1,702 421 3MB
Pages 343 Page size 432 x 686.88 pts Year 2003
Ocular Neuroprotection
edited by
Leonard A. Levin University of Wisconsin-Madison Madison, Wisconsin, U.S.A.
Adriana Di Polo University of Montreal Montreal, Quebec, Canada
MARCEL
MARCEL DEKKER, INC. DEKKER
-
NEWYORK BASEL
Although great care has been taken to provide accurate and current information, neither the author(s) nor the publisher, nor anyone else associated with this publication, shall be liable for any loss, damage, or liability directly or indirectly caused or alleged to be caused by this book. The material contained herein is not intended to provide specific advice or recommendations for any specific situation. Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 0-8247-4070-X This book is printed on acid-free paper. Headquarters Marcel Dekker, Inc. 270 Madison Avenue, New York, NY 10016 tel: 212-696-9000; fax: 212-685-4540 Eastern Hemisphere Distribution Marcel Dekker AG Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland tel: 41-61-260-6300; fax: 41-61-260-6333 World Wide Web http:/ /www.dekker.com The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/Professional Marketing at the headquarters address above. Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved. 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 and retrieval system, without permission in writing from the publisher. Current printing (last digit): 10 9 8 7 6 5 4 3 2 1 PRINTED IN THE UNITED STATES OF AMERICA
Preface
Neuroprotection comprises a broad range of therapeutic strategies aimed at preventing death of neurons in trauma or disease. As applied to ocular disorders, neuroprotection has received enormous attention in past years because of its potential for treating diseases that were heretofore untreatable or difficult to treat, including disorders primarily involving death of photoreceptors (such as in macular degeneration, retinitis pigmentosa, and retinal detachment) or retinal ganglion cells (such as in glaucoma). The goal of this book is to provide the reader with up-to-date ocular neuroprotection research methods that are used in leading laboratories worldwide. Chapters 1 through 15 focus on research methods at the laboratory level, with emphasis on various cell culture and animal models, classes of neuroprotective drugs, biochemical and cellular outcome measures, and applicability to human disease. Chapters 16 through 18 focus on how clinical trials of neuroprotection may be carried out, concentrating on issues of recruitment, outcome measures, and requirements of regulatory agencies. The information conveyed in this book is intended to be practical. The chapters address the nuts and bolts of the actual methods themselves, along with potential problems, solutions to those problems, potential shortcuts (and contraindications to shortcuts), and specific suppliers, where appropriate. The book is not intended to be an exhaustive presentation of research data or theories, but rather to provide a focused description of useful methods for conducting neuroprotection research. This book will acquaint those considering research in the area with the major techniques, tools, and issues in ocular neuroprotection, while investigators with experience in neuroprotection research may benefit from exposure to new models, methodologies, and clinical approaches. Regardless of the reader’s experience, this volume will be an invaluable resource. iii
iv
Preface
We sincerely thank all the contributors—leading scientists in their fields— who have generously shared their knowledge and expertise in each chapter. Leonard A. Levin Adriana Di Polo
Contents
Preface Contributors
iii ix
I. In Vitro Models 1. Culture of Retinal Neurons James D. Lindsey
1
II. In Vivo Models of Optic Nerve Damage 2. Crush Injury of the Optic Nerve Michal Schwartz and Eti Yoles
13
3. Intraocular Pressure Elevation: Vein Cauterization Sansar C. Sharma
23
4. Intraocular Pressure Elevation: Injecting Hypertonic Saline into Episcleral Veins John C. Morrison, Elaine C. Johnson, Lijun Jia, and William O. Cepurna 5. Intraocular Pressure Elevation: Laser Photocoagulation of the Trabecular Meshwork B’Ann T. Gabelt, James N. Ver Hoeve, and Paul L. Kaufman
31
47 v
vi
Contents
III. In Vivo Models of Retinal Damage 6.
Light-Induced Retinal Degeneration Daniel T. Organisciak, R. M. Darrow, and L. S. Barsalou
85
7.
Retinal Detachment Ward M. Peterson
109
8.
Retinal Ischemia Manuel Vidal-Sanz, Marı´a P. Lafuente, Inmaculada Selle´sNavarro, Marı´a E. Rodrı´guez, Sergio Mayor-Torroglosa, Marı´a P. Villegas-Pe´rez
129
IV. Neuroprotective Strategies 9.
10.
Drug Delivery Robert W. Nickells and Cassandra L. Schlamp
153
Recombinant Viral Vectors Przemyslaw Sapieha and Adriana Di Polo
167
V. Analysis of Neuronal Survival and Function 11.
Quantification of Retinal Cells Christine A. Curcio and Kenneth R. Sloan
189
12.
Ex Vivo and Whole-Mount Retinal Preparations Arthur J. Weber
205
13.
Detection of Single-Cell Apoptosis William G. Tatton, Ruth M. E. Chalmers-Redman, and Nadine A. Tatton
225
14.
Imaging of Retinal Ganglion Cells Joshua P. Vrabec and Leonard A. Levin
241
15.
Evaluation of Retinal Function: Electroretinography Marc He´bert and Pierre Lachapelle
249
Contents
vii
VI. Clinical Trials 16. Evaluation of Visual Outcome Pamela A. Sample
273
17. Clinical Trials in Neuroprotection Scott M. Whitcup
291
18. Regulatory Issues in Clinical Trials Anthony C. Arnold
303
Index
313
Contributors
Anthony C. Arnold, M.D. Neuro-Ophthalmology Division, Department of Ophthalmology, Jules Stein Eye Institute, David Geffen School of Medicine, UCLA, Los Angeles, California, U.S.A. L. S. Barsalou Wright State University, Dayton, Ohio, U.S.A. William O. Cepurna, B.S. Department of Ophthalmology, Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, U.S.A. Ruth M. E. Chalmers-Redman New York, U.S.A.
Mount Sinai School of Medicine, New York,
Christine A. Curcio, Ph.D. Department of Ophthalmology, School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, U.S.A. R. M. Darrow
Wright State University, Dayton, Ohio, U.S.A.
Adriana Di Polo, Ph.D. Department of Pathology and Cell Biology, University of Montreal, Montreal, Quebec, Canada B’Ann T. Gabelt, M.S. Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. Marc He´bert, Ph.D. University of Laval, Quebec City, Quebec, Canada Lijun Jia, M.D., Ph.D. Department of Ophthalmology, Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, U.S.A. Elaine C. Johnson, Sc.D. Department of Ophthalmology, Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, U.S.A. Paul L. Kaufman, M.D. Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. ix
x
Contributors
Pierre Lachapelle, Ph.D. Department of Ophthalmology, McGill University, Montreal, Quebec, Canada Marı´a P. Lafuente, M.D., Ph.D. de Murcia, Murcia, Spain
Department of Ophthalmology, Universidad
Leonard A. Levin, M.D., Ph.D. Departments of Ophthalmology and Visual Sciences, Neurology, and Neurological Surgery, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. James D. Lindsey, Ph.D. Hamilton Glaucoma Center and Department of Ophthalmology, School of Medicine, University of California, San Diego, La Jolla, California, U.S.A. Sergio Mayor-Torroglosa, O.D. de Murcia, Murcia, Spain
Department of Ophthalmology, Universidad
John C. Morrison, M.D. Casey Eye Institute, Oregon Health & Science University, Portland, Oregon, U.S.A. Robert W. Nickells, Ph.D. Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. Daniel T. Organisciak, Ph.D. Department of Biochemistry and Molecular Biology, Wright State University, Dayton, Ohio, U.S.A. Ward M. Peterson, Ph.D. Preclinical Programs, Inspire Pharmaceuticals, Durham, North Carolina, U.S.A. Marı´a E. Rodrı´guez, M.D., Ph.D. dad de Murcia, Murcia, Spain
Department of Ophthalmology, Universi-
Pamela A. Sample, Ph.D. Department of Ophthalmology, University of California, San Diego, La Jolla, California, U.S.A. Przemyslaw Sapieha Department of Pathology and Cellular Biology, University of Monteal, Monteal, Quebec, Canada Cassandra L. Schlamp, Ph.D. Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. Michal Schwartz The Weizmann Institute of Science, Rehovot, Israel Inmaculada Selle´s-Navarro, M.D., Ph.D. Universidad de Murcia, Murcia, Spain
Department of Ophthalmology,
Sansar C. Sharma, Ph.D. Department of Ophthalmology, New York Medical College, Valhalla, New York, U.S.A.
Contributors
xi
Kenneth R. Sloan, Ph.D. Department of Computer and Information Sciences, University of Alabama at Birmingham, Birmingham, Alabama, U.S.A. Nadine A. Tatton Mount Sinai School of Medicine, New York, New York, U.S.A. William G. Tatton U.S.A.
Mount Sinai School of Medicine, New York, New York,
James N. Ver Hoeve, Ph.D. Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. Manuel Vidal-Sanz, M.D., Ph.D. de Murcia, Murcia, Spain
Department of Ophthalmology, Universidad
Marı´a P. Villegas-Pe´rez, M.D., Ph.D. versidad de Murcia, Murcia, Spain
Department of Ophthalmology, Uni-
Joshua P. Vrabec Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, Madison, Wisconsin, U.S.A. Arthur J. Weber, Ph.D. Department of Physiology, Michigan State University, East Lansing, Michigan, U.S.A. Scott M. Whitcup, M.D. Allergan, Inc., Irvine, and Jules Stein Eye Institute, David Geffen School of Medicine, UCLA, Los Angeles, California, U.S.A. Eti Yoles
The Weizmann Institute of Science, Rehovot, Israel
1 Culture of Retinal Neurons James D. Lindsey Hamilton Glaucoma Center School of Medicine University of California, San Diego La Jolla, California, U.S.A.
I.
INTRODUCTION
The culture of retinal tissue and cells has led to new insights regarding the influence of neurotransmitters, growth factors, hormones, and other biological molecules on retinal neuron survival and differentiation, as well as the role of the extracellular matrix in retinal cell function; it has also provided important screening tools to investigate potential neuroprotective treatments [1–7]. A particular advantage of many culture systems is they can permit visual assessment of cell shape, viability, and synaptogenesis of living retinal neurons [7–11]. Through the use of specialized dyes, it is possible to monitor changes of intracellular calcium, apoptosis markers, and other cytoplasmic molecules within living retinal cells (see Chap. 14) [12,13]. Also, the effects of drug treatments can be assessed in the absence of systemic responses or dilution that would be encountered in vivo. Despite these attractions, however, it can be challenging to established retinal neuron cultures based on the abbreviated methods often presented in published articles. Hence, this chapter discusses the design of successful retinal neuron culture systems, provides details for certain common procedures that often are overly simplified or omitted in the methods of published articles, and addresses the identification and correction of problems that are occasionally encountered during the culture of retinal neurons. In addition, several resources are iden1
2
Lindsey
tified that provide further information useful for the design and evaluation of experiments using retinal neuron cultures.
II. DESIGN CONSIDERATIONS A.
Culture Type
The type of culture chosen is often constrained by the cellular response that is to be studied. For example, the effects of experimental treatments on retinal neuron survival and differentiation are easily studied in monolayer cell cultures. In these cultures, a completely dissociated retinal cell suspension is placed in tissue culture dishes, and within 6 to 8 hours, the neurons attach to the dish surface and begin to differentiate neuritic processes. Because each neuron can be directly inspected using an inverted microscope, it is relatively easy to determine whether the neuron is alive and whether process growth has been altered. These features also can be studied in explant retinal culture systems in which some of the original retinal tissue organization is retained. However, because it is often difficult to inspect individual living neurons in explants, assessment of survival and differentiation can be better accomplished using appropriate biochemical or histological analyses. B.
Developmental Age
The developmental age of the retina also can be important. For example, growing cells from embryonic retina cells are often easier to maintain in culture than cells from adult retina. It is also possible to expand the number of embryonic retinal cells by encouraging cell division. In contrast, cells from older embryo retina or adult retina typically are post-mitotic. Hence, it can be challenging to obtain large numbers of retinal cells for culture experiments from adult retina. However, such experiments can be important because the cellular responses of embryonic retinal cells can vary from the responses of adult retinal cells. C.
Culture Media
The various culture media contain different types and amount of salts, amino acids, energy sources, buffers, and vitamins. For example, Dulbecco’s modified Eagles medium (DMEM), which can be used for culturing embryonic and newborn rat or mouse retinal neurons, contains 7 salts, 13 amino acids, 8 vitamins, glucose, and pyruvate. In contrast, Medium 199, which has been used for culturing chick retinal neurons, contains 11 salts, 21 amino acids, 17 vitamins, and glucose, plus 17 additional ingredients, but no pyruvate. For the successful culture of retinal neurons expressing glutamate or aspartate receptors, it may be beneficial
Culture of Retinal Neurons
3
to eliminate these amino acids from the medium because they could induce excitotoxic stress. Neurobasal medium (available from Gibco BRL) is one such medium that has been formulated without glutamate or aspartate [14]. It is important to note that the pH of tissue culture medium often is regulated by exposing the medium to an atmosphere containing a specific amount of carbon dioxide. For example, DMEM is designed to buffer correctly in an atmosphere containing 10% carbon dioxide. However, media such as Ham’s F12 and RPMI 1640 require 5% carbon dioxide. The correct concentration of carbon dioxide for a particular medium can be obtained from the medium supplier or from previously published protocols. As each medium has its advantages and limitations, the best medium for each retinal neuron type should be determined empirically. D. Media Supplements The performance of media is often enhanced with supplements including hormones, growth factors, cofactors, lipids, carrier proteins, and natural fluids. Hormones that often are beneficial include insulin, progesterone, and hydrocortisone. Growth factors that have been helpful for the maintenance of certain retinal neurons include fibroblast growth factor–2, nerve growth factor, brain-derived neurotrophic factor, ciliary neurotrophic factor, neurotrophin-3, and neurotrophin-4 [3,15,16]. Useful cofactors, molecules that facilitate certain biological reactions, include compounds such as sodium selenite and copper sulfate. Lipids that have been used in certain retinal neuron cultures include linoleic acid and docosahexaenoic acid [10,17]. Carrier proteins that have proved to be beneficial include transferrin and serum albumin. Note that the addition of unbuffered solutions of certain supplements can change the media pH. Natural fluids such as sera, amniotic fluid, and cerebrospinal fluid have been used as supplements. These contain hormones, growth factors, and other potentially beneficial molecules. It should be noted that the composition of natural fluids can vary among suppliers and even among lots from the same supplier. With natural fluids, as with purified supplements, the optimal amount to add can vary among retinal neuron types and culture systems. Hence, the best mixture of supplements and/or natural fluids needs to be determined for each case. E.
Substrate Enhancements
The performance of retinal cultures often can be improved by application of specialized coatings to the culture dishes. The first type of coating is the charge modifiers and includes poly-l-lysine and poly-l-ornithine. Many retinal neurons express glycoproteins that contain negatively charged sialic acid. As a result, they do not bind strongly to negatively charged tissue culture plastic. Poly-llysine and poly-l-ornithine are positively charged synthetic amino acid polymers.
4
Lindsey
Coating the culture dish with either of these compounds changes the charge of the tissue culture surface and thereby promotes neuronal attachment [18]. The second type of coating includes extracellular matrix molecules such as collagens, fibronectin, laminin, and heparin sulfate proteoglycan. These molecules can bind to specific cell surface receptors, facilitate growth factor action, and can often dramatically improve neuronal differentiation and survival [19,20]. The third type of coating includes molecules with specific affinities such as antibodies and lectins. These molecules can facilitate or modulate binding of certain subsets of cells present within a dissociated retinal cell suspension [18]. As with media and supplements, the optimal coating should be determined for each retinal cell type and culture system. F.
Co-Cultures
The survival of certain retinal cell types has been improved by co-culturing with other cell types such as astrocytes, or with target neurons such as lateral geniculate neurons in the case of retinal ganglion cells [6]. In the case of astrocytes, a confluent monolayer of astrocytes is often generated prior to seeding the retinal neurons (adding the dissociated cell suspension to the cultures). Because retinal neurons can grow on top of the astrocytes in these cultures, their survival and differentiation may be easily evaluated with the inverted microscope. Retinal neuron aggregate cultures also can be generated in the presence of other cell types. This will be further discussed below. Co-cultures have been useful for investigating the cellular basis of beneficial cell interactions. G.
Cell Density
The density of cells placed within monolayer cell cultures can influence cell survival and differentiation [21]. Usually, increased survival is observed with increased cell density. This is because the retinal neurons produce molecules that are beneficial for their survival in vitro. If cell density is too high, however, differentiation can be inhibited. Also, cell clumping can become problematic in high-density cultures. This can make it difficult to assess survival changes and changes in cell structure.
III. SELECTED METHODS A.
Isolation of Chick Retina
This protocol is appropriate for chick embryos ranging in age from 7 to 12 days. All instruments must be sterilized before use. After being briefly wiped with 70% ethanol, the rounded end of the egg is opened using blunt forceps, and the shell
Culture of Retinal Neurons
5
membrane over the embryo is removed with fine forceps. Using fine-curved forceps, the vascular connections to yolk membranes are broken. Next, the embryo is lifted out using blunt-curved forceps and placed in a 60 mm dish. After separation of the head, the eyelid rudiments are either pulled off with fine forceps or removed using spring-action scissors. Then the eyes are lifted from their sockets using the curved portion of curved sharp forceps and transferred to a 60-mm dish containing Hank’s buffered saline solution (HBSS). Any orbital tissues such as muscle and fat tissue is then removed from the surface of the eyes. Next, the tips of sharp forceps are inserted into the globe near the optic nerve to create a small opening. With two pairs of sharp forceps, the edge of the sclera at the opening is grasped at opposite sides of the opening and, by spreading, the sclera and retinal pigment epithelium (RPE) are peeled from the globe—like removing a thin glove from a hand. The lens and cornea typically come off with the sclera/RPE. The center of the vitreous gel is grasped with blunt forceps, and the retina is gently teased from the vitreous gel using sharp forceps. The retina is transferred to another 60 mm dish containing HBSS by gently lifting with either fire-polished glass hooks or spread curved forceps. Nonretinal tissue present on small portions of the retina are sliced off using sharp sterile tungsten needles that have been glued with epoxy to glass rod handles. The cleaned retina is transferred to another 60 mm dish containing HBSS. At this point, the retina is ready to be sliced for sliced cultures, diced with tungsten needles and then placed in explant cultures, or diced and then enzymatically dissociated into a single cell suspension. This protocol can be adapted to newborn and adult rat eyes by using appropriate sterile surgical technique to remove eyes and then proceeding as described to isolate the retina. B. Generating Retinal Neuron Monolayer Cultures This method is adapted from a protocol for retinal neuron cultures that promotes the survival and differentiation of several types of retinal neurons [10]. Precoat tissue culture dishes with a charge modifier (e.g., by filling 35 mm dishes with 0.1 mg/mL 30–70 kDa poly-l-ornithine dissolved in purified water and sterile filtered; incubate at least 1 h at room temperature). Add an extracellular matrix coating (e.g., 0.5 µg/mL laminin dissolved in DMEM; incubate in 10% CO2 incubator at 37oC, overnight). Prepare media fresh, add supplements (e.g., Medium 199 containing 10% heat-inactivated fetal bovine serum [FBS], 110 µg/ mL linoleic acid–albumin, 100 U/mL penicillin, and 2 mM glutamine) and place in humidified incubator (5% CO2, 37°C) to equilibrate temperature and pH. Isolate chick retina and dice into small pieces (1 mm squares) as described above. The retinal pieces are transferred to a test tube and allowed to settle to the bottom. The HBSS is replaced with calcium- and magnesium-free HBSS (CMF-HBSS) and incubated for 10 minutes in a 37°C water bath. After two
6
Lindsey
rinses (changes) in CMF-HBSS, a solution containing 0.25% trypsin and 1.85 U/ml DNAse (dissolved in CMF-HBSS) is added and the tissue is incubated at 37°C for 20 min. After trypsin is removed, the retinal pieces are rinsed with HBSS containing 10% fetal bovine serum (FBS). After 1 mL of 10% FBS/HBSS is added, the tissue is triturated (sucked in and out of a Pasteur pipette in which the tip has been reduced by flame polishing) several times until it dissociates into a single cell suspension. Avoid excessive trituration that can reduce viability. Examine a drop of the suspension placed on a microscope slide using a microscope with phase optics. If there are cell clumps present, these can be removed by passing the cell suspension through a sterile Nytex filter fabric affixed to a sterile glass tube. The concentration of cells in the suspension is determined by adding one drop of suspension to each side of a Neubauer ruling hemacytometer. After placing the hemacytometer on a conventional (upright) phase microscope, cells visible in a 1 mm square area are counted (a group of 16 squares present at the corners of the ruling). A button-activated counter facilitates tallying the count. Multiply the count obtained by 10,000 to obtain cells per milliliter. Total yield is calculated by multiplying the suspension concentration by the total suspension volume. After appropriate dilution with final media, the cell suspension is added to the culture dishes. In the case of chick retinal neurons seeded in 35 mm dishes described above, dilute the suspension to 400,000 cells/mL and add 1 mL per dish (add to the 1 mL already present in the dish). The dishes are then placed into the incubator.
IV. ALTERNATIVE CULTURE FORMATS A.
Retinal Explant Cultures
Retinal explant cultures can be useful for examining neurite outgrowth as well as neuronal differentiation [19,22,23]. Typically, isolated retina is diced into pieces and the pieces are placed in a drop of medium containing 40 to 50% serum on the bottom of a culture dish. After allowing the serum to clot, which holds the explant in place, add culture medium. Within 1 to 2 days, glia are observed growing out of the explant. Several days later, neurites are observed radiating from the explants. These cultures have been used to evaluate interactions between growing axons and extracellular matrix, the effects of growth factors, and the role of glia in axon extension. B.
Retinal Slice Cultures
Cellular contacts and interactions are typically well preserved in slice cultures [24,25]. The usual approach is to lay the isolated retina on a cellulose filter mem-
Culture of Retinal Neurons
7
brane and transfer it to the sterilized stage of a tissue chopper, where slices 100– 400 µm thick are cut. The slices, along with the attached membrane, are then transferred into culture medium in a culture dish and placed into an incubator. These cultures preserve the association of retinal cells with their microenvironment and neighboring cell types. Within the slice, synaptic associations also are retained. Cells within retinal slices can be studied using electrophysiological recordings. Subsets of cells, such as amacrine cells, can be identified and studied within these slice cultures using specific immunohistochemical markers. C. Reaggregation Cultures The presence of adhesion- or contact-mediated interactions can be studied in reaggregation cultures [26–28]. A retinal cell suspension is prepared as described above. Aliquots of the suspension are transferred to reaggregation culture dishes or small (e.g., 25 mL), sterile Erlenmeyer flasks and then placed in a gyratory shaker contained within a 37°C incubator. The size of the aggregate is dependent on cell density, gyration amplitude (10–25 mm), and gyration frequency (50– 80 rpm). By mixing the retinal cell suspension with other cell type suspensions (such as from lateral geniculate), target cellular interactions can be studied. Reaggregate cultures can be evaluated by biochemical analysis, by counting, or by histology. D. Enriched Photoreceptor Cultures Methods have been developed to generate mixed retinal neuron cultures from embryonic chick or mouse eyes that contain identifiable photoreceptors as well as multipolar neurons (neurons that express multiple neurites) [10,29]. The photoreceptors are readily identified by their extension of a single neurite, their expression of opsin, and, in the case of chick photoreceptors, a cytoplasmic lipid droplet visible by phase microscopy. Adding 2 mM kainic acid or 1–2 nM β-bungarotoxin to the cultures can eliminate up to 70% of the multipolar neurons without apparent effect on the photoreceptors [30,31]. Combined treatments with these agents does not show additive effects or potentiation between the toxins [31]. Another method useful for generating cultures of purified photoreceptors separates these cells from other retinal neurons during isolation through the use of vibratome sectioning [32]. E.
Purified Retinal Ganglion Cell Cultures
Retinal ganglion cells can be purified from retinal cell suspensions using cell panning [33] and then cultured in chemically defined media [2]. This technique relies on retinal ganglion cell expression of the cell surface marker protein Thy-
8
Lindsey
1 and the absence of expression of an antigen recognized by anti-macrophage antibody. A retinal cell suspension is generated as described above. This suspension is incubated with monoclonal mouse–anti-macrophage antibodies and then poured into a large petri dish precoated with goat–anti-mouse IgG antibodies. Nonadherent cells are collected and then incubated in a second petri dish precoated with anti–Thy-1 antibody. After repeated washing with HBSS to remove nonadherent cells, the remaining cells are greater than 95% retinal ganglion cells. The attached cells are then released from the panning dish using mild trypsin digestion and placed in monolayer cell culture.
V.
TROUBLESHOOTING
A.
Pathogen Contamination
The appearance of yeast, fungi, and bacteria in cultures will alter results and eventually destroy the cultures. Yeast infection appears as a proliferation of round or ovoid particles within the cultures. Fungi form filamentous threads (mycelia) that often are branched. Bacteria appear as clumps of small rods or particles. Excellent pictures of cultures infected with these microbes are present in a cell culture manual by Freshney [34]. Mycoplasma also can infect cultures and reduce viability. It is not possible to see mycoplasma by microscopic inspection. However, they can be detected by staining cultures using the dye Hoechst 33258 or by use of a commercial test kit (available from Gibco BRL). Antibiotics rarely can control an established infection. Usually, the best course of action is to discard the contaminated cultures and suspected media, resterilize the equipment and workspaces of the culture laboratory, and then resume with new media and plates. B.
Chemical Contamination
Abnormal performance of solutions used in tissue culture procedures can reflect chemical contamination. For example, if glassware used to prepare media is insufficiently rinsed after washing, residual detergent can be left on the glassware walls and then contaminate solutions placed in the glassware. A failure in the performance of the laboratory water purifier also can introduce chemical contamination. Other possible sources of chemical contamination include a pH electrode that is inadvertently used for measuring nontissue culture solutions, and spilled chemicals on the pan of the laboratory microbalance used to weigh purified media supplements. C.
Inactivation of Tissue Culture Reagents
Several tissue culture reagents have a relatively short shelf life. For example, glutamine in solution becomes deactivated within several days. Hence, frozen
Culture of Retinal Neurons
9
stock glutamine often is added to media just before use. Certain peptide supplements can become deactivated by repeated freeze-thaw. Hence, they should be stored in a non–frost-free freezer. Certain enzymes, such as trypsin, can become deactivated by auto-digestion if kept at room temperature or 37°C for an extended period. Hence, frozen enzyme solutions used in tissue dissociation should be thawed in the water bath just before they are needed. D. Incorrect Incubator Settings or Function Incorrect concentration of carbon dioxide within the incubator will produce incorrect pH within the media. As mentioned above, the correct carbon dioxide concentration for a particular media can be obtained from the supplier. Correct performance of the incubator controls can be confirmed using a Fyrite flue-gas analyzer (available from Fisher Scientific). A small air sample is withdrawn from the incubator chamber through a special port without opening the incubator door. This air sample is exposed to an alkali solution that absorbs the carbon dioxide. The change in the volume of the air sample is measured to determine carbon dioxide concentration in the original sample. Regular testing of incubators will often identify performance problems prior to the loss of valuable experiments. E.
Phototoxicity
Examination of cultures using an inverted microscope is often the best way to monitor the status of retinal monolayer, explant, or slice cultures. However, the cells are easily damaged by too much light exposure. Use of a green filter can reduce the more-harmful short wavelength light and thus protect the cultures during examination. VI. RECOMMENDED RESOURCES Excellent books on basic cell culture technique are available that discuss the culture laboratory, aseptic technique, selection of culture supplies, sterilization, basic considerations in the generation of primary cultures, characterization of cell cultures, and contamination control [34,35]. A classic reference for the formulation of chemically defined media for neuronal cultures is a book series edited by Barnes, Sirbasku, and Sato [36]. Also helpful are several articles in recent editions of Methods in Cell Biology [37–45]. VII. CONCLUDING COMMENTS First, start with an established protocol, if available. If one that accomplishes what you want is not available, start with the most similar one you can find and
10
Lindsey
then move toward your goal by making and evaluating changes one at a time. Isolate your cell culture activities from other laboratory activities as much as possible. Standardize all procedures. Meticulously document the source, age, lot number, and usage of all media components. Minimize delay in the generation of cultures. Regularly examine cultures after they have become attached (this usually occurs within 8 h). Do not examine too many cultures at one time as they are sensitive to pH and temperature shock. Finally, document your results as completely as possible with each experiment, even when it seems that it did not work. This will often help considerably in identifying problems and developing improved methods for retinal cell cultures.
REFERENCES 1. 2.
3.
4. 5.
6.
7.
8. 9.
10. 11.
12.
de Araujo EG, Linden R. Trophic factors produced by retinal cells increase the survival of retinal ganglion cells in vitro. Eur J Neurosci 1993;5:1181–1188. Lindsey JD, Weinreb RN. Survival and differentiation of purified retinal ganglion cells in a chemically defined microenvironment. Invest Ophthalmol Vis Sci 1994; 35:3640–3648. Kashiwagi F, Kashiwagi K, Iizuka Y, Tsukahara S. Effects of brain-derived neurotrophic factor and neurotrophin-4 on isolated cultured retinal ganglion cells: evaluation by flow cytometry. Invest Ophthalmol Vis Sci 2000;41:2373–2377. Shindler KS, Zurakowski D, Dreyer EB. Caspase inhibitors block zinc-chelator induced death of retinal ganglion cells. Neuroreport 2000;11:2299–2302. Yasuyoshi H, Kashii S, Zhang S, et al. Protective effect of bradykinin against glutamate neurotoxicity in cultured rat retinal neurons. Invest Ophthalmol Vis Sci 2000; 41:2273–2278. Morgan J, Caprioli J, Koseki Y. Nitric oxide mediates excitotoxic and anoxic damage in rat retinal ganglion cells cocultured with astroglia. Arch Ophthalmol 1999; 117:1524–1529. Saga T, Scheurer D, Adler R. Development and maintenance of outer segments by isolated chick embryo photoreceptor cells in culture. Invest Ophthalmol Vis Sci 1996;37:561–573. Romano C, Price MT, Olney JW. Delayed excitotoxic neurodegeneration induced by excitatory amino acid agonists in isolated retina. J Neurochem 1995;65:59–67. Sherry DM, St JR, Townes-Anderson E. Morphologic and neurochemical target selectivity of regenerating adult photoreceptors in vitro. J Comp Neurol 1996;376: 476–488. Adler R, Lindsey JD, Elsner CL. Expression of cone-like properties by chick embryo neural retina cells in glial-free monolayer cultures. J Cell Biol 1984;99:1173–1178. Madreperla SA, Adler R. Opposing microtubule- and actin-dependent forces in the development and maintenance of structural polarity in retinal photoreceptors. Dev Biol 1989;131:149–60. Stalmans P, Himpens B. Confocal imaging of Ca⫹⫹ signaling in cultured rat retinal
Culture of Retinal Neurons
13.
14. 15.
16.
17. 18.
19. 20.
21. 22.
23.
24. 25. 26.
27.
28. 29.
11
pigment epithelial cells during mechanical and pharmacologic stimulation. Invest Ophthalmol Vis Sci 1997;38:176–187. Marin N, Romero B, Bosch-Morell F, Llansola M, Felipo V, Roma J, Romero FJ. Beta-amyloid-induced activation of caspase-3 in primary cultures of rat neurons. Mech Ageing Dev 2000;119:63–67. Brewer G. Isolation and culture of adult rat hippocampal neurons. J Neurosci Methods 1997;71:143–155. Heidinger V, Hicks D, Sahel J, Dreyfus H. Peptide growth factors but not ganglioside protect against excitotoxicity in rat retinal neurons in vitro. Brain Res 1997; 767:279–288. Meyer-Franke A, Kaplan M, Pfrieger F, Barres B. Characterization of the signaling interactions that promote the survival and growth of developing retinal ganglion cells in culture. Neuron 1995;15:805–819. Rotstein N, Aveldano M, Politi L. Essentiality of docosahexaenoic acid in retina photoreceptor cell development. Lipids (Suppl) 1999;34:S115. Adler R. Regulation of neurite growth in purified retina neuronal cultures: effects of PNPF, a substratum-bound, neurite-promoting factor. J Neurosci Res 1982;8: 165–177. Carri NG, Perris R, Johansson S, Ebendal T. Differential outgrowth of retinal neurites on purified extracellular matrix molecules. J Neurosci Res 1988;19:428–439. Hewitt AT, Lindsey JD, Carbott D, Adler R. Photoreceptor survival-promoting activity in interphotoreceptor matrix preparations: characterization and partial purification. Exp Eye Res 1990;50:79–88. Tezel GM, Seigel GM, Wax MB. Density-dependent resistance to apoptosis in retinal cells. Curr Eye Res 1999;19:377–388. Rowe-Rendleman C, Mitchell CK, Habrecht M, Redburn DA. Expression and downregulation of the GABAergic phenotype in explants of cultured rabbit retina. Invest Ophthalmol Vis Sci 1996;37:1074–1083. Germer A, Jahnke C, Mack A, Enzmann V, Reichenbach A. Modification of glutamine synthetase expression by mammalian Muller (glial) cells in retinal organ cultures. Neuroreport 1997;8:3067–3072. Feigenspan A, Bormann J, Wassle H. Organotypic slice culture of the mammalian retina. Vis Neurosci 1993;10:203–217. Hoff A, Hammerle H, Schlosshauer B. Organotypic culture system of chicken retina. Brain Res Protocols 1999;4:237–248. Vollmer G, Layer PG, Gierer A. Reaggregation of embryonic chick retina cells: pigment epithelial cells induce a high order of stratification. Neurosci Lett 1984; 48:191–196. Pearlman EM, Seigel GM, Notter MF. Induction of c-fos by excitatory amino acids in developing chick retina is affected by changes in cellular interactions. J Neurosci Res 1993;36:252–259. Watanabe T, Voyvodic JT, Chan-Ling T, et al. Differentiation and morphogenesis in pellet cultures of developing rat retinal cells. J Comp Neurol 1997;377:341–350. Politi LE, Lehar M, Adler R. Development of neonatal mouse retinal neurons and photoreceptors in low density cell culture. Invest Ophthalmol Vis Sci 1988;29:534– 543.
12 30.
31. 32.
33.
34. 35. 36.
37. 38. 39. 40. 41. 42.
43. 44.
45.
Lindsey Abrams L, Politi LE, Adler R. Differential susceptibility of isolated mouse retinal neurons and photoreceptors to kainic acid toxicity. In vitro studies. Invest Ophthalmol Vis Sci 1989;30:2300–2308. Politi LE, Adler R. Generation of enriched populations of cultured photoreceptor cells. Invest Ophthalmol Vis Sci 1986;27:656–665. Fontaine V, Kinkl N, Sahel J, Dreyfus H, Hicks D. Survival of purified rat photoreceptors in vitro is stimulated directly by fibroblast growth factor-2. J Neurosci 1998; 18:9662–9672. Barres BA, Silverstein BE, Corey DP, Chun LL. Immunological, morphological, and electrophysiological variation among retinal ganglion cells purified by panning. Neuron 1988;1:791–803. Freshney RI. Culture of Animal Cells: A Manual of Basic Technique. New York: Wiley-Liss, 2000. Mather J. Introduction to cell and tissue culture: theory and technique. In: Introductory Cell and Molecular Biology Techniques. New York: Plenum Press, 1998. Barnes D, Sirbasku D, Sato G. Methods for preparation of media, supplements, and substrata for serum-free animal cell culture. In: Cell Culture Methods for Molecular and Cell Biology. New York: A.R. Liss, 1984. Finlay D, Wilkinson G, Kypta R, de Curtis I, Reichardt L. Retinal cultures. Methods Cell Biol 1996;51:265–283. Helmrich A, Barnes D. Animal cell culture equipment and techniques. Methods Cell Biol 1998;57:3–17. Lincoln CK, Gabridge MG. Cell culture contamination: sources, consequences, prevention, and elimination. Methods Cell Biol 1998;57:49–65. Loo DT, Rillema JR. Measurement of cell death. Methods Cell Biol 1998;57:251– 264. Malicki J. Development of the retina. Methods Cell Biol 1999;59:273–299. Mather JP. Making informed choices: medium, serum, and serum-free medium. How to choose the appropriate medium and culture system for the model you wish to create. Methods Cell Biol 1998;57:19–30. Mattson MP, Barger SW, Begley JG, Mark RJ. Calcium, free radicals, and excitotoxic neuronal death in primary cell culture. Methods Cell Biol 1995;46:187–216. Mills JC, Wang S, Erecinska M, Pittman RN. Use of cultured neurons and neuronal cell lines to study morphological, biochemical, and molecular changes occurring in cell death. Methods Cell Biol 1995;46:217–242. Moore A, Donahue CJ, Bauer KD, Mather JP. Simultaneous measurement of cell cycle and apoptotic cell death. Methods Cell Biol 1998;57:265–278.
2 Crush Injury of the Optic Nerve Michal Schwartz and Eti Yoles The Weizmann Institute of Science Rehovot, Israel
I.
INTRODUCTION
Optic neuropathies are chronic neurodegenerative diseases of the optic nerve [1–3], in which degeneration of the axons leads eventually to death of their corresponding cell bodies, the retinal ganglion cells (RGCs). The mechanism underlying the progression of disease is not yet fully understood but probably involves the activity of physiological compounds that become cytotoxic when their normal concentrations are exceeded. In order to study the mechanisms of RGC death, identify the nerve-derived mediators of toxicity causing the ongoing spread of damage, and screen compounds for their neuroprotective potential (i.e., their ability to arrest or reduce this secondary degeneration), suitable animal models are needed. We have established a rat model in which the optic nerve is subjected to a well-calibrated partial crush injury of the required severity [4,5]. Use of the model has made it possible to demonstrate self-propagating secondary degeneration [5], identify some of the mediators of degeneration common to many neurodegenerative disorders [6], study the molecular mechanisms underlying RGC death, and discover processes of neuroprotection [7–12]. Molecular mechanisms can also be studied in the severely crushed optic nerve of the mouse, an easily obtained model in which the availability of transgenic mice can be exploited for studies of the effects of relevant genes on RGC survival [13,14]. 13
14
Schwartz and Yoles
Figure 1 Anatomy of the rat visual system.
II. ANATOMY OF THE RAT VISUAL SYSTEM The axons of RGCs in the rat and mice, after exiting the retina through the optic nerve head, form the optic nerve. At the optic chiasm most of the fibers cross to the contralateral optic tract to reach the optic tectum (superior colliculus). Fewer than 10% of the fibers beyond the optic chiasm are ipsilateral [15]. Nearly all of the RGCs in each tract project to the superior colliculus on that side, and fewer than 40% of them have collateral projections to the dorsal lateral geniculate nucleus (LGN) [15–17].
III. PARTIAL CRUSH INJURY OF THE RAT OPTIC NERVE A.
Surgical Exposure of the Optic Nerve Intraorbitally
The intraorbital part of the optic nerve is longer in rodents than in other species, making it relatively easy to carry out experimental manipulations without impinging on adjacent tissues or harming the nerve itself. All surgical procedures are done under general anesthesia. We use a binocular operating microscope; the conjuctiva is incised lateral to the cornea, the retractor bulbi muscles are separated using curved blunt forceps, and the optic nerve is identified and exposed near
Crush Injury of the Optic Nerve
15
Figure 2 Cross-action forceps.
the eyeball by blunt dissection for 2.5–3 mm, and care is taken not to stretch the nerve. B. Calibrated Crush Injury A reproducible crush injury of graded severity is inflicted on the optic nerve by the use of pre-calibrated cross-acting (self-closing) forceps, which open when the handles are pressed and close when the handles are released [4]. The force exerted by the grasping jaws (and thus the severity of the crush lesion inflicted) is adjusted by varying the number of revolutions of the screw attached to the handle. Using the forceps, a moderate, mild, or very mild crush injury is inflicted on the exposed optic nerve about 1 mm distal to the eye, for a period of 30 seconds.
IV. SEVERE CRUSH INJURY OF THE MOUSE OPTIC NERVE To identify and characterize the molecules participating in the process of RGC death, it is necessary to devise an animal model that allows molecular manipulation. Establishment of the mouse model makes it possible to study the effects of severe optic nerve injury in genetically manipulated mice. For this purpose, all RGCs must be labeled 72 h before optic nerve crush. With the aid of a binocular operating microscope, the conjunctiva over the posterior pole of the eye of the anesthetized mouse is incised. The optic nerve is exposed by gentle blunt dissection between the surrounding muscle and the retrobulbar region, as described above for the rat. Using cross-action forceps and taking care not to interfere with the blood supply, we then crush the nerve for 2 s.
16
V.
Schwartz and Yoles
RETROGRADE LABELING OF RETINAL GANGLION CELLS
Because of the anatomical construction of the visual system, retinal ganglion cell (RGC) survival at any time after axonal injury can be quantified by the use of retrograde neuronal tracers. Properties of the tracers selected for this purpose should include (1) lack of any effect on neuronal viability and activity; (2) intense fluorescence; (3) resistance to fading; (4) absence of diffusion from labeled cells; and (5) relatively prolonged survival time. The RGCs that survive an optic nerve crush injury, and are potentially capable of being rescued by neuroprotective therapy, are the cell bodies of damaged fibers and of intact fibers that escaped the injury. To determine the total number of surviving RGCs, the protocol of choice is of the labeling prior to the injury. To assess the number of surviving RGCs with still-intact fibers the protocol of choice is the post-injury labeling. These two protocols are done by employing the retrograde labeling procedures, as described in the following section. A.
Labeling of All Retinal Ganglion Cells Prior to Injury
The total number of RGCs in the retina is determined after stereotactic injection of a fluorescent dye to the superior colliculus of both hemispheres, where almost all of the optic axons form synapses. 1. The Rat Model The anaesthetized rat is placed in a stereotactic device, the skull is exposed and kept dry and clean, and the bregma is identified and marked (see diagram below). The designated point of injection is 6 mm rostral to the bregma and 1.2 mm lateral to the midline. A window is drilled in the scalp above the designated coordinates in both hemispheres. Using a Hamilton syringe, 2 µL of the neurotracer dye FluoroGold [18] (5% solution in saline; Fluorochrome, Denver, CO), which meets all of the criteria mentioned above, is injected into the superior colliculus 3.8 mm, 4 mm, and 4.2 mm below the bony surface, at a rate of 1 µL/min at each of the three depths. The needle is then slowly withdrawn and the skin is sutured. 2. The Mouse Model Anesthetized mice are placed in a stereotactic device, the skull is exposed and kept dry and clean, and the bregma is identified and marked. The designated point of injection is at a depth of 2 mm from the brain surface, 2.92 mm posterior to the bregma and 0.5 mm lateral to the midline. A window is drilled in the scalp above the designated coordinates in both hemispheres. With a Hamilton syringe,
Crush Injury of the Optic Nerve
17
Figure 3 The rat skull.
the neurotracer dye FluoroGold (5% solution in saline) is then applied (1 µL, at a rate of 0.5 µL/min) and the skin over the wound is sutured. B. Post-injury Labeling of Cell Bodies of Rescude Fibers Post-injury application of the fluorescent lipophilic dye 4-(4-(didecylamino)styryl)-N-methylpyridinium iodide (4-Di-10-Asp) [19] (Molecular Probes, Europe BV) distal to the site of optic nerve lesion results in the labeling of RGCs with intact axons, as only axons whose continuity is preserved across the site of injury are capable of transferring the dye to RGC bodies. At different times after crush injury, the optic nerve is reexposed intraorbitally as described above. With the use of a 27-G syringe, a small hole is made in the dura 1 mm from the distal border of the site of injury, and the axons are cut to allow dye uptake. Solid crystals (0.2–0.4 mm diameter) of the dye are deposited at the cut edge of the optic nerve. Five days after dye application, the number of labeled RGCs is determined. The dye application procedure has no effect on RGC survival during the period until retinal excision [5].
Figure 4 Application of a dye distally to the lesion site.
18
Schwartz and Yoles
Figure 5 Whole-mounted flattened retina.
VI. COUNTING OF LABELED RGCs At the end of the experimental period, the rats or mice are killed and their eyes are excised into petri dishes containing phosphate-buffered saline (PBS). The retina is detached from the eye without the vitreous body and fixed in freshly prepared 4% paraformaldehyde. Four cuts are made in the fixed retina to allow flattening of the retina onto a nitrocellulose filter. Labeled RGCs are counted using the fluorescent microscope. It should be noted that RGC density across the rat retina ranges from about 1000 cells/mm2 at the periphery to 6000 cells/mm2 in the center. However, over most of the retina, except at the outer periphery, the average density is about 3000 cells/mm2 [5,15]. Nevertheless, after optic nerve injury the rate of RGC death is higher at the periphery than at the center of the retina [5]. Accordingly, labeled RGCs are counted in four to six fields at the same distance from the center of the retina, at a magnification of ⫻250. The numbers of labeled RGCs per field are averaged, and the mean number of RGCs per square millimeter is calculated.
VII. ELECTROPHYSIOLOGICAL MEASUREMENTS The visual evoked potential (VEP) response to light indicates the integrity of an animal’s visual system and can be used to assess the effects of injury and treatment on the system’s functional integrity. Only those axons that escaped the primary lesion and remain intact, with or without protection from secondary de-
Crush Injury of the Optic Nerve
19
generative processes, are capable of conducting action potentials. The pattern of field potentials in response to light stimulation is recorded from the primary visual cortex. The potential evoked by the light originates in the retina and is propagated along the surviving axons to reach the superior colliculus, their target in the brain. Electrodes are implanted above the primary visual cortex as follows: The anesthetized rat is placed in a small stereotactic instrument and two holes are drilled in the skull. Through each hole an electrode is implanted epidurally, with the dura kept intact to minimize cortical damage. The electrodes are gold contact pins (Wire-Pro, Salem, NJ) soldered to stainless steel screws, which are screwed into the holes and cemented to the skull with acrylic cement. An electrode inserted through a hole drilled in the nasal bone is used as a reference point. The second hole is in area V1 (primary visual cortex), with coordinates bregma ⫺8 mm and lateral 3 mm. Field potentials, before and after injury, are recorded from the visual cortex in response to stroboscopic light stimulation (xenon flash tube 4 W/s, 1–2 ms duration, 0.3 Hz). The signal evoked in the cortex is amplified 1000 times with a microelectrode AC amplifier, model 1800 (AM Systems) and digitized (12 bits, 5000 samples/s) with an MIO16–9 board and the LabView 2.2.1 data acquisition and management system (National Instruments). Potentials should be presented as the means of six recordings, three with and three without light, each involving 60 light flashes. VIII. SOME PRACTICAL TIPS 1. When incising the conjunctiva, make sure that you do it as far as possible from the limbal area, which has abundant vasculature. An incision at this site might cause massive bleeding, obscuring the area close to the optic nerve. 2. While exposing the optic nerve, separate it as much as possible from the adjacent fat and fascia. 3. While labeling the RGCs with a lipophilic fluorescent dye such as 4Di-10-Asp, make sure that the dye is completely immersed in the hole you have made in the dura sheath by injecting a drop of incomplete Freund’s adjuvant. 4. If the view of the retina under the microscope is too blurred to count the cells, the problem might be caused by one or more of the following: The vitreous body is still attached to the retina. The paraformaldehyde solution in which the retina is soaked is not fresh. Dye application process went wrong.
20
Schwartz and Yoles
REFERENCES 1.
2.
3. 4.
5. 6. 7.
8. 9.
10.
11.
12.
13.
14.
15.
Schwartz M, Belkin M, Yoles E, Solomon A. Potential treatment modalities for glaucomatous neuropathy: neuroprotection and neuroregeneration. J Glaucoma 1996; 5:427–432. Schwartz M, Yoles E, Levin LA. “Axogenic” and “somagenic” neurodegenerative diseases: definitions and therapeutic implications. Mol Med Today 1999; 5:470– 473. Schwartz M, Yoles E. Optic nerve degeneration and potential neuroprotection: implications for glaucoma (abst). Eur J Ophthalmol 1999; 9 (suppl 1): S9–11. Assia E, Rosner M, Belkin M, Solomon A, Schwartz M. Temporal parameters of low energy laser irradiation for optimal delay of post-traumatic degeneration of rat optic nerve. Brain Res 1989; 476:205–212. Yoles E, Schwartz M. Degeneration of spared axons following partial white matter lesion: implications for optic nerve neuropathies. Exp Neurol 1998; 153:1–7. Yoles E, Schwartz M. Elevation of intraocular glutamate levels in rats with partial lesion of the optic nerve. Arch Ophthalmol 1998; 116:906–910. Schwartz M, Moalem G, Leibowitz-Amit R, Cohen IR. Innate and adaptive immune responses can be beneficial for CNS repair. Trends Neurosci 1999; 22:295– 299. Schwartz M, Cohen IR. Autoimmunity can benefit self-maintenance. Immunol Today 2000; 21:265–268. Yoles E, Belkin M, Schwartz M. HU-211, a nonpsychotropic cannabinoid, produces short- and long-term neuroprotection after optic nerve axotomy. J Neurotrauma 1996; 13:49–57. Yoles E, Muller S, Schwartz M. NMDA-receptor antagonist protects neurons from secondary degeneration after partial optic nerve crush [published erratum appears in J Neurotrauma 1999 Apr;16(4):345]. J Neurotrauma 1997; 14:665– 675. Moalem G, Yoles E, Leibowitz-Amit R, Muller-Gilor S, Mor F, Cohen IR, Schwartz M. Autoimmune T cells retard the loss of function in injured rat optic nerves. J Neuroimmunol 2000; 106:189–197. Kipnis J, Yoles E, Porat Z, Cohen A, Mor F, Sela M, Cohen IR, Schwartz M.T cell immunity to copolymer 1 confers neuroprotection on the damaged optic nerve: possible therapy for optic neuropathies. Proc Natl Acad Sci USA 2000; 97:7446– 7451. Levkovitch-Verbin H, Harris-Cerruti C, Groner Y, Wheeler LA, Schwartz M, Yoles E. RGC death in mice after optic nerve crush injury: oxidative stress and neuroprotection. Invest Ophthalmol Vis Sci 2000; 41:4169–4174. Fisher J, Levkovitch-Verbin H, Schori H, Yoles E, Butovsky O, Kaye JF, Ben-Nun A, Schwartz M.Vaccination for neuroprotection in the mouse optic nerve: implications for optic neuropathies. J Neurosci 2001; 21:136–142. Linden R, Perry VH. Massive retinotectal projection in rats. Brain Res 1983; 272: 145–149.
Crush Injury of the Optic Nerve 16.
17. 18. 19.
21
Dreher B, Sefton AJ, Ni SY. Nisbett G. The morphology, number, distribution and central projections of Class I retinal ganglion cells in albino and hooded rats. Brain Behav Evol 1985; 26:10–48. Martin PR. The projection of different retinal ganglion cell classes to the dorsal lateral geniculate nucleus in the hooded rat. Exp Brain Res 1986; 62:77–88. Schmued LC, Fallon JH. Fluoro-Gold: a new fluorescent retrograde axonal tracer with numerous unique properties. Brain Res 1986; 377:147–154. Fritzsch B, Wilm C. Dextran amines in neuronal tracing. Trends Neurosci 1990; 13: 14.
3 Intraocular Pressure Elevation: Vein Cauterization Sansar C. Sharma New York Medical College Valhalla, New York, U.S.A.
I.
INTRODUCTION
Glaucoma is an optic neuropathy in which damage primarily occurs to the optic nerve axons and the retinal ganglion cells. The death of neurons and damage to the axons is related to the increase in intraocular pressure. The need to develop a suitable animal model stems from the fact that experimentation on monkeys (the ideal animal to study progression of a pathology) is expensive. In an ideal situation, an animal model should provide results that are consistent and cost effective. Our model is based on the concept that obstructing the outflow of aqueous humor would mimic the disease process, thereby providing conditions suitable for studying the effect of various neuroprotective agents. In the following sections we describe a method by which chronic elevation of intraocular pressure is induced in the rat eye. The procedure is simple and reproducible, with little or no inflammatory response.
II. ANTERIOR DRAINAGE PATHWAYS OF THE RAT EYE One of the major pathways is via the anterior chamber angle and into the Schlemm’s canal. The Schlemm’s canal is in communication with the venous 23
24
Sharma
plexus via the collector channels. The venous plexus drains into the episcleral veins, which are radially oriented and are called radial veins. They run posteriorly and merge with the large vessels near the extraocular muscle insertion. The large vessels join posteriorly with the inferior and superior ophthalmic veins and enter the cavernous sinus. Any interruption of the episcleral drainage will cause a rise in IOP and, possibly, secondary glaucoma. It is therefore post-trabecular glaucoma. Because the outflow is affected, this procedure may lead to open-angle glaucoma [1,2]. If one increases the post-trabecular meshwork resistance by cauterizing the deep episcleral veins, one would expect an increase in IOP. Described below are the essential steps in creating elevation of IOP in adult rat eyes. This method is reliable and reproducible. The range of elevated IOP pressure correlates with the number of cauterized veins. This model does not involve introduction of exogenous material in the eye.
III. PROCEDURE 1. Adult Wistar rats (250–300 g) are anesthetized with intraperitoneal injection of 0.1 to 0.15 mL mixture of acepromazine maleate (1.2 mg/kg), xylazine (8 mg/kg), and ketamine (40 mg/kg). Anesthetic injection can be given intramuscularly; however, it takes between 15 and 20 min to induce deep anesthetic condition to perform surgery. Certain hyperactive rats may require a second or third injection of the anesthetic mixture administered in much smaller doses than the initial dose. 2. To maintain the animals’ body temperature during the procedure, they should be placed on a heated pad. 3. For each animal, the contralateral eye should serve as a comparative control and should either have a sham operation or be left undisturbed. Because of individual variations in IOP, pre- and postoperative IOP should be compared between normal and experimental eyes for each animal. 4. In order to keep the eye open, place the sutures on the eyelids. 5. Four to five aqueous-containing radial veins emerge from the circumferential venous plexus in the rat eye. The supranasal vein should be located first. In order to stabilize the globe and expose the veins, make an incision 2–3 mm long within the limbal periphery using sterile microscissors. Then make two radial relaxing incisions at the edges of the initial incision and recess the tissue posteriorly to expose the underlying extraocular muscle. 6. Isolate the muscle (superior rectus) and anchor with a suture to expose the underlying episcleral-radial vein. Special attention should be paid
Intraocular Pressure Elevation: Vein Cauterization
25
Figure 1 The photograph shows the location of the episcleral radial (arrow) merging caudally with the ciliary vein. Following the incision to conjunctiva in the rat eye, superior rectus muscle is anchored by a suture (black) and pulled forward to expose the circumferential veins.
to minimize the blood loss and damage to the conjunctiva and, especially, the sclera. Aqueous-containing radial veins lie slightly deeper to the ocular muscles. The radial veins can be distinguished by their darker color. The radial veins travel on the surface of the sclera (Fig. 1). 7. The radial veins merge with the ciliary veins at about one-third the distance between the episcleral venous plexus and the optic nerve head. Near the junction, isolate the radial vein with the least damage to the sclera and put an open number-five forcep under it to create a bridge. Apply the cautery at the center of the isolated vein. The cautery tip should be about the same size as the outer diameter of the vein. The purpose of this procedure is to minimize thermal damage to the sclera during cauterization. Different material, such as wood or metal, can be placed under the isolated vein to protect from heat. As the vein is cauterized, a global mass appears on both ends of the severed vein. At the proximal portion of the cauterized vein, there is a distention of the radial vein, as the drainage is blocked. Even a
26
Sharma
8.
9.
10.
11.
12.
minor leak of the fluid at the cauterized end will encourage neovascularization. Incidentally, this happens in 10 to 15% of the surgeries. In each of these cases, we have observed formation of microvessels within a week. Measure IOP 2 to 3 days after surgery and check the sites of cauterization; if any are leaky, recauterize them. A successful operation requires that vessels not be reconnected. Another method is to isolate the vein, ligate it, and then cauterize. In Wistar rats there are four to five major venous trunks that run posteriorly in each eye. Usually they are equidistant around the circumference of the globe. On the dorsal aspect of the globe there are two major vessels equidistant from the superior oblique and the superior rectus muscles. Near the superior rectus muscle, the radial vein is located closer to the temporal border. Similarly, there are two episcleral venous trunks located on the ventral aspect of the eye. One trunk is located between inferior rectus and inferior oblique and the other located at the inferior border of the lateral rectus muscle. Occasionally, some rats have a fifth venous trunk that is located deeper than the medial rectus. In order to elevate IOP to approximately double the value of normal IOP, it is advisable to cauterize a minimum of two and a maximum of three vessels. If all venous trunks are cauterized, the eye becomes necrotic within a week. Expose the second and third venous trunks by manipulating the anchored superior rectus muscle or superior oblique muscle. The isolation procedure is similar to that described above. After the venous occlusion, flush the eyes with saline and apply antibiotic ophthalmic ointment. The veins can be cauterized at the junction of the episcleral vein and ciliary vein (Fig. 2, incision 2) which leads to elevated IOP. This procedure does not lead to any change in constriction and or dilation of the pupil. No inflammation was noticed in the anterior aspect of the eye. Some practitioners of this model have utilized neovascularizaton inhibitors such as 5-fluorouracil applied subconjunctively for the first few days after venous occlusion. It is highly advisable not to measure the IOP on the first day after surgery, as chances of dislodging the cauterized plug are very high. Monitoring IOP can begin within 2– 3 days, when conjuctival incisions are healed. Following the occlusion of three veins, the IOP is usually in the range of 28 to 30 mmHg (normal average being 13.0 mmHg). Within 2 to 3 days following surgery, there is a small drop in the IOP. Subsequently, the elevated IOP remains high (22 to 25 mmHg) for the longest period of study (120 days).
Intraocular Pressure Elevation: Vein Cauterization
27
Figure 2 Diagramatic representation of the rat eye showing circumferential venous plexus (ven. plex.) and the Schlemm canal (Schl. can.). Notice the anchoring of superior rectus (sup. rec.) muscle. Collector channels (coll. cha.) connect the Schlemm canal with the venous plexus, which subsequently exit via radial veins on the scleral surface, and carry aqueous (rad. vein. aqu.). Interruptions 1 and 2 on the scleral vessel mark the point where cauterization leading to the elevation of IOP can be made.
IV. METHODS FOR MEASURING IOP For rats in the awake state, animals are held on a flat surface gently with minimal pressure applied to the shoulder. Press gently on the head. Apply topical anesthesia (a drop or two of proparacaine), and measure the IOP. Using Tonopen X-L (Mentor Ophthalmics) applanation is made and a reading is noted. The Tonopen probe usually requires three to four applications to the cornea before its processor is activated. Routinely, four readings are obtained and averaged and the mean values recorded. The IOP of each experimental eye is compared with the contralateral unoperated eye. For the sake of consistency and to exclude variations in IOP due to diurnal cycle, measurements should be made at one set time. We measure IOP between 9 and 10 a.m. Measurements are made twice a week in
28
Sharma
the first 2 weeks following the surgery and once a week thereafter for the duration of the experiments. When using Mentor 1 pneumotonometer, we mildly anesthetize the rat, with the same anesthetic used for the initial surgery. Three consecutive IOP measurements are recorded. Mentor 1 pneumotonometer provides consistent IOP preferences, and experience of each investigator will play a greater role in choosing the method of IOP measurement. In anesthetized Wistar rats the mean IOP of the normal eye is 12.5 mmHg, whereas in restrained but awake rats it is 13.5 mmHg. The range of normal IOP in Wistar rats is between 10 and 16 mmHg.
V.
ADVANTAGES AND DISADVANTAGES 1.
2.
3.
This procedure requires careful surgical manipulations and experience. In initial trials, our success rate was 25%. After realizing the formation of new vessels at the occluded end and being exceedingly careful in all surgical procedures, we reached a success rate of 80 to 90%. If more than one rat is transferred to a cage, this usually leads to some aggressive behavior, which amplifies the chance of reopening the wound. It is preferable to keep rats isolated, one per cage. In less than 5% of cases, corneal lesions or hemorrhaging of the globe occurred. Affected animals should be excluded from any studies.
The elevated IOP can be maintained consistently in a large number of animals for a considerable period of time. Therefore, this procedure leads to the development of an excellent experimental system for studying the mechanisms of cell death of retinal ganglion cells and the effects of various neuroprotective agents. Using this model system, we have studied the effects of neurotrophins and gene transfer to the retinal ganglion cells in glaucomatous animals and animals with optic nerve transections [3–8]. Prelabeling of all retinal ganglion cells exclusively in the retina was achieved by use of Fluorogold (Fluorochrome Inc., Englewood, CO) following protocols described in the above references.
VI. CONCLUSIONS The procedure described above serves as a simplified means of inducing chronic elevation of intraocular pressure and experimental glaucoma in rodents. This procedure clearly allows viewing of the optic disc [9] and does not induce a strong inflammatory response within the eye [10]. This procedure has furthered our understanding of the relationship between IOP and optic neuropathy on one hand
Intraocular Pressure Elevation: Vein Cauterization
29
and the evaluation of various neuroprotective agents on the other—and will continue to do so.
REFERENCES 1. 2. 3. 4.
5.
6.
7.
8.
9. 10.
Minas TF, Podos SM. Familial glaucoma associated with elevated episcleral venous pressure. Arch Ophthalmol 1968; 80:201–208. McGovern RA. Glaucoma associated with raised episcleral venous pressure. In: Cairns JE, ed. Glaucoma. Vol. 2. Florida: Grune & Stratton. 1986:711–728. Chaudhary P, Ahmed FAK, Sharma SC. MK801—A neuroprotectant in rat hypertensive eye. Brain Res 1998; 792:154–158. Chaudhary P, Ahmed FAK, Quebada P, Sharma SC. Caspase inhibitors block the retinal ganglion cell death following optic nerve transection. Mol Brain Res 1999; 67:36–45. Garcia-Valenzuela E, Sharma SC. Rescue of retinal ganglion cells following axotomy induced apoptosis through TRK oncogene transfer. Neuroreport 1998; 9: 3165–3170. Garcia-Valenzuela E, Shareef S, Walsh J, Sharma SC. Programmed cell death of retinal ganglion cells during experimental glaucoma. Exp Eye Res 1995; 61:33– 44. Ko ML, Hu DN, Ritch R, Sharma SC. The combined effect of brain derived neurotrophic factor and a free radical scavanger in experimental glaucoma. Invest Ophthalmol Vis Sci 2000; 47:2967–2971. Shareef SR, Garcia-Valenzuela E, Salierno A, Walsh J, Sharma SC. Chronic ocular hypertension following episcleral venous occlusion in rats. Exp Eye Res 1995; 61: 379–382. Sawada A, Neufeld AR. Confirmation of the rat model of chronic moderately elevated intraocular pressure. Exp Eye Res 1999; 69:525–531. Mittag TW, Danias J, Pohorenec G, Yuan HY, Burakgazi E, Chalmers-Redman R, Podos SM, Tatton WG. Retinal damage after 3 to 4 months of elevated intraocular pressure in a rat glaucoma model. Invest Ophthalmol Vis Sci 2000; 41:3451–3459.
4 Intraocular Pressure Elevation: Injecting Hypertonic Saline into Episcleral Veins John C. Morrison, Elaine C. Johnson, Lijun Jia, William O. Cepurna Casey Eye Institute Oregon Health & Science University Portland, Oregon, U.S.A.
I.
INTRODUCTION
Recent studies have affirmed the effectiveness of aggressively treating intraocular pressure (IOP) for stabilizing visual field loss in patients with advanced glaucomatous optic nerve damage [1]. However, many patients will still suffer progressive visual field loss despite what may appear to be adequate lowering of IOP. It is for these patients that effective neuroprotection offers the best hope for preserved vision. Some of these patients already have advanced glaucomatous optic nerve damage. For others, it may not be possible to achieve maximal pressure lowering because of medication side effects, surgical failure, or unacceptable risks of surgery. All of these considerations, as well as the observation that patients with normal tension glaucoma can also benefit from aggressive IOP control [2,3], suggest that understanding the mechanisms of pressure-induced optic nerve damage may be important for developing effective neuroprotection. This would be particularly true if some of these mechanisms are irreversible, rendering the remaining optic nerve fibers vulnerable to otherwise normal IOP. It is clear that modeling optic nerve damage by creating chronically elevated intraocular pressure (IOP) offers an important opportunity to develop meth31
32
Morrison et al.
ods to protect or preserve the optic nerve in glaucoma. The optic nerve head is a highly complicated structure, consisting of axons surrounded by glial cells passing through a pressure gradient, with a unique vascular supply. Because of this, such models must be created in the intact, functioning eye. Once such a model is in place, it provides a reproducible system for testing the effectiveness of potential neuroprotective compounds. II. OVERVIEW OF METHODS IN THE CONTEXT OF NEUROPROTECTION The primate model of trabecular meshwork sclerosis from argon laser treatment has proven highly useful over the past 3 decades [4–6]. The advantages of this approach are that the primate optic nerve anatomy is very similar to that of the human, and the pathology of chronically elevated IOP in monkeys bears many similarities to that of human glaucoma. However, the cost and difficulty of working with this model precludes its use for studies assessing the cellular response to IOP and other experimental manipulations, where individual variability necessitates the use of large numbers of animals. This is also true for studies designed to test the effect of potential neuroprotective agents. The development of chronic IOP elevation in laboratory rats provides an alternative model, which has many advantages [7–9]. First, these animals are easy to handle, allowing frequent measurement of IOP with the animal awake. Second, their relatively low cost makes it possible to use them in studies requiring large numbers of animals. Third, a large body of knowledge based primarily on the rat already exists with regard to the cellular and molecular biology of central nervous system and optic nerve damage. This provides opportunities for understanding the cellular mechanisms of pressure-induced optic nerve damage. Finally, the anatomy of the rat optic nerve head bears several features in common with that of the human [10,11] and the pathology of elevated pressure in the model described here has many similarities to that of human glaucoma [12]. A reliable model of pressure-induced optic nerve damage must have three components: a method of creating elevated IOP; a reliable, unbiased method of measuring IOP that will provide objective understanding of the pressure experienced by the eye and the optic nerve head; and a rapid, reproducible system for assessing the resulting optic nerve damage. By using all of these features such a model can be used to develop a reliable understanding of the mechanism of pressureinduced optic nerve damage and to assess potential neuroprotective therapies. Three basic methods of creating chronically elevated IOP in rats have been described. These include cauterization of large episcleral veins [13], laser photocoagulation of the angle vessels [14], and episcleral vein injection of hypertonic saline [7]. The former two methods are described and discussed in other chapters of this book. This chapter will describe our method of producing scarring of the aqueous humor outflow pathways using hypertonic saline injection of the episcleral veins.
Intraocular Pressure Elevation: Hypertonic Saline
33
In addition to the method itself, we will present our experience with measuring IOP and determining the extent of optic nerve damage produced by this technique.
III. CREATING AQUEOUS OUTFLOW OBSTRUCTION BY HYPERTONIC SALINE INJECTION OF EPISCLERAL VEINS A. Procedure Objectives The major objective in creating a chronic model of elevated IOP is to selectively obstruct aqueous outflow. In the rat, this is complicated by the close proximity of the aqueous outflow pathways within the angle of the eye and the blood supply to the ciliary body that comes from the major arterial circle of the iris. A too vigorous destruction of the outflow system can easily compromise the blood supply to the ciliary body and produce unwanted hypotony. Our method produces selective injury to this outflow system, based on the anatomy of aqueous humor outflow. In the rat, the primary route of aqueous humor outflow is through the trabecular meshwork and into Schlemm’s canal (Fig. 1). Following this, it escapes through numerous aqueous collector channels and into the circular venous plexus
Figure 1 Schematic of the anatomy of aqueous humor outflow in the rat eye. Aqueous humor (dots) moves into Schlemm’s canal (SC), through trans-scleral collector channels and into the limbal venous plexus. From here, flow occurs posteriorly within radial aqueous-containing veins (which also contain blood) (AV). The arterial supply to the anterior segment is also illustrated, showing arterial supply from anterior ciliary arteries (ACA) and long posterior ciliary arteries (LPCA). These arterioles interconnect via a circular limbal artery. The iris (I) and ciliary process (CP) arterial blood supply arise from the major arterial circle of the iris. (From Ref. 15.)
34
Morrison et al.
Table 1
Injection Equipment
Operating microscope (at least 16⫻3 with 10⫻3 oculars) with foot-drive focus Dumont #5 forceps (0.05 ⫻3 0.01 mm) Curved Vannas scissors (Storz) Curved Mosquito hemostatic forceps Heavy tissue (iris) scissors Weck cell sponges Dumont #7, reverse action curved forceps (blunted tip) Pump (settings to microliters per minute) Plastic (delrin) ring Hypertonic saline (0.22 Fm millipore filter) Microneedle
that encircles the limbus. Numerous radial veins drain aqueous (and blood) away from this plexus, posteriorly within the episclera [15]. We create isolated scarring of the trabecular meshwork by injecting a sclerosing agent (hypertonic saline) into the episcleral vein in retrograde fashion. By applying a resilient plastic ring around the equator of the eye, with a gap straddling the vein to be injected, the other episcleral veins are temporarily blocked off. The injection, via a microneedle into episcleral veins, is confined to the limbal plexus, and the saline is forced into Schlemm’s canal and the trabecular meshwork. We have found that the needle insertion is best done by hand, as a micromanipulator is too cumbersome and not easily adaptable to the anatomical variations in limbal vasculature. B.
Equipment
The necessary equipment for the entire procedure is listed in Table 1. Methods of manufacturing the microneedle used for injecting hypertonic saline and a description of the plastic ring are presented in detail below. C.
Microneedle Construction
The equipment needed to construct the microneedle is listed in Table 2. The steps in constructing the microneedle consist of heating the polyethylene tubing over a bunsen burner and stretching it to an even taper. This produces a segment leading from normal diameter to a thinnest dimension that is approximately 15 to 20 cm long. A 5–6 mm section of glass microneedle (pulled out on a needle puller) is then inserted into the tip of the tubing and a drop of the glue is applied
Intraocular Pressure Elevation: Hypertonic Saline Table 2
35
Equipment for Manufacturing the Microneedle
P50 tubing 10 Fl borosilicate micropipettes Bunsen burner Epoxy glue Needle puller (to pull micropipette to approx 50 µm outer diameter and 30 µm inner diameter 23-G needle (tip broken off) 1 cc syringe Abrasive wheel (Dremel tool)
to the junction. This microneedle must have little or no taper, so that the needle will provide an effective seal of the vessel wall where it is inserted. A tapered needle, as seen with a standard micropipette, will leak if the needle is not held perfectly still within the vessel, and it is allowed to slide back and forth. A small amount of the glue will naturally migrate between the tubing and the needle, so an equal length segment of glass must be inside the tubing to prevent the glue from covering and plugging up the internal opening of the glass needle. The large end of the tubing is then swedged onto the 23-G needle shaft (with the tip broken off) and secured with a drop of glue (Fig. 2). Once this is entirely dry (24 hs), the needle is beveled using the fine abrasive wheel on a dremel tool. Water applied to the stone provides good lubrication and produces a smooth, sharp bevel. When a needle is no longer usable, it can be cut off the end of the tubing and a new one glued in its place. This allows the tubing, which can be difficult to pull to a taper, to be reused several times. D. Plastic Ring The plastic ring is essential for isolating the injection to the limbus. It compresses all of the veins draining aqueous away from the limbus, except the one being injected, and confines the saline to the limbal vessels and Schlemm’s canal. This ring is manufactured from Delrin stock using a mini lathe. Dimensions and appearance of the current ring are shown in Figure 3. There is a 1 mm wide groove machined in the inner aspect of the ring. This gives the ring a U shape when cut in cross section and improves its stability when applied to the equator of the eye. A small gap is cut in this ring, allowing the ring to straddle the vein to be injected. We have found that this gap is most easily cut with a razor blade, with the sides beveled away from each other. This allows better access to the vessel with the microinstruments, both for dissection and for the needle placement.
36
Morrison et al.
Figure 2 Microneedle and syringe construction (above), with detail of needle attachment to tubing (below). (From Ref. 7.)
E.
Steps of Episcleral Vein Injection 1.
2.
3.
A lateral canthotomy is made to improve access to the eye following a 5 s compression with a hemostat. This heals quickly and does not require suturing. The best position for the ring is at the equator of the eye, and it should be placed under the dissecting or surgical microscope. The ring is spread, with a fine mosquito forceps, just large enough to slip over the equator, then slid off the forceps with a finger. This should immediately blanch out the limbal vessels (except possibly the artery) for a few minutes, due to aqueous being forced out of the anterior chamber. The plexus and episcleral veins will refill with blood once IOP falls enough to allow blood to refill the limbal plexus, either from filling of the long posterior ciliary artery, the anterior ciliary artery, or both. At this point, any possible routes of saline escape from the limbus should be identified and the ring shifted to occlude them, still leaving the injection vein unobstructed. Generally, the most accessible and largest veins are located superiorly.
Intraocular Pressure Elevation: Hypertonic Saline
(a)
37
(b)
Figure 3 Delrin ring used to isolate saline injection to the limbal vasculature and aqueous outflow pathways. (a) Appearance of the ring, both whole and when cut in cross section; (b) overall dimensions of the ring, as well as the gap cut in the ring, with beveled sides to improve access to the vessel.
Thus, the animal is best positioned on his chest, with the superior aspect of the globe pointed up. Excellent illumination is essential, particularly when working with the higher magnifications. An assistant, who is operating and timing the injection pump, can rotate the eye down to improve access to the vein, approximately at the equator. 4. The vessel wall is first exposed through a conjunctival incision. This is performed using the Dumont forceps to put traction on the conjunctiva or Tenons just over the vessel. Complete removal of connective tissue improves the chances of inserting the needle into the vessel lumen, without creating a false passage between the outside of the vessel wall and perivascular adventitia. 5. Once the vessel is exposed, the surgeon grasps the needle, bevel up, with curved, reverse action forceps. The best place to grasp the needle is at the bead of glue that forms at the junction of the needle and the tubing. The glue protects the tubing and the glass from being crushed by the forceps. It is also helpful to place a groove in the inner edges
38
Morrison et al.
6.
7.
8.
9.
of the forceps blades. This provides a more positive grasp of the needle and keeps the needle from pivoting, since the sides of the glue joint are usually curved. Needle insertion is performed by grasping the vein with Dumont forceps posterior to the intended insertion site. This stabilizes the vein and increases its size by temporarily obstructing the venous blood as it flows away from the limbal plexus. It is best to choose a straight section of the vein to decrease the chance that the needle will be passed out of the vein accidentally. The technique for needle insertion involves placing the tip of the needle (which is held nearly parallel to the vein) against the vein wall. By advancing the needle toward the limbus slightly, the tip will engage the vessel wall. This is usually aided by slightly loosening tension on the vein, thus allowing it to gain its maximum size. Once engaged, the needle is passed forward into the lumen of the vein. Generally, this requires increasing tension on the vein slightly, so that it will not move. The needle can then be advanced into the vessel lumen, approximately 2 to 3 mm. Often, this is accompanied by a slight reflux of blood into the needle. Tipping the needle up slightly after entering the vessel lumen will also decrease the chance of accidentally passing the tip out of the opposite wall of the vein. The injection is begun once the needle is safely in the vessel. Accidental extravasation usually prevents successful cannulation and injection, although it can still be attempted. A failed injection can sometimes be remedied by finding and injecting an alternative vessel, or going to the fellow eye. Usually, the eye has to be abandoned, but injection can be reattempted in 1 to 2 weeks. The injection is timed by noting when the pressure in the needle builds enough to fill the limbal vessels. This is the point at which timing starts. At the end of the injection time, the needle is removed and the vessel clamped with the Dumont forceps for several seconds to prevent reflux of the saline. Typically, good signs of a successful injection are immediate blanching of the limbal vasculature and deepening of the anterior chamber. A temporary lens opacity usually develops due to osmotic effects of the hyertonic saline in the anterior chamber. All of these are consistent with the saline being forced into Schlemm’s canal and across the trabecular meshwork. Following the injection, the limbal vessels will refill slowly, usually beginning with the limbal artery. Antibiotic ointment is instilled without suturing and the animal allowed to recover from the anesthesia. We do not typically use post-operative steroids or repeated dosing with antibiotics.
Intraocular Pressure Elevation: Hypertonic Saline
39
The typical postoperative course involves a mild amount of inflammation, usually manifested by slight corneal haze for a day or two. This almost always is self-limited. Following the injection, the IOP may initially show a brief elevation, but generally is normal or low for a few days. This is then followed by a gradual increase. Most eyes will demonstrate a rise in IOP by 10 days to 2 weeks after the injection. This is most likely due to gradual scarring of the trabecular meshwork and angle, producing a variable degree of angle closure. Once aqueous humor formation returns to normal, the closed angle is not able to accommodate the increased flow, resulting in elevated IOP. We have found that, when performed on animals housed in a 12-h light: dark cycle, the scarring of the aqueous ouflow pathways produces elevated pressures that can vary markedly with the circadian cycle [16]. This can range from the low 20s to the mid 40s from the light to the dark phase of the cycle, representing a doubling of the normal circadian rhythm [17]. We have also found that placing the animals in a constant, low level light environment will increase the mean IOP to approximately 27 mm Hg and minimize the extent of circadian fluctuation. In this situation, the typical range of mean IOP following the procedure varies from the low 30s to low 40s, as determined by the Tonopen in awake animals (see below). The success rate of producing an elevated pressure with this method is quite high. Review of a recent group of 40 eyes shows that pressures were significantly elevated in 33 eyes following a single injection. A second injection is always possible in eyes that fail to respond, and it can increase the success rate even further [7]. F.
Factors That Influence the Success of a Given Injection in Producing Elevated IOP
Three main factors influence the success of saline injection. These factors appear to affect each other, and experience, correlated with success in producing elevated IOP, is required to find the proper combination of saline concentration, injection force, and duration of injection. 1. Molarity of the saline solution. We have found that a 1.75–2.5 M solution produces a reliable increase in IOP. Concentrations higher than this will result in excessive inflammation and, often, long-lasting hypotony. 2. Force of injection. This influences the ability of the saline to gain access to the trabecular meshwork. Too low a force will fail to perfuse the angle structures sufficiently. An excessive force may produce too much inflammation, with a result similar to that of using too high a concentration of the saline. This may be due to subtle interconnections that exist between the episcleral vasculature and emissary veins that
40
Morrison et al.
3.
G.
drain venous blood from the ciliary body to the episclera. A high injection force may also cause perfusion of the ciliary processes with the saline. We have found that using a pump for this injection, aiming for a force sufficient to inject at a rate of 50 µL over 30 s, helps achieve more reproducible injections. Duration of injection. We generally try to inject for 15 to 30 s. Longer times will produce excessive inflammation.
Reasons for Failure of Injection to Produce Elevated IOP
These generally correspond to inadequate delivery of the saline into the outflow system. They often correlate with a lack of anterior chamber deepening and lens opacity, and with poor blanching of the limbal vasculature. The most common cause of this is if the ring fails to adequately occlude all vessels leading away from the limbus. This may be due to improper placement of the ring, or anatomic peculiarities of vessels that make it impossible to adequately occlude all of these vessels. Occasionally there are large, deep veins that appear to drain directly from the ciliary body, emptying into the limbal plexus. We generally avoid these if it is not possible to occlude the vessel with the ring, as these vessels are large enough to allow much of the saline to drain away from the limbal plexus. Accidental injection of these vessels may also produce inadvertent injection of the ciliary body vasculature, which could produce ciliary body shutdown and hypotony.
IV. IOP MEASUREMENT IOP measurement must be done in an unbiased fashion, using a method that is objective and calibrated to actual IOP [18,19]. We use the Tonopen XL tonometer for measuring IOP. This is done using topical anesthesia, taking 10 consecutive readings and then using the average as the measured IOP. Acceptable readings are those that register immediately upon contact of the tonometer tip with the cornea, using firm but not excessive force. The eye itself should be moved posterior slightly with each applanation of the tonometer. The operator must learn to recognize valid readings and objectively ignore all those that do not meet these criteria. A very light touch with the corneal tear film may produce a low single digit reading that is not accurate. In addition, readings that occur when the tip breaks contact with the cornea (“off ” readings) and the instrument generated averages are also not accurate. The key challenge with this tonometry is that the high curvature of the rat cornea requires that the
Intraocular Pressure Elevation: Hypertonic Saline
41
tonometer be held exactly perpendicular to the corneal surface. This positioning is acquired with practice. Measuring IOP in the rat eye requires considerable experience and this skill is best acquired by first calibrating the instrument on a cannulated rat eye connected to a low pressure transducer and an extra syringe to adjust the IOP, described below. With this setup, the individual learns, through immediate feedback, how to judge an acceptable reading. With time and practice, the ability to measure IOP in an unbiased fashion improves, and this can be seen by a steadily diminishing standard deviation in the measurements obtained. Finally, we have found that general anesthetics can produce a rapid and substantial decrease in measured IOP [20]. Because of this, we now measure all IOPs with animals in the awake state, using only topical anesthesia. This allows the best understanding of the pressure to which the eye and optic nerve are actually exposed. This technique also requires considerable practice. However, Brown Norway rats are very docile and rapidly become accustomed to these measurements. Once this technique is mastered, measuring awake IOP is actually much faster, and obviously more accurate, than under any general anesthetic. Periodic “refresher” calibration sessions are useful, even after the practitioner becomes skillful with this technique. These help the practitioner maintain the ability to sense acceptable readings and help uncover any possible systematic errors that may develop. Such sessions should also be done whenever the Tonopen is serviced or a new one is purchased.
V.
TONOMETER CALIBRATION
The tonometer is best calibrated in the living rat eye. The objective is to manipulate the IOP in known amounts while allowing simultaneous measurements with the tonometer. To do this, the eye of an anesthetized rat is cannulated and connected to a manifold equipped with 3-way stopcocks. The manifold is also connected to a low pressure transducer and to a Hamilton syringe for altering IOP. A third port is connected to a 60 cc syringe to allow periodic refilling of the system, as needed, while another is connected to a sphygmomanometer for calibrating the transducer and chart recorder. By keeping the system between the eye, the transducer and the chart recorder open at the same time, actual IOP can be rapidly manipulated by one operator, who also records tonometer readings made by another directly on the chart strip as they are made. With this arrangement, IOP measurements can be obtained with the tonopen and compared immediately to actual IOP, obtained through the transducer. This provides immediate feedback, helping the Tonopen operator improve technique and learn to recognize valid IOP readings in an unbiased fashion.
42
Morrison et al.
The most difficult part of this technique is in cannulating the anterior chamber without damaging the lens or iris. We first make a beveled incision into the peripheral cornea from the temporal side using a sharp 23-G needle, so that the beveled tip enters the anterior chamber but the needle does not go in all the way. A blunted 23-G needle attached to the tubing that leads to the manifold is then inserted into the eye, bevel down first. It is helpful to have an assistant gently inject through the needle to keep the anterior chamber inflated. Once the needle is in the eye, the bevel is rotated anterior to keep the tip from being blocked by iris. By making the initial incision slightly small, the seal of the blunted needle will be tight and will not leak. This can be maintained for a reasonably long time, and the pressure range from 10 to 50 can be sampled several times, although it is possible that higher pressures may cause these thin corneas to be even more thin, thus altering the fine calibration of the instrument. The correlations that result from these experiments demonstrate a linear relationship between actual IOP and the measured, Tonopen IOP (Fig. 4). As
Figure 4 Example of an actual Tonopen calibration curve. Tonometer readings at each IOP level are plotted against the actual IOP, as measured with a transducer connected to the anterior chamber. Regression of this relationship demonstrates a good fit to a straight line. This indicates that Tonopen readings can be reliably used to determine actual IOP.
Intraocular Pressure Elevation: Hypertonic Saline
43
with all tonometers in all eyes, we have found that the Tonopen readings are less than actual pressure at the higher end of this range.
VI. ASSESSING NERVE DAMAGE AND ITS RELATIONSHIP TO IOP LEVEL Optic nerve damage can be assessed using many techniques. We use a qualitative grading scale of damage determined by histological examination of a crosssection of the myelinated portion of the optic nerve. This is rapid, reproducible, easily taught to others, and correlates well with actual numbers of axons lost, as determined by careful ultrastructure analyses. This grading system is described in Table 3. When using this system to assess injury, four or five masked graders record their assessment of the grade, and the average grade is calculated as the score for that nerve. When using this system to compare the nerve injury grade to IOP level between two groups of animals, we determine the best fit line for one group (e.g., the control group) to generate a formula to predict each data point for both groups. The difference between observed and expected values for each data point are then calculated. The mean difference values for each group can then be compared by t-test to determine the significance of the differences. Alternatively, data can be linearized before statistical analysis. Sample size analyses using data like this indicate that group sizes of 20 will provide a test with sufficient power to detect a 10% difference in the number of optic nerve axons between groups.
Table 3 Injury Grade 1 2 3 4
5
Grading System for Assessing Extent of Optic Nerve Cross-Section Description Normal optic nerve morphology with, at most, only a few randomly scattered degenerating axons Densely staining, degenerating axons appear focally, with a few axonal swellings Numerous degenerating axons and axonal swellings appear to spread away from the focal area. Central damage tends to exceed that in the periphery Numerous degenerating and axonal swellings appear throughout the nerve, interspersed with apparently normal axons. Numbers of degenerating and normal axons appear to be approximately equivalent Degenerating axons and axonal swellings make up nearly the entire mass of the nerve, with scattered, apparently normal axons. Gliosis may appear in severe cases
44
Morrison et al.
VII. ADVANTAGES AND DISADVANTAGES The major advantages of the method discussed here is the low cost of purchasing and maintaining laboratory rats. This allows relatively large numbers of animals to be used for studies, both of injury mechanism as well as for testing neuroprotection effectiveness of new agents. In this manner, it is possible to minimize difficulties imposed by individual variation in response to elevated IOP that exists among animals. The equipment needed for the injections and for making the microneedle is standard and readily available. A second advantage of this method is that it clearly relies on obstructing aqueous outflow, which is also the primary mechanism by which elevated IOP occurs in glaucoma. This is supported by our studies of the aqueous outflow system of the rat eye, histologic documentation of angle closure in these specimens, and the increased fluctuation of IOP that we see in these animals [17]. In the latter respect, this model clearly mimics human glaucoma, in which increased IOP fluctuation is a characteristic finding in many patients. A third advantage of this model is that these animals are easy to work with and allow us to measure IOP while they are awake. This provides a very accurate assessment of the IOP experienced by the eye throughout a given experiment. This presents a distinct advantage over obtaining IOP from animals that are under the influence of general anesthetics, commonly employed for studies of primate glaucoma models. Fourth, experience to date strongly suggests that the optic nerve damage experienced by these animals results solely from the pressure elevation. This is supported by our observation of no optic nerve damage in eyes that received an injection but failed to develop elevated IOP. In addition, we have found a very strong correlation between the extent of IOP elevation and optic nerve damage. This validates the accuracy of our pressure measurements, as well as our method of determining optic nerve damage. In addition, we have found that specific cellular responses, such as message production for Thy-1 [21] and neurofilament protein are closely linked to the extent of pressure elevation. Finally, this model provides a unique opportunity to understand the sequence of cellular events that follows elevation of IOP. The entire knowledge of neuropathology and cell biology of CNS injury developed in other rat CNS models is now available to the vision researcher to understand the cell biology of pressure-induced optic nerve damage. This provides a powerful opportunity to develop directed, rational strategies of neuroprotection for patients with extensive glaucomatous optic nerve damage who have progressive loss of visual function despite maximally controlled IOP. As stated above, the primary difficulty with creating elevated IOP in these small eyes lies in the challenge of scarring the aqueous outflow pathways without damaging the immediately adjacent structures that are necessary for maintaining aqueous humor formation. Damage to the latter will often produce hypotony, rather than ocular hypertension.
Intraocular Pressure Elevation: Hypertonic Saline
45
A potential disadvantage, common to all methods of modeling glaucoma in rats, lies in the difficulty of obtaining accurate measurements of IOP. Due to fluctuations of IOP, a characteristic of all eyes with aqueous outflow obstruction, frequent daily measurements are essential. Measuring IOP in awake rats is an important skill that requires considerable experience. However, once this skill is acquired, we have found that measuring IOP in awake animals is faster and less traumatic for the animals than using general anesthetics. Disadvantages of our method in particular primarily lie in the surgical skill needed to successfully cannulate these small vessels for injection and the time and patience needed to learn to measure IOP in rats. However, with practice and careful attention to detail, most researchers should be able to acquire this skill. Several laboratories have already successfully reproduced this model. However, this success requires a strong commitment, and nearly all of the individuals learning this technique have included in a personal visit, often more than once, to our laboratory to gain first-hand knowledge of these skills.
VIII. CONCLUSIONS Our method of producing chronic elevation of IOP by episcleral vein injection of hypertonic saline is both reproducible and reliable. It combines the advantages of using laboratory rats with a simple, elegant method of scarring the aqueous humor outflow pathways, while minimizing the effects on other anterior segment structures. Our experience with this model demonstrates that the resulting pressure elevation has many similarities to that of human glaucoma, as does the pathology of the optic nerve damage. With careful assessment of IOP and the extent of optic nerve damage, it is possible to understand the relationship between pressure and optic nerve damage. This understanding will allow us to use this model to test the effectiveness of potential neuroprotective agents and improve our understanding of the mechanism of pressure-induced optic nerve damage.
REFERENCES 1.
2.
3.
The AGIS Investigators: The advanced glaucoma intervention study (AGIS): 7. The relationship between control of intraocular pressure and visual field deterioration. Am J Ophthalmol 2000; 130:429–440. The Collaborative Normal-Tension Glaucoma Study Group. The effectiveness of intraocular pressure reduction in the treatment of normal-tension glaucoma. Am J Ophthalmol 1998; 126:498–505. The Collaborative Normal-Tension Study Group. Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Am J Ophthalmol 1999; 126:487– 497.
46 4. 5.
6.
7.
8. 9. 10.
11. 12.
13. 14.
15. 16.
17. 18. 19. 20.
21.
Morrison et al. Gaasterland D, Kupfer C. Experimental glaucoma in the rhesus monkey. Investigative Ophthalmol 1974; 13:455–457. Quigley HA, Addicks EM. Chronic experimental glaucoma in primates: I: Production of elevated intraocular pressure by anterior chamber injection of autologous ghost red blood cells. Invest Ophthalmol Vis Sci 1980;19:126–136. Quigley HA, Addicks EM. Chronic experimental glaucoma in primates: II: Effect of extended intraocular pressure on optic nerve head and axonal transport. Invest Ophthalmol Vis Sci 1980; 19:137–152. Morrison JC, Moore CG, Deppmeier LMH, Gold BG, Meshul CK, Johnson EC. A rat model of chronic pressure-induced optic nerve damage. Exp Eye Res 1997; 63: 85–96. Morrison JC, Johnson EC, Cepurna W. Animal models in glaucoma research. Ophthalmic Practice 1998; 16:12–20. Morrison JC, Cepurna WO, Johnson EC. Modeling glaucomatous optic nerve damage. Ophthalmol Clin North Am 1999; 39:29–41. Morrison JC, Farrell SK, Johnson EC, Deppmeier LMH, Moore CG, Grossmann E. Structure and composition of the rodent lamina cribrosa. Exp Eye Res 1995; 60: 127–135. Morrison JC, Johnson EC, Funk R. The microvasculature of the rat optic nerve head. Invest Ophthalmol Vis Sci 1999; 40:1702–1709. Johnson EC, Morrison JC, Farrell SK, Deppmeier LMH, Moore CG, McGinty MR. The effect of chronically elevated intraocular pressure on the rat optic nerve head extracellular matrix. Exp Eye Res 1996; 62:663–674. Shareef SR, Garcia-Valenzuela E, Salierno A, Sharma S. Chronic ocular hypertension following episcleral venous occlusion in rats. Exp Eye Res 1995; 61:379–382. Schori H, Kipnis J, Yoles E, WoldeMussie E, Ruiz G, Wheeler LA, Schwartz M. Vaccination for protection of retinal ganglion cells against death from glutamate cytotoxicity and ocular hypertension: implications for glaucoma. Proc Natl Acad Sci USA 2001; 98:3398–3403. Morrison JC, Fraunfelder FW, Milne ST, Moore CG. Limbal microvasculature of the rat eye. Invest Ophthalmol Vis Sci 1995; 36:751–756. Jia L, Cepurna WO, Johnson EC, Morrison JC. Patterns of intraocular pressure elevation after aqueous humor outflow obstruction in rats. Invest Ophthalmol Vis Sci 2000; 41:1380–1385. Moore CG, Johnson EC, Morrison JC. Circadian rhythm of intraocular pressure in the rat. Curr Eye Res 1996; 15:185-191. Moore CG, Milne S, Morrison JC. Non-invasive measurement of rat IOP with the TonoPen. Invest Ophthalmol Vis Sci 1993; 34:363–369. Moore CG, Epley D, Milne ST, Morrison JC. Chronic non-invasive measurement of intraocular pressure in the rat eye. Curr Eye Res 1995; 14:711–717. Jia L, Cepurna WO, Johnson EC, Morrison JC. Effect of general anesthetics on IOP in rats with experimental aqueous outflow obstruction. Invest Ophthalmol Vis Sci 2000; 41:3415–3419. Schlamp CL, Johnson EC, Li Y, Morrison JC, Nickells RW. Changes in Thy1 gene expression associated with damaged retinal ganglion cells. Mol Vis 2001; 7:192–201.
5 Intraocular Pressure Elevation: Laser Photocoagulation of the Trabecular Meshwork B’Ann T. Gabelt, James N. Ver Hoeve, and Paul L. Kaufman University of Wisconsin–Madison Madison, Wisconsin, U.S.A.
I.
INTRODUCTION
Glaucomas occur widely throughout the animal kingdom. However, primary glaucomas usually involve a genetic predisposition and tend to occur most commonly in domesticated animal species (for review, see Ref. 1). Development of animal models with different forms of spontaneous glaucoma is time consuming and expensive. Experimental glaucoma was induced in animals as early as 1905 [2]. More current experimental animal models of glaucoma have been reviewed by Gelatt [3] and are only briefly summarized below. However, there are limited numbers of in vivo animal models of experimental glaucoma that are useful for evaluating the pathophysiology and potential therapy of human glaucomatous optic neuropathy. The eye and visual system of the macaque monkey more closely resemble that of the human, and the monkey model of ocular hypertension with its resulting optic neuropathy is generally acknowledged to best reflect the optic neurodegeneration associated with human glaucoma. This primate model will be reviewed in detail, including the application of current clinical and basic science technologies. Experimental glaucoma in nonhuman primates has permitted study of aqueous humor dynamics, glaucomatous changes in the visual pathways from the photoreceptors to the visual cortex, and anterior and posterior ocular segment pharmacologic effects. Nonhuman primate species used include cynomolgus monkeys (Macaca fascicularis), owl monkeys (Aotus trivirgatus), rhesus mon47
48
Gabelt et al.
keys (Macaca mulatta), and squirrel monkeys (Saimiri sciurea), but the most work has been done in cynomolgus and rhesus.
II. MONKEY MODELS A.
Trabecular Laser
1. Historical Most current studies are done with laser-induced glaucoma model in the rhesus and cynomolgus monkey, which was first described by Gaasterland and Kupfer in 1974 [4]. Ba´ra´ny, studying large numbers of vervet monkeys in Uganda, had noted elevated intraocular pressure (IOP) and glaucomatous cupping in an occasional monkey following apparent ocular trauma and postulated its cause as trauma-induced damage and malfunction of the trabecular meshwork (TM) (personal communication). Gaasterland and Kupfer reasoned correctly that controlled trauma and scarification confined to the TM could be induced by the application of high energy laser burns delivered by standard clinical methods, and could elevate IOP similar to human traumatic glaucoma, but without inducing other anterior segment abnormalities that could themselves impair vision and preclude evaluation of the posterior segment. 2. Method Laser photocoagulation of the TM is conducted in the anesthetized monkey. Typically, ketamine (10 mg/kg, I.M.) or ketamine plus diazepam (1 mg/kg, I.M.) or acepromazine (0.2–1 mg/kg, I.M.) are sufficient to minimize eye movements during the procedure. If additional short duration sedation is necessary, methohexital sodium (10 mg/kg, I.M.) can be added and will usually last about 1 h. Monkeys are placed prone on a wooden, plastic, or metal board with a post on which a custom-fabricated head holder is mounted. The head holder has a metal bar that fits in the monkey’s mouth, can be adjusted to hold the head securely, and can be further adjusted to position the eye for lasering. Alternatively, an assistant can simply hold the head in the proper position. Topical anesthetic is administered and a custom-fabricated [5] mirrored Goniolens (Ocular Instruments) filled with hydroxypropyl methylcellulose (Gonak, Akorn, Buffalo Grove, IL), is placed on the eye. The custom lenses have a smaller diameter lens to easily fit through the small palpebral fissure. Also, the curvature of the lens and angle of the mirrors are different due to the smaller eye. In some larger animals, a standard adult or pediatric Goldmann-type lens can be used or a canthotomy can be performed to facilitate lens placement. However, in smaller monkeys, the procedure is greatly aided by the smaller and properly constructed monkey-specific lenses.
Intraocular Pressure Elevation: Laser Photocoagulation Table 1
49
Laser Parameters for Producing Experimental Glaucoma in Monkeys
Laser Type Argon (e.g., Coherent, model 900) Red Diode (e.g., Iris Medical Oculight SLX) Green Diode (e.g., Iris Medical Oculight GL)
Power (mW)
Duration (ms)
Spot size (mm)
Number of burns
1500
500–1000
50
50–250
1250
500–1000
75
50–250
1000
500–1000
75
50–250
A standard clinical argon laser (e.g., Coherent, model 900), or portable red diode (Iris Medical Oculight SLX) or green diode (Iris Medical Oculight GL) may be used. Typical settings are as follows: argon—1500 mW power, 500– 1000 ms duration, 50 µm spot size, 50–250 burns; red diode—1250 mW, 500– 1000 ms, 75µm, 50–250 burns [6]; green diode—1000 mW, 500–1000 ms, 75 µm, 50–250 burns (Table 1). Contiguous burns are placed in the mid-TM over 180—270° of the circumference of the TM in each session. Care should be taken to avoid burn spread posteriorly over the ciliary muscle, as this will increase the post-treatment inflammatory reaction and prolong the post-treatment hypotony (see below). Scarification of the anterior ciliary muscle may obstruct uveoscleral outflow and result in greater IOP elevation. These factors are very variable and hard to control, and placing the center of the burn at the junction of the pigmented and nonpigmented portion of the TM seems like the best compromise. At least one quadrant is left untreated to avoid very high IOP rises. Immediately after laser treatment, moderate iridocyclitis occurs with resultant ocular hypotony, which usually resolves within 3–4 weeks. IOP will then usually return to normal, or rise above normal if the session was effective (Fig. 1). However, additional treatment sessions are usually necessary, and the number of sessions required and their effectiveness can vary greatly between monkeys. The intensity and location of subsequent sessions can nonetheless be titrated to the target IOP. Unless one is striving for a very high IOP, it is generally advisable never to treat the entire 360° circumference with contiguous burns at any one sitting. We always leave at least 90° untreated at each sitting, although the “untreated” quadrant may have been treated at a prior session. In our experience, it is usually necessary to treat the entire circumference at least once, however the sessions are split. If no sustained pressure rise is achieved after the first session and resolution of the posttreatment inflammation, we will again treat 270°, encompassing the previously untreated quadrant. Third and subsequent sessions are titrated according to the response and the desired target IOP. A final IOP of within 10 mmHg of target can usually be obtained by varying the treatment strategy. Only rarely does one
50
Gabelt et al.
Figure 1 IOP and its control following argon laser treatment of the TM in one eye (experimental glaucoma) of rhesus monkey 1035. This monkey was unusual in that only one laser session was required to elevate the IOP; usually two or three sessions are required. Topical Timoptic-XE 0.5% was given once daily or every other day where indicated to maintain IOP at the desired level. Lowering IOP pharmacologically (with Timoptic, Alphagan, Trusopt and PGF2α-1-isopropylester) for 1 week prior to sacrifice resulted in no change in the surprisingly modest disc cupping (experimental glaucoma C/D⫽0.4; Control C/D⫽0.2), thus suggesting that there was actual loss of tissue, rather than simply a pressure-induced mechanical backward bowing of the elastic lamina cribrosa. The axonal loss for experimental glaucoma was mild (0.28). Reproduced with permission. (From Ref. 10.)
encounter an animal in which the IOP never rises with multiple treatments. Typically, 2–3 treatment sessions are required to achieve a sustained IOP rise. Some animals may need to be re-treated after a sustained period of IOP elevation that is then followed by a gradual decrease in IOP. If IOP exceeds the desired level or if the monkey displays any signs of discomfort (usually when IOP exceeds 60 mmHg by Goldmann applanation tonometry), standard antiglaucoma medications can be administered once or twice a day. Monkeys can be trained to enter an “iron maiden” (a modified squeeze cage) that has been adapted to immobilize the conscious monkey while tilting it to a vertical position so that the eyelid may be retracted and the medications dropped onto the cornea. Alternatively, for uncooperative monkeys, 5–10 mg/kg ketamine I.M. can be given for sedation and the medications then administered.
Intraocular Pressure Elevation: Laser Photocoagulation
51
However, some monkeys may show a decrease in appetite with frequent anesthesia so this must be assessed on a case-by-case basis. The medications that have been used successfully alone or in combination include Timoptic-XE (0.5% timolol maleate in gel-forming vehicle, Merck & Co, West Point, PA); Alphagan (0.2% brimonidine tartrate, Allergan, Irvine, CA); Trusopt (2% dorzolamide hydrochloride, Merck); and PGF2α-1-isopropylester (2 µg in 5 µl saline, Caymen Chemical Co, Ann Arbor, MI) or Xalatan (0.005% latanoprost, Pharmacia Corp, Peapack, NJ). If necessary, acetazolamide (5 mg/kg, Ben Venue Laboratories, Bedford, OH) I.M. has been given once or twice daily. 3. Clinical IOP under ketamine anesthesia is monitored weekly or more frequently if medications are being implemented to target a specific pressure range. IOP is measured with a minified Goldmann (Haag-Streit, Ko¨niz, Switzerland) applanation tonometer [7]. Occasionally, these are backed up by measurements with a Tonopen XL (Mentor O&O, Norwell, MA) if corneal edema or neovascularization, or head and eye movements under ketamine anesthesia preclude readings with the Goldmann [8]. Others have also used pneumotonometry for IOP determinations [9]. In some cases, IOP can fluctuate greatly. Slit lamp biomicroscopy of the anterior and posterior segments, including stereoscopic optic disc evaluation with a fundus lens are performed once a month when IOP has stabilized. The size, shape, and pallor of the optic disc, the cup-to-disc ratio, and the retinal nerve fiber layer are evaluated. The pupil in the lasered eye usually becomes dilated, probably due to damaging the parasympathetic motor nerves to the iris by the laser treatments (energy spread to the anterior ciliary muscle through which the nerves travel to reach the iris). Iridolenticular adhesions may develop independently of corneal changes or duration of IOP elevation. Anterior and posterior synechiae are often observed. However, IOP may be elevated even if most or all of the angle remains open and there is no papillary block. The TM is invariably heavily pigmented. As the duration of IOP elevation becomes longer, some animals may develop corneal edema followed, in some cases, by neovascularization of the cornea. Cupping of the optic nerve head, with posterior bowing of the lamina cribrosa is typical [4]. Optic disc cupping occurs more rapidly in the monkey than in the human. For a given monkey, this is dependent on the IOP elevation and duration. However, results can fluctuate between monkeys. In order to accurately assess cupping due to loss of neural/glial tissue versus posterior bowing of the elastic monkey lamina, the IOP should be lowered. This can be done with the combinations of pharmacological agents mentioned above to control IOP. Alternatively, I.V. mannitol (1.5 g/kg) can be administered over a 30 min period. If
52
Gabelt et al.
IOP is still too high, give I.V. acetazolamide (5 mg/kg) and wait 30 min more. If IOP is still greater than 20 mmHg, give I.V. methohexital (5 mg/kg) or pentobarbital (10 mg/kg). If IOP after mannitol is still greater than 40 mmHg, give both acetazolamide and methohexital or pentobarbital. Fundus stereo photography (Topcon TRC 50IA fundus camera (Topcon America Corporation, Paramus, NJ) can be used to document the time course of the changes. An example of glaucomatous damage is shown (Fig. 2) after 4 months of elevated pressure where the entire optic disc surface is excavated; the disc margin is undermined; there is substantial peripapillary atrophy; and the retinal nerve fiber layer is substantially attenuated [10]. Other instrumentation used to image the optic nerve in humans produce results that are more difficult to interpret in the monkeys due to a lack of corrections for the smaller eye, steeper corneal curvature, uncompensated corneal birefringence, and poor ocular fixation. Confocal scanning laser ophthalmoscopy (TopSS Topographic Scanning System, Laser Diagnostic Technologies, Inc., San Diego, CA) [11,12] and Heidelberg retinal tomography (HRT; Heidelberg Engineering, Heidelberg, Germany) [13,14] can be used for optic disc and peripapillary retinal contour analysis. Scanning laser polarimeter (GDx, Laser Diagnostic Technologies, Inc., San Diego, CA) to assess retinal nerve fiber layer thickness may be used for generalized qualitative evaluation of differences between the eyes of a given monkey. Adaptations of the HRT and GDx are being made for more accurate and quantitative use in the monkey and even for rodents [15] (R. Weinreb, personal communication). Optical coherence tomography (OCT) is under study for evaluation of retinal nerve fiber layer in glaucoma in humans [16– 18] and may be applicable to monkeys. For all these photographic/imaging procedures, pupils are dilated with 2.5% phenylephrine HCl (Mydfrin, Alcon, Ft. Worth, TX) and 1% tropicamide (Mydriacyl, Alcon). Anesthesia for these procedures is ketamine (10 mg/kg, I.M.) ⫹ acepromazine (0.2–1 mg/kg I.M.), ⫹ methohexital sodium (15 mg/kg, I.M.) if needed to eliminate eye movements. 4. Perimetry Behavioral perimetry in monkeys shows the same intersubject variability in the effects of elevated IOP on visual field sensitivities that are common with hightension glaucoma or ocular hypertension patients [19]. Perimetry regimens with either white or monochromatic stimuli are not useful predictors of ganglion cell loss until a substantial proportion of cells have died. The variance in ganglion cell loss is large for mild defects, which would be diagnostic of early glaucoma, and for visual field locations near the fovea, where sensitivity losses occur relatively late in the disease process [20]. Monkeys with laser-induced glaucoma exhibit the same type of Humphrey visual field defects as do glaucomatous humans.
Intraocular Pressure Elevation: Laser Photocoagulation
53
Figure 2 (A) Fundus photographs of experimental glaucoma and control (B) eyes of a rhesus monkey at 4.5 months after unilateral laser-induced IOP elevation (experimental glaucoma ⫽ ⬃40 mmHg; control ⫽ ⬃20 mmHg). The entire experimental glaucoma disc surface is excavated, the disc margins are undermined (white arrowhead), there is substantial peripapillary atrophy (white arrow), and the retinal nerve fiber layer (asterisks in control) is substantially attenuated (white star). (C) Fundus photographs of ONT and of control (D) eyes of a rhesus monkey, 3.5 wk after transection, 1 week prior to sacrifice. Note normal retinal vasculature, absence of retinal or vitreous hemorrhage, and presence of pallor but absence of disc cupping and very early, mild attenuation of nerve fiber layer (absence of striations emanating from the temporal disc margin) as a result of ONT. (Reproduced with permission from Ref. 10.)
54
Gabelt et al.
Spectral sensitivity defects occur in experimental glaucoma similar to those found in patients with glaucoma. Elevated IOP resulted in short wavelength sensitivity losses. The optimum condition identifying the greatest short wavelength sensitivity reduction is a yellow background of moderate intensity. In the early stages of experimental glaucoma, the cone mechanisms and the rod mechanism typically showed decreased test and field sensitivities. In advanced stages of experimental glaucoma, the largest sensitivity losses are in the longer wavelength, red-green opponent mechanisms [21]. 5. Electrophysiology Objective, noninvasive electrophysiological measures of retinal ganglion cell (RGC) function may be of value for monitoring glaucomatous damage in humans and for providing comparable functional measures in experimental animal models. Ganzfeld ERG. The a- and b-waves of the full-field ERG reflect a massed electrical response from the entire retina that is dominated by photoreceptor and bipolar activity with little apparent contribution from the ganglion cell layer [22]. Most ERG studies of human patients with glaucoma found no correlation between visual loss and scotopic or photopic full-field ERG [23–26] perhaps due to variability in severity, disease progression, and treatment regimens. This is of less concern in laser-induced experimental glaucoma in nonhuman primates. In agreement with human investigations, several studies of experimental glaucoma have found no effect on the scotopic a- and b-waves of the traditional full-field ERG [27–31]. However, some recent investigations in human glaucoma patients do find alterations of certain features of the a- and b-waves [32–36] suggesting that layers of the retina distal to the ganglion cells may be involved in glaucoma. Support for outer retinal changes are recent studies showing swelling of cell bodies in the outer nuclear and outer plexiform layers in both human and experimental glaucoma [37]. Oscillatory Potentials. The dark-adapted full-field ERG elicited by a bright flash can be filtered to reveal time-locked high frequency wavelets riding on the a- and b-waves known as oscillatory potentials. Oscillatory potentials are thought to be generated within the inner plexiform layer [38]. Several studies have found that oscillatory potentials are reduced in human glaucoma [34,39,40]. Investigations of oscillatory potentials in nonhuman primate models of glaucoma are mixed [28,41]. Photopic Negative Response. Another feature of the full-field ERG that appears to be affected in glaucoma is a negative wave that follows the b-wave recorded under photopic conditions. In some studies the photopic negative re-
Intraocular Pressure Elevation: Laser Photocoagulation
55
sponse was found to be greatly reduced in nonhuman primate experimental glaucoma [31,42], and in human patients [26,43,44] whereas other investigators found no consistent changes in patients with advanced glaucoma [45]. Scotopic Threshold Response. The scotopic threshold response is the dark-adapted ERG to a very dim flash of light that first appears as a cornea negative wave and peaks approximately 200 ms after stimulation [46]. On the basis of pharmacologic manipulations, the scotopic threshold response is thought to reflect proximal, or inner retinal activity [46]. However, the scotopic threshold response is not abolished in humans with long-standing optic atrophy or in cats with optic nerve transection (ONT) [47], suggesting that a substantial part of the scotopic threshold response may be due to amacrine cell activity. The scotopic threshold response in human glaucoma patients is highly variable and generally unchanged in humans with glaucomatous field loss [48]. In contrast, the scotopic threshold response is greatly altered or abolished in nonhuman primate experimental glaucoma [28]. Frishman [28] suggests the scotopic threshold response may receive contributions from both rod amacrine (a short-latency component) and ganglion cells (a long-latency component). There appears to be marked species differences in the relative contribution of the two cell types. Pattern ERG. The pattern pattern ERG is a small amplitude ERG response to pattern “reversal” (e.g., exchange of the black and white checks of a chessboard pattern) or onset of a pattern in which there is no global change in luminance. It is not a full-field stimulus; rather the pattern ERG is the summed response from a large area of the central field. Many studies have shown that the pattern ERG reflects ganglion cell activation [22,49,50]. The pattern ERG is reduced in human diseases of the optic nerve [51], including glaucoma [52–63]. Changes in the pattern ERG also have been reported in ocular hypertension [57,58,64–70]; see Korth [71] for a review of the pattern ERG in glaucoma. The pattern ERG has also been examined in animal models of glaucoma. Reduction of pattern ERG amplitude is correlated with the degree of disc cupping in monkeys with chronic ocular hypertension [27,41]. The pattern ERG is not used extensively in the diagnosis and monitoring of disease progression, perhaps due to the technical difficulties with recording small amplitude signals and the finding that substantial peripheral loss must occur before the pattern ERG is affected [71]. In addition, some of the inconsistencies among these studies have been attributed to differences in the relative insensitivity of the full-field ERG to ganglion cell loss and the inability of the mass ERG response to reflect localized or patchy loss of function that is characteristic in glaucoma. Several novel electrophysiological techniques have been introduced that appear promising for detecting and monitoring glaucoma because they are thought to reflect inner retinal activity.
56
Gabelt et al.
Multifocal ERG (mERG). The mERG technique derives the local electrical response of many small patches of the retina (typically 103) using a sparse binary m-sequence cross correlation technique introduced by Sutter [72,73]. The technique has attracted considerable attention in the study of glaucoma for two reasons. First, it can image the patchy, localized areas of retinal dysfunction that characterize the glaucomatous visual field loss. Second, the binary m-sequence method permits a single recording to be analyzed in first-order (linear) and higherorder components, or “kernels.” Higher-order kernels reflect complex nonlinear dynamic responses that are presumed to originate in layers of the retina proximal to photoreceptors, possibly including ganglion cells. Chan [74] found reductions in the amplitudes of both the first- and secondorder kernels in humans with glaucoma. Delays in mERG waveforms also have been found in primary open angle glaucoma [75–77]. Amplitude differences were not evident in these studies. The relationship between the mERG and glaucomatous field losses as measured by standard automated perimetry has not been established [78–80]. The mERG has also been examined in experimental glaucoma. In normal eyes of rhesus monkeys, the first-order mERG response kernel contains more prominent oscillatory potentials (⬎ 60 Hz) than in comparable recordings in humans [81,82]. The rhesus macaque mERG oscillatory potentials are larger in central than in peripheral locations. The second-order kernel also contains larger oscillatory potentials than seen in human recordings. In addition, there are prominent naso-temporal variations in the oscillatory potentials of both kernels; responses near the optic nerve head have larger amplitude oscillatory potentials than temporal locations [81,82]. One approach to reversibly simulating glaucoma in nonhuman primates is to pharmacologically suppress sodium-based spiking activity of the inner retinal with tetrodotoxin and NMDA. Four effects on the mERG have been noted following intravitreal administration of tetrodotoxin and NMDA: (1) a marked increase in the amplitude of the mERG first and second order responses; (2) retinotopic changes, with the largest increased amplitudes occurring in the foveal region; and (3) a reduction of the prominent oscillatory potentials; and (4) removal of the naso-temporal variations. These findings differ from human glaucoma studies that found no effect or reduced amplitude foveal responses. In nonhuman primates with advanced experimental glaucoma, Frishman found the effects were similar to those produced by suppressing inner retinal activity with tetrodotoxin and NMDA [82]. In addition, the later portion of the first order kernel waveform was altered, lacking a dip after the large positive wave, similar to the changes seen in the photopic negative response. mERG changes increased over the time course of glaucoma and were more diffusely distributed across the visual field [82]. Another study of nonhuman primates with
Intraocular Pressure Elevation: Laser Photocoagulation
57
IOPs elevated for over 16 months, and histologically documented ganglion cell loss, found the amplitude of both first and second order mERG responses were attenuated and were highly correlated with RGC density (Fig. 3) [29,30]. A similar effect was found in nonhuman primates in the early stages of experimental glaucoma [83]. mERG after 4 weeks of elevated IOP (⬎25 mmHg) showed an attenuation of the negative waveform complex at 40–70 ms following the prominent positive (P1) wave. A similar alteration of these waves following ONT suggests this waveform feature of the mERG, to some extent, reflects ganglion cell activity [83]. Visual Evoked Potential (VEP). Amplitude reductions and delays in peak latency of the VEP elicited by patterned stimuli have been reported in human glaucoma [52,60,84–94]. In addition, distortion of waveforms, rendering them unscoreable using traditional metrics, were found in a high proportion of glaucoma patients [52,89]. The correlation with standard automated perimetry defects has been attempted but has yielded poor results [71,84]. Pattern VEP has been considered less sensitive than standard automated perimetry for the early detection of glaucoma. Recently, a multifocal VEP (mVEP) method has been introduced that appears to correlate well with field loss [80,95]. When all factors are taken into account, the mVEP appears to capture standard automated perimetry field losses with exceptional accuracy. Its role in early detection remains to be established. The mVEP has not been reported in studies of experimental glaucoma. 6. Aqueous Humor Dynamics Outflow facility can be measured by constant pressure perfusion [96], Schiotz tonography [97,98], pneumotonography [99] and by a fluorophotometric technique [9,100]. In vivo constant pressure outflow facility measurement decreases from prelaser baseline of 0.33 µL/min/mmHg–0.75 µL/min/mmHg, to 0.02 µL/ min/mmHg–0.11 µL/min/mmHg post-laser [4]. Toris also showed tonographic outflow facility was decreased by 71% at 36–75 days and fluorophotometric outflow facility was decreased by 63% at least 1.7 years later [9]. Outflow facility by constant pressure perfusion in a group of 16 cynomolgus monkeys at 1–5 months post laser, when IOP averaged 32.9 ⫾ 3.5 mmHg in lasered versus 17.0 ⫾ 1.2 mmHg in control eyes, was 0.049 ⫾ 0.012 µL/min/mmHg versus 0.424 ⫾ 0.038 µL/min/mmHg, respectively (Kiland J, Kaufman PL, unpublished data). Uveoscleral outflow measured with tracers or calculated was increased at least one year after laser treatment [9]. This could be partially due to the persistent low-grade inflammation that may be associated with chronic endogenous prostaglandin release. Also, artificial openings (small cyclodialysis clefts) between the scleral spur and the ciliary body may have resulted from the laser burns, which
58
Gabelt et al.
(1) Figure 3 Part 1. First order mERG responses from both eyes of a monkey with severe ocular hypertension. (A) Normotensive (OS) eye. Trace array of 61 responses, each of which represents the local retinal response corresponding to the stimulus element at that location in the stimulus field. The central seven traces (shaded) represent responses from macular retina extending to approximately 8° retinal eccentricity. The surrounding 54 traces (unshaded) represent responses from perimacular retina extending from approximately 8 to 25° retinal eccentricity. Calibration bars 200 nV, 100 ms. Note that amplitude of individual traces is expressed in units of volts because each trace represents the response from retinal areas of the same size. (B) Macular and perimacular responses obtained by summation of either the central seven or surrounding 54 response traces, respectively, from (A). Amplitude measures for the five peaks of the macular response were made as indicated. Calibration bars, 5 nV/degree squared (deg2) and 25 ms. Note that macular and perimacular response amplitude is expressed as response density (volts/unit retinal area), because these responses reflect retinal stimulus areas of different size. (C) Sixty-one response array obtained from the hypertensive (OD) eye of the same animal whose responses are shown in (A) and (B). Calibration as for (A). (D) Macular (top trace) and perimacular (bottom trace) responses obtained from (C); calibration as for (B). Part 2. Second-order mERG responses from the same recordings that produced the first-order responses shown in Part 1. (A) As for Part 1A. (B) As for Part 1B. A single peak-to-peak measure of second-order macular response amplitude was made as indicated. Calibration bars, 2.5 nV/deg2 and 25 ms. (C) As for Part 1C. (D) As for Part 1D.
Intraocular Pressure Elevation: Laser Photocoagulation
59
(2)
(3) Part 3. Comparison of macular and perimacular responses from the hypertensive (OD, light traces) and normotensive (OS, heavy traces) eyes shown in Parts 1 and 2. Histological analysis showed that normalized (OD/OS) perifoveal RGC density in the hypertensive eye was 0.11. First order response calibration bars, 5 nV/deg2 and 25 ms; second-order calibration bars, 2.5 nV/deg2 and 25 ms. (From Ref. 30.)
60
Gabelt et al.
could increase uveoscleral outflow [101]. Cyclodialysis clefts can also increase uveoscleral facility in monkeys [101]. The increased uveoscleral outflow in experimental glaucoma was not sufficient to prevent the IOP rise [9]. Apparent aqueous humor flow in lasered eyes measured by fluorophotometry significantly decreased by 46% at 36–75 days after laser at a time when IOP was increased by 17.0 ⫾ 9.3 mmHg [9]. If real, the aqueous flow decrease may have been caused by inflammation and damage to the ciliary processes [102]. Aqueous flow returned to normal when measured at least one year after laser treatment [9]. Accumulated protein in and around the ciliary body would likely not hinder egress of aqueous humor. If anything, it would slightly increase osmotic pressure, which would draw more water, although this effect would likely be small. A breakdown of the blood-aqueous barrier should also not affect aqueous humor production [103]. However, both increased protein concentration and blood–aqueous barrier breakdown could affect fluorophotometric measurement of flow. If fluorescein binds to albumin and is quenched, the fluorophotometer would not “see” as much as is actually present. Also, the increased scatter from macromolecules could increase the fluorescence signal (by multiple scatter of excitation light and fluorescence into the measurement window, and by some scattered excitation light passing the barrier filter, which is not 100% efficient) leading to an overestimate of fluorescein concentration. It is difficult to predict how both of these processes would affect the fluorophotometric estimate of flow rate. The flow estimate is proportional to the change in mass of fluorescein in the cornea and the anterior chamber divided by the mean concentration in the anterior chamber on an interval. If fluorescein concentration in the anterior chamber is over- or underestimated, flow rate would be under- or overestimated respectively. Thus, quenching could increase the estimate of flow, whereas scatter might decrease it. Breakdown of the blood–aqueous barrier would also allow more fluorescein to diffuse directly into the blood. One assumes that this loss normally accounts for 8% or less of fluorescein loss [104]. Diffusional loss in an eye with a compromised blood–aqueous barrier has not been measured, but it would likely be higher. This increased loss would lead to an overestimate of flow rate rather than an underestimate and lower flow, as Toris and Pederson observed. One might also expect that aqueous humor flow would decrease during inflammation. A diminished flow would increase concentration of macromolecules that respond to the insult and help repair the damage, and would slow the release of any toxins into the venous blood. Reduced flow would give the TM more time to break down and remove debris and invading substances. The increased protein concentration in the aqueous humor may reflect this lower flow rate as well as breakdown of the blood–aqueous barrier, although it would be difficult to measure either independently.
Intraocular Pressure Elevation: Laser Photocoagulation
61
Flow rate could be measured in an inflamed eye by using a large radioactive tracer, large enough not to cross the corneal endothelium or to leave the anterior chamber by diffusional pathways. Any fluorescence methods to measure flow would likely be difficult because of the increased scatter. 7. Blood Flow Blood flow in the retina, optic nerve head, and retrobulbar optic nerve measured with tritiated iodoantipyrine did not differ between control and glaucomatous eyes [105]. 8. Vitreous Sampling/Injections/Glutamate Visualization of the area of the vitreous for injections and for sampling can be accomplished by placing the monkey on its back, dilating the pupil, and placing a well made of tubing on the cornea and filling it with gonioscopic gel. Under microscopic visualization, a 23-G needle attached to a tuberculin syringe can be inserted and 0.1–0.2 mL of vitreous withdrawn from the desired location without producing hemorrhage. The pressure in the eye can be restored by then injecting viscoelastic material into the vitreous. Drugs of interest can be delivered to the vitreous with a similar set up but using a smaller (30-G) needle and injecting, usually, up to 50 µL of material. Vitreous glutamate was shown to be elevated in experimental glaucoma of 18–32 weeks duration to concentrations potentially toxic to RGCs [106]. Anterior and posterior vitreous levels were, respectively, 59.7 ⫾ 7.3 and 80.3 ⫾ 7.8 µmol/ L in experimental glaucoma and 12.3 ⫾ 1.5 and 12.3 ⫾ 2.3 µmol/L in control eyes. However, in a group of 20 cynomolgus and rhesus monkeys with laserinduced glaucoma from 3–51.7 weeks, glutamate levels in the posterior vitreous sampled from near the posterior pole were no different between the experimental glaucoma and contraleral normal control eyes (experimental glaucoma ⫽ 32.9 ⫾ 6.8 µM; control ⫽ 36.0 ⫾ 7.6 µM) (Kaufman, unpublished data). 9. Neuroprotection Memantine, a noncompetitive NMDA receptor antagonist (NMDA-type glutamatergic open-channel blocker), was administered orally to monkeys with experimental glaucoma for 15 months. Significantly less loss of visually evoked cortical potential amplitude was demonstrated in memantine-treated compared with untreated animals with experimental glaucoma [107]. Ganglion cell survival was enhanced in regions of the retina where the photoreceptors had been focally destroyed by retinal laser photocoagulation prior to the induction of experimental glaucoma in the monkey [108]. These findings demonstrate that RGCs can be protected from pressure-induced damage in the
62
Gabelt et al.
monkey and that the outer retina may be involved in the pathophysiology of glaucomatous optic neuropathy. The monkey is a good model for such studies. Morphological changes in the photoreceptors have been observed in experimental glaucoma in the monkey [37]. These changes are remarkably similar to what has been found in chronic human glaucoma [37]. In both cases, there is selective, ischemic-like swelling of the L/M-cones (red- and green-sensitive cones), perhaps due to decreased choroidal blood flow [109,110]. (The L/M-cones are numerically the dominant cone population since the S- or blue-sensitive cones make up only about 9% of all cones.) At the molecular level, preliminary studies have shown decreased production of mRNA that encodes for the L/M-cone opsins in both human glaucoma and experimental glaucoma in monkeys [111]. Rhodopsin mRNA levels, by contrast, remain relatively unaffected. Interestingly, the L/Mcones that have the greatest reduction in mRNA in monkeys may be those in the arcuate region. Independent evidence for outer retinal injury in human as well as experimental glaucoma has now been obtained using the mERG [75,112]. Delays in an early feature (N1) of the mERG seem to be most prominent in the arcuate region, as well. 10. IOP-Lowering Strategies The experimental glaucoma monkey has been used to investigate the efficacy and mechanism of clinical and potential glaucoma therapeutic agents. Timolol (0.5%), epinephrine (2%), pilocarpine (4%), vanadate (1%), PGF2α (500 µg), and forskolin (1%) were all found to significantly decrease IOP after single or multiple treatments [113,114]. Other prostaglandin derivatives have been tested extensively in this model [115,116], including some of the newest antiglaucoma therapies (e.g., latanoprost) [116,117]. The combined effects of brimonidine, dorzolamide, or latanoprost with timolol were all found to decrease IOP more than timolol alone [118]. Trabeculectomy surgery to construct a guarded fistula between the anterior chamber of the eye and the subconjunctival space is often complicated by the ability of the surrounding tissue to scar over the fistula, thus blocking the route of exit for the aqueous and reducing the ability of the treatment to lower IOP. Strategies to prevent wound healing and the scarring process can be studied in experimental glaucoma in monkeys. Typically, mitomycin C has been used in humans for this purpose, but recently studies in monkeys using a recombinant adenovirus carrying the coding region for human p21 has shown this to be just as effective without the other adverse effects often encountered with mitomycin C. 11. Histological Findings Light and electron microscopic examinations of the filtration angle treated by laser several months previously indicate “blunting” of the trabecular beams and
Intraocular Pressure Elevation: Laser Photocoagulation
63
Figure 4 (A) Anterior chamber angle after laser photocoagulation on two occasions. IOP, 40 mmHg. Noteworthy are scarring of TM and adjacent tissue (arrow), obliteration of canal of Schlemm, and pigment dispersion with melanin phagocytosis within iris root and TM. (B) Anterior chamber angle of control monkey eye. Normal trabecular area (small arrow) separates anterior chamber from canal of Schlemm (large arrow). (From Ref. 4.)
scattered peripheral anterior synechiae [119]. Scarring of the TM and obliteration of the canal of Schlemm is observed in treated eyes (Fig. 4) [4]. (Kaufman, Carassa, Albert, Tamm, unpublished data). In the retina, there is a selective loss of ganglion cells and thinning of the nerve fiber layer (Fig. 5) [4]. The large RGCs located in the midperiphery and fovea appear to be preferentially damaged in laser-induced experimental glaucoma [120,121]. In the optic nerve, optic nerve fibers larger than the mean axon diameter atrophy more rapidly in the glaucomatous eye [122]. RGCs in glaucomatous eyes undergo a pattern of degeneration that, morphologically, originates with the dendritic arbor of the cell. Both midget and parasol cells show signs of atrophy, although parasol cells appeared to be slightly more sensitive to the degenerative effects of prolonged IOP elevation [123]. Parafoveal cone loss was not found in experimental glaucoma in which there was extensive damage to RGCs [124]. Small RGCs that project to the parvocellular layers of the lateral geniculate nucleus belong to the P pathway or the color system. Large RGCs project to the magnocellular layers and belong to the M pathway or the luminance system. Rapid-phase axonal transport, measured by radioactive labeling, to the dorsal lateral geniculate body in monkeys with chronic experimental glaucoma is de-
64
Gabelt et al.
Figure 5 (A) Macular retina and optic nerve head of treated monkey eye from Figure 4A. There is selective loss of ganglion cells with thinning of the nerve fiber layer (small arrow) and cupping of the nervehead with posterior bowing of the lamina cribrosa (large arrow). Splitting of the thin retina at the juncture with optic nerve is an artifact. (B) Macular retina and optic nervehead of control monkey eye. A thick layer of ganglion cells is present next to the fovea (small arrow). There is no cupping of the nervehead; the lamina cribrosa goes straight across the anterior optic nerve (large arrow). (A and B at same low-power magnification.) (From Ref. 4.)
(1) Figure 6 Part 1. Primate retinogeniculate pathway showing the regions of the LGN examined and the approximate locations of their retinal inputs. Layers 1 and 2 are the M-layers, and layers 3 through 6 are the p-layers. Ganglion cells in nasal retina project to the contralateral LGN, whereas those in the temporal retina project ipsilaterally. The nasal region of the LGN receives its afferent input from ganglion cells in the superior retina (B, C), and the temporal region of the LGN is innervated by ganglion cells located in inferior retina (A, D). (From Ref. 126.)
Intraocular Pressure Elevation: Laser Photocoagulation
65
(2) Figure 6 (cont.) Part 2. Photomicrographs of cresyl violet-stained coronal sections from the right LGN of a normal rhesus monkey (A) and animals that had the pressure in one eye elevated for 2.5 (B), 8 (C), and 24 (D) weeks. In all cases, the nasal region of the nucleus is to the left, and the temporal region is to the right. Layers 1, 4, and 6 receive retinal input from the normal contralateral eye, whereas layers 2, 3, and 5 are innervated by the glaucomatous ipsilateral eye. Prolonged elevation of IOP resulted in a decrease in the size and Nissl substance within neurons receiving input from the glaucomatous eye, resulting in their pale appearance. Scale bar, 500 µm. (From Ref. 126.)
creased more in the magnocellular layers than in the parvocellular layers [125]. Unilateral elevation of IOP severely affects the size, density, and number of neurons in those LGN layers receiving input from the affected eye. One study of increased IOP for 2.5–27 weeks shows a greater degenerative effect on magnocellular than parvocellular regions [126]; while another study showed equal loss of neurons in the magnocellular and parvocellular layers with experimentally increased IOP for 14 months (Fig. 6) [127]. Significant loss of CaMKII-a immunoreactivity (indicating blue-on neurons) between and within the principal layers of the LGN occurs in all ocular hypertensive monkeys with or without significant
66
Gabelt et al.
(3) Figure 6 (cont.) Part 3. Low-power microphotographs of coronal sections of the left LGN from control (left) and glaucomatous (14 months) (right) cynomolgus monkeys, immunostained for parvalbumin. All six layers in the control are strongly immunoreactive for parvalbumin as indicated by numbers. There is overall shrinkage of the LGN and a decrease in immunoreactivity in parvocellular layers 4 and 6 in the glaucomatous LGN compared with control. The bar indicates 0.5 mm. (From Ref. 127.)
optic nerve fiber loss [128]. Changes in the distribution of certain neurochemicals occurs in the visual cortex of glaucomatous primates, suggesting cortical plasticity in recovery from glaucomatous visual damage [129]. Cytochrome oxidase reactivity in the neurons in the LGN was reduced to the same degree in both the P- and M-cellular layers with increasing severity of experimental glaucoma [130]. In eyes with IOP increased for 14–60 months, ganglion cells and nitrergic nerve fibers in the choroid are significantly reduced. Whole mount preparations of the retina stained for NADPH diaphorase revealed a significant reduction in positively stained amacrine cells, reduction in diameter of arterioles and changes in the staining pattern of the retinal vasculature, particularly in the perimacular region [131]. Vascular and glial changes of the retrolaminar optic nerve was studied in monkeys with elevated pressures for 1 to 4 years. Independently of axon loss, the number of capillaries remained constant or decreased only slightly. Some vessels, especially in the most severely damaged regions, were occluded.
Intraocular Pressure Elevation: Laser Photocoagulation
67
The density of glial cells increased. In glaucomatous optic nerves, the density of alphaB-crystallin and glial fibrillary acidic protein-positive cells significantly increased [132]. Optic nerve head changes have been evaluated in the laser-induced chronic glaucoma monkey model by several methods. Optic disc cupping occurs symmetrically in early stages of experimental glaucoma but occurs predominantly in the vertical axis in later stages. Spontaneous reduction of IOP led to a reversal of cupping, which was significantly less in later stages of glaucoma [133]. Cupping in the monkey may have both a compliance component and a true tissue loss component, analogous to the situation in human infants with congenital glaucoma [134]. Lowering the IOP and re-examination may be the only way to tell. When IOP is acutely lowered, there is significantly less anterior movement of optic nerve heads with larger deeper cups than those with smaller cups [135]. Glaucomatous nerve heads showed increased labeling for collagen type IV along the margins of beams in the lamina cribrosa, due to accumulation of basement membrane–like materials. Material in the pores of the laminar beams labeled with antibodies to collagen types I, II and IV. These changes were not seen in optic nerve transected eyes [136]. In glaucomatous nerve heads, there is a major disruption of the lamina cribrosa beam structure, including a decrease in collagen density [137,138]. The destruction of collagen beams accompanies an accumulation and enlargement of collagen-associated proteoglycan filaments. Accumulation of chondroitin sulfate proteoglycans was most evident. Prominent filamentous heparan sulfate/heparin proteoglycans were also noted in thickened astrocytic and vascular basal laminae [139]. Elastin is an important component of the ECM of the lamina cribrosa that provides resiliency and deformability to the tissue. The elastic properties of the lamina cribrosa are important for buffering the constant fluctuations in IOP [140]. Abnormal elastin synthesis by astrocytes of the lamina cribrosa in experimental glaucoma optic neuropathy was shown to be specific to elevated IOP and not secondary to axonal loss, as demonstrated by comparison with optic nerve transected eyes [10].
B. Other Monkey Models 1. Optic Nerve Transection Optic nerve transection (ONT) can be used to distinguish between elevated IOP– induced changes versus those due to the loss of axons that occurs in transection. Six animals underwent transection of the optic nerve in one eye, preserving the central vessels as verified by the absence of hemorrhage by indirect ophthalmoscopy at the completion of the transection and again several days later. Briefly, an oculoplastic surgeon performs a lateral orbitotomy [141] using pentobarbital
68
Gabelt et al.
(15 mg/kg I.V. or 35 mg/kg I.M.) or isofluorane (1.5–2% inhalation) anesthesia. The intraconal space is entered by gentle dissection between the lateral and superior recti muscles, under 2.5⫻-loupe magnification. A malleable retractor is used to gently retract the globe medially. At all times, pressure on the globe is kept as light as possible and pressure is released for a few seconds every 2 to 3 mins. Under visualization with an operating microscope, the optic nerve is exposed and a sickle knife is used to make a 3 mm linear incision in the dura parallel to the nerve, as far posteriorly as practical (at least 15 mm posterior to the globe) in order to avoid damage to the central retinal artery. Dural vessels and use of cautery are avoided. Neurosurgical angled fine scissors are then used to extend the incision posteriorly several millimeters. The scissors are then inserted within the dural sheath, and the nerve transected (two cuts each two-thirds through the nerve or complete transection) under direct visualization [10]. The nerve may be transected within the sheath, or the sheath and the nerve can be transected as one unit. The ciliary ganglion is used as a marker to verify that we are sufficiently far posterior. The transection is performed approximately 12 mm posterior to the globe in rhesus monkeys. The retina is then observed by direct and indirect ophthalmoscopy, to ensure that no central retinal artery occlusion occurred. The wounds are closed and the animals treated with systemic benzathine and procaine penicillin (30,000 U/kg, Phoenix Pharmaceutical, Inc., St. Joseph, MO) daily for 5 days and systemic methylprednisolone acetate (Depo-Medrol, 1 mg/kg I.M. Pharmacia Corp, Pepack, NJ) for 3 weeks, tapering to 0.1 mg/kg over the course of 7–10 days. An alternative to the prolonged depomedrol treatment is dexamethasone (1 mg/kg, I.M., Phoenix Pharmaceutical, Inc., St. Joseph, MO) administered for 3–5 days. Analgesia is provided with buprenorphine HCl (0.1 mg/kg, I.M., Abbott Labs, Chicago, IL) every 12 h for 3 days, if necessary. Disc pallor and loss/attenuation of the nerve fiber layer may be evident within one month (Fig. 2) although, in most cases, indirect ophthalmoscopy, slit lamp biomicroscopic funduscopy, and stereoscopic fundus photography detect no clear-cut changes in the fundus of eyes approximately 1 month after ONT compared with their presurgical baseline or with their contralateral control. Changes in the clinical appearance of the optic disc begin to be visible 5 weeks after experimental optic nerve trauma [142]. Consistent with ONT, an afferent pupillary defect develops in the ONT eye. The pupil in the transected eye was generally larger than in the control eye, and the consensual response to light was weak in some ONT eyes throughout the 4 weeks. The dilated pupil and absent or weak consensual response to light may have resulted from damage to the ciliary ganglion or the parasympathetic efferent fibers traveling with the ciliary nerves during the dissection, so that the iris sphincter muscle received reduced innervation. No systematic alteration in IOP is measured in ONT eyes, although transient elevations may occur, presumably related to orbital swelling and pressure on the globe, as can occur after orbitotomy in humans.
Intraocular Pressure Elevation: Laser Photocoagulation
69
As mentioned previously, ONT does not elicit responses in astrocytes of the optic nerve head nor alter the structure of the tissues as seen after experimental glaucoma [10]. 2. Endothelin Endothelin-1 delivered to the perineural region of the anterior optic nerve via osmotically driven minipumps has been used to study the effect of chronic vasoconstriction and reduced blood flow to the optic nerve. Optic nerve blood flow determined by colored microspheres after 7 days for endothelin-1 administration was significantly decreased [143]. Chronic ischemia for up to 2 to 6 months resulted in diffuse loss of axons without a change in the IOP [144]. Confocal scanning laser ophthalmoscopic topographic analysis 3–5 months after the onset of ischemia showed increased cup area, cup volume, and mean cup depth. mERG after 5 months showed functional changes consistent with inner-retinal damage [145]. 3. Steroids Early attempts to induce IOP elevation and outflow facility reduction by chronic administration of glucocorticosteroids in a nonhuman primate, analogous to steroid-induced ocular hypertension in the human, were not successful [146]. More recently, steroid-induced ocular hypertension has been induced for the first time in cynomolgus monkeys (Fig. 7) [147]. In this model, monkeys received topical ocular drops (10 µL) of 0.1% dexamethasone 3 times a day for 28 days; 45% (5 of 11) of monkeys tested developed and maintained ocular hypertension of greater than 5 mmHg, similar to the frequency in humans [148,149]. Discontinuation of dexamethasone treatment resulted in IOP returning to normal in 2 weeks. The response in responder monkeys was reproducible but nonresponder monkeys remained nonresponders. This study identified no statistically significant evidence for a link between myocilin mutations and steroid-induced ocular hypertension in both monkeys and humans. The use of sustained-release intraocular steroid implants is awaiting testing in monkeys as another possible mode for inducing ocular hypertension in this species. 4. Particles/Gels Intracameral injection of ghost red blood cells causes abrupt IOP elevations of 70–90 mmHg lasting 2 to 42 days [150,151]. IOP elevation of 45 mmHg to 70 mmHg for only 2 to 4 days could cause significant retinal ganglion and axonal degeneration. Cross-linked polyacrylamide (microgels), an altered form of the synthetic viscoelastic Orcolon, elevated IOP by 5–10 mmHg and decreased out-
70
Gabelt et al.
Figure 7 (A) IOP time course of cynomolgus monkeys during first DEX treatment. Eleven monkeys were treated with topical glucocorticoid (10 µL of 0.1% DEX) three times a day for 28 days. IOP measurements were recorded every 3 to 4 days during 4 weeks of glucocorticoid treatment and for an additional 2 weeks. An elevation of IOP in excess of 5 mmHg was observed in five monkeys that were considered steroid responders. (B) IOP time course during the second DEX treatment. After a 6-month washout period, 10 of the 11 monkeys were re-treated with the same regimen of DEX. Elevated IOP was observed in the same monkeys that had steroid-induced elevations in IOP during the first course of glucocorticoids. Filled circle—IOP of steroid responders at each time point; open circles—IOP of the nonresponding monkeys. Data are mean ⫾ SD. (From Ref. 147.)
Intraocular Pressure Elevation: Laser Photocoagulation
71
flow facility by 40–50% for 1–2 months following exchange of the anterior chamber with microgel solution [152]. This was due to accumulation of microgels in the cribriform meshwork and beneath the inner wall of Schlemm’s canal with no cellular alterations or inflammatory infiltrate. Repeated intracameral injections of 50–100 µL of sterile 10 µm latex microspheres has recently been used to maintain elevated IOP by TM obstruction for up to 3 years in rhesus monkeys (Fig. 8) [153]. This has the advantage of producing gradual, chronic elevations of IOP without the need for expensive
Figure 8 Histological analysis showing the initial localization of beads injected into the anterior chamber to the region of the TM (top) and their intense packing density following multiple injections (bottom). Beads distributed to the iris had no adverse effect on its function. (From Ref. 153.)
72
Gabelt et al.
ophthalmic equipment, without producing inflammation, and without compromising visibility of the optic disc. 5. Enzymes Injection of alpha chymotrypsin into the posterior chamber rapidly increases IOP and is also accompanied by iridocyclitis, lens displacement, ciliary body atrophy, anterior chamber angle deformation, and peripheral anterior synechiae [154– 156]. IOP elevation can persist for at least 6 months or longer.
III. CONCLUDING REMARKS Laser-induced ocular hypertension is currently the most commonly used and most extensively characterized model for experimental glaucoma in primates. It mimics the optic neuropathy affecting the entire visual pathway, analogous to the human disease, and also amplifies the IOP-lowering effect of drugs. It has thus provided, and will no doubt continue to provide, useful pathophysiological and therapeutic insights relevant to the human disease. Other promising primate models including bead injections and steroid administration await more comprehensive characterization to enhance our understanding of the situations in which they will be most useful.
ACKNOWLEDGMENTS We thank Drs. Jay McLaren and Richard Brubaker for their contributions to the aqueous humor formation section and Dr. T. Michael Nork for his contribution to the photoreceptor section.
REFERENCES 1. 2.
3. 4. 5.
Gelatt KN, Brooks DE, Samuelson DA. Comparative glaucomatology I: The spontaneous glaucomas. J Glauc 1998; 7:187–201. ¨ ber Blutgefasse des Auges bie Blaukom und U ¨ ber experimetnteles Bartels M. U Glaukom durch Versperrung von vorderen Blutbahnen. Z Augenheilkunde 1905; 103:258, 458. Gelatt KN, Brooks DE, Samuelson DA. Comparative glaucomatology II: The experimental galucomas. J Glauc 1998; 7:282–294. Gaasterland D, Kupfer C. Experimental glaucoma in the rhesus monkey. Invest Ophthalmol 1974; 13:455–457. Kaufman PL, Wallow IH. Minified diagnostic contact lenses for biomicroscopic
Intraocular Pressure Elevation: Laser Photocoagulation
6.
7. 8.
9.
10.
11.
12.
13.
14. 15.
16.
17.
18.
19. 20.
73
examination and photocoagulation of the anterior and posterior segment in small primates. Exp Eye Res 1985; 40:883–885. Wang RF, Schumer RA, Serle JB, Podos SM. A comparison of argon laser and diode laser photocoagulation of the trabecular meshwork to produce the glaucoma monkey model. J Glauc 1998; 7:45–49. Kaufman PL, Davis GE. “Minified” Goldmann applanating prism for tonometry in monkeys and humans. Arch Ophthalmol 1980; 98:542–546. Peterson JA, Kiland JA, Croft MA, Kaufman PL. Intraocular pressure measurement in cynomolgus monkeys: Tono-Pen versus manometry. Invest Ophthalmol Vis Sci 1996; 37:1197–1199. Toris CB, Zhan GL, Wang YL, Zhao J, McLaughlin MA, Camras CB, Yablonski ME. Aqueous humor dynamics in monkeys with laser-induced glaucoma. J Ocular Pharmacol Ther 2000; 16:19–27. Pena JDO, Agapova O, Gabelt BT, Levin LA, Lucarelli M, Kaufman PL, Hernandez MR. Increased elastin expression in astrocytes of the lamina cribrosa in response to elevated intraocular pressure. Invest Ophthalmol Vis Sci 2001; 42:2303– 2314. Geyer O, Michaeli-Cohen A, Silver DM, et al. Reproducibility of topographic measures of the glaucomatous optic nerve head. Br J Ophthalmol 1998; 82:14– 17. Burgoyne CF, Quigley HA, Thompson HW, Vitale S, Varma R. Early changes in optic disc compliance and surface position in experimental glaucoma. Ophthalmology 1995; 102:1800–1809. Chauhan BC, Le Blanc RP, McCormick TA, Rogers JB. Test-retest variability of topographic measurements with confocal scanning laser tomography in patients with glaucoma and control subjects. Am J Ophthalmol 1994; 118:9–15. Mikelberg FS, Wijsman K, Schulzer M. Reproducibility of topographic parameters obtained with the Heidelberg Retina Tomograph. J Glaucoma 1993; 2:101–103. Pan J, Tremblay F, Archibald ML, Kelly MEM, Chauhan BC. Correlation of intraocular pressure elevation with optic nerve head topography, retinal function and histopathology in a chronic pressure-related model of optic nerve damage. Invest Ophthalmol Vis Sci 2001; 42(ARVO Abstracts):S829(Abs nr 4451). Drexler W, Morgner U, Ghanta RK, Kartner FX, Schuman JS, Fujimoto JG. Ultrahigh-resolution ophthalmic cortical coherence tomography. Nat Med 2001; 7:502– 507. Schuman JS, Hee MR, Arya AV, Pedut-Kloizman T, Puliafito CA, Fujimoto JG, Swanson EA. Optical coherence tomography: a new tool for glaucoma diagnosis. Curr Opin Ophthalmol 1995; 6(2):89–95. Hoh ST, Greenfield DS, Mistlberger A, Liebmann JM, Ishikawa H, Ritch R. Optical coherence tomography and scanning laser polarimetry in normal, ocular hypertensive, and glaucomatous eyes. Am J Ophthalmol 2000; 129(2):129–135. Harwerth RS, Smith ELR, DeSantis L. Experimental glaucoma: perimetric field defects and intraocular pressure. J Glaucoma 1997; 6:390–401. Harwerth RS, Carter-Dawson L, Shen F, Smith ELr, Crawford ML. Ganglion cell losses underlying visual field defects from experimental glaucoma. Invest Ophthalmol Vis Sci 1999; 40:2242–2250.
74 21.
22. 23. 24. 25.
26.
27.
28.
29.
30.
31.
32. 33.
34.
35.
36.
Gabelt et al. Kalloniatis M, Harwerth RS, Smith ELR, DeSantis L. Color vision anamalies following experimental glaucoma in monkeys. Ophthalm Physiol Optics 1993; 13: 56–67. Maffei L, Fiorentini A, Bisti S, Hollander H. Pattern ERG in the monkey after section of the optic nerve. Exp Brain Res 1985; 59:423–425. Leydhecker G. The electroretinogram in glaucomatous eyes. Br J Ophthalmol 1950; 34:500–505. Francois J. L’electrographie dans le glaucoma. Acta Ophthalmol 1953; 31:205– 218. Holopigian K, Greenstein VC, Seiple W, Hood DC, Ritch R. Electrophysiologic assessment of photoreceptor function in patients with primary open-angle glaucoma. J Glaucoma 2000; 9:163–168. Viswanathan S, Frishman LJ, Robson JG, Walters JW. The photopic negative response of the flash electroretinogram in primary open angle glaucoma. Invest Ophthalmol Vis Sci 2001; 42:514–522. Marx MS, Podos SM, Bodis-Wollner I, et al. Flash and pattern electroretinograms in normal and laser-induced glaucomatous primate eyes. Invest Ophthalnol Vis Sci 1986; 27:378–386. Frishman LJ, shen FF, Du L, et al. The scotopic electroretinogram of macaque after retinal ganglion cell loss from experimental glaucoma. Invest Opthahlmol Vis Sci 1996; 37:125–141. Hare W, Ton H, Woldemussie E, Ruiz G, Feldmann B, Wijono M. Electrophysiological and histological measures of retinal injury in chronic ocular hypertensive monkeys. Eur J Ophthalmol 1999; 9(Suppl 1):S30–S33. Hare WA, Ton H, Ruiz G, Feldmann B, Wijono M, WoldeMussie E. Characterization of retinal injury using ERG measures obtained with both conventional and multifocal methods in chronic ocular hypertensive primates. Invest Ophthalmol Vis Sci 2001; 42:127–136. Viswanathan S, Frishman LJ, Robson JG, Harwerth RS, Smith ELR. The photopic negative response of the macaque electroretinogram: reduction by experimental glaucoma. Invest Ophthalmol Vis Sci 1999; 40:1124–1136. Fazio DT, Heckenlively JR, Martin DA, Crhristensen RE. The electroretinogram in advanced open-angle glaucoma. Doc Ophthalmol 1986; 63:45–54. Vaegan Graham SL, Goldberg I, Buckland L, Hollows FC. Flash and pattern electroretinogram changes with optic atrophy and glaucoma. Exp Eye Res 1995; 60: 697–706. Vaegan Graham SL, Goldberg I, Millar TJ. Selective reduction of oscillatory potentials and pattern electroretinograms after retinal ganglion cell damage by disease in humans or by kainic acid toxicity in cats. Doc Ophthalmol 1991; 77:237– 253. Velten IM, Korth M, Horn FK. The a-wave of the dark adapted electroretinogram in glaucomas: are photoreceptors affected? Br J Ophthalmol 2001; 85:397– 402. Velten IM, Horn FK, Korth M, Velten K. The b-wave of the dark adapted flash electroretinogram in patients with advanced asymmetrical glaucoma and normal subjects. Br J Ophthalmol 2001; 85:403–409.
Intraocular Pressure Elevation: Laser Photocoagulation 37.
38. 39. 40. 41.
42.
43.
44.
45. 46. 47. 48. 49. 50.
51.
52.
53. 54.
55.
75
Nork TM, Ver Hoeve JN, Poulsen GL, et al. Swelling and loss of photoreceptors in chronic human and experimental glaucomas. Arch Ophthalmol 2000; 118:235– 245. Ogden TE. The oscillatory waves of the primate electroretinogram. Vision Res 1973; 13:1059–1074. Gur M, Zeevi YY, Gielik M, Neumann E. Changes in the oscillatory potentials of the electroretinogram in glaucoma. Curr Eye Res 1987; 6:457–466. Wanger P, Persson HE. Pattern-reversal electroretinograms in unilateral glaucoma. Invest Ophthalmol Vis Sci 1983; 24:749–753. Johnson MA, Drum BA, Quigley HA, Sanchez RM, Dunkelberger GR. Patternevoked potentials and optic nerve fiber loss in monocular laser-induced glaucoma. Invest Ophthalmol Vis Sci 1989; 30:897–907. Viswanathan S, Frishman LJ, Robson JG. The uniform field and pattern ERG in macaques with experimental glaucoma: removal of spiking activity. Invest Ophthalmol Vis Sci 2000; 41:2797–2810. Colotto A, Falsini B, Salgarello T, Iarossi G, Galan ME, Scullica L. Photopic negative response of the human ERG: losses associated with glaucomatous damage. Invest Ophthalmol Vis Sci 2000; 41:2205–2211. Drasdo N, Adelbasi YH, Chiti Z, Mortlock KE, Morgan JE, North RV. The scone PHNR and pattern ERG in primary open angle glaucoma. Invest Ophthalmol Vis Sci, 2001; 42: p. 1266–1272. Cursiefen C, Korth M, Horn FK. The negative response of the flash electroretinogram in glaucoma. Doc Ophthalmol 2001; 103:1–12. Sieving PA, Nino C. Scotopic threshold response (STR) of the human electroretinogram. Invest Ophthalmol Vis Sci 1988; 29:1608–1614. Seiving PA. Retinal ganglion cell loss does not abolish the scotopic threshold response (STR) of the cat and human ERG. Clin Vis Sci 1991; 6:149–158. Korth M, Nguyen NX, Horn FK, Martus P. Scotopic threshold response and scotopic PII in glaucoma. Invest Ophthalmol Vis Sci 1994; 35:619–625. Maffei L, Fiorentini A. Electroretinographic response to alternating gratings after section of the optic nerve. Science 1981; 211:953–955. Baker C, Hess R, Olsen B, Zrenner E. Current source density analysis of linear and non-linear components of the primate electroretinogram. J Physiol 1988; 407: 155–176. Fiorentini A, Maffei L, Pirchio M, Spinelli D, Porciatti V. The ERG in response to alternating gratings in patients with diseases of the peripheral visual pathway. Invest Ophthalmol Vis Sci 1981; 21:490–493. Marx MS, Podos SM, Bodis-Wollner I, Lee PY, Wang RF, Severin C. Signs of early damage in glaucomatous monkey eyes: low spatial frequency losses in the pattern ERG and VEP. Exp Eye Res 1988; 46:173–184. Bach M, Hiss P, Rover J. Check-size specific changes of pattern electroretinogram in patients with early open-angle glaucoma. Doc Ophthalmol 1988; 69:315–322. Korth M, Horn FK, Storck B, Jonas J. The pattern-evoked electroretinogram (PERG): age-related alterations and changes in glaucoma. Graefes Arch Clin Exp Ophthalmol 1989; 227:123–130. Odom JV, Feghali JG, Jin Jc, Weinstein GW. Visual function deficits in glaucoma.
76
56.
57.
58.
59.
60.
61.
62.
63. 64.
65.
66.
67. 68. 69. 70. 71. 72.
Gabelt et al. Electroretinogram pattern and luminance nonlinearities. Arch Ophthalmol 1990; 108:222–227. Ringens PJ, Vijfvinkel-Bruinenga S, van Lith GH. The pattern-elicited electroretinogram. I. A tool in the early detection of glaucoma? Ophthalmologica 1986; 192: 171–175. Weinstein GW, Arden GB, Hitchings RA, Ryan S, Calthorpe CM, Odom JV. The pattern electroretinogram (PERG) in ocular hypertension and glaucoma. Arch Ophthalmol 1988; 106:923–928. Ruben ST, Arden GB, O’Sullivan F, Hitchings RA. Pattern electroretinogram and peripheral colour contrast thresholds in ocular hypertension and glaucoma: comparison and correlation of results. Br J Ophthalmol 1995; 79:326–331. Graham SL, Wong VA, Drance SM, Mikelberg FS. Pattern electroretinograms from hemifields in normal subjects and patients with glaucoma. Invest Ophthalmol Vis Sci 1994; 35:3347–3356. Bobak P, Bodis-Wollner I, Harnois C, Maffei L, Mylin L, Podos SM, Thornton J. Pattern electroretinograms and visual-evoked potentials in glaucoma and multiple sclerosis. Am J Ophthalmol 1983; 96:72–83. Trick GL. Retinal potentials in patients with primary open angle glaucoma: physiological evidence for temporal frequency tuning deficits. Invest Ophthalmol Vis Sci 1985; 26:1750–1758. Trick GL. Pattern electroretinogram: an electrophysiological technique applicable to primary open-angle glaucoma and ocular hypertension. J Glauc 1992; 1:271– 279. Bach M, Pfieffer N, Birkner-Binder D. Pattern-electroretinogram reflects diffuse retinal damage in early glaucoma. Clin Vis Sci 1992; 7:335–340. Porciatti V, Falsini B, Brunori S, Colotto A, Moretti G. Pattern electroretinogram as a function of spatial frequency in ocular hypertension and early glaucoma. Doc Ophthalmol 1987; 65:349–355. Pfeiffer N, Bach M. The pattern-electroretinogram in glaucoma and ocular hypertension: a cross-sectional and longitudinal study. Ger J Ophthalmol 1992; 1:35– 40. Neoh C, Kaye SB, Brown M, Ansons AM, Wishart P. Pattern electroretinogram and automated perimetry in patients with glaucoma and ocular hypertension. Br J Ophthalmol 1994; 78:359–362. Bach M, Speidel-Faux A. Pattern electroretinogram in glaucoma and ocular hypertension. Doc Ophthalmol 1989; 73:173–181. Wanger P, Persson HE. Pattern-reversal electroretinograms in ocular hypertension. Doc Ophthalmol 1985; 61:27–31. Trick GL. PRRP abnormalities in glaucoma and ocular hypertension. Invest Ophthalmol Vis Sci 1986; 27:271–279. Ambrosio G, Areinzo G, Aurilla P, Colasanti A, Fusco R. Pattern electroretinograms in ocular hypertension. Doc Ophthalmol 1988; 69:161–165. Korth M. The value of electrophysiological testing in glaucomatous diseases. J Glaucoma 1997; 6:331–343. Sutter EE, Tran D. The field topography of ERG components in man I. The photopic luminance response. Vision Res 1992; 32:433–446.
Intraocular Pressure Elevation: Laser Photocoagulation 73. 74. 75.
76.
77.
78.
79. 80. 81.
82.
83.
84. 85.
86.
87. 88.
89.
77
Sutter EE. The fast-m transform: a fast computation of cross-correlations with binary m-sequences. SIAM J Comput 1991; 20:686–694. Chan HL, Brown B. Multifocal ERG changes in glaucoma. Ophthalmol Physiol Opt 1999; 19:306–316. Ver Hoeve JN, Murdock TJ, Heatley GA, Nork TM. Increased latency of early multifocal ERG response in glaucoma. Invest Ophthalmol Vis Sci 2000; 41(ARVO Abstracts):S520(Abs nr 2772). Palmowski AM, Allgayer R, Heinemann-Vemaleken B. The multifocal ERG in open angle glaucoma–a comparison of high and low contrast recordings in highand low-tension open angle glaucoma. Doc Ophthalmol 2000; 101:35–49. Hasegawa S, Takagi M, Usui T, Takada R, Abe H. Waveform changes of the firstorder multifocal electroretinogram in patients with glaucoma. Invest Ophthalmol Vis Sci 2000; 41:1597–1603. Fortune B, Johnson CA, Cioffi GA. The topographic relationship between multifocal electroretinographic and behavioral perimetric measures of function in glaucoma. Opton Vis Sci 2001; 78:206–214. Sutter EE, Bearse MAJ. The optic nerve head component of the human ERG. Vision Res 1999; 39:419–436. Hood DC, Zhang X. Multifocal ERG and VEP responses and visual fields: comparing disease-related changes. Doc Ophthalmol 2000; 100:115–137. Hood DC, Frishnman LJ, Viswanathan S, Robson JG, Ahmed J. Evidence for a ganglion cell contribution to the primate electroretinogram (ERG): effects of TTX on the multifocal ERG in macaque. Vis Neurosci 1999; 16:411–416. Frishman LJ, Saszik S, Harwerth RS, Viswanathan S, Li Y, Smith ELR, Robson JG, Barnes G. Effects of experimental glaucoma in macaques on the multifocal ERG. Multifocal ERG in laser-induced glaucoma. Doc Ophthalmol 2000; 100: 231–251. Kim CBY, Ver Hoeve JN, Nickells RW, Kaufman PL, Heatley GA, Kiland JA, Nork TM. Multifocal ERG in experimental glaucoma: relation to optic nerve transection. Invest Ophthalmol Vis Sci 2001; 42(ARVO Abstracts):S718(Abs nr 3864). Cappin JM, Nissim S. Visual evoked responses in the assessment of field defects in glaucoma. Arch Ophthalmol 1975; 93:9–18. Atkin A, Bodis-Wollner I, Podos SM, Wolkstein M, Mylin L, Nitzberg S. Flicker threshold and pattern VEP latency in ocular hypertension and glaucoma. Invest Ophthalmol Vis Sci 1983; 24:1524–1528. Towle VL, Moskowitz A, Sokol S, Schwartz B. The visual evoked potential in glaucoma and ocular hypertension: effects of check size, field size, and stimulation rate. Invest Ophthalmol Vis Sci 1983; 24:175–183. Watts MT, Good PA, O’Neill EC. The flash stimulated VEP in the diagnosis of glaucoma. Eye 1989; 3:732–737. Holopigian K, Seiple W, Mayron C, Koty R, Lorenzo M. Electrophysiological and psychophysical flicker sensitivity in patients with primary open-angle glaucoma and ocular hypertension. Invest Ophthalmol Vis Sci 1990; 31:1863–1868. Bray LC, Mitchell KW, Howe JW. Prognostic significance of the pattern visual evoked potential in ocular hypertension. Br J Ophthalmol 1991; 75:79–83.
78 90. 91. 92.
93. 94. 95. 96.
97. 98.
99. 100.
101.
102. 103.
104. 105. 106.
107.
108.
Gabelt et al. Galloway NR, Barber C. The transient pattern onset VEP in glaucoma. Ophthalmol (Proc Series) 1981; 27:95–101. Howe JW, Mitchell KW. Electrophysiologically determined contrast sensitivity in patients with ocular hypertension. Doc Ophthalmol 1992; 80:31–41. Parisi V, Bucci M. Visual evoked potentials after photostress in patients with primary open angle glaucoma and ocular hypertension. Invest Ophthalmol Vis Sci 1992; 33:436–442. Ermers HJM, De Heer LJ, Van Lith GHM. VECPs in patients with glaucoma. Doc Ophthalmol (Proc Series) 1974; 4:387–393. Schmeisser ET, Smith TJ. High-frequency flicker visual-evoked potential losses in glaucoma. Ophthalmology 1989; 96:620–623. Klistorner A, Graham SL. Objective perimetry in glaucoma. Ophthalmology 2000; 107:2283–2299. Ba´ra´ny EH. Simultaneous measurements of changing intraocular pressure and outflow facility in the vervet monkey by constant pressure infusion. Invest Ophthalmol 1964; 3:135–143. Kolker AE, Hetherington JJ. Outflow facility, in Becker-Shaffer’s Diagnosis and Therapy of the Glaucomas. St Louis: C.V. Mosby; 1983; 68–83. Croft MA, Oyen MJ, Gange SJ, Fisher MR, Kaufman PL. Aging effects on accommodation and outflow facility responses to pilocarpine in humans. Arch Ophthalmol 1996; 114:586–592. Langham ME, Leydhecker W, Krieglstein G, Waller W. Pneumatonographic studies on normal and glaucomatus eyes. Adv Ophthalmol 1976; 32:108–133. Yablonski ME, Cook DJ, Gray J. A fluorophotometric study of the effect of argon laser trabeculoplasty on aqueous humor dynamics. Am J Ophthalmol 1985; 99: 579–582. Suguro K, Toris CB, Pederson JE. Uveoscleral outflow following cyclodialysis in the monkey eye using a fluorescent tracer. Invest Ophthalmol Vis Sci 1985; 26: 810–813. Toris CB, Pederson JE. Aqueous humor dynamics in experimental iridocyclitis. Invest Ophthalmol Vis Sci 1987; 28:477–481. Brubaker RF. Measurement of aqueous flow by fluorophotometry. In: Ritch R, Shields MB, Krupin T, eds. The Glaucomas—Basic Sciences. St. Louis: Mosby; 1996; 447–454. McLaren JW, Ziai M, Brubaker RF. A simple three-compartment model of anterior segment kinetics. Exp Eye Res 1993; 56:355–366. Quigley HA, Hobman RM, Sanchez R, Addicks EM. Optic nerve head blood flow in chronic experimental glaucoma. Arch Ophthalmol 1985; 103:956–962. Dreyer EB, Zurakowski D, Schumer RA, Podos SM, Lipton SA. Elevated glutamate levels in the vitreous body of humans and monkeys with glaucoma. Arch Ophthalmol 1996; 114:299–305. Hare W, WoldeMussie E, Lai R, Ton H, Ruiz G, Feldmann B, Wijono M, Chun T, Wheeler L. Efficacy and safety of memantine, an NMDA-type open-channel blocker, for reduction of retinal injury asociated with experimental glaucoma in rat and monkey. Surv Ophthalmol 2001; 45:S284–S289. Nork TM, Poulsen GL, Nickells RW, Ver Hoeve JN, Cho NC, Levin LA, Lucarelli
Intraocular Pressure Elevation: Laser Photocoagulation
109. 110.
111.
112.
113.
114.
115.
116.
117.
118.
119. 120. 121.
122.
123. 124.
79
MJ. Protection of ganglion cells in experimental glaucoma by retinal laser photocoagulation. Arch Ophthalmol 2000; 118:1242–1250. Alm A, Bill A. Blood flow and oxygen extraction in the cat uvea at normal and high intraocular pressures. Acta Physiol Scand 1970; 80:19–28. Yancey CM, Linsenmeier RA. Oxygen distribution and consumption in the cat retina at increased intraocular pressure. Invest Ophthalmol Vis Sci 1989; 30:600– 611. Poulsen GL, Schlamp CL, Nork TM, Nickells RW. Changes in photoreceptor gene expression in glaucoma. Invest Ophthalmol Vis Sci 1999; 40(ARVO Abstracts)(4): S788(Abs nr 4148). Ver Hoeve JN, Kim CBY, Heatley GA, Kaufman PL, Kiland JA, Nickells RW, Nork TM. Delay of an early multifocal ERG wave feature in experimental glaucoma. Invest Ophthalmol Vis Sci 2001; 42:S148(Abs nr 789). Lee P-Y, Podos SM, Howard-Williams JR, Severin CH, Rose AD, Siegel MJ. Pharmacological testing in the laser-induced monkey glaucoma model. Curr Eye Res 1985; 4(7):775–781. Lee P-Y, Podos SM, Serle JB, Camras CB, Severin CH. Intraocular pressure effects of multiple doses of drugs applied to glaucomatous monkey eyes. Arch Ophthalmol 1987; 105:249–252. Wang RF, Camras CB, Lee PY, Podos SM, Bito LZ. Effects of prostaglandins F2 alpha, A2, and their esters in glaucomatous monkey eyes. Invest Ophthalmol Vis Sci 1990; 31:2466–2470. Wang RF, Podos SM, Serle JB, Mittag TW, Ventosa F, Becker B. Effects of latanoprost or 8-iso prostaglandin E2 alone and in combination on intraocular pressure in glaucomatous monkey eyes. Arch Ophthalmol 2000; 118:74–77. Serle JB, Podos SM, Kitazawa Y, Wang R-F. A camparative study of latanoprost (Xalatan) and isopropyl unoporostone (Rescula) in normal and glaucomatous monkey eyes. Jpn J Ophthalmol 1998; 42:95–100. Wang RF, Serle JB, Gagliuso DJ, Podos SM. Comparison of the ocular hypotensive effect of brimonidine, dorzolamide, latanoprost, or artificial tears added to timolol in glaucomatous monkey eyes. J Glauc 2000; 9:458–462. Radius RL, Pederson JE. Laser-induced primate glaucoma II. histopathology. Arch Ophthalmol 1984; 102:1693–1698. Glovinsky Y, Quigley HA, Pease ME. Foveal ganglion cell loss is size dependent in experimental glaucoma. Invest Ophthalmol Vis Sci 1993; 32:395–400. Glovinsky Y, Quigley HA, Dunkelberger GR. Retinal ganglion cell loss is size dependent in experimental glaucoma. Invest Ophthalmol Vis Sci 1991; 32:484– 491. Quigley HA, Sanchez RM, Dunkelberger GR, L’Hernault NL, Baginski TA. Chronic glaucoma selectively damages large optic nerve fibers. Invest Ophthalmol Vis Sci 1987; 28:913–920. Weber AJ, Kaufman PL, Hubbard WC. Morphology of single ganglion cells in the glaucomatous primate retina. Invest Ophthalmol Vis Sci 1998; 39:2304–2320. Wygnanski T, Desatnik H, Quigley HA, Glovinsky Y. Comparison of ganglion cell loss and cone loss in experimental glaucoma. Am J Ophthalmol 1995; 120: 184–189.
80 125.
126.
127.
128.
129.
130.
131.
132.
133.
134. 135.
136.
137.
138.
139.
Gabelt et al. Dandona L, Hendrickson A, Quigley HA. Selective effects of experimental glaucoma on axonal transport by retinal ganglion cells to the dorsal lateral geniculate nucleus. Invest Ophthalmol Vis Sci 1991; 32:1593–1599. Weber AJ, Chen H, Hubbard WC, Kaufman PL. Experimental glaucoma and cell size, density, and number in the primate dorsal lateral geniculate nucleus. Invest Ophthalmol Vis Sci 2000; 41:1370–1379. Yu¨cel YH, Zhang Q, Gupta N, Kaufman PL, Weinreb RN. Loss of neurons in magno-and parvocellular layers of the lateral geniculate nucleus in glaucoma. Arch Ophthalmol 2000; 118:378–384. Yu¨cel YH, Zhang Q, Weinreb RB, Kaufman PL, Gupta N. Atrophy of relay neurons in magnocellular and parvocellular layers in the lateral geniculate nucleus in glaucoma. Invest Ophthalmol Vis Sci 2001; 42:3216–3222. Lam DY, Kaufman PL, Gabelt BT, Matsubara JA. Cortical consequences of elevated intraocular pressure in a primate model of glaucoma. Invest Ophthalmol Vis Sci 2000; 41(ARVO Abstracts):S831(Abs nr 4413). Crawford ML, Harwerth RS, Smith ELR, Shen F, Carter-Dawson L. Galucoma in primates: cytochrome oxidase reactivity in parvo- and magnocellular pathways. Invest Ophthalmol Vis Sci 2000; 41(7):1791–1802. May CA, Hayreh SS, Ossoining K, Kaufman PL, Lu¨tjen-Drecoll E. Choroidal and retinal vasculature and its innervation in laser-induced glaucoma in rhesus monkeys. Ophthalmologica 1997; 211:161–171. Furuyoshi N, Furuyoshi M, May CA, Hayreh SS, Alm A, Lu¨tjen-Drecoll E. Vascular and glial changes in the retrolaminar optic nerve in glaucomatous monkey eyes. Ophthalmologica 2000; 214:24–32. Shirakashi, M, Abe, H, Sawaguchi, S, and Iwata, K, The relationship between deterioration and reversal of optic disc cupping in monkeys with chronic experimental high-pressure glaucoma. Graefes Arch Clin Exp Ophthalmol 1998; 236(7): 546–552. Quigley HA. The pathogenesis of reversible cupping in congenital glaucoma. Am J Ophthalmol 1977; 84(3):358–370. Coleman AL, Quigley HA, Vitale S, Dunkelberger G. Displacement of the optic nerve head by acute changes in intraocular pressure in monkey eyes. Ophthalmology 1991; 98:35–40. Morrison JC, Dorman-Pease ME, Dunkelberger GR, Quigley HA. Optic nerve head extracellular matrix in primary optic atrophy and experimental glaucoma. Arch Ophthalmol 1990; 108:1020–1024. Quigley HA, Doran-Pease ME, Brown AE. Quantitative study of collagen and elastin of the optic nerve head and sclera in human and experimental monkey glaucoma. Curr Eye Res 1991; 10:877–888. Sawaguchi S, Yue BY, Fukuchi T, Abe H, Suda K, Kaiya T, Iwata K. Collagen fibrillar network in the optic nerve head of normal monkey eyes and monkey eyes with laser-induced glaucoma—a scanning electron microscopic study. Curr Eye Res 1999; 18:143–149. Fukuchi T, Sawaguchi S, Yue BY, Iwata K, Hara H, Kaiya T. Sulfated proteoglycans in the lamina cribrosa of normal monkey eyes and monkey eyes with laserinduced glaucoma. Exp Eye Res 1994; 58:231–243.
Intraocular Pressure Elevation: Laser Photocoagulation 140.
141.
142. 143. 144. 145.
146. 147.
148. 149. 150.
151. 152.
153. 154. 155. 156. 157.
81
Zeimer RC. Circadian variations in intraocular pressure. In: Ritch R, Shields MB, Krupin T, eds The Glaucomas: Basic Sciences. 2d ed. St Louis: Mosby: 1996; 429–445. Gonnering RS, Dortzbach RK, Erickson KA, Kaufman PL. The cynomolgus monkey as a model for orbital research. II. Anatomic effects of lateral orbitotomy. Curr Eye Res 1984; 3:541–555. Anderson DR. Ascending and descending optic atrophy produced experimentally in squirrel monkeys. Am J Ophthalmol 1973; 76:693–711. Orgul S, Cioffi GA, Bacon DR, Van Buskirk EM. An endothelin-1-induced model of chronic optic nerve ischemia in rhesus monkeys. J Glaucoma 1996; 5:135–138. Cioffi GA, Sullivan P. The effect of chronic ischemia on the primate optic nerve. Eur J Ophthalmol 1999; 9:S34–S36. Wang L, Cioffi GA, Hernandez MR, Fortune B, Cull G, Oguri A, Van Buskirk EM. Topographic, functional and histological changes of the optic nerve in a primate chronic ischemia model. Invest Ophthalmol Vis Sci 2000; 41(ARVO Abstracts):S898(Abs nr 4778). Armaly MG. Aqueous outflow facility in monkeys and the effect of topical corticoids. Invest Ophthalmol 1964; 3:534–538. Fingert JH, Clark AF, Craig JE, Alward WLM, Snibson GR, McLaughlin M, Tuttle L, Mackey DA, Sheffield VC, Stone EM. Evaluation of the myocilin (MYOC) glaucoma gene in monkey and human steroid-induced ocular hypertension. Invest Ophthalmol Vis Sci 2001; 42:145–152. Armaly MF. Statistical attributes of the steroid hypertensive response in the clinically normal eye. Invest Ophthalmol Vis Sci 1965; 4:187–197. Becker B. Intraocular pressure response to topical corticosteroids. Invest Ophthalmol Vis Sci 1965; 4:198–205. Quigley HA, Addicks EM. Chronic experimental glaucoma in primates I. Production of elevated intraocular pressure by anterior chamber injection of autologous ghost red blood cells. Invest Opthalmol Vis Sci 1980; 19:126–136. Hayreh SS, March W. Pathogenesis of block of rapid orthograde axonal transprot by elevated intraocular pressure. Exp Eye Res 1979; 28:515–523. Kaufman PL, Lu¨tjen-Drecoll E, Hubbard WC, Erickson KA. Obstruction of aqueous humor outflow by cross-linked polyacrylamide microgels in bovine, monkey and human eyes. Ophthalmology 1994; 101:1672–1679. Weber AJ, Zelenak D. Experimental glaucoma in the primate induced by latex microspheres. J Neurosci Meth 2001; 111:39–48. Kalvin NH, Hamasaki DI, Gass JDM. Experimental glaucoma in monkeys. Arch Ophthalmol 1966; 76:82–103. Zimmerman LE, de Venecia G, Hamasaki DI. Pathology of the optic nerve in experimental acute glaucoma. Invest Ophthalmol 1967; 6:109–125. Lampert PW, Vogel MH, Zimmerman LE. Pathology of the optic nerve in experimental acute glaucoma. Invest Ophthalmol Vis Sci 1978; 7:199–213. Vareilles P, Silverstone D, Plazonnet B, Le Douarec JC, Sears ML, Stone CA. Comparison of the effects of timolol and other adrenergic agents on intraocular pressure in the rabbit. Invest Ophthalmol 1977; 16:987–996.
82
Gabelt et al.
BIBLIOGRAPHY FOR EXPERIMENTAL GLAUCOMA MODELS IN OTHER SPECIES 1.
2. 3.
4.
5. 6. 7.
8. 9.
10. 11.
12. 13. 14.
15. 16.
17.
McDonald TO, Hodges JW, Borgmann AR, Leaders FE. The water-loading test in rabbits: a method to detect potential ocular hypotensive drugs. Arch Ophthalmol 1969; 82:381–384. Thorpe RM, Kolker AE. A tonographic study of water loading in rabbits. Arch Ophthalmol 1967; 77:238–243. Bonomi L. Methods to produce ocular hypertension in animals. In: Saettone MS, Bucci G, Soeiser P, eds. Ophthalmic Drug Delivery, Biopharmaceutical, Technological and Clinical Aspects. Padova: Liviana Press, 1989. Himber K, De Burlet G, Andermann G. Effects of adrenergic agents on alpha-chymotrypsin-induced ocular hypertension in albino and pigmented rabbits: a comparative study. J Ocular Pharm 1989; 5:93–98. Sears D, Sears M. Blood-aqueous barrier and alpha-chymotrpsin glaucoma in rabbits. Am J Ophthalmol 1974; 77:378–383. Gherezghiher T, March WF, Mordquist RE, Koss MC. Laser-induced glaucoma in rabbits. Exp Eye Res 1986; 43:885–894. Vartiainen J, MacDonald E, Urttie A, Rouhiainen H, Virtanen R. Dexmedetomidine-induced ocular hypotension in rabbits with normal or elevated intraocular pressures. Invest Ophthalmol Vis Sci 1992; 33:2019–2023. Tuovinen E, Liesmaa M, Esila¨ R. The influence of corticosteroids on intraocular pressure in rabbits. Acta Ophthalmol 1966; 44:901–905. Majzoub U. The effect of steroids on aqueous dynamics of rabbit eyes. In: Drug Mechanisms in Glaucoma. Paterson G, Miller SJH, Paterson GD, eds. Boston: Little, Brown: 93–103. Levene RZ, Rothberger M, Rosenberg S. Corticosteroid glaucoma in the rabbit. Am J Ophthalmol 1974; 78:505–510. Knepper PA, Breen M, Weinstein HG, Blacik LJ. Intraocular pressure and glycosaminoglycan distribution in the rabbit eye: effect of age and dexamethasone. Exp Eye Res 1978; 27:567–575. Rohen JW, Linner E, Witmer R. Electron microscopic studies on the trabecular meshwork in two cases of corticosteroid glaucoma. Exp Eye Res 1973; 17:19–31. Radius RL, Bade B. Pressure-induced optic nerve axonal transport interruption in cat eyes. Arch Ophthalmol 1981; 99:2163–2174. Sossi N, Anderson DR. Effect of elevated intraocular pressure on blood flow. Occurrence in cat optic nerve head studied with iodoantipyrine I-125. Arch Ophthalmol 1983; 101:98–101. Weinstein JM, Duckrow RB, Beard D, Brennan RW. Regional optic nerve blood flow and its autoregulation. Invest Ophthalmol Vis Sci 1983; 24:1559–1565. Sossi N, Anderson DR. Blockage of axonal transport in optic nerve induced by elevation of intraocular pressure: Effect of arterial hypertension induced by angiotensin I. Arch Ophthalmol 1983; 101:94–97. Whitley RD, Albert RA, McDaniel GR, Brewer RM, Mora EC, Henderson RA. Photo-induced buphthalmic avian eyes. I. Continuous fluorescent light. Poult Sci 1984; 3:1537–1542.
Intraocular Pressure Elevation: Laser Photocoagulation 18. 19. 20. 21. 22.
23. 24.
25. 26. 27.
28.
29.
30.
31.
32. 33. 34.
35. 36.
83
Lauber JK, Cheng KM. Heritable susceptibility to environmentally induced glaucoma in several mutants of Japanese quail. J Hered 1989; 80:268–271. Wilkinson JL, Hodos W. Intraocular pressure and eye enlargement in chicks. Curr Eye Res 1991; 10:163–168. Lauber JK. Light-induced avian glaucoma as an animal model for human primary glaucoma. J Ocular Pharm 1987; 3:77–100. Barrie KP, Gelatt KN, Gum GG, Samuelson DA. Effects of alpha chymotrypsin on the canine eye. Am J Vet Res 1982; 43:207–216. Smith PJ, Brooks DE, Lazarus JA, Kubilis PS, Gelatt KN. Ocular hypertension following cataract surgery in dogs: 139 cases. J Am Vet Med Assoc 1992–1993; 209:105–111. Moore CG, Milne ST, Morrison JC. Noninvasive measurement of rat intraocular pressure with the TonoPen. Invest Ophthalmol Vis Sci 1993; 34:363–369. Niwa N, Ogura K, Suganuma A, Sagami F, Matsubara Y. A method for measuring intraocular pressure in anesthetized rats using a simple tonometer. Animal Eye Res 1995; 14:23–29. Cohan BE, Bohr DF. Goldmann applanation tonometry in the conscious rat. Invest Ophthalmol Vis Sci 2001; 42:340–342. Sawada A, Neufeld AH. Confirmation of the rat model of chronic, moderately elevated intraocular pressure. Exp Eye Res 1999; 69:525–531. Silver DM, Quigley HA, Castelbuono AC, Pease ME, Kerrigan-Baumrind LA, Crowhurst C. Rat intraocular pressure using a miniaturized pneumatic tonometer. Invest Ophthalmol Vis Sci 2001; 42(ARVO Abstracts)(4):S830(Abs nr 4456). Morrison JC, Moore CG, Deppmeier LMH, Gold BG, Meshul CK, Johnson EC. A rat model of chronic pressure-induced optic nerve damage. Exp Eye Res 1997; 16: 85–96. Shareef SR, Garcia-Valenzuela E, Salierno A, Walsh J, Sharma SC. Chronic ocular hypertension following episcleral venous occlusion in rats. Exp Eye Res 1995; 61: 379–382. Ueda J, Sawaguchi S, Hanyu T, et al. Experimental glaucoma model in the rat induced by laser trabecular photocoagulation after an intracameral injection of India ink. Jpn J Ophthalmol 1998; 42:337–344. Mittag TW, Danias J, Pohorenec G, Yuan H-M, Burakgazi E, Chalmers-Redman R, Podos SM, Tatton WG. Retinal damage after 3 to 4 months of elevated intraocular pressure in a rat glaucoma model. Invest Ophthalmol Vis Sci 2000; 41:3451–3459. Morrison J, Farrell S, Johnson E, Deppmeier L, Moore CG, Grossmann E. Structure and composition of the rodent lamina cribrosa. Exp Eye Res 1995; 60(2):127–135. Garcia-Valenzuela E, Shareef S, Walsh J, Sharma SC. Programmed cell death of retinal ganglion cells during experimental glaucoma. Exp Eye Res 1995; 61(1):33–44. Klo¨cker N, Cellerino A, Ba¨hr M. Free radical scavenging and inhibition of nitric oxide synthase potentiates the neurotrophic effects of brain-derived neurotrophic factor on axotomized retinal ganglion cells in vivo. J Neurosci 1998; 18:1038–1046. Buys YM, Trope GE, Tatton WG. (-)-Deprenyl increases the survival of rat retinal ganglion cells after optic nerve crush. Curr Eye Res 1995; 14:119–126. John SW, Anderson MG, Smith RS. Mouse genetics: a tool to help unlock the mechanisms of glaucoma. J Glauc 1999; 8(6):400–412.
84 37.
38.
39. 40.
41. 42.
43. 44.
45.
Gabelt et al. John SWM, Smith RS, Savinova OV, Hawes NL, Chang B, Turnbull D, Davisson M, Roderick TH, Heckenlively JR. Essential iris atrophy, pigment dispersion, and glaucoma in DBA/2J mice. Invest Ophthalmol Vis Sci 1998; 39:951–962. Bayer AU, Neuhardt T, May AC, Martus P, Maag K-P, Brodie S, Lu¨tjen-Drecoll E, Podos SM, Mittag T. Retinal morphology and ERG response in the DBA/2NNia mouse model of angle-closure glaucoma. Invest Ophthalmol Vis Sci 2001; 42: 1258–1265. John SWM, Hagaman JR, MacTaggart TE, Peng L, Smithes O. Intraocular pressure in inbred mouse strains. Invest Ophthalmol Vis Sci 1997; 38(1):249–253. Avila MY, Carre DA, Stone RA, Civan MM. Reliable measurement of mouse intraocular pressure by a servo-null micropipette system. Invest Ophthalmol Vis Sci 2001; 42:1841–1846. Cohan BE, Bohr DF. Measurement of intraocular pressure in awake mice. Invest Ophthalmol Vis Sci 2001; 42:2560–2562. Takahashi H, Noda S, Imamura Y, Nagasawa A, Kubota R, Mashima Y, Kudoh J, Oguchi Y, Shimizu N. Mouse myocilin (Myoc) gene expression in ocular tissues. Biochem Biophys Res Commun 1998; 248(1):104–109. Smith RS, Bechtold L, John SWM. Ultrastructure of mouse trabecular meshwork. Invest Ophthalmol Vis Sci [ARVO Abstract], 1998; 39:S705. Abstract nr 3238. Dietlein TS, Jacobi PC, Krieglstein GK. Experimentally induced outflow obstruction in the mouse eye [ARVO Abstract]. Invest Ophthalmol Vis Sci 1998; 39(4): S704. Abstract nr 3233. Fisher J, Levkovitch-Verbin H, Schori H, Yoles E, Butovsky O, Kaye JF, BenNun A, Schwartz M. Vaccination for neuroprotection in the mouse optic nerve: implications for optic neuropathies. J Neurosci 2001; 21(1):136–142.
6 Light-Induced Retinal Degeneration Daniel T. Organisciak, R. M. Darrow, and L. S. Barsalou Wright State University Dayton, Ohio, U.S.A.
I.
INTRODUCTION
Acute light-induced photoreceptor cell degeneration has been studied for over 35 years as a model of visual cell loss arising from genetic and age-related disease or light-induced trauma during ocular surgery. The rationale for these studies has been based largely on morphological similarities between the end stages of disease and the appearance of the retina after experimental retinal degenerations. While much has been learned, the recent advent of transgenic animal models of human disease and retinal gene knockouts has resulted in renewed interest in the light-damage model. In addition to using acute intense light to test the relevance of genetically modified animals as models of human retinal degenerations, we have also gained insights into the mechanism of retinal light damage. We now know that visual cell loss, whether of genetic origin, age-related, or light-induced, involves programmed cell death. We also know that light exposure induces an oxidative stress in the retina that may trigger apoptosis, leading to photoreceptor cell degeneration. Accordingly, genetic predisposition and environmental light appear to produce synergistic effects on the survival or demise of visual cells. The great advantage of acute light damage, however, is that if done correctly it can result in the synchronous involvement of numerous photoreceptors, whereas genetic or age-related degenerations involve fewer photoreceptors over much longer periods of time. Retinal light damage, therefore, provides an opportunity 85
86
Organisciak et al.
to study cell death during a shortened time course and to determine the order of events leading to retinal degeneration.
II. GENERAL OVERVIEW The prevention of light-induced retinal degeneration has been described in numerous studies employing natural or synthetic antioxidants, dietary alterations of retinol or essential fatty acids, and even manipulation of the light-rearing environment or the start time of light treatment. These various forms of neuroprotection have implications for therapeutic strategies designed to treat or prevent retinal disease. To this end, the following is provided as an experimental guide for acute light-induced retinal degeneration in animal models. It is intended to maximize experimental outcomes, to reduce variables, and to minimize mistakes (most of which we have already made). We begin with the selection of animal models followed by exposure paradigms and expected outcomes. Finally, we address ocular neuroprotection by looking at endogenous processes in the retina and exogenous treatments. The interested reader is referred to additional comprehensive reviews or descriptions on mechanisms of retinal light damage and cellular death [1–8].
III. METHODS/PROCEDURES; ADVANTAGES/ DISADVANTAGES A.
Animal Selection and Maintenance
1. Species Most animals will undergo retinal cell loss from intense light, provided that light exposure conditions are optimized and other precautions are taken. Commonly used animal models are rodents, which, because of their rod-dominant retinas and because they are nocturnal, incur retinal damage at light levels only one- to two-fold greater than ordinary room lighting. In addition, rats or mice are often selected for transgene expression and gene knockouts, which further increases their importance. Less frequently used are rabbits. Although their retinas can be damaged by visible light, the intensity required is much greater than for rats and most successful studies have employed more energetic blue light [9] or UV wavelengths. In our hands, we have never been able to induce retinal light damage in albino rabbits with green light exposures that cause massive visual cell loss in rats. Diurnal species including, ground squirrels [10], birds [11], miniature pigs [12], and monkeys [13] have been studied for the effects of UV or visible light on rod or cone cells, although only monkeys have a foveal region akin to
Light-Induced Retinal Degeneration
87
the human retina. Notwithstanding their cost and the need for anesthesia, some of these species are difficult to breed in captivity and are most often acquired from the wild. Using captured animals increases experimental variability, and captivity is potentially much more stressful than for typical laboratory animals. The intensity of light required to induce retinal damage in diurnal animals is also higher, often mW or greater, than the µW levels used for nocturnal rodents. Similarly, longer duration light exposure at higher light energy levels can lead to damage in the inner retinal layers, which compromises interpretations regarding photoreceptor cell loss. Other species have not been studied as extensively as mammalian models. These include fish, amphibians, crustaceans, and insects such as drosophila. However, the zebra-fish model developed by Dowling and associates [14] and albino and pigmented trout [15] have been used in light-damage studies. Similarly, transgenic frogs developed by Papermaster [16] and Drosophila mutants offer additional models for such studies. Although the use of such models is beyond the scope of this chapter, additional information can be found in Photostasis and Related Phenomena [17] as well as other sources [18–21]. Among the rodent models, light damage typically spares both the cone cells and the inner retinal layers [22–24]. Depending on the intensity of light and the duration of exposure, the inferior hemisphere of the rat eye is also less susceptible to damage than the superior hemisphere [25,26]. The same appears to be true for mice [23,24] and for transgenic rodents [27,28], although c-fos knockout mice do not exhibit the same pattern of retinal light damage [29]. Whereas genetically manipulated mice are more widely available and have been used to study specific proteins that may be involved in retinal light damage, in general, mice are more difficult to damage than rats. In an earlier study, LaVail et al. [23] demonstrated a genetic predisposition among mice that leads to retinal light damage following continuous exposure for 1–6 weeks. Using young Royal College of Surgeons (RCS) rats, we reached similar conclusions in animals exposed to light for only 8 h [30]. The reasons for the greater light damage susceptibility of rats are not clear but may be related to the packing of photoreceptors within the eye and the relative rhodopsin concentrations of the two species (0.35 vs. 2.0 nmol/eye) 2. Pigmentation Where practical, albino animals are a better choice than pigmented animals because of their relative light damage susceptibilities. Pigmented animals require mydriatics and about twofold longer exposure times (at equal light intensity) than albinos to exhibit the same degree of retinal damage. This is because light is effectively screened by melanin and only enters the pigmented eye through the dilated pupil. In albino rats or mice, light enters the eye through the sclera as well as the pupil and, although reduced in intensity, still penetrates the closed
88
Organisciak et al.
eyelids of sleeping animals. Obviously, the choice of mydriatic and its duration of action, relative to the length of light exposure needs to be considered. Irrespective of this limitation, the mechanism of action of light on photoreceptors is the same in albino and pigmented animals, and numerous studies have shown that melanin is unlikely to cause visible-light induced retinal degeneration (see Ref. 3 for references therein). Our best experience has been with male albino rats, although, in our hands, female Sprague-Dawley or albino RCS animals are equally susceptible to light damage. This may not be true, however, for other strains of albino rats, transgenic rats having rhodopsin mutations, and some mice. To minimize variability, it is recommended that a single sex of experimental animal be used where possible. 3. Age and Rearing Conditions Irrespective of the choice of animals to be used, variables associated with lightrearing conditions and age can impact the outcome of acute retinal light damage. All too often animals obtained from commercial sources are used within several days of arrival and without consideration for rearing light differences and dark adaptation. For example, with respect to rodent models, light levels in many laboratory animal facilities range from over 1000 lux white light at the top of animal cage racks to less than 120 lux on the bottom shelf. This results in differences in rhodopsin levels among experimental animals and may result in visual cell loss in the long run. Because complete rod outer segment rhodopsin turnover in rodents occurs in about 9–10 days we suggest a minimum of 2 weeks acclimation, at a defined light level, before experimentation. A nice example of the differential effects of light rearing environment on light damage has been published [31,32]. These authors found that rats reared under higher light intensities are more resistant to retinal light damage than rats from lower light level environments. However, at the highest light levels, visual cell loss also occurred prior to intense light treatment. This variable is even more critical in studies of age-related differences in light damage susceptibility. Commercial suppliers normally rotate animals from the top to bottom shelves of the cage racks, but obtaining rats of different ages at different times can easily lead to differences in rhodopsin and other ocular proteins which are up- or down-regulated by light and dark rearing conditions [33]. Nevertheless, animals should be well dark adapted (12–16 h), to maximize rhodopsin content before intense light treatment, and all exposures should be initiated at the same time of the day. In our animal facility, cyclic light–reared rats are housed on cage racks equipped with multiple 7 W night lights (G.E. 7C7, indicator bulbs; Cisco Electric Co. Columbus, OH) in multiple outlet strips attached to cage racks above each shelf. By wiring these outlet strips in parallel and with the use of a commercial
Light-Induced Retinal Degeneration Table 1
89
Species Selection for Retinal Light Damage Studies
Animal Type Nocturnal (rats/mice)
Diurnal (pigs, birds, squirrels, monkeys) Other models (Fish, amphibians, insects)
Advantages
Notes
Easily damaged by light (µW); albino and pigmented strains; transgenic/knockouts Rod/cone cell models of light damage; some transgenics available Transgenics being developed or mutants; known rod/outer segment regeneration after damage (some)
Rearing conditions need to be carefully controlled
Cost, anesthesia, breeding difficult, UV and mW levels of light required Retinal cell types different from mammalian, not well studied
dimmer switch and timer, we are able to set light levels uniformly to any desired illuminance (e.g., 20–30 lux) as well as on/off times (e.g., 8 a.m./8 p.m.) for the entire colony of animals. The 12-h cyclic light period is also adjusted for seasonal changes in time such as daylight savings time. Overhead lighting in our animal room has been disconnected. Normally, rats, purchased as weanlings or born in the facility, are maintained until the age of 60 days (⬃4 rod renewal cycles) before use. A similar room for dark rearing of rats has an internal, light-protected door for access and red Plexiglas panels in the overhead light fixtures. These lights are on for less than 30 min per day for routine animal care. Because the red Plexiglas panels (2423) transmit light greater than 600 nm (Fig. 3), rhodopsin is not bleached in dark-reared rats during normal care and feeding periods. Table 1 summarizes some of the considerations and conditions discussed in animal selection and maintenance B. Acute Light Exposure Paradigms Based on the foregoing and our experience, the focus of this section will be on the rat model of light damage. Because retinal light damage exhibits reciprocity and the bleaching of rhodopsin triggers visual cell loss [25], both intensity of light and duration of exposure need to be considered along with other factors that can influence the extent of photoreceptor cell damage. 1. Light Chamber Design The relative ease with which rats incur retinal light damage often leads investigators to overlook experimental conditions that can help to reduce variability
90
Organisciak et al.
and increase the reliability of their conclusions. One variable is the uniformity of light during exposure. As intense light is a stress for experimental animals, they will seek the lowest level of light and attempt to shield their eyes during exposure. This may be the corners of rectangular cages or plastic exposure chambers. In wire top cages with light from above, animals will often remain under their water bottles or food supply, if provided on top of the wire cage cover. Figure 1 depicts the type of light chambers we use to induce retinal degeneration in rats. Our light chambers are made from 1/8 in. thick Plexiglas, molded to produce a 24 in. long, 6 in. OD cylinder (Dayton Plastics, Dayton, OH). Shown here is green 2092 Plexiglas [25], but other types of plastics can be used depending on the wavelengths of light desired. The overall advantage of Plexiglas, or other plastics, is that UV is effectively absorbed by most of these materials (see Fig. 3). This reduces the potential for conjunctivitis in experimental animals [25]. The Plexiglas exposure chamber is mounted in a cradle and is surrounded by 7–12 in. circular 32W fluorescent bulbs (G.E. “Cool white” FC 12T9-CW). This provides 360° of relatively uniform light during exposure periods. The chamber contains an internal wire mesh floor and is tipped 10° to eliminate wastes. A small 4 in. electric pancake fan mounted at the rear is run at approximately 50% line voltage to reduce heating during light exposure. Our chambers
Figure 1 Plexiglas light exposure chamber for rats.
Light-Induced Retinal Degeneration
91
are equipped with on/off timers (EC71D/305 Paragon Electric Corp., Two Rivers, WI), which can be programmed for different start/stop times or intermittent light exposures. It is important that the electric fans are on only during light periods. We use four green chambers with light levels of 1200–1500 lux (corneal irradiance 170 or 200 µW/cm2) measured internally with a light meter (IL 1400A, International Light Inc., Newburyport, MA). Lower light levels can be achieved by wrapping the cylinders with commercial screening, which acts like a neutral density filter (see Fig. 1). Hyperthermic light exposures are conducted in four vertical 9 in. OD round chambers, each surrounded by three circular 12 in. fluorescent bulbs. Rats are maintained at approximately 80% relative humidity and increased temperature for 2 h before, as well as during, light exposure (required to raise core body temperature). The cabinet contains heaters set for the desired hyperthermic exposure. The chamber shown in Figure 2 is from W.K. Noell and was used in his original light damage study. See Noell et al. [25] and Organisciak et al. [34] for details.
Figure 2 Hyperthermic light exposure chamber.
92
Organisciak et al.
Well dark adapted rats are given food and water ad libitum during light exposure. The animals, two per chamber or one for hyperthermic exposure, are also unrestrained and unanesthetized during exposure. Typically, light exposure starts at 9 a.m., although exposures can be conducted at any time provided start times are consistent. Following light exposure, our rats are maintained in darkness for as long as 2 weeks before end point measurements of visual cell loss, whereas many investigators simply return animals to their normal rearing light environment. This does not appear to alter the outcome of most measurements of visual cell loss. 2. Factors That Influence the Extent of Light Damage Aside from the principle of reciprocity, which implies that time and intensity have an inverse relationship with respect to light damage, many other characteristics influence the extent of visual cell loss—for example, the wavelengths of light. Figure 3A shows the transmission spectra for several types of Plexiglas, or other plastic materials. Green Plexiglas 2092 and blue 2424 each have an approximately 100 nm band pass, but both bleach rhodopsin with the same efficiency, and bleaching is dependent on light intensity. Figure 3B shows that the energy for blue and green light is approximately the same at the higher green intensities,
Figure 3 Characteristics of plastics used in light chambers. (A) Transmission spectra of various plastic materials. (B) Rhodopsin bleaching in rats exposed to light in chambers composed of green or blue Plexiglas at different light intensities.
Light-Induced Retinal Degeneration
93
despite the differences in photometric units (e.g., 1500 lux vs. 170 lux). Accordingly, at similar corneal irradiances, rhodopsin is bleached by about 90% in 5 min. However, in a recent study with 403 nm blue light, retinal was found to be photoisomerized and rhodopsin photoregenerated [35]. Whatever light filter is employed, however, it is important to keep in mind that plastic pigments age and chambers should be replaced periodically (e.g., see old green 2092 [Fig. 3A]). Likewise, fluorescent bulbs also “age” and should be replaced on a regular schedule. Most often, continuous light exposures are used, although intermittent light and hyperthermic treatment both enhance the extent of light damage in rats (Fig. 4). These light-exposure paradigms have been utilized in an attempt to produce synchronous visual cell involvement. For the exposure conditions shown in Figure 4, and with unanesthetized albino rats, we find that hyperthermia is 18to 24-fold more damaging than continuous exposure under euthermic conditions. However, see de Lint et al. [36], who measured fundoscopic lesions in hyperthermic-pigmented rats under anesthesia and found only a twofold difference. In our hands, intermittent light treatment is twofold more damaging than continuous light exposure of the same intensity and duration [37].
Figure 4 Paradigms of intense light exposure. Weanling rats are maintained in dim cyclic light or darkness for 40 days before exposure to intense light. When antioxidants, or drugs, are given (normal injections) these normally precede light treatment.
94
Organisciak et al.
The prior light-rearing conditions of experimental animals, age, and diet are additional factors that influence light damage. Irrespective of the type of light exposure used, dark-reared rats are about two- to threefold more susceptible to light damage than comparable cyclic light reared animals. Dark-reared rats also incur RPE cell damage, whereas in cyclic light–reared rats the RPE is less involved. Young cyclic light–reared rats (⬍25 days) are generally quite resistant to light damage compared with adult animals [38]. Among adult rats we found a twofold increase in visual cell loss at 12 months versus 2-month-old animals after 24 h green light exposures. Ironically, dark rearing resulted in no age-related increase in retinal light damage, but those animals were still more susceptible to damage than cyclic light reared animals [39]. In rats fed vitamin A–deficient diets, or in those fed essential fatty acid–deficient diets, retinal light damage is also decreased [40,41]. Recently we tested the start time of light exposure as a variable. Our findings indicate that light exposure beginning at 1 a.m. causes oneto twofold more damage than light exposure at 9 a.m. [42]. With the exception of mydriatics, anesthetics such as halothane also reduce light damage susceptibility in rodents by preventing rhodopsin regeneration [43]. These factors have been summarized in Table 2. C.
Outcomes of Retinal Light Damage
Depending on the types of information sought and the measurements used, conclusions regarding the extent and nature of light-induced cell loss can vary. Among the typical techniques used—histology, electrophysiology, and biochemistry—each has its limitations, and correlative measurements are recommended. In this section we point out some of these limitations and other problems associated with interpretations based on the time course of retinal degeneration. Our
Table 2 Some Experimental Factors Which Alter Retinal Light Damage in Rats Reduced light damage • • • • • • • •
Cyclic light reared Young (2 mo.) Euthermic 9 a.m. start time Continuous light Vitamin A deficient Deficient DHA (22 : 6) Pigmented (dilated)
Enhanced light damage • • • • • • • •
Dark reared Old (12 mo.) Hyperthermic 1 a.m. start time Intermittent light Normal diet Normal DHA levels Albino
Fold increase 2–3 2 18–24 1–2 2 1–2 1–2 2
Light-Induced Retinal Degeneration
95
approach will be a quantitative assessment of visual cell loss arising from the synchronous involvement of photoreceptors during intense light exposure. 1. Measurements of Visual Cell Loss Much of our understanding of lights’ pathological effects on the retina has come from histology. It is the most widely used approach, providing information on the types of cells involved and their locations within quadrants of the retina. No other approach is as good in describing organelle damage within cells during or after light exposure. From such studies we know that the entire visual cell becomes involved during damage, including the ROS, inner segment mitochondrial rich region, nucleus, and even the synaptic end plate. Depending on the type of light and the animal model used RPE damage also occurs, either simultaneously or shortly after visual cell damage [6,38]. The major drawback is that histology is descriptive and is often presented as a “snapshot” that is interpreted as representative damage in the entire tissue. While this may be true, the examples chosen frequently are those that depict the greatest morphological changes. Accordingly, histological measurements that assess visual cell loss by morphometric analysis over the entire eye, including ONL thickness, ROS length, and/or rod cell nuclear counts, are required. Such approaches require serial sectioning and proper cell alignment for accurate conclusions regarding ONL volume and ROS length. Obviously this is painstaking and tedious work, requiring considerable expertise. One of the first examples of this approach was published by Rapp and Williams [26] in their detailed analysis of the superior and inferior hemispheres of lightdamaged rat eyes. Electroretinography is one of the best examples of unbiased functional analysis of photoreceptor survival and loss. Its strength lies in its repetitive use for following ERG changes in the same animals. By comparing scotopic A- and Bwave amplitudes, latencies, and thresholds, which reflect the summed electrical signals generated primarily by intact photoreceptors and Muller cells, much can be learned about the extent of visual cell damage and its time course. Photopic analysis, to determine cone cell responses, provides valuable functional measurements of histologically identifiable cell types that survive light damage. This is particularly useful in cone-dominant animal models because purified cone pigments have never been isolated from the retina. Unfortunately, simple ERG analysis and morphometry of visual cell loss do not always correlate as well as expected. This may be related to RPE cell damage and the subsequent shunting of current generated in response to light stimuli [3]. Despite this limitation, electroretinography, using Ganzfeld illumination, provides an average summated response from the entire population of synaptically intact photoreceptors remaining after light damage. In one of the few comprehensive studies comparing ERG analysis and morphometry in light-damaged rat retinas Sugawara, Sieving, and
96
Organisciak et al.
Bush [44] report a high degree of correlation between the two techniques and discuss potential drawbacks in electroretinography. The interested reader is referred to this thorough study. Classical biochemistry offers many opportunities to assess retinal damage and cell loss because of the variety of measurements that can be made with the same tissue. As with electroretinography and morphometry, the two eyes of the same animal can also provide correlative assessments of light damage. This is based on many studies showing that both eyes have identical photoreceptor cell numbers and components, and that when intense light exposures are properly conducted the extent of visual cell loss is the same. Typically, biochemical measurements are “average responses” determining changes in the entire tissue that remains following damage or cell loss. Because acute light-induced retinal degeneration induces synchronous cell involvement, much has been learned about the mechanism(s) of light damage and cell death from these techniques. The relative ease by which biochemical measurements can be made is another advantage, but the choice of measurement and the time course of damage can lead to erroneous conclusions. Because rhodopsin is essentially contained within the ROS of rod photoreceptors and damaged cellular material is removed by the RPE, it can be used as an endpoint determination of rod cell survival following light damage. When compared with the rhodopsin level in similarly dark adapted, but unexposed, littermate controls, the extent of visual cell loss can reasonably be quantified. Although this technique works well in normal rats, in RCS rats its use alone is unsatisfactory. Because phagocytosis is deficient in these animals, rhodopsin accumulates in the partially degraded photoreceptor debris, and estimates of visual cell loss by DNA measurements differ by as much as 50% from simple rhodopsin determinations [30]. Through the activation of endonucleases, DNA is degraded in dying photoreceptors, even in the absence of phagocytosis by RPE. However, DNA is also present in cells of the inner nuclear layers. To quantitate photoreceptor cell DNA losses, therefore, it is necessary to subtract the DNA content of the inner retina layers. Based on this approach and using old RCS rat retinas, which have lost their photoreceptor cells, we determined that about 70% of retinal DNA is contained within the ONL. By using this subtraction technique, reasonable conclusions about visual cell loss and DNA fragmentation in photoreceptors can be achieved from retinal DNA measurements following light damage. Similarly, about 50% of retinal protein and lipid is contained within the photoreceptor cell layer of the adult rat retina. Because extensive light-induced visual cell loss can result in the loss of up to one half of the retinal protein, this needs to be controlled for in typical enzyme measurements. For example, the activity of trans retinol dehydrogenase, which is present in photoreceptors, is decreased by as much as 25% following light exposure, but the Muller cell en-
Light-Induced Retinal Degeneration
97
zyme glutamine synthetase (GS) is unaffected by light [45]. These measurements need to be reported per retina, because GS activity per milligram of protein would show about a 50% increase; clearly a different and erroneous conclusion. Therefore, despite the relative ease and advantage of biochemical measurements, knowledge of the distribution of retinal proteins, enzymes, DNA, or lipid within the normal retina is essential for quantitative assessments of visual cell loss in degenerating tissues. 2. Time Course of Damage The relative order of events leading to retinal damage and degeneration is important, not only from a mechanistic standpoint but also for possible therapeutic interventions. Time course measurements beginning with the light-mediated bleach of rhodopsin (the trigger for light damage) and proceeding through the final common pathway of cell death can also lead to reliable conclusions regarding endpoint determinations of retinal cell loss. As cellular apoptosis appears to be a primary cause of retinal degenerations in animal models and in human disease we can also learn much from its measurement. There are, however, potential problems associated with time course measurements of retinal light damage and cell loss. The decision to use one type of measurement or another in light damage studies rests on the important variable of when the determination is made. In rats a somewhat normal ERG is recordable for several hours after intense light treatment and rhodopsin regenerates in darkness to about 60% of control over a period of 6 h to 1 day even though photoreceptors are severely damaged [25,63]. Subsequently, the ERG declines and the level of rhodopsin decreases as photoreceptors die and degenerate over the next 4–6 days. Ten to 14 days after intense light, most of the degenerated photoreceptors have been removed and repair in the surviving cells is nearly complete. At this time morphometry, ERG, and biochemical measurements more accurately reflect total visual cell loss. Over the same 2-week “recovery period,” retinal DNA levels decline from about 90% of control immediately after light exposure to a final level commensurate with the extent of visual cell loss and the percent of rhodopsin loss. Thus, although light leads to DNA degradation, often there is little measurable loss during a typical light exposure period. Conclusions regarding the nature of DNA degradation also depend on the timing of the measurements and the technique used. An example of this is shown in Figure 5, which depicts neutral and alkaline gel electrophoretic analysis of DNA from rat retinas. Using neutral gels, a pattern of DNA ladders is present in cyclic light– reared rats immediately after 24 h of light exposure. This pattern is not as apparent in dark-reared rat retinas until 2 days after a 3 h light treatment. In both cases DNA ladders are not present 4 to 16 days after light treatment. Alkaline gel
98
Organisciak et al.
Figure 5 Gel electrophoresis of rat retinal DNA. (A,B) neutral agarose gel electrophoresis (C,D) alkaline agarose gels. Each lane was loaded with a total of 2 µg DNA from three different rat retinas. (Adapted from Ref. 46.)
electrophoresis, which detects single strand breaks in DNA and larger fragments, shows a pattern of DNA breakdown that coincides with the apoptotic pattern in cyclic light rats, but that precedes the same pattern in dark reared rat retinas by at least 12 h. What does this mean? Simply put, if DNA electrophoresis is conducted 4–16 days after light exposure, no evidence of the nature of DNA degradation is detectable. Furthermore, by neutral gel analysis, apoptosis could be detected in one group of rats after 12 h in darkness but is barely detectable in the other group of animals. While this example is for DNA [46], similar time-dependent differences in retinal gene expression and protein or enzyme analysis can be found. The bottom line is that conclusions based on single time-point determinations of the extent or nature of pathological changes in the retina can vary considerably depending on when those measurements are made. D.
Prevention of Retinal Light Damage
Having discussed some variables associated with light-induced retinal degeneration, the goal of this section is experimental design related to its prevention. The literature is replete with claims that one treatment or another reduces the extent of visual cell loss from light, but upon further analysis these findings are often
Light-Induced Retinal Degeneration
99
viewed as marginal. That is not to say that extending visual cell life by, say, 5 to 10% is not important. As almost any clinician can attest, there are patient quality-of-life benefits from preserving vision in age-related or genetically based conditions. Because the mechanisms of retinal cell loss in human disorders and experimental retinal degenerations bear a striking resemblance, well-designed interventions in animal models of light damage offer hope for future clinical trials. With an aim toward improving the efficacy of drug, antioxidant, or dietary treatments, we discuss some of the problems associated with these in animal models. 1. Endogenous or Adaptive Processes The relative susceptibility or resistance of the retina to light damage depends on a variety of conditions already discussed (see Table 2). These include rearing conditions and age, both of which contribute to changes in retinal gene expression. Because these changes precede intense light treatment, they are endogenous in nature and can greatly affect experimental outcomes. Simple differences in rhodopsin levels, or other visual cell proteins, which are altered by rearing conditions or age, indicate that control measurements should be made before light damage experiments are done. The same is true for transgenics or retinal gene knockout models as these can alter in unexpected ways other proteins or processes in the retina. For example, transgenic rd mice overexpressing Bcl-2 were found to be resistant to retinal light damage, but also to have lower rhodopsin levels than controls; a possible reason for the finding [47]. C-fos knockout mice also have reduced rhodopsin levels and fewer photoreceptor cells than c-fos ⫹/⫹ animals [48], although the basic conclusion regarding the importance of this transcription factor in light-induced apoptotic cell death is unchanged. In addition to providing mechanistic insights into light damage, the use of genetically modified animal models has implications regarding the progression of inherited or age-related retinal disease. Early work with RCS rats, which have a recessive mutation in phagocytosis of ROS tips by RPE, showed that both lightrearing conditions [49] and intense light exposure [50] accelerate the rate of retinal degeneration. In albino mice, LaVail et al. [23] found light damage susceptible and resistant strains, a finding that was confirmed in F1 heterozygotes [24]. In some cases of these “spontaneous mutations,” retinal light sensitivity led to studies into the underlying gene defect. However, with prescribed transgene expression or knockout models, questions about specific proteins and the role of light environment can be addressed more directly. Using transgenic mice expressing a mutated SOD gene, Mittag et al. [51] found that they were more susceptible to retinal light damage than nontransgenic animals. In mice [27] and in rats [28] with rhodopsin mutations, light rearing and/or intense exposures were found to influence the rates of retinal degeneration. Arrestin knockout mice exhibit re-
100
Organisciak et al.
duced retinal light damage [52] and c-fos knockouts [29] and the absence of RPE-65, which is required for rhodopsin regeneration [53], prevents light damage [54]. These studies and others [47,55] have provided useful insights into the mechanisms of inherited retinal degenerations and retinal light damage. Using transgenic rats with a P23H rhodopsin mutation or an S-334 truncation, provided by M. LaVail, we compared their relative susceptibilities to retinal light damage. Figure 6 is an example of results from rats reared in different light environments and exposed to intense green light. Overall, P23H line 3 rats were more susceptible than line 2 rats, which have a slower rate of spontaneous retinal degeneration (LaVail, personal communication), S-334 ter line 4 and line 9 rats were equally damaged by light when previously reared in dim cyclic light. Why a folding mutation in the amino terminal region of rhodopsin leads to greater light damage susceptibility than the loss of 15 amino acids from the C-terminal region is unknown at present. However, these transgenic rats appear to be good models of some forms of human retinitis pigmentosa because the loss of rod photoreceptors is accelerated by intense light [56]. In recent studies we found that normal rats exhibit both increased light damage susceptibility and resistance to light damage, depending only on the time of day that light exposure begins [42]. A circadian response was also found to exist in transgenic P23H and S-334 ter rats [8]. This indicates that endogenous factors normally expressed in the retina contribute to light damage susceptibility and that systemic or local signals in the retina may regulate their expression. A variety of growth factors also provide neuroprotection in rats given intraocular injections prior to light exposure [57]. As these neurotrophic factors are normally given in microgram concentrations 2 days before a 1-week intense light treatment, it appears that they elicit changes in the expression of other retinal proteins that provide protection. In a follow-up study, LaVail and associates found that β-FGF induced a dose-dependent expression of the transcription factors c-fos and c-jun in isolated rat Muller cells. [58]. There are two important points to be made from studies such as these. The first is that these factors work at low concentrations in a dose- and time-dependent manner in rats. The same may not hold for other rodents, such as mice, which reemphasizes the need for Cyclic Reared Normal ≅ ⬍ S-334, line 9 ⫽ line 4 ⬍ P23H, line 2 ⬍⬍ line 3 Dark Reared Normal ⬍ S-334, line 4 ⬍ line 9 ⬍ P23H, line 2 ⬍⬍ line 3 Figure 6 Relative light damage susceptibility in different rat models. Results are based on end point determinations of rhodopsin and DNA in comparison to their respective unexposed controls. Normal Sprague-Dawley rats are least susceptible to light damage.
Light-Induced Retinal Degeneration
101
careful species selection. The second point is that trauma to the injected eye can induce a stress response, as shown by partial protection in the eyes of vehicle injected animals. LaVail et al. [57] controlled for this partial response, further supporting their important conclusions regarding neurotrophic factors. 2. Exogenous Treatments or Factors A variety of natural or synthetic compounds have been shown to provide protection against retinal light damage. In fact, a colleague once asked, “is there anything that does not protect the retina against light damage?”(personal communication, B. Winkler). Quite simply, the answer is yes. The same compounds that exhibit protection provide no effect when given at the wrong time, at the wrong concentration, or in the wrong way. There are also a number of variables associated with exogenous agents and neuroprotection in the light damage model. These include systemic versus local effects, concentration, uptake and half-life or clearance rates, as well as time and intensity relationships during light exposure. In this section we describe some of these variables in the rat model. First and foremost the question of efficacy is a function of the extent of light-induced damage. In our model the goal is to induce 50% photoreceptor cell loss in untreated or vehicle-injected rats. In doing so, it is possible to reliably determine whether a drug protects against or enhances retinal light damage by measuring an increase or decrease in visual cell survival. Too much damage can overcome a protective effect and too little can mask the effect. In this type of experiment, the variable most often applied is duration of exposure at a fixed light intensity. With the use of a 50% cell response, it is also possible to compare the relative efficacy of a variety of drugs or closely related compounds. Using such an approach we were able to demonstrate that the synthetic antioxidant dimethylthiourea (DMTU) was more effective in reducing light damage than lascorbic acid, a natural antioxidant present in the retina [59]. It is important to note that the d-steroisomer of ascorbate, which is an antioxidant but not an enzymatic cofactor in mammals, was as effective as l-ascorbate when given at an equal concentration [59]. A second variable is tissue uptake and the half-life of administered compounds. For cyclic light–reared rats treated with DMTU, we found complete protection following light exposures lasting for up to 48 h. By measuring tissue levels, we found that DMTU has a half-life of about 24 h [60], whereas ascorbic acid has a much shorter half-life (4–8 h). DMTU levels in the retina were also about fivefold higher than l-ascorbate 10 min after IP injection. As serum levels of drugs can also be high, it is important to use perfused animals for accurate retinal tissue level measurements. We estimate that the rat retina contains about 10 µL of blood. Accordingly, to demonstrate uptake in a tissue such as the retina, if the animal is not perfused, the concentration of drug should be higher than in blood.
102
Organisciak et al.
A surprising number of studies report protection against retinal light damage without measurements of the nature described above. Thus, it is never really clear whether the administered compound is itself active in preventing the damage or whether it exerts a systemic, perhaps stress, effect leading to protection. Ideally, an agent that exhibits a primary effect not only will be present in the retina for a period of time but also will be found at lower levels in the eyes of light-treated animals than in unexposed controls. The level of ascorbic acid, for example, can be shown to decrease in the rat retina during light exposure and to be effective only when given before the start of light treatment [61,62]. To achieve effective levels of drugs in the eye, the route of administration and the type of compound are also important. We have had good success with water-soluble antioxidants such as DMTU and ascorbic acid given I.P., I.V., or I.O. The kinetics of uptake for ascorbate in retina or RPE are similar for I.P. and I.O. injections although, as expected, the concentrations in the two tissues and the time courses of their losses were different [62]. The same is not true, however, for lipid soluble materials such as β-carotene or other carotenoids. The reasons for this are not entirely clear. Using I.P. or oral gavage of these materials in rats, we have been unsuccessful in demonstrating efficient retinal uptake. It appears that detergents will be required to solubilize these lipophylic materials for possible tissue uptake and light damage studies. Finally, a comment about dose response curves. Whether one chooses to measure tissue uptake or not, or to determine serum levels and clearance rates, a dose response curve for any drug or compound can be informative and useful in
Figure 7 Dose-response curves for antioxidants used in light damage. Rhodopsin measured 2 weeks after intense light exposure. (A) 8 h (䊊) or 3 h (䊉) of intermittent light. (B) 24 h (䊊) or 12 h (䊉) continuous light. (䊊, 䊉) cyclic light and dark reared rats. (A) (䉭, 䉱) urea; (B) (䊐) dehydroascorbic acid; (A) (䊐, ■) unexposed controls ⫾ S.D. (n ⫽ 8–12).
Light-Induced Retinal Degeneration
103
studies of neuroprotection. These responses can be different for different animal models. As shown in Figure 7, the level of antioxidant required for maximal effectiveness is not the same for rats reared in dim cyclic light or in darkness. First the level of tissue damage in the two types of rats is different. Second, the concentration of injected DMTU or ascorbate required to reach a plateau effect is greater in dark reared rats [60,61]. This is important because working with lower concentrations will not achieve the same level of protection, compromising the drugs’ effect. In practice, we use a concentration well above the minimum to achieve protection because this gives a more uniform response among test subjects. This translates into more reliable results with fewer animals required for light damage studies. IV. CONCLUSIONS In the preceding we have attempted to provide an experimental outline for studies related to acute retinal light damage and neuroprotection in vivo. By addressing some problems and pitfalls we hope that future investigators will avoid some of the common mistakes that we and others have made. While a number of mechanistic interpretations might be derived from such studies, the effects of light on genetically modified animals, and the prevention of vision loss, has implications well beyond the use of the models described here. For the present these experimental outcomes will need to be extrapolated to the human condition. However, with a better understanding of the variables associated with acute light damage, we may find simple dietary or other therapeutic approaches to neuroprotection for those afflicted with genetic or age related vision loss. ACKNOWLEDGMENTS Supported by NIH grant EY-01959; Ohio Lions Research Foundation; and M. Petticrew, Springfield, OH. REFERENCES 1. 2. 3.
Lanum J. The damaging effects of light on the retina: Emperical findings, theoretical and practical implications. Survey Ophthal 1978; 22:221–249. Kremers JJM, van Norren D. Two classes of photochemical damage of the retina. Lasers and Light in Ophthalmology 1988; 2:41–52. Organisciak DT, Winkler BS. Retinal light damage: Practical and theoretical considerations. In: Osborne N, Chader G, eds. Progress in Retinal and Ocular Research. Oxford, England: Pergamon Press; 1994:1–29.
104
Organisciak et al.
4.
Rapp LM. Retinal phototoxicity. In: Chang LW, Dyer RS, eds. Handbook of Neurotoxicology. New York: Marcel Dekker; 1995: 32:963–1003 Reme´ CE, Weller M, Szczesny P, Munz K, Hafezi F, Reinboth JJ, Clausen M. Light-induced apoptosis in the rat retina in vivo: morphologic features, threshold and time course. In: Anderson RE, LaVail MM, Hollyfield JG, eds. Degenerative Diseases of the Retina. New York: Plenum Press; 1995:19–25. Reme´ CE, Hafezi F, Marti K, Munz, Reinboth JJ. Light damage to retina and pigment epithelium. In: Marmor MF, Wolfensberger TJ, eds. The Retinal Pigment Epithelium: Function and Disease. New York: Oxford University Press; 1998:563–586. Winkler BS, Boulton ME, Gottsch JD, Sternberg P. Oxidative damage and agerelated macular degeneration. Mol Vis 1999; 5:1–19. Organisciak DT, Sarna T. Genetic, age and light induced degenerations of the retina and retinal pigment epithelium. In: Coohill TP, Valenzena DP, eds. Photobiology for the 21st Century. Overland Park, Kansas: Valdenmar Publishing Co.; 2001:81– 100 Lawwill T. Effects of prolonged exposure of rabbit retina to low intensity light. Invest Ophthalmol Vis Sci 1973; 12:45–51. Collier RJ, Zigman S. Comparison of retinal photochemical lesions after exposure to near-UV or short-wavelength visible radiation. In: LaVail MM, Anderson RE, Hollyfield JG, eds. Inherited and Environmentally Induced Retinal Degenerations. New York: Alan R. Liss; 1989:569–575. Fite KV, Bengston L, Donaghey B. Experimental light damage increases lipofuscin in the retinal pigment epithelium of Japanese quail. Exp Eye Res 1993; 57:449– 460. Dureau P, Jeanny JC, Clerc B, Dufier J-L, Courtois Y. Long term light-induced retinal degeneration in the miniature pig. Mol Vis 1996; 2:7–14. Sperling HG. Prolonged intense spectral light effects on Rhesus retina. In: Williams TP, Baker BN, eds. The Effects of Constant Light on Visual Processes. New York: Plenum Press; 1980:195–214. Li L, Dowling JE. Zebrafish visual sensitivity is regulated by a circadian clock. Vis Neurosci 1998; 15:851–857. Allen DM, Hallows TE. Solar pruning of retinal rods in albino rainbow trout. Vis Neurosci 1997; 14:589–600 Moritz OL, Tom BM, Knox BE, Papermaster DS. Fluorescent photoreceptors of transgenic xenopres laevis imaged in vivo by two microscopy techniques. Invest Ophthalmol Vis Sci 1999; 40:3276–3280. Williams TP, Thistle AB, Eds. Photostasis and Related Phenomena. New York: Plenum Press; 1998. Meyer-Rochow VB. Light induced damage to photoreceptors of spiny lobsters and other crustaceans. Crustaceana 1994; 67:95–109. Kashiwagi T, Meyer-Rochow VB, Nishimura K, Eguchi E. Fatty acid composition and ultra structure of photoreceptive membrane in the crayfish under conditions of thermal and photic stress. J Comp Physiol B 1997; 167:1–8. Allen DM, Pipes C, Deramus K, Hallows TE. A comparison of light-induced rod degeneration in two teleost models. In Hollyfield et al., eds. Retinal Degenerative Diseases and Experimental Therapy. New York: Plenum Publishers; 1999:337–350.
5.
6.
7. 8.
9. 10.
11.
12. 13.
14. 15. 16.
17. 18. 19.
20.
Light-Induced Retinal Degeneration 21.
22. 23.
24.
25. 26.
27.
28.
29.
30.
31. 32. 33.
34. 35.
36. 37.
38.
105
Yoon J, Ben-Ami HG, Hong YS, Park S, Strong KK, Bowman J, Geng C, Baek KB, Minke B, Pak WL. Novel mechanism of massive photoreceptor degeneration caused by mutations in the trp gene of Drosophilia. J Neurosci 2000; 20:649–659. LaVail MM. Survival of some photoreceptor cells in albino rats following longterm exposure to continuous light. Invest Ophthalmol Vis Sci 1976; 15:64–70. LaVail MM, Gorrin GM, Repaci MA. Strain differences in sensitivity to lightinduced photoreceptor degeneration in albino mice. Curr Eye Res 1987; 6:825– 834. LaVail MM, Gorrin GM, Repaci MA, Thomas LA, Ginsberg HM. Genetic regulation of light damage to photoreceptors. Invest Ophthalmol Vis Sci 1987; 28:1043– 1048. Noell WK, Walker VS, Kang BS, Berman S. Retinal damage by light in rats. Invest. Ophthalmol 1966; 5:450–473. Rapp LM,Williams TP. A parametric study of retinal light damage in albino and pigmented rats. In: Williams TP, Baker BN, eds. The Effects of Constant Light on Visual Processes. New York: Plenum Press; 1980:133–159. Naash ML, Peachey NS, Li Z-Y, Gryczan CC, Goto Y, Blanks J, Milam AH, Ripps H. Light-induced acceleration of photoreceptor cell degeneration in transgenic mice expressing mutant rhodopsin. Invest Ophthalmol Vis Sci 1996; 37:775–782. Green ES, Meng MD, LaVail MM, Flannery JG. Characterization of rhodopsin missorting and constitutive activation in a transgenic rat model of retinitis pigmentosa. Invest Ophthalmol Vis Sci 2000; 41:1546–1553. Hafezi F, Steinbach JP, Marti A, Munz K, Wang ZQ, Wagner EF, Aguzzi A, Reme´ CE. The absence of c-fos prevents light-induced apoptotic cell death of photoreceptors in retinal degeneration in vivo. Nat Med 1997; 3:346–349. Organisciak, DT, Li M, Darrow RM, Farber DB. Photoreceptor cell damage by light in young Royal College of Surgeons rats. Curr Eye Res 1999; 2:188– 196. Penn JS, Anderson RE. Effect of light history on rod outer segment membrane composition in the rat. Exp. Eye Res 1987; 44:767–778. Penn JS, Naash ML, Anderson RE. Effect of light history on retinal antioxidants and light damage susceptibility in the rat. Exp Eye Res 1987; 44:779–788. Organisciak DT, Xie A, Wang H-M, Jiang Y-L, Darrow RM, Donoso LA. Adaptive changes in visual cell transduction protein levels: Effect of light. Exp Eye Res 1991; 53:773–779. Organisciak DT, Darrow RM, Noell WK, Blanks JC. Hyperthermia accelerates retinal light damage in rats. Invest Ophthalmol Vis Sci 1995; 36:997–1008. Grimm C, Reme´ CE, Rol PO, Williams TP. Blue light’s effects on rhodopsin: photoreversal of bleaching in living rat eyes. Invest Ophthalmol Vis Sci 2000; 41:3984– 3990. De Lint PJ, van Norren D, Toebosch AMW. Effect of body temperature on threshold for retinal light damage. Invest Ophthalmol Vis Sci 1992; 33:2382–2387. Organisciak DT, Jiang Y-L, Wang H-M, Pickford M, Blanks JC. Retinal light damage in rats exposed to intermittent light. Comparison with continuous exposure. Invest Ophthalmol Vis Sci 1989; 30:795–805. Noell WK. There are different kinds of retinal light damage in rats. In: Williams
106
39.
40. 41.
42. 43.
44.
45.
46.
47. 48.
49. 50. 51. 52.
53.
54.
55.
Organisciak et al. TP, Baker BN, eds. The Effects of Constant Light on Visual Processes. New York: Plenum Press; 1980:3–28. Organisciak DT, Darrow RM, Barsalou L, Darrow RA, Kutty RK, Kutty G, Wiggert BN. Light history and age-related changes in retinal light damage. Invest Ophthalmol Vis Sci 1998; 39:1107–1116. Noell WK, Albrecht R. Irreversible effects of visible light on the retina: Role of vitamin A. Science 1971; 172:76–80. Organisciak DT, Darrow RM, Jiang YL, Blanks JC. Retinal light damage in rats with altered levels of rod outer segment docosahexaenoate. Invest Ophthalmol Vis Sci 1996; 37:2243–2257. Organisciak DT, Darrow RM, Barsalou L, Kutty RK, Wiggert B. Circadian dependent retinal light damage in rats. Invest Ophthalmol Vis Sci 2000; 41:3694–3701. Keller C, Grimm C, Wenzel A, Hafezi F, Reme´ CE. Halothane anesthesia inhibits rhodopsin regeneration and prevents photoreceptor degeneration in the rodent retina. Invest Ophthalmol Vis Sci 2001; 2001; 42:476–480. Sugawara T, Sieving PA, Bush RA. Quantitative relationship of the scotopic and photopic ERG to photoreceptor cell loss in light damaged rats. Exp Eye Res 2000; 70:693–705. Darrow RA, Darrow RM, Organisciak DT. Biochemical characterization of cell specific enzymes in light-exposed rat retinas: Oxidative loss of all-trans retinol dehydrogenase activity. Curr Eye Res 1997; 16:144–151. Organisciak DT, Darrow RA, Barsalou L, Darrow RM, Lininger LA. Light induced damage in the retina: differential effects of dimethylthiourea on photoreceptor survival, apoptosis and DNA oxidation. Photochem Photobiol 1999; 70:261–268. Joseph RM, Li T. Overexpression of Bcl-2 or Bcl-xL transgenes and photoreceptor degeneration. Invest Ophthalmol Vis Sci 1996; 37:2434–2446. Kueng-Hitz, N, Grimm C, Lansel N, Hafezi F, He L, Fox DA, Reme´ CE, Niemeyer G, Wenzel A. The retina of cfos -/- mice: electrophysiologic, morphologic and biochemical aspects. Invest Ophthalmol Vis Sci 2000; 41:909–916. LaVail MM, Battelle BA. Influence of eye pigmentation and light deprivation on inherited retinal dystrophy in the rat. Exp Eye Res 1975; 21:167–192. Noell WK. Hereditary retinal degeneration and damage by light. In: Estratto dagli Atti Simposio di Oftalmologia Pediatrica. Marccari Parma; 1974:322–329. Mittag TW, Bayer AU, LaVail MM. Light-induced retinal damage in mice carrying a mutated SOD 1 gene. Exp Eye Res 1999; 69:677–683. Chen J, Simon MI, Matthes MT, Yasumura D, LaVail MM. Increased susceptibility to light damage in an arrestin knockout model of Oguchi disease (Stationary night blindness). Invest Ophthalmol Vis Sci 1999; 40:2978-2982. Redmond MT, Yu S, Lee E, Bok D, Hamasaki D, Chen N, Goletz P, Ma J-X, Crouch R, Pfeifer K. Rpe65 is necessary for production of 11-cis-vitamin A in the retinal visual cycle. Nat Genet 1998; 20: 344–351. Grimm C, Wenzel A, Hafezi F, Yu S, Redmond M. Protection of Rpe65-deficient mice identifies rhodopsin as mediator of light-induced retinal degeneration. Nat Genet 2000; 25:63–66. Hobson AH, Donovan M, Humphries MM, Tuohy G, McNally N, Carmody R, Cotter T, Farrar GJ, Kenna PF, Humphries P. Apoptotic photoreceptor cell death
Light-Induced Retinal Degeneration
56.
57.
58.
59.
60.
61. 62.
63.
107
in the rhodopsin knockout mouse in the presence and absence of c-fos. Exp Eye Res 2000; 71:247–254. Heckenlively JR, Rodriguez JA, Daiger SP. Autosomal dominant sectoral retinitis pigmentosa: Two families with transversion mutations in codon 23 of rhodopsin. Arch Ophthalmol 1991; 109:84–91. LaVail MM, Unoki K, Yasumura D, Matthes MT, Yancopoulos GD, Steinberg RH. Multiple growth factors, cytokines, and neurotrophins rescue photoreceptors from the damaging effects of constant light. Proc Natl Acad Sci USA 1992; 89:11249– 11253. Cao W, Li F, Steinberg RH, LaVail MM. Induction of c-fos and c-jun mRNA expression by basic fibroblast growth factor in cultured rat Muller cells. Invest Ophthalmol Vis Sci 1998; 39: 656–573. Organisciak DT, Darrow RM, Bicknell IR, Jiang Y-L, Pickford M, Blanks JC. Protection against retinal light damage by natural and synthetic antioxidants. In: Anderson RE, Hollyfield MM, LaVail MM, eds. Retinal Degenerations. Boca Raton, FL: CRC Press; 1991:189–201. Organisciak DT, Darrow RM, Jiang YL, Marak GE, Blanks JC. Protection by dimethylthiourea against retinal light damage in rats. Invest Ophthalmol Vis Sci 1992 33:1599–1609. Organisciak DT, Wang HM, Li ZY, Tso MOM. The protective effect of ascorbate in retinal light damage of rats. Invest Ophthalmol Vis Sci 1985; 26:1580–1588. Organisciak DT, Bicknell IR, Darrow RM. The effects of L- and D- ascorbic acid administration on retinal tissue levels and light damage in rats. Curr Eye Res 1992; 11:231–241. Organisciak DT, Wang H-M, Xie A, Reeves DS, Donoso LA. Intense-light mediated alterations in rat rod outer segment lipids and proteins. In: LaVail MM, Anderson RE, Hollyfield JG, eds. Inherited and Environmentally Induced Retinal Degenerations. New York: Alan Liss, Inc., 1989:493–512.
7 Retinal Detachment Ward M. Peterson Inspire Pharmaceuticals Durham, North Carolina, U.S.A.
I.
INTRODUCTION
Retinal detachment is an acute pathological endpoint of many ocular conditions and often results in permanent impairment of central vision if the macula is detached. Models of experimental retinal detachment (RD) have been developed to understand the disease process and to provide insight into various aspects of basic retinal biology. For reviews of significant findings made in the context of experimental RD, see Refs. [1–3]. Although naturally occurring retinal detachments have been observed in cats, dogs, and monkeys, these occur too infrequently to be useful as reliable model systems [4–8]. To mimic various aspects of the clinical condition, a number of experimental models of RD have been developed using surgical and, to a lesser extent, pharmacological means. On the cellular level, sustained retinal alterations and photoreceptor degeneration are common outcomes of prolonged detachments (3). The significance of retinopathic changes in causing protracted or permanent visual dysfunction in patients suffering from retinal detachment is well appreciated [3,9]. However, investigations of experimental models of RD in the context of either reducing the severity or enhancing the recovery of retinal function are considerably limited. 109
110
Peterson
II. CLINICAL RETINAL DETACHMENT Based on general etiological parameters, retinal detachments are thought of as belonging to one of three major forms: rhegmatogenous, traction, and serous [10,11]. In the exudative form of age-related macular degeneration (AMD), choroidal neovascularization can lead to leakage of fluid and growth of blood vessels in the subretinal space, which can produce localized macular separations. Models of AMD are beyond the scope of this chapter. Photoreceptors receive oxygen and glucose from the nearby choriocapillaris and even a shallow detachment can result in photoreceptor hypoxia and ischemia [12,13]. Because of their high metabolic activity, photoreceptors are very susceptible to prolonged hypoxic conditions [14]. Whatever the cause of the RD, failure to achieve retinal reattachment in a timely manner often leads to irreversible loss of vision in the affected region. For example, prolonged detachment of the macula results in significant cone alteration or degeneration in the fovea and invariably leads to permanent loss of visual acuity, distortion of color vision, and metamorphopsia, even following successful retinal reattachment [9]. Rhegmatogenous retinal detachment is the most common form of retinal detachment and occurs as a result of tears or holes in the retina. In the older eye, the vitreous loses some of its gel-like characteristics and can become liquefied, and these retinal holes or tears permit liquefied vitreal fluid to enter and accumulate in the subretinal space, thus creating and enlarging the detachment. A variety of surgical techniques (scleral buckle, pneumatic retinopexy, and vitrectomy) are used to reattach the retina. Retinal reattachment surgery often affords an initial success rate of 70 to 90% and a final success rate of higher than 95% following repeat surgeries [15,16]. Scleral buckle surgery is the most common technique for treating rhegmatogenous RD and can result in overall reattachment success rates higher than 90% following a single operation. Thus retinal reattachment surgery is thought to be largely successful. However, success in RD surgery is measured primarily, if not exclusively, in terms of achieving retinal reattachment. Unfortunately, this anatomical success does not parallel recovery in visual function. In a retrospective study, Burton found that of patients treated successfully with a single operative procedure, 42% achieved better than 20/50 visual acuity. For cases involving patients with macula-off detachments, only 20% of reattached macula achieved postoperative visual acuities of 20/50 or better [9]. Although decreased postoperative visual acuity can occur as a result of proliferative vitreoretinopathy, cystoid macular edema, or macular pucker, it is generally thought that degenerative loss of macular cone function is the primary underlying cause of permanent visual acuity impairment and color distortion following “successful” reattachment surgery [9,15]. Tractional forces within the vitreous body or along the inner surface of the retina can exert sufficient mechanical force on the retina to pull it away from the
Retinal Detachment
111
RPE, thereby creating a traction retinal detachment. Traction RD is the second most common form of retinal detachment and is thought to result in part from an inappropriate cellular “wound healing” response along the vitreal-retinal and retinal-RPE interface. Proliferating, hypertrophic, reactive, or otherwise altered glial cells (Mu¨ller cells and astrocytes), RPE, pericytes, and endothelial cells are believed to contribute to the formation of the tractional forces [17]. These cells can proliferate and form a clinically evident membrane structure that contracts and pulls either perpendicularly (via interactions with vitreous strands) or tangentially to the retinal surface. Traction RD occurs in diabetic retinopathy, penetrating trauma (secondary to internal bleeding), retinopathy of prematurity and proliferative vitreoretinopathy [10,11]. It can also lead to permanent visual loss, and is generally treated with vitrectomy, which allows the surgeon to remove the tractional forces. Tractional forces can also contribute to the formation of a retinal tear by pulling a patch of retina from the RPE, usually creating a “horseshoe” tear and increasing the likelihood for the development of rhegmatogenous RD. Diseases and abnormalities of the choroid or RPE can lead to serous or exudative retinal detachment [10]. Serous RD occurs without retinal holes or tears and without apparent tractional forces on the retina, and it is thought in most cases to originate from choroidal fluid leaking into the subretinal space across a compromised RPE [10,18]. Conditions that can lead to serous detachments include central serous retinopathy, severe hypertension, toxemia of pregnancy, Harada’s disease, and various choroidal inflammatory disorders. Treatments for serous RD are generally nonsurgical. Many aspects of rhegmatogenous, traction, and serous RD are studied by employing an experimental model of non-rhegmatogenous retinal detachment, which has been used extensively to study photoreceptor degeneration, retinal alterations, and, more recently, to test potential therapeutic approaches for preventing retinopathic changes. This model is described below.
III. EXPERIMENTAL NON-RHEGMATOGENOUS RETINAL DETACHMENT A common method for experimentally creating a retinal detachment is to introduce a small needle or micropipette in the anterior globe (usually through the pars plana), advance the needle or micropipette through the vitreous and into the retina, and then gently inject fluid into the subretinal space (Fig. 1A). The resultant detachment is best described as a non-rhegmatogenous retinal detachment— non-rhegmatogenous because the retinal hole caused by the needle is sufficiently small to prevent shunting of subretinal and vitreal fluid [36]. Experimental nonrhegmatogenous RD is sometimes referred to as a “subretinal bleb” or “serous retinal detachment,” although it is important to recognize that this model does
112
Peterson
Figure 1 Diagrammatic representation of techniques used to create experimental non-rhegmatogenous and rhegmatogenous retinal detachments. (A) The micropipette or small needle (usually 32-G) is inserted through the pars plana, slowly advanced into the vitreous, and into the retina. A small outflow is maintained across the needle tip to prevent obstruction by the vitreal gel, but this is usually done with very small diameter micropipette tips. After creation of the subretinal bleb and retinal separation, the micropipette or needle is retracted, leaving a hole that either seals on its own (perhaps through localized blood-clotting) or a hole that is otherwise too small for exchange of subretinal and intravitreal fluid. (B) In preparation for the creation of a retinal hole, the vitreous is collapsed by aspirating and reinjecting hyaluronidase with the needle that is inserted intravitreally through the pars plana (top panel). The retinal hole and surrounding detachment are subsequently created by pressure injection of vitreous against the retina (middle and bottom panels), resulting in a rhegmatogenous retinal detachment [35].
Retinal Detachment
113
not formally mimic many aspects of a true serous RD. The model is used to study many aspects of retina/RPE interactions, including rates of subretinal fluid reabsorption, forces of retinal adhesion, metabolic and pharmacological effects on the RPE fluid pump, and subcellular and cellular effects on the retina and RPE [1,19–23]. Such retinal detachments are also invariably produced when delivering compounds, proteins, and viral vectors (for gene-based therapy) into the subretinal space [24,25]. More recently, large bullous or complete non-rhegmatogenous retinal detachments are deliberately produced in the clinic as part of an evolving surgical technique called macular translocation surgery for the treatment of age-related macular degeneration [26]. In this approach, the entire macula is detached from the underlying diseased RPE and moved to a presumably healthier region of the RPE. Experimental models of non-rhegmatogenous RD have been produced in rabbit, cat, monkey, and, more recently, rat, ground squirrel, and pig [23,27–33]. A. Preoperative Procedure Animals undergo general anesthesia in preparation for retinal detachment surgery. A variety of anesthetic combinations have been used for rabbits (singly or in various combinations): thorazin, pentobarbital, urethane, ketamine, and acepromazine maleate. Cats: combinations of ketamine and sodium pentobarbital; or ketamine and acepromazine maleate. Monkeys: combinations of ketamine and xylazine; phencyclidine hydrochloride and sodium pentobarbital; or sodium pentobarbital alone. Rats: a combination of buprenophrine, sodium pentobarbital, and atropine. Topical anesthesia is administered via retrobulbar injection of xylocaine in rabbits and cats. To visualize the injection and fundus, pupils are dilated using (singly or in various combinations): phenylephrine, cyclopentolate, tropicamide, and atropine. Some of the pioneering or prototypical models for experiment RD can be found in these references: [23,30,31,34,35]. B. Surgical Procedure: Retinal Detachment Modifications of a basic surgical procedure are used for preparing the eye and creating a retinal detachment; these approaches often depend on the species and the investigator. Delineation of the myriad of surgical approaches used will not be attempted here, but a number of general methodologies are discussed instead. In rabbits, the superior sclera is penetrated with a needle (usually 22-G) 3–4 mm behind the limbus to avoid inserting through the retina, and a glass micropipette (15–50 µm tip diameter) or small needle (32-G or smaller) is inserted through the scleral opening and advanced into the vitreous under the fine control of a micromanipulator [23,36]. For micropipettes, a small outflow of fluid can be maintained across the tip by applying air pressure at 10–20 lb/in.2 to prevent
114
Peterson
vitreous gel from obstructing the tip. The tip is advanced until the retina is gently penetrated and a detachment begins to form. Small and large subretinal blebs can be made this way. In cats, a number of modifications of a basic surgical technique for inducing large retinal detachments have been reported [13,30,31]. The procedure begins with a pre-detachment preparation phase in which an extracapsular lens extraction is performed through an 180° corneal incision. This is followed by excision of the posterior capsule and a partial vitrectomy or, alternatively, the posterior capsule can be left intact. The vitrectomy affords better control of the height and extent of the induced RD. The cornea is sutured closed and allowed to heal for several weeks before the actual retinal detachment is created. For RD surgery, an infusion cannula is sewn in place in the cornea (for example, in the inferotemporal quadrant). A puncture incision is then made with a 20-G needle in an adjacent quadrant (superotemporal) of the cornea, and the posterior capsule and vitreous are removed. Fluid-gas exchange is performed following the vitrectomy. Alternatively, the lens and vitreous can be left in place [13]. A glass micropipette with a flat 80–100 µm tip diameter is mounted on a micromanipulator and then inserted into the 20-G incision. The retinal detachment can then be made in a manner similar to that used in rabbits. Variations of this technique have been adapted for use in rhesus monkeys [28]. A rat model of retinal detachment has recently been developed and used for subretinal fluid reabsorption studies [33]. Retinal detachments were made in Long-Evans rats by first inserting a 26-G guidance needle behind the limbus and into the vitreous. A 33-G flat-tip needle attached to a Hamilton syringe is then inserted into the barrel of the 26-G guidance needle to create the detachment and to inject solutions into the vitreous. A specially designed double convex lens for the rat eye is placed on the cornea to view injections and the fundus. Although the 26-G needle used to penetrate the rat eye may seem rather large relative to the size of the eye, this approach does not appear to cause significant changes in intraocular pressure or other complications and seems amenable for studying retinal detachments over a 24-h period. C.
Surgical Procedure: Retinal Reattachment
In some cases, it is necessary to reproducibly induce retinal reattachment. In cats and monkeys, a gas/air exchange procedure has been developed to induce retinal reattachment and theoretically should be applicable for use in other species with eyes of similar or larger sizes [28,31]. In this approach, animals are first given general anesthesia. The eye is then prepared for a mixture of sulfur hexafluoride (SF6) gas and air exchange via an infusion cannula and a drainage cannula at different quadrants of the globe. In the cat, the infusion cannula is placed in the inferotemporal quadrant of the cornea approximately 1.5 mm from the limbus.
Retinal Detachment
115
(The infusion cannula can be inserted through a 20-G incision, for example, and secured with 5.0 Dacron mattress suture.) A second 20-G incision is then made in the superotemporal quadrant, in which a 20-G blunt beveled tip on a Charles fluted needle is inserted. This allows for a mixture of 50% or 75% SF6 gas and air to be flushed through the eye until complete exchange is achieved. D. Solutions A variety of buffers and Ringer’s solutions have been used to experimentally induce retinal detachments. Table 1 shows the effects of some of these buffers and solutions on the duration of detachments made in a number of species. The results shown in Table 1 were taken from various published reports and may represent laboratory-specific findings. Other factors such as height and extent of the induced RD are also important factors in affecting the duration of the detachment. Reattachment of experimental non-rhegmatogenous RD usually occurs over a period of hours to days, assuming that a physiologically based solution is used to create the detachment. Rabbits and rats appear to spontaneously reattach over a period of hours, and cats and monkeys can take many hours or even days [1,27,30,36]. For studies of prolonged, controllable, and semi-reproducible retinal detachments, a useful technique has been developed that enables the researcher to create detachments of fixed lengths of time and consists of adding sodium hyaluronate in the infusion solution; this prevents the retina from reat-
Table 1 Approximate Duration of Non-Rhegmatogenous Retinal Detachments Following Subretinal Injection of Different Solutions in Various Species Species Rhesus monkeys Rhesus monkeys Rhesus monkeys Rhesus monkeys Cat Cat New Zealand rabbits New Zealand rabbits Rat
Solutions Balanced salt saline (BSS) Glutathione-bicarbonate Ringer’s solution Autologous serum Silicone oil Balanced salt saline BSS ⫹ 0.25% Na hyaluronate 1.5% solution of Na hyaluronate Ames solution Modified phosphate buffered saline
Detachment period ⬃7 days Complete reabsorption, 2–7 days Gradual reabsorption, ⬎3 weeks No reabsorption ⬍3 days No reabsorption No reabsorption, at least up to 29 days Few hours At least 24 hours
116
Peterson
taching [22,30,55]. In cats, for example, retinal detachments of up to 14 months in duration have been made with this approach [30]. Furthermore, retinal reattachment can be achieved by using the gas/fluid exchange technique described above. In the aforementioned rat model, it was noted that the constancy of the apparent subretinal bleb size over time depended on the relative osmolarity of the blood serum, thus suggesting that osmotic differences across the retina and RPE play a significant role in determining the rates of subretinal fluid reabsorption [33]. The investigators exploited this finding to devise a modified phosphate buffered saline solution that allowed subretinal blebs to remain relatively constant, at least for an initial 24-h period. This control over the constancy in bleb size has been useful, for example, in evaluating pharmaceutical approaches for enhancing subretinal fluid reabsorption [33]. E.
Species Considerations
There are significant anatomical differences between human eyes and the eyes of various species used for experimental RD studies, and relevant differences should be taken into account when interpreting the results of animal studies in the context of human retinal detachments. Rabbit retinas are rod-dominant, lack a cone-rich macula, and contain extremely limited intrinsic blood vessels that reside along the nasal-to-temporal myelin wing. The rabbit retina is thought to receive virtually all of its oxygen from the choriocapillaris. Thus a detachment would render the entire retina hypoxic, rather than merely the outer retina. The retinal vascular bed in the cat more closely resembles that of humans, and although the cat retina contains a higher population of cones than the rabbit retina, it is nevertheless rod-dominant and also lacks a cone-dominant macula. Rhesus and cynomolgus monkeys contain true cone-dominant macula and would represent ideal animal models for human RD. The retinas of ground squirrels contains 85 to 90% cones, their color vision is dichromatic, and their cone system has been well characterized electrophysiologically. The vascular bed is extensive, and so squirrels appear to be an attractive alternative to monkeys for RD research. Rats also contain an extensive vascular bed, but the retina is rod-dominant (⬎95%) and lacks a macular region. Finally, the porcine eye, which also does not contain a true cone-dominant macula, appears to be an attractive model for studying neurodegeneration and neuroprotection in the context of RD. The size of the eye, the retinal vasculature system, lens/vitreal volume ratio, and the presence of a cone-enriched macula (or visual streak) give strong credence for studying cone alterations in this species. F.
Relevance of Model
There were some early concerns about whether the retinal hole created by the small needle or micropipette was sufficiently large to create a true shunt for
Retinal Detachment
117
subretinal and vitreal fluid. If so, the model could no longer be thought of as “non-rhegmatogenous” and the ability to use the model to estimate RPE fluid absorption, for example, is questionable. To address this concern, Marmor and colleagues showed that in rabbits the creation of retinal holes with a 15–25 µm diameter micropipette had no effect on the rates of subretinal fluid reabsorption [36]. Sealing the hole with cyanoacrylate, mucilage, or an air bubble did not affect the rate of subretinal fluid reabsorption, nor did the creation of multiple holes in the same bleb. Pederson and MacLellan showed that in rhesus monkeys, the creation of a subretinal bleb containing a small retinotomy resulted in spontaneous retinal reattachment, whether or not the eye had a vitrectomy [27,34]. In rats, Mamanishkis and colleagues showed that fluorescein injected into the subretinal space did not diffuse into the vitreous through the hole [33]. These results all provide strong experimental support that small retinal holes do not create a shunt for subretinal and vitreal fluid, hence validating the concept that such retinal detachments are non-rhegmatogenous in nature. In humans, detachments enlarge over a period of hours to days, whereas in the lab, detachments are usually created in less than a minute. Rhegmatogenous RD in an untreated human eye often progresses to total detachment. In the absence of a significant retinal hole or tear, experimental non-rhegmatogenous RD tends to reattach gradually over a period of hours to a few days, and thus requires additional constituents (such as hyaluronic acid) in the buffer formulations used for subretinal injections to maintain prolonged retinal separation. Because clinical retinal detachments tend to occur in an older eye, the state of the human vitreous is significantly different from that of the experimental models. The vitreal collagen framework collapses in the aging eye, resulting in syneresis, liquefaction, and pooling of fluid within the vitreous gel. Gel liquefaction is highly correlated with posterior vitreous detachments, which in turn sets up tractional forces on areas of vitreoretinal adhesion [37]. Such traction on areas of vitreoretinal adhesion may cause retinal tears and subsequent retinal detachment. Thus, some of the important differences between RD in humans and in animal models are the origin and context of the diseased condition, particularly the role of the vitreous. Despite these limitations, a number of important findings made in the lab have demonstrable correlative findings in humans, and have yielded further insight into the pathology of RD above and beyond what might be expected from clinical or patho-clinical studies alone. Some of these findings are described below. G.
Retinal Changes
For an excellent review of alterations in retina and RPE associated with experimental retinal detachment and reattachment, see Ref. 3. Research in animal models of RD has focused considerably more on rods than cones, and very little of the work on cones is performed on the cone-dominant macula. Of the animal
118
Peterson
models employed in experimental studies, only certain primate species (including the rhesus macaques) contain a cone-dominant macula. The retina of the nocturnal owl monkey, which has been used in a series of RD studies, contains a macular region that is primarily made up of rods [38]. Rod and cone outer segments exhibit a number of well-appreciated differences in terms of their structural and biochemical interactions with the interphotoreceptor matrix and the RPE apical membrane [39,40]. Recent studies provide some evidence that rods and cones respond differently to retinal detachment and reattachment. Surviving rods continue to express high levels of rhodopsin and other cell-specific markers in a detached retina, whereas cones respond with a rapid reduction in a variety of protein and cellular markers, such as short wavelength (S)-cone opsin, calbindin D, carbonic anhydrase, peanut agglutinin, and possibly medium to long (M,L)-cone opsin [3,41– 43]. Labeling of apoptotic cell death with the TUNEL (TdT-mediated duTPbiotin nick end labeling) technique demonstrated that both rods and cones die during retinal detachments [42,44,45]. However the significance of detachmentmediated reductions in measurable expression of cone-specific markers remains elusive in the context of cone cell death. That is, reductions in the expression of these markers do not necessarily indicate inexorable death of cones. Further investigation into the time-course and extent of recovery of these markers needs to be conducted to appreciate the correlation, if any, between the decrease in cone-specific cell markers and cone death [42]. There is some experimental and clinical evidence that S-cones are more susceptible to prolonged or irreversible damage than M,L-cones in both human and experimental retinal detachment. Distortion in blue-yellow color vision has been described as a relatively common visual defect following successful retinal reattachment. Using enzyme histochemistry for carbonic anhydase and immunocytochemical localization of S antigen to differentially identify blue cones and red/green cones, Nork and colleagues provided evidence for increased loss of blue cones in human rhegmatogenous RD [43]. Clinical electroretinography (ERG) has also provided some insight into alterations in cone-mediated function in patients. Cone ERG results from 19 patients taken prior to and at multiple time points following successful retinal reattachment surgery provide evidence that the capacity for S-cone b-wave amplitudes to recover is significantly reduced when compared with those of L- and M-cone b-wave amplitudes [46,47]. In another study, multifocal ERG recordings taken from detached and attached areas of the retina before retinal reattachment surgery were compared with similar recordings taken from the same areas after successful reattachment surgery [48]. These ERG recordings were then compared with corresponding visual field measures. The authors found that improvement in ERG responses from both areas were relatively modest and surprisingly did not correlate with the more significant improvements in visual field results. Taken together, these ERG recordings in human RD point to the utility of conventional full-field as well as multifocal
Retinal Detachment
119
ERG in extracting information about retinal function, but also suggest that ERG parameters may not correlate with visual function. There have been some published reports on experimental RD that have focused on scotopic and photopic ERG responses during detachment and following reattachment [48–52]. Kim and colleagues recorded photopic focal ERG from small retinal detachments in adult rabbits and showed that b-wave amplitudes decreased significantly 30 min following retinal detachment and recovered back to baseline 3 days later [52]. However, immediately following the detachment, b-wave implicit times increased and remained significantly higher than control eyes even up to 28 days following reattachment. Because reattachment in rabbits occurs within a few hours following the induction of a small, bullous detachment, these results indicate that changes in retinal cone function persisted for weeks following retina/RPE reapposition and that photopic b-wave implicit time may be a useful parameter to assess loss and recovery of cone function. ERG recordings made from eyes of ground squirrels with RD of varying sizes showed a clear depression in cone-mediated ERG function in the detachment zone [77]. However, unlike observations made in human RD, no apparent differences in Scone and M-cone ERG amplitudes were observed in squirrel. The reasons for this discrepancy remain unresolved, but may be a result of species differences or differences in experimental versus surgical conditions [77]. Although most of the basic and clinical research on retinal detachments have focused on changes in photoreceptors, significant alterations in the inner retina and RPE have also been observed and are believed to play a significant role in the aptly named “retinopathy of detachment” [3]. One of the leading causes of failure in retinal detachment surgery is the development of proliferative vitreoretinopathy, which is thought to be mediated by de-differentiated, proliferating, and hypertrophic non-neuronal cells, including retinal astrocytes, Mu¨ller, and RPE cells. The proliferative response can lead to eventual contraction of cells on the vitreal surface of the retina and result in traction retinal detachment. Understanding the causes of proliferative vitreoretinopathy may provide some insight into ways of preventing its development after successful reattachment surgery [15,17].
IV. POTENTIAL THERAPEUTIC OPPORTUNITIES A variety of experimental parameters have been employed to quantitatively or semiquantitatively define the retinopathic effects of experimental retinal detachment and to measure the efficacy of a potential therapeutic approach. A number of these parameters are listed here: Immunohistochemical labeling of rhodopsin and cone-opsin Quantification of cell proliferation with MIB-1 antibody
120
Peterson
Labeling of Mu¨ller cells using antibodies directed against glial fibrillary acidic protein (GFAP), cellular retinaldehyde binding protein, and glutamine synthetase Glutamate receptor GluR-2 immunohistochemical labeling Outer nuclear layer thickness and cell counts Outer segment length Quantification of apoptosis using the TUNEL technique Full-field and multi-focal ERG Scotopic and photopic ERG In the context of experimental RD, it has been only in recent years that studies have been conducted to investigate new approaches for either reducing the severity of retinopathic changes or improving retinal recovery following reattachment. Such experimental or neuroprotective therapies aimed at treating retinal changes in the context of RD are not as well defined as those strategies developed in the context of experimental models of light-induced and inherited photoreceptor degeneration [53]. Retinal detachments and photoreceptor degenerative diseases share a number of similar features, including time-dependent alterations in Mu¨ller cells (reactive gliosis and hypertrophy), photoreceptor cells (particularly synaptic changes), and ultimately apoptosis as the mechanism of photoreceptor cell death [54]. A number of the same survival factors (ciliary neurotrophic factor, brain-derived neurotrophic factor, pigment epithelium-derived factor, glial-derived neurotrophic factor, basic fibroblast growth factor, etc.) that enhance photoreceptor survival and function in the context of experimental models of photoreceptor degeneration are likely to have a similar effect in experimental RD. Recent work has shown that brain-derived neurotrophic factor (BDNF) and glial-derived neurotrophic factor (GDNF) can mitigate the progression of a number of pathological retinal changes associated with experimental retinal detachments in the cat [55,56]. Lewis and colleagues showed that BDNF enhanced the organization and increased the length of outer segments and significantly reduced the proliferative changes of Mu¨ller cells while the retina was detached for 7 or 28 days. However, BDNF failed to reduce the overall photoreceptor cell death in this model. In a separate study, intravitreal GDNF was shown to increase the mean outer segment lengths in the detached retina when compared with control detachments [56]. A potentially important difference between models of photoreceptor degenerative diseases and experimental RD in the context of retinal neuroprotection may be the role of supplemental oxygen. In a pair of recently published studies, cat retinas were detached for three days, during which some animals were housed under hyperoxic conditions (70% ambient oxygen) and remaining animals were housed under normoxic conditions (21% ambient oxygen) [13,57]. Oxygen sup-
Retinal Detachment
121
plementation was shown to promote the survival of photoreceptor cells, enhance the organizational structures of their outer segments, reduce Mu¨ller cell proliferation and hypertrophy, and normalize various aspects of the glutamate signaling system. These studies elegantly illustrate the importance of oxygen supplementation in decelerating a number of photoreceptor and retinal degenerative processes in RD. More recently, work in a ground squirrel model demonstrated that hyperoxia also rescued cone photoreceptors in this cone-dominant species, suggesting that survival of cones with hyperoxia is not necessarily secondary to survival of rods [58]. These studies on the effects of hyperoxia on photoreceptor survival provide a strong rationale for testing oxygen supplementation for enhancing visual outcome associated with the treatment of retinal detachments in the clinic. Two factors play important roles in recovery of central vision following macula-off detachments: preoperative visual acuity and duration of the detachment [9]. In the absence of significant degeneration of the fovea, preoperative visual acuity is correlated with the height of the fovea from the RPE. Therefore, these clinical observations suggest that the ability to reduce the distance between the fovea and RPE and to speed retinal reattachment may improve visual outcome following reattachment. The RPE normally absorbs fluid in the subretinal-tochoroidal direction, and this RPE fluid “pump” function can be enhanced in vitro and in vivo by pharmacological means [1,59,60]. Thus, a pharmacological approach that stimulates the RPE pump could facilitate the removal of extraneous fluid in the subretinal space in RD and may provide a therapeutic approach for reducing the spread and height of the retinal separation. If sufficiently robust, pharmacological stimulation may reattach the retina without surgery; alternatively, it may be used as a surgical adjunct to minimize the spread of the detachment or facilitate the rate of reattachment. Carbonic anhydrase inhibitors acetazolamide and benzolamide have been shown to enhance subretinal fluid reabsorption in experimental subretinal blebs, presumably by activating the RPE fluid pump, and thereby facilitate retinal reattachment in rabbits [61,62]. However, acetazolamide is poorly tolerated, thus pointing to the need for alternative pharmacological means for stimulating the RPE pump. The apical membrane of bovine RPE contains metabotropic P2Y2 receptors that respond to the natural ligands adenosine 5′-triphosphate (ATP) and uridine 5′-triphosphate (UTP) by activating membrane transport mechanisms and stimulating net apical-to-basolateral fluid absorption in vitro [63]. A synthetic P2Y2 receptor agonist INS37217, when delivered subretinally or intravitreally, has recently been shown to significantly enhance subretinal fluid reabsorption and retinal reattachment following experimental RD in rabbit [64] and rat [33]. Recent work in a mouse model of experimental RD also showed that INS37217 significantly enhanced the recovery rate of retinal function (as determined using standard ERG techniques) following retinal reattachment [78]. In the clinic, a pharmacological agent such as INS37217 that limits the spread of retinal detachments
122
Peterson
and enhances retinal reattachment may provide a novel strategy for improving visual outcome following surgery.
V.
MODELS OF RHEGMATOGENOUS, SEROUS, AND TRACTION RETINAL DETACHMENT
Machemer, Aaberg, and Norton developed an experimental model of rhegmatogenous retinal detachment in the owl monkey [8,29,35]. They defined the conditions that are necessary to produce and sustain long-standing rhegmatogenous RD, and concluded that alterations of the vitreous structure and creation of a relatively large retinal hole are required. Their basic technique is described in Figure 1B. They compared the histological effects of experimental verses naturally occurring retinal detachments in the owl monkey and found many similar features, including cystoid degeneration of the outer plexiform layer, degeneration of the outer and inner nuclear layers, and appearance of flat and multi-nucleated RPE cells [8]. Their model produced cystoid spaces and edema in the inner retina that developed and enlarged with time, and subsequently reabsorbed and disappeared following surgical reattachment of the retina [28]. These cystoid spaces were much larger and prominent than those reported from the experimental models of non-rhegmatogenous RD described by other investigators and do not appear to be artifacts of histological processing. It is unclear if these findings are species dependent, a result of surgical manipulation, or caused by the large retinal hole, but the observations of cystoid spaces and retinal edema are clinically relevant because they are found in long-standing human retinal detachments. Pederson and colleagues published a series of papers in 1982 to 1986 that examined, among other things, some of the technical differences used in various approaches for creating rhegmatogenous versus non-rhegmatogenous RD [27, 34,65–67]. They were able to produce long term retinal detachments in rhesus monkeys by inducing a large bullous detachment (1 mL of 20% autologous serum in Ringer’s solution injected into the subretinal space), creating a large 2–3 mm retinal hole with a hooked tip of a 25-G needle, and following up with a subtotal vitrectomy. Without the vitrectomy, retinal detachments tended to spontaneously reattach [27]. In their model of rhegmatogenous RD, they found a transient and marked decrease in intraocular pressure and an increase in the rate of fluorescein disappearance from the vitreous. Based on these and other findings, they concluded that there is a posterior-directed flow of fluid through the retinal hole and that the RPE fluid pump function is increased. Many models of traction RD have been developed using different methodologies [2,17,68–71]. A reproducible model of traction RD in the cat was created by first inducing a serous retinal detachment using the rose bengal method (see next paragraph), and then injecting dermal fibroblasts into the vitreous [68]. Trac-
Retinal Detachment
123
tion RD was produced within the first 2 weeks after fibroblast injection, with characteristics of vitreoretinal strands and epiretinal proliferation. Retinal degeneration was observed in this model. The causes of the retinal degeneration may be attributed to multiple origins, including the associated photodynamic treatment for inducing the serous detachment, vascular thrombosis, and the prolonged detachment itself. Experimental traction RD was also observed following injection of autologous fibroblasts into the vitreous cavity of rabbits, and induction of perforating ocular injury in rabbits, pigs, and monkeys [69–71]. Serous retinal detachment is difficult to model. A reproducible model for serous RD utilizes intravenous administration of the dye rose bengal, which immediately photosensitizes the choroid and RPE and can be photochemically activated in the eye using a xenon photocoagulator filtered to a central wavelength of approximately 550 nm [72]. Adjusting the irradiance, duration of light exposure, and the dose of the rose bengal can control the peak and size of the resultant serous detachment. In the cat, Wilson and colleagues showed that serous retinal detachments as small as a focal retinal bleb (lasting 1 h) to a massive bullous detachment comprising the entire retina (lasting 4 months) were induced with this model. Considerable ischemic degeneration of the detached retina was observed in the region of the photothrombosis at 3 days following photodynamic injury, but the retina outside this region appeared relatively undamaged. For detachments that lasted for at least 7 days, there was a complete loss of photoreceptor outer segments, and for detachments that lasted at least 14 days, the neural retina was thin and gliotic with loss of the outer layers. Significant occlusion of blood vessels in the retina was also observed after 14 days of detachment. The need to photosensitize the choroid and induce significant photoactivated damage to the retina, RPE, and choroid indicate major disparities between the model and clinical serous RD. Experimental serous RD has also been created following occlusion of choroidal circulation with microsphere embolization and in experimental malignant hypertension [73–76]. To my knowledge, no systematic studies of experimental serous, traction, or rhegmatogenous retinal detachments have been conducted in the context of preventing retinal degeneration.
VI. CONCLUSION Many experimental models of retinal detachment have been developed in the past few decades, and a number of them have been used to carefully define and understand various aspects of retinal alteration, remodeling, and degeneration. Quantitative and semiquantitative descriptions of retinal changes in experimental retinal detachment are important because they provide the investigator with measurable parameters to evaluate neuroprotective and other vision-salvaging strategies. There remains a significant medical need to enhance visual outcomes fol-
124
Peterson
lowing anatomically successful retinal reattachment procedures. Because macula-off detachments are more frequently seen in the clinic and because loss of central visual acuity and color distortion are common following successful macular reattachment surgery, a better understanding of the cellular and electrophysiological effects of experimentally detached macula is needed. In Burton’s study, only 2% of patients operated on within 5 days of macular detachment were able to achieve postoperative visual acuity of 20/20 [9]. This 2% figure is illuminating: it tells us that in macula-off detachments it is indeed possible to restore visual acuity back to normal. The researcher is therefore motivated to find ways of utilizing experimental models of retinal detachment to develop new therapies that enable the capacity of foveal cones to recover fully. However, the remaining figure of 98% is daunting—it is a measure of the challenges that lay ahead. ACKNOWLEDGMENT I would like to thank Drs. Steve Fisher and Sheldon Miller for their valuable input and feedback on this chapter. REFERENCES 1. 2. 3.
4. 5. 6.
7. 8.
9.
Marmor MF. Control of subretinal fluid: experimental and clinical studies. Eye 1990; 4:340–344. Martini B. Proliferative vitreo-retinal disorders: experimental models in vivo and in vitro. Acta Ophthalmol Suppl 1992; 201:1–63. Fisher SK, Stone J, Rex TS, Linberg KA, Lewis GP. Experimental retinal detachment: a paradigm for understanding the effects of induced photoreceptor degeneration. Prog Brain Res 2001; 131:679–698. Roberts SR. Detachment of the retina in animals. J Am Vet Med Ass 1959; 135: 423–435. MacMillan AD. Acquired retinal folds in the cat. J Am Vet Med Assoc 1976; 168: 1015–1020. Blair NP, Dodge JT, Schmidt GM. Rhegmatogenous retinal detachment in Labrador retrievers. I. Development of retinal tears and detachment. Arch Ophthalmol 1985; 103:842–847. Blair NP, Dodge JT, Schmidt GM. Rhegmatogenous retinal detachment in Labrador retrievers. II. Proliferative vitreoretinopathy. Arch Ophthalmol 1985; 103:848–854. Aaberg TM, Machemer R. Correlation of naturally occurring detachments with long-term retinal detachment in the owl monkey. Am J Ophthalmol 1970; 69:640– 650. Burton TC. Recovery of visual acuity after retinal detachment involving the macula. Trans Am Ophthalmol Soc 1982; 80:475–497.
Retinal Detachment 10. 11. 12.
13.
14. 15. 16. 17. 18. 19. 20. 21.
22.
23.
24.
25.
26. 27. 28.
125
Anand R, Tasman WS. Nonrhegmatogenous retinal detachment. In: Glaser BM, ed. Retina. Vol. 3. St. Louis: Mosby, 1994:2463–2488. Hay A, Lander MB. Types of pathogenetic mechanisms of retinal detachment. In: Glaser BM, ed. Retina. Vol. 3. St. Louis: Mosby, 1994:1971–1977. Winkler B. The intermediary metabolism of the retina: biochemical and functional aspects. In: Anderson RE, ed. Biochemistry of the Eye. San Francisco: American Academy of Ophthalmology 1983:227–242. Mervin K, Valter K, Maslim J, Lewis G, Fisher S, Stone J. Limiting photoreceptor death and deconstruction during experimental retinal detachment: the value of oxygen supplementation. Am J Ophthalmol 1999; 128:155–164. Linsenmeier RA, Braun RD. Oxygen distribution and consumption in the cat retina during normoxia and hypoxemia. J Gen Physiol 1992; 99:177–197. Wilkinson CP, Rice TA. Results of retinal reattachment surgery. In: Craven L, ed. Michels Retinal Detachment. St. Louis: Mosby, 1997:935–972. The repair of rhegmatogenous retinal detachments. American Academy of Ophthalmology. Ophthalmology 1990; 97:1562–1572. Ryan SJ. Traction retinal detachment. XLIX Edward Jackson Memorial Lecture. Am J Ophthalmol 1993; 115:1–20. Marmor MF, Yao XY. Conditions necessary for the formation of serous detachment. Experimental evidence from the cat. Arch Ophthalmol 1994; 112:830–838. Marmor MF, Yao XY. The metabolic dependency of retinal adhesion in rabbit and primate. Arch Ophthalmol 1995; 113:232–238. Negi A, Kawano S, Marmor MF. Effects of intraocular pressure and other factors on subretinal fluid resorption. Invest Ophthalmol Vis Sci 1987; 28:2099–2102. Anderson DH, Stern WH, Fisher SK, Erickson PA, Borgula GA. The onset of pigment epithelial proliferation after retinal detachment. Invest Ophthalmol Vis Sci 1981; 21:10–16. Berglin L, Algvere PV, Seregard S. Photoreceptor decay over time and apoptosis in experimental retinal detachment. Graefes Arch Clin Exp Ophthalmol 1997; 235: 306–312. Marmor MF, Abdul-Rahim AS, Cohen DS. The effect of metabolic inhibitors on retinal adhesion and subretinal fluid resorption. Invest Ophthalmol Vis Sci 1980; 19:893–903. LaVail MM, Unoki K, Yasumura D, Matthes MT, Yancopoulos GD, Steinberg RH. Multiple growth factors, cytokines, and neurotrophins rescue photoreceptors from the damaging effects of constant light. Proc Natl Acad Sci USA 1992; 89:11249– 11253. Flannery JG, Zolotukhin S, Vaquero MI, LaVail MM, Muzyczka N, Hauswirth WW. Efficient photoreceptor-targeted gene expression in vivo by recombinant adeno-associated virus. Proc Natl Acad Sci USA 1997; 94:6916–6921. Machemer R. Macular translocation. Am J Ophthalmol 1998; 125:698–700. Pederson JE, Cantrill HL, Cameron JD. Experimental retinal detachment. II. Role of the vitreous. Arch Ophthalmol 1982; 100:1155–1159. Guerin CJ, Anderson DH, Fariss RN, Fisher SK. Retinal reattachment of the primate macula. Photoreceptor recovery after short-term detachment. Invest Ophthalmol Vis Sci 1989; 30:1708–1725.
126
Peterson
29.
Machemer R, Norton EW. Experimental retinal detachment and reattachment: I. Methods, clinical picture and histology. Bibl Ophthalmol 1969; 79:80–90. Anderson DH, Stern WH, Fisher SK, Erickson PA, Borgula GA. Retinal detachment in the cat: the pigment epithelial-photoreceptor interface. Invest Ophthalmol Vis Sci 1983; 24:906–926. Anderson DH, Guerin CJ, Erickson PA, Stern WH, Fisher SK. Morphological recovery in the reattached retina. Invest Ophthalmol Vis Sci 1986; 27:168–183. Linberg KA, Lewis GP, Charteris DG, Fisher SK. Experimental detachment in a cone dominant retina. Invest Ophthalmol Vis Sci 1999; 40:S951. Maminishkis A, Jalickee S, Blaug SA, Rymer J, Yerxa BR, Peterson WM, Miller SS. The P2Y2 receptor agonist INS37217 stimulates RPE fluids transport in vitro and retinal reattachment in rat. Invest Ophthalmol Vis Sci 2002; 43:3555–3566. Pederson JE, MacLellan HM. Experimental retinal detachment. I. Effect of subretinal fluid composition on reabsorption rate and intraocular pressure. Arch Ophthalmol 1982; 100:1150–1154. Machemer R, Norton EW. Experimental retinal detachment in the owl monkey. I. Methods of production and clinical picture. Am J Ophthalmol 1968; 66:388–396. Marmor MF, Porteus M, Negi A, Immel J. Validation of a model of non-rhegmatogenous retinal detachment. Curr Eye Res 1984; 3:515–518. Wilkinson CP, Rice TA. Vitreoretinal precursors of retinal detachment. In: Craven L, ed. Micels Retinal Detachment. St. Louis: Mosby, 1997. Hamasaki DI. An anatomical and electrophysiological study of the retina of the owl monkey, Aotes trivirgatus. J Comp Neurol 1967; 130:163–174. Johnson LV, Hageman GS, Blanks JC. Interphotoreceptor matrix domains ensheath vertebrate cone photoreceptor cells. Invest Ophthalmol Vis Sci 1986; 27:129– 135. Fisher SK, Steinberg RH. Origin and organization of pigment epithelial apical projections to cones in cat retina. J Comp Neurol 1982; 206:131–145. Fariss RN, Molday RS, Fisher SK, Matsumoto B. Evidence from normal and degenerating photoreceptors that two outer segment integral membrane proteins have separate transport pathways. J Comp Neurol 1997; 387:148–156. Linberg KA, Lewis GP, Shaaw C, Rex TS, Fisher SK. Distribution of S- and Mcones in normal and experimentally detached cat retina. J Comp Neurol 2001; 430: 343–356. Nork TM, Millecchia LL, Strickland BD, Linberg JV, Chao GM. Selective loss of blue cones and rods in human retinal detachment. Arch Ophthalmol 1995; 113: 1066–1073. Cook B, Lewis GP, Fisher SK, Adler R. Apoptotic photoreceptor degeneration in experimental retinal detachment. Invest Ophthalmol Vis Sci 1995; 36:990–996. Chang CJ, Lai WW, Edward DP, Tso MO. Apoptotic photoreceptor cell death after traumatic retinal detachment in humans. Arch Ophthalmol 1995; 113:880–886. Yamamoto S, Hayashi M, Takeuchi S. Cone electroretinograms in response to color stimuli after successful retinal detachment surgery. Jpn J Ophthalmol 1998; 42:314– 317. Hayashi M, Yamamoto S. Changes of cone electroretinograms to colour flash stimuli after successful retinal detachment surgery. Br J Ophthalmol 2001; 85:410–413.
30.
31. 32. 33.
34.
35. 36. 37. 38. 39.
40. 41.
42.
43.
44. 45. 46.
47.
Retinal Detachment 48. 49. 50.
51.
52. 53. 54. 55.
56.
57.
58.
59.
60.
61. 62.
63.
64.
65.
127
Sasoh M, Yoshida S, Kuze M, Uji Y. The multifocal electroretinogram in retinal detachment. Doc Ophthalmol 1997; 94:239–252. Kamei S. [The recovery of the local ERG recorded from reattached retina after retinal detachment]. Nippon Ganka Gakkai Zasshi 1992; 96:776–783. Mori T, Kamei S, Sugawara T, Tazawa Y. [Changes in retinal electrical response due to different size of experimental retinal detachment]. Nippon Ganka Gakkai Zasshi 1991; 95:1248–1251. Imai K, Loewenstein A, de Juan E, Jr. Translocation of the retina for management of subfoveal choroidal neovascularization I: experimental studies in the rabbit eye. Am J Ophthalmol 1998; 125:627–634. Kim SD, Nao-i N, Maruiwa F, Sawada A. Electrical responses from locally detached retina and its recovery after reattachment. Ophthalmologica 1996; 210:195–199. Luthert PJ, Chong NH. Photoreceptor rescue. Eye 1998; 12:591–596. Milam AH, Li ZY, Fariss RN. Histopathology of the human retina in retinitis pigmentosa. Prog Retin Eye Res 1998; 17:175–205. Lewis GP, Linberg KA, Geller SF, Guerin CJ, Fisher SK. Effects of the neurotrophin brain-derived neurotrophic factor in an experimental model of retinal detachment. Invest Ophthalmol Vis Sci 1999; 40:1530–1544. Lewis GP, Fisher SK. Glial cell-line derived neurotrophic factor (GDNF): retinal distribution and rescue effects on photoreceptors following experimental retinal detachment. Invest Ophthalmol Vis Sci 1998; 39:S572. Lewis G, Mervin K, Valter K, et al. Limiting the proliferation and reactivity of retinal Muller cells during experimental retinal detachment: the value of oxygen supplementation. Am J Ophthalmol 1999; 128:165–172. Sakai T, Lewis GP, Linberg KA, Fisher SK. The ability of hyperoxia to limit the effect of experimental detachment in cone-dominant retina. Invest Ophthalmol Vis Sci 2001; 42:3264–3273. Hughes BA, Gallemore RP, Miller SS. Transport mechanisms in the retinal pigment epithelium. In: Marmor MF, Wolfensberger TJ, eds. The Retinal Pigment Epithelium. New York: Oxford University Press, 1998:103–134. Peterson WM, Quong JN, Miller SS. Mechanisms of fluid transport in retinal pigment epithelium. The Third Great Basin Visual Science Symposium on Retinal Research 1998; III:34–42. Marmor MF, Maack T. Enhancement of retinal adhesion and subretinal fluid resorption by acetazolamide. Invest Ophthalmol Vis Sci 1982; 23:121–124. Wolfensberger TJ, Chiang RK, Takeuchi A, Marmor MF. Inhibition of membranebound carbonic anhydrase enhances subretinal fluid absorption and retinal adhesiveness. Graefes Arch Clin Exp Ophthalmol 2000; 238:76–80. Peterson WM, Meggyesy C, Yu K, Miller SS. Extracellular ATP activates calcium signaling, ion, and fluid transport in retinal pigment epithelium. J Neurosci 1997; 17:2324–2337. Meyer CH, Hotta K, Peterson WM, Toth CA, Jaffe GJ. Effect of INS37217, a P2Y2 receptor agonist, on experimental retinal detachment and electroretinogram in adult rabbits. Invest Ophthalmol Vis Sci 2002; 43:3567–3574. Cantrill HL, Pederson JE. Experimental retinal detachment. III. Vitreous fluorophotometry. Arch Ophthalmol 1982; 100:1810–1813.
128
Peterson
66.
Pederson JE, Cantrill HL. Experimental retinal detachment. V. Fluid movement through the retinal hole. Arch Ophthalmol 1984; 102:136–139. Kroll AJ, Machemer R. Experimental retinal detachment and reattachment in the rhesus monkey. Electron microscopic comparison of rods and cones. Am J Ophthalmol 1969; 68:58-77. Wilson CA, Khawly JA, Hatchell DL, Machemer R. Experimental traction retinal detachment in the cat. Graefes Arch Clin Exp Ophthalmol 1991; 229:568–573. Fastenberg DM, Diddie KR, Sorgente N, Ryan SJ. A comparison of different cellular inocula in an experimental model of massive periretinal proliferation. Am J Ophthalmol 1982; 93:559–564. Cleary PE, Minckler DS, Ryan SJ. Ultrastructure of traction retinal detachment in rhesus monkey eyes after a posterior penetrating ocular injury. Am J Ophthalmol 1980; 90:829–845. Gregor Z, Ryan SJ. Combined posterior contusion and penetrating injury in the pig eye. II. Histological features. Br J Ophthalmol 1982; 66:799–804. Wilson CA, Royster AJ, Tiedeman JS, Hatchell DL. Exudative retinal detachment after photodynamic injury. Arch Ophthalmol 1991; 109:125–134. Collier RH. Experimental embolic ischemia of the choroid. Arch Ophthalmol 1967; 77:683–692. Gaudric A, Sterkers M, Coscas G. Retinal detachment after choroidal ischemia. Am J Ophthalmol 1987; 104:364–372. Stern WH, Ernest JT. Microsphere occlusion of the choriocapillaris in rhesus monkeys. Am J Ophthalmol 1974; 78:438–448. Kishi S, Tso MO, Hayreh SS. Fundus lesions in malignant hypertension. I. A pathologic study of experimental hypertensive choroidopathy. Arch Ophthalmol 1985; 103:1189-1197. Jacobs GH, Calderone JB, Jakai T, Lewis GP, Fisher SK. An animal model for studying cone function in retinal detachment. Doc Ophthalmol 2002; 104:119–132. Nour M, Alvarez K, Quiambao A, Peterson WM, Al-Ubaidi MR, Naasm MI. Effects of P2Y2 receptor agonist (INS37217) on recovery of retinal morphology and function following subretinal injections in mice: potential use in gene therapy delivery. Invest Ophthalmol Vis Sci 2002; 43: abstract #4610.
67.
68. 69.
70.
71. 72. 73. 74. 75. 76.
77. 78.
8 Retinal Ischemia Manuel Vidal-Sanz, Marı´a P. Lafuente, Inmaculada Selle´sNavarro, Marı´a E. Rodrı´guez, Sergio Mayor-Torroglosa, and Marı´a P. Villegas-Pe´rez Universidad de Murcia Murcia, Spain
I.
INTRODUCTION
In recent years, an increasing amount of experimental work has been directed toward the search for substances (e.g., trophic factors, pharmaceutical compounds) that could diminish or delay neuronal degeneration secondary to central nervous system injury or genetic diseases. The concept of neuroprotection as a possible therapy for degenerative diseases of unknown mechanisms has gained acceptance among the scientific community. Yet we are still at the stage of gathering bench-laboratory information necessary to understand the neuronal response to injury, and thus it might take some time before neuroprotection becomes an option for clinical management. We will describe some of the methodological approaches used in our laboratory to study the fate of retinal ganglion cells after ischemia and will discuss the problems, advantages, and disadvantages of these methods. Specifically, we will concentrate on two protocols recently used in our laboratory to induce transient ischemia of the retina. Our aim is to provide the researcher with a detailed guide on these two experimental models to study the fate of retinal ganglion cells after ischemic injury and neuroprotection. 129
130
Vidal-Sanz et al.
II. METHODS A.
Animal Care and Anesthesia
We have mainly used adult female Sprague-Dawley rats (200–225 g) for our experiments. The rats were obtained from the breeding colony of Murcia University or from Harlan Interfauna lbe´rica (Barcelona, Spain). Animal care and experimental procedures were done according to institutional guidelines, European Union regulations, and policies on the use of animals of the association for research and vision in ophthalmology (ARVO). All surgical manipulations were carried out using general anesthesia. Different types of anesthesia were used. For retrograde labeling, animals were administered a mixture of ketamine (75 mg/kg) (Ketolar, Parke-Davis, S.L. Barcelona, Spain) and xylazine (10 mg/kg) (Rompu´n, Bayer, S.A. Barcelona, Spain) in sterile saline. For neuroprotection studies, animals were anesthetized with intraperitoneal injections of 7% chloral hydrate in saline (0.42 mg/g body weight), because both ketamine and xylazine have been reported to have neuroprotective effects on ischemia-induced neuronal damage [1,2]. After experimental manipulations, animals were placed in their cages to recover from anesthesia and a steroid-antibiotic ointment containing neomycin and dexamethasone (Fludronef, Iquinosa, Madrid, Spain) was applied over the ocular surface to prevent corneal desiccation. Rats were fed ad libitum and kept in cages in temperature-controlled rooms with a 12-h light/dark cycle (light period from 8 a.m. to 8 p.m.). Light intensity in the cages ranged from 8 to 24 lux. An operating microscope and standard microsurgical instruments were used for all experimental procedures. B.
Is the Retina Suitable for Neuroprotection Studies?
The retina is a highly specialized part of the central nervous system that takes its position in the front of the head during development. From this location, the retina looks at the outside world and captures electromagnetic waves within the visible spectrum; these are transduced into electrical signals that are in turn processed through the different retinal layers to produce a retinal output. The retinal ganglion cell (RGC) axons convey this output along the optic nerve toward several relay stations in the midbrain, where it is further analyzed and sent over to the visual cortex where visual perception occurs. Several characteristics of the retinocollicular system of the adult rodent make it suitable for in vivo studies of neuronal survival after injury and administration of neuroprotective substances.
Retinal Ischemia
131
1. Location The retina is encapsulated within the eye globe, which is located in the orbit, and is easily accessible for experimental manipulations. Moreover, the retina is somewhat isolated from the rest of the CNS, and thus experimental manipulations of one retina do not usually involve damage to the other retina or other parts of the brain, and this allows animal survival for short- and long-term studies. 2. Natural Reservoir The eye globe may act as a reservoir for substances injected into the vitreous. These substances may be used to treat the retinal cells, or to label selectively the entire axonal population of RGCs, thus allowing identification of retinal projections into the brain [3]. 3. Vascular Supply In the rat, the ophthalmic artery trifurcates into two posterior ciliary arteries (nasal and temporal) and one central retinal artery (CRA). The CRA penetrates into the eye through the inferior and nasal aspect of the optic nerve [4–7]. Seven or eight terminal radial branches of the CRA are the main vascular supply to the inner layers of the retina. The choroidal blood vessels provide nutrients to the outermost one-third of the retina, and originate from the two posterior ciliary arteries. Arterioles arising from the posterior ciliary arteries, as well as from the CRA, are the main source of blood flow to the optic nerve head [6]. Because the retina is encapsulated within the sclera, which is a semi-rigid structure, increments of the intraocular pressure directly affect retinal blood flow. In addition, the ophthalmic artery may be identified intraorbitally on its course along the ON sheath before it trifurcates and is dissected and ligated, thus interrupting retinal and choroidal blood supply. 4. Optic Nerve Manipulations The intra-orbital segment of the optic nerve can be readily identified, dissected, and injured. The ophthalmic artery does not enter the ON, like in most mammals, but courses within the ON sheath, and this allows experimental manipulations of the ON without compromising retinal blood supply [8,9]. 5. Identification of Retinal Neurons The great majority of RGCs in the albino rat send their axon to the superficial layers of the superior colliculi (SCi) [10] and a proportion (35%) of these also send a collateral to the dorsal lateral geniculate nuclei (dLGNi) [11]. Application
132
Vidal-Sanz et al.
of retrogradely transported tracers to the main retinorecipient territories (e.g., the SCi), or to their cut axons in the ON results in the selective labeling of the RGC population. Because the retina may be examined as a flat mount, the entire RGC population can be observed in one preparation, and this facilitates quantitative estimations of the RGC population in the normal or experimental retina. In addition to RGCs, the retina comprises five other different classes of neurons, whose cell somata and processes are all confined to the retina. The cell somata of these are organized into three nuclear layers (outer, inner, and RGC layer) with two intermingled synaptic layers (outer and inner plexiform layers). This layered structure may be readily examined in retinal cross-sections. Furthermore, specific classes and subclasses of retinal neurons and cells may be identified using retrograde labeling techniques combined with immunocytochemical techniques [12]. C.
Retrograde Labeling of the RGC Population
It is difficult to distinguish RGCs from the many displaced amacrine cells in the RGC layer [13] on the basis only of classical anatomical methods [12]. Furthermore, identification of injured RGCs is hindered because these may suffer phenotypic changes, including morphological [14], immunocytochemical [8,12], and molecular [15,16]. To identify the RGC population we have used retrogradely transported neuronal tracers applied to the terminals or cut axons of the RGCs. Identifying and counting RGCs is important when evaluating whether a substance may have neuroprotective effects. 1. Use of FluoroGold as Retograde RGC Tracer The RGC population was retrogradely labeled from the superior colliculi (SCi) with the fluorescent tracer FluoroGold (FG) (Fluorochrome Inc., Engelwood, CO) following protocols that were originally described in 1988 [17]. In brief: The Midbrain was exposed, the pia mater overlying both SCi was gently removed, and a small piece of gelatin sponge (Spongostan Film, Ferrosan, Denmark) soaked in a solution of 3% FG and 10% dimethyl sulfoxide in saline was laid over the SCi. Because most RGCs in the rat project to the SCi [10], this procedure results in the labeling of almost the entire RGC population in both retinas [17]. Previous studies from our laboratory [18,19] have shown that following FG application to the SCi, some RGCs appear labeled as early as 3 days after dye application, and most RGCs appear labeled by 7 days after dye application. At this time, the densities of FG-labeled RGCs are similar to those obtained when other fluorescent and nonfluorescent retrogradely transported tracers are applied to the main retinorecipient target regions in the brain [12,17,20]. Furthermore, FG application to the SCi [19] or the lateral rectus muscle [21] results in the labeling of the entire population of RGCs or abducens motoneurons, respectively,
Retinal Ischemia
133
for up to 4 weeks after tracer application without leakage or apparent fading. Longer survival periods, however, were associated with a decrease in the numbers of labeled cells [19,21], and by 3 months only one-third of the population of abducens motoneurons was still labeled with FG [21]. At this time point, the entire population of abducens motoneurons could be labeled again if FG was reinjected into the ipsilateral rectus muscle, suggesting that FG is not toxic to neurons but disappears slowly from the neurons with increasing survival periods [21]. 2. Use of di-ASP as Retrograde RGC Tracer To identify RGCs in long-term studies we have used two different methods. In a first method, we labeled the RGC population with dil, a lipophilic carbocyanine which persists within the RGC somata for periods of up to 21 months after tracer application without leakage or fading [17,20]. A second method consists of applying retrograde tracers to the axons [12,22] or terminals [23] of RGCs, shortly before animal processing. Here we describe the use of DiAsp applied intraorbitally to the ocular stump of the transected ON. The lipophilic dye 4Di-10Asp [DiAsp, D29, N-4-4-4-didecylaminostyryl-N-methylpyridinium iodide, Molecular Probes, Eugene, OR] can be applied to the RGC target territories or their cut axons to retrogradely label RGCs, both in control and experimental animals [22,24,25]. In brief: Three days before sacrifice, the left ON is exposed in the orbit, dissected from its surrounding sheaths, and cut close to its origin in the optic disc [8]. DiAsp is then applied to the ocular stump of the intraorbitally transected ON. To facilitate dye uptake by cut axons, crystals of DiAsp were previously dissolved in dimethylformamide and the solvent allowed to evaporate. Intraorbital transection of the ON induces RGC death [12,20], but this first appears between day 4 and 5 after axotomy [18,26]. Therefore, it is likely that the densities of DiAsp-labeled RGCs 3 days after dye application would not be affected by axotomy-induced RGC death. However, we cannot exclude the possibility that transient ischemia of the retina prior to DiAsp application shortens this 4–5 day period before the onset of cell death. D. Induction of Retinal Ischemia A variety of procedures have been described to induce retinal ischemia in laboratory animals. These may be divided into two main groups. One consists of increasing the intraocular pressure (IOP) above systolic levels to interrupt blood flow within the eye, and the other consists of the ligature of the blood vessels supplying the retina. Retinal ischemia may be permanent or transient if reperfusion is allowed (for review, see Table 3, Ref. 27). In the following section we describe the methods used in our laboratory to induce transient ischemia of the retina by elevation of the IOP [19] or by selective ligature of the ophthalmic vessels [28–32].
134
Vidal-Sanz et al.
1. General Considerations Ischemia-induced CNS damage is influenced by body temperature, which should be kept constant throughout the duration of the experiment. This is also the case for experiments aimed at studying ischemia-induced RGC death [33]. In a previous study, in which the room temperature was maintained constant at 23°C, the body temperature of a group of rats anesthetized and subjected to transient retinal ischemia was monitored with a rectal probe. The body temperature fluctuations in these animals ranged from 35.4°C right after initiation of retinal ischemia to 33.8°C 2 h later [19]. Because in a large series of similarly prepared experiments the densities of RGCs surviving different periods of ischemia were very consistent, it is possible that small body temperature fluctuations did not interfere with RGC survival [19]. Furthermore, the temperature within the eye was not monitored; thus, we ignore whether these small changes in body temperature also influenced retinal temperature and if this in turn affected RGC survival in those experiments. The retinal blood flow may be monitored through the operating microscope. Inspection of the eye fundus is facilitated by pupil dilation with a topical drop of 1% tropicamide (Cusi Laboratories, El Masnou, Barcelona, Spain). Corneal desiccation throughout the experiment is avoided by applying on the cornea a solution of 2% hydroxypropylmethylcellulose (Gonioftal 4000, Cusi Laboratories, El Masnou, Barcelona, Spain). The placement of a coverslip over the cornea facilitates direct visualization of the eye fundus through the microscope. The eye fundus of the albino rat shows the central retinal artery dividing into six or seven radial vessels as well as six or seven radial veins that converge toward the disc into the central retinal vein. 2. Increase of the Intraocular Pressure Above Systolic Levels We induced retinal ischemia in the left eye by increasing the intraocular pressure (IOP) above systolic arterial levels (see below). In these animals, as well as in those in which ischemia was induced by selective ligature of the ophthalmic vessels, the right intact eye served as control with the animals under deep anesthesia, two nylon monofilament 6/0 sutures were placed on the superior and inferior bulbar conjunctiva of the left eye close to the corneo-scleral limbus. These sutures were pulled tangentially in opposite directions until retinal blood flow was interrupted completely; this was assessed by examination of the eye fundus. The sutures were then tied to a metal frame, specially devised for these studies, to maintain the IOP above systolic levels. Inspection of the eye fundus was required throughout the duration of the ischemic period to ascertain blood flow arrest (Fig. 1a). Retinal ischemia is characterized by pallor of the iris and intense pallor of the eye fundus, as well as by blood flow arrest within radial retinal vessels. These were either empty or showing fragmented columns of arrested red blood cells.
Retinal Ischemia
135
Figure 1 Schematic representations of the methods employed to induce transient ischemia of the retina by increasing the intraocular pressure above systolic levels (a) or by selective ligature of the ophthalmic vessels (b). Both procedures induce choroidal and retinal blood flow interruption. (a) Two sutures were placed on the superior and inferior bulbar conjunctiva close to the corneo-scleral limbus. These sutures were pulled tangentially in opposite directions to increase the intraocular pressure until retinal blood flow was interrupted completely. This was assessed by examination of the fundus. (b) The optic nerve (ON) is exposed in the orbit and the dural sheath opened longitudinally. A fine 10/0 suture is placed between the ON and the sheath and tied around the sheath avoiding damage to the ON. Because the ON contains the ophthalmic artery, this procedure interrupts retinal blood flow.
136
Vidal-Sanz et al.
At the end of ischemia, the sutures were released slowly and this allowed restoration of the retinal blood flow. Reestablishment of blood flow was characterized by increasing movement of the fragmented columns of red cells, the progressive filling of the radial vessels to their normal pattern, and the pink appearance of the fundus. Reestablishment of blood flow within the retina, if not spontaneous, may be facilitated with gentle eye massage. Animals in which restoration of the retinal blood flow did not occur, or was associated with retinal or vitreous hemorrhages, were discarded from the study. Regular findings at the termination of the period of ischemia were a moderate edema of conjunctiva and cornea [19]. 3. Selective Ligature of the Ophthalmic Vessels Under deep anesthesia, the left optic nerve was exposed in the orbit and the ON sheath was opened longitudinally [8] (Figs. 1b, 2a). A fine 10/0 monofilament suture was carefully introduced between the ON and the sheath, and tied around the sheath, avoiding damage to the ON (Fig. 2b). Because the ON sheath contains the ophthalmic artery [6], this procedure interrupts retinal and choroidal blood flow. The ligature may be released after different ischemic intervals to allow retinal reperfusion. The eye fundus may be examined directly through the operating microscope to assess retinal blood flow before, during, and after retinal ischemia (Fig.3) [30–32]. E.
Administration of Neuroprotective Substances
Substances to be tested for their neuroprotective effects may be administered before, during, or after the insult. These may be given systemically or topically or delivered directly into the vitreous [34]. Topical instillation is a simple procedure that may be done in the alert animal. For direct administration into the vitreous, the animal is sedated with anesthesia, the pupil is dilated, and topical anesthetic eyedrops are also instilled on the eye. A 30-G needle is used to produce a penetrating puncture through the conjunctiva, sclera, choroid, and retina at approximately 1–2 mm from the corneo-scleral limbus. The needle of 5 µL Hamilton microsyringe is then introduced through this puncture into the vitreous and directed toward the posterior pole to avoid injury to the lens [34]. Direct injury to the lens results in lens opacity and may have neuroprotective effects on RGC survival [35,36]. Injection volumes were not greater than 5 µL, and saline was used in most instances as vehicle. F.
Tissue Processing
After various survival times, animals were given an overdose of anesthetics (ketamine and xylazine) and perfused through the heart first and briefly with 0.9%
Retinal Ischemia
137
Figure 2 Micrographs taken with the operating microscope illustrating the procedures employed for selective ligature of the ophthalmic vessels. (a) Micrograph illustrating the posterior part of the eye, the optic nerve head (*) and its surrounding dural sheath. (b) The dural sheath has been opened longitudinally and a fine suture is tied around the sheath avoiding damage to the ON (*).
138
Vidal-Sanz et al. Figure 3 Micrographs illustrating the eye fundus of the retina before ischemia (a), during ischemia (b), and shortly after reperfusion (c). These micrographs were taken through a video camera connected to the operating microscope. (a) A normal eye fundus shows typical radial retinal vessels. (b) During the period of transient ischemia induced by selective ligature of the ophthalmic vessels, these appear empty or with fragmented columns of red cells that do not move. (c) Shortly after release of the ligature around the ophthalmic vessels, blood flow is restored within radial retinal vessels.
Retinal Ischemia
139
NaCl followed by 4% paraformaldehyde in 0.1 M phosphate buffer. Both eyes were enucleated and an 8/0 suture was placed in the superior aspect of the conjunctiva near the limbus for orientation purposes. The retina may be examined in cross-sections or in whole-mounts. In the following section we describe the methodology employed to obtain flattened whole-mounts. 1. Retinal Whole-Mounts The retinas were dissected out and prepared as flattened whole-mounts by making four radial cuts, the deepest one indicating the superior pole of the retina and the others indicating the nasal, temporal, and inferior pole of the retina, respectively. Flattened whole-mounts were transferred to a piece of filter paper and this was immersed in the same fixative for 1 h. The retinas were released from the filter paper, washed in the buffer solution, and mounted vitreal side up on subbed slides. For experiments in which DiAsp was used to retrogradely label RGCs, 0.1 M sodium carbonate buffer (pH 9) was used as mounting media. In animals in which FG was used as retrograde tracer, mounting media consisted of 50% glycerol in 0.1 M sodium carbonate buffer (pH 9) containing 0.04% of p-phenylenediamine [37]. Retinas were examined under fluorescence microscopy (Axiophot, Zeiss, Oberkochen, Germany) with fluorescein (BP 450-490, LP520) and ultraviolet (BP 365/12, LP 397) filters that allow the observation of the green-yellowish fluorescence of DiAsp and of the white-gold fluorescence of FG, respectively. It is our experience that examination of DiAsp should be done within the first 24 h after processing the animal because the dye tends to dissipate and fade away, thus losing its fluorescence quality soon after mounting. 2. Photographing and Counting Labeled RGCs Densities of surviving RGCs in the flat mounted retinas were estimated following previously described methods to determine densities of RGCs within the central regions of the retina [12,17–20,30], where RGCs are the most prevalent [38,39]. In brief, FG-labeled RGCs were counted in a masked fashion from printed micrographs (⫻400) taken from 12 standard rectangular areas of 0.0864 mm2 of each retina situated, three in every retinal quadrant, at approximately 0.875, 1.925, and 2.975 mm, respectively, from the optic disc. A labeled RGC was counted if the whole cell or its nucleus was visible within the micrograph. Counts in the 12 regions were averaged and divided by the area of the picture to obtain a mean RGC density (cells/mm2) per retina. For groups of similarly treated retinas, results were reported as mean RGC densities ⫾ SD (standard deviation). Cell counts were done in a masked fashion from printed micrographs; the identity of the retinas that led to the micrographs was not known until cell counts
140
Vidal-Sanz et al.
from different groups were finished. For statistical analysis, RGC densities were compared using two non-parametric tests: the one-way ANOVA Kruskal-Wallis and Mann-Whitney tests, with P values of ⬍0.05 considered statistically significant. 3. General Appearance of Retinas Labeled with FG or Di-ASP In control retinas, only the retinal cells labeled with FG have the typical characteristics of RGCs. These are observed in the RGC layer, but there is a small proportion of displaced RGCs (Dogiel corpuscles) that may be observed in the inner plexiform or inner nuclear layers. FG-labeled RGCs have the typical punctate and diffuse FG fluorescence, which delineates their soma and occasionally initial segments of their primary dendrites (Fig. 4a, b). In a recent study, the densities of FG-labeled RGCs in the control retinas of large series of rats were estimated following the above-described methods [29,30]. Although there were small variations in the mean densities of FG-labeled RGCs in some control retinas, overall these were rather consistent [29,30] (Fig. 5). These variations can be attributed to slight differences in tracer application or in the efficiency of the batches of tracer employed, and also have been observed in previous studies from this laboratory [18,19]. Nevertheless, in the experimental groups, the densities obtained for the right intact retinas may be used as 100% survival for their contralateral, left operated retinas. In the experimental left retinas subjected to various periods of ischemia, in addition to RGCs, microglial cells also appear intensely labeled with FG. These cells, which can be easily distinguished from RGCs on the basis of their morphology (cells with bright fluorescence in their small soma and multiple fine tortuous processes), phagocytose the debris of degenerating FG-labeled RGCs and thus become FG-labeled [24]. FG-labeled microglial cells have been found in a number of studies in which FG was applied to the SCI before retinal injury in the rat [18,19,36] and goldfish [40]. In control and experimental retinas in which we used DiAsp as retrograde tracer, the only cells that appeared DiAsp-labeled were RGCs. These showed typical punctate and diffuse DiAsp fluorescence delineating the soma, axon, and, occasionally, primary and secondary RGC dendrites (Figs. 6, 7). The mean densities of RGCs labeled with di-ASP applied to intraorbital ON stump in a control group of nine rats (22) was somewhat smaller than the densities of labeled RGCs when other nonlipophilic tracers were applied to the SC or the ON (17–19). This could be explained by the limited solubility of the dye, which resulted in lower labeling efficiency. Because DiAsp application involves transection of the ON, the low numbers of DiAsp-labeled RGCs found in our experiments could also be explained by the axotomy-induced retrograde degeneration of RGCs. However, this is not likely because axotomy-induced RGC death first appears between
Retinal Ischemia
141
Figure 4 Retinal ganglion cells (RGCs) retrogradely labeled with FluoroGold (FG) in whole-mounted retinas. FG was applied to the superior colliculi one week prior to animal processing. (a) The central region of the retina in the superior temporal quadrant of this retina shows FG-labeled RGCs separated by radially oriented blood vessels and axon bundles (⫻218). (b) At higher magnification, FGlabeled RGCs show the typical diffuse and granular fluorescence depicting their somata (⫻436).
142
Vidal-Sanz et al.
Figure 5 Micrograph of the whole-mounted right retina in an adult PVG rat 1 week after FluoroGold application to both superior colliculi. FG-labeled RGCs are evenly distributed throughout the retinal quadrants. The micrograph was prepared with the aid of a motorized stage on a photomicroscope with a high-resolution camera connected to an image analysis system with an automatic frame grabber device (Image-Pro Plus, V4.1; Media Cybernetics, Silver Spring, MD, USA). The superior aspect of the retinas is between 1 and 2 o’clock orientation. (Scale bar ⫽ 1 mm.)
4 and 5 days after ON section close to the eye [18,26], and we estimated RGC densities 3 days after dye application [22]. Moreover, in experiments in which DiAsp was applied to the SC to retrogradely label RGCs, Thanos [24] reported densities of DiAsp-labeled RGCs similar to those obtained when we applied DiAsp to the cut ON, indicating that DiAsp has lower labeling efficiency than other neuronal tracers. Finally, the absence of DiAsp-labeled microglial cells in our experiments further supports that there is no RGC death in the first 3 days after ON section.
Retinal Ischemia
143
Figure 6 Retinal ganglion cells (RGCs) retrogradely labeled with DiAsp applied to the ocular stump of the intraorbitally transected optic nerve 3 days before animal processing. The central retina of the superior temporal quadrant shows typical DiAsp labeling of retinal fiber bundles as well as the somata and primary dendrites of RGCs (⫻441).
Figure 7 Retinal ganglion cells (RGCs) retrogradely labeled with DiAsp applied to the ocular stump of the intraorbitally transected optic nerve 3 days before animal processing. The central region of the inferior-nasal quadrant shows typical intense DiAsp labeling of RGC axonal bundles and somata (⫻59).
144
Vidal-Sanz et al.
III. DISCUSSION Several requirements need to be met for the above-mentioned protocols to be useful. These include, among others (1) reproducibility of the same insult given to the retina; (2) consistency of the results produced with a given type of insult; and (3) reliable estimates of the effects of the insult in the survival of the RGC population and/or other retinal neurons. A.
Requirements Needed for the Protocols to Be Useful
1. Reproducible Insult For an insult to be reproducible, it should be performed equally every time and should also lead to similar results. Thus, an identical procedure should be performed every time that transient ischemia is given to the retina, whether this is induced by elevated IOP or by ligature of the ophthalmic vessels. A number of investigators monitor the level of the IOP reached to induce blood arrest with the use of intraocular cannulas connected to a manometer system [41–46]. In our experiments the eye fundus was constantly inspected to ensure flow arrest, but accurate measurements of the IOP were not done. Nevertheless, studies in which a number of animals were treated with similar periods of ischemia showed consistent numbers of surviving RGCs [19], suggesting that small fluctuations in the intraocular pressure reached to arrest blood supply did not affect RGC survival substantially. Ligature of the ophthalmic vessels seems to be a rather consistent method to induce blood arrest into the retina [30]. The reproducibility of this method relies on the ability of the investigator to perform in identical fashion the dissection and ligature of the ophthalmic vessels that run within the ON sheath. Another variable to be taken into account when performing these studies is the difference in the period of time (usually a few minutes) from ligature release to full retinal reperfusion. 2. Consistency of Injury-Induced Cell Loss The results of a large series of experiments in which transient ischemia of the rat retina was induced by elevated IOP were highly consistent for the different groups analyzed after different survival (reperfusion) intervals [19]. Similarly, consistent results in the amount of RGC loss were observed in another series of experiments in which transient ischemia of the rat retina was induced by ligature of the ophthalmic vessels and RGCs were labeled with DiAsp [31] or FG [29,30,32]. Altogether, these studies indicate that RGC survival after different periods of transient ischemia and survival intervals is a highly consistent and predictable finding.
Retinal Ischemia
145
3. Reliable Estimates of Surviving Cells The studies mentioned above were based on counts of cells retrogradely labeled with dyes applied to their cut axons or their targets. The efficiency of the tracer employed—that is, the proportion of retinal ganglion cells that becomes retrogradely labeled—should be well established. Another important issue is to be able to reproduce consistently similar proportions of labeled retinal ganglion cells every time the dye is applied. The techniques employed in these experiments sampled similar regions of the retinas located within the central retina, where higher densities of RGCs are observed [13], but overlook RGC densities within more peripheral areas of the retina. While this is a reliable and consistent method to obtain RGC densities, a more accurate estimate should include counting all the FG-labeled RGCs within each retina, but this is major job to be done manually. RGC counts with an automatic image analysis system of a digital image of the retina has been precluded by technical difficulties in identifying, single FG-labeled neurons, FGlabeled microglial cells, and FG-labeled cells that are located close together as it happens for the central regions of the retina (see Fig. 4a,b). Recent advances in image-analysis may permit in the future more detailed analysis off the RGC population [23,39], including the identification of different types of RGCs, as defined by their morphological properties [47]. Screening studies on the effects of neuroprotectants with the above-mentioned techniques rely on the interpretation of labeled RGCs as alive, rescued neurons. In some instances, neuronal cells were labeled prior to injury with tracers applied to their main targets in the brain (e.g., Refs. 34,30). In other instances these cells were labeled after injury with tracers applied to their targets [23], or their cut axons [31]. While this assumption is generally accepted for many of these studies [18,19,23,29–32,34,35], we cannot be certain that rescued RGCs retain all their normal physiological properties. Other studies have shown that retinal cells may undergo functional and metabolic deficits before they actually die [16,48,49]. For example, chronic ligature of both carotid arteries did not induce evident loss of photoreceptors until 9 months later, but functional correlates indicated that both the a and b waves of the electroretinogram were clearly pathological, as early as 90 days after bilateral carotid occlusion [50]. This highlights the need for further studies to ascertain functional viability of retinal neurons surviving ischemic injury [51]. B. Advantages and Disadvantages of These Methods 1. Increasing IOP Above Systolic Levels Increasing the IOP above systolic levels with two sutures pulling from both sides of the eye has the advantage of being a noninvasive method. No puncture is given
146
Vidal-Sanz et al.
to the eye and/or the retina, and therefore nonspecific neuroprotective effects [35] that may originate as a consequence of lens injury [36] are discarded. However, the main disadvantage of this method is that it is tedious and time consuming; constant observation of the eye fundus of the retina to ascertain blood arrest is required to adjust tension on the sutures when necessary. The duration of the period of transient ischemia influences both the amount and pattern of RGC loss [19]. This fact should be taken into account when designing experiments aimed at investigating possible neuroprotective effects. Furthermore, we ignore whether in addition to blood flow arrest, increased intraocular pressure also induces compression or axotomy-like damage to retinal axons. If this were to be the case, an additional axotomy-like insult is given to the retina. 2. Ligature of the Ophthalmic Vessels Transient ischemia of the retina may be induced with methods that do not increase the intraocular pressure. Transient ligature of the carotid [48,52] and vertebral arteries [53] induces retinal ischemia but may also affect the brain. Clamping or ligating the ophthalmic vessels together with the ON have been used to induce retinal ischemia [35,54–57]. However, these were short-term experiments and, in addition to retinal ischemia, these procedures also induced ON axotomy. For the purpose of our studies, aimed at investigating the short- and long-term effects of retinal ischemia, we have preferred to induce retinal ischemia by the selective ligature of the ophthalmic vessels, avoiding direct mechanical damage to ON fibers. Selective ligature of the ophthalmic vessels was first used in monkeys by Hamasaki and Kroll [58] and later applied to rats [28–32,51]. Selective ligature of the ophthalmic vessels induces in the retina pathological findings that are similar to those found after ischemia induced by elevated IOP, including the loss of RGCs and the thinning of the IPL, INL, and OPL [59]. This method does not increase the IOP and, thus, there is no intraocular compression of retinal axons, avoiding axotomy-like injury to RGC fibers. The effects of ligature of the ophthalmic vessels on the ON head were not studied, but it is likely that transient ligature of the ophthalmic vessels also induces transient ischemia of the ON head. The main source of blood flow to the ON head in the rat comes from branches of the posterior ciliary arteries as well as from arterioles arising from the central retinal artery [6], and thus it is likely that ligature of the ophthalmic vessels affects blood flow within the ON head. C.
Further Implications of These Models
Retinal ischemia induced by any of the above-discussed methods induces an early loss of RGCs [19,31]. The loss of RGCs that appears shortly after retinal ischemia may be prevented, at least in part, with the use of different neuroprotective agents.
Retinal Ischemia
147
A recent article reviews several studies on RGC neuroprotection after retinal ischemia [27]. The variety of drugs employed in these experiments may be taken as an index of the many steps involved in RGC death that may be halted, and further points to the multiple factors involved in the process of ischemia-induced RGC death (see Table 7, Ref. 27). The techniques described here do not address questions related to the specific mechanisms of action of putative neuroprotectants. An efficient neuroprotective drug should diminish injury-induced RGC death, and this may be accomplished either through a direct effect on injured RGCs or indirectly by activating other non-RGC neurons or nonneuronal cells of the retina. Thus, whether the neuroprotective effect of a given drug is directly mediated through the RGCs themselves, or indirectly mediated through other cells in the retina, should be tackled with additional methods. In addition to the early loss of RGCs, retinal ischemia also induces a protracted loss of RGCs [19,31]. The mechanism by which the initial insult triggers such a long-lasting and prolonged process of RGC death is unknown. Whether injury triggers the immediate death of a subset of RGCs, and this in turn leads to secondary RGC death is currently unknown [31]. In this context, it is worth noticing that ideal neuroprotection should prevent both the early as well as the secondary slow loss of RGCs that were not primed to die in the first instance but that disappear with time [29–32]. Therapeutic interventions after CNS ischemia are at present mainly directed toward the so-called penumbra zone, a region of incomplete ischemia that surrounds the core of the infarcted region, which is most susceptible to reperfusion damage. The above-discussed models involve complete ischemia of the retina, and there may be differences in susceptibility of different retinal neurons depending on their location within the retina [55,60]. However, the concept of penumbra zone after ischemia may not be fully reproducible in these models of retinal ischemia. Moreover, a few minutes of ischemia induce irreversible damage in the CNS [61–63], whereas in the retinal models examined in these studies (increase of the IOP and SLOV), the transient period of ischemia required to produce RGC death is substantially longer. For example, 5 min of global cerebral ischemia induced by bilateral carotid artery occlusion are enough to produce neuronal death of nearly all the CA1 pyramidal neurons within the gerbil hippocampus after 5 min global cerebral ischemia [61,62]. In contrast, periods of 45–60 min of pressure-induced retinal ischemia are required to induce RGC death [19]. Transient ischemia of the retina has been used as a basic research model for glaucoma because ischemia may be involved in the pathogenic mechanism of this disease. In addition, both ischemia and glaucoma involve progressive loss of RGCs [19]. However, both the above-discussed methods to induce retinal ischemia elicit a rapid and massive RGC loss, and this is clearly not the case for chronic neurodegenerative conditions such as glaucoma, where RGC death appears to be a progressive but much slower event. Moreover, retinal ischemia also
148
Vidal-Sanz et al.
induces the loss of other non-RGC neurons [59] and this may not be the case for glaucoma [64]. Ligature of the ophthalmic vessels does not fully resemble the clinical situation of acute interruption of the blood flow either by central artery or vein occlusion. In these diseases, the outer retina is usually not affected, while ligature of the ophthalmic vessels interrupts blood supply to the retina and choroid, thus leading to different pathological findings. Preliminary results indicate that after ligature of the ophthalmic vessels, not only the RGC layer but the other retinal layers were severely affected by the initial period of transient ischemia [65]. The ligature of the ophthalmic vessels and the pressure-increased models to induce retinal ischemia do not fully reproduce human conditions. Nevertheless, these are useful models in neurobiology research to induce progressive neuronal degeneration of the RGC population, and are suitable models to examine the consequences of such insults in the different retinal layers after retinal ischemia. The fact that other retinal layers also degenerate with time after these insults may provide an opportunity to explore the progressive loss of other non-RGC neurons of the retina, including the inner and outer nuclear layer neurons. Moreover, these models may be used as powerful in vivo tools to screen for compounds that may have relevant neuroprotective effects against ischemia-induced neuronal cell death [29–32,65]. IV. CONCLUSIONS The above described methods to induce retinal ischemia cannot be extrapolated to clinical conditions in the human eye. Nevertheless, they provide a powerful tool to examine the fate of retinal neurons after ischemia and to further explore future therapeutic interventions to lessen the effects of transient ischemia. ACKNOWLEDGMENTS This work was supported by research grants from the Regional Government of Murcia (Fundacio´n Se´neca, PI-92/00540/FS/01), the Spanish Ministry of Science and Technology (BFI2002-03742), The Spanish Ministry of Health (03/13; FIS PI020407), and the European Union (QLK6-CT-2000-00569 and QLK6-CT2001-00385). The authors thank A. Avile´s, M. E. Aguilera, and J. M. Bernal for technical support. REFERENCES 1.
Wen R, Cheng T, Li T, Cao W, Steinberg RH. α2-Adrenergic agonists induced basic fibroblast growth factor expression in photoreceptors in vivo and ameliorate light damage. J Neurosci 1996; 16:5986–5992.
Retinal Ischemia 2.
3.
4. 5. 6. 7.
8.
9. 10. 11. 12.
13. 14.
15.
16.
17.
18.
19.
149
Hoffman WE, Pelligrino D, Werner C, Kochs E, Albrecht RF, Schulte, Esch J. Ketamine decreases plasma catecholamines and improves outcome from incomplete cerebral ischemia in rats. Anesthesiology 1992; 76:755–762. Avile´s-Trigueros M, Sauve´ Y, Lund RD, Vidal-Sanz M. Selective innervation of retinorecipient brainstem nuclei by retinal ganglion cell axons regenerating through peripheral nerve grafts in adult rats. J Neurosci 2000; 20:361–374. Forrester J, Peters A. Nerve fibers in optic nerve of rat. Nature 1967; 214:245–247. Janes RG, Bounds GW. The blood vessels of the rat’s eye. Am J Anat 1955; 96: 357–373. Morrison JC, Johnson EC, Cepurna WO, Funk RHW. Microvasculature of the rat optic nerve head. Invest Ophthalmol Vis Sci 1999; 40:1702–1709. Sugiyama K, Zhao-Bin G, Kawase C, Yamamoto T, Kitazawa Y. Optic nerve and peripapillary choroidal microvasculature of the rat eye. Invest Ophthalmol Vis Sci 1999; 40:3084–3090. Vidal-Sanz M, Bray GM, Villegas-Pe´rez MP, Thanos S, Aguayo AJ. Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat. J Neurosci 1987; 9:2894–2909. Vidal-Sanz M, Villegas-Pe´rez MP, Bray GM, Aguayo AJ. Use of peripheral nerve grafts to study regeneration after CNS injury. Neuroprotocols 1993; 3:29–33. Linden R, Perry VH. Massive retinotectal projection in rats. Brain Res 1983; 272: 145–149. Martin PR. The projection of different retinal ganglion cell classes to the dorsal geniculate nucleus in the hooded rat. Exp Brain Res 1986; 62:77–88. Villegas-Pe´rez MP, Vidal-Sanz M, Bray GM, Aguayo AJ. Influences of peripheral nerve grafts on the survival and regrowth of axotomized retinal ganglion cells in adult rats. J Neurosci 1988; 8:265–280. Perry VH. Evidence for an amacrine cell system in the ganglion cell layer of the rat retina. Neuroscience 1981; 6:931–944. Thanos S. Alterations in the morphology of ganglion cell dendrites in the adult retina after optic transection and grafting of the peripheral nerve segments. Cell Tissue Res 1988; 254:599–609. McKerracher L, Essagian C, Aguayo AJ. Temporal changes in β-tubulin and neurofilament mRNA levels after transection of adult rat retinal ganglion cell axons in the optic nerve. J Neurosci 1993; 13:2617–2626. Schlamp CL, Johnson EC, John YL, Morrison C, Nickells RW. Changes in Thy1 gene expression associated with damaged retinal ganglion cells. Mol Vis 2001; 7:192–201. Vidal-Sanz M, Villegas-Pe´rez MP, Bray GM, Aguayo AJ. Persistent retrograde labeling of adult rat retinal ganglion cells with the carbocyanine dye cil. Exp Neurol 1988; 102:92–101. Peinado-Ramo´n P, Salvador M, Villegas-Pe´rez MP, Vidal-Sanz M. Effects of axotomy and intraocular administration of BDNF, NT-3 and NT-4 on the survival of adult rat retinal ganglion cells: a quantitative in vivo study. Invest Ophthalmol Vis Sci 1996; 37:489–500. Selle´s I, Villegas-Pe´rez MP, Salvador M, Ruiz-Go´mez JM, Vidal-Sanz M. Retinal ganglion cell death after different transient periods of pressure-induced ischemia
150
20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
32.
Vidal-Sanz et al. and survival intervals: a quantitative in vivo study. Invest Ophthalmol Vis Sci 1996; 37:2002–2024. Villegas-Pe´rez MP, Vidal-Sanz M, Rasminsky M, Bray GM, Aguayo AJ. Rapid and protracted phases of retinal ganglion cell loss follow axotomy in the optic nerve of adult rats. J Neurobiol 1993; 24:23–36. Go´mez-Ramirez AM, Villegas-Pe´rez MP, Miralles de Imperial J, Salvador-Silva M, Vidal-Sanz M. Effects of intramuscular injection of botulinum toxin and doxorubicin on the survival of abducens motoneurons. Invest Ophthalmol Vis Sci 1999; 40:414–424. Vidal-Sanz M, Lafuente MP, Miralles de Imperial J, Villegas-Pe´rez MP. Long term effects of transient ischemia induced by selective ligature of the ophthalmic vessels on retinal ganglion cell survival (abstr). Invest Ophthalmol Vis Sci 1999; 40(4): S480. Mittag TW, Danias J, Pohorenec G, Yuan H-M, Burakgazi R, Redman-Chalmers R, Podos SM, Tatton WG. Retinal damage after 3 to 4 months of elevated intraocular pressure in a rat glaucoma model. Invest Ophthalmol Vis Sci 2000; 41:3451–3459. Thanos S. The relationship of microglial cells to dying neurons during natural neuronal cell death and axotomy-induced degeneration of the rat retina. Eur J Neurosci 1991; 3:1189–1207 Thanos S. Adult retinofugal axons regenerating through peripheral nerve grafts can restore the light-induced pupilloconstriction reflex. Eur J Neurosci 1992; 4:691–699. Berkelaar M, Clarke DB, Wang Y-C, Bray GM, Aguayo AJ. Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats. J Neurosci 1994; 14:4368–4374. Osborne NN, Ugarte M, Chao M, Chidlow G, Bae JH, Wood JPM, Nash MS. Neuroprotection in relation to retinal ischemia and relevance to glaucoma. Surv Ophthalmol 1999; 43: S102–S128. Otori Y, Shimada S, Morimura H, Ishimoto I, Tohyama M, Tano Y. Expression of c-fos and c-jun mRNA following transient retinal ischemia: an approach using ligation of the retinal central artery in the rat. Surv Ophthalmol 1997; 42(suppl 1):S96– S104. Vidal-Sanz M, Lafuente MP, Mayor S, Miralles de Imperial J, Villegas-Pe´rez MP. Retinal ganglion cell death induced by retinal ischemia: Neuroprotective effecs of two alpha-2 agonists. Surv Ophthalmol 2001; 45(suppl 3):S261–S267. Lafuente MP, Villegas-Pe´rez MP, Sobrado-Calvo P, Garcı´a-Avile´s A, Miralles de Imperial J, Vidal-Sanz M. Neuroprotective effects of α2-selective adrenergic agonist against ischemia-induced retinal ganglion cell death. Invest Ophthalmol Vis Sci 2001; 42:2074–2084. Lafuente MP, Villegas-Pe´rez MP, Selle´s-Navarro I, Mayor-Torroglosa S, Miralles de Imperial J, Vidal-Sanz M. Retinal ganglion cell death after acute retinal ischemia is an ongoing process whose severity and duration depends on the duration of the insult. Neuroscience 2002; 109:157–168. Lafuente MP, Villegas-Pe´rez MP, Mayor S, Aguilera ME, Miralles de Imperial J, Vidal-Sanz M. Neuroprotective effects of brimonidine against transient ischemia induced-retinal ganglion cell death: a dose response in vivo study. Exp Eye Res 2002; 74:181–189.
Retinal Ischemia 33.
34.
35.
36. 37.
38. 39.
40. 41. 42. 43. 44. 45.
46.
47. 48.
49. 50.
151
Faberowski N, Stefa´nsson E, Davidson RC. Local hypothermia protects the retina from ischemia: a quantitative study in the rat. Invest Ophthalmol Vis Sci 1989; 30: 2309–2313. Vidal-Sanz M, Lafuente M, Sobrado-Calvo P, Selle´s-Navarro I, Rodriguez E, Mayor-Torroglosa S, Villegas-Pe´rez MP. Death and neuroprotection of retinal ganglion cells after different types of injury. Neurotoxicity Res 2000; 2:215–227. Mansour-Robaey S, Clarke DB, Wang YC, Bray GM, Aguayo AJ. Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells. Proc Natl Acad Sci USA 1994; 91: 1632–1636. Leon S, Yin Y, Nguyen J, Irwin N, Benowitz L. Lens injury stimulates axon regeneration in the mature rat optic nerve. J Neurosci 2000; 20:4615–4626. Dodd J, Solter D, Jessell TM. Monoclonal antibodies against carbohydrate differentiation antigens identify subsets of primary sensory neurons. Nature 1984; 311:469– 472. Perry VH, Henderson Z, Linden R. Postnatal changes in retinal ganglion cells and optic axon populations in the pigmented rat. J Comp Neurol 1983; 219:356–368. Danias J, Shen F, Goldblum D, Chen B, Ramos-Esteban J, Podos SM, Mittag T. Cytoarchitecture of the retinal ganglion cells in the rat. Invest Ophthalmol Vis Sci 2002; 43:587–594. Salvador-Silva M, Vidal-Sanz M, Villegas-Pe´rez MP. Microglial cells in the retina of Carassius auratus: effects of optic nerve crush. J Comp Neurol 2000; 417:431–447. Johnson NF, Foulds WS. The effects of total acute ischemia on the structure of the rabbit retina. Exp Eye Res 1978; 27:45–49. Bu¨chi ER, Suivaizdis I, Fu J. Pressure-induced retinal ischemia in rats: an experimental model for quantitative study. Ophthalmologica 1991; 203:138–147. Takahashi K, Lam TT, Edwar DP, Bu¨chi ER, Tso MOM. Protective effects of flunarizine on ischemic injury in the rat retina. Arch Ophthalmol 1992; 110:862–870. Zang C, Takahashi K, Lam TT, Tso MOM. Effects of basic fibroblast growth factor in retinal ischemia. Invest Opthalmol Vis Sci 1994; 35:3163–3168. Osborne NN, Larsen A, Barnett NL. Influence of excitatory amino acids and ischemia on rat retinal choline acetyltransferase-containing cells. Invest Ophthalmol Vis Sci 1995; 36:1692–1700. Adachi K, Kashii S, Masai H, Ueda M, Morizane C, Kaneda K, Kume T, Akaike A, Honda Y. Mechanism of the pathogenesis of glutamate neurotoxicity in retinal ischemia. Graefe Arch Clin Exp Ophthalmol 1998; 236:766–774. Fukuda Y. A three group classification of rat retinal ganglion cells; histological and physiological studies. Brain Res 1977; 119:327–334. Barnett NL, Osborne NN. Prolonged bilateral carotid artery occlusion induces electrophysiological and immunohistochemical changes to the rat retina without causing histological damage. Exp Eye Res 1995; 61:83–90. Nash MS, Osborne NN. Assessment of Thy-1 mRNA levels as an index of retinal ganglion cell damage. Invest Ophthalmol Vis Sci 1999; 40:1293–1298. Osborne NN, Safa R, Nash MS. Photoreceptors are preferentially affected in the rat retina following permanent occlusion of the carotid arteries. Vision Res 1999b; 39:3995–4002.
152
Vidal-Sanz et al.
51.
Lafuente MP, Mayor-Torroglosa S, Miralles de Imperial J, Villegas-Pe´rez MP, Vidal-Sanz M. Transient ischemia of the retina results in altered retrograde axoplasmic transport: Neuroprotection with brimonidine. Exp Neurol 2002; 178:243–258. Miller L, Perry D. Opiate receptor subtype binding in gerbil hippocampus is altered by forebrain ischemia. Brain Res 1989; 495:367–372. Benfenati F, Merlo-Pich E, Zoli M, Grimaldi R, Fuxe K, Agnati LF. Changes in striatum MU and Deta opoid receptors after transient forebrain ischemia: a quantitative autoradiographic study. Brain Res 1991; 546:171–175. Steffa´nsson E, Wilson CA, Schoen T, Kuwabara T. Experimental ischemia induces cell mitosis in the adult rat retina. Invest Ophthalmol Vis Sci 1988; 29:1050–1055. Crosson CE, Willis JA, Potter DE. Effect of the calcium antagonist, nifedipine, on ischemic retinal dysfunction. J Ocul Pharmacol Ther 1990; 6:293–299. Szabo ME, Droy-Lefaix MT, Doly M, Braquet P. Free radical-mediated effects in reperfusion injury: a histologic study with superoxide dismutase and EGB 761 in rat retina. Ophthalmic Res 1991; 23:225–234. Tsujikawa A, Ogura Y, Hiroshiba N, Miyamoto K, Kiryu J, Honda Y. In vivo evalution of leukocyte dynamics in retinal ischemia reperfusion injury. Invest Ophthalmol Vis Sci 1998; 39:793–800. Hamasaki DI, Kroll AJ. Experimental central retinal artery occlusion. An electrophysiological study. Arch Ophthalmol 1968; 80:243–248. Villegas-Pe´rez MP, Rodriguez ME, Garcia A, Coll L, Aguilera ME, Miralles de Imperial J, Vidal-Sanz M. Transient ischemia of the retina: long-term neuroprotection of inner retinal layers with a selective alpha-2 agonist (AGN190342, Brimonidine) (abstr). Ophthalmic Res 2000; 32(suppl 2):48. Kaneda K, Kashii S, Kurosawa T, Kaneko S, Akaike A, Honda Y, Minami M, Satoh M. Apoptotic DNA fragmentation and upregulation of Bax induced by transient ischemia of the rat retina. Brain Res 1999; 815:11–20. Kirino T. Delayed neuronal death in the gerbil hippocampus following ischemia. Brain Res 1982; 239:57–69. O’Neill MJ, Hicks CA, Ward MA, Osborne DJ, Wishart G, Mathews KS, McLaughlin DP, Samford JA, McCarty DR, Patrick KE, Roman C, Fleisch JH, Gilmore J, Boot JR. LY393615 a novel neuronal Ca2⫹ and Na⫹ channel blocker with neuroprotective effects in models of in vitro and in vivo cerebral ischemia. Brain Res 2001; 888:138–149. Sommer C, Roth SU, Kuhn R, Kiessling M. Metabotropic glutamate receptor subtypes are differentially expressed after transient cerebral ischemia without, during and after tolerance induction in the gerbil hippocampus. Brain Res 2000; 827:172– 180. Kendell KR, Quigley HA, Kerrigan LA, Pease ME, Quigley EN. Primary openangle glaucoma is not associated with photoreceptor cell loss. Invest Ophthalmol Vis Sci 1995; 36:200–201. Rodriguez ME, Villegas-Pe´rez MP, Garcı´a A, Coll L, Aguilera E, Miralles de Imperial J, Lafuente MP, Vidal-Sanz M. Transient ischemia of the retina: inner retinal layers neuroprotection with a selective alpha-2 agonist (AGN 19034, Brimonidine) (abstr). Invest Ophthalmol Vis Sci 2001; 42(4):S747.
52. 53.
54. 55. 56.
57.
58. 59.
60.
61. 62.
63.
64.
65.
9 Drug Delivery Robert W. Nickells and Cassandra L. Schlamp University of Wisconsin–Madison Madison, Wisconsin, U.S.A.
I.
INTRODUCTION
The growing concept of being able to protect neurons from damaging stimuli, or being able to rescue them from actively executing a cell death program, carries with it the requirement of being able to deliver the appropriate drug or agent to the target cells. In the eye, this means being able to apply an effective dose of an agent to the retina and optic nerve. In some cases, it may also mean that the drug be applied continuously or for a sustained period of time in order to achieve a therapeutic effect. This chapter describes methods that have been used by various laboratories to deliver agents, most notably small pharmacological molecules, successfully. Each method has its advantages and disadvantages, which will be discussed. There are four basic modes of application to deliver a drug to the back of the eye. These are trans-corneal or topical, intravitreal, trans-scleral, and systemic [1].
II. TRANS-CORNEAL APPLICATION OF DRUGS There is a great deal of interest in the trans-corneal route of drug administration because it is an easy method of employment for patients and, in the case of glaucoma, could complement a patient’s regular use of eyedrops. Also, some of the ophthalmic drugs currently used to lower intraocular pressure appear to have a limited neuroprotective effect, making the use of these agents commercially important. 153
154
A.
Nickells and Schlamp
Application of Drug and Analysis of Effect
The delivery of drugs to the posterior pole of the eye by topical application to the cornea faces several challenges, including a long diffusion pathway to the target tissue, corneal impermeability to large molecules, and convectional flow of intraocular fluid in the opposite direction. Protocols designed to examine the effects of drugs delivered to the cornea must therefore demonstrate penetration to the posterior pole and biological efficacy. A recent and careful study by Mizuno et al. [2] to test the effects of nipradilol met these challenges in a study using male Japanese White rabbits.
B.
Penetration Studies
Wake rabbits were placed in restraining cages. Radiolabeled [3-14C]-nipradilol (Amersham Pharmacia Biotech) was applied to the lower cul-de-sac of the right eye twice daily (100 µL of a 1 % solution) for 7 days. Care was taken to apply the agent at the same times each day. Thirty minutes after the final application, the rabbits were anesthetized with phenobarbital and the eyes enucleated. The basic control for these experiments was to apply saline (or appropriate vehicle) to fellow eyes and to eyes of rabbits that did not receive radiolabeled agent. Enucleation was performed by the following procedure to minimize contamination of ocular tissues with the radiolabeled nipradilol. An incision was made along the outer edge of the orbit and the eyelids were clamped together with a surgical clip. The entire globe, complete with eyelids and surrounding tissues, was then removed as a pouch. The pouch was then immediately immersed in hexane and solid carbon dioxide for 2 min and then stored at ⫺15°C. Distribution of the [3-14C]-nipradilol was analyzed by dissecting the tissues of the eye under semi-frozen conditions, solubilizing them, and counting them in a scintillation counter. The actual amount of agent that had penetrated different tissues was calculated from the quench-corrected measurement of radioactivity and the known specific activity of the radiolabeled compound. Note that quench corrections are routinely performed automatically by modern scintillation counters and specific activity should be determined as a function of the half-life and the age of the isotope being used. Tissues that were analyzed in the Mizuno study included anterior chamber components (cornea, aqueous humor, iris, and lens) and posterior chamber components (vitreous and retina/choroid). After two applications every day for 7 days, the vast majority of labeled nipradilol was localized to the cornea (1284fold over controls) and iris, with moderate levels being detected in the aqueous humor. Substantially less nipradilol was detected in the lens, vitreous, and retina/ choroid (threefold over controls), but the levels in the retina were still significantly higher than in control eyes. Also, the effective concentration of nipradilol
Drug Delivery
155
in this tissue was estimated at 0.12 µM, which was in the clinically effective range of the drug. C. Efficacy Studies In many studies, efficacy is measured as the ability to prevent loss of target cells in the retina. It is important to note, however, that topically applied drugs could achieve these effects through systemic means (i.e., affecting ocular blood flow) rather than by directly acting on the target cells. To test the physiological effect of this dose of nipradilol, Mizuno et al. utilized a bioassay that took advantage of the antagonist effect of nipradilol on endothelin-1. Nipradilol has a vasodilating activity, whereas endothelin-1 has a vasoconstricting effect. Fundus photography was utilized to determine the diameter of the two major retinal arteries at the rim of the optic nerve head. The change in artery diameter was monitored with successive fundus photographs after an intravitreal injection of 20 µL of 5 ⫻ 10⫺8 M endothelin-1. These experiments showed that topical application of nipradilol provided a significant reduction in the vasoconstriction effects on intravitreally injected endothelin-1 up to 60 min, indicating that sufficient amounts of the topically applied drug had penetrated to target endothelial cells at the back of the eye. D. The Route of Trans-Corneal Delivery to the Back of the Eye Radiolabeled-drug studies suggest that the route of drugs administered to the cornea to the back of the eye is periocular, rather than through diffusion across the cornea and on into the vitreous. The evidence indicating this is the very low level of accumulation of drug in the vitreous. In addition, nipradilol (molecular weight of 326.35) is small enough to penetrate through the sclera, where the permeability constant is inversely proportional to the molecular weight for solutes between 285 and 70,000 daltons (see Sec. IV). Additionally, studies have shown that both timolol and betaxolol accumulate in the connective tissue of Tenon’s capsule in patients with prolonged usage of these drugs as topical drops [3], supporting this mode of access to the posterior pole.
III. INTRAVITREAL DELIVERY Intravitreal application is the most direct method for delivering a compound to the cells of the retina. It has been used to test recombinant viruses carrying neuroprotective genes [4], recombinant proteins such as growth factors [5–7], and small compounds such as flupirtine [8].
156
A.
Nickells and Schlamp
Intravitreal Injections
Injections into the vitreous are relatively straightforward and in most animals can be accomplished with a suitably small gauge needle. Injections are always made in anesthetized animals. Rats can be injected with a 27–30 G needle with little difficulty. Mouse eyes are more difficult to inject into and require a two-stage process of first piercing the conjunctiva and sclera with a 30-G needle followed by passing a glass micropipette through the hole into the eye [9]. Ideally the micropipette should be pulled mechanically using Drummond borosillicate glass capillary tubing (Cat. No. 3-000-210-G/ID 0.704 mm). Injections are easiest using needles with a long tapered point rather than a steep point typical of pipettes used for single cell injections or electrophysiology. Hand-pulled glass pipettes can also be used successfully. A total maximum volume of 2 µL can be injected into the mouse eye without damaging the retina. There are reports that up to 10 µL injections can be made into the rat eye. There are several disadvantages to intravitreal injections. The first is the transient time of exposure to the drug, since the turnover rate of fluids (and hence clearance of a drug) in the vitreous is quite high [1]. This may necessitate repeated injections, which are not trivial on small eyes, although, such have been performed successfully in the rat eye for flupirtine [8]. Repeated injections also run the risk of a variety of unwanted side effects, including cataract formation, endophthalmitis, retinal detachment, and vitreal hemorrhage [l0]. Alternatively, sustained-release devices have been implanted into the vitreous to allow a more continuous exposure of small molecules [11], and osmotic mini-pumps have been used successfully to provide continuous application of larger molecules such as brain-derived neurotrophic factor [12]. If the neuroprotective agent is a protein, long-term delivery can also be achieved by introduction of a transgene for the molecule into surrogate retinal cells by recombinant viral gene therapy (see Chapter 10). A second problem with intravitreal injections is leakage of the injected fluid from the injection site. This is particularly true in the mouse eye if the initial puncture wound is too large, resulting in a poor seal around the glass pipette. Last, great care must be taken not to damage anterior chamber structures, including the lens, with the injection needle. This can significantly complicate results by secondarily creating a neuroprotective environment in the eye [13]. Lens damage is often marked by cataract formation a few days after the injection and affected animals should be excluded from the data set in experiments employing intravitreal injections. In the mouse, making the initial puncture at a site just posterior to the limbus and then inserting the micropipette used for the injection at a relatively steep angle can avoid this damage. Care must then be taken to avoid penetrating the retina. It is recommended that initial attempts be made using a tracer dye to determine the success of the injections. With practice, it is possible to achieve highly consistent results.
Drug Delivery
157
IV. TRANS-SCLERAL DELIVERY Trans-scleral delivery of compounds is gaining favor as a method of applying neuroprotective agents to the retina, because compounds can be delivered to specific regions of the posterior pole without the complications of repeated intravitreal injections or the requirement for applying high concentrations of drug necessary for both trans-corneal and systemic delivery methods. The sclera is an elastic tissue composed of collagen fibrils and proteoglycans, which has a highly porous nature. Several in vitro diffusion studies using isolated pieces of sclera indicate that the rate of diffusion across this tissue is inversely proportional to the molecular weight of compounds between 285 and 70,000 molecular weight [1,14]). The molecular radius of molecules also influences diffusion, such that more globular compounds diffuse at a higher rate than linear structures of the same molecular mass [10]. More recent studies indicate that even large biomolecules, such as immunoglobulins, can penetrate the sclera [15] and retain their biological activity [16]. Methods of applying compounds to the eye for scleral diffusion fall into two categories depending on the amount and time course of drug delivery required. For short, one-time applications, drug is injected subconjunctivally as a single bolus. Diffusion can occur over a time period of minutes, but this rate can be influenced by factors such as molecular mass, as indicated above, and by scleral thickness, which varies from being relatively thin at the equator of a human eye (averaging 0.39 mm) to relatively thick (1.0 mm) at the optic nerve [17]. Prolonged drug release can be achieved by using a sustained-release method. Like intravitreal delivery, sustained-release methods range from osmotic pumps to preparations of drug in slow-release matrices such as cyanoacrylate adhesive. In addition, Tenon’s fibroblasts are readily transduced with recombinant DNA using adenoviruses [18] or naked plasmids absorbed into a collagen shield [19], methods that can be used to deliver genes encoding proteins small enough to penetrate the sclera. Because the transduced fibroblasts have been shown to express a transgene for several weeks post infection, this method may be used to achieve a biological alternative to a sustained-release device. Adenovirus infection can be achieved simply by a sub-conjunctival injection of virus, although this method often results in a diffuse spread of the infection around the globe (Fig. 1). A more discrete area of transduction can be achieved by soaking a cellulose sponge with virus and placing it under a flap of the Tenon’s capsule and conjunctiva in the region of interest. Significant levels of transduction are obtained using as little as a 5-min exposure. Like any other sustained-release device, the drawback with the gene transfer method is that it is not permanent and the application of recombinant virus would have to be repeated after between 14 and 30 days, when transgene expression becomes quiescent and/or the transduced cells are eliminated. Times of persistence may vary depending on the nature of the transgene
158
Nickells and Schlamp
Figure 1 Gross morphology of an enucleated rabbit eye (OD) showing distribution of a recombinant adenovirus injected subconjunctivally 4 days previously. The superior part of the eye is indicated with a suture in the cornea and temporal is to the left of the micrograph. A bolus of 100 µL (approximately 1010 virus particles) of recombinant adenovirus carrying the reporter gene β-galactosidase was injected by introducing a needle under the temporal conjunctiva and working it superiorly to the point indicated by the arrow. After 4 days, the animal was sacrificed and the eye enucleated, fixed briefly in paraformaldehyde and then stained for β-galactosidase activity. Stained cells are evident along the area of injection and needle track, while cells in the inferior nasal conjunctiva are not transduced. Histological analysis of transduced conjunctiva show that Tenon’s fibroblasts are readily transduced by adenovirus and therefore may be suitable targets for gene therapy using secreted small proteins that could diffuse across the sclera.
and the promoter being used to drive expression. In addition, this method must employ a transgene construct that allows for secretion of the target molecule.
V.
SYSTEMIC DELIVERY
Several systemic routes of drug delivery have been used successfully in ocular neuroprotection studies, including intraperitoneal or intravenous injections and oral delivery. Other modes of delivery include subcutaneous and intramuscular
Drug Delivery
159
injections. In general, systemic delivery is the preferred method for protocols that require repeated applications, although this requires the use of high concentrations of drug, which may have undesired side effects. Detailed descriptions of systemic delivery methods for different animals can be found in the appropriate volume of the Laboratory Animal Pocket Reference series. Methods used on rats and mice are described here. A. Oral Delivery One of the most dramatic uses of oral delivery for a neuroprotective agent was reported recently by Neufeld and colleagues [20]. In this study, the nitric oxide synthase inhibitor aminoguanidine was introduced into the drinking water of rats, half of whom had been made ocular hypertensive using the vortex vein occlusion method. The amount of drug taken in by each animal was determined by measuring the amount of water consumed. In this particular paradigm, the water was changed 3 times a week, and the rats were kept on the drug for a period of 6 months. Based on the concentration of drug put into the water (2.0 g/L) and the amount consumed (in mL/day), the authors estimated that the rats consumed about 60 mg of drug per day. Although this approach was successfully used in this study, drawbacks are several-fold. First, it is limited to drugs that are water-soluble. Second, it is important that the drug does not impair the behavior of the animal’s drinking habits, primarily by making the water unpalatable. This was a concern in the Neufeld study, which found that rats consuming aminoguanidine drank an average of approximately 25% less water than rats drinking untreated water, although this difference was not statistically significant based on the variation in the data set. Third, this method may be prone to overestimating the amount of drug consumed by not accounting for gravity-induced seepage of water from the drinking bottle. In a simple experiment testing for gravity flow from standard feeding bottles, it was estimated that 1–2 mL of water is lost per day in conditions where the bottles were undisturbed. It is likely that this loss is greater when animals are agitating the nozzles during drinking. More controlled drug studies interested in testing the use of oral delivery employ the gavage technique [21]. In this method, a drug is taken up as a metered dose in a syringe, which is then attached to a standard feeding or gavage needle with a blunt tip. In mice, it is possible to use a 30–22 G needle, either straight or curved depending on the size of the mouse, attached to a tuberculin syringe. Up to 100 µL can be introduced into the stomach of even small mice. To feed a mouse, it is essential that the animal is firmly restrained with complete immobilization of the head. The unanesthetized animal is fed by grasping it dorsally and articulating its head down and away from the holder to open the airway and esophagus. The needle is inserted down the esophagus into the
160
Nickells and Schlamp
stomach and the fluid is administered slowly (Fig. 2). With practice, an experienced technician can feed several hundred animals in a day using this method. The primary complication with gavage is entering the needle down the trachea and into the lungs. If this occurs the needle may become obstructed and the animal may begin to cough or struggle vigorously after the fluid is introduced. Fluid may also come out of the nose. At this point the needle should be withdrawn immediately. If it appears that fluid has been injected into the lungs, the animal should be euthanized. Another complication is the rupture of the esophagus or trachea due to excessive force when entering the needle. The resulting tissue damage can lead to hydrothorax after fluid application. The gavage method was
Figure 2 Ref. 21.)
Diagram showing the gavage method for drug delivery in mice. (From
Drug Delivery
161
used successfully to introduce FK506 into rats to test its neuroprotective effects on retinal ganglion cells after optic nerve crush [22]. The advantages of the gavage method is greater control over the dose and the timing of application of a target drug. The drawback with gavage feeding is that it is more labor-intensive for the researcher and disruptive to the animals, although they will become accustomed to repeated feedings by this method. B. Intraperitoneal Delivery Intraperitoneal injections can be a very efficient method of systemic drug delivery [21]. Injections are made into the caudal left abdominal quadrant to avoid the cecum in the right. The animal should be held dorsally with the head and tilted away and down from the holder (Fig. 3). The needle should be introduced with a quick firm motion to ensure passage through the abdominal musculature. After
Figure 3 21.)
Diagram showing intraperitoneal injections into the mouse. (From Ref.
162
Nickells and Schlamp
Figure 4 Diagram of subcutaneous injections into the mouse. (From Ref. 21.)
penetration, pull back on the syringe plunger. The presence of any fluid is indicative of penetration of an abdominal organ and the needle should be repositioned. Depending on the compound, injection into an abdominal organ should have no serious consequences other than altering the rate of absorption of the drug. Up to 1 mL of fluid can be introduced by this method into a mouse, but between 100 and 400 µL is typical. This method has been used successfully to test the neuroprotective effects of brimonidine on retinal ganglion cells [23,24]. An alternative to intraperitoneal injections is subcutaneous administration of drug (Fig. 4). In general, access of an injected compound to the bloodstream is less rapid than with intraperitoneal injections, but this is usually not an issue with experiments that require prolonged drug exposure. In addition, subcutaneous injections are less likely to damage the animal, and therefore may be more desirable for conditions that require repeated dosing. C.
Intravenous Delivery
Drugs introduced intravenously (I.V.) can reach the retina very rapidly. In humans, for example, fluorescein injected into a vein of the arm reaches the retinal arteriole system in 12–15 s and completely fills the vasculature of the retina and choroid within 25 s in a healthy person. The best veins for delivering drugs vary with the animal used. The most accessible site for an I.V. injection in the mouse and rat is in one of the lateral tail veins, which are easily found although quite small [21]. Injections are typically carried out by placing the mouse in a restraining device with a hole that allows the tail to stick out (Fig. 5). Warming the tail will dilate the vein for better detection. Injections are made with a 27-G
Drug Delivery
163
Figure 5 Diagram of intravenous injection into a mouse through the lateral tail vein. The mouse is constrained in a chamber with the tail extending through a hole. (From Ref. 21.)
needle (or smaller), which is inserted into the vein and extended 1–2 mm. A lack of resistance to the injected material is indicative of successful entry into the vessel. In some cases, it may be possible to see blanching of the vein as the injected fluid fills it. The major drawback with this method of delivery is the inability to successfully perform repeated injections on some animals (such as mice) because of scarring of the veins. ACKNOWLEDGMENTS This work was supported in part by NIH grant R29 EY 12223, a grant from the American Health Assistance Foundation, and by an unrestricted grant from Research to Prevent Blindness to the Department of Ophthalmology at the University of Wisconsin. The authors are grateful for the helpful discussions of Drs. Alyson Jarvis, T. Michael Nork, and Barbara Faha, and the assistance of Mr. Lou Kohl. REFERENCES 1.
Geroski DH, Edelhauser HF. Drug delivery for posterior segment eye disease. Invest Ophthalmol Vis Sci 2000; 41:961–964.
164
Nickells and Schlamp
2.
Mizuno K, Koide T, Yoshimura M, Araie M. Neuroprotective effect and intraocular penetration of nipradilol, a β-blocker with nitric oxide donation action. Invest Ophthalmol Vis Sci 2001; 42:688–694. Sponsel WE, Terry S, Khuu HD, Lam KW, Frenzel H. Periocular accumulation of timolol and betaxolol in glaucoma patients under long-term therapy. Surv Ophthalmol 1999; 43(suppl 1):S210–S213. Di Polo A, Aigner LJ, Dunn RJ, Bray GM, Aguayo AJ. Prolonged delivery of brain-derived neurotrophic factor by adenovirus-infected Mu¨ller cells temporarily rescues injured retinal ganglion cells. Proc Natl Acad Sci USA 1998; 95:3978– 3983. Mansour-Robaey S, Clarke DB, Wang Y-C, Bray GM, Aguayo AJ. Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells. Proc Natl Acad Sci USA 1994; 91: 1632–1636. Mey J, Thanos S. Intravitreal injections of neurotrophic factors support the survival of axotomized retinal ganglion cells in adult rats in vivo. Brain Res 1993; 602:304– 317. Sievers J, Hausmann B, Unsicker K, Berry M. Fibroblast growth factors promote the survival of adult rat retinal ganglion cells after transection of the optic nerve. Neurosci Lett 1987; 76:157–162. Nash MS, Wood JPM, Melena J, Osborne NN. Flupirtine ameliorates ischaemiclike death of rat retinal ganglion cells by preventing calcium influx. Brain Res 2000; 856:236–239. Li Y, Schlamp CL, Nickells RW. Experimental induction of retinal ganglion cell death in adult mice. Invest Ophthalmol Vis Sci 1999; 40:1004–1008. Lang JC. Ocular drug delivery: conventional ocular formulations. Adv Drug Delivery Res 1995; 16:39–43. Sanborn GE, Anand R, Torti RE. Sustained-release ganciclovir therapy for treatment of cytomegalovirus retinitis. Arch Ophthalmol 1992; 110:188–195. Sawai H, Clarke DB, Kittlerova P, Bray GM, Aguayo AL Brain-derived neurotrophic factor and neurotrophin-4/5 stimulate growth of axonal branches from regenerating retinal ganglion cells. J Neurosci 1996; 16:3887–3894. Leon S, Yin Y, Nguyen J, Irwin N, Benowitz LI. Lens injury stimulates axon regeneration in the mature rat optic nerve. J Neurosci 2000; 20:4615–4626. Olsen TW, Edelhauser HF, Lim JI, Geroski DH. Human scleral permeability: effects of age, cryotherapy, transscleral diode laser, and surigcal thinning. Invest Ophthalmol Vis Sci 1995; 36:1893–1903. Ambati J, Canakis CS, Miller JW, Gragoudas ES, Edwards A, Weissgold DJ, et al. Diffusion of high molecular weight compounds through the sclera. Invest Ophthalmol Vis Sci 2000; 41:1181–1185. Ambati J, Gragoudas ES, Miller JW, You TT, Miyamoto K, Delori FC, et al. Transscleral delivery of bioactive protein to the choroid and retina. Invest Ophthalmol Vis Sci 2000; 41:1186–1191. Olsen TW, Aaberg SY, Geroski DH, Edelhauser HF. Human sclera: thickness and surface area. Am J Ophthalmol 1998; 125:237–241. Perkins TW, Faha B, Kiland J, Poulsen G, Brumback L, Sinha D, et al. Effect of
3.
4.
5.
6.
7.
8.
9. 10. 11. 12.
13. 14.
15.
16.
17. 18.
Drug Delivery
19.
20.
21. 22. 23.
24.
165
p21 gene therapy on glaucoma filtering surgery in rabbits. Invest Ophthalmol Vis Sci 2000; 41:S580. Angella GJ, Sherwood MB, Balasubramanian L, Doyle JW, Smith MF, van Setten G, et al. Enhanced short-term plasmid transfection of filtration surgery tissues. Invest Ophthalmol Vis Sci 2000; 41:4158–4162. Neufeld AH, Sawada A, Becker B. Inhibition of nitric-oxide synthase 2 by aminoguanidine provides neuroprotection of retinal ganglion cells in a rat model of chronic glaucoma. Proc Natl Acad Sci USA 1999; 96:9944–9948. Suckow MA, Danneman P, Brayton C. The Laboratory Mouse. Boca Raton: CRC Press; 2001. Freeman EE, Grosskreutz CL. The effects of FK506 on retinal ganglion cells after optic nerve crush. Invest Ophthalmol Vis Sci 2000; 41:1111–1115. Yoles E, Wheeler LA, Schwartz M. Alpha2-adrenoreceptor agonists are neuroprotective in a rat model of optic nerve degeneration. Invest Ophthalmol Vis Sci 1999; 40:65–73. Donello JE, Padillo EU, Webster ML, Wheeler LA, Gil DW. Alpha2-adrenoreceptor agonists inhibit vitreal glutamate and aspartate accumulation and preserve retinal function after transient ischemia. J Pharm Exp Ther 2001; 296:216–223.
10 Recombinant Viral Vectors Przemyslaw Sapieha and Adriana Di Polo University of Montreal Montreal, Quebec, Canada
I.
INTRODUCTION
Gene transfer technology is gaining ground as a tool to investigate and promote neuroprotection in the retina. In the past few years, a number of studies have established proof-of-principle for the efficacy of gene delivery using viral vectors to enhance neuronal survival in animal models of retinal diseases. Several factors have contributed to the progress in this area, such as the elucidation of the genetic basis of inherited retinal diseases, the availability of natural, experimental, or transgenic animal models, and the development of recombinant viral vectors suitable for in vivo gene delivery. The design of appropriate neuroprotective strategies is the first step in tackling the complex problem of neuroprotection in the retina. A sensible strategy should consider several factors, including (1) the cell type affected; (2) the mechanism and cause of death; (3) the appropriate target cell for gene transfer; (4) the time-course of death; (5) the developmental stage of the experimental animal at the moment of therapy; (6) the optimal viral vector for gene delivery; (7) the time-course of vector-mediated transgene expression; and (8) the available animal models. There may be as many neuroprotective schemes as there are retinal diseases or injury models. The goal of this chapter is to provide the reader with an up-to-date account of current methods and tools for viral gene transfer that may serve as a guide to investigate neuroprotection in the retina. 167
168
Sapieha and Di Polo
II. NEUROPROTECTIVE STRATEGIES A variety of gene delivery strategies have been explored to promote neuronal survival in the retina following injury or disease. A gene transfer protocol that leads to the complete and permanent rescue of degenerating retinal neurons is yet to be established. However, the following strategies successfully enhanced cell survival by delaying, in some cases considerably, the time-course of neurodegeneration. A.
Gene Supplementation
Many retinal disorders are caused by genetic defects that result in a lack of one or more essential gene products. Delivery of a healthy copy of the mutated gene to the affected cell may restore loss-of-function deficits and lead to neuroprotection. This approach has been tested in the retinal degeneration (rd) mouse, a model of autosomal recessive human retinitis pigmentosa (RP). Defects in the rod specific cGMP phosphodiesterase β-subunit (β-PDE) underlie irreversible and rapid photoreceptor loss in this model [1,2]. Transient rescue of the rd phenotype was achieved following delivery of the wild-type β-PDE cDNA into the retina using several viral vectors; adenovirus (Ad) [3] encapsidated Ad (gutted vector) [4], adeno-associated virus (AAV) [5], and lentivirus (LV) [6]. Furthermore, Ali et al. demonstrated that supplementation of the peripherin-2 gene using AAV vectors resulted in the preservation of outer segment ultrastructure and function for up to 10 months in the rds mouse [7], a model for autosomal dominant RP and macular dystrophy. B.
Ribozyme Therapy
Retinal diseases caused by dominantly inherited mutations may result in the production of abnormal gene products that affect cellular trafficking, metabolism, and function. Cell death due to accumulation of these toxic products can be minimized by delivery of ribozymes, small RNA molecules that can be designed to cleave mutant RNA transcripts while leaving the wild-type mRNAs intact. Lewin et al. demonstrated morphological and functional protection of photoreceptors following AAV-delivery of ribozymes in a mutant rhodopsin transgenic rat model of autosomal dominant RP [8]. Long-term studies on the effectiveness of this form of therapy indicated that the progression of photoreceptor loss was considerably delayed for up to 8 months of age [9]. C.
Neurotrophic Factor Therapy
The main limitation of the gene correction therapies mentioned above is that the genetic basis of prevalent forms of retinal disorders, such as glaucoma or age-
Recombinant Viral Vectors
169
related macular degeneration, remains unknown. In addition, retinal diseases that arise from mutations in several genes (polygenic diseases) or a combination of environmental and genetic factors represent a challenge in the design of potential gene therapies. Alternative therapies would involve the use of neurotrophic factors, with a broad spectrum of action, capable of providing a more generic form of neuroprotection. Neurotrophic factors bind to specific cell surface receptors triggering intracellular signaling pathways that lead to neuronal survival [10– 12]. Because most neurotrophic factors are secreted and diffuse well within the retina, this type of approach allows one to select either the affected cell or its supporting cells as targets for gene transfer. For example, delivery of the brainderived neurotrophic factor (BDNF) gene to Mu¨ller glial cells using an Ad vector resulted in temporary protection of retinal ganglion cells in an optic nerve axotomy rat model [13]. Ad-mediated delivery of ciliary neurotrophic factor (CNTF) has been shown to slow photoreceptor loss in the rd and rds mouse [14,15]. Basic fibroblast growth factor (bFGF) delivered by Ad vectors delayed retinal degeneration in the Royal College of Surgeons (RCS) rat [16]. More recently, AAV-mediated delivery of bFGF delayed photoreceptor cell death in transgenic rats carrying a mutant rhodopsin [17] and retinal ganglion cell loss after axotomy [18]. D. Inhibition of Apoptosis Apoptosis or programmed cell death appears to be a common mechanism of neuronal loss in the injured or degenerating retina. Retinal ganglion cells have been shown to die by apoptosis after optic nerve axotomy and in experimental glaucoma [19–21]. Photoreceptors also die by apoptosis in several inherited mouse models of RP [22,23], the RCS rat [24], light-induced retinal damage [25] and experimental retinal detachment in the rat [26]. The identification of intracellular components of the cell death program (e.g., bcl-2 family members and caspases) has motivated experimental strategies involving transfer of antiapoptotic genes. For example, gene delivery of bcl-2 using Ad resulted in moderate protection of photoreceptors in the rd mouse model [27]. Recently, Ad-mediated gene transfer of the X-linked inhibitor of apoptosis protein (XIAP) resulted in partial protection of retinal ganglion cells following axotomy [28] and high intraocular pressure [29].
III. SELECTION OF A GENE DELIVERY SYSTEM Genes can be introduced into cells via nonviral and viral vectors. Nonviral methods include liposome-mediated DNA transfer, DNA carried on ballistic metal particles (“gene gun”), and micro-injection techniques [30]. These gene transfer
Photoreceptors Retinal ganglion cells RPE
RPE Photoreceptors
Adenoassociated virus (AAV)
Lentivirus (LV)
Efficient Can infect stem cells
Low Requires helper virus
Efficient
Infection in vitro
Peak: N/D Duration: ⬃12 weeks (longest time examined)
Peak: ⬃5–7 days Duration: 1–3 weeks Several months with immunosuppression Plateau: 2–4 weeks Duration: 3 months to ⬎ 1 year
Time-course retinal transgene expression
wtAAV: Chromosomal integration rAAV: chromosomal integration in vitro N/D in vivo wt LV: Chromosomal integration rLV: Chromosomal integration
wt Ad: Episomal rAd: Episomal
Localization of transferred DNA
P3/P4 Biosafety Level Facility
P1/P2 Biosafety Level Facility
P2 Biosafety Level Facility
Biosafety (NIH guidelines)
1010 –1011 pfu/ mL
1012 –1014 pp/ mL 1010 –1012 ip/ mL
108 –109 iu/mL Packaging cell lines: 106 iu/mL
⬃4.5 kb
⬃10 kb
Titer stocks
⬃8 kb
Cloning capacity
Abbreviations: RPE: retinal pigment epithelium; r: recombinant; wt: wild-type; pfu: plaque forming unit; pp: physical particles; ip: infectious particles; iu: infectious units; N/D: not determined.
Mu¨ller cells RPE Photoreceptors (early development)
Adenovirus (Ad)
Main target retinal cells in vivo
Viral Vectors Widely Used in Retinal Neuroprotection Research
Viral vector
Table 1
170 Sapieha and Di Polo
Recombinant Viral Vectors
171
methods often yield low DNA transduction rates that result in limited transgene expression in vivo. In contrast, viral vectors have the ability to efficiently infect proliferating and post-mitotic cells and support adequate transgene expression levels. Improvements in the design of these vectors and the methods to increase viral titers and purity have contributed to their popularity. What makes a viral system suitable for therapeutic use? Let us imagine an ideal vector with the following characteristics: (1) it can transduce a large number of target cells; (2) it mediates gene expression specifically in the target cells; (3) it allows stable expression of the delivered gene product at adequate levels; (4) it is safe, entailing no toxic or immunogenic response in the target tissue; (5) it has no limitation in the size of DNA that it can accommodate; and (6) it is easy to produce in large volumes of high purity, high titer stocks. Although such a perfect vector does not yet exist, there are a number of useful viral vectors available for gene transfer research. Here we will focus on Ad, AAV, and LV vectors, the most widely used vectors in retinal neuroprotection research (Table 1). The following sections are not intended to endorse any vector in particular but to highlight their advantages and disadvantages for different gene transfer applications.
IV. ADENOVIRUS (Ad) Ad vectors are an efficient tool to study the effect of in vitro and in vivo gene expression on neuroprotection. The major advantages of this system are that Ad can efficiently infect post-mitotic cells and that it can be easily grown to high titers. Ad contains a linear double-stranded DNA genome of approximately 36 kilobases (kb) encapsidated in an icosahedral protein shell. Immediate early genes (E1, E2, E3, and E4) orchestrate viral gene transcription and suppression of the host immune response, whereas late genes are necessary for viral assembly [31]. Most recombinant Ad belong to the group C Ad type 2 or 5. Ad vectors were initially generated with deletions of the early region 1 (∆E1), which contains genes required for virus replication [32], rendering vector replication defective and more suitable for gene transfer into mammalian cells. A major disadvantage of these early Ad vectors is the strong cytotoxic and immune response elicited upon infection of the host cells [33]. Recent versions of Ad vectors have been produced in which the entire viral genome, except for the terminal repeat regions required for viral assembly, has been replaced by exogenous gene sequences. These so-called “gutless” vectors exhibit considerably reduced immune response [34,35], but can only be produced in the presence of a helper virus that provides all the proteins required for viral replication [36]. Because it is often difficult to fully separate the “gutted” vector from the helper virus, we will focus here on novel methods to produce recombinant Ad vectors without the need for helper virus (see below).
172
A.
Sapieha and Di Polo
Ad Tropism in the Retina
Although Ad vectors show efficient transduction of a wide variety of cell types in vitro, the cellular tropism for viral infection in vivo appears to be more complex. Our studies using Ad vectors injected into the vitreous chamber of adult rat eyes demonstrated that Muller cells, the predominant glial cell in the retina, is the main target for Ad infection in vivo [13]. This approach has proved to be useful for delivery of genes encoding diffusible neurotrophins to promote neuroprotection of axotomized retinal ganglion cells. Ad vectors have also been shown to efficiently transduce the retinal pigment epithelium (RPE) following subretinal injections [37,38]. Together, these studies indicate that nonneuronal cells in the adult retina are the preferred cellular targets for recombinant Ad. Nevertheless, some experimental conditions support limited Ad transduction of retinal neurons. For example, intraocular administration of Ad in animals at early developmental stages may result in modest infection of photoreceptors [3]. In addition, introduction of Ad to the brain (e.g., superior colliculi) or to the transected optic nerve stump results in retrograde transport of viral particles that mediate gene expression in retinal ganglion cells [28,39]. Because all these studies involved transgenes directed by the ubiquitous cytomegalovirus (CMV) promoter, it is likely that the presence of cellular receptors for Ad in nonneuronal retinal cell types mediates the observed viral tropism in vivo. The coxsackievirus and adenovirus receptor (CAR) protein involved in Ad attachment and infection has been identified [40]. In addition, integrins αvβ3 and αbβ5 participate in Ad internalization [41]. The specific cellular localization of these receptors in the retina remains to be defined.
B.
Immunological Response to Ad and Transgene Expression
Most studies have shown robust Ad-mediated transgene expression detectable soon after vector administration, but this expression is transient and decreases within a few weeks. This has been attributed mainly to a host immune reaction to viral and transgene products in which the infected cells are targeted for rapid T-cell mediated clearance [42,43]. This is supported by the observation that Admediated gene expression is prolonged in immunodeficient animals [42]. It is possible that other factors such as promoter silencing or loss of viral DNA, since the Ad genome remains episomal, contribute to this limited expression [44]. As such, Ad represents an attractive system in gene therapy applications that require the expression of a gene product for a limited amount of time. For applications that require sustained transgene expression, several approaches have been explored. Repeated administration of Ad to enhance transgene expression has proved to be inefficient due to the production of neutralizing antiviral antibodies
Recombinant Viral Vectors
173
[45,46]. Alternative strategies have been implemented to by-pass the natural immune response. Daily administration of immune-suppressant drugs, delivered by subcutaneous injections [13] or osmotic minipumps, can effectively sustain transgene expression in the retina for several months. Similarly, coadministration of modulators of the immune response, such as CTLA4-Ig, prolong Ad-mediated gene expression [47]. The ocular immune response elicited by Ad has been shown to depend on the route of administration of the vector. It is now clear that injection of Ad vectors into the vitreous space leads to a stronger immunological reaction than subretinal injection [3,42]. This finding highlights important differences in the ocular immune privilege response between the intravitreal and subretinal compartments following Ad injection. It is interesting that, the immune response triggered by intravitreal administration of Ad vectors has been shown to promote a moderate degree of neuroprotection [13,43]. This has been partially attributed to the production of cytokines and neurotrophic factors by activated T cells [48]. C. Preparation of Ad Vectors E1 or E1/E3 deleted Ad vectors are widely used in most gene transfer applications. Most methods to generate such vectors rely on standard molecular biology and tissue culture techniques. Traditional protocols for the production of recombinant Ad involve homologous recombination of two pieces of DNA: (1) a plasmid carrying the gene of interest, usually replacing the E1 genes, and the 5′ end of the Ad genome, and (2) the 3′ end of the Ad genome. Both DNA constructs are cotransfected into low passage 293 cells, a human embryonic kidney cell line engineered to express the E1 gene products in trans to allow replication and propagation of Ad [49]. Recombination events result in the production of functional viral particles that have incorporated the gene of interest into their genome. Single viral clones are identified as plaques using standard overlay techniques and are further characterized by polymerase chain reaction (PCR) or dot-blot analysis of viral DNA. Viral clones containing the gene of interest are then propagated in 293 cells. The purification of Ad preparations is routinely done by cesium chloride gradient ultracentrifugation and the viral titer is established using standard plaque titration assays. A detailed protocol to prepare Ad using this method has been described elsewhere [50]. Commercial kits for the production of Ad vector using this technique are available (Qbiogene, Carlsbad, CA). This approach has proved to be useful but time-consuming due to the low efficiency of homologous recombination in mammalian cells and the need for repeated rounds of plaque purification to eliminate contamination with wild-type virus. A modification of this approach has been developed where Ad is reconstituted in yeast [51] or bacteria [52,53], which are endowed with a more efficient recombination machinery and are easier to manipulate than mammalian cells. In
174
Sapieha and Di Polo
addition, the recombinant viral DNA is isolated from single clones, allowing one to generate homogeneous Ad preparations, free of contaminating wild-type virus, following transfection into 293 cells. Recently, a novel method that does not require homologous recombination has been developed to generate E1/E3- or E1/E4-deleted recombinant Ad by in vitro ligation [54,55]. First, the gene of interest is inserted into a shuttle plasmid to generate an expression cassette. Then, this cassette is excised and ligated into another vector containing the complete Ad genome with E1/E3 or E1/E4 deletions. The resulting recombinant Ad DNA is grown and purified following transformation into bacteria. This DNA is then used to transfect 293 cells where viral particles carrying the desired expression cassette are propagated. Because transfection is done with DNA from a single clone, there is no need to screen plaques following transfection. Commercial kits to generate recombinant Ad by this method are available (Adeno-X, Clontech Laboratories, Palo Alto, CA). One drawback of this technique is the need to transform bacteria and transfect 293 cells with large plasmids, which may result in poor yields. Alternative methods using cosmid technology have been designed to select clones containing full-size genomes while excluding small and incomplete DNAs that are often produced following transformation of bacteria with large plasmids [56].
V.
ADENO-ASSOCIATED VIRUS (AAV)
AAV is a member of the parvoviridae family initially identified as a contaminant of Ad stocks. AAV requires a helper virus (e.g., Ad or herpes simplex virus) for replication. The wild-type virus houses a single-stranded genome of 4680 base pairs (bp) containing two genes, rep and cap, that encode proteins involved in replication and encapsidation, respectively. The AAV genome is flanked by two identical 145-bp inverted terminal repeats (ITRs), which are essential for packaging, replication, or integration. Recombinant AAV vectors derived from human parvovirus AAV-2 have been produced by substituting all viral sequences, except for the ITRs, for a transgene of interest [57,58]. However, packaging of functional AAV particles requires the presence of the rep and cap proteins typically provided in trans. The recombinant AAV system has several advantages for in vivo gene transfer research: (1) it is not pathogenic and has not been implicated in the etiology of any known human disease [59]; (2) mediates long-term transgene expression that persists for several months in vivo [60,61]; (3) the absence of viral sequences results in minimal immune response or cytotoxicity in the target tissues [62,63]; and (4) it can efficiently infect post-mitotic cells in vivo [64]. In the absence of helper virus, wild-type AAV can integrate at a specific site on the q arm of chromosome 19 to establish latent infection [65]. However, the lack of rep proteins has been shown to compromise integration specificity
Recombinant Viral Vectors
175
leading to random insertion of recombinant AAV [66,67]. Although viral integration into the genome may contribute to the stability of AAV-mediated transgene expression, a careful evaluation of the risks associated with insertional mutagenesis is required before implementing AAA-based therapies. A disadvantage of AAV vectors has been the size constraint for packaging genes larger than 4.7 kb. Although methods have been developed to increase the size of delivered transgenes by trans-splicing of two independent vectors coadministered to the same tissue [68], this remains a limitation of the AAV system. The laborious work needed to produce AAV vector stocks has often been regarded as a disadvantage of this vector system; however, recent availability of reagents and improvements in the protocols, described below, have greatly facilitated the preparation of high-titer and pure AAV stocks. A. AAV Tropism in the Retina AAV-mediated gene expression can be restricted to photoreceptor cells when under the control of a well-characterized murine rhodopsin promoter sequence [69]. More recently, retinal ganglion cells have been identified as the primary targets for AAV infection in the inner retina following intravitreal injection of viral vectors [70]. Thus, subretinal injection of AAV vectors results mainly in gene transfer to photoreceptors and RPE cells, whereas intravitreal injection allows infection of cells in the ganglion cell layer. Unlike Ad, AAV appears to have a preferential tropism for retinal neurons rather than glial cells. This is consistent with studies in the brain showing AAV transduction of subsets of neurons rather than astrocytes, oligodendrocytes, or microglia [71,72]. It is interesting that genetic modification of capsid proteins has been shown to allow AAV targeting of cells normally resistant to infection [73]. B. Time Course of AAV-Mediated Transgene Expression Recombinant AAV vectors have a slow onset of detectable transgene expression in the retina, which typically reaches a plateau between 1 and 8 weeks following administration of the vector depending on the animal species [74] In rodents, peak expression is normally found 3 to 6 weeks postinjection of the vector [60,70,75,76]. Similar time-dependent increases in AAV-mediated expression have been observed in brain [77], muscle [78], and liver [79]. The mechanism for this in vivo delay to reach peak expression levels in the retina has not been determined, but may be related to the requirement of the single-stranded AAV genome to be converted to a double-stranded form for the transgene to be expressed [80,81]. This feature of AAV-mediated gene expression should be taken into consideration when designing neuroprotective strategies. Experimental para-
176
Sapieha and Di Polo
digms with early onset and fast retinal neurodegeneration may be less amenable to AAV-based therapies. Interestingly, AAV mediates long-term transgene expression that can persist in the retina for at least one year after vector administration [60,61]. C.
Preparation of AAV Vectors
Because AAV depends on the presence of certain Ad helper genes to replicate and mount an effective infection, traditional methods have relied on the use of Ad for packaging recombinant AAV. These protocols involve cotransfection of two constructs, a vector plasmid carrying the transgene expression cassette, and a helper plasmid that provides the rep and cap genes, into cells infected with helper Ad. Although separation of the resulting recombinant AAV from Ad has been routinely performed on cesium chloride (CsCl) gradients, via column chromatography, or by heat-inactivation, Ad remains a frequent contaminant of these preparations. The presence of helper virus can result in unwanted cellular host immune response [82], confounding the interpretation of the results following in vivo gene transfer of contaminated AAV stocks. These concerns were addressed by the development of constructs that provide the essential helper genes in a plasmid, but lack the structural and replication Ad genes [63,83–85). For example, the AAV rep and cap genes and the Ad helper genes have been introduced in a single helper plasmid, pDG [83], eliminating the need for Ad infection. In addition, the use of helper plasmids has been shown to improve the titer of recombinant AAV stocks possible due to strong cap gene expression that enhances virus encapsidation [86]. Current methods of AAV preparation in most laboratories are based on cotransfection of low passage 293 cells with a vector plasmid containing the transgene of interest flanked by the ITRs and a helper plasmid. Once the AAV particles are assembled within the host cells, the virus is extracted by freezing and thawing the cells and subsequently purified. Traditional AAV vector purification protocols that involve precipitation of the virus with ammonium sulfate followed by repeated rounds of CsCl density gradient centrifugation lead to poor viral recovery and low infectivity. A new purification strategy based on density gradient purification using iodixanol has been reported [87]. Iodixanol (Nycomed Pharma, Roskilde, Denmark) is a nonionic iodinated density gradient media, which, unlike CsCl gradients, does not promote aggregation of AAV and cell proteins that normally make the purification of AAV particles difficult. The fraction containing the AAV is further purified on a heparin/agarose column. This step relies on the high affinity of this virus for heparan sulfate proteoglycan [88], its main cellular receptor for attachment and internalization [89]. The eluted fraction containing the virus is then concentrated and its titer is estimated. A detailed protocol describing this procedure has been described elsewhere [87].
Recombinant Viral Vectors
177
In the absence of helper virus, AAV cannot mount a productive (lytic) infection. Exposure of cells to AAV does not result in the formation of plaques as with Ad. Consequently, indirect methods are used to estimate how much virus is present in a given stock. Because not all virion particles are necessarily capable of successful infection, virus quality is often assessed as the ratio of physical particles to infectious particles in a given stock. A quantitative competitive (QC)PCR method is currently used to estimate total viral particles or copies of recombinant AAV genomes [90]. DNA dot blot analysis can also be used to estimate total viral particles [91], but it is more time-consuming and unreliable than PCRbased methods. Infectious particles are determined by a infectious center assay (ICA) using the C12 cell line [92] that contains integrated wild-type AAV rep and cap genes. ICA involves infection of C12 cells with the AAV preparation followed by infection with Ad. The recombinant AAV genome is amplified in those cells successfully infected by AAV virions. Subsequently, the number of cells expressing detectable amounts of recombinant AAV DNA are quantified using radiolabeled probes [87]. Methods for manufacturing high-titer AAV stocks based on transient transfection of 293 cells in the absence of infectious helper virus have proved particularly useful for preclinical studies that require rapid testing of therapeutic genes in a variety of animal models. While this method provides versatility, it is laborious and expensive to scale up due to the large amounts of DNA needed for transfection and the low efficiency of cotransfection protocols. As AAV vectors move toward the clinical arena, improved methods to produce large quantities of pure, high-titer stocks are required. Packaging cell lines engineered to contain the AAV rep and cap genes have been particularly useful [92,93]. The transgene of interest can be delivered to packaging cells via a recombinant Ad to generate recombinant AAV stocks in the presence of helper Ad [94,95]. Alternatively, the rep and cap genes have been delivered via a recombinant herpesvirus to cells that carry a stably integrated AAV provirus with the transgene of choice [96]. These strategies are promising for scaling up AAV stocks, but they inevitably result in contamination with helper viruses and require thorough purification prior to any clinical application.
VI. LENTIVIRUS Lentivirus (LV), a genus of retroviruses, consists of two identical single-stranded RNA molecules and enzymes required for replication within a viral protein core. Following virus internalization, the viral RNA is reverse-transcribed into doublestranded DNA and transported to the cell nucleus [97]. Viral DNA is then permanently integrated into the host genome to become a provirus. The retrovirus genome contains gag, pol, and env genes flanked by long-terminal repeats (LTRs).
178
Sapieha and Di Polo
These genes encode proteins essential for replication, encapsidation, internalization, and reverse transcription. Replication-deficient recombinant retroviral vectors have been generated by substituting all viral genes for a transgene of interest with the exception of the cis-acting sequences required for vector propagation, such as the reverse transcription initiation site and the packaging site [98]. Functional recombinant retrovirus particles can be generated in culture when the gag, pol, and env gene products are provided in trans. Most retroviral vectors can only transfer genes into cells that are actively proliferating [99]. Thus, their use in neuroprotective strategies that typically involve gene transfer into fully differentiated cells is rather limited. An exception to this rule are LVs, such as the human immunodeficiency virus (HIV), which can efficiently infect nonmitotic cells [100]. This ability relies on nuclear localization signals in the preintegration complex that allow entry into the nucleus without the need for nuclear membrane fragmentation [99]. Other advantages of the LV system are its relatively large cloning capacity, close to 10 kb; its ability to mediate high levels of transgene expression in vivo; and the lack of immune response in the target tissues [101,102]. Recently, recombinant LV was shown to efficiently infect hematopoietic stem cells, extending its potential use as a therapeutic vector [103]. The main concern with LV vector systems is the risk of generating replication competent recombinant (RCR) virus during the production of viral stocks. Because HIV is a human pathogen, considerable work has been done to increase biosafety of LV production systems. The latest methods to generate safer HIVbased vector systems are described below. Other concerns include low vector titers and the risks associated with insertional mutagenesis as the vector integrates into the host genome. The issue of transgene silencing still requires further investigation. The lack of immune response associated with LV recombinant infection and its ability to stably integrate into host DNA are promising features for persistent gene expression. A systematic study of the time course of expression mediated by LV vectors in the retina should resolve this issue. A.
Cellular Tropism of LV Vectors
LV-mediated gene transfer and expression in the retina in vivo was first characterized by subretinal injection of a vector carrying the green fluorescent protein (GFP) gene [104]. Using the ubiquitous CMV promoter, the main cellular targets of LV vectors were shown to be photoreceptors and RPE cells with some bipolar and Mu¨ller cells. When a rhodopsin-specific promoter was used, transgene expression was restricted to the photoreceptor layer [6,103]. The infection pattern of LV vectors following intravitreal injection remains to be characterized. An attractive feature of HIV-based vectors is their ability to efficiently infect cells in vitro. For example, RPE primary cultures have been transduced
Recombinant Viral Vectors
179
with LV to express GFP. Transgene expression persisted in the infected RPE cells following their transplantation into the subretinal space of a host [105]. HIV-based vectors have been shown to efficiently transduce human CD34⫹ hematopoietic stem cells that were capable of long-term engraftment of nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice [103]. Recently, neural stem cells have been isolated from the RPE in the ciliary margin of the adult eye [106]. These cells can express retinal specific markers when they are differentiated in vitro. Thus, ex vivo transduction of stem cells using HIV-based vectors followed by retinal transplantation may be useful in the design of neuroprotective strategies. B. Preparation of LV Vectors LV has a more complex genome and, consequently, a more complicated replication cycle than other retroviruses. In addition to the three structural gag, pol, and env genes, LV has additional regulatory genes: vif, vpr, vpu, tat, rev, and nef genes [107]. Traditional methods to generate recombinant LV vectors involved a vector plasmid with all viral genes deleted, containing only the transgene of interest flanked by the essential cis-acting elements. The viral proteins required for virus propagation were provided in trans by cotransfection of a helper plasmid into 293 cells. This procedure resulted in low titer preparations and in high risk for generating RCR virus. This is a major concern because the considerable overlap that exists between cis-acting sequences in HIV vectors and helper constructs increases the chance of homologous recombination leading to the production of RCR viral particles. A better understanding of the HIV replication cycle has led to new advances in the field of LV vector packaging. The latest generation of LV vectors have only three of the nine HIV-1 genes: gag, which encodes the main structural proteins; pol, responsible for the production of viral replication proteins and rev, encoding a regulator required for gag and pol gene expression. The HIV envelope has been substituted by other viral envelope proteins such as the vesicular stomatitis virus G-protein (VSV-G) due to its high stability and broad tropism [108]. Methods to generate recombinant LV particles from three or four separate transcriptional units containing gag/pol, rev, VSV-G, and the transgene of interest following transfection into 293 cells have been recently described [101,109]. By distributing the required sequences in multiple plasmids, the risk of creation of RCR virions was minimized. In addition, the overlap between vector and helper sequences was markedly reduced. The safety level for production of LV has been further improved by the development of self-inactivating (SIN) HIV-1-derived vectors. Two independent research groups have reported HIV-based vectors in which the regulatory sequences (promoter and enhancer elements) contained in the 3′ LTR have been
180
Sapieha and Di Polo
deleted [110,111]. By eliminating these cis-acting elements, the virus loses its capacity for gene expression directed by both LTRs, resulting in inactivation of the integrated virus in the infected cells. In addition, the expression of the transgene of interest can only be controlled by an internal heterologous promoter. A stable, high-output packaging cell line to produce this latest generation of LV vectors has been recently reported [112]. These results represent the latest advancement in the efforts to render HIV-based vectors safe for clinical use. Although these novel vectors greatly minimize the risk of RCR virus infection to users during manipulation and preparation, the biosafety of LV vectors in vivo has yet to be demonstrated.
VII. SUMMARY Several independent research laboratories have recently established proof-ofprinciple for the efficacy of gene therapy for neuroprotection in animal models of retinal injury or disease. Three viral systems, Ad, AAV, and LV, have been used for most gene delivery applications to the retina both in vitro and in vivo. The pros and cons inherent in each of these vectors need to be carefully weighed when designing a gene transfer strategy. Recent progress in the methods to create these recombinant viruses has greatly increased the ability of researchers to generate high-titer and pure stocks while reducing production times. In addition, the likelihood of adverse reactions in the host tissues and the risk of RCR virus generation have been minimized. As these viral vectors progress toward preclinical and potential clinical human trials, highly stringent standards for their efficacy, biosafety, ease for scaling up production and purity will be expected. Future in vivo characterization of these viral vectors will be essential to fully assess if these requirements are met for clinical applications.
ACKNOWLEDGMENTS Supported by the Canadian Institutes of Health Research (CIHR) and the Foundation Fighting Blindness. A.D.P. is a scholar of Fonds de la Recherche en Sante´ du Que´bec.
REFERENCES 1.
Bowes C, Li T, Danciger M, Baxter LC, Applebury ML, Farber DB. Retinal degeneration in the rd mouse is caused by a defect in the beta subunit of rod cGMPphosphodiesterase. Nature 1990; 347:677–680.
Recombinant Viral Vectors 2.
3.
4.
5.
6.
7.
8.
9.
10. 11. 12. 13.
14.
15.
16.
181
Pittler SJ, Baehr W. Identification of a nonsense mutation in the rod photoreceptor cGMP phosphodiesterase beta-subunit gene of the rd mouse. Proc Natl Acad Sci USA 1991; 88:8322–8326. Bennet J, Pakola S, Zheng Y, Maguire AM. Humoral response after administration of El-deleted adenoviruses: immune privilege of the subretinal space. Hum Gene Ther 1996; 7:1763–1769. Kumar-Singh R, Farber DB. Encapsidated adenovirus mini-chromosome-mediated delivery of genes to the retina: application to the rescue of photoreceptor degeneration. Hum Mol Genet 1998; 7:1893–1900. Jomary C, Vincent K, Grist J, Neal M, Jones S. Rescue of photoreceptor function by AAV-mediated gene transfer ina mouse model of inherited retinal degeneration. Gene Ther 1997; 4:683–690. Takahashi M, Miyoshi H, Verma IM, Gage FH. Rescue from photoreceptor degeneration in the rd mouse by human immunodeficiency virus vector-mediated gene transfer. J Virol 1999; 73:7812–7816. Ali RR, Sarra GM, Stephens C, Alwis MD, Bainbridge JW, Munro PM, Fauser S, Reichel MB, Kinnon C, Hunt DM, Bhattacharya SS, Thrasher AJ. Restoration of photoreceptor ultrastructure and function in retinal degeneration slow mice by gene therapy. Nat Genet 2000; 25:306–310. Lewin AS, Drenser KA, Hauswirth WW, Nishikawa S, Yasumura D, Flannery JG, and LaVail MM. Ribozyme rescue of photoreceptor cells in a transgenic rat model of autosomal dominant retinitis pigmentosa. Nat Med 1998; 4:967–971. LaVail MM, Yasumura D, Matthes MT, Drenser KA, Flannery JG, Lewin AS, and Hauswirth WW. Ribozyme rescue of photoreceptor cells in P23H transgenic rats: long-term survival and late-stage therapy. Proc Natl Acad Sci USA 2000; 97:11488–11493. Lewin GR, Barde Y-A. Physiology of the neurotrophins. Ann Rev Neurosci 1996; 19:289–317. Kaplan DR, Miller FD. Neurotrophin signal transduction in the nervous system. Curr Opin Neurobiol 2000; 10:381–391. Huang E, Reichardt LF. Neurotrophins: roles in neuronal development and function. Ann Rev Neurosci 2001; 24:677–736. Di Polo A, Aigner LJ, Dunn RJ, Bray GM, Aguayo AJ. Prolonged delivery of brain-derived neurotrophic factor by adenovirus-infected Mu¨ller cells temporarily rescues injured retinal ganglion cells. Proc Natl Acad Sci USA 1998; 95:3978– 3983. Cayouette M, Gravel C. Adenovirus-mediated gene transfer of ciliary neurotrophic factor can preven photoreceptor degeneration in the retinal degeneration (rd) mouse. Hum Gene Ther 1997; 8:423–430. Cayouette M, Behn D, Sendtner M, Lachapelle P, Gravel C. lntraocular gene transfer of ciliary neurotrophic factor prevents death and increases responsiveness of rod photoreceptors in the retinal degeneration slow mouse. J Neurosci 1998; 18: 9282–9293. Akimoto M, Miyatake S, Kogishi J, Hangai M, Okazaki K, Takahashi JC, Saiki M, Iwaki M, Honda Y. Adenovirally expressed basic fibroblast growth factor rescues photoreceptor cells in RCS rats. Invest Ophthalmol Vis Sci 1999; 40:273–279.
182 17.
18.
19.
20.
21.
22. 23.
24.
25. 26. 27.
28.
29.
30. 31. 32.
33.
Sapieha and Di Polo Lau D, McGee LH, Zhou S, Rendahl KG. Mannin WC, Escobedo JA, Flannery JG. Retinal degeneration is slowed in transgenic rats by AAV-mediated delivery of FGF-2. Invest Ophthalmol Vis Sci 2000; 41:3622–3633. Sapieha M, Cheng L, Rendahl K, Ladner M, Manning WC. Adeno-associated virus-mediated delivery of FGF-2, FGF-5, and FGF-18 genes protects injured retinal ganglion cells in vivo. Invest Ophthalmol Vis Sci 2001; 42:S23. Berkelaar M, Clarke DB, Wang Y-C, Bray GM, Aguayo AJ. Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats. J Neurosci 1994; 14:4368–4374. Garcia-Valenzuela E, Shareef S, Walsh J, Sharma SC. Programmed cell death or retinal ganglion cells during experimental glaucoma. Exp Eye Res 1995; 61:33– 44. Quigley HA, Nickells RW, Kerrigan LA, Pease ME, Thibault DJ, Zack DJ. Retinal ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis. Invest Opthalmol Vis Sci 1995; 36:774–786. Chang GQ, Hao Y, Wong F. Apoptosis: final common pathway of photoreceptor death in rd, rds, and rhodopsin mutant mice. Neuron 1993; 11:595–605. Portera-Cailliau C, Sung CH, Nathans J, Adler R. Apoptotic photoreceptor cell death in mouse models of retinitis pigmentosa. Proc Natl Acad Sci USA 1994; 91:974–978. Tso MO, Zhang C, Abler AS, Chang CJ, Wong F, Chang GQ, Lam TT. Apoptosis leads photoreceptor degeneration in inherited retinal dystrophy of RCS rats. Invest Ophthalmol Vis Sci 1994; 35:2693–2699. Abler AS, Chang CJ, Ful J, Tso MO, Lam TT. Photic injury triggers apoptosis of photoreceptor cells. Res Com Mol Pathol Pharmacol 1996; 92:177–189. Cook BE, Lewis GP, Fisher SK, Adler R. Apoptotic photoreceptor degeneration in experimental retinal detachment. Invest Ophthalmol Vis Sci 1995; 36:990–996. Bennett J, Zeng Y, Bajwa R, Klatt L, Li Y, Maguire AM. Adenovirus-mediated delivery of rhodopsin-promoted bcl-2 results in a delay in photoreceptor cell death in the rd/rd mouse. Gene Ther 1998; 5:1156–114. Kugler S, Straten G, Kreppel F, Isenmann S, Liston P, Bahr M. The X-linked inhibitor of apoptosis (XIAP) prevents cell death in axotomized CNS neurons in vivo. Cell Death Diff 2000; 7:815–824. McKinnon SJ, Ransom NL, Tahzib NG, Lehman DM, Hauswirth WW. Intravitreal gene therapy using adeno-associated virus (AAV) delivery of X-linked inhibitor of apoptosis protein (XIAP) protects optic nerve axons in a hypertensive rat model of glaucoma. Invest Ophthalmol Vis Sci 2001; 42:S410. Mahato RI, Smith LC, Rolland A. Pharmaceutical perspectives of non viral gene therapy. Adv Genet 1999; 41:95–156. Shenk T. Group C adenoviruses as vectors or gene therapy. In: Kapplitt MG, Loewy AD, eds. Viral Vectors. San Diego: Academic Press, 1995:43–54. Bett AJ, Haddara W, Prevec L, F.L.G. An efficient and flexible system for construction of adenovirus vectors with insertions or deletions in early regions 1 and 3. Proc Natl Acad Sci USA 1994; 91:8802–8806. Kovesdi I, Brough DE, Bruder JT, Wickham TJ. Adenoviral vectors for gene transfer. Curr Opin Biotechnol 1997; 8:583–589.
Recombinant Viral Vectors 34.
35.
36.
37.
38.
39. 40.
41.
42.
43.
44.
45.
46.
47.
48.
183
Schiedner G, Morral N, Parks RJ, Wu Y, Koopmans SC, Langston C, Graham FL, Beaudet AL, Kochanek S. Genomic DNA transfer with a high-capacity adenovirus vector results in improved in vivo gene expression and decreases toxicity. Nat Genet 1998; 18:180–183. Morsy MA, Gu M, Motzel S, Zhao J, Lin J, Su Q, Allen H, Franlin L, Partks RJ, and Granaham FL. An adenoviral vector deleted for all viral coding sequences results in enhanced safety and extended expression of a leptin transgene. Proc Nat Acad Sci USA 1998; 95:7866–7871. Parks RJ, Chen L, Anton M, et al. A helper-dependent adenovirus vector system: removal of helper virus by Cre-mediated excision of the viral packaging signal Proc Natl Acad Sci USA 1996; 93:13565–13570. Bennett J, Wilson J, Sun D, Forbes B, Maguire A. Adenovirus vector-mediated in vivo gene transfer into adult murine retina. Invest Ophthalmol Vis Sci 1994; 35:2535–2542. Li T, Adamian M, Roof DJ, Berson EL, T.P.D, Roessler BJ, Davidson BL. In vivo transfer of a reporter gene to the retina mediated by an adenoviral vector. Invest Ophthalmol Vis Sci l994; 35:2543–2549. Cayouette M, Gravel C. Adenovirus-mediated gene transfer to retinal ganglion cells. Invest Ophthalmol Vis Sci 1996; 37:2022–2028. Bergelson JM, Cunningham JA, Droguett G, Kurt-Jones EA, Krithivas A, Hong JS, Horwitz MS, Crowell RL, Finberg RW. Isolation of common receptor for coxsackie B viruses and Adenoviruses 2 and 5. Science 1997; 275:1320–1323. Wickman TJ, Mathias P, Cheresh DA, and Nemerow GR. Integrins avβ3 and avβ5 promote adenovirus internalization but not virus attachment. Cell 1993; 73:309– 319. Hoffman LM, Maguire AM, Bennett J. Cell-mediated immune response and stability of intraocular transgene expression after adenovirus-mediated delivery. Invest Ophthalmol Vis Sci 1997; 38:2224–2233. Reichel MB, Ali RR, Thrasher AJ, Hunt DM, Bhattacharya SS, Baker D. Immune responses limit adenovirally mediated gene expression in the adult mouse eye. Gene Ther 1998; 5:1038–1046. Isenmann S, Engel S, Ku¨gler S, Gravel C, Weller M, Bahr M. Intravitreal adenovirual gene transfer evokes an immune response in the retina that is directed against the heterologous lacZ transgene product but does not limit transgene expression. Brain Res 2001; 892:229–240. Yang Y, Nunes FA, Berencsi K, Furth EE, Gonczol E, Wilson JM. Cellular immunity to viral antigens limits El-deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 1994; 91:4407–4411. Yang Y, Li Q, Ertl HC, Wilson JM. Cellular and humoral immune responses to viral antigens create barriers to lung-directed gene therapy with recombinant adenoviruses. J Virol 1995; 69:2004–2015. Ali RR, Reichel MB, Byrnes AP, Stephens CJ, Thrasher AJ, Baker D, Hunt DM, Bhattacharya SS. Co-injection of adenovirus expressing CTLA4-Ig prolongs adenovirally-mediated lacZ reporter gene expression in the mouse retina. Gene Ther 1998; 5:1561–1565. Noalem G, Gdalyahu A, Shani Y, Otten U, Lazarovici P, Cohen IR, Schwartz M.
184
49. 50. 51. 52.
53.
54. 55. 56. 57.
58.
59. 60.
61.
62.
63. 64. 65.
66.
Sapieha and Di Polo Production of neurotrophins by activated T cells: implications for neuroprotective autoimmunity. J Autoimmun 2000; 15:331–345. Graham FL, Smiley J, Russell WC, Nairn R. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Virol 1977; 36:59–74. Csaky KG. Preparation of recombinant adenoviruses. In: Rakoczy PE, ed. Vision Research Protocols, 2001. Totowa, NJ: Humana Press Inc. 141–153. Ketner G, Spencer F, Tugendreich S, Connelly C, Hieter P. Proc Natl Acad Sci USA 1994; 91:6186–6190. Crouzet J, Naudin L, Orsini C, Vigne E, Ferrero L, Le Roux A, Benoit P, Latta M, Torrent C, Branellec D, Denefle P, Mayaux JF, Perricaudet M, Yeh P. Recombinational construction in Escherichia coli of infectious adenovirral genomes. Proc Natl Acad Sci USA 1997; 94:1414–1419. He TC, Zhou S, Sa Costa LT, Yu J, Kinzler KW, Vogelstein B. A simplified system for generating recombinant adenovirus. Proc Natl Acad Sci USA 1998; 95:2509–2514. Mizuguchi H, Kay MA. Efficient construction of a recombinant adenovirus vector by an improved in vitro ligation. Hum Gene Ther 1998; 9:2577–2583. Mizuguchi H, Kay MA. A simple method for constructing El- and El/E4- deleted recombinant adenoviral vectors. Hum Gene Ther 1999; 10:2013–2017. Danthinne X, Werth E. New tools for the generation of E1- and/or E3-substituted adenoviral vectors. Gene Ther 2000; 7:80–87. McLaughlin S, Collis P, Hermonat P, Muzyczka N. Adeno-associated virus general transduction vectors: analysis of proviral structures. J Virol 1988; 62:1963– 1973. Samulski RJ, Chang LS, Shenk T. Helper-free stocks of recombinant adeno-associated viruses: normal integration does not require viral gene expression. J Virol 1989; 9:3822–3828. Berns KI, Bohensky RA. Adeno-associated viruses: an update. Adv Virus Res 1987; 32:243–306. Guy J, Qi X, Muzyczka N, Hauswirth WW. Reporter expression persists 1 year after adeno-associated virus-mediated gene transfer to the optic nerve. Arch Ophthalmol 1999; 117:929–937. Dudus L, Anand V, Acland GM, Chen S-J, Wilson JM, Fisher KJ, Maguire AM, Bennett J. Persistent transgene product in retina, optic nerve and brain after intraocular injection of rAAV. Vis Res 1999; 39:2545–2553. Xiao X, Li J, Samulski RJ. Long-term and efficient in vivo gene transfer into muscle tissue of immunocompetent mice with an rAAV vector. J Virol 1996; 70: 8098–8108. Xiao X, Li J, Samulski RJ. Production of high-titer recombinant adeno-associated virus vectors in the absence of helper adenovirus. J Virol 1998; 72:2224–2232. Podsakoff G, Wong KK, Chatterjee S. Efficient gene transfer into nondividing cells by adeno-associated virus-based vectors. J Virol 1994; 68:5656–5666. Kotin RM, Siniscalco M, Samulski RJ, Zhu XD, Hunter L, Laughlin CA, McLaughlin S, Muzyczka N, Rocchi M, Berns KI. Site-specific integration by adeno-associated virus. Proc Natl Acad Sci 1990; 87:2211–2215. Flotte TR, Afione SA, Zeitlin PL. Adeno-associated virus vector gene expression
Recombinant Viral Vectors
67.
68.
69.
70.
71.
72. 73.
74.
75.
76.
77.
78.
79.
80.
81.
185
occurs in non-dividing cells in the absence of vector DNA integration. Am J Respir Cell Mol 1994; 11:517–521. Ponnazhagan S, Erikson D, Kearns WG, Zhou SZ, Nahreini P, Wang XS, Srivastava A. Lack of site-specific integration of the recombinant adeno-associated virus 2 genomes in human cells. Hum Gene Ther 1997; 8:275–284. Yan Z, Zhang Y, Duan D, Engelhardt F. Trans-splicing vectors expand the utility of adeno-associated virus for gene therapy. Proc Natl Acad Sci USA 2000; 97: 6716–6721. Flannery JG, Zolotukhin S, Vaquero MI, LaVail MM, Muzyczka N, Hauswirth WW. Efficient photoreceptor-targeted gene expression in vivo by recombinant adeno-associated virus. Proc Natl Acad Sci l997; 94:6916–6921. Cheng L, Sapieha P, Kittle´rova P, Hauswirth WW, and Di Polo A. TrkB gene transfer protects retinal ganglion cells from axotomy-induced death in vivo. J Neurosci 2002, in press. Kaplitt MG, Leone P, Samulski RJ, Xiao X, Pfaff DW, O’Malley KL, During MJ. Long-term gene expression and phenotypic correction using adeno-associated virus vectors in the mammalian brain. Nat Genet 1994; 8:148–154. Bartlett JS, Samulski RJ, McCown TJ. Selective and rapid uptake of adeno-associated virus type 2 in brain. Hum Gene Ther 1998; 9:1181–1186. Girod A, Ried M, Wobus C, Lahm H, Leike K, Kleinschmidt J, Dele´age G, Hallek M. Genetic capsid modifictaion allow efficient re-targeting of adeno-associated virus type 2. Nat Med 1999; 9:1052–1056. Bennett J, Anad V, Acland GM, Maguire AM. Cross-species comparison of in vivo reporter gene expression after recombinant adeno-associated virus-mediated retinal transduction. Met Enzymol 2000; 316:777–789. Bennett J, Duan D, Engelhardt JF, Maguire AM. Real-time, noninvasive in vivo assessment of adeno-associated virus-mediated retinal transduction. Invest Ophthalmol Vis Sci 1997; 38:2857–2863. Ali RR, Reichel MB, De Alwis M, Kanuga N, Kinnon C, Levinsky RJ, Hunt DM, Bhattacharya SS, Thrasher AJ. Adeno-associated virus gene transfer to mouse retina. Hum Gene Ther 1998; 9:81–86. Lo WD, Qu G, Sferra TJ, Clark R, Chen R, Johnson PR. Adeno-associated virusmediated gene transfer to the brain: duration and modulation of expression. Hum Gene Ther 1999; 10:201–213. Fisher KJ, Jooss K, Alston J, Yang Y, Haecker SE, High K, Pathak R, Raper SE, Wilson JM. Recombinant adeno-associated virus for muscle directed gene therapy. Nat Med 1997; 3:306–312. Snyder RO, Spratt SK, Lagarde C, Bohl D, Kaspar B, Sloan B, et al. Efficient and stable adeno-associated virus-mediated transduction in the skeletal muscle of adult immunocompetent mice. Hum Gene Ther 1997; 8:1891–1900. Ferrari FK, Samulski T, Shenk T, Samulski RJ. Second-strand synthesis is a ratelimiting step for efficient transduction by recombinant adeno-associated virus vectors. J Virol 1996; 70:3227–3234. Malik AK, Monahan PE, Allen DL, Chen B-G, Samulski RJ, Kurachi K. Kinetics of recombinant adeno-associated virus-mediated gene transfer. J Virol 2000; 74: 3555–3565.
186 82.
83.
84.
85.
86.
87.
88. 89.
90.
91.
92. 93. 94.
95.
96.
Sapieha and Di Polo Monahan PE, Samulski RJ, Tazelaar J, Xiao X, Nichols TC, Bellinger DA, Read MS, Walsh CE. Direct intramuscular injection with recombinant AAV vectors results in sustained expression in a dog model of hemophilia. Gene Ther 1998; 5:40–49. Grimm D, Kern A, Rittner K, Kleinschmidt JA. Novel tools for production and purification of recombinant adeno-associated virus vectors. Hum Gene Ther 1998; 9:2745–2760. Matsushita T, Elliger S, Elliger C, Podsakoff G, Villarreal L, Kutzman GJ, Iwaki Y, Colosi P. Adeno-associated virus vectors can be efficiently produced without helper virus. Gene Ther 1998; 5:938–945. Salvetti A, Ore´ve S, Chadeuf G, Favre D, Cherel Y, Championn-Arnaud P, DavidAmeline J, Moullier P. Factors influencing recombinant adeno-associated virus production. Hum Gene Ther 1998; 9:695–706. Grimm D, Kleinschmidt JA. Progress in adeno-associated virus type 2 vector production: promises and prospects for clinical use. Hum Gene Ther 1999; 10:2445– 2450. Hauswirth WW, Lewin AS, Zolotukhin S, Muzyczka N. Production and purification of recombinant adeno-associated virus. Meth Enzymol 2000; 316:743– 761. Summerford C, Samulski EJ. Viral receptors and vector purification: new approaches for generating clinical-grade reagents. Nat Med 1999; 5:587–588. Summerford C, Samulski RJ. Membrane-associated heparan sulfate proteoglycan is a receptor for adeno-associated virus type 2 virions. J Virol 1998; 72:1428– 1445. Conway JE, Zolotukhin S, Muzyczka N, Hayward GS, Byrne BJ. Recombinant adeno-associated virus type 2 replication and packaging is entirely supported by a herpes simplex virus type 1 amplicon expressing Rep and Cap. J Virol 1997; 71:8780–8789. Snyder RO, Miao CH, Patijn GA, Spratt SK, Danos O, Nagy D, Gown AM, Winther B, Meuse L, Cohen LK, Thompson AR, Kay MA. Persistent and therapeutic concentrations of human factor LX in mice after hepatic gene transfer of recombinant AAV vectors. Nat Genet 1997; 16:270–276. Clark KR, Voulgaropoulou F, Fraley SE, Wilson JM. Cell lines for the production of recombinant adenp-associated virus. Hum Gene Ther 1995; 6:1329–1341. Inoue N, Russell DW. Packaging cells based on inducible gene amplification for the production of adeno-associated virus vectors. J Virol l998; 72:7024–7031. Gao GP, Qu G, Faust LZ, Engdahl RK, Xiao W, Hughes JV, Zoltick PW, Wilson JM. High-titer adeno-associated viral vectors from a Rep/Cep cell line and hybrid shuttle virus. Hum Gene Ther 1998; 9:2353–2362. Liu XL, Clark KR, Johnson PR. Production of recombinant adeno-associated virus vectors using a packaging cell line and a hybrid recombinant adenovirus. Gene Ther 1999; 6:293–299. Conway JE, Rhys AP, Zolothukin CMJ, Zolothukin S, Muzyczka N, Hayward GS, Byrne BJ. High-titer recombinant adeno-associated virus production utilizing a recombinant herpes simplex virus type I vector expressing AAV-2 Rep and Cap. Gene Ther 1999; 6:986–993.
Recombinant Viral Vectors 97. 98.
99. 100. 101.
102.
103.
104.
105.
106.
107. 108.
109.
110. 111.
112.
187
Panganiban AT. Retroviral reverse transcription and DNA integration. Semin Virol 1990; 1:187–194. Coffin JM. Retroviridae: the viruses and their replication. In: Fields B, Knipe D, Howley P, eds. Fundamental Virology. Philadelphia: Lippincott-Raven, 1999: 763–843. Roe T, Reynolds TC, Yu G, Brown PO. Integration of murine leukemia virus DNA depends on mitosis. EMBO J 1993; 12:2099–2108. Lewis PF, Emerman M. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J Virol 1994; 68:510–516. Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, Verma IH, Trono D. In vivo gene delivery and stable transduction of nondividing cells by lentivirus vectors. Science 1996; 272:263–267. Kafri T, Blomer U, Peterson DA, Gage FH, Verma IM. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nat Genet 1997; 17:314–317. Miyoshi H, Smith KA, Mosier DE, Verma IM, Torbett BE. Transduction of human CD34⫹ cells that mediate long-term engraftment of NOD/SCID mice by HIV vectors. Science 1999; 283:682–686. Miyoshi H, Takahashi M, Gage FH, Verma IM. Stable and efficient gene transfer into the retina using an HIV-based lentiviral vector. Proc Natl Acad Sci USA 1997; 94:10319–10323. Lai C-C, Gouras P, Doi K, Lu F, Kjeldbye H, Goff SP, Pawliuk R, Leboulch P, Tsang SH. Tracking RPE transplants labeled with retroviral gene transfer with green fluorescent protein. Invest Ophthalmol Vis Sci 1999; 40:2141–2146. Tropepe V, Coles BLK, Chiasson BJ, Horsford DJ, Elia AJ, McInnes RR, van der Kooy D. Retinal stem cells in the adult mammalian eye. Science 2000; 287:2032– 2036. Trono D. HIV accessory proteins: leading role for the supporting cast. Cell 1995; 3:389–394. Yee J, Miyanohara A, LaPorte P, Bouic K, Burns JC, Friedman T. A general method for the generation of high-titer, pantropic retroviral vectors: highly efficient infection of primary hepatocytes. Proc Natl Acad Sci USA 1994; 91:9564– 9568. Dull T, Zufferey R, Kelly M, Mandel RJ, Nguyen M, Trono D, Naldini I. A thirdgeneration lentivirus vector with a conditional packaging system. J Virol 1998; 72:8463–8471. Miyoshi H, Blo¨mer U, Takahashi M, Gagem F H, Verma IM. Development of a self-inactivating lentivirus vector. J Virol 1998; 72:8150–8157. Zufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L, Trono D. Selfinactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 1998; 72:9873–9880. Xu K, Ma H, McCown TJ, Verma IM, Kafri T. Generation of a stable cell line producing high-titer self-inactivating lentiviral vectors. Mol Ther 2001; 3:97–104.
11 Quantification of Retinal Cells Christine A. Curcio and Kenneth R. Sloan University of Alabama at Birmingham Birmingham, Alabama, U.S.A.
I.
INTRODUCTION
A necessary milestone in establishing the safety and efficacy of new neuroprotective interventions is histological evaluation of the retina in treatment and control groups. It is likely that any treatment will be quantitative in its effect—that is, that a disease process will be slowed rather than completely halted. This means that the evaluation will have be quantitative rather than qualitative in nature and will have to be designed to detect small but potentially significant effects. The purpose of this chapter is to familiarize the reader with methods of counting cells in retinal whole mounts and sections. Examples will come from our work on photoreceptors and ganglion cells in human retina and the effects of aging and degeneration on these cell populations [1–6]. It should be emphasized that counting cell bodies is only the beginning of a complete evaluation of neural retina following treatments, because even modest loss could be accompanied by functionally significant changes at the subcellular or synaptic level. This chapter does not address the methods required for such studies.
II. WHAT ARE YOU MEASURING? TOTAL CELLS VERSUS TOPOGRAPHY The total number of cells of a given type within the retina is a single number that is easy to understand and easy to analyze using straightforward statistical 189
190
Curcio and Sloan
techniques for assessing between-group differences (e.g., t-tests or analysis of variance). The total number of cells in the retina is determined by counting cells in small samples and dividing the counts in each sample by the area of the sample to achieve density (number of cells per square millimeter of retinal surface, or planimetric density). The density is multiplied by total retinal area to achieve total number. Samples must be representative of the whole tissue, and their locations must be selected with some knowledge of the variations in cell density expected with retinal position. The topography of cell loss (in a disease) or reduction of cell loss (for a treatment) can be analyzed further to identify regionally specific effects. Our studies of aging human macula involved both determining the total number of cells and identifying changes in the topography of cell density [3,4]. Figure 1 shows that the total number of cones in the central 0.8-mm diameter of macula was remarkably constant over adulthood (31,200 ⫾ 3,100 cones). The use of total numbers enabled us to avoid the high between-individual variability that is present in the peak density of foveal cones, in the very center of the fovea. In the same eyes the total number of rods within the central 8-mm diameter decreased 30%. We used this approach to demonstrate that the loss of rods in aging human retina was selective for the parafovea, the details of which were then elucidated with maps of the differences between young and old eyes. The specifics of determining total cell number and analyzing retinal topography will be explained further below.
Figure 1 Number of cones in fovea and rods in human macula. (A) Cone density was integrated over a 0.8 mm diameter circle centered on the foveal center. Solid circles, donor eyes. Open circles, normal eyes surgically removed from patients with craniofacial tumors. (B) Rod density was integrated over a 8-mm-diameter circle centered on the foveal center in donor eyes. (From Ref. 4.)
Quantification of Retinal Cells
191
III. WHOLE MOUNTS VERSUS SECTIONS: WHICH SHOULD YOU USE? Morphological assessment typically involves histological sections prepared using conventional techniques such as paraffin embedding, cryosectioning, or epoxy embedding for electron microscopy. However, unlike most tissues, the retina can be studied as a whole mount, offering distinct advantages for quantitative studies. In a retinal whole mount, topography is preserved, with major landmarks still in place. With the exceptions noted below (Sec. VIII), the total population of a homogeneous cell type occupies layers that are only one to two cells deep, permitting its visualization within a small range of focus. Cells in individual layers can be visualized using type-specific antibodies or by distinctive fluorescence patterns achieved with nucleic acid dyes such as DAPI or ethidium homodimer (Molecular Probes). Shrinkage can be negligible in whole mounts except around cut edges. Large samples are easily attainable, so that the study of minority cell populations and the detection of small effects are possible. High-contrast images of specific cell types that are achievable in a limited focal plane should be the goal, as these images may be conducive to either automated counting or counting that can be done by relatively low-level personnel. Photoreceptors present a challenge to whole-mount studies, because their extreme compartmentalization is reflected in the composition of the retinal layers viewed en face. 1. Inner segments. In human retina, we counted photoreceptors at the level of the inner segments, the cross-sections of which form a mosaic in a single focal plane. Cone inner segments are distinguishable from rods by their threefold larger diameter and their more highly refractile properties in most parts of human retina. However, in retinal degenerations, the morphology of photoreceptors changes as outer segments are lost and the inner segment shortens and broadens. Therefore, in our study of eyes with age-related maculopathy [5], we validated counts of rods and cones made on the basis of size differences of inner segments using carbonic anhydrase histochemistry [7], which stains cone nuclei even in degenerated retina. In other species, the difference between rod and cone inner segments is not as prominent as it is in human retina (e.g., rabbit [8]) or inner segments may form tiers rather than a mosaic (e.g., pig [9]). In these species, examination of multiple focal planes is required to distinguish between rod and cone inner segments. 2. Nuclei. Cone nuclei occupy a single layer in the outer nuclear layer of most mammalian retinas, but rod nuclei form multiple rows that require many focal planes for analysis. 3. Outer segment. The first widely used photoreceptor-specific antibodies were directed against proteins associated with transduction in the outer segment.
192
Curcio and Sloan
Other investigators have used lectins that bind cone and rod sheaths. The validity of these markers is beyond the scope of this chapter. However, it is worth mentioning that outer segments are fragile and could be easily lost in the preparation of a whole mount. Therefore, with outer segment–specific markers one should verify that the vast majority of inner segments have an attached outer segment, which we did when counting short-wavelength–sensitive cones in human retina. Some cone markers like calbindin stain cones in their entirety [l0]. The major challenge to accurate counts of ganglion cells is the substantial minority population of displaced amacrine cells, which may overlap in size and morphology with ganglion cells. The proportion of ganglion cell layer neurons that are amacrine cells varies among species and within a species, with retinal region [11]. The sine qua non of a retinal ganglion cell is the presence of a projection axon to the brain. Some markers (e.g., BRN3A and BRN3B [12]) have been used in conjunction with injections of a retrograde tracer in the brain to identify ganglion cells. However, tracer injections only reach some of retinorecipient targets in the brain, and within a given target, not all ganglion cell terminals may take up and transport tracer. Therefore, the gold standard for validating retinal ganglion cell counts is comparison to optic nerve axon counts in the same eye [13]. If the total number of ganglion cells is the only desired outcome variable (and not overall topography, or counts of a particular subtypes), one should count optic nerve axons rather than ganglion cell bodies. It is possible to obtain unbiased counts of optic nerve axons using high-throughput morphometric techniques developed for neurogenetic studies by Williams and colleagues [14]. These authors determined total axon number from the total cross-sectional area of the nerve and the number of axons within unbiased counting frames on electron micrographs that were systematically placed within the nerve. Axons were counted directly on the negatives. Despite the many advantages of retinal whole mounts for quantification, sections will continue to play an important role in assessing the retina in neuroprotection studies. Stained sections currently afford greater detail than wholemounts and are amenable to multiple marker studies in single animals. Epoxyembedded tissues can be used for electron as well as light microscopy. There is a wide range of routine, special, and advanced stains and high-throughput processing technologies designed for paraffin sections, the mainstay of clinical pathology laboratories. Finally, it should be stressed that a complete census of retinal cells currently requires a combination of techniques. In a study that has provided the most complete enumeration of the neurons in the C57B1/6 mouse retina, photoreceptors were counted in wholes mounts, interneurons and Mu¨ller cells were counted in vertically oriented ultrathin sections examined by electron microscopy and optical sections examined by confocal microscopy, and ganglion cells were counted in electron micrographs of optic nerve sections [13].
Quantification of Retinal Cells
193
IV. VISUALIZING TISSUE FOR COUNTS The investigator has several choices for imaging retinal tissue for cell counts. These include making counts directly from a microscope using an ocular reticule to delimit counting fields, making counts from a video image generated by a camera mounted on a microscope, producing photographs that can be printed or scanned for off-line analysis, and capturing digitized images for off-line analysis. In our studies of photoreceptors and ganglion cell layer neurons in human retina, we viewed tissue with a computer-video-microscope system that included a 60⫻ 1.4 N.A. oil-immersion objective, Nomarski differential interference contrast optics, a digitizing tablet, and a computer-controlled stepper motor stage. We viewed the tissue on a video monitor that provided 2000–3000⫻ final magnification, depending on the objective used. The video image was combined with custom graphical overlays that were manipulated using a digitizing tablet to mark counted cells. This arrangement allowed us to optimize the focus for counting for photoreceptor inner segments and to count while focusing through the thickness of the human macular ganglion cells layer. A computer controlled the movements of the stage so that the retinal sheet could be sampled systematically in a manner that substantially eliminated the operator error inherent with manual stage control. Many of these functions, which required custom software in our studies, are now available in commercial stand-alone packages (Image Lab, Metamorph) or are bundled with image processing software available for confocal microscopy.
V.
SAMPLING CONSIDERATIONS
The past 15 years have seen significant advances in stereology, the branch of applied mathematics that seeks to deduce unbiased information about threedimensional tissue structure from two-dimensional sections. These methods eliminate bias (i.e., systematic over- or under-counts) and have practical implications for investigators in neuroprotection research. A thorough discussion of the mathematical foundations of spatial sampling is beyond the scope of this chapter. A. Sampling Windows and Exclusion Rules A counting window is an artificial boundary on the visible part of the tissue in which counts are made. In order to ensure statistical independence of samples, cells should be countable in only one window. The science of counting windows developed for sections through three-dimensional tissues (see Sec. VII) also apply to two-dimensional retinal whole mounts. For instructional purposes, it is useful to imagine a retinal whole mount that is completely covered by contiguous count-
194
Curcio and Sloan
Figure 2 Unbiased counting frames. A restricted field of view of tissue, delimited by thin dashed lines, contains profiles of cells cut in cross-section. In addition to the profiles completely within the frame, one counts all profiles partially within the frame as long as they do not touch or cross the solid exclusion lines. Thus, only the hatched cells are counted as “in.” (From Ref. 15.)
ing windows, like wallpaper. In this example, many cells fall on boundaries between adjacent windows, but they can be considered “in” only one window in order to avoid overcounts, and thus rules for determining what is “in” are required. In the unbiased counting window shown in Figure 2 [15], one counts all the profiles completely inside the frame provided that they do not in any way touch or intersect the full-drawn exclusion edges or their extensions. In our studies, cells that crossed the left and bottom borders of the counting window were considered “in,” and cells that crossed the right and top borders of the counting window were considered “out.” These rules apply regardless of whether counting windows are contiguous or widely spaced. The size of the counting window in tissue dimensions is determined by calibrating the optical and video components of the system with a calibrated micrometer slide.
B.
Systematic Random Sampling for Determining Total Cell Number
For statistically meaningful counts, the sites where counts are made must be chosen independently of their content [15]. That is, sites cannot be chosen because they have features of interest. The choice of sampling pattern across the retina is dictated by whether totals or topography is the outcome variable of interest.
Quantification of Retinal Cells
195
If the total number of cells is desired, then the method used in a recent study of cone survival in rd/rd mouse retina is appropriate [16]. These investigators randomly positioned a regular grid over the tissue, and they counted cells in windows located at 50 evenly spaced sites within the grid. In this approach, the regional specializations containing higher density of cells are sampled in proportion to their retinal area. A regional specialization such as a fovea or area centralis is small relative to the entire retina and contributes little to the total number of cells. In human retina, the small and densely packed foveal cones constitute only 0.7% of the total number of cones (32,000 vs. 4.6 million), because they occupy only a tiny area. It would therefore be possible to skip the entire fovea in a systematic sampling scheme and not appreciably change the total number of cones. However, topographic information is clearly lost by such an approach.
C. Systematic Sampling for Visualizing Topography If analysis of retinal topography is planned, then the sampling scheme must take into account the presence of regional specializations. An efficient sampling scheme is one that balances the competing goals of avoiding artifacts caused by interpolating densities between widely spaced points (undersampling) and to avoid collecting more data than necessary to achieve a density estimate with tolerable error (oversampling). Cell densities in primate retina are approximately radially symmetric around the fovea and change most rapidly near the fovea. In our topography studies we counted photoreceptors and/or ganglion cells at 100– 120 locations that were closely spaced in the fovea and less closely spaced away from the fovea. We used a spiral pattern that evenly tesselated the retinal surface (Fig. 3).
D. How Much Is Enough? It is possible to obtain smooth contour maps, reflective of low variance data, with remarkably few samples. In our studies of aging and degeneration, we counted cells within a 12-mm-diameter area centered on the fovea. We counted photoreceptors in a single 39-µm-square counting window and ganglion cell layer neurons in adjacent 39-µm-square windows until a total of 15 cells was obtained. Approximately 1385 cones (0.17% of the total within central 12 mm diameter), and 1815 ganglion cell layer neurons (0.3% of the total) were counted in these studies. We found empirically that this sample size was adequate to create a smooth map (see Sec. VI.B). In a recent study of mouse retina [16], 2.85% of the total number of counts were sampled. The validity of the sample size in this case was determined
196
Curcio and Sloan
Figure 3 Sampling points used to produce smooth maps of cell density in human retina. (A) Cells were counted at each point. Center of spiral pattern is the point of highest cone density in the fovea. (B) Central portion of A, showing a denser grid of sampling points in the fovea. (From Ref. 17.)
by showing that counting twice or three times as many cells did not change the outcome. It is possible that even fewer cells could have been counted.
VI. METHODS FOR ANALYZING TOPOGRAPHIC CHANGES A.
Retinal Model
The retina, being part of sphere, must be cut to be flattened. This is not a problem for generating counts of total numbers [16]. On the other hand, the cut edges may interfere with topographic analysis. For our topography studies, we developed a method for reconstructing the human retina by aligning the cut edges of a threepiece whole mount using major retinal vessels as landmarks [17]. Our studies of aging and degeneration used only the macula, and reconstruction of the entire retina was not necessary. In both cases (whole retina or macula), we found it convenient to use a digital model consisting of locations on the retinal sphere that were indexed by spherical coordinates and an associated cell density [17]. The spherical coordinates were referenced by the center of highest cone density in the foveal center, and the directions nasal [l80°] and temporal [0°] referred to the appropriate sides of a line passing through the foveal center and perpendicular to a line through the fovea and optic disc center of a standard left eye. This coordinate system allowed us to combine maps from eyes of different individuals. Note that
Quantification of Retinal Cells
197
our coordinate system was ideally suited for human retina, because the fovea, the point of highest cone density, and the optic disc were readily identified internal landmarks. However, a coordinate system for a species without grossly visible retinal landmarks should be done in reference to eye position in the head, a process that begins with marking directions on the eyes when they are still in the skull. Data points in our retinal model were connected into a mesh of triangular patches which closely approximated a sphere. A value at any point within a patch could be determined by calculating a weighted average of the values at its three vertices, yielding an acceptable approximation to the value found in the tissue itself at that location. Computer graphic display techniques were used to produce a variety of images. The values associated with vertices of the triangular patches were linearly interpolated across the patch. Video look-up tables were manipulated to produce false-colored smooth surfaces, terraced surfaces in false color (Fig. 6, Ref. 4) and gray scale (Fig. 4), and traditional black-and-white contour maps [18]. Plots of average cells per square millimeter as a function of eccentricity along selected meridians and maps of average density were created by resampling models of individual eyes at a set of standard locations, in which retinal directions were preserved. The total number of cells in selected regions was calculated by multiplying the mean density of each triangular patch on the model by its area on the spherical surface of the retina. B. Graphical Methods A map is the most direct way to inspect retinal topography. We displayed our data in the polar azimuthal equidistant projection, familiar from human perimetry, which preserves radial distances and distorts circumferential distances. In addition to maps of individual eyes we created maps of mean cell density in a group of eyes by resampling data from individual eyes at standard locations that reflected the weighting of the original sample points—that is, with more locations near the fovca (Fig. 5A,B) [4]. A difference map is a comprehensive and effective way of seeing overall differences in retinal topography. We created maps of the difference between groups using a measure of local variability, so that a difference in a location where variability was high would be displayed less prominently than a difference in a location where variability was low. In our study of aging human macula [4], cell density in an older age group (test population) was compared to a younger age group (reference population). We computed the difference between log (density) at each standard location in all possible pairs of test and reference eyes. The mean of the pairwise differences at each location was used to create a colorcoded map. Figure 4C shows a gray-scale version of the difference map, and a color version is shown in Figure 6G of Ref. 4. This approach was used to demonstrate the exquisite localization of age-related rod loss in the human macula.
Figure 4 Rod topography in young and elderly eyes. (A) Donor eyes, 27–37 yr. (B) Donor eyes, 82–93 yr. (C) Differences between eyes in A and B. Maps are displayed as fundus view of left eyes. Black oval denotes the optic disc. Rings of isoeccentricity are at 2 mm intervals. Gray scale in C indicates rods per square millimeter in panels A and B and the mean difference in density between older and younger eyes in panel C, as follows. As applied to panels A and B, the gray scale indicates 16 discrete bands of 12,500 rods/mm2 between 0 (black) and 200,000 (white). Gray scale applied to panel C indicates 16 bands of 0.02 log units between ⫺0.16 (black, or 31% lower) and 0.16 (white, or 44% higher). In panel C, light shades indicate locations where the older eyes have higher mean density than the younger eyes, and dark shades indicate where they have lower mean density. There is an annulus of deep rod loss in the parafovea. (From Ref. 4.)
198 Curcio and Sloan
Quantification of Retinal Cells
199
Figure 5 Rod and cone loss in exudative age-related macular degeneration. (A) Sampling pattern for photoreceptor counts in eyes with exudative age-related maculopathy (ARM). The small oval indicates the optic disc. The large irregular pattern indicates an area of almost complete photoreceptor loss overlying fibrovascular scars and atrophy of the retinal pigment epithelium (RPE). To quantify rod and cone loss, counts were made at 0.1 mm intervals along 4–8 arbitrarily placed meridians (hatched lines). Counts were also made at comparable locations in 3–4 age-matched control eyes. (B) Rods but not cones are lost around the margins of fibrovascular scar and RPE atrophy in an eye with exudative ARM. (Dark circles, rods; gray circles, cones.) (From Ref. 5.)
Figure 6 Disector. A block of retinal tissue (A) containing particles a–e is cut into serial sections (B). Hatched profiles in B–D are cells that are transected by the sectioning plane. A pair of sections, distance t apart, is drawn from the stack. The top (reference) plane contains an unbiased counting frame of area A (C), and the lower (look-up) section does not (D). The volume of the disector is A ⫻ t. Only cell b is counted as “in,” because it falls within the counting frame in the reference section (C), and it is not present in the look-up section (D). This diagram shows a physical disector (i.e., identifying cells in separate sections) An optical disector involves counting cells identified in the top and bottom focal planes of the same thick section. (From Ref. 23.)
200
C.
Curcio and Sloan
Specialized Methods for Regional Photoreceptor Loss
Figure 5 shows a different systematic sampling strategy employed to assess the relative rate of cone and rod degeneration surrounding areas of nearly complete photoreceptor loss vitread to fibrovascular scars and RPE atrophy in eyes with late exudative age-related maculopathy [5]. Here we counted photoreceptors at 0.1 mm intervals along arbitrarily chosen tracks that crossed the boundary between intact and degenerated photoreceptor mosaic. Photoreceptor degeneration was expressed as loss relative to controls as a function of distance from the margin of the intact photoreceptor mosaic. Counts were compared to those in matched locations in control eyes using the same measure described in the previous section for maps. In the figure, the dashed line indicates the variability in cell density among control eyes, calculated as follows. At each retinal location, we computed the differences between log(density) for each pairwise comparison of one randomly chosen control eye and the other controls. This computation, done for both cones and rods, established 95% confidence intervals for differences among the controls. Results were similar regardless of which control was chosen. Then we computed the differences in log(density) for pairwise comparisons of the ARM eye and the controls. Directional differences between ARM eyes and controls that fell below the lower confidence limit for controls were considered significant loss. We then reported the percentage of counting sites with significant loss and the percentage of sites with loss where either rod or cone loss predominated.
VII. COUNTS IN SECTIONS It is possible to count cells in sections through the full thickness of the retina, (i.e., in the plane orthogonal to a retinal whole mount). Two different approaches have been employed for counting retinal cells in sections, depending the layer of interest. In both cases, one must carefully attend to the location of samples in order to ensure that counts are taken from comparable locations (see Sec. VI A). In the mammalian retina, photoreceptor nuclei form a thick layer in the outer nuclear layer (ONL), typically with a single row of cones external to multiple rows of rods. Many studies of retinal degeneration and treatments in animal models have quantified photoreceptors as either rows of ONL nuclei or thickness of ONL layer [19]. Use of ONL thickness compares favorably with counts of photoreceptor nuclei and can be done quickly [20]. This approach is suitable for quantifying the more numerous rods but is less suitable for quantifying the less numerous cones, because the sample of cones is very small in single sections [21]. Note that counts of cells along the length of a retinal section (typically expressed as
Quantification of Retinal Cells
201
cells/100 µm) assumes that the counted particles (usually nuclei) do not differ in size between treatment and control groups, which may not be true for degenerating photoreceptors. Counts along single slices also assume the overall size of the retina does not differ between groups. It is possible to section the retina serially and determine a total number of cells using a subsample of sections in order to obtain a sample size comparable to that achievable in a whole-mount. Such an approach is feasible for very small eyes, such as embryonic tissues. The current standard for cell counting in sections is the disector, an unbiased counting method that uses counts obtained in parallel planes separated by a known distance in the tissue (Fig. 6). The rule is to count only those particles that appear within an unbiased counting frame in one plane (the reference plane) but not in the matched plane (the look-up plane). The number of nuclei (conventionally called Q⫺ ) is contained within the volume of the disector. The volume of the disector is equal to the area of the counting frame, multiplied by the distance between planes. If the specimen has been sampled systematically by multiple disectors, the number per volume is calculated by summing the area and Q⫺ over all disectors. Reference and look-up planes are separated by a distance that is one-quarter to one-third that of the particles being counted. There should be only one counted particle per cell, and the fewest assumptions are required for smooth convex objects, so cell nuclei are the typical counting particle. In order to count nuclei that are 8 µm in diameter, planes that are 2–3 µm apart are required. An efficient way to implement a disector is to track through successive focal planes in a thick slice of tissue, a method known as the optical disector. It is possible to use a high numerical aperture objective to identify the reference and look-up planes a relatively thick section (ⱖ25 µm). Confocal microscopy can also be used to optically section tissue for counting of fluorescent cells and will likely be used more frequently for this purpose as these instruments become more widely available. The alternative to an optical disector is the physical disector, which involves either semi-thin or ultrathin sections prepared for transmission conventional electron microscopy. In order to identify the same cells in the reference and look-up planes, the sections are aligned by reference to local fiducial landmarks such as blood vessels. A physical disector is difficult to apply to the ONL of the retina. The rod nuclei are morphologically homogeneous and form a paracrystalline array like ball bearings in a box, and there are no nearby blood vessels. Thus, the reference and look-up planes can be aligned spuriously, introducing inaccuracies into the number of cells missing from the look-up plane. However, in the inner nuclear and ganglion cell layers, the presence of capillaries and multiple, distinctive cell populations together facilitate the chore of aligning sections. Disector methods have been successfully used for determining cell densities in these layers [13], and they are the method of choice in contemporary morphometric studies of brain [22].
202
Curcio and Sloan
VIII. CHALLENGES OF THE MACULA For either whole-mounts or sections, the primate macula presents unique challenges for cell counting, due to its extreme thickness and the presence of the fovea. In the macula of human and monkey, ganglion cells are up to six cells deep. In whole mounts, conventional Nissl stains that involve dehydration of tissue reduce thickness by factors up to two-thirds [2]. We used unstained tissue that was cleared with dimethyl sulfoxide rather than ethanol and xylene and viewed with differential interference contrast optics [2,3,6]. We continuously focused through the tissue while counting nucleoli of ganglion cell layer neurons. Studies using fluorescent markers for inner retinal cells would require multiple image planes. On the photoreceptor side of the fovea, the external fovea is an inward dip, where photoreceptor inner segments do not form a single plane. Therefore, macular photoreceptors, like the ganglion cells, also require continuous adjustment of focus to find the optimal plane for counting. Another challenge to cell counting in the macula is the sharp changes in cell density near the foveal center. The cone distribution has a very small (100 µm or less) area of very high density in the foveal center that drops by 90% within 1 mm. Therefore, counts in this area require precisely specified sample locations. Poor control of location can insert noise into cell counts and make effects difficult to detect. In our whole-mount studies, we set the point of highest cone density as the foveal center and used a computer-controlled stepper-stage on the microscope, so that position relative to the foveal center could be specified accurately. In studies requiring sections, we sectioned serially into the fovea so that the center could be identified by the absence of ganglion cell and inner nuclear layers, and the absence of Henle fibers cut in cross-section. Other challenges to studies using the macula include long incubation times required for complete penetration of the retina by immunoreagents, the absence of certain well-established markers in the foveal region [l0], and the predilection for postmortem swelling in inner retina that can impede the identification of the foveal center. ACKNOWLEDGMENTS Supported by NIH grant EY06109 and unrestricted funds from Research to Prevent Blindness, Inc., to the Department of Ophthalmology, University of Alabama School of Medicine. REFERENCES 1.
Curcio CA, Sloan KR, Kalina RE, Hendrickson AE. Human photoreceptor topography. J Comp Neurol 1990; 292:497–523.
Quantification of Retinal Cells 2. 3. 4.
5. 6. 7.
8. 9. 10.
11. 12.
13. 14.
15.
16.
17. 18.
19.
203
Curcio CA, Allen KA. Topography of ganglion cells in human retina. J Comp Neurol 1990; 300:5–25. Curcio CA, Drucker DN. Retinal ganglion cells in Alzheimer’s disease and aging. Ann Neurol 1993; 33:248–257. Curcio CA, Millican CL, Allen KA, Kalina RE. Aging of the human photoreceptor mosaic: evidence for selective vulnerability of rods in central retina. Invest Ophthalmol Vis Sci 1993; 34:3278–3296. Curcio CA, Medeiros NE, Millican CL. Photoreceptor loss in age-related macular degeneration. Invest Ophthalmol Vis Sci 1996; 37:1236–1249. Medeiros NE, Curcio CA. Preservation of ganglion cell layer neurons in age-related macular degeneration. Invest Ophthalmol Vis Sci 2001; 42:795–803. Nork TM, McCormick SA, Chao G-M, Odom JV. Distribution of carbonic anhydrase among human photoreceptors. Invest Ophthalmol Vis Sci 1990; 31:1451– 458. Young HM, Vaney DI. Rod-signal interneurons in the rabbit retina: 1. Rod bipolar cells. J Comp Neurol 1991; 310:139–153. Miller WH, Snyder AW. The tiered vertebrate retina. Vision Res 1977; 17:239– 255. Haley TL, Pochet R, Baizer L, Burton MD, Parmentier M, Crabb JW, Polans AS. Calbindin D-28K immunoreactivity of human cone cells varies with retinal position. Vis Neurosci 1995; 12:301–307. Kolb H. Amacrine cells of the mammalian retina: neurocircuitry and functional roles. Eye 1997; 11:904–923. Xiang M, Zhou L, Macke JP, Yoshioka T, Hendry SH, Eddy RL, Shows TB. The Brn-3 family of POU-domain factors: primary structure, binding specificity, and expression in subsets of retinal ganglion cells and somatosensory neurons. J Neurosci 1995; 15:4762–4785. Jeon C-J, Strettoi E, Masland RH. The major cell populations of the mouse retina. J Neurosci 1998; 18:8936–8946. Williams RW, Strom RC, Rice DS, Goldowitz D. Genetic and environmental control of variation in retinal ganglion cell number in mice. J Neurosci 1996; 16:7193– 7205. Gundersen HJG, Bendtsen TF, Korbo L, Marcussen N, Moller A, Nielsen K, Nyengaard JR. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. Acta Pathol Microbiol Immunol Scand 1988; 96:857–881. Mohand-Said S, Deudon-Combe A, Hicks D, Simonutti M, Forster V, Fintz AC, Leveillard T. Normal retina releases a diffusible factor stimulating cone survival in the retinal degeneration mouse. Proc Natl Acad Sci USA 1998; 95:8357–8362. Curcio CA, Sloan KR, Meyer D. Computer methods for sampling, reconstruction display and analysis. Vision Res 1989; 19:529–540. Curcio CA, Sloan KR, Jr., Packer O, Hendrickson AE, Kalina RE. Distribution of cones in human and monkey retina: individual variability and radial asymmetry. Science 1987; 236:579–582. LaVail MM, Yasumura D, Matthes MT, Drenser KA, Flannery JG, Lewin AS, Hauswirth WW. Ribozyme rescue of photoreceptor cells in P23H transgenic rats:
204
20.
21.
22.
23.
Curcio and Sloan long-term survival and late-stage therapy. Proc Natl Acad Sci USA 2000; 97: 11488–11493. Michon JJ, Li ZL, Shioura N, Anderson RJ, Tso MM. A comparative study of methods of photoreceptor morphometry. Invest Ophthalmol Vis Sci 1991; 32:280– 284. LaVail MM, Matthes MT, Yasamura D, Steinberg RH. Variability in rate of cone degeneration in the retinal degeneration (rd/rd) mouse. Exp Eye Res 1997; 65:45– 50. West MJ, Slomianka L, Gunderson HJG. Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator. Anat Rec 1991; 231:482–497. Mayhew TM, Gundersen HJG. ‘If you assume, you can make an ass out of u and me’: a decade of the disector for stereological counting of particles in 3D space. J Anat 1996; 188:1–15.
12 Ex Vivo and Whole-Mount Retinal Preparations Arthur J. Weber Michigan State University East Lansing, Michigan, U.S.A.
I.
INTRODUCTION
Despite its relatively small size, the brain is an enormously complex organ that poses many challenges to those trying to understand its organization, function, and dysfunction. Fortunately, however, the brain comprises a number of more simple and well-defined systems, and many of these have proved to be good models for studying complex brain mechanisms. The visual system is one area of the brain that has received considerable attention over the years, in part because its functional integrity can be assessed easily using light stimulation, but also because it is naturally divided between the eye and the rest of the brain. In addition, several primary components of the central visual pathway, including the eye, optic nerve, and visual cortex, are readily accessible for experimental manipulation and/or therapeutic intervention.
II. THE RETINA AS A MODEL FOR STUDYING NEUROPROTECTION Over the past several years considerable research has focused on the structure and function of the normal retina [1–5] and following injury to either the optic 205
206
Weber
nerve [6–13] or visual cortex [14–19]. Because it is a relatively isolated organ, and because the anatomy, physiology, and pharmacology of many of its neurons are well defined, the retina provides a good model for studying degeneration and neuroprotection within the central nervous system. In addition, the retina can be studied in vivo, or removed from the eye and examined for an extended period of time as an isolated tissue preparation [20–25]. Because removal from the eye does not disrupt its cellular organization, neurons within the isolated retina retain their normal spatial and connective relations (synaptic and gap junctional), and their responses to light stimulation. This is in contrast to other isolated brain preparations where tissue acquisition commonly results in a significant disruption of afferent and efferent connections, neurons at the tissue margins are damaged during isolation, and the functional integrity of the isolated preparation can be assessed only by means of electrical stimulation. The maintenance of normal cellular relations within the isolated retina also is important with respect to neuroprotection, where survival mechanisms need not involve only direct interactions between the drug and the injured neurons. The cellular organization of the vertebrate retina, which is conserved across most species, also makes it an ideal tissue for use in degeneration and neuroprotection studies. Although the retina contains several million neurons, these are arranged into three distinct layers of nerve cells, separated by two layers of synaptic connections. And while many subclasses of retinal neurons have been described, it is generally agreed that most retinas contain only five major classes of neurons. From outside to inside, these include the photoreceptors, horizontal, bipolar, and amacrine cells, and the ganglion cells [1]. Of these different cell types, perhaps the best studied are the ganglion cells. These neurons represent the final stage of visual processing within the retina and their axons form the optic nerve, a primary site of injury in many optic neuropathies. In addition, ganglion cells and their axons give rise to the different functional streams that compose the central visual pathway [2,5] Finally, because they are relatively large and form a distinct layer near the inner retinal surface, ganglion cells are more accessible, both in vivo and in the isolated retina than most other neurons [1].
III. GANGLION CELL CLASSIFICATIONS AND METHODS OF STUDYING To date, at least thirty morphologically distinct types of ganglion cells have been reported in the vertebrate retina [26–28]. Despite this relatively large number, however, our current understanding of ganglion cell structure and function is based almost exclusively on a select group of neurons from the cat and primate retinae. In the cat, the three major anatomical classes of ganglion cells are the
Ex Vivo and Whole-Mount Retinal Preparations
207
alpha, beta, and gamma cells, and there is direct intracellular evidence that these correspond with the Y-, X- and W-cells described physiologically [2,29–31]. Of these different classes of ganglion cells, the alpha and beta cells have been described most completely. In brief, alpha cells represent about 5% of the ganglion cells in the cat retina. They have the largest cell bodies (25–35 µm diameters), and large, radially oriented, dendritic arbors that commonly originate from five or six primary dendrites and display a regular pattern of branching. Beta-cells comprise about 55% of the ganglion cells in the cat retina. They have mediumsized somata (18–20 µm diameters), and their small to medium-sized dendritic trees often originate from a single primary dendrite that then gives rise to a compact, bushy dendritic arbor. Functionally, Y-cells have high contrast gain, large receptive fields, fast conducting axons, and are considered to be involved primarily with object detection. By contrast, X-cells have small receptive fields, are of highest density in central retina, and are considered to be involved primarily with the analysis of fine detail. The primate retina also contains three major classes of ganglion cells. Anatomically, these are the parasol, midget, and small-field bistratified cells [4,5,20,22,25,32–36]. Parasol cells represent approximately 10% of the ganglion cells in the primate retina. At all retinal eccentricities, the somata and dendritic fields of these neurons are among the largest in the ganglion cell layer. Similar to the alpha cells of the cat retina, the dendritic arbors of primate parasol cells commonly originate from three or four large primary processes that branch regularly and form a radially symmetric arbor. Functionally, parasol cells have large receptive fields with rapidly conducting axons, and they respond best to achromatic stimuli of high temporal and low spatial frequency [3,5,37–40]. Combined with their relatively uniform distribution acrosss the retina, these neurons also are considered to be involved primarily with object detection. Midget ganglion cells represent about 80% of the ganglion cells in the primate retina. They have medium-sized somata and their small to medium-sized dendritic trees often originate from a single dendrite that then gives rise to a compact, bushy dendritic arbor. Midget ganglion cells have small receptive fields and respond best to chromatic stimuli (primarily red–green) of high spatial and low temporal frequency [3,5,37-40]. Based on these characteristics, and their high density and one-toone synaptic arrangements with midget bipolar cells within the fovea, these neurons are considered to subserve fine spatial discrimination. Small-field bistratified cells represent about 5–8% of the ganglion cells in the primate retina. These neurons have somata and dendritic fields that are similar in size to those of surrounding parasol cells, and they respond best to short-wavelength (blue) stimuli [4,5,36]. For more than 100 years, anatomists have been interested in the structure and classification of neurons in different regions of the brain, including the retina. Some of the earliest anatomical work in the retina was conducted by Tartuferi
208
Weber
and Ramo´n y Cajal using modifications of the osmium-dichromate-silver staining method of Golgi, and by Dogiel, using the methylene blue staining routine of Ehrlich [41]. More recent studies aimed at defining the different morphological classes of ganglion cells in the vertebrate retina also have relied on these early techniques. In addition, neurofibrillar stains have been used to study the morphologies of ganglion cells in the normal retina [42], and following damage to the optic nerve [8]. A number of different retrograde tracing methods also have been used to examine ganglion cell morphology. Some of these have involved direct application of various dyes and neuronal tracers to either the intact retina or the severed optic nerve [43–46]; others have used micro-injections of tracers into specific retinal target structures in order to define the central projection patterns of specific ganglion cell types [47–49]. The major advantage of these staining techniques over the more traditional Nissl-staining methods [50] is that they provide detailed information about not only the cell soma but also the dendritic arbor. This enhances the accuracy of ganglion cell classification and provides a finer level of detail for determining injury- and/or treatment-related changes in ganglion cell morphology. Surprisingly, to date few of the techniques described above have been applied to studies of retinal ganglion cell degeneration or neuroprotection. Most likely this is because of their capriciousness. Although silver staining methods reveal exquisite morphological detail, they often stain only select subsets of neurons. In addition, the pattern of staining can be highly variable across retinas, or even within the same retina. Various approaches using retrograde tracers also have been hampered by the fact that they often label only small populations of ganglion cells, the labeling is inconsistent and nonuniform across and within retinas, and the tracer often fails to label fine distal dendrites. Furthermore, in those cases where the insult occurs at the level of the optic nerve, the potential for labeling ganglion cells from a central target are greatly diminished or eliminated. One approach that has been used with relatively good success to study retinal degeneration and neuroprotection in the rat involves placing a piece of gelatin sponge soaked with either Fluoro-Gold (Fluorochrome, Englewood, CO) or the carbocyanine dye DiI (Molecular Probes, Eugene, OR) over the superior colliculus and visual thalamus in vivo [11,51]. Because both of these retrograde tracers are capable of labeling ganglion cells long term, it is possible to prelabel ganglion cells, induce the retinal injury, and still provide an adequate survival period for assessing the treatment strategy. One limitation to this approach, however, is that it is most practical only in small vertebrates, such as the rat, where the central target nuclei can be easily accessed and covered with the tracer-soaked sponge. In addition, the tracers label primarily the cell soma, and the fate of tracer released from degenerating ganglion cells is always a concern. Finally, recent evidence indicates that, following long-term survival periods (⬎6 wk), Fluoro-
Ex Vivo and Whole-Mount Retinal Preparations
209
Gold migrates not only from ganglion cells but from the retina in general (Bal Chauhan, personal communication).
IV. THE ISOLATED RETINA PREPARATION In our analyses of retinal ganglion cells from cats with neonatal visual cortex damage [19] and monkeys with experimental glaucoma [25], we sought to avoid the limitations of the silver impregnation and retrograde labeling techniques by combining an isolated retina preparation with intracellular staining techniques. This approach provided several advantages. First, it allowed us to visualize and target single neurons. This was especially important in the cats with neonatal visual cortex damage, where there was a significant reduction in the number of surviving ganglion cells. Second, using the isolated retina preparation we were able to examine ganglion cells from matched areas of each retina, thereby reducing sample variability. And finally, like the silver staining methods, the intracellular approach allowed us to label and compare qualitatively and quantitatively morphological features associated not only with the cell soma but also the dendritic tree and intraretinal segment of the axon. Similar procedures are used to isolate and maintain our cat and primate retinas. These are outlined diagrammatically in Figure 1 and described in more detail below. Following the desired survival period, each animal receives an overdose of pentobarbital sodium. The eyes then are removed and placed into the same solution that will be used to sustain them during the course of the experiment. We have used both Ames media [20–22,46] and the artificial cerebral spinal fluid (aCSF) described by Saito [29] with equal success. If the eyes must be transported, we have found it best to either leave them intact, or design a small chamber for providing oxygenation during transport. The anterior segment of each eye (from the ora serrata forward) is removed using a scalpel blade and pair of small scissors, and the resulting posterior eyecup is placed into a beaker of aCSF (pH 7.4). The beaker contains a stainless steel basket that holds the tissue off the bottom, and the aCSF is oxygenated with 95% O2 and 5% CO2 at room temperature using a glass gas dispersion tube with a fritted ending. Separation of the anterior and posterior segments is performed in a petri dish over gauze that has been moistened lightly with aCSF to prevent the tissue from sticking. Good separation, with complete removal of the vitreous, seems to work best if the initial incision is made 2–3 mm posterior to the ora serrata, and the two segments then are folded back away from each other using two pair of forceps with serrated tips. The goal is to have the vitreous remain firmly attached to the anterior segment and removed from the posterior segment by peeling it away from the retinal surface; pulling the vitreous perpendicular to the surface can
210
Weber
Figure 1 Schematic diagram outlining the steps used in preparing the isolated, living retina preparation. The posterior eyecup is stored and dissected in oxygenated artificial cerebrospinal fluid (aCSF) prior to being placed, ganglion cell layer up, into the recording/injection chamber. This chamber, which also receives oxygenated aCSF, fits onto the stage of an upright microscope. Single neurons in the isolated retina are viewed and targeted for intracellular analysis and injection by prestaining with the vital dye Acridine Orange. A high-resolution video monitor is used to provide visual stimulation of the retina, and an intracellular amplfier is used to obtain biophysical measurements.
result in mechanical damage to the retina. If vitreous remains attached to the posterior eyecup, it often can be removed prior to placing the retina into the recording and injection chamber using the reverse side of a pair of curved forceps and grasping the vitreous in an area of the retina that will not be examined. Treating the retina briefly with a low concentration (0.1%) of collagenase (Type II, C-6885, Sigma Chemical, St. Louis, MO) also has been used [20]. Again, it is important to try to peel the vitreous from the retina by detaching it from the margin of the eyecup and drawing it horizontally across the surface of the retina. A clean retinal surface is imperative for good oxygenation of the tissue, clean microelectrode penetrations and recordings of ganglion cells, and successful postinjection processing. Once a posterior eyecup with a clean retinal surface is obtained, one can either separate the retina from the choroid and sclera using a fine (# 00) artist’s
Ex Vivo and Whole-Mount Retinal Preparations
211
brush, remove only the sclera using forceps and scissors, or leave the entire eyecup intact. Complete isolation of the retina is suitable for intracellular staining studies, but not those involving electrophysiological recordings, where dissection-related damage to the photoreceptor and pigment epithelia layers could affect light responsiveness. For these studies it is best to either remove only the sclera or leave the eyecup intact. Final preparation of the retina for placement into the injection/recording chamber is performed in a petri dish containing aCSF. The eyecup is placed on a submersed, perforated, stainless steel platform, and a bubble stone is used to maintain oxygenation. Regardless of which type of isolated retina preparation will be used, three to four radial cuts, made away from the area of interest, are used to flatten the tissue; the optic disc, fovea, and retinal blood vessels serve as references for positioning these incisions. The retina then is placed, ganglion cell layer up, into the injection/recording chamber, which has been filled with oxygenated aCSF. While many different types of chambers are available commercially, we prefer one of our own design. The chamber consists of a rectangular piece of Plexiglas that has a circular well in the center. A pair of concentric Plexiglas rings fit into the well. The outer ring contains strands of nylon stocking that form a platform and hold the tissue above the chamber bottom. The inner ring also contains strands of nylon stocking, and when press fit into the outer ring, forms a “sandwich” that holds the tissue securely in place. The chamber then is mounted onto the stage of an upright microscope (Nikon Optiphot-2) equipped with epifluorescence. The tissue is perfused with warm (36°C), oxygenated aCSF using a gravity flow system. An aspirator bottle containing the aCSF is placed approximately 24 in. above the tissue preparation, and the solution is oxygenated with 95% O2 /5% CO2 using a glass gas dispersion tube. The aCSF then passes through a flow meter and water jacket where the flow rate is adjusted to 4–6 cc/min and the solution is warmed just prior to delivery into the tissue chamber. The temperature of the water jacket, which consists of a series of stainless steel tubes traversing a water-filled compartment, is adjusted using a temperature-controlled circulating pump (Isotemp 2100, Fisher Scientific). The aCSF is warmed by making five passes through the Plexiglas water jacket before being dispensed to the tissue chamber. The solution flows over the tissue, and then is drawn off from a separate well using a vacuum pump and sidearm flask. Not drawing the aCSF directly from the injection/recording chamber prevents fluctuations in the chamber fluid level. The stainless steel needle used to draw off the fluid is moveable, thereby allowing one to adjust the depth of the aCSF in the chamber. This is of particular importance for the physiological recordings, where fluid depth affects the capacitance of the electrode, and therefore the quality of the electrical signals recorded. In all cases, the retinal dissections are performed in dim light, and the retinas are allowed 30 min in compete darkness in the injection/recording chamber before being studied. Single ganglion cells are viewed under epifluorescence us-
212
Weber
ing a 40⫻ water immersion objective (Nikon, N.A. 0.55) with a working distance of about 1.6 mm. In order to aid visualization, the retinas are treated periodically with two to three drops of the vital dye Acridine Orange (A-4912, Sigma, 1mM in aCSF, see Fig. 1). This often is not necessary when working with a completely isolated cat retina, where individual ganglion cells can be viewed under standard illumination using either differential contrast optics, or by simply misadjusting the microscope condenser. However, because of the increased number of ganglion cell axons, it is difficult to view individual cells in the primate retina without Acridine prestaining. When ganglion cells are viewed using epifluorescence, a neutral density filter (ND4) is used to reduce the intensity of the mercury vapor light reaching the tissue, and cells are viewed only long enough to either align the electrode with the selected cell (physiology studies), or quickly penetrate the cell (anatomical studies).
V.
ELECTRODE PREPARATION AND INTRACELLULAR STAINING
Intracellular injections and/or recordings are made using glass microelectrodes and a four-axis hydraulic micromanipulator (Narishighe, Japan). The glass micropipettes are pulled on a Flaming-Brown P-87 horizontal micropipette puller (Sutter Instruments, Novato, CA). For anatomical studies, we have found that a 3% solution of the fluorescent dye Lucifer Yellow CH (Sigma, L-0259) in 0.1 M LiCl (pH 7.6) works well for labeling both cat and primate ganglion cells. For combined electrophysiological and anatomical studies, we fill the pipettes with a solution containing 2% Neurobiotin (Vector Labs, Burlingame, CA) and 0.05% pyranine (Aldrich, Milwaukee, WI) in 1 M potassium acetate buffer (pH 7.6); Lucifer Yellow CH precipitates in 1 M potassium acetate, and therefore we do not use it with this buffer. However, it may remain soluble in ⬍ 200 mM acetate buffer. Because both fluorescent dyes have excitation and emission wavelengths similar to Acridine Orange, it is possible to view the electrode tip and the prestained ganglion cells using the same epifluorescence filter combination (B3A/DM505, Nikon). This makes it easier to target single ganglion cells and results in a more stable preparation, because filter cubes do not need to be exchanged during cell penetrations. The electrodes are beveled from an initial resistance of 80–100 MΩ to a final resistance of about 35–45 MΩ using a K.T. Brown type beveler (BV-10, Sutter Instruments). For intracellular injection with Lucifer Yellow CH, the electrode is positioned adjacent to the cell soma and advanced slowly along its axis until it is seen to penetrate the cell soma. Successful penetration of the cell membrane is recognized by the sudden filling of the soma by the small amount of dye that leaks from the electrode tip. Complete filling of the soma, dendritic tree, and intraretinal axon then is achieved by passing 1–5 nA of
Ex Vivo and Whole-Mount Retinal Preparations
213
negative current through the electrode. Typically-2 min is required to fill parasol, midget, and beta ganglion cells, whereas 2–3 min is needed for the larger alpha cells. Intracellular penetration and recording using the potassium acetate-Neurobiotin electrodes is accomplished using a similar strategy; however, here the negative capacitance circuitry of the intracellular amplifier (AxoClamp 2B, Axon Instruments, Union City, CA) is used to gently “ring” the electrode tip and aid with penetration of the cell membrane. Successful penetration is indicated by a negative deflection of about 50–60 mV in the recorded electrode potential, and large intracellular spikes. Injection of Neurobiotin is achieved using positive current pulses (1–2 nA) of 100–200 ms duration, delivered for 3–5 min. Regardless of the injection procedure, it is imperative that optimal electrode configuration and injection parameters be established for the intracellular dyes and buffers being used, and the cells being studied. Nonoptimal conditions can result in morphological artifacts that could be misinterpreted as degenerative changes. In addition, because it is not uncommon for a bond to develop between the cell membrane and electrode, quick withdrawl of the electrode is needed in order to prevent the cell soma from being damaged mechanically or excised from the tissue. Upon completion of the injections and/or recordings, the retina is removed from the chamber and drop-fixed in a solution containing 4% paraformaldehyde in 0.1 M phosphate buffer. Retinas containing Lucifer Yellow CH-filled ganglion cells then are washed with buffer, mounted onto subbed slides, and coverslipped using DePeX (BDH Laboratory, Poole, England), a non-autofluorescing mounting medium. Neurobiotin-labeled retinas are washed with 0.01 M phosphate buffered saline, dissected from the sclera and pigment epithelium using a fine artist’s brush (# 00), and processed using Vector Lab’s Vectastain ABC Elite (PK-6100) and DAB substrate (SK-4100) kits. These retinas also are then mounted and coverslipped using DePeX.
VI. RETINAL SAMPLING, MAPPING, AND ANALYSIS Although the intracellular technique provides detailed filling of single ganglion cells, it does not permit large numbers of neurons to be sampled across the entire retina. Therefore, one typically needs to select a specific area of the retina for study. This also is necessary because ganglion cell size is not constant, but increases with increased retinal eccentricity. Thus, for accurate morphological comparisons to be made, it is necessary to match ganglion cell samples for retinal location. Fortunately, the ability to visualize specific retinal landmarks, such as the optic disc, fovea, area centralis, and retinal blood vessel pattern, in the isolated retina make it possible to coordinate cell sampling across retinae.
214
Weber
In order that qualitative and quantitative comparisons be made for retinal ganglion cells of like retinal eccentricity, the position of each injected cell relative to the location of the optic disc and either fovea (primate) or area centralis (cat) is mapped in the fixed and mounted retina using a microscope stage digitizing system (AccuStage, Shoreview, MN). Because this system employs a set of mechanical stage encoders, accurate reconstruction of the retinal map and cell locations is not influenced by changes in microscope objective magnification. In the case of the fluorescent-labeled neurons, this map also serves as a guide during the capture of individual cell images using the confocal microscope. The application of confocal microscopy greatly enhances one’s ability to acquire and analyze ganglion cell morphological data efficiently; complete high-resolution cell reconstructions can be achieved in minutes, instead of hours, as needed using more conventional techniques. In addition, because the images are captured as a set of digital optical slices, it is possible not only to view each cell in three-dimensions from any perspective but also to compress the optical slices into a single image that includes the soma, dendritic tree, and intraretinal axon (Fig. 2). This feature is very useful given the highly three-dimensional structure of ganglion
Figure 2 Samples of parasol and midget ganglion cells from glaucomatous eyes that were injected intracellularly with the fluorescent dye Lucifer Yellow CH and captured using confocal microscopy. Note that the earliest changes are associated with the dendritic processes. (Adapted from Ref. 25.)
Ex Vivo and Whole-Mount Retinal Preparations
215
cells, and one that is not readily available using standard video image capture and analysis systems. Aside from the ease of capture and the ability to freely manipulate cell images, a third advantage of being able to apply confocal microscopy is that most quantitative measurements can be made directly from the digital images using the system’s own morphological analysis software. This enhances the efficiency of the analysis and reduces the potential for magnification-based errors. However, since most confocal image formats now are compatible with current image analysis packages, cell images can be downloaded and analyzed offline to reduce cost and increase flexibility. For our purposes, we combined intracellular staining and both video-enhanced and confocal microscopy to compare ganglion cell morphologies following either neonatal damage to visual cortex in the cat [19] or experimental glaucoma in primates [25]. Qualitative and quantitative ganglion cell comparisons included the assessment of differences in dendritic field organization, cell body size, dendritic field area, and axon diameter for ganglion cells of specific classes. In the cat studies, we found that a selective loss of beta ganglion cells due to neonatal damage to visual cortex affects only the development of surviving beta, and not alpha, ganglion cells. This suggests that the dendritic fields of ganglion cells in the vertebrate retina achieve their final morphologies based on competitive interactions among ganglion cells of the same, and not different, classes. With respect to the primate studies, application of the intracellular staining technique allowed us to demonstrate for the first time that the earliest signs of ganglion cell degeneration in glaucoma occur at the level of the distal dendritic tree, and that the pattern of degeneration is similar for both midget and parasol type cells. In brief, the degenerative changes include a thinning of the proximal and distal dendrites, abrupt reductions in dendritic process diameter at branch points, and a general decrease in the complexity of the dendritic tree (Fig. 2). These findings have at least three significant implications. First, they indicate that the onset of glaucoma-related ganglion cell death occurs earlier than previously thought based on Nissl-stained estimates of ganglion cell loss alone. Second, because ganglion cells receive all of their input from more distal retinal elements via their dendrites, it is reasonable to assume that early deficits in ganglion cell function also occur; this is the focus of our structurefunction studies. And finally, since changes at the level of the cell soma occur later, they suggest a window of opportunity for effective neuroprotection intervention. Because our structure-function studies employ the use of a nonfluorescent dye, the ganglion cells for this work are analyzed using the Neurolucida/ NeuroExplorer cell reconstruction and analysis system (MicroBrightField, Inc., Colchester, VT). This system incorporates a motorized stage and video-based (Hamamatsu C-5985 chilled CCD camera) image analysis software. Individual cells that have been filled with Neurobiotin are reconstructed three-dimensionally
216
Weber
from a live video image by tracing and recording the spatial relations of each cell component (e.g., soma, dendrite, dendritic branch point, spine). Once the data have been entered, it is possible to rotate the cell image in any plane to view the vertical and horizontal dimensions of the cell. It also is possible to apply a number of standard analysis routines, including Sholl and polar analyses of the cell’s dendritic tree. These provide valuable information concerning the branching pattern and complexity of the dendritic arbor, and whether it shows any asymmetry (Fig. 3). In addition, it is possible to compare differences in dendritic spine content, number of dendritic nodes, number and type (primary, secondary, etc.) of dendritic branches and branch length, surface area of the soma and dendritic field, and dendritic field volume. All of these features are important because the structural integrity of the soma and dendritic field influences the spatial and temporal response properties of ganglion cells. Although the current focus is to correlate degenerative changes in ganglion cell morphology with changes in ganglion cell function, these techniques also can be used to establish whether different neuroprotective treatment strategies result in the recovery of normal ganglion cell structure-function following retinal injury.
Figure 3 Summary of the anatomical and physiological data that can be obtained using the isolated retina preparation. Both biophysical and visual stimulation information are collected, and these data then are compared with the ganglion cell’s structural integrity.
Ex Vivo and Whole-Mount Retinal Preparations
217
VII. USE OF THE ISOLATED RETINA PREPARATION FOR BIOPHYSICAL AND VISUAL STIMULATION STUDIES As noted previously, one of the main advantages of working with the isolated retina preparation is that its neural circuitry is preserved. Thus one can analyze not only the morphology, but also the intrinsic and visual response properties of single neurons (Fig. 3). Biophysical analyses, which evaluate membrane integrity, include measurements of resting membrane potential, membrane time constants, threshold levels for activation, and rate of firing in response to intracellularly applied steps of depolarizing current. In our setup, the stimuli used for these tests are presented, and the cellular responses collected and analyzed, using a commercial stimulus presentation and data acquisition and analysis package (pClamp6, Axon Instruments). Visual stimulation of the isolated retina can be achieved by presenting light stimuli (flashing or patterned) via either the camera port or the condenser of the microscope. Because we do not remove the sclera from the retina preparation, our system involves positioning a high-resolution XYZ video monitor (Tektronix 608) over the camera port of the microscope (Fig. 1). The video monitor is aligned with the center of the camera port, and the cell to be studied is positioned in the center of the microscope field. Visual patterns consisting of drifting and counter-phased light and dark bars are projected onto the isolated retina through the 40⫻ objective. The various patterns are generated and presented in random order by computer control of a Picasso CRT Image Synthesizer (Innisfree, Fenstanton, Cambridgeshire, England). Sinusoidal-modulated color diodes also are used to examine the response properties of primate ganglion cells.
VIII. USE OF THE WHOLE-MOUNTED RETINA FOR STUDYING NEUROPROTECTION As a first assessment of neuronal degeneration or survival following a particular manipulation or treatment strategy, it often is useful to start by examining a large population of neurons over a wide region of the tissue being studied. Since this is not possible using the isolated retina-intracellular technique, we used Nisslstained retinal whole mounts in our initial studies of retinal ganglion cell survival following optic nerve injury and brain-derived neurotrophic factor (BDNF: Regeneron Pharmaceuticals, Tarrytown, NY) treatment in the cat [13]. The retinal whole mount is a unique tissue preparation, and it is ideally suited for studying changes in ganglion cell size, number, and density. Cats too, offer several advantages for use in neuroprotection studies. First, as noted previously, the morpholo-
218
Weber
gies and central projection patterns of cat ganglion cells are comparable to those of the primate. Second, the cat eye, and particularly the vitreal chamber, is similar in size to that of the primate. This provides for a more direct comparison between drug application and neuronal survival than is possible following the treatment of different sized eyes, where dose and diffusion differences might be important limiting factors. Finally, it is relatively easy to achieve uniform Nissl staining of ganglion cells in the cat retina (Fig. 4). This is not always the case in the primate, where the higher number of ganglion cells, thicker nerve fiber layer, and often more difficult to remove vitreous can result in one having to stain for a specific area of interest.
Figure 4 Schematic and photomicrograph showing the approximate region of the cat retina used to study changes in ganglion cells size, number, and density following optic nerve crush and treatment with BDNF. Note the clarity of ganglion cell staining (Nissl) in the cat retina and the uniformity of cell size at the retinal eccentricity used (dotted line). OD: optic disc; AC: area centralis. (Adapted from Ref. 13.)
Ex Vivo and Whole-Mount Retinal Preparations
219
A central issue in any study that involves the measurement of ganglion cell numbers is cell identification. In the cat, as with most vertebrate retinas the ganglion cell layer also contains a significant number of amacrine cells [52–54]. While most ganglion cells in the rat can be identified using the sponge-based retrograde labeling method described previously, this approach is not readily applicable in the cat due to the greater depth and spatial separation of the colliculus and visual thalamus. Nevertheless, in the cat one can distinguish most ganglion cells and “displaced” amacrine cells based on soma size and cytoplasmic differences; the amacrine cells are among the smallest neurons in the ganglion cell layer, they show a low ratio of cytoplasmic-to-nuclear volume, and they often contain a prominent basophilic nuclear fold [52]. In our analysis of the neuroprotective effects of BDNF in cats following optic nerve injury, we selected for quantitative analysis a region of the retina that occupied 1.74 mm2 and was located 3.0 mm above and 1.5 mm temporal to the area centralis. This region was chosen because of the relatively constant size and density of ganglion cells in this area of the cat retina (Fig. 4)[26]. Welldefined retinal landmarks such as the optic disc, area centralis, and retinal blood vessel pattern were used, along with the microscope stage digitizer, to properly orient each retina on the microscope stage, and to standardize the starting point and stage movements for cell sampling. From the starting point, 42 digital images (41,000 µm2 /image) then were obtained systematically using the Hamamatsu high-resolution video camera and a 40⫻ objective. The retinal images were collected as three dorsal-ventral passes composed of 14 images each. Double counting was avoided by separating each sample column horizontally by 500 µm (using the Digitizer readout), and vertically by using each previous image as a reference for the next. Cell size, density, and number were determined directly from the digital images using an image analysis and measurement software package (Image Pro Plus, Media Cybernetics). By standardizing the measurements and using a region of the retina that could be identified reliably and sampled systematically across retinas, we were able to demonstrate the neuroprotective ability of BDNF in an eye comparable in size to that of the primate. Further, by testing different amounts of the drug, we were able to demonstrate not only that 30 µg of BDNF is needed to obtain maximum ganglion cell survival in the cat, but also that increasing the drug dose results in a decrease, rather than increase, in the level of neuroprotection (Fig. 5). Because the sample areas were standardized for each retina, changes in cell density mirrored the changes in cell number. Cell size measurements suggested a complex response among the different classes of ganglion cells. While 30 µg of BDNF retained the largest number of ganglion cells, 90 µg minimized the loss of medium-sized neurons and retained normal proportions of large, medium, and small ganglion cells.
220
Weber
Figure 5 Photomicrographs and graph demonstrating changes in ganglion cell number following optic nerve crush and either no treatment or treatment with different levels of BDNF. Treatment with 30 µg of BDNF resulted in the greatest cell survival, primarily by rescuing the medium-sized beta cells. Atrophic ganglion cells in the untreated retina are recognized by their irregular shape and clumped chromatin (arrows). (Adapted from Ref. 13.)
IX. SUMMARY In this chapter I have attempted to highlight some of the advantages of using the retina as a model for studying neuronal degeneration and neuroprotection. I also have tried, without going into detail, to make the reader aware of some of the more traditional approaches that have been used to study retinal ganglion cell morphology over the years. Although many of these might be considered outdated, there remain areas where such traditional methods could be applied to current studies of retinal degeneration and neuroprotection. Our use of the isolated retina and intracellular staining method, along with the Nissl-stained retinal whole mount, reflects both ends of the spectrum. The highly traditional Nisslstained approach continues to serve as a quick and reliable means for making a
Ex Vivo and Whole-Mount Retinal Preparations
221
first assessment of neuronal survival based on the evaluation of a large number of cells. The intracellular approach provides a more detailed analysis of the cellular changes that occur at the single cell level. It is hoped that the procedures described here will provide a foundation for others also seeking ways to enhance their understanding of the injured retina and its responses to different neuroprotection treatment strategies.
ACKNOWLEDGMENTS Supported by NIH/NEI grant EY11159 and grants from the Michigan State University Foundation, The Glaucoma Foundation, Alcon Laboratories, and The American Health Assistance Foundation.
REFERENCES 1. 2.
3.
4. 5. 6.
7.
8.
9.
10.
Dowling, JE. The Retina: An Approachable Part of the Brain. Cambridge, MA: Belknap Press of Harvard University Press, 1987. Sherman SM. Functional organization of the W-, Y-, and W-type retinal pathways in the cat: a review and hypothesis. Prog Psychobiol Physiol Psychol 1985; 11: 233–314. Kaplan E, Lee BB, Shapley RM. New views of primate retinal function. In: Osborne N, Chader G, ed. Progress in Retinal Research. Oxford: Pergamon, 1990:273– 336. Dacey DM. Physiology, morphology, and spatial densities of identified ganglion cell types in primate retina. Ciba Foundation Symposium 1994; 412–428. Rodieck RW, The First Steps in Seeing. Sunderland, MA: Sinauer, 1998. Cottee LJ, FitzGibbon T, Westland K, Burke W. Long survival of retinal ganglion cells in the cat after selective crush of the optic nerve. Eur J Neurosci 1991; 3: 1245–1254. Mey J, Thanos S. Intravitreal injections of neurotrophic factors support the survival and regrowth of axotomized retinal ganglion cells. Brain Res 1993; 602:304– 317. Silveira LCL, Russelakis-Carneiro M, Perry VH. The ganglion cell response to optic nerve injury in the cat: differential responses revealed by neurofibrillar staining. J Neurocytol 1994; 23:75–86. Mansour-Robaey S, Clarke DB, Wang Y-C, Bray GM, Aguayo AJ. Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells in adult rats in vivo. Proc Natl Acad Sci USA 1994; 91:1632–1636. Watanabe M, Sawai H, Fukuda Y. Number and dendritic morphology of retinal ganglion cells that survived after axotomy in adult cats. J Neurobiol 1995; 27:189– 203.
222
Weber
11.
Peinado-Ramo´n PM, Salvador MP, Villegas-Pe´rez MP, Vidal-Sanz M. Effects of axotomy and intraocular administration of NT-4, NT-3, and brain-derived neurotrophic factor on the survival of adult rat retinal ganglion cells. Invest Ophthalmol Vis Sci 1996; 37:489–500. Sawai H, Clarke DB, Kittelerova P, Bray GM, Aguayo AJ. Brain-derived neurotrophic factor and neurotrophin-4/5 stimulate growth of axonal branches from regenerating retinal ganglion cells. J Neurosci 1996; l6:3887–3894. Chen H, Weber AJ. BDNF enhances retinal ganglion cell survival in cats with optic nerve damage. Invest Ophthalmol Vis Sci 2001; 45:966–974. Pearson HE, Labar DL, Payne BR, Cornwell P, Aggarwal N. Transneuronal retrograde degeneration in the cat retina following neonatal ablation of visual cortex. Brain Res 1981; 212:470–475. Sherman SM, Spear PD. Organization of visual pathways in normal and visually deprived cats. Physiol Rev 1982; 62:738–855. Kalil RE. Removal of visual cortex in the cat: effects on the morphological development of the retino-geniculo-cortical pathway. In: Stone J, Dreher B, Rapaport DH, eds. Development of Visual Pathways in Mammals. New York: Alan R. Liss, 1984: 257–274. Payne BR, Pearson HE, Cornwell P. Transneuronal degeneration of beta retinal ganglion cells in the cat. Proc R Soc 1984; 222:15–32. Kalil RE. Reorganization of retinogeniculate connections in the cat following damage to the visual cortex. In: Bjorkland A, Aguayo AJ, Ottoson D, eds. Brain Repair. London: MacMillan Press, 1990; 285–307. Weber AJ, Kalil RE, Stanford LR. Dendritic field development of retinal ganglion cells in the cat following neonatal damage to visual cortex: evidence for cell class specific interactions. J Comp Neurol 1998; 390:470–480. Watanabe M, Rodieck RW. Parasol and midget ganglion cell of the primate retina. J Comp Neurol 1989; 289:434–454. Watanabe M, Sawai H, Fukuda Y. Number, distribution, and morphology of retinal ganglion cells with axons regenerated into peripheral nerve graft in adult cats. J Neurosci 1993; 13:2105–2117. Dacey D, Lee BB. The “blue-on” opponent pathway in primate retina originates from a distinct bistratified ganglion cell type. Nature 1994; 367:731–735. Izumi Y, Benz AM, Kirby CO, Labruyere J, Zorumski CF, Price MT, Olney JW. An ex vivo rat retinal preparation for excitoxicity studies. J Neurosci Meth 1995; 60:219–225. Romano C, Chen Q, Olney JW. The intact isolated (ex vivo) retina as a model system for the study of excitotoxicity. Progr Retinal Eye Res 1998; 17:465– 483. Weber AJ, Kaufman PL, Hubbard WC. Morphology of single ganglion cells in the glaucomatous primate retina. Invest Ophthalmol Vis Sci 1998; 39:2304–2320. Boycott BB, Wa¨ssle H. The morphological types of ganglion cells of the domestic cat’s retina. J Physiol (Lond) 1974; 240:397–419. Kolb H, Nelson R, Mariani A. Amacrine cells, bipolar cells and ganglion cells of the cat retina: a Golgi study. Vision Res 1981; 21:1081–1114.
12.
13. 14.
15. 16.
17. 18.
19.
20. 21.
22. 23.
24.
25. 26. 27.
Ex Vivo and Whole-Mount Retinal Preparations 28. 29. 30.
31. 32. 33. 34. 35.
36. 37. 38.
39. 40. 41. 42. 43.
44. 45. 46. 47. 48.
223
Isayama T, Berson DM, Pu M. Theta ganglion cell type of cat retina. J Comp Neurol 2000; 417:32–48. Saito H. Morphology and physiologically identified X-, Y-, and W-type retinal ganglion cells of the cat. J Comp Neurol 1983; 221:279–288. Fukuda Y, Hsiao C-F, Watanabe M, Ito H. Morphological correlates of physiologically identified Y-, X-, and W-cells in cat retina. J Neurophysiol 1984; 52:999– 1013. Stanford LR, Sherman SM. Structure/function relationships of retinal ganglion cells in the cat. Brain Res 1984; 297:381–386. Rodieck RW, Binmoeller KF, Dineen J. Parasol and midget ganglion cells of the human retina. J Comp Neurol 1985; 233:115–132. Kolb H, Linberg KA, Fisher SK. Neurons of the human retina: a Golgi study. J Comp Neurol 1992; 318:147–187. Dacey DM, Petersen MR. Dendritic field size and morphology of midget and parasol ganglion cells in the human retina. Proc Natl Acad Sci USA 1992; 89:9666–9670. Rodieck RW, Watanabe M. Survey of the morphology of macaque retinal ganglion cells that project to the pretectum superior colliculus, and parvicellular laminae of the lateral geniculate nucleus. J Comp Neurol 1993; 338:289–303. Dacey DM. Morphology of a small-field bistratified ganglion cell type in the macaque and human retina. Vis Neurosci 1993; 10:1081–1098. DeMonasterio FM, Gouras P. Functional properties of ganglion cells of the rhesus monkey retina. J Physiol (Lond) 1975; 251:167–195. Kaplan E, Shapley RM. The primate retina contains two types of ganglion cells, with high and low contrasts sensitivity. Proc Natl Acad Sci USA 1986; 83:2755– 2757. Merigan WH. Chromatic and achromatic vision of macques: role of the P pathway. J Neurosci 1989; 9:776–783. Lee BB. Receptive field structure in the primate retina (Minireview). Vision Res 1996; 36:631–644. Cajal SRY. The structure of the retina. In: Thorpe SA, Glickstein M, eds. Springfield, IL: Charles C Thomas, 1972. Wa¨ssle H, Peichl L, Boycott BB. Morphology and topography of on- and off-alpha cells in the cat retina. Proc R Soc Lond B 1981; 212:157–175. Villegas-Pe´rez MP, Vidal Sanz M, Bray GM, Aguayo AJ. Influences of peripheral nerve grafts on the survival and regrowth of axotomized retinal ganglion cells in adult rats. J Neurosci 1998; 8:265–280. Watanabe M, Sawai H, Fukuda Y. Axonal regeneration of retinal ganglion cells in the cat geniculocortical pathway. Brain Res 1991; 560:330–333. Mey J, Thanos S. Intravitreal injections of neurotrophic support the survival of axotomized retinal ganglion cells in adult rats in vivo. Brain Res 1993; 602:304–317. Zhan XJ, Troy JB. An, efficient method that reveals both the dendrites and the soma mosaics of retinal ganglion cells. J Neurosci Meth 1997; 72:109–116. Leventhal AG, Rodieck RW, Dreher B. Retinal ganglion cell classes in the old world monkey: morphology and central projections. Science 1981; 213:1139–1142. Perry VH, Oehler R, Cowey A. Retinal ganglion cells that project to the dorsal
224
49. 50. 51.
52.
53.
54.
Weber lateral geniculate nucleus in the macaque monkey. Neuroscience 1984; 12:1101– 1123. Martin PR. The projection of different retinal ganglion cell classes to the dorsal lateral geniculate nucleus in the hooded rat. Exp Brain Res 1986; 62:77–88. Stone J. The number and distribution of ganglion cells in the cat’s retina. J Comp Neurol 1978; 180:753–772. Vidal-Sanz M, Villegas-Pe´rez MP, Bray GM, Aguayo AJ. Persistent retrograde labeling of adult rat retinal ganglion cells with the carbocyanine dye Dil. Exp Neurol 1988; 102:92–101. Hughes A. Population magnitudes and distribution of the major modal classes of cat retinal ganglion cell as estimated from HRP filling and a systematic survey of the soma diameter spectra for classical neurons. J Comp Neurol 1981; 197:303– 339. Wong ROL, Hughes A. The morphology, number, and distribution of a large population of confirmed displaced amacrine cells in the adult cat retina. J Comp Neurol 1987; 255:159–177. Vaney DI. The mosaic of amacrine cells in the mammalian retina. In: Osborne N, Chader G, eds. Progress in Retinal Research Oxford: Pergamon, 1999:49–99.
13 Detection of Single-Cell Apoptosis William G. Tatton, Ruth M. E. Chalmers-Redman, and Nadine A. Tatton Mount Sinai School of Medicine New York, New York, U.S.A.
I.
INTRODUCTION
Our initial definitions of the apoptotic process were based on a set of morphological changes that occurred in the cell nucleus and cytoplasm. As first described by Kerr using electron microscopy [1] the earliest visible change in the apoptotic cell was the aggregation of chromatin into compact masses along the nuclear membrane. Eventually, more and more compact granular masses appeared and filled the nucleus, combined with a gradual reduction in nuclear volume. At the same time the cytoplasm displayed progressive condensation, but with preservation of organelles. “Apoptotic bodies” consisting of discrete spherical or ovoid fragments containing highly condensed chromatin were then phagocytosed and lysed by nearby cells. Initially these changes were thought to encompass the entire apoptotic process, but now it is clear that apoptotic bodies are part of the final degradative phase of apoptosis (Fig. 1). Interestingly, the appearance of these morphological changes is still used as a standard by which to determine whether cells have died via apoptosis. Now, however, we recognize that it is probably of greater importance to determine what apoptotic signaling pathway the cell has followed because knowledge of the pathway will ultimately provide us with information necessary to design agents to selectively block apoptosis. It is worth noting that not all apoptotic pathways require the involvement of the mitochondria. 225
226
Tatton et al.
Figure 1 Schematic illustrating the three phases of apoptosis: the initiation phase, decisional phase, and degradative phase. Four hypothetical initiating insults are shown: three that involve a mitochondrial decision phase and are mitochondrially dependent and a fourth that is mitochondrially independent. The multiple initiation and decision phase signaling pathways converge onto common events in the degradation phase. The diagram is meant to emphasize the role of mitochondrial decisional processes in some forms of apoptosis, including the role of BAD, BAX, and BCL-2-like proteins in the mitochondrial decisional process. (From Ref. 39.) The diagram is not meant to fully illustrate all apoptosis signaling pathways.
II. APOPTOTIC DEGRADATION FACTORS RELEASED BY DECREASED MITOCHONDRIAL MEMBRANE PERMEABILITY At least two factors that signal for apoptotic degradation and that are released from mitochondria have been identified: cytochrome c (CytC) and apoptosis initiation factor (AIF), a 50 kD flavoprotein. Released CytC interacts with apoptosis protease activating factor-1 (Apaf-1) [2], dATP/ATP and procaspase 9 to form a complex known as the apoptosome [3], in which procaspase 9 is converted to caspase 9 (see Fig. 1). Caspase 9 then converts procaspase 3 to activated caspase 3 along with activating caspase 7. The apoptosome seems to function as a multicaspase activating complex (see Ref. 4). Caspase 3 in turn activates DNA frag-
Detecting Single-Cell Apoptosis
227
mentation factor [5], an endonuclease activator which enables DNA cleavage and acinus which enables chromatin condensation [6]. The second factor released from mitochondria, AIF, induces DNA loss, peripheral chromatin condensation, and digestion of chromatin into 50 kbp-fragments [7]. The caspase-3 inhibitor ACE-DVAD-FMC inhibits chromatin condensation induced by AIF. The proteases that occur in the early stages of the apoptotic caspase-activation cascade, pro-caspase 2 and 9, have been found to be concentrated in the intermembranous space of mitochondria [8].
III. METHODS OF DETECTING SINGLE-CELL APOPTOSIS We have developed a number of protocols to examine apoptosis in situ, with particular interest in human postmortem tissue. These include a fluorescent double-labeling method to simultaneously visualize DNA fragmentation and apoptotic chromatin condensation [9] and several microwave-based antigen retrieval methods for immunocytochemistry of apoptosis-related proteins on paraffin sections. We have also modified these protocols for use on fixed, frozen sections and on cultured cells when examining apoptotic changes in different model systems. Gavrieli first published a method of in situ end labeling [10] that allowed the visualization of DNA fragmentation in tissue sections. Popularly known as the TUNEL method, or ISEL (in situ end labeling), this was a breakthrough in the examination of apoptotic cell death in tissue sections and in cultured cells. Unfortunately, ISEL when used alone, cannot provide unequivocal identification of an apoptotic nucleus. Endonucleases that digest DNA can create single- and/or double-strand breaks reflected as high molecular weight (50–300 kbp) DNA fragments on gel electrophoresis [11]. Often the DNA digestion continues, with the ultimate production of low molecular weight oligonucleosome-sized fragments [12,13]. Single-strand DNA breaks likely accumulate in the linker regions between oligonucleosomes prior to the double-strand DNA breaks which create a “ladder” pattern when observed by gel electrophoresis [14]. The ISEL method was considered to be selective in detecting these double-stranded breaks between oligonucleosomes. However, the terminal transferase enzyme (TdT) commonly used for ISEL is generally supplied with a protocol that requires increased cobalt chloride (2.5 mM) in the reaction mix, allowing TdT to label double-strand breaks (protruding, blunt, and recessed ends) as well as single-strand DNA breaks [15,16]. It is important to note that cells dying by necrosis have also been reported to contain singleand/or double-strand breaks, and thus the ISEL method may not always distinguish between the two modes of cell death [17–19]. A second, independent method is therefore required to confirm apoptosis where there is demonstrable
228
Tatton et al.
DNA fragmentation. The cyanine dye YOYO-1 (Molecular Probes) binds to DNA (and RNA) and allows the visualization of apoptotic chromatin condensation—an event that is independent of DNA fragmentation. Use of confocal laser microscopy has offered us the additional advantage of high resolution of these nuclear labels (see Refs. 9,20,21). IV. ISEL/YOYO FOR HUMAN BRAIN POSTMORTEM PARAFFIN SECTIONS Standard, 5-µm-thick paraffin sections are stored at 4°C after sectioning. Slides are run through a standard series of baths to remove all paraffin and rehydrate the sections. 1. 2. 3. 4. 5. 6. 7.
8. 9.
10. 11.
Xylene 1—15 min. Xylene 2—10 min. 100% Ethanol 1—10 min. 100% Ethanol 2—10 min. 50% Ethanol—10 min. dH2O—10 min. Antigen Unmasking Solution (Vector Labs). Dilute 3.77 mL of this stock solution in 400 mL distilled H2O in a glass container (a Pyrex measuring cup is good). Add eight slides in a plastic rack. Microwave slides for 3.5 – 4 min, power setting 7—final temperature reached is approximately 60°C. Temperature should not be above 62°C in order to avoid producing DNA degeneration due to excessive heating (i.e., 65°C or more) and nonspecific labeling. Let slides cool, submerged in the buffer for 30 min on the bench. Important: Only eight slides should be done at one time. We have found that if there are more slides in the rack, the heating is uneven across all slides, resulting in unreliable labeling. Power settings and time vary with each microwave to achieve a final temperature of 60°C (we use an 850 W Kenmore with a turntable). 3% H2O2 /Methanol, room temperature, l0 min. PBS rinse—3 min. RNAseA digest (100 µg/mL in 2⫻ SSC) at 37°C for 25 min). Digest time should be decreased for retinal sections and for embryonic tissue. The RNAse digest decreases background since terminal transferase (TdT) can tail from single-strand RNA. 2⫻ SSC rinse, 5 min. Proteinase K digest (20 µg/mL in T.E. pH 8) at 37°C for 3–5 min. Too long a digest time will increase nonspecific labeling. We do a very brief digest for embryonic tissue (or retina), always less than 3 min.
Detecting Single-Cell Apoptosis
229
12. 0.1 M glycine/0.1 M Tris pH 7.2, ice-cold, 10–15 min. Stops the enzyme digest and also helps reduce nonspecific background. 13. Equilibration buffer (Intergen), room temperature, 20–30 min. Blot off excess before adding reaction mix. 14. TdT/BODIPY-dUTP reaction mix, at 37°C for 60 min. Thirty minutes provides inadequate and unreliable labeling; 90 min does not provide any appreciable increase in signal to noise ratio. Cover each tissue section with a parafilm “coverslip” (allows the spread of a small volume of labeling reaction mix evenly over section). Can use 12 µL on a mouse brain cross-section, 20 µL on most human SNc sections (unless it is really big, then 25 µL). 15. 2⫻ SSC stop bath, 3⫻ 10 min, prewarmed to 37°C. Slides are washed in 400 mL volumes of 2⫻ SSC to effectively remove nonspecifically bound BODIPY-dUTP. It is important to use a large volume to remove excess label and that the temperature be 37°C. 16. PBS, 2⫻ 5 min. Important to rinse all 2⫻ SSC away to get good YOYO staining. 17. YOYO (1 :500 in PBS), 30 min, dark, RT. 18. PBS rinses, 5⫻. 19. Use Aquamount (Gurr) or GelMount (Biomeda) for coverslips—both are water soluble and dry quickly. Let slides dry in fume hood overnight, then store in fridge. We have found that the best color/brightness will be in the first 72 h (apoptotic nuclei appear orange if you have a filter that lets you do simultaneous red/green imaging). After 72 h, the fluorescence is a little less bright, but still good for at least another 3–4 months. In order to get consistent labeling it is important that all concentrations of enzymes and their buffers are accurate and are applied at the stated temperature. This is also essential for the 2⫻ SSC stop baths that follow the labeling reaction. We prewarm all digestion enzymes before applying to the slides. The TdT reaction mix is made up in the sequence given below just before use. The TdT enzyme is stored at ⫺20°C and should be kept in a freezer block or on ice when in use. We use less enzyme and dUTP in our labeling reaction than is given in the product spec sheet. We have tested a broad range and found that 1.4 µL of dUTP and 1.4 µL of TdT per 100 µL reaction volume gave as good or better labeling on tissue sections than the volumes recommended in the manufacturer’s spec sheet.
TdT Reaction Mix 20 µL 5⫻ TdT buffer 10 µL CoCl2 67.2µL dH2O
230
Tatton et al.
1.4 µL 1.4 µL 100 µL
BODIPY-TR-14-dUTP (Molecular Probes) TdT (Boehringer Mannheim/Rocher)
We tested different manufacturer’s TdT samples; Boehringer/Roche was the most reliable and gave the best labeling. The sample includes the enzyme, reaction buffer, and CoCl2. Tdt: Boehringer/Roche, cat. # 220 582 BODIPY-TR-dUTP: Molecular Probes, cat. # C7618 YOYO-1: molecular probes, cat # Y3601 Equilibration Buffer: Intergen, cat # S7106 Other useful things to control for successful labeling: 1. A short postmortem interval for all pathology cases (ours were less than 6 h on average). You will notice in many European publications the postmortem intervals are 20 h or greater—this cannot produce reliable ISEL because many nuclei will label because of extensive DNA degradation that occurs postmortem. Try to get less than a 12 h postmortem interval if you can. 2. The tissue should be fixed in buffered formalin and kept refrigerated to provide good DNA preservation for ISEL. Note: We saw no difference in labeling with tissue kept in fixative for 6 months, 2 years, or 3 weeks when we initially tested human brain cryosections. 3. Proteinase K digestion is critical—too long and your background signal is increased (and it is impossible to get rid of), too short and you don’t get good access to the DNA in the nucleus. Always do a test run on a few slides, trying different times. Embryonic tissue is more fragile, needs maybe about half the digest time of adult brain. With some fragile tissues, we have found that a brief exposure to Neuropore (Trevigen) offers good permeabilization. Again the time of exposure must be tested (i.e., 5–15 min).
V.
IMMUNOCYTOCHEMISTRY FOR APOPTOSISRELATED PROTEINS—HUMAN POSTMORTEM PARAFFIN SECTIONS
A.
GAPDH (Monoclonal Antibody—Chemicon)
We and others found that GAPDH (glyceraldehyde 3-phosphate dehydrogenase) plays a role in the apoptotic pathway of some cells. In our partially neuronally differentiated PC12 cells that enter apoptosis after serum and NGF withdrawal, we found that there is an increase in cytoplasmic GAPDH as well as nuclear accumulation of this protein. Blocking GAPDH nuclear accumulation, either by antisense oligonucleotides [22] or with deprenyl-related propargylamines [23],
Detecting Single-Cell Apoptosis
231
reduced the incidence of apoptotic nuclei in vitro. It is not yet clear what role GAPDH plays in the nucleus of cells entering apoptosis. We do know that nuclear GAPDH accumulation occurs before the nuclear degradative changes. This protocol works well for adult brain tissue, but we have noted that we get better results if sections are stored cold, and reacted soon after being cut. Dewax sections according to standard protocol (see ISEL/YOYO for details). 1. Antigen Unmasking Solution (Vector Labs). 400 mL in Pyrex measuring cup. Heat on High power for about 5 min to bring to boil. Add slides in plastic rack and heat at Power setting 8 for 40–60 s to reach boiling, the let slides boil for 1 min only. Let slides cool in buffer for 30–40 min on benchtop. 2. PBS rinse, 1⫻ 5 min. 3. Methanol, ⫺20°C, 10 min. Rinse with PBS. 4. RNAse A digest, 100 µg/mL in 2⫻ SSC, prewarmed to 37°C, for 3 min only! This step allows antibody access to GAPDH, which is often found in association with RNA in the cell. 5. Ice-cold 0.1 M glycine/0.1 M Tris buffer (pH 7.2) for 5 min to stop enzyme digest and decrease background. 6. PBS rinse, room temperature, 5 min. 7. 10% Normal Goat Serum/0.2% Tween 20, 15 min to block. 8. Primary Antibody (1: 200) in 1% NGS/0.2% Tween 20/0.1 M PBS overnight, 4°C. Individual sections are covered with a parafilm “coverslip.” 9. PBS rinses, room temperature, 4⫻. 10. Secondary Antibody, goat anti-mouse IgG–Alexa 594 (1:200), in 1% NGS/0.2% Tween 20/0.1 M PBS, 1 h 37°C. 11. PBS rinse, 3⫻. 12. Coverslip with Gelmount (Biomeda) or Aquamount (Gurr) as described above. 1. Human Retinal Sections We have found that the microwave heat treatment can be too harsh (as detailed for GAPDH) and have often substituted a brief (about 10 min, room temperature) Neuropore permeabilization step instead or no permeabilization step depending on the quality of the case samples. Care trust be taken not to overexpose retinal sections to this agent. We also find that there is a high level of nonspecific, dull background autofluorescence in human adult (aged) retina largely in the red range that we do not see in the brain/spinal cord sections. This is distinct from the bright autofluorescent granules that accumulate in retinal ganglion cell neurons. For this reason, we do not usually do more than one antibody on human postmortem sections
232
Tatton et al.
and often use DAPI as a nuclear counterstain (bright blue) or YOYO (bright green) to visualize the chromatin pattern. Confocal imaging allows the section to be viewed in the far red range and thus distinguish nonspecific autofluorescent signal from antibody-specific signal (see Ref. 20, for details). Note that without an RNase A digest, YOYO will also stain cytoplasmic RNA. 2. Fluorescent Secondary Antibody Conjugates We routinely use the Alexa dyes from molecular probes. They come in a broad range of excitation/emission wavelengths, the colors are bright, intense and hold up extremely well to confocal laser scanning. As a general rule, we have found that YOYO staining should be done after the secondary antibody step. B.
Caspase 3-Active Fragment
Although there are a few antibodies now available to the activated fragment of caspase 3, we have found that we get excellent, reproducible results with the New England BioLabs product. Control protein samples are also provided in case you want to use this product for immunoblots. We have also used an antibody against caspase 3 (recognizes both pro-caspase 3 and activated caspase 3) from Pharmingen, which is also very reliable (see Ref. 20). For use on human brain sections, we still use a microwave antigen retrieval method (as described for GAPDH) but do not use any subsequent enzyme digests. The protocol described below has been used on paraffin sections of human retina. Dewax according to the standard protocol. 1. Rinse sections in PBS for 5 min, room temperature. 2. Block 10% NGS/02.% Tween 20/0.1 M PBS, 1 h RT. Alternately, substitute a 10 min permeabilization step with Neuropore followed by 20 min blocking step with 10% NGS/0.1 M PBS. Blot off excess. 3. Primary antibody (rabbit polyclonal) 1:100 in 1% NGS/0.2% Tween/ PBS, overnight at 4°C. Sections are covered with a parafilm coverslip. 4. PBS rinse, 2⫻ 5 min. 5. Secondary antibody—Alexa 594 goat anti-rabbit 1: 200 in 1% NGS/ 0.2% Tween/PBS, 40 min, 37°C. 6. PBS rinse, 2⫻ 5 min. 7. YOYO (1:500) in PBS, 30 min, RT, in the dark. 8. PBS rinses, 3⫻ 5 min. 9. Coverslip with Gelmount or Aquamount. C.
Caspase 3 or Bax—Rapid Method for Brightfield Microscopy
This is an alternate protocol developed for human brain paraffin sections and is an adaptation of the Biomeda Autoprobe kit, which gives an intense one-step
Detecting Single-Cell Apoptosis
233
labeling for brightfield viewing. Unlike the Autoprobe original protocol, we have extended some of the times because of antigen accessibility in sections from old formalin-fixed blocks (as opposed to freshly perfused tissue). We have also increased the primary antibody incubation to overnight at 4°C. Note that the pepsin reagent digest times would likely need some changes in incubation times if used for human retinal sections. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
12. 13. 14.
15. 16. 17. 18. 19. 20. 21. 22.
Xylene 1—15 min. Xylene 2—10 min. 100% Ethanol—10 min. 100% Ethanol—10 min. Endoblocker (1 mL in 5 mL 100% ethanol), 5 min, room temperature. 95% Ethanol—3 min. 50% Ethanol—3 min. d H2O—5 min. Pepsin reagent 12 min at 37°C (prewarmed). 1⫻ AutoBuffer (AB), three quick rinses. Antigen Unmasking Solution (Vector labs) 3.77 mL stock AUS in 400 mL dist. H2O. Bring solution to boiling (in glass Pyrex measuring cup (4-cup size), at high power, about 5 min. Add 8–10 slides in plastic rack to solution, bring to boil in approximately 30 s (power setting 8) and then let boil for 1 min only. Let slides sit in buffer on the benchtop for 30–40 min. 1⫻ AB 5 min RT Tissue conditioner (Biomeda), 1 drop/mL AB, 10 min, room temperature. Primary antibody—Caspase 3 (1 :500, Pharmingen) or Bax (Santa Cruz) (1: 200) in primary antibody diluting buffer (Biomeda) O/N 4°C. AB rinse, 2 ⫻ 5 min. Secondary antibody, universal antibody (Biomeda kit), 40 min, 37°C AB rinse, 2 ⫻. Peroxidase reagent, 37°C, 30 min. AB rinse, 1⫻. Working chromogen solution, 15 min. Dist H2O, stop bath, 2 ⫻ 5 min. Coverlip with Gelmount or Aquamount (Gurr).
Useful information. You can likely find a different antigen retrieval method for every day of the month when you search the literature. We found that it helps to start with the mildest approach possible (i.e., use an unmasking solution at room temperature first) and then gear up to more aggressive heat treatments and then possibly choose to combine heat treatment with some type of enzymatic digest. We have found that we will vary our retrieval protocols according to
234
Tatton et al.
whether we are working with adult tissue or embryonic tissue (gentler methods), brain or retina (gentler methods). But most important is to find out what your antigen of interest does in the cell. Is it found in a membrane, is it bound to RNA, is it found in vesicles, in the nucleus? All of these will help determine what you need to do to make it accessible, often with a limited number of trial runs. VI. METHODS FOR EXAMINING APOPTOTIC CHANGES IN CULTURED CELLS Many of our detection methods were first developed for in vitro models of apoptosis. These include ISEL/YOYO, measurement of mitochondrial membrane potential, immunocytochemistry for “apoptotic” proteins, DNA-binding dyes to visualize chromatin condensation, and so on. We generally use YOYO-1 for looking at chromatin condensation for tissue sections or cultured cells. This choice has really been determined by the capabilities of our krypton-argon laser. The Hoechst dye (bisbenziamide) and DAPI are also both excellent dyes for examining nuclear DNA and will not stain cytoplasmic RNA. Both are bright blue and require UV excitation, are long-lasting and slow to quench (see molecular probes). Either can be applied to tissue sections following antibody incubations in place of YOYO-1. A.
Hoechst Dye or YOYO-1 for Chromatin Condensation For monolayer cell culture: 1. 2. 3. 4.
5. 6.
B.
Wash cells 3⫻ with PBS to remove serum/media. Fix cells with 4% paraformaldehyde, 30 min on ice. Rinse well with PBS, 3⫻. Add Hoechst dye (5 µg/mL PBS), 20–30 min at room temperature, in the dark. Or use YOYO-1 (1 :1000 in PBS) for 20–30 min at room temperature, in the dark. Wash 5⫻ PBS. Mount in Aquamount (Gurr). We have found that Gelmount is not a good choice for cultured cells stained with YOYO-1. It appears to “bleed out” of the cells after 24 h; therefore, we recommend Aquamount for mounting cell culture coverslips.
ISEL/YOYO for Monolayer Cell Cultures 1. Wash cells 3⫻ with PBS to remove serum/media. 2. Fix cells with 4% paraformaldehyde, 30 min, on ice. With neuronal cultures we find it best to keep cells in fixative in the fridge overnight after the initial 30 min on ice, then rinse with PBS the next day.
Detecting Single-Cell Apoptosis
235
3. Rinse well with PBS, 3⫻. 4. Dip in 100% methanol, 1–2 min, room temperature. It is important to keep this step short as apoptotic nuclei can dissociate from the remaining cell soma in cultured cells. 5. PBS rinse, 1⫻, 5 min. 6. RNAseA digest (100 µ/mL in 2⫻ SSC), 10 min, 37°C 7. Rinse, 2⫻ SSC, 2 ⫻ 5 min. 8. 0.1 M glycine/0.1 M Tris, pH 7.2, 20 min RT. Note that the proteinase K digest is omitted (we found that it really served to detach cells from the coverslip rather than improve ISEL signal). 9. Equilibration buffer (Intergen), 15–30 min, room temperature. 10. TdT/BODIPY dUTP-red/reaction mix, 60 min, 37°C. 11. Rinse, 2⫻ SSC (prewarmed), 3⫻ 10 min, 37°C. 12. PBS rinse, 2⫻ 5 min. 13. YOYO-1 (1 :1000 in PBS), 20–30 min, room temperature, in the dark. 14. PBS rinses, 5⫻. 15. Aquamount coverslips, let dry in fume hood, then store in fridge. We grow cells on 22-mm square coverslips for ISEL/YOYO or 12 mm round coverslips. The 22-mm coverslips are kept in 35-mm petri dishes and are easy to flush with different solutions. A parafilm coverslip can easily be applied to these in order to use a minimal volume of TdT reaction mix (30–40 µL). If using the round coverslip, we have found it easiest to transfer them to a ceramic well plate rather than keep them in a 24-well plastic tray. We invert them in each well, adding reagents under the coverslip (it will save on TdT reaction mix volume—use about 70 µL).
VII. MITOCHONDRIAL MEMBRANE POTENTIAL AS A MEASURE OF APOPTOTIC CHANGE There is an increasing body of evidence that mitochondria play a critical decisional role in number of forms of apoptosis including those initiated in neurons by withdrawal of trophic factors, hypoxia/ischemia, and glutamate excitotoxicity. Using partially differentiated PC 12 cells, apoptosis is induced by removal of NGF and serum. We have found that there is a significant decline in mitochondrial membrane potential after trophic withdrawal, and that this precedes the appearance of nuclear apoptotic degradation. We have used a fixable mitochondrial potentiometric dye, chloromethyltetramethylrosamine (CMTMR or Mitotracker orange) to measure Ψ∆ M. Recently, the use of CMTMR and related dyes has been questioned (see Ref. 24). The mitochondrial uptake of cationic dyes used to estimate ψ∆M depends on the difference between the plasma membrane potential and Ψ∆M , and on the
236
Tatton et al.
time of incubation [25]. We [26–30] and others [31–34] have used rhodamine derivatives, with a chloromethyl moiety that binds to matrix thiols for this purpose. These derivatives offer the advantage that fixatives bind them to matrix proteins allowing immunocytochemistry to be subsequently carried out in the same cells. The purported disadvantage in using the fixable dyes has been thought to result from the dyes not being Nernstian in their outward movement from mitochondria (i.e., if Ψ∆M decreases during the incubation time, the dyes will not leave mitochondria in proportion to the extent of the decreases). Accordingly, the fixable dyes were proposed to offer less sensitive measures of Ψ∆M [24]. Furthermore, the binding of the fixable dyes to thiols has led to the consideration of the possibility that reactive oxygen species (ROS) levels rather than Ψ∆M might determine their in situ fluorescence [35] and that the dyes might damage mitochondria and open the PTP, dissipating Ψ∆M [36,37]. We have compared the sensitivity of the fixable dye CMTMR to other well-established mitochondrial potentiometric dyes and found that if cells are incubated with CMTMR for short periods before fixation (i.e., l5–30 min), they offer sensitive measures of Ψ∆M that are in agreement with those provided by other dyes such as JC-1. Ideally, Ψ∆M should be measured using more than one dye in order to confirm that observed changes in fluorescence can be reproduced. However other potentiometric dyes such as TMRM and JC-1 (molecular probes) require the capability of live cell imaging which may not be possible for many laboratories. For this reason we have only supplied a protocol for CMTMR. CMTMR (Mitotracker Orange) enters mitochondria proportionally to the potential difference between the cytoplasmic compartment and mitochondrial matrix. After entering mitochondria, the chloromethyl groups of CMTMR react with thiols on proteins and peptides to form aldehyde-fixable conjugates and remain sequestered in the mitochondria after permeabilization and fixation [33,38]. The CMTMR mitochondrial fluorescence represents the highest level of potential difference between the mitochondrion and adjacent cytoplasmic compartment (or the nuclear compartment) of the living cell during the period of dye exposure prior to fixation. CMTMR fluorescence intensity as an estimate of Ψ∆M is measured and plotted as a frequency distribution (see Ref. 30 for details). A.
Mitotracker (Orange) plus YOYO-1 • 20 ⫻ 50 µg vials; MW 427.37 • Add 850 µL sterile DMSO (N.B. — the full 850 µL will not fit in vial from Molecular Probes); gives a stock solution of 0.1 mM. • Store as 25 µL aliquots at ⫺20°C, protect from light and moisture. 1. Dilute mitotracker stock 1:1000 in prewarmed media. 2. Aspirate media from cells and add mitotracker-media for 15 min, return cells to 37°C, CO2 incubator.
Detecting Single-Cell Apoptosis
237
3. 4. 5. 6. 7. 8.
Rinse 3⫻ in HBSS. Fix cells with 4% paraformaldehyde, for 20–30 min on ice. Rinse 2–3⫻ in PBS. Expose cells for 1–2 min to methanol, room temperature. Rinse cells with PBS, 2⫻. Incubate cells with YOYO-1 (1:1000 in PBS) for 30 min, room temperature. 9. Rinse cells in PBS, 5⫻. 10. Mount coverslips with Aquamount. Once dry, store slides at 4°C. B. Mitotracker, Bax (or GAPDH), and YOYO-1 Note that for cells that have been incubated with CMTMR and fixed, no detergent should be used in later steps (such as Triton or Tween) because it will result in CMTMR bleeding out of the mitochondria. CMTMR-treated cells may be stored in PBS at 4°C for at least 1 week, if necessary. In the following protocol we have used Bax antibody (Santa Cruz, N-20 fragment), Bcl-2 (Santa Cruz), or GAPDH (Chemicon), but virtually any antibody may be used that does not require a detergent permeabilization to access its antigen. Note that we have also exposed CMTMR-treated cells to a cold ethanol postfix without any problems, and we have also used a diluted (1:3) solution of Neuropore for 5 min when a permeabilization step was required with no damage to CMTMR. 1. 2. 3. 4. 5. 6. 7.
8. 9. 10. 11.
Rinse cell-coverslips with PBS, 1⫻ 5 min. Expose briefly to methanol, 1–2 min, room temperature. Rinse in PBS, 2⫻ 5 min. Block with 10% NGS/PBS for 20 min, room temperature. Primary antibody Bax or Bcl-2 (1:200) or GAPDH (1 :500) in 1% NGS/PBS, overnight, 4°C. Rinse cells with PBS, 2⫻ 5 min. Secondary antibody, goat anti-mouse or goat anti-rabbit Cy5-IgG (or Alexa 690), 1:200 in 1% NGS/PBS for 1 h, room temperature (or 40 min, 37°C). Rinse cells with PBS, 2⫻ 5 min. YOYO-1 (1 :1000 in PBS) for 20–30 min, room temperature. Rinse cells with PBS, 5⫻. Mount coverslips with Aquamount, let air dry, and then store at 4°C, in the dark.
This protocol allows for triple labeling of the cell. Note that to visualize Cy5, laser excitation of the fluorophore will be necessary. CMTMR emits in a range similar to Texas Red or TRITC; therefore, appropriate secondary antibody conjugates must be chosen that do not overlap with CMTMR or with YOYO-1. Alter-
238
Tatton et al.
nately, a double-label may be employed—CMTMR with YOYO-1 or DAPI, CMTMR with any other antibody using an Alexa 488 -IgG conjugate (if no confocal is available). Note that GAPDH is used at 1:500 dilution on cultured cells, but at 1: 200 on tissue sections.
REFERENCES 1. 2.
3.
4.
5.
6.
7.
8.
9.
10.
11. 12.
13.
Kerr JFR, Wyllie AH, Currie AR. Apoptosis, a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972; 26:239–257. Zou H, Henzel WJ, Liu XS, Lutschg A, Wang XD. Apaf-1, a human protein homologous to C-elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell 1997; 90:405–413. Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 1997; 91:479–489. Bratton SB, MacFarlane M, Cain K, Cohen GM. Protein complexes activate distinct caspase cascades in death receptor and stress-induced apoptosis. Exp Cell Res 2000; 256:27–33. Liu XS, Zhou H, Slaughter C, Wang XD. DFF, a heterodimeric protein that functions downstream of caspase-3 to trigger DNA fragmentation during apoptosis. Cell 1997; 89:175–184. Sahara S, Aoto M, Eguchi Y, Imamoto N, Yoneda Y, Tsujimoto Y. Acinus is a caspase 3-activated protein required for apoptotic chromatin condensation. Nature 1999; 401:168–173. Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers GM, Mangion J, Jacotot E, Constantini P, Loeffler M, Larochette N, Goodlett DR, Aebersold R, Siderovski DP, Penninger JM, Kroemer G. Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 1999; 397:441–446. Susin SA, Lorenzo HK, Zamzami N, Marzo I, Brenner C, Larochette N, Prevost MC, Alzari PM, Kroemer G. Mitochondrial release of caspase-2 and -9 during the apoptotic process. J Exp Med 1999; 189:381–394. Tatton NA, Maclean-Fraser A, Tatton WG, Perl DP, Olanow CW. A flourescent double-labeling method to detect and confirm apoptotic nuclei in Parkinson’s disease. Ann Neurol 1998; 44: S142–148. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in-situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992; 119: 493–501. Walker PR, Sikorska M. Eudonuclease activities, chromatin structure, and DNA degradationn in apoptosis. Biochem Cell Biol 1994; 72:615–623. Cohen GM, Sun XM, Snowden RT, Dinsdale D, Skilleter DN. Key morphological features of apoptosis may occur in the absence of internucleosomal DNA fragmentation. Biochem J 1992; 286:331–334. Oberhammer F, Wilson JW, Dive C, Morris ID, Hickman JA, Wakeling AE, Walker PR, Sikorska M. Apoptotic death in epithelial cells: cleavage of DNA to 300 and/
Detecting Single-Cell Apoptosis
14.
15. 16.
17.
18.
19.
20.
21.
22.
23.
24.
25. 26.
27.
28.
239
or 50 kb fragments prior to or in the absence of internucleosomal fragmentation. EMBO 1993; 12:3679–3684. Peitsch MC, Polzar B, Stephan H, et al. Characterization of the endogenous deoxyribonuclease involved in nuclear DNA degradation during apoptosis programmed cell death. EMBO 1993; 12:371–377. Deibel M, Coleman MS. Biochemical properties of purified human terminal deoxynucleotidyltransferase. J Biol Chem 1980; 255:4206–4212. Tabor S, Struhl K, Scharf SJ, Gelfand DH. Enzymatic manipulation of DNA and RNA. In: Current Protocols in Molecular Biology. New York: Wiley & Sons, Inc, 1999. Didenko VV, Hornsby PJ. Presence of double-strand breaks with single-base 3′overhangs in cells undergoing apoptosis but not necrosis. J Cell Biol 1996; 135:1369– 1376. Dypbukt JM, Ankarcrona M, Burkitt M, Sjoholm A, Orrenius S, Nicoteri P. Different pro-oxidant levels stimulate growth, trigger apoptosis or produce necrosis of insulin-secreting RINmSF cells: the role of intracellular polyamines. J Biol Chem 1994; 269:30553–30560. Funk A, Rudel J, Fellbaum C, Hofler H. Specific in situ end labeling of apoptosis shows different rates of programmed cell death in non-Hodgkin lymphomas. Verh Dtsch Ges Pathol 1994; 78:318–320. Tatton NA. Increased caspase 3 and bax immunoreactivity accompany nuclear GAPDH translocation neuronal apoptosis in Parkinson’s Disease. Exp Neurol 2000; 166:29–43. Tatton NA, Kish SJ. In situ detection of apoptotic nuclei in the substantia nigra compacta of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated mice using terminal deoxynucleotidyl transferase labelling and acridine orange staining. Neuroscience 1997; 77:1037–1048. Ishitani R, Chuang DM. Glyceraldehyde-3-phosphate dehydrogenase antisense oligodeoxynucleotides protect against cytosine arabinonucleoside-induced apoptosis in cultured cerebellar neurons. Proc Natl Acad Sci USA 1996; 93:9937–9941. Carlile GW, Chalmers-Redman RM, Tatton NA, Pong A, Borden KE, Tatton WG. Reduced apoptosis after nerve growth factor and serum withdrawal: conversion of tetrameric glyceraldehyde-3-phosphate dehydrogenase to a dimer. Mol Pharmacol 2000; 57:2–12. Bernardi P, Scorrano L, Colonna R, Petronilli V, Di Lisa F. Mitochondria and cell death: mechanistic aspects and methodological issues. Eur J Biochem 1999; 264: 687–701. Rottenberg H, Wu S. Quantitative assay by flow cytometry of the mitochondrial membrane potential in intact cells. Biochim Biophys Acta 1998; 1404:393–404. Chalmers-Redman RM, Fraser AD, Carlile GW, Pong A, Tatton WG. Glucose protection from MPP⫹-induced apoptosis depends on mitochondrial membrane potential and ATP synthase. Biochem Biophys Res Commun 1999; 257:440–447. Grant SM, Shankar SL, Chalmers-Redman RM, Tatton WG, Szyf M, Cuello AC. Mitochondrial abnormalities in neuroectodermal cells stably expressing human amyloid precursor (hAPP751). Neuroreport 1999; 10:41–46. Jing Y, Dai J, Chalmers-Redman RM, Waxman S. Arsenic troxide selectively in-
240
29.
30.
31.
32.
33. 34. 35. 36.
37.
38. 39.
Tatton et al. duces acute promyelocytic leukemia cell apoptosis via a hydrogen peroxide-dependent pathway. Blood 1999; 94:2102–2111. Sugrue MM, Wang Y, Rideout H, Chalmers-Redman RM, Tatton WG. Reduced mitochondrial membrane potential and altered responsiveness of a mitochondrial membrane megachannel in p53-induced senescence. Biochem Biophys Res Commun 1999; 261:123–130. Wadia JS, Chalmers-Redman RME, Ju WJH, Carlile GW, Phillips JL, Fraser AD, Tatton WG. Mitochondrial membrane potential and nuclear chages in apoptosis caused by serum and nerve growth factor withrawal: time course and modification by (-)-deprenyl. J Neurosci 1998; 18:932–947. Gilmore K, Wilson M. The use of chloromethyl-X-rosamine (Mitotracker red) to measure loss of mitochondrial membrane potential in apoptotic cell is incompatible with cell fixation. Cytometry 1999; 36:355–8. Macho A, Decaudin D, Castedo M, Hirsch T, Susin SA, Zamzami N, Kroemer G. Chloromethyl-X-Rosamine is an aldehyde-fixable potential-sensitive fluorochrome for the detection of early apoptosis [see comments]. Cytometry 1996; 25:333–340. Poot M, Gibson LL, Singer VL. Detection of apoptosis in live cells by MitoTracker red CMXRos and SYTO dye flow cytometry. Cytomery 1997; 27:358–364. Poot M, Pierce RC. Detection of apoptosis and changes in mitochondrial membrane potential with chloromethyl-X-rosamine. Cytomery 1999; 36:359–360. Ferlini C, Scambia G, Fattorossi A. Is chloromethyl-X-rosamine useful in measuring mitochondrial transmembrane potential? Cytometry 1998; 31:74–75. Minamikawa T, Sriratana A, Williams DA, Browser DN, Hill JS, Nagley P. Chloromethyl-X-rosamine (Mito Tracker Red) photosentises mitochondria and induces apoptosis in intract human cell. J Cell Sci 1999; 112:2419–2430. Scorrano L, Petronilli V, Colonna R, Di Lisa F, Bernardi P. Chloromethyltetramethylrosamine (Mitotracker Orange) induces the mitochondrial permeability transition and inhibits respiratory complex, I: Implications for the mechanism of cytochrome c release. J Biol Chem 1991; 274:24657–24663. Calarco PG. Polarization of mitochondria in the unfertilized mouse oocyte. Dev Genet 1995; 16:36–43. Tatton WG, Olanow CW. Apoptosis in neurodegenerative diseases: the role of mitochondria. Biochim Biophys Acta 1999; 1410:195–213.
14 Imaging of Retinal Ganglion Cells Joshua P. Vrabec and Leonard A. Levin University of Wisconsin–Madison Madison, Wisconsin, U.S.A.
I.
INTRODUCTION
In recent years, the study of neuronal apoptosis has expanded into the field of digital technology. Many aspects of neuronal physiology, such as ion flux, generation of reactive oxygen species, and changes in membrane potential can now be observed in real time. As a result, digital imaging of individual neurons labeled with condition-dependent fluorescent dyes will most certainly help define the temporal relationships of the many interlinked apoptotic pathways. Neuronal apoptosis involves the generation of reactive oxygen species (ROS) and a loss of mitochondrial membrane potential (∆ΨM) via the opening of the mitochondrial permeability transition pore (MPTP) [1,2]. Although the precise relationships of these events to the grand scheme of the apoptotic cascade are not clear, they likely are key events in the initiation of apoptosis. Therefore, blocking either ROS release or MPTP opening may be attractive pharmacological targets for neurodegenerative diseases in the future. In order to better elucidate these events in real time, we describe our method for digitally imaging rat retinal ganglion cells (RGCs) using fluorescent light microscopy. Methods for detecting release of superoxide anion and for detecting changes in the mitochondrial membrane potential are outlined below. Similar methods can be used for imaging other reactive oxygen species, calcium levels, pH, and other intracellular processes [3–6].
241
242
Vrabec and Levin
II. RETROGRADE LABELING OF RETINAL GANGLION CELLS In order to differentiate retinal ganglion cells from other retinal neurons in mixed culture, we take advantage of the unique anatomical connections between the brain and the cell bodies of RGCs. When an appropriate dye is injected into the superior colliculus of an anesthetized rat, the dye will flow via retrograde axonal transport in the axons of RGCs back to the cell bodies located in the inner retina [7]. Take care to select a fluorescent probe that will work in this manner. DiI-C18, DAPI, and several other dyes are available, but it is important that the dye chosen have an excitation/emission spectrum that will not interfere with the experimental probe. For example, if a red and green emitting dye such as JC-1 will be used for imaging ∆ΨM , then you should select a blue-violet dye (e.g., DAPI) for the retrograde labeling of RGCs. In the methods described below for RGC imaging of both ∆ΨM and superoxide anion, DAPI was used for retrograde labeling.
III. RETINAL GANGLION CELL CULTURE Long-Evans rat pups, age P3, are injected with either DiI-C18 (Molecular Probes) or DAPI (Molecular Probes) into each superior colliculus for the retrograde labeling of retinal ganglion cells [8]. After waiting for at least 2 days, pups are sacrificed by CO2 asphyxiation and their retinas are dissected. The retinas are then dissociated according to previous methods [8], and plated on a poly-L-lysinetreated eight-well chambered cover glass for 2 h 37°C, 5% CO2, 80% humidity. After the incubation period is complete, follow the specific JC-1 or HEt staining protocols for imaging. Cultures are incubated for 2 h for two reasons. First, the dissociated cell mix that is added to the well needs time to become attached to the poly-L-lysine– coated surface of the wells. Second, some of the cells in primary neuronal cultures such as these begin to undergo stress- or axotomy-related apoptosis, which can be as much as 50% by 24 h [8,9]. Imaging soon after culture reduces the possibility that the cell being imaged will be in the process of dying from axotomy, although this obviously cannot be excluded. Longer-term cultures are more likely to have significant amounts of apoptosis.
IV. SETTING UP FOR IMAGING A.
Choice of Dyes
JC-1 (Molecular Probes) is a dual-emission fluorescent probe that associates with the mitochondrial inner membrane. The dye exists as two forms (monomer and
Imaging of RGCs
243
J-aggregates) depending on the membrane potential of the inner mitochondrial membrane with which the dye is associated. The monomer associates with the inner membrane regardless of membrane potential, but J-aggregates will only form if an adequate membrane potential is present. While both forms of JC-1 have identical excitation spectra, each form has its own unique emission spectrum. The monomer form emits at 535 nm and the J-aggregates have a peak emission at 580 nm. With the proper imaging equipment, these unique characteristics allow one to measure the emission intensity of each form of JC-1 and determine the membrane status of mitochondria in intact RGCs. Specifically, one can measure the ratio of 580/535 nm emission over time to detect changes in ∆ΨM brought on by specific pharmacologic treatments. Dihydroethidium (HEt) is a dye that has a peak emission around 580–590 nm when converted to its active form by superoxide anion. When used in this method, it can be used to detect the superoxide burst occurring during apoptosis [10,11]. B. Equipment Needed Listed below is the equipment necessary for imaging over short time intervals, and also the additional devices needed if one is to perform imaging experiments that will last several hours (microscope stage incubator, pH control, etc.). While each device can be purchased separately, there are some companies that will provide most, if not all, the necessary imaging equipment in a bundled package. Equipment necessary for imaging: 1. Inverted fluorescent microscope (Zeiss Axiovert S100 or a similar model) fitted with appropriate filters (discussed below) and 100⫻ oilimmersion objective 2. Sutter Instruments Lambda 10-2 filter wheel (or a similar model) with filters. To image with JC-1, a dual-emission probe, the filter wheel should be fitted along the emission light path (i.e., between the microscope and digital camera). 3. Dye-specific filter sets (obtain from Chroma or Omega Optical) a. RGC identification DAPI: Excitation: 330 nm Dichroic: 400 nm Emission: 450 nm b. ∆ΨM measurement JC-1: Excitation: 480 nm Dichroic: 505 nm Emission: 535 nm and 580 nm filters (Lambda 10-2 filter wheel) c. Superoxide anion measurement
244
Vrabec and Levin
HEt: Excitation: 515 nm Dichroic: 565 nm Emission: 580 nm long pass 4.
5. 6. 7. 8.
1.0 log (10%) neutral density filter. This should be fitted in the path of the microscope’s light source, and will cut down on the fading of JC-1. Photometrics SenSys digital camera UniBlitz shutter driver and shutter (important component for reducing the amount of dye fading) Pentium III computer or better, loaded with MetaFluor 4.0 software (Universal Imaging Corporation), or later version Lab-Tek II chambered cover glass for cell culture (0.2–0.5 mL working volume, 0.7 cm2 growth area) (Fisher Scientific)
Optional accessories for long-term studies (all but the CO2 tank and regulator are obtained from Harvard Apparatus): CSMI microscope plate incubator (#65-0101) TC-202A temperature controller w/ thermistor (#65-0045) pH controller (#70-2116) Watson Marlow 205CA perfusion pump (#72-0501) Perfusion tubing (#72-0571) CO2 tank with regulator
V.
MEASURING MITOCHONDRIAL MEMBRANE POTENTIAL OR SUPEROXIDE ANION IN CULTURED RETINAL GANGLION CELLS
A.
Dye Treatment
Due to the dyes’ extreme sensitivity to light, all procedures involving JC-1 or HEt should be carried out in the dark. Prepare a 4 mM (JC-1) or 3.2 mM (HEt) stock solution in dimethyl sulfoxide (DMSO). Aliquots should be stored at ⫺20°C and protected from light. I.
Staining protocol for JC-1 1. Thaw out the 4 mM stock solution of JC-1 in the 37°C water bath. 2. Add 7 mL of Neurobasal medium to a 15 mL Falcon tube, and place in a 37°C water bath. 3. Once thawed, add 5.37 µL of the 4 mM stock to 7 mL of Neurobasal medium in a sterile incubator. Important: Do not add the JC-1 directly to the media. Tilt the tube containing the media at a 45° angle, then
Imaging of RGCs
4. 5.
6. 7. 8.
9.
10. 11.
245
add the JC-1 to the inside lip of the tube, taking care to keep the JC-1 away from the media. Then, carefully close the tube, and IMMEDIATELY vortex the solution. Now that the 2-h incubation is complete, remove the plate from the incubator. Using a sterile glass pipette, carefully aspirate the culture media from a well (touching only the side of the well), and immediately add 400 µL of the JC-1 treating solution (prepared above). Repeat with the other wells on the plate. Place the plate in the 37° incubator for 15 min. While cells are in the incubator, add 200 µL of 1 M HEPES buffer to 9.8 mL of HBSS and vortex to make a 20 mM HEPES washing solution. If buffering is not desirable, use HBSS alone. After the 15 min incubation is complete, aspirate a well, and immediately add 400 µL of the HEPES/HBSS wash to the well. Repeat with the other wells on the plate. Wash cells a second time: Aspirate a well, and this time add 350 µL of HEPES/HBSS solution to the well. Repeat with the other wells. Place plate on microscope stage and prepare for imaging.
2. Staining protocol for HEt HEt begins a slow conversion to ethidium after it contacts cell media. Therefore, it is only loaded into cells just before imaging is to commence. 1. All procedures done in darkness/minimal light. 2. Thaw HEt aliquot (3.2 mM in DMSO). 3. Make a 1000-fold dilution with the imaging media (clear HBSS and chemical treatment) as the diluent. 4. Place the 3.2 µM HEt solution in 37° water bath 15 min prior to cell culture incubation completion. 5. Remove plate of cells from incubator after 2 h incubation. 6. Aspirate the media from each well, adding 350 µL of 3.2 µM HEt solution immediately after each aspiration. 7. Incubate plate at 37° for 10 min. 8. Place plate on microscope stage and prepare for imaging. B. Imaging of Mitochondrial Membrane Potential in Retinal Ganglion Cells Due to differences in the number and size of mitochondria in cells, the absolute amount of JC-1 fluorescence emitted will vary. As a consequence, for a population of cells in a well, the changes in ∆ΨM after an appropriate stimulus will also
246
Vrabec and Levin
vary substantially. Therefore, one cannot simply measure the average fluorescence from a well if the purpose is to measure the timing of acute changes in ∆ΨM . Instead, individual cells should be observed for these types of experiments, and this can be achieved by selecting specific regions for measurement using the MetaFluor software. An advantage of measuring ∆ΨM in this way is that small changes in ∆ΨM due to pharmacological treatments can accurately be detected. However, the number of RGCs in a single field of a mixed retinal culture under a 100⫻ objective is low, and does not allow for the measurement of many cells simultaneously. A section below describing data analysis discusses how to combine experiments to obtain significance. During imaging, the digital camera, filter wheel, and microscope shutter are all under the control of the imaging software program (MetaFluor), which integrates all of their functions. This allows the experimenter to take a relatively passive role during the actual acquisitions of images. Before beginning any experiment, the program must be set up properly. Variables such as acquisition number, time between acquisitions, camera settings, optical filters to use, and so on must be decided on beforehand. The advantage is that MetaFluor does the rest after a simple mouse click starts the imaging process. Remember that for measuring ∆ΨM , two separate emission filters are needed (535 nm ⫽ areas of low ∆ΨM , 580 nm ⫽ areas of high ∆ΨM ). Thus, a typical imaging cycle for JC-1 will involve taking one picture at 535 nm, followed by a picture at 580 nm. MetaFluor can then automatically calculate a ratio of the 580 nm and 535 nm fluorescent intensities (580/535 nm). The data can be transferred while imaging to Microsoft Excel, for further analysis after the experiment. A representative from the software company is an excellent resource for describing the complete capabilities of the software package. The experiments measuring ∆ΨM start by obtaining images once per minute for 5 min and calculating the ratio of 580/535 nm. After this initial baseline fluorescence has been determined, a chemical treatment is added to the well. Then, additional images are acquired once per minute for 20 min, up to 12 h (long-term imaging equipment needed, see above.) Chemical treatments are selected for their ability (proven or proposed) to alter ∆ΨM in some fashion. An excellent control is 10 µM valinomycin, a K⫹ ionophore for the mitochondrial inner membrane, which causes collapse of the ∆ΨM within minutes. 1. Imaging Protocol 1. 2.
Using an appropriate filter set for DAPI, locate an RGC in the well. Switch to the 100⫻ oil-immersion lens, and refocus. The nucleus should be stained a bright, vivid blue-violet color. Check that the cell appears intact and living. Note: Other light purple–colored cells may be found in the well, but
Imaging of RGCs
3.
4.
5. 6.
247
these are not retinal ganglion cells. Some DAPI-stained RGCs die and lyse during the incubation period, releasing small amounts of DAPI that may be taken up by nearby cells (bipolar cells, macrophages, etc.). These cells that have secondarily taken up DAPI appear as very faint purple cells scattered throughout the plate. If one concentrates on locating only cells that have their nucleus stained (not the entire cytoplasm), and that staining is an intense purple (not faint), then there should be no confusing the RGCs from other retinal cells in this system. Change the filter block to the position for JC-1 excitation. The cell should have bright orange spots or streaks (these are the stained mitochondria) against a bright blue background. If the cell appears entirely green, it is probably dead as a result of the dissociation procedure, and should not be studied. Focus on the cell, and again check that it is centered in the field of view. Manually close the shutter and divert the microscope light to the filter wheel/digital camera apparatus Using the imaging software, complete one cycle of imaging. The digital camera will acquire one image of the fluorescent intensity at both 535 nm and 580 nm. Using the imaging software, encircle a region of the just-imaged cell for measurement. Start the experiment, which will continue to image the same cell to the specifications of the program setting while making fluorescent measurement.
2. Imaging Protocol for Dihydroethidium (HEt) This procedure is essentially identical to that used for imaging ∆ΨM with JC-1. 3. Common Problems with Obtaining an Image • The light path in the microscope was not properly diverted to the filter wheel/digital camera apparatus. • The shutter did not open. Light should pass through the plate during the image acquisition—if it did not, then the shutter did not open properly. This usually can be fixed by turning the power for the shutter controller off, then back on after a few seconds. Or, try hitting the Reset button on the shutter controller a few times, then try imaging again. • The proper filters are not in the filter block and/or filter wheel. • The cell moved out of the camera’s field of view. Oil microscopy using a 100⫻ objective can be tricky. Any slight jarring of the microscope apparatus or vibration of the table it is placed on may cause the cell to move. Addition of chemical treatments by pipetting into the well while
248
Vrabec and Levin
imaging may also cause problems. Make sure the microscope is on a very stable, heavy table. The poly-L-lysine solution used to adhere the cells should be in good condition. If necessary, addition of agents by perfusion pump may be necessary.
REFERENCES 1. 2.
3.
4. 5. 6. 7.
8. 9. 10.
11.
Zamzami N, Susin SA, Marchetti P, Hirsch T, Gomez-Monterrey I, Castedo M, et al. Mitochondrial control of nuclear apoptosis. J Exp Med 1996; 183(4):1533–1544. Marchetti P, Castedo M, Susin SA, Zamzami N, Hirsch T, Macho A, et al. Mitochondrial permeability transition is a central coordinating event of apoptosis. J Exp Med 1996; 184(3):1155–1160. Zochowski M, Wachowiak M, Falk CX, Cohen LB, Lam YW, Antic S, et al. Imaging membrane potential with voltage-sensitive dyes. Biol Bull 2000; 198(1):1– 21. Frank J, Biesalski HK, Dominici S, Pompella A. The visualization of oxidant stress in tissues and isolated cells. Histol Histopathol 2000; 15(1):173–184. Takahashi A, Camacho P, Lechleiter JD, Herman B. Measurement of intracellular calcium. Physiol Rev 1999; 79(4):1089–1125. Johnson I. Fluorescent probes for living cells. Histochem J 1998; 30(3):123–140. Vidal-Sanz M, Villegas-Perez MP, Bray GM, Aguayo AJ. Persistent retrograde labeling of adult rat retinal ganglion cells with the carbocyanine dye diI. Exp Neurol 1988; 102(1):92–101. Levin LA, Clark JA, Johns LK. Effect of lipid peroxidation inhibition on retinal ganglion cell death. Invest Ophthalmol Vis Sci 1996; 37:2744–2749. Kortuem K, Geiger LK, Levin LA. Differential susceptibility of retinal ganglion cells to reactive oxygen species. Invest Ophthalmol Vis Sci 2000; 41:3176–3182. Budd SL, Castilho RF, Nicholls DG. Mitochondrial membrane potential and hydroethidine-monitored superoxide generation in cultured cerebellar granule cells. FEBS Lett 1997; 415(1):21–24. Scanlon JM, Reynolds IJ. Effects of oxidants and glutamate receptor activation on mitochondrial membrane potential in rat forebrain neurons. J Neurochem 1998; 71(6):2392–2400.
15 Evaluation of Retinal Function: Electroretinography Marc He´bert University of Laval Quebec City, Quebec, Canada
Pierre Lachapelle McGill University Montreal, Quebec, Canada
I.
INTRODUCTION
The purpose of this chapter is to provide the reader with a working knowledge of functional testing of the retina by describing the most common procedure used for that purpose—namely, the electroretinogram whether evoked to a diffuse (flash ERG) or structured (pattern ERG and multifocal ERG) light stimulus. It is not, however, the scope of this chapter to discuss in great details the origin of the different components of the ERG, especially since for several of them this is still an unresolved issue. The electroretinogram (ERG) identifies the biopotential that is generated by the retina in response to a light stimulus. To date, the ERG still represents the test of choice to investigate the functioning retina whether in its normal state or altered either as a result of pathology, aging, or intoxication (medication or otherwise), to name a few. Given the neural origin of the retina, its functioning is often affected by neurotoxic or neuroactive components. On numerous occasions the ERG was shown to be a precious ally to the ophthalmologist and neuroscientist who wished (1) to detect the effect of neuroactive components on the function of the retina; (2) investigate the long-term sequels that these components had on the functional integrity of the retina; and (3) examine the possible revers249
250
He´bert and Lachapelle
ibility of the effect either spontaneously or as a result of therapy. Similarly, the ERG was also shown to be a significant adjunct to the clinician’s armamentarium as it not only permitted one to diagnose and characterize a pathology, even in instances where there were no other obvious clinical signs, but also the information gathered was instrumental in identifying the possible site of malfunction. In the following sections, we will describe the different techniques used to record and analyze the electroretinogram. The accent will be on the human ERG because the techniques used are now standardized. However, it should be noted that human techniques can be used with little modifications on all sorts of animal species, such as: dogs [1] rats [2], mice [3], pigs [4], birds [5], and rabbits [6], to name a few. Although our aim is not to conduct a review of the literature to support this claim, modified human ERG techniques were previously used to study the toxicity of inhaled trichloroethylene on the rabbit retina [7,8], determine the short- and long-term sequels of postnatal oxygen exposure in newborn rats (animal model of human retinopathy of prematurity) [9,10], evaluate the beneficial effect of an intravitreal injection of ciliary neurotrophic factor (CNTF) in rescuing the degenerating photoreceptors in RDS mice [3], and, more recently, study the potentially harming effect of Sildafil (Viagra) on the retinal function of mice affected with a retinal degeneration similar to a form of human retinitis pigmentosa (RP) [11]. Similarly, the electroretinogram is frequently used to quantify the functional consequences of the new genotypes that are generated through selective breeding or created through genetic manipulations [12].
II. GENERAL OVERVIEW OF ELECTRORETINOGRAPHY The electroretinogram (ERG) represents the electrical response that is generated by the entire retina when stimulated by a flash of light of adequate energy. It is often compared to the electrocardiogram (ECG) in that it is similarly composed of a series of waves that are presumed to originate from different retinal cells. However, unlike the ECG, which represents the ongoing activity of the myocardium, in order to generate an ERG response one must stimulate the retina with light. The ERG is thus an evoked potential generated by the excited retina and recorded at a distance (usually the cornea) from the latter. It represents a weighted average of all the retinal cells excited by the flash of light. It is fair to say that the ERG does not exist as such, unlike the ECG. It is a creation that enables us to synchronize the retinal activity and consequently greatly facilitate its study. This concept is of extreme importance if one wishes not only to understand the origin of the ERG but also to appreciate the need to normalize the ERG acquisition parameters in order to extract as much meaningful functional information as possible. Continuing with the ECG analogy, in the latter
Evaluation of Retinal Function
251
case it is the electrode position that is critical for inter-laboratory comparisons, whereas with the ERG it is the stimulating parameters that are of the utmost importance. Added to the above, one must also take into consideration the unique physiology of the retina, which permits it to work under a wide range of luminance levels. This is for the great part due to the two types of photoreceptors: rods (for night time or scotopic vision) and cones (for daytime or photopic vision), which are normally found in the retinas of most higher vertebrates. One of the aims of electroretinography, whether used as an investigative tool or as a diagnostic test, is to separate the cone ERG from the rod ERG in order to distinctly assess the function of each type of photoreceptor. As we will see later on, standardized ERG protocols were written with that specific aim in mind.
III. COMPONENTS OF THE ERG AND THEIR ORIGINS The electroretinogram was the first biopotential ever recorded. As early as 1865, Holmgren published a study in which he reported the recording of the electrical response of the eye of frogs upon excitation by a light stimulus [13]. Some 12 years later, Dewar was first to report human ERG recordings [13]. It is, however, Granit who contributed most to our understanding of the ERG wave genesis. In his classic study, which made use of ether intoxication in cats, Granit showed that the ERG was basically composed of three different processes, each of which gives rise to a different wave [14]. The three processes, named PI, PII, and PIII to reflect their order of disappearance as the level of anesthesia deepened, were shown to correspond to the c-wave, b-wave, and a-wave, respectively. With the refinement of the stimulating and recording techniques, new waves were added to the three studied by Granit. However, it is customary in modern electroretinography to concentrate on the two major waves of the ERG: namely, the a- and b-waves as well as the oscillatory potentials (OPs), which are components of high frequency and low voltage normally seen on the rising limb of the b-wave (see Fig. 1 for ERG waves identification) [15]. In the following paragraphs we will briefly review the postulated site of origin of these waves. A. The a-Wave As shown at Figure 1, the a-wave, which is negative in polarity, is the first major component of the ERG. To date, the consensus situates the generator of the awave at the level of the photoreceptors [13]. Depending on the state of retinal adaptation, either the cones alone or the cones and the rods will contribute to the genesis of the a-wave [13]. It should be noted that there is no pure rod ERG a-
252
He´bert and Lachapelle
Figure 1 Representative photopic broadband (recording bandwidth: 1–500 Hz) electroretinogram recorded from a sedated Guinea pig. The vertical arrow points to the onset of the stimulus. Components of the ERG are identified as follows: awave (a), b-wave (b), and oscillatory potentials (OPs). The amplitude of the a-wave is measured from the prestimulus baseline to the trough of the a-wave, whereas that of the b-wave is measured from the trough of the a-wave to the peak of the b-wave. Peak times are measured from flash onset to peak. Vertical calibration: 30 µV. Horizontal calibration: 20 ms.
wave as such because in scotopic condition, an a-wave is recordable only in response to flashes of light of intensities which are in the photopic range (see Sec. V). These responses are usually referred to as mixed cone and rod ERG because both photoreceptors are claimed to contribute to their genesis. Analysis of the leading edge of the a-wave, using a computational approach, is often used to further characterize the physiology of the photoreceptor response as measured with the ERG [16]. It is of interest to note that in some animal species (such as rats and mice, for example), the photopic (cone) ERG is devoid of a recordable
Evaluation of Retinal Function
253
a-wave despite a large b-wave. The exact reason for this remains obscure to date. By convention, the amplitude of the a-wave is measured from baseline to the trough of the a-wave. B. The b-Wave This is the most prominent and consequently the most important component of the ERG response. As shown in Figure 1, the b-wave is positive in polarity and said to be generated as a second-order potential by the Mu¨ller cells, which are glial cells running through the entire thickness of the retina, from the outer to the inner limiting membrane [13]. Sieving et al. (1994) suggested that the photopic ERG would be generated as a result of the synchronized activation of the ON-depolarizing bipolar cells (ON-DBC) and OFF hyperpolarizing bipolar cells (OFF-HBC), each contributing in sequence to the shaping of the b-wave [17]. The ON-DBC provides the impetus to push the b-wave up (i.e., from the trough of the a-wave to the peak of the b-wave) while the OFF-HBC limits the amplitude of the b-wave by pulling the retinal potential from the b-wave peak to a baseline value. This concept of ERG b-wave genesis is known as the PUSH-PULL hypothesis. By convention, the amplitude of the b-wave is measured from the trough of the a-wave (when present) to the peak of the b-wave. C. The Oscillatory Potentials The oscillatory potentials (OPs) identify the high-frequency components of the ERG, which are often seen as ripples on the ascending limb of the b-waves (Fig. 1) [18]. Fourier analysis reveals that while the a- and b-waves of the ERG are of a frequency domain of about 60 Hz or less, the OPs are usually of a frequency domain greater than 100 Hz. As we will see later, this difference in frequency domain is put to a use when one wishes to selectively record the OPs with minimal contamination from the slower a- and b-waves of the ERG. Practically all the retinal components, with the exception of the photoreceptors and the Mu¨ller cells, have been suggested as possible candidate for the genesis of the OPs [18]. Although there is yet no consensus as to their exact origin, it is important to note that there is more and more evidence to suggest that the terminology “oscillatory potentials” is probably a misnomer, because there are studies published to date that clearly show that the OPs do not (collectively) represent an oscillation such as the vibration of a retinal element or membrane, as the name could suggest. Past studies have shown that the number of OPs will vary as a function of stimulus intensity [19] (see also Fig. 4), inter-stimulus interval [20], retinal adaptation [21,22], pharmacological manipulation [23,24], and pathology [25,26], to name a few. Results from the above-mentioned studies clearly suggest that each OP would be generated by a different, and most probably functionally independent
He´bert and Lachapelle
254
retinal element. Compared to the other ERG components, the OPs were also shown to be significantly more sensitive to alteration in the retinal environment, whether acquired or innate [18,25]. Finally, as it currently stands, flash-evoked ERG responses do not include component(s) specifically assigned to the activity of the ganglion cells or beyond. There have been reports suggesting that some long-latency post–b-wave components (photopic i-wave or long-latency OPs) might be attributable to the activation of the retinal ganglion cells and/or the optic nerve including the chiasm [27– 29]. However, these claims were not verified elsewhere.
IV. RECORDING PROCEDURE A.
Preparation of the Subject
As indicated at the onset, the electroretinogram is an evoked potential. Consequently, in order to record, one must synchronize the recording of the retinal potential with the stimulus that will generate this response. To achieve this goal, several criteria must be met, the most important of which being full cooperation from the subject. Taking again the ECG analogy, the heart will continue to beat and thus an ECG recording will be possible (maybe with some difficulties) whether the subject (human or animal) is compliant or not. This is somewhat different for the ERG because the subject must cooperate to a certain degree if one wishes to record good quality and reproducible responses. If the subject voluntarily keeps its eyes shut, an adequate ERG recording is simply impossible. It is for that reason that, in animal experimentation, the subjects are systematically anesthetized for the procedure and not for the pain that might be involved, because it is, depending on the type of recording electrodes, an almost harmless procedure. In this laboratory we have successfully used a mixture of ketamine and xylazine to record ERGs from a variety of animal species including rats, mice, Guinea pigs, rabbits, and birds. Human subjects, on the other hand, need not be anesthetized or even sedated—even newborns or young infants. At least this has been our experience of more than 25 years of recording ERGs in a pediatric center. B.
Electrode and Recording Parameters
1. Electrode Types and Position (active, reference, ground) According to the ERG standard of the International Society for Clinical Electrophysiology of Vision (ISCEV) [15], the active electrode should be corneal or as close as possible to the cornea. It is for that reason that the electrode of choice is the contact lens electrode (Henkes, Burian-Allen, etc.; see Fig. 2, top). The
Evaluation of Retinal Function
255
reason for this is simple. As mentioned above, the ERG is an evoked potential that is recorded on the surface of the eye but originates from the retina. Therefore, the recording site is at some distance from the origin of the potential. Consequently, the closer we are from the source of the potential, the larger this potential will be. This is true of all ERGs, human or animal, and for that reason contact lenses have been developed for several animal species, including the mouse. Another advantage to the contact lens electrodes is the fact that their construction also includes a blepharostat whose function is to keep the eyelids open during the procedure, thus optimizing the delivery of light to the retina (see Fig. 2, top). Contact lens electrodes must, however, be used with a topical anesthetic, as they are painful to wear, and also with a viscous interface (usually methylcellulose) in order to protect the cornea from a possible abrasion. These electrodes cannot be worn for more than 20–30 min because the risk of corneal abrasion increases with time. Also, several sizes must be purchased in order to accommodate the different sizes of eyes (especially when pediatric and adult patients are seen in the same clinic) as well as the waiting time due to sterilization (in a clinical setup). The latter point is of importance as these electrodes are quite expensive. As an alternative to the contact lens electrode, the fiber electrodes are gradually becoming more and more popular, the most popular one being the DTL fiber electrode, which is made of a nylon yarn coated with silver (Fig. 2, middle) [30,31]. Fiber electrodes offer several advantages, such as painless even after several hours of wear (in humans), no need for topical anesthesia and disposable. They must, however, be used with great care and their positioning controlled as even minor changes in position will have a significant impact on the amplitude of the resulting ERG. Regardless of the type of electrode used, with human subjects, reference and ground electrodes (surface electrodes) are usually positioned at the external canthi (or forehead) and earlobes (earclip electrodes) respectively; for animal subjects, they are respectively placed in the mouth and inserted in the tail (needle electrodes) or subcutaneously in the leg. Electrode impedance should be minimized as much as possible (⬍5KΩ). 2. Delivering the Stimulus: The Ganzfeld The electroretinogram being an evoked response, a good ERG examination, whether for clinical or investigative purpose, will try to record the electrical response from as much retinal tissue as possible. It is for that reason that it is recommended to use a light diffuser (or Ganzfeld: Fig. 2, bottom) whose purpose, by design, is to diffuse the stimulus (and light adapting background light when used) to retinal eccentricities as far as the ora serata [32]. This is even more important in experimental situations where the rod function must be accurately assessed because, as we know, rods are most numerous in the peripheral retina. The same applies to animal experimentation where human Ganzfeld or stimulator
256
He´bert and Lachapelle
Evaluation of Retinal Function
257
Table 1 Normative Data Obtained in 40 Normal Subjects (22 men; 18 women, age: 5–82) Amplitude (µV) average (lower-upper limits)
Peak time (ms) average (lower-upper limits)
SF (Standard flash)
a-wave
b-wave
a-wave
b-wave
PHOTOPIC
28.8 (20–38) na
13.4 (11.5–14.8) na na
30 (26–32) 26.7 (24.4–29.6) 67.4 (57–82) 40.2
SF Single flash
⫺2.5 log SF (pure rod)
na
110 (85–140) 90 (65–120) 150
SF (mixed rodcone)
165
(110–220) 250
14.2
(120–270)
(160–365)
(12.7–15.8)
SF Flicker 30hz SCOTOPIC
(32–46.3)
na: not applicable (e.g., wave not present in response).
specifically designed for animal experimentation are used [33]. It is important to stress here that an adequate stimulator with diffusing possibilities should always be used if one seeks reproducible responses of high diagnostic potentials. Placing the subject (human or animal) directly in front of the light source (flash or other) is clearly insufficient. 3. Amplification and Recording Bandwidth As stated above, the electroretinogram is a far field potential in that it represents the electrical activity evoked from the retina but recorded at the cornea—that is, in humans, about 25 mm from the source. The amplitude of the resulting biopoFigure 2 In order to obtain signals of the highest possible quality, it is imperative that the recording electrode be in contact with the eye, whether the ERGs are recorded from human or animal subjects. This is best achieved with the corneal contact lens (top picture), which normally includes a built-in blepharostat in order to keep the eyelids open. Use of these electrodes requires topical anesthesia and thus limits the wearing time to some 30 min or so. Alternative ERG electrodes, such as the DTL fiber electrode (second picture: DTL fiber is seen running on margin of the lower lid: arrow head), placed deep in the conjunctival bag significantly extends the recording time because they are basically painless as well as harmless. Once the eye returns to its primary position (third picture), the electrode does not interfere with the optical axis. Finally, ERGs should always be evoked in full field (Ganzfeld: bottom picture) or full field–like condition in order to stimulate as much of the retina as possible.
258
He´bert and Lachapelle
Figure 3 Typical ERG responses representative of the ISCEV standards. Vertical line(s) indicate flash onset. Photopic responses (background set at 25 cd/m2: single flash (A), oscillatory potentials (B), flicker 30 Hz (C). Scotopic responses: pure rod (D), mixed rod-cone (E), and oscillatory potentials (F). Recordings were obtained using a UTAS-3000 system (LKC Technologies, Inc, Gaithersburg, MD) along with DTL fiber electrodes (DTL Plus ElectrodeTM, Retina Technologies, Scranton, PA) and dilated pupils. Bandwidths were between 1 and 500 Hz for the broadband signals (A,C,D,E) or 75 and 500 Hz for the OPs (B,F). SF ⫽ ISCEV standard flash which was set at 3 cd/m2 /s. DA ⫽ dark adaptation. Vertical calibration: 25 µV (photopic tracings) and 50 µV (scotopic tracings). Horizontal calibration: 25 ms.
tential is in the range of approximately 100 µV. This value varies with the intensity of stimulation, the level of retinal adaptation (photopic usually yields smaller responses compared to scotopic: see Table 1), the type of recording electrodes (usually the largest responses are obtained with contact lens electrodes) [34] as well as species. Given the relatively low voltage generated, it is necessary to amplify this signal in order to obtain a measurable response. Similarly, it was
Evaluation of Retinal Function
259
Figure 4 Photopic ERG luminance-response function. A gradual increase in the intensity of the stimulus (bottom to top) augments the amplitude of the a- and bwaves of the broadband ERG (left column). However, while the amplitude of the a-wave continues to grow regularly, that of the b-wave reaches a maximum at .64 log cd⋅m⫺2⋅s, after which it decreases, a phenomenon known as the photopic hill. Similarly, the timing of the a-wave shortens with brighter flashes, whereas that the b-wave increases. The photopic waveforms also include another component identified as the i-wave, the origin of which is still debated [27]. A similar increase in flash energy will not only impact on the amplitude of the OPs but also on the number. At threshold, the OP response includes only one major OP—namely, OP2 (OP1 identifies the small notch seen prior to OP2 in some tracings). A gradual increase in intensity will add OP3 and OP4 to this initial response. Horizontal calibration: 25 ms. Vertical calibration: 50 (ERG) and 25 (OPs) µV.
mentioned above that the ERG response includes several components, some of which are easily differentiated with their temporal frequency domain. Consequently, in order to record as complete an ERG response as possible, it is recommended to amplify the retinal signal at least 1000 times within a bandwidth of 1–500 Hz at least. However, if the selective recording of the OPs is sought, the low frequency cutoff of the recording bandwidth should be increased to 75–100 Hz while keeping the upper limit to 500 Hz. In the latter case, the amplification should be augmented to 10,000 times to compensate for the attenuation in signal that will result from restricting the recording bandwidth (see Figs. 3 and 4).
He´bert and Lachapelle
260
V.
PROTOCOLS FOR DIFFERENTIATION OF THE CONE AND ROD SYSTEMS (THE ISCEV STANDARDS)
One of the main objectives when using the ERG is the differentiation of the cone and rod systems. According to the standards of the International Society for Clinical Electrophysiology of Vision (ISCEV) [15], in order to reach this goal at least five basic responses should be acquired, two in the photopic state (lightadapted retina) and three in the scotopic state (dark-adapted retina): 1. 2. 3. 4. 5.
Photopic single flash response (standard flash) Photopic 30 Hz flicker response (standard flash) Scotopic dim flash rod response (intensity 2.5 log unit below the standard flash) Scotopic mixed rod-cone response (standard flash) Scotopic oscillatory potentials (standard flash, bandpass 75–500 Hz)
Examples of typical ERGs obtained according to the ISCEV standards are presented in Figure 3. A complete description of the ISCEV standards can be found on the Society’s website at www.iscev.org A.
Photopic Single Flash Response
It should be noted that ISCEV does not require the use of a specific level of background illumination nor a specific intensity of flash in order to generate the standard flash. Rather the standard suggests a range of photopic background (17– 34 cd⋅m⫺2) and flash intensity (1.5–3.0 cd⋅m⫺2⋅s), an approach that was recently questioned [35]. Responses illustrated in Figure 3 were evoked to a flash of 3 cd⋅m⫺2⋅s delivered against a background set at 25 cd⋅m⫺2. It is recommended to expose the retina to the rod desensitizing background at least 5 min before initiating the cone ERG evaluation. This period of adaptation should be increased to 10 min, at least, if the recording of the photopic responses immediately follows the dark adaptation period in order to avoid the light adaptation effect that was previously shown to affect the amplitude and the timing of most of the photopic ERG components [21,22]. As shown in Figure 4, with an increase in the strength of the stimulus (from bottom to top) there is a gradual increase in the amplitude of the ERG a- and bwaves. However, while the amplitude of the a-wave appears to grow steadily, that of the b-wave reaches a maximum at approximately 0.64 log cd⋅m⫺2⋅s, following which further increases in flash intensity will yield b-waves of gradually smaller voltages. This unique feature of the photopic ERG luminance-response function is known as the Photopic Hill [35,36]. Also of interest is to note that
Evaluation of Retinal Function
261
while the peak time of the a wave shortens with brighter flashes, that of the bwave increases. Finally, as mentioned above (Sec. III.C), in response to progressively brighter flashes, the oscillatory potentials increase in number as well as in amplitude.
B. Photopic Flicker Response Although in most instances the single-flash photopic ERG response will be sufficient to identify most of the cone-related retinal disorders, the ISCEV standard also recommends the use of a stimulus flickering at a rate of 30 flashes per second as the best means to isolate a cone ERG free from any possible rod-mediated contamination (Fig. 3, tracing C). Analysis of the resulting waveforms will include amplitude (from trough to peak) as well as peak time measurements (see Table 1).
C. Scotopic ERG Responses As shown in Figure 5, with time spent in dark-adaptation, there is a gradual increase in the amplitude of the resulting ERG response as a result of the slow regeneration of rhodopsin, which takes approximately 30 min (Fig. 5, tracing 1). However, the ERG response becomes much more stable once the regeneration of the rod’s photopigment is completed (Fig. 5, tracing 2). Consequently, in order to assess the rod function specifically, the intensity of the flash stimulus must be significantly attenuated in order to account for the gain in sensitivity brought by the dark-adaptation process. The intensity recommended is one that is 2.5 logunit dimmer than that used in photopic condition. Typically, the ERG evoked to this intensity of stimulation will be devoid of a recordable a wave (see Fig. 3, tracing D, and Fig. 5, tracing 1), while use of an intensity of stimulation within the photopic range will yield an ERG identified as the scotopic mixed rod-cone response (Fig. 3, tracing E, and Fig. 5, tracing 2). It should be noted that if averaging is considered, the interstimulus interval should be of at least 2 s for the dimmer stimuli and at least 5 s when the brighter flashes are used. Also, responses to the first flash should be discarded as they often show a conditioning flash effect [37,38]. The ISCEV standard also recommends that the OPs should be obtained in scotopic condition (Fig. 3, tracing F) because they are usually of larger amplitudes than those obtained in photopic condition. However, the latter recommendation should not prevent one from recording the OPs against a photopic environment as well, as there is a growing body of evidence supporting the view that photopic and scotopic OPs might originate from different retinal structures or pathways [26,39].
262
He´bert and Lachapelle
Figure 5 Tracing 1 illustrates the progressive growth in amplitude of the b-wave as a result of gradual dark-adaptation from onset of darkness (t ⫽ 0) to 30 min (t ⫽ 30). The waveform was obtained by the superposition of ERGs in response to flashes of light (intensity: ⫺3.5 log cd⋅m⫺2 ⋅ s) delivered at the rate of 1 flash each 5 s. At tracing 2, the same procedure was adopted for the following period of 30 min [e.g., from 30 min (t ⫽ 30) to 60 min (t ⫽ 60) of dark adaptation]. The intensity of the flash was also raised by 1 log-unit, thus explaining the different morphology.
As mentioned above, scotopic ERG waveforms evoked to dim flashes only include a b-wave (Fig. 6, tracing ⫺3.8 and dimmer). With increasing intensities, the amplitude of the b-wave increases rapidly until a plateau is reached. This plateau usually correlates with the appearance of an a wave suggestive of a cone contribution to the response (tracing ⫺3.0). When plotted on a graph (see Fig. 7), the relationship between the amplitude of the b-wave and the intensity of the
Evaluation of Retinal Function
263
Figure 6 Representative scotopic luminance-response function obtained from a normal human subject. As the intensity of the stimulus grows (from bottom to top), the amplitude of the b-wave increases gradually but, unlike in photopic condition (see Fig. 4), its peak time shortens. At threshold, the ERG includes only a b-wave; the a-wave appearing only in responses evoked to brighter flashes (brighter than ⫺3.0 log-unit). Again, as in photopic conditions, the a-wave peak time shortens with gradually brighter flashes. Also, several oscillatory potentials are seen riding on the ascending limb of the b-wave. Vertical arrow points to the onset of the flash. Calibration: 50 ms (horizontal) and 100 µV (vertical).
stimulus (luminance-response function curve) adopts the shape of a sigmoidal function, which can be best described with the Naka-Rushton equation [40] V/V max ⫽ I n /(I n ⫹ K n) where V represents the amplitude of the b-wave evoked from a flash of intensity I, n represents the slope of the function (which is usually close to 1 in scotopic condition) and K (or retinal sensitivity) the intensity of stimulation necessary to produce a b-wave half of maximal amplitude (Vmax). K and Vmax parameters have been used on several occasions to further characterize the scotopic function. These measurements were shown to be highly reproducible in test-retest conditions [41].
264
He´bert and Lachapelle
Figure 7 Scotopic luminance-response function curve (Naka-Rushton) obtained by plotting the b-wave amplitude data (ordinate in microvolts) against the intensity of the stimulus (abscissa in log intensity) with the use of sigmoidal curve fit software. Data obtained from both eyes (OD: right eye, OS: left eye) were plotted to illustrate high reproducibility. Vmax identifies maximal rod b-wave amplitude (as per equation) while K identifies the retinal sensitivity which is the intensity of the stimulus needed to evoke a b wave half the amplitude of Vmax.
VI. FACTORS AFFECTING THE ERG The ISCEV standard makes use of a very brief flashing stimuli (usually ⬍5 ms). Longer stimuli will add an OFF effect, which will contaminate the response and thus complicate the analysis; that is of course unless the experimenter wishes to separate the ON-ERG from the OFF-ERG [42,43]. Although the ISCEV standards are based on white light only, there are also specialized ERG protocols specifically aimed at isolating the activity of the short, medium, and long wavelength sensitive cones [44]. Use of these protocols might be helpful in both clinical and research settings. Normal aging also influences the ERG parameters. In humans, it is estimated the amplitude of the b-wave decreases by about 2.5 µV per year from age 10 to 70 [45]. A diurnal variation in the ERG b wave has also been reported [46]. It is therefore suggested to record the ERGs at about the same time of day, especially if a follow-up study is considered.
Evaluation of Retinal Function
265
Retinal pathologies, whether acquired or inherited, which have an impact on the normal processing of the visual stimulus can also alter the genesis of the ERG response. In order to appreciate the possible consequences, it is suggested to measure the amplitude and peak times of the major ERG components (a- and b-waves and OPs) and compare these parameters with normative data. Previous studies have shown that, depending on the retinal anomaly, the waves of the ERG could be affected in amplitude and/or timing. Also, examining each ERG component separately allows one to distinguish wave-specific anomalies. Previous studies have also shown that individual OPs can be specifically affected by a given disease process (or experimentally) while the remaining OPs are unaltered [25,26].
VII. INTERPRETATION OF THE ELECTRORETINOGRAM The electroretinogram is used to assess the functional integrity of the retina. With that in mind, one must remember that the retina contains two types of photoreceptors—namely, the cones and the rods—and that the activity of each can be specifically isolated with the use of specific stimulating conditions as advocated by the ISCEV standards. This is of the utmost importance, given that most retinal disorders (acquired or inherited) will often initially affect the normal functioning of one of the two photoreceptor populations. In order to determine the normalcy of the response, two parameters are considered: namely, the peak time and the amplitude. The peak time is defined as the time separating the onset of the stimulus from the maximal amplitude (or peak) of the wave under consideration. For example the peak time of the a wave refers to the time elapsed between the onset of the flash and the first major negative peak of the ERG response (or a-wave), while the peak time of the b-wave identifies the time separating the flash onset and the highest positive peak of the ERG wave (or b-wave), and so on. The same logic also applies to the individual oscillatory potentials (OPs). Similarly, the amplitude of a given wave is always measured from the preceding trough to peak except for the a-wave, which is measured from the prestimulus baseline. Thus, the amplitude of the b-wave is measured from the trough of the a-wave to the peak of the b-wave, while the amplitude of the OP3 for example is measured from the trough between OP2 and OP3 and the peak of OP3. Once the amplitudes and peak times of the different ERG waves and OPs are obtained, they are compared to normative data to ascertain normalcy. If one considers only amplitude and peak time measurements, the following diagnostic categories are possible: (1) normal amplitude and timing, (2) normal amplitude and delayed timing, (3) low amplitude and normal timing, and (4) low amplitude and delayed timing.
He´bert and Lachapelle
266
It should be noted that faster than normal responses are usually considered as a variation of normal. The above four categories can be applied to any of the ERG waves and OPs, although (as a rule) the analysis has been traditionally limited to the b-wave (photopic and scotopic). It is only recently that the a wave as well as the OPs (photopic and scotopic) have received increased attention and that their use was shown to increase the diagnostic potential of the ERG. Use of the above method of analysis of the ERG will not only allow the investigator to identify cone- and/or rod-specific retinopathies but also localize the retinal site most affected by comparing a- and b-wave measurements. Similarly, since individual OPs were previously shown to be differently affected either by the stimulus parameters or pathologies, analysis of each OP individually is also strongly suggested in order to increase the diagnostic potential of the ERG.
VIII. OTHER ELECTRODIAGNOSTIC TECHNIQUES WITH MORE SPECIFIC OR LIMITED USE A.
The Pattern-ERG (PERG)
The pattern ERG is the retinal response evoked by viewing (mono or binocularly) an alternating reversible checkerboard pattern. It is claimed to be generated at the level of the retinal ganglion cell and/or optic nerve [47]. The PERG is made of two principal components: P50 and N95 (see Fig. 8), which letters and numbers correspond to the polarity (positive or negative) and usual timing of the peak or trough of the response observed in normal individuals. P50 is sensitive to the luminance and appears to be generated in part by the same generators of the fullfield ERG, whereas N95, which is sensitive to the contrast and spatial frequency of the stimulus, is more specialized to the ganglion cells. The overall amplitude of the PERG is relatively small and range from 0.5 to 8 µV in normal individuals. This low amplitude accounts for the difficulty in achieving good recordings and explains why the technique is used routinely only by a handful of laboratories. PERG has received research attention because it can detect selectively macular and inner retina dysfunctions that go undetected with the full-field ERG. The pattern ERG is considered a good test for macular function but requires very good technical skills and experience with the technique. It should be noted that the PERG needs an averaging of about 100–400 responses in order to achieve a good signal-to-noise ratio. Fixation and refraction is also very important, which makes the technique difficult to use in infants and patients with low visual acuity (see the ISCEV standards for proper PERG recording). B.
The Multifocal ERG (mfERG)
The mfERG is a relatively new technique that allows for the assessment of localized retinal dysfunction within a 40 to 50° field that covers approximately 23%
Evaluation of Retinal Function
267
Figure 8 Representative example of a PERG obtained binocularly in a female (26 years old) using a checkerboard pattern presented on a monitor with a field size of 16° ⫻ 16° and reversals of 4.5/s and average of 800 responses (UTAS 3000 LKC Technologies Inc., Gaithersburg, MD). The largest amplitude, observed at N95, is close to 8 µV.
of the total cone population [48]. The stimulus is a pattern (typically presented on a computer screen that is composed of an array of hexagons that alternate independently between black and white state in a pseudo-random sequence described mathematically as an m-sequence [49]. At any point in time, half of the hexagons are white and half are black, which allows for a constant luminance. The most common matrix is composed of either 61, 103, and 241 hexagons that increase in size with distance from the center to compensate for the decrease of cone density with eccentricity [50]. This allows the production of focal ERG responses of similar amplitudes independently if they are recorded in the macula or para-macula. At a fast rate of stimulation of 75 Hz, each hexagon is renewed every 13.3 ms. For the viewer, the stimuli appear as little lights flickering in a random manner. The electricity generated by the retina in response to the highspeed stimulation is recorded as a continuum (strand of ERGs) using the same electrodes that are used for the full-field ERG. Focal ERGS are extracted from each location using a cross-correlation technique [51]. This is made possible as each hexagon follows the same exact sequence of black and white presentation, although each location begins at a different point in the cycle. From our experience, using a DTL electrode and a matrix of 61 hexagons, good results are achieved with a 4-min protocol. However, because the patient cannot blink during the stimulation and has to maintain a perfect fixation at a cross in the center of
268
He´bert and Lachapelle
Figure 9 Representative example of a mfERG recording obtained with a matrix of 61 hexagons, which subtended a field of about 40° (VERISTM, Science 4, ElectroDiagnostic Imaging, Inc, San Mateo, CA). In (A), a 3-D topographical map (scalar plot) of local response density. Observe the central peak (fovea), which has the highest density of photoreceptors as well as a noticeable depression caused by the blind spot seen on the far left-hand side. In (B), a trace arrays of all 61 electroretinograms are presented. Observe that the overall amplitude at each location is maintained because the hexagons are scaled with eccentricity (larger in the periphery). Having similar amplitudes allows the detection of area of abnormalities just by looking at the trace arrays. Responses can also be averaged into rings, quadrants, or hemiretinal areas, to name a few (not presented). The total recording was 3 min 38 s, achieved using 16 slightly overlapping segments 15 s long (signal amplification: 100,000; bandwidth: 10–100 Hz). The mfERG recording was obtained from the left eye of a male (35 years old) dilated with Tropicamide 0.8% using a DTL fiber electrode.
the screen, the 4-min protocol is split into 16 segments of 15 s duration (30 s if using a contact lens). Depending on how the patient handles that task, the 4-min protocol can be completed in 5 to 15 min. In contrast to the full-field ERG, which necessitates minimal cooperation from the patients, cooperation is crucial to achieve the mfERG. Therefore, it would be difficult to perform this type of recording in infants. As with the pattern ERG, refraction is also important. Recently, ISCEV introduced the first guidelines for the mfERG. In these guidelines it is recommended that when the hexagons are in the white state they should produce a luminance of 200 cd/m2, whereas when they are in the black state they should be close to a luminance of 0 cd/m2. The background should be
Evaluation of Retinal Function
269
set at 100 cd/m2, yielding to an overall mean luminance of 100 cd/m2. Amplification is usually 100,000 times and the bandwidth set at 1–300 Hz or 10–300 Hz. Typical views of the recording are presented in Figure 9. Looking at Figure 9 (right-hand side), it is clear that mfERG response shares some resemblance with the full-field cone ERG as it is composed of a negative deflection (N1) followed by a positive deflection (P1). However, in the mfERG, the peak of the response occurs at an earlier time and it is devoid of the fast oscillation observed in the full-field ERG, namely, the oscillatory potentials. This difference in the waveforms is likely due to the stimulation paradigm that is employed and how the data analysis is performed. In contrast to the full-field ERG in which the final response is composed of consecutive responses that are simply averaged together, in the mfERG the final response is composed of additions and subtractions of preselected responses that depend on the analysis selected. For instance, with the first-order analysis (most commonly used), only white hexagons that were followed by a black hexagon are averaged together at each location. With the second-order analysis, we are measuring the impact of successive flashes on the mfERG response [51]. Similar to the full-field ERG, the generators of the firstorder analysis of the mfERG appear to be the photoreceptors and bipolar cells. Higher order analyses that take into account the effect of consecutive flashes are believed to be indicators of ganglions cell activity as well as optic nerve function [51]. Studies are still needed to confirm the exact components and generators of the mfERG. The research interest in mfERG is that it provides a spatial resolution of the macula that cannot be achieved with the full-field ERG. Of note, animal research can be performed with mfERG, as shown by reports on monkeys, cats, rats, and pigs.
IX. CONCLUDING REMARKS The purpose of this chapter was to provide the reader with standardized means to assess the retinal function in clinical or experimental conditions. Assessing the functional status of the retina using noninvasive procedures is relatively simple. The methods described can yield a wealth of extemely valuable information that can be used to make a diagnosis whether the retinal disorder was acquired through experimentation or otherwise, or inherited. However, in order to achieve the highest level of diagnostic accuracy possible, users are urged to create their own sets of normative data for all the species and age ranges under analysis.
ACKNOWLEDGMENTS This work was supported by grants-in-aid from McGill University-Montreal Children’s Hospital Research Institute, the Canadian Institutes of Health Research
He´bert and Lachapelle
270
(Grant MT-12153 and MT-13383), the FCAR-GRENE and the Vision Network of the FRSQ. Thanks are due to Olga Dembinska, Julie Racine, and Marianne Rufiange for providing some of the illustrations.
REFERENCES 1. 2.
3.
4.
5.
6.
7. 8.
9.
10. 11.
12.
13. 14. 15.
Rosolen S, Rigaudie`re F, Lachapelle P. A practical method to obtain reproducible binocular electroretinograms in dogs. Doc Ophthalmol 2002; 105:83–92. Lachapelle P, Benoit J, Dembinska O, Rojas LM, Almazan G, Chemtob S. Persistent functional and structural retinal anomalies in newborn rats exposed to hyperoxia. Can J Physiol Pharmacol 1999; 77:48–55. Cayouette M, Behn D, Sendtner M, Lachapelle P, Gravel C. Functional rescue of photoreceptors in the retinal degeneration slow (rds) mouse by CNTF. J Neurosci 1998; 18:9282–9293. Rosolen SG, Rigaudie`re F, Saint-Macary G, Lachapelle P. Recording the photopic electroretinogram from conscious Yucatan micropigs. Doc Ophthalmol 2000; 98: 197–205. Rojas LM, McNeil R, Cabana T, Lachapelle P. Diurnal and nocturnal visual capabilities in shorebirds as a function of their feeding strategies. Brain Behav Evol 1999; 53:29–43. Gorfinkel J, Lachapelle P. Maturation of the photopic b-wave and oscillatory potentials of the electroretinogram in the neonatal rabbit. Can J Ophthalmol 1990; 25: 138–144. Blain L, Lachapelle P, Molotchnikoff S. The effect of acute trichloroethylene exposure on electroretinogram components. J Neurotoxicol Teratol 1990; 12:633–636. Blain L, Lachapelle P. Comparative effect of chronic trichloroethylene exposure on the electroretinogram components and oscillatory potentials. Neurotoxicol 1994; 15: 627–632. Dembinska O, Rojas LM, Varma DR, Chemtob S, Lachapelle P. Graded contribution of retinal maturation to the development of oxygen-induced retinopathy in rats. Invest Ophthalmol Vis Sci 2001; 42:1111–1118. Reynaud X, Hansen RM, Fulton AB. Effect of prior oxygen exposure on the electroretinographic responses of infant rats. Invest Ophthalmol Vis Sci 1995; 36:2071–2079. Behn D, Potter MJ. Sildenafil-mediated reduction in retinal function in heterozygous mice lacking the γ-subunit of phosphodiesterase. Invest Ophthalmol Vis Sci 2001; 42:523–527. Jaissle GB, May CA, Reinhard J, Kohler K, Fauser S, Lu¨tjen-Drecoll E, Zrenner E, Seeliger MW. Evaluation of the rhodposin knockout mouse as a model of pure cone function. Invest Ophthalmol Vis Sci 2001; 42:506–513. Armington JC. The Electroretinogram. New York: Academic Press, 1974. Granit R. The components of the retinal action potential in mammals and their relation to the discharge in the optic nerve. J Physiol 1933; 77:207–239. Marmor MF, Zrenner E. Standard for clinical electroretinography (1999 update). Doc Ophthalmol 1999; 97:143–156.
Evaluation of Retinal Function 16.
17.
18. 19. 20. 21. 22.
23. 24. 25. 26.
27. 28. 29. 30. 31. 32. 33. 34. 35.
271
Hood DC, Birch DG. Assessing abnormal rod photoreceptor activity with the awave of the electroretinogram: applications and methods. Doc Ophthalmol 1997; 92:253–267. Sieving PA, Murayama K, Naarendorp F. Push-pull model of the primate electroretinogram : a role for hyperpolarizing neurons in shaping the b-wave. Vis Neurosci 1994; 11:519–532. Wachtmeister LL. Oscillatory potentials in the retina: what do they reveal. Prog Retinal Eye Res 1998; 17:485–521. Lachapelle P. Evidence for an intensity-coding oscillatory potential in the human electroretinogram. Vision Res 1991; 31:767–774. Lachapelle P. The effect of slow flicker on the human photopic oscillatory potentials. Vision Res 1991; 31:1851–1857. Lachapelle P. Analysis of the photopic electroretinogram recorded before and after dark-adaptation. Can J Ophthalmol 1987; 22:354–361. Peachey NS, Alexander KR, Fishman GA, Derlacki DJ. Properties of the human cone system electroretinogram during light adaptation. Appl Optics 1989; 28:1145– 1150. Lachapelle P, Blain L, Quigley MG, Polomeno RC, Molotchnikoff S. The effect of diphenylhydantoin on the electroretinogram. Doc Ophthalmol 1990; 73:359–368. Guite´ P, Lachapelle P. The effect of 2-amino-4-phosphonobutyric acid on the oscillatory potentials of the electroretinogram. Doc Ophthalmol 1990; 75:125–133. Lachapelle P. The human suprathreshold photopic oscillatory potentials: Method of analysis and clinical illustration. Doc Ophthalmol 1994; 88:1–25. Lachapelle P, Benoit J, Rousseau S, McKerral M, Polomeno RC, Little JM, Koenekoop R. Evidence supportive of a functional discrimination between photopic oscillatory potentials as revealed with cone and rod mediated retinopathies. Doc Ophthalmol 1998; 95:35–54. Rousseau S, McKerral M, Lachapelle P. The i-wave: bridging flash and pattern electroretinography. EEG Clin Neurophysiol (suppl) 1996; 46:167–173. Lachapelle P. A possible contribution of the optic nerve to the photopic oscillatory potentials. Clin Vision Sci 1990; 5:421–426. Gouras P, Mackay CJ. Detecting early postchiasmatic visually evoked responses. Clin Vision Sci 1988; 3:119–124. Dawson WW, Trick GL, Litzkow CA. Improved electrode for electroretinography. Invest Ophthalmol 1979; 18:988–991. He´bert M, Vaegan, Lachapelle P. Reproducibility of ERG responses obtained with the DTL electrode. Vision Res 1999; 39:1069–1070. Gouras P. Electroretinography: some basic principles. Invest Ophthalmol 1970; 9: 557–569. Lachapelle P, Blain L. A new speculum electrode for electroretinography. J Neurosci Methods. 1990; 32:245–249. Lachapelle P, Benoit J, Little JM, Lachapelle B. Recording the oscillatory potentials with the DTL electrode. Doc Ophthalmol 1993; 83:119–130. Lachapelle P, Rufiange M, Dembinska O. A physiological basis for definition of the ISCEV ERG standard flash (SF) based on the photopic hill. Doc Ophthalmol 2001; 102:157–162.
272
He´bert and Lachapelle
36.
Wali N, Leguire LE. The photopic hill: a new phenomenon of the light adapted electroretinogram. Doc Ophthalmol 1992; 80:335–342. Lachapelle P, Benoit J, Blain L, Guite´ P, Roy MS. The oscillatory potentials in response to stimuli of photopic intensities delivered in dark-adaptation: an explanation for the conditioning flash effect. Vision Res 1990; 30:503–513. Peachey NS, Alexander KR, Fishman GA. Rod and cone system contributions to oscillatory potentials: an explanation for the conditioning flash effect. Vision Res 1987; 27:859–866. Rousseau S, Lachapelle P. Origin of the oscillatory potentials recorded at the onset of dark adaptation. Doc Ophthalmol 1999; 99:135–150. Naka KI, Rushton WAH. S-potentials from colour units in the retina of fish (cyprinidae). J Physiol 1966; 185:536–555. He´bert M, Lachapelle P, Dumont M. Reproducibility of Naka-Rushton parameters derived from electroretinograms recorded with DTL electrodes. Doc Ophthalmol 1996; 91:333–342. Nagata M. Studies on the photopic ERG of the human retina. Jpn J Ophthalmol 1963; 7:96–124. Seiple W, Holopigian K. The OFF response of the human electroretinogram does not contribute to the brief flash b-wave. Vis Neurosci 1994; 11:667–673. Gouras P, Mackay CJ, Ivert L, Mittl RN, Neuwirth J, Eggers H. Computer-assisted spectral electroretinography in vitrectomy patients. Ophthalmology 1985; 92:83– 90. Weleber RG. The effect of age on the human cone and rod ganzfeld electroretinogram. Invest Ophthalmol Vis Sci 1981; 20:392–399. Birch DG, Berson EL, Sanberg MA. Diurnal rhythm in the human rod ERG. Invest Ophthalmol Vis Sci 1984; 25:236–238. Maffei L, Fiorentini A, Bisti S, Holla¨nder H. Pattern ERG in the monkey after section of the optic nerve. Exp Brain Res 1985; 59:423–425. Hood DC. Assessing retinal function with the multifocal technique. Prog Retinal Eye Res 2000; 19:607–46. Sutter EE, Tran D. The field topography of ERG component in man-I. The photopic luminance response. Vision Res 1992; 32:433–466. Curcio CA, Sloan Jr. KR, Kalina RE, Hendrickson AE. Human photoreceptor topography. J Comp Neurol 1990; 292:497–523. Sutter E. The interpretation of multifocal binary kernels. Doc Ophthalmol 2000; 100:49–75.
37.
38.
39. 40. 41.
42. 43. 44.
45. 46. 47. 48. 49. 50. 51.
16 Evaluation of Visual Outcome Pamela A. Sample University of California, San Diego La Jolla, California, U.S.A.
I.
INTRODUCTION
Until recently, clinical trials for glaucoma have focused on medical or surgical techniques for lowering intraocular pressure (IOP). The outcome measure of choice in many of these trials was a criterion reduction in IOP. The effectiveness of a given intervention could be measured in a matter of a few months at most. Very recently, potential new neuroprotection therapies designed to protect undamaged but at-risk axons and ganglion cells, or to rescue those that are marginally damaged, have been put forward [1]. Neuroprotection studies, however, may take much longer to determine whether these treatments prevent slowly progressing glaucomatous damage. To determine whether ganglion cells have survived and maintained function due to neuroprotective therapy, outcome measures of optic nerve appearance and visual function will be necessary. Lowering intraocular pressure cannot answer this question. Visual function testing is the only thing that can assess effectiveness in the living human eye. Even if measures of optic nerve structure show no progression, it does not guarantee that function is spared or has recovered. There are two approaches to determining the best tests to measure outcome in neuroprotection trials. We need to know what are the best tests available to us right now. This review will present several candidate tests currently available. We also need to be aware that each of these tests has disadvantages that reduce 273
274 Table 1
Sample Advantages and Disadvantages of Test Procedures
Availability Standardized test Optimized procedures Normative database Statistical analysis package Monitors fixation Reliability indices Self-calibrating equipment Test time under 6 min Patients prefer it Validated by studies Ganglion cell specific Isolation known Longitudinal studies Early detection Identifies progression Over/underclassifies OHT Correlates with optic nerve disease Works for a variety of eye disease Variability Lens, pupil, refraction/affect
SAP
SWAP
FDT
HPRP
1 yes yes yes* 1* yes yes yes yes/SITA 3 yes no none yes no 3 under 3 3 3 2
2 yes yes yes yes yes yes yes no 4 yes yes* yes* yes yes 1 no 1 1 4 4
3 yes ? yes yes no yes yes yes 2 yes yes no no yes ? over ? ? 2 1
4 no ? yes 4 no yes no yes 1 yes yes no yes no 2 ? 2 2 1 3
* ⫽ obviously best test 1–4: Ranking best test to worst test ? ⫽ not yet known
their ability to detect the small changes in visual outcome that are expected in these trials. I will review the advantages and disadvantages of each test (summarized in Table 1) and highlight areas of research needed to correct the disadvantages.
II. CANDIDATE VISUAL FUNCTION TESTS Measures of visual function include electrophysiology, visual acuity, and various other psychophysical measures targeted at peripheral retina. Electrophysiological measures, although promising [2–4], still need modification and continued study to determine if they will be useful for detecting the subtle changes over time required for assessing neuroprotection. In addition, the commercially available
Evaluation of Visual Outcome
275
testing equipment (VERIS, Tomey, Nagoya, Japan) is not currently present in many of the centers and offices participating in the clinical trials. Visual acuity may be a useful outcome measure for studies of macular degeneration or optic neuropathies that affect foveal function, but it is not the measure of choice for primary open angle glaucoma because the central macula is spared until the later stages of the disease. Currently, perimetry remains the measure of choice for studies of glaucoma progression. This review will discuss the rationale for different perimetric techniques and the potential for following change in clinical trials of neuroprotection. The focus will be on commercially available tests with standardized procedures. The advantages and disadvantages of each are summarized in Table 1. Ganglion cells and their axons, which form the optic nerve, are the primary sites of damage due to glaucoma. There are roughly 1.25 million ganglion cells in the human retina with a minimum of 22 subtypes [5]. The three major ganglion cell types—magnocellular, parvocellular, and small-bistratified—compose up to 90% of all retinal ganglion cells in macaque and human eyes. Each of the three ganglion cell types has distinctive features that have been used in devising tests to detect and follow loss of vision due to glaucoma (Fig. 1). In addition, the morphological differences among these three cell types may indicate that one type is more amenable to neuroprotection than another, and appropriate measures of each subtype could be crucial to determining if this is the case.
A. Psychophysical Tests of Peripheral Visual Field 1. Standard Achromatic Automated Perimetry Standard achromatic automated perimetry (SAP) is the most commonly used version of perimetry for clinical diagnosis as well as in clinical trials. SAP utilizes a small (0.47°) 200 ms flash of white light as the target presented on a dim background (10 cd/m2 or 31.5 asb). The target is randomly presented to several locations (54 in program 24-2 and 76 in program 30-2) using a Humphrey Visual Field Analyzer II (Humphrey Instruments, Dublin, CA). Similar test patterns are available on Octopus perimeters (Interzeag AG, Zurich, Switzerland). Results in this section will focus on those obtained with the Humphrey unit. While a virtual congruence between visual field damage and loss of optic nerve fibers has been assumed, significant damage to the ganglion cells may occur before the disease can be recognized by visual field abnormalities [6,7]. By the time visual field loss is initially detected by Goldmann kinetic perimetry, 40% to 50% of retinal ganglion cells may have suffered irreversible damage [6]. Before SAP first demonstrates abnormalities, 20 to 40% of ganglion cells may be lost [8,9]. These findings spurred several studies designed to find more sensitive measures of vision loss, measures that target specific subtypes of retinal ganglion cells.
276
Sample
Figure 1 An oversimplified schematic depicting the separation of visual pathways serving different visual functions from cones through to the LGN. Cones are depicted by triangles at the top. Their inputs are combined at the retinal ganglion cell level to form the blue-yellow opponent on ganglion cell (small bistratified), the redgreen opponent ganglion cell (midget), which also handles high resolution tasks, and the achromatic black-white opponent cells (parasol). The axons from each type of ganglion cell project to a different region in the lateral geniculate nucleus (LGN) and that is how they get their other names, koniocellular ganglion cell, parvocellular ganglion cells, and magnocellular ganglion cells. The visual functions preferred by each ganglion cell are listed at the bottom of each pathway. Each pathway can handle other visual functions, but not as efficiently as it does those listed beneath it.
It can be argued that SAP is nonspecific for ganglion cell type, and detection of a white flash of light can be mediated through many types of retinal ganglion cells. This allows a redundancy at a given retinal location. Meaning if one type of ganglion cell is damaged, but the others are not, the signal can still be detected. Each of the tests described below, on the other hand, attempts to evaluate one type of visual function as a surrogate for isolating a specific ganglion cell subtype (see Fig. 1). If successful, this means that when the subtype of ganglion cell under test is damaged, the other subtypes can not easily detect the target. This has been referred to as reduced redundancy [10,11].
Evaluation of Visual Outcome
277
2. Short-Wavelength Automated Perimetry Short-wavelength automated perimetry (SWAP) is a modification of SAP available on the Humphrey visual field analyzer and the Octopus perimeter utilizing the same test programs as SAP [12]. SWAP utilizes a 440 nm, narrow band, 1.8° target at 200 ms duration on a bright 100 cd/m2 yellow background to selectively test the short-wavelength sensitive pathway. The test provides a dynamic range of approximately 35 dB and 15 dB of isolation before the next most sensitive mechanism can detect the target, most likely the middle-wavelength sensitive pathway [12,13]. At the ganglion cell level, the patient’s response to this test is mediated by the small bistratified blue-yellow ganglion cells. Their dendritic field size, soma size, and axon size are slightly smaller than that of the magnocellular cells in humans [14]. Small bistratified ganglion cells are fewer in number and compose 6–10% of ganglion cells in peripheral retina [14]. The small bistratified cells receive their input from the blue-cone bipolar cells [15]. These cells respond in a sustained manner and have blue-yellow color opponency. They prefer shortwavelength stimuli, which are stressed in SWAP by adapting rods and the other two cone inputs using the bright yellow background. The key word here is “prefer.” For example, ganglion cells that prefer high temporal frequencies can respond to lower temporal frequencies under certain stimulus conditions, but their maximal response will be in the higher temporal frequency range and they will be more sensitive than the other ganglion cell types to these stimuli. SWAP has more than 13 years of longitudinal evaluation and has been shown by several independent studies to be a more effective test than SAP for early detection of glaucoma-related field loss [16–18]. SWAP also identifies progression 1–3 years prior to detection by standard visual fields [19–21], and works well in advanced cases not complicated by the presence of cataracts [22]. Although SWAP has higher test-retest variability than SAP, and more than is desirable for long-term follow-up for progression of glaucomatous vision loss [23, 24], it has consistently been shown superior to SAP for identifying progression [20,21]. 3. High-Pass Resolution Perimetry It was originally thought that SWAP was testing blue-yellow ganglion cells that pass through the parvocellular pathways of vision. Since SWAP was developed, it has been determined that the blue-yellow ganglion cells responsible for processing the stimuli of SWAP are most likely not parvocellular, but are the small bistratified ganglion cells whose axons project to the interlaminar layers of the lateral geniculate nucleus of the thalamus [14,25–27]. Parvocellular ganglion cells are the most numerous, making up about 70% of the total. These cells are
278
Sample
also referred to as midget ganglion cells. These cells are distributed throughout the retina, and they have the smallest cell bodies, axons, and dendritic field sizes of the three types [14]. These cells handle acuity and resolution tasks. They prefer stimuli with low temporal frequencies and high spatial frequencies, along with color [28]. If we wish to include a test more likely to be detected by the parvocellular pathway, we can look to High-Pass Resolution Perimetry. This test is a resolution task. High-pass resolution visual fields were developed by Lars Frisen, M.D., and are tested using the Ophthimus High-Pass Resolution Perimeter (HighTech Vision, Malmo¨, Sweden) [29]. The test presents spatially filtered rings across 50 test locations in a 30° visual field. Fourteen different ring step sizes are used, with the smallest subtending 0.8° visual angle and each successive larger ring size spaced 0.1 log units apart. Thresholds are designated as the smallest ring size that can be resolved by the patient. The ring targets consist of a dark border (15 cd/m2) surrounding a light core (25 cd/m2) and are designed such that the space-average luminance across the ring target is equiluminant with the background of the display (20 cd/m2). In this manner, resolution rather than detection thresholds are measured and the target is both seen and resolved instantaneously or not seen at all. High-Pass Resolution Perimetry was found to be comparable to standard fields for detecting vision loss [30] and superior for identifying change over time [31]. It is a very patient-friendly test, taking about 5 min and giving feedback for correct responses. The major drawback to its acceptance is the lack of standardization across test units. 4. Frequency-Doubling Technology Perimetry Frequency-doubling technology perimetry (FDT) [32,33] is based on the frequency-doubling illusion [34,35], which occurs when viewing a counterphased grating with a low spatial frequency and a high temporal rate. Above threshold, the percept is double the spatial frequency of the actual physical grating [35]. This illusion has been attributed to a subset of the magnocellular ganglion cells, which are nonlinear in their response properties [36]. Magnocellular ganglion cells are fewer in number and compose about 8– 10% of the ganglion cell population. These cells are also referred to as parasol ganglion cells. These cells are distributed across the retina, and compared to the parvocellular cells they have larger cell bodies, axons, and dendritic field sizes with consequently larger receptive field center sizes. The magnocellular cells’ axons project to the magnocellular layers of the lateral geniculate nucleus [37]. They respond in a transient fashion to visual stimulation, and their axons have a relatively faster conduction velocity compared to the other two cell types [38– 41]. These cells prefer stimuli with high temporal frequencies, low spatial fre-
Evaluation of Visual Outcome
279
quencies, and motion. There is some debate about whether FDT at threshold is measuring a small subset of nonlinear magnocellular cells (estimated at about 3% of the ganglion cells), or if the target is more likely detectable due to its flicker component [42–44] by the full complement of magnocellular cells (still only about 10% of the population). At threshold, the percept is not always of a grating, either perceptually doubled or veridical, but sometimes is described as a “shimmering” or “flickering” [45,46]. Either way, early evidence has shown the test is sensitive to early glaucomatous defects and correlates well with SAP for mean defect [32,47–51]. Frequency-doubling perimetry is measured with a new instrument, the Humphrey FDT Visual Field Instrument using Welch-Allyn Frequency doubling technology (Skaneateles Falls, NY). The targets consist of a 0.25 cycle per degree sinusoidal grating that undergoes 25 Hz counterphase flicker. The test uses a modified binary search staircase threshold procedure to measure the contrast needed for detection of the stimulus. Each grating target is a square subtending about 10° in diameter. Targets are presented in one of 17 test areas located within the central 20° radius of visual field (program C-20). With a shift in fixation point location, the range can be extended to 30° and two additional locations in the nasal step area (program N-30). FDT has some advantages over SAP and SWAP. The test time is about one-half of the time required for a full threshold 24-2 field, primarily due to the smaller number of test locations used. The results are less affected by blur, pupil size differences if always greater than 2 mm diameter, or bifocal correction [52], and it has lower test-retest variability than SAP [53]. It is similar to SAP and SWAP in that statistical analysis packages can be developed to give global indices, such as MD and PSD, and pattern deviation probability plots can be derived. The major drawback to FDT for clinical trials that look for change over time is that it is too new to have longitudinal data. We do not yet know if the small number of test locations will provide sufficient resolution for following change over time. Johnson and colleagues are currently working on a modification of the test to increase the number of test targets [54]. While this may improve the test ability to detect change over time, it will also increase the test time. The advantages and disadvantages of each test are summarized in Table 1. If a test is obviously the best, an asterisk denotes it. For example, SAP has the most extensive normative database and FDT is least affected by changes in pupil size. If the tests could be ranked, they are ranked from best (1) to worst (4). For example, SWAP has been found more useful than SAP for a variety of eye diseases but it is most affected by lens opacities. If we do not yet know something it is indicated by a ?. For example, we do not know if FDT will be good for following change over time because it is too new for extensive longitudinal studies. These advantages and disadvantages have had some influence on whether
280
Sample
these tests have been selected for use in current clinical trials. To my knowledge, there is no study currently using HPRP. Although this test has numerous advantages, it has not received widespread acceptance due to the lack of standardized test equipment and no self-calibrating feature. Results obtained on the same eye may vary from one unit to another. The other three tests are all involved in clinical trials at present. SWAP is not an outcome measure but is an ancillary test in the NEI-sponsored Ocular Hypertension Treatment Study (OHTS) [55] and FDT is a secondary outcome measure in an Allergan-sponsored clinical trial of the drug Memantine. This study should help us determine the effectiveness of FDT in longitudinal studies assessing change in visual function. In nearly all clinical trials involving perimetry, SAP is the method of choice to date. SAP is currently the standard of care for visual field testing. The availability, standardization, and sophisticated statistical analyses are useful in clinical trials. However, in clinical trials that must identify small changes in visual performance, it may be less than ideal. SWAP identifies change 1 to 3 years sooner than SAP [20,21] and the results indicating change are more likely to be repeatable [56,57]. However, the increased variability and full threshold only strategy of SWAP have been detrimental to its acceptance. SITA (see below) for SWAP, when available, should go a long way toward alleviating these problems as it has reduced both test time and variability for SAP. 5. Swedish Interactive Thresholding Algorithm Swedish interactive thresholding algorithm (SITA) is a new way of obtaining threshold for SAP [58,59]. It has led to an increased interest in SAP despite evidence that the visual function specific subtests, SWAP and FDT, are much more sensitive for detection [51] and SWAP and HPRP are better for following progression [20,21,31]. This new enthusiasm for standard fields is based on two attributes of SITA. The test time for a visual field is reduced in half relative to full threshold SAP [59–61] and the test-retest variability may also be reduced, which should make the results more reliable for measuring change over time [62–64]. Because SITA is a change in threshold procedure, which utilizes information collected during the test to better determine the intensity of the next flash at a given location [65,66], it easily could be applied to other forms of perimetry, such as SWAP. Studies comparing SITA with the original full threshold algorithm indicate that SITA will show approximately 1 dB better thresholds than standard full threshold testing, but with a statistically deeper defect on the pattern and total deviation plots in glaucoma patients [60,62,66,67]. The smaller intersubject variance and greater reproducibility in patient eyes may mean shallower defects are needed in SITA fields for statistical significance to be reached than for the standard full threshold algorithm, but this has not been tested [62,63]. For trials that have already begun testing with full threshold strategy, it is not advisable to
Evaluation of Visual Outcome
281
switch to SITA. Clinically, when switching from standard full threshold to the SITA strategy to follow patients, it is necessary to obtain new baseline fields and to make comparisons between the two types of algorithms based on the total and pattern deviation probability plots rather than on absolute threshold values [64]. Clinical trials often initiate treatment after obtaining baselines, so there is no opportunity to reassess baselines with SITA. However, trials that were initiated after the development of SITA, such as the Neuroprotection in Ischemic Optic Neuropathy Trial sponsored by Allergan, have selected SITA for use with SAP as a secondary efficacy measure.
III. THE DEFINITION OF CHANGE IN CLINICAL TRIALS Because it is often necessary in clinical trials to determine if progression is occurring before a series of five to seven fields can be obtained, linear regression has not been the method of choice. Instead, the statistical methods for identifying field progression relative to two baseline visits provided by the GCP and global indices of Statpac II have been incorporated into several large clinical trials: (1) Normal Tension Glaucoma Study (NTG); (2) Early Manifest Glaucoma Trial (EMGT); (3) Advanced Glaucoma Intervention Study (AGIS); (4) Collaborative Initial Glaucoma Treatment Study (CIGTS); and (5) Ocular Hypertension Treatment Study (OHTS). A. Normal Tension Glaucoma Study The Normal Tension Glaucoma Study was developed to assess the effect of lowering intraocular pressure on the progression rate of normal tension glaucoma [68]. To be eligible for this study, patient eyes had to have glaucomatous excavation of the optic disc and a field defect, standard achromatic perimetry, consisting of a cluster of three non-edge points depressed by 5 dB, with one of the points also depressed by 10 dB. This defect had to be confirmed by two of three baseline fields performed within a 4-week window. Patients with a history of IOP greater than 24 mmHg were excluded. Progression was suspected when (1) at least two contiguous points within or adjacent to a baseline defect showed a reduction is sensitivity from baseline of at least 10 dB, or three times the average baseline short-term fluctuation for that patient, whichever is greater; (2) the sensitivity of each suspected point is outside the range of values observed during baseline testing; or (3) when a defect occurs in a previously normal part of the field. To reach a definitive decision of progression, four confirmatory tests were required. This large number of confirmatory fields was a consequence of looking for smaller field changes, which were
282
Sample
reliable indicators of glaucomatous visual field loss and not just indicative of physiologic long-term fluctuation [68]. B.
The Early Manifest Glaucoma Trial
The Early Manifest Glaucoma Trial (EMGT) was developed to assess the effectiveness of reducing intraocular pressure in early, previously untreated open-angle glaucoma. Because EMGT uses field progression as a study endpoint, a progression algorithm was developed [69]. As mentioned previously, the GCP from Statpac was modified so scoring is based on the pattern deviation probability map. For the EMGT, there is an initial screening, two pre-intervention field tests, and two baseline visits. The two baseline visits must have a GHT “outside normal limits” due to the same sectors or a “borderline” GHT on two consecutive field exams with obvious localized change to the optic disc [69]. Progression requires three or more points flagged by the pattern deviation version of the GCP analysis. Only if these same points are confirmed on two subsequent visual fields is progression verified. C.
Advanced Glaucoma Intervention Study
The Advanced Glaucoma Intervention Study (AGIS) algorithm was developed to determine eligibility for the AGIS study and to evaluate disease progression in patients with more advanced glaucomatous visual fields [70]. The AGIS scoring system is based on the concepts that (1) multiple defects can occur in the upper, lower, and nasal hemifields; (2) a defect requires two or more adjacent defective points; (3) the severity of depression must be greater than changes due to variability; and (4) the defect must be caused by glaucoma. The AGIS score is calculated by totaling the number of adjacent depressed test points in the upper, lower, and nasal hemifields compared with age-matched standardized normal eyes in the total deviation printout of Statpac II. The score becomes larger with increases in the number of depressed test sites and with increasing depth of defect. The final AGIS score ranges from 0 to 20. Two pre-intervention field tests are conducted less than 60 days apart to determine subject eligibility. An AGIS score between 1 and 16 and a reliability score less than 3 are required on the first field for subject inclusion. The second field is used as the baseline for subsequent tests. Progression is quantified as an increase in score from baseline reference by four or more points on two consecutive reliable fields. D.
Collaborative Initial Glaucoma Treatment Study
The Collaborative Initial Glaucoma Treatment Study (CIGTS) uses a scoring system modified from AGIS [71]. In brief, scoring is based on (1) the total devia-
Evaluation of Visual Outcome
283
tion probability plot, which adjusts the total deviation values at each point relative to the most normal region in the visual field; (2) each abnormal test location must be accompanied by at least two adjacent abnormal points; and (3) each abnormal point is given a score from one to four based on the probability level (5% to 0.5%) of the three contiguous depressed points. The value for each of the 52 locations within the visual field is combined to get a maximum possible score of 208. The total score is then divided by a conversion factor (10.4) to get a final score ranging from 0 to 20. Two pre-intervention field tests must be reliable (reliability score ⬍ 4); the GHT must read “outside normal limits”; there must be at least three contiguous points on the total deviation plot with a p ⬍ 0.02; and if the points are in the nasal field, they cannot cross the horizontal midline. The two pre-intervention field scores are averaged to create a baseline CIGTS score. If the two baseline fields are more than 7 points apart, then a third field is conducted and the three are used to compute a baseline CIGTS score. Progression is quantified as an increase in score from baseline reference by three or more points on two consecutive reliable fields. E.
The Ocular Hypertension Treatment Study
The previously mentioned studies all involve assessment of change in an already abnormal visual field. The Ocular Hypertension Treatment Study (OHTS) [55] looks for change from a designation of normal visual field to abnormal visual field using SAP 30-2 full threshold visual fields. Fields are identified as abnormal if they have a glaucoma hemifield test results of “outside normal limits” or a corrected pattern standard deviation at the 5% probability or worse with a cluster of abnormal points consistent with glaucoma. Two confirming fields are required to call the fields glaucomatous. What these descriptions of progression or change make clear is that there is no agreed-upon method for identifying progression and each clinical trial has developed methods specific to the study population involved. A few studies have compared some of the methods [72,73]. Regardless of the algorithm employed, it is still difficult to identify and confirm change without sufficient follow-up.
IV. FOLLOWING CHANGE MORE EFFECTIVELY In addition to the obvious need to reduce the variability inherent in psychophysical tests of visual function and to confirm changes seen, there are other strategies that may assist us in identifying change more reliably. Each ganglion cell axon crosses the retina and enters the optic nerve. These fibers travel in bundles in a specific pattern. Several studies have correlated the location of visual field defect to location of optic nerve damage with good results [74–79]. We have found these
284
Sample
correlations hold for focal defects found on SAP, SWAP, and non-commercially available motion automated perimetry (MAP) [46,80,81]. In addition, there is an important and direct relationship between these measures of visual function and location of damage. In a study comparing SAP, SWAP, FDT, and MAP, we found that when a given individual had vision loss on more than one test, the same area of the visual field was affected [51]. Finally, in early glaucoma there is little evidence for secondary effects of neural degeneration as demonstrated by a lack of crossover of defect at the horizontal midline [82]. This does not rule out the presence of secondary degeneration but may be indicative of the limitations of our current testing strategies. Since we now have information about different visual functions, location of visual field loss and the relationship to optic nerve damage, how can we use this information to improve tests for monitoring the effectiveness of neuroprotection therapies in clinical trials? Here we are not only interested in detection of a defect but also in change over time. Will those eyes receiving the neuroprotective agent show less progression of glaucoma than those that do not receive neuroprotection? For this, we also must understand something about how visual fields progress. To determine typical patterns of visual field progression and their relationship to the 24-2 field grid, we prospectively followed, in a multicenter study, 115 glaucoma patients, who were experienced with fields, having two abnormal baseline visual fields, abnormal optic nerves, and repeated testing for fields that progressed [83]. Progression was categorized as (1) deepening of an existing scotoma; (2) expansion of an existing scotoma; or (3) a new scotoma. Three methods for classifying progression were used: (1) a clinically determined method; (2) the glaucoma change probability (GCP) based on total deviation (TD); and (3) the GCP based on pattern deviation (PD). Regardless of the classification method used, the glaucomatous visual fields progressed in the area of the visual field where baseline testing showed an existing scotoma. Because an individual’s location of defect overlaps on different visual function tests and progression occurs in this area, we can maximize finding repeatable change by concentrating testing on the defective areas determined at study entry.
V.
CONCLUSIONS
Tests of visual function and measures of optic nerve structure that take into account the location of damage at baseline for each individual and incorporate this information in their designs for following change should be most successful for determining the effectiveness of neuroprotection. Since there is no gold standard for progression, it is still necessary for each clinical trial to identify an algorithm that best fits the needs of the study population under test. It would be prudent to incorporate a variety of visual function tests in these studies, because
Evaluation of Visual Outcome
285
we do not know if one type of ganglion cell may be more susceptible to a given neuroprotection therapy than another. It will be necessary to continue to improve our methods for identifying change in visual fields. This includes improving visual field techniques and improved analysis of results over time. As it stands now, more time for follow-up will be necessary to determine the effectiveness of a given neuroprotective agent than was needed in past studies to show the effectiveness of IOP-lowering agents. This is because glaucoma progresses slowly and possible change requires confirmation on more than one subsequent visit.
ACKNOWLEDGMENTS This work is supported by NEI EY08208 and a Lew R. Wasserman award from Research to Prevent Blindness (PAS).
REFERENCES 1. 2. 3.
4.
5. 6.
7. 8.
9.
10.
Kaufman PL, Gabelt BT, Cynander M. Introductory comments on neuroprotection. Surv Ophthalmol 1999; 43:S89–90. Sutter EE, Bearse MA. The optic nerve head component of the human ERG. Vision Res 1998; 39:419–436. Hood DC, Greenstein VC, Holopigian K, et al. An attempt to detect glaucomatous damage to the inner retina with the multifocal erg. Invest Ophthalmol 2000; 41: 1570–1579. Hasegawa S, Takagi M, Usui T, Takada R, Abe H. Waveform changes of the firstorder multifocal electroretinogram in patients with glaucoma. Invest Ophthalmol Vis Sci 2000; 41:1597–1603. Rodieck R. The First Steps in Seeing. Sunderland: Sinauer Associates, 1998:256. Quigley H, Addicks E, Green W. Optic nerve damage in human glaucoma: III. Quantitative correlation of nerve fiber loss and visual field defect in glaucoma, ischemic neuropathy, papilledema and toxic neuropathy. Arch Ophthalmol 1982; 100: 135–146. Drance S, Anderson D. Automatic Perimetry in Glaucoma: A Practical Guide. New York: Grune & Stratton, 1985:36. Quigley HA, Dunkelberger GR, Green WR. Retinal ganglion cell atrophy correlated with automated perimetry in human eyes with glaucoma. Am J Ophthalmol 1989; 107:453–64. Harwerth R, Carter-Dawson L, Shen F, Smith E, Crawford M. Ganglion cell losses underlying visual field defects from experimental glaucoma. Invest Ophthalmol Vis Sci 1999; 40:2242–2250. Johnson CA. Selective versus non-selective losses in glaucoma. J Glaucoma 1994; 3:S32–S44.
286
Sample
11.
Sample PA, Madrid M, Weinreb RN. Evidence for a variety of functional defects in glaucoma suspect eyes. J Glaucoma 1994; 3:S5–S18. Sample PA, Johnson CA, Haegerstrom-Portnoy G, Adams AJ. Optimum parameters for short-wavelength automated perimetry. J Glaucoma 1996; 5:375–83. Sample PA, Weinreb RN. Color perimetry for assessment of primary open-angle glaucoma. Invest Ophthalmol Vis Sci 1990; 31:1869–1875. Dacey DM. Morphology of a small-field bistratified ganglion cell type in the macaque and human retina. Vis Neurosci 1993; 10:1081–1098. Kouyama N, Marshak DW. Bipolar cells specific for blue cones in the macaque retina. J Neurosci 1992; 12:1233–1252. Sample PA. Short-wavelength automated perimetry: its role in the clinic and for understanding ganglion cell function. Progr Retinal Res 2000; 19:369–383. Sample PA, Taylor JD, Martinez GA, Lusky M, Weinreb RN. Short-wavelength color visual fields in glaucoma suspects at risk. Am J Ophthalmol 1993; 115:225– 233. Johnson CA, Adams AJ, Casson EJ, Brandt JD. Blue-on-yellow perimetry can predict the development of glaucomatous visual field loss. Arch Ophthalmol 1993; 111: 645–650. Sample PA, Martinez GA, Weinreb RN. Color visual fields: a 5 year prospective study in eyes with primary open angle glaucoma. In: Mills R, eds. Perimetry Update 1992/1993: Proceedings of the Xth International Perimetric Society Meeting. Amsterdam: Kugler, 1993:473–476. Sample PA, Weinreb RN. Progressive color visual field loss in glaucoma. Invest Ophthalmol Vis Sci 1992; 33:2068–2071. Johnson CA, Adam AJ, Casson EJ, Brandt JD. Progression of early glaucomatous visual field loss for blue-on-yellow and standard white-on-white automated perimetry. Arch Ophthalmol 1993; 111:651–656. Sample PA, Plummer DJ, Weinreb RN. Standard achromatic perimetry vs. shortwavelength automated perimetry for following glaucoma. In: Kriglestein GK, eds. Glaucoma Update V. Berlin: Springer-Verlag, 1995:197–204. Sample PA, Cook JN, Weinreb RN. Variability and sensitivity of short-wavelength color visual fields in normal and glaucoma eyes. Opt Soc Am 1993:292–295. Kwon YH, Park HJ, Aliza JAP, Ugurlu S, Caprioli J. Test-retest variability of blueon-yellow perimetry is greater than white-on-white perimetry in normal subjects. Am J Ophthalmol 1998; 126:29–36. Dacey D, Lee B. The ‘blue-on’ opponent pathway in primate retina originates from a distinct bistratified ganglion cell type. Nature 1994; 367:731–735. Dacey D. Circuitry for color coding in the primate retina. Proc Natl Acad Sci USA 1996; 93:582–588. Martin PR, White AJ, Goodchild AK, Wilder HD, Sefton AE. Evidence that blueon cells are part of the third geniculocortical pathway in primates. Proc Natl Acad Sci USA 1997; 27:5900–5905. Livingstone M, Hubel D. Segregation of form, color, movement, and depth: anatomy, physiology, and perception. Science 1988; 240:740–749. Frisen L, Nikolajeff F. Properties of high-pass resolution perimetry targets. Acta Ophthalmol 1993; 71:320–326.
12. 13. 14. 15. 16. 17.
18.
19.
20. 21.
22.
23. 24.
25. 26. 27.
28. 29.
Evaluation of Visual Outcome 30.
31.
32. 33. 34. 35. 36. 37. 38. 39. 40. 41.
42. 43. 44. 45. 46.
47.
48.
49.
287
Martinez GA, Sample PA, Weinreb RN. Comparison of high-pass resolution perimetry and standard automated perimetry in glaucoma. Am J Ophthalmol 1995; 119: 195–201. Chauhan B, House P, McCormick T, Raymond B, LeBlanc R. Comparison of conventional and high-pass resolution perimetry in a prospective study of patients with glaucoma and healthy eyes. Arch Ophthalmol 1999; 117:24–33. Johnson CA, Samuels SJ. Screening for glaucomatous visual field loss with frequency-doubling perimetry. Invest Ophthalmol Vis Sci 1997; 38:413–425. Maddess T, Henry GH. Performance of nonlinear visual units in ocular hypertension and glaucoma. Clin Vis Sci 1992; 7:371–383. Kelly D. Frequency doubling in visual responses. J Opt Soc Am A 1966; 56:1628– 1633. Kelly DH. Nonlinear visual responses to flickering sinusoidal gratings. J Opt Soc Am 1981; 71:1051–1055. Maddess T, Hemmi JM, James AC. Evidence for spatial aliasing effects in the Y-like cells of the magnocellular visual pathway. Vis Res 1998; 38:1843–1859. Lee BB, Kremers J, Yeh T. Receptive fields of primate retinal ganglion cells studied with a novel approach. Vis Neurosci 1998; 15:161–175. Schiller PH, Logothetis NK. The color-opponent and broad-band channels of the primate visual system. Trends Neurosci 1990; 13:392–398. Schiller PH, Logothetis NK, Charles ER. Role of the color-opponent and broadband channels in vision. Vis Neurosci 1990; 5:321–346. Schiller PH, Logothetis NK, Charles ER. Parallel pathways in the visual system: their role in perception at isoluminance. Neuropsychologia 1991; 29:433–441. Shapley R, Kaplan E, Soodak R. Spatial summation and contrast sensitivity of X and Y cells in the lateral geniculate nucleus of the macaque. Nature 1981; 292: 543–545. Kulikowski JJ, Tolhurst DJ. Psychophysical evidence for sustained and transient detectors in human vision. J Physiol 1973;232:149–162. Tyler CW. Observations on spatial-frequency doubling. Perception 1974; 3:81–86. Tyler CW. Specific deficits of flicker sensitivity in glaucoma and ocular hypertension. Invest Ophthalmol Vis Sci 1981; 20:204–212. Flood T, Flanagan JG. The influence of detection versus resolution threshold criteria in frequency doubling perimetry. J Acad Optom 1998; 75:S31. Bosworth CF, Sample PA, Williams JM, Zangwill L, Lee BL. Spatial relationship of motion automated perimetry and optic disc topography in patients with glaucomatous optic neuropathy. J Glaucoma 1999; 8:281–289. Sponsel W, Arango S, Trigo Y, Mensah J. Clinical classification of glaucomatous visual field loss by frequency doubling perimetry. Am J Ophthalmol 1998; 125: 830–836. Serguhn S, Spiegel D. Comparison of frequency doubling perimetry to standard achromatic automated perimetry in glaucoma. Invest Ophthalmol Vis Sci (ARVO Abstract) 1999; 40:S842. Kondo Y, Yamamoto T, Sato Y, Matsubara M, Kitizawa Y. A frequency-doubling perimetric study in normal-tension glaucoma with hemifield defect. J Glaucoma 1998; 7:261–265.
288
Sample
50.
Quigley HA. Identification of glaucoma-related visual field abnormality with the screening protocol of frequency doubling technology. Am J Ophthalmol 1998; 125: 819–829. Sample PA, Bosworth CF, Blumenthal EZ, Girkin C, Weinreb RN. Visual function specific perimetry for indirect comparison of different ganglion cell populations in glaucoma. Invest Ophthalmol Vis Sci 2000; 41:1783–1790. Johnson CA, Demirel S. The role of spatial and temporal factors in frequencydoubling perimetry. In: Wall M, Heijl A, eds. Perimetry Update 1996/97: Proceedings of the XIIth International Perimetric Society Meeting. Amsterdam: Kugler, 1997:13–19. Chauhan D, Johnson C. Test-retest variability of frequency-doubling perimetry and conventional perimetry in glaucoma patients and normal subjects. Invest Ophthalmol Vis Sci 1999; 40:648–656. Johnson C, Cioffi G, Van Buskirk E. Frequency doubling technology perimetry using a 24-2 stimulus presentation pattern. Optom Vis Sci 1999; 76:571–581. Gordon M, Kass M. The ocular hypertension treatment study: design and baseline description of the participants. Arch Ophthalmol 1999; 117:573–583. Sample PA, Emdadi A, Kono Y, Weinreb RN. Repeatability of abnormality and progression in glaucomatous standard and SWAP visual fields. In: Wall M, Heijl A, eds. The XIIIth International Perimetric Society Meeting. Amsterdam: Kugler, 1998:57. Demirel S, Johnson CA. Incidence of short-wavelength automated perimetry (SWAP) deficits in ocular hypertensive eyes. North Am Perimetr Soc Meeting Proc 1998. Bengtsson B, Olsson J, Heijl A, Rootzen H. A new generation of algorithms for computerized threshold perimetry SITA. Acta Ophthalmol Scand 1997; 75:368– 375. Bengtsson B, Heijl A. Evaluation of a new perimetric threshold strategy, SITA, in patients with manifest and suspect glaucoma. Acta Ophthalmol Scand 1998; 76: 268–272. Wild JM, Pacey IE, Hancock SA, Cunliffe IA. Between algorithm, between-individual differences in normal perimetric sensitivity: full threshold, FASTPAC, and SITA. Swedish Interactive Threshold algorithm. Invest Ophthalmol Vis Sci 1999; 40:1152–1161. Wild JM, Pacey IE, O’Neill EC, Cunliffe IA. The SITA perimetric threshold algorithms in glaucoma. Invest Ophthalmol Vis Sci 1999; 40:1998–2009. Bengtsson B, Heijl A. Inter-subject variability and normal limits of the SITA Standard, SITA Fast, and the Humphrey Full Threshold computerized perimetry strategies, SITA STATPAC. Acta Ophthalmol Scand 1999; 77:125–129. Heijl A, Bengtsson B. Third generation rapid algorithms for static computerized perimetry, SITA. In: Kriglestein G, eds. Glaucoma Update VI. Berlin: SpringerVerlag, 2000:125–131. Heijl A, Bengtsson B, Patella VM. Glaucoma follow-up when converting from long to short perimetric threshold tests. Arch Ophthalmol 2000; 118:489–493. Olsson J, Rootzen H. An image model for quantal response analysis in perimetry. Scand J Statistics 1994; 21:375–387. Anderson D, Patella V. Automated Static Perimetry. 2d ed. New York: Mosby, 1999:85–87.
51.
52.
53.
54. 55. 56.
57. 58.
59.
60.
61. 62.
63.
64. 65. 66.
Evaluation of Visual Outcome 67.
68. 69. 70. 71.
72.
73. 74. 75.
76.
77.
78.
79.
80. 81.
82.
83.
289
Sharma AK, Goldberg I, Graham SL, Mohsin M. Comparison of the Humphrey Swedish interactive thresholding algorithm (SITA) and full threshold strategies. J Glaucoma 2000; 9:20–27. Schulzer M, The NTGS group. Errors in the diagnosis of visual field progression in normal-tension glaucoma. Ophthalmology 1994; 101:1589–1594. Leske C, Heijl A, Hyman L, Bengtsson B, The EMGT Group. Early Manifest Glaucoma Trial. Ophthalmology 1999; 106:2144–2153. Investigators TA. The Advanced Glaucoma Intervention Study (AGIS): 2. Visual field scoring and reliability. Ophthalmology 1994; 101:1445–1455. Musch DC, Lichter PR, Guire KE, Standardi CL. The Collaborative Initial Glaucoma Treatment Study: study design, methods, and baseline characteristics of enrolled patients. Ophthalmology 1999; 106:653–662. Katz J, Congdon N, Friedman D. Methodological variations in estimating apparent progressive visual field loss in clinical trials of glaucoma treatment. Arch Ophthalmol 1999; 117:1137–1142. Katz J. Scoring systems for measuring progression of visual field loss in clinical trials of glaucoma treatment. Ophthalmology 1999; 106:391–395. Weber J, Dannheim F, Dannheim D. The topographical relationship between optic disc and visual field in glaucoma. Acta Ophthalmol 1990; 68:568–574. Teesalu P, Airaksinen PJ, Tuulonen A. Blue-on-yellow visual field and retinal nerve fiber layer in ocular hypertension and glaucoma. Ophthalmology 1998; 105:2077– 2081. Teesalu P, Vihanninjoke K, Airaksinen J, Tuulonen A, Laara E. Correlation of blueon-yellow visual fields with scanning confocal laser optic disc measurements. Invest Ophthalmol Vis Sci 1997; 38:2452–2459. Polo V, Abecia E, Pablo LE, et al. Short-wavelength automated perimetry and retinal nerve fiber layer evaluation in suspected cases of glaucoma. Arch Ophthalmol 1998; 116:1295–1298. Girkin CA, Emdadi A, Sample PA, et al. Short-wavelength automated perimetry and standard perimetry in detection of progressive optic disc cupping. Arch Ophthalmol 2000; 118:1231–1236. Mansberger SL, Sample PA, Zangwill L, Weinreb RN. Achromatic and shortwavelength automated perimetry in patients with glaucomatous large cups. Arch Ophthalmol 1999; 117:1473–1477. Yamagishi N, Anton A, Sample PA, et al. Mapping structural damage of the optic disc to visual field defect in glaucoma. Am J Ophthalmol 1997; 123:667–676. Anton A, Yamagishi N, Zangwill L, Sample PA, Weinreb RN. Mapping structural to functional damage in glaucoma with standard automated perimetry and confocal scanning laser ophthalmoscopy. Am J Ophthalmol 1998; 125:436–446. Boden C, Sample PA, Weinreb RN. Early glaucoma visual field loss rarely crosses the horizontal midline: implications for ganglion cell death. Invest Ophthalmol Vis Sci (ARVO abstract) 2001; 42:In press. Sample PA, Bosworth CF, Pascual J, Vasile C, Weinreb RN. Patterns of glaucomatous visual field progression. Invest Ophthalmol Vis Sci (ARVO abstract) 2000; 41:S103,#539.
17 Clinical Trials in Neuroprotection Scott M. Whitcup Allergan, Inc. Irvine and Jules Stein Eye Institute David Geffen School of Medicine, UCLA Los Angeles, California, U.S.A.
I.
INTRODUCTION
A clinical trial is a planned experiment in humans designed to assess the safety and/or efficacy of a treatment. The well-designed clinical trial should control for bias that can corrupt the interpretation of clinical data. Unfortunately, a great deal of medical practice is based on anecdotal clinical reports or poorly designed clinical studies. Much of the scientific dogma we read in textbooks is actually based on a retrospective review of inconclusive data obtained from a handful of patients. This is especially true of new therapeutic areas in medicine, where experience and published data are lacking.
II. METHODS OF CLINICAL STUDIES There are four basic types of clinical studies: case series, case-control studies, cohort studies, and randomized clinical trials. Case reports or case series are usually retrospective reviews that detail the clinical findings and outcomes of patients with a particular disease. Although these reports can help define the manifestations of a disease, they are prone to bias and can mislead the reader. Since the data are collected retrospectively from patient charts, critical information is often missing. The disease may not be well defined in the report, and 291
292
Whitcup
some of the reported patients may actually have a different condition. There may be bias in acquiring the patients, and the reported patients may be dissimilar from the general patient population. Case series tend to be written by specialists who treat patients with more severe or atypical diseases. Importantly, case series lack a control group for comparison. For example, a physician could report two patients with non-arteritic ischemic optic neuropathy (NAION) who had substantial improvement in their vision after starting multivitamins. It would be difficult to conclude that vitamins improve vision in patients with this disease without knowing how many patients with NAION on vitamins had no improvement in vision. Occasionally, investigators will try to compensate for the lack of an appropriate control group in a study by comparing their study results to a group of historical controls. The investigators agree that a control group is needed, but are still reluctant to randomly assign patients to the new treatment or to a standard therapy or placebo. There are a number of reasons for this reluctance. First, it is much more difficult to conduct a well-controlled, randomized clinical trial. A protocol needs to be written, institutional review board (IRB) approval is required, and the methods for patient randomization, conduct of the trial, collection of the data, and analysis of the results need to be detailed. Second, many investigators truly believe that the new treatment is better, and that it would be unethical to keep patients from receiving the new treatment. The main problem with the use of historical controls is that data from historical controls tend to be biased. Data from historical controls are often collected differently from patients enrolled in a trial and followed prospectively (information bias). Patients in a trial can also differ clinically from the patients in a historical control group (selection bias), not only in recognized important clinical parameters like disease severity, but also in potentially unrecognized or undocumented parameters that could affect a clinical outcome—for example, diet or other environmental factors. There are numerous other sources of bias in clinical studies [1]. Observer bias leads to a systematic alteration in the measuring of a response in patients. Unvalidated or inappropriate instruments for measurement can also bias the results of a study. In a properly designed trial, controlling for confounding factors can minimize bias. Randomization, for example, can help to balance these factors in the treatment groups. Importantly, randomization helps to balance unrecognized sources of bias between groups. In the case-control study the investigator compares a group of patients with a given disorder to a control group without this condition. The clinical records of both groups are then compared to see if certain factors occur more commonly in one group. A classic example of a case control study would compare the smoking history of a group of patients with lung cancer and an age- and sex-matched control group. One can then calculate an odds ratio that states the relative risk for a condition like lung cancer given a specific risk factor like smoking. For
Clinical Trials in Neuroprotection
293
example, an odds ratio of 6.7 would mean that people who smoke are 6.7 times more likely to develop lung cancer than people who do not smoke. Case-control studies, although more powerful than case series, also rely on a retrospective review of patient records. Again, bias may systematically alter the data and lead to inappropriate conclusions. There may be a recording bias in the information collected from patients and controls. For example, patients with cancer may spend more time thinking about their medical history and reasons why they might have developed their disease than would a person without the disease. Physicians may spend a great deal of time detailing clinical information from patients that may not be collected from the controls. Despite this potential bias, well-conducted case-control studies can provide useful clinical information, especially when standard procedures for data collection are followed. Furthermore, case-control studies may be the only feasible method for studying certain rare conditions. Cohort studies identify two groups of patients; for example, one cohort receives a treatment and the other cohort does not receive the therapy. The two groups of patients or cohorts are then followed prospectively for the development of a specific outcome. However, because the treatment is not randomly assigned, the two groups of patients may differ greatly in certain critical clinical parameters. For example, maybe the treatment is given only to the most severely ill patients who have “nothing to lose.” These patients may be unlikely to respond to any treatment, no matter how effective. Pharmaceutical drug development therefore includes a number of clinical studies, but final determination of safety and efficacy is based predominantly on pivotal randomized clinical trials. Clinical studies during the development of a new medicine are often divided into four phases. Phase 1 clinical trials are the initial safety trials of a new medicine. These are usually conducted in normal volunteers, often in males. In the field of cancer, Phase 1 trials are often conducted in more severely ill patients. The trials can be open label, where patients and investigators are unmasked to the treatment allocation. Multiple doses may be tested in a Phase 1 trial, often starting with the lowest dose and escalating to higher dosages if they are tolerated. Phase 2 trials are designed to study the safety and efficacy of a new medication. These trials are often double-masked, where both the patient and investigators do not know what treatment is being administered. Classically these studies are called double-blind studies; however, in ophthalmology we prefer the term double-masked, since it is difficult to get a patient with an eye disease to enroll in a study with double-blind in the title. Phase 2 trials typically have more patients than Phase 1 trials, are conducted in patients with the disease, but still may examine several dosages or treatment regimens. The Phase 3 clinical trial is the pivotal clinical study for the approval of the medication. These studies almost always are larger randomized clinical trials comparing the new medication to the standard treatment or to placebo. The
294
Whitcup
United States Food and Drug Administration (FDA) usually requires two Phase 3 trials prior to the approval of a new drug. Studies conducted after a medicine is approved and marketed are called Phase 4 trials. These studies are conducted in patient populations for which the medicine is intended and may compare the medicine to currently available therapies. These studies are also used to elucidate additional clinical data that supplements the results of the Phase 1, 2, and 3 trials. The randomized clinical trial provides the most robust evidence about the safety and efficacy of a new treatment. Because patients are randomly assigned to the new treatment or to the control treatment, if the number of patients in the study is large enough, the treatment groups are usually similar. This is a critical point, since treatment outcome could be affected if the groups differed in clinically relevant parameters. Although one could try to control or compensate for imbalances using certain statistical analyses, this only works for the parameters that are thought to affect outcome and for which data are available. As stated before, randomization is powerful because it controls for both known and unknown sources of bias.
III. ISSUES IN THE DESIGN AND CONDUCT OF CLINICAL TRIALS Table 1 lists the components of a well-designed clinical study. It is important that the investigators be experienced in conducting clinical trials and have clinical expertise in the disease being studied. The primary outcome variable of the study, also known as the primary endpoint, should be clearly stated. This primary outcome variable is the main parameter on which the investigator plans to judge the efficacy of the intervention. Therefore, it is important to prospectively choose the primary outcome variable, even if multiple clinical outcomes are examined. The procedure for enrolling patients into the study should be detailed, and the inclusion and exclusion criteria listed, because the patient population will determine how generalizable the results are to a larger patient population outside of the trial. For example, results of a potential neuroprotective medication studied in patients with narrow angle glaucoma and intraocular pressures of 40 mmHg or above may not be generalizable to patients with open angle glaucoma with pressures in the range of 22 to 30 mmHg. The treatment and dosing regimen should be clearly stated. Choice of the control treatment is also critical to the value of the study. Patients in the control group should be treated according to the current best standard of care. If no proven treatment is available, a placebo could be considered. The dosage of the control regimen should also be appropriately chosen. It would be inappropriate to compare a new glaucoma medication to pilocarpine 1% dosed once daily.
Clinical Trials in Neuroprotection Table 1
295
Components of a Randomized Clinical Trial
Study approved by Institutional Review Board. Appropriate informed consent obtained from patients. Disease well defined with specific diagnostic criteria. Patient population well defined with specific inclusion and exclusion criteria. Patients randomly assigned to new treatment and control treatment according to standardized procedures. Patients and investigators appropriately masked from treatment assignment. Sample size accurately determined to control for type I and type II error. Outcome measures specified and minimum differences to be considered as clinically important detailed. Procedures for the conduct of the trial well detailed. Timing of study visits and collection of data strictly specified. Statistical analysis plan specified prior to locking the database and unmasking of treatment assignments. Results of an intent-to-treat analysis, where all randomized patients are included in the analysis should be provided, even if additional analyses performed.
It is extremely important to perform appropriate sample size calculations for all clinical trials. Sample size is based not only on the event rate expected in the two groups, but also on the desired level of protection against type I and type II error. Type I error (alpha) occurs when the study falsely concludes that the therapies tested are different when in fact they are the same. Especially when a standard therapy for a disease currently exists, most clinical trials protect more strongly against this type of error, since one would not want a new treatment to be wrongly administered when an effective therapy is available. Most studies limit the possibility of type I error to less than 0.05 (5%). Type II error (beta) occurs when a study falsely concludes that there is no difference between the treatments when in fact a difference exists. Typically, type II error for many clinical trials is set at 0.2 (20%). This means that there is a 20% chance that the treatments are different although the study shows no significant difference. The number of patients greatly affects type 2 error. Statistical power (1-beta) is the chance of proving the difference between the two groups that is defined in the sample size calculations. Many studies in the literature are underpowered. They do not have sufficient patients to have a reasonable chance of detecting a meaningful difference between the two groups. A small study that concludes that there is no significant difference between two treatments can be extremely misleading. The treatment effect could be 40% higher with the new treatment, but if the numbers of patients are small and the variability of response high, this difference may not be statistically significant, yet clearly be clinically meaningful. In analyzing study results, look at the actual data, and do not be misled by an insignificant
296
Whitcup
P value, especially if the study is inadequately powered to detect a clinically meaningful difference. Remember that clinical studies can never definitively prove that two treatments are the same.
IV. CLINICAL TRIALS OF NEUROPROTECTION IN OPHTHALMOLOGY Proof of neuroprotection in ophthalmology will require a well-designed clinical trial employing many of the strategies detailed above. The most definitive evidence will require a randomized clinical trial comparing the potential neuroprotective treatment to a control treatment or placebo. Although there are difficulties in conducting any clinical trial, studies of potential neuroprotective medications will encompass several additional challenges. There are very few randomized clinical trials of neuroprotective medications in medicine. One such study, the Glycine Antagonist in Neuroprotection (GAIN) Americas trial, was a randomized, double-masked placebo-controlled trial conducted to examine the efficacy of gavestinel, an antagonist of the glycine site of the N-methyl-d-aspartate receptor, as a neuroprotective therapy for acute ischemic stroke [2]. The main outcome measure was functional capability at 3 months, measured by the Barthel Index. This study concluded that gavestinel administered up to 6 h after an acute ischemic stroke did not improve functional outcome at 3 months. Memantine, an uncompetitive N-methyl-d-aspartate (NMDA) antagonist has been studied in patients with severe dementia, both Alzheimer type and vascular type [3]. In this study, neuroprotection with memantine led to functional improvement and reduced care dependence in severely demented patients. There are even fewer randomized controlled clinical trials of neuroprotection in ophthalmology. One example is the Ischemic Optic Neuropathy Decompression Trial (IONDT) [4]. This was a National Eye Institute–sponsored multicenter clinical trial designed to assess the safety and efficacy of optic nerve decompression surgery compared with careful follow-up alone in patients with nonarteritic ischemic optic neuropathy (NAION). Prior to this study, several nonrandomized trials showed benefits of optic nerve decompression; however, none of these were randomized studies [5–8]. Interestingly, results from the IONDT showed that patients assigned to surgery did no better when compared with patients assigned to careful follow-up [4]. Improved visual acuity of three or more lines of visual acuity was achieved by 23.6% of the surgery group compared with 42.7% of the careful follow-up group. In fact, patients receiving surgery had a significantly greater risk of losing three or more lines of vision at 6 months: 23.9% in the surgery group worsened compared with 12.4% in the careful followup group.
Clinical Trials in Neuroprotection
297
The IONDT illustrates the need for a well-controlled clinical study to adequately assess neuroprotective therapy. Prior to this trial, improvement in visual acuity in patients with NAION was thought to be rare: less than 10% [9,10]. In the IONDT, 42.7% of patients in the careful follow-up group had a three line or greater improvement in visual acuity [4]. Furthermore, optic nerve decompression surgery was found to be ineffective and potentially harmful to patients with this disease. A. Endpoints A critical factor in designing clinical trials for neuroprotection is endpoint selection. The primary outcome measures in the IONDT were gain or loss of three or more lines of visual acuity on the New York Lighthouse chart at 6 months after randomization [4]. This is a functional endpoint that has been used in a number of ophthalmology trials and is felt to be clinically meaningful. Visual acuity has also been used as a standard clinical endpoint for clinical trials in macular degeneration, where central acuity can be affected early in the course of the disease. Unfortunately, changes in central acuity may be an insensitive endpoint for many ophthalmic diseases. For example, central visual acuity loss occurs relatively late in the course of some diseases including glaucoma and retinitis pigmentosa. Although visual field loss can be used as a functional endpoint, visual field loss progresses slowly and may require many years before meaningful changes occur. Identification and validation of surrogate endpoints will improve our ability to assess neuroprotective therapies. In a sense, intraocular pressure has been used as a surrogate endpoint for glaucoma treatments. Without well-controlled data, lowering intraocular pressure was felt to reduce the risk of vision loss in patients with glaucoma. Data from randomized clinical trials that support the benefits of lowering intraocular pressure are now becoming available. In a recent report from the Advanced Glaucoma Intervention Study (AGIS), investigators showed that eyes with intraocular pressure less than 18 mmHg for 100% of visits over 6 years had mean changes from baseline in visual field defect score close to zero during follow-up, whereas eyes with intraocular pressure less than 18 mmHg for less than 50% of visits had an estimated worsening over follow-up of 0.63 units of visual field defect score [11]. There are a number of new methods for assessing visual function that may be useful as endpoints in clinical trials for neuroprotection. Depending on the disease, endpoints may change at differing rates and have a large impact on length of clinical trials and the required sample size. Other studies have documented that rates of change differ among measures of visual function. In a trial of patients with retinitis pigmentosa, investigators assessed changes in measures of visual function in patients over time [12]. The smallest amount of change occurred for
298
Whitcup
visual acuity and hue discrimination, and the greatest amount of change occurred for visual field area. Electrophysiologic testing, newer methods of visual field testing, and contrast sensitivity are just a few examples of potential endpoints for clinical trials in neuroprotection. The electroretinogram was used as the primary outcome measure in a randomized clinical trial of vitamin A and vitamin E supplementation for retinitis pigmentosa [13]. This was a National Eye Institute–sponsored randomized, double-masked trial to determine whether supplements of vitamin A or vitamin E alone or in combination affect the course of retinitis pigmentosa. In this study, the main outcome measure was the cone electroretinogram amplitude. Patients receiving 15,000 IU/day of vitamin A were 32% less likely to have a decline in amplitude of 50% or more from baseline than those not receiving this dosage (P ⫽ 0.03). Although not statistically significant, similar trends were observed for rates of decline of visual field area. These data support the potential benefit of identifying endpoints that may be more sensitive or less variable in clinical trials. More accurate and sensitive measures of visual function will improve our ability to test neuroprotective therapies. Use of the scanning laser ophthalmoscope to assess perimetry [14] or of the multifocal electroretinogram [15] may be useful endpoints in trials of retinal disease. Similarly, new methods of assessing glaucomatous damage such as the Heidelberg retinal tomograph (HRT), the GDx nerve fiber analyzer (GDx), and the optical coherence tomograph (OCT) may be important measures of visual function in neuroprotection trials in patients with glaucoma [16–19]. New methods of measuring visual field using short-wavelength automated perimetry or multifocal visual evoked potential could also improve our ability to accurate assess changes in visual function in these patients [20,21]. However, it is important to remember that all of these endpoints need to be extensively evaluated and validated before widespread use in clinical trials. Increasingly, there is an interest in assessing the effect of new therapies on quality of life. Questionnaires are often employed to determine quality of life; however, it is important to make sure that the questionnaires are validated for the disease of interest. The visual function questionnaire is one such quality of life measure that has been validated in a number of diseases, including diabetes mellitus, macular degeneration, glaucoma, and cytomegalovirus retinitis [22]. B.
Neuroprotection and Glaucoma
A number of scientists are trying to develop neuroprotective therapies for glaucoma. However, data suggest that lowering IOP can decrease visual loss, and in a sense, provide neuroprotection. Therefore, in studying neuroprotection in glaucoma, it will be important to assess preservation of vision while controlling for intraocular pressure. For example, there are a number of laboratory studies of glaucoma medications that demonstrate neuroprotection in animal models of
Clinical Trials in Neuroprotection
299
optic nerve disease. It is important to compare these drugs to control medications that similarly lower IOP but reportedly have no neuroprotective effects. For example, in an animal model of ocular hypertension, systemic administration of brimonidine or timolol had equivalent effect on IOP [23]. Nevertheless, brimonidine significantly reduced the progressive loss of retinal ganglion cells by greater than 50%, whereas timolol had no effect. These considerations should also apply to human clinical trials. Two Phase 3 randomized, double-masked clinical trials are in progress assessing the neuroprotective effects of memantine, a drug that blocks the NMDA receptor. In this trial, the neuroprotective effects of the drug will be assessed independently from IOP. Although definitive proof of a neuroprotective medication in ophthalmology will probably require data from a randomized, clinical trial, there are some instances where an open-label trial of a medication could lead to believable evidence. In diseases where the clinical outcome is well documented with little to no variability, an open-label trial with historical controls could provide convincing data, using historical controls. Unfortunately, there are few such conditions. One potential disorder is central retinal artery occlusion in patients lacking a cilioretinal artery. If a treatment were studied that preserved 20/20 vision in even a relatively small number of patients, the data could be fairly convincing, because the visual outcome of this condition is usually catastrophic. However, there is still the opportunity for bias, and one should be very careful about misinterpreting data from trials without a concurrent control group and random assignment of treatment.
V.
CONCLUSIONS
In conclusion, neuroprotection offers an exciting therapeutic approach to a number of diseases. Clearly, patients with disorders affecting the retina or optic nerve may benefit from neuroprotective medications. One must be careful in interpreting data from small, uncontrolled studies. As can be seen, there are a multitude of issues that must be considered. Well-designed clinical trials with validated endpoints will provide the best insight on the neuroprotective effects of medications and provide new treatment options that, one hopes, will save vision.
REFERENCES 1. 2.
Sackett DL. Bias in analytic research. J Chronic Dis 1979; 32:51–63. Sacco RL, DeRosa JT, Haley ED Jr, Levin B, Ordronneau P, Phillips SJ, Rundek T, Snipes RG, Thompson JL. The Glycine Antagonist in Neuroprotection Americas Investigators. Glycine antagonist in neuroprotection for patients with acute stroke: GAIN Americas: a randomized controlled trial. JAMA 2001; 285:1719–1728.
300
Whitcup
3.
Winblad B, Poritis HN. Memantine in severe demintia: results of the 9M-Best Study (benefit and efficacy in severely demented patients during treatment with memantine). Int J Geriatr Psychiatry 1999; 14:135–146. The Ischemic Optic Neuropathy Decompression Trial Research Group. Optic nerve decompression surgery for nonarteritic anterior ischemic optic neuropathy (NAION) is not effective and may be harmful. JAMA 1995; 273:625–632. Sergott RC, Cohen MS, Bosley TM, Savino PJ. Optic nerve decompression may improve the progressive form of nonarteritic ischemic optic neuropathy. Arch Ophthalmol 1989; 107:1743–1754. Kelman SE, Elman MJ. Optic nerve sheath decompression for non-arteritic ischemic optic neuropathy improves multiple visual function parameters. Arch Ophthalmol 1991; 109:667–671. Spoor TC, Wilkinson MJ, Ramocki JM. Optic nerve sheath decompression for the treatment of progressive nonarteritic ischemic optic neuropathy. Am J Ophthalmol 1991; 111:724–728. Spoor TC, McHenry JG, Lau-Sickon L. Progressive and static nonarteritic ischemic optic neuropathy treated by optic nerve sheath decompression. Ophthalmology 1993; 100:306–311. Boghen DR, Glaser JS. Ischemic optic neuropathy: a clinical profile and natural history. Brain 1975; 98:689–708. Repka MX, Savino PJ, Schatz NJ, Sergott RC. Clinical profile and long-term implications of anterior ischemic optic neuropathy. Am J Ophthalmol 1983; 96:478– 483. The AGIS Investigators. The advanced glaucoma intervention study (AGIS): 7. The relationship between control of intraocular pressure and visual field deterioration. Am J Ophthalmol 2000; 130:429–440. Holopigian K, Greenstein V, Seiple W, Carr RE. Rates of change differ among measures of visual function in patients with retinitis pigmentosa. Ophthalmology 1996; 103:398–405. Berson EL, Rosner B, Sandberg MA, et al. A randomized trial of vitamin A and vitamin E supplementation for retinitis pigmentosa. Arch Ophthalmol 1993; 111: 761–772. Sunness JS, Schuchard RA, Shen N, Rubin GS, Dagnelie G, Haselwood DM. Landmark-driven fundus perimetry using the scanning laser ophthalmoscope. Invest Ophthalmol Vis Sci 1995; 36:1863–1874. Hood DC, Seiple W, Hologigian K, Greenstein V. A comparison of the components of the multifocal and full-field ERGs. Vis Neurosci 1997; 14:533–544. Miglior S, Casula M, Guareschi M, Marchetti I, Iester M, Orzalesi N. Clinical ability of Heidelberg retinal tomograph examination to detect glaucomatous visual field changes. Ophthalmology 2001; 108:1621–1627. Zangwill LM, Bowd C, Berry CC, Williams J, Blumenthal EZ, Sanchez-Galena CA, Vasile C, Weinreb RN. Discriminating between normal and glaucomatous eyes using the Heidelberg retina tomograph, GDx Nerve Fiber Analyzer, and Optical Coherence Tomograph. Arch Ophthalmol 2001; 119:985–993. Wollstein G, Garway-Heath DF, Fontana L, Hitchings RA. Identifying early glaucomatous changes. Comparison between expert clinical assessment of optic disc pho-
4.
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15. 16.
17.
18.
Clinical Trials in Neuroprotection
19.
20.
21. 22.
23.
301
tographs and confocal scanning ophthalmoscopy. Ophthalmology 2000; 107:2272– 2277. Bowd C, Zangwill LM, Berry CC, Blumenthal EZ, Vasile C, Sanchez-Galeana C, Bosworth CF, Sample PA, Weinreb RN. Detecting early glaucoma by assessment of retinal nerve fiber layer thickness and visual function. Invest Ophthalmol Vis Sci 2001; 42:1993–2003. Hutchings N, Hosking SL, Wild JM, Flanagan JG. Long-term fluctuation in shortwavelength automated perimetry in glaucoma suspects and glaucoma patients. Invest Ophthalmol Vis Sci 2001; 42:2332–2337. Klistorner A, Graham SL. Objective perimetry in glaucoma. Ophthalmology 2000; 107:2283–2299. Mangione CM, Lee PP, Gutierrez PR, Spritzer K, Berry S, Hays RD. Development of the 25-item National Eye Institute Visual Function Questionnaire. Arch Ophthalmol 2001; 119:1050–1058. WoldeMussie E, Ruiz G, Wijono M, Wheeler L. Brimonidine is neuroprotective to ganglion cells in rats with laser-induced chronic ocular hypertension. Invest Ophthalmol Vis Sci 2001; 42:2849–2855.
18 Regulatory Issues in Clinical Trials Anthony C. Arnold Jules Stein Eye Institute David Geffen School of Medicine, UCLA Los Angeles, California, U.S.A.
I.
INTRODUCTION
Investigators who perform clinical trials on human subjects in the United States must recognize certain established principles with regard to recruiting and maintaining subjects and must adhere to the policies of both local (Institutional Review Board, IRB) and federal (Food & Drug Administration, FDA) regulatory agencies. Coordination with sponsoring pharmaceutical company policies is essential in all aspects, particularly as compromises may be necessary to satisfy the requirements of all agencies involved.
II. RECRUITING AND MAINTAINING SUBJECTS Recruitment of subjects is often one of the most difficult features of a clinical trial. Regulatory issues connected with IRB review of recruitment will be discussed below. Key features for effectively maximizing entry of subjects into a trial include: 1. Awareness of both referral sources and subjects. Informational pamphlets for patients and physicians, individual letters to referral sources, CME lectures by investigators, publicity campaigns such as press releases and radio and television interviews, may all increase awareness of the disease to be studied and the specific clinical trial being developed. Professional publicists may be 303
304
Arnold
valuable assets in the development and implementation of an organized plan for dissemination of information. Referral sources should be encouraged to utilize their patient databases to identify subjects at risk or with the disease, but privacy issues regarding contact with such patients (see below) must be addressed. 2. Motivation of referral sources and subjects. The disease being investigated may be currently untreatable, and referral to a trial may provide a possible new hope for the subject as well as prestige for the referring physician by association with the center. However, both subjects and referring physicians must be invested in answering the question of effectiveness rather than simply obtaining a possible therapy, because a placebo or a standard therapy will be a possible result of entry; receipt of the study medication is not guaranteed. Subjects may be paid a reasonable amount for participation, but coercion by too high a sum must be avoided. Similarly, referral sources may receive some perquisites for their participation, including reimbursement for time in examination, but specific referral fees or “kickbacks” for referral of subjects are not acceptable. Referral sources must be comfortable that patients will be returned to their care following the trial. 3. Availability of clinical center personnel. A critical factor in subject recruitment is the ease with which subjects may be referred to the nearest Clinical Center. A single telephone call to the center should allow the referring physician to establish the subject’s probable eligibility and obtain instructions for timely evaluation of that subject without further demands on the referring physician’s time. This requires Center staff that is adequately trained, motivated to facilitate entry of subjects, skilled at interpersonal relations, and available to devote the necessary time to the trial. Additionally, ethical considerations in recruitment become important for clinical trials. 1. Overzealous recruitment. There is, of course, ample motivation for an investigator to recruit as many subjects for the trial as possible, for reasons related to rapid completion of the trial, prestige of high recruitment, and financial gain. The risks of overzealous recruitment include a. Coercion of subjects by counseling that slants information to encourage entry into the trial; b. Bias of data obtained (entry of subjects with incorrect diagnosis, poor followup for subjects who lack true motivation due to inaccurate counseling, and entering subjects after a time window for treatment in acute disease, thus biasing results of therapy negatively). c. Exposure of subjects to undue risk in cases of incorrect diagnosis or missed time window, when there may be no scientific basis for therapy. Advertisements or flyers must not be misleading. Examples of such improprieties include claims of safety and efficacy, referring to “new treatment” or “free medical treatment,” when in actuality the subject simply will not be charged for participating. 2. Treatment outside the trial. If a study drug is available outside the study but is not FDA-approved for the studied disease or at the dosage being
Regulatory Issues
305
studied, its use for the disorder may not be supported by the FDA except under special conditions. However, when the study drug is available and approved, for example when the study is comparing it to another therapy, subjects must be advised that it is available outside the trial. While such advice may impede recruitment, it is ethically required; the IRB will generally address this issue in the informed consent process.
III. INSTITUTIONAL REVIEW BOARD ISSUES Human medical research in the United States is guided by ethical principles developed and outlined in the World Medical Association Declaration of Helsinki (1964) and in the Belmont Report, a 1979 document created by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. All research involving human subjects in the United States is subject to Institutional Review Board (IRB) approval. The basis for IRB review of research is contained in the Code of Federal Regulations, Title 45, Part 46 (45 CFR 46). The IRB approval process is often complex and time-consuming, but is deemed essential for the adequate protection of human subjects, particularly where scientific and financial stakes are high. The investigator must become familiar with IRB requirements of the institution sponsoring or supporting the research; these may vary widely depending on the IRB constituency (private vs. public). Regarding the conduct of clinical trials, the following aspects are highlighted: A. Informed Consent Informed consent is considered to be a process, not an isolated, one-time counseling session between investigator and subject. This ongoing dialogue should provide continuing disclosure of risks and benefits as new information becomes available. It begins with the initial description of research activity in the recruitment phase and continues through the subject’s final participation in the research activity, with updates as necessary regarding new findings, risks, and benefits detected as the trial progresses. Such updates may include complications occurring during the trial, new treatment options that become available, or decisions by sponsor to provide study drug to subjects without charge following the study. The updates may require pre-approval by the IRB. The informed consent form (ICF) is the documentation of this process. In today’s clinical research environment, many elements of the ICF are required, and both IRBs and pharmaceutical sponsors develop templates for investigators’ use in clinical research. A great deal of time and effort may be saved by reconciling differences in required wording that may exist between the two versions prior to IRB submission.
306
B.
Arnold
Recruitment
In addition to the recruitment issues discussed above, certain points are of significant concern to the IRB, particularly with regard to privacy issues and the possibility of coercion. Areas of focus in reviewing studies include: Selection. Unless there is a specific medical justification that may be provided, subjects for a trial should not be preselected on the basis of age, race, gender, or other potentially biasing feature. Method of identification. The use of medical records or any information that is not in the public domain to identify potential subjects on the basis of existing medical conditions violates individual privacy rights. Similarly, the investigator should request that referring physicians approach their own potential subject patients to inform them of possible research, allowing them to make their own decision as to whether to contact the investigator, rather than asking the referring physician to identify specific potential subjects without their knowledge and prior approval. In some cases, the use of a flyer or information sheet, requiring the potential subject to initiate contact with the investigator, may be required. Such recruitment tools must not mislead potential subjects into believing that a study treatment is superior to current therapy. These flyers must have IRB approval. Avoiding coercion. This is especially critical in the common situation in which potential subjects are under the care of the investigator and may be concerned that their medical care may be jeopardized if they do not participate in research. A statement is generally placed into the ICF to explain the potential conflict which may exist between an investigator’s interest in a study versus an individual patient’s care. Additionally, coercion in the form of excessively high payment to subjects must be avoided. Amount and method of payment should be detailed in the ICF.
C.
Definition of “Minimal Risk”
“Minimal risk,” as defined by federal regulations, is a specifically described situation, “where the probability and magnitude of harm or discomfort anticipated in the proposed research are not greater, in and of themselves, than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.” It is important to note, in the preparation of IRB and FDA documents, that this may not necessarily correspond to the investigator’s intuitive concept of minimal risk. For example, even though a proposed research procedure may entail little physical or psychological risk to the subject, if that
Regulatory Issues
307
risk is more than encountered in “routine activities” for that subject, it may involve more than “minimal risk” and may require additional approval procedures. D. Statement of Emergency Care and Compensation All research that presents more than the strict definition of “minimal risk” as noted above requires a statement regarding “Emergency Care and Compensation for Injury.” This statement generally outlines the responsibility of the institution, such as a university, supporting the clinical center for a trial, in case of injury to a subject during a trial. Specific wording is generally required by the institution, and it may vary depending on whether the sponsor is a public (governmental— e.g., NIH) entity or a private (industry—e.g., pharmaceutical company). The sponsoring agency may also require specific wording, and the two may conflict. In clinical trials conducted by a pharmaceutical company in a university-based clinical center, agreement must be reached between the two organizations regarding financial responsibility for emergency care resulting from injury during the conduct of a trial. It may either be outlined in the ICF or in the clinical trial agreement (CTA) contract detailing the financial reimbursement from the pharmaceutical company to the university for the conduct of the trial. It is not acceptable to require that such emergency care be provided by third-party payers. E.
Reporting Adverse Events
An adverse event (AE) is defined as “an undesirable and unintended, although not necessarily unexpected, result of therapy or other intervention.” It may or may not be drug-related. Most IRBs require the reporting of such events, in writing, within a specified period of time, usually within 2–5 days, depending on the severity of the event. Typical adverse events involve injury, problems in the consent process, and other violations of protocol. The investigator is responsible for determining whether the event requires a change in the research protocol and/ or ICF, and recommending such changes when indicated. Subjects should be instructed to report any unusual medical occurrence to the PI as soon as possible. Determinations as to whether an incident is considered an adverse event, whether an AE is considered serious, and whether it is causally related to study drug are critical issues. Incidents considered non-AE include those that occur before or later than 4 weeks after dosing and worsening of disease that is considered within the normal variation of clinical course for the disease. Serious AEs are specifically defined by the FDA as those that result in or prolong current hospitalization, produce persistent or significant incapac-
308
Arnold
ity or disability, result in congenital anomaly, are life-threatening, or result in death. The severity of an AE is a separate issue. Other medical events not specifically meeting these criteria may be considered serious if they jeopardize the subject. If life-threatening, these must be reported to FDA by the pharmaceutical company within 7 days, and any SAE must be reported within 15 days. Causality for AEs is generally an issue for the FDA to decide, the guideline being a “reasonable possibility” that the event was caused by the drug.
F.
Investigator Training and Certification in Human Research
As of October 1, 2000, the NIH instituted a policy requiring all proposals for contracts and grants for research involving human subjects to certify that all key personnel have received education on the protection of human subjects. In many university research environments, this same requirement is applied by the IRB to all human research, including that sponsored by sources other than NIH. Both live teaching sessions at research institutions and web-based training programs are available to fulfill this requirement. Certification is time-limited, with duration for the current programs 2 years.
IV. FOOD AND DRUG ADMINISTRATION ISSUES A.
Investigational New Drug Exemption
Federal law prohibits the distribution of new drugs, biologics, and medical devices until the Food and Drug Administration (FDA) has reviewed clinical data and determined that a product is safe and effective for a specific use in human patients. Testing of a new drug requires an exemption from that law, and the sponsoring agent must apply to the FDA for an Investigational New Drug (IND) exemption before tests with human subjects may begin. If the investigator, rather than a pharmaceutical company, is the developer of the drug, that person or the sponsoring institution may be responsible for submitting an IND application to the FDA and providing the response to the IRB. In certain cases in which an approved drug is being tested for a slightly modified use, the FDA may issue a response indicating that an IND is not required. In general, however, approved drugs tested for new indications or at new doses require an IND exemption. Any study of such a drug must strictly adhere to the submitted protocol; changes must be submitted in writing and approved by the FDA and IRB.
Regulatory Issues
309
B. Classification of Trials The FDA classifies clinical trials according to the following schema: Phase 1. Initial introduction of an investigational new drug into humans. Designed to determine pharmacologic and metabolic actions, side effects with increasing doses (establish safe dosage), and gain early evidence for effectiveness. Study is very closely monitored, often involves healthy volunteers and a few patients with disease, usually 20–80 subjects. The goal is to gain sufficient data to design a valid Phase 2 study. Phase 2. Controlled study to evaluate effectiveness for a specific indication and to determine short-term risks and side effects. Closely monitored, usually several hundred subjects. Phase 3. Large-scale controlled study of the drug administered as it would be when marketed, often involving several hundred to thousand subjects. Performed after preliminary evidence of effectiveness has been obtained, it is designed to evaluate effectiveness, safety, and appropriate dosage, to provide a basis for labeling. Following this study, a sponsor typically applies for FDA approval to market the drug. Most large-scale clinical trials fall in this category. Phase 4. Concurrent with approval, the FDA may seek agreement from the sponsor to conduct post-marketing (Phase 4) studies to evaluate other doses, indications, or duration for drug use. C. Controlled Studies The FDA describes five categories of controls which may be used in drug trials: 1. 2. 3. 4. 5.
Placebo concurrent Dose-comparison concurrent No-treatment concurrent Active-treatment concurrent Historical
The FDA requires that the study design be adequate to obtain valid results, but it does not indicate a preference for type of control. There is misunderstanding regarding this issue. The use of placebo in a clinical trial may be unethical if there is proven effective therapy for the disease and it is withheld in favor of placebo. Many believe the FDA prefers placebo for comparison of effect, requiring it as a feature of randomized clinical trials. Placebo would be expected to have an advantage over the use of an active-treatment control, as the difference between effects of study drug and placebo may be greater than that between study drug and active-treatment control, thus allowing smaller sample size to
310
Arnold
demonstrate significance. The FDA has indicated that an active-treatment control trial “in which a finding of no difference between study drug and active-control would be considered evidence of effectiveness of the new agent.” It has also observed that such a design may be incapable in certain settings of allowing a conclusion to be drawn, due to the statistical difficulty of proving no difference. The FDA has not, however, as is commonly suggested, mandated the use of placebo for the study of new drugs. The risks of the use of placebo or no-treatment controls in specific diseases must be weighed against the risk that an activecontrol design may expose the subject to the risks of a trial without a reasonable chance to obtain useful information. D.
Financial Disclosure
Investigators who participate in clinical trials must disclose all financial ties to the sponsoring pharmaceutical company, whether relatively minor, such as limited honoraria for lectures; more substantial, such as ongoing consulting agreements, research grants or equipment purchases; or major, such as full-time employment by the company or equity interest in the company. Proprietary interest in a test product, such as patents, copyright, trademark, or licensing agreements must also be disclosed. Such disclosure must be reported to the IRB, the FDA, and may be required to be documented in the ICF. The FDA monitors data from investigators with such financial links extremely carefully. In certain universities, potential conflict of interest or the appearance of such conflict is highly scrutinized. A positive financial disclosure by an investigator may prompt detailed review for such conflicts, and a determination as to whether the investigator may take part in a sponsored trial or whether he or she must sever all such financial ties during the conduct of the trial may be made by the university. E.
Reporting Adverse Events
Specific procedures are outlined in the CFR for reporting the occurrence of an adverse event during a clinical trial. Investigators must familiarize themselves and comply with these regulations.
BIBLIOGRAPHY 1.
2.
18th World Medical Assembly. World Medical Association Declaration of Helsinki: Recommendations guiding medical doctors in biomedical research involving human subjects. Helsinki, Finland, June 1964. National Commission for the Protection of Human Subjects of Biomedical and Be-
Regulatory Issues
3. 4. 5.
311
havioral Research. The Belmont Report: Ethical principles and guidelines for the protection of human subjects of research. OPRR Reports, April 18, 1979. Code of Federal Regulations. Title 45 (45 CFR 46): Protection of Human Subjects. Revised June 18, 1991. Code of Federal Regulations. Title 21 (21 CFR): Food and Drugs. Revised April 1, 1999. FDA Information Sheets, October 1, 1995.
Index
Abducens nerve, 132, 133 Acepromazine, 24, 48, 52, 113 Acetazolamide, 51, 52, 121 Acinus, 227 Acridine orange, 210, 212 Ad libitum, 92, 130 Adeno-associated virus, 168, 170, 171, 174, 175, 176, 177, 180 Adenovirus, 62, 157, 158, 168, 170, 171, 172 Advanced Glaucoma Intervention Study, 281, 282, 297 Adverse event, 307, 308 Agarose, 98, 176 Age-related macular degeneration (ARMD), 110, 113, 199 Alpha-crystallin, 67 Alphagan (brimonidine), 50, 51 Amacrine cell, 7, 55, 66, 132, 192, 206, 219 Aminoguanidine, 159 Antiglaucoma, 50, 62 Antisense, 230 Apoptosis, 1, 85, 97, 98, 120, 169, 225, 226, 227, 230, 231, 234, 235, 241, 242, 243 Apoptosis initiation factor (AIF), 226, 227 Apoptosis protease activating factor-1 (APAF-1), 226 Apoptosome, 226
Applanation, 27, 40, 50, 51 Aquamount, 229, 231, 232, 233, 234, 235, 237 Aqueous humor, 23, 32, 33, 37, 39, 44, 45, 47, 57, 60, 72, 154 Arcuate, 62 Area centralis, 195, 213, 214, 218, 219 Argon laser, 32, 49, 50 Arrestin, 99 Artificial CSF, 209, 210, 211, 212 Ascorbate, 101, 102, 103 Ascorbic acid, 101, 102, 103 Aspartate, 2, 3 Astrocyte, 4, 67, 69, 111, 119, 175 Autofluorescence, 231, 232 Axoclamp, 213 Axon, 6, 13, 14, 16, 17, 18, 19, 23, 32, 43, 63, 66, 67, 69, 130, 131, 132, 133, 140, 141, 145, 146, 192, 206, 207, 209, 212, 213, 214, 215, 217, 242, 273, 275, 277, 278, 283 Axotomy, 133, 146, 169, 172, 242 Azimuthal, 197 Bandpass, 260 Basic fibroblast growth factor, 169 Bax, 226, 233, 237 Bcl-2, 99, 169, 237 Benzolamide, 121 Betaxolol (Betoptic), 155 313
314 Biomicroscopy, 51, 68 Bipolar cell, 207, 247, 253, 269, 277 Bisbenziamide, 234 Bistratified cell, 207, 275, 277 Blepharostat, 255, 257 BODIPY, 235 Brain-derived neurotrophic factor (BDNF), 3, 120, 156, 169, 217, 218, 219, 220 Bregma, 16, 19 Brimonidine (Alphagan), 50, 51, 62, 162, 299 Brn3a, 192 Brn3b, 192 Bullous, 113, 119, 122, 123 Buprenorphine, 68, 113 CA1, 147 Calbindin, 118, 192 Cannula, 38, 41, 42, 45, 114, 115, 144 Canthi, 255 Canthotomy, 36, 48 Capsid, 175 Carbocyanine, 133, 208 Carbonic anhydrase, 118, 121, 191 Carotenoids, 102 Caspase, 169, 226, 232 Cautery, 25, 68 CCD, 215 CD34, 179 Cecum, 161 Cell culture, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 173, 178, 234, 242, 244, 245, 246 Cell density, 4, 7, 190, 195, 196, 197, 200, 202, 219 Central retinal artery, 131 c-fos, 87, 99, 100 cGMP, 168 Chloromethyl-tetramethylrosamine (CMTMR), 235, 236, 237, 238 Chondroitin, 67 Choriocapillaris, 110, 116 Choroid, 62, 66, 110, 111, 123, 131, 135, 136, 148, 154, 162, 210 Chromogen, 233 Chymotrypsin, 72
Index Ciliary neurotrophic factor (CNTF), 3, 120, 169, 250 Cilioretinal, 299 c-jun, 100 Clinical trials, 99, 273, 274, 276, 279, 280, 281, 283, 284, 291, 292, 293, 294, 295, 296, 297, 298, 299, 303, 304, 305, 307, 309, 310 Cobalt chloride, 227, 229, 230 Co-culture, 4 Collaborative Initial Glaucoma Treatment Study (CIGTS), 281, 282, 283 Collagenase, 210 Cone photoreceptors, 62, 116, 117, 118, 121, 124, 190, 191, 192, 195, 199, 200, 251, 264, 265, 275 Confocal, 52, 69, 192, 193, 201, 214, 215, 228, 232, 238 Conjunctiva, 14, 26, 37, 257 Contamination, 8, 9, 154, 173, 177, 253, 261 Cornea, 5, 14, 27, 40, 42, 50, 51, 55, 60, 61, 114, 134, 136, 154, 155, 158, 250, 254, 255 Cosmid, 174 Cotransfection, 173, 176, 177, 179 Counterphase, 278, 279 Counterstain, 232 Coverslip, 134, 213, 229, 231, 232, 233, 234, 235, 237 Coxsackievirus, 172 Cribriform, 71 Cross-action forceps, 15 Cryosection, 191, 230 CsCl, 176 Culture media, 2, 3, 245 Cy5, 237 Cyanoacrylate, 117, 157 Cyclodialysis, 57, 60 Cyclopentolate, 113 Cynomologus monkey, 47, 48, 57, 61, 66, 69, 70, 116 Cystoid macular edema, 110, 122, 303 Cytochrome c, 226
Index Cytochrome oxidase, 66 Cytomegalovirus, 172, 178 Cytotoxicity, 13, 171, 174 DAPI, 191, 232, 234, 238, 242, 243, 246, 247 Dehydroascorbic acid, 102 Dehydrogenase, 96, 230 Dendritic, 63, 207, 208, 209, 212, 214, 215, 216, 277, 278 Depo-Medrol, 68 Dexamethasone, 68, 69, 70, 130 Diaphorase, 66 DiAsp, 17, 19, 133, 139, 140, 142, 143, 144 Dichroic, 243, 244 Dihydroethidium (HEt), 243, 247 DiI, 208 Dimethyl sulfoxide (DMSO), 132, 202, 236, 244, 245 Dimethylformamide, 133 Dimethylthiourea (DMTU), 101, 102, 103 Disector, 199, 201 DNAse, 6 Docosahexaenoic acid, 3 Dorzolamide (Trusopt), 50, 51, 62 Dremel, 35 tool, 35 Dulbecco’s modified Eagles media (DMEM), 2, 3, 5 Dura, 17, 19, 68, 135, 137 dUTP, 229 Earclip, 255 Early Manifest Glaucoma Trial (EMGT), 281, 282 Eclampsia, 111 Elastin, 67 Electrocardiogram, 250, 254 Electrode, 8, 19, 211, 212, 213, 251, 254, 255, 257, 267, 268 Electroretinogram (ERG), 54, 55, 95, 96, 97, 118, 119, 120, 145, 249, 250, 251, 252, 253, 254, 255, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 298
315 Embolization, 123 Encapsidation, 168, 171, 174, 176, 178 Endonuclease, 96, 227 Endophthalmitis, 156 Endothelin, 69 Enucleation, 154 Epifluorescence, 211, 212 Epiretinal, 122 Episclera, 24, 25, 26, 31, 32, 33, 34, 36, 39, 40, 45 episcleral vein, 24, 26, 31, 32, 33, 34, 36, 45 Episomal, 170, 172 Equiluminant, 278 Ethidium, 191, 245 Euthermic, 93, 94 Excitotoxicity, 3, 56, 61, 235, 279, 296, 299 Explant, 2, 5, 6, 9 cultures, 5, 6 Extracapsular, 114 Extracellular matrix, 1, 4, 5, 6, 67 Exudative, 110, 111, 199, 200 Fetal Bovine Serum (FBS), 5, 6 Fibroblast growth factor (FGF), 3, 120, 169 Fibronectin, 4 FK506, 161 Flavoprotein, 226 Fluorescein, 60, 117, 122, 139, 162 FluoroGold, 16, 17, 28, 132, 133, 139, 140, 141, 142, 144 Fluorophore, 16, 28, 132, 208, 237 Fluorophotometry, 57, 60 Flupirtine, 155, 156 Food and Drug Administration (FDA), 294, 303, 305, 306, 307, 308, 309, 310 Forceps, 4, 5, 14, 15, 25, 34, 36, 37, 38, 209, 210, 211 Forskolin, 62 Fovea, 52, 56, 63, 64, 86, 110, 121, 124, 190, 195, 196, 197, 202, 207, 211, 213, 214, 268, 274 Fundus, 51, 52, 53, 68, 113, 114, 134, 135, 136, 138, 144, 146, 155, 198
316 Funduscopy, 68 Fyrite, 9 Ganzfeld, 54, 95, 255, 257 GAPDH, 230, 231, 232, 237, 238 Gavage, 102, 159, 160, 161 Gavestinel, 296 GDx, 52, 298 GelMount, 229, 231, 232, 233, 234 Gene transfer, 28, 157, 167, 168, 169, 171, 173, 174, 175, 176, 178, 180 Glaucoma, 1, 23, 24, 28, 31, 32, 44, 45, 47, 48, 49, 50, 52, 53, 54, 55, 56, 57, 60, 61, 62, 63, 65, 66, 67, 69, 72, 147, 148, 153, 168, 169, 209, 214, 215, 273, 274, 275, 277, 278, 280, 281, 282, 283, 284, 294, 297, 298 hemifield test, 282 Glia, 6, 32, 51, 66, 67, 111, 119, 169, 172, 175, 253 fibrillary acidic protein (GFAP), 67, 119 -derived neurotrophic factor (GDNF), 120 Gliosis, 43, 120 Gliotic, 123 Glucocorticoid, 69, 70 Glutamate, 2, 3, 61, 119, 120, 235 Glutamine synthetase, 97, 119 Glyceraldehyde, 230 Glycine, 229, 231, 235, 296 Glycoprotein, 3 Golgi, 208 Gonioftal, 134 Goniolens, 48 Gonioscopic, 61 Green fluorescent protein (GFP), 178, 179 Growth factors, 1, 3, 6, 100, 155 Hanks balanced salt solution (HBSS), 5, 6, 8, 237, 245 Heidelberg retinal tomograph (HRT), 52, 298 Hemacytometer, 6
Index Hematopoietic, 178, 179 Hemifield, 282, 283 Hemiretinal, 268 Henle, 202 Heparan, 67, 176 HEPES, 245 Herpesvirus, 177 HEt, 242, 243, 244, 245, 247 Heterologous, 180 Heterozygote, 99 High-Pass Resolution Perimetry (HPRP), 277, 278, 279, 280 Histochemistry, 118, 191 Hoechst dye, 8, 234 Homodimer, 191 Hyaluronic acid, 115, 117 Hyaluronidase, 112 Hydrogen peroxide, 228 Hydroxypropylmethylcellulose, 134 Hyperoxia, 120, 121 Hyperthermia, 91, 92, 93, 94 Hypertonic saline, 31, 32, 33, 34, 45 Hypertrophic, 111, 119 Hypotony, 33, 39, 40, 44, 49 Icosahedral, 171 Imaging software, 246, 247 MetaFluor, 244, 246 Metamorph, 193 Immune response, 23, 28, 171, 172, 173, 174, 176, 178 Immunoblotting, 232 Immunocytochemistry, 118, 132, 227, 230, 234, 236 Immunodeficient, 172, 179 Immunohistochemistry, 7, 119 Immunoreactivity, 65, 66 Immunoreagents, 202 Immunostaining, 66 Immunosuppression, 170, 173 In situ end labeling, 227, 228, 230, 231, 234, 235 Incubator settings, 5, 6, 7, 9, 236, 243, 244, 245 Infarction, 147 Inner plexiform layer, 54, 132, 146
Index Innervation, 68 Insertional, 175, 178 Institutional review board (IRB), 292, 295, 303, 305, 306, 307, 308, 310 Integrin, 172 Interlaminar, 277 International Society for Clinical Electrophysiology of Vision (ISCEV), 254, 258, 260, 261, 264, 265, 266, 268 Interneurons, 192 Interphotoreceptor, 118 Intracameral, 69, 71 Intraconal, 68 Intraocular pressure (IOP), 23, 24, 26, 27, 28, 31, 32, 33, 36, 39, 40, 41, 42, 43, 44, 45, 48, 49, 50, 51, 52, 53, 54, 57, 60, 62, 63, 65, 66, 67, 68, 69, 70, 71, 72, 114, 122, 131, 133, 134, 135, 144, 145, 146, 147, 153, 169, 273, 281, 294, 297, 298, 299 Intraorbital, 14, 17, 131, 133, 140, 143 Intraperitoneal, 24, 130, 158, 161, 162 Intraretinal, 209, 212, 214 Intravenous, 123, 158, 162, 163 Intravitreal, 56, 112, 120, 121, 153, 155, 156, 157, 173, 175, 178, 250 Inverted terminal repeat, 174, 176 Iodixanol, 176 Iodoantipyrine, 61 Ionophore, 246 IOP measurement, 28, 40, 41, 70 Iridocyclitis, 49, 72 Iridolenticular, 51 Ischemic Optic Neuropathy Decompression Trial (IONDT), 296, 297 ISEL, 227, 228, 230, 231, 234, 235 Isoeccentricity, 198 Isofluorane, 68 Kainic acid, 7 Ketamine, 24, 48, 50, 51, 52, 113, 130, 136, 254 Koniocellular, 275
317 Lamina cribrosa, 50, 51, 64, 67 Laminin, 4, 5 Laser, 32, 48, 49, 50, 51, 52, 57, 60, 61, 62, 63, 69, 228, 232, 234, 237, 298 Latanoprost (Xalatan), 51, 62 Lateral geniculate nucleus (LGN), 4, 7, 14, 63, 64, 65, 66, 131, 275, 277, 278 Lectin, 4, 192 Lens, 5, 38, 40, 42, 48, 51, 72, 114, 116, 136, 146, 154, 156, 246, 254, 255, 257, 258, 268, 279 Lentivirus, 168, 170, 177 Light, dark cycle, 39 damage, 9, 85, 86, 87, 88, 89, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103 exposure, 9, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 123 Limbus, 19, 24, 33, 34, 35, 36, 37, 38, 40, 113, 114, 134, 135, 136, 139, 156 Linoleic, 3, 5 Lipophilic, 17, 19, 102, 133 Long posterior ciliary artery, 33, 36 Long-terminal repeats, 177, 179, 180 Macaca mulatta, 48 Macula, 58, 59, 64, 109, 110, 113, 116, 117, 118, 123, 168, 169, 190, 193, 196, 197, 202, 266, 267, 269, 274, 297, 298 edema, 110 Maculopathy, 191, 199, 200 Magnocellular (M-cell), 63, 65, 275, 277, 278 Mannitol, 51, 52 Media supplements, 3, 8 Memantine, 61, 279, 296, 299 Metabotropic, 121 receptors, 121 Metamorphopsia, 110 Methohexital sodium, 48, 52
318 Methylene blue, 208 Methylprednisolone, 68 Microelectrode, 19, 210 Microgel, 69, 71 Microglia, 140, 142, 145, 175 Microinstruments, 35 Microneedle, 34, 35, 36, 44 Microscissors, 24 Microspheres, 69, 71, 123 Microsyringe, 136 Microvessel, 26 Minipumps, 69, 173 Mitochondria, 95, 225, 226, 227, 234, 235, 236, 237, 241, 242, 243, 244, 245, 246, 247 permeability transition pore, 236, 241 Mitomycin, 62 MitoTracker, 235, 236, 237 Monolayer cultures, 5 Morphometry, 95, 96, 97, 192, 201 Motoneurons, 132, 133 Mu¨ller cell, 111, 119, 120, 169, 170, 178, 192, 253 Multifocal ERG, 56, 57, 58, 62, 69, 118, 249, 266, 268, 269 Multifocal VEP, 57 Multipolar, 7 Mycoplasma, 8 Mydriatic, 52, 87, 88, 94 Myelin, 43, 116 Myocilin, 69 NADPH, 66 ND4, 212 Neovascularization, 26, 51, 110 Nernstian, 236 Nerve fiber analyzer, 298 Nerve growth factor, 3, 230, 235 Neurite, 2, 6, 7 Neurobasal, 3, 244 Neurobiotin, 212, 213, 215 Neurodegeneration, 47, 116, 168, 176 Neurofibrillar, 208 Neurofilament, 44 Neurogenetic, 192
Index Neuroprotection, 1, 13, 16, 23, 28, 29, 31, 32, 44, 45, 61, 86, 100, 101, 103, 116, 120, 121, 123, 129, 130, 132, 136, 145, 146, 147, 148, 153, 155, 156, 157, 158, 159, 161, 162, 167, 168, 169, 170, 171, 172, 173, 175, 178, 179, 180, 189, 192, 193, 205, 206, 208, 215, 216, 217, 219, 220, 221, 273, 274, 276, 280, 284, 291, 294, 296, 297, 298, 299, 303 Neurotoxic, 249 Neurotracer, 16, 17 Neurotrophic factors, 100, 101, 129, 169, 173, 235 Neurotrophin, 3, 28, 100, 101, 168, 169, 172, 173 neurotrophin-3 (NT-3), 3 neurotrophin-4 (NT-4), 3 Nipradilol, 154, 155 Nitrergic, 66 Nitric oxide synthase (NOS), 159 NMDA, 56, 61, 296, 299 Nomarski, 193 Nonarteritic anterior ischemic optic neuropathy (NAION), 292, 296, 297 Nonresponder, 69, 70 Normal tension glaucoma (NTG), 31, 281 Normotensive, 58, 59 Normoxia, 120 Ocular Hypertension Treatment Study (OHTS), 279, 281, 283 Oculoplastic, 67 Oligodendrocyte, 175 Oligonucleosome, 227 Oligonucleotides, 230 Ophthalmic vessels ligature, 133, 134, 135, 136, 137, 138, 144, 146, 148 Ophthalmoscopy, 52, 67, 68, 69 Opponency, 277 Opsin, 7, 62, 118 Optic nerve crush, 15, 16, 161, 218, 220
Index Optic nerve head, 14, 25, 32, 51, 56, 61, 64, 67, 69, 131, 137, 155 Optic nerve transection, 28, 53, 55, 57, 67, 68, 69, 133, 140 Optical coherence tomograph (OCT), 52, 298 Ora serrata, 209 Orbitotomy, 67, 68 Oscillatory potentials (OPs), 54, 56, 251, 252, 253, 254, 258, 259, 260, 261, 263, 265, 266, 269 Outer nuclear layer (ONL), 95, 96, 119, 148, 191, 200, 201 Outer plexiform layer (OPL), 54, 122, 146 Outflow, 23, 24, 32, 33, 39, 40, 44, 45, 49, 57, 60, 69, 112, 113 Overexpress, 99 Oversampling, 195 Palpebral, 48 Paracrystalline, 201 Parafilm, 229, 231, 232, 235 Paraformaldehyde, 18, 19, 139, 158, 213, 234, 237 Parafovea, 63, 190, 198 Parvalbumin, 66 Parvocellular (P-cell), 63, 65, 66, 275, 277, 278 Parvoviridae, 174 Pattern electroretinogram (PERG), 266, 267 PC12, 230 pClamp6, 217 Penumbra, 147 Pericyte, 111 Perifoveal, 59 Perimacular, 58, 59, 66 Perimetry, 52, 56, 57, 197, 274, 276, 277, 278, 280, 281, 283, 298 frequency-doubling technology (FDT), 278, 279, 280, 283 Perineural, 69 Periocular, 155 Peripapillary, 52, 53 Perivascular, 37 Permeability transition pore, 236, 241
319 Peroxidase, 233 PGF2α, 62 PGF2α-1-isopropylester, 50, 51 Phagocytosis, 63, 96, 99, 140, 225 Phenobarbital, 154 Phenylephrine, 52, 113 Phosphodiesterase, 168 Photoactivation, 123 Photochemically, 123 Photocoagulator, 123 Photoisomerized, 93 Photopic, 54, 56, 95, 119, 120, 251, 252, 253, 254, 258, 259, 260, 261, 263, 266 Photopigment, 261 Photoreceptor cultures, 7 Photostasis, 87 Photothrombosis, 123 Phototoxicity, 9, 85, 86, 87, 88, 89, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103 Pia mater, 132 Pigmentation, 87, 88 Pilocarpine, 62, 294 Planimetric, 190 Pneumotonography, 57 Pneumotonometry, 28, 51 Polarimeter, 52 Polyacrylamide, 69 poly-l-lysine, 3, 248 poly-l-ornithine, 3, 5 Polymerase chain reaction, 173 Potentiometric dye, 235, 236 Prelabel, 28, 208 Primary visual cortex, 19 Procaspase, 226 Proparacaine, 27 Propargylamines, 230 Proteinase, 228, 230, 235 Proteoglycan, 4, 67, 157, 176 Punctate, 140 Pyranine, 212 Pyruvate, 2 Radiolabeling, 154, 177 rds mouse, 168, 169, 250
320 Reactive oxygen species (ROS), 95, 96, 99, 236, 241 Reaggregation, 7 cultures, 7 Recti muscle, 25, 26, 68, 132, 133 Retina, 2, 4, 5, 6, 14, 16, 18, 19, 28, 54, 56, 58, 61, 62, 63, 64, 66, 68, 85, 86, 87, 95, 96, 97, 98, 99, 100, 101, 102, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 129, 130, 131, 132, 133, 135, 136, 138, 139, 141, 142, 143, 144, 145, 146, 147, 148, 153, 154, 155, 156, 157, 162, 167, 168, 169,172, 173, 175, 176, 178, 180, 189, 190, 191, 192, 193, 194, 195, 196, 197, 200, 201, 202, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 228, 231, 232, 234, 242, 249, 250, 251, 253, 255, 257, 258, 260, 265, 266, 267, 269, 274, 275, 277, 278, 283, 299 blood flow, 61, 62, 69, 131, 133, 134, 135, 136, 138, 146, 148, 155 detachment, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 156, 169 ganglion cell (RGC), 4, 7, 8, 13, 14, 15, 16, 17, 18, 19, 23, 28, 54, 57, 59, 61, 63, 129, 130, 131, 132, 133, 134, 136, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 161, 162, 169, 170, 172, 175, 192, 205, 208, 209, 214, 217, 220, 231, 241, 242, 243, 244, 245, 246, 247, 254, 266, 275, 276, 299 cultures, 7 ischemia, 69, 110, 129, 133, 134, 135, 136, 138, 140, 144, 146, 147, 148, 235
Index pigment epithelium (RPE), 5, 94, 95, 96, 99, 102, 111, 113, 116, 117, 118, 119, 121, 122, 123, 170, 172, 175, 178, 179, 199, 200 reattachment, 110, 114, 116, 117, 118, 121, 123 slice cultures, 6, 7 whole-mount, 18, 141, 142, 189, 191, 192, 193, 196, 201, 202, 205, 217, 220 Retinaldehyde, 119 Retinitis pigmentosa, 100, 168, 250, 297, 298 Retinocollicular, 130 Retinogeniculate, 64 Retinopexy, 110 Retinorecipient, 132, 192 Retinotomy, 117 Retinotopic, 56 Retrobulbar, 15, 61, 113 Retrograde labeling, 16, 130, 132, 133, 139, 140, 192, 208, 209, 219, 242 dyes, DAPI, 191, 232, 234, 238, 242, 243, 246, 247 DiAsp, 17, 19, 133, 139, 140, 142, 143, 144 DiI, 208 FluoroGold, 16, 17, 28, 132, 141, 142 Retrolaminar, 66 Rhegmatogenous, 110, 111, 112, 117, 118, 121, 122, 123 Rhodamine, 236, 237 Rhodopsin, 62, 87, 88, 89, 92, 93, 94, 96, 97, 99, 100, 102, 118, 119, 168, 169, 175, 261 Ribozyme, 168 RNAse, 228, 231, 232, 235 Rod photoreceptors, 8, 117, 118, 121, 190, 191, 198, 199, 200, 251, 255, 277 Rose bengal, 122, 123 Rostral, 16 Royal College of Surgeons (RCS) rat, 87, 88, 96, 99, 169 RPMI, 3
Index Sampling, 61, 193, 194, 195, 196, 199, 200, 213, 219 Schlemm’s canal, 23, 27, 33, 34, 35, 38, 63, 71 SCID, 179 Sclera, 5, 25, 87, 113, 131, 136, 155, 156, 157, 158, 210, 211, 213, 217 Sclerosing, 34 Scotoma, 284 Scotopic, 54, 55, 95, 119, 120, 251, 252, 258, 260, 261, 262, 263, 264, 266 Secondary degeneration, 13, 283 Selenite, 3 Sialic acid, 3 Sigmoidal, 263, 264 Sildafil, 250 StatPac, 281, 282 Stereology, 193 Stereotactic, 16, 19 Subconjunctival, 26, 62, 157, 158 Subretinal, 110, 111, 112, 113, 114, 115, 116, 117, 121, 122, 172, 173, 175, 178, 179 Superior colliculi, 14, 16, 19, 131, 132, 141, 142, 172, 208, 242 Superoxide, 241, 242, 243, 244 Swedish Interactive Thresholding Algorithm (SITA), 280 Synaptogenesis, 1 Synechiae, 51, 63, 72 Syneresis, 117 Tectum, 14 Tenon, 37, 155, 157, 158 Terminal deoxynucleotidyl transferase (TdT), 227, 228, 229, 230, 235 Tetrodotoxin, 56 Thermistor, 244 Thiols, 236 Timolol (Timoptic), 50, 51, 62, 155, 299 TMRM, 236 Tonography, 57 Tonometry, 40, 41, 42, 43, 50, 51
321 Tonopen, 27, 39, 40, 41, 42, 43, 51 TopSS, 52 Trabecular, 32, 33, 34, 38, 39, 48, 62, 63 Trabeculectomy, 62 Trans-corneal, 153, 155, 157 Transfection, 174, 177, 179 Transferase, 227, 228 Transferrin, 3 Transgenic, 86, 89, 99, 156, 157, 158, 167, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180 Trans-scleral, 33, 153, 157 TRITC, 237 Trituration, 6 Trophic factor, 129, 235 Trophic withdrawal, 235 Tropicamide, 52, 113, 134, 268 Trusopt (dorzolamide), 50, 51 Trypsin, 6, 8, 9 TUNEL, 118, 120, 227 Tween, 231, 232, 237 Ultrastructure, 43, 168 Ultrathin, 192, 201 Undersampling, 195 UniBlitz shutter, 244 Uveoscleral, 49, 57, 60 Valinomycin, 246 Vanadate, 62 Vannas scissors, 34 Vasoconstriction, 155 Vasodilation, 155 Vectastain, 213 Venous plexus, 24, 25, 27, 33 VERIS, 268, 274 Vervet, 48 vesicular stomatitis virus (VSV), 179 Vibratome, 7 Virion, 177, 179 Viscoelastic, 61, 69 Visual evoked potential (VEP), 18, 57, 298 Vitrectomy, 110, 111, 114, 117, 122 Vitreoretinal, 117, 122 Vitreoretinopathy, 110, 111, 119
322 Vitreous, 5, 112, 122, 156, 218 Vmax, 263,
Index 18, 19, 53, 61, 110, 111, 113, 114, 116, 117, 119, 131, 136, 139, 154, 155, 172, 173, 200, 209, 210,
Xalatan (latanoprost), 51, 62 XIAP, 169 Xylazine, 24, 113, 130, 136, 254 Xylene, 202, 228, 233 Xylocaine, 113
264
Wistar, 23, 24, 26, 28
YOYO-1, 228, 230, 234, 235, 236, 237, 238
About the Editors
LEONARD A. LEVIN is Associate Professor of Ophthalmology and Visual Sciences, Neurology, and Neurological Surgery at University of Wisconsin Medical School–Madison, where he studies mechanisms of retinal ganglion cell death. He primarily focuses on the role axonal damage plays in inducing loss of retinal ganglion cells, an area common to both neuro-ophthalmology and glaucoma, and is the principal investigator on a National Eye Institute–supported grant to study this topic. Dr. Levin is a Research to Prevent Blindness Dolly Green Scholar and an associate editor of the Archives of Ophthalmology. He received the A.B. degree (1980) in applied mathematics, the Ph.D. degree (1988) in neurobiology, and the M.D. degree (1988) from Harvard University, Cambridge, Massachusetts. He did a residency in ophthalmology and a fellowship in neuro-ophthalmology at the Massachusetts Eye and Ear Infirmary. ADRIANA DI POLO is Assistant Professor of Pathology and Cell Biology at the Universite´ de Montre´al, Quebec, Canada. A major focus of her laboratory is to develop gene therapy strategies to promote cell survival and regeneration of adult retinal ganglion cells, the neuronal population that dies in glaucoma. She received the B.Sc. degree (1989) from the Universidad Central de Venezuela, Caracas, and the Ph.D. degree (1995) from the University of California School of Medicine, Los Angeles.