sign in sign up
<<
Home , Prion

Investigating pathogenic mechanisms associated with prion and α-synuclein misfolding

Cathryn Laura Ugalde BSc(Hons)

Submitted in total fulfilment for the requirements of the degree of:

Doctor of Philosophy August 2017

Department of Biochemistry and , University of Melbourne Department of , University of Melbourne The Florey Institute of Neuroscience and Mental Health, University of Melbourne Department of Biochemistry and Genetics, La Trobe University

ABSTRACT

Neuronal loss and the aggregation of misfolded prion protein (PrPSc) and α-synuclein (αsyn) in the are hallmarks of prion and synucleinopathy disorders (such as Parkinson’s , , with Lewy body), respectively. PrPSc is unusual because it is the major component of prions; ‘proteinacious infectious particles’ that are the causative agent in prion disease. Many features of prion disease have been directly attributed to PrPSc including its ability to propagate, causing disease in susceptible and be transmissible. Another feature of PrPSc is strain variability; where the conformation of the misfolded species ascribes the clinical profile of disease that develops. Neurotoxicity is also intimately associated with the generation of PrPSc, however, the precise mechanisms underlying its development in prion disease are not resolved. Mounting evidence suggests that αsyn shares similar pathogenic mechanisms to PrP in terms of its ability to misfold, propagate and cause disease in susceptible animals. Like PrP, the mechanisms associated with the toxicity of misfolded αsyn are not well defined. This thesis studied the pathogenic mechanisms of PrP and αsyn misfolding with particular interest in how these cause neurotoxicity through the development of and ex vivo systems that model pathogenic aspects of their respective disorders. A potential neurotoxic mechanism of misfolded αsyn was found using the protein misfolding cyclic amplification (PMCA) assay to produce a heterogeneous population of misfolded species and their pathogenicity examined in cultured neuronal cells. Misfolded αsyn was found to bind to the lipid cardiolipin that is virtually exclusively expressed in mitochondria. Extensive analysis revealed misfolded αsyn causes hyperactive respiration in the mitochondria of live cells without causing any functional deficit. This robust data gives strong support for the as a target for misfolded αsyn and reveals a potential up-stream pathogenic pathway of neuronal loss in Parkinson’s disease. Next, a model of prion disease was established using organotypic slice cultures. Upon with a mouse-adapted human prion strain, brain slices were shown to propagate resistant PrP and develop neurotoxicity. The relevance of protein misfolding in these systems was assessed using the small molecule Anle138b; which has been previously reported to alter misfolding of PrP and αsyn, and be efficacious to prion infected animals and models of PD. This molecule was found to modulate the misfolding of both proteins, which in the case of PrP, led to a rescue of prion-induced

i neuronal loss in organotypic brain slice cultures. Collectively this work robustly shows misfolded αsyn and PrP can cause substantial alterations to they are exposed to. Findings from this work contribute to our general understanding on these proteins in disease and highlight both similarities and differences in their misfolding and pathogenicity.

ii

DECLARATION

This is to certify that;

i) The thesis comprises only my work towards the PhD except where indicated in the Preface ii) Due acknowledgement has been made in the text to all other material used iii) The thesis is less than 100,000 works in length, exclusive of tables, bibliographies and appendices

Cathryn Laura Ugalde

Department of Biochemistry and Molecular Biology, University of Melbourne

Department of Pathology, University of Melbourne

The Florey Institute of Neuroscience and Mental Health, University of Melbourne

Department of Biochemistry and Genetics, La Trobe University

iii

PREFACE

I acknowledge the work presented in this thesis carried out by others:

 Andrew Hill (Department of Biochemistry and Genetics, La Trobe University), David Finkelstein (The Florey Institute of Neuroscience and Mental Health, University of Melbourne) and Victoria Lawson (Department of Pathology, University of Melbourne) are co-authors of the peer-reviewed publication ‘Pathogenic mechanisms of prion protein, -β and α-synuclein misfolding: the prion concept and neurotoxicity of protein oligomers’ and the published book chapter ‘Neurotoxicity in prion disease: relationships with other neurodegenerative and outlooks for therapeutic intervention’ (details of these works are described in ‘Published works arising from thesis’, page vi). They provided critical feedback during the writing of the manuscripts. Exerts from both these written works are presented in this thesis.

 Paul Fisher, Sarah Annesley and Katelyn Mroczek (Department of Physiology, Anatomy and , La Trobe University) assisted in the cellular respiratory control and glycolysis experiments using the Seahorse XF Analyser. The instrument belongs to Paul Fisher. Sarah and Katelyn prepared the XF media and drugs used in these experiments and operated the instrument.

 Victoria Lawson and Laura Ellett (Department of Pathology, University of Melbourne) supplied the M1000-infected and normal brain .

 Roberto Cappai (Department of Pathology, The University of Melbourne) provided the wild-type αsyn constructs αsyn:pDNA3+ and αsyn:pRSET B and the mutant αsyn construct A53T:pRSET B.

 Irene Volitakis (Florey Institute for Neuroscience and Mental Health, University of Melbourne) processed the samples analysed by inductively coupled plasma-mass spectrometry shown in Table 3.1.

iv

 Peter Lock (Department of Biochemistry and Genetics, La Trobe University) and Benjamin Scicluna (Department of Biochemistry and Genetics, La Trobe University) took the transmission electron microscopy images shown in Figure 3.2C and Figure 3.4C.

v

ACKNOWLEDGEMENTS

The acknowledgements are one of the most important sections of a thesis. When it comes to writing them, it’s a simple task to mentally list those who have been influential to the completion of this thesis. It is much harder, however, to convey this in a way that you feel is appropriate. I hope this section reflects correctly the gratitude I have for those mentioned herein who have helped me in some way during my PhD journey.

First and foremost I thank my supervisors; Andy Hill, Vicki Lawson and David Finkelstein. I don’t think I could’ve had a better team of supervisors. You all contributed immeasurably to the development of my PhD project and have been important mentors to me during my time as a PhD student. Thank you for all the support over the years. You all challenged me, backed me to drive my projects and, in doing so, taught me valuable lessons about being an assertive researcher. Thank you also for the candid discussions on my data and their implications; our conversations – boarderline debates were some of the most enjoyable moments of my PhD.

Thank you to the members of my committee who have followed me along the journey; my committee chairs Heung-Chin Cheung and Dominic Ng and committee members Kelly Rogers and Joe Ciccotosto. Your helpful discussions over the years were appreciated.

To the current and past lab members of the Hill lab who I’ve worked with over the years; Mitch, Ting, Cam, Lesley, Laura, Robyn, Ben, Jacob, Amir, Mark; thank you for the friendships, support and for always having something to talk about other than research! Particular thanks to Cam and Ting, who as fellow PhD students have been great supports over the years, and Mitch, who helped my project in so many little ways over the last few years.

I would also like to thank Paul Fisher, Sarah Annesley and Katelyn Mroczek for their dedication to the collaboration on the experiments using the Seahorse XF Analyser. As an expert mitochondria lab, their time, teachings and resources were instrumental to the collection of some of the data in this thesis. In particular, thank you to Sarah who dedicated a lot of time early on in the collaboration when we weren’t sure of its outcome.

I have worked with many amazing researchers who taught me invaluable lessons as a researcher and were so supportive over the years. I have a huge amount of gratitude for my fellow-prion researcher Laura Ellett. Thank you for all of the fun and laughs (through the HIs and LOs!). More importantly, thank you for the scientific discussions, having a positive

vi attitude to virtually everything and all the tasty chai lattes! Thank you also to my friends who I worked with at Monash University and Department of Veterinary and Agricultual Sciences (University of Melbourne) before I started my PhD. I was lucky to be trained, work with and be friends with some of the best researchers I’ve ever met. Thank you to Kim, Kelly, Ben, Arthi, Donna, Laura and Pamu. You were all instrumental to my PhD experience.

I was financially supported by the CJD Support Group Network (CJDSGN) through various scholarships. This organisation supports families affected by prion disease and those at risk of developing the disease. They also aim to educate people about these disorders and raise money for prion research. A special thank you to the family and friends of Carol Willesee, Silva Coelho, David Matthews and Frank Burton, who contributed financially to awards I recieved. The CJDSGN wouldn’t be what it is today without Suzanne Solvyns and David Ralston who are tireless advocates for spreading awareness about the prion disorders. Their commitment and dedication is second to none and the support they give families is nothing short of extraordinary. I am honoured to consider myself a part of the CJDSGN family; I thank Suzanne and David for supporting me, inviting me to be a part of your wonderful annual conference and allowing me to represent young prion researchers in Australia.

My last acknowledgment is for my family. Mum, Dad, Anna, Sarah, Bayden and Arlo. It goes without saying you have all been my biggest support during my PhD and I appreciate everything you’ve done for me over the years. Thank you for the patience, love and always being so enthusiastic about my PhD project and journey.

vii

PUBLISHED WORKS ARISING FROM THIS THESIS

Ugalde CL, Lawson VA, Finkelstein DI, Hill AF; (2015) Neurotoxicity in prion disease: relationships with other neurodegenerative diseases and outlooks for therapeutic intervention. Giachin G, Legame G, editors. In: The prion phenomena in neurodegenerative disease: new frontiers in neuroscience. U.S.A: Nova Science Publishers

Ugalde CL, Finkelstein DI, Lawson VA, Hill AF; (2016) Pathogenic mechanisms of prion protein, amyloid-β and α-synuclein misfolding: the prion concept and neurotoxicity of protein oligomers, Journal of Neurochemistry, 139(2):162-80

viii

OTHER PUBLICATIONS ARISING DURING PHD CANDIDATURE

Jamasbi E, Ciccotosto GD, Tailhades J, Robins-Browne RM, Ugalde CL, Sharples RA, Patil N, Wade JD, Hossain MA, Separovic F; (2015) Site of fluorescent label modifies interaction of melittin with live cells and model membranes, Biochimica et Biophysica Acta, 1848(10PtA):2031-9

Van der Velden J, Harkness LM, Barker DM, Barcham GJ, Ugalde CL, Koumoundouros E, Bao H, Organ LA, Tokanovic A, Burgess JK, Snibson KJ; (2016) The effects of Tumstatin on vascularity, airway and lung function in an experimental model of chronic asthma, Scientific Reports, 6:26309

ix

TABLE OF CONTENTS

ABSTRACT ...... i

DECLARATION ...... iii

PREFACE ...... iv

ACKNOWLEDGEMENTS ...... vi

PUBLISHED WORKS ARISING FROM THIS THESIS ...... viii

OTHER PUBLICATIONS ARISING DURING PHD CANDIDATURE ...... ix

TABLE OF CONTENTS ...... x

ABBREVIATIONS ...... xiv

LIST OF FIGURES ...... xxii

LIST OF TABLES ...... xxiv

CHAPTER 1: INTRODUCTION ...... 1 General Background on PrPC, amyloid-β (Aβ) and α-synuclein (αsyn) ...... 4 PrPC ...... 4 Aβ ...... 5 αsyn ...... 6 PrPSc: the Conventional Prion ...... 9 Prion-like Properties of Misfolded Aβ...... 13 Prion-like Properties of Misfolded αsyn...... 16 Should We Expand the Prion Concept? ...... 23 Mechanisms of Neurotoxicity in Proteinopathies ...... 24 Toxic Oligomers...... 25 Aβ oligomers ...... 25 αsyn oligomers ...... 27 PrP oligomers ...... 28 Other Potential Neurotoxic Agents in Prion Disease ...... 29 The generation of a candidate toxic PrP Species; PrPL ...... 30 Altered PrPC signalling ...... 33 Alterations to function ...... 36 Mislocalisation of PrP ...... 37 The unfolded protein response (UPR) ...... 38 Summary & Aims ...... 43

CHAPTER 2: MATERIALS & METHODS ...... 45 Approvals and Animal Ethics ...... 45 Molecular Biology Procedures ...... 45 Transformation of αsyn:pcDNA3+ into competent cells ...... 45

x

Amplification and purification of DNA midipreps ...... 46 Methods using Recombinant αsyn ...... 47 Lyophilisation and preparation of recombinant protein for αsyn-PMCA ...... 47 αsyn PMCA ...... 47 Measuring kinetics of αsyn fibrillization using ThT fluorescence ...... 48 PK digestion of PMCA-generated αsyn ...... 48 Circular dichroism spectroscopy ...... 49 Lipid binding assays ...... 49 Preparation of Compounds Added to αsyn PMCA ...... 49 Anle138b ...... 49 Cardiolipin ...... 49 Procedures ...... 50 Maintenance of adherent cell lines ...... 50 Passaging and seeding of cells ...... 50 Freezing and thawing cells ...... 50 Transfection of SH-SY5Y cells to produce stable cell lines ...... 51 Detection of αsyn in transfected SH-SY5Y cells using immunofluorescence ...... 51 Lactate dehydrogenase (LDH) cytotoxicity assay ...... 52 Preparation of samples for measuring cellular readouts using Seahorse XF Analyser .... 53 Measuring cellular respiratory control (CRC) in SH-SY5Y cells ...... 53 Measuring glycolysis in SH-SY5Y cells ...... 53 Lysis of cultured cells ...... 54 Organotypic Culturing of tga20 Brain Slices ...... 54 The use of animals in this study ...... 54 Preparation of buffers for brain slicing ...... 54 Cerebellum isolation and sectioning using a tissue chopper ...... 55 Organotypic culturing of brain slices ...... 56 Measuring viability in live organotypic brain slices ...... 56 Collection and lysis of organotypic brain slices ...... 57 Protein Biochemistry...... 57 Protein Quantification...... 57 Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) ...... 58 Coomassie brilliant blue staining on gel electrophoresis ...... 58 SDS-PAGE western immunoblot analysis using PVDF membrane ...... 58 Immunodetection ...... 59 Enhanced chemiluminescence (ECL) ...... 59 List of and fluorescent dyes used in this thesis (Table 2.1) ...... 60 Prion Infection Experiments ...... 61 Safety procedures for handling prions ...... 61 Preparation of infectious M1000 brain homogenates ...... 61

xi

Infection of tga20 brain slices with M1000 ...... 61 PrP PMCA ...... 62 Amplification of PrPSc in singular M1000-infected brain slices using PrP PMCA ...... 62 Treatment of M1000-infected organotypic brain slices with Anle138b ...... 62 PK digestion of M1000-infected brain slices or PrP PMCA reaction products ...... 62 Data Analysis ...... 63

CHAPTER 3: IDENTIFYING POTENTIAL NEUROTOXIC MECHANISMS OF MISFOLDED αSYN 64 Introduction ...... 64 The association of αsyn with lipids ...... 64 Misfolded αsyn in experimental use ...... 67 Technical aspects of the PrP- and αsyn-PMCA assays ...... 68 Aims ...... 70 Results ...... 70 PMCA can induce misfolding of PrPC into PrPSc ...... 70 PMCA can produce misfolded αsyn species in prion conversion buffer (PCB) ...... 71 Fibrillization of αsyn by PMCA can occur using non-toxic buffers ...... 75 Characterisation of PMCA-induced misfolding of αsyn dissolved in PBS+150 mM NaCl (PBSN)...... 77 PBSN buffer is non-toxic to untransfected neuronal SH-SY5Y and αsyn-overexpressing SH-SY5Y cells ...... 81 Assessment on the affinity of misfolded αsyn to bind lipid species ...... 82 Development of assays to characterise αsyn misfolding using PMCA ...... 85 Cardiolipin accelerates PMCA-induced αsyn misfolding ...... 87 The effect of PMCA-generated misfolded αsyn on cellular respiratory control (CRC) in SH-SY5Y cells ...... 88 PMCA-generated misfolded αsyn do not alter glycolysis in SH-SY5Y cells ...... 97 Discussion ...... 100 Conclusion ...... 107

CHAPTER 4: THE DEVELOPMENT OF A MODEL OF PRION DISEASE USING ORGANOTYPIC BRAIN SLICE CULTURES ...... 108 Introduction ...... 108 Common tools to study prion disease ...... 108 The importance of using models that develop propagation and toxicity to study prion disease ...... 110 Organotypic brain slice cultures (OBSCs) ...... 112 Aims ...... 113 Results ...... 114 Organotypic culturing of tga20 cerebellar slices ...... 114 Cerebellar OBSCs from tga20 mice propagate PrPRes upon prion infection ...... 120 PrPRes can be identified in single OBSC exposed to M1000 prions ...... 123

xii

M1000-infected OBSCs develop neurotoxicity ...... 125 Discussion ...... 128 Conclusion ...... 133

CHAPTER 5: DETERMING THE RELEVANCE OF PROTEIN MISFOLDING TO THE PATHOGENICITY OF αSYN AND PRP USING ANLE138B ...... 134 Introduction ...... 134 Identification of molecules that are effective for therapeutic intervention in prion disease ...... 134 Anti-prion molecules that show efficacy towards other NDP-associated proteins ...... 138 Aims ...... 140 Results ...... 140 Anle138b reduces PrPRes in M1000-infected OBSCs...... 140 The expression of total PrP and GFAP are altered by Anle138b in M1000-infected OBSCs ...... 144 Anle138b protects M1000-infected OBSCs from neuronal vulnerability ...... 146 Anle138b modulates PMCA-induced αsyn fibrillization ...... 148 No further modulation of mitochondrial respiration is seen in PMCA-generated misfolded αsyn species formed in the presence of Anle138b ...... 153 Discussion ...... 156 Conclusion ...... 163

CHAPTER 6: GENERAL DISCUSSION ...... 164 Overview of main findings ...... 164 The similarities and differences in how PrP and αsyn misfold ...... 167 The toxic mechanisms of misfolded neurodegenerative proteinopathy-associated proteins: direct mechanisms, via an interaction with normal monomeric protein or both? ...... 168 Toxic mechanisms of αsyn in synucleinopathies ...... 170 Towards effective therapeutic strategies for NDPs ...... 173 Protein-directed antibodies ...... 175 Combined treatment strategies ...... 176 Conclusion ...... 177

REFERENCES ...... 178

xiii

ABBREVIATIONS

% Percent αsyn α-synuclein oC Degrees Celsius μg Microgram μM Micromolar μm Micrometre μL Microliter AU Arbitrary units Aβ Amyloid-β AD Alzheimer’s disease ADP Adenosine diphosphate AICD Amyloid precursor protein intracellular ALS Amyotrophic lateral sclerosis ApoE Apolipoprotein E APP Amyloid precursor protein ATF4 Activating factor 4 ATF6 Activating transcription factor 6 ATP Adenosine triphosphate AUC Analytical ultracentrifugation BACE1 β-site amyloid precursor protein cleaving 1 BBB Blood brain barrier BCA Bicinchoninic acid BCII Class 2 biosafety cabinet BME Basal medium essential BSA Bovine serum albumin BSE Bovine spongiform C Cholesterol CA Cardiolipin

xiv

CAA Cerebral amyloid angiopathy Cas CRISPR-associated system CD Circular dichroism CJD Creutzfeldt-Jakob disease cm Centimeter CNS Central nervous system Co Cobalt Con Control (brain tissue) Cpd-B Compound B CRC Cellular respiratory control CRISPR Clustered regularly interspaced short palindromic repeats CSF Cerebral spinal fluid ctmPrP Transmembrane PrP with the C-terminus facing into the cytoplasm Cu CWD Cyt c Cytochrome c CytPrP Cytosolic PrP DAPI 4',6-Diamidino-2-Phenylindole, Dihydrochloride ddH2O Double distilled H2O DG Diacylglycerol dH2O Distilled H2O DHA Docosahexaenoic acid DIC Differential interference contrast DLB Dementia with Lewy body DMEM Dulbeco’s modified eagle medium DMSO Dimethyl sulfoxide DNA Deoxyribonucleic acid dPBS Dulbecco’s PBS (calcium and magnesium free PBS) Dpl Doppel DPS Days post sectioning

xv

DY Hamster prion strain drowsy e- Electron ECAR Extracellular acidification rate ECL Enhanced chemiluminescence eIF2α The α subunit of eukaryotic factor-2 ER Endoplasmic reticulum ERAD ER associated degradation ETC Electron transport chain EtOH Ethanol fCJD Familial CJD FCS Fetal calf serum Fe Iron FFI Fatal familial FIJI ImageJ – software for image analysis FTD Fronto-temporal dementia g Gravitational force g Gram GAPDH Glyceraldehyde 3-phosphate dehydrogenase GBSS Grey’s balanced salt solution GBSSK Grey’s balanced salt solution supplemented with kynurenic acid GCI Glial cytoplasmic inclusion GFAP Glial fibrillary acidic protein GPCR G-protein coupled receptor GPI Glycosylphosphatidylinositol GM Monosialotetrahexosylganglioside GSS Gerstmann-Sträussler-Scheinker syndrome h Hour H Hoeschst 33342 H+ Proton HIV Human immunodeficiency

xvi

HRP Horse radish peroxidase HY Hamster prion strain Hyper i.p Intraperitoneal injection iAPP Islet amyloid polypeptide iCJD Iatrongenic CJD ICP-MS Inductively coupled plasma-mass spectrometry IMM Inner-mitochondrial membrane IMS Intermembrane space INF-BH Infected brain homogenate IRE1 Inositol-requiring enzyme 1 KA Kynurenic acid kDa Kilo dalton KO Knockout LB Lewy body LCP Luminescent conjugated polymers LDH Lactate dehydrogenase LN Lewy neurite LTP Long term potentiation LTD Long term depression LuB Luria Broth M Molar M1000 Mouse-adapted familial prion strain M1000 MAM Mitochondrial-associated ER membrane MEM Minimal essential medium MeOH Methanol min Minute MES 2-(N-morpholino)ethanesulfonic acid MeOH Methanol mg Milligram mGlu5 metabotrophic glutamate-5

xvii

MGRN1 Mahogunin miRNA Micro RNA mL Millilitre mm Millimeter mM Millimolar MM Mitochondrial Matrix Mn Manganese MPTP N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine MSA Multiple system atrophy MSDS Material safety data sheet Na Sodium NAC Non-amyloid component NaCl Sodium chloride NBH Normal brain homogenate NDP Neurodegenerative proteinopathy nm Nanometer NMDA N-Methyl D-Aspartate NSF N-ethylmaleimide-sensitive factor ntmPrP Transmembrane PrP with the N-terminus facing into the cytoplasm

O2 Molecular oxygen O/N Overnight OBSCs Organotypic brain slice cultures OCR Oxygen consumption rate OMM Outer mitochondrial membrane P Phosphorous PA Phosphatidic acid PBS Phosphate buffered saline PBSN PBS + 150 mM NaCl PBS-T PBS containing 0.1% Tween 20 PC Phosphatidylcholine

xviii

PC2 Physical containment level 2 PCB Prion conversion buffer PCR Polymerase chain reaction PD Parkinson’s disease PDD PD with dementia PE PERK Protein kinase RNA-like endoplasmic reticulum PG Phosphatidylglycerol Pi Inorganic phosphate PI Propidium iodide

PIP PtdIns(4)P3

PIP2 PtdIns(4,5)P2

PIP3 PtdIns(3,4,5)P3 PK PPS Pentosan polysulfate PMCA Protein misfolding cyclic amplification PMF Proton motive force PMSF Phenylmethylsulfonyl fluoride pQC Pre-emptive quality control PRNP Human prion protein PrP Prion protein (both PrPC and PrPSc) PrPC Normal cellular PrP PrPSc Disease-associated PrP PrPRes PK resistant PrP PS Phosphatidylserine PSD-95 Post synaptic density 95 protein PUFA Polyunsaturated fatty acids PVDF Polyvinylidene fluoride Q Quinone RFU Relative fluorescent units

xix

ROS Reactive oxygen species RNA Ribonucleic acid RT Room temperature sec Second SEM Standard error of the mean shRNA Short hairpin RNA siRNA Small interfering RNA SCEPA cell in the end point format assay sCJD Sporadic CJD SCM Slice culture media (for organotypic brain slices) SDS Sodium dodecyl sulphate SDS-PAGE SDS-polyacrylamide gel electrophoresis SF Sulfatide SH-SY5Y Human neuroblastoma cell line SIFT Scanning for intensity fluorescent targets SM Sphingomyelin SN Substantia nigra SNAP25 25 kDa synaptosomal associated protein SNARE Soluble NSF attachment receptors SNCA Human alpha-synuclein gene SOD1 1 SUVs Small unilamellar vesicles ROS Reactive oxygen species t-SNARE Target SNARE TBS Tris-buffered saline TBSN TBS + 150 mM NaCl TCA Tricarboxylic acid TEM Transmission electron microscopy TG Triglyceride ThT Thioflavin T

xx

TOM40 Translocase of outer membrane 40 TSE Transmissible spongiform encephalopathy Untreat Untreated UPR Unfolded protein response V Volts v-SNARE Vesicle SNARE v/v Volume to volume VAMP2 Vesicle associated membrane protein 2 vCJD Variant CJD vs. Versus w/v Weight to volume XBP1 X-box binding protein 1 XF Extracellular flux Zn Zinc

xxi

LIST OF FIGURES

Figure 1.1 Nucleation-dependent model of prion replication Figure 1.2. Disease-associated accumulation of misfolded protein in the central nervous system Figure 1.3: Human PrPC, αsyn and Aβ and the generation of the Aβ peptide by the proteolytic processing of APP Figure 1.4: Propagation of misfolded protein in rodent models following intracerebral inoculation of a misfolded seed Figure 1.5: The hypothetical generation of PrPL Figure 1.6: The flexible tail of PrP modulates the toxicity of PrP-directed antibodies in brain slices Figure 1.7: The unfolded protein response (UPR) Figure 3.1: Protein misfolding cyclic amplification (PMCA) of PrP Figure 3.2: αsyn PMCA Figure 3.3: PMCA-induced αsyn fibrillization can occur under non-toxic conditions Figure 3.4: Production of misfolded αsyn species in PBSN buffer produces comparable misfolded content to αsyn species generated in prion conversion buffer Figure 3.5: PBSN is non-toxic to untransfected SH-SY5Y and αsyn-overexpressing SH-SY5Y cells after 24 h Figure 3.6: PMCA-generated αsyn selectively binds to cardiolipin Figure 3.7: αsyn PMCA-induced rises in ThT are time-dependent allowing kinetic studies on its fibrillization and 72 h PMCA-products adopt partial PK resistance Figure 3.8: The effect of cardiolipin on αsyn fibrillization. Figure 3.9: Oxidative phosphorylation and measuring cellular respiratory control (CRC) via changes in oxygen consumption rate (OCR) Figure 3.10: PMCA-generated misfolded αsyn causes mitochondria to become hyperactive in SH-SY5Y cells Figure 3.11: PMCA-generated misfolded αsyn do not cause functional deficits in mitochondria of SH-SY5Y cells Figure 3.12: Glycolysis and and measuring glycolytic potential via changes in extracellular acidification rate (ECAR). Figure 3.13: PMCA-generated misfolded αsyn do not affect glycolysis in SH-SY5Y cells

xxii

Figure 4.1: Workflow of preparation and maintenance of organotypic brain slice cultures (OBSCs) Figure 4.2: Organotypic tga20 cerebellar brain slices are viable in culture Figure 4.3: Cell populations are dynamic in tga20 OBSCs Figure 4.4: PrPRes can be detected in viable OBSCs infected with M1000 prions Figure 4.5: Representative images used for statistical analysis shown in Figure 4.4B Figure 4.6: Singular M1000-infected brain slices amplify PrPRes using PMCA Figure 4.7: M1000-infected OBSCs develop neurotoxicity Figure 5.1. Structure of several compounds that possess prion-alleviating properties Figure 5.2: Anle138b reduces PrPRes in viable M1000-infected OBSCs Figure 5.3: Representative images used for statistical analysis shown in Figure 5.2A Figure 5.4: The effect of Anle138b on the expression of PrP and GFAP in viable M1000- infected OBSCs Figure 5.5: Anle138b alleviates M1000-induced neuronal vulnerability in OBSCs Figure 5.6: Anle138b increases PMCA-generated PK-resistant αsyn that do not propagate their altered phenotype through serial rounds of PMCA Figure 5.7: The ability of Anle138b to modulate PMCA-generated αsyn species is dependent on protein misfolding Figure 5.8: The addition of Anle138b in PMCA-induced αsyn misfolding does not alter the changes to mitochondrial respiration caused by misfolded αsyn

xxiii

LIST OF TABLES

Table 2.1: List of antibodies and fluorescent dyes used in this thesis Table 3.1: Inductively coupled plasma – mass spectrometry (ICP-MS) of PBSN and PBSN containing αsyn (PBSN+αsyn) Table 3.2: Values of significant differences identified in Figure 3.10 Table 4.1: Human trials performed for therapeutic intervention of prion disease. Table 4.2: Values of significant differences identified in Figure 4.7 Table 5.1: Values of significant differences identified in Figure 5.8

xxiv

CHAPTER 1: INTRODUCTION

Neurodegenerative proteinopathies (NDPs) are a group of disorders distinguished by the intra- or extracellular accumulation of specific proteins as β-sheet rich aggregates in the central nervous system (CNS), and associated neuronal vulnerability. These disorders include synucleinopathies [such as Parkinson’s disease (PD), dementia with Lewy body (DLB) and multiple system atrophy (MSA)], transmissible spongiform (TSEs; also known as prion diseases), Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), among others.

The pathogeneses of NDPs are extremely complex and the underlying biochemical processes that are implicated in these disorders are poorly understood. This limited understanding translates into poor prognostic outcomes. All NDPs are incurable, with current therapeutic strategies only aimed at alleviating symptoms rather than curative treatment. NDPs are also difficult to diagnose; this usually occurs after the onset of clinical symptoms and often is only confirmed by post mortem analysis by determining the type and location of misfolded protein, and profile of neuronal loss. Given that the development of clinical disease is an end- stage phenotype whereby substantial cell loss has already occurred, further understanding into the pathogenic pathways associated with these disorders are crucial to enable early diagnosis and effective therapeutic intervention.

Although the various NDPs are distinguished by disease-specific pathology and clinical presentations, they share conspicuous similarities. They are largely age-related disorders with most cases being of sporadic origin. The normal physiological function of the proteins that aggregate in disease is unclear, however they are all highly enriched in the CNS. Critically the protein that aggregates in disease can be directly associated with disease pathogenesis given that in the gene encoding the protein, or its precursor forms, cause early- onset familial disease. These similar features may reflect common pathogenic mechanisms are shared amongst the NDP disorders.

Of the proteinopathies, the unusual mechanisms governing the pathogenic nature of prion diseases are particularly striking. They represent a group of infectious NDPs where the causative and transmissible agent of disease is the prion or ‘proteinaceous infectious particle’ [1] that is composed almost exclusively of PrPSc; a misfolded isoform of the normal cellular prion protein, PrPC. The ability for a misfolded protein to be the pathogenic agent in disease

1 is referred to as the protein only hypothesis and dictates that prion disease occurs via the propagation of PrPSc, whereby PrPSc induces conformational misfolding in susceptible PrPC causing nucleation-dependent, autocatalytic misfolding [2]. This continuous seeding allows amplification of misfolded aggregates, forming first small soluble misfolded oligomers that extend forming protofibrils and mature fibrils. This continuous cycle of seeding allows prions to propagate within the host resulting in accumulation of β-sheet rich PrPSc in the CNS and widespread damage to neurons. During prion propagation, extension of the growing PrPSc aggregate can cause the aggregate to break, thus creating more nucleation seeds to amplify propagation (Figure 1.1). Prion propagation occurs during a long silent that can span decades, which is followed by the rapid onset of clinical disease and death.

Accumulating evidence suggests misfolded proteins associated with other NDPs share conserved molecular and biochemical properties with PrPSc that influences how they misfold, aggregate and propagate in disease. Referred to as the prion concept, this suggests that the description of a ‘prion’ may indeed also define other NDP-associated proteins. Some of the strongest evidence for prion-like behaviour exists for amyloid-β (Aβ) and α-synuclein (αsyn), proteins associated with AD and synucleinopathies (Figure 1.2), respectively [3]. Indeed, detailing of the stereotypical spread of these proteins in the CNS over the course of disease provided the first evidence for propagation of pathogenic αsyn and Aβ seeds [4-6]. It is important to note that in addition to Aβ and αsyn, many other proteins associated with NDPs have also been shown to share similarities with prions, including superoxide dismutase 1 (SOD1), huntingtin protein, and tau [7-10].

Whether or not other NDP-associated proteins aside from PrPSc adhere to the definition of a prion remains the subject of conjecture. As such, studying the similarities (and differences) of relatable proteins are important avenues of research to contribute to the delineation of what are conserved properties of aggregation-prone proteins and what are unique, protein-specific properties. This thesis explores pathogenenic mechanisms associated with PrP and αsyn misfolding with particular attention on features of disease where increased understanding may prove beneficial to the generation of effective therapeutic strategies. An important detail to clarify for this thesis is the terminology of the word ‘prion’, which has been used to describe misfolded proteins aside from PrPSc in several scientific articles [11-15]. Herein the only protein I refer to as a prion is the misfolded and transmissible PrPSc.

2

Figure 1.1 Nucleation-dependent model of prion replication. Misfolding of normal cellular prion protein, PrPC converts it into its transmissible, disease-associated isoform PrPSc. PrPSc induces nucleation-dependent misfolding in other PrPC leading to amplification and propagation of transmissible prions. The growing aggregates extend into oligomers, protofibrils and then fibrils that form the protein aggregates characteristic of disease. A natural component of prion propagation is breakage of the growing PrPSc aggregate, which produces more nucleation sites for templated misfolding.

3

General Background on PrPC, amyloid-β (Aβ) and α-synuclein (αsyn)

PrPC PrPC is a glycoprotein typically found attached to the via its glycosylphosphatidylinositol (GPI) anchor where it predominately resides within resistant microdomains [16]. It is strongly expressed in both neurons and glial cells of the CNS [17] and detected in low amounts in various peripheral compartments [18].

Encoded by the PRNP gene in humans, full length PrP is 253 amino acids in length, however mature PrP is shorter due to post-translational processing. The 22 N-terminal signal sequence is removed upon its entry to the endoplasmic reticulum (ER), while cleavage of the 23 amino acid C-terminal sequence occurs upon attachment of its GPI anchor [19, 20]. PrPC has a secondary structure that consists of two β-sheet and three α-helices [21, 22] and contains several core domains (Figure 1.3A). Closest to its anchor site, the C-terminal region is referred to as the globular domain (amino acid residues: 124 – 230) while the N-terminal is the unstructured flexible tail (amino acid residues: 23 – 123). The flexible tail contains a region of 4-5 glycine-rich octapeptides, spanning residues 59-89, referred to as the octapeptide repeat region. Polymorphisms lacking one octapeptide repeat region occurs in approximately 1% of the population [23] whereas supernumerary insertions of various lengths in this region are linked to familial forms of prion disease [24-27]. Linking the globular domain to the flexible tail is the hydrophobic region (amino acid residues: 112-133) that contains a palindromic region and glycine rich region. This region is almost perfectly conserved amongst [28]. Two N-linked glycosylation sites are also present on the molecule at positions 181 and 197 in the human amino acid sequence [29]. The occupancy of these sites is variable, leading to un-, mono- and di-glycosylated forms of the protein. This distinguishing feature transcribes a characteristic tri-banding pattern of PrP when detected using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

The normal physiological function of PrP is unknown. Because PrP knockout mice are viable and have no overt phenotype [30], PrP is not considered developmentally essential. Proposed functions of PrP include roles in: immune system regulation [31-41], circadian rhythm [42- 44], copper homeostasis [45-51], protection from stress [47, 52-61], [62, 63] and peripheral neuronal maintenance [64, 65]. The latter function has recently garnered renewed interest in a study demonstrating a role of PrP in the maintenance of Schwann cells; cells necessary for the integrity of peripheral neurons by ensuring appropriate myelination of

4 axons. In this study the flexible tail of PrP was found to cause a dose-dependent increase in cAMP in both primary and immortalised lines [65]. This observation was found to be due to an association of PrP with the G protein-coupled receptor (GPCR), Adgrg6; cAMP activation occurs in immortalised cells overexpressing Adgrg6 but not other GPCRs and the effect of the flexible tail of PrP is suppressed following ablation of Adgrg6 from cultured Schwann cells [65]. The identification of a receptor for PrPC is strong evidence that implicates it in a role in maintaining the integrity of peripheral neurons.

Aβ Aβ is a peptide produced by the sequential proteolytic cleavage of the integral membrane protein, amyloid precursor protein (APP) [66]. Two distinct processing pathways of APP may occur depending on the actions of its cleavage α-, β- and γ-secretases [67]. As the name suggests, the generation of Aβ occurs via the plaque-forming (amyloidogenic) pathway, where β-secretase (also known as BACE1) cleaves APP causing the generation of a membrane bound C-terminal fragment (CTF), CTFβ and a soluble N-terminal fragment (sAPPβ). CTFβ can be further proteolysed by γ-secretase, leaving the membrane-bound C- terminal fragment amyloid precursor protein intracellular domain (AICD), whilst liberating the soluble N-terminal fragment, Aβ (Figure 1.3B).

APP processing may also follow the non-plaque-forming (non-amyloidogenic) pathway. Here α-secretase cleaves APP to produce a secreted N-terminal fragment (sAPPα) and a C-terminal fragment, CTFα, of APP. The membrane-bound CTFα is another target of γ-secretase, which upon cleavage yields a soluble N-terminal fragment (p3) and AICD (Figure 1.3B). An important distinguishing feature of the amyloidogenic and non-amyloidogenic pathways is the differential cleavage sites of BACE1 and α-secretase; BACE1 targets a residue closer to the N-terminus of the APP which results in CTFβ being longer in length than CTFα. This means that upon proteolysis by γ-secretase, the Aβ fragment is longer in length compared to p3. This increased length is considered pertinent to Aβ’s propensity to misfold and aggregate.

Site specificity of γ-secretase is imperfect and hence can produce Aβ peptides that differ in length by several amino acid lengths. The most abundant Aβ species are those produced from cleavage at position 40 of CTFβ, referred to as Aβ40, which constitutes around 80 – 90 % of the total Aβ species. The next most common cleavage site is position 42, with Aβ42 comprising around 5 – 10% of the total pool of Aβ species. The ratio of Aβ42/ Aβ40 is found

5 elevated in familial AD patients [68, 69] and given that Aβ42 is more susceptible to fibrillization [70] it is generally considered the relevant pathogenic Aβ species in AD. However, in addition to Aβ40 and Aβ42, various longer Aβ species have been identified in AD brain [71], with recent studies showing Aβ43 to be more pathogenic than Aβ42 [72].

The biological function of Aβ is not resolved. Originally considered to be an unimportant peptide only expressed in aged brain, its identification in various bodily fluids [73] and released from cultured cells [74] argues Aβ has a physiological function. Examination of APP has led to proposed functions of Aβ associated with neuronal maintenance: transgenic APP knockout mice and mice expressing low levels of APP show a range of cognitive deficits [75-79] and administration of APP-directed antibodies impair memory formation in vivo [80, 81]. However, given that these studies are on APP, it cannot be confirmed the phenotype observed is due to Aβ production and not another product of APP metabolism. Several studies have studied Aβ which show evidence for it being involved in neurotransmission and memory [82, 83] and protection from toxic stress [84, 85].

αsyn αsyn is a small acidic protein that is highly expressed in neurons where it is largely localised to presynaptic terminals [86, 87]. The protein is encoded by the gene SNCA in humans [88] and contains three core regions which span its entire 140 amino acid length: an unstructured N-terminus (amino acid: 1-61), a central non-amyloid component (NAC) (amino acid: 61-95) and C-terminus (amino acid: 96-140) (Figure 1.3A). Here, the amphipathic N-terminal and NAC region contain seven repeat regions composed of imperfect KTKEGV hexameric motifs, while the C-terminal contains 10 Glu and 5 Asp residues and hence has a high net negative charge. The central hydrophobic NAC region was appropriately named following its identification within plaques of AD patients, as a component that was distinct to the previously identified Aβ protein [89, 90]. NAC peptides are capable of self-aggregation as well as seeding Aβ aggregation [91-93] and these misfolded species are toxic to immortalised neuronal cells [93]. As such, the NAC region is considered to be the highly pathogenic region of the protein.

Early reports described αsyn to have no or little secondary structure under physiological conditions [94-99], however it has since been shown to harbour weak secondary structures under certain native environments [100, 101]. The protein may also undergo various post-

6 translational modifications such as serine/threonine and tyrosine phosphorylation [102-106], N-terminal acetylation [102], ubiquitination [102, 107, 108], sumoylation [109], tyrosine nitration [110], transglutamination [111-113], and methionine oxidation [114].

αsyn belongs to the family of synucleins that also includes β- and γ-synuclein. A high degree of sequence homology is shared between the α and β proteins [115] and similar expression patterns are found throughout brain regions [116, 117]. Less homology is shared with γ- synuclein, which while also found in the brain [118], it has high expression in the peripheral nervous system [118, 119]. In contrast to αsyn, neither β or γ-synuclein harbour any inherent propensity to fibrillize [120]. Interestingly, β- and γ-synuclein have been implicated with neurodegenerative conditions: both proteins have been associated with neuronal pathology in PD and DLB brain [121] and γ-synuclein expression has been found to be altered in the retina of AD patients [122]. The mechanisms implicating either protein in these disorders are unknown, and hence further examination into the relevance of β- and γ-synuclein in NDPs is merited.

There lacks consensus on the primary function of αsyn, with the protein being reported to be involved in various and diverse biological processes. Proposed functions include roles in: regulation of glucose levels [123-126], activity [127-131], antioxidant [132-134], neuronal differentiation [135, 136], suppression of apoptosis [137] and regulation of dopamine synthesis [138, 139]. Many of these actions appear reflective on its strong expression in presynaptic terminals [86, 87], with some of the strongest evidence reporting the protein as essential to soluble N-ethylmaleimide-sensitive factor (NSF) attachment receptor (SNARE) complex assembly: a multimeric protein unit that is involved in the docking and fusion of synaptic vesicles with the presynaptic membrane in neurons [127-129, 131].

7

Figure 1.2. Disease-associated accumulation of misfolded protein in the central nervous system. The accumulation of PrPSc, Aβ and αsyn in the central nervous system is associated with prion, Alzheimer’s and synucleinopathy disease, respectively. PrPSc is a misfolded form of PrPC that is usually attached to the cell membrane, while Aβ is a peptide fragment produced from the proteolytic cleavage of amyloid precursor protein (APP). In disease, both PrPSc and misfolded Aβ are found aggregated in the extracellular space. αsyn is localised intracellularly and its misfolded aggregates can occur within neurons (as Lewy Bodies; LB) and oligodendroglia (as Glial Cytoplasmic Inclusions; GCI) in synucleinopathy disorders.

8

PrPSc: the Conventional Prion

Prion diseases represent a group of transmissible, invariably lethal NDPs that affect both humans and animals. In humans, prion disease can be caused by mutations in PRNP, resulting in familial Creutzfeldt-Jakob disease (CJD), Gerstmann-Stäussler-Scheinker (GSS) syndrome or fatal familial insomnia (FFI). These diseases can also be contracted following exposure to prions and include variant CJD (vCJD), and iatrogenic CJD (iCJD), or they may arise from an unknown origin (sporadic CJD; sCJD). Non-human prion diseases include scrapie in sheep and , bovine spongiform encephalopathy (BSE) in and chronic wasting disease (CWD) in deer and . The pathology of prion disease is characterised by extracellular deposition of aggregated PrPSc (Figure 1.2), along with and vacuolation including associated neuropathological spongiform change in the CNS.

The infectious nature of prions is well established to transmit disease both within and across species. Early studies showed brain homogenate from scrapie-infected sheep could successfully transmit disease to goats [140] and mice [141]. Likewise, numerous studies demonstrated human prion disease could be transmitted to non-human primates [142-144]. In humans, acquired prion disease has been associated with several serious epidemics. Kuru, one of the most prominent forms of prion disease, afflicted the native Fore tribes of Papua New Guinea. In these populations, ritualistic was identified as the route of transmission and cessation of this practice has virtually eradicated new cases of disease [145]. The finding that consumption of BSE-infected beef was responsible for vCJD [146, 147] led to one of the largest health crises of the 20th century, causing widespread economic damage to countries found to have a contaminated beef-industry and concern of large-scale public exposure to the due to the long incubation period of disease. Additionally various iCJD cases have been documented with routes of unintentional human-to-human transmission including the use of inappropriately sterilised surgical equipment [148, 149], dura mater grafts [150, 151], corneal grafting [152, 153], blood transfusion [154, 155] and use of human cadaveric pituitary-derived growth hormone or gonadotrophin [156, 157] (for an extensive list of references on human-to-human transmission of prion disease, refer to: [158]). Of these, the highest number of iCJD cases are associated with dura mater grafts and hormone administration [159]. These cases of acquired disease have led to major reform in the handling of biological samples in the and medical industry to reduce contamination and prevent future outbreaks of prion disease.

9

The conversion of PrPC to PrPSc is concomitant with substantial structural and biophysical alterations to the molecule. Upon misfolding, the α-helix rich PrPC switches to one primarily composed of β-sheet [160, 161]. These changes lead to an increased resistance to degradation by heat and ; with the latter feature being exploited for its detection by incubation with proteinase K (PK). In addition, it is well accepted not all PrPSc are identical and rather, PrPSc encompass a range of various sized PrPSc quasi-species [162] that harbour unique properties; for example, the most infectious prion particles appear to be those of small masses consisting of 14-28 molecules [163, 164].

Numerous data demonstrate that propagation of PrPSc via conformational misfolding of PrPC is an essential component of disease pathogenesis. Transgenic PrPC knockout mice are resistant to infection by prions [165] and ablating PrPC expression in prion-infected mice after the onset of clinical symptoms halts disease progression and rescues early neuropathological changes [166]. The development of assays that study the conversion of PrPC to PrPSc in cell free systems provides additional insight into the misfolding of the protein. In the cell free conversion assay, radio-labelled recombinant PrPC adopts PrPSc-like features such as resistance to PK when incubated in the presence of partially denatured PrPSc derived from prion infected hamster [167]. Another cell free assay, the protein misfolding cyclic amplification (PMCA) assay, shows amplification of PK-resistant PrP (referred to as PrPRes) occurs when excess PrPC is subjected to repeated rounds of sonication and incubation in the presence of a small PrPSc seed [168]. When subjected to PMCA, normal brain homogenate seeded with a small amount of prion-infected brain homogenate amplifies PK- resistant, insoluble PrPSc-like species, that transmits disease when inoculated into wild-type mice [169]. Collectively these studies demonstrate the requirement of PrPC in the generation of PrPSc and in the propagation and transmission of prions in vivo.

An important component to proving the prion hypothesis required demonstration that PrPSc alone was a transmissible element that could induce the pathology of prion disease. Studies showed recombinant PrPC may be misfolded to form PrPSc-like species that causes disease when inoculated into PrP-overexpressing mice [170] or in wild-type mice when the protein is misfolded in the presence of normal brain homogenate [171], however variations existed in the pathology and long incubation period required for presentation of disease indicating PrPC alone may not be the sole pathogenic agent in traditional prion disease. An explanation for this variation was suggested by the observation that additional factors, or co-factors, could contribute to the misfolding of PrPC. Using PMCA, PrPSc-like species may be formed from

10 normal brain homogenate or purified brain-derived PrPC combined with RNA species [172, 173], both that cause disease in wild-type rodents. The latter finding, along with others, show polyanionic and lipid species are required for efficient misfolding of PrPC and infectivity [172, 174-177]. By the addition of co-factors, misfolding recombinant PrP in the presence of RNA and lipids produces de novo prions that are highly infectious in vivo and manifest disease in wild-type mice which faithfully replicates conventional prions [178]. These studies confirm an altered conformation of PrP is the major component responsible for infectivity of prions, however the requirement of co-factors suggests other molecules may form a natural component of a transmissible prion.

An unusual feature of prion disease is strain variability, whereby unique patterns of clinical presentation, incubation period, PrPSc deposition and neuronal vulnerability are displayed amongst prions [179]. These characteristics are conserved amongst serial passage in inbred mice and are not an artefact of the host given that prion strains may be reisolated in mice following passage in intermediate species [180]. Unlike conventional where strain variability is a result of genomic alteration, it is unusual for multiple disease phenotypes to be enciphered by a polypeptide chain. Important knowledge into strain variability came from elegant studies on two hamster-adapted prion strains Hyper (HY) and Drowsy (DY). These two strains arose from the same original animal and were propagated through hamsters with identical PRNP sequence, however while hamsters inoculated with HY prions develop and hyperaesthesia, hamsters infected with DY prions show lethargy and a longer incubation period [181]. These divergent clinical features are coupled with unique pathological profiles of between the two strains [181] and biochemical analysis of the of animals infected with HY or DY reveal differences in electrophoretic mobility of PrPRes [182], suggesting the structure of PrPSc is altered between the two strains. This observation is further supported by the finding that these strains also exhibit different infrared spectrum of absorption in the range that correlates to β-sheet content [183], indicating the secondary structures of HY and DY are dissimilar. Structural variability of PrPSc is also likely to occur in the human condition, where PrPSc present in sporadic prion diseased brain are distinguished by changes in electrophoretic mobility which appears to correlate with pathological and clinical features of disease [184]. Although the general consensus is that the conformer of the misfolded protein encodes its disease phenotype, the origin of unique strains and how these conformational changes produce the unique clinical and pathological profiles is still poorly understood.

11

Figure 1.3: Human PrPC, αsyn and Aβ and the generation of the Aβ peptide by the proteolytic processing of APP. (A) Full length human PrPC is a glycoprotein that contains several structured regions (two β-sheet and three α-helices) and several core domains. αsyn contains several core regions and has seven imperfect KTKEGV repeats. Aβ is a small peptide that can vary by several amino acid lengths. (B) The peptide Aβ is produced from the proteolytic cleavage of APP by β- and γ –secretase following the amyloidogenic pathway. APP may also be processed by α- and γ-secretase via the non- amyloidogenic pathway which produces p3; a shorter peptide than Aβ.

12

Prion-like Properties of Misfolded Aβ

Alzheimer’s disease is a slow progressing NDP that is the most common cause of dementia. It is associated with the accumulation of primarily composed of Aβ in the extracellular space of the CNS [185] (Figure 1.2). Along with Aβ deposition, AD is associated with the accumulation of neurofibrillary tangles consisting of hyperphosphorylated forms of the microtubule-associated protein, tau. Aside from age, the biggest known risk factor for AD is the expression of the ε4 allele of the lipoprotein ApoE (ApoE4) [186] that worsens the trajectory of cognitive decline in association with Aβ burden in preclinical AD individuals [187-189].

Some of the first indications of prion-like behaviour of Aβ came from primate studies, where Aβ deposition was found in the CNS of marmosets 6-7 years after intracerebral inoculation of human AD tissue [190], suggesting pathogenic Aβ seeds may propagate within its host similar to prions. Further support for this concept came following the generation of transgenic animals. Mice expressing APP containing mutations associated with human familial AD spontaneously develop Aβ pathology and are used to model disease, with two of the most widely used transgenic mice Tg2576 [191] and APP23 [192] expressing the Swedish familial K670M/N671L. Unilateral inoculation of dilute brain homogenate from human AD patient brain into the and neocortex of young Tg2576 mice accelerates the deposition of Aβ, with extensive Aβ-pathology found 5 months post injection whilst limited or no deposition occurred in Tg2576 counterparts injected with age-matched non-AD or young non-AD brain extract [193]. Under the same experimental conditions, a similar finding was observed by aging the mice to 12 months, where although Aβ deposition was present bilaterally, protein deposition was more extensive in the hemisphere inoculated with AD brain [194]. Consistent with the ability of Aβ-rich brain homogenate to induce this pathology, inoculating young presymptomatic mice with brain homogenate from aged counterparts that have developed extensive Aβ deposition accelerates the generation of Aβ deposition [195].

Since these findings, numerous studies have highlighted the relevance of Aβ in accelerating pathology in AD-model mice. Removal of Aβ from aged APP23 brain homogenate via immunodepletion or immunisation against Aβ abrogates the ability of extracts to trigger Aβ pathology [195] and pre-treating human AD brain homogenate with magnetic beads coupled to a compound that binds misfolded protein decreases Aβ levels in the homogenate and reduces its seeding potential in vivo [196]. Consistent with these observations, the

13 concentration of Aβ in the seed and its expression in the brain determines the time and severity of Aβ deposition, and is independent of the age of the mouse when inoculated [195, 197]. Aβ seeds derived from of patients with amyloid burden without dementia maintain the capacity to induce pathology in AD-model mice, confirming the importance of Aβ seeds, and not other factors associated with dementia in AD, in inducing pathology [198]. Additional evidence for Aβ being the causative agent for accelerated Aβ deposition comes from the finding that synthetic misfolded Aβ is capable of triggering Aβ pathology in AD- model mice [12, 199]. The development of Aβ-pathology in transgenic APP-model mice has been studied where Aβ propagation follows a systematic spread with induction of pathology corresponding to the limbic connectome [200]. Collectively these studies give strong support for Aβ being the agent driving the accelerated Aβ pathology seen in AD-model mice and supports the concept that Aβ may propagate similar to prions.

The transmissible nature of Aβ seeds share several similarities with prions. While lacking any significant evidence, an alternate theory to the causative agent of prion diseases was one of viral origin [201-203]. This is now considered implausible, due in part to studies confirming the causative agent is highly resistant to the virus neutraliser, formaldehye [204, 205]. Similar to PrPSc, the ability of Aβ seeds to propagate in vivo is maintained following treatment with [206]. Likewise misfolded Aβ harbours the ability to be transmitted on steel wires [207]; a concept developed to replicate transmission of prions via surgical equipment [208, 209]. It has also been reported small aggregates of Aβ are the most infectious [210]. Although synthetic Aβ may induce AD pathology in vivo [12, 199], it is markedly less effective than brain homogenate [199], which may indicate biological co-factors are required for efficient propagation of Aβ in vivo. The propagation of Aβ in vivo has also been confirmed following injection of Aβ seeds via the intraperitoneal route [211, 212] consistent with propagation of pathogenic seeds from the periphery to the CNS.

Although compelling, there are several caveats to using AD-model mice to study Aβ propagation. Acceleration of an existing Aβ pathology in transgenic mice after exposure to an Aβ extract does not necessarily prove that Aβ is transmissible; however the generation of Aβ pathology in transgenic mice expressing wild-type human APP or mutant human APP that would not develop spontaneous Aβ pathology in the experimental timeframe does support the transmissible properties of Aβ aggregates [197, 213, 214]. In contrast cerebral spinal fluid (CSF) from AD patients that harbour much higher levels of Aβ compared to brain extracts did not induce Aβ pathology in young APP23 mice [215], which may indicate co-factors

14 associated with Aβ in the brain but not CSF may influence the ability of Aβ to propagate pathology. An additional consideration using these transgenic mice is that although Aβ deposition can be accelerated by the addition of Aβ seeds, neurological dysfunction is absent. This highlights a major limitation to the use of transgenic mutant Aβ animals, which do not develop this feature of disease associated with protein aggregation, to model AD.

Evidence of transmission of Aβ in humans has been reported where unusual deposition of Aβ was found in the brains of iCJD patients who contracted disease following human cadaveric pituitary-derived growth hormone treatment [216]. Of a small sample group of eight, four contained extensive deposition of Aβ in the CNS parenchyma and varying degrees of cerebral amyloid angiopathy (CAA); the pathological deposition of Aβ in blood vessels that occurs in approximately 80% of AD patients [217]. An additional three cases had sparse Aβ deposition or Aβ deposition in association with PrP plaques. Given the young age of the patients (age range: 36 – 51 years), it is extremely unlikely the Aβ pathology arose sporadically. A possible explanation for this observation is that Aβ was also present in the cadaver-derived growth hormone. As further evidence that Aβ deposition occurred through a transmission event, the growth-hormone recipients did not have any known factors that may have predisposed them to Aβ pathology, including known AD-associated genetic mutations and positive ApoE4 status. It is possible that the Aβ deposition was a result of cross-seeding by PrPSc, however in the majority of cases co-localisation was not found and post mortem pathological assessment of a larger number (35) of patients with genetic or sporadic prion disease failed to identify Aβ pathology in comparable age groups, with the exception of two cases that were positive for the ApoE4 allele [216]. As proof-of-principle for the presence of Aβ in pituitary-derived growth hormone, deposited Aβ has been found in the pituitary gland of AD patients [216, 218]. In support of this, another study has reported unusual Aβ pathology in iCJD patients who contracted disease after dura mater grafting [219]. Five out of seven patients (age range: 28 - 63 years) reported Aβ pathology and CAA consistent with AD, which was significantly greater than the frequency of Aβ pathology seen in 21 age- matched controls. In both cases, although it is intriguing to postulate iatrogenic transmission of Aβ has occurred, the overarching manifestation of iCJD makes it impossible to ascertain whether this transmission would have resulted in clinical AD.

Strain variation may also be a feature of Aβ. In vitro fibrillization of monomeric Aβ seeded with human brain from AD patients that exhibited different clinical and pathological profiles produces fibrils with unique morphological profiles [220]. Differences can also be found

15 amongst different transgenic mutant APP mice such as APP23 and APP/PS1 mice; a transgenic mouse line that contains the APP Swedish mutation along with a mutation in the AD-associated gene PSEN1 at position L166P. Aβ isolated from these mice harbour unique profiles in morphology, binding-affinities to amyloid dye and Aβ peptide ratios. Consistent with the concept of prion strains, inoculation of APP/PS1 brain homogenate into APP23 induces the formation of Aβ aggregates with similar characteristics of the inoculum [221]. A similar finding occurs using recombinant protein, where Aβ species induce differential plaque characteristics and Aβ peptide ratios in mouse brain when produced in the absence or presence of detergent [12]. In addition, unique profiles of Aβ pathology develop in transgenic mice inoculated with human brain from patients carrying different familial AD-associated APP mutations. The Artic mutation (E693G) occurs within the Aβ sequence of APP and disease is characterised by increased protofibril formation and distinct AD pathology, whereas the Swedish mutation (K670M/N671L) is located outside the Aβ sequence, causing disease associated with the overproduction of wild-type Aβ and typical AD pathology that is similar to sporadic disease. Transgenic APP23 mice inoculated with brain homogenate containing the Artic mutation exhibited unique aggregation characteristics and regional deposition profiles compared to those inoculated with homogenate derived from sporadic or Swedish AD brain with one of the most striking differences found in the morphology of CAA in the [13]. These apparently strain-specific features are conserved following serial passage in mice [13].

Prion-like Properties of Misfolded αsyn

The deposition of misfolded αsyn in the CNS is a pathological hallmark of synucleinopathies and includes PD, DLB and MSA, among others. Although overlapping clinical presentations can make correct diagnosis difficult, synucleinopathies are distinguished by unique profiles of cell types and brain regions susceptible to develop αsyn pathology and the locality of neuronal vulnerability [222]. For example, the pathology of PD and DLB is characterised by intraneuronal deposits of αsyn called Lewy bodies (LB) or Lewy neurites (LN) [223-227] whereas MSA-associated αsyn is distinguished predominantly by cytoplasmic inclusions called glial cytoplasmic inclusions (GCIs) and these occur in a type of glial cell called oligodendroglia [228-230] (Figure 1.2). In PD, the neurons that are most vulnerable are

16 dopaminergic cells of the substantia nigra whereas in MSA and DLB the profile of neuronal loss is more widespread.

Disease-associated changes to the post-translational modifications of αsyn is a common feature amongst the synucleinopathies. Most notably, high levels of αsyn phosphorylated at serine reside at position 129 (pSer129) are found in both LBs and GCIs [231]. Under normal physiological conditions pSer129 accounts for a low abundance in the overall pool of αsyn, however within LBs it accounts for over 90% of the total protein [102]. The biochemical processes that lead to elevated phosphorylation of αsyn in disease and the consequence of such an alteration are unknown. Recently a study showed that compared to wild-type αsyn, the expression of mutants that do not undergo phophorylation (S129A and S129G) is more toxic to cells and produces the formation of αsyn aggregates that are larger [232]. Hence it is possible the phosphorylation of αsyn is a protective mechanism to remove protein aggregates from the cell.

The expression of αsyn protein is associated with disease as duplications [233] or triplications [234] in SNCA, and various point mutations cause early onset familial PD. To date, five mutations in αsyn have been found associated with early onset PD: A53T [235], A30P [236], E46K [237], H50Q [238, 239] and G51D [240]. Similar to the human condition, transgenic animal models that express elevated levels of human wild-type αsyn, or αsyn containing PD- associated mutations spontaneously develop αsyn pathology and neurological illness and are thus used to model disease [241-244]. While transgenic animal models expressing mutant αsyn are typically used to study disease associated with protein aggregation, many other transgenic systems have been developed to study other features of disease such as dopaminergic cell loss and motor impairment [245].

Some of the earliest evidence that αsyn may have the characteristics of a prion came from autopsy of PD patients who, as part of a clinical trial, had fetal nigral dopaminergic nerve cells grafted into their brain. The presence of αsyn deposits in the transplanted tissue suggested in the 11-16 years post-surgery, αsyn transmission occurred from the host to the grafted tissue [246, 247]. Contrasting reports on fetal transplants in PD patients have described grafted regions remain absent of αsyn pathology for up to 14 years post-surgery [248], however numerous in vitro and in vivo data support propagation of αsyn seeds. In immortalised cells overexpressing mutant A53T αsyn protein, exposure to misfolded αsyn species leads to punctate staining of αsyn suggestive of propagation of disease-associated

17 inclusions [249]. Similarly, using cationic-liposomes to aid transfection, fluorescently-tagged fibrillar αsyn has been shown to enter immortalised cells overexpressing αsyn and induce changes in endogenous protein consistent with pathogenic αsyn such as phosphorylation, ubiquitination and insolubility [250]. A caveat to these studies is the requirement for αsyn to be overexpressed in the cells, which does not model disease where the majority of synucleinopathies express normal levels of αsyn protein, and the use of artificial vehicle reagents to aid the delivery of misfolded protein into the cell. However, a similar finding has also been found following extracellular exposure of primary neuronal cultures from wild-type mice to fibrillar αsyn species without the aid of vehicle reagents, which resulted in the recruitment of endogenous αsyn into insoluble LB-like inclusions and associated changes to neuronal vulnerability including decreases in synaptic protein and neuronal excitability [251].

Several other studies support in vitro propagation of αsyn. Co-culturing two cell lines expressing αsyn containing different fluorescent tags results in the progressive emergence of double-labelled cells [252] and treating tagged αsyn cells with misfolded αsyn tagged with a different probe leads to significant co-localisation in cultured cells [253]. In immortalised cells expressing tagged-αsyn, incubation with human brain derived from MSA patients causes the tagged-αsyn to aggregate [11, 14]. In these cultures, αsyn aggregates have been shown to transverse in both retrograde and anterograde direction and transfer to neighbouring neurons and glial cells [249, 252, 254]. Collectively these studies give strong evidence for prion-like transmission of αsyn in vitro.

Prion-like propagation of αsyn is further supported by in vivo studies inoculating αsyn seeds into susceptible mice. In transgenic mice expressing homozygous mutant A53T αsyn (M83+/+), inoculation of fibrillar recombinant αsyn or brain homogenate from aged, terminal mice, into the brain of young pre-symptomatic mice accelerates the development of αsyn CNS pathology and clinical disease [255-260]. A similar observation is found following intramuscular or intravenous injection of αsyn seeds [257, 261]. In these studies, similar to the human condition, aggregated αsyn is largely phosphorylated and other pathological features of disease are present, including reactive astrogliosis and microgliosis. Consistent with the observations in human studies, grafting dopaminergic neurons into transgenic mice overexpressing human αsyn leads to the identification of αsyn aggregation in the grafted tissue [252].

18

Seeded propagation has also been reported after inoculation of misfolded recombinant αsyn in wild-type rodents [262-267] and transgenic mice overexpressing wild-type αsyn (M20+/+) that do not spontaneously develop inclusions over their lifetime [267, 268]. In some studies, motor impairment has been reported to accompany propagation [262, 266], and accordingly, treating mice with antibodies targeting misfolded αsyn via intraperitoneal injection reduces neuronal loss and delays the onset of clinical disease [266]. Similarly, human MSA brain may induce propagation of αsyn and clinical disease in hemizygous M83+/- mice, which do not spontaneously develop inclusions or clinical disease in most experimental timeframes [11, 269]. Although the propagation of αsyn in vivo has been extensively reported, with αsyn pathology extending along white matter tracks to regions distant from the injection site, its identification may be associated with a technical artefact. Antibodies targeting phosphorylated αsyn (pSer128/81A) have been shown to cross-react with neurofilament subunit L that is abundant in white matter [258], hence previously reported models of propagation may include histological artefacts falsely attributed to phosphorylated αsyn. This observation may be a matter of dose threshold of the , as others have found no cross- reactivity [262, 263, 265, 266, 270], including minimal pSer128/81A staining in a αsyn brain and primary neuronal cultures [251, 262]. Regardless, the specificity of these antibodies requires further investigation and it is an important technical issue to clarify.

Although the evidence for prion-like propagation of αsyn is extensive, variation exists in the incubation period, pattern of αsyn pathology and mouse line used. For example, whilst M83+/+ mice develop extensive αsyn pathology following injection with recombinant fibrillar αsyn, in other transgenic mouse lines, such as M47+/+ (expressing human αsyn E46K mutation), αsyn deposition is largely restricted to the injection site [258]. This may indicate artefacts associated with αsyn aggregation with the expression of certain transgenes, or that the normal wild-type αsyn cannot be patterned and amplified using the E46K mutated αsyn. Furthermore, it has recently been reported injection of spinal cord homogenate from wild- type mice induces robust CNS αsyn pathology in M83+/- mice [271], suggesting inherent characteristics of the transgenic M83 mouse may contribute to αsyn pathology. Although propagation of αsyn pathology has also been reported in wild-type animals [262-267], similar results are unable to be achieved in other studies [258], which may, in part, be associated with the use of the pSer128/81A antibody or the type of seed injected being unable to interact with the native αsyn. An important consideration is that mouse αsyn contains a threonine at

19 position 53 and given that most studies using transgenic αsyn mice express the transgene on a wild-type background, species-specific contributions by endogenous αsyn may be a confounding factor in these studies. However, recently it has been shown transgenic mice expressing wild-type human αsyn on a knockout αsyn background are also conducive to CNS αsyn propagation following injection of human brain homogenate from DLB and MSA patients [272], giving evidence for αsyn propagation being a conserved observation in various genetic backgrounds.

Similar to prions, lipids may represent a co-factor to enhance the fibrillization of αsyn; lipid- rich exosomes accelerate the fibrillization of monomeric αsyn protein [273] and appear to act by stimulating primary nucleation [274]. Misfolded αsyn species may be produced using PMCA [275], however unlike PrP misfolding, aggregation of αsyn may be induced easily from recombinant wild-type protein in the absence of co-factors or serial rounds of propagation. Prions do not exclusively form amyloid and may exist as various soluble and insoluble species, and similarly the capacity of αsyn to spread in vivo is conserved in both fibrillar and non-amyloidogenic seeds [267]. The ability to generate various αsyn species from the same starting material in vitro is reminiscent of strain variation amongst prions. Various types of mature fibrillar αsyn with distinst morphology may be easily produce from recombinant protein and two populations of misfolded αsyn, ribbons and fibrils, exert differential abilities to cross-seed fibrillization of monomeric αsyn or and induce toxicity in immortalised cells [253, 270]. Following injection into rat brain, ribbons induce more abundant deposition of phosphorylated αsyn, whilst fibrils were shown to represent the most toxic species, causing cell death and associated motor impairment after injection [257]. Brain tissue from MSA patients exhibited differences in triggering aggregation of αsyn in vitro [14] and αsyn isolated from human PD brain show differential cleavage products following degradation by PK [270]. Intriguingly, recent evidence also demonstrates differences in transmissibility between αsyn derived from various human conditions. In M83+/- mice, inoculation with human MSA, but not PD brain homogenate induces disease consisting of deposited phosphorylated αsyn, progressive neurologic dysfunction, astrogliosis and microgliosis [11]. The aggregation of αsyn in these mice following inoculation with MSA brain extract was predominantly observed as cytoplasmic inclusions located in neuronal populations, which is in contrast to its deposition in oligodendroglia characteristic of MSA [11]. The pathological phenotype induced in these animals is independent of total αsyn protein and instead correlates to the capacity of the brain homogenate to induce aggregation

20 in cultured cells. Furthermore, the clinical disease that develops in MSA-infected mice is transmissible, which causes disease when inoculated into young M83+/- mice. Given the ability of MSA-derived αsyn to induce disease that is transmissible, it has been suggested αsyn associated with this disease is, by definition, a prion [11, 269].

21

Figure 1.4: Propagation of misfolded protein in rodent models following intracerebral inoculation of a misfolded seed. Propagation of PrPSc, Aβ and αsyn has been studied in various mouse models. Images represent the endogenous aggregation of protein in the brain of rodents that occurs over their lifetime (green line) and aggregation that occurs following the injection of a seed (red line) (A) Wild-type mouse models do not spontaneously develop protein aggregation over their lifespan and the induction of a seed induces aggregation of protein (B) Transgenic mouse models that spontaneously develop protein aggregation and the incorporation of the seed accelerates the deposition of protein. (C) Transgenic mouse models that do not spontaneously develop inclusions over the experimental timeframe, and the induction of a seed induces protein aggregation.

22

Should We Expand the Prion Concept?

Mounting evidence appears to draw parallels between αsyn, Aβ and PrPSc, with studies supporting prion-like seeding, propagation and transmissibility of pathogenic proteins causing clinical disease in animals, however caveats exist associated with the use of transgenic mice to adequately model disease. For example, the possibility of other agents aside from αsyn to cause aggregation raises important considerations when evoking prion- like mechanisms and reflects hazards associated with using transgenic mice where aggregation of protein is an inherent characteristic of that model. Studies that use transgenic mice that remain asymptomatic may alleviate these issues, however it is not clear whether expressing lower levels of the transgene are entirely innocuous. In prion disease, transmission of disease into susceptible wild-type animals may readily be achieved, and represent bona fide models of disease. Therefore the gold-standard for investigating prion-like propagation and transmission of misfolded protein would require the use of non-transgenic animals. Numerous data support the ability of certain αsyn species to propagate and transmit disease in wild-type mice [262-267], however in the case of Aβ, equivalent studies require the use of transgenic APP animals (Figure 1.4). Taking this into consideration, as the current evidence stands, no other protein aside from PrPSc has conclusively been confirmed as a prion and hence the term prion should pertain to PrPSc only.

If conclusive evidence arises showing other misfolded proteins constitute a prion, there would undoubtedly be substantial implications on the health and safety regulations of these diseases. However it is important to bear in mind that prions derived from different proteins will likely have different biochemical characteristics. Recently it has been shown misfolded αsyn are markedly easier to decontaminate from surfaces and do not require the stringent techniques required for decontamination of prions [276], and as such the potential for untoward public health risks, such as unintentional transmission of misfolded αsyn, seems unlikely.

23

Mechanisms of Neurotoxicity in Proteinopathies

A consistent feature that accompanies the accumulation of protein in the human CNS is the progressive damage and loss of neurons. In NDPs, although disease-specific differences are found in the tempo and localisation of neurotoxicity, similarities are found which may indicate conserved mechanisms underlie the generation of cell death. This may be reflected, in part, by conserved fibrillization properties, where fibril formation involves the misfolding of normal monomeric protein and extension into various species (oligomers, protofibrils) prior to its maturation into mature fibrils (Figure 1.1). In this regard, extrapolation of knowledge on the toxicity associated with the misfolding of one NDP-associated protein may prove rewarding in deciphering general neurotoxic mechanisms that arise when a protein misfolds in a cell.

Historically, the presence of misfolded proteins in the brain led to the belief that the sheer aggregation of protein was responsible for neurotoxicity in NDPs, however today this is considered an oversimplification. In human prion disease, large variation exists in the protein deposition seen in the CNS at post mortem, where in some cases occurs with minimal PrPSc load [277, 278]. These findings are supported by animal models of disease where transgenic mice expressing mutations associated with familial prion disease have less aggregated protein in their brain at terminal stage disease compared to wild-type mice inoculated with prions [279], or PrPSc is entirely absent [280]. Prion infection in mice expressing PrPC lacking a GPI-anchor causes disease characterised by a long incubation period prior to the development of clinical symptoms, despite the accumulation of a large amount of PrPSc in the CNS [281]. In addition, under certain conditions, subclinical infection can occur where PrPSc propagation occurs without the development of clinical symptoms or neurodegeneration. Mice infected with hamster prions remain asymptomatic despite having prion titres in their brain equivalent to end-stage disease in conventional mice [282]. Other states of subclinical infection are achieved when low inoculum or oral infection routes are used [283-285].

Disassociation between protein aggregation and neurotoxicity is similarly observed in AD and synucleinopathies. In human AD brain, cell loss is not associated with regions of protein deposition [286, 287] and in certain transgenic animal models of AD, death can occur in the absence of amyloid formation [288]. Likewise, insoluble αsyn does not correlate well with the degree of neurodegeneration in human brain [289-291]. This observation extends to

24 animal models where in transgenic rodent and fly models of PD, cell loss can occur in the absence of protein aggregation [241, 292]. These studies suggest that deposition of misfolded protein is unlikely to be the toxic agent and instead may be an unrelated endpoint to disease pathogenesis.

In prion disease, toxicity is it not considered associated with a loss of function of the protein because PrP knockout mice exhibit no neurodegenerative phenotype [30, 293]. Instead neurotoxicity involves an interaction between PrPC and PrPSc, in view of the finding that only cells that express PrPC may be damaged by PrPSc. Prion-infected PrP knockout mice hosting grafted PrP-expressing tissue only develop neurodegeneration in the PrP-expressing tissue, whilst PrP deposition is widespread across the brain [294]. Similarly, ablation of PrP in mice with established prion infection halts disease development and reverses clinical symptoms, despite these mice harbouring levels of PrPSc in the brain equivalent to end-stage disease [166]. These studies give strong evidence for the conversion of PrPC to PrPSc, and not the sheer aggregation of protein, being a major component to neurotoxicity. These findings may be relevant to neurotoxic mechanisms of αsyn and Aβ misfolding, and thus unifying knowledge may help shed light on general toxic mechanisms of folded protein.

Toxic Oligomers

Some of the strongest evidence exists for small soluble oligomers of misfolded protein being the neurotoxic agent in proteinopathies. These species are highly unstable compared to the organized β-sheet rich structures formed by mature fibrils and by virtue of size, smaller aggregates have a higher relative exposed surface area than larger mature aggregates [295]. Because of this, it has been suggested the recruitment of misfolded protein into aggregates is a mechanism employed by the cell to sequester small toxic species away from vital cellular compartments where they cause substantial damage. This may represent a common mechanism shared amongst the various proteinopathies.

Aβ oligomers Neurodegeneration in AD has been correlated to the pool of soluble Aβ species [296, 297]. These findings are supported by in vitro studies showing particularly aggressive forms of AD,

25 such as the Arctic mutation, produces larger quantities of prefibrillar species [298], an observation which is also seen when Aβ fibrillization occurs in the presence of ApoE4 [299]. In wild-type mice, oligomeric Aβ preparations, but not insoluble amyloid plaque cores, alters synaptic plasticity by blocking long term potentiation (LTP) and activating long term depression (LTD) [300, 301]; electrophysiological correlates of learning and memory. They also trigger dendritic spine retraction [302, 303], reduce dendritic spine density [301] and cause memory deficits [301, 304]. The ability of Aβ species to induce toxicity in cultured cells is enhanced in heterogeneous populations of Aβ species compared to discrete species, with the toxicity of protofibrils being linked to the expression of monomeric protein [305]. This finding suggests ongoing fibril formation involving interactions between misfolded species and monomeric Aβ is pertinent to toxicity. Collectively these studies give strong evidence for oligomeric species of Aβ causing damage to neurons.

The mechanisms underlying Aβ oligomeric-induced toxicity has also been studied, where the N-Methyl D-Aspartate (NMDA) and metabotrophic glutamate-5 (mGlu5) receptors have been associated with the development of LTP and LTD respectively [301, 303]. Interestingly, the modulation of the mGlu5 receptor appears to involve PrPC, which acts as a receptor for Aβ oligomers and complexes with mGlu5 [306]. Accordingly, PrP-directed antibodies inhibit Aβ-oligomer induced toxicity [307]. In addition, it has been reported PrP knockout mice are resistant to toxicity elicited by Aβ oligomers [306], however follow up studies have been unable to observe such a striking effect [308-311]. Aside from acting on neuronal receptors, Aβ oligomers may also cause toxicity by inducing hyperphosphorylation of tau, leading to collapse of microtubule cytoskeleton and neuritic dystrophy [312].

The conformation of the toxic Aβ species is still unclear. Investigations into the toxicity of structurally defined oligomers demonstrate neurotoxicity in cultured cells increases with oligomer order for species larger than a monomer [313]; a finding which is supported by studies showing the smallest toxic Aβ species is the dimer [301, 312]. However others report dimers correlate poorly with toxicity in cultured neurons and instead trimers and tetramers may be the smallest toxic agents [314]. Trimers are also believed to be the basic component of Aβ*56 [304]; a non-fibrillar 56-kDa Aβ oligomeric species that has also been implicated in toxicity. This species is naturally present in human brain and CSF, where it correlates with soluble tau and negatively correlates with several synaptic proteins in cognitively intact individuals [315, 316]. In rats, Aβ*56 induces transient cognitive dysfunction independent of amyloid deposition and neuronal loss [304] and as such Aβ*56 may contribute to the early

26 deficits seen in AD, prior to the accumulation of protein indicative of disease. Given Aβ*56 is likely to be a dodecamer (based on its apparent molecular weight on SDS-PAGE), further support for it being the toxic agent comes from studies highlighting the role of large oligomeric structures of Aβ as potent toxic oligomers [317-319]. However, there is a technical issue that complicates this interpretation because the presence of Aβ*56 can be produced by filtration or separating Aβ species via electrophoresis in the presence of SDS [320]. Regardless, given that there is no evidence Aβ*56 alone can induce persistent neurological dysfunction, its generation is not considered necessary for toxicity associated with Aβ misfolding.

αsyn oligomers In human DLB brain, the presence of αsyn aggregates at presynaptic terminals, rather than LB abundance, correlates with decreases in presynaptic markers and dendritic spine loss [321]. Similarly, soluble oligomers are found enriched in regions of PD brain found to have deregulated markers of protein synthesis [322]. In rodents, the expression of mutant αsyn that preferentially forms oligomers is more toxic than mutations which more readily form fibrils [323]. In primary neuronal cultures and transgenic invertebrate (Caenorhabditis elegans and Drosphilia melanogaster) model systems, the expression of variants of αsyn that increase its propensity to oligomerise and decrease the formation of mature fibrils are more toxic than wild-type αsyn [324]. Misfolded αsyn species produced in the presence of a compound that promotes formation of large fibrillar aggregates and reduces oligomer species exhibits reduced seeding potential and toxicity to cultured cells compared to untreated fibrils [325]. Various oligomeric species produced from synthetic wild-type αsyn display varying degrees of toxicity to cultured cells, with annular oligomeric species, 40 – 45 nm in diameter, reported as the most toxic [326]. Compared to preparations of fibrillar αsyn, purified oligomeric species are significantly more toxic to primary neuronal cultures, associated with the activation of reactive oxygen species (ROS) [327]. Collectively these studies give strong evidence for oligomers being the most toxic αsyn species.

Studies have investigated the mechanism of toxicity by αsyn species. Prefibrillar αsyn oligomers have been shown to cause Golgi apparatus fragmentation and trafficking malfunction [328] and mitochondrial impairment [329] in vitro. The identification of the mitochondrion as a target for αsyn oligomers aligns with a plethora of data linking

27 mitochondria dysfunction to synucleinopathies; in human PD brain deficits are found in mitochondrial complex I activity [330-332] and several that cause familial PD are associated with mitochondria functioning, including PINK1, LRRK2, Parkin and DJ-1 [333]. Additionally, chemicals that induce mitochondria dysfunction are capable of manifesting PD in humans and in animal models in the absence of αsyn deposition [334-336]. Therefore, it is possible that mitochondrial dysfunction is a central mechanism associated with toxicity in PD, and possibly other synucleinopathy disorders.

Another potential site for αsyn oligomer-induced damage is the cellular membrane. It is well known αsyn has a high affinity to bind to lipid species [337] and several in vitro studies suggest pore formation and/or permeabilisation of the membrane causing channel leakage as toxic mechanisms by αsyn oligomer or protofibril species [326, 338]. Likewise αsyn may cause toxicity at the cell membrane by interacting with important solute pumps. In cultured neurons, exogenously added αsyn oligomers can form clusters on the cell membrane and associate with the α3 subunit of Na+/K+-ATPase, causing alterations to the distribution of the receptor and deficits in Na+ gradient and pumping potential [339]. Given that mutations in the gene encoding the α3 subunit of Na+/K+-ATPase are associated with rapid-onset dystonia Parkinsonism [340-342], an adverse interaction with αsyn and the receptor is likely to be relevant to at least a subset of human synucleinopathy conditions.

Although αsyn oligomers have been shown to be toxic to neurons, they may not represent the only toxic αsyn species. Fibrils also form clusters on neuronal membranes and associate with the α3 subunit of Na+/K+-ATPase with higher affinity than oligomers [339]. They are also reported to exhibit the highest degree of toxicity compared to oligomers and ribbons when injected into rat brain [257]; an observation that is supported by several in vitro studies [253, 343]. Therefore, the relevance of oligomeric, and larger αsyn species to toxicity needs further investigation.

PrP oligomers Although difficulties in producing and isolating oligomeric PrP species have hampered studies on toxic PrP-oligomers, it is likely the structure of oligomers are conserved amongst amyloid-forming proteins. Support for this comes from the finding that synthetic antibodies are capable of binding oligomeric species of amyloid-forming protein irrespective of primary sequence including PrP, αsyn, Aβ and islet amyloid polypeptide (iAPP) [344]. Similarly,

28 small molecules that are reported to bind to and inhibit oligomer formation have been reported efficacious to both prion and PD-animal models [345, 346].

Studies using synthetic PrP show strong support for small aggregates being the toxic species. Species reminiscent of oligomers have increased toxicity in vitro [164, 347, 348] and in vivo [164] compared to larger fibrillar species. During PrPC misfolding, it is believed monomeric PrP adopts an α-helical conformation prior to its dimerisation and extension into oligomers [349]. Monomeric recombinant α-helical PrP is highly toxic to murine neuronal but not fibroblast cell cultures and causes toxicity in a dose-dependent manner when injected into mouse brain [350]. These studies are supported by kinetic studies showing oligomeric species that exhibit PrPSc features such as protease resistance, increase with the development of clinical disease in prion-infected mice [351].

Although the evidence for oligomers being the toxic agent is compelling, no common mechanism of action has been identified. This may be a result of the differential endogenous localisation of the proteins and/or different biochemical or morphological properties of the putative toxic species. The latter factor is particularly relevant given that ‘oligomers’ refers to essentially any species larger than a monomer and thus represents a wide range of various sized species. In addition, given that these species exist in equilibrium with each other, distinguishing one species as toxic over another will be challenging. This was reflected by studies showing the reported putative toxic Aβ dimer rapidly aggregates into metastable protofibrils [352] that may be more toxic in nature. An additional pertinent consideration in elucidating the toxic species is the potential for artificial manipulation of fibrillization by laboratory processes or reagents. The popular laboratory solvent SDS is capable of inducing artificial aggregation of Aβ [353] and αsyn [354]; which in the case of αsyn follows an atypical fibrillization pathway. Thus the in vitro generation of oligomers or preparations of tissue are important considerations when deciphering the toxic oligomer.

Other Potential Neurotoxic Agents in Prion Disease

Given that prion disease may be a paradigm for other NDPs, it is important to consider other proposed toxic mechanisms of PrP that may be extrapolated to relatable proteins. The high degree of heterogeneity in the clinical presentation of prion disease suggests the generation of

29 toxicity may not be caused by a single neurotoxic agent; rather multiple overlapping pathways of toxicity may exist to explain all forms of disease. In addition to toxic oligomers, several other candidate neurotoxic agents have been proposed for prion-induced neurotoxicity [355]. While there is evidence some of these proposed mechanisms are also relevant to the toxicity of other NDP-associated proteins, others appear likely to be toxic mechanisms unique to the PrP protein.

The generation of a candidate toxic PrP Species; PrPL Although prion disease can affect a wide range of animals, transmission between species is accompanied by a ‘transmission barrier’ whereby the incubation period is distinctly longer, or does not cause clinical disease, when infecting prions are from a different species. This lag- phase decreases following subsequent passages as it adapts to the host species [356]. These observations, along with the identification of subclinical infection, suggest a dissociation between what is infectious in prion disease (propagating PrPSc) and the toxic agent. An explanation for this phenomenon is that the neurotoxic agent is a conformation of PrP distinct from that of PrPC and PrPSc. Referred to as PrP* [357, 358] or PrPL, denoting the lethal conformer of PrP [359, 360], it is generated either as an intermediate or by-product of the conversion of PrPC to PrPSc (Figure 1.5). According to this hypothesis, subclinical infection would occur when differences in primary sequences of host and infectious PrPSc translate to a slow conversion rate of PrPC to PrPSc, which would mean either maturation of the toxic intermediate PrPL to PrPSc occurs faster than its generation, or the slow catalytic ability of PrPSc to form PrPL from PrPC as a toxic by-product. In subsequent passages however, when infectious PrPSc is host-adapted, PrPL production is increased which facilitates the generation of disease that presents with neurotoxicity and a clinical phenotype.

Kinetic studies further support an uncoupling of infectivity and toxicity. Rises in infectious PrPSc titres during disease does not linearly correlate with the development of clinical symptoms or neurodegeneration [1, 361-363]. This observation was described further in large cohort studies showing prion propagation could be separated into two phases: an initial phase where infectivity exponentially increases, which is followed by a second phase, distinguished by a plateau in the rising PrPSc titres [362, 363]. This plateau is independent of PrPC concentration, given that transgenic mice expressing various levels of PrPC are reported to have equivalent PrPSc titres upon the saturation of infectivity [362, 363]. This is in contrast to

30 the generation of toxicity which is associated with PrPC expression: mice with higher levels of PrP have a symptomatic period that is shorter as a proportion of the overall disease course compared to lower PrP-expressing counterparts [364]. In agreement to this, large cohort studies show the duration of the aforementioned plateau period is inversely correlated with PrPC expression and determines the time until clinical onset [362, 363]. These studies suggest infectivity and toxicity are two different processes with unique characteristics. A logical explanation for this uncoupling is differential PrPL abundance between the infectious and toxic phases.

A caveat to the PrPL hypothesis is that the morphology of the molecule is completely unknown and, given that no PrP isoforms aside from PrPC and PrPSc have ever been reported to occur in vivo, its existence remains purely hypothetical. Still as mentioned previously, not all PrPSc molecules are identical and as such various sized PrP variants are thought to exist in vivo. The identification of disease associated PK-sensitive PrPSc [365] is one such example and indeed is an attractive candidate for PrPL. The production of PK-sensitive PrPSc via propagation increases exponentially from the plateau in infectivity through until terminal disease [363] and certain mutations in PrP bias towards the generation of these species, which translates into reduced survival rates when inoculated into mice [366]. However a definition of what constitutes PK-sensitive PrPSc is lacking and as such various biochemical approaches have been used to separate out PK-sensitive PrPSc from PrPC and whether or not they isolate the same population is yet to be established. Other quantitative measures rely on indirect measurements by subtractive calculations of PrPRes from total PrP, which in order for correctness assumes PrPC levels remain unchanged during disease progression. Consensus for this facet of disease is lacking and while some have reported mRNA levels of PRNP remain unchanged during disease progression [367, 368], others report a decrease in PrPC concentration (which would lead to an underestimation of PK-sensitive PrPSc during infection) [369], or more detrimentally, an increase in PRNP mRNA and/or total PrP levels throughout disease until terminal stages [370-372]. Hence whether or not PK-sensitive PrPSc, or indeed any other putative PrPL population which is distinct from PrPSc, is directly involved in the generation of neurotoxicity remains to be conclusively established.

31

Figure 1.5: The hypothetical generation of PrPL. (A) PrPL may be generated as a toxic intermediate in the conversion of PrPC to PrPSc or (B) as a toxic by-product of prion conversion. Adapted from [360].

32

Altered PrPC signalling Changes to the status quo of PrPC functioning and/or signalling are also thought to underlie toxicity in prion disease. Evidence for this is demonstrated by the lethal toxicity of structural PrP mutants where expressions of transgenic PrP containing large amino-proximal deletions in PrP (such as residues 32-121 and 32-134) on a PrP knockout background are highly toxic in mice, with ataxia and cerebellar granule cell degeneration presenting and death occurring by 3-4 months of age [373]. This observed toxicity is thought to be a consequence of the deletion of the hydrophobic region of the PrP molecule given that total ablation of the hydrophobic region is extremely toxic whilst mutations covering only half the region are innocuous [374]. The hydrophobic region is implicated in the conversion process of PrPC to PrPSc where it undergoes significant structural rearrangement: amino acid substitutions in this region significantly alter the infectivity of PrPSc [375] and antibodies targeting this region bind PrPC but not PrPSc [376]. Further insight comes from the toxicity observed in transgenic mice expressing Doppel (Dpl); a homolog of PrP which shows structural conformation to the globular domain but lack any alignment to the flexible tail [377]. The neurotoxicity of Dpl and the PrP deletion mutations may be overcome by co-expression with PrPC. These findings suggest these toxic mutants may act by altering the normal function of PrPC.

Several mechanisms have been postulated to describe how altered PrPC functioning may contribute to toxicity in prion disease. One possibility is that PrPC has neuroprotective properties that are abrogated by its conversion to PrPSc. Subtle phenotypic changes are observed in PrP knockout mice, which have higher levels of apoptotic cell death following ischemic or traumatic brain damage [378-381]. Primary neuronal cultures from PrP knockout mice are highly sensitive to apoptosis following serum withdrawal, which can be rescued by the expression of PrPC or the anti-apoptotic molecule Bcl-2 [382] and human neurons are resistant to the pro-apoptotic stimuli Bax, when Bax is co-expressed with wild-type PrPC [383]. The ability of PrP to rescue cells from apoptosis or PrP knockout mice from brain injury is ablated when PrP lacks the octapeptide repeat region of the flexible tail [383-385]. Consistent with this, deleting or blocking the function of this region with antibodies slows disease progression in rodent animal models [386, 387]. However, given that the octapeptide repeat region is dispensable for the protective action of PrPC towards both Dpl and toxic PrP mutants [373, 388], a neuroprotective role for PrPC is not considered associated with its ability to suppress toxicity of these molecules.

33

Another major hypothesis in the field is that PrPC itself acts as a receptor for PrPSc, and their interaction triggers the activation of a neurotoxic cascade by PrPC. This hypothesis is supported by reports that antibodies targeting certain regions of the PrP molecule can cause rapid neurodegeneration in brain tissue [389, 390]. One study used a comprehensive panel of monoclonal antibodies targeting various regions of the PrP molecule to show those targeting the globular domain, particularly α-helices 1 and 3 (such as POM1), are toxic in nanomolar concentrations to organotypic brain slice cultures (OBSCs) prepared from PrP overexpressing mice [390]. As with prion disease, the toxicity mediated by POM1 is PrPC dependent (Figure 1.6A-B). These antibodies do not act by distorting the conformation of PrP, and given that various small antibody fragments could achieve equivalent toxicity, including small recombinant single chain Fv fragments, the resulting toxicity is not due to cross-linking of PrPC or complement [390]. Additionally, the toxic signalling pathways elicited by these antibodies overlaps with those triggered by prions: POM1 treated and prion infected OBSCs both induce the expression of reactive oxygen species (ROS) and calpains, and are protected by ROS scavengers and calpain-specific inhibitors [390-392]. Both activate components of the protein kinase RNA (PKR)-like ER kinase (PERK) signalling pathway [392]; a pathway reported as a downstream neurotoxic pathway in prion disease [372] (discussed further in: The Unfolded Protein Response). Similarities are also found at the transcriptional level, with 80% of genes found down-regulated being conserved between the models [392]. Further support for PrPC being the upstream neurotoxic agent was demonstrated in studies showing the toxicity of POM1 is mediated by the flexible tail of PrPC. OBSCs prepared from transgenic mice expressing PrP that lack a flexible tail, PrP(Δ32-93), are resistant to the toxic effects of POM1, which also occurs when wild-type PrP overexpressing brain slices are co- treated with POM1 and antibodies, such as POM2, that target the octapeptide repeat region of the flexible tail [390] (Figure 1.6C-D). The binding of POM2 to PrP appears to disrupt any putative native conformation of the octapeptide repeat region [393]. Similarly, prion infected OBSCs treated with POM2 are protected from neuronal loss [392]. These studies reveal PrPC is capable of adopting a toxic gain-of-function, which is triggered by the flexible tail following stimulus in the globular domain, possibly following attachment to the cellular membrane (Figure 1.6E). This mechanism may also explain the toxicity of the PrP deletion mutants where PrP(Δ94-134) transgenic mice have a small, but significant, increase in median survival time when injected with POM2 compared to non-injected counterparts [390].

34

Figure 1.6: The flexible tail of PrP modulates the toxicity of PrP-directed antibodies in brain slices. (A) The expression of PrPC is essential for toxicity in prion disease in vivo. (B) A similar phenotype can be triggered in wild-type PrP overexpressing brain slices by antibodies, such as POM1, which target the globular domain of PrP. (C) POM1 mediated toxicity requires a functioning flexible tail: transgenic mice expressing amino-truncated PrP, PrP(Δ32-93), are resistant to the effects of POM1 (D) as are wild-type PrP overexpressing brain slices co-treated with POM1 and antibodies targeting the flexible tail, such as POM2. (E) This may mean toxicity is triggered by the flexible tail, possibly by alteration to the native conformation causing attachment to the cell membrane.

35

The capacity of the flexible tail to mediate a toxic signal fails to explain the toxicity of Dpl, or the observation that the toxic effects of Dpl or PrP mutants are abrogated by co-expression with PrPC. Further insight may come from the finding that mice expressing anchorless PrP succumb to a variant of prion disease characterised by a long incubation period and atypical neuropathology where extensive PrPSc accumulation occurs but there is no evidence of spongiform change [281]. Given the overtly differential disease course and pathology in anchorless PrP transgenic mice compared to wild-type PrP counterparts, it is possible different mechanisms of neurodegeneration exist in these mice whereby the generation of ‘acute’ toxicity, per se, requires PrPC to be attached to the cellular membrane. This further supports the idea that numerous pathways may contribute to prion disease-associated neuronal death, which may depend on the spatial and temporal status of PrPC.

Alterations to Endoplasmic Reticulum function Alterations to the homeostatic functioning of the ER may also contribute to prion-disease induced toxicity. The ER is essential for the co-translational or post-translational modification of various secretory and transmembrane proteins. Once targeted to the ER, these nascent proteins are modified and folded by chaperones and folding factors in order to form their proper tertiary structure, after which they are either targeted to the membrane or released from the cell via the secretory pathway. This process is strictly regulated by quality control mechanisms and disruption of this system leads to a state referred to as ‘ER stress’. ER stress can be triggered by various stress stimuli, such as accumulation of misfolded protein, amino acid deprivation and viral replication [394]. Numerous studies suggest alterations to the normal processing of PrP within the ER [395-399] and/or the activation of pathways associated with ER stress [372] can contribute to toxicity in prion disease.

36

Mislocalisation of PrP Aside from PrP that is attached to the cellular membrane via a GPI anchor, several other topological variants of PrP exist. Random errors in processing can lead to the hydrophobic region of PrP adopting a transmembrane conformation with its C- or N-terminus facing the inside of the lumen of the ER, resulting in ctmPrP or ntmPrP respectively [395, 400, 401]. While under normal conditions these topological conformers make up a small percent of the overall pool of PrP and are innocuous, even slight over expression of ctmPrP, or PrP mutations that preferentially produce it, lead to a non-transmissible neurodegenerative phenotype in mice that arises independently of PrPSc accumulation [395, 396]. In addition, PrP may be mislocalised to the cytosol if the degradative capacity of the proteasome is impaired [402, 403], or by stress triggering retrograde transport of PrP from the ER to the cytosol during ER associated degradation (ERAD) [404, 405]: a controlled process to remove unwanted or terminally misfolded protein by targeting them to the cytosol for proteasomal degradation. Increased expression of cytosolic PrP (cytPrP) in both neuronal and non-neuronal cells produces PrPSc-like self-perpetuating, amorphous aggregates that are highly toxic [397-399] and transgenic mice expressing mutant PrP lacking both its C- and N-terminus, which leads to accumulation of PrP in the cytosol, develop neurological illness characterised by substantial loss of granule cerebellar neurons and [398]. It has been reported both ctmPrP and cytPrP may cause toxicity by binding to and disrupting the actions of Mahogunin (MGRN1), a cytosolic E4 ligase that is involved in the ubiquitination and removal of misfolded proteins [406]. In accordance to this, MGRN1 knockout mice exhibit pathological commonalities with prion disease such as spongiform degeneration [407]. However, follow up studies have so far failed to find a role of MGRN1 in prion disease, as neither the up- regulation nor ablation of MGRN1 alters the pathology or course of disease [408, 409].

The mechanisms of ctmPrP or cytPrP production have been investigated. As a general rule increasing the hydrophobicity of PrP increases the production of transmembrane PrP variants [395, 410], which may explain toxicity in certain familial forms of prion disease. Alternatively, PrPSc-induced chronic ER stress has been proposed to trigger cytPrP accumulation via the activation of a pre-emptive quality control (pQC) system that acts to halt translocation of protein and increase retrotranslocation and degradation in the cytosol. If persistently activated, pQC may increase the level of cytPrP beyond the degradative capacity of the proteasome, resulting in accumulation of toxic PrP in the cytosol. In support of this, transgenic mice expressing a PrP variant that has low translocation potential (thus modelling

37 a state of chronic activation of pQC) develop a mild neurodegenerative phenotype similar, in some aspects, with prion disease pathology [411]. In addition, it is feasible that proteasome inhibition caused by PrP oligomers [412, 413] may also lead to the accumulation of cytPrP, and hence provide a mechanism of action for oligomeric-PrP toxicity. However, a role for ctmPrP or cytPrP in prion disease-associated neurotoxicity is questioned by the observation that in vivo models are only able to recapitulate certain aspects of disease. Several studies report the expression of cytPrP is non-toxic, perhaps even protective, in cell culture and animal models [414, 415], hence any effects elicited by these variants in prion disease are not considered to be the sole contributors to toxicity.

The unfolded protein response (UPR) A common feature of ER stress is protein misfolding in the ER, which in order to prevent perturbations in normal cellular activities, leads to the activation of the unfolded protein response (UPR). The UPR is a self-protective mechanism which aims to restore normal ER homeostasis by activating chaperones and enzymes associated with the folding pathway, activating ERAD and suppressing global protein synthesis. Conversely, under conditions of chronic or irreversible stress, the UPR can trigger apoptotic cell death, thus it represents a complex regulatory system that can signal both pro- and anti-cell survival signals. There are three signalling transduction components of the UPR: PERK, inositol-requiring enzyme 1 (IRE1), and activating transcription factor 6 (ATF6) [394]. They are all activated by the heat-shock protein Grp78/BiP, which is reported to be constitutively bound to its signalling molecule in an ‘inactive’ state and dissociates to ‘activate’ its signalling molecule following the detection of rising levels of misfolded protein [416, 417]. Following activation, PERK oligomerises and trans-autophosphorylates the cytosolic portion of its receptor complex, which leads to the activation and phosphorylation of its target protein, the α subunit of factor-2 (eIF2α). Phosphorylated eIF2α (p-eIF2α) triggers a suppression of global protein synthesis and induces the activation of activating transcription factor 4 (ATF4) that enters the nucleus to modulate target UPR genes. ATF4 regulates the functions of CHOP, which is believed to activate apoptotic pathways, and GADD34, which serves to provide a negative feedback loop by dephosphorylating eIF2α. IRE1 also activates via oligomerisation and trans-autophosphorylation, which leads to cleavage and subsequent splicing of mRNA that encodes the transcription factor, x-box binding protein 1 (XBP1). Two splice variants of XBP1 are translated with opposing actions: one that activates UPR

38 target genes, the other which represses them. Activation of the anchored transcription factor, ATF6, results in its translocation to the Golgi apparatus via a CREB/ATF bZIP transcription factor domain, where it is cleaved by proteases to yield a cytosolic fragment known ATF6f (‘f’ for fragment). ATF6f is capable of entering the nucleus where it also activates UPR target (Figure 1.7).

UPR activation by PrP Grp78/BiP is found elevated in human CJD brain [418] and its expression is elevated in cells following exposure to PrPSc [418] or the toxic PrP fragment PrP(106-126) [419]. Downstream components of PERK signalling, p-eIF2α and ATF4, are elevated in prion diseased OBSCs [392]. In prion infected mice, the rising levels of misfolded PrPSc correlates with increases in components of PERK signalling; phosphorylated PERK (p-PERK) and p- eIF2α, as well as decreases in synaptic proteins, leading to neuronal death [372]. In this model the overexpression of GADD35 or the knockdown of PrP was able to suppress PrPSc- induced toxicity, whereas treatment with salubrinal, a compound that inhibits the dephosphorylation of p-eIF2α, exacerbated neurodegenerative symptoms and enhanced disease progression [372]. In agreement to these findings, treatment with drugs targeting the PERK pathway impede the development of clinical symptoms and neuropathology in prion diseased mice [420-422]. Collectively these studies suggest PERK-mediated signalling greatly contributes to prion disease-induced neurotoxicity.

What causes the activation of PERK signalling in prion disease is unknown. The observation that toxic globular domain-directed antibodies activate components of the PERK signalling pathway [392] argues against PrPSc accumulation activating PERK signalling in prion disease. Given that some report PrP mRNA levels increase during prion infection [372] it is possible increased synthesis of native PrP activates the UPR. This is supported by studies showing the overexpression of protein production may trigger UPR markers [423]. However given that others report PrP mRNA levels to be unchanged throughout disease progression [367, 368], further elucidation is necessary to determine the activator/s of UPR signalling in prion disease. In addition, the mechanism of cell death by eIF2α may be multifaceted. Aside from a suppression of protein synthesis, cleaved caspase-12 (mouse homologue to human caspase-4) also correlates with neuronal cell loss in mouse models of prion disease [372, 418], which may indicate the activation of apoptosis. However neither ablation of caspase-12 [424], nor modulation of apoptotic mediators by Bax deletion or Bcl-2 overexpression [425]

39 alters disease progression. Studies examining the effect of PERK signalling on gene expression by knocking down various components of the signalling pathway reveal nearly half of PERK-dependent targets are ATF4 independent [426], hence it cannot be excluded that other unknown downstream effectors also contribute to eIF2α-mediated neuronal cell death in prion disease.

UPR activation by Aβ PERK-mediated neurotoxicity may have an equivalent role in other neurodegenerative diseases. In AD, Grp78/BiP, p-PERK and p-eIF2α are all found elevated in diseased brain [427, 428] and mice overexpressing human ApoE4 [429]. In transgenic AD-model mice expressing the lethal APP/PS1 mutation, the observed decrease in synaptic proteins was found to be concomitant with increased p-eIF2α levels. Strikingly, the conditional deletion of PERK was able to rescue synaptic protein synthesis which improved behavioural spatial memory and suppressed the potentiation of LTP [430], hence giving strong evidence PERK signalling is involved in neurodegeneration in AD. The contributions of p-eIF2α to these effects may be multifaceted given that, in contrast to the reduced translation of protein synthesis elicited by p-eIF2α, it has also been shown to directly increase the translation of BACE1 [431-435]. In agreement to these findings, reductions in PERK-dependent eIF2α phosphorylation suppress any elevation in BACE1 [436].

40

Figure 1.7: The unfolded protein response (UPR). Rising levels of misfolded protein triggers the activation of three signalling components of the UPR in the endoplasmic reticulum (ER). Activation of IRE1 occurs following trans- autophosphorylation of the receptor complex that in turn cleaves mRNA encoding XBP1. PERK also trans-autophosphorylates following activation, which in turn phosphorylates and activates its target eIF2α. The main outputs for peIF2α are a reduction in total protein synthesis and activation of transcription factor ATF4. Activated ATF6 transits to the Golgi apparatus where it is cleaved, releasing ATF6 fragment (ATF6f). XBP1s, AT4 and ATF6f are all capable of entering the nucleus to activate target UPR genes associated with various pro- and anti-survival signals. In particular, ATF4 can induce the expression of GADD34 that generates a negative feedback loop by targeting eIF2α for dephosphorylation.

41

UPR activation by αsyn In PD brain, p-PERK and p-eIF2α are found elevated in the substantia nigra compared to control brain, with p-PERK immunoreactivity co-localising with areas of high αsyn expression [437]. Numerous studies support a role of αsyn in directly activating the PERK pathway. The accumulation of αsyn in dopaminergic cells leads to increases in both Grp78/BiP and ATF4 [438]. Manganese-induced ER stress causes αsyn to oligomerise and activate the PERK pathway in OBSCs [439] and ablation of αsyn in this model leads to a reduction in PERK-mediated signalling, while the levels of pIRE1 and XBP1 mRNA splicing remained unchanged [440]. It has also been proposed monomeric αsyn within the ER activates PERK signalling by binding directly to Grp78/BiP [438]. Cells expressing mutant A53T αsyn harbour elevated levels of p-eIF2α, Grp78/BiP and CHOP and increased cell death relative to wild-type cells, which in contrast to prion disease models, is partially reversed by treatment with salubrinal or inhibition of caspase-12 [441]. This suggests the PERK pathway may have contrasting effects in prion and PD. In support of this, A53T transgenic mice exhibit an attenuation in αsyn-mediated cell death following treatment with salubrinal [442].

Collectively these studies give strong evidence that PERK-mediated signalling is implicated in toxicity associated with various NDPs; however, it seems to have contrasting roles whereby it can act in either a protective or harmful capacity. The observed contrasting effect of salubrinal between prion disease and PD is one such example of this and highlights the complex nature of UPR signalling. Additionally although strong evidence suggests the PERK branch of the UPR is involved in neurodegeneration in mice infected with prions [372], the expressions of p-PERK and p-eIF2α are variable in post mortem brain from patients with prion disease [443, 444] and appears to be weakly activated in mice infected with sCJD or vCJD [444]. Therefore, further investigations are needed to fully establish the role the UPR plays in neurotoxicity in prion disease and relatable NDPs in the human condition.

42

Summary & Aims

A growing body of evidence is finding similar pathogenic mechanisms associated with the misfolding of Aβ, αsyn and the well-studied PrP. Studies show that similar to prions, seeds of Aβ and αsyn are capable of inducing aggregation in normal protein that propagate within the CNS, which in certain cases, causes clinical disease in susceptible animals. The concept of strain variation may also be a conserved feature of these proteins, where differences in structure of misfolded species encode unique disease phenotypes. Current evidence suggests certain αsyn species are more prion-like than others, where in contrast to αsyn derived from PD brain, MSA-associated αsyn is efficient at propagating in vitro and in vivo, causing disease in susceptible transgenic mice [11]. This is an interesting finding given that, similar to prion disease, the clinical progression of MSA is rapid, whereas AD and PD are slower progressing disorders. Although compelling, many prion-like mechanisms of αsyn and Aβ identified are burdened by conflicting findings and/or lack of conclusive data. This, in part, reflects experimental considerations in current studies, such as the ability of transgenic animals to accurately model disease and assess prion-like characteristics of aberrant proteins and the use of misfolded recombinant protein to accurately model biologically-relevant species. Therefore, important future studies should include the use wild-type animals and robust biochemical techniques that investigate the different species of pathogenic protein aggregates, to define those that are most proficient at propagating, seeding and causing toxicity.

An adequate explanation to the numerous data identifying both similarities and differences between aggregation-prone proteins is that some pathogenic mechanisms are protein-specific, while others represent general mechanisms associated with protein misfolding. Neurotoxicity is a good example of this, where while numerous data implicate pathways that are seemingly conserved amongst various NDP-associated proteins (such as oligomers and the activation of ER-stress pathways), other potential toxic agents have found no equivalent and appear intimately linked to the biochemical and biophysical characteristics of the protein. As such, further clarification on the similarities (and differences) between relatable proteins is essential, not only with regards to the prion concept, but also to ascertain features of disease that may be leveraged to further our understanding on a broad range of NDPs.

Aside from PrPSc, the strongest evidence for a NDP-associated protein constituting a prion exists for αsyn. As such this thesis aimed to further discrimate the pathogenic mechanisms of

43

PrP and αsyn misfolding. The hypothesis of this thesis is that relatable pathogenic mechanisms are shared by misfolded PrP and αsyn, which may be exploited to further our understanding on general mechanisms of protein misfolding. To address this hypothesis, I exploited the prion-like features of αsyn to generate aggregated proteins using the prion assay, PMCA; adapting the protocol described by others [275] to allow fibrillization to occur under non-toxic conditions. Upon completion of this aim, the next objective (also Chapter 3) was to identify potential mechanisms of toxicity of PMCA-generated misfolded αsyn. The next aim (Chapter 4) was to develop a model to study prion disease using OBSCs. The final aim (Chapter 5) was to use the systems developed in Chapter 3 and Chapter 4 to determine the relevance of protein misfolding to the generation of the pathogenic readouts developed previously in this thesis.

44

CHAPTER 2: MATERIALS & METHODS

Biosafety Approvals and Animal Ethics

The handling of all compounds used in this thesis conformed to the recommended safety procedures outlined in the chemical’s material safety data sheet. Work was performed in laboratories that had PC2 certification. Studies using animals were carried out in strict accordance of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes issued by the National Health and Medical Research Council. All experiments performed were with the approval of the Animal Ethics Committees of the University of Melbourne (ID: 1312999.1) and La Trobe University (ID: AEC15-34).

Molecular Biology Procedures

The open reading frame of the wild-type and mutant (A53T) human αsyn gene SNCA had previously been cloned into the pcDNA3+ and pRSET B , under the control of a cytomegalovirus and T7 promotor, respectively. Both plasmids contain an ampicillin resistance gene for selection in bacterial systems, while pcDNA3+ also comprises a neomycin resistance gene used to positively select transfected cells in culture. Both vectors were kind gifts from Professor Roberto Cappai (The Department of Pathology, The University of Melbourne). The pcDNA3+ vector was used to produce stable cell lines over expressing wild-type αsyn, while the pRSET B vector was used to produce recombinant protein (wild-type and mutant A53T αsyn) [445].

Transformation of αsyn:pcDNA3+ into competent cells All transformation and amplification procedures followed aseptic technique. The αsyn:pcDNA3+ vector was transformed into MAX Efficiency® DH5α™ strain of Escherichia Coli ( Technologies) by mixing 1 μL of the DNA construct with 100 μL of DH5α cells. The DNA/cell mix was incubated on ice for 30 min, heat shocked at 42oC for 30 sec and placed on ice to recover for 2 min. 250 μL of Luria Broth (LuB; 10 g/L bacterial tryptone (Merck, Kilsyth, VIC, Australia), 5 g/L extract, 5 g/L NaCl, pH 7.5) was then added and the solution incubated further at 37oC for 1 h with agitation using a Thermomixer

45

(Eppendorf). Transformed cells were plated on LuB agar plates (LuB + 15 g/L Agar, Oxoid – Thermo Scientific, Scoresby, VIC, Australia) containing ampicillin (100 μg/mL, Astral Scientific, Gymea, NSW, Australia) and incubated overnight at 37oC to grow colonies.

Amplification and purification of plasmid DNA midipreps Single colonies were picked and amplified in 100 mL LuB broth containing ampicillin (100 μg/mL) overnight at 37oC in a shaker incubator. Extraction of plasmid DNA was then performed using a PureYieldTM Plasmid Midiprep System (Promega) as per the manufacturer’s instructions. Briefly, cell suspensions were pelleted at 5, 000 g for 10 min and resuspended in cell resuspension buffer. Cells were then lysed following the addition of cell lysis solution, mixed and incubated for 3 min, room temperature (RT). Neutralisation buffer was added and waste cellular debris pelleted following centrifugation at 15, 000 g for 15 min, RT. The binding and clearing columns were then assembled and placed into a vacuum manifold. After centrifugation, the plasmid-containing supernatant was purified and collected into the binding column by filling the clearing column with supernatant and vacuum applied until all solution had passed through both columns. The clearing column was removed and discarded and 5 mL of endotoxin removal wash, then column wash solution was passed through the binding column one after the other aided by vacuum. The membrane was then dried, and column placed in a 50 mL Falcon™ tube (Thermo Fisher Scientific) and eluted by centrifugation after addition of 600 μL nuclease-free water and centrifugation at 2,000 g for 5 min, RT using a swinging bucket rotor.

Successful purification was confirmed by running the plasmid on an agarose gel. A 1% (w/v) agarose gel containing DNA stain (1:10 000, SYBR Safe DNA gel stain, Invitrogen) was loaded with 1 μL of plasmid diluted in DNA loading dye alongside the original plasmid for 1 h, 100 V. Bands corresponding to the correct size of the plasmid were visualised under UV using a ImageQuant 350 (GE Healthcare). concentration was determined using a nanodrop (Biospec-nano, Shimadzu Biotech).

46

Methods using Recombinant αsyn

Lyophilisation and preparation of recombinant protein for αsyn-PMCA Recombinant human αsyn was purchased from Monash Protein Production facility, Monash University Clayton VIC. Briefly, protein was produced using wild-type or mutant A53T αsyn:pRSET B transformed into BL12 Gold cells and amplified in LuB broth containing ampicillin. Protein was purified using a Hitrap Q column and dialysed against water with four buffer changes. Protein concentration was measured and diluted to 1 mg/mL. Protein was supplied as 1 mL aliquots. Purchased protein was lyophilised overnight using a benchtop freeze dry system (Labcono, U.S.A) and stored at -80oC until reconstituted in the desired buffer. Buffers used in this thesis were: prion conversion buffer (PCB; 1% (v/v) Triton-X 100, 150 mM NaCl, cOmplete ULTRA protease inhibitor cocktail tablet (1 per 10 mL; Roche)/phosphate buffered saline (PBS; Amresco, Solon, OH, U.S.A)), PBS, PBS+150 mM NaCl (PBSN), tris buffered saline (TBS; Amresco, Solon, OH, U.S.A) and TBS+150 mM NaCl (TBSN). For reconstitution of protein, 10 – 16 mg protein was dissolved in 1 mL buffer and centrifuged at 122, 500 g for 30 min, 4oC (Optima™ MAX Ultracentrifuge, Beckman Coulter) to sediment unwanted debris. Protein concentration of the supernatant was determined by measuring protein absorbance at 280 nm using a photometer (BioPhotometer, Eppendorf) and employing the Beer-Lambert law. Protein was diluted to a final concentration of 90 μM and 60 μL aliquoted into PCR tubes and stored at -80oC until experimental use in αsyn PMCA.

αsyn PMCA Pre-prepared 60 μL αsyn aliquots were taken from -80oC and thawed at RT. Tubes were briefly spun in a benchtop centrifuge (Microfuge®16 centrifuge, Beckman Coulter), 37±3 mg beads (1.0 mm zirconia/silica beads, BioSpec Products) were added and tubes sealed with Parafilm (Pechiney Plastic Packaging Company, Chicago, IL, USA). Samples were placed into a 96-tube rack in a microplate cup-horn sonicator (Misonix 4000) that was filled with water using a circulating water bath to maintain a steady state temperature of 37oC. The rack was adapted in-house to ensure tubes in the holder sat 3 mm above the sonicator plate and the water level in the cup-horn was adjusted to be the level of the bottom of the rack (approximately 1.5 cm). PMCA reactions using αsyn consisted of 20 sec sonications every 29 min 40 sec for varying lengths of time. Power setting 7 was used for all αsyn PMCA

47 experiments. To perform serial rounds of PMCA, PMCA products were diluted 1/10 in PBSN and 5 μL added to fresh PCR tubes containing 60 μL of 90 μM αsyn and PMCA performed as per standard αsyn PMCA conditions.

Measuring kinetics of αsyn fibrillization using ThT fluorescence For studies investigating kinetics of αsyn fibrillization, individual PCR tubes containing 60 μL of 90 μM αsyn were placed into the 96-tube rack of a running sonicator that was continuously cycling. This was done at specified intervals to enable all tubes to be removed at the same time such that their total process time corresponded to the time-point of interest (e.g. to measure αsyn fibrillization at 8 h a tube was placed into a running sonicator 8 h prior to the completion of the cycle). Extent of fibrillization was quantified in PMCA products by reactivity to the fluorescent compound, Thioflavin T (ThT, Sigma Aldrich). Here protein solution was diluted 1/25 in ThT solution (20 μM ThT, 50 mM Glycine in H2O, pH 8.5 with KOH) for a final volume of 250 μL in a well of a 96-well black polystyrene plate (Nunc™, Thermo Fisher Scientific). Fluorescence was measured using a Varioskan plate reader (Thermo Fischer Scientific) with 450 nm excitation and 480 nm emission settings. Triplicate readings were averaged and values determined following subtraction of recordings of ThT solution void of sample to account for background fluorescence. When determining the effect of a compound on PMCA-induced αsyn fibrillization using ThT, separate tubes containing the compound in 60 μL of PBSN only was added to PMCA at the longest (24 or 72 h) time- point or left at -80oC (the 0 h time-point). These served as controls to confirm no ThT signal was attributed to the compound in the presence and absence of PMCA.

PK digestion of PMCA-generated αsyn 10 mM stock PK (Astral Scientific) was diluted in PBSN to 6 x the desired final concentration and 4 μL added to 20 μL of the PMCA product prior to incubation at 37 oC for 30 min with occasional agitation. Replacement of PK with PBSN in equivalent samples acted as non-PK controls and these samples were kept at RT during the digestion of PK-containing samples. Following incubation, all samples were diluted in 4x NuPAGE LDS sample buffer (Life Technologies) and boiled at 100oC for 10 min prior to separation of species using SDS- PAGE.

48

Circular dichroism spectroscopy Conformational changes in αsyn following PMCA were measured using a circular dichroism (CD) spectrometer (Jasco J-815 CD Spectrometer), with a spectral range of 190 - 260 nm and step size of 1 nm. Analysis was performed using a 1 mm-path length cuvette on protein diluted in dH2O to a final concentration of 1 mg/mL. CD readings were taken on αsyn protein exposed to PMCA or left at -80oC (0 h time-point/non-PMCA). The spectrum of buffer alone in an equivalent dilution in dH2O was also taken to confirm that no inherent spectra came from the PBSN solution.

Lipid binding assays Lipid strips dotted with 15 long chain (>diC16) highly pure synthetic lipid analogues were purchased from Echelon Inc. and used to determine the affinity of PMCA-generated misfolded or monomeric (non-PMCA) αsyn to lipid species. Strips were blocked in 5% skim milk (2 h, RT), prior to incubation with 10 μg/mL αsyn protein diluted in blocking buffer for 2 h, RT. Extent of binding was measured using immunoblot by detection of αsyn using MJFR1 antibody (see sections: Immunodetection and Enhanced Chemiluminescence).

Preparation of Compounds Added to αsyn PMCA

Anle138b Anle138b (Sigma-Aldrich) was dissolved in dimethyl sulfoxide (DMSO) to a final concentration of 10 mM and aliquots stored at -80 oC. For αsyn PMCA reactions, stock Anle138b was thawed at RT and diluted in DMSO to 1.3, 13 and 130 μM and 5 μL added to PCR tubes containing 60 μL of 90 μM αsyn for a final concentration of 0.1, 1 and 10 μM Anle138b.

Cardiolipin 500 μg of synthetic 16:0 cardiolipin (Echelon, Inc.) was diluted in a small volume of methanol (MeOH; 50 μL) and solubilised by sonication (Misonix 3000) at 45 oC, 10 min.

49

Lipid/MeOH was diluted in RT PBSN to a final concentration of 1 mg/mL or 0.1 mg/mL and 5 μL added to PCR tubes containing 60 µL of 90 µM αsyn dissolved PBSN.

Cell Culture Procedures

Maintenance of adherent cell lines All cell culture was performed in a class II biosafety cabinet using aseptic technique. The human neuroblastoma cell line SH-SY5Y [446] was used for all experiments on cultured cells in this thesis. This cell line is a subclone of the cell line SK-N-SH originally isolated from a biopsy taken from a 4 year old female with neuroblastoma [447]. SH- SY5Y cells were maintained in Dulbeccos modified eagle medium (DMEM; Life Technologies) supplemented with 10% fetal calf serum (FCS; Life Technologies), 1% GlutaMAX (Life Technologies) and 1% penicillin/streptomycin 100 x (Life Technologies). In this thesis, complete media is the term given to DMEM containing these added components. Culture media was replaced with fresh, prewarmed complete media three times a week. Cells were grown in tissue culture flasks (Nunc™ – Thermo Scientific) in a o humidified environment at 37 C with 5% CO2.

Passaging and seeding of cells Cells were grown to >85% confluency prior to passaging or seeding. Prewarmed complete media was the only solution used for this process. To passage SH-SY5Y cells, monolayers were gently washed once with media, and cells detached by spritzing with media. Cells were then resuspended in media prior to being re-plated into culture flasks at the desired concentration. For seeding known numbers of cells, cells in suspension were counted using an automatic cell counter (Scepter Cell Counter, Merck Millipore), diluted in media to achieve the required concentration and plated into a new tissue culture flask or plate of the desired size.

Freezing and thawing cells To thaw cells, vials stored in liquid nitrogen were removed and warmed briefly on dry ice. Frozen cells were slowly thawed in a 37oC water bath and removed immediately once the last

50 crystals began to thaw. The cell suspension was added to 10 mL of prewarmed complete media then centrifuged for 3 min, 700 g. The supernatant was aspirated and cells resuspended in 10 mL of prewarmed complete media before being transferred to a flask. Media was replaced the next day to remove trace DMSO. For freezing stocks, cells were grown to 95% confluency and cells collected by spritzing. Multiple flasks were usually prepared and cells pooled and centrifuged for 3 min, 700 g. The supernatant was aspirated and cells resuspended in 10 ml of cold 1:1 complete media:freezing media (FM; 40% complete media, 50% FCS, 10% DMSO (v/v)). Cells were centrifuged again and supernatant aspirated prior to resuspension of cells in cold FM. 1 ml of the cell suspension was then quickly transferred to pre-chilled, labelled cryogenic tubes (Greiner Bio-One). These tubes were stored overnight at -80oC before being transferred to liquid nitrogen for storage the following day.

Transfection of SH-SY5Y cells to produce stable cell lines SH-SY5Y cells were seeded in a 12-well tissue culture plate at a density of 1.5 x 105 for use 48 h later. Cells were serum starved in 750 μL FCS-depleted media for 4 h prior to transfection. Varying amounts of plasmid (αsyn:pcDNA3+) and lipofectamine transfection reagent (Thermo Fisher Scientific) were combined separately in 125 μL Opti-MEM (Thermo Fisher Scientific) per well, before being combined, mixed well and left to incubate for 20 min, RT. 250 μL of the combined solution was then added dropwise to each well and o incubated in 37 C, 5% CO2 for 6 h before being replaced with fresh complete media. Cells were left to recover for 48 h before addition of the selective antibiotic G418 (300 μg/mL; Thermo Fisher Scientific) into complete media without added penicillin/streptomycin. Expression of αsyn in clones derived from transfected single cell colonies was monitored by western immunoblot analysis, with the cell line expressing the highest level of αsyn used for further experiments. The αsyn overexpressing SH-SY5Y cell line used in this thesis was derived from clone A4, and hence named as such.

Detection of αsyn in transfected SH-SY5Y cells using immunofluorescence 3 x 104 A4 and untransfected SH-SY5Y cells were seeded into 8-well Lab-Tek®II Chamber Slides (Nalge Nunc, Naperville, IL, U.S.A) and left for 48 h to strongly adhere. Cells were washed once in PBS and fixed with 4% (w/v) paraformaldehyde (PFA; Sigma-Aldrich) in PBS for 15 min, RT. Cells were washed three times in PBS to remove residual PFA and

51 permeabilised in 0.5% (v/v) Triton X-100/PBS for 2 min, RT. Cells were washed again as described above and blocked in buffer (2% (w/v) bovine serum albumin (BSA; Sigma- Aldrich)/PBS) for 30 min, RT. Blocking buffer was removed and cells incubated with primary antibody (MJFR1 to detect αsyn) diluted in blocking buffer for 2 h, RT, followed by five washes in PBS. Fluorphore-conjugated (anti-rabbit, 568-Alexa Fluor conjugated) secondary antibody and 4’,6-diamidino-2-phenylindole (DAPI), also diluted in blocking buffer were then added to the chamber wells and left for 2 h, RT before a final five washes in PBS. Walls of the chamber slide were removed and mounted with glass cover slips (Mediglass, Taren Point, NSW, Australia) with ProLong™ Gold Antifade Mountant (Thermo-Fisher). Slides were inverted and left to dry overnight in the dark before being sealed with nail polish (Sally Hansen).

Lactate dehydrogenase (LDH) cytotoxicity assay A lactate dehydrogenase (LDH) cytotoxicity assay Kit (Thermo Fisher Scientific) was used to determine the cytotoxicity of buffers used in PMCA to A4 and SH-SY5Y cells. The protocol followed was as per the manufacturer’s instructions. Briefly, 1 x 104 cells were seeded into a 96-well tissue culture plate and 24 h later triplicate wells were treated with 10 μL of PBSN or

PCB diluted in H2O to achieve a final concentration of 0.5, 1, 5, or 10% final volume of buffer. Additionally, one set of triplicate wells was treated with H2O to act as a control for spontaneous cell death and one set was left untreated to be treated later with lysis buffer to measure maximum cell death (maximum LDH activity). Cells were incubated under standard cell culture conditions for 24 h. Following incubation, 10 μL lysis buffer was added to the set of untreated triplicate wells and incubated again for a further 45 min. 50 μL of media from all wells was then transferred to a 96-well flat-bottom plate. Reaction mixture (50 μL) was added to each well and incubated for 30 min RT protected from light followed by the addition of 50 μL of stop solution. Absorbance was measured at 490 nm and 680 nm to obtain sample fluorescence and background fluorescence respectively. To determine LDH activity, the 680 nm absorbance was subtracted from the 490 nm value. Percent cytotoxicity of a sample was determined using the formula: % toxicity = (sample LDH activity – spontaneous LDH activity)/ (maximum LDH activity – spontaneous LDH activity)*100

52

Preparation of samples for measuring cellular readouts using Seahorse XF Analyser The Seahorse Extracellular Flux (XF) Analyser (Seahorse Biosciences, North Billerica, MA, U.S.A) used in this project belongs to the Paul Fisher Lab, La Trobe University. The XF assay medium and all drugs used to measure Cellular Respiratory Control (CRC) and glycolysis were made by Dr Sarah Annesley and Ms Katelyn Mroczek (Fisher Lab, La Trobe University), and they also ran the Seahorse instrument.

Confluent SH-SY5Y cells grown in T75 flasks were harvested, pelleted and resuspended in XF assay medium. PMCA-generated misfolded and monomeric (-80oC) αsyn protein or PBSN only (26.25 μL per well) was mixed with lipofectamine (3 μL per well) and the combined solution added to the cells (2 x 105 cells per well) in suspension and mixed gently. Equivalent cell densities were similarly left untreated. Cells and protein (525 μL in total) were then plated into the well of a Seahorse XF24 plate that had been pre-coated (and subsequently air dried) with Matrigel (356231, Corning) diluted 1:2 in XF assay medium. Cells were left to adhere for 1 h at 37oC prior to CRC or glycolysis analysis with Seahorse XF Analyser.

Measuring cellular respiratory control (CRC) in SH-SY5Y cells CRC in live SH-SY5Y cells was measured via changes in the oxygen consumption rate (OCR) following the sequential addition of pharmacological agents (oligomycin, CCCP, rotenone, antimycin A). Between each compound treatment, the average of three measurement cycles of OCR was taken, each cycle including a 3 min mix step, 2 min wait and a measurement time of 3 min. In doing so, several parameters of mitochondrial respiration were obtained: basal respiration, ATP-linked respiration, proton leak respiration, maximum respiration rate and spare respiratory capacity. Each condition tested had a minimum of three replicate wells, with the average of the values taken per experiment.

Measuring glycolysis in SH-SY5Y cells Glycolysis was calculated from changes in pH associated with the extracellular acidification rate (ECAR) in live SH-SY5Y cells. Several parameters on glycolytic potential were measured: basal glycolysis, starved glycolysis, glycolytic capacity and spare glycolytic capacity, all measured following the successive addition of pharmacological agents (glucose, oligomycin, rotenone & antimycin-A, 2-deoxy-D glucose).

53

Lysis of cultured cells All steps to lyse cultured cells were performed on ice using pre-chilled reagents. Pelleted cells were washed twice with cold PBS and resuspended in lysis buffer (150 mM NaCl, 50 mM Tris pH 7.4, 1% (v/v) Triton X-100, 1% (w/v) sodium deoxycholate (Sigma-Aldrich) and cOmplete ULTRA protease inhibitor in ddH2O). Cells were left to lyse for 20 min with occasional agitation, then unwanted debris and nuclear material removed by centrifugation (1, 000 g, 3 min). The supernatant was collected and either used immediately in protein determination and western immunoblot analysis, or stored at -80oC for future use.

Organotypic Culturing of tga20 Brain Slices

All brain slice culture techniques were performed in a class II biosafety cabinet using aseptic technique. In order to enable slices to recover in culture, the total process time of all brain slicing experiments did not exceed 4 h. All dissection tools were sterilised by and all equipment wiped with 80% (v/v) ethanol (EtOH) and sterilised by UV prior to experimental use.

The use of animals in this study Tga20 PrP overexpressing mice on a C57BL/6 background [364] were bred at the Bio21 Animal Facility, The University of Melbourne and La Trobe Animal Facility, La Trobe University. Colonies were maintained with homozygous mating pairings. All mice were housed at a constant temperature of 22oC in a 12 h light-dark cycle, in shoebox cages with wood shaving bedding. Fed and water was supplied ab libitum.

Preparation of buffers for brain slicing Cutting solution and brain slice culture media (SCM) were made in-house and stored at 4oC for a maximum period of two months. All perishable components of buffers were stored as sterile 50 mL aliquots to avoid changes to pH associated with repeated exposure to air upon opening of containers. Buffers used in the sectioning and culturing of brain slices all had physiological pH of 7.2 – 7.4 and were filter sterilized with a 0.22 μm vacuum driven

54

Stericup® Filter Unit (Merck Millipore). The recipies of the cutting solution and the SCM are as follows:

The cutting solution, Grey Balanced Salt Solution (GBSS): 137 mM NaCl, 5 mM KCl, 0.845 mM Na2HPO4, 1.5mM CaCl2.2H2O, 0.66mM KH2PO4, 0.28mM MgSO4.7H2O, 1.0 mM

MgCl2.6H20, 2.7mM NaHCO3 in sterile ddH2O. GBSS was supplemented with 25 mM kynurenic acid (Sigma-Aldrich) and 33 mM glucose (45% (w/v) in H2O, sterilized, Sigma- Aldrich) immediately prior to use in brain sectioning. This complete solution is referred to as GBSSK.

The slice culture media (SCM): 100 mL 2x concentration of minimal essential medium (MEM; Gibco by Life Technologies; bought in powder form and made up as 2x stock concentration), 100 mL basal medium eagle (BME; Gibco by Life Technologies), 100 mL heat inactivated horse serum (Gibco by Life Technologies), 4 mL glutamax, 4 mL penicillin/streptomycin, 5.5 mL glucose (45% (w/v) solution), 86.5 mL ddH2O.

Cerebellum isolation and sectioning using a tissue chopper Tga20 mouse pups (9-10 days of age) were culled via decapitation using large scissors and the head wiped with 80% (v/v) EtOH. Using fine scissors, a long incision was made under the skin from the back of the head laterally down the midline to the forefront of the head, and skin carefully peeled back to reveal the skullcap. Remaining cervical vertebrae at the back of the skull was removed and cuts were made either side of the base of the skull and at the forefront in between the eye sockets. The skullcap was separated down its centre with shallow incisions which avoided cutting too deep and damaging the brain tissue underneath. The two resulting hemispheres of the skullcap were then levered back from the incision line to the sides of the skull revealing the brain. Jeweller’s tweezers were used to pinch and isolate the cerebellum, which was immediately placed in an ice slurry of GBSSK for 2 min (a timer was used). The brain tissue was then placed onto the stage of a tissue chopper (McIlwin Tissue Chopper) on top of two pieces of circular Whatman® filter paper (Sigma-Aldrich) pre-moistened with two drops of GBSSK and cut to the diameter of the stage. An additional two drops of GBSSK was added to the brain tissue upon its placement on the stage to prevent the tissue drying out. Tissue sections 400 µm thick were cut serially, with each slice collected as soon as it was cut using a small paintbrush and transferred immediately to ice-cold GBSSK with the aid of a blunt spatula. Slices were then either transferred directly to

55

Millicell® cell culture inserts (Millipore) for culturing or first exposed to M1000 prions prior to culturing.

Organotypic culturing of brain slices Brain slices were maintained in SCM on 30 mm Millicell® cell culture inserts; hydrophilic PTFE cell culture inserts with pore size of 0.4 μm. The inserts have small plastic supports (approximate height of 1 mm) that allows the insert to sit slightly raised when placed in a well of a tissue culture plate. This setup allows nutrient transfer to occur via capillary action of the media in the outside well through the membrane to the slice, whilst maintaining sufficient oxygen transfer to slice. Brain slices in GBSSK were transferred into the inserts using disposable, sterile plastic Pasteur pipettes that has been cut to produce a blunt-end large enough to draw up and move a whole brain slice without causing damage. Slices were transferred one at a time with a total of 3 – 6 slices plated per insert. Excess GBSSK in the insert was removed using a P1000 pipette and the insert was carefully placed into a well of a 6-well plate containing 1000 μL SCM. Slices were housed in a humidified environment at o 37 C with 5% CO2. Media was changed three times a week, where SCM was removed from the outer well and fresh prewarmed SCM (1000 μL) carefully added. After every media change, the insert was inspected to ensure no large bubbles had formed on the underside of the insert between the membrane and media, which would hinder the transfer of media to the slice.

Measuring viability in live organotypic brain slices Live brain slices were incubated with SCM containing propidium iodide (PI; 5 μg/mL; Life Technologies) and Hoechst 33342 (5 μg/mL; Life Technologies) for 30 min and visualised on an inverted widefield fluorescent microscope (DMI 6000 B, Leica). Slices treated with 5 μM staurosporine (Enzo Life Sciences) 48 h prior to imaging served as a positive control for cell death. Images were taken using 10 and 20 x objective lens at random locations around the slice. Both PI and Hoechst 33342 are non-toxic and hence viable slices could be recovered by washing in ice cold PBS and harvested for further analysis via SDS-PAGE and western immunoblot.

56

Collection and lysis of organotypic brain slices Lysis was performed on pooled organotypic brain slices (usually 6 per treatment group). Tissue was collected by placing the culture plate containing slices on ice and replacing the SCM with cold dPBS. This step was repeated an additional two times to remove residual SCM. dPBS (500 μL) was also carefully pipetted into the insert and removed to wash the surface of the slices. More dPBS (500 μL) was then added into the insert and slices detached from the membrane with the aid of a cell scraper (Corning®). Tissue and dPBS was then collected into Eppendorf tubes and spun for 10 min, 4oC at max speed, and supernatant discarded. An additional 500 μL dPBS was added to the insert and again the cell scraper used to collect any remainder tissue and centrifuged again. Supernatant was again discarded and the remaining pellet either lysed and protein concentration determined, or stored at -80oC for future use. For lysis, the pellet was carefully resuspended in lysis buffer (150 – 300 μL, depending on pellet size) and subjected to three freeze/thaw cycles. For uninfected slices, the lysis buffer was the same detailed in section: Lysis of cultured cells. For infected slices a different protease inhibitor cocktail was used, where the cOmplete ULTRA protease inhibitor cocktail tablet was replaced by cOmplete Mini protease inhibitor cocktail tablet (Roche; 1 per 10 mL). This substitution is due to the ULTRA tablet containing a protease that inhibits the activity of PK.

Protein Biochemistry

Protein Quantification The bicinchonic acid (BCA) assay was used to determine protein content of protein lysates. A standard curve was used in every assay using 25 μL of eight dilutions ranging from 2000

μg/mL to 25 μg/mL BCA in ddH2O, as well as a ddH2O only sample. Lysates were diluted 1/10 in ddH2O and 25 μL of all samples incubated with 200 μL BCA working reagent (50:1 BCA Reagent 1: Reagent 2) at 37oC for 30 min in the dark. Protein concentration of each lysate was determined relative to the linear equation of the standards. Triplicate absorbance readings were taken for each standard and sample, read at 562 nm using a Varioskan plate reader (Thermo Fischer Scientific).

57

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) For cell lysates and biological tissue, lysates were diluted to the desired final concentration (5 – 40 μg) in 1x sample buffer (200 mM Tris pH 6.8, 8% (w/v) SDS, 40% (v/v) glycerol, 0.08% (w/v) bromophenol blue) containing 5% (v/v) β-mercaptoethanol (Sigma-Aldrich). Samples were heated at 100oC for 5 min and centrifuged for 2 min at 16, 000 g prior to being resolved through precast SDS-PAGE 4-12% NuPAGE Novex Bis-Tris 10% Bis-Tris gels (Life Technologies) using an XCell SureLock tank (Life Technologies) filled with 1x NuPAGE MES SDS running buffer (50 mM 2-(N-morpholino)ethanesulfonic acid, 50 mM Tris, 0.1% (w/v) SDS, 1 mM EDTA, pH 7.3; Life Technologies). Electrophoresis was performed at a constant 170 V for 50 min. To monitor the extent of protein separation, a prestained protein marker (SeeBlue Plus2 Prestained Standard, Invitrogen) was resolved alongside the lysates. For recombinant αsyn, samples were diluted 1/20 in 37oC buffer, diluted in 4x LDS Sample Buffer (NuPAGE®, Life Technologies) and boiled at 100oC for 10 min prior to being resolved on SDS-PAGE, as per the conditions above.

Coomassie brilliant blue staining on gel electrophoresis SDS-PAGE gel containing resolved proteins was incubated with Coomassie brilliant blue stain (0.1% (w/v) Coomassie brilliant blue R-250 (Sigma-Aldrich) in 50% MeOH, 15% glacial acetic acid, 35% ddH2O (v/v)) overnight with constant agitation. Proteins were detected the next day after several washes in destain (20% EtOH, 10% glacial acetic acid,

70% dH2O(v/v)). Following destain, the gel was transferred to a white plastic insert and imaged using a developing dock (ChemiDoc™ MP Imaging System, Bio-Rad).

SDS-PAGE western immunoblot analysis using PVDF membrane Proteins were transferred from the separating gel onto 0.45 um polyvinyl difluoride (PVDF) membrane (Millipore) using a Criterion transfer cell (Bio-Rad, USA). The transfer was achieved by cutting four pieces of Whatman 3MM filter paper (GE Healthcare) and one piece of PVDF membrane to the approximate size of the separating gel. The Whatman filter paper was presoaked in western blot transfer buffer (25 mM Tris, 190 mM Glycine, 20% MeOH

(v/v)/dH2O) while the PVDF membrane was activated in MeOH, before also being soaked in transfer buffer prior to transfer cell cassette assembly. The gel, once removed from its cast was trimmed to remove raised edges and also soaked in transfer buffer. The cassette was then

58 assembled in the following order: x2 apparatus pads, x2 Whatman filter paper, gel, PVDF membrane, x2 Whatman filter paper, x1 apparatus pad. Throughout this process the pads and filter paper were continuously gently rolled with a roller to remove any air bubbles. The closed cassette was placed in the gel tank with an ice pack and filled with transfer buffer. Electrophoresis was performed at a constant 100 V for 50 - 60 min in a cold room (constant outside temperature of 4oC).

Immunodetection Membranes containing transferred protein were blocked in 5% (w/v) skim milk (Diploma Instant Skim Milk Powder; Fonterra, Mount Waverly, VIC, Australia) dissolved in PBST

(137 mM NaCl, 2.7 mM KCl, 6.5 mM Na2HPO4, 1.5 mM KH2PO4, 0.1% (v/v) Tween-20 (Ajax Finechem; Thermo Scientific), pH 7.4) at RT for 1.5 h with gentle rocking. The membrane was then probed with primary antibody in 2.5% (w/v) skim milk/PBST. The membrane was then washed four times in PBST for 30 min (5 min x 3, 15 min x 1) then probed with secondary antibody for 1 h at RT in either 2.5% (w/v) skim milk/PBST. The membrane was then further washed four times as previously described, washed once in PBS and developed using Enhanced Chemiluminescence (ECL).

Enhanced chemiluminescence (ECL) To visualise horseradish peroxidase (HRP)-conjugated antibody binding to the membrane, two commercial ECL kits were used; Clarity™ ECL Western Blotting Substrate (Bio-Rad) or SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Fisher). The majority of blots were developed using Clarity™, while SuperSignal™ was used on membranes that required higher sensitivity. For both systems, the two ECL reagents (A and B) were mixed at a ratio of 1:1 and then incubated with the membrane at RT for 5 min, with occasional agitation. Chemiluminescence was captured using a developing dock (ChemiDoc™ MP Imaging System, Bio-Rad). Following ECL capture, membranes were resoaked in PBST and either dried for storage or re-probed. If the second protein of interest was similar in molecular weight or was raised in the same host the membrane was stripped before re-probing. In this instance, 10 mL of stripping buffer (2% (v/v) HCl in dH2O) was incubated with the membrane for 10 min with gentle rocking prior to being washed well in PBST and blocked again in blocking buffer before starting the immunodection assay again.

59

List of antibodies and fluorescent dyes used in this thesis (Table 2.1)

Primary antibodies:

Antibody Type Dilution Company & reference ID αsyn [MJFR1] Rabbit Monoclonal 1/10,000 Abcam (ab138501)

PrP (03R19.2) Rabbit Monoclonal 1/25,000 made in-house [448]

GFAP Rabbit Polyclonal 1/10,000 Abcam (ab7260) PSD-95 Mouse Monoclonal 1/500 Merck (MAB1598)

Synaptophysin Mouse Monoclonal 1/1,000 Cell Signalling (12270)

VAMP2 Rabbit Monoclonal 1/1,000 Cell Signalling (13508) SNAP25 Rabbit Polyclonal 1/1,000 Abcam (AB41455)

β-actin Mouse Monoclonal 1/10,000 Cell Signalling (8H10D10)

GAPDH Mouse Monoclonal 1/50,000 Life Technologies (MA5-15738) All primary antibodies were diluted in 2.5% skim milk. All incubations with membranes were performed overnight at 4oC, except β-actin and GAPDH which were incubated with the membrane for 1 h RT.

Secondary antibodies:

Antibody Dilution Company & reference ID Anti-Rabbit IgG, HRP-linked whole antibody (from donkey) 1/10,000-1/25,000 GE Healthcare (NA934) Anti-Mouse IgG, HRP-linked whole antibody (from sheep) 1/5,000-1/25,000 GE Healthcare (NA931)

Anti- IgG, HRP-linked whole antibody (from rabbit) 1/10,000 Sigma (A5420)

Anti-rabbit IgG H&L Alexa Fluor®568 (from goat) 1/500 Abcam (175471)

Fluorescent dyes: Name Dilution Company & reference ID

Propidium iodide (1 mg/mL solution in water) 1/200 Life Technologies (P3566)

Hoeschst 33342, trihydrochloride, trihydrate (10 mg/mL in water) 1/2,000 Life Technologies (H3570)

DAPI (stored as 1 mg/mL in water) 1/1000 Life Technologies (D1306)

60

Prion Infection Experiments

Safety procedures for handling prions Experimental procedures dealing with infectious prions were performed in a designated containment room under PC2 guidelines. Required personal protective equipment included double gloves, gown, and safety glasses that were worn at all times. All handling of biological material was performed in a cytotoxic class II biosafety hood and waste processed with chemical and/or physical decontamination. Liquid waste containing prions was diluted in NaOH to a final concentration of >2 M and left to soak for a minimum of 12 h. Solid waste was autoclaved at 134oC for 20 min. Tissue culture plates and tips were soaked in NaOH prior to autoclaving as per the conditions outlined above.

Preparation of infectious M1000 brain homogenates Brain homogenate from terminal Balb/C mice infected with M1000 prions [449], supplied by Associate Professor Victoria Lawson (Department of Pathology, The University of Melbourne) was the only source of infectious material used in this thesis. For infection experiments, M1000-infected or normal brain homogenate (NBH; supplied by Associate Professor Victoria Lawson or isolated from mice bred at La Trobe University) was made by weighing whole brain in closed screw cap tubes and adding dPBS for a final concentration of 10% (w/v). Tissue was homogenized by sequential passaging through needles of increasing gauge (18 – 26G). Aliquots were stored at -80oC until experimental use. Homogenate used in PrP PMCA reactions followed the same protocol detailed above except PCB was used as the buffer the tissue was homogenised in.

Infection of tga20 brain slices with M1000 M1000 brain homogenate or NBH was diluted to a final concentration of 1% (v/v) in GBSSK. Freshly cut free floating slices were transferred from GBSSK to GBSSK containing homogenate in a 6-well plate using a blunt ended Pasteur pipette and incubated at 4oC for 1 h. Buffer was removed and slices washed well in GBSSK three times prior to transferring slices to culture inserts as described previously (see section: Organotypic culturing of brain slices).

61

PrP PMCA A standard PrP PMCA was performed by combining NBH (90 μL of 10% (w/v)/PCB) with a small seed of M1000 infected brain homogenate (10 μL of 10% (w/v)/PCB) in PCR tubes. Tubes were sealed with Parafilm and placed in a sonicator as per the procedure for αsyn PMCA. PMCA reactions using PrP consisted of 30 sec sonications every 29 min 30 sec for a total process time of 24 h (24 h equating to one PMCA cycle). Power setting 7 was used for all PrP PMCA experiments. Confirming the relevance of PrPC and PrPSc in the substrate and seed respectively was achieved by running PrP PMCA reactions where the NBH substrate was replaced with PrP knockout brain (supplied by Associate Professor Victoria Lawson or isolated from mice bred at La Trobe University) and the M1000 seed was replaced with NBH.

Amplification of PrPSc in singular M1000-infected brain slices using PrP PMCA Single brain slices were collected using a cell scraper and collected in Eppendorf tubes. Slices were homogenised in 10 μL PCB and lysed by three freeze/thaw cycles. After each thaw, careful pipetting was employed to avoid foaming of the buffer. Samples were added to 90 μL of 10% (w/v) NBH in PCB and subjected to one PrP PMCA cycle. Upon completion of one PrP PMCA cycle, a second serial PMCA cycle was achieved by spiking the PMCA reaction product (10 μL) into fresh PCR tubes containing 90 μL of NBH/PCB, and subjecting the samples to another round of PMCA (referred to as the 2nd PMCA cycle).

Treatment of M1000-infected organotypic brain slices with Anle138b Stock Anle138b (10 mM, for stock preparation, see: Preparation of compounds added to PMCA) was diluted in DMSO to 1 mM or 100 μM and diluted 1:1000 in warm SCM for a final concentration of 1 μM and 0.1 μM respectively. Media containing compound was immediately used on organotypic brain slice cultures and used in every media change (three treatments per week).

PK digestion of M1000-infected brain slices or PrP PMCA reaction products For PK digestion of M1000-infected brain slices, 20 μg of protein was diluted to 20 μL in lysis buffer prior to incubation with 10 μL of PK (60 μg/mL PK diluted in dPBS) for a final

62 concentration of 20 μg/mL PK. To detect de novo protease resistant PrP generated in the PMCA assay reaction products, 20 μL of PMCA sample was incubated with 10 μL of PK (150 μg/mL PK diluted in dPBS) for a final concentration of 50 μg/mL PK.

Digestion was performed for 1 h at 37oC with gentle rocking. For non-PK samples, equivalent amount of protein was combined with 10 μL dPBS and left on ice for the duration of the PK incubation period. PK proteolysis was stopped by addition of phenylmethylsulfonyl fluoride (PMSF; Sigma-Aldrich;) for a final concentration of 5 mM (using a stock of 100 mM PMSF in MeOH). Samples were mixed well, left on ice for 5 min and 11 μL sample buffer containing 5% β-mercaptoethanol added and 20 μL loaded onto a 4-12% Bis-Tris pre-cast gel for SDS-PAGE and western immunoblot analysis.

Data Analysis

Densitometric analysis was performed on unsaturated western immunoblot images using Image Lab. To quantify protein abundance in biological tissue, band densities (expressed as arbitrary units, AU) were normalised to house-keeping genes (β-actin or GAPDH). For quantification of fluorescent PI signal in images taken of live organotypic brain slices, percent area PI staining was determined using FIJI [450], performed on triplicate samples. All statistical analysis was performed using GraphPad Prism, using a statistical criterion of 0.05. When values from at least three independent replicates were combined, they were depicted as mean±standard error of the mean (SEM).

63

CHAPTER 3: IDENTIFYING POTENTIAL NEUROTOXIC MECHANISMS OF MISFOLDED αSYN

Introduction

In synucleinopathies, the generation of neuronal loss and its associated clinical presentation is highly variable [222]. While this diversification is most obvious between the categorical subtypes of synucleinopathies (e.g. PD vs. MSA), variation is also found within a given synucleinopathy. Indeed, in the case of PD, the clinical guidelines used to diagnose disease carry a high degree of error, with a recent systematic review and meta-analysis reporting a pooled diagnostic accuracy of 80.6% in specialised clinics [451]. Similar to prion disease, this broad spectrum of clinical and pathological profiles associated with synucleinopathies supports the idea that numerous mechanisms of toxicity may be employed by misfolded αsyn depending on its structure and/or locality.

The aim of this chapter was to investigate mechanisms of neurotoxicity of misfolded αsyn. A large body of evidence indicates αsyn can directly cause damage to target organelles. Proposed sites where αsyn can cause damage include the cell membrane [326, 338, 339, 452, 453], ER or Golgi apparatus [328, 454, 455] and the mitochondria [329, 456-462], among others. An interesting persistent connection made between αsyn and toxicity is an association with lipids. Indeed, this is considered relevant particularly to αsyn’s ability to cause damage at the lipid-rich cell membrane and mitochondria [463].

The association of αsyn with lipids Several lipid classes have been identified as binding partners for αsyn, which in some cases appear to contribute to the misfolding and/or pathogenicity of the protein. In particular, many studies have demonstrated an interaction of αsyn with polyunsaturated fatty acids (PUFA). Recombinant αsyn harbours an increased propensity to aggregate when exposed to both free forms of PUFA and those esterified with phospholipids [464, 465]. Treating cultured neurons with PUFAs causes an elevation in the formation of αsyn oligomers [466] which go on to form higher order aggregates. This finding is specific to the class of lipid given that a similar effect could not be achieved upon treatment with either monounsaturated or saturated fatty acids [466]. Critically, in this system the formation of PUFA-induced oligomers is also

64 associated with cytotoxicity [467]. These findings in cultured cells are supported by studies using recombinant protein, where the chronic treatment of αsyn with the PUFA, docosahexaenoic acid (DHA), induces α-helical folding in the protein prior to its conversion to fibrillar species [465, 468]. PUFAs may also be regulated by αsyn in disease. Elevated levels of PUFAs are observed in soluble brain fractions in PD and DLB brain [469]. In the αsyn knockout mice, the PUFAs DHA and α-linolenic acid are downregulated [469]. Taken together, these studies suggest that an association of αsyn with PUFAs may contribute to both healthy normal function and disease pathogenesis.

PUFAs are also particularly sensitive to lipid peroxidation, which is a feature of PD [470]. A product of lipid peroxidation, 4-hydroxy-2-nonenal has been implicated in various detrimental processes in disease: it can generate protein adducts within LBs in neurons [471] and alter dopamine transport, which contributes to the PD-associated feature of reduced dopamine levels [472]. Hence peroxidation of PUFAs may directly augment disease pathogenesis. However, a recent study reveals a protective role of PUFAs by chemical modulation of αsyn. In the presence of DHA, αsyn is modified at position His50, forming a covalent adduct [473]. This suggests a role of the protein in sequestering free radicals and hence an association of αsyn with PUFAs may be neuroprotective. While any connection remains not well defined, this finding aligns with other published works that suggest a neuroprotective role of αsyn [131, 137, 474].

In addition to PUFAs, several other lipid classes have been shown to associate with αsyn. αsyn has a greater affinity for synthetic vesicles containing phosphatidylethanolamine (PE) compared to those that are phosphatidylcholine (PC)-rich [475] and several studies show no binding of αsyn to preparations containing solely PC [476-480]. In preparations of recombinant αsyn mixed with synthetic vesicles composed of PE and Phosphatidylserine (PS) (1:1 w/w), the concentration of vesicles dose-dependently increased the abundance of dimeric αsyn species in the pelleted insoluble fraction compared to the lipid-free supernatent [475] and hence this may suggest dimers are the relevant species that interacts with these lipids. Numerous studies show monomeric wild-type or mutant αsyn preferentially binds to vesicles made partially of phosphatidic acid (PA) [476, 481, 482]. Specifically, αsyn has higher affinity for PA than PS [476, 481, 483, 484], and the binding of αsyn to PA-rich membranes stabilizes the secondary structure and increases the α-helix content of the protein [483]. The reasons for these differential binding affinities of αsyn to the various lipid classes are likely multifaceted, however lipid structure is considered to be important. In the case of PS, its

65 bulky head group is thought to sterically interfere with αsyn binding. Competition binding may also an attributing factor in mixed lipid populations; a notion that is supported by the finding that the binding of αsyn to PA may be augmented upon incubation with PE [482].

The monosialotetrahexosyl-gangliosides (GM) is another lipid class shown to associate with αsyn [485-488], however discrepancies exist in which subclass it has the greatest affinity for. Monomeric preparations of αsyn strongly bind to GM1 where it induces α-helical folding in the protein that inhibits its fibrillization [485, 489]. In cultured cells, the action of inhibitors which impede internalisation of αsyn may be reversed upon exposure to GM1 [487] and this system could be prevented upon disruption of lipid raft structures [487]. Hence GM1 is considered to be a vehicle of αsyn internalisation within lipid domains. Others show a stronger interaction of αsyn with GM3, which is able to inhibit channel formation by αsyn [486, 488]; a feature previously ascribed to misfolded αsyn at membranes that cause membrane permeability and toxicity [452, 453].

Cardiolipin (CA) is a lipid which has also been shown to bind to αsyn, with several studies showing preparations of monomeric protein interacts with synthetic vesicles containing CA with higher affinity than those lacking the lipid [490, 491]. CA-only vesicles have the highest affinity for oligomeric αsyn [491] and hence these species may be the most relevant to this interaction. Investigations into the ability of CA-containing vesicles to modulate αsyn fibrillization have reported no effect [273]. Cholesterol (C) may also associate with αsyn, given that a peptide fragment of αsyn (67-78) binds to C and is highly toxic to cultured neurons [492]. Recently, cholesterol has been shown to facilitate the binding of oligomer αsyn with physiologically relevant membranes [478], however its precise mechanism for this is unknown.

It is clear αsyn has a high affinity towards various lipid species. Upon binding, some lipids are able to accelerate folding in αsyn, which in some cases leads to increased pathogenicity of the protein. Many studies investigating such interactions use recombinant monomeric protein, however it is important to note that in the studies that don’t fully characterise the structure of the protein, it is possible that the protein associating with the lipid are actually small oligomers that have formed via spontaneous aggregation in solution. This highlights an important consideration in interpreting which species of αsyn interacts with lipids. Further investigations will be important to define which species of αsyn lipids preferentially interact

66 with within a biological environment and the consequence this interaction has to αsyn-related disorders.

Furthermore, while many lipid classes have been implicated to associate with αsyn, variations exist between the ways these interactions are studied. Most studies investigate protein:lipid interactions using small unilamellar vesicles (SUVs) or planar lipid bilayers. Membrane curvature is a major contributor to the affinity αsyn has to membranes, with enhanced membrane binding being observed in membranes with increased curvature [481, 493-495]. This is presumably due to the smaller size creating an increase in the number of ‘packing defects’ [496-498]; random protein binding sites that arise on the membrane due to the exposure of the hydrophobic acyl chain interior. Because this is likely to induce a degree of artificial error, SUVs may only be an appropriate tool to study potential biological interactions αsyn has with highly curved intracellular membranes. A potential candidate for this type of interaction is an association of αsyn with synaptic vesicles. Studies showing support of this interaction in vivo report the two in close proximity in human brain [242, 499]. However, it is likely this is not the only type of lipid structure αsyn associates with within a cell. In particular, aside from being packed within lipid bilayers, free forms of lipids are found with a cell and hence systems modelling biological membranes may not appropriately reflect or identify all possible interactions αsyn has with lipids within a cellular environment. This is an important consideration when interpreting an association with, or role of lipids to the pathogenicity of αsyn.

Misfolded αsyn in experimental use Studying the pathogenicity of misfolded αsyn requires its production or isolation. Misfolded αsyn used in experiments are derived either from biological systems (animal or human in origin) or produced from bacterially expressed recombinant protein. While the use of αsyn derived from biological systems is preferred, low yields following extraction and the aforementioned concerns associated with the use of SDS to aid their isolation, makes it a challenging task. An additional issue arises in using human-derived αsyn when normal brain tissue is required as a control. Given that synucleinopathies are generally age-related disorders, it can be a surprisingly difficult task to obtain aged-matched control tissue free of aggregated protein (the most common being contamination by aggregated Aβ in preclinical AD brain). Several animal studies have investigated the propagation of αsyn and

67 development of clinical disease induced by inoculation of terminal PD-model mouse brain [255, 262], however these studies utilise substantial quantities of brain tissue to achieve an effect. For these reasons, recombinant protein is the source of most of the fibrillar αsyn in experimental use.

It is well established misfolding may be easily induced in recombinant αsyn by minor changes to its environment, including decreases in pH and heat [95]. As such, fibrillar protein can be easily produced in a laboratory by exposing αsyn in neutral buffer to continuous shaking at an ambient temperature. This method produces large quantities of homogenous fibrillar protein after 5-7 days. Variations in the preparation of αsyn, the buffer it is reconstituted in or the addition of compounds facilitate the production of other defined species (such as oligomers or ribbons), with several studies characterising the properties of well-defined αsyn structural species [253, 257]. While these studies are useful to attribute pathogenic properties to specific conformations of misfolded αsyn, a caveat to these systems is that these techniques only produce one type of misfolded species. These homogenous populations unlikely reflect the numerous conformations found in human disease and hence its ability to model in vivo disease, including any inter-species interactions that occur, is insufficient. Recently the PMCA has been shown to produce misfolded αsyn species of various sizes [275] and hence this assay may represent a useful tool to rapidly produce large quantities of misfolded αsyn that better reflects its human derived equivalents.

Technical aspects of the PrP- and αsyn-PMCA assays Developed to study PrP misfolding [168], the rationale of the PMCA exploits the nucleation and polymerisation properties of PrPSc to accelerate prion replication in an in vitro environment. A typical PMCA involves combining a small seed of infectious material with substrate containing excess PrPC and subjecting this reaction to multiple rounds of sonication and incubation. Here, the sonication step causes fragmentation in the PrPSc fibrils, which increases the nucleation sites for polymerisation to occur during the incubation period via incorporation of PrP into the growing aggregate. As long as PrPC is in excess, the PMCA allows continual exponential amplification of PrPSc (Figure 3.1A). A typical PMCA sees the reaction mixture subjected to sonication for 30 sec, every 29 min 30 sec for a total process time of 24 h. This is referred to as one PMCA cycle.

68

The technical aspects of PrP PMCA are well established with several standardised methods reported in the literature [500, 501]. Because of the presence of essential co-factors naturally present within the CNS of biological , the most common source of substrate is brain homogenate from a mouse that does not harbour prion infection (referred to normal brain homogenate, NBH). While less effective, cell lysate has also been shown to be an adequate substrate for conversion [502]. The extent of misfolding of PrPC to PrPSc in PMCA is usually quantified by PK digestion where the abundance of products resistant to proteolysis (PrPRes species) is compared between a sample subjected to PMCA and an equivalent non- PMCA sample that has instead been stored at -80oC for the duration of the PMCA cycle. Thus, the difference in PrPRes abundance between these samples is attributed to PMCA- induced PrP conversion. When identifying PrPSc in biological samples with PK, an important technical detail to clarify is that while PrPRes is indicative of PrPSc, PrPRes does not reflect the total PrPSc population in a homogenate given that PrPSc species that are sensitive to PK are also known to exist [365].

Several studies have adapted PMCA for its use with αsyn [275, 503, 504]. While minor variations are found between these protocols, it was Herva et al. that showed the effective production of various misfolded αsyn species using PMCA [275]. This protocol details the rapid fibrillization of monomeric recombinant αsyn protein within the equivalent timeframe of one PrP PMCA cycle (24 h) (Figure 3.2A). PMCA-induced fibrillization of αsyn may be achieved using the same buffer, temperature, total process time and sonication strength as PrP methodologies. Modification to the PrP system includes a shortened sonication step, with 20 sec sonications every 29 min 40 sec and the addition of beads to the reaction mixture to aid the disruption of the growing aggregates by sonication (Figure 3.2A). Consistent with the susceptibility of αsyn to spontaneously misfold, an additional major difference is that compared to PrP, the misfolding of αsyn does not require the presence of a seed containing misfolded protein. However the addition of a preformed αsyn seed in the PMCA accelerates fibrillization in monomeric protein [275].

While the main output of a PrP PMCA is to study the extent of misfolding and/or the presence of PrPRes products, the ability to make various sized αsyn species using PMCA presents it as a useful tool to produce αsyn for further analysis. However currently a limitation to this is the buffer which lyophilised protein is reconstituted in, which contains a high concentration of detergent (1% (v/v) Triton X-100). As such, at the present this system

69 does not easily facilitate studies on the pathogenic properties of these species formed in biological in vitro or in vivo systems.

Aims

PMCA has been shown to produce various sized αsyn species. This system presents a novel approach to produce misfolded proteins that are more biologically relevant than other fibrillization methods. Current studies on the pathogenicity of species formed via PMCA are limited to low concentrations of dilute protein due to a high concentration of detergent in the conversion buffer. As such, the first part of this chapter was to adapt the PMCA to produce misfolded αsyn species under non-toxic conditions. The second aim of the chapter was to use PMCA-generated αsyn products to investigate potential mechanisms of αsyn-induced neurotoxicity.

Results

PMCA can induce misfolding of PrPC into PrPSc Before attempting to produce misfolded αsyn using PMCA, the system was first confirmed to be working by showing amplification of PrPRes in a standard PrP PMCA reaction (Figure 3.1A). Here NBH was spiked with a small amount of brain homogenate from a terminal prion infected mouse (INF BH) and subjected to a 24 h PrP PMCA or stored at -80oC (to serve as the non-PMCA control). Successful PMCA-induced conversion of PrPC to PrPSc was confirmed by PK digestion, where the sample exposed to PMCA contained more PrPRes compared to the non-PMCA control (Figure 3.1B). Because the PrPSc molecule contains PK resistant and sensitive regions, its exposure to PK results in partial truncation of the protein. As such, via western immunoblot, PrPRes is distinguished from undigested PrP by running at a slightly lower molecular weight. Because of this, INF BH samples exposed to proteolysis (+PK) and left undigested (-PK) were run alongside the PK digested PMCA reaction products to confirm that they are PrPRes species (Figure 3.1B). These controls are useful to include in

70 all studies identifying PrPRes and thus has been included in all western immunoblots identifying digested PrPRes products in this thesis.

The relevance of PrPC and PrPSc in the substrate and seed, respectively, can be easily determined by substituting the PrPC-rich NBH with homogenate derived from PrP knockout brain or the infectious seed with NBH. The result of these exchanges is a loss in the ability of PrPSc to amplify via misfolding of PrPC or no amplification due to the lack of a seed to catalyse the reaction. These controls were included in the PMCA reaction and are shown in Figure 3.1B. Collectively theses studies confirm the PrP PMCA reaction is working effectively.

PMCA can produce misfolded αsyn species in prion conversion buffer (PCB) Generating misfolded αsyn species using PMCA was first attempted following the protocol outlined by Herva, et al. [275] (Figure 3.2A). The buffer used by Herva et al. to dissolve αsyn is routinely used as the conversion buffer for PrP PMCA reactions and hence hereafter is referred to as the prion conversion buffer (PCB). The extent of fibrillization following PMCA was assessed using western immunoblot, Thioflavin T (ThT) fluorescence and transmission electron microscopy (TEM). Western immunoblot was employed to identify small conformations of misfolded αsyn that may be resolved by SDS-PAGE. The results from the western immunoblot analysis showed samples subjected to PMCA contained a range of various sized species, with monomeric protein observed at 14 kDa and dimer, trimer and tetramer oligomers identified by their increase in molecular weight by intervals of the weight of the monomer (Figure 3.2B). Species greater than a tetramer, separated by the size of one monomeric unit were identified all the way up to the high molecular weight range of the gel. The abundance of these species represented a ‘laddering’ effect, with monomeric species representing the most abundant species, followed by dimers, then trimers, and so on, with each subsequent elevation in order being slightly lower in abundance than the last. Densitometric analysis was used to determine the percent abundance of each defined oligomer in the PMCA sample. This showed the abundance of monomer, dimer, trimer, tetramer and pooled higher molecular weight oligomers larger than a tetramer to represent 41, 15, 11, 7 and 26 percent of the total protein in the lane, respectively (Figure 3.2B). While this may not include any large species that were not resolved by SDS-PAGE, it indicates PMCA can produce an abundant amount of oligomeric species. In comparison, a non-PMCA sample

71 stored at -80oC during the PMCA process was almost entirely composed of monomeric protein (Figure 3.2B). A small amount of immunoreactivity was observed in the high molecular weight range of the -80oC sample, however given that these species could not be separated by SDS-PAGE and the sample was not exposed to PMCA, they are likely to be amorphous aggregates. The distribution of monomeric species to those larger than a monomer in the -80oC sample was 82 and 18 percent, respectively.

TEM was next used to confirm the presence of large mature αsyn fibrils. Consistent with the TEM images reported by Herva et al. [275], PMCA generated αsyn fibrils were found to be rod-like in structure. No aggregates were observed in the -80oC sample (Figure 3.2C).

The compound ThT exhibits fluorescence upon binding to β-sheet structures [505, 506], and hence is universally used to detect amyloid. Upon exposure to 24 h PMCA, αsyn adopted strong ThT reactivity, while the monomeric -80oC sample displayed no fluorescent signal (- 80oC vs PMCA; 1.06±0.298 and 755±37.99 relative fluorescent units; RFU) (Fig 3.2D). Collectively these data confirm PMCA is a useful tool to produce a range of various-sized αsyn species.

72

Figure 3.1: Protein misfolding cyclic amplification (PMCA) of PrP. (A) In a PrP PMCA reaction, a small PrPSc seed (infectious brain homogenate; INF BH) is mixed with substrate containing excess PrPC (normal brain homogenate, NBH). The sample is subjected to intermittent rounds of sonication and incubation allowing for in vitro propagation of PrPSc to occur. One round of PMCA refers to a reaction that consists of 30 sec sonications every 29 min 30 sec for a total process time of 24 h. (B) Validation of the PMCA technique can be shown by enhanced levels of PrPRes in a sample exposed to PMCA compared to a non-PMCA control. Comparable samples where the seed is replaced with NBH or substitution of the PrPC-containing substrate with PrP knockout brain (KO-BH) confirms both PrPSc and PrPC as important contributors for efficient PMCA-induced PrPC misfolding.

73

Figure 3.2: αsyn PMCA. (A) Workflow of αsyn PMCA: Lyophilised recombinant wild-type protein was reconstituted in buffer to 90 µM and 60 µL aliquoted into PCR tubes. Beads (37±3 mg) were added to the tubes, which were sealed with Parafilm and placed in a heated (37oC) water bath connected to a timed sonicator. Sonications were set to occur for 20 sec every 29 min 40 sec for a total process time of 24 h (B-D) Extent of fibrillization in protein exposed to PMCA and non-PMCA controls (-80oC) was assessed by (B) western immunoblot analysis using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123) (C) transmission electron microscopy (representative images) and (D) ThT fluorescence, where the values obtained for each experimental replicate represented the average fluorescence of triplicate wells after subtraction of a blank well to account for background fluorescence (RFU= relative fluorescent units). Data presented as mean±SEM (n=3).

74

Fibrillization of αsyn by PMCA can occur using non-toxic buffers In order to progress to studies on the pathogenicity of PMCA-generated αsyn, it was necessary to isolate these species in a buffer that did not contain detergent. Given the small size and highly dynamic nature of oligomers, buffer exchange or centrifugation were not considered viable options to collect or maintain the integrity of the small, oligomeric species. Instead, the replacement of the conversion buffer, PCB, with a non-toxic analogue was pursued. In order to address this, the extent of PMCA-induced fibrillization of αsyn was assessed following its reconstitution in several buffers assumed to be non-toxic. The following buffers were tested: PBS, PBS+150 mM NaCl (PBSN), TBS and TBS+150 mM NaCl (TBSN). These samples were exposed to 24 h PMCA as per the same conditions detailed in Figure 3.2 and were compared to an equivalent sample of αsyn dissolved in PCB buffer (αsyn:PCB).

Western immunoblot and ThT were used as readouts to compare the extent of αsyn fibrillization in the various buffers. The ThT data revealed all buffers were able to produce species that contained ThT-reactive conformations. Of these, PBS and TBS (which lack the additional 150 mM NaCl) had slightly lower relative fluorescent signals identified using ThT than those buffers containing the higher salt concentration (24 h PMCA of PCB, PBS, PBSN, TBS and TBSN: 755.5±37.99, 433±8.391, 813.5±65.04, 502.8±71.91 and 695.9±69.6 RFU, respectively) (Figure 3.3A). This observation is consistent with studies showing salt concentrations are positively associated with αsyn fibrillization [507]. As expected, no ThT- reactivity was observed in any non-PMCA samples stored at -80oC (Figure 3.3A).

Resolving samples on SDS-PAGE and immunodetection of αsyn gave further information on the αsyn species produced by the buffers. This assay revealed differences in the efficiency of the buffers to induce fibrillization, with PCB clearly being the most proficient buffer for producing oligomeric species (Figure 3.3B). By comparison, weak fibrillization was observed in all non-toxic buffers exposed to PMCA (Figure 3.3B). The degree of fibrillization was variable amongst the non-toxic buffers with PBSN being the most effective non-toxic buffer for αsyn misfolding. No aggregation was observed in the -80oC samples, except for a small amount in the high molecular weight range of αsyn:PCB. This finding is consistent with what was observed in Figure 3.2B.

75

Figure 3.3: PMCA-induced αsyn fibrillization can occur under non-toxic conditions. Lypholised recombinant wild-type protein was reconstituted in PCB, PBS, PBSN, TBS or TBSN and subjected to αsyn PMCA for a total process time of 24 h. Equivalent samples not subjected to PMCA (-80oC) served as the non-PMCA monomeric controls. Extent of fibrillization in PMCA-exposed samples was assessed using (A) ThT fluorescence, where the values obtained for each experimental replicate represented the average fluorescence of triplicate wells after subtraction of a blank well to account for background fluorescence (RFU= relative fluorescent units). Data presented as mean±SEM (n=3), and (B) western immunoblot analysis using αsyn- specific monoclonal antibody MJFR1 (amino acid specificity: 118-123).

76

Characterisation of PMCA-induced misfolding of αsyn dissolved in PBS+150 mM NaCl (PBSN) Given that none of the non-toxic buffers were as effective as PCB in facilitating misfolding of αsyn (Figure 3.3), the protocol was further adapted with the most efficient non-toxic buffer, PBSN. The first modification tested was total process time, in order to determine if PBSN may produce comparable misfolded αsyn content to PCB if exposed to PMCA for a longer period of time.

To this end, an incremental time-course experiment was performed where αsyn:PBSN was subjected to PMCA for a total process time of: 0, 24, 48 and 72 h. These samples were compared to αsyn:PCB exposed to 24 h PMCA. Here, western immunoblot analysis revealed the abundance of misfolded species produced in PBSN continues to amplify beyond a 24 h PMCA process time, with time-dependent elevations in misfolded content observed in samples exposed to PMCA for 48 h and 72 h. Following exposure to PMCA for 72 h, the presence of oligomeric species in the αsyn:PBSN was comparable to 24 h αsyn:PCB (Figure 3.4A). Similar to 24 h αsyn:PCB, 72 h αsyn:PBSN was ThT-reactive (PCB 24 h vs. PBSN 72 h; 755.5±37.99 and 740.3±67.36 RFU) (Figure 3.4B). Collectively these studies confirm 72 h is an appropriate PMCA process time to produce misfolded species in PBSN.

The species formed by αsyn:PBSN were further characterised by TEM and circular dichroism (CD). Using TEM, the morphology fibrillar species produced in PBSN after PMCA for 72 h (Figure 3.4C) was found to be consistent with those produced in PCB (Figure 3.2C and Herva et al. [275]). CD spectroscopy measures the difference between left- and right-handed circularly polarised light that occurs when a molecule contains multiple light-absorbing groups (chiral chromophores). The CD spectrum of a molecule is measured over a range of wavelengths with α and β secondary structures corresponding to specific absorbance profiles. Measuring the structural elements of a molecule in this way allows the tracking of changes in the structure that may be imparted over time. The ordered misfolding of αsyn has been studied using CD, with changes in secondary structure observed upon its aggregation using shaking/incubation [508, 509] and PMCA (using PCB) [275]. Measuring the CD spectra of αsyn produced in PBSN after a 72 h PMCA reaction showed high negative values at 180 – 220 nm and low negative values in the range of 190 – 200 nm, indicative of high β-sheet content and a low amount of disorder, respectively. In both these regions, the CD of monomeric protein not exposed to PMCA (0 h) was reversed, revealing the protein to be largely disordered with little secondary structure. The resulting CD spectra is consistent with

77 the organised rearrangement of αsyn upon misfolding and confirms the misfolding of αsyn in this system is equivalent to traditional techniques of fibrillization [508, 509] and PMCA- induced fibrillization in PCB [275].

Various transition metals have been shown to accelerate αsyn fibrillization. These include copper (Cu) [510], iron (Fe) [510], cobalt (Co) [510], manganese (Mn) [510], and zinc (Zn) [511], among others. Critically, Cu and Fe have been shown to not only accelerate fibrillization, but also produce species that are thinner and more ‘network-like’ than fibrils produced in their absence [512]. Because of this, the concentration of metals within the PBSN buffer was assessed to determine if these metals might be contributing to misfolding in this system. Inductively coupled plasma-mass spectrometry (ICP-MS) was employed to determine abundances of several common elements. Consistent with using a PBS-based buffer, sodium (Na) and phosphorus (P) were the most abundant metals in the samples containing buffer alone (PBSN) and buffer containing 90 µM αsyn (PBSN+αsyn) (Table 3.1). Several metals shown to modulate αsyn fibrillization were recorded as being at or below the detection limit of the instrument: Mn, Fe, Co, Zn (Table 3.1). Importantly, low values were recorded for both Fe and Cu and hence any contribution to αsyn misfolding by relevant metals is negligible.

78

Figure 3.4: Production of misfolded αsyn species in PBSN buffer produces comparable misfolded content to αsyn species generated in prion conversion buffer. Lypholised recombinant wild-type protein was reconstituted in PBSN or PCB and subjected to PMCA for a total process time of 0, 24, 48 or 72 h. Extent of fibrillization in samples was assessed using (A) western immunoblot analysis using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123) and (B) ThT fluorescence, where the values obtained for each experimental replicate represented the average fluorescence of triplicate wells after subtraction of a blank well to account for background fluorescence (RFU= relative fluorescent units). Data presented as mean±SEM (n=3). (C) Fibrils formed in αsyn:PBSN after 72 h PMCA were characterised by transmission electron microscopy (representative images). (D) Secondary structure of 72 h PMCA αsyn:PBSN and non-PMCA monomeric (0 h αsyn:PBSN) protein was determined using circular dichroism spectroscopy (n=1, representative spectra).

79

Table 3.1: Inductively coupled plasma – mass spectrometry (ICP-MS) of PBSN and PBSN containing αsyn (PBSN+αsyn). * depicts values at or below the limit of the instrument

80

PBSN buffer is non-toxic to untransfected neuronal SH-SY5Y and αsyn-overexpressing SH-SY5Y cells A necessary component to confirming PBSN is a suitable buffer to produce misfolded αsyn for this project was to confirm PBSN is non-toxic when applied to cultured cells. To address this, the ability of the buffer alone to cause cell death in cultured cells was measured via the release of lactate dehydrogenase (LDH) into conditioned media. LDH is an enzyme that is a natural component of the tricarboxylic acid (TCA) cycle, and is exclusively located within the cytoplasm of a cell. Its presence outside of its cellular environment (i.e. in cell culture media) is associated with cell death by its release from dying cells that have a compromised cellular membrane [513]. Commercial LDH cell cytotoxicity kits are able to measure the percentage of cell death using indicator probes that react with the enzyme. Here, the degree of toxicity caused by a known additive is calculated as the amount of LDH it causes cells to release, compared to the LDH release from cells treated with a known toxic compound (lysis buffer treated) minus the cell’s spontaneous LDH release (water treated). LDH release after exposure to PCB or PBSN was assessed in untransfected neuroblastoma SH-SY5Y cells or SH-SY5Y cells stably expressing human wild-type αsyn (A4). Prior to LDH measurements, confirmation of elevated αsyn expression in A4 cells, compared to untransfected SH-SY5Y counterparts, came from western immunoblot and immunofluorescence using αsyn–specific antibody MJFR1 (Figure 3.5A). Both cell lines were then treated with PCB or PBSN such that the total volume of culture media was composed of 0.5, 1, 5 or 10% bufffer and LDH measured 24 h later (Figure 3.5B). Results showed a clear dose-dependent increase in toxicity of PCB, with 10% PCB eliciting a near-100% cell death response for both untransfected SH- SY5Y (LDH values of 0.5, 1, 5 and 10% PCB: 40.09±7.72, 60.45±0.96, 84.95±5.46 and 97.29±6.48 % cytotoxicity, respectively) and A4 cells (LDH values of 0.5, 1, 5 and 10% PCB: 42.44±8.06, 56.06±3.39, 75.21±6.54 and 91.21±1.96, respectively). In contrast, PBSN showed negligible toxicity at all concentrations tested for untransfected SH-SY5Y (LDH values of 0.5, 1, 5 and 10% PBSN: -3.10±3.08, -1.87±1.45, -1.05±1.90 and 0.93±1.81 % cytotoxicity, respectively) and A4 cells (LDH values of 0.5, 1, 5 and 10% PBSN: -5.22±4.75, -5.79±4.16, -0.47±2.89 and 5.21±3.26, respectively).

The cytotoxicity measured by LDH assay for both cell lines was confirmed by differential interference contrast (DIC) images of cells exposed to water, lysis buffer, PCB and PBSN (Figure 3.5C). These experiments confirm PBSN is a suitable buffer to produce αsyn species

81 under non-toxic conditions, which greatly improves the capacity of the PMCA to be used as a tool to study the pathogenicity of misfolded αsyn species.

From this point forward, all experiments detailing misfolded αsyn uses species produced in PBSN after exposure to PMCA for 72 h. This is referred to as the ‘PMCA’ sample, while monomeric, non-PMCA protein is represented as the ‘-80oC’ sample.

Assessment on the affinity of misfolded αsyn to bind lipid species The development of a method to produce misfolded αsyn under non-toxic conditions using PMCA allowed the commencement of the second aim of this chapter, which was to identify potential neurotoxic mechanisms of misfolded αsyn. While numerous mechanisms of toxicity have been reported, a common theme unifying several of these mechanisms appears to involve the interaction of misfolded αsyn with lipid species. Therefore, to gain insight on potential mechanisms to pursue, I screened the affinity of misfolded αsyn to interact with various lipids.

This was studied using a hydrophobic membrane that had been dotted with 15 long chain (>diC16) highly pure synthetic lipid analogues and a blank control. This membrane was incubated with misfolded (PMCA) or monomeric (-80oC) αsyn and the degree of protein binding to the lipids assessed via immunoblot using the αsyn-specific antibody MJFR1. Binding affinity was quantified as the percent αsyn binding over the blank control. Results revealed that, compared to the -80oC sample, PMCA αsyn exhibited a general elevation in both lipid-specific and non-specific binding to the membrane (Figure 3.6). After normalisation to the blank control, misfolded αsyn was found to have no affinity to 11 of the lipids (Figure 3.6A-B). A few classes showed weak binding potential, with PC, sphingomyelin (SM) and sulfatide (SF) having a percent binding of lipid normalised to blank of 110, 105 and 109, respectively. By far, αsyn exhibited the strongest affinity to CA, which had an elevation of 142% over the blank control (Fig 3.6A-B). Interestingly, the interaction of αsyn with these lipid species was conformation specific, with no reactivity with any lipid observed after incubation with the -80oC sample (Figure 3.6C). Given that this sample contains monomeric protein and no ordered aggregates, it can be inferred that the protein interacting with lipids in the PMCA sample are ordered misfolded structures.

82

Figure 3.5: PBSN is non-toxic to untransfected SH-SY5Y and αsyn-overexpressing SH-SY5Y cells after 24 h. (A) SH- SY5Y cells were stably transfected with human wild-type αsyn. The single cell colony, A4, was shown to express elevated levels of αsyn compared to untransfected SH-SY5Y cells via western immunoblot and immunofluorescence using αsyn- specific monoclonal antibody MJFR1 (amino acid specificity: 118-123). (B) Toxicity of PCB and PBSN in SH-SY5Y and A4 cells was measured by assessing LDH levels via absorbance. Values were expressed as a percent cytotoxicity calculated as the percent absorbance compared to equivalent cells treated with lysis buffer (maximum cell death). Deduction of water treated samples accounted for spontaneous LDH release. All buffers were measured using triplicate reads. Data presented as mean±SEM (n=3). (C) Following treatment with water, lysis buffer, PBSN or PCB differential interference contrast images of cells were taken. n=1, representative images.

83

Figure 3.6: PMCA-generated αsyn selectively binds to cardiolipin. A hydrophobic membrane strip dotted with 15 different lipids (long-chain >diC16 synthetic analogues) was incubated with PMCA-generated αsyn (A) Binding of αsyn to lipids was assessed via western immunoblot using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123). (B) Extent of binding was quantified as % over blank. (C) Unlike PMCA-generated misfolded αsyn, monomeric αsyn (-80oC) did not show affinity to any lipid class. n=1

84

Development of assays to characterise αsyn misfolding using PMCA The identification of CA as a lipid class with high binding affinity to misfolded αsyn merited further analysis into whether this lipid can modulate αsyn fibrillization. To do this required the development of assays to characterise species formed by PMCA. ThT is routinely used to study kinetics of fibrillization and hence its use may identify changes to PMCA-induced αsyn misfolding associated with the addition of compounds. Additionally it has been shown that species formed by PMCA are more PK resistant [275] which may be a useful assay to detect changes in the population of species formed. To this end, these two assays were developed to facilitate further studies on the interaction between CA and αsyn.

To use ThT-reactivity to measure αsyn fibrillization, αsyn protein was added sequentially to PMCA such that the total time in PMCA was 4, 8, 24 or 72 h. A sample stored at -80oC was used for the 0 h time-point. Measuring ThT fluorescence in these samples confirmed a time- dependent increase in ThT-signal in αsyn protein exposed to PMCA (Figure 3.7A). Mutant A53T αsyn is known to misfold faster than wild-type protein [509, 514-516] and as such was used as a comparative to test in this system. Similar to published findings, mutant A53T αsyn fibrillized faster than wild-type protein. Rises in ThT for A53T αsyn, but not wild-type protein, was observed at 4 h (A53T vs. WT at 4 h; 334.3±77.02 and 1.169±0.715 RFU), while the earliest time-point to detect rises in ThT for wild-type protein was following 8 h of PMCA (Figure 3.7A).

The resistance of PMCA-generated αsyn to protease digestion was determined by exposing PMCA or -80oC samples to 0, 1, 10 or 50 µg/mL PK in PBSN prior to SDS-PAGE and western immunoblot to identify PK-resistant αsyn products. This assay showed that similar to previously published findings [275], PMCA-generated species are more resistant to PK digestion than -80oC αsyn, with immunoreactivity persisting in all PMCA samples in all concentrations of PK tested – up to 50 µg/mL PK. As expected, the abundance of protein was dependent on the concentration of PK. Compared to the PMCA products, -80oC αsyn was almost entirely digested in 1 µg/mL PK, and no protein was found upon treatment with 10 or 50 µg/mL PK (Figure 3.7B). Thus, these assays are suitable to detect any differences imparted on the PMCA-induced fibrillization of αsyn by CA.

85

Figure 3.7: αsyn PMCA-induced rises in ThT are time-dependent allowing kinetic studies on its fibrillization and 72 h PMCA-products adopt partial PK resistance. (A) Wild-type or mutant A53T αsyn was reconstituted in PBSN and subjected to αsyn PMCA for a total process time of 0, 2, 4, 8, 24 or 72 h. Fibrillization was assessed using ThT fluorescence, where the values obtained for each experimental replicate represented the average fluorescence of triplicate wells after subtraction of a blank well to account for background fluorescence (RFU= relative fluorescent units). Data presented as mean±SEM (n=3). (B) The biochemical properties of wild-type αsyn exposed to 72 h PMCA (PMCA) or not (-80oC) was assessed by its resistance to proteolysis following digestion in PK at final concentration of 0, 1, 10 or 50 µg/mL. PK-resistant species were detected by immunoblot using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123). n=1.

86

Cardiolipin accelerates PMCA-induced αsyn misfolding Having established assays to characterise PMCA products and αsyn fibrillization, the next experiment was to determine whether CA could modulate either of these readouts. The CA used for these studies was the same product dotted onto the lipid membrane strip used in Figure 3.6; synthetic 16:0 CA (Echelon, Inc).

After reconstitution in MeOH and PBSN, CA was added to αsyn:PBSN for a final concentration of 77 µg/mL and 7.7 µg/mL. MeOH/PBSN alone was added to tubes with equivalent protein to serve as a control for the effect of the buffer. The ability of CA to modulate αsyn fibrillization was determined by measuring ThT fluoresence in individual tubes exposed to PMCA for a total process time of: 0, 6, 8, 10, 12, 24 or 72 h. Compared to tubes containing αsyn with MeOH/PBSN (buffer only) and 7.7 µg/mL CA, the addition of 77 µg/mL CA accelerated αsyn fibrillization. This was demonstrated by an elevation in ThT first identified after 8 h (MeOH/PBSN, 7.7 and 77 µg/mL CA: 7.67±3.63, 0.833±0.29 and 206.1±101.5 RFU, respectively). This elevation in ThT by αsyn+77 µg/mL CA was sustained, with an increased ThT signal compared to both other αsyn-containing groups found in all subsequent time-points up until 72 h (MeOH/PBSN, 7.7 and 77 µg/mL CA: 537.8±45.59, 487±12.62 and 1048±60.85 RFU, respectively) Importantly neither the lipid or MeOH/PBSN exhibited any florescence signal in the absence of αsyn (Figure 3.8A).

The effect of CA on the misfolding of αsyn was next assessed using PK digestion. PMCA of αsyn was performed in reactions containing CA (as per the conditions outlined above) or buffer for 72 h prior to being exposed to 0, 2 or 10 µg/mL PK. These samples were compared to corresponding tubes that did not undergo PMCA (-80oC). Results showed in the absence of proteolysis or low concentrations of PK (2 µg/mL), no discernible difference in αsyn species could be noted between the treatment groups. However, upon treatment with 10 µg/mL PK a marginal increase in both high and low molecular weight protein was observed in the PMCA sample containing 77 µg/mL CA, compared to all other groups (Figure 3.8B). Densitometric analysis revealed, in the PMCA samples digested with 10 µg/mL PK, the total protein abundance in the sample containing 77 µg/mL CA was 1.38 fold higher than its buffer control (MeOH/PBSN) equivalent. This altered expression appeared to be due to alterations in PK resistant material in the molecular weight range of 90 – 100 kDa and 15 – 20 kDa. On the western immunoblot in Figure 3.8B, these regions are identified by red arrows. Given that elevated reactivity was observed in both the high and low molecular region, these two

87 populations of PK resistant protein likely reflect the large species that were able to resolve further through the gel following PK digestion and their concomitant cleavage products.

Both 2 and 10 µg/mL PK digested all protein in all -80oC groups indicating neither the CA nor buffer interferred with the activity of PK in this experiment.

The effect of PMCA-generated misfolded αsyn on cellular respiratory control (CRC) in SH-SY5Y cells CA was named following its identification in bovine heart muscle [517]. It is a diphosphatidyglycerol lipid that is conical in shape, with a small anionic head group and large hydrophobic tail. Critically, the expression of CA is almost exclusively localised to the inner mitochondrial membrane (IMM), where it is also biosynthesised. Hence, while it is expressed in all cells that undergo mitochondrial respiration, the discovery of CA in cardiac tissue was likely linked to the high mitochondrial density of these cells.

The main biological function of the mitochondria is oxidative phosphorylation; a sequence of reactions that occur in the IMM that generates the majority of the cell’s energy needs in the form of adenosine triphosphate (ATP). A fundamental component of oxidative phosphorylation is the electron transport chain (ETC), a series of four complexes (I, II, III and IV) that accept electrons from energy-rich molecules NADH and FADH2 in the mitochondrial matrix (MM). These donated electrons are shuttled from their entry point at complex I or II to complex III, then to complex IV via a series of redox reactions with each complex that receives the electrons having a greater affinity for them than the complex they are donated from. The electron carriers ubiquinone and cytochrome c aid the transport of electrons from complex I or II to III and III to IV, respectively. From complex IV, the electrons are released back into the MM where they combine with molecular oxygen (O2) and protons (H+), producing water as a by-product. The process of passing of electrons through these complexes results in the release of protons (H+) from complex I, III and IV into the intermembrane space (IMS). Critically, the pumping of protons into the IMS causes a proton gradient referred to as the proton motive force (PMF). This PMF causes H+ to flow back to the MM through ATP Synthase (also called complex V), which in doing so drives the production of ATP from adenosine diphosphate (ADP) and inorganic phosphate (Pi). An image depicting oxidative phosphorylation is shown in Figure 3.9A.

88

Figure 3.8: The effect of cardiolipin on αsyn fibrillization. Cardiolipin (CA) was dissolved in MeOH/PBSN and added to wild-type αsyn for a final concentration of 7.7 and 77 µg/mL CA. MeOH/PBSN alone added to αsyn:PBSN served to control for the effect of the buffer. (A) Samples were subjected to PMCA for a total process time of 0, 6, 8, 10, 12, 24 and 72 h. Tubes containing PBSN only (no αsyn) with buffer or CA exposed to PMCA for 0 and 72 h served to detect any inherent fluorescence caused by CA or the buffer. Extent of fibrillization was assessed by measuring ThT fluorescence, where the values obtained for each experimental replicate represented the average fluorescence of triplicate wells after subtraction of a blank well to account for background fluorescence (RFU= relative fluorescent units). Data presented as mean±SEM (n=3). (B) The biochemical properties of αsyn species formed after 72 h PMCA (PMCA) or not (-80oC) was assessed by its resistance to proteolysis following digestion in PK at final concentration of 0, 2 or 10 µg/mL. PK-resistant species were detected by immunoblot using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123). Areas of differential PK resistance between PMCA-generated species are identified by red arrows. n=1.

89

The finding that misfolded αsyn binds selectively to CA (Figure 3.6) suggests that mitochondria are a target for misfolded αsyn. To this end I sought to assess the effect misfolded αsyn has on mitochondrial respiration. One of the most powerful methods to assess mitochondrial respiration is by measuring cellular respiratory control (CRC). CRC measures changes in respiration over time in live cells following the incremental addition of compounds that selectively inhibit components of oxidative phosphorylation. In this regard, information of the functioning of each of the components of oxidative phosphorylation may be obtained. Unlike standard techniques that quantify products of respiration at a single time- point, the ability of CRC to measure flux changes in the mitochondria make it a powerful tool to identify both overt and subtle alterations to respiration.

The pharmacological chemicals used in CRC measurements generally include an uncoupling agent, and compounds that inhibit ATP-synthase, complex I and complex II. Examples of these inhibitors are CCCP, oligomycin, rotenone and antimycin-A, respectively and were the inhibitors used in all CRC experiments in this thesis. The data obtained from such measurements allows the following parameters to be measured in a single experiment: basal respiration, ATP synthesis, maximum (max) respiration, complex I activity, complex II activity, respiratory control ratio, proton leak, coupling efficiency and non-mitochondrial respiration. The respiration of the individual complexes may be further converted to percent change from basal or max respiration (depending on the readout in question) to give the functional readout of that complex. A schematic of a typical CRC experiment is shown in Figure 3.9, including the formulas used to determine readouts of mitochondrial respiration and a typical flux pattern observed over the course of an experiment (Figure 3.9B).

CRC analysis was performed using the Seahorse Extracellular Flux (XF) Analyser (Agilent Technologies). In this system, respiration is measured via changes in oxygen (referred to as oxygen consumption rate; OCR) in the media in wells containing adhered cells. This is performed using a solid-state sensor probe that detects changes in dissolved oxygen in a portion of the media over time, with multiple readings taken to increase the sensitivity of the assay.

The ability of misfolded αsyn to alter mitochondrial respiration was measured in naïve SH- SY5Y cells. Cells were incubated with misfolded αsyn premixed with lipofectamine and seeded into a Seahorse plate (XFe24). Baseline readings were taken on adhered cells, prior to the addition of inhibitor compounds which were sequentially added in the following order:

90 oligomycin, CCCP, rotenone, antimycin-A. In addition to cells exposed to misfolded αsyn, other treatment groups included cells incubated with monomeric αsyn (-80oC), buffer alone (PBSN) and left untreated (Untreat). Each experimental replicate represented the average of five wells per treatment group.

Results showed cells exposed to misfolded αsyn had an elevated basal OCR compared to both monomeric and buffer-treated cells (Figure 3.10 and Table 3.2). This significant elevation of OCR in the presence of misfolded αsyn was maintained in individual components of mitochondrial respiration; ATP synthesis, max OCR, complex I activity and non-mitochondrial respiration (Figure 3.10 and Table 3.2). The graphical representation of this data is shown in Figure 3.10 and the raw values, SEM and p-values for all significant comparisons identified are shown in Table 3.2. Importantly, no significant difference was found between -80oC, PBSN and untreated groups in any readout. While elevated upon exposure to misfolded αsyn compared to the control groups, no significant difference was found between any groups of complex II (Figure 3.10E). However, this is a typical observation when measuring the respiration of this complex following inhibition of complex I, given the low overall contribution to OCR from complex II. An additional important technical detail to clarify is that given that complex II is measured last in the CRC measurement, its readout refers to the OCR of complex II linked to complex III.

To summarise this data, misfolded αsyn causes broad-spectrum elevations to all components of respiration in mitochondria of SH-SY5Y cells. This hyperactivity was specific to misfolded αsyn and was not observed in cells treated with monomeric protein, buffer alone or left untreated.

91

Figure 3.9: Oxidative phosphorylation and measuring cellular respiratory control (CRC) via changes in oxygen consumption rate (OCR). (A) Mitochondrial respiration occurs via oxidation phosphorylation: a process whereby electrons are passed from energy-rich donors to complexes of the ETC that sit within the inner mitochondrial membrane. Here NADH or FADH2 donate a pair of electrons to complex I and complex II, respectively. A series of redox reactions sees these electrons be passed from their point of entry, to complex III and complex IV, prior to being released back into the mitochondrial matrix where they combine with O2 and H+ forming water as a by-product. The passing of electrons through the ETC causes the release of protons into the intermembrane space, which produces a proton gradient referred to as the PMF. The PMF allows protons to re-enter the mitochondrial matrix via ATP synthase, which drives the production of ATP from ADP and inorganic phosphate (Pi). (B) Parameters of mitochondrial respiration may be obtained in live cells by measuring OCR following the sequential addition of inhibitors of oxidative phosphorylation. These fluxes in OCR correlate to various mitochondrial readouts, including functional readouts derived from formulations of basal parameters. Abbreviations used: ETC; electron transport chain. PMF; proton motive force, IMM: inner mitochondrial membrane, e-: electron, H+: proton, O2: molecular oxygen, ATP: adenosine triphosphate, ADP: adenosine diphosphate, Pi: inorganic phosphate, Q: ubiquinone, Cyt c: cytochrome c

92

Figure 3.10: PMCA-generated misfolded αsyn causes mitochondria to become hyperactive in SH-SY5Y cells. SH-SY5Y cells were incubated with PMCA-generated αsyn, monomeric αsyn (-80oC), buffer alone (PBSN) or left untreated. Media- containing cells were plated into a well of a Seahorse XF24 plate and mitochondria respiration measured in adhered cells using the Seahorse XF Analyser. This was performed by detecting changes in oxygen (referred to as the oxygen consumption rate; OCR) following the addition of pharmacological agents: oligomycin, CCCP, rotenone and antimycin-A. In doing so, the following parameters were measured: (A) basal OCR, (B) ATP synthesis, (C) max OCR, (D) complex I activity, (E) complex II activity and (F) non-mitochondrial respiration. The average of five wells was taken for each sample per experimental replicate. Data presented as mean±SEM (n=7-15). Statistical significance was examined by ANOVA and Tukey’s multiple comparisons test with a statistical criterion of 0.05. * p<0.05, **p<0.01

93

Comparison Values±SEM (pmol/min) p-value Significance PMCA vs. -80oC 163.8±11.82 and 108.9±8.28 0.0049 ** Basal PMCA vs. PBSN 163.8±11.82 and 92.86±6.39 0.0035 **

PMCA vs. -80oC 206±18.65 and 139±12.2 0.0146 * ATP PMCA vs. PBSN 206±18.65 and 113.5±17.68 0.0061 ** Synthesis PMCA vs. Untreat 206±18.65 and 112.6±16.26 0.0032 **

PMCA vs. -80oC 113.1±9.26 and 71.5±5.49 0.0094 ** Max PMCA vs. PBSN 113.1±9.26 and 61.46±7.72 0.0204 *

PMCA vs. -80oC 170±17.69 and 116.3±11.11 0.0446 *

Complex I PMCA vs. PBSN 170±17.69 and 95.2±18.23 0.0213 *

PMCA vs. Untreat 170±17.69 and 96.99±14.73 0.0061 **

PMCA vs. -80oC 117.4±16.37 and 119.4±10.91 0.0227 * Non- mitochondrial PMCA vs. PBSN 117.4±16.37 and 96.65±17.6 0.0094 ** respiration PMCA vs. Untreat 117.4±16.37 and 99.49±15.77 0.0025 ** Table 3.2: Values of significant differences identified in Figure 3.10. Statistical significance was examined by ANOVA and Tukey’s multiple comparisons test with a statistical criterion of 0.05. * p<0.05, **p<0.01

94

Further analysis of the OCR values obtained in Figure 3.10 was performed to determine whether the alterations in respiration identified were associated with concomitant functional deficits. Depending on the measurement, this was interpreted by the normalisation of values to either the baseline measurements (% basal OCR) or the maximum when exposed to CCCP (% max OCR). The following functional readouts were calculated: basal (% max OCR), ATP (% basal), max (% basal), complex I (% max OCR), Non-mitochondrial respiration (% max OCR) and spare capacity.

No difference was seen amongst any experimental group (Figure 3.11), indicating that misfolded αsyn is not modulating the ability of the mitochondria to respire. Therefore, while misfolded αsyn causes mitochondrial hyperactivity, it does not translate to any kind of observable dysfunction in the mitochondria of SH-SY5Y cells.

95

Figure 3.11: PMCA-generated misfolded αsyn do not cause functional deficits in mitochondria of SH-SY5Y cells. SH- SY5Y cells were incubated with PMCA-generated αsyn, monomeric αsyn (-80oC), buffer alone (PBSN) or left untreated. Media-containing cells were plated into a well of a Seahorse XF24 plate pre-coated with Matrigel. The Seahorse XF Analyser measured mitochondria respiration by detecting changes in oxygen (referred to as the oxygen consumption rate; OCR) following the addition of pharmacological agents: oligomycin, CCCP, rotenone and antimycin-A. The functioning of mitochondria was determined as parameters expressed as % change max or basal OCR (depending on the functional readout). Spare capacity was measured as the difference between max and basal OCR. The average of five wells was taken for each sample per experimental replicate. Data presented as mean±SEM (n=7-15) and statistical significance examined by ANOVA and Tukey’s multiple comparisons test with a statistical criterion of 0.05. No statistical significance was found for all experimental comparisons.

96

PMCA-generated misfolded αsyn do not alter glycolysis in SH-SY5Y cells Following on from the observations seen in Figure 3.10, I next explored whether other components of cellular respiration were similarly affected by misfolded αsyn. Glycolysis is an important upstream component to cellular respiration; it produces a small amount of the total ATP synthesised by the cell and critically produces acetyl co-enzyme A, which feeds into the TCA cycle, producing the electron donors FADH2 and NADH that are essential for oxidative phosphorylation. Given the intimate association between glycolysis and oxidative phosphorylation, the glycolytic potential of SH-SY5Y cells was measured upon exposure to misfolded αsyn.

In addition to the probe that measures oxygen consumption, the Seahorse XF Analyser contains a pH probe that measures the extracellular acidification (ECAR) rate of cells. Within a cellular environment, ECAR is largely attributed to glycolysis and similar to the fluxes in cellular respiration, the glycolytic potential of cells may be tested by the sequential addition of compounds: glucose, oligomycin, rotenone & antimycin-A and 2-deoxy-D glucose. Basic parameters are calculated from this, including: starved ECAR, basal ECAR, glycolytic capacity and spare glycolytic capacity. An image depicting glycolysis, the order of compounds used in this experiment and a typical ECAR trajectory are shown in Figure 3.12.

Glycolytic potential of SH-SY5Y cells was measured upon exposure to PMCA-generated misfolded αsyn (PMCA), monomeric αsyn (-80oC), buffer alone (PBSN) and untreated (Untreat) with the same treatment regime as the CRC experiments. Following three independent experiments, no difference was found between any of the experimental groups in all glycolytic parameters (Figure 3.13).

97

Figure 3.12: Glycolysis and and measuring glycolytic potential via changes in extracellular acidification rate (ECAR). (A) Glycolysis is an upstream component to mitochondrial respiration where one molecule of glucose is catalyzed to two molecules of pyruvate. The oxidation of pyruvate forms acetyl coenzyme A (Acetyl CoA); an essential component to the tricarboxylic acid (TCA) cycle. A main function of the TCA cycle is the production of the energy rich NADH and FADH2, which are electron donors in the electron transport chain (ETC) (B) Parameters of glycolysis may be obtained in live cells by measuring extracellular acidification rate (ECAR) following the sequential addition of glucose, inhibitors to the ETC and inhibitors of glycolysis. These fluxes in ECAR correlate to various glycolytic parameters. Abbreviations used in figure; Acetyl CoA: acetyl co-enzyme A, TCA: tricarboxylic acid, ETC: electron transport chain

98

Figure 3.13: PMCA-generated misfolded αsyn do not affect glycolysis in SH-SY5Y cells. SH-SY5Y cells were incubated with PMCA-generated αsyn, monomeric αsyn (-80oC), buffer alone (PBSN) or left untreated. Media-containing cells were plated into a well of a Seahorse XF24 plate pre-coated with Matrigel. The Seahorse XF Analyser measured glycolytic potential by detecting changes in pH (that corresponds to the extracellular acidification rate of a cell; ECAR) in the absence and presence of glucose, and following the addition of pharmacological agents: oligomycin + antimycin-A and 2-deoxy-D glucose. In doing so, the following parameters were measured: basal ECAR (A), starved ECAR (B), glycolytic capacity (C), spare glycolytic capacity (D). The average of five wells was taken for each sample per experimental replicate. Data presented as mean±SEM (n=3) and statistical significance examined by ANOVA and Tukey’s multiple comparisons test with a statistical criterion of 0.05. No statistical significance was found for all experimental comparisons.

99

Discussion

Originally developed to study PrP misfolding, the PMCA assay has been adapted to induce misfolding of αsyn, producing a range of various sized species [275]. It was of interest to use this system to study the pathogenicity of misfolded αsyn, however further analysis on the species formed was limited due to a high concentration of detergent in the conversion buffer. Hence this chapter had two components: 1) to adapt the published protocol of PMCA- induced αsyn misfolding to enable their production to occur under non-toxic conditions and 2) to use this newly reformed system to investigate potential mechanisms of toxicity associated with misfolded αsyn. Replication of the original protocol was first achieved, with species produced via PMCA being characterised using a range of readouts (ThT, TEM and western immunoblot). This confirmed the system as a suitable tool to produce misfolded αsyn. The effectiveness of PMCA-induced αsyn misfolding when reconstituted in non-toxic buffers was next tested, and adaption of the protocol performed with the most efficient non- toxic buffer, PBSN. New parameters were determined to produce comparable misfolded content to the initial buffer, with 72 h being the optimised time to misfold αsyn in PBSN. The ability of PBSN to be well tolerated in neuronal cells was the final assessment in the adaptation of the protocol and allowed further studies into the cellular targets of misfolded αsyn.

The major finding of this chapter was that PMCA-generated αsyn species affects mitochondrial respiration. Mitochondrial dysfunction is a central feature of PD and consistent with the findings of this chapter, numerous studies have implicated αsyn in directly modulating mitochondrial readouts in various animal [456-461] and cell culture models of disease [462, 518, 519]. Hence this work aligns with the concept that the mitochondrion is a central organelle targeted by αsyn.

In SH-SY5Y cells, I showed misfolded αsyn caused elevations in OCR at every complex, however this hyperactivity did not translate to any functional deficit. Interestingly, this finding is in contrast to the general paradigm that αsyn decreases mitochondrial activity. Transgenic mice overexpressing wild-type αsyn have reduced ATP production and associated elevations in ROS and oxidative mitochondrial damage [456, 457, 518] while primary cultures derived from transgenic mice expressing human mutant A53T αsyn have impaired mitochondrial membrane potential and reduced maximum respiration [459]. The expression of A53T exclusively in dopaminergic neurons show reductions in substrate-specific

100 respiration [461]. In addition to overall decreases in mitochondrial respiration, αsyn has been implicated in causing functional deficits; particularly at the level of complex I. Cells expressing wild-type or mutant αsyn have deficits in complex I [461, 462] and comparing respiration in neuronal cells which harbour deficits in specific complexes implicate complex I in misfolded αsyn-induced reductions to respiration [518]. Reductions in complex I activity are also observed in post mortem PD brain [330-332] and compounds that target complex I cause environmental PD in humans [334-336]. Likewise compounds which target this complex are used as inducible models of PD in animals, such as N-methyl-4-phenyl-1,2,3,6- tetrahydropyridine (MPTP) [520, 521], 6-hydroxydopamine [522], paraquat [523-525] and rotenone [336]. Rotenone is so selective that it is routinely used as the complex I inhibitor in CRC experiments (including the CRC experiments in this thesis) [526], and hence its use as a model of PD strongly implicates complex I deficiency in synucleinopathies.

In order to delineate these divergent findings, it is important to consider factors that may contribute to variations in results. A major influence is likely to be the model used, with cell culture systems, transgenic animal models and human disease all harbouring very different cellular environments. Reporting an acute insult to naïve, healthy cells upon treatment with misfolded protein is likely dissimilar to the disease in humans which has been slowly progressing over years of clinical dormancy. Similarly, given that the overexpression of protein in cultured cells does not result in intracellular aggregation of protein, it also is unlike the human condition. In this regard, it could be argued the introduction of misfolded protein into naïve cells is more biologically relevant than a cellular environment where the clearance machinery is adequately removing protein aggregates. Regardless, while each system reports on an effect of αsyn, its relevance to in vivo PD is uncertain. However this also reflects an important point with the data collected on human samples, which report the status of the mitochondria at post mortem and hence terminal stages of disease. Detailing this stage of disease, while relevant, may not be an adequate representation of mitochondrial health for the majority of the disease process. Considering this, it is possible the mitochondria are able to adapt to the insult caused by αsyn for the majority of the disease course, before being overwhelmed at end-stage disease. Such a scenario would align with the findings of this chapter. Importantly, enhanced respiratory activity with no functional deficit has recently been reported to occur in immortalised lymphocytes derived from ante mortem PD patients [527]. This elevation was found to be independent of age, disease duration or disease severity and hence supports the theory that the mitochondria are adaptive in disease. As such, the

101 dogma that mitochondria dysfunction is a persistent, overwhelming feature to PD may ultimately be revised.

Another relevant point to studies reporting on mitochondria function is the analysis method used. Most studies measure respiration by quantification of components of oxidative phosphorylation at a single time-point, often in lysed or fixed cells. While relevant to the sample in question, the mitochondria are highly dynamic and a single snapshot of mitochondrial health provides little information on the respiratory control of the cell. Likewise, measuring these readouts on unviable cells makes the assay highly susceptible to artefacts associated with the processing of the sample. Therefore, systems that measure flux changes in live cells are considered the strongest and most relevant approach to study mitochondria function. This chapter used the Seahorse XF Analyser to measure CRC in live cells exposed to misfolded αsyn and from this a large amount of information was obtained on relevant parameters of respiration. To date, only one other study has measured CRC in live cells exposed to misfolded αsyn [518]. Here, the treatment of αsyn was found to decrease respiration and the use of cells with deficiencies in various complexes was employed to implicate complex I in this effect. While further research is needed to resolve the reasons why this data contrasts with those presented here, it is important to note this study did not find significance in either mitochondrial respiration or complex deficiencies [518]. This is likely associated with high statistical error reported in the published study and therefore should be repeated to confirm the response. It is important to note that in the CRC experiments of this chapter, data were consistent and highly reproducible.

This chapter did not elucidate whether the changes in CRC upon exposure to misfolded αsyn were due to a direct or indirect mechanism. The observation in Figure 3.6 that PMCA- generated αsyn species selectively binds to CA is support for αsyn modulating mitochondria respiration via a direct interaction with this lipid. CA is commonly referred to as the signature phospholipid of the mitochondria. It is predominately expressed in the IMM, which contains more than 40 fold higher levels of CA than the outer mitochondrial membrane (OMM) [528]. Given its locality within the IMM, it is not surprising CA plays an important role in mitochondrial bioenergetics. Within the ETC, CA has been shown to be required for optimal activity of all components of oxidative phosphorylation: complex I [529-531], complex II [532], complex III [529, 533, 534] and IV [535] and ATP synthase [536]. For a number of these complexes, CA is an integral component to the complex with crystallography confirming CA bound tightly to complex III [534] and complex IV [537]. CA is also

102 pertinent to the assembly of the ETC. The complexes of the ETC do not lie in uniform along the IMM and rather, they form higher order structures called respiratory super-complexes [538-540]. The formation of such structures is thought to increase efficiency of electron flow. Dimeric forms of CA have been shown to be important to the activity and stability of the super-complexes [541-543]. Finally CA also plays roles in the functioning of other relevant proteins of oxidative phosphorylation. It is required for optimal activity of the ADP/ATP carrier [544, 545]; a protein responsible for the shuttling of ATP from the MM to the IMS. Crystallography studies confirm a tight interaction with two [546] or three [547] CA molecules bound per one molecule of ADP/ATP. CA also associates with the carrier molecule cytochrome c (cyt c) [548, 549] that transports electrons from complex III to complex IV. Hence the role of CA in mitochondrial respiration is multifaceted. Therefore, consistent with the CRC data reported in this chapter, if an interaction between αsyn and CA occurs in a biological setting, it would likely result in broad spectrum changes to respiration.

Further support for a direct interaction of αsyn with the mitochondria comes from many studies showing mutant or wild-type αsyn associates with isolated mitochondria [519, 550] and mitochondria in vivo [461, 462, 519, 551, 552]. Also, αsyn has been shown to bind to membranes mimicking mitochondrial membranes (with a preference for those containing CA) [490, 491].

A relevant question to the theory of a direct interaction of αsyn with the mitochondria, is how would αsyn get into the IMM to elicit damage? The N-terminus of αsyn contains a cryptic mitochondrial targeting sequence and as such it has been postulated this region aids the translocation of αsyn across the OMM to the IMM [462]. Alternatively the mitochondrial import receptor, translocase of outer membrane 40 (TOM40) has been implicated in the transport of αsyn into the IMM, given that antibodies to the protein block mitochondrial import of αsyn [462]. Further to this, TOM40 was identified in a pull down assay with a synthetic C-terminal αsyn peptide [553]. Finally the presence of pore-like structures on the OMM has been observed in the mitochondria isolated from the substania nigra of rats selectively overexpressing wild-type αsyn in this region [460]. This may aid the diffusion of αsyn aggregates into the IMM and/or contribute to the deficits associated with a direct interaction of αsyn with the mitochondrion.

Given that the PMCA produces a heterogeneous population of αsyn species, this chapter did not distinguish which species of misfolded αsyn is causing the changes to mitochondrial

103 respiration. An attractive candidate is the oligomeric conformations given that these species have been shown to fragment CA-rich membranes [491], however, other species may be relevant. In transgenic mice overexpressing wild-type αsyn, the predominant αsyn species that accumulated in the mitochondria causing deficits was monomeric full-length N- terminally acetylated αsyn [456]. The latter study highlights an important consideration to using recombinant protein to study pathogenic mechanisms of αsyn, which are not exposed to the numerous post-translational modifications endogenous αsyn is susceptible to. In cultured cells expressing mutant A53T αsyn, both the stimulation of ROS production and alterations to mitochondrial respiration highly correlate with levels of phosphorylation at Ser129 [554] and therefore the production of misfolded species using PMCA may not model all pathogenic subtypes of αsyn that present in a cell and cause damage in disease.

The ability of αsyn to associate with lipids has been described previously in this chapter. Studies reporting αsyn:lipid interactions primarily use monomeric αsyn which is interesting given that the lipid affinity assay employed in this chapter (Figure 3.6) showed no detectable binding of monomeric αsyn with any lipid class. This may indicate that while monomeric αsyn associates with many lipids, changes to its affinity and possibility specificity occurs upon misfolding. An additional important factor to this is the type of assay used to study binding. SUVs and planar lipid bilayers are often used to study αsyn:lipid interactions. In this regard, the use of a lipid strip in this chapter may be considered a limitation because it does not model the complex three-dimensional orientation of lipids within membranes. However, as mentioned previously, systems that model membranes are likely only relevant to potential biological interactions αsyn has with that type of membrane. Given that CA is found in many areas of the mitochondria and clearly not just embedded within a membrane, using models of biological membranes may be unsuitable to study the interaction αsyn has with this lipid. Taking this into consideration, the lipid blot assay is arguably the best model to use to detect this interaction.

This chapter also revealed CA is capable of accelerating αsyn fibrillization and producing αsyn species that are more PK resistant. While the modest changes to PK resistance signifies a change in the population of species formed, it does not indicate whether this sift is solely due to an acceleration in fibrillization or whether CA modulates the formation of species. The former consideration is likely given that lipids are natural accelerators of fibrillization to many proteins by acting as a scaffold that binds and concentrates protein for enhanced protein-protein interaction. This is considered relevant to the role of lipids in PrP misfolding

104 and hence may be a unified feature of amyloid-forming proteins. An interaction with lipids has also been reported to trigger folding in monomeric protein, as in the case of the PUFA, DHA [465, 468]. Clarification on the effect of CA in this system may come from structural analysis on the species formed in its presence. Techniques such as analytical ultracentrifugation (AUC) may reveal more detail on the species and this system has previously been used to characterise the conformation of several oligomeric αsyn preparations [327]. Hence it may be a powerful tool to detect changes in αsyn species produced with PMCA. An additional important experiment will be to determine whether the effect of CA is lipid-specific, or whether other lipid classes may induce similar profiles in αsyn. Nominating several species with varying binding affinities from the lipid blot assay (Figure 3.6) will be useful to align data obtained from numerous technical assays presented in this thesis.

Given that this current work does not confirm a direct interaction between misfolded αsyn and mitochondria, it is possible indirect mechanisms are driving the enhanced mitochondrial respiratory activity. As a general rule, enhanced respiration to all complexes without altering function implicates an indirect mechanism. The most likely candidates for this are those that trigger an increase in ATP production. Autophagic flux is a major modulator of cellular respiration and thus the detection and clearance of misfolded αsyn may be a major contributing pathway to the changes in respiration seen here. Another potential indirect mechanism would be via an association of αsyn with mitochondrial-associated ER membranes (MAMs). Recent evidence suggests both mutant and wild-type αsyn localises in MAMs [555, 556]. Here they are capable of disrupting the linkage between the ER and mitochondria [556] and trigger mitochondrial fragmentation [555]. This causes reductions in transport of Ca2+ from the ER to the mitochondria. Ca2+ is an important molecule in the TCA cycle and hence this results in decreases in ATP synthesis [556]. To address these mechanisms, the expression of autophagy markers (e.g. p62, LC3I, LC3II, ATG18) should be inspected, along with conformational studies on the localisation of misfolded αsyn in this model.

The finding that glycolysis was unaffected upon exposure to PMCA-generated αsyn reveals important information on the potential mechanism of action. It suggests that if mitochondrial respiration was enhanced because of increased energy demands following exposure to misfolded αsyn, it was not enough to also drive changes in glycolysis. This is consistent with the fact that no mitochondrial deficits were identified and hence glycolysis was not activated

105 as a compensatory mechanism. It also eliminates changes to glycolysis as a driver to the effects seen. From here, complementary studies will be useful to investigate other processes linked to mitochondria respiration. Thorough assessment of mitochondrial morphology, locality and health should be performed by employing fluorescent compounds (such as MitoTracker® or tetramethylrhodamine, methyl ester) to determine localisation, morphology and permeability of the mitochondrial membrane. TEM would also be suitable for morphological analysis. Likewise, metabolomics will be useful to identify any other components of respiration that may be playing a role in this system. Additionally, assessment on other processes controlled by the mitochondria should be monitored. Apoptosis is tightly regulated by the mitochondria and is triggered by the release of cytochrome c from the IMM. CA has been shown to be essential for this process by stimulating peroxidase activity of cytochrome c, triggering H2O2 dependent peroxidation that in turn oxidises CA [548, 549]; an essential component for the release of cytochrome c from the membrane. Additionally, ROS production is a major feature of PD, which is considered to be driven predominately by mitochondrial dysfunction [557], likely by electron leak at locations where they are passed from complex I or II to complex III with ubiquinone, resulting in their interaction with O2. However elevated ROS is also found in immortalized lymphocytes from ante mortem PD patients that harbour mitochondria with increased respiration and no functional deficit [527]. Therefore, if ROS levels are found to be also raised by misfolded αsyn in association with elevated respiration, it will have major implications on the understanding on the pathogenesis of disease and may suggest that ongoing elevations in OCR, and not deficits in respiration, is a major cause of oxidative damage in neuronal cells in PD.

106

Conclusion

The aim of this chapter was to develop a system to produce a heterogeneous population of misfolded αsyn species under non-toxic conditions and to investigate neurotoxic mechanisms associated with these species. I have developed and thoroughly characterised the generation of αsyn species in the non-toxic buffer PBSN using PMCA. These misfolded species were found to associate with the mitochondrial-specific lipid CA. CA accelerates αsyn fibrillization and alters the formation of αsyn species produced by PMCA. Critically PMCA- generated misfolded αsyn was shown to enhance mitochondrial respiratory activity in neuronal SH-SY5Y cells. This finding is paradigm shifting and may have important implications to our understanding on the pathogenesis of PD.

107

CHAPTER 4: THE DEVELOPMENT OF A MODEL OF PRION DISEASE USING ORGANOTYPIC BRAIN SLICE CULTURES

Introduction

Having developed a system to study pathogenic αsyn in Chapter 3, the aim of this chapter was to likewise do the same for PrP. An important distinction between proposed toxic mechanisms of PrP and αsyn misfolding is that while misfolded αsyn may cause direct damage to target organelles, the toxicity of PrPSc requires the presence of PrPC [166, 294] (discussed in Chapter 1). This suggests that the generation of toxicity in prion disease is intimately linked to the propagation of PrPSc. Because of this, in this chapter a model of prion disease was developed that shows the generation of neurotoxicity in association with propagation of PrPSc.

Common tools to study prion disease Traditional assays to study prion disease are intracerebral inoculation of animals and in vitro cultured cells. In both systems, disease is studied following exposure to an infectious sample that is highly enriched in PrPSc, such as brain homogenate from a terminal infected mouse or lysate from chronically infected immortalised cells.

The choice of model used generally depends on the feature of disease being studied. In vitro studies are robust, experimentally rapid and comparatively much cheaper than in vivo systems. Cultured cells exposed to prions amplify misfolded protein over several passages and therefore they are generally used for studying cellular mechanisms of prion infectivity and propagation of misfolded protein. Limitations are that few cell lines are conducive to prion infection [558] (the factors which contribute to a cell’s susceptibility to prions remains largely obscure) and of those a small number are neuronal in origin. Critically, prion infected cells do not develop neurotoxicity associated with prion propagation. While the reason for this has not been identified, it is considered to be due in part to the dilution effect of of immortalised cells [558]. Furthermore, a monolayer of homogenous cells is not an adequate representation of the multitude of different cell types and intracellular interactions present in a functioning, three-dimensional CNS and as such the chance of artificial or

108 incomplete responses is high. However while cultured cells are not vulnerable to neurotoxicty associated with propagation of prions, they have been shown to develop acute, PrP-dependent toxicity involving the activation of ROS when exposed to prions [559]. Given the known relevance of chronic propagation to PrPSc-induced toxicity in established disease [166, 294], the mechanisms of toxicity are likely different between acute toxicity in cultured cells and that which is associated with propagation in animals (discussed in more detail in Chapter 6). Hence while in vitro systems are useful tools to study certain features of disease, they do not develop the full spectrum of features of advanced disease. This fact should be taken into consideration when choosing the model to study aspects of PrPSc biology.

In vivo prion infection are bona fide models of disease. In these studies, animals are usually exposed to prions via intracerebral injection, however introduction of PrPSc into peripheral sites also manifests complete disease once prions penetrate the BBB. As per disease in humans, prion infected animals experience a preclinical incubation period which is followed by the development of neurological symptoms. While the precise pathological and clinical symptoms that present in prion-infected animals are strain dependent, the manifestations of disease within a strain is highly reproducible. In mice, clinical disease usually presents initially as subtle alterations in burrowing, nesting and grooming that rapidly progresses to large-scale declines in memory based tasks, cognition and co-ordination [560]. The pathology in the brains of prion- infected mice exhibit all hallmarks of disease including extracellular PrPSc deposition, neuronal loss and spongiform change [560]. Consistent with the protein-only hypothesis, the expression level of PrPC negatively correlates to incubation time [364, 370, 561]. Because of this, transgenic animals expressing elevated levels of PrPC are often used to study prion disease pathogenesis within a shortened time-frame, with one of the most widely used mouse lines being the transgenic tga20, which express PrPC twelve fold above wild-type counterparts [364]. Despite this, these experiments still take several months and complete experiments usually require the use of a large number of mice. This makes these experiments labour intensive and expensive compared to in vitro experiments.

109

The importance of using models that develop propagation and toxicity to study prion disease The fact that cell culture in vitro models of prion disease do not develop neurotoxicity has had significant consequences when extrapolating experimental findings using this system to the human condition. This is eloquently described by considering the history of therapeutic use in humans to treat disease. A large number of compounds have been found that exhibit prion-alleviating properties either in vitro or in vivo, with several of these compounds being trialled in humans (Table 4.1). To date, none have proved overwhelmingly successful. The unsuccessful results from human studies are in part due to the difficulties in translation from in vitro systems to the human condition. This was exemplified by the promising in vitro results for the antimalarial and antipsychotic compounds quinacrine and chlorpromazine. Both compounds were found to inhibit PrPSc propagation in cultured cells [562-564] and given their safe efficacy in humans and ability to cross the blood brain barrier (BBB), quinacrine was soon trialled in humans to treat prion disease despite a lack of in vivo evidence or known mechanism of action. Quinacrine was unsuccessful in improving disease parameters either individually [565-567], or in a combined treatment strategy with chlorpromazine [568, 569] (Table 4.1). After human use was started, reports emerged showing quinacrine failed to extend survival in prion-infected mice [570-572] and its treatment is capable of generating drug-resistant prions [572] and can amplify prion replication in certain prion diseases [573]. These failed therapeutic attempts demonstrate that a reduction in PrPSc in vitro is unlikely to be an appropriate readout of efficacy for the human condition and emphasises the importance of modelling neurotoxicity associated with prion propagation to study disease pathogenesis.

The prion field would greatly benefit from the use of models that develop all critical facets of disease that are higher through-put than in vivo systems, whilst being robust and reproducible similar to in vitro techniques. One such system that appears to fit this profile is the use of organotypic brain slices cultures (OBSCs) to model disease.

110

Effect of compound in Human study Compound Outcome Comments animal models design

Anti-viral compound No effect on survival Observational [576, No effect on trialled when disease Amantadine [574, 575] 577] patient survival believed to be of viral origin No effect on No justification given for Flupirtine RCT-DB [578] patient survival its use in humans

Observational [565- No effect on survival or Variable effect on Treatment in humans 567] Quinacrine PrPSc accumulation survival started before animal [570-572, 579] No effect on testing RCT-DB [566] patient survival

Combined Case report [568, No effect on Chlorpromazine 569] patient survival and Quinacrine

Observational [581, Increase patient 582] survival

No effect on Increase survival to RCT-DB [583] patient survival Doxycycline peripheral and CNS prions Currently [580] Preventative (EU underway in FFI clinical trial register: presymptomatic 2010-022233-28) carriers

Observational [588, Variable effects Surgical complications Intraventricular Increase survival to 589] on survival common and toxicity infusion peripheral and CNS prions observed in some human of PPS [571, 584-587] Case report [590- Possible increase trials 593] patient survival

Table 4.1: Human trials performed for therapeutic intervention of prion disease. RCT-DB: Randomized double blind placebo-controlled clinical trial; CNS: central nervous system; FFI: Fatal Familial Insomnia; PPS: Pentosan Polysulfate.

111

Organotypic brain slice cultures (OBSCs) Organotypic culturing is a process whereby tissue, which has been removed from an , continues to develop and function as if it was still part of the intact organism. The process involves maintaining biological tissue ex vivo in such a way that its basic structural and connective organisation of the intact organ is conserved. Organotypic culturing has proved invaluable in studies into how cells, particularly neurons, develop. While the technical aspects of organotypic culturing has undergone many iterations throughout the 20th century, the notion that tissue may survive ex vivo first came from seminal studies by Ross Granville Harrison, who in 1907 showed the survival and growth of frog embryonic nerve tissue when cultured in frog lymphatic fluid [594]. Referred to as the ‘hanging drop method’, his system involved combining the two in droplets on a sterile glass slide. Here the lymphatic fluid provided nutrients to the nerve and protected it from drying out. Once the fluid clotted, the slide containing the drop could be inverted allowing for its visualisation. While only culturing the tissue for a few days, his work on outgrowth significantly contributed to our understanding of nerve cell biology in modern day neuroscience.

The decades proceeding Harrison’s work saw an expansion in research of ex vivo culturing. While numerous researchers were performing similar experiments, the term ‘organotypic’ was first referenced by Reinbold (1954) in a publication on differentiation of the chick embryo eye [595]. This was followed by subsequent studies by others applying a similar technique to heart and intestine tissue [596, 597]. Organotypic culturing on brain tissue was first published in 1970 on the cerebellum [598-600] and following these reports, many studies aimed to optimise growing conditions of CNS tissue. Some of the early methods used a chamber or tissue plate to maintain viable organotypic cultures [601, 602], however an improved method was developed following extensive work led by B.H. Gähwiler and colleagues who established a system to culture brain slices organotypically using the roller drum technique [603-607]. This system adheres thin brain slices on glass coverslips using clotted plasma, which are then rotated slowly in a roller-drum at ambient conditions. These slices thin out over time resulting in a quasi-monolayer of cells. While this progressive thinning facilitates high resolution imaging of the cultures, the lack of three-dimensional integrity is considered a limitation when modelling the in vivo state [603]. Again, this method was later expanded upon by L. Stopinni et al. [608], who cultivated brain tissue on semiporous membranes at the interface between air and culture media. Importantly this system allows adequate oxygen saturation (it had previously been noted submerging the tissue significantly reduces the viability of cultures), whilst capillary motion

112 enables nutrient transfer from an exterior pool of culture media through the membrane to the slice. This model successfully preserves the three-dimensional structure allowing for morphological and biochemical studies [608]. Today, semiporous membranes are the most commonly used system for organotypic culturing.

The numerous iterations in technical details surrounding organotypic culturing has revealed basic parameters required for organotypic culturing: a solid foundation for structural support of the tissue, nutrient-rich culture medium, exposure to and sufficient oxygenation and ambient temperature and humidity [609]. In OBSCs, neurons are electrically active and as such are conducive to multi-electrophysiological recordings and simulations [610-614]. This makes them a powerful tool to study neuronal status in health and disease.

OBSCs have been shown useful to study prion disease: infected brain slices develop all critical features of disease including propagation of PrPSc [391, 615-618], spongiform change [391, 618] and neurotoxicity [391, 617]. Brain slices are conducive to infection from a range of prions strains [391, 616], with brain tissue derived from both transgenic tga20 [391, 615-617] and wild-type [618] mice. Propagation of prions has been reported to be up to six times faster than equivalent in vivo inoculation experiments [616, 617]. The capacity of this system to produce multiple experimental slices from one animal has further benefits including reducing the number of animals required for large-scale experiments and offers the ability to easily have internal controls from the same animal if desired.

Aims

Limitations exist in both in vitro and in vivo systems to study prion disease. While prion- infection in animals generates bona fide disease, these experiments are generally low through-put and expensive compared to in vitro techniques. The development of systems that have higher throughput and model all pertinent features of disease, particularly neurotoxicity, will be beneficial to further our understanding of disease. The aim of this chapter was to develop an ex vivo model of prion replication and neurotoxicity using OBSCs infected with the prion strain M1000.

113

Results

Organotypic culturing of tga20 cerebellar slices The first aim of this chapter was to develop a method to successfully maintain organotypic brain slices in culture for extended periods of time. In anticipation of modelling prion disease upon establishment of a reliable protocol, the tga20 mouse line [364] was used in all experiments using organotypic cultures.

I established a method of successful organotypic culturing using the cerebellum of 9-10 day old tga20 pups. This region has routinely been used in prion-infected OBSC experiments [391, 392, 615-618]. It is an appropriate brain region to use to model neurotoxicity given that over 75% of CNS neurons are located within the cerebellum [619, 620] and compared to other brain compartments, the distribution of neurons in the cerebellum is highly homogeneous [621, 622].

The developed protocol included three key steps; isolation of the brain, sectioning using a tissue chopper and culturing. For the isolation step, the cerebellum was rapidly isolated and cooled in a cutting solution of neutral buffer containing kynurenic acid; an anti-excitotoxic molecule that likely acts as an antagonist of excitatory receptors [623]. To section the tissue, the cerebellum was mounted onto the stage of a tissue chopper and 400 μm thick sections made. The culturing/incubation step involved transferring multiple slices onto the semi- porous membrane of specialised cell culture inserts which were placed into standard culture plates containing specialised brain slice media (Figure 4.1A).

o Slices were cultured at 37 C with 5% CO2 for up to 70 days, with media being replaced three times a week. During this time, the morphology of the slices changed significantly, most notable was a progressive spreading out of the tissue and concomitant reduction in the thickness of the tissue over the incubation period. Representative images revealing the morphological changes in OBSCs I cultured are shown in Figure 4.1B.

114

Figure 4.1: Workflow of preparation and maintenance of organotypic brain slice cultures (OBSCs). (A) There are three main components to the production of organotypic brain slices: tissue isolation, sectioning and culture/incubation. Harvesting tissue involved removing the cerebellum from a 9-10 day old tga20 pup, snap cooling it in cutting solution and placement onto the stage of a tissue chopper. The tissue was then sectioned and transferred to specialised inserts that are placed in tissue culture plates that contain culture media. (B) Slices were cultured for up to 70 days, during which time the slices underwent morphological change visualised in (B).

115

A major component to confirming brain slices are organotypic was to show that they are viable in culture. For this, viability was measured using the fluorescent dyes propidium iodide (PI) and Hoechst 33342. PI binds DNA and RNA and is non-cell membrane permeable and Hoechst 33342 also binds both DNA and RNA but is cell membrane permeable and hence these dyes detect dead and total cells, respectively. Here Hoechst 33342 was used to confirm that PI images were taken in cell-dense areas, while PI was used for the quantification whereby viability was measured as the difference in PI signal between naïve, untreated brain slices and slices that had been treated with the apoptosis-inducing agent, staurosporine for 48 h prior to the experiment. In this regard, low PI values (percent area fluorescence) in OBSCs compared to high values of a staurosporine-treated slice indicates viability. In these experiments, determining positive fluorescence in the staurosporine-treated slice is critical to confirm the live slices as organotypic because a non-viable slice treated with staurosporine will be unable to achieve any fluorescence upon incubation with PI.

Viability was assessed in three slices per treatment group at 1, 7 and 70 days post sectioning (DPS; Figure 4.2). The percent area PI staining was high in slices at 1 DPS (55.89±9.85 percent area PI). This is a common observation with OBSCs considered to be due to the stress induced by the cutting process. Given that this time-point is shorter than the required time necessary to induce apoptosis with staurosporine, quantification of viability cannot be performed. Viability measurements at 7 DPS demonstrated the slices settled quickly in culture, with the PI signal significantly elevated in staurosporine-treated slices compared to untreated counterparts (Untreated vs. Staurosporine; 3.60±3.59 and 73.67±14.82 percent area PI, p=0.0101). This observation continued until 70 DPS where the slices remained viable in culture determined by low PI fluorescence compared to a staurosporine-treated counterpart (Untreated vs. Staurosporine: 3.34±1.69 and 34.57±2.80 percent area PI, p=0.007).

Taken together, the PI viability experiments confirm the protocol developed successfully enables slices to be cultured organotypically for up to 70 DPS.

116

Figure 4.2: Organotypic tga20 cerebellar brain slices are viable in culture. Viability was measured by quantifying propidium iodide and Hoechst 33342 labelling in live slices. Slices exposed to staurosporine for 48 h prior to visualisation served as a positive control of cell death. Images were taken on an inverted wide-field microscope (Leica DM16000) using 10x and 20x objective lens. Graphs are presented as mean±SEM (n=3). Statistical significance was examined by Student’s t test with a statistical criterion of 0.05. * p<0.05, *** p<0.001

117

Given that the degree of neuronal vulnerability will be assessed in prion-infected OBSCs, it was pertinent to determine the changes that inherently occur to neurons in healthy slices over time. The expression of neuronal proteins relevant to neurodegeneration (both pre- and post- synaptic) was measured in OBSCs harvested at 1, 7, 21, 42, 56 and 70 DPS and compared to tga20 brain tissue obtained from a pup the same age as those used for OBSCs experiments (9- 10 days old). The following neuronal markers were used: post synaptic density-95 (PSD-95), synaptophysin, the 25 kDa synaptosomal associated protein (SNAP25) and vesicle-associated membrane protein 2 (VAMP2).

Immunoblot revealed expression of all neuronal markers in OBSCs was reduced compared to fresh tissue isolated from tga20 brain (Figure 4.3). This decrease occurred immediately once in culture (at 1 DPS) and was maintained throughout all time-points examined. The strongest reduction was found in synaptophysin, with very low or no expression identified after 7 DPS. While the expression of all neuronal markers did not recover to levels seen in vivo over the time period examined, their exact levels appeared to fluctuate over time in culture (Figure 4.3).

Given that PrP expression will be assessed in OBSCs upon prion infection, changes in endogenous PrP levels was also measured at all time-points. PrP was detected in all samples which, similar to the neuronal markers, was variable over the experimental period. Changes to the astrocyte marker glial fibrillary acidic protein (GFAP) was also measured. Strikingly, the expression of GFAP was highly elevated compared to the fresh tissue. This increase was marginal at 1 DPS and higher in all other time-points (Figure 4.3). This is consistent with others who have reported significant elevations in GFAP protein in slices once they become organotypic [624, 625].

Probing this panel of markers in OBSCs revealed important information about the changes imparted on slices in culture. The large elevation in GFAP signal with increased time gave further evidence that the protocol developed successfully produces slices that are organotypic. The fluctuations on protein expression revealed OBSCs to be dynamic, living systems.

118

Figure 4.3: Cell populations are dynamic in tga20 OBSCs. OBSCs prepared from tga20 pups were maintained for up to 70 days. Pooled slices (6) were collected, lysed and resolved using SDS-PAGE. Fresh tissue isolated from a tga20 pup was resolved alongside OBSC samples. Proteins were transferred to PVDF membrane and western immunoblot performed to detect changes to cell populations in OBSCs over time in culture. Protein markers used were: PSD-95, synaptophysin (Synapt), SNAP25 and VAMP2, PrP and GFAP. β-actin was probed for as a loading control. n=1.

119

Cerebellar OBSCs from tga20 mice propagate PrPRes upon prion infection Having developed a reliable protocol to culture brain slices organotypically, their ability to adopt infection and propagate PrPRes after exposure to prions was next assessed. The prion strain used for infection experiments in this thesis is M1000, a human familial prion strain [626] that has become mouse adapted following serial passages in inbred mice. M1000 was originally isolated from a patient with GSS and is known as the Fukoaka-1 prion strain [627].

For the infection experiments, sectioned tga20 brain slices were exposed to 1% (w/v) M1000 or NBH for 1 h at 4oC. Slices were subsequently washed well to remove residual PrPSc and cultured as per standard culture conditions (Figure 4.4A). The viability of M1000-infected brain slices was assessed by PI/H staining at 1, 7 and 70 DPS. Similar to uninfected, naïve slices (Figure 4.2), a high degree of PI positive staining was evident 1 DPS in both NBH and M1000- treated slices (NBH and M1000; 49.22±10.4 and 53.73±20.65 percent area PI, respectively) (Figure 4.4B). No significant difference was found between the two groups. By 7 DPS, both treatment groups were viable with significantly lower PI staining compared to staurosporine-treated slices [(Staurosporine vs. NBH; 90.62±5.50 and 0.023±0.01 percent area PI, p=0.0001) and (Staurosporine vs. M1000; 90.62±5.50 and 10.4±8.903 percent area PI, p=0.0002). This observation was similarly observed at 70 DPS, with both M1000- and NBH-treated slices showing low PI values compared to their staurosporine-treated equivalent [(Staurosporine vs. NBH; 72.19±6.52 and 0.200±0.105 percent area PI, p=0.0001) and (Staurosporine vs. M1000; 72.19±6.52 and 0.014±0.005 percent area PI, p=0.0001)] (Figure 4.4B). The graphical representation of this data is shown in Figure 4.4B, while representative images used in the analysis of PI staining are shown in Figure 4.5.

The ability of M1000-infected brain slices to propagate PrPRes was determined by detecting PK resistant PrP products in NBH and M1000-infected slices (6 per group) collected at 1, 7, 21, 42, 56 and 70 DPS. Digested samples were resolved on SDS-PAGE and PrP detected by western immunoblot. Results showed a time dependent elevation in PrPRes in M1000-infected slices, which was first detected at 42 DPS and continued to increase at both 56 and 70 DPS. By comparison, no detectable PrPRes was identified in slices exposed to NBH. In order to confirm the presence of PrP species in all samples, the experiment was repeated in the absence of PK (Figure 4.4C).

120

Figure 4.4: PrPRes can be detected in viable OBSCs infected with M1000 prions. (A) The protocol for infection experiments involved exposing freshly cut tga20 slices to 1% (w/v) M1000- or NBH prior to removal of residual homogenate and plating as per standard organotypic technique. (B) Viability was measured by quantifying propidium iodide and Hoechst 33342 labelling in live slices. Slices exposed to staurosporine for 48 h prior to visualisation served as a positive control of cell death. Graphs are presented as mean±SEM (n=3). Statistical significance was examined by Student’s t test with a statistical criterion of 0.05 *** p<0.001, ****p<0.0001. (C) Slices (6) were pooled, lysed and either exposed to PK or left undigested prior to western immunoblot for detection of PrP. PrP expression was detected using 03R19 antibody. n=1.

121

Figure 4.5: Representative images used for statistical analysis shown in Figure 4.4B. Images were taken on an inverted wide-field microscope (Leica DM16000) using 10x and 20x objective lens.

122

PrPRes can be identified in single OBSC exposed to M1000 prions It was of interest to determine whether the PrPRes generated in M1000-infected OBSCs can seed amplification of PrPRes using PMCA. Here, individual NBH- or M1000-infected slices were collected at the same DPS as previous experiments (1, 7, 21, 42, 56, and 70), lysed in PCB and added to NBH substrate as per the standard PrP PMCA protocol (refer to Chapter 3 for details, Figure 3.1). Samples were subjected to one round of PMCA and PrPRes detected in samples via PK digestion (50 µg/mL final concentration of PK) and western immunoblot analysis. Results showed M1000-infected brain slices contained PrPRes at 56 and 70 DPI (Figure 4.6A).

To determine if PrPRes could be detected at earlier time-points, a serial PMCA was performed. Here sample that had undergone one round of PMCA (samples from Figure 4.6A) was spiked into fresh tubes containing NBH in PCB and exposed to another round of PMCA. PK digestion and immunoblot of these samples again showed PrPRes in 56 and 70 DPI, however it was unable to increase the sensitivity to include earlier time-points (Figure 4.6B).

In addition to confirming the ability of M1000-infected OBSCs to seed amplification of PrPRes in the PMCA, this experiment demonstrates that misfolded protein may be detected using very low amounts of protein; this experiment used one brain slice, whereas the detection of PrPRes by PK digestion alone (Figure 4.4C) used six pooled slices per time-point. Hence the use of this assay to detect PrPRes is advantageous because it reduces the number of slices needed to detect prion infectivity.

123

Figure 4.6: Singular M1000-infected brain slices amplify PrPRes using PMCA. (A) Singular brain slices from NBH- or M1000-infected slices were lysed and added to NBH homogenised in PCB. Samples were exposed to one round PrP PMCA prior to PK digestion (50 µg/mL final concentration) and western immunoblot performed to detect PrPRes. (B) After one round of PMCA, samples were spiked into fresh tubes containing NBH and another 24 h PrP PMCA performed (two rounds total PMCA) prior to detection of PrP as above. n=1. * indicates non-specific binding of the antibody

124

M1000-infected OBSCs develop neurotoxicity Neuronal vulnerability in M1000-infected OBSCs was next assessed. Both NBH- and M1000-infected OBSCs were collected at 1 or 70 DPS and expression of neuronal markers quantified by immunoblot. For this, the neuronal markers utilised in initial characterisation experiments on uninfected slices (Figure 4.3) were used: PSD-95, synaptophysin, SNAP25 and VAMP2.

Because substantial variation was found in neuronal markers over time in uninfected slices (for example, see Figure 4.3), samples collected from 1 and 70 DPS were run on separate blots to facilitate densitometric analysis. Initial experiments aimed at quantifying neuronal expression at 70 DPS failed because the difference between NBH- and M1000-infected groups was so significant that both bands could not be visualised at the same time without saturation of the NBH signal Therefore, to enable quantification of the 70 DPS samples, two fold more protein was loaded for the prion infected group compared to uninfected NBH group (10 and 5 µg respectively). This issue was not observed in samples from 1 DPS and therefore equivalent protein loading was used at this time-point (5 µg for each treatment group).

Densitometry and statistical analysis was performed on three individual experiments. At 1 DPS, no significant difference was found in any of the neuronal markers. In contrast, at 70 DPS, a decrease in all neuronal markers was found in prion-infected OBSCs with the fold change over NBH for PSD-95, synaptophysin, SNAP25 and VAMP2 being: 3.55, 18.64, 4.82 and 3.22 AU, respectively. This difference was statistically significant for all neuronal markers. The graphical representation of this data is shown in Figure 4.7 and the raw values, SEM and p-values for all comparisons are shown in Table 4.2.

Collectively, these results demonstrate clearly there are synaptic changes associated with infection, where M1000 causes a selective loss of key neuronal proteins of both the pre- and post-synapse.

125

Figure 4.7: M1000-infected OBSCs develop neurotoxicity. Tga20 slices exposed to M1000 or NBH were cultured for 1 or 70 days. Pooled slices (6) were collected, lysed and resolved using SDS-PAGE and immunoblot analysis performed to detect the expression of neuronal markers: PSD-95, synaptophysin (Synapt), SNAP25 and VAMP2. β-actin was probed for as a loading control. Graphs are presented as mean±SEM (n=3). Statistical significance was examined by Student’s t test with a statistical criterion of 0.05. * p<0.05, *** p<0.001

126

Neuronal Marker Values ± SEM (AU) p-value Significance PSD-95 1.493±0.283 and 0.4207±0.014 0.0194 *

Synapt 1.995±0.107 and 0.107±0.025 0.0232 *

SNAP25 1.426±0.283 and 0.296±0.1473 0.024 *

VAMP2 1.936±0.084 and 0.6002±0.088 0.0004 ***

Table 4.2: Values of significant differences identified in Figure 4.7. Statistical significance was examined by Student’s t test with a statistical criterion of 0.05. * p<0.05, *** p<0.001

127

Discussion

Inoculating susceptible animals with prions leads to bona fide disease, making it a powerful tool to study disease that is reflective of the human condition. While this experimental system is the gold-standard, in vivo studies are generally labour intensive, expensive and lower in through-put than comparable in vitro techniques. Models of prion disease that reduce or eliminate the issues associated with in vivo systems while being more powerful than using cultured cells are attractive options to further our understanding on disease pathogenesis. This chapter showed the development and characterisation of a model of prion disease using OBSCs.

The first component of this chapter was to develop a robust, reproducible technique to culture naïve brain slices organotypically. I showed the successful development of this technique using cerebellar slices derived from transgenic tga20 pups. In culture, these slices were found to be viable for long periods of time (up to 70 days). This was shown by standard techniques of quantifying dead cells (via PI fluorescence) and supported by observed changes to cellular populations consistent with organotypic culturing. Here one of the most striking changes was elevations in the astrocyte marker GFAP in OBSCs; a finding that has been reported by others [624, 625]. Structural support is one of the main functional roles of astrocytes in the CNS and hence its up regulation in OBSCs is likely to recover the slice from the loss of structure associated with the sectioning process. The loss of a functioning BBB may also be a trigger for the up regulation of GFAP in these slices. Astrocytes play an important role in the generation of the BBB: numerous studies show cultured endothelial cells are able to adopt properties of the BBB upon exposure to astrocytes [628-632] and accordingly, HIV-infected astrocytes damage the integrity of the BBB [633]. The ability of OBSCs to undergo such drastic biophysical changes once in culture confirms them as a viable, functional tissue that is able to respond to various stress stimuli. While this is strong support for their ability to replicate in vivo systems, this also carries associated caveats. Such significant alterations to cellular populations would drastically alter the ratios of cellular populations beyond what would occur in an intact organism. OBSCs also progressively spread out and thin over time (example in Figure 4.1), which would also significantly influence the cellular microenvironment of the tissue.

Following the development of a protocol to successfully culture OBSCs, the cultures were exposed to M1000 prions and the development of hallmarks associated with in vivo disease

128 assessed. Following infection, viable OBSCs were shown to propagate PrPRes and develop neuronal injury consistent with what occurs in a diseased organism. To my knowledge, this is the first time that a mouse-adapted human prion strain has been used to infect OBSCs, and the characterisation studies performed in this chapter will be valuable for future experiments using the M1000 prion strain in this system. The propagation rate of PrPRes in cerebellar OBSCs upon infection with M1000 appears to be slower than other prion strains. Studies investigating the ability of cerebellar OBSCs to be infected by prions (determined by the abundance of PrPRes at 35 DPS) reveal variations in PrPRes abundance associated with strain type. Of 15 strains tested (which included strains where the original inoculum was derived from scrapie, BSE and CWD) the most infectious strains were found to be the mouse-adapted scrapie strains RML, 79A and 139A, while several strains showed no detectable PrPRes within the experimental timeframe [616]. Accordingly, RML is the most widely used strain to infect OBSCs, with several labs reporting the detection of PrPRes in tga20 OBSCs within 21 days in culture [391, 615-617, 634] and subsequent associated neuronal vulnerability by 42 [391], 45 [392] or 69 DPS [617]. This is in contrast to M1000 where, reported in this chapter, PrPRes in tga20 OBSCs and neurotoxicity were first identified at 42 and 70 DPS, respectively. While RML would be a useful control to use in this chapter to compare to M1000, the use of scrapie or BSE-derived prion strains are restricted in Australia and can only be used under high levels of physical containment. Hence M1000 was the only strain used for this thesis.

The reasons for differential infectivity and/or rates of propagation between strains in OBSCs are unknown, however a contributing factor may be the brain region the prions are exposed to. The spread of PrPSc over the course of disease is strain-specific and in human diseased brain, the deposition of PrPSc and associated pathology in the cerebellum is highly variable [635, 636]. As such, it is possible that neurons in the cerebellum are less susceptible to infection with the M1000 prion strain than other strains. Proof of principle for variation in the capacity of strains to infect the cerebellum comes from studies mapping the spread of PrPSc deposition in mice where compared to RML, the 22L strain spreads to the cerebellum earlier and carries higher PrPSc burden in this region in in vivo disease [637]. This is an interesting observation given that to date the only strain that has been shown to effectively amplify PrPRes in wild-type cerebellar OBSCs is 22L [618]; with all other OBSC models of prion disease using slices derived from tga20 mice. Hence these studies support the concept that strains that efficiently penetrates the cerebellum in vivo appear to also be highly effective in OBSCs. This is particularly interesting because it suggests that the susceptibility of the

129 cerebellum to adopt infection in disease is not solely associated with the ability of PrPSc to spread to this region in an intact organism. Instead, it is also associated with a strain’s ability to infect and propagate within the cerebellum following direct exposure to prions. While these considerations are useful to delinerate the ability of brain regions to adopt prion infection, the relevance of strain specificity to prion infection in OBSCs is confounded by variables such as titre of the infectious material in the inoculum and the time the slices are exposed to the initial infectious inoculum. Hence without performing controlled infection assays using material that has been quantified by infectivity (using assays such as the scrapie cell in the end point format assay, SCEPA), it is challenging to directly compare between studies. Regardless, the finding in this chapter that a mouse-adapted human strain can infect OBSCs is the first time this strain type has been shown effective in this system. This is an important observation that permits further use of the M1000 strain in OBSCs and supports the use of other adapted human strains in this system.

The neuronal markers used in this chapter comment on the status of various components of a functioning neuron. Postsynaptic density 95 (PSD-95) was chosen as it is one of the key components of excitatory post synaptic density [638] and is involved in synaptic plasticity [639]. The 25 kDa synaptosomal associated protein (SNAP25) and vesicle-associated membrane protein 2 (VAMP2) are both components of the SNARE complex formation [640]; the term given to various soluble N-ethylmaleimide-sensitive factor (NSF) attachment receptors (SNAREs) that collectively co-ordinate the release of neurotransmitters from the presynaptic terminal. Here synaptic vesicles containing neurotransmitters harbour vesicle SNAREs (v-SNAREs) on their exterior surface, which interact with target SNAREs (t- SNAREs) anchored in the pre-synaptic membrane, allowing for the neurotransmitters to be released into the synaptic cleft. VAMP2 and SNAP25 are essential v- and t- SNAREs, respectively, and directly interact with each other upon the docking of the vesicle with the presynaptic membrane [641-644]. In addition to roles in exocytosis, both proteins are implicated in endocytosis at the synapse [645, 646] and hence play multifaceted roles in synaptic transmission. Synaptophysin is a calcium binding glycoprotein and is the most abundant protein on synaptic vesicles [647-649]. Indeed its expression is so high that it is used as a general marker of the presynaptic compartment.

In uninfected naïve OBSCs, all neuronal markers were down regulated compared to freshly isolated tga20 tissue (Figure 4.3). This is consistent with the changes in neuronal networks imparted on OBSCs, which undergo significant neurodegenerative and regenerative processes

130 post sectioning. This has been observed using electrophysiological recordings, where OBSCs are shown to form new functional synaptic contacts within 3 – 6 days in culture [614]. Synaptophysin appeared most considerably modulated by the sectioning and culturing process, with no detectable expression observed after 7 DPS when processed alongside fresh tga20 tissue. Given that both synaptophysin and VAMP2 are expressed on synaptic vesicles, it is curious that their expression did not align. This may be due to additional factors that alter their expression, such as stress induced from the cutting process. Neuronal proteins are differentially modulated by stress, which is capable of causing significant and long lasting effects to their expression. Numerous studies report significant decreases in synaptophysin protein or mRNA expression in mice following acute injury [650] and immobilization stress [651-653]. Hence the large reduction in synaptophysin and its divergence to other presynaptic proteins could be due to confounding factors aside from neuronal connectivity.

At 70 DPS, all neuronal markers were significantly decreased in M1000-infected OBSCs compared to the control tissue exposed to NBH. This is consistent with rodent animal models of disease in whole organisms that similarly show decreases in the protein or mRNA of PSD- 95 [372, 654], synaptophysin [655-657], VAMP2 [372, 657] and SNAP25 [655]. Similarly, in the cerebellum of human CJD patients, both synaptophysin and SNAP25 are found decreased [658]. In this chapter the fold change difference in neuronal markers between M1000- and NBH-treated samples at 70 DPS was comparable for PSD-95, VAMP2 and SNAP25 (a range of around 3.5), whilst for synaptophysin this fold change was much higher (18.64). Given the strong expression of synaptophysin at the presynaptic terminal, it indicates this region as highly affected in disease. In support of this is the finding that prion-infected OBSCs have significant reductions in dendritic spine density compared to uninfected controls [617]. Again, an important point is that its modulation is not proportional to the other presynaptic markers, VAMP2 and SNAP25. This may reflect the higher starting expression of synaptophysin, contributions by other modulators of its expression (such as stress, discussed above) or its exclusive localisation at the pre-synapse. Indeed, while VAMP2 and SNAP25 are largely presynaptic markers, they both appear to have physiological roles in postsynaptic compartments. VAMP2 has been found within the post-synapse where it is involved in the trafficking of essential receptor subunits [659]. It remains unclear whether SNAP25 is expressed in the post-synapse, however it appears to have important functional roles in the region which suggests it is likely to also be localised there to a certain degree [660]. Hence while the reductions in PSD-95 upon prion infection clearly confirms both

131 neuronal compartments are affected by M1000, if they are modulated to varying degrees this may explain the variation in expression of the predominately presynaptic neuronal markers seen in this chapter.

While immunoblot to a panel of neuronal markers confirmed the development of neurotoxicity in M1000-infected OBSCs, it was not associated with elevations in PI fluorescence. This is in contrast to other studies using OBSCs, where in RML-infected OBSCs a negative correlation is seen between PI staining and the abundance of the neuronal marker, NeuN [391]. This may be due to differences in the prion strain and the degree or type of neurotoxicity between these systems. Given that PI detects RNA and DNA in cells that have a compromised membrane, these are likely removed from the culture system relatively quickly during media changes, and hence PI staining likely identifies cell death only for a short period of time. In this regard, at 70 DPS in M1000-infected OBSCs, the rate of cell death may be too low and/or too early to detect a significant change in PI staining. Additionally if the cell death is neuronal specific it may be too late to detect these changes given that significant cell death has already occurred. Hence while PI is a useful tool to measure viability compared to staurosporine-treated slices, it is not surprising that it could not detect any M1000-induced damage. This is further supported by the observation that the prion-induced elevations in PI in OBSCs reported by others is marginal, with PI (percent area) staining of infected slices at 42 DPS being around 6, compared to values around 60 for staurosporine-treated slices [391].

132

Conclusion

Neurotoxicity is an essential component to prion disease pathogenesis. History has taught us that the use of model systems that do not develop neurotoxicity upon prion infection are unlikely to be reflective of the human condition. This chapter aimed to develop a model of prion disease that shows neurodegeneration in association with prion propagation using OBSCs. Herein I showed the development of a robust protocol to culture OBSCs for long periods of time (up to 70 days). Critically, I showed these slices are conducive to prion infection, wherein thorough characterisation on the propagation of PrPRes and neuronal health upon infection with M1000 prions gave a detailed view on development of disease in this system. This is the first time the prion strain M1000 has been used to infect OBSCs and as such are important characterisation studies that permit further experimental use of this prion strain. The development of this model is further used to study the misfolding and toxicity of PrP in Chapter 5.

133

CHAPTER 5: DETERMING THE RELEVANCE OF PROTEIN MISFOLDING TO THE PATHOGENICITY OF αSYN AND PRP USING ANLE138B

Introduction

Chaper 3 and 4 showed the development of in vitro and ex vivo models that report the ability of misfolded αsyn and PrP to alter normal neuronal functioning. Following on from this, I sought to use these systems as platforms to directly compare the relevance of a pathogenic feature presumed to be conserved between the two proteins. As mentioned in Chapter 1, the misfolding of PrP and αsyn is considered an upstream process that enables both proteins to adopt pathogenic features associated with disease [3]. Various drugs have been shown to modulate the misfolding of NDP-associated proteins or potential downstream neurotoxic pathways, which translates to improved disease parameters in animal models. Because of this, this chapter aimed to test whether a small molecule that recognises pathogenic αsyn and PrP conformers can modify their aggregation properties and their pathogenicity as defined in Chapter 3 and Chapter 4. Such work has important implications to further our understanding on the generation of pathogenic features of these proteins but also, critically, to identify molecules that may be therapeutic in disease.

Identification of molecules that are effective for therapeutic intervention in prion disease To date, one of the most promising molecules trialled in animals and humans to treat prion disease has been the polyanion, pentosan polysulfate (PPS). PPS interferes with prion replication and is efficacious in animal models to prions administered to both peripheral compartments [584-587] and the CNS [571]. However human trials have had variable results [588-593] (Table 4.1) considered to be due, in part, to the inability of the drug to effectively cross the BBB. This requires the compound to be administered intraventrically and hence is a complication to the extrapolation of PPS to treat disease in humans. More recent promising small compounds have been demonstrated to transverse the BBB. Compound-B (cpd-B) extends survival of mouse-adapted scrapie infected mice by 2.3% when treatment is administered orally from the day of intracerebral inoculation with prions [661]. Due to the potential of cpd-B to generate toxic metabolites [662], less-toxic analogues have been

134 generated with similar anti-prion effects, however they do not appear effective against human prion strains [663]. The compound IND24 significantly extends the lifespan of mice infected with mouse-adapted scrapie prions, however IND24 was ineffective to human prion strains, and has the capacity to generate drug resistant prions [664]. Luminescent conjugated polymers (LCPs) show high affinity to bind amyloid and certain compounds are capable of crossing the BBB [665]. Various LCPs have been shown to reduce PrPSc load in infected brain homogenate and prion-infected OBSCs and mice, where they appear to act via hyperstabilizing PrPSc deposits [634, 666]. Structure-activity studies reveal a five-thiophene moiety and charged side groups as important to the anti-prion activity of LCPs, with the most potent compound identified being LIN5001 [666] (Figure 5.1). In mice, LIN5001 extends survival by 36.4% compared to control animals when treatment is initiated 7 days prior to intracerebral inoculation of RML prions, and by 20.5% when treatment starts 20 days post infection [666]. However, in this study LCPs were administered via the intraventricular route and as such its ability to surpass the BBB is unclear.

To date, one of the most efficacious drugs identified to treat prion disease in animals is Anle138b. The anti-prion properties of this molecule were identified by performing two library screens of 10,000 compounds using the assay scanning for intensely fluorescent targets (SIFT) that detects molecules that disrupt the association between PrPC and PrPSc. Positive hits using SIFT were then further tested in vitro using a cell-based dot blot assay to measure its ability to reduce PrPSc propagation. From these studies, cluster analysis revealed a lead structure of 3,5-diphenyl-pyrazole derivatives and a further focused library (150 compounds) was produced and screened using SIFT, cell culture, PrP PMCA and animal models of prion disease. This robust set of experiments identified numerous potential hits, with Anle138b found to have the highest anti-prion activity in vivo and in vitro [345]. Extensive analysis has been performed on the ability of Anle138b to modulate PrP misfolding. It is able to reduce PrPSc amplification in PMCA of prions derived from mouse- adapted scrapie (RML, ME7), mouse-adapted BSE (301C) and human (sCJD, vCJD) strains [345]. The ability of Anle138b to treat all stages of disease was assessed in RML-infected mice in studies where treatment was initiated from the day of infection, at 80 days post infection or at 120 days post infection. These time points were chosen because by 80 days post infection untreated mice start to develop subtle phenotypic symptoms associated with prion disease and have PrPSc deposition in their brain, while by 120 days post infection clinical disease is obvious and the disease is considered to be in an advanced stage.

135

Compared to DMSO-treated controls, Anle138b treated mice had a median survival of 1.98, 1.44 and 1.30 fold higher when treatment commenced at the day of infection, day 80 and day 120, respectively [345]. The ability of Anle138b to be efficacious when treatment is started at advanced stages of disease makes it one of the strongest anti-prion compounds identified to date. Sucrose gradient centrifugation of PrPRes species revealed Anle138b to be effective at reducing both large aggregates and small oligomers [345], and hence it was suggested Anle138b acts as an oligomer modulator.

Targeting the neurotoxic agent would likely be an advantageous therapeutic strategy, and given the plethora of data suggesting aggregated PrPSc is not the toxic agent, compounds acting to reduce PrPSc load may be ineffective. For example, PPS does not interfere with the toxicity of the PrP-directed antibody POM1 [392] and therefore if PrPC is the upstream neurotoxic agent, PPS will likely be of little therapeutic benefit. Compounds that act on potential neurotoxic pathways in prion disease show great promise. The PERK-inhibitor compound GSK2606414 suppresses the manifestations of behavioural deficits and prion- disease associated neuropathology in mice when daily treatment was commenced in parallel with the first signs of synapse loss in diseased mice [420]. GSK2606414 easily crosses the BBB so can be administered orally, and has a high specificity for PERK [420, 667]. Another compound, ISR1B, which acts downstream of eIF2α, has been shown to be therapeutic to mouse models of prion disease [421], however unfortunately both compounds are not well tolerated in vivo. Recently other compounds, trazodone hydrochloride and dibenzoylmethane, have been shown to target the UPR and are efficacious to prion infected animals [422]. Both act downstream of eIF2α, are bioavailable and readily surpass the BBB. Critically, both drugs show no signs of peripheral toxicity reported by other UPR-modulating compounds and given that trazodone is a licensed antidepressant drug routinely used to treat sleep disturbances in AD [668], it is an attractive compound to extrapolate to treat the human condition.

Several efficacious compounds discussed in this chapter are shown in Figure 5.1.

136

Figure 5.1. Structure of several compounds that possess prion-alleviating properties. The large polyanionic compound, pentosan polysulfate (PPS) is one of the most widely used anti-prion compounds, however its efficacy is limited in the human condition due to difficulties surpassing the blood brain barrier. Newer compounds have been developed that show efficacy to prion-infected rodents or OBSCs, such as: compound-B (cpd-B), IND24, 5001 (a luminescent conjugated polymer, LCP), Anle138b, GSK2606414 and trazodone hydrochloride.

137

Anti-prion molecules that show efficacy towards other NDP-associated proteins Numerous compounds that reduce PrPSc burden and/or extend preclinical incubation time in prion-diseased animals have likewise been shown to be therapeutic to other animal models of NDP. The identification and use of such compounds are valuble tools to study the similarities between NDP-associated proteins.

While several compounds such as GSK2606414, ISR1B and trazodone that target the PERK- arm of the UPR have been described to be effective in prion infected mice [420-422], several also appear suitable treatments for models of tauopathy. In a transgenic mouse line overexpressing a repressible form of human mutant tau (P101L), GSK2606414 offers neuroprotection and suppresses clinical disease [669]. Both trazodone and dibenzoylmethane are similarly therapeutic in this mouse line [422] and hence the UPR is likely a relevant neurotoxic pathway associated with pathogenic forms of several NDP-associated proteins.

Anle138b has also been shown to be efficacious to pathogenic αsyn and tau, with varying degrees of effectiveness in mouse models of PD [345, 346], MSA [670] and tauopathy [671]. The treatment of transgenic mice overexpressing αsyn containing the A30P mutation in neurons with Anle138b causes a lag in the generation of disease-associated motor deficits, an extended preclinical incubation period and less aggregated αsyn deposits in the brain compared to DMSO-treated counterparts when treatment is commenced from 8 weeks of age [345]. This translates to an increase in the median of disease-free survival by 10 weeks. Strikingly, by 69 weeks of age, transgenic A30P mice treated with Anle138b have reduced amounts of oligomeric αsyn species in their brain [345]. This was shown using sucrose gradient density where the presence of αsyn species derived from Anle138b-treated mice were found in similar fractions to that of recombinant monomeric αsyn protein. In contrast to this, the species extracted from DMSO-treated mice mirrored the profile of Fe-induced αsyn oligomers [345], suggesting that similar to PrP, Anle138b reduces αsyn oligomeric structures. In addition to its efficacy when using an early treatment regime, Anle138b also extends the survival in this mouse line when treatment commences after the onset of clinical symptoms (350 days of age) [346]. Likewise, Anle138b also appears effective to a MPTP toxicity model of PD, by protecting neurons in the substantia nigra from MPTP-induced neurotoxicity [345]. However, whilst these studies show Anle138b is therapeutic in various animal models of PD, its effect is not as striking in mouse models of MSA. Transgenic mice overexpressing wild- type αsyn in oligodendroglia [672, 673], which are intoxicated with 3-nitropropionic acid are used to model MSA which develop neurological dysfunction and pathology similar to the

138 human condition. Using this acute model, the treatment with Anle138b only caused minor improvements to clinical symptoms and was not found to alter the pathological deposition of protein or neuronal loss [670].

Anle138b has also shown to be effective in tauopathy mouse models. This was observed using the transgenic mouse model PS19 that express human tau containing a P301S mutation. These mice develop neurofibrillary tangles, neuronal loss and glial activation that is associated with age-related clinical symptoms from around 7 months of age [674]. Anle138b treatment initiated post weaning led to reductions in insoluble tau load in the brain and a rescue of tau-associated neuropathology including neuronal loss and astrogliosis. Anle138b also improved cognition and extended survival by 2.9 fold at 12 months of age (a time-point where 80% of untreated PS19 had succumbed to disease) [671].

Anle138b is a stable molecule that readily passes through the BBB when introduced into the periphery. Because of this, most of the reported studies using Anle138b administer the compound as a food supplement, which leads to high levels in the serum and brain within hours of consumption [345]. Other routes have also been shown effective for Anle138b administration including intraperitoneal injection (i.p) and oral gavage, however compared to its incorporation into food, this leads to more rapid clearing of the molecule following i.p administration and lower bioavailability in the serum and brain following oral gavage [345]. Nonetheless using all treatment groups Anle138b reaches around 3 fold higher levels in the brain compared to the serum [345], demonstrating excellent BBB penetration.

The effect of Anle138b on protein fibrillization has been studied. In vitro studies show Anle138b can directly bind to misfolded PrP [345], αsyn [345, 675] and tau [671]. The ability of Anle138b to reduce the abundance of low molecular weight PrPRes in prion infected brain and αsyn oligomers in transgenic A30P mice initially led to the suggestion that it was an oligomer modulator, however more recently, in the case of αsyn, it has been shown to bind to aggregates at hydrophobic sites [675]. Given that oligomers harbour weak structural arrangement, this may suggest that Anle138b binds to aggregates of a higher order than oligomers. As such, the influence of Anle138b on protein misfolding needs further analysis.

139

Aims

Anle138b has been reported to bind to and modulate the fibrillization of both PrP and αsyn and it is hypothesised that this binding interferes with further fibrillization. It is unclear however, what conformation Anle138b recognises, the effect this has on fibrils formed and exactly how this modulates disease progression. This chapter investigates the ability of Anle138b to modulate PrP and αsyn fibrillization and the effect this has on the pathogenic properties of the misfolded protein. This will be studied by employing the models established in Chapter 3 and Chapter 4.

Results

Anle138b reduces PrPRes in M1000-infected OBSCs The OBSC model of prion disease developed in Chapter 4 was first used to study the efficacy of Anle138b in this system. An M1000-infection experiment was performed on pooled slices as described in Chapter 4 (see Figure 4.4) which were then separated into the four following treatment groups: untreated, DMSO (vehicle control), 0.1 µM Anle138b and 1.0 µM Anle138b. Anle138b treatment was performed by diluting the compound in DMSO and adding it to SCM at a concentration of 1:1000 for a final concentration in media of 0.1 µM or 1 µM. The equivalent volume of DMSO was added to SCM for the DMSO-only vehicle control. Compound-free SCM and Anle138b- or DMSO-containing SCM were used for the duration of the culturing process starting from 4 h post sectioning. Six slices per treatment group were collected at 1, 42, 56 and 70 DPS.

To confirm that Anle138b was well tolerated by the OBSCs, viability was assessed at 70 DPS for all treatment groups. The PI staining of slices from all treatment groups was found to be significantly lower than staurosporine-treated slices (Staurosporine vs. M1000, +DMSO, +0.1 µM Anle138b or +1.0 µM Anle138b; 22.48±10.63 vs. 0.23±0.12, 0.85±0.49, 0.51±0.21 or 0.87±0.44 percent area PI, respectively. All groups: p<0.0001) (Figure 5.2A). Representative images used in PI analysis are shown in Figure 5.3. This experiment confirmed that neither Anle138b nor DMSO compromise the viability of M1000-infected OBSCs, even after extended periods of time in culture.

140

The ability of Anle138b to modulate the propagation of PrPSc in M1000-infected OBSCs was next assessed. This was evaluated by performing PK digestion on all treatment groups and time-points (1, 42, 56 and 70 DPS) to detect PrPRes, a marker of prion propagation. An important component to analysing results of the PK digestion experiment was to first clarify whether Anle138b is able to modulate the sensitivity of PK digestion or immunodetection of PrP in the OBSCs. To determine this, M1000-infected brain homogenate was incubated with 0.1 µM or 1.0 µM Anle138b for 1 h prior to PK digestion. Negative controls included infected brain homogenate incubated with dPBS or DMSO; buffers that were used to dilute PK and Anle138b to their desired concentration, respectively. Equivalent samples were left undigested. Western immunoblot of all samples revealed no differences in digestion patterns following PK treatment (Figure 5.2B) and therefore Anle138b was confirmed to not modulate the digestion of PrP or sensitivity of the western immunoblot assay.

Due to low amounts of protein recovered following extended periods of time in culture, all analysis on Anle138b-treated OBSCs was performed on a pool of six slices. While this restricted the ability to perform statistical analysis (all studies using Anle138b-treated OBSCs are n=1), it was deemed the most appropriate way to fully characterise the changes Anle138b imparts on the M1000-infected OBSCs.

All treatment groups and time points were next exposed to PK digestion (20 µg/mL final PK concentration) and western immunoblot performed. For all treatment groups, low levels of PrPRes was identified at 42 DPS, which was elevated by 56 DPS. A difference was seen between the treatment groups by 70 DPS, where while the PrPRes in the control groups (M1000 and +DMSO) continued to increase from 56 to 70 DPS, a plateau was observed in the PrPRes of both Anle138b treated groups (Figure 5.2C). Consistent with this, comparing all treatment groups at 70 DPS, reveal less PrPRes in the Anle138b-treated groups. Taken together, this data reveal Anle reduces the propagation and abundance of PrPRes in OBSCs.

141

Figure 5.2: Anle138b reduces PrPRes in viable M1000-infected OBSCs. Tga20 OBSCs exposed to M1000 prions were cultured in the presence of 0.1 or 1 µM Anle138b for up to 70 days. Untreated infected slices (M1000) and vehicle treated slices (+DMSO) were used as controls. (A) Viability was measured by quantifying propidium iodide and Hoechst 33342 labelling in live slices. Slices exposed to staurosporine for 48 h prior to visualisation served as a positive control of cell death. Graphs are presented as mean±SEM (n=6). Statistical analysis of the difference was examined by one-way ANOVA using Dunnett’s multiple comparisons test, with a statistical criterion of 0.05. **** p<0.0001 (B) Two sets of infected brain homogenate (INF BH) was incubated with 0.1 µM or 1.0 µM Anle138b, DMSO or dPBS for 1 h at RT. One set each was then incubated with PK (20 µg/mL final concentration) or dPBS prior to samples being resolved on SDS-PAGE and western immunoblot performed to detect PrP. (C) Slices (6) were pooled, lysed and PK digestion performed as above, prior to SDS- PAGE and western immunoblot for detection of PrPRes. PrP was detected using 03R19 antibody. Quantification of PrPRes was performed for OBSCs cultured for 56 and 70 days. n=1, 6 pooled slices.

142

Figure 5.3: Representative images used for statistical analysis shown in Figure 5.2A. Images were taken on an inverted wide-field microscope (Leica DM16000) using 10x and 20x objective lens.

143

The expression of total PrP and GFAP are altered by Anle138b in M1000-infected OBSCs To see if the effect Anle138b has on PrPRes production was associated with reduced PrP expression or affected the astrocytes, undigested samples used in Figure 5.2C were resolved by SDS-PAGE and immunoblot performed to detect total PrP and GFAP expression. Protein expression for DMSO-, 0.1 µM Anle138b- and 1 µM Anle138b-treated groups were expressed as fold change over untreated M1000-infected brain slices, following normalisation to a β-actin loading control (Figure 5.4).

Similar to previous observations (Figure 4.3), the expression of GFAP in OBSCs was found to change dramatically over time, with low expression at 1 DPS, which was elevated at all other time points (42, 56 and 70 DPS) (Figure 5.4). Interestingly, a strong, dose-dependent elevation in GFAP expression was observed in Anle138b-treated samples at 1 DPS compared to M1000 or DMSO-treated slices (GFAP/M1000-treated control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 1.05, 3.31 and 6.10 AU, respectively) (Figure 5.4). While at the later time points no obvious change in abundance of GFAP was found, inspection of the isoforms of GFAP present (in the low exposure blot) revealed a potential decrease in low molecular weight GFAP isoforms at 70 DPS related to Anle138b treatment (Figure 5.4A).

Western immunoblot analysis of PrP expression reveals important information of the total pool of PrP in prion-infected samples. Unlike PrPRes, which is solely composed of PrPSc, total PrP contains all PrP species in the sample: PrPC and PrPSc (both PrPRes and PK sensitive PrPSc). At 1 DPS, similar to GFAP, PrP in Anle138b-treated samples was elevated in a dose- dependent manner over the M1000 control (PrP/M1000-treated control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 1.12, 1.79 and 3.13 AU, respectively). Comparing the expression of PrP at other time points revealed a potential decrease in PrP at 42 DPS in the 0.1 µM Anle138b-treated samples compared to other groups (PrP expression/M1000-treated control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 0.95, 0.43 and 0.90 AU, respectively). At 56 DPS, both Anle138b-treated groups had decreased expression compared to the M1000 controls (PrP/M1000-treated control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 1.31, 0.89 and 0.42 AU, respectively) while at 70 DPS, equivalent PrP levels were found between +DMSO and +0.1 µM Anle138b-treated samples, with less expression seen in slices treated with +1.0 µM Anle138b (PrP expression/M1000-treated control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 0.88, 0.89 and 0.42 AU, respectively) (Figure 5.4B).

144

Figure 5.4: The effect of Anle138b on the expression of PrP and GFAP in viable M1000-infected OBSCs. Tga20 OBSCs exposed to M1000 prions were cultured in the presence of 0.1 or 1 µM Anle138b for up to 70 days. Untreated infected slices (M1000) and DMSO treated slices (+DMSO) were used as controls. (A) Pooled (6) slices were lysed and protein was resolved on SDS-PAGE and immunoblot performed to detect GFAP and PrP (with 03R19 antibody). The expression of β-actin was probed for as a loading control. (B) Quantification of proteins was expressed as fold-change over untreated M1000-infected OBSCs. n=1, 6 pooled slices. Due to n=1, no statistical analysis was performed.

145

Anle138b protects M1000-infected OBSCs from neuronal vulnerability Given that Anle138b was effective at reducing the abundance of PrPRes in M1000-infected OBSCs (Figure 5.2C), its effect on M1000-induced changes to neuronal populations was next assessed. Expression analysis of the panel of neuronal markers used previously (Figure 4.3 and Figure 4.7) was performed on all treatment groups at each time-point collected (1, 42, 56 and 70 DPS). Protein abundance was normalised to β-actin loading control and expression for DMSO-, 0.1 µM Anle138b- and 1.0 µM Anle138b-treated groups shown as fold change over untreated M1000-infected OBSCs (Figure 5.5).

At 1 DPS, while no difference was observed amongst the treatment groups with PSD-95 and synaptophysin, a small dose-dependent increase in Anle138b treated samples was observed in the expression of VAMP2 and SNAP25: [(VAMP2/M1000-infected control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 0.56, 0.71 and 0.95 AU, respectively) and (SNAP25/M1000-infected control for +DMSO, +0.1 µM Anle138b and +1.0 µM Anle138b; 0.52, 0.71 and 0.91 AU, respectively)].

No obvious difference was observed between any treatment groups with any neuronal marker at 42 and 56 DPS, however at 70 DPS the levels of all proteins were elevated in Anle138b- treated groups compared to untreated M1000-infected OBSCs. Quantifying this revealed a fold-change increase of ~2 for: VAMP2 (VAMP2/M1000-infected control for +0.1 µM Anle138b: 1.86 and +1.0 µM Anle138b: 1.99 AU), SNAP25 (SNAP25/M1000-infected control for +0.1 µM Anle138b: 1.82 and +1.0 µM Anle138b: 2.10 AU) and PSD-95 (PSD- 95/M1000-infected control for +0.1 µM Anle138b: 1.86 and +1.0 µM Anle138b: 1.46 AU). The fold-change in Anle138b-treated groups compared to untreated M1000-infected OBSCs was much higher for synaptophysin which were 14.43 and 25.40 for +0.1 and +1.0 µM Anle138b-treated samples, respectively (Figure 5.5B). As expected, the abundance of all neuronal markers in DMSO-treated slices at 70 DPS was equivalent to the untreated control group with a grouped mean fold change±SEM of 1.01±0.1522 (n=4). Hence this experiment showed M1000-induced neuronal loss was alleviated upon Anle138b treatment.

146

Figure 5.5: Anle138b alleviates M1000-induced neuronal vulnerability in OBSCs. Tga20 OBSCs exposed to M1000 prions were cultured in the presence of 0.1 or 1 µM Anle138b for up to 70 days. Untreated infected slices (M1000) and DMSO treated slices (+DMSO) were used as controls. (A) Pooled (6) slices were lysed and protein was resolved on SDS-PAGE and immunoblot performed to detect the expression of neuronal markers: PSD-95, synaptophysin, SNAP25 and VAMP2. β-actin was probed for as a loading control. (B) Quantification of proteins was expressed as fold-change over untreated M1000- infected OBSCs. n=1, 6 pooled slices. Due to n=1, no statistical analysis was performed.

147

Anle138b modulates PMCA-induced αsyn fibrillization In addition to being effective in prion-infected mice, Anle138b has been shown to be efficacious to pathogenic αsyn in PD and MSA animal models [345, 346, 670]. In Chapter 3, the αsyn PMCA was adapted to produce misfolded αsyn species under non-toxic conditions, which were then shown to modulate mitochondrial respiration in immortalised neuronal cells. The PMCA was also used as a tool to study the influence the lipid cardiolipin has on the fibrillization of αsyn. Using these assays, I sought to determine whether Anle138b modulates αsyn fibrillization and if so, whether the altered species formed caused changes to the effect PMCA-generated misfolded αsyn has on mitochondrial respiration.

To determine the effect Anle138b has on αsyn fibrillization, the molecule was dissolved in DMSO and added αsyn for a final concentration of 1 or 10 µM. Samples containing the equivalent volume of PBSN or DMSO were used as non-Anle138b treated controls. All samples were subjected to PMCA, while duplicate reactions were stored at -80oC for the duration of the PMCA reaction to serve as the non-PMCA, monomeric equivalents. After PMCA, all samples were subjected to PK digestion (a final concentration of 10 µg/mL) or mixed with an equivalent volume of PBSN. Samples were then resolved on SDS-PAGE and proteins identified either by western immunoblot (Figure 5.6A) or Coomassie brilliant blue staining to detect total protein (Figure 5.6B). Both detection systems showed αsyn fibrillization was altered in the presence of 10 µM Anle138b, where an increased abundance of oligomers was observed compared to all other treatment groups (Figure 5.6A-B). Strikingly, this translated to increased levels of PK resistant material following proteolysis (Figure 5.6A-B). Interestingly, the methods of detection (western immunoblot vs. Coomassie staining) showed differences in the abundance of species. This was most notable in the samples following PK digestion, where in PMCA-αsyn treated with 10 µM Anle138b, the immunoblot could detect large quantities of αsyn product in the high molecular weight range (Figure 5.6A). In contrast, these species were less distinct using Coomassie staining that detects total protein, which instead showed strong reactivity to small molecular weight species (Figure 5.6B). Digestion of samples stored at -80oC confirmed that at all concentrations used, Anle138b was not interfering with the PK digestion itself.

It was important to distinguish whether the altered αsyn species formed by Anle138b were able to propagate the changed phenotype characterised by enhanced PK resistance and concentration of low-order oligomers. To address this, a serial PMCA was performed where material from the eight reactions (that underwent a single round of PMCA) shown in Figure

148

5.6A-B were diluted and spiked into fresh monomeric αsyn. These were then subjected to another round of αsyn PMCA, prior to incubation with PK or PBSN following the protocol described above. Results of the western immunoblot showed the distinct species formed by 10 µM Anle138b were not able to transfer their phenotype upon serial rounds of PMCA (Figure 5.6C), hence indicating that the Anle138b-induced change imparted on αsyn requires the presence of the molecule to maintain its effect.

149

Figure 5.6: Anle138b increases PMCA-generated PK-resistant αsyn that do not propagate their altered phenotype through serial rounds of PMCA. Anle138b diluted in DMSO was added to αsyn for a final concentration of 1 or 10 µM. Control samples were αsyn added with an equivalent volume of buffer (PBSN) or DMSO. Samples were exposed to PMCA for 72 h. Duplicate samples kept at -80oC for the duration of the PMCA reaction were the monomeric samples. All samples were incubated with PK (10 µg/mL final concentration) or PBSN prior to being resolved on SDS-PAGE. αsyn products were detected via (A) western immunoblot using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123) or (B) Coomassie brilliant blue staining. (C) All samples (8) were diluted in PBSN and added to fresh αsyn and exposed to another 72 h round of PMCA. Following PMCA, reactions were digested in PK (or left in PBSN) and αsyn products detected by SDS-PAGE and western immunoblot as per the conditions detailed above. n=1.

150

To further investigate how Anle138b modulates αsyn misfolding, the next objective was to determine if Anle138b exerts its effect on the misfolding process or is able to directly modulate formed fibrils. For this, Anle138b (0.1, 1.0 or 10 µM) was either added to αsyn prior to the PMCA reaction (as per Figure 5.6, compound added at: 0 h) or it was added for 1 h to mature αsyn fibrils that had been exposed to PMCA for 71 h (compound added at: 71 h). Additional experimental groups included the replacement of Anle138b with PBSN or DMSO. Because a standard αsyn PMCA has a total process time of 72 h, the treatment of Anle138b to mature fibrils for 1 h after 71 h in the PMCA meant the incubation of Anle138b with mature fibrils lapsed when the fibrils treated with Anle138b from 0 h were finished in the PMCA. This allowed both treatment groups to be processed at the same time. This experimental regime was also applied to monomeric samples that were stored at -80oC.

At the completion of the 72 h process time, all samples were resolved by SDS-PAGE and species identified by western immunoblot. Similar to Figure 5.6A, 10 µM Anle138b increased the abundance of oligomeric αsyn species when Anle138b was added at 0 h, however a similar was not observed when Anle138b was added at 71 h (Figure 5.7A). This suggests that Anle138b acts by modulating αsyn fibrillization and does not have any effect on preformed species.

To see if treatments had an effect on the PK resistance of αsyn, PK digestion was performed on all 16 samples shown in Figure 5.7A. For this, samples were divided into PMCA and - 80oC groups and western immunoblot performed in the presence and absence of PK. Protein exposed to PMCA again showed increased PK resistant material when 10 µM Anle138b was added at 0 h, however treatment with the compound at 71 h failed to have a noticeable effect on the digestion patterns (Figure 5.7B). PK digestion on the samples stored at -80oC confirmed Anle138b treatment was not modulating the susceptibility of αsyn to PK digestion (Figure 5.7C).

An interesting observation in Figure 5.7A-B was that all samples that were removed from PMCA after 71 h had substantially fewer small oligomeric species than comparable groups that were exposed to PMCA for 72 h. This is likely due to the samples being left at RT and unsonicated for the duration of the incubation period with Anle138b and demonstrates that oligomers in this system are highly dynamic.

151

Figure 5.7: The ability of Anle138b to modulate PMCA-generated αsyn species is dependent on protein misfolding. (A) Anle138b diluted in DMSO was added to αsyn for a final concentration of 1 or 10 µM. Control samples were αsyn added with an equivalent volume of buffer (PBSN) or DMSO. αsyn was exposed to treatment groups either prior to a 72 h PMCA or after 71 h PMCA (compound added at 0 h or 71 h, respectively). Duplicate samples kept at -80oC for the duration of the PMCA reaction were the monomeric samples. (A) After 72 h, all samples were resolved on SDS-PAGE and αsyn detected using western immunoblot using αsyn-specific monoclonal antibody MJFR1 (amino acid specificity: 118-123). (B) PMCA- generated and (C) -80oC samples were digested in PK (or left in PBSN) and αsyn species detected by SDS-PAGE and western immunoblot as per the conditions detailed above. n=1.

152

No further modulation of mitochondrial respiration is seen in PMCA-generated misfolded αsyn species formed in the presence of Anle138b Having shown Anle138b is capable of producing an altered population of misfolded αsyn, I then sought to determine whether the Anle138b-specific alterations to fibrillization translated to changes in pathogenicity of misfolded αsyn at the mitochondria. Here, the experimental setup was as per that described in Chapter 3 (see Figure 3.9 and Figure 3.10) and included the following six treatment groups: PMCA, PMCA+DMSO, PMCA+Anle138b, -80oC, - 80oC+DMSO and -80oC+Anle138b. Because no obvious change to PMCA-induced αsyn fibrillization was seen in the low concentrations of Anle138b tested (0.1 and 1 µM), the highest concentration of Anle138b (10 µM) was used for the PMCA and -80oC+Anle138b groups in these experiments.

Following CRC analysis using the Seahorse XF Analyser, no significant changes were observed between PMCA+Anle138b and PMCA or PMCA+DMSO in any mitochondrial readout (Figure 5.8). While not significant, a constant trend was seen where DMSO- containing samples (both +DMSO and +Anle138b) enhanced the respiration compared to samples that were DMSO-free, suggesting that the concentration of DMSO used in this experiment was elevating mitochondrial respiration in this system.

Consistent with the results of Chapter 3, significant differences were found between PMCA and -80oC sample groups (Figure 5.8). Values, SEM and statistical significance of the differences found are shown in Table 5.1.

153

Figure 5.8: The addition of Anle138b in PMCA-induced αsyn misfolding does not alter the changes to mitochondrial respiration caused by misfolded αsyn. SH-SY5Y cells were incubated with PMCA-generated αsyn made in the presence of 10 µM Anle138b, DMSO only or untreated, or equivalent treatment groups of -80oC αsyn. Media-containing cells were plated into a well of a Seahorse XF24 plate pre-coated with Matrigel. The Seahorse XF Analyser measured mitochondria respiration by detecting changes in oxygen (referred to as the oxygen consumption rate; OCR) following the addition of pharmacological agents: oligomycin, CCCP, rotenone and antimycin-A. In doing so, the following parameters were measured: (A) basal OCR, (B) ATP synthesis, (C) max OCR, (D) complex I activity, (E) complex II activity and (F) non-mitochondrial respiration. The average of three to five wells was taken for each sample per experimental replicate. Data presented as mean±SEM (n=5-15). Statistical significance was examined by ANOVA and Tukey’s multiple comparisons test with a statistical criterion of 0.05. * p<0.05, **p<0.01

154

Comparison Values ± SEM (pmol/min) p-value Significance PMCA vs. -80oC 163.8±11.82 and 108.9±8.284 0.0069 **

Basal PMCA+DMSO vs. -80oC 182.6±24.81 and 108.9±8.284 0.0118 *

PMCA+Anle vs. -80oC 183.5±22.18 and 108.9±8.284 0.0104 *

ATP Synthesis PMCA vs. -80oC 206±18.65 and 139±12.2 0.0445 *

PMCA vs. -80oC 113.1±9.264 and 71.5±5.489 0.0115 *

Max PMCA+DMSO vs. -80oC 130.1±21.93 and 71.5±5.489 0.0118 *

PMCA+Anle vs. -80oC 130.1±20.04 and 71.5±5.489 0.0118 * Table 5.1: Values of significant differences identified in Figure 5.8. Statistical significance was examined by ANOVA and Tukey’s multiple comparisons test with a statistical criterion of 0.05. * p<0.05, **p<0.01

155

Discussion

The small molecule Anle138b has been reported to attenuate disease progression in various animal models of NDPs. In particular, it reduces the burden of PrPSc in prion infected mice and misfolded αsyn in transgenic animal models of PD [345], which in both cases is associated with longer pre-clinical incubation times and extended survival [345, 346]. The ability of Anle138b to modulate the fibrillization of both compounds made it an attractive tool to directly study the relevance of protein misfolding to the pathogenicity of PrP and αsyn in the systems developed in this thesis. Similar to published results [345], one of the major findings of this chapter was that Anle138b reduced PrPRes in OBSCs and protected slices from neurotoxicity associated with prion propagation. The effect of Anle138b in prion models has not been shown using OBSCs and hence the findings of this chapter revealed a novel approach to studying this molecule. Another important finding of this chapter was the effect Anle138b has on αsyn misfolding in the PMCA. Here it caused a striking modulation to the fibrillization of the protein, by producing species that are traditionally thought to confer pathogenicity: increasing oligomeric structures and enhancing PK resistance of the species formed. This observation is in contrast to the reported studies showing Anle138b reduces oligomer content of both PrPSc and pathogenic αsyn. In these studies, sucrose gradient centrifugation was employed to show Anle138b reduces the abundance of both high and low molecular weight PrPRes species in mouse prion diseased brain [345] and, using the same method of analysis, Anle138b-treatment to mutant A30P αsyn transgenic mice caused a shift in aggregated αsyn profiles within the brain in line with it reducing low molecular weight aggregates [345]. The reasons behind the contrasting findings between the published studies showing reduced αsyn oligomers following Anle138b treatment [345] and those presented in this chapter reporting an increase, are unknown. It may suggest Anle138b can have divergent effects on αsyn misfolding depending on the system used; with its modulation of the misfolding of recombinant protein using PMCA being divergent to its effect in transgenic mutant αsyn mice. Alternatively, technical considerations to the published studies may mean Anle138b is not effective at reducing oligomeric αsyn structures. The latter notion appears to be supported by the finding that Anle138b shows affinity to hydrophobic regions on the misfolded αsyn aggregate [675] and given that this is not considered to be a structural component to the highly dynamic oligomeric species, it may indicate Anle138b acts on higher order aggregates. The findings from this chapter are important to further elucidate a

156 mechanism of action of Anle138b, and to determine the effect such alterations have on the pathogenicity associated with the misfolding of PrP and αsyn.

The original paper detailing the identification of Anle138b and its effect on PrP misfolding is, to date, the only publication reporting its efficacy to pathogenic PrPSc [345] and hence its assessment using M1000-infected OBSCs in this chapter is important to contribute to the understanding of this molecule. While Anle138b caused reductions in PrPRes load at 70 DPS, it appeared to have a marginal effect at 56 DPS. This is interesting given that detecting PrPRes as a marker of PrPSc propagation in M1000-infected OBSCs in Chapter 4 (see Figure 4.4C) confirmed extensive propagation occurs by 56 DPS. Hence this data would seem to indicate Anle138b does not reduce the rate of propagation at early stages of disease. Further to this, at 56 DPS Anle138b exerts no change to the neuronal markers. While this observation may be interpreted as the compound offering no neuroprotection at this time-point, given that the expression of the neuronal markers was not determined at 56 DPS in M1000-infected OBSCs, the extent of neuronal vulnerability at this time-point is unknown. As such, it is possible that if Anle138b modulates the toxic agent at 56 DPS, the degree of toxicity caused by M1000 may be too low to detect any discernible difference between the treatment groups. Clarity on the effect of Anle138b at 56 DPS would come from analysis on M1000-infected OBSCs and NBH-treated controls to determine the status of neuronal health at this time- point.

This chapter did not define the structure or size of the PrPRes species formed in the presence of Anle138b. As mentioned in Chapter 1, the abundance of PK sensitive PrPSc can be inferred from the difference between PrPRes and total PrP. This may be relevant to the current data, where the preliminary analysis shows a decrease in PrPRes compared to control groups at 70 DPS in M1000-infected OBSCs treated with 0.1 µM Anle138b (Figure 5.2C) yet concomitant reductions in total PrP were not observed (Figure 5.4C). This may indicate Anle138b either increases PrPC, PK sensitive PrPSc or both in this system. It has been shown that PrPC expression is unchanged following Anle138b treatment in wild-type mice [345] and hence it is more likely that, similar to its effect on αsyn, Anle138b increases the abundance of oligomeric PK sensitive PrPSc structures. However a consideration to calculating PK sensitive PrPSc via the difference between PrPRes from total PrP using western immunoblot is that the values for PrPRes only reflects digested products that resolve through SDS-PAGE to the level of the misfolded monomeric protein. In this regard, if Anle138b acts as a stabiliser of protein aggregates, it may not necessarily decrease PrPSc abundance but produce PrPSc fibrils that are

157 longer and/or more resistant to proteolysis. Elongation of fibrils may therefore produce PrPRes species that are unable to resolve through SDS-PAGE after PK digestion. This is an interesting consideration given that while Anle138b was reported to decrease both small and large PrPRes aggregates separated by sucrose gradient [345], these were identified by standard PK digestion techniques identifying the PrPRes monomer and hence does not eliminate the possibility that larger undetected species exist in the low sucrose concentration fractions. While intriguing, this does not easily explain the reductions in PrPRes also found at higher sucrose gradient fractions that are reflective of oligomeric conformations, unless these small structures are likewise too stable to be degraded by PK. Regardless, to further elucidate the findings of this chapter, the samples processed in Figure 5.2C would benefit from further analysis that includes steps to disaggregate any potential large PrPRes aggregates following PK digestion and western immunoblot analysis of the entire resolved gel. The addition of such steps could potentially reveal a shift in PrPRes identified (either at the monomeric level or as aggregated species at a higher molecular weight range) in this assay due to large aggregates being further disassembled, allowing them to resolve through SDS-PAGE and be detected.

If Anle138b extends the length of PrPSc fibrils, it likely means the compound acts as a stabiliser of fibrillar aggregates. This mechanism of action would have significant benefits on disease pathogenesis where the stabilisation of PrPSc fibrils would reduce breakage of the fibril and hence reduce propagation by decreasing the pool of nucleation sites for PrPSc to recruit and misfold PrPC. This reduction in the rate of propagation could therefore be a major contributing factor to the protection from neurotoxicity exerted by Anle138b either by reducing the abundance of toxic species formed via propagation and/or slowing the rate of propagation itself which triggers neurotoxicity.

The notion that Anle138b may stabilise misfolded aggregates is supported by the αsyn data generated in this chapter showing Anle138b enhances the PK resistance of species formed. Unlike PrPSc, misfolded αsyn that harbour resistance to PK are not largely reduced to its monomeric conformation upon proteolysis and rather are identified as a smear in the high molecular weight region of the gel following electrophoresis. In the case of αsyn, the morphology of the Anle138b-treated species conferring enhanced PK resistance may be similar to untreated fibrils but more stable and/or they may be longer due to reduced breakages in the growing aggregate upon sonication. In this regard, it is possible that

158

Anle138b has the same mechanism of action to the misfolding of PrP and αsyn where it acts as a stabiliser of misfolded structures.

A consideration to this hypothesis is why Anle138b then causes an increase in αsyn oligomeric structures in the PMCA? Given that the misfolded αsyn species formed in PMCA exist in equilibrium with each other, it is probable that modulation of one species will have dramatic effects to those not directly targeted by Anle138b. Considering this, if Anle138b binds to mature αsyn structures, it may cause an overall reduction in the rate the aggregates extend from oligomers to higher ordered species. This would cause a bottle-neck at the level of oligomers which are continuously being produced, unaffected, from monomeric protein. Another possibility is that Anle138b is binding to and stabilising the oligomeric structures which would increase their abundance by blocking their ability to disassemble back to monomeric structures (a likely process that would occur when these species are in equilibrium) or it may trap any generation of oligomer, including off-pathway oligomers and hence the overall pool would still increase. While plausible, this latter scenario does not adequately explain the enhanced PK resistance of αsyn species in this system nor align with the data reporting Anle138b binds to hydrophobic pockets on the αsyn aggregate [675]. Regardless, the data presented in this chapter show the altered species produced by Anle138b are not able to propagate the changed phenotype, demonstrating Anle138b directly modulates protein misfolding. Taken together, the most likely scenario is that Anle138 stabilises αsyn aggregation by binding to mature species. While this appears to be the most appropriate explanation of the data in this chapter, further studies should be performed on Anle138b- treated αsyn species, where assays such as TEM and AUC will be useful to further characterise the species formed in the presence of this molecule.

An interesting observation in this chapter is the effect Anle138 has on PrP and GFAP expression at 1 DPS in M1000-infected OBSCs, with a strong dose dependent increase seen in both protein markers. This may indicate Anle138b directly modulates the expression of these proteins, however as discussed previously, there is no evidence this is the case for PrPC [345], and hence considered unlikely. A more likely possibility is that Anle138b protects slices from acute stress caused by exposure to M1000 prions. Given that PrP is highly expressed in neurons, the elevation in PrP could indicate less neuronal loss by Anle138b-treated slices at this time-point. Consistent with this, compared to DMSO-treated OBSCs, Anle138b-treated OBSCs exhibited a dose dependent elevation in two neuronal markers (VAMP2 and SNAP25) at this time-point (Figure 5.5). The markers that exhibited these elevations are important given

159 that no difference was identified in synaptophysin, which would be consistent with the confounding factor of general stress modulating this protein’s expression (discussed in Chapter 4). The notion that prions may cause acute damage in this model is supported by a study showing M1000 can be acutely toxic in cultured murine neuronal cells following infection. This response is dose-dependent, dependent on PrPC expression and associated with elevated ROS [559]. Consistent with M1000 prions causing acute toxicity in OBSCs, measuring the viability in NBH and M1000-infected samples revealed slightly higher levels of PI staining in the infected samples at 7 DPS (Figure 4.4B) which may indicate a lag in recovery in the infected samples associated with acute toxicity. This finding aligns with the GFAP expression. It has been shown that uninfected OBSCs rapidly recruit GFAP once in culture (Figure 4.3), and hence the elevated expression of GFAP at 1 DPS upon treatment with Anle138b in M1000- infected OBSCs may indicate infection delays this upregulation, which is rescued upon addition of the molecule. This should be further examined by thoroughly characterising the changes to GFAP upon exposure to M1000 compared to NBH-treated controls in early time- points. Nonetheless, collectively the preliminary data in this chapter supports the notion that Anle138b can protect OBSCs from early M1000-associated stress.

Given that Anle138b causes such a dramatic change to the population of αsyn species formed using PMCA, I sought to determine if this alteration in species translated to changes in the ability of misfolded αsyn to modulate mitochondria function in neuronal SH-SY5Y cells. No difference was seen between fibrils formed in the presence of Anle138b to those produced with DMSO, or left untreated (fibrils only). While this may indicate Anle138b does not target the toxic αsyn species, there are several confounding factors to consider in this system. Firstly, it appears that DMSO causes an elevation in OCR and hence the solvent may mask any small modulation caused by the altered αsyn profiles. Along with this, it is pertinent to note the PK digestion experiments do not give any information on the type/structure or abundance of the modulated species and hence the extent of these changes may be outside the threshold limits to cause any significant pathogenic change. Alternatively, it is possible in this system Anle138b is not targeting the pathogenic species, which given its ability to be protective to PrPSc shown in this chapter, would indicate either differential actions of Anle138b or a divergence in the species that cause toxicity between the proteins.

Figure 5.6B demonstrated for the first time that PMCA-generated αsyn species can be detected using an assay that identifies total protein (Coomassie brilliant blue stain). Because the same samples were also processed using western immunoblot (Figure 5.6A), this allowed for the

160 direct comparison of these assays to detect misfolded αsyn. Critically, similar to the western immunoblot assay, PMCA-generated species identified by Coomassie stain were found to exhibit a laddering effect of misfolded αsyn species with monomeric protein representing the most abundant species, with a progressive reduction in abundance with every addition of monomeric unit to the conformation. This confirms that the PMCA-generated species exist in equilibrium and this pattern previously observed using western immunoblot (Figures 3.2, 3.3, 3.4, 3.7 and 3.8) is not associated with any artefacts of antibody binding, such as reduced recognition due to the burying of epitopes into the aggregate. Comparing these figures also revealed differences in the distribution of species identified between these assays, where in contrast to the antibody detection system, the Coomassie staining revealed a greater abundance in small digestion products following proteolysis with PK and less high molecular weight species (Figure 5.6A-B). Given that the Coomassie staining detected the low molecular weight species, it is likely these species are cleavage products that do not contain the epitope the αsyn antibody recognises (MJFR1 antibody, amino acid resides: 118-123). Support for this comes from analysis of the ability of PK digestion products of fibrillar αsyn to be detected using the αsyn-directed antibodies NAC1, LB509 and syn102. The antibody NAC1 was raised against a peptide corresponding to residues 75-91 [676], LB509 is an antibody that has an epitope within residues 115-121/122 [677] and syn102, raised against recombinant αsyn, recognises the region of 131-140 of αsyn [228, 678]. Following PK digestion of fibrillar recombinant αsyn, one study reported NAC1 showed strong reactivity for low molecular weight products of αsyn (less than 10 kDa) [679]. In contrast, Syn102 could not detect anything lower than the size of the monomer (17 kDa) and LB509 had intermediate binding, with low expression of a few species between 10 and 19 kDa [679]. Hence this study demonstrated the PK protein fragments identified via SDS-PAGE and western immunoblot were αsyn that had undergone C-terminal cleavage. This aligns with the findings of this chapter where the antibody used, MJFR1 (residues: 118-123), was unable to detect these small species in Figure 5.6A. Taking this into account, the Coomassie stain used in Figure 5.6B reveals Anle138b dose dependently enhances the generation of C-terminally cleaved αsyn products following PK digestion. This again confirms Anle138b alters species and produces distinct αsyn fragments upon PK digestion. Given that Coomassie staining detected no low molecular weight products in the monomeric controls, the digestion products are not fragments of monomeric protein nor is Anle138b protecting the monomeric species from total digestion and hence it can be inferred that the cleavage products promoted by Anle138b are derived from a misfolded conformation. Given that Figure 5.6A shows Anle138b treatment (10 µM) causes an elevation of PK resistant 161 species in the high molecular weight range detected via western immunoblot, it is likely the small cleavage products identified via Coomassie are cleavage products from these species. Alternatively, they may represent oligomeric species that harbour enhanced PK resistance. Again, this supports the notion that Anle138b stabilises and/or extends the length of misfolded species.

Several lines of evidence demonstrates that Anle138b has differential effects depending on the type of transgenic animal model it is used in. In a PD animal model expressing mutant A30P αsyn in neurons, Anle138b improves cognition and reduces protein deposition [345], which translates to extended survival time [346]. It is also efficacious to PD animal models generated using MPTP [345]. However, in a MSA animal model, Anle138b only exerts partial protection to functional readouts and gives no protection to neuronal vulnerability [670]. This may indicate that the efficacy of Anle138b is dependent of the status of the misfolded αsyn and/or the cell-type used in the particular experiment. The latter consideration is particularly relevant given that recently Anle138b has been shown to cause cell death in melanoma cells [680]. Melanoma is a malignancy that, compared to other cancers, PD patients and their relatives have an increased risk of developing [681-686]. Despite this, no genetic association has been found associating the two pathogeneses [687-689]. Immortalised melanoma-derived cells, which highly express αsyn and contain both monomeric and oligomeric species, are highly susceptible to cell death following treatment with Anle138b [680]. Here, Anle138b caused morphological changes and reduced proliferation of the cells, which was linked with dysregulation in autophagy (detected by immunoblot of p62/SQSTM1) and reduced mitochondrial membrane potential [680]. These effects were clearly demonstrated using concentrations of Anle138b (10 µM) that were lower than those used in neuronal immortalised cultured cells which showed no signs of toxicity [345]. While it is intriguing to postulate the effect Anle138b has on αsyn in melanoma cells, a limitation to this study was that there was no evidence showing Anle138b directly associated with αsyn, and hence potential off-target effects by Anle138b need to be considered for this cell line. This is particularly relevant given that Anle138b has been shown to have divergent effects on autophagy in various cell types and disease models; in melanoma cells Anle138b modulates p62/SQSTM1 whereas in mouse model of tauopathy no changes in autophagy markers (p62/SQSTM1 and LC3) were found [671]. These findings may indicate that the effect of Anle138b is cell type specific however it may also indicate Anle138b has different mechanisms of action between proteins or conformations of proteins. While both scenarios may explain the contrasting weak protective capacity of Anle138b in MSA [670]

162 compared to PD animal models [345], it clearly demonstrates more work is needed to confirm the effect Anle138b has on αsyn misfolding and the resulting pathogenic profiles this imparts on the misfolded species formed.

Conclusion

The aim of this chapter was to use Anle138b as a tool to study the effect of protein misfolding on the pathogenic properties of PrP and αsyn. Consistent with published studies, Anle138b was found to be an effective anti-prion compound in M1000-infected OBSCs by reducing PrPRes and protecting slices from neurotoxicity associated with prion propagation. Interestingly, it also identified a potential role of Anle138b in protecting slices from acute toxicity caused by M1000 prions. In the αsyn PMCA, Anle138b modulated αsyn species by increasing the concentration of oligomeric αsyn and enhancing the PK resistance of species formed. However, these did not translate to any functional alterations to the effect of misfolded αsyn on mitochondrial respiration. Taken together, the best interpretation of this data is that Anle138b acts as a stabiliser of misfolded structures. In the case of PrP, this results in a protection from PrPSc- induced toxicity, however the evidence of a therapeutic mode-of-action of Anle138b to pathogenic αsyn is less clear and requires further research to fully ascertain its effect.

163

CHAPTER 6: GENERAL DISCUSSION

Overview of main findings

Neuronal vulnerability and the deposition of misfolded PrP (PrPSc) and αsyn in the CNS are hallmarks of prion and synucleinopathy disorders, respectively. An intriguing feature of PrPSc is that it is the sole [170] or main component [178, 690, 691] of prions that are the causative, infectious agent in prion disease. Here, critical features of prion disease have been directly attributed to PrPSc including its ability to propagate via nucleation-dependent conformational misfolding of PrPC (Figure 1.1) and be transmissible, causing disease when inoculated into susceptible animals. The notion of strain variability is also a critical feature of PrPSc, whereby the clinical disease phenotype is transcribed by the conformation of the misfolded protein. Neurotoxicity is also a major feature associated with PrPC misfolding, however its generation in disease is not well understood. Today mounting evidence suggests αsyn may act in the same capacity as PrP in terms of its ability to misfold, propagate and be transmissible causing disease in certain animals [3]. The extent of the similarities identified between PrP and αsyn have led to the suggestion in the scientific field that αsyn is, by definition, a prion [11, 14, 269].

The main objective of this thesis was to explore the pathogenic features of PrPC and αsyn misfolding, with a particular focus on the way these proteins contribute to the generation of neurotoxicity. Similar to PrPSc, the toxic mechanisms associated with αsyn misfolding are ill defined and hence further understanding on this facet of disease is important to further our general understanding on these proteins and facilitate further discrimination of the prion concept. Also, knowledge of the neurotoxic agent/s at the molecular level that result in disease will be essential for the development of effective therapeutic strategies for these currently incurable disorders. In this thesis these proteins were studied following the development of robust systems to produce misfolded αsyn protein and test its toxicity in cultured cells, and a model prion disease using organotypic brain slice cultures.

164

The major findings presented in this thesis were: i) PMCA-generated misfolded αsyn binds to cardiolipin and modulates mitochondrial respiration in neuronal cells ii) Organotypic brain slice cultures are amendable to infection by mouse-adapted human prions that cause disease associated with the propagation of PrPSc and neurotoxicity iii) Anle138b modulates the misfolding of both proteins, which in the case of PrP, leads to the protection of organotypic brain slice cultures from prion-induced neurotoxicity.

One of the strongest aspects of this thesis was the development and use of powerful techniques to study PrP and αsyn. A limitation to studying the toxicity of αsyn is that traditional techniques to produce misfolded forms of this protein generates a homogenous population of misfolded protein. In human disease, numerous misfolded species exist and hence the production and use of defined species of a specific conformation is unlikely an appropriate representation of in vivo derived, disease-associated αsyn. To this end, one of the initial aims of the project was to develop a system to produce a heterogeneous population of misfolded αsyn species. For this I used the prion assay, PMCA, which has been shown to produce a range of various sized αsyn species [275]. A limitation to the reported protocol was a high concentration of detergent in the conversion buffer, and hence after establishment of the published protocol I adapted the system to produce misfolded protein under non-toxic conditions. Successful optimisation of this system allowed experimental analysis on the pathogenicity of misfolded species formed, where PMCA-generated αsyn was found to bind to cardiolipin and target the mitochondria in cultured neuronal cells. Here they caused hyperactive respiration of mitochondria (Figure 3.10) without any associated functional deficit (Figure 3.11). This finding is in contrast to the mitochondrial dysfunction reported in aggressive transgenic models expressing αsyn in vitro and in vivo [456, 457, 459, 461, 518] and post mortem Parkinson’s disease (PD) brain [330-332, 462]. However the results of this thesis align with the recent observation of mitochondrial hyperactivity in immortalised lymphocytes derived from ante mortem PD patients [527] and given the robust results of my data in a system using naïve wild-type cells, the data presented in this thesis are strong evidence that hyperactive mitochondria is a major, relevant pathogenic process that occurs in the brain in early PD. This finding has far-reaching implications that suggest that mitochondria adapt to αsyn-related insult for the majority of disease course in PD and, if this hyperactivity is associated with elevated ROS production in neurons, it indicates that hyperactivity rather than dysfunction is the cause of oxidative damage in the brain in PD.

165

Next, organotypic brain slice cultures were used to develop a model of prion disease. While in vivo animal systems develop authentic disease upon prion infection, they carry limitations associated with experimental cost and are labour intensive. This presents the need for alternate systems to study these diseases. Prion infected organotypic brain slice cultures have been shown to develop all critical features of disease [391] and hence I developed this system as part of my project. The capacity of organotypic brain slice cultures to be infected by human-derived strains has not been reported previously and so the use and characterisation of disease caused by the mouse-adapted human familial prion strain, M1000, in this thesis is a novel finding that confirms infection is achievable with this strain type. Critically M1000-infected slices developed pertinent features of disease including propagation of PrPRes and neurotoxicity. The development of neurotoxicity was measured using a panel of markers targeting various components of synaptic transmission to thoroughly study what components of neuronal health are most vulnerable in this system.

Finally, the relevance of protein misfolding to the pathogenic properties of PrP and αsyn was studied using the small molecule Anle138b. Anle138b has been shown to be protective to both prion infected animals [345] and transgenic models of PD [345, 346]. While initial reports suggest Anle138b is an oligomer modulator [345], more recent evidence indicates it may instead recognise higher order aggregates [675]. As such, the work using Anle138b in this thesis had two main components i) to further define the mechanism of action of Anle138b in PrP and αsyn misfolding and ii) to identify the consequence of this effect on pathogenic readouts developed previously in the thesis. Critically, Anle138b was shown to modulate PrP misfolding by reducing PK-resistant PrP, which led to a suppression of neurotoxicity associated with prion propagation in M1000-infected organotypic brain slice cultures (Figure 5.5). Interestingly, indications of a protective role of Anle138b to acute PrPSc toxicity were also identified (Figure 5.4). In the αsyn PMCA, Anle138b increased the presence of oligomeric structures and PK resistance of misfolded αsyn, but did not alter the effect misfolded αsyn has on mitochondrial respiration. Assuming a conserved mechanism of action between the two proteins, this work demonstrates that PK resistance and/or enhancement of specific types of misfolded proteins does not necessarily correlate with pathogenicity. Collectively, the most appropriate interpretation of the data is that Anle138b acts to stabilise protein structures during the misfolding and aggregation of both proteins. However, the point or structure in the aggregation pathway that it targets has yet to be determined. Regardless, the data presented in this thesis and by others [345] show strong evidence the change Anle138b imparts on PrP

166 misfolding is neuroprotective, however its efficacy to misfolded αsyn is less clear. In this regard, the effect of Anle138b on αsyn misfolding and critically, whether the altered species are non-toxic, less toxic or even more toxic requires further attention.

The similarities and differences in how PrP and αsyn misfold

The hypothesis of this thesis was that similarities exist between PrP and αsyn misfolding, which may be exploited to further our understanding of disease pathogenesis. Work presented in this thesis confirms that the misfolding of both PrP and αsyn is associated with altered neuronal function. The demonstration that Anle138b modulates the type and/or abundance of misfolded species for both proteins further supports the hypothesis of this thesis and argues that the pathways of misfolding are similar between these proteins. However, while this may be applicable to the morphology of structures formed and hence pertain the ability of molecules that recognise a conformation-specific target to modulate protein misfolding, it is pertinent to note that the way these proteins misfold remains starkly different. This is clearly demonstrated via the use of the PMCA in this thesis. While αsyn is able to spontaneously misfold in a solution composed exclusively of recombinant protein (Figure 3.2), effective PrP misfolding requires the addition of a preformed seed (Figure 3.1 and [168, 169]). This is a strong divergent feature that distinguishes PrP from αsyn whereby the small, largely disordered αsyn protein requires a low activation energy to misfold, whereas the well-defined structural conformation of PrPC means much higher activation energies are required for misfolding to occur. As such, although the physical characteristics of misfolded structures may be the same between PrP and αsyn, it does not mean the mechanism of misfolding or the stability of the misfolded structures is equivalent. This concept may also be relevant to the way these species exist in equilibrium with each other. While the use of the PMCA demonstrates sonication-induced breakage of fibrils accelerates the formation of PrPSc and αsyn aggregates, the breakage of PrPSc aggregates into smaller species that occurs in disease once the fibrils reach a certain length is likely mechanistically dissimilar to misfolded αsyn that can easily expand or contract into higher or lower ordered conformations. Further evidence for alterations in the biophysical properties between misfolded αsyn and PrPSc comes from the finding that in contrast to PrPSc, misfolded αsyn is markedly easier to decontaminate [276]. As such, in addition to using compounds that modulate fibrillization, further assessment into the ways these proteins misfold and their ability

167 to be pathogenic should be performed by studying the ability of these structures to disassemble. This would give important insight into the way species exist in equilibrium and how certain species are generated, which could be leveraged to modulate the generation of pathogenic conformations.

An additional relevant link between PrP and αsyn misfolding described in this thesis is the role of lipids to this process. The misfolding of pure preparations of PrP and the generation of prions that are transmissible can be achieved by the addition of essential co-factors; lipids and polyanionic species [178]. This thesis revealed the lipid cardiolipin accelerates αsyn misfolding (Figure 3.8) and hence for both proteins lipids may facilitate protein misfolding. In terms of PrP, the misfolding of the protein in the absence of lipids produces species that cause disease associated with a long incubation period and atypical pathology [170, 171] and because of this lipids are considered to be a necessary component of a transmissible prion. While the fact that pure preparations of misfolded recombinant αsyn protein can cause clinical disease in susceptible rodents [257, 261] indicates that lipids are not indispensable for misfolded αsyn to confer pathogenicity, it is yet to be established whether these species generates a different type of disease compared to that which generates organically in an organism. As mentioned previously in this thesis, it would be interesting to determine whether lipids impart any alterations to the αsyn species formed or whether it simply acts as an accelerant of fibrillization. Regardless, the role of lipids to PrP misfolding is likely to facilitate protein misfolding by acting as a scaffold for enhanced protein-protein interactions and, given the aforementioned differences in activation energies, its presence is likely more apposite to PrP misfolding.

The toxic mechanisms of misfolded neurodegenerative proteinopathy- associated proteins: direct mechanisms, via an interaction with normal monomeric protein or both?

This thesis supports the concept that NDP-associated proteins can cause toxicity via numerous mechanisms. A critical consideration to the ability of these proteins to be neurotoxic is whether misfolded structures directly cause damage to the target cell, or whether their toxicity involves the propagation of misfolded protein and/or interactions between protein species. In prion disease, several critical studies confirm that a cell must express PrPC in order for it to succumb to prion-induced toxicity [166, 294]. This argues that PrPSc alone is not wholly toxic and instead

168 the development of toxicity somehow relies on an interaction between PrPC and PrPSc. Potential mechanisms for this in prion disease were described in Chapter 1 [355].

The resolute requirement of normal monomeric protein for PrPSc-induced toxicity may not be as steadfast for other NDP-associated proteins. In this thesis, several mechanisms of misfolded αsyn-induced toxicity have been proposed that appear to not require normal monomeric αsyn. The finding in this thesis that misfolded αsyn, which modulates mitochondrial respiration, also associates with cardiolipin suggests that misfolded αsyn may directly damage the mitochondria via this interaction. This is in line with studies that show interactions between misfolded αsyn and lipid membranes can cause disruptions to the membrane and toxicity [326, 338, 491]. Assuming the effect of αsyn at the mitochondria identified in this thesis is an upstream mechanism that causes neurotoxicity in PD, this would demonstrate a critical divergence in the abilities of these proteins to cause toxicity. However, given that SH-SY5Y cells express endogenous αsyn and its monomeric state is still present in the PMCA-generated heterogeneous mixed population, it is challenging to determine the influence of normal monomeric αsyn in this system. The finding that disease-associated αsyn can propagate within wild-type and transgenic αsyn A53T mutant mice causing disease [11, 255-262], appears to suggest that αsyn – like prions, has the capacity to induce toxicity associated with its propagation. However, these studies do not clarify whether this propagation produces a defined species that directly cause damage or whether the act of propagation itself is associated with toxicity. Further analysis on the influence of normal monomeric protein in these systems should come from studies modulating αsyn expression in established diseased states to determine the contribution of an interaction between misfolded species and normal monomeric αsyn to the development of neurotoxicity.

Given that widespread neurotoxicity occurs late in prion disease, the overarching toxic pathways associated with PrPSc-induced damage in established disease are likely associated with chronic propagation. However, this thesis and other published data show evidence that prions can also be acutely toxic. Anle138b appeared to protect organotypic brain slice cultures from early M1000-induced damage (Figure 5.4), which is consistent with the finding that cultured cells exposed to M1000 prions develop PrP-dependent acute cell death associated with ROS activation [559]. ROS activation is not considered to a major contributor to neurotoxicity in advanced prion disease: astrocytes rather than neurons have elevated ROS products in end- stage prion-infected mice and CJD brain [692] and over the course of prion disease, oxidative stress is highest at the first identification of PrPRes and lower at terminal disease [449]. Taken

169 together, these studies support the concept that PrPSc can employ multiple mechanisms of toxicity. Variations between the mechanisms of acute toxicity of prions and that which develops in association with chronic propagation may be associated with the status of the cell, where the initial insult of high levels PrPSc into naïve cells may cause significant damage prior to the clearance of the majority of the seeds and the activation of adaptive mechanisms by the cell. While these adaptive mechanisms may initially be beneficial, they may also be a driver of these toxic effects in progressed disease. Considering this, it would be interesting to compare the relevance of key potential toxic agents in both acute and chronic disease states. This includes membrane bound PrPC, and the UPR; both which have been proposed as neurotoxic agents in disease [372, 390] and are foreseeable players in acute toxicity. Regardless, the finding that prions also appear to modulate astrocyte up-regulation in organotypic brain slice cultures (Figure 5.4) indicates that acute toxicity of prions may not be neuron specific. This is an important observation and indicates the lower PrP expression found astrocytes is not protective, nor is any other potential neuron-specific agent/s required for this toxicity. Taking this into consideration, further analysis of the health of CNS cell types in organotypic brain slice cultures in early prion infection would be useful, including the use of organotypic brain slice cultures derived from global and cell type-specific knockout PrP transgenic mice to establish the relevance of PrP expression in this system. Taken together, work presented in this thesis and published in the literature seem to support the notion that NDP-associated proteins can elicit damage via multiple mechanisms. However, it appears the central requirement of normal monomeric protein (like PrPC) to the generation of toxicity is, for now, prion specific.

Toxic mechanisms of αsyn in synucleinopathies

This thesis proposes that the mitochondrion is a major site targeted by misfolded αsyn. While it is well established that this organelle is implicated in PD [693], it is interesting that it is not considered a major pathogenic feature in other synucleinopathies, such as multiple system atrophy (MSA) [694]. Given that αsyn aggregation occurs in distinct compartments between these diseases and the type of protein aggregation that forms (Lewy bodies vs. glial cytoplasmic inclusions) are morphologically distinct, it may suggest numerous mechanisms of toxicity exist based on the conformation and/or locality of the misfolded protein. This is particularly relevant given that neurons, the cell type αsyn is found aggregated in in PD, have an extraordinarily

170 high and variable metabolic rate. As such, this cell type may be more susceptible to any minor modulation of mitochondrial respiration caused by misfolded αsyn. In such a scenario, it is intriguing to interpret early mitochondrial damage as an upstream mechanism that causes the aggregation of αsyn as Lewy bodies within neurons to sequester misfolded species away from the mitochondria. However, while the locality of misfolded αsyn within neurons may be linked to altered mitochondrial functioning in PD, dementia with Lewy bodies (DLB) also presents with Lewy bodies and similar to MSA, it is also not traditionally considered to be associated with mitochondrial deficits. Indeed a systematic review finds only weak evidence of an association of the mitochondria in this disease [695]. Insight into the differences between PD and DLB may come from considering the cell type susceptible to toxicity in these diseases, where PD is distinguished by the selective loss of dopaminergic neurons of the substania nigra. It is well established that dopamine enhances the production of β-sheet negative, oligomeric αsyn [445, 696, 697] and hence if this is the pathogenic conformation that modulates mitochondrial respiration, it may generate a larger quantity of the pathogenic species that target the mitochondria in these systems. Further research should be conducted into the capacity of αsyn to be pathogenic in dopaminergic systems. This may be achieved using dopamine- expressing cell types, such as the dopaminergic rat neuronal cell line N27 [698] or creating an enhanced dopaminergic environment in SH-SY5Y cells via their differentiation using retinoic acid which upregulates the expression of dopaminergic markers and enhances the cell’s susceptibility to dopaminergic toxins [699].

An additional interesting distinction between the synucleinopathies is the development of dementia. While only a small subset of PD patients present with this symptom; referred to as PD with dementia (PDD), it is a frequent symptom of MSA and DLB [222]. Given that dementia is also a common feature of prion and Alzheimer’s disease (AD) it may suggest the type of neuronal vulnerability is the same in these NDPs. In AD, synapse loss correlates with the onset and severity of dementia [700-702] and various synaptic proteins predict cognitive decline in both AD and DLB [703]. Similarly, the loss of VAMP2 and monomeric αsyn correlate with the duration of dementia in DLB, PDD and AD [704] and hence synapse loss is likely an important component to NDPs which present with dementia. It is unknown how αsyn could variably target synapses between the synucleinopathies. A potential explanation to this distinction is the existence of strain variability: a notion that is supported by studies showing discrete populations of αsyn have defined pathogenic properties [11, 14, 253, 257, 270]. In this capacity, a defined type of misfolded αsyn may preferentially cause damage to

171 the synapse. Given that αsyn is localised at the pre-synapse, it should also be considered that synapse loss may be triggered by αsyn associated with propagation in this area of the cell and/or localisation of αsyn away from the synapse upon misfolding. Support for the latter factor as a relevant theory comes from studies showing αsyn plays an important role at the synapse; αsyn has been shown to directly bind VAMP2 and enhance SNARE complex assembly in vitro and in vivo [128] and ablation of the protein causes age-dependent impairment of complex assembly [705]. The involvement of αsyn at the SNARE complex requires the presence of the lipid regulator arachidonic acid [705] and interestingly it has been shown αsyn may modulate synaptic transmission by cross-bridging the lipid phosphatidylserine to VAMP2 to facilitate SNARE-dependent vesicle docking [706]. Hence it could be expected that a loss of αsyn in these areas would cause drastic effects to neuronal transmission and health.

Further to this, MSA and prion disease are both defined as NDPs that are rapid in onset of clinical disease and progression, and hence it is intriguing to speculate that the act of propagation itself rather than direct toxicity of misfolded αsyn may be a unifying process of these disorders. In this regard, this could reveal a distinction to pathogenic misfolding of αsyn in PD, where the rate of propagation may be slow enough that the cell is able to replenish lost protein and maintain the locality of monomeric αsyn at the synapse and therefore other neurotoxic mechanisms present, such as enhanced ROS production associated with mitochondrial hyperactivity [527] that makes it the most damaging toxic mechanism in dopaminergic cells. This scenario would explain the divergent neurotoxic processes of αsyn within the spectrum of synucleinopathies where in one system propagation prevails as a major upstream toxic process, whereas in another oxidative damage caused by direct interactions of misfolded αsyn with target organelles is more relevant. While it is important to consider the divergent toxic mechanisms presented amongst the synucleinopathies, similar downstream toxic mechanisms may still exist. For example, PERK activation is found in both PD and MSA, where its activation co-localises with regions of αsyn deposition [437, 707]. Given that this mechanism is also activated in certain prion diseases [372], this suggests that similar neurotoxic pathways may present in a range of diseases with distinctly different pathological and clinical profiles. Collectively it is likely that numerous mechanisms are able to cause damage when a protein misfolds in the cell, and as such the difficult task will be to identify the dominant pathway/s in any given biological system. Future research should focus

172 on the contribution of each potential toxic agent to identify the main driver/s of toxicity in disease and hence, reveal the best target for therapeutic intervention on a case-by-case basis.

Towards effective therapeutic strategies for NDPs

An appropriate extension to the considerations of pathogenic mechanisms of PrP and αsyn misfolding described in this thesis is to discuss key therapeutic strategies to treat disease. Given the centrality of protein misfolding to the activation of toxic pathways by NDP- associated proteins, numerous therapeutic strategies involve targeting the protein prone to misfold to inhibit its ability to fibrillize and/or reduce the pool of monomeric protein and thus limit propagation. Discussed in Chapter 5, many compounds that recognise PrP and modulate its fibrillization are efficacious in prion disease [355]. The majority of these compounds appear to target PrPSc, which offers the major advantage of avoiding potential unknown side effects associated with the loss of the monomeric protein. However, due to the adaptive nature of PrP misfolding, a common issue using these compounds is the generation of drug resistant prions. This resistance has been shown to extend to the activity of other anti-prion agent’s potency which target the same structure; for example, both cpd-B and quinacrine are less effective to IND24-resistant prions [664]. This may be relevant to the efficacy of Anle138b, and should be examined to see if it can occur in any capacity to pathogenic PrP and αsyn misfolding. Nonetheless it reveals an important consideration when targeting misfolded conformations.

An alternate approach to treat these diseases is by modulating the expression of the NDP- associated protein. In the case of PrP, this is demonstrated by the ability to rescue mice with established prion disease by ablating the expression of PrPC [166] and given that PrP knockout mice have no overt phenotype [30], it suggests targeting PrPC may be a viable therapeutic strategy. This has been achieved by employing RNAi, where a study using lentiviral mediated shRNA against PrPC in prion infected mice showed a single treatment of shRNA injected into each hippocampus when the first signs of neuropathological change occurred prevented the onset of behavioural deficits and reduced spongiform degeneration and neuronal loss [708]. These neuropathological changes were observed in sites surrounding the injection site and despite the small region of brain exposed to the RNAi, the mean

173 survival of shRNA treated mice was 23.5% higher than untreated animals [708]. This demonstrates that even small reductions in PrPC expression are likely to be therapeutic in prion disease. While this may be an appropriate strategy to treat prion disease, it is important to note that such an equivalent may not be beneficial for disease systems presenting with misfolded αsyn, where the efficacy of targeting the monomeric protein may hinge on αsyn’s importance at the synapse.

Nonetheless, targeting the gene encoding the relevant protein will likely be a suitable strategy to treat familial forms of disease. Allele-specific siRNA show great potential where preliminary studies have been performed for familial forms of NDPs; AD [709], PD [710] ALS [711] and Huntington’s disease [712] with promising results. Additionally the CRISPR/Cas-9 system has been utilised to correct a mutation associated with cystic fibrosis in patient-derived cultured intestinal cells, which when assayed in organoid culture, showed a rescue of disease-associated defects [713]. These encouraging studies for treating single-gene hereditary defects support the potential for tailored patient-specific therapy, which is considered a strong therapeutic strategy. However, whilst being an attractive approach to use, limitations to the use of RNAi as therapies are that they do not readily cross the BBB and have the potential to trigger off-target effects and/or toxicity due to over-loading the RNAi pathway. Various alterations to naked RNAi may improve these parameters, such as designing siRNA duplexes to decrease off-target effects, or fusing it to neuron specific markers such as rabies virus glycoprotein to aid its delivery to the CNS. Alternatively, other methods of knockdown may reduce toxicity; miRNA appear generally better tolerated than siRNA or shRNA [714] and synthetic miRNA sequences have been used to target PRNP mRNA to produce PrPC knockdown in cultured cells [715]. Additionally many studies have investigated using siRNA delivery vectors [716]. In particular, the naturally occurring RNA carriers exosomes show great potential in allowing the safe transport of siRNA over large distances where intravenous injections of exosomes fused to rabies virus glycoprotein are targeted to the CNS where they deliver siRNA cargo specifically to neurons [717]. This system has been successfully utilised to reduce levels of BACE1 in mouse brain by over 60% in the striatum, cortex and midbrain of wild-type mice [717]. In another study, exosomes were efficiently targeted to the brain via the intranasal route [718], suggesting a non-invasive mode of administration for RNAi molecules.

174

Protein-directed antibodies A well-used method to target NDP-associated proteins is via antibody administration to immunize against the protein or modulate its fibrillization. In cultured cells and cell-free systems, monoclonal antibodies that target PrP abolish prion propagation [719-724]. Their effectiveness extends to in vivo animal models where two antibodies ICSM35 and ICSM18, that recognise residues 93-105 and 143-153 of PrP respectively, have been shown to completely ameliorate propagation of peripheral prions in infected mice when treatment starts before PrPSc reaches the brain [725]. This striking result is the only therapeutic strategy to date shown to clinically cure prion disease in PrP-expressing mice. ICSM18 and ICSM35 are non-toxic to mice [726] and appear to act by stabilising the conformation of PrPC [727]. A humanised version of ICSM18, referred to as PRN100, has been produced which is non-toxic at intravenous doses of up to 200 mg/kg in cynomolgus macaques [307], giving evidence for its safe use in human trials. Given its large size, the efficacy of PrP-directed antibodies decreases once prions reach the CNS and as such, its ability to cross the BBB is considered a large therapeutic challenge. However, the observation that intraperitoneal injection of the PrP antibody 6D11 (amino acid specificity: 93-109) in APP/PS1 transgenic mice results in a reduction in behavioural deficits, gives proof-of-principle evidence that peripherally administered antibodies can reach the CNS [728]. In agreement to this, studies investigating Aβ-directed antibodies as a therapeutic for AD, showed small amounts of peripherally- administered antibodies were found in the brain [729, 730].

Similar to PrP, antibodies have been shown to be effective to other NDP models where immunotherapy shows efficacy in vitro and in vivo to pathogenic αsyn [266, 731-735], Aβ [195, 729, 730], tau [736-738] and SOD1 [739]. In progress or completed phase I trials to treat PD and MSA in humans involve immunisation with αsyn-directed antibodies, such as PRX002 (Prothena; ClinicalTrials.gov Identifier: NCT02157714) and BIIB054 (Biogen; ClinicalTrials.gov Identifier: NCT02459886) or via active immunisation against αsyn using PD01A and PD03A (AFFiRiS; ClinicalTrials.gov Identifiers: NCT02270489, NCT02618941, NCT02758730, NCT01568099, NCT02267434). While most of these initial studies have been tolerance tests and assessment of CNS penetration of the antibodies in vivo, they show promising results that facilitates further analysis of their efficacy in human disease [740, 741].

Despite the plethora of data showing protein-directed antibodies as efficacious, concern is raised by reports showing that under certain conditions PrP antibodies can be neurotoxic

175

[389, 390]. The antibodies D13 and P, which align to the hydrophobic region of PrP (residues 95-105), have been reported to be toxic to wild-type mice, with widespread neuronal loss observed 24 hours post hippocampal injection [389]. In the case of D13, this has been extensively studied using a range of doses in wild-type mice [742], however the findings of these experiments are in contrast to others which report it to be non-toxic [726]. Similarly 6D11 cures cells of prion infection and prolongs the incubation period of disease in vivo [743]. Antibodies targeting α-helix 1 (amino acid residues: 145-155) and 3 (amino acid residues: 200-219) of the globular domain of PrP are highly toxic to PrP-overexpressing organotypic brain slice cultures [390, 392], however the α-helix 1 region aligns to the binding region of the non-toxic, therapeutic ICSM18 [725, 726]. An antibody, W226, which recognises a proximal region and has a 59% homology to ICSM18, is able to prevent prion replication in cultured cells, yet exerts no protection to mice from peripheral prion infection [744]. The reason for the contrasting effects of these antibodies has not been elucidated. It may be a matter of dose threshold: either of the concentration of the antibody or expression of available PrPC to bind. Additional factors may include the species specificity, binding affinity, mode-of-action, or the differences in epitope recognition of the antibody. Enlightenment on regions to target for effective intervention may come from clinical trials underway aimed at identifying PrP autoantibodies in carriers of PRNP mutations (ClinicalTrials.gov Identifier: NCT02837705). Nonetheless, the observations that under certain conditions PrP-directed antibodies have the capacity to be toxic should be addressed prior to its use in humans.

Combined treatment strategies Given that it is likely multiple toxic mechanisms may be employed by misfolded protein, a one-dimensional treatment regime may not be the most effective approach to treat NDPs. In this regard, an attractive strategy to treat disease is via a combined treatment strategy that targets various aspects considered relevant to the generation of toxicity. For example, the use of agents that reduce the expression of normal monomeric protein alongside ones that recognise misfolded structures and modulate fibrillization may be highly effective and should be tested in appropriate animal systems. This treatment regime may also alleviate the development of undesired facets of compound use such as the generation of drug-resistant prions. Likewise, the use of drugs that target potential downstream toxic pathways, such as the UPR, will likely be beneficial in the disease types that present with this form of toxicity.

176

The consideration of using combined treatment strategies, of course, does not diminish the great need for further understanding into the neurotoxicity of NDP-associated proteins, particularly in terms of the involvement of strain variability and propagation to the activation of various toxic mechanisms. In doing so, the hope is that this will lead to discrimination of the relevance of these features to neurotoxicity to allow tailored patient-specific therapies that target the relevant, dominant neurotoxic agent/s for the specific condition.

Conclusion

In conclusion, the work presented in this thesis has demonstrated the misfolding of αsyn and PrP is a pathogenic feature of their respective disorders that is able to cause significant alterations to cells exposed to misfolded protein. While many similarities were identified with regards to how these proteins misfold, this thesis equally demonstrated critical differences in the role of these proteins in disease. This reflects the complex nature of NDPs and is an important consideration when extending the term ‘prion’ to other NDP-associated proteins. Further understanding of the relationship and interplay between critical facets of disease will be essential to further our understanding of these diseases and identify the precise pathogenic mechanisms that can be targeted for effective therapeutic intervention.

177

REFERENCES

1. Prusiner, S.B., Novel proteinaceous infectious particles cause scrapie. Science, 1982. 216(4542): p. 136-144.

2. Prusiner, S.B., Prions. Proceedings of the National Academy of Sciences of the United States of America, 1998. 95(23): p. 13363-13383.

3. Ugalde, C.L., et al., Pathogenic mechanisms of prion protein, amyloid-β and α- synuclein misfolding: the prion concept and neurotoxicity of protein oligomers. J Neurochem, 2016. 139(2): p. 162-180.

4. Braak, H., et al., Staging of brain pathology related to sporadic Parkinson's disease. Neurobiology of Aging, 2003. 24(2): p. 197-211.

5. Braak, H. and E. Braak, Neuropathological staging of Alzheimer-related changes. Acta Neuropathologica, 1991. 82(4): p. 239-259.

6. Thal, D.R., et al., Phases of Aβ-deposition in the human brain and its relevance for the development of AD. Neurology, 2002. 58(12): p. 1791-800.

7. Lee, S. and H.-J. Kim, Prion-like mechanism in amyotrophic lateral sclerosis: are protein aggregates the key? Experimental Neurobiology, 2015. 24(1): p. 1-7.

8. Holmes, B.B. and M.I. Diamond, Prion-like properties of tau protein: the importance of extracellular tau as a therapeutic target. The Journal Of Biological Chemistry, 2014. 289(29): p. 19855-19861.

9. Costanzo, M. and C. Zurzolo, The cell biology of prion-like spread of protein aggregates: mechanisms and implication in neurodegeneration. The Biochemical Journal, 2013. 452(1): p. 1-17.

10. Brundin, P., R. Melki, and R. Kopito, Prion-like transmission of protein aggregates in neurodegenerative diseases. Nature reviews. Molecular cell biology, 2010. 11(4): p. 301-307.

11. Prusiner, S.B., et al., Evidence for α-synuclein prions causing multiple system atrophy in humans with parkinsonism. Proceedings Of The National Academy Of Sciences Of The United States Of America, 2015. 112(38): p. E5308-E5317.

12. Stöhr, J., et al., Distinct synthetic Aβ prion strains producing different amyloid deposits in bigenic mice. Proc Natl Acad Sci U S A, 2014. 111(28): p. 10329-34.

13. Watts, J.C., et al., Serial propagation of distinct strains of Aβ prions from Alzheimer's disease patients. Proceedings of the National Academy of Sciences of the United States of America, 2014. 111(28): p. 10323-10328.

14. Woerman, A.L., et al., Propagation of prions causing synucleinopathies in cultured cells. Proceedings Of The National Academy Of Sciences Of The United States Of America, 2015. 112(35): p. E4949-E4958.

178

15. Prusiner, S.B., A unifying role for prions in neurodegenerative diseases. Science, 2012. 336(6088): p. 1511-1513.

16. Vey, M., et al., Subcellular colocalization of the cellular and scrapie prion proteins in caveolae-like membranous domains. Proceedings of the National Academy of Sciences of the United States of America, 1996. 93(25): p. 14945-14949.

17. Bendheim, P.E., et al., Nearly ubiquitous tissue distribution of the scrapie agent precursor protein. Neurology, 1992. 42(1): p. 149-156.

18. Ford, M.J., et al., Selective expression of prion protein in peripheral tissues of the adult mouse. Neuroscience, 2002. 113(1): p. 177-192.

19. Stahl, N., et al., Identification of glycoinositol phospholipid linked and truncated forms of the scrapie prion protein. Biochemistry, 1990. 29(38): p. 8879-8884.

20. Stahl, N., et al., Scrapie prion protein contains a phosphatidylinositol . Cell, 1987. 51(2): p. 229-240.

21. Riek, R., et al., NMR structure of the mouse prion protein domain PrP(121-231). Nature, 1996. 382(6587): p. 180-182.

22. Zahn, R., et al., NMR solution structure of the human prion protein. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(1): p. 145-150.

23. Palmer, M.S., et al., Deletions in the prion protein gene are not associated with CJD. Human Molecular Genetics, 1993. 2(5): p. 541-544.

24. Krasemann, S., et al., Prion disease associated with a novel nine octapeptide repeat insertion in the PRNP gene. Molecular Brain Research, 1995. 34(1): p. 173-176.

25. Mead, S., et al., Inherited prion disease with six octapeptide repeat insertional mutation - molecular analysis of phenotypic heterogeneity. Brain, 2006. 129: p. 2297- 2317.

26. Pietrini, V., et al., Creutzfeldt-Jakob disease with a novel extra-repeat insertional mutation in the PRNP gene. Neurology, 2003. 61(9): p. 1288-1291.

27. Campbell, T.A., et al., A prion disease with a novel 96-base pair insertional mutation in the prion protein gene. Neurology, 1996. 46(3): p. 761-766.

28. Schatzl, H.M., et al., Prion protein gene variation among primates. Journal of Molecular Biology, 1995. 245(4): p. 362-374.

29. Haraguchi, T., et al., Asparagine-linked glycosylation of the scrapie and cellular prion proteins. Archives of Biochemistry and Biophysics, 1989. 274(1): p. 1-13.

30. Bueler, H., et al., Normal development and behavior of mice lacking the neuronal cell-surface PrP protein. Nature, 1992. 356(6370): p. 577-582.

179

31. Zhang, C.C., et al., Prion protein is expressed on long-term repopulating hematopoietic stem cells and is important for their self-renewal. Proceedings of the National Academy of Sciences of the United States of America, 2006. 103(7): p. 2184-2189.

32. Alais, S., et al., Functional mechanisms of the cellular prion protein (PrPC) associated anti-HIV-1 properties. Cellular and Molecular Life Sciences, 2012. 69(8): p. 1331-1352.

33. Leblanc, P., D. Baas, and J.L. Darlix, Analysis of the interactions between HIV-1 and the cellular prion protein in a human cell line. Journal of Molecular Biology, 2004. 337(4): p. 1035-1051.

34. Roberts, T.K., et al., PrPC, the cellular isoform of the human prion protein, is a novel biomarker of HIV-associated neurocognitive impairment and mediates neuroinflammation. American Journal of Pathology, 2010. 177(4): p. 1848-1860.

35. Gabus, C., et al., The prion protein has RNA binding and chaperoning properties characteristic of nucleocapsid protein NCp7 of HIV-1. Journal of Biological Chemistry, 2001. 276(22): p. 19301-19309.

36. Nikles, D., et al., Circumventing tolerance to the prion protein (PrP): vaccination with PrP-displaying particles induces humoral immune responses against the native form of cellular PrP. Journal of , 2005. 79(7): p. 4033-4042.

37. Lotscher, M., et al., Induced prion protein controls immune-activated in the mouse spleen. Plos One, 2007. 2(11).

38. Nakamura, Y., et al., Transfection of prion protein gene suppresses coxsackievirus B3 replication in prion protein gene-deficient cells. Journal of General Virology, 2003. 84: p. 3495-3502.

39. Korom, M., et al., A proautophagic antiviral role for the cellular prion protein identified by infection with a herpes simplex virus 1 ICP34.5 mutant. Journal of Virology, 2013. 87(10): p. 5882-5894.

40. Durig, J., et al., Differential constitutive and activation-dependent expression of prion protein in human peripheral blood leucocytes. British Journal of Haematology, 2000. 108(3): p. 488-496.

41. Li, R.L., et al., The expression and potential function of cellular prion protein in human lymphocytes. Cellular Immunology, 2001. 207(1): p. 49-58.

42. Tobler, I., et al., Altered circadian activity rhythms and sleep in mice devoid of prion protein. Nature, 1996. 380(6575): p. 639-642.

43. Huber, R., T. Deboer, and I. Tobler, Sleep deprivation in prion protein deficient mice and control mice: genotype dependent regional rebound. Neuroreport, 2002. 13(1): p. 1-4.

44. Huber, R., T. Deboer, and I. Tobler, Prion protein: a role in sleep regulation? Journal of Sleep Research, 1999. 8: p. 30-36.

180

45. Thompsett, A.R., et al., High affinity binding between copper and full-length prion protein identified by two different techniques. Journal of Biological Chemistry, 2005. 280(52): p. 42750-42758.

46. Pauly, P.C. and D.A. Harris, Copper stimulates endocytosis of the prion protein. Journal of Biological Chemistry, 1998. 273(50): p. 33107-33110.

47. Brown, D.R., et al., The cellular prion protein binds copper in vivo. Nature, 1997. 390(6661): p. 684-687.

48. Haigh, C.L., et al., PrPC-related signal transduction is influenced by copper, membrane integrity and the α cleavage site. Cell Research, 2009. 19(9): p. 1062- 1078.

49. Siggs, O.M., et al., Disruption of copper homeostasis due to a mutation of Atp7a delays the onset of prion disease. Proceedings of the National Academy of Sciences of the United States of America, 2012. 109(34): p. 13733-13738.

50. Canello, T., et al., Copper is toxic to PrP-ablated mice and exacerbates disease in a mouse model of E200K genetic prion disease. Neurobiology of Disease, 2012. 45(3): p. 1010-1017.

51. Quaglio, E., R. Chiesa, and D.A. Harris, Copper converts the cellular prion protein into a protease-resistant species that is distinct from scrapie isoform. Journal of Biological Chemistry, 2001. 276(14): p. 11432-11438.

52. Shyu, W.C., et al., Hypoglycemia enhances the expression of prion protein and heat- shock protein 70 in a mouse neuroblastoma cell line. Journal of Neuroscience Research, 2005. 80(6): p. 887-894.

53. Shyu, W.C., et al., Heat shock modulates prion protein expression in human NT-2 cells. Neuroreport, 2000. 11(4): p. 771-774.

54. Shyu, W.C., et al., Overexpression of PrPC by adenovirus-mediated gene targeting reduces ischemic injury in a stroke rat model. Journal of Neuroscience, 2005. 25(39): p. 8967-8977.

55. Shyu, W.C., et al., Hyperbaric oxygen enhances the expression of prion protein and heat shock protein 70 in a mouse neuroblastoma cell line. Cellular and Molecular Neurobiology, 2004. 24(2): p. 257-268.

56. White, A.R., et al., Prion protein-deficient neurons reveal lower glutathione reductase activity and increased susceptibility to hydrogen peroxide toxicity. American Journal of Pathology, 1999. 155(5): p. 1723-1730.

57. Klamt, F., et al., Imbalance of antioxidant defense in mice lacking cellular prion protein. Free Radical Biology and Medicine, 2001. 30(10): p. 1137-1144.

58. Bertuchi, F.R., et al., PrPC displays an essential protective role from oxidative stress in an astrocyte cell line derived from PrPC knockout mice. Biochemical and Biophysical Research Communications, 2012. 418(1): p. 27-32.

181

59. Milhavet, O., et al., Prion infection impairs the cellular response to oxidative stress. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(25): p. 13937-13942.

60. Arlt, S., et al., Increased lipid peroxidation in cerebrospinal fluid and plasma from patients with Creutzfeldt-Jakob disease. Neurobiology of Disease, 2002. 10(2): p. 150-156.

61. Guentchev, M., et al., Evidence for oxidative stress in experimental prion disease. Neurobiology of Disease, 2000. 7(4): p. 270-273.

62. Criado, J.R., et al., Mice devoid of prion protein have cognitive deficits that are rescued by reconstitution of PrP in neurons. Neurobiology of Disease, 2005. 19(1-2): p. 255-265.

63. Nishida, N., et al., Prion protein is necessary for latent learning and long-term memory retention. Cellular and Molecular Neurobiology, 1997. 17(5): p. 537-545.

64. Bremer, J., et al., Axonal prion protein is required for peripheral maintenance. Nature Neuroscience, 2010. 13(3): p. 310-U9.

65. Küffer, A., et al., The prion protein is an agonistic ligand of the G protein-coupled receptor Adgrg6. Nature, 2016. 536(7617): p. 464-+.

66. Kang, J., et al., The precursor of Alzheimer's disease amyloid A4 protein resembles a cell-surface receptor. Nature, 1987. 325(6106): p. 733-736.

67. Haass, C. and D.J. Selkoe, Cellular processing of β-Amyloid Precursor Protein and the genesis of Amyloid β-Peptide. Cell, 1993. 75(6): p. 1039-1042.

68. Borchelt, D.R., et al., Familial Alzheimer's disease-linked presenilin 1 variants elevate Aβ1-42/1-40 ratio in vitro and in vivo. Neuron, 1996. 17(5): p. 1005-1013.

69. Scheuner, D., et al., Secreted amyloid β-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease. Nature Medicine, 1996. 2(8): p. 864-870.

70. Burdick, D., et al., Assembly and aggregation properties of synthetic Alzheimer's A4/β amyloid peptide analogs. Journal of Biological Chemistry, 1992. 267(1): p. 546-554.

71. Miravalle, L., et al., Amino-terminally truncated Aβ peptide species are the main component of cotton wool plaques. Biochemistry, 2005. 44(32): p. 10810-10821.

72. Saito, T., et al., Potent amyloidogenicity and pathogenicity of Aβ43. Nature Neuroscience, 2011. 14(8): p. 1023-U120.

73. Seubert, P., et al., Isolation and quantification of soluble Alzheimer's β-peptide from biological fluids. Nature, 1992. 359(6393): p. 325-327.

74. Haass, C., et al., Amyloid β-peptide is produced by cultured cells during normal metabolism. Nature, 1992. 359(6393): p. 322-325.

182

75. Dawson, G.R., et al., Age-related cognitive deficits, impaired long-term potentiation and reduction in synaptic marker density in mice lacking the β-amyloid precursor protein. Neuroscience, 1999. 90(1): p. 1-13.

76. Muller, U., et al., Behavioral and anatomical deficits in mice homozygous for a modified β amyloid precursor protein gene. Cell, 1994. 79(5): p. 755-765.

77. Tremml, P., et al., Neurobehavioral development, adult openfield exploration and swimming navigation learning in mice with a modified β-amyloid precursor protein gene. Behavioural Brain Research, 1998. 95(1): p. 65-76.

78. Tremml, P., et al., Enriched early experiences of mice underexpressing the β-amyloid precursor protein restore spatial learning capabilities but not normal openfield behavior of adult animals. Genes Brain and Behavior, 2002. 1(4): p. 230-241.

79. Senechal, Y., P.H. Kelly, and K.K. Dev, Amyloid precursor protein knockout mice show age-dependent deficits in passive avoidance learning. Behavioural Brain Research, 2008. 186(1): p. 126-132.

80. Doyle, E., et al., Intraventricular infusions of antibodies to amyloid-β-protein precursor impair the aquisition of a passive avoidance reponse in the rat. Neuroscience Letters, 1990. 115(1): p. 97-102.

81. Huber, G., et al., Involvement of amyloid precursor protein in memory formation in the rat - an indirect antibody approach. Brain Research, 1993. 603(2): p. 348-352.

82. Puzzo, D., et al., Picomolar amyloid-β positively modulates synaptic plasticity and memory in hippocampus. Journal of Neuroscience, 2008. 28(53): p. 14537-14545.

83. Morley, J.E., et al., A physiological role for amyloid-β protein: enhancement of learning and memory. Journal of Alzheimers Disease, 2010. 19(2): p. 441-449.

84. , L.D., et al., The production of amyloid β peptide is a critical requirement for the viability of central neurons. Journal of Neuroscience, 2003. 23(13): p. 5531-5535.

85. Minniti, A.N., et al., Intracellular amyloid formation in muscle cells of Aβ-transgenic Caenorhabditis elegans: determinants and physiological role in copper detoxification. Molecular Neurodegeneration, 2009. 4.

86. Murphy, D.D., et al., Synucleins are developmentally expressed, and α-synuclein regulates the size of the presynaptic vesicular pool in primary hippocampal neurons. Journal of Neuroscience, 2000. 20(9): p. 3214-3220.

87. Jakes, R., M.G. Spillantini, and M. Goedert, Identification of 2 distinct synucleins from human brain. Febs Letters, 1994. 345(1): p. 27-32.

88. Spillantini, M.G., A. Divane, and M. Goedert, Assignment of human α-synuclein (SNCA) and β-synuclein (SNCB) genes to 4q21 and 5q35. Genomics, 1995. 27(2): p. 379-381.

183

89. Ueda, K., et al., Molecular-cloning of cDNA-encoding an unrecognized component of amyloid in Alzheimer disease. Proceedings of the National Academy of Sciences of the United States of America, 1993. 90(23): p. 11282-11286.

90. Masliah, E., et al., Altered presynaptic protein NACP is associated with plaque formation and neurodegeneration in Alzheimer's disease. American Journal of Pathology, 1996. 148(1): p. 201-210.

91. Han, H.Y., P.H. Weinreb, and P.T. Lansbury, The core Alzheimer's peptide NAC forms amyloid firbils which seed and are seeded by β-amyloid - Is NAC a common trigger or target in neurodegenerative disease? Chemistry & Biology, 1995. 2(3): p. 163-169.

92. Iwai, A., et al., Non-Aβ component of Alzheimer's disease amyloid (NAC) is amyloidogenic. Biochemistry, 1995. 34(32): p. 10139-10145.

93. El-Agnaf, O.M.A., et al., Aggregates from mutant and wild-type α-synuclein proteins and NAC peptide induce apoptotic cell death in human neuroblastoma cells by formation of β-sheet and amyloid-like filaments. Febs Letters, 1998. 440(1-2): p. 71- 75.

94. Weinreb, P.H., et al., NACP, a protein implicated in Alzheimer's disease and learning, is natively unfolded. Biochemistry, 1996. 35(43): p. 13709-13715.

95. Uversky, V.N., J. Li, and A.L. Fink, Evidence for a partially folded intermediate in α- synuclein fibril formation. Journal of Biological Chemistry, 2001. 276(14): p. 10737- 10744.

96. Dedmon, M.M., et al., Mapping long-range interactions in α-synuclein using spin- label NMR and ensemble molecular dynamics simulations. Journal of the American Chemical Society, 2005. 127(2): p. 476-477.

97. Wu, K.P., et al., Characterization of conformational and dynamic properties of natively unfolded human and mouse α-synuclein ensembles by NMR: implication for aggregation. Journal of Molecular Biology, 2008. 378(5): p. 1104-1115.

98. Maiti, N.C., et al., Raman spectroscopic characterization of secondary structure in natively unfolded proteins: α-synuclein. Journal of the American Chemical Society, 2004. 126(8): p. 2399-2408.

99. Fauvet, B., et al., α-synuclein in central nervous system and from erythrocytes, mammalian cells, and Escherichia coli exists predominantly as disordered monomer. Journal of Biological Chemistry, 2012. 287(19): p. 15345-15364.

100. Bartels, T., J.G. Choi, and D.J. Selkoe, α-synuclein occurs physiologically as a helically folded tetramer that resists aggregation. Nature, 2011. 477(7362): p. 107- U123.

101. Wang, W., et al., A soluble α-synuclein construct forms a dynamic tetramer. Proceedings of the National Academy of Sciences of the United States of America, 2011. 108(43): p. 17797-17802.

184

102. Anderson, J.P., et al., Phosphorylation of Ser-129 is the dominant pathological modification of α-synuclein in familial and sporadic Lewy body disease. Journal of Biological Chemistry, 2006. 281(40): p. 29739-29752.

103. Okochi, M., et al., Constitutive phosphorylation of the Parkinson's disease associated α-synuclein. Journal of Biological Chemistry, 2000. 275(1): p. 390-397.

104. Pronin, A.N., et al., Synucleins are a novel class of substrates for G protein-coupled receptor kinases. Journal of Biological Chemistry, 2000. 275(34): p. 26515-26522.

105. Ellis, C.E., et al., α-synuclein is phosphorylated by members of the Src family of protein-tyrosine kinases. Journal of Biological Chemistry, 2001. 276(6): p. 3879- 3884.

106. Nakamura, T., et al., Activated Fyn phosphorylates α-synuclein at tyrosine residue 125. Biochemical and Biophysical Research Communications, 2001. 280(4): p. 1085- 1092.

107. Hasegawa, M., et al., Phosphorylated α-synuclein is ubiquitinated in α- synucleinopathy lesions. Journal of Biological Chemistry, 2002. 277(50): p. 49071- 49076.

108. Tofaris, G.K., et al., Ubiquitination of α-synuclein in Lewy bodies is a pathological event not associated with impairment of proteasome function. Journal of Biological Chemistry, 2003. 278(45): p. 44405-44411.

109. Dorval, V. and P.E. Fraser, Small ubiquitin-like modifier (SUMO) modification of natively unfolded proteins tau and α-synuclein. Journal of Biological Chemistry, 2006. 281(15): p. 9919-9924.

110. Giasson, B.I., et al., Oxidative damage linked to neurodegeneration by selective α- synuclein nitration in synucleinopathy lesions. Science, 2000. 290(5493): p. 985-989.

111. Citron, B.A., et al., Protein crosslinking, tissue transglutaminase, alternative splicing and neurodegeneration. Neurochemistry International, 2002. 40(1): p. 69-78.

112. Junn, E., et al., Tissue transglutaminase-induced aggregation of α-synuclein: implications for Lewy body formation in Parkinson's disease and dementia with Lewy bodies. Proceedings of the National Academy of Sciences of the United States of America, 2003. 100(4): p. 2047-2052.

113. Andringa, G., et al., Tissue trans glutaminase catalyzes the formation of α-synuclein crosslinks in Parkinson's disease. Faseb Journal, 2004. 18(3): p. 932-+.

114. Uversky, V.N., et al., Methionine oxidation inhibits fibrillation of human α-synuclein in vitro. Febs Letters, 2002. 517(1-3): p. 239-244.

115. George, J.M., The synucleins. Genome Biology, 2002. 3(1).

116. Iwai, A., et al., The precursor protein of non-Aβ component of Alzheimer's disease amyloid is a presynaptic protein of the central-nervous-system. Neuron, 1995. 14(2): p. 467-475.

185

117. Nakajo, S., et al., Localization of the phosphoneuroprotein 14 (PNP 14) and its messengers RNA expression in rat brain determined by immunocytochemistry and in situ hybridization. Molecular Brain Research, 1994. 27(1): p. 81-86.

118. Buchman, V.L., et al., Persyn, a member of the synuclein family, has a distinct pattern of expression in the developing nervous system. Journal of Neuroscience, 1998. 18(22): p. 9335-9341.

119. Ji, H.J., et al., Identification of a breast cancer-specific gene, BCSG1, by direct differential cDNA sequencing. Cancer Research, 1997. 57(4): p. 759-764.

120. Biere, A.L., et al., Parkinson's disease-associated α-sylnuclein is more fibrillogenic than β- and γ-synuclein and cannot cross-seed its homologs. Journal of Biological Chemistry, 2000. 275(44): p. 34574-34579.

121. Galvin, J.E., et al., Axon pathology in Parkinson's disease and Lewy body dementia hippocampus contains α-, β-, and γ-synuclein. Proceedings of the National Academy of Sciences of the United States of America, 1999. 96(23): p. 13450-13455.

122. Surguchov, A., et al., Synucleins in ocular tissues. Journal of Neuroscience Research, 2001. 65(1): p. 68-77.

123. Geng, X.H., et al., α-synuclein binds the K-ATP channel at insulin-secretory granules and inhibits insulin secretion. American Journal of Physiology-Endocrinology and Metabolism, 2011. 300(2): p. E276-E286.

124. Sharma, S.K., et al., Insulin-degrading enzyme prevents α-synuclein fibril formation in a nonproteolytical manner. Scientific Reports, 2015. 5.

125. Rodriguez-Araujo, G., et al., α-synuclein elicits glucose uptake and utilization in adipocytes through the Gab1/PI3K/Akt transduction pathway. Cellular and Molecular Life Sciences, 2013. 70(6): p. 1123-1133.

126. Rodriguez-Araujo, G., et al., Low α-synuclein levels in the blood are associated with insulin resistance. Scientific Reports, 2015. 5.

127. Burre, J., M. Sharma, and T.C. Sudhof, α-synuclein assembles into higher-order multimers upon membrane binding to promote SNARE complex formation. Proceedings of the National Academy of Sciences of the United States of America, 2014. 111(40): p. E4274-E4283.

128. Burre, J., et al., α-synuclein promotes SNARE-complex assembly in vivo and in vitro. Science, 2010. 329(5999): p. 1663-1667.

129. Choi, B.K., et al., Large α-synuclein oligomers inhibit neuronal SNARE-mediated vesicle docking. Proceedings of the National Academy of Sciences of the United States of America, 2013. 110(10): p. 4087-4092.

130. Park, S.M., et al., Distinct roles of the N-terminal-binding domain and the C- terminal-solubilizing domain of α-synuclein, a molecular chaperone. Journal of Biological Chemistry, 2002. 277(32): p. 28512-28520.

186

131. Chandra, S., et al., α-synuclein cooperates with CSPα in preventing neurodegeneration. Cell, 2005. 123(3): p. 383-396.

132. Zhu, M., et al., α-synuclein can function as an antioxidant preventing oxidation of unsaturated lipid in vesicles. Biochemistry, 2006. 45(26): p. 8135-8142.

133. Latchoumycandane, C., et al., Protein kinase Cδ is a key downstream mediator of manganese-induced apoptosis in dopaminergic neuronal cells. Journal of Pharmacology and Experimental Therapeutics, 2005. 313(1): p. 46-55.

134. Hashimoto, M., et al., α-synuclein protects against oxidative stress via inactivation of the c-Jun N-terminal kinase stress-signaling pathway in neuronal cells. Journal of Biological Chemistry, 2002. 277(13): p. 11465-11472.

135. Ostrerova, N., et al., α-synuclein shares physical and functional homology with 14-3-3 proteins. Journal of Neuroscience, 1999. 19(14): p. 5782-5791.

136. Chen, R.H.C., et al., α-synuclein membrane association is regulated by the Rab3a recycling machinery and presynaptic activity. Journal of Biological Chemistry, 2013. 288(11): p. 7438-7449.

137. Jin, H.J., et al., α-synuclein negatively regulates protein kinase Cδ expression to suppress apoptosis in dopaminergic neurons by reducing p300 histone acetyltransferase activity. Journal of Neuroscience, 2011. 31(6): p. 2035-2051.

138. Peng, X.M.M., et al., α-synuclein activation of protein phosphatase 2A reduces tyrosine hydroxylase phosphorylation in dopaminergic cells. Journal of Cell Science, 2005. 118(15): p. 3523-3530.

139. Yang, W., et al., α-synuclein overexpression increases phospho-protein phosphatase 2A levels via formation of calmodulin/Src complex. Neurochemistry International, 2013. 63(3): p. 180-194.

140. Cuillé, J. and P.L. Chelle, La maladie dite tremblante du mouton est-elle inocuable? Comptes Rendus Hebdomadaires Des Seances De L Academie Des Sciences, 1936. 203: p. 1552-1554.

141. Chandler, R.L., Encephalopathy in mice produced by inoculation with scrapie brain material. Lancet, 1961. 1(7191): p. 1378-9.

142. Gajdusek, D.C., C.J. Gibbs, and M. Alpers, Experimental transmission of a Kuru-like syndrome to chimpanzees. Nature, 1966. 209(5025): p. 794-796.

143. Gibbs, C.J., et al., Creutzfeldt-Jakob disease (Spongiform Encephalopathy) - transmission to chimpanzee. Science, 1968. 161(3839): p. 388-&.

144. Masters, C.L., D.C. Gajdusek, and C.J. Gibbs, Jr., Creutzfeldt-Jakob disease virus isolations from the Gerstmann-Sträussler syndrome with an analysis of the various forms of amyloid plaque deposition in the virus-induced spongiform encephalopathies. Brain, 1981. 104(3): p. 559-88.

187

145. Collinge, J. and M.P. Alpers, Introduction. The end of kuru: 50 years of research into an extraordinary disease. Philosophical Transactions of the Royal Society B: Biological Sciences, 2008. 363(1510): p. 3607-3612.

146. Will, R.G., et al., A new variant of Creutzfeldt-Jakob disease in the UK. Lancet, 1996. 347(9006): p. 921-925.

147. Hill, A.F., et al., The same prion strain causes vCJD and BSE. Nature, 1997. 389(6650): p. 448-450.

148. Bernoulli, C., et al., Danger of accidental person-to-person transmission of Creutzfeldt-Jakob disease by surgery. Lancet, 1977. 1(8009): p. 478-479.

149. El Hachimi, K.H., et al., Putative neurosurgical transmission of Creutzfeldt-Jakob disease with analysis of donor and recipient: agent strains. Comptes Rendus De L Academie Des Sciences Serie Iii-Sciences De La Vie-Life Sciences, 1997. 320(4): p. 319-328.

150. Miyashita, K., et al., Creutzfeldt-Jakob disease in a patient with a cadaveric dural graft. Neurology, 1991. 41(6): p. 940-941.

151. Willison, H.J., A.N. Gale, and J.E. McLaughlin, Creutzfeldt-Jakob disease following cadaveric dura-mater graft. Journal of Neurology Neurosurgery and Psychiatry, 1991. 54(10): p. 940-940.

152. Duffy, P., et al., Possible person-to-person transmission of Creutzfeldt-Jakob disease. New England Journal of Medicine, 1974. 290(12): p. 692-693.

153. Heckmann, J.G., et al., Transmission of Creutzfeldt-Jakob disease via a corneal transplant. Journal of Neurology Neurosurgery and Psychiatry, 1997. 63(3): p. 388- 390.

154. Llewelyn, C.A., et al., Possible transmission of variant Creutzfeldt-Jakob disease by blood transfusion. Lancet, 2004. 363(9407): p. 417-421.

155. Peden, A.H., et al., Preclinical vCJD after blood transfusion in a PRNP codon 129 heterozygous patient. Lancet, 2004. 364(9433): p. 527-9.

156. Cochius, J.I., et al., Creutzfeldt-Jakob disease in a recipient of human pituitary- derived gonadotropin. Australian and New Zealand Journal of Medicine, 1990. 20(4): p. 592-593.

157. Powell-Jackson, J., et al., Creutzfeldt-Jakob disease after administration of human growth-hormone. Lancet, 1985. 2(8449): p. 244-246.

158. Brown, P., et al., Iatrogenic Creutzfeldt-Jakob disease - the waning of an era. Neurology, 2006. 67(3): p. 389-393.

159. Brown, P., et al., latrogenic Creutzfeldt-Jakob disease, final assessment. Emerging Infectious Diseases, 2012. 18(6): p. 901-907.

188

160. Borchelt, D.R., et al., Scrapie and cellular prion proteins differ in their kinetic of synthesis and topology in cultured-cells. Journal of Cell Biology, 1990. 110(3): p. 743-752.

161. Pan, K.M., et al., Conversion of α-helices into β-sheets features in the formation of the scrapie prion proteins. Proceedings of the National Academy of Sciences of the United States of America, 1993. 90(23): p. 10962-10966.

162. Li, J.L., et al., Darwinian of prions in cell culture. Science, 2010. 327(5967): p. 869-872.

163. Silveira, J.R., et al., The most infectious prion protein particles. Nature, 2005. 437(7056): p. 257-261.

164. Simoneau, S., et al., In vitro and in vivo neurotoxicity of prion protein oligomers. Plos Pathogens, 2007. 3(8): p. 1175-1186.

165. Bueler, H., et al., Mice devoid of PrP are resistant to scrapie. Cell, 1993. 73(7): p. 1339-1347.

166. Mallucci, G., et al., Depleting neuronal PrP in prion infection prevents disease and reverses spongiosis. Science, 2003. 302(5646): p. 871-874.

167. Kocisko, D.A., et al., Cell-free formation of protease-resistant prion protein. Nature, 1994. 370(6489): p. 471-474.

168. Saborio, G.P., B. Permanne, and C. Soto, Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding. Nature, 2001. 411(6839): p. 810-813.

169. Castilla, J., et al., In vitro generation of infectious scrapie prions. Cell, 2005. 121(2): p. 195-206.

170. Legname, G., et al., Synthetic mammalian prions. Science, 2004. 305(5684): p. 673- 676.

171. Makarava, N., et al., Recombinant prion protein induces a new transmissible prion disease in wild-type animals. Acta Neuropathol, 2010. 119(2): p. 177-87.

172. Deleault, N.R., et al., Formation of native prions from minimal components in vitro. 2007, National Academy of Sciences. p. 9741.

173. Barria, M.A., et al., De novo generation of infectious prions in vitro produces a new disease phenotype. Plos Pathogens, 2009. 5(5).

174. Deleault, N.R., et al., Protease-resistant prion protein amplification reconstituted with partially purified substrates and synthetic polyanions. J Biol Chem, 2005. 280(29): p. 26873-9.

175. Deleault, N.R., R.W. Lucassen, and S. Supattapone, RNA molecules stimulate prion protein conversion. Nature, 2003. 425(6959): p. 717-20.

189

176. Geoghegan, J.C., et al., Selective incorporation of polyanionic molecules into hamster prions. J Biol Chem, 2007. 282(50): p. 36341-53.

177. Wang, F., et al., Lipid interaction converts prion protein to a PrPSc-like proteinase K-resistant conformation under physiological conditions. Biochemistry, 2007. 46(23): p. 7045-53.

178. Wang, F., et al., Generating a prion with bacterially expressed recombinant prion protein. Science, 2010. 327(5969): p. 1132-1135.

179. Morales, R., K. Abid, and C. Soto, The prion strain phenomenon: molecular basis and unprecedented features. Biochim Biophys Acta, 2007. 1772(6): p. 681-91.

180. Bruce, M., et al., Transmission of bovine spongiform encephalopathy and scrapie to mice: strain variation and the species barrier. Philos Trans R Soc Lond B Biol Sci, 1994. 343(1306): p. 405-11.

181. Bessen, R.A. and R.F. Marsh, Biochemical and physical properties of the prion protein from two strains of the transmissible mink encephalopathy agent. J Virol, 1992. 66(4): p. 2096-101.

182. Bessen, R.A. and R.F. Marsh, Identification of two biologically distinct strains of transmissible mink encephalopathy in hamsters. J Gen Virol, 1992. 73 ( Pt 2): p. 329- 34.

183. Caughey, B., G.J. Raymond, and R.A. Bessen, Strain-dependent differences in β-sheet conformations of abnormal prion protein. Journal of Biological Chemistry, 1998. 273(48): p. 32230.

184. Hill, A.F., et al., Molecular classification of sporadic Creutzfeldt-Jakob disease. Brain, 2003. 126(Pt 6): p. 1333-46.

185. Masters, C.L., et al., Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proceedings of the National Academy of Sciences of the United States of America, 1985. 82(12): p. 4245-4249.

186. Corder, E.H., et al., Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science, 1993. 261(5123): p. 921-923.

187. Mormino, E.C., et al., Amyloid and ApoE epsilon 4 interact to influence short-term decline in preclinical Alzheimer disease. Neurology, 2014. 82(20): p. 1760-1767.

188. Lim, Y.Y., et al., ApoE and BDNF polymorphisms moderate amyloid β-related cognitive decline in preclinical Alzheimer's disease. Molecular Psychiatry, 2015. 20(11): p. 1322-1328.

189. Lim, Y.Y., et al., ApoE epsilon 4 moderates amyloid-related memory decline in preclinical Alzheimer's disease. Neurobiology of Aging, 2015. 36(3): p. 1239-1244.

190. Baker, H.F., et al., Evidence for the experimental transmission of cerebral β- to primates. Int J Exp Pathol, 1993. 74(5): p. 441-54.

190

191. Hsiao, K., et al., Correlative memory deficits, Aβ elevation, and amyloid plaques in transgenic mice. Science, 1996. 274(5284): p. 99-102.

192. Sturchler-Pierrat, C., et al., Two amyloid precursor protein transgenic mouse models with Alzheimer disease-like pathology. Proc Natl Acad Sci U S A, 1997. 94(24): p. 13287-92.

193. Kane, M.D., et al., Evidence for seeding of β-amyloid by intracerebral infusion of Alzheimer brain extracts in β-amyloid precursor protein transgenic mice. Journal of Neuroscience, 2000. 20(10): p. 3606-3611.

194. Walker, L.C., et al., Exogenous induction of cerebral β-amyloidosis in βAPP- transgenic mice. Peptides, 2002. 23(7): p. 1241-7.

195. Meyer-Luehmann, M., et al., Exogenous induction of cerebral β-amyloidogenesis is governed by agent and host. Science, 2006. 313(5794): p. 1781-1784.

196. Duran-Aniotz, C., et al., Aggregate-depleted brain fails to induce Aβ deposition in a mouse model of Alzheimer's disease. PLoS ONE, 2014. 9(2): p. e89014.

197. Hamaguchi, T., et al., The presence of Aβ seeds, and not age per se, is critical to the initiation of Aβ deposition in the brain. Acta Neuropathologica, 2012. 123(1): p. 31- 37.

198. Duran-Aniotz, C., et al., Brains from non-Alzheimer's individuals containing amyloid deposits accelerate Aβ deposition in vivo. Acta Neuropathol Commun, 2013. 1: p. 76.

199. Stöhr, J., et al., Purified and synthetic Alzheimer's amyloid β (Aβ) prions. Proc Natl Acad Sci U S A, 2012. 109(27): p. 11025-30.

200. Ye, L., et al., Progression of seed-induced Aβ deposition within the limbic connectome. Brain Pathol, 2015. 25(6): p. 743-52.

201. Gajdusek, D.C., diseases of the central nervous system. American Journal of Clinical Pathology, 1971. 53(3): p. 320-32.

202. Rohwer, R.G., Scrapie infectious agent is virus-like in size and susceptibility to inactivation. Nature, 1984. 308(5960): p. 658-62.

203. Rohwer, R.G., The scrapie agent: "a virus by any other name". Curr Top Microbiol Immunol, 1991. 172: p. 195-232.

204. Gordon, W.S., Advances in veterinary research. Vet Rec, 1946. 58(47): p. 516-25.

205. Pattison, I.H., Resistance of the scrapie agent to formalin. J Comp Pathol, 1965. 75: p. 159-64.

206. Fritschi, S.K., et al., Aβ seeds resist inactivation by formaldehyde. Acta Neuropathol, 2014. 128(4): p. 477-84.

191

207. Eisele, Y.S., et al., Induction of cerebral β-amyloidosis: Intracerebral versus systemic Aβ inoculation. Proceedings of the National Academy of Sciences of the United States of America, 2009. 106(31): p. 12926-12931.

208. Flechsig, E., et al., Transmission of scrapie by steel-surface-bound prions. Molecular Medicine, 2001. 7(10): p. 679-684.

209. Zobeley, E., et al., Infectivity of scrapie prions bound to a stainless steel surface. Molecular Medicine, 1999. 5(4): p. 240-243.

210. Langer, F., et al., Soluble Aβ seeds are potent inducers of cerebral β-amyloid deposition. Journal of Neuroscience, 2011. 31(41): p. 14488-14495.

211. Eisele, Y.S., et al., Peripherally applied Aβ-containing inoculates induce cerebral β- amyloidosis. Science, 2010. 330(6006): p. 980-982.

212. Eisele, Y.S., et al., Multiple factors contribute to the peripheral induction of cerebral β-amyloidosis. J Neurosci, 2014. 34(31): p. 10264-73.

213. Morales, R., et al., De novo induction of amyloid-β deposition in vivo. Mol Psychiatry, 2012. 17(12): p. 1347-53.

214. Rosen, R.F., et al., Exogenous seeding of cerebral β-amyloid deposition in βAPP- transgenic rats. Journal of Neurochemistry, 2012. 120(5): p. 660-666.

215. Fritschi, S.K., et al., Highly potent soluble amyloid-β seeds in human Alzheimer brain but not cerebrospinal fluid. Brain, 2014. 137(Pt 11): p. 2909-15.

216. Jaunmuktane, Z., et al., Evidence for human transmission of amyloid-β pathology and cerebral amyloid angiopathy. Nature, 2015. 525(7568): p. 247-250.

217. Serrano-Pozo, A., et al., Neuropathological alterations in Alzheimer disease. Cold Spring Harb Perspect Med, 2011. 1(1): p. a006189.

218. Irwin, D.J., et al., Evaluation of potential infectivity of Alzheimer and Parkinson disease proteins in recipients of cadaver-derived human growth hormone. JAMA Neurol, 2013. 70(4): p. 462-8.

219. Frontzek, K., et al., Amyloid-β pathology and cerebral amyloid angiopathy are frequent in iatrogenic Creutzfeldt-Jakob disease after dural grafting. Swiss Med Wkly, 2016. 146: p. w14287.

220. Lu, J.X., et al., Molecular structure of β-amyloid fibrils in Alzheimer's disease brain tissue. Cell, 2013. 154(6): p. 1257-68.

221. Heilbronner, G.A., et al., Seeded strain-like transmission of β-amyloid morphotypes in APP transgenic mice. EMBO Reports, 2013: p. 1017.

222. McCann, H., et al., α-synucleinopathy phenotypes. Parkinsonism & Related Disorders, 2014. 20, Supplement 1: p. S62-S67.

192

223. Spillantini, M.G., et al., α-synuclein in Lewy bodies. Nature, 1997. 388(6645): p. 839- 840.

224. Baba, M., et al., Aggregation of α-synuclein in Lewy bodies of sporadic Parkinson's disease and dementia with lewy bodies. American Journal of Pathology, 1998. 152(4): p. 879-884.

225. Irizarry, M.C., et al., Nigral and cortical Lewy bodies and dystrophic nigral neurites in Parkinson's disease and cortical Lewy body disease contain α-synuclein immunoreactivity. Journal of Neuropathology and Experimental Neurology, 1998. 57(4): p. 334-337.

226. Wakabayashi, K., et al., NACP, a presynaptic protein, immunoreactivity in Lewy bodies in Parkinson's disease. Neurosci Lett, 1997. 239(1): p. 45-8.

227. Spillantini, M.G., et al., α-synuclein in filamentous inclusions of Lewy bodies from Parkinson's disease and dementia with Lewy bodies. Proc Natl Acad Sci U S A, 1998. 95(11): p. 6469-73.

228. Tu, P.H., et al., Glial cytoplasmic inclusions in white matter oligodendrocytes of multiple system atrophy brains contain insoluble α-synuclein. Annals of Neurology, 1998. 44(3): p. 415-422.

229. Wakabayashi, K., et al., α-synuclein immunoreactivity in glial cytoplasmic inclusions in multiple system atrophy. Neurosci Lett, 1998. 249(2-3): p. 180-2.

230. Spillantini, M.G., et al., Filamentous α-synuclein inclusions link multiple system atrophy with Parkinson's disease and dementia with Lewy bodies. Neuroscience Letters, 1998. 251(3): p. 205-208.

231. Fujiwara, H., et al., α-synuclein is phosphorylated in synucleinopathy lesions. Nature Cell Biology, 2002. 4(2): p. 160-164.

232. Tenreiro, S., et al., Phosphorylation modulates clearance of α-synuclein inclusions in a yeast model of Parkinson's disease. PLoS Genetics, 2014. 10(5): p. e1004302.

233. Chartier-Harlin, M.C., et al., α-synuclein locus duplication as a cause of familial Parkinson's disease. Lancet, 2004. 364(9440): p. 1167-1169.

234. Singleton, A.B., et al., α-synuclein locus triplication causes Parkinson's disease. Science, 2003. 302(5646): p. 841-841.

235. Polymeropoulos, M.H., et al., Mutation in the α-synuclein gene identified in families with Parkinson's disease. Science, 1997. 276(5321): p. 2045-2047.

236. Kruger, R., et al., Ala30Pro mutation in the gene encoding α-synuclein in Parkinson's disease. Nature Genetics, 1998. 18(2): p. 106-108.

237. Zarranz, J.J., et al., The new mutation, E46K, of α-synuclein causes Parkinson and Lewy body dementia. Annals of Neurology, 2004. 55(2): p. 164-173.

193

238. Appel-Cresswell, S., et al., α-synuclein p.H50Q, a novel pathogenic mutation for Parkinson's disease. Movement Disorders, 2013. 28(6): p. 811-813.

239. Proukakis, C., et al., A novel α-synuclein missense mutation in Parkinson disease. Neurology, 2013. 80(11): p. 1062-1064.

240. Lesage, S., et al., G51D α-synuclein mutation causes a novel Parkinsonian-pyramidal syndrome. Annals of Neurology, 2013. 73(4): p. 459-471.

241. Masliah, E., et al., Dopaminergic loss and inclusion body formation in α-synuclein mice: implications for neurodegenerative disorders. Science, 2000. 287(5456): p. 1265-1269.

242. Kahle, P.J., et al., Subcellular localization of wild-type and Parkinson's disease- associated mutant α-synuclein in human and transgenic mouse brain. Journal of Neuroscience, 2000. 20(17): p. 6365-6373.

243. Feany, M.B. and W.W. Bender, A Drosophila model of Parkinson's disease. Nature, 2000. 404(6776): p. 394-398.

244. van der Putten, H., et al., Neuropathology in mice expressing human α-synuclein. Journal of Neuroscience, 2000. 20(16): p. 6021-6029.

245. Chesselet, M.F. and F. Richter, Modelling of Parkinson's disease in mice. Lancet Neurology, 2011. 10(12): p. 1108-1118.

246. Kordower, J.H., et al., Lewy body-like pathology in long-term embryonic nigral transplants in Parkinson's disease. Nature Medicine, 2008. 14(5): p. 504-506.

247. Li, J.Y., et al., Lewy bodies in grafted neurons in subjects with Parkinson's disease suggest host-to-graft disease propagation. Nature Medicine, 2008. 14(5): p. 501-503.

248. Mendez, I., et al., Dopamine neurons implanted into people with Parkinson's disease survive without pathology for 14 years. Nature Medicine, 2008. 14(5): p. 507-509.

249. Danzer, K.M., et al., Seeding induced by α-synuclein oligomers provides evidence for spreading of α-synuclein pathology. Journal of Neurochemistry, 2009. 111(1): p. 192- 203.

250. Luk, K.C., et al., Exogenous α-synuclein fibrils seed the formation of Lewy body-like intracellular inclusions in cultured cells. Proceedings of the National Academy of Sciences of the United States of America, 2009. 106(47): p. 20051-20056.

251. Volpicelli-Daley, L.A., et al., Exogenous α-synuclein fibrils induce Lewy body pathology leading to synaptic dysfunction and neuron death. Neuron, 2011. 72(1): p. 57-71.

252. Hansen, C., et al., α-synuclein propagates from mouse brain to grafted dopaminergic neurons and seeds aggregation in cultured human cells. Journal of Clinical Investigation, 2011. 121(2): p. 715-725.

194

253. Bousset, L., et al., Structural and functional characterization of two α-synuclein strains. Nature Communications, 2013. 4: p. 2575-2575.

254. Desplats, P., et al., Inclusion formation and neuronal cell death through neuron-to- neuron transmission of α-synuclein. Proceedings of the National Academy of Sciences of the United States of America, 2009. 106(31): p. 13010-13015.

255. Luk, K.C., et al., Intracerebral inoculation of pathological α-synuclein initiates a rapidly progressive neurodegenerative α-synucleinopathy in mice. Journal of Experimental Medicine, 2012. 209(5): p. 975-986.

256. Mougenot, A.L., et al., Prion-like acceleration of a synucleinopathy in a transgenic mouse model. Neurobiology of Aging, 2012. 33(9): p. 2225-2228.

257. Peelaerts, W., et al., α-synuclein strains cause distinct synucleinopathies after local and systemic administration. Nature, 2015. 522(7556): p. 340-344.

258. Sacino, A.N., et al., Amyloidogenic α-synuclein seeds do not invariably induce rapid, widespread pathology in mice. Acta Neuropathologica, 2014(5): p. 645.

259. Rutherford, N.J., et al., Studies of lipopolysaccharide effects on the induction of α- synuclein pathology by exogenous fibrils in transgenic mice. Mol Neurodegener, 2015. 10: p. 32.

260. Bétemps, D., et al., α-synuclein spreading in M83 mice brain revealed by detection of pathological α-synuclein by enhanced ELISA. Acta Neuropathologica Communications, 2014. 2(1): p. 1-14.

261. Sacino, A.N., et al., Intramuscular injection of α-synuclein induces CNS α-synuclein pathology and a rapid-onset motor phenotype in transgenic mice. Proceedings of the National Academy of Sciences of the United States, 2014(29): p. 10732.

262. Luk, K.C., et al., Pathological α-synuclein transmission initiates Parkinson-like neurodegeneration in nontransgenic mice. Science, 2012. 338(6109): p. 949-953.

263. Paumier, K.L., et al., Intrastriatal injection of pre-formed mouse α-synuclein fibrils into rats triggers α-synuclein pathology and bilateral nigrostriatal degeneration. Neurobiology of Disease, 2015. 82: p. 185-199.

264. Masuda-Suzukake, M., et al., Pathological α-synuclein propagates through neural networks. Acta Neuropathol Commun, 2014. 2: p. 88.

265. Masuda-Suzukake, M., et al., Prion-like spreading of pathological α-synuclein in brain. Brain, 2013. 136: p. 1128-1138.

266. Tran, H.T., et al., α-synuclein immunotherapy blocks uptake and templated propagation of misfolded α-synuclein and neurodegeneration. Cell Reports, 2014. 7(6): p. 2054-2065.

267. Sacino, A.N., et al., Induction of CNS α-synuclein pathology by fibrillar and non- amyloidogenic recombinant α-synuclein. Acta Neuropathologica Communications, 2013. 1: p. 38-38.

195

268. Sacino, A.N., et al., Brain injection of α-synuclein induces multiple proteinopathies, gliosis, and a neuronal injury marker. J Neurosci, 2014. 34(37): p. 12368-78.

269. Watts, J.C., et al., Transmission of multiple system atrophy prions to transgenic mice. Proceedings Of The National Academy Of Sciences Of The United States Of America, 2013. 110(48): p. 19555-19560.

270. Guo, J.L., et al., Distinct α-synuclein strains differentially promote tau inclusions in neurons. Cell, 2013. 154(1): p. 103-17.

271. Sacino, A.N., et al., Non-prion-type transmission in A53T α-synuclein transgenic mice: a normal component of spinal homogenates from naïve non-transgenic mice induces robust α-synuclein pathology. Acta Neuropathologica, 2015.

272. Bernis, M.E., et al., Prion-like propagation of human brain-derived α-synuclein in transgenic mice expressing human wild-type α-synuclein. Acta Neuropathol Commun, 2015. 3(1): p. 75.

273. Grey, M., et al., Acceleration of α-synuclein aggregation by exosomes. J Biol Chem, 2015. 290(5): p. 2969-82.

274. Galvagnion, C., et al., Lipid vesicles trigger α-synuclein aggregation by stimulating primary nucleation. Nat Chem Biol, 2015. 11(3): p. 229-234.

275. Herva, M.E., et al., Anti-amyloid compounds inhibit α-synuclein aggregation induced by Protein misfolding cyclic amplification (PMCA). The Journal Of Biological Chemistry, 2014.

276. Bousset, L., et al., An efficient procedure for removal and inactivation of α-synuclein assemblies from laboratory materials. J Parkinsons Dis, 2015.

277. Medori, R., et al., Fatal familial insomnia, a prion disease with a mutation at codon- 178 of the prion protein gene. New England Journal of Medicine, 1992. 326(7): p. 444-449.

278. Collinge, J., et al., Prion dementia without characteristic pathology. Lancet, 1990. 336(8706): p. 7-9.

279. Manson, J.C., et al., A single amino acid alteration (101L) introduced into murine PrP dramatically alters incubation time of transmissible spongiform encephalopathy. Embo Journal, 1999. 18(23): p. 6855-6864.

280. Hsiao, K.K., et al., Spontaneous neurodegeneration in transgenic mice with mutant prion protein. Science, 1990. 250(4987): p. 1587-1590.

281. Chesebro, B., et al., Fatal transmissible amyloid encephalopathy: a new type of prion disease associated with lack of prion protein membrane anchoring. Plos Pathogens, 2010. 6(3).

282. Hill, A.F., et al., Species-barrier-independent prion replication in apparently resistant species. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(18): p. 10248-10253.

196

283. Thackray, A.M., et al., Chronic subclinical prion disease induced by low-dose inoculum. Journal of Virology, 2002. 76(5): p. 2510-2517.

284. Thackray, A.M., M.A. Klein, and R. Bujdoso, Subclinical prion disease induced by oral inoculation. Journal of Virology, 2003. 77(14): p. 7991-7998.

285. Collins, S.J., et al., Extended period of asymptomatic prion disease after low dose inoculation: assessment of detection methods and implications for infection control. Neurobiology of Disease, 2005. 20(2): p. 336-346.

286. Giannakopoulos, P., et al., Tangle and neuron numbers, but not amyloid load, predict cognitive status in Alzheimer's disease. Neurology, 2003. 60(9): p. 1495-1500.

287. Bennett, D.A., et al., Neurofibrillary tangles mediate the association of amyloid load with clinical Alzheimer disease and level of cognitive function. Archives of Neurology, 2004. 61(3): p. 378-384.

288. Carlson, G.A., et al., Genetic modification of the phenotypes produced by amyloid precursor protein overexpression in transgenic mice. Human Molecular Genetics, 1997. 6(11): p. 1951-1959.

289. Hughes, A.J., et al., Accuracy of clinical diagnosis of idiopathic Parkinson's disease: a clinicopathological study of 100 cases. Journal of Neurology Neurosurgery and Psychiatry, 1992. 55(3): p. 181-184.

290. Tompkins, M.M. and W.D. Hill, Contribution of somal Lewy bodies to neuronal death. Brain Research, 1997. 775(1-2): p. 24-29.

291. Parkkinen, L., et al., α-synuclein pathology does not predict extrapyramidal symptoms or dementia. Annals of Neurology, 2005. 57(1): p. 82-91.

292. Auluck, P.K., et al., Chaperone suppression of α-synuclein toxicity in a Drosophila model for Parkinson's disease. Science, 2002. 295(5556): p. 865-868.

293. Mallucci, G.R., et al., Post-natal knockout of prion protein alters hippocampal CA1 properties, but does not result in neurodegeneration. Embo Journal, 2002. 21(3): p. 202-210.

294. Brandner, S., et al., Normal host prion protein necessary for scrapie-induced neurotoxicity. Nature, 1996. 379(6563): p. 339-343.

295. Chiti, F. and C.M. Dobson, Protein misfolding, functional amyloid, and human disease, in Annual Review of Biochemistry. 2006. p. 333-366.

296. McLean, C.A., et al., Soluble pool of Aβ amyloid as a determinant of severity of neurodegeneration in Alzheimer's disease. Annals of Neurology, 1999. 46(6): p. 860- 866.

297. Mc Donald, J.M., et al., The presence of sodium dodecyl sulphate-stable Aβ dimers is strongly associated with Alzheimer-type dementia. Brain, 2010. 133: p. 1328-1341.

197

298. Nilsberth, C., et al., The 'Arctic' APP mutation (E693G) causes Alzheimer's disease by enhanced Aβ protofibril formation. Nature Neuroscience, 2001. 4(9): p. 887-893.

299. Cerf, E., et al., High ability of apolipoprotein E4 to stabilize amyloid-β peptide oligomers, the pathological entities responsible for Alzheimer's disease. Faseb Journal, 2011. 25(5): p. 1585-1595.

300. Walsh, D.M., et al., Naturally secreted oligomers of amyloid β protein potently inhibit hippocampal long-term potentiation in vivo. Nature, 2002. 416(6880): p. 535-539.

301. Shankar, G.M., et al., Amyloid-β protein dimers isolated directly from Alzheimer's brains impair synaptic plasticity and memory. Nature Medicine, 2008. 14(8): p. 837- 842.

302. Lacor, P.N., et al., Aβ oligomer-induced aberrations in synapse composition, shape, and density provide a molecular basis for loss of connectivity in Alzheimer's disease. Journal of Neuroscience, 2007. 27(4): p. 796-807.

303. Shankar, G.M., et al., Natural oligomers of the Alzheimer amyloid-β protein induce reversible synapse loss by modulating an NMDA-type glutamate receptor-dependent signaling pathway. Journal of Neuroscience, 2007. 27(11): p. 2866-2875.

304. Lesne, S., et al., A specific amyloid-β protein assembly in the brain impairs memory. Nature, 2006. 440(7082): p. 352-357.

305. Jan, A., et al., Aβ42 neurotoxicity is mediated by ongoing nucleated polymerization process rather than by discrete Aβ42 species. Journal of Biological Chemistry, 2011. 286(10): p. 8585-8596.

306. Lauren, J., et al., Cellular prion protein mediates impairment of synaptic plasticity by amyloid-β oligomers. Nature, 2009. 457(7233): p. 1128-U84.

307. Klyubin, I., et al., Peripheral administration of a humanized anti-PrP antibody blocks Alzheimer's disease Aβ synaptotoxicity. Journal of Neuroscience, 2014. 34(18): p. 6140-6145.

308. Balducci, C., et al., Synthetic amyloid-β oligomers impair long-term memory independently of cellular prion protein. Proceedings of the National Academy of Sciences of the United States of America, 2010. 107(5): p. 2295-2300.

309. Calella, A.M., et al., Prion protein and Aβ-related synaptic toxicity impairment. Embo Molecular Medicine, 2010. 2(8): p. 306-314.

310. Cisse, M., et al., Ablation of cellular prion protein does not ameliorate abnormal neural network activity or cognitive dysfunction in the J20 line of human amyloid precursor protein transgenic mice. Journal of Neuroscience, 2011. 31(29): p. 10427- 10431.

311. Kessels, H.W., et al., The prion protein as a receptor for amyloid-β. Nature, 2010. 466(7308): p. E3-E4.

198

312. Jin, M., et al., Soluble amyloid β-protein dimers isolated from Alzheimer cortex directly induce tau hyperphosphorylation and neuritic degeneration. Proceedings of the National Academy of Sciences of the United States of America, 2011. 108(14): p. 5819-5824.

313. Ono, K., M.M. Condron, and D.B. Teplow, Structure-neurotoxicity relationships of amyloid β-protein oligomers. Proceedings of the National Academy of Sciences of the United States of America, 2009. 106(35): p. 14745-14750.

314. Jana, M.K., et al., Membrane bound tetramer and trimer Aβ oligomeric species correlate with toxicity towards cultured neurons. Journal of Neurochemistry, 2015. 136(3): p. 594-608.

315. Handoko, M., et al., Correlation of specific amyloid-β oligomers with tau in cerebrospinal fluid from cognitively normal older adults. Jama Neurology, 2013. 70(5): p. 594-599.

316. Lesne, S.E., et al., Brain amyloid-β oligomers in ageing and Alzheimer's disease. Brain, 2013. 136: p. 1383-1398.

317. Gong, Y.S., et al., Alzheimer's disease-affected brain: Presence of oligomeric A β ligands (ADDLs) suggests a molecular basis for reversible memory loss. Proceedings of the National Academy of Sciences of the United States of America, 2003. 100(18): p. 10417-10422.

318. Bitan, G., et al., Amyloid β-protein (Aβ) assembly: Aβ40 and Aβ42 oligomerize through distinct pathways. Proceedings of the National Academy of Sciences of the United States of America, 2003. 100(1): p. 330-335.

319. Bernstein, S.L., et al., Amyloid-β protein oligomerization and the importance of tetramers and dodecamers in the aetiology of Alzheimer's disease. Nature Chemistry, 2009. 1(4): p. 326-331.

320. Robb, E., et al., High order W02-reactive stable oligomers of amyloid-β are produced in vivo and in vitro via dialysis and filtration of synthetic amyloid-β monomer. J Alzheimers Dis, 2015. 44(1): p. 69-78.

321. Kramer, M.L. and W.J. Schulz-Schaeffer, Presynaptic α-synuclein aggregates, not Lewy bodies, cause neurodegeneration in dementia with Lewy bodies. Journal of Neuroscience, 2007. 27(6): p. 1405-1410.

322. Garcia-Esparcia, P., et al., Altered machinery of protein synthesis is region- and stage-dependent and is associated with α-synuclein oligomers in Parkinson's disease. Acta Neuropathol Commun, 2015. 3(1): p. 76.

323. Winner, B., et al., In vivo demonstration that α-synuclein oligomers are toxic. Proceedings of the National Academy of Sciences of the United States of America, 2011. 108(10): p. 4194-4199.

324. Karpinar, D.P., et al., Pre-fibrillar α-synuclein variants with impaired β-structure increase neurotoxicity in Parkinson's disease models. Embo Journal, 2009. 28(20): p. 3256-3268.

199

325. Lam, H.T., et al., Stabilization of α-synuclein fibril clusters prevents fragmentation and reduces seeding activity and toxicity. Biochemistry, 2016.

326. Danzer, K.M., et al., Different species of α-synuclein oligomers induce calcium influx and seeding. Journal of Neuroscience, 2007. 27(34): p. 9220-9232.

327. Chen, S.W., et al., Structural characterization of toxic oligomers that are kinetically trapped during α-synuclein fibril formation. Proceedings of the National Academy of Sciences of the United States of America, 2015. 112(16): p. E1994-E2003.

328. Gosavi, N., et al., Golgi fragmentation occurs in the cells with prefibrillar α-synuclein aggregates and precedes the formation of fibrillar inclusion. Journal of Biological Chemistry, 2002. 277(50): p. 48984-48992.

329. Luth, E.S., et al., Soluble, prefibrillar α-synuclein oligomers promote complex I- dependent, Ca2+-induced mitochondrial dysfunction. Journal of Biological Chemistry, 2014. 289(31): p. 21490-21507.

330. Schapira, A.H.V., et al., Mitochondrial complex I deficiency in Parkinson's disease. Lancet, 1989. 1(8649): p. 1269-1269.

331. Parker, W.D., S.J. Boyson, and J.K. Parks, Abnormalities of the electron transport chain in idiopathic Parkinson's disease. Annals of Neurology, 1989. 26(6): p. 719- 723.

332. Janetzky, B., et al., Unaltered aconitase activity, but decreased complex I activity in substantia nigra pars compacta of patients with Parkinson's disease. Neuroscience Letters, 1994. 169(1-2): p. 126-128.

333. Camilleri, A. and N. Vassallo, The centrality of mitochondria in the pathogenesis and treatment of Parkinson's disease. Cns Neuroscience & Therapeutics, 2014. 20(7): p. 591-602.

334. Langston, J.W. and P.A. Ballard, Parkinson's disease in a chemist working with 1- methyl-4-phenyl-1,2,5,6-tetrahydropyridine. New England Journal of Medicine, 1983. 309(5): p. 310-310.

335. Burns, R.S., et al., A primate model of Parkinsonism: selective destruction of dopaminergic neurons in the pars compacta of the substantia nigra by n-methyl-4- phenyl-1,2,3,6-tetrahydropyridine. Proceedings of the National Academy of Sciences of the United States of America-Biological Sciences, 1983. 80(14): p. 4546-4550.

336. Betarbet, R., et al., Chronic systemic pesticide exposure reproduces features of Parkinson's disease. Nature Neuroscience, 2000. 3(12): p. 1301-1306.

337. Ruiperez, V., F. Darios, and B. Davletov, α-synuclein, lipids and Parkinson's disease. Progress in Lipid Research, 2010. 49(4): p. 420-428.

338. Volles, M.J., et al., Vesicle permeabilization by protofibrillar α-synuclein: implications for the pathogenesis and treatment of Parkinson's disease. Biochemistry, 2001(26): p. 7812.

200

339. Shrivastava, A.N., et al., α-synuclein assemblies sequester neuronal α3-Na+/K+- ATPase and impair Na+ gradient. The EMBO Journal, 2015.

340. de Carvalho Aguiar, P., et al., Mutations in the Na+/K+ ATPase α3 gene ATP1A3 are associated with rapid-onset dystonia parkinsonism. Neuron, 2004. 43: p. 169-175.

341. Rodacker, V., M. Toustrup-Jensen, and B. Vilsen, Mutations Phe785Leu and Thr618Met in Na+,K+-ATPase, associated with familial rapid-onset dystonia parkinsonism, interfere with Na+ interaction by distinct mechanisms. The Journal Of Biological Chemistry, 2006. 281(27): p. 18539-18548.

342. Clapcote, S.J., et al., Mutation I810N in the α3 isoform of Na+,K+-ATPase causes impairments in the sodium pump and hyperexcitability in the CNS. Proceedings Of The National Academy Of Sciences Of The United States Of America, 2009. 106(33): p. 14085-14090.

343. Pieri, L., et al., Fibrillar α-synuclein and huntingtin exon 1 assemblies are toxic to the cells. Biophys J, 2012. 102(12): p. 2894-905.

344. Kayed, R., et al., Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis. Science, 2003. 300(5618): p. 486-489.

345. Wagner, J., et al., Anle138b: a novel oligomer modulator for disease-modifying therapy of neurodegenerative diseases such as prion and Parkinson's disease. Acta Neuropathologica, 2013. 125(6): p. 795-813.

346. Levin, J., et al., The oligomer modulator anle138b inhibits disease progression in a Parkinson mouse model even with treatment started after disease onset. Acta Neuropathologica, 2014. 127(5): p. 779-780.

347. Kazlauskaite, J., et al., An unusual soluble β-turn-rich conformation of prion is involved in fibril formation and toxic to neuronal cells. Biochemical and Biophysical Research Communications, 2005. 328(1): p. 292-305.

348. Novitskaya, V., et al., Amyloid fibrils of mammalian prion protein are highly toxic to cultured cells and primary neurons. Journal of Biological Chemistry, 2006. 281(19): p. 13828-13836.

349. Jansen, K., et al., Structural intermediates in the putative pathway from the cellular prion protein to the pathogenic form. Biological Chemistry, 2001. 382(4): p. 683-691.

350. Zhou, M.H., et al., Highly neurotoxic monomeric α-helical prion protein. Proceedings of the National Academy of Sciences of the United States of America, 2012. 109(8): p. 3113-3118.

351. Mays, C.E., et al., Prion infectivity plateaus and conversion to symptomatic disease originate from falling precursor levels and increased oligomeric PrPSc species. Journal Of Virology, 2015.

352. O'Nuallain, B., et al., Amyloid β-protein dimers rapidly form stable synaptotoxic protofibrils. Journal of Neuroscience, 2010. 30(43): p. 14411-14419.

201

353. Bitan, G., et al., Neurotoxic protein oligomers - what you see is not always what you get. Amyloid-Journal of Disorders, 2005. 12(2): p. 88-95.

354. Giehm, L., et al., SDS-induced fibrillation of α-synuclein: an alternative fibrillation pathway. Journal of Molecular Biology, 2010. 401(1): p. 115-133.

355. Ugalde, C.L., et al., Neurotoxicity in prion disease: relationships with other neurodegenerative diseases and outlooks for therapeutic intervention, in The prion phenomena in neurodegenerative disease: new frontiers in neuroscience, G. Giachin and G. Legame, Editors. 2015, Nova Science Publishers: U.S.A.

356. Kimberlin, R.H. and C.A. Walker, Evidence that transmission of one source of scrapie agent to hamsters involves separation of agent strains from a mixture. Journal of General Virology, 1978. 39(JUN): p. 487-496.

357. Weissmann, C., Spongiform encephalopathies - the prion's progress. Nature, 1991. 349(6310): p. 569-571.

358. Cohen, F.E., et al., Structual clues to prion replication. Science, 1994. 264(5158): p. 530-531.

359. Hill, A.F. and J. Collinge, Subclinical prion infection. Trends in Microbiology, 2003. 11(12): p. 578-584.

360. Collinge, J. and A.R. Clarke, A general model of prion strains and their pathogenicity. Science, 2007. 318(5852): p. 930-936.

361. Baringer, J.R., K.A. Bowman, and S.B. Prusiner, Replication of the scrapie agent in hamster brain precedes neuronal vacuolation. Journal of Neuropathology and Experimental Neurology, 1983. 42(5): p. 539-547.

362. Sandberg, M.K., et al., Prion propagation and toxicity in vivo occur in two distinct mechanistic phases. Nature, 2011. 470(7335): p. 540-542.

363. Sandberg, M.K., et al., Prion neuropathology follows the accumulation of alternate prion protein isoforms after infective titre has peaked. Nature Communications, 2014. 5.

364. Fischer, M., et al., Prion protein (PrP) with amino-proximal deletions restoring susceptibility of PrP knockout mice to scrapie. Embo Journal, 1996. 15(6): p. 1255- 1264.

365. Pastrana, M.A., et al., Isolation and characterization of a proteinase K-sensitive PrPSc fraction. Biochemistry, 2006. 45(51): p. 15710-15717.

366. Coleman, B.M., et al., Pathogenic mutations within the hydrophobic domain of the prion protein lead to the formation of protease-sensitive prion species with increased lethality. Journal of Virology, 2014. 88(5): p. 2690-2703.

367. Hwang, D., et al., A systems approach to prion disease. Molecular Systems Biology, 2009. 5.

202

368. Zampieri, M., et al., A system-level approach for deciphering the transcriptional response to prion infection. , 2011. 27(24): p. 3407-3414.

369. Mays, C.E., et al., Prion disease tempo determined by host-dependent substrate reduction. Journal of Clinical Investigation, 2014. 124(2): p. 847-858.

370. Bueler, H., et al., High prion and PrPSc levels but delayed-onset of disease in scrapie-inoclulated mice heterozygous for a disrupted PrP gene. Molecular Medicine, 1994. 1(1): p. 19-30.

371. Schulz-Schaeffer, W.J., et al., The paraffin-embedded tissue blot detects PrPSC early in the incubation time in prion diseases. American Journal of Pathology, 2000. 156(1): p. 51-56.

372. Moreno, J.A., et al., Sustained translational repression by eIF2 α-P mediates prion neurodegeneration. Nature, 2012. 485(7399): p. 507-U119.

373. Shmerling, D., et al., Expression of amino-terminally truncated PrP in the mouse leading to ataxia and specific cerebellar lesions. Cell, 1998. 93(2): p. 203-214.

374. Baumann, F., et al., Lethal recessive myelin toxicity of prion protein lacking its central domain. Embo Journal, 2007. 26(2): p. 538-547.

375. Harrison, C.F., et al., Conservation of a glycine-rich region in the prion protein is required for uptake of prion infectivity. Journal of Biological Chemistry, 2010. 285(26): p. 20213-20223.

376. Peretz, D., et al., A conformational transition at the N terminus of the prion protein features in formation of the scrapie isoform. Journal of Molecular Biology, 1997. 273(3): p. 614-622.

377. Moore, R.C., et al., Doppel-induced in transgenic mice. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98(26): p. 15288-15293.

378. Hoshino, S., et al., Prions prevent brain damage after experimental brain injury: a preliminary report, in Brain Edema Xii, T. Kuroiwa, et al., Editors. 2003. p. 297-299.

379. Weise, J., et al., Upregulation of cellular prion protein (PrPc) after focal cerebral ischemia and influence of lesion severity. Neuroscience Letters, 2004. 372(1-2): p. 146-150.

380. Sakurai-Yamashita, Y., et al., Female-specific neuroprotection against transient brain ischemia observed in mice devoid of prion protein is abolished by ectopic expression of prion protein-like protein. Neuroscience, 2005. 136(1): p. 281-287.

381. McLennan, N.F., et al., Prion protein accumulation and neuroprotection in hypoxic brain damage. American Journal of Pathology, 2004. 165(1): p. 227-235.

382. Kuwahara, C., et al., Prions prevent neuronal cell-line death. Nature, 1999. 400(6741): p. 225-226.

203

383. Bounhar, Y., et al., Prion protein protects human neurons against Bax-mediated apoptosis. Journal of Biological Chemistry, 2001. 276(42): p. 39145-39149.

384. Mitteregger, G., et al., The role of the octarepeat region in neuroprotective function of the cellular prion protein. Brain Pathology, 2007. 17(2): p. 174-183.

385. Sakudo, A., et al., Octapeptide repeat region and N-terminal half of hydrophobic region of prion protein (PrP) mediate PrP-dependent activation of superoxide dismutase. Biochemical and Biophysical Research Communications, 2005. 326(3): p. 600-606.

386. Flechsig, E., et al., Prion protein devoid of the octapeptide repeat region restores susceptibility to scrapie in PrP knockout mice. Neuron, 2000. 27(2): p. 399-408.

387. Scalabrino, G., et al., The octapeptide repeat region and cobalamin-deficient polyneuropathy of the rat. Muscle & Nerve, 2011. 44(6): p. 957-967.

388. Yoshikawa, D., et al., Dominant-negative effects of the N-terminal half of prion protein on neurotoxicity of prion protein-like protein/doppel in mice. Journal of Biological Chemistry, 2008. 283(35): p. 24202-24211.

389. Solforosi, L., et al., Cross-linking cellular prion protein triggers neuronal apoptosis in vivo. Science, 2004. 303(5663): p. 1514-1516.

390. Sonati, T., et al., The toxicity of antiprion antibodies is mediated by the flexible tail of the prion protein. Nature, 2013. 501(7465): p. 102-+.

391. Falsig, J., et al., Prion pathogenesis is faithfully reproduced in cerebellar organotypic slice cultures. Plos Pathogens, 2012. 8(11).

392. Herrmann, U.S., et al., Prion and anti-PrP antibodies trigger converging neurotoxic pathways. PLoS Pathog, 2015. 11(2): p. e1004662.

393. Swayampakula, M., et al., The crystal structure of an octapeptide repeat of the prion protein in complex with a Fab fragment of the POM2 antibody. Protein Science, 2013. 22(7): p. 893-903.

394. Ron, D. and P. Walter, Signal integration in the endoplasmic reticulum unfolded protein response. Nature Reviews Molecular Cell Biology, 2007. 8(7): p. 519-529.

395. Hegde, R.S., et al., A transmembrane form of the prion protein in neurodegenerative disease. Science, 1998. 279(5352): p. 827-834.

396. Stewart, R.S., et al., Neurodegenerative illness in transgenic mice expressing a transmembrane form of the prion protein. Journal of Neuroscience, 2005. 25(13): p. 3469-3477.

397. Ma, J.Y. and S. Lindquist, Conversion of PrP to a self-perpetuating PrPSc-like conformation in the cytosol. Science, 2002. 298(5599): p. 1785-1788.

398. Ma, J.Y., R. Wollmann, and S. Lindquist, Neurotoxicity and neurodegeneration when PrP accumulates in the cytosol. Science, 2002. 298(5599): p. 1781-1785.

204

399. Grenier, C., et al., Molecular morphology and toxicity of cytoplasmic prion protein aggregates in neuronal and non-neuronal cells. Journal of Neurochemistry, 2006. 97(5): p. 1456-1466.

400. Hegde, R.S., et al., Transmissible and genetic prion diseases share a common pathway of neurodegeneration. Nature, 1999. 402(6763): p. 822-826.

401. Stewart, R.S., B. Drisaldi, and D.A. Harris, A transmembrane form of the prion protein contains an uncleaved signal peptide and is retained in the endoplasmic reticululm. Molecular Biology of the Cell, 2001. 12(4): p. 881-889.

402. Drisaldi, B., et al., Mutant PrP is delayed in its exit from the endoplasmic reticulum, but neither wild-type nor mutant PrP undergoes retrotranslocation prior to proteasomal degradation. Journal of Biological Chemistry, 2003. 278(24): p. 21732- 21743.

403. Rane, N.S., J.L. Yonkovich, and R.S. Hegde, Protection from cytosolic prion protein toxicity by modulation of protein translocation. Embo Journal, 2004. 23(23): p. 4550- 4559.

404. Ma, J.Y. and S. Lindquist, Wild-type PrP and a mutant associated with prion disease are subject to retrograde transport and proteasome degradation. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98(26): p. 14955-14960.

405. Yedidia, Y., et al., Proteasomes and ubiquitin are involved in the turnover of the wild- type prion protein. Embo Journal, 2001. 20(19): p. 5383-5391.

406. Chakrabarti, O. and R.S. Hegde, Functional depletion of mahogunin by cytosolically exposed prion protein contributes to neurodegeneration. Cell, 2009. 137(6): p. 1136- 1147.

407. He, L., et al., Spongiform degeneration in mahoganoid mutant mice. Science, 2003. 299(5607): p. 710-712.

408. Silvius, D., et al., Levels of the mahogunin ring finger 1 E3 ubiquitin ligase do not influence prion disease. Plos One, 2013. 8(1).

409. Gunn, T.M. and G.A. Carlson, RML prions act through Mahogunin and Attractin- independent pathways. Prion, 2013. 7(4): p. 267-271.

410. Yost, C.S., et al., Non-hydrophobic extracytoplasmic determinant of stop transfer in the prion protein. Nature, 1990. 343(6259): p. 669-672.

411. Rane, N.S., et al., Reduced translocation of nascent prion protein during ER stress contributes to neurodegeneration. Developmental Cell, 2008. 15(3): p. 359-370.

412. Kristiansen, M., et al., Disease-associated prion protein oligomers inhibit the 26S proteasome. Molecular Cell, 2007. 26(2): p. 175-188.

205

413. Deriziotis, P., et al., Misfolded PrP impairs the UPS by interaction with the 20S proteasome and inhibition of substrate entry. Embo Journal, 2011. 30(15): p. 3065- 3077.

414. Roucou, X., et al., Cytosolic prion protein is not toxic and protects against Bax- mediated cell death in human primary neurons. Journal of Biological Chemistry, 2003. 278(42): p. 40877-40881.

415. Thackray, A.M., et al., Cytosolic PrP can participate in prion-mediated toxicity. Journal of Virology, 2014. 88(14): p. 8129-8138.

416. Bertolotti, A., et al., Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nature Cell Biology, 2000. 2(6): p. 326-332.

417. Shen, J.S., et al., Stable binding of ATF6 to BiP in the endoplasmic reticulum stress response. Molecular and Cellular Biology, 2005. 25(3): p. 921-932.

418. Hetz, C., et al., Caspase-12 and endoplasmic reticulum stress mediate neurotoxicity of pathological prion protein. Embo Journal, 2003. 22(20): p. 5435-5445.

419. Ferreiro, E., et al., An endoplasmic-reticulum-specific apoptotic pathway is involved in prion and amyloid-β peptides neurotoxicity. Neurobiology of Disease, 2006. 23(3): p. 669-678.

420. Moreno, J.A., et al., Oral treatment targeting the unfolded protein response prevents neurodegeneration and clinical disease in prion-infected mice. Science Translational Medicine, 2013. 5(206).

421. Halliday, M., et al., Partial restoration of protein synthesis rates by the small molecule ISRIB prevents neurodegeneration without pancreatic toxicity. Cell Death & Disease, 2015. 6: p. e1672-e1672.

422. Halliday, M., et al., Repurposed drugs targeting eIF2 α-P-mediated translational repression prevent neurodegeneration in mice. Brain, 2017. 140: p. 1768-1783.

423. Kober, L., C. Zehe, and J. Bode, Development of a novel ER stress based selection system for the isolation of highly productive clones. Biotechnology and Bioengineering, 2012. 109(10): p. 2599-2611.

424. Steele, A.D., et al., Prion pathogenesis is independent of caspase-12. Prion, 2007. 1(4): p. 243-247.

425. Steele, A.D., et al., Diminishing apoptosis by deletion of bax or overexpression of Bcl-2 does not protect against infectious prion toxicity in vivo. Journal of Neuroscience, 2007. 27(47): p. 13022-13027.

426. Harding, H.P., et al., An integrated stress response regulates amino acid metabolism and resistance to oxidative stress. Molecular Cell, 2003. 11(3): p. 619-633.

427. Hoozemans, J.J.M., et al., The unfolded protein response is activated in pretangle neurons in Alzheimer's disease hippocampus. American Journal of Pathology, 2009. 174(4): p. 1241-1251.

206

428. Hoozemans, J.J.M., et al., The unfolded protein response is activated in Alzheimer's disease. Acta Neuropathologica, 2005. 110(2): p. 165-172.

429. Segev, Y., D.M. Michaelson, and K. Rosenblum, ApoE epsilon 4 is associated with eIF2α phosphorylation and impaired learning in young mice. Neurobiology of Aging, 2013. 34(3): p. 863-872.

430. Ma, T., et al., Suppression of eIF2 α kinases alleviates Alzheimer's disease-related plasticity and memory deficits. Nature Neuroscience, 2013. 16(9): p. 1299-U185.

431. O'Connor, T., et al., Phosphorylation of the translation initiation factor eIF2α increases BACE1 levels and promotes amyloidogenesis. Neuron, 2008. 60(6): p. 988- 1009.

432. Mihailovich, M., et al., Complex of BACE1 involves upstream AUGs and stimulatory elements within the 5 ' untranslated region. Nucleic Acids Research, 2007. 35(9): p. 2975-2985.

433. Lammich, S., et al., Expression of the Alzheimer protease BACE1 is suppressed via its 5 '-untranslated region. Embo Reports, 2004. 5(6): p. 620-625.

434. Devi, L. and M. Ohno, Phospho-eIF2 α level is important for determining abilities of BACE1 reduction to rescue cholinergic neurodegeneration and memory defects in 5XFAD mice. Plos One, 2010. 5(9).

435. Tonelli, D.D., et al., Translational regulation of BACE-1 expression in neuronal and non-neuronal cells. Nucleic Acids Research, 2004. 32(5): p. 1808-1817.

436. Devi, L. and M. Ohno, PERK mediates eIF2 α phosphorylation responsible for BACE1 elevation, CREB dysfunction and neurodegeneration in a mouse model of Alzheimer's disease. Neurobiology of Aging, 2014. 35(10): p. 2272-2281.

437. Hoozemans, J.J.M., et al., Activation of the unfolded protein response in Parkinson's disease. Biochemical and Biophysical Research Communications, 2007. 354(3): p. 707-711.

438. Bellucci, A., et al., Induction of the unfolded protein response by α-synuclein in experimental models of Parkinson's disease. Journal of Neurochemistry, 2011. 116(4): p. 588-605.

439. Xu, B., et al., Oxidative stress involvement in manganese-induced α-synuclein oligomerization in organotypic brain slice cultures. Toxicology, 2013. 305: p. 71-78.

440. Xu, B., et al., α-synuclein is involved in manganese-induced ER stress via PERK signal pathway in organotypic brain slice cultures. Molecular Neurobiology, 2014. 49(1): p. 399-412.

441. Smith, W.W., et al., Endoplasmic reticulum stress and mitochondrial cell death pathways mediate A53T mutant α-synuclein-induced toxicity. Human Molecular Genetics, 2005. 14(24): p. 3801-3811.

207

442. Colla, E., et al., Endoplasmic reticulum stress is important for the manifestations of α- synucleinopathy in vivo. Journal of Neuroscience, 2012. 32(10): p. 3306-3320.

443. Wiersma, V.I., et al., Activation of the unfolded protein response and granulovacuolar degeneration are not common features of human prion pathology. Acta Neuropathologica Communications, 2016. 4.

444. Unterberger, U., et al., Endoplasmic reticulum stress features are prominent in Alzheimer disease but not in prion diseases in vivo. Journal of Neuropathology and Experimental Neurology, 2006. 65(4): p. 348-357.

445. Cappai, R., et al., Dopamine promotes α-synuclein aggregation into SDS-resistant soluble oligomers via a distinct folding pathway. FASEB journal, 2005.

446. Biedler, J.L., et al., Multiple neurotransmitter synthesis by human neuroblastoma cell lines and clones. Cancer Research, 1978. 38(11): p. 3751-3757.

447. Biedler, J.L., L. Helson, and B.A. Spengler, Morphology and growth, tumorigenicity of human neuroblastoma cells in continuous culture. Cancer Research, 1973. 33(11): p. 2643-2652.

448. Lawson, V.A., et al., Mouse-adapted sporadic human Creutzfeldt-Jakob disease prions propagate in cell culture. International Journal of Biochemistry & Cell Biology, 2008. 40(12): p. 2793-2801.

449. Brazier, M.W., et al., Correlative studies support lipid peroxidation is linked to PrPres propagation as art early primary pathogenic event in prion disease. Brain Research Bulletin, 2006. 68(5): p. 346-354.

450. Schindelin, J., et al., Fiji: an open-source platform for biological-image analysis. Nature Methods, 2012. 9(7): p. 676-682.

451. Rizzo, G., et al., Accuracy of clinical diagnosis of Parkinson disease: a systematic review and meta-analysis. Neurology, 2016. 86(6): p. 566-576.

452. Furukawa, K., et al., Plasma membrane ion permeability induced by mutant α- synuclein contributes to the degeneration of neural cells. Journal of Neurochemistry, 2006. 97(4): p. 1071-1077.

453. Zakharov, S.D., et al., Helical α-synuclein forms highly conductive ion channels. Biochemistry, 2007. 46(50): p. 14369-14379.

454. Cooper, A.A., et al., α-synuclein blocks ER-Golgi traffic and Rab1 rescues neuron loss in Parkinson's models. Science, 2006. 313(5785): p. 324-328.

455. Gitler, A.D., et al., The Parkinson's disease protein α-synuclein disrupts cellular Rab homeostasis. Proceedings of the National Academy of Sciences of the United States of America, 2008. 105(1): p. 145-150.

456. Sarafian, T.A., et al., Impairment of mitochondria in adult mouse brain overexpressing predominantly full-length, N-terminally acetylated human α- synuclein. Plos One, 2013. 8(5).

208

457. Bender, A., et al., TOM40 mediates mitochondrial dysfunction induced by α-synuclein accumulation in Parkinson's disease. Plos One, 2013. 8(4).

458. Martin, L.J., et al., Parkinson's disease α-synuclein transgenic mice develop neuronal mitochondrial degeneration and cell death. Journal of Neuroscience, 2006. 26(1): p. 41-50.

459. Li, L., et al., Human A53T α-synuclein causes reversible deficits in mitochondrial function and dynamics in primary mouse cortical neurons. Plos One, 2013. 8(12).

460. Zhu, Y.G., et al., α-synuclein overexpression impairs mitochondrial function by associating with adenylate translocator. International Journal of Biochemistry & Cell Biology, 2011. 43(5): p. 732-741.

461. Chinta, S.J., et al., Mitochondrial α-synuclein accumulation impairs complex I function in dopaminergic neurons and results in increased mitophagy in vivo. Neuroscience Letters, 2010. 486(3): p. 235-239.

462. Devi, L., et al., Mitochondrial import and accumulation of α-synuclein impair complex I in human dopaminergic neuronal cultures and Parkinson disease brain. Journal of Biological Chemistry, 2008. 283(14): p. 9089-9100.

463. Snead, D. and D. Eliezer, α-synuclein function and dysfunction on cellular membranes. Experimental Neurobiology, 2014. 23(4): p. 292-313.

464. Perrin, R.J., et al., Exposure to long chain polyunsaturated fatty acids triggers rapid multimerization of synucleins. Journal of Biological Chemistry, 2001. 276(45): p. 41958-41962.

465. Broersen, K., et al., α-synuclein adopts an α-helical conformation in the presence of polyunsaturated fatty acids to hinder micelle formation. Biochemistry, 2006. 45(51): p. 15610-15616.

466. Sharon, R., et al., The formation of highly soluble oligomers of α-synuclein is regulated by fatty acids and enhanced in Parkinson's disease. Neuron, 2003. 37(4): p. 583-595.

467. Assayag, K., et al., Polyunsaturated fatty acids induce α-synuclein-related pathogenic changes in neuronal cells. American Journal of Pathology, 2007. 171(6): p. 2000- 2011.

468. De Franceschi, G., et al., Molecular insights into the interaction between α-synuclein and docosahexaenoic acid. Journal of Molecular Biology, 2009. 394(1): p. 94-107.

469. Sharon, R., et al., Altered fatty acid composition of dopaminergic neurons expressing α-synuclein and human brains with α-synucleinopathies. Journal of Biological Chemistry, 2003. 278(50): p. 49874-49881.

470. Tsang, A.H.K. and K.K.K. Chung, Oxidative and nitrosative stress in Parkinson's disease. Biochimica Et Biophysica Acta-Molecular Basis of Disease, 2009. 1792(7): p. 643-650.

209

471. Dalfo, E., et al., Evidence of oxidative stress in the neocortex in incidental Lewy body disease. Journal of Neuropathology and Experimental Neurology, 2005. 64(9): p. 816- 830.

472. Morel, P., et al., Effects of 4-hydroxynonenal, a lipid peroxidation product, on dopamine transport and Na+/K+ ATPase in rat striatal synaptosomes. Neurochemistry International, 1998. 33(6): p. 531-540.

473. De Franceschi, G., et al., α-synuclein structural features inhibit harmful polyunsaturated fatty acid oxidation, suggesting roles in neuroprotection. The Journal of Biological Chemistry, 2017. 292(17): p. 6927-6937.

474. Aoki, R. and Y.R. Li, α-synuclein promotes neuroprotection through NF-κB-mediated transcriptional regulation of protein kinase Cδ. Sci Signal, 2011. 4(195): p. jc6.

475. Jo, E.J., et al., Defective membrane interactions of familial Parkinson's disease mutant A30P α-synuclein. Journal of Molecular Biology, 2002. 315(4): p. 799-807.

476. Bussell, R. and D. Eliezer, Effects of Parkinson's disease-linked mutations on the structure of lipid-associated α-synuclein. Biochemistry, 2004. 43(16): p. 4810-4818.

477. van Maarschalkerweerd, A., et al., Protein/lipid coaggregates are formed during α- synuclein-induced disruption of lipid bilayers. Biomacromolecules, 2014. 15(10): p. 3643-3654.

478. van Maarschalkerweerd, A., V. Vetri, and B. Vestergaard, Cholesterol facilitates interactions between α-synuclein oligomers and charge-neutral membranes. Febs Letters, 2015. 589(19): p. 2661-2667.

479. van Rooijen, B.D., M.M.A.E. Claessens, and V. Subramaniam, Lipid bilayer disruption by oligomeric α-synuclein depends on bilayer charge and accessibility of the hydrophobic core. Biochimica Et Biophysica Acta-Biomembranes, 2009. 1788(6): p. 1271-1278.

480. Giehm, L., et al., Low-resolution structure of a vesicle disrupting α-synuclein oligomer that accumulates during fibrillation. Proceedings of the National Academy of Sciences of the United States of America, 2011. 108(8): p. 3246-3251.

481. Rhoades, E., et al., Quantification of α-synuclein binding to lipid vesicles using fluorescence correlation spectroscopy. Biophysical Journal, 2006. 90(12): p. 4692- 4700.

482. Jo, E.J., et al., α-synuclein membrane interactions and lipid specificity. Journal of Biological Chemistry, 2000. 275(44): p. 34328-34334.

483. Davidson, W.S., et al., Stabilization of α-synuclein secondary structure upon binding to synthetic membranes. Journal of Biological Chemistry, 1998. 273(16): p. 9443- 9449.

484. Perrin, R.J., et al., Interaction of human α-synuclein and Parkinson's disease variants with phospholipids - structural analysis using site-directed mutagenesis. Journal of Biological Chemistry, 2000. 275(44): p. 34393-34398.

210

485. Martinez, Z., et al., GM1 specifically interacts with α-synuclein and inhibits fibrillation. Biochemistry, 2007. 46(7): p. 1868-1877.

486. Di Pasquale, E., et al., Altered ion channel formation by the Parkinson's-disease- linked E46K mutant of α-synuclein is corrected by GM3 but not by GM1 gangliosides. J Mol Biol, 2010. 397(1): p. 202-18.

487. Park, J.Y., et al., On the mechanism of internalization of α-synuclein into microglia: roles of ganglioside GM1 and lipid raft. Journal of Neurochemistry, 2009. 110(1): p. 400-411.

488. Fantini, J. and N. Yahi, Molecular basis for the glycosphingolipid-binding specificity of α-synuclein: key role of tyrosine 39 in membrane insertion. Journal of Molecular Biology, 2011. 408(4): p. 654-669.

489. Bartels, T., et al., N-α-acetylation of α-synuclein increases its helical folding propensity, GM1 binding specificity and resistance to aggregation. Plos One, 2014. 9(7).

490. Zigoneanu, I.G., et al., Interaction of α-synuclein with vesicles that mimic mitochondrial membranes. Biochimica Et Biophysica Acta-Biomembranes, 2012. 1818(3): p. 512-519.

491. Nakamura, K., et al., Direct membrane association drives mitochondrial fission by the Parkinson disease-associated protein α-synuclein. Journal of Biological Chemistry, 2011. 286(23): p. 20710-20726.

492. Fantini, J., D. Carlus, and N. Yahi, The fusogenic tilted peptide (67-78) of α-synuclein is a cholesterol binding domain. Biochimica Et Biophysica Acta-Biomembranes, 2011. 1808(10): p. 2343-2351.

493. Middleton, E.R. and E. Rhoades, Effects of curvature and composition on α-synuclein binding to lipid vesicles. Biophysical Journal, 2010. 99(7): p. 2279-2288.

494. Pranke, I.M., et al., α-synuclein and ALPS motifs are membrane curvature sensors whose contrasting chemistry mediates selective vesicle binding. Journal of Cell Biology, 2011. 194(1): p. 88-102.

495. Kjaer, L., et al., The influence of vesicle size and composition on α-synuclein structure and stability. Biophysical Journal, 2009. 96(7): p. 2857-2870.

496. Jensen, M.B., et al., Membrane curvature sensing by amphipathic helices: a single liposome study using α-synuclein and annexin B12. Journal of Biological Chemistry, 2011. 286(49): p. 42603-42614.

497. Cui, H.S., E. Lyman, and G.A. Voth, Mechanism of membrane curvature sensing by amphipathic helix containing proteins. Biophysical Journal, 2011. 100(5): p. 1271- 1279.

498. Nuscher, B., et al., α-synuclein has a high affinity for packing defects in a bilayer membrane - a thermodynamics study. Journal of Biological Chemistry, 2004. 279(21): p. 21966-21975.

211

499. Irizarry, M.C., et al., Characterization of the precursor protein of the non-Aβ component of senile plaques (NACP) in the human central nervous system. J Neuropathol Exp Neurol, 1996. 55(8): p. 889-95.

500. Barria, M.A., D. Gonzalez-Romero, and C. Soto, Cyclic amplification of prion protein misfolding. Methods in molecular biology (Clifton, N.J.), 2012. 849: p. 199-212.

501. Saunders, S.E., J.C. Bartz, and R.A. Shikiya, Protein misfolding cyclic amplification of prions. Jove-Journal of Visualized Experiments, 2012(69).

502. Mays, C.E., et al., In vitro amplification of misfolded prion protein using lysate of cultured cells. Plos One, 2011. 6(3).

503. Roostaee, A., et al., Aggregation and neurotoxicity of recombinant α-synuclein aggregates initiated by dimerization. Molecular Neurodegeneration, 2013. 8.

504. Shahnawaz, M., et al., Development of a biochemical diagnosis of Parkinson disease by detection of α-synuclein misfolded aggregates in cerebrospinal fluid. Jama Neurology, 2017. 74(2): p. 163-172.

505. Vassar, P.S. and C.F.A. Culling, Fluorescent stains, with special reference to amyloid and connective tissue. Archives of Pathology, 1959. 68(5): p. 487-498.

506. Biancalana, M. and S. Koide, Molecular mechanism of Thioflavin-T binding to amyloid fibrils. Biochimica Et Biophysica Acta-Proteins and Proteomics, 2010. 1804(7): p. 1405-1412.

507. Hoyer, W., et al., Dependence of α-synuclein aggregate morphology on solution conditions. Journal of Molecular Biology, 2002. 322(2): p. 383-393.

508. Kamiyoshihara, T., et al., Observation of multiple intermediates in α-synuclein fibril formation by singular value decomposition analysis. Biochem Biophys Res Commun, 2007. 355(2): p. 398-403.

509. Serpell, L.C., et al., Fiber diffraction of synthetic α-synuclein filaments shows amyloid-like cross-β conformation. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(9): p. 4897-4902.

510. Uversky, V.N., J. Li, and A.L. Fink, Metal-triggered structural transformations, aggregation, and fibrillation of human α-synuclein - a possible molecular link between Parkinson's disease and heavy metal exposure. Journal of Biological Chemistry, 2001. 276(47): p. 44284-44296.

511. Munishkina, L.A., A.L. Fink, and V.N. Uversky, Concerted action of metals and macromolecular crowding on the fibrillation of α-synuclein. Protein and Peptide Letters, 2008. 15(10): p. 1079-1085.

512. Bharathi, S.S. Indi, and K.S.J. Rao, Copper- and iron-induced differential fibril formation in α-synuclein: TEM study. Neuroscience Letters, 2007. 424(2): p. 78-82.

513. Korzeniewski, C. and D.M. Callewaert, An enzyme-release assay for natural cytotoxicity. Journal of Immunological Methods, 1983. 64(3): p. 313-320.

212

514. Narhi, L., et al., Both familial Parkinson's disease mutations accelerate α-synuclein aggregation. Journal of Biological Chemistry, 1999. 274(14): p. 9843-9846.

515. Conway, K.A., et al., Acceleration of oligomerization, not fibrillization, is a shared property of both α-synuclein mutations linked to early-onset Parkinson's disease: implications for pathogenesis and therapy. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(2): p. 571-576.

516. Greenbaum, E.A., et al., The E46K mutation in α-synuclein increases amyloid fibril formation. Journal of Biological Chemistry, 2005. 280(9): p. 7800-7807.

517. Pangborn, M.C., Isolation and purification of a serologically active phospholipid from beef heart. Journal of Biological Chemistry, 1942. 143(1): p. 247-256.

518. Reeve, A.K., et al., Aggregated α-synuclein and complex I deficiency: exploration of their relationship in differentiated neurons. Cell Death & Disease, 2015. 6: p. e1820- e1820.

519. Parihar, M.S., et al., Mitochondrial association of α-synuclein causes oxidative stress. Cellular and Molecular Life Sciences, 2008. 65(7-8): p. 1272-1284.

520. Nicklas, W.J., I. Vyas, and R.E. Heikkila, Inhibition of NADH-linked oxidation in brain mitochondria by 1-methyl-4-phenyl-pyridine, a metabolite of the neurotoxin, 1- methyl-4-phenyl-1,2,5,6-tetrahydropyridine. Life Sciences, 1985. 36(26): p. 2503- 2508.

521. Tipton, K.F. and T.P. Singer, Advances in our understanding of the mechanisms of the neurotoxicity of MPTP and related-compounds. Journal of Neurochemistry, 1993. 61(4): p. 1191-1206.

522. Glinka, Y., M. Gassen, and M.B.H. Youdim, Mechanism of 6-hydroxydopamine neurotoxicity. Journal of Neural Transmission-Supplement, 1997(50): p. 55-66.

523. Fukushima, T., et al., Mechanism of cytotoxicity of paraquat I: NADH oxidation and paraquat radical formation via complex I. Experimental and Toxicologic Pathology, 1993. 45(5-6): p. 345-349.

524. Tawara, T., et al., Effects of paraquat on mitochondrial electron transport system and catecholamine contents in rat brain. Archives of Toxicology, 1996. 70(9): p. 585-589.

525. Cocheme, H.M. and M.P. Murphy, Complex I is the major site of mitochondrial superoxide production by paraquat. Journal of Biological Chemistry, 2008. 283(4): p. 1786-1798.

526. Brand, Martin D. and David G. Nicholls, Assessing mitochondrial dysfunction in cells. Biochemical Journal, 2011. 435(Pt 2): p. 297-312.

527. Annesley, S.J., et al., Immortalized Parkinson's disease lymphocytes have enhanced mitochondrial respiratory activity. Disease Models & Mechanisms, 2016. 9(11): p. 1295-1305.

213

528. de Kroon, A., et al., Phospholipid composition of highly purified mitochondrial outer membranes of rat liver and Neurospora crassa. Is cardiolipin present in the mitochondrial outer membrane? Biochimica Et Biophysica Acta-Biomembranes, 1997. 1325(1): p. 108-116.

529. Fry, M. and D.E. Green, Cardiolipin requirement for electron transfer in complex I and complex III of the mitochondrial respiratory chain. Journal of Biological Chemistry, 1981. 256(4): p. 1874-1880.

530. Paradies, G., et al., Reactive oxygen species affect mitochondrial electron transport complex I activity through oxidative cardiolipin damage. Gene, 2002. 286(1): p. 135- 141.

531. Drose, S., K. Zwicker, and U. Brandt, Full recovery of the NADH : ubiquinone activity of complex I (NADH : ubiquinone oxidoreductase) from Yarrowia lipolytica by the addition of phospholipids. Biochimica Et Biophysica Acta-Bioenergetics, 2002. 1556(1): p. 65-72.

532. Schwall, C.T., V.L. Greenwood, and N.N. Alder, The stability and activity of respiratory Complex II is cardiolipin-dependent. Biochimica Et Biophysica Acta- Bioenergetics, 2012. 1817(9): p. 1588-1596.

533. Gomez, B. and N.C. Robinson, Phospholipase digestion of bound cardiolipin reversibly inactivates bovine cytochrome bc(1). Biochemistry, 1999. 38(28): p. 9031- 9038.

534. Lange, C., et al., Specific roles of protein-phospholipid interactions in the yeast cytochrome bc(1) complex structure. Embo Journal, 2001. 20(23): p. 6591-6600.

535. Robinson, N.C., Functiona binding of cardiolipin to cytochrome c oxidase. Journal of Bioenergetics and Biomembranes, 1993. 25(2): p. 153-163.

536. Eble, K.S., et al., Tightly associated cardiolipin in the bovine heart mitochondria ATP synthase as analyzed by 31P nuclear magnetic resonance spectroscopy. Journal of Biological Chemistry, 1990. 265(32): p. 19434-19440.

537. Ozawa, T., M. Tanaka, and T. Wakabayashi, Crystallization of mitochondrial cytochrome oxidase. Proceedings of the National Academy of Sciences of the United States of America-Biological Sciences, 1982. 79(23): p. 7175-7179.

538. Acin-Perez, R., et al., Respiratory active mitochondrial supercomplexes. Molecular Cell, 2008. 32(4): p. 529-539.

539. Schagger, H., Respiratory chain supercomplexes of mitochondria and . Biochimica Et Biophysica Acta-Bioenergetics, 2002. 1555(1-3): p. 154-159.

540. Wittig, I., et al., Supercomplexes and subcomplexes of mitochondrial oxidative phosphorylation. Biochimica Et Biophysica Acta-Bioenergetics, 2006. 1757(9-10): p. 1066-1072.

214

541. Claypool, S.M., et al., Cardiolipin defines the interactome of the major ADP/ATP carrier protein of the mitochondrial inner membrane. Journal of Cell Biology, 2008. 182(5): p. 937-950.

542. Pfeiffer, K., et al., Cardiolipin stabilizes respiratory chain supercomplexes. Journal of Biological Chemistry, 2003. 278(52): p. 52873-52880.

543. Zhang, M., E. Mileykovskaya, and W. Dowhan, Gluing the respiratory chain together - cardiolipin is required for supercomplex formation in the inner mitochondrial membrane. Journal of Biological Chemistry, 2002. 277(46): p. 43553-43556.

544. Brandolin, G., et al., Kinetic, binding and ultrastructural properties of the beef heart adenine nucleotide carrier protein after incorporation into phospholipid vesicles. Biochim Biophys Acta, 1980. 592(3): p. 592-614.

545. Krämer, R. and M. Klingenberg, Enhancement of reconstituted ADP,ATP exchange activity by phosphatidylethanolamine and by anionic phospholipids. FEBS Letters, 1980. 119(2): p. 257-260.

546. Pebay-Peyroula, E., et al., Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside. Nature, 2003. 426(6962): p. 39-44.

547. Nury, H., et al., Relations between structure and function of the mitochondrial ADP/ATP carrier, in Annual Review of Biochemistry. 2006. p. 713-741.

548. Kagan, V.E., et al., Cytochrome c acts as a cardiolipin oxygenase required for release of proapoptotic factors. Nature Chemical Biology, 2005. 1(4): p. 223-232.

549. Petrosillo, G., et al., Reactive oxygen species generated from the mitochondrial electron transport chain induce cytochrome c dissociation from beef-heart submitochondrial particles via cardiolipin peroxidation. Possible role in the apoptosis. Febs Letters, 2001. 509(3): p. 435-438.

550. Nakamura, K., et al., Optical reporters for the conformation of α-synuclein reveal a specific interaction with mitochondria. Journal of Neuroscience, 2008. 28(47): p. 12305-12317.

551. Cole, N.B., et al., Mitochondrial translocation of α-synuclein is promoted by intracellular acidification. Experimental Cell Research, 2008. 314(10): p. 2076-2089.

552. Shavali, S., et al., Mitochondrial localization of α-synuclein protein in α-synuclein overexpressing cells. Neuroscience Letters, 2008. 439(2): p. 125-128.

553. McFarland, M.A., et al., Proteomics analysis identifies phosphorylation-dependent α- synuclein protein interactions. Molecular & Cellular Proteomics, 2008. 7(11): p. 2123-2137.

554. Perfeito, R., et al., Linking α-synuclein phosphorylation to reactive oxygen species formation and mitochondrial dysfunction in SH-SY5Y cells. Molecular and Cellular Neuroscience, 2014. 62: p. 51-59.

215

555. Guardia-Laguarta, C., et al., α-synuclein Is localized to mitochondria-associated ER membranes. Journal of Neuroscience, 2014. 34(1): p. 249-259.

556. Paillusson, S., et al., α-synuclein binds to the ER-mitochondria tethering protein VAPB to disrupt Ca2+ homeostasis and mitochondrial ATP production. Acta Neuropathologica, 2017. 134(1): p. 129-149.

557. Blesa, J., et al., Oxidative stress and Parkinson's disease. Frontiers in Neuroanatomy, 2015. 9.

558. Vilette, D., Cell models of prion infection. Veterinary Research, 2008. 39(4).

559. Haigh, C.L., et al., Acute exposure to prion infection induces transient oxidative stress progressing to be cumulatively deleterious with chronic propagation in vitro. Free Radical Biology and Medicine, 2011. 51(3): p. 594-608.

560. Watts, J.C. and S.B. Prusiner, Mouse models for studying the formation and propagation of prions. J Biol Chem, 2014. 289(29): p. 19841-9.

561. Scott, M., et al., Transgenic mice expressing hamster prion protein produce species- specific scrapie infectivity and amyloid plaques. Cell, 1989. 59(5): p. 847-857.

562. Korth, C., et al., Acridine and phenothiazine derivatives as pharmacotherapeutics for prion disease. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98(17): p. 9836-9841.

563. Kocisko, D.A., et al., New inhibitors of scrapie-associated prion protein formation in a library of 2,000 drugs and natural products. Journal of Virology, 2003. 77(19): p. 10288-10294.

564. Doh-Ura, K., T. Iwaki, and B. Caughey, Lysosomotropic agents and cysteine protease inhibitors inhibit scrapie-associated prion protein accumulation. Journal of Virology, 2000. 74(10): p. 4894-4897.

565. Collinge, J., et al., Safety and efficacy of quinacrine in human prion disease (PRION- 1 study): a patient-preference trial. Lancet Neurology, 2009. 8(4): p. 334-344.

566. Geschwind, M.D., et al., Quinacrine treatment trial for sporadic Creutzfeldt-Jakob disease. Neurology, 2013. 81(23): p. 2015-2023.

567. Haik, S., et al., Compassionate use of quinacrine in Creutzfeldt-Jakob disease fails to show significant effects. Neurology, 2004. 63(12): p. 2413-2415.

568. Martinez-Lage, J.F., et al., Creutzfeldt-Jakob disease acquired via a dural graft: failure of therapy with quinacrine and chlorpromazine. Surgical Neurology, 2005. 64(6): p. 542-545.

569. Benito-Leon, J., Combined quinacrine and chlorpromazine therapy in fatal familial insomnia. Clinical Neuropharmacology, 2004. 27(4): p. 201-203.

570. Collins, S.J., et al., Quinacrine does not prolong survival in a murine Creutzfeldt- Jakob disease model. Annals of Neurology, 2002. 52(4): p. 503-506.

216

571. Doh-Ura, K., et al., Treatment of transmissible spongiform encephalopathy by intraventricular drug infusion in animal models. Journal of Virology, 2004. 78(10): p. 4999-5006.

572. Ghaemmaghami, S., et al., Continuous quinacrine treatment results in the formation of drug-resistant prions. Plos Pathogens, 2009. 5(11).

573. Bian, J.F., H.E. Kang, and G.C. Telling, Quinacrine promotes replication and conformational mutation of chronic wasting disease prions. Proceedings of the National Academy of Sciences of the United States of America, 2014. 111(16): p. 6028-6033.

574. Kimberlin, R.H. and C.A. Walker, Antiviral compound effective against experimental scrapie. Lancet, 1979. 2(8142): p. 591-592.

575. Tateishi, J., Antibiotics and antivirals do not modify experimentally-induced Creutzfeldt-Jakob disease in mice. Journal of Neurology Neurosurgery and Psychiatry, 1981. 44(8): p. 723-724.

576. Terzano, M.G., et al., The effect of amantadine on arousal and EEG patterns in Creutzfeldt-Jakob disease. Archives of Neurology, 1983. 40(9): p. 555-559.

577. Neri, G., et al., Amantadine in Creutzfeldt-Jakob disease. Review of the literature and case contribution. Riv Neurobiol, 1984. 30(1): p. 47-56.

578. Otto, M., et al., Efficacy of flupirtine on cognitive function in patients with CJD - a double-blind study. Neurology, 2004. 62(5): p. 714-718.

579. Barret, A., et al., Evaluation of quinacrine treatment for prion diseases. Journal of Virology, 2003. 77(15): p. 8462-8469.

580. De Luigi, A., et al., The efficacy of tetracyclines in peripheral and intracerebral prion infection. Plos One, 2008. 3(3).

581. Tagliavini, F., Prion therapy: tetracyclic compounds in animal models and patients with Creutzfeldt-Jakob disease. Alzheimer's & Dementia, 2008. 4(4, Supplement): p. T149-T150.

582. Zerr, I., Therapeutic trials in human Transmissible Spongiform Encephalopathies: recent advances and problems to address. Infectious Disorders - Drug Targets, 2009. 9(1): p. 92-99.

583. Haik, S., et al., Doxycycline in Creutzfeldt-Jakob disease: a phase 2, randomised, double-blind, placebo-controlled trial. Lancet Neurology, 2014. 13(2): p. 150-158.

584. Diringer, H. and B. Ehlers, Chemoprophylaxis of scrapie in mice. Journal of General Virology, 1991. 72: p. 457-460.

585. Ehlers, B. and H. Diringer, Dextran sulfate-500 delays and prevents mouse scrapie by impairment of agent replication in spleen. Journal of General Virology, 1984. 65(AUG): p. 1325-1330.

217

586. Ladogana, A., et al., Sulphate polyanions prolong the incubation period of scrapie- infected hamsters. Journal of General Virology, 1992. 73: p. 661-665.

587. Farquhar, C., A. Dickinson, and M. Bruce, Prophylactic potential of pentosan polysulphate in transmissible spongiform encephalopathies. Lancet, 1999. 353(9147): p. 117-117.

588. Tsuboi, Y., K. Doh-ura, and T. Yamada, Continuous intraventricular infusion of pentosan polysulfate: clinical trial against prion diseases. Neuropathology, 2009. 29(5): p. 632-636.

589. Bone, I., et al., Intraventricular pentosan polysulphate in human prion disease: an observational study in the UK. European Journal of Neurology, 2008. 15(5): p. 458- 464.

590. Newman, P.K., et al., Postmortem findings in a case of variant Creutzfeldt-Jakob disease treated with intraventricular pentosan polysulfate. Journal of Neurology Neurosurgery and Psychiatry, 2014. 85(8): p. 919-922.

591. Parry, A., et al., Long term survival in a patient with variant Creutzfeldt-Jakob disease treated with intraventricular pentosan polysulphate. Journal of Neurology Neurosurgery and Psychiatry, 2007. 78(7): p. 733-734.

592. Terada, T., et al., Less protease-resistant PrP in a patient with sporadic CJD treated with intraventricular pentosan polysulphate. Acta Neurologica Scandinavica, 2010. 121(2): p. 127-130.

593. Whittle, I.R., R.S.G. Knight, and R.G. Will, Unsuccessful intraventricular pentosan polysulphate treatment of variant Creutzfeldt-Jakob disease. Acta Neurochirurgica, 2006. 148(6): p. 677-679.

594. Harrison, R.G., Observations on the living developing nerve fiber. Proceedings of the Society for Experimental Biology and Medicine, 1907. 4: p. 140-143.

595. Reinbold, R., Organotypic differentiation of the eye of the chick embryo in vitro. Comptes Rendus Des Seances De La Societe De Biologie Et De Ses Filiales, 1954. 148(15-1): p. 1493-1495.

596. Monesi, V., Differentiation of argyrophil and argentaffin cells in organotypic cultures of embryonic chick intestine. Journal of Embryology and Experimental Morphology, 1960. 8(3): p. 302-313.

597. Loffredo Sampaolo, C., Chick embryo heart in organotypic culture. Boll Soc Ital Biol Sper, 1956. 32(12): p. 1580-2.

598. Wolf, M.K., Anatomy of cultured mouse cerebellum. II. Organotypic migration of granule cells demonstrated by silver impregnation of normal and mutant cultures. Journal of Comparative Neurology, 1970. 140(3): p. 281-&.

599. Hauw, J.J., B. Berger, and R. Escourolle, Presence of synapse in organotypical culture in vitro from human cerebellum. Comptes Rendus Hebdomadaires Des Seances De L Academie Des Sciences Serie D, 1972. 274(2): p. 264-+.

218

600. Wolf, M.K., Anatomy of cultured mouse cerebellum. I. Golgi and electron microscopic demonstrations of granule cells, their afferent and efferent synapses. Journal of Comparative Neurology, 1970. 140(3): p. 261-&.

601. Ansevin, K.D. and B.V. Lipps, Tissue plate - new technique for long-term organotypic culture. In Vitro-Journal of the Tissue Culture Association, 1973. 8(6): p. 483-488.

602. Boyd, W.H., Chamber for organotypic culture - adapted for growing large volumes of tissue. Stain Technology, 1971. 46(2): p. 85-&.

603. Gahwiler, B.H., Organotypic monoloayer-cultures of nervous-tissue. Journal of Neuroscience Methods, 1981. 4(4): p. 329-342.

604. Gahwiler, B.H., Slice cultures of cerebellar, hippocampal and hypothalamic tissue. Experientia, 1984. 40(3): p. 235-243.

605. Gahwiler, B.H., Morphological differentiation of nerve cells in thin organotypic cultures derived from rat hippocamus and cerebellum. Proceedings of the Royal Society Series B-Biological Sciences, 1981. 211(1184): p. 287-+.

606. Gahwiler, B.H., Organotypic cultures of neural tissue. Trends in Neurosciences, 1988. 11(11): p. 484-489.

607. Gahwiler, B.H., et al., Continuous presence of nerve growth factor is required for maintenance of cholinergic septal neurons in organotypic slice cultures. Neuroscience, 1990. 36(1): p. 27-31.

608. Stoppini, L., P.A. Buchs, and D. Muller, A simple method for organotypic cultures of nervous-tissue. Journal of Neuroscience Methods, 1991. 37(2): p. 173-182.

609. Gahwiler, B.H., et al., Organotypic slice cultures: a technique has come of age. Trends in Neurosciences, 1997. 20(10): p. 471-477.

610. Egert, U., et al., A novel organotypic long-term culture of the rat hippocampus on substrate-integrated multielectrode arrays. Brain Research Protocols, 1998. 2(4): p. 229-242.

611. Jahnsen, H., et al., Coupling of organotypic brain slice cultures to silicon-based arrays of electrodes. Methods-a Companion to Methods in Enzymology, 1999. 18(2): p. 160-172.

612. Karpiak, V.C. and D. Plenz, Preparation and maintenance of organotypic cultures for multi-electrode array recordings. Curr Protoc Neurosci, 2002. Chapter 6: p. Unit 6.15.

613. Dong, H.W. and D.V. Buonomano, A technique for repeated recordings in cortical organotypic slices. Journal of Neuroscience Methods, 2005. 146(1): p. 69-75.

614. Stoppini, L., P.A. Buchs, and D. Muller, Lesion-induced neurite sprouting and synapse formation in hippocampal organotypic cultures. Neuroscience, 1993. 57(4): p. 985-994.

219

615. Falsig, J. and A. Aguzzi, The prion organotypic slice culture assay - POSCA. Nature Protocols, 2008. 3(4): p. 555-562.

616. Falsig, J., et al., A versatile prion replication assay in organotypic brain slices. Nature Neuroscience, 2008. 11(1): p. 109-117.

617. Campeau, J.L., et al., Early increase and late decrease of purkinje cell dendritic spine density in prion-infected organotypic mouse cerebellar cultures. Plos One, 2013. 8(12).

618. Wolf, H., et al., Deposition pattern and subcellular distribution of disease-associated prion protein in cerebellar organotypic slice cultures infected with scrapie. Front Neurosci, 2015. 9: p. 410.

619. Williams, R.W. and K. Herrup, The control of neuron number. Annu Rev Neurosci, 1988. 11: p. 423-53.

620. Herculano-Houzel, S., The Human Brain in Numbers: A Linearly Scaled-up Primate Brain. Frontiers in Human Neuroscience, 2009. 3: p. 31.

621. Rockel, A.J., R.W. Hiorns, and T.P. Powell, The basic uniformity in structure of the neocortex. Brain, 1980. 103(2): p. 221-44.

622. Herculano-Houzel, S., C. Watson, and G. Paxinos, Distribution of neurons in functional areas of the mouse cerebral cortex reveals quantitatively different cortical zones. Front Neuroanat, 2013. 7: p. 35.

623. Stone, T.W., Neuropharmacology of quinolinic and kynurenic acids. Pharmacological Reviews, 1993. 45(3): p. 309-379.

624. Staal, J.A., et al., Characterization of cortical neuronal and glial alterations during culture of organotypic whole brain slices from neonatal and mature mice. Plos One, 2011. 6(7).

625. Bahr, B.A., Long-term hippocampal slices: A model system for investigating synaptic mechanisms and pathological processes. Journal of Neuroscience Research, 1995. 42(3): p. 294-305.

626. Brazier, M.W., et al., Manganese chelation therapy extends survival in a mouse model of M1000 prion disease. Journal of Neurochemistry, 2010. 114(2): p. 440-451.

627. Tateishi, J., et al., Transmission of chronic spongiform encephalopathy with kuru plaques from humans to small rodents. Annals of Neurology, 1979. 5(6): p. 581-584.

628. Janzer, R.C. and M.C. Raff, Astrocytes induce blood-brain barrier properties in endothelial cells. Nature, 1987. 325(6101): p. 253-7.

629. Sobue, K., et al., Induction of blood-brain barrier properties in immortalized bovine brain endothelial cells by astrocytic factors. Neuroscience Research, 1999. 35(2): p. 155-164.

220

630. Tontsch, U. and H.C. Bauer, Glial-cells and neurons induce blood brain barrier related enzymes in cultured cerebral endothelial cells. Brain Research, 1991. 539(2): p. 247-253.

631. Isobe, I., et al., Astrocytic contributions to blood-brain barrier (BBB) formation by endothelial cells: a possible use of aortic endothelial cell for in vitro BBB model. Neurochemistry International, 1996. 28(5-6): p. 523-533.

632. Hayashi, Y., et al., Induction of various blood-brain barrier properties in non-neural endothelial cells by close apposition to co-cultured astrocytes. Glia, 1997. 19(1): p. 13-26.

633. Eugenin, E.A., et al., Human immunodeficiency virus infection of human astrocytes disrupts blood-brain barrier integrity by a gap junction-dependent mechanism. Journal of Neuroscience, 2011. 31(26): p. 9456-9465.

634. Margalith, I., et al., Polythiophenes inhibit prion propagation by stabilizing prion protein (PrP) aggregates. Journal of Biological Chemistry, 2012. 287(23): p. 18872- 18887.

635. Parchi, P., et al., Molecular basis of phenotypic variability in sporadic Creutzfeldt- Jakob disease. Ann Neurol, 1996. 39(6): p. 767-78.

636. Parchi, P., et al., Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects. Ann Neurol, 1999. 46(2): p. 224- 33.

637. Karapetyan, Y.E., et al., Prion strain discrimination based on rapid in vivo amplification and analysis by the cell panel assay. PLOS ONE, 2009. 4(5): p. e5730.

638. Chen, X.B., et al., Mass of the postsynaptic density and enumeration of three key molecules. Proceedings of the National Academy of Sciences of the United States of America, 2005. 102(32): p. 11551-11556.

639. Migaud, M., et al., Enhanced long-term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein. Nature, 1998. 396(6710): p. 433-439.

640. Sollner, T., et al., SNAP receptors implicated in vesicle targeting and fusion. Nature, 1993. 362(6418): p. 318-324.

641. Johnston, P.A. and T.C. Sudhof, The multisubunit structure of synaptophysin - relationship between bonding and homooligomerization. Journal of Biological Chemistry, 1990. 265(15): p. 8869-8873.

642. Calakos, N. and R.H. Scheller, Vesicle-associated membrane protein and synaptophysin are associated on the synaptic vesicle. Journal of Biological Chemistry, 1994. 269(40): p. 24534-24537.

643. Edelmann, L., et al., Synaptobrevin binding to synaptophysin - a potential mechanism for controlling the exocytotic fusion machine. Embo Journal, 1995. 14(2): p. 224-231.

221

644. Washbourne, P., G. Schiavo, and C. Montecucco, Vesicle-associated membrane protein-2 (synaptobrevin-2) forms a complex with synaptophysin. Biochemical Journal, 1995. 305: p. 721-724.

645. Zhang, Z., et al., The SNARE proteins SNAP25 and Synaptobrevin are involved in endocytosis at hippocampal synapses. Journal of Neuroscience, 2013. 33(21): p. 9169-9175.

646. Deak, F., et al., Synaptobrevin is essential for fast synaptic-vesicle endocytosis. Nature Cell Biology, 2004. 6(11): p. 1102-U21.

647. Jahn, R., et al., A 38,000-dalton membrane protein (p38) present in synaptic vesicles. Proceedings of the National Academy of Sciences of the United States of America, 1985. 82(12): p. 4137-4141.

648. Wiedenmann, B. and W.W. Franke, Identification and localization of synaptophysin, an integral membrane glyocoprotein of WR 38,000 characteristic of presynaptic vesicles. Cell, 1985. 41(3): p. 1017-1028.

649. Rehm, H., B. Wiedenmann, and H. Betz, Molecular characterization of synaptophysin, a major calcium-binding protein of the synaptic vesicle membrane. Embo Journal, 1986. 5(3): p. 535-541.

650. Wang, Y.C., et al., [The alterations in expression of the synaptophysin in rat brain trauma injury model-an experimental Study.]. Sichuan Da Xue Xue Bao Yi Xue Ban, 2016. 47(6): p. 852-856.

651. Thome, J., et al., Stress differentially regulates synaptophysin and synaptotagmin expression in hippocampus. Biological Psychiatry, 2001. 50(10): p. 809-812.

652. Hao, Y.X., A. Shabanpoor, and G.A. Metz, Stress and corticosterone alter synaptic plasticity in a rat model of Parkinson's disease. Neuroscience Letters, 2017. 651: p. 79-87.

653. Meng, Z.Z., et al., Effect of xiaoyaosan decoction on learning and memory deficit in rats induced by chronic immobilization stress. Evidence-Based Complementary and Alternative Medicine, 2013.

654. Boese, A.S., et al., MicroRNA abundance is altered in synaptoneurosomes during prion disease. Molecular and Cellular Neuroscience, 2016. 71: p. 13-24.

655. Siso, S., et al., Abnormal synaptic protein expression and cell death in murine scrapie. Acta Neuropathologica, 2002. 103(6): p. 615-626.

656. Cunningham, C., et al., Synaptic changes characterize early behavioural signs in the ME7 model of murine prion disease. European Journal of Neuroscience, 2003. 17(10): p. 2147-2155.

657. Davies, M.J., et al., Reduced expression of the presynaptic co-chaperone cysteine string protein α (CSP α) does not exacerbate experimentally-induced ME7 prion disease. Neuroscience Letters, 2015. 589: p. 138-143.

222

658. Ferrer, I., Synaptic pathology and cell death in the cerebellum in Creutzfeldt-Jakob disease. Cerebellum, 2002. 1(3): p. 213-22.

659. Hussain, S. and S. Davanger, Postsynaptic VAMP/synaptobrevin facilitates differential vesicle trafficking of GluA1 and GluA2 AMPA receptor subunits. Plos One, 2015. 10(10).

660. Antonucci, F., et al., SNAP-25, a known presynaptic protein with emerging postsynaptic functions. Frontiers in Synaptic Neuroscience, 2016. 8: p. 7.

661. Kawasaki, Y., et al., Orally administered amyloidophilic compound is effective in prolonging the incubation periods of animals cerebrally infected with prion diseases in a prion strain-dependent manner. Journal of Virology, 2007. 81(23): p. 12889- 12898.

662. Hwu, J.R., et al., Aminyl and iminyl radicals from arylhydrazones in the photo- induced DNA cleavage. Bioorganic & Medicinal Chemistry, 2004. 12(10): p. 2509- 2515.

663. Lu, D., et al., Biaryl amides and hydrazones as therapeutics for prion disease in transgenic mice. Journal of Pharmacology and Experimental Therapeutics, 2013. 347(2): p. 325-338.

664. Berry, D.B., et al., Drug resistance confounding prion therapeutics. Proceedings of the National Academy of Sciences of the United States of America, 2013. 110(44): p. E4160-E4169.

665. Aslund, A., et al., Novel pentameric thiophene derivatives for in vitro and in vivo optical imaging of a plethora of protein aggregates in cerebral amyloidoses. Acs Chemical Biology, 2009. 4(8): p. 673-684.

666. Herrmann, U.S., et al., Structure-based drug design identifies polythiophenes as antiprion compounds. Sci Transl Med, 2015. 7(299): p. 299ra123.

667. Axten, J.M., et al., Discovery of 7-Methyl-5-(1-{ 3-(trifluoromethyl)phenyl acetyl}- 2,3-dihydro-1H-indol-5 -yl)-7H-pyrrolo 2,3-d pyrimidin-4-amine (GSK2606414), a Potent and Selective First-in-Class Inhibitor of Protein Kinase R (PKR)-like Endoplasmic Reticulum Kinase (PERK). Journal of Medicinal Chemistry, 2012. 55(16): p. 7193-7207.

668. McCleery, J., D.A. Cohen, and A.L. Sharpley, Pharmacotherapies for sleep disturbances in dementia. Cochrane Database of Systematic Reviews, 2016(11).

669. Radford, H., et al., PERK inhibition prevents tau-mediated neurodegeneration in a mouse model of frontotemporal dementia. Acta Neuropathologica, 2015. 130(5): p. 633-642.

670. Fellner, L., et al., Anle138b partly ameliorates motor deficits despite failure of neuroprotection in a model of advanced multiple system atrophy. Frontiers in Neuroscience, 2016. 10.

223

671. Wagner, J., et al., Reducing tau aggregates with anle138b delays disease progression in a mouse model of tauopathies. Acta Neuropathologica, 2015. 130(5): p. 619-631.

672. Kahle, P.J., et al., Hyperphosphorylation and insolubility of α-synuclein in transgenic mouse oligodendrocytes. Embo Reports, 2002. 3(6): p. 583-588.

673. Stemberger, S., et al., Targeted overexpression of human α-synuclein in oligodendroglia induces lesions linked to MSA-like progressive autonomic failure. Experimental Neurology, 2010. 224(2): p. 459-464.

674. Yoshiyama, Y., et al., Synapse loss and microglial activation precede tangles in a P301S tauopathy mouse model. Neuron, 2007. 53(3): p. 337-351.

675. Deeg, A.A., et al., Anle138b and related compounds are aggregation specific fluorescence markers and reveal high affinity binding to α-synuclein aggregates. Biochimica Et Biophysica Acta-General Subjects, 2015. 1850(9): p. 1884-1890.

676. Culvenor, J.G., et al., Non-Aβ component of Alzheimer's disease amyloid (NAC) revisited - NAC and α-synuclein are not associated with Aβ amyloid. American Journal of Pathology, 1999. 155(4): p. 1173-1181.

677. Jakes, R., et al., Epitope mapping of LB509, a monoclonal antibody directed against human α-synuclein. Neuroscience Letters, 1999. 269(1): p. 13-16.

678. Lippa, C.F., et al., Lewy bodies contain altered α-synuclein in brains of many familiar Alzheimer's disease patients with mutations in presenilin and amyloid precursor protein genes. American Journal of Pathology, 1998. 153(5): p. 1365-1370.

679. Miake, H., et al., Biochemical characterization of the core structure of α-synuclein filaments. Journal of Biological Chemistry, 2002. 277(21): p. 19213-19219.

680. Turriani, E., et al., Treatment with diphenyl–pyrazole compound anle138b/c reveals that α-synuclein protects melanoma cells from autophagic cell death. Proceedings of the National Academy of Sciences, 2017.

681. Bertoni, J.M., et al., Increased melanoma risk in Parkinson disease a prospective clinicopathological study. Archives of Neurology, 2010. 67(3): p. 347-352.

682. Paisan-Ruiz, C. and H. Houlden, Common pathogenic pathways in melanoma and Parkinson disease. Neurology, 2010. 75(18): p. 1653-1655.

683. Ferreira, J.J., et al., Skin cancer and Parkinson's disease. Movement Disorders, 2010. 25(2): p. 139-148.

684. Bajaj, A., J.A. Driver, and E.S. Schernhammer, Parkinson's disease and cancer risk: a systematic review and meta-analysis. Cancer Causes & Control, 2010. 21(5): p. 697-707.

685. Olsen, J.H., et al., Parkinson disease and malignant melanoma in first-degree relatives of patients with early-onset melanoma. Epidemiology, 2011. 22(1): p. 109- 112.

224

686. Kareus, S.A., et al., Shared predispositions of Parkinsonism and cancer a population- based pedigree-linked study. Archives of Neurology, 2012. 69(12): p. 1572-1577.

687. Meng, S.S., et al., No association between Parkinson disease alleles and the risk of melanoma. Cancer Epidemiology Biomarkers & Prevention, 2012. 21(1): p. 243-245.

688. Dong, J., et al., Susceptibility loci for pigmentation and melanoma in relation to Parkinson's disease. Neurobiology of Aging, 2014. 35(6).

689. Wirdefeldt, K., et al., Parkinson's disease and cancer: a register-based family study. American Journal of Epidemiology, 2014. 179(1): p. 85-94.

690. Weissmann, C., A unified theory of prion propagation. Nature, 1991. 352(6337): p. 679-683.

691. Couzin, J., Biomedicine. An end to the prion debate? Don't count on it. Science, 2004. 305(5684): p. 589-589.

692. Andreoletti, O., et al., Astrocytes accumulate 4-hydroxynonenal adducts in murine scrapie and human Creutzfeldt-Jakob disease. Neurobiol Dis, 2002. 11(3): p. 386-93.

693. Bose, A. and M.F. Beal, Mitochondrial dysfunction in Parkinson's disease. J Neurochem, 2016. 139 Suppl 1: p. 216-231.

694. Gu, M., et al., Mitochondrial respiratory chain function in multiple system atrophy. Movement Disorders, 1997. 12(3): p. 418-422.

695. Spano, M., et al., The possible involvement of mitochondrial dysfunctions in Lewy body dementia: a systematic review. Functional Neurology, 2015. 30(3): p. 151-158.

696. Conway, K.A., et al., Kinetic stabilization of the α-synuclein protofibril by a dopamine-α-synuclein adduct. Science, 2001. 294(5545): p. 1346-9.

697. Lee, H.J., et al., Dopamine promotes formation and secretion of non-fibrillar α- synuclein oligomers. Experimental and Molecular Medicine, 2011. 43(4): p. 216-222.

698. Prasad, K.N., et al., Establishment and characterization of immortalized clonal cell lines from fetal rat mesencephalic tissue. In Vitro Cell Dev Biol Anim, 1994. 30a(9): p. 596-603.

699. Cheung, Y.T., et al., Effects of all-trans-retinoic acid on human SH-SY5Y neuroblastoma as in vitro model in neurotoxicity research. Neurotoxicology, 2009. 30(1): p. 127-35.

700. Davies, C.A., et al., A quantitative morphometric analysis of the neuronal and synaptic content of the frontal and temporal cortex in patients with Alzheimer's disease. Journal of the Neurological Sciences, 1987. 78(2): p. 151-164.

701. DeKosky, S.T. and S.W. Scheff, Synapse loss in frontal cortex biopsies in Alzheimer's disease: correlation with cognitive severity. Ann Neurol, 1990. 27(5): p. 457-64.

225

702. Terry, R.D., et al., Physical basis of cognitive alterations in Alzheimer's disease - synapse loss is the major correlate of cognitive impairment. Annals of Neurology, 1991. 30(4): p. 572-580.

703. Bereczki, E., et al., Synaptic proteins predict cognitive decline in Alzheimer's disease and Lewy body dementia. Alzheimers Dement, 2016. 12(11): p. 1149-1158.

704. Vallortigara, J., et al., Decreased levels of VAMP2 and monomeric α-synuclein correlate with duration of dementia. Journal of Alzheimers Disease, 2016. 50(1): p. 101-110.

705. Darios, F., et al., α-synuclein sequesters arachidonic acid to modulate SNARE- mediated exocytosis. Embo Reports, 2010. 11(7): p. 528-533.

706. Lou, X.C., et al., α-synuclein may cross-bridge v-SNARE and acidic phospholipids to facilitate SNARE-dependent vesicle docking. Biochemical Journal, 2017. 474: p. 2039-2049.

707. Makioka, K., et al., Involvement of endoplasmic reticulum stress defined by activated unfolded protein response in multiple system atrophy. Journal of the Neurological Sciences, 2010. 297(1-2): p. 60-65.

708. White, M.D., et al., Single treatment with RNAi against prion protein rescues early neuronal dysfunction and prolongs survival in mice with prion disease. Proceedings of the National Academy of Sciences of the United States of America, 2008. 105(29): p. 10238-10243.

709. Feng, X.R., et al., Allele-specific silencing of Alzheimer's disease genes - The amyloid precursor protein genes with Swedish or London mutations. Gene, 2006. 371(1): p. 68-74.

710. de Ynigo-Mojado, L., I. Martin-Ruiz, and J.D. Sutherland, Efficient allele-specific targeting of LRRK2 R1441 mutations mediated by RNAi. Plos One, 2011. 6(6).

711. Wang, H.Y., et al., Therapeutic delivered by a chemically modified small interfering RNA against mutant SOD1 slows amyotrophic lateral sclerosis progression. Journal of Biological Chemistry, 2008. 283(23): p. 15845-15852.

712. Carroll, J.B., et al., Potent and selective antisense oligonucleotides targeting single- nucleotide polymorphisms in the Huntington disease gene/allele-specific silencing of mutant huntingtin. Molecular Therapy, 2011. 19(12): p. 2178-2185.

713. Schwank, G., et al., Functional repair of CFTR by CRISPR/Cas9 in intestinal stem cell organoids of Cystic Fibrosis patients. Cell Stem Cell, 2013. 13(6): p. 653-658.

714. Boudreau, R.L., et al., Nonallele-specific silencing of mutant and wild-type huntingtin demonstrates therapeutic efficacy in Huntington's disease mice. Molecular Therapy, 2009. 17(6): p. 1053-1063.

715. Kang, S.G., et al., Establishment and characterization of Prnp knockdown neuroblastoma cells using dual microRNA-mediated RNA interference. Prion, 2011. 5(2): p. 93-102.

226

716. Boese, A.S., et al., Small RNA drugs for prion disease: a new frontier. Expert Opinion on Drug Discovery, 2013. 8(10): p. 1265-1284.

717. Alvarez-Erviti, L., et al., Delivery of siRNA to the mouse brain by systemic injection of targeted exosomes. Nature Biotechnology, 2011. 29(4): p. 341-U179.

718. Zhuang, X.Y., et al., Treatment of brain inflammatory diseases by delivering exosome encapsulated anti-inflammatory drugs from the nasal region to the brain. Molecular Therapy, 2011. 19(10): p. 1769-1779.

719. Enari, M., E. Flechsig, and C. Weissmann, Scrapie prion protein accumulation by scrapie-infected neuroblastoma cells abrogated by exposure to a prion protein antibody. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98(16): p. 9295-9299.

720. Horiuchi, M. and B. Caughey, Specific binding of normal prion protein to the scrapie form via a localized domain initiates its conversion to the protease-resistant state. Embo Journal, 1999. 18(12): p. 3193-3203.

721. Feraudet, C., et al., Screening of 145 anti-PrP monoclonal antibodies for their capacity to inhibit PrPSc replication in infected cells. Journal of Biological Chemistry, 2005. 280(12): p. 11247-11258.

722. Pankiewicz, J., et al., Clearance and prevention of prion infection in cell culture by anti-PrP antibodies. European Journal of Neuroscience, 2006. 23(10): p. 2635-2647.

723. Peretz, D., et al., Antibodies inhibit prion propagation and clear cell cultures of prion infectivity. Nature, 2001. 412(6848): p. 739-743.

724. Perrier, V., et al., Anti-PrP antibodies block PrPSc replication in prion-infected cell cultures by accelerating PrP degradation. Journal of Neurochemistry, 2004. 89(2): p. 454-463.

725. White, A.R., et al., Monoclonal antibodies inhibit prion replication and delay the development of prion disease. Nature, 2003. 422(6927): p. 80-83.

726. Klohn, P.C., et al., PrP antibodies do not trigger mouse hippocampal neuron apoptosis. Science, 2012. 335(6064): p. 52-52.

727. Antonyuk, S.V., et al., Crystal structure of human prion protein bound to a therapeutic antibody. Proceedings of the National Academy of Sciences of the United States of America, 2009. 106(8): p. 2554-2558.

728. Chung, E., et al., Anti-PrPC monoclonal antibody infusion as a novel treatment for cognitive deficits in an Alzheimer's disease model mouse. Bmc Neuroscience, 2010. 11.

729. Bard, F., et al., Peripherally administered antibodies against amyloid β-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nature Medicine, 2000. 6(8): p. 916-919.

227

730. Banks, W.A., et al., Passage of amyloid β protein antibody across the blood-brain barrier in a mouse model of Alzheimer's disease. Peptides, 2002. 23(12): p. 2223- 2226.

731. Mandler, M., et al., Next-generation active immunization approach for synucleinopathies: implications for Parkinson's disease clinical trials. Acta Neuropathologica, 2014. 127(6): p. 861-879.

732. Games, D., et al., Reducing C-terminal-truncated α-synuclein by immunotherapy attenuates neurodegeneration and propagation in Parkinson's disease-like models. Journal of Neuroscience, 2014. 34(28): p. 9441-9454.

733. Bae, E.-J., et al., Antibody-aided clearance of extracellular α-synuclein prevents cell- to-cell aggregate transmission. Journal of Neuroscience, 2012. 32(39): p. 13454- 13469.

734. Masliah, E., et al., Effects of α-synuclein immunization in a mouse model of Parkinson's disease. Neuron, 2005. 46(6): p. 857-868.

735. Masliah, E., et al., Passive immunization reduces behavioral and neuropathological deficits in an α-synuclein transgenic model of Lewy body disease. Plos One, 2011. 6(4).

736. Boutajangout, A., et al., Passive immunization targeting pathological phospho-tau protein in a mouse model reduces functional decline and clears tau aggregates from the brain. Journal of Neurochemistry, 2011. 118(4): p. 658-667.

737. d'Abramo, C., et al., Tau passive immunotherapy in mutant P301L mice: antibody affinity versus specificity. Plos One, 2013. 8(4).

738. Yanamandra, K., et al., Anti-tau antibodies that block tau aggregate seeding in vitro markedly decrease pathology and improve cognition in vivo. Neuron, 2013. 80(2): p. 402-414.

739. Gros-Louis, F., et al., Intracerebroventricular infusion of monoclonal antibody or its derived Fab fragment against misfolded forms of SOD1 mutant delays mortality in a mouse model of ALS. Journal of Neurochemistry, 2010. 113(5): p. 1188-1199.

740. AFFiRiS, AFFiRiS announces top line results of first-in-human clinical study using AFFITOPE® PD03A, confirming immunogenicity and safety profile in Parkinson’s disease patients. 2017.

741. Prothena, Clinical results presented from Prothena's phase 1b study of PRX002/RG7935 demonstrating robust antibody CNS penetration and significant reduction of free serum α-synuclein in patients with Parkinson's disease. 2017, Prothena.

742. Reimann, R.R., et al., Differential toxicity of antibodies to the prion protein. PLoS Pathog, 2016. 12(1): p. e1005401.

228

743. Sadowski, M.J., et al., Anti-PrP Mab 6D11 suppresses PrPSc replication in prion infected myeloid precursor line FDC-P1/22L and in the lymphoreticular system in vivo. Neurobiology of Disease, 2009. 34(2): p. 267-278.

744. Petsch, B., et al., Biological effects and use of PrPSc- and PrP-specific antibodies generated by immunization with purified full-length native mouse prions. Journal of Virology, 2011. 85(9): p. 4538-4546.

229

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s: Ugalde, Cathryn

Title: Investigating pathogenic mechanisms associated with prion protein and -synuclein misfolding

Date: 2017

Persistent Link: http://hdl.handle.net/11343/198019

File Description: Thesis: Investigating pathogenic mechanisms associated with prion protein and -synuclein misfolding

Terms and Conditions: Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the copyright owner. The work may not be altered without permission from the copyright owner. Readers may only download, print and save electronic copies of whole works for their own personal non-commercial use. Any use that exceeds these limits requires permission from the copyright owner. Attribution is essential when quoting or paraphrasing from these works.