University of Southern California Dissertations and Theses

Gene therapy for motor neuron degeneration in murine tissue culture and transgenic mouse models of familial amyotrophic lateral sclerosis

(USC Thesis Other)

Gene therapy for motor neuron degeneration in murine tissue culture and transgenic mouse models of familial amyotrophic lateral sclerosis

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. U M I films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UM I a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, M l 48106-1346 USA 800-521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. R eproduced with permission of the copyright owner. Further reproduction prohibited without permission. GENE THERAPY FOR MOTOR NEURON DEGENERATION IN MURINE TISSUE CULTURE AND TRANSGENIC MOUSE MODELS OF FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS ©2002 by Brian Francis Roehmholdt A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements of the Degree DOCTOR OF PHILOSOPHY (CELL AND NEUROBIOLOGY) May 2002 Brian Francis Roehmholdt Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UM I Number: 3073841 __ ___ ( f i ) UMI UMI Microform 3073841 Copyright 2003 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA The Graduate School U niversity Park LOS ANGELES, CALIFORNIA 90089-1695 This dissertation, w ritten b y Brian Francis Roehmholdt_________________ Under the direction o f h .ls.. Dissertation Com m ittee, and approved b y a ll its members, has been presented to and accepted b y The Graduate School, in p a rtia l fulfillm ent o f requirem ents for the degree o f DOCTOR OF PHILOSOPHY o f Graduate Studies Date l DISSER TA TION COMMITTEE Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Brian Francis Roehmholdt Leslie P. Weiner, M.D7 Judy A. Garner, Ph.D. GENE THERAPY FOR MOTOR NEURON DEGENERATION IN MURINE TISSUE CULTURE AND TRANSGENIC MOUSE MODELS OF FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS (FALS) Since the identification nearly a decade ago of superoxide dismutase-1 (SOD I) mutations in a subset of FALS cases intense research has provided insight into the pathogenic mechanisms involved in selective motor neuron death. Conflicting reports have produced evidence both for and against apoptosis as the final cell death pathway. This study first showed that hydrogen peroxide (H2O2 ) might produce a caspase- independent form of apoptosis in a motor neuron cell line (Mn-1), providing a model of motor neuron death with similarities to cell death in FALS. Next Mn-1 cells transfected with a mutant SOD 1 were shown to have an increased susceptibility to oxidative-damage induced apoptosis compared to controls. This supports the role of oxidative damage in the pathogenesis of FALS, and also lends credence to the theory that FALS cell death is apoptotic. Third, it was shown that a gene therapy vector containing human bcl-2 (Ad- bcl-2) was able to infect differentiated MNGA cells and protect them from apoptotic cell death induced by treatment with H2O2 Finally, treating FALS transgenic mice with intra-muscular injections of Ad-bcl-2 tested the protective ability of Ad-bcl-2 in vivo. Treated mice had a significant delay in disease onset compared to controls. The effect of Ad-bcl-2, however, did not extend the course of disease or prevent motor neuron loss. Based on the theory that disease onset in FALS is related to oxidative damage while I Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. isc progression is the result of glutamate excitotoxicity, the effect of hcl-2 in delaying onset but not preventing progression could be explained by a direct anti-oxidant function of bcl-2 rather than its usual anti-apoptotic role. The data indicate that while bcl-2 is a promising therapy for treatment of motor neuron degeneration in FALS, it will be necessary to use therapy that affects other elements of oxidative/excitotoxic-induced cell death along with bcl-2 to both delay onset and hinder disease progression. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DEDICATION AND ACKNOWLEDGEMENTS This dissertation was possible because of the support of many friends, colleagues, and loved ones. Special thanks to the members of my dissertation committee for their guidance and the St. Matthias High School family who endured my research while I worked there as Vice Principal and Principal. This dissertation is dedicated to Haley Varela, my beloved goddaughter. May you also be inspired to attain a terminal degree in your field of study. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DEDICATION To my fiancee, Katie Page, who endured late nights at the lab and late nights at the computer, along with countless fits of near despair and a final moment of joy when I hit save for the last time. Thank you for your support, encouragement, and patience. And to my parents, Gary and Kathy Roehmholdt, who have always stood by me and encouraged me to pursue all of my dreams, no matter where they take me. I would never have made it this far without your support and guidance. Thank you. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS I would like to thank my co-advisors, Dr. Leslie P. Weiner and Dr. Judy A. Gamer, for their unending support and guidance over the past four years. You were always willing to take time to teach me, even when I thought that I already knew the answer. I would also like to thank the remaining members of my committee: Dr. Thomas McNeill, Dr. Hans-Jiirgen Fiille, and Dr. Nori Kasahara. From all of you I have learned invaluable lessons in science and research that will help me throughout my career. Finally I would like to thank the innumerable people whom I have pestered over the years for one reason or another, especially the staff of the Department of Neurology and the Department of Cell and Neurobiology. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS Pape DEDICATION ii ACKNOWLEDGEMENTS iii LIST OF TABLES vii LIST OF FIGURES viii CHAPTER 1: BACKGROUND AND SIGNIFICANCE Introduction 1 Neurodegenerative disorders and cell death 2 Apoptosis, caspases, and the bcl-2 family 4 Assays for apoptosis 11 Amyotrophic lateral sclerosis and excitotoxicity 12 FALS and SOD 1 14 Motor neurons and models of neurodegeneration 16 Motor neuron markers 21 Treatment of ALS 26 Overview of thesis work 29 CHAPTER 2: CHARACTERIZATION OF HYDROGEN PEROXIDE- INDUCED CELL DEATH IN THE Mn-1 MOTOR NEURON CELL LINE INTRODUCTION 33 RESULTS Mn-1 cell culture and differentiation 38 RT-PCR for motor neuron markers 38 RT-PCR for apoptotic markers 44 Detection of apoptosis by flow cytometry 48 Analysis of DNA ladder formation 48 Analysis of apoptosis by immunoblot for activated caspase-3 SO Analysis of apoptosis by colorimetric enzyme assay SO Dose-response relationship for H2O2 by flow cytometry 53 RT-PCR for apoptotic markers after H2O2 treatment 56 DISCUSSION 59 iv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EXPERIMENTAL PROCEDURES Cell culture 72 RT-PCR 72 Induction of apoptosis 74 Annexin V-propidium iodide staining 74 Assessment of DNA ladder 75 Immunoblot for caspase-3 activation 76 Colorimetric enzyme assay for caspase-3 activation 77 CHAPTER 3: BCL-2 INHIBITS CASPASE-INDEPENDENT APOPTOSIS IN A TISSUE CULTURE MODEL OF FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS INTRODUCTION 78 RESULTS Transient expression of WT-SODI and G93 A-SOD1 86 Stable expression of WT-SOD 1 and G93 A-SOD 1 86 Expression of apoptotic markers in MNWT and MNGA cell lines 89 Trypan blue assay for cell death 89 DNA ladder in MNWT and MNGA cells 95 Effect of hydrogen peroxide (H2O2) on expression of apoptotic markers 95 Expression of a human bcl-2 transgene from an adenoviral vector in Mn-1, MNWT, and MNGA cells 99 Trypan blue assay for the effects of adenoviral- mediated bcl-2 expression on H202-induced cell death 99 RT-PCR for the effects of adenoviral-mediated bcl-2 expression on apoptotic markers during H2 0 2 -induced cell death 102 DISCUSSION 106 EXPERIMENTAL PROCEDURES Mn-1 cell culture and differentiation 115 Transient and stable transfection of Mn-1 cells 115 RT-PCR 117 Immunoblot procedure using enhanced chemifluorescence 118 Induction of apoptosis 119 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Trypan blue assay for cell viability Assessment of DNA ladder Infection of cells with adenoviral vector 119 120 121 CHAPTER 4: AN ADENOVIRAL-BCL-2 VECTOR DELIVERED VIA INTRA-MUSCULAR INJECTION RESULTS IN DELAYED ONSET OF DISEASE IN A MOUSE MODEL OF FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS INTRODUCTION 123 RESULTS Disease progression in untreated G93A mice 129 Immunohistochemistry for expression of human bcl-2 134 Safety of Ad-lacZ and Ad-bcl-2 in control mice 137 Assessment of disease onset and progression in bcl-2 treated mice 141 Determination of motor neuron number in treated and control mice 146 DISCUSSION 150 EXPERIMENTAL PROCEDURES Transgenic mice and experimental design 157 Intra-muscular injection of adenoviral vectors 158 Clinical symptoms 159 Collection of spinal cords 159 Histology 160 Immunohistochemistry for human bcl-2 160 CHAPTER 5: SUMMARY AND CONCLUSIONS 162 REFERENCES 168 APPENDIX A 178 APPENDIX B 183 vi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 RT-PCR for expression of motor neuron cell surface receptors in non-differentiated and differentiated Mn-1 cells RT-PCR for expression of intracellular motor neuron markers in non-differentiated and differentiated Mn-1 cells RT-PCR for expression of motor neuron developmental markers in non-differentiated and differentiated Mn-1 cells RT-PCR for expression of pro-apoptotic markers in non- differentiated and differentiated Mn-1 cells RT-PCR for expression of anti-apoptotic markers in non- differentiated and differentiated Mn-l cells Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 3.11 Figure 3.12 Figure 3.13 Figure 4.1 Figure 4.2 Figure 4.3 RT-PCR for anti-apoptotic markers in Mn-1, MNWT, and MNGA cell lines 91 Trypan blue assay for non-differentiated Mn-1, MNWT, and MNGA cell viability after treatment with hydrogen peroxide 93 Trypan blue assay for differentiated Mn-1, MNWT, and MNGA cell viability after treatment with hydrogen peroxide 94 DNA ladder formation in differentiated MNWT and MNGA cells after treatment with hydrogen peroxide 96 RT-PCR for pro-apoptotic markers in MNWT and MNGA cell lines at 3 hours and 24 hours after treatment with 200 pM hydrogen peroxide 97 RT-PCR for anti-apoptotic markers in MNWT and MNGA cell lines at 3 hours and 24 hours after treatment with 200 pM hydrogen peroxide 100 Immunoblot for human bcl-2 in Mn-1, MNWT, and MNGA cells 24 hours after infection with Ad-bcl-2 101 Trypan blue assay for viable cells after treatment with 200 or 500 pM hydrogen peroxide 103 RT-PCR for pro-apoptotic markers in non-infected, Ad-lacZ infected, or Ad-bcl-2 infected MNGA cells 3 hours after treatment with 200 pM hydrogen peroxide 104 RT-PCR for anti-apoptotic markers in non-infected, Ad-lacZ infected, or Ad-bcl-2 infected MNGA cells 3 hours after treatment with 200 pM hydrogen peroxide 105 Average weekly weights for mice from pre-disease onset (12 weeks) to end-stage (18 weeks) 130 Representative photomicrographs of lumbar spinal cord anterior homs at the experimental end-stage 131 Total number of motor neurons in 10 lumbar cord sections from B6SJL, WT-SOD1, and G93A-SOD1 mice at end-stage 135 ix Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 4.10 Figure 4.11 Figure 4.12 Figure 4.13 Figure 4.14 Total number of motor neurons in 10 lumbar cord sections from G93 A-SOD 1 mice pre-disease onset, at disease onset, at disease midpoint, and at disease end-point Representative photomicrograph of immunohistochemical staining for human bcl-2 in mouse lumbar motor neurons Average weekly weights for B6SJL mice injected with saline, Ad-lacZ, or Ad-bcl-2 Average weekly weights for WT-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2 Total number of motor neurons in 10 lumbar cord sections from B6SJL mice injected with saline, Ad-lacZ, or Ad-bcl-2 Total number of motor neurons in 10 lumbar cord sections from WT-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2 Age of disease onset in G93 A-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2 Length of disease course in G93 A-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2 Average weekly weights for G93A-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2 Total number of motor neurons in 10 lumbar cord sections From G93A-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2. Sections were collected at disease onset Total number of motor neurons in 10 lumbar cord sections From G93 A-SOD 1 mice injected with saline, Ad-lacZ, or Ad-bcl-2. Sections were collected at disease end-stage Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1 BACKGROUND AND SIGNIFICANCE Introduction Amyotrophic lateral sclerosis (ALS), more commonly known as Lou Gehrig’s disease, is a degenerative disease of the central nervous system. Motor neurons located in the anterior horn of the spinal cord and neurons of the motor cortex that form the lateral corticospinal tracts die, resulting in a clinical disease characterized by mixed upper and lower motor neuron symptoms. The ultimate result of motor neuron degeneration is paralysis and death. Currently, the cause of neuronal death in ALS is not understood. Like many neurological disorders, ALS can be either sporadic (SALS) or familial. Although clinically indistinguishable from each other, familial ALS (FALS) has been associated with a variety of genetic mutations. Nearly twenty percent of FALS cases have a mutation in the cytoplasmic copper/zinc superoxide dismutase gene (SOD1) on chromosome 21q. The connection between SOD I mutations and motor neuron death has been studied in tissue culture and in transgenic mouse models, but the link remains elusive. One question that has been studied in depth is the nature of cell death in FALS, whether it is necrotic, characterized by a toxic insult resulting in cell lysis, or apoptotic, proceeding by an aberrant programmed death mechanism. Most current evidence points towards the latter and so potential therapy for FALS has centered on 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. prevention of apoptosis. At the center of such therapy lies a family of proteins called the B-cell lymphoma-2 (Bcl-2) family that has been shown to alter and even inhibit apoptotic mechanisms in many cell death scenarios. Neurodegenerative disorders and cell death Neurodegenerative disorders are a group of neurological disorders characterized by loss of specific subsets of neurons, resulting in clinical manifestations ranging from cognitive decline to complete muscular paralysis, and invariably ending in death. The majority of cases arise sporadically, but a small percentage is inherited. Many of the genetic alterations associated with hereditary forms of disease have been discovered in recent years, allowing researchers to study and manipulate the molecular and genetic mechanisms thought to be responsible. Insight gained from such research in the past decade has brought new hope for treatment of these devastating illnesses, none of which can currently be cured or prevented. Neurodegeneration is a term that includes several types of neuronal damage, including axonal degeneration, accumulation of insoluble abnormal intracellular proteins, and ceil death. Cell death is traditionally classified as either necrotic or apoptotic. Loss of plasma membrane integrity and spilling of cell contents, resulting in inflammation and further tissue damage characterize necrosis (Saikumar et al., 1999). Apoptosis is a form of programmed cell death (PCD), in which a cell 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. receives a signal to die and responds by initiating an active pathway that requires energy and often involves de novo protein synthesis (Lincz, 1998). The plasma membrane remains intact thus preventing an inflammatory response. In most multicellular organisms, PCD is a normal physiological process, and is essential for proper development and maturation. In the development of the human nervous system, nearly 50% of the neurons originally formed die by PCD, mostly prior to birth (Rubin, 1997). Such a powerful mechanism to remove cells has great potential for harmful effects if there is perturbation of the normal process or reversion to a developmental process, as is thought to be the case in many instances of neurodegeneration. Kerr, et al first used the term apoptosis, from the Greek word for leaves “falling o ff’ a tree, to describe the morphological changes in a particular form of PCD (Kerr et al., 1972). Apoptosis is now used to describe a type of cell death characterized by a series of morphological, biochemical, and molecular events that occur in many cell types, although not all events occur in all cell types or forms of PCD. The primary morphological features of apoptosis as first described by Kerr, et al are plasma membrane blebbing to form apoptotic bodies, cell shrinkage, and chromatin condensation (Kerr et al., 1972). The apoptotic bodies are membrane- enclosed fragments of cytoplasm and organelles that are phagocytosed by neighboring cells resulting in lack of inflammation. Since this initial description nearly thirty years ago, several biochemical and molecular events have been described which correlate with the morphological changes. These include DNA 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cleavage into multiples of 180-200 base-pair sized fragments (DNA ladder), corresponding to the distance between nucleosomes, loss of plasma membrane asymmetry, and activation of an intracellular cascade involving proteases (blebbing), mitochondrial alterations, and a growing family of regulatory proteins. Apoptosis, caspases, and the bcl-2 family In the past decade a large family of regulatory proteins involved in apoptosis, called the bcl-2 family, has been discovered. These are divided in to pro- and anti- apoptotic members. Bcl-2, the best-characterized anti-apoptotic member of the family, was originally found to be an oncogene in B-cell lymphoma. It was discovered that, unlike in most cancers known at the time, the defect was not in unbridled cell division, but rather in aberrant cell death, such that cells over expressing bcl-2 did not die (Brown, 1997). Thus bcl-2 was identified as the first mammalian anti-cell death gene. Since then a family of related proteins has been elucidated, including several more anti-apoptotic proteins, such as bcl-xl, bcl-w, and A l, and a series of pro-apoptotic molecules, including bax, bak, bik, bad, bim, and Hrk (Holinger et al., 1999; Huang and Strasser, 2000; Lin et al., 1993; Pelligrini and Strasser, 1999; Yin et al., 1994). The function of such a large variety of regulatory proteins and their interaction is not known, however it is thought that many of the regulatory molecules dominate in a specific cell type and/or at a certain stage of the 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. life cycle. For example, bcl-2 is expressed in the developing nervous system, but bcl-xl predominates in the adult nervous system and bcl-2 expression is minimal (Gonzalez-Garcia et al., 1995). A great deal has been discovered in recent years regarding the function of the bcl-2 family in regulating apoptosis. Bcl-2 is able to inhibit apoptosis caused by numerous stimuli, including serum and growth factor withdrawal, chemotherapeutic agents, irradiation, and glucocorticoid treatment (Saikumar et al., 1999). The action of bcl-2 appears to be dependent on its location on the outer mitochondrial membrane. If bcl-2 is not allowed to target to the mitochondria, then its anti- apoptotic effect is limited (Minn et al., 1998). Similarly, bax exerts its pro-apoptotic function through association with the mitochondrial membrane (Marzo et al., 1998). This points to the mitochondria as key players in apoptosis. How anti-apoptotic bcl-2-like molecules and pro-apoptotic molecules, such as bax, interact to regulate cell death at the mitochondrial membrane is currently the focus of intense research. Although there are a variety of stimuli and signaling pathways that can result in apoptosis, the final effectors in the majority of cases appear to be a family of proteases called caspases. Caspases cleave at specific aspartate residues, and are themselves formed as procaspases that must be cleaved to become active (Deveraux et al., 1999; Vaux and Strasser, L996). Procaspases appear to be constitutively present in the cell. If trophic factors required for survival are removed, a signaling pathway is activated that results in the cleavage of procaspases to form active caspases. The active caspases in turn activate downstream caspases in 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a manner reminiscent of the complement cascade, or cleave a variety of intracellular targets that produce the characteristics of apoptosis and result in cell death (Allen et al., 1998; Pettman and Henderson, 1998; Saikumar et al., 1999). Cleavage of a specific intracellular protein target can result in either activation or destruction of the target. Many structural proteins, including nuclear lamins and actin, are cleaved, resulting in loss of structural support and cellular disintegration. The DNA repair enzyme poly (ADP)-ribosylating protein (PARP) is inactivated by cleavage, while the inhibitor for an apoptosis related DNAse is cleaved (ICAD), allowing activation of the DNAse (CAD), which in turn is responsible for cleavage of the DNA into the characteristic oligonucleosomal sized fragments. This cascade also affects a number of regulatory proteins, resulting in up- regulation of the pro-apoptotic c-Jun and stress-activated protein kinase pathways and inhibition of the anti-apoptotic PI3 kinase pathway (Saikumar et al., 1999). In addition to removal of environmental survival signals, a number of stimuli can act via a family of death receptors, including Fas and tumor necrosis factor receptor, which act to directly activate the caspase cascade (Allen et al., 1998). Apoptosis is best characterized in the nematode Caenorhabditis elegans. During development, 1090 somatic cells are formed, out of which 131 die by PCD (Thornberry, 1996). The molecular pathway has been elucidated for the effectors in this system, and centers on two molecules, ced-3 and ced-4. Ced-3 is a caspase, and in C. elegans, is responsible for producing the characteristic morphological changes of apoptosis by cleaving various intracellular targets. Ced-4 binds ced-3, and is 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. responsible for activation of ced-3 by forming multimeric complexes composed of ced-3 and ced-4 that allow for self-activation of ced-3. This multimeric complex is termed the apoptosome, and its formation can be inhibited by a third molecule called ced-9, which binds ced-4 and prevents the formation of the apoptosome and thus activation of ced-3. Several mammalian homologues have been found for the various ced proteins, including a whole family of caspases related to ced-3 and another family of ced-9 related molecules that form the bcl-2 family (Lincz, 1998; Saikumar et al., 1999). A recently identified ced-4 homolog, termed Apaf-l, appears to be the link between bcl-2 and the caspases. In the apoptosome of C. elegans, ced-4 prevents the activation of a caspase-9 like initiator caspase, ced-3. Bcl-2 is thought to similarly interact with Apaf-l to prevent activation of caspase-9 in mammalian cells. Bcl-2 thus prevents the activation of the caspase cascade, and apoptosis does not proceed (Lincz, 1998). Despite the similarity, however, the mammalian pathway is not as simple as that of C. elegans. Successful activation of the caspases by Apaf-1 requires the presence of additional factors, including ATP and cytochrome c, a protein involved in the respiratory chain of the mitochondrial inner membrane (Saikumar et al., 1999). The release of cytochrome c from the mitochondrial intermembrane space requires loss of mitochondrial outer membrane integrity. Another feature of apoptosis is termed the permeability transition, in which the mitochondria lose their membrane potential due to an alteration in permeability. This alteration is due to the formation of the permeability pore, the components of 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. which have yet to be completely identified. Recent evidence points to a role for bax in the opening of the transition pore, which would partially account for its role as a pro-apoptotic protein (Marzo et al., 1998). Bcl-2 has not been localized to the pore as of yet, but its mitochondrial location coincides with the location of the pore complex, so it is thought that bcl-2 exerts its anti-apoptotic regulatory effect, at least in part, by interacting with one or more of the pore components (Minn et al., 1998). According to one recent model (figure 1.1), an increase in bax after an apoptotic stimulus would cause bcl-2, which in the non-apoptotic state binds to and sequesters Apaf-l, to bind excess Bax. This results in the release of Apaf-l, which can then interact with procaspase-9. With the opening of the transition pore facilitated by bax and the release of cytochrome c and other activators of apoptosis, the Apaf- l/procaspase-9 complex (the apoptosome) could activate the caspase cascade, resulting in cell death (Adams and Cory, 1998). To add to the complexity, recent evidence indicates that bcl-2 can have a direct anti-oxidant effect distinct from its role in binding pro-apoptotic molecules (Kane et al., 1993; Tyurina et al., 1997; Wiedau-Pazos et al., 1996). Despite an incomplete understanding of the apoptotic process and the role of the Bcl-2 family, with the current level of knowledge one can still begin to envision possible strategies to alter the course of apoptotic cell death in human disease. 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.1. Diagrammatic representation of the current theory of bcl-2, bax, and caspase interactions to produce apoptosis. In a non-apoptotic cell, bcl-2 is localized to the mitochondria and binds Apaf-l (1). When the cell receives an apoptotic insult (2), a pathway is activated that results in up-regulation of bax transcription. The increased levels of bax have two possible functions (4). Cytoplasmic bax translocates to the mitochondrial membrane and binds bcl-2 (or bcl-xl), interfering with bcl-2-Apaf-l binding. Evidence for the second function is not conclusive, but it is thought that bax is able to cause the release of cytochrome c (cyt c) from the mitochondrial intermembrane space, possibly by forming a pore. Free Apaf-l and cytochrome c are able to interact with procaspase-9 (procas 9) and dATP to form the apoptosome (5). The apoptosome allows multiple procaspase-9 molecules to become active caspase-9 (cas 9) (6), which in turn cleaves procaspase-3 (procas 3) to form active caspase-3 (cas 3). Active caspase-3 will cleave a variety of intracellular targets, including additional caspases, structural proteins, and DNA repair and regulatory enzymes, producing the apoptotic phenotype and eventual cell death. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Apoptotic insult (oxidative stress, UV, growth factor withdrawal) Active cas 3 9. Cleavage of targets £> to produce apoptosis (additional caspases, structural proteins, repair and regulatory enzymes). Procas 3 Active cas9 f dATP Apoptosome Cyt C^. W * Increase bax transcription MITOCHONDRIA NUCLEUS o Assays for apoptosis A number of biochemical and molecular changes characteristic of apoptosis form the basis of assays that distinguish cell death by apoptosis from cell death by other processes. One of the most common, and often considered a hallmark of apoptosis, is detection of DNA cleavage into 200 base-pair fragments (corresponding to nucleosomes) after activation of the apoptotic cascade (Bortner et al., 1995). DNA from apoptotic cells thus produces a pattern of bands on an agarose gel approximately 200 base pairs apart, called a DNA ladder. A second assay, for the loss of membrane asymmetry, is also used to detect apoptosis. Phosphatidylserine (PS) is usually found on the intracellular surface of the plasma membrane. During the apoptotic process, PS is exposed on the otherwise intact external surface of the plasma membrane. A lipophilic protein called annexin V will bind preferentially to PS, and when coupled to a fluorescent marker can be used to identify apoptotic cells by flow cytometry (Maiese and Vincent, 2000). Third, activation of caspases is assayed by a number of different methods, including sodium-dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblotting to detect the smaller, activated enzyme form and colorimetric assays for enzyme activity (Armstrong et al., 1997; Takadera et al., 1999). This is by no means a comprehensive list of the techniques used to detect apoptosis, and as new processes are elucidated in apoptotic pathways more tests will certainly be devised. 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Amyotrophic Lateral Sclerosis and Excitotoxicity ALS is a progressive neurodegenerative disorder in which there is selective loss of motor neurons in the anterior horn of the spinal cord and neurons of the motor cortex that form the lateral corticospinal tracts. ALS has a prevalence of about 5 cases per 100,000, with a median age of onset of 55 and a median survival of three years (Ince et al., 1998). Death is usually due to complications of paralysis including compromised respiratory function (Cudkowicz et al., 1997). Clinically there is a mixture of upper and lower motor neuron disease symptoms. Lower motor neuron disease produces weakness and atrophy of limb muscles, fasciculations, and cramps. Upper motor neuron disease results in weakness, spasticity, and hyperreflexia. Onset of disease is usually asymmetric and localized, then progressive in a contiguous manner (de Belleroche et al., 1995). The majority of cases are sporadic, but nearly ten percent are inherited in an autosomal dominant fashion and are termed familial ALS (FALS). FALS is clinically indistinguishable from sporadic ALS (SALS) (Deng et al., 1993). Recent evidence seems to suggest that motor neuron death in ALS may be the result of excitotoxicity (Lin et al., 1998; Ludolph et al., 2000). This term applies to over-stimulation of a neuron by excessive excitatory input. The primary excitatory neurotransmitter in the mammalian central nervous system (CNS) is glutamate. Glutamate binds to a family of receptors that include sodium and calcium channels. In normal excitatory transmission, binding of glutamate opens the channel, allowing 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. influx of sodium via a receptor subtype called the AMPA/kainate receptor. If enough ion channels open the nerve terminal will depolarize, producing an action potential. Over-stimulation of the AMPA/kainate receptor by glutamate produces a prolonged depolarization, which removes the endogenous blockade of a second glutamate receptor subtype, the NMDA receptor. Removal of the blockade allows glutamate to bind to and open NMDA receptors, which results in an influx of calcium ions (Doble, 1999). Calcium is a strong intracellular messenger, and in excess can cause activation of enzymes, such as proteases, kinases and phosphatases, as well as activating second messenger cascades that lead to cell lysis and death (Coyle and Puttfarcken, 1993). In addition to its role as an intracellular messenger, calcium can have a direct effect on cellular homeostasis by preventing repolarization. The inability of a cell to repolarize will result in an altered membrane potential. As the cell attempts to repolarize energy stores will deplete, resulting in collapse of energy production. At this stage, excitotoxicity becomes a positive feedback system. The inability to produce energy causes glutamate reuptake mechanisms to fail and intracellular stores of glutamate to be released, which in turn further increases extra cellular stores of glutamate to act on receptors. In addition, this loss of cellular integrity provides a mechanism to spread excitotoxic cell death to neighboring cells (Coyle and Puttfarcken, 1993). In SALS there is evidence for an impaired glutamate reuptake mechanism by the glial cells (Lin et al., 1998; Ludolph et al., 2000; Meyer et al., 1998). This would result in excess glutamate at the synapse, allowing excitotoxicity to proceed as outlined above. 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FALS and SOD1 In approximately twenty percent of FALS cases a mutation is found in the cytoplasmic copper/zinc superoxide dismutase (SOD1) (Rosen et al., 1993). To date over sixty SOD1 mutations have been found in FALS, almost all single base pair substitutions resulting in a full length, functional enzyme (Orrell, 2000). SOD I normally converts the superoxide free radical, OV-, to hydrogen peroxide (H2O2) and oxygen (O2) (McNamara and Fridovich, 1993). Free radicals such as OV - and the hydroxyl radical ( OH), along with H2O2 and peroxynitrite (ONOO '), are termed reactive oxygen species (ROS) and are normal byproducts of cellular metabolism (Robberecht, 2000). ROS can interact with and damage a variety of intracellular targets including lipid membranes, proteins, and nucleic acids (Dawson and Dawson, 1998). Due to the potentially damaging nature of such molecules, cells have developed several methods of defense against ROS, including three types of SOD that perform the function described above, and catalase and glutathione peroxidase, which detoxify H202(Lipton et al., 1993; Robberecht, 2000). When the link between certain cases of FALS and SOD1 mutations was discovered, an early theory was that mutant SOD1 had decreased activity, resulting in increased levels of intracellular ROS, which would damage a variety of intracellular structures and impair mitochondrial energy production (McNamara and Fridovich, 1993). As described above, loss of cellular energy equilibrium results in loss of cell integrity, decreased function of cellular ion pumps, increased intracellular 14 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. calcium, and cell death. However a loss of function mutation is inconsistent with the autosomal dominant form of inheritance of FALS, and recent evidence has shown that although many of the SOD1 mutants do have decreased activity, at least one has normal function and still causes FALS (de Belleroche et al., 1995). In addition, transgenic knockout mice lacking SOD1 do not develop motor neuron disease, implying that it is the mutation that causes disease (Robberecht, 2000). The current theory is that mutant SOD1 has gained a toxic function. Transgenic mice over-expressing a wild type human SOD1 do not develop disease, while mice expressing a mutant SOD1 become progressively paralyzed in a manner similar to human ALS (Brujin et al., 1998; Cleveland et al., 1996; Gurney et al., 1994; Ripps et al., 1995; Wong et al., 1995). This indicates that the gain of function is not simply an increase in the normal dismutase action of the enzyme, supported by the aforementioned decrease in activity of most of the mutants, but rather the gain of a novel toxic function. Two hypotheses have been proposed to explain the gain of function. The first is an increase in a normally low-level peroxidase reaction by SOD1, producing OV - from H1O2 . The second involves an increased ability of peroxynitrite, ONOO', formed by the interaction of OV - and nitric oxide (NO), to interact with OV -, producing a highly reactive nitronium intermediate. This intermediate could react with tyrosine in proteins, potentially resulting in inactivation of certain key cellular proteins and signaling pathways (Cleveland, 1996). Either method, or both in concert, could contribute to an altered cellular oxidative state and cellular and mitochondrial damage, producing cell death. Indeed, signs of oxidative 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. damage in both human ALS spinal cord and transgenic mouse models seem to indicate that excess ROS play a part in the disease mechanism (Robberecht, 2000). Motor neurons and models of neurodegeneration One of the most confusing aspects of neurodegenerative disorders is the specificity of cell death. Familial forms of neurodegeneration, including FALS, are often linked to a mutation that is present in cells throughout the body, yet in most instances only a specific subset of cells is affected in the disease state. In FALS, mutant SOD1 is ubiquitously expressed but only motor neurons degenerate. This hints at an inherent susceptibility of motor neurons to oxidative damage that is absent in other cells of the body. Several hypotheses have been proposed to explain why motor neurons respond differently to this oxidative stress than other cells. Motor neurons lack two calcium-binding proteins, calbindin and parvalbumin, which might normally help to sequester excess intracellular calcium caused by oxidative damage or excitotoxic mechanisms. Two groups of motor neurons that are spared in ALS, Onuf’s nucleus and oculomotor neurons, both express normal levels of these two proteins, thus supporting the theory that lack of calcium binding proteins predisposes most motor neurons to calcium mediated cell death (Doble, 1999). In addition, some groups have reported that the levels of the AMPA/kainate glutamate receptor subunit 2 (GluR2) are decreased in motor neurons (Shaw et al., 1999). GluR2 can form a functional glutamate receptor with other AMPA receptor subunits, resulting in a 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. calcium impermeable receptor complex. The absence of GluR2 in motor neurons would indicate that the functional AMPA receptor allows the passage of calcium into the cell. This would account for the evidence that in motor neurons degeneration appears to proceed via AMPA/kainate receptors rather than NMDA receptors, as described above (Coyle and Puttfarcken, 1993). One missing piece of the puzzle, however, is the link between glutamate receptor activation and the type of cell death found in ALS. Evidence points to apoptosis as the primary means of cell death in ALS, while the excitotoxic damage described above seems to produce cell death by lysis, which is necrosis. The link may be the particular subtype of AMPA/kainate receptor found in motor neurons, which upon activation produces an intracellular calcium cascade favoring apoptosis, as opposed to NMDA receptor mediated necrosis (Doble, 1999). How activation of an alternate form of the glutamate receptor can produce one type of cell death over another is an open area for future study. The current question is whether cell death in ALS is apoptotic, or whether, as has been hypothesized, it lies somewhere along the spectrum between necrosis and apoptosis. The answer to this question has important therapeutic implications. The evidence accumulated thus far for a cell death mechanism in ALS has proven inconclusive. Some studies have found increased levels of Bax and decreased levels of Bd-2 in motor neurons of SALS spinal cords, while others have found no changes (Mu et al., 1996; Troost et al., 1995). DNA fragmentation has been demonstrated in SALS spinal cord tissue by a technique called in situ end 17 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. labeling (ISEL), which enzymatically labels cleaved 3’ ends of DNA (Troost et al., 1995). Using a related technique called TUNEL another group could not confirm this finding (He and Strong, 2000). Finally, only one group has shown caspase activation in human ALS tissue (Martin, 1999). To further address the question of apoptotic versus necrotic cell death models of FALS have been created in both transgenic mice and in tissue culture. Mice expressing a variety of mutations found in FALS have been created, including SOD1 mutations (glycine to arginine at positions 37,85, and 86, and glycine to alanine at position 93) and over-expression of the neurofilament light chain (Brujin et al., 1998; Cleveland et al., 1996; Gurney et al., 1994; Ripps et al., 1995; Wong et al., 1995). Each of the models reproduces some of the characteristics of human motor neuron disease, although none are exact clinically or pathologically. The most commonly used transgenic model has a glycine to alanine mutation at position 93 (G93A) of SOD I, developed by Gurney, et al. Clinically the mice develop hind limb weakness at 3 to 4 months of age, followed by tremor, and progress to complete paralysis and death around 5 months of age (Gurney et al., 1994; Vukosavic et al., 1999). Cytopathological studies comparing the motor neuron degeneration of G93A FALS mice to human FALS spinal cord sections show a number of similarities, including anterior horn atrophy, gliosis, axonal swelling, neurofilament aggregation, and Lewy-type bodies (intra-cytoplasmic inclusions) (Dal Canto and Gurney, 1994; Tu et al., 1996). 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Attempts to identify the type of motor neuron death in the G93A mouse model have provided mixed results, paralleling similar attempts in human ALS. One group reported that G93A mice lack standard signs of apoptosis, including evidence of DNA strand breaks and activation of the apoptosis associated protease caspase-3, while another group found evidence of both caspase-3 and a related protease caspase-1, to be activated in G93A motor neurons (Li et al., 2000; Mighelli et al., 1999). In addition, several groups have shown that expression of bcl-2 can delay onset of motor neuron disease in G93A mice (Assouz et al., 2000; Kostic et al., 1997; Mighelli et al., 1999). This does not confirm that the process is apoptotic, however, because bcl-2 has also been postulated to possess anti-cell death functions not related to known anti-apoptotic pathways (Kane et al., 1993; Tyurina et al., L997). In tissue culture, non-motor neuron cell lines, including two neuroblastoma lines (SH-SY5Y and N2a), a pheochromacytoma line (PC 12) and yeast cells have been transfected with both wild-type and mutant forms of SOD I (Carri et al., 1997; Ghadge et al., 1997; Pasinelli et al., 1998; Rabizadeh et al., 1995). In SH-SY5Y cells, over-expression of a mutant SOD1 produces altered mitochondrial function, an increase in intracellular calcium, increased production of ROS, and decreased levels of bcl-2. Although each of these changes could be related to an apoptotic process, each one has also been associated with necrosis, with the exception of decreased bcl- 2. However, as previously mentioned, bcl-2 may have functions unrelated to apoptosis, and it might be these activities that are altered in the presence of mutant 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SOD1. In N2a cells, expression of mutant SOD L resulted in activation of caspase-1, but the contribution of caspase-1 to apoptosis is still controversial (Gurney et al., 2000; Wyllie, 1997). Despite each of these objections, none of the data eliminates the possibility of an apoptotic process, and these models are useful in examining the effects of mutant SOD I in neural cells. However it is necessary to keep in mind that the pathway to apoptosis differs from cell to cell, and therefore the best model of motor neuron death is in a motor neuron. The ideal situation is to produce primary cultures from FALS mice, which has been done with some success (Assouz et al., 2000). Isolation of motor neurons from mice is technically difficult and time consuming, though, and thus limits the number of experimental variables that can be addressed. As with many other cell types that are difficult to culture, a transformed cell line could be used to create a model. Again, there is a problem. Adult motor neurons are terminally differentiated cells and cannot be made immortal by transformation. This difficulty has been overcome by creating fusion cell lines. One such line, the Mn-1 cell line, developed by Salazar-Grueso, et al, is a fusion of embryonic day 13 mouse spinal cord motor neurons with the N18TG2 neuroblastoma line. The cells can be differentiated to a neuronal phenotype, showing flattened cell bodies and long processes, by reduction of serum present in the culture medium, and will express several motor neuron markers, such as acetylcholine esterase, choline acetyl transferase, and the neurofilament light, medium, and heavy chains (Salazar-Grueso et al., 1991). Although these 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. characteristics serve to minimally identify the Mn-1 line as motor neuronal, it is necessary to perform further characterization in order to use it as a model for SOD1- mediated cell death. As mentioned previously each type of cell responds differently to apoptotic stimuli, so a clear understanding of the origin and present state of the Mn-1 line is needed. Motor neuron markers The field of motor neuron research has produced few definitive markers. At best, one can look for expression patterns that are found, if not exclusively, then at least consistently, in motor neurons. These can be divided into 3 groups: receptors, general intracellular proteins, and developmental stage and location specific markers. Receptors characteristic of motor neurons include the neuronal cell adhesion molecule (NCAM), the AMPA-specific glutamate receptor, normally composed of subunits 1 through 4 (GluRl-4), the androgen receptor, and the tyrosine kinase receptors B (trkB) and C (trkC) (Chen and Chiu, 1992; Sar and Stumpf, 1997; Sendtner et al., 2000; Williams et al., 1997). One hypothesis for the susceptibility of motor neurons to cell death in ALS is a decreased amount of the GluR2 subunit (Shaw et al., 1999). The GluR2 subunit renders the AMPA glutamate receptor impermeable to calcium. Glutamate toxicity via AMPA receptors has been implicated in cases of sporadic ALS (Vandenberghe et al., 2000). The androgen receptor is also normally expressed on adult motor neurons. One of the peculiarities 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the Mn-1 line is the reported lack of expression of the androgen receptor. This unusual characteristic has been exploited to develop a model of another motor neuron disorder, Kennedy’s disease (Brooks et al., 1998). Finally, trkB and trkC are members of the tyrosine kinase receptor family that mediates responses to a variety of molecules, called neurotrophins, which are involved in neuronal growth, differentiation, and death (Sendtner et al., 2000). There are also a variety of intracellular proteins that are consistent with motor neuron lineage, including neuron-specific enolase (NSE) (Nogami et al., 1998). As negative markers, two intermediate filament proteins, vimentin (Vim) and glial fibrillary acidic protein (GFAP) should not be expressed in adult motor neurons (Boyne et al., 1996; Vaquero et al., 1992). Finally, as mentioned before, neither calbindin nor parvalbumin are expressed in motor neurons. Last, there are developmental stage and location specific markers that will provide the most information regarding motor neuron lineage. The field of motor neuron development has identified several markers that, if expressed concurrently or sequentially during development, help to pinpoint the embryonic origin of spinal cord precursor cells (Goulding, 1998; Tanabe et al., 1998; Tsuchida et al., 1994). Some of these proteins are expressed throughout the life of the neuron, while others have a limited time span. Looking at expression of these markers in the Mn-1 cell line in non-differentiated and differentiated states may provide confirmation of their motor neuronal origin. It cannot be expected, however, that all such markers will be found, due to the temporal expression patterns associated with several of them. In 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. addition, the use of a neuroblastoma parent will likely present some markers that would not normally be expected from a motor neuron, and perhaps even down- regulate others that should be present. In the developing neural tube, precursor cells first respond to graded levels of sonic hedge-hog (Shh), an inductive protein secreted by the notocord. The level of Shh received by precursor cells results in two populations of cells, each expressing a high level of a specific homeodomain protein. Cells with a high level of Pax6 form a pool of ventral somatic motor neuron (SMN) progenitors, while Nkx2.2-expressing cells will generate visceral motor neurons. Although the precursors tend to produce one protein or the other almost exclusively, some precursors do transiently produce both Pax6 and Nkx2.2. Pax6 expression is dominant, however, and even in the presence of transiently expressed Nkx2.2 is thought to produce somatic motor neurons (Ericson et al., 1997). Pax6 positive cells exit the cell cycle and begin to express a member of the Lim homeodomain family, islet-1 (Isll). Expression of Isll is common to all SMN progenitors, although it is not expressed throughout the adult life of all SMNs. At this point, the expression of various other members of the Lim family results in division of SMN progenitors into columns within the developing ventral cord (figure 1.2). The cord is divided into medial and lateral, with each of these regions having medial and lateral divisions also. These four regions are identified as medial column medial (MCM), medial column lateral (MCL), lateral column medial (LCM), and lateral column lateral (LCL). Based on the expression patterns of four 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.2. Diagram showing the location and expression patterns of embryonic motor neuron precursors in the developing spinal cord. Note that cells of the LCL co-express isll along with Isl2 and Liml for a brief period of time during development. Abbreviations: medial column medial (MCM); medial column lateral (MCL); lateral column medial (LCM); lateral column lateral (LCL); islet-1 (Isll); islet-2 (Isl2). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cranial Medial Cervical Thoracic Lumbar Caudal MCM (Isll, Isl2, Lim3) MCL (Isll, Isl2) Lateral LCM (Isll, Isl2) LCL asI2, Liml) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 Lim family members, the embryonic origin of a SMN precursor can be identified. The MCM SMN precursors express Isll, a related protein islet-2 (Isl2), and a third Lim family member, Lim3. The MCL and LCM cells both express Isl L and Isl2, but not Lim3, and the LCL cells express Isl2 and a fourth Lim family member, Liml. The LCL cells will transiently co-express Isll with Isl2 and Liml, but at later stages will down-regulate Isll expression. A fifth group of cells can be identified by expression of Isll without Isl2, and these cells will become visceral motor neurons (Pfaff and Kintner, 1998; Tsuchida et al., 1994). Examining the Mn-1 cell line in undifferentiated and differentiated states for expression of these markers should provide confirmation of their motor neuronal origin and thus of their utility as a model for motor neuron disease. Treatment of ALS The eventual goal of modeling any disease is to provide enough clues to the etiology and process that a specific and effective therapy can be developed. None of the neurodegenerative disorders can currently be cured or prevented, and in most cases only the symptoms can be treated, not the underlying disorder. ALS is no exception. The mounting evidence that glutamate excitotoxicity plays a role in ALS has led to the development of a glutamate release inhibitor called riluzole that is currently the only approved drug therapy for ALS. Use of riluzole has shown a slight increase in lifespan and decrease in muscle deterioration, however it is neither 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a long-term therapy nor a cure (Doble, 1999). Another drug that interferes with glutamatergic transmission, gabapentin, has been shown to be slightly neuroprotective in human ALS (Doble, 1999). In mouse models of FALS, riluzole prolongs survival, but does not delay onset of disease, while an antioxidant, vitamin E, delays clinical onset and progression, but does not prolong survival. This suggests that the altered oxidative state due to mutant SOD 1 and excitotoxicity may act at different stages of the disease, and may need to be treated separately. In the same mouse model, gabapentin was able to prolong survival (Gurney et al., 1996). Each of the current treatments for ALS attempts to affect the pharmacological mechanisms of disease. With the multitude of new information regarding disease mechanisms and genetic and molecular control of cellular death processes, the next step in therapy for ALS, and indeed for many diseases, is a gene therapy approach. In recent years gene therapy has progressed from a theoretical dream to a clinical reality. A number of diseases have been the focus of clinical trials (Engelhard et al., 1998; Harris and Lemoine, 1996). Gene therapy for the central nervous system (CNS) is also progressing, but presents some difficulties not encountered with dividing and removable cells (i.e. cells that can be taken out of the body and returned after treatment). For this reason, viral vectors are the delivery vehicles that currently hold the most promise for gene therapy of the CNS. Viruses can infect cells and deliver genes without the need to manipulate environmental conditions in a culture dish. Adenoviruses are even able to infect non-dividing cells such as neurons, and produce long-term (although not permanent) expression of exogenous gene products 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Karpati et alM 1996). Another advantage of the adenoviral vector for gene therapy of the CNS, specifically of the spinal motor neurons, is that intra-muscular injection of the vector results in uptake and retrograde transport along the motor neurons (Warita et al., 1998). The advent of gene therapy and adenoviral vectors, then, provides the opportunity to treat motor neuron degeneration by delivery of an exogenous gene or copy DNA to correct the endogenous cellular malfunction. In FALS, there are two possible strategies for gene therapy: either correct the cause of the problem (i.e. the SOD1 mutation) or the end result (i.e. cell death). The latter choice is the better of the two for several reasons. First, SOD1 mutations are only a small fraction of total ALS cases. In addition, even within cases expressing a mutant SOD1, there have been over 50 different mutations discovered. This makes targeting any one mutation in the SOD1 gene of extremely limited benefit. Instead, an approach that can alter the cellular response to the SOD1 mutation would be of greater use. The question is, what gene product can be increased to prevent cell death? The answer is being provided by the recent insights in to the mechanisms of apoptosis. Members of the bcl-2 family are ideal candidates for gene therapy of apoptotic cell death. One recent study has shown that bcl-2 expressed from adeno-associated virus injected directly into FALS mouse spinal cord can delay cell death and onset of symptoms in the area adjacent to the injection (Assouz et al., 2000). If cell death in neurodegenerative disorders is indeed apoptotic, then altering levels of these regulatory proteins could be an effective treatment for neurodegeneration. 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. OVERVIEW OF THESIS WORK Since the identification nearly a decade ago of SOD1 mutations in a subset of FALS cases intense research has provided insight into the pathogenic mechanisms involved in selective motor neuron death. Three main questions have been addressed. The first centers on the mechanisms leading to cell death and disease. Two theories have emerged that may be responsible for different aspects of SODl- associated disease. The first is based on evidence of oxidative damage to a variety of intracellular components and the apparent toxic gain of function due to SOD 1 mutations (Liu et al., 1999). Oxidative damage may result in the initial damage to motor neurons that produces disease onset. The second theory involves both familial and sporadic forms of ALS and centers on neuronal damage caused by glutamate excitotoxicity (Doble, 1999). In FALS it has been hypothesized that the initial insult to motor neurons is oxidative, while progression of disease is the result of excitotoxicity. The end result is motor neuron death and progressive paralysis. The second question involves the selective nature of cell death in ALS. Despite the ubiquitous expression of SOD1, only motor neurons are affected by expression of mutant SOD1 in FALS, and the same selectivity is found in SALS with no known mutation. Several theories have been suggested to explain these findings, including decreased or absent expression of a glutamate receptor subunit and two calcium binding proteins, calbindin and parvalbumin (Doble, 1999; Shaw et al., 1999). Third, the type of cell death in ALS has been the center of much debate. Conflicting 29 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reports have produced evidence both for and against apoptosis as the final death pathway due to oxidative or glutamatergic damage. The answer to this question could have important therapeutic implications. In chapter 2 ,1 describe my identification of a cell line for use as a motor neuronal model of cell death, and then present the results of my investigation into the apoptotic properties of this system. I have shown that the Mn-l cell line appears to be derived from embryonic motor neuron precursors ultimately responsible for innervation of limb muscles, and as such should be useful for studying motor neuron cell death as it relates to neuromuscular disease. I also looked at the effect of a variety of apoptotic agents on Mn-1 cells. The results of this study showed that treatment with H2O2 produces a caspase-independent form of apoptosis, providing a model of motor neuron cell death that might have similarities to cell death in FALS. To further study the disease mechanisms involved in FALS, I designed a tissue culture model of mutant SOD 1-expressing motor neurons using the Mn-L cell line. This work is described in chapter 3. The glycine to alanine mutation at position 93 (G93A) of SOD1 is found in cases of FALS and has also been used to generate a transgenic mouse model of motor neuron degeneration, so 1 chose this mutant to produce the MNGA cell line [Gurney, 1994]. Using this line I was able to show that motor neurons expressing G93A-SOD1 have an increased susceptibility to oxidative-damage induced apoptosis compared to Mn-l cells or cells stably expressing the wild-type SOD1 (MNWT). This supports the role of oxidative damage in the pathogenesis of FALS, and also lends credence to the theory that cell 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. death in this disease is apoptotic. Similar to the type of cell death found in Mn-l cells after an oxidative insult, cell death in MNGA cells is apoptosis but lacks caspase activation. These findings correlate with similar results found in human and murine forms of ALS (Martin, 1999; Mighelli et al., 1999). Finally I looked at the ability of an adenoviral gene therapy vector containing cDNA for human bcl-2 (Ad- bcl-2) to prevent cell death in MNGA cells exposed to oxidative stress. The results show that Ad-bcl-2 is able to infect differentiated MNGA cells and protect them from apoptotic cell death induced by treatment with H2O2. Glutamate excitotoxicity is not present in this system, suggesting that bcl-2 can act via an anti-oxidant pathway rather than an anti-glutamatergic pathway to prevent motor neuron apoptosis. To see if the ability of the Ad-bcl-2 vector to prevent oxidative-damage induced apoptosis in MNGA cells could be extended to an in vivo model of FALS I treated G93 A transgenic mice with intra-muscular injections of Ad-bcl-2. Chapter 4 describes the results of these experiments. In mice injected with Ad-bcl-2 before signs of clinical disease there is a significant delay in disease onset compared to mice injected with saline or with an adenovirus containing the lacZ gene (Ad-lacZ). The effect of bcl-2, however, does not seem to extend the course of disease or to prevent motor neuron loss. These results indicate that bcl-2 may be acting in a role additional to that of an anti-apoptotic protein. This is supported by recent evidence showing that bci-2 can have a direct anti-oxidant function unrelated to its anti- apoptotic properties and by data from the MNGA cell line presented in chapter 2 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Kane et al., 1993; Tyurina et al., 1997). Based on the theory that disease onset in FALS is related to oxidative damage while disease progression is the result of glutamate excitotoxicity, the effect of bcl-2 in delaying onset but not preventing progression can be explained by this anti-oxidant function. The data indicate that while bcl-2 is a promising therapy for treatment of motor neuron degeneration in FALS, it will be necessary to use therapy that affects other elements of oxidative/excitotoxic-induced cell death along with bcl-2 to both delay onset and hinder disease progression. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2 CHARACTERIZATION OF HYDROGEN PEROXIDE-INDUCED CELL DEATH IN THE Mn-l MOTOR NEURON CELL LINE INTRODUCTION Amyotrophic lateral sclerosis (ALS), more commonly known as Lou Gehrig’s disease, is a progressive neurodegenerative disorder in which there is selective loss of motor neurons in the anterior horn of the spinal cord and neurons of the motor cortex that form the lateral corticospinal tracts (Ince et al., 1998). ALS has a prevalence of about 5 cases per 100,000, with a median age of onset of 55 and a median survival of three years. Death is due to complications of paralysis including compromised respiratory function (Cudkowicz et al., 1997). Clinically there is a mixture of upper and lower motor neuron symptoms. Lower motor neuron disease produces weakness and atrophy of limb muscles, fasciculations, and cramps. Upper motor neuron disease results in weakness, spasticity, and hyperreflexia. Onset of disease is usually asymmetric and localized, then progressive in a contiguous manner (de Belleroche et al., 1995). The majority of cases are sporadic, but nearly ten percent are inherited in an autosomal dominant fashion and are termed familial ALS (FALS). FALS is clinically indistinguishable from sporadic ALS (SALS) (Deng et al., 1993). 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Recent evidence seems to suggest that motor neuron death in ALS may be the result of glutamate excitotoxicity (Lin et al., 1998; Ludolph et al., 2000). Over stimulation of glutamate receptors produces a prolonged depolarization, which results in an influx of calcium ions (Doble, 1999). Calcium is a strong intracellular messenger, and in excess can cause activation of enzymes and second messenger cascades that lead to cell lysis and death (Coyle and Puttfarcken, 1993). As part of this process calcium causes increased production of reactive oxygen species (ROS) (Doble, 1999). ROS include free radicals such as OV - and the hydroxyl radical ( OH), along with H2O2 and peroxynitrite (ONOO ') (Robberecht, 2000). ROS can interact with and damage a variety of intracellular targets (Dawson and Dawson, 1998). Recent studies show lipid, DNA, and protein oxidation in ALS tissue, strengthening the theory of glutamate excitotoxicity acting via increased intracellular calcium (Liu et al., 1999). In the familial form of the disease, FALS, a subset of cases have been linked to mutations in cytoplasmic copper/zinc superoxide dismutase (SOD1) (Rosen et al., 1993). SOD1 normally converts the superoxide free radical, OV-, to hydrogen peroxide (H2O2) and oxygen (O2) (McNamara and Fridovich, 1993). The mutations are thought to confer a toxic function on SOD1 rather than causing disease as a result of decreased enzyme activity (de Belleroche et al., 1995; Robberecht, 2000). Two hypotheses have been proposed to explain the gain of function. The first is an increase in a normally low-level peroxidase reaction by SOD1, producing OV - from H2O2 . The second involves an increased ability of peroxynitrite, ONOO', formed by 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the interaction of O2 - and nitric oxide (NO), to interact with OV producing a highly reactive nitronium intermediate. This intermediate could react with tyrosine in proteins, potentially resulting in inactivation of certain key cellular proteins and signaling pathways (Cleveland, 1996). Either method, or both in concert, could contribute to an altered cellular oxidative state and cellular and mitochondrial damage, producing cell death, and providing a link between the sporadic and familial forms of ALS. Despite advances in understanding the intracellular pathways that initiate cell death in ALS, the question of which cell death pathway is activated, necrosis or apoptosis, still remains. The answer to this question could have important therapeutic implications. The evidence accumulated thus far for a cell death mechanism in ALS has proven inconclusive. Some studies have found increased levels of bax and decreased levels of bcl-2 in motor neurons of SALS spinal cords, while others have found no changes (Mu et al., 1996; Troost et al., 1995). DNA fragmentation has been demonstrated in SALS spinal cord tissue by a technique called in situ end labeling (ISEL), which enzymatically labels cleaved 3’ ends of DNA (Troost et al., 1995). Using a related technique called TUNEL another group could not confirm this finding (He and Strong, 2000). Finally, only one group has shown caspase activation in human ALS tissue (Martin, 1999). In tissue culture, non-motor neuron cell lines, including two neuroblastoma lines (SH-SY5Y and N2a), a pheochromacytoma line (PC12) and yeast cells have been transfected with both wild-type and mutant forms of SOD1 in an attempt to 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. model FALS (Carri et al., 1997; Ghadge et al., 1997; Pasinelli et al., 1998; Rabizadeh et al., 1995). In SH-SY5Y cells, over-expression of a mutant SOD1 produces altered mitochondrial function, an increase in intracellular calcium, increased production of ROS, and decreased levels of bcl-2. Although each of these changes could be related to an apoptotic process, each one has also been associated with necrosis, with the exception of decreased bcl-2. However bcl-2 may have an anti-oxidant function unrelated to its role in apoptosis, and it might be this function that is altered in the presence of mutant SOD1 (Kane et al., 1993; Tyurina et al., 1997). In N2a cells, expression of mutant SOD1 resulted in activation of caspase-1, but the contribution of caspase-1 to apoptosis is still controversial (Gurney et al., 2000; Wyllie, 1997). Despite each of these objections, none of the data eliminates the possibility of an apoptotic process, and these models are useful in examining the effects of mutant SOD1 in neural cells. However it is necessary to keep in mind that the pathway to apoptosis differs from cell to cell, and therefore the best model of motor neuron death is in a bona fide motor neuron. The ideal situation would be to produce primary cultures from FALS mice, which has been done with some success (Assouz et al., 2000). Isolation of motor neurons from mice is technically difficult and time consuming, though, and thus limits the number of experimental variables that can be addressed. As with many other cell types that are difficult to culture, a transformed cell line can be used to create a model. Adult motor neurons are terminally differentiated cells though, and do not spontaneously undergo malignant transformation. This 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. difficulty has been overcome by creating fusion cell lines. One such line, the Mn-1 cell line, developed by Salazar-Grueso, et al, is a fusion of embryonic day 13 mouse spinal cord motor neurons with the N18TG2 neuroblastoma line. The cells can be differentiated to a neuronal phenotype by reduction of serum present in the culture medium, and will express several motor neuron markers, such as acetylcholine esterase, choline acetyl transferase, and the neurofilament light, medium, and heavy chains (Salazar-Grueso et al., 1991). Although these characteristics serve to minimally identify the Mn-1 line as motor neuronal, it is necessary to perform further characterization in order to use it as a model for SODl-mediated cell death. As mentioned previously each type of cell responds differently to apoptotic stimuli, so a clear understanding of the origin and present state of the Mn-L line is needed. in this study the Mn-1 cell line is characterized for motor neuronal and apoptotic characteristics in order to determine its utility as a model system for motor neuron cell death. In addition the effects of a variety of apoptotic stimuli are tested for their ability to induce cell death in Mn-1 cells. Finally, an oxidative model of motor neuron death is established using hydrogen peroxide to induce apoptosis. 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESULTS Mn-1 cell culture and differentiation Cells grown in DMEM with 10% FCS had polygonal cell bodies with occasional short processes. After growth in DMEM with 2% FCS (differentiation medium) for 72 hours cells showed an altered morphology consistent with differentiation (Brooks et al., 1998). The cell bodies were larger and the processes were considerably extended (Figure 2.1). Cells grown for 72 hours in differentiation medium also appeared to stop dividing as determined by cell counts before and after plating. Differentiated cells survived at least 6 days with regular changes of medium. Reverse transcriptase-polymerase chain reaction (RT-PCR) for motor neuron markers To determine if the Mn-1 cell line expressed markers consistent with a motor neuron lineage, three categories of expression were examined in both non differentiated and differentiated cells. All RT-PCR results were assayed as described in experimental procedures. High expression in comparison to the -actin control is scored as (+++), moderate as (++), low as (+), minimal as (+/-), and absent as (-). Figure 2.2 shows representative bands for each level of expression. The first group 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.1.a. Non-differentiated Mn-1 cells with occasional short processes, b. After 72 hours in differentiation medium (DMEM with 2% FCS), cell bodies appear larger and long processes have formed. 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2.2. Representative RT-PCR samples for semi-quantitative analysis of band strength. Bands are compared to a P-actin standard run with each set of samples and given a rating based on relative intensity levels. The levels are as follows: high (equal to or exceeding p-actin intensity) (+++), moderate (++), low (+), minimal (+/-), and negative (-). 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. looked at cell surface receptors, including neuronal cell adhesion molecule (NCAM), glutamate AMPA/kainate receptor subunits 1-4 (GluRl-4), tyrosine kinase receptors trkB and trkC, and the death receptor Fas (table 2.1). RT-PCR for NCAM showed a low (+) level of expression in non-differentiated cells and minimal (+ /-) expression in differentiated cells. AMPA receptor subunit expression showed that GluRl is minimally expressed (+/-) in both non-differentiated and differentiated states, while GluR2 appears to decrease after differentiation (from + to +/-). GluR3 is absent entirely, and GluR4 is present in low levels in both states (+). RNA from mouse spinal cord was a positive control for GluR3. The tyrosine kinase receptor trkB showed minimal expression (+/-) in non-differentiated and differentiated cells, while trkC was expressed at low levels (+). Fas was expressed at low levels (+) in non- differentiated cells and minimally (+/-) in differentiated cells. The second category of markers studied was general intracellular proteins, including an enzyme marker, neuron specific enolase (NSE), two calcium binding proteins, calbindin (Calb) and parvalbumin (Parv), and two intermediate filament proteins not normally found in adult motor neurons, vimentin (Vim) and glial Fibrillary acidic protein (GFAP) (table 2.2). Levels of NSE were low (+) in both cellular states, whereas calbindin and parvalbumin were negative in both. The positive control for calbindin was mouse cerebral cortex and for parvalbumin was mouse spinal cord. Vim expression was high (+++) in both cellular states, while GFAP was completely negative. The positive control for GFAP was mouse cerebral cortex. 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NCAM GluRl GluR2 GluR3 GluR4 TrkB TrkC Fas Non- differentiated + + / - + - + + / - + + Differentiated + / - + /- + /- - + + / - + + / - Table 2.1. RT-PCR for expression of motor neuron cell surface receptors in non- differentiated and differentiated Mn-1 cells. Semi-quantitative expression levels are defined in figure 2.2. Abbreviations used: neuronal cell adhesion molecule (NCAM), AMPA/kainate glutamate receptor subunit 1 (GluRl), subunit 2 (GluR2), subunit 3 (GluR3), subunit 4 (GluR4), tyrosine kinase receptor B (TrkB), tyrosine kinase receptor C (Trk C). 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NSE Calb Parv Vim GFAP Non- differentiated + - - +++ - Differentiated + - - +++ - Table 2.2. RT-PCR for expression of intracellular motor neuron markers non-differentiated and differentiated Mn-1 cells. Semi-quantitative expression levels are defined in figure 2.2. Abbreviations used: neuron specific enolase (NSE), calbindin (Caib), parvalbumin (Parv), vimentin (Vim), glial fibrillary acidic protein (GFAP). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The third category of markers contains homeodomain proteins involved in motor neuron differentiation (table 2.3). These include Pax6 and Nkx2.2, and four members of the Lim family of proteins: Isll, Isl2, Liml, and Lim3. Pax6 was expressed at low levels (+) in both non-differentiated and differentiated states, while Nkx2.2 was minimally expressed (+/-). Isll expression was moderate (++), Isl2 was low (+), Liml was low (+), and Lim3 was minimal (+/-) in both cellular states. RT-PCR for apoptotic markers In addition to looking at the motor neuronal characteristics of the Mn-1 line, I screened non-differentiated and differentiated cells for apoptotic markers. These can be divided into pro- and anti-apoptotic markers. The pro-apoptotic markers are bax, bad, and bid, along with four members of the caspase family of proteases, caspase-1 (Casl), caspase-3 (Cas3), caspase-8 (Cas8), and caspase-9 (Cas9) (table 2.4). The anti-apoptotic markers screened for include bcl-2, bcl-xl, and BAG-1 (table 2.5). Bax expression was moderate (++) in both states, while bad and bid were both minimally expressed (+/-). Caspase-1 expression was absent in non-differentiated cells, and minimally expressed in differentiated cells (+/-). Caspase-3 expression was moderate in both states, while caspase-8 expression was high (+++). Caspase-9 expression was minimal (+/-) in non-differentiated cells and low (+) in differentiated cells. For anti-apoptotic markers, bcl-2 expression was low (+) in non-differentiated 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pax6 Nkx2.2 Isll Isl2 Liml Lim3 Non- differentiated + + / - ++ + + + / - Differentiated + + /- ++ + + + / - Table 2.3. RT-PCR for expression of motor neuron developmental markers in non- differentiated and differentiated Mn-t cells. Semi-quantitative expression levels are defined in figure 2.2. Abbreviations used: islet-1 (Isll), islet-2 (Isl-2). 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bax Bad Bid Casl Cas3 Cas8 Cas9 Non- differentiated ++ + /- + / - - ++ +++ + /- Differentiated ++ + / - + / - + / - ++ +++ + Table 2.4. RT-PCR for expression of pro-apoptotic markers in non-differentiated and differentiated Mn-1 cells. Semi-quantitative expression levels are defined in figure 2.2. Abbreviations used: caspase-1 (Casl). caspase-3 (Cas3), caspase-8 (Cas8), caspase-9 (Cas9). 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bcl-2 Bcl-xl BAG1 Non- differentiated + ++ ++ Differentiated + /- ++ ++ Table 2.5. RT-PCR for expression of anti-apoptotic markers in non- differentiated and differentiated Mn-1 cells. Semi-quantitative expression levels are defined in figure 2.2. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cells and minimal (+/-) in differentiated cells. Bcl-xl and BAG-1 were expressed at moderate levels (++) in both states. Detection of apoptosis by flow cytometry After Mn-1 cells were screened for motor neuronal characteristics and apoptotic markers, a variety of apoptotic stimuli were applied to determine if apoptosis could be induced and, if so, what the strongest stimulus was. The three stimuli were sodium nitroprusside (SNP), camptothecin (Ct), and hydrogen peroxide (H1O2). Cells were treated with 300 M SNP, 2 M Ct, and a low (100 M) and high (500 M) concentration of H2O2, as described in experimental procedures. Cells were collected, stained with AV-F1TC and PI, and analyzed by flow cytometry. Cells positive for AV-FITC alone are considered to be in the early stage of apoptosis. Data was analyzed by Student’s t-test for comparison of differences between individual means. Figure 2.3 shows the results. All 4 treatments produced staining that was significantly higher than the control. SNP and 500 M H2O2 showed the highest percentages of AV-FITC stained cells, indicating around 17% of cells in both samples were in the early stages of apoptosis at their respective collection times. Analysis of DNA ladder formation To confirm the nature of cell death occurring by each treatment, DNA from each sample was extracted and examined for a DNA ladder pattern by agarose gel 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0 I I Treatment ■ Control ■ SNP (p< 0.001) ■ CT (p<0.005) □ H202 lOOuM (p<0.02) ■ H202 SO O uM (p<0.001) Figure 2.3 Percent Annexin-V-FITC (AV-FITC) positive cells after treatment with apoptotic inducers. Vertical axis is percent. Data are the average of 8 trials +/- S.E.M. Statistical analysis was performed using Student’s t-test to compare individual sample means to the control mean. 49 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. electrophoresis (Figure 2.4). In contrast to the flow cytometry results, the only treatment that was able to induce a DNA ladder pattern, indicating apoptosis, was a high concentration of HjOi. Even though SNP, Ct, and 100 M H2O2 showed signs of apoptosis by AV-FITC staining, none were able to induce a DNA ladder. Analysis of apoptosis by immunoblot for activated caspase-3 Since the pathways activated during apoptosis vary between cell types and not ail markers of apoptosis can be induced in all cell types, a third marker for apoptosis was used. In many apoptotic pathways, the final effector molecule responsible for producing several of the phenotypic characteristics of apoptosis is the caspase-3 protease. The inactive procaspase 3 is a 32 kD protein. When it is cleaved, a smaller 17 kD active form can be detected by immunoblot. Figure 2.5 shows the results of the immunoblot for active caspase 3 after all 4 apoptotic stimuli. Procaspase-3 can be detected in each sample, but the active 17 kD cleavage product is not present in any sample. Analysis of apoptosis by colorimetric enzyme assay Since only 500 M H2O2 was able to induce both AV-FITC staining and a DNA ladder, another, more sensitive caspase assay was attempted using H2O2 only. Active caspase-3 will cleave the synthetic substrate DEVD-p-nitroaniline (DEVD- 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123 bps SNP Ct H2 0 2 H2 0 2 100 uM 500 uM Figure 2.4. DNA ladder formation after treatment with apoptotic inducers. DNA was extracted from samples after treatment as described in experimental procedures. 10 pi of resuspended DNA was run on a 1.0% agarose gel containing ethidium bromide and visualized under UV light. Abbreviations used: 123 base-pair ladder (123 bps), sodium nitroprusside (SNP), camptothecin (Ct), hydrogen peroxide (H2 0 2). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SNP Ct H2O2 H2 0 2 Control 100 pM 500 pM 44— 29— ^ ----- Procaspase-3 2 0 - - ^ __ Active 15— Caspase-3 5 - - Figure 2.5. Immunoblot for detection of the 32 kD procaspase-3 and the 17 kD active caspase-3. Samples were electrophoresed on a 15% SDS-PAGE gel and transferred to nitrocellulose. Blots were developed using the ECL system (Amersham-Pharmacia). Samples were loaded in duplicate. Abbreviations used: sodium nitroprusside (SNP), camptothecin (Ct), hydrogen peroxide (H2O2). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pNA), releasing the chromophore p-nitroaniline. Absorbencies measured at 405 nm in cell lysates incubated with DEVD-pNA correlate with levels of active caspase-3. After the standard treatment procedure, 500 M H2O2 did not induce caspase activity detectable above control, consistent with the results of the immunoblot (Figure 2.6). Dose-response relationship of H2O2 by flow cytometry After screening by flow cytometry, DNA ladder formation, and caspase-3 activation, 500 M H2O2 was chosen as the best method of inducing apoptosis. Cells were treated with concentrations of H2O2 ranging from 0 to 1000 M and assayed by flow cytometry for apoptosis in order to generate a dose-response curve. Figure 2.7a shows that treatment with 500 M H2O2 produced the highest level of AV-FITC staining, indicating that this was the optimal treatment dose for inducing apoptosis. Above 500 M H2O2 AV-FITC staining decreased, reaching control levels again with 1000 M H2O2. This seemed unusual, so overall cell death was estimated by adding the single stained AV-FITC population to the double stained (AV-FITC + PI positive) population. Double stained cells are a combination of necrotic and late apoptotic cells, and so when added to the single stained AV-FITC population provide an estimate of overall cell death. Figure 2.7b shows that a nearly linear dose- response curve is obtained for overall cell death between 0 and 1000 M H2O2 . 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.297 0.287 Treatment ■ Control □ H202 SO O uiY l (p>0.05) Figure 2.6. Caspase-3 colorimetric enzyme assay to detect active caspase-3 after treatment with 500 pM H2O2 . Values are the mean absorbencies at 405 nm for 4 independent samples run in duplicate. Data was analyzed using Student’s t-test for comparison of means. 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30 25 a ) ^ 20 i C ' x a > § IS H < 10 5 * t , , — ,---------------- 0 200 400 600 800 1000 1200 H202 cone (uM) 100 90 80 <n a > o 70 CL T] > < 60 50 40 30 20 0 200 400 600 800 1000 1200 H202 cone (uM) Figure 2.7. Dose-response curve for treatment with H1O2. a. Percent AV-FITC positive cells by flow cytometric analysis, b. Percent AV-FITC single stained plus AV-FITC/PI double stained cells by flow cytometric analysis. Values are the mean +/- standard deviation for 8 independent trials run in duplicate. 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RT-PCR for apoptotic markers after H2O2 treatment The lack of caspase-3 activation in Mn-1 cells after treatment with apoptotic stimuli prompted a closer examination of the apoptotic pathways involved in H2O2- mediated cell death. RT-PCR was used to look for changes in levels of pro- and anti-apoptotic markers at 3 and 24 hours after treatment. The pro-apoptotic markers measured were bax, caspase-1, caspase-3, and caspase-9. The anti-apoptotic markers were bcl-2 and bcl-xl. Figure 2.8a shows a steady level of bax expression from control through 24 hours post-treatment. Figure 2.8b shows that caspase-1 is minimally expressed in untreated controls, and not detectable in either 3 or 24 hour treated samples. Caspase-3 (figure 2.8c) and caspase-9 (figure 2.8d) both show an unusual pattern, with a decrease in mRNA levels at 3 hours after treatment and a return to control levels by 24 hours after treatment. On the anti-apoptotic side, levels of bcl-2 are minimal in untreated controls and seem to decrease in 3 hour and 24 hour treated samples (figure 2.9a). Bcl-xl expression, like bax, does not change from control through 24 hours post-treatment (figure 2.9b). 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Co 3h 24h Co 3h 24h Figure 2.8. RT-PCR showing levels of pro-apoptotic markers in Mn-1 cells in control, 3 hours post-treatment, and 24 hours post-treatment. A. Bax B. Caspase- l(Casl) C. Caspase-3(Cas3) D. Caspase-9 (Cas9). Abbreviations: beta-actin (P~ actin), control (Co), 3 hours (3h), 24 hours (24h). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P-actin - p-actin Co 3h 24h Co 3h 24h Figure 2.9. RT-PCR for anti-apoptotic markers in Mn-1 cells at control, 3 hours post-treatment, and 24 hours post-treatment. A. Bcl-2 B. Bcl-xl. Abbreviations: beta-actin (P-actin), control (Co), 3 hours (3h), 24 hours (24h). 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DISCUSSION The goal of this study was to identify a cell line that could be used to study motor neuron degeneration. There are three requisite characteristics that are necessary in order to develop a successful model. First, the cell line should have a demonstrable motor neuron lineage. Although primary cultures would provide the closest model to in vivo motor neuron function, this technique is difficult and labor intensive, and the number of variables that can be studied at any one time is limited. In addition, each time new motor neurons are harvested, the number and type of motor neurons would be different, making it difficult to standardize any experiments. A motor neuron cell line, on the other hand, can be grown easily, and may provide more consistency from one experiment to another. While there are caveats that must be kept in mind when they are used, they may ultimately be a better choice for use in experiments such as those described here. Most cell lines are monoclonal populations of cells that change very little over several passages. This provides a steady model upon which one can optimize experimental conditions. In addition, if there are alterations to the cell line after several generations, it is always possible to return to an earlier passage that is identical to the original cells. The main drawback to using an immortalized cell line is that by its very nature it is abnormal, having lost the endogenous restraints on cell division normally present. For this reason it is necessary to determine how closely a cell line resembles the original cell type by looking for expression of markers characteristic of the 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sought after cell. The Mn-1 cell line is an immortalized cell line created by fusing embryonic day 13 primary mouse motor neurons with a mouse neuroblastoma. After screening, one monoclonal population was found to have a neuronal morphology and to express several markers that are consistent with a motor neuron lineage, including acetylcholinesterase (AChE), choline acetyl transferase (ChAT), and the heavy, medium, and light neurofilament proteins (NF-H, NF-M, NF-L) (Salazar-Grueso et al., 1991). This served to identify the Mn-1 line as a model for motor neurons. In successive experiments it was shown that the androgen receptor, a cell surface receptor usually found on motor neurons, was not expressed (Brooks et al., 1998). This irregularity casts doubt on the nature of the Mn-1 cell line. Although expression of the above markers is consistent with a motor neuron lineage, none of them provides absolute identification. Therefore it was necessary to perform further characterization to attempt to establish the lineage of the Mn-l cell line. The first step was to continue with the general characterization of neuronal markers consistent with, but not definitive for motor neuron lineage. These markers can be split into two basic categories: cell surface receptors and general intracellular proteins. NCAM is a general cell surface marker found on most neurons, including motor neurons (Chen and Chiu, 1992). The Mn-1 cell line showed low to minimal levels of expression of NCAM, which is consistent with neuronal lineage. A second type of cell surface receptor, glutamate receptors, are normally found in adult motor neurons, and the AMPA/kainate receptor subtype has been implicated in excitotoxic damage in ALS. One hypothesis is that AMPA/kainate 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. receptors in motor neurons lack the GluR2 subunit, which confers calcium impermeability to the receptor in most regions of the CNS. Without this subunit, the receptor becomes permeable to calcium, producing an increase in intracellular calcium that appears to be the link between glutamate over-stimulation and cell death (Doble, 1999; Vandenberghe et al., 2000). The Mn-1 cell line showed only minimal to low levels of expression of any of the subunits, and lacked the GluR3 subunit entirely. Although this was not the pattern expected for a motor neuron, it does not rule out motor neuron lineage since a cell line is not expected to express every marker characteristic of the native cell type, and lack of expression of GluR3 could be specific to an as yet unidentified subset of motor neurons, similar to low or absent expression of GluR2. In addition, minimal expression of the AMPA/kainate receptor allows for a unique opportunity to study disease mechanisms in ALS. Recent evidence indicates that disease onset and disease progression in ALS may be the result of two separate processes. Disease onset has been linked to oxidative damage whereas disease progress has been linked to excitotoxicity (Facchinetti et al., 1998). A motor neuron model that does not respond to AMPA/kainate receptor stimulation could be used to study the contribution of oxidative stress to cell death separate from glutamate toxicity. Another family of receptors involved in cell death is the neurotrophin receptor family. Neurotrophins are signaling molecules involved in neuronal growth, differentiation, and death (Majdan and Miller, 1999). In general, neurotrophins help neurons establish connections with target cells during 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. development, and then are required by neurons throughout life to survive (Kemic and Parada, 2000). Removal of neurotrophins has been shown to induce cell death in a variety of neuronal populations, while treatment with neurotrophins supports motor neuron survival in vitro and can interfere with disease progression in some mouse models of motor neuron degeneration (Chao et al., 1998; Kernic and Parada, 2000; Sendtner et al., 2000). As a result of these studies, neurotrophins have been used in trials to treat ALS (Louvel et al., 1997). These trials have been largely unsuccessful, however, due to side effects and delivery problems (Kernic and Parada, 2000). Clearly more research is needed in this potential area of therapy for ALS. Motor neurons respond to three specific neurotrophins: brain-derived neurotrophic factor (BDNF), neurotrophin 3 (NT-3), and neurotrophin 4/5 (NT-4/5). BDNF and NT-4/5 act via the tyrosine kinase receptor trkB, while NT-3 binds to a related receptor, trkC. In Mn-1 cells both trkB and trkC are expressed, providing a possible model to study the effects of neurotrophins on motor neuron cell death. Finally, Fas is a member of the tumor necrosis factor receptor family, which is involved in mediating apoptosis via an intracellular pathway that directly activates the caspase cascade (Lincz, 1998). Fas ligand (FasL) has recently been found to be involved in mediating normal embryonic cell death in motor neurons via a motor neuronal Fas receptor (Sendtner et al., 2000). Physiologic cell death during development is another pathway being studied to help understand aberrant cell death in adult motor neurons. Fas was expressed at low levels in undifferentiated neurons, and decreased to minimal expression in differentiated cells, consistent with a 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. function in early, embryonic neurons. This pattern of expression in Mn-l cells may provide a means to study physiologic cell death in motor neuron development and its relevance to adult motor neuron disease. The second category of proteins used to characterize the Mn-1 cell line was intracellular proteins. A general neuronal marker, neuron-specific enolase, was expressed at moderate levels, while a marker for glial cells, GFAP, was negative, as would be expected in neurons. Two cytosolic calcium binding proteins, calbindin and parvalbumin, have been implicated in motor neuron susceptibility to glutamate excitotoxicity and increased intracellular calcium. Motor neurons lack both of these proteins, and for this reason it is thought that they are unable to buffer levels of intracellular calcium that most neurons can handle, leading to increased motor neuron death in situations of glutamate over-stimulation. The Mn-l cell line lacks both calbindin and parvalbumin, consistent with native motor neuron expression patterns, adding another variable that can be examined in the motor neuron death pathway in Mn-l cells. One anomaly did arise in screening for this set of markers. Vimentin, an intermediate filament protein that is not normally found in adult motor neurons in significant quantities, was strongly expressed in both non-differentiated and differentiated Mn-l cells. This could be due to the nature of the neuroblastoma parental line. While all of the markers studied are consistent with motor lineage, and often useful for studying motor neuron disease, no marker identified so far is definitive for distinguishing a motor neuron from other neuronal types. Potentially the most useful 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. data for identification of motor neurons may come from the field of motor neuron development. Six homeodomain proteins have been identified that are expressed during and after motor neuron development, and certain combinations have been used to characterize the embryonic location of developing motor neurons (Goulding, 1998; Tanabe et al., 1998; Tsuchida et al., 1994). The first two are Pax6 and Nkx2.2. Pax6 is induced in ventral spinal cord progenitor cells that are destined to become somatic motor neurons (SMNs). Nkx2.2 is expressed in cells destined to become visceral motor neurons. In the Mn-l cell line Pax6 expression is low, and Nkx2.2 expression is minimal. Although it seems contradictory to express both markers, some Pax6 populations have been found where Nkx2.2 expression occurs transiently during development. In these cases, Pax6 appears to dominate, possibly by directly repressing Nkx2.2, sending the cells down a path to become SMNs (Ericson et al., 1997). Pax6, although important to somatic motor neuron development, is also expressed by some classes of ventral intemeurons (Ericson et al., 1997). It is therefore necessary to look further for definitive evidence of a motor neuron lineage. This can be accomplished by screening Mn-l cells for expression of 4 members of the Lim homeodomain family. These four markers, Isll, Isl2, Liml, and Lim3, are expressed variably in SMNs destined to innervate various regions of musculature. The developing spinal cord can be divided in to four columns of SMNs: medial and lateral regions are first defined, and then each of these is split again into medial and lateral. The four columns are termed medial column medial (MCM), medial column 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lateral (MCL), lateral column medial (LCM), and lateral column lateral (LCL). Each column will innervate a distinct group of muscles. The MCM neurons will innervate axial musculature near the spinal cord, MCL neurons will project to muscles of the body wall, LCM neurons will innervate limb muscles developed from the ventral muscle mass, and LCL will innervate limb muscles derived from the dorsal muscle mass (Tsuchida et al., 1994). The specificity of projections is thought to be determined by variable expression of the Lim proteins (see figure 1.2). MCM neurons express Isll, Isl2, and Lim3, both MCL and LCM cells express Isll and Isl2 without Lim3, and LCL cells express Isll, Isl2, and Liml. Depending on the stage of development, Isll expression may already be down-regulated in LCL cells, although there is a period of co-expression of Isll, Isl2, and Liml (Tsuchida et al., 1994). Mn-l cells expressed moderate levels of Isll, confirming their identity as cells of motor neuron origin. Isl2 and Liml were both expressed at low levels, and expression of the fourth Lim family member, Lim3, was minimal to non-existent in both non-differentiated and differentiated cells. The combination of Isll, Isl2, and Liml identify the Mn-l cell line as having originated in the cell population destined to be part of the LCL, removed at a developmental stage before expression of Isll had been down- regulated. LCL neurons will innervate limb muscles, and so the Mn-l cell line is derived from cells that are directly affected in ALS, providing an excellent model of motor neuron degeneration. 65 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The second feature necessary in a cell line that will be used to study motor neuron degeneration is the expression of appropriate apoptotic markers. Apoptosis has been implicated in several forms of neurodegeneration, including ALS, but the data has not been definitive (He and Strong, 2000; Mighelli et al., 1999; Mu et al., 1996; Troost et al., 1995). Cells used to model motor neuronal death must express certain pro- and anti-apoptotic markers to be useful in differentiating between apoptosis and necrosis. Of the pro-apoptotic markers screened, expression of bax, the prototypical pro-apoptotic marker, was moderate in both non-differentiated and differentiated states, while expression of both bad and bid was minimal. This is consistent with studies showing that bax is constitutively present in the cytoplasm, but is held in check by anti-apoptotic factors (Pettman and Henderson, 1998). Another pro-apoptotic marker, caspase-1, also known as interleukin 1- converting enzyme (ICE), is primarily attributed with a role in inflammation. Some studies have postulated a role for it in apoptosis, but this is still controversial (Ankarcrona et al., 1994; Gurney et al., 2000; Pasinelli et al., 1998). Mn-l cells were negative for caspase-1 in the non-differentiated state, but showed low expression in the differentiated state. This indicates that, if caspase-l is involved in apoptosis, differentiated Mn-l cells may be more primed for apoptosis than non- differentiated cells. Caspase-3 expression was moderate while caspase-8 expression was high in both non-differentiated and differentiated states. Caspase-3 is considered the final effector in classical apoptotic pathways, and is responsible for cleaving a wide array 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of intracellular targets, ultimately resulting in characteristic features of apoptosis, such as DNA fragmentation and formation of apoptotic bodies (Allen et al., 1998; Lincz, 1998). Caspase-8, along with caspase-9, is considered an initiator caspase. Initiator caspases are upstream from caspases-1 and -3, and responsible for activating the downstream caspases. Upstream caspases can be activated either directly by death receptor binding (for example, via Fas-FasL interactions), as in the case of caspase-8, or indirectly by cellular insults, for caspase-9 (Lincz, 1998). High levels of caspase-8 expression indicate that the Mn-l cell line might be useful for studying receptor-mediated cell death found during embryogenesis. Caspase-9 is the initiator caspase implicated in apoptosis due to cellular insults, including oxidative dysregulation (Saikumar et al., 1999). Non-differentiated Mn-l cells express caspase-9 minimally and expression increases to low levels in differentiated cells, supporting the idea that differentiated cells are more primed for apoptosis than non- differentiated. Anti-apoptotic markers included the prototype of this family, bcl-2, a closely related molecule, bcl-xl, and a non-bcl-2 family member, BAG-1. Bcl-2 expression was low to minimal in Mn-l cells, while bcl-xl and BAG-1 expression were both moderate. This is consistent with reports that in the adult nervous system bcl-2 is down-regulated and bcl-xl dominates (Gonzalez-Garcia et al., 1995). BAG-1, which does not share homology with the bcl-2 family, was expressed at moderate levels in Mn-l cells. Although the mechanism of action is currently unknown, BAG-1 has been shown to interact with bcl-2 to prevent apoptosis. Overall, the Mn-l line 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. expresses markers appropriate to the study of apoptotic pathways, fulfilling the second requirement for designing a model of motor neuron cell death. The final requirement is the ability of the cell line to respond to apoptotic signals. Three different tests were used to look for the occurrence of apoptosis due to a variety of insults. Non-differentiated cells were treated with camptothecin, which acts as a topoisomerase I inhibitor, or one of two oxidative insults, SNP, which generates NO, or H2O2, which acts by increasing levels of intracellular ROS. By flow cytometry, SNP and 500 M H2O2 were nearly equal in inducing apoptosis, but only H2O2 was able to induce a DNA ladder. The implication is that some features of apoptosis are induced by some insults, while others are not. This is inconsistent with the classical view of apoptosis, in which cell death occurs accompanied by a characteristic set of morphological and biochemical alterations. To further define the apoptotic pathway involved with these insults, two different assays were run for activated caspase-3. Both were negative, confirming the fact that an atypical apoptotic pathway is activated in Mn-l cells. Only 500 M H2O2 was able to induce 2 out of 3 of the standard markers for apoptosis, indicating that this insult is the best choice to further study apoptotic cell death in a motor neuron model. Mn-l cells were treated with concentrations ranging from 0 to 1000 M H2O2, and assayed for apoptosis by flow cytometry. The highest percentage of AV- FITC single stained cells was induced with 500 M, with a drop in AV-FITC positive cells at higher concentrations. One possible explanation for this is that as 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the concentration of H2O2 increases, cell death shifts from an apoptotic to a necrotic form, with both types occurring at some concentrations. Levels of apoptotic markers before and after treatment with H2O2 were also examined by RT-PCR. The markers examined were bax, caspase-1, caspase-3, caspase-9, bcl-2, and bcl-xl. Bax expression was unchanged throughout the treatment period, indicating either that bax is not involved in this form of cell death, or that levels of bax protein are sufficient to promote cell death without an increase in transcription. Caspase- 1 was expressed at extremely low levels in untreated controls, and was not detectable at all 3 hours after treatment. This indicates that caspase-1 is not involved in apoptotic cell death induced in Mn-l cells by H2O2 . In contrast, both caspase-3 and caspase-9 show an unusual pattern of expression, with a decrease in apparent expression 3 hours after treatment and an increase to control levels by 24 hours. This could be explained by an increase in translation without an increase in transcription by 3 hours after treatment. The cell could be stockpiling procaspases after the initial insult. By 24 hours, the level of caspase expression has increased, indicating transcription of new message, or a decrease in translation with a steady level of transcription throughout. The meaning of this is unclear, however, since caspase-3 activation does not seem to play a part in apoptosis induced by H2O2 in this system. The increase in translation, resulting in an apparent decrease in message, might be a false start by the cell, in which any insult results in increased translation of caspases, but later signals result in a caspase- independent pathway, so levels of mRNA once again rise to pre-treatment quantities. 69 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The fact that the anti-apoptotic marker bcl-2 is expressed at only minimal or less levels, suggests that it does not seem to play a strong role in Mn-l cell apoptosis. Bcl-xl, a molecule with similar structure and function to bcl-2, but with a higher level of expression in the adult nervous system, is steady throughout, similar to bax. Again, this could indicate that bcl-xl is not involved in regulating this form of cell death, or that the balance between transcription and translation are adequate for the cellular processes occurring after treatment with 500 M H2O2 . In conclusion, in this study the Mn-l line has been shown to be of likely motor neuronal origin in that the expression patterns of homeobox genes mimic what is seen in the developing spinal cord, particularly those ceils characteristic of the LCL. The first symptoms in many cases of ALS involve asymmetric paralysis in the limbs, and so a cell line derived from the anterior horn cell precursors that would innervate these muscles provides an excellent system in which to model the disease. In addition, I showed that the Mn-l cell line expresses an array of pro- and anti- apoptotic markers. The expression of these genes indicates that Mn-l cells are capable of undergoing apoptosis should the appropriate signal be received. Finally, 500 M H2O2 was found to provide an oxidative insult that appears to induce a caspase-independent form of apoptosis in Mn-l cells. This is important because oxidative dysregulation has been implicated in ALS, and the evidence that an oxidative insult produces an apoptotic phenotype in a motor neuron model supports the theory that apoptosis contributes to cell death in ALS. The apoptotic pathway, however, is not a classical pathway involving activation of caspase-3, but rather 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. seems to proceed via an alternate caspase-independent pathway that still produces DNA fragmentation. This could have important therapeutic implications for attempts to prevent neuronal death by interfering with apoptotic mechanisms. Caspase inhibitors would not be effective, while the existence of an alternate path suggests the possibility of blocking cell death at an as yet unidentified step. Further study into the nature of cell death produced by oxidative insult in Mn-l cells is necessary to in order to better understand the process of motor neuron degeneration found in ALS. 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EXPERIMENTAL PROCEDURES Cell Culture Mn-l cells were a gift from Kenneth Fischbeck. Cells were grown in Dulbecco’s modification of Eagle’s medium (DMEM) (Gibco) containing 10% fetal calf serum (FCS) (growth medium). Cells were split by washing once with phosphate buffered saline (PBS) without calcium and magnesium (Irvine Scientific) and then incubated with 0.05% trypsin-EDTA for 3 minutes at 37°C. Cells were split 1:3 once per week. For use in experiments cells were allowed to grow for 48 hours before any treatments were administered. To produce the differentiated phenotype cells were plated in DMEM containing 2% FCS (differentiation medium) for 72 hours before being used in experiments. Differentiation was confirmed by extension of processes and cessation of cell division. Treatments were performed in the same medium (high or low serum) in which the cells were grown. Reverse Transcriptase-Polymerase Chain Reaction Total RNA was extracted from cells using the RNeasy Mini Kit (Qiagen) following the manufacturer’s instructions. After extraction RNA was resuspended in RNAse-ffee water and the concentration was determined using a spectrophotometer. 100 picograms (pg) of each sample were used for reverse transcriptase-polymerase 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. chain reaction (RT-PCR). RT-PCR was performed using the OneStep RT-PCR Kit (Qiagen) according to the manufacturer’s instructions. The basic program for * reverse transcription and amplification was as follows: 50°C for 30 minutes, then 95°C for 15 minutes, followed by 35 cycles of 94°C for 30 seconds, 50 to 65°C (depending on the primer pair) for 30 seconds, and 72°C for I minute, finished with a 10 minute final extension at 72°C. Primers were designed based on mRNA sequences available from GenBank. -actin primers for controls were purchased from Promega. For a list of individual primer sequences, annealing temperatures, and expected product lengths see Appendix A. After RT-PCR 10 1 of product was visualized on a 1.0 % agarose gel containing ethidium bromide. Images were captured using a Kodak Eagle Eye II system. Results were assessed based on a semi-quantitative analysis of intensity in comparison to a -actin control. Bands were assigned a score of +++ (high) for intensity matching or exceeding the -actin control, ++ (moderate), + (low), +/- (minimal), or - for absent bands. Primers producing absent bands were tested against other tissue, including mouse spinal cord, mouse cerebral cortex, and a mouse neuroblastoma cell line, N2a, to insure that the primers were functional. Results are the estimated average of three independent trials. Representative gels from each RT-PCR can be found in Appendix B. 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Induction of apoptosis To induce apoptosis cells were treated with sodium nitroprusside (SNP) (Calbiochem), camptothecin (Sigma), or hydrogen peroxide (Sigma). For treatment with SNP, medium was replaced with appropriate high or low serum medium containing 300 M SNP and cells were incubated at 37°C for 10 minutes. After incubation the medium containing SNP was removed and replaced with fresh medium and the cells were allowed to incubate at 37°C for 6 hours (Toku et al., 1998). Camptothecin was added to fresh medium at a concentration of 2 M and cells were allowed to incubate for 6 hours at 37°C (adapted from Pharmingen’s Apoptosis Instruction Manual, 1998). Hydrogen peroxide (30% stock) was diluted in water and added to fresh medium in concentrations ranging from 100 to 1000 M and cells were incubated for 3 to 24 hours at 37°C. Annexin V-Propidium iodide staining For flow cytometry 1 x 106 cells were plated in appropriate medium in a T-25 flask. After treatment, the supemate was collected into 15 ml tubes on ice and the flasks were washed once with PBS without calcium and magnesium. Cells were trypsinized and added to the supernate to insure that detached cells were included. Cells were spun down, washed twice with PBS, and then treated with annexin V- fluorescein isothiocyanate (AV-FITC) and propidium iodide (PI) according to the 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. directions for the TACS Annexin V-FITC Apoptosis Detection Kit (R&D Systems). Samples were assayed on a FACStar flow cytometer. Results are the average of 8 independent trials. Statistical analysis was performed using either Student’s t-test for comparison between individual means or ANOVA for multiple means. Assessment of DNA ladder Extraction of DNA for assessment of DNA ladder formation was performed essentially as described with minor modifications (Toku et al., 1998). 3 x 106 cells were plated in a 100-mm dish in appropriate medium. Collection was the same as for flow cytometry. After the cells were washed, the cell pellets were lysed for 10 minutes in hypotonic lysis buffer (lOmM Tris-chloride, pH 7.4, lOmM EDTA, 0.5% Triton X-100) on ice. Lysates were centrifuged at 14,000 x g for 25 minutes at 4°C to pellet intact chromatin. The supemate was transferred to a clean tube and incubated with 35 g of RNAse A (Boehringer-Mannheim) for I hour at 37°C, followed by 40 g of proteinase K (Boehringer-Mannheim) for 1 hour at 37°C. After incubation the lysate was mixed with an equal volume of 1:1 phenol chloroform, vortexed, and centrifuged for 5 minutes at 14,000 x g at room temperature. The aqueous portion was transferred to a clean tube and incubated with an equal volume of isopropanol, 50 M sodium chloride, and 5 nanograms (ng) per microliter (1) of glycogen (Boehringer-Mannheim) at -20°C for 18 hours. After incubation samples were centrifuged at 14,000 x g for 20 minutes, air dried, and 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. resuspended in TE buffer (lOmM Tris, ImM EDTA, pH 7.4). Resuspended DNA was visualized as described for RT-PCR. Immunoblot for active caspase-3 Cells were plated at a density of 1 x 106 cells per plate in 100-mm plates. After treatment cells were collected as described previously. Cell pellets were then lysed in a triple detergent lysis buffer (50 raM Tris-chloride, pH7.4, 150 mM NaCl, 5 mM EDTA, 0.5% NP-40,0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate) on ice for 30 minutes. The lysate was centrifuged at maximum speed in a microcentrifuge at 4°C for 15 minutes. The cleared supernatant was transferred to a clean tube and total protein was assayed for according to the method of Lowry (Bensadoun and Weinstein, 1976). Lanes were loaded with 100 g of total cellular protein, separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and transferred overnight at 4°C to a nitrocellulose membrane (Schleicher and Schuell). Immunoblots were performed as described by Harlow and Lane. Blots were incubated with goat anti-mouse caspase-3 antibody (sc-1224) (SantaCruz) at a dilution of 1:200. The secondary antibody was horseradish peroxidase derivitized rabbit anti-goat IgG (sc-2020) (SantaCruz) at a dilution of 1:10,000. The Enhanced Chemiluminescence (ECL) kit (Amersham) was used to visualize the secondary antibody. Blots were exposed to Hyperfilm ECL (Amersham) and developed. 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Colorimetric enzyme assay for active caspase-3 Activation of procaspase-3 to caspase-3 can be assayed for using a synthetic substrate that, when cleaved by caspase-3 in cell lysates, produces a characteristic color change. The intensity of this color change can be determined by spectrophotometric analysis at 405nm and correlated to the amount of active caspase-3. 5 x 106 cells were plated in T-75 flasks and treated with 500 M HiCK 24 hours after treatment, cells were collected and processed according to the directions for the Caspase-3 colorimetric assay kit (R&D Systems). Data represent the mean values obtained from 4 independent samples run in duplicate. Statistical analysis was performed using Student’s t-test to compare individual means. 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3 Bcl-2 INHIBITS CASPASE-INDEPENDENT APOPTOSIS IN A MURINE TISSUE CULTURE MODEL OF FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS INTRODUCTION Amyotrophic lateral sclerosis (ALS) is a degenerative disease of the central nervous system. Motor neurons located in the anterior horn of the spinal cord and neurons of the motor cortex that form the lateral corticospinal tracts die, resulting in a clinical disease characterized by mixed upper and lower motor neuron symptoms. The end result of motor neuron degeneration is paralysis and death. The ultimate cause of neuronal death in ALS is not yet understood. Like many neurological disorders, ALS can be either sporadic (SALS) or familial. Although clinically indistinguishable from each other, familial ALS (FALS) has been associated with a variety of genetic mutations. Nearly twenty percent of FALS cases have a mutation in the cytoplasmic copper/zinc superoxide dismutase gene (SOD 1) on chromosome 21q and are inherited in an autosomal dominant fashion. The connection between SOD1 mutations and motor neuron death has been studied in tissue culture and in transgenic mouse models, but the link remains elusive. 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. When the association between certain cases of FALS and SOD1 mutations was discovered, an early theory was proposed that mutant SOD1 had decreased activity, resulting in increased levels of intracellular reactive oxygen species (ROS), such as the superoxide anion and hydroxyl radicals, which would damage a variety of intracellular structures and impair mitochondrial energy production (McNamara and Fridovich, 1993). Loss of cellular energy equilibrium results in loss of cell integrity, decreased function of cellular ion pumps, increased intracellular calcium, and cell death (Coyle and Puttfarcken, 1993; Doble, 1999). However, a loss of function mutation is inconsistent with the autosomal dominant form of inheritance of FALS, and recent evidence has shown that although many of the SOD1 mutants do have decreased activity, at least one has normal function and still causes FALS (de Belleroche et al., 1995). In addition, transgenic knockout mice lacking SOD1 do not develop motor neuron disease, implying that it is the mutation that causes disease (Robberecht, 2000). The current theory is that mutant SOD1 has gained a toxic function. Transgenic mice over-expressing a wild type human SOD1 do not develop disease, while mice expressing a mutant SOD1 become progressively paralyzed in a manner similar to human ALS (Brujin et al., 1998; Cleveland et al., 1996; Gurney et al., 1994; Ripps et al., 1995; Wong et al., 1995). This indicates that the gain of function is not simply an increase in the normal dismutase action of the enzyme, supported by the aforementioned decrease in activity of most of the mutants, but rather the gain of a novel toxic function. Two hypotheses have been proposed to explain the gain of 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. function. The first is an increase in a normally low-level peroxidase reaction by SOD1, producing OV - from H2O2 . The second involves an increased ability of peroxynitrite, ONOO\ formed by the interaction of OV - and nitric oxide (NO), to interact with O2 producing a highly reactive nitronium intermediate. This intermediate could react with tyrosine in proteins, potentially resulting in inactivation of certain key cellular proteins and signaling pathways (Cleveland, 1996). Either method, or both in concert, could contribute to an altered cellular oxidative state and cellular and mitochondrial damage, producing cell death. Indeed, signs of oxidative damage in both human ALS spinal cord and transgenic mouse models seem to indicate that excess ROS play a part in the disease mechanism (Robberecht, 2000). One question that has been studied in depth is the nature of cell death in FALS, whether it is necrotic, characterized by a toxic insult resulting in cell lysis, or apoptotic, proceeding by an aberrant programmed cell death mechanism. This question has important therapeutic implications. There is mounting evidence that cell death in ALS proceeds via a programmed cell death mechanism, but the data thus far has not clearly identified it as apoptosis. Some studies have found increased levels of Bax and decreased levels of Bcl-2 in motor neurons of SALS spinal cords, while others have found no changes (Mu et al., 1996; Troost et al., 1995). DNA fragmentation has been demonstrated in SALS spinal cord tissue by a technique called in situ end labeling (ISEL), which enzymatically labels cleaved 3’ ends of DNA (Troost et al., 1995). Using a related technique called TUNEL another group 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. could not confirm this finding (He and Strong, 2000). Finally, only one group has shown caspase activation in human ALS tissue (Martin, 1999). To further address the question of apoptotic versus necrotic cell death, models of FALS have been created in both transgenic mice and in tissue culture. Mice expressing a variety of mutations found in FALS have been created, including SODL mutations (glycine to arginine at positions 37, 85, and 86, and glycine to alanine at position 93) and over-expression of the neurofilament light chain (Brujin et al., 1998; Cleveland et al., 1996; Gurney et al., 1994; Ripps et al., 1995; Wong et al., 1995). Cytopathological studies comparing the motor neuron degeneration in FALS mice with a glycine to alanine mutation at position 93 (G93A) to human FALS spinal cord sections show a number of similarities, including anterior horn atrophy, gliosis, axonal swelling, neurofilament aggregation, and Lewy-type bodies (intra-cytoplasmic inclusions) (Dal Canto and Gurney, 1994; Tu et al., 1996). Attempts to identify the type of motor neuron death in the G93A mouse model have provided mixed results, paralleling similar attempts in human ALS. One group reported that G93A mice lack standard signs of apoptosis, including evidence of DNA strand breaks and activation of the apoptosis-associated protease, caspase-3, while another group found evidence of both caspase-3 and a related protease caspase-1, to be activated in G93A motor neurons (Li et al., 2000; Mighelli et al., 1999). In addition, several groups have shown that expression of the anti-apoptotic protein bcl-2 can alter progression of motor neuron degeneration in G93A mice (Assouz et al., 2000; Kostic et al., 1997; Vukosavic et al., 2000). This does not 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. confirm that the process is apoptotic, though, because bcl-2 has also been postulated to possess anti-cell death functions not related to known anti-apoptotic pathways (Kane et al., 1993; Tyurina et al., 1997). In tissue culture, non-motor neuron cell lines, including two neuroblastoma lines (SH-SY5Y and N2a), a pheochromacytoma line (PC12) and yeast cells have been transfected with both wild-type and mutant forms of SOD1 (Carri et al., 1997; Ghadge et al., 1997; Pasinelli et al., 1998; Rabizadeh et al., 1995). In SH-SY5Y cells, over-expression of a mutant SOD1 produces altered mitochondrial function, an increase in intracellular calcium, increased production of ROS, and decreased levels of bcl-2. Although each of these changes could be related to an apoptotic process, each one has also been associated with necrosis, with the exception of decreased bcl- 2. However, as mentioned previously, bcl-2 may have functions unrelated to apoptosis, and it might be these activities that are altered in the presence of mutant SOD1. In N2a cells, expression of mutant SOD1 resulted in activation of caspase-1, but the contribution of caspase-1 to apoptosis is still controversial (Gurney et al., 2000; Wyllie, 1997). Despite each of these objections, none of the data eliminates the possibility of an apoptotic process, and these models are useful in examining the effects of mutant SOD1 in neural cells. It is necessary to keep in mind that the pathway to apoptosis differs from cell to cell, and therefore the best model of motor neuron death would be a bona fide motor neuron. Primary cultures from FALS mice could serve as a source of motor neurons, which has been done with some success (Assouz et al., 2000). However, 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. isolation of motor neurons from mice is technically difficult and time consuming, and thus limits the number of experimental variables that can be addressed. In addition, each time cells are harvested the composition and type of motor neurons is different, and this introduces another variable into any experiments performed. To overcome the difficulties inherent in working with primary cultures, a motor neuron cell line, the Mn-l line, has been designed. This cell line allows a greater number of variables to be tested by using a cell type that remains stable over several generations. In chapter 2 the Mn-l cell line was shown to express homeodomain genes consistent with an origin in the motor neuron precursor population that forms the lateral column lateral (LCL) in the developing spinal cord. These neurons will innervate limb muscles, and so are among the neurons that would be affected in ALS. In addition, it was shown that the Mn-l cell line expresses a variety of apoptotic markers, and that an oxidative insult can produce features of apoptosis. The form of apoptosis, however, does not appear to follow the classical pathway, but rather proceeds via a caspase-independent path. Despite the overall uncertainty of the cell death pathway in ALS, most data suggests that a form of apoptosis is involved. As a result, gene therapy aimed at prevention of cell death by inhibition of apoptotic processes has been the focus of several recent studies. The bcl-2 family includes a number of proteins involved in the regulation of apoptosis, and contains both pro- and anti-apoptotic members. The prototype of the anti-apoptotic members is bcl-2 itself. Bcl-2 has been shown to 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. interfere with apoptosis in a variety of systems, including neurons (Saikumar et al., 1999). Two recent studies using the G93A mouse show that bcl-2 can have a modulatory effect on motor neuron death. The first study used transgenic mice over expressing bcl-2 crossed with G93 A FALS mice to show that expression of bcl-2 could delay onset of disease and increase the number of anterior horn motor neurons during disease progression. Bcl-2 was unable, however, to alter the length of disease, and by end-stage both bcl-2-expressing and control mice showed equivalent losses of motor neurons (Kostic et al., 1997). Assouz, et al, used an adeno- associated viral vector to express bcl-2 after direct injection of the virus into the lumbar spinal cord. They were able to show an increased number of motor neurons and a delay in the onset of symptoms in areas adjacent to the site of injection (Assouz et al., 2000). Although the data is useful in showing that bcl-2 can alter motor neuron death in a model of FALS, the therapeutic utility of either of the above approaches is limited. A more desirable goal would be to use a delivery system that does not involve such invasive procedures. A recent study showed that intra-muscular injection of an adenoviral vector containing the gene for lacZ resulted in retrograde axonal transport in motor neurons and expression of the lacZ protein. This approach is clearly superior to highly invasive procedures. Additionally, the use of an adenovirus instead of an adeno-associated virus would result in a transient expression of the gene product delivered. Although this can be a disadvantage of 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. adenoviral vectors, it may be that transient expression of bcl-2 is preferable to uncontrolled constitutive expression. If expression is lost, treatment can be repeated, but it is more difficult to turn off expression once begun. The goal of this study was to create a model of SOD 1-associated FALS in the Mn-1 cell line using a human SOD1 cDNA containing the G93A mutation. Mn-1 cells expressing either wild-type SOD1 (MNWT cells) or G93A SOD I (MNGA cells) were treated with H2O2 and examined for signs of cell death by apoptosis. In addition, the effect of an adenoviral vector containing the human bcl-2 cDNA on cell death in MNGA cells was studied. 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESULTS Transient expression of WT-SOD1 and G93A-SOD1 To test the ability of the Mn-1 cell line to express either the WT-SOD1 or the G93A-SOD1, cells were transfected with one of the plasmids as described in experimental procedures. Beginning at 24 hours, expression was analyzed by RT- PCR using primers specific for the human form of SOD1. Figure 3.1 shows that both WT-SOD1 and G93A-SOD1 can be expressed, and that expression lasts for up to 5 days without a significant decrease in levels. Stable expression of WT-SOD1 and G93A-SOD1 After confirming that Mn-1 cells could express both WT-SOD1 and G93A- SOD1, stable cell lines were created as described in experimental procedures. RT- PCR for expression of WT-SOD1 in the MNWT clonal line and G93A-SOD1 in the MNGA clonal line confirmed expression of the appropriate mRNA (Figure 3.2a). Expression of SOD1 proteins was confirmed by immunoblot with ECF (Figure 3.2b). 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NT MT W1 G l W2 G2 W3 G3 W4 G4 W5 G5 hSODl 6-actin Figure 3.1. RT-PCR for human SOD1 in transiently transfected Mn-l cells. Samples were collected beginning 24 hours after transfection, and every day after up to five days. Abbreviations: human SOD1 (hSODl), WT-SOD1 (W), G93A-SOD1 (G), non-transfected (NT), mock transfected (using an empty pcDNA3.1+ plasmid) (MT); days are indicated by number (1 for day 1, 2 for day 2, and so on). (3-actin is the beta actin control. 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MNGA hSODl r hSODl MNGA hSODl Figure 3.2. Stable expression of human WT-S0D1 in MNWT cells and G93A- SOD1 in MNGA cells. A. RT-PCR for human SOD1 (hSODl). B. Immunoblot using ECF for human SOD1. Positive human SOD1 control for RT-PCR was the human retinoblastoma cell line Weri-Rb-1 (WR1), and for immunoblot was human recombinant SOD1 (rhSODl) (Sigma). 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Expression of apoptotic markers in MNYVT and MNGA cell lines After establishing that the MNWT and MNGA cell lines were expressing the proper transgene, RT-PCR was performed to see if expression of either form of SOD1 resulted in alteration of apoptotic markers. The Mn-1 line was previously characterized for expression of the pro-apoptotic markers bax and caspases-1,-3, and -9, and the anti-apoptotic markers bcl-2 and bcl-xl, and a semi-quantitative scale for evaluating expression was described (see chapter 2). Bax expression seems to have increased in both MNWT and MNGA cells compared to Mn-1 cells from low (+) to moderate (++) levels, (Figure 3.3a), while caspase-1 (Figure 3.3b) and bcl-2 (Figure 3.4a) levels remained barely detectable, with minimal expression (+ / -) in all 3 lines. Levels of caspase-3 (figure 3.3c) were also unaffected by either WT-SOD1 or G93A- SOD1 expression, remaining at moderate levels (++). Caspase-9 (figure 3.3d) showed the most interesting pattern, with low (+) levels of expression in Mn-1 cells and no expression in either MNWT or MNGA cells. Bcl-xl expression (figure 3.4b) was similar in Mn-1 and MNWT cells, and showed a slight increase (from + to ++) in MNGA cells. Trypan blue assay for cell death After identifying the apoptotic markers present in the MNWT and MNGA cells, an oxidative insult using H2O2 was applied. It was shown previously that this 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P-actin -► ♦ P-actin C. 123 bps M W G D. 123 bps m w G Cas3 - -Cas9 4 -P-actin P-actin -► Figure 3.3. RT-PCR for pro-apoptotic markers in Mn-1, MNWT, and MNGA cell lines. A. Bax. B. Caspase-1 (Casl). C. Caspase-3 (Cas 3). D. Caspase-9 (Cas 9). Abbreviations: 123 base-pair ladder (123 bps), beta-actin (P-actin). Mn-1 (M), MNWT (W), MNGA (G). 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. B. 123 bps m w G 123 bps M W G P-actin ♦ < • P-actin Figure 3.4. RT-PCR for anti-apoptotic markers in Mn-1, MNWT, and MNGA cell lines. A. Bcl-2. B. Bcl-xl. Abbreviations: 123 base-pair ladder (123 bps), beta actin (P-actin), Mn-l (M), MNWT (W), MNGA (G). 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. treatment induced an apoptotic phenotype in Mn-1 cells (see chapter 2). Mn-1, MNWT, and MNGA cells in both non-differentiated (figure 3.5) and differentiated (figure 3.6) states were treated with increasing doses of H2O2 , ranging from 0 to 500 M. After 24 hours viable cells were identified by trypan blue exclusion, and the percentage of viable cells in the population was determined. Figure 3.5 shows that for non-differentiated cells, Mn-1 cells are less susceptible to oxidative-stress induced cell death than either MNWT or MNGA cells, with statistical significance first demonstrated at 100 M (p < 0.001). There is no statistically significant difference in percentage cell death between MNWT and MNGA cells until 400 M, when MNGA cells appear to be more susceptible than MNWT cells (p < 0.001). For differentiated cells, the percentage of viable cells is lower at all concentrations of H2O2 , indicating that differentiated cells are more vulnerable to oxidative-stress induced cell death (figure 3.6). A change is also seen in relative levels between cell lines. Mn-1 and MNWT cells appear equally susceptible (no statistical difference at any treatment point), while MNGA cells are more vulnerable, with statistically significant differences seen at all levels between 100 M H202(p < 0.001) and 500 M. At 500 M, most of the cells in all three lines are dead and significance cannot be determined. 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 a > o Q J .o ro > 80 - 60 - 40 20 0 - I ----------------------------1 --------------------------- : ----------------------------1 ----------------------------1 ----------------------------1 --------------------------- 0 100 200 300 400 500 600 H2 0 2 concentration (^M) Mn-1 MNWT — A— MNGA Figure 3.5. Trypan blue assay for non-differentiated Mn-1, MNWT, and MNGA cell viability after treatment with H2O2 . Statistical analysis was performed using one-way ANOVA. MNWT and MNGA lines are significantly different from Mn-1 (p < 0.001) from 100 pM through 500 pM. MNWT and MNGA lines are only significantly different from each other above 400 pM (p < 0.001). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 -i---------------------------1 ---------------------------1 ---------------------------1 ---------------------------1 --------------------------- 1 --------------------------- 0 100 200 300 400 500 600 H2 0 2 concentration (pM) Mn-1 MNWT — A— MNGA Figure 3.6. Trypan blue assay for differentiated Mn-1, MNWT, and MNGA cell viability after treatment with H2O2 . Statistical analysis was performed using one way ANOVA. The MNGA cell line is statistically different from both the Mn-1 and MNWT lines from 100 pM (p <0.01) through 400 pM. The Mn-1 and MNWT cell lines do not deviate significantly from each other at any point. At 500 pM there is no difference between any of the cell lines. 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DNA ladder in MNWT and MNGA cells After showing that differentiated MNGA cells were more susceptible to oxidative-induced cell death than either MNWT or Mn-1 cells, the mechanism of cell death was further investigated. In Mn-1 cells, H2O2 produced a caspase-independent apoptotic phenotype. Differentiated MNWT and MNGA cells treated with increasing doses of H2O2 were examined for the formation of a DNA ladder, a hallmark of apoptotic cell death. Figure 3.7 shows that in MNWT cells, a DNA ladder is not apparent until 500 M, while in MNGA cells this pattern can be seen with doses as low as 200 M H2O2. Effect of hydrogen peroxide on expression of apoptotic markers To further characterize the pathway involved in H202-induced cell death in differentiated MNWT and MNGA cells bax, caspases-1,-3, and -9, bcl-2, and bcl-xl expression levels were studied by RT-PCR at 3 and 24 hours post-treatment. Each cell line was treated with 200 M H2O2 and samples were collected at specified times. In both MNWT and MNGA cells, levels of bax showed a decreasing trend from control (++) to 24 hours post treatment (+) (figure 3.8a). Caspase-L levels (figure 3.8b) are barely detectable prior to treatment (+ / -), and disappear entirely by 3 hours after treatment. Caspase-3 levels (figure 3.8c) are moderate (++) in controls and do not appear to change significantly in either MNWT or MNGA cells 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.7. DNA ladder formation in differentiated MNWT and MNGA cells after treatment with increasing levels of H2O2 . MNWT cells do not show a ladder pattern until 500 pM, while MNGA cells show a ladder beginning at 200 pM, indicating that differentiated MNGA cells are more susceptible to oxidative-stress induced apoptosis than differentiated MNWT cells. Abbreviations: 123 base pair ladder (123 bps), control (Co). Numbers above gels correspond to the dose of H2O2 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123 bps WCo W3h W 24H GCo G3h G24h Bax M l ■ * - P-actin WCo W3h W24h GCo G3h G24h 4- P-actin Figure 3.8. RT-PCR for pro-apoptotic markers in MNWT and MNGA cell lines at 3 hours and 24 hours after treatment with 200 pM H2O2. A. Bax. B. Caspase-1 (Casl). C. Caspase-3 (Cas 3). D. Caspase-9 (Cas 9). Abbreviations: beta actin (p- actin), MNWT (W), MNGA (G), control (Co), 3 hours (3h), 24 hours (24h). 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3.8. (continued) 123 bps WCo W3h W24h GCo G3h G24h ^-C as3 ■ * - P-actin D. 123 bps WCo W3h W 24H GCo G3h G 24H ♦Cas9 P-actin 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. throughout the time examined. Caspase-9 (figure 3.8d), which is not expressed in control cells, is also not expressed in either cell line after treatment. Bcl-2, expressed at minimal levels (+ / -) in control cells, does not appear to alter significantly during the course of treatment (figure 3.9a). Bcl-xl levels, similar to bax levels, show a decreasing trend in both MNWT (from + to + / -) and MNGA cells (from ++ to + / -) over the time period studied (figure3.9b). Expression of a human bcl-2 transgene from an adenoviral vector in Mn-1, MNWT, and MNGA cells In order to see if bcl-2, an anti-apoptotic protein, could interfere with the cell death pathway induced by H2O2, cells were infected with an adenoviral vector containing the human bcl-2 cDNA. Infection was performed as described in experimental procedures on differentiated Mn-1, MNWT, and MNGA cells. After 24 hours, expression of the protein was confirmed in all three cell lines by immunoblot using ECF for visualization (Figure 3.10). Trypan blue assay for effect of adenoviral mediated bcl-2 expression on H2O2- induced cell death MNGA cells were grown in 24 well plates in differentiation medium for 48 hours. After 48 hours cells were infected with an adenoviral vector containing either 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. WCo W3h W24h G24h WCo W3h W24h GCo G3h G24h ■4-P-actin Figure 3.9. RT-PCR for anti-apoptotic markers in MNWT and MNGA cell lines at 3 hours and 24 hours after treatment with 200 pM H2O2. A. Bcl-2. B. Bcl-xl. Abbreviations: beta actin (P-actin), MNWT (W), MNGA (G), control (Co), 3 hours (3h), 24 hours (24h). 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I MR-32 MCo WCo GCo M+B W+B G+B Figure 3.10. Immunoblot for human Bcl-2 in Mn-1, MNWT and MNGA cell lines 24 hours after infection. The human neuroblastoma line IMR-32 was used as a positive control. Uninfected samples are listed as control (Co). Infected samples are listed as M+B (Mn-1 + Bcl-2), W+B (MNWT + Bcl-2), and G+B (MNGA + Bcl- 2).The blot was developed using the ECF system (Amersham). 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. bcl-2 (Ad-bcl-2), or lacZ (Ad-lacZ) as a control. 24 hours post-infection cells were treated with 200 M H2O2 for another 24 hours, and then assayed for viability by trypan blue counts. Cells infected with Ad-lacZ showed no increased survival in comparison to control cells, while those infected with Ad-bcl-2 showed nearly a 30% increase in viable cells over both control and Ad-lacZ infected cells (p < 0.001) (Figure 3.11). RT-PCR for effect of adenoviral mediated bcl-2 expression on apoptotic markers during l^Ch-induced cell death MNGA cells were grown and treated as described above for trypan blue assays, and 3 hours post-treatment were collected for RNA extraction. RT-PCR was performed to look at levels of bax, caspases-1,-3, and -9, bcl-2, and bcl-xl in cells treated with or without previous infection by either Ad-lacZ or Ad-bcl-2. Figure 3.12a shows that levels of bax are decreased (from ++ to + / -) in l^C^-treated MNGA cells expressing Ad-bcl-2 in comparison to uninfected cells or cells infected with Ad-lacZ. Neither caspase-1 (figure 3.12b) nor caspase-9 (figure 3.12d) were detectable in the infected samples. Both caspase-3 (figure 3.12c) and bcl-2 (figure 3.13a) expression were detectable, but neither showed any significant changes. Bcl- xl expression, however, disappeared entirely in cells expressing human bcl-2, but was detectable in control and Ad-lacZ-infected samples (figure 3.13b). 102 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 Non-infected Ad-lacZ Ad-bcl-2 Control 200 pM 500 pM Figure 3.11. Trypan blue assay for viable cells after treatment with 200 or 500 pM H2 O2 . MNGA cells were non-infected, infected with Ad-lacZ, or infected with Ad- bcl-2. Statistical analysis was performed using one-way ANOVA. In the untreated group, there was no difference in cell viability. At 200 pM, Ad-bcl-2 infected cells showed a significant increase in viability in comparison to both non-infected and Ad-lacZ infected cells (p < 0.001). At 500 pM statistical significance cold not be demonstrated 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A. B. 123 bps CU CH L B 123 bps CU CH L B P-actin -► •4- P-actin C. D. 123 bps CU CH L B 123 bps CU CH L B 4 -P-actin Figure 3.12. RT-PCR for pro-apoptotic markers in non-infected, Ad-lacZ infected, or Ad-bcl-2 infected MNGA cells 3 hours after treatment with 200 pM H2O2 . A. Bax. B. Caspase-1 (Casl). C. Caspase-3 (Cas3). D. Caspase-9 (Cas9). Abbreviations: 123 base-pair ladder (123 bps); beta-actin (P-actin); control untreated (CU); control H2O2 treated (CH); Ad-lacZ infected, H2O2 treated (L); Ad-bcl-2 infected, H2O2 treated (B). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P-actin ♦ 1 k . i u t * -» f c W K 4 * . m m m m Figure 3.13. RT-PCR for anti-apoptotic markers in non-infected, Ad-lacZ infected, or Ad-bcl-2 infected MNGA cells 3 hours after treatment with 200 pM H2O2 . A. Bcl-2. B. Bcl-xl. Abbreviations: 123 base-pair ladder (123 bps); beta actin (p- actin); control untreated (CU); control H2O2 treated (CH); Ad-lacZ infected, H2O2 treated (L); Ad-bcl-2 infected, H2O2 treated (B). 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DISCUSSION The goal of this study was to create a tissue culture model of mutant SOD1- associated cell death in the Mn-1 motor neuron cell line that could be used to gain insight into pathogenic mechanisms in FALS. In addition, the efficacy of using an adenoviral vector containing the human bcl-2 cDNA to interfere with cell death in this model was evaluated. The Mn-1 line had previously been shown to express homeodomain genes consistent with an origin in precursor cells of the lateral column lateral (LCL) of the developing spinal cord (see chapter 2). These cells ultimately become the anterior horn motor neurons responsible for innervation of the limb muscles. In ALS the anterior horn motor neurons are affected, and an early sign of disease is often asymmetric paralysis of the limbs (Walling, 1999). Therefore it was determined that the Mn-1 line is an appropriate cell line in which to model motor neuron degeneration. Mutations in SOD1 have been found in nearly 20% of FALS cases, and one of these is a glycine to alanine mutation at position 93 (G93A). How the presence of these mutations results in disease is not known, although it appears that a gain of function, rather than a loss of enzyme activity, is responsible (Marx, 1996). There are two hypotheses for this gain of function, both of which involve excess reactive oxygen species (ROS) (Cleveland, 1996). ROS are molecules such as free radicals and nitric oxide (NO) that can interact and damage a variety of cellular components, and it is exactly this type of damage that has been found in both sporadic and 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. familial forms of ALS (Robberecht, 2000). The type of cell death resulting in ALS after such damage, however, is still unclear (He and Strong, 2000; Mu et al., 1996; Troost et al., 1995). Experiments with the Mn-1 line show that an oxidative insult results in a caspase-independent form of apoptosis. To test the effects of mutant SOD1 in the context of oxidative-stress induced apoptosis I developed a cell culture model of FALS. Mn-1 cells were transfected with either a wild-type (WT-SOD1) or a mutant human copper/zinc superoxide dismutase (G93A-SOD1) and studied for signs of apoptosis after treatment with H2O2. The first step in creating this model was to transiently transfect Mn-1 cells with a plasmid containing either WT-SOD1 or G93A-SODI. Mn-1 cells expressed mRNA from both plasmids for up to 5 days following transfection without a noticeable decrease in expression levels, at which time stably transfected subclones were produced. Two lines showing strong levels of expression by RT-PCR were chosen and named MNWT, which expressed the WT-SOD1, and MNGA, which expressed the G93A-SOD1 mutant. Expression of protein was confirmed by immunoblotting with ECF. RT-PCR was used to screen the MNWT and MNGA cell lines for expression of a panel of apoptotic markers used previously with the Mn-1 line (see chapter 2). These include the pro-apoptotic markers bax, caspase-1, caspase-3, caspase-8, and caspase-9, and the anti-apoptotic markers bcl-2 and bcl-xl. In comparison to the native Mn-1 line, levels of bax increased in both MNWT and MNGA cells, suggesting a possible predisposition to apoptotic insults in these lines as compared to 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the parent line. This is not likely to account for the increased susceptibility of motor neurons seen in FALS, however, since both WT-SOD1 and G93A-SOD1 result in an increase in bax. Caspase-1, along with bcl-2, was expressed at barely detectable levels in the Mn-1 line and remained so in both the MNWT and MNGA lines. This supports earlier findings in the Mn-1 line that neither caspase-l nor bcl-2 seem to play a large role in this model of motor neuron death. Caspase-3 levels, although higher than caspase-1, also did not change in the presence of either WT-SOD1 or G93A-SOD1. This suggests that, although more bax is expressed, either caspase-3 levels are not increased until apoptosis is initiated, or caspase-3 is not involved in the type of cell death to which these lines are predisposed. Caspase-9 showed the most striking pattern. Although Mn-1 cells expressed low levels of caspase-9 mRNA, transfection with either form of human SOD1 resulted in complete down-regulation of caspase-9. This suggests that the primary pathway to oxidative-stress induced apoptosis, in which caspase-9 acts to initiate the caspase cascade, is not present in either MNWT or MNGA cells. The possibility exists that, after treatment with an apoptotic inducer, caspase-9 expression is up- regulated, but as an initiator caspase, one would expect at least a low level of constitutive expression in order to begin the cascade leading to cell death. In combination with low levels of caspase-1 and unaltered levels of caspase-3, the lack of caspase-9 in MNWT and MNGA cells further supports the theory of a caspase- independent form of cell death present in motor neurons. 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bcl-xl expression was similar in Mn-1 and MNWT cells, and slightly increased in MNWT. This could indicate that cells expressing the G93A-SOD1 are stressed and are attempting to combat the increased stress by up-regulating bcl-xl. In the presence of the WT-SOD1 there is no need to increase anti-apoptotic gene expression because SOD1 is itself anti-apoptotic in oxidative stress situations (Facchinetti et al., 1998). In the presence of G93A-SOD1, however, there is a gain of function mutation producing increased levels of ROS along with decreased enzymatic function, and so the cell needs to up-regulate levels of bcl-xl in comparison to either Mn-1 or MNWT in order to survive under normal conditions. These results show that there is a difference in the apoptotic states of motor neurons expressing a WT-SOD1 versus a mutant G93A-SOD1. After the cell lines were characterized for apoptotic markers, their responses to oxidative-stress induced cell death were examined. This had been characterized previously in the Mn-1 cell line (see chapter 2), where the response was determined to be a caspase-independent form of apoptosis. Trypan blue counts were performed after treatment with H2O2 in order to determine if there were differences in percentage cell death among Mn-1, MNWT, and MNGA cells. In non-differentiated cells, expression of any form of SOD1 appears to increase Mn-1 susceptibility to cell death at treatment with levels as low as 100 M H2C > 2 (p < 0.001). On the other hand, there was no detectable difference in percentage cell death between MNWT and MNGA cells at levels below 400 M H2 0 2(p< 0.001). 109 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Differentiated MNGA cells, however, were significantly more susceptible to oxidative-induced cell death than either Mn-1 or MNWT cells (p < 0.001). These data show that in the context of a differentiated cell, the expression of G93A-SOD1 alters the cellular response to oxidative damage, while expression of WT-SOD1 has no effect. This is consistent with the increased susceptibility found in ALS in adult motor neurons expressing mutant SOD1, and in transgenic mouse models where expression of mutant SOD1, but not WT-SOD1, produces disease (Brujin et al., 1998; Gurney et al., 1994; Ripps et al., 1995; Wong et al., 1995). Based on this, the MNGA cell line appears to have reproduced the susceptibility of motor neurons found in SODl-associated FALS. The mechanism of cell death induced by H2O2 does have bearing on the value of this model of FALS, and it was thus necessary to determine whether the cells died by necrosis or a form of apoptosis. A DNA ladder was found in treated differentiated MNWT and MNGA cells, confirming that cell death in this case was by apoptosis. There was a striking difference, however, in the intensity of treatment necessary to produce the ladder. In MNWT cells, a DNA ladder was not apparent at levels below 500 M H2O2, while in MNGA cells a ladder was seen with doses as low as 200 M. This suggests not only that cell death induced by H2O2 is apoptotic, but also that the MNGA line may be more susceptible to this form of apoptosis than the MNWT line. This supports the current theory that motor neuron cell death in FALS could be a result of oxidative insults that result in apoptosis. The MNGA cell 110 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. line may be, therefore, a suitable model in which to study the disease mechanisms of FALS motor neuron death. The expression of apoptotic markers in MNWT and MNGA cells after treatment with HzC^also has bearing on the value of the cell line as a model of apoptotic motor neuron death. MNWT and MNGA cells are similar to Mn-1 cells in that caspases do not appear to play a significant role in the cell death process, even though the process is apoptotic. In fact, none of the caspases screened showed any difference in expression after H2O2 treatment. Only caspase-3 was expressed at greater than minimal levels in control cells. Although the amount of caspase-3 translation could be matched by transcription to produce a steady level of mRNA, the data so far do not support any role for caspases in this model of cell death. The conclusion must be that similar to the Mn-1 parent line, neither MNWT nor MNGA use a caspase-dependent pathway to initiate apoptosis in response to oxidative stress. Interestingly, caspase activation is difficult to demonstrate in both human and murine forms of ALS, although the significance of this observation is unclear (Li et al., 2000; Martin, 1999; Mighelli et al., 1999). The levels of bax and bcl-xl both showed a decreasing trend in MNWT and MNGA cells over the treatment period. This could indicate that the level of translation of each of these antagonistic proteins increases as the cell decides whether to live or die, without a concomitant increase in transcription. In the context of the findings outlined above, it was important to elucidate the effect of the anti-apoptotic bcl-2 protein on progression of apoptosis in MNGA cells. I l l Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bcl-2 has been shown to interfere in apoptotic pathways in a wide variety of cell types, including neurons, and has recently been shown to delay disease onset and prevent motor neuron loss in FALS mice (Assouz et al., 2000; Kostic et al., 1997). One study delivered an adeno-associated virus containing bcl-2 via direct injection into the spinal cord. Although the results showed an increase in anterior horn motor neurons relative to control mice and a delay in onset of symptoms in the area surrounding the injection, this is not a clinically acceptable means of delivering bcl-2 to motor neurons. Studies have shown that adenoviruses can be transported retrograde in motor neurons after intra-muscular injection and result in exogenous gene expression (Warita et al., 1998). After treatment of MNGA cells with either an Ad-bcl-2 construct or an Ad-lacZ construct as a control, trypan blue studies revealed that Ad-bcl-2 infected MNGA cells were significantly less susceptible to H2O2- induced cell death than either uninfected or Ad-lacZ infected controls. This supports the recent evidence that bcl-2 can interfere with mutant SOD-l-associated motor neuron degeneration and also shows that delivery via an adenoviral vector appears to not harm a motor neuron-like cell. Evaluation of expression of apoptotic markers in Ad-lacZ and Ad-bcl-2- infected MNGA cells after treatment with H2O2 revealed that the majority of the markers, including the caspases and endogenous bcl-2, were either not present or did not change in the presence of exogenous bcl-2. Two changes did occur, though. First, bax levels were decreased in Ad-bcl-2 treated cells compared to Ad-lacZ- infected and uninfected controls 3 hours after treatment. This indicates that anti- 112 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. apoptotic bcl-2 may be acting via down-regulation of bax, and supports the idea that the cell death mechanism, despite the lack of caspase involvement, is apoptotic. As previously mentioned, however, bcl-2 may have an anti-oxidant function, and in the presence of increased ROS due to mutant SOD 1 the anti-oxidant mechanism may be involved in protecting the cell by down-regulating a pro-apoptotic factor. Clearly this area needs more study. Second, in the presence of exogenous bcl-2, expression of endogenous bcl-xl was undetectable. Since bcl-2 and bcl-xl share many similar functions, it is possible that only one can be expressed at high levels at a time. Bcl-xl is normally found in adult motor neurons, while bcl-2 is absent. This same pattern was found in Mn-1, MNWT, and MNGA cells. Complete down-regulation of bcl-xl after infection with Ad-bcl-2 in MNGA cells indicates that this could also occur in vivo in motor neurons. As long as bcl-2 is over-expressed, down-regulation of bcl-xl may have no effect. However, if expression of bcl-2 diminishes before bcl-xl can be up-regulated, then the motor neurons might be left without an endogenous anti-apoptotic defense mechanism. Several recent studies have shown that although treatment with bcl-2 can delay apoptosis, in many cases the cell will eventually die by either apoptosis or necrosis. It is possible that the loss of endogenous anti-apoptotic factors due to high expression of exogenous bcl-2 may contribute to this. Each cell has its normal complement of pro- and anti-apoptotic factors, and even though bcl-2 may be able to replace some of the functions of bcl-xl, it may not be able to compensate entirely for 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a complete absence of bcl-xl. This is something to keep in mind when using bcl-2 for gene therapy to alter the course of cell death in neuronal systems. In conclusion, in this study I developed and characterized the cell death pathways of a tissue culture model of G93A-SODl-associated FALS, the MNGA cell line. This model appears to respond to oxidative stress with a caspase- independent form of apoptosis, which supports the proposed role of apoptosis as the cell death mechanism in ALS. The MNGA cell line will be useful in future studies to elucidate the exact pathways involved motor neuron degeneration. In addition, I tested the efficacy of adenoviral-mediated expression of bcl-2 to interfere with H2O2- induced cell death in MNGA cells. Like many other trials using bcl-2, cells expressing exogenous bcl-2 were significantly less susceptible to cell death than control cells. Based upon this and other recent studies showing that bcl-2 can interfere with motor neuron death in FALS mice, it is possible to envision using an adenoviral vector to deliver bcl-2 to the spinal cord of FALS mice via an intra muscular injection. Although a promising strategy, and one worth pursuing, it is necessary to keep in mind that there may be alterations in endogenous expression of anti-apoptotic factors, and that this may open the door to cellular damage that bcl-2 cannot prevent. 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EXPERIMENTAL PROCEDURES Mn-L cell culture and differentiation Mn-1 cells were a gift from Kenneth Fischbeck. Cells were grown in Dulbecco’s modification of Eagle’s medium (DMEM) (Gibco) containing 10% fetal calf serum (FCS) (growth medium). Cells were split by washing once with phosphate buffered saline (PBS) without calcium and magnesium (Irvine Scientific) and then incubated with 0.05% trypsin-EDTA for 3 minutes at 37°C. Cells were split 1:3 once per week. For use in experiments cells were allowed to grow for 48 hours before any treatments were administered. To produce the differentiated phenotype cells were plated in DMEM containing 2% FCS (differentiation medium) for 72 hours before being used in experiments. Differentiation was confirmed by extension of processes and cessation of cell division. Treatments were performed in the same medium (high or low serum) in which the cells were grown. For MNWT and MNGA cells, growth or differentiation medium was supplemented with 250 g/ml G418 sulfate (Invitrogen). Transient and stable transfection of Mn-1 cells Mn-1 cells were transfected with a pcDNA 3.1+ vector (Invitrogen) containing either a wild-type SOD1 (pWT-SODl) or a G93A SOD1 cDNA (pG93A- 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SOD1). The cDNAs were provided by Teepu Siddique in pBluescript plasmids. Each cDNA was removed by restriction digest at the EcoRl site in the pBluescript multicloning site (MCS) and inserted into the corresponding site in pcDNA 3.1+. Mini- and midi-preps of plasmid DNA were grown in subcloning efficiency DH5 competent cells (Gibco). The plasmid containing the G93A-SOD1 cDNA was grown at 33°C to prevent toxicity to the bacteria. Plasmid DNA was extracted using the Wizard plus Mini or Midi Prep kits (Promega). The inserts were sequenced to confirm their orientation and the presence of the G93A mutation. Transfections were performed using the Lipofectamine Plus kit (GIBCO) according to the manufacturer’s directions. Transient transfections were assayed for expression after 24 hours. For stable transfections, the same procedure was followed, but on the day after the transfection the medium was changed to DMEM with 10% FCS and 500g/ml G418 sulfate. The selection medium was changed every 3 days until colonies of resistant cells were evident. Colonies were picked and subcloned, and subclones were screened for expression of human SOD1 (hSODl) by RT-PCR. Two subclones were chosen and given the names MNWT, for the cell line expressing the WT-SOD1, and MNGA, for the cell line expressing the G93A-SOD1 mutant. 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. producing absent bands were tested against other tissue, including mouse spinal cord, mouse cerebral cortex, and a mouse neuroblastoma cell line, N2a, to insure that the primers were functional. Results are the estimated average of three independent trials. Immunoblot procedure using enhanced chemifluorescence Cells were plated at a density of 1 x 106 cells per plate in 100-mm plates. For analysis of hSODl and hbcl-2 expression, cell monolayers were washed twice with PBS with calcium and magnesium and then lysed with a triple detergent R1PA buffer (50 mM Tris-chloride, pH7.4,150 mM NaCl, 5 mM EDTA, 0.5% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate) on ice for 30 minutes. The lysate was then centrifuged at maximum speed in a microcentrifuge at 4°C for 15 minutes. The cleared supernatant was transferred to a clean tube and total protein was assayed for according to the method of Lowry (Bensadoun and Weinstein, 1976). For hSODl 400g of total cellular protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred overnight at 4°C to a polyvinyl difluoride membrane (PVDF) (Pall). The same procedure was used for hbcl-2 except that 200 g of total cellular protein were used. Immunoblots were developed using either a mouse anti-human SOD1 antibody (NovoCastra) (NCL-SOD1) at a dilution of 1:500 or a mouse anti-human bcl-2 antibody (sc-509) (SantaCruz) at a dilution of 1:200. The secondary antibody was 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. horseradish peroxidase derivitized rabbit anti-mouse IgG, provided with the Enhanced Chemifluorescence (ECF) (Amersham) immunoblot developing kit. The protocol for ECF development was followed from this point, including incubation with the anti-fluorescein antibody conjugated to alkaline phosphatase. After incubation with the ECF substrate blots were dried and scanned on a Molecular Dynamics Storm 860 scanner and analyzed using Molecular Dynamics ImageQuant software. Induction of apoptosis To induce apoptosis in cells for trypan blue and DNA ladder assays cells were treated with hydrogen peroxide (H2O2) (Sigma) diluted in water and added to fresh medium (growth or differentiation) in concentrations ranging from 0 to 1000 M for Mn-1 cells and 0 to 500 M for MNWT and MNGA cells. Samples were incubated for 24 hours at 37°C. For RT-PCR analysis of apoptotic markers and trypan blue analysis of exogenous bcl-2 effects on cell death, cells were treated with 200 M H2O2 for the indicated times. Trypan blue assay for cell viability Cells were plated at a concentration of 1 x 105 cells per well in 24 well plates and treated as described for each experiment. 24 hours after H 2O2 treatment the 119 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. medium was aspirated and placed in 15 ml tubes on ice, cells were washed once with PBS without calcium and magnesium, and trypsinized. The cells were then transferred to the 15 ml tube containing the appropriate supernate, and the tubes were centrifuged for 5 minutes at 1200 x g at 4°C. The supernate was aspirated and cells were resuspended in 2001 PBS. An equal volume of 4% trypan blue (Sigma) was mixed with the resuspended cells and cells were counted on a hemacytometer. Five squares were counted for each sample in duplicate, and experiments were repeated 8 times. Data was analyzed by one-way ANOVA followed by appropriate post hoc tests. Assessment of DNA ladder Extraction of DNA for assessment of DNA ladder formation was performed essentially as described with minor modifications (Toku et al., 1998). 3 x 106 cells were plated in a 100-mm dish in appropriate medium. Collection was the same as for the trypan blue assay up until the first centrifugation step. The cells were then washed twice with PBS and the cell pellets were lysed for 10 minutes in hypotonic lysis buffer (lOmM Tris-chloride, pH 7.4, lOmM EDTA, 0.5% Triton X-100) on ice. Lysates were centrifuged at 14,000 x g for 25 minutes at 4°C to pellet intact chromatin. The supernatant was transferred to a clean tube and incubated with 35 g of RNAse A (Boehringer-Mannheim) for 1 hour at 37°C, followed by 40 g of proteinase K (Boehringer-Mannheim) for 1 hour at 37°C. After incubation the lysate 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. was mixed with an equal volume of 1:1 phenol chloroform, vortexed, and centrifuged for 5 minutes at 14,000 x g at room temperature. The aqueous portion was transferred to a clean tube and incubated with an equal volume of isopropanol, 50 M sodium chloride, and 5 nanograms (ng) per microliter of glycogen (Boehringer-Mannheim) at -20°C for 18 hours. After incubation samples were centrifuged at 14,000 x g for 20 minutes, air dried, and resuspended in TE buffer (lOmM Tris, ImM EDTA, pH 7.4). Resuspended DNA was visualized the same as for RT-PCR. Infection of cells with adenoviral vectors Cells were infected with an adenoviral vector containing either lacZ (Ad- lacZ) or human bcl-2 cDNA (Ad-bcl-2). The Ad-lacZ construct was provided by Mel Trousdale, and the Ad-bcl-2 construct by Jeannie Chen. Both viruses were grown and titered in the Trousdale lab. For infection, the medium was removed from cells and replaced with 1/10 of the normal amount. Virus was added to this at a multiplicity of infection (MOI) of 25, and the cells were incubated for two hours at 37°C with rocking. After 2 hours, enough medium was added to bring the amount up to the normal volume. For experiments examining the effects of bcl-2 on H2O2- induced cell death, cells were differentiated for 48 hours, and then infected with the appropriate vector. 24 hours post-infection cells were treated with H2O2 according to 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the standard protocol outlined above, and results were assayed 3 or 24 hours after treatment. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C H A PT E R 4 AN ADENOVIRAL-BCL-2 VECTOR DELIVERED VIA INTRA-MUSCULAR INJECTION RESULTS IN DELAYED ONSET OF DISEASE IN A MOUSE MODEL OF FAMILIAL AMYOTROPHIC LATERAL SCLEROSIS INTRODUCTION Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder in which there is selective loss of motor neurons in the anterior horn of the spinal cord and neurons of the motor cortex that form the lateral corticospinal tracts (Ince et al., 1998). ALS has a prevalence of about 5 cases per 100,000, with a median age of onset of 55 and a median survival of three years. Death is due to complications of paralysis including compromised respiratory function (Cudkowicz et al., 1997). Clinically there is a mixture of upper and lower motor neuron symptoms. Lower motor neuron disease produces weakness and atrophy of limb muscles, fasciculations, and cramps. Upper motor neuron disease results in weakness, spasticity, and hyperreflexia. Onset of disease is usually asymmetric and localized, then progressive in a contiguous manner (de Belleroche et al., 1995). The majority of cases are sporadic, but nearly ten percent are inherited in an autosomal dominant fashion and are termed familial ALS (FALS). FALS is clinically indistinguishable from sporadic ALS (SALS) (Deng et al., 1993). 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Approximately twenty percent of FALS cases have a mutation in the cytoplasmic copper/zinc superoxide dismutase gene (SOD1) on chromosome 21q (Rosen et al., 1993). The connection between SOD1 mutations and motor neuron death has been studied in tissue culture and in transgenic mouse models, but the link remains elusive. One question that has been studied in depth is the nature of cell death in FALS, whether it is necrotic, characterized by a toxic insult resulting in cell lysis, or apoptotic, proceeding by an aberrant programmed death mechanism. The answer to this question may have important therapeutic implications. Currently there is no cure for ALS. Treatment is limited to palliative care and the glutamate release inhibitor riluzole. Riluzole was one of a number of anti- glutamatergic agents tested after studies implicated glutamate excitotoxicity as a factor in disease pathogenesis (Lin et al., 1998; Ludolph et al., 2000). Although there is slight prolongation of life in ALS patients on riluzole, the effects are minimal, and so the search for an effective treatment continues. One possibility is gene therapy to correct or bypass the disease process at the cellular level. A promising candidate for gene therapy to prevent cell death is the anti-apoptotic protein bcl-2. The ability of increased levels of bcl-2 to prevent apoptosis has been demonstrated in a variety of systems, including neurons (Saikumar et al., 1999). The bcl-2 family is composed of pro- and anti-apoptotic members that contain several homologous regions involved in formation of homo- and heterodimers. Dimerization between pro- and anti-apoptotic proteins is thought to regulate cell 124 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. death decisions (Korsmeyer, 1995; Yin et al., 1994). This underscores the importance of the relative ratios between these antagonistic proteins, and increased levels on the anti-apoptotic side would be expected to be beneficial in preventing apoptosis. In addition, bcl-2 is thought to have an anti-oxidant effect separate from its role in apoptosis (Kane et al., 1993; Tyurina et al., 1997). If cell death in FALS is apoptotic then the combination of anti-apoptotic and anti-oxidant properties of bcl-2 makes this a promising candidate for gene therapy. To address the question of apoptotic versus necrotic cell death, several cell lines, including two neuroblastoma lines (SH-SY5Y and N2a), a pheochromacytoma line (PC12) and yeast cells, have been transfected with a mutant SOD1 containing one of the mutations found in FALS (Carri et al., 1997; Ghadge et al., 1997; Pasinelli et al., L998; Rabizadeh et al., 1995). Although much of the evidence supports apoptosis as the death mechanism in these models, overall the results have been controversial. One reason for this could be the fact that none of the lines used have been motor neurons. In chapter 3 I described the development of a tissue culture motor model of FALS that was produced by stably transfecting Mn-1 motor neurons with a mutant SOD1 that has a glycine to alanine mutation at position 93 (MNGA cells). In this model, oxidative stress in the form of H2O2 produces caspase- independent apoptotic cell death. This supports the theory that motor neuron death in FALS, which has been linked to oxidative stress and increased levels of reactive oxygen species (ROS), is apoptotic. 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In addition to tissue culture models, a number of transgenic mouse models have been designed that model motor neuron degeneration, including SOD1 mutations (glycine to arginine at positions 37, 85, and 86, and glycine to alanine at position 93) and over-expression of the neurofilament light chain (Brujin et al., 1998; Cleveland et al., 1996; Gurney et al., 1996; Ripps et al., 1995; Wong et al., 1995). Cytopathological studies comparing motor neuron degeneration in FALS mice with a G93A mutation (G93A mice) to human FALS spinal cord sections show a number of similarities, including anterior horn atrophy, gliosis, axonal swelling, neurofilament aggregation, and Lewy-type bodies (intra-cytoplasmic inclusions) (Dal Canto and Gurney, 1994; Tu et al., 1996). Attempts to define the mechanism of motor neuron death in the G93A mouse model have been inconclusive, paralleling similar attempts in tissue culture. One group reported that G93A mice lack typical signs of apoptosis, including evidence of DNA strand breaks and activation of the apoptosis-associated protease caspase-3, while others found evidence of activation of both caspase-3 and a related protease caspase-1 in G93A mouse motor neurons (Li et al., 2000; Mighelli et al., 1999). Despite this conflicting data there is no evidence that excludes apoptosis as the means of cell death in ALS. Possibly some form of apoptosis is involved in ALS that does not have all the morphological and biochemical features normally associated with this type of cell death. Despite uncertainty surrounding the exact pathway of cell death in ALS, attempts have been made to treat dying motor neurons with increased levels of bcl-2. In chapter 3 it was found that bcl-2 could reduce 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oxidative-stress induced apoptosis in MNGA cells. Two other recent studies have looked at the effect of bcl-2 in the G93A mouse. Both showed that bcl-2 could have a modulatory effect on motor neuron degeneration. The first study crossed transgenic mice over-expressing bcl-2 with G93A FALS mice, and showed that expression of bcl-2 could delay onset of disease and increase the number of anterior horn motor neurons during disease progression. Bcl-2 was unable, however, to alter the length of the disease, and by end-stage both bcl-2-expressing and control mice showed equivalent losses of motor neurons (Kostic et al., 1997). Although this study provides evidence that bcl-2 can affect motor neuron degeneration in a model of FALS, the therapeutic utility of such an approach is limited. The second study used an adeno-associated viral vector to express bcl-2 after direct injection of the virus into the lumbar spinal cord. They observed an increased number of motor neurons and a delay in the onset of symptoms in areas adjacent to the site of injection (Assouz et al., 2000). Again, although the study demonstrated a beneficial effect of bcl-2, the clinical utility of intra-spinal injections is limited. Instead, a delivery system needs to be developed that does not involve such invasive procedures. One recent study showed that intra-muscular (i.m.) injection of an adenoviral vector containing the gene for lacZ resulted in retrograde axonal transport in motor neurons and expression of the lacZ protein (Warita et al., 1998). This is approach provides an alternative to highly invasive procedures, and the use of an adenovirus instead of an adeno-associated virus will result in a transient expression of the gene product delivered. Although this is often cited as a disadvantage of adenoviral 127 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vectors, it may be that transient expression of bcl-2 is better than uncontrolled constitutive expression. If expression is lost, treatment can be repeated, but it is more difficult to turn off expression once begun. In muscle cells, which are infected at the site of injection, uncontrolled bcl-2 expression could result in transformation and tumor growth, while in post-mitotic neurons, the altered pro- to anti-apoptotic protein ratios could result in increased cell death. In this study I injected G93A FALS mice with an adenoviral vector containing cDNA for human bcl-2 (Ad-bcl-2) intra-muscularly, with the object of introducing the bcl-2 cDNA into the motor neurons innervating that muscle. Mice were treated 2 weeks before estimated onset of symptoms with Ad-bcl-2, Ad-lacZ, or saline injections into fore-and hind limbs bilaterally. Onset and progression of disease was evaluated in each group. In addition, control mice from the background strain, B6SJL, and from a strain expressing the wild-type human SOD I (WT-SOD1), were treated with both vectors to determine the effect of increased bcl-2 or lacZ in a healthy animal. 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. RESULTS Disease progression in untreated G93A mice The clinical progression of disease in G93 A mice is well documented, as is the lack of disease in mice over-expressing the wild-type SOD1 (Andreassen et al., 2000; Assouz et al., 1997; Gurney et al., 1994). In this experiment G93A-SOD1, WT-SODl, and background B6SJL mice were followed for clinical disease progression including onset of symptoms and weight loss. Figure 4.1 shows average weekly weights for mice from each strain up to 18 weeks of age, at which point all untreated G93A mice had reached endpoint. The graph shows that while both B6SJL and WTSOD1 mice had a steady increase in weight during this time, G93A mice showed a slight increase in weight until 15 weeks, then a decreasing trend up through week 18. By week 17 the difference in weights between G93A mice and both control lines had reached significance (p < 0.01) and remained significant through end-stage. Figure 4.2 shows representative photomicrographs of lumbar spinal cord anterior horns from each strain of mice at approximately 18 weeks of age. Figures 4.2a and 4.2b show that B6SJL and WTSOD1 mice had similar numbers of healthy looking motor neurons, while G93A mice (figure 4.2c) had almost none. This correlates with disease progression as indicated by clinical symptoms and weight loss. 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26 t 25- 24- 23- C O E 22 (0 o > 21 - 05 Q ) 20- £ 19- 18- 17- 16- 11 12 13 14 15 16 Age (weeks) 17 18 19 B B C wrc — 4 — QfiC Figure 4.1. Average weekly weights for mice from pre-disease onset (12 weeks) to end-stage (18 weeks). B6SJL control mice (B6C) and WT-SOD1 control mice (WTC) show a consistent increase in weight over the period of observation. G93 A- SOD1 mice (GAC) show a slight increase up to 15 weeks, followed by a steady decrease over the next 3 weeks. Differences in average weekly weight between G93 A mice and control mice become statistically significant at 16 weeks and remain significant through end-stage (p < 0.05). Data are expressed as means +/-S.E.M. 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.2. Representative photomicrographs of lumbar spinal cord anterior horns at the experimental end-stage. Sections were stained for Nissl. A. B6SJL. B. WT- SOD1.C. G93A-SOD1. 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. / • ' u . • W : ' W , ' • ' • . * * £ % . • v l i i * I* • / * ' / % - -V -? - T , ^ r * ' * :!lUr *.* * • * / ; • • ^ , * *v 'V . \ 4 % i • • • v ;. ^ ^ - • • " ' . , • * i * . , • ' ■ , . ' . i ' i ’. : - . ' ’ . • ' *.v', * • .st 4* H . T"A . ^ ' v -4 « ,-* • T^&EkR > 4 ** h :« l * sr5 . V l f e 132 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.2. (continued) C. .. > * * fc * * L * • ' ' - - - ' • * « * V . ft ' • * b * ' r* » A * % ♦ • . * r v , ' . V f s * * ^ •? « . • -; r . 4 t » . J ' . J , ' ■ ‘ \» : • 3 ? , ♦<. , . v * . i/* ' i ? \*> *’ ,V^-< 'f ’ ^ r ^ \ ' / • «t * . «*'*? - 3 * - v . ^ : ,1 ‘ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission All mice were sacrificed at G93A disease endpoint (see experimental procedures), and 10 sections spaced 240 m apart from individual lumbar cords were stained for Nissl using cresyl violate acetate. Motor neurons in each anterior horn were counted as described in experimental procedures. Numbers given are total anterior horn motor neurons from both anterior horns in 10 sections. Figure 4.3 shows the average number of neurons in each species at end-stage. These numbers correlate with the photomicrographs seen in figure 2, with no difference between B6SJL (B6CL, n=7) and WTSOD1 (WTCL, n=7) mice, but a drastic and significant decrease in G93A mice (GACL, n=5) (p = 0.004). Figure 4.4 shows progression of disease in G93A mice, beginning with mice sacrificed at time 0 (GACE, n=5), at disease onset (GACO, n=5), at disease midpoint (GACM, n=3), and at disease endpoint (GACL, n=5). Overall there was a significant decrease in motor neuron number from control to end-stage (p= 0.017), although no significance was detected between adjacent time points. Immunohistochemistry for expression of hbcl-2 Mice from each strain were infected via intra-muscular (i.m.) injection with an adenovirus containing either cDNA for human bcl-2 (Ad-bcl-2) or a lacZ gene (Ad-lacZ). Control mice from each species were given injections of sterile normal saline (NS). Immunohistochemistry for human bcl-2 showed positive anterior horn motor neurons in all 3 species, indicating that retrograde axonal transport of virus 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 - c n B6C WTC GAC Figure 4.3. Total number of motor neurons in 10 lumbar cord sections from B6SJL (B6C), WT-SOD1 (WTC), and G93A-SOD1 (GAC) mice at end-stage. GAC mice have significantly fewer neurons at end stage than either B6C or WTC, correlating with the advanced stage of paralysis found at end-stage (*p= 0.004). Data are expressed as means +/-S.E.M. 135 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. < 20 G A C E G A C O G A C M G A C L Figure 4.4. Total number of motor neurons in 10 lumbar cord sections from G93A- SOD1 mice pre-disease onset (GACE), at disease onset (GACO), at disease midpoint (GACM), and at end-stage (GACL). As disease progresses there is a steady decrease in the number of motor neurons, with an overall significant difference between GACE and GACL (*p= 0.017). No significance was demonstrated between adjacent time points. Data are expressed as means +/-S.E.M. 136 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and expression of gene product was successful. Figure 4.5 shows a representative photomicrograph of a bcl-2 positive motor neuron in a G93A mouse sacrificed at disease onset. The number of positive neurons in mice at disease onset was estimated at around 1% of total neurons observed. Bcl-2 positive neurons were also observed at disease midpoint and disease endpoint in G93A mice, although these were rare. Mice were injected at 10 weeks of age, indicating that expression of bcl-2 in some cases lasted for 8 weeks. In general, however, it is assumed that expression of bcl-2 was transient, and the effects exerted would be early in the pathological progression. Safety of Ad-lacZ and Ad-bcl2 in control mice Over-expression of exogenous proteins in the nervous system could theoretically result in unintended pathology. Therefore it was necessary to treat mice from all three species with both vectors to determine if either vector would harm non-motor neuron disease mice. Figures 4.6 and 4.7 show graphs of the average weekly weights of B6SJ1 (B6C, n=5) (4.6) and WTSOD1 (WTC, n=5) (4.7) mice in each treatment group. Mice in all categories showed an increase in weight with no difference between treated and untreated controls, indicating that neither of the vectors appeared to harm control animals. To confirm that motor neurons were not inadvertently damaged by treatment, sections from mice each in group at the end stage of the experiment were stained for Nissl as described above and compared to 137 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.5. Representative photomicrograph of immunohistochemical staining for human bcl-2 in mouse lumbar motor neurons. A. Positive staining for bcl-2 in a G93A mouse at disease onset (arrow). B. Negative neurons in B6SJL non-infected control mouse. Sections were counterstained with Mayer’s hematoxylin. 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 27- 26- 25- ( /) E 24- c c O ) 23- •C O J < u 22- £ 2 1 - 20- 19- 18- 1 1 — I -- 12 — i— 13 14 15 1 6 /Age (weeks) 17 1 8 19 Figure 4.6. Average weekly weights for B6SJL mice injected with saline (B6C), Ad-lacZ (B6L), or Ad-bcI-2 (B6B). Mice in all 3 treatment groups show a consistent increase in weight over the period of observation. There is no statistical difference between treatment groups at any time point. Data are expressed as means +/-S.E.M. 139 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 27- rf wrc wn_ wtb 2 0 - 19- 1 8 1 1 12 13 1 4 15 1 6 17 18 1 9 Age (weeks) Figure 4.7. Average weekly weights for WT-SOD1 mice injected with saline (WTC), Ad-lacZ (WTL), or Ad-bcl-2 (WTB). Mice in all 3 treatment groups show a consistent increase in weight over the period of observation. There is no statistical difference between treatment groups at any time point. Data are expressed as means +/-S.E.M. 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pretreatment controls. Numbers of motor neurons did not differ significantly between any of the treatment groups in either B6SJL (B6C, n=5) (figure 4.8) or WTSOD1 (WTC, n=5) (figure 4.9), confirming that treatment with an adenoviral vector via i.m. injection did not harm normal mice. In addition, no inflammation was evident nor did mice develop tumors by gross inspection at the injection sites. Assessment of disease onset and progression in bcl-2 treated mice To test the ability of adenovirally-delivered bcl-2 to affect motor neuron disease in the G93A FALS model, mice were injected with Ad-lacZ, Ad-bcl-2, or NS for control and observed for signs of disease onset as indicated by hind limb weakness and tremor (Gurney et al., 1994). Figure 4.10 shows the average age of onset (in days) of each treatment group (GAC, n=12; GAL, n=9; GAB, n=9). Mice treated with Ad-bcl-2 had a delay in disease onset of 8.5 days (p=0.01) over control and Ad-lacZ treated mice. A subset of mice from each treatment group (GAC, n=4; GAL, n=3; GAB, n=7) were allowed to survive to end-stage, and the length of disease course was determined. These results are shown in figure 4.11. No difference was found between any of the treatment groups, indicating that although disease onset is delayed, once initiated the disease proceeds to end-stage in the same amount of time irrespective of the treatment given. Since there was no apparent difference in disease length by clinical symptoms, differences in the general health of the animals in each group were 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 - to B6C B6L B6B Figure 4.8. Total number of motor neurons in 10 lumbar cord sections from B6SJL mice injected with saline (B6C), Ad-lacZ (B6L), or Ad-bcl-2 (B6B). There is no difference between treatment groups, indicating that neither of the viral vectors resulted in loss of motor neurons in WT-SOD1 mice. Data are expressed as means +/-S.E.M. 142 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 - W TC W TL WTB Figure 4.9. Total number of motor neurons in 10 lumbar cord sections from WT- SOD1 mice injected with saline (WTC), Ad-lacZ (WTL), or Ad-bcl-2 (WTB). There is no difference between treatment groups, indicating that neither of the viral vectors resulted in loss of motor neurons in WT-SOD1 mice. Data are expressed as means +/-S.E.M. 143 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110 GAC GAL GAB Figure 4.10. Age of disease onset in G93A-SOD1 mice injected with saline (GAC), Ad-lacZ (GAL), or Ad-bcl-2 (GAB). GAB mice have a delayed onset of 8.5 days compared to untreated and Ad-lacZ treated controls (p= 0.01). Data are expressed as means +/-S.E.M. 144 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0 ) 30 a > 25 C/3 “ GAC GAL GAB Figure 4.11. Length of disease course in G93A-SOD1 mice injected with saline (GAC), Ad-lacZ (GAL), or Ad-bcl-2 (GAB). There is no statistical difference between treatment groups, indicating that bcl-2 does not prolong survival in FALS mice. Data are expressed as means +/-S.E.M. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. approximated by examining average weekly weights. Although no statistical difference was found between the groups up to week 18, 2 mice from the bcl-2 treated group were still alive at this time point (figure 4.12). The data on weight loss after week 18 cannot be interpreted because the mice in both control groups had already reached end-stage. Determination of motor neuron number in treated and control mice Since there was no difference seen in progression of disease by clinical symptoms or by weight loss in mice treated with Ad-bcl-2, the number of motor neurons observed by Nissl staining in each treatment group was quantified at onset and end-stage. Regardless of disease stage (onset or end-stage), there was no difference in numbers of motor neurons between treatment groups (figure 4.13 and 4.14) (GACE, n=5; GACO, n=5; GALO, n=4; GABO, n=5; GACL, n=5; GALL, n=4; GABL, n=5). 146 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 6 24 C / 3 E 2 2 2 O ) o > ^ 20 1 8 1 6 1 1 12 1 3 1 4 1 5 Age (weeks) 1 6 1 7 1 8 1 9 Figure 4.12. Average weekly weights for G93A-SOD1 mice injected with saline (GAC), Ad-lacZ (GAL), or Ad-bcl-2 (GAB). Mice in all 3 treatment groups show a decline beginning around 15 weeks of age. There is no statistical difference between treatment groups at any time point. Data are expressed as means +/-S.E.M. 147 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60 - 50 - 1 / 1 GACO GALO GABO Figure 4.13. Total number of motor neurons in 10 lumbar cord sections from G93A-SODlmice injected with saline (GACO), Ad-lacZ (GALO), or Ad-bcl-2 (GABO). Sections were collected at disease onset. There is no difference between treatment groups, indicating that bcl-2 does not delay disease onset by preventing loss of motor neurons. Data are expressed as means +/-S.E.M. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 - GALL GABL Figure 4.14. Total number of motor neurons in 10 lumbar cord sections from G93A-SODlmice injected with saline (GACL), Ad-lacZ (GALL), or Ad-bcl-2 (GABL). Sections were collected at disease end-stage. There is no difference between treatment groups, indicating that bcl-2 does not have an effect on motor neuron survival in this model. Data are expressed as means +/-S-E.M. 149 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DISCUSSION The goal of this study was to determine if an adenoviral vector containing human bcl-2 cDNA could be delivered to motor neurons via i.m. injection and produce an alteration in disease course in G93A FALS mice. Bcl-2 has been shown to interfere with cell death in a variety of systems, including neurons, and has recently been shown to delay disease onset and inhibit motor neuron death in the G93A FALS mouse (Assouz et al., 2000; Kostic et al., 1997). One study crossed bcl-2 mice with FALS mice to increase endogenous bcl-2 levels, while another injected an adeno-associated viral vector containing bcl-2 directly in to the lumbar spinal cord. Both studies were successful in showing alteration of motor neuron disease by increased bcl-2 expression, but neither method of delivery is clinically useful. One recent study showed that i.m injection of an adenoviral vector containing the lacZ gene resulted in expression of lacZ in motor neurons of G93A mice (Warita et al., 1998). The non-invasive nature of this method makes it ideal for delivery of bcl-2 to motor neurons in FALS mice. Disease progression in G93A mice has been well characterized (Andreassen et al., 2000; Assouz et al., 1997; Gurney et al., 1994; Vaux and Strasser, 1996). The first signs of disease occur around 90 days of age and include hind limb weakness and tremor. Over the next 4 to 6 weeks mice show increased tremor, gait abnormalities, and finally progressive paralysis of hind and fore limbs. G93A mice in the current study followed the normal pattern of disease, with an average age of 150 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. onset of 92.4 days, and an average length of disease of 31.2 days. Average weekly weight did not begin decreasing until after 15 weeks of age, which approximates the clinical midpoint of disease when gait abnormalities are first seen. After paralysis begins, progression is rapid and weight loss becomes more marked until end-stage. End-stage G93A mice had significantly fewer motor neurons in lumbar spinal cord than age-matched controls, correlating with the severe clinical disease seen at this point. Looking at lumbar spinal cord motor neurons in G93A mice before onset, at onset, at disease midpoint, and at end-stage, one sees a statistically significant decrease overall. Between each individual time point, however, no significance was demonstrated so an estimate of relative time course of motor neuron loss cannot be made. Mice from all 3 strains were given i.m. injections of Ad-lacZ, Ad-bcl-2, or NS into proximal limb muscles. At various time points mice were sacrificed and spinal cord sections stained for expression of human bcl-2. Bcl-2 staining was found in motor neurons in all three species at end-stage (18 weeks), although these neurons were rare. One of the characteristics of adenoviral-mediated gene therapy is that expression is transient (Ghadge et al., 1995). This could explain the paucity of positive neurons in end-point mice. In G93A mice at onset of disease (around 12-13 weeks of age) approximately 1% of neurons were positive for human bcl-2. Although this is not a large percentage, it correlates with data found in studies that used i.m. injection of adenovirus to deliver lacZ to motor neurons, indicating that the 151 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. delivery method is effective. In future studies it may be necessary to perform multiple injections in each limb to increase the number of motor groups infected. One concern of gene therapy is that delivery of an exogenous gene could be harmful to the target cells or the organism as a whole, either as a direct result of gene expression or as a by-product of the delivery system. This is especially true in the case of bcl-2 because the ratio of pro- and anti-apoptotic proteins is thought to be important in determining whether a cell will live or die (Adams and Cory, 1998). Although delivery of bcl-2 to a cell that is destined to undergo apoptosis may be an effective way to balance the ratio between apoptotic proteins, delivery of bcl-2 to a healthy cell may not be desirable. Either the pro- to anti-apoptotic protein ratio could be thrown off, resulting in cell death instead of cell rescue, or, in a dividing cell, bcl-2 could become oncogenic, resulting in malignant transformation and tumor formation. To determine the safety of the adenoviral constructs used in this study, mice from all 3 strains were infected with each of the vectors and observed for clinical abnormalities including motor neuron dysfunction, weight loss, and tumor formation. None of the mice showed evidence of any of these problems, and all appeared healthy and active until the end of the experiment. Another concern of using Ad-bcl-2 stems from the inflammatory reaction often associated with adenoviral vectors. No signs of inflammation were found in spinal cord sections of treated mice, consistent with observations in other studies that the CNS is protected from the usual inflammatory response (Ghadge et al., 1995). 152 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The main question addressed by this study is whether bcl-2 delivered via an adenoviral vector can alter disease progression in G93 A mice. Average onset in Ad- bcl-2 infected mice was delayed by 8.5 days, indicating that bcl-2 was able to interfere with mechanisms involved in disease onset. This effect is similar to other treatments that have successfully delayed onset of disease symptoms in G93A mice, including the anti-oxidant vitamin E and intra-spinal injection of an adeno-associated virus-bcl-2 vector (Assouz et al., 2000; Gurney et al., 1996). Despite the success of bcl-2 in delaying onset of disease, disease progression was unaffected. The length of disease was the same in treated and control mice, and weight loss was also similar in all three groups. G93A mice crossed with mice over-expressing bcl-2 (G93A/bcl-2 mice) show a similar pattern, with delayed onset but no prolongation of disease (Kostic et al., 1997). It is important to note, however, that 2 mice in the Ad-bcl-2 treated group were still alive after the control mice in both untreated and Ad-lacZ treated groups had reached end-stage. Although there is no statistical significance associated with this observation, weight loss in bcl-2 treated mice seemed to be declining less rapidly at end stage than in control mice. One possibility to explain the effect of bcl-2 on onset and its lack of effect on progression is that different mechanisms are involved in each of these processes. It has been hypothesized that disease in FALS mice, and by extension in human FALS, has two components. Onset is thought to be the result of oxidative dysregulation in motor neurons, while progression is associated with glutamate excitotoxicity (Doble, 1999). These pathways feed in to each other creating a downward spiral of disease 153 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. progression. Bcl-2 is thought to have an anti-oxidant effect separate from its role in apoptosis, and it may be that the function of bcl-2 is to prevent oxidative damage in FALS motor neurons, thus delaying onset of disease. Once disease has begun, however, bcl-2 may no longer be involved in the pathway leading to cell damage and death, and so no change is seen in length of the disease course. Alternatively, even if bcl-2 does play a role in preventing motor neuron death after disease onset, it may be that too many neurons do not express excess bcl-2, and the overwhelming cycle of glutamate excitotoxicity and oxidative damage may eventually win out. In order to determine whether bcl-2 simply interferes with oxidative damage at the onset of disease or if its presence can rescue motor neurons, cell counts were performed at onset and end-stage in treated and control mice. These data show that even as early as disease onset there was no observable difference in numbers of motor neurons between treated and untreated mice, and this observation was confirmed at end-stage. This is in contrast to two other studies that used bcl-2 to treat G93A mice (Assouz et al., 2000; Kostic et al., 1997). Both found that over- expression of bcl-2 resulted in decreased motor neuron death. Each of these studies, however, used high levels of bcl-2 over-expression that affected multiple cell types in the spinal cord. It has been hypothesized that activated glial cells contribute to motor neuron degeneration in the G93A mouse, possibly by producing excess ROS (Hall et al., 1998). The ability of bcl-2 to rescue motor neurons from cell death in these studies may be the result of bcl-2 expression in glial cells inhibiting production of ROS, thus allowing motor neurons to survive longer than in the current study. In 154 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. support of this is the fact that in G93A/bcl-2 mice, the number of motor neurons at end-stage was identical to that found in non-hybrid G93A mice, indicating that the effect of bcl-2 on motor neuron survival was transient. In the current study expression of bcl-2 was limited to neurons, thus the effect of bcl-2 over-expression on motor neuron function and survival could be isolated. The results show that bcl-2 does not directly act to increase survival when expressed in motor neurons alone. This evidence supports the theory that bcl-2 function is limited to disease onset, possibly by acting as an anti-oxidant, but does not rescue neurons from cell death once disease has begun. One recent study has suggested that disease onset does not involve motor neuron loss, but rather decreased motor neuron function and axonal damage, and that the majority of motor neuron loss occurs at the later stages of disease, possibly correlating with onset of paralysis (Kennel et al., 1996). Bcl-2 may have a role in preventing this decreased function by limiting early oxidative damage within motor neurons and thus delaying clinical onset but not affecting later motor neuron death and disease progression. In conclusion, in this study I have demonstrated the ability of an adenoviral vector to deliver a human bcl-2 cDNA to mouse motor neurons via i.m. injection and express protein for up to 8 weeks. In addition I have shown that expression of bcl-2 by this method results in a significant delay in motor neuron disease onset in FALS mice. The effect of bcl-2, however, seems to be limited to the early stages of disease, possibly involving an anti-oxidant role for bcl-2 resulting in a delay in motor neuron dysfunction that is manifested as a delayed onset of disease. After disease 155 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. onset, however, the presence of bcl-2 seems to have no effect on disease progression or motor neuron survival. Therefore bcl-2, although a partially effective therapy for treatment of motor neuron degeneration in FALS mice, may need to be combined with other therapies that can counteract the mechanisms involved in disease progression to effectively treat motor neuron disease. 156 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EXPERIMENTAL PROCEDURES Transgenic mice and experimental design Transgenic mice over-expressing human Cu/Zn superoxide dismutase (SOD1) containing a glycine to alanine mutation at position 93 (G93ASOD1) were purchased from Jackson Laboratories, along with mice over-expressing the wild-type human SOD1 (WTSOD1), and background B6SJL mice. Mice were kept in the vivarium at the Keck School of Medicine of the University of Southern California. They were housed in plastic filter-top cages in an isolated room with a 12 hour light.dark cycle. All mice had free access to food and water. Mice were observed three times a week for signs of motor neuron disease and weight loss. After disease onset mice were observed daily. A blinded observer confirmed clinical symptoms. All procedures conformed to the university’s Internal Animal Care and Use Committee (LACUC) and Biosafety Committee guidelines. B6SJL and WTSOD1 control mice were split into 4 groups. The first group from each strain was sacrificed without injection to serve as an early control (B6CE or WTCE, n=5 for each strain). The second group of controls was treated with normal saline and sacrificed at the end of the experiment as late controls (B6CL or WTCL, n=5 for each strain). The third and fourth groups were treated with Ad-lacZ (B6LL or WTLL, n=4 for each species) or Ad-bcl-2 (B6BL or WTBL, n=7 for each species), respectively, and followed to the end of the experiment. G93A mice were 157 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. split into 10 groups. In addition to the above groups (GACE early controls, n=5; GACL late controls, n=4; GALL lacZ end stage, n=3; and GABL bcl-2 end stage, n=6) there were mice from each treatment (control, lacZ, or bcl-2) sacrificed at disease onset (GACO, n=5; GALO, n=4; GABO, n=5) and at estimated disease midpoint (GACM, n=3; GALM, n=2; GABM, n=3). One mouse in the G93A bcl-2 end stage group (GABL) died from disease before tissue could be collected. Intra-muscular injection of adenoviral vectors Approximately 2 weeks before onset of disease mice were treated with an adenovirus containing either cDNA for human bcl-2 (Ad-bcl-2, courtesy of Dr. Jeannie Chen, Keck School of Medicine) or the lacZ gene (Ad-lacZ, courtesy of Dr. Mel Trousdale, Keck School of Medicine). Before injection mice were deeply anesthetized with Avertin following IACUC protocol. 2 x 108 plaque forming units (p.f.u.) in a volume of 20 1 were injected into fore and hind limb musculature bilaterally. Control mice were injected with 201 of sterile normal saline. All mice survived the procedure with no apparent nerve or muscle damage. After injection mice were observed as described above. 158 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Clinical symptoms Clinical progression in the G93A mouse has been previously characterized (Andreassen et al., 2000; Assouz et al., 1997; Gurney et al., 1994; Kennel et al., 1996). Disease onset is characterized by hind limb weakness followed by hind limb tremor. Disease progression involves onset of gait abnormalities (waddling) followed by paralysis in the limbs. End stage was defined by the inability of a mouse to right itself within 10 seconds after being turned on its side (Andreassen et al., 2000). For data on disease onset mice had to show initial symptoms for 2 consecutive examinations, and the first day was used as the date of onset. Collection of spinal cords At designated times, mice were sacrificed and spinal cords collected as follows. Mice were deeply anesthetized using Avertin and perfused with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) via intracardiac catheterization. After perfusion, the entire spinal column was removed from the animal and a laminectomy performed. At this point the spinal tissue was post-fixed overnight in 4% paraformaldehyde at 4°C. On day 2 the spinal cord was removed from the remainder of the spinal column, washed for 3 x 10 minutes in ice-cold 0.1 M phosphate buffer (pH 7.4), and placed in cryoprotectant (30% sucrose in 0.1 M phosphate buffer, pH7.4) overnight at 4°C. On day 3 cords were cut into cervical 159 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. enlargement, thoracic and lumbar enlargement regions and frozen in OCT compound. Histology 20 um-thick serial sections from each lumbar enlargement were cut on a cryostat, placed on poly-L-lysine coated slides and stored at -80°C until staining. Ten sections per mouse, each spaced 240 um apart, were stained for Nissl substance with 0.25% cresyl violet acetate (Sigma). Images were captured using a Leica DMR microscope attached to a SPOT camera (Diagnostic Instruments) and analyzed with Adobe Photoshop 6.0. Both anterior horns from each section were photographed. The region of the anterior horn was defined by staining and by drawing a bisector through the central canal in the coronal plane. Cell counts were performed using an optical dissector and according to the principles of stereology (West, 1999). The examiner was blinded to treatment groups and species. Immunohistochemistry for human bcl-2 Sections from each treatment group were also stained for expression of human bcl-2 in motor neurons. The antibody was a mouse monoclonal anti-human bcl-2 (sc-509) (Santa Cruz) antibody at a dilution of 1:100. To prevent high background levels, the M.O.M. staining kit (Vector labs) for mouse antibodies on 160 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mouse tissue was used. Images were captured as described above. A blinded observer performed assessment of positive staining. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTERS SUMMARY AND CONCLUSIONS Motor neuron death in SODl-associated FALS appears to be the result of an aberrant cellular oxidative state (Liu et al., 1999; Wiedau-Pazos et al., 1996). The connection between oxidative damage and cell death is not yet known, although damage to mitochondria and increased intracellular calcium appear to play an important role (Doble, 1999). These findings are consistent with initiation of a programmed cell death mechanism, but evidence for classical apoptosis in FALS has been contradictory (Martin, 1999; Mu et al., 1996; Troost et al., 1995). In this thesis I looked for signs of apoptosis in both the Mn-1 cell line and the MNGA cell line, a tissue culture model of SOD-l-associated FALS that was developed in this lab and described in chapter 3. In addition, I used a gene therapy approach to treat cell death in the MNGA model and in a murine model of SODl-associated FALS by increasing levels of the anti-apoptotic protein bcl-2. The work presented here helps to clarify the type of apoptotic cell death found in SOD-1 associated motor neuron disease. It also shows that adenoviral-mediated gene therapy using bcl-2 can rescue MNGA cells in culture from oxidative damage-induced cell death and can interfere with disease onset in G93A-FALS mice. In chapter 2 1 characterized the Mn-1 cell line for functional and developmental motor neuron markers. The results showed that Mn-1 cells might be 162 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. derived from an embryonic cell population that will eventually innervate limb musculature. Since paralysis of the limbs is an early sign of ALS, I looked at the response of Mn-1 cells to oxidative stress, which has been implicated in disease pathogenesis in ALS (Liu et al., 1999). I found that treatment with H2O2 produced Mn-1 cell death with some, but not all, of the features normally associated with apoptosis. Most notably lacking was activation of the caspase cascade. Instances of caspase-independent apoptosis have been described, and one of the controversies surrounding the type of cell death in ALS has been an inability to consistently demonstrate caspase activity (Li et al., 2000; Martin, 1999; Mighelli et al., 1999). Based on the results presented in chapter 1 1 concluded that the Mn-1 line was an appropriate cell line for modeling oxidative damage-induced apoptosis in motor neurons. In chapter 3 I used the Mn-1 motor neuron line to create stable cell lines expressing either the human wild-type SOD1 (MNWT cells) or a mutant SOD1 containing a glycine to alanine mutation at position 93 (MNGA cells), which has been found in SODl-associated FALS. These cell lines respond to oxidative stress similar to the Mn-1 parental line, undergoing a caspase-independent form of apoptosis. The presence of mutant SOD1 in differentiated MNGA cells conferred increased susceptibility to oxidative-stress compared to Mn-1 or MNWT cells, paralleling the effect of mutant SOD1 in human and murine forms of FALS. From this I concluded thai the MNGA cell line could be used as a tissue culture model to study the mechanisms of cell death in SODl-associated FALS. In addition I infected 163 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MNGA cells with an adenoviral vector containing cDNA for the anti-apoptotic human bcl-2 protein (Ad-bcl-2) to see if increased levels of bcl-2 could alter the specific type of apoptosis found in this model. The results showed that differentiated MNGA cells infected with Ad-bcl-2 24 hours before treatment with H2O2 were significantly less susceptible to oxidative-stress induced apoptosis than non-infected cells or cells infected with an adenovirus containing the lacZ gene (Ad-lacZ). The method of action of bcl-2 in this system is unknown. Although normally associated with an anti-apoptotic function, recent studies suggest that bcl-2 may have a direct anti-oxidant role, and it may be this function that protects MNGA cells from H202-induced apoptosis (Kane et al., 1993; Tyurina et al., 1997). One unexpected finding of treatment with bcl-2 was down-regulation of endogenous bcl-xl. Bcl-2 is normally replaced by bcl-xl as the major anti-apoptotic protein in the adult central nervous system (Gonzalez-Garcia et al., 1995). Over-expression of bcl-2 in differentiated motor neurons may hinder the normal cellular function of bcl-xl. This could be detrimental to bcl-2 infected cells if there is a divergence of function between bcl-2 and bcl-xl and might explain why the anti-apoptotic effect of bcl-2 seems to be temporally limited. In future studies it might be useful to examine the effects of bcl-xl over-expression in this model. The expression of an anti-apoptotic protein that is found in adult motor neurons may have a longer or more pronounced effect on motor neuron survival, and cause fewer potentially deleterious alterations, than one that is foreign to the normal function of the cell. 164 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In chapter 4 1 looked at the efficacy of Ad-bci-2 in treating a murine form of FALS in the G93A mouse, which expresses the same mutant SOD1 used to create the MNGA cell line. Motor neuron death in FALS mice, similar to the findings in human ALS, has not been definitively identified as apoptotic. Despite this, two recent studies have shown that expression of the anti-apoptotic protein bcl-2 in G93A mice can delay onset of disease and protect motor neurons from death (Assouz et al., 2000; Kostic et al., 1997). Although successful in demonstrating a therapeutic role for bcl-2 in FALS mouse motor neuron degeneration, neither method of gene delivery is clinically applicable. Furthermore it is possible that the ability of bcl-2 to limit motor neuron death in these studies is the result of bcl-2 expression in glia in addition to motor neurons. One recent study has shown that Ad-lacZ delivered as an intra-muscular injection is transported retrograde and expressed in anterior horn motor neurons of G93A mice (Warita et al., 1998). I decided to use this approach to deliver bcl-2 to motor neurons in G93A mice. The results showed a significant delay in onset of disease in mice treated with Ad-bcl-2 in comparison to Ad-lacZ-treated and control mice, but no change in overall disease progression and motor neuron survival. One possibility is that bci-2 is acting as an anti-oxidant rather than as an anti-apoptotic protein. This is supported by recent evidence implicating two different mechanisms of disease in FALS. Onset of disease has been associated with oxidative damage to intracellular structures of motor neurons while progression of disease seems to result from glutamate excitotoxicity compounded by further oxidative damage, possibly 165 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. leading to motor neuron apoptosis (Doble, 1999). Bcl-2 may be able to interfere with the first step in the disease process, resulting in delayed onset, but be unable to prolong disease and protect motor neurons after onset. If bcl-2 is acting in an anti oxidant pathway rather than as an anti-apoptotic protein no conclusions can be reached from this data regarding the form of cell death in the G93A mouse. Although bcl-2 is a promising gene therapy candidate for treatment of motor neuron disease based on these results, it is evident that bcl-2 alone cannot provide full protection against motor neuron degeneration. Future studies will need to identify therapies to be used in combination with bcl-2 to both delay onset and halt progression of motor neuron disease. In summary, in this thesis 1 established a tissue culture model of FALS motor neuron death (the MNGA cell line) and showed that oxidative damage induced cell death in both Mn-1 motor neuron and MNGA cells is a caspase-independent form of apoptosis. In addition I was able to show that over-expression of bcl-2 in MNGA cells is protective in this model of cell death. Unexpectedly, over-expression of bcl- 2 resulted in down-regulation of bcl-xl, the primary anti-apoptotic bcl-2 family member expressed in the adult nervous system, suggesting that bcl-2 may interfere with normal adult motor neuron anti-cell death pathways. I also showed that an adenovirus-bcl-2 vector delivered to motor neurons via intra-muscular injection delayed onset of disease in G93A FALS mice, but was unable to prolong survival of motor neurons or extend the course of disease. Although the exact function of bcl-2 in this system is unknown, based on the results of experiments in the MNGA cell 166 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. line and the pattern of protection of bcl-2 in the G93A mouse, it can be hypothesized that bcl-2 is acting as an anti-oxidant rather than as an anti-apoptotic or anti- excitotoxic factor to protect motor neurons in these models. 167 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES Adams, J. M., and Cory, S. (1998). The Bcl-2 protein family: arbiters of cell survival, Science 281, 1322-1326. Allen, R. T., Cluck, M. W., and Agrawal, D. K. (1998). Mechanisms controlling cellular suicide: role of Bcl-2 and caspases, Cellular and Molecular Life Sciences 54, 427-445. Andreassen, O. A., Ferrante, R. J., Klivenyi, P., Klein, A. M., Shinobu, L. A., Epstein, C. J., and Beal, M. F. (2000). Partial deficiency of manganese superoxide dismutase exacerbates a transgenic mouse model of amyotrophic lateral sclerosis, Annals of Neurology 4 7 ,447-455. Ankarcrona, M., Dypbukt, J. M., Brune, B., and Nicotera, P. (1994). Interleukin-IB- induced nitric oxide production activates apoptosis in pancreatic RINm5F cells, Experimental Cell Research 213, 172-177. Armstrong, R. C., Aja, T. J., Hoang, K. D., Gaur, S., Bai, X., Alnemri, E. S., Litwack, G., Karanewsky, D. S., Fritz, L. C., and Tomaselli, K. J. (1997). Activation of the CED3/ICE-related protease CPP32 in cerebellar granule neurons undergoing apoptosis but not necrosis, The Journal of Neuroscience 17,553-562. Assouz, M., Hottinger, A., Patema, J.-C., Zum, A. D., Aebischer, P., and Bueler, H. (2000). Increased motomeuron survival and improved neuromuscular function in transgenic ALS mice after intraspinal injection of an adeno-associated virus encoding Bcl-2, Human Molecular Genetics 9, 803-811. Assouz, M., Leclerc, N., Gumey, M. E., Warter, J.-M., Poindron, P., and Borg, J. (1997). Progressive motor neuron impairment in an animal model of familial amyotrophic lateral sclerosis, Muscle and Nerve 20,45-51. Bensadoun, A., and Weinstein, D. (1976). Assay of proteins in the presence of interfering materials, Analytical Biochemistry 70, 241-250. Bortner, C. D., Oldenberg, N. B. E., and Cidlowski, J. A. (1995). The role of DNA fragmentation in apoptosis, Trends in Cell Biology 5,21-26. 168 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Boyne, L. J., Fischer, I., and Shea, T. B. (1996). Role of vimentin in early stages of neuritogenesis in cultured hippocampal neurons, International Journal of Developmental Neuroscience 14, 739-748. Brooks, B. P., Merry, D. E., Paulson, H. L., Lieberman, A. P., Kolson, D. L., and Fischbeck, K. H. (1998). A cell culture model for androgen effects in motor neurons, Journal of Neurochemistry 7 0 ,1054-1060. Brown, R. (1997). The bcl-2 family of proteins, British Medical Bulletin 5 3 ,466- 477. Brujin, L. I., Houseweart, M. K., Kato, S., Anderson, S. D., Ohama, E., Reaume, A. G., Scott, R. W., and Cleveland, D. W. (1998). Aggregation and motor neuron toxicity of an ALS-linked SOD1 mutant independent from wild-type SOD I, Science 2 81,1851-1854. Carri, M. T., Ferri, A., Battistoni, A., Famhy, L., Gabbianelli, R., Poccia, F., and Rotilio, G. (1997). Expression of a Cu,Zn superoxide dismutase typical of familial amyotrophic lateral sclerosis induces mitochondrial alteration and increase of cytosolic calcium concentration in transfected neuroblastoma SH-SY5Y cells, FEBS Letters 414,365-368. Chao, M., Casaccia-Bonnefil, P., Carter, B., Chittka, A., Kong, H., and Yoon, S. O. (1998). Neurotrophin receptors: mediators of life and death, Brain Research Reviews 26, 295-301. Chen, E. W., and Chiu, A. Y. (1992). Early stages in the development of spinal motor neurons, Journal of Comparative Neurology 320, 291-303. Cleveland, D. W. (1996). Neuronal growth and death: order and disorder in the axoplasm, Cell 84, 663-666. Cleveland, D. W., Brujin, L. I., Wong, P. C., Marszalek, J. R., Vechio, J. D., Lee, M. K., Xu, X.-S., Borchelt, D. R., Sisodia, S., and Price, D. L. (1996). Mechanisms of selective motor neuron death in transgenic mouse models of motor neuron disease, Neurology 47, S54-S62. Coyle, J. T., and Puttfarcken, P. (1993). Oxidative stress, glutamate, and neurodegenerative disorders, Science 262, 689-695. Cudkowicz, M. E., McKenna-Yasek, D., Sapp, P. E., Chin, W., Geller, B., Hayden, D. L., Schoenfeld, D. A., Hosier, B. A., Horvitz, H. R., and Brown, R. H. (1997). Epidemiology of mutations in superoxide dismutase in amyotrophic lateral sclerosis, Aimals of Neurology 41,210-221. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dal Canto, M. C., and Gurney, M. E. (1994). Development of central nervous system pathology in a murine transgenic model of human amyotrophic lateral sclerosis, American Journal of Pathology 145,1271-1279. Dawson, V. L., and Dawson, T. M. (1998). Nitric oxide in neurodegeneration, Progress in Brain Research 118,215-229. de Belleroche, J., Orrell, R., and King, A. (1995). Familial amyotrophic lateral sclerosis/ motor neurone disease (FALS): a review of current developments, Journal of Medical Genetics 3 2 ,84-87. Deng, H.-X., Hentati, A., Tainer, J. A., Iqbal, Z., Cayabyab, A., Hung, W.-Y., Getzoff, E. D., Hu, P., Herzfeldt, B., Roos, R. P., et al. (1993). Amyotrophic lateral sclerosis and structural defects in Cu,Zn superoxide dismutase, Science 261, 1047- 1051. Deveraux, Q. L., Stennicke, H. R., Salvesen, G. S., and Reed, J. C. (1999). Endogenous inhibitors of caspases, Journal of Clinical Immunology 19, 388-398. Doble, A. (1999). The role of excitotoxicity in neurodegenerative disease: implications for therapy, Pharmacological Therapy 8 1 ,163-221. Engelhard, H. H., Egli, M., and Rozental, J. M. (1998). Use of antisense vectors and oligodeoxynucleotides in neuro-oncology, Pediatric Neurosurgery 28, 279-285. Ericson, J., Rashbass, P., Schedl, A., Brenner-Morton, S., Kawakami, A., van Heyningen, V., Jessell, T. M., and Briscoe, J. (1997). Pax6 controls progenitor cell identity and neuronal fate in response to graded Shh signalling, Cell 9 0 ,169-180. Facchinetti, F., Dawson, V. L., and Dawson, T. M. (1998). Free radicals as mediators of neuronal injury, Cellular and Molecular Neurobiology 18, 667-682. Ghadge, G. D., Lee, J. P., Bindokas, V. P., Jordan, J., Ma, L., Miller, R. J., and Roos, R. P. (1997). Mutant superoxide dismutase-1-linked familial amyotrophic lateral sclerosis: molecular mechanisms of neuronal death and protection, The Journal of Neuroscience 17,8756-8766. Ghadge, G. D., Roos, R. P., Kang, U. J., Wollman, R., Fishman, P. S., Kalynych, A. M., Barr, E., and Leiden, J. M. (1995). CNS gene delivery by retrograde transport of recombinant replication-defective adenoviruses, Gene Therapy 2, 132-137. Gonzalez-Garcia, m., Garcia, I., Ding, L., Sue, O. S., Boise, L. H., Thompson, C. B., and Nunez, G. (1995). Bcl-x is expressed in embryonic and postnatal neural tissues and functions to prevent neuronal cell death, Proceedings of the National Academy of Sciences USA 92,4304-4308. 170 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Goulding, M. (1998). Specifying motor neurons and their connections, Neuron 21, 943-946. Gurney, M. E., Cutting, F. B., Zhai, P., Doble, A., Taylor, C. P., Andrus, P. K., and Hall, E. D. (1996). Benefit of vitamin E, Riluzole, and gabapentin in a transgenic model of familial amyotrophic lateral sclerosis, Annals of Neurology 3 9 ,147-157. Gurney, M. E., Pu, H., Chiu, A. Y., Dal Canto, M. C., Polchow, C. Y., Alexander, D. D., Caliendo, J., Hentati, A., Kwon, Y. W., Deng, H.-X., et al. (1994). Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation, Science 264,1772-1775. Gurney, M. E., Tomasselli, A. G., and Heinrikson, R. L. (2000). Stay the executioner's hand, Science 288, 283-284. Hall, E. D., Oostveen, J. A., and Gurney, M. E. (1998). Relationship of microglial and astrocytic activation to disease onset and progression in a transgenic model of familial ALS, Glia 23, 249-256. Harris, J. D., and Lemoine, N. R. (1996). Stratehies for targeted gene therapy, Trends in Genetics 12,400-405. He, B. P., and Strong, M. J. (2000). Motor neuronal death in sporadic amyotrophic lateral sclerosis (ALS) is not apoptotic. A comparative study of ALS and chronic aluminium chloride neurotoxicity in New Zealand white rabbits, Neuropathology and Applied Neurobiology 2 6 ,150-160. Holinger, E. P., Chittenden, T., and Lutz, R. J. (1999). Bak BH3 peptides antagonize bcl-xl function and induce apoptosis through cytochrome c-independent activation of caspses, The Journal of Biological Chemistry 274,13298-13304. Huang, D. C. S., and Strasser, A. (2000). BH3-only proteins-Essential initiators of apoptotic cell death, Cell 103,839-842. Ince, P. G., Lowe, J., and Shaw, P. J. (1998). Amyotrophic lateral sclerosis: current issues in classification, pathogenesis and molecular pathology, Neuropathology and Applied Neurobiology 2 4 ,104-117. Kane, D. J., Sarafian, T. A., Anton, R., Hahn, H., Gralla, E. B., Valentine, J. S., Ord, T., and Bredesen, D. E. (1993). Bcl-2 inhibition of neural death: decreased generation of reactive oxygen species, Science 262,1274-1277. Karpati, G., Lochmuller, H., Nalbantoglu, J., and Durham, H. (1996). The principles of gene therapy for the nervous system, Trends in Neuroscience 19,49-54. 171 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Kennel, P. F., Finiels, F., Revah, F., and Mallet, J. (1996). Neuromuscular function impairment is not caused by motor neurone loss in FALS mice: an electromyographic study, Neuroreport 7,1427-1431. Kemic, S. G., and Parada, L. F. (2000). The molecular basis for understanding neurotrophins and their relevance to neurologic disease, Archives of Neurology 57, 654-657. Kerr, J. F. R., Wyllie, A. H., and Currie, A. R. (1972). Apoptosis: A basic biological phenomenon with wide-ranging implications in tissue kinetics, British Journal of Cancer 26, 239-257. Korsmeyer, S. J. (1995). Regulators of cell death, Trends in Genetics 11, 101-105. Kostic, V., Jackson-Lewis, V., de Bilbao, F., Dubois-Dauphin, M., and Przedborski, S. (1997). Bcl2: Prolonging life in a transgenic mouse model of familial amyotrophic lateral sclerosis, Science 277, 559-562. Li, M., Ona, V. O., Geugan, C., Chen, M., Jackson-Lewis, V., Andrews, L. J., Olszewski, A. J., Stieg, P. E., Lee, J. P., Przedborski, S., and Friedlander, R. M. (2000). Functional role of caspase-1 and caspase-3 in an ALS transgenic mouse model, Science 288, 335-339. Lin, C.-L. G., Bristol, L. A., Dykes-Hoberg, M., Crawford, T., Clawson, L., and Rothstein, J. D. (1998). Aberrant RNA processing in a neurodegenerative disease: the cause for absent EAAT2, a glutamate transporter, in amyotrophic lateral sclerosis, Neuron 20,589-602. Lin, E. Y., Orlofsky, A., Berger, M. S., and Prystowsky, M. B. (1993). Characterization of A l, a novel hemopoietic-specific early-response gene with sequence similarity to bcl-2, The Journal of Immunology 151,1979-1988. Lincz, L. F. (1998). Deciphering the apoptotic pathway: All roads lead to death, Immunology and Cell Biology 76,1-19. Lipton, S. A., Choi, Y.-B., Pan, Z.-H., Lei, S. Z., Chen, H.-S. V., Sucher, N. J., Loscalzo, J., Singel. D. J., and Stamler, J. S. (1993). A redox-based mechanism for the neuroprotective and neurodestructive effects of nitric oxide and related nitroso- compounds, Nature 364, 626-632. Liu, D., Wen, J., Liu, J., and Li, L. (1999). The roles of free radicals in amyotrophic lateral sclerosis: reactive oxygen species and elevated oxidation of protein, DNA, and membrane phospholipids, The FASEB Journal 13,2318-2328. 172 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nogami, N., Takatsu, A., Endo, N., and Ishiyama, I. (1998). Immunohistochemistry of neuron-specific enolase in neurons of the medulla oblongata from human autopsies, Acta Histochemica 100,371-282. Orrell, R. W. (2000). Amyotrophic lateral sclerosis: copper/zinc superoxide dismutase (SOD1) gene mutations, Neuromuscular Disorders 10,63-68. Pasinelli, P., Borchelt, D. R., Houseweart, M. K., Cleveland, D. W., and Brown, R. H. (1998). Caspase-1 is activated in neural cells and tissue wioth amyotrophic lateral sclerosis-associated mutations in copper-zinc superoxide dismutase, Proceedings of the National Academy of Sciences USA 95, 15763-15768. Pelligrini, M., and Strasser, A. (1999). A portrait of the Bcl-2 protein family: life, death, and the whole picture, Journal of Clinical Immunology 19, 365-377. Pettman, B., and Henderson, C. E. (1998). Neuronal cell death, Neuron 20,633-647. Pfaff, S. L., and ICintner, C. (1998). Neuronal diversification: development of motor neuron subtypes, Current Opinion in Neurobiology 8, 27-36. Rabizadeh, S., Gralla, E. B., Borchelt, D. R., Gwinn, R., Valentine, J. S., Sisodia, S., Wong, P., Lee, M., Hahn, H., and Bredesen, D. E. (1995). Mutations associated with amyotrophic lateral sclerosis convert superoxide dismutase from an antiapoptotic gene to a proapoptotic gene: studies in yeast and neural cells, Proceedings of the National Academy of Sciences USA 92, 3024-3028. Ripps, M. E., Huntley, G. W., Hof, P. R., Morrison, J. H., and Gordon, J. W. (1995). Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis, Proceedings of the National Academy of Sciences USA 92,689-693. Robberecht, W. (2000). Oxidative stress in amyotrophic lateral sclerosis, Journal of Neurology 247 [Suppl 1), I/1-I-6. Rosen, D. R., Siddique, T., Patterson, D., Figlewicz, D. A., Sapp, P., Hentati, A., Donaldson, D., Goto, J., O'Regan, J. P., Deng, H.-X., et al. (1993). Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis, Nature 362,59-62. Rubin, L. L. (1997). Neuronal cell death: when, why, and how, British Medical Bulletin 5 3 ,617-631. 174 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Saikumar, P., Dong, Z., Mikhialov, V., Denton, M., Weinberg, J. M., and Venkatacbalam, M. A. (1999). Apoptosis: Definition, mechanisms, and relevance to disease, American Journal of Medicine 107,489-506. Salazar-Grueso, E. F., Kim, S., and Kim, H. (1991). Embryonic mouse spinal cord motor neuron hybrid cells, Neuroreport 2,505-508. Sar, M., and Stumpf, W. E. (1997). Androgen concentration in motor neurons of cranila nerves and spinal cord, Science 197, 77-78. Sendtner, M., Pei, G., Beck, M., Schweizer, U., and Wiese, S. (2000). Developmental motoneuron cell death and neurotrophic factors, Cell Tissue Research 301, 71-84. Shaw, P. J., Williams, T. L., Slade, J. Y., Eggett, C. J., and Ince, P. G. (1999). Low expression of GluR2 AMPA receptor subunit protein by human motor neurons, Neuroreport 10,261-265. Takadera, T., Matsuda, I., and Ohyashiki, T. (1999). Apoptotic cell death and caspase-3 activation induced by N-methyl-D-aspartate receptor antagonists and their prevention by insulin-like growth factor I, Journal of Neurochemistry 73,548-556. Tanabe, Y., William, C., and Jessell, T. M. (1998). Specification of motor neuron identity by the MNR2 homeodomain protein, Cell 95, 67-80. Thomberry, N. A. (1996). The caspase family of cysteine proteases, British Medical Bulletin 53, 478-490. Toku, K., Tanaka, J., Yano, H., Desaki, J., Zhang, B., Yang, L., Ishihara, K., Sakanaka, M., and Maeda, N. (1998). Microglial cells prevent NO-induced neuronal apoptosis in vitro, Journal of Neuroscience Research 53,415-425. Troost, D., Aten, J., Morsink, F., and de Jong, J. M. B. V. (1995). Apoptosis in amyotrophic lateral sclerosis is not restricted to motor neurons. Bcl-2 expression is increased in unaffected post-central gyrus, Neuropathology and Applied Neurobiology 21,498-504. Tsuchida, T., Ensini, M., Morton, S. B., Baldassare, M., Edlund, T., Jessell, T. M., and Pfaff, S. L. (1994). Topographic organization of embryonic motor neurons defined by expression of LIM homeobox genes, Cell 79,957-970. 175 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tu, P.-H., Raju, P., Robinson, K. A., Gurney, M. E., Trojanowski, J. Q., and Lee, V. M.-Y. (1996). Transgenic mice carrying a human mutant superoxide dismutase transgene develop neuronal cytoskeletal pathology resembling human amyotrophic lateral sclerosis lesions, Proceedings of the National Academy of Sciences USA 93, 3155-3160. Tyurina, Y. Y., Tyurin, V. A., Carta, G., Quinn, P. J., Schor, N. F., and Kagan, V. E. (1997). Direct evidence for antioxidant effect of bcl-2 in PC12 rat pheochromacytoma cells, Archives of Biochemistry and Biophysics 344,413-423. Vandenberghe, W., Robberecht, W., and Brorson, J. R. (2000). AMPA receptor calcium permeability, GluR2 expression, and selective motoneuron vulnerability, The Journal of Neuroscience 2 0 ,123-132. Vaquero, J., Coca, M., Oya, S., Arias, A., Zurita, M., and Morales, C. (1992). Expression of neuronal and glial markers in so-called oligodendroglial tumors induced by transplacental administration of ethyl-nitrosourea in the rat, Histology and Histopathology 7,647-651. Vaux, D. L., and Strasser, A. (1996). The molecular biology of apoptosis, Proceedings of the National Academy of Sciences USA 93, 2239-2244. Vukosavic, S., Dubois-Dauphin, M., Romero, N., and Przedborski, S. (1999). Bax and Bcl-2 interaction in a transgenic mouse model of familial amyotrophic lateral sclerosis, Journal of Neurochemistry 73,2460-2468. Vukosavic, S., Stefanis, L., Jackson-Lewis, V., Guegan, C., Romero, N., Chen, C., Dubois-Dauphin, M., and Przedborski, S. (2000). Delaying caspase activation by bcl-2: A clue to disease retardation in a transgenic mouse model of amyotrophic lateral sclerosis, The Journal of Neuroscience 2 0 ,9119-9125. Walling, A. D. (1999). Amyotrophic lateral sclerosis: Lou Gehrig's disease, American Family Physician. Warita, H., Abe, K., Setoguchi, Y., and Itoyama, Y. (1998). Expression of an adenovirus-mediated E. coli lacZ gene in skeletal muscles and spinal motor neurons of transgenic mice with a mutant superoxide dismutase gene, Neuroscience Letters 246, 153-156. West, M. J. (1999). Stereological methods for estimating the total number of neurons and synapses: issues of precision and bias, Trends in Neuroscience 22,51-61. 176 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wiedau-Pazos, M., Trudell, J. R., Altenbach, C., Kane, D. J., Hubbell, W. L., and Bredesen, D. E. (1996). Expression of bcl-2 inhibits cellular radical generation, Free Radical Research 24, 205-212. Williams, T. L., Day, N. C., Ince, P. G., Kamboj, R. K., and Shaw, P. J. (1997). Calcium-permeable alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptors: A molecular determinant of selective vulnerability in amyotrophic lateral sclerosis, Annals of Neurology 42, 200-207. Wong, P. C., Pardo, C. A., Borchelt, D. R., Lee, M. K., Copeland, N. G., Jenkins, N. A., Sisodia, S., Cleveland, D. W., and Price, D. L. (1995). An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria, Neuron 14, 1105-1116. Wyllie, A. H. (1997). Apoptosis: an overview, British Medical Bulletin 53,451-465. Yin, X.-M., Oltvai, Z. N., and Korsmeyer, S. J. (1994). Heterodimerization with bax is required for bcl-2 to repress cell death, Nature 369, 331-338. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Marker 1. N-CAM* 2. GluRl* 3. GluR2* 4. GluR3* 5. GluR4* APPENDIX A. List of primer sequences, annealing temperatures, and expected product lengths from RT-PCR for Mn-1 markers (see chapter 2). Primers T, Product length f-5’ GGC TGG CGG CGC TCA ATG G 3’ 60° r-5’ GAT GGC CCC GGT GCT CAG G 3’ f-5’ AGA AAG GGC GCC GGA CCA ACT A 3’ 60° r-5’ GCC ACC GCC ACA TCT GCT CTT C 3’ f-5’ AAG GGG CGC TGA TCA AGA ATA CA 3’ 55° r-5’ CCG CAG CAG AAT CCA GCA CA 3’ f-5’ GGC GCA AGG GAT CCA GAG ACT AA 3’ 55° r-5’ GCA CCC AAG GAA AAC CAA AGA CTG 3’ f-5’ TGG TGT CAG CGT GGT CTT GTT CCT 3’ r-5’ GCC ACG CCC TCA GCT GTA GTT CT 3’ 55° 419 bp 403bp 407bp 404bp 471 bp 00 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6. TrkB* 7. TrkC* 8. Fas 9. NSE* 10. Calb* 11. Parv* 12. Vim* 13. GFAP* f-5’ GCG CGG CTC TGG GGC TTA TG 3’ 60° 481bp r-5’ CCT GAG TGT CGG GGC TGG ATT TAG 3’ f-5’ TGC CCA GCC AAG TGT AGT TTC T 3’ 60° 528bp r-5’ CCG CGC CTC CCC CTG TT 3’ f-5’ TGG AAA CAA ACT GCA CCC TGA C 3’ 55° 417bp r-5’ TGC CCT CCT TGA TGT TAT TTT CTC 3’ f-5 ’ ATC GCA CCG GCC CTC ATC AG 3 ’ 60° 419bp r-5’ CCG CCT TCA TCC CCC ACA TTA GT 3’ f-5’ TCG ACG CTG ACG GAA GTG 3’ 55° 447bp r-5’ GTA GTA ACC TGG CCA TCT CTG TCA 3’ f-5’ TCA AGA AGG CGA TAG GAG 3’ 50° 299bp r-5’ TTA CGT TTC AGC CAC CAG 3’ f-5’ AGC CGC AGC CTC TAT TCC TCA TC 3’ 60° 463bp r-5’ GGG TGC TIT CGG CTT CCT CTC 3’ f-5’ CCT CCG CCA AGC CAA ACA CGA A 3’ 60° 435bp r-5’ TCA CCA TCC CGC ATC TCC ACA GTC 3’ vO Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14. Pax6 f-5’ CGG GAA AGA CTA GCA GCC AAA AT 3’ 55° 412bp r-5’ TGT GCG GAG GGG TGT AGG TAT CAT 3’ 15. Nkx2.2 f-5’ CCG AAG GGC CAG AGG AGG AGA GC 3’ 65° 515bp r-5’ CGG CGA GGG CAG AGG CGT CAC C 3’ 16. Isl 1 * f-5’ CTG CGG GAG GAT GGG CTT TTC T 3’ 60° 508bp r-5’ AGG GCG GCT GGT AAC TTT G 3’ 17. Isl2* f-5’ CAC GCC GCC TGC CTC AAG 3’ 60° 549bp r-5’ GCT CAA GCC GGT CAT CTC TAC TA 3’ 18. Liml f-5’ TCC CGG GAA GGC A AG CTC TAC TGT 3’ 55° 418bp r-5’ GTC CCC TAC GTT TGG CAC CTA 19. Lim3 f-5’ GCC GCG CCC AGG ACT TCG TGT A 3’ 65° 449bp r-5’ GGC GTC GCT GTC TTG TCC CTC CTG 3’ 20. Bax f-5’ GAG CAG CCG CCC CAG GAT G 3’ 60° 417bp r-5’ GGT GAG CGA GGC GGT GAG GAC 3’ 21. Bad f-5’ GAC CAG CAG CCC AGA GTA TGT TCC 3’ 60° 481 bp r-5’ TTG CCC AAG TTT CGA TCC CAC CAG 3’ 00 o Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22. Bid 23. Casl* 24. Cas3* 25. Cas8* 26. Cas9* 27. Bcl-2 28. Bcl-xl 29. BAG1 f-5’ GGG GGC CAA GCA CAT CAC A 3’ 60° r-5’ TGG AAG ACA TCA CGG AGC AAA GA 3’ f-5 ’ GAT CTG GGG TAT ACC CTG AAA GTG 3 ’ 50° r- 5’ TGA AGA GCA GAA AGC AAT AAA ATC 3’ f-5’ CTC GCT CTG GTA CGG ATG TGG A 3’ 55° r-5’ TGC TGC AAA GGG ACT GGA TGA A 3’ f-5’ AGC GCA GAC CAC AAG AAC AAA G 3’ 55° r-5’ CAC GCC AGT CAG GAT GCT AAG A 3’ f-5’ CAG GCC GGT GGA CAT TGG IT 3’ 60° r-5’ GGC AGC CGC TCC CGT TGA AG 3’ f-5’ CGC CGG GCT GGG GAT GAC TTC T 3’ 60° r-5’ CAC TTG TGG CCC AGG TAT GC 3’ f-5’ GGA GAC CCC CAG TGC CAT CAA T 3’ 60° r-5’ GTC CCA GCC GCC GTT CTC C 3’ f-5’ TGC AAG CCG CGG GTG AAG AAG AAA 60° r-5’ CCA ATT AAC ATG ACT CGG CAA CCA 483bp 414bp 477bp 438bp 440bp 428bp 433bp 485bp 0 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. '''Abbreviations: neuronal cell adhesion molecule (NCAM); AMPA/kainate glutamate receptor subunits 1-4 (GluRl-4); tyrosine kinase receptor B (trkB) and C (trkC); neuron- specific enolase (NSE); calbindin (Calb); parvalbumin (Parv); vimentin (Vim); glial fibrillary acidic protein (GFAP); islet-1 (lsll); islet-2 (Isl2); caspase-1 (Casl); caspase-3 (Cas3); caspase-8 (Cas8); caspase-9 (Cas9). 00 K > APPENDIX B. Representative agarose gels from RT-PCR for Mn-1 cell characterization (chapter 2). A. B. □□ s c an D. G. □n nn qd H. Figure B.l Representative agarose gels from RT-PCR for Mn-l cell surface receptors. Left panel from each figure shows non-differentiated results, right panel shows differentiated. Top panel represents the RT-PCR product, bottom panel is the P-actin control for each sample. A. Neuronal cell adhesion molecule (NCAM). B-E. AMPA/kainate glutamate receptor subunits 1-4 (GluR 1-4), respectively. F.- G. Tyrosine kinase receptors B and C (trkB and trkC), respectively. H. Fas. 183 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. □ □ □ □ Q Q D. □ □ E. Figure B.2. Representative agarose gels from RT-PCR for Mn-1 intracellular proteins. Left panel from each figure shows non-differentiated results, right panel shows differentiated. Top panel represents the RT-PCR product, bottom panel is the (3-actin control for each sample. A. Neuron-specific enolase (NSE) B. Calbindin. C. Parvalbumin. D. Vimentin. E. Glial fibrillary acidic protein (GFAP). 184 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. □ □ S B q n D. E. F. B C D C S ^2 C l C3 Figure B.3. Representative agarose gels from RT-PCR for Mn-1 developmental markers. Left panel from each figure shows non-differentiated results, right panel shows differentiated. Top panel represents the RT-PCR product, bottom panel is the P-actin control for each sample. A. Pax6 B. Nkx2.2. C. Islet-1 (Isll). D. Islet-2 (Isl2). E. Liml F. Lim3. 185 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. □ □ □□ □□ D. F. G. Figure B.4. Representative agarose gels from RT-PCR for Mn-1 pro-apoptotic markers. Left panel from each figure shows non-differentiated results, right panel shows differentiated. Top panel represents the RT-PCR product, bottom panel is the P-actin control for each sample. A. Bax B. Bad. C. Bid. D. Caspase-1 (Casl). E. Caspase-3 (Cas3) F. Caspase-8 (Cas8). G. Caspase-9 (Cas9). 186 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. n n □ □ □□ Figure B.5. Representative agarose gels from RT-PCR for Mn-1 anti-apoptotic markers. Left panel from each figure shows non-differentiated results, right panel shows differentiated. Top panel represents the RT-PCR product, bottom panel is the P-actin control for each sample. A. Bcl-2 B. Bcl-xl. C. BAG-1. 187 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B □ □ □ Figure B.6. Representative agarose gels from RT-PCR showing positive controls for markers that were negative in Mn-1 cells. Top panel represents the RT-PCR product, bottom panel is the (3-actin control for each sample. A. AMPA/kainate glutamate receptor subunit 3 (GluR3), positive control from mouse spinal cord. B. Calbindin, positive control from mouse cerebral cortex. C. Parvalbumin, positive control from mouse spinal cord. D. Glial fibrillary acidic protein (GFAP), positive control from mouse cerebral cortex. 188 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Development of replication -competent retroviral vectors for efficient, targeted gene therapy of cancer

PDF

Glial activation and neurotoxicity: Age-related cytokine expression, ApoJ-induced microglial activation, and role of peroxynitrite in mediating neuron death

PDF

FGF8, isoform b, is an etiological factor in prostate tumorigenesis

PDF

A review of molecular conjugates and their use in gene therapy with the presentation of a model experiment: Gene therapy with novel fusion proteins that target breast cancer cells

PDF

Development of novel adenovirus-retroelement vectors and their applications beyond gene therapy

PDF

Hippocampal response to injury

PDF

Development of HA -pseudotyped retroviral vectors for cell -specific gene delivery

PDF

Activin-mediated growth inhibition in prostate cancer LNCaP cells by transcriptional regulation of apoptosis-related genes

PDF

Changes in growth factor and immune cell distributions in the lacrimal gland of pregnant rabbits following corneal injury

PDF

Identification and cloning of developmentally regulated genetic loci in transgenic mice by screening for novel expression pattern of the lacZ reporter gene

PDF

A dynamic apical actin cytoskeleton facilitates exocytosis of tear proteins in rabbit lacrimal acinar epithelial cells

PDF

Involvement of the secretory pathway for AMPA receptors in the expression of LTP

PDF

Fluid balance and neuropeptides in the lateral hypothalamic area

PDF

Signaling cascades: a functional characterization of cone arrestin and a differential gene expression analysis of developing retinal ganglion cells

PDF

Expression of matrix metalloproteinases and their inhibitors in the muscles of amyotrophic lateral sclerosis and control patients

PDF

Amelogenin domains in a self-assembly process

PDF

Morphological and functional evaluation of hESC-RPE cell transplantation in RCS rats

PDF

Neurotrophic actions of TGFɑ, NT4/5 and bFGF in the developing and adult intact and lesioned rat nigrostriatal system

PDF

Essential role of the carboxyl-terminus for proper rhodopsin trafficking and "enlightenment" to the pathway(s) causing retinal degeneration in a mouse model expressing a truncated rhodopsin mutant

PDF

Identification of oncogenes cooperating in murine mammary tumorigenesis and transgenic mouse models of breast cancer

Asset Metadata

Creator Roehmholdt, Brian Francis (author)

Core Title Gene therapy for motor neuron degeneration in murine tissue culture and transgenic mouse models of familial amyotrophic lateral sclerosis

School Graduate School

Degree Doctor of Philosophy

Degree Program Cell and Neurobiology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, neuroscience,health sciences, pathology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Garner, Judy A. (committee chair), Weiner, Leslie P. (committee chair), Fulle, Hans-Jurgen (committee member), Kasahara, Noriyuki (committee member), McNeill, Thomas (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-230364

Unique identifier UC11334727

Identifier 3073841.pdf (filename),usctheses-c16-230364 (legacy record id)

Legacy Identifier 3073841.pdf

Dmrecord 230364

Document Type Dissertation

Rights Roehmholdt, Brian Francis

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA