TARGETED LENTIVIRAL-MEDIATED DELIVERY OF PROGRANULIN cDNA IN A

GENETIC MODEL OF AMYOTROPHIC LATERAL SCLEROSIS

by

Pierre Zwiegers

B.Sc., The University of British Columbia, 2009

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

in

THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES

(Experimental Medicine)

THE UNIVERSITY OF BRITISH COLUMBIA

(Vancouver)

November 2015

© Pierre Zwiegers, 2015

ii

Abstract

Amyotrophic lateral sclerosis is a fatally progressive neurodegenerative disease characterized by

the loss of motor neurons in both the brain and spinal cord. Neuronal cell death leads to a lack of

muscle innervation which eventually gives rise to paralysis. Patients typically survive less than

five years following the initial onset; with respiratory complications ultimately causal in death.

The majority of ALS cases are of unknown etiology, with around 5% linked to aberrant genetic

mutations. The most well characterized mutations account for approximately 1-2% of all ALS

cases, and are linked to the genetic locus for superoxide dismutase-1 (SOD1).

The severity of disease and the concomitant lack of effective therapeutic options necessitate

significant research efforts in search of viable treatment options that would significantly impede

and/or reverse the ultimate loss of motor neurons. Progranulin (PGRN) may be one such an

option since the protein is a secreted growth factor that has demonstrated neuroprotective

outcomes in models of Alzheimer’s- and Parkinson’s disease and may generally be important in

the long term survival of neurons.

Employing a lentiviral-mediated delivery mechanism at a stage prior to disease onset, we set out

to assess the potential neurotrophic properties of exogenously delivered PGRN cDNA when

targeted to neuronal subsets within the lumbar spinal cord of mice expressing mutant SOD1. The

resultant outcomes at both the behavioural and neuropathological levels did not demonstrate any

significant protective effect from the lentiviral delivery of exogenous PGRN. PGRN-treated

cohorts did not exhibit an increase in overall survival, a decrease in gliosis, or an increase in

iii

neuronal viability when compared to the GFP-lentiviral-injected control groups. This study

suggests that PGRN delivery at an early stage of ALS neurodegeneration preceding the

phenotypic expression of disease may not be a viable therapeutic option in ameliorating the ALS

degenerative cascade. Technical caveats to this interpretation are discussed. One of these

includes the unexpected reduction in the copy number of the mSOD1 gene in transgenic mice,

which ultimately presenting with a protracted progression of the disease. The present study thus

additionally underscores some of the challenges faced in pre-clinical therapeutic development

using murine ALS models.

iv

Preface

All of the work presented in relation to this thesis was conducted in the laboratory of the Neural

Dynamics Research Group at the VGH Research Pavilion, under the direct supervision of Dr.

Christopher A. Shaw. Dr. Shaw conceived of the original project and provided the space and

resources to conduct the experiment. The author of this thesis was responsible for all

experimental manipulation including- but not limited to- generating the animal colony under

study, the transfection of subjects, temporal behavioural monitoring, as well as tissue collection,

histological processing, and analysis. All of the methods and viral constructs used in this project

were approved by the University of British Columbia’s Animal Care and Biosafety Committees,

respectively (Breeding certificates: A07-0018; A10-0383; Experimental animal handling

certificates: A07-0213, A11-0342; Biohazard approval certificate: B10-0094).

Publications arising from work presented in this thesis:

(1) Zwiegers P, Lee G, Shaw C A: Reduction in hSOD1 copy number significantly impacts ALS

phenotype presentation in G37R (line 29) mice: implications for the assessment of putative

therapeutic agents. J Negat Results Biomed 2014, 13:14. The author generated and maintained

the colony of animals from the commercial breeders, collected experimental data from

behavioural assays, reviewed the literature, performed all statistical analyses, generated the

graphics, and produced the initial draft of the manuscript. GL had previously generated the

colony of collaborator-derived animals for an independent investigation, and provided the

genotyping and lifespan data for comparative analysis. Additionally, GL provided critical

revisions that were integral to the manuscript. CAS conceptualized the original investigation as

well as revised the manuscript.

v

(2) Zwiegers P, Shaw CA: Disparity of outcomes: the limits of modeling amyotrophic lateral

sclerosis in murine models and translating results clinically. J Controv Biomed Res 2015, 1:4–

22. The author performed the literature review, produced the manuscript, and generated the

figures. CAS contributed editorial changes.

Chapter-specific items:

Chapter 1. Figure 1.3, E1, and Figures 1.1, 1.2 and Tables 1.1, 1.2 have been previously

published (Zwiegers et al. 2014; Zwiegers & Shaw 2015, respectively), alongside the derivation

of relevant sections of text and figure captions. As listed above, I am the principal author on both

manuscripts, and have secured permission from all co-authors for inclusion in this thesis.

Chapter 3. Figures 3.8 and 3.10 as well as the associated text and ideas have been previously

published as outlined above (Zwiegers et al. 2014). The work is included here with permission

from the associated co-authors.

Chapter 4. Sections of chapter 4 (including tables 4.1, 4.2, 4.3, and figure 4.1) have been

previously published (Zwiegers et al. 2014; Zwiegers & Shaw 2015). Permission for use has

been granted by the co-authors.

Appendix C. Figure C.1, Panels B and C were produced and provided by Drs. CA Shaw, Denis

G. Kay, J Van Kampen and G Lee. Figure C.2 was produced in collaboration with D. Kwok as

part of a preliminary investigation; D. Kwok aided in viral delivery, while the author was

responsible for tissue collection and histological processing..

vi

Appendix D. The emGFP and PGRN lentiviral constructs, DNA sequences, and diagrams were

provided by Invitrogen Inc.

Appendix E. Figure E.1 is based on data provided by the Jackson Laboratory and published in

Zwiegers et al, 2014).

Appendix F Previously published as Zwiegers et al, 2015. Permission for use has been granted

by the co-author.

vii

Table of Contents

Abstract...........................................................................................................................................ii

Preface............................................................................................................................................iv

Table of Contents .........................................................................................................................vii

List of Tables ...............................................................................................................................xiv

List of Figures...............................................................................................................................xv

List of Abbreviations .................................................................................................................xvii

Acknowledgements ...................................................................................................................xviii

Dedication .....................................................................................................................................xx

Chapter 1: Introduction ................................................................................................................1

1.1 Overview of Amyotrophic Lateral Sclerosis .................................................................. 1

1.2 Genetic Heterogeneity of Amyotrophic Lateral Sclerosis .............................................. 3

1.3 ALS and SOD1 Defects: Overview and Mechanistic Hypotheses ................................. 6

1.3.1 Mitochondrial Dysfunction: Altered Energy Metabolism, Oxidative Stress, and

Apoptosis ................................................................................................................................ 8

1.3.2 Excitotoxicity ............................................................................................................ 12

1.3.3 Endoplasmic Reticulum Stress and Altered Proteasomal Functioning..................... 13

1.3.4 Non-cell Autonomous Contributions to Motor Neuron Cell Death.......................... 14

1.4 Modeling ALS In Vivo: Transgenic Mice Overexpressing Mutant Human SOD1 ...... 16

1.4.1 Mice Expressing the Glycine to Alanine Substitution at Codon 93 ......................... 17

1.4.2 Mice Expressing the Glycine to Arginine Substitution at Codon 37 ........................ 21

1.5 Caveats Inherent to ALS In Vivo Model Systems ........................................................ 23

viii

1.5.1 Transgene Copy Number Variation .......................................................................... 24

1.5.2 Limited Applicability to Clinical ALS Cases ........................................................... 25

1.6 The Properties and Diverse Biological Functions of Progranulin ................................ 26

1.7 Neurodegenerative Disease and Progranulin ................................................................ 28

1.7.1 Ubiquitin-positive Frontotemporal Lobar Degeneration .......................................... 28

1.7.2 Amyotrophic Lateral Sclerosis.................................................................................. 29

1.8 Previous Pre-clinical Applications of Progranulin in the CNS..................................... 30

1.9 Research Theme, Objectives, and Experimental Hypothesis ....................................... 31

1.9.1 Research Theme ........................................................................................................ 31

1.9.1.1 Objectives.......................................................................................................... 31

1.9.1.2 Hypothesis......................................................................................................... 32

Chapter 2: Materials and Methods ............................................................................................33

2.1 Materials and Methods.................................................................................................. 33

2.1.1 Colony Generation and Animal Husbandry.............................................................. 33

2.1.1.1 Breeding Wild-type and mSOD1 Transgenic Progeny from Commercially-

Obtained Breeders............................................................................................................. 34

2.2 Quantitative Genotyping ............................................................................................... 35

2.2.1 Distinguishing Wild-type from Transgenic mSOD1 Animals.................................. 35

2.2.2 End-stage Confirmation of GFP and PGRN Delivery .............................................. 36

2.3 Vector Design ............................................................................................................... 37

2.4 Lentiviral Administration Targeting Motor Neurons in the Lumbar Spinal Cord:

Bilateral Gastrocnemius Muscle Injections .............................................................................. 38

2.5 Assessing Disease Onset and Progression .................................................................... 39

ix

2.5.1 Body Weight Changes .............................................................................................. 40

2.5.2 Wire Hang ................................................................................................................. 40

2.5.3 Leg Extension ........................................................................................................... 41

2.6 Euthanasia and Organ Harvesting................................................................................. 41

2.7 Tissue Processing and Histological Assessments ......................................................... 42

2.7.1 Tissue Processing and Cryosectioning...................................................................... 42

2.7.2 Cresyl Violet Assay: Staining for Neuronal Nissl Substance ................................... 43

2.7.2.1 Differentiating Between Lumbar Spinal Cord Levels ...................................... 44

2.7.2.2 Neuronal Counts With CellProfilerTM Cell Image Analysis Software ............. 44

2.7.2.3 Neuronal Morphological Analysis .................................................................... 45

2.8 Immunohistological Assays .......................................................................................... 47

2.9 Statistical Analysis ........................................................................................................ 49

Chapter 3: Exploiting Lentiviral Delivery Mechanisms: the Effects of Targeting the

Lumbar Spinal Cord For Progranulin Overexpression ...........................................................50

3.1 Results ........................................................................................................................... 50

3.1.1 Overall Phenotypic Outcomes .................................................................................. 55

3.1.1.1 Body Weight of Mice........................................................................................ 55

3.1.1.2 Temporal Changes in Measured Behavioural Indices ...................................... 58

3.1.1.3 Latency to Fall from an Elevated Grid.............................................................. 58

3.1.1.4 Leg Extension Reflex Score.............................................................................. 61

3.1.1.5 Overall Lifespan and Disease Onset Comparisons Between mSOD1 Groups . 63

3.1.2 Histological Analyses ............................................................................................... 66

3.1.2.1 Neuronal Assessment throughout the Lumbar Spinal Cord ............................. 66

x

3.1.2.2 Morphological Characteristics of Neurons within the L3-L5 Lumbar Spinal

Cord Segment.................................................................................................................... 68

3.1.2.3 Probing for Viable Cholinergic Neurons within the L3-L5 Lumbar Spinal Cord

Segment…......................................................................................................................... 73

3.1.2.4 End-stage Glial Cell Changes within the L3-L5 Lumbar Spinal Cord

Segment…......................................................................................................................... 74

3.1.2.4.1 Astrogliosis.................................................................................................. 74

3.1.2.4.2 Microgliosis................................................................................................. 75

3.1.3 Transgenic Progeny Derived from Commercial Breeders Showed a Reduction in

Mutant Human SOD1 Copy Number and a Concomitant Increase in Overall Lifespan ...... 77

3.1.3.1 ΔCT Differences in Progeny Derived from Commercial vs. Collaborator

Breeders... ......................................................................................................................... 77

3.1.3.2 ΔCT Comparisons Stratified According to Sex and Treatment Group ............. 79

3.1.3.3 Lifespan is Negatively Correlated with the Degree of mSOD1 Transgene

Presence.. .......................................................................................................................... 80

3.1.4 End-stage Probing for Lentiviral Transduction Efficiency and Expression ............. 82

3.2 Discussion of Experimental Findings ........................................................................... 82

3.2.1 Decreased mSOD1 Copy Number of the G37R Locus Delays Disease Onset and

Extends Lifespan in Transgenic Animals ............................................................................. 82

3.2.2 Early-Stage Lentiviral-mediated Progranulin Delivery Does Not Attenuate the Onset

or Progression of an ALS-like Phenotype in mSOD1 Transgenic Mice .............................. 85

xi

3.2.2.1 Progranulin Transgene Delivery Does Not Delay Disease Onset, Alter

Behavioural Phenotypic Abnormalities, or Extend Survival in mSOD1 Animals when

Targeted at an Early Stage of the Disease Cascade. ......................................................... 86

3.2.2.2 Early Progranulin Delivery Targeting the Lumbar Spinal Cord Does Not

Mitigate Neuronal Degeneration or Neuroinflammatory Processes at the Diseased End-

stage……. ......................................................................................................................... 88

3.3 Conclusion .................................................................................................................... 93

Chapter 4: General Discussion and Future Studies..................................................................94

4.1 General Discussion ....................................................................................................... 94

4.1.1 Potential Sources of Error and Limitations in Experimental Design........................ 95

4.1.1.1 Reduced mSOD1 Copy Numbers in Derived Transgenic G37R Progeny ....... 95

4.1.1.2 High Variability in Behavioral and Neuropathological Outcomes ................... 95

4.1.1.3 Lack of Confirmation of PGRN/GFP Transgene Integration at End-stage ...... 97

4.1.1.4 Manually Organizing Lumbar Spinal Tissues into Discrete Segments ............ 98

4.2 Future Studies ............................................................................................................... 99

4.2.1 Progranulin as an ALS Therapeutic .......................................................................... 99

4.2.1.1 Targeting Multiple Spinal Cord Regions at the Late Stage of Disease

Pathogenesis...................................................................................................................... 99

4.2.1.2 Neuronal Specific Overexpression of Progranulin in a Murine Model of

ALS……. ........................................................................................................................ 100

4.2.1.3 Investigate the Potential for Progranulin as an Epidermal Biomarker in ALS

Progression...................................................................................................................... 100

4.2.2 Considerations for the Clinical Translation of ALS Research Efforts ................... 102

xii

4.2.2.1 Limitations of Clinical ALS Trials and Modeling Disease in Mouse Models

Expressing Mutant Human SOD1 .................................................................................. 102

4.2.2.2 Models Representative of Clinical ALS ......................................................... 104

4.2.2.3 Impediments to Replicability between Pre-clinical Studies ........................... 106

4.2.2.4 Species Differences and Drug Metabolism..................................................... 108

4.2.3 Concluding Remarks............................................................................................... 109

References ...................................................................................................................................123

Appendices ..................................................................................................................................139

Appendix A Chapter 1 Supplementary Materials ................................................................... 139

A.1 Neuropathological Hallmarks of Amyotrophic Lateral Sclerosis........................... 139

A.1.1 sALS Neuropathology ................................................................................ 139

A.1.2 fALS Neuropathology................................................................................. 140

A.2 Clinical Phenotypic Heterogeneity of Amyotrophic Lateral Sclerosis ................... 141

A.2.1 Extent of Upper and Lower Motor Neuron Involvement ........................... 142

A.2.2 Initial Site of Onset ..................................................................................... 143

A.2.3 Rate of Disease Progression........................................................................ 144

A.2.4 Degree of Cognitive Impairment ................................................................ 145

A.3 Binding Partners for Progranulin ............................................................................ 146

A.3.1 Sortilin: Trafficking Extracellular Progranulin into the Endolysosome ..... 148

A.3.2 Progranulin Acts as a Co-factor for Endolysosomal TLR9-mediated

Responses.................................................................................................................... 149

A.3.3 Progranulin: an Antagonist of the TNF Receptor-mediated Signaling

Cascade… ................................................................................................................... 149

xiii

Appendix B List of Chemical Reagents.................................................................................. 151

Appendix C Preliminary Findings .......................................................................................... 153

C.1 Neuroprotective Effects of Progranulin .................................................................. 153

C.2 Targeted GFP-expression in Lumbar Cord Motor Neurons Following Lentiviral

Transduction........................................................................................................................ 154

Appendix D Viral Constructs.................................................................................................. 155

D.1 Progranulin Lentiviral Construct and Gene Sequence ............................................ 155

D.2 GFP Lentiviral Construct and Gene Sequence ....................................................... 159

Appendix E Relative hSOD1 Copy Number in Male Breeders.............................................. 163

Appendix F Chapter 4 Supplementary Materials.................................................................... 164

F.1 Future Considerations for Modeling Disease and Testing Clinically-relevant ALS

Therapeutics ........................................................................................................................ 164

F.2 Clinically-relevant Biomarkers: Assessing Disease Progression and Therapeutic

Efficacy ............................................................................................................................... 164

F.3 Development of Alternate Models to More Effectively Mirror the Diverse

Etiological Factors that Underlie ALS Pathogenesis .......................................................... 167

F.4 Clinical Trials Should be Designed and Stratified so that the Therapeutic Effect of

Any Agent is Tested in a Patient Subset that is Homologous for Set Criteria.................... 169

xiv

List of Tables

Table 1.1. Select phenotypic correlates of disease progression in the G93A murine model is

dependent on transgene integration within the genome................................................................ 18

Table 1.2. Select phenotypic correlates of disease progression in the G37R murine model is

dependent on transgene integration within the genome................................................................ 22

Table 2.1. Experimental animals stratified by sex, genotype, treatment and group ..................... 35

Table 3.1 Summary of behavioral and neuropathological outcomes in male cohorts .................. 52

Table 3.2 Summary of behavioural outcomes in female cohorts.................................................. 54

Table 3.3 Mean diameter of healthy and atrophying motor neurons across genotype and

treatment groups............................................................................................................................ 72

Table 4.1 Therapeutic efficacy outcomes in pre-clinical mSOD1 models. ................................ 110

Table 4.2. Therapeutic drug effect on lifespan of patients in select clinical trials. .................... 112

Table 4.3. Therapeutic drug effect on functional outcomes of patients in select clinical trials. 117

xv

List of Figures

Figure 1.1. Graphical representation approximating the frequencies of genomic causes of ALS in

the affected patient population. ....................................................................................................... 4

Figure 1.2. Mutations within the SOD1 locus result in amino acid substitutions throughout the

polypeptide...................................................................................................................................... 5

Figure 1.3. Select mutant SOD1 pathomechamisms contributing to neuronal degeneration. ...... 10

Figure 1.4. Effect of mSOD1 transgene levels on the expression of an ALS phenotype and

expected therapeutic outcomes based on disease severity. ........................................................... 25

Figure 2.1. Graphical overview of the experimental timeline ...................................................... 34

Figure 2.2. Schematic of lentiviral administration into gastrocnemius muscles and the retrograde

transport of transgene constructs into the spinal cord................................................................... 39

Figure 2.3. Image processing and CellProfiler analysis of cresyl-violet stained spinal cord

sections.......................................................................................................................................... 46

Figure 3.1 Temporal animal weight changes in transgenic mSOD1 and wild-type mice. ........... 58

Figure 3.2 Latency to fall from an elevated grid in mSOD1 and wild-type mice over time ........ 61

Figure 3.3. Leg extension reflex scores over time in both mSOD1 and wild-type animals ......... 63

Figure 3.4 Comparisons between disease onset and survival in transgenic mSOD1 animals ...... 65

Figure 3.5 Neuronal viability across regions of the lumbar spinal cord of male wild-type and

transgenic mSOD1 animals........................................................................................................... 68

Figure 3.6 Characterization of neuronal morphology between treatment and genotype groups .. 71

Figure 3.7 Immunohistological assays within the ventral horn of tissues from the L3-L5 region

assessing phenotype severity and the effect of transgene delivery ............................................... 77

xvi

Figure 3.8 Comparisons of quantitative genotyping for the mutant hSOD1 locus in two

independently-derived colonies. ................................................................................................... 79

Figure 3.9 Comparisons of quantitative genotyping for the mutant hSOD1 locus across sex and

treatment group. ............................................................................................................................ 80

Figure 3.10 Effect of mSOD1 transgene dosage on lifespan of transgenic animals in two

independently-derived colonies. ................................................................................................... 82

Figure 4.1 Literature survey for details regarding the genomic SOD1 G37R locus in line 29

animals. ....................................................................................................................................... 108

Figure A.1 Putative extracellular binding partners mediate both cellular uptake and varied

intracellular effects...................................................................................................................... 147

Figure C.1 Preliminary findings exploring the neuroprotective properties of progranulin ........ 153

Figure C.2 Temporal expression of retrogradely transported GFP cDNA in lumbar spinal cord

(LSc) tissues following gastrocnemius muscle viral transfection............................................... 154

Figure E.1 Cycle threshold values of original commercial male breeders utilized in seeding our

transgenic G37R (line 29) colony as assayed by Jackson Laboratory ........................................ 163

Figure F.1 A ‘model-up’ approach in designing clinically effective ALS therapeutics. ............ 169

xvii

List of Abbreviations

ALS: amyotrophic lateral sclerosis

ChAT: choline acetyl transferase

ERAD: endoplamic reticulum associated degradation

FS: functional score

FTLD: frontotemporal lobar degeneration

G37R: mutant SOD1 with a Glycine to Arginine substitution at codon 37

G93A: mutant SOD1 with a Glycine to Alanine substitution at codon 93

GFAP: glial acidic fibrillary acidic protein

GFP: green fluorescent protein

HI: Hyaline inclusion

Iba-1: ionized calcium-binding adapter molecule-1

LBHI: Lewy-body like hyaline inclusion

mSOD1: mutant SOD1; refers to mice expressing the mutant form of the human enzyme

ODN: oligodeoxynucleotides

PGRN: progranulin

RHI: Round hyaline inclusion

SOD1: Cu/Zn superoxide dismutase 1

TDP-43: transactive response DNA binding Protein, 43 kDa

TNFα: tumour necrosis factor alpha

UPR: unfolded protein response

VH: ventral horn

SLI: Skein-like inclusion

xviii

Acknowledgements

I am eternally grateful to my supervisory committee Drs. Christopher A. Shaw, Shernaz X.

Bamji, and Charles Krieger for their invaluable input throughout the course of my time in the

Experimental Medicine graduate program at UBC. To Dr. Shaw specifically, I wish to extend my

heartfelt gratitude for giving me the opportunity so many years ago as an undergraduate student

to embark on a journey that as ultimately culminated in the genesis of this manuscript. Thank

you for the learning experience, all of the support, resources, and motivational chats when I was

frustrated with enigmatic data.

To members of the Neural Dynamics Research Group both past and present: thank-you! You

have all been such a great inspiration to me and have had a hand igniting my love/hate

relationship with the fickle mistress that is scientific research. Drs. Reyniel Cruz-Aguado, Jason

Wilson, Grace Lee, Lucija Tomljenovic, Yongling Li, as well as Dominica Kwok, Agripina

Suarez, Rena Tabata, Trisha Kostesky, Darryl Bannon, Mike Petrik, Jessica Holbek, Jess

Morrice, Sneha Seth, Bellars Chan, Dominik Sommerfeld, Brett Hilton, Opeyemi Banjo, Janice

Yoo, Nikolas Kokan and Clara Setiawan. Thank you for all the great times: whether we were

harvesting tissues, running way too many Western blots at the same time, or staying in the lab at

unholy hours, it was a blast. Finally, to the mice who so valiantly gave their lives for science: I

salute you!

xix

To my family, both the Zwiegers and Loewen clans; your support, love, and encouragement

throughout this project have been greatly appreciated. I am also forever indebted to my wife,

Victoria. Your love and support was the impetus for the completion of this work.

To the many friends I have not seen much: thank-you for your patience. See, this is what I was

actually working on whenever I could not commit to an engagement!

Lastly, this work would not have been possible without the generous support of both ALS

Society of Canada and the Luther Allyn Shourds Dean Estate.

xx

Dedication

The work contained herein is firstly dedicated to the courageous patients and their families who

daily take up the fight against this devastating disease. The famous neurologist Jean-Martin

Charcot once addressed his medical students and proclaimed regarding the lack of effective ALS

treatment options in the nineteenth century: “[l]et us keep looking in spite of everything. Let us

keep searching. It is indeed the best method of finding, and perhaps thanks to our efforts, the

verdict we will give [the ALS] patient tomorrow will not be the same we must give […] today”.

With your support, the ALS research community will work tirelessly to understand the

underlying mechanisms so as to rationally design targeted and effective therapeutics. In the near

future, medical science will deem this to be a treatable condition.

Secondly, I wish to dedicate this work to my late father, Pieter Andries Jacobus Zwiegers. Your

time on earth expired far too early, and I can only hope that the path I have chosen would have

made you proud.

1

Chapter 1: Introduction

1.1 Overview of Amyotrophic Lateral Sclerosis

Amyotrophic lateral sclerosis (ALS) is the most prevalent form of motor neuron disease and was

first described by the French neurologist Jean-Martin Charcot in the mid-nineteenth century who

defined the condition based on strict clinicopathological findings (Goetz 2000; Rowland 2001).

The disease is an age-related, multifactorial, and neurodegenerative disorder that primarily

manifests due to the atrophying of motor neurons in the brain and spinal cord that control

voluntary movement. In North America, ALS has an incidence of 2 per 100,000 individuals and

this rate is expected to increase with the aging demographic (Chiò et al. 2013). Due to the

progressive nature of the disease, the overt symptomology worsens temporally and ultimately

results in the denervation of muscles that control respiratory function. As a result of respiratory

complications, patients typically die within 2-4 years following the initial onset (Chiò et al.

2013).

To date, multiple etiological factors have been described; however the vast majority of ALS

cases are due to unknown predisposing factors. Inheritable mutations in oligogenic loci account

for a modest 5% of all ALS cases (termed familial, or “fALS”), but are the primary tools utilized

to investigate disease pathomechanisms and putative therapeutics. The vast majority of cases,

however, are likely due to unknown exogenous environmental factors (and to a lesser degree

spontaneous de novo mutations) that can result in an insult that drives the system towards a

phenotype (termed sporadic, or “sALS”) that is virtually indistinguishable from fALS (Kabashi

et al. 2007; Byrne et al. 2011).

2

Multiple research endeavours have served to illustrate not only the diverse neuropathological and

consequent clinical phenotypic outcomes that serve to distinguish fALS from sALS, but have

shown that ALS can present along a spectrum of disease manifestation (See Appendix A sections

A.1-A.2). As discussed below and further expounded on in Chapter 4 and Appendix A, this

reductionist paradigm of investigating ALS primarily in the context of specific genomic

anomalies does not take into account the potential heterogeneic pathology and/or disease

pathogeneses that ultimately manifests in an ALS-phenotype, nor does it account for any

possible gene-environment interactions. Neuropathologically, familial ALS often shows positive

immunolabelling for misfolded SOD1 epitopes, which is a finding not always recapitulated in

sporadic forms of ALS (Liu et al. 2009; Pokrishevsky et al. 2012; Rotunno & Bosco 2013; Ayers

et al. 2014). Conversely, cytoplasmic TDP-43 positive inclusions is common in sALS, but is a

scarce finding in the mutant SOD1-linked familial form (Mackenzie et al. 2007). As summarised

in Cervenakova et al, there are additional clinical features that critically distinguish fALS from

sporadic forms of the disease (2000). Clinically, fALS presents with an age of onset that is 10

years earlier compared to sALS. The majority of fALS cases show signs of limb onset without

bulbar involvement, and less than 20% of fALS cases show signs of both lower motor neuron

dysfunction in conjunction with upper motor neuron degeneration.

Following more than two decades of research, ALS has been a condition largely intractable to

therapeutic intervention. With more than 50 clinical trials conducted to date, Riluzole—with a

modest 2-3 month increase in lifespan—remains the only FDA-approved therapeutic in the

treatment of disease (Miller et al. 2012; Mitsumoto et al. 2014). Aside from advances in

respiratory support and parenteral nutrition, the clinical prognosis of an ALS patient remains

3

largely unchanged from what it was in Charcot’s day. With the current lack of effective

therapeutics, it is critical to expedite research endeavours that investigate novel approaches in

attenuating and/or preventing the progressive loss of motor neurons and thus abrogate the

accompanying functional deficits.

1.2 Genetic Heterogeneity of Amyotrophic Lateral Sclerosis

As with the wide-ranging clinical and neuropathological phenotypes that are described in

Appendix A , the underlying genetic causes are heterogeneous in nature. Approximately 5% of

ALS cases are due to an inheritable genetic defect. To date, mutations in more than 20 gene loci

have been linked to familial forms of the disease (reviewed in Harms & Baloh, 2013; Marangi &

Traynor, 2014). Common genetic causes of ALS include the recently characterized C9orf72

intronic hexanucleotide repeat expansion (fALS: 30-50%; sALS: ~10%), as well as mutations in

the genetic loci for the Cu/Zn superoxide dismutase 1 enzyme (SOD1; fALS: 10-20%;

sALS:~3%), TAR DNA binding protein (TARDBP; fALS: 0-62%; sALS:~2%), and Fused in

Sarcoma (FUS; fALS:1-13%; sALS: ~2%) RNA-binding protein (Figure 1.1; Harms & Baloh,

2013).

4

Figure 1.1. Graphical representation approximating the frequencies of genomic causes of ALS in the affected patient population.

Frequency of causes are expressed per 100 ALS patients and do not account for sex differences. Mutations in the loci for C9ORF72, SOD1, and TDP-43 account for a large proportion of inheritable cases, however the vast majority of ALS cases are of a sporadic nature. Infographic generated from the upper frequency limit listed in Harms and Baloh (2013).

Focusing exclusively on SOD1: nearly twenty percent of fALS cases are linked to over 180

missense mutations throughout the exogenic sequence encoding the enzyme (Figure 1.2;

Valentine, Doucette, & Zittin Potter, 2005; ALSoD, 2015). The substitution of amino acids at

most of these loci impart on SOD1 a novel toxic-gain-of- function property that predisposes the

system to an ALS phenotype (Ratovitski et al. 1999; Harms & Baloh 2013). This genetic

variation is underscored by phenotypic heterogeneity in that some missense mutations result in a

clinically better (e.g. H46R) or worse (e.g. A4V) prognosis, whereas phenotypic differences (i.e.

age of onset and disease duration) can even be observed in patients with the same SOD1-

missense mutation (Andersen et al. 1997; Kato 2008; Chiò et al. 2011).

5

Figure 1.2. Mutations within the SOD1 locus result in amino acid substitutions throughout the polypeptide.

A) Pymol1-generated graphical visualization of the SOD1 protein depicting unique amino acid substitution sites. B) Primary amino acid sequence identifying sites of more than 150 amino acid substitutions. Sites of the mutant residues depicted are derived from the ALS Online Genetic Database (2015). Although the vast majority of missense mutations impart on the enzyme a toxic property that is associated with neurodegenerative changes, there is a degree of phenotypic heterogeneity associated with the various genotypes that affects the rate of disease progression.

For the vast majority (≥ 90%) of ALS cases, however, the phenotype manifests primarily as a

result of unknown etiological factors. These sporadic (sALS) forms are purported to be clinically

and neuropathologically indistinguishable from the inherited cases (Kabashi et al. 2007; Ravits et

al. 2013). This general similarity, as a result, has served to establish the notion that a common

pathway involving SOD1 mediates neurodegenerative disease processes inherent to both

sporadic and familial forms of ALS. Thus it is thought that if the underlying pathological

mechanistic cascade inherent to SOD1-fALS can be understood, that this knowledge can be

translated to the wider ALS population and used to develop an effective therapeutic agent

1The PyMOL Molecular Graphics System. Schrödinger, LLC

6

1.3 ALS and SOD1 Defects: Overview and Mechanistic Hypotheses

Following the initial description of mutations in the SOD1 locus, work establishing a causal link

between ALS pathobiology and SOD1 have focused on the role it plays in vivo. SOD1 is a

soluble homodimeric metalloenzyme that is located in both the cytosol and mitochondrial

intermembrane space, and plays an integral role in the catalysis of free oxygen radicals into

hydrogen peroxide and molecular oxygen (Bruijn et al. 2004; Valentine et al. 2005). Post-

translational modifications are required to activate the enzyme, and these include: binding of

copper and zinc ions at the appropriate residues, dimerization of two identical SOD1 monomers,

and stabilization via disulfide-bond formation (Vehviläinen et al. 2014). The catalysis of the

superoxide radical into molecular oxygen and hydrogen peroxide is dependent on the bound

copper ion which is successively reduced and oxidized by the reactive oxygen species (Bruijn et

al. 2004). The SOD1 enzyme is generally an incredibly stable polypeptide, with the protein

remaining stable and copper-bound, even in 8M urea (Wong et al. 1995). In addition to the

exposure to oxidative stress, SOD1 protein products need to be transported through the

axoplasm, the size of which can greatly exceed the length of the neuronal cell body.

Anterograde transport of SOD1 is accomplished by exploiting slow component-b, and as such,

the enzyme needs to exhibit an extended half-life so as to continually scavenge free radicals

along the length of the axon (Borchelt et al. 1998; Valentine et al. 2005).

7

Within this milieu of oxidative stress and prolonged half-life, it is expected that aberrant

conformational changes can occur at the tertiary and quaternary levels of protein organization. In

vitro studies have shown that under conditions of oxidative stress, the enzyme will be

predisposed to changes in its native conformational state, and that both mutant and wild-type

SOD1 exhibit an increased propensity to misfold and aggregate (Rakhit et al. 2002; 2004). To

date, the subset of fALS cases linked to the SOD1 locus arise from point mutations, and confers

on the protein not only a reduced half-life (Hoffman et al. 1996), but a novel toxic property that

ultimately overwhelms any compensatory cellular defensive mechanisms (Gurney 1997).

Subsequently, a current hypothesis posits that the underlying pathobiology of ALS arise due to

an aberrant structural conformational change of the SOD1 protein, which is borne out of either a

genetic predisposition to misfolding, or an exogenous environmental insult targeted at

destabilizing the homodimeric protein. As intriguing as the possibility of research outcomes

applying equally to both sporadic and inherited forms of ALS, this may not be completely

accurate. Liu et al (2009) found that fALS and sALS cases could be clearly differentiated via

immunolabelling with an antibody specific for an epitope exposed when the SOD1 dimer

becomes destabilized. This, finding, however, is highly dependent on the epitope that the

antibody is specific for (Forsberg et al. 2010; Rotunno & Bosco 2013).

To date, the exact pathomechanism(s) underlying the phenotype observed in cases of mutant

SOD1 remains unresolved. Multiple aberrant pathways have been identified and particular

aspects have been targeted by therapeutics. It is highly probable that there is an intricate interplay

among these various pathways that ultimately culminate in the prototypical neurodegenerative

8

changes that are observed. Select pathways are considered in the following sub-sections (See

Figure 1.3 below).

1.3.1 Mitochondrial Dysfunction: Altered Energy Metabolism, Oxidative Stress, and

Apoptosis

SOD1 can be localized to both the cytosol and the mitochondrial intermembrane space (Shaw

2005). The mitochondrial organelle is crucial in generating the intracellular energy stores

required to drive cellular biochemistry, maintaining calcium homeostasis, and plays an important

role in regulating apoptosis (Shi et al. 2010). Dysfunction of mitochondria would ultimately

perturb the energy balance of the system and contribute to cellular degeneration. Clinically, in

ALS patients, aberrant mitochondrial pathology show a dense aggregation of the organelle

localized to the presynaptic terminal (Sasaki & Iwata 1996). Murine models of SOD1-ALS show

evidence of aberrant mitochondrial pathology; most notably mitochondrial swelling and

vacuolization (Dal Canto & Gurney 1994; Wong et al. 1995).

Mitochondria are a principal source of reactive oxygen species (ROS) within the cell on account

of reactions that metabolize oxygen (Barber et al. 2006; Vehviläinen et al. 2014). These species

include both hydrogen peroxide and the superoxide anion. Oxidative stress arises when there is a

perturbation between the balance of ROS production and removal (Turner et al. 2013). As

reviewed by Barber et al, multiple studies have implicated oxidative stress in the pathogenesis

of ALS (2006).

9

A)

C)

B)

D)

10

Figure 1.3. Select mutant SOD1 pathomechamisms contributing to neuronal degeneration.2

A) In the context of mutant SOD1, excessive glutamate excitotoxicity can arise due to a decrease in glutamate re-uptake; with the resultant calcium ion influx driving the cellular death cascade. B) Mutant SOD1 has been shown to interact and sequester Derlin-1 which typically is involved in shuttling misfolded proteins from the endoplasmic reticulum into the cytosol for ubiquitin-mediated proteasomal degradation. With Derlin-1 rendered non-functional, misfolded proteins accumulate within the ER, inhibiting the ER-associated degradation pathway and resulting in ER stress that ultimately activates cell death pathways. C) Non-cell autonomous processes have long been suspected to play a role in mSOD1-mediated neurodegeneration. Cytotoxic and pro-inflammatory substances produced by activated microglia act on the neuronal population, and excessive stimulation gives rise to cell death. D) Mutant SOD1 has been found to accumulate within the mitochondrial intermembrane space and can contribute to cellular degeneration through multiple pathways: i. Excessive reactive oxygen species formation can damage a variety of cellular components. ii. Mutant SOD1drives the cytosolic localization of cytochrome C from the mitochondrion, which activates a plethora of caspases involved in the apoptotic cascade.

It is expected that the hydrogen peroxide produced by mitochondrial SOD1 leads to oxidative

damage since the organelle lacks the scavenging systems required to further reduce hydrogen

peroxide (Vehviläinen et al. 2014). Superoxide and hydrogen peroxide can ultimately produce

peroxynitrite and the hydroxyl radical, respectively. These reactive species can damage various

constituents of the cell such as DNA, proteins, and lipids; thus oxidative damage can ultimately

result in mutations, altered protein conformations, and changes in cell membrane properties

(Barber et al. 2006). If the cell is unable to completely process and eliminate the superoxide and

hydrogen peroxide species, cellular function and viability are negatively affected.

Functionally, in vivo model systems expressing the G93A missense mutation present with

evidence for a decrease in mitochondrial membrane potential, an increase in the concentration of

Cellular elements for Figure 1.4 provided by Servier medical art and modified under a Creative Commons Attribution 3.0 Unported License.

11

cytosolic calcium ions, as well as an increased susceptibility to oxidative stress (Carrì et al. 1997;

Kruman et al. 1999). None of these experimental findings clearly delineate whether mutant

SOD1 is driving these aberrant cellular changes or whether these are secondary to another causal

cellular insult. To address this discrepancy, an in vitro model system expressing mutant SOD1 in

various organelles has further characterized the importance of mitochondrial dysfunction in the

pathogenesis of SOD1 ALS. Mitochondrial localization of the mutant enzyme is crucial in

inducing the release of cytochrome C into the cytosol which drives the activation of caspases that

ultimately culminate in apoptotic cell death (Figure 1.4; Takeuchi et al. 2002).

There is experimental evidence to suggest that mutant SOD1 can directly interact with the anti-

apoptotic factor Bcl-2, and prevent it from inhibiting the apoptotic pathway (Pasinelli et al.

2004). The prototypical apoptosis pathway involves a plethora of cell signaling players.

Typically, Bcl-2 binds to- and inhibits BID (BH3 interacting-domain death agonist) from

activating the downstream apoptotic cascade. In response to pro-apoptotic signals, BID interacts

with BAX (Bcl2-associated x protein), which leads to the insertion of BAX into the

mitochondrial outer membrane. BAX insertion ultimately leads to pore formation and results in

the release of cytochrome C from within the mitochondria which results in the activation of a

plethora of executioner caspases that ultimately result in cell death.

Follow-up research complicated the straightforward interpretation of mSOD1 inhibiting the pro-

survival signaling cascade and subsequently resulting in programmed cell death by suggesting

that neuronal damage is independent of the apoptotic pathway (Gould et al. 2006). In light of this

finding, further evidence exploring the previously described interaction between mSOD1 and

12

Bcl-2, suggested that this association induced a toxic conformational change in the structure of

the pro-survival protein (Pedrini et al. 2010). The toxic form of Bcl-2 is expected to result in

mitochondrial dyshomeostasis in a manner that is independent of the classic apoptotic pathway

(Pedrini et al. 2010).

1.3.2 Excitotoxicity

Glutamate is a major excitatory neurotransmitter within the CNS (reviewed in Danbolt 2001).

Once released into the synaptic cleft, glutamate relays the action potential to the adjacent target

neuron and the original signal is further propagated towards the target site. Under normal

physiological conditions, an intricate system of neuronal and glial glutamate transporters result

in the expedient endocytic uptake of extracellular glutamate, and so remove the excitatory

stimulus acting on the post-synaptic cell. Increased levels of excess glutamate in the extracellular

milieu will result in excitotoxic activity of the post-synaptic cell due to aberrant Ca2+

homeostasis within the cytosol arising from excessive membrane depolarization mediated

through glutamate receptors (Figure 1.4; Julien 2001). Increased cytosolic Ca2+ ions are

compartmentalized within the mitochondrion, which can result in alterations of mitochondrial

permeability, production of reactive oxygen species (ROS) and/or activating a plethora of

enzymes involved in cell death pathways (Bruijn et al. 2004; Jaiswal et al. 2009; Zhivotovsky &

Orrenius 2011). Increased intracellular levels of Ca2+ can further result in the activation of nitric

oxide synthase (NOS), which subsequently produces nitric oxide (NO), a potent free radical that

can react with the superoxide anion (Hand & Rouleau 2002). The resultant peroxynitrite radical

can subsequently further propagate cellular dysfunction by damaging and inactivating various

proteins. Preliminary evidence invoking a role for glutamate in the pathogenesis of ALS included

13

an increase in peripheral levels (and a concomitant decease in CNS tissue) in ALS patients

(Plaitakis & Constantakakis 1993; Camu et al. 1993), reduced glutamate uptake in patient-

derived synaptosomes (Rothstein et al. 1992), and as well as a reduction in levels of the

astrocytic glutamate transporter, EAAT2 (Julien 2001).

1.3.3 Endoplasmic Reticulum Stress and Altered Proteasomal Functioning

As described in Appendix A.1, ALS neuropathological findings are often associated with

abnormal aggregates within the cytosol that are composed of misfolded proteins. These

misfolded proteins are typically tagged with ubiquitin and routed to the proteasome for

proteolytic degradation. These findings typically arise due to changes derived from altered

endoplasmic reticulum (ER) functioning.

Typically, the ER plays a critical role in the secretory pathway. Proteins destined for exocytosis

are modified via various chaperones so as to adopt the appropriate conformations that are

required to mediate proper function. ER dyshomeostasis would ultimately result in the

accumulation of misfolded proteins within the lumen of the ER; resulting in ER stress (Matus et

al. 2011). ER stress invokes the unfolded protein response (UPR) pathway which strengthens the

cells ability to produce properly folded peptides and to degrade proteins with abnormal

conformations (Matus et al. 2011). As pointed out by Matus et al, there are various lines of

evidence to suggest that ER stress plays an important role in the pathogenesis of both sporadic-

and familial ALS by way of increased measures of ER stress markers (2011).

14

Once invoked, the UPR increases the transcription of various proteins in an attempt to normalize

cellular functioning. Additional protein chaperones are recruited to mediate proper protein

folding in the ER, while an increase in proteins related to the ER-associated degradation (ERAD)

pathway tags misfolded proteins with ubiquitin for proteosomal degradation within the cytosol.

Ultimately, when left unchecked, prolonged ER stress leads to cell death via the induction of

apoptosis (Matus et al. 2011).

With its propensity to misfold and aggregate, mutant SOD1 (and WT SOD1 under oxidative

conditions) would be thought to be a prime candidate to induce ER stress. One possible pathway

with experimental evidence involves SOD1 causing a perturbation within the ERAD cascade

(Figure 1.3). As summarized by Polymenidou and Cleveland, mutant SOD1 has been shown to

directly interact with Derlin-1, a component of the ERAD pathway (2008). The interaction with

Derlin-1 is typically critical in mediating both the translocation of misfolded proteins from the

ER lumen to the cytosol, and the subsequent protein degradation (Polymenidou & Cleveland

2008). The mutant SOD1 association with Derlin-1 inhibits the ERAD pathway, exacerbates ER

stress by an increased accumulation of misfolded proteins within the ER lumen, and ultimately

leads to the initiation of the apoptotic cascade.

1.3.4 Non-cell Autonomous Contributions to Motor Neuron Cell Death

Long-standing neuropathological observations of astro- and microgliosis have served as key

hallmarks of neurodegeneration; however the unique contribution(s) of each of these cell-types

to the disease phenotype has been largely unknown. In the context of mutant SOD1, the crucial

15

role that these non-neuronal cells play in ALS pathobiology has become increasingly established

through use of recent novel research strategies.

Chimeric mice comprised of mixtures of both mutant SOD1 and wild type-expressing cells

affected the presentation of the ALS phenotype (Clement et al. 2003). This group demonstrated

that mSOD1 motor neurons exhibited a greater degree of survivability when surrounded by a

higher proportion of WT non-neuronal cells. However, to differentiate between the specific

contributions mediated through astrocytic/microglial expression of mSOD1, subsequent research

groups adopted strategies that specifically altered transgene expression in various cell-types.

Targeted depletion of mutant SOD1 in microglia (but not motor neurons) did not alter the onset

of the ALS phenotype, but significantly attenuated the rate of disease progression (Boillée et al.

2006). Exclusive mSOD1 expression within astrocytes did not lead to the development of motor

deficits or abnormal neuronal morphological changes (Gong et al. 2000). A targeted decrease in

astrocytic mSOD1 expression did not delay disease onset, however this reduction slowed late-

stage disease progression in an mSOD1 model (Yamanaka et al. 2008).

Mutant SOD1 expression within motor neurons has been shown to be a determinant of disease

onset in ALS murine models. Deletion of mutant SOD1 expression within motor neurons delays

the onset of disease, but does not attenuate the resulting progressive neurodegenerative changes

(Boillée et al. 2006; Ferraiuolo et al. 2011). Targeted expression of mSOD1 in neuronal cells

(within the context of a non-ALS murine model) results in neurodegeneration, which is

dependent on the level of mutant transgene expression (Jaarsma et al. 2008).

16

Together, these data suggest that ALS pathology is a non-cell autonomous process where glial

cells play a pivotal role in establishing the aberrant phenotype. Mutant SOD1-expressing

microglia and astrocytes interact with motor neurons to influence disease progression, while the

mSOD1-mediated insult in neuronal cells determine the initial onset of disease (Ferraiuolo et al.

2011).

1.4 Modeling ALS In Vivo: Transgenic Mice Overexpressing Mutant Human SOD12 3

Animal models delineating disease-course specific parameters are of indisputable importance in

the study of neurodegenerative conditions. Theoretically, these model systems not only afford for

a glimpse into the mechanistic underpinnings of disease pathogenesis, but provide for a

homogenous study population in which to test potential therapeutic interventions. However,

utilizing an animal model is not without certain caveats—a topic that is briefly discussed below

and further explored in Chapter 4:.

To explore the mechanistic underpinnings of fALS associated with mutations in the SOD1 locus,

animal models have been invaluable in determining that it is not a reduction in enzymatic

activity that underlies ALS pathogenesis, but the acquisition of novel toxic properties (Gurney

1997). Considered below are two commonly used models of mutant SOD1-expressing mice

(mSOD1) utilized in pre-clinical development. To date, however, multiple additional transgenic

lines carrying various mSOD1 missense mutations or additional loci implicated in ALS have

2Section 1.4 based on work published in Zwiegers et al, 2014; Zwiegers and Shaw, 2015

17

been developed (Kato, 2008; Joyce, Fratta, Fisher, & Acevedo-Arozena, 2011; Van Den Bosch,

2011).

1.4.1 Mice Expressing the Glycine to Alanine Substitution at Codon 93

Following the identification of multiple missense mutations resulting in SOD1-mediated fALS

by Rosen and colleagues (1993), Gurney et al. (1994) developed the G93A transgenic model:

one of the most widely used murine models used to evaluate the toxicity and efficacy of

therapeutic agents against ALS (disease-related phenotypic outcomes summarized in Table 1.1).

In brief, the research group amplified and excised the region of the genome encoding codon 93

that contained a nucleotide substitution that would give rise to a missense sequence coding for

alanine in lieu of glycine. Using this DNA fragment, the 14.5-kb human SOD1 (hSOD1) gene

carrying the requisite regulatory sequences was reassembled and microinjected into fertilized

(C57BL6XSJL)F1 embryos; thus generating the transgenic founder lines. As described by

Gurney et al. (1994), the founder line exhibiting the greatest amount of increased SOD1 activity

(G1) started to show evidence of deterioration by approximately 3 months of age. At this point,

mice presented with evidence of deteriorating stride length accompanied by signs of hind limb

weakness. By approximately 150 days of age, animals reached a paralyzed state in which

mobility was impeded to a degree where they were unable to seek out nourishment.

Neuropathologically, changes were of a progressive nature (Gurney et al. 1994; Dal Canto &

Gurney 1994; Dal Canto & Gurney 1995). Overall, G1 animals demonstrated a reduction in

choline acetyltransferase (ChAT) positive motor neurons, as well as a loss of myelinated axons

18

in the ventral motor roots within lumbar sections of the spinal cord. At the outset, vacuolar degeneration is observed within the motor

neurons, but with time, alongside a marked reduction of neuronal cell bodies, pathological vacuoles are present within the surrounding

neuropil with a marked deterioration of the anterior horn.

Table 1.1. Select phenotypic correlates of disease progression in the G93A murine model is dependent on transgene integration within the genome

Age at pathology (d) Genetic modification (line)

# of gene copies

Relative SOD1 Protein Levels (brain) (hSOD1 protein (ng)/µg total brain protein)

Relative increase in SOD1 activity (U/µg total brain protein)

Mean ALS onset

Mean age at death

Motor neuron vacuolation

Motor neuron loss

G93A (G1) ᵡ,1,2,3,4 18 4.1 42.6 90-120 153-185 73-163 >180 G93A (G1H)4,5,6 25 3.36† 3.04ᵟ 90 136 37 80 G93A (G1L) ᵡ,7 13 -- -- 200 251 200* 230 G93A (G5)1,4 4 1.3 27 -- -- -- -- G93A (G5/G5)ǂ,4,8 10 1.28† 1.61ᵟ 300 >400 ** >300 G93A (G12)1,4 2.2 1.1 19.5 -- -- -- -- G93A (G20)1,3 1.7 0.7 16.9 -- 332-354 -- ~350

ᵡPublications can erroneously refer to the original G1 line as G1L; it is imperative to confirm the copy number of the strain n question †SOD1 protein (ng) per µg soluble brain protein δ Activity (U) per µg soluble brain protein ǂ Bred to homozygosity * V. Little evidence of vacuolar pathology at this late stage ** V. Little vacuolar pathology evident at end stage of phenotype 1Gurney et al., 1994; 2Dal Canto & Gurney, 1994; 3Dal Canto & Gurney, 1995; 4Kato, 2008; 5Gurney, 1997; 6Chiu et al., 1995; 7Zhang et al., 1997; 8Dal Canto & Gurney, 1997

19

During the original investigation, Gurney only followed the transgenic animals up until

approximately 5 months of age: a point at which the additional lines (G5, G12, G20) expressing

lower levels of the G93A mutant SOD1 had not progressed to the point of establishing a

phenotype at either the behavioural or neuropathological level (1994). Later, establishing a role

for an effect of gene dosage, Dal Canto and Gurney (1995) demonstrated that animals expressing

a six-fold decrease in mutant G93A hSOD1 expression (line G20 compared to G1) were

accompanied by an alteration in the expression of an ALS phenotype. It thus became evident that

the timing and rate of disease progression was highly dependent on transgene expression within

these models.

Animals derived from line G20 demonstrated attenuated disease progression in that these mice

presented with onset of disease at a later stage, as well as failed to decline as rapidly as seen with

animals from the G1 line (Table 1.1; Dal Canto & Gurney, 1995). At nearly one year of age,

mice derived from line G20 exhibited symptomology consistent with that of end-stage ALS:

significant motor neuron loss, degeneration of the anterior horns, Lewy-like axonal inclusions, as

well as a marked absence of vacuoles (Dal Canto & Gurney 1995).

Pathological evidence for vacuolar degeneration in animals that highly express the mutant G93A

hSOD1 locus does not recapitulate observations of end-state pathology in human cases, and is

thus suspected to be a toxic artefact from the transgene being significantly over expressed (Dal

Canto & Gurney, 1995). Corroboration of this notion came with the establishment of a

homozygous G5/G5 line (see Table 1.1) that expresses the G93A locus to a lesser degree (10

copies vs. 18) as is seen in the G1 line (Dal Canto & Gurney, 1995). Clinical signs of hind limb

20

tremors in these animals initiated at approximately 300 days of age, with paralysis setting in

nearly 100 days later. Pathologically at 120 days, axonal swellings – attributed to

microvacuoles—were occasionally observed; with no adverse presentation in the neuropil or

deficits in neuronal ultrastructure. As the phenotype became more established, the size of axonal

swellings credited to aggregation of neurofilamentous bundles, increased. At 300 days of age,

marked deficits in neuronal number became established; with those remaining presenting with

ubiquitin positive intraneuronal inclusions that were reminiscent of Lewy bodies.

Due to rare recombination events that have either expanded or retracted the number of copies

randomly inserted into the genome, the G1 line has borne two additional sub-lines.

A line with a 40% expansion in copy number (25 copies of the G93A locus) show evidence of a

more severe ALS-pathology (Chiu et al., 1995; Table 1.1). These high copy number, or G1H

mice, starts to show evidence of neurodegeneration as early as 90 days of age with a tremor that

becomes increasingly pronounced and ultimately results in severe hind limb paralysis more than

a month later. At the pathological level, these animals exhibit the canonical aberrations hereunto

described to the G93A model, but at an accelerated pace. There is evidence of vacuolation in

spinal motor neurons present after 37 days, a decrease in myelinated axons after day 80, and loss

of ChAT-positive motor neurons after 90 days (Chiu et al. 1995).

On the other hand, mice carrying 13 copies of the G93A locus (a 30% reduction in transgene

copies from the original line; designated line G1L) exhibited an attenuated ALS phenotype

(Gurney, 1997). As described by Zhang et al (1997), animals derived from this line initially

show signs of muscle weakness at 200 days of age, with complete paralysis established 50 days

21

later. In this line, evidence of vacuolated motor neurons were scant, however intraneuronal

inclusions of neurofilamentous aggregates increased with age, and neuronal cell loss was

significant after 230 days.

This early work has established that the number of transgene copies randomly inserted into the

genome can exert a pronounced effect in modifying the rate of disease onset as well as the

progression of the deteriorating phenotype, since it is a determinant of the expression level of the

mutant locus.

1.4.2 Mice Expressing the Glycine to Arginine Substitution at Codon 37

Originally developed by Wong et al. (1995), the transgenic G37R line retaining full SOD1

specific activity was generated via PCR-directed mutagenesis. Briefly, a plasmid carrying a 12kb

DNA fragment encoding wild-type human SOD1 (hSOD1) was digested with restriction

endonucleases. A fragment containing the second exon was sub-cloned into an additional vector

and subsequently exposed to a mutagenic strategy at codon 37 using specific primer sets.

Following confirmation of the desired mutation, the complete gene-transcriptionally regulated by

endogenous promoters- was ultimately reassembled and the 12kb gene fragment carrying the

G37R mutation was microinjected into (C57Bl/6J X C3H/HeJ)F2 mouse embryos. Four

transgenic founder lines with variable expression levels of the mutant enzyme were shown to

develop phenotypic characteristics reminiscent of ALS (Table 1.2). Of these, G37R SOD1 line

29 animals were shown to express a moderate seven-fold increase in spinal cord SOD1 activity,

with the onset of motor deficits evident at 6-8 months of age.

22

Table 1.2. Select phenotypic correlates of disease progression in the G37R murine model is dependent on transgene integration within the genome

Age at pathology (d)

Genetic modification (line)

Relative SOD1 Protein Levels (human/mouse) spinal cord

Total increase in SOD1 activity

ALS onset

Motor neuron vacuolation

Motor neuron loss

Reactive astrogliosis

Mean age at death

G37R(42)1,2 12.3 14.5 106-121 77 105 77 155-176 G37R(9)1 6.2 9 152-183 70 105 35 -- G37R(106)1 5.3 7.2 167-228 189 126 77 -- G37R(29)1,2,3 5 7 183-243 none 133 77 348-380 WT hSOD11 10.5 9.2 none by

548d none none none --

1(Wong et al. 1995); 2(Nguyen et al. 2001); 3(Ezzi et al. 2010) Generally, transgenic animals that highly expressed the mutant locus developed normally and showed no signs of an adverse

phenotype until approximately 3.5 – 6 months of age (Wong et al. 1995). Initial signs included minor gait perturbations where a

reduction in spontaneous locomotion was observed accompanied by a difficulty in extending the hind limbs when suspended by the

tail. Temporally, symptoms progressively worsened, ultimately resulting in a significant reduction in the animal’s weight, as well as

significant muscle atrophy that was accompanied by paralysis of the hind limbs. Extended observation of additional lines of transgenic

animals that express either lower levels of the mutant G37R hSOD1 locus, or increased levels of wild-type hSOD1 (at 9 and 18

months, respectively), did not show any behavioral or neuropathological correlates of ALS pathogenesis (Wong et al. 1995).

23

Neuropathologically, a progressively degenerative cascade is observed within motor neurons at

various levels of the spinal cord and brainstem (Table 1.2; Wong et al., 1995). Initially, up until 5

weeks of age, no anomalous pathology in terms of motor neuron degeneration is observed. At 11

weeks of age, transgenic lines expressing a greater amount of the mutant hSOD1 locus present

with evidence of vacuolar degeneration of axons and dendritic processes. Nine weeks later (20

weeks of age), these animals show evidence of motor neuron loss at the level of the lumbar

spinal cord. Accompanying this neuronal loss is widespread muscle atrophy and muscle

denervation that is typified at the behavioural level with marked deficits in measures of

performance (i.e. leg extension reflex score, reduced stride length, impeded rotarod performance

etc.). Furthermore, reactive astrogliosis is evident in all lines carrying mutant G37R hSOD1 loci

that express the gene to an appreciable degree, with some lines showing evidence as early as 5

weeks of age. On the contrary, it is important to note that animals generated to over express the

wild-type hSOD1 locus fail to recapitulate any features of motor neuron degeneration and do not

present with ALS-associated pathologies (Table 1.2).

1.5 Caveats Inherent to ALS In Vivo Model Systems3 4

There is a stark paucity of positive clinical outcomes based on therapeutics developed in these

murine models. Many pre-clinical trials indicate a positive outcome in terms of the overall

survivability; however, there are various caveats that negatively affect the clinical translation of

the positive pre-clinical findings that are discussed in Chapter 4. As explored later, chief among

43Section 1.5 based on work published in Zwiegers et al, 2014;Zwiegers and Shaw, 2015

24

these issues being variations in transgene copy number along with a limited applicability to cases

of human ALS.

1.5.1 Transgene Copy Number Variation

As explored in the preceding sections, due to the random nature of transgene integration when

these mSOD1 animal models are established, various lines expressing the mutant enzyme to

varying degrees were established. Furthermore, intralocus cross-over events during meiotic

recombination could further expand or retract the number of gene copies within the genome. For

the G93A line of animals, this is a relative rare occurrence; with approximately 3% of generated

progeny showing signs of altered transgene levels compared to the original parental line

(Alexander et al. 2004). Fluctuations in transgene levels ultimately present at the phenotypic

level; with expansions increasing disease severity, and retractions attenuating the expected

phenotypic disease course. As depicted in Figure 1.4, this is anticipated to have severe

consequences on experimental outcomes since any measure of pre-clinical therapeutic efficacy

may just be consequent to transgene level variation. Thus, depending on the constitution of the

various cohorts used in an experimental paradigm, different outcomes between control and

treatment groups may be an inherent consequence of the underlying transgene dosage.

Furthermore, unreported variations in transgene levels undermine replicative studies between

independent research groups as significant alterations in copy number can potentially bias the

results obtained.

25

Figure 1.4. Effect of mSOD1 transgene levels on the expression of an ALS phenotype and expected therapeutic outcomes based on disease severity.

Proposed model explaining pre-clinical outcomes assessed in transgenic animals. It is possible that the therapeutic effect (or lack thereof) seen in these animal models may primarily be a function of variations in mutant locus copy number and the corresponding diseased phenotype. Outliers at either end of the transgene expression spectrum will undermine replication studies and effective clinical translation.

1.5.2 Limited Applicability to Clinical ALS Cases

ALS clinical cases account for an estimated worldwide prevalence rate of approximately 4-

7/100,000 population (Chiò et al. 2013). This roughly corresponds to a total of 500,000 total

ALS cases worldwide. Taking into account only SOD1-linked familial ALS; 5000-10,000

patients worldwide would be expected to carry a mutant form of the SOD1 locus. As explored in

section 1.2 and A.1-A.2, there is evidence to suggest that the underlying pathobiology inherent to

mutant SOD1 phenotype expression is fundamentally different from the majority of ALS cases.

Subsequently, a model predicated upon a specific genetic causality may not have widespread

applicability to the ALS community and thus undermine endeavours of clinical translation.

26

1.6 The Properties and Diverse Biological Functions of Progranulin

Progranulin (PGRN) is a conserved, secreted cysteine-rich protein with a multitude of cellular

effects including wound healing, cell proliferation, and modulating the inflammatory response

(De Muynck & Van Damme 2011). The 8 kb region located on chromosome 17q21, has become

a topic of vested interest in the study of neurodegenerative disease as the pleiotropic product may

be important in the long-term survival of nerve cells (Ahmed et al. 2007; Cruts & Van

Broeckhoven 2008). Due to post-translational processing, the secreted growth factor, is heavily

glycosylated and appears as an approximately 90 kDa protein band following sodium dodecyl

sulfate polyacrylamide gel electrophoresis (He & Bateman 2003). The protein is derived from 13

exons and consists of 7.5 tandem repeats of a conserved 12-cysteinyl granulin motif (C-X5-6-C-

X5-CC-X8-CC-X6-CC-X5-CC-X5-C-X5-6-C; De Muynck & Van Damme 2011). Proteolytic

cleavage of full-length PGRN gives rise to several 6 kDa granulin (GRN) peptides.

Progranulin is expressed in multiple cell types including mitotically active epithelia, immune

cells, microglia and various neuronal subtypes (Daniel et al. 2000; He & Bateman 2003; Ahmed

et al. 2007; Ryan et al. 2009). Interestingly, both the full- length protein and the cleaved peptides

induce a multitude of effects on cellular physiology (Figure A.1 ; He & Bateman 2003). These

include: male-specific brain differentiation, wound healing, inflammation, and the regulation of

tumorigenesis. Many of the physiological effects at the cellular level are mediated through a

myriad of intra- and/or extracellular interactions between PGRN (or the various cleaved GRN

peptides) and binding partners including Sortilin, Toll-like receptor 9, and the receptor for Tissue

Necrosis Factor (Figure A.1; For additional information, see section A.3 for a description of

PGRNs interaction with these proteins).

27

Evidence for the neurotrophic properties of PGRN has steadily accumulated within the research

literature over the past several years and this has indicated that the pleiotropic protein exhibits

various properties that are atypical of other more well-characterized growth factors. In a model

of traumatic brain-injury, PGRN mRNA expression was upregulated 24 hours following the

insult, while brain derived neurotrophic factor and neuregulin showed a more immediate

response with mRNA levels increasing within 3 hours of the injury (Matzilevich et al. 2002). In

an axotomy model of acute neuronal injury, changes in PGRN expression was shown to be cell-

type dependent (Moisse et al. 2009). Neuronal PGRN expression decreased 7 days following the

injury, while microglial expression was markedly upregulated: both returning to baseline levels

28 days after the acute injury.

Studying the growth factor in culture has further elucidated certain aspects of its function.

Exogenous addition of recombinant full- length PGRN or the cleaved GRN E domain to primary

cultures of rat cortical and motor neurons exhibited increased neuronal survival and neurite

outgrowth (Van Damme et al. 2008). Overexpression of PGRN in NSC-34 cells has been shown

to protect transduced cells from serum-deprived cell death for a prolonged period of up to two

months (Ryan et al. 2009). Furthermore, increased PGRN expression was found to be induced as

a stress response to hypoxic and acidic environments of cultured fibroblasts (Guerra et al. 2007).

The cytoprotective effect mediated by the exogenous addition of PGRN has been found to

involve the activation of cell survival signaling cascades. In primary neuronal cells exposed to

excitotoxic or oxidative stressors, PGRN was found to activate both the extracellular signal-

28

regulated kinases (ERK) and protein kinase B (PKB aka Akt) (Xu et al. 2011). The authors

further showed that depending on the nature of the toxic insult, inhibitors of these signaling

pathways ablated the neuroprotective effects of the exogenously added PGRN (Xu et al. 2011).

1.7 Neurodegenerative Disease and Progranulin

1.7.1 Ubiquitin-positive Frontotemporal Lobar Degeneration

Clinical relevance of the growth factor was originally identified by two independent research

groups that implicated null mutations within the PGRN sequence as the genetic basis for tau-

negative ubiquitin-positive frontotemporal lobar degeneration (FTLD-U), which materializes due

to a state of haploinsufficiency (Cruts et al. 2006; Baker et al. 2006).

Frontotemporal lobar degeneration encompasses a clinically heterogeneous subset of conditions

that arise due to progressive cortical atrophy present in the frontal and temporal lobes of affected

patients. Frontotemporal dementia (FTD) is the most prevalent form of non-Alzheimer's

degenerative dementia; accounting for up to 20% of all dementia cases, with an age of onset

typically between the fifth and seventh decade of life (Neary et al. 2005). To date, more than 60

pathogenic PGRN mutations have been implicated in FTLD pathophysiology and are confirmed

or expected to be loss-of-function (LOF) alleles (Cruts et al. 2012; Petkau & Leavitt 2014).

It is therefore anticipated that ubiquitin positive inclusions associated with FTLD is manifested

due to the resulting haploinsufficiency in which inadequate amounts of the gene product is

produced to promote optimal cellular functioning.

29

1.7.2 Amyotrophic Lateral Sclerosis

Although a proportion of ALS patients meet the clinical criteria for a diagnosis of FTLD, to date,

PGRN mutations have not been found to be a genetic determinant causal to the development of

ALS pathogenesis (Swinnen & Robberecht 2014; Petkau & Leavitt 2014). Previous work has

described a patient presenting with frontotemporal dementia lacking any clinical signs of motor

neuron dysfunction, even though there was a family history of autosomal dominantly inherited

ALS (Spina et al. 2007). Additional work has described novel variations of the gene in a single

case of ALS-FTD and sALS; suggesting that changes in this loci are not a prevalent cause of

motor neuron damage (Schymick et al. 2007). Genetic variability at this locus can, however, play

a role in the phenotypic expression of the neurodegenerative cascade. A study of Belgian ALS

patients suggested that polymorphisms in the PGRN gene can be a modifier of ALS disease

progression in that disease phenotype presents at an earlier age of onset coupled with a more

rapid decline of survival (Sleegers et al. 2008).

Clinically, PGRN levels in the cerebrospinal fluid and plasma of newly-diagnosed ALS patients

do not differ from that of control subjects which would indicate that the expression levels of the

secreted factor remains unchanged at the onset of the diseased state (Philips et al. 2010). Post-

mortem immunohistochemical analysis of end-stage ALS cases, indicate an increase of

progranulin expression in both neurons and glia within areas of neurodegeneration (Irwin et al.

2009). Recapitulating an ALS phenotype in the G93A model, Philips and colleagues were able to

not only demonstrate that PGRN levels in the CSF increase over the disease course, but as the

phenotype deteriorates, expression of progranulin within the spinal cord increases, and that this

upregulation correlates with an enhanced expression of the gene within the microglia (2010).

30

What is unknown at this point is whether this late stage upregulation of the PGRN product is

causal to pathogenesis or a mechanism to mitigate additional neuronal damage mediated via the

toxic nature of mSOD1.

1.8 Previous Pre-clinical Applications of Progranulin in the CNS

On account of the important developing role for PGRN in the CNS, several pre-clinical

therapeutic strategies have employed the growth factor in an attempt to attenuate

neurodegenerative changes.

Two independent groups have explored the therapeutic efficacy of enhancing PGRN expression

in murine models of Alzheimer’s disease (AD). Lentiviral-mediated delivery of PGRN cDNA

directly into the dentate gyrus of AD mice lowered the amyloid β (AB) plaque burden, and

concomitantly prevented neuronal loss and memory deficits (Minami et al. 2014). Utilizing a

similar strategy in an additional AD model, Van Kampen and Kay demonstrated a reduced AB

plaque burden along with a decrease in neuroinflammation and a purported preservation of

synapses (2011).

Further exploring PGRNs pleiotropic properties, Van Kampen et al delivered PGRN cDNA

directly into the substantia nigra of male C57Bl6 mice, who were subsequently exposed to a

parkinsonism-inducing neurotoxin (2014). Overexpression of PGRN was shown to preserve

nigrostriatal neurons as well as influence some of the characteristic pathophysiological events

underlying Parkinson’s disease, namely apoptosis and neuroinflammation (Van Kampen et al.

2014).

31

1.9 Research Theme, Objectives, and Experimental Hypothesis

1.9.1 Research Theme

Multiple experimental studies have explored the pleiotropic and neurotrophic properties of

PGRN. Loss-of-function mutations have been implicated in FTLD (Ahmed et al. 2007), while

exogenous addition of the growth factor has produced a beneficial effect in models of arthritis,

Parkinson’s- and Alzheimer’s disease (Tang et al. 2011; Van Kampen & Kay 2011; Minami et

al. 2014; Van Kampen et al. 2014). Preliminary in vitro work in our laboratory have indicated a

neuroprotective role of the PGRN product in a MPTP model of cytotoxicity (Figure C.1A), while

lentiviral-mediated upregulation in an ALS murine model approaching disease end stage showed

a significant attenuation in the loss of motor neurons (Figure C.1B,C). We thus wanted to

explore the effect of upregulating exogenously delivered PGRN cDNA in an ALS murine model

at a time prior to the onset of significant behavioral deficits and the concomitant loss of motor

neurons.

1.9.1.1 Objectives

The objective of this study was to assess the behavioural and neuropathological outcomes

following the upregulation of exogenously delivered PGRN prior to phenotypic onset in an ALS

murine model. A plethora of behavioural tests and histological assays were employed to

determine whether both behavioral and neuropathological correlates of disease were alleviated

following PGRN administration.

32

This study aims to contribute to our current understanding of the potential neurotrophic effects

mediated through progranulin and assess whether the growth factor can therapeutically be

applied in the context of remediating and/or attenuating ALS pathogenesis.

1.9.1.2 Hypothesis

Lentiviral-mediated retrograde targeting of PGRN cDNA into the motor neurons of G37R (line

29) animals from the gastrocnemius muscle injection site—at a stage prior to prototypical

phenotypic expression—is predicted to:

(i) lessen the severity of the expressed behavioural phenotype

(ii) attenuate motor neuron loss

(iii) diminish the severity of astrogliosis and microgliosis

33

Chapter 2: Materials and Methods

2.1 Materials and Methods 5

2.1.1 Colony Generation and Animal Husbandry

All breeding, handling, and experimental manipulation of research animals were in accordance

with the established guidelines of the Canadian Council on Animal Care, and approved under the

auspices of UBC’s Animal Care Committee (Breeding certificates: A07-0018; A10-0383;

Experimental animal handling certificates: A07-0213, A11-0342). Mice (maximum 5 per cage)

were housed at the Jack Bell Research Centre (JBRC; Vancouver, British Columbia), at a

constant temperature of 21-22°C, and exposed to a 12-hour light/dark cycle. Food (Purina

LabDiet ®) and water were available ad libitum. General animal husbandry and daily animal

monitoring was provided by JBRC staff, while our lab was responsible for all experimental

procedures (see Figure 2.1 for an outline of the experimental timeline). At 3-4 months of age,

animals were transferred to containment level 2 (CL-2) for lentiviral delivery, and were housed

there until euthanized. At the onset of phenotype expression in mSOD transgenic animals, a

hydrated, high caloric gelatinous substrate was added to the cage bottom as required for

palliative care.

5 See Appendix B for a list of chemical reagents utilized in this study

34

Figure 2.1. Graphical overview of the experimental timeline

A colony of mutant SOD1 (G37R line 29) and wild-type conspecifics were generated. Following baseline behavioural measures, lentiviral administration of a GFP or PGRN cDNA construct was accomplished through injection into the gastrocnemius muscles. Behavioural monitoring continued until the end-stage. Following euthanasia, tissues were processed and histologically assessed. 2.1.1.1 Breeding Wild-type and mSOD1 Transgenic Progeny from Commercially-

Obtained Breeders

A colony of mice heterozygous for the G37R (line 29) locus were generated from breeding pairs

obtained commercially from the Jackson Laboratory (Bar Harbor, MI). A total of six male

B6.Cg-Tg(SOD1*G37R)29Dpr/J (G37R; Jackson Laboratory) breeders were crossed with wild-

type congenic C57BL/6 females (Jackson Laboratory) to generate progeny that seeded the

colony. Briefly, in cages of group-housed females (3-4 per cage), bedding material from the male

35

to be used in copulation was added. Three days later, the male was introduced into the cage, and

remained there for a maximum of two weeks. At this point, each female was individually housed

and monitored for the delivery of a litter. After approximately 3.5- 4 weeks of age, ear punch

samples were collected from progeny and animals were weaned. Two consecutive rounds of

breeding were required to generate sufficient transgenic animals for the current study under

consideration (Table 2.1). Due to facility space limitations, progeny were group-housed based on

sex after weaning.

Table 2.1. Experimental animals stratified by sex, genotype, treatment and group

Viral administration

Jun-11 Aug-11

Sex Genotype Lentiviral construct

N (cohort #1)

N (cohort #2)

Total N

Male Wild-type GFP 12 0 12

Wild-type PGRN 11 1 12

G37R GFP 6 5 11

G37R PGRN 7 6 13

Female Wild-type GFP 9 3 12

Wild-type PGRN 10 2 12

G37R GFP 7 4 11

G37R PGRN 7 5 12

2.2 Quantitative Genotyping

2.2.1 Distinguishing Wild-type from Transgenic mSOD1 Animals

To confirm mSOD1 transgene presence, ear tissue samples were obtained from progeny at the

time of weaning. At 3.5- 4 weeks of age, the offspring were anesthetized in a closed chamber

with 2% isoflurane (Baxter, ON) and an ear punch apparatus (Jack Bell Research Centre) was

used to excise viable tissue specimens and permanently identify each subject. Animals were

36

observed 15- 30 minutes after tissue sampling to ensure that each subject had recovered from

anesthesia.

Tissue samples were quantitatively genotyped against the c-Jun housekeeping gene via an

automated genotyping platform developed by Transnetyx Inc. (Cordova, TN; mSOD1 primer

sets: forward primer CAGTAACTGAGAGTTTACCCTTTGGT; reverse primer

CACACTAATGCTCTGGGAAGAAAGA; Cordova, TN).

Results are reported as the relative difference between the cycle threshold values (CT) found in

assays for the housekeeping and target gene of interest, raised to the second power to account for

the doubling of target DNA during each cycle of the PCR process (i.e. Signal = 2 (CTcJUN –

CTmSOD1)).

2.2.2 End-stage Confirmation of GFP and PGRN Delivery

To assess the presence of the delivered transgene within the spinal cord at the experimental end-

stage, we again contracted out quantitative genotyping to Transnetyx (PGRN primer sets:

forward primer TCACTGTGTCTGGGACTTCCA; reverse primer

GCAGCAGTGGTAGCCATCA; GFP primer sets: forward primer

CGTCGTCCTTGAAGAAGATGGT; reverse primer CACATGAAGCAGCACGACTT).

Glass slides mounted with spinal cord tissue sections (10x 10µm thick sections per slide; see

Section 2.7.1 for cryosectioning information) were acclimated to room temperature for 15

37

minutes; dried at 40˚C for 15 minutes, rinsed in 1x phosphate-buffered saline (1x PBS) for 5

minutes, and dried at 37˚C for an additional 30 minutes.

Dried tissue sections (20 per animal; n=3 male animals per group) from each animal were

scraped off of each slide with a clean razor blade (VWR) and submitted to Transnetyx for

automated real-time PCR.

2.3 Vector Design

All experimental manipulations, procedures and biosafety considerations involving lentiviral

preparations were approved by the Biosafety Committee at the University of British Columbia

(Biohazard approval certificate: B10-0094). Vector design and viral preparation were contracted

out to Invitrogen Inc. (Carlsbad, CA) and Applied Biological Materials (Richmond, BC).

In short, a third-generation lentiviral vector encoding either Progranulin (PGRN; nm_008175;

Appendix D.1) or emerald GFP (GFP; Appendix D.2) cDNA was generated in the context of

three packaging plasmids which coded for the VSV-G envelope protein, Gag/Pol proteins, and

the Rev accessory protein. Viral particles are rendered inactive by way of a truncation within the

U3 region of the 3’ long terminal repeat which attenuates the potential oncogenic nature of the

retrovirus, as well as limits transgenic expression to the gene of interest.

38

2.4 Lentiviral Administration Targeting Motor Neurons in the Lumbar Spinal Cord:

Bilateral Gastrocnemius Muscle Injections

At approximately 3.5 months of age (range 91-116 days), under isoflurane-induced anesthesia

(3% 02, 1.5-2.5% isoflurane), both rear limbs were shaved and prepared for viral injection within

a biosafety cabinet. Ophthalmic ointment (Jack Bell Research Centre) was applied to both eyes

to prevent complications from ocular desiccation, and reapplied throughout the procedure as

needed. Ten microliters (5x2µL injections) of a commercially-designed lentiviral vector was

administered directly into each cleanly-shaven gastrocnemius muscle via a Hamilton syringe

(Reno, NV). The target muscle was chosen so as facilitate retrograde transport of the viral

construct to the lumbar spinal cord, specifically, regions L3-L5 (Figure 2.2; Nakajima et al.

2008).

39

Figure 2.2. Schematic of lentiviral administration into gastrocnemius muscles and the retrograde transport of transgene constructs into the spinal cord.6

At approximately 3.5 months of age, bilateral injections of the GFP or PGRN viral constructs targeting both gastrocnemius muscles were administered to anesthetised animals. Anterograde transport of the viral vector ultimately facilitated expression within motor neurons of the lumbar spinal cord.

Each lentiviral preparation was administered at 4.0x108 transduction units (TU) per mL and

contained either a GFP or Progranulin cDNA construct (Applied Biological Materials, BC),

resulting in each animal receiving a total of 8.0x106 functional vector particles

2.5 Assessing Disease Onset and Progression

Characteristic phenotypic changes in mice modeling ALS are instrumental in delineating the

disease course and to establish whether a tested therapeutic intervention attenuates the

progressive nature of the disease. Starting at 9 weeks of age, baseline measures in animals were

obtained weekly for 5 consecutive weeks. Following viral administration, animals were housed

in a CL-2 facility adhering to CL-3 operating procedures as delineated by UBC’s Biosafety

Program and approved by the internal Biosafety Committee. Periodic (bi-weekly to monthly)

assessment of weight, latency to fall from an inverted elevated grid, and a score on the leg

extension reflex test were utilized to assess the onset of an ALS phenotype in the transgenic

animals and to monitor disease progression.

Once transferred to the containment unit, mice were singly tested for each of the behavioural

parameters between 7 a.m. and 5 p.m. Animal handling was limited to a biological safety cabinet

Elements for Figure 2.2 provided by Servier medical art and modified under a Creative Commons Attribution 3.0 Unported License.

40

as per restrictions prescribed by the animal housing facility to limit subject exposure to other

viral/bacterial pathogens studied in the containment unit. At each trial, animals were randomized

at the cage-level, with the experimenter blinded to which viral construct was administered. Due

to the temporal progression of the ALS phenotype, the experimenter was eventually able to

distinguish between transgenic and wild-type conspecifics.

2.5.1 Body Weight Changes

At each time point body weight data was collected with an electronic scale. In transgenic ALS

animals, a temporally progressive reduction in body weight is a prototypical feature of disease

pathogenesis (Wong et al. 1995; Weydt et al. 2003; Dupuis et al. 2004; Kieran et al. 2007;

Oliván et al. 2015). Tracking temporal weight changes thus provides for an objective assessment

of disease progression. Disease onset was retroactively determined to be the time point at which

the animal had reached its peak body weight prior to phenotype-related atrophy (Boillée et al.

2006; Wang, Popko & Roos 2014).

2.5.2 Wire Hang

Measuring the latency to fall from an inverted grid is considered an index of neuromuscular

strength and motor coordination (Paylor et al. 1999; Oliván et al. 2015). In models of ALS, this

test is typically used to distinguish early phenotype-related deficits (Miana-Mena et al. 2005;

Günther et al. 2012). To conduct the wire hang test, mice were placed on a 30 cm elevated wire-

grid, inverted 180° over a soft surface, and the latency to fall recorded for a maximum trial run of

60 seconds (Tabata et al. 2008).

41

2.5.3 Leg Extension

The leg extension reflex test is used to evaluate the presence or absence of hind limb extension

when mice are suspended from the base of the tail as a measure of neuronal dysfunction

(Barnéoud & Curet 1999). Our group has modified this test to discriminate more subtle

differences in an environmental model of ALS, employing a scale from 0 (full retraction) to 4

(normal extension) based on the response of the hind limbs shown by each mouse (Wilson et al.

2004; Tabata et al. 2008). The scaled test was defined as follows; a score of four: complete

extension of both limbs (normal). Three: two limbs extended with some tremors and/or punching

of one limb. Two: one limb extended, one retracted, or tremors in both limbs. One: one limb

retracted and tremors in the other. Zero: both hind limbs completely retracted.

2.6 Euthanasia and Organ Harvesting

Transgenic animals were euthanized for subsequent histological assays at the experimental end

point when hind limb paresis was fully established. At this point subjects are unable to correct

themselves within 20-30 seconds when placed on their sides (loss of righting reflex), and the

ambulatory difficulties significantly impede proper nutrition and hydration. Wild-type

conspecifics were sacrificed 2-3 months following euthanasia of the final set of transgenic

animals to confirm that these did not unexpectedly develop ALS-like tremors and paralysis (i.e.

due to an uncharacterized low mSOD1 copy number).

Mice were deeply anesthetized in an enclosed container with high dose of isoflurane (4-5%;

Baxter, Mississauga, ON) until each animal appeared to be unconsciousness. After confirming

that each animal had reached a surgical plane of anesthesia, a thoracotomy was performed, and

the heart exposed. A needle delivering phosphate-buffered saline (PBS, pH 7.4) or a fixative

42

solution of 10% buffered formalin (pH 6.9-7.1; Fisher, Ottawa, ON) was inserted into the left

ventricle, while an incision to the right atrium allowed for the complete discharge of blood once

the perfusion was initiated.

Mice were first transcardially perfused with PBS for 2-5 minutes until the effluent was devoid of

blood. Next, a fixative solution of 10% buffered formalin was delivered for 2-5 minutes until

fixation tremors ceased and the musculature appeared stiff. Each perfusion was accomplished

with the aid of a Masterflex peristaltic pump (Cole-Parmer Canada Inc., QC) so that the

PBS/formalin solution was delivered at a constant rate. Prior to each perfusion, the tubing

connected to the peristaltic pump was flushed thoroughly with PBS to remove any remaining

fixative in the system.

Tissue specimens (brain, spinal cord, liver, kidney, and gastrocnemius muscle) were dissected

out and deposited in scintillation vials filled with 10% buffered formalin.

2.7 Tissue Processing and Histological Assessments

2.7.1 Tissue Processing and Cryosectioning

Immediately following sample collection, tissues were fixed overnight in 10% buffered formalin

(Fisher, Ottawa, ON) at 4˚C. The formalin solution was subsequently decanted and the tissues

rinsed in PBS (pH 7.4). Cryopreservation was accomplished by equilibrating tissues in

successive increasing concentrations of sucrose dissolved in PBS (5% (10 min), 10% (10 min),

and 25% (overnight)). Equilibration was taken to be once the tissues sank to the bottom of the

scintillation tubes, at which point the samples would be transferred to the subsequent higher

43

sucrose concentration. Fully equilibrated (i.e. following 25% sucrose exposure) spinal cord

tissues were segmented into 3 portions roughly corresponding to the cervical, thoracic and

lumbar regions. Each spinal cord specimen was placed in a cryomold (Sakura Finetek USA Inc,

Torrance, CA); covered with an optimum cutting temperature (O.C.T) embedding medium

(VWR, Edmonton, AB), and frozen by immersion in an isopentane bath at -80˚C. Additional

samples were collected (brain, liver, kidney, gasctrocnemius muscles) and frozen in O.C.T

medium; however these were not further analyzed in the current study. Tissues were stored

frozen at -80˚C until processed for cryosectioning.

Serial transverse lumbar spinal cord sections (10 µm-thick) were cut with a Leica CM3050 S

cryostat (Leica Biosystems, Concord, ON) and mounted on Superfrost Plus charged glass slides

(Fisher, Ottawa, ON). Each lumbar cord segment was sectioned into 1.0mm sets which

corresponded to 10 slides with 10 tissue sections mounted on each. Spinal cord sections on each

slide were thus 100 µm apart. Subsequent to cryosectioning, tissue slides were stored at -20˚C

until required for further histological processing.

2.7.2 Cresyl Violet Assay: Staining for Neuronal Nissl Substance

To aid in neuronal quantification, assessing neuron morphology and identifying the lumbar

spinal cord region of interest, tissue slides were stained for Nissl bodies with cresyl violet

acetate. Briefly, slides were acclimated to room temperature, rinsed in PBS (2x 2min) and post-

fixed in 80% acetone (2 min) (Fisher). Tissue sections were subsequently re-hydrated in ethanol

(95%: 5 min; 70%: 2 min), rinsed in ddH2O (2 min), stained with a filtered cresyl-violet solution

(5 min; 0.1g cresyl violet acetate, 100ml ddH2O, 0.5ml glacial acetic acid), rinsed in distilled

44

water (2 min), and then dehydrated in a graded series of ethanol (70%: 0.5 min 95%: 0.5 min

100%: 2x 0.5min). Slides were finally cleared in xylene (Fisher, Ottawa, ON) and cover-slipped

(VWR, Edmonton, AB), in Permount mounting media (Fisher).

2.7.2.1 Differentiating Between Lumbar Spinal Cord Levels

Gross anatomical differences along the cervical-to-sacral axis were identified with the aid of a

mouse spinal cord atlas; these included the size and shape of the ventral horn, gracile fasiculus,

and spinal lamina (Anderson et al. 2009). Based on these morphological differences, the lumbar

spinal cord was grouped into L1-L2, L3-5, and L6 segments. Cresyl violet-stained tissue sections

on a slide from each 1.0 mm segment (i.e. 10 sections of tissue, 100 µm apart) of spinal cord was

examined and grouped into one of these designated lumbar cord levels. To aid subsequent

analyses, each set of slides corresponding to a 1.0 mm segment of lumbar spinal cord tissue was

thus stratified into these groupings along the L1-L6 axis.

2.7.2.2 Neuronal Counts With CellProfilerTM Cell Image Analysis Software

Neuronal viability was assessed in cresyl-violet stained tissue sections from regions L1-L6 of

male animals (minimum n=3 animals per group, average of sixteen 100 µm apart tissue sections

per lumbar spinal cord region). Images of each ventral horn were captured (Motic Images

Advanced 3.2, Motic, BC) at 10x magnification under a light microscope (B5 Professional

series, Motic) and saved as JPG files. Each image was cropped so as to be restricted exclusively

to each respective ventral horn, modified using a consistent set of parameters to both invert

colours and increase contrast between neurons and the background (Figure 2.3A). Modified

images were subsequently analyzed with CellProfilerTM image analysis software, which has been

45

used to identify and measure a variety of biological objects (Lamprecht et al. 2007). Based on

inputted parameters, the analysis software automatically identifies and counts objects of interest.

The software was programmed to identify two populations of neuronal cells: those bounded by

an equivalent diameter of 6-8pixels (diameter: 13-17 µm; area: 130-240 µm2), and 8-16 pixels

(diameter 17-35 µm; area: 240-950 µm2) within each ventral horn (Figure 2.3B). The combined

cell-size range encompasses both putative alpha and gamma motor neurons (Friese et al. 2009),

as well as is expected to account for atrophying alpha motor neurons. The experimenter visually

confirmed with the original image file that each identified object was neuron-like in morphology,

and adjusted the quantification accordingly. Cell counts from each cell population within both

ventral horns were calculated and averaged across all tissue sections from each animal.

2.7.2.3 Neuronal Morphological Analysis

For male animals in the L3-L5 cord region, motor neuron diameters were calculated as the

average of two lines bisecting the neuronal cell body in 10 µm-thick sections that were at least

100 µm apart (n=3 per group; 3-7 sections per animal). Images at 40x were obtained using a light

microscope and cell diameters (as the average length of two lines bisecting the cell body)

measured with the ImageJ software platform (NIH, MD). Cell bodies were characterized as

healthy when they presented with distinctive neuron-like morphology, a distinct nucleolus and

cytoplasmic Nissl granules. Atrophying cell bodies were those that appeared both shrunken in

appearance, typically hyperchromatic, and devoid of a nucleolus. Data from both ventral horns

were combined and averaged across all tissue sections for each particular animal.

46

Figure 2.3. Image processing and CellProfiler analysis of cresyl-violet stained spinal cord sections.

Lumbar spinal cord sections were stratified into regions L1-L2, L3-L5, and L6. Ten micrometer-thick tissue sections from each region were stained with cresyl violet and images at 10x magnification were obtained for further processing. A) Images were cropped so as only include the ventral horn and manipulated so that motor neurons were contrasted from a black background. Neuronal quantification and labeling of positive hits based on cell size was accomplished by CellProfiler. B) Both ventral horns (VHs) from

A) B)

47

each tissue section was analyzed by CellProfiler and cells with an area of 130-950um2 (6-16 pixels in diameter) were included in the counts. The output for each VH was visually confirmed with the original image, and neuron counts amended as needed.

2.8 Immunohistological Assays

Tissue sections for immunohistology were processed as follows: slides were acclimated to room

temperature (RT) for a minimum of 30 minutes and then dried at 50˚C for 30 minutes to limit the

number of tissues sloughing off during subsequent processing steps. Tissues were re-hydrated in

ddH2O (3 min) and then exposed to heat-induced antigen retrieval (10mM Citric Acid, 0.05%

Tween-20, pH 6.0) for 10 minutes. Following PBST (0.1% Tween-20 in PBS) rinses (2x 5 min),

endogenous peroxidase activity was quenched by incubating with 3% H2O2 (diluted in methanol)

for 15 minutes. Sections were then rinsed with additional PBST (2x 5 min washes) and

subsequently incubated for 60 minutes at RT in a blocking-permeabilization solution consisting

of PBST supplemented with either 10% normal goat serum (NGS; i.e. for rabbit anti-GFAP, GFP

and Iba-1 antibodies), or 10% normal horse serum (NHS; i.e. for the goat anti-ChAT antibody).

Following the incubation period, the blocking-permeabilization solution was removed and 10%

normal sera containing the requisite antibody at an appropriate dilution was added (in 10% NGS:

rabbit anti-GFAP (Abcam) at 1:200, GFP (Abcam) at 1:50; Iba-1 (Wako) at 1:100; in 10% NHS:

goat anti-ChAT (Millipore) at 1:100). In each experiment, subsets of tissues were incubated in

the absence of a primary antibody to test for non-specific binding interactions. Slides were

incubated overnight at 4˚C in a humidified chamber. Following the incubation period, tissue

sections were rinsed in PBST (3x 5 min) and incubated for 60 minutes at RT with the designated

biotinylated secondary antibody at a 1:200 dilution made in normal sera (goat anti-rabbit in 10%

NGS (GFAP, GFP, and Iba-1); horse anti-goat in 10%NHS (ChAT)). Following a PBST rinse

48

(2x 5 min), slides were incubated with the biotin-HRP conjugate prepared from the Vectastain

ABC kit (Vector Laboratories, Burlingame, CA) for 30 minutes. The 3,3’-Diaminobenzidine

(DAB) peroxidase substrate (Sigma, Saint Louis, MI) was added to the slides following a final

PBST rinse (2x 5 min) and observed until a distinct colour change was established (an incubation

period of 5-10 minutes). A subsequent ddH2O rinse (2 min) quenched the precipitation reaction

and attenuated any additional black-brown product formation. Next, hematoxylin as a

counterstain was employed to enhance the contrast of the immunohistological assays. Briefly,

following the ddH2O rinse, slides were stained with hematoxylin (Fisher, ON) for 1 minute;

rinsed under running water; immersed in 1% lithium carbonate (BDH Chemicals) for 30

seconds; rinsed with running water again; progressed through a series of dehydrating graded-

alcohols (70%, 90%, 2x 100% ethanol) for 30 seconds, and finally cleared in xylene (Fisher,

ON) prior to coverslipping (VWR, Edmonton, AB) in Permount mounting media (Fisher).

Immunohistological quantification was confined to the gray matter of the ventral horn (VH)

which was demarcated with the aid of an artificial line laterally bisecting the central canal.

ChAT-positive motor neurons from both VHs were counted directly under 10x magnification

(B5 Professional Series light microscope, Motic). Results are reported as an average of both the

left and right ventral horn, with a minimum of 4 tissue sections per animal (maximum of 10).

GFAP and Iba-1 immunoassays were quantified by digitally transposing a 6 x 6 counting grid

over a 640 x 512 pixels image of the ventral horn taken at 40x. The average number of positive

cells identified in each quadrant from a minimum of 3 sections (maximum 10) per animal within

both VHs is reported per mm2. GFP-positive staining was qualitatively assessed within the

ventral horn under 10x and 40x magnification.

49

2.9 Statistical Analysis

Data were analyzed utilizing either the GraphPad Prism (San Diego, CA) or the Statistica (Tulsa,

OK) software platform. Differences in cycle threshold values between sourced animals were

assessed with the Kruskal-Wallis non-parametric analysis followed by Dunn’s Multiple

Comparison test.

Differences in survival and onset of phenotype were assessed by the log-rank (Mantel-Cox) test.

All animals euthanized due to non-ALS pathologies (e.g. as a result of other complications

including severe dermatitis, abnormal growths etc.) were excluded from survival and histological

analyses (Scott et al. 2008).

Behavioral assays (LE, WH) and animal weight were assessed at each time point via one-way

ANOVA/ Kruskal-Wallis and reported as mean +/- standard error of the mean (SEM).

Histological assays (Nissl-positive motor neuron counts, ChAT, GFAP, and Iba-1

quantification), were combined for both hemispheres and averaged across the number of tissue

sections and analyzed via one-way ANOVA or Kruskal-Wallis non-parametric if the condition

for normality was not satisfied.. Histological data are expressed as mean +/- standard deviation

(SD).

Motor neuron diameter differences for both healthy and atrophying cells were analyzed by one-

way ANOVA and the data reported as mean +/- SD. Differences in the distribution of cell

diameters was assessed via the Mann-Whitney test statistic.

50

Chapter 3: Exploiting Lentiviral Delivery Mechanisms: the Effects of

Targeting the Lumbar Spinal Cord for Progranulin Overexpression

Hypothesis:

Lentiviral-mediated upregulation of progranulin within motor neurons of the lumbar spinal cord

at a stage prior to the phenotypic onset of ALS symptoms would, (i), attenuate the progressive

degenerative changes that are typically experienced in animal models that over-express mutant

SOD1, and (ii), concurrently both mitigate neuronal loss and diminish evidence for excessive

gliosis within the spinal cord.

Specific Aims:

1.To determine the effect of progranulin upregulation on the behavioural indices and overall

…phenotypic disease presentation in mutant SOD1 animals

2. To assess the degree of neuronal viability and gliosis following progranulin upregulation once

…the degenerative phenotype has been fully established.

3. Investigate the effect of mSOD1 copy number on the presentation of an ALS phenotype in the

…transgenic animals used in this experiment

3.1 Results

In order to investigate the long-term neurotrophic properties of exogenously delivered

progranulin, transgenic mice carrying the G37R mutant human SOD1 allele were generated

alongside wild-type conspecifics. At approximately 3.5 months of age, each gastrocnemius

muscle received five 2µL injections containing a lentiviral preparation carrying a GFP or PGRN

51

cDNA construct. Multiple injections were employed so as to maximize the number of transduced

neurons. For short-term gene expression (up to 1 month), this strategy has been previously

employed by our research group to successfully target motor neurons within the lumbar region of

the spinal cord (Figures C.1, C.2). Following lentiviral delivery, pre-injection (i.e. baseline)

behavioural assays continued until the experimental endpoint where transgenic animals exhibited

a distinguishable ALS phenotype (on average, 15 months following viral administration), with

wild-type littermates euthanized shortly afterwards. The behavioural tests were utilized so as to

assess overt phenotypic differences in wild-type vs. transgenic mSOD1 animals and to discern

any therapeutic effect mediated by progranulin upregulation. Following tissue collection and

immunohistological processing, motor neuron viability and markers for neurodegeneration were

assessed to ascertain any positive effects related to transgene delivery. All neuropathological

and/or behavioural outcomes are summarized in Table 3.1, or Table 3.2 for male and female

animals, respectively.

52

Table 3.1 Summary of behavioral and neuropathological outcomes in male cohorts

WILD-TYPE mSOD1 G37R

GFP PGRN % difference GFP PGRN % difference Lentiviral delivery

Mean age (d) at injection 105 +/- 8.8 95 +/- 4.5 -9.6%** 106 +/- 7.9 99 +/- 7.2 -6.70%n.s. Combined n at injection 12 12 -- 11 13 --

Disease onset and lifespan

Mean age (d) at onset of ALS phenotype -- -- 357 +/- 66.5 335.8 +/- 75.4 -5.94% Mean disease (d) duration -- -- 242 +/- 77.0 267 +/- 51.0 +10.3%

Mean age at end-stage ALS -- -- 599 +/- 61.2 603 +/- 34.8 +0.7% N at experimental end stage/end point 9 11 -- 6 12 --

Onset analysis; median age (d) -- -- 344 315 -8.4%n.s. Survival analysis; median age (d) -- -- 609 607 -0.3%n.s.

Temporal phenotypic outcomes

Weight no significant temporal trend no significant temporal trend Wire Hang no significant temporal trend no significant temporal trend

Leg Extension no significant temporal trend no significant temporal trend

End-stage neuropathology (n of animals)

Overall motor neuron assessment in the lumbar spinal cord

L1-L2 19.7 +/- 7.3 (3) 26.9 +/- 7.3 (5) +36.5%n.s. 11.9 +/- 2.7 (4) 13.3 +/- 3.3 (6) +11.8%n.s. L3-L5 29.8 +/- 5.2 (5) 29.9 +/- 2.4 (7) +1.4%n.s. 12.3 +/- 4.5 (4) 15.4 +/- 3.0 (7) +25.2%n.s.

L6 23.1 +/- 6.7 (4) 24.6 +/- 4.3 (6) +6.5%n.s. 18.0 +/- 2.6 (3) 16.9 +/- 2.5 (7) -6.1%n.s.

Targeted end-stage assessment of lumbar spinal cord region L3-L5

Motor neuron morphology

Diameter of healthy neurons 21.5 +/- 0.4 (3) 24.0 +/- 1.3 (3) +11.6% 17.2 +/- 1.5 (3) 15.7 +/- 0.4 (3) -8.9% Diameter of atrophying neurons 15.8 +/- 1.8 (3) 16.5 +/- 0.8 (3) +4.8% 13.0 +/- 0.5 (3) 12.1 +/- 0.4 (3) +7.0%

Cumulative distribution function; median diameter healthy neurons 20.94 22.78 +8.8%* 16.36 14.86 -9.2%n.s.

53

WILD-TYPE mSOD1 G37R

GFP PGRN % difference GFP PGRN % difference Cumulative distribution function;

median diameter atrophying neurons 14.07 14.97 +6.4%n.s. 11.96 11.66 -2.5%n.s.

Immunohistology

Surviving ChAT +ve cholinergic neurons 8.7 +/- 3.4 (5) 10.5 +/- 4.3 (6) +20.7%n.s. 5.1 +/- 1.3 (5) 4.3 +/- 1.8 (6) -15.7%n.s.

GFAP: degree of astroglial proliferation/mm2 96.9 +/- 57.4 (4) 64.5 +/- 26.7 (7) -33.4%n.s. 463.5 +/- 124.4 (5) 472.1 +/- 65.5 +1.9%n.s.

Iba-1: degree of microglial proliferation/mm2 248.0 +/- 52.8 (6) 251.5 +/- 100.4 (7) +1.4%n.s. 664.1 +/- 77.9 (5)

690.0 +/- 168.2 (7) +3.9%n.s.

Quantitative genotyping of the mSOD1

gene locus (n of animals)

Signal 2^ΔCT -- -- -- 12.6 +/- 2.8 (11) 12.6 +/- 2.4 (13) 0%n.s.

Comparisons made between PGRN and GFP cohorts within wild-type or mSOD1 mice. n.s. non-significant;*p<0.05; **p<0.01

54

Table 3.2 Summary of behavioural outcomes in female cohorts

WILD-TYPE mSOD1 G37R GFP PGRN % difference GFP PGRN % difference

Lentiviral delivery

Mean age (d) at injection 109 +/- 6.2 101 +/- 4.8 -7.6%** 104 +/- 8.2 101 +/- 3.8 -2.00%n.s. Combined n at injection 12 12 -- 11 12 --

Disease onset and lifespan

Mean age (d) at onset of ALS phenotype -- -- 274 +/- 90.5 294 +/- 27.7 +7.3% Mean disease (d) duration -- -- 231 +/- 33.3 268 +/- 36.5 +16.0%

Mean age at end-stage ALS -- -- 505 +/- 109.8 563 +/- 19.6 +11.5% Number of animals at experimental end stage/end point 10 8 -- 9 9 --

Onset analysis; median age (d) -- -- 287 294 +2.4%n.s. Survival analysis; median age (d) -- -- 532 573 +7.71%n.s.

Temporal phenotypic outcomes

Weight no significant temporal trend no significant temporal trend Wire Hang no significant temporal trend no significant temporal trend

Leg Extension no significant temporal trend no significant temporal trend

Quantitative genotyping of the mSOD1 gene locus (n of animals)

Signal 2^ΔCT -- -- -- 13.1+/-8.1(11) 11.6+/-3.8(12) -11.50%n.s.

Comparisons made between PGRN and GFP cohorts within wild-type or mSOD1 mice. n.s. non-significant;*p<0.05; **p<0.01

55

3.1.1 Overall Phenotypic Outcomes

3.1.1.1 Body Weight of Mice

Measures of body weight commencing at baseline showed a steady increase until disease-

associated atrophy in transgenic mSOD1 animals bifurcated the curve and showed a steady

decline in overall body mass (Figure 3.1A and B, significant deviations from wild-type animals

are denoted by a solid or dotted line). Compared to the female cohort, transgenic male animals

initiated the prototypical decrease in weight somewhat later (69 vs. 61 weeks of age). For male

animals stratified according to genotype, there were no measured weight differences between

either treatment group at any time point as measured by one-way ANOVA (or Kruskal-Wallis if

test for equal variances failed). Repeated measures ANOVA on male G37R subjects excluding

the final time point, did not denote a significant effect due to treatment [F(1,4)= 0.91, p=0.40],

nor was there a significant interaction between time and treatment [F(39, 156)= 0.70, p=0.90].

Similarly, in wild-type males (excluding animals that prematurely died; 3 from the GFP group

and 1 from the PGRN), there was no significant treatment effect [F(1, 9= 0.46, p=0.52], or an

interaction between time and treatment [F(41, 369)= 0.23, p=1.0] as analyzed by repeated

measures ANOVA.

Similar to male subjects, female animals organized according to genotype did not show a

significant effect between PGRN and GFP groups at any time point. Repeated measures

ANOVA on female G37R subjects (censoring one G37R-PGRN subject euthanized after trial

30), did not denote a significant effect due to treatment [F(1,1)= 51.1, p=0.09, nor was there a

significant interaction between time and treatment [F(37, 37)= 1.25, p=0.25]. Similarly, in wild-

type females (excluding animals that prematurely died; 1 from the GFP group and 2 from the

56

PGRN group), repeated measures ANOVA established no significant treatment effect [F(1, 8)=

0.57, p=0.47]. However, a significant interaction effect between time and treatment was found

between wild-type females treated with PGRN or GFP [F(41, 328)= 1.57, p=0.03]. Overall, this

measure did not show a positive effect of PGRN transgene delivery on the phenotype-associated

loss of body mass.

57

0 9 10 11 12 13 18 18 19 21 23 53 61

57 64 65 69 67 71 73 75 78 81 79 83 86 89 87 Trial Age (w)

(A)

(B)

0 9 10 11 12 13 18 18 19 21 23 53 61

57 64 65 69 67 71 73 75 78 81 79 83 Trial Age (w)

58

Figure 3.1 Temporal animal weight changes in transgenic mSOD1 and wild-type mice.

(A) WT and G37R male animals demonstrated a temporally progressive increase in body weight from measures initiating at baseline. At approximately 69 weeks of age (Trial 31), ALS-phenotype degenerative changes initiated a steady decline in body mass within the G37R cohort of animals. Within either genotype, no demonstrable significant difference in outcome was mediated by early PGRN administration (One-way ANOVA/Kruskal-Wallis/Repeated measures ANOVA). (B) WT and G37R female animals demonstrated a temporally progressive increase in body weight from measures initiating at baseline. At approximately week 61 of age (Trial 27), ALS-phenotype degenerative changes initiated a steady decline in body mass within the G37R cohort of animals, which was distinguishable from WT counterparts from 69-71 weeks of age. Within either genotype, no demonstrable significant difference in outcome was mediated by early PGRN administration (One-way ANOVA/Kruskal-Wallis/Repeated measures ANOVA). Solid line: Significant difference between G37R-GFP and at least one WT counterpart; Dotted line: Significant difference between G37R-PGRN and at least one WT counterpart. Mean +/- SEM shown.

3.1.1.2 Temporal Changes in Measured Behavioural Indices

3.1.1.3 Latency to Fall from an Elevated Grid

When inverted, the latency to fall is taken to be a measure of neuromuscular strength. Ideally, the

test would be expected to clearly differentiate between wildtype and mSOD1 animals as the

disease phenotype would ultimately lead to muscle atrophy that negatively affects performance.

In male animals, the data were highly variable and analysis at each time point (one-way

ANOVA/Kruskal-Wallis non-parametric analyses depending on the outcome of the test for equal

variances) failed to differentiate between both genotype and treatment group, primarily on

account of the large variation inherent to the recorded data (Figure 3.2A). Temporally, a decrease

in performance was evidenced within both genotypes. Repeated measures ANOVA restricted to

treatment groups within either male genotype did not differentiate a statistically significant

treatment effect, or an interaction between time and treatment [G37R: Treatment: F(1,4)= 1.96,

p=0.23, Time*Treatment: F(40, 160)= 1.37, p=0.09; WT: Treatment: F(1,13)= 0.05, p=0.83,

Time*Treatment: F(41, 533)= 1.11, p=0.30].

59

To a similar degree, data obtained from female subjects were highly variable. Unlike the male

cohort, female animals from either genotype did not demonstrate as robust a decline in

performance. Analysis at each time point (one-way ANOVA or Kruskal-Wallis non-parametric

analysis) started to differentiate between WT and transgenic mSOD1 animals at trial #33 (dotted

or solid line, Figure 3.2B) , with data from trial 36, the only point at which a significant

difference between PGRN and GFP treatment groups in the G37R cohort were established

(Figure 3.2B). This purported difference between G37R-PGRN and G37R-GFP groups at trial

#36 is an experimental artefact as the standard deviation in the GFP group was significantly

minimized on account of several animals reaching an experimental end point during the

preceding two weeks. A repeated measure ANOVA for the female cohorts was not performed on

account of a lack of variance in the measure at various time points. Taken together, PGRN

transgene delivery did not positively influence performance on the inverted wire hang test.

60

(A)

(B)

61

Figure 3.2 Latency to fall from an elevated grid in mSOD1 and wild-type mice over time

(A) Male animals from both WT and G37R cohorts demonstrated a temporal decline in their ability to adhere to an inverted grid, with PGRN administration mediating no positive effect overall (repeated measures ANOVA), or at any given time point (One-way ANOVA/Kruskal-Wallis). (B) WT and G37R female animals exhibited a marginally delayed onset in the latency to fall deficit compared to male counterparts. Time points nearing the experimental end point showed a significant defect in transgenic vs. wildtype animals (One-way ANOVA/Kruskal Wallis), however a temporal difference in the GFP vs PGRN cohorts were not established.Solid line: Significant difference between G37R-GFP and at least one WT counterpart; Dotted line: Significant difference between G37R-PGRN and at least one WT counterpart. Mean +/- SEM shown.

3.1.1.4 Leg Extension Reflex Score

The leg extension reflex test did not differentiate a temporally significant effect mediated by

PGRN administration (Figure 3.3). Male animals appeared to show a deficit in the leg extension

score at an earlier stage of the disease cascade compared to female counterparts. Due to the high

degree of variability in the data, at no point did male animals show a significant difference in the

leg extension reflex score, nor was the test sensitive enough to distinguish between wildtype and

transgenic mSOD1 animals (Figure 3.3A). For female conspecifics, the phenotypic effect of the

mutant SOD1 gene locus became apparent at the final stages of the disease cascade, however,

PGRN delivery did not mediate a positive effect on the outcome of the hind limb reflex test

(Figure 3.3B).

62

(A)

(B)

63

Figure 3.3. Leg extension reflex scores over time in both mSOD1 and wild-type animals

(A) In male animnals, the leg extension reflex score did not mediate a significnat difference between wildtype and transgenic groups. One-way ANOVA/Kruskal Wallis analysis at each timepoint did not show a significant treatment effect. (B) Significant differences between PGRN-treated female transgenic and wildtype animals were established at later timepoints in conjunction with a significant difference between PGRN- and GFP-treated transgenics at the final two measures. Dotted line: Significant difference between G37R-PGRN and at least one WT counterpart. Mean +/- SEM shown.

3.1.1.5 Overall Lifespan and Disease Onset Comparisons Between mSOD1 Groups

The experimental endpoint for transgenic mSOD1 animals was taken to be the time point at

which the animal was exhibiting severe hind limb paresis with the inability to self-correct once

placed on its side (Section 2.6). Onset of an ALS-like phenotype was retroactively determined to

be the time at which the animal reached its peak body weight, prior to the progressive decline

that is prototypical of the ALS model (Section 2.5). Body weight as a measure of disease onset

was utilized over functional scores (WH or LE) since it provided the earliest and most robust

measure distinguishing between wild-type and mSOD1 conspecifics. Female G37R-PGRN

animals showed a slight delay in median onset (294 vs 287d) and an increase in survivability

(573 vs 532d) compared to GFP-treated counterparts.

Overall, compared to female animals, male progeny exhibited a further delayed onset and a more

attenuated rate of deterioration. Unlike the female cohort, G37R-PGRN male animals presented

with disease onset somewhat earlier (315 vs 343d) and showed a similar median survival time

(607 vs 609d) compared to the G37R-GFP group. In either male or female cohort, the PGRN

treated group exhibited an non-significant increase in median disease duration compared to the

GFP cohort (male animals: 26.5d; female animals: 34.5d).

64

The Kaplan-Meier curves show that the median time of both onset and survival in either male or

female G37R animals does not differ significantly between the GFP and PGRN treated cohorts

(Figure 3.4). Mantel-Cox log-rank test comparing G37R-PGRN and G37R-GFP groups within

either sex did not indicate statistically significant differences in survival or disease onset. For

male transgenic conspecifics, the survival time statistic was χ2 =0.00175, df=1, p = 0.967, while

those for onset analyses was χ2 = 0.328, df=1, p = 0.567. Similarly, for female transgenic

counterparts, the survival time statistic was χ2 =0.01221, df=1, p = 0.912, while those for onset

was χ2 = 0.5145, df=1, p = 0.473. Thus, regardless of gender in transgenic mSOD1 animals,

PGRN construct delivery did not statistically influence the overall survival time or delay the

onset of symptoms in comparison with the GFP-treated cohort.

65

Figure 3.4 Comparisons between disease onset and survival in transgenic mSOD1 animals

(A) Kaplan-Meier graphs depicting the onset and overall lifespan in transgenic G37R male and (B) female animals in GFP and PGRN treated cohorts. Log-rank (Mantel-Cox) analyses did not show a significant difference in onset or survival between PGRN and GFP cohorts in either male or female animals.Male G37R-GFP median onset(d): 344, median survival(d): 609; G37R-PGRN: 315, 607. Female G37R-GFP median onset (d): 287, median survival (d): 532; G37R-PGRN: 294, 573

(A)

(B)

66

3.1.2 Histological Analyses

3.1.2.1 Neuronal Assessment throughout the Lumbar Spinal Cord

Cresyl violet staining of the neuronal nissl substance throughout the lumbar spinal cord in male

animals was employed to assess overall neuronal viability following transgene delivery. In

addition to assessing neuronal counts in the lentiviral-targeted L3-L5 region, we attempted to

gauge any off-target effects on account of the secretory properties of PGRN by computing

neuronal counts in the L1-L2 and L6 segments of the spinal cord.

Image analysis software (CellProfilerTM) identified and counted cells within the ventral horn in

serial 100µm-apart, 10µm-thick sections with an area of 130-950 um2, which were subsequently

manually confirmed by the experimenter. Kruskal-Wallis non-parametric analyses (followed by

Dunn’s Multiple Comparison Tests for post hoc analysis) were conducted on the basis of at least

one experimental group failing to pass the Kolmogorov-Smirnov or D’Agostino-Pearson

omnibus tests for normality.

In the L1-L2 region of the lumbar spinal cord, PGRN delivery into animals of either genotype

demonstrated a slight, but non-significant (p>0.05) increase in mean viable motor neurons

compared to their GFP counterparts (Figure 3.5; 37% increase in WT-PGRN (mean of 26.9 +/-

7.3 (SD) neurons) vs WT-GFP (19.7 +/- 7.3); 12% increase in G37R-PGRN (13.3 +/- 3.3) vs

G37R-GFP (11.9 +/- 2.7)).

When stratified according to genotype, in the targeted L3-L5 region, there were no significant

differences within either treatment group, however G37R-PGRN animals, on average, showed a

67

12% increase in viable motor neuron counts compared to G37R-GFP conspecifics (Figure 3.5;

WT-PGRN: average of 29.9 +/- 2.4 neurons) vs WT-GFP (29.8 +/- 5.2); G37R-PGRN (15.4 +/-

3.0) vs G37R-GFP (12.3 +/- 4.5)).

Similarly, in the L6 region of the lumbar spinal cord, comparisons between treatment groups

within the context of either genotype did not yield a statistically significant difference. In

contrast to the other segments within the spinal cord, G37R-PGRN treated animals showed

evidence of a slight decrease of viable motor neuron counts when compared to GFP-treated

counterparts (Figure 3.5; 7% decrease in the average number of neurons (+/- SD) in G37R-

PGRN (16.9 +/- 2.5) vs. G37R-GFP (18.0 +/- 2.6); 7% increase in WT-PGRN (24.6 +/- 4.3) vs.

WT-GFP (23.1+/-6.7)).

* * * * * *

68

Figure 3.5 Neuronal viability across regions of the lumbar spinal cord of male wild-type and transgenic mSOD1 animals. Motor neurons within the ventral horn restricted to lumbar segments L1-L6 were quantified with CellProfiler cell image analysis software. Ten micrometer–thick, serially sectioned, and 100µm-apart tissue sections were utilized for the analysis. Cresyl violet stained neuronal cells identified with an area between 130-950µm2 (a cellular diameter between 13-35µm) were included in the cell counts and averaged across all tissue sections from an animal in any given lumbar region. In comparison to the WT-GFP group, wild-type animals that received PGRN injections showed a trend for increased neuronal numbers in both L1-L2 and L6 regions (37% and 7%, respectively). For transgenic G37R animals receiving PGRN, regions L1-L2 and L3-L5 demonstrated an increase (12% and 25%) in survival numbers, while region L6 showed an overall 6% decrease in neuronal counts compared to the G37R-GFP cohort. Kruskal-Wallis non-parametric analysis followed by Dunn’s multiple comparisons did not compute any statistically significant differences (p>0.05) between treatment groups within the context of the WT or G37R genotype. Comparisons that were statistically significant within the specified regions between genotypes are as follows (* p<0.05): L1-L2 (WT-PGRN vs. G37R-GFP, p<0.05); L3-L5 (WT-GFP vs. G37R-GFP, p<0.05; WT-GFP vs. G37R-PGRN, p<0.05; WT-PGRN vs. G37R-GFP, p<0.05; WT-PGRN vs. G37R-PGRN, p<0.05); L6 (WT-PGRN vs. G37R-PGRN, p<0.05) ); n = number of animals in each group 3.1.2.2 Morphological Characteristics of Neurons within the L3-L5 Lumbar Spinal Cord

Segment

Nissl staining of L3-L5 lumbar spinal cord tissue sections facilitated the classification of

neuronal cell bodies as healthy or atrophying based on strict morphological characterizations as

set out in section 2.7.2.3. Briefly, healthy motor neurons were taken to be those that presented

with a distinctive nucleolus, cytoplasmic nissl granules, and a “motor neuron-like”

morphological shape with processes protruding from the cell body (Inset, Figure 3.6A).

Atrophying motor neurons appeared shrunken in appearance, lacked a visible nucleolus, and

typically presented with a darker cresyl violet staining pattern (Inset, Figure 3.6A). Qualitatively,

both healthy and atrophying neurons from transgenic G37R animals presented with smaller cell

bodies in comparison to wild-type mice. The cumulative distributions of cell diameter in both of

69

these neuronal sub-types were shifted to the left compared to wild-type conspecifics, which is

further underscored by the smaller mean size of both cell populations in the transgenic groups

(Figure 3.6A, B, Table 3.3). Comparisons of the cumulative distributions of transgenic mSOD1

animals did not differentiate a significant treatment effect on cell body size mediated by PGRN

treatment in either healthy or atrophying neurons (healthy neurons: Mann-Whitney U=1536,

n1=n2=3, p=0.0844, two-tailed; atrophying neurons: Mann-Whitney U=11700, n1= n2=3,

p=0.1935, two-tailed). For the healthy motor neuron sub-type in wild-type animals, PGRN

administration caused an overall shift towards cell bodies with larger diameters (Mann-Whitney

U=8379, n1= n2=3, p=0.029, two-tailed), an observation not recapitulated with atrophying motor

neurons (Mann-Whitney U=8568, n1= n2=3, p= 0.1069, two-tailed).

Overall changes in the frequency distribution of motor neuron sizes appeared consistent within

either genotype (Figure 3.6 C, D). Healthy motor neurons within either wild-type cohort

presented with larger soma sizes, while cells classified as being atrophied within the transgenic

sets showed an increase in frequency of cells with smaller diameters, as compared to wild-type

counterparts.

70

(A)

(B)

71

Figure 3.6 Characterization of neuronal morphology between treatment and genotype groups

(A) Cumulative distribution function illustrates the changes in the cell diameter of both healthy and atrophying motor neurons across all groups. Transgenic mSOD1 animals presented with cells of a smaller diameter in both neuronal subtypes, compared to

(C)

(D)

72

their wild-type counterparts. Progranulin administration caused a significant shift towards cell bodies with larger diameters in WT animals (p<0.05), however a similar effect was not established in the G37R cohort, nor did the exogenous delivery of PGRN significantly influence changes in the cell diameter of atrophying motor neurons. Inset: cresyl violet-stained 40x micrographs within the ventral horn depicting representative healthy (†) and atrophying ǂ neuronal cells. Scale bar = 20µm. (B) Mean size (+/-SD) of healthy (G37R-GFP: 17.2 +/- 1.5, G37R-PGRN: 15.7 +/- 0.4; WT-GFP: 21.5+/- 0.4, WT-PGRN 24.0+/- 1.3) and atrophying (G37R-GFP: 13.0+/-0.5, G37R-PGRN: 12.1 +/- 0.4; WT-GFP: 15.8+/-1.8, WT-PGRN 16.5 +/- 0.8) neurons. (C) Changes in the frequency distribution of healthy motor neuron sizes showed that neurons from G37R animals were generally skewed towards smaller diameters in comparison to WT counterparts. (D) Changes in the frequency distribution of atrophying motor neuron sizes showed that neurons from G37R animals were skewed towards smaller diameters in comparison to WT counterparts.

Table 3.3 Mean diameter of healthy and atrophying motor neurons across genotype and treatment groups

Healthy neuronal diameters

AVG SD # of

animals # of neurons

counted Difference between

medians1 WT-GFP 21.5 0.4 3 129

WT-PGRN 24.0 1.3 3 153 +1.845 G37R-GFP 17.2 1.5 3 56

G37R-PGRN 15.7 0.4 3 67 -1.498

Atrophying neuronal diameters

AVG SD # of animals

# of neurons counted

Difference between medians1

WT-GFP 15.8 1.8 3 144 WT-PGRN 16.5 0.8 3 134 +0.8975

G37R-GFP 13.0 0.5 3 153 G37R-PGRN 12.1 0.4 3 167 -0.03

1Comparing treatment groups within either genotype

73

3.1.2.3 Probing for Viable Cholinergic Neurons within the L3-L5 Lumbar Spinal Cord

Segment

Motor neurons considered cholinergic were identified with a polyclonal anti-ChAT antibody in

male animals. Within the L3-L5 lumbar spinal cord region, motor neurons restricted to the

ventral horn were tallied under 10x magnification. When compared to wild-type counterparts,

transgenic animals showed evidence of a more than 50% reduction in ChAT positive motor

neurons; recapitulating the finding of neuronal viability assessed via cresyl violet in the L3-L5

region (Section 3.1.2.1; Figure 3.5). Overall, in wild-type animals, PGRN administration

mediated a non-significant 20.7% increase in neuronal counts, while paradoxically contributing

to a 15.7% decrease in mSOD1 cholinergic neurons (Figure 3.7A).

A one-way ANOVA was conducted to compare the effect of transgene delivery on cholinergic

neuronal viability. There was a significant effect of exogenous gene delivery on motor neuron

counts between groups [F(3, 18)=5.585, p=0.0069]. However, post hoc comparisons using the

Tukey multiple comparisons test indicated that there was no statistically significant effect

mediated by PGRN administration within either genotype when compared to the respective GFP

cohort (WT-GFP µ= 8.7 +/- 3.4, n=5; WT-PGRN µ= 10.5 +/- 4.3, n=6; G37R-GFP µ= 5.1 +/-

1.3, n=5; G37R-PGRN µ= 4.3+/-1.8, n=6). Comparisons between WT-PGRN and either

treatment group within the G37R cohort were significant at the p<0.05 level. Taken together,

these results suggest that PGRN transgene delivery early on in the mSOD1 disease cascade does

not mitigate cholinergic neuron loss at the neurodegenerative end stage.

74

3.1.2.4 End-stage Glial Cell Changes within the L3-L5 Lumbar Spinal Cord Segment

A critical hallmark of ALS pathogenesis in transgenic mSOD1 models is the proliferation of glial

cells surrounding motor neurons as the overt phenotype progressively deteriorates.

Immunohistologically assessing the degree of microglial or astrocytic proliferation was

conducted so as to establish whether the initial PGRN delivery would attenuate the expected

prototypical neuroinflammatory cascade.

3.1.2.4.1 Astrogliosis

Antibodies directed against the glial fibrillary acidic protein (GFAP) identified astrocytes within

the ventral horn of tissue sections restricted to the L3-L5 region of the spinal cord from male

animals. Images from the ventral horn were captured at 40x, and with the aid of a superimposed

counting grid, positive cells were tabulated. As expected, transgenic mSOD1 animals showed

more than a 500% increase in positive GFAP immunolabeling in comparison to wild-type

littermates (Figure 3.7B).

A one-way ANOVA was conducted to compare the effect of transgene delivery on astroglial

proliferation. There was a significant effect between the genotypes of various treatment groups

[F(3, 19)=55.12, p<0.001]. Post hoc comparisons using the Tukey multiple comparisons test

indicated that the mean number (+/- SD) of positive GFAP cells/mm2 were significantly different

(p<0.001) between WT-GFP (96.9 +/- 57.4) and G37R-GFP (463.5 +/- 124.4) or G37R-PGRN

(472.1+/-65.5), as well as between WT-PGRN (64.5+/-26.7) and G37R-GFP (463.5 +/- 124.4) or

G37R-PGRN (472.1+/-65.5). However, there was no statistically significant effect mediated by

PGRN administration within either genotype when compared to the respective GFP cohort.

75

These results suggest that early PGRN targeting of the lumbar spinal cord does not mitigate the

underlying proliferation of astrocytes that are eventually established with the progressively

deteriorating disease cascade.

3.1.2.4.2 Microgliosis

As with changes in the distribution of astrocytes above, the effect of PGRN on the degree of

microglial proliferation was assessed in the ventral horn of tissue sections mapped to the L3-L5

region of the spinal cord. Transgenic male mSOD1 animals showed more than a 250% increase

in positive Iba-1 immunolabeling in comparison to wild-type littermates (Figure 3.7C). One-way

ANOVA was conducted to compare the effect of transgene delivery on microglial proliferation.

There was a significant effect between the genotypes of various treatment groups [F(3,

21)=30.07, p<0.001]. Post hoc comparisons using the Tukey multiple comparisons test indicated

that the mean number (+/- SD) of positive Iba-1 cells/mm2 were significantly different (p<0.001)

between WT-GFP (248.0 +/- 52.8) and G37R-GFP (664.1 +/- 77.9) or G37R-PGRN (690.0 +/-

168.2), as well as between WT-PGRN (251.5 +/-100.4) and G37R-GFP (664.1 +/- 77.9) or

G37R-PGRN (690.0 +/- 168.2). However, there was no statistically significant effect mediated

by PGRN administration within either genotype when compared to the respective GFP cohort

(p>0.05). These results suggest that early PGRN targeting of the lumbar spinal cord does not

mitigate the underlying microglial proliferation that is typically associated with the advanced

stages of disease presentation in murine ALS models.

76

(A)

(B)

(C)

* *

** ** **

**

** ** **

**

+20.7%

-15.7%

-33.4%

+1.9%

+1.4%

+3.9%

77

Figure 3.7 Immunohistological assays within the ventral horn of tissues from the L3-L5 region assessing phenotype severity and the effect of transgene delivery

(A) ChAT quantification of motor neurons did not distinguish a significant positive effect of PGRN treatment in either WT or G37R cohorts. WT-GFP: 8.7 +/- 3.4; WT-PGRN: 10.5 +/- 4.3; G37R-GFP: 5.1 +/- 1.3; G37R-PGRN: 4.3 +/- 1.8. (B) Astroglial and (C) microglial proliferation measured via GFAP or Iba-1 immunoreactivity, respectively, was unaffected by the early PGRN intervention within intra-genotype comparisons. Degree of GFAP immunoreactivity per mm2 for each of the groups was as follows: WT-GFP: 96.9 +/- 57.4; WT-PGRN: 64.5 +/- 26.7; G37R-GFP: 463.5 +/- 124.4; G37R-PGRN: 472.1 +/- 65.5. Microglial proliferation as measured by Iba-1 reactivity per mm2: WT-GFP: 248.0 +/- 52.8; WT-PGRN: 251.5 +/- 100.4; G37R-GFP: 664.1 +/- 77.9; G37R-PGRN: 690.0 +/- 168.2. Graphs represent mean +/- SD, significant results from one-way ANOVA followed by Tukey’s post hoc assessment denoted as *p<0.05, **p<0.001. Images taken at 100x, scale bar = 10µm.

3.1.3 Transgenic Progeny Derived from Commercial Breeders Showed a Reduction in

Mutant Human SOD1 Copy Number and a Concomitant Increase in Overall Lifespan4 7

G37R transgenic mSOD1 animals originally derived from line 29 were selected in this study to

assess the therapeutic efficacy of growth factor upregulation in mice that did not deteriorate as

rapidly as the prototypical G93A model animals. In contrast to the expected phenotypic disease

course , the delayed onset and attenuated disease progression in the transgenic cohort generated

for this study warranted further investigation. To this aim, we here compare these transgenic

mice to- an independently derived colony of G37R(29) animals from breeders provided by a

local collaborator.

3.1.3.1 ΔCT Differences in Progeny Derived from Commercial vs. Collaborator Breeders

Progeny for this study were generated from breeders obtained commercially through the Jackson

Laboratory. Based on historical control data, these breeders were deemed to maintain the G37R

locus at comparable levels from when the colony was initially established at the commercial

4A version of section 3.1.3 has been previously published in Zwiegers et al, 2014

78

facility (Figure E.1). Transgenic conspecifics derived from these breeders demonstrated a mean

signal (2ΔCT) of 12.6 and 12.3 for male and female animals, respectively that was relatively

tightly clustered (Figure 3.8). In contrast, a cohort of G37R(line 29) animals previously derived

in our laboratory from breeder sets provided by a local collaborator, showed a near three-fold

increase in relative hSOD1 copy numbers (33.5, male; 34.0, female; Figure 3.8). Kruskal-Wallis

non-parametric analysis followed by Dunn’s Multiple Comparison Test determined that

differences in the relative abundance of the transgene locus between sexes within either the

commercial- or collaborator-derived progeny group were not statistically significant (p>0.05).

However, both intra- and inter-sex comparisons between the two progeny groups revealed a

significant difference in hSOD1 transgene copy number, with the commercially-derived progeny

presenting with a significant reduction in the hSOD1 copy number (p<0.01 or P<0.001; Figure

3.8)

*** ***

** ***

79

Figure 3.8 Comparisons of quantitative genotyping for the mutant hSOD1 locus in two independently-derived colonies. Progeny derived from commercial breeders exhibited a near three-fold decrease in the genomic G37R signature compared to collaborator-derived offspring. Results are presented as the relative difference between the PCR cycle threshold (ΔCT) values of the housekeeping (c-JUN) and target (mutant SOD1) genes, raised to the second power. Kruskal-Wallis non-parametric analysis followed by Dunn’s Multiple Comparison test, indicated significant inter-source differences (p < 0.001, 0.01) while failing to differentiate loci heterogeneity from within each source. Collaborator- ♂ vs. Commercial-♂ (p < 0.001); Collaborator- ♂ vs. Commercial-♀ (p < 0.001); Collaborator- ♀ vs. Commercial-♂ (p < 0.01); Collaborator- ♀ vs. Commercial-♀ (p < 0.001). 3.1.3.2 ΔCT Comparisons Stratified According to Sex and Treatment Group

There was no significant difference in the levels of the hSOD1 transgene across the GFP or

PGRN treatment groups, within either sex, as measured by one-way ANOVA ([F(3,43) =

0.2081, p= 0.8903] Figure 3.9). Thus, transgenic male and female animals receiving the lentiviral

administration carried similar copy numbers for the hSOD1 locus, and all are expected to express

the ALS phenotype to a similar degree (Male: 12.6 +/- 2.8[GFP]; 12.6 +/- 2.4 [PGRN]; Female:

13.1 +/- 8.1 [GFP]; 11.6 +/- 3.8 [PGRN].

80

Figure 3.9 Comparisons of quantitative genotyping for the mutant hSOD1 locus across sex and treatment group. Stratification of qPCR results from commercially-derived progeny according to sex and viral construct delivered. Results are presented as the relative difference between the PCR cycle threshold (ΔCT) values of the housekeeping (c-JUN) and target (mutant SOD1) genes, raised to the second power. No significant inter-group differences were distinguishable via one-way ANOVA.

3.1.3.3 Lifespan is Negatively Correlated with the Degree of mSOD1 Transgene Presence

With the temporally progressive nature of the degenerative changes that are established in

transgenic mSOD1 animals, significant hind limb paresis warrants humane termination of

experimental subjects. The age of phenotypic onset and degree of longevity can be taken as a

measure of disease severity. Comparisons of the survival curves between collaborator- and

commercially-derived progeny indicate that the relative amount of mSOD1 within the genome

can have a significant effect on the overall presentation of the ALS phenotype. The Kaplan-

Meier survival curve shows that the median survival time for the commercially-derived G37R

progeny is significantly longer than that of the collaborator-derived colony, with an overall 46%

increase in the average lifespan (Figure 3.10A). The median survival times for commercially-

derived progeny were 603 and 560 days for males and females, respectively. Collaborator-

derived animals experienced a shorter survival time, with male animals surviving for a median of

393 days, and females for 394 days. Intra-sex comparisons with Mantel-Cox statistical analysis

showed that commercially-derived progeny exhibited a statistically significant increase in overall

lifespan. Mantel-Cox statistics comparing males from both colonies were χ2 =34.57, df=1,

p<0.0001, while those for female transgenics were χ2 = 18.6, df=1, p<0.0001.

81

Linear regression analysis of pooled data (Figure 3.10B) demonstrates a slight negative correlation between the overall lifespan of

research subjects and the relative abundance of the mutant hSOD1 locus (Y = -0.07449*X + 56.09; r2=0.4427, p= 0.0001).

(A)

B)

82

Figure 3.10 Effect of mSOD1 transgene dosage on lifespan of transgenic animals in two independently-derived colonies.

(A) Kaplan-Meier survival analysis. Transgenic animals with a lower CT signal value exhibited an increased survival to the humane experimental endpoint The Mantel-Cox log-rank test indicated a significant difference between animal source and lifespan. Inter-sex comparisons show that collaborator-derived progeny with a higher CT signal value reached the endpoint (inability to self-correct) after a median age of 393 days, while commercially-derived progeny with a lower CT signal value reached the endpoint much later, after a median age of 583 days. *** p<0.001. Collaborator- ♂ n= 8; Collaborator- ♀ n= 5; Commercial-♂ n= 18; Commercial-♀ n= 17. The table insert summarizes all statistical comparisons. B) Linear regression analysis reveals a negative correlation (r2=0.4427) between relative mutant SOD1 copy number and lifespan (significant, non-zero slope, p<0.0001; male and female animals from either cohort are combined). Collaborator n=11; Commercial n=35

3.1.4 End-stage Probing for Lentiviral Transduction Efficiency and Expression

Outsourced automated quantitative PCR (data not shown) on tissue sections from the targeted

L3-L5 region for either the GFP or PGRN sequence, failed to yield positive results compared to

negative controls (n=3 per group). Similarly, assessing immunoreactivity for the GFP gene

product in the L3-L5 region did not demonstrate any degree of GFP staining (data not shown;

n=3 per GFP group).

3.2 Discussion of Experimental Findings

3.2.1 Decreased mSOD1 Copy Number of the G37R Locus Delays Disease Onset and

Extends Lifespan in Transgenic Animals5 8

The transgenic animals utilized in this study were derived from a colony of commercially-

obtained breeders that had most likely experienced an uncharacterized and unreported drop in the

mSOD1 gene locus when the colony was originally deposited into the repository (The Jackson

5Derived from work published in Zwiegers et al., 2014

83

Laboratory, personal communication). As explored in section 1.5.1, once a transgenic line is

established, fluctuations in transgene copy number can ultimately modify the phenotypic

presentation of the expected disease. Rare intraloci cross-over events during meiotic

recombination could act to further expand or retract the relative abundance of a particular locus

of interest. Within the G93A model, transgene copy number reduction is relatively rare and

estimated to occur in 0.1-3% of the generated progeny (Gurney 1997; Alexander et al. 2004).

Theoretically, a breeding strategy employing multiple transgenic breeder animals as described in

section 2.1.1.1 would limit the chance occurrence of a reduction in copy number in the resultant

progeny. Thus, given the relatively tight clustering of the mSOD1 signal (2ΔCT) within the

commercially-derived progeny used in this study (male (mean+/-SD): 12.6 +/-2.5; female:

12.3+/-6.1; Figure 3.8), it implicates a drop in copy number in the original commercial breeder

colony as being causal to the attenuated progression of the diseased phenotype in the generated

offspring. Furthermore, as illustrated in figure E.1, the male breeder animals employed in this

study exhibited comparative levels of the mSOD1 transgene when compared to historical control

samples when the colony was first established in the Jax repository. This further supports the

notion of an uncharacterized drop in the mSOD1 copy number when the breeding colony was

established at the repository, since successive progeny derived from the this colony all exhibited

similar ΔCT values.

The degree of mSOD1 transgene presence does to a large degree correspond to the amount of

mutant gene product produced and the resultant manifestation of the diseased phenotype.

Typically, G37R animals derived from line 29 exhibit a mean lifespan of around 350-380 days

(Wong et al. 1995; Nguyen et al. 2001; Ezzi et al. 2010). Male and female transgenic animals

84

generated for lentiviral delivery in this study exhibited an increased lifespan (Figure 3.10);

surviving for a mean of 602 days (range: 484-670) and 536 (days; range: 273-624), respectively.

Although animals from both sexes exhibited the same relative level of the mSOD1 transgene

(Figure 3.9), the finding that female transgenic animals showcased a steeper rate of decline and

reached the humane experimental endpoint prior to male counterparts is an observation not

corroborated in the research literature. Typically, female mSOD1 animals survive longer than

their male counterparts (Alexander et al. 2004; Cervetto et al. 2013). Animals exhibiting similar

survival outcomes as described here were found in subjects with a reduced copy number from a

colony of G93A mutant SOD1 animals. Animals carrying 4 copies of the G93A locus presented

with a delay in disease onset and ultimately warranted end-stage euthanasia at 625 +/- 60.8 days

(Alexander et al. 2004).

When compared to a previous colony of transgenic G37R(line 29) animals that were derived

from a local collaborator and more closely mimicked the prototypical disease course, the newly

generated animals presented with a near 3-fold drop in copy number and showed a significant

46% increase in mean overall lifespan (Figure 3.10). Ultimately, this gave rise to transgenic

progeny exhibiting a milder presentation of the ALS phenotype with a delayed onset and a

concomitant attenuated progression of the disease resulting in an extended survival time. This

unexpected outcome provided an opportunity to assess the lentiviral-mediated upregulation of

progranulin in the context of an ALS model presenting with a less severe phenotypic

manifestation of the disease. However, a possible caveat to this outcome lies in the fact that the

drop in copy number allows for a more prolonged progression of the aberrant ALS phenotype.

Targeted lentiviral administration into vulnerable neuronal populations at an early phase of the

85

progressive disease cascade may negatively affect the stable expression of exogenous cDNA,

since these targeted cells may degenerate as the phenotype aggressively worsens

3.2.2 Early-Stage Lentiviral-mediated Progranulin Delivery Does Not Attenuate the

Onset or Progression of an ALS-like Phenotype in mSOD1 Transgenic Mice

Stratification of animals based on sex and treatment group revealed that subjects used in the

current study presented with similar relative mSOD1 transgene levels (12.4 +/- 4.6, all cohorts

combined; Figure 3.9). Animals would thus not be predicted to present with a significant

variation in the expected ALS phenotype between treatment groups within each sex due to

differences in mSOD1 transgene levels. Any difference in the onset, overall survivability, and

markers of neurodegeneration would thus be consequent to the effects mediated by the early-

stage lentiviral delivery of exogenous progranulin cDNA.

Early lentiviral-mediated PGRN upregulation targeting lumbar spinal cord motor neurons prior

to the phenotypic onset was hypothesized to mitigate the prototypical disease cascade; extending

survival and dampening the neurodegenerative response. Under the experimental conditions

employed in this study with an animal model exhibiting a less severe ALS phenotype, we were

unable to demonstrate the upregulation of exogenous cDNA at end-stage and concomitantly

failed to validate a positive outcome in any behavioural or neuropathological measure.

86

3.2.2.1 Progranulin Transgene Delivery Does Not Delay Disease Onset, Alter

Behavioural Phenotypic Abnormalities, or Extend Survival in mSOD1 Animals when

Targeted at an Early Stage of the Disease Cascade.

Intra-sex comparison of treatment groups did not indicate a significant effect mediated by early-

stage PGRN cDNA targeting of lumbar spinal cord motor neurons. Male transgenic animals

(Figure 3.4A) showed a similar overall survivability between treatment groups (mean+/-SD;

PGRN: 603 +/- 34.8; GFP: 599 +/- 61.2), even though animals transfected with PGRN cDNA

presented with an onset of disease 20 days prior to the GFP cohort (PGRN: 335.8 +/- 75.4; GFP:

356.5 +/- 66.5). In contrast, female transgenic animals (Figure 3.4B) transfected with PGRN

cDNA showed a non-significant delay in disease onset by 20 days (PGRN: 294 +/- 27.7; GFP

274 +/- 90.5) and a concomitant increase in mean lifespan by 58 days (PGRN: 563.11 +/- 19.6;

GFP: 505 +/- 109.8). Similarly, due to the high degree of variability in the data, behavioural

measures as assessed by the latency to fall from an inverted grid or the leg extension reflex score

failed to discern a positive effect on the ALS phenotype between GFP and PGRN treated groups

(Figure 3.2Figure 3.3).

In the context of ALS model systems, these negative findings are corroborated with other

investigations exploring the potential application of upregulating the neurotrophic factor as a

putative therapeutic agent in attenuating disease presentation. Transgenic G93A mSOD1 animals

cross-bred to constitutively over-express human PGRN in all cell types presented with increased

PGRN mRNA and protein levels within the CNS; however, this did not attenuate disease onset

nor positively influence overall survival rates (Herdewyn et al. 2013). Similarly, a

complementary strategy exploiting the sustained intracerebroventricular (ICV) delivery of

87

recombinant human PGRN to G93A rats prior to disease onset, failed to delay the onset of motor

symptoms or extend survival time (Herdewyn et al. 2013). Although exogenous PGRN delivery

in these studies did not further exacerbate the degenerative cascade, potential limitations of these

investigations do limit our understanding regarding the neurotrophic properties of the growth

factor in the context of ALS. Firstly, with the widespread upregulation of human PGRN in all

cell types (most notably glial cells), the degree of proteolytic processing into the various granulin

domains was not assessed, and it is thus unknown as to how effective this strategy is in

promoting a neuroprotective response in mSOD1 motor neurons. Secondly, neuronal uptake of

ICV-mediated delivery of the recombinant protein was not confirmed, nor was neuronal viability

or markers of neurodegeneration assessed within the context of either treatment strategy. Thirdly,

therapeutic efficacy may be inhibited within the context of the severe neurodegenerative insult

mediated by the highly expressed G93A SOD1 locus.

However, it remains a possibility that targeting mSOD1-mediated neuronal degenerative

processes with increased progranulin levels may not be a viable treatment strategy for SOD1-

fALS. In a zebrafish model system of mutant TDP-43 induced axonopathy, overexpression of

human PGRN was shown to both increase axonal outgrowth length, as well as decrease the

aberrant axonal branching induced by mutant TDP-43 expression (Laird et al. 2010). Utilizing

the same model system, Laird et al showed that mutant human SOD1 (A4V mutation)

overexpression induced an aberrant axonopathy, however, PGRN upregulation did not mediate a

rescue of the defective phenotype (2010). Taken together, this further suggests that the

mechanistic cascade underlying mutant TDP-43 and SOD1 neurodegeneration do not converge

on a common pathway (Herdewyn et al. 2013). Exogenous PGRN upregulation may thus only

88

mediate a positive effect in TDP-43 proteinopathies. With the finding that a large proportion of

sALS patients exhibit extranuclear translocation of TDP-43 (Mackenzie et al. 2007) and that

fALS cases can be linked to mutations in the TDP-43 gene locus (Harms & Baloh 2013), PGRN

upregulation may be of some benefit in these non-SOD1 cases of ALS.

3.2.2.2 Early Progranulin Delivery Targeting the Lumbar Spinal Cord Does Not

Mitigate Neuronal Degeneration or Neuroinflammatory Processes at the Diseased End-

stage

On further consideration, the lack of positive outcomes in the behavioural measures and overall

survivability are likely an outcome of the experimental paradigm utilized. Lentiviral-mediated

neuronal targeting under the parameters described in section 2.4, would limit expression within a

subset of motor neurons in the lumbar spinal cord. Due to the wide-spread progressive

neurodegenerative changes occurring in the ALS model, targeting a region in the spinal cord

innervating one particular muscle group would unlikely halt the global degenerative process.

Thus, assessing neuropathological outcomes was utilized to glean whether PGRN upregulation

promoted neuronal survivability and attenuated the expected neuroinflammatory processes

inherent to mSOD1 models.

End-stage neuronal viability assessed by tabulating Nissl-positive cell bodies within the ventral

horn of the spinal cord did not indicate a statistically significant effect of PGRN treatment;

however, overall neuronal counts do suggest the possibility of a positive effect. Quantification of

motor neurons in the targeted L3-L5 region demonstrated a 25% increase in surviving motor

neurons at disease end stage in G37R-PGRN transfected animals compared to the GFP injected

89

cohort, while no difference between treatment groups in wild-type animals were found (Figure

3.5).

Similarly, in tissue sections corresponding to region L1-L2 regardless of genotype, early PGRN

delivery mediated a 37% and 12% increase in neuronal viability in wild-type or transgenic

animals, respectively (Figure 3.5). Finally, in the L6 region, PGRN transgene delivery mediated

a 7% increase in neuronal viability, while paradoxically, being associated with a 6% decrease in

the transgenic G37R group (Figure 3.5). As expected, in comparison to end-stage cresyl-violet

stained motor neurons within the lumbar spinal cord of more prototypical transgenic mSOD1

models (G93A: Wang, Popko, Tixier, et al. 2014 , G85R: Wang, Popko & Roos 2014), the G37R

animals utilized here, showed evidence of an increase in overall neuronal viability

Paradoxically, the outcome of neuronal counts as assessed by cresyl violet staining in region L3-

L5 is a finding not corroborated by immunolabeling for cholinergic motor neurons (Figure

3.7A). Early stage PGRN delivery mediated a non-significant 20% increase in ChAT positive

neuronal counts in wild-type animals, however in the transgenic G37R cohorts, PGRN

administration resulted in a near 20% reduction in viable cholinergic motor neurons.

Originally, motor neuron counts were stratified according to whether they were considered to be

alpha (240-950um2) or gamma (130-240um2) motor neurons. Upon further consideration these

counts were combined to generate an overall estimate of motor neurons within the ventral horn.

A potential drawback of this strategy is that there is no differentiation between healthy and

atrophying motor neurons. Further characterizing motor neurons within the targeted L3-L5

90

region as “healthy” or “atrophying” indicated that transgenic mSOD1 mice presented with cells

of a smaller diameter compared to their wild-type counterparts (Figure 3.6C,D). PGRN

administration was shown to mediate a significant shift of healthy motor neurons towards a

larger diameter within wild-type animals; a finding not recapitulated in the G37R cohort (Figure

3.6A).

Astro- and microglial proliferation are hallmarks of the degenerative processes inherent to ALS

pathogenesis (Lasiene & Yamanaka 2011). Although the transgenic G37R animals generated for

the study considered here experienced a drop in mSOD1 copy number and concomitantly

experienced a less severe ALS-like phenotype, there was a marked increase in glial cell

proliferation in comparison to wild-type counterparts within the ventral horn of regions L3-L5 at

disease end-stage (Figure 3.7B,C). In comparison to both the prototypical G93A and G85R ALS

models, the transgenic G37R animals used in this study showed evidence of a diminished glial

cell proliferative state within the lumbar spinal cord (Wang, Popko, Tixier, et al. 2014; Wang,

Popko & Roos 2014). Taken together, this further suggests that the transgenic cohorts used in

this study presented with an attenuated ALS-like phenotype.

Early-stage progranulin intervention did not mitigate the proliferation of either astrocytes or

microglia to a significant degree in the transgenic cohorts. In comparison to wild-type

counterparts, however, astrocytic proliferation appeared somewhat diminished in the PGRN-

treated cohort, while mediating no effect on the number of microglia.

91

In a progressively degenerative disease such as ALS, approximating any potential therapeutic

effect at the end-stage of disease pathogenesis does not necessarily adequately capture any

potentially efficacious effect at an earlier time point of the disease cascade. This is further

complicated by the fact that the use of ALS murine models has identified various compensatory

mechanisms that are deployed in response to the neurodegenerative insult mediated by the

mutant SOD1 protein. For example, in the G93A model an early compensatory response to the

ER stress experienced by motor neurons is the perinuclear re-organization of protein synthesis;

which aims to maintain neuronal survival by preserving RNA translational processes (Riancho et

al. 2014). This presumably occurs as the system attempts to normalize the toxic insult and stave

off the initiation of the apoptotic cascade. However, ultimately any inherent mechanism(s)

attempting to buffer against the degenerative cascade are overwhelmed, and the biological

system is driven to neuronal cell death, muscle atrophy, and the resultant display of paralysis.

Thus, overwhelming ER stress ultimately leads to apoptotic cell death (Flamment et al. 2012). In

the context of a therapeutic such as upregulated PGRN exploiting endogenous cell processes for

production and secretion, any untoward effect on the protein producing machinery will

negatively affect the therapeutic outcome.

Preliminary findings generated in our laboratory targeting mSOD1 transgenic motor neurons at a

later stage (12.1 months of age) of the disease cascade just 10 days prior to the onset of paralysis

(Figure C.1), demonstrated positive outcomes in ChAT positive neuronal viability at disease end

stage. In contrast, the current study assessing the applicability of PGRN upregulation at an earlier

stage of the disease cascade (approximately 3.5 months of age) was unable to establish a positive

long-term neuroprotective effect in an animal model presenting with an attenuated form of the

92

ALS phenotype. This suggests that the timing of viral administration and subsequent PGRN

upregulation is instrumental in mitigating the aberrant ALS-like phenotype in mSOD1 models.

What remains a possibility, however, is that the stable, long-term expression of the PGRN cDNA

construct was not achieved through the experimental paradigm utilized. The VSV-G

pseudotyped, HIV-derived lentiviral vector employed in this study was selected based on both

the preliminary work produced by our group, as well as the mounting body of evidence

suggesting its applicability for long-term stable transgene expression. Direct injection of an HIV-

1-based lentiviral vector into the murine brain initially demonstrated successful transduction of

terminally differentiated neurons more than 30 days after the initial transfection (Naldini et al.

1996). Similarly, assessing expression levels four weeks following lentiviral-mediated delivery

of a target gene directly into the thoracic spinal cord, suggested that lentiviral vectors

demonstrated the most stable transgene integration and expression in comparison to retroviral or

adenoviral vectors (Abdellatif et al. 2006). Employing a transgene delivery strategy into the

spinal cord reminiscent of the technique used in this study, Mentis et al. demonstrated transgene

expression within the lumbar spinal cord two-weeks following the initial muscle targeting

(2006). Furthermore, long-term expression of lentiviral-delivered cDNA within the brain, has

been confirmed 4 months following the initial vector delivery (Hottinger et al. 2000).

Taken together, it is thus expected that exploiting the inherent properties of an HIV-1-based

lentiviral vector would promote long-term and stable expression in targeted neuronal cells.

However, given the findings of this study, it may be plausible that when the targeted cells are

exposed to a temporally progressive neurodegenerative cascade – and ultimately cell death – that

93

the production of an exogenous gene product is halted, and/or rendered ineffective. The lack of

GFP-immunopositive cells at disease end stage is in stark contrast to our preliminary findings

(Figure C.2) assessing transgene expression up to a month following transfection. This outcome

underscores a significant challenge to the current study design, since temporal changes in long-

term transgene expression would alter the degree of availability of the neurotrophic factor

3.3 Conclusion

We targeted lumbar spinal cord motor neurons for the upregulated expression of progranulin by

an HIV-1-based lentivirus delivery mechanism. Transgenic mSOD1 animals were targeted at

stage where no overt ALS-like phenotype had yet been established so as to transduce healthy and

intact motor neurons. Upregulation of the growth factor was hypothesized to mitigate the

neurodegenerative cascade and not only result in an increase in survival, but attenuate aberrant

neurodegenerative processes and ultimately show evidence for increased neuronal viability and

reduced gliosis. Exogenous delivery of progranulin at 3.5- 4 months of age failed to not only

alter the eventual phenotypic presentation of hind limb paralysis and shortened survival, but also

did not demonstrate a diminution of the neurodegenerative cascade. Our findings here further

demonstrate that a reduction in the mutant human SOD1 gene locus gives rise to an attenuated

presentation of the expected ALS phenotype, where animals exhibit a delay in disease onset and

concomitant increase in overall lifespan.

94

Chapter 4: General Discussion and Future Studies

4.1 General Discussion

Amyotrophic lateral sclerosis is a progressive neurodegenerative disorder of primarily unknown

etiology which concomitantly lacks effective therapeutic options. Progranulin is a secreted

growth factor with an array of pleiotropic properties affecting cellular biochemistry, which has

borne out positive outcomes when applied to in vivo models of Alzheimer’s- and Parkinson’s

disease (Van Kampen & Kay 2011; Minami et al. 2014; Van Kampen et al. 2014). Preliminary

work by our group had set out to assess whether the neurotrophic properties inherent to the

growth factor would be of benefit in the context of late stage ALS pathogenesis. With the

experimental design described in this manuscript (Chapter 2:), we set out to assess the potential

neurotrophic properties of exogenously delivered PGRN cDNA when targeted to neuronal

subsets within the lumbar spinal cord at a time point early in the disease cascade. We

hypothesized that lentiviral-mediated neuronal targeting encoding PGRN cDNA at a stage prior

to significant neuronal loss would attenuate the expected progressive disease cascade.

The resultant outcomes at both the behavioural and neuropathological levels failed to distinguish

any significant effect mediated by the lentiviral delivery of progranulin at an early stage of the

ALS disease cascade (Chapter 3:). Furthermore, on account of a significant drop in the copy

number of the mSOD1 locus, transgenic animals utilized in this study presented with a protracted

progression of the disease. Thus the present study further underscores some of the challenges

faced by pre-clinical therapeutic development using murine ALS models.

95

4.1.1 Potential Sources of Error and Limitations in Experimental Design

Various factors inherent to the study design in question have contributed as potential sources of

error, thus complicating interpretations of the resultant data sets:

4.1.1.1 Reduced mSOD1 Copy Numbers in Derived Transgenic G37R Progeny

A major factor affecting the outcomes in the current study was the finding that the cohorts of

commercially-derived G37R(line 29) transgenic animals carried the mSOD1 transgene to a much

lower degree than expected. Untoward consequences arising from this reality included not only

an increase in overall lifespan, but a somewhat variable presentation in disease onset.

Additionally, the unexpected increase in lifespan would be expected to negatively affect the

stable integration of the GFP/PGRN transgenes, as neuronal degeneration could ultimately lead

to the loss of transfected neurons expressing the exogenous gene of interest. In the end, the major

impediment arising out of the reduced mSOD1 transgene levels, is the prohibitive effect that this

finding has on replicative studies with “true” G37R(line29) animals. That said, directly arising

from our findings here, Jax’s database for this colony has been updated to reflect the attenuated

presentation of ALS. Thus, any future researchers deriving transgenic animals from this colony

are well aware of the prolonged disease cascade that these animals experience.

4.1.1.2 High Variability in Behavioral and Neuropathological Outcomes

A major drawback to interpretation of the work presented here is the high degree of variability

that is evident in both behavioural and neuropathological measures. Mice from the WT and

G37R cohorts were generated under the same conditions from related male breeder animals, and

housed in similar group-matched conditions (a subset of male animals were singly housed due to

96

in-fighting), however, variability between research subjects and/or groups persisted during the

study.

Due to the sheer number of animals, the baseline behavioural assessments were split across two

successive dates. Following lentiviral administration, the experimenter became adept at

processing all of the research subjects within the same day, and thus for the remainder of the

trials, all animals were tested within the same day, thus limiting inter-day variability.

Once transferred into the containment unit, an added caveat affecting data variability was that all

animal manipulation and behavioural measures were restricted to a biological safety cabinet

(BSC) as required by the Biosafety committee. Presumably, the auditory stimuli provided by the

BSC would negatively affect animal welfare as it may be expected to increase stress hormone

levels in the animals (Turner et al. 2005). Furthermore, since behavioural assessments were

conducted on weekdays between 7am and 5pm - corresponding to the resting phase of the

circadian cycle - this testing paradigm is likely not ideal (Roedel et al. 2006). In future studies to

reduce animal stress and limit variability in a specific behavioural measure, automated nightly

measurements of spontaneous activity on a cage running wheel should be incorporated into the

battery of behavioural tests of singly-housed subjects (Bennett et al. 2014)

Due to the associated costs of long-term housing within the containment unit, we incorporated a

bare minimum of 10-12 animals per group in the study. On account of our previous experience

with this model, we did not expect the animals to present with an attenuated progression of the

disease and thus did not account for the loss of animals from sporadic causes such as dermatitis.

97

Not only did the loss of animals throughout the study further influence data variability, but the

wide range in the overall survivability of transgenic animals can be expected to affect the

variability in the end-stage assessment of neuropathological outcomes. Finally, challenges

inherent to the processing and preparation of tissue specimens further limited the available viable

samples for immunohistological analysis. Given the opportunity to build on this experiment, we

would bolster the power of the study by excluding WT animals and increasing the number of

subjects within each transgenic cohort (n of 24-30 per group).

4.1.1.3 Lack of Confirmation of PGRN/GFP Transgene Integration at End-stage

Preliminary data generated by our group justifying the current study had indicated that stable

transgene integration can be observed a month following gastrocnemius-mediated lentiviral

delivery (Figure C.2). Additionally, knowing that lentiviral-mediated transgene delivery has been

found to stably transfect targeted cells for the long term in other contexts (e.g. Naldini et al.

1996; Hottinger et al. 2000; Van Kampen et al. 2014), we presumed that once stable genomic

integration had occurred, that the transgene would be expressed until the neuronal cell was no

longer viable. However, the inability to confirm transgene integration at the experimental end-

stage via qPCR for either locus is a significant drawback to the current study, and, is most likely

a reflection of the technical challenges that are inherent to performing quantitative genotyping on

fixed and processed tissues. Additionally, in contrast to our preliminary findings (Figure C.2),

the inability to demonstrate any GFP immunopositive labeling in the L3-L5 region at the

experimental end point further underscores the challenges of long-term transgene integration and

expression within a progressively neurodegenerative model. Any future studies assessing the

98

therapeutic efficacy of transgene upregulation would be best served by the temporal collection

and analyses of tissue samples for the gene and/or gene product of interest.

4.1.1.4 Manually Organizing Lumbar Spinal Tissues into Discrete Segments

A concerted effort was taken to differentiate between the L1-L6 segments of the spinal cord

tissue sections used in histological assays. As detailed in section 2.7, 1.0mm subsets of cresyl

violet stained 10µm-thick tissue sections (separated by 100µm) were visualized, manually

compared to a spinal cord atlas, and segmented into L1-L2, L3-L5, and L6 regions based on

anatomical landmarks. It is possible that sets of slides bordering a neighbouring lumbar spinal

cord segment would contain sections from both regions. This was retroactively addressed during

the image analysis of overall neuronal counts, when each tissue section would be stratified

according to the various regions more accurately. Only nuanced changes needed to be

incorporated at this stage; speaking to the effective organization of the spinal cord at the original

assessment. When targeting the L3-L5 region exclusively for ChAT, GFP, GFAP, and Iba-1

immunoreactivity, the updated organization of tissue slides were utilized to more precisely target

the region of interest. Future studies assessing gene delivery to discrete regions of the spinal cord

could be conducted more expediently if spinal cord tissues are directly transected out at the level

of the vertebrae.

99

4.2 Future Studies

4.2.1 Progranulin as an ALS Therapeutic

4.2.1.1 Targeting Multiple Spinal Cord Regions at the Late Stage of Disease

Pathogenesis

Given the previous success with our preliminary studies targeting the lumbar spinal cord at an

advanced stage of the disease cascade in mSOD1 animals (Appendix C), thoroughly further

investigating this finding would be meritorious in assessing how the timing of lentiviral

administration affects the disease cascade. Targeting neurons 2-3 months prior to the expected

phenotypic development would assess the efficacy of PGRN transgene delivery at a stage when

neuronal damage has been initiated. This approach has the additional benefit of limiting the

encumbrance of ensuring significant long-term transgene expression, all the while, limiting the

number of deaths from unexpected factors. A limitation of the current study is the sole targeting

of motor neurons innervating the gastrocnemius muscles. Realistically, due to the relatively low

numbers of neurons transduced (i.e. a specific subset primarily in the L3-L5 region), this would

not present with an overall alteration of the ALS behavioural phenotype. Concurrent targeting of

multiple muscle groups in transgenic mSOD1 animals, would allow for widespread delivery of

the PGRN transgene throughout the thoracic to cervical spinal cord (Nakajima et al. 2008). With

the widespread delivery of the transgene prior to disease onset, behavioural measures and

neuropathological correlates may be more significantly affected by PGRN upregulation.

Furthermore, additional work assessing this experimental paradigm in the context of the mutant

TDP-43 model would serve to establish whether the growth factor could ameliorate the ALS-like

deficits in a TDP-43 proteinopathy.

100

4.2.1.2 Neuronal Specific Overexpression of Progranulin in a Murine Model of ALS

Specific overexpression of the progranulin construct exclusively restricted to motor neuron

populations throughout the spinal cord would serve to assess the neurotrophic properties of the

growth factor. This strategy would allow for the long-term and stable expression of the

transgene, while assessing the therapeutic effect of neuron-specific PGRN expression throughout

the spinal cord. Adopting an approach similar to that previously described (Feng et al. 2000;

Jaarsma et al. 2008), PGRN cDNA would be cloned into a Thy1.2 cassette exhibiting a neuron-

specific expression pattern. The resultant construct will be used to generate a line of animals

overexpressing PGRN which will subsequently be crossed with mSOD1 animals; generating

double transgenic mice. As with section 4.2.1.1, it would be expected that the widespread

neuron-specific overexpression of PGRN would attenuate behavioural and neuropathological

outcomes of the disease cascade in comparison to mSOD1 animals. Additional follow-up work

would recapitulate a similar strategy utilizing mutant TDP-43 models to assess the growth

factor’s effect on a non-SOD1 linked model of ALS.

4.2.1.3 Investigate the Potential for Progranulin as an Epidermal Biomarker in ALS

Progression6 9

As described in section 1.7.2, PGRN levels in CSF and plasma samples from newly-diagnosed

ALS patients, do not differ from that of controls (Philips et al. 2010). Post-mortem analysis of

clinical cases show evidence of increased PGRN expression levels that are confined to areas of

neurodegeneration (Irwin et al. 2009).

6This experimental paradigm has recently been submitted by our research group in consideration for funding

101

ALS patients typically do not present with bed-sores, which is indicative of putative biochemical

changes within the epidermis (Ono 2000). In sALS cases, an increase in the expression of PGRN

has been identified within the epidermal layer of the skin. This increased expression of PGRN

has been positively correlated with the duration of the disease, which suggests that as the disease

progresses, expression of PGRN in the skin increases (Yasui et al. 2011). What remains

unknown is whether this temporal increase in the expression of the neurotrophic factor is causal-

or in response to- the exacerbating degenerative cascade or whether symptomology is established

after some unknown threshold level of PGRN expression is exceeded. Furthermore, it is unclear

as to whether such increase in expression within the epidermis is observed in familial cases.

In the G93A animal model, Philips et al were able to show (1) that PGRN levels in the CSF

temporally increase over the disease course, and (2), that as the ALS phenotype worsens, levels

of PGRN within the spinal cord increases primarily due to enhanced microglial expression

(2010). What thus remains to be elucidated is whether PGRN levels within murine skin are

altered to a similar degree as in sALS patients, and the degree to which this correlates with

neural degeneration.

To investigate the clinical utility of temporal epidermal PGRN as a predictive biomarker for

disease progression, transgenic mSOD1 animals would be euthanized at points throughout the

disease cascade. Skin samples would be assayed for PGRN expression, while spinal tissues

would be analyzed for neuronal viability and neuroinflammatory markers. Correlative

comparisons between PGRN expression and the degree of neurodegeneration could then be used

to assess whether epidermal PGRN levels can function as a predictive biomarker in this animal

102

model. Positive outcomes from this avenue of research could be further investigated in the

clinical setting and if borne out, would provide for a minimally invasive assay that could predict

the degree of neuronal loss, and lead to earlier diagnoses.

4.2.2 Considerations for the Clinical Translation of ALS Research Efforts 10

4.2.2.1 Limitations of Clinical ALS Trials and Modeling Disease in Mouse Models

Expressing Mutant Human SOD17 11

To date, various therapeutics have been investigated for the ability to attenuate and/or ameliorate

the pathological disease cascade induced due to ALS neuropathy (Wilkins et al. 2011;

Mitsumoto et al. 2014). Each of these agents addresses a unique hypothetical mechanism which

is purported to play a key role in establishing the eventual disease state, but have failed to

demonstrate effective translation at the clinical level. Selected outcomes from both pre-clinical

and clinical trials are summarized for riluzole, minocycline, and creatine in tables 4.1- 4.3.

In short, various pre-clinical mSOD1 murine models testing these three therapeutic agents have

demonstrated a positive effect while failing to recapitulate positive outcomes in clinical subjects.

To date, riluzole remains the only drug approved in the treatment of ALS, but only exerts a

modest 2-3 month increase in lifespan (Miller et al. 2012). The discrepancy between pre-clinical

results and clinical translation are multifactorial in nature. Mounting evidence supports a

heterogeneous cascade of events that underlie the inherent pathomechanism(s) of ALS (Turner et

al. 2013). Clinically, patients present with a phenotypically heterogeneous disease that can affect

10 See Appendix E for an additional discussion with respect to parametric considerations vital to clinical translation 7A version of section 4.2 has previously been published as part of Zwiegers and Shaw, 2015.

103

site of onset, rate of disease progression, upper and/or lower motor neuron involvement, and

whether the phenotype is strictly behavioural or elicits some form of cognitive impairment

(Appendix A.2; Ravits et al. 2013). It is probable that inadequate stratification of the patient

population in terms of phenotypic variability (and perhaps unknown genetic and environmental

causal factors) could undermine robust clinical outcomes, since any specific treatment effects

may only be applicable to a subset of ALS patients (Beghi et al. 2011; Mitsumoto et al. 2014).

The theoretical utility of animal models is in their ability to mimic the underlying disease

cascade and thus act as a platform for therapeutic development. A model system should be

representative of the patient population as a whole so that any pre-clinical drug development

should be widely applicable when translated clinically. As pointed out in section 1.5.2, there is

roughly a global prevalence rate for ALS of 4-7 per 100 000 population; thus around half a

million patients would be affected world-wide (Chiò et al. 2013; Figure 1.1). Familial ALS cases

due to mutant SOD1 loci would then account for approximately 5000-10,000 patients. Pre-

clinical development based solely on a model of mutant SOD1 thus has the potential to greatly

impede translational applicability since (i) such a small percentage of patients that comprise the

clinical trial will be carrying a mutant copy of the SOD1 gene, and (ii) logistically enrolling

SOD1-affected fALS patients in a trial of sufficient power to properly test a developed therapy

would not be feasible.

Further, as is evidenced in summary tables 4.1- 4.3, there is limited proof for translational

applicability of therapeutics developed in mSOD1 models of ALS: a caveat not restricted to

neurodegenerative disease research. Generally a positive pre-clinical effect has less than a 40%

104

chance of recapitulating a similar clinical outcome (Hackam & Redelmeier 2006), with some

animal models not even accurately portraying the inherent disease pathomechanism(s) (Seok et

al. 2013).

Limited independent validation studies, publication bias and deficits in good experimental design

may also account for some of the discrepancy in outcomes between clinical and pre-clinical

mSOD1 studies (Benatar 2007). Furthermore, as Scott et al poignantly discuss, failing to control

for specified biological factors in ALS models has likely resulted in the measurement of chance

variability (i.e. “noise”), and has thus resulted in false positive results (2008). To underscore the

importance of these variables, an optimized study paradigm investigating the efficacy of

Riluzole, minocycline, or creatine was unable to replicate the positive pre-clinical outcomes

summarized in table 4.1 (Scott et al. 2008).

In short, due to the complex interplay of etiological factors, effectively modeling ALS has

proven to be quite challenging. Thus, the discrepancy in translating therapeutics clinically is

likely multifactorial in nature and will be the focus of the remaining discussion. Of primary

concern here is how well the in vivo model mimics clinical ALS, impediments hindering

replicability between pre-clinical studies, as well as the importance of species differences in drug

metabolism.

4.2.2.2 Models Representative of Clinical ALS

Focusing on the original line of G93A mSOD1 animals with the greatest number of transgene

integrations (G1), neuropathological changes are of a progressively degenerative nature (Table

105

1.1; Gurney et al. 1994; Dal Canto & Gurney 1994; Dal Canto et al. 1995). At the outset,

vacuolar degeneration is observed within motor neurons, but with time, alongside a marked

reduction of neuronal cell bodies, pathological vacuoles are present within the surrounding

neuropil with a marked deterioration of the anterior horn. As discussed in section 1.4.1, evidence

for vacuolar degeneration in animals highly expressing the mutant G93A SOD1 locus does not

reflect the reality of end-state pathology in human cases, and has been suggested to be a toxic

artefact arising from the transgene being significantly overexpressed (Dal Canto et al. 1995).

Thus, it is possible that the marked overexpression of the mutant transgene results in a more

severe disease pathology which may not reliably measure a positive or negative clinical outcome

for translational purposes. Within the context of an accelerated disease cascade, it is probable

that failed therapeutic agents at the pre-clinical stage may have had some clinical utility in the

protracted form of human disease. However, without strong pre-clinical data in these progressive

models, candidate therapeutics will not be promoted to the clinical trial stage.

Expression of the aberrant phenotype in SOD1 fALS models is driven by the significant

overexpression of the mutant hSOD1 locus (Table 1.1and Table 1.2). In stark contrast, familial

forms of the disease (SOD1-linked fALS) are inherited mostly in an autosomal dominant manner

and thus present with ALS in the context of one aberrant copy. Transgenic lines carrying

significantly fewer copies of various gain-of-function mutations do not develop disease after

prolonged observation (Gurney et al. 1994; Wong et al. 1995), or give rise to a significantly

protracted disease course (Dal Canto & Gurney 1997). For instance the SOD1 A4V strain of

mice carry the most common SOD1 mutation found in North America (Saeed et al. 2009), yet

fail to recapitulate the requisite phenotypic correlates of disease (Gurney et al. 1994). This is

106

primarily thought to be a consequence of decreased transgene expression levels. Of the two A4V

lines developed, the line expressing the mutant protein to a higher degree (20% greater than the

lower expressing line) was shown to exhibit an affected phenotype only when crossbred with a

line overexpressing wild-type hSOD1; suggesting that a threshold level of misfolded SOD1

needs to be exceeded prior to phenotypic onset (Deng et al. 2006).

An additional impediment in utilizing the model to predict clinical efficacy is in the timing of

drug administration. Typically the timing of drug delivery in these models is prior to phenotypic

onset and thus provides a wide therapeutic window for the effect to be realized (Table 4.1). In

lieu of biomarkers which specifically gauge the onset and progression of the disease, a typical

therapeutic intervention initiated in the clinic will only be administered following symptom

onset; most notably at a stage of the disease when marked neuronal cell loss has already

occurred. Due to the intractable and rapidly progressive nature of ALS, it should not be

surprising that many promising pre-clinical therapeutics have not been able to attenuate and/or

ameliorate the disease when initiated at an advanced stage when translated to the clinic.

4.2.2.3 Impediments to Replicability between Pre-clinical Studies

The literature is rife with studies exemplifying the relationship between mutant hSOD1 transgene

levels incorporated into the genome and the degree of phenotypic severity. Our experience with

the G37R model animals in this study underscores this dilemma, in that we had generated a

colony of animals from commercial breeders with an unreported drop in copy number. As

expected, these presented with a concomitant increase in lifespan and delay in disease

progression (Section 3.1.3, Zwiegers et al,. 2014). Transgene level variation can arise due to

107

meiotic recombination. In the G93A mSOD1 model, this is relatively infrequent with

recombination accounting for transgene fluctuations in minor percentage of progeny generated

(Alexander et al. 2004). Due to random meiotic events, the original G93A G1 line has spawned

two additional sub-lines with either a 40% expansion (25 copies) or 30% retraction (13 copies)

of the mutant hSOD1 locus; each showing variations in phenotype severity on account of altered

transgene expression levels (Chiu et al. 1995; Zhang et al. 1997). Experimentally, variations in

transgene dosage are detrimental to research efforts. First, small colonies are likely to be

adversely impacted and would serve limited applicability in replication studies. Second,

undetected fluctuations in individual copy number will give rise to animals exhibiting varied

phenotypic expression of disease severity, respond differently to therapeutic intervention, and

contribute to variability in the data. Should an experimental cohort by chance be comprised of

outliers with either higher/lower copy numbers, a tested therapeutic effect may be primarily a

consequence of disease severity linked to transgene levels (Figure 1.4). Without dutifully

reporting on the number of integrated transgenes/mSOD1 expression levels, the issue is further

compounded when independent groups attempt to replicate previously published results. In the

case of G37R(line 29) animals, a review of the relevant literature over the preceding two decades

have indicated that critical information regarding transgene copy number and mutant SOD1

expression levels have not been reported (Figure 4.1).

108

Figure 4.1. Literature survey for details regarding the genomic SOD1 G37R locus in line 29 animals.

A review of publications (n=25) listed in the Mouse Genome Informatics database over the course of the preceding two decades in reference to G37R line 29 animals show (1) discrete copy numbers have not been characterized, (2), only half of the studies assess SOD1 protein and/or aggregate levels, and (3), that even if the genotype were assessed, no values regarding the relative transgene presence are reported. 4.2.2.4 Species Differences and Drug Metabolism

The incongruent findings between clinical and pre-clinical studies may further be explained by

underlying physiological differences in metabolizing specific therapeutics. To the authors’

knowledge, no investigation has yet been conducted to study whether any putative ALS

pharmacological agent is processed similarly in both mice and man, and thus whether it is

bioactive to the same degree in these different species. Although it is used to induce a form of

parkinsomism, systemic administration of the neurotoxic agent MPTP exemplifies species-

specific differences in metabolic activity (Kalaria et al. 1987). In humans and primate models,

intravenous delivery mediates an acute parkinsonian syndrome virtually indistinguishable from

idiopathic Parkinson’s disease. Rodent models however have proven to be somewhat more

109

resilient to the toxic insult accompanying systemic administration. In rats, because of peripheral

enzymatic catabolism of MPTP by monoamine oxidase, the polar MPP+ metabolite cannot cross

the blood-brain-barrier and thus fails to selectively induce dopaminergic neuron degeneration

(Kalaria et al. 1987). Without dutifully assessing the possibility of cross-species drug metabolic

differences, it is possible that any pre-clinical effect may not be fully recapitulated in the clinic.

4.2.3 Concluding Remarks

Targeted lentiviral delivery of PGRN cDNA into motor neurons of the lumbar spinal cord in an

mSOD1 ALS model expressing an attenuated form of the disease, did not demonstrate an

appreciable remediation in overall survival or neuropathological changes compared to a GFP-

injected cohort. This study underscores the challenges inherent to long term exogenous transgene

expression in a temporally progressive neurodegenerative model, as well as, highlights the

dependence of ALS phenotypic expression on the degree of mutant SOD1 levels. Future clinical

translation of putative therapies should be restricted to pre-clinical data demonstrating positive

outcomes in multiple model systems. Adopting such a strategy would allow for a tested

therapeutic to be more widely applicable to the ALS patient population at large.

110

Table 4.1 Therapeutic efficacy outcomes in pre-clinical mSOD1 models.

Strain N

Timing of drug administration

Treatment (admin route)

Onset (d) Mean +/-

SD

% delay in disease onset

lifespan (d) Mean +/- SD

% increase in lifespan

RILUZOLE mechanism: anti-glutametergic

Gurney et al, 1996 (49) Tg(SODl*G93A) 1Gur 9 control 0 µg/mL water (Diet) 95 +/-12

134 +/- 8

10 presymp; 50d 100 µg/mL water (Diet) 98 +/- 11 3.16% 148 +/- 14 10 .45%*

Gurney et al, 1998 (50) Tg(SODl*G93A) 1Gur 11 control 0 mg/kg body weight/day (Diet) n.d.

127 +/- 18.9

11 presymp; 42-43d

12 +/- 0.5 mg/kg body weight/day (Diet) n.d. n/a 129 +/- 14.6 1.58%

10 presymp; 42-43d

24 +/- 1.3 mg/kg body weight/day (Diet) n.d. n/a 140 +/-13.6 10.24%

11 presymp; 42-43d

44 +/- 2 mg/kg body weight/day (Diet) n.d. n/a 139 +/- 9.0 9.449%*

Snow et al, 2003 (71) Tg(SOD1*G93A)dl1Gur 17 presymp; 40d control (Diet) approx. 180

13 presymp; 40d

100ug/mL in drinking water (Diet) approx. 192

approx. 6.67%* n.d n.d

MINOCYCLINE mechanism: antibiotic/

anti-inflammatory, anti-apoptotic

Kriz et al, 2002 (54) Tg(SOD1*G37R)29Dpr 12 control control (Diet)

311.5 +/- 45.4

343.56 +/-36.4

17

late presympt; 7/9months 0.1% w/w; 1g/kg chow (Diet)

316.26 +/- 47.9 1.54%

365.4 +/- 60.6 6.36%*

Van Den Bosch et al, 2002 (55) Tg(SODl*G93A) 1Gur 7 presymp , 70d 0 mg/kg n.d.

130.3 +/- 4.5

7 presymp , 70d 25 mg/kg n.d. n.d. 142.9 +/- 7.9 9.67%**

7 presymp , 70d 50 mg/kg n.d. n.d. 150.9 +/- 9.8 15.8%***

Zhu et al, 2002 (56)φ Tg(SODl*G93A) 1Gurǂ 10 presymp, 35d control - saline (IP) 90.3 +/- 7.0

125.6 +/- 10.8

10 presymp, 35d 10mg/kg body weight/day (IP) 109.0 +/- 4.7 20.71%*** 136.8 +/- 3.8 8.92%**

8 presymp, 42d control - saline (IP) n.d.

126.3 +/- 7.6

8 presymp, 42d 11mg/kg body weight/day (IP) n.d.

139.0 +/- 5.9 10.06%*

111

Strain N

Timing of drug administration

Treatment (admin route)

Onset (d) Mean +/-

SD

% delay in disease onset

lifespan (d) Mean +/- SD

% increase in lifespan

Zhang et al, 2003 (72) Tg(SODl*G93A) 1Gurǂ 10

presymp; 21d (diet), 28d (IP)

control diet + saline injections ( IP) 94.2 +/- 6.8

126.3 +/- 4.2

10 presymp; 28d 22mg/kg/d (IP) 113 +/- 8.8 19.96%* 142.2 +/-4.9 12.59%*

CREATINE

mechanism: limits mitochondrial damage; important in ATP prod'n

Klivenyi et al, 1999 (70)φ Tg(SODl*G93A) 1Gur 6 control control (Diet) n.d.

143.7 +/- 5.6

7 presymp; 70d 1% w/w (Diet) n.d. n/a 157.2 +/- 7.4 9.4%*

7 presymp; 70d 2% w/w (Diet) n.d. n/a

169.3 +/- 12.4 17.81%**

Snow et al, 2003 (71) Tg(SOD1*G93A)dl1Gur 17 presympt; 40d control (Diet) approx. 180

14 presymp; 40d 2% w/w (Diet) approx. 193

approx. 7.2%* n.d n.d.

Zhang et al, 2003 (72) Tg(SODl*G93A) 1Gurǂ 10

presymp; 21d (diet), 28d (IP)

control diet + saline injections ( IP) 94.2 +/- 6.8

126.3 +/- 4.2

10 presymp; 21d 2% w/w (Diet) 111 +/- 4.7 17.83%* 141.9 +/- 4.3 n.d.

Combinatory Treatments Snow et al, 2003 (71) Tg(SOD1*G93A)dl1Gur 17 presymp; 40d control (Diet) approx. 180

n.d n.d.

15 presymp; 40d

riluzole:100µg/mL in drinking water + 2% creatine w/w in chow (Diet) approx. 194

approx 7.78%* n.d n.d.

Zhang et al, 2003 (72) Tg(SODl*G93A) 1Gurǂ 10

presymp; 21d (diet), 28d (IP)

control diet + saline injections (Diet + IP) 94.2 +/- 6.8

126.3 +/- 4.2

10

presymp; 21d (diet), 28d (IP)

2% creatine + 22mg/kg/d minocycline (Diet + IP) 122 +/- 8.9 29.51%* 157.2 +/- 4.1 24.47%*

112

n.d: not determined/fully described φ: Variance assumed to be reported as SEM; converted into SD and reported here ǂ: Not clearly delineated; assumed to be G1H based on mean survival data --: no specific details listed presymp: treatment initiated at a presymptomatic stage P-values (compared to transgenic control): * p≤ 0.05; **p≤ 0.01*** ; p≤0.001

Table 4.2. Therapeutic drug effect on lifespan of patients in select clinical trials.

Survival Outcome

Study design

Study duration Treatment Initial N

Age Mean +/-

SD % survival diff from control

% Δ from con

p-val

RILUZOLE

% survival at 12

months diff from control

% Δ from con

p-val

Lacomblez et al, 1996 (45)

double-blind,

placebo-controlled ≤18 months control 242 53-61 62.80%

50 mg/d 237 55-61 70.50% +7.70% +12.26% n.s

100mg/d 236 55-60 73.70% +10.90% +17.36%

0.016

200mg/d 244 56-59 72.50% +9.70% +15.45%

0.027

% survival at 18 months

diff from control

% Δ from con

p-val

control 242 53-61 50.40%

50 mg/d 237 55-61 55.30% +4.90% +9.72% n.s

100mg/d 236 55-60 56.80% +6.40% +12.70%

< 0.05

200mg/d 244 56-59 57.80% +7.40% +14.68% n.s

% survival at 21 months

diff from control

% Δ from con

p-val

113

Survival Outcome

Study design

Study duration Treatment Initial N

Age Mean +/-

SD % survival diff from control

% Δ from con

p-val

Bensimon et al, 1994 (44)

double-blind, placebo-controlled ≤21 months control 78 46-69 37.18%

100mg/d 77 46-68 49.35% +12.17% 32.73%

0.046

% survival at 6 months

diff from control

% Δ from con

p-val

Zoccolella et al, 2007 (48)

retrospective cohort study

>24 months, up to 60 months control 53 19-80 84.90%

riluzole trxǂ 73 32-80 93.20% +8.3% +9.78% n.s

% survival at 12 months

diff from control

% Δ from con

p-val

control 53 19-80 67.90%

riluzole trxǂ 73 32-80 79.50% +11.6% +17.08% n.s

% survival at 18 months

diff from control

% Δ from con

p-val

control 53 19-80 --

riluzole trxǂ 73 32-80 -- +4.6% -- n.s.

% survival at 12 months (BO)

diff from control

% Δ from con

p-val

control 13 (BO) 19-80 46.15%

riluzole trxǂ 20 (BO) 32-80 75.00% +28.85% +62.51% --

% survival at 6 diff from % Δ p-

114

Survival Outcome

Study design

Study duration Treatment Initial N

Age Mean +/-

SD % survival diff from control

% Δ from con

p-val

months (BO) control from con val

control 13 (BO) 19-80 69.23%

riluzole trxǂ 20 (BO) 32-80 95% +25.77% +37.18% 0.04

% survival (patients aged >70

yrs) diff from control

% Δ from con

p-val

control 14 19-80 43%

riluzole trxǂ 20 32-80 70% +27.00% +62.79 n.s

survival (rapidly progressive cohort)

diff from control

% Δ from con

p-val

control 13 19-80 18.8 months

riluzole trxǂ 22 32-80 20.5 months 1.7 months +9.04% n.s

% survival at 60 months

diff from control

% Δ from con

p-val

Traynor et al, 2003 (47)

retrospective cohort study 60 months control 97 43-92 23.71%

riluzole trxǂ 149 26-86 32.21% +8.5% 35.85% n.s.

% survival at 18 months

diff from control

% Δ from con

p-val

control 97 43-92 --

riluzole trxǂ 149 26-86 -- +5.1% -- n.s.

% survival at 12 months

diff from control

% Δ from con

p-val

control 97 43-92 --

riluzole trxǂ 149 26-86 -- +15% --

0.015

% survival at 6 diff from % Δ p-

115

Survival Outcome

Study design

Study duration Treatment Initial N

Age Mean +/-

SD % survival diff from control

% Δ from con

p-val

months control from con val

control 97 43-92 --

riluzole trxǂ 149 26-86 -- +23% --

0.015

% survival <18 months (BO

patients) diff from control

% Δ from con

p-val

control 57 (BO) 38.60%

riluzole trxǂ 80 (BO) 45.00% +6.4% +16.59% 0.02

8

CREATINE

% survival at 16 months

diff from control

% Δ from con

p-val

Groeneveld et al, 2003 (65)

double-blind, placebo-controlled

16 months, prematurely halted control 87 56-68 12.64%

10g/d 88 44-67 13.64% +1.00% +7.90% n.s.

% survival at 12 months

diff from control

% Δ from con

p-val

control 87 56-68 41.38%

10g/d 88 44-67 43.18% +1.8% 4.35% n.s.

% survival at 6 months

diff from control

% Δ from con

p-val

Shefner et al, 2004 (73)

double-blind, placebo- 6 months control 54 47-72 89.00%

116

Survival Outcome

Study design

Study duration Treatment Initial N

Age Mean +/-

SD % survival diff from control

% Δ from con

p-val

controlled

20g/d for 5d then 5g/d 50 48-70 96.00% +7.00% +7.90% n.s.

MINOCYCLINE

% survival at 9 months

diff from control

% Δ from con

p-val

Gordon et al, 2007 (59)

double-blind, placebo-controlled

9 months; followed ≤ 32 months control 206 47-69 84.47%

start @ 100mg 2x/d; inc to ≤ 400mg/d 206 47-70 80.10% -0.0437 -5.17% n.d*

% survival at 12 months

diff from control

% Δ from con

p-val

control 206 47-69 63.59%

start @ 100mg 2x/d; inc to ≤ 400mg/d 206 47-70 57.77% -5.82% -9.15% n.d

% survival at 12 months

diff from control

% Δ from con

p-val

control 206 47-69 17.96%

start @ 100mg 2x/d; inc to ≤ 400mg/d 206 47-70 19.42% -1.46% -8.13% n.d

117

BO: bulbar onset ǂ: dosage not specified n.s: not significant n.d: not determined/fully described *:not significant due to hazard ratio analysis at this time point

Table 4.3. Therapeutic drug effect on functional outcomes of patients in select clinical trials.

Functional Outcome

Study design Study

duration Treatment N Age

Mean +/-SD % Δ Functional

Outcome diff from control

% Δ from con p-val

RILUZOLE

Manual Muscle Testing

(rate of deterioration) diff from control

% Δ from con p-val

Lacomblez et al, 1996 (45)

double-blind, placebo-controlled

≤18 months

control 242 53-61

--

50 mg/d 237 55-61 -- -- -- n.s

100mg/d 236 55-60 -- -- -- n.s

200mg/d 244 56-59 -- -- -- n.s

Bulbar & Limb Norris scales

(rate of deteriration) diff from control

% Δ from con p-val

control 242 53-61 --

50 mg/d 237 55-61 -- -- -- n.s.

100mg/d 236 55-60 -- -- -- n.s.

200mg/d 244 56-59 -- -- -- n.s.

118

Functional Outcome

Study design Study

duration Treatment N Age

Mean +/-SD % Δ Functional

Outcome diff from control

% Δ from con p-val

Limb Function (mean annual rate of

deterioration) diff from control

% Δ from con p-val

Bensimon et al, 1994 (44)

double-blind, placebo-controlled

≤21 months

control 78 46-69

28.10%

100mg/d 77 46-68 21.80% -6.30% -22.42% n.s

Bulbar Function (mean annual rate of

deterioration) diff from control

% Δ from con p-val

control 78 46-69 12.30%

100mg/d 77 46-68 9.80% -2.50% -20.33% n.s

Muscle Strength (mean annual rate of

deterioration) diff from control

% Δ from con p-val

control 78 46-69 34.40%

100mg/d 77 46-68 22.90% -11.50% -33.43% 0.028

CREATINE

Muscle Strength (mean rate of decline/month)

diff from control

Slope % Δ

from con p-val Groeneveld et al, 2003 (65)

double-blind, placebo-controlled

16 months, halted

control 87 56-68

-0.076

10g/d 88 44-67 -0.087 -0.011 +14.47% n.s.

Vital Capacity (mean rate of decline/month)

diff from control

% Δ from con p-val

119

Functional Outcome

Study design Study

duration Treatment N Age

Mean +/-SD % Δ Functional

Outcome diff from control

% Δ from con p-val

control 87 56-68 41.38%

10g/d 88 44-67 43.18% +1.8% 4.35% n.s.

ALSFRS (median rate of decline/month)

diff from control

% Δ from con p-val

control 87 56-68 -1.02

10g/d 88 44-67 -0.86 0.16 -15.67% n.s

Muscle Strength (mean rate of decline/month)

diff from control

% Δ from con p-val

Shefner et al, 2004 (73)

double-blind, placebo-controlled

6 months control 54 47-72

--

20g/d for 5d then 5g/d

50 48-70 -- -0.018 -- n.s.

ALSFRS-R (mean rate of decline/month)

diff from control

% Δ from con p-val

control 54 47-72 --

20g/d for 5d then 5g/d

50 48-70 -- 0.042 -- n.s.

MUNE (mean rate of decline/month)

diff from control

% Δ from con p-val

control 54 47-72 --

20g/d for 5d then 5g/d

50 48-70 -- -0.303 -- n.s.

Muscle strength (mean rate of

diff from control

% Δ from con p-val

120

Functional Outcome

Study design Study

duration Treatment N Age

Mean +/-SD % Δ Functional

Outcome diff from control

% Δ from con p-val

decline/day)

Rosenfeld et al, 2008 (74)

double-blind, placebo-controlled

9 months control 54 48-70

-0.07

10g/d for 5d, then 5g/d

53 46-66

-0.08 -0.01 +14.29% n.s.

Vital Capacity (mean rate of decline/month)

diff from control

% Δ from con p-val

control 54 48-70 --

10g/d for 5d, then 5g/d

53 46-66 -- -- -- n.s.

ALSFRS-R (mean rate of decline/month)

diff from control

% Δ from con p-val

control 54 48-70 --

10g/d for 5d, then 5g/d

53 46-66 -- -- -- n.s.

MINOCYCLINE

ALSFRS-R (mean rate of decline/month)

diff from control

% Δ from con p-val

Gordon et al, 2007 (59)

double-blind,

placebo-controlled

9 months; followed

for ≤ 32

months

control 206 47-69 -1.04

start at 100mg 2x/d;

increased to ≤

206 47-70 -1.30 -0.26 +25% 0.005

121

Functional Outcome

Study design Study

duration Treatment N Age

Mean +/-SD % Δ Functional

Outcome diff from control

% Δ from con p-val

400mg/d

Vital Capacity (mean rate of decline/month)

diff from control

% Δ from con p-val

control 206 47-69 -3.01

start at 100mg 2x/d;

increased to ≤ 400mg/d

206 47-70 -3.48 -0.47 +15.60%

n.s.

Muscle strength (mean rate of decline/month)

diff from control

% Δ from con p-val

control 206 47-69 -0.26

start at 100mg 2x/d;

increased to ≤ 400mg/d

206 47-70 -0.30 -0.04 +15.38% n.s.

Gordon et al, 2004 (57)

double-blind,

placebo-controlled

6 months ALSFRS-R

(mean rate of decline/month)

diff from control

% Δ from con p-val

control 9 42-66 --

200mg/d 10 46-71 -- -0.01 -- n.s.

Vital Capacity (mean rate of decline/month)

diff from control

% Δ from con p-val

control 9 42-66 --

200mg/d 10 46-71 -- -1.81 -- n.s.

122

Functional Outcome

Study design Study

duration Treatment N Age

Mean +/-SD % Δ Functional

Outcome diff from control

% Δ from con p-val

Muscle strength (mean rate of decline/month)

diff from control

% Δ from con p-val

control 9 42-66 --

200mg/d 10 46-71 -- +0.65 -- n.s.

ALSFRS-R (mean rate of decline/month)

diff from control

% Δ from con p-val

dble-blind, placebo-

con, cross-over

8 months control 5 49-66

--

start 100mg 2x/d;

inc to ≤ 400mg/d

18 46-68

-- -0.78 -- 0.047

Vital Capacity (mean rate of decline/month)

diff from control

% Δ from con p-val

control 5 49-66 --

start 100mg 2x/d;

inc to ≤ 400mg/d

18 46-68

-- -0.13 -- n.s.

n.s: not significant --: no specific details listed/unclear ALSFRS(-R): ALS Functional Rating Scale (Revised) MUNE: Motor Unit Number Estimation.

123

References

Abdellatif, A. a et al., 2006. Gene Delivery to the Spinal Cord: Comparison Between Lentiviral, Adenoviral, and Retroviral Vector Delivery Systems. J Neurosci Res, 84(3), pp.553–567.

Ahmed, Z. et al., 2007. Progranulin in frontotemporal lobar degeneration and neuroinflammation. Journal of neuroinflammation, 4, p.7.

Al-Chalabi, A. et al., 2012. The genetics and neuropathology of amyotrophic lateral sclerosis. Acta Neuropathologica, 124(3), pp.339–352.

Alexander, G.M. et al., 2004. Effect of transgene copy number on survival in the G93A SOD1 transgenic mouse model of ALS. Molecular Brain Research, 130(1-2), pp.7–15.

ALSoD, 2015. ALS Online Genetics Database. [online] Available at: http://alsod.iop.kcl.ac.uk/Index.aspx. Available at: http://alsod.iop.kcl.ac.uk/.

Andersen, P.M. et al., 1997. Phenotypic heterogeneity in motor neuron disease patients with CuZn-superoxide dismutase mutations in Scandinavia. Brain : a journal of neurology, 120 ( Pt 1, pp.1723–1737.

Anderson, C.R. et al., 2009. The Spinal Cord, Elsevier. Available at: http://www.sciencedirect.com/science/article/pii/B9780123742476500018 [Accessed August 27, 2015].

Ayers, J.I. et al., 2014. Conformational specificity of the C4F6 SOD1 antibody; low frequency of reactivity in sporadic ALS cases. Acta neuropathologica communications, 2(1), p.55. Available at: http://www.actaneurocomms.org/content/2/1/55.

Baker, M. et al., 2006. Mutations in progranulin cause tau-negative frontotemporal dementia linked to chromosome 17. Nature, 442(7105), pp.916–919.

Barber, S.C., Mead, R.J. & Shaw, P.J., 2006. Oxidative stress in ALS: A mechanism of neurodegeneration and a therapeutic target. Biochimica et Biophysica Acta - Molecular Basis of Disease, 1762(11-12), pp.1051–1067.

Barnéoud, P. & Curet, O., 1999. Beneficial effects of lysine acetylsalicylate, a soluble salt of aspirin, on motor performance in a transgenic model of amyotrophic lateral sclerosis. Experimental neurology, 155(2), pp.243–251.

Beghi, E. et al., 2011. The epidemiology and treatment of ALS: focus on the heterogeneity of the disease and critical appraisal of therapeutic trials. Amyotrophic lateral sclerosis : official publication of the World Federation of Neurology Research Group on Motor Neuron Diseases, 12(1), pp.1–10.

124

Benatar, M., 2007. Lost in translation: Treatment trials in the SOD1 mouse and in human ALS. Neurobiology of Disease, 26(1), pp.1–13.

Bennett, E.J. et al., 2014. Early Detection of Motor Dysfunction in the SOD1G93A Mouse Model of Amyotrophic Lateral Sclerosis (ALS) Using Home Cage Running Wheels. PLoS ONE, 9(9), p.e107918. Available at: http://dx.plos.org/10.1371/journal.pone.0107918.

Blokhuis, A.M. et al., 2013. Protein aggregation in amyotrophic lateral sclerosis. Acta Neuropathologica, 125(6), pp.777–794.

Boillée, S. et al., 2006. Onset and progression in inherited ALS determined by motor neurons and microglia. Science (New York, N.Y.), 312(5778), pp.1389–1392.

Borchelt, D.R. et al., 1998. Axonal transport of mutant superoxide dismutase 1 and focal axonal abnormalities in the proximal axons of transgenic mice. Neurobiology of disease, 5(1), pp.27–35.

Van Den Bosch, L., 2011. Genetic rodent models of amyotrophic lateral sclerosis. Journal of Biomedicine and Biotechnology, 2011(Table 1).

Brooks, B.R. et al., 2000. El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotrophic lateral sclerosis and other motor neuron disorders : official publication of the World Federation of Neurology, Research Group on Motor Neuron Diseases, 1(5), pp.293–299.

Bruijn, L.I., Miller, T.M. & Cleveland, D.W., 2004. Unraveling the mechanisms involved in motor neuron degeneration in ALS. Annual review of neuroscience, 27, pp.723–749.

Burkhardt, M.F. et al., 2013. A cellular model for sporadic ALS using patient-derived induced pluripotent stem cells. Molecular and Cellular Neuroscience, 56, pp.355–364. Available at: http://dx.doi.org/10.1016/j.mcn.2013.07.007.

Byrne, S. et al., 2011. Rate of familial amyotrophic lateral sclerosis: a systematic review and meta-analysis. Journal of neurology, neurosurgery, and psychiatry, 82(6), pp.623–627.

Camu, W., Billiard, M. & Baldy-Moulinier, M., 1993. Fasting plasma and CSF amino acid levels in amyotrophic lateral sclerosis: a subtype analysis. Acta neurologica Scandinavica, 88(1), pp.51–55.

Carrì, M.T. et al., 1997. Expression of a Cu,Zn superoxide dismutase typical of familial amyotrophic lateral sclerosis induces mitochondrial alteration and increase of cytosolic Ca2+ concentration in transfected neuroblastoma SH-SY5Y cells. FEBS Letters, 414(2), pp.365–368.

125

Cedarbaum, J.M. et al., 1999. The ALSFRS-R: A revised ALS functional rating scale that incorporates assessments of respiratory function. Journal of the Neurological Sciences, 169(1-2), pp.13–21.

Cervenakova, L. et al., 2000. Progressive muscular atrophy variant of familial amyotrophic lateral sclerosis (PMA/ALS). Journal of the neurological sciences, 177(2), pp.124–130.

Cervetto, C. et al., 2013. Motor neuron dysfunction in a mouse model of ALS: Gender-dependent effect of P2X7 antagonism. Toxicology, 311(1-2), pp.69–77. Available at: http://dx.doi.org/10.1016/j.tox.2013.04.004.

Chen, X. et al., 2013. Progranulin does not bind tumor necrosis factor (TNF) receptors and is not a direct regulator of TNF-dependent signaling or bioactivity in immune or neuronal cells. The Journal of neuroscience : the official journal of the Society for Neuroscience, 33(21), pp.9202–13. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3707136&tool=pmcentrez&rendertype=abstract.

Chew, J. et al., 2015. C9ORF72 repeat expansions in mice cause TDP-43 pathology, neuronal loss, and behavioral deficits. Science, pp.1–22. Available at: http://www.sciencemag.org/cgi/doi/10.1126/science.aaa9344.

Chiò, a. et al., 2013. Global epidemiology of amyotrophic lateral sclerosis: A systematic review of the published literature. Neuroepidemiology, 41(2), pp.118–130.

Chiò, A. et al., 2011. Phenotypic heterogeneity of amyotrophic lateral sclerosis: a population based study. Journal of neurology, neurosurgery, and psychiatry, 82(7), pp.740–746.

Chiu, a Y. et al., 1995. Age-dependent penetrance of disease in a transgenic mouse model of familial amyotrophic lateral sclerosis. Molecular and cellular neurosciences, 6(4), pp.349–362.

Clement, a M. et al., 2003. Wild-type nonneuronal cells extend survival of SOD1 mutant motor neurons in ALS mice. Science (New York, N.Y.), 302(5642), pp.113–117.

Cruts, M. et al., 2006. Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked to chromosome 17q21. Nature, 442(7105), pp.920–924.

Cruts, M. & Van Broeckhoven, C., 2008. Loss of progranulin function in frontotemporal lobar degeneration. Trends in Genetics, 24(4), pp.186–194.

Cruts, M., Theuns, J. & Van Broeckhoven, C., 2012. Locus-specific mutation databases for neurodegenerative brain diseases. Human Mutation, 33(9), pp.1340–1344.

126

Cudkowicz, M.E. et al., 1998. Limited corticospinal tract involvement in amyotrophic lateral sclerosis subjects with the A4V mutation in the copper/zinc superoxide dismutase gene. Annals of Neurology, 43(6), pp.703–710.

Dal Canto, M.C. et al., 1995. Neuropathological changes in two lines of mice carrying a transgene for mutant human Cu, Zn SOD, and in mice overexpressing wild-type human SOD: A model of familial amyotrophic lateral sclerosis (FALS). Journal of Neuropathology and Experimental Neurology, 54(3), p.442.

Dal Canto, M.C. & Gurney, M.E., 1997. A low expressor line of transgenic mice carrying a mutant human Cu,Zn superoxide dismutase (SOD1) gene develops pathological changes that most closely resemble those in human amyotrophic lateral sclerosis. Acta Neuropathologica, 93(6), pp.537–550.

Dal Canto, M.C. & Gurney, M.E., 1994. Development of central nervous system pathology in a murine transgenic model of human amyotrophic lateral sclerosis. The American journal of pathology, 145(6), pp.1271–1279.

Dal Canto, M.C. & Gurney, M.E., 1995. Neuropathological changes in two lines of mice carrying a transgene for mutant human Cu,Zn SOD, and in mice overexpressing wild type human SOD: A model of familial amyotrophic lateral sclerosis (FALS). Brain Research, 676(1), pp.25–40.

Van Damme, P. et al., 2008. Progranulin functions as a neurotrophic factor to regulate neurite outgrowth and enhance neuronal survival. Journal of Cell Biology, 181(1), pp.37–41.

Danbolt, N.C., 2001. Glutamate uptake. Progress in neurobiology, 65(1), pp.1–105.

Daniel, R. et al., 2000. Cellular localization of gene expression for progranulin. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society, 48(7), pp.999–1009.

DeJesus-Hernandez, M. et al., 2011. Expanded GGGGCC Hexanucleotide Repeat in Noncoding Region of C9ORF72 Causes Chromosome 9p-Linked FTD and ALS. Neuron, 72(2), pp.245–256. Available at: http://dx.doi.org/10.1016/j.neuron.2011.09.011.

Deng, H.-X. et al., 2006. Conversion to the amyotrophic lateral sclerosis phenotype is associated with intermolecular linked insoluble aggregates of SOD1 in mitochondria. Proceedings of the National Academy of Sciences of the United States of America, 103(18), pp.7142–7147.

Dupuis, L. et al., 2004. Evidence for defective energy homeostasis in amyotrophic lateral sclerosis: benefit of a high-energy diet in a transgenic mouse model. Proceedings of the National Academy of Sciences of the United States of America, 101(30), pp.11159–11164.

127

Etemadi, N. et al., 2013. Progranulin does not inhibit TNF and lymphotoxin-α signalling through TNF receptor 1. Immunology and cell biology, 91(10), pp.661–4. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24100384.

Ezzi, S.A. et al., 2010. Neuronal over-expression of chromogranin A accelerates disease onset in a mouse model of ALS. Journal of Neurochemistry, 115(5), pp.1102–1111.

Feng, G. et al., 2000. Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron, 28(1), pp.41–51. Available at: http://www.ncbi.nlm.nih.gov/pubmed/11086982.

Ferraiuolo, L. et al., 2011. Molecular pathways of motor neuron injury in amyotrophic lateral sclerosis. Nature Reviews Neurology, 7(11), pp.616–630. Available at: http://dx.doi.org/10.1038/nrneurol.2011.152.

Flamment, M. et al., 2012. New insights into ER stress-induced insulin resistance. Trends in Endocrinology and Metabolism, 23(8), pp.381–390.

Forsberg, K. et al., 2010. Novel antibodies reveal inclusions containing non-native SOD1 in sporadic ALS patients. PLoS ONE, 5(7), pp.1–9.

Friese, A. et al., 2009. Gamma and alpha motor neurons distinguished by expression of transcription factor Err3. Proceedings of the National Academy of Sciences of the United States of America, 106(32), pp.13588–13593.

Gass, J. et al., 2012. Progranulin regulates neuronal outgrowth independent of Sortilin. Molecular Neurodegeneration, 7(1), p.33.

Goetz, C.G., 2000. Amyotrophic lateral sclerosis: Early contributions of Jean-Martin Charcot. Muscle and Nerve, 23(3), pp.336–343.

Gong, Y.H. et al., 2000. Restricted expression of G86R Cu/Zn superoxide dismutase in astrocytes results in astrocytosis but does not cause motoneuron degeneration. The Journal of neuroscience : the official journal of the Society for Neuroscience, 20(2), pp.660–665.

Gordon, P.H. et al., 2009. Clinical features that distinguish PLS, upper motor neuron-dominant ALS, and typical ALS. Neurology, 72(22), pp.1948–1952.

Gould, T.W. et al., 2006. Complete dissociation of motor neuron death from motor dysfunction by Bax deletion in a mouse model of ALS. The Journal of neuroscience : the official journal of the Society for Neuroscience, 26(34), pp.8774–8786.

Guerra, R.R. et al., 2007. Progranulin is a stress-response factor in fibroblasts subjected to hypoxia and acidosis. Growth factors (Chur, Switzerland), 25(4), pp.280–285.

128

Günther, R. et al., 2012. Clinical Testing and Spinal Cord Removal in a Mouse Model for Amyotrophic Lateral Sclerosis (ALS). Journal of Visualized Experiments, (61), pp.4–7.

Gurney, M.E. et al., 1994. Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science (New York, N.Y.), 264(5166), pp.1772–1775.

Gurney, M.E., 1997. The use of transgenic mouse models of amyotrophic lateral sclerosis in preclinical drug studies. Journal of the Neurological Sciences, 152(SUPPL. 1), pp.67–73.

Hackam, D.G. & Redelmeier, D.A., 2006. Translation of Research Evidence From Animals to Humans. Journal of the American Medical Association, 296(14), pp.1731–1732.

Hand, C.K. & Rouleau, G.A., 2002. Familial amyotrophic lateral sclerosis. Muscle Nerve, 25(2), pp.135–159. Available at: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=11870681.

Hardiman, O., van den Berg, L.H. & Kiernan, M.C., 2011. Clinical diagnosis and management of amyotrophic lateral sclerosis. Nature Reviews Neurology, 7(11), pp.639–649. Available at: http://dx.doi.org/10.1038/nrneurol.2011.153.

Harms, M.B. & Baloh, R.H., 2013. Clinical Neurogenetics: Amyotrophic Lateral Sclerosis. Neurologic Clinics, 31(4), pp.929–950.

He, Z. & Bateman, A., 2003. Progranulin (granulin-epithelin precursor, PC-cell-derived growth factor, acrogranin) mediates tissue repair and tumorigenesis. Journal of Molecular Medicine, 81(10), pp.600–612.

Herdewyn, S. et al., 2013. Progranulin does not affect motor neuron degeneration in mutant SOD1 mice and rats. Neurobiology of Aging, 34(10), pp.2302–2303. Available at: http://dx.doi.org/10.1016/j.neurobiolaging.2013.03.027.

Hoffman, E.K. et al., 1996. Proteasome inhibition enhances the stability of mouse Cu/Zn superoxide dismutase with mutations linked to familial amyotrophic lateral sclerosis. Journal of the Neurological Sciences, 139(1), pp.15–20.

Hottinger, A.F. et al., 2000. Complete and long-term rescue of lesioned adult motoneurons by lentiviral-mediated expression of glial cell line-derived neurotrophic factor in the facial nucleus. The Journal of neuroscience : the official journal of the Society for Neuroscience, 20(15), pp.5587–5593.

Hu, F. et al., 2010. Sortilin-mediated endocytosis determines levels of the frontotemporal dementia protein, progranulin. Neuron, 68(4), pp.654–667. Available at: http://dx.doi.org/10.1016/j.neuron.2010.09.034.

129

Hukema, R.K. et al., 2014. A new inducible transgenic mouse model for C9orf72-associated GGGGCC repeat expansion supports a gain-of-function mechanism in C9orf72-associated ALS and FTD. Acta Neuropathologica Communications, 2(1), pp.2–5. Available at: http://www.actaneurocomms.org/content/2/1/166.

Ince, P.G. et al., 2003. Corticospinal tract degeneration in the progressive muscular atrophy variant of ALS. Neurology, 60(8), pp.1252–1258.

Irwin, D., Lippa, C.F. & Rosso, a., 2009. Progranulin (PGRN) expression in ALS: An immunohistochemical study. Journal of the Neurological Sciences, 276(1-2), pp.9–13. Available at: http://dx.doi.org/10.1016/j.jns.2008.08.024.

Jaarsma, D. et al., 2008. Neuron-specific expression of mutant superoxide dismutase is sufficient to induce amyotrophic lateral sclerosis in transgenic mice. The Journal of neuroscience : the official journal of the Society for Neuroscience, 28(9), pp.2075–2088.

Jaiswal, M.K. et al., 2009. Impairment of mitochondrial calcium handling in a mtSOD1 cell culture model of motoneuron disease. BMC neuroscience, 10(1), p.64.

Joyce, P.I. et al., 2011. SOD1 and TDP-43 animal models of amyotrophic lateral sclerosis: Recent advances in understanding disease toward the development of clinical treatments. Mammalian Genome, 22(7-8), pp.420–448.

Julien, J.-P., 2001. Amyotrophic Lateral Sclerosis: Unfolding the Toxicity of the Misfolded. Cell, 104(4), pp.581–591.

Kabashi, E. et al., 2007. Oxidized/misfolded superoxide dismutase-1: The cause of all amyotrophic lateral sclerosis? Annals of Neurology, 62(6), pp.553–559.

Kalaria, R.N., Mitchell, M.J. & Harik, S.I., 1987. Correlation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine neurotoxicity with blood-brain barrier monoamine oxidase activity. Proceedings of the National Academy of Sciences of the United States of America, 84(10), pp.3521–3525.

Van Kampen, J. & Kay, D., 2011. Progranulin gene therapy prevents plaque formation and synapse loss in a rodent model of AD. Alzheimer’s & Dementia, 7(4), pp.e7–e8. Available at: http://dx.doi.org/10.1016/j.jalz.2011.09.021.

Van Kampen, J.M., Baranowski, D. & Kay, D.G., 2014. Progranulin gene delivery protects dopaminergic neurons in a mouse model of Parkinson’s disease. PLoS ONE, 9(5).

Kato, S., 2008. Amyotrophic lateral sclerosis models and human neuropathology: Similarities and differences. Acta Neuropathologica, 115(1), pp.97–114.

130

Kelley, B.J. et al., 2009. Prominent phenotypic variability associated with mutations in Progranulin. Neurobiology of Aging, 30(5), pp.739–751.

Kieran, D. et al., 2007. Deletion of the BH3-only protein puma protects motoneurons from ER stress-induced apoptosis and delays motoneuron loss in ALS mice. Proceedings of the National Academy of Sciences of the United States of America, 104(51), pp.20606–20611.

Kihira, T. et al., 2004. ALS-like skin changes in mice on a chronic low-Ca/Mg high-Al diet. Journal of the Neurological Sciences, 219(1-2), pp.7–14.

Kimura, F. et al., 2006. Progression rate of ALSFRS-R at time of diagnosis predicts survival time in ALS. Neurology, 66(2), pp.265–267.

Kruman, I.I. et al., 1999. ALS-linked Cu/Zn-SOD mutation increases vulnerability of motor neurons to excitotoxicity by a mechanism involving increased oxidative stress and perturbed calcium homeostasis. Experimental neurology, 160(1), pp.28–39.

Labra, J. et al., 2015. Rate of disease progression: a prognostic biomarker in ALS. Journal of Neurology, Neurosurgery & Psychiatry, p.jnnp–2015–310998. Available at: http://jnnp.bmj.com/lookup/doi/10.1136/jnnp-2015-310998.

Laird, A.S. et al., 2010. Progranulin is neurotrophic in vivo and protects against a mutant TDP-43 induced axonopathy. PLoS ONE, 5(10), pp.1–7.

Lamprecht, M.R., Sabatini, D.M. & Carpenter, A.E., 2007. CellProfiler: Free, versatile software for automated biological image analysis. BioTechniques, 42(1), pp.71–75.

Lasiene, J. & Yamanaka, K., 2011. Glial cells in amyotrophic lateral sclerosis. Neurology Research International, 2011.

Liu, H.N. et al., 2009. Lack of evidence of monomer/misfolded superoxide dismutase-1 in sporadic amyotrophic lateral sclerosis. Annals of Neurology, 66(1), pp.75–80.

Macchi, Z. et al., 2015. A multi-center screening trial of rasagiline in patients with amyotrophic lateral sclerosis: Possible mitochondrial biomarker target engagement. Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration, (November 2014), pp.1–8. Available at: http://informahealthcare.com/doi/abs/10.3109/21678421.2015.1026826.

Mackenzie, I.R. a et al., 2007. Pathological TDP-43 distinguishes sporadic amyotrophic lateral sclerosis from amyotrophic lateral sclerosis with SOD1 mutations. Annals of Neurology, 61(5), pp.427–434.

Mandrioli, J. et al., 2006. Amyotrophic lateral sclerosis: prognostic indicators of survival. Amyotrophic lateral sclerosis : official publication of the World Federation of Neurology Research Group on Motor Neuron Diseases, 7(4), pp.211–220.

131

Marangi, G. & Traynor, B.J., 2014. Genetic causes of amyotrophic lateral sclerosis: New genetic analysis methodologies entailing new opportunities and challenges. Brain Research, 1607, pp.75–93. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0006899314013614.

Matsuda, T. et al., 2015. TLR9 signalling in microglia attenuates seizure-induced aberrant neurogenesis in the adult hippocampus. Nature Communications, 6, p.6514. Available at: http://www.nature.com/doifinder/10.1038/ncomms7514.

Matus, S., Glimcher, L.H. & Hetz, C., 2011. Protein folding stress in neurodegenerative diseases: A glimpse into the ER. Current Opinion in Cell Biology, 23(2), pp.239–252. Available at: http://dx.doi.org/10.1016/j.ceb.2011.01.003.

Matzilevich, D. a. et al., 2002. High-density microarray analysis of hippocampal gene expression following experimental brain injury. Journal of Neuroscience Research, 67(5), pp.646–663.

Mentis, G.Z. et al., 2006. Transduction of motor neurons and muscle fibers by intramuscular injection of HIV-1-based vectors pseudotyped with select rabies virus glycoproteins. Journal of Neuroscience Methods, 157(2), pp.208–217.

Miana-Mena, F.J. et al., 2005. Optimal methods to characterize the G93A mouse model of ALS. Amyotrophic lateral sclerosis and other motor neuron disorders : official publication of the World Federation of Neurology, Research Group on Motor Neuron Diseases, 6(1), pp.55–62.

Miller, R., Mitchell, J. & Moore, D., 2012. Riluzole for amyotrophic lateral sclerosis ( ALS )/ motor neuron disease ( MND ) ( Review ). , (3).

Minami, S.S. et al., 2014. Progranulin protects against amyloid β deposition and toxicity in Alzheimer’s disease mouse models. Nature Medicine, 20(10), pp.1157–1164.

Mitsumoto, H., Brooks, B.R. & Silani, V., 2014. Clinical trials in amyotrophic lateral sclerosis: why so many negative trials and how can trials be improved? The Lancet Neurology, 13(11), pp.1127–1138. Available at: http://linkinghub.elsevier.com/retrieve/pii/S1474442214701292.

Moisse, K. et al., 2009. Divergent patterns of cytosolic TDP-43 and neuronal progranulin expression following axotomy: Implications for TDP-43 in the physiological response to neuronal injury. Brain Research, 1249, pp.202–211. Available at: http://dx.doi.org/10.1016/j.brainres.2008.10.021.

De Muynck, L. & Van Damme, P., 2011. Cellular Effects of Progranulin in Health and Disease. Journal of Molecular Neuroscience, 45, pp.549–560.

Nakajima, H. et al., 2008. Target muscles for retrograde gene delivery to specific spinal cord segments. Neuroscience Letters, 435(1), pp.1–6.

132

Naldini, L. et al., 1996. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science (New York, N.Y.), 272(5259), pp.263–267.

Naudé, P.J.W. et al., 2011. Tumor necrosis factor receptor cross-talk. FEBS Journal, 278(6), pp.888–898.

Neary, D., Snowden, J. & Mann, D., 2005. Frontotemporal dementia. Lancet Neurology, 4(11), pp.771–780.

Neumann, M. et al., 2006. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science (New York, N.Y.), 314(5796), pp.130–133.

Nguyen, A.D. et al., 2013. Progranulin: At the interface of neurodegenerative and metabolic diseases. Trends in Endocrinology and Metabolism, 24(12), pp.597–606. Available at: http://dx.doi.org/10.1016/j.tem.2013.08.003.

Nguyen, M.D., Larivière, R.C. & Julien, J.P., 2001. Deregulation of Cdk5 in a mouse model of ALS: Toxicity alleviated by perikaryal neurofilament inclusions. Neuron, 30(1), pp.135–147.

Oliván, S. et al., 2015. Comparative study of behavioural tests in the SOD1G93A mouse model of amyotrophic lateral sclerosis. Experimental Animals, 64(2), pp.147–153.

Ono, S., 2000. The skin in amyotrophic lateral sclerosis. Amyotrophic lateral sclerosis and other motor neuron disorders : official publication of the World Federation of Neurology, Research Group on Motor Neuron Diseases, 1(3), pp.191–199.

Park, B. et al., 2011. Granulin Is a Soluble Cofactor for Toll-like Receptor 9 Signaling. Immunity, 34(4), pp.505–513.

Park, B. et al., 2009. Proteolytic cleavage in an endolysosomal compartment is required for Toll-like receptor 9 activation. Nature immunology, 9(12), pp.1407–1414. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2735466&tool=pmcentrez&rendertype=abstract.

Pasinelli, P. et al., 2004. Amyotrophic lateral sclerosis-associated SOD1 mutant proteins bind and aggregate with Bcl-2 in spinal cord mitochondria. Neuron, 43(1), pp.19–30.

Paylor, R. et al., 1999. Impaired Learning and Motor Behavior in Heterozygous Pafah1b1 (Lis1) Mutant Mice. Learning & Memory, 6(5), pp.521–537.

Pedrini, S. et al., 2010. ALS-linked mutant SOD1 damages mitochondria by promoting conformational changes in Bcl-2. Human Molecular Genetics, 19(15), pp.2974–2986.

133

Petkau, T.L. & Leavitt, B.R., 2014. Progranulin in neurodegenerative disease. Trends in Neurosciences, 37(7), pp.388–398. Available at: http://dx.doi.org/10.1016/j.tins.2014.04.003.

Philips, T. et al., 2010. Microglial upregulation of progranulin as a marker of motor neuron degeneration. Journal of neuropathology and experimental neurology, 69(12), pp.1191–1200.

Plaitakis, a & Constantakakis, E., 1993. Altered metabolism of excitatory amino acids, N-acetyl-aspartate and N-acetyl-aspartyl-glutamate in amyotrophic lateral sclerosis. Brain research bulletin, 30(3-4), pp.381–386.

Pokrishevsky, E. et al., 2012. Aberrant localization of FUS and TDP43 is associated with misfolding of SOD1 in amyotrophic lateral sclerosis. PLoS ONE, 7(4), pp.1–9.

Polymenidou, M. & Cleveland, D.W., 2008. Motor neuron disease: The curious ways of ALS. Nature, 454(7202), pp.284–285.

Pupillo, E. et al., 2014. Long-term survival in amyotrophic lateral sclerosis: A population-based study. Annals of Neurology, 75(2), pp.287–297.

Rakhit, R. et al., 2004. Monomeric Cu,Zn-superoxide Dismutase Is a Common Misfolding Intermediate in the Oxidation Models of Sporadic and Familial Amyotrophic Lateral Sclerosis. Journal of Biological Chemistry, 279(15), pp.15499–15504.

Rakhit, R. et al., 2002. Oxidation-induced misfolding and aggregation of superoxide dismutase and its implications for amyotrophic lateral sclerosis. Journal of Biological Chemistry, 277(49), pp.47551–47556.

Ratovitski, T. et al., 1999. Variation in the biochemical/biophysical properties of mutant superoxide dismutase 1 enzymes and the rate of disease progression in familial amyotrophic lateral sclerosis kindreds. Human Molecular Genetics, 8(8), pp.1451–1460.

Ravits, J. et al., 2013. Deciphering amyotrophic lateral sclerosis: what phenotype, neuropathology and genetics are telling us about pathogenesis. Amyotrophic lateral sclerosis & frontotemporal degeneration, 14 Suppl 1, pp.5–18. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3779649&tool=pmcentrez&rendertype=abstract.

Renton, A.E. et al., 2011. A hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron, 72(2), pp.257–268.

Renton, A.E., Chiò, A. & Traynor, B.J., 2014. State of play in amyotrophic lateral sclerosis genetics. Nature neuroscience, 17(1), pp.17–23. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24369373.

134

Riancho, J. et al., 2014. Compensatory Motor Neuron Response to Chromatolysis in the Murine hSOD1G93A Model of Amyotrophic Lateral Sclerosis. Frontiers in Cellular Neuroscience, 8, pp.1–13. Available at: http://journal.frontiersin.org/article/10.3389/fncel.2014.00346/abstract.

Richard, J.-P. & Maragakis, N.J., 2014. Induced pluripotent stem cells from ALS patients for disease modeling. Brain Research, 1607, pp.15–25. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0006899314012268.

Robberecht, W. & Philips, T., 2013. The changing scene of amyotrophic lateral sclerosis. Nature reviews. Neuroscience, 14(4), pp.248–64. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23463272.

Robelin, L. & Gonzalez De Aguilar, J.L., 2014. Blood biomarkers for amyotrophic lateral sclerosis: Myth or reality? BioMed Research International, 2014.

Roedel, A. et al., 2006. Effects of light or dark phase testing on behavioural and cognitive performance in DBA mice. Laboratory animals, 40(4), pp.371–381.

Rosen, D.R. et al., 1993. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature, 362(6415), pp.59–62. Available at: http://www.nature.com/nature/journal/v362/n6415/abs/362059a0.html\nhttp://www.ncbi.nlm.nih.gov/pubmed/8446170.

Rothstein, J.D., Martin, L.J. & Kuncl, R.W., 1992. Decreased Glutamate transport by the brain and spinal cord in amyotrophic lateral sclerosis. The New England Journal of Medicine, 326(22), pp.1464–1468.

Rotunno, M.S. & Bosco, D. a, 2013. An emerging role for misfolded wild-type SOD1 in sporadic ALS pathogenesis. Frontiers in cellular neuroscience, 7(December), p.253. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3863749&tool=pmcentrez&rendertype=abstract.

Rowland, L.P., 2001. How Amyotrophic Lateral Sclerosis Got Its Name. Archives of Neurology, 58(3), pp.512–515. Available at: http://archneur.jamanetwork.com/article.aspx?articleid=778657.

Ryan, C.L. et al., 2009. Progranulin is expressed within motor neurons and promotes neuronal cell survival. BMC neuroscience, 10, p.130.

Ryberg, H. & Bowser, R., 2008. Title: Protein biomarkers for amyotrophic lateral sclerosis Author(s): Henrik Ryberg and Robert Bowser Source: Expert Review of Proteomics, 5(2), pp.249–262.

135

Saeed, M. et al., 2009. Age and founder effect of SOD1 A4V mutation causing ALS. Neurology, 72(19), pp.1634–1639.

Sasaki, S. & Iwata, M., 1996. Ultrastructural study of synapses in the anterior horn neurons of patients with amyotrophic lateral sclerosis. Neuroscience letters, 204(1-2), pp.53–56.

Schulz, J.D., Wilson, J.M.. & Shaw, C.A., 2003. A Murine Model of ALS-PDC with Behavioral and Neuropathological Features of Parkinsonism. Annals of the New York Academy of Sciences, 991(1), pp.326–329.

Schymick, J.C. et al., 2007. Progranulin mutations and amyotrophic lateral sclerosis or amyotrophic lateral sclerosis-frontotemporal dementia phenotypes. Journal of neurology, neurosurgery, and psychiatry, 78(7), pp.754–756.

Scott, S. et al., 2008. Design, power, and interpretation of studies in the standard murine model of ALS. Amyotrophic lateral sclerosis : official publication of the World Federation of Neurology Research Group on Motor Neuron Diseases, 9(1), pp.4–15.

Seok, J. et al., 2013. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proceedings of the National Academy of Sciences of the United States of America, 110(9), pp.3507–12. Available at: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3587220&tool=pmcentrez&rendertype=abstract.

Shaw, P.J., 2005. Molecular and cellular pathways of neurodegeneration in motor neurone disease. Journal of neurology, neurosurgery, and psychiatry, 76(8), pp.1046–1057.

Shi, P. et al., 2010. Mitochondrial dysfunction in amyotrophic lateral sclerosis. Biochimica et biophysica acta, 1802(1), pp.45–51.

Singer, M.A. et al., 2005. Primary lateral sclerosis Clinical and laboratory features in 25 patients. Journal of clinical Neuromuscular Disease, 7(1), pp.1–9. Available at: http://journals.lww.com/jcnmd/Abstract/2005/09000/Primary_Lateral_Sclerosis__Clinical_and_Laboratory.1.aspx.

Sleegers, K. et al., 2008. Progranulin genetic variability contributes to amyotrophic lateral sclerosis. Neurology, 71, pp.253–259.

Spina, S. et al., 2007. Clinicopathologic features of frontotemporal dementia with. Neurology, pp.1–8.

Swash, M., Desai, J. & Misra, V.P., 1999. What is primary lateral sclerosis? Journal of the Neurological Sciences, 170(1), pp.5–10.

136

Van Swieten, J.C. & Heutink, P., 2008. Mutations in progranulin (GRN) within the spectrum of clinical and pathological phenotypes of frontotemporal dementia. The Lancet Neurology, 7(10), pp.965–974.

Swinnen, B. & Robberecht, W., 2014. The phenotypic variability of amyotrophic lateral sclerosis. Nature Reviews Neurology, 10(11), pp.661–670. Available at: http://www.nature.com/doifinder/10.1038/nrneurol.2014.184.

Tabata, R.C. et al., 2008. Chronic exposure to dietary sterol glucosides is neurotoxic to motor neurons and induces an ALS-PDC phenotype. NeuroMolecular Medicine, 10(1), pp.24–39.

Tai, H.-C. & Schuman, E.M., 2008. Ubiquitin, the proteasome and protein degradation in neuronal function and dysfunction. Nature reviews. Neuroscience, 9(11), pp.826–838.

Takeuchi, H. et al., 2002. Mitochondrial localization of mutant superoxide dismutase 1 triggers caspase-dependent cell death in a cellular model of familial amyotrophic lateral sclerosis. Journal of Biological Chemistry, 277(52), pp.50966–50972.

Tang, W. et al., 2011. The growth factor progranulin binds to TNF receptors and is therapeutic against inflammatory arthritis in mice. Science (New York, N.Y.), 332(6028), pp.478–484.

Turner, J.G. et al., 2005. Hearing in laboratory animals: Strain differences and nonauditory effects of noise. Comparative Medicine, 55(1), pp.12–23.

Turner, M.R. et al., 2013. Controversies and priorities in amyotrophic lateral sclerosis. The Lancet Neurology, 12(3), pp.310–322. Available at: http://dx.doi.org/10.1016/S1474-4422(13)70036-X.

Turner, M.R. et al., 2013. Mechanisms, models and biomarkers in amyotrophic lateral sclerosis. Amyotrophic lateral sclerosis & frontotemporal degeneration, 14 Suppl 1, pp.19–32. Available at: http://www.ncbi.nlm.nih.gov/pubmed/23678877.

Vaccaro, A. et al., 2012. Methylene blue protects against TDP-43 and FUS neuronal toxicity in C. elegans and D. rerio. PLoS ONE, 7(7).

Valentine, J.S., Doucette, P. a & Zittin Potter, S., 2005. Copper-zinc superoxide dismutase and amyotrophic lateral sclerosis. Annual review of biochemistry, 74, pp.563–593.

Vargas, M. & Johnson, J., 2010. Astrogliosis in Amyotrophic Lateral Sclerosis and therapeutic potential of astrocytes. Neurotherapeutics, 7(4), pp.471–481. Available at: http://www.sciencedirect.com/science/article/pii/S1933721310000723.

Vehviläinen, P., Koistinaho, J. & Gundars, G., 2014. Mechanisms of mutant SOD1 induced mitochondrial toxicity in amyotrophic lateral sclerosis. Frontiers in cellular neuroscience, 8, pp.1–7. Available at:

137

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=4023018&tool=pmcentrez&rendertype=abstract.

Wang, L., Popko, B., Tixier, E., et al., 2014. Guanabenz, which enhances the unfolded protein response, ameliorates mutant SOD1-induced amyotrophic lateral sclerosis. Neurobiology of Disease, 71, pp.317–324. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0969996114002411.

Wang, L., Popko, B. & Roos, R.P., 2014. An enhanced integrated stress response ameliorates mutant SOD1-induced ALS. Human Molecular Genetics, 23(10), pp.2629–2638.

Wegorzewska, I. et al., 2009. TDP-43 mutant transgenic mice develop features of ALS and frontotemporal lobar degeneration. Proceedings of the National Academy of Sciences of the United States of America, 106(44), pp.18809–18814.

Weydt, P. et al., 2003. Assessing disease onset and progression in the SOD1 mouse model of ALS. Neuroreport, 14(7), pp.1051–1054.

Wilkins, H.M. et al., 2011. Poor Correlation Between Drug Efficacies In The Mutant SOD1 Mouse Model Versus Clinical Trials Of ALS Necessitates The Development Of Novel Animal Models for Sporadic Motor Neuron Disease,

Williams, R.O. et al., 1995. Successful therapy of collagen- induced arthritis with TNF receptor-IgG fusion protein and combination with anti-CD4. Immunology, 84(3), pp.433–439.

Wilson, J.M.B. et al., 2002. Behavioral and neurological correlates of ALS-parkinsonism dementia complex in adult mice fed washed cycad flour. Neuromolecular medicine, 1(3), pp.207–221.

Wilson, J.M.B. et al., 2004. Quantitative measurement of neurodegeneration in an ALS-PDC model using MR microscopy. NeuroImage, 23(1), pp.336–343.

Wong, P.C. et al., 1995. An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria. Neuron, 14(6), pp.1105–1116.

Xu, J. et al., 2011. Extracellular progranulin protects cortical neurons from toxic insults by activating survival signaling. Neurobiology of Aging, 32(12), pp.5–16.

Yachnis, A.T. & Riviera-Zengotita, M.L., 2014. Neuropathology: a volume in the high yield pathology series, Philadelphia: Saunders.

Yamanaka, K. et al., 2008. Astrocytes as determinants of disease progression in inherited amyotrophic lateral sclerosis. Nature neuroscience, 11(3), pp.251–253.

138

Yamashita, S. & Ando, Y., 2015. Genotype-phenotype relationship in hereditary amyotrophic lateral sclerosis. Translational Neurodegeneration, 4(1), p.13. Available at: http://www.translationalneurodegeneration.com/content/4/1/13.

Yasui, K. et al., 2011. Increased progranulin in the skin of amyotrophic lateral sclerosis : An immunohistochemical study. Journal of the Neurological Sciences, 309(1-2), pp.110–114. Available at: http://dx.doi.org/10.1016/j.jns.2011.07.003.

Zhang, B. et al., 1997. Neurofilaments and orthograde transport are reduced in ventral root axons of transgenic mice that express human SOD1 with a G93A mutation. Journal of Cell Biology, 139(5), pp.1307–1315.

Zheng, Y. et al., 2011. C-terminus of progranulin interacts with the beta-propeller region of sortilin to regulate progranulin trafficking. PLoS ONE, 6(6).

Zhivotovsky, B. & Orrenius, S., 2011. Calcium and cell death mechanisms: A perspective from the cell death community. Cell Calcium, 50(3), pp.211–221. Available at: http://dx.doi.org/10.1016/j.ceca.2011.03.003.

Zwiegers, P., Lee, G. & Shaw, C. a, 2014. Reduction in hSOD1 copy number significantly impacts ALS phenotype presentation in G37R (line 29) mice: implications for the assessment of putative therapeutic agents. Journal of Negative Results in BioMedicine, 13(1), p.14. Available at: http://www.jnrbm.com/content/13/1/14.

139

Appendices

Appendix A Chapter 1 Supplementary Materials

A.1 Neuropathological Hallmarks of Amyotrophic Lateral Sclerosis

A.1.1 sALS Neuropathology

At the terminal end stage, the surviving motor neurons at autopsy show various pathological

changes that are indicative of neurodegeneration. A prominent feature of ALS is the presence of

ubiquitin positive cytoplasmic aggregates that are typically indicative of proteosomal

dysfunction (Tai & Schuman 2008; Al-Chalabi et al. 2012; Blokhuis et al. 2013). Gross

pathological abnormalities include signs of a degenerating spinal cord that presents with

shrunken anterior nerve roots (Yachnis & Riviera-Zengotita 2014). Evidence for increased

astrocytic and microglial proliferation are further indicative of the underlying neurodegenerative

changes that ultimately culminates in neuronal loss (Vargas & Johnson 2010; Lasiene &

Yamanaka 2011)

In sALS, important intracytoplasmic pathological features include Bunina body formation as

well as both skein-like and round hyaline inclusions (Kato 2008). Bunina bodies are small

granular inclusions that are usually restricted to the cytoplasm and dendrites, which are not

immunoreactive for ubiquitin (Kato 2008). Skein-like inclusions (SLIs) are thread-like

structures within the cytoplasm, whereas round hyaline inclusions (RHIs) are spherical

intracytoplasmic deposits (Yachnis & Riviera-Zengotita 2014). Both SLIs and RHIs show

positive ubiquitin immunolabeling (Kato 2008).

140

As previously indicated, additional immunohistological features are apparent that serves to

distinguish sporadic cases from fALS. Abnormal translocation from the nucleus leads to TDP-

43 positive cytoplasmic inclusions. This is a unique feature of sALS which fails to present in

SOD1 fALS cases, but is observed in patients with non-SOD1 fALS (e.g. in cases of TARDBP-

and C9ORF72 fALS) (Mackenzie et al. 2007; Al-Chalabi et al. 2012). Additionally, pathological

aggregates in sALS are typically immunonegative for epitopes selective for misfolded forms of

SOD1 (reviewed in Forsberg et al. 2010; Rotunno & Bosco 2013). This finding, however, is

highly dependent on the peptide sequence used to generate the antibody, and as such, non-

native/misfolded SOD1-positive immunolabeling has been described in some cases of sALS

(Forsberg et al. 2010).

A.1.2 fALS Neuropathology

Aberrant neuropathology in fALS can be broadly divided into two types: those that present with

pathological changes that are similar to sALS, and those that show posterior column involvement

and lack SLIs (which is typical in cases of SOD1-linked fALS) (Kato 2008).

Non-SOD1 fALS present with findings similar to sALS cases: Bunina bodies with ubiquitin

positive SLIs/RHIs which typically lack the presence of SOD1aggregates (Kato 2008).

As Kato points out, cases with posterior column involvement show degenerative changes in the

middle region of the column along with some features of sALS neuropathology (2008). Kato

further points out that certain SOD1-fALS cases with long survival times lack signs of Bunina

body formation, but show evidence for astrocytic hyaline inclusions (HI), while HI in neuronal

141

cells appear characteristically Lewy-body like (LBHI). These intracytoplasmic inclusions are

strongly immunoreactive for ubiquitin deposits and show positive labeling for SOD1.

Familial cases of ALS can be further subdivided based on the identification of the protein(s)

associated with the dense ubiquitin aggregates; which are typically a function of the underlying

genetic mutation. As previously mentioned, SOD1-fALS present with inclusions that are positive

for the mutant protein, but lack evidence of aberrant TDP-43 localization (Mackenzie et al.

2007). Additionally, SOD1-fALS protein aggregates can also be positive for UBQLN2 and

C9ORF72 peptides (Blokhuis et al. 2013). Familial cases linked to TDP43 show additional

immunopositive labeling for FUS and OPTN but are negative for C9ORF72 (Blokhuis et al.

2013). Interestingly, C9ORF72-fALS lacks labeling of the mutant protein, but can present with

TDP-43, OPTN, and UBQLN2 aggregates (Blokhuis et al. 2013).

A.2 Clinical Phenotypic Heterogeneity of Amyotrophic Lateral Sclerosis

Temporally progressive muscle weakness initiated in discrete CNS regions (i.e. brainstem,

cervical-, thoracic-, and lumbosacral regions of spinal cord) are prototypical symptoms of

disease (Brooks et al. 2000; Ravits et al. 2013). Clinical presentation of ALS has long been

shown to differ in the extent of upper and lower motor neuron involvement, the initial site of

onset, the rate of disease progression, as well as in the degree of cognitive impairment. This

heterogeneity informs the current understanding of ALS as a syndrome and has led to the

stratification of patient cohorts in contemporary clinical trials to more appropriately investigate

therapeutic compounds.

142

A.2.1 Extent of Upper and Lower Motor Neuron Involvement

A cardinal feature of ALS involves the degenerative changes that affect neurons of the motor

cortex (upper motor neurons, or UMNs) and/or those that directly innervate skeletal muscles and

extend from the brainstem or spinal cord (lower motor neurons, or LMNs). In “classical ALS”

the involvement of U- and LMNs give rise to the characteristically progressive weakness that

initiates at specific body regions (Ravits et al. 2013). The degree of upper and/or lower motor

neuron involvement varies in ALS and this gives rise to a wide array of clinical phenotypes that

present on a continuous spectrum with primary lateral sclerosis and progressive muscular

atrophy occupying the extremes (Swinnen & Robberecht 2014).

Primary lateral sclerosis (PLS) denotes a rare condition that predominantly manifests as a result

UMN involvement. In the majority of PLS patients, muscle weakness typically first manifests in

the legs and progresses symmetrically to the bulbar and arm muscles (Ravits et al. 2013).

Compared to classical ALS, PLS with strict UMN involvement shows an attenuated rate of

progression which concomitantly presents with a decreased disease burden and an associated

increase in lifespan (Singer et al. 2005; Swinnen & Robberecht 2014). On the other hand, PLS

patients who develop signs of LMN degeneration typically present with an accelerated disease

course (Gordon et al. 2009). The degree of LMN involvement in the pathophysiology of PLS

remains contentious: in a subset of patients with a clinical diagnosis of PLS, findings at autopsy

revealed degenerative changes in LMNs, which could eventually have predisposed the patients to

present with a more ALS-like clinicopathology (Swash et al. 1999). PLS can thus essentially be

thought of as ALS lacking LMN involvement; with the latter phenotype expected to become

143

clinically established during the course of disease pathogenesis as additional LMNs degenerate

(Swash et al. 1999; Swinnen & Robberecht 2014).

In contrast to PLS, progressive muscular atrophy (PMA) affects primarily the LMNs and present

clinically with paralysis and muscle atrophy with onset in any limb region (Cervenakova et al.

2000; Ravits et al. 2013; Swinnen & Robberecht 2014). Typically there is no UMN involvement,

however a proportion of patients with PMA will ultimately present with signs of UMN

degeneration: an observation linking this phenotype to ALS (Ince et al. 2003; Ravits et al. 2013).

Clinical presentation of PMA lacking any UMN involvement show signs of a lessened disease

severity, and thus, better prognosis (Swinnen & Robberecht 2014). Further underscoring the

putative link between PMA and ALS is the finding that certain SOD1-linked forms of fALS

(specifically the A4V and D101N mutations) show signs of progressive limb weakness without

any clinical indication of upper motor neuron degeneration (Cudkowicz et al. 1998;

Cervenakova et al. 2000).

A.2.2 Initial Site of Onset

Symptoms of muscle weakness can start in the upper or lower limbs, as well as in the bulbar

musculature. Clinical observations have shown that the initial site of onset influences the severity

of the disease cascade, which underscores the use of this clinical parameter in stratifying patient

cohorts for clinical trials. The majority of patients (65%) show signs of asymmetric limb onset

(LO) which present with muscle weakness in the legs or arms accompanied by fasciculations,

while an additional 30% present with a bulbar phenotype (BO) where the initial symptoms

include dysarthria and dysphagia (Hardiman et al. 2011; Ravits et al. 2013; Swinnen &

144

Robberecht 2014). Patients with a definitive onset in the bulbar muscles suffer from a poorer

prognosis compared to their limb-onset counterparts. When diagnosed, these BO patients are

typically older and show a median survival time of two years, while LO patients exhibit a mean

survival of 2.6 years (Chiò et al. 2011). The poorer prognosis in BO patients is thought primarily

to arise from the fact that this cohort is expected to experience issues with aspiration and show

earlier signs of respiratory dysfunction (Swinnen & Robberecht 2014).

A.2.3 Rate of Disease Progression

The rates at which degenerative changes become increasingly established during the course of

disease are a major confounding factor in conducting clinical trials. Further underscoring the

underlying phenotypic heterogeneity, ALS patients present along a continuum of disease

progression. On one end, patients with a rapidly deteriorating phenotype show evidence of

survival times that are less than 3 years after the initial disease onset (Yamashita & Ando 2015).

Patients experiencing this phenotypically aggressive form of ALS typically are diagnosed at a

later stage of life, is bulbar in onset with symptoms that first manifest in the upper limbs

(Mandrioli et al. 2006).

At the other end of the spectrum, patients can present with an attenuated rate of disease

progression and survive with ALS symptoms for more than a decade (Pupillo et al. 2014). As

Pupillo et al point out, certain prognostic factors associated with this lessened disease burden

include signs of spinal onset, younger age at diagnosis and male gender (2014). Differences in

disease progression are evident even between the various fALS mutations. Case in point: patients

with the A4V mutation typically survive for a year following disease onset, while those carrying

145

the D90A SOD1 mutant, show a more attenuated rate of disease progression (Renton et al. 2014;

Yamashita & Ando 2015).

Observations such as these have led to clinical trial protocols that attempt to stratify patients

according to the approximate rate of disease progression. The revised ALS functional rating

scale (ALSFRS-R) is an instrument that is frequently used to assess the disease burden

experienced by patients (Cedarbaum et al. 1999). One of the limitations of this measure is that it

does not take into account the effect of symptom duration (Kimura et al. 2006). Expressing the

ALSFRS-R score as a function of disease duration provides a clinically-relevant measure of

disease progression (ΔFS; change in functional score) that can be employed to stratify patient

cohorts (Kimura et al. 2006; Labra et al. 2015). ΔFS scores at the initial clinical assessment was

determined to be a significant predictor of overall survival and was useful in delineating cohorts

of patients with varying clinical outcomes (Labra et al. 2015). The group not only found an

inverse correlation between median survival times and ΔFS scores, but showed that higher ΔFS

scores were associated with a group of patients that were generally older, showed predominant

signs of bulbar involvement and experienced a shorter duration of disease.

A.2.4 Degree of Cognitive Impairment

Over the previous decade, multiple research outcomes have indicated that ALS and

frontotemporal lobar dementia (FTLD) exist on a spectrum of clinical manifestation ranging

from solely motor neuron involvement to purely frontotemporal neuron degeneration (Swinnen

& Robberecht 2014). As summarized by Swinnen and Robberecht, clues regarding the potential

overlap were clinically evident (2014). Nearly a quarter of all ALS patients meet the criteria for

146

clinical FTLD and it is likely that a larger proportion show some milder frontal lobe

involvement. On the flipside, nearly 50% of FTD patients show some signs of motor system

involvement. Aside from the clinical overlap between these conditions, the linkage was more

clearly established when the shared underlying pathology was described for some cases.

In 2006, the identification of ubiquitinated inclusions that were positive for translocated TDP-43

in cases of sALS and FTLD suggested a putative link between apparently discordant disorders of

the CNS (Neumann et al. 2006). Subsequent immunohistochemical assessments were able to

demonstrate TDP-43 positive cytoplasmic inclusions in cases of sALS/fALS with dementia, but

failed to show positive TDP-43 immunoreactivity in cases of SOD1-ALS (Mackenzie et al.

2007).

Further strengthening the link between the extremes of the ALS-FTD clinical spectrum was the

2011 discovery of the intronic hexanucleotide repeat expansion in the C9ORF72 locus (Renton

et al. 2011; DeJesus-Hernandez et al. 2011). This repeat expansion explains nearly 40% of fALS

and approximately a quarter of inherited FTD cases (Renton et al. 2014).

A.3 Binding Partners for Progranulin

Due to the pleiotropic properties exhibited by this unique gene product in both its full length- and

cleaved peptide- forms, insight can be gleaned by identifying those proteins that this motif can

interact with and through which it can mediate intracellular signaling cascades (Figure A.1)

147

Figure A.1 Putative extracellular binding partners mediate both cellular uptake and varied intracellular effects 12

i. Progranulin can act as an antagonist to TNFα-mediated signaling and thus inhibits the activation of cytokines that normally result in cell death. ii. PGRN interacts with pathogenic CpG ODNs and facilitates TLR9 recognition of the pathogenic motif which ultimately results in the activation of the immune cell. iii. Extracellular neuronal sortilin facilitates receptor-mediated endocytosis and thus regulates extracellular levels of PGRN. Inset: Full-length PGRN or cleaved GRNs evoke specific cellular responses.

Cellular elements for Figure A.1 provided by Servier medical art and modified under a Creative Commons Attribution 3.0 Unported License.

148

A.3.1 Sortilin: Trafficking Extracellular Progranulin into the Endolysosome

Sortilin, a member within the Vps10 family, has been localized to expression within motor

neurons (Hu et al. 2010). On account of PGRN being a secreted growth factor, binding partners

at the cell-surface is of crucial importance in order to facilitate the growth factor’s multiple

effects on cellular physiology and biochemistry. Hu and colleagues have described an endocytic

pathway that targets extracellular PGRN to lysosomal localization through Sortilin, and this

receptor has been shown to modulate extracellular levels in vivo (2010). Recent work has shown

that the C-terminal GRN E domain specifically interacts with the β-propeller region of sortilin,

and that this interaction facilitates the intracellular trafficking of the growth factor (Zheng et al.

2011). Whether the binding of PGRN to sortilin directly mediates any downstream effects on

cellular physiology remains to be fully elucidated. Current evidence suggests that the sortilin

receptor is crucial in routing extracellular PGRN through the endolysosomal pathway and that a

yet uncharacterized interaction mediates the observed neurotrophic effects at the cellular level.

Thus, a binding interaction with sortilin is not the sole route through which PGRN exerts its

cellular neurotrophic effect. Gass et al demonstrated that neuronal outgrowth in Grn -/- primary

hippocampal neuronal cultures can be restored with the exogenous addition of recombinant

PGRN; however, this positive effect on cellular morphology is still evident when exogenous

PGRN is added to SORT -/- primary hippocampal cultures (2012). It is clear that PGRN plays a

crucial role in modulating neurite outgrowth, but that this effect is independent of the endocytic

uptake mediated via the sortilin receptor.

149

A.3.2 Progranulin Acts as a Co-factor for Endolysosomal TLR9-mediated Responses

The innate immune system demarcates a finely orchestrated cascade of events designed to

respond to conserved bacterial or viral pathogenic motifs. One such pattern recognizing receptor,

Toll-like receptor 9 (TLR9), binds to CpG oligodeoxynucleotide motifs (CpG-ODN) and

initiates a signaling cascade that ultimately drives the cellular production of cytokines that aid in

the innate and eventual adaptive immune responses. To date, TLR9 expression in the CNS has

been described and found to be localized to microglia (Matsuda et al. 2015). Produced as a

precursor protein within the endoplasmic reticulum, TLR9 signaling is primed following

proteolytic cleavage within the endolysomal compartment (Park et al. 2009). Park and colleagues

were further able to demonstrate that progranulin (and/or cleaved granulins) acts as co-factor;

playing a critical role in mediating an innate immune response to foreign DNA particulates

through TLR9-mediated signaling (2011). Addition of exogenous PGRN to progranulin-deficient

macrophages not only restored the delivery of CpG-ODNs to the endolysomal compartment, but

aided in the interaction of TLR9 and the pathogenic motif (Park et al. 2011).

A.3.3 Progranulin: an Antagonist of the TNF Receptor-mediated Signaling Cascade

Signals denoting critical information regarding the extracellular milieu are of great importance in

regulating various effects on cellular physiology. TNFα is a cytokine that typically plays a role in

immune cell regulation and is primarily involved in inflammatory processes and initiating

apoptosis. The intracellular signaling cascade mediated through the interaction of the TNFα

ligand and its associated cell-surface receptor (TNFR1 and/or TNFR2) play a crucial role in

determining the ultimate fate of the cell. TNFR1 is ubiquitously expressed primarily within cells

150

of the immune system, while TNFR2 expression is restricted and can be found in astrocytes,

microglia, and certain neuronal sub-types (Naudé et al. 2011).

Using a yeast two-hybrid strategy, Tang et al have demonstrated that domains within the

progranulin peptide can show selective antagonism of TNFα directed cell signaling mediated via

TNF receptors (2011). The PGRN FAC domains and the associated linker regions were shown to

exhibit an increased (decreased) binding affinity to TNFR2 (TNFR1) compared to TNFα, and

thus exert multiple effects on the cellular signaling cascade. The group was able to demonstrate

that the delivery of recombinant PGRN in PGRN-deficient mice could reverse collagen-induced

arthritis (Tang et al. 2011).

Further investigating the minimal residues required for TNFR binding, Atsttrin, a fusion protein

comprised of approximately 2/3 of the FAC construct and the associated linker regions, has been

found to be sufficient in interfering with the TNFα-mediated expression of inflammatory arthritis

in murine models (Tang et al. 2011). Thus binding of particular PGRN domains and TNFR2

plays a pivotal role in ameliorating inflammatory disease processes (Figure A.1).

The inhibitory PGRN interaction with TNFα receptors, however, is not without controversy.

Follow up studies using recombinant forms of PGRN have not been able to replicate the

interaction with TNFRs (Chen et al. 2013; Etemadi et al. 2013). A potential explanatory variable

underscoring this discrepancy may the conformational state of the commercially-obtained

recombinant proteins. Alternatively, the original interaction observed by Tang may have been an

experimental artefact and the inhibitory effects of PGRN may be independent of directly

151

inhibiting the TNFα cascade. On account of the anti-inflammatory properties mediated via

PGRN, the inhibitory interaction with TNFRs provides for an alluring explanation; however

more work is required to clarify PGRNs role in TNFα-mediated inflammatory processes

(Nguyen et al. 2013).

Appendix B List of Chemical Reagents

Chemicals/Reagents (abbreviation, formula) Source Acetic Acid (CH3COOH) Sigma Acetone (C3H6O) Fisher Aerrane (isoflurane) Baxter Buffered Formalin, 10% (CH2O) Fisher Citric acid monohydrate (C6H10O8) Fisher Cresyl violet acetate (C18H15N3O3) Sigma Ethanol (EtOH, C2H6O) VGH Glacial acetic acid (CH3COOH) Sigma Hematoxylin (C16H14O6) Fisher Hydrogen peroxide, 30% (H2O2) Fisher Lithium carbonate (Li2CO3) BDH Methanol (MeOH, CH4O) Fisher PGRN and GFP lentiviral construct design, initial production Invitrogen PGRN and GFP lentiviral construct production ABM Potassium chloride (KCl) Fisher Potassium phosphate monobasic (KH2PO4) Fisher Sodium cholride (NaCl) Fisher Sodiumphosphate dibasic (Na2HPO4) Fisher Sucrose (C12H22O11) Fisher Triton X-100 (C14H22O(C2H4O)n (n=9-10)) Fisher Tween-20 (C58H114O26) Fisher Xylene (C8H10) Fisher

152

Immunohistological Reagents Antibodies and Assorted Histological supplies (Abbr; Cat#) Source Concentration Goat α-Choline Acetyltransferase (ChAT; AB144P) Millipore 1:100 Goat α-Rabbit biotinylated IgG (BA-1000) Vector 1:200 Horse α-Goat biotinylated IgG(BA-9500) Vector 1:200 Normal Goat Serum (NGS; S-1000) Vector 10% Normal Horse Serum (NHS; S-2000) Vector 10% Permount mounting media Fisher -- Rabbit α-Green Fluorescent Protein (GFP, ab183734) Abcam 1:50 Rabbit α-Glial Fibrillary Acidic Protein (GFAP, ab68428) Abcam 1:200 Rabbit α-Ionized Ca2+ binding adaptor molecule-1 (Iba-1; 019-19741) Wako 1:100 SIGMAFAST 3,3'-Diaminobenzidine tablets Sigma -- Vectastain Elite ABC kit Vector --

General Laboratory Supplies and Equipment Other Supplies and Laboratory Equipment Source B5 Professional Series light microscope Motic Cryostat (CM3050 S) Leica Biosystems Dissecting equipment Shaw Lab; Jack Bell

Research Centre Glass coverslips VWR Needles and syringes BD Biosciences Optimum Cutting Temperature (O.C.T) medium VWR Peristaltic pump Masterflex pH meter (Symphony SB70P) VWR Pipettes Gilson Razor blades VWR Superfrost Plus charged glass slides Fisher Scientific Ten and 20 µL syringes Hamilton Tissue-Tek cryomolds Sakura Finetek USA Inc Transgenic and wild-type breeder sets The Jackson Laboratory Wirehang apparatus Shaw Lab

153

Appendix C Preliminary Findings

C.1 Neuroprotective Effects of Progranulin

Figure C.1 Preliminary findings exploring the neuroprotective properties of progranulin

(A) Cell survival assay (MTT colorimetric assay) in PC-12 cells evaluating the protective properties of 4nM PGRN in the presence of 100-1000 µM of MPTP. There was a significant increase in the percentage of surviving cells when PGRN was present for each of the four doses of MPTP treatment. Mean ± SEM and Student t-test results are shown.(B) ChAT immunolabeling of motor neurons in the ventral horn of lumbar spinal cord sections. G37R (line 29) mice and wild-type conspecifics received bilateral injections of Progranulin (PGRN) or GFP-containing lentiviral constructs directly into the gastrocnemius muscles (8.2x107 IU/mL) at 52.5 weeks of age and were euthanized 3.5 weeks later. (C) Quantification ChAT immunolabeling revealed a significant neuroprotective effect of PGRN in the fALS group. PGRN cDNA injected animals expressing mutant SOD1 exhibit more surviving neurons compared to the conspecific group receiving control GFP injections. Panels B and C were provided by Drs. CA Shaw, Denis G. Kay, J Van Kampen & G Lee.

154

C.2 Targeted GFP-expression in Lumbar Cord Motor Neurons Following Lentiviral

Transduction

Figure C.2 Temporal expression of retrogradely transported GFP cDNA in lumbar spinal cord (LSc) tissues following gastrocnemius muscle viral transfection

Male CD-1 animals received bilateral administration of a GFP-expressing lentivirus into both gastrocnemius muscles (5x2µL lentiviral vector at 4.0x108 TU/mL per muscle). One-, two- and four weeks post-transfection, transduction efficiency was qualitatively assessed via anti-GFP activity in lumbar cord motor neurons; establishing the applicability of retrograde targeting of motor neurons via intramuscular delivery. Images taken at 100x magnification.

155

Appendix D Viral Constructs

D.1 Progranulin Lentiviral Construct and Gene Sequence

pLenti6/V 5-mGranulin vector Sequence:

1 aagggctcga gtctagaggg cccgcggttc gaaggtaagc ctatccctaa 51 ccctctcctc ggtctcgatt ctacgcgtac cggttagtaa tgagtttgga 101 attaattctg tggaatgtgt gtcagttagg gtgtggaaag tccccaggct 151 ccccaggcag gcagaagtat gcaaagcatg catctcaatt agtcagcaac 201 caggtgtgga aagtccccag gctccccagc aggcagaagt atgcaaagca 251 tgcatctcaa ttagtcagca accatagtcc cgcccctaac tccgcccatc 301 ccgcccctaa ctccgcccag ttccgcccat tctccgcccc atggctgact 351 aatttttttt atttatgcag aggccgaggc cgcctctgcc tctgagctat 401 tccagaagta gtgaggaggc ttttttggag gcctaggctt ttgcaaaaag 451 ctcccgggag cttgtatatc cattttcgga tctgatcagc acgtgttgac 501 aattaatcat cggcatagta tatcggcata gtataatacg acaaggtgag 551 gaactaaacc atggccaagc ctttgtctca agaagaatcc accctcattg 601 aaagagcaac ggctacaatc aacagcatcc ccatctctga agactacagc 651 gtcgccagcg cagctctctc tagcgacggc cgcatcttca ctggtgtcaa 701 tgtatatcat tttactgggg gaccttgtgc agaactcgtg gtgctgggca 751 ctgctgctgc tgcggcagct ggcaacctga cttgtatcgt cgcgatcgga 801 aatgagaaca ggggcatctt gagcccctgc ggacggtgcc gacaggtgct

156

851 tctcgatctg catcctggga tcaaagccat agtgaaggac agtgatggac 901 agccgacggc agttgggatt cgtgaattgc tgccctctgg ttatgtgtgg 951 gagggctaag cacaattcga gctcggtacc tttaagacca atgacttaca 1001 aggcagctgt agatcttagc cactttttaa aagaaaaggg gggactggaa 1051 gggctaattc actcccaacg aagacaagat ctgctttttg cttgtactgg 1101 gtctctctgg ttagaccaga tctgagcctg ggagctctct ggctaactag 1151 ggaacccact gcttaagcct caataaagct tgccttgagt gcttcaagta 1201 gtgtgtgccc gtctgttgtg tgactctggt aactagagat ccctcagacc 1251 cttttagtca gtgtggaaaa tctctagcag tagtagttca tgtcatctta 1301 ttattcagta tttataactt gcaaagaaat gaatatcaga gagtgagagg 1351 aacttgttta ttgcagctta taatggttac aaataaagca atagcatcac 1401 aaatttcaca aataaagcat ttttttcact gcattctagt tgtggtttgt 1451 ccaaactcat caatgtatct tatcatgtct ggctctagct atcccgcccc 1501 taactccgcc catcccgccc ctaactccgc ccagttccgc ccattctccg 1551 ccccatggct gactaatttt ttttatttat gcagaggccg aggccgcctc 1601 ggcctctgag ctattccaga agtagtgagg aggctttttt ggaggcctag 1651 ggacgtaccc aattcgccct atagtgagtc gtattacgcg cgctcactgg 1701 ccgtcgtttt acaacgtcgt gactgggaaa accctggcgt tacccaactt 1751 aatcgccttg cagcacatcc ccctttcgcc agctggcgta atagcgaaga 1801 ggcccgcacc gatcgccctt cccaacagtt gcgcagcctg aatggcgaat 1851 gggacgcgcc ctgtagcggc gcattaagcg cggcgggtgt ggtggttacg 1901 cgcagcgtga ccgctacact tgccagcgcc ctagcgcccg ctcctttcgc 1951 tttcttccct tcctttctcg ccacgttcgc cggctttccc cgtcaagctc 2001 taaatcgggg gctcccttta gggttccgat ttagtgcttt acggcacctc 2051 gaccccaaaa aacttgatta gggtgatggt tcacgtagtg ggccatcgcc 2101 ctgatagacg gtttttcgcc ctttgacgtt ggagtccacg ttctttaata 2151 gtggactctt gttccaaact ggaacaacac tcaaccctat ctcggtctat 2201 tcttttgatt tataagggat tttgccgatt tcggcctatt ggttaaaaaa 2251 tgagctgatt taacaaaaat ttaacgcgaa ttttaacaaa atattaacgc 2301 ttacaattta ggtggcactt ttcggggaaa tgtgcgcgga Acccctattt 2351 gtttattttt ctaaatacat tcaaatatgt atccgctcat gagacaataa 2401 ccctgataaa tgcttcaata atattgaaaa aggaagagta tgagtattca 2451 acatttccgt gtcgccctta ttcccttttt tgcggcattt tgccttcctg 2501 tttttgctca cccagaaacg ctggtgaaag taaaagatgc tgaagatcag 2551 ttgggtgcac gagtgggtta catcgaactg gatctcaaca gcggtaagat 2601 ccttgagagt tttcgccccg aagaacgttt tccaatgatg agcactttta 2651 aagttctgct atgtggcgcg gtattatccc gtattgacgc cgggcaagag 2701 caactcggtc gccgcataca ctattctcag aatgacttgg ttgagtactc 2751 accagtcaca gaaaagcatc ttacggatgg catgacagta agagaattat 2801 gcagtgctgc cataaccatg agtgataaca ctgcggccaa cttacttctg 2851 acaacgatcg gaggaccgaa ggagctaacc gcttttttgc acaacatggg 2901 ggatcatgta actcgccttg atcgttggga accggagctg aatgaagcca 2951 taccaaacga cgagcgtgac accacgatgc ctgtagcaat ggcaacaacg 3001 ttgcgcaaac tattaactgg cgaactactt actctagctt cccggcaaca 3051 attaatagac tggatggagg cggataaagt tgcaggacca cttctgcgct

157

3101 cggcccttcc ggctggctgg tttattgctg ataaatctgg agccggtgag 3151 cgtgggtctc gcggtatcat tgcagcactg gggccagatg gtaagccctc 3201 ccgtatcgta gttatctaca cgacggggag tcaggcaact atggatgaac 3251 gaaatagaca gatcgctgag ataggtgcct cactgattaa gcattggtaa 3301 ctgtcagacc aagtttactc atatatactt tagattgatt taaaacttca 3351 tttttaattt aaaaggatct aggtgaagat cctttttgat aatctcatga 3401 ccaaaatccc ttaacgtgag ttttcgttcc actgagcgtc agaccccgta 3451 gaaaagatca aaggatcttc ttgagatcct ttttttctgc gcgtaatctg 3501 ctgcttgcaa acaaaaaaac caccgctacc agcggtggtt tgtttgccgg 3551 atcaagagct accaactctt tttccgaagg taactggctt cagcagagcg 3601 cagataccaa atactgttct tctagtgtag ccgtagttag gccaccactt 3651 caagaactct gtagcaccgc ctacatacct cgctctgcta atcctgttac 3701 cagtggctgc tgccagtggc gataagtcgt gtcttaccgg gttggactca 3751 agacgatagt taccggataa ggcgcagcgg tcgggctgaa cggggggttc 3801 gtgcacacag cccagcttgg agcgaacgac ctacaccgaa ctgagatacc 3851 tacagcgtga gctatgagaa agcgccacgc ttcccgaagg gagaaaggcg 3901 gacaggtatc cggtaagcgg cagggtcgga acaggagagc gcacgaggga 3951 gcttccaggg ggaaacgcct ggtatcttta tagtcctgtc gggtttcgcc 4001 acctctgact tgagcgtcga tttttgtgat gctcgtcagg ggggcggagc 4051 ctatggaaaa acgccagcaa cgcggccttt ttacggttcc tggccttttg 4101 ctggcctttt gctcacatgt tctttcctgc gttatcccct gattctgtgg 4151 ataaccgtat taccgccttt gagtgagctg ataccgctcg ccgcagccga 4201 acgaccgagc gcagcgagtc agtgagcgag gaagcggaag agcgcccaat 4251 acgcaaaccg cctctccccg cgcgttggcc gattcattaa tgcagctggc 4301 acgacaggtt tcccgactgg aaagcgggca gtgagcgcaa cgcaattaat 4351 gtgagttagc tcactcatta ggcaccccag gctttacact ttatgcttcc 4401 ggctcgtatg ttgtgtggaa ttgtgagcgg ataacaattt cacacaggaa 4451 acagctatga ccatgattac gccaagcgcg caattaaccc tcactaaagg 4501 gaacaaaagc tggagctgca agcttaatgt agtcttatgc aatactcttg 4551 tagtcttgca acatggtaac gatgagttag caacatgcct tacaaggaga 4601 gaaaaagcac cgtgcatgcc gattggtgga agtaaggtgg tacgatcgtg 4651 ccttattagg aaggcaacag acgggtctga catggattgg acgaaccact 4701 gaattgccgc attgcagaga tattgtattt aagtgcctag ctcgatacat 4751 aaacgggtct ctctggttag accagatctg agcctgggag ctctctggct 4801 aactagggaa cccactgctt aagcctcaat aaagcttgcc ttgagtgctt 4851 caagtagtgt gtgcccgtct gttgtgtgac tctggtaact agagatccct 4901 cagacccttt tagtcagtgt ggaaaatctc tagcagtggc gcccgaacag 4951 ggacttgaaa gcgaaaggga qaccagagga gctctctcga cgcaggactc 5001 ggcttgctga agcgcgcacg gcaagaggcg aggggcggcg actggtgagt 5051 acgccaaaaa ttttgactatg cggaggctag aaggagagag atgggtgcga 5101 gagcgtcagt attaagcggg ggagaattag atcgcgatgg gaaaaaattc 5151 ggttaaggcc agggggaaag aaaaaatata aattaaaaca tatagtatgg 5201 gcaagcaggg agctagaacg attcgcagtt aatcctggcc tgttagaaac 5251 atcagaaggc tgtagacaaa tactgggaca gctacaacca tcccttcaga 5301 caggatcaga agaacttaga tcattatata atacagtagc aaccctctat

158

5351 tgtgtgcatc aaaggataga gataaaagac accaaggaag ctttagacaa 5401 gatagaggaa gagcaaaaca aaagtaagac caccgcacag caagcggccg 5451 ctgatcttca gacctggagg aggagatatg agggacaatt ggagaagtga 5501 attatataaa tataaagtag taaaaattga accattagga gtagcaccca 5551 ccaaggcaaa gagaagagtg gtgcagagag aaaaaagagc agtgggaata 5601 ggagctttgt tccttgggtt cttgggagca gcaggaagca ctatgggcgc 5651 agcgtcaatg acgctgacgg tacaggccag acaattattg tctggtatag 5701 tgcagcagca gaacaatttg ctgagggcta ttgaggcgca acagcatctg 5751 ttgcaactca cagtctgggg catcaagcag ctccaggcaa gaatcctggc 5801 tgtggaaaga tacctaaagg atcaacagct cctggggatt tggggttgct 5851 ctggaaaact catttgcacc actgctgtgc cttggaatgc tagttggagt 5901 aataaatctc tggaacagat ttggaatcac acgacctgga tggagtggga 5951 cagagaaatt aacaattaca caagcttaat acactcctta attgaagaat 6001 cgcaaaacca gcaagaaaag aatgaacaag aattattgga attagataaa 6051 tgggcaagtt tgtggaattg gtttaacata acaaattggc tgtggtatat 6101 aaaattattc ataatgatag taggaggctt ggtaggttta agaatagttt 6151 ttgctgtact ttctatagtg aatagagtta ggcagggata ttcaccatta 6201 tcgtttcaga cccacctccc aaccccgagg ggacccgaca ggcccgaagg 6251 aatagaagaa gaaggtggag agagagacag agacagatcc attcgattag 6301 tgaacggatc tcgacggtat cgataagctt gggagttccg cgttacataa 6351 cttacggtaa atggcccgcc tggctgaccg cccaacgacc cccgcccatt 6401 gacgtcaata atgacgtatg ttcccatagt aacgccaata gggactttcc 6451 attgacgtca atgggtggag tatttacggt aaactgccca cttggcagta 6501 catcaagtgt atcatatgcc aagtacgccc cctattgacg tcaatgacgg 6551 taaatggccc gcctggcatt atgcccagta catgacctta tgggactttc 6601 ctacttggca gtacatctac gtattagtca tcgctattac catggtgatg 6651 cggttttggc agtacatcaa tgggcgtgga tagcggtttg actcacgggg 6701 atttccaagt ctccacccca ttgacgtcaa tgggagtttg ttttggcacc 6751 aaaatcaacg ggactttcca aaatgtcgta acaactccgc cccattgacg 6801 caaatgggcg gtaggcgtgt acggtgggag gtctatataa gcagagctcg 6851 tttagtgaac cgtcagatcg cctggagacg ccatccacgc tgttttgacc 6901 tccatagaag acaccgactc tagaggatcc actagtccag tgtggtggaa 6951 ttgatccctt caccatgtgg gtcctgatga gctggctggc cttcgcggca 7001 gggctggtag ccggaacaca gtgtccagat gggcagttct gccctgttgc 7051 ctgctgcctt gaccagggag gagccaacta cagctgctgt aaccctcttc 7101 tggacacatg gcctagaata acgagccatc atctagatgg ctcctgccag 7151 acccatggcc actgtcctgc tggctattct tgtcttctca ctgtgtctgg 7201 gacttccagc tgctgcccgt tctctaaggg tgtgtcttgt ggtgatggct 7251 accactgctg cccccagggc ttccactgta gtgcagatgg gaaatcctgc 7301 ttccagatgt cagataaccc cttgggtgct gtccagtgtc ctgggagcca 7351 gtttgaatgt cctgactctg ccacctgctg cattatggtt gatggttcgt 7401 ggggatgttg tcccatgccc caggcctctt gctgtgaaga cagagtgcat 7451 tgctgtcccc atggggcctc ctgtgacctg gttcacacac gatgcgtttc 7501 acccacgggc acccacaccc tactaaagaa gttccctgca caaaagacca 7551 acagggcagt gtctttgcct ttttctgtcg tgtgccctga tgctaagacc

159

7601 cagtgtcccg atgattctac ctgctgtgag ctacccactg ggaagtatgg 7651 ctgctgtcca atgcccaatg ccatctgctg ttccgaccac ctgcactgct 7701 gcccccagga cactgtatgt gacctgatcc agagtaagtg cctatccaag 7751 aactacacca cggatctcct gaccaagctg cctggatacc cagtgaagga 7801 ggtgaagtgc gacatggagg tgagctgccc tgaaggatat acctgctgcc 7851 gcctcaacac tggggcctgg ggctgctgtc catttgccaa ggccgtgtgt 7901 tgtgaggatc acattcattg ctgcccggca gggtttcagt gtcacacaga 7951 gaaaggaacc tgcgaaatgg gtatcctcca agtaccctgg atgaagaagg 8001 tcatagcccc cctccgcctg ccagacccac agatcttgaa gagtgataca 8051 ccttgtgatg acttcactag gtgtcctaca aacaatacct gctgcaaact 8101 caattctggg gactggggct gctgtcccat cccagaggct gtctgctgct 8151 cagacaacca gcattgctgc cctcagggct tcacatgtct ggctcagggg 8201 tactgtcaga agggagacac aatggtggct ggcctggaga agatacctgc 8251 ccgccagaca accccgctcc aaattggaga tatcggttgt gaccagcata 8301 ccagctgccc agtagggcaa acctgctgcc caagcctcaa gggaagttgg 8351 gcctgctgcc agctgcccca tgctgtgtgc tgtgaggacc ggcagcactg 8401 ttgcccggcc gggtacacct gcaatgtgaa ggcgaggacc tgtgagaagg 8451 atgtcgattt tatccagcct cccgtgctcc tgaccctcgg ccctaaggtt 8501 gggaatgtgg agtgtggaga agggcatttc tgccatgata accagacctg 8551 ttgtaaagac agtgcaggag tctgggcctg ctgtccctac ctaaagggtg 8601 tctgctgtag agatggacgt cactgttgcc ccggtggctt ccactgttca- 8651 gccaggggaa ccaagtgttt gcgaaagaag attcctcgct gggacatgtt 8701 tttgagggat ccggtcccaa gaccgctact gtag

D.2 GFP Lentiviral Construct and Gene Sequence

EmGFP

160

pLenti6.2/GW/EMGFP vector Sequence:

ATGTAGTCTTATGCAATACTCTTGTAGTCTTGCAACATGGTAACGATGAGTTAGCAACATGCCTTACAAGGAGAGAAAAAGCACCGTGCATGCCGATTGGTGGAAGTAAGGTGGTACGATCGTGCCTTATTAGGAAGGCAACAGACGGGTCTGACATGGATTGGACGAACCACTGAATTGCCGCATTGCAGAGATATTGTATTTAAGTGCCTAGCTCGATACATAAACGGGTCTCTCTGGTTAGACCAGATCTGAGCCTGGGAGCTCTCTGGCTAACTAGGGAACCCACTGCTTAAGCCTCAATAAAGCTTGCCTTGAGTGCTTCAAGTAGTGTGTGCCCGTCTGTTGTGTGACTCTGGTAACTAGAGATCCCTCAGACCCTTTTAGTCAGTGTGGAAAATCTCTAGCAGTGGCGCCCGAACAGGGACTTGAAAGCGAAAGGGAAACCAGAGGAGCTCTCTCGACGCAGGACTCGGCTTGCTGAAGCGCGCACGGCAAGAGGCGAGGGGCGGCGACTGGTGAGTACGCCAAAAATTTTGACTAGCGGAGGCTAGAAGGAGAGAGATGGGTGCGAGAGCGTCAGTATTAAGCGGGGGAGAATTAGATCGCGATGGGAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAAAAATATAAATTAAAACATATAGTATGGGCAAGCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTGTTAGAAACATCAGAAGGCTGTAGACAAATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAGTAGCAACCCTCTATTGTGTGCATCAAAGGATAGAGATAAAAGACACCAAGGAAGCTTTAGACAAGATAGAGGAAGAGCAAAACAAAAGTAAGACCACCGCACAGCAAGCGGCCGCTGATCTTCAGACCTGGAGGAGGAGATATGAGGGACAATTGGAGAAGTGAATTATATAAATATAAAGTAGTAAAAATTGAACCATTAGGAGTAGCACCCACCAAGGCAAAGAGAAGAGTGGTGCAGAGAGAAAAAAGAGCAGTGGGAATAGGAGCTTTGTTCCTTGGGTTCTTGGGAGCAGCAGGAAGCACTATGGGCGCAGCGTCAATGACGCTGACGGTACAGGCCAGACAATTATTGTCTGGTATAGTGCAGCAGCAGAACAATTTGCTGAGGGCTATTGAGGCGCAACAGCATCTGTTGCAACTCACATCTGGGGCATCAAGCAGCTCCAGGCAAGAATCCTGGCTGTGGAAAGATACCTAAAGGATCAACAGCTCCTGGGGATTTGGGGTTGCTCTGGAAAACTCATTTGCACCACTGCTGTGCCTTGGAATGCTAGTTGGAGTAATAAATCTCTGGAACAGATTTGGAATCACACGACCTGGATGGAGTGGGACAGAGAAATTAACAATTACACAAGCTTAATACACTCCTTAATTGAAGAATCGCAAAACCAGCAAGAAAAGAATGAACAAGAATTATTGGAATTAGATAAATGGGCAAGTTTGTGGAATTGGTTTAACATAACAAATTGGCTGTGGTATATAAAATTATTCATAATGATAGTAGGAGGCTTGGTAGGTTTAAGAATAGTTTTTGCTGTACTTTCTATAGTGAATAGAGTTAGGCAGGGATATTCACCATTATCGTTTCAGACCCACCTCCCAACCCCGAGGGGACCCGACAGGCCCGAAGGAATAGAAGAAGAAGGTGGAGAGAGAGACAGAGACAGATCCATTCGATTAGTGAACGGATCTCGACGGTATCGATAAGCTTGGGAGTTCCGCGTTACATAACTTACGGTAAATGGCCCGCCTGGCTGACCGCCCAACGACCCCCGCCCATTGACGTCAATAATGACGTATGTTCCCATAGTAACGCCAATAGGGACTTTCCATTGACGTCAATGGGTGGAGTATTTACGGTAAACTGCCCACTTGGCAGTACATCAAGTGTATCATATGCCAAGTACGCCCCCTATTGACGTCAATGACGGTAAATGGCCCGCCTGGCATTATGCCCAGTACATGACCTTATGGGACTTTCCTACTTGGCAGTACATCTACGTATTAGTCATCGCTATTACCATGGTGATGCGGTTTTGGCAGTACATCAATGGGCGTGGATAGCGGTTTGACTCACGGGGATTTCCAAGTCTCCACCCCATTGACGTC

161

AATGGGAGTTTGTTTTGGCACCAAAATCAACGGGACTTTCCAAAATGTCGTAACAACTCCGCCCCATTGACGCAAATGGGCGGTAGGCGTGTACGGTGGGAGGTCTATATAAGCAGAGCTCGTTTAGTGAACCGTCAGATCGCCTGGAGACGCCATCCACGCTGTTTTGACCTCCATAGAAGACACCGACTCTAGAGGATCCACTAGTCCAGTGTGGTGGAATTCTGCAGATATCAACAAGTTTGTACAAAAAAGCAGGCTCCACCATGGTGAGCAAGGGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCTGGTCGAGCTGGACGGCGACGTAAACGGCCACAAGTTCAGCGTGTCCGGCGAGGGCGAGGGCGATGCCACCTACGGCAAGCTGACCCTGAAGTTCATCTGCACCACCGGCAAGCTGCCCGTGCCCTGGCCCACCCTCGTGACCACCTTCACCTACGGCGTGCAGTGCTTCGCCCGCTACCCCGACCACATGAAGCAGCACGACTTCTTCAAGTCCGCCATGCCCGAAGGCTACGTCCAGGAGCGCACCATCTTCTTCAAGGACGACGGCAACTACAAGACCCGCGCCGAGGTGAAGTTCGAGGGCGACACCCTGGTGAACCGCATCGAGCTGAAGGGCATCGACTTCAAGGAGGACGGCAACATCCTGGGGCACAAGCTGGAGTACAACTACAACAGCCACAAGGTCTATATCACCGCCGACAAGCAGAAGAACGGCATCAAGGTGAACTTCAAGACCCGCCACAACATCGAGGACGGCAGCGTGCAGCTCGCCGACCACTACCAGCAGAACACCCCCATCGGCGACGGCCCCGTGCTGCTGCCCGACAACCACTACCTGAGCACCCAGTCCGCCCTGAGCAAAGACCCCAACGAGAAGCGCGATCACATGGTCCTGCTGGAGTTCGTGACCGCCGCCGGGATCACTCTCGGCATGGACGAGCTGTACAAGTAAACCCAGCTTTCTTGTACAAAGTGGTTGATATCCAGCACAGTGGCGGCCGCTCGAGTCTAGAGGGCCCGCGGTTCGAAGGTAAGCCTATCCCTAACCCTCTCCTCGGTCTCGATTCTACGCGTACCGGTTAGTAATGACGATGGTACCTACCGGGTAGGGGAGGCGCTTTTCCCAAGGCAGTCTGGAGCATGCGCTTTAGCAGCCCCGCTGGGCACTTGGCGCTACACAAGTGGCCTCTGGCCTCGCACACATTCCACATCCACCGGTAGGCGCCAACCGGCTCCGTTCTTTGGTGGCCCCTTCGCGCCACCTTCTACTCCTCCCCTAGTCAGGAAGTTCCCCCCCGCCCCGCAGCTCGCGTCGTGCAGGACGTGACAAATGGAAGTAGCACGTCTCACTAGTCTCGTGCAGATGGACAGCACCGCTGAGCAATGGAAGCDGGTAGGCCTTTGGGGCAGCGGCCAATAGCAGCTTTGCTCCTTCGCTTTCTGGGCTCAGAGGCTGGGAAGGGGTGGGTCCGGGGGCGGGCTCAGGGGCGGGCTCAGGGGCGGGGCGGGCGCCCGAAGTCCTCCGGAGGCCCGGCATTCTGCACGCTTCAAAAGCGCACGTCTGCCGCGCTGTTCTCCTCTTCCTCATCTCCGGGCCTTTCGACTCTAGACACGTGTTGACAATTAATCATCGGCATAGTATATCGGCATAGTATAATACGACAAGGTGAGGAACTAAACCATGGCCAAGCCTTTGTCTCAAGAAGAATCCACCCTCATTGAAAGAGCAACGGCTACAATCAACAGCATCCCCATCTCTGAAGACTACAGCGTCGCCAGCGCAGCTCTCTCTAGCGACGGCCGCATCTTCACTGGTGTCAATGTATATCATTTTACTGGGGGACCTTGTGCAGAACTCGTGGTGCTGGGCACTGCTGCTGCTGCGGCAGCTGGCAACCTGACTTGTATCGTCGCGATCGGAAATGAGAACAGGGGCATCTTGAGCCCCTGCGGACGGTGCCGACAGGTGCTTCTCGATCTGCATCCTGGGATCAAAGCCATAGTGAAGGACAGTGATGGACAGCCGACGGCAGTTGGGATTCGTGAATTGCTGCCCTCTGGTTATGTGTGGGAGGGCTAAGCACAATTCGAGCTCGGTACCTTTAAGACCAATGACTTACAAGGCAGCTGTAGATCTTAGCCACTTTTTAAAAGAAAAGGGGGGACTGGAAGGGCTAATTCACTCCCAACGAAGACAAGATCTGCTTTTTGCTTGTACTGGGTCTCTCTGGTTAGACCAGATCTGAGCCTGGGAGCTCTCTGGCTAACTAGGGAACCCACTGCTTAAGCCTCAATAAAGCTTGCCTTGAGTGCTTCAAGTAGTGTGTGCCCGTCTGTTGTGTGACTCTGGTAACTAGAGATCCCTCAGACCCTTTTAGTCAGTGTGGAAAATCTCTAGCAGTAGTAGTTCATGTCATCTTATTATTCAGTATTTATAACTTGCAAAGAAATGAATATCAGAGAGTGAGAGGAACTTGTTTATTGCAGCTTATAATGGTTACAAATAAAGCA

162

ATAGCATCACAAATTTCACAAATAAAGCATTTTTTTCACTGCATTCTAGTTGTGGTTTGTCCAAACTCATCAATGTATCTTATCATGTCTGGCTCTAGCTATCCCGCCCCTAACTCCGCCCATCCCGCCCCTAACTCCGCCCAGTTCCGCCCATTCTCCGCCCCATGGCTGACTAATTTTTTTTATTTATGCAGAGGCCGAGGCCGCCTCGGCCTCTGAGCTATTCCAGAAGTAGTGAGGAGGCTTTTTTGGAGGCCTAGGGACGTACCCAATTCGCCCTATAGTGAGTCGTATTACGCGCGCTCACTGGCCGTCGTTTTACAACGTCGTGACTGGGAAAACCCTGGCGTTACCCAACTTAATCGCCTTGCAGCACATCCCCCTTTCGCCAGCTGGCGTAATAGCGAAGAGGCCCGCACCGATCGCCCTTCCCAACAGTTGCGCAGCCTGAATGGCGAATGGGACGCGCCCTGTAGCGGCGCATTAAGCGCGGCGGGTGTGGTGGTTACGCGCAGCGTGACCGCTACACTTGCCAGCGCCCTAGCGCCCGCTCCTTTCGCTTTCTTCCCTTCCTTTCTCGCCACGTTCGCCGGCTTTCCCCGTCAAGCTCTAAATCGGGGGCTCCCTTTAGGGTTCCGATTTAGTGCTTTACGGCACCTCGACCCCAAAAAACTTGATTAGGGTGATGGTTCACGTAGTGGGCCATCGCCCTGATAGACGGTTTTTCGCCCTTTGACGTTGGAGTCCACGTTCTTTAATAGTGGACTCTTGTTCCAAACTGGAACAACACTCAACCCTATCTCGGTCTATTCTTTTGATTTATAAGGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAAAATTTAACGCGAATTTTAACAAAATATTAACGCTTACAATTTAGGTGGCACTTTTCGGGGAAATGTGCGCGGAACCCCTATTTGTTTATTTTTCTAAATACATTCAAATATGTATCCGCTCATGAGACAATAACCCTGATAAATGCTTCAATAATATTGAAAAAGGAAGAGTATGAGTATTCAACATTTCCGTGTCGCCCTTATTCCCTTTTTTGCGGCATTTTGCCTTCCTGTTTTTGCTCACCCAGAAACGCTGGTGAAAGTAAAAGATGCTGAAGATCAGTTGGGTGCACGAGTGGGTTACATCGAACTGGATCTCAACAGCGGTAAGATCCTTGAGAGTTTTCGCCCCGAAGAACGTTTTCCAATGATGAGCACTTTTAAAGTTCTGCTATGTGGCGCGGTATTATCCCGTATTGACGCCGGGCAAGAGCAACTCGGTCGCCGCATACACTATTCTCAGAATGACTTGGTTGAGTACTCACCAGTCACAGAAAAGCATCTTACGGATGGCATGACAGTAAGAGAATTATGCAGTGCTGCCATAACCATGAGTGATAACACTGCGGCCAACTTACTTCTGACAACGATCGGAGGACCGAAGGAGCTAACCGCTTTTTTGCACAACATGGGGGATCATGTAACTCGCCTTGATCGTTGGGAACCGGAGCTGAATGAAGCCATACCAAACGACGAGCGTGACACCACGATGCCTGTAGCAATGGCAACAACGTTGCGCAAACTATTAACTGGCGAACTACTTACTCTAGCTTCCCGGCAACAATTAATAGACTGGATGGAGGCGGATAAAGTTGCAGGACCACTTCTGCGCTCGGCCCTTCCGGCTGGCTGGTTTATTGCTGATAAATCTGGAGCCGGTGAGCGTGGGTCTCGCGGTATCATTGCAGCACTGGGGCCAGATGGTAAGCCCTCCCGTATCGTAGTTATCTACACGACGGGGAGTCAGGCAACTATGGATGAACGAAATAGACAGATCGCTGAGATAGGTGCCTCACTGATTAAGCATTGGTAACTGTCAGACCAAGTTTACTCATATATACTTTAGATTGATTTAAAACTTCATTTTTAATTTAAAAGGATCTAGGTGAAGATCCTTTTTGATAATCTCATGACCAAAATCCCTTAACGTGAGTTTTCGTTCCACTGAGCGTCAGACCCCGTAGAAAAGATCAAAGGATCTTCTTGAGATCCTTTTTTTCTGCGCGTAATCTGCTGCTTGCAAACAAAAAAACCACCGCTACCAGCGGTGGTTTGTTTGCCGGATCAAGAGCTACCAACTCTTTTTCCGAAGGTAACTGGCTTCAGCAGAGCGCAGATACCAAATACTGTTCTTCTAGTGTAGCCGTAGTTAGGCCACCACTTCAAGAACTCTGTAGCACCGCCTACATACCTCGCTCTGCTAATCCTGTTACCAGTGGCTGCTGCCAGTGGCGATAAGTCGTGTCTTACCGGGTTGGACTCAAGACGATAGTTACCGGATAAGGCGCAGCGGTCGGGCTGAACGGGGGGTTCGTGCACACAGCCCAGCTTGGAGCGAACGACCTACACCGAACTGAGATACCTACAGCGTGAGCTATGAGAAAGCGCCACGCTTCCCGAAGGGAGAAAGGCGGACAGGTATCCGGTAAGCGGCAGGGTCGGAACAGGAGAGCGCACGAGGGAGC

163

TTCCAGGGGGAAACGCCTGGTATCTTTATAGTCCTGTCGGGTTTCGCCACCTCTGACTTGAGCGTCGATTTTTGTGATGCTCGTCAGGGGGGCGGAGCCTATGGAAAAACGCCAGCAACGCGGCCTTTTTACGGTTCCTGGCCTTTTGCTGGCCTTTTGCTCACATGTTCTTTCCTGCGTTATCCCCTGATTCTGTGGATAACCGTATTACCGCCTTTGAGTGAGCTGATACCGCTCGCCGCAGCCGAACGACCGAGCGCAGCGAGTCAGTGAGCGAGGAAGCGGAAGAGCGCCCAATACGCAAACCGCCTCTCCCCGCGCGTTGGCCGATTCATTAATGCAGCTGGCACGACAGGTTTCCCGACTGGAAAGCGGGCAGTGAGCGCAACGCAATTAATGTGAGTTAGCTCACTCATTAGGCACCCCAGGCTTTACACTTTATGCTTCCGGCTCGTATGTTGTGTGGAATTGTGAGCGGATAACAATTTCACACAGGAAACAGCTATGACCATGATTACGCCAAGCGCGCAATTAACCCTCACTAAAGGGAACAAAAGCTGGAGCTGCAAGCTT

Appendix E Relative hSOD1 Copy Number in Male Breeders

Figure E.1 Cycle threshold values of original commercial male breeders utilized in seeding our transgenic G37R (line 29) colony as assayed by Jackson Laboratory

Cycle threshold values of original commercial male breeders utilized in seeding our transgenic G37R (line 29) colony as assayed by the Jackson laboratory. Results are plotted against CT values of historical control samples from G37R (line 29 and 42), G93A, and G85R animals. G37R (line 29) animals exhibit some of the lowest copy numbers of the hSOD1 transgene. Breeder males (o) that seeded our colony were found to be well within a 0.5 deviation in CT values in comparison with historical controls (•). As such, these breeders were deemed to retain comparable levels of the transgene as historical G37R (line 29) controls (outlined region); Mean +/- SD shown. Data provided by Transnetyx.

164

Appendix F Chapter 4 Supplementary Materials

F.1 Future Considerations for Modeling Disease and Testing Clinically-relevant ALS

Therapeutics 13

Due to the definitive need for therapeutics targeting the neurodegenerative cascade(s) inherent to

ALS pathogenesis, it is of crucial importance that non-productive research endeavours are

expediently identified and resources redirected to more promising avenues. To date, the

reductionist approach of understanding ALS as an mSOD1-mediated insult has not borne out an

effective treatment strategy. In contrast to the current approach of studying potential therapeutics

in the context of a specific genetic causality, a disease such as ALS with a complex etiology

necessitates a multi-pronged strategy. Herein I broadly consider future strategies that could

greatly facilitate the development and testing of potential ALS therapies.

F.2 Clinically-relevant Biomarkers: Assessing Disease Progression and Therapeutic

Efficacy

There is a desperate need for the development of clinically-relevant biomarkers that will not only

aid in an earlier diagnosis of disease, but also provide information with regard to disease

progression. If effective therapeutics can then be targeted to an earlier stage of the

neurodegenerative cascade, it is plausible to suspect that neuronal integrity can be maintained for

a prolonged period.

13 A version of Appendix F has previously been published as part of Zwiegers and Shaw, 2015.

165

To date, various potential peripheral blood biomarkers have been identified, but none have been

successfully translated into clinical use (Reviewed in Robelin & Gonzalez De Aguilar 2014).

Identification of such biomarkers would not only aid in diagnosing patients at an earlier stage of

the disease, but also provide for an alternate measure of the effectiveness of any putative

therapeutic agents.

Having a distinct correlative marker of disease progression might also shorten clinical trials since

the effectiveness of a drug could be more efficiently assessed (Robelin & Gonzalez De Aguilar

2014; Ryberg & Bowser 2008). With shorter clinical trials, patients could then participate in

more studies, thus allowing for more therapeutics to be clinically assessed and a treatment

regimen to be fast-tracked. In addition, predictive biomarkers would allow for an earlier

diagnosis of disease. This could facilitate an earlier intervention strategy in the progressive

neurodegenerative cascade and theoretically mitigate additional CNS damage.

Blood provides for an easily-obtained biological fluid which constitutes a viable source for

biomarker discovery, with the added caveat that blood biomarkers may not directly correlate

with motor neuron degeneration (Robelin & Gonzalez De Aguilar 2014). As Robelin and De

Aguilar discuss, both the blood-brain barrier and the blood-cerebrospinal fluid barrier may act as

an impediment to the crossing of relevant biomarkers into systemic circulation (2014). There is

thus a possibility that surrogate peripheral biomarkers may not adequately reflect the underlying

degenerative cascade, or provide a direct measure of the intended therapeutic effect within the

CNS. On account of the complex neurodegenerative cascade at play during ALS pathogenesis, it

166

may be more appropriate to identify panels of biomarkers that clearly distinguish ALS from

related disorders (Robelin & Gonzalez De Aguilar 2014).

A recently conducted clinical trial of rasagiline –an inhibitor of monoamine oxidase B–

demonstrates the application of potential blood biomarkers in assessing the clinical outcome of a

therapeutic agent (Macchi et al. 2015). Prior work had demonstrated that the drug had specific

beneficial effects on mitochondrial function, decreased oxidative damage, and inhibited

apoptosis. As previously pointed out, mitochondrial dysfunction is a clinical hallmark of ALS

pathogenesis (Section 1.3.1). Thus, measuring particular biomarkers related to mitochondrial

function (e.g. changes in mitochondrial membrane potential, oxidative stress, and the relative

abundance of pro-survival/pro-apoptotic signals) may allow for a biochemically-relevant

measure of therapeutic effectiveness in ALS clinical trials (Macchi et al. 2015). The trial did not

demonstrate an improvement in the functional ALSFRS-R scores after 12 months compared to

historical controls (when corrected for symptom duration). However, peripheral lymphocytes

showed evidence of an increased mitochondrial membrane potential, decreased oxidative stress,

and an increased Bcl-2:Bax ratio; indicative of pro-survival cell signaling. A major caveat to the

use of peripheral biomarkers as a surrogate indicator for activity within the central nervous

system is the degree to which the therapeutic agent is able to penetrate, and act on cells within

the CNS. In the case of rasagiline, the use of a peripheral mitochondrial biomarker is expected to

be an adequate indicator of CNS activity since the therapeutic agent has demonstrated peripheral

distribution and CNS penetration (Macchi et al. 2015).

167

F.3 Development of Alternate Models to More Effectively Mirror the Diverse Etiological

Factors that Underlie ALS Pathogenesis

A critical problem lies in developing animal models which are more representative of ALS

patients. To date, primarily on account of the unknown etiology, there is a lack of models that are

representative of the sporadic form of the disease. Our group has previously investigated the

dietary administration of cycad flour or extracted sterol glucosides as causative agents in the

purported disease cascade (Tabata et al. 2008; Schulz et al. 2003; Wilson et al. 2002), while

others have assessed low Calcium/Magnesium, high Aluminum diets (Kihira et al. 2004).

Putative ALS therapeutics have yet to be formally investigated within the context of these

environmental model systems. Furthermore, there needs to be a renewed effort to develop and

validate additional animal models with suspected environmental etiological factors.

On account of our increased understanding of the underlying genetic causal factors at play in

ALS (Figure 1.1), there have been multiple additional murine models developed since the advent

of the mutant SOD1 mouse. These include mice expressing mutant TDP-43 or the C9ORF72-

associated hexanucleotide repeat expansions (Wegorzewska et al. 2009; Hukema et al. 2014;

Chew et al. 2015). Characterizing these additional systems allow for a more representative model

of the overall ALS population, as it takes into account the multiple underlying pathological

changes that are implicated in disease pathogenesis

However, a central caveat to conducting pre-clinical studies in animal models is the inability to

rapidly screen multiple therapeutic agents in a single assay. To address this, an approach

involves modeling the disease with nerve cells derived from induced pluripotent stem cells from

168

sALS and/or fALS patients (Burkhardt et al. 2013; Richard & Maragakis 2014). Not only could

this strategy allow for the screening of a plethora of neuroprotective compounds, but it may pave

the way for determining patient-specific drug responses to ALS therapeutics.

An alternate approach employs the use of small invertebrate and vertebrate model systems

expressing mutant forms of loci associated with ALS as platforms for pre-clinical drug

development. Zebrafish and Caenorhabditis elegans animal models that express mutant forms

of human TDP-43 or FUS allow for the production of large numbers of animals that can be

housed in multi-well plates, tested with an array of potential therapeutic compounds, and

assessed whether or not the disease phenotype is attenuated (Vaccaro et al. 2012). Strategies such

as these may allow for the rapid identification of potential therapeutic agents that can ultimately

be applied at the clinical level.

Realistically, due to the heterogeneity of the pathological mechanism(s) at play, it is quite

unlikely that any one model would be sufficient in mimicking all of the relevant underlying

pathobiology. What should be adopted instead is a multi-pronged approach where a potential

therapeutic is validated in multiple model systems testing specific outcomes prior to clinical

translation (Figure F.1). Adopting this strategy would enhance the likelihood of success since a

treatment with positive outcomes in multiple model systems will be of greater relevance to the

heterogeneous patient population.

169

Figure F.1 A ‘model-up’ approach in designing clinically effective ALS therapeutics. 14

A potentially effective strategy to identify putative therapeutic options is to screen multiple compounds in either small invertebrate/vertebrate ALS models systems, or alternatively in patient stem cell-derived neuronal cells. Therapeutic agents demonstrating appreciable neuroprotective properties in either case are further validated in several ALS murine models. Therapeutics demonstrating the most promising outcomes in multiple murine models are then further validated in the ALS patient population.

F.4 Clinical Trials Should be Designed and Stratified so that the Therapeutic Effect of

Any Agent is Tested in a Patient Subset that is Homologous for Set Criteria

Clinical ALS trials should be conducted in a manner that reflects the pathobiological

heterogeneity underlying the neurodegenerative cascade. As discussed elsewhere, current

evidence suggests that the disease presents along a continuum which ranges from “pure” ALS to

Elements for Figure F.1 provided by Servier medical art and modified under a Creative Commons Attribution 3.0 Unported License.

170

frontotemporal dementia (Turner et al. 2013; Robberecht & Philips 2013). Employing a

heterogeneous study population in a clinical trial with undefined disease mechanisms will not see

positive outcomes unless all disease pathways converge for the ultimate phenotypic expression.

It is possible that this may not be the case in ALS. That said, stratifying patient cohorts based on

both the underlying causal mechanism(s) and degree of disease progression (i.e. due to validated

biomarkers), would allow clinical researchers to develop targeted strategies that address definite

causal mechanisms within a specific context. As our understanding of the complex etiological

factors in ALS evolves over the coming years, we can investigate multiple therapeutic

compounds acting on diverse disease mechanisms and perhaps more effectively intervene in a

large proportion of the affected populace.