USF Libraries
USF Digital Collections

Molecular intervention in mouse models of amyotrophic lateral sclerosis and alzheimer's disease - neuropathology and behavior

MISSING IMAGE

Material Information

Title:
Molecular intervention in mouse models of amyotrophic lateral sclerosis and alzheimer's disease - neuropathology and behavior
Physical Description:
Book
Language:
English
Creator:
Bennett, Steven
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Neurodegeneration
Intracranial
Intrathecal
Inflammation
And microglia
Dissertations, Academic -- Medical Sciences -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Neurodegeneration describes the progressive loss of structure and function of neurons, leading ultimately to cell and organism death. Although the initiating factors of neurodegenerative diseases such as Alzheimer's, Parkinson's, Huntington's, and Amyotrophic Lateral Sclerosis may be different, they share common pathophysiologies. Proteinopathies, as these diseases are now termed, are characterized by atypical deposits of proteins, often due to misfolding. Associated with these deposits are dysfunctional mitochondria, oxidative stress, disrupted axonal transport, inflammation, and apoptotic cell death. If this occurs in motor neurons, as in ALS, ataxia precedes death with little or no change in cognition. On the other hand, if the deposits are found in cortical neurons, as in Alzheimer's disease, the outcome is dementia and motor function remains largely intact. Each disease is selective for particular types of neurons and brain regions. Although research has elucidated much of the molecular biology involved in these diseases, their initiating causes remain largely unknown. Most of our current understanding originated with the identification of gene mutations that cause rare familial forms of these diseases. As a result, numerous strains of transgenic animals have been developed to study neurodegenerative disease phenomena and were central to the studies presented in this body of work. Novel routes of drug and gene delivery are described here as well as characterization of the mouse models studied. In particular, this work demonstrates that the blood brain barrier is disrupted in ALS followed by the formation of autorosettes in ALS mice. In various Alzheimer's disease mouse models, it was demonstrated that the acute phase reactant alpha-1-antichymotrypsin (ACT) not only interacts with amyloid plaques, but also induces tau phosphorylation in vivo; tying together these disease hallmarks. It was also shown that small fragments of Aβ (1-11) could disrupt the formation of mature amyloid plaques in these mice. Lastly, it was demonstrated that mature plaques could also be decreased by intracranial delivery of granulocyte-macrophage stimulating factor (GM-CSF). My dissertation research goal was to understand and develop these treatment strategies based on protein disaggregation, neuroprotection, and inflammation, meanwhile developing novel methods for targeted delivery of molecules into the CNS of mice.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Steven Bennett.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0003178
usfldc handle - e14.3178
System ID:
SFS0027494:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003178
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Bennett, Steven.
0 245
Molecular intervention in mouse models of amyotrophic lateral sclerosis and alzheimer's disease neuropathology and behavior
h [electronic resource] /
by Steven Bennett.
260
[Tampa, Fla] :
b University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: Neurodegeneration describes the progressive loss of structure and function of neurons, leading ultimately to cell and organism death. Although the initiating factors of neurodegenerative diseases such as Alzheimer's, Parkinson's, Huntington's, and Amyotrophic Lateral Sclerosis may be different, they share common pathophysiologies. Proteinopathies, as these diseases are now termed, are characterized by atypical deposits of proteins, often due to misfolding. Associated with these deposits are dysfunctional mitochondria, oxidative stress, disrupted axonal transport, inflammation, and apoptotic cell death. If this occurs in motor neurons, as in ALS, ataxia precedes death with little or no change in cognition. On the other hand, if the deposits are found in cortical neurons, as in Alzheimer's disease, the outcome is dementia and motor function remains largely intact. Each disease is selective for particular types of neurons and brain regions. Although research has elucidated much of the molecular biology involved in these diseases, their initiating causes remain largely unknown. Most of our current understanding originated with the identification of gene mutations that cause rare familial forms of these diseases. As a result, numerous strains of transgenic animals have been developed to study neurodegenerative disease phenomena and were central to the studies presented in this body of work. Novel routes of drug and gene delivery are described here as well as characterization of the mouse models studied. In particular, this work demonstrates that the blood brain barrier is disrupted in ALS followed by the formation of autorosettes in ALS mice. In various Alzheimer's disease mouse models, it was demonstrated that the acute phase reactant alpha-1-antichymotrypsin (ACT) not only interacts with amyloid plaques, but also induces tau phosphorylation in vivo; tying together these disease hallmarks. It was also shown that small fragments of Aβ (1-11) could disrupt the formation of mature amyloid plaques in these mice. Lastly, it was demonstrated that mature plaques could also be decreased by intracranial delivery of granulocyte-macrophage stimulating factor (GM-CSF). My dissertation research goal was to understand and develop these treatment strategies based on protein disaggregation, neuroprotection, and inflammation, meanwhile developing novel methods for targeted delivery of molecules into the CNS of mice.
590
Advisor: Huntington Potter, Ph.D.
653
Neurodegeneration
Intracranial
Intrathecal
Inflammation
And microglia
690
Dissertations, Academic
z USF
x Medical Sciences
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3178



PAGE 1

Molecular Intervention in Mouse Models of Amyotrophi c Lateral Sclerosis and Alzheimers Disease Neuropathology and Behavior by Steven Prescott Bennett A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Molecular Medicine College of Medicine University of South Florida Major Professor: Hunt ington Potter, Ph.D. Samuel Saporta, Ph.D. Jun Tan, M.D., Ph.D. Ronald Keller, Ph.D. Craig Doupnik, Ph.D. Eric Bennett, Ph.D. Peter Neame, Ph.D. Date of Approval: October 14, 2009 Keywords: neurodegeneration, intracra nial, intrathecal, inflammation, and microglia Copyright 2009, Steven Prescott Bennett

PAGE 2

ACKNOWLEDGEMENTS Thanks are due first to my major pr ofessor, Dr. Huntington Potter, for his unrelenting support, keen insight, and dilig ent mentorship. My sincere thanks go to other committee members; Dr. Samuel Saporta, Dr. Jun Tan, and Dr. Ronald K. Keller, Dr. Craig Doupnik, Dr. Eric Benne tt, and Dr. Peter Neame. I would further like to thank my wife, Laur a, for her continuous and overwhelming support during this process. In addition, I would like to thank my family for their love and dedication to my education. Lastly, I would like to recognize those individuals who have directly or indirectly made this Ph.D. possible, especially: Dr. Inge Wefes, Dr. Jaya Padmanabhan, Timothy Boyd, Tina Fiorelli, and Michelle Norden.

PAGE 3

i TABLE OF CONTENTS List of Tables vi List of Figures vii List of Abbreviations x Abstract xiii Chapter 1 Introduction 1 1.1. Introduction to neurodegenerative disease 1 1.2. Amyotrophic Lateral Sclerosis 3 1.2.1. Amyotrophic Lateral Sclerosis 3 1.2.2. Major Genes and Pr oteins in ALS 4 1.2.3. Genes and Major Mouse Models of ALS 5 1.2.3.1. SOD1 5 1.2.3.2. ALS2 6 1.2.3.3. TDP-43 7 1.2.4. ALS and excitotoxicity 8 1.2.5. ALS and the cytoskeleton 8 1.2.6. ALS and intracellul ar inclusions 9 1.2.7. ALS and inflammation 10 1.2.8. ALS and cellular respiration 11 1.2.9. ALS and axonal transport 12 1.2.10. ALS and the vascular system 13 1.2.11. ALS and the blood brain barrier 13 1.2.12. Environmental components of ALS 14 1.2.13. ALS and neuronal death 15 1.3. Alzheimers Disease 15 1.3.1. Alzheimers Pathology 15 1.3.2. Major Genes and Proteins in AD 16 1.3.3. Extracellular Aggregates in AD 18 1.3.4. AD and the cytoskeleton (tau) 20 1.3.5. Major Mouse Models of AD 21 1.3.6. AD and inflammation 22 1.3.7. GM-CSF and Alzheimers disease 23 1.3.8. Environmental components of AD 24

PAGE 4

ii 1.3.9. AD and the vascular system 24 1.3.10. AD and oxidative stress 25 1.3.11. Integration of many pathways 25 1.3.12. AD and neuronal death 26 1.4. Research RationaleAmyotrophic Lateral Sclerosis 26 1.4.1. Transmission and induction of ALS 27 1.4.2. Disruption of the blood brain barrier 28 1.4.3. Neuroprotective effects of DTG (1,3-di-o-tolylguanidine) 28 1.4.4. Novel therapeutic laminin derivative, KDI 29 1.5. Research RationaleAlzheimers Disease 29 1.5.1. Development of a bilatera l infusion system in mice 29 1.5.2. Intracranial infusion of KDI, a glutamate receptor antagonist and electroporation of a KDI expressing plasmid 30 1.5.3. Alzheimers Disease and -1 antichymotrypsin 31 1.5.3.1. ACT and tauopathies 31 1.5.3.2. Inhibition of the ACT-A interaction 31 1.5.4. Colony stimulating factors and amyloid clearance and cognitive enhancement 32 Chapter 2 Transmissibility of Misfolded Superoxide Dismutase Inclusions in a Mouse Model of ALS 33 2.1. Introduction 33 2.2. Materials and Methods 34 2.2.1. Intrathecal Catheter Construction 34 2.2.2. Spinal Cord Extracts 35 2.2.3. Osmotic Minipumps 36 2.2.4. Mice 36 2.2.5. Intrathecal Infusions 36 2.2.6. Histology and Immunohistochemistry 37 2.2.7. Behavior 38 2.3. Results and Conclusions 38 2.4. Figures 40 Chapter 3 Evidence of Compromised Bl ood-Spinal Cord Barrier in Early and Late Symptomatic SOD1 Mice Modeling ALS 45 3.1. Abstract 45 3.2. Introduction 46 3.3. Results 48 3.3.1. Disease symptom progression in G93A mice 48 3.3.2. Motor neurons in the cervical and lumbar spinal cords of G93A mice 49 3.3.3. Fluorescent detection of Evans Blue in the cervical and lumbar spinal cords of G93A mice 50

PAGE 5

iii 3.3.4. Immunochistochemical char acteristics of basement membranes, endothelial cells, a nd astrocytes in the spinal cords of G93A mice 50 3.4. Discussion 52 3.5. Materials and Methods 56 3.5.1. Animals 56 3.5.2. Characteristics of disease progression 57 3.5.3. Evans Blue dye 57 3.5.4. Euthanatasia and tissue preparation 58 3.5.5. Immunofluorescence staining 58 3.5.6. Staining of motor neurons in the spinal cord 59 3.5.7. Statistical analysis 59 3.6. Figures 60 3.7. References 80 Chapter 4 The Neuroprotective Effect s of the Sigma Receptor Agonist DTG (1,3-Di-O-Tolylguanidine) in a Mouse Model of ALS 86 4.1. Introduction 86 4.2. Materials and Methods 88 4.2.1. Osmotic Minipumps and subcutaneous implantation 88 4.2.2. Mice 88 4.2.3. Histology and Immunohistochemistry 89 4.2.4. Behavior 89 4.3. Results and Conclusions 90 4.4. Figures 91 Chapter 5 The Neuroprotective Effects of the Carboxyl Terminus of Gamma Laminin, KDI, on a Mouse Model of ALS 94 5.1. Introduction 94 5.2. Materials and Methods 96 5.2.1. Intrathecal Catheter Assembly 96 5.2.2. Osmotic Minipumps 96 5.2.3. Mice 97 5.2.4. Intrathecal Infusions 97 5.2.5. Histology and Immunohistochemistry 98 5.2.6. Behavior 98 5.3. Results and Conclusion 99 5.4. Figures 101 Chapter 6 A Novel Technique for Simulta neous Bilateral Brain Infusions in a Mouse Model of Neurodegenerative Disease 106 6.1. Abstract 106 6.2. Introduction 107 6.3. Materials and Methods 109 6.3.1. Catheter Construction 109

PAGE 6

iv 6.3.2. Transgenic Mice 110 6.3.3. Materials 110 6.3.4. Intracranial Infusions 110 6.3.5. Preparation of Brain Extracts and Western Blot Analysis 112 6.3.6. Histology and Immunohistochemistry 112 6.4. Results 113 6.4.1. Animal Recovery 113 6.4.2. Bilateral Brain Infusion 114 6.4.3. Interaction of ACT and amyloid 115 6.5. Discussion 116 6.6. References 120 6.7. Figures 122 Chapter 7 The Neuroprotective Effects of the Carboxyl Terminus of Gamma Laminin, KDI, on a Mouse Model of Alzheimers Disease 131 7.1. Introduction 131 7.2. Materials and Methods 133 7.2.1. Catheter Construction 133 7.2.2. Osmotic Minipumps 134 7.2.3. Transgenic Mice 134 7.2.4. Intracranial Infusions 134 7.2.5. Electroporation 135 7.2.6. Behavior 136 7.2.7. Preparation of Brain Extracts and Western Blot Analysis 137 7.2.8. Histology and Immunohistochemistry 137 7.3. Results and Conclusions 138 7.4. Figures 140 Chapter 8 ACT Induces the Phosphoryla tion of Tau in Mouse Models of AD 144 8.1. Introduction 144 8.2. Materials and Methods 145 8.2.1. Catheter Construction 145 8.2.2. Transgenic Mice 146 8.2.3. Materials 147 8.2.4. Intracranial Infusions 147 8.2.5. Preparation of Brain Extracts and Western Blot Analysis 148 8.2.6. Histology and Immunohistochemistry 149 8.3. Results and Conclusions 149 8.4. Figures 152 Chapter 9 The Effects of Chronic In tracerebral Infusion of ACT on Amyloid Plaques: Iinhibition with A Related Peptides 158 9.1. Introduction 158 9.2. Materials and Methods 160 9.2.1. Catheter Construction 160

PAGE 7

v 9.2.2. Transgenic Mice 161 9.2.3. Materials 161 9.2.4. Intracranial Infusions 162 9.2.5. Histology and Immunohistochemistry 163 9.3. Results and Conclusions 163 9.4. Figures 166 Chapter 10 Arthritis-Induced Hematopoiesis Repairs Pathology and Memory in Alzheimers Mice 171 10.1. Abstract 171 10.2. Introduction 172 10.3. Materials and Methods 175 10.3.1. Transgenic Mouse Studi es Involving Intracerebral Administration of CSFs 175 10.3.2. Immunohistochemistry and Image Analysis of Intrahippocampal-injected Mice 176 10.3.3. Behavioral Transgenic Mouse Study Involving GM-CSF Treatment 176 10.3.4. Immunohistochemistry and Image Analysis of Subcutaneous GM-CSF-treated Mice 177 10.4. Results 178 10.4.1. Intracerebral administration of CSFs 178 10.4.2. Daily subcutaneous injection of GM-CSF 179 10.5. Discussion 181 10.6. References 183 10.7. Figures 187 Chapter 11Discussion 197 References 207 Appendices 234 Appendix ALymphopenia and Spontaneous Autorosette Formation in an SOD1 Model of ALS 235 Appendix BArthritis-Induced He matopoiesis Repairs Pathology and Memory in Alzheimers MiceSupplementary Figures 248 About the Author End Page

PAGE 8

vi LIST OF TABLES Table 1 SOD1 aggregate infusion into ALS micesurvival data 44 Table 2 DTG infusion into ALS mi ce-weight and survival data 91 Table 3 KDI infusion into ALS mice-survival data 101 Table 4 Complete blood cell count and white blood cell differential in SOD1 G93A mice as compared to control mice 246

PAGE 9

vii LIST OF FIGURES Figure 1Mouse intrathecal catheter assembly 40 Figure 2Western blot of high mol ecular weight aggregates from 130day SOD1 G93A mice prior to infusion. 41 Figure 3Rodent rotarod behavioral apparatus 42 Figure 4Immunohistochemical comparison of SOD1 G93A and nontransgenic spinal cords 43 Figure 5Characteristics of di sease progression in G93A mice 60 Figure 6Motor neurons in the cervical spinal cord of G93A mice at early and late stage of dis ease (cresyl violet staining) 62 Figure 7Motor neurons in the lumb ar spinal cord of G93A mice at early and late stage of dis ease (cresyl violet staining). 64 Figure 8Evans Blue fluorescence in th e cervical spinal cord of G93A mice at early and late stages of disease 66 Figure 9Evans Blue fluorescence in the lumbar spinal cord of G93A mice at early and late stages of disease 68 Figure 10Immunofluores cence staining for lamini n in the cervical spinal cord of G93A mice at ea rly and late stages of disease 70 Figure 11Immunofluores cence staining for laminin in the lumbar spinal cord of G93A mice at ea rly and late stages of disease 72 Figure 12Immunofluorescen ce staining for Glut-1 in the cervical and lumbar spinal cords of G93A mi ce at early and late stages of disease 74 Figure 13Immunohistochemical staini ng for endothelial cells (CD146) and astrocytes (GFAP) in the ce rvical spinal cord of G93A mice at early and late stages of disease 76

PAGE 10

viii Figure 14Immunohistochemical staini ng for endothelial cells (CD146) and astrocytes (GFAP) in the lumbar spinal cord of G93A mice at early and late stages of disease 78 Figure 15Lifespan of DTG and sh am treated SOD1 G93A mice 92 Figure 16Fluoro-jade stai ning of late stage SOD1 G93A spinal tracts 93 Figure 17Promotion of neurite outgrowth in presence of KDI 102 Figure 18Stereo microscopic images of laminectomized rats with and without KDI 103 Figure 19Hematoxylin eosin stained longitudinal sections of the rat spinal cord at site of injury 104 Figure 20Thioflavis S staining on br ain sections from 6 PS/APP mice demonstrating plaque variability 122 Figure 21Mouse intracranial catheter assembly 124 Figure 22Intracranial catheters in use with osmotic minipumps 125 Figure 23Montage of unilatera lly ACT infused PS/APP mouse 126 Figure 24ACT immunostaining on PS/APP and nontransgenic mice 128 Figure 25Western blot of PS/APP mice receiving ACT infusions into the lateral ventricles 129 Figure 26Immunohistochemical comparison of PS/APP/ACT transgenic mouse plaques and ACT infused PS/APP mouse 130 Figure 27Electroporation of EGFP expressing plasmid into the mouse brain 140 Figure 28Schematic of radial arm water maze 141 Figure 29Preliminary latency plots of KDI treated and untreated mice in the RAWM 142 Figure 30Preliminary error plots of KDI treated and untreated mice in the RAWM 143 Figure 31ACT infusion increases PHF-1 in htau mice 152 Figure 32Montage of htau mouse brain receiving ACT unilaterally 153

PAGE 11

ix Figure 33Western Blot of ACT infused htau mice showing increases in Pser202 154 Figure 34ACT infusion into htau mice leads to dephosphorylation of GSK-3 155 Figure 35ACT infusion into htau mice leads to dephosphorylation of ERK 156 Figure 36ACT infusion into P DAPP mice induces phosphorylation of Thr231 157 Figure 376E10 staining of PS/APP mouse brains following A 1-11 infusion 166 Figure 384G8 staining of PS/ APP mouse brains following A 1-11 infusion 167 Figure 39Semiquantitative analys is of 6E10 and 4G8 staining in PS/APP mice following A 1-11 infusion 168 Figure 40Iba1 microglial staining comparison from A 1-11 infused hemispheres to sham treated contralateral hemisphere 169 Figure 41Intrahippocampal in jection of GM-CSFand aCSF 187 Figure 42Behavioral analysis following daily subcutaneous GM-CSF injections 189 Figure 43Amyloid deposition in subc utaneous GM-CSF-injected mice 191 Figure 44Microglial immunostain ing in subcutaneous GM-CSFinjected mice 193 Figure 45Synaptophysin immunost aining in subcutaneous GM-CSFinjected mice 195 Figure 46Morphological analysis of peripheral blood cells (Giemsa staining). Spontaneous autorose tte-formations were found in G93A mice 247

PAGE 12

x LIST OF ABBREVIATIONS AD Alzheimers disease A Amyloid beta ALS Amyotrophic lateral sclerosis CAA Cerebral amyloid angiopathy CNS Central nervous system CSF Cerebrospinal fluid aCSF Artificial cerebrospinal fluid APP Amyloid precursor protein IL Interleukin TNFTumor necrosis factor alpha IFN Interferon NSAID Non-steroidal anti-inflammatory drugs CC Cingulate cortex H Hippocampus EC Entorhinal cortex GFAP Glial fibr ial acidic protein ELISA Enzyme-linked immunosorbent assay IgG Immunoglobulin G

PAGE 13

xi TGF Tumor growth factor BBB Blood-brain barrier i.c.v intracerebrovenricular i.p. Interperitoneal MN Motor neuron GM-CSF Granulocyte Macrophage colony stimulating factor SOD1 Superoxide dismutase 1 ApoE ApolipoproteinE ACT -1 antichymotrypsin Serpin Serine protease inhibitor RAWM Radial arm water maze PS1 Presenilin 1 NFT Neurofibrillary tangles PHF Paired helical filaments FTD Frontal temporal dementia CBD Cortical basal degeneration PSP pProgressive supranuclear palsy pSer Phosphorylated serine pThr Phosphorylated threonine MAPT Microtubule asso ciated protein tau DTG 1,3-di-o-tolylguanidine ERK Extracellular related kinase GSK-3 Glycogen synthase kinase 3

PAGE 14

xii MCAO Middle cerebral arterial occlusion NO Nitric oxide LPS Lipopolysaccharide PrPc Normal prion protein PrPsc Scrapie or mi sfolded prion protein TSE Transmissible spongiform encephalopathy WT Wild type KDI Lysine-aspartic acid-isoleucine NMDA N-methyl-D-aspartic acid AMPA -amino-3-hydroxyl-5-methyl-4 -isoxazole-propionate LTP Long term potentiation EGFP Enhanced green fluorescent protein TDP-43 TARP DNA bi nding protein 43 kDa DOB Date of birth DOD Date of death NTG Nontransgenic LS Lifespan C3 Complement C3 convertase

PAGE 15

xiii MOLECULAR INTERVENTION IN MOUSE MODELS OF AMYOTROPHIC LATERAL SCLEROSIS AND ALZHEIMERS DISEASE NEUROPATHOLOGY AND BEHAVIOR Steven Prescott Bennett ABSTRACT Neurodegeneration describes the progressive loss of structure and function of neurons, leading ultimately to cell and organi sm death. Although the initiating factors of neurodegenerative diseases such as Alzh eimers, Parkinsons, Huntingtons, and Amyotrophic Lateral Sclerosis may be differe nt, they share common pathophysiologies. Proteinopathies, as these diseases are now te rmed, are characterized by atypical deposits of proteins, often due to misfolding. Asso ciated with these deposits are dysfunctional mitochondria, oxidative stress disrupted axonal transport, inflammation, and apoptotic cell death. If this occurs in motor neurons, as in ALS, ataxia precedes death with little or no change in cognition. On the other hand, if the deposits are found in cortical neurons, as in Alzheimers disease, the outcome is dementia and motor function remains largely intact. Each disease is selective for part icular types of neurons and brain regions. Although research has elucidated much of the molecular biology involved in these diseases, their initiating cau ses remain largely unknown. Most of our current understanding origin ated with the identification of gene mutations that cause rare familial forms of these diseases. As a result, numerous strains of transgenic animals have been deve loped to study neurodegenerative disease phenomena and were central to the studies pres ented in this body of work. Novel routes of drug and gene delivery are described here as well as characterization of the mouse models studied. In particular this work demonstrates th at the blood brain barrier is

PAGE 16

xiv disrupted in ALS followed by the formation of autorosettes in ALS mice. In various Alzheimers disease mouse models, it was demonstrated that the acute phase reactant alpha-1-antichymotrypsin (ACT) not only inte racts with amyloid plaques, but also induces tau phosphorylation in vivo ; tying together these disease hallmarks. It was also shown that small fragments of A (1-11) could disrupt the fo rmation of mature amyloid plaques in these mice. Lastly, it was demons trated that mature plaques could also be decreased by intracranial delivery of gra nulocyte-macrophage stimulating factor (GMCSF). My dissertation research goal was to understand and develop these treatment strategies based on protein disaggrega tion, neuroprotection, and inflammation, meanwhile developing novel methods for targeted delivery of molecule s into the CNS of mice.

PAGE 17

1 CHAPTER 1 INTRODUCTION 1.1. Introduction to neurodegenerative disease While all diseases, especia lly those that are fatal, ar e devastating, there is an additional sense of loss when those diseases a ffect our minds often considered the core that makes us human. Although the exact pathophysiology of different neurodegenerative diseases varies, they share certain comm onalities, and all involve selective vulnerability of neurons in distin ct brain regions. For example, there are properties of motor neurons that make them selectively vulnerabl e to cell death in amyotrophic lateral sclerosis, and properties of cortical and hippocampal neurons that make these neurons susceptible to Alzheimers disease. In either case, the outcome is progressive neuronal death that is terminal to the patient. Collectively, neurodegenerative diseases ar e a leading cause of death in the U.S. There are currently over 5 million Alzheimers cases, 1.5 million Parkinsons, 30,000 ALS, and 30,000 Huntingtons patients. In addition 150,000 people are annually diagnosed as at risk patients and about 300 cases of prion disease are counted annually (Centers of Disease Control and Prevention). Despite decades of research dedicated to each of these diseases, there are no cures to date. Indeed, those drugs currently prescribed provide modest, if any, alleviati on of symptoms and may delay death for just weeks or months. Therefore, novel therapies are urgently needed and strongly sought, some with the intent to be effective for multiple neurodegenerative disorders. The most common molecular mechanism of neurodegenerative diseases, and a primary focus of my work, is th e misfolding of proteins that leads to extracellular and/or intracellular proteinaceous aggregates. T ypically, these inclusions/plaques include molecular chaperones and polyubiquitinated proteins, indicating that the cellular

PAGE 18

2 mechanisms for protein folding, quality cont rol and protein degradation are overwhelmed in protein misfolding diseases (PMDs)/proteinopathies (rev iewed in Soto and Estrada, 2008). Another component of neurodegenerative diseases is inflammation, which is both a response to aggregates and contributes to their formation. The ac tivation of microglia and astrocytes and the subsequent release of various cytokines and chemokines induce a local, chronic state of inflammation in the affected brain regions (McGeer and McGeer, 1995). The involvement of astrocytes tie s inflammation to another aspect of neurodegeneration: glutamate excitotoxicity (Abraham et al., 2001). Astrocytes are known to surround glutaminergic synapses where they express GLT-1 (EAAT2) glutamate transport molecules which function to rapidly clear this neurotransmitter. Failure to adequately clear glutamate can lead to abnormally elevated glutamate levels at the synapse, in turn causing surges of calci um into post-synaptic neurons (reviewed in Amara and Fontana 2002). Although, Ca2+ entry partly underlies the process of synaptic plasticity, disruption of ca lcium homeostasis by excess Ca2+ has a pronounced deleterious impact on neurons and surrounding glia. Ca lcium reacts within the cell, altering enzymatic activities, inducing damage to DNA and organelles, and negatively affecting axonal transport and cytoskeletal orga nization (Gennarelli et al., 1993). Excess intracellular calcium is also one of several conditions that lead to dysregulation of mitochondria. Mitochondria l dysfunction is problematic because it burdens the cell with reactive oxygen species (ROS) and induces oxidative stress. Moreover, an intrinsic apoptotic pathway can be activated by the rel ease of cytochrome c from the mitochondrial inner membrane, l eading to programmed cell death (PCD) or apoptosis (reviewed by Onyango et al., 2006). Indeed, neuronal apoptosis is a common, final event in virtually all neurodegenerative diseases. While all neurodegenerative diseases ha ve a genetic component, only 5-10% of patients represent cases where the genetic mu tation leading to the autosomal dominantly inherited familial forms of both AD a nd ALS are known. Additional genetic and epigenetic components as well as environmental contributions still have to be determined. It is for this reason that most of the knowledge acquired about ALS and AD comes from the finding and analysis of genetic mutations throughout human populations. Because

PAGE 19

3 the phenotypes of familial and sporadic cases are so similar for each disease, those mutations have been the most useful m eans to study disease progression and develop possible therapeutic interventions for both fam ilial and sporadic forms of the diseases. Protective as well as risk enhancing environm ental or life style factors for each disease have also been described. For example, educational/environmental enrichment and developing Alzheimers disease seem to be nega tively correlated, i.e. these factors have a protective effect, while in ALS a positive corr elation, a disease causing/enhancing impact of physical exertion as has been reported for example with soldiers and athletes (Belli and Vanacore, 2005; Haley, 2003). Once mutations in the disease-related ge nes were identified and the genes were cloned, in vitro and in vivo models were developed to test various hypotheses on disease progression and therapy. Cell culture and or ganotypic tissue culture have proven useful in answering very targeted scientific questions. Howeve r, for the study of protein aggregation in CNS tissues, a more systemic approach with animal studies was needed. Accordingly, several different lines of transgenic mice have been developed over the past 15 years to analyze the formation of protein a ggregates in vivo. Such mouse models of AD and ALS were the main subjects of my work which investigated causes and treatments for Protein Misfol ding Diseases in the brain. 1.2. Amyotrophic Lateral Sclerosis 1.2.1. ALS Pathology Amyotrophic Lateral Sclerosis (ALS) is currently the most common motor neuron disease, affecting roughly 30,000 pe ople in the U.S. with a lif etime risk of 1 in 2000. Originally called Charcots disease after th e French neurobiologist and physician JeanMartin Charot, who first char acterized the disease in 1869, it has more recently been named Lou Gehrigs disease after the famous Yankees baseball player who was diagnosed with ALS in 1939 and died 2 years la ter at the age of 37. ALS is characterized by disruption of the cytoskeleton in motor neurons (Williamson et al., 1998), reactive

PAGE 20

4 gliosis (Leigh and Swash, 1991), accumulation of intracellular protei naceous inclusions (Bruijn et al., 1998), neurofilament abnor malities and formation of axonal spheroids (Carpenter, 1968; Gonatas et al., 1992; Hirano et al., 1984; Leigh et al., 1991), loss of myelinated fibers in the ventral roots of co rticospinal tracts (Delisle and Carpenter, 1984), and disruption of the blood brain barrier (Garbuzova-Davis et al., 2007c). Selective death of motor neurons occurs in th e motor cortex, brainstem, and spinal cords of affected individuals (R owland and Shneider, 2001b). ALS is a devastating disease, causing spasticity, hyperrefl exia, generalized weakness, muscle atrophy and paralysis (Mulde r et al., 1986), all lead ing to death within 2-5 years of onset, most commonly from respir atory failure. ALS primarily targets the large caliber alpha motor neurons in the anterior horn of the spinal cord, 35-100 m in diameter (Ravits et al., 2007). The averag e age of diagnosis varies, but generally symptoms occur between 40-60 y ears of age (Hirtz et al., 2007). 90-95% of ALS cases are sporadic, that is, there are no known gene tic or environmental causes, despite the high prevalence of the disease in certain populations (Belli and Vanacore, 2005; Haley, 2003; Horner et al., 2003; Mulder and Kurland, 1987; Vanacore et al., 2006). Most knowledge about the disease has so far been gained from mutations in 6 differe nt genes that underlie ALS (Gros-Louis et al., 2006). These geneti c predispositions accounts for only 5-10% of all ALS cases, but because the phenotypes of familial and sporadic ALS are so similar, these mutations have been valuable tools in understanding the disease pathogenenesis. 1.2.2. Major Genes and Proteins in ALS The most common gene that harbors mu tations leading to familial ALS is superoxide dismutase (SOD1) (Rosen, 1993). SOD1 functions as a 32 kD homodimer and each subunit of 153 amino acids is encoded by 5 exons on chromosome 21q22 (Siddique et al., 1991). The pr imary function of SOD1 is to convert toxic superoxide radicals, produced by mitochondria duri ng oxidative phosphorylation, to molecular oxygen and hydrogen peroxide. Appropriately, the protein is found in the cytosol, nucleus, mitochondrial inner membrane, and per oxisomes of cells to exert its antioxidant

PAGE 21

5 effects (Bruijn et al., 2004b). The precise role of SOD1 in the progression of ALS is not completely known. However, the prevalence and number of mutations in the human gene encoding SOD1 strongly indicated a role fo r the protein in the disease progression. Although the protein is re latively small, there ha ve been over 150 mutations identified in hSOD1 since the first was identified over 15 years ago (Rosen, 1993). Collectively, mutations in SOD1 account fo r about 3% of all ALS cases (Andersen, 2006). SOD1 is a ubiquitously expressed Cu/Zn metalloenzyme with zinc providing structural stability and copper coordinating the alternating transfer of electrons to both oxidize and reduce superoxide (Beckman et al., 1993). Each SOD1 monomer contains one Cu/Zn active site. However, mutations wi thin this active site do not lead to any different ALS phenotype than mutations in othe r parts of the molecule (Borchelt et al., 1994; Bowling et al., 1995). Therefore it is likely not the activity of the enzyme itself that promotes the disease, but rather a stru ctural change in the molecule with strong downstream effects. This information, together with the fact that SOD1 knockout mice do not develop disease symptoms, supports the hyp othesis that the invol vement of SOD1 in the pathogenesis of ALS is a result of a toxic-gain -offunction (Shaw, 2005). This view leads to the conclusion that irrespective of the specific mu tation, mutant SOD1 has taken on a new function, independent of its enzymatic activity, and that this function is toxic to the cell. The inheritance pattern of SOD1 mutations, like most neurodegenerative diseases, is autosomal domina nt (Siddique et al., 1996) 1.2.3. Genes and Major Mouse Models of ALS 1.2.3.1. SOD1 Currently 12 mouse models of ALS have been developed that harbor some of the indentified mutant forms of human SOD 1 and are used to study ALS and potential therapeutics. The most intensely studied ge netic variations are three point mutations leading to the single amino acid substituti ons G93A (Gurney, 1994), G37R (Wong et al., 1995), G85R (Bruijn et al., 1997), all of which are used to study disease progression at

PAGE 22

6 different rates. The G93A mouse model, for example, begins showing symptoms as early as 10 weeks and the animals die at 13-16 weeks. The G85R mode l, on the other hand, begins showing symptoms at 5 months a nd the mice die by 8-9 months. There are advantages to all different SOD1 mice, becau se some studies require longer therapeutic windows and others aim fo r fast responses. The pathophysiology of SOD1 mutant mi ce is very similar to that of ALS patients, i.e. the primary pathological event is a progressi ve loss of both upper and lower motor neurons (Gurney, 1994) and mitochondri al vacuolization (Dalcanto and Gurney, 1995) leading to severe hind limb weakness. These events are accompanied by reactive microgliosis and astrocytosis and localized inflammation (Hall et al., 1998a), the formation of intracellular proteinaceous inclusions (Johnston et al., 2000), neurofilament positive inclusions (Tu et al., 1996), and fragmentation of the Golgi apparatus (Mourelatos et al., 1996). Other events are glutamate excitotoxicity (Rothstein et al., 1990), abnormal axonal transport (Williamson and Cleveland, 1999), and abnormal regulation of growth factors (Lambrechts et al., 2003). Importantl y, in all cases the pathological cascade culminates finally in the activation of caspase 3 (Li et al., 2000; Pasinelli et al., 2000; Vukosavic et al., 2000), and the induction of cell death. Although it is generally disputed if aggr egate formation is the cause, a byproduct, or even a protective mechanism in neurodegenerative disease, there is no doubt that the presence of inclusions in ALS and the downstream effects of mutant SOD1 on subcellular organelles as well as the activation of programmed cell death (PCD ) make SOD1 a very attractive molecule for the investigation of this disease. 1.2.3.2. ALS2 In another ALS related gene, ALS2 (als in) some disease related mutations have been identified. ALS2 is located on chromosome 2 and was recently shown to cause rare, juvenile onset forms of motor neuron di sease which progress slowly, and strongly resemble the symptoms of ALS (Hadano et al., 2001; Yang et al., 2001). The ALS2 encoded protein, alsin, is a 184-kD protein that has been shown to have 3 putative

PAGE 23

7 guanine-nucleotide-exchange f actor domains (GEFs). Alsin binds to the periphery of endosomes and is involved with the Ras supe rfamily of G protein receptors (Yamanaka et al., 2003). Specifically, GEFs catalyze the disso ciation of GDP from G-proteins and act, in vitro as an exchange factor for Rab5 (Otomo et al., 2003), a molecule known to be involved in endosomal trafficki ng. Alsin is enriched in ne rvous tissue and the mutations that have been identified so far are very unstable (Yamanaka et al., 2003), suggesting a loss of activity in one or more of the GEF domains. Recently, animal models harboring mutations in alsin (ALS2) have been generated for further investigation. 1.2.3.3. TDP-43 Another protein that has drawn some attention over th e past few years is TDP-43 (TAR DNA Binding protein 43) TDP-43 is a 43 kDa prot ein encoded by the TARDBP gene located on chromosome 1 and is expre ssed ubiquitously, incl uding the CNS, where it is found in neuronal an d glial nuclei and to a lesser exte nt in the cytoplasm (Geser et al., 2009). The diverse functions of TDP43 are not completely understood, but it is believed to stabilize low molecular weight neurofilaments (Strong et al., 2007), modulate cdk6 expression (Ayala et al., 2008a; Ayala et al., 2008b), act as a scaffold for nuclear bodies (Wang et al., 2004), and affect micr oRNA biogenesis (Gregor y et al., 2004), cell cycle, apoptosis (Ayala et al., 2008b), a nd mRNA transport and regulation of its translation at synapses (W ang et al., 2004). The protein, which can be abnormally phosphorylated and misfolded, was originally discovered in patients with Frontotemporal Dementia (FTD) and has recently been discove red in the inclusions of both familial and sporadic cases of ALS (Arai et al., 2006; Neumann et al ., 2006). Research results involving the role(s) of TDP43 in neurodegenerative disease have been somewhat conflicting, but its presence in the inclusions of ALS pa tients has made it a very appealing target for therapeutic interventions.

PAGE 24

8 1.2.4. ALS and excitotoxicity The most convincing link between sporad ic and familial forms of ALS involves glutamate excitotoxicity (Rothstein et al ., 1990). The neurotransmitter glutamate has been implicated in ALS because excess glutamate levels were discovered in the cerebrospinal fluid (CSF) of roughly half of the ALS patients sampled (Rothstein et al., 1991; Rothstein et al., 1990; Shaw et al., 1995 ). Excess glutamate induces repetitive firing of neurons leading to increases of intracellular calcium levels through increased permeability of calcium-permeable glutam ate receptors (Beers et al., 2001). A supporting phenomenon towards glutamate exce ss as a possible cause for the selective death of motor neurons in ALS is the fact that excess glutamate is cleared much less efficiently from synapses in sporadic and familial ALS cases and accordingly in most ALS cases, glutamate clearance is dysregulat ed. Typically, fingerlike projections from astrocytes rich in the glutamate transpor ter EAAT2 (GLT-1) surround synapses in order to rapidly clear glutamate (R othstein et al., 1996; Tanaka et al., 1997). This transport molecule, however, shows pronounced down-re gulation in ALS (in response to caspase 3), supporting glutamate excitotoxicity as a potential cause for ALS. Riluzole, a drug that functions to deactivate NMDA receptors and reduce synaptic glutamate levels, is currently the only effective drug for ALS that is FDA approved (Gur ney et al., 1998). Further studies revealed that glutamat e may also be responsible for other neurodegenerative diseases, namely Alzhei mers and Parkinson s disease (Beal, 1998; Hynd et al., 2004). 1.2.5. ALS and the cytoskeleton Most neurodegenerative proteinopathies, including both sporadic and familial ALS (Carpenter, 1968; Chou and Fakadej, 1971; Hirano, 1991; Hirano et al., 1984) show some disturbation of neurofilament content and assembly. Neurofilaments are neuronspecific intermediate filament s and are the most abundant structural proteins found in neurons. They are thought to add rigidity, tensile strength and intracellular guidance for

PAGE 25

9 transport along axons and dendr ites. Neurofilaments are cyto skeleton proteins that act together to form and maintain cell shape and facilitate the transport of various molecules and organelles within the cytoplasm of neur ons. Abnormal assembly of neurofilaments is of particular consequence in ALS as they are primarily responsibl e for establishing the proper diameter in axons (Lee and Clevela nd, 1996). Lower alpha motor axons are the thickest in the body and proper Neurofilament assembly are critical for adequate neurotransmission through these fibers. Abnormal assembly is prevalent in sporadic ALS cases and in most mouse models. Normally, Neurofilament assembly occurs between NF-L (light), NF-M (medium), and NF-H (hea vy) chains (in order of molecular weight68kDa, 160 kDA, and 200kDa, respectively) (H arris et al., 1991). The first event is dimerization of NF-L filaments or dimeriza tion of NF-L with either NF-M or NF-H filaments, followed by assembly of dimers in a staggered antiparallel fashion allowing NF-H and NF-M filaments to associate (Fuchs and Weber, 1994). Elongation occurs by head to tail assembly of NF-L filament s and phosphorylation occurs on the NF-M and NF-H filaments, the latter of which is one of the most extensively phosphorylated proteins in neurons (Strong et al., 2005b). In ALS, neurofilament assembly can be altered drastically due to reduced steady state le vels of NF-L in affected motor neurons. Moreover, mutant SOD1 has recently been shown to destabilize NF-L, providing a potential connection between familial ALS muta tions and alteration in NF ratios (Ge et al., 2005). 1.2.6. ALS and intracellular inclusions The formation of intracellular inclusions has so far been the most prominent feature of ALS, even though there is still no clear understa nding of their importance or contribution to the disease. In familial cases of ALS, the inclusions are immunopositive for SOD1. Otherwise, they are comprised ma inly of Neurofilaments (Liu et al., 2004), ubiquitin (Bruijn et al., 1998; Jonsson et al., 2004; Mather et al., 1993), peripherin (Robertson et al., 2001), dorfi n (Niwa et al., 2002), molecular chaperones (HSP 40 and HSP 90) (Namba et al., 1991; Shinder et al ., 2001; Takeuchi et al., 2002) and other

PAGE 26

10 proteins. These inclusions, similar to the amyloid deposits in AD and the -synuclein positive plaques found in Parkinsons (Shimura et al., 2001), have been subject of intensive studies to elucidat e the pathophysiolgy of this ne urodegenerative disease. While the cause for these plaques/inclusions is still unknown, much emphasis has been placed on investigating those deposits and agen ts that can exacerbate or ameliorate plaque growth as well as their relative e ffects on behavior and pa thology. Intracellular deposits in ALS could negativel y affect the cell in many ways They could contribute to the loss of protein function, thereby possibl y enhancing further deposition and causing other downstream effects within the cell, such as mitochondrial dysfunction (Dal Canto and Gurney, 1994; Kong and Xu, 1998; Wong et al., 1995). Another pot ential effect of protein aggregation might be the disruption of active sites and thereby enzymatic activity of the involved proteins. In ALS, altering binding regions for copper and zinc in the active site of SOD1 was found to cause protei n misfolding and aberrant activity regarding specific substrates (Wang et al., 2003). It has been proposed, for example, that mutant SOD1 interacts with other molecules, such as nitric oxide (Estevez et al., 1999), and forms toxic byproducts, such as peroxynitrate. Peroxynitrate can then react with tyrosine residues to form nitrotyrosine, and thereby block the activity of the nitrated protein (Beckman et al., 1993). Another consequence of protein aggregati on is the binding of various chaperones, thereby leading to the depletion of, for example, of HSP 40 and HSP70. A related observation describes evidence that in pr oteinopathies and inclusions/deposits the proteasome, the mayor protein degradation m achinery in the cell, is overwhelmed by an overload of ubiquitinated prot eins targeted for degrad ation (Leigh et al., 1991). 1.2.7. ALS and inflammation The activation of microglia and astr ocytes has led researchers to the understanding that ALS is not a cell autonomous disease (Cleme nt et al., 2003). i.e. the disease does not progress solely within neur ons, but is exacerbated by supporting glia in the brain. Astrocytes and microglia have also been shown to express mutant SOD1 and

PAGE 27

11 astrocytes even develop in clusions similar to those found in neurons. Activated microglia are known to release proinflammatory cytokines such as TNF NO, Cox-2, and IL-1 and IL-6 (Sargsyan et al., 2005), all of which are a part of a localized immune response. TNF can drive peripherin expression in neurons (found in inclusions) (Robertson et al., 2001), again providing a link between non-neuronal cells and ALS. Transforming growth factor beta 1 ( TGF -1) and Macrophage colony-stimulating factor (MCSF) are actually upregulat ed and released by microglia in presymptomatic mice (Elliott, 2001). Initial studi es involving administration of LPS into ALS mice, a known microglial activator, worsened the symptoms a nd shortened lifespan (W eydt et al., 2004). Studies have also shown that mutant SOD1 in microglia and astrocytes accelerates the release of these molecules and worsens th e disease, whereas normal supporting cells surrounding a mutant SOD1 expressing neuron greatly slowed disease progression, but had no change on the onset of the diseas e. Similarly, introducing mutant SOD1expressing microglia and astrocytes to norma l neurons did not induce motor neuron death suggesting that the supporting cells play a ro le in the disease, but are not the cause (Boillee et al., 2006a; Boillee et al., 2006b). So me current therapies are being used to target microglial activation. Minocycline, a broad spectrum antibiotic, has been proven to inhibit microglial activation in vivo (Van Den Bosch et al., 2002; Zhu et al., 2002) and its use for treating ALS has been pro posed for clinical investigation. 1.2.8. ALS and cellular respiration The role of mitochondria in the pathol ogy of ALS is not entirely clear, but the presence of vacuolated mitochondria in AL S patients and mouse models (Dalcanto and Gurney, 1995; Wong et al., 1995) and the concom itant interruption of cellular respiration in the presence of mutant SOD1 provides a strong indication that the organelle is involved in the disease process. Specifically, cellular respiration is reduced significantly from both decreased activity of cytoch rome oxidase, leading to decreased ATP production, and release of cytochrome c in response to oxidative stress and caspase activation (Jung et al., 2002). Moreover, mutant SOD1 has been shown to bind within

PAGE 28

12 the mitochondrial inner membrane space in spin al cord motor neurons (MNs) (Higgins et al., 2002). Lastly, partial deficiency in th e mitochondrial matrix manganese-containing SOD2 enzyme exacerbated ALS-like pathology in SOD1 mutant mice (Zimmerman et al., 2007). Current therapies for ALS target ed to mitochondria in clude the non-protein amino acid creatine, to enhance mitochondria l energy storage (Adhihe tty and Beal, 2008; Klivenyi et al., 1999), and the antibiotic minocycline. Minocyc line, a tetracycline derivative previously shown to inhibit micr oglial activation, has also been proven to inhibit the release of cytochrome c from mitochondria; a final step of programmed cell death (Kriz et al., 2002). Desp ite promising effects in mice, neither therapy has provided patients with alleviation of sy mptoms or extended lifespan. 1.2.9. ALS and axonal transport Another deficiency in ALS, impaired axonal transport, is common to neurodegenerative diseases. Given the as ymmetrical nature of motor neurons, an enormous metabolic load is placed on cell bod ies to transport molecules as far as 1 meter (fast and slow transport) to and from synapses (Bruijn et al., 2004b). The retrograde, dynein-mediated, transport of components such as multivesicular bodies and trophic factors (NGF, VEGF) back to the cell body is impaired by mutant SOD-1, leading to the conclusion that diminished transport corre lates well with the development of motor neuron disease (Williamson et al., 1998). Dynein has many roles, including positioning of the ER and Golgi, and the assembly of the mitotic spindle. Two point mutations have been identified in dynein that cause motor neuron diseas e (Hafezparast et al., 2003). Moreover, the disruption of the dynactin complex, an activator of cytoplasmic dynein, inhibits retrograde axonal tran sport and leads to a late-ons et progressive motor neuron disease (King and Schroer, 2000). A singl e point mutation in the p150 subunit of dynactin has similarly been linked to a lowe r motor neuron disorder (Puls et al., 2003).

PAGE 29

13 1.2.10. ALS and the vascular system Unexpected risk factors contributing to ALS have been revealed in growth and neurotrophic factors. Since astrocytes and microglia are such important sources of these factors, damage to supporti ng cells and interrup tion of the trophic support provided by them is a primary underlying cause of motor neurodegeneration. VEGF, a growth factor known to control the growth and permeability of blood vessels and to be protective of motor neurons, has been found to induce motor neuron disease following targeted deletion of the hypoxia-response element (Oosthuyse et al., 2001). Under normal conditions, hypoxia will induce VE GF expression under transcri ption factors that respond to the oxygen tension in order to maintain or restore vascular perf usion and trigger the growth of blood vessels. If this function is ablated in mice, however, classic ALS symptoms will ensue within 5 months of age (Lambrechts et al., 2003). Neurotrophic factors, on the othe r hand, though not shown to cause ALS pathology can ameliorate symptoms in mice. The most prominent of these factors are CNTF (Anand et al., 1995), GDNF (Elliott and Snider, 1996), and IGF-1 (Oppenheim, 1996), all of which have been used for re trograde AAV delivery from muscle to motor neurons, though the only positive effects seen on disease onset and survival were attributed to IGF-1 (Kaspar et al., 2003). 1.2.11. ALS and the blood brain barrier Recently, it was discovered that the blood brain barrier is compromised in ALS patients (Garbuzova-Davis et al., 2007c). Ind eed, impairment of the BBB occurs in most neurodegenerative diseases (Popescu et al ., 2009). Capillary endothelial cells are distributed along the length of the vesse ls, completely encircling the lumen and connected via adherents and tight junctions to form a diffusion barrier into the brain for most blood-borne substances (Ueno, 2007). Increasing evidence points to immune system involvement in ALS pathogenesis. In the spinal cord and brain of both ALS patients and animal models, the presence of T-cell lymphocytes (Alexianu et al., 2001;

PAGE 30

14 Engelhardt et al., 1993a), deposits of IgG (Donnenfeld et al., 1984a; Engelhardt et al., 2005b), complement components C3 and C4 (Engelhardt and Appel, 1990), and monocyte/macrophage and dendri tic cells (Troost et al., 1990 a) were observed that may have a critical role in motor neuron damage. IgG was detected in the perikarya of motor neurons of the lumbar spinal cord in mice after intraperitone al injection of IgG derived from sera of ALS patients. Moreover, th e uptake of IgG in multivesicular bodies in endothelial cells in the affect ed areas of the spinal cord was found in both ALS patients and mice injected with human ALS IgG. A dditionally, significantly increased levels of albumin, IgG, and C3c have been noted in the cerebrospinal fl uid of ALS patients (Annunziata and Volpi, 1985a; Leonardi et al., 1984a). These findings suggest that barrier permeability is affected in ALS. These phenomena were also identified in spinal cord microvessels of G93A mice at early ( 13 weeks of age) and late (17 weeks of age) stage disease, although more leakage, visible with Evans Blue dye, was found in lumbar spinal cords of mice at terminal st age disease. Additionally, basement membrane disruption was noted at both early and late st age disease, as shown by the loss of laminin staining in the G93A mice. Down regulation of Glut-1 and CD146 e xpressions in spinal cord endothelial cells was also found in G 93A mice at early and late stage disease, suggesting an altered endothelial lining leading to vascular leakage (Garbuzova-Davis et al., 2007c) 1.2.12. Environmental components of ALS Although unexplained, there have been e nvironmental links to acquiring ALS. First described by Mulder and Kurland in 1987, there are clusters of the disease in given populations that increase the chance of developing ALS. In the 1950s for example, there was a 50-fold increase in developing ALS on the island of Guam. More recently there has been a 2 fold increase in ALS patients returning from duty in the Gulf Wars (Haley, 2003; Horner et al., 2003). Additionally, it was reported that ALS was unusually common amongst Italian soccer player s (Belli and Vanacore, 2005; Vanacore et al., 2006). Factors such as length of time playing/fighti ng, intensity of exercise, and

PAGE 31

15 use of drugs or supplements were taken in to account, though nothing was conclusively identified as causing ALS. 1.2.13. ALS and neuronal death Apoptosis or PCD, is a common pathway for neuronal death in neurodegenerative disease and is the final event in ALS. As a presymptomatic event, Caspase 1 was shown to be activated in neurons (Pasinelli et al ., 2000; Vukosavic et al., 2000). Activation of caspase 1 is later followed by activation of caspase 3 in as trocytes and motor neurons (Guegan et al., 2001). Caspase 3 subsequently triggers the release of cytochrome C from the mitochondrial respiratory chain, which activates caspa se 9, known as intrinsic caspase activation. Caspase 9 may act as an e ffector for subsequent activation of caspase 7 and increased release/activ ation of caspase 3 (executio ner caspases), which are responsible for the degradation of cellular constituents and cell death. Alternatively, apoptosis can be stimulated extrinsically by extracellular queues activating caspase 8 followed by caspase 3 and then following the same path as the intrinsic pathway. Further evidence to support apoptosis in ALS was dete rmined by long-term, intrathecal delivery of the pan-caspase inhibitor (N-benzylocar bonyl-Val-Ala-Aspfl uoromethylketone) or zVAD-fmk, which significantly slowed the prog ression of the disease in G93A mice (Li et al., 2000). Further support for apop tosis in ALS was demonstrated by BCL-2 overexpression in this same model. BCL-2 (B-cell lymphoma 2) is a known antiapoptotic factor that slowed onset and promoted survival of G93A mice to a similar extent as the pan-caspase inhibi tor zVAD-fmk (Kostic et al., 1997). 1.3. Alzheimers Disease 1.3.1. Alzheimers Pathology Alzheimers disease is currently the most common form of neurodegeneration, affecting over 5 million people nationwide. Approximately 1 person in 8 will be

PAGE 32

16 diagnosed with AD by the age of 65, and 1 in 2 by the age of 85 (National Institutes of Health U.S. Department of Health and Human Services). The disease is characterized, initially, by loss of short term memory and attention span, although affected individuals function with relative normal capacity for up to 3 years following diagnosis. Typically, individuals die within 7-10 years of AD diagnosis, during which time they gradually withdraw from daily life. Slowly and ine xorably, the inability to remember names and follow conversations culminates to the inability to carry out familiar tasks, to reason, and to exercise judgment. It is a gradual, but severe cognitive decline, leading ultimately to dementia and death. The cause of death, in most cases, is AD directly, but some AD patients die of cancer, stroke, heart disease, and pneumonia as their general health deteriorates. Pathologically, AD is characterized by ne uronal loss with extr acellular deposits of amyloid and intracellular accumulation of ne urofibrillary tangles in the hippocampus, amygdala, and cerebral cortex (Selkoe, 1986). Curiously, these classical hallmarks of AD were already described in 1907 by Alois Alzheimer himself following an autopsy of a woman who had died with se vere dementia. It was not until 75 years later, however, that the biochemical tools were develope d to determine the components of these proteinaceous aggregates (Glennar and Wong 1984)( reviewed by Avila et al., 2004). Moreover, it has since been discove red that other pr ocesses common to neurodegeneration including ex citotoxicity, oxidative stress, inflammation, and apoptosis all have a role in the pathogenesis of AD. 1.3.2. Major Genes and Proteins in AD Several proteins have been identified within the amyloid plaques including A apolipoprotein E (apoE) (Wisniewsk i and Frangione, 1992) and Alpha 1antichymotrypsin (ACT) (Abraham et al., 1988), LRP (low densitity lipoprotein receptorrelated protein), activated compliment, and numerous cytokines. A is the primary component of the plaques and is a 38-43 am ino acid peptide proteo lytically derived from the amyloid precursor protein (APP) (G lennar and Wong, 1984)(Hardy and Selkoe,

PAGE 33

17 2002). APP is a transmembrane protein of unknown function expressed in almost all tissues. In mammals, there are three ma jor isoforms, APP695, APP751, and APP770, which vary by the presence or absence of the Kunitz protease inhibitor (KPI) domain and a chondroitin sulfate glycosaminoglycan (CS GA G) attachment site (Sisodia and Price, 1995). In the brain, primarily APP695 is expressed on the surface of neurons. APP is processed proteolytically by three enzymes, , and -secretases to generate either amyloidgenic or non-amy loidogenic peptides. During amyloidogenic processing, APP is cleaved first by -secretase, releasing a la rge N-terminal ectodomain, sAPP into the extracellular space. The re maining C-terminal fragment of 99 amino acids still bound is then cleaved by -secretase in the cell membrane to yield A at 39, 40, or 42 amino acids in length. -Secretase cleavage between Val637 (in the APP695 isoform) and Ile638 generates A 40, and cleavage after Ala6 39 results in production of Ab42. The C terminus remaining in the membrane would be expected to be 59 or 57 amino acids in length, respectively, alt hough only a 50 amino acid sequence has been identified. This APP intracellular domain (A ICD) presumably plays an important role in the regulation of multiple genes that are involved in the disease development (S omnicki and Le niak 2008). Gamma-secretase has been discovered to exert further action by cleaving at the -site of APP (Leu646-Val647), followed by cleavage at the -site (Val643 and Ile644). Finally the peptide is cut at the -site, releasing both A and AICD. In the non-amyloidgenic pathway, the production of A peptides is precluded because the APP ectodomain is cleaved by -secretase at the -site of APP (Lys613-Leu614) which resides within the A sequence. -secretase cleaves APP in to the released ectodomain (sAPP ) and a remaining membrane bound fragment C83. The remaining fragment is then processed by -secretase, similar to the amyloidgenic pathway, to generate AICD. Some mutations in the APP gene result in an increased production of pathogenic forms of A (Tanzi et al., 1991). To date th ere are 160 known mutations in APP and related cleavage enzymes, presenilin 1 (PS1) a nd presenilin 2 (PS2), that lead to familial AD. In fact, most inherited forms of AD re sult from mutations in PS1. Together with APH-1 (anterior pharyx defective 1 protein ) and nicastrin, th e presenilins constitute an integral part of the -Secretase, an aspartyl protease w ith low specificity that cleaves

PAGE 34

18 substrates within the sequence of their transmembrane domains. PS1 has nine transmembrane spanning domains and is thought to harbor the active site of -secretase. Two highly conserved aspartate residues in PS1 (257 and 385) within the 6th and 7th transmembrane domains of PS1 comprise the ca talytic core of the enzyme. Nicastrin is necessary for the proper assembly of the -secretase complex and has been proposed to modulate the traffic of substrates to the activ e site, and to serve as a substrate receptor (Wolfe, 2008) The functions of APH1 and PEN2 are not fully elucidated. APH1 has been implicated in stabilizing PS1 an d plays a role during assembly of the -Secretase complex, while PEN2 stabilizes the final comp lex and is involved in PS1 endoproteolysis (Spasic and Annaert, 2008). Because processing of APP is effected by secretase activity, mutations in individual components of this protein complex can lead to increases in A similar to mutations in APP. ApoE4 is the major genetic risk factor for Alzheimers disease. While it serves systemically as a lipid transport protein, in the brain, ApoE is rele ased by astrocytes and microglia as part of an immune response. It also plays a key role in the maintenance and repair of neurons. The 4 allele of ApoE confers the greate st risk of AD at the lowest age of onset. ApoE2, on the other hand, is associated with the lowest risk and highest age of onset, and apoE3 with intermediate risk and age of onset (Strittmatter et al., 1993). The exact mechanism behind the effect of ApoE isoforms as risk factors for AD remains under investigation. 1.3.3. Extracellular Aggregates in AD A is prone to aggregation under physio logical conditions. The C-terminal 12 amino acid residues from the transmembrane region of the A peptide are hydrophobic and promote self-aggregation and polymeriza tion into amyloid fibrils. The use of in A denotes its propensity to form cross -sheet arrays and polymerize into amyloid fibrils. Specifically, 1-42 has a stronger propensity for aggregation than the more common 1-40 peptide and is the key seed-forming A species in amyloid fibrils and plaques.

PAGE 35

19 A peptides behave aberrantly in AD, lik ely leading to a protein misfolding event that serves to nucleate further misfolding. Because of the numerous ways that an amino acid sequence can orient in 3 dimensions, mi sfolding of proteins is a common event in cells. The first line of defense against th e formation of misfolded proteins is their prevention via the utilization of molecula r chaperones. Chaperones associate with nascent polypeptides during translation as they emerge from the ribosome and promote proper folding. In the event that a protein is misfolded faster than a chaperone can prevent, a second line of defense, the ubiquitin/proteasome system, is employed. Misfolded proteins are ubiquitinylated and thereby (i n most cases) labeled for degradation by the proteasome. Both ch aperones and ubiquitin are found in amyloid plaques (as well as intracellular tangles), suggesting that these systems were overwhelmed in affected cells (Pallares-Trujillo et al., 1998) Alzheimers amyloid fibrils are fila mentous structures comprised of A about 10 nm wide and 100-1000 nm long. Previous stud ies using x-ray fiber diffraction analysis originally determined the presence of a -sheet structure within fibrils. These ribbon-like sheets form when A peptides align perpendicular to the length of the filament and are held together with hydrogen bonds that run pa rallel to the length of the fibril. These fibrils then aggregate with themselves and other molecules, such as Apolipoprotein E (ApoE)(Wisniewski and Frangione, 1992) and ACT (Ma et al., 1994), to form insoluble amyloid plaques that average 50 m in diameter (Rambaran and Serpell, 2008). Amyloid plaques have been the focu s of AD research for over 100 years. However, more recent data suggest that it might not be mainly the plaques that are neurotoxic, but the more soluble forms of A Oligomers, trimers, and dimers of A have been shown to be toxic to cultured neurons since th ey can readily pass through the cell membrane and cause neuronal damage (W alsh and Selkoe, 2007). Moreover, it is unclear if disaggregation of plaques could po ssibly lead to an increase in soluble A species. In either view, however, it is cl ear that the toxic starter molecule is A whether in aggregates, oligomers or, most likely, bot h. Therapies aimed at reducing monomeric A would reduce the amount available for oligomers and fibrils alike.

PAGE 36

20 1.3.4. AD and the cytoskeleton (tau) In addition to extracellular plaques, another central hallmark of Alzheimers disease is the accumulation of intracellular aggregates of hyperphosphorylated tau. Tau is a cytosolic protein involved in the assemb ly and stability of microtubules. In AD and other tauopathies, tau metabolism is alte red by overexpression of tau, mis-sense mutations, and posttranslational modificat ions. Upon phosphorylation, tau undergoes misfolding events, dissociates from micro tubules, and aggregates into filaments (reviewed in Chung, 2009). These filaments are the primary component of neurofibrillary tangles whic h disrupt the cytoskeleton a nd inhibit axonal transport, leading to cell death as reviewed (Gendron and Petrucelli, 2009). Tau is encoded by 16 exons located on ch romosome 17 and is primarily expressed in neurons. Exons 2, 3, and 10 are alternativel y spliced, leading to 6 isoforms of tau. Exon 10 encodes one of the regions involved in the binding of tau to microtubules, and alternative splicing of this e xon produces tau isoforms cont aining either three (tau 3R without exon 10) or four (tau 4R with exon 10) microtubule bindi ng domains (Andreadis, 2005). The contribution of the various tau isoforms to the development of tauopathies is evidenced by several tau mutations that cause frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17) (Kar et al., 2005). In th is disease, an increase in the splice variant of exon 10 leads to an elevated ratio of 4R w ith respect to 3R isoforms. This effect is mainly attributed to intronic mutations that result in the forced expression of exon 10. In contrast, in Picks disease tau mutations have been id entified that lead to decreased expression of exon 10. Aggregates are predominantly tau filaments composed of the 3R isoform. In AD, no mutations causing changes in tau isoforms have been identified, although increased expression of the 4R tau isoform has been detected in the affected brain region. Imbalance among tau is oforms (3R:4R) and the types of neurons involved are distinct for each tauopathy (Hernandez and Avila, 2007). Several phosphorylation sites have been identified in tauopa thies and respective kinases are implicated in the disease pathogenesis. Primar ily, tau is phosphorylated by glycogen synthase kinase 3 (GSK 3), cdk5, CAMKII (Wang et al., 2007) and less

PAGE 37

21 commonly by ERK1, CK1, and PKA (Li et al., 2004; Liu et al., 2006; Pei et al., 2002; Trojanowski et al., 1993). In total, 45 phos phorylation sites have been identified that contribute to AD pathogenesis. They represent more th an half of the total 85 phosphorylation sites identified on tau. Regulation of tau also occurs by dephosphorylation via PP2A and, to a lesser ex tent, PP1 and PP2B. Pin-1, a member of the peptidyl-prolyl cis-trans isomerase family of proteins, is invol ved in the assembly, folding and transport of certain proteins, including tau. Pin1 bi nds tau when it is phosphorylated at the AD-related thr 231 residu e and facilitates its dephosphorylation by PP2A (Butterfield et al., 2006). It is therefore through phosphorylation and dephosphorylation events that tau can be regulated in the CNS and effect aggregation. 1.3.5. Major Mouse Models of AD In order to study Alzheimers disease, ma ny different strains of transgenic mouse models have been developed over the past several years. The animals have been generated following the identification of numerous genes involved in familial (early onset) AD, where the patient develops the dis ease before the age of 65. In some studies, Tg2576 mice were used, which harbor the Swe dish mutations in the Amyloid Precursor Protein (APP) gene (Mullan et al., 1992). These mutations (K670D and M671L) lead to increased processing of the APP gene, resulting in the extracellular accumulation of peptides A 1-40 and A 1-42, the latter of which is more toxic. One of the main pathological hallmarks of AD is the formation of neuritic plaques, termed amyloid, found primarily in the hippocampus and cortex and comprised predominantly of the A peptides. Tg2576 mice also show cognitive impair ment similar to that of AD patients. Typically, plaques are detectable in their brains at approx imately 12 months of age and correlate with behavioral de ficits (Dewachter et al., 2001). Mouse models of Alzheimers disease ha ve been generated for many of the APP and PS1 mutations identified in the human population. Another common model is based on an APP gene in which valine is encoded instead of a phenylalanine at position 717, known as the London mutation (V717F) (Crawford et al., 1991). The PDAPP mouse is

PAGE 38

22 used throughout the experiments presented he re. There are also mutations in PS1 and PS2 genes that give rise to altered processivity of APP and increased generation of A peptides and correlating plaque load. PS1 M 146L is the model used in our laboratory. Double or triple transgenic mice harbor several of these mutated genes to determine the roles of these proteins in AD and aid in the design of therapeutics. Additionally, transgenic mice have been generated to model the other hallmark of ADtauopathy. htau mice (B6.CgMapttm1(GFP)Klt Tg(MAPT)8cPdav/J) are homozygous for knock out of the targeted endogenous mtau alleles and hemizygous for the transgene. Although no endogenous mtau is detected, all six isoforms (including both 3R and 4R forms) of human MAPT are expressed. H yperphosphorylated tau is detected in cell bodies and dendrites by three months of age. Paired helical filaments of aggregated insoluble MAPT can be isolated from brai n tissue as early as two months of age (Andorfer et al., 2003). 1.3.6. AD and inflammation ApoE and ACT are a part of a local inflammatory process in the brain. A in turn, has domains that bind ACT and ApoE, a nd these proteins serve to catalyze the conversion of A into filaments in vitro (Ma et al., 1996; Wisniewski et al., 1994 ) and in vivo to form amyloid deposits in the formation of beta-pleated sheets (Bales et al., 1999; Nilsson et al., 2001). ACT is a 68 kDa acute phase serum glycoprotein from the family of serine protease inhibitors (serpins). Systemically, it is released by hepatocyes and monocytes in response to inflammation, and is released by neutrophils to function as a suicide inhibitor of cathepsin G (Kalsheker, 1996). In the brain, ACT is known to be produced by and released from astrocyt es in response to IL-1 stimulation by microg lia, following trauma, infection, or amyloidoses (Abraham, 2001; Das and Potter, 1995). ACT is highly upregulated in AD brain and is involved both in the formation of amyloid plaques (Abraham et al., 1988) and in the phosphoryl ation of tau (Padmanabhan et al., 2006).

PAGE 39

23 Microglia have been traditionally regarded as resident CNS macrophages. They comprise about 10% of the adult CNS cell population (Pessac et al ., 2001). Microglia normally remain in a quiescent state in the healthy CNS, and are characterized morphologically by small cell bodies and elaborate processes. However, upon activation in response to in fection or CNS injury, microglia experience a change in morphology including shortening of cellular pro cesses and enlargement of their somas. Activated microglia up-regulate a host of cell surface activation antigens (major histocompatibility complex (MHC) class I and II, B7.1, B7.2, and CD40) and produce innate cytokines and chemokines (e.g., tumor necrosis factor(TNF), interleukin (IL)-1, IL-6, IL-12, and IL-18 (Janeway a nd Medzhitov, 2002) (Medzhitov and Janeway, 2000). 1.3.7. GM-CSF and Alzheimers disease Numerous studies have found Rheumatoid ar thritis (RA) to be a negative risk factor for Alzheimers disease (AD) pathoge nesis (Myllykangas-Luosujarvi and Isomaki, 1994), but the mechanisms for RAs pr otective effect ar e unknown. Although two decades of research have intensely focuse d on anti-inflammatory approaches against AD pathogenesis, NSAID clinical tr ials have failed to show therapeutic benefit in AD. This suggests that cyclooxygenase 2 activation is not the causative agent for the inverse relationship between RA and AD (McGeer et al., 2006). This outcome suggests that it might be intrinsic factors within RA pathoge nesis that protect against AD. Because RA is a peripheral disease that prevents the onset of a neurodegenerative disease, the most logical approach suggests immunological ph enomena. Keeping in mind that it seems unlikely for specific secreted factors to be produced in the synovial pannus and maintain sufficient concentrations during travel th rough the circulatory sy stem and across the blood brain barrier to inhibit AD pathogenesis, we determined that up-regulation of local cellular populations in RA would have the hi ghest potential to enter into the brain and inhibit the development of AD pathology a nd/or neuronal dysfunc tion. We therefore

PAGE 40

24 focused our attention primarily on leukocytes in volved in both diseases, and secondarily on their secreted inflammatory mediators. 1.3.8. Environmental co mponents of AD Similar to ALS, sporadic AD cases ar e thought to have an environmental component in addition to genetic risk factors. It is generally be lieved that individuals who remain physically and menta lly active, maintain strong soci al ties, and have a history of higher education are less lik ely to develop AD. Accordi ngly, groups that are isolated and have limited social interact ions have a strong propensit y to develop the disease. Studies with transgenic mice (PS/APP) in en riched environments (large cages, larger groups of animals, toys) compared to standa rd housing (individually caged), indicated that the enriched animals performed better in working memory tasks (RAWM) than controls (Costa et al., 2007). In addition to lifes tyle, nutritional components have also been implicated in conferring risks or benef its to the development of AD. For example, Resveratrol, a polyphenol in re d wine, epigallocatechin-3-ga llate (EGCG) in green tea, caffeine, and moderate ethanol consumpti on have all been shown to decrease the likelihood of developing AD (Arendash et al., 2006; Marambaud et al., 2005; RezaiZadeh et al., 2008). However, their exact mech anisms of interference have still to be determined. 1.3.9. AD and the vascular system The importance of environmental and me dical components for the prevention and cure of Alzheimers disease le ads to the question of how the blood brain barrier (BBB) is involved in the development and progression of Alzheimers disease. It has been known that the BBB in AD, similar to ALS, is comp romised. In fact, vascular dysfunction has been implicated in the progression of AD patients for over 20 years (Kalaria, 1992). Cerebral hypoperfusion and impaired LR P or ApoE-mediated transport of A across the BBB likely contribute to the progression of AD as well. In cerebral amyloid angiopathy

PAGE 41

25 (CAA), amyloid deposits are found lining the mi crovessels within the brain, potentially decreasing cerebral blood flow (CBF). Recen t data from clinical imaging experiments (MRI, SPECT) have shown that AD patients ha ve a resting CBF that is significantly reduced when compared to healthy contro ls (Bell and Zlokovic, 2009). The selective permeability of the BBB to certain molecules is of critical importance for maintaining homeostasis within the CNS. Disruption of this barrier, as seen in ALS and AD, can inhibit normal transport of molecules to a nd from the brain, or deteriorate and allow indiscriminate interaction between the br ain and periphery, exacerbating inflammation. 1.3.10. AD and oxidative stress Both fibrillar and oligomeric A species stimulate the formation of reactive oxygen species, including free radicals, within neurons (Moreira et al., 2008). Oxidative stress occurs when the antioxidant defens e system is overpowered by free radicals. Oxygen and nitrogen based molecules with unpaired electrons are very unstable and highly reactive. If they are not removed prope rly, they will react w ith lipids, proteins, and nucleic acids and disrupt membrane fluidity, enzymatic activity, and cause damage to organelles and DNA. Alternatively, dysfunc tional mitochondria could alter enzymatic activity in the antioxidant defense system. E ither way, there is a clear role of oxidative stress in AD pathogenesis as reviewed (Swerdlow, 2009). 1.3.11. Integration of many pathways Although it has been known for quite some time that both plaque and tangle pathologies exist in AD patients, it has not always been clear if the events occur independently or in close relatio nship. It is now known that A can interact with neurons to induce kinase activity and tau phosphorylation (Zheng et al., 2002). Moreover, A has been shown to exert similar e ffects intracellularly (Laferla et al., 2006). Therefore, according to the amyloi d cascade hypothesis of AD (Hardy, 2006), the

PAGE 42

26 overproduction of A leads directly to the formation of amyloid plaques and triggers intracellular aggregation of tau, leading to neuronal death. Other evidence also supports the fact th at amyloid and tau pathologies occur not completely independently. ACT, a protein prev iously described to be upregulated in the CSF of AD patients and a component of amyl oid plaques, has recently been shown to induce tau pathology. Cortical neurons isolated from human embryos and neonatal mice treated with ACT showed increased tau phosphorylation at the AD rele vant sites ser 202, thr 231, and thr 396/404 (PHF-1) that was tri ggered by the cellular kinases ERK1 and GSK3 (Padmanabhan et al., 2006). Moreover, it has been show n that ACT expression is upregulated in other tauopathies (FTDB, PSP, CBD). Collectively, these data support that ACT can induce the phosphorylation of tau, providing another pathological link between the seminal hallmarks of AD, namely plaques and tangles. 1.3.12. AD and neuronal death Apoptosis is the last pathway activat ed in AD and other neurodegenerative diseases, other biochemical processes are activated earlier in th e disease progression. Like ALS, apoptosis can be activated by the intrinsic and extrinsic pathways and is the final event in the disease progression. 1.4. Research RationaleAmyo trophic Lateral Sclerosis As described above, Amyotrophic Lateral Sclerosis and Alzheimers disease are neurodegenerative diseases that share certain features and therefore might be prevented or cured with related procedures or medications. It is with this perspective that I focused my research on the promotion and inhibiti on of protein aggreg ation in Amyotrophic Lateral Sclerosis and Alzheimers disease.

PAGE 43

27 1.4.1. Transmission and induction of ALS Proteinopathies, i.e. diseases that are based on diverse forms of protein aggregation and misfolding or the secondary modifications that are based on those features, can be caused by environmental and/ or genetic components. For example, prion diseases involve the transformation of speci fic proteins into mo re aggregate-prone conformations, leading to the nucleati on of beta-pleated sheet structures. Based on the transmissible nature of pri on diseases after i ngestion of a prion protein and given the aggressive and progressive nature of ALS, we sought to determine if the intracellular inclusions in ALS that contain SOD1, ubiquiti n, and neurofilaments, could be isolated from the spinal cords of affected transgenic ALS mouse models and injected into young transgenic mice that do not yet carry the pathoge nic form of SOD1. The goal was to assess if the disease could be induced in the host. If these high molecular weight proteins c ould be isolated from ALS mi ce and delivered via a single bolus injection into the lateral ventricles of recipient mice, new avenues might be opened to study this devastating disease. The literature on spinal cord diseases re ported the route of protein delivery we had thought out for this experiment as rath er problematic. Because the CNS relies on passive diffusion and the animals movement to circulate CSF, delivering small amounts of a test substance into the lateral ventricl es was not likely to diffuse the substance in sufficient quantity to the lumbar cisternthe initial site of degeneration in ALS. Additionally, it was questionable if exposure to a single inject ion could trigger a response in the animals. For these reasons, we pref erred to develop a tec hnique allowing chronic infusion of homogenates into the lumbar cister n at L3 and L4. Catheters were fashioned to be placed directly into the intrathecal spac e, sutured into place, and attached to osmotic minipumps for 2 and 4 week delivery. Changes in motor ability (rotarod, latency to/ability to turn over) as well as life-span and enhan cement of pathology (increased inclusions and vacuolization of dysfunc tional mitochondria) were investigated.

PAGE 44

28 1.4.2. Disruption of th e blood brain barrier The blood-brain barrier (BBB), blood-spin al cord barrier (BSCB), and bloodcerebrospinal fluid barrier (BCSFB) control cer ebral/spinal cord homeostasis by selective transport of molecules and cells.. In the sp inal cord and brain of both ALS patients and animal models, infiltration of T-cell lymphocytes, monocyte-derived macrophages and dendritic cells, and IgG deposits have been observed that may have a critical role in motor neuron damage. Additionally, increased levels of albumin and IgG have been found in the cerebrospinal fluid in ALS patie nts. These findings suggest altered barrier permeability in ALS. Recently, using electron microscopy, we showed disruption of the BBB and BSCB in areas of motor neuron degene ration in the brain a nd spinal cord in G93ASOD1 mice modeling ALS at both early and late stages of disease (GarbuzovaDavis et al., 2007c). Examination of the capill ary ultrastructure reve aled endothelial cell degeneration, which, along with astrocyte al teration, compromised the BBB and BSCB. However, the effect of these alterations upon ba rrier function in ALS is still unclear. The aim of this study was to determine the func tional competence of the BSCB in G93A mice at different stages of disease 1.4.3. Neuroprotective effects of DTG (1,3-di-o-tolylguanidine) DTG (1,3-di-o-tolylguanidine) is a high affinity sigma receptor agonist that has often been reported to ameliorate stroke damage in rodents as reviewed (Willing and Pennypacker, 2007). We therefore thought to investigate if DTG could serve as a potential treatment to reduce the robust pa thology in ALS mouse models. Because DTG was known to pass the blood brain barrier (BBB) direct delivery into the CNS was not required and subcutaneous release over a period of 28 days was expected to be sufficient. In mouse models of stroke, DTG had been re ported to block the inflammatory response evoked by MCAO, as measured by the decreased number of reactive astrocytes and activated microglia (Ajmo et al., 2006). Neur onal survival is specifically enhanced if DTG is administered 24 hours after an ischemic stroke. Because the efficacy of DTG for

PAGE 45

29 stroke treatment was likely the result of co mbined neuroprotective and anti-inflammatory properties of sigma receptors, and because th ese receptors are found ubiquitously in the CNS, we hypothesized that enhancing the ac tivity of sigma receptors might provide an effective means for the treatment of ALS as well. 1.4.4. Novel therapeutic la minin derivative, KDI Laminin is a heterotrimeric extrace llular matrix glycopr otein involved in numerous biological processes including cel l adhesion, differentiation, migration, and, most importantly, neurite outgrowth. Our collabora tor, Paivi Liesi, elucidated that it is, in fact, the soluble tripeptide fragment compos ed of the amino acids lysine, aspartic acid and isoleucine (KDI) at the ca rboxy terminus of the laminin -chain that promotes neurite outgrowth (Liebkind et al., 2003). They showed that KDI coul d protect the brain from a kainic acid induced lesion in a dose dependent manner (Wiksten et al., 2004a) and that local administration of KDI to a fully transected rat spinal cord would actually heal the tissue, promote the growth of neurons th rough the glial scar, and restore lower limb function to the animals (Wiksten et al., 2004a). Because the pharm acokinetics of this tripeptide had not yet been esta blished, direct delivery into the intrathecal space (lumbar cistern) via customized cathete rs instead of systemic appli cation was desired. Kainate is a glutamate analog and glutamate excitotoxici ty (dysregulation) is a common feature of both sporadic and familial ALS. The prospect of using KDI to treat ALS was therefore worthwhile for 2 reasons: Neuroprotection from dysregulated glutamate excitotoxicity and re-growth of damaged motor neurons via laminins and integrins. 1.5. Research RationaleAlzheimers Disease 1.5.1. Development of a bilateral infusion system in mice Many lines of transgenic mice harboring the gene for human amyloid precursor protein (APP) with different mutations causing familial Alzheimers disease have been

PAGE 46

30 developed over the past decade to study pla que deposition and other aspects of AD. However, variations in size, density (fibrilla r and dense core), pla que number, and total amyloid load between animals of the same age and genotype have been reported, and significant pathological analys es thus often require very large numbers of mice. Therefore, to study AD therapies for example, it was imperative to develop a technique that would allow each brain hemisphere to r eceive different infusions. To this end, we developed a catheter technique that facilitates simultaneous bilateral infusion in mouse brains, thereby using the contralateral hemis phere of the same animal as an internal control. The infusions occurred over a period of 28 days with less than 5% mortality. 1.5.2. Intracranial infusion of KDI, a glut amate receptor antagonist and electroporation of a KDI expressing plasmid A) Memantine was the fi rst in a novel class of Alzheimer's disease medications that acts on the glutamatergic system by blocking NMDA glutamate receptors. Blocking these receptors causes an upregulation in dormant AMPA receptors in post-synaptic neurons. In the dentate gyrus, this upregulation is part of the process of neurogenesis. Trying KDI as potential therapeutic for Alzheimers disease was based on the noncompetitive, ionotropic glutamate receptor antagonist properties of KDI. KDI was infused intracranially to induce its effect while conferring general neuroprotection to increased glutamate levels. B) In vivo electroporation has proven to be an eff ective means of delivering plasmids to virtually any tissue or organ without the need for attenu ated viral vectors. The nucleotides encoding the six amino acids representing the longe r c-terminus of -1 laminin (EIIKDI) were successfully cloned into PCDNA 3.1+. After optimizing the electroporation conditions with an EGFP e xpression vector, the plasmid encoding the KDI-peptide was introduced for sustained expression into the hippocampus of PS/APP mice via electroporation.

PAGE 47

31 1.5.3. Alzheimers Disease and -1 antichymotrypsin 1.5.3.1. ACT and tauopathies Recent work in our lab has shown, in v itro, that tau becomes phosphorylated at several AD-related sites when primary ne uronal cultures are incubated with ACT (Padmanabhan et al., 2006). In other, non-Alzheimers human tauopathies, ACT is also upregulated, suggesting that it may have a ro le in tying together the mayor players in Alzheimers pathology, i.e. amyloid and tau. In this experiment, th e same technique was employed as previously (bilateral infusions ), but the infusions were performed in different transgenic animals. Here I used mice expressing the human tau protein (htau), and knocked out for the endogenous mouse tau (mtau), to determine if ACT has a similar effect on the phosphorylation of tau in vivo as was shown in cell culture, potentially leading to NFTs. 1.5.3.2. Inhibition of the ACT-A interaction ACT is a serine protease inhibitor that has been show to act as a pathological chaperone for the development of Alzheimers disease. ACT promotes the fibrillization of A in vitro, is upregulated in the CSF of AD patients and is an inte gral part of amyloid plaques. It is not known, however, how AC T reacts with exis ting plaques. While transgenic animals are useful for st udying these phenomena, they have two disadvantages: they are c onstructed to overexpress the transgene throughout their lifetime and show a strong variability in plaque load between animals. To view ACTinduced changes in plaque number, plaque size, and plaque mo rphology and to allow direct comparison in the same animal, ACT wa s infused unilaterally into the parenchyma of the (ipsilateral) hippocampus and CSF as control into th e (contralateral) hippocampus of the same animal. The purpose of this experiment, ultimately, is to test various compounds for their ability to inhibit the ACT-A interaction. Previous in vitro work showed that fragments of A when incubated with ACT or ApoE and full length A can

PAGE 48

32 inhibit the formation of fibrils and increase survival of transfected cells (Ma et al., 1996). In this experiment, fragment 1-11 which is known to be the ACT binding fragment of A was co-infused with ACT into PS/APP mi ce to disrupt amyloid formation. Other experiments entailed infusion of the 1-11 pe ptide alone to see if it might inhibit the endogenous ACT-equivalent in the mouse. 1.5.4. Colony stimulating factors and amyl oid clearance and cognitive enhancement Numerous studies have found Rheumatoid ar thritis (RA) to be a negative risk factor for Alzheimers disease (AD), but the mechanisms for RAs protective effect are unknown. Much research has proposed that medication with NSAIDs, which is common for RA patients, is protectiv e against AD. However most NSAID clinical trials have proven unsuccessful in AD patients and have th erefore led to the hypot hesis that intrinsic components of RA pathogenesis may be protec tive against AD. In these experiments, we proposed that cytokines are upregulated in RA and confer neuroprotection to AD. Multiple colony stimulating factors that induc e leukocyte populations and are involved in RAs pathogenesis, such as GM-CSF, are upregualted in RA patients and were the natural target of our investigation. Fi rst we performed bolus intrahippocampal administration of rmGM-CSF into one hemisphere and control aCSF into the other hemisphere to compare plaque load and reactiv e gliosis. In a second approach we used sub-cutaneous rmGM-CSF injection into aged cognitively-impaired PS/APP mice in an attempt to reverse the cognitive impairment. Both standard radial-arm water maze task and a novel Cognitive Interference task we re employed to measure any cognitive enhancement.

PAGE 49

33 CHAPTER 2 TRANSMISSIBILITY OF MISFOLDED SUPEROXIDE DISMUTASE INCLUSIONS IN A MOUSE MODEL OF ALS 2.1. Introduction My first investigations related to AL S focused on the proteinaceous inclusions formed within neurons and astrocytes in the CN S of ALS patients. Of particular interest was the possibility that deposits of mutant or misfolded SOD1 might carry the capacity to transmit the disease in a prion like fashion. In prion disease, the native prion protein, PrPc, (PrPc,= cellular prion pr otein) features a predominan tly alpha-helical structure, while PrPSc, as in scrapie, a prion disease in sheep, features a pre dominantly beta-pleated sheet structure. The only difference between these proteins is a conformational change and according to the protein only hypothesi s, the agent responsible for misfolding native prion protein to the aggregate prone form is the misfol ded protein itself (Prusiner, 1998). PrPSc initiates and sustains the propa gation of the pathogenic misfolding of the PrPc, molecule in vivo and causes neurodegene ration in a classic ma nner: Progressive loss of neurons, local inflammation (microglio sis/astrocytosis), deposition of abnormally folded prion protein (PrPsc), and, notably, tran smissibility in most forms of the disease (transmissible spongiform encephalopathyTSE). Seeded disease transmissibility/promo tion has also been demonstrated in Alzheimers mouse models. However, the di seases could only be induced if the mice were genetically predisposed to develop the disease. In young APP transgenic mice, cerebral amyloidosis can be triggered with intr acranial infusion of d ilute cortical extracts from AD patients and they develop amyloid pathology within a few months A -rich brain extracts from APP transgenic mice produce seeding similar to that from human extracts, indicating that the inducing agent is not uniquely presen t in the human brain

PAGE 50

34 (Walker et al., 2002). Infusion of brain extr acts from WT or NTG mice into APP mice, or extracts from APP transgenic mice into Wt or NTG mice induced no phenotypic change, however, there is evidence that injecting brain homogenates from plaque producing mice induces tau pathology in tau transgenic mice (Bolmont et al., 2007). Moreover, direct infusion of A 1-42 into AD transgenic mice, although sufficient to cause cognitive decline and neuronal death, will not form plaques. Taken together, the data from a number of neurodegenerative diseases indicated that abnormal protein conformation could be acquired and induced in other animals. I therefore sought to determine if a transmissible element could be identified within the motor neuron inclusions of ALS mice. Spinal cord homogenates from symptomatic SOD1 G93A mice were homogenized, determ ined to contain high molecular weight species of SOD1, and injected into the late ral ventricles of SOD1 and NTG mice in an attempt to induce or speed the developmen t of the ALS phenotype. Additionally, the extracts were infused directly into the lumbar cistern of these animals. Finally, spinal cord (SC) homogenates from symptomatic ALS mice were intrathecally infused into presyptomatic SOD1 transgenic mice with the in tention to hasten the disease onset and/or progression. 2.2. Materials and Methods 2.2.1. Intrathecal Catheter Construction Recent work by our and other groups has s hown that compounds that are injected into the cerebrospinal fluid (CSF) do not diffuse efficiently throughout the spinal cord (unpublished observations). It wa s therefore necessary to infu se the peptide continuously into the CSF of the intrath ecal space (lumbar cistern) of the spinal cord of the ALS models. This approach turned out to be a very arduous task for many research groups, and successful long-term intrathecal catheteri zation in mice, partic ularly in conjunction with Alzet osmotic minipumps, had not been reported at this time. Due to meticulous attention to surgical details and assi stance from the ALS Therapy Development

PAGE 51

35 Foundation, I succeeded in performing long-te rm intrathecal cath eterization in mice. To this end, intrathecal catheter s were fashioned by first inse rting a 10 cm, 0.005 diameter tungsten wire (Small Parts, Inc. Miramar, Fl) into a 3 cm length of polyethylene tubing (PE-10, Becton Dickinson, Sparks, MD.) with an internal diameter of 0.011 and stretching the material so it was snug with the wire. The un-stretched portion of the PE10 (outer diameter 0.024) was then inserted into a 4 cm long pol yvinyl tubing (Durect Corp. Cupertino, Ca.) (internal diameter 0.027) with approximately 2 cm of overlap. The space difference of 0.003 in the step-up to the larger polyvinyl tubing is filled with Locktite 454 adhesive (Plastics One. Roa noke, VA.) and secured with a silk suture (figure 1). After the assemblies were left to dry overnight, sterile water was forced through to assure that the lines were not occluded or damaged. 2.2.2. Spinal Cord Extracts After an overdose of sodium pentobarb ital (~150mg/kg) (i.p.) 6 animals (4 hSOD1 G93A mice and 2 hSOD1WT ) were transcardially perf used with 0.9% saline. The spinal cords were microdissected from a group of 5 late disease stage animals and 2 age-matched controls, immediately dounce homogenized in cold 300 M sucrose, aliquoted, and frozen at -80oC until use. Equal amounts of protein from each sample were mixed with 2x sample buffer (Invitrogen, Carlsbad, CA), boiled for 5 min, and subjected to 10-20% Tris Tricine SDS ge l electrophoresis (In vitrogen). The electrophoresed proteins were wet transferred onto 0.2 m nitrocellulose (Whatman, Dassel, Germany) and probed with a rabbi t anti-mouse polyclonal antibody to SOD1 (1:1000). The secondary antibody was horse radish peroxidase-conjugated donkey antirabbit IgG (Pierce, Rockford, IL). The immunolabelled proteins were visualized with an enhanced chemiluminesence detection kit (S uperSignal pico, Pierce, Rockford, IL). After different exposures, autoradiographs were developed with a Konica/Minolta SRX101A (Tokyo, Japan) and scanned using a Ricoh Aficio MP3010. Homogenates that showed mostly higher weight SOD1 immunoreac tivity were considered to be aggregates and used for subsequent infusions (figure 2).

PAGE 52

36 2.2.3. Osmotic Minipumps Sterile Alzet osmotic minipumps-model 2004 (Durect Corp. Cupertino, Ca.), with a flow rate of 0.25 L/hr were filled with 200 g of total protein from homogenates of G93A mutant and wt transgenic SOD1 mi ce, diluted in artificial CSF (Harvard Apparatus, Holliston, MA) and incubated at 37oC in sterile saline for 48 hours. Typically, catheters are attached and primed at this time, but the tungsten wire needed to remain in place for stability du ring insertion into the spine. 2.2.4. Mice Another 10 SOD1 G93A mice a nd 4 controls (C57/B6 backgr ound) were purchased from Jackson Labs (Bar Harbor, ME.). These mice express 18 copies of mutant SOD1 and develop ALS pathology at appr oximately 100 days. Their av erage lifespan is about 130 days and the phenotype closely resembles a more aggressive familial form of the disease. This model was chosen because at the time it was widely accepted by ALS investigators and the therapeutic/pathological window wa s short enough to run multiple studies. 2.2.5. Intrathecal Infusions All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (90 days old25g) were anesthetized with 1-2% isoflurane and shaved and scrubbed with 10% Betadine so lution at the site of incision. A small (3 cm) incision was made, exposing the lumbar region of the vertebral column, and double bladed scissors were used to form a subcutaneous pocket along the back of the animal between the scapula and towards the head. A puncture was then made between the L2 and L3 vertebrae with a 23 gauge needle that allows the small, stretched portion of PE-10 (stabilized with the tungsten wire) to be in serted 0.9 cm into the intrathecal space and being stopped by the silk suture used to adjoin the two different sized tubings. Accurate

PAGE 53

37 placement of the catheter was visualized in ev ery instance by a brief tail flick or muscle twitch, at which point the tungsten wire was removed and an Alzet pump attached. Another silk suture was used to anchor the catheter into the superficial erector musculature dorsal the vertebral column. Th e placement of the second suture into the musculature immediately caudal to the first assembly stitch on the catheter prevented them from being pulled out during the course of the experiment. The pumps infused for 28 days, around which time the animals were at the terminal stages of paralysis. 2.2.6. Histology and Immunohistochemistry Mouse spinal cords were carefully removed and fixed in 10% formaldehyde (formalin) for 72 hours and then passed thr ough a series of sucrose solutions (10-30%) over another 72 hours. Tissues were frozen to the peltier stage (Physitemp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzer land) and sectioned coronally at 25 m. After incubating the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X-100, and 0.02% sodium azide) for 60 min, primary antibodies against SOD1 (1:500) were applied and incubated at 4 o C overnight. After thorough washing, the sections were incubated with secondary antibody, Alexa 488 and 594 fluorophores (1:1000, 1:4000-Invitrogen) in the da rk at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,000, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and s ealed with Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). Cresyl violet staining was used to analyze the general morphology of motor neurons and vacuolated m itochondria (figure 3). The slides were first stained with cresyl viol et (Sigma, St. Louis, MO1 gram/400mL water at pH 3.6) for 3 minutes, rinsed in water, and placed in 95% ethanol, successively, 3 times, for 4 minutes each. Then they were placed into 100% ethanol twice fo r 2 minutes each and xylene twice for 2 minutes each before bei ng coverslipped with Cytoseal (Electron Microscopy Sciences, Fort Washington, PA) and left to dry at ambient temperature overnight. The images were analyzed on a Zeiss Imager Z1 microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axiovision 4.7.

PAGE 54

38 2.2.7. Behavior Motor deficits and the health of the animals were assessed throughout the experiment. Body weights before and after a 28 day treatment were measured daily as well as the biweekly performa nce on a rotarod test. The ro tarod is designed to assess motor coordination, balance and equilibrium. The test mouse is placed on the rod and the rotarod accelerates gradually to a pre-set maximum speed. The mices performance abilities are recorded as latencies in falling from the rod. The rotarod consists of a semienclosed chamber which contains a beam ma de of ribbed plastic and flanked by round plates on either side to prevent any escape. The rod is suspended at a height of 20 cm above the counter and the mouse is placed on top of the beam, facing in the orientation opposite to that of its rotation, so that forw ard locomotion is necessary to avoid falling (figure 4). The rotarod spins consistently at 1 of 10 pre-set speeds (#5 was chosen) during three 2-minute trials. Latencies for the mice to fall from the rod are recorded automatically once the fallen animal interrupt s a laser beam, thereby stopping the timer. Each mouse is given 3 trials with a 20 minute interval betw een trials. Once it is evident that the mouse is unable to perform,the animal is removed from the study. 2.3. Results and Conclusions Mating experiments in which mice harbor ing the wild type human SOD1 gene were mated with mice harboring the wild t ype human SOD 1 gene G85R resulted in offspring with a more aggressive phenotype than the G85R alone (Wang et al., 2009). Disease onset time and age of death were both significantly reduced in the litter, suggesting that G85Rs interaction with WT SOD1 enhances its pathogenicity. With similar rationale, we sought to determine if there was an element within the inclusions, if not mSOD1 itself, that could, similar to prion diseases induce a misf olding event, thereby accelerating the disease process. In these experiments, we wished to dete rmine if the induction of misfolding in the spinal cords of SOD1 G93A, a nother mutant human SOD1, was transmissible, i.e. if a misfolded protein from a host can be introdu ced into a recipient to nucleate further

PAGE 55

39 misfolding and cause neurodegeneration. Transm issibility, not necessar ily infectivity, of misfolded proteins might provide insight into how the process of nucleation occurs in proteinopathies. Recent work with tau tangl es, for example, in AD demonstrates that once the tangles are injected into the brain of a presymptomatic mouse, newly formed tangles form immediately, indicating that th e first exposure to the tangles triggers a positive feedback loop of tangle fo rmation (Clavaguera et al., 2009) We sought to determine if exposure to mutant, aggregated SOD would nucleate the formation of characteristic intracellular in clusions in the spinal cord of ALS mice. This was approached first with intracerebr oventricular bolus injections of homogenized spinal cords (figure 2) from symptomatic mice into WT and non-symptomatic mice in order to accelerate the phenotype. We found, however, that this was an ineffective means of delivery in this model, since the pathology begins in the anterior spinal column and molecules travel in the CSF diffusely. For this reason, customized catheters in combination with osmotic minipumps needed to be made in order to chronically and effectively deliver the homogenates directly to the target tissue and to infuse the homogenates for a month rather instead of single injections at a distal site. Table 1 below shows that mutant mice receiving mutant homogenates lived an average of 129.6 days whereas mutant mice receiving WT homogenates lived an average of 124 days. The average lifespan of animals in this colony was 128. No significant change in disease onset (or time of death) we re apparent between the groups behaviorally (data not showndevice figure 4), and histological examinat ion of the spinal cords was not warranted. Furthermore, no other group ha s reported transmissible elements of ALS since these experiments were conducted. In sum, unlike AD and prion disease, SOD1 inclusions cannot seed or nucleate the fo rmation of new aggregates, suggesting a divergence in pathology from other neurodege nerative diseases. At the same time intrathecal delivery of molecule s into mice, at th is time, a novel tec hnique only practiced by a few researchers was proven to be a valuable tool for drug delivery.

PAGE 56

2.4. Figures Figure 1. Intrathecal catheter assembly 40

PAGE 57

Figure 2. Western Blot of spinal cord homogenates from nontransgenic, SOD1WT, and mSOD1 G93A mice which were subsequently used for chronic intrathecal infusions 41

PAGE 58

SOD1 G93A Nontransgenic Figure 3 20 m sections from the lumbar region of spinal cords of nontransgenic (left) and SOD1 G93A mice (right) at 5x magnifi cation. Below is increased magnification (40x) from the ventral horn showing dying mo tor neurons with mitochondrial vacuoles in the SOD1 G93A mouse (right). 42

PAGE 59

Figure 4. A standard rotarod for individua l mice (Med Associates Inc.) 43

PAGE 60

Table 1. Comparing lifespan from SOD1 G93A mi ce infused with SOD1 G93A spinal cord homogenates (AGG-M) to animals infused with WT infusions (AGG-S) 44

PAGE 61

45 CHAPTER 3 EVIDENCE OF COMPROMISED BLOOD-SPINAL CORD BARRIER IN EARLY AND LATE SYMP TOMATIC SOD1 MICE MODELING ALS Svitlana Garbuzova-Davis a-d*, Samuel Saporta a,b,d, Edward Haller d Irina Kolomey a, Steven P. Bennett f,g, Huntington Potter f,g, Paul R. Sanberg a-e 3.1. Abstract The blood-brain barrier (BBB), blood-spin al cord barrier (BSCB), and bloodcerebrospinal fluid barrier (BCSFB) control cer ebral/spinal cord homeostasis by selective transport of molecules and cells from the syst emic compartment. In the spinal cord and brain of both ALS patients and animal models, infiltration of T-cell lymphocytes, monocyte-derived macrophages and dendritic ce lls, and IgG deposits have been observed that may have a critical role in motor neur on damage. Additionally, increased levels of albumin and IgG have been found in the cerebrospinal fluid in ALS patients. These findings suggest altered ba rrier permeability in ALS. Recently, we showed disruption of the BBB and BSCB in ar eas of motor neuron degeneration in the brain and spinal cord in G93A SOD1 mice modeling ALS at bot h early and late stages of disease using electron microscopy. Examinati on of capillary ultrastructure revealed endothelial cell degeneration, which, along with astrocyte alteration, compromised the BBB and BSCB. However, the effect of these alterations upon barrier function in ALS is still unclear. The aim of this study was to determine the functional competence of the BSCB in G93A mice at different stages of disease. Methodology/Principal Findings : Evans Blue (EB) dye was intravenously inject ed into ALS mice at early or late stage disease. Vascular leakage and the condition of basement membranes, endothelial cells,

PAGE 62

46 and astrocytes were investigated in cervical and lumbar spinal cords using immunohistochemistry. Results showed EB leak age in spinal cord microvessels from all G93A mice, indicating dysfunction in e ndothelia and basement membranes and confirming our previous ultrastructura l findings on BSCB disruption. Additionally, downregulation of Glut-1 and CD146 expressions in the endo thelial cells of the BSCB were found which may relate to vascular leakage. Conclu sions/Significance : Results suggest that the BSCB is compromised in areas of motor neuron degeneration in ALS mice at both early and late stages of the disease. 3.2. Introduction The central nervous system (CNS) is an immunologically privileged zone, which is normally protected from entr y of immune cells and seru m proteins by the blood-brain barrier (BBB), blood-spinal co rd barrier (BSCB), and bloodcerebrospinal fluid barrier (BCSFB). These barriers are specialized struct ures of the CNS that control cerebral/spinal cord homeostasis by selective transport of molecules and cells from the systemic compartment (reviewed in (Ballabh et al., 2004; Engelhardt, 2006; Pardridge, 1999, 1988; Rebenko-Moll et al., 2006; Sharma, 2005; Vorbrodt and Dobrogowska, 2003). This control is possible du e to the unique structural elem ents of the microvasculature endothelial cells of brain capil laries and epithelial cells of the choroids plexus, astrocyte end-feet, and pericytes. Brai n (spinal cord) capillary endot helial cells are distributed along the length of the vessels completely encircling th e lumen and connected via adherens and tight junctions. Adherens junctions support inte rcellular adhesion, as tight junctions form a diffusion barrier for mo st blood-borne substances. The basement membrane (i.e. basal lamina), surrounding th e endothelial cells and pericytes, supports the abluminal surface of the endothelium. Th e basement membrane, which consists of laminin, fibronectin, collagens, and proteo glycans (Dermietzel and Krause, 1991; Rutka et al., 1988), separates adjacent tissues, acting as a barrier to the passage of macromolecules and cell migration. The astr ocyte perivascular end-feet ensheathing approximately 95% of the vessel wall appear to have an important role for maintenance

PAGE 63

47 of the BBB (Prat et al., 2001). Thus, functiona l integrity of all BBB/BSCB elements is critical for protection of the CNS fr om various harmful blood substances. Impairment of the BBB occurs in various pathological CNS conditions. Accumulation of collagen in vascular basement membranes and focal necrotic changes in endothelial cells were found in Alzheimers patients (Claudio, 1996). Degradation of the extracellular matrix may be concomitant with BBB disruption and tissue softening, setting the stage for the most pronounced fo rms of brain swelling and leading to the development of severe cerebral edema over subs equent hours and days in stroke patients (reviewed in (Avata and Ropper, 2002)). In multiple sclerosis (MS), which is characterized by inflammatory lesions within the CNS, BBB disruption enables leakage of the serum protein fibrinogen into the br ain parenchyma (Kirk et al., 2003; Kwon and Prineas, 1994) and may precede myelin damage aggravating the inflammatory process. Recently, Vos et al. (Vos et al., 2005) demons trated that BBB dysfunction in MS patients is apparent not only in focal lesions but al so in diffuse abnormalities in white matter which were detected by postmortem MR I. BSCB breakdown was found in an experimental model of traumatic spinal cord injury (Noble and Wr athall, 1989; Olsson et al., 1992). It has been shown that spinal nerv e lesions also alter BSCB function (Gordh et al., 2006). Results of these studies showed in creased microvascular permeability leading to blood protein extravasation and the formation of vasogenic edema that play important roles in the pathophysiology of the di seased or injured spinal cord. Amyotrophic lateral sclerosis (ALS) is a progressive de generative disease affecting motor neurons in the spinal cord, motor cortex and brainstem that leads to paralysis and death within five years of di sease onset (Miller et al., 2000; Rowland and Shneider, 2001a). Although numerous hypothese s about the etiology and pathogenesis of ALS have been proposed (reviewed in (Bru ijn et al., 2004a; Cleveland and Rothstein, 2001; Strong et al., 2005a)), increasing evidence points to immune sy stem involvement in ALS pathogenesis (Alexianu, 1995; Alexianu et al., 2001; Appe l et al., 1994; Kawamata et al., 1992). In the spinal cord and brain of both ALS patients and animal models, the presence of T-cell lymphocytes (Alexianu et al., 2001; Engelhardt et al., 1993b), deposits of IgG (Donnenfeld et al., 1984b; Engelhardt and Appel, 1990; Engelhardt et al., 2005a;

PAGE 64

48 Mohamed et al., 2002), complement component s C3 and C4 (Donnenfeld et al., 1984b), and monocyte/macrophage and dendritic cells (Engelhardt et al., 1993b; Henkel et al., 2004; Troost et al., 1990b) were observed that ma y have a critical role in motor neuron damage. Recently, IgG was detected in the perikarya of motor neurons of the lumbar spinal cord in mice after in traperitoneal inject ion of IgG derived from sera of ALS patients (Engelhardt et al., 2005a). Moreover, the uptake of IgG in multivesicular bodies in endothelial cells in the affected areas of the spinal cord was found in both ALS patients and mice injected with human ALS IgG. Additionally, significan tly increased levels of albumin, IgG, and C3c have been noted in the cerebrospinal fluid of AL S patients (Annunziata and Volpi, 1985b; Apostolski et al., 1991; Leonardi et al., 1984b; Meucci et al., 1993). These findings suggest that barrier permeability may be affected in ALS. Recently, we showed disruption of the BBB and BSCB in areas of motor neuron degeneration in the brain and spinal cord in G93A SOD1 mice modeling ALS at both early and late stages of disease using electron microscopy (Garbuzova-Davis et al., 2007b). Examination of capillary ultrastructure revealed endothelial ce ll degeneration, which, along with astrocyte alteration, compromised the BBB and BSCB. Howe ver, the effect of these alterations upon barrier function in ALS is still unclear. The aim of this study was to determine the functional competence of the BSCB in G93A SOD1 mice modeling ALS at different stages of disease. 3.3. Results 3.3.1. Disease symptom progression in G93A mice The G93A mice were monitored weekly fo r initial disease symptoms and disease symptom progression by body weight and extens ion reflex, beginning at 7 weeks of age. Body weights of G93A mice gradually in creased until 11 weeks of age and then stabilized for 2 weeks (12-13 w eeks of age). Tremor, an initial sign of disease, was seen in some mice as early as 13 weeks of age. After this time, mice consistently lost body

PAGE 65

49 weight due to muscle atrophy. By 18 weeks of age, G93A mice had lost approximately 14% (p=0.06) of their 13-weeks-of-age body mass, dropping from 25.4.38 g to 21.8.55 g (Figure 5, A). Eighteen weeks ol d G93A mice weighed 28% less than C57BL/6J mice of the same age (p=0.007), which continued to maintain their body weights until 20 weeks of age. Deterioration in extension reflex started to appear in G93A mice at 13 weeks of age (Figure 5, B). During the next 5 weeks, mice demonstrated slowly declining hindlimb ex tension and by 18 weeks of age, showed no extension (p<0.001) and exhibi ted hindlimb paralysis. Thus G93A mice at about 13 weeks of age showed initial si gns of disease such as trem or, weight loss, and reduced hindlimb extension. Terminal stage of diseas e in these mice was observed at 17-18 weeks of age, as demonstrated by complete hind limb paralysis, significant reduction of body weight and absence of hindlimb extension. 3.3.2. Motor neurons in the cervical and lumbar spinal cords of G93A mice Nissl body staining was performed to id entify motor neuron condition in the spinal cords of G93A mice in ea rly and late stages of diseas e. In control C57BL/6J mice, many healthy motor neurons with large soma an d neuritic processes were visible in the cervical and lumbar ventral horn of the spin al cords at 12-13 weeks of age (Figure 6A, Figure 7A) and 19-20 weeks of age (Figure 6B, Figure 7B). In the cervical spinal cord of G93A mice, numerous motor neurons with vacu olization were found at 13 weeks of age, concurrent with initial disease symptoms (Figure 2C). When disease symptoms progressed and mice were paraly zed, usually at 17-18 weeks of age, the number of motor neurons had decreased and motor neurons of various sizes displayed vacuolization (Figure 6D). Only a few healthy motor neur ons were identified at this time. More dramatic changes in motor neuron condition we re found in the lumbar spinal cords of ALS mice. Most motor neurons in 13 week old mice showed signs of degeneration (Figure 7C). In G93A mice at 17-18 weeks of age, the ventral horn was essentially devoid of motor neurons (Figure 7D).

PAGE 66

50 3.3.3. Fluorescent detection of Evans Blue in the cervical and lumbar spinal cords of G93A mice Evans Blue, the tracer used to assess disr uption of the BSCB, was injected into G93A mice at initial (1 3 weeks of age) and late (17-18 w eeks of age) stages of disease symptoms prior to euthanasia. C57BL/6J mice were also administered the EB dye. In the cervical spinal cord, EB was clearly detected within the blood vessels in the control C57BL/6J mice at 12-13 weeks of age (Figur e 8A, B, C) or in the cross-sectioned capillaries at 19-20 weeks of age (Figur e 8D, E). Vascular leakage of EB was distinguished in G93A mice with early disease symptoms (Fi gure 8F, G) and at end-stage of disease (Figure 8H, I, J) when more EB extravasation was noted. Interestingly, vessel permeability was perceived in both initial (Fi gure 8F) and late (Figure 8I) symptomatic G93A mice. In the lumbar sp inal cord, EB dye was observe d intravascularly in control C57BL/6J mice at 12-13 weeks of age (Figur e 9A, B) and 19-20 weeks of age (Figure 9C, D), similar to results in the cervical spinal cord, while EB extravasation abnormalities were found in G93A mice at 13 weeks of age (Figure 9E, F). Significant EB diffusion into the parenchyma of the lumbar spinal cord from many blood vessels was detected in G93A mice at end-stage of disease, 17-18 weeks of age (Figure 9G, H). Considerable vessel permeability was seen in early (Figure 9F) as well as late (Figure 9G) symptomatic G93A mice. Thus, vascular leakage of EB was detected in G93A mice not only at endstage of disease but also concurre nt with early disease symptoms. 3.3.4. Immunochistochemical characteristics of basement membranes, endothelial cells, and astrocytes in the spinal cords of G93A mice Immunofluorescent staining for laminin-1 (major non-collagenous basement membrane glycoprotein) demonstrated well or ganized microvasculature networks in the cervical and lumbar ventral horn of the spinal cords in C57BL/6J mice at 12-13 weeks of age (Figure 10A, Figure 11A) and 19-20 weeks of age (Figure 10B, Figure 11B). In these control animals, many capillaries of different calibers were visibl e by laminin detection.

PAGE 67

51 Conversely, a marked reduction of labeled ve ssels was observed in both cervical and lumbar spinal cords of G93A mice at early (Figure 10C, Figure 11C) and late (Figure 10D, Figure 11D) stages of disease, suggesting a loss of vascularizat ion or disruption of vascular basement membrane integrity. In so me cervical spinal cord sections of G93A mice with initial disease symp toms, blurred spots around capillaries were apparent in the ventral horn of the spinal co rd (Figure 10C), probably due to vascular leakage. Immunofluorescent staining for Glut-1 (g lucose transporter 1) showed high expression of Glut-1 in microva scular endothelia of the cervi cal and lumbar ventral horn of the spinal cords in C57BL/ 6J mice at 12-13 weeks of age (Figure 12A, B, H, I) and at 19-20 weeks of age (Figure 12C, J). The Glut-1 immunoreaction of endothelial cells lining the many capillaries of different calibers was low, or none, in both cervical and lumbar spinal cords of G93A mi ce at early (Figure 12D, E, K, L) and late (Figure 12F, G, M, N) stages of disease. Immunohistochemically, endothelial cells (CD146) and astrocytes (GFAP) were of normal appearance in the cervical and lumb ar ventral horn of the spinal cords in the control C57BL/6J mice at 1213 weeks of age (data not sh own) and 19-20 weeks of age (Figure 13A, B; Figure 14A, B, C). Delin eated astrocytes and astrocytes with perivascular end-feet on the vessel wall we re clearly observed. However, endothelia surrounding capillaries were partially revealed in the cervical spinal cord of G93A mice at initial (Figure 13C, D) or late (Figure 13E, F) stages of disease. Although CD146 antigen expression was detected in some e ndothelial cells of early symptomatic G93A mice, indistinct immunoreaction for CD146 wa s observed in G93A mice at end-stage of disease. Additionally, swollen endothelial ce lls (Figure 13F) were found in the cervical spinal cord of late symptomatic G93A mice. A small number of delineated astrocytes were also noted. In the lumbar spinal cord of G93A mice at initial (Figure 14D, E) and late (Figure 14F, G) stages of disease, decreased or unclear CD146 antigen expression was noted in endothelial cells. Notably, increased astrocyte activation in both cervical (Figure 13F) and lumbar (Figure 14F, G) spinal cords was detected in G93A mice at late stage of disease.

PAGE 68

52 3.4. Discussion In the present study, we investigated th e functional competence of the BSCB in G93A SOD1 mice modeling ALS at differe nt stages of disease. We observed microscopic evidence of BSCB impairment in ce rvical and lumbar spinal cords, areas of motor neuron degeneration, of ALS mice at in itial disease symptoms and, more severely, at late stage disease. Our data show EB leakage in cervical/lumbar spinal cord microvessels in G93A mice at early (13 week s of age) and late (17-18 weeks of age) stage disease. More leakage was found in lumb ar spinal cords of mice at terminal stage disease. Additionally, basement membrane di sruption was noted at both early and late stage disease, as shown by the loss of laminin staining in the G93A mice. Downregulation of Glut-1 and CD146 expressions in spinal cord endothelial cells was also found in G93A mice at early and late stage disease and may relate to altered endothelial lining leading to vascular leakage. Small numbers of delineated astrocytes were also established. These results conf irm our previous ultrastructural findings (Garbuzova-Davis et al., 2007b) on disruption of the BSCB showing functional incompetence of BSCB structural elements in ALS mice. Significant death of motor neurons in G93A mice occurs at the onset of clinical disease (90 days) and by end-stage disease (136 days), mice show up to a 50% loss of cervical and lumbar motor neurons (Chiu et al., 1995; Gurney et al., 1994; Weydt et al., 2003). In G93A mice, motor deficits have been observed in tests of muscle strength and coordination as early as 8 w eeks of age (Barnoud et al ., 1997). These results extend those of our previous stud ies (Chen, 2003; Garbuzova-Dav is et al., 2001; GarbuzovaDavis et al., 2002), showing initial signs of disease, such as tremor, weight loss, and reduced hindlimb extension, in G93A mice at about 13 weeks (90 days) of age. At this age, numerous vacuolizated motor neurons were found in the cervical lumbar spinal cord and most motor neurons in the lumbar spin al cord showed signs of degeneration. The primary BBB/BSCB function is contro l of the CNS homeostasis by selective transport of molecules and cells from the systemic compartment. Substances with a molecular weight higher than 400 Da gene rally cannot cross th e barriers by free

PAGE 69

53 diffusion. However, certain endogenous larg e molecules, such as insulin, leptin, transferrin, and insulin-like growth factors, enter the br ain from blood via specific endothelial carrier-mediated or receptor-media ted transporters (revi ewed in (Ballabh et al., 2004; Pardridge, 2005, 1999, 1988)). Recently, Ig G was detected in the perikarya of motor neurons of the lumbar spinal cord in mice 24 hours after intraperitoneal injection of IgG derived from sera of ALS patients (E ngelhardt et al., 2005a). The injected IgG was found in the axon terminals of the lumbar ventral horn motor ne urons, localizing in the microtubules and rough endoplasmic retic ulum. Furthermore, IgG was similarly detected in spinal cord motor neurons of ALS patients. There was also evidence of IgG intake in endothelial cells in affected areas of the spinal cord in both ALS patients and mice injected with human ALS IgG. Engelhar dt et al. (Engelhardt et al., 2005a) suggest that there may be multiple antibodies targeti ng a variety of epitopes of motor neurons in ALS. The molecular weight of IgG is 150,000 Da and it is unlikely that these molecules could cross an intact brain cap illary endothelium even by re ceptor-mediated transcytosis. However, Pirttila et al. (Pirtt ila et al., 2004) showed that insulin-like growth factor (IGF)1, IGF binding protein-2, or nitric oxide were not elevated in CSF of ALS patients, suggesting that there is not a major di sruption in the BCSFB. In another report (Kirkinezos et al., 2004 ), an ALS mouse model with a permissive BBB was created by crossing G93A mouse with the mdr1a/b knockout mouse and showed that cyclosporine A (CsA), which cannot cross an intact BBB, BSCB or BCSFB, reached the CNS when delivered intraperitoneally into this combin ed mouse model. Since the authors did not investigate BBB or BSCB condition in the original transgenic G93A mice, it is possible that disruption or dysfunction of these barr iers occur in ALS. The 1200 Da molecular weight of CsA is much smaller than that of IgG. Further investig ation is needed to resolve this apparent discrepancy in ability to cross the BBB/BSCB. Our finding of Evans blue extravasion in early symptomatic G93A mice may suggest that large molecules such as IgG a nd other blood proteins a ppear in the spinal cord due to vascular leakage, one po ssible mechanism accelerating motor neuron damage. However, it is unclear if BBB/BSCB disruption appears prior to motor neuron degeneration or as result of motor neuron dysfunction. Al so, differences between the

PAGE 70

54 BBB and BSCB in endothelial protein concentr ations may impact observed pathological changes in G93A mice. It has been shown that microvascular endothelial cells, isolated from murine spinal cord, morphologically similar to BBB endothelial cells, express reduced amounts of several prominent BBB prot eins such as tight junction-associated proteins ZO-1 and occluding, a dherens junction-associated pr oteins beta-catenin and VEcadherin, and the efflux transporter P-gl ycoprotein (Racke et al., 2006). Reduction in immunofluorescent labeling of basement membrane of affected G93A mice suggests possible membrane disrup tion. The basement membrane is part of the extracellular matrix and is composed of collagens, proteoglycans, elastin and several glycoproteins, of which laminin is the mo st abundant (Dermietzel and Krause, 1991; Rutka et al., 1988). Reduced laminin labeling wa s observed in cervical and lumbar spinal cords of both early and late symptomatic G 93A mice, possibly indi cating vascularization changes leading to capillary wall permeability. Interestingly, Ono et al. (Ono et al., 1998) showed fragmented and widely separated collagen bundles in capilla ries and decreased amounts of collagen in postmortem posterior ha lf of the lateral funiculus and in the anterior horn of cervical enlargements fr om patients with sporadic ALS. Although the role of these aberrations in the pathogenesis of ALS remains to be determined, the authors suggested that abnormalities of collagen in the perivascular spaces of capillaries may be secondary to neuronal degenera tion as a underlying mechanism in ALS. It is well known that glucose transpor t through the BBB (BSCB) is mediated by glucose transporter isoform 1 (G lut-1) (Gerhart et al., 1989; Pardridge, 1991). Glut-1 is associated mainly with the brain capillar y endothelial cells and is asymmetrically distributed between the luminal and ablumi nal membranes (Farrell and Pardridge, 1991). This asymmetric intracellular pool of glucose transporter ma y provide for rapid transport of glucose across the abluminal plasmalemma to the brain parenchyma (Stewart et al., 1994). The alteration of Glut-1 may be relate d to the pathogenesis of microvascular permeability as has been shown, for example, in cerebral edema (reviewed in (Dwyer et al., 2002)). In the present study, we found low a nd mostly absent expression of Glut-1 in capillaries of both cervical and lumbar spinal cords of G93A mice at early and late stages of disease. This downregulation of Glut-1 expression in the endothelial cells of the BSCB

PAGE 71

55 may be related to altered endothelial lining leading to vascular leakage. Alternatively, decreased Glut-1 expression may result from aggravated alterations of the BSCB in G93A mice. Although additional experiments su ch as quantitative analysis of Glut-1 distribution and density in th e endothelial plasma membranes are needed to elucidate the regulatory mechanisms of Glut-1 expression in the spinal co rd, the present study indicates that alteration of Glut-1 could be involved in the pathogenesis of ALS. Another of our findings was that endothel ia surrounding capillaries were partially revealed by CD146 antigen expression in the ce rvical and lumbar spinal cords of G93A mice at initial and, more markedly, at late st ages of disease. More over, small numbers of delineated astrocytes were established. These results may indicate that degeneration or, at least, partial dysfunction, of non-neuronal ce lls in ALS occurs. Evidence of widespread inflammatory reactions in ALS already ex ists. The presence of monocyte/macrophage cells, activated microglia, and reactive astrocytes was established in th e spinal cord tissue of most ALS patients (Engelh ardt et al., 1993b; Henkel et al., 2004; Ince et al., 1996; Sasaki et al., 2000; Troost et al., 1990b). In a mouse model of ALS, immune/inflammatory responses (Hall et al ., 1998b; Hensley et al., 2006) are present even before any evidence of motor dysf unction (Alexianu et al ., 2001; Almer et al., 2001). Strategically, astrocytes are located at the interface between the blood vessels and the brain as well as in the spinal cord parenchyma, influencing both the entry of blood cells into the CNS and the activity of inva ding cells once they have entered the brain parenchyma (reviewed in (Aschner, 1998; Mu cke and Eddleston, 1993)). It has been shown that activated astrocytes and microglia in response to signals derived from the immune system or generated within the CNS produce various inflammatory molecules that may increase the permeability of the e ndothelial cell barrier (Prat et al., 2001). Inhibition of microglia activation, for example, as recently shown in vitro and in vivo using minocycline, may protect the brain after ischemic stroke by improving BBB viability and integrity (Yenari et al., 2006). It is possible that glial cell activation in ALS could lead to vessel leakage. Additionally, decreased numbers of delineated astrocytes and their perivascular end-feet at the blood capillaries could affect vessel permeability.

PAGE 72

56 Thus, our results confirm our previous u ltrastructural findings on disruption of the BSCB showing functional incompetence of BSCB structural elements in ALS mice. A breakdown in the BSCB is clearly indicated by EB leakage in cervical/lumbar spinal cord microvessels in G93A mice at early and late stages of disease. Laminin labeling suggests that basement membrane of vessels in the spinal cords of the dis eased G93A mice may be affected. Additionally, downregulation of Glut-1 and CD146 expressions in the endothelial cells of the BSCB may be related to altered en dothelial lining leading to vascular leakage. Degeneration of astr ocytes could influence BSCB integrity. Importantly, is BSCB breakdown a primary or secondary mechanism to motor neuron degeneration in G93A mice? Demonstrati ng BSCB disruption prior to the onset of disease symptoms and other pathological pr ocesses would indicate that BSCB disruption plays a primary role in ALS pathogenesis. 3.5. Materials and Methods 3.5.1. Animals All described procedures were approved by the Institutional Animal Care and Use Committee at USF and conducted in compliance with the Guide for the Care and Use of Laboratory Animals Transgenic male mice B6SJL-TgN (SOD1-G93A) 1GUR (G93A; Jackson Laboratories), over-express ing human SOD1, carrying the Gly93 Ala mutation, were used. Fourteen mice at 7 weeks of ag e were assessed on se nsitive indicators of degenerative state (body weight and extension reflex). Six C57BL/6J male mice of the same age were used as controls. All mice we re maintained on a 12:12 h dark:light cycle (light on at 06:00 hours). Room temperature was 23 C. Food and water was available ad libitum.

PAGE 73

57 3.5.2. Characteristics of disease progression Body weight was measured weekly throughout the study. Extension reflex was also observed weekly, as we previously described (Garbuzova-Davis et al., 2003a; Garbuzova-Davis et al., 2003b). Briefly, extens ion of the hindlimbs was observed while the mouse was suspended by its tail. A score (0-2) was given to each mouse indicating: normal (2), partial (1) or absent (0) hi ndlimb extension. Eight G93A mice were euthanatized at initial signs of disease (tremor, weight loss, and reduced hindlimb extension) at approximately 13 weeks of age, while six we re allowed to reach the endpoint of hindlimb paralysis at 17-18 weeks of age, when they were also euthanatized. Measures of body weight and extension refl ex were recorded for C57BL/6J mice according to the same schedule as G93A mice. C57BL/6J mice were euthanatized at 1213 weeks (n=3) or 19-20 weeks (n=3) of age. 3.5.3. Evans Blue dye Evans Blue dye (EB, Aldrich Chemical ), 961 Da, was used as a tracer for assessing BBB disruption (Saria and Lundber g, 1983). G93A mice with initial or late stage disease and C57BL/6J mice were intravenously injected with 2% EB in saline solution via the jugular vein 30-40 min prior to euthanasia. This su rgical procedure was performed as we previously described (Garbuzova-Davis et al., 2003b). Briefly, mice were anesthetized with Isoflurane deliv ered using a calibrate d vaporizer equipped induction chamber and nose cone and administered at 2-5% in O2 (2 L/min) to induce anesthesia and then decreased to 2% to maintain the anesthesia. The jugular vein was exposed and isolated using blunt dissection. Th e vein was ligated and a 31-gauge needle, attached to a 100-l Hamilton syringe, was plac ed into the lumen of the vein and sutured in place. Evans Blue was delivered (0.2 ml/100g, 40l/mouse) during 2 min. The needle was withdrawn, the suture tightened, a nd the incision clos ed with Vetbond.

PAGE 74

58 3.5.4. Euthanatasia and tissue preparation Euthanatasia of all mice was achieved unde r deep pentobarbital anesthesia and perfusion was not performed to avoid mech anical disruption of blood capillaries. The cervical/lumbar spinal cords were removed, fi xed in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB), pH 7.2, and then cryoprotected in 20% sucrose in 0.1 M PB (pH 7.2) overnight. Coronal sections of the cervical (C2-C3) and lumbar (L4-L5) spinal cords were cut at 30 m in a cryostat. 3.5.5. Immunofluorescence staining For identification of EB leakage, serial tissue sections of the spinal cords were thaw-mounted on slides, washed with deioni zed water to remove the freezing medium, and then rinsed several times in phosphatebuffered saline (PBS). The slides were coverslipped with Vectashi eld (Vector) and examined under an epifluorescence microscope. Some spinal cord tissues were used for immunofluorescent analysis of the basement membrane (laminin), endotheli al cells (Glut-1, CD146), and astrocytes (GFAP). Briefly, sections of the cervical/lumbar spinal cords were labeled with rat antimouse laminin-1 ( and chains) monoclonal antibody (1 :50, Chemicon), rabbit antimouse Glut-1 polyclonal anti body (1:100, Alpha Diagnostic Int. ), or double-stained with mouse monoclonal CD146 (1:30, Chemicon) and rabbit polyclonal antibodies against glial fibrillary acidic prot ein (GFAP, 1:500, Dako). The next day, the slides were incubated for 2 hrs with appropriate secondary antibodies conjugated to either rhodamine (1:1500, Alexa 594, Molecular Probes) or FITC (1:500-700, Alexa 488, Molecular Probes) and, after several rins es in PBS, coverslipped with Vectashield or Vectashield with DAPI (Vector) and examined under epifluorescence using an Olympus BX60 microscope.

PAGE 75

59 3.5.6. Staining of motor neurons in the spinal cord Coronal sections of the C2-C3 and L4-L5 spinal cords were rinsed in PBS and then stained with 0.1% cresyl violet (30 sec) for routine histological analysis of motor neurons. Sections were then washed seve ral times in deionized water, air-dried, dehydrated, and coverslipped using Permount. 3.5.7. Statistical analysis Data are presented as means SEM. The nonparametric Mann-Whitney unpaired test was used to compare medians. Acknowledgments: This work was supported by the USF Internal Awards Program.

PAGE 76

3.6. Figures 60

PAGE 77

61 Figure 5. Characteristics of disease progression in G93A mice (A) Body weight and (B) extension reflex of G93A and control C 57BL/6J mice. G93A mice at about 13 weeks of age showed initial signs of disease such as weight loss and reduced hindlimb extension. Terminal stage of disease wa s observed at 17-18 weeks of age, as demonstrated by complete hindlimb paralysis, significant reduction of body weight and absence of hindlimb extension. Arrows indicate the age of mice when euthanatasia was performed. The five pointed star in A indicates difference (p =0.06) in body weights between G93A mice at 13 weeks of age and 18 weeks of age; the four pointed star indicates a significant diffe rence (p=0.007) in body weights between G93A and C57BL/6J mice at 18 weeks of age. The four pointed star in B indicates a significant difference in extension reflex (p<0.001) be tween G93A mice at 13 weeks of age and 18 weeks of age.

PAGE 78

62

PAGE 79

63 Figure 6. Motor neurons in the cervical spinal cord of G93A mice at ea rly and late stage of disease (cresyl violet staining). In the cervical spinal cord, many healthy motor neurons with large soma and neuritic proce sses were identified in the control C57BL/6J mice at (A) 12-13 weeks of age and (B) 19-20 weeks of age. In G93A mice, numerous motor neurons with vacuoliza tion (asterisks) were found at (C) 13 weeks of age and (D) decreased numbers of motor neurons were not ed in 17-18 week old mice. Motor neurons of various sizes displayed vacuolization (asterisks). S cale bar on left side is 200 m, right side is 50 m.

PAGE 80

64

PAGE 81

65 Figure 7. Motor neurons in the lumbar spinal cord of G93A mice at ea rly and late stage of disease (cresyl violet staining) In the lumbar spinal cord, C57BL/6J mice at (A) 12-13 weeks of age and (B) 19-20 weeks of age showed nume rous motor neurons with strong Nissl body staining. Most degenerated or swo llen motor neurons (asterisks) were found in G93A mice at (C) early (13 weeks of age) and (D) late (17-18 weeks of age) stages of disease; most surviving motor neurons we re small. Scale bar on left side is 200 m, right side is 50 m.

PAGE 82

66

PAGE 83

67 Figure 8. Evans Blue fluorescence in the cervical sp inal cord of G93A mice at early and late stages of disease. In the cervical spinal cord, EB was clearly det ected within the blood vessels (red, arrowheads) in the control C57BL/6J mice at (A, B, C) 12-13 weeks of age or (D, E) in the lumen of vessels (brillian t green) at 19-20 weeks of age. In G93A mice, vascular leakage of EB (red, arrows) was detected (F, G) at early (13 weeks of age) disease symptoms and (H, I, J) at end-stage of disease (17-18 weeks of age) when more EB extravasation was seen. Arrowheads in F and I indicate vessel permeability. Scale bar in A-J is 25 m.

PAGE 84

68

PAGE 85

69 Figure 9 Evans Blue fluorescence in the lumbar spinal cord of G93A mice at early and late stages of disease. In the lumbar spinal cord, EB dye (red, arrowheads) was determined intravascularly in the control C57BL/6J at (A, B) 12-13 weeks of age and (C, D) 19-20 weeks of age similar to the cervical spinal cord. EB extr avasation abnormalities were found in G93A mice at (E, F) 13 week s of age (red, arrows). (G, H) Significant EB diffusion (red, arrows) into the parenchyma of the lumbar spinal cord from many blood vessels was detected in G93A mice at e nd-stage of disease (1 7-18 weeks of age). Arrowheads in F and G indicate vessel permeability. Scale bar in A-H is 25 m.

PAGE 86

70

PAGE 87

71 Figure 10. Immunofluorescence staining for laminin in the cervical spinal cord of G93A mice at early and la te stages of disease Many blood vessels of different diameter were immunoreactive for laminin-1 (red ) in the control C57BL/6J mice at (A) 12-13 weeks of age and (B) 19-20 weeks of age. In G93A mice at (C) initial or (D) late stages of disease, capillaries appear to be le ss numerous. In some early symptomatic G93A mice, (C) blurry spots around capillaries were found. The nuclei in A-D are shown with DAPI. Scale bar in A, B, C, D is 200 m; inserts a, b, c, d is 50 m.

PAGE 88

72

PAGE 89

73 Figure 11. Immunofluorescence staining for laminin in the lumbar spinal cord of G93A mice at early and late stages of disease Various laminin-positive vessels (red) were observed in the control C57BL/6J mice at (A) 12-13 weeks of age and (B) 19-20 weeks of age similar to cervical spinal cord resu lts. Fewer blood vessels were labeled in G93A mice at (C) early or (D) end-stage of dis ease. The nuclei in A-D are shown with DAPI. Scale bar in A, B, C, D is 200 m; inserts a, b, c, d is 50 m.

PAGE 90

74

PAGE 91

75 e 0 was found in G93A ice at (K), (L) early or (M), (N) end-stage of disease. The nuclei in H-N are shown ith DAPI. Scale bar in J, M N is 50 m; H I K L is 25 m. Figure 12 Immunofluorescence staini ng for Glut-1 in the cervical and lumbar spinal cords of G93A mice at early and late stages of disease Cervical spinal cord. High expression of Glut-1 (red) was determined in endothelial lining of many blood vessels of various diameters in the cervical spinal cord of the contro l C57BL/6J mice at (A), (B) 1213 weeks of age and (C) 19-20 weeks of age. In G93A mice at (D), (E) initial or (F), (G) late stages of disease, immunoreaction for Glut -1 in the endothelial cells appear to b low, or nonexistent. The nuclei in A-G are shown with DAPI. Scale bar in A and E is 5 m; B C D F G is 25 m. Outline of white dots indicates configuration of blood vessels. Lumbar spinal cord. Similar to the cervical spin al cord, most Glut-1-positive endothelial cells (red) were observe d in the control C57BL/6J mice at (H), (I) 12-13 weeks of age and (J) 19-20 weeks of age. Less Glut-1 expression m w

PAGE 92

76

PAGE 93

77 cytes l or (E, F) late stages of disease. Note: increased astrocyte activation the cervical spinal cord (F asterisks ) was detected in G93A mice at late stage of isease. The nuclei in A, C, and E are shown with DAPI. Scale bar in A, C, E is 50 m; D, F is 25 m. Figure 13. Immunohistochemical staining for endothe lial cells (CD146) and astro (GFAP) in the cervical spinal cord of G93A mice at early and late stages of disease. (A, B) Normal appearance of endothelial cells (green, arrowheads) and delineated astrocytes (red, asterisk) was observed in the control C57BL/6J mice at 19-20 weeks of age. Endothelia (green, arrowheads) surrounding capillaries were partially revealed in G93A mice at (C, D) initia in d B

PAGE 94

78

PAGE 95

79 B, arrowh eads) was observed. Note: increased astrocyte activation the lumbar spinal cord (F, G asterisks ) was detected in G93A mice at late stage of isease. The nuclei in A, C, D, and F are shown with DAPI. Scale bar in A, C, D, F is 50 m; B, E, G is 25 m. Figure 14. Immunohistochemical staining for endothe lial cells (CD146) and astrocytes (GFAP) in the lumbar spinal cord of G93A mice at early and late stages of disease (A, C) Similar to cervical spinal cord, endothelial cells (green, arrowheads) and astrocytes (red, asterisk) in C57BL/6J mice at 19-20 w eeks of age appeared normal. In G93A mice at (D, E) early or (F, G) end-stage of disease, decreas ed CD146 antigen expression by endothelial cells (green, in d

PAGE 96

80 .7. References lexianu ME. The role of immune pro cesses in amyotrophic lateral sclerosis mouse model of familial LS correlates with disease progr ession. Neurology, 2001; 57: 1282-9. P, Chen C, Przedborski Increased expression of the pro-in flammatory enzyme cyclooxygenase-2 in nnunziata P, Volpi N. High levels of C3c in the cerebrospinal fluid from amyotrophic postolski S, Nicolic J, Bugarski-Prokopljevi c C, Miletic V, Pavlovic S, Filipovic S. ppel SH, Smith RG, Engelhardt JI, Stef ani E. Evidence for autoimmunity in onses in the CNS. euroToxicol., 1998; 19: 269-82. 2002; 9: 113-24. allabh P, Braun A, Nedergaard M. The bl ood-brain barrier: an overview structure, arnoud P, Lolivier J, Sanger DJ, Scatton B, Moser P. Quantitative motor assessment in s involved in motor euron degeneration in ALS. A nnu. Rev. Neurosci., 2004; 27: 723-49. antigens in cultured human mbilical cord blood cells. Exp Neurol, 2003; 181: 89. rney ME. Age; 91: 6-14. 3 A pathogenesis. Rom J Neurol Psychiatry, 1995; 33: 215-27. Alexianu ME, Kozovska M, Appel SH. Immune reactivity in a A Almer G, Guegan C, Teismann P, Naini A, Rosoklija G, Hays A S amyotrophic lateral sclerosis. Ann Neurol, 2001; 49: 176-85. A lateral sclerosis pati ents. Acta Neurol. Scand., 1985; 72: 61-4. A Serum and CSF immunological findings in ALS. Acta Neurol. Scand., 1991; 83: 96-8. A amyotrophic lateral sclerosis. J Neurol Sci, 1994; 124 Suppl: 14-9. Aschner A. Astrocytes as mediators of imm une and inflammatory resp N Avata C, Ropper AH. Ischaemic brain oede ma. J. Clin. Neurosci., B regulation, and clinical implications Neurobiol. Dis ., 2004; 16: 1-13. B FALS mice: a longitudinal study. NeuroReport, 1997; 8: 2861-5. Bruijn L, Miller TM, Cleveland DW. Unraveling the mechanism n Chen N, Song, S., Hudson, J.E., Willing, A.E., Garbuzova-Davis, S., Sanches-Ramos, J.,Sanberg, P.R., Zigova, T. Neural and he matopoietic u Chiu AY, Zhai P, Dal Canto MC, Peters TM, Kwon YW, Prattis SM, Gu dependent penetrance of disease in a tran sgenic mouse model of familial amyotrophic lateral sclerosis. Mol Ce ll Neurosci, 1995; 6: 349-62. Claudio L. Ultrastructural f eatures of the blood-brain ba rrier in biopsy tissue from Alzheimer's disease patients. Acta Neuropathol. (Berl.), 1996

PAGE 97

81 leveland DW, Rothstein JD. From Charcot to Lou Gehrig: deciphering selective motor ermietzel R, Krause D. Molecular anatom y of the blood-brain barrier as defined by onnenfeld H, Kascsak RJ, Bartfeld H. Deposits of IgG and C3 in the spinal cord and wyer D, Vannucci SJ, Simpson IA. Expre ssion, regulation, and functional role of ngelhardt B. Regulation of immune cell entr y into the central nervous system. Results ngelhardt JI, Appel SH. IgG reactivity in the spinal cord and motor cortex in ngelhardt JI, Soos J, Obal I, Vigh L, Siklos L. Subcellular localizat ion of IgG from sera myotrophic lateral sclerosis. Arch Neurol, 1993; 50: 30-6. lly and ablumenal membranes: an electron icroscopic immunogold study. Proc Natl Acad Sci USA, 1991; 88: 5779-83. odeling ALS. Brain Res., 2007; 1157: 126-37. h apparent otor deficit. Amyotroph Lateral Scler Ot her Motor Neuron Diso rd, 2001; 2: 175-80. model & Stem Cell Research, 2003a; 2: 255-70. C neuron death in ALS. Nat Rev Neurosci, 2001; 2: 806-19. D immunocytochemistry. Int. Re v. Cytol., 1991; 127: 57-109. D motor cortex of ALS patients. J. Neuroimmunol., 1984; 6: 51-7. D glucose transporters (GLUTs) in br ain. Int Rev Neurobiol, 2002; 51: 159-88. E Probl. Cell Differ., 2006; 43: 259-80. E amyotrophic lateral sclerosis. Arch. Neurol., 1990; 47: 1210-6. E of ALS patients in the nervous system Acta Neurol Scand, 2005; 112: 126-33. Engelhardt JI, Tajti J, Appel SH. Lymphocy tic infiltrates in the spinal cord in a Farrell CL, Pardridge WM. Blood-brain barrie r glucose transporter is asymmetrica distributed on brain capillary endothelial lu minal m Garbuzova-Davis S, Haller E, Saporta S, Kolomey I, Nicosia SV, Sanberg PR. Ultrastructure of blood-brai n barrier and blood-spinal co rd barrier in SOD1 mice m Garbuzova-Davis S, Willing AE, Milliken M, Saporta S, Sowerby B, Cahill DW, Sanberg PR. Intraspinal implantation of hNT neurons into SOD1 mice wit m Garbuzova-Davis S, Willing AE, Milliken M, Sa porta S, Zigova T, Cahill DW, Sanberg PR. Positive effect of transplantation of hN T neurons (NTera 2/D1 cell-line) in a of familial amyotrophic lateral scle rosis. Exp Neurol, 2002; 174: 169-80. Garbuzova-Davis S, Willing AE, Zigova T, Sa porta S, Justen EB, Lane JC, Hudson JE, Chen N, Davis CD, Sanberg PR. Intravenous Administration of Human Umbilical Cord Blood Cells in a Mouse Model of Amyotrophic Lateral Sclerosis: Distribution, Migration, and Diffenentiation. Journal of He matotherapy 1

PAGE 98

82 n odel of ALS: Distribution, migration and differe ntiation. Journal of ematotherapy and Stem Cell ResearcH, 2003b: 255-70. c ose transporter localization brain using light and el ectron immunocytochemistry. J Neurosci Res, 1989; 22: 464rrier function nd activated astrocytes in the rat. Pain, 2006; 124: 211-21. Kwon YW, Deng HX, et al. Motor neuron degeneration in mice that express a uman Cu,Zn superoxide dismutase mutation. Science, 1994; 264: 1772-5. ic activation que 004; 55: 221-35. 3A-SOD1 ouse model of amyotrophic lateral sclerosi s. Antioxid Redox Signal, 2006; 8: 2075-87. mutation in exon 4 of the Cu/Zn s uperoxide dismutase gene: pathological and munocytochemical changes. Acta Ne uropathol (Berl), 1996; 92: 395-403. athol, 1992; 140: 69107. l l calibers of vessel and associ ated with blood-brain barrier akage and active demyelinati on. J. Pathol., 2003; 201: 319-27. ood-brain barrier benefits fr om systemic cyclosporine A treatment. J eurochem, 2004; 88: 821-6. Garbuzova-Davis S, Willing AE, Zigova T, Sa porta S, Justen EB, Lane JE, Hudson JE, Hart CD, Sanberg PR. Intravenous administra tion of human umbilical cord blood cells i a mouse m H Gerhart DZ, Le Vasseur RJ, Br oderius MA, Drewes LR. Glu in 72. Gordh T, Chu H, Sharma HS. Spinal nerve le sion alters blood-spinal cord ba a Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, Alexander DD, Caliendo J, Hentati A h Hall ED, Oostveen JA, Gurney ME. Relationshi p of microglial and astrocyt to disease onset and progression in a transg enic model of familial ALS. Glia, 1998; 23: 249-56. Henkel JS, Engelhardt JI, Siklos L, Simpson EP, Kim SH, Pan T, Goodman JC, Siddu T, Beers DR, Appel SH. Presence of de ndritic cells, MCP-1, and activated microglia/macrophages in amyotrophic lateral sc lerosis spinal cord tissue. Ann Neurol, 2 Hensley K, Mhatre M, Mou S, Pye QN, St ewart C, West M, Williamson KS. On the relation of oxidative stress to neuroinflamma tion: lessons learned from the G9 m Ince PG, Shaw PJ, Slade JY, Jones C, Hudgs on P. Familial amyotrophic lateral sclerosis with a im Kawamata T, Akiyama H, Yamada T, McGeer PL. Immunologic reactions in amyotrophic lateral sclerosis brain and spinal cord tissue. Am J P 7 Kirk J, Plumb J, Mirakhur M, McQuaid S. Tight junctional abnormality in multiple sclerosis white matter affects a le Kirkinezos IG, Hernandez D, Bradley WG, Moraes CT. An ALS mouse model with a permeable bl N

PAGE 99

83 Neurol., 1994; 3: 625-36. of J. Neurol. Sci., 1993; 118: 73-8. ohamed HA, Mosier DR, Zou LL, Siklos L, Alexianu ME, Engelhardt JI, Beers DR, Le transmitter release in motor eurons. J. Neurosci. Res., 2002; 69: 110-6. entral nervous system. FA SEB J., 1993; 7: 1226-32. 97-203. M. Collagen abnormalities in the spinal cord from patients with myotrophic lateral sclerosis. J. Neurol, Sci., 1998; 160: 140-7. 1; 2: 419-26. barrier biology and methodology. J. NeuroVirol., 1999; 5: 56-69. Kwon EE, Prineas JW. Blood-brain barrier abnormalities in longstanding multiple sclerosis lesions. An immunohistochemical study. J. Neuropathol Exp 5 Leonardi A, Abbruzzese G, Arata L, Cocito L, Vische M. Cerebrospinal fluid (CSF) findings in amyotrophic lateral sclero sis. J. Neurol., 1984; 231: 75-8. Meucci G, Rossi G, Bettini R, Montanaro D, Gironelli L, Voci L, Bianchi F. Laser nephelometric evaluation of albumin, IgG a nd a2-macroglobulin: applications to study alterations of the blood-brain barrier Miller RG, Anderson FJ, Bradley WG, Brooks BR, Mitsumoto H, Munsat TL, Ringel SP. The ALS patient care database: goals, desi gn, and early results. ALS C.A.R.E. Study Group. Neurology, 2000; 54: 53-7. M WD, Appel SH. Immunoglobulin Fc gamma r eceptor promotes immunoglobulin uptake immunoglobulin-mediated calcium increase, a nd neuro n Mucke L, Eddleston M. Astrocytes in infec tious and immune-mediated diseases of the c Noble LJ, Wrathall JR. Distribution and time c ourse of protein extravasation in the rat spinal cord after contusive in jury. Brain Res, 1989; 482: 57-66. Olsson Y, Sharma HS, Pettersson A, Ce rvos-Navarro J. Release of endogenous neurochemicals may increase vascular permeability, induce edema and influence cell changes in trauma to the spinal cord. Prog Brain Re s, 1992; 91: 1 Ono S, Imai T, Munakata S, Takahashi K, Kanda F, Hashimoto K, Yamano T, Shimizu N, Nagao K, Yamauchi a Pardridge WM. Advances in cell biology of blood-brain barrier transport. Semin. Cell Biol., 199 Pardridge WM. The blood-brain barrier and neurotherapeutics. Ne uroRx, 2005; 2: 1-2. Pardridge WM. Blood-brain 5 Pardridge WM. Recent advances in blood-brai n barrier transport. Ann. Rev. Pharmacol. Toxicol., 1988; 28: 25-39.

PAGE 100

84 h nding protein-2 or nitric oxide are not increased in MS r ALS. Acta Neurol Scand., 2004; 109: 337-41. Antel JP. G lial cell influence on the human blood-brain arrier. Glia, 2001; 36: 145-55. 2006; 533: 57-68. ous system: heaven (or hell) is in the detail s. Curr. Opin. Immunol., 2006; 8: 683-9. 4: 688-700. eripheral nervous systems: structure a nd function. J. Neurosurg., 1988; 69: 155-70. d morphological valuation of vascular permeability in animal tissues. J Neurosci Methods, 1983; 8: 41-9. M. iNOS and nitrotyrosine immunoreactivity in myotrophic lateral sclerosis. Ne urosci. Lett., 2000; 291: 44-8. -spinal cord barrier in traumatic injury and repair. urr Pharm Des, 2005; 11: 1353-89. n barrier ultrastructure. icrosc Res Tech, 1994; 27: 516-27. is: a Exp. Neurol., 2005; 64: 649-64. cs 6: 401-10. Pirttila T, Vanhatalo S, Turpeinen U, Riikone n R. Cerebrospinal fluid insulin-like growt factor-1, insulin growth factor bi o Prat A, Biernacki K, Wosik K, b Racke K, Juergens UR, Matthiesen S. C ontrol by cholinergic mechanisms. Eur J Pharmacol, Rebenko-Moll NM, Liu L, Cardona A, Rans ohoff RM. Chemokines, mononuclear cells and the nerv 1 Rowland LP, Shneider NA. Amyotrophic late ral sclerosis. New Engl J Med, 2001; 34 1 Rutka JT, Apodaca G, Stern R, Rosenblum M. The extracellular matrix of the central and p Saria A, Lundberg JM. Evans blue fluores cence: quantitative an e Sasaki S, Shibata N, Komori T, Iwata a Sharma HS. Pathophysiology of blood C Stewart PA, Hayakawa K, Farrell CL. Quantita tion of blood-brai M Strong MJ, Kesavapany S, Pant HC. The pat hobiology of amyotrophic la teral scleros proteinopathy? J. Neuropathol. Troost D, Van den Oord JJ, Vianney de Jong JMB. Immunohistochemical characteristi of the inflammatory infiltration in amyotr ophic lateral sclerosis. Neuropathol. Appl. Neurobiol., 1990; 1 Vorbrodt AW, Dobrogowska DH. Molecular an atomy of intercellular junctions in the brain endothelial and epithelial barriers: electron microsc opist's view. Brain Res. Rev., 2003; 42: 221-42.

PAGE 101

85 os CMP, Geurts JJG, Montagne L, van Haaste rt ES, B L, van der Valk P, Barkhof F, eydt P, Hong SY, Kliot M, Moller T. A ssessing disease onset and progression in the SOD1 mouse model of ALS. NeuroReport, 2003; 14: 1051-4. Yenari MA, Xu L, Tang XN, Qi ao Y, Giffard RG. Microglia potentiate damage to bloodbrain barrier constituents: improvement by minocycline in vivo and in vitro. Stroke, 2006; 37: 1087-93. V de Vries HE. Blood-brain barrier alterations in both focal and diffuse abnormalities on postmortem MRI in multiple sclerosi s. Neurobiol Dis, 2005; 20: 953-60. W

PAGE 102

86 CHAPTER 4 THE NEUROPROTECTI VE EFFECTS OF THE SIGMA RECEPTOR AGONIST DTG (1,3-DI-O-TOLYLGUANIDINE) ON A MOUSE MODEL OF ALS 4.1. Introduction Other aspects of ALS have been of inte rest to our lab including inflammation and excitotoxicity. Treatments that have been be neficial in rodent m odels of stroke, for example, have potential therapeutic benefits to ALS models. It has been shown by our colleagues (Keith Pennypacker et al.) th at in middle cerebral artery occlusion ( MCAO) sigma receptor agonists are both neuroprotec tive and anti-inflammatory (Ajmo et al., 2006). We proposed that sigma receptor activ ation will protect motor neurons from excitotoxicity and microglia mediated in flammation, the key pathological components shared by stroke and ALS. Specifically, we wi shed to test whether this class of drugs was effective at reducing motor neuron dege neration and secondary inflammation in the SOD1 G93A mouse model of ALS. Inflammatory responses are common to al l types of neuronal injury and enhance neurodegeneration (DeLegge and Smoke, 2008). This response is mediated by microglia, the endogenous macrophage in the CNS, by as trocytes, and by infiltrating immune cells, which include macrophages and lymphocytes. Resting, nonphogocytic, microglia do not contribute to neurodegenera tion. However, upon injury-i nduced activation, they can contribute to the destruction of neurons with the release of TNF(Bezzi et al., 2001; Hide et al., 2000; Noda et al., 2000), leading to the activation of the tr anscription of proinflammatory genes, such as inducible ni tric oxide synthase and proinflammatory cytokines (Gebicke-Haerter, 2001). Histological analysis of spinal sections obtained from SOD1 transgenic mice have shown a thre e fold increase in activated microglia per slice compared to their wild type littermates. Release of tumor necrosis factor alpha

PAGE 103

87 (TNF) and nitric oxide (NO) can induce glutamate release from astrocytes (Shibata and Kobayashi, 2008). Further neur odegeneration is elicited through macrophages and other immune cells infiltrating into the ALS-damaged area of the spinal cord. Sigma receptors, first identified by Martin et al. 1976, are found in numerous tissues and organs. These pleiotropic receptors have since been classified into two types on the basis of pharmacology, sigma-1 and sigma-2, (B ourrie et al., 2004). Both subtypes have high to moderate affinity for antipsychotics, such as haloperidol, and guanidines, such as DTG (1,3-di-o-tolylguanidine). Sigma-1 receptors have two putative transmembrane domains and are different than any ot her known mammalian protein. While the endogenous ligand for sigma receptors has not been clearly identifie d, various candidates have been suggested. Progesterone, for exam ple, and other neurosteroids, including pregnenolone and dehydroepiandrosterone, ha ve high affinity for sigma receptors (Maurice et al., 1997). Given th e general neuroprotective prop erties of neurosteroids and their abundance in the corte x, these sigma ligands would be obvious candidates to be investigated as possible endogenous liga nds to reduce microgliosis and ensuing inflammation. The current FDA approved therapy for ALS, Riluzole, is thought to modulate excitotoxicity by acting as a glutamate re ceptor antagonist, although use of this drug results only in a three month increase in li fespan. Riluzole treatment, however, does nothing for the secondary inflammatory processe s, which are critical in the pathology of neurodegenerative disorders su ch as ALS, AD, and stroke. Taken together, effective treatment regimens for ALS (and possibly ot her neurodegenerative disorders) should include strategies to minimize both excitotoxicity and inflammation. Recently, it has been shown that microglia express sigma-1 receptors, making DTG a molecule of therapeutic interest in ALS (H all et al., 2009). Additionally, sigma receptor activation has been proven to modulate cytoki ne progression and inhi bit the secretion of TNF, NO, and interferonby macrophages induced by lipopolysachharide (LPS) injection (Bourrie et al., 1995). Overall, DTG is an exceptional candidate for ALS treatment because it promotes neuronal viability even after neurodegenerative and inflammatory processes have already begun. Furthermore, recent experiments have

PAGE 104

88 shown that stimulation of sigma receptors decr eases the excitability of neurons. Thus, the treatment is two-fold in modulating the excitability of neurons and modulating neuroinflammation. In the following experime nts, life expectancy, weight retention and motor skills of SOD1 mice were examined following chronic systemic administration of DTG. 4.2. Materials and Methods 4.2.1. Osmotic Minipumps and subcutaneous implantation Sterile Alzet osmotic minipumps-model 2004 (Durect Corp. Cupertino, Ca.), with a flow rate of 0.25 L/hr were filled with 100mM DTG (1,3-di-o-tolylguanidinea sigma 1 receptor agonist) in 3% lactic aci d as vehicle and incubated at 37oC in sterile saline for 48 hours. All procedures involving experi mentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (90 days old25g) were anesthetized with 1-2% isoflurane and shaved and scrubbed with 10% Betadine so lution at the site of incision. A small (3 cm) incision was made, exposing the neck, and double bladed scissors were used to form a subcutaneous pocket along the back of the animal between the scapula and towards the tail for placement of the pump. 4.2.2. Mice 40 SOD1 G93A mice (C57/B6 background) and 10 NTG controls were purchased from Jackson Labs (Bar Harbor, ME.). Thes e mice express 18 copies of mutant SOD1 and develop ALS pathology at approximat ely 100 days. The average lifespan is approximately 130 days and the phenotype clos ely resembles a more aggressive familial form of the disease. The model was chosen because it was so widely accepted in the field at the time and the therapeutic/pathologica l window was short enough to run multiple studies in succession. All surg eries were performed at dis ease onset (~100 days) and run

PAGE 105

89 for 30 days or until the animal reached endpoint stage. Half of th e mice received sham infusions (3% lactic acid only). 4.2.3. Histology and Immunohistochemistry Mouse spinal cords were carefully removed and fixed in 10% formaldehyde (formalin) for 72 hours and then passed thr ough a series of sucrose solutions (10-30%) over another 72 hours. Tissues were frozen to the peltier stage (Physitemp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzer land) and sectioned coronally at 25 m. After incubating the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X-100, and 0.02% sodium azide) for 60 min, primary antibodies against SOD1 (1:500) were applied and incubated at 4o C overnight. After thorough washing, the sections were incubated with secondary antibody, Alexa 488 and 594 fluorophores (1:1000, 1:4000-Invitrogen) in the da rk at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,000, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and s ealed with Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). The images were analyzed on a Zeiss Imager Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axiovision 4.7. FluoroJade stai ning was used to identify dying motorneurons and Cresyl violet staining was used for genera l morphology of motor neurons and vacuolated mitochondria (as previously described). 4.2.4. Behavior Motor deficits and health of the animal s were assessed throughout the experiment. Body weights were taken biweekly during th e 28 day treatment as well as daily performance on a rototod test. The rotor od is designed to assess motor coordination, balance and equilibrium. The mouse is pla ced on the rod and the rotorod accelerates gradually to a pre-set speed. Latencies for the mice to fall from the rod are recorded. The rototod consists of a semi-enclosed chamber, which contains a beam made of ribbed

PAGE 106

90 plastic and flanked by round plates on either si de to prevent any escape. The rod is fixed at a height of 20 cm above the counter and the mouse is placed on t op of the beam facing in the orientation opposite to that of its rota tion, so that forward locomotion is necessary to avoid falling. The rotorod spins consisten tly at 1 of 10 pre-set speeds (#5 was chosen) for three 2-minute trials. Latencies in falling from the rod are recorded automatically once the fallen animal interrupts a laser ther eby stopping the timer. Each mouse has 3 trials with a 20 minute interval between trials Once it is evident that the mice are unable to perform, animals are removed from the study. 4.3. Results and Conclusions Although preliminary data (n=10) with DTG indicated that prophylactic treatment of SOD1 G93A mice with DTG increased life span and retained body weight compared to sham treated littermates, we were unable to confirm either observation in this study (n=40). As table 2 below indicates, the av erage lifespan for the DTG treated group was ~134 days as compared to the sham treated group which was ~130 days. This modest increase was not statistically si gnificant (figure 15). The beha vioral data (not shown) and weight retention data were similar in that no significant detectable differences could be found between the two groups. Immunohistoche mical analysis, though not performed to any great extent given the outcome of the study, indicate using Flurojad e, that most motor neurons were dead or dying by the end of th e experiment, proposed earlier as scortched earth (figure 16). That is, the damage done to the spinal cord motor neurons was so great, that a minor therapeutic benefit could not be observed. If these studies were to be repeated, administering DTG prior to disease endpoint or using a different model with a larger therapeutic window may be more effective.

PAGE 107

4.4. Figures Table 2 Lifespan and body mass of mice receiving DTG or sham infusions 91

PAGE 108

Figure 15 Graphical representation of lifespan (Y-axis) in table 2 showing modest benefit with DTG, but without significance. 92

PAGE 109

93 A B Figure 16 The scortched earth of degenerating ne urons (as well as glia) in the spinal cord of SOD1 G93A mice were detected by Fluoro-Jade staining. A) Fluoro-Jade detects tracts of degenerating neurons. B) At higher magnification, individual neurons are visible. (Borrowed from an ALSA gran t applicationPennypacker et al. 2006)

PAGE 110

94 CHAPTER 5 THE NEUROPROTECTI VE EFFECTS OF THE CARBOXYL TERMINUS OF GAMMA LAMININ, KDI, ON A MOUSE MODEL OF ALS 5.1. Introduction Because the exact cause of Amyotrophic Lateral Sclerosis (ALS) is unknown, effective treatments are yet to be developed. It is for this reason that multiple approaches for therapies have been investigated. Although ALS is a proteinopathy, glutaminergic dysregulation and motor neuron death also o ccur and therefore various neuroprotectants have been of interest. In the following expe riments, we investigated the use of laminin, namely a fragment of -laminin, as a therapy for ALS. Laminin is a heterotrimeric extracellular matrix glycoprotein involved in numerous biological proc esses. Recently, it has been elucidated that it is, in fact, the so luble tripeptide fragment-lysine, aspartic acid, and isoleucine (KDI) on th e carboxyl terminus of the -laminin chain that promotes neurite outgrowth (Liesi et al., 2001). To date, there are 15 known isoforms of laminin that are expressed in mammals in a tissue-specific manner. Five different subunits have been described, 4 and 3 which arrange as a cruciform heterotrimer ( ) of ~900KD in total mass as intact laminin molecules (Miner, 2008). The molecules are the main non-collagenous components of basement memb ranes and also play diverse roles in the extracellular matrix relati ng to cell structure, adhesion, differentiation, migration, signaling, metastasis, and neurit e outgrowth. It has also been shown that laminin is upregulated during development and repair of most tissues, and portions of the molecule can maintain biological activity following cleavage events. Some laminin cleavage products retain biological activity (Wiksten et al., 2003). Over the past decade, our collaborator, Pivi Liesi, has worked extens ively with laminin. In particular, she has elucidated that the carboxyl terminus of 1-laminin is responsibl e for neurite outgrowth

PAGE 111

95 both in vitro and in vivo (Liebkind et al., 2003; Wiksten et al., 2004b). Original work focused on the decapeptide of the molecule exerting these effects; however, subsequent studies of the hexapeptide a nd finally the tripeptide, KDI, still showed activity as a neurite outgrowth factor. KDI has also prove n to be a glutamate receptor antagonist via patch clamp studies. In these studies, it was demonstrated that KDI non-competitively inhibited ionotropic glutamate receptors (at 1mg/mL) ~100% of AMPA -alpha-amino-3hydroxy-5-methyl-4-isoxazole propionic ac id and 50% of NMDAN-Methyl-DAspartate receptors (Moykkynen et al., 2005). In vitro work on primary human embryonic spinal cord cultures and organotypic slices of human embryonic spinal cord embedded in Matrigel showed m easurable tracts of extending neurites. The most recent experiment performed by Liesi et. al. (2006) showed that rat spinal cords, following transsection, could heal with the continued administration of KDI to the site. In this experiment, glial scarring was minimal and the animals were in most cases able to completely recover from paralysis (Wiksten et al., 2004b). The molecule has thus shown promise as a therapeutic for neurodegenera tion, specifically ALS and AD, and we have subsequently investigated its effect in di fferent mouse models of neurodegenerative disease. Because it is not yet known whether KDI can cross the blood-brain barrier following systemic administration, direct infusi on of the molecule in to the cerebrospinal fluid (CSF) of these mice was required. Recent work by other groups has proven, however, that compounds in the CSF do not diffuse efficiently. It was therefore necessary to continuously infuse the peptide into the intrathe cal space (lumbar cistern) of the spinal cord of the mice. This has pr oven to be an arduous task for many research groups and was achieved here with assi stance from the ALS Therapy Development Foundation. To the authors knowledge, successf ul long-term intrathecal catheterization in mice, particularly in conjunction with Alzet osmotic minipumps, has not been reported elsewhere. The pathologies of ALS begin in the lowe r lumbar region of the spinal column and progress rostrally causing hindlimb paralysis and eventually death. A series of transgenic animals have been developed over the years to mimic the disease; the most

PAGE 112

96 frequently used is the SOD1 G93A model. This animal has a glycine to alanine substitution at the 93rd po sition of the 153 amino acid SOD1 monomer. The mice become paralyzed at approximately 100 days and live another 20-30 days with ALS-like symptoms. Due to the aggressive disease prog ression in this model, they were the most commonly used model to investigate the diseas e at the time. The use of KDI in this mouse model of ALS was promising for 2 reason s: First, because KDI was shown to be a neurite outgrowth promoter (Liebkind et al ., 2003), enhanced survival, if not MN regrowth was anticipated in these mice dying of motor neuron loss. Second, given KDIs potent ability, similar to riluzole and meman tine, to lower synaptic glutamate levels, it was reasonable to anticipate KDI might reduce excitotoxic insults in ALS. 5.2. Materials and Methods 5.2.1. Intrathecal Catheter Assembly Intrathecal catheters were fashioned as described previously by first inserting a 10cm, 0.005 diameter tungsten wire (Small Parts, Inc. Miramar, Fl) into a 3 cm length of polyethylene tubing (PE-10, Becton Dickin son, Sparks, MD.) with an internal diameter of 0.011 and stretching the materi al so it was snug with the wire. The unstretched portion of the PE-10 (outer diam eter 0.024) was next inserted into a 4cm long polyvinyl tubing (Durect Corp. Cuperti no, Ca.) (internal diameter 0.027) with approximately 2 cm of overlap. The differe nce of 0.003 in the step-up to the larger polyvinyl tubing was filled w ith Locktite 454 adhesive (Plastics One. Roanoke, VA.) and secured with a silk suture. The assemblies were then left to dry overnight and sterile water was forced through to assure that the lines were not occluded or damaged. 5.2.2. Osmotic Minipumps Sterile Alzet osmotic minipumps-model 2004 (Durect Corp. Cupertino, Ca.), with a flow rate of 0.25 L/hr were filled with 200ug (1mg/m L) of KDI (NeoMPS, San Diego,

PAGE 113

97 CA) diluted in artificial C SF (Harvard Apparatus, Holliston, MA) and incubated at 37oC in sterile saline for 48 hours. Typically, cathete rs were attached and primed at this time, but the tungsten wire needed to remain in place for stability during insertion into the spine. 5.2.3. Mice 20 SOD1 G93A mice (C57/B6 background) were purchased from Jackson Labs (Bar Harbor, ME.). These mice express 18 copies of mutant SOD1 and develop ALS pathology at approximately 100 days. The av erage lifespan is about 130 days, and the phenotype closely resembles a more aggressive familial form of the disease. This ALS mouse model was chosen because at the time it was widely accepted in the field and the therapeutic/pathological wi ndow was short enough to run high throughput studies in succession very timely. All surgeries were performed at disease onset (~100 days) and run for 28 days or until the anim al reached endpoint stage. 5.2.4. Intrathecal Infusions All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (90 days old25g) were anesthetized with 1-2% isoflurane and shaved and scrubbed with 10% Betadine so lution at the site of incision. A small (3 cm) incision was made, exposing the lumbar region of the vertebral column, and double bladed scissors were used to form a subcutaneous pocket along the back of the animal between the scapula and towards the head. A puncture was then made between the L2 and L3 vertebrae with a 23 gauge needle that allows the small, stretched portion of PE-10 (stabilized with the tungsten wire) to be in serted 0.9 cm into the intrathecal space and stopped by the silk suture used to adjoin the two different sized tubings. Accurate placement of the catheter was visualized in ev ery instance by a brief tail flick or muscle twitch, at which point the tungsten wire was removed and Alzet pump attached. Another

PAGE 114

98 silk suture was used to anchor the catheter into the superficial erector musculature dorsal the vertebral column. The placement of the second suture into the musculature immediately caudal to the first assembly st itch on the catheter prevented them from being pulled out during the course of the experiment. The pumps infused for 28 days, around which time the animals were at th e terminal stages of paralysis. 5.2.5. Histology and Immunohistochemistry Mouse spinal cords were carefully removed and fixed in 10% formaldehyde (formalin) for 72 hours and then passed thr ough a series of sucrose solutions (10-30%) over another 72 hours. Tissues were frozen to the peltier stage (Physitemp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzer land) and sectioned coronally at 25 m. After incubating the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X-100, and 0.02% sodium azide) for 60 min, primary antibodies against SOD1 (1:500) were applied and incubated at 4 o C overnight. After thorough washing, the sections were incubated with secondary antibody, Alexa 488 and 594 fluorophores (1:1000, 1:4000-Invitrogen) in the da rk at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,000, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and s ealed with Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). FluoroJade staining was used to identify dying motorneurons. The images were analyzed on a Zeiss Imager Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axiovision 4.7. 5.2.6. Behavior Motor deficits and the animals health were assessed throughout the experiment. Body weights were taken daily during the 28 day treatment as well as biweekly performance on a rototod test. The rotor od is designed to assess motor coordination, balance and equilibrium. The mouse is pla ced on the rod and the rotorod accelerates gradually to a pre-set speed. Latencies for the mice to fall from the rod are recorded.

PAGE 115

99 The rotorod consists of a semi-enclosed chamber which contains a beam made of ribbed plastic and flanked by round plates on either si de to prevent any escape. The rod is fixed at a height of 20 cm above the counter, and the mouse is placed on top of the beam facing in the orientation opposite to that of its rota tion, so that forward locomotion is necessary to avoid falling. The rotorod spins consisten tly at 1 of 10 pre-set speeds (#5 was chosen) during three 2-minute trials. Latencies for the mice to fall from the rod are recorded automatically once the fallen animal interrupt s a laser, thereby stopping the timer. Each mouse was given 3 trials with a 20 minute in terval between trials. Once it was evident that the mice were no longer able to perform the task, they were removed from the study. 5.3. Results and Conclusion KDI was of particular intere st to us as a therapeutic for ALS for several reasons. First and foremost, it was demonstrated both in vitro and in vivo that KDI was a potent neurite outgrowth factor. Prev ious work by Liesi et all dem onstrated that neurons would extend long processes when embedded in ma trigel containing KDI as compared to standard media (figure 17). More work fr om her laboratory reveal ed that KDI, following a full transection of rat spinal cords at L5-L6, could restore the animals ability to walk within a couple of weeks with drastic changes in gross mor phology of the SC (figure 18). Hematoxylin eosin staining reveled that gl ial scarring was also minimized and that neurons were actually able to grow through the wound (figure 19). Further investigation from her lab revealed, using patch clamp studi es on human neocorti cal neurons, that KDI was a potent inhibitor of glutamate receptors (both AMPA and NMDA). The only FDA approved drug for ALS, riluzole, functions by lowering synaptic levels of glutamate. Memantine is another glutamate reducing agent that has shown benefit to transgenic ALS mice. Glutamate excitotoxicity unequivocally plays a role in the pathogenesis of ALS and the therapeutic effects of these compounds supports that. We were confident that we had a small molecule of similar biological capability that would benefit the animals in two ways: reduced glutamate concentrations at synapses and promotion of axonal growth.

PAGE 116

100 As indicated in the table below (table 3), KDI had no significant effect (p=0.62) on lifespan in the treated cohort. The animals were monitored daily and given a score 110 based on their condition. Th eir ability to walk, to groom, and to control bodily functions were taken into account and a c linical endpoint was es tablished when the animal was paralyzed completely, unable to attain food and water. There were no differences between the groups in this analysis regarding the progression of the disease in theses animals (data not shown). Rota rod performance in these studies were inconclusive. Extensive hist ological and biochemical anal ysis of tissues were not warranted, at this time, based on survival a nd progression data and studies with KDI and ALS were halted. Fluorojade analysis revealed a scortched earth eff ect in the tissues of mice at the end stage in previous studies. That is, the number of dying or dead motor neurons was so high that any changes induced by KDI would have been imperceivable. Transgenic mouse models, we have found, are limited in most cases to answering the questions that they were designed for. The SOD1 G93A mouse model, for example, has about 18-22 copies of the mutant gene being expressed, which does not correlate in any way with fALS cases, let alone sporadic cases in which the inclusions have little or no WT SOD1 bound (Bruijn et al., 2004b). Moreover, using a model of protein aggregation to study glutamate reducing com pounds can be misleading in assessing the effects. There is some justification that glutamate is dysregulated in these mice based upon levels of mutant SOD1, though results have varied in that determination (Tortarolo et al., 2006) (Tovar-y-Rom o and Tapia, 2006). In retrospe ct, the phenotype in this model was so aggressive that it wa s difficult to test therapeutic compounds with it. The mice would begin to be paralyzed at 90 days of age, which was quite young for the surgeries involved. Then they would quickly decline a nd need to be euthanized as quickly as 1520 days later (statistically no different than untreated animals). The value of the SOD1 G93A mouse to the ALS research commun ity has since been questioned. Personal communication with the inventor of the S OD1 G93A transgenic mouse in July 2006 revealed that his lab no longer uses this pa rticular mouse for AL S studies supporting our decision to discontinue working with them.

PAGE 117

5.4. Figures Table 3. Lifespan of KDI-treated (blue) and sham treated (black) SOD1 G93A mice 101

PAGE 118

Figure 17 Liesi et al. 2003 (previous work and rationale for experiment) TUJ1immunoreactive (tubulin/axon labeling) neurons from the human embryonic spinal cord on top of the injured monolayers of the human spinal cord glial cells after 48 hr in vitro. In a, a neuron in the lower righ t corner of the photograph ex tends a long neurite in the presence of 0.1 g/ml of the KDI peptide in the cultu re medium. Note that the moderate TUJ1 immunoreactivity of the majority of the underlying glial cells cannot be seen because of dark exposure of the photograph. Only the intense neuronal and some more intense glial immunoreactivity can be seen. In b a spinal cord neuron fails to extend long neurites when no KDI peptide is presen t in the culture medium. Scale bar = 17 m. 102

PAGE 119

Figure 18. Liesi et al. 2004 (previous work and rationale for experiment) Stereo microscopic images of parafo rmaldehyde-fixed KDI-treated ( a, b ) and placebo-treated ( c, d ) spinal cords 14 weeks after operation. Dorsal images (a, c) show a connective tissue scar that forms at the site of laminectomy (arrow). It is obvious from these stereo images that dorsal (a) and ventral (b) poles of the spinal cord of the KDI-treated animal (motor score 3/4) are both better preserved than are do rsal (c) and ventral (d ) poles of the spinal cord of the placebotreated animal (motor score 1/1). Placebo-treated spinal cord consists of semitranslucent bloody opaque tissue at the transection site whereas KDI-treated spinal cord looks more intact. 103

PAGE 120

104

PAGE 121

105 Figure 19 Liesi et al. 2004 (previous work and rationale for experiment) Examples of longitudinally cut hematoxylin-eosin-stained paraffin sections (4 m) through the ventral spinal cord show extent of damage in KDI-treated ( ac ) and placebo-treated (df ) animals. The motor score of each animal after 14 weeks is indicated in the left upper corner of each figure. Results indicate that spinal cords were transected completely and confirm that all KDI treated spinal cords (even the one in c) show less damage and smaller scar areas than do any placebo-treate d spinal cords. KDI treatment thus reduces tissue damage and cavity formation and improves the chances for the tissue to heal. Scale bar 500 m.

PAGE 122

106 CHAPTER 6 A NOVEL TECHNIQU E FOR SIMULTANE OUS BILATERAL BRAIN INFUSIONS IN A MOUSE MODEL OF NEURODEGENERATIVE DISEASE Steven P. Bennett1,2, Tim Boyd1,2 Michelle Norden1,2,3, Jaya Padmanabhan1,2, Peter Neame2, Inge Wefes2,3, Huntington Potter1,2,3 1-Johnnie B. Byrd Sr. Alzheimers Center and Research Institute, Tampa Florida 33613 2-University of South Florida, Department of Molecula r Medicine, Tampa 33612 3-Eric Pfeiffer Suncoast Alzheimers and Gerontology Center 6.1. Abstract A common problem faced by researcher s using transgenic models to study disease is the phenotypic variabil ity that exists within a gr oup or colony of animals. Significant pathological analyses thus often require large nu mbers of mice to perform. Many lines of transgenic mice harboring the gene for human amyloid precursor protein (APP) with different mutations causing familial Alzheimers disease have been developed over the past decade to study pla que deposition and other aspects of AD. However, variations in size, density, pla que number, and total amyloid load between animals of the same age and genotype have been identified by our lab and others. Therefore, to study the effects of compounds on amyloid pathology, it was imperative to develop a technique that would allow each brain hemisphere to receive different infusions. We have developed ca theters that facilitate simult aneous bilateral infusion in mouse brains, thereby using the contralatera l hemisphere of the same animal as an internal control while studying, for exam ple, the effect of compounds on amyloid plaques, a pathological hallmark of the progr ession of Alzheimers disease (AD). Several

PAGE 123

107 molecules have been identified within the plaques including the major component, the A peptide, and two inflammation-related protei ns, apolipoprotein E (apoE) and the serine protease inhibitor -1 antichymotrypsin (ACT). In these experiments, ACT was infused unilaterally over a period of 28 days into the parenchyma and lateral ventricles of PS/APP mice and observed to associate with amyloid plaques, with minimal mortality. Utilizing the ACT/A interaction, an outline of this proced ure is discussed here in detail. 6.2. Introduction Alzheimers disease is the most comm on form of neurodegeneration, affecting over 5 million people nationwide. Approximately 1 person in 8 will be diagnosed with AD by the age of 65, and 1 in 2 by the age of 85 (National Institutes of Health U.S. Department of Health and Human Services). The disease is characterized by severe cognitive decline, leading to dementia and death within an aver age of 8 years from diagnosis. Pathologically, the disease outco me is massive neurona l loss, extracellular deposits of amyloid, and intr acellular accumulation of neurof ibrillary tangles (Selkoe, 1986). Several proteins have been id entified within the plaques including A apolipoprotein E (apoE) (Wisniewsk i and Frangione, 1992) and Alpha 1antichymotrypsin (ACT) (Abr aham et al., 1988). A is the primary component of the plaques and is a 40-43 amino acid peptide pr oteolytically derived from the amyloid precursor protein (APP) (Hardy and Selkoe, 2002). Mutations in the APP gene that cause AD result in an increased producti on of pathogenic forms of A (Tanzi et al., 1991). ApoE and ACT are part of a local in flammatory process in the brain. A binds to ACT and ApoE, and these proteins serv e to catalyze the conversion of A into filaments in vitro (Ma et al., 1996; Wisnie wski et al., 1994), and in vivo to form beta-pleated structured amyloid deposits (Bales et al., 1999; Nil sson et al., 2001). ACT is a 68 kDa acute phase serum glycoprotein from the family of serine protease inhibitors (serpins). Systemically, it is released by hepatocyes and monocytes in response to inflammation and functions as a su icide inhibitor of cathepsin G released by neutrophils (Kalsheker, 1996). In the brain, ACT is known to be produced by and

PAGE 124

108 released from astrocyt es in response to IL-1 stimulation by microglia following trauma, infection, or amyloidoses (Abraham, 2001; Das and Potter, 1995). ACT is highly upregulated in AD brain and is involved both in the formation of amyloid plaques (Abraham et al., 1988) and in the phosphoryl ation of tau (Padmana bhan et al., 2006). Using a transgenic mouse carrying a mutated human presenilin gene (PS) and a mutated human APP gene (PS/APP), we observed the co-loc alization of ACT on existing amyloid plaques in adult PS/APP mice, utilizi ng a novel brain infusion technique. Amyloid plaque load can va ry significantly from animal to animal of the same APP transgenic genotype as well as from hemi sphere to hemisphere of the same animal (Figure 20). We analyzed the brains of 6 PS/APP mice at 10 months of age at 5 sections per brain front to back for variations in plaque load. Although variab ility does exist from hemisphere to hemisphere in this model per a given section (11%), we have demonstrated that the variability from animal to animal is almost 3 times that from that of total hemisphere to hemisphere comparison (78% fr om mean compared to 28%, respectively) (Figure 20B). For this reason, we designe d a bilateral delivery technique that would allow one of the two hemispheres to serve as an internal control. This technique allows researchers to use fewer mice than an experime nt of this nature would typically require and with more accurate results. Thus ACT, for the purposes of our work, could be delivered to one hemisphere and the vehicle, artificial cerebrospinal fluid (aCSF), to the other. Current systems for drug/compound de livery to the mouse brain use pedestal cannulae which are commercially available, but tend to be bulky and irritating to the animal, restrictive to the researcher, and lead to high mortality rates. Furthermore, use of two independent pedestal-based cannulae infusi ons in the same mouse is not possible due to their large size. Therefore, we designed and constructed novel catheters that are implanted subcutaneously and contoured to th e skull. They are minimally invasive and virtually eliminate iatrogenic mortality. Our data show that this method is reliable for studying the chronic delivery of ACT into PS/APP mice to study its effects on amyloid pathology. Moreover, because the catheters are custom made, they can easily be adapted to study the impact of virtually a ny compound in other mouse models of

PAGE 125

109 neurodegeneration or brain disease as well as on non-transgenic mice in order to study the effects of these compounds on normal physiology. 6.3. Materials and Methods 6.3.1. Catheter Construction The catheters (patent pendingPCT/US08/ 73974) used for this experiment were made by first taking a 1.25 cm length of 30 gauge stainless steel tubing (Small Parts, Inc. Miramar, Fl) and carefully inserting it into a 20g needle (tip removed) to the appropriate depth under a dissecting microscope (Leica, Heerbruug, Switzerland), and bending it at 2.5 mm to approximately 90 degrees being ve ry careful not to crimp the tube. The remaining length of the tube was bent agai n at 5 mm to an angle of 120-160 degrees, approximating the contour of each animals s kull. These metal cannulae were inserted into a 3 cm length of polyethylene (PE-10) tu bing with an intern al diameter of 0.28 mm and an outer diameter of 0.61 mm. One centime ter of the PE-10 was then itself inserted into a 4.5 cm length of sterile polyvinyl tubing (PV-50-I.D. 0.69 mm/ O.D. 1.14 mm, Durect Corp., Cupertino, Ca), held in pl ace using a bead of Locktite 454 adhesive (Plastics One. Roanoke, VA.), and cured overn ight. The following day, sterile water was forced through each catheter assembly to ensu re that the lines were not obstructed. The osmotic pumps were then filled with solution, attached to the catheters and primed and implanted as described below. The proven ca theters were later custom manufactured by Braintree Scientific (Brain tree, MA) with some minor changes: The adhesive was supplanted by heat sealing PE-10 to PE-50 tubing and, because the diameters of the PV50 and new PE-50 were not exactly the same, a sheath of silicone (SIL 047) was placed over the PE-50 with an extra 2 mm overhang to attach it to the flow modulator of the pump, forming a tight, leak-proof seal (Figure 21).

PAGE 126

110 6.3.2. Transgenic Mice PS/APP (presenilin 1/amyloid precursor protein) mice and PS/APP/ACT mice were generated by crossing heterozygous PDGF-hAPP (V71 7F) mice with PDGF-hPS1 (M146L) on both Swiss Webster and C57BL/ 6 backgrounds. In some cases, PS/APP mice were bred with mice harboring a ge ne for human ACT (hACT) under a GFAP promoter (Nilsson et al., 2001) were used to be compared to ACT-infused animals. Genotyping was performed using compara tive real-time PCR (Bio-Rad iCyclerHercules, CA). Pathogenically, these Alzh eimer's mouse models are characterized by robust accumulation of amyloid plaques and the development of microgliosis between 610 months. The mice used in these studies ranged from 9-10 months of age. 6.3.3. Materials Purified human -1 antichymotrypsin (ACT) was purchased lyophilized from Fitzgerald (Concord, MA) and wa s reconstituted in artificial CSF (Harvard Apparatus, Holliston, MA) to a concentration of 1mg/ mL. ACT was infused directly into the hippocampal parenchyma or lateral ventricl es for 28 days using the novel catheters attached to Alzet osmotic minipumps (Al zet model 1004, Durect Corp. Cupertino, CA) with an average flow rate of 0.12 L/hour. The pumps and catheters were submerged in 0.9% sterile saline at 37oC and primed for 48 hours prior to implantation. 6.3.4. Intracranial Infusions All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (9 month-old PS/APP25-35g) were anesthetized with 12% isoflurane, shaved and scrubbed with 10% Betadine solution at the site of incision, and placed into a dual arm stereotaxic frame (Kopf Instruments. Tujunga, Ca.). A small (5 cm) incision was made, exposing the skull and neck, and double bladed scissors (10

PAGE 127

111 cm curved Strabismus, Fine Science Tools, Foster City, CA.) were used to form a subcutaneous pocket along the back of the an imal into which 2 osmotic minipumps were inserted with catheters attached. One of th e stereotaxic arms held an Ideal Micro-drill (Roboz Surgical Instrument Co., Gaithers burg, MD) holding a 0.32 mm diameter carbide drill bit. Using the drill bit on the stereota xic arm to find the proper coordinates (from Bregma -2.2-2.5 mm anterior-pos terior, +/2.2-2.5 mm medial -lateral), two holes were carefully drilled into the skull and a 30 ga uge needle from a Hamilton syringe (Hamilton Co., Reno, Nevada) was attached to the second arm of the stereotaxic frame. The needle was inserted to the appropriate depth ( 2.2-2.5 mm corresponding to the posterior portion of the CA1/CA2 hippocampus) and allowed to sit for 5 minutes for the surrounding tissue to adjust. Alternatively, holes were drille d corresponding to the late ral ventricles at the following coordinates: from Bregma -0.2 mm anterior-pos terior, +/1.0 mm mediallateral, and to a depth of 2.2-2.5 mm. After removing the needle, the tips of the cannulae were held directly over the holes with forceps and then gently inserted straight into the preformed holes. Pulling back slightly on th e catheters by the base of the osmotic pumps while inserting the cannulae provided stability which prevented them from moving and potentially damaging tissue while they were a ffixed. The cannulae were firmly affixed to the skull with Locktite 454 adhesive (Pla stics One. Roanoke, VA.) and secured down with a piece of nitrile, appr oximately 1 cm in diameter. After the adhesive cured, the scalp was closed with 6-7 silk sutures. Fo r analgesia, the animal received an immediate subcutaneous dose of ketoprofen (10mg/kg) and up to every 6 hours, as needed, for a maximum of 48 hours post-operatively. After the 28 day period, the animals were given an overdose (~150mg/kg) of sodium pentobarbita l (i.p.) and were tran scardially perfused with 0.9% saline. The brains were carefully removed and analyzed as outlined below. At this time, the integrity of the catheters was confirmed again by forcing water through them, and the pumps were also determined to be empty.

PAGE 128

112 6.3.5. Preparation of Brain Extracts and Western Blot Analysis The neocortex and hippocampus were each microdissected on ice, immediately snap frozen in powdered dr y ice, and stored at -80o C until use. The tissue samples were then dounce homogenized in cold RIPA buf fer (50 mM TrisHCl pH7.4, 150 mM NaCl, 2 mM EDTA, 1% NP-40, 0.1% SDS) in the pr esence of a protease inhibitor cocktail (Complete mini protease inhibito r tablet -Roche Diagnostics, Indianapolis, IN). Samples were subsequently spun at 20,000 g for 30 min (Eppendorf 5417R, Westbury, NY) and the supernatants removed, ali quoted, and frozen at -80oC. Equal amounts of protein from each sample were mixed with 2x sample buffer (Invitrogen, Carlsbad, CA), boiled for 5 min, and subjected to 10-20% Tris Tricine SDS gel electrophor esis (Invitrogen). Purified hACT and nontransgenic brain homogenates were used as co ntrols. The electrophoresed proteins were wet transferred onto 0.2 m nitrocellulose (Whatman, Dassel, Germany) and probed with a rabbit anti-mouse polyclonal antibody to ACT (1:1000Dako, Glostrop, Denmark). The secondary antibody was horseradish peroxidase-conjugated donkey anti-rabbit IgG (Pierce, Rockford, IL ). The immuno-labelled proteins were visualized with an enhanced chemiluminesence detection kit (SuperSignal pico, Pierce, Rockford, IL). After different exposures, autoradiographs were developed with a Konica/Minolta SRX101A (Tokyo, Japan), scanned using a Ricoh Aficio MP3010 (Tokyo, Japan) and quantified using Image J. 6.3.6. Histology and Immunohistochemistry Mouse brain tissues were fixed in 10% formaldehyde (formalin) for 72 hours and then passed through a series of sucrose soluti ons (10-30%) over anothe r 72 hours. Brains were frozen to the peltier stage (Physit emp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzerland) and s ectioned coronally at 25 m. After incubating the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X100, and 0.02% sodium azide) for 60 min, primar y antibodies against ACT (1:500Dako, Glostrop, Denmark) and A (6E10,1:1000Covance, Princeton, NJ) were applied and

PAGE 129

113 incubated at 4 o C overnight. After thorough washing, th e sections were incubated with secondary antibodies, Alexa 488 and 594 fluor ophores (1:1000, 1:4000-In vitrogen) in the dark at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,000, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and sealed with Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). The images were analyzed on a Zeiss Imager Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axi ovision 4.7 software and quantified using Image J. Based on previous plaque analyses performed in our lab and others, for animal to animal, hemisphere to hemisphere pla que comparison, 5 sections from 6 mice (30 total) were stained using Thioflavin S to label the beta-pleated amyloid deposits from anterior to posterior hippocampus in each hemisphere of each mouse. Since the analysis is pairwise, i.e. every data point is compared to every data point, this number was sufficient to show relative variance. Corona l sections of brain (25m) were mounted to slides and let to adhere overnight. The s ections were then rinsed with deionized water and submerged in a 1% Thioflavin S aqueous solution (Sigma-Aldrich, St. Louis, MO) for 5 minutes. The sections were differentia ted in 70% ethanol for 5 minutes, rehydrated in 30% ethanol for 5 minutes, washed with deionized water coverslipped, imaged, and quantified as described previously. 6.4. Results 6.4.1. Animal Recovery Intracranial delivery of molecules, though invasive, is a common procedure to confirm results from in vitro experiments in an in vivo system and to test various compounds in transgenic animal s via targeted delivery. As an additional advantage, the novel method described here bypasses concer ns related to the blood-brain barrier permeability of the administered compound. AC T, a serine protease inhibitor that is known to contribute to the formation of am yloid pathology in Alzheimer's disease, was infused into the hippocampi or lateral ve ntricles of 9-month-old PS/APP mice for a

PAGE 130

114 period of 28 days. ACT was infused into one hemisphere and aCSF (vehicle) into the other hemisphere, thus allowing a comparativ e analysis to be made within the same animal. In all cases, the animals recovered very quickly from the procedure and resumed eating and grooming within minutes of awaken ing. All animals were monitored closely during the 4 week infusion. In only a few inst ances did animals scratch at the incisions, but in no case were they able to open the wound and remove or damage any of the implanted pumps and catheters. In most cases the hair grew back within 2 weeks, and the animals became virtually indistinguishab le from non-treated animals, with the exception of bulges on the haunches from the implanted osmotic pumps (Figure 22A). The implants caused minimal discomfort to the animals, and procedural mortality was practically eliminated from these studies. When harvesting tissues following the 28 day infusion, the catheters were checked to ensure they were still connected to the pumps (Figure 22B) and to the cannulae (Figure 22C). The nitrile cap is indicated by the arrow in figure 22D. The entire assembly is show n attached to an Alzet osmotic minipump model 1004 (Durect Corp., Cupertino, CA.) in Figure 22E. Finally, when removing the brains only 2 small holes are visible (Fi gure 22F) showing that there is no lateral movement of the cannulae during infusion. No serious damage or impairment was caused by the cannula, although tis sue is displaced where inse rted. Figure 23 shows, on the left hemisphere, exactly where the cannula was placed and that surrounding tissue was intact following infusion. Figure 24 A a nd B show a region of the hippocampus anterior to the infusion site which also does not show signs of damage. To be noted, the gauge of the cannulae used in this experiment (30g or 0.30 mm O.D.) is the same of those used in commercially available cannulae (Alzet, PlasticsOne), but without the irritation to the animal, inflammation, occasional motor impairment, and high mortality rates associated with them. 6.4.2. Bilateral Brain Infusion Due to the increased variability in plaque load between animals, even within a largely inbred transgenic mouse line, it is of ten difficult to compare treated and untreated

PAGE 131

115 mice with a reasonably small number of animals. This phenomenon can be caused by phenotypic variability and/or differential expression of transg enes. Upon analysis of 6 untreated PS/APP mice at 10 months of age, we found that plaque load can vary significantly from animal to animal (78% mean standard deviation in pixel count). This fact makes it difficult to draw conclusions on the effects that different molecules might have on amyloid deposition or other features of the transgenic brain when comparing animals where hemispheres vary much less (28%) when compared. Figure 20A and B demonstrates the variation in plaque load between hemispheres and within animals of this genotype. Therefore, in order to study promoters and inhibitors of AD pathology directly in the brain, it would be preferable (almost twice the effici ency) to compare the individually-treated hemispheres of each anim al rather than to compare the brains of littermates or matched cohorts from other litters. Western blots and immunocytochemical techniques demonstrated that our cathete rs functioned properly and that ACT was delivered successfully to the ventricles and parenchyma, re spectively (Figures 23, 24, and 25). If ACT is infused into a PS/APP mouse that has plaques for it to interact with, it binds to the plaques and diffuses very little (Figure 24C & D). However, if ACT is delivered into the hemispheres of nontransgeni c mice it diffuses, but still remains in the intended hemisphere (Figure 24A and B). 6.4.3. Interaction of ACT and amyloid It is well established that ACT is localized, together with the A peptide, in the amyloid plaques of Alzheimers disease (Abraham et al, 1988). However, the mechanism and effect of th is interaction are still unde r investigation. Previous in vitro experiments indicated that ACT can bind to A and become integrated into amyloid filaments as they are formed, and the same mechanism was presumed to occur in vivo (Mucke et al., 2000). Using the applicat ion method described here, we successfully delivered ACT into individual brain hemis pheres of adult PS/APP mice and found that ACT binds to the perimeter of existing pl aques in a similar manner as found for ACT association with amyloid pla ques in PS/APP/ACT transgenic mice (Nilsson et al., 2001)

PAGE 132

116 (Figure 26). While the applic ation of exogeneous ACT did no t lead to increased amyloid load during the time frame of these experiments, it should be noted that ACT was infused for only 28 days. A longer exposure could resu lt in an increased plaque load and is currently being investigated. At the same time, while ACT is known to enhance plaque formation, the perimetrical bind ing of ACT to pre-exisiting A plaques, as shown here, might indicate that ACT involvement in AD pathogenesis not only occurs concurrent with plaque formation but also afterwards. This conclusion is consistent with the fact that ACT is highly upregulat ed in adult AD brains (Abr aham et al, 1988) and also induces the phosphorylation of ta u (Padmanabhan et al, 2006). Although there is evidence that ACT's involvement in AD is not restricted to its participation in plaque formation, the pur pose of using ACT in this model was to demonstrate the utility of our infusion system The catheters and ca nnulae described here served as an outstanding t ool for targeted delivery of ACT into individual brain hemispheres and will facilitate further examination of the mechanistic involvement of ACT and other molecules in the pat hology of Alzheimer's disease. 6.5. Discussion The variability of plaque load (i.e. size, density, and number) between brain hemispheres in PS/APP mice, though a factor to c onsider in these types of experiments, is substantially less than plaque variability from mouse to mouse (Figure 20A and B.). Based on our analysis of 6 untreated, unrelated animals in our colony, we determined the plaque load data from 5 brain sectio ns through 6 ten-month-old PS/APP mice and compared the standard deviation (plaque va riance) between sides of the same slice (11%), between total hemispheres of the same animal (28%), and between individual animals (78%). All of the data from the left hemisphere were compared with the right for each mouse (Hemi-Hemi, 1 data point for each m ouse); the left half of a single slice is compared with the right side (Left-Right, 1 data point fo r each slice); and a permuted comparison of each mouse with every other mouse (Mouse-Mouse, 1 data point for each pair of mice). Where data were aggregated, (Hemi-Hemi and Mouse-Mouse), the mean of

PAGE 133

117 the aggregated counts was used to make the standard deviation comparable to the single slice data (standard deviation of two numbers). The data indi cate that evaluation of whole hemispheres in the same mouse is about 4 times less varied than comparing different animals to each other. Because of this variability, it was advantageous to develop a technique for the delivery of drugs and small molecules that would allow the treatment of individual brain hemispheres. By expos ing one hemisphere to an experimental compound and the contralateral hemisphere to the vehicle only, our novel technique described here easily allows each animal to serve as its own histological and biochemical control. Previously, bilateral treatment, either into lateral ventricles or brain parenchyma was difficult to accomplish with mice. Because transgenic mice provide an important experimental system for disease st udies, such as Alzheimers disease, it was imperative to develop a durable and reliable technique for simultaneous bilateral delivery into the brain that is well-tolerated by mi ce. Commercially available bilateral delivery devices for mice are pedestal cannulae, which have proven problematic for a variety of reasons: 1) The pedestals are bulky and requi re large amounts of adhesive to fix into place because the cannula surface that comes into contact with the skull is flat while the skull is curved. This design requires that e nough adhesive be applied to fill the gaps and to secure the applicator in place. Frequently, the adhesive does not have a chance to cure completely and reacts with surrounding tissue, causing irritation and the need for animals to be removed from the experiment. 2) The pedestals and tubing remain partially exposed, allowing the animals access to the catheters potentially compromising experiments and also causing concern for their health (inflammation, ca chexia, death). 3) The bilateral pedestals have limited flexibility fo r the spacing of the catheters. If they are designed with fixed distances and fixed depths, their placement in a 25 gram mouse is very different compared to a 35 gram mous e, leading to improper placement of the cannula. 4) Other systems require specialized arms to lower the catheter into place, whereas these are best manipul ated individually by hand to be reliably and reproducibly positioned. 5) Moreover, the entire catheter s and pumps we used can be completely removed without terminating th e life of the animal. This is generally not true of

PAGE 134

118 commercially available catheters, which, when removed cause damage to the skull and do not leave the scalp in healt hy enough condition to be suture d. 6) Lastly, since these catheters remain completely subcutaneous, be havioral testing can be safely performed while an animal is being treated, which is not a possibility with pedestals. Overall, the stereotaxic placement of pedestal cannulae, part icularly bilateral pedestal cannulae, into a mouse brain is very restrictive and problematic. Our catheter system described here fa cilitates targeted compound delivery to individual mouse hemispheres and is remarkab ly adaptable to reaching any region of the mouse brain and overcomes the challenges lis ted above. Because the stainless steel catheters can be fashioned to any depth and are implanted individually, variability in the size of the mouse brains can be easily accounted for and acc urate catheter placement and proper contouring is ensured. Most importa ntly, the entire system is implanted subcutaneously with little if any protrusi on. The use of the non-re active nitrile cap to secure the catheters in place requires only a fraction of the adhesive compared to the pedestal method, thereby giving ample time for it to cure properly before the scalp is sutured. Our observations dem onstrate that, in most cases, the animals re-grow hair over the scalp in a very short time and never scra tch or over-groom the site or damage the catheters. The mortality in experiments using this method is closer to 2-3% as opposed to 30-40% using conventional techniques over the 28-day infusion period. Here we demonstrated that ACT binds to pre-existing amyloid plaques in APP transgenic mice only in the infused hemisphe re. We also showed that ACT could be delivered both to the parenchyma of the m ouse hippocampus and into lateral ventricles using our new catheters. Cha nges in total plaque load, how ever, in the presence of ACT were not observed over the 28 day infusions. Although ACT is known to be involved in the process of fibrillization of A both in vitro and in vivo we did show that mature plaques retain binding capacity for ACT. The morphology of the amyloid plaques after ACT infusion strongly resembles the plaque morphology in PS/APP/ACT transgenic mice (Figure 26). Our observation suggests th at our delivery system is an effective method to test various compounds in mouse models of neurodegenerative and other

PAGE 135

119 diseases. Ongoing investigations with inhibitors of the A /ACT interaction will help to determine ACT's role in AD both dur ing and beyond plaque promotion. In sum, using our novel catheter system, we have studied the infusion of a protein known to be a component of AD pathology into a mouse model of AD. The potential application of this system, however, is quite broad. While we have focused on intrahippocampal and intracerebroventricu lar (ICV) delivery, our catheters can conceivably be inserted in a ny depth and be bent to any angle to deliver compounds to any region of the brain in other mouse models of neurodegenerative disease. Bilateral pedestal cannulas do not provide users with such adaptability, since accounting for variability in animal size poses a problem Likewise, using 2 independent pedestal catheters to achieve bilateral delivery in mi ce is prohibited due to the large size of the pedestals. The technique described here ev en allows for the implantation of 2 cannulae per hemisphere (4 per brain), which would allow delivery of multiple agents to either hemisphere in the same mouse with minimum mortality. Acknowledgements: Support for this work was generously provided for by the Johnnie B. Byrd Sr. Alzheimers Center and Research Institute.

PAGE 136

120 6.6. References Abraham CR. Reactive astrocytes and alpha1-antichymotrypsin in Alzheimer's disease. Neurobiol Aging, 2001; 22: 931-6. Abraham CR, Selkoe DJ, Potter H. Immunochemi cal identification of the serine protease inhibitor alpha 1-antichymotrypsin in the brai n amyloid deposits of Alzheimer's disease. Cell, 1988; 52: 487-501. Bales KR, Verina T, Cummins DJ, Du Y, D odel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghetti B, Paul SM. Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mous e model of Alzheimer's disease. Proc Natl Acad Sci U S A, 1999; 96: 15233-8. Das S, Potter H. Expression of the Alzheimer amyloid-promoting factor antichymotrypsin is induced in human as trocytes by IL-1. Neuron, 1995; 14: 447-56. Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeu tics. Science, 2002; 297: 353-6. Kalsheker NA. Alpha 1-antichymotrypsin. Int J Biochem Cell Biol, 1996; 28: 961-4. Ma J, Brewer HB, Jr., Potter H. Alzhei mer A beta neurotoxicity: promotion by antichymotrypsin, ApoE4; inhibition by A be ta-related peptides. Neurobiol Aging, 1996; 17: 773-80. Mucke L, Yu GQ, McConlogue L, Rockenstei n EM, Abraham CR, Masliah E. Astroglial expression of human alpha(1)-antichymotryps in enhances alzheimer-like pathology in amyloid protein precursor transgenic mice. Am J Pathol, 2000; 157: 2003-10. Nilsson LN, Bales KR, DiCarlo G, Gordon MN, Morgan D, Paul SM, Potter H. Alpha-1antichymotrypsin promotes beta-sheet amyloi d plaque deposition in a transgenic mouse model of Alzheimer's disease. J Neurosci, 2001; 21: 1444-51. Padmanabhan J, Levy M, Dickson DW, Potter H. Alpha1-antichymotrypsin, an inflammatory protein overexpressed in Alzheimer's disease brain, induces tau phosphorylation in neurons. Brain, 2006; 129: 3020-34. Selkoe DJ. Altered structural proteins in plaq ues and tangles: what do they tell us about the biology of Alzheimer's disease? Neurobiol Agi ng, 1986; 7: 425-32. Tanzi RE, George-Hyslop PS, Gusella JF. Molecular genetics of Alzheimer disease amyloid. J Biol Chem, 1991; 266: 20579-82.

PAGE 137

121 Wisniewski T, Castano EM, Golabek A, Vogel T, Frangione B. Acceleration of Alzheimer's fibril formation by apolipoprotei n E in vitro. Am J Pathol, 1994; 145: 10305. Wisniewski T, Frangione B. Apolipoprotein E: a pathological chaperone protein in patients with cerebral and systemic am yloid. Neurosci Lett, 1992; 135: 235-8.

PAGE 138

6.7. Figures 122

PAGE 139

Figure 20. A) Thioflavin S staining on 25 m coronal sections (montages) of 3 different PS/APP mice at 10 months of age showing plaque variability in the hippocampus and neighboring cortex. B) Scatter-plot of plaque load data from 5 brain sections through 6 ten-month-old PS/APP mice comparing standard deviation (plaque variance) between total hemispheres of the same animal (28%), between sides of the same section (11%) and between animals (78%). All of the data from the left hemisphere are compared with the right for each mouse (Hemi-Hemi, 1 data point for each mouse); the left half of a single slice is compared with the right side (L eft-Right, 1 data point for each slice); and a permuted comparison of each mouse with every other mouse (Mouse-Mouse, 1 data point for each pair of mice). Where data were aggregated, (Hemi-Hemi and Mouse-Mouse), the mean of the aggregated counts was used to make the standard deviation comparable to the single slice data (standar d deviation of two numbers). 123

PAGE 140

Figure 21. Catheter assembly. AFirst bend of 30g stainless steel tubing. BSecond contour bend. Cpolyethylene PE -10 tubing. DHeat seal. EPE-50. FSilicone sleeve (SIL047). 124

PAGE 141

Figure 22. A 10 month-old PS/APP mouse with 2 osmotic pumps following 28 days of infusion while still active (A) and after transcardial perf usion showing the Alzet pumps (B) and cannulae (C) still securely attached. The arrow in figure D shows the nitrile cap used to affix the catheters to the skull and Fi gure E shows the entire assembly attached to an Alzet 1004 osmotic minipump. When the brain is removed, only small holes are visible where the cannulae were inserted (F). 125

PAGE 142

126

PAGE 143

Figure 23 A) A montage of images taken at 10x from 25 m sections of an animal treated with ACT on the left and aCSF contro l on the right. The top section is immunolabelled with the A antibody 6E10 and the bottom with ACT antibodies. Inlays (40x) show A in green, ACT in red, and Ho echst nuclear stain in blue. B) Semi-quantitative analyses of ACT and A immunostaining in the hippocampus and cortex of PS/APP mice (n= 4) infused with ACT into the parenchyma of one hemisphere and aCSF in the contralateral hemisphere. 127

PAGE 144

Figure 24. Montage images of 25 m coronal sections of 10 month old nontransgenic (A & B ) and PS/APP mice (C & D) immuno-labeled with and antibody against 1 antichymotrypsin following 28 day infusion dem onstrating that ACT stays on the infused hemisphere with or without the presence of plaques. 128

PAGE 145

Figure 25 A) Western Blot of PS/APP of RIPA homogenized mouse hemispheres receiving ACT (red arrows) or aCSF (white arrows) into lateral ventricles (ICV). The green arrow denotes ACT taken from an osmotic pump implanted for 28 days to show minimal degradation of the protein during the experiment. B) Quantification of ACT in hemispheres infused with ACT or aCSF. 129

PAGE 146

Figure 26. Comparison of ACT associated with amyloid plaques (40x) from a PS/APP/ACT transgenic mouse (A) and a PS/APP ACT-infused mouse (B) both at 9 months of age. Im muno-labeling of A is shown in green (6E10 antibody), ACT in red, and Hoechst nuclear stain in blue. 130

PAGE 147

131 CHAPTER 7 THE NEUROPRO TECTIVE EFFECTS OF THE CARBOXYL TERMINUS OF GAMMA LAMI NIN, KDI, ON A MOUSE MODEL OF ALZHEIMERS DISEASE 7.1. Introduction Although AD is classified as a protei nopathy, the two curr ently FDA approved medications used to treat the disease, Aricept and Namenda, act on distinctly different aspects of the disease. Aricept (donezepil HCL) is an acety lcholinesterase inhibitor and is believed to improve cognition in patients by increasing synaptic c oncentrations of the neurotransmitter acetylcholine. The exact mechanism of this effect is not known, but it has been suggested that Aricept increases the expression of IG F-1 in hippocampal neurons, increases neurogenesis in the dentate gyrus, and increases cerebral blood flow in parenchymal arterioles (Narimatsu et al ., 2009). Namenda (memantine) functions differently as its primary function is believe d to be inhibition of NMDA receptors in glutamate neurotransmission. Blocking thes e receptors in a noncompetitive, voltage mediated (ionotropic) manner thereby inhibi ting the prolonged influx of Ca2+ ions that induce excitotoxic damage to the neuron. Over the past decade, our collaborator, Pivi Liesi, has worked extensively with laminins. In particular, she has elucidated that the carboxyl terminus of 1-laminin is responsible for neurite outgrowth both in vitro and in vivo To date, there are 15 known isoforms of the laminin expressed in mammals in a tissue-specific manner. Five different subunits have been described, 4 and 3 which arrange as a cr uciform heterotrimer ( ) of ~900KD in total mass as intact laminin molecules. The molecules are the main non-collagenous components of basement memb ranes and also play diverse roles in the extracellular matrix relati ng to cell structure, adhesion, differentiation, migration, signaling, metastasis, and neurit e outgrowth. It has also been shown that laminin is

PAGE 148

132 upregulated during development and repair of most tissues and portions of the molecule can maintain biological activity following cleavage events. Original work focused on the gamma laminin decapeptide exerting these neuroregenerative and neuroprotective effect s; however, subsequent studies of the hexapeptide and finally the tripeptide, KDI, still showed activity as a neurite outgrowth factor and functioned as a non-competitive ionotropic glutamate receptor antagonist. In vitro work, using patch clamp analysis, reveal ed that KDI could inhibit ~50% of NMDA receptors and 100% of AMPA receptors (Moykkynen et al., 2005). Further work showed in vivo that pretreatment of the rat hippocampus with bolus injections of KDI protected the brain region from subsequent injections of kainic acid, a glutamate analog (Wiksten et al., 2004a). If indeed dysr egulation of glutamate is an etiology found in AD and Namenda has proven to have some efficacy, we proposed that KDI would work in a similar manner to ameliorate the sy mptoms of AD and improve cognition. In these experiments, KDI was used to treat PS/APP mice. Since it is not yet known whether this molecule can cross the blood-brain barrier following systemic administration, direct infusion of the molecu le into the hippocampal parenchyma of these models was required. In order to determ ine if KDI could improve working memory, similar to memantine, the animals were test ed in a RAWM followi ng 28 days of chronic infusion. Preliminary data suggested improve ment with treatment of KDI, but without change in plaque load. Therefore, since KDI was reported to bind glutamate receptors, another avenue was investig atedLong term potentiation. Long-term potentiation (LTP) is a molecular mechanism involving synapse strengthening upon recent patterns of activity. It serves as a model for memory and learning in several brain areas, described first in the hippocampus (McKay, 1997). Briefl y, it involves the bind ing of glutamate to ionotropic glutamate receptors, post-syna ptic depolarization, influx of calcium, phosphorylation of Thr286 and constitutive activ ation of CaM Kinase II, resulting in the upregulation of more AMPA receptors and ac tivation of relevant genes (Schlett, 2006). During the time that the KDI infusions were being performed, EIIKDI was cloned into an expression vector. This was done in anticipation of achieving longer term expression of the molecule in the brain. In vivo electroporation has been used in

PAGE 149

133 numerous tissues for gene expression (Helle r and Heller, 2006). Briefly, the technique works by injecting plasmid into tissue and surr ounding the site with an electric field to perforate the cells and allow the plasmid to enter cells and be expressed. This method has many advantages over traditional means of transfection that i nvolve developing viral vectors and contending with an immune response. Electroporation is effective as a proof-of principle that introduced genes will be a) expressed and b) exert a physiological response within th e cell before developing mo re complicated transfection strategies. Previous work in ours and other la bs has demonstrated th at this technique is safe in the CNS of mice and was optimized first using an EGFP expression vector (figure 27). 7.2. Materials and Methods 7.2.1. Catheter Construction The catheters (patent pendingPCT/US08/ 73974) used for this experiment were made by first taking a 1.25 cm length of 30 gauge stainless steel tubing (Small Parts, Inc. Miramar, Fl) and carefully bending it at 2.5 mm under a dissecting microscope (Leica, Heerbruug, Switzerland), to approximately 90 degrees, being careful not to crimp the tube. The remaining length of the tube was bent again at 5 mm to an angle of 120-160 degrees, approximating the contour of each animals skull. These metal cannulae were inserted into a 3 cm length of polyethylene (PE-10) tubing with an internal diameter of 0.28 mm and an outer diameter of 0.61 mm. One centimeter of the PE-10 was then itself inserted into a 4.5 cm lengt h of sterile polyvinyl tubing (PV-50-I.D. 0.69 mm/ O.D. 1.14 mm, Durect Corp., Cupertino, Ca), held in place using a bead of Locktite 454 adhesive (Plastics One. Roanoke, VA.), and cured overn ight. The following day, sterile water was forced through each catheter assembly to ensu re that the lines were not obstructed. The osmotic pumps were then filled with solution, attached to the catheters and primed and implanted as described below. The proven ca theters were later custom manufactured by Braintree Scientific (Brain tree, MA) with some minor changes: The adhesive was

PAGE 150

134 supplanted by heat sealing PE-10 to PE-50 tubing and, because the diameters of the PV50 and new PE-50 were not exactly the same, a sheath of silicone (SIL 047) was placed over the PE-50 with an extra 2 mm overhang to attach it to the flow modulator of the pump, forming a tight, leak-proof seal. 7.2.2. Osmotic Minipumps Sterile Alzet osmotic minipumps-model 2004 (D urect Corp. Cupertino, Ca.), with a flow rate of 0.25uL/hr were filled with 200ug (1mg/mL) of KDI (NeoMPS, San Diego, CA) diluted in artificial CSF (Harvard Appa ratus, Holliston, MA) and incubated at 37oC in sterile saline for 48 hours. T ypically, catheters are attached and primed at this time, but the tungsten wire needed to remain in place fo r stability during insertion into the spine. 7.2.3. Transgenic Mice PS/APP (presenilin 1/amyloid precursor pr otein) mice were generated by crossing heterozygous PDGF-hAPP (V717F) mice w ith PDGF-hPS1 (M146L) on both Swiss Webster and C57BL/6 backgrounds. Genotyping was performed using comparative realtime PCR (Bio-Rad iCyclerHercules, CA). Pathogenically, these Alzheimer's mouse models are characterized by robust accu mulation of amyloid plaques and the development of microgliosis be tween 6-8 months of age. 7.2.4. Intracranial Infusions All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (8 month-old PS/APP25-35g) were anesthetized with 1-2% isoflurane, shaved and scrubbed with 10% Beta dine solution at the site of incision, and placed into a stereotaxic frame (Kopf Inst ruments. Tujunga, Ca.). A small (3 cm) incision was made, exposing the skull and neck, and double bladed scissors were used to

PAGE 151

135 form a subcutaneous pocket along the back of the animal into which 2 osmotic minipumps were inserted with catheters attach ed. Two holes were drilled into the skull (from Bregma -2.2-2.5 mm anterior-posterior +/2.2-2.5 mm medial -lateral, and the 30 gauge catheters were inserted to a dept h of 2.2-2.5 mm) corresponding to the posterior portion of the CA3 hippocampus. Alternativel y, holes were drilled corresponding to the lateral ventricles (from Bregma -0.2 mm ante rior-posterior, +/1.0 mm medial-lateral, and to a depth of 2.2-2.5 mm). Once they we re inserted, the cannulae were affixed to the skull with Locktite 454 adhesive (Plastics One. Roanoke, VA.) and secured down with a piece of nitrile, approximately 1 cm in diamet er (Fig 3D). After the adhesive cured, the scalp was closed with 6-7 silk sutures. Fo r analgesia, the animal received an immediate dose of ketoprofen (10mg/kg) and up to every 6 hours, as needed, for a maximum of 48 hours post-operatively. After the 28 day pe riod, the animals were given an overdose (~150mg/kg) of sodium pentobarb ital (i.p.) and were transcar dially perfused with 0.9% saline. The brains were carefully removed and analyzed as outlined below. At this time, the integrity of the catheters was confirmed again by forcing water through them, and the pumps were also determined to be empty. 7.2.5. Electroporation The mouse codons for the longer peptide, EIIKDI, were determined and oligomers were synthesized with a small portio n of a leader sequence to a length of 98 nucleotides (IDT, San Jose, CA). The sens e and antisense strands were annealed and blunt-end cloned into a PCDNA 3.1+ expression vector between the XbaI and ecoRV sites and confirmation of insertion, c opy, and orientation was confirmed by DNA sequencing. An EGFP clone was used to optim ize the electroporation procedure as it was roughly the same size (5 kB) and provided eas y visibility of transfection efficiency (figure 26) The procedure is similar to what was mentioned above. Animals (8 monthold PS/APP25-35g) were anesthetized with 1-2% isoflurane, shaved and scrubbed with 10% Betadine solution at the site of incisi on, and placed into a stereotaxic frame (Kopf Instruments. Tujunga, Ca.). A small (3 cm ) incision was made, exposing the skull and

PAGE 152

136 landmarks. 3 holes were made, the first at from Bregma -2.2-2.5 mm anterior-posterior, + 2.2-2.5 mm medial-lateral, corresponding to the CA3 region of the hippocampus. The other 2 holes were drilled at 1mm caudal and 1mm rostral to the first hole and were for placement of the electrodes (at a depth of 2.5mm). 5 g of EGFP DNA (1 g/ L) was injected into the center hole to a depth of 2.5 mm and the plasmid is pulsed into the surrounding cells. The square pulse generato r (BTX, Hollister, Maine) is then set to 225V with 10 pulses, 2Ms each and 250Ms inte rvals between pulses. The animal was given an overdose (~150mg/kg) of sodium pentobarbital (i.p.) 24 hours later and the brain carefully removed under a fluores cence dissecting scope (Leica, Heerbruug, Switzerland) and photographed with a Lei ca stereoscopic camera (Leica, Heerbruug, Switzerland). EGFP was electroporated both in the hippocampus and subventricular zone (figure 27). 7.2.6. Behavior For the RAWM task of spatial working memo ry (Arendash et al., 2007; Arendash et al., 2001; Ethell et al., 2006), an aluminum insert was placed into a 100cm circular pool to create 6 radially-distributed swim arms em anating from a central circular swim area (Figure 28). An assortment of 2-D and 3-D visual cues surrounded the pool. The number of errors prior to locating which one of the 6 swim arms contained a submerged escape platform (9cm diameter) was determined for 5 trials/day. There was a 30-min time delay between the 4th trial (T4; final acquisition trial) and 5th trial (T5; memory retention trial). The platform location was changed daily to a di fferent arm, with different start arms for each of the 5 trials semi-randomly selected from the remaining 5 swim arms. During each trial (60 s maximum), the mouse was returned to that trials start arm upon swimming into an incorrect arm and the num ber time required to locate the submerged platform was recorded. If the mouse did not fi nd the platform within a 60-s trial, it was guided to the platform for a 30-s stay. The numbers of errors and escape latency during trials 4 and 5 are both considered indices of working memory and are temporally similar

PAGE 153

137 to standard registration/recall testing of specific items used clinically in evaluating AD patients. 7.2.7. Preparation of Brain Extracts and Western Blot Analysis The neocortex and hippocampus were each microdissected on ice, immediately snap frozen in powdered dr y ice, and stored at -80o C until use. The tissue samples were then dounce homogenized in cold RIPA buf fer (50 mM TrisHCl pH7.4, 150 mM NaCl, 2 mM EDTA, 1% NP-40, 0.1% SDS) in the pr esence of a protease inhibitor cocktail (Complete mini protease inhibito r tablet -Roche Diagnostics, Indianapolis, IN). Samples were subsequently spun at 20,000 g for 30 min (Eppendorf 5417R, Westbury, NY) and the supernatants removed, ali quoted, and frozen at -80oC. Equal amounts of protein from each sample were mixed with 2x sample buffer (Invitrogen, Carlsbad, CA), boiled for 5 min, and subjected to 10-20% Tris Tricin e SDS gel electrophores is (Invitrogen). A and plaques were immunolabeled using 6E10 and 4G8 antibodies (1:1000) (Covance, Princeton. NJ) The secondary antibody was horseradish peroxidase-conjugated donkey anti-rabbit IgG (Pierce, Rockford, IL). Th e immuno-labelled protei ns were visualized with an enhanced chemiluminesence detecti on kit (SuperSignal pico, Pierce, Rockford, IL). After different exposures, autoradiogr aphs were developed with a Konica/Minolta SRX101A (Tokyo, Japan), scanned using a Ricoh Aficio MP3010 (Tokyo, Japan). Quantification was not performed due to a lack of correlation between plaques of treated and untreated animals. 7.2.8. Histology and Immunohistochemistry Mouse brain tissues were fixed in 10% formaldehyde (formalin) for 72 hours and then passed through a series of sucrose soluti ons (10-30%) over anothe r 72 hours. Brains were frozen to the peltier stage (Physit emp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzerland) and s ectioned coronally at 25 m. After incubating the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X-

PAGE 154

138 100, and 0.02% sodium azide) for 60 min, primar y antibodies against ACT (1:500Dako, Glostrop, Denmark) CAMKII (1:500), pCAM KII (1:1000), GlutR1 (1:1000), pGlutR1 (1:1000), synaptophysin (1:250), synaptopodi n (undiluted), synaptobrevin, PSD-95 (1:500), -tubulin (1:1000) and GAP-43 (1:500) and A (6E10,1:1000Covance, Princeton, NJ) were applied and incubated at 4 o C overnight. After thorough washing, the sections were incubated with seconda ry antibodies, Alexa 488 and 594 fluorophores (1:1000, 1:4000-Invitrogen) in the dark at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,00 0, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and sealed w ith Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). The images were analyzed on a Zeiss Imager Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axiovision 4.7 software and quantified using Image J. 7.3. Results and Conclusions Memantine (Namenda) functions through i nhibition of NMDA receptors in order to reduce the effects of glutamate on post synap tic neurons. Glutamate excitotoxicity is a common pathology for most neurodegenerative di seases and is generally accepted as part of the AD pathogenesis (Schaeffer and Gattaz, 2008). This issue, however, is debated since earlier experiments with transgenic mi ce were somewhat conflicting and the effects of memantine on AD patients are modest at best (Farlow et al., 2008). In these experiments, we used the tripeptide car boxyl terminus of gamma laminin, KDI, to determine if a) glutamate excitotoxicity is a factor in a mouse models of Alzheimers and b) if KDI would reduce the ex citotoxic damage and reverse cognitive decline in these mice. Previous work with the peptide fragment revealed that it is a potent non competitive ionotropic glutamate receptor an tagonist that inactivates most AMPA receptors and ~50% of NMDA receptors (Moykkynen et al., 2005). In another set of experiments, this same group showed that pr etreatment of the ra t hippocampus with KDI followed by an injection of kainate, a gl utamate analog, prevented excitoxicity and neuron loss in a dose-dependent manner. This ev idence, taken together with the fact that

PAGE 155

139 KDI is a potent neurite outgrowth factor, indicated it had excellent therapeutic potential in AD. Initially, the transgenic KDI-treated gr oup was showing improvement in cognition from RAWM trials (figures 29 and 30). This was true when comparing individual animals before and after treatment as well as between sham and treated groups of animals. This prompted some excitement and although no change in plaque load was detected, the idea that cognition was improved through the glutamate pathway led us towards thinking about LTP as a possible mechanism for cognitive enhancement. LTP has been identified as a possible molecular mechanism for learning in the DG. Briefly, glutamate binds to AMPA and NMDA receptors and triggers the influx of sodium and calcium, respectively into the postsynaptic cel l, which causes a short-lived depolarization called the excitatory postsynaptic potential (EPSP). With repeated stimulation, the next EPSP will occur prior to the decay of the previo us. The sum of this is that postsynaptic neurons can be more readily depolarized fo llowing a succession of EPSPs and the influx of ions causes a temporary upregulation of the cellular kinases CamKII, PKC, and PKA which function to phosphorylate dormant AMPA receptors, bringing them activated to the synapse and essentially amplifying the si gnal in the post-synaptic neuron (Kerchner and Nicoll, 2008). To this end, changes in expression of CamKII, pCamKII, GluR1, and pGluR1 were looked into, but not observed. Similarly, there were no detectable changes in synaptophysin, synaptotagmin, synaptobrev in or GAP-43, suggesting that there was not increased synaptogenesis occuring in the treated mice. Furthermore, subsequent behavioral trials failed to repeat what was seen in the first cohort and it was decideded, conclusively, that KDI was of no therapeutic benefit to AD mice.

PAGE 156

7.4. Figures 40x Figure 27 Demonstration of successful EGFP vector electroporation in the hippocampus and subventricular zone of mouse brain. 140

PAGE 157

Figure 28. Schematic of radial arm water maze fo r rodents in testing working (shortterm) memory. 141

PAGE 158

Figure 29. Preliminary latency plots in the RAWM of KDI treated and untreated Tg2576 AD mouse models. Re sults were not repeatable in 2 subsequent cohortsno change was s een between groups (data not shown). 142

PAGE 159

Figure 30. Preliminary error plots in th e RAWM of KDI treated and untreated Tg2576 AD mouse models. Re sults were not repeatable in 2 subsequent cohortsnochange was seen between groups (data not shown). 143

PAGE 160

144 CHAPTER 8 ACT INDUCES THE PHOSPHORYLATION OF TAU IN MOUSE MODELS OF AD 8.1. Introduction Alpha 1-antichymotrypsin (ACT) is an acute phase serum glycoprotein and serine protease inhibitor that has been shown conc lusively to accelerate amyloid fibril and plaque formation in vitro and in vivo (Ma et al., 1994; Nilsson et al., 2004). ACT is upregulated in the injured and AD brain a nd released by astrocytes surrounding the neurotoxic insult. Whereas ACT has a direct role in the in the formation of amyloid plaques, recent work from our lab has show n that it can also i nduce the phosphorylation of tau, which is an important step in the fo rmation of neurofibrill ary tangles (NFTs), the other classical protein misfolding hallmark of AD. NFTs are comprised mainly of the microtubule binding protein tau. Tau activ ity is regulated by phosphorylation and hyperphosphorylation of tau promotes its mi sfolding and dissociati on from microtubules. ptau assembles into paired helical filame nts (PHF) that eventually aggregate into insoluble NFTs causing disruption of the cy toskeleton, abnormal axonal transport, and neurotoxicity. The upregulation of so luble ACT in the AD brain, the resulting hyperphoshorylation of tau and the inclusion of ACT in the formation of amyloid plaques, make ACT a link between the amyloidogenic and and tauopathic pathways in Alzheimers disease. Further evidence for ACTs involvement in the phosphorylation of tau comes from examination of post mortem brain ti ssue from individuals of other tauopathies. Known to be closely involved with AD pa thology, ACT was found to be upregulated and colocalized to tau deposits in frontotemporal dementia (FTD ), corticobasal degeneration (CBD), and progressive super nuclear palsy (PSP) patients in neurons and astrocytes.

PAGE 161

145 Moreover, these inclusions are all PHF-1 i mmunopositive, suggesting that an interaction between ACT and tau might be partly responsible for the pathogen ic events in these diseases. ACT is a suicide protease inhib itor, which binds to its target, changes its conformation, and is then, t ogether with the bound prot ease, endocytosed. Although ACT has been found in astrocytes (Abraham, 2001), it was initially not found in neurons. While early northern blot analyses failed to detect ACT mRNA in neurons, subsequent immunocytochemical analyses demonstrated its presence both in neur ons and astrocytes. This observation strongly suggests that ACT, after inhibition of its target protease, enters neurons and then interacts with tau. Alternatively, ta u phosphorylation could occur through signal transduction cascades once ACT binds to a receptor, though no evidence for these hypotheses has been yet obtained. Nonetheless, human cortical neurons and mouse embryonic neurons treated with ACT demonstrate increased phosphorylation of tau at pSer202, Pthr 231, and pser 396/ 404 (PHF-1) as well as intraneuronal ACT staining (Padmanabhan et al., 2006). In the following experiments, we wished to intracranially infuse ACT into the hippocampal parenchyma of transgenic mice th at expresse human tau, but no mouse tau, to determine if tau phosphor ylation and NFT formation would ensue. ACT is well established in the AD field as an amyloid promot er. If it could also be proven to be a part of the ensuing Alzheimers tauopathy it would link these two pathologies. Demonstrating that ACT serves as such a li nk might make it a viable therapeutic target for multiple forms of tauopathies and neurodegeneration. 8.2. Materials and Methods 8.2.1. Catheter Construction The catheters (patent pendingPCT/US08/ 73974) used for this experiment were made by first taking a 1.25 cm length of 30 gauge stainless steel tubing (Small Parts, Inc. Miramar, Fl) and carefully bending it at 2.5 mm under a dissecting microscope (Leica, Heerbruug, Switzerland), to approximately 90 degrees, being careful not to crimp the

PAGE 162

146 tube. The remaining length of the tube was bent again at 5 mm to an angle of 120-160 degrees, approximating the contour of each animals skull. The metal cannulae were inserted into a 3 cm length of polyethylene (PE-10) tubing with an internal diameter of 0.28 mm and an outer diameter of 0.61 mm. One centimeter of the PE-10 was then itself inserted into a 4.5 cm lengt h of sterile polyvinyl tubing (PV-50-I.D. 0.69 mm/ O.D. 1.14 mm, Durect Corp., Cupertino, Ca), held in place using a bead of Locktite 454 adhesive (Plastics One. Roanoke, VA.), and cured overn ight. The following day, sterile water was forced through each catheter assembly to ensu re that the lines were not obstructed. The osmotic pumps were then filled with solution, attached to the catheters and primed and implanted as described below. The proven ca theters were later custom manufactured by Braintree Scientific (Brain tree, MA) with some minor changes: The adhesive was supplanted by heat sealing PE-10 to PE-50 tubing and, because the diameters of the PV50 and new PE-50 were not exactly the same, a sheath of silicone (SIL 047) was placed over the PE-50 with an extra 2 mm overhang to attach it to the flow modulator of the pump, forming a tight, leak-proof seal. 8.2.2. Transgenic Mice 10 htau mice (B6.Cg-Mapttm1(GFP)Klt Tg(MAPT)8cPdav/J) were purchased from Jackson Labs. These mice are homozygous for knock out of the the targeted endogenous mtau alleles and hemizygous for the tr ansgene. Although no endogenous mtau is detected, all six isoforms (including both 3R and 4R forms) of human MAPT are expressed. Hyperphosphorylated tau is det ected in cell bodies and dendrites by three months. Paired helical filaments of aggregated insoluble MAPT can be isolated from brain tissue as early as two m onths of age. Overall, these mice develop tau pathology very quickly and are fertile and viable for studies where they are bred with other AD transgenics.

PAGE 163

147 8.2.3. Materials Purified human -1 antichymotrypsin (ACT) was purchased lyophilized from Fitzgerald (Concord, MA) and wa s reconstituted in artificial CSF (Harvard Apparatus, Holliston, MA) to a concentration of 1mg/ mL. ACT was infused directly into the hippocampal parenchyma for 28 days using the n ovel catheters attached to Alzet osmotic minipumps (Alzet model 1004, Durect Corp. Cupertino, CA) with an average flow rate of 0.12 L/hour. The pumps and catheters were submerged in 0.9% sterile saline at 37oC and primed for 48 hours prior to implantation. 8.2.4. Intracranial Infusions All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (3 month-old ht au20-25g) were anesth etized with 1-2% isoflurane, shaved and scrubbed with 10% Beta dine solution at the site of incision, and placed into a stereotaxic frame (Kopf Inst ruments. Tujunga, Ca.). A small (3 cm) incision was made, exposing the skull and neck, and double bladed scissors were used to form a subcutaneous pocket along the back of the animal into which 2 osmotic minipumps were inserted with catheters attach ed. Two holes were drilled into the skull (from Bregma -2.2-2.5 mm anterior-posterior +/2.2-2.5 mm medial -lateral, and the 30 gauge catheters were inserted to a dept h of 2.2-2.5 mm) corresponding to the posterior portion of the CA3 hippocampus. Once they were inserted, the cannulae were affixed to the skull with Locktite 454 adhesive (Pla stics One. Roanoke, VA.) and secured down with a piece of nitrile, appr oximately 1 cm in diameter (Fig 3D). After the adhesive cured, the scalp was closed with 4-6 silk sutures. For anal gesia, the animal received an immediate dose of ketoprofen (10mg/kg) a nd up to every 6 hours, as needed, for a maximum of 48 hours post-operatively. After the 28 day period, the animals were given an overdose (~150mg/kg) of sodium pentobarbita l (i.p.) and were tran scardially perfused with 0.9% saline. The brains were carefully removed and analyzed as outlined below. At

PAGE 164

148 this time, the integrity of the catheters wa s confirmed again by forcing water through, and the pumps were also determined to be empty. 8.2.5. Preparation of Brain Extracts and Western Blot Analysis The neocortex and hippocampus were each microdissected on ice, immediately snap frozen in powdered dr y ice, and stored at -80o C until use. The tissue samples were then dounce homogenized in cold RIPA buf fer (50 mM TrisHCl pH7.4, 150 mM NaCl, 2 mM EDTA, 1% NP-40, 0.1% SDS) in the pr esence of a protease inhibitor cocktail (Complete mini protease inhibito r tablet -Roche Diagnostics, Indianapolis, IN). Samples were subsequently spun at 20,000 g for 30 min (Eppendorf 5417R, Westbury, NY) and the supernatants removed, ali quoted, and frozen at -80oC. Equal amounts of protein from each sample were mixed with 2x sample buffer (Invitrogen, Carlsbad, CA), boiled for 5 min, and subjected to 10-20% Tris Tricine SDS gel electrophor esis (Invitrogen). Purified hACT and nontransgenic brain homogenates were used as co ntrols. The electrophoresed proteins were wet transferred onto 0.2 m nitrocellulose (Whatman, Dassel, Germany) and probed with a rabbit anti-mouse polyclonal antibody to monoclonal/polyclonal antibodies to PHF-1 (1:1000Peter Davies, Albert Einstein In st. NY), pSer202/CP-13 (1:1000-Peter Davies), MC-1 (1:500Peter Davies), Tg5 total tau (1:1000Peter Davies), pGSK3 / (1:1000Cell Signaling, Beverly, MA), pAKT (1:1000Cell Signaling), pERK (1:1000Cell Signalin g), pThr231 (1:1000-Invitrogen), pSer422 (1:1000Invitrogen), pSer262 (1:1000Invitrogen), AT100 (1:1000ThermoScientific, Waltham, MA), ACT (1:1000Dako, Glostrop, Denmark). The secondary antibodies are horseradish peroxidase-conj ugated goat anti-mouse IgG and donkey anti-rabbit IgG (Pierce, Rockford, Il) and the bands were vi sualized with an enhance SuperSignal pico (Pierce, Rockford, Il). The immuno-labelled proteins were visualized with an enhanced chemiluminesence detection kit (SuperSignal pico, Pierce, Rockford, IL). After different exposures, autoradiographs were devel oped with a Konica/Minolta SRX101A (Tokyo, Japan), scanned using a Ricoh Aficio MP3010 (Tokyo, Japan) and quantified using

PAGE 165

149 Image J. Alternatively, the secondary antibod ies used were infrared (Licor, Lincoln, NE) and visualized at both 680 and 800 nm using the Odyssey (Licor). 8.2.6. Histology and Immunohistochemistry Mouse brain tissues were fixed in 10% formaldehyde (formalin) for 72 hours and then passed through a series of sucrose soluti ons (10-30%) over anothe r 72 hours. Brains were frozen to the peltier stage (Physit emp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzerland) and s ectioned coronally at 25 m. After incuba ting the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X100, and 0.02% sodium azide) for 60 min, a set of primary monoclonal/polyclonal antibodies was applied. Primary antibodies (m ono and polyclonal) incl uded antibodies to PHF-1 (1:250Peter Davies, Albert Eins tein Inst. NY), pSer202/CP-13 (1:250-Peter Davies), MC-1 (1:100Peter Davies), T g5 total tau (1:250Peter Davies), pGSK3 / (1:250Cell Signaling, Beverly, MA), pAKT (1 :250Cell Signaling), pERK (1:250Cell Signaling), pThr231 (1:250-Invitrogen), pSer422 (1:250Invitroge n), pSer262 (1:250Invitrogen), AT-100 (1:250ThermoScient ific, Waltham, MA), ACT (1:500Dako, Glostrop, Denmark). were applied and incubated at 4 o C overnight. After thorough washing, the sections were exposed to secondary antibodies, Alexa 488 and 594 fluorophores (1:1000, 1:4000-Invitrogen) in the da rk at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,000, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and s ealed with Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). The images were analyzed on a Zeiss Imager Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axiovision 4.7 software. 8.3. Results and Conclusions It was also discovered recently that ACT is not only upregulated in AD, but other tauopathies including FTDP, CBD, and PSP (Padmanabhan et al., 2006). When

PAGE 166

150 considering that ACT is highly overexpressed in the AD brain and is part of amyloid plaques, taken together with ACTs role in tauopathies, we hypothesized that ACT could be a trigger for tau phosphoryla tion in AD. A primary focus in AD research has been finding relationships between the two primary hallmarks of the disease, amyloid and tau. Whereas it was believed for some time that these pathologies exis ted independently of each other, new evidence suggests that both am yloid aggregates and intracellular AB can stimulate the phosphorylation of tau. In these experiments, we demonstrate that ACT, an acute phase serum glycoprotein, cont ributes to phosphorylation of tau in vivo Previously, it was demonstrated in huma n and PDAPP mouse cortical neurons that adding ACT to the culture media resulted in AD-specific phosphorylations of tau known to contribute to AD pathogenesis. When ACT is chronically infused into the hippocampus of an htau mouse, these same phosphorylation events occur, although not due to the activity of the same intracellular kinases. Immunohistochemical analyses revealed that increased levels of phosphor-tau were found in the hippocampus of the ACT infu sed hemisphere of 5 month old htau mice and in 12 month old PDAPP mice. Speci fically, strong staining for pSer202 (CP-13 antibody) was detected in htau mice (figure 32) and pThr231 in PDAPP Mice (figure 36). Western blot analysis confirmed the results in the htau mice. Brain homogenates from ACT-treated and untreated htau brain hemispheres were run side by side. Pser202 was increased in each case where ACT was infuse d (figure 33). Moreover, it was also determined that there was an increase in pSer396/Thr404 (PHF-1) in the presence of ACT (figure 31), a phosphorylation event closely associated with pre-tangle formation. This site has been shown to be phosphorylated by GSK-3 and activation of GSK-3 (indicated by dephosphorylation at 51KDa) is ev ident in this study (figure 34). All of these data are consistent with previous in vitro data. One difference between the studies was that ERK was activated (indicated by in creased phosphorylation at 42/44KDa) in the presence of ACT in vitro and appears inactivated in vivo following ACT infusions (figure 35). The exact mechanism by which tau becomes hyperphosphorylated in Alzheimers disease is not known. In vit ro analysis of tau phosphoryla tion have shown that several kinases can induce phosphorylation of tau at numerous sites, roughly half of which are

PAGE 167

151 observed to be hyperphosphorylat ed in Alzheimers disease brains. Inhibition of GSK3 for example, was shown to reduce the phosphorylation of ta u and increase cell viability. GSK-3 amongst other kinases, is elevated in the brains of Alzheimer patients, which suggests that molecules, in this case inflammation-related, can trigger their activation, contributing to tau hyperphosphorylation and tangle formation. In these experiments, we build upon our previous work in which we propose that the increased A level in Alzheimers disease brai n activates microglia to express inflammatory proteins like IL-1B. IL-1B, in turn, activates astrocytes to release large amounts of ACT into the surr ounding parenchyma (Abraham, 2001). ACT then triggers increased expression or activati on of certain kinases, such as GSK-3, which results in tau hyperphosphorylation, tangle formation and ne uronal degeneration (Hooper et al., 2008). Therefore, our data support the theory of an inflammatory cascade in AD, brought on by A More experiments are cu rrently underway to determine which other kinases are activated in response to ACT that result in tau phosphorylat ion. Inhibition of kinase activity is currently being pur sued as a potential therapy for AD (Perez-Martinez, 2009). These results suggest that ACT could be a ca ndidate as an upstream target in both the formation of amyloid plaque s and phosphorylation of tau. Taking together the in vitro and in vivo tau data, these results suggest that inhibiting ACT upstream from tangle formati on (and amyloid plaque formationchapter 9) could be a potenti al therapy for AD.

PAGE 168

8.4. Figures R L L R L L R L R R htau htau htau NTG NTG Figure 31. ACT infusion induces incr eases in phospho tau (PHF-1 ) in htau mice. Whole brain homogenates from ACT treated (right) and untreated (left) hemispheres of htau mice are run alongside each other in a west ern blot and visualized using infrared secondary antibodies (Lycor). The white ar rows indicate where ACT (in red) was infused. PHF-1 is shown in green and the si gnal intensity increases in the presence of ACT. Data is not quantified due to the small number of animals used. 152

PAGE 169

Figure 32. Immunohistochmical analysis of an ht au mouse following ACT infusion (left side). The montage was taken at 5x and the inlays taken at 40x. ACT is shown in red, pSer202 is shown in green and is upregulated in the presence of ACT. Data is not quantified due to the small number of animals used. 153

PAGE 170

Figure 33. Whole brain homogenates fr om ACT treated) and untreated hemispheres of htau mice are run alongside eachother in a wester n blot and visualized using an infrared secondary antibody (Lycor). PSer202 signal intens ity (green) increases in the presence of ACT. 154

PAGE 171

ACT untreated Figure 34. ACT infusion leads to a dephosphorylat ion, thus activation, of GSK3 as indicated by the arrows on a westen blot of whole brain homogenates from htau mice. 155

PAGE 172

pERK 42/44 KD ACT 68KD Figure 35. ACT infusion leads to decreased phosphor ylation, thus inactivation, of ERK as indicated by the arrows on a westen blot of whole brain homogenates from htau mice. 156

PAGE 173

Figure 36. Infusion of ACT into the hippocampus of a PDAPP mouse induces phosphorylation of endogenous mouse tau at Thr 231 (left) as compared to aCSF into the contralateral hemisphere (right). 157

PAGE 174

158 CHAPTER 9 THE EFFECTS OF CH RONIC INTRACEREBRAL INFUSION OF ACT ON AMYLOID PLAQUES: INHIBITION WITH AETA RELATED PEPTIDES 9.1. Introduction Previous work in our laboratory and ot hers have demonstrated that inflammation and an acute phase response are initiated in the extracellular periphery of amyloid plaques (Abraham et al., 1988; Potter et al ., 2001). This inflammation appears to be initated when increased levels of oligomeric/fibrillar A activate microglia, inducing them to release cytokines such as IL-1 IL6 and TNF IL1 and IL6, in turn trigger the production and release of ACT and apoE from astrocytes (Das and Potter, 1995). These two products of inflammation have been term ed pathological chaperones because they accelerate the fibrillization of the Alzheimer A peptide and subsequent amyloid plaque formation. ACT, a serine protease inhibito r (serpin) is serum glycoprotein that is normally made in the liver in response to infl ammation and inhibits proteases in order to maintain homeostasis and prevent undue tissue destruction during the acute phase response. ACT is upregulated significantly in the brain of AD patie nts (Porcellini et al., 2008). ApoE is also upregulated and released from astrocytes and microglia during brain damage and inflammation. The ApoE4 isoform has been shown to be the strongest heritable risk factor for th e disease and is also released by astrocytes and microglia during the acute phase response (Abraham, 2001 ). Evidence that polymorphisms in the ACT gene are also risk factors for developi ng is accumulating but does not come near the strength of the apoE4 eff ect (Nilsson et al., 2004). It has been demonstrated that synthetic A peptides self-polymerize in vitro under various physiological conditions to form filame nts that resemble the beta-pleated amyloid

PAGE 175

159 structures found in AD by transmission electron microscopy (TEM) (Castelletto et al., 2008) and attenuated total reflection-FTIR spectroscopy (Cerf et al., 2009). These preparations of filaments have also proven to be neurotoxic to primary neuronal cell cultures. The polymerization of A it was found, could be accelerated by ACT, and to a greater extent by ApoE4. Moreover, it has been established that ACT and ApoE each bind to different regions on the AB peptide (Ma et al., 1994). ACT binds to the hydrophilic N-terminus (amino acids 2-9) a nd ApoE to the 12-28 region of the A peptide. These small fragments, when incubated with full length A and their respective pathological chaperones (ACT and ApoE) inhibit the catalyzed polymerization of fibrils in a dose-responsive manner. When these react ions were added to cultures of human cortical neurons, it was found th at the two inhibitory peptid es prevented apoE and ACT respectively from promoting A polymerization and neurotox icity, thus enhancing cell survival, as visualized by MTT assays (Ma et al., 1996). Work with transgenic mice has extende d and supported the cell culture findings described above. Breeding PDAPP mice with and without endogenous mouse ApoE and and with and without a hACT transgene de monstrated that both ACT and ApoE act independently, but synergistically to facili tate the formation of amyloid plaques. Moreover, it was determined that the regi ons of the brain containing plaques were somewhat different. In the absence of ApoE, ACT-positive plaques were localized to the hippocampus and in the presence of ApoE, ACT-positive plaques were found in the hippocampus and cortex (Nilsson et al., 2001). These experiments also showed that cognitive deficits in the P DAPP mice, as assayed by the Mo rris Radial Arm Water Maze and the Radial Arm Water Maze, depende d on the presence of ACT and/or apoEpromoted amyloid deposits (Bales et al., 1999; Nil sson et al., 2001) In December 2006, Wiesniewski et al. created a compound named A 12-28P representing a modified 1228 peptide that reduced amyloid burden when injected systemically into APP-expressing mice. Th e modifications to the peptide included exchanging the valine in position 18 for proline, thus rendering the peptide nonfibrillogenic and preventing the molecule from binding to plaques. Additionally, AB12-28P was synthesized in a slightly modified version by using D-amino acids and

PAGE 176

160 end-protected by acetylation and amidation of the N and C termini, respectively. These modifications decreased the potential for i mmunogenicity and extended the serum halflife of the peptide. AB12-28P was, however, still able to cross the BBB and to inhibit apoE/A binding with these modifications. Specifically, the compound was able to decrease amyloid burden by 51.6% and 45.6% in the cortex and hippocampus, respectively, of 7 month old Tg2576 mice and 50.7% and 40.0% in PS/APP (PS-Tg2576) mice (Sadowski et al., 2006). Evidentally, the ability of A 12-28 to serve as a decoy peptide and prevent apoE from binding to A in vitro was duplicated in vivo and suggests that the modified peptide designed by Wisniewski and colleagues is a potential therapy for Alzheimers disease. Because the effect of the interaction of A peptides with ACT on the formation of amyloid was well established, we desired to de termine whether this interaction could be exploited to develop a novel Alzheimer thera py. We first studied the effects of ACT on existing plaques by infusing purified human AC T intracranially into PDAPP mice at ~ 8 months of age and assessing the morphology of the plaques. In the next phase of the research, we wished to determine whether the N-terminus of Ab could inhibit the ACT/A interaction in a manner similar to the inhibition of the ApoE/A interaction in vivo by the modified 12-28 peptide. To this end, experiments were designed in which ACT was co-infused with A 1-11 into the hemispheres of PS/APP mice as was the A 1-11 fragment alone. 9.2. Materials and Methods 9.2.1. Catheter Construction The catheters (patent pendingPCT/US08/ 73974) used for this experiment were made by first taking a 1.25 cm length of 30 gauge stainless steel tubing (Small Parts, Inc. Miramar, Fl) and carefully bending it at 2.5 mm under a dissecting microscope (Leica, Heerbruug, Switzerland), to approximately 90 degrees, being careful not to crimp the tube. The remaining length of the tube was bent again at 5 mm to an angle of 120-160

PAGE 177

161 degrees, approximating the contour of each animals skull. These metal cannulae were inserted into a 3 cm length of polyethylene (PE-10) tubing with an internal diameter of 0.28 mm and an outer diameter of 0.61 mm. One centimeter of the PE-10 was then itself inserted into a 4.5 cm lengt h of sterile polyvinyl tubing (PV-50-I.D. 0.69 mm/ O.D. 1.14 mm, Durect Corp., Cupertino, Ca), held in place using a bead of Locktite 454 adhesive (Plastics One. Roanoke, VA.), and cured overn ight. The following day, sterile water was forced through each catheter assembly to ensu re that the lines were not obstructed. The osmotic pumps were then filled with solution, attached to the catheters and primed and implanted as described below. The proven ca theters were later custom manufactured by Braintree Scientific (Brain tree, MA) with some minor changes: The adhesive was supplanted by heat sealing PE-10 to PE-50 tubing and, because the diameters of the PV50 and new PE-50 were not exactly the same, a sheath of silicone (SIL 047) was placed over the PE-50 with an extra 2 mm overhang to attach it to the flow modulator of the pump, forming a tight, leak-proof seal. 9.2.2. Transgenic Mice PS/APP (presenilin 1/amyloid precursor protein) mice and PS/APP/ACT mice were generated by crossing heterozygous PDGF-hAPP (V71 7F) mice with PDGF-hPS1 (M146L) on both Swiss Webster and C57BL/ 6 backgrounds. Genotyping was performed using comparative real-time PCR (Bio-Rad iC yclerHercules, CA). Pathologically, these Alzheimer's mouse models are characterized by robust accumulation of amyloid plaques and the development of microgliosi s between 6-8 months of age. 9.2.3. Materials Purified human -1 antichymotrypsin (ACT) was purchased lyophilized from Fitzgerald (Concord, MA) and wa s reconstituted in artificial CSF (Harvard Apparatus, Holliston, MA) to a concentration of 1mg/mL. The A 1-11 was suspended at 5 mg/mL (Rpeptide, Bogart, GA) ACT was infused directly into the hippocampal parenchyma for

PAGE 178

162 28 days using the novel cathete rs attached to Alzet osmotic minipumps (Alzet model 1004, Durect Corp. Cupertino, CA) with an average flow rate of 0.12 L/hour. The pumps and catheters were submer ged in 0.9% sterile saline at 37oC and primed for 48 hours prior to implantation. 9.2.4. Intracranial Infusions All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Animals (8 month-old PS/APP25-35g) were anesthetized with 12% isoflurane, shaved and scrubbed with 10% Betadine solution at the site of incision, and placed into a stereotaxic frame (Kopf In struments. Tujunga, Ca.). A small (5 cm) incision was made, exposing the skull and neck, and double bladed scissors were used to form a subcutaneous pocket along the back of the animal into which 2 osmotic minipumps were inserted with catheters attach ed. Two holes were drilled into the skull (from Bregma -2.2-2.5 mm anterior-posterior +/2.2-2.5 mm medial -lateral, and the 30 gauge catheters were inserted to a dept h of 2.2-2.5 mm) corresponding to the posterior portion of the CA3 hippocampus. Alternativel y, holes were drilled corresponding to the lateral ventricles (from Bregma -0.2 mm ante rior-posterior, +/1.0 mm medial-lateral, and to a depth of 2.2-2.5 mm). Once they were inserted, the cannulae were affixed to the skull with Locktite 454 adhesive (Plastics One. Roanoke, VA.) and secured down with a piece of nitrile, approximately 1 cm in diamet er (Fig 3D). After the adhesive cured, the scalp was closed with 6-7 silk sutures. Fo r analgesia, the animal received an immediate dose of ketoprofen (10mg/kg) and additional doses once every 6 hours, as needed, for a maximum of 48 hours post-operatively. After the 28 day period, the animals were given an overdose (~150mg/kg) of sodium pentobarbita l (i.p.) and were tran scardially perfused with 0.9% saline. The brains were carefully removed and analyzed as outlined below. At this time, the integrity of the catheters wa s confirmed again by forcing water through, and the pumps were also determined to be empty.

PAGE 179

163 9.2.5. Histology and Immunohistochemistry Mouse brain tissues were fixed in 10% formaldehyde (formalin) for 72 hours and then passed through a series of sucrose soluti ons (10-30%) over anothe r 72 hours. Brains were frozen to the peltier stage (Physit emp, Clifton, NJ) of a histoslide (Leica, Heerbruug, Switzerland) and s ectioned coronally at 25 m. After incuba ting the sections with blocking buffer (Tris-buffered saline with 10% normal goat serum, 0.1% Triton X100, and 0.02% sodium azide) for 60 mi n, primary antibodies against A (6E10 and 4G8,1:1000Covance, Princeton, NJ) as well as Iba1(1:1000) (Wako Chemicals, USA. Inc., Richmond, VA.) were applied and incubated at 4 o C overnight. After thorough washing, the sections were incubated w ith secondary antibodies, Alexa 488 and 594 fluorophores (1:1000, 1:4000-Invitrogen) in the da rk at ambient temperature for 2 hours. This incubation was followed by a Hoechst (1:10,000, Sigma, St. Louis, MO) nuclear stain, and the sections were washed and s ealed with Gel-Mount (Electron Microscopy Sciences, Hatfield, PA). The images were analyzed on a Zeiss Imager Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany) using Axiovision 4.7 software and quantified using Image J. 9.3. Results and Conclusions Previous work from our and othe r laboratories has shown that A is the primary component of amyloid plaques found in the AD br ain. It has been shown in vitro that A can self polymerize into fibrils and that this process is accelerated by extracellular pathological chaperones (Wis niewski and Frangione, 1992). The term pathological chaperone which was coined by Thomas Wiesnieski in 1992 is used to describe molecules that promote a pathological change in a protein of interest, in our case A Molecular chaperones, on the other hand, are responsible for promoting the folding of proteins into their physiologica l, rather than a pa thological conformation. Previous work has demonstrated that both ACT and ApoE ha ve roles as pathological chaperones in AD. Not only can they accelerate the formation of A fibrils in vitro (Ma et al., 1994), but

PAGE 180

164 transgenic animal studies have also reveal ed that these molecules bind directly to A fibrils and become part of the ensui ng amyloid plaques (Nilsson et al., 2004). ApoE and ACT are both upregulated in th e AD brain and released by residing glia as part of the chronic inflammation induced by A (or A oligomers). The normal psysiological roles of these molecules are also known. ApoE is rele ased by hepatocytes and is the major binding protei n of LDL during cholesterol tr ansport to tissues. ACT, on the other hand, is released by hepatocytes and neutrophils to inhibit cathepsin G during inflammation to remodel damaged tissues a nd possibly engulf pathogens (Kalsheker, 1996). ACT is a serine protease inhibitor that is produced in concert with se veral serine proteases that would otherwis e go out of homeostatic balanc e during growth and repair and cause damage to cells and tissues. The normal roles of ApoE and ACT in the brain, however, are not as clear. One possibility for the presence of ACT was proposed by our lab some time ago: Because the structural homology between cathepsin G, chymotrypsin and A is striking, A might be recognized by ACT as a protease and bound to inhibit its action (Ma et al., 1994). To interfere with the interaction of ACT and A thereby theformation of amyloid, novel strategies have been developed to inhibit the binding of these molecules. Previous therapeutic approaches have been focused on inhibiting the secretase enzymes that overproduce A on immunizing patients against the A peptide, or on manipulating cholinergic and glutaminergic neurotransmi ssion. Inhibiting th e interaction of A with ApoE and ACT respectively, was a new approach that has also proven effective in mouse models. Early experiments showed ACT and ApoE could accelerate the formation of A fibrils in vitro, but also that this promotion could be inhibite d with small fragments of the A peptide itself. The amino acids that correspond to aminoacids 12-28 in the A 1-42 peptide, attenuated the formation of fibrils significantly when incubated with ApoE (Ma et al., 1996). Similarly, when A 1-42 was incubated with ACT and the amino acids that correspond to positions 2-9 of the A 1-42 fragment, attenuated fibril formation was observed as well. Additionally, when these r eactions were cultured with human cortical neurons, MTT cell survival assays determined that the cells were more viable with increasing concentrations of the inhibiti ng peptides. In sum, these experiments

PAGE 181

165 demonstrated that decoy fragments of A could bind ApoE and ACT, thereby preventing these proteins from binding to full length A and accelerating the formation of amyloid. The next phase of this research, in vivo, was carried out by Wiesnieski et al. in 2006. In this set of experiments, the modified 12-28 peptide, previously described, was delivered systemically in mice harboring the Swedish APP mutation and was found to drastically reduce amyloid load and improve cognition. In my experiments described here, the rationale was very similar in that the an A fragment of 11 amino acids, covering the 2-9 region of A was used to interrupt the A /ACT interaction. However, in these experiments, the peptide was deliv ered without modification and via direct infusion into the hippocampus of PS1/PDAPP mi ce. Though preliminary, the delivery of the 1-11 peptide in a small group of animals (n=4) virtually eliminated plaques in the vicinity of the infusions (figures 37 a nd 38quantitation figure 39) as well as the presence of microglia (Iba1 stainingfigur e 40). The use of bot h 6E10 (figure 37) and 4G8 (figure 38) antibodies to show plaque clearance is notable because they bind the 116 and 17-24 residues on A respectively, which correspond to the binding sites of ACT and ApoE. One might expect that the epitope for ACT be bound to the decoy peptide and not available for the antibody, but staining wa s strong using both antibodies. Because these mice did not carry the human ACT gene, the interpretation of this finding is not perfectly clear, but it is known that mice ha ve a serpin homologue (Serpin A3N) that is found in plaques that cross-reacts with the ACT antibody. Perhaps the A 1-11 peptide binds to this endogenous protein, and prevents it from promoting amyloid filament formation. Moreover, it appears that the application of the A 1-11 peptide not only interferes with the formation of plaques, but also contributes to the clearance of already existing plaques.. The lack of microglial staining on the treated hemisphere would suggest that microglia engulfed amyloid deposit s and then departed. Further clarification is currently being pursued with additional experiments. Howe ver, it is already evident from this study and the work perf ormed by Wiesniewski et al. that A fragments make effective decoys to bind to the pathol ogical chaperones ApoE and ACT and provide another potential therapy for AD.

PAGE 182

9.4. Figures A 1-11 infusion site #981 Figure 37. Infusion site of A 1-11 into 8 month old PS/APP mice (n=4) immunolabeled with 6E10 showing marked plaque reduction. 166

PAGE 183

A 1-11 infusion site #981 Figure 38. Infusion site of A 1-11 into 8 month old PS/APP mice (n=4) immunolabeled with 4G8 showing mark ed plaque reduction. 167

PAGE 184

0.00 1.00 2.00 3.00 4.00 5.00 6.00 4G8 6E10 control (aCSF) ab1-11 Figure 39. Semi-quantification of 4 mice treated with A 1-11 stained with both 6E10 and 4G8 antibodies, with significance of p= 0.024 and p= 0.058 respectively. 168

PAGE 185

Figure 40. A) Immunostaining of A 1-11 treated mice with the micr oglial marker Iba1 showing an absence of microglia on the treated side 169

PAGE 186

0 0.5 1 1.5 2 2.5 3 3.5 4control (aCSF) A 1-11 Figure 40. B) Quantification (semi) of above imm unohistochemical analysis of Iba1 (n=2) 170

PAGE 187

171 CHAPTER 10 ARTHRITIS-INDUCED HEMATOPOIESIS REPAIRS PATHOLOGY AND MEMORY IN ALZHEIMERS MICE *Tim D. Boyd, PhDc 1,2,3, *Steven P. Bennett, PhD1,2,3, Takashi Mori, DVM,PhD,7 Nickolas Governatori, BS5, Melissa Runfeldt,BS5, Michelle Norden1, Jaya Padmanabhan, PhD1,2, Peter Neame, PhD1,2, Inge Wefes, PhD2,3, Juan Sanchez-Ramos, PhD, MD1,4, Gary W. Arendash, PhD5,6 Huntington Potter, PhD1,2,3,6 *Both authors contributed equally to this work 10.1. Abstract Objectives: Rheumatoid arthritis (RA) is a negative risk factor for the development of Alzheimers disease (AD). Wh ile commonly assumed that RA patients usage of nonsteroidal anti-inflammatory drugs (NSAIDs) helped prevent onset and progression of AD, NSAID clinical trials have proven unsuccessf ul in AD patients. To determine whether intrinsic factors within RA pathogenesis itself may underlie RAs protective effect, we investigated the activity of colony-stimula ting factors, up-regulated in RA, on the pathology and behavior of transgenic AD mice. Methods : 5 g bolus injections of macrophage, gra nulocyte, and granul ocyte-macrophage colonystimulating factors (M-CSF, G-CSF, or GM-CSF) were administered unilaterally into the hippocampus of aged cognitively-impaired AD mice and the resulting amyloid load reductions determined one week later, using the artificial cere brospinal fluid (aCSF)injected contralateral sides as controls. 20 da ily subcutaneous injec tions of 5g of GMCSF (the most amyloid-reducing CSF in the bolus experiment) were administered to balanced cohorts of AD mice after assessment in a battery of cognitive tests. Reductions in amyloid load and improvements in cognitive function were assessed.

PAGE 188

172 Results : G-CSF and more signifi cantly, GM-CSF reduced amyloidosis throughout the treated brain hemisphere one week following bolus administration to AD mice. Moreover, subcutaneous GM-CSF administratio n significantly reduced brain amyloidosis and completely reversed the cognitive impairment, while increasing hippocampal synaptic area and microglial density. Interpretation: These findings, along with two decad es of accrued safety data using Leukine, recombinant human GM-CSF, in el derly leukopenic patients, suggest that Leukine should be tested as a treatment to reverse cerebral amyloid pathology and cognitive impairment in AD. 10.2. Introduction Although numerous studies have reported Rh eumatoid arthritis (RA) to reduce the risk of Alzheimers disease (AD), the mechan isms for RAs protec tive effect are still unknown (McGeer et al., 2006). These repo rts have commonly assumed that RA patients usage of non-steroidal anti-inflamm atory drugs (NSAIDs) may help prevent the onset and progression of AD (McGeer et al., 200 6), but the largest NSAID clinical trials have not demonstrated efficacy in reducing the incidence of dementia, and recently Naproxen was reported to be detrimental, wi th increased risk of cardiovascular and cerebrovascular events (Martin et al., 2008). These results sugge sted to us that intrinsic, probably immunological factor s within RA pathogenesis itself, may underlie RAs protective effect against AD. We surmised that up-regulated local cellular populations in RA would have the highest poten tial to enter into the brai n and inhibit the development of AD pathology and/or neuronal dysfunction. Alzheimers disease is an age-related, pr ogressive neurodegenerative disorder that presents as increasing decline in cognitive and executive function. Alzheimer dementia is associated with cerebrovascular dysf unction (Humpel and Marksteiner, 2005), extracellular accumulation of amyloid (A ) peptides in the br ain parenchyma and vasculature walls (Rhodin et al., 2000; Scheuner et al., 1996) (predominantly A 1-42 and A 1-40), and intraneuronal accumu lation of neurofibrillary tangles consisting of

PAGE 189

173 hyperphosphorylated Tau proteins (Pad manabhan et al., 2006). Associated neuroinflammation may contribute to AD pat hogenesis (Griffin et al., 1989), as the inflammatory proteins apolipoprotein E (apoE) and 1-Antichymotrypsin (ACT) catalyze the polymerization of A peptides into amyloid filaments in vivo and in vitro (Ma et al., 1996; Nilsson et al., 2004; Padmanabhan et al., 2006; Potter et al., 2001; Wisniewski et al., 1994). Conversely, it has also been shown th at amyloid plaques form rapidly and then become decorated by microglia (Koenigsknech t-Talboo et al., 2008; Meyer-Luehmann et al., 2008), both resident and bone marrow-derived, suggesting an ability and intention to remove amyloid (Malm et al., 2005; Simard and Rivest, 2004; Simard et al., 2006). Thus it is unclear whether neuroinflammation is de leterious or beneficial in the AD brain, and indeed the role of microglia in AD is co mplex and may involve different states of activation with different activities. Rheumatoid arthritis is an autoimmune disease in which inflamed synovial tissue and highly vascularized pannus form, irreparably damaging the cartilage and bone. In this inflammatory pannus, leukocyte populat ions are greatly expanded, and many proinflammatory factors are produced that wo rk together in feed-forward mechanisms, further increasing leukocytosis, cytokine/c hemokine release, osteoclastogenesis, angiogenesis, and autoantibody production (rheu matoid factors and anti-citrullinated protein antibodies) (Schellekens et al., 2000; Szekanecz and Koch, 2007). Additionally, the adaptive immune system pres ents a Th17 phenotype within CD4+ lymphocytes, with ultimate production of interleukin 17 (IL-17) wh ich is then responsible for inducing much of the pro-inflammatory effects (Cox et al., 2008; Parsonage et al., 2008). Further enhancements of leukocyte populations come fr om increased expression of structurallyunrelated colony-stimulating factors: M-CSF (macrophage), G-CSF (granulocyte), and GM-CSF (granulocyte-macrophage) (Kawaji et al., 1995; Nakamura et al., 2000; Olszewski et al., 2001; Xu et al., 1989). Although up-regulated leukocytes in RA coul d potentially enter into the brain and inhibit development of AD pathology and/or neuronal dysfunction, lymphocytic infiltrates into AD patient brai ns have not been reported. The lack of infiltration suggests that proliferation and activati on of the innate immune system might be responsible for

PAGE 190

174 preventing AD pathology in RA patients. Evidence supporting the innate immune systems role in AD pathogenesis show that complement proteins are up-regulated in AD brain, and that inhibition of C3 convertase significantly in creases amyloid pathology in AD mice (Wyss-Coray et al., 2002). Bone marrowderived microglia also play a critical role in restricting amyloid deposition, but th is association and many associated receptors and enzymes, such as CD36, scavenger recepto r A, and receptor for advanced glycation end products, or neprilysin, insulysin, and matr ix metalloproteinases, decline with age, while AD pathology increases (El Khoury et al ., 2007; Hickman et al., 2008; Simard et al., 2006). To investigate the interplay of the innate immune system and AD, we studied the effects on AD pathology of the three colony-stimulating factors (M-CSF, G-CSF, and GM-CSF), which are up-regulated during RA pathogenesis (Kawaji et al., 1995; Nakamura et al., 2000; Olszewski et al., 2001; Xu et al., 1989). These CSFs enhance the survival and function of their respective leukocytes and drive thei r proliferation and differentiation from myeloi d lineage precursors. GM-CSF induces dendritic cells, macrophages, and granulocytes (neutrophils basophils, and eosinophils), while M-CSF and G-CSF respectively induce the macrophage and granulocyte subsets of the innate immune system. These innate cells have the ability to diapedese from the circulatory system and to differentiate further into vari ous specialized immune cells within organs (microglia, Langerhans cells, etc.). GM-CSF and G-CSF are also known to be involved in erythropoiesis, and GM-CSF and erythropoietin act synerg istically in the maturation and proliferation of the bur st-forming and colony-forming erythroid units to the normoblast stage of erythropoiesis (Fisher, 2003). Circulating A binds to complement opsonin C3b in an antibody-independent fash ion, and C3-opsonized particles bind to the complement receptor CR1on erythrocytes and to CR1g on liver-resident kupffer macrophages (Helmy et al., 2006; Rogers et al., 2006). Thus GM-CSF could function in both the peripheral clearance of A and in bone marrow-derived microglial activity, since it is involved in the proliferation, differe ntiation, and maintenance of most innate leukocytes.

PAGE 191

175 Here, we report on experiments that invest igated the effect of CSF administration on amyloid plaque deposition, microglial activ ation, synaptic func tion, and associated cognitive decline in a mouse model of AD. Our results, particularly with GM-CSF, provide a compelling explanati on for RAs inverse relationship with AD. Moreover, the reduction of amyloidosis and enhancement of cognition by GM-CSF warrant clinical investigation of Leukine for the treatment of Mild Cognitive Impaired (MCI) and AD patients, especially with Le ukines long-standing safety hi story in leukopenic patients. 10.3. Materials and Methods All procedures involving experimentation on animals were performed in accordance with the guidelines set forth by the University of South Florida Animal Care and Use Committee. Transgene detections were performed using QPCR (Bio-Rad iCycler, Hercules, CA). 10.3.1. Transgenic Mouse Studies Involving Intracerebral Administration of CSFs PS/APP mice in this study, which begi n accumulating robust amyloid plaques at 6-8 months, were generated by crossing he terozygous PDGF-hAPP (V717F) mice with PDGF-hPS1 (M146L) on both Swiss We bster and C57BL/6 backgrounds. M-CSF was bilaterally infused directly into the lateral ventricles (5 g/day) for 14 days using a novel intracranial cat heter infusion system (Benne tt, 2009). This completely subcutaneously-contained system allows bilateral intracerebral infusion of test substances ipsilaterally and vehicle c ontralaterally, and overcomes the problem of amyloidosis variance between animals, effectively making each animal its own control (Supplementary Figure 1). All three CSFs were stereotaxically-inject ed (5 g/injection) into the (ipsilateral) hippocampus, with aCSF vehicle injected contra laterally into four PS/APP mice each (all 10-12 months old, 25-35 g, both genders). Two holes were drilled into the skull (from bregma -2.5 mm anterior -posterior, +/2.5 mm medial-lateral, and the 30 gauge needle inserted to a depth of 2.5 mm). Mice were perf used with 0.9% cold

PAGE 192

176 saline 7 days later and their brains placed in 10% neutral buffered formalin. Recombinant mouse GM-CSF (rmGM-CSF), recombinant murine G-CSF (rmG-CSF), and recombinant mouse M-CSF (rmM-CSF) (R&D Systems, Minneapolis, MN) will be referred to as GM-CSF, G-CSF, a nd M-CSF throughout this publication. 10.3.2. Immunohistochemistry and Image Analysis of Intrahippocampal-injected Mice Formalin-fixed brains were either cor onally cryosectioned at 14-m, or paraffinembedded and sectioned at 5-m, with stan dard deparaffination and antigen retrieval steps (boiled in 10mM Sodium Citrate buffer for 20 minutes) performed before immunohistochemical staining of 5 sections at 150 m intervals. To significantly reduce cost of reagents and antibodies with paraffin-embedded slides, a novel magnetic immunohistochemical staining device was developed (patent pending, Tech ID# 09A015). Standard fluorescent immunohistoche mical techniques used primary anti-A antibodies 6E10 (Covance, Emeryville, CA, 1:1000), and MabTechs 3740-5 (MabTech, Cincinnati, OH, 1:5000) to immunolabel amyloid deposition coupled with Alexa fluorophore-labelled secondary antibodies (Molecular Probes, Eugene, OR, 1:1000, 1:4000), and Hoechst (Sigma) nuclear stai ning. Immunofluorescence was detected on a Zeiss Imager.Z1 microscope (Oberkochen, Ge rmany) using Axiovision 4.7 software and digital images quantified using ImageJ (Supplementary Figure 5-6 Appendix B). 10.3.3. Behavioral Transgenic Mouse St udy Involving GM-CSF Treatment Mice in this study were derived from th e Florida Alzheimers Disease Research Center mouse colony, wherein heterozygous mice carrying the mutant APPK670N, M671L gene (APPsw) are routinely crossed wi th heterozygous PS1 (Tg line 6.2) mice to obtain APPsw/PS1, APPsw, PS1, and non-transg enic (NT) genotype offspring with a mixed C57/B6/SW/SJL background. Eleven A PPsw, 4 APPsw/PS1, and 17 NT mice, all 12-months old, were selected and evaluated for 8 days in the RAWM task of working memory as previously described (Arendash et al., 2001) and Suppl ementary Figure 7).

PAGE 193

177 Numerous experiments have revealed that various genotypes of AD mice perform equally once they reach cognitive impa irment. Thus, the 15 Tg mice were divided into two groups, balanced in RAWM performance, with 2 APPsw/PS1 mice included in each group. The 17 NT mice were also divided into two groups, balanced in RAWM performance. Two weeks following pre-tr eatment testing, one group of Tg mice ( n = 7) and one group of NT mice ( n = 9) were started on a 10-day treatment protocol with GM-CSF (5 g/day given subcutaneously), while animals in the control Tg and NT groups ( n = 8 per group) concurrently received daily vehicle (saline) treatment subcutaneously. On the 11th day of injections, all mice began four days of RAWM evaluation, were given 2 days of rest, then ev aluated in 4 days of Cognitive Interference task testing as previously described (Echev erria et al., 2009; Loew enstein et al., 2004). Daily GM-CSF and saline injections were c ontinued throughout the behavioral testing period. After completion of behavioral testing at 20 days into treatment, all mice were euthanatized, brains fixed as described above, and paraffin-embedded. Careful visual examination of all tissues upon necropsy revealed no morphological abnormalities, and the mice tolerated daily subcutaneous injectio ns well. Each analysis was done by a single examiner blinded to sample identities, and st atistical analyses were performed by a single examiner blinded to treatment group identitie s. The code was not broken until analyses were completed. 10.3.4. Immunohistochemistry and Image Analysis of Subcutaneous GM-CSF-treated Mice Five 5m sections (150m apart) were made of formalin-fixed, paraffinembedded sections throughout the hippocampus of each mouse and immunoreactivity was developed using the Vectastain ABC Elite kit (Vector Laboratories, Burlingame, CA) coupled with the diaminobenzidine reaction, according to the manufacturers protocol. Immunostaining us ed biotinylated anti-A clone 4G8 (Covance, Emeryville, CA, 1:200), synaptophysin (DAKO, Carpinte ria, CA, undiluted), and Iba1 (Wako, Richmond, VA, 1:1000) as primary antibodies. Images were acquired using an Olympus

PAGE 194

178 BX60 microscope and digital images were qua ntified using SimplePCI software (Compix Inc., Imaging Systems, Cranberry Township, PA), according to previous methods (Sanchez-Ramos et al., 2009). Each analysis was done by a si ngle investigator blinded to sample identities and genotype. 10.4. Results 10.4.1. Intracerebral administration of CSFs Following bilateral intracerebroventricular infusion of M-CSF for two weeks into PS/APP mice, immunohistochemical analysis of both experimental and control mice showed considerable variances of amyloid deposition between mice of similar age (Supplementary Figure 1-Appendix B), signi ficantly compromising our ability to determine M-CSFs effect in a limited m ouse cohort. While improving our drug delivery system by developing novel bilateral brain infusion catheters (Be nnett, 2009), we found that parenchymally-infused recombinant pept ides remained localized to the infused hemisphere. These findings led us to administer the CSFs as a unilateral intrahippocampal bolus with a c ontralateral injection of vehicle as cont rol, thus obviating the need for large numbers of transgenic mice and age-matched littermate controls to obtain statistical significance. Each CSF was stereotaxically injected into the hippocampus of 4 mice, with artificial cerebrospinal flui d vehicle (aCSF) injected contralaterally. The mice were sacr ificed 7 days post-injection. M-CSF injections into mice resulted in sw elling of the entire hemisphere as compared to the control side, and in one m ouse, an apparent hyperplasia had formed at the injection site (Supplementary Figure 2Appendix B). Quantification of amyloid plaque loads from anterior to posterior of each mouse showed similar deposition in the M-CSF-injected hemispheres as compared to the control sides (d ata not shown). In contrast to M-CSF, G-CSF injections did not induce swelli ng and showed some modest reductions of amyloid deposition (Supplemen tary Figure 3Appendix B), which was

PAGE 195

179 subsequently corroborated by independent observations from fellow investigators (Sanchez-Ramos et al., 2009). GM-CSF-injections, however, demonstrat ed pronounced decreases in amyloid deposition, as compared to control hemisphere s, in visual observati ons of coronal tissue sections (Figure 41a; Supplementary Figure 4Appendix B). Quantification of amyloid plaques anterior to posterior revealed significant reductions within individual mice and overall significant reductions for all pla que parameters measured (Figure 41b and Supplementary Figure 5Appendix B). 10.4.2. Daily subcutaneous injection of GM-CSF Based upon the positive results from intrah ippocampal injections, we investigated the effect of subcutaneous GM-CSF injection on AD pathology and cognitive function. Prior to GM-CSF treatment, APPsw+PS1(Tg) mice were first confirmed by RAWM testing to be cognitively-impaired for work ing memory. Both the non-transgenic control mice (NT) and the Tg mice were then sub-divided into two cognitively-balanced groups, for either GM-CSF or saline treatment. RAWM testing post-injection re-confirmed that Tg control mice were substantially impair ed compared to NT control mice. This impairment was evident in individual blocks of testing, but also over all 4 days of testing (Figure 42a). In sharp contrast, GM-CSF-treated Tg mice performed equally well or better than NT control mice during individual blocks and overall. GM-CSF-treated NT mice performed as well as or slightly be tter than NT contro ls (Figure 42a). Before evaluation in the Cognitive Inte rference Task, the mice rested two days. This task mimics human interference testi ng, which discriminates between normal aged, MCI, and AD patients (Loewenstein et al., 2004 ). In all four cognitive interference measures assessed over 4 days of testing (Figure 42b), Tg control mice were clearly impaired compared to NT mice, and Tg mice treated with GM-CSF exhibited significantly better three-trial recall and delaye d recall compared to Tg controls. Indeed, for all four cognitive measures, GM-CSF-treated transgenic AD mice performed similarly to NT mice. A partic ularly strong effect of GM-C SF treatment in Tg mice was

PAGE 196

180 evident for the proactive interf erence measure during the first half of testing (Figure 42c), wherein GM-CSF-treated Tg mice performed substantially be tter than Tg controls and identically to both groups of NT mice. Pro active interference sus ceptibility has been reported to be a more sensitive marker for differentiating MCI and AD patients from aged normals than traditional measures of delayed r ecall and rate of forgetting (Loewenstein et al., 2004). Parenthetically, even the GM-CSFtreated NT mice showed a trend towards improved cognition in behavioral studies, albe it not statistically significant. Subsequent analysis of brains from Tg mice of this study revealed that GM-CSF treatment induced large reductions in amyloid burdens within entorhinal cortex ( 55%) and hippocampal ( 57%) compared to control Tg mice (Figure 43). The improved cognitive function and re duced cortical amyloidosis of GM-CSFtreated Tg mice were paralleled by increased microglial density as compared to salinetreated Tg mice (Figure 44), implying an a ugmented ability to bind and remove amyloid deposition (El Khoury et al., 2007; Hickman et al., 2008). The GM-CSF-treated Tg mice similarly demonstrated increased synaptophys in immunoreactivity in both CA1 and CA3 regions (Figure 45), indicating increased syna ptic area in these hippocampal areas. Prior work has shown that adult neural stem cells in hippocampal dentate gyrus (DG) express GM-CSF receptors, and GM-CSF increases neur onal differentiation of these cells in a dose-dependent fashion (Kruger et al., 2007). Thus, one mechanism for the observed GM-CSF-induced cognitive improvement is enhanced removal of deposited A in hippocampus, with ensuing neuronal growth/s ynaptic differentiation of DG mossy fiber innervation to CA3, resulting in increased innervation/synaptogenesis of Schaffer collaterals into CA1. Removal of deposited A from entorhinal cortex may also increase perforant pathway viability to hippocam pal projection fields in DG and CA1. Thus GM-CSF-induced reduction of amyloidosis and enhancement of hippocampal/entorhinal cortex circuitry, crit ical for working (short-term) memory, may underlie GM-CSFs reversal of working memo ry impairment in Alzheimers Tg mice.

PAGE 197

181 10.5. Discussion Since peripheral leukocyte popul ations are increased in RA and possess the ability to infiltrate into the brain, we initially investigated M-CSF, G-CSF, and GM-CSF to determine which CSF might affect amyloidosis. In the vasculature, all three CSFs work to drive the proliferation, differentiation, and survival of their respective innate leukocytes from monocytic precursors. In our study, we found different functional effects for each intrahippocampalinjected CSF. In the M-CSF injected mi ce, there was no effect on amyloidosis but pathological changes were noticed, such as swelling and hyperplasia (Supplementary Figure 2Appendix B). Parenchymal overexpre ssion of M-CSF in any organ is probably not advisable as overexpression of M-CSF a nd/or its receptor within mammary glands has similarly resulted in tumor formation and hyperplasia (Kirma et al., 2004). In a study by Boissonneault et al. (2009), the authors published that chronic intraperitoneal injection of M-CSF prevents and reverses amyloid deposition and cognitive impairment and induces brain accumulation of bone marro w-derived microglia (Boissonneault et al., 2009). Although the authors did not relate their findings to RAs inve rse relationship with AD, their data provides evidence that up -regulated M-CSF in RA pathogenesis and systemic administration of M-CSF in AD patients may impart protection against AD onset or progression. Peripheral administration of G-CSF has also been recently found to ameliorate amyloid pathology and reduce c ognitive defects in AD models (SanchezRamos et al., 2009), which corroborates our observations of modest amyloid reduction by intrahippocampal injection. Although the M-CSF and G-CSF findi ngs are encouraging, our GM-CSF intrahippocampal injections into an AD mouse model demonstrated a much more pronounced reduction of amloidosis. Furthermor e, subcutaneous administration of GMCSF resulted in reversal of cognitive impairment resulting in function similar to that of wild-type mice, and an averag e of about 50% reduction of amyloidosis. Explanations for these very robust effects are multiple, incl uding the aforementioned augmentation of peripheral erythropoietic amyloid-clearance mechanisms (Fisher, 2003; Helmy et al.,

PAGE 198

182 2006; Rogers et al., 2006), as well as incr eased neurogenesis (K ruger et al., 2007), increased cerebral angiogenesis (Schneider et al., 2007), neuroprotection from apoptosis (Schabitz et al., 2008), reduction in amyloi dosis (Figures 41, 43, and Supplementary Figure 5Appendix B) and increased neur onal plasticity (Figur e 45). These effects mirror those reported for G-CSF (Sanchez-Ramo s et al., 2009), and supports the current practice of interchangeable prescription of either recombinant human GM-CSF (Leukine) or G-CSF (Granocyte, Neupogen, and Neulas ta) into patients with depressed bone marrow function. G-CSF primarily treats neutropenia while GM-CSF treats all leukopenia, and both have long records of safety data from two decades of FDAapproved usage. Rare adverse events are us ually mild febrile incidents that quickly subside upon cessation of administration. G-CSF is currently in clinical trial for stroke and was recently approved for an AD Phase II clinical trial. However, GM-CSF/Leukine is more effective in the AD mouse mode l and while neutrophils are short-lived leukocytes, and the fact that GM-CSF induces up-regulation of all inna te cells means that it could potentially impart prolonged protective effects against AD. The beneficial effects of CSFs, especial ly GM-CSF in mouse models of AD, point to a potential new approach to AD therapy and support our hypothesis that up-regulated leukocytosis, induced in RA, may impart the pr otective effect of RA against AD. Indeed, they also indicate, with the failure of NSAID clinical trials in AD which encouraged us to develop and test our hypothesis that intrinsic pathogenic properties of RA are protective against AD, that age-linked depressed hematopoiesis may be etiological for AD pathogenesis. Acknowledgements: Support for this work was provided for by the Johnnie B. Byrd Sr. Alzheimers Center and Research Institute, the Eric Pfeiffer Ch air for research on Alzheimers disease, and the Florida Alzheimers Disease Research Center (P50AG25711).

PAGE 199

183 10.6. References Arendash GW, King DL, Gordon MN, Morgan D, Hatcher JM, Hope CE, Diamond DM. Progressive, age-related behavi oral impairments in transgenic mice carrying both mutant amyloid precursor protein and presenilin-1 transgenes. Brain Res, 2001; 891: 42-53. Bennett SP, Boyd T.D., Norden M., Padmana bhan J., Neame P., Wefes I., Potter H. A Novel Technique for Simultane ous Bilateral Brain Infusi ons in a Mouse Model of Neurodegenerative Disease Journa l of Neuroscience Methods, 2009. Boissonneault V, Filali M, Lessa rd M, Relton J, Wong G, Rive st S. Powerful beneficial effects of macrophage colony-stimulating factor on beta-amyloid deposition and cognitive impairment in Alzheime r's disease. Brain, 2009; 132: 1078-92. Cox CA, Shi G, Yin H, Vistica BP, Wawrousek EF, Chan CC, Gery I. Both Th1 and Th17 are immunopathogenic but differ in other key biological act ivities. J Immunol, 2008; 180: 7414-22. Echeverria V, Burgess S, Gamble-George J, Zeitlin R, Lin X, Cao C, Arendash GW. Sorafenib inhibits nuclear factor kappa B, d ecreases inducible nitric oxide synthase and cyclooxygenase-2 expression, and restores working memory in APPswe mice. Neuroscience, 2009; 162: 1220-31. El Khoury J, Toft M, Hickman SE, Means TK Terada K, Geula C, Luster AD. Ccr2 deficiency impairs microglial accumulation and accelerates progression of Alzheimerlike disease. Nat Med, 2007; 13: 432-8. Fisher JW. Erythropoietin: physiology and pharmacology update. Exp Biol Med (Maywood), 2003; 228: 1-14. Griffin WS, Stanley LC, Ling C, White L, MacLeod V, Perrot LJ, White CL, 3rd, Araoz C. Brain interleukin 1 and S-100 immunoreactivity are elevated in Down syndrome and Alzheimer disease. Proc Natl Acad Sci U S A, 1989; 86: 7611-5. Helmy KY, Katschke KJ, Jr., Gorgani NN, Klja vin NM, Elliott JM, Diehl L, Scales SJ, Ghilardi N, van Lookeren Campagne M. CRIg: a macrophage complement receptor required for phagocytosis of circulat ing pathogens. Cell, 2006; 124: 915-27. Hickman SE, Allison EK, El Khoury J. Micr oglial dysfunction and defective betaamyloid clearance pathways in aging Alzheimer's disease mice. J Neurosci, 2008; 28: 8354-60. Humpel C, Marksteiner J. Ce rebrovascular damage as a cause for Alzheimer's disease. Curr Neurovasc Res, 2005; 2: 341-7.

PAGE 200

184 Kawaji H, Yokomura K, Kikuchi K, Somoto Y, Shirai Y. [Macrophage colonystimulating factor in patients with rheuma toid arthritis]. Nippon Ika Daigaku Zasshi, 1995; 62: 260-70. Kirma N, Luthra R, Jones J, Liu YG, Nair HB, Mandava U, Tekmal RR. Overexpression of the colony-stimulating factor (CSF-1) and/or its receptor c-fms in mammary glands of transgenic mice results in hyperplasia and tumor formation. Cancer Res, 2004; 64: 416270. Koenigsknecht-Talboo J, Meyer-Luehmann M, Parsadanian M, Garcia-Alloza M, Finn MB, Hyman BT, Bacskai BJ, Holtzman DM. Rapid microglial response around amyloid pathology after systemic anti-Abeta antibody administration in PDAPP mice. J Neurosci, 2008; 28: 14156-64. Kruger C, Laage R, Pitzer C, Schabitz WR, Schneider A. The hematopoietic factor GMCSF (granulocyte-macrophage colony-stim ulating factor) promotes neuronal differentiation of adult neural stem ce lls in vitro. BMC Neurosci, 2007; 8: 88. Loewenstein DA, Acevedo A, Luis C, Cr um T, Barker WW, Duara R. Semantic interference deficits and the detection of mild Alzheimer' s disease and mild cognitive impairment without dementia. J In t Neuropsychol Soc, 2004; 10: 91-100. Ma J, Brewer HB, Jr., Potter H. Alzhei mer A beta neurotoxicity: promotion by antichymotrypsin, ApoE4; inhibition by A be ta-related peptides. Neurobiol Aging, 1996; 17: 773-80. Malm TM, Koistinaho M, Parepalo M, Vatane n T, Ooka A, Karlsson S, Koistinaho J. Bone-marrow-derived cells contribute to the re cruitment of microglial cells in response to beta-amyloid deposition in APP/PS1 double tr ansgenic Alzheimer mice. Neurobiol Dis, 2005; 18: 134-42. Martin BK, Szekely C, Brandt J, Piantadosi S, Breitner JC, Craft S, Evans D, Green R, Mullan M. Cognitive function over time in the Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT): re sults of a randomized, contro lled trial of naproxen and celecoxib. Arch Neur ol, 2008; 65: 896-905. McGeer PL, Rogers J, McGeer EG. Infl ammation, anti-inflammatory agents and Alzheimer disease: the last 12 years. J Alzheimers Di s, 2006; 9: 271-6. Meyer-Luehmann M, Spires-Jones TL, Prad a C, Garcia-Alloza M, de Calignon A, Rozkalne A, Koenigsknecht-Talboo J, Ho ltzman DM, Bacskai BJ, Hyman BT. Rapid appearance and local toxicity of amyloid-beta plaques in a mouse model of Alzheimer's disease. Nature, 2008; 451: 720-4.

PAGE 201

185 Nakamura H, Ueki Y, Sakito S, Matsum oto K, Yano M, Miyake S, Tominaga T, Tominaga M, Eguchi K. Hi gh serum and synovial fluid granulocyte colony stimulating factor (G-CSF) concentrations in patients with rheumatoid arthritis. Clin Exp Rheumatol, 2000; 18: 713-8. Nilsson LN, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM, Po tter H. Cognitive impairment in PDAPP mice depends on ApoE and ACT-catalyzed amyl oid formation. Neurobiol Aging, 2004; 25: 1153-67. Olszewski WL, Pazdur J, Kubasiewicz E, Zaleska M, Cooke CJ, Miller NE. Lymph draining from foot joints in rheumatoid arth ritis provides insight in to local cytokine and chemokine production and transport to lymph nodes. Arthritis Rh eum, 2001; 44: 541-9. Padmanabhan J, Levy M, Dickson DW, Potter H. Alpha1-antichymotrypsin, an inflammatory protein overexpressed in Alzheimer's disease brain, induces tau phosphorylation in neurons. Brain, 2006; 129: 3020-34. Parsonage G, Filer A, Bik M, Hardie D, Lax S, Howlett K, Church LD, Raza K, Wong SH, Trebilcock E, Scheel-Toellner D, Sa lmon M, Lord JM, Buckley CD. Prolonged, granulocyte-macrophage colony-stimulating factor-dependent, neutrophil survival following rheumatoid synovial fibroblast act ivation by IL-17 and TNFalpha. Arthritis Res Ther, 2008; 10: R47. Potter H, Wefes IM, Nilsson LN. The inflammation-induced pathological chaperones ACT and apo-E are necessary catalysts of Alzheimer amyloid formation. Neurobiol Aging, 2001; 22: 923-30. Rhodin J, Thomas T, Bryant M, Sutton ET. Animal model of Alzheimer-like vascular pathology and inflammatory reaction. Ann N Y Acad Sci, 2000; 903: 345-52. Rogers J, Li R, Mastroeni D, Grover A, Le onard B, Ahern G, Cao P, Kolody H, Vedders L, Kolb WP, Sabbagh M. Peripheral clearance of amyloid beta peptide by complement C3-dependent adherence to erythrocytes. Neurobiol Aging, 2006; 27: 1733-9. Sanchez-Ramos J, Song S, Sava V, Catlow B, Lin X, Mori T, Cao C, Arendash GW. Granulocyte Colony Stimulating Factor (G-C SF) Decreases Brain Amyloid Burdenand Reverses Cognitive Impairment in Alzheimer's Mice. Neuroscience, 2009. Schabitz WR, Kruger C, Pitzer C, Weber D, Laage R, Gassler N, Aronowski J, Mier W, Kirsch F, Dittgen T, Bach A, Sommer C, Sc hneider A. A neuroprotective function for the hematopoietic protein granulocyte-macrophage colony stimulating factor (GM-CSF). J Cereb Blood Flow Metab, 2008; 28: 29-43.

PAGE 202

186 Schellekens GA, Visser H, de Jong BA, va n den Hoogen FH, Hazes JM, Breedveld FC, van Venrooij WJ. The diagnostic properties of rheumatoid arthritis antibodies recognizing a cyclic citru llinated peptide. Arthritis Rheum, 2000; 43: 155-63. Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Levy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D, Younkin S. Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutatio ns linked to familial Alzheimer's disease. Nat Med, 1996; 2: 864-70. Schneider UC, Schilling L, Schroeck H, Nebe CT, Vajkoczy P, Woitzik J. Granulocytemacrophage colony-stimulating factor-induced vessel growth restores cerebral blood supply after bilateral ca rotid artery occlusion. Stroke, 2007; 38: 1320-8. Simard AR, Rivest S. Bone marrow stem ce lls have the ability to populate the entire central nervous system into fully differentiated parenchymal microglia. FASEB J, 2004; 18: 998-1000. Simard AR, Soulet D, Gowing G, Julien JP, Rivest S. Bone marrow-derived microglia play a critical role in restri cting senile plaque formation in Alzheimer's disease. Neuron, 2006; 49: 489-502. Szekanecz Z, Koch AE. Macrophages and their products in rheumatoid arthritis. Curr Opin Rheumatol, 2007; 19: 289-95. Wisniewski T, Castano EM, Golabek A, Vogel T, Frangione B. Acceleration of Alzheimer's fibril formation by apolipoprotei n E in vitro. Am J Pathol, 1994; 145: 10305. Wyss-Coray T, Yan F, Lin AH, Lambris JD, Alexander JJ, Quigg RJ, Masliah E. Prominent neurodegeneration and increased pl aque formation in complement-inhibited Alzheimer's mice. Proc Natl Acad Sci U S A, 2002; 99: 10837-42. Xu WD, Firestein GS, Taetle R, Kaushans ky K, Zvaifler NJ. Cytokines in chronic inflammatory arthritis. II. Granulocytemacrophage colony-stimulating factor in rheumatoid synovial effusions. J Clin Invest, 1989; 83: 876-82.

PAGE 203

10.7. Figures 187

PAGE 204

188 Figure 41. Intrahippocampal injection of GM-CSF (left) and aCSF (right) (a) Representative coronal tissue cryo-sectione d at 14 m and stained with MabTech A /Alexa 546. Image is a montage of about 145 pictures taken at 10X. White spots indicate amyloid plaque immunolabelling (see Supplementary Figure 4 Appendix B for representative montaged sections of all 4 mice). (b) Overall plaque reductions seen in all 4 plaque parameters measured from 5 quantif ied sections per mouse (n = 4 mice). Each analyzed picture per coronal section was th resholded equally to the same standard deviation from the histogram m ean, and analyzed for area, pe rimeter, feret diameter, and integrated density parameters of each plaque. Area and Perimeter data were calculated from the total number of plaque values in each hemisphere per section, and Feret Diameter and Integrated Density were calcula ted from the average values of the plaques measured in each hemisphere per section. Each section quantified contained analysis of 15-25 individual 10X pictures per hemisphere with less pictures quantified in the anterior brain and more in posterior br ain (see Supplementary Figure 6 Appendix B). Error bars are Standard Error of the Mea n. Statistical significance from ipsilateral GMCSF administration versus contralateral aCSF-injection hemispheres was obtained by paired Students t-test with p values (Area: p < 1.11E-07; Perimeter: p < 1.41E-06; Feret Diameter: p < 2.36E-09; Integrated Density: p < 1.11E-07).

PAGE 205

189

PAGE 206

190 Figure 42. Behavioral analysis following daily subcutaneous GM-CSF injections. (a) Standard Radial Arm Water Maze errors Tg control mice (n = 8) show substantial impairment on working memory trials T4 and T5 compared to NT control mice (n = 8) in individual blocks of testi ng (upper), and over all 4 days of testing (lower). GM-CSFtreated Tg mice (n = 7) performed as well as or better than NT control mice on working memory trials T4 and T5 during individua l blocks and over all. GM-CSF-treated NT mice (n = 9) performed similarly to or slightly better than NT controls (Note significantly better performance of NT+GMCSF group versus NT group for T4 of Block 1), although this effect was not significant overall. St atistical significance determined by one-way ANOVA for the 8-days of pre-treatment testin g (four 2-day blocks) and the 4-days of post-treatment testing (two 2-day blocks), w ith evaluation for indi vidual blocks, as well as over all blocks. Thereafter, post hoc pair -by-pair differences between groups were resolved with the Fisher LSD (least significant difference) test. ( **p < 0.05 or higher significance versus all other groups; p < 0.05 or higher sign ificance versus Tg+GMCSF and NT+GM-CSF). (b) Cognitive Interference Task. Overall (4 Days) Tg control mice are impaired compared to NT mice on all four cognitive measures assessed. GMCSF-treated Tg mice exhibited significantly better 3-trial recall (A1-A3) and delayed recall (A5) compared to Tg controls and performed simila rly to NT mice in all four cognitive measures. GM-CSF treatment of NT mice did not result in significantly better performance compared to NT controls, although trends for a benefici al GM-CSF effect in NT mice were evident overall. For statistical analysis, both 2-day bl ocks were analyzed separately, as were all four days coll ectively. One way ANOVAs were employed for each of the four behavioral measures analyze d, followed by post hoc Fishers LSD test to determine significant group differences at p < 0.05. (*Tg significantly different from NT+GM-CSF, **Tg significantly diffe rent from all other groups). (c) Cognitive Interference Task. Proactive Inte rference testing (First 2 days) GM-CSF-treated Tg mice performed significantly better than Tg contro ls and equally to NT and GM-CSF-treated NT mice. Statistical signifi cance determined by one-way ANOVA (**p < 0.05 or higher significance versus a ll other groups).

PAGE 207

191

PAGE 208

192 Figure 43. Amyloid deposition in subcutan eous GM-CSF-injected mice. (a-d) Photomicrographs of coronal 5-m paraffi n-embedded sections immunolabelled with anti-A antibody (clone 4G8 in Entorrhinal cort ex (E) and hippocampus (H)). Since the 4G8 antibody was obtained with biotin label, the secondary step of the ABC protocol was omitted. However, treatment with 70% formic acid prior to the pre-blocking step was necessary. Phosphate-buffered saline (0.1 mM pH 7.4) was used instead of primary antibody or ABC reagent as a negative control. Pictures are repres entative of amyloid load closest to the mean of th e GM-CSFor saline-treated Tg groups. Scale bar = 50 m. (e) Percent of amyloid burden from the average of five 5m sections (150m apart) through both anatomic regions of interest (h ippocampus and entorhinal cortex) per mouse of GM-CSF-treated (n = 5) ve rsus salinetreated (n = 6). A threshold opti cal density was obtained that discriminated staining from background, and each region of interest was manually edited to eliminate artifacts. Data are reported as percentage of immunolabeled area captured (positive pixels) relative to the full area captured (total pixels). Statistical significance was determined by two-tailed homos cedastic Students t-test with a p value of < 0.05 considered significant: Entorhinal cortex (*p < 0.026), and Hippocampus (p = 0.12).

PAGE 209

193

PAGE 210

194 Figure 44. Microglial immunostaining in subcut aneous GM-CSF-injected mice. (a-d) Photomicrographs of coronal 5m para ffin-embedded sections immunolabelled with Iba-1 antibody in Entorrhinal cortex (E ) and hippocampus (H). Normal rabbit serum was used instead of primary antibody or ABC reagent as a negative control. Pictures are representative of Iba-1 immunolabelling closest to the mean of the GM-CSFor saline controltreated groups. Scale bar = 50m. (e) Percent of Iba1 burden from the average of five 5-m sections (150 m apart) through bot h anatomic regions of interest (H and EC) per mouse of GM-CSF-treated (n = 5) versus saline-treated (n = 6). A threshold optical density was obtained that discriminated staining from background, and each region of interest was manually edited to eliminate artifacts. Data are reported as percentage of immunolabeled area captured (pos itive pixels) relative to the full area captured (total pixels). Differences between the two groups were statistically significant by two-tailed homoscedastic Students t-test with a p va lue of <0.05 considered significant: H(p < 0.02), EC(p < 0.05).

PAGE 211

195

PAGE 212

196 Figure 45. Synaptophysin immunostaining in subc utaneous GM-CSF-injected mice. (a-d) Photomicrographs of coronal 5-m para ffin-embedded sections immunolabelled with anti-synaptophysin antibody. Normal ra bbit serum was used instead of primary antibody or ABC reagent as a negative cont rol. Pictures are representative of synaptophysin immunolabelling closest to the mean of the GM-CSFor saline controltreated groups. Scale bar = 50 m. (e) Percent of synaptophysin immunoreactivity from the average of 5 sections per mouse of GM-C SF-treated (n = 5) versus saline controltreated (n = 6). To evaluate synaptophysin im munoreactivity, after the mode of all images was converted to gray scale, the average intensity of positive signals from each image was quantified in the CA1 and CA3 regions of hippocampus as a relative number from zero (white) to 255 (black). Albeit numerically small, differences between the two groups were statistically significant by two-tailed hom oscedastic Students t-test with a p value of < 0.05 considered significant: CA1(p < 0.0013), CA3(p < 0.0023).

PAGE 213

197 CHAPTER 11 DISCUSSION Neurodegeneration describes the selective and progressive loss of structure and function in neurons. Although the initiating factors, symtoms, and pathologies of neurodegenerative diseases such as Alzheime rs disease, Parkinsons, Huntingtons, and Amyotrophic Lateral Sclerosis may be differe nt, they share common pathophysiologies. Primary hallmarks of these pathologies are atypical accumulations of protein deposits, often due to misfolding. Often associat ed with these deposits are dysfunctional mitochondria, oxidative stress disrupted axonal transport, inflammation, and apoptotic cell death (DeLegge and Smoke, 2008; Gebick e-Haerter, 2001; Gendron and Petrucelli, 2009; Swerdlow, 2009). If these disruptions oc cur in motor neurons, as in ALS, motor function is impaired and if the deposits are f ound in cortical neurons, as in Alzheimers disease, the outcome is dementia (Soto, 2003; Weisman et al., 2006). Why these neurons are selectively vulnerable in each disease is unknown and, unfortunately for many people, preventions and therapies remain elusive. In the work described here, mouse models of neurodegenerative disease were used to investigate promoter s and inhibitor of neurodegeneration in the proteinopathies ALS and AD. Mouse models are a common and useful re search tool for investigating disease pathology and developing potential therapies prior to clinical trials in humans. In comparison to other animals, their short lifespan averaging 2.5 years, high reproductive rate, and relative ease of modifi cation for the incorporation and expression of transgenes make mice ideal for studying human diseases. Additionally, controlled and standardized environments keep behavioral testing rath er objective and every mouse is completely nave to each experiment. Therefore, respons es to novel environments are controlled and readily quantifiable. Another benefit of mice, though a potentially confounding variable, is their hardy immune system. They are very capable of resisting iatrogenic infection

PAGE 214

198 during intracranial and intrathecal pene tration as well as during subcutaneous implantation of osmotic pumps in conditions where complete sterility is simply not possible. There are also some clear disadvantages and challenges to using transgenic mouse models. In both AD and ALS, only 5-10% of the disease is caused directly by gene mutations (Gros-Louis et al., 2006; Hardy, 2006). The efficacy of genetic approaches to the study of neurodegenerative diseases is also limited by the fact that many such disorders have a wide display of phenotype s and varied etiolo gy. In addition, the interpretation of results obtaine d in animal models has to in tegrate the fact that often nonphysiological levels of selected proteins such as A or SOD1 are used to recapitulate a disease in a mouse within a reasonable experi mental time frame. Moreover, because it is not yet known if the proteinaceous aggregates in these diseases are causes, byproducts, or neuroprotective factors, one must be mindful of the exact query to the system and of the interpretation of the results In addition, while it is reasonable to try possible interventions on already esta blished mouse models instead of engaging in the time consuming and tedious process of producing al ternative mouse models, approaches with suboptimal mouse models can sometimes fail to produce conclusive answers. It may be in this context that several of the experime nts conducted here did not result in favorable outcomes. For example, many animals in my experiments were designed to make amyloid plaques, tau tangles, and SOD1 incl usions. Naturally studies that focused on exacerbating or ameliorating these phenotypes were more successful than testing compounds involved in a different, though important, facet of the disease. The focus of my work presented he re has been on mouse models of neurodegenerative diseases, taking the good w ith the bad. I successfully studied the permeability of the blood brain barrier and the spontaneous formation of autorosettes in the SOD1 G93A mouse. Other positive result s were obtained from practical, engineering approaches to working with the mice. In trathecal delivery of compounds in mice is arduous, and historically expe riments that required deliver y of compounds into the CSF would enter through the lateral ventricles. Over some time, I have developed a tool and technique to effectively and re producibly deliver compounds dire ctly to the site of early

PAGE 215

199 ALS pathology in the lumbar region of the spin al cord. After years of struggling with commercially available intracranial catheter s and experiencing high mortality rates, infections, physical damage to hardware or the animal, and pain/discomfort, enough incentive to devise a better means presented itself. Not only can the patented intracranial catheter assembly obviate concer n for these pitfalls, but it solved another major issue with the AD mice: plaque variability within th e colony. If one hemisphere could serve as the others control, differen ces in amyloid load between genetically identical animals could be compensated. In another novel a pplication, though not a novel concept, I used in vivo electroporation, into the CNS, a procedure that had not been published for such use at the time I started out. Although it may app ear rather crude, electroporation is an excellent tool for proof of principle gene delivery prio r to designing attenuated viral vectors (Heller and Heller, 2006). Ultimately, scientific success was generall y achieved by directly focusing queries on what the mouse models were designed for (inclusions, plaques, tangles) or characterization of these pathologies in mice. Human tau mice were used to investigate phosphorylation and tangle formation, and PS/APP mice were used to study the effects of ACT and plaque build up. Most noteworthy ar e the experiments that reduce plaque load in the animals, which was achieved with 2 completely different approaches (A 1-11 and GM-CSF). It is still unknown whether plaq ues and tangles are the cause of AD or a bystander effect, but they are nonrtheless the focal point of most AD research. The first set of experiments focused on a well-studied and relatively common aspect of neurodegenerative di sease: transmissibility. In 1994, Stanley Pruisner received the Nobel Prize for ground breaking work w ith prion protein. Bovine spongiform encephalopathy (BSE), scrapie of sheep, and Creutzfeldt-Jakob dis ease (CJD) of humans are the most commonly known prion diseases (Pru siner, 1998). Prions are proteins that can acquire a misfolded conformation that nucleates the misfolding of other normally folded prion proteins and form beta sheet amyloid structures (Wic kner et al., 2008). It has since been discovered that amyloid pla ques from post-mortem human or mouse brain, once suspended and injected into the brai n of a young PDAPP tran sgenic mouse, can trigger the early seeding of plaques (Walker et al., 2002). More recently, it has been

PAGE 216

200 shown that a misfolded tau mutant could seed tangle formation when injected into an htau mouse (Clavaguera et al., 2009). We hypothesi zed that such an outcome could also be triggered by SOD1 aggregates to initiate ALS. After homogenizing spinal cords from end-stage SOD1 G93A animals, the homogenate s were infused into the lumbar cistern of younger transgenics to see if ALS pathology coul d be seeded or accelerated. Prior to our experiments, there were conflicting reports from breeding experiments with mutant and wild type SOD1 mice stating that the mutant SOD1 may exert a conf ormational change in the WT and accelerate the disease (Wang et al., 2009), whereas other groups found no difference (Bruijn et al., 2004b). When SOD1 aggregates were infused in our experiments, we could not detect any changes in the behavior or lifes pan in the animals. Our negative result could be explained as fo llows: 1) Aggregates were infused into the CSF and not directly into the tissue out of concern about causing damage to the animal. 2) There was already a ceiling effect with ~18 copies of mutant SOD1 being expressed, and the infusions could not enhance this eff ect any further. Moreover, the human SOD1 gene is flanked by Alu repeats (retrotrans posons) (Maguire et al ., 2006) and colonies would eventually begin to lose copy number as a result. As mentioned in Chapter 2, Mark Gurney, the investigator who disc overed the G93A mutation, disclosed in a personal communication that his own lab has not use this SOD1 transgenic mouse model for quite some time. The S OD1 G93A mouse was designed to form inclusions and develop an aggressive paralytic phenotype af ter only 100 days of age and persists for 30 days until the animals had to be euthani zed. Although this therapeutic window seems appealing to researchers for successive high throughput drug studies, the most effective therapies have been siRNA (Rizvanov et al., 20 09) and antisense oligonucleotides (Smith et al., 2006). In addition, these approaches would only be useful in reducing the amount of mutant protein that is be ing heavily overproduced. Acco rdingly, these therapies might only translate to individuals w ith that particular SOD1 mutation, while there are in fact over 150 ALS-causing SOD1 mutations known (Turner and Talbot, 2008). While inflammation and excitotoxicity are very clearly implicated in ALS pathogenesis, we came to conclude that SOD1 aggregation ha s the most dominant effect. Therefore DTG and KDI, though potentially useful compounds, ha d only marginal if any effect on this

PAGE 217

201 mouse model. However, given the potential of both of these compounds, further research in other models (G85R, G37R), even prophyl actically, may be warranted but was not pursued by our lab. As previously stated, some of our obse rvations in the SOD1 G93A mouse were noteworthy. Inflammation is a ha llmark of all neurodegenerati ve disorders and is evident in this model from activation of microglia and astrocytes (DeL egge and Smoke, 2008; Gebicke-Haerter, 2001). Recently it was also suggested that the adaptive immune system plays a role in ALS (Banerje e et al., 2008). In chapter 3, we show that the blood brain barrier is compromised in the G93A mouse. Previous work showed that T cells (Donnenfeld et al., 1984b; Engelhardt and Appel, 1990; Engelhardt et al., 2005a; Mohamed et al., 2002) and IgG (Alexianu et al., 2001; Engelhardt et al., 1993b) were actually present in high amount s in the CSF of ALS patient s and also in these mice (Garbuzova-Davis et al., 2007a). The increa sed permeability of the BBB in later stages of the disease would provide a logical expl anation for this phe nomenon. Furthermore, we have identified the formation of autorose ttes in this model (appendix A), which has been shown to be a T cell marker. Taken together, our observations show that ALS may involve activation of both the innate and adap tive immune systems. Unfortunately, the model was of little use for the st udy of potential therapeutics. It was at this time that I shifted my efforts from studies on ALS to investigations related to Alzheimers disease. Initially I transferred the idea of testing KDI, the tripepetide of -laminin, as a potential therapeutic for neuodegenerative diseases to the treatment of cognitive impairment. We inves tigated the effects of KDI on plaque load in treated PS/APP and Tg2576 mice, no changes were observed. Nonetheless, the reported potent in vitro effects of KDI as a neurit e outgrowth factor and NMDA/AMPA receptor antagonist (Moykkynen et al., 2005) (similar to Na menda) left us optimistic that it could provide some benefit to the mice. Because the first cohort of KDI treated animals, as shown in chapter 7, showed marked restoration in cognitive abilities as tested with the RAWM, the effects of KDI on LTP and synapt ogenesis were investigated. While the brains from the first cohort were analyze d, the second cohort of animals was starting on their 28 day infusions. Using immunohistochemical analyses, synaptic markers including

PAGE 218

202 synaptophysin, synaptotagmin, synaptopodi n, synaptobrevin, GAP-43, and PSD-95 antibodies revealed no changes between treated and untreated mice; the possibility of delivering bilaterally was not yet available at this time. In another hypothesis we speculated that we were interfering with the glutamate neurotransmission in such a manner as to stimulate the phosphorylation, t hus activation, of dormant AMPA receptors leading to the possibility that KDI was inducing LTP. GluRI, pGluRI, CamKII, and pCamKII levels were compared between KDI treated and untreated mice, but no differences were observed. Moreover, additi onal behavior tests that were performed twice with new animal cohorts, but did not show significant diffe rences between KDItreated and untreated animals either. We interpreted this outcome as follows: 1) If the data published on KDI and its ability to func tion as a potent non-competitive ionotropic glutamate receptor antagonist is accurate, then our results just show little relevance of glutamate dysregulation in AD, i.e KDI does not yield a therapeutic effect in AD, which, at the same time, helps to explain Namenda s modest benefit. 2) The chosen mouse model was possibly not the best system to inve stigate the relevance of excitotoxicity in AD. 3) KDIs effects had, for the most part, been limited to in vitro studies and cannot be translated to the mice. When studying the htau mice, we focuse d directly on phosphorylation events that lead to tangle formation. Tau has ~45 AD-related phosphorylation sites, and numerous kinases and phosphotases are responsible for regulating the phosphorylation status and thus the conformation of tau. Previous in vitro work from our laboratory showed that in most tauopathies, including CBD, PSP, FTDP-17, and AD, antichymotrypsin is upregulated and subsequently found in a ggregates in post mortem brain tissues (Padmanabhan et al., 2006). This observation was of particular interest to us, because ACT is well-established to have a role in A fibrillization in vitro (Ma et al., 1996) and amyloid plaque formation in transgenic mice (Nilsson et al., 2001). ACT, like ApoE, has been termed a pathological chaperone (Wisniewski and Frangione, 1992) in the development of plaques in AD. In chapte r 8, I demonstrated how chronic intracranial infusion of ACT into an htau mouse increa sed levels of pSer 202, PHF-1, and pThr231 in tau and that the intracellular kinase, GSK3, wa s activated. GSK3 has been implicated in

PAGE 219

203 the phosphorylation of several sites on tau, including pThr231 and pSer394/404 (PHF-1) (Hooper et al., 2008), and my findings are c onsistent with these observations. One notable difference between my study and prev ious in vitro work was that pERK was found in much lower amounts (dea ctivated) in after infusion of ACT. Thus in cultured mouse cortical neurons, adding ACT to the me dia increased levels of pERK in a dose dependent fashion (Padmanabhan et al., 2006) whereas in vivo, the opposite effect seemed to occur. These results are from a pilot study of 5 mice, only 4 of which clearly received ACT. This work will be follow ed up as a full study with more animals in anticipation of revealing othe r phosphorylation sites as well as putative kinases. Additionally, it is not yet know n if it is the physical pres ence or protease inhibiting activity of ACT that induces these events and whether they occur from ACT interacting with elements on the cell surface of neurons or after endocytosis. These possibilities would need to be addressed in order to determine how to be st inhibit ACT-mediated tau phosphorylation and tangle formation. ACT is of particular in terest because it provides a link between the hallmark pathologies of AD. It is directly involved in the fibrillization of A and the formation of plaques and has been shown to induce tau phosphrylation by GSK-3 at sites known to be involved in Alzheimers disease (Padmanabha n et al., 2006). Therefore, inhibition or downregulation of ACT, which is highly upr egulated during, AD may be of therapeutic interest. Early observations from our lab suggest that A has strong homology in the Nterminus with both cathepsin G and chymotryps in, both of which are serine proteases and known serine protease targets of ACT (Abraham and Potter, 1989). It is therefore conceivable that ACT is upregulated to bind A in order to sequest er it from forming plaques or oligomers, but that ACT become s outcompeted by the amount of substrates and overwhelmed in the process. Historically, ridding the brain of pl aques and tangles has been the major therapeutic goal, either thr ough stimulation or inhibition of the innate and adaptive immune systems (Maccioni et al., 2009) or mo re direct mechanisms. The involvement of ACT and ApoE in the formation of amyloi d plaques and concomitant activation of microglia and astrocytes and release of cytokines and chemokines s upport the theory that,

PAGE 220

204 as in all neurodegenerative diseases, inflamma tion is a major aspect of AD pathogenesis. Whether this is a primary or secondary event to the formation of oligomers and plaques is uncertain, as well as whether or not it is bene ficial or detrimental to have triggered an immunological response. Microglia, for exam ple, are known to release proinflammatory cytokines (Hanisch, 2002) and also to engulf amyloid (Herbe r et al., 2007). In chapter 9, I describe the use of a decoy A peptide (1-11) to reduce the amount of amyloid formed. The rationale for this approach came from previous in vitro work from our laboratory in which the amount of A fibrils was enhanced in the presence of ApoE and ACT, but that such catalyzed fibrilization could be attenuated by adding A fragments 12-28 and 2-9, respectively. These peptides include the binding regions for ApoE and ACT on full length A and they, might block these proteins from binding A and rendering the reaction mi xture less toxic to cultured neurons, especially with increasing amounts of the decoy peptides (M a et al., 1996). In 2006, Wisniewski et al. published a study in which they had taken the 12-28 peptide and, with some modification, injected the molecule systemically into Tg2576 mice (with and without PS1 M146L). This treatment resulted in a significant reduc tion in plaque load and improved cognition in behavioral tasks (Sadowski et al., 2006). The same rationale was applied in our attempt to interfer with the interaction of ACT and A when we introduced the peptide fragment 1-11 (containing the 2-9 binding site of ACT on A ) in PS/APP mice as they were developing plaques. The peptide was in fused intrahippocampally into a small pilot of PS/APP mice, unilaterally, with aCSF to the contralateral hemisphere for direct comparison of plaque load. As shown in th e figures of chapter 10, the peptide not only significantly reduced new plaque formation, but appears to have cleared existing plaques as shown with both the 6E10 antibody (binds A 1-16) and 4G8 antibody (binds A 1724). Most likely, the peptide binds to the A CT equivalent in the mouse or serpin A3N, which is identified in the plaques of these mice with anti-ACT antibodies. The mechanism for the clearance is unknown, but Ib a1 microglial staining is virtually absent from the treated hemisphere, indicating that these cells are not present or that the respective epitope is not accessible. Further investigation is needed to determine the putative mechanisms of clearance as well as to infuse the peptide bilaterally for

PAGE 221

205 behavioral analyses. Ultimately, modificat ion of the peptide to reduce immunogenicity and increase its half life, could allow syst emic dosing and yield a potential AD therapy. In another part of my dissertation research, I investig ated the negative correlation between patients with Rheumatoid arthritis and AD. We hypothesized that there were factors upregulated in the autoimmune disease that c onferred neuroprotection and reduced if not eliminated the risk of de veloping AD. Although up-regulated leukocytes in RA could potentially enter into the brai n and inhibit development of AD pathology and/or neuronal dysfunction, lym phocytic infiltrates into AD pa tient brains have not been reported. The lack of infiltrati on suggests that activation of the innate immune system might be responsible for preven ting AD pathology in RA patients Originally, 3 cytokines were identified as upr egulated in RA and of potenti al interest: G-CSF, M-CSF, and GM-CSF.. These CSFs enhance the surviv al of their respectiv e leukocytes and drive their proliferation and diffe rentiation from monocytic precursors. M-CSF and G-CSF induce specific subsets of the innate imm une system, while GM-CSF induces the full range of innate cells. Bolus injections of 5 g per hemisphere showed significant plaque reductions after only 1 week (see chapter 11). This success led us to begin a subcutaneous dosing regime and performed behavioral analyses. The improved cognitive function and reduced cortical amyloidosis of GM-CSF-treat ed Tg mice was paralleled by increased synaptophysin immunoreactivity in both CA1 and CA3, indicating increased synaptic density in these hippocampal areas. Further i nvestigation into GM-C SF to elucidate its mechanism in AD is in progress as well as the initiation of clinical trials with Leukine an FDA approved drug of recombinant GM-CSF currently in use for recovery from chemotherapy. It is a remarkable and unfortunate fact that despite the enourmous knowledge that has been acquired about ALS and AD, effective cures or even treatments for both diseases have been elusive. Much debate exists on whether protei n aggregation is an initiating factor in neurodegeneration or a si de effect. Similar de bate questions whether activation of the immune system is benefi cial or detrimental to CNS function in neurodegenerative diseases. Our research suggests the former. In AD, some reports state that cognition is closely linked to plaque load in the mice (Gordon et al., 2001),

PAGE 222

206 suggesting that, even without neuron loss, the plaques cause behavioral deficits. Conversely, other studies show no change in plaque load with improved cognition (Costa et al., 2007). Nonetheless, th e studies discussed here focused primarily on the use of transgenic mouse models of protein aggregation and on proc edures that could exacerbate or inhibit the effect. The specific focus of my work was on the development of novel means of drug and gene delivery into transgenic mice. The fact that certain compounds did not affect protein aggregation still leav es the possibility that they might have therapeutic potential if tested in other systems. However, unquestionably, the delivery systems developed along the way, have imp lications that reach beyond AD and ALS as they can each be readily adapted for investig ations with virtually all mouse models of neurodegenerative disease.

PAGE 223

207 REFERENCES Abraham CR. Reactive astrocytes and alpha1-antichymotrypsin in Alzheimer's disease. Neurobiol Aging, 2001; 22: 931-6. Abraham CR, Potter H. Alpha 1-antichymotrypsin in brain aging and disease. Prog Clin Biol Res, 1989; 317: 1037-48. Abraham CR, Selkoe DJ, Potter H. Immunochemi cal identification of the serine protease inhibitor alpha 1-antichymotrypsin in the brai n amyloid deposits of Alzheimer's disease. Cell, 1988; 52: 487-501. Adhihetty PJ, Beal MF. Creatine and its potenti al therapeutic value for targeting cellular energy impairment in neurodegenerative dis eases. Neuromolecular Med, 2008; 10: 27590. Ajmo CT, Jr., Vernon DO, Collier L, Pennypacker KR, Cuevas J. Sigma receptor activation reduces infarct si ze at 24 hours after permanent middle cerebral artery occlusion in rats. Curr Ne urovasc Res, 2006; 3: 89-98. Alexianu ME. The role of immune pro cesses in amyotrophic lateral sclerosis pathogenesis. Rom J Neurol Psychiatry, 1995; 33: 215-27. Alexianu ME, Kozovska M, Appel SH. Immune reactivity in a mouse model of familial ALS correlates with disease progr ession. Neurology, 2001; 57: 1282-9. Almer G, Guegan C, Teismann P, Naini A, Rosoklija G, Hays AP, Chen C, Przedborski S. Increased expression of the pro-in flammatory enzyme cyclooxygenase-2 in amyotrophic lateral sclerosis. Ann Neurol, 2001; 49: 176-85. Anand P, Parrett A, Martin J, Zeman S, Foley P, Swash M, Leigh PN, Cedarbaum JM, Lindsay RM, Williams-Chestnut RE, et al. Re gional changes of ciliary neurotrophic factor and nerve growth factor levels in post mortem spinal cord and cerebral cortex from patients with motor disease. Nat Med, 1995; 1: 168-72. Andersen PM. Amyotrophic lateral sclerosis associated with muta tions in the CuZn superoxide dismutase gene. Curr. Ne urol. Neurosci. Rep., 2006; 6: 37-46.

PAGE 224

208 Andorfer C, Kress Y, Espinoza M, de Silva R, Tucker KL, Barde YA, Duff K, Davies P. Hyperphosphorylation and aggregation of tau in mice expressing normal human tau isoforms. J Neurochem, 2003; 86: 582-90. Andreadis A. Tau gene alternative splic ing: expression patt erns, regulation and modulation of function in normal brain a nd neurodegenerative diseases. Biochim Biophys Acta, 2005; 1739: 91-103. Annunziata P, Volpi N. High levels of C3c in the cerebrospinal fluid from amyotrophic lateral sclerosis pati ents. Acta Neurol S cand, 1985a; 72: 61-4. Annunziata P, Volpi N. High levels of C3c in the cerebrospinal fluid from amyotrophic lateral sclerosis pati ents. Acta Neurol. S cand., 1985b; 72: 61-4. Apostolski S, Nicolic J, Bugarski-Prokopljevi c C, Miletic V, Pavlovic S, Filipovic S. Serum and CSF immunological findings in ALS. Acta Neurol. Scand., 1991; 83: 96-8. Appel SH, Smith RG, Engelhardt JI, Stef ani E. Evidence for autoimmunity in amyotrophic lateral sclerosis. J Neurol Sci, 1994; 124 Suppl: 14-9. Arai T, Hasegawa M, Akiyama H, Ikeda K, Nonaka T, Mori H, Mann D, Tsuchiya K, Yoshida M, Hashizume Y, Oda T. TDP-43 is a component of ubiquitin-positive taunegative inclusions in frontotemporal l obar degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res Commun, 2006; 351: 602-11. Arendash GW, Jensen MT, Salem N, Jr., Hussein N, Cracchiolo J, Dickson A, Leighty R, Potter H. A diet high in omega-3 fatty acids does not improve or protect cognitive performance in Alzheimer's transgenic mice. Neuroscience, 2007; 149: 286-302. Arendash GW, King DL, Gordon MN, Morgan D, Hatcher JM, Hope CE, Diamond DM. Progressive, age-related behavi oral impairments in transgenic mice carrying both mutant amyloid precursor protein and presenilin-1 transgenes. Brain Res, 2001; 891: 42-53. Arendash GW, Schleif W, Rezai-Zadeh K, Jackson EK, Zacharia LC, Cracchiolo JR, Shippy D, Tan J. Caffeine protects Alzheimer's mice against cognitive impairment and reduces brain beta-amyloid produc tion. Neuroscience, 2006; 142: 941-52. Aschner A. Astrocytes as mediators of imm une and inflammatory responses in the CNS. NeuroToxicol., 1998; 19: 269-82. Avata C, Ropper AH. Ischaemic brain oede ma. J. Clin. Neurosci., 2002; 9: 113-24. Ayala YM, Misteli T, Baralle FE. TDP-43 regulates retinoblastoma protein phosphorylation through the repression of cyc lin-dependent kinase 6 expression. Proc Natl Acad Sci U S A, 2008a; 105: 3785-9.

PAGE 225

209 Ayala YM, Zago P, D'Ambrogio A, Xu YF, Petru celli L, Buratti E, Baralle FE. Structural determinants of the cellular localization a nd shuttling of TDP-43. J Cell Sci, 2008b; 121: 3778-85. Bales KR, Verina T, Cummins DJ, Du Y, D odel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghetti B, Paul SM. Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mous e model of Alzheimer's disease. Proc Natl Acad Sci U S A, 1999; 96: 15233-8. Ballabh P, Braun A, Nedergaard M. The bl ood-brain barrier: an overview structure, regulation, and clinical implications Neurobiol. Dis ., 2004; 16: 1-13. Banerjee R, Mosley RL, Reynolds AD, Dh ar A, Jackson-Lewis V, Gordon PH, Przedborski S, Gendelman HE. Adaptive immune neuroprotection in G93A-SOD1 amyotrophic lateral sclerosis mi ce. PLoS ONE, 2008; 3: e2740. Barnoud P, Lolivier J, Sanger DJ, Scatton B, Moser P. Quantitative motor assessment in FALS mice: a longitudinal study. NeuroReport, 1997; 8: 2861-5. Beal MF. Excitotoxicity and nitric oxide in Parkinson's disease pathogenesis. Ann Neurol, 1998; 44: S110-4. Beckman JS, Carson M, Smith CD, Koppenol WH. ALS, SOD and peroxynitrite. Nature, 1993; 364: 584. Beers DR, Ho BK, Siklos L, Alexianu ME Mosier DR, Mohamed AH, Otsuka Y, Kozovska ME, McAlhany RE, Smith RG, Appe l SH. Parvalbumin overexpression alters immune-mediated increases in intracellula r calcium, and delays disease onset in a transgenic model of familial amyotrophic late ral sclerosis. J. Neurochem., 2001; 79: 499509. Bell RD, Zlokovic BV. Neurovascular mechanisms and blood-brain ba rrier disorder in Alzheimer's disease. Acta Neuropathol, 2009. Belli S, Vanacore N. Proportionate mortality of Italian soccer pl ayers: is amyotrophic lateral sclerosis an occu pational disease? Eur J Epidemiol, 2005; 20: 237-42. Bennett SP, Boyd T.D., Norden M., Padmana bhan J., Neame P., Wefes I., Potter H. A Novel Technique for Simultane ous Bilateral Brain Infusi ons in a Mouse Model of Neurodegenerative Disease Journa l of Neuroscience Methods, 2009. Bezzi P, Domercq M, Brambilla L, Galli R, Schols D, De Clercq E, Vescovi A, Bagetta G, Kollias G, Meldolesi J, Volterra A. CXCR4-Activated astrocyte glutamate release via TNF alpha: amplification by micr oglia triggers neurtoxicit y. Nature Neurosci, 2001; 4: 702-10.

PAGE 226

210 Boillee S, Vande Velde C, Cleveland DW. AL S: a disease of motor neurons and their nonneuronal neighbors. Ne uron, 2006a; 52: 39-59. Boillee S, Yamanaka K, Lobsiger CS, Copela nd NG, Jenkins NA, Kassiotis G, Kollias G, Cleveland DW. Onset and progression in i nherited ALS determined by motor neurons and microglia. Science, 2006b; 312: 1389-92. Boissonneault V, Filali M, Lessa rd M, Relton J, Wong G, Rive st S. Powerful beneficial effects of macrophage colony-stimulating factor on beta-amyloid deposition and cognitive impairment in Alzheime r's disease. Brain, 2009; 132: 1078-92. Bolmont T, Clavaguera F, Meyer-Luehmann M, Herzig MC, Radde R, Staufenbiel M, Lewis J, Hutton M, Tolnay M, Jucker M. Induction of tau pat hology by intracerebral infusion of amyloid-beta -cont aining brain extract and by am yloid-beta deposition in APP x Tau transgenic mice. Am J Pathol, 2007; 171: 2012-20. Borchelt DR, Lee MK, Slunt HS, Guarnieri M, Xu ZS, Wong PC, Brown RH, Jr., Price DL, Sisodia SS, Cleveland DW. Superoxide dismutase 1 with mutations linked to familial amyotrophic lateral scle rosis possesses significant activity. Proc Natl Acad Sci U S A, 1994; 91: 8292-6. Bourrie B, Bouaboula M, Benoit JM, Derocq JM, Esclangon M, Le Fur G, Casellas P. Enhancement of endotoxin-induced interl eukin-10 production by SR 31747A, a sigma ligand. Eur J Immunol, 1995; 25: 2882-7. Bourrie B, Bribes E, Derocq JM, Vidal H, Casellas P. Sigma receptor ligands: applications in inflammation and oncology. Curr Opin Investig Drugs, 2004; 5: 1158-63. Bowling AC, Barkowski EE, McKenna-Yasek D, Sapp P, Horvitz HR, Beal MF, Brown RH, Jr. Superoxide dismutase concentration and activity in familial amyotrophic lateral sclerosis. J Neurochem, 1995; 64: 2366-9. Bruijn L, Miller TM, Cleveland DW. Unraveling the mechanisms involved in motor neuron degeneration in ALS. A nnu. Rev. Neurosci., 2004a; 27: 723-49. Bruijn LI, Becher MW, Lee MK, Anderson KL, Jenkins NA, Copeland NG, Sisodia SS, Rothstein JD, Borchelt DR, Price DL, Clev eland DW. ALS-linked SOD1 mutant G85R mediates damage to astrocytes and promot es rapidly progressive disease with SOD1containing inclusions. Neuron, 1997; 18: 327-38. Bruijn LI, Houseweart MK, Kato S, Anderson KL, Anderson SD, Ohama E, Reaume AG, Scott RW, Cleveland DW. Aggregation and motor neuron toxicity of an ALS-linked SOD1 mutant independent from wild-t ype SOD1. Science, 1998; 281: 1851-4.

PAGE 227

211 Bruijn LI, Miller TM, Cleveland DW. Unra veling the mechanisms involved in motor neuron degeneration in ALS. A nnu Rev Neurosci, 2004b; 27: 723-49. Butterfield DA, Abdul HM, Opii W, Newman SF, Joshi G, Ansari MA, Sultana R. Pin1 in Alzheimer's disease. J Neurochem, 2006; 98: 1697-706. Carpenter S. Proximal axonal enlargement in motor neuron disease. Neurology, 1968; 18: 841-51. Castelletto V, Hamley IW, Harris PJ. Self-asse mbly in aqueous solution of a modified amyloid beta peptide fragment. Biophys Chem, 2008; 138: 29-35. Cerf E, Sarroukh R, Tamamizu-Kato S, Breydo L, Derclaye S, Dufrene Y, Narayanaswami V, Goormaghtigh E, Ruyssch aert JM, Raussens V. Anti-parallel betasheet a signature structure of the oligomeric amyloid-beta peptide. Biochem J, 2009. Chen N, Song, S., Hudson, J.E., Willing, A.E., Garbuzova-Davis, S., Sanches-Ramos, J.,Sanberg, P.R., Zigova, T. Neural and he matopoietic antigens in cultured human umbilical cord blood cells. Exp Neurol, 2003; 181: 89. Chiu AY, Zhai P, Dal Canto MC, Peters TM, Kwon YW, Prattis SM, Gurney ME. Agedependent penetrance of disease in a tran sgenic mouse model of familial amyotrophic lateral sclerosis. Mol Ce ll Neurosci, 1995; 6: 349-62. Chou SM, Fakadej AV. Ultrastructure of chro matolytic motoneurons and anterior spinal roots in a case of Werdnig-Hoffmann disease. J Neuropathol E xp Neurol, 1971; 30: 36879. Claudio L. Ultrastructural f eatures of the blood-brain ba rrier in biopsy tissue from Alzheimer's disease patients. Acta Neuropathol. (Berl.), 1996; 91: 6-14. Clavaguera F, Bolmont T, Crowther RA, Abra mowski D, Frank S, Probst A, Fraser G, Stalder AK, Beibel M, Staufenbiel M, Juck er M, Goedert M, Tolnay M. Transmission and spreading of tauopathy in transg enic mouse brain. Nat Cell Biol, 2009. Clement AM, Nguyen MD, Roberts EA, Garcia ML, Boillee S, Rule M, McMahon AP, Doucette W, Siwek D, Ferrante RJ, Brown RH Jr., Julien JP, Goldstein LS, Cleveland DW. Wild-type nonneuronal cells extend survival of SOD1 mutant motor neurons in ALS mice. Science, 2003; 302: 113-7. Cleveland DW, Rothstein JD. From Charcot to Lou Gehrig: deciphering selective motor neuron death in ALS. Nat Rev Neurosci, 2001; 2: 806-19.

PAGE 228

212 Costa DA, Cracchiolo JR, Bachstetter AD, Hughes TF, Bales KR, Paul SM, Mervis RF, Arendash GW, Potter H. Enrichment improve s cognition in AD mice by amyloid-related and unrelated mechanisms. Ne urobiol Aging, 2007; 28: 831-44. Cox CA, Shi G, Yin H, Vistica BP, Wawrousek EF, Chan CC, Gery I. Both Th1 and Th17 are immunopathogenic but differ in other key biological act ivities. J Immunol, 2008; 180: 7414-22. Crawford F, Hardy J, Mullan M, Goate A, Hughes D, Fidani L, Roques P, Rossor M, Chartier-Harlin MC. Sequencing of exons 16 and 17 of the beta-amyloid precursor protein gene in 14 families with early onset Alzheimer's disease fails to reveal mutations in the beta-amyloid sequence. Neurosci Lett, 1991; 133: 1-2. Dal Canto MC, Gurney ME. Development of central nervous system pathology in a murine transgenic model of human amyotrophi c lateral sclerosis. Am J Pathol, 1994; 145: 1271-9. Dalcanto MC, Gurney ME. NEUROPAT HOLOGICAL CHANGES IN 2 LINES OF MICE CARRYING A TRANSGENE FOR MU TANT HUMAN CU,ZN SOD, AND IN MICE OVEREXPRESSING WILD-TYPE HUMAN SOD A MODEL OF FAMILIAL AMYOTROPHIC-LATERAL-SCLEROSIS (F ALS). Brain Res., 1995; 676: 25-40. Das S, Potter H. Expression of the Alzheimer amyloid-promoting factor antichymotrypsin is induced in human as trocytes by IL-1. Neuron, 1995; 14: 447-56. DeLegge MH, Smoke A. Neurodegeneration an d inflammation. Nutr Clin Pract, 2008; 23: 35-41. Delisle MB, Carpenter S. Neurofibrillary axonal swellings and amyotrophic lateral sclerosis. J Neurol Sci, 1984; 63: 241-50. Dermietzel R, Krause D. Molecular anatom y of the blood-brain barrier as defined by immunocytochemistry. Int. Re v. Cytol., 1991; 127: 57-109. Dewachter I, Moechars D, van Dorpe J, Tesse ur I, Van den Haute C, Spittaels K, Van Leuven F. Modelling Alzheimer's disease in multiple transgenic mice. Biochem Soc Symp, 2001: 203-10. Donnenfeld H, Kascsak RJ, Bartfeld H. Deposits of IgG and C3 in the spinal cord and motor cortex of ALS patients. J Neuroimmunol, 1984a; 6: 51-7. Donnenfeld H, Kascsak RJ, Bartfeld H. Deposits of IgG and C3 in the spinal cord and motor cortex of ALS patients. J. Neuroimmunol., 1984b; 6: 51-7.

PAGE 229

213 Dwyer D, Vannucci SJ, Simpson IA. Expre ssion, regulation, and functional role of glucose transporters (GLUTs) in br ain. Int Rev Neurobiol, 2002; 51: 159-88. Echeverria V, Burgess S, Gamble-George J, Zeitlin R, Lin X, Cao C, Arendash GW. Sorafenib inhibits nuclear factor kappa B, d ecreases inducible nitric oxide synthase and cyclooxygenase-2 expression, and restores working memory in APPswe mice. Neuroscience, 2009; 162: 1220-31. El Khoury J, Toft M, Hickman SE, Means TK Terada K, Geula C, Luster AD. Ccr2 deficiency impairs microglial accumulation and accelerates progression of Alzheimerlike disease. Nat Med, 2007; 13: 432-8. Elliott JL. Cytokine upregulation in a murine model of familial amyotrophic lateral sclerosis. Brain Res Mol Brain Res, 2001; 95: 172-8. Elliott JL, Snider WD. Motor neuron growth factors. Neurology, 1996; 47: S47-53. Engelhardt B. Regulation of immune cell entr y into the central nervous system. Results Probl. Cell Differ., 2006; 43: 259-80. Engelhardt JI, Appel SH. IgG reactivity in the spinal cord and motor cortex in amyotrophic lateral sclerosis. Arch. Neurol., 1990; 47: 1210-6. Engelhardt JI, Soos J, Obal I, Vigh L, Siklos L. Subcellular localizat ion of IgG from sera of ALS patients in the nervous system Acta Neurol Scand, 2005a; 112: 126-33. Engelhardt JI, Tajti J, Appel SH. Lymphocy tic infiltrates in the spinal cord in amyotrophic lateral sclerosis. Arch Neurol, 1993a; 50: 30-6. Estevez AG, Crow JP, Sampson JB, Reiter C, Zhuang Y, Richardson GJ, Tarpey MM, Barbeito L, Beckman JS. Induction of nitric oxide-dependent apoptos is in motor neurons by zinc-deficient superoxide di smutase. Science, 1999; 286: 2498-500. Ethell DW, Shippy D, Cao C, Cracchiolo JR Runfeldt M, Blake B, Arendash GW. Abeta-specific T-cells reverse cognitive dec line and synaptic loss in Alzheimer's mice. Neurobiol Dis, 2006; 23: 351-61. Farlow MR, Miller ML, Pejovic V. Treat ment options in Alzheimer's disease: maximizing benefit, managing expectations Dement Geriatr Cogn Disord, 2008; 25: 408-22. Farrell CL, Pardridge WM. Blood-brain barrie r glucose transporter is asymmetrically distributed on brain capillary endothelial lu minal and ablumenal membranes: an electron microscopic immunogold study. Proc Natl Acad Sci USA, 1991; 88: 5779-83.

PAGE 230

214 Fisher JW. Erythropoietin: physiology and pharmacology update. Exp Biol Med (Maywood), 2003; 228: 1-14. Fuchs E, Weber K. Intermediate filaments: structure, dynamics, function, and disease. Annu Rev Biochem, 1994; 63: 345-82. Garbuzova-Davis S, Haller E, Saporta S, Kolomey I, Nicosia SV, Sanberg PR. Ultrastructure of blood-brai n barrier and blood-spinal co rd barrier in SOD1 mice modeling ALS. Brain Res, 2007a; 1157: 126-37. Garbuzova-Davis S, Saporta S, Haller E, Ko lomey I, Bennett SP, Potter H, Sanberg PR. Evidence of compromised blood-spinal cord ba rrier in early and late symptomatic SOD1 mice modeling ALS. PLoS ONE, 2007c; 2: e1205. Garbuzova-Davis S, Willing AE, Milliken M, Saporta S, Sowerby B, Cahill DW, Sanberg PR. Intraspinal implantation of hN T neurons into SOD1 mice with apparent motor deficit. Amyotroph Lateral Scler Ot her Motor Neuron Diso rd, 2001; 2: 175-80. Garbuzova-Davis S, Willing AE, Zigova T, Sa porta S, Justen EB, Lane JC, Hudson JE, Chen N, Davis CD, Sanberg PR. Intravenous Administration of Human Umbilical Cord Blood Cells in a Mouse Model of Amyotrophic Lateral Sclerosis: Distribution, Migration, and Diffenentiation. Journal of He matotherapy & Stem Cell Research, 2003a; 12: 255-70. Ge WW, Wen W, Strong W, Leys tra-Lantz C, Strong MJ. Mutant copper-zinc superoxide dismutase binds to and destabilizes human lo w molecular weight neurofilament mRNA. J Biol Chem, 2005; 280: 118-24. Gebicke-Haerter PJ. Microglia in neurodege neration: molecular aspects. Microsc Res Tech, 2001; 54: 47-58. Gendron TF, Petrucelli L. Th e role of tau in neurodegeneration. Mol Neurodegener, 2009; 4: 13. Gerhart DZ, Le Vasseur RJ, Br oderius MA, Drewes LR. Gluc ose transporter localization in brain using light and el ectron immunocytochemistry. J Neurosci Res, 1989; 22: 46472. Geser F, Martinez-Lage M, Kwong LK, Lee VM, Trojanowski JQ. Amyotrophic lateral sclerosis, frontotemporal dementia and beyond: the TDP-43 diseases. J Neurol, 2009. Gonatas NK, Stieber A, Mourelatos Z, Chen Y, Gonatas JO, Appel SH, Hays AP, Hickey WF, Hauw JJ. Fragmentation of the Golgi apparatus of motor neurons in amyotrophic lateral sclerosis. Am J Pathol, 1992; 140: 731-7.

PAGE 231

215 Gordh T, Chu H, Sharma HS. Spinal nerve le sion alters blood-spinal cord barrier function and activated astrocytes in the rat. Pain, 2006; 124: 211-21. Gordon MN, King DL, Diamond DM, Jantzen PT, Boyett KV, Hope CE, Hatcher JM, DiCarlo G, Gottschall WP, Morgan D, Are ndash GW. Correlation between cognitive deficits and Abeta deposits in transgenic APP+PS1 mice. Neurobiol Aging, 2001; 22: 377-85. Gregory RI, Yan KP, Amuthan G, Chendrimada T, Doratotaj B, Cooch N, Shiekhattar R. The Microprocessor complex mediates the genesis of microRNAs. Nature, 2004; 432: 235-40. Griffin WS, Stanley LC, Ling C, White L, MacLeod V, Perrot LJ, White CL, 3rd, Araoz C. Brain interleukin 1 and S-100 immunoreactivity are elevated in Down syndrome and Alzheimer disease. Proc Natl Acad Sci U S A, 1989; 86: 7611-5. Gros-Louis F, Gaspar C, Rouleau GA. Gene tics of familial and sporadic amyotrophic lateral sclerosis. Biochim Biophys Acta, 2006; 1762: 956-72. Guegan C, Vila M, Rosoklija G, Hays AP, Przedborski S. Recruitment of the mitochondrial-dependent apoptotic pathway in amyotrophic lateral scle rosis. J Neurosci, 2001; 21: 6569-76. Gurney ME. Transgenic-mouse model of amyot rophic lateral sclerosis. N Engl J Med, 1994; 331: 1721-2. Gurney ME, Fleck TJ, Himes CS, Hall ED. Riluzole preserves motor function in a transgenic model of familial amyotrophic la teral sclerosis. Neurology, 1998; 50: 62-6. Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, Alexander DD, Caliendo J, Hentati A, Kwon YW, Deng HX, et al. Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science, 1994; 264: 1772-5. Hadano S, Hand CK, Osuga H, Yanagisawa Y, Otomo A, Devon RS, Miyamoto N, Showguchi-Miyata J, Okada Y, Singaraja R, Figlewicz DA, Kwiatkowski T, Hosler BA, Sagie T, Skaug J, Nasir J, Brown RH, Jr., Scherer SW, Rouleau GA, Hayden MR, Ikeda JE. A gene encoding a putative GTPase regulator is mutated in familial amyotrophic lateral sclerosis 2. Nat Genet, 2001; 29: 166-73. Hafezparast M, Klocke R, Ruhrberg C, Ma rquardt A, Ahmad-Annuar A, Bowen S, Lalli G, Witherden AS, Hummerich H, Nicholson S, Morgan PJ, Oozageer R, Priestley JV, Averill S, King VR, Ball S, Peters J, Toda T, Yamamoto A, Hiraoka Y, Augustin M, Korthaus D, Wattler S, Wabnitz P, Dickneite C, Lampel S, Boehme F, Peraus G, Popp A, Rudelius M, Schlegel J, Fuchs H, Hrabe de Angelis M, Schiavo G, Shima DT, Russ AP,

PAGE 232

216 Stumm G, Martin JE, Fisher EM. Mutations in dynein link motor neuron degeneration to defects in retrograde transpor t. Science, 2003; 300: 808-12. Haley RW. Excess incidence of ALS in young Gulf War veterans. Neurology, 2003; 61: 750-6. Hall AA, Herrera Y, Ajmo CT, Jr., Cuevas J, Pennypacker KR. Sigma receptors suppress multiple aspects of microglial act ivation. Glia, 2009; 57: 744-54. Hall ED, Oostveen JA, Gurney ME. Relationshi p of microglial and astrocytic activation to disease onset and progression in a transg enic model of familial ALS. Glia, 1998b; 23: 249-56. Hanisch UK. Microglia as a source and ta rget of cytokines. Glia, 2002; 40: 140-55. Hardy J. Has the amyloid cascade hypothesis for Alzheimer's disease been proved? Curr Alzheimer Res, 2006; 3: 71-3. Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeu tics. Science, 2002; 297: 353-6. Harris J, Ayyub C, Shaw G. A molecular di ssection of the carboxyter minal tails of the major neurofilament subunits NF-M and NF-H. J Neurosci Res, 1991; 30: 47-62. Heller LC, Heller R. In vivo electroporation for gene therapy. Hum Gene Ther, 2006; 17: 890-7. Helmy KY, Katschke KJ, Jr., Gorgani NN, Klja vin NM, Elliott JM, Diehl L, Scales SJ, Ghilardi N, van Lookeren Campagne M. CRIg: a macrophage complement receptor required for phagocytosis of circulat ing pathogens. Cell, 2006; 124: 915-27. Henkel JS, Engelhardt JI, Siklos L, Simpson EP, Kim SH, Pan T, Goodman JC, Sidduque T, Beers DR, Appel SH. Presence of de ndritic cells, MCP-1, and activated microglia/macrophages in amyotrophic lateral sc lerosis spinal cord tissue. Ann Neurol, 2004; 55: 221-35. Hensley K, Mhatre M, Mou S, Pye QN, St ewart C, West M, Williamson KS. On the relation of oxidative stress to neuroinflamma tion: lessons learned from the G93A-SOD1 mouse model of amyotrophic lateral sclerosi s. Antioxid Redox Signal, 2006; 8: 2075-87. Herber DL, Mercer M, Roth LM, Symmonds K, Maloney J, Wilson N, Freeman MJ, Morgan D, Gordon MN. Microg lial activation is required for Abeta clearance after intracranial injection of lipopolysaccharide in APP transgenic mice. J Neuroimmune Pharmacol, 2007; 2: 222-31.

PAGE 233

217 Hernandez F, Avila J. Tauopathies. Cell Mol Life Sci, 2007; 64: 2219-33. Hickman SE, Allison EK, El Khoury J. Micr oglial dysfunction and defective betaamyloid clearance pathways in aging Alzheimer's disease mice. J Neurosci, 2008; 28: 8354-60. Hide I, Tanaka M, Inoue A, Nakajima K, K ohsaka S, Inoue K, Nakata Y. Extracellular ATP triggers tumor necrosis factor-alpha re lease from rat microglia. J Neurochem, 2000; 75: 965-72. Higgins CM, Jung C, Ding H, Xu Z. Mutant Cu, Zn superoxide dismutase that causes motoneuron degeneration is pr esent in mitochondria in th e CNS. J Neurosci, 2002; 22: RC215. Hirano A. Cytopathology of amyotrophic late ral sclerosis. Adv Neurol, 1991; 56: 91-101. Hirano A, Nakano I, Kurland LT, Muld er DW, Holley PW, Saccomanno G. Fine structural study of neurofibrillary changes in a family with amyotrophic lateral sclerosis. J Neuropathol Exp Neurol, 1984; 43: 471-80. Hirtz D, Thurman DJ, Gwinn-Hardy K, Mohamed M, Chaudhuri AR, Zalutsky R. How common are the "common" neurologic disorders? Neurology, 2007; 68: 326-37. Hooper C, Killick R, Lovestone S. The GS K3 hypothesis of Alzheimer's disease. J Neurochem, 2008; 104: 1433-9. Horner RD, Kamins KG, Feussner JR, Gr ambow SC, Hoff-Lindquist J, Harati Y, Mitsumoto H, Pascuzzi R, Spencer PS, Tim R, Howard D, Smith TC, Ryan MA, Coffman CJ, Kasarskis EJ. Occurrence of am yotrophic lateral sclerosis among Gulf War veterans. Neurology, 2003; 61: 742-9. Humpel C, Marksteiner J. Ce rebrovascular damage as a cause for Alzheimer's disease. Curr Neurovasc Res, 2005; 2: 341-7. Hynd MR, Scott HL, Dodd PR. Glutamat e-mediated excitotoxicity and neurodegeneration in Alzheimer's dis ease. Neurochem Int, 2004; 45: 583-95. Ince PG, Shaw PJ, Slade JY, Jones C, Hudgs on P. Familial amyotrophic lateral sclerosis with a mutation in exon 4 of the Cu/Zn s uperoxide dismutase gene: pathological and immunocytochemical changes. Acta Ne uropathol (Berl), 1996; 92: 395-403. Janeway CA, Jr., Medzhitov R. Innate immune recognition. Annu Rev Immunol, 2002; 20: 197-216.

PAGE 234

218 Johnston JA, Dalton MJ, Gurney ME, Kopito RR. Formation of high molecular weight complexes of mutant Cu, Zn-superoxide dismutase in a mouse model for familial amyotrophic lateral sclerosis. Proc Na tl Acad Sci U S A, 2000; 97: 12571-6. Jonsson PA, Ernhill K, Andersen PM, Bergemal m D, Brannstrom T, Gredal O, Nilsson P, Marklund SL. Minute quantities of misfolde d mutant superoxide dismutase-1 cause amyotrophic lateral sclerosi s. Brain, 2004; 127: 73-88. Jung CW, Higgins CMJ, Xu ZS. Mitochondrial electron transport chain complex dysfunction in a transgenic mouse model for amyotrophic lateral sclerosis. J. Neurochem., 2002; 83: 535-45. Kalaria RN. The blood-brain barr ier and cerebral microcircula tion in Alzheimer disease. Cerebrovasc Brain Metab Rev, 1992; 4: 226-60. Kalsheker NA. Alpha 1-antichymotrypsin. Int J Biochem Cell Biol, 1996; 28: 961-4. Kar A, Kuo D, He R, Zhou J, Wu JY. Ta u alternative splicing and frontotemporal dementia. Alzheimer Dis Assoc Disord, 2005; 19 Suppl 1: S29-36. Kaspar BK, Llado J, Sherkat N, Rothstein JD, Gage FH. Retrograde viral delivery of IGF-1 prolongs survival in a mouse ALS model. Science, 2003; 301: 839-42. Kawaji H, Yokomura K, Kikuchi K, Somoto Y, Shirai Y. [Macrophage colonystimulating factor in patients with rheuma toid arthritis]. Nippon Ika Daigaku Zasshi, 1995; 62: 260-70. Kawamata T, Akiyama H, Yamada T, McGeer PL. Immunologic reactions in amyotrophic lateral sclerosis brain and spinal cord tissue. Am J Pathol, 1992; 140: 691707. Kerchner GA, Nicoll RA. Silent synapses and the emergence of a postsynaptic mechanism for LTP. Nat Rev Neurosci, 2008; 9: 813-25. King SJ, Schroer TA. Dynactin increases th e processivity of the cytoplasmic dynein motor. Nat Cell Biol, 2000; 2: 20-4. Kirk J, Plumb J, Mirakhur M, McQuaid S. Tight junctional abnormality in multiple sclerosis white matter affects al l calibers of vessel and associ ated with blood-brain barrier leakage and active demyelinati on. J. Pathol., 2003; 201: 319-27. Kirkinezos IG, Hernandez D, Bradley WG, Moraes CT. An ALS mouse model with a permeable blood-brain barrier benefits fr om systemic cyclosporine A treatment. J Neurochem, 2004; 88: 821-6.

PAGE 235

219 Kirma N, Luthra R, Jones J, Liu YG, Nair HB, Mandava U, Tekmal RR. Overexpression of the colony-stimulating factor (CSF-1) and/or its receptor c-fms in mammary glands of transgenic mice results in hyperplasia and tumor formation. Cancer Res, 2004; 64: 416270. Klivenyi P, Ferrante RJ, Matthews RT Bogdanov MB, Klein AM, Andreassen OA, Mueller G, Wermer M, Kaddurah-Daouk R, Beal MF. Neuroprotective effects of creatine in a transgenic animal model of amyotrophi c lateral sclero sis. Nat Med, 1999; 5: 347-50. Koenigsknecht-Talboo J, Meyer-Luehmann M, Parsadanian M, Garcia-Alloza M, Finn MB, Hyman BT, Bacskai BJ, Holtzman DM. Rapid microglial response around amyloid pathology after systemic anti-Abeta antibody administration in PDAPP mice. J Neurosci, 2008; 28: 14156-64. Kong J, Xu Z. Massive mitochondrial degeneration in motor neurons triggers the onset of amyotrophic lateral sclerosis in mice expres sing a mutant SOD1. J Neurosci, 1998; 18: 3241-50. Kostic V, Jackson-Lewis V, de Bilbao F, Dubois-Dauphin M, Przedborski S. Bcl-2: prolonging life in a transgenic mouse model of familial amyotrophic lateral sclerosis. Science, 1997; 277: 559-62. Kriz J, Nguyen MD, Julien JP. Minocycline slows disease progression in a mouse model of amyotrophic lateral sclerosis. Neurobiol Dis, 2002; 10: 268-78. Kruger C, Laage R, Pitzer C, Schabitz WR, Schneider A. The hematopoietic factor GMCSF (granulocyte-macrophage colony-stim ulating factor) promotes neuronal differentiation of adult neural stem ce lls in vitro. BMC Neurosci, 2007; 8: 88. Kwon EE, Prineas JW. Blood-brain barrier abnormalities in longstanding multiple sclerosis lesions. An immunohistochemical study. J. Neuropathol Exp. Neurol., 1994; 53: 625-36. Lambrechts D, Storkebaum E, Morimoto M, Del-Favero J, Desmet F, Marklund SL, Wyns S, Thijs V, Andersson J, van Marion I, Al-Chalabi A, Bornes S, Musson R, Hansen V, Beckman L, Adolfsson R, Pall HS, Prats H, Vermeire S, Rutgeerts P, Katayama S, Awata T, Leigh N, Lang-Lazdunski L, Dewerc hin M, Shaw C, Moons L, Vlietinck R, Morrison KE, Robberecht W, Van Broeckhoven C, Collen D, Andersen PM, Carmeliet P. VEGF is a modifier of amyotrophic lateral sclerosis in mice and humans and protects motoneurons against ischemic d eath. Nat Genet, 2003; 34: 383-94. Lee MK, Cleveland DW. Neuronal intermedia te filaments. Annu Rev Neurosci, 1996; 19: 187-217.

PAGE 236

220 Leigh PN, Swash M. Cytoskeletal pathol ogy in motor neuron diseases. Adv Neurol, 1991; 56: 115-24. Leigh PN, Whitwell H, Garofalo O, Buller J, Swash M, Martin JE, Gallo JM, Weller RO, Anderton BH. Ubiquitin-immunoreactive intraneu ronal inclusions in amyotrophic lateral sclerosis. Morphology, distribution, and specificity. Brain, 1991; 114 ( Pt 2): 775-88. Leonardi A, Abbruzzese G, Arata L, Cocito L, Vische M. Cerebrospinal fluid (CSF) findings in amyotrophic lateral sclero sis. J. Neurol., 1984b; 231: 75-8. Li G, Yin H, Kuret J. Casein kinase 1 delta phosphor ylates tau and disrupts its binding to microtubules. J Biol Chem, 2004; 279: 15938-45. Li M, Ona VO, Guegan C, Chen M, Jackson-Lewis V, Andrews LJ, Olszewski AJ, Stieg PE, Lee JP, Przedborski S, Friedlander RM. F unctional role of caspase-1 and caspase-3 in an ALS transgenic mouse model. Science, 2000; 288: 335-9. Liebkind R, Laatikainen T, Liesi P. Is th e soluble KDI domain of gamma1 laminin a regeneration factor for the mammalian central nervous system? J Neurosci Res, 2003; 73: 637-43. Liesi P, Laatikainen T, Wright JM. Biol ogically active sequence (KDI) mediates the neurite outgrowth function of the gamma-1 ch ain of laminin-1. J Neurosci Res, 2001; 66: 1047-53. Liu F, Liang Z, Shi J, Yin D, El-Akkad E, Grundke-Iqbal I, Iqbal K, Gong CX. PKA modulates GSK-3betaand cdk5-catalyzed phos phorylation of tau in siteand kinasespecific manners. FEBS Lett, 2006; 580: 6269-74. Liu Q, Xie F, Siedlak SL, Nunomura A, Honda K, Moreira PI, Zhua X, Smith MA, Perry G. Neurofilament proteins in neurodegenera tive diseases. Cell Mol Life Sci, 2004; 61: 3057-75. Loewenstein DA, Acevedo A, Luis C, Cr um T, Barker WW, Duara R. Semantic interference deficits and the detection of mild Alzheimer' s disease and mild cognitive impairment without dementia. J In t Neuropsychol Soc, 2004; 10: 91-100. Ma J, Brewer HB, Jr., Potter H. Alzhei mer A beta neurotoxicity: promotion by antichymotrypsin, ApoE4; inhibition by A be ta-related peptides. Neurobiol Aging, 1996; 17: 773-80. Ma J, Yee A, Brewer HB, Jr., Das S, Potter H. Amyloid-associated proteins alpha 1antichymotrypsin and apolipoprotein E promote assembly of Alzheime r beta-protein into filaments. Nature, 1994; 372: 92-4.

PAGE 237

221 Maccioni RB, Rojo LE, Fernandez JA, Kuljis RO. The role of neuroimmunomodulation in Alzheimer's disease. Ann N Y Acad Sci, 2009; 1153: 240-6. Maguire DJ, Oey H, McCabe M. Alu sequen ces in the human respirome. Adv Exp Med Biol, 2006; 578: 73-9. Malm TM, Koistinaho M, Parepalo M, Vatane n T, Ooka A, Karlsson S, Koistinaho J. Bone-marrow-derived cells contribute to the re cruitment of microglial cells in response to beta-amyloid deposition in APP/PS1 double tr ansgenic Alzheimer mice. Neurobiol Dis, 2005; 18: 134-42. Marambaud P, Zhao H, Davies P. Resveratro l promotes clearance of Alzheimer's disease amyloid-beta peptides. J Biol Chem, 2005; 280: 37377-82. Martin BK, Szekely C, Brandt J, Piantadosi S, Breitner JC, Craft S, Evans D, Green R, Mullan M. Cognitive function over time in the Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT): re sults of a randomized, contro lled trial of naproxen and celecoxib. Arch Neur ol, 2008; 65: 896-905. Mather K, Martin JE, Swash M, Vowles G, Brown A, Leigh PN. Histochemical and immunocytochemical study of ubiquitinated ne uronal inclusions in amyotrophic lateral sclerosis. Neuropathol Appl Neurobiol, 1993; 19: 141-5. Maurice T, Junien JL, Privat A. Dehydroepiandrosterone sulfate attenuates dizocilpineinduced learning impairment in mice via sigm a 1-receptors. Behav Brain Res, 1997; 83: 159-64. McGeer PL, Rogers J, McGeer EG. Infl ammation, anti-inflammatory agents and Alzheimer disease: the last 12 years. J Alzheimers Di s, 2006; 9: 271-6. McKay R. Stem cells in the central ne rvous system. Science, 1997; 276: 66-71. Medzhitov R, Janeway C, Jr. The Toll recepto r family and microbial recognition. Trends Microbiol, 2000; 8: 452-6. Meucci G, Rossi G, Bettini R, Montanaro D, Gironelli L, Voci L, Bianchi F. Laser nephelometric evaluation of albumin, IgG a nd a2-macroglobulin: applications to study of alterations of the blood-brain barrier. J. Neurol. Sci., 1993; 118: 73-8. Meyer-Luehmann M, Spires-Jones TL, Prad a C, Garcia-Alloza M, de Calignon A, Rozkalne A, Koenigsknecht-Talboo J, Ho ltzman DM, Bacskai BJ, Hyman BT. Rapid appearance and local toxicity of amyloid-beta plaques in a mouse model of Alzheimer's disease. Nature, 2008; 451: 720-4.

PAGE 238

222 Miller RG, Anderson FJ, Bradley WG, Brooks BR, Mitsumoto H, Munsat TL, Ringel SP. The ALS patient care database: goals, desi gn, and early results. ALS C.A.R.E. Study Group. Neurology, 2000; 54: 53-7. Miner JH. Laminins and their roles in ma mmals. Microsc Res Tech, 2008; 71: 349-56. Mohamed HA, Mosier DR, Zou LL, Siklos L, Alexianu ME, Engelhardt JI, Beers DR, Le WD, Appel SH. Immunoglobulin Fc gamma r eceptor promotes immunoglobulin uptake, immunoglobulin-mediated calcium increase, a nd neurotransmitter release in motor neurons. J. Neurosci. Res., 2002; 69: 110-6. Moreira PI, Santos MS, Oliveira CR, Shenk JC, Nunomura A, Smith MA, Zhu X, Perry G. Alzheimer disease and the role of free radi cals in the pathogenesis of the disease. CNS Neurol Disord Drug Ta rgets, 2008; 7: 3-10. Mourelatos Z, Gonatas NK, Stieber A, Gurn ey ME, DalCanto MC. The Golgi apparatus of spinal cord motor neurons in transgen ic mice expressing mutant Cu,Zn superoxide dismutase becomes fragmented in early, precli nical stages of the disease. Proc. Natl. Acad. Sci. U. S. A., 1996; 93: 5472-7. Moykkynen T, Liebkind R, Sjoberg J, Korp i ER, Liesi P. The neuroprotective KDI domain of gamma 1-laminin is a universal and potent inhi bitor of ionotropic glutamate receptors. J Neurosci Res, 2005; 81: 797-804. Mucke L, Eddleston M. Astrocytes in infec tious and immune-mediated diseases of the central nervous system. FA SEB J., 1993; 7: 1226-32. Mucke L, Yu GQ, McConlogue L, Rockenstei n EM, Abraham CR, Masliah E. Astroglial expression of human alpha(1)-antichymotryps in enhances alzheimer-like pathology in amyloid protein precursor transgenic mice. Am J Pathol, 2000; 157: 2003-10. Mulder DW, Kurland LT. Motor neuron disease: epidemiologic studies. Adv Exp Med Biol, 1987; 209: 325-32. Mulder DW, Kurland LT, Offord KP, Beard CM. Familial adult motor neuron disease: amyotrophic lateral sclerosi s. Neurology, 1986; 36: 511-7. Mullan M, Crawford F, Axelman K, Houlde n H, Lilius L, Winblad B, Lannfelt L. A pathogenic mutation for probable Alzheimer's di sease in the APP gene at the N-terminus of beta-amyloid. Nat Genet, 1992; 1: 345-7. Myllykangas-Luosujarvi R, Isomaki H. Alzheime r's disease and rheumatoid arthritis. Br J Rheumatol, 1994; 33: 501-2.

PAGE 239

223 Nakamura H, Ueki Y, Sakito S, Matsum oto K, Yano M, Miyake S, Tominaga T, Tominaga M, Eguchi K. Hi gh serum and synovial fluid granulocyte colony stimulating factor (G-CSF) concentrations in patients with rheumatoid arthritis. Clin Exp Rheumatol, 2000; 18: 713-8. Namba Y, Tomonaga M, Ohtsuka K, Oda M, Ikeda K. [HSP 70 is associated with abnormal cytoplasmic inclusions characteristi c of neurodegenerative diseases]. No To Shinkei, 1991; 43: 57-60. Narimatsu N, Harada N, Kurihara H, Naka gata N, Sobue K, Okajima K. Donepezil improves cognitive function in mice by increasing the production of insulin-like growth factor-I in the hippocampus. J Ph armacol Exp Ther, 2009; 330: 2-12. Neumann M, Sampathu DM, Kwong LK, Truax AC, Micsenyi MC, Chou TT, Bruce J, Schuck T, Grossman M, Clark CM, McClus key LF, Miller BL, Masliah E, Mackenzie IR, Feldman H, Feiden W, Kretzschmar HA, Trojanowski JQ, Lee VM. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science, 2006; 314: 130-3. Nilsson LN, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM, Po tter H. Cognitive impairment in PDAPP mice depends on ApoE and ACT-catalyzed amyl oid formation. Neurobiol Aging, 2004; 25: 1153-67. Nilsson LN, Bales KR, DiCarlo G, Gordon MN, Morgan D, Paul SM, Potter H. Alpha-1antichymotrypsin promotes beta-sheet amyloi d plaque deposition in a transgenic mouse model of Alzheimer's disease. J Neurosci, 2001; 21: 1444-51. Niwa J, Ishigaki S, Hishikawa N, Yamamoto M, Doyu M, Murata S, Tanaka K, Taniguchi N, Sobue G. Dorfin ubiquitylates mutant SOD1 and prevents mutant SOD1mediated neurotoxicity. J Biol Chem, 2002; 277: 36793-8. Noble LJ, Wrathall JR. Distribution and time c ourse of protein extravasation in the rat spinal cord after contusive in jury. Brain Res, 1989; 482: 57-66. Noda M, Nakanishi H, Nabekura J, Akaike N. AMPA-Kainate subtypes of glutamate receptor in rat cerebral microg lia. J Neurosci, 2000; 20: 251-8. Olsson Y, Sharma HS, Pettersson A, Ce rvos-Navarro J. Release of endogenous neurochemicals may increase vascular permeability, induce edema and influence cell changes in trauma to the spinal cord. Prog Brain Re s, 1992; 91: 197-203. Olszewski WL, Pazdur J, Kubasiewicz E, Zaleska M, Cooke CJ, Miller NE. Lymph draining from foot joints in rheumatoid arth ritis provides insight in to local cytokine and chemokine production and transport to lymph nodes. Arthritis Rh eum, 2001; 44: 541-9.

PAGE 240

224 Ono S, Imai T, Munakata S, Takahashi K, Kanda F, Hashimoto K, Yamano T, Shimizu N, Nagao K, Yamauchi M. Collagen abnormalities in the spinal cord from patients with amyotrophic lateral sclerosis. J. Neurol, Sci., 1998; 160: 140-7. Oosthuyse B, Moons L, Storkebaum E, Beck H, Nuyens D, Brusselmans K, Van Dorpe J, Hellings P, Gorselink M, Heymans S, Theilmeier G, Dewerchin M, Laudenbach V, Vermylen P, Raat H, Acker T, Vleminckx V, Van Den Bosch L, Cashman N, Fujisawa H, Drost MR, Sciot R, Bruyninckx F, Hicklin DJ, Ince C, Gressens P, Lupu F, Plate KH, Robberecht W, Herbert JM, Collen D, Carm eliet P. Deletion of the hypoxia-response element in the vascular endothelial grow th factor promoter causes motor neuron degeneration. Nat Genet, 2001; 28: 131-8. Oppenheim RW. Neurotrophic survival molecu les for motoneurons: an embarrassment of riches. Neuron, 1996; 17: 195-7. Otomo A, Hadano S, Okada T, Mizumura H, Kunita R, Nishijima H, Showguchi-Miyata J, Yanagisawa Y, Kohiki E, Suga E, Yas uda M, Osuga H, Nishimoto T, Narumiya S, Ikeda JE. ALS2, a novel guanine nucleotide ex change factor for the small GTPase Rab5, is implicated in endosomal dynamics. Hum Mol Genet, 2003; 12: 1671-87. Padmanabhan J, Levy M, Dickson DW, Potter H. Alpha1-antichymotrypsin, an inflammatory protein overexpressed in Alzheimer's disease brain, induces tau phosphorylation in neurons. Brain, 2006; 129: 3020-34. Pallares-Trujillo J, LopezSoriano FJ, Argiles JM. The involvement of the ubiquitin system in Alzheimer's disease (rev iew). Int J Mol Med, 1998; 2: 3-15. Pardridge WM. Advances in cell biology of blood-brain barrier transport. Semin. Cell Biol., 1991; 2: 419-26. Pardridge WM. The blood-brain barrier and neurotherapeutics. Ne uroRx, 2005; 2: 1-2. Pardridge WM. Blood-brain barrier biology and methodology. J. NeuroVirol., 1999; 5: 556-69. Pardridge WM. Recent advances in blood-brai n barrier transport. Ann. Rev. Pharmacol. Toxicol., 1988; 28: 25-39. Parsonage G, Filer A, Bik M, Hardie D, Lax S, Howlett K, Church LD, Raza K, Wong SH, Trebilcock E, Scheel-Toellner D, Sa lmon M, Lord JM, Buckley CD. Prolonged, granulocyte-macrophage colony-stimulating factor-dependent, neutrophil survival following rheumatoid synovial fibroblast act ivation by IL-17 and TNFalpha. Arthritis Res Ther, 2008; 10: R47.

PAGE 241

225 Pasinelli P, Houseweart MK Brown RH, Jr., Cleveland DW. Caspase-1 and -3 are sequentially activated in motor neuron deat h in Cu,Zn superoxide dismutase-mediated familial amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A, 2000; 97: 13901-6. Pei JJ, Braak H, An WL, Winblad B, Co wburn RF, Iqbal K, Grundke-Iqbal I. Upregulation of mitogen-activated protein kinases ERK1/2 and MEK1/2 is associated with the progression of neurofibrillary degenerati on in Alzheimer's disease. Brain Res Mol Brain Res, 2002; 109: 45-55. Perez-Martinez D. [The role of lithium in neurodegenerative diseases: new registries for old actors.]. Neurol ogia, 2009; 24: 143-6. Pessac B, Godin I, Alliot F. [Microglia: origin and developmen t]. Bull Acad Natl Med, 2001; 185: 337-46; discussion 46-7. Pirttila T, Vanhatalo S, Turpeinen U, Riikone n R. Cerebrospinal fluid insulin-like growth factor-1, insulin growth factor binding protein-2 or nitric oxide are not increased in MS or ALS. Acta Neurol Scand., 2004; 109: 337-41. Popescu BO, Toescu EC, Popescu LM, Bajenaru O, Muresanu DF, Schultzberg M, Bogdanovic N. Blood-brain barrier alterations in ageing and dementia. J Neurol Sci, 2009. Porcellini E, Davis EJ, Chiappelli M, Ianni E, Di Stefano G, Forti P, Ravaglia G, Licastro F. Elevated plasma levels of alpha-1-antichymotrypsin in age-related cognitive decline and Alzheimer's disease: a potential therap eutic target. Curr Pharm Des, 2008; 14: 265964. Potter H, Wefes IM, Nilsson LN. The inflammation-induced pathological chaperones ACT and apo-E are necessary catalysts of Alzheimer amyloid formation. Neurobiol Aging, 2001; 22: 923-30. Prat A, Biernacki K, Wosik K, Antel JP. G lial cell influence on the human blood-brain barrier. Glia, 2001; 36: 145-55. Prusiner SB. The prion diseases Brain Pathol, 1998; 8: 499-513. Puls I, Jonnakuty C, LaMonte BH, Holzbaur EL, Tokito M, Mann E, Floeter MK, Bidus K, Drayna D, Oh SJ, Brown RH, Jr., Ludlow CL, Fischbeck KH. Mutant dynactin in motor neuron disease. Nat Genet, 2003; 33: 455-6. Racke K, Juergens UR, Matthiesen S. C ontrol by cholinergic mechanisms. Eur J Pharmacol, 2006; 533: 57-68.

PAGE 242

226 Rambaran RN, Serpell LC. Amyloid fibrils: abnormal protein asse mbly. Prion, 2008; 2: 112-7. Ravits J, Laurie P, Fan YX, Moore DH. Im plications of ALS focality Rostral-caudal distribution of lower motor neuron loss postmortem. Lippincott Williams & Wilkins, 2007: 1576-82. Rebenko-Moll NM, Liu L, Cardona A, Rans ohoff RM. Chemokines, mononuclear cells and the nervous system: heaven (or hell) is in the detail s. Curr. Opin. Immunol., 2006; 18: 683-9. Rezai-Zadeh K, Arendash GW, Hou H, Fernand ez F, Jensen M, Runfeldt M, Shytle RD, Tan J. Green tea epigallocatechin-3-gallate (EGCG) reduces beta-amyloid mediated cognitive impairment and modulates tau path ology in Alzheimer transgenic mice. Brain Res, 2008; 1214: 177-87. Rhodin J, Thomas T, Bryant M, Sutton ET. Animal model of Alzheimer-like vascular pathology and inflammatory reaction. Ann N Y Acad Sci, 2000; 903: 345-52. Rizvanov AA, Mukhamedyarov MA, Palotas A, Islamov RR. Retrogradely transported siRNA silences human mutant SOD1 in spinal cord motor neurons. Exp Brain Res, 2009; 195: 1-4. Robertson J, Beaulieu JM, Doroudchi MM, Durham HD, Julien JP, Mushynski WE. Apoptotic death of neurons exhibiting peripherin aggregates is mediated by the proinflammatory cytokine tumor necrosis factor-alpha. J Cell Biol, 2001; 155: 217-26. Rogers J, Li R, Mastroeni D, Grover A, Le onard B, Ahern G, Cao P, Kolody H, Vedders L, Kolb WP, Sabbagh M. Peripheral clearance of amyloid beta peptide by complement C3-dependent adherence to erythrocytes. Neurobiol Aging, 2006; 27: 1733-9. Rosen DR. Mutations in Cu/Zn superoxide dism utase gene are associated with familial amyotrophic lateral sclero sis. Nature, 1993; 364: 362. Rothstein JD, Dykes-Hoberg M, Pardo CA, Bristol LA, Jin L, Kuncl RW, Kanai Y, Hediger MA, Wang Y, Schielke JP, Welty DF. Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and cl earance of glutamate. Neuron, 1996; 16: 675-86. Rothstein JD, Kuncl R, Chaudhry V, Claw son L, Cornblath DR, Coyle JT, Drachman DB. Excitatory amino acids in amyotrophic lateral sclerosis: an update. Ann Neurol, 1991; 30: 224-5.

PAGE 243

227 Rothstein JD, Tsai G, Kuncl RW, Clawson L, Cornblath DR, Drachman DB, Pestronk A, Stauch BL, Coyle JT. Abnormal excitatory amino acid metabolism in amyotrophic lateral sclerosis. Ann Neurol, 1990; 28: 18-25. Rowland LP, Shneider NA. Amyotrophic lateral sclerosis. New Engl J Med, 2001a; 344: 1688-700. Rutka JT, Apodaca G, Stern R, Rosenblum M. The extracellular matrix of the central and peripheral nervous systems: structure a nd function. J. Neurosurg., 1988; 69: 155-70. Sadowski MJ, Pankiewicz J, Scholtzova H, Mehta PD, Prelli F, Quartermain D, Wisniewski T. Blocking the apolipoprotein E/amyloid-beta intera ction as a potential therapeutic approach for Alzheimer's diseas e. Proc Natl Acad Sci U S A, 2006; 103: 18787-92. Sanchez-Ramos J, Song S, Sava V, Catlow B, Lin X, Mori T, Cao C, Arendash GW. Granulocyte Colony Stimulating Factor (G-C SF) Decreases Brain Amyloid Burdenand Reverses Cognitive Impairment in Alzheimer's Mice. Neuroscience, 2009. Sargsyan SA, Monk PN, Shaw PJ. Microglia as potential contributors to motor neuron injury in amyotrophic lateral sc lerosis. Glia, 2005; 51: 241-53. Saria A, Lundberg JM. Evans blue fluores cence: quantitative and morphological evaluation of vascular permeability in animal tissues. J Neurosci Methods, 1983; 8: 41-9. Sasaki S, Shibata N, Komori T, Iwata M. iNOS and nitrotyrosine immunoreactivity in amyotrophic lateral sclerosis. Ne urosci. Lett., 2000; 291: 44-8. Schabitz WR, Kruger C, Pitzer C, Weber D, Laage R, Gassler N, Aronowski J, Mier W, Kirsch F, Dittgen T, Bach A, Sommer C, Sc hneider A. A neuroprotective function for the hematopoietic protein granulocyte-macrophage colony stimulating factor (GM-CSF). J Cereb Blood Flow Metab, 2008; 28: 29-43. Schaeffer EL, Gattaz WF. Cholinergic and gl utamatergic alterati ons beginning at the early stages of Alzheimer disease: participation of the phospholipase A2 enzyme. Psychopharmacology (Berl), 2008; 198: 1-27. Schellekens GA, Visser H, de Jong BA, va n den Hoogen FH, Hazes JM, Breedveld FC, van Venrooij WJ. The diagnostic properties of rheumatoid arthritis antibodies recognizing a cyclic citru llinated peptide. Arthritis Rheum, 2000; 43: 155-63.

PAGE 244

228 Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Levy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D, Younkin S. Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutatio ns linked to familial Alzheimer's disease. Nat Med, 1996; 2: 864-70. Schlett K. Glutamate as a modulator of em bryonic and adult neurogenesis. Curr Top Med Chem, 2006; 6: 949-60. Schneider UC, Schilling L, Schroeck H, Nebe CT, Vajkoczy P, Woitzik J. Granulocytemacrophage colony-stimulating factor-induced vessel growth restores cerebral blood supply after bilateral ca rotid artery occlusion. Stroke, 2007; 38: 1320-8. Selkoe DJ. Altered structural proteins in plaq ues and tangles: what do they tell us about the biology of Alzheimer's disease? Neurobiol Agi ng, 1986; 7: 425-32. Sharma HS. Pathophysiology of blood-spinal cord barrier in traumatic injury and repair. Curr Pharm Des, 2005; 11: 1353-89. Shaw PJ. Molecular and cellular pathways of neurodegeneration in motor neurone disease. J Neurol Neurosur g Psychiatry, 2005; 76: 1046-57. Shaw PJ, Forrest V, Ince PG, Richardson JP, Wastell HJ. CSF and plasma amino acid levels in motor neuron disease: elevation of CSF glutamate in a subset of patients. Neurodegeneration, 1995; 4: 209-16. Shibata N, Kobayashi M. [The role for oxi dative stress in neurodegenerative diseases]. Brain Nerve, 2008; 60: 157-70. Shimura H, Schlossmacher MG, Hattori N, Fr osch MP, Trockenbacher A, Schneider R, Mizuno Y, Kosik KS, Selkoe DJ. Ubiquitina tion of a new form of alpha-synuclein by parkin from human brain: implications for Parkinson's disease. Science, 2001; 293: 2639. Shinder GA, Lacourse MC, Minotti S, Durham HD. Mutant Cu/Zn-superoxide dismutase proteins have altered solubility and interact with heat shock/stress pr oteins in models of amyotrophic lateral sclerosis. J Biol Chem, 2001; 276: 12791-6. Siddique T, Figlewicz DA, Pericak-Vance MA, Haines JL, Rouleau G, Jeffers AJ, Sapp P, Hung WY, Bebout J, McKenna-Yasek D, et al. Linkage of a gene causing familial amyotrophic lateral sclerosi s to chromosome 21 and ev idence of genetic-locus heterogeneity. N Engl J Med, 1991; 324: 1381-4.

PAGE 245

229 Siddique T, Nijhawan D, Hentati A. Molecu lar genetic basis of familial ALS. Neurology, 1996; 47: S27-34; discussion S-5. Simard AR, Rivest S. Bone marrow stem ce lls have the ability to populate the entire central nervous system into fully differentiated parenchymal microglia. FASEB J, 2004; 18: 998-1000. Simard AR, Soulet D, Gowing G, Julien JP, Rivest S. Bone marrow-derived microglia play a critical role in restri cting senile plaque formation in Alzheimer's disease. Neuron, 2006; 49: 489-502. Sisodia SS, Price DL. Role of the beta-amyloid protein in Alzheimer's disease. FASEB J, 1995; 9: 366-70. Smith RA, Miller TM, Yamanaka K, Monia BP, Condon TP, Hung G, Lobsiger CS, Ward CM, McAlonis-Downes M, Wei H, Wancewicz EV, Bennett CF, Cleveland DW. Antisense oligonucleotide therapy for neurodegenerative disease. J Clin Invest, 2006; 116: 2290-6. Soto C. Unfolding the role of protein misf olding in neurodegenerative diseases. Nat Rev Neurosci, 2003; 4: 49-60. Spasic D, Annaert W. Building gamma-secret ase: the bits and pieces. J Cell Sci, 2008; 121: 413-20. Stewart PA, Hayakawa K, Farrell CL. Quantita tion of blood-brain barrier ultrastructure. Microsc Res Tech, 1994; 27: 516-27. Strittmatter WJ, Weisgraber KH, Huang DY, Dong LM, Salvesen GS, Pericak-Vance M, Schmechel D, Saunders AM, Goldgaber D, Roses AD. Binding of human apolipoprotein E to synthetic amyloid beta peptide: isofor m-specific effects and implications for lateonset Alzheimer disease. Proc Natl Acad Sci U S A, 1993; 90: 8098-102. Strong MJ, Kesavapany S, Pant HC. The pat hobiology of amyotrophic la teral sclerosis: a proteinopathy? J Neuropathol Exp Neurol, 2005b; 64: 649-64. Strong MJ, Volkening K, Hammo nd R, Yang W, Strong W, Leystra-Lantz C, Shoesmith C. TDP43 is a human low molecular wei ght neurofilament (hNFL) mRNA-binding protein. Mol Cell Neurosci, 2007; 35: 320-7. Swerdlow RH. The Neurodegenerative Mito chondriopathies. J Alzheimers Dis, 2009. Szekanecz Z, Koch AE. Macrophages and their products in rheumatoid arthritis. Curr Opin Rheumatol, 2007; 19: 289-95.

PAGE 246

230 Takeuchi H, Kobayashi Y, Yoshihara T, Ni wa J, Doyu M, Ohtsuka K, Sobue G. Hsp70 and Hsp40 improve neurite outgrowth and suppr ess intracytoplasmic aggregate formation in cultured neuronal cells expressing mutant SOD1. Brain Res, 2002; 949: 11-22. Tanaka K, Watase K, Manabe T, Yamada K, Watanabe M, Takahashi K, Iwama H, Nishikawa T, Ichihara N, Kikuchi T, Okuyama S, Kawashima N, Hori S, Takimoto M, Wada K. Epilepsy and exacerba tion of brain injury in mice lacking the glutamate transporter GLT-1. Scie nce, 1997; 276: 1699-702. Tanzi RE, George-Hyslop PS, Gusella JF. Mo lecular genetics of Alzheimer disease amyloid. J Biol Chem, 1991; 266: 20579-82. Tortarolo M, Grignaschi G, Calvaresi N, Zennaro E, Spaltro G, Colovic M, Fracasso C, Guiso G, Elger B, Schneider H, Seilheimer B, Caccia S, Bendotti C. Glutamate AMPA receptors change in motor neurons of SOD1G93A transgenic mice and their inhibition by a noncompetitive antagonist am eliorates the progression of amytrophic lateral sclerosislike disease. J Neurosci Res, 2006; 83: 134-46. Tovar-y-Romo LB, Tapia R. Cerebral neurons of transgenic ALS mi ce are vulnerable to glutamate release stimulation but not to increased extracellular glutamate due to transport blockade. Exp Neurol, 2006; 199: 281-90. Trojanowski JQ, Mawal-Dewan M, Schmidt ML Martin J, Lee VM. Localization of the mitogen activated protein kinase ERK2 in Al zheimer's disease neurof ibrillary tangles and senile plaque neurites. Brain Res, 1993; 618: 333-7. Troost D, Van den Oord JJ, Vianney de J ong JM. Immunohistochemical characterization of the inflammatory infiltrate in amyotr ophic lateral sclerosis. Neuropathol Appl Neurobiol, 1990a; 16: 401-10. Tu PH, Raju P, Robinson KA, Gurney ME, Trojanowski JQ, Lee VMY. Transgenic mice carrying a human mutant supero xide dismutase transgene deve lop neuronal cytoskeletal pathology resembling human amyotrophic lateral sclerosis lesions. Proc. Natl. Acad. Sci. U. S. A., 1996; 93: 3155-60. Turner BJ, Talbot K. Transgenics, toxicity and therapeutics in rodent models of mutant SOD1-mediated familial ALS. Prog Neurobiol, 2008; 85: 94-134. Ueno M. Molecular anatomy of the brain e ndothelial barrier: an overview of the distributional features. Cu rr Med Chem, 2007; 14: 1199-206. Van Den Bosch L, Tilkin P, Lemmens G, Robberecht W. Minocycline delays disease onset and mortality in a transgenic model of ALS. Neuroreport, 2002; 13: 1067-70.

PAGE 247

231 Vanacore N, Binazzi A, Bottazzi M, Belli S. Amyotrophic lateral scle rosis in an Italian professional soccer player. Parkinsonism Relat Disord, 2006; 12: 327-9. Vorbrodt AW, Dobrogowska DH. Molecular an atomy of intercellular junctions in the brain endothelial and epithelial barriers: electron microsc opist's view. Brain Res. Rev., 2003; 42: 221-42. Vos CMP, Geurts JJG, Montagne L, van Haaste rt ES, B L, van der Valk P, Barkhof F, de Vries HE. Blood-brain barrier alterations in both focal and diffuse abnormalities on postmortem MRI in multiple sclerosi s. Neurobiol Dis, 2005; 20: 953-60. Vukosavic S, Stefanis L, Jackson-Lewis V, Guegan C, Romero N, Chen C, DuboisDauphin M, Przedborski S. Delaying caspase activation by Bcl-2: A clue to disease retardation in a transgenic mouse model of amyotrophic lateral sclerosis. J Neurosci, 2000; 20: 9119-25. Walker LC, Bian F, Callahan MJ, Lipinski WJ, Durham RA, LeVine H. Modeling Alzheimer's disease and other proteopathies in vivo: is seeding the key? Amino Acids, 2002; 23: 87-93. Walsh DM, Selkoe DJ. A beta oligomers a decade of discovery. J Neurochem, 2007; 101: 1172-84. Wang HY, Wang IF, Bose J, Shen CK. Structural diversity and functio nal implications of the eukaryotic TDP gene fam ily. Genomics, 2004; 83: 130-9. Wang J, Slunt H, Gonzales V, Fromholt D, Coonfield M, Copeland NG, Jenkins NA, Borchelt DR. Copper-binding-site-null SOD1 cau ses ALS in transgenic mice: aggregates of non-native SOD1 delineate a common feature. Hum Mol Genet, 2003; 12: 2753-64. Wang JZ, Grundke-Iqbal I, Iqbal K. Kinases and phosphatases and tau sites involved in Alzheimer neurofibrillary degenera tion. Eur J Neurosci, 2007; 25: 59-68. Wang L, Deng HX, Grisotti G, Zhai H, Siddique T, Roos RP. Wild-type SOD1 overexpression accelerates disease onset of a G85R SOD1 mouse. Hum Mol Genet, 2009; 18: 1642-51. Weisman D, Hakimian E, Ho GJ. Interl eukins, inflammation, and mechanisms of Alzheimer's disease. Vi tam Horm, 2006; 74: 505-30. Weydt P, Hong SY, Kliot M, Moller T. A ssessing disease onset and progression in the SOD1 mouse model of ALS. NeuroReport, 2003; 14: 1051-4. Weydt P, Yuen EC, Ransom BR, Moller T. Increased cytotoxic potential of microglia from ALS-transgenic mice. Glia, 2004; 48: 179-82.

PAGE 248

232 Wickner RB, Shewmaker F, Kryndushkin D, Ed skes HK. Protein inheritance (prions) based on parallel in-register beta-sheet amyloid structures. Bioessays, 2008; 30: 955-64. Wiksten M, Liebkind R, Laatikainen T, Lies i P. Gamma 1 laminin and its biologically active KDI-domain may guide axons in the floor plate of human embryonic spinal cord. J Neurosci Res, 2003; 71: 338-52. Wiksten M, Vaananen A, Liebkind R, Rauha la P, Liesi P. Soluble KDI domain of gamma1 laminin protects adult hippocampus from excitotoxicity of kainic acid. J Neurosci Res, 2004a; 78: 411-9. Wiksten M, Vaananen AJ, Liebkind R, Liesi P. Regeneration of adult rat spinal cord is promoted by the soluble KDI domain of ga mma1 laminin. J Neurosci Res, 2004b; 78: 403-10. Williamson TL, Bruijn LI, Zhu Q, Anderson KL, Anderson SD, Julien JP, Cleveland DW. Absence of neurofilaments reduces the selective vulnerability of motor neurons and slows disease caused by a familial amyotrophi c lateral sclerosis-linked superoxide dismutase 1 mutant. Proc Natl A cad Sci U S A, 1998; 95: 9631-6. Williamson TL, Cleveland DW. Slowing of axonal transport is a very early event in the toxicity of ALS-linked SOD1 mutants to mo tor neurons. Nat. Neurosci., 1999; 2: 50-6. Willing AE, Pennypacker KR. Alternate approach to understanding the molecular mechanisms of stroke-induced injury Histol Histopathol 2007; 22: 697-701. Wisniewski T, Castano EM, Golabek A, Vogel T, Frangione B. Acceleration of Alzheimer's fibril formation by apolipoprotei n E in vitro. Am J Pathol, 1994; 145: 10305. Wisniewski T, Frangione B. Apolipoprotein E: a pathological chaperone protein in patients with cerebral and systemic am yloid. Neurosci Lett, 1992; 135: 235-8. Wolfe MS. Gamma-secretase: structure, function, and modulation for Alzheimer's disease. Curr Top Med Chem, 2008; 8: 2-8. Wong PC, Pardo CA, Borchelt DR, Lee MK Copeland NG, Jenkins NA, Sisodia SS, Cleveland DW, Price DL. An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria. Neuron, 1995; 14: 1105-16. Wyss-Coray T, Yan F, Lin AH, Lambris JD, Alexander JJ, Quigg RJ, Masliah E. Prominent neurodegeneration and increased pl aque formation in complement-inhibited Alzheimer's mice. Proc Natl Acad Sci U S A, 2002; 99: 10837-42.

PAGE 249

233 Xu WD, Firestein GS, Taetle R, Kaushans ky K, Zvaifler NJ. Cytokines in chronic inflammatory arthritis. II. Granulocytemacrophage colony-stimulating factor in rheumatoid synovial effusions. J Clin Invest, 1989; 83: 876-82. Yamanaka K, Vande Velde C, Eymard-Pierre E, Bertini E, Boespflug-Tanguy O, Cleveland DW. Unstable mutants in the pe ripheral endosomal membrane component ALS2 cause early-onset motor neuron disease. Proc Natl Acad Sci U S A, 2003; 100: 16041-6. Yang Y, Hentati A, Deng HX, Dabbagh O, Sa saki T, Hirano M, Hung WY, Ouahchi K, Yan J, Azim AC, Cole N, Gascon G, Yagmour A, Ben-Hamida M, Pericak-Vance M, Hentati F, Siddique T. The gene encoding al sin, a protein with three guanine-nucleotide exchange factor domains, is mutated in a form of recessive amyotrophic lateral sclerosis. Nat Genet, 2001; 29: 160-5. Yenari MA, Xu L, Tang XN, Qi ao Y, Giffard RG. Microglia potentiate damage to bloodbrain barrier constituents: improvement by minocycline in vivo and in vitro. Stroke, 2006; 37: 1087-93. Zheng WH, Bastianetto S, Mennicken F, Ma W, Kar S. Amyloid beta peptide induces tau phosphorylation and loss of cholinergic neurons in rat primary septal cultures. Neuroscience, 2002; 115: 201-11. Zhu S, Stavrovskaya IG, Drozda M, Kim BY Ona V, Li M, Sarang S, Liu AS, Hartley DM, Wu DC, Gullans S, Ferrante RJ, Prze dborski S, Kristal BS, Friedlander RM. Minocycline inhibits cytochrome c release a nd delays progression of amyotrophic lateral sclerosis in mice. Na ture, 2002; 417: 74-8. Zimmerman MC, Oberley LW, Flanagan SW. Mu tant SOD1-induced neuronal toxicity is mediated by increased mitochondrial super oxide levels. J Neuroc hem, 2007; 102: 609-18.

PAGE 250

234 APPENDICES

PAGE 251

235 Appendix ALymphopenia a nd Spontaneous Autorosette Formation in an SOD1 Model of ALS Oleg I. Kuzmenok a, Paul R. Sanberg a,b,d,e,f, Tammy G. Desjarlais a, Steven P. Bennett g Svitlana N. Garbuzova-Davis a,b,c,d,* Abstract Amyotrophic lateral sclerosis (ALS) is a ne urodegenerative disorder characterized by motoneuron degeneration. Increasing evidence suggests immune system involvement in ALS pathogenesis but information about peripheral blood characte ristics has been lacking. We evaluated hematological and mo rphological parameters in peripheral blood of G93A SOD1 mice. A signifi cant decrease in white blood cells was found at the end stage of disease. The lym phocyte reduction may suggest immunodeficiency in ALS. Spontaneously forming rosettes with au tologous erythrocytes were noted in approximately 28% of lymphocytes in SOD1 mice. To our knowledge, this is the first study characterizing hematology and revealing autorosettes in the S OD1 mouse model of ALS at the terminal phase of disease. Introduction Currently, amyotrophic lateral sclerosis (ALS) is the most frequently occurring motor neuron disease lacking effective therapy (Mul der et al., 1986). About 10% of cases are dominantly inherited and about 20% of those arise from muta tions in the gene for Cu/Zn superoxide dismutase (SOD1). The pathogeni c mechanism of ALS is unknown and there

PAGE 252

236 Appendix A: (Continued) is no identified biological marker of disease progression that can be monitored in ALS patients. In ALS spinal cord tissue, the pr esence of T cells (Ale xianu et al., 2001), IgG deposits (Mohamed et al., 2002; Engelhardt and Appel, 1990), monocytes and activated macrophages (Henkel et al., 2004), suggests that these immune e ffectors are involved in the inflammatory process. Inflammato ry response occurs in the CNS through mechanisms that differ from systemic inflam mation (Consilvio et al., 2004), as described for ALS patients (McGeer and McGeer, 2002). Several transgenic mice modeling ALS expressing mutated human SOD1 have been cr eated. The first and most widely used is the G93A mouse that overe xpresses human SOD1, carryi ng the Gly93 Ala mutation and that develops an ALSlike motor neuron di sease. G93A SOD1 mi ce show motor neuron pathology similar to ALS patients (Gurney et al., 1994). Still, we lack information regarding the hematological characterist ics of both ALS patients and animals. Characteristics of complete blood count ( CBC) and white blood cell (WBC) differential may therefore help elucidatethe type of pathogenic processes present during disease, very important for developing a treatment stra tegy. Previously, an increasing number of circulating neutrophils found in ALS patients had been described (D esport et al., 2001). However, statistically significant differences between the number of neutrophils in ALS patients and neutrophils normal range were not shown. The aim of the present work was to evaluate the complex he matological and morphological parameters in peripheral blood from th e G93A SOD1 mouse model of ALS2. Materials and methods Mice Transgenic male mice B6SJL-TgN 1GUR (G93A) obtained from Jackson Laboratories, Bar Harbor, MA, overexpressing human S OD1, carrying the Gly93 to Ala mutation (Gurney et al., 1994), were used (n =12). Transgenic mice (BL6/SJL) carrying the normal

PAGE 253

237 Appendix A: (Continued) allele for human SOD1 gene (n =5, hTg) and C57BL/6 animals (n =7), matched by age and gender to experimental animals were us ed as controls. All animal experiment procedures were approved by the Institutional Animal Care and Use Committee at USF and conducted in compliance with the Guide for the Care and Use of Laboratory Animals. Blood samples The blood samples were obtained from G93A mi ce at the terminal stage of disease when mice developed hind limb paralysis (18 week s). Samples were also obtained from hTg and C57BL/6 mice of the same age, 18 weeks. Mice were sacrificed under deep chloral hydrate (10%) an esthesia and about 300 Al of bl ood was taken transcardially and collected into blood collection tubes w ith EDTA(K3) (Sherwood Medical, MO). Analyses for CBC and WBC di fferential were performed by Antech Diagnostics. (NY, USA). Additionally, blood smears (3 from each animal) were taken from the tail vein for morphological analysis. The smears were dried for 30 min, fixed in methanol for 7 min, then stained by Giemsa (SigmaAldr ich, GS80, St. Louis, MO) as previously described (Brown and Febiger, 1993). Morphological evaluation of blood smears After staining, blood smears were rinsed several times in distilled water and coverslipped with Permount (Fisher Scientific Fair Lawn, NJ). The morphology of the peripheral blood cells was examined under an Olympus BX-60 microscope. The images were analyzed by Image-Pro Plus versi on 4.1 for Windows software (Silver Spring, MD). Rosette counts were performed us ing the full area of the slide.

PAGE 254

238 Appendix A: (Continued) Statistical analysis All data on CBC count and WBC differentia l were analyzed by ANOVA (analysis of variance). The results are presented as m eans TSEM. Significance between the results for different groups was estimated us ing the two-tailed Students t-test. Results Hematological analysis The blood samples were obtained from G93A mi ce at the terminal stage of disease (about 18 weeks of age). All G93A mice had seve re disease progression (complete hind limb paralysis) at the time of bl ood collection. The main characte ristics of the red and white blood cells from G93A and control (hTg and C 57BL/6) groups of mi ce are presented in Table 1. The red blood cell counts did not differ between G93A mice and the two control groups. However, the platelet numbers were increased in G93A mi ce without statistical significance. In G93A mice, WBC numbers were drasti cally reduced (0.7_103/AL, p <0.001) compared to control hTg mi ce (2.6_103/AL) and C57BL/6 (3.8_103/AL). Lymphocyte density in G93A mice (0.26_103/AL) was 8.4 times lower than in hTg mice (2.22_103/AL) and 10 times lower than in C57BL/6 mice (2.66_103/AL). The absolute number of circulating monocytes and eosi nophils also decreased ( p <0.05) in G93A mice. The basophil number was significantly increased in G93A mice compared to hTg mice ( p <0.05), but not to C57BL/6 mice. This dramatic reduction in the number of lymphocytes in peripheral blood may suggest severe immunodeficiency in ALS. Leukocytes are typically recr uited from circulation into local sites of inflammation, where the cells perform their functions to re duce tissue alterations. However, neutrophil levels, which can reflect acute inflammation in peripheral blood of G93A mice, did not significantly differ from control animals.

PAGE 255

239 Appendix A: (Continued) Morphological examination of the peripheral blood in SOD1 mice Routine blood smears can still be useful a nd informative if current views on disease pathogenesis are incorporated during thei r perusal. Morphologi cal examination of peripheral blood from G93A mice showed a number of lymphocytes spontaneously forming rosettes with autologous erythrocyt es or autorosette-forming lymphocytes (aRFL) (Fig. 1A,B). The mean percentage of aRFL was 27.7T3.3 in the peripheral blood from G93A mice. The average number of er ythrocytes which formed rosettes around one lymphocyte in G93A mice was 8.6T3.2. A squeezed shape of erythrocytes in rosettes was noted. The aRFL were found in two thirds of G93A mice but not in the peripheral blood of C57BL/6 (Fig. 1C) or hTg (Fig. 1D,E) control animals. Discussion It has been shown that the im mune system is involved in ALS in human and G93A mice (Alexianu et al., 2001; Mohamed et al., 2002; Engelhardt and Appel, 1990). The clinical significance and degree of involvement of the immune system in ALS is still controversial. Nyland and Nae ss (1978) reported that T lym phocytes in the peripheral blood of ALS patients were not significantly reduced. However, othe r clinical studies found decreased totals of blood lymphocytes and increased percenta ges of B-cells and IgM (Khondkarian et al., 1985) and reductions of T mu cells, which bear Fc receptor for IgM (Westall et al., 1983). Moreover, Provincia li et al. (1988) showed significant marked lymphopenia, reduction of CD2, CD8 and Leu 7 positive cells and an increase of the CD4/CD8 ratio in ALS patients at early disease stage. From these results, it is clear that immunopathological reactions in ALS ar e occurring and may depend upon duration, severity, and cause of disease. However, bi ological markers of th is disease are still uncertain. Utilizing CBC and WBC differential, we were the first to find a significant

PAGE 256

240 Appendix A: (Continued) decrease of WBC numbers in G93A SOD1 mice at the end stage of disease. Decreased WBC in G93A mice was mainly due to declin es in lymphocyte density, 8.4 times lower than in hTg mice and 10 times lower than in C57BL/6 control mice. Significantly decreased numbers of peripheral monocytes a nd eosinophils in G93A mice versus control animals were also noted. These findings ma y suggest severe immunodeficiency in these animals at the terminal stage of disease. Since the transgenic G 93A mice have a high number (20) of copies of mu tant SOD1, it is possible to suggest that pe ripheral blood results are be linked to ALS gene mutations. This SOD1-med iated toxicity may involve a cascade of events including depletion of peri pheral lymphocytes. Also, we cannot exclude the possibility that consumption of these cells is extremely high, causing their disappearance fromperipheral blood. Monocytop enia and eosinopenia can be considered as prognostic features in ALS. An eo sinophil is a type of phagocyte that produces the pro-inflammatory protein histamin e, as well as nerve growth factor (NGF) and neurotrophin-3 (NT-3) (Kobayashi et al., 2002). Although most eosinophilic leukocytes are localized in mucosal tissues, perhaps these cells can migrate from blood into neural tissue following chemo-attraction si gnals, as demonstrated for T-cells (CD4, CD8) (Alexianu et al., 2001) and monocyt e/macrophages (CD14, CD68) (Henkel et al., 2004). Also, mucosal immune responses could be involved in ALS pathogenesis and that the redistribution of white cells from the blood may be mediated by mucosal chemo-kines (Belyakov and Berzofsky, 2004). However, the significance of mucosal immunity in the pathogenesis of ALS is likely linked to the pos sibility of viral in fection. The role of basophils is not completely understood. Bas ophils circulate for a few hours in blood and migrate into tissues where they may reside for several weeks as mast cells. Basophilia commonly indicates the presence of myelopro liferative disease as opposed to leukemoid reaction but can be a consequence of alle rgy, hypersensitivity reaction, metabolic or endocrine disorders (Bochner and Schleime r, 2001), chronic basophilic leukemia, or Hodgkins disease (Pardanani et al., 2003). Basophils comple te their differentiation in bone marrow and appear in the peripheral bl ood as mature basophils. Recently, these

PAGE 257

241 Appendix A: (Continued) blood cells have been shown to be important regulators of humoral immune responses by their influence on B cell function (Mack et al., 2005) and they can be recruited to sites of inflammation (Wedemeyer et al., 2000). Obse rved basophilia in G 93A mice may reflect the inflammatory process in the spinal co rd, involving mast cells, macrophages and T cells, as recently described in ALS patient s (Graves et al., 2004). The link between basophilia and lymphopenia observed in G93A mice at end-stage disease needs further investigation, and we will initially focu s upon possible alterations in bone marrow function. Interestingly, red blood cell parameters in ALS mice did not statistically differ from both control groups of animals. Only one hematological parameter has been previously noted in ALS patients, an incr eased number of circulating neutrophils correlating with hypermetabolism (Desport et al., 2001). The reason for the association between elevated neutrophil levels and hype rmetabolism is still unknown. However, our data on neutrophils contained in periphera l blood of G93A mice did not support this neutrophil observation in patients. Additionall y, the platelet numbers were increased in G93A mice without statistical significance. However, these results have coincident with recently published data (Kiktenko et al., 2005), that platelet activation was found in ALS patients. Platelet activation was more pronoun ced in ALS than in multiple sclerosis (Kiktenko et al., 2005). It is clear that the pr esence of white blood cells in the peripheral blood has been significantly reduced in G 93A mice at end stage disease, but the mechanism underlying this decline is unclear. Since white blood cells ultimately originate from bone marrow, it is possible to suggest that production of these cells has been impaired. However, we cannot exclude poten tial involvement of secondary lymphoid organs such as the spleen or lymph nodes in this pathology. Presen tly, we are performing phenotypic characterization of lymphocytes in peripheral blood, primary and secondary lymphoid organs and degenerating areas of th e brain and spinal cord in G93A mice modeling ALS. Correlation of lymphocyte dens ities among these locations is currently under investigation. It will also be important to determine if reductions in leukocytes, especially lymphocytes, in pe ripheral blood correlate with disease progression. Our other

PAGE 258

242 Appendix A: (Continued) finding was spontaneous aRFL in G93A mice at the terminal stage of disease. Rosetteformation has been previously noted in so me pathological conditi ons. The ability of lymphocytes to bind with autologous erythrocyt es has been observed in patients with autoimmune thyroiditis and cancer (Endo et al ., 1984; Tamura et al ., 1984; Ichikawa et al., 1983). Subpopulations of T-lymphocytes (C D4) forming aRFL in peripheral blood from patients with thyroid diseases (E ndo et al., 1984), lupus erythematous (SLE) (Ichikawa et al., 1983) and non-Hodgkins lym phomas (Tamura et al., 1984) has been reported. Finally, aRFL evaluation has been pr oposed as a criterion of autoimmune pathology. However, the usual protocol fo r aRFL evaluation includes the mitogeninduced activation of lymphocytes in vitro by pokeweed and phytogemagglutinin mitogens for rosette formation (Ichikawa et al., 1983). Our results, showing aRFL in ALS mice without lymphocyte stimulation by mitoge ns or other activators, may indicate involvement of an autoimmune mechanism in the etiopathology of ALS. As is well known, autologous rosette incidence is also abnormally high in nude (congenitally athymic) mice. The high numbers of aR FL found after adult thymectomy can be normalized by injection of thymic hormones, suggesting that aRFL might be associated with immature T-cell precursors (Cha rreire and Bach, 1975). The phenomenon of spontaneous aRFL noted in G93A mice suggest s that early precursors from thymus or bone marrow may be involved in forming aR FL. However, phenotypical analysis of lymphocyte maturation should be performed to confirm this involvement and a future study will include this analysis. We hypothesize that the lymphopoietic system is involved in ALS and that impairment of this system critically influences immune system function. This is the first evidence that peripheral blood of transgenic G93A mice contains autologous rosette-forming lymphocytes Further investigation of autoimmunity and pathologically significant antigens in ALS is warranted.

PAGE 259

243 Appendix A: (Continued) Acknowledgements: This work was supported by NIH STTR (Phase I) grant 1R41NS46870-01A1. References 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. Alexianu, M.E., Kozovska, M., Appel, S.H ., 2001. Immune reactivity in a mouse model of familial ALS correlates with disease progression. Neurology 57, 1282 1289. Belyakov, I.M., Berzofsky, J.A., 2004. Immunobio logy of mucosal HIV infection and the basis for development of a new generation of mucosal AIDS vaccines. Immunity 20, 247. Bochner, B.S., Schleimer, R.P., 2001. Mast ce lls, basophils, and eo sinophils: distinct but overlapping pathways for recr uitment. Immunol. Rev. 179, 5 O.I. Kuzmenok et al. / Journal of Neuroimmunology 172 (2006) 132 135 Brown, A., Febiger, L., 1993. Hematology: Principles and Procedures, 6th edR Lea and Febiger, Philadelphia, p. 101. Charreire, J., Bach, J.F., 1975. Binding of auto logous erythrocytes to immature T-cells. Proc. Natl. Acad. Sci. U. S. A. 72, 3201. Consilvio, C., Vincent, A.M., Feldman, E.L., 2004. Neuroinf lammation, COX-2, and ALSa dual role? Exp. Neurol. 187, 1 10. Desport, J.C., Preux, P.M., Magy, L., Boirie, Y. Vallat, J.M., Beaufrere, B., Couratier, P., 2001. Factors correlated with hypermetabo lism in patients with amyotrophic lateral sclerosis. Am. J. Clin. Nutr. 74, 328. Endo, Y., Aratake, Y., Kotani, T., Kuribayash i, T., Tamura, K., Ohtaki, S., 1984. Increase in autorosette-forming lymphocytes in t hyroid diseases. Endocrinol. Jpn. 31, 627 634.

PAGE 260

244 Appendix A: (Continued) Engelhardt, J.I., Appel, S.H., 1990. IgG reactivit y in the spinal cord and motor cortex in amyotrophic lateral sclerosi s. Arch. Neurol. 47, 1210 1216. Graves, M.C., Fiala, M., Dinglasan, L.A., Liu, N.Q., Sayre, J., Chiappelli, F., van Kooten, C., Vinters, H.V., 2004. Inflammation in amyotrophic lateral sclerosis spinal cord and brain is mediated by activated m acrophages, mast cells and T cells. Amyotroph. Lateral Scler. Other Mot. Neuron Disord. 5, 213 219. Gurney, M.E., Pu, H., Chiu, A.Y., Dal Canto, M.C., Polchow, C.Y., Henkel, J.S., Engelhardt, J.I., Siklos, L., Simpson, E.P., Kim, S.H., Pan, T., Goodman, J.C., Sidduque, T., Beers, D.R., Appel, S.H., 2004. Presence of dendritic cells, MCP-1, and activated microglia/macrophages in amyotrophi c lateral sclerosis sp inal cord tissue. Ann. Neurol. 55, 221. Ichikawa, Y., Lavastida, M.T., Gonzalez, E.B., Daniels, J.C., 1983. Further characterization of mitogen-i nduced autorosette-forming ce lls: correlati on with T-cell subsets and with lymphocyte proliferative res ponses to mitogens. Cell. Immunol. 76, 351. Khondkarian, O.A., Gannushkina, I.V., Zavalishin, I.A., Zhirnova, I.G., Nevskaia, O.M., 1985. Immunologic abnormalities in amyotrophic late ral sclerosis. Zh. Nevrol. Psikhiatr. Im. S. S. Korsakova. 85, 195R (Russian). Kiktenko, A., Zlobina, G., Brusov, O., Zakha rova, M., 2005. Structure of peripheral blood platelets surface in patients with amyotrophic lateral sclerosis and multiple sclerosis. Zh. Nevrol. Psikhiatr. Im S. S. Korsakova. 105, 40R (Russian). Kobayashi, H., Gleich, G.J., Butterfield, J.H., Kita, H., 2002. Human eosinophils produce neurotrophins and secrete nerve growth factor on immunologic stimuli. Blood 99, 2214 2220. Mack, M., Schneider, M.A., Moll, C., Cihak, J., Bruhl, H., Ellwart, J.W., Hogarth, M.P., Stangassinger, M., Schlondorff, D., 2005. Identification of antigen-c apturing cells as basophils. J. Immunol. 174, 735 741. McGeer, P., McGeer, E., 2002. Inflammatory pr ocesses in amyotrophic lateral sclerosis. Muscle Nerve 26, 459 470.

PAGE 261

245 Appendix A: (Continued) Mohamed, H.A., Mosier, D.R., Zou, L.L., Sikl os, L., Alexianu, M.E., Engelhardt, J.I., Beers, D.R., Le, W.D., Appel, S.H., 2002. ImmunoglobulinFc gamma receptor promotes immunoglobulin uptake, immunoglobulinmediated calcium increase, and neurotransmitter release in motor neurons. J. Neurosci. Res. 69, 110 116. Mulder, D.W., Kurland, L.T., Offord, K.P., Beard, C.M., 1986. Familial adult motor neuron disease: amyotrophic late ral sclerosis. Neurology 36, 511 Nyland, H., Naess, A., 1978. T lymphocytes in the peripheral blood from patients with neurological diseases. Ac ta Neurol. Scand. 58, 272 279. Pardanani, A.D., Morice, W.G., Hoyer, J.D., Tefferi, A., 20 03. Chronic basophilic leukemia: a distinct clinico-pathologi c entity. Eur. J. Haematol. 71, 18 22 Provinciali, L., Laurenzi, M.A., Vesprini, L ., Giovagnoli, A.R., Bartocci, C., Montroni, M., Bagnarelli, P., Clementi, M., Varaldo, P. E., 1988. Immunity assessment in the early state of amyotrophic lateral sc lerosis: a study of virus antibodies and lymphocyte subsets. Acta Neurol. Scand. 78, 449 454. Tamura, K., Aratake, Y., Endo, Y., Ohtaki, S ., 1984. Clinical signifi cance of autorosetteforming cells in T-cell lymphoma. Leuk. Res. 8, 275 283. Wedemeyer, J., Tsai, M., Galli, S.J., 2000. Roles of mast cells and basophils in innate and acquired immunity. Curr. Opin. Immunol. 12, 624 631. Westall, F.C., Rubin, R., Nieder, J., Jablecki, C., 1983. Low percentage of T mu cells in amyotrophic lateral sclerosis. Immunol. Lett. 7, 139 140.

PAGE 262

Appendix A: (Continued) Table 4. Complete blood cell count and white blood cell differential in SOD1 G93A mice as compared to control mice 246

PAGE 263

Appendix A: (Continued) Figure 46. Morphological analysis of periphe ral blood cells (Giemsa staining). Spontaneous autorosette-formations were f ound in G93A mice (A, B, arrows) at end stage of disease. A squeezed shape of er ythrocytes in rosett es was noted. Normal morphological characteristics of peripheral bl ood cells were noted in C57BL/6 (C) and BL6/SJL (hTg) control animals (D, E). 1 autorosette forming by lymphocytes; 2 lymphocyte, 3platelet, 4 reticulocyte; 5monocyte; 6segmented neutrophil; 7 basophil; 8erythrocyte. Scale bar in AE is 25 Am. 247

PAGE 264

Appendix BArthritis-Induced Hematopo iesis Repairs Pathology and Memory in Alzheimers MiceSupplementary Figures 248

PAGE 265

249 Appendix B: (Continued) Supplementary Figure 1. Significant variation of amyloid plaque load between mice. Animals (PS/APP, all 8.8 -9.6 months, number ed sequentially according to date of birth, 25-35 g, both genders) were anesth etized with 1-2% isoflurane shaved and scrubbed with 10% Betadine solution at the site of incisi on, and placed into a stereotaxic frame (Kopf Instruments, Tujunga, Ca.). A small (3 cm ) incision was made, exposing the skull, and curved Strabysmus surgical scissors were us ed to form a subcutaneous pocket along the animals back into which 2 osmotic minipu mps (Alzet model 1004, average flow rate of 0.12 L/hour, Durect Corp., Cupertino, CA) were inserted. Two holes were drilled into the skull (from Bregma -0.1 mm anterior-posterior, +/0.9 mm medial-lateral), and 30 gauge catheters were insert ed at a depth of 3.0 mm, co rresponding to the lateral ventricles. Leading from the Alzet pump wa s a proprietary catheter system (patent pending PCT/US08/73974) with the delivery tips fashioned to the contours of the skull rather than the commercially-available pede stal cannula. The cannul ae are affixed to the skull using Locktite 454 adhesi ve (Plastics One, Roanoke, VA) and secured with 1 cm diameter nitrile, followed by silk sutures to close the scalp. After 2 weeks of bilateral intracerebroventricular infusions of M-CSF, mice were perfus ed, brain tissues were fixed in 10% neutral buffered formalin, and cr yosectioned at 14 m. Standard fluorescent immunohistochemistry used 6E10/Alexa 488 a nd Hoechst nuclear st ain (blue). Bright green spots indicate amyloid plaques. Pi ctures taken at 5X. Mouse numbers 148,160,171, and 211 received M-CSF and mice 164, 170, 176, and 177 received aCSF.

PAGE 266

Appendix B: (Continued) 250

PAGE 267

Appendix B: (Continued) 251

PAGE 268

252 Appendix B: (Continued) Supplementary Figure 2. Intrahippocampal injection of M-CSF (left) and aCSF (right). ( A ) The image is a montage of ~35 5X pictures and is representative of the effects seen from anterior hippocampus to posterior in all 4 M-CSF-injected mice. ( B ) This photo shows enlargement of the M-CSF-injected le ft hemisphere, as seen following saline perfusion. Note the small bump at the site of injection (arrow). ( C ) Image shows cyst or tumor-like growth formed in the needle track at the site of M-CSF injection. Cryosectioned at 14 m and stained with 6E 10/Alexa 488 and Hoechst. Picture taken at 20X.

PAGE 269

Appendix B: (Continued) 253

PAGE 270

254 Appendix B: (Continued) Supplementary Figure 3. Intrahippocampal injection of G-CSF (left) and aCSF (right) Amyloid plaques indicated as white spots. Visual observation of amyloid plaques show a modest reduction of plaque in the left G-CSF-injected he misphere. Cryosectioned at 14 m and stained with 6E10/Alexa546 and Hoechst nuclear stain. Images are montages of ~35 5X pictures each. Sections numbered 1 through 6 and correspond with anterior to posterior.

PAGE 271

Appendix B: (Continued) 255

PAGE 272

Appendix B: (Continued) 256

PAGE 273

257 Appendix B: (Continued) Supplementary Figure 4. Intrahippocampal injection of GM-CSF(left) and aCSF(right). Representative sections of each mouse proxima l to injection site. Ti ssue sections stained with MabTech -A /Alexa 488. White spots indicate amyloid plaque immunolabelling. Images are montages of about 145 pictures taken at 10X. Figures 4A-C are from 14 m frozen sections, and 4D is from a 5 m paraffin-embedded section.

PAGE 274

Appendix B: (Continued) 258

PAGE 275

Appendix B: (Continued) 259

PAGE 276

Appendix B: (Continued) 260

PAGE 277

Appendix B: (Continued) 261

PAGE 278

Appendix B: (Continued) 262

PAGE 279

Appendix B: (Continued) 263

PAGE 280

264 Appendix B: (Continued) Supplementary Figure 5. Quantification of reduced amyloid deposition in GM CSF injected left hemispheres versus aCSF injected contralateral right hemispheres. There were 5 sections per mouse quantified. Each montaged section contained over 140 10X pictures and of these, 15 25 pictures pe r hemisphere were selected to quantify as described in Supplementary Figure 6 online. Al l pictures per section were taken at the same exposure on a Zeiss Imager.Z1 fluorescence microscope with a Zeiss Axiocam Mrm camera (Oberkochen, Germany ) using Axiovision 4.7 software. Each figure shows total or average values from the 5 sections /mouse with significance per individual mouse and overall. Error bars are SEM : (a b ) plaque areas (c d ) perimeter values ( e ) average feret diameters ( f ) average integrated densities

PAGE 281

Appendix B: (Continued) 265

PAGE 282

266 Appendix B: (Continued) Supplementary Figure 7. Behavioral Tasks Radial Arm Water Maze For the RAWM task of spatial working memory, an aluminum insert was placed into a 100cm circular pool to create 6 radially-dis tributed swim arms emanating from a central circular swim area. An assortment of 2-D and 3-D visual cues surrounded the pool. The number of errors pr ior to locating which one of the 6 swim arms contained a submerged escape platform (9cm diameter) was determined for 5 trials/day. There was a 30-min time delay between the 4th trial (T4; fi nal acquisition trial) and 5th trial (T5; memory rete ntion trial). The platform lo cation was changed daily to a different arm, with different start arms fo r each of the 5 trials semi-randomly selected from the remaining 5 swim arms. During each trial (60 s maximum), the mouse was returned to that trials start arm upon swi mming into an incorrect arm and the latency time required to locate the submerged plat form was recorded. If the mouse did not find the platform within a 60-s trial, it wa s guided to the platform for a 30-s stay. The numbers of errors and escape latency during tria ls 4 and 5 are both considered indices of working memory and are temporally similar to standard registration/recall testing of specific items used clinically in evaluating AD patients. Cognitive Interference Task This task was designed to mimic, measure-for-measure, a cognitive interference task recently utilized clinically to discri minate between normal aged, MCI, and AD patients. The task invol ves two RAWM set-ups in two different rooms, with two sets of visual cues different from those utilized in standard RAWM testing. The task requires animals to remember a set of visual cues (in RAWM-A), so that following interference with a different set of cues (in RAWM-B), the initial set of cues can be recalled to successfully solve the RA WM task. Five behavioral measures were examined: A1-A3 (Composite three-trial recall score from first 3 trials performed in RAWM-A), B (proactive interfer ence measure attained from a single trial in RAWM-B), A4 (retroactive interference measure attained during a single trial in RAWM-A), and A5 (delayed-recall measure attained from a single trial in RAWM-A following a 20-min

PAGE 283

267 Appendix B: (Continued) delay between A4 and A5). As with the sta ndard RAWM task, this interference task involves the platform location being change d daily to a different arm for both RAWM set-ups. For A1 and B trials, the animal is initially allowed one minute to find the platform on their own before being guided to the platform. Then the actual trial is performed in each case. As with the standard RAWM task, animals were given 60s to find the escape platform per trial, with the number of errors and es cape latency recorded per trial.

PAGE 284

ABOUT THE AUTHOR Steven Prescott Bennett received his Bachelors degree in Bi ology with a strong minor in Chemistry from Eckerd College in 1997. After 2 years working in a DNA sequencing laboratory, he began his research career at the Roskamp research institute. In the fall of 2001 he joined the Medical Science Ph.D. program at the University of South Florida in the laboratory of Dr. Huntingt on Potter where he specialized in drug and gene delivery to mouse models of amyotrophic lateral sclerosi s (ALS) and Alzheimers disease (AD). His formal appointment was in the department of Molecular Medicine within college of Medicine. He successfully defended his doc toral dissertation on October 14, 2009 at the University of South Florida.