeled Streptavidin Protocols 2 and 3 were developed by C. Hauser-Kronberger and G.W. Hacker and co-workers (Salzburg) using combined Cy3 and Nanogold-labeled streptavidin,24 but are also effective with fluorescein and Nanogold-labeled streptavidin. They are based on their method developed for in situ hybridization detection with Nanogold®.12,32 Materials and Reagents • PBS: 10× PBS (Mg2+ and Ca2+ free), pH 7.6: 11.36 g Na2HPO4, 2.72 g KH2PO4, 87.0 g NaCl in 800 mL distilled water. Adjust pH with concentrated NaOH and add distilled water to a final volume of 1 L. • Standard sodium citrate buffer (SSC): 175.32 g NaCl and 88.23 g sodium citrate in 800 mL distilled water. Adjust pH with NaOH to 7.0 and add distilled water to a final volume of 1 L. • Prehybridization buffer: 50% deionized formamide/10% dextran sulfate in 2× SSC. • Lugol’s iodine (Cat. No. L-6146; Merck, Darmstadt, Germany). • Gelatin (45%): Cold Water Fish

Combined Fluorescent and Gold Probes Gelatin (Cat. No. G-7765; SigmaAldrich, Steinheim, Germany). • Fluorescein–Nanogold- or Cy3–Nanogold-labeled streptavidin (Nanoprobes). • Silver acetate AMG reagents (see below). • Reagent grade alcohols. • Ultrapure water. • Proteinase K (Cat. No. 1373196; Roche Molecular Biochemicals, Mannheim, Germany). • Biotinylated DNA probe (e.g., Enzo, New York, NY, USA). • Coplin jars or equivalents. • Standard laboratory pipettors. • Slide warmers and/or incubators (37° and 50°C capable). • Heating block (92°–95°C capable). • Nuclear stains. • Mounting media: Permount or DPX (BDH Chemicals, Poole, England, UK).

Staining Procedure 1. Deparaffinize sections from formaldehyde-fixed tissue in fresh xylene (2 times for 15 min each). 2. Rinse and rehydrate in graded alcohols and distilled water (2–3 min each). 3. Soak in PBS (20 mmol/L, pH 7.6) for 3 min. 4. Incubate with 0.1 mg/mL proteinase K in PBS at 37°C for about 8 min. The duration is critical and has to be tested very carefully, depending on tissue, fixation, and other factors. 5. Rinse in 2 changes of PBS, for 3 min. 6. Permeabilize with 0.3% Triton X-100 in PBS for 15 min. 7. Wash in PBS for 2 min. 8. Rinse in 2 changes of distilled water, dehydrate with graded alcohols (50%, 70%, and 98% isopropanol) for 1 min each, and air-dry the sections.

Figure 7.2. HeLa cells. Fluorescence photomicrographs of labeled HeLa cells (A) stained with monoclonal primary antibody against SC35 pre-mRNA splicing factor, followed by secondary fluorescein and Nanogold-labeled antimouse Fab′. (B) Control stained with fluorescein and Nanogold-labeled antimouse Fab′ only (primary antibody omitted). Final original magnification = 1000×, bar = 10 µm. (Reproduced with permission from Powell, R.D. et al., 1997, J. Histochem. Cytochem., 45:947–956.)

111

Gold and Silver Staining 9. Prehybridize with 1:1 mixture of deionized formamide and 20% dextran sulfate in 2× SSC at 50°C for 5 min. 10. Carefully shake off excess prehybridization block. 11. Add one drop of biotinylated DNA probe on the section and cover with a small coverslip. Avoid air bubbles. 12. Heat sections on heating block at 92°–94°C for 8 to 10 min to denature DNA. 13. Incubate in a moist chamber at 37°C overnight (or for at least 2 h). 14. Posthybridization washes (5 min each): 2 changes of 4× SSC (first wash to remove coverslips), 2× SSC, 0.1× SSC, 0.05× SSC, and then distilled water. 15. Put slides into Lugol’s iodine solution for 5 min. 16. Wash in tap water and then distilled water.

Figure 7.3. Transmission electron micrographs of labeled HeLa cells. (A) Cells stained with monoclonal primary antibody against SC35 pre-mRNA splicing factor, followed by secondary fluorescein and Nanogold-labeled antimouse Fab′. (B) Control stained with fluorescein and Nanogold-labeled antimouse Fab′ only (primary antibody omitted). Bar = 1 µm. (Final manification ×12,000.) (Reproduced with permission from Powell, R.D. et al., 1997, J. Histochem. Cytochem., 45:947–956.)

112

17. Put into 2.5% sodium thiosulfate for a few seconds until sections are colorless. Then wash in tap water for 5 min and distilled water for 2 min. 18. Immerse in PBS containing 0.1% fish gelatin (45% concentrate) and 0.1% Tween 20 for 5 min. 19. Incubate sections with fluorescein– Nanogold- or Cy3–Nanogold-labeled streptavidin diluted 1:200 to 1:500 in PBS containing 1% bovine serum albumin (BSA) at room temperature for 60 min. 20. Wash in 3 changes of PBS containing 0.1% fish gelatin and 0.1% Tween 20 for 5 min each. 21. Repeatedly wash in distilled water for at least 10 min altogether, the last 2 rinses in ultrapure water (EM grade). 22. Perform silver acetate AMG (see Addendum) or GoldEnhance develop-

Combined Fluorescent and Gold Probes ment (Nanoprobes; see also Chapter 3). 23. Rinse carefully in tap water for at least 3 min. After silver amplification, sections can be counterstained with nuclear fast red, dehydrated, and mounted in Permount or in DPX. Do not use Eukitt (Electron Microscopy Sciences; this can change the intensity of silver stains after optimal development, sometimes even after weeks and months). Protocol 3. In Situ Hybridization, Tyramide Signal Amplification and Detection with Combined Fluorescein–Nanogold- or Cy3– Nanogold-Labeled Streptavidin Materials and Reagents This procedure will require the same materials, reagents, and equipment used for in situ hybridization detection (see Protocol 2). Tyramide signal amplification will additionally require the following • Hydrogen peroxide. • Methanol. • Tween 20. • Dimethyl sulfoxide (DMSO). • Fluorescein isothiocytrate (FITC) or biotin haptenated probe. • GenPoint in situ streptavidin-biotinperoxidase complex kit (DAKO, Glostrup, DK, and Carpinteria, CA, USA), which includes blocking powder and biotinylated tyramide. Staining Procedure 1. Deparaffinize sections from formaldehyde-fixed tissue in fresh xylene (2 times for 15 min each). 2. Rinse in absolute ethanol (2 times for 5 min each), then 95% ethanol (2 times for 5 min each), followed by 2 changes of double-distilled water. 3. Immerse the slides in Target Retrieval solution (Cat. No. S1700; DAKO) at

95°C for 40 min, then let the slides cool in the same solution for 20 min. 4. Rinse the slides in several changes of double-distilled water, then incubate them with proteinase K (Cat. No. S3004; Dako) diluted 1:5000 in 50 mmol/L Tris-HCl buffer (pH 7.6) for 5 min at room temperature. Alternatively, steps 3 and 4 may be replaced by pretreatment with 0.1 mg/mL proteinase K in 50 mmol/L Tris-HCl buffer (pH 7.6) at 37°C for about 8 min (optimal duration should be tested). 5. Wash slides in double-distilled water (3 changes for 5 min each). 6. Treat with 3% H2O2 in methanol at room temperature for 30 min. 7. Wash slides in double-distilled water for 10 min. 8. Put slides into Lugol’s iodine solution for 5 min, then wash in double-distilled water. 9. Put into 2.5% sodium thiosulfate for a few seconds until sections are colorless, then wash in double-distilled water (2 times for 5 min each). 10.Allow slides to air-dry. 11.Add one drop of biotinylated DNA probe on the section and cover with a small glass coverslip. Avoid air bubbles. 12.Heat sections on heating block at 92° to 94°C for 5 min to denature DNA. 13.Incubate in a moist chamber at 37°C overnight (or for at least 1 h). 14.Remove the coverslip by soaking slides in a TBST (Tris-buffered saline containing Tween 20) bath for 5 min. 15.Incubate slides in Stringent Wash (provided in the GenPoint kit) for 20 min at 55°C. 16. Drain off section, wipe area around section dry, and surround it with a PAP-pen (Dako-Pen) (Cat. No. S-2002; Dako). 17. Immerse slides in TBST for 5 min. 113

Gold and Silver Staining 18. Apply primary streptavidin-horseradish peroxidase (HRP), diluted 1:800 in the diluent (from GenPoint kit) to sections and incubate in a moist chamber for 15 min at room temperature. 19. Wash in 3 changes of TBST for 5 min each. 20.Apply ready-to-use biotinyl-tyramide solution (from GenPoint kit) and incubate in a moist chamber for 15 min at room temperature. 21.Wash in 3 changes of TBST-gelatin (TBST containing 0.1% fish gelatin pH 7.6) for 5 min each. 22.Incubate the sections with fluorescein– Nanogold- or Cy3–Nanogold-labeled streptavidin diluted 1:250 in PBS containing 1% BSA at room temperature for 60 min. 23.Wash in 3 changes of TBST-gelatin for 5 min each. 24.Repeatedly wash in ultrapure water (EM grade). 25.Perform silver acetate AMG or GoldEnhance development according to the instructions of the manufacturer. Note: The development process may be stopped by simply washing the sections in distilled water (several changes), or more effectively by a brief wash in 2%–5% sodium thiosulfate, followed by water. 26.After AMG amplification, sections can be counterstained with hematoxilin and eosin and/or nuclear fast red, dehydrated, and mounted in Permount or in DPX. Avoid the use of Eukitt. This can change the intensity of silver stains after optimal development, sometimes even after weeks and months. Protocol Addendum: Silver Acetate Autometallography12 The procedure described here was modified for use with Nanogold reagents. 114

Materials and Reagents • Solutions A and B should be freshly prepared for every run. Solution A: Dissolve 80 mg silver acetate (85140; Fluka, Buchs, Switzerland) in 40 mL of glass double-distilled water. Silver acetate crystals can be dissolved by continuous stirring within about 15 min. • Citrate buffer: Dissolve 23.5 g of trisodium citrate dihydrate and 25.5 g citric acid monohydrate in 850 mL of deionized or distilled water. This buffer can be kept at 4°C for at least 2 to 3 weeks. Before use, adjust to pH 3.8 with citric acid solution. • Solution B: Dissolve 200 mg hydroquinone in 40 mL citrate buffer. • Enhancement solution: Just before use, mix solution A with solution B. Staining Procedure 1. Silver amplification: Place the slides vertically in a glass container (preferably with about 80 mL volume and up to 19 slides; Schiefferdecker-type) and cover them with the mixture of solutions A and B. Staining intensity can be checked in the LM during the amplification process, which usually takes about 5 to 20 min, depending on primary antibody or nucleic acid probe concentration, incubation conditions, and the amount of accessible antigen or nucleic acid sequence in question. 2. Stop enhancement by washing in distilled water (several changes). 3. After stopping the enhancement process, slides can be examined in an LM more carefully. If staining intensity is still too low, wash slides for one more time in double-distilled water and develop further in enhancement solution. Figure 7.4 shows two examples of the

Combined Fluorescent and Gold Probes results obtained by this method. In this specimen, human papillomavirus type 16 (HPV-16) DNA is visualized in CaSki cells and in SiHa cells; the latter are known to contain only 1 to 2 copies of the target DNA, and the staining is clearly visualized by both fluorescence and LM. TECHNICAL HINTS AND DISCUSSION The optimum conditions for labeling with combined fluorescent and Nanogold

probes generally will be close to those for using Nanogold probes in similar procedures, and this should be used as the starting point when using the combined fluorescent and Nanogold probes. The protocols above are intended as guidelines. Combined fluorescent and Nanogold probes are entirely covalently linked and stabilized. Therefore, the problems sometimes found with colloidal gold probes, which include aggregation leading to poor antigen access,13 nonspecific binding,3 and probe dissociation, which gives rise to lower labeling,19 will usually be greatly

Figure 7.4. In situ hybridization for HPV-16 DNA with biotinylated tyramides and Cy3–Nanogold-labeled streptavidin. Top row: CaSki cells (A) Fluorescence (bar = 20 µm), (B) Light micrograph after gold AMG (bar = 10 µm), (C) Electron micrograph of same specimen (bar = 1 µm). Bottom row: SiHa cells by (D) fluorescence, (E) LM after gold AMG (bar = 10 µm), (F) fluorescence, and (G) light micrographs of control with biotin-tyramide omitted (bar = 20 µm).

115

Gold and Silver Staining reduced or negligible for probes containing Nanogold. The probes are stable to a wide range of buffers and experimental conditions, and may therefore be used in many staining and processing protocols. Thiols have a strong affinity towards gold and can degrade Nanogold. Therefore, thiol-containing reagents, such as dithiothreitol (DTT), should be avoided when staining with Nanogold-containing probes. If thiols must be used, concentrations should be kept below 1 mM and exposure times kept below 10 min. In addition, some reports suggest that Nanogold may be affected by exposure to high temperatures. However, a detailed study by UV visible spectroscopy has shown that about 80% of Nanogold is still intact even after several h at 100°C.15 The risk from such conditions may be avoided by performing AMG before exposure. When higher concentrations of the combined fluorescent and Nanogold probe are required to obtain sufficiently intense fluorescent signals, this can result in higher background signals upon silver–salt-based AMG. This may be avoided by washing thoroughly with sodium citrate buffer before enhancement. Where HQ Silver is used, 0.02 M sodium citrate buffer at pH 7.0 has been found to be most effective; in preparations utilizing the Danscher silver enhancement protocol,7 0.02 M sodium citrate buffer adjusted to pH 3.5 was most effective. In some procedures, poor development has been found upon AMG. Results may be improved by changing from commercial silver enhancement reagents to the Danscher formulation7 or Hacker’s silver acetate modifications,12 or by substituting formaldehyde for glutaraldehyde in postfixation. Combined fluorescein and Nanogold probes have sometimes shown a nonspecific affinity for nucleic acids or other nuclear materials, possibly due to the presence of 116

the hydrophobic fluorescein in close proximity to Nanogold, which possesses similarly hydrophobic aromatic substituents, which might interact with the bases of nucleic acids. Treatment with either a detergent, such as Tween 20, or an amphiphilic reagent, such as benzamidine or 1,2,3heptanetriol,20 may reduce this effect. Enlargement of the Nanogold by AMG might be expected to reduce the fluorescence yield, since the enlarged particle will have higher extinction coefficients. However, Robinson and Vandré have shown that short periods of silver enhancement (1–2 min), which are sufficient to render the Nanogold particles visible by EM, are tolerated, and sufficient fluorescence is retained for useful correlation of the data found by fluorescence and other optical microscopy methods.26 Combined fluorescent and Nanogold probes are only one of a number of new probes that have been made possible by the development of covalent metal cluster and nanoparticle labeling technology. Other metal clusters adapted for use as microscopy labels include 2 nm platinum clusters. These are larger and more readily visualized than Nanogold even without AMG.22 The smaller tetrairidium cluster has also been used in conjunction with image processing for highresolution structural studies of large protein assemblies such as viral capsids.5 Larger probe labels prepared using the same covalent cross-linking rationale used for Nanogold would possess many advantages for EM applications, and because covalent linking enables the preparation of probes containing molecules such as lipids1,16 and oligonucleotides2 as well as the antibodies and proteins used for colloidal gold conjugation, it would expand the range of probes available. We have recently prepared a covalently linked 10 nm gold probe, and this was successfully used to label the proteins that make up the polar tube apparatus in microsporida

Combined Fluorescent and Gold Probes spores.11 Further development of this technology is currently in progress, and it is hoped that such probes will be made available commercially in the future. ACKNOWLEDGMENTS We are grateful to Dr. D.L. Spector, S. Kaurin, and J. McCann of Cold Spring Harbor Laboratory for developing Protocol 1 for cellular labeling and for helpful discussions in resolving background issues. I am also grateful to Drs. G.W. Hacker, C. Hauser-Kronberger (both Salzburg), and A.L.M. Cheung (Hong Kong) for the in situ hybridization protocols. This work was supported by the Office of Biological and Environmental Research of the US Department of Energy under Prime Contract No. DE-AC02-98CH10886 with Brookhaven National Laboratory by National Institutes of Health Grant 2P41RR61777 and by Small Business Innovation Research Grant Nos. GM48328, GM56090, and GM60067 from the National Institute of General Medical Sciences, National Institute of Health. REFERENCES 1.Adler-Moore, J. 1994. AmBisome targeting to fungal infections. Bone Marrow Transplant. 14:S3-S7. 2.Alivisatos, A. P., K.P. Johnsson, X. Ping, T.E. Wilson, C.J. Loweth, M.P. Bruchez, Jr., and P.G. Schultz. 1996. Organization of ‘nanocrystal molecules’ using DNA. Nature 382:609-611. 3.Behnke, O, T. Ammitzboll, H. Jessen, M. Klokker, K. Nilausen, J. Tranum-Jensen, and L. Olsson. 1986. Non-specific binding of protein-stabilized gold sols as a source of error in immunocytochemistry. Eur. J. Cell Biol. 41:326-338. 4.Bernhard, W. 1969. A new staining procedure for electron microscopical cytology. J. Ultrastruct. Res. 27:250-265. 5.Cheng, N., J.F. Conway, N.R. Watts, J.F. Hainfeld, V. Joshi, R.D. Powell, S.J. Stahl, P.E. Wingfield, and A.C. Steven. 1999. Tetrairidium, a 4-atom cluster, is readily visible as a density label in 3D cryo-EM maps of proteins at 10-25 Å resolution. J. Struct. Biol. 127:169-176.

6.Danscher, G. 1981. Histochemical demonstration of heavy metals. A revised version of the silver sulphide method suitable for both light and electron microscopy. Histochemistry 71:1-16. 7.Danscher, G. 1981. Localization of gold in biological tissue. A photochemical method for light and electron microscopy. Histochemistry 71:81-88. 8.Forster, Th. 1948. Zwischenmolekulare Energiewandung und Fluoreszenz. Ann. Physik. 2:55-75. 9.Fu, X.-D. and T. Maniatis. 1990. Factor required for mammalian spliceosome assembly is localized to discrete regions in the nucleus. Nature 343:437-441. 10.Gilerovitch, H.G., G.A. Bishop, J.S. King, and R.W. Burry. 1995. The use of electron microscopic immunocytochemistry with silver-enhanced 1.4-nm gold particles to localize GAD in the cerebellar nuclei. J. Histochem. Cytochem. 43:337-343. 11.Gutierrez, E., R.D. Powell, J.F. Hainfeld, and P.M. Takvorian. 1999. A covalently linked 10 nm gold immunoprobe, p. 1324-1325. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mansfield, and R.L. Price (Eds.), Proc. 57th Ann. Mtg., Micros. Soc. Amer. Springer-Verlag, New York. 12.Hacker, G.W., C. Hauser-Kronberger, I. Zehbe, H. Su, A. Schiechl, O. Dietze, and R. Tubbs. 1997. In situ localization of DNA and RNA sequences: supersensitive in situ hybridization using streptavidinNanogold-silver staining: minireview, protocols, and possible applications. Cell Vision 4:54-65. 13.Hainfeld, J.F. 1990. STEM analysis of Janssen Auroprobe One, p. 954. In G.W. Bailey (Ed.), Proc. XII Int. Cong. Elec. Microsc. San Francisco Press, San Francisco. 14.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40:177184. 15.Hainfeld, J.F. and F.R. Furuya. 1995. Silver enhancement of Nanogold and undecagold, p. 71-96. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 16.Hainfeld, J.F., F.R. Furuya, and R.D. Powell. 1999. Metallosomes. J. Struct. Biol. 127:152-160. 17.Hainfeld, J.F. and R.D. Powell. 1997. Nanogold technology: new frontiers in gold labeling. Cell Vision 4:408-432. 18.Hainfeld, J.F., R.D. Powell, J.K. Stein, G.W. Hacker, C. Hauser-Kronberger, A.L.M. Cheung, and C. Schöfer. 1997. Gold-based autometallography, p. 486487. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mansfield, and R.L. Price (Eds.), Proc. 57th Ann. Mtg., Micros. Soc. Amer. Springer-Verlag, New York. 19.Kramarcy, N.R. and R. Sealock. 1990. Commercial preparations of colloidal-gold-antibody complexes frequently contain free active antibody. J. Histochem. Cytochem. 39:37-39. 20.Michel, H. 1991. Crysallization of Membrane Proteins. CRC Press, Boca Raton. 21.Powell, R.D., C.M.R. Halsey, and J.F. Hainfeld. 1998. Combined fluorescent and gold immunoprobes: reagents and methods for correlative light and electron microscopy. Micros. Res. Tech. 42:2-12.

117

Gold and Silver Staining 22.Powell, R.D., C.M.R. Halsey, W. Liu, V.N. Joshi, and J.F. Hainfeld. 1999. Giant platinum clusters: 2 nm covalent metal cluster labels. J. Struct. Biol. 127:177184. 23.Powell, R.D., C.M.R. Halsey, D.L. Spector, S.L. Kaurin, J. McCann, and J.F. Hainfeld. 1997. A covalent fluorescent-gold immunoprobe: simultaneous detection of a pre-mRNA splicing factor by light and electron microscopy. J. Histochem. Cytochem. 45:947956. 24.Powell, R.D., V.N. Joshi, C.M.R. Halsey, J.F. Hainfeld, G.W. Hacker, C. Hauser-Kronberger, W.H. Muss, and P.M. Takvorian. 1999. Combined Cy3Nanogold conjugates for immunocytochemistry and in situ hybridization, p. 1324-1325. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mansfield, and R.L. Price (Eds.), Proc. 57th Ann. Mtg., Micros. Soc. Amer. Springer-Verlag, New York. 25.Reynolds, E.C. 1963. The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J. Cell Biol. 17:208-213. 26.Robinson, J.M. and D.D. Vandré. 1997. Efficient immunocytochemical labeling of leukocyte microtubules with FluoroNanogold: an important tool for correlative microscopy. J. Histochem. Cytochem. 45:631-642. 27.Segond von Banchet, G. and B. Heppelmann. 1995. Non-radioactive localization of supstance P binding

118

site in rat brain and spinal cord using peptides labeled with 1.4-nm gold particles. J. Histochem. Cytochem. 43:821-827. 28.Sun, X.J., L.P. Tolbert, and J.G. Hildebrand. 1995. Using laser scanning confocal microscopy as a guide for electron microscopic study: a simple method for correlation of light and electron microscopy. J. Histochem. Cytochem. 43:329-335. 29.Takizawa, T. and J.M. Robinson. 1994. Use of 1.4nm immunogold particles for immunocytochemistry on ultra-thin cryosections. J. Histochem. Cytochem. 42:1615-1623. 30.Takizawa, T., K. Suzuki, and J.M. Robinson. 1998. Correlative microscopy using FluoroNanogold on ultrathin cryosections: proof of principle. J. Histochem. Cytochem. 46:1097-1102. 31.Vandré, D.D. and R.W. Burry. 1992. Immunoelectron microscopic localization of phosphoproteins associated with the mitotic spindle. J. Histochem. Cytochem. 40:1837-1847. 32.Zehbe, I., G.W. Hacker, H. Su, C. Hauser-Kronberger, J.F. Hainfeld, and R. Tubbs. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, streptavidin-Nanogold, and silver acetate autometallography. Detection of single-copy human papilloma virus. Am. J. Pathol. 150:1553-1561.

8

Detection of Human Papillomavirus by In Situ Hybridization Using an In Situ PCR Thermal Cycler Peter Jackson and Fraser A. Lewis

INTRODUCTION A major problem confronting researchers attempting to carry out in situ hybridization reactions is the lack of instrumentation dedicated to in situ work. Problems arise when attempting to contain the reaction components on the section during the hybridization process. Attempts to solve this problem using coverslips and a variety of sealing components can lead to failure. This is usually due to failure of the sealing components, allowing the reaction mixture to leak out from under the coverslip, and the subsequent drying of the specimen. This problem has been addressed by the by the introduction of the GeneAmp in situ PCR System 1000 (PE Biosystems, Foster City, CA, USA). Although the instrument is designed for in situ polymerase chain reaction (PCR) use, the system has been successfully adapted for use in in situ hybridization reactions. The instrument is a thermal cycler that can cycle 30 sections (10 slides with 3 sections per slide) at a time. The slides are held firmly against a series of vertical slots comprising the block of the thermal cycler to

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

ensure precise temperature control and consistency. The system also provides a unique containment system for localizing reaction components on the section. An assembly tool (Figure 8.1) is used to assemble AmpliCover discs and clips (both from PE Biosystems) onto 1.2-mm silanecoated slides to create a small air bubblefree standardized reaction chamber over the section and hybridization reaction mixture. The AmpliCover clips hold the discs firmly in place and remain sealed throughout the hybridization process. PROTOCOLS DNA Probe Human papilloma virus 6b probe (H. Zur Hausen, Heidelberg, Germany). Placental DNA (Oncor, Gaithersburg, MD, USA) was used as a positive control probe, and plasmid pBR 322 (Life Technologies, Gaithersburg, MD, USA) was used as a negative control. The probes were biotinylated by nick translation with biotin-11deoxyuridine triphosphate.

119

Gold and Silver Staining Nick Translation of DNA Probe Materials and Reagents • Solution A: 0.2 mmoL each of dATP, dCTP, and dGTP in 500 mmol/L Tris-HCl, pH 7.8, 50 mmol/L MgCl2, 100 mmol/l/L 2-mercaptoethanol, 100 µg nuclease-free bovine serum albumin (BSA). • 1 µg purified recombinant plasmid DNA in water. • Biotin-11-dUTP in 100 mmol/L TrisHCl, pH 7.5, 1 mmol/L EDTA. • Solution B: 0.4 U/mL DNA poly-

merase 1, 40 pg/µL DNase 1 in 50 mmol/L Tris-HCl, pH 7.5, 5 mmol/L magnesium acetate, 1 mmol/L 2-mercaptoethanol, 0.1 µmol/L phenylmethylsulphonyl fluoride, 50% glycerol, 100 µg/mL BSA. • Stop buffer: 0.3 mol/L EDTA, pH 8.0. • 5% sodium dodecyl sulphate (SDS). Method 1. Into a 1.5-mL microcentrifuge tube sitting on ice, pipet the following: 5 µL

Figure 8.1. Assembly tool of PE Biosystems Thermocycler IS-1000.

120

Detection of Human Papillomavirus Solution A, 1 µg DNA, 2.5 µL biotin11-dUTP, and y µL water to give a final volume of 45 mL. 2. Add 5 µL Solution B and mix gently. Centrifuge briefly to bring liquid to bottom of tube. 3. Incubate at 15°C for 90 min. 4. Add 5 µL stop buffer and 1.25 µL of 5% SDS. 5. Make up to 100 µL with water. Purification of Labeled Probe 1. Plug the bottom of a 1-mL disposable syringe with a small quantity of glass wool and fill with Sephadex G-50 (Amersham Pharmacia Biotech, Piscataway, NJ, USA) in sodium-Tris-acetate (STE) buffer. Insert the syringe into a 15-mL centrifuge tube and centrifuge at 1800 rpm (1200× g) for 4 min. 2. Continue to add more Sephadex suspension and recentrifuge (at 1800 rpm for 4 min each time) until the volume of packed Sephadex is approximately 0.9 mL. 3. Add 100 µL STE to the column and centrifuge at 1800 rpm for 4 min. 4. Decap a 500-µL microcentrifuge tube and attach it to the bottom of the syringe. 5. Dilute the biotinylated probe with 50 µL STE buffer. 6. Apply the 100 µL of probe to the top of the column attached to the microcentrifuge tube and spin at 1800 rpm for 4 min. 7. The eluent collected in the tube contains the biotinylated probe, while the unincorporated nucleotides remain bound to the column. Store the probe at -20°C in 10-µL aliquots. Alternatively, premade columns such as ProbeQuant G-50 (Amersham Pharmacia Biotech) can be used, with the proce-

dure being similar to that described. The biotinylated probes were used at a concentration of 200 ng/mL in a hybridization buffer containing 2× SCC (sodium chloride, sodium citrate), 5% dextran sulphate, 0.2% dried milk powder, and 50% formamide (1× SCC = 0.15 mol/L sodium chloride, 0.015 mol/L sodium citrate). Preparation of Paraffin Sections Five-micrometer-thick sections of formalin-fixed paraffin blocks were cut onto silane-coated slides (PE Biosystems) and hot plated for 24 h at 60°C. Sections were deparaffinized in xylene at 37°C for 30 min and in xylene at room temperature (twice) for 10 min each, then rinsed in 100% ethanol and allowed to air dry. The dried sections can be stored at room temperature until required for hybridization. Pretreatment of Tissue Sections 1. Prepare for the protease digest by placing a plastic staining dish in a precision water bath set at 37°C containing 250 mL proteinase K buffer (50 mmol/L Tris-HCl, pH 7.6, 5 mM EDTA) and 4 µg/mL proteinase K (PE Biosystems). 2. Immerse sections in 0.02 mol/L HCl for 10 min. 3. Wash sections for 2 min in phosphatebuffered saline (PBS). 4. Immerse slides in PBS containing 0.01% Triton X-100 for 3 min. 5. Wash sections for 2 min in PBS. 6. Immerse slides in 50 mM Tris-HCl, 5 mmol/L EDTA at 42°C for 2 min. 7. Transfer slides to prewarmed proteinase K solution and incubate for exactly 30 min. 8. Transfer the sections immediately into 250 mL of 2× SSC prewarmed to 80°C and incubate for 5 min. 121

Gold and Silver Staining 9. Rinse slides in water. 10. Dehydrate the sections through graded alcohols to 100% ethanol over a 10 min period. The slides may be stored in the ethanol at 4°C for several days until required for hybridization.

clip (Figure 8.2) using the assembly tool. Once assembled, transfer the slides to the IS 1000 instrument programmed for a denaturing step (94°C for 10 min) followed by a hybridization step (37°C for 16 h).

Hybridization of Pretreated Slides

Posthybridization Washes

1. Prepare the probe at a concentration of 200 ng/mL in a hybridization buffer containing 2× SSC, 5% dextran sulfate, 0.2% (wt/vol) milk powder, 50% formamide. Make up the probe fresh, as it does not store long term in this buffer. 2. Apply 50 mL of the hybridization mixture to each section and seal in place with an AmpliCover disc and

Figure 8.2. Special glass slides (PE Biosystems) with amplicovers in place.

122

1. Remove the slides from the IS 1000 and remove the clips and discs. 2. Transfer the slides to 2× SSC at room temperature for 5 min. 3. Sequentially wash the slides as follows. a. 2× SSC at 60°C for 10 min. b. 0.2× SSC at room temperature for 2 min.

Detection of Human Papillomavirus c. 0.2× SSC at 42°C for 10 min. d. 0.1× SSC at room temperature for 2 min. e. Rinse in distilled water. Detection of Hybridization Signal 1. Immerse sections in Lugol’s iodine for 2 min. Rinse in water and decolorize in 2.5% sodium thiosulphate solution. Wash sections in water for 2 min.

2. Wash sections in Tris-buffered saline (TBS) pH 7.6, for 2 min. 3. Treat sections with a 1 in 10 dilution of casein (Vector Laboratories, Peterborough, UK). 4. Rinse in TBS, pH 7.6. 5. Apply goat antibiotin labeled with 1 nm colloidal gold (British Biocell International, Cardiff, UK) diluted 1 in 100 in 1 in 50 casein in TBS, pH 7.6 for 1 h. 6. Wash in TBS, pH 7.6, for 2 min.

Figure 8.3. Total DNA probe as positive control in a paraffin section of human anal condyloma. Objective 16×. (Original magnification ×178.)

Figure 8.4. HPV6 in paraffin section of human anal condyloma. Objective 16×. (Original magnification ×178.)

123

Gold and Silver Staining Sensitivity may be increased by using a second incubation of rabbit antigoat 1 nm colloidal gold-labeled antibody for 1 h. 7. Wash sections in TBS, pH 7.6 (2 times for 5 min). 8. Wash sections in distilled water (2 times for 5 min). 9. Silver enhance using the LM/EM Silver Enhancing kit (British Biocell International) using light micro-

scopical control. 10. Wash sections in distilled water (2 times for 5 min). 11. Fix sections in 2.5% sodium thiosulphate solution for 2 min. 12. Wash sections in tap water for 2 min. 13. Counterstain as desired. 14. Dehydrate through alcohol, clear in xylene, and mount (Figures 8.3 and 8.4).

Figure 8.5. Epipolarization microscopy of total DNA probe in paraffin section of human anal condyloma. Objective 16×. (Original magnification ×178.)

Figure 8.6. Epipolarization microscopy of HPV6 probe in paraffin section of human condyloma. Objective 16×. (Original magnification ×178.)

124

Detection of Human Papillomavirus DISCUSSION The in situ hybridization technique described and its immunogold detection system are a modification of the procedure previously described by Lewis et al.5 and Jackson et al.4 However, the immunogold–silver staining (IGSS) method2 produces a signal that is insoluble in commonly used laboratory dehydrating and clearing reagents, is visible with low power light microscopy, and which can be enhanced by epipolarization.3 There are several advantages associated with the use of epipolarization microscopy over the use of histochemical reactions involving the use of enzymes to produce a colored end product. The gold–silver complex emits light which is easy to perceive. The signal is stable, and archival preparations can be reexamined at will (Figures 8.5 and 6). The counterstain (cell morphology) and cytochemical reactions can be appreciated simultaneously. The increased visibility of the reaction product ensures that a low signal is easily seen. The inclusion of the Lugol’s iodine followed by decolorization in sodium thiosulphate is an essential step in the technique. Lugol’s iodine is usually employed in histopathology to remove mercury pigment from sections fixed in a mercury-containing fixative. The inclusion of this step in the technique for formalin-fixed sections has been found to be of the utmost importance, although the rationale is not fully understood. The use of casein as a protein block has been found to greatly minimize nonspecific antibody binding and eliminates the need to use animal serum, which traditionally has been used for this purpose. Immunoglobin-labeled with 5 nm colloidal gold particles may be regarded as protein-coated gold particles. Reduction of the gold marker particle size to 1 nm results in a reduction of the overall probe

size, greater penetrating ability, and less steric hindrance. The smaller the gold size the more silver-enhanceable gold label reaches the antigenic site. Immunogold reagents of 1 nm have been manufactured so that the ratio between probe antibody and gold particle is smaller than or equal to 1. Each probe antibody has at least one 1 nm gold particle absorbed to it and may be likened to fluorescence-labeled antibodies. The use of light sensitive autometallography solutions to silver enhance colloidal gold particles, as originally described by Danscher in 1981,1 necessitated the use of a photographic room and safe lights. In the protocol presented above, silver enhancement was carried out using an LM/EM Silver Enhancing kit. The use of the light-stable silver enhancing kit dispenses with the need to carry out the silver enhancement process using photographic dark room conditions. Silver enhancement was carried out under normal laboratory lighting conditions and at room temperature. After 5 min, the slides were washed in distilled water and examined by light microscopy. If the probe was found to be suboptimally demonstrated, then the silver enhancer was reapplied. Sections were checked in this way at 1-min intervals until the probes were clearly visible under the light microscope. Overintensification results in an unacceptable degree of background staining. After washing and fixing the silver deposit in sodium thiosulphate, the label is stable and insoluble in alcohol and xylene, and sections can be counterstained, dehydrated, and cleared. Alternative light-insensitive silver enhancement procedures are described in Chapters 2 and 3 of this volume. REFERENCES 1.Danscher, G. 1981. Localization of gold biological tissue. A photochemical method for light and electron microscopy. Histochemistry 71:81-88. 2.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird.

125

Gold and Silver Staining 1983. Immunogold–silver staining: new method of immunostaining with advanced sensitivity. J. Histochem. Cytochem. 31:938-994. 3.Jackson, P., F.A. Lewis, and M. Wells. 1989. In situ hybridization technique using an immunogold–silver staining system. Histochem. J. 21:425-428. 4.Jackson, P., D.A. Dockey, F.A. Lewis, and M. Wells. 1990. Application of 1 nm gold probes on paraffin wax

126

sections for in situ hybridization histochemistry. J. Clin. Pathol. 43:810-812. 5.Lewis, F.A., S. Griffiths, R. Dunnicliffe, M. Wells, N. Dudding, and C.C. Bird. 1987. Sensitive in situ hybridization technique using biotin-streptavidinpolyalkaline phosphatase complex. J Clin. Pathol. 40:163-166.

9

Supersensitive In Situ Hybridization by Tyramide Signal Amplification and Nanogold® Silver Staining: The Contribution of Autometallography and Catalyzed Reporter Deposition to the Rejuvenation of In Situ Hybridization Raymond R. Tubbs, James Pettay, Thomas Grogan, Annie L.M. Cheung, Richard D. Powell, James Hainfeld, Cornelia Hauser-Kronberger, and Gerhard W. Hacker

INTRODUCTION It is peculiar that in situ hybridization (ISH), a technique with many similarities to immunohistochemistry (IHC), has not enjoyed the phenomenal growth in both basic research and clinical applications as has its sister technique IHC. Since the late 1970s, when immunoperoxidase techniques began to be applied to routine diagnostic material and to numerous research applications, there has been a natural evolution of the IHC procedure. Namely, only a few primary antibodies were available commercially at the onset, and only one indirect and the peroxidase-antiperoxidase

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

(PAP) technique detection systems were in place. With the advent of avidin–biotin detection systems and monoclonal antibodies, and a viable commercial market, extraordinary growth of the procedure’s applications in clinical research and diagnostic pathology occurred during the subsequent two decades. Today, IHC is automated and widely used for research purposes and, to a large extent, has become a routine diagnostic “special stain” in most clinical laboratories. During the same period, ISH enjoyed very little growth in both research and diagnostic applications. What has accounted for this lack of maturation of the technique?

127

Gold and Silver Staining The success of IHC is part of the reason measuring a gene’s encoded protein routinely and inexpensively, particularly as automation evolved, rendered IHC a more viable choice in many instances. Inherent comparative sensitivity of the procedures has also clearly been a factor. Unfortunately, the chromogenic procedures in place are often insufficiently sensitive to detect the relatively low amounts of DNA and RNA levels at which the clinical utility is to be found. But ISH is enjoying a renaissance, as reflected in Protocols 1 through 6 and Figures 9.1 through 9.4. There are multiple reasons for this rejuvenation. First, great advances have been achieved through enhanced reporter systems. Detection systems, now of either fluorescence, chromogenic, or autometallographic type, can utilize variations of tyramide signal amplification (TSA).2,8,10,14–16,19,20,25 Also known as catalyzed reporter deposition (CARD), TSA exploits the catalytic action of peroxidase on tyramide conjugates. Single copies of the human papillomavirus (HPV) can now be routinely detected in human cells with this technology25 Advances in cell conditioning have also greatly enhanced the practical sensitivity of ISH, and when used rationally in combination with TSA, tremendous analytical strength is achieved. The detection of individual or amplified endogenous genes such as the oncogene Her-2/neu, with proper cell conditioning and signal amplification, can be done by either fluorescent or brightfield methods. Autometallography itself also provides for enhanced signal intensity with clean background, particularly when Nanogold covalent particles are used.413,17,18,21,22 Perhaps the greatest boon to the renaissance of ISH will be the identification of truly remarkable new applications for the technology, such as routine brightfield in situ target detection with morphologic correlation, or unique combinations of fluorescence, autometallography, and 128

chromogenic preparations.24 With the advent of automation, brightfield ISH applications can potentially be combined with IHC, whereby endogenous and the gene’s encoded protein can be simultaneously visualized. One such technique has recently been developed (see Protocol 5 and Figure 9.3) and was named CODFISH (concomitant oncoprotein detection with fluorescence in situ hybridization [FISH]). Autometallography and subtracted unique sequence yeast artificial chromosome (YAC) probe-based ISH, may also potentially be combined with IHC. Finally, given the explosive growth of knowledge of the human genome, largescale screening via tissue microarrays and the advent of new probe technologies such

Figure 9.1. Squamous cell carcinoma of the uterine cervix with genome-integrated HPV 16/18 copies. Nearly every tumor cell nucleus shows one black dot, very likely representing a single HPV 16/18 copy detected as in Protocol 2.

Super Sensitive In Situ Hybridization as subtractive unique sequence ISH probes derived from bacterial artificial chromosomes (BACs) or YACs dramatically augment signal intensity by virtue of probe size, sequence specificity, and subtraction of cross-hybridization polymorphic loci.1,3 Even chromosomal deletions can be profiled in interphase FISH or chromogenic in situ hybridization (CISH) format, thereby opening even greater vistas for the revised and rejuvenated ISH procedure. REAGENTS, SUPPLIES, AND INSTRUMENTATION The majority of reagents necessary to perform ultrasensitive autometallographic ISH are available commercially. In fact, patents are held for some of the key components of the technology, precluding at least in house “home brew” format synthe-

sis of the reagents, at least if clinical applications are anticipated. Reagents for autometallography are available from Nanoprobes (Yaphank, NY, USA) (http://www.nanoprobes.com/), and it is also possible to use alternative silver enhancers such as silver acetate autometallography.9 For procedures that depend upon tyramide-based amplification, the basic patent is owned by NEN Life Science Products (Boston, MA, USA) as the renaissance kit. Manufacturers such as DAKO (Carpinteria, CA, USA) have adapted the tyramide technology to a stable kit format for detection of DNA targets, which is available as the GenPoint kit. All the other reagents are available from standard chemical manufacturers and are mainstays of the typical molecular pathology laboratory. These include phosphate-buffered saline (PBS), standard sodium citrate buffer (SSC), graded alcohols,

Figure 9.2. Hodgkin lymphoma mononuclear cell variants demonstrating Epstein-Barr Virus associated RNA (EBER). Cells containing EBER RNA of EBV are labeled in black, using the procedure described in Protocol 3.

129

Gold and Silver Staining organic solvents, and similar reagents. A variety of cell conditioning approaches have been advocated in the literature. In general, these approaches utilize heat and a variety of buffer solutions or chelating agents. Steamers, pressure cookers, water baths, and microwave treatments have all been advocated. As a general rule, each ISH system will usually have its own peculiar requirements for cell conditioning (heated epitope retrieval, antigen recovery, etc.), and emperical approaches using combinations of buffers and duration of heat are usually necessary. More recently, an automated instrument has become available (Discovery and Benchmark; Ventana Medical Systems, Tucson, AZ, USA) for which cell conditioning prior to ISH are performed on-line by the instrument. This approach includes removal of paraffin and cell conditioning in the presence of proprietary buffers. Consistency of use of this instrument should have a very beneficial effect on reproducible performance of ISH techniques. Probes for ISH may consist of genomic DNA probes, ribroprobes, or oligonucleotide probes. Many excellent reagents are available from commercial sources, but much of what is currently performed is based upon home brew probe reagents for both clinical research and some clinical applications. Whether of commercial or home brew origin, the performance of each probe, and the detection system utilized, must be carefully validated. Guidelines for such validation have been previously published. Multiple detection systems for fluorescent and chromogenic–autometallographic assays are available. For example, direct conjugates for FISH analysis, including up to 4color direct conjugate systems, are available from Vysis (Downers Grove, IL, USA). Some preliminary work has been done in the commercial sector to develop chromogenic ISH probes, and these are available commercially from several sources (DAKO, 130

Zymed Laboratories [South San Francisco, CA, USA], Ventana Medical Systems, Vector Laboratories [Burlingame, CA, USA], Research Genetics [Huntsville, AL, USA], Enzo Diagnostics [Farmingdale, NY, USA], and others). Once again, it must be emphasized that each probe, and the detection used for that system, must be carefully validated using published guidelines. Stability of biotin–tyramide conjugates is a serious issue, but has largely been achieved by its two manufacturers, NEN Life Science Products and DAKO. In particular, the DAKO GenPoint kit has achieved a remarkably stable tyramide reagent supported by that system, and in combination with ultrapure streptavidin–peroxidase, provides for excellent and reproducible results. Other immunogenic compounds can be labeled with tyramide including fluorescein isothio-

Figure 9.3. (See color insert Figure 9.3 following page 78.) Fluorescence confocal photomicrograph of CODFISH preparation. Paraffin section of cell line known to display amplification of the Her-2/neu gene (green) and its encoded protein (red). Amplification–overexpression is identified (amplified gene copy is green, increased Her-2/neu oncoprotein is red).

Super Sensitive In Situ Hybridization

A

B

Figure 9.4. GOLDSUISH preparations. Autometallography using subtractive unique sequence ISH identifies 1 or 2 copies of Her-2/neu in a subpopulation of normal truncated nuclei (A; arrows) and overexpression–overamplification in an invasive ductal carcinoma (B).

131

Gold and Silver Staining cyanate (FITC), digoxigenin, and dinitrophenol (DNP). Also, the nucleotides can be directly labeled with peroxidase. Several commercial sources will provide custom conjugation services for these haptens. PROTOCOLS Note: The protocols included in this chapter undergo frequent refinement, and new similar techniques are added at the following web site: (http://www.sbg.ac.at/ kgg/protocols/protocols.htm). Protocol 1. DNA In Situ Hybridization with Streptavidin–Nanogold Background and Purpose This protocol allows a reliable and very sensitive detection of a few copies of HPV, cytomegalovirus (CMV), Epstein Barr virus (EBV), herpes simplex virus (HSV), and other DNA viruses in routinely formalin-fixed and paraffin-embedded tissues. The method described below has great potential as a robust and fast routine method for diagnostic purposes in all cases where conventional ISH is not sensitive enough, and where actual single-copy sensitivity is not needed. The protocol also works well with FluoroNanogold (Nanoprobes), thus allowing visualization of the very same preparation by fluorescent microscopy (FISH) and transmitted light microscopy. Nanogold preparations can be transferred to the electron microscope.21 Nanogold was developed by Dr. James F. Hainfeld17 and is available from and patented by Nanoprobes. Solutions • PBS: 10× PBS (Mg2+- and Ca2+-free), pH 7.6: 11.36 g Na2HPO4, 2.72 g 132

KH2PO4, 87.0 g NaCl in 800 mL distilled water. Adjust pH with concentrated NaOH and add distilled water to a final volume of 1 L. • SSC: 175.32 g NaCl and 88.23 g sodium citrate in 800 mL distilled water. Adjust pH with NaOH to 7.0 and add distilled water to a final volume of 1 L. Procedure 1. Deparaffinize sections from formaldehyde-fixed tissue in fresh xylene (2 times for 15 min each). 2. Rinse and rehydrate in graded alcohols and distilled water (2–3 min each). 3. Soak in PBS (20 mM, pH 7.6) for 3 min. 4. Incubate with 0.1 mg/mL proteinase K (Code No. 1373196; Roche Molecular Biochemicals, Mannheim, Germany) in PBS at 37°C for about 8 min. The duration is critical and has to be tested very carefully, depending on tissue, fixation, and other factors. 5. Rinse in 2 changes of PBS for 3 min. 6. Permeabilize with 0.3% Triton X-100 in PBS for 15 min. 7. Wash in PBS for 2 min. 8. Rinse in 2 changes of distilled water, dehydrate with graded alcohols (50%, 70%, and 98% isopropanol) for 1 min each and air-dry the sections. 9. Prehybridize with 1:1 mixture of deionized formamide and 20% dextran sulfate in 2× SSC at 50°C for 5 min. 10. Carefully shake off excess prehybridization block. 11. Add one drop of biotinylated DNA probe on the section and cover with a small coverslip. Avoid air bubbles. 12. Heat sections on heating block at 92° to 94°C for 8 to 10 min to denature DNA. 13. Incubate in a moist chamber at 37°C

Super Sensitive In Situ Hybridization overnight (or for at least 2 h). 14. Posthybridization washes (5 min each): 2 changes of 4× SSC (first wash to remove coverslips), 2× SSC, 0.1× SSC, 0.05× SSC, and then distilled water. 15. Put slides into Lugol’s iodine solution (Merck, Darmstadt, Germany) for 5 min. 16. Wash in tap water and then distilled water. 17. Put into 2.5% sodium thiosulfate for a few seconds until sections are colorless. Then wash in tap water for 5 min and distilled water for 2 min. 18. Immerse in PBS containing 0.1% fish gelatin (45% concentrate; Cat. No. G7765; Sigma-Aldrich, Steinheim, Germany) and 0.1% Tween 20 for 5 min. 19. Incubate sections with streptavidin– Nanogold (Nanoprobes) diluted 1:200 to 1:500 in PBS containing 1% bovine serum albumin (BSA) at room temperature for 60 min. 20. Wash in 3 changes of PBS containing 0.1% fish gelatin and 0.1% Tween 20 for 5 min each. 21. Repeatedly wash in distilled water for at least 10 min altogether, the last 2 rinses in ultrapure water (electron microscopy [EM]-grade). 22. Perform silver acetate autometallography or GoldEnhance development (Nanoprobes). The procedure for silver acetate is given here: a. Solutions A and B should be freshly prepared for every run. Solution A: dissolve 80 mg silver acetate (Code No. 85140; Fluka Chemical, Buchs, Switzerland) in 40 mL of glass double-distilled water. Silver acetate crystals can be dissolved by continuous stirring within about 15 min. b. Citrate buffer: dissolve 23.5 g of trisodium citrate dihydrate and

25.5 citric acid monohydrate in 850 mL or deionized or distilled water. This buffer can be kept at 4°C for at least 2 to 3 weeks. Before use, adjust to pH 3.8 with citric acid solution. c. Solution B: dissolve 200 mg hydroquinone in 40 mL citrate buffer. d. Enhancement solution: just before use, mix solution A with solution B. e. Silver amplification: place the slides vertically in a glass container (preferably with about 80 mL volume and up to 19 slides; Schiefferdecker-type) and cover them with the mixture of solutions A and B. Staining intensity can be checked in the light microscope during the amplification process, which usually takes about 5 to 20 min depending on primary antibody or nucleic acid probe concentration, incubation conditions, and the amount of accessible antigen or nucleic acid sequence in question. f. Photographic fixer has been used in combination with colloidal gold to stop the enhancement process immediately. Alternatively, a 2.5% aqueous solution of sodium thiosulfate can be used. Urgent: It has, however, turned out that both these fixing treatments can be harmful when working with labeled tyramides or with Nanogold. We therefore now recommend to stop enhancement simply by washing in distilled water. g. After stopping the enhancement process, slides can be examined in a light microscope more carefully. If staining intensity is still too low, wash slides for one more time in double-distilled water and develop further in enhancement solution. 133

Gold and Silver Staining h. After washing in distilled water, sections can be counterstained with nuclear fast red, or hematoxilin, and/or eosin. i. After washing, dehydration in graded alcohols, and clearing in xylene, Permount (Fisher Scientific, Pittsburgh, PA, USA) are the preferred mounting media (see companion chapters in this book, especially Chapter 1). Protocol 2: DNA In Situ Hybridization with Labeled Tyramides and Streptavidin–Nanogold Background and Purpose This method is based on the superb properties of Nanogold and of the catalyzed reporter deposition–tyramide signal amplification system. It allows a supersensitive and relatively reliable detection of even single copies of HPV in routinely formaldehyde-fixed and paraffin-embedded tissue specimens, as well as on formalin-fixed cytological preparations. Other DNA viruses, as well as mRNA and other RNA stainings, have been tested too (see Protocol 3). The protocol given here was developed for formalin-fixed paraffin sections glued onto silanized glass slides. Cytocentrifuge preparations can also be used with the following procedure, but steps 1 and 2 should be eliminated (start at step 3). Cytospins should be air-dried prior to fixation in either neutral-buffered formalin or absolute ethanol. It must be mentioned that this method is not yet completely problem-free. We experienced repeatedly that the outcome strongly depends on the individual tissue and fixation conditions applied, as well as on the quality and type of hybridization probe used. As the method is extremely sensitive, this also means that even minor 134

tissue fixation problems or minor probe and reagent quality differences yield massive background staining, thereby sometimes masking structures to be diagnosed. Quite often, intra-nuclear staining of fibroblasts has also been noted. The protocol is therefore given as a guideline for one’s own experiments. Concerning reporter molecules, biotin-labeled cDNA probes (Enzo Diagnostics) often gave excellent results when demonstrating HPVs. The quality of the probe for hybridization is of enormous relevance here; small impurities can lead to massive background staining. Riboprobes labeled with FITC have also given good staining in our system (see Protocol 3). It has to be mentioned that we were not yet successful to detect digoxigenin-labeled probes with the tyramide system; one possible explanation may be steric hindrance. Solutions CARD, developed by Dr. Mark Bobrow, is patented by Perkin Elmer Life Sciences under the term TSA (http:// www.lifesciences.perkinelmer.com). We have successfully used biotinylated tyramides (BTs) from the TSA-Indirect kit for ISH. A licensed tyramide product that worked well in our tests is available from DAKO, contained in the GenPoint In Situ kit (Cat. No. K0620). Procedure 1. Deparaffinize sections from formaldehyde-fixed tissue in fresh xylene (2 times for 15 min each). 2. Rinse in absolute ethanol (2 times for 5 min each). 3. Treat with 3% H2O2 in methanol at room temperature for 30 min. 4. Rinse in double-distilled (ultrapure) water for 10 sec and then in PBS for 3 min.

Super Sensitive In Situ Hybridization 5. Incubate sections with 0.1 mg/mL proteinase K in PBS at 37°C for about 8 min (optimal duration should be tested). This treatment may partly destroy tissue morphology. However, it is necessary to open up binding sites for the probe to reach full sensitivity. Combination with microwave treatment may be feasible. 6. Wash in 2 changes of PBS for 3 min each, then wash with ultrapure water for 10 sec. 7. Dehydrate with graded alcohols (50%, 70%, absolute ethanol) for 5 min each and air-dry the sections. 8. Prehybridize with 1:1 mixture of deionized formamide and 20% dextran sulfate in 2× SSC at 50°C for 5 min. 9. Carefully shake off the excess prehybridization block. 10. Add one drop of biotinylated DNA probe on the section and cover with a small coverslip. Avoid air bubbles. 11. Heat sections on heating block at 92° to 94°C for 8 to 10 min to denature DNA. 12. Incubate in a moist chamber at 37°C overnight (or for at least 2 h). 13. Posthybridization washes (5 min each): 2 changes of 2× SSC (first wash to remove coverslips), 0.5× SSC, 0.2× SSC, and then distilled water. 14. Put slides into Lugol’s iodine solution for 5 min. 15. Wash in tap water and then doubledistilled water. 16. Put into 2.5% sodium thiosulfate for a few seconds until sections are colorless. Then wash in distilled water for 2 min. 17. Drain off section, wipe area around the section dry, and surround it with a DAKO-Pen (Cat. No. S-2002; DAKO). 18. Incubate with blocking solution at 37°C for 30 min. Blocking solution is 4× SSC containing 5% casein sodium salt (Cat.

No. C-8654; Sigma) or 0.5% blocking powder (from Renaissance TSA-indirect ISH kit; Cat. No. NEL730; Perkin Elmer Life Sciences). 19. Briefly wash in 4× SSC containing 0.05% Tween 20 for 2 min. 20. Incubate with streptavidin–biotin–peroxidase complex (e.g., StreptABComplex/HRP Duet kit; Cat. No. K0492; DAKO) at room temperature for 30 min. This complex is dissolved in the above blocking solution at a concentration of 1:200. 21. Wash in 3 changes of 4× SSC containing 0.05% Tween 20 for 2 min each, followed by 2 changes of PBS for 2 min each. 22. Incubate the sections with BT at room temperature for exactly 10 (with BT from Renaissance TSA-indirect ISH kit) or 15 min (with BT from GenPoint In Situ kit). The BT reagent in the GenPoint kit is ready-to-use. For the TSA-indirect kit, a stock solution of BT is prepared by adding 100 mL ethanol to the lyophilized reagent and is diluted 1/50 to 1/100 with the supplied amplification diluent mixed with distilled water at 1:1, as described in the kit. According to the guidelines supplied, the working solution should contain 1 or 0.5 mg of BT per mL diluent, consisting of 0.2 mol/L TrisHCl, 10 mmol/L imidazole, pH 8.8, and 0.01% H2O2. 23. Wash in 4 changes of PBS containing 0.05% Tween 20 and 20% dimethyl sulfoxide (DMSO) at room temperature for 3 min each. 24. Immerse in PBS-gelatin (PBS containing 0.1% fish gelatin) for 5 min. 25. Incubate the sections with streptavidin–Nanogold diluted 1:750 in PBS containing 1% BSA at room temperature for 60 min. 26. Wash in 3 changes of PBS-gelatin for 5 135

Gold and Silver Staining min each. 27. Repeatedly wash in ultrapure water (EM-grade). 28. Perform silver acetate autometallography (see Protocol 1). Note: For all applications where Nanogold is used instead of colloidal gold, it is crucial that slides are not dipped into sodium thiosulfate solution to stop the silver enhancement process. Instead, the development process should be interrupted by simply washing the sections in distilled water (several changes). Sodium thiosulfate would remove the black Nanogold silver–gold staining already obtained. 29. After autometallographic amplification, sections can be counterstained with hematoxylin and eosin and/or nuclear fast red, dehydrated, and mounted in Permount or in DPX. Protocol 3. RNA In Situ Hybridization with Labeled Tyramides and Streptavidin–Nanogold7,23 Practical Considerations The same precautions as given in Protocol 2 have to be taken. As the following protocol is the working procedure for RNA, additional guidelines are necessary. RNA is extremely labile and especially susceptible to destruction by RNase. RNA loss from tissue may occur following delays in primary fixation or also because of production of endonucleases by bacteria-contaminated water baths and other equipment used in preparing tissues. In routine histopathological diagnosis, there is little the molecular pathologist can do to control fixation conditions beyond admonition of his or her colleagues and attention to processing protocols. Great care should be 136

exercised to keep microtome blades and water baths clean and free of bacterial contamination, and sterile water should be used during the hybridization step. Either riboprobes or synthetic oligonucleotides labeled with a reporter molecule such as FITC or biotin may be used. Digoxigenin as the reporter molecule did not work in our tyramide signal-amplified ISH protocols. Hybridization conditions and nucleotide content must be appropriate for RNA rather than DNA hybridization. The protocol given here for FITC as the reporter molecule was developed for formalin-fixed paraffin sections on silanized glass slides. Cytocentrifuge preparations can also be used with the following procedure, but steps 1 and 2 should be eliminated (start at step 3). Cytospins should be air-dried prior to fixation in either neutral-buffered formalin or absolute ethanol. Solutions • PBS: 10× PBS (Mg2+- and Ca2+-free), pH 7.6: 11.36 g Na2HPO4, 2.72 g KH2PO4, 87.0 g NaCl in 800 mL distilled water. Adjust pH with concentrated NaOH and add distilled water to a final volume of 1 L. • SSC: 175.32 g NaCl and 88.23 g sodium citrate in 800 mL distilled water. Adjust pH with NaOH to 7.0 and add distilled water to a final volume of 1 L. Procedure 1. Deparaffinize formaldehyde-fixed sections in fresh xylene (2 times for 15 min each). 2. Rinse in absolute ethanol (2 times for 5 min each). 3. Treat with 3% H2O2 in methanol at room temperature for 30 min. 4. Rinse in double-distilled (ultrapure) water for 10 sec and then in PBS for 3 min.

Super Sensitive In Situ Hybridization 5. Incubate sections with 0.1 mg/mL proteinase K in PBS at 37°C for about 8 min. Optimal concentration and duration is critical and should be tested carefully. This treatment may partly destroy tissue morphology. However, it is necessary to open up binding sites for the probe to reach full sensitivity. Combination with microwave treatment may be feasible. 6. Wash in 2 changes of PBS for 3 min each, then wash with ultrapure water for 10 sec. 7. Dehydrate with graded alcohols (50%, 70%, absolute ethanol) for 5 min each and air-dry the sections. 8. Prehybridize with 1:1 mixture of deionized formamide and 20% dextran sulfate in 2× SSC at 50°C for 5 min. 9. Carefully shake off the excess prehybridization block. 10. Add one drop of FITC haptenated ribonucleotide or antisense oligonucleotide probe on the section and cover with a small coverslip. Avoid air bubbles. 11. Hybridize in a moist chamber at 37°C overnight (or for at least 2 h). 12. Posthybridization washes (5 min each): 2 changes of 2× SSC (first wash to remove coverslips), 0.5× SSC, 0.2× SSC, and then distilled water. 13. Wash in PBS for 3 min. 14. Drain off section, wipe area around the section dry, and surround it with a DAKO-Pen. 15. Incubate with mouse monoclonal antiFITC antibody (Cat. No. BA-9200; Roche Molecular Biochemicals) diluted in PBS at a working dilution of 1:1000 at room temperature for 30 min. 16. Wash in PBS twice for 3 min each. 17. Incubate with biotinylated goat antimouse IgG antibody (Vector Laborato-

ries) diluted in PBS at a working dilution of 1:200 at room temperature for 30 min. 18. Wash in PBS twice for 3 min each. 19. Put slides into Lugol’s iodine solution for 5 min. 20. Wash in tap water and then doubledistilled water. 21. Put into 2.5% sodium thiosulfate for a few seconds until sections are colorless. Then wash in distilled water for 2 min. 22. Incubate with blocking solution at 37°C for 30 min. Blocking solution is 4× SSC containing 5% casein sodium salt or 0.5% blocking powder (from Renaissance TSA-indirect ISH kit). 23. Briefly wash in 4× SSC containing 0.05% Tween 20 for 2 min. 24. Incubate with streptavidin–biotin–peroxidase complex (e.g., from StreptABComplex/HRP Duet kit) at room temperature for 30 min. The complex is dissolved in the above blocking solution at a concentration of 1:200. 25. Wash in 3 changes of 4× SSC containing 0.05% Tween 20 for 2 min each, followed by 2 changes of PBS for 2 min each. 26. Incubate the sections with BT at room temperature for exactly 10 (with BT from Renaissance TSA-indirect ISH kit) or 15 min (with BT from GenPoint in situ kit). For the TSA-indirect kit, a stock solution of BT is prepared by adding 100 mL ethanol to the lyophilized reagent and is diluted 1/50 to 1/100 with the supplied amplification diluent mixed with distilled water at 1:1, as described in the kit. According to the guidelines supplied, the working solution should contain 1 or 0.5 mg of BT per mL diluent, consisting of 0.2 M Tris-HCl, 10 mM imidazole, pH 8.8, and 0.01% H2O2. 27. Wash in 4 changes of PBS containing 0.05% Tween 20 and 20% DMSO at 137

Gold and Silver Staining room temperature for 3 min each. 28. Immerse in PBS-gelatin (PBS containing 0.1% fish gelatin) for 5 min. 29. Incubate the sections with streptavidin–Nanogold diluted 1:750 in PBS containing 1% BSA at room temperature for 60 min. 30. Wash in PBS-gelatin for 5 min (3 times for 5 min each). 31. Repeatedly wash in ultrapure water (EM-grade). 32. Perform silver acetate autometallography (see Protocol 1). Protocol 4. DNA In Situ Hybridization with the GenPoint Kit in Combination with Streptavidin–Nanogold Background and Purpose The method combines the convenience of using a commercially available TSA kit (GenPoint Catalyzed Signal Amplification [CSA] System for In situ Hybridization) with the superior quality of Nanogold silver–gold detection and has been used for biotinylated or FITC-conjugated probes in DNA or RNA ISH. Solutions In addition to some of the solutions given in the above protocols: • Preparation of Tris-buffered saline with Tween (TBST) (10× concentrated): 3.029 g Tris, 17.532 g sodium chloride, 5 mL Tween 20. Make up to 500 mL with double-distilled water and adjust pH to 7.6. Procedure 1. Deparaffinize sections from formaldehyde-fixed tissue in fresh xylene (2 times for 15 min each). 2. Rinse in absolute ethanol (2 times for 138

5 min each), then 95% ethanol (2 times for 5 min each), followed by 2 changes of double-distilled water. 3. Immerse the slides in Target Retrieval solution (Cat. No. S1700; DAKO) at 95°C for 40 min, then let the slides cool in the same solution for 20 min. 4. Rinse the slides in several changes of double-distilled water, then incubate them with proteinase K diluted 1:5000 in 50 mM Tris-HCl buffer (pH 7.6) for 5 min at room temperature. Alternatively, steps 3 and 4 may be replaced by pretreatment with 0.1 mg/mL proteinase K in 50 mM TrisHCl buffer (pH 7.6) at 37°C for about 8 min (optimal duration should be tested carefully). 5. Wash slides in double-distilled water (3 changes for 5 min each). 6. Treat with 3% H2O2 in methanol at room temperature for 30 min. 7. Wash slides in double-distilled water for 10 min. 8. Put slides into Lugol’s iodine solution for 5 min, then wash in double-distilled water. 9. Put into 2.5% sodium thiosulfate for a few seconds until sections are colorless, then wash in double-distilled water (2 times for 5 min each). 10. Allow slides to air-dry. 11. Add one drop of biotinylated DNA probe on the section and cover with a small glass coverslip. Avoid air bubbles. 12. Heat sections on heating block at 92° to 94°C for 5 min to denature DNA. 13. Incubate in a moist chamber at 37°C overnight (or for at least 1 h). 14. Remove the coverslip by soaking slides in a TBST bath for 5 min. 15. Incubate slides in Stringent Wash (provided in the GenPoint kit) for 20 min at 55°C.

Super Sensitive In Situ Hybridization 16. Drain off section, wipe area around the section dry, and surround it with a DAKO-pen. 17. Immerse slides in TBST for 5 min. 18. Apply primary streptavidin–horseradish peroxidase (HRP), diluted 1:800 in the diluent (from GenPoint kit) to sections and incubate in a moist chamber for 15 min at room temperature. 19. Wash in 3 changes of TBST for 5 min each. 20. Apply ready-to-use biotinyl–tyramide solution (from GenPoint kit) and incubate in a moist chamber for 15 min at room temperature. 21. Wash in 3 changes of TBST-gelatin (TBST containing 0.1% fish gelatin, pH 7.6) for 5 min each. 22. Incubate the sections with streptavidin–Nanogold diluted 1:250 in PBS containing 1% BSA at room temperature for 60 min. 23. Wash in 3 changes of TBST-gelatin for 5 min each. 24. Repeatedly wash in ultrapure water (EM-grade). 25. Perform autometallography (See Protocol 1). 26. After autometallographic amplification, sections can be counterstained with hematoxylin and eosin and/or nuclear fast red, dehydrated, and mounted in Permount or in DPX. Protocol 5. CODFISH Background and Purpose This protocol was developed as a response to the documented problems correlating Her-2/neu overexpressions–amplifications defined by IHC and FISH (24). CODFISH permits the simultaneous detection of cell membrane associated Her-2/neu oncoprotein (by IHC) and gene copy enu-

meration (by DNA FISH). The assay is especially suitable for an algorythm whereby brightfield IHC is used for screening. About 70% of breast carcinoma cases will be overtly negative for Her-2/neu by IHC, so FISH is done for only 20% to 30% of cases which show 2+ or 3+ IHC staining, the groups presently requiring FISH for gene copy quantitation. The CODFISH procedure is also provided as an illustration of future possible gold–silver-based systems using the same principles. At this point, the detection system for CODFISH is based on fluorescence. But eventual extension to a combination of autometallography and brightfield microscopy are clearly possible and are currently being pursued by our laboratory as well as others. The CODFISH approach combines gene enumeration by FISH with semiquantitative assessment of corresponding oncoprotein overexpression with alkaline phosphatase-based immunohistochemistry, exploiting the unique bifunctional properties of fast red K. This chromogen provides a bright pink-red color reaction by conventional optical microscopy, and when viewed with a rhodamine excitation-range wavelength reveals a brilliant red nonquenching flourescence. A combination of FITC and fast red K in the flourescence mode provides excellent visualization of the encoded protein and amplification of the gene (Figure 9.3). This approach may have particular value in an algorhythmic system where false positives may complicate interpretation. This appears to be especially true for the Her-2/neu oncogene system. IHC false positives are a serious complication of the laboratory assessment for eligibility for Herceptin (Trastuzumab; Genentech, San Francisco, CA, USA) therapy. Even greater utility will be achieved when this system becomes adaptable to complete bright field visualization. 139

Gold and Silver Staining Solutions • Target retrieval solution. • CB11 monoclonal antibody to Her-2/neu (Ventana Medical Systems/ Vector Laboratories). • Red alkaline phosphatase detection kit (Ventana Medical Systems). • Her-2/neu Dig FISH kit (Ventana Medical Systems). • Wash solutions. Procedure 1. Unstained paraffin sections on electrostatically charged slides are deparaffinized and rehydrated in 3 changes each of xylene, absolute alcohol, and 95% and 80% alcohol, and then placed in 1× PBS for 5 min. 2. Cell conditioning is achieved through the use of microwaving in citrate buffer for 15 min, followed by cooling of the slides in the citrate solution at room temperature for 20 min. 3. The monoclonal antibody CB11 at manufacturer’s concentration is dispensed onto the sections and allowed to incubate for 1 h at room temperature. 4. Sections are then washed in 1× PBS, 2 changes of 3 min each, and biotinylated antimouse IgG (Ventana Medical Systems) applied at a concentration of 15 µg/mL for 30 min. 5. Two changes of 1× PBS wash at room temperature are followed by the addition of avidin–alkaline phosphatase (Ventana Medical Systems) at a concentration of 3000 µg/mL for 30 min. 6. Sections are then washed twice for 3 min each, in 1× PBS at room temperature, and then the chromogen reaction product developed from the fast red K proprietary solution (Ventana Medical Systems), 10 min at room temperature. 140

7. Sections are then washed briefly in distilled water and counterstained with hemotioxylin for approximately 30 sec. Sections are then dehydrated in graded alcohols. 8. Ten microliters of digoxigenin-labeled probe in proprietary probe solution (Oncor [Gaithersburg, MD, USA]/Ventana Medical Systems) are applied to the section and covered with a coverslip. 9. The probe solution and target tissue are codenatured at 90°C for 6 min followed by overnight hybridization at 37°C in a humidified chamber. 10. The coverslip is removed by soaking the slides in 2× SSC for 5 min at room temperature. 11. Washes of 0.5× SSC for 5 min at 72°C follow. 12. Slides are prewashed in 1× PBS containing 0.5% Tween 20 for 3 min at room temperature. 13. FITC-antidigoxigenin 1:50 in 1× PBS, pH 7.6, containing 0.5% BSA is applied to the section and allowed to incubate for 1 h. 14. The sections are then washed 3 times for 5 min each in 1× PBS containing 0.5% Tween 20. 15. The sections are then counterstained with 20 µL 4′6-diamidine-2-phenylindole (DAPI) in antifade solution (Oncor) and mounted for FISH analysis. 16. The fast red K bright red chromogenic reaction product of alkaline phosphatase also displays brilliant pink-red fluorescence on excitation using rhodamine filters. The CODFISH Her2/neu copy number in 20 cells is counted in two representative fields and averaged. Note: Digoxigenin as the reporter molecule may present problems when used with colloidal or clustered gold. Instead of

Super Sensitive In Situ Hybridization digoxigenin, it is better to use biotin, FITC, or other molecules in such combinations.

system whereby protein and gene copy amplification can be simultaneously assessed with the CODFISH method.

Protocol 6. GOLDFISH: Autometallographic Subtractive Unique Sequence In Situ Hybridization

Solutions

Background and Purpose The GOLDFISH protocol was developed as a response to the documented problems correlating Her-2/neu overexpressions–amplifications defined by IHC and FISH. GOLDFISH offers a potential advantage over CODFISH and simple IHC for Her-2/neu protein detection. The evaluation is all done by conventional brightfield microscopy; no special equipment is required. Also, there is potential for a brightfield 2-color systems analogous to CODFISH, by the addition of alkaline phosphatase-based IHC employing fast red K as the choromogen with or without FluoroNanogold (Nanoprobes). Such 2color brightfield applications are currently being developed in our and others’ laboratories. This procedure outlines the system whereby single copies of the endogenous oncogene Her-2/neu can be detected by conventional optical microscopy. Autometallography is especially useful in this instance, since the very discrete gold–silver detection products are sharply defined as compared to enzymatic chromogenic peroxide and alkaline phosphate detection systems. Our initial experiment focused upon the use of tyramide conjugates for this purpose. Unfortunately, the system was not readily adaptable for this particular gene, since unexpectedly high levels of background and excessive confluence of detection products were observed. Current investigation is focused upon combining this autometallographic procedure with IHC in a 2-color

• Target retrieval solution. • Her-2/neu Dig FISH kit (Zymed Laboratories, South San Francisco, CA, USA). • PBS wash solution. • Goat antimouse IgG–Nanogold. • Autometallography solutions (Nanoprobes or as in Protocol 1). Protocol 1. Unstained paraffin sections on electrostatically changed slides are deparaffinized and rehydrated in 3 changes each of xylene, absolute alcohol, and 95% alcohol and cell conditioning achieved through the use of microwaving in citrate buffer for 15 min, followed by cooling of the slides in the citrate solution at room temperature for 20 min. 2. Sections are then washed briefly in distilled water and then dehydrated in graded alcohols. 3. Ten microliters of digoxigenin-labeled subtracted unique sequence YACderived megabase probe in proprietary probe solution (Zymed) is applied to the section and covered with a coverslip. 4. The probe solution and target tissue are codenatured at 90°C for 6 min followed by overnight hybridization at 37°C in a humidified chamber. 5. The coverslip is removed by soaking the slides in 2× SSC for 5 min at room temperature. 6. Washes of 0.5× SSC for 5 min at 72°C follow. 7. Slides are prewashed in 1× PBS con141

Gold and Silver Staining taining 0.5% Tween 20 for 3 min at room temperature. 8. FITC-antidigoxigenin 1:50 in 1× BCS containing 0.5% BSA is applied to the section, per slide, and allowed to incubate for 1 h. 9. The sections are then washed 3 times for 5 min each in 1× PBS containing 0.5% Tween 20. 10. The sections are then overlayed with mouse anti-FITC (Roche Molecular Biochemicals) at 1:800 dilution in PBS with 0.5% Tween 20. 11. Following PBS washing, 3 times for 5 min each in 1× PBS containing 0.5% Tween, biotinylated goat antimouse IgG-Nanogold at 1:200 is applied for 30 min with intervening PBS washes. 12. Autometallography is performed with silver acetate or lactate developing solution for 8 min, and the reaction is stopped by immersion in distilled water. 13. The slides are then dehydrated in graded alcohols and xylene and covered with a coverslip. 14. Her-2/neu copy number is displayed as black spherical nuclear granules (Figure 9.4, A and B). DISCUSSION AND TECHNICAL HINTS The explosive growth of knowledge of the human genome, the greatly improved access to large-scale screening via tissue microarrays, and the advent of new probe technologies such as subtractive, unique sequence ISH probes derived from BACs or YACs that dramatically augment signal intensity by virtue of probe size and sequence specificity, are revolutionizing the technique of ISH.1,3,19,20 Even chromosomal deletions can be profiled in interphase FISH or CISH format, opening even 142

greater vistas for the revised and rejuvenated procedure, ISH. The purpose of this chapter is not to review the technique of ISH in technical detail. There are several excellent reviews of the method extant in the published literature. Only the highlights of particular importance to the autometallographic modifications of ISH are noted here. The value of appropriate cell conditioning in accounting for the remarkable improvements in ISH results cannot be overemphasized. Microwaving, simple heat retrieval, steam systems, pressure cookers, and simple water bath heat conditioning, have all improved the ability to see fewer copies of DNA/RNA targets with a little or no background staining. Automation of ISH by the newly manufactured Discovery in the Nexus Plus staining robot accomplishes all of these steps, also including deparaffinization and counterstaining, directly on-line on the instruments. And these instruments also allow for separate heating and buffer treatments of each of 20 separate slides, each stained by a unique protocol. All of the general principles underlying successful ISH experiments apply to the autometallographic modifications of the procedure. For example, careful attention to probe characteristics, stringency of washing solutions, hybridization conditions, and accuracy of denaturalization temperatures and conditions, must all be rigorously controlled and carefully monitored for successful staining, just as with FISH and chromogenic ISH systems. Hybridization requirements vary for probes in all systems; generally reflecting melting temperatures as a consequence of a GC base ratios. The formula Tm - 5°C = 0.1 × SSC @ 60°C can be used as general guideline for determination of the appropriate melting temperature of the probe, but in our experi-

Super Sensitive In Situ Hybridization ence, the formula is only useful for “ballparking” experimental conditions—nearly always, additional empiric experiments are required to find the “sweet spot.” When using tyramide systems, the contribution of endogenous peroxidase cannot be overemphasized. Even the smallest amount of peroxidase or pseudoperoxidase will result in catalysis of the tyramide conjugate and will seriously contribute to background staining.25 When using oligonucleotides directly conjugated to peroxidase, it is critical that one modifies the system so that the peroxidase activity of the conjugate is preserved. Temperatures above approximately 52°C will destroy peroxidase activity from the very expensive probe reagent. Concerning reporter molecules and probe size, we have extensively tested biotin-labeled cDNA probes for the detection of viral DNA. In these tests, we noted that some of the probes gave excellent results and in most cases were ISH stained using Protocols 1 through 4 (e.g., HPV 16/18; Enzo). However, we also noted that some other biotinylated cDNA probes (e.g., HPV 31/33/51; Enzo) most often lead to unacceptable background staining when used in combination with labeled tyramides, thus making efficient reading of the sections impossible. In case of RNA, we have successfully worked with self-made FITC-labeled riboprobes (Protocol 3). Using digoxigenin as the probe label, in many experiments, successful staining was never obtained when the TSA system was applied. The reason for this incompatibility is not yet clarified, but may be explained by steric hindrance. Probe size is also an important consideration. Whereas cDNAs and riboprobes give excellent sensitivity in many situations, oligonucleotides have not yet been extensively tested by us. Earlier, only some 20-bp long oligonucleotides could be made and labeled by one or two biotin molecules only. However, today it is relatively easy to synthesize oligoprobes of

much larger size, and methods have been proposed that allow a multiple labeling with reporter molecules (e.g., using the LabelIT nucleic acid labeling kits; Cat. No. MIR3400; Mirus, Madison, WI, USA). Such systems should make it possible now to detect single copies of genes or message by TSA by custom synthesis probes, which can then be optimally adapted for each specific application. For the autometallographic development, whether gold-enhanced or silver acetate–lactate solutions are used (see Chapters 2 and 3), it is best to first optimize staining by monitoring with conventional optical microscopy. But eventually, for assurance of reproducable performance of the procedure, conditions should be finalized in such a way that a standard time for a silver development can be identified and followed. It is imperative that sodium thiosulfate not be used to stop the silver reactions, as this reagent will most often destroy the visualization product that has been developed. On the other hand, extraction of endogenous trace metals with Lugol’s iodine solution and bleaching with thiosulfate early in the ISH procedure is absolutely imperative for proper perfomance of autometallographic ISH.7,23,25 REFERENCES 1.Bubendorf, L., J. Kononen, P. Koivisto, P. Schraml, H. Moch, T.C. Gasser, N. Willi, M.J. Mihatsch, G. Sauter, and O.P. Kallioniemi. 1999. Survey of gene amplifications during prostate cancer progression by high-throughout fluorescence in situ hybridization on tissue microarrays. Cancer Res. 59:803-806. 2.Cheung, A.L.M., A.H. Graf, C. Hauser-Kronberger, O. Dietze, R. Tubbs, and G.W. Hacker. 1999. Detection of human papillomavirus in cervical carcinoma: comparison of peroxidase, Nanogold, and catalyzed reporter deposition (CARD)-Nanogold in situ hybridization. Mod. Pathol. 12:689-696. 3.Davidson, J.M., T.W. Morgan, B.L. Hsi, S. Xiao, and J.A. Fletcher. 1998. Subtracted, unique-sequence in situ hybridization. Am. J. Pathol. 153:1401-1409. 4.Danscher, G., G.W. Hacker, J.O. Norgaard, and L. Grimelius. 1993. Autometallographic silver amplification of colloid gold. J. Histotechnol. 16:201-207.

143

Gold and Silver Staining 5.Danscher, G., G.W. Hacker, C. Hauser-Kronberger, and L. Grimelius. 1995. Trends in autometallographic silver amplification of colloidal gold particles, p. 11-18. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Methods and Applications. CRC Press, Boca Raton. 6.De Bault, L.E. and B.L. Wang. 1994. Localization of mRNA by in situ transcription and immunogold–silver staining. Cell Vision 1:67-70. 7.Fujimori, O., T. Ueda, and K. Yamada. 1994. Effects of iodine pretreatment of sections upon immunogold–silver staining in light microscopic immunohistochemistry. Okajimas Folia Anat. Jpn. 71:319-323. 8.Graf, A.-H., A.L.M. Cheung, C. Hauser-Kronberger, N. Dandachi, R.R. Tubbs, O. Dietze, and G.W. Hacker. 2000. Clinical relevance of HPV 16/18 testing methods in cervical squamous cell carcinoma. Appl. Immunohist. Mol. Morphol. (AIMM) (In press). 9.Hacker, G.W., L. Grimelius, G. Danscher, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc tissues. J. Histotechnol. 11:213-221. 10.Hacker, G.W., A.-H. Graf, C. Hauser-Kronberger, G. Wirnsberger, A. Schiechl, G. Bernatzky, U. Wittauer, H. Su, H. Adam, J. Thurner, et al. 1993. Application of silver acetate autometallography and gold-silver staining methods for in situ DNA hybridization. Chin. Med. J. 106:83-92. 11.Hacker, G.W. and G. Danscher. 1994. Recent advances immunogold–silver staining-autometallography. Cell Vision 1:102-109. 12.Hacker, G.W., G. Danscher, L. Grimelius, C. HauserKronberger, W. H. Muss, J. Gu, and O. Dietze. 1995. Silver staining techniques, with special reference to the use of different silver salts in light and electron micorscopical immunogold–silver staining, p. 20-45. In M.A. Hayat (Ed.), Immunogold–Silver Staining: Methods and Applications. CRC Press, Boca Raton. 13.Hacker, G.W., I. Zehbe, J. Hainfeld, J. Sallstrom, C. Hauser-Kronberger, A.-H. Graf, H. Su, O. Dietze, and O. Bagasra. 1996. High performance Nanogold in situ hybridization and in situ PCR. Cell Vision 3:209214. 14.Hacker, G.W. and G. Danscher. 1997. Immunogold–silver staining-autometallography: recent developments and protocols, p. 41-54. In J. Gu (Ed.), Analytical Morphology—Theory, Protocols, and Applications. Eaton Publishing, Natick, MA. 15.Hacker, G.W., C. Hauser-Kronberger, I. Zehbe, H.

144

Su, A. Schiechl, O. Dietze, and R. Tubbs. 1997. In situ localization of DNA and RNA sequences: supersensitive in situ hybridization using streptavidinNanogold silver staining: mini review, protocols, and possible applications. Cell Vision 4:54-67. 16.Hacker, G.W. 1998. High-performance Nanogold-silver in situ hybridisation. Eur. J. Histochem. 42:111120. 17.Hainfeld, J.F. 1987. A small gold-conjugated antibody label: improved resolution for electron microscopy. Science 236:450-453. 18.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 30:177184. 19.Kononen, J., L. Bubendorf, A. Kallioniemi, M. Barlund, P. Schraml, S Leighton, J. Torhorst, M.J. Mihatsch, G. Sauter, and O.P. Kallioniemi. 1998. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat. Med. 4:844-847. 20.Moch, H., P. Schraml, L. Bubendorf, M. Mirlacher, J. Kononen, T. Gasser, M.J. Mihatsch, O.P. Kallioniemi, and G. Sauter. 1999. High-throughput tissue microarray analysis to evaluate genes uncovered by cDNA microarray screening in renal cell carcinoma. Am. J. Pathol. 154:981-986. 21.Powell, R.D., C.M.R. Halsey, D.L. Spector, S.L. Kaurin, J. McCann, and J.F. Hainfeld. 1997. A covalent flourescent-gold immunoprobe: simultaneous detection of a pre-mRNA splicing factor by light and electron microscopy. J. Histochem. Cytochem. 45:947956. 22.Rufner, R., N.E. Carson, M. Forte, G. Danscher, J. Gu, and G.W. Hacker. 1995. Autometallography for immunogold–silver staining in light and electron microscopy. Cell Vision 2:327-333. 23.Tbakhi, A. G. Totos, C. Hauser-Kronberger, J. Pettay, D. Baunoch, G.W. Hacker, and R.R. Tubbs. 1998. Fixation conditions for DNA and RNA in situ hybridization: a reassessment of molecular morphology dogma. Am. J. Pathol. 152:35-41. 24.Tubbs, R.R., J. Pettay, T. Grogan, M. Stoler, and P. Roche. 2000. CODFISH (concomitant oncoprotein detection and fluorescence in situ hybridization). J. Mol. Diag. 2:78-83. 25.Zehbe I., G.W. Hacker, H. Su, C. Hauser-Kronberger, J.F. Hainfeld, and R.R. Tubbs. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, straptavidin-Nanogold, and silver acetate autometallography: detection of single-copy human papillomavirus. Am. J. Pathol. 150:1553-1561.

10

Immunogold Labeling Techniques for Transmission Electron Microscopy: Applications in Cell and Molecular Biology Laura Lossi, Patrizia Aimar, Elena Beltramo, and Adalberto Merighi

INTRODUCTION Since its introduction as an electrondense marker for electron microscopy in 1971,41 colloidal and recently also clustered gold have become the markers of choice for the in situ detection of a wide number of cellular and extracellular molecules by transmission electron microscopy (TEM). In this chapter, we briefly review the basic postembedding immunogold staining (IGS) and immunogold–silver staining (IGSS) methodology and then discuss some more recent applications based on the combination of colloidal gold immunocytochemistry (ICC) with enzyme histochemistry, tracing techniques, and transferase–immunogold labeling procedures for use in cell and molecular biology. A number of publications appeared in the past few years in which the basic IGS methods have been extensively reviewed.9–11,14,56,72,74,89,91,114–116 In this chapter, we will focus our attention on the more recent exploitations of the technique with particular emphasis on applications in cell and molecular biology. Since our research interest has always been devoted 0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

to the study of the central nervous system (CNS), the examples of application described here will mainly be related to analysis of central neurons. Nevertheless, we are confident that the protocols discussed in this chapter will be helpful also to the general biologist and in clinical research. A brief summary of the state-ofthe-art will help to put our work in perspective. Readers are referred to the existing literature for a more detailed discussion of primary fixation and for comparison between postembedding and pre-embedding procedures. 9,10,72,

74,89,91,96,106,111,114–116,129

Basic Postembedding Immunogold Labeling Methods General Principles The main factors to be considered for successful immunolabeling may be summarized as follows: (1) retention of antigens within the tissue and preservation of their antigenicity; (2) preservation of adequate ultrastructure; and (3) absence of a barrier likely to prevent penetration of the 145

Gold and Silver Staining antibodies into the tissue and interaction with their respective antigens. The achievement of the optimal balance of these three factors is often a rather demanding task. It seems obviously pointless to obtain labeling in the absence of good ultrastructure, and vice versa. Therefore, primary fixation represents a necessary preliminary step to all ICC procedures, although ultrathin frozen sections of unfixed material or cryosubstitution techniques might represent useful alternatives.26,35–38,46,88,112 However, these methods are very expensive, show many technical difficulties, and their general practicability is still limited. After primary fixation, ICC procedures can be carried out following the pre- or postembedding approach. As an alternative, some of the problems related to the embedding procedures can be avoided by cutting ultrathin frozen sections of fixed tissue, although this procedure is also very demanding in terms of costs and technical skill.65,76,140 As we will discuss in one of the following sections, the use of smallsized colloidal gold particles and/or gold clusters, in conjunction with silver intensification procedures in a pre-embedding protocol, represents an appealing alternative for localization of particularly labile antigens, since it combines together several advantages that are inherent to either the pre- or postembedding methods. The main advantages and disadvantages of these approaches are summarized in Table 10.1. Using immunogold postembedding staining methods, it is necessary to be aware that problems will be encountered, which are not only related to fixation and the labeling procedure, but also to all the preparative stages of postfixation, dehydration, resin infiltration, and curing. All these aspects have been extensively reviewed in a previous publication.91 Nonetheless, we intend to summarize here some aspects of particular relevance and mention some more recent findings that 146

may represent significant ameliorations of the existing protocols. It seems worthy to remember that embedding can be carried out using both hydrophobic epoxy resins and hydrophilic acrylic embedding media.25 For most tissue, the quality of tissue preservation is definitely superior after epoxy resin embedding and osmium postfixation, although embedding in hydrophilic media results in higher labeling efficiency. This latter is a consequence of less detrimental effects on tissue antigenicity.23 Nevertheless, a method of Epon preparation has recently been described that allowed for obtaining approximately the same IGS labeling for epoxy sections as for acrylic sections without any etching.22 Osmium postfixation represents a fundamental step in postembedding ICC of epoxy sections. Proteins and some macromolecular carbohydrates are well preserved after aldehyde fixation. In general, however, lipids remain soluble in organic solvents such as alcohols or acetone57,101 and are extracted during embedding procedures unless secondary fixation with osmium tetroxide is applied or aldehyde-fixed tissue is embedded at low temperatures using hydrophilic resins.23,25,144 We have been forced into the use of osmium postfixation because it was necessary for adequate preservation of the delicate membrane structures of different synapses and subcellular organelles within the CNS. Recently, an osmium-free method of Epon embedment that preserves both ultrastructure and antigenicity for postembedding ICC of CNS has been introduced.102 The method is based on primary fixation with high level glutardialdehyde and replacement of osmium postfixation with tannic acid followed by other heavy metals and p-phenylenediamine. Nevertheless, to obtain the maximum advantage from the use of postembedding techniques, osmium postfixation

Immunogold Labeling Techniques should be used whenever possible, irrespective of the tissue under study. Theodosis and coworkers have been among the first to successfully apply a postembedding IGS procedure for localization of neuropeptides in central neurons after osmium postfixation.135,136 Since then, numerous papers have followed, and authors have experienced that pretreatments with strong oxidizers are usually required to restore antigenicity of osmicated material. Different chemicals have been employed, but sodium metaperiodate15 gives the most satisfactory results. Treatment with sodium metaperiodate restores tissue antigenicity by removing osmium that has not reacted with tissue components.25 It also enhances the hydrophilic nature of the resin by oxidizing the hydrophobic alkane side chains to alcohols, aldehydes, and acids,25 thus allowing an easier antigen–antibody interaction. Etching the sections by immersion in a saturated aqueous solution of sodium metaperiodate is a critical step, and we have previously dealt with problems inherent to this aspect of the labeling procedure. A suitable exposure time for the metaperiodate solution has to be a compromise between that needed to restore enough antigenicity and that in which the sections would be damaged or, eventually, destroyed.89,91 It seems of relevance the finding that, by using a sodium citratebuffered sodium metaperiodate solution at high temperature, a significantly higher gold labeling density is achieved.131 There is increasing evidence that at least certain peptides can be successfully labeled after osmium postfixation without any pretreatment of the tissue and conventional149 or modified22 epoxy embedding. As alternatives to the use of osmium postfixation and conventional dehydration and embedding procedures, several other protocols have been proposed in the past years, but they have not yet entered in routine use.

Single Staining In our laboratory, single immunolabeling is performed directly on uncoated grids using an indirect method with immunoglobulin G (IgG)-gold or, less frequently, protein A (or protein G)-gold (see Figures 10.1A, 10.4A, and the working Protocol 1). All immunostaining procedures are carried out using 20- to 50-µL drops of solutions on inverted petri dishes with parafilm sheets taped to the inside of the lid. Grids must be wet throughout the entire labeling procedure. It is advisable to place a strip of filter paper soaked with water within the petri dish to create a humid chamber and thus avoid evaporation of immunoreactants. Epon and/or Araldite sections of unosmicated material are etched in hydrogen peroxide,15 while those from osmicated tissue are pretreated with sodium metaperiodate to remove osmium tetroxide (but see above). Grids are subsequently washed with 0.5 mol/L Tris-buffered saline (TBS), pH 7.2 to 7.6, containing 1% Triton X100 to facilitate contact between the hydrophobic surface of the sections and immunoreagents. They are then incubated in a normal serum from the donor species of the gold-conjugated IgG or with 1% ovalbumin in TBS when protein A (protein G)-gold probes are used. Other proteins can be added to TBS to reduce background staining instead of, or in addition to ovalbumin (see Table 10.4 in Reference 91). Grids are subsequently blotted with fiber-free filter paper and directly transferred into the primary antiserum without washing in buffers. Incubation in the primary antiserum is usually carried out overnight at room temperature, however its length and the titer of primary antisera have to be varied according to the different primary antibodies used. Criteria for primary antiserum incubation follow the same general principles as all ICC procedures. 147

Gold and Silver Staining We use 0.5 mol/L TBS (pH 7.2–7.6) containing 1% bovine serum albumin (BSA) as a diluting buffer for both primary antisera and colloidal gold probes and as a rinsing buffer in the next washes (TBSBSA). Tris buffer is preferred to other buffers because it has a stabilizing effect on colloidal gold sols.59 The high molarity of the buffer reduces nonspecific background, which is a consequence of electrostatic interactions between the negatively charged gold sols and positively charged groups present in the structure of epoxy resins. BSA is added to the buffer for two reasons: (1) to reduce nonspecific background, and (2) to stabilize the gold probe suspensions. To obtain cleaner preparations, we recommend to jet wash grids with a syringe or a wash bottle between each change of a buffer, rather than only floating them on buffer drops. After an extensive wash in TBS-BSA, grids are incubated in the gold conjugates, usually IgGgold or protein A (protein G)-gold, rinsed in TBS-BSA, and postfixed in 2.5% glutardialdehyde in sodium cacodylate buffer to fix the immunoreactants. Finally, grids are extensively washed in double-distilled water. All buffers must be removed prior to counterstaining because uranyl acetate precipitates in the presence of traces of saline. Multiple Staining The possibility to localize two or more antigens in the same tissue section is one of the major advantages associated with the use of postembedding IGS procedures (Figures 10.3 and 10.4B). The labeling efficiency of any of multiple labeling protocols described below is lower when compared to that of the single labeling methods, considering that all multiple labeling techniques suffer from problems which are inherent to: (1) spatial competition (steric hindrance) between primary antisera and/or gold 148

probes, and (2) the necessity to obtain a proper exposure of all antigens to be immunostained at the surface(s) of the section. The advantages and disadvantages inherent to each procedure in terms of technical simplicity and labeling efficiency have already been discussed in detail.74,91 Simultaneous Labeling The simultaneous double (multiple) labeling method should or can be used when primary antisera raised in different species are available.134 In practical terms, this is simply a single labeling procedure in which grids are incubated with mixtures of primary antisera, followed by a mixture of the appropriate colloidal gold-labeled IgGs of convenient sizes (in general 10 and 20 nm) (Figure 10.3). The major advantage of this method is that it is technically very simple and specific. Double-Face Labeling When ultrathin plastic sections are collected onto uncoated grids, both faces are available for immunostaining considering that there is not penetration of immunoreagents into the resins. This principle is at the basis of the so-called double-face labeling method, which is used for visualization of two different antigens on the same ultrathin section using primary antisera (antibodies) raised in the same species. In its original formulation, this was performed using protein A-gold complexes.9 A modification of the original protocol was subsequently described in which IgG-gold probes were used.145 Irrespective of the gold probe used, each face of the grid is immunostained separately according to the single labeling procedure using different primary antibodies and colloidal gold conjugates with particles of different sizes (e.g., 10 and 20 nm). A careful manipulation of

Immunogold Labeling Techniques the grid is necessary to avoid both sides being wet by the same reagent, thus giving rise to false positive results. This can be more easily avoided if a formvar film is placed on the face of the grid used in the first immunolabeling procedure.92 Sequential Double Protein A-Gold Labeling The sequential double protein A-gold procedure45,113,114 is carried out on a single face of the grid. As for the double-face labeling method, this procedure can be used when the primary antibodies available are both raised in the same species. First, a single labeling is performed using gold particles of small diameter (3–5 nm), then, unreacted immunoglobulins are saturated with an excess of free protein A, and finally, another single labeling procedure is carried out with gold particles of larger size (10 nm). To avoid false double positive results, it is of paramount importance to saturate all free protein A binding sites. Paraformaldehyde Vapor Technique Binding sites of unreacted immunoglobulins, but not of tissue antigens, are denaturated using hot paraformaldehyde vapors.148 This theoretically enables one to use two or more primary antibodies raised in the same species for multiple staining. The procedure is carried out on a single face of the grid. After a single labeling procedure using gold particles of small diameter, grids are treated with hot paraformaldehyde, and a second single labeling procedure is carried out with gold particles of large size. Other Multiple Labeling Methods It has been reported that silver intensification of colloidal gold particles allows dou-

ble labeling with primary antisera raised in the same species.16 It is also possible to combine any of the techniques described above to perform multiple staining.34,92 Spatial Resolution of Postembedding IGS Methods The use of postembedding IGS techniques gives the possibility of localizing antigens to different subcellular compartments more precisely than after other ultrastructural ICC methods. Therefore, the spatial resolution of the procedures used becomes a critical factor that must be carefully considered to determine the actual site in which the antigen(s) under study is (are) localized. The spatial resolution of the technique is defined as the distance between the center of the gold particle and the epitope recognized by the primary antibody. The minimum value that can theoretically be achieved with the protein A-gold technique is about 16 nm using a colloidal gold particle of 3 nm.114 Using the IgG-gold technique, the spatial resolution is slightly less satisfactory since IgGs are bigger than protein A.100 For example, using IgG-gold conjugates with gold particles of 10 nm, the resolution is approximately 21 nm. Therefore, the bigger the gold particle, the less satisfactory is the spatial resolution of the technique. Gold clusters of small size covalently attached to Fab′ fragments (Nanogold; Nanoprobes, Yaphank, NY, USA) offer a better resolution since Fab′ fragments are about 1/3 the size of IgGs.55 Using electron spectroscopic imaging for high resolution ICC with boronated protein A and various postembedding protocols, a spatial resolution in the magnitude of 0.5 nm was achieved.12,109 However, this approach is unlikely to enter into routine use, even though it represents a major improvement in resolution over the more commonly employed techniques. 149

Immunolabeling Method

Advantages

Disadvantages

Pre-Embedding

Large areas of the tissue may be immunostained and examined.

Only the more superficial sections of tissue blocks show optimal labeling due to poor penetration of immunoreactants.

It is possible to flat-embed the material in such a way that the same immunostained section is used for both light and electron microscopic examination.

Treatments, which are necessary to improve penetration of immunoreagents, have a more or less deleterious effect on the ultrastructure.

The sensitivity is very high and allows the detection of low concentration antigens.

The electron dense product of the peroxidase reaction diffuses from the original site of deposition and/or masks the underlying structures and therefore does not allow a reliable subcellular localization of antigens. Multiple labeling experiments are relatively difficult to perform.

Pre-Embedding with Ultra Small Size Gold

Large areas of the tissue may be immunostained and examined.

Only the more superficial sections of tissue blocks show optimal labeling due to poor penetration of immunoreactants.

It is possible to flat-embed the material in such a way that the same immunostained section is used for both light and electron microscopic examination.

Treatments, which are necessary to improve penetration of immunoreagents, have a more or less deleterious effect on the ultrastructure.

The sensitivity is very high and allows the detection of low concentration antigens.

Multiple labeling experiments are relatively difficult to perform.

Reliable subcellular localization of molecules under study is obtained.

Gold and Silver Staining

150

Table 10.1. Comparative Evaluation of Pre- and Postembedding Procedures

Table 10.1. (Continued) Postembedding

There are no penetration problems, since the cells are open to the immunoreactants, and these are surface reactions.

The antigenicity of the tissue is generally reduced by the osmium postfixation, the embedding, and the curing procedure.

Reliable subcellular localization of molecules under study is obtained.

Smaller areas of tissue can be immunostained compared to pre-embedding procedures.

Gold particles can be prepared in different sizes allowing simultaneous visualization of more than one antigen in the same preparation and even within the same cell organelle.

It is more difficult to correlate light and electron microscopic findings on the same material than after pre-embedding immunostaining.

Different (serial) sections from the same block can be labeled with different primary antisera and/or used for immunocytochemical controls.

Immunogold Labeling Techniques

151

Gold and Silver Staining Quantification of IGS In epoxy resin embedded material, gold labeling is restricted to antigens exposed to the surface of the ultrathin sections.9,13 Since immunoreagents do not penetrate sections here, tissue permeability is not influencing the distribution of labeling as after pre-embedding ICC. This makes it possible to quantify the gold particle label over different subcellular compartments. Quantitative measurement of gold labeling in ultrathin frozen section requires a threedimensional analysis, since in this case immunoreactants do penetrate tissues.76,104,105 Results are usually expressed in terms of the number of gold particles per area of tissue and should be corrected for background labeling over empty resin.39, 40,51,60,92,114,125,127,129 Computerized systems and dedicated software have been developed to this purpose.7,17,39,108 IGSS at the Ultrastructural Level The basic principles of IGSS described for light microscopical ICC also hold for TEM applications (Figure 10.1B). The technique of silver enhancement, originally termed autometallography (AMG), has

been established since the beginning of the 1980s mainly from the work of Danscher and Hacker (reviewed in References.53,54,117 Silver intensification can be used for producing different-sized gold particles in multiple labeling procedures (see above) and as a single step procedure after immunogold labeling to facilitate the visualization of gold particles at low magnification.63,70,71,130 However, the most relevant ultrastructural applications of IGSS techniques are based on the use of very small (about 1 nm) colloidal gold particles or gold clusters (Nanogold)55 for: (1) ICC labeling, (2) molecular biology (see below), and (3) neuronal tracing procedures. In tracing procedures, (stereotaxic) injections in vivo of 1 nm colloidal gold particles coupled to appropriate protein carriers are performed. The gold–carrier complex is anterogradely and/or retrogradely transported by neurons and then visualized in the electron microscope after silver intensification.103 The potential application of this method in combination with postembedding immunogold labeling procedures is discussed in one of the following sections. We will discuss below the main advantages and disadvantages of the use of silverintensified 1 nm colloidal gold particles or

Figure 10.1. Schematic drawing illustrating the principles of the indirect immunogold (A) and immunogold–silver (B) labeling procedures.

152

Immunogold Labeling Techniques Nanogold in pre-embedding ICC labeling. More detailed information about the characteristics and application of Nanogold is found in Chapters 6 and 10. This approach has been introduced in the recent past to overcome some of the problems that are inherent to the “classic” pre- and postembedding immunogold labeling methods (see Table 10.1). The general procedure consists of primary aldehyde fixation with very low concentration glutaraldehyde, usually below 0.5%, followed by incubation with primary antibody at optimal titer. The ICC reaction is then visualized by incubating sections with appropriate 1 nm colloidal gold or Nanogold reagents. After silver intensification, the tissue is postfixed in higher concentration glutardialdehyde (optional), osmium tetroxide, and embedded in resin. This approach somehow conjugates together many of the advantages of the classic pre- and postembedding methods, in particular: (1) it allows the labeling of very sensitive antigens that otherwise could not be detected by a postembedding approach; (2) it does not suffer from most of the penetration problems inherent to pre-embedding procedures considering the small size of the probes employed; and (3) the reaction end-product depicts very precisely the subcellular site of antigen localization in a fashion comparable to that of postembedding gold labeling80 (Figure 10.2). Despite some technical difficulties, there are numerous examples in which this method has been successfully used to label very labile antigens in several cells and tissues.5,28,29,132,143 Controls Controls are necessary to assess the specificity of the immunostaining and the quality of the detection method used as in any other ICC procedure. This matter has

been extensively reviewed in the existing literature.73 Combined Postembedding Immunogold Labeling Methods and Cell–Molecular Biology Techniques As thoroughly discussed in the previous sections, single and/or multiple postembedding immunogold labeling can now often be performed after high glutaraldehyde, osmium, and epoxy embedment, i.e., using a material which was subjected to an almost conventional preparative procedure. This has opened the way to combine the immunogold techniques with several other methods relevant to cell and molecular biology. These combined approaches have a wide range of potential applications and can be theoretically employed to solve a number of problems in different fields of basic and applied research. Combination with Pre-Embedding Procedures Postembedding immunogold labeling procedures have been successfully applied in combination with pre-embedding ICC labeling. This approach has been particularly useful for the study of the connectivity and synaptic interactions of central neurons.28,29,49,50,98 In general terms, a pre-embedding labeling procedure is performed first, after low or mild aldehyde fixation, using a primary antibody directed towards the more labile antigen to be detected and a suitable chromogen (usually 3,3′-diaminobenzidine [DAB]). Then, stronger aldehyde fixation, osmication, and embedding followed. The second, more resistant, antigen under study is eventually visualized using a classical postembedding staining protocol. The entire procedure is technically very demanding and not always so straightforward. The choice of such an approach 153

Gold and Silver Staining mainly depends on the impossibility to visualize the more labile antigen with a postembedding protocol. However, there are several other variables to consider, such as, for example, the need to analyze the ultrastructural relationship between membrane-bound receptors, which are better visualized by a pre-embedding approach, and their ligands in synapses, which are ideally labeled by postembedding gold ICC. Combination with Enzyme Histochemical Techniques Again this approach combines a pre- and a postembedding procedure, but the first preembedding labeling is a histochemical reaction, which is used to reveal endogenous enzyme activities. In this case, primary fixation must be compatible with retention of the activity of the enzyme(s) under study. Moreover, there is the need of a proper chromogen for ultrastructural visualization of the enzyme–substrate reaction. Fortunately, a number of enzymes, such as acetylcholinesterase,18,126 alkaline phosphatase,48,52, 128 β-galactosidase,123 and nicotinamide adenine dinucleotide phosphate-diaphorase (NADPHd)2,19,133 retain their activities in tissues fixed for ultrastructural examination. The possibility to label nitric oxide (NO)-producing cells by means of a NADPH-d reaction modified for ultrastructural use in combination with postembedding IGS3,19 seems to be of particular interest. The free radical gas NO is a recently identified intracellular messenger molecule which appears to be involved in a wide range of physiological and pathological processes.121 Several lines of evidence indicate that in aldehyde-fixed tissues, NADPH-d corresponds to NO synthase (NOS), the key enzyme in the synthesis of NO.146 The histochemical method for NADPH-d has been widely used to investigate the light microscopic distribution of 154

cells containing NOS. The NADPH-d reaction has been modified for ultrastructural use, basically by substituting the nonosmiophilic chromogen nitroblue tetrazolium (NBT) (currently used for light microscopy), with 2-(2′-benzothiazolyl)-5styryl-3-(4′-phthalhydrazidyl) tetrazolium chloride (BSPT), a non-osmiophilic tetrazolium salt that, unlike NBT, yields an osmiophilic formazan.2,31,133,150 We have combined the pre-embedding visualization of NO-producing neurons in the spinal dorsal horn (by means of a modified NADPH-d reaction) with postembedding immunogold labeling (Figure 10.3) of the calcitonin gene-related peptide (CGRP), substance P, and glutamate to better characterize the synaptic relationship of these molecules in pain processing.3 In this case, primary fixation could be carried out using high concentrations of glutaraldehyde. The reaction was performed on vibratome sections, since the enzyme activity was inhibited by the curing temperature of the resin (see Protocol 2). The choice of NBT or BSPT as a chromogen for the diaphorase reaction at the ultrastructural level has been discussed in another recent publication.2 An additional field of further exploitation of postembedding gold labeling methods relies on the possibility to combine them with the ultrastructural visualization of enzymes such as β-galactosidase and human placental alkaline phosphatase, which have been engineered as reporter genes in transgenic organisms.52,123 Combination with Neuroanatomical Tract-Tracing Methods The combination of postembedding IGS labeling and neuroanatomical tracttracing methods has represented a useful tool in the study of neuronal connections during the past years. There are several possibilities to trace neuronal connections at

Immunogold Labeling Techniques the light level. Any of the methods so far developed for light microscopy has inherent advantages and disadvantages, but, in general terms, to be successfully employed in conjunction with postembedding ultrastructural cytochemistry, a good preservation of tissue cytology is mandatory together with the need to employ an electrondense tracer. Successful tracing of neuronal pathways at the ultrastructural level has been achieved using: (1) Golgi impregnation;126 (2) retrograde labeling with either free horseradish peroxidase (HRP) or HRP conjugated to plant lectins or other electron dense markers such as colloidal gold particles;6,32,79,82–84,86,107,111 (3) anterograde labeling with free HRP20,21,30,64,75 or biotinylated dextran amine;122 and (4) HRP microiontophoretic filling of single neurons or fibers, following electrophysiological characterization.4,85,110 There are numerous examples in the literature in which these methods have been employed in conjuction with postembedding immunogold labeling.90,97,118,122 Other approaches such as: (1) anterograde tracing using radiolabeled amino acids;58 (2) degeneration methods;67,68 and (3) microiontophoretic filling of single neurons under visual control93 can also be potentially combined with postembedding immunogold labeling. As a general protocol, all these procedures involved the administration of the tracer in vivo or in a slice preparation, followed by primary fixation in low concentration glutardialdehyde and the preembedding visualization of the tracer by enzyme histochemistry. Tissue is then postfixed in higher glutardialdehyde concentration, osmicated, and embedded. Finally a classical IGS procedure is employed to detect directly on tissue sections the antigen(s) of interest. Since the procedure is generally carried out on vibratome slices, correlative light and electronic studies are made possible.

Combination with Freeze-Fracture Replication and Freeze-Etching Freeze-fracture replication and freezeetching techniques both involve the production of a vacuum-deposited replica from a fracture plane in a frozen specimen.112 Nowadays, these are established methods in transmission (and scanning) electron microscopy, particularly in cell and membrane biology. In the recent past, several different approaches have been developed to combine IGS with freezefracture,42,43,61 freeze-etching,147 and deep freezing.62 The methods employed so far are very heterogeneous, and IGS has been performed either before or after freeze-fracture, or even directly on platinum–carbon replicas of the freeze-fractured membranes. The specific technical aspects of these methods have already been reviewed in the recent past.141 Combination with Molecular Biology Techniques We will review in this paragraph the most relevant applications based on the use of postembedding colloidal gold methods in ultrastructural molecular biology. In situ hybridization labeling procedures, in situ polymerase chain reaction (PCR) and in situ self-sustained sequence replicationbased amplification (3SR) at the ultrastructural level can also rely on the use of goldlabeled probes generally followed by silver intensification141 and can be successfully combined with postembedding immunogold labeling.155,156 Also, the combination of in situ reverse transcription PCR (RTPCR) and immunogold cytochemistry has recently been described for receptor localization at the light level,87 which has a great potential for ultrastructural applications. Although of particular relevance for molecular biology applications, all these 155

Figure 10.2. Pre-embedding immunogold labeling. Glial fibrillary acidic protein (GFAP) in astrocytes of the rat hippocampus using AMG-amplified 1.4 nm gold clusters (Nanogold). The label is clearly visible over the glial filaments in the astrocytic processes (A) and cell bodies (B). Abbreviations: cb = cell body; gf = glial filaments. Scale bars—0.5 µm.

156

Immunogold Labeling Techniques

Figure 10.3. Combined pre-embedding visualization of NADPH-d and postembedding immunogold labeling of sensory neuropeptides. Rat spinal dorsal horn sections from different segmental levels (A,B). The NADPH-d positive dendrites contain floccular NBT deposits. Two peptide-immunoreactive axonal endings are shown. They contain a number of large dense core vesicles (LGVs) which are immunolabeled using a rat monoclonal antibody against substance P (10 nm gold) and a rabbit polyclonal antiserum against CGRP (20 nm gold). Some double-labeled LGVs are shown at higher magnification in the inserts. Abbreviations: d = dendrite. Scale bars—0.5 µm; inserts—0.25 µm.

157

Gold and Silver Staining methods will not be taken into consideration in this review for the sake of brevity. The availability of nucleic acids at the surface of epoxy sections allows for specific labeling with exogenous transferases and modified nucleotides. Moreover, the use of nonisotopic nucleotide analogues has considerably improved the potential applications of these procedures and the possibility to combine them with the immunogold labeling methods. In general terms, molecular biology techniques at the ultrastructural level are two-step procedures, which are often referred to as transferase-immunogold techniques. In these, an enzymatic (transferase) reaction is first performed, which allows for the labeling of the nucleic acids. Then, an IGS procedure is used to reveal the incorporated nucleotide analogues.137–139 In the past years, these techniques have been mainly used to study in situ the ultrastructural distribution of DNA and RNA in various cell types and tissues. To label DNA, a terminal deoxynucleotidyl transferase (TdT)-immunogold technique was developed, while RNA was labeled with a polyadenilate nucleotidyl transferase (PnT)-immunogold technique.137,138 TdT is an unusual type of DNA polymerase found only in lymphocyte precursors at early stages of their differentiation.27 It catalyzes the template-independent addition of deoxyribonucleoside triphosphates to the 3′ OH ends of double- or single-stranded DNA.27 In the original protocols, nucleotides to be added by TdT on ultrathin sections were biotin- or brominelabeled, and either an antibiotin antibody or a monoclonal antibromodeoxyuridine antibody in conjuction with appropriate gold probes were used to visualize the reaction in the electron microscope.139 PnT and biotinylated ATP were used to label RNA.139 PnT is a primer-dependent polymerase isolated from Escherichia coli. It catalyzes the addition of 5′ adenosine 158

monophosphate to the 3′ OH end of single-stranded RNA using ATP as its substrate.124 The potential value and applications of these techniques have increased significantly in the recent past for several reasons. First, the introduction of the digoxigenin labeling method for detection of nucleic acids has represented a major breakthrough in terms of specificity and sensitivity for use of transferase-immunogold procedures. Digoxigenin (DIG) is a steroid isolated from digitalis plants. AntiDIG antibodies are highly specific, since the only natural source of the steroid are the blossoms and leaves of Digitalis spp. They can be easily conjugated to fluorochromes, enzymes such as HRP or alkaline phosphatase, and colloidal gold particles, to reveal the presence of DIG-labeled nucleotides in tissue sections or blots. DIG is usually linked to the C-5 position of uridine nucleotides via a spacer arm containing 11 carbon atoms. In hybridization procedures, the DIG-labeled nucleotides may be incorporated, at a defined density, into nucleic acid probes by DNA polymerases, RNA polymerases, and TdT. DIG label may be added by random primed labeling, nick translation, PCR, 3′ end labeling–tailing, or in vitro transcription.1,81,120 Second, a number of monoclonal antibodies have been developed against thymidine analogues such as bromodeoxyuridine (BrdU) and iododeoxyuridine (IdU), which have opened the possibility to perform cell cycle kinetic studies at the light level,8,95,119,154 but also at ultrastructural level directly on ultrathin epoxy sections of osmicated tissues.24,69 Third, a series of transferase reactions have recently been developed at the light level to visualize apoptotic cells directly on tissue sections after in situ nick translation with DNA polymerase I47 or unmodified T7 polymerase,151 end labeling with TdT,44,47 or ligation of DIG-labeled double-stranded

Immunogold Labeling Techniques DNA fragments with T4 DNA ligase.33 Apoptosis is a gene-regulated process of cell death in which the cell is an active participant in its own demise.66,152 This form of “cell suicide” is most often found during embryonic development, but also in normal cell and tissue turnover, maintenance of immune tolerance, and endocrine-regulated tissue atrophy. Fragmentation of DNA into low molecular weight oligomers is a biochemical hallmark of apoptosis.153 Cleavage of the DNA may yield doublestranded mono- and/or oligonucleosomes of low molecular weight, as well as “nicks” (single strand breaks) in high molecular weight DNA. These DNA strand breaks can be detected by enzymatic labeling of the free 3′ OH termini with modified nucleotides, usually dUTP linked to a fluorochrome, biotin, DIG, etc. DNA polymerase I catalyzes the template-dependent addition of nucleotides when one strand of a double-stranded DNA molecule is nicked. Theoretically, by such an approach, not only the apoptotic DNA is detected, but also the random fragmentation of DNA by multiple endonucleases occurring in cellular necrosis. Unmodified T7 DNA polymerase is a highly processive DNA template-dependent 5′→3′ polymerase and a 3′→5′ exonuclease for 3′ OH acceptor groups at recessed, blunt, or overhanging ends. To detect apoptotic cells, the enzyme is utilized, first, to generate 5′ overhangs from any 3′ OH acceptor groups at DNA strand breaks and second, to end-fill the overhangs and incorporate biotinylated dATP.151. T4 DNA ligase was used to ligate in situ double-stranded DNA fragments labeled in vitro with DIG (or Texas Red) to DNA double-strand breaks with singlebase 3′ overhangs as well as blunt ends in apoptotic nuclei.33 This approach has been shown to be very specific for apoptotic DNA, avoiding the false positive results

that can derive by DNA damage from necrosis, in vitro autolysis, peroxide toxicity, and heating. The nick end labeling with TdT (TUNEL = TdT-mediated dUTP nick end labeling) is considered to be more sensitive, faster, and specific than the template-dependent nick translation. For ultrastructural examination, TdT was used to enzymatically link 11-digoxigeninated-dUTP to fragmented DNA in apoptotic cells in conjunction with anti-DIG goldlabeled antibodies94,99 (Figure 10.4B). We have recently used this procedure in combination with IGS of proliferating cells after in vivo incorporation of BrdU or IdU. This approach permitted the simultaneous determination of cell kinetic parameters of proliferating and apoptotic cells in the postnatal cerebellum.77 The method is particularly interesting since it allows for preferential labeling of cells at early stages of apoptosis and/or pre-apoptotic cells, i.e., cells without any typical ultrastructural sign of apoptosis. We are at present trying to modify the T4 DNA ligase procedure for use at the ultrastructural level, since this method should offer the best results in terms of specificity for apoptotic DNA cleavage.33 Clearly, these procedures have a great potential of exploitation in the future. Indeed, it is very often difficult to assess with certainty the nature of apoptotic cells at the light level, although transferase techniques can be combined with ICC labeling.78,142 However, apoptotic cells are generally labeled at a stage of immaturity in which any specific antigen is not expressed yet, or, in the opposite, they are already so heavily damaged that they do not retain specific antigens to give some clues on their origin or nature. Some of these problems are easily bypassed after ultrastructural examination, considering that different cell types have generally very distinctive cytological features, even at very early stages of differentiation. 159

Gold and Silver Staining

Figure 10.4. Immunogold labeling of exogenously administered BrdU. Postnatal rabbit cerebellum using an anti-BrdU monospecific monoclonal antibody (20 nm gold). (A) Single BrdU labeling of a granule cell precursor in the external granular layer (EGL). Gold particles are dispersed throughout the nucleus, at times associated with the condensed chromatin. (B) Simultaneous visualization of BrdU and DNA fragmentation in an apoptotic cell within the EGL. The cell nucleus has been labeled for DNA fragmentation using a postembedding TUNEL procedure with a 10 nm gold-labeled anti-DIG antibody (arrows). The apoptotic bodies also contain 20 nm gold particles (arrow head) indicative of BrdU incorporation. The asterisks indicate the areas shown at higher magnification in the corresponding inserts. Scale bars: (A) 0.5 µm; insert, 0.25 µm; (B) 1 µm; insert, 0.1 µm.

160

Immunogold Labeling Techniques PROTOCOLS Protocol 1. Single Immunogold Labeling Procedure On Grid All solutions must be filtered with disposable filters (0.22 µm; Millipore, Bedford, MA, USA). All steps are performed directly on grid. Grids are incubated on drops (30–50 µL) of solutions. 1. Collect sections onto precleaned uncoated 200 to 300 mesh nickel grids (Electron Microscopy Science, Fort Washington, PA, USA). 2. Float grids on drops of a saturated aqueous solution of sodium metaperiodate (Sigma, St.Louis, MO, USA) for 5 to 15 min at room temperature. 3. Rinse in 0.5 mol/L TBS, pH 7.0 to 7.6, containing 1% Triton X-100. 4. Incubate for 60 min at room temperature in TBS containing 1% BSA (Fraction V; Sigma) (TBS-BSA) to which 10% normal serum from the donor species of the gold-conjugated IgGs has been added. If protein A-gold complexes are used, incubate in TBS-BSA containing 1% egg albumin (Sigma). 5. Transfer onto drops of primary antisera at optimal titer for 24 to 72 h at 4°C. Thoroughly wash in TBS-BSA and incubate for 60 min at 37°C with the appropriate gold conjugate (British BioCell, Cardiff, Wales, UK) diluted 1:15 in TBS-BSA. 6. Extensively rinse in TBS-BSA and postfix for 10 min at room temperature in 2.5% glutardialdehyde in cacodylate buffer. 7. Wash grids in double-distilled water and allow to dry protected from dust. 8. Counterstain with Reynold’s lead citrate and uranyl acetate.

Protocol 2. Electron Microscopical NADPH-Diaphorase Histochemistry 1. Preincubate Vibratome sections (50–200 µm) in 0.1 mol/L Sörensen buffer, pH 7.4, for 10 min at room temperature. 2. Transfer sections in the same buffer containing 0.25% Triton X-100, 1 mg/mL β-NADPH (Sigma), and 0.2 mg/mL NBT (Fluka Chemicals, St.Louis, MO, USA). 3. Monitor the reaction under the microscope and stop it by transferring sections in Sörensen buffer. 4. Postfix for 1 h at 4°C in osmium ferrocyanide 5. Thoroughly wash in maleate buffer, pH 5.2, and stain for 1 h at 4°C with 1% uranyl acetate in maleate buffer, pH 6.0. 6. Extensively rinse in maleate buffer, pH 5.2, and dehydrate in increasing concentrations of ethanol. 7. Flat-embed in Araldite. CONCLUSION In 1996, the silver anniversary of colloidal gold was celebrated. Since 1971, when Faulk and Taylor first published their “colloidal gold method for the electron microscope”,41 more than 25 years of colloidal gold marker system for histochemistry and ICC had passed. Nowadays, colloidal gold methods have entered multiple fields of cell and molecular biology and proved to be irreplaceable tools for conjugating together the morphological and functional analysis of living structures. We are confident that these methods will be further expanded and perfected, so that we can still rely on them after entering the third millennium.

161

Gold and Silver Staining ACKNOWLEDGMENTS Much of the work described here was supported by grants from the University of Torino, the Italian Ministero dell’Università e della Ricerca Scientifica e Tecnologica (MURST—Cofin98) and Consiglio Nazionale delle Ricerche (CNR). We are greatly indebted to Dr. Jim Hainfeld, Nanoprobes, Inc., for his generous gift of Nanogold reagents and the GoldEnhance EM gold amplification reagent. REFERENCES 1.Anonymous. 1996. Nonradioactive in situ hybridization application manual. Boehringer Mannheim Biochemica, Mannheim. 2.Aimar, P., I. Barajon, and A. Merighi. 1998. Ultrastructural features and synaptic connections of NADPH-diaphorase positive neurons in the rat spinal cord. Eur. J. Anat. 2:27-34. 3.Aimar, P., L. Pasti, G. Carmignoto, and A. Merighi. 1998. Nitric oxide-producing islet cells modulate the release of sensory neuropeptides in the rat substantia gelatinosa. J. Neurosci. 18:10375-10388. 4.Alvarez, F.J., A.M. Kavookjian, and R. Light. 1992. Synaptic interactions between GABA-immunoreactive profiles and the terminals of functionally defined myelinated nociceptors in the monkey and cat spinal cord. J. Neurosci. 12:2901-2917. 5.Anderson, K.D., E.J. Karle, and A. Reiner. 1994. A pre-embedding triple-label electron microscopic immunohistochemical method as applied to the study of multiple inputs to defined tegmental neurons. J. Histochem. Cytochem. 42:49-56. 6.Basbaum, A.I. 1989. A rapid and simple silver enhancement procedure for ultrastructural localization of the retrograde tracer WGAapoHRP-Au and its use in double-label studies with postembedding immunocytochemistry. J. Histochem. Cytochem. 37:18111815. 7.Beier, K. and H.D. Fahimi. 1985. Automatic determination of labeling density in protein A-gold immunocytochemical preparations using an image analyzer. Application to peroxisomal enzymes. Histochemistry 82:99-100. 8.Belecky-Adams, T., B. Cook, and R. Adler. 1996. Correlation between terminal mitosis and differentiated fate of retinal precursor cells in vivo and in vitro: analysis with the “window-labeling” technique. Dev. Biol. 178:304-315. 9.Bendayan, M. 1984. Protein A-gold electron microscopic immunocytochemistry: methods, applications, and limitations. J. Electron Microsc. Tech. 1:243-270.

162

10.Bendayan, M. 1987. Introduction of the protein Ggold complex for high-resolution immunocytochemistry. J. Electron Microsc. Tech. 6:7-13. 11.Bendayan, M. 1995. Colloidal gold postembedding immunocytochemistry. Prog. Histochem. Cytochem. 29:1-159. 12.Bendayan, M., R.F. Barth, D. Gingras, I. Londono, P.T. Robinson, F. Alam, D.M. Adalms, and L. Mattiazzi. 1989. Electron spectroscopic imaging for high resolution immunocytochemistry: use of boronated protein A. J. Histochem. Cytochem. 37:573-580. 13.Bendayan, M., A. Nanci, and F.W.K. Kan. 1987. Effect of tissue processing on colloidal gold cytochemistry. J. Histochem. Cytochem. 35:983-996. 14.Bendayan, M. and H. Sthephens. 1984. Double labeling cytochemistry applying the protein A-gold technique, p. 143-154. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam. 15.Bendayan, M. and M. Zollinger. 1983. Ultrastructural localization of antigenic sites on osmium-fixed tissues appliying the protein A-gold technique. J. Histochem. Cytochem. 31:101-109. 16.Bienz, K., D. Egger, and L. Pasamontes. 1986. Electron microscopic immunocytochemistry. Silver enhancement of colloidal gold marker allows double labeling with the same primary antibody. J. Histochem. Cytochem. 34:1337-1342. 17.Blackstad, T.W., T. Karagulle, and O.P. Ottersen. 1990. Morforel, a computer program for two-dimensional analysis of micrographs of biological specimens, with emphasis on immunogold preparations. Comput. Biol. Med. 20:15-34. 18.Bolam, J.P., C.A. Ingham, and A.D. Smith. 1984. The section-Golgi-impregnation procedure-3. Combination of Golgi-impregnation with enzyme histochemistry and electron microscopy to characterize acetylcholinesterase-containing neurons in the rat neostriatum. Neuroscience 12:711-718. 19.Bouwens, L. and G. Kloppel. 1994. Cytochemical localization of NADPH-diaphorase in the four types of pancreatic islet cell. Histochemistry 101:209-214. 20.Broman, J. and F. Ådahl. 1994. Evidence for vesicular storage of glutamate in primary afferent terminals. Neuroreport 5:1801-1804. 21.Broman, J., J. Westman, and O.P. Ottersen. 1990. Ascending afferents to lateral cervical nucleus are enriched in glutamate-like immunoreactivity: a combined anterograde transport-immunogold study in the cat. Brain Res. 520:178-191. 22.Brorson, S.H. and F. Skjorten. 1996a. Improved technique for immunoelectron microscopy. How to prepare epoxy resin to obtain approximately the same immunogold labeling for epoxy sections as for acrylic sections without any etching. Micron 27:211-217. 23.Brorson, S.H. and F. Skjorten. 1996b. The theoretical relationship of immunogold labeling on acrylic sections and epoxy sections. Micron 27:193-201. 24.Casasco, A., A. Calligaro, and M. Casasco. 1992. Proliferative and functional stages of rat ameloblast differentiation as revealed by combined immunocytochem-

Immunogold Labeling Techniques istry against enamel matrix protein and bromodeoxyuridine. Cell Tissue Res. 270:415-423. 25.Causton, B.E. 1984. The choice of resins for electron immunocytochemistry, p. 29-36. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam. 26.Chang, J.P. and M. Yokama. 1970. A modified section freeze substitution technique. J. Histochem. Cytochem. 18:683-684. 27.Chang, L.M.S. and F.J. Bollum. 1986. Molecular biology of terminal transferase. Crit. Rev. Biochem. 21:2752. 28.Cheng, P.Y., L.Y. Liu Chen, C. Chen, and V.M. Pickel. 1996. Immunolabeling of Mu opioid receptors in the rat nucleus of the solitary tract: extrasynaptic plasmalemmal localization and association with Leu5enkephalin. J. Comp. Neurol. 371:522-536. 29.Cheng, P.Y., A. Moriwaki, J.B. Wang, G.R. Uhl, and V.M. Pickel. 1996. Ultrastructural localization of muopioid receptors in the superficial layers of the rat cervical spinal cord: extrasynaptic localization and proximity to Leu5-enkephalin. Brain. Res. 731:141-154. 30.Cruz, F., D. Lima, and A. Coimbra. 1987. Several morphological types of terminal arborizations of primary afferents in laminae I-II of the rat spinal cord, as shown after HRP labeling and Golgi impregnation. J. Comp. Neurol. 261:221-236. 31.Darius, S., G. Wolf, P.L. Huang, and M.C. Fishman. 1995. Localization of NADPH-diaphorase/nitric oxide synthase in the rat retina: an electron microscopic study. Brain Res. 690:231-235. 32.De Zeeuw, C.I., J.C. Holstege, F. Calkoen, T.J.H. Ruigrok, and J. Voogd. 1988. A new combination of WGA-HRP anterograde tracing and GABA immunocytochemistry applied to afferents of the cat inferior olive at the ultrastructural level. Brain Res. 447:369375. 33.Didenko, V.V. and P.J. Hornsby. 1996. Presence of double-strand breaks with single-base 3′ overhangs in cells undergoing apoptosis but not necrosis. J. Cell Biol. 135:1369-1376. 34.Doerr-Schott, J. and M.C. Lichte. 1986. A triple ultrastructural immunogold staining method. Application to the simultaneous demonstration of three hypophyseal hormones. J. Histochem. Cytochem. 34:1101-1104. 35.Dudek, R.W. and A.F. Boyne. 1986. An excursion through the ultrastructural world of quick-frozen pancreatic islets. Am. J. Anat. 175:217-243. 36.Dudek, R.W., A.F. Boyne, and M. Freinkel. 1981. Quick-freeze fixation and freeze-drying of isolated rat pancreatic islets: application to the ultrastructural localization of inorganic phosphate in the pancreatic beta cell. J. Histochem. Cytochem. 29:321-325. 37.Dudek, R.W., G.V. Childs, and A.F. Boyne. 1982. Quick-freezing and freeze drying in preparation for high quality morphology and immunocytochemistry at the ultrastructural level: application to pancreatic beta cells. J. Histochem. Cytochem. 30:129-138. 38.Dudek, R.W., I.M. Varndell, and J.M. Polak. 1984. Quick-freeze fixation and freeze drying for electron

microscopic immunocytochemistry, p. 235-248. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam. 39.Eneström, S. and B. Kniola. 1990. Quantitiative ultrastructural immunocytochemistry using a computerized image analysis system. Stain Technol. 65:263-278. 40.Eneström, S. and B. Kniola. 1995. Resin embedding for quantitative immunoelectron microscopy. A comparative computerized image analysis. Biotech. Histochem. 70:135-146. 41.Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemistry 8:1081-1083. 42.Fujimoto, K. 1995. Freeze-fracture replica electron microscopy combined with SDS digestion for cytochemical labeling of integral membrane proteins. Application to the immunogold labeling of intercellular junctional complexes. J. Cell. Sci. 108:3443-3449. 43.Fujimoto, K., M. Umeda, and T. Fujimoto. 1996. Transmembrane phospholipid distribution revealed by freeze-fracture replica labeling. J. Cell. Sci. 109:24532460. 44.Gavrieli Y., Y. Sherman, and S.A. Ben-Sasson. 1992. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119:493-501. 45.Geuze, H.J., J.W. Slot, R.C.T. Scheffer, and P.A. van der Ley. 1981. Use of colloidal gold particles in double-labeling immunoelectron microscopy of ultrathin frozen sections. J. Cell Biol. 89:653-665. 46.Gingras, D. and M. Bendayan. 1994. Compartmentalization of secretory proteins in pancreatic zymogen granules as revealed by immunolabeling on cryo-fixed and molecular distillation processed tissue. Biol. Cell 81:153-163. 47.Gold, R., M. Schmidt, G. Rothe, H. Zischler, H. Breitschopf, H. Wekerle, and H. Lassmann. 1993. Detection of DNA fragmentation in apoptosis: application of in situ nick translation to cell culture systems and tissue sections. J. Histochem. Cytochem. 41:10231030. 48.Gomez, S. and A. Boyde. 1994. Correlated alkaline phosphatase histochemistry and quantitative backscattered electron imaging in the study of rat incisor ameloblasts and enamel mineralization. Microsc. Res. Tech. 29:29-36. 49.Gong, L.W., Y.-Q. Ding, D. Wang, H.X. Zheng, B.Z. Qin, J.S. Li, T. Kaneko, and N. Mizuno. 1997. GABAergic synapses on mu-opioid receptor-expressing neurons in the superficial dorsal horn: an electron microscopy study in the cat spinal cord. Neurosci. Lett. 227:33-36. 50.Gracy, K.N. and V.M. Pickel. 1996. Ultrastructural immunocytochemical localization of the N-methyl-Daspartate receptor and tyrosine hydroxylase in the shell of the rat nucleus accumbens. Brain. Res. 739:169181. 51.Gundersen, V., O. Shupliakov, L. Brodin, O.P. Ottersen, and J. Storm-Mathisen. 1995. Quantification of excitatory amino acid uptake at intact glutamatergic synapses by immunocytochemistry of exoge-

163

Gold and Silver Staining nous D-aspartate. J. Neurosci. 15:4417-4428. 52.Gustincich, S., A. Feigenspan, D.K. Wu, L.J. Koopman, and E. Raviola. 1997. Control of dopamine release in the retina: a transgenic approach to neural networks. Neuron 18:723-726. 53.Hacker, G.W. and G. Danscher. 1994. Recent advances in immunogold-silver staining-autometallography. Cell Vision 1:102-109. 54.Hacker, G.W., W.H. Muss, C. Hauser-Kronberger, C. Dansher, R. Rufner, J. Gu, H. Su, A. Andreasen, M. Stoltenberg, and O. Dietze. 1996. Electron microscopical autometallography: immunogold-silver staining (IGSS) and heavy-metal histochemistry. Comp. Meth. Enzymol. 10:257-269. 55.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4 nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40:177184. 56.Halton, D.W. and G.P. Brennan. 1997. Ultrastructural localization of peptides using immunogold labeling. Methods Mol. Biol. 73:293-303. 57.Hayat, M.A. 1977. Principles and Techniques of Electron Microscopy, Vol. 1. Biological Applications. Edward Arnold, New York. 58.Holstege, J.C. and J.M. Vrensen. 1988. Anterograde tracing in the brain using autoradiography and HRPhistochemistry. A comparison at the ultrastructural level. J. Microscopy 150:233-243. 59.Horisberger, M. and J. Rosset. 1977. Colloidal gold, a useful marker for transmission and scanning electron microscopy. J. Histochem. Cytochem. 25:295-305. 60.Howell, K.E., U. Reuter-Carlson, E. Devaney, J.P. Luzio, and S.D. Fuller. 1987. One antigen one gold? A quantitative analysis of immunogold labeling of plasma membrane 5′-nucleotidase in frozen thin sections. Eur. J. Cell Biol. 44:318-326. 61.Hulser, D.F., B. Rehkopf, and O. Traub. 1997. Dispersed and aggregated gap junction channels identified by immunogold labeling of freeze-fractured membranes. Exp. Cell. Res. 233:240-251. 62.Isobe, Y., M. Nakatsugawa, G.R. Hou, and L.F. Lemanski. 1994. Three-dimensional distribution of desmin and vimentin in cultured hamster cardiomiocytes using the immunogold deep-etching replica technique. Histochemistry 101:155-168. 63.Juiz, J.M., M.L. Campos, R.H. Helfert, and R.A. Altschuler. 1996. Silver intensification of immunocolloidal gold on ultrathin plastic sections applied to the study of the neuronal distribution of GABA and glycine. J. Hirnforsch. 37:51-56. 64.Kechagias, S. and J. Broman. 1994. Compartmentation of glutamate and glutamine in the lateral cervical nucleus: further evidence for glutamate as a spinocervical tract neurotransmitter. J. Comp. Neurol. 340:531540. 65.Keller, G.A., K.T. Tokuyasu, A.H. Dutton, and S.J. Singer. 1984. An improved procedure for immunoelectronmicroscopy: ultrathin plastic embedding of immunolabeled ultrathin frozen sections. Proc. Natl. Acad. Sci. USA 81:5744-5747. 66.Kerr, J.F., A.H. Wyllie, and A.R. Currie. 1972. Apoptosis: a basic biological phenomenon with wide-rang-

164

ing implications in tissue kinetics. Br. J. Cancer 26:239-257. 67.Knyihar-Csillik, E., B. Csillik, and A.B. Oestreicher. 1992. Light and electron microscopic localization of B50 (GAP43) in the rat spinal cord during transganglionic degenerative atrophy and regeneration. J. Neurosci. Res. 32:93-109. 68.Knyihar-Csillik, E., B. Csillik, and P. Rakic. 1982. Ultrastructure of normal and degenerating glomerular terminals of dorsal root axons in the substantia gelatinosa of the Rhesus monkey. J. Comp. Neurol. 210:357-375. 69.Kondo, K. and T. Makita. 1996. Electron microscopic immunogold labeling of bromodeoxyuridine (BrdU) in routine electron microscopy. Acta Histochem. Cytochem. 29:115-119. 70.Krenacs, T. and L. Dux. 1994. Silver-enhanced immunogold labeling of calcium-ATPase in sarcoplasmic reticulum of skeletal muscle [letter, comment]. J. Histochem. Cytochem. 42:967-968. 71.Krenacs, T., M. van Dartel, E. Lindhout, and M. Rosendaal. 1997. Direct cell/cell communication in the lymphoid germinal center: connexin43 gap junctions functionally couple follicular dendritic cells to each other and to B lymphocytes. Eur. J. Immunol. 27:1489-1497. 72.Larsson, L.-I. 1981. Peptide immunocytochemistry. Prog. Histochem. Cytochem. 13:1-85. 73.Larsson, L.-I. 1983. Methods for immunocytochemistry of neurohormonal peptides, p. 147-209. In A. Björklund and T. Hökfelt, (Eds.), Handbook of Chemical Neuroanatomy, Vol. 1. Methods in Chemical Neuroanatomy. Elsevier, Amsterdam. 74.Larsson, L.-I. 1996. Immunogold labeling of neuroendocrine peptides with special reference to antibody specificity and multiple staining techniques. Histochem. Cell. Biol. 106:93-103. 75.Light, A.R. and A.M. Kavookjian. 1988. Electron microscopic localization of peptide-like immunoreactivity in labeled dorsal root terminals in the spinal substantia gelatinosa of the monkey, p. 57-69. In F. Cervero, G.J. Bennett, and P.M. Headley (Eds.), Processing of Sensory Information in the Superficial Dorsal Horn of the Spinal Cord. Plenum Press, New York. 76.Merighi A., L. Lossi, and S. Mioletti. 2000. Undifferentiated progenitors and granule cell precursors undergo apoptosis within a few hours of their generation, but represent an endogenous source of brain-derived neurotropic factor for the postnatal cerebellar cortex. Soc. Neurosci. 26:598. 77.Lossi, L. and A. Merighi. Cell cycle kinetic of proliferating and apoptotic granule cell progenitors in the rabbit postnatal cerebellum. 78.Lossi, L., D. Zagzag, M.A. Greco, and A. Merighi. 1998. Apoptosis of undifferentiated progenitors and granule cell precursors in the postnatal human cerebellar cortex correlates with expression of BCL-2, ICE and CPP-32 proteins. J. Comp. Neurol. 399:359-372. 79.Lue, J.H., Y.F. Jiang Shieh, J.Y. Shieh, and C.Y. Wen. 1996. The synaptic interrelationships between primary afferent terminals, cuneothalamic relay neurons and GABA-immunoreactive boutons in the rat cuneate

Immunogold Labeling Techniques nucleus. Neurosci. Res. 24:363-371. 80.Majdoubi, M.E., M.H. Metz Boutigue, P. Garcia Sablone, D.T. Theodosis, and D. Aunis. 1996. Immunocytochemical localization of chromogranin A in the normal and stimulated hypothalamo-neurohypophysial system of the rat. J. Neurocytol. 25:405-416. 81.Martin, R., C. Hoover, S. Grimme, C. Grogan, J. Höltke, and C. Kessler. 1990. A highly sensitive, nonradioactive DNA labeling and detection system. BioTechniques 9:762-768. 82.Maxwell, D.J., W.H. Christie, A.D. Short, J. StörmMathisen, and O.P. Ottersen. 1990. Central boutons of glomeruli in the spinal cord of the cat are enriched with L-glutamate-like immunoreactivity. Neuroscience 36:83-104. 83.Maxwell, D.J., W.M. Christie, O.P. Ottersen, and J. Störm-Mathisen. 1990. Terminals of group Ia primary afferent fibers in Clarke’s column are enriched with Lglutamate-like immunoreactivity. Brain Res. 510:346350. 84.Maxwell, D.J., W.M. Christie, A.D. Short, and A.G. Brown. 1990. Direct observations of synapses between GABA-immunoreactive boutons and muscle afferent terminals in lamina VI of cat spinal cord. Brain Res. 530:215-222. 85.Maxwell, D.J., R. Kerr, E. Jankowska, and J.S. Riddell. 1997. Synaptic connections of dorsal horn group II spinal interneurons: synapses formed with the interneurons and by their axon collaterals. J. Comp. Neurol. 380:51-69. 86.Maxwell, D.J., O.P. Ottersen, and J. Storm-Mathisen. 1995. Synaptic organization of excitatory and inhibitory boutons associated with spinal neurons which project through the dorsal columns of the cat. Brain Res. 676:103-112. 87.Mee, A.P., L.K. Davenport, J.A. Hoyland, M. Davies, and E.B. Mawer. 1996. Novel and sensitive detection systems for the vitamin D receptor—in situ-reverse transcriptase-polymerase chain reaction and immunogold cytochemistry. J. Mol. Endocrinol. 16:183-195. 88.Menco, B.P., A.M. Cunningham, P. Qasba, N. Levy, and R.R. Reed. 1997. Putative odor receptors localize in cilia of olfactory receptor cells in rat and mouse: a freeze-substitution ultrastructural study. J. Neurocytol. 26:297-312. 89.Merighi, A. 1992. Postembedding electron microscopic immunocytochemistry, p. 51-87. In J.M. Polak and J.V. Priestly (Eds.), Electron Microscopic Immunocytochemistry. Oxford University Press, London. 90.Merighi, A., F. Cruz, and A. Coimbra. 1992. Immunocytochemical staining of neuropeptides in terminal arborization of primary afferent fibers anterogradely labeled and identified at light and electron microscopic levels. J. Neurosci. Methods 42:105-113. 91.Merighi, A. and J.M. Polak. 1993. Postembedding immunogold staining, p. 229-264 In A.C. Cuello, (Ed.), Immunohistochemistry II. John Wiley & Sons, London. 92.Merighi, A., J.M. Polak, and D.T. Theodosis. 1991. Ultrastructural visualization of glutamate and aspartate immunoreactivities in the rat dorsal horn with special reference to the co-localization of glutamate, substance

P, and calcitonin gene-related peptide. Neuroscience 40:67-80. 93.Merighi, A., E. Raviola, and R.F. Dacheux. 1996. Connections of two types of flat cone bipolars in the rabbit retina. J. Comp. Neurol. 371:164-178. 94.Migheli, A., A. Attanasio, and D. Schiffer. 1995. Ultrastructural detection of DNA strand breaks in apoptotic neural cell by in situ end-labeling techniques. J. Pathol. 176:27-35. 95.Miller, M.A., C.M. Mazewski, Y. Sheikh, L.M. White, G.A. Yanik, D.M. Hyams, B.C. Lampkin, and A. Raza. 1991. Simultaneous immunohistochemical detection of IUdR and BrdU infused intravenously to cancer patients. J. Histochem. Cytochem. 39:407-412. 96.Morris, J.F., D.V. Pow, and F.D. Shaw. 1989. Strategies for the ultrastructural study of peptide containing neurons, p. 83-124. In G. Fink and A.J. Harmar (Eds.), Neuropeptides: A Methodology. John Wiley & Sons, Chichester. 97.Negyessy, L., J. Takacs, I. Divac, and J. Hamori. 1994. A combined Golgi and postembedding GABA and glutamate electron microscopic study of the nucleus dorsomedialis thalami of the rat. Neurobiology 2:325-341. 98.Nirenberg, M.J., J. Chan, R.A. Vaughan, G.R. Uhl, M.J. Kuhar, and V.M. Pickel. 1997. Immunogold localization of the dopamine transporter: an ultrastructural study of the rat ventral tegmental area. J. Neurosci. 17:5255-5262. 99.Nishikawa, S. and F. Sasaki. 1995. DNA localization in nuclear fragments of apoptotic ameloblasts using anti-DNA immunoelectron microscopy: programmed cell death of ameloblasts. Histochem. Cell Biol. 104:151-159. 100.Ottersen, O.P. and C.R. Bramham. 1988. Quantitative electron microscopic immunocytochemistry of excitatory amino acids, p. 93-100. In E.A. Cavalheiro, J. Lehmann, and L. Turski (Eds.), Frontiers in Excitatory Amino Acids Research. Alan R. Liss, New York. 101.Pease, D.C. 1964. Histological Techniques for Electron Microscopy. Academic Press, New York. 102.Phend, K.D., A. Rustioni, and R.J. Weinberg. 1995. An osmium-free method of EPON embedment that preserves both ultrastructure and antigenicity for postembedding immunocytochemistry. J. Histochem. Cytochem. 43:283-292. 103.Pickel, V.M., J. Chan, and C. Aoki. 1993. Electron microscopic immunocytochemical labeling of endogenous and/or transported antigens in rat brain using silver intensified one-nanometre colloidal gold, p. 265280. In A.C. Cuello (Ed.), Immunohistochemistry II. John Wiley & Sons, Chichester. 104.Posthuma, G., J.W. Slot, and H.J. Geuze. 1984. Immunocytochemical assays of amylase and chymotrypsinogen in rat pancreas secretory granules. Efficacy of using immunogold-labeled ultrathin cryosections to estimate relative protein concentrations. J. Histochem. Cytochem. 32:1028-1034. 105.Posthuma, G., J.W. Slot, and H.J. Geuze. 1987. Usefulness of the immunogold technique in quantitation of a soluble protein in ultra-thin sections. J. Histochem. Cytochem. 35:405-411.

165

Gold and Silver Staining 106.Priestley, J.V. 1984. Pre-embedding ultrastructural immunocytochemistry: immunoenzyme techniques, p. 37-52. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam. 107.Priestley, J.V. and A.C. Cuello. 1989. Ultrastructural and neurochemical analysis of synaptic input to trigemino-thalamic projection neurones in lamina I of the rat: a combined immunocytochemical and retrograde labeling study. J. Comp. Neurol. 285:467-486. 108.Pulczynski, S. and O. Myhre Jensen. 1994. Quantitation of immunogold with an interactive image analysis system: a new, practical approach to antibodyinduced modulation, internalization, and intracellular transport of surface antigens in viable hematopoietic cells. Anal. Quant. Cytol. Histol. 16:393-399. 109.Qualmann, B., M.M. Kessels, F. Klobasa, P.W. Jungblut, and W.D. Sierralta. 1996. Electron spectroscopic imaging of antigens by reaction with boronated antibodies. J. Microsc. 183:69-77. 110.Rèthelyi, M., A.R. Light, and E.R. Perl. 1989. Synaptic ultrastructure of functionally and morphologically characterized neurons of the superficial spinal dorsal horn of cat. J. Neurosci. 9:1846-1863. 111.Ribeiro-Da-Silva, A., J.V. Priestley, and A.C. Cuello. 1993. Pre-embedding ultrastructural immunocytochemistry, p. 181-228. In A.C. Cuello (Ed.), Immunohistochemistry II. John Wiley & Sons, Chichester. 112.Robards, A.W. and U.B. Sleytr. 1985. Low temperature methods in biological electron microscopy, p. 1551. In A.M. Glauert (Ed.), Practical Methods in Electron Microscopy. Elsevier, Amsterdam. 113.Roth, J. 1982a. The preparation of protein A-gold complexes with 3 nm and 15 nm gold particles and their use in labeling multiple antigens on ultrathin sections. Histochem. J. 14:791-801. 114.Roth, J. 1982b. The protein A-gold (pAg) technique—a qualitative and quantitative approach for antigen localization on thin sections, p. 107-134. In G.R. Bullock and P. Petrusz (Eds.), Techniques in Immunohistochemistry, Vol. I. Academic Press, London. 115.Roth, J. 1984. The protein A-gold technique for antigen localization in tissue sections by light and electron microscopy, p. 113-122. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam. 116.Roth, J. 1996. The silver anniversary of gold: 25 years of the colloidal gold marker system for immunocytochemistry and histochemistry. Histochem. Cell. Biol. 106:1-8. 117.Rufner, R., N.E. Carson, M. Forte, G. Danscher, J. Gu, and G.W. Hacker. 1995. Autometallography for immunogold-silver staining in light and electron microscopy. Cell Vision 2:327-333. 118.Saha, S., T.F. Batten, and P.N. McWilliam. 1995. Glutamate-immunoreactivity in identified vagal afferent terminals of the cat: a study combining horseradish peroxidase tracing and postembedding electron microscopic immunogold staining. Exp. Physiol. 80:193202. 119.Sato, Y., T. Ito, A. Nozawa, and M. Kanisawa. 1995.

166

Bromodeoxyuridine and iododeoxyuridine double immunostaining for epoxy resin sections. Biotech. Histochem. 70:169-174. 120.Schmitz, G.G., T. Walter, R. Seibl, and C. Kessler. 1991. Non-radioactive labeling of oligonucleotides in vitro with the hapten-digoxigenin by tailing with terminal tranferase. Ann. Biochem. 192:222-231. 121.Schuman, E.M. and D.V. Madison. 1994. Nitric oxide and synaptic function. Annu. Rev. Neurosci. 17:153-183. 122.Schwarz, C. and Y. Schmitz. 1997. Projection from the cerebellar lateral nucleus to precerebellar nuclei in the mossy fiber pathway is glutamatergic: a study combining anterograde tracing with immunogold labeling in the rat. J. Comp. Neurol. 381:320-334. 123.Sekerkova, G., Z. Katarova, F. Joo, J.R. Wolff, S. Prodan, and G. Szabo. 1997. Visualization of betagalactosidase by enzyme and immunohistochemistry in the olfactory bulb of transgenic mice carrying the LacZ transgene. J. Histochem. Cytochem. 45:1147-1155. 124.Sippel, A.E. 1973. Purification and characterization of adenosine triphosphate: ribonucleic acid adenyl transferase from Escherichia coli. Eur. J. Biochem. 37:31-40. 125.Slot, J.W., G. Posthuma, L.Y. Chang, J.D. Crapo, and H.J. Geuze. 1989. Quantitative aspects of immunogold labeling in embedded and non-embedded sections. Am. J. Anat. 185:271-281. 126.Somogyi, P. 1990. Synaptic connections of neurones identified by golgi impregnation: characterization by immunocytochemical, enzyme histochemical, and degeneration methods. J. Electron Microsc. Tech. 15:332-351. 127.Somogyi, P., K. Halasy, J. Somogyi, J. StörmMathisen, and O.P. Ottersen. 1986. Quantification of immunogold labeling reveals enrichment of glutamate in mossy and parallel fibre terminals in cat cerebellum. Neuroscience 19:1045-1050. 128.Song, Z.M., S.J. Brookes, and M. Costa. 1994. Characterization of alkaline phosphatase-reactive neurons in the guinea-pig small intestine. Neuroscience 63:1153-1167. 129.Stirling, J.W. 1990. Immuno- and affinity probes for electron microscopy: a review of labeling and preparation techniques [review]. J. Histochem. Cytochem. 38:145-158. 130.Stirling, J.W. 1993. Use of tannic acid and silver enhancer to improve staining for electron microscopy and immunogold labeling. J. Histochem. Cytochem. 41:643-648. 131.Stirling, J.W. and P.S. Graff. 1995. Antigen unmasking for immunoelectron microscopy: labeling is improved by treating with sodium ethoxide or sodium metaperiodate, then heating on retrieval medium. J. Histochem. Cytochem. 43:115-123. 132.Takizawa, T. and J.M. Robinson. 1994. Use of 1.4nm immunogold particles for immunocytochemistry on ultra-thin cryosections. J. Histochem. Cytochem. 42:1615-1623. 133.Tang, F.R., C.K. Tan, and E.A. Ling. 1995. The distribution of NADPH-d in the central grey region (lamina X) of rat upper thoracic spinal cord. J. Neurocytol.

Immunogold Labeling Techniques 24:735-743. 134.Tapia, F.J., I.M. Varndell, L. Probert, J. De Mey, and J.M. Polak. 1983. Double immunogold staining method for the simultaneous ultrastructural localization of regulatory peptides. J. Histochem. Cytochem. 31:977-981. 135.Theodosis, D.T., D.B. Chapman, C. Montagnese, D.A. Poulain, and J.F. Morris. 1986. Structural plasticity in the hypothalamic supraoptic nucleus at lactation affects oxitocin but not vasopressin-secreting neurons. Neuroscience 17:661-678. 136.Theodosis, D.T., L. Paut, and M.L. Tappaz. 1986. Immunocytochemical analysis of the GABAergic innervation of oxytocin- and vasopressin-secreting neurons in the rat supraoptic nucleus. Neuroscience 19:207-222. 137.Thiry, M. 1992. Highly sensitive immunodetection of DNA on sections with exogenous terminal deoxynucleotidyl transferase and non-isotopic nucleotide analogues. J. Histochem. Cytochem. 40:411-419. 138.Thiry, M. 1993. Immunodetection of RNA on ultrathin sections incubated with polydenylated nucleotidyl tranferase. J. Histochem. Cytochem. 41:657-665. 139.Thiry, M. 1995. Nucleic acid compartmentalization within the cell nucleus by in situ transferase-immunogold techniques. Microsc. Res. Tech. 31:4-21. 140.Tokuyasu, K.T. 1984. Immuno-cryoultramicrotomy, p. 71-82. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam. 141.Torrisi, M.R. and P. Mancini. 1996. Freeze-fracture immunogold labeling. Histochem. Cell. Biol. 106:1930. 142.Torunsciolo, D.R.Z., R.E. Schimdt, and K.A. Roth. 1995. Simultaneous detection of TdT-mediated dUTP-biotyn nick end-labeling (TUNEL) positive cells and multiple immunohistochemical markers in single tissue sections. BioTechniques 19:800-805. 143.Van Bockstaele, E.J. and E.E. Colago. 1996. Selective distribution of the NMDA-R1 glutamate receptor in astrocytes and presynaptic axon terminals in the nucleus locus coeruleus of the rat brain: an immunoelectron microscopic study. J. Comp. Neurol. 369:483496. 144.van Lookeren Campagne, M., A.B. Oestreicher, T.P. Van Der Krift, W.H. Gispen, and A.J. Verkleij. 1991. Freeze-substitution and lowicryl HM20 embedding of fixed rat brain: suitability for immunogold ultrastructural localization of neural antigens. J. Histochem. Cytochem. 39:1267-1279. 145.Varndell, I.M. and J.M. Polak. 1984. Double immunostaining procedures: techniques and applications, p. 155-177. In J.M. Polak and I.M. Varndell (Eds.), Immunolabeling for Electron Microscopy. Elsevier, Amsterdam.

146.Vincent, S.R. 1995. Localization of nitric oxide neurons in the central nervous system, p. 83-102. In S.R. Vincent (Ed.), Nitric Oxide in the Nervous System. Academic Press, London. 147.Wakayama, Y., M. Murahashi, T. Jimi, H. Kojima, S. Shibuya, and H. Oniki. 1997. Immunogold and freeze etch electron microscopic studies of merosin localization in basal lamina of human skeletal muscle fibers. Acta Neuropathol. (Berl.) 93:34-42. 148.Wang, B.L. and L.-I. Larsson. 1985. Simultaneous demonstration of multiple antigens by indirect immunofluorescence or immunogold staining. Novel light and electron microscopical double and triple staining method employing primary antibodies from the same species. Histochemistry 83:47-56. 149.Wharton, J., S. Gulbenkian, A. Merighi, D.M. Kuhn, R. Jahn, K.M. Taylor, and J.M. Polak. 1988. Immunocytochemical and ultrastructural localization of peptide-containing nerves and myocardial cells in the human atrial appendage. Cell Tissue Res. 254:155166. 150.Wolf, G., S. Würdig, and G. Schünzel. 1992. Nitric oxide synthase in rat brain is predominantly located at neuronal endoplasmic reticulum: an electron microscopic demonstration of NADPH-diaphorase activity. Neurosci. Lett. 147:63-66. 151.Wood, K.A., B. Dipasquale, and R.J. Youle. 1993. In situ labeling of granule cells for apoptosis-associated DNA fragmentation reveals different mechanisms of cell loss in developing cerebellum. Neuron 11:621632. 152.Wyllie, A.H. 1980. Glucocorticoid-induced tymocyte apoptosis is associated with endogenous endonuclease activation. Nature 284:555-556. 153.Wyllie, A.H., R.G. Morris, A.L. Smith, and D. Dunlop. 1984. Chromatin cleavage in apoptosis: association with condensed chromatin morphology and dependence on macromolecular synthesis. J. Pathol. 142:67-77. 154.Yanik, G.A., N. Yousuf, M.A. Miller, S.H. Swerdlow, B.C. Lampkin, and A. Raza. 1992. In vivo determination of cell cicle kinetics of non-Hodgkin’s lymphomas using iododeoxyuridine and bromodeoxyuridine. J. Histochem. Cytochem. 40:723-728. 155.Zehbe, I., G.W. Hacker, H. Su, C. Hauser Kronberger, J.F. Hainfeld, and R. Tubbs. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, streptavidin-Nanogold, and silver acetate autometallography: detection of single-copy human papillomavirus. Am. J. Pathol. 150:1553-1561. 156.Zehbe, I., J.F. Sallstrom, M. Evander, K. Edlund, E. Rylander, G. Wadell, and E. Wilander. 1996. Nonradioisotopic detection and typing of human papillomaviruses by use of polymerase chain reaction and single-strand conformation polymorphism. Diagn. Mol. Pathol. 5:206-213.

167

11

Pre-Embedding Immunoelectron Microscopy with Nanogold® Immunolabeling, Silver Enhancement, and Its Stabilization by Gold Hajime Sawada and Michiyo Esaki

INTRODUCTION Immunohistochemistry and related techniques are versatile methods to elucidate the localization of various molecules, thereby giving insight into their functions.23 Although the techniques are applicable for both light and electron microscopy (LM and EM), EM immunohistochemistry is often difficult to perform, and the interpretation of the results is challenging.20 In our laboratory, we have been searching for a simpler method for immunoelectron microscopy17 and possibly for correlative LM and EM.18 We found a pre-embedding technique using Nanogold® (Nanoprobes, Yaphank, NY, USA) that is very satisfactory both for sensitivity and for the preservation of ultrastructure. Since the first application of the technique, this and similar techniques have been tested by several groups, who found it to be useful.1,10,16,19,21,25 In this chapter, we introduce a routine protocol for pre-embedding immunoelectron microscopy, which is very similar in

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

technique to immunofluorescence microscopy and thus is easy to apply, has good sensitivity, and is relatively easy to interpret. PROTOCOLS Materials and Reagents • Nanogold-labeled secondary antibodies (Nanoprobes). • Chloroauric acid [hydrogen tetrachloroaurate(III)tetrahydrate]. (Nakarai Tesque, Kyoto, Japan). • Silver acetate (e.g., Nakarai Tesque). • Hydroquinone (e.g., Fuji Photo Film, Tokyo, Japan). Overview of the Whole Procedure In the following section, our most recent entire protocol for Nanogoldlabeled immunoelectron microscopy is given step by step (Table 11.1). The first part (steps 1 to 4) is very similar to LM immunohistochemistry on cryostat sections. Silver enhancement

169

Gold and Silver Staining Table 11.1. Schematic Protocol of Pre-Embedding Immunoelectron Microscopy 1. Fixation ↓ 2. Cryoprotection and freezing ↓ 3. Cryostat sectioning ↓ 4. Immunolabeling with primary antibodies and Nanogold-labeled secondary antibodies ↓ 5. Silver enhancement Take LM photomicrographs ↓ 6. Gold toning ↓ 7. Postfixation, dehydration, and substitution ↓ 8. Flat embedding in Epon Take LM photomicrographs for correlation ↓ 9. Ultrathin sectioning, staining, and observation Take EM photomicrographs

(autometallography; AMG) (step 5) and gold toning (step 6) are somewhat unique, but are not very difficult. Postfixation and dehydration (step 7) are almost the same as conventional EM. We devised an embedding step (step 8) in order to obtain photographs of the same area with LM and EM, taking advantage of the signal visible with both microscope types. The silver enhancement and gold toning steps (steps 5 and 6) are critical. For the remainder of the protocol, a preferred AMG method may be substituted, e.g., the silver- or gold-salt based AMG techniques described in Chapters 2 and 3. Procedure 1. Fixation: Tissues are fixed with periodate-lysine-paraformaldehyde (PLP) fixative (14) either by perfusion or immersion and left in the fixative for 2 to 6 h at 170

room temperature. A longer fixation period often deteriorates antigenicity. Other fixatives, such as 8% formaldehyde, or 0.2% glutardialdehyde, 2% formaldehyde in 0.1 M phosphate buffer24,25 can be used depending on the antigens and antibodies. Use of glutardialdehyde improves the structural preservation but mars antigenicity. 2. Cryoprotection and freezing: Immerse the tissues in a solution of 30 g sucrose added to 100 mL of phosphatebuffered saline (PBS) (thus, this is not 30%) for 1 h to overnight for cryoprotection. When the tissues sink to the bottom, they are ready for freezing. Freeze the tissues by dipping them in liquid nitrogen. The tissues can be stored for more than a year if they are kept at -70°C or colder, and dehydration is prevented by covering the surface with plastic wraps.

Pre-Embedding Immunoelectron Microscopy 3. Mounting in a cryostat and sectioning: In order to mount tissue specimens on cryostat specimen chucks, the tissues are thawed to room temperature and immersed in OCT compound (TissueTek; Miles Scientific, Chicago, IL, USA) for 5 to 10 min until the specimens fit well into the compound. Otherwise, cracks may form between the tissue specimens and surrounding OCT compounds during sectioning. The specimen with OCT compound is put in a small molding cup made with household aluminum foil, placed on a specimen chuck, and frozen in liquid nitrogen. Soaking for too long in liquid nitrogen produces cracks within the specimen. Therefore, we usually soak the specimen for 10 to 20 sec and then transfer it into a cryostat chamber and leave it on a cold stage cooled at about -30°C. The specimen is then set on the cryomicrotome, and sections are cut at -20°C. Sections between 6 to 20 µm thick are usual in our laboratory. The sections are then placed on either silane (2-aminopropyl-triethoxysilane)or poly-L-lysine-coated glass slides (both are commercially available, e.g., from Polysciences, Warrington, PA, USA) and air-dried with an electric fan for 10 min. 4. Immunolabeling: In order to minimize the use of precious antibodies, the sections are encircled with a peroxidase anti-peroxidase (PAP) pen (Agar Scientific Ltd., Essex, England, UK). After removing the OCT compound by immersing the glass slides in PBS, the sections are blocked with appropriate blocking solutions, such as 10% bovine serum albumin (BSA) in PBS, or 1% blocking reagent (Roche Molecular Biochemicals, Mannheim, Germany), or 50% calf serum. The sections are then treated with primary antibodies diluted in the blocking

solution (20–100 µL) for 2 h at room temperature with gentle shaking. After rinsing in PBS (with or without 0.1% Tween 20) 3 times for 5 min each, the sections are treated with Nanogold-labeled secondary antibodies diluted 1:50 to 1:100 in the blocking solution and incubated for 1 to 2 h at room temperature with gentle shaking. The sections are rinsed with PBS (with or without 0.1% Tween 20) 3 times for 5 min each and fixed with 1.2% glutardialdehyde, 2% formaldehyde in 0.1 M phosphate buffer, pH 7.4, for at least 30 min. The sections can be stored after this step in a refrigerator. 5. Silver enhancement:3,7,15 The sections are rinsed in double-distilled water (RDW) twice, since avoiding ions is essential for AMG. Do not use buffers such as phosphates that will make a precipitate with silver solution. Mix equal amounts of silver salt solution (solution A) and the reduction buffer (solution B) just before use. Apply the mixture to the sections.7,15 Solution A: 100 mg silver acetate in 50 mL RDW Solution B: 1.4 g trisodium citrate dihydrate, 1.5 g citric acid monohydrate, 0.25 g hydroquinone, add distilled water to 50 mL. Solutions A and B can be stored at -70°C in the dark for more than a month. The pH of solution B should be about pH 3.8; citric acid can be used to adjust the pH. If the pH is higher than 3.9, nonspecific reactions may occur. Although darkness is not necessary for silver acetate enhancement, it is advisable to avoid unnecessary illumination. Enhancement is usually done at room temperature for 10 to 15 min. The duration of enhancement is determined by observing the sections under an 171

Gold and Silver Staining inverted LM. An additional reference section is placed on the experimental slides, treated with a reference antibody, and processed with the experimental sections. The optimum enhancement time is judged from its color. Usually a light brown to brown colored signal is sufficient for EM. After appropriate enhancement, the sections are rinsed in RDW several times. If LM photomicrographs are needed, they should be taken at this step, since the visible signal made by silver enhancement may disappear after gold toning. 6. Gold toning:2,16,17 Since the silver shell made around the Nanogold particles is not stable in osmium tetroxide solution, it is recommended to replace it with gold. An aqueous solution of chloroauric acid (0.1%) is placed on the sections. Within 1 min, the brown-black color of silver disappears, and the sections acquire original color. We usually treat sections for 2 min and then quickly rinse them with RDW. Afterwards, the sections are washed in RDW about 5 times for 2 min each, followed by the osmium tetroxide step. If one stores the sections in RDW for a longer time, the sections will change in color to pink, even at 4°C. This seems to be the reason for very small background flecks often observed on many of the structures under EM. Although they can be distinguished from the signals by the difference in size, it is best to avoid them. 7. Postfixation, en bloc staining, and dehydration: The sections are treated with 1% osmium tetroxide in 0.1 M phosphate buffer, pH 7.4, for 1 to 2 h at 4°C as a routine specimen for EM. They can be stained en bloc with 0.5% uranyl acetate solution. Dehydration is performed with ascending concentrations of ethanol (e.g., 10 min each for 70%, 80%, 90%, 95%, 3 172

× 100% ethanol). The sections are then treated briefly in propylene oxide twice. 8. Embedding:18 It is very important that the sections do not dry. Being aware of this, take the glass slide from the propylene oxide and place Epon resin on top of sections as quick as possible. After 2 min, apply fresh Epon to the sections again and mount the sections between the glass slide and an Aclar film (Ted Pella, Redding, CA, USA) with spacers (such as 2 thicknesses of filter papers) to keep some distance between the Aclar film and the glass slide. One may use polyester overhead projector films (University Coop, Japan) in place of Aclar films. Use sufficient amount of Epon to fill the spaces. Place the glass slides on a Teflon oven sheet and polymerize the Epon films in a 60°C oven for 48 h. 9. Trimming and thin sectioning:18 After hardening, the Aclar films can easily be peeled off from Epon sheets with forceps and/or razor blades. The Epon surface is absolutely smooth and is suitable for observation under LM. Select the areas for EM observation under LM and mark them with a marker pen. Place the glass slides with the Epon sheets containing specimen on a heater block at 70° to 80°C. After a while, the Epon sheets will soften and then can be cut easily with a razor blade and peeled off the glasses.12 Photographs are taken at this step for correlative LM and EM. After trimming, the sections are glued to the specimen blocks for ultramicrotomy. It is advantageous to put the section side opposite to the side used for glueing. Then the final trimming is done, and ultrathin sections are cut from the blocks. We usually obtain usable sections containing tissues almost from the very beginning.

Pre-Embedding Immunoelectron Microscopy Mount the sections on copper grids, stain, and observe them as you would conventional thin section EM. Two electron micrographs taken with the present method are shown in Figures 11.1 and 11.2. TECHNICAL HINTS AND DISCUSSION Penetration of Probes into Specimens and Their Sensitivity Nanogold is a small probe, having a diameter of 1.4 nm, and even with conjugated Fab′ fragment, it is smaller than primary antibodies.8,9 It is one of the smallest immunoprobes commercially available at present and has excellent penetrability.11 When one uses cryostat sections, such Nanogold reagents can penetrate into the whole thickness of the sections,17 i.e., into extracellular spaces, cellular cytoplasm, and even into nuclear matrices. Although Nanogold may not penetrate through intact plasma membranes and other cyto-

plasmic membranes, one can use detergents such as Triton X-100 to obtain better accessibility of antigens and probes without significant deterioration of ultrastructures.25 Using Nanogold, it was possible for us to reliably detect in the EM most of the antigens also detectable by indirect immunofluorescence LM. However, judging from the reaction with antilaminin antibodies, which labeled the basal laminae facing clefts in the tissues better than those in intact tissue blocks (data not shown), there is a limitation of diffusion even within 20-µm sections. Visibility of Probes after Silver Enhancement and Gold Toning Nanogold is not visible under conventional transmission EM,11 and enlarging by silver enhancement and stabilization by gold substitution compromised the usage of osmium tetroxide with structural preservation and visibility of signals. The drawback is the splitting of each signal into several grains, which hinders quantitative

Figure 11.1. Rat seminiferous tubule labeled with polyclonal antilaminin serum. Nanogold-labeled antirabbit IgG was used to detect the primary antibodies sitting on their antigenic sites of the tissue section. Many gold–silver grains are seen along the basal laminae of the seminiferous tubule [spermatogoniae (G), open arrow], outer surface of myoid cells (M), and lymphatic endothelial cells (L). In some areas, the basal lamina of the myoid cells and lymphatic endothelial cells can be distinguished (arrows). Some of the labels are composed of clusters of several gold particles (circles), probably derived from single silver-amplified Nanogold particles. Bar = 1 µm.

173

Gold and Silver Staining study. Use of a higher concentration of gold toning solutions (such as 1%) seemed to prevent splitting somehow, though this was not tested critically. There is also a possibility that the signals might disappear during gold toning. However, in our experience, only a few signals if any disappeared during this step. Consideration of Methods for Silver Enhancement Other silver enhancement methods, such as those using silver lactate, can be substituted for silver acetate AMG,5,22 but many of these techniques need darkness and are therefore inconvenient for routine use. Some silver enhancement methods are also known for partly unspecific reactions (argyrophilic–argentaffin type). For more information on this, see Chapters 2 and 3.

The inexact controllability of enhancement size make double labeling with two Nanogold layers difficult. In order to control the size of the amplified gold particles after AMG, we tried to control the temperature more accurately and to use alternative enhancement methods,4,13 but also with this, we have not been able to obtain uniform size of gold–silver particles yet.17 For double labeling, however, there are various methods that can be used in combination with Nanogold.11 Silver enhancements for more than 20 min, or twice enhancement, are not recommended because of high background visible even under LM. Other Applications Although most of our studies with the present methods were done on immuno-

Figure 11.2. A stereo paired photomicrograph of the interstitial area of the rat testis. Labeling was obtained with monoclonal anti-type VI collagen antibodies and subsequent Nanogold antimouse IgG. The section is 0.5 µm thick. Although type VI collagen fibrils are not visible in this preparation, one can trace their course from the labeling, which often shows a periodicity of about 100 nm (arrowheads). Their course is distinct from the fibrillar collagens with 64 nm periodicity (open arrows). C, cytoplasm of a fibroblastic cell. Bar = 500 nm.

174

Pre-Embedding Immunoelectron Microscopy

Figure 11.3. JY3 macrophage cells treated with biotinylated horseradish peroxidase (HRP) before fixation. After fixation, they were permeabilized with Triton X-100, treated with streptavidin–Nanogold, silver-enhanced, gold-substituted, postfixed, dehydrated, and embedded in Epon.25 The cells also had phagocytosed latex beads. Internalized Nanogold particles are seen along the membranes of phagosomes (P) and attached membranes. They are also found along the plasma membrane, but not along the membranes of nuclear envelopes or rough endoplasmic reticulum (arrowheads). N, nucleus. Bar = 1 µm.

electron microscopy, one can use Nanogold for any procedure that needs high resolution localization; for example, for internalization of labeled molecules (Figure 11.3), lectin histochemistry25 and in situ hybridization.6 REFERENCES 1.Aoki, T., H. Hagiwara, and T. Fujimoto. 1997. Peculiar distribution of fodrin in fat-storing cells. Exp. Cell Res. 234:313-320. 2.Arai, R., M. Geffard, and A. Calas. 1992. Intensification of labelings of the immunogold–silver staining method by gold toning. Brain Res. Bull. 28:343-345. 3.Baschong, W. and Y.D. Stierhof. 1998. Preparation, use, and enlargement of ultrasmall gold particles in immunoelectron microscopy. Microsc. Res. Tech. 42:66-79. 4.Bienz, K., D. Egger, and L. Pasamontes. 1986. Electron microscopic immunocytochemistry: Silver enhancement of colloidal gold marker allows double labeling with the same primary antibody. J. Histochem. Cytochem. 34:1337-1342. 5.Danscher, G. and I.O.R. Norgaard. 1983. Light microscopic visualization of colloidal gold on resinembedded tissue. J. Histochem. Cytochem. 31:13941398.

6.Hacker, G.W. 1998. High performance Nanogold-silver in situ hybridisation. Eur. J. Histochem. 42:111120. 7.Hacker, G.W., L. Grimelius, G. Danscher, G. Bernatzky, W. Muss, H. Adam, and J. Thurner. 1988. Silver acetate autometallography: an alternative enhancement technique for immunogold–silver staining (IGSS) and silver amplification of gold, silver, mercury, and zinc in tissues. J. Histotechnol. 11:213-221. 8.Hainfeld, J.F. 1987. A small gold-conjugated antibody label: improved resolution for electron microscopy. Science 236:450-453. 9.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40:177184. 10.Hasson, T., P.G. Gillespie, J.A. Garcia, R.B. Macdonald, Y. Zhao, A.G. Yee, M.S. Mooseker, and D.P. Corey. 1997. Unconventional myosins in inner-ear sensory epithelia. J. Cell Biol. 137:1287-1307. 11.Koeck, P.J.B. and K.R. Leonard. 1996. Improved immune double-labeling for cell and structural biology. Micron 27:157-165. 12.Kraft, L.M., K. Joyce, and E.D. D’Amelio. 1983. Removal of histological sections from glass for electron microscopy: use of Quetol 651 resin and heat. Stain Technol. 58:41-43. 13.Lah, J.J., D.M. Hayes, and R.W. Burry. 1990. A neutral pH silver development method for the visualization of 1-nanometer gold particles in pre-embedding electron microscopic immunocytochemistry. J. Histochem. Cytochem. 38:503-508. 14.McLean, I.W. and P.K. Nakane. 1974. Periodatelysine-paraformaldehyde fixative: a new fixative for immunoelectron microscopy. J. Histochem. Cytochem. 22:1077-1083. 15.Murata, F., S. Tsuyama, K. Ihida, N. Kashio, M. Kawano, and S.S. Li. 1992. Sulfated glycoconjugates demonstrated in combination with high iron diamine thiocarbohydrazide silver proteinate and silver acetate physical development. J. Electron Microsc. 41:14-20. 16.Pohl, K. and Y.D. Stierhoff. 1998. Action of Gold chloride (gold toning) on silver-enhanced 1nm gold markers. Microsc. Res. Tech. 42:59-65. 17.Sawada, H. and M. Esaki. 1994. Use of nanogold followed by silver enhancement and gold toning for postembedding immunolocalization in osmium-fixed, Epon-embedded tissues. J. Electron Microsc. 43:361366. 18.Sawada, H. and M. Esaki. 1998. A simple flat embedding method for the correlative light and electron microscopic immunocytochemistry. J. Electron Microsc. 47:535-537. 19.Sawada, H. and M. Esaki. 2000. A practical technique to postfix nanogold-immunolabeled specimens with osmium and to embed them in Epon for electron microscopy. J. Histochem. Cytochem. 48:493-498. 20.Sawada, H., H. Stukenbrok, D. Kerjaschki, and M.G. Farquhar. 1986. Epithelial polyanion-podocalyxin is found on the sides but not the soles of the foot processes of the glomerular epithelium. Am. J. Pathol. 125:309-318.

175

Gold and Silver Staining 21.Shin, B.C., K. Fujikura, T. Suzuki, S. Tanaka, and K. Takata. 1997. Glucose-transporter Glut3 in the rat placental barrier—a possible machinery for the transplacental transfer of glucose. Endocrinology 138:3997-4004. 22.Stierhof, Y.-D., B.R. Humbel, and H. Schwarz. 1991. Suitability of different silver enhancement methods applied to 1 nm colloidal gold particles: an immunoelectron microscopic study. J. Electron Microsc. Tech. 17:336-343. 23.Stirling, J.W. 1990. Immuno- and affinity probes for

176

electron microscopy: a review of labeling and preparation techniques. J. Histochem. Cytochem. 38:145157. 24.Tokuyasu, K.T. 1986. Application of cryoultramicrotomy to immunocytochemistry. J. Microsc. 143:139149. 25.Yoshida, K., M. Ono, and H. Sawada. 1999. Lipopolysaccharide-induced vacuoles in macrophages: their origin is plasma membrane-derived organelles and endoplasmic reticulum, but not lysosomes. J. Endotoxin Res. 5:1-28.

12

Gold Cluster Immunoprobes: Light and Electron Microscopy John M. Robinson, Toshihiro Takizawa, and Dale D. Vandré

INTRODUCTION Immunocytochemistry (ICC) is a valuable set of methodologies routinely used in basic and applied biological research and for diagnostic purposes in the pathology laboratory. The various ICC labeling procedures can be classified according to the detection system employed (i.e., fluorescence, chromogenic, or particulate) and whether labeling occurs before or after embedding the sample. ICC localization can be achieved at both the light microscopy (LM) and electron microscopy (EM) levels. Detection of cell or tissue antigens, by means of ICC, relies upon the specificity of the antigen–antibody interaction. The information obtained by immunolabeling, at the LM level, can be enormous and sufficient for many purposes. However, there are situations in which the increased spatial resolution afforded by the electron microscope is desirable. The two major methods employed in the detection of antigen–antibody binding at the EM level are particulate probes and chemical reactions that generate electron dense markers or markers that can be rendered electron dense (e.g., oxidative polymeriza-

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

tion of 3,3-diaminobenzidine [DAB]). The particulate probes most often used are colloidal gold. Colloidal gold particles, readily generated in the laboratory, are unstable in the presence of electrolytes. However, these particles can be stabilized by electrostatic adsorption of various macromolecules, including antibodies. It is the adsorption of macromolecules to colloidal gold generating specific probes that permits affinitylabeling procedures (ICC, lectin cytochemistry, etc.) (e.g., see References 8 and 16). Colloidal gold probes that are routinely used in immunoelectron microscopy range from 5 to 15 nm in diameter (for reviews see References1,4,16,21. Smaller colloidal gold probes (1–3 nm diameter) are also in use.2,3,27 These latter ultrasmall gold probes are useful in situations where greater penetration of the gold into a sample is desirable. The drawback to these ultrasmall colloidal gold particles is that they are difficult to visualize in routine EM of cells and tissues. Indeed, the 1 nm particles must have their size enhanced for routine observations. This size increase is usually achieved with silverbased autometallography (AMG; also referred to as silver enhancement).5,6,10 However, size enhancement of ultrasmall

177

Gold and Silver Staining gold probes with gold has also been introduced.14 An alternative to ultrasmall colloidal gold is represented by metal–cluster compounds (e.g., gold–cluster compounds).11–13 These gold–cluster probes (e.g., Nanogold; Nanoprobes, Yaphank, NY, USA) are discrete chemical compounds in the 1 nm size range. They differ in several ways from colloidal gold. For example, antibodies (or a variety of other molecules) can be covalently linked to gold cluster compounds rather than association with colloidal gold through electrostatic interactions. We have used Nanogold (NG) and fluorescently labeled NG, which is known as FluoroNanogold (FNG; Nanoprobes), in several studies that combine LM and EM level ICC. The methods we have employed form the basis of this chapter. We will review three different labeling techniques in which gold–cluster immunoprobes have been employed. These are, (1) labeling microtubules in cells heavily extracted with detergent, (2) labeling the membrane protein caveolin-1 in cells gently permeabilized with detergent, and (3) labeling ultrathin cryosections of human neutrophils for correlative LM and EM. PROTOCOLS Materials and Reagents Most of the chemicals and supplies referred to are routinely used in cell biological or biochemical laboratories. Thus, only manufacturers of the more specialized reagents and supplies will be mentioned. Gelatin (type A), gum Arabic, and bovine IgG were from Sigma (St. Louis, MO, USA). The monoclonal mouse anti-human myeloperoxidase (clone MPO-7) was from DAKO (Glostrup, DK). Monoclonal mouse anti-α-tubulin and β-tubulin were from Amersham Pharmacia Biotech 178

(Piscataway, NJ, USA). The chicken anticaveolin-1 was prepared in one of our laboratories (J.M.R.). Nanogold and FluoroNanogold labeled with affinity-purified goat antimouse Fab′ or streptavidin was obtained from Nanoprobes. Conventional secondary antibodies labeled with fluorescein isothiocyanate (FITC) or biotin were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA, USA). The Polymorphoprep cell isolation medium was from Accurate Chemicals (Westbury, NY, USA). Mowiol was obtained from Polysciences (Warrington, PA, USA). Maxtaform “finder” grids were from Graticules (Tonbridge, Kent, England, UK). Protocol 1. LM and EM Level Localization of Tubulin in Leukocyte Microtubules Preparation of Human Leukocytes 1. Isolate human leukocytes from freshly drawn heparanized blood from healthy donors as we have described.7 2. Layer blood samples onto Polymorphoprep and centrifuge at 2000 rpm for 30 min at room temperature using an HS-4 rotor in a RC 2-B centrifuge (Sorvall, Newtown, CT, USA). 3. Carefully remove the cellular layer containing neutrophils and eosinophils and equilibrate with an equal volume of 0.45% NaCl. 4. Collect the mononuclear cell fraction in a similar fashion. 5. Equilibrate both cell fractions with excess Hank’s balanced salt solution (HBSS) buffer (5.37 mmol/L KCl, 5.55 mmol/L glucose, 0.44 mM anhydrous KH2PO4, 137 mM NaCl, 0.34 mM anhydrous Na2HPO4, 4.17 mM NaHCO3, 0.1% bovine serum albumin [BSA], pH 7.2).

Gold Cluster Immunoprobes 6. Gently pellet cells in a clinical centrifuge. 7. Resuspend the resulting pellet in 1 mL of distilled water for 30 sec in order to lyse contaminating red blood cells. 8. Mix cells with 12 mL HBSS to restore isotonicity. 9. Pellet isolated cells and resuspend in neutrophil buffer [10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 150 mM NaCl, 5 mM KOH, 1.2 mM MgCl2, 1.3 mM CaCl2, 5.5 mM glucose, pH 7.5]. 10. Allow cells to adhere to clean glass coverslips for 10 min at 37°C. 11. Use cells attached to coverslips in immunolabeling procedures. Immunocytochemical Labeling 1. Rinse cells in phosphate-buffered saline (PBS) (149 mM NaCl, 1.5 mM KH2PO4, 2.7 mM KCl, 6.5 mM Na2HPO4, pH 7.4). 2. Fix cells in 0.7% glutaraldehyde in PBS for 15 min at 22°C. 3. Rinse cells at least 3 times in PBS. 4. Extract fixed cells with 0.5% sodium dodecyl sulfate (SDS) in PBS for 15 min at 22°C. An extraction step with 0.5% Triton X-100 in PBS for 15 min prior to SDS treatment can also be used. 5. Wash cells at least 5 times with PBS. 6. Reduce unreacted aldehyde groups by treatment with NaBH4 [1–2 mg/mL in Tris-buffered saline (TBS) (0.9% NaCl, 10 mM Tris-base, pH 7.4) over a 20–30 min period with 2 changes of freshly prepared NaBH4]. It should be noted that NaBH4 in TBS generates copious bubbles. Care must be taken to prevent coverslips from floating. 7. Wash cells 3 times in PBS.

8. Minimize nonspecific antibody binding by incubation with 4% normal goat serum in PBS for 1 h at 22°C. 9. Incubate cells with primary antibody for 2 h at 22°C. In our experiments, a mixture of clones DM1A and DM1B, mouse monoclonal antibodies specific for α- and β-tubulin, respectively, were used. The antibody concentration used is not given since the optimal concentration should be determined independently in each laboratory employing these methods. 10. Wash cells in 6 to 8 changes of PBS for a period of 1 h. 11. Incubate cells for 2 h at 22°C with secondary antibodies diluted in PBS containing 4% normal goat serum. In our studies, we have used conventional fluorescently labeled goat antimouse antibodies as well as the small immunogold probes NG and FNG. As with primary antibodies, the optimal concentration of secondary antibodies should be determined independently. 12. Wash cells in PBS 6 to 8 times for a period of 1 h. Examination of Cells by Fluorescence Microscopy 1. Mount coverslips containing cells in Mowiol medium containing p-phenylenediamine (1 mg/mL) as an antiphotobleaching agent on microscope slides. 2. Examine cells labeled with conventional fluorescently tagged secondary antibodies or with FNG by fluorescence microscopy. Silver Enhancement of Nanogold or FluoroNanogold 1. Wash coverslips 3 times in 50 mmol/L MES buffer (2-[N-morpholino]-ethanesulfonic acid) (pH 6.15) to remove chlorine from the samples. 179

Gold and Silver Staining 2. Carry out the silver enhancement reaction using the method of Burry5 for varying times (e.g., 1–15 min). The reaction mixture consists of 50 mmol/L MES buffer (pH 6.15) containing 50% gum Arabic, 0.03% n-propylgallate, and 0.11% silver lactate. 3. Do the silver enhancement reaction at 22°C under red safe light conditions in a dark room. 4. Stop the silver enhancement reaction by treatment with a neutral pH fixer solution for 15 min. This solution consists of 250 mM sodium thiosulfate and 20 mM HEPES, pH 7.4. 5. Wash the samples 3 times with PBS. 6. Refix cells with 2% glutaraldehyde in PBS for 10 min at 22°C. 7. Wash cells 3 times in PBS. Examination of Cells by Brightfield LM 1. Following the silver enhancement reaction, mount coverslips on microscope slides in Mowiol lacking antiphotobleaching agents. 2. Examine cells by brightfield LM.

Protocol 2. Immunoelectron Microscopic Localization of Caveolin-1 in Cultured HUVEC following Pre-Embedding Labeling Preparation of Cells 1. Grow human umbilical vein endothelial cells (HUVEC) in M199 medium supplemented with 20% fetal bovine serum (FBS), 0.2% HEPES, 0.5% cow brainderived endothelial cell growth factor, and 0.02% penicillin–streptomycin and use between passages 3 and 8. Grow cells on sterilized 12-mm glass coverslips in 24-well culture dishes. 2. Rinse coverslips in warm PBS containing calcium, magnesium, and glucose (PBS+) (138 mM NaCl, 2.7 mM KCl, 16.2 mM Na2HPO4, 1.47 mM KH2PO4, 0.90 mM CaCl2, 0.5 mM MgCl2, 7.5 mM D-glucose, pH 7.35) to remove serum. 3. Fix cells in 4% paraformaldehyde freshly prepared in PBS lacking calcium, magnesium, and glucose (PBS-) for 1 h at 22°C. 4. Wash cells in PBS- (at least 4 times).

Preparation of Cells for EM 1. Following silver enhancement, wash cells 3 times in 100 mM cacodylate buffer, pH 7.2. 2. Incubate cells in 0.1% OsO4 in cacodylate buffer for 15 min. More recently we have found that phosphate buffers lacking chloride are superior to cacodylate buffer in this step. 3. Wash the coverslips 3 times in cacodylate buffer. 4. Dehydrate cells in ethanol and embed in Epon using routine methods as we have described previously.19 5. Cut thin sections and collect on EM grids, then examine with or without heavy metal contrast staining. 180

Immunocytochemical Labeling of Cells 1. Permeabilize fixed cells by treatment with 0.1% saponin in PBS- for 15 min at 22°C. It must be noted that saponin permeabilization is reversible. Therefore, saponin (0.1%) must be included in all ICC solutions including the washing medium. 2. Incubate cells for 1 h at 22°C with PBScontaining 5% normal goat serum to block nonspecific protein binding. 3. Incubate cells with anti-caveolin-1 antibodies (in our experiments a chicken antibody that recognizes the N-terminal region of the molecule was used).

Gold Cluster Immunoprobes 4. Wash cells extensively (at least 8 times during a 1 h period with PBS- containing saponin). 5. Incubate cells for 2 h at 22°C with a donkey antichicken antibody to which biotin is conjugated. 6. Wash cells extensively as previously indicated. 7. Incubate cells with streptavidin-labeled FNG for 2 h at 22°C. 8. Wash the samples extensively in PBS- as previously indicated. Examination of Cells by Fluorescence Microscopy 1. Mount coverslips in Mowiol containing an antiphotobleaching agent on glass microscope slides (as discussed in Protocol 1). 2. Evaluate the samples for labeling efficiency by fluorescence microscopy before progressing to the EM portion of the preparation. Silver Enhancement of FluoroNanogold 1. Refix cells in 1.5 % glutaraldehyde in PBS- for 15 min at 22°C. 2. Wash the samples at least 4 times in PBS-. 3. Carry out the silver enhancement reaction (as discussed in Protocol 1). Light and Electron Microscopy 1. Prepare samples for examination by brightfield or other LM methods (e.g., phase contrast, differential interference contrast) by mounting coverslips in Mowiol lacking antiphotobleaching agents. 2. Prepare the samples for EM in the same manner as described previously (see Protocol 1).

Protocol 3. Procedure for Correlative Fluorescence and Electron Microscopy on Ultrathin Cryosectioned Neutrophils Preparation of Cells 1. Isolate cells from venous blood as described previously (see Protocol 1). Other cell isolation procedures can also be used.25,26 2. Fix cells for ICC (human neutrophils were fixed with 4% paraformaldehyde in 0.1 mol/L cacodylate buffer, pH 7.4, containing 5% sucrose or with the mixture of 4% paraformaldehyde and 0.1% glutaraldehyde in the same buffer for 120 min at 22°C). 3. Wash cells in 0.1 M cacodylate buffer, pH 7.4, containing 5% sucrose to stop fixation (at least 3 times). Preparation of Ultrathin Cryosections 1. Embed cells by centrifugation into 10% gelatin dissolved in the cacodylate wash buffer. 2. Solidify the gelatin at 4°C and cut it into small pieces. 3. Infiltrate gelatin pieces containing cells with a graded series of sucrose (0.17– 2.3 M) in 0.1 M cacodylate buffer, pH 7.4, overnight at 4°C. 4. Place cells in the gelatin pieces on specimen pins that can be mounted in the ultracryomicrotome. 5. Rapidly freeze the cells by plunging the specimen pins into a slush of liquid nitrogen. Store samples in liquid nitrogen until used. 6. Cut ultrathin cryosections (approximately 90 nm in thickness). Use these cryomicrotome settings as a guide: knife at -100°C, specimen at -110°C, and the chamber at -120°C. 7. Collect sections on formvar-coated nickel EM grids using a transfer solution 181

Gold and Silver Staining of 2 M sucrose containing 0.75% gelatin in 0.1 M cacodylate buffer (pH 7.4).25 (We use special finder grids to facilitate cell location by LM and EM.) Immunocytochemical Labeling on Ultrathin Cryosections with FluoroNanogold 1. Incubate grids in a blocking medium for 1 h at 22°C (the solution contains 1% nonfat dry milk and 1 mg/mL bovine IgG in PBS-). 2. Incubate sections with primary antibody for 3 to 4 h at 22°C. 3. Wash grids in PBS- at least 8 times. 4. Incubate grids with FNG as the secondary antibody for 3 to 4 h at 22°C. 5. Wash grids in PBS- at least 8 times. LM Examination of the Fluorescence Signal from FluoroNanogold 1. Immerse grids in a small drop of 50% glycerol/PBS- (pH 8.0) containing antiphotobleaching reagents, 1 mg/mL p-phenylenediamine or 10 mg/mL n-propyl gallate on a microscope slide. 2. Prepare a temporary slide by overlaying with a coverslip. 3. Examine the slide by epifluorescence and differential interference contrast (DIC) microscopy. 4. Record images of selected cell profiles. Silver Enhancement of FluoroNanogold 1. Disassemble the temporary slide preparations. 2. Wash grids in PBS- 3 times. 3. Wash grids in 50 mM MES buffer (pH 6.15) to remove chlorine (as mentioned in Protocol 1).

182

4. Subject the FNG-labeled sections to the silver enhancement reaction (as described in Protocol 1). 5. Wash grids in PBS-. 6. Refix sections with 2% glutaraldehyde in PBS- for 10 min at 22°C. 7. Wash grids in PBS-. Negative Staining and EM Examination of the Gold Signal from FluoroNanogold 1. Wash grids in distilled water. 2. Negatively stain the ultrathin sections in phosphotungstic acid by the method of Sakai et al.22 3. Stabilize the formvar supporting film with evaporated carbon. 4. Examine the grids with the EM. Record the same cell profiles that had been photographed by fluorescence microscopy in the EM. TECHNICAL HINTS AND DISCUSSION Methods for use of gold–cluster immunoprobes in three different ICC preparations have been presented. In the first case, microtubules have been localized in human leukocytes using a pre-embedding labeling procedure. This method is useful for both LM and EM level detection of these cytoskeletal elements. We have compared several preparative procedures (i.e., different fixatives and permeabilization conditions) for obtaining the optimal immunofluorescence labeling of leukocyte microtubules.7 We find that 0.7% glutaraldehyde fixation was the best of the fixatives that we tried. Moreover, we found that extraction of fixed leukocytes with SDS led to optimal labeling and was essential for successful immunolabeling of neutrophil microtubules. The

Gold Cluster Immunoprobes microtubule labeling intensity achieved using NG or FNG and a silver enhancement procedure compared favorably to that obtained with routine immunofluorescence labeling (Figure 12.1). The signal from silver-enhanced gold clusters is amenable to analysis by several imaging modalities of optical microscopy, including the reflectance mode in confocal microscopy.20 These preparations are more stable than material processed for immunofluorescence and are not subject to photobleaching as are many immunofluorescence preparations. However, it should be noted that immunofluorescence does have certain advantages over silver-enhanced gold. These include easier quantifi-

cation of the ICC signal with fluorescence and the analysis of thicker specimens by confocal microscopy. In summary, we find that labeling of leukocyte microtubules with gold–cluster immunoprobes, especially the fluorescently tagged ones, is a useful approach to the study of these structures by LM and EM. Localization of membrane associated proteins, with pre-embedding techniques, places greater constraints on the ICC procedures than does the localization of structural elements such as components of the cytoskeleton. This is because membranes are more susceptible to damage and alteration than are some other components of the cell. This is particularly so

Figure 12.1. Use of gold–cluster immunoprobes for LM and EM immunolocalization of leukocyte microtubules. (A) A human monocyte labeled with antibodies to tubulin and then NG was used as the reporter system. The sample was subsequently subjected to a silver enhancement reaction. Note that the microtubules are readily evident in this brightfield image. (B) The distribution of microtubules in a human monocyte (cell on left) and neutrophil (cell on right) as determined by conventional immunofluorescence. Note that the number of monocyte microtubules is similar to that observed in cells where they were detected with silver-enhanced gold– cluster probes. (C) An electron micrograph of a thin section from a human monocyte is shown. In this case, microtubules are detected with antibodies to tubulin and with the silverenhanced gold cluster immunoprobes (arrows). Identical EM results have been obtained with both NG and FNG. Note the extensive labeling along the microtubule segments present in this thin section.

183

Gold and Silver Staining in the context of detergents used for cell permeabilization. Excess permeabilization (i.e., extraction) can damage membranes dramatically, even when aldehyde fixatives are used. In our studies, localizing components associated with membranes or membrane bounded compartments, we utilize gentle permeabilization that still allows penetration of the immunological reagents. We routinely employ saponin (0.1%) for

this purpose. As noted earlier, saponin permeabilization is reversible and must be included in all of the solutions including the washing buffer. In this chapter, we present results on the ultrastructural localization of caveolin-1 in cultured endothelial cells. Caveolin-1 is a marker protein for specialized microdomains of the plasma membrane that are present in many cell types and are known

Figure 12.2. ICC localization of caveolin-1 in HUVEC. (A) The distribution of caveolin-1 in a whole HUVEC as determined by immunofluorescence. The fluorescence signal is localized to a large number of small punctate structures. (B) An electron micrograph of a cell prepared for routine morphological analysis (no ICC was conducted). This thin section, which is tangential to the cell surface, has a number of small vesicular structures that have the characteristic appearance of caveolae (arrows). (C) An electron micrograph of a HUVEC prepared for pre-embedding localization of caveolin-1. The detection system was silver-enhanced FNG. Note that the caveolae are heavily decorated with silver-enhanced gold particles (arrows).

184

Gold Cluster Immunoprobes as caveolae. These membrane microdomains have a characteristic ultrastructural appearance. Using gold–cluster immunoprobes, we have achieved extensive labeling of caveolae with a pre-embedding proce-

Figure 12.3. Localization of myeloperoxidase (MPO) in the same ultrathin cryosection of human neutrophils by fluorescence and EM using anti-MPO and then FNG as the reporter system. (A) A low magnification fluorescence image of an ultrathin cryosection; the FNG-signal is observed as granule-like fluorescent spots. (B) Low magnification electron micrograph of the same cells shown in panel A following the silver enhancement reaction and negative staining. A portion of an EM grid bar is shown (*).

dure (Figure 12.2). Indeed, the labeling that we achieve is more extensive than that reported previously. In other studies using ultrathin cryosections, the intensity of labeling for caveolins 1 and 2 with colloidal gold as the reporter system has been very low.23,24 The fact that we obtain such extensive labeling in our pre-embedding procedure underscores the idea that the small gold–cluster probes penetrate thoroughly into saponin-permeabilized cells and have high labeling efficiency. Correlative microscopy refers to the situation in which the same structure or label is examined by two or more imaging techniques in order to obtain the unique information provided by each of them. In cell biological studies, this often involves fluorescence and EM. The combination of a fluorochrome and colloidal gold into a single probe would appear to be the labeling reagent of choice for correlative ICC. However, few studies have used these single probes.15 The reason for this appears to be related to the reduction in, or complete

Figure 12.4. Comparing the fluorescence and silver-enhanced gold signals from FNG-labeled neutrophils. The cell was prepared in the same manner as in Figure 12.3. (A) Higher-magnification fluorescence image of a portion of the cell indicated with an arrow in the inset. Specific fluorescence spots showing the localization of MPO (arrows) are labeled 1 and 2. (Inset) Low magnification fluorescence image of the cell profile demonstrating the FNG signal. (B) Higher magnification electron micrograph of the same cell shown in the inset of panel A. Granule profiles containing silver-enhanced FNG (arrows) are labeled 1’ and 2’. (Inset) Low magnification electron micrograph of the same cell profile shown in the inset of panel A. Note that there is a one-to-one correspondence between the fluorescent spots and the FNG-labeled granules (i.e., the azurophillic granules). Other intracellular granules that are MPO negative are unlabeled (arrowheads) and a portion of the nucleus (n) is evident.

185

Gold and Silver Staining loss of, the fluorescence signal from these probes.9,18 Powell and colleagues17 report that this loss of fluorescence is due to fluorescence resonance energy transfer with the colloidal gold particles serving as the acceptor. An alternative approach to this problem is to construct a fluorescently tagged gold– cluster immunoprobe (i.e., FNG).18 We have verified that one can observe both the fluorescence signal and the silver-enhanced gold signal from FNG.20 We have also used the FNG probe in a more rigorous test of its ability to serve as a useful reagent in correlative fluorescence and EM level ICC. We used ultrathin cryosections with human neutrophils as the model system.26 In these experiments, ultrathin cryosections are collected on EM finder grids. They are then labeled with the primary antibody, which is followed with FNG. The labeled grids are initially observed with the fluorescence microscope, and images of selected cell profiles are collected. The grids are subsequently subjected to a silver enhancement reaction and then prepared for EM. We readily observe the fluorescence signal from these 90-nm thick sections and the silverenhanced gold signal (Figures 12.3 and 12.4). The fluorescence signal and the silver-enhanced gold signal display a one-toone correspondence, thus verifying that FNG can be used as a high resolution probe for correlative LM and EM. The examples presented herein illustrate that gold–cluster immunoprobes are valuable additions to the arsenal of ICC labeling reagents. These probes have high labeling efficiency when compared to colloidal gold and will penetrate into samples when colloidal gold (≥5 nm) fails to do so. In addition, the availability of fluorescently tagged gold–cluster immunprobes allows for high resolution correlative fluorescence and EM.

186

REFERENCES 1.Albrecht, R.M., O.E. Olorundare, S.R. Simmons, J.C. Loftus, and D.F. Mosher. 1992. Use of correlative microscopy with colloidal gold labeling to demonstrate platelet receptor distribution and movement. Methods Enzymol. 215:456-479. 2.Baschong, W., J.J. Lucocq, and J. Roth. 1985. “Thiocyanate gold”: small (2-3 nm) colloidal gold for affinity cytochemical labelling in electron microscopy. Histochemistry 83:409-411. 3.Baschong, W. and N.G. Wrigley. 1990. Small colloidal gold conjugated to Fab fragments or to immunoglobulin G as high-resolution labels for electron microscopy: a technical overview. J. Electron Microsc. Tech. 14:313-323. 4.Bendayan, M. 1995. Colloidal gold post-embedding ICC. Prog. Histochem. Cytochem. 29:1-159. 5.Burry, R.W. 1995. Pre-embedding immunocytochemistry with silver-enhanced small gold particles, p. 217230. In M.A. Hayat (Ed.), Immunogold-Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 6.Danscher, G., G.W. Hacker, C. Hauser-Kronberger, and L. Grimelius. 1995. Trends in autometallographic silver amplification of colloidal gold particles, p. 11-18. In M.A. Hayat (Ed.), Immunogold-Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 7.Ding, M., J.M. Robinson, B.C. Behrens, and D.D. Vandré. 1995. The microtubule cytoskeleton in human phagocytic leukocytes is a highly dynamic structure. Eur. J. Cell Biol. 66:234-245. 8.Geoghegan, W.D. and G.A. Ackerman. 1977. Adsorption of horseradish peroxidase, ovomucoid, and antiimmunoglobulin to colloidal gold for the indirect detection of wheat germ agglutinin and goat antihuman immunoglobulin G on cell surfaces at the electron microscopic level: a new method, theory, and application. J. Histochem. Cytochem. 25:1187-1200. 9.Goodman, S.L., K. Park, and R.M. Albrecht. 1991. A correlative approach to colloidal gold labeling with video-enhanced LM, low voltage scanning electron microscopy, and high-voltage electron microscopy, p. 369-409. In M.A. Hyatt (Ed.), Colloidal Gold: Principles, Methods, and Applications, Vol 3. Academic Press, San Diego. 10.Hacker, G.W., G. Danscher, L. Grimelius, C. HauserKronberger, W.H. Muss, A. Schiechl, J. Gu, and O. Dietze. 1995. Silver staining techniques with special reference to the use of different silver salts in light and electron microscopical immunogold-silver staining, p. 19-45. In M.A. Hayat (Ed.), Immunogold-Silver Staining: Principles, Methods, and Applications. CRC Press, Boca Raton. 11.Hainfeld, J.F. 1987. A small gold-conjugated antibody label: improved resolution for electron microscopy. Science 263:450-453. 12.Hainfeld, J.F. 1988 Gold cluster-labeled antibodies. Nature 333:281-282.

Gold Cluster Immunoprobes 13.Hainfeld, J.F. and F.R. Furuya. 1992. A 1.4-nm gold cluster covalently attached to antibodies improves immunolabeling. J. Histochem. Cytochem. 40:177184. 14.Hainfeld, J.F. et al., 1999. Gold-based autometallography. Microsc. Microanal. 5(Suppl 2):486-487. 15.Horisberger, M. 1981. Colloidal gold: a cytochemical marker for light and fluorescence microscopy and transmission electron microscopy. Scan. Electron Microsc. 2:9-31. 16.Horisberger, M. 1992. Colloidal gold and its application in cell biology. Int. Rev. Cytol. 136:227-287. 17.Powell, R.D., C.M.R. Halsey, and J.F. Hainfeld. 1998. Combined fluorescent and gold immunoprobes: reagents and methods for correlative light and electron microscopy. Microsc. Res. Tech. 42:2-12. 18.Powell, R.D, C.M.R. Halsey, D.L. Spector, S.L. Kaurin, J. McCann, and J.F. Hainfeld. 1997. A covalent fluorescent-gold immunoprobe: simultaneous detection of a pre-mRNA splicing factor by light and electron microscopy. J. Histochem. Cytochem. 45:947956. 19.Robinson, J.M., T. Okada, J.J. Castellot, and M.J. Karnovsky. 1986. Unusual lysosomes in aortic smooth muscle cells: presence in living and rapidly frozen cells. J. Cell Biol. 102:1615-1622. 20.Robinson, J.M. and D.D. Vandré. 1997. Efficient immunocytochemical labeling of leukocyte microtubules with FluoroNanogold: an important tool for correlative microscopy. J. Histochem Cytochem. 45:631-642.

21.Roth, J. 1996. The silver anniversary of gold: 25 years of the colloidal gold marker system for immunocytochemistry and histochemistry. Histochem. Cell Biol. 106:1-8. 22.Sakai, T., T. Saruwatari, O. Fukushima, and T. Saito. 1995. The covering method: an improved negative staining method for ultrathin cryosections of tissue. J. Electron Microsc. 44:479-484. 23.Scherer, P.E., R.Y. Lewis, D. Volonté, J.A. Engelman, F. Galbiati, J. Couet, D.S. Kohtz, E. van Donselaar, P. Peters, and M.P. Lisanti. 1997. Cell-type and tissuespecific expression of caveolin-2. Caveolins 1 and 2 colocalize and form a stable hetero-oligomeric complex in vivo. J. Biol. Chem. 272:29337-29346. 24.Stang, E., J. Kartenbeck, and R.G. Parton. 1997. Major histocompatibility complex class I molecules mediate association of SV40 with caveolae. Mol. Biol. Cell 8:47-57. 25.Takizawa, T. and J.M. Robinson. 1994. Composition of the transfer medium is crucial for high-resolution immunocytochemistry of cryosectioned of human neutrophils. J. Histochem. Cytochem. 42:1157-1159. 26.Takizawa, T., K. Suzuki, and J.M. Robinson. 1998. Correlative microscopy using fluoronanogold on ultrathin cryosections: proof of principle. J. Histochem. Cytochem. 46:1097-1106. 27.Van de Plas, P. and J.L.M. Leunissen. 1993. Ultrasmall gold probes: characteristics and use in immuno (cyto)chemical studies. Methods Cell Biol. 37:241257.

187

13

Highly Sensitive Ultrastructural Immunogold Detection Using Tyramide and Gold Enhancement Christian Schöfer and Klara Weipoltshammer

INTRODUCTION Immunoelectron microscopy using proteins conjugated with gold particles is the preferred tool for high-resolution detection of probing molecules at the electron microscopic level.4 The gold particles are electron dense and small in size, so they offer a spatial resolution superior to other detection methods such as the precipitation of diaminobenzidine (DAB).5 Colloidal gold grains with diameters of 5 to 20 nm are widely used. They can readily be recognized at moderate magnifications and are easily distinguishable from underlying structures even after staining of the specimen. This approach has limitations in sensitivity, which are especially apparent for the detection of molecules that are present in low copy numbers at the surface of ultrathin sections. Among the factors that might contribute to the fact that not all epitopes are detected are: (1) sterical hindrance of binding of the immunoreactive site of the antibody due to the conjugation with the comparatively large gold grain; (2) impairment of antigen binding by the charged surface of colloidal gold grains;

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

and (3) decoupling of gold grain and antibody during washes and staining. Two different strategies have been developed to increase the sensitivity of immunogold detection procedures at the ultrastructural level. In the first approach, the binding sites for the gold-labeled antibody are amplified. This became possible with the tyramide amplification technique also known as catalytic reporter deposition (CARD) or tyramide signal amplification (TSA; Perkin Elmer Life Sciences, Boston, MA, USA). The method was developed for biochemical immunoassays2 and adapted for light microscopy (LM)1,11 and electron microscopy (EM).10,12 It relies on the peroxidase-dependent activation of the phenolic part of tyramides. It is thought that these activated intermediates are short-lived and bind to electron rich moieties of proteins in close vicinity of a peroxidase-labeled antibody. Labeled tyramides (e.g., biotin-labeled tyramides) are then detected with the immunogold technique. Since the binding sites for immunogold detection are amplified, the probability of labeling more epitopes of the molecule in question is higher, which results in increased sensitivity.

189

Gold and Silver Staining The second strategy relies on the fact that the sensitivity of immunogold electron microscopy is strongly influenced by the size of the gold particle used. Therefore, smaller gold grains with diameters in the range of 1 nm (referred to as ultrasmall gold particles) are used to enhance sensitivity. The obvious reason is that the interfering effects of the gold particles for antigen binding are reduced if the grain size becomes smaller. Two different kinds of gold particles are used: colloidal gold grains of 1 nm diameter and Nanogold grains (Nanoprobes, Yaphank, NY, USA) of 1.4 nm diameters. For both types of gold label, it will be necessary to enlarge the particles by means of autometallography (AMG) in order to clearly discern them from the underlying structures.3 This procedure relies on the precipitation of noble metal ions onto the surface of gold particles, which causes them to grow in size. Silver ions are widely used for this purpose. In combination with immunogold technique, it has been designated as immunogold–silver staining (IGSS).6 Most recently, gold–ion-based AMG was introduced instead of silver– salt-based AMG.8 In the following, three protocols are given: (1) a standard EM embedding protocol suitable for performing postembedding immunogold detection; (2) a protocol for TSA/CARD amplification and detection with 5 nm colloidal goldlabeled antibodies; and (3) a protocol for deposition of gold onto 1.4 nm Nanogold particles. The latter protocol is intended to replace the detection step of the former protocol, thereby further enhancing sensitivity of the detection procedure. Of course it can also be applied independently from CARD. The protocols given are designed for immunogold detection at the surface of ultrathin sections. 190

PROTOCOLS Protocol 1. Processing of Samples for Immuneolectron Microscopy Materials and Reagents • Dissection instruments. • A shallow glass vessel with wide opening for dissections. • Small glass vessels with lids (approximately 2-mL volume) or microcentrifuge tubes. • Gelatin capsules or microcentrifuge tubes. • Crushed ice. • Incubator adjusted to 55°C. • Microtome for cutting ultrathin sections. • Gold grids for mounting the sections. If AMG is planned, use copper or nickel grids. • Phosphate buffer (0.1 M): prepare a 0.1 M NaH2PO4 solution and a 0.1 M Na2HPO4 solution. Titrate the latter solution by adding NaH2PO4 until a pH of 7.4 is reached. Put on ice. • Fixative: 4% formaldehyde plus 0.5% glutardialdehyde in 0.1 M phosphate buffer (pH 7.4). Preparation of fixative: adjust pH value of 25 mL 0.1 M phosphate buffer to above 8.0. Heat buffer to above 70°C (do not boil) and add 4 g paraformaldehyde. Stir and wait until solution becomes clear. Cool down, filter the solution, and adjust pH to 7.4. Mix 2.5 mL of this 16% fixative solution with 7.5 mL 0.1 M phosphate buffer and add 200 µL glutardialdehyde from a commercially available 25% solution. Prepare immediately before use and place on ice. Caution: Paraformaldehyde powder, formaldehyde, and glutardialdehyde are hazardous. Prepare fixative in a fume hood and wear gloves. • Ethanol 30%, 70%, 80%, 90%, and

Highly Sensitive Ultrastructural Immunogold Detection 100%. • Acrylic resin LR White medium grade (London Resin Company, London, England, UK). Caution: Irritating to the skin, wear gloves. Protocol for the Processing of Samples for Immunoelectron Microscopy 1. Excise small quantities of tissue and transfer it to the glass vessel filled with fixative. Try to keep the time lapse between biopsy and immersion of tissue into the fixative as short as possible for the benefit of structural preservation. Further dissect specimen in the fixative to pieces of about 1 mm3. Transfer the pieces to a small glass vessel (or microcentrifuge tube). Highly vascularized tissues may be shortly washed in phosphate buffer before transferring to the fixative. 2. Fix the tissue blocks in ample volumes of fixative for 30 min on ice. 3. Wash with phosphate buffer for 20 min on ice and repeat twice. 4. Dehydrate samples in a graded series of ethanol (30%, 70%, 80%, 90%, and 2 × 100%) for 15 min each on ice. 5. Incubate with a mixture of absolute ethanol and LR White resin (1:1 vol/vol; mix well) for 15 min on ice. 6. Incubate with pure LR White. Change the resin about 6 times with at least a 1 h interval. Allow infiltration of tissue blocks with LR White resin overnight at 4°C (e.g., 2 changes of LR White on the first day and 4 changes the next day). 7. For polymerization, transfer the samples into gelatin capsules or to a new microcentrifuge tube (the walls of the old one may have been etched by the resin resulting in incomplete polymerization). Fill up the entire volume of the capsule or tube with LR White and close the lid. Since LR White polymer-

izes only in the absence of oxygen, it is critical that the volume of trapped air under the lid is kept at a minimum. 8. Place tubes or capsules into an incubator adjusted to 55°C overnight. 9. Remove the polymerized blocks and cut ultrathin sections with an ultramicrotome. Mount sections on dried grids (store in ethanol to keep them clean). The mounted sections should be allowed to dry thoroughly in order to prevent the loss of sections during further handling. Protocol 2. Immunogold Detection Using CARD/TSA Amplification Materials and Reagents • Parafilm. • Filter paper. • Good quality forceps for handling grids. • Phosphate-buffered saline (PBS): 150 mM NaCl, 1.5 mM NaH2PO4, 15 mM Na 2HPO4, pH 7.4 (a 10× stock solution can be stored for weeks). • PBSB: 1% bovine serum albumin (BSA) plus 1% normal goat serum (NGS; optional) in PBS. Prepare two batches: one with pH 7.4 and another with pH 8.2. • PBT: PBS supplemented with 0.05% Tween 20. Prepare two batches: one with pH 7.4 and another with pH 8.2. • Primary antibody specific for the detection of the epitope in question. If no information about the application of this antibody at the EM level is available, it is helpful if it has been tested on Paraffin sections. This will at least tell if the antigenicity of the epitope is maintained after formaldehyde fixation. The optimal dilution for use in EM has to be determined experi191

Gold and Silver Staining mentally; usually the antibody concentration is higher than for LM. The antibody should be diluted with PBSB. • Secondary peroxidase-labeled antibody specific against the primary antibody. • TNT: 0.15 M NaCl, 0.1 M Tris-HCl, pH 7.5, 0.05% Tween 20. • CARD kit containing biotin-labeled tyramides, amplification diluent, and blocking reagent (TSA Biotin System, Perkin Elmer Life Sciences). • 5 nm colloidal gold-conjugated antibiotin antibody (BioCell, Cardiff, England, UK) diluted 1:20 in PBSB, pH 8.2. Other grain sizes (10 and 20 nm) may be used as well. Gold-labeled streptavidin can also be used. • Distilled water. • Aqueous uranyl acetate (4% in double-distilled water, filtered). Lead citrate may be used subsequently, but will make it more difficult to detect the gold grains. Protocol for Immunogold Detection Using CARD Amplification All steps can be performed in drops or floating on the surface of drops that have been placed on Parafilm. Volumes can be as little as 10 µL per drop (20 µL are more convenient for handling). Forceps of good quality are highly recommended for manipulation of the grids. In the following, a generally applicable protocol is given. See Results section for an experiment using specific primary and secondary antibodies. 1. Put grids in PBSB (pH 7.4) for 30 min. 2. Incubate grids with primary antibody for 1 h. 3. Wash grids with PBT (pH 7.4) for 3 times for 10 min. 4. Incubate grids with secondary antibody for 1 h. 192

5. Wash grids in PBT (pH 7.4) 3 times for 10 min. 6. Wash with TNT for 10 min. 7. Incubate grids with the supplied blocking reagent for 30 min. 8. Shortly rinse grids with TNT. 9. Prepare the tyramide reaction solution: dilute the labeled tyramides 1:50 with the amplification diluent. 10. Incubate grids with the tyramide reaction solution for 5 to 8 min. Note that the incubation time is critical for resolution and specificity of the amplification procedure. Optimal incubation times should be tested. 11. Wash grids immediately with TNT 2 times for 10 min. 12. Wash grids with PBT (pH 8.2) for 10 min. 13. Incubate grids with 5 nm colloidal goldlabeled anti-biotin antibody for 1 h. 14. Wash grids with PBT (pH 8.2) 3 times for 10 min. 15. Wash grids with distilled water 3 times for 2 min. 16. Incubate grids with 4% aqueous uranyl acetate for 20 min. 17. Rinse grids 3 times with double-distilled water and dry on filter paper. Protocol 3. Immunogold Detection Using CARD/TSA Amplification, Nanogold Detection, and Gold-Based AMG In the following, the procedure will be described as an alternative detection variant of the previous protocol. In principle, it can of course also be used without the CARD amplification steps. Materials and Reagents • All items of the previous protocol are required.

Highly Sensitive Ultrastructural Immunogold Detection • Nanogold-labeled streptavidin (Nanoprobes). Dilute 1:100 with PBSB, pH 7.4. Ultrasmall colloidal gold-conjugated antibodies can be used instead. • Gold enlargement kit (GoldEnhance EM formulation; Nanoprobes) containing so-called enhancer (solution A), activator (solution B), and initiator (solution C). • Sodium thiosulfate (1%). Protocol for Immunogold Detection Using CARD Amplification, Nanogold Detection, and Gold-Based AMG Start with steps 1 to 11 of the previous protocol: 12. Wash grids in PBT (pH 7.4) for 10 min. 13. Incubate with Nanogold-labeled streptavidin for 1 h. 14. Start preparing the GoldEnhance solution 15 min before use. Mix solution A and solution B 1:1 (vol/vol). Add solution C immediately before use (A:B:C = 1:1:1; vol/vol/vol). 15. Wash with PBT (pH 7.4) 3 times for 10 min. 16. Rinse with distilled water 3 times for 2 min. 17. Incubate grids with enlargement solution for 4 to 10 min. Incubation time is critical; the optimal time should be tested. 18. Rinse grids shortly with distilled water. 19. Incubate grids with sodium thiosulfate for 2 min. 20. Rinse grids with distilled water 3 times for 5 min. 21. Incubate grids in 4% aqueous uranyl acetate for 20 min. 22. Rinse grids 3 times with double-distilled water and dry on filter paper.

Controls Controls for the CARD reaction may include omitting the primary antibody, omitting the peroxidase-labeled antibody, omitting the tyramides, or replacing them with unlabeled tyramides. For controls of the gold enhancement reaction omit the primary antibody or the gold-labeled antibody. RESULTS Embedding in hydrophilic resins is favorable for performing reactions that take place in aqueous solutions. As an example, biopsies of mouse and human testes were embedded as outlined in the first protocol. Figure 13.1 shows a comparison of testicular cells embedded in LR White and in the hydrophobic resin Epon. Embedding in hydrophilic resins preserves the ultrastructure reasonably well, although Epon polymerizes more homogeneous, thus yielding better structural preservation than LR White. The described detection protocols were tested with an epitope of well-known distribution. DNA was detected in nuclei of elongated spermatids of human testis embedded in LR White. The primary antibody used was a monoclonal anti-DNA antibody (Roche Molecular Biochemicals, Basel, Switzerland). It was diluted 1:5 in PBSB. For the CARD reaction, the secondary antibody was a goat antimouse peroxidase-labeled antibody (Sigma, St. Louis, MO, USA). It was diluted 1:50 in PBSB. The incubation time for the CARD reaction was 6 min at room temperature. Figure 13.2 shows comparison of conventional immunogold detection and CARD. The increase in labeling is apparent after CARD, while specificity and 193

Gold and Silver Staining spatial resolution are comparable. For demonstration of Nanogold labeling and gold enhancement, Nanogold-labeled streptavidin was used after CARD (as above). It was diluted 1:40 in PBSB. The gold-based AMG reaction was allowed to proceed for 6 or 8 min at room temperature. Figure 13.3 shows the result and demonstrates the influence of prolonged incubation time on the growth of gold grains. TECHNICAL HINTS AND DISCUSSION The outlined protocols enhance the sensitivity of the immunogold detection at the EM level. The presented methods will be most helpful to obtain better results if conventional detection procedures result in low signal densities. In case immunoelectron microscopy with the presented methods (postembedding immunogold detection) does not lead

to satisfactory results, it might be necessary to perform a pre-embedding approach. Pre-embedding will be necessary if the antigenic sequences became inaccessible or deteriorated by the strong cross-linking fixative glutardialdehyde or by the resin used. In short, the immunoreaction and the immunogold detection are performed prior to embedding for EM. The application of CARD and ultrasmall gold in the preembedding approach has been demonstrated recently.9 Nevertheless, with this technique, penetration of gold-labeled antibodies into cells, and especially into nuclei, is a crucial problem. Besides, it is also a laborious procedure. In contrast, the postembedding approach offers good structural preservation, is comparatively easy to use, and will be the method of choice for many immunoreactions. The choice of fixative is extremely important for performing immunoreactions. Especially, the concentration of the strong cross-linking agent glutardialdehyde in the fixative is a limiting factor for

Figure 13.1. Comparison of elongated spermatids from mouse testis embedded in the hydrophilic resin LR White (A) and in the hydrophobic resin Epon (B). Both tissue samples were fixed as in the first protocol. No postfixation with osmium tetroxide was performed. LR White sections were stained in 4% aqueous uranyl acetate, and Epon sections were stained in 4% methanolic uranyl acetate. Note the better structural preservation in panel B, especially concerning the membranes. Calibration bar = 200 nm.

194

Highly Sensitive Ultrastructural Immunogold Detection preservation of the antigenicity of the protein sequence in question. Glutardialdehyde concentrations as low as 0.01% can be used, although some deterioration of ultrastructure occurs at this concentration. Formaldehyde concentration can be lowered to 2%. Never use osmium tetroxide as postfixative for the postembedding approach. In order to carry out the immunoreactions at the surface of sections, hydrophilic acryl-based resins are highly preferable for embedding the samples. Suitable products include Lowicryl (Polysciences, Warrington, PA, USA), LR Gold (London Resin Company), LR White, Unicryl (Polysciences), etc. Some of these resins are cured at high temperature (LR White) or at low temperature (Unicryl) or very low temperatures (Lowicryl, LR Gold, LR White) with UV radiation. Embedding at very low temperatures should be preferred if fixation-sensitive and temperaturelabile antigens and enzymes are studied. The outlined protocol for EM embedding of tissue blocks can of course also be used to embed cells, e.g., cultured monolayers. The cells need to be detached from culture flasks and spun down between

incubations in microcentrifuge tubes. As already mentioned, 5 to 10 nm colloidal gold conjugates are widely used as a detection system. They are easy to use but also have some disadvantages. The colloidal gold particle has a charged surface, and due to the small distance between the gold particle and the conjugated antibody with its antigen-reactive site, interference with antigen binding can occur. Furthermore, the forces involved in coupling the antibody to the gold particle decrease with the surface of the particle resulting in possible loss of gold label. The latter is particularly true for ultrasmall gold particles. In contrast, Nanogold particles are slightly larger, but are covalently bound to the antibody or streptavidin molecules. The efficiency of EM immunogold detection with colloidal gold particles can be enhanced if the pH value of the incubation solution is raised to about 8.0. Furthermore, NGS can be added, and the concentration of sodium ions in the antibody incubation solution can be raised to about 1 M. The use of ultrasmall gold labels increases the sensitivity of antigen binding but requires subsequent enlargement of the

Figure 13.2. Conventional and CARD-enhanced immunogold detection of DNA in human elongated spermatids at the surface of LR White-embedded ultrathin sections. A monoclonal anti-DNA antibody was used as outlined in Results. Conventional detection (A) with an antimouse 5 nm colloidal gold-labeled antibody. After tyramide signal enhancement (as in Protocols), a significant increase in the number of gold particles is obvious (B). Spatial resolution is only slightly lower than in conventional detection, which is expected for signal amplification methods (some label is located at the cytoplasmic side of the nuclear membrane). Calibration bar = 100 nm.

195

Gold and Silver Staining

Figure 13.3. Immunogold detection of DNA in human elongated spermatids with Nanogold and subsequent gold enhancement after CARD (see Protocols). Comparison of two different incubation times for gold enhancement. After 6 min of incubation at room temperature (A), the gold particles are sufficiently enlarged to be conveniently viewed at magnifications in the range of 20 000×. After 8 min of incubation at room temperature (B), individual gold grains fused to form large gold clusters. Calibration bar = 100 nm.

gold grains before viewing. The deposition of gold onto gold particles instead of silver has advantages in lower background staining and in the ease of handling. For example, gold toning, a procedure to stabilize the gold–silver complexes after silver deposition is no longer necessary because deposited gold is insensitive to low pH values present, e.g., during uranyl acetate staining. Nevertheless, silver ions for AMG have been successfully used at the EM level, yielding very good results. For appropriate protocols for performing IGSS.7,14 Obviously, the incubation time for gold amplification by AMG is critical. This is especially true for the detection of epitopes that are believed to occur highly clustered. After prolonged incubation times, the neighboring gold grains will merge, forming a large cluster and thus making the interpretation difficult. Performing a time series of gold enhancement should help in this respect. Incubation time is also crucial for the 196

CARD reaction. Too long reaction times lead to decreased spatial resolution and background staining. Again, performing a time series is recommended. If background is a problem, high molecular weight polymers such as dextran sulfate may be added to the reaction solution.10,12,13 Protocols for preparing labeled tyramides are available.1,10 The methods presented lead to higher labeling densities and increased sensitivity and are therefore most useful in applications where epitopes present in low copy number need to be detected at highest resolution. ACKNOWLEDGMENTS We thank M. Almeder for excellent technical assistance and are grateful to Drs. G.W. Hacker and C. Hauser-Kronberger for their helpful comments concerning the use of silver and gold for signal enhancement in immunogold EM.

Highly Sensitive Ultrastructural Immunogold Detection REFERENCES 1.Adams, J.C. 1992. Biotin amplification of biotin and horseradish peroxidase signals in histochemical stains. J. Histochem. Cytochem. 40:1457-1463. 2.Bobrow, M.N., T.D. Harris, K.J. Shaughnessy, and G.J. Litt. 1989. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J. Immunol. Methods 125:279-285. 3.Danscher, G. and S.O. Norgaard. 1983. Light microscopic visualization of colloidal gold on resin embedded tissue. J. Histochem. Cytochem. 31:394-398. 4.Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemistry 8:10081-10083. 5.Graham, R.C., Jr. and M.J. Karnovsky. 1966. The early stages of absorption of injected horseradish peroxidase in the proximal tubule of mouse kidney, ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem. 14:291-302. 6.Hacker, G.W., D.R. Springall, N.S. Van, A.E. Bishop, L. Grimelius, and J.M. Polak. 1985. The immunogold-silver staining method. A powerful tool in histopathology. Virchows Arch. A 406:449-461. 7.Hacker, G.W. 1998. High performance Nanogold-silver in situ hybridisation. Eur J. Histochem. 42:111120. 8.Hainfeld, J.F., R.D. Powell, J.L. Stein, G.W. Hacker, C. Hauser-Kronberger, A. Cheung, and C. Schöfer. 1999. Gold-based autometallography, p. 486-487. In G.W. Bailey, W.G. Jerome, S. McKernan, J.F. Mans-

field, and R.L. Price (Eds.), Proc. 57th Ann. Mtg., Micros. Soc. Amer. Springer, New York. 9.Humbel, B.M., M.D.M. De Jong, W.H. Muller, and A.J. Verkleij. 1998. Pre-embedding immunolabeling for electron microscopy: an evaluation of permeabilization methods and markers. Microsc. Res. Tech. 42:4358. 10.Mayer, G. and M. Bendayan. 1999. Immunogold signal amplification: application of the CARD approach to electron microscopy. J. Histochem. Cytochem. 47:421-429. 11.Raap, A.K., M.P.C. van de Corput, R.A.W. Vervenne, R.P.M. van Gijlswijk, H.J. Tanke, and J. Wiegant. 1995. Ultra-sensitive FISH using peroxidase-mediated deposition of biotin- or fluorochrome tyramides. Hum. Mol. Gen. 4:529-534. 12.Schöfer, C., K. Weipoltshammer, M. Almeder, and F. Wachtler. 1997. Signal amplification at the ultrastructural level using biotinylated tyramides and immunogold detection. Histochem. Cell Biol. 108:313-319. 13.van Gijlswijk, R.P.M., J. Wiegant, A.K. Raap, and H.J. Tanke. 1996. Improved localization of fluorescent tyramides for fluorescence in situ hybridization using dextran sulfate and polyvinyl alcohol. J. Histochem. Cytochem. 44:389-392. 14.Zehbe, I., G.W. Hacker, H. Su, C. Hauser-Kronberger, J.F. Hainfeld, and R. Tubbs. 1997. Sensitive in situ hybridization with catalyzed reporter deposition, streptavidin-nanogold, and silver acetate autometallography. Am. J. Path. 150:1553-1561.

197

14

In Situ Hybridization at the Electron Microscopic Level Christian Schöfer and Klara Weipoltshammer

INTRODUCTION The detection of nucleic acids at the ultrastructural level has a long standing history. Histochemical methods such as the EDTA-regressive staining1 to detect RNPs or the osmium-amine staining2,11 to detect DNA were among the first techniques introduced to electron microscopy (EM) to localize nucleic acids in general. Indirect detection procedures using enzymatic digestion with DNases and RNases offered an alternative approach. Later on, methods derived from molecular biology were added to the repertoire of methods such as the terminal deoxynucleotidyl transferase reaction for 3′ end labeling of DNA14 or in situ nick translation for detection of DNA.13 Immune reactions using an anti-DNA antibody offer an easy-to-use alternative for labeling DNA in EM.15 The above mentioned methods contain two limitations: (1) they can only be used to detect DNA or RNA in general; and (2) most of them label preferentially DNA rather than RNA. If the high resolution localization of specific sequences of nucleic acids is required, EM in situ hybridization (EM-ISH) needs to be performed. EMISH was introduced in the 1970s4,7 in combination with autoradiographic detec-

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

tion. Later on, nonradioactively labeled probes became available9 making detection techniques possible that are based on immune reactions. Diaminobenzidine (DAB) precipitation5 and later on goldconjugated antibodies3 were used to localize the probes. Particularly the latter method resulted in better spatial resolution and made the lengthy and hazardous procedure of autoradiography obsolete. EMISH can be used to detect RNAs and DNAs. In the case of DNA, the technique will practically be restricted to sequences occurring in higher copy numbers due to the obvious fact that an ultrathin section contains only a comparatively small amount of DNA. There are two different strategies for EM-ISH: (1) the reactions are carried out at the surface of ultrathin sections (postembedding); and (2) the ISH is essentially carried out as for light microscopy (LM) but using an immunogold detection system followed by embedding suitable for EM (pre-embedding). The postembedding technique has several advantages, such as better structural preservation and easier performance. The following protocols will outline techniques necessary for postembedding EM-ISH to detect DNA and RNA. Colloidal gold-labeled antibodies

199

Gold and Silver Staining (10 nm) are used to detect the bound probes. For further increasing the sensitivity of the detection system, see Chapter 13. PROTOCOLS Protocol 1. Processing of Samples for EM-ISH The protocol outlined below is designed for the embedding of tissue blocks. It can also be applied to cells, e.g., cultured monolayers. The cells need to be detached from culture flasks and then spun down between incubations in microcentrifuge tubes. Materials and Reagents • Dissection instruments. • A shallow glass vessel with wide opening for dissections. • Small glass vessels with lids (approximately 2-mL volume) or microcentrifuge tubes. • Gelatin capsules or microcentrifuge tubes. • Crushed ice. • Incubator adjusted to 55°C. • Microtome for cutting ultrathin sections. • Gold grids for mounting the sections. • RNase-free water (for EM-ISH to detect RNA only): prepare 0.1% diethylpyrocarbonate (DEPC; Fluka Chemical, Milwaukee, WI, USA) in distilled water. Incubate at 37°C for 2 h. A commercially available solution of DEPC is diluted 1:1000 in distilled water. Caution: Active DEPC is a suspected carcinogen. DEPC water can be autoclaved to reduce the risk of handling. • Phosphate buffer (0.1 M): prepare a 0.1 M NaH2PO4 solution and a 0.1 M Na2HPO4 solution. Titrate the latter 200

solution by adding NaH2PO4 until a pH of 7.4 is reached. Put on ice. Use RNase-free water if RNA EM-ISH is planned. • Fixative: 4% formaldehyde plus 0.5% glutardialdehyde in 0.1 M phosphate buffer (pH 7.4). Preparation of fixative: adjust pH value of 25 mL 0.1 M phosphate buffer to above 8.0. Heat buffer to above 70°C (do not boil) and add 4 g paraformaldehyde. Stir and wait until solution becomes clear. Cool down, filter the solution, and adjust pH to 7.4. Mix 2.5 mL of this 16% fixative solution with 7.5 mL 0.1 M phosphate buffer and add 200 µL glutardialdehyde from a commercially available 25% solution. Prepare immediately before use and place on ice. Caution: Paraformaldehyde powder, formaldehyde, and glutardialdehyde are hazardous. Prepare fixative in a fume hood and wear gloves. • Ethanol 30%, 70%, 80%, 90%, 100%. • Acrylic resin LR White medium grade (London Resin Company, London, England, UK). Caution: Irritating to the skin, wear gloves. Procedure 1. Excise small quantities of tissue and transfer it to the glass vessel filled with fixative. Try to keep the time lapse between biopsy and immersion of tissue in the fixative as short as possible for the benefit of structural preservation and RNA integrity. Further dissect the specimen in the fixative to pieces of about 1 mm3. Transfer the pieces to a small glass vessel (or microcentrifuge tube). Highly vascularized tissues may be shortly washed in phosphate buffer before transferring to the fixative. Fix the tissue blocks in ample volumes of fixative for 30 min on ice.

In Situ Hybridization at the Electron Microscopic Level 2. Wash with phosphate buffer for 20 min on ice and repeat twice. 3. Dehydrate samples in a graded series of ethanol (30%, 70%, 80%, 90%, 2× 100%) for 15 min each on ice. 4. Incubate with a mixture of absolute ethanol and LR White resin (1:1 vol/vol; mix well) for 15 min on ice. 5. Incubate with pure LR White. Change the resin about 6 times with at least a 1 h interval. Allow infiltration of tissue blocks with LR White resin over night at 4°C (e.g., 2 changes of LR White on the first day and 4 changes the next day). 6. For polymerization, transfer the samples to gelatin capsules or to a new microcentrifuge tube (the walls of the old one may have been etched by the resin resulting in incomplete polymerization). Fill up the entire volume of the capsule or tube with LR White and close the lid. Since LR White polymerizes only in absence of oxygen, it is critical that the volume of trapped air under the lid is kept at a minimum. 7. Place tubes or capsules into an incubator adjusted to 55°C overnight. 8. Remove the polymerized blocks and cut ultrathin sections with an ultramicrotome. Mount sections on dried grids (store in ethanol to keep them clean). The mounted sections should be allowed to dry thoroughly in order to prevent loss of sections during further handling. For RNA EM-ISH, avoid longer storage of sections. Protocol 2. Probe Labeling for EM-ISH Materials and Reagents • Water bath that keeps water temperature stable at 15°C for 90 min; cooling water bath is recommended (alternative: ice cooling).

• Double-distilled water. • Microcentrifuge tubes. • Centrifuge suitable for spinning microcentrifuge tubes at 40 ,000×g. • Probes: single- and double-stranded DNA as well as RNA can be used as probes for EM-ISH. A variety of different protocols for probe labeling exist. Below, the nick translation method is described requiring doublestranded DNA targets [e.g., cloned fragments, polymerase chain reaction (PCR) products]. The labeled probes can be used to detect DNA as well as RNA. Probes (eukaryotic DNA; 0.5–1.0 µg) in less than 16 µL volume of distilled water are required. • Nick translation: DNA polymerase I, DNase I, all four dNTPs, a labeled dNTP, reaction buffer, and the probe. Here, a commercially available kit was used (Biotin nick translation mix; Roche Molecular Biochemicals, Basel, Switzerland) to generate biotinlabeled probes. The mixture contains a 5× concentrate of: reaction buffer, both enzymes, 0.25 mM each of dATP, dCTP, dGTP, and 0.17 mM dTTP, 0.08 mM biotin-16-dUTP. The protocol follows basically the instructions of the manufacturer. • Ammonium acetate (4 M) in distilled water. • Isopropanol. • Ethanol (70%; store cold). Procedure 1. Adjust a cooling water bath to 15°C. 2. Place microcentrifuge tube on ice. Calculate the volume of probe to obtain 0.5 to 1 µg of DNA. Add probe to the tube, add 4 µL of the nick translation mixture, and fill up with distilled water to 20 µL. 201

Gold and Silver Staining 3. Put microcentrifuge tube into water bath for 90 min. 4. Add 20 µL ammonium acetate and 40 µL isopropanol and incubate for 5 min at room temperature. 5. Spin down at 40 ,000×g for 20 min. 6. Remove supernatant and wash pellet with ice-cold ethanol (70%), then spin down again. 7. Remove ethanol and dissolve the DNA in 20 µL distilled water (may need hours). 8. Determine final concentration of DNA photometrically (a loss of 20% can be assumed) and store for further use at -20°C. Protocol 3. Electron Microscopic In Situ Hybridization Materials and Reagents • Good quality forceps for handling the grids. • Shaker. • Two incubators; one adjustable in the range of 37° to 45°C, the other one between 65° and 95°C. • Heated shaking water bath. • Exicator box. • Glass dishes with wide opening for convenient handling of grids. Vessels should be big enough to contain more than 1 mL of volume. • Moist chamber: glass box with lid for horizontally positioned glass slides. • Glass slides and coverslips. • Distilled water or RNase-free water (for EM-ISH to detect RNA only, see Protocol 1). • 2× standard sodium citrate (SSC): 300 mM NaCl, 30 mM sodium citrate, pH 7.2 (dilute a 20× stock solution with distilled water; stock can be stored for months at room temperature). 202

• RNase A (for EM-ISH to detect DNA; optional for EM-ISH to detect RNA): 100 µg/mL RNase A (Roche Molecular Biochemicals) in 2× SSC (100-µL aliquots of a stock solution of 1 mg/mL can be stored frozen for weeks; dilute with 900 µL 2× SSC). • TrisCaCl2: 20 mmol/L Tris, 2 mM CaCl2, pH 7.4 (a 10× stock solution can be stored for weeks at 4°C). • Proteinase K: 0.5 µg/mL proteinase K (Roche Molecular Biochemicals) in Tris-CaCl2 (100-µL aliquots of 5 µg/mL can be stored frozen and diluted with 900 µL Tris-CaCl2). • Phosphate-buffered saline (PBS): 150 mM NaCl, 1.5 mM NaH2PO4, 15 mM Na2HPO4, pH 7.4 (a 10× stock solution can be stored for weeks). • PBS/MgCl2: 50 mM MgCl2 in PBS, pH 7.2. Adjust the pH slowly in order to prevent precipitation of MgCl2 and filter the solution. • 4% formaldehyde: see Protocol 1 (omit glutardialdehyde). • Formamide (Acros Organics, Geel, Belgium): formamide degrades during prolonged storage; store in dark. Small quantities of formamide from which the (pre)hybridization mixture will be prepared should be stored in vacuum in an exicator box containing dehydrating substances (e.g., silicagel). Likewise, it can be poured through an ion exchange resin before use. Caution: Formamide is irritating to the skin and suspected to be teratogenic. Wear gloves. • FSSC: 50% formamide in 2× SSC. • Prehybridization mixture (for RNA hybridization only): 200 ng/µL denatured herring sperm DNA (Sigma, St. Louis, MO, USA), 200 ng/µL yeast tRNA (Sigma), 10% dextran sulfate (average molecular weight, 500 000;

In Situ Hybridization at the Electron Microscopic Level Sigma) in 50% formamide in 2× SSC. Dextran sulfate dissolves slowly at room temperature; mix for a few hours. Solution can be stored in the refrigerator. • Hybridization mixture: same as prehybridization mixture plus 2 ng/µL biotin-labeled probe. EM-ISH DNA Detection Procedure If not stated otherwise, all steps are performed at room temperature. 1. Incubate grids with RNase A at 37°C for 1 h. This step is necessary if expressed sequences are to be detected in order to avoid binding to transcripts; accessibility of probes to target DNA is also facilitated. 2. Wash in 2× SSC twice for 10 min on a shaker. 3. Wash in Tris-CaCl2 for 10 min (Ca++ is a cofactor of proteinase K) on a shaker. 4. Incubate grids in covered glass dishes with proteinase K at 37°C for 10 min to digest proteins. This is crucial for accessibility of probes to target DNA. 5. Wash in PBS/MgCl2 twice for 10 min (Mg++ reduces the activity of proteinase K) on a shaker. 6. Incubate grids in 4% formaldehyde for 10 min. This will stabilize the structure. 7. Wash in PBS twice for 10 min on a shaker to remove residual formaldehyde. 8. Rinse in distilled water 3 times for 2 min on a shaker. 9. Preparation for hybridization: fill the moist chamber with FSSC and heat to 85°C. Place the grids onto a clean and dry glass slide. Remove water that might have been transferred to the slide with filter paper. Immediately afterwards, put the hybridization mixture (10 µL) onto

the grids and place a coverslip (20 × 20 mm) on top of it. With a coverslip of this size, it will be possible to process a maximum of about 20 grids. Take care to avoid the formation of any air bubbles under the coverslip. The grids should not lie on top of each other. 10. Place the slide in an incubator adjusted to 85°C for 10 min to denature both the target and the probe. 11. Put the slide in the hot moist chamber and transfer it to an incubator adjusted to 37°C to allow annealing of probes overnight (minimum of 6 h). 12. Warm up a shaking water bath containing vessels filled with FSSC, 2× SSC, and small dishes to 42°C. 13. Transfer grids to dishes and wash grids in FSSC at 42°C for 3 × 10 min under gentle shaking (stringency washes). 14. Wash grids in 2× SSC at 42°C for 10 min under gentle shaking to remove dextran sulfate and formamide. 15. Wash grids in 2× SSC at room temperature twice for 10 min on a shaker. Hybridization is thereby finished, and the detection procedure may start immediately (see last protocol). EM-ISH RNA Detection Procedure The RNA detection procedure can be performed in a similar way as hybridizations to detect DNA with a few modifications. All solutions until hybridization should be prepared with RNase-free water. Wear gloves for handling and use sterile glassware and autoclaved pipet tips. Start with steps 3 to 8 exactly as outlined in the previous protocol, then proceed as follows: 1. Preheat a moist chamber filled with FSSC to 42°C. 2. Place the grids onto a clean and dry glass 203

Gold and Silver Staining slide. Remove residual liquid that might have been transferred to the slide with filter paper. Put the prehybridization mixture (10 µL) onto the grids and place a coverslip (20 × 20 mm) on top of it. With a coverslip of this size, it will be possible to process about 20 grids. Take care to avoid the formation of any air bubbles under the coverslip. The grids should not lie on top of each other. 3. Incubate the grids in prehybridization mixture at 42°C for 1 h for blocking of unspecific binding. 4. Remove slides from moist chamber and transfer the box to an incubator adjusted to 65°C. 5. Boil the hybridization mixture for 10 min in order to melt double-stranded probes, place on ice immediately to prevent annealing. 6. Transfer the grids onto a new clean and dry glass slide. Remove prehybridization mixture that might have been transferred to the slide with filter paper. Put the hybridization mixture (10 µL) onto the grids in the same way as in step 2. 7. Place the slides in the incubator containing the moist chamber and incubate at 65°C for 10 min in order to open hairpin loops formed by the target RNAs. 8. Put slides into the moist chamber and incubate at 42°C to allow hybridization overnight (minimum of 6 h). 9. Wash grids in FSSC at 42°C 3 times for 10 min (stringency washes). 10. Wash grids in 2× SSC at 42°C for 10 min to remove dextran sulfate and formamide. 11. Wash grids in 2× SSC at room temperature twice for 10 min. Perform detection of bound probes as outlined in the next protocol. 204

Protocol 4. Immunogold Detection of Bound Probes A comparatively fast protocol is given here. For methods to increase the sensitivity of the detection procedures, see Chapter 13. Materials and Reagents • Parafilm. • Filter paper. • Good quality forceps for handling grids. • Distilled water. • PBS: see Materials and Reagents for EM-ISH. • PBSB: 1% bovine serum albumin (BSA) plus 1% normal goat serum (NGS; optional) in PBS, pH 8.2. • PBT: PBS supplemented with 0.05% Tween 20, pH 8.2. • 10 nm colloidal gold-conjugated antibiotin antibody (raised in donkey; BioCell, Cardiff, Wales, UK) diluted 1:20 in PBSB, pH 8.2. Other grain sizes (5 and 20 nm) may be used as well. • Aqueous uranyl acetate (4% in double-distilled water, filtered). Lead citrate may be used subsequently, but will make it more difficult to detect the gold grains. Procedure for Immunogold Detection of Bound Probes All steps can be performed in drops or floating on the surface of drops that have been placed on Parafilm at room temperature in a moist chamber. Volumes can be as little as 10 µL per drop (20 µL are more convenient for handling). 1. Put grids in PBSB for 30 min. 2. Incubate grids in 10 nm colloidal goldlabeled anti-biotin antibody for 1 h.

In Situ Hybridization at the Electron Microscopic Level 3. Wash grids in PBT 3 times for 10 min. 4. Wash grids in distilled water 3 times for 2 min. 5. Incubate grids in uranyl acetate for 20 min. 6. Rinse grids 3 times with double-distilled water and dry on filter paper. Grids are ready for EM inspection. Controls Negative controls for EM-ISH to detect DNA can be obtained by digestion of target DNA with DNase I during pretreatment. Accordingly, RNase A digestion can be used to remove target RNA for EM-ISH to detect RNA. For this type of EM-ISH, sense probes can be used for control purposes if available. Another control experiment is to omit the labeled probe or replace it with nonmatching probes. This will also help to check the specificity of the detection system. Performing the detection procedure on untreated sections might be necessary in order to check if the conjugates of colloidal gold and antibodies are degraded. If possible, it is preferable to perform ISH at the EM and LM level in parallel. This will combine the advantages of both LM (high sensitivity, 3-D information) and EM (very high resolution). The information that can be obtained about the conditions and the specificity and sensitivity of an ISH experiment at the LM level will help to perform and evaluate EM-ISH. RESULTS Here, we present examples of EM-ISH experiments performed with human ribosomal DNA fragments as probes. Ribosomal nucleic acids are predominantly located in the nucleolus of cells, which is responsible for production of preribosome particles. This offers the possibility to study the dis-

tribution of rDNA and rRNA in relation to a comparatively complex ultrastructure. The nucleolus consists of several compartments. The most prominent components are the fibrillar centers, the dense fibrillar component, and the granular component. The latter compartment is free of rDNA and densely packed with ribosomal RNA. A biopsy of human testis and stimulated human lymphocytes were embedded in LR White resin as described. The same DNA probe was used to detect rDNA and rRNA. It was a probe of the transcribed unit of human ribosomal DNA spanning most of 28S and including 5.8S, the internal transcribed spacers, and part of 18S rDNA (EcoRI defined A-fragment was a gift from Prof. J. Sylvester, Nemours Children’s Clinic, Jacksonville, FL, USA). Figure 14.1 shows the results of EMISH to detect ribosomal DNA and ribosomal RNA in nucleoli of stimulated human lymphocytes. Figure 14.2 demonstrates the appearance of EM-ISH in pachytene stage of spermatogenesis. The nucleolus is in the process of inactivation, and the nucleolar components are well separated. This example is particularly well suited to demonstrate the different distribution of signal after detection of either rDNA or rRNA. TECHNICAL HINTS AND DISCUSSION EM-ISH can be performed for the high resolution localization of RNAs and DNAs. In practice, the detection of DNAs will be restricted to sequences present in high copy numbers considering the fact that it will need some 100 to 200 serial ultrathin sections through an entire average round mammalian nucleus. Pre-embedding EM-ISH is not recommended, because the ultrastructure appears to be deteriorated after the harsh treat205

Gold and Silver Staining ments involved in performing EM-ISH. This is particularly true for EM-ISH to detect DNA. Several kinds of probes can be used for EM-ISH. DNA probes can be double or single stranded such as cloned DNA, cosmid probes, yeast artificial chromosome (YAC) clones, PCR products, cDNAs, or end-labeled oligonucleotides. DNA probes are easy to handle because they are less prone to degradation than RNA probes. Among the available labeling techniques for double-stranded DNA probes, random priming and nick translation are most widely used.8 It is not recommended to directly label DNA during PCR, since it appears that Taq DNA polymerases have difficulties in incorporation of labeled nucleotides used for EM-ISH (see below). DNA probes are suitable for the detection of both DNA and RNA. However, RNA probes form more stable hybrids with RNA than DNA. This allows higher stringency conditions reducing the amount of mismatched probes. RNA probes can be obtained by in vitro transcription of DNA

inserted into a vector containing promoters for the synthesis of sense and antisense RNA. An example would be a vector containing a SP6 and a T7 polymerase promoter. RNase-free conditions throughout hybridization are mandatory using RNA probes. RNase inhibitors can be added to the probe stock solution. Recently, an alternative type of probe became available. Peptide nucleic acids (PNAs) replace the sugar backbone of DNA with a peptide and are considered to be more stable throughout EM-ISH. Biotin9 and digoxigenin6 are preferentially used as nonradioactive labels for EM-ISH. Gold-conjugated antibodies binding directly to these two labels offer a specific and fast way to detect the bound probes. Alternatively, fluorochrome-labeled probes, as used for fluorescent in situ hybridization (FISH), can be used for EM-ISH if specific antibodies directed against this fluorochrome are available. The most important steps to perform a successful EM-ISH are:12 (1) fixation of tissue or cells; (2) protein digestion (both

Figure 14.1. EM-ISH to detect ribosomal nucleic acids in stimulated human lymphocytes. Detection of ribosomal DNA (A) reveals that the signal is preferentially localized in the dense fibrillar component of nucleoli. After detection of ribosomal RNA (B), the gold grains are distributed over dense fibrillar component and granular component. In addition, the cytoplasmic ribosomes are also labeled. DF, dense fibrillar component; FC, fibrillar center; GC, granular component. Calibration bar = 200 nm.

206

In Situ Hybridization at the Electron Microscopic Level tightly connected); and (3) quality of the probe. In the following, the various steps of the procedures are shortly discussed. The fixation of specimen is of utmost importance for the success of EM-ISH. In short, better preserved ultrastructure will be accompanied by lower signal density and vice versa. It is therefore imperative to find an optimal balance between structural preservation and probe accessibility. The choice of fixative will also affect further steps, in particular, protein digestion. The fixative given here should therefore be understood as a suggestion to start with. It might be advisable to make a series of different concentrations of fixative. Especially, the concentration of the strong cross-linking agent glutardialdehyde can be varied. Never use osmium tetroxide because it will prevent binding of probes. Performing reactions in aqueous solutions requires embedding the specimen in hydrophilic resins such as Lowicryl (Poly-

sciences, Warrington, PA, USA), LR Gold (London Resin Company), LR White, Unicryl (Polysciences), etc. The temperature at which polymerization of the resin takes place is of less importance for EMISH, since nucleic acids are comparatively insensitive to elevated temperatures. The ultrathin sections should be mounted on thoroughly cleaned and dried gold grids (grids can be stored in ethanol or acetone). Meshed grids (400 mesh) proved to be good. Do not use grids coated with supporting films, because the film might tear during the following treatments. Apart from fixation, the pretreatment prior to hybridization is of great importance. Protein digestion is particularly important in order to facilitate the access of the probes to the target. The strength of the digestion step is connected with the degree of protein cross-linking during fixation. It is crucial to find a good balance between the two steps. Proteinase K is the most widely used enzyme. Pepsin can be

Figure 14.2. EM-ISH to detect ribosomal nucleic acids in pachytene stage of human spermatogenesis. Attached to the nucleolar components the NORs (nucleolus organizer region; a part of chromosomes housing nontranscribed rDNA) can be seen. Note the differences in distribution of label between detection rDNA (A) and rRNA (B). In panel A, the majority of signal is distributed over the NOR and absent from the granular component. This example shows, that the RNase digestion was complete, and that specifically, rDNA was detected, although a probe from the transcribed part of rDNA was used. A reverse distribution of label can be seen in panel B. Label is concentrated over dense fibrillar component and granular component, whereas the NOR is free of gold grains. DF, dense fibrillar component; FC, fibrillar center; GC, granular component. Calibration bar = 200 nm.

207

Gold and Silver Staining used as well and increases sensitivity of the EM-ISH but has a stronger impact on ultrastructure due to the high acidity of the enzyme solution. For EM-ISH to detect cytoplasmic RNA, the degree of protein digestion can be lower than for EM-ISH to detect DNA. Alternatively, detergents like saponin can be used instead of enzymes for this purpose. For EM-ISH to detect DNA, the separation of the double-stranded target DNA as well as the separation of the probe DNA (if double stranded) can be performed simultaneously. A formula to calculate the melting temperature of the sequence in question is:10 TM = [0.41 × (%GC)] - 500/n - [0.61 × (% formamide)] + (16.6 × log M) + 81.5

where TM is the dissociation temperature (melting temperature) in °C, GC is the content of GC basepair bindings in the sequence in percent, n is the number of nucleotides (about 500 if the probe was labeled by nick translation), and M is the ionic strength of the hybridization mixture (sodium ions in SSC) in mol/L. However, this formula was generated for immobilized naked DNA. Therefore, the TM for EMISH at the surface of resin-embedded tissue should be well above the calculated value (10°–15°C). The hybridization is typically performed at 37°C. Hybridization temperature together with TM and probe length influence the resulting basepair mismatch between probe and target (stringency). For EM-ISH to detect RNA, elevated temperature is applied prior to hybridization to open secondary structures of target RNA that might interfere with probe binding such as hairpin loops. If RNA probes are used, the strand melting of target and probe can be performed simultaneously. The resulting stringency after the hybridization step is relatively low. In addition, probes might be unspecifically bound to cellular structures. In order to increase 208

the stringency, the so-called stringency washes are performed. This step offers the possibility to accurately tune the desired stringency. In case of EM-ISH to detect RNA, a step using RNase A can be performed after the stringency washes to remove single-stranded RNA including unspecifically bound probes. Two methods for the detection of labeled probes are used at the EM level: peroxidaseconjugated antibodies and subsequent deposition of DAB (5) or antibodies conjugated to gold particles (3). The DAB method offers good sensitivity, but the signal is faint, and the resolution is relatively poor. The immunogold method results in the best resolution and is therefore most widely used. Usually, colloidal gold-conjugated antibodies with grain sizes between 5 and 20 nm are used. The use of antibodies labeled with different grain sizes makes double labeling experiments possible. For protocols to further enhance the sensitivity of the immunogold detection procedure, see companion chapters in this volume. Some general properties of EM should be kept in mind for the interpretation of the results of EM-ISH: (1) The lateral resolution is far higher than in LM, but information about the third dimension is lacking. In practical terms that means that a high number of cells has to be scrutinized to obtain information about the distribution of nucleic acids in the cells; (2) Only those molecules that are exposed at the surface of a section are available for detection. The detected molecule in a single section therefore does not reflect the overall distribution. Again, the evaluation of many cells of one type and of many sections circumvents this problem; and (3) Immunogold detection allows quantification of gold grains because of the particulate nature of gold grains. Quantification can be performed by evaluating the signal density over a particular cellular compartment. Evaluation of signal density requires area measurements which are most

In Situ Hybridization at the Electron Microscopic Level conveniently done with the help of a commercially available morphometry software. The results of quantification should be compared to one another and to background densities with the help of statistics. At first, it should be analyzed if the signal over an area is distributed according to a Gaussian (normal) distribution. If this is the case, a Student’s t-test might be used. Otherwise, a nonparametrical test, for example a Kolmogoroff-Smirnoff test, should be preferred. ACKNOWLEDGMENTS We thank M. Almeder for expert technical assistance in performing ISHs at the EM level. REFERENCES 1.Bernhard, W. 1969. A new staining procedure for electron microscopical cytology. J. Ultrastruct. Res. 27:250-265. 2.Cogliati, R. and A. Gautier. 1973. Demonstration of DNA and polysaccharides using a new “Schiff type” reagent. C.R. Hebd. Seances. Acad. Sci. D. 276:30413044. 3.Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemistry 8:10081-10083. 4.Geuskens, M. and E. May. 1974. Ultrastructural localization of SV40 viral DNA in cells, during lytic infection, by in situ molecular hybridization. Exp. Cell Res. 87:175-185. 5.Graham, R.C., Jr. and M.J. Karnovsky. 1966. The

early stages of absorption of injected horseradish peroxidase in the proximal tubule of mouse kidney, ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem. 14:291-302. 6.Heiles, H.B., E. Genersch, C. Kessler, R. Neumann, and H.J. Eggers. 1988. In situ hybridization with digoxigenin-labeled DNA of human papillomaviruses (HPV 16/18) in HeLa and SiHa cells. BioTechniques 6:978-981. 7.Jacob, J., M.H. Moar, K. Gillies, D. Macleod, and K.W. Jones. 1976. Molecular hybridization of RNA and DNA in situ visualization at the electron microscope level. J. Microsc. 106:185-198. 8.Maniatis, T., E.F. Fritsch, and J. Sambrook. 1982. Molecular Cloning. A Laboratory Manual. CSH Laboratory Press, Cold Spring Harbor, NY. 9.Manning, J.E., N.D. Hershey, T.R. Broker, M. Pellegrini, H.K. Mitchell, and N. Davidson. 1975. A new method of in situ hybridization. Chromosoma 53:107117. 10.Meinkoth, J. and G. Wahl. 1984. Hybridization of nucleic acids immobilized on solid supports. Anal. Biochem. 138:267-284. 11.Puvion, E. and W. Bernhard. 1975. Ribonucleoprotein components in liver cell nuclei as visualized by cryoultramicrotomy. J. Cell Biol. 67:200-214. 12.Schöfer, C., K. Weipoltshammer, C. Hauser-Kronberger, and F. Wachtler. 1997. High resolution detection of nucleic acids at the electron microscopic level. Review of in situ hybridization technology, the use of gold, and catalyzed reporter deposition (CARD). Cell Vision 4:443-454. 13.Thiry, M. 1991. in situ nick translation at the electron microscopic level: a tool for studying the location of DNAse I-sensitive regions within the cell. J. Histochem. Cytochem. 39:871-874. 14.Thiry, M. 1992. Highly sensitive immunodetection of DNA on sections with exogenous terminal deoxynucleotidyl transferase and non-isotopic nucleotide analogues. J. Histochem. Cytochem. 40:411-419. 15.Walker, S.E., R.H. Gray, M. Fulton, R.D. Wigley, and B. Schnitzer. 1978. Palmerston North mice, a new animal model of systemic lupus erythematosus. J. Lab. Clin. Med. 92:932-945.

209

15

Immunogold–Silver Staining for Scanning Electron Microscopy in Cancer Research Bao-Le Wang, Nina Flay, and Gerhard W. Hacker

INTRODUCTION Cell surface components, especially glycoproteins and glycolipids, have versatile functions involved in cell–cell and cell– matrices recognition, adhesion, signal transduction, cell growth, and motility.6,8,9 It is important to understand the essential roles of these components under physiological and pathological circumstances by identifying and localizing cell surface-specific macromolecules and studying their relationship with cellular structures. Scanning electron microscopy (SEM) has the advantage of offering three-dimensional images of cell surface ultrastructures. Combined with immunocytochemical methods, immuno-SEM is becoming one of the most potent tools for the characterization of cell surface components. Since Faulk and Taylor7 introduced colloidal gold particles as transmission electron microscope (TEM) immunocytochemical labels in 1971, and Horisberger 0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

et al.14 thereafter demonstrated their applicability in SEM in 1975, colloidal gold probes have been increasingly used in immunoelectron microscopy (immunoEM) because they have a number of advantages to other particulate markers such as ferritin and iron–dextran complexes, as well as enzymatic markers.12,18 Their high electron density and strong emission of secondary electrons make colloidal gold probes easily detectable under the EM. Homodispersed colloidal gold particles can be prepared in different sizes to adapt to various applications including multiple labeling immuno-EM.24,25 To be directly identifiable under a conventional scanning electron microscope with limited resolution power, the size of colloidal gold particles needs to be large enough. Usually 50 to 60 nm are the most convenient sizes, though particles as small as 20 nm have also been used successfully.12 However, the immunolabeling sensitivity is inversely related to the size of the 211

Gold and Silver Staining gold particle, and theoretically, gold particles of the smallest diameter should give the best labeling efficiency, because steric hindrance would be minimized.5,15 Danscher3 in 1981 introduced silver enhancement (autometallography, AMG), using a “physical developer” based on silver lactate to deposit metallic silver on colloidal gold particles, thereby enabling the originally invisible tiny gold particles to be visualized under the light microscope (LM). In 1983, Holgate et al. used Danscher’s method to silver enhance colloidal gold-adsorbed immunoglobulins applied in immunohistochemistry, and called this the immunogold–silver staining (IGSS) method.10 Based on this method, Scopsi et al.19 introduced the application of silver-enhanced colloidal gold probes for SEM in 1986. This application of IGSS has the advantage of increased detection efficiency while maintaining high sensitivity by using gold particles of smaller size, such as 5 to 7 nm in diameter. In recent years, Dr. Springer’s group has intensively investigated the role of T (Thomsen-Friedenreich) and Tn epitopes in pathogenesis of carcinoma. T and Tn are progenitors of O-linked carbohydrate structures of glycoproteins on fully differentiated normal epithelial and blood cell surfaces. T and Tn are masked by extended heterosaccharide chains and are inaccessible to the immune system in normal and benign epithelial cells. However, in transformed malignant epithelial cells, T and Tn epitopes, resulting from aberrant glycosylation, are uncovered and therefore immunodetectable in about 90% of all adenocarcinomas.20–23 Previous studies revealed that T and Tn antigens on the cancer cell surface contribute actively to cancer cell aggressiveness.21 The extent of T and Tn antigen expression in cancer cells is related to the prognosis.26 To further understand the role of T and Tn epi212

topes in carcinoma cell adhesion, invasion, locomotion, and metastasis, we employed an IGSS method to identify and localize T and Tn epitopes on human breast carcinoma cells and observed their relationship to the cell surface structures at the SEM level.27 The detailed procedures for SEMIGSS are as follows: PROTOCOLS Materials and Reagents • Glass coverslips (round, 12 mm in diameter; Fisher Scientific, Pittsburgh, PA, USA). • Twenty-four well cell culture plates (Becton Dickinson Labware, Bedford, MA, USA). • Transwell chambers (6.5 mm in diameter with 8 µm pore size membrane; Corning Costar, Acton, MA, USA). • Carcinoma cell lines: human breast carcinoma cell line HTB 24 from ATCC (Rockville, MD, USA); HBC 6558 was generated and grown in H.M. Bligh Laboratories (North Chicago, IL, USA) (27). • Poly-D-lysine (Sigma, St. Louis, MO, USA). • Matrigel (Becton Dickinson Labware). • DFCI-1 cell culture medium (see Supporting Protocol A). • Hanks’ balanced salt solution (HBSS) containing 0.014% CaCl2 (Life Technologies, Gaithersburg, MD, USA). • Dulbecco’s phosphate-buffered saline (DPBS; Sigma). • Paraformaldehyde (Electron Microscopy Sciences, Fort Washington, PA, USA). • Glutardialdehyde (Electron Microscopy Sciences). • Bovine serum albumin (BSA, Fraction V; Sigma). • Glycine (Sigma).

Immunogold–Silver Staining for Scanning Electron Microscopy • Primary antibodies: mouse monoclonal antibodies against T and Tn (RS1-114, HT-8, BaGS-3, BaGS5).26 • Secondary antibodies: colloidal gold (10 nm)-conjugated goat antimouse IgG + IgM [Amersham Pharmacia Biotech, Piscataway, NJ, USA; or prepared as previously reported].25 • Tris-HCl-buffered saline (TBS, 0.02 M, pH 8.2; Sigma). • Silver enhancement solution (see Supporting Protocol B). • Cell culture-needed conditions and apparatus. • Aluminum stubs. • Critical point drying apparatus (Polaron E3000; Polaron Instruments, Hatfield, PA, USA). • Palladium sputter coater (Polaron E5150; Polaron Instruments). • Scanning electron microscope (JEOL 35C; JEOL USA, Peabody, MA, USA). Protocol 1. Specimen Preparation All cell culture-related procedures are carried out under sterile conditions. 1a. Preparation of poly-D-lysine-coated glass coverslips: autoclave-sterilized 12mm round glass coverslips are placed in 24-well cell culture plates. The coverslips are covered by 1 mL/well of 0.1 mg/mL solution of poly-D-lysine in distilled water for 15 min at room temperature. Then the poly-D-lysine solution is discarded, and the coverslips are rinsed three times with DPBS and are ready for use. 1b. Preparation of Matrigel-coated transwell membranes: transwell chambers (6.5 mm in diameter) with 8 µm pore size membrane are coated with Matrigel, a basement membrane extract as follows: to each upper cham-

ber, 100 µL of a 1 mg/mL Matrigel in DFCI-1 medium is added, and then the chamber is incubated for 5 h at 37°C. Remove excess medium, and the chamber is ready to use. 2a. Growth of cancer cells on glass coverslips: plate cancer cells (HTB 24, HBC 6558) at 1 to 2 × 104 cells in 1 mL of DFCI-1 medium per coverslip per well (see step 1a.), incubate at 37°C, 5% CO2, and humidified atmosphere for 3 to 4 days until cells reach 50% confluence. 2b. Growth of cancer cells on transwell membranes: plate cancer cells at 2 to 5 × 104 cells in 100 µL of DFCI-1 medium to each upper chamber and add 600 µL of DFCI-1 alone to each lower chamber. Incubate at 37°C, 5% CO2, and humidified atmosphere for 2 to 3 days. The cells on the upper surface of the transwell membrane are removed by gentle wiping with a cotton swab, while the cells that penetrated to the lower surface of the membrane are used for further processing. 3. Cancer cells grown on coverslips or on Matrigel-coated transwell membranes are rinsed with HBSS three times, then processed in the same way as in the procedures described in protocols 2 and 3. It is important to never let the cells become dry during the whole process until the step after the critical point drying. 4. Fix the cells by immersing the coverslips or transwell membrane in 4% paraformaldehyde and 0.5% glutardialdehyde in DPBS at room temperature for 1 h. 5. Rinse the cells three times for 5 min each with DPBS. 6. Incubate the cells in 0.1% glycine in DPBS for 30 min to quench any remaining free aldehyde. 213

Gold and Silver Staining 7. Rinse the cells three times for 5 min each with DPBS. Protocol 2. Immunogold–Silver Staining for SEM 1. The cells are incubated with 1% BSA in DPBS for 15 min. 2. Remove the BSA solution. Then, the specimens are incubated with properly diluted primary antibodies in 1% BSA in DPBS. We use a cocktail of monoclonal antibodies, anti-T: RS1-114, 1:200; HT-8, 1:400, or anti-Tn: BaGS3, 1:80; BaGS-5, 1:80, at room temperature for 1 h. For immunocytochemistry (ICC) controls, either omit the primary antibodies or replace them by antibodies pre-absorbed to their specific and immunologically related antigens. (In our case: adsorption with group O, T, or Tn red blood cells.) 3. Rinse the cells three times for 5 min each with DPBS. 4. Rinse the cells with 0.02 mol/L TBS, pH 8.2, containing 0.1% BSA, for 1 min. 5. Incubate the cells with properly diluted gold-conjugated secondary antibody (in our case it is 10 nm gold-conjugated goat antimouse IgG + IgM in 0.02 mol/L TBS-BSA) for 45 min. 6. Rinse the cells three times for 5 min each with 0.02 mol/L TBS-BSA. 7. Rinse the cells three times for 5 min each with distilled water. 8. Incubate the cells with silver developer solution (see Supporting Protocol B) for 15 min. 9. Rinse the cells three times for 5 min each with distilled water. 10. Refix the cells with 2% glutardialdehyde in DPBS for 30 min. 11. Rinse the cells three times for 5 min each with distilled water. 214

Protocol 3. Processing Specimen For SEM 1. Dehydrate the cells with a series of ascending concentrations of ethanol (EtOH): At this point, the transwell membranes containing the stained cells are removed by cutting with a scalpel along the wall edge of the transwells. Put them back in the lower chambers, ensuring that the sides with cells face upward. Then process as coverslips: 30% EtOH 10 min (2 changes during the 10 min) 50% EtOH 10 min (2 changes during the 10 min) 70% EtOH 10 min (2 changes during the 10 min) 95% EtOH 10 min (2 changes during the 10 min) 100% EtOH 15 min (3 changes during the 15 min) 2. At this point, specimens either on circular glass coverslips or on transwell membranes are ready for critical point drying. Keep the cells submerged in 100% EtOH at all times to prevent air drying. Place the coverslips or transwell membranes in appropriate holders for critical point drying. The following procedure has proven successful: a. Individually wrap each specimen containing the cells in lint-free optical lens paper and load the specimens into the trough. The samples can be placed within or between mesh baskets designed to fit into the trough. For three baskets, five samples can be processed at once by placing the three samples in the three baskets and the other two samples between the baskets. The position of the samples with respect to location in or between baskets should be noted to avoid mix-up of the samples.

Immunogold–Silver Staining for Scanning Electron Microscopy b. Fill the trough to the brim to ensure that the specimens are completely immersed in the transitional fluid. The entire trough is covered with a sliding screen to make sure that the contents are maintained firmly in their assigned places. 3. Put the holders with specimens in the critical point drying apparatus (Polaron E3000) to perform the critical point drying (see supporting protocol C). 4. Open critical point drying specimen chamber and remove specimens. Samples are now dry and ready for mounting and sputter coating. 5. Mount the coverslips or transwell membranes on aluminum stubs with double stick tape. In order to get the correct specimen size to fit the aluminum stub, use a diamond pen to score and break each coverslip in half. For transwell membranes, use a small cut portion. 6. Paint edges of the coverslips or transwell membranes with colloidal silver paste, to be sure that there is electrical contact between the metal stub and the top surface of the glass coverslip or transwell membrane. Place the specimens in a drying oven at 40°C for a few minutes. 7. Put the specimens into the sputter coater (Polaron E5150), coat with gold–palladium (see supporting protocol D). 8. Store the coated specimens in a desiccator at room temperature until examined. 9. The specimens are observed under a SEM (e.g., JEOL 35C) in secondary electron image mode at accelerating voltage of 15 kV. Supporting Protocol A. Preparation of DFCI-1 Epithelial Cell Culture Medium Recipe for 2000 mL, according to Band and Sager,1 with slight modification:

Dissolve: Alpha MEM (Cat. No. 12000-022; Life Technologies) 1 package for 1 L in 800 mL distilled water Ham’s F-12 (Cat. No. 21700-075; Life Technologies) 1 package for 1 L in 800 mL distilled water Mix the above two solutions and then add the components in Table 15.1. Adjust the final volume to 2000 mL using distilled water. Mix well, filter through a 0.22-µm sterile filter, and store at 4°C for up to 3 weeks. The pH should range from approximately 7.1 to 7.2. Fungizone is an optional supplement. Supporting Protocol B. Preparation of Autometallographic Silver Enhancement Solution: According to Danscher,3 with Slight Modification Recipe for 100 mL: mix the solutions in the following order. The silver lactate solution should be prepared and only added immediately before use, and should be protected from light. 1. 2.55 g citric acid monohydrate plus 2.35 g trisodium citrate dihydrate plus 50 mL distilled water. 2. 50% gum arabic in 20 mL distilled water. 3. 0.85 g hydroquinone plus 15 mL distilled water. 4. 0.11 g silver lactate plus 15 mL distilled water. Above reagents are all from Sigma. Instead of this silver enhancement procedure, other types of AMG gold amplification procedures can be used, e.g., the more light-insensitive silver acetate AMG developer (see Chapter 2). Supporting Protocol C. Critical Point Drying 1. Initially, the water bath temperature 215

Gold and Silver Staining Table 15.1. Components for DFCI-1 Epithelial Cell Culture Medium

Components 1. Sodium bicarbonate (Cat. No. 11810-025; Life Technologies)

Stock Solution

Add

powder

3.36 g

Final Concentration 1.68 mg/mL

2. EGF (Cat. No. E9644; Sigma)

100 µg/mL (dH2O)

250 µL

12.5 ng/mL

3. Triiodo-thyronine (Cat. No. T6397; Sigma)

0.1 mM (dH2O)

200 µL

10 nM

4. HEPES buffer (Cat. No. 15630-080; Life Technologies)

1M

20 mL

10 mM

5. Ascorbic acid (Cat. No. A4403; Sigma)

100 mM (dH2O)

1 mL

50 µM

6. Estradiol (Cat. No. E2758; Sigma)

100 µM (95%EtOH)

40 µL

2 nM

7. Insulin (Cat. No. I1882; Sigma)

5 mg/mL (dH2O)

400 µL

1 µg/mL

8. Hydrocortisone (Cat. No. H0888; Sigma)

10 mg/mL (95% EtOH)

200 µL

2.8 µM

9. Ethanolamine (Cat. No. E0135; Sigma)

100 mM (DPBS)

2 mL

0.1 mM

10. Phosphoethanolamine (Cat. No. P0503; Sigma)

100 mM (DPBS)

2 mL

0.1 mM

11. Transferrin (Cat. No. T1147; Sigma)

10 mg/mL (DPBS)

2 mL

10 µg/mL

12. L-Glutamine (Cat. No. 25030-081; Life Technologies)

200 mM

20 mL

2 mM

13. Penicillin–Streptomycin (Cat. No.

10 000 U/mL

20 mL

100 U/mL

15140-122; Life Technologies)

10 000 µg/mL

100 µg/mL

14. Sodium selenite (Cat. No. S5261; Sigma)

300 µM (dH2O)

100 µL

15 nM

15. Cholera toxin (Cat. No. C3012; Sigma)

5 µg/mL (DPBS)

400 µL

1 ng/mL

20 mL

1%

16. FBS (Cat. No. 16000-044; Life Technologies)

—

17. Bovine pituitary extract (Cat. No. 13028-014; Life Technologies)

25 mg/tube

25 mg

12.5 µg/mL

18. Fungizone (Cat. No. 15290-018; Life Technologies)

250 µg/mL

4 mL

0.5 µg/mL

dH2O, distilled water; EGF, epidermal growth factor; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; FBS, fetal bovine serum.

should be set so that the temperature inside the critical point dryer is maintained between 18° and 20°C. 2. Use only 99.8% (bone dry) CO2 with a siphon (dip) tube. Open the CO2 valve, increasing the pressure very slowly to a maximum pressure of 800 psi within the chamber. The upper (gaseous) vent 216

should be slightly open to allow gases to exit. When the chamber is filled completely with fluid, the CO2 will begin to exit through the lower (fluid) vent tube. Fluid CO2 resembles snow in its appearance. 3. Flush the chamber for 2 to 4 min, maintaining a balanced fluid level (at

Immunogold–Silver Staining for Scanning Electron Microscopy midwindow on the critical point drying apparatus) between the CO2 entering the chamber and the transitional fluid exiting the chamber. In this way, specimens remain immersed at all times, while the transitional fluid is being replaced by liquid CO2. This is important to prevent drying of the cells. Looking through the window, turbulence between the transitional fluid being flushed out and the CO2 being introduced should be observed. 4. Infiltrate (purge) for 25 to 30 min at 20°C. During this step, all valves are closed and the level of fluid as seen in the window should remain constant. A drop in fluid level suggests there is a leak in the apparatus. Loss of fluid without introduction of more fluid will result in premature drying of samples and must be avoided. 5. Repeat the flush for 2 to 4 min, introducing fresh liquid CO2, and venting off the infiltrating fluid. Continue to monitor the fluid level, which should remain at midwindow. At the end of this step, the specimens should be completely infiltrated with liquid CO2 with no transitional fluid remaining. Close all inlets and outlets. 6. To completely dry the samples, evaporate the CO2 by elevating both temperature and pressure of the gas to its critical point. The critical point of CO2 is reached at 38°C and 1200 psi. The limits should be set at these levels and care taken to assure that neither the temperature nor the pressure exceeds these limits. During this step the CO2 should be heated slowly until the critical point of temperature and pressure is achieved; at this point, evaporation of the gas will occur as surface tension is lost. It is recommended that the temperature and pressure be carefully monitored so that the temperature increases at a rate of

one or two degrees per minute and no faster. The fluid level in the window will slowly rise. When the critical point for CO2 is reached, the level will disappear from the window, signifying that the CO2 has completely evaporated. The samples should be completely dry when the critical point is reached. 7. Open the top (gaseous) vent slightly to allow very slow venting. Supporting Protocol D. Sputter Coating with Gold–Palladium A sputter coater (Polaron E5150) is attached to (1) an argon gas cylinder with gas regulator, and (2) a rotary vacuum pump. 1. After the dried specimens have been placed in the chamber, the chamber lid is firmly closed, and the vacuum pump is turned on. Usually, within 1 min, a vacuum reading of 0.2 Torr should be reached. This is a weak vacuum. A lag in time necessary to achieve this vacuum reading suggests there is a leak in one of the lines, and this must be corrected before proceeding further. Pumping to the necessary strong vacuum of 0.06 Torr will take an additional 25 to 30 min. 2. At a vacuum reading of 0.06 Torr, leak the argon gas by (1) opening the argon gas cylinder, and (2) opening the leak valve on the sputter coater all the way for about 30 sec before closing the valve. The needle on the vacuum indicator should swing toward a stronger reading. The greater the number of times that this leak step is repeated, the stronger the final vacuum will be. For the actual coating procedure, a final vacuum of 0.04 Torr is necessary. 3. To coat the specimens: a. Turn the dial to “Set HT.” b. Set the kV reading to 2.4 microamps. 217

Gold and Silver Staining c. Turn on the argon until the microamp reading is at 2. At this step, the vacuum reading will be around 0.06 Torr. Maintain it at this level by adjusting the argon valve and the kV constantly but slightly. If the vacuum is not strong enough, it will automatically shut off the electricity and another vacuum must be pulled. If the microamp reading is too high, the fuse will blow. d. A lavender glow will appear in the chamber. Allow the specimens to be coated for 90 sec. e. Turn off the electric current, the HT, the argon gas, and the rotary pump. f. Gently vent the chamber and remove the specimens. Store them in a dry place.

cally at the adhesion plaque area, adjacent to the base of adhesion fibers (Figures 15.1–15.4). The gold–silver particles had strong emission of secondary electrons. Therefore, they were generally brighter than the intrinsic cellular structure. Since the positive labeling was concentrated at the adhesion plaque area, the gold particles may have clumped together close enough to form coalescence during silver enhancement; this resulted in various sizes of the gold–silver particles. The majority of them ranged from 250 to 1000 nm; the larger particles usually showed slightly irregular

RESULTS Under the scanning electron microscope, the fine structures of the breast carcinoma cells grown on both glass coverslips and transwell membranes were well preserved and showed similar features, though the cancer cells grown on transwell membranes after pore penetration had more raised and elongated shape. The many steps of IGSS procedure did not cause damage to the ultrastructures on cancer cell surfaces, including microvilli, lamellipodia, microspikes, and retraction fibers. In our case, a characteristic feature of some cancer cells was noticed. On the edge of these cancer cells, a kind of “anchoring-type” adhesion plaque structure showed numerous dense, straight adhesion fibers stretched out from the focal region of plasma membrane and anchored to the substratum (Figures 15.1–15.4). The positive labeling for T and Tn antigens, resulting from the IGSS, revealed that the gold–silver particles were not randomly distributed on the cancer cell surface, but aggregated specifi218

Figure 15.1. SEM photomicrograph of cancer cell line HTB 24 grown on glass coverslips. IGSS for Tn epitopes. Positive labeling aggregated at adhesion plaque area. (Original magnification ×2000.) (Reprinted with permission from Wang, B.L. et al., J. Submicrosc. Cytol. Pathol., 30:503–509.)

Figure 15.2. SEM photomicrograph of cancer cell line HTB 24 grown on glass coverslips. IGSS for T epitopes. Positive labeling aggregated at adhesion plaque area. (Original magnification ×3600.) (Reprinted with permission from Wang, B.L. et al., J. Submicrosc. Cytol. Pathol., 30:503–509.)

Immunogold–Silver Staining for Scanning Electron Microscopy shapes. These features made them easily distinguishable from the cell surface structures even at lower magnification (Figures 15.1, 15.3, and 15.4). The background caused by nonspecific precipitation of silver was not obvious. The ICC specificity controls were negative. TECHNICAL HINTS AND DISCUSSION Through application of IGSS for local-

Figure 15.3. SEM photomicrograph of cancer cell line HBC 6558 grown on glass coverslips. IGSS for T epitopes. Positive labeling aggregated at adhesion plaque area. (Original magnification ×1800.) (Reprinted with permission from Wang, B.L. et al., J. Submicrosc. Cytol. Pathol., 30:503–509.)

Figure 15.4. SEM photomicrograph of cancer cell line HBC 6558 grown on transwell membrane coated with Matrigel. IGSS for T epitopes. Positive labeling aggregated at adhesion plaque area. (Original magnification ×2000.) (Reprinted with permission from Wang, B.L. et al., J. Submicrosc. Cytol. Pathol., 30:503–509.)

izing T and Tn antigens on human breast cancer cell surface at the SEM level, our results demonstrated again that immunoSEM can provide valuable information about the relationship between certain cell surface macromolecules and structures. At first, colloidal gold label was introduced into SEM application without using a silver enhancement procedure, and the most convenient sizes of the gold particles used in these studies were approximately 50 to 60 nm.11–14 When applied on specimens with relatively smooth surfaces such as yeast cells or erythrocytes, such gold particles can be easily identified, especially on specimens without metal coating. The advantage of using colloidal gold label without AMG is that it allows double immunolabeling with different sizes of gold particles and quantitative evaluation.11,12 However, when studying larger specimens, like cancer cells grown on coverslips or transwell membranes, a conductive coating is needed to prevent electrostatic charging and to increase the contrast of threedimensional images of the cell surface architecture. In this case, small colloidal gold particles of 15 to 20 nm become difficult to resolve in secondary electron image mode, as they become “buried” under the coating, but using larger gold particles may reduce the labeling sensitivity.2,12 To solve this problem, the employment of silver enhancement is an approach that also provides higher detection efficiency while retaining the high labeling sensitivity of small gold particles of 5 or 10 nm (e.g., see Reference 19). To recognize small colloidal gold particles ambiguous in secondary electron image (SEI) mode, the backscattered electron image (BEI) mode was applied in SEM.5 However, the structural details of cell surface are not satisfactory in BEI. By using IGSS methodology, the small gold particles introduced to the target–antigen site are progressively enlarged and can be 219

Gold and Silver Staining efficiently detected in SEI mode, while the cell surface ultrastructures as three-dimensional images can also be observed. Therefore, analyzing the relationship between the cell surface-specific macromolecules and the ultrastructures becomes feasible. For SEM-IGSS, selection and manipulation of the prospective final size of gold–silver particles should be determined, depending upon the aims and objectives to be investigated. Many factors may affect the gold–silver particle growth rate and the final size during the silver enhancement procedure, such as the concentration of the protective colloid of gum arabic, the concentration and type of the different silver salts used, the development time, the temperature, the light strength, the pH value, and the ionic strength of the developing solution (see Chapters 2 and 3).15 As pointed out by by Scopsi et al.,19 silver nitrate, due to its high disintegration coefficient, gives a speed of reaction higher than that obtained with silver lactate. However, several authors have also found that silver nitrate can lead to nonspecific (nonimmune-related) reactions and therefore has to be applied with caution, using various specificity controls (see Chapters 2 and 3). Also, shorter development times should be used with silver nitrate to achieve the same particle size as usually given with silver lactate- or silver acetate-based AMG. It is convenient to perform silver development at room temperature (18°–20°C). To keep the results reproducible and consistent, one has to make sure that the gum arabic stored frozen is brought to room temperature before use. When the other factors, like pH and ionic strength are reproducibly maintained, the most important factors to control the final size of the gold–silver particles are the concentration of gum arabic and the development time. Previous studies16,19 revealed that without protective colloid in the AMG development solution, the silver development 220

becomes less controllable after 5 min, and high background will result from nonspecific precipitation of crystal-like silver particles. In Danscher’s original recipe,3 the final concentration of protective colloid gum arabic is 30%. When small colloidal gold particles (5–7 nm) were used in IGSS, Scopsi et al.19 reported that 15% of gum arabic provided satisfactory control of the progressive silver enlargement of gold particles. After 60 min, the diameter of the particles was increased about 15-fold, but their round shape was preserved. In our case, using 10 nm colloidal gold particles and 10% gum arabic for IGSS, the particles were enlarged about 30-fold after 15 min without nonspecific precipitate formation. Selective balance among the original gold particle size, the concentration of gum arabic, the development time, and the final size of gold–silver particles should favor distinguishing the marker from fine cellular structures, while exposing the specimen to a low pH development solution for as short a time as possible in IGSS procedures. An alternative AMG developer based on gold–salt, working under neutral pH conditions, was recently introduced under the brand name GoldEnhance (Nanoprobes, Yaphank, NY, USA) and allows control of particle size in a more exact way than earlier developers (Chapter 3). To stabilize cell surface fine architecture, and preserve antigenicity of target macromolecules, proper application of fixative is also crucial. Without prefixation prior to the many steps of IGSS, the cell surface ultrastructures may be easily distorted, and some intracytoplasmic antigens may leak out and cause artifacts.11,13 However, very strong fixation may also denature the fixative-sensitive antigens. We found that using 4% paraformaldehyde and 0.5% glutardialdehyde fixation for 1 h before the IGSS process can effectively prevent cell surface structure changes while preserving the T and Tn immunoreactivity satisfacto-

Immunogold–Silver Staining for Scanning Electron Microscopy rily. After IGSS, some loss of cell surface markers may result from mechanical shearing during preparation of marked specimens for SEM evaluation. Postfixation with 2% glutardialdehyde can minimize the structural damage and consolidate the attachment of gold–silver particles during the drying procedures. Though some earlier SEM studies suggested that critical point drying was unnecessary,2 we found that in agreement with the previous reports,4,17 the critical point drying procedures are necessary for preserving well the overall three-dimensional ultrastructures of the cells, especially the fine adhesion fibers and microvilli, keeping the original spatial relation between the gold–silver particles and cell structures, and avoiding various degrees of cell collapse, cracks, and loss of surface details observed in air-dried specimens. ACKNOWLEDGMENTS This chapter is dedicated to Dr. Georg F. Springer, the founder of Heather M. Bligh Cancer Research Laboratories at Finch University of Health Sciences, The Chicago Medical School. The research work reported in this chapter was supported by the Heather M. Bligh Cancer Fund. REFERENCES 1.Band, V. and R. Sager. 1989. Distinctive traits of normal and tumor-derived human mammary epithelial cells expressed in a medium that supports long-term growth of both cell types. Proc. Natl. Acad. Sci. USA 86:1249-1253. 2.Catt, J.W., M.A. Peacock, and F.L. Harrison. 1985. Surface localization of an endogenous lectin in rabbit bone marrow. Histochem. J. 17:189-199. 3.Danscher, G. 1981. Histochemical demonstration of heavy metals. A revised version of the sulphide silver method suitable for both light and electron microscopy. Histochemistry 71:1-16. 4.De Bault, L.E. 1973. A critical point drying technique for SEM of tissue culture cells grown on plastic substratum, p. 317-324. In O. Johari (Ed.), Scanning

Electron Microscopy, Part III. IIT Research Institute, Chicago. 5.De Harven, E. and D. Soligo. 1986. Scanning electron microscopy of cell surface antigens labeled with colloidal gold. Am. J. Anat. 175:277-287. 6.DiCorleto, P.E. and C.A. De La Motte. 1989. Role of cell surface carbohydrate moieties in monocytic cell adhesion to endothelium in vitro. J. Immunology 143:3666-3672. 7.Faulk, W.P. and G.M. Taylor. 1971. An immunocolloid method for the electron microscope. Immunochemistry 8:1081-1083. 8.Fukuda, M. 1996. Possible roles of tumor-associated carbohydrate antigens. Cancer Res. 56:2237-2244. 9.Hakomori, S.-i. 1996. Tumor malignancy defined by aberrant glycosylation and sphingo (glyco) lipid metabolism. Cancer Res. 56:5309-5318. 10.Holgate, C.S., P. Jackson, P.N. Cowen, and C.C. Bird. 1983. Immunogold–silver staining: new method of immunostaining with enhanced sensitivity. J. Histochem. Cytochem. 31:938-944. 11.Horisberger, M. 1979. Evaluation of colloidal gold as a cytochemical marker for transmission and scanning electron microscopy. Biol. Cellulaire 36:253-258. 12.Horisberger, M. 1984. Electron-opaque markers: a review, p. 17-26. In J.M. Polak and I.M. Varndell (Eds.), Immunolabelling for Electron Microscopy. Elsevier, New York. 13.Horisberger, M. and J. Rosset. 1977. Colloidal gold, a useful marker for transmission and scanning electron microscopy. J. Histochem. Cytochem. 25:295-305. 14.Horisberger, M., J. Rosset, and H. Bauer. 1975. Colloidal gold granules as markers for cell surface receptors in the scanning electron microscope. Experimentia 31:1147-1149. 15.Lackie, P.M., R.J. Hennessy, G.W. Hacker, and J.M. Polak. 1985. Investigation of immunogold–silver staining by electron microscopy. Histochemistry 83:545-550. 16.Larsson, L.-I. 1988. Immunocytochemistry: Theory and Practice. p. 107-110. CRC Press, Boca Raton. 17.Lewis, E.R., L. Jackson, and T. Scott. 1975. Comparison of miscibilites and critical-point drying properties of various intermediate and transitional fluids, p. 317324. In O. Johari (Ed.), Scanning Electron Microscopy. IIT Research Institute, Chicago. 18.Molday, R.S. and P. Mather. 1980. A review of cell surface markers and labeling techniques for scanning electron microscopy. Histochem. J. 12:273-315. 19.Scopsi, L., L.-I. Larsson, L. Bastholm, and M. Hartvig Nielsen. 1986. Silver-enhanced colloidal gold probes as markers for scanning electron microscopy. Histochemistry 86:35-41. 20.Springer, G.F. 1984. T and Tn, general carcinoma autoantigens. Science 224:1198-1206. 21.Springer, G.F. 1989. Tn epitope (N-acetyl-D-galactosamineα-O-serine/threonine) density in primary breast carcinoma: a functional predictor of aggressiveness. Mol. Immunol. 26:1-5. 22.Springer, G.F. 1997. Immunoreactive T and Tn epitopes in cancer diagnosis, prognosis and immunotherapy. J. Mol. Med. 75:594-602.

221

Gold and Silver Staining 23.Springer, G.F., P.R. Desai, and I. Banatwala. 1975. Blood group MN antigens and precursors in normal and malignant human breast glandular tissue. J. Natl. Cancer Inst. 54:335-339. 24.Wang, B.L. and L.-I. Larsson. 1985. Simultaneous demonstration of multiple antigens by indirect immunofluorescence or immunogold staining. Novel light and electron microscopical double and triple staining method employing primary antibodies from the same species. Histochemistry 83:47-56. 25.Wang, B.L., L. Scopsi, M. Hartvig Nilsen, and L.-I. Larsson. 1985. Simplified purification and testing of colloidal gold probes. Histochemistry 83:109-115.

222

26.Wang, B.L., G.F. Springer, and S.C. Carlstedt. 1997. Quantitative computerized image analysis of Tn and T (Thomsen-Friedenreich) epitopes in prognostication of human breast carcinoma. J. Histochem. Cytochem. 45:1393-1400. 27.Wang, B.L., G.F. Springer, and L.C. Harwick. 1998. T (Thomsen-Friedenreich) and Tn epitope location and their spatial relations to adhesion plaques on human breast carcinoma cells: immunogold–silver staining studies at scanning electron microscopic level. J. Submicrosc. Cytol. Pathol. 30:503-509.

16

Immunogold–Silver Scanning Electron Microscopy Using Glycerol Liquid Substitution ∨

Jozef Samaj, Hans-Jürgen Ensikat, Wilhelm Barthlott, and Dieter Volkmann

INTRODUCTION Scanning electron microscopy (SEM) in combination with cytochemistry based on gold–silver techniques is a method of choice for detection and visualization of diverse molecules (polysaccharides, proteins, proteoglycans, and glycoproteins) located at cell surfaces. The main advantage of this combination resides in the three-dimensional view offered by SEM and the possibility to study at high resolution topographical distributions of target molecules tagged with gold–silver labeling at relatively broad surface or even intracellular areas. Immunogold SEM is a valuable method because it shows the marker distribution on entire tissue and/or cell surfaces, which is otherwise difficult to reconstruct from sections prepared for transmission electron microscopy. Gold particles are suitable and stable markers for SEM because of their chemical properties, uniform size, and strong emission of secondary and backscattered electrons. They can be easily prepared or commercially

0-8493-1392-9/02/$0.00+$1.50 © 2002 by CRC Press LLC

obtained in a wide range of sizes from 1 to 150 nm. For the first time, the powerful combination of SEM and cytochemistry was used in the middle 1970s by Horisberger et al.23,24 and Horisberger and Rosset,22 to localize polysaccharides on yeast cells and protoplasts using gold particles as labels for lectins or antibodies. Soon after, the immunogold SEM method was used for investigating animal and human biological samples.16,17,25,35 Exceptionally, plant samples (protoplasts) were also investigated using gold-tagged lectin (concanavalin A) in SEM.7,8 It was also demonstrated that immunogold SEM is suitable for localization of two or more antigens at cell surfaces.23,25,27 In the first attempts for immunogold-SEM studies, a standard secondary electron detector and metal- or carbon-coated samples were used. The immunogold-SEM method was further improved when the back-scattered electron imaging (BEI) mode was first employed in SEM by Trejdosiewicz et al.49 BEI is based on the high electron back-scattering coeffi-

223

Gold and Silver Staining cient of gold related to its high atomic number (Z = 79). This approach considerably increases contrast during gold marker detection on biological samples, which all have low back-scattering coefficients. Moreover, when the BEI signal was mixed with the secondary electron imaging (SEI) signal, an optimum correlation between the marker distribution and preservation of cell surface molecules was demonstrated on carboncoated11,47 and chromium-coated19 samples. Rather big gold particles (about 50 nm in diameter) were considered to be of optimal size for immunogold SEM, especially for overview examinations at lower magnifications. Smaller gold particles were not clearly visible in conventional microscopes. However, Walther et al.50 and Walther and Müller51 used an optimized BEI detector and showed that 5 to 15 nm gold particles can be unambiguously detected by SEM. The invention of silver enhancement (autometallography; AMG) helped to further improve the immunogold-SEM method. AMG can enhance the signal by precisely increasing the size of very small gold particles (1–5 nm in diameter) to specific desirable sizes.4,10,45 Because of steric hindrance effects, the bigger colloidal gold particles have limited access to the target molecules and therefore can label fewer epitopes–antigenic sites than smaller gold particles.26,51 When the gold particle size is only 1 to 5 nm compared to larger ones, the labeling density is increased.20,21 Furthermore, the AMG-amplified gold particles, in comparison to their gold counterparts of similar size, can be better recognized and viewed with BEI.19,45 Recently, it was demonstrated that glycerol, which has a very low vapor pressure and sufficient electrical conductivity, can be used to substitute water in biological samples.13,37 The use of liquid substitution brings a further improvement for broad range of specimens. Such samples infiltrated 224

with glycerol can be observed directly in the SEM without any drying. Due to the inherent electrical conductivity of glycerol, the samples can be examined without conductive coating, which improves the contrast of the marker. Furthermore, liquid substitution avoids the appearance of certain artificial structures, which occur during critical point drying or freeze drying typically, on specimens with a high water content, e.g., on mucilage layers. Glycerol infiltrated samples usually have smoother surfaces on which gold markers are clearly visible. We developed a simple and reliable protocol based on glycerol substitution for immunogold–silver localization of plant surface molecules at the SEM level. This method was applied successfully for the localization of different arabinogalactan– protein epitopes at surfaces of plant cells.41,42 In this chapter, we provide a detailed procedure and application of this immunolocalization SEM method as applied for plant cells. We also discuss its possible applications for different biological fields including immunolocalization of other surface-located molecules (e.g., proteoglycans, proteins, or glycoproteins) in various eukaryotic cells. PROTOCOLS Plant Material and Tissue Culture Grains of maize (Zea mays L., cv. Alarik) obtained from Force Limagrain (Darmstadt, Germany) were soaked and germinated in moist filter paper at 24°C in darkness. Seedlings with straight primary roots, 30 to 35 mm long, were selected for immunolabeling. Maize embryogenic callus was induced as described previously39 and maintained on MS medium33 supplemented with 12.7 µM 2,4-dichloro-phenoxy acetic acid and 0.088 mol/L sucrose at 24°C in darkness.

Immunogold–Silver Scanning Electron Microscopy Materials and Reagents • Use deionized (preferably double-)distilled water throughout the protocol. • Glass vials (approximately 10-mL capacity) with caps. • Single-edge razor blades. • Dissection instruments (tweezers, scalpel, needle). • Pasteur pipets. • Fabric (linen, cotton). • Membrane filters (nitrocellulose, 0.1 µm). • Desiccator. • Silica gel (as desiccant). • Conductive glue (conductive carbon or silver dag). • Filter paper. • Dissection stereomicroscope (e.g., Wild, Heerbrugg, Switzerland). • Device for continuous glycerol infiltration (Figure 16.1; Plano W Plannet GmbH, Marburg, Germany). • Coolable SEM specimen holder;37

Figure 16.1. Scheme of device for continuous glycerol infiltration of samples.

Plano W Plannet GmbH). • Scanning electron microscope: the Stereoscan S200 SEM (Leo, Cambridge, England, UK) was used here, equipped with a detector for backscattered electrons. • Glycerol (87%, 100%). • Phosphate-buffered saline (PBS) (140 mM NaCl, 2.7 mM KCl, 6.5 mM Na2HPO4•2H2O, 1.5 mM KH2PO4, 3 mM NaN3, pH 7.3). • Cytoskeleton-stabilizing buffer (CSB) [50 mM piperazine-N,N′-bis(2-ethane sulfonic acid (PIPES), 5 mM MgSO4, 5 mM EGTA, 3 mM NaN3, pH 7.0]. First prepare 0.5 mM stock solution of EGTA (3.8 g of EGTA in 10 mL of distilled water, add solid KOH until solution is clear, add distilled water to 20 mL, set pH to 8.0 using KOH). Put 15.1 g of PIPES free acid to 700 mL of distilled water, add solid KOH until solution is clear, add 1.23 g of MgSO4•7H2O, 0.2 g of NaN3, and 10 mL of 0.5 M EGTA stock solution, add water to final volume 1000 mL, set pH to 7.0 using KOH. • Fixation buffer (3.7% formaldehyde and 0.25% glutardialdehyde in PBS or CSB). • Aldehyde-blocking buffer (50 mM glycine in PBS). • Protein-blocking buffer: 5% bovine serum albumin (BSA) (Fraction V; Sigma, St. Louis, MO, USA) and 0.1% fish gelatin in PBS. • Incubation buffer (1% BSA in PBS). • Silver enhancement kit (BioCell, Cardiff, England, UK) or one of the AMG developers as described in this book. • Primary antibodies. We used monoclonal antibodies raised in rat against three different arabinogalactan–proteins epitopes: JIM4 and MAC207 recognizing carbohydrate epitopes with 225

Gold and Silver Staining GlcpA-β(1→3)-D-GalpA-α(1→2)-LRha residues and LM2 recognizing carbohydrate epitope-containing β-linked glucuronic acid, kindly provided by Dr. Paul Knox, University of Leeds, England, UK. • Secondary antibodies (goat or rabbit antirat IgGs adsorbed to 5 nm gold particles; BioCell). Protocol for Tissue Preparation and Immunolabeling Procedure All steps are performed at room temperature. 1. Cut pieces of roots (maximum of 10 mm long) and dissect callus pieces possessing several individual clumps of cells (maximum of 5 × 5 × 5 mm) using dissection instruments and razor blades directly into plastic petri dishes filled with fixation buffer. 2. Transfer samples to glass vials and fix with fixation buffer using vacuum infiltration for 1 h. 3. Wash out fixation buffer thoroughly with PBS (4 times for 10 min each). 4. Block aldehydes with aldehyde-blocking buffer for 30 min. 5. Rinse samples with PBS for 10 min. 6. Block proteins with protein-blocking buffer for 30 min. 7. Incubate samples with primary antibodies (JIM4, MAC207, and LM2) diluted 1:50 in incubation buffer for 1 h. 8. Wash samples with PBS (6 times for 5 min each). 9. Incubate samples with secondary antibodies (adsorbed to 5 nm colloidal gold particles) diluted 1:100 in incubation solution for 1 h. 10. Wash samples again with PBS (4 times for 10 min each) and with deionized dis226

tilled water (4 times for 10 min each). 11. AMG amplify gold particles with silver enhancement kit according to manufacturer instructions (BioCell). Alternatively, other AMG developers can be used. Optimal enhancement time depends on temperature and should be tested experimentally for every new tissue type. Optional: A better control of size and uniformity of silver-enhanced particles can be achieved when the protecting colloid gum Arabic resuspended in sodium citrate buffer (0.735 g to 25 mL of citrate buffer, pH set to 5.5 with citric acid, store at -20°C) is added to the mixture of two components of silver enhancement kit in the ratio 1:1 (vol/vol). Further information about acidic developers containing gum Arabic is available in literature.10,45 12. Stop silver enhancement by rinsing samples with deionized distilled water. 13. Substitute samples with glycerol according to the method described by Ensikat and Barthlott.13 Glycerol infiltration can be performed either with immersed samples (stepped procedure) or with samples placed on fabric (glycerol infiltration from below; continuous procedure). For immersed samples, use graded glycerol solutions (10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% and 87% for 1–3 h each). For infiltration from below, use device for continuous substitution of water by glycerol (Figure 16.1). Unicellular organisms and small samples should be first immobilized on nitrocellulose membrane.37 14. Remove excess glycerol by blotting samples on filter paper (check under stereomicroscope). 15. Mount samples on SEM metal stubs using conductive glue. 16. Store samples in desiccator with silica gel.

Immunogold–Silver Scanning Electron Microscopy Controls As negative controls, samples incubated in the absence of primary antibody were used in order to check nonspecific binding of secondary antibody. The negative control used for the maize roots samples was the incubation of the JIM4 antibody, which does not bind to roots and root hairs.42 As positive control for maize root hairs, the MAC 207 antibody, which does not bind to root hairs, but binds to the remaining trichoblast surface, was used.42 As negative immunodepletion control for maize root hairs, the LM2 binding to root hairs can be avoided by pre-incubating of the LM2 antibody with the corresponding hapten inhibitor (methyl-β-D-glucuronoside) containing glucuronic residues.46 A mixture of LM2 and hapten (applied as a first antibody) should give negative results. Microscopy Gold–silver-labeled glycerol-substituted specimens are observed using the BEI for marker detection and the SEI for topographical view. Due to the electrical conductivity of the glycerol, the samples do not need a conductive coating. Since the specimens are usually covered by a thin glycerol layer, this lasts only a few minutes until it evaporates in the high vacuum in the SEM, and the specimen surfaces become visible. When the surfaces are dry, the specimen can be transferred to a cooled holder to reduce further evaporation of glycerol and to increase the sample stability (between 0°–10°C, it is thermally isolated; keep sample cold for longer observations). The contrast of such uncoated samples in SEI is often favorable at low acceleration voltages of 5 to 10 kV. Specimens are usually stable for observation for 30 to 120 min (due to the slow glycerol evaporation). Afterwards, they can be rewetted with new glycerol if necessary.

Optional Alternative for Sample Preparation Procedure for Critical Point Drying 1. Cut pieces of roots (maximum of 10 mm long) and dissect callus pieces possessing several individual clumps of cells (maximum 5 × 5 × 5 mm) using dissection instruments and razor blades directly into a plastic petri dish filled with fixation buffer. 2. Transfer samples to glass vials and fix with fixation buffer using vacuum infiltration for 1 h. 3. Wash away fixation buffer thoroughly with PBS (4 times for 10 min each). 4. Dehydrate samples with graded acetone or ethanol (15%, 30%, 70%, 90%, 96%, and 3 × 100% for 30 min each). 5. Change ethanol for amyl acetate. 6. Dry samples using critical point dryer (e.g., CPD 020; Balzers Instruments, Balzers, Liechtenstein) with CO2. 7. Mount samples on SEM metal stubs using conductive glue. 8. Coat samples with thin layer of gold (10–15 nm) using a sputter coater. RESULTS Reliability of Glycerol Substitution Method In preliminary experiments, we have tested the structural preservation of the mucilage surfaces of calli and roots using two sample preparation techniques: (1) critical point drying, and (2) glycerol substitution.41 While critical point drying causes artifactual appearance of mucilage surface layers, the glycerol substitution method provides an excellent mucilage surface preservation, which is likely much closer to native conditions, both in calli 227

Gold and Silver Staining and roots. This is due to the gentle and continuous replacement of water by glycerol and absence of harsh drying in the glycerol-based procedure. In our studies, these results unambiguously confirmed the reliability of the glycerol substitution method used for sample preparation. The surfaces of glycerol-substituted samples appeared smoother, which helped in the visualization of the silver-enhanced gold particles after immunolabeling. Localization of JIM4 Epitope at Surfaces of Callus Cells Silver-amplified gold particles associated with JIM4 epitope occurred preferentially at surfaces of embryogenic callus cells (Figures 16.2 A, B, and D; arrows). In controls (with primary antibody omitted during immunolabeling), embryogenic callus cells were either not or only weakly labeled, confirming the specificity of the immunogold-SEM method (Figure 16.2C). The immunogold–silver labeling itself has not altered the surface structure of embryogenic cells if compared to control non-labeled cells substituted with glycerol (data not shown; see Figure 16.1E in Reference 41). Detailed investigations revealed that individual silver-enhanced gold particles depicting JIM4 epitope are aligned along filamentous structures located at the surface of embryogenic cells (Figure 16.2D, arrows). Localization of LM2 and MAC207 Epitopes at Root Hair and Root Surfaces The specific localization of LM2 epitope at root hair surfaces was characteristic for all of their developmental stages. It was first visualized using the immunogold–silver technique in combination with stereomicroscope observation.42 For more 228

detailed studies, we used the high resolution technique of glycerol substitution of immunolabeled samples for SEM. This technique allowed good structural preservation of young initiating root hairs. Immunogold SEM revealed that the LM2 epitope is abundant at the surfaces of initiating root hairs, but almost absent from the remaining trichoblast surface (Figure 16.2E). In some cases, this epitope was enriched on tips of initiating hairs, while in others it was not, and this is probably related to different growth characteristics of individual root hairs. Already in young root hairs, the LM2 epitope was deposited in the outer root hair cell wall with a specific pattern represented by holes without LM2 epitope and curly bands organized in network-like structures enriched with the LM2 epitope (Figure 16.2E). As a positive control, we used the MAC207 antibody, which labels root surface but not root hairs, and as a negative control, we applied JIM4 antibody, which does not bind to the root surface (data not shown). The glycerol substitution method also allowed the preservation of fine root surface structures including the mucilage layer covering the root cap and meristematic epidermal cells (Figure 16.2F). We found that the MAC207 epitope predominantly occurs at the root epidermal surface (Figure 16.2G), though it was occasionally visualized at the surface of root cap mucilage. Detailed investigations revealed that the MAC207 epitope was arranged along filament-like structures and stripes at the epidermal surface, and these were oriented along with the root growth axis (Figure 16.2G). TECHNICAL HINTS AND DISCUSSION Arabinogalactan–proteins (AGPs) represent an immensely heterogeneous class of

Immunogold–Silver Scanning Electron Microscopy

Figure 16.2. Immunolocalization of different AGP epitopes at surfaces of maize callus cells (A–D), root hairs (E), and root epidermis (F–G) as revealed by immunogold–silver SEM based on glycerol substitution. (A, B, and D) Labeling of embryogenic cells with JIM4 antibody shows that the JIM4 epitope is aligned along filamentous structures (see arrows on panel D). Panels A and D represent mixed BEI/SEI images, and panel B represents a BEI image corresponding to panel A. Panel D shows a detail of panel A with labeled filamentous structures (small arrows). (C) Control embryogenic callus cells treated without primary antibody show very little nonspecific immunolabeling (mixed BEI/SEI image). (E). Labeling of root hair (indicated by asterisk) with LM2 antibody. (F and G) SEI (F) and corresponding BEI (G) observations of root surface labeled with MAC207. Note that only the epidermal root surface contains this epitope. Root cap covered with a thick mucilage layer is indicated by asterisk. Calibration bars: 2.5 µm for panels A, B, and C; 1 µm for panel D; 6 µm for panel E; and 20 µm for panels F and G. Panels A, B, D, F, and G are reproduced from Reference 41 with kind permission of Blackwell Science Ltd., and panel E is reproduced from Reference 42 with kind permission of The Japanese Society of Plant Physiologists.

229

Gold and Silver Staining plant-specific proteoglycans.29 Using histochemical and immunolocalization approaches, AGPs have been located at surfaces of plant organs, e.g., maize roots,3,41 where they form a continuous sheet around the root body surrounding it from environment. This layer, the outer pellicle, has fibrillar organization when observed with transmission electron microscopy.1 AGPs have been found in cell walls of tip-growing cells like germinating pollen tubes30,38 and root hairs.42 They were also identified as a component of secretion products of the transmitting tract9 and root cap52 cells. Plant cells cultured in vitro possess different AGPs in their cell walls43,48 or secrete some AGPs to culture media.34 The precise function of AGPs remains obscure, but data supporting their possible role in cell–cell signaling32,44,53 and plant morphogenesis34 have been accumulating recently. Some nondirect evidence also indicates possible links of AGPs to the cytoskeleton. For example, microtubular cytoskeletal toxins (such as colchicine and trifluoralin) and cold treatment can destroy outermost extracellular layers,5,12 which were shown to be composed of AGPs in some plant cells cultured in vitro, e.g., in maize callus.43 Previously, it has been found that outer proteinaceous extracellular matrix layers and networks are also structural markers of maize embryogenic callus.39,43 It has been suggested that they play an important role in plant morphogenesis.40 On the other hand, inhibitors of AGP biosynthesis, such as dihydroxyprolin, can affect the actin cytoskeleton (F. Balus∨ka and D. Volkmann, personal communication). However, plasma membrane receptor molecules (similar to integrins in animal cells), that link AGPs to the cytoskeleton, remain to be identified in plants. Until now, AGPs were localized in situ using histochemistry with β-glucosyl Yariv reagent or immunofluorescence and immu230

nogold transmission electron microscopy with AGP-specific antibodies.28,34 Here, we tested the reliability of glycerol substitution for structural preservation of mucilage at plant surfaces, and we developed an associated immunogold labeling technique for arabinogalactan–proteins. We studied the spatial distributions of AGP epitopes on the surface of maize embryogenic calli and roots using monoclonal antibodies JIM4, LM2, and MAC207. For this purpose, a new immunogold-SEM method was employed, which is based on liquid substitution of samples with glycerol.41 We showed that AGPs are components of the outer extracellular fibrillar layers and networks that cover the surface of roots41 and cells undergoing somatic embryogenesis.41,43 Simultaneously with technical improvements, the immunogold-SEM method came to be used in immunolocalization studies performed on various biological samples. Immunocytochemistry in combination with SEM was used for investigations of topographical distributions of different molecules at surfaces of prokaryotic cells like Escherichia coli45 and various eukaryotic cells including yeast, plant, animal, and human cells.14,20,21 Different target molecules were visualized and studied using gold-tagged lectins or antibodies, including different polysaccharide moieties (to which specific lectins bind), proteins, glycoproteins, proteoglycans, and enzymes (e.g., fibronectin, collagen, acid phosphatase, acetyl cholinesterase). Most of the studies were focused on cell surface receptor–ligand interactions, expression of cell surface lectin-binding sites, and surface distribution of extracellular matrix (ECM) components. For example, Trejdosiewicz et al.49 performed immunogold SEM for detection of fibronectin, a human fibroblast cell-surface molecule. These authors also compared the above mentioned

Immunogold–Silver Scanning Electron Microscopy method with immunoperoxidase and immunofluorescence labeling techniques. They found immunogold SEM to be the superior method because of its high resolution and possibility to quantify labeling density. Later, Birrell et al.4 used silver enhancement of colloidal gold in combination with photoelectron microscopy for investigation of human fibronectin at fibroblast surfaces. They found that silver enhancement increased label contrast, leading to better fibronectin visualization. However, the application of immunogold SEM is not restricted exclusively to cell surfaces. It was used also for studies of cytoskeleton components like microtubules2,15,21 and actin microfilaments,36 and intracellular antigens like cytokeratin.20 The method was also applied to study intracellular ligand epitopes.18 There is a variety of sample preparation techniques for SEM based on air drying, critical point drying or freeze drying, which can be combined with immunogold labeling. The proper choice is dependent on the sample nature and on the goals of investigation. Here, we employed the glycerol substitution method for sample preparation of immunogold-labeled samples to study plant cell surfaces. This method has certain advantages for observation and imaging of biological samples. It is suitable especially for examinations of dry plant surfaces (e.g., epicuticular wax on the leaf surface), unicellular organisms and small samples (immobilized to nitrocellulose membrane), and samples with high water content (e.g., gels and mucilage).13,41 Importantly, it also provides necessary resolution for molecular cytological studies, and it can be combined with elemental analysis (EDX) or immunogold–silver techniques including statistical (computerized) evaluation of label density over larger areas. We demonstrated that glycerol substitution can preserve fragile immunolabeled surface structures like fila-

ments with associated AGP epitopes41 or intracellular actin microfilaments.36 In concert with others,11,49 we have shown that BEI noticeably increased the intensity and contrast of the signal and allowed the unambiguous identification of the immunogold labeling. When BEI was used in combination with SEI (as a mixed signal), it revealed better corresponding cell morphology in a material-dependent contrast. However, the glycerol substitution method also has some shortcomings. It seems to be best suited for rather flat samples. Some soft and complex structures may collapse, and fine details are sometimes hardly visible (e.g., microvilli of human or animal cells can be flattened). Some of these problems can be reduced or avoided, however, by optimization of the fixation procedure, use of continuous rather than stepped glycerol substitution, or alternatively, by substitution with triethylene glycol, which has lower viscosity as glycerol and therefore penetrates better to the sample.13 Nevertheless, the method has been used successfully for a wide range of different specimens, such as plant leaves and roots, human erythrocytes, diverse animal tissues, cell cultures, unicellular organisms, and also nonbiological water-containing samples. We recommend that in specific cases, results of both drying and liquid substitution should be compared. In conclusion, our new silver-enhanced immunogold-SEM method for plant surfaces, based on glycerol substitution, provides sufficient resolution as well as necessary structural and antigenic preservation of biological samples and can be used for diverse surface and intracellular antigens of prokaryotic and eukaryotic cells (including bacteria, yeast, lower and higher plants, animals, and humans). Importantly, samples substituted with glycerol 231

Gold and Silver Staining do not need metal coating, which could mask smaller gold or silver-enhanced gold particles. Thus, gold–silver markers of antigenic sites can be detected with high reliability. This method can be advantageous for fine immunolocalization studies, because fragile thin surface layers and filamentous structures can easily be altered or damaged by other preparation techniques, e.g., by critical point drying conventionally used for SEM,6 by immersing specimens in cooling agent or freeze drying used for low temperature SEM13 or by exposure of the sample to low temperature.31 We hope that recent improvements of the immunogold–silver SEM methods open the way for its routine use in topographical immunolocalization studies coupled to stereo-pair SEM imaging and computerized quantitative analysis of label density over the entire cell surface or relatively large tissue area. ACKNOWLEDGMENTS ∨

J.S. is grateful to the Alexander von Humboldt Foundation (Bonn, Germany) for his research fellowship. This work was supported by Agravis (Bonn, Germany) through Deutsche Agentur für Raumfahrtangelegenheiten (DARA, Bonn, Germany) and the Ministerium für Wissenschaft und Forschung (MWF, Düsseldorf, Germany). We thank Dr. Paul Knox (University of Leeds, England, UK) for providing us with the monoclonal antibodies JIM4, MAC207, and LM2, and for his collaboration. REFERENCES 1.Abeysekera, R.M. and M.E. McCully. 1993. The epidermal surface of the maize root tip. I. Development in normal roots. New Phytol. 125:413-429. 2.Albrecht, R.M., J. Prudent, S.R. Simmons, J. Pawley, and J. Choate. 1989. Observations of colloidal gold labelled platelet microtubules: high voltage electron

232

microscopy and low voltage-high resolution scanning electron microscopy. Scanning Microsc. 3:273-278. 3.Bacic, A., S.F. Moody, and A.E. Clarke. 1986. Structural analysis of secreted root slime from maize (Zea mays L.). Plant Physiol. 80:771-777. 4.Birrell, G.B., D.L. Habliston, K.K. Hedberg, and O.H. Griffith. 1986. Silver-enhanced colloidal gold as a cell surface marker for photoelectron microscopy. J. Histochem. Cytochem. 34:339-345. ∨ 5.Bobák, M., A. Hlavacka, M. Ovecka, and J. Samaj. 1999. Effect of trifluralin and colchicine on the extracellular matrix surface networks during early stages of direct somatic embryogenesis of Drosera rotundifolia L. J. Plant Physiol. 155:387-392. 6.Boyde, A. 1978. Pros and cons of critical point drying and freeze drying for SEM. Scan. Electron Microsc. 303-314. 7.Burgess, J. and P.J. Linstead. 1976. Ultrastructural studies of the binding of Concanavalin A to the plasmalemma of higher plant protoplasts. Planta 130:7983. 8.Burgess, J. and P.J. Linstead. 1977. Membrane mobility and the Concanavalin A binding system of the plasmalemma of higher plant protoplasts. Planta 136:253259. 9.Cheung, A.Y., H. Wang, and H.M. Wu. 1995. A floral transmitting tissue-specific glycoprotein attracts pollen tubes and stimulates their growth. Cell 82:383-393. 10.Danscher, G. 1981. Histochemical demonstration of heavy metals. A revised version of the sulphide silver method suitable for both light and electron microscopy. Histochem. 71:1-16. 11.De Harven, E., R. Leung, and H. Christensen. 1984. A novel approach for scanning electron microscopy of colloidal gold-labeled cell surfaces. J. Cell Biol. 99:5357. 12.Dubois, T., J. Dubois, M. Guedira, A. Diop, and J. Vasseur. 1992. SEM characterization of an extracellular matrix around somatic proembryos in roots of Cichorium. Ann. Bot. 70:119-124. 13.Ensikat, H.J. and W. Barthlott. 1993. Liquid substitution: a versatile procedure for SEM specimen preparation of biological materials without drying and coating. J. Microsc. 172:195-203. 14.Erlandsen, S.L., C. Frethem, and R. Aurata. 1990. Workshop on high-resolution immunocytochemistry of cell surfaces using field emission SEM. J. Histochem. Cytochem. 38:1779-1780. 15.Goode, D. and T.K. Maugel. 1987. Backscattered electron imaging of immunogold-labelled and silver enhanced microtubules in cultured mammalian cells. J. Electron Microsc. Tech. 5:263-273. 16.Goodman, S.L., G.M. Hodges, L.K. Trejdosiewicz, and D.C. Livingston. 1979. Colloidal gold probes. A further evaluation. Scan. Electron Microsc. 2:619-628. 17.Goodman, S.L., G.M. Hodges, and D.C. Livingston. 1980. A review of the colloidal gold marker system. Scan. Electron Microsc. 3:133-146. 18.Gross, D.K. and U. De Boni. 1990. Colloidal gold labeling of intracellular ligands in dorsal root sensory neurons, visualized by scanning electron microscopy. J. Histochem. Cytochem. 38:775-784.

Immunogold–Silver Scanning Electron Microscopy 19.Herter, P., G. Laube, J. Gronczewski, and W.W. Minuth. 1993. Silver-enhanced colloidal-gold labeling of rabbit kidney collecting-duct cell surfaces imaged by scanning electron microscopy. J. Microsc. 171:107115. 20.Hodges, G.M., J. Southgate, and E.C. Toulson. 1987. Colloidal gold—a powerful tool in scanning electron microscope immunocytochemistry: an overview of bioaplications. Scan. Microsc. 1:301-318. 21.Hodges, G.M. and K.E. Carr. 1990. Colloidal gold immunocytochemistry using SEM. Eur. Microsc. Anal. 6:17-20. 22.Horisberger, M. and J. Rosset. 1976. Localization of wheat germ agglutinin receptor sites on yeast cells by scanning electron microscopy. Experientia 32:9981000. 23.Horisberger, M., J. Rosset, and H. Bauer. 1975. Colloidal gold granules as markers for cell surface receptors in the scanning electron microscopy. Experientia 31:1147-1149. 24.Horisberger, M., J. Rosset, and H. Bauer. 1976. Localization of mannan at the surface of yeast protoplasts by scanning electron microscopy. Arch. Microbiol. 109:9-14. 25.Horisberger, M. and J. Rosset. 1977. Gold granules, a useful marker for SEM. Scan. Electron Microsc. 2:7582. 26.Horisberger, M. and M. Tacchini-Vonlanthen. 1983. Stability and steric hindrance of lectin-labelled gold markers in transmission and scanning electron microscopy, p. 189-197. In T.C. Bog-Hansen and G.A. Spengler (Eds.), Lectins, Vol. 3. De Gruyter, Berlin. 27.Hoyer, L.C., J.C. Lee, and C. Bucana. 1979. Scanning immunoelectron microscopy for identification and mapping of two or more antigens on cell surfaces. Scan. Electron Microsc. 3:629-636. 28.Knox, J.P. 1997. The use of antibodies to study the architecture and developmental regulation of plant cell walls. Int. Rev. Cytol. 171:79-120. 29.Knox, J.P. 1999. Intriguing, complex, and everywhere: getting to grips with arabinogalactan-proteins. Trends Plant Sci. 4:123-125. 30.Li, Y.Q., C. Faleri, A. Geitmann, H.-Q. Zhang, and M. Cresti. 1995. Immunogold localization of arabinogalactan proteins, unesterified and esterified pectins in pollen grains and polllen tubes of Nicotiana tabacum L. Protoplasma 189:26-36. 31.Lin, P.S., D.F.H. Wallach, and S. Tsai. 1973. Temperature-induced variations in the surface topology of cultured lymphocytes are revealed by scanning electron microscopy. Proc. Natl. Acad. Sci. USA 70:2492-2496. 32.McCabe, P.F., T.A. Valentine, S. Forsberg, and R.I. Pennell. 1997. Soluble signals from cells identified at the cell wall establish a developmental pathway in carrot. Plant Cell 9:2225-2241. 33.Murashige, T. and F. Skoog. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant 15:473-497. 34.Nothnagel, E.A. 1997. Proteoglycans and related components in plant cells. Int. Rev. Cytol. 174:195-291. 35.Nurdin, A.T., M. Horisberger, E. Savariau, and J.P. Caen. 1980. Visualization of lectin binding sites on the

surface of human platelets using lectins absorbed to gold granules. Experientia 36:1215-1217. 36.Reichelt, S., H.J. Ensikat, W. Barthlott, and D. Volkmann. 1995. Visualization of immunogold-labelled cytoskeletal proteins by scanning electron microscopy. Eur. J. Cell Biol. 67:89-93. 37.Robards, A.W. and A.J. Wilson. 1996. Procedures in Electron Microscopy. Liquid Substitution Methods, p. 67-76. John Wiley & Sons, Chichester. 38.Roy, S., G.Y. Jauh, P.K. Hepler, and E.M. Lord. 1998. Effects of Yariv phenylglycoside on cell wall assembly in the lily pollen tube. Planta 204:450-458. ∨ 39.Samaj, J.,∨ M. Bobák, A. Blehová, J. Kris∨tín, and O. Auxtová-Samajová. 1995. Developmental SEM observations on an extracellular matrix in embryogenic calli of Drosera rotundifolia and Zea mays. Protoplasma 186:45-49. ∨ 40.Samaj, J., M. Bobák, M. Ovecka, A. Blehová, and A. Pretová. 1997. Structural Features of Plant Morphogenesis In Vitro, p. 122. Veda, Bratislava. ∨ 41. Samaj, J., H.J. Ensikat, F. Balus∨ka, J.P. Knox, W. Barthlott, and D. Volkmann. 1999a. Immunogold localization of plant surface arabinogalactan-proteins using glycerol liquid substitution and scanning electron microscopy. J. Microsc. 193:150-157. ∨ 42.Samaj, J., M. Braun, F. Balus∨ka, H.J. Ensikat, Y. Tsumuraya, and D. Volkmann. 1999b. Specific localization of arabinogalactan-protein epitopes at the surface of maize root hairs. Plant Cell Physiol. 40:874883. ∨ 43.Samaj, J., F. Balus∨ka, M. Bobák, and D. Volkmann. 1999c. Extracellular matrix surface network of embryogenic units of friable maize callus contains arabinogalactan-proteins recognized by monoclonal antibody JIM4. Plant Cell Rep. 18:369-374. 44.Schultz, C., P. Gilson, D. Oxley, J. Youl, and A. Bacic. 1998. GPI-anchors on arabinogalactan proteins: implications for signalling in plants. Trends Plant Sci. 3:426431. 45.Scopsi, L., L.I. Larsson, L. Bastholm, and M.H. Nielsen. 1986. Silver-enhanced colloidal gold probes as markers for scanning electron microscopy. Histochemistry 86:35-41. 46.Smallwood, M., E.A. Yates, W.G.T. Willats, H. Martin, and J.P. Knox. 1996. Immunochemical comparison of membrane-associated and secreted arabinogalactan-proteins in rice and carrot. Planta 198:452-459. 47.Soligo, D., E. de Harven, M.T. Nava, and G. Lamberthenghi-Deliliers. 1986. Immunocytochemistry with back-scattered electrons, p. 289-297. In M. Mueller, R.P. Becker, A. Boyde, and J.J. Wolosewick (Eds.), The Science of Biological Specimen Preparation. SEM. AMF O’Hare, Chicago. 48.Stacey, N.J., K. Roberts, and J.P. Knox. 1990. Patterns of expression of the JIM4 arabinogalactan-protein epitope in cell cultures and during somatic embryogenesis in Daucus carota L. Planta 180:285-292. 49.Trejdosiewicz, L.K., M.A. Smolira, G.M. Hodges, S.L. Goodman, and D.C. Livingston. 1981. Cell surface distribution of fibronectin in cultures of fibroblasts and bladder derived epithelium: SEM-immunogold localization compared to immunoperoxidase and

233

Gold and Silver Staining immunofluorescence. J. Microsc. 123:227-236. 50.Walther, P., S. Kriz, M. Müller, B.H. Ariano, U. Brodbeck, P. Ott, and M.E. Schweingruber. 1984. Detection of protein A gold 15 nm marked surface antigens by backscattered electrons. Scan. Electron Microsc. 3:1257-1266. 51.Walther, P. and M. Müller. 1985. Detection of small (5-15 nm) gold-labelled surface antigens by backsattered electrons, p. 538-541. Proc. EMSA. San Francisco Press, San Francisco.

234

52.Winicur, Z.M., G.F. Zhang, and A. Staehelin. 1998. Auxin deprivation induces synchronous Golgi differentiation in suspension-cultured tobacco BY-2 cells. Plant Physiol. 117:501-513. 53.Youl, J.J., A. Bacic, and D. Oxley. 1998. Arabinogalactan-proteins from Nicotiana alata and Pyrus communis contain glycosylphosphatidylinositol membrane anchors. Proc. Natl. Acad. Sci. USA 95:7921-7926.

INDEX A

B

Acetic acid, as AMG stop reagent, 44 Adenohypophysis, double PAGS, 80 Adrenalin, 3 Adrenals, silver localization, 18 AGP epitopes, 228–230 Aldehyde-based fixatives, 63–64, 116, 170, 194–195, 200, See also Fixatives Alkaline phosphatase, 65, 154 Alzheimer’s disease, 24 Amines, silver staining, 5 Amino acids, See also Peptides; specific amino acids fixative-induced cross-linking, 64 Nanogold labeling, 93 Ammoniacal silver solution, 3, 6, 9–10 Ammonium molybdate, 105 Amyloid b Nanogold complex, 40 Antibodies, See also specific immunogold applications gold-labeled, 60–61, See also Immunogold staining (IGS) and labeling immunogold-silver staining applications, 59–60, See also Immunogold-silver staining PAGS specificity control, 76 Antigen detection, using PAGS, See Protein A-gold-silver (PAGS) staining Antigen labeling, dot blots for checking, 30 Antigen retrieval methods, 49–50, 53–54, 64 Antilaminin, 173 Apoptosis, 159 Arabinogalactan-proteins (AGPs), 228–230 Araldite, 147 Argentaffin/argyrophyllic reactions, 2 neuroendocrine cell types, 4 nonspecific reactions, 62, 72, 174 Argyrosis, 18 Astrocytes, 50, 156 Aurothioglucose, 16, 105 Aurothiomalate, 15, 16 Automated ISH, 142 Autometallography (AMG), 13–24, 29–45, 107, 128, 177, See Gold enhancement autometallography; Silver enhancement autometallography bismuth detection, 14, 20–21 Danscher’s developer, 16, 62, 72, 116, 212. See also Silver lactate gold localization, 15–16 IGSS, 61–63 metal selenides/sulfides, 13–15, 19–20, 22 schematic representation, 16 silver localization, 18–19 zinc localization, 21–23 Avidin-biotin-IGSS, 65

Back developing, 43 Backscattered electron image (BEI) mode, 219, 223–224 Benzamidine, 116 Beta (β) cells, 21 Beta-galactosidase, 65, 154 Beta-mercaptoethanol (BME), 39, 87 Biotin labeling ATP, 158 cDNA probes, 143 EM-ISH, 206 tyramides, 130, 134, 138, 189, See also Tyramide signal amplification Bismuth detection, 14, 20–21, 24 Blood-brain barrier gold penetration, 15 silver penetration, 18 Blotting IGSS applications, 65 silver-enhanced Nanogold, 30, 37–39 Brain bismuth localization, 20–21 BrdU visualization, 160 glial fibrillary acidic protein (GFAP), 156 indirect IGSS, 50 K+ channels, 34 mercury distribution, 20 neocortical vessels, 24 silver localization, 18–19 zinc localization, 21–24 BrdU, 158, 160 Breast carcinoma, 50, 52, 139–141 immunogold-silver staining applications for SEM, 212–219 Brightfield microscopy combined ISH-IHC applications, 128 gold visualization, 71 tubulin localization protocol, 180 two-color system, 141 Bromodeoxyuridine (BrdU), 158, 160 Bromohydroquinone, 58, 72 BSA, 148 BSPT, 154

C Canada balsam, 10 Cancer cells HBC 6558 micrograph, 219 HTB 24 micrograph, 218

235

Gold and Silver Staining immunogold-silver staining applications for SEM, 211–221 preparation protocol, 213–214 protocols, 212–218 results, 218–219 Carbowax, 39 Casein, as protein block, 125 Catalyzed reporter deposition (CARD), 128, 134, 189, See also Tyramide signal amplification immunogold detection protocol, 191–193 immunogold detection results, 193–194 incubation time, 196 Caveolin-1 localization, 180–181, 184–185 Cell conditioning, 130, 142 Cell cultures caveolin-1 localization protocol for HUVEC, 180–181, 184–186 HeLa cells for combined gold-fluorescent labeling, 109 Cell-molecular biology, combined immunogold labeling methods, 153–159, See also Immunogold staining (IGS) and labeling Cells in suspension, Nanogold labeling, 89 Cell surface components, 211 Cervical carcinoma, 36, 128 Chloride buffer removal, 32–33 Chlorolauric acid, 172 Cholecystokinin (CCK) cells, 4 Chromaffin cells, 2, 3, 5–6, 18 Chromagranin A, 4–5, 50 Chromogenic in situ hybridization (CISH), 129 Citrate buffer, 43, 48, 53–54, 72 Classical argyrosis, 18 Clustered gold, 145 immunoprobes, 177–186, See Immunogold staining (IGS) and labeling manufacturers, 54 Nanogold, See Nanogold CODFISH, 128, 130, 139–141 Colchicine, 230 Collagenous structures, 62, 174 Colloidal gold, 47, 145 AMG, 13, See also Silver enhancement autometallography antibody labeling, 60–61 electron microscopy applications, 177, 211, See Immunoelectron microscopy EM-ISH, 208 IGSS ultrastructural applications, 152–153, See also Immunogold-silver staining Nanogold double labeling, 86 SEM application, 219–220 indirect IGSS, 53–55, 57, 60 manufacturers, 54

236

Nanogold double labeling, 44, 86 Nanogold vs., 42, 51–52, 85, 99–100, 178, 195 particle size and ultrasmall particles, 100, 125, 177–178, 190, 194, 195 protein A-gold, 71, 74, See also Protein A-gold-silver (PAGS) staining vascular infusion, 23 Colloidal silver, 18, 19, 23 Color photography, 63 Concanavalin A, 223 Connective tissue, silver localization, 18 Controls EM-ISH, 205 IGSS SEM, 227 IGSS specificity, 60 immunogold staining, 153 PAGS specificity, 76 Coomassie® blue, 30 Copper sulfide or selenide, 14 Correlative fluorescence and electron microscopy, 185–186, See also Fluorescence microscopy; Multiple staining or labeling Critical point drying, 215–217, 221 Cryosection protocol for neutrophils, 181–182 Cysteine residue, Nanogold labeling, 93 Cy3-Nanogold-labeled streptavidin, 53, 110–115 Cytoseal, 10

D Danscher, Gorm, 1, 61 Desmin, 49 Detergents, 43, 59, 116, 173, 184 Developers, AMG, See Silver enhancement autometallography Development time, 41–42, 116, 220 PAGS, 74 3,3′-Diaminobenzidine (DAB), 63, 65, 153, 177, 189, 208 Digoxigenin, 132, 134, 136, 143, 158–159, 206 Dihydroxyprolin, 230 Dinitrophenol (DNP), 132 Dipalmitoyl phosphatidylethanolamine (DPPE) Nanogold, 96–97, 105 Disodium EDTA, 43 Disulfide bond reduction, 92, 93, 107 Dithiothreitol (DTT), 39, 87, 116 DNA apoptosis, 159 cautions for stop reagent use, 43 combined immunogold labeling and molecular biology, 158–159 in situ hybridization (ISH), 130 detection procedure, 203 EM-ISH, 199–209, See also Electron microscopy

Index in situ hybridization GenPoint kit and streptavidin-Nanogold, 138–139 HPV detection, 119–124 labeled tyramides and streptavidin-Nanogold, 134–136 probe labeling, 201–202, 206 results, 205 technical guidelines, 208 using streptavidin-Nanogold, 132–134 in situ PCR, 53 nick translation, 120–121, 158, 199, 201, 206 visualization in mammary carcinoma, 52 DNA polymerase, 158, 159, 201 Dopa, 3 Dopamine, 3, 4 Dot blots, 30 Double-face immunogold labeling, 148–149 Double labeling, See Multiple staining or labeling; specific techniques DPPE-Nanogold, 96–97, 105 DPX, 10 DTNB, 94

E EDTA-regressive staining, 199 Electron microscopy (EM), 177, See also Scanning electron microscopy; Transmission electron microscopy AMG development time and, 41–42 combined LM tubulin localization protocol, 178–180, 182–183 correlative fluorescence and electron microscopy, 181–182, 185–186 double-labeled Nanogold-colloidal gold, 86 gold-enhanced Nanogold, 30, 34–35, 41 immunogold labeling techniques, 145–167, See Immunoelectron microscopy; Immunogold staining (IGS) and labeling laser Nanogold-type probes, 116 Nanogold direct viewing, 97–99 Nanogold labeling for post-embedding, 89–90 Nanogold labeling for pre-embedding, 89 Nanogold visualization, 29, 86, 103, 105, 173–174 pre-embedding immunoelectron microscopy with immunogold labeling, 169–176, See Immunoelectron microscopy recommended silver enhancement procedure for Nanogold, 87 Electron microscopy in situ hybridization (EM-ISH), 199–209 controls, 205 DNA detection procedure, 203 fixation, 207 immunogold detection protocol, 204–205 ISH protocol, 202–203

nucleic acid probes, 201–202, 206 postembedding, 199–201 pre-embedding, 199, 205–206 protein digestion, 207–208 resolution issues, 208–209 results, 205 RNA detection procedure, 203–204 steps, 206–208 technical guidelines for nucleic acid detection, 208 Electron spectroscopic imaging, 149 Electrophoretic analysis, 40 Ellman’s reagent, 94 Endocrine organs, 16 Energy-filtered TEM, 29 Entellan, 10 Enterochromaffin cells, 2, 3 Enterochromaffin-like (ECL) cells, 5–6 Enzyme histochemistry combined immunogold labeling, 154 signal quality, 63 IGSS applications, 52 Epipolarization microscopy, 125 Epon, 146, 147 Epoxy resin, 146, 152 Epstein Barr virus, 129, 132 Estrogen receptors, 57 Eukitt, 10 Euparal, 10

F Fab′ fragments, 29, 42, 51, 102, 173 IGS spatial resolution, 149 monomaleimido-Nanogold labeling, 91–92 RBC, 103 Farmer’s solution, 44, 76 Fast red K, 141 Fc fragments, 82 Fibroblasts intra-nuclear staining, 134 silver localization, 18 Fibronectin, 230–231 FISH, 128, 130, 132, 139–142 CODFISH, 128, 130, 139–141 EM-ISH, 206 GOLDFISH, 141–142 Fixatives amino acid cross-linking, 64 EM-ISH, 207 gold-cluster microprobe methods and, 182 IGSS and, 63–64 IGSS SEM applications, 220–221 pre-labeling quenching, 87 RNA and, 136 silver staining for neuroendocrine cells, 2 Fluorescein isothiocyanate (FITC), See also

237

Gold and Silver Staining Fluorescence-gold multiple labeling; Fluorescence microscopy DNA ISH protocol using GenPoint kit and streptavidin-Nanogold, 138–139 RNA ISH protocol using labeled tyramides and streptavidin-Nanogold, 136 tyramide labeling, 130, 132 Fluorescein isothioriboprobes, 134, 143 Fluorescence, immunofluorescence vs. silver-enhanced gold, 183 Fluorescence-gold multiple labeling, 107–118, See also FluoroNanogold background signals, 116 double-labeled streptavidin-Nanogold, 61 immunolabeling cellular components, 108–110 ISH and detection protocol, 110–115 multiple immunostaining using IGSS, 65 nonspecific affinity for nuclear materials, 116 optimum conditions, 115–117 poor development, 116 Fluorescence in situ hybridization (FISH), 128, 130, 132, 139–142 Fluorescence microscopy caveolin-1 localization, 181 correlative electron microscopy, cryosectioned neutrophils, 181–182, 185–186 HeLa cells, 111 tubulin localization protocol, 179 FluoroNanogold, 53, 61, 132 caveolin-1 localization protocol, 181 combined EM-LM immunocytochemistry applications, 178–180 correlative EM, 182, 186 negative staining, 182 Formaldehyde, 63–64, 200 Freeze-etching, 155 Freeze-fracture replication, 155 Freezing protocol, 170 Frozen sections, unsuitability for immunogold-silver staining, 49, 63

G β-Galactosidase, 65, 154 Ganglion, 51 Gastric carcinoid, 8 Gastric inhibitory peptide (GIP) cells, 5 Gastrointestinal mucosa, neuroendocrine cell staining, 3, 4, 5, 7, 8 Gelatin, 39 Gels, silver-enhanced Nanogold, 30 GeneAmp®, 119 GenPoint® Catalyzed Signal Amplification (CSA) kit, 129, 134, 138–139 Glial fibrillary acidic protein (GFAP), 156 Glucagon, 4 Glucose-molybdate, 105

238

Glutardialdehyde, 63–64, 116, 170, 194–195, 200 Glycerine/gelatin, 10 Glycerol liquid substitution, 224, 231–232 reliability, 227–228, 230 Gold-palladium sputter-coating protocol, 217–218 Gold blood-brain barrier and, 15 clusters, See Clustered gold; Nanogold colloidal, See Colloidal gold light microscopy visualization, 48, 71 localization by AMG, 15–16 sulfides/selenides, 14 transplacental transport, 16 Gold enhancement autometallography (AMG), 17, 30, 41, 190 immunogold detection, 193, 196 incubation time, 196 in situ hybridization specimens, 35 protocol for EM, 34–35 protocol for LM, 35–37 silver enhancement vs., 41 stop reagent, 43 Gold immunoblotting, 37–39 Gold ion reduction, 15 Gold therapy, 15 Gold toning, 33, 63, 172 Grimelius silver staining methods, 2, 4–5 background staining, 10 microwave procedure, 9 protocol, 7–9 Gum Arabic, 39, 42, 61, 62, 72, 220

H HBC 6558, 219 HeLa cells, 109, 111, 112 Hematoxilin and eosin (H & E), 50 Hepatocyte, 17 Herpes simplex virus (HSV), 132 Hippocampus, 156 Histamine cells, 2, 5 Hodgkin lymphoma, 129 Horseradish peroxidase (HRP), See Peroxidase; Peroxidase-antiperoxidase HQ silver, 116 AMG protocols, 31, 33–34 technical recommendations, 42–43 HTB 24, 218 Human papillomavirus (HPV) detection, 36, 115, 119–126, 128, 132 Human umbilical vein endothelial cells (HUVEC), caveolin-1 localization, 180–181, 184–185 Hydrofluoric acid, 109–110 Hydrogen sulfide (H2S) gas method, 24 Hydroquinone, 15, 48 5-Hydroxytryptamine (5-HT), See Serotonin

Index I Immunoassays, immunogold-silver staining applications, 65 Immunoblotting, silver enhancement protocol, 37–39 Immunocytochemistry or immunohistochemistry (ICC or IHC), 47, 177, See also specific methods AMG protocol using HQ Silver, 33–34 antigen retrieval, 64 combination of postembedding and pre-embedding procedures, 153–154 combined ISH applications, 128 correlative fluorescence and electron microscopy, 181–182, 185–186 electron microscopy, See Immunoelectron microscopy general immunogold labeling principles, 145–147, See also Immunogold staining (IGS) and labeling immunogold-silver staining, 47, See Immunogoldsilver staining ISH popularity vs., 127–128 labeling procedure categories, 177 Nanogold applications, 100, See Nanogold Nanogold-labeled S-ABC IHC method and, 58 PAG techniques, 71–84, See Protein A-gold-silver (PAGS) staining pre-embedding immunoelectron microscopy with Nanogold immunolabeling, 169–176, See Immunoelectron microscopy reagent-producing companies, 50 two-color brightfield microscopy system, 141 PAG techniques, 71, See Protein A-gold-silver (PAGS) staining Immunoelectron microscopy (IEM), 47, 177, 189, See also Electron microscopy; Immunogoldsilver scanning electron microscopy; Scanning electron microscopy caveolin-1 localization after pre-embedding labeling, 180–181, 184–185 colloidal gold probe properties, 177 combined EM-LM level tubulin localization protocol, 178–180, 182–183 immunogold-silver staining for SEM in cancer research, 211–221 ISH applications, See Electron microscopy in situ hybridization PAG techniques, 71, See also Protein A-gold-silver (PAGS) staining pre-embedding immunohistochemistry with Nanogold immunolabeling, 169–176 gold toning, 172 other applications, 175 silver enhancement, 171–172, 174 tissue preparation, 170–171 ultrastructural immunogold detection, 189–197,

199–209, See also Electron microscopy in situ hybridization; Immunogold detection colloidal gold vs. Nanogold, 195 embedding protocol for postembedding detection, 190–191, 200–201 pre-embedding, 194, 199 results, 193–194 TSA/CARD, Nanogold detection, and goldbased AMG, 192–193 TSA/CARD protocol, 191–193 Immunoglobulin G (IgG) disulfide bond reduction, 92, 93 Fab′ fragment labeling with Nanogold, 91, See also Fab′ fragments immunogold staining, 147 monomaleimido-Nanogold labeling, 92–93 Nanogold labeling, 29 Immunogold detection, 189–197, See also Immunoelectron microscopy EM-ISH, See Electron microscopy in situ hybridization embedding protocol for postembedding detection, 190–191, 200–201 fixative, 194–195 gold-enhanced AMG, 196 pre-embedding, 194, 199 results, 193–194 TSA/CARD protocol, 191–193 Immunogold-silver scanning electron microscopy, 211–221, 223–232 arabinogalactan-protein epitope localization, 228–230 back-scattered electron imaging (BEI), 223–224 cancer research applications, 211–221 controls, 227 critical point drying, 215–217, 221, 227 fibronectin detection, 230–231 fixatives, 220–221 glycerol substitution, 231–232 gold-silver particle size and, 220 IGSS protocol, 214 immunoperoxidase and immunofluorescence resolution vs., 231 materials and reagents, 212–213, 224–225 microscopy, 227 potential applications, 231 reliability of glycerol substitution, 227–228, 230 results, 218–219 secondary electron image (SEI) mode, 219–220, 224 silver lactate preparation protocol, 215 specimen preparation protocol, 213–214 specimen processing protocol, 214–215 tissue preparation protocol, 226 Immunogold-silver staining (IGSS), 13, 16, 47–69, 152–153, 190, 196, 212 alternative AMG developer, 220

239

Gold and Silver Staining AMG, 61–63 antigen retrieval methods, 49–50, 53–54, 64 applications, 59–60 blotting applications, 65 comparative sensitivity, 48 early studies and procedures, 47–50 glutaraldehyde postfixation, 61 gold toning, 63 immunoassay applications, 65 immunogold reagents, 60–61 indirect method, 49, 53–55, 57 iodine treatment, 58–59, 74–76 labeled bridge IGSS method, 57–58 lectin-gold labeling, 64–65 limitations, 50–51 methods comparison, 57–58 multiple immunostaining, 65 Nanogold fluorescence, 53, See also FluoroNanogold Nanogold properties, 51–52 Nanogold vs. colloidal gold, 51–52 nerve tracing, 64 non-IHC applications, 52–53 PAP sensitivity vs., 65, 84 primary antibodies, 59–60 protective colloid, 220, See also Gum Arabic protein A-gold (PAG) methods, 58, 71–84, See Protein A-gold-silver (PAGS) staining SEM applications, See Immunogold-silver scanning electron microscopy silver deposits, 63 specificity controls, 60 streptavidin-biotin-complex (S-ABC) and, 49–50, 51, 58 streptavidin-Nanogold protocol, 55–57 tissue fixatives, 63–64 ultrastructural level applications, 152–153 unsuitability of frozen sections, 49, 63 washing buffers, detergents, and gelatin, 59 Immunogold staining (IGS) and labeling, 47, 71, 145–167, See also Immunogold-silver staining arabinogalactan-protein epitope localization, 228–230 combined cellular and molecular biology techniques, 153–159 enzyme histochemistry, 154 freeze-fracture replication and freeze-etching, 155 molecular biology, 155, 158–159 neuroanatomical tract tracing, 154–155 pre-embedding ICC, 153–154 combined EM-LM level tubulin localization protocol, 178–180, 182–183 controls, 153 electron microscopy, See Immunoelectron microscopy general immunocytochemistry principles, 145–147

240

gold cluster immunoprobe microscopic applications, 177–186 multiple staining, 148–149 double-face labeling, 148–149 paraformaldehyde vapor technique, 149 sequential double protein A-gold labeling, 149 simultaneous labeling, 148 osmium postfixation, 146–147 postembedding and pre-embedding comparison, 150–151 postembedding labeling methods, 145–147 combination of postembedding and preembedding procedures, 153–154 combined cellular and molecular biology techniques, 153–159 electron microscopical NADPH-d histochemistry, 161 labeling procedure on grid, 161 multiple staining, 148–149 single staining, 147–148 spatial resolution, 149 pre-embedding immunoelectron microscopy with Nanogold, 169–176, See Immunoelectron microscopy ultrastructural immunogold detection, 189–197, 199–209, See also Immunoelectron microscopy; Immunogold detection Immunohistochemistry, See Immunocytochemistry or immunohistochemistry Immunoperoxidase immunohistochemistry, double PAGS immunostaining protocol, 83–84 Indirect IGSS method, 49 procedure using colloidal gold, 53–55, 57, 60 In situ hybridization (ISH) automation, 142 autometallographic subtractive unique sequence ISH protocol, 129–130, 141–142 biotin-labeled cDNA probes, 134 cell conditioning, 128, 130, 142 chromogenic in situ hybridization (CISH), 129 CODFISH, 128, 130, 139–141 combined brightfield IHC applications, 128 combined immunogold labeling, 155 DNA nick translation, 120–121, 158, 199, 201, 206 DNA probe protocol, 119–124, See also DNA electron microscopy (EM-ISH), 199–209, See Electron microscopy in situ hybridization gold enhancement AMG, 35 GOLDFISH, 141–142 human papillomavirus detection, 119–126, 128 multiple detection systems, 130 in situ PCR thermal cycler, 119 new applications, 128 popularity as research technique, 127–128 probes for, 130 CODFISH, 139–141

Index DNA using GenPoint kit and streptavidinNanogold, 138–139 melting temperature, 142 RNA using labeled tyramides and streptavidinNanogold, 136–138 viral DNA using labeled tyramides and streptavidin-Nanogold, 132–136 viral DNA using streptavidin-Nanogold, 132–134 reagents and supplies, 129–132 silver enhancement procedure, 125 tissue preparation, 121–122 web site, 132 tyramide signal amplification (TSA or CARD), 18, 113–115, 128, See Tyramide signal amplification In situ PCR, 53 combined immunogold labeling, 155 reverse transcription PCR (RT-PCR), 155 thermal cycler and HPV detection, 119 In situ self-sustained sequence replication (3SR), 155 In situ transcription, IGSS applications, 52 Insulin, Nanogold labeling, 103 Insulin receptor, 103 Insulin cells, 4 Iodine treatment, 58–59, 74–76, 125, 143 Iodoacetic acid, radiolabeled, 94 Iododeoxyuridine (IdU), 158

J JIM4 epitope, 228–230 JY3 macrophage, 175

K Kidney, silver localization, 18 Ki-67, 50, 52, 53, 57 Kupffer cells, 18

L Labeled bridge IGSS method, 57–58 Laser Nanogold-type probes, 116 Lead citrate, 29, 91, 105 Lectin, 52, 64–65, 223 Leukocyte microtubules, 183 Leukocyte preparation protocol, 178–179 Light microscopy (LM) caveolin-1 localization protocol, 181 combined EM-LM level tubulin localization protocol, 178–180, 182–183 correlative electron microscopy, 181–182, 185–186 fluorescence applications, See Fluorescence microscopy gold enhancement of Nanogold, 35–37

gold visualization, 47–48, 71 imaging modalities for silver-enhanced gold visualization, 183 Nanogold labeling for, 87–88 PAG staining method, 71, See Protein A-gold-silver (PAGS) staining recommended silver enhancement procedure for Nanogold, 87 silver visualization, 48 Lighting conditions, 16, 38, 40–41, 61, 62 Liposomes, DPPE-Nanogold labeling, 96–97, 104, 105 Liver, silver localization, 18 LM2 epitope, 228–230 Lugol’s iodine treatment, 58–59, 74–76, 125, 143 Lung, 63 Lymphocytes EM-ISH, 206 microtubules, 183 Lysosomes, 21

M MAC207 epitope, 228–230 Maize, 224, 229 MAM-6, 48 Mammary carcinoma, 52, 53, See Breast carcinoma Masson staining, 2–3 background staining, 10 microwave procedure, 7 protocol, 6–7 Melanin, 3 Melting temperature, 142 β-Mercaptoethanol (BME), 39, 87 Mercury, 13, 14, 19–20, 23–24 Methylamine vanadate (NanoVan), 91, 105 Microscopic methods, See Electron microscopy; Light microscopy Microtubules, 100, 178–180, 182–183 cytoskeletal toxins, 230 Microwave irradiation, silver staining procedure, 1, 2, 3, 4 protocols, 7, 9 Midgut carcinoid, 4, 6 Molecular biology, combined postembedding immunogold labeling, 155, 158–159 Monocyte microtubules, 183 Monomaleimido-Nanogold labeling, 91–92, 107 Fab′ fragment, 91–92 IgG, 92–93 other proteins, 93–94 Mono-NHS (N-hydroxysuccinimide) Nanogold labeling, 94–96, 102 peptides, 94–95 proteins, 94–95

241

Gold and Silver Staining Mounting media for silver staining, 10 Multiple staining or labeling antigen staining using silver-enhanced Nanogold, 44 combined fluorescent-gold probes, 107–118, See also Fluorescence-gold multiple labeling; FluoroNanogold immunolabeling cellular components, 108–110 ISH and detection protocol, 110–113 optimum conditions, 115–117 EM and LM level tubulin localization, 178–180 silver-enhanced Nanogold, 44 immunogold-silver staining, 65 Nanogold-colloidal gold for EM, 86 PAGS combined PAGS-immunoperoxidase immunohistochemistry protocol, 83–84 single-PAG, two-step development protocol, 76–78 two PAG, one-step development protocol, 78–82 postembedding immunogold labeling, 148–149 double-face labeling, 148–149 paraformaldehyde vapor technique, 149 sequential double labeling, 149 simultaneous labeling, 148 2-color brightfield applications, 141 Myeloperoxidase (MPO) localization, 185

N NADPH-diaphorase (NADPH-d), 154, 157, 161 Nanogold®, 17–18, 85, 107, 128 colloidal gold vs., 42, 51–52, 85, 99–100, 178, 195 combined EM-LM immunocytochemistry applications, 178–180 combined fluorescent labels, 107–118, See also Fluorescence-gold multiple labeling; FluoroNanogold background signals, 116 immunolabeling cellular components, 108–110 ISH and detection protocol, 110–113 nonspecific affinity for nuclear materials, 116 optimum conditions, 115–117 poor development, 116 conjugates, 99–100, 107 background staining, 43 covalent bonding, 85, 99 EM visualization, 29, 86, 173–174 Fab′ labeling, See Fab′ fragments gold enhancement, 30, See also Gold enhancement autometallography for LM, 35–37 for EM, 34–35 IGSS with streptavidin-Nanogold protocol, 55–57

242

immunolabeling, general discussion, 101–102 penetrability, 173 pH optimum, 51 pre-embedding immunoelectron microscopy immunolabeling, 169–176, See under Immunoelectron microscopy silver enhancement/AMG, 29–45, See also Silver enhancement autometallography double labeling with immunocolloidal gold, 44 technical hints, 39–44 multiple antigen staining, 44 protocols immunoblotting, 37–38 gold toning, 33 HQ Silver, 31, 33–34 OsO4, 31–33 reagents, 42–43 size distribution, 42 storage, 86–87 streptavidin conjugate, See Streptavidin-Nanogold temperature effects, 116 thiols and, 38–39, 86–87, 102, 116 web site, 129 Nanogold, microscopic uses, 85–106 EM direct viewing, 97–99 labeling protocols colloidal gold double labeling for EM, 90 negative stains, 90–91 TEM post-embedding, 89–90 TEM pre-embedding, 89 liposomes, with DPPE-Nanogold, 96–97, 105 LM, 88 using monomaleimido-Nanogold, 91–92 Fab′ fragment, 91–92 IgG, 92–93 other proteins, 93–94 using mono-NHS Nanogold, 94–96, 102 peptides, 95–96 proteins, 94–95 materials and reagents, 87–88 negative staining, 105 recommendations for improved visibility, 91 STEM, 103, 105 Nanogold-silver staining (NGSS), 58–59, See Immunogold-silver staining antibody labeling, 60–61 NanoVan, 91, 105 Negative staining FluoroNanogold signal, 182 using with Nanogold, 90–91, 105 Neuroanatomical tract tracing, 64, 154–155 Neuroendocrine cells, silver stains for identifying, 1–11, See also Silver staining Neuroendocrine tumors, 4 Neurons, silver localization, 18–19 Neuropeptides, immunogold-silver staining, 49 Neutrophil preparation protocol, 181–182 Nexus Plus staining robot, 142

Index N-hydroxysuccinimide (NHS) Nanogold labeling, 94–96, 102 Nick translation of DNA, 120–121, 158, 199, 201, 206 Nitric oxide (NO), 154 Nitric oxide synthase (NOS), 154 Nitroblue tetrazolium (NBT), 154 Noradrenalin, 3, 4 N-propyl gallate (NPG), 42

O Oligonucleotides, 130, 143 Osmium-amine staining, 199 Osmium postfixation, 146–147 Osmium tetroxide (OsO4), 29, 30, 31–33, 43, 90, 195 Ovaries, silver localization, 18

P PAGS, See Protein A-gold-silver (PAGS) staining Pancreatic cells, 4–5, 21, 76, 77, 82 Pancreatic polypeptide (PP) cells, 4 PAP, See Peroxidase-antiperoxidase Parabenzoquinone, 63 Paraformaldehyde serotonin reaction, 3 vapor technique, 149 Pepsin, 207 Peptides fixative-induced cross-linking, 64 labeling using Mono-NHS Nanogold, 95–96 peptide nucleic acids, 206 postembedding immunogold labeling, 147 Permount, 10 Peroxidase neuroanatomical tract tracing, 155 reagent suppliers, 130 S-ABC peroxidase, immunogold-silver staining sensitivity vs., 51, 58 tyramide systems and, 143 Peroxidase-antiperoxidase (PAP), 171 IGSS sensitivity vs., 48, 65, 84 Peroxidase-diaminobenzidine-peroxide, AMG effects, 65 Pertex, 10 p53, 50, 57 pH, Nanogold vs. colloidal gold properties, 51 Photo fix, AMG reaction stopping, 44 Photographic development, 13, 62 Physical development, 2, 13, 47, 62 PIXE, See Proton-induced X-ray emission Placental transport of gold, 16 of silver, 19

Plant cells, immunogold-silver scanning electron microscopy, 223–232 Platinum clusters, 116 Polyclonal antilaminin, 173 Polyethylene glycol, 39 Polymerase chain reaction (PCR), in situ, See In situ PCR Polypeptide Y (PYY) cells, 3 Postembedding EM embedding protocol for immunogold detection, 190–191, 200–201 EM-ISH, 199–201 immunoelectron microscopy, PAG techniques, 71 immunogold labeling, 145–161, See Immunogold staining (IGS) and labeling Nanogold labeling for TEM, 89–90 Potassium ion (K+) channels, 34 Pre-embedding immunocytochemistry, 150 combined postembedding immunogold labeling procedures, 153–154 EM-ISH, 199, 205–206 HQ Silver enhancement of nanogold, 33–34 immunoelectron localization of caveolin-1, 180–181, 184–185 immunoelectron microscopy with immunogold labeling, 169–176, See Immunoelectron microscopy immunogold detection, 150, 194, 199 membrane-associated components and, 183–184 Nanogold labeling for TEM, 89 Progesterone receptors, 57 Proliferation markers, 50 Proteasome-amyloid b Nanogold complex, 40 Protein A-gold immunolabeling, 147 sequential double labeling for postembedding IGS, 149 spatial resolution, 149 Protein A-gold-silver (PAGS) staining, 58, 71–84 AMG protocol, 72–73 detection sensitivity, 71 developers, 72, 77, 82 development time, 74 double staining method, 72 developers, 77, 82 immunoperoxidase immunohistochemistry protocol, 83–84 single-PAG, two-step development protocol, 76–78 two-PAG, one-step development protocol, 78–82 iodine treatment, 74–76 silver reduction for reducing background staining, 76 specificity controls, 76 staining protocol, 73–76 tissue preparation protocol, 72 Proteinase K, 207

243

Gold and Silver Staining Protein digestion, 207–208 Protein labeling, using Mono-NHS Nanogold, 94–95 Proteins, fixative-induced cross-linking, 64 Proton-induced X-ray emission (PIXE), 13 Protoplasts, 223

R Red blood cells, 100 Reducing solutions, 8, 9 Riboprobes, 130, 134, 143 RNA combined immunogold labeling and molecular biology, 158 EM-ISH, 199–209, See also Electron microscopy in situ hybridization detection procedure, 203–204 probe labeling, 201–202, 206 results, 205 technical guidelines, 208 FITC-labeled riboprobes, 143 fixation cautions, 136 in situ hybridization protocol using labeled tyramides and streptavidin-Nanogold, 136–138 in situ PCR, 53 stop reagent cautions, 43 RNase labeling, 14, 16, 17

S S-ABC peroxidase, immunogold-silver staining sensitivity vs., 51, 58 Saponin, 43, 184 Scanning electron microscopy (SEM), 211, See also Electron microscopy Au-Pd sputter coating protocol, 217–218 backscattered electron image (BEI) mode, 219, 223–224 cancer research applications, 211–221 immunogold SEM, 211–221, 223–232, See also Immunogold-silver scanning electron microscopy Nanogold visualization, 29, 103–105 secondary electron image (SEI) mode, 219–220, 224 use of colloidal gold, 219–220 Scanning-transmission electron microscopy (STEM), Nanogold visualization, 29, 103–105 Secondary electron image (SEI) mode, 219–220, 224 Secretin, 3 Selenides, AMG localization, 13–15, 19–23 Self-sustained sequence replication (3SR), 155 Seminiferous tubules, 23, 173 Sensory neurons, 16, 157 Sequential double protein A-gold labeling, 149

244

Serotonin (5-hydroxytryptamine; 5-HT), 2, 3, 4–5 Sevier-Munger silver staining method, 2, 5–6 background staining, 10 protocol, 9–10 Silver, See also Immunogold-silver staining AMG, 15, See Silver enhancement autometallography biological marker, 23 deposits in IGSS sections, 63 LM visualization, 48 localization by AMG, 18–19 medical uses, 18 reduction for PAGS, 76 solutions, 6, 7, 9 sulfide/selenide, 13, 14, 18, 19, 21 Silver acetate, 38, 40, 62, 116, 129, 174 double PAGS and, 82 solution preparation procedure, 114 Silver bromide, 62 Silver enhancement autometallography (AMG), 13, 114, 152–153, 177, 190, 212 AMG chemistry, 15 bismuth detection, 14, 20–21 catalytic center, 13, 15, 16 caveolin-1 localization protocol, 181 chemical background, 15 CODFISH applications, 128, 130, 139–141 commercial kits, 17 correlative EM and fluorescence microscopy, 182, 186 Danscher’s formulation, 16, 62, 72, 116, 212 developers, 14, 16–17, 39–40, 61, 62, 72, 116, 129, 212, See also Silver acetate; Silver lactate; Silver nitrate cautions, 38–39 double PAGS methods, 77, 82 IGSS applications, 220 light sensitivity, 16, 38, 40–41, 61, 62 LM applications, 87 development process, 61 development time, 41–42, 220 double PAGS staining, 72 lectin-gold labeling, 64–65 gold enhancement vs., 41, See also Gold enhancement autometallography gold localization, 15–16 history and early procedures, 13–14, 61 IGSS, 47–48, 61–63, See Immunogold-silver staining procedures, 62 with streptavidin-Nanogold protocol, 56 ultrastructural applications, 152–153 immunogold-silver SEM, 224–232, See also Immunogold-silver scanning electron microscopy iodine effects, 74 ISH procedures, 125, 127–144, See In situ hybridization

Index lectin-gold labeling, 64–65 lighting conditions, 16, 61, 38, 40–41, 62, 125 lung IHC problems, 63 mercury localization, 14, 19–20 practical applications, 23–24 metal selenides/sulfides, 13–15, 19–20 Nanogold applications, 29–45, See also Nanogold background staining, 43 double labeling with immunocolloidal gold, 44 gels and blots, 30 protocols immunoblotting, 37–38 gold toning, 33 HQ Silver, 31 HQ Silver and pre-embedding immunocytochemistry, 33–34 OsO4, 31–33 reagents, 42–43 size distribution, 42 technical hints, 39–44 time course, 31 Nanogold visualization, 173–174 novel practical applications, 23–24 pre-embedding immunoelectron microscopy, 171–172, 174 procedure overview, 29–30 protective colloid, 39, 61, 62, 220, See also Gum Arabic protocol for PAGS, 72–73 protocol with silver acetate, 114–115 recommended procedures for microscopy with Nanogold, 87 protein A-gold (PAG) methods, 58, 71, 72–73, See also Protein A-gold-silver (PAGS) staining selenium method, 15 SEM applications, 219, See also Immunogold-silver scanning electron microscopy; Scanning electron microscopy gold-silver particle size and, 220 signal properties, 63 silver lactate method for counterstaining, 23 silver localization, 18–19 stopping reagent, 15, 43–44, 56–57, 136 subtractive unique sequence ISH (GOLDFISH), 129–130, 141–142 tissue fixatives, 63–64 tubulin localization protocol, 179–180 ultrastructural applications, 152 unspecific argyrophilic-argentaffin reactions, 62, 72, 174 UV exposure and, 18 vascular infusion of colloidal gold, 23 visualization, imaging modalities, 183 water purity issues, 63 zinc localization, 21–23 Silver lactate, 16–17, 23, 38, 39, 62, 72, 174, 212 development time, 220 solution preparation protocol, 215

Silver nitrate, 62, 72 development time, 220 medical use, 18 PAGS, 77 Silver staining, 1–11, See also Immunogold silver staining; Silver enhancement autometallography background staining, 10 neuroendocrine cell type identification, 1–11 fixatives, 2 Grimelius stain, 2, 4–5 protocol, 7–9 Masson staining, 2–3 protocol, 6–7 microwave irradiation, 2, 3, 4, 7, 9, 10 Sevier-Munger stain, 2, 5–6 protocol, 9–10 technical hints, 10 Simultaneous immunogold labeling, 148 Skin discoloration, classical argyrosis, 18 Sodium metaperiodate, 147 Sodium thiosulfate, 43–44, 59, 125, 136, 143 Somatostatin cells, 4 Spermatids, 194, 195, 196 Spinal cord bismuth localization, 21 mercury localization, 20 silver localization, 19 Spinal ganglia, 16 Sputter coating protocol, 217–218 Squamous cell carcinoma, 128 Steroid receptors, 50 Stopping reagent, 15, 43–44, 56–57, 136, 143 Streptavidin-biotin-complex (S-ABC), 49–50 Streptavidin-biotin-IGSS methods, 58 Streptavidin-Nanogold, 29, 51 antibody labeling, 60–61 double-labeled, protocol for ISH and detection, 110–115 fluorescence double labeling, 61, 110–115 immunogold-silver staining process, 61 RNA ISH protocol, 136–138 viral DNA ISH protocols, 132–136 Streptavidin-peroxidase, reagent suppliers, 130 Striated muscle, 49 Substance P, 102 Subtractive unique sequence ISH, 129–130, 141–142 Sulfhydryl determination, 94 Sulfides, AMG localization, 13–15, 19–21

T T and Tn antigens, 212, 218 Temperature effects on Nanogold, 116 Terminal deoxynucleotidyl transferase, 199 Testis, 23, 35, 173, 174 Tetrairidium, 105, 116

245

Gold and Silver Staining Texas Red®, 159 T4 DNA ligase, 159 Thiols, Nanogold and, 38–39, 86–87, 102, 116 Thiosulfate, 15, 43–44, 59, 125, 136, 143 Thymidines, monoclonal antibodies, 158 Timm’s method, 13–14, 21 Tissue fixatives, immunogold-silver staining applications, 63–64 Tissue preparation protocols immunogold SEM protocol for protoplasts, 226 for ISH, 121–122 for PAGS, 72 Transmission electron microscopy (TEM), See also Electron microscopy HeLa cells, 112 immunogold labeling techniques, 145–167, See Immunogold staining (IGS) and labeling immunogold-silver staining applications, 152–153 Nanogold labeling for post-embedding, 89–90 Nanogold labeling for pre-embedding, 89 Nanogold visualization, 29, 86, 105 RBCs, 103 scanning-transmission electron microscopy (STEM), Nanogold visualization, 29, 103–105 Trifluoralin, 230 Tris, 148 Tris-buffered saline (TBS), 59 Tubulin localization, 178–180, 182–183 Tumor suppressor gene proteins, 50 Tumors, neuroendocrine, silver staining, 4 Tween, 43, 116 Tyramide, fluorescein labeling, 130 Tyramide signal amplification (TSA, CARD), 18, 128, 134, 189 biotin complex stability, 130 immunogold detection protocol, 191–193 immunogold detection results, 193–194 incubation time, 196 kits, 134, 138–139 Nanogold-streptavidin protocols DNA ISH using GenPoint kit, 138–139 double-labeled ISH, 110–115 RNA ISH, 136–138 viral DNA ISH, 134–136 peroxidases and, 143 reagents and kits, 129–132 technical recommendations, 143 web site, 134

246

U Ultrasmall colloidal gold particles, 100, 177–178, 190, 194, 195 Ultrathin cryosectioned neutrophils, 181–182 myeloperoxidase localization, 185 Ultrathin sections for EM-ISH, 207 Ultraviolet (UV) irradiation AMG staining and, 18 gold ion reduction, 15 Umbilical vein endothelial cells (HUVEC), caveolin-1 localization, 180–181, 184–185 Undecagold, 105 Unspecific argyrophilic-argentaffin reactions, 62, 72, 174 Uranyl acetate, 91, 105 Urinary bladder ganglion, 51

V Vanadium-based stain, 91, 105 Vascular infusion of colloidal gold, 23 Vessel detection (neocortical), 24

W Washing buffer systems, 59 Water purity, 63 Water substitution, by glycerol, 224–232 Web sites gold macromolecule manufacturers, 54 ISH, 132 Nanoprobes, 129 TSA, 134 Wheat germ agglutinin (WGA), 64

Z Zinc localization by AMG, 13, 14, 15, 19, 21–23 ion pumps, 21 practical applications, 24 Zinc-enriched neurons (ZEN), 15, 21–23 terminals, 21 vesicles, 22