Glial Cells in Health and Disease

Glial cells in health and disease

Anjali Balakrishnan

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Biochemistry University of Toronto

Glial cells in health and disease

Anjali Balakrishnan

Doctor of Philosophy

Department of Biochemistry University of Toronto

2020

Abstract

Glial cells are often considered to be ‘supporting cast’ members in the nervous system, with ancillary roles in providing nutrient and structural support to neurons. However, glial cells have many essential roles, including the myelination of nerves to allow information to be transmitted rapidly and efficiently. My thesis has largely focused on the role of myelinating glial cells in health and disease. I first studied a population of malignant oligodendrocyte-like cells that form glial tumors in the central nervous system, called oligodendroglioma. For the rest of my thesis,

I focused on Schwann cells in the peripheral nervous system. In Chapter 2 of my thesis, I describe my investigation into the role of extracellular vesicles in controlling oligodendroglioma growth by mediating heterotypic and homotypic cell-cell interactions. I revealed that oligodendroglioma tumor cells secrete extracellular vesicles that carry cytotoxic cargo to induce cell death in neighboring cells. Furthermore, I implicated a gene involved in extracellular vesicle biogenesis, SMPD3, in negatively regulating oligodendroglioma growth by controlling the synthesis of extracellular vesicles (Chapter 2). I then studied the development of Schwann cells, and their transition into repair Schwann cells post nerve injury.

I characterized the dynamic expression patterns of a panel of transcriptional regulators during development and in repair Schwann cells post-injury (Chapter 3). I then used this panel of markers to ask whether the ets domain transcription factor Etv5, expressed transiently in

Schwann cell precursors, played a role in regulating Schwann cell development and in repair

Schwann cells by using a hypomorphic Etv5 mutant mouse model (Chapter 4). While Etv5 mutants had no apparent defects in Schwann cell development, I describe several important caveats and future considerations. Finally, I performed the first steps towards developing a non-integrative, triple transcription factor mediated lineage conversion strategy for the generation of induced repair Schwann cells from mouse embryonic fibroblasts (Chapter 5). In conclusion, I have gained new insights on how glial cells in a healthy and diseased state are regulated. My findings have therapeutic implications for the treatment of oligodendroglioma tumors in the central nervous system, and for peripheral nerve repair.

iii

Acknowledgments

First and foremost, I want to send out my love and gratitude to my supervisor, Dr. Carol Schuurmans. Thank you for taking my call that eventful afternoon and accepting me as a graduate student! Thank you for your patience, and for spoiling us with the time you take to train each one of us individually. Carol has been an empowering role model for women in STEM. I feel hopeful that I will be able to take her legacy forward in every role I pursue in the future. I want to thank my wonderful supervisory committee, Dr. Alain Dabdoub and Dr. Micheal Ohh for being a strong support system for me at the University of Toronto! I also want to thank Dr. Jennifer Chan, Dr. Jeff Biernaskie, and Dr. Hon Leong for their scientific counsel. I had the incredible opportunity to be a student at two wonderful universities through my graduate career, and I remain grateful for all the splendid experiences.

I would like to thank Dr. Dawn Zinyk, who has helped me from the first day I entered the lab. I thank her for showing me the ropes around the lab, all the hundred cloning steps that we performed together, and for patiently answering all my doubts. I want to thank Dr. Yacine Touahri for his friendship and help with nearly everything through the last seven years. It has been a wonderful bond which I will cherish in the times to come. I want to give a shout-out to all the brilliant students I have worked with- Humna Noman, Sajeevan Sujanthan, Marielle Balanaser. I have learnt something from all of them. I miss my daily discussions with you Humna! I am blessed to have worked along with all my wonderful lab mates over the years, with a special shout-out to- Dr. Rajiv Dixit, Natasha Klenin, Nobuhiko Tachibana, Luke Ajay David, Mary Chute, Sisu Han, Dr. Vorapin Chinchalongporn, and Lakshmy Vasan. You all made graduate life so much better and memorable! A special note of thanks to my closest friends- Chinmayee and Anupama. I do not know how I would have survived in Calgary without your companionship and miss you both terribly in Toronto!

I would like to thank Shakir Hasan, for just about everything ever. Thank you for bearing with all my mood swings and standing by patiently even when things got tough. I wish our journey ahead is a fabulous one! And finally, I would like to thank my family- Mom, Dad, Amit, Swati, Arun, and Arunima for being the most caring family ever. Thank you for always supporting my dreams, ambitions, and rebellions. Distance makes the heart grow fonder….and it did! I am blessed to be a part of the Balakrishnan household! I hope I make you proud!

Dedication This thesis is dedicated to my Mom and Dad. I am so proud to be your daughter. Ever thine, Ever mine, Ever ours.

Table of Contents

Acknowledgments...... iv

List of Tables ...... xii

List of Figures ...... xiii

Appendices ...... xvii

List of Abbreviations ...... xviii

Dissemination of work arising from this Thesis ...... xxiii

Other contributions during my thesis work ...... xxv

Introduction ...... 1

1.1 Glial cells: An overview ...... 2

1.2 Glioma: An Introduction ...... 2

1.2.1 General introduction on glioma subtypes ...... 2

1.2.2 Oligodendroglioma ...... 2

1.3 Extracellular vesicles: An Introduction ...... 4

1.3.1 A brief primer on EV classification, isolation, and biogenesis ...... 5

1.3.2 EV functions in the ‘healthy’ nervous system ...... 9

1.3.3 Roles of EVs in glioma ...... 13

1.4 Schwann Cells: An Introduction ...... 19

1.4.1 Glial cells of the peripheral nervous system: Schwann cells ...... 19

1.4.2 Genes regulating Schwann cell specification and differentiation ...... 27

1.4.3 Nerve growth factor receptor and Neuregulin signaling: Role in Schwann cell myelination ...... 31

1.4.4 Peripheral nerve injury and role of Schwann cells post injury ...... 33

1.4.5 Phenotype of a repair Schwann cell ...... 35

1.4.6 Identifying an alternate source for Schwann cells: reprogramming somatic cells to generate Schwann cells ...... 38

1.5 Hypothesis and specific aims of the thesis ...... 44

SMPD3-mediated extracellular vesicle biogenesis inhibits oligodendroglioma tumor growth ...... 45

2.1 Abstract ...... 46

2.2 Introduction ...... 46

2.3 Methods...... 47

2.3.1 Patient-derived tumor tissues and cells and study approval ...... 47

2.3.2 The Cancer Genome Atlas (TCGA) survey ...... 48

2.3.3 Animals ...... 48

2.3.4 BT088 and BT054 cell culture ...... 48

2.3.5 Small molecule inhibitors ...... 49

2.3.6 Mouse NSC isolation and culture ...... 49

2.3.7 AnnexinV- PI Apoptosis assay ...... 49

2.3.8 Conditioned media ...... 49

2.3.9 Incucyte live cell imaging ...... 50

2.3.10 CO-BT088 co-cultures ...... 50

2.3.11 Pellet assay ...... 51

2.3.12 Density gradient Ultracentrifugation ...... 51

2.3.13 Scanning Electron Microscopy ...... 51

2.3.14 Transmission Electron Microscopy ...... 52

2.3.15 Nanosight tracking analysis (NTA) and Nano-flow cytometry ...... 52

2.3.16 Molecular Cloning ...... 52

2.3.17 Transduction and transfection ...... 53

2.3.18 Tumor xenografts ...... 54

2.3.19 Western blotting ...... 54

2.3.20 Tissue Processing and Immunostaining ...... 54

2.3.21 Mass Spectrometry...... 55

2.3.22 Image analysis ...... 55

vii

2.3.23 Statistical analysis ...... 56

2.4 Results ...... 57

2.4.1 Oligodendroglioma cells exert non-cell autonomous effects within the tumor niche ...... 57

2.4.2 ODG secretomes have distinct bioactive effects but both include cytotoxic EVs ...... 60

2.4.3 ODG EVs induce proliferation followed by apoptosis ...... 63

2.4.4 ODG cells secrete bioactive EVs mainly in the 40-200 nm exosome size range ...... 66

2.4.5 Proteomic profiling reveals ODG cells secrete distinct vesicular cargo .... 69

2.4.6 Lower SMPD3 expression is associated with poor prognosis in oligodendroglioma patients ...... 71

2.4.7 SMPD3 inhibits ODG cell proliferation in vitro ...... 73

2.4.8 Knockdown of SMPD3 promotes proliferation of ODG cells in vitro ...... 76

2.4.9 SMPD3 knockdown facilitates ODG growth in vivo ...... 80

2.4.10 SMPD3 knockdown increases ODG invasiveness and growth in human cerebral organoid ...... 84

2.5 Discussion ...... 87

Temporal analysis of gene expression in the murine Schwann cell lineage and the acutely injured postnatal nerve ...... 90

3.2 Introduction ...... 92

3.3 Material and methods ...... 94

3.3.1 Animals ...... 94

3.3.2 Embryo processing...... 94

3.3.3 Surgery ...... 95

3.3.4 Immunohistochemistry ...... 95

3.3.5 RNA in situ hybridization ...... 96

3.3.6 Microscopy and image processing ...... 96

3.4 Results ...... 97

3.4.1 Expression of Schwann cell lineage markers in neural crest cells...... 98

viii

3.4.2 Expression of Schwann cell lineage markers in migratory NCC precursors ...... 102

3.4.3 Expression of Schwann cell lineage markers in Schwann cell precursors ...... 109

3.4.4 Expression of Schwann cell lineage markers in immature Schwann cells ...... 114

3.4.5 Expression of Schwann cell lineage markers in pro-myelinating Schwann cells ...... 117

3.4.6 Expression of Schwann cell lineage markers in mature neonatal (P7) and adult (P65) myelinating and non-myelinating Schwann cells ...... 121

3.4.7 Nerve injury triggers adult Schwann cells to recapitulate a unique pattern of embryonic glial lineage transcription factors ...... 130

3.5 Discussion ...... 134

3.5.1 Sustained expression of Sox9 and Sox10 across the Schwann cell lineage ...... 136

3.5.2 Transcriptional regulators expressed at early stages in the Schwann cell lineage ...... 137

3.5.3 Transcriptional regulators expressed at late stages in the Schwann cell lineage ...... 138

3.5.4 Injury activates SC lineage genes that recapitulate features of both Schwann cell precursors and pro-myelinating Schwann cells ...... 139

Etv5 is not required for Schwann cell development but is required to regulate the Schwann cell response to peripheral nerve injury ...... 143

4.1 Abstract ...... 144

4.2 Introduction ...... 145

4.3 Material and methods ...... 146

4.3.1 Animals and genotyping ...... 146

4.3.2 Embryo collection ...... 147

4.3.3 Peripheral nerve crush...... 147

4.3.4 Immunohistochemistry ...... 147

4.3.5 RNA in situ hybridization...... 148

4.3.6 Microscopy and image processing ...... 148

4.3.7 Statistical analysis ...... 148

4.4 Results ...... 149

4.4.1 Schwann cell precursors develop normally in E12.5 Etv5-/- peripheral ganglia ...... 149

4.4.2 Immature Schwann cells develop normally in E15.5 Etv5-/- peripheral ganglia ...... 152

4.4.3 Late immature Schwann cells/pro-myelinating Schwann cells are detected in E18.5 Etv5-/- peripheral ganglia and in the dorsal and ventral roots .... 153

4.4.4 Schwann cells populate the postnatal sciatic nerve in Etv5-/- pups and respond to injury normally with an expansion in number ...... 155

4.5 Discussion ...... 157

Glial transcription factor-based conversion of mouse embryonic fibroblasts to a repair Schwann cell identity ...... 159

5.1 Abstract ...... 160

5.2 Introduction ...... 161

5.3 Material and methods ...... 163

5.3.1 Vector Construction ...... 163

5.3.2 Animals and genotyping ...... 163

5.3.3 Cell culture and cell isolation ...... 163

5.3.4 Cell Transfection ...... 164

5.3.5 Dorsal root ganglia-Schwann cell co-culture assay ...... 164

5.3.6 Fluorescence-Activated Cell Sorting (FACs) ...... 165

5.3.7 Western blotting ...... 165

5.3.8 Quantitative reverse transcription polymerase chain reaction (RTqPCR) ...... 165

5.3.9 Tissue processing and Immunofluorescence assay ...... 166

5.4 Results ...... 166

5.4.1 Establishing a panel of markers to assess lineage conversion ...... 166

5.4.2 Mis-expression of Sox10 induces Schwann cell specific marker expression in fibroblasts...... 169

5.4.3 Design of triple transcription factor expression vectors for induction of a Schwann cell identity in fibroblasts ...... 172

5.4.4 Sox10-Jun-Sox2 (T2) combined with glial media induces expression of embryonic and repair specific Schwann cell markers...... 175

5.4.5 Ngfr+O4+ Schwann-like cells present with improved Schwann cell specific marker transcript expression ...... 179

5.4.6 Optimization of Dorsal root ganglia-Schwann cell co-culture assay for Schwann cell association and myelination assessment ...... 182

5.5 Discussion ...... 184

Discussion ...... 187

6.1 Summary of present findings ...... 188

6.1.1 Chapter 2: SMPD3-mediated extracellular vesicle biogenesis inhibits oligodendroglioma tumor growth ...... 188

6.1.2 Chapters 3-5: Schwann cells in a healthy and injured nerve ...... 195

6.2 Conclusions ...... 205

Appendices ...... 207

References ...... 218

List of Tables

Table 2.1: Summary of ODG patient tumors analyzed.

xii

List of Figures

Chapter 1:

Figure 1-1: Schematic of EV-mediated signaling between neighboring cells.

Figure 1.2: Schematic of cell-cell interactions in the brain with and without tumor cells.

Figure 1.3. Glial cells in the peripheral nervous system.

Figure 1.4. Trunk neural crest cell migration.

Figure 1.5. Cellular transitions steps in the Schwann cell lineage.

Figure 1.6. Schwann cells in peripheral nerve repair.

Figure 1.7. Direct cellular reprogramming approach to generate induced Schwann cells.

Chapter 2:

Figure 2.1. IDH mutant oligodendroglioma display non-cell autonomous activation of proliferation and receptor tyrosine kinase signaling.

Figure 2.2. Oligodendroglioma EVs induce apoptosis in neural stem cells.

Figure 2.3. Patient-derived oligodendroglioma cells secrete soluble and EV-enclosed bioactive factors.

Figure 2.4. Patient-derived oligodendroglioma cells secrete EV-enclosed bioactive factors which induce cytotoxic effects.

Figure 2.5. Oligodendroglioma EVs induce apoptosis in neural stem cells.

Figure 2.6. Bioactive exosomes are secreted from oligodendroglioma cell lines.

Figure 2.7. Proteomic profiling reveals distinct BT088 and BT054 vesiculomes and identifies VEGF signaling as a targetable growth-promoting pathway.

Figure 2.8. Poor prognosis in oligodendroglioma patients associated with low SMPD3.

xiii

Figure 2.9. High SMPD3 expression inhibits oligodendroglioma cell growth in a suspension system.

Figure 2.10. High SMPD3 expression inhibits oligodendroglioma cell growth.

Figure 2.11. Low SMPD3 expression promotes oligodendroglioma cell growth.

Figure 2.12. SMPD3 knockdown is growth enhancing and recapitulated by GW4869, a pharmacological inhibitor.

Figure 2.13. SMPD3 knockdown facilitates oligodendroglioma growth in vivo.

Figure 2.14. Confirming SMPD3 knockdown of xenografted oligodendroglioma cells.

Figure 2.15. SMPD3 knockdown facilitates oligodendroglioma growth in human cerebral organoids.

Chapter 3:

Figure 3.1. Schematic representation of the different phases of Schwann cell development.

Figure 3.2. Expression of SC lineage markers in E9.0 NCCs.

Figure 3.3 Expression of SC lineage markers in E9.0 NCCs.

Figure 3.4. Expression of SC lineage markers in E10.5 NCC precursors.

Figure 3.5 Co-labelling of SC lineage markers and NeuN in the E10.5 trunk.

Figure 3.6. Expression of Sox9 and Sox10 in wild type cortices.

Figure 3.7. Expression of SC lineage markers in E10.5 NCC precursors.

Figure 3.8. Expression of SC lineage markers in E12.5 SCPs.

Figure 3.9. Co-labelling of SC lineage markers and NeuN in the E12.5 trunk.

Figure 3.10. Expression of SC lineage markers in E12.5 SCPs.

Figure 3.11. Expression of SC lineage markers in E14.5 iSCs.

Figure 3.12. Expression of SC lineage markers in in E14.5 iSCs. xiv

Figure 3.13. Expression of SC lineage markers in E18.5 late immature/pro-myelinating SCs.

Figure 3.14. Expression of SC lineage markers in E18.5 late immature/pro-myelinating SCs.

Figure 3.15. Expression of SC lineage markers in SCs.

Figure 3.16. Expression of SC lineage markers in P7 mature SCs.

Figure 3.17. Expression of SC lineage markers in P7 myelinating/non-myelinating SCs.

Figure 3.18. Expression of SC lineage markers in adult P65 uninjured sciatic nerve.

Figure 3.19. Expression of SC lineage markers absent in the P65 adult uninjured and injured nerve.

Figure 3.20. Expression of SC lineage markers in P65 sciatic nerve after acute nerve injury.

Figure 3.21 Developmental glial-lineage genes are up-regulated after acute nerve injury.

Figure 3.22. Expression of pro-myelinating genes Yy1 and Nfatc4 is sustained in proliferating de-differentiated SCs.

Figure 3.23 Summary of temporal expression profiles of key transcription factors in the SC lineage.

Chapter 4:

Figure 4.1. Schwann cell transcriptional lineage markers are expressed normally in E12.5 Etv5- /- Schwann cell precursors (SCPs).

Figure 4.2. Schwann cell non-transcription factor lineage markers are expressed normally in E12.5 Etv5-/- Schwann cell precursors (SCPs).

Figure 4.3. Schwann cell lineage markers are expressed normally in E15.5 Etv5-/- immature Schwann cells (iSCs).

Figure 4.4. Schwann cell lineage markers are expressed normally in E18.5 Etv5-/- late immature/pro-myelinating Schwann cells.

Figure 4.5. Increase in Sox10+ Schwann cells post-injury in Etv5-/- sciatic nerve.

Chapter 5:

Figure 5.1. Establishing a qPCR primer panel to test Schwann cell lineage conversion

Figure 5.2. Overexpression of Sox10 in MEFs induces embryonic Schwann cell markers.

Figure 5.3. Validating triple transcription expression vectors, T1 and T2.

Figure 5.4. Schwann cell-specific markers induced by Sox10, T1 and T2 misexpression in MEFs.

Figure 5.5. Schwann cell-specific proteins induced by Sox10, T1 and T2 misexpression in MEFs.

Figure 5.6. Ngfr+O4+ Schwann-like cells exhibit improved SC specific marker gene expression.

Figure 5.7. DRG-Schwann cell co-culture assay.

Chapter 6:

Figure 6.1. Schwann cell-specific markers induced by Sox10, T1 and T2 misexpression in Skin derived precursor cells.

xvi

Appendices

Table A.2.1 Tabulation of results.

Table A.2.2 Mass spectrometry analysis protocol settings.

xvii

List of Abbreviations

ANOVA analysis of variance AP2 Activating enhancer binding protein 2 alpha factor ARRDC1 arrestin-domain-containing protein 1 aSMase Acid Sphingomyelinase AUP Animal utilization protocol B2m beta-2 microglobulin BBB Blood-brain-barrier BC Boundary cap BDNF brain derived neurotrophic factor BFABP fatty acid binding protein 7, brain BrdU bromodeoxyuridine BTPCs Brain tumor propagating cells Ca2+ calcium ion CD1 CD1 mouse strain Cdh19 Cadherin 19 cDNA complementary DNA CIC Capicua Transcriptional Repressor CLIC1 Chloride intracellular channel-1 CM Conditioned media CM-EV CM minus EV CNS Central Nervous System CNTF ciliary neurotrophic factor CO Cerebral organoid Col1a1 Collagen, type I, alpha 1 CRD Cysteine rich domain CSC Cancer stem cell CREB cAMP response element binding protein Cx32 Connexin 32 CyaA Cytosine arabinoside DAPI 4′,6-diamidino-2-phenylindole DIV days in vitro DMEM Dulbecco's Modified Eagle's Medium dn Dominant negative DNA deoxyribonucleic acid dpi days post injury DRG Dorsal root ganglion E embryonic day EDTA ethylenediaminetetraacetic acid EGF epidermal growth factor EGFP enhanced green fluorescence protein Egr1 Early growth response 1 Egr2 Early growth response 2

xviii

EM electron microscopy eNSCs Embryonic Neural Stem Cells Erbb2 Erb-b2 receptor tyrosine kinase 2 Erbb3 Erb-b2 receptor tyrosine kinase 3 Erk extracellular signal-regulated kinase Endosomal sorting complexes required for transport ESCRT machinery EtOH ethanol Etv5 ETS Variant Transcription Factor 5 EV extracellular vesicle FACS fluorescence activated cell sorting FasL Fas Ligand FBS fetal bovine serum FGF fibroblast growth factor fl floxed FM Fresh media g Relative Centrifugal Force in units of times gravity GAP43 Growth associated protein 43 Gapdh glyceraldehyde-3-phosphate dehydrogenase GBM Glioblastoma or Glioblastoma Multiforme GFAP glial fibrillary acidic protein GFP green fluorescence protein GSC Glioma stem cell H2O2 hydrogen peroxide HBSS Hanks' Balanced Salt solution HCl hydrochloric acid HEK293T human embryonic kidney cells 293 hiPSC(s) Human induced pluripotent stem cell(s) HNA Human nuclear antigen HOG Human oligodendroglioma Hprt Hypoxanthine-guanine phosphoribosyl transferase hr(s) hour(s) IDH1 Isocitrate dehydrogenase 1 IDH2 Isocitrate dehydrogenase 2 IDHm IDHMR132H IHC immunohistochemistry iPSC(s) induced pluripotent stem cell(s) iSC(s) Immature Schwann cell(s) ISEV International society for extracellular vesicles Jun Jun proto-oncogene KD knockdown Lam Laminin lEV Large Extracellular vesicles lncRNA Long non-coding RNA

xix

LPS Lipopolysaccharide M molar MAG Myelin associated glycoprotein MBP Myelin basic protein mEV medium Extracellular vesicles mg milligram Mg2+ magnesium ion MgCl2 magnesium chloride min(s) minute(s) miRNA micro RNA ml millilitre mm millimetre mM millimolar MPZ Myelin protein zero mRNA messenger RNA ms millisecond MSC Mesenchymal Stem Cell MSE Myelinating Schwann cell element mV millivolt MVB multivesicular bodies N normal (acid/base concentration) NaCl sodium chloride NaF sodium fluoride NaH2PO4 sodium phosphate monobasic NaOV sodium orthovanadate NCC(s) Neural crest cell(s) ncRNA non-coding RNA Nfatc4 Nuclear factor of activated T-cells, cytoplasmic 4 NFκB nuclear factor kappa B NGF Nerve growth factor Ngfr Nerve growth factor receptor nm nanometre nM nanomolar Nrg1 Neuregulin1 NSC(s) neural stem cell(s) NSG Nod Scid Gamma nSMase2 Neutral sphingomyelinase2 N-terminal/terminus amino-terminal/terminus (amino acid chain) Ntrk Neurotrophic receptor tyrosine kinase O/N overnight OCT optimal cutting temperature ODG Oligodendroglioma Olig1 Oligodendrocyte transcription factor 1 Olig2 Oligodendrocyte transcription factor 2

OPC(s) Oligodendrocyte precursor cell(s) P postnatal day p p-value PAGE polyacrylamide gel electrophoresis Pax3 paired homeobox 3 PBS Phosphate buffered saline PBT Phosphate buffered saline/0.1% Triton X PCR polymerase chain reaction PDL Poly-D-Lysine Pen/Strep penicillin/streptomycin PFA paraformaldehyde PI Propidium iodide PI3K phosphatidylinositol-3-kinase PKC protein kinase C PKCd protein kinase C delta type Pmp22 Peripheral myelin protein 22 PMSF phenylmethylsulfonyl fluoride PNS Peripheral nervous system Pou3f1 POU domain, class 3, transcription factor 1 Pou3f2 POU domain, class 3, transcription factor 2 Pou3f3 POU domain, class 3, transcription factor 3 PVDF PolyVinylidene DiFluoride qPCR quantitative polymerase chain reaction Rcf Relative centrifugal field RNA ribonucleic acid Rpm rotation per minute RT room temperature RTK receptor tyrosine kinase s.e.m. Standard error of the mean S100b/S100 S100 protein, beta polypeptide SC(s) Schwann cell(s) SCP Schwann cell precursor Scr Scrambled SDS sodium dodecyl sulphate sEV small Extracellular vesicles SGC(s) Satellite glial cell(s) Shh Sonic hedgehog SKP(s) Skin derived precursor cell(s) SMPD3 sphingomyelin phosphodiesterase3 sn Spinal nerve SOD Superoxide dismutase Sox10 sex determining region Y-box 10 Sox2 sex determining region Y-box 2 Sox9 sex determining region Y-box 9

xxi srRNA self-replicative RNA STR short tandem repeat SGZ subgranular zone TBS tris buffer saline TCGA The cancer genome atlas TF(s) Transcription factor(s) Tfap2a Transcription factor AP2a Thy1 Thymus cell antigen 1, theta TMRE Tetramethylrhodamine, ethyl ester TNF Tumor necrosis factor TRAIL Tumor necrosis factor-related apoptosis-inducing ligand V volt VEE Venezuelan Equine Encephalitis VPA Valproic acid V-SVZ ventricular-subventricular zone WB Western blot WHO World Health Organization wt/WT wild-type Yy1 Yin Yang 1 β-actin beta actin μg microlitre μl microgram µm micrometre µM micromolar

xxii

Dissemination of work arising from this Thesis

Chapter 1. Balakrishnan A, Roy S, Fleming T, Leong HS, Schuurmans C. 2020. The Emerging Role of Extracellular Vesicles in the Glioma Microenvironment: Biogenesis and Clinical Relevance. Cancers 2020, 12, 1964. doi: 10.3390/cancers12071964.

I wrote majority of the review article, except the last two sections that were written by Sabrina Roy and Hon Leong. All illustrations were generated by Taylor Fleming. I assembled the first draft, which was edited by Carol Schuurmans before submission.

Chapter 2: Balakrishnan A, Adnani L, Chinchalongporn V, Vasan L, Prokopchuk O, A, Chen M, El-Sehemy Olender T, Sheikh T, Islam R, Sujanthan S, Zinyk D, Comanita L, Kan B, Fleming T, Leong HS, Morshead C, Brand M, Wallace V, Chan J, Schuurmans C. SMPD3-mediated extracellular vesicle biogenesis inhibits oligodendroglioma growth. (bioRxiv pre-print server; MS ID#: BIORXIV/2020/202200).

The project was initiated by Lata Adnani, and after her graduation, I took over, repeating her initial experiments for validation, while also greatly expanding the assays performed. All experiments in the current manuscript/thesis chapter were performed by me, with the following exceptions: (1) Cerebral organoid-co-cultures: These experiments were performed and analyzed by Vorapin Chinchalongporn, Dawn Zinyk, Alex Prokopchuk and Lakshmy Vasan in the Schuurmans lab, with help generating cerebral organoids from Rehnuma Islam in Cindi Morshead’s lab. (2) Patient sections: Myra Chen in Jennifer Chan’s lab performed performed immunostaining of human ODG tissues. (3) Mouse xenografts: Ahmed El-Sehemy and Lacrimioara Comanita in Valerie Wallace’s lab performed the xenografting experiments. (4) Proteomics: Thomas Olender in Marjorie Brand’s lab carried out proteomic data analysis. (5) TCGA analyses: Tanveer Sheikh in Jennifer Chan’s lab and Alex Prokopchuk in the Schuurmans’ lab performed this analysis. (6) Pellet assay and GW4869 treatment assay: The cre-based EV transfer assay, and GW4869 high dose treatment assay were data from Lata Adnani’s work. (7) Technical assistance: Sajeevan Sujanthan and Boris Kan were undergraduates who assisted with neurosphere assays, cell quantification, Western blots, and cell line maintenance. (8) Schematics: Taylor Fleming generated all the illustrations. I wrote the manuscript, which was edited by all authors.

xxiii

Chapter 3: *Balakrishnan A, *Stykel M, Touahri Y, Stratton JA, Biernaskie J, Schuurmans C. 2016. Temporal analysis of gene expression in the murine Schwann cell lineage and the acutely injured postnatal nerve. PLoS One. Apr 8;11(4): e0153256. doi: 10.1371/journal. pone.0153256. eCollection 2016. PMID:27058953. * equal contribution

I initiated this project and performed all the embryonic and uninjured nerve experiments, cell counting and data analysis. Morgan Stykel performed the injured nerve experiments under Dr. Biernaskie’s guidance. Yacine Touahri performed embryo tissue collection and contributed with data analysis. The first draft was assembled by me with contributions from Morgan Stykel, Yacine Touahri, Dr. Biernaskie, Dr. Schuurmans. I wrote the manuscript, which was edited by all authors.

Chapter 4. Balakrishnan A, Belfiore L, Vasan L, Touahri Y, Stykel M, Fleming T, Midha R, Biernaskie J, Schuurmans C. Etv5 is not required for Schwann cell development but is required to regulate the Schwann cell response to peripheral nerve injury. (Manuscript prepared for submission)

I initiated this project and performed all the embryonic and uninjured nerve experiments and data analysis. Yacine Touahri performed embryo tissue collection, and Lakshmy Vasan contributed with data analysis. Morgan Stykel performed the injured nerve experiments under Dr. Biernaskie’s and Dr. Midha’s guidance. Taylor Fleming generated the illustrations. I wrote the manuscript, which was edited by all authors.

Chapter 5: Balakrishnan A, Belfiore L, Touahri Y, Noman H, Zinyk D, Biernaskie J, Schuurmans C. Glial transcription factor-based conversion of mouse embryonic fibroblasts to a repair Schwann cell identity. (Manuscript in preparation)

I initiated this project and performed all the experiments and data analysis, with the following exceptions. Lauren Belfiore contributed to cell culture isolation and maintenance. Yacine Touahri contributed to qPCR experiments. Humna Noman contributed to qPCR and in vitro dorsal root ganglia experiments. Dawn Zinyk contributed to plasmid construction. Dr. Biernaskie provided intellectual support and in experimental design. I wrote the thesis chapter.

xxiv

Other contributions during my thesis work

Other publications from PhD program (published/in press)

1. Aslanpour S, Rosin JM, Balakrishnan A, Schuurmans C, Kurrasch DM. 2020. Ascl1 is required to specify a subset of ventromedial hypothalamic neurons. Development. 2020. 147: dev180067 doi: 10.1242/dev.180067. PMID: 32253239.

I performed cell sorting, qPCR, and immunostaining experiments.

2. Adnani L, Dixit R, Chen X, Balakrishnan A, Modi H, Touahri Y, Logan C, Schuurmans C. 2018. Plag1 and Plagl2 have overlapping and distinct functions in telencephalic development. Biology Open. 2018 Nov 26;7(11).1-16. pii: bio038661. doi: 10.1242/bio.038661. PMID: 30361413.

I performed the qPCR experiments.

3. *Han S, *Dennis D, *Balakrishnan A, Dixit R, Britz O, Zinyk D, Guillemot F, Kurrasch D, Schuurmans C. 2018. A non-canonical role for the proneural gene Neurog1 as a negative regulator of neocortical neurogenesis. Development. 2018 Oct 1;145(19). 1-13. pii: dev157719. doi: 10.1242/dev.157719. PMID: 30201687. *equal contribution

I was involved in sample collection, performed immunostaining, RNA in situ hybridization, and qPCR experiments in wild type vs Neurog1-/- samples, and data quantification. I also performed the PLA assay.

4. *Wilkinson G, *Dennis D, *Li S, Dixit R, Adnani L, Balakrishnan A, Kovach C, Gruenig N, Dyck R, Schuurmans C. 2017. Neurog2 and Ascl1 together regulate a postmitotic derepression circuit to govern laminar fate specification in the neocortex. Proc Natl Acad Sci U S A. 2017 Jun 20;114(25): E4934-E4943. doi: 10.1073/pnas. 1701495114. PMID:28584103.

I contributed towards data quantification on knockout mice

xxv

5. Hoghooghi V, Palmer AL, Frederick A, Jiang Y, Merkens JE, Balakrishnan A, Finlay TM, Grubb A, Levy E, Gordon P, Jirik FR, Nguyen MD, Schuurmans C, Visser F, Ousman S. 2020. Cystatin C plays a sex-dependent detrimental role in MOG35-55-induced experimental autoimmune encephalomyelitis. (Cell Reports; In Press)

I performed RNA in situ hybridization experiments

xxvi

Other publications from PhD program (under revision)

1. Touahri Y, Balakrishnan A, Adnani L, Dixit R, Tachibana N, David LA, Chute M, Zinyk D, Chan J, Wallace VA, Biernaskie J, Schuurmans C. 2020. Direct reprogramming of somatic cells to an induced retinal progenitor cell-like fate using a triple transcription factor cocktail. (Stem Cell; Under revision)

I contributed towards cell line generation, maintenance, and qPCR experiments.

xxvii

Introduction

Contents of this chapter have been submitted for publication:

Modified from “Balakrishnan A, Roy S, Fleming T, Leong HS, Schuurmans C. 2020. The Emerging Role of Extracellular Vesicles in the Glioma Microenvironment: Biogenesis and Clinical Relevance. Cancers 2020, 12, 1964. doi: 10.3390/cancers12071964.

Link: https://www.mdpi.com/2072-6694/12/7/1964/htm

1.1 Glial cells: An overview

The vertebrate nervous system is divided into two parts - the peripheral (PNS) and central (CNS) nervous systems. These two regions of the nervous system differ in many ways, including in their glial content. The focus of my PhD has been on the myelinating glial cells, which include oligodendrocytes in the CNS and Schwann cells in the PNS.

1.2 Glioma: An Introduction

1.2.1 General introduction on glioma subtypes

Gliomas are primary brain tumors, composed of cells with gene expression profiles and morphological characteristics similar to glial cells (e.g. astrocytes, oligodendrocytes)1,2. Glioma subtypes include oligodendroglioma (ODG), astrocytoma, glioblastoma, ependymomas, as well as schwannoma and neurofibroma1,2. Depending on the speed and extent of tumor growth, histological features, and tumor invasiveness gliomas have been graded between Grade I-IV1,3. Low-grade gliomas (Grade II-III) include astrocytoma and ODG3,4, with the key identifying features including (but not limited to) mutation in the isocitrate dehydrogenase (IDH) 1 or IDH2 genes (in nearly 70% of low-grade gliomas5), deletion of chromosomal arms 1p/19q, mutation in Capicua (CIC), a repressor of Ets transcription factors6,7. In addition, other low grade gliomas have mutations in telomerase reverse transcriptase promoter8, or in tumor protein 53 (in astrocytomas)9-11. With respect to survival, patients with low-grade glioma exhibit better prognosis compared to higher grade gliomas (like glioblastoma)3. My study focused on ODG and the interaction of ODG cells with other ODG cells as well as non-neoplastic cells in the glioma microenvironment.

1.2.2 Oligodendroglioma

1.2.2.1 Epidemiology

According to the 2016 World Health Organization classification of tumors of the CNS, ODG is classically defined as “a diffusely infiltrating, slow-growing glioma with isocitrate dehydrogenase (IDH) 1 or IDH2 mutation and codeletion of chromosomal arms 1p and 19q12. Studies have reported that amongst brain and CNS tumors, ODGs are prevalent in approximately 1.4% patients (sample size =392,982; all age groups) with 53 as the median age of the patients. In addition, ODGs have been reported in 1.5% of patients between 15-19 years of age (sample size = 7259) based on studies spanning 2011-2015 in Northern America13. There

2 were no independent statistical studies obtainable for the incidence rate of ODGs in Canada when this report was being generated.

1.2.2.2 Classification, molecular, and histological features

WHO classification of tumors of the brain and the CNS has identified ODGs under two main groups depending on the degree of malignancy. IDH-mutant, 1p/19q co-deleted ODG fall under grade II gliomas, while anaplastic ODG (IDH-mutant, 1p/19q co-deleted) are categorized under grade III gliomas13. Astrocytomas on the other hand are diagnosed by 1p/19q codeletion but do not always present with the characteristic IDH mutation4. While IDH1 and IDH2 enzymes bring about conversion of isocitrate to alpha ketoglutarate, IDH1 (R132H) and IDH2 (R140, R172) mutants bring about conversion of alpha ketoglutarate into D-2-hydroxyglutarate (2-HG)14,15. The increase in levels of 2-HG result in inhibition of DNA- and histone- demethylases bringing about global methylation of CpG island16,17 and cell differentiation arrest18. The inhibition of DNA- and histone-demethylases and thereby the increased methylation status of cells has been proposed to promote tumorigenicity in cells19,20. Interestingly, presence of IDH 1/2 mutation has been associated with better patient prognosis compared to IDH wild type gliomas21.

Histologically, ODGs generally present as a diffuse infiltrative tumor composed of small, rounded cells with uniformly rounded nuclei in both low- and anaplastic-grade ODGs22. Additionally, the tumors present with a branching vascular network which gives the tumor a wired appearance and have been found to be sparsely infiltrated with GFAP+ reactive astrocytes22.

1.2.2.3 Treatment and prognosis

ODG patients present with partial to tonic-clonic seizures depending upon the location and stage of the tumor23. Grade II ODGs (untreated cases) grow at an annual rate of 3.8–4.4 mm/year24 with patients presenting with seizures and focal deficits25. The median survival estimate of patients with low grade IDH mutant ODG ranges from 12-14 years26,27. In contrast, low grade IDH mutant astrocytoma patients exhibit median survival estimate of 9-13 years, which further worsens in high grade gliomas e.g. GBM (~4 years)28,29. The favorable survival times in ODGs are attributed in part to the ever-improving treatment options available in the form of surgery, chemotherapy, and radiation30. However, radiotherapeutic intervention has been reported to lead to cognitive deficits in treated patients, and hence is not preferred as a

stand-alone treatment approach30. Ideal treatment strategies for patients can be designed by basing the plan on the molecular marker information available on the tumor. Although surgery is the first mode of approach for resecting ODG tumors, better outcomes have been noted in patients treated with combination therapy- i.e. radiation and chemotherapy31.

1.3 Extracellular vesicles: An Introduction

The first hint that extracellular vesicles (EVs) might exist and have biological activity was the 1967 observation that “platelet dust” participates in clot formation32. This was followed in the 1980s by the use of electron microscopy to visualize EVs emanating from late endosomes in reticulocytes33,34. These vesicles were found to carry transferrin receptors, and were eventually termed ‘exosomes’35. Seminal work in the 1990s then uncovered a role for EVs derived from B cell lymphocytes36 and dendritic cells37 in antigen presentation to T cells. These reports gave birth to an entire new field of research that was predicated on the notion that EVs have biological actions and are not solely ‘waste carriers’, as initially thought36,37. Since then, EVs have been shown to mediate intercellular communication between neighboring and distantly located cells in a vast array of biological contexts (reviewed extensively in References38-43). The specific roles that EVs play in intercellular communication is due to their selective and cell-type-specific loading of lipids and nucleic acids, including DNA, messenger RNA (mRNA), microRNA (miRNA) and non-coding RNA (ncRNA)44,45. EVs also carry protein cargo, which may be packaged into EVs non-specifically, but nevertheless serve as a readout of cellular state and can also influence the physiology of recipient cells44,45. Notably, encapsulating vulnerable cargo in vesicular structures renders them inaccessible from degradation by ribonucleases, deoxy-ribonucleases and proteases in the extracellular space, as these enzymes cannot traverse the EV lipid bilayer; as such, EV biogenesis has been an important evolutionary advancement that has allowed for complex intracellular signaling to take place in multicellular organisms46-48.

All cells have unique physiologies and molecular identities, and so too do their derivative EVs. EV cargo differs depending on cell type and a cells’ physiological status and may thus be used to trace the cell of origin45. Moreover, as “highly-stable reservoirs of disease biomarkers” (Exocarta; http://www.exocarta.org/), EVs may serve as valuable indicators of health and disease43. When EVs are secreted and enter the peripheral circulation, they may contain biomarkers that carry mutational, cell signaling and microenvironmental information that could be used as readouts of a therapeutic response. Given that a current limitation in neuro-oncology

4 is that tumor progression can only be monitored radiologically, new non-invasive measurements of disease progression could revolutionize patient care. In the context of glioma, the ability to detect the molecular state of a brain tumor in patient biofluids would significantly facilitate patient diagnosis, disease stratification and treatment monitoring in a non-invasive fashion. Early support for this notion came from the demonstration that cells derived from human glioblastoma tissue secrete EVs in vitro48, and the content of EVs in serum and cerebrospinal fluid (CSF) (e.g., presence of amplified epidermal growth factor receptor (EGFR) in EVs in CSF of glioblastoma patients49) differs between patients with and without glioma48-51.

In this section, we summarize recent literature describing the roles of EVs in mediating neural cell communication, especially in a glioma context, and describe the potential clinical utility of EVs for glioma subtyping and as biomarkers for glioma detection and therapeutic monitoring. The use of ‘humanized’ rodent models has galvanized this research, with patient- derived tumor xenografting now a mainstay approach to model glioma, including for the study of EVs in disease progression and as biomarkers.

1.3.1 A brief primer on EV classification, isolation, and biogenesis

1.3.1.1 EV classification and isolation

In healthy and diseased states, cells release EVs of different sizes and intracellular origins. The result is a heterogeneous mix of membranous vesicles, collectively termed “EVs”. While the definition of EV has evolved over time, the International Society for Extracellular Vesicles (ISEV) currently defines EVs as naturally released, non-replicative particles that are delimited by a lipid bilayer52. However, nomenclature in the field has been historically ‘muddled’, with researchers initially classifying EVs based on physical characteristics, composition, and even cell of origin52. Terminologies such as exosomes, microparticles, and microvesicles have also been used interchangeably, further complicating the field. The ISEV thus set out to standardize the naming system, now classifying EVs (the generic term) based on physical characteristics, with particles <200 nm in diameter called small EVs (sEVs) while larger particles (>200 nm) are termed medium (mEVs; between 200-400 nm) and large EVs (lEVs; >400 nm)52 (Fig. 1.1). Several markers have been identified that label these vesicular structures, including the tetraspanins, with CD81 enriched in sEVs, CD9 in sEVs and mEVs, and CD63 in EVs of all sizes40. 5

Figure 1.1. Schematic of EV-mediated signaling between neighboring cells.

Generation of EVs from a host cell, depicting small (sEV; 40–200 nm), and medium to large (m/lEV; 0.2–1 µm) EVs and apoptotic bodies (1–5 µm). sEVs find their origin in the early endosomal compartment which give rise to multivesicular bodies (MVBs). MVBs fuse with the plasma membrane to secrete sEVs. Biogenesis of sEVs can occur via the endosomal sorting complex related to transport (ESCRT)-dependent pathway or ESCRT-independent pathway. Release of sEVs involves proteins such as RabGTPases (e.g., Rab11, Rab35, Rab27a/27b). Cargo in MVBs fated for degradation are processed via lysosomes. m/lEVs are released by budding of the plasma membrane regulated by proteins including Arf6 and RhoA GTPase, while dying cells undergo blebbing and form cell protrusions to release apoptotic bodies. Exomeres (~35 nm) are also released by cells via unknown mechanisms. A magnified sEV is presented on the right, demonstrating commonly found sEV cargo (DNA, messenger RNA (mRNA), microRNA (miRNA), non-coding RNA (ncRNA), long non-coding RNA (lncRNA), proteins, enzymes, heat shock proteins). Recipient cells internalize sEVs via endocytosis.

EVs can be isolated from different sources (e.g. conditioned media from cultured cells, or biofluids, such as urine, serum, blood, etc.) using a growing list of protocols. EV isolation strategies include sequential ultracentrifugation, high resolution density gradient ultracentrifugation53, direct immunoaffinity capture using canonical EV specific antibodies (CD63,CD81,CD9)44, immunoprecipitation using magnetic beads54, ultrafiltration55, and the use of numerous commercial EV isolation kits based on immunophenotypes53,56. As EV isolation methods have been the topic of other excellent recent reviews57,58 they will not be further elaborated on herein. However, one important consideration is that several of the less specific EV isolation methods (e.g. sequential ultracentrifugation) also collect non-vesicular particles (exomeres, ~35-50 nm) that are not membrane enclosed, but which may contribute to the molecular and biological phenotypes associated with these preparations59,60. Heat shock proteins (HSP) like HSP90, HSP13, Histone H2A,H3,H4, as well as Argonaute proteins (Ago1-3) are enriched in non-vesicular particles compared to exosomes44, which may help to identify the presence of exomeres in EV preparations.

1.3.1.2 Small and medium/large EVs

Small EVs (sEVs; diameter: <200nm), previously known as exosomes, are derived from multivesicular bodies (MVBs)61,62. Within MVBs, endosomal membranes invaginate to form intraluminal vesicles (ILVs) that fuse with the cellular plasma membrane to secrete the enclosed vesicles61,62 (Fig. 1.1). SEVs have an approximate diameter of 40-200 nm, with densities ranging from 1.08-1.13 g/ml depending on the cell-of-origin44,63. The functions attributed to sEVs in cancer are diverse, ranging from aiding tumor progression and metastases, to promoting tumor dormancy 61,64. The potential use of sEVs for cancer immunotherapy, nano- vaccines and as diagnostic tools is also currently being explored65,66.

Medium or large EVs (mEVs or lEVs) (diameter: 200-1000 nm) were previously called microparticles, microvesicles, ectosomes and oncosomes67. They are generated by outward budding and fission of the plasma membrane68. Similar to sEVs, m/lEVs can also transfer cytosolic components to neighboring cells48,69-71, with their contents protected from extracellular degradation by an enveloping lipid bilayer (Fig. 1.1)68. M/lEVs form at specific sites in the membrane where lipids and proteins aggregate in microdomains72. Cytoplasmic Ca2+ levels are also elevated in ‘budding’ cells, promoting clustering and activation of phospholipid scramblases, which move lipids across the plasma membrane, and floppases, which transport lipids such as phosphatidylserine from the inner to the outer membrane 7 leaflet72,73. Notably, presentation of lipids such as phosphatidylserine on the outer membrane leaflet serves as an ‘eat-me’ signal that alerts microglia in the brain (or macrophage elsewhere in the body) that a cell is under stress and should be eliminated45,72. Annexin A1 has recently been reported to be a distinguishing marker expressed in m/lEVs, and not in sEVs44.

Large EVs also include larger apoptotic bodies (diameter: 1-5 µm) that are released from cells undergoing fragmentation due to programmed cell death74. With the onset of apoptosis, cells undergo a cascade of structural changes (apoptotic cell disassembly) involving blebbing and protrusion of the cell membrane, followed by release of apoptotic bodies (Fig. 1.1)74. These EVs tend to fall in the larger size range, however smaller apoptotic bodies (called apoptotic vesicles) in the size range of sEVs have also been identified75. Similar to other m/lEVs, apoptotic bodies present phosphatidylserine on their surfaces, so that they are quickly cleared75.

The list of vesicle types continues to grow as methods of analysis become more sophisticated and classification criteria more granular. Other categories of EVs include autophagic EVs, released via autophagy dependent pathways76 and arrestin-domain-containing protein 1 (ARRDC1)-mediated m/lEVs77,78. In the below sections, we will use the updated sEV and m/lEV terminology that we apply to studies that defined vesicles as exosomes or microparticles/microvesicles, respectively.

1.3.1.3 ESCRT-dependent and ESCRT-independent pathways for EV biogenesis

At least 10-15 proteins are known to play a functional role in EV formation40, with the major players summarized herein. The most common sEV biogenesis route involves an endosomal sorting complex related to transport (ESCRT)-dependent pathway (extensively described in other reviews;79,80). Briefly, ESCRT-dependent sEV biogenesis and release involves the concerted actions of ESCRT complex proteins- ESCRT-0, -I, II, III62,81. ESCRT-0 proteins have ubiquitin binding sites that recognize and sequester ubiquitinylated cargo in the late endosomal membrane. ESCRT-0 proteins then sequentially recruit ESCRT-I, -II and -III complex proteins using accessory proteins like ALIX79,81. Deformation of the endosomal membrane to form vesicular buds containing cytosolic cargo involves ESCRT-I and II complexes81, while vesicle scission and release of ILVs into MVBs involves ESCRT-III complex proteins62,79,82.

ESCRT-independent mechanisms also sort cargo into ILVs of MVBs62,79,81,83-87. One such mechanism involves neutral sphingomyelinase2 (nSMase2), an enzyme that acts upon sphingolipids in lipid-raft microdomains within the endosomal membrane60,83. nSMase2 hydrolyses sphingomyelin into phosphocholine and ceramide88,89. Ceramide is enriched in sEVs and is involved in targeting cargo into ILVs and packaging of sEVs83. Sphingosine-1- phosphate, a ceramide metabolism by-product, also exhibits a role in sorting cytosolic cargo into ILVs 90. Budding of ILVs is promoted by ceramide courtesy of their cone-shaped structure that promotes membrane bending91, which in turn induces smaller microdomains within endosomal membranes to merge into larger domains83,92. A reduction in nSMase2 activity using a pharmacological inhibitor, GW4869, blocks the packaging of protein83 and miRNAs93 into sEVs. After sEV synthesis, Rab GTPases (Rab27a, Rab27b, Rab11, Rab35) are involved in sEV secretion and recycling of proteins between the endosomal compartment and plasma membrane94-96.

Biogenic pathways for other vesicle types are also beginning to be elucidated, such as acidic sphingomyelinase (aSMase), which is activated and recruited to the plasma membrane to promote m/lEV shedding91,97. Sphingomyelin enriched in the outer leaflet of the membrane is hydrolyzed by aSMase, destabilizing the membrane and facilitating m/lEV shedding97,98. Other regulators involved in m/lEV formation and release include Arf6, involved in endosomal recycling99 and the small GTPase, RhoA100.

1.3.2 EV functions in the ‘healthy’ nervous system

Gliomas are composed of a mixture of malignant glial cells ('tumor cells') as well as as non- malignant neurons and glial cells ('neural cells'), and a variety of other inflammatory and vascular cells ('stromal cells'). Tumor cells carry driver mutations causally implicated in oncogenesis, but other cells in the microenvironment also participate in tumor cell proliferation and growth. A solid understanding of EV production by ‘normal’ brain cells is required as these vesicles can impact tumor growth and progression. As published literature on brain EVs is vast, our survey is not comprehensive, but rather gives a flavor for the types of roles that EVs play in a ‘healthy’ brain (Fig. 1.2).

1.3.2.1 Neural stem and progenitor cell derived EVs

In the adult brain, neural stem cells (NSCs) are restricted to a few neurogenic zones that repopulate specific brain regions throughout life, including the ventricular-subventricular zone (V-SVZ) in the forebrain that repopulates the murine olfactory bulb and human striatum, the subgranular zone (SGZ) that repopulates the mouse/human dentate gyrus, and the mediobasal hypothalamus that gives rise to new hypothalamic neurons101-103. Outside of these niches, the adult NSC response is limited, and while some neuroblasts are produced in response to injury in other brain regions104,105, few new neurons survive and integrate to register any meaningful recovery106. Nevertheless, NSC-derived EVs have been studied extensively with regards to their neuroprotective properties in injury models, such as stroke107,108, and for their ability to modulate microglia activity109. Reprogrammed NSCs (derived from mouse fibroblasts and astrocytes in vitro) were also found to secrete EVs that promote their own proliferation by activation of MEK/ERK signaling110. SEVs secreted from human induced pluripotent stem cells (hiPSCs)-derived neurons also regulate neural circuit assembly111. Finally, a recent study found that sEVs secreted by hypothalamic NSCs into the CSF slow down the aging process in rodent models in vivo103. Taken together, these studies and others support the idea that NSC- derived EVs may have therapeutic value for the treatment of brain injury or neurodegenerative disease, as reviewed elsewhere112, and possibly for the study of longevity. With respect to glioma, brain tumor propagating cells (BTPCs) are thought to arise from adult NSCs, and the adult NSC niche may support the growth and division of BTPCs113. Thus, understanding how NSCs signal to tumor cells is essential to devise strategies to block tumor cell proliferation.

1.3.2.2 Neuron-derived EVs

Neuronal sEV release was initially detected in cultured cortical cells that received a membrane depolarization stimulus, with glutamate receptor subunits part of the sEV cargo114. Other studies have since confirmed that EVs are released from neurons in response to glutamatergic activity115. SEV release during neuronal firing has been proposed to have a ‘waste disposal’ role, removing miRNAs to reduce their silencing effect in ‘active’ neurons116. In addition, sEVs released from cortical neurons in cell culture are primarily taken up by neuronal and not glial cells and aid inter-neuronal communication117. Cultured neurons form close knit networks that uptake neuronal EVs and simultaneously re-secrete EVs (via endogenous secretory endosomes) to aid EV spreading118, and in this manner promote neural circuit development 111.

Recently, neuronal sEV release has been confirmed in vivo using a transgenic CD63-GFP reporter line, which suggested that activity dependent sEV release occurs in post-synaptic soma and dendrites119. Contrasting to in vitro studies, this transgenic animal revealed that neuronal EVs, carrying miR-124 in its cargo, are taken up by astroglia, resulting in reduced Glt1 expression levels119. Neuronal EVs have also been described as carriers of miRNAs that act non-cell autonomously in other studies116,120,121. For example, sEV release of miR-132 promotes vascular integrity in the zebrafish brain by targeting eef2k in endothelial cells121. Similarly, sEV release of signaling proteins in neuronal EVs have been conjectured to play a role in intercellular communication, as well as for disposal of these proteins 114,122-124. Neuronal secretion of growth factors such as VEGF and FGF2 in sEVs125, could also impact glioma growth.

1.3.2.3 Astrocyte-derived EVs

Astrocytes are multi-functional macroglial cells that have a wide host of different functions - they provide structural and trophic support for neurons, contribute to the blood-brain-barrier (BBB), and maintain myelin integrity126. Astrocytes were initially shown to shed m/lEVs via aSMase in response to P2X7 receptor-ATP ligand binding97. EVs derived from two-day-old rat cortical astrocytic cultures carry various growth factors (e.g. VEGF, FGF2)127 and promote neurite outgrowth and neuronal survival128. The neuroprotective effects of astrocytic EVs have been studied extensively, identifying critical cargo that is transferred to neuronal cells, including neuroglobin129, synapsin, which is released during oxidative stress and ischaemic conditions128, and ApolipoproteinD, also released in response to oxidative stress130. Astrocytic EVs also promote the differentiation of rat oligodendrocyte precursor cells (OPCs) in vitro, and interestingly, as astrocytes age, this capacity declines131. A key study described the role of astrocytic EVs in promoting proliferation and survival of breast cancer and melanoma cells disseminating to the brain using an orthotopic mouse xenograft model system132. Tumor cells metastasizing to the brain were found to promote astrogliosis, resulting in the release of miR- 19a loaded sEVs from reactive astrocytes. Increased miR-19a levels in the brain microenvironment reduce PTEN expression (miR-19a target), an important tumor suppressor, culminating in tumor cell growth and inhibition of tumor cell apoptosis132. This study demonstrated the important role that astrocytic EVs play in tumorigenesis and highlighted the importance of considering EV secretion by ‘normal’ cells in the brain tumor microenvironment when studying disease mechanism.

1.3.2.4 Oligodendrocyte-derived EVs

Oligodendrocytes are myelinating glial cells in the central nervous system133 and they release sEVs in a Ca2+ dependent manner, upon receiving neuronal stimuli96,134-136. Exocytosis of EVs by an OPC cell line, Oli-neu, is brought about by Rab35 GTPase and the GTPase activating protein TBC1D10A–C96. SEVs released from oligodendrocytes are characteristically enriched in myelin proteins (e.g. phospholipid protein (PLP), 2’, 3’-cyclic nucleotide 3’- phosphodiesterase (CNP))83,96,134,136,137. Thus, EV secretion by oligodendroglial cells permits expulsion of excess myelin proteins, and importantly plays a role in promoting neuron- oligodendrocyte communication136. Indeed, neuronal uptake and functional retrieval of oligodendroglial sEVs occurs at axonal and somadendritic sites136 and influences neuronal gene expression (Plp, Ier3, Vgf, Bdnf), provides metabolic support and enhances neuronal activity138. Oligodendrocyte-derived EVs also exhibit a protective function in neurons against oxidation induced stress or nutrient deprivation, potentially mediated by EV delivery of catalase and superoxide dismutase (SOD) 1 enzymes136,138. Surprisingly, oligodendrocyte sEVs inhibit oligodendrocyte differentiation in culture137. EVs released by oligodendroglial precursor cell line, Oli-neu, are also selectively taken up by microglia via micropinocytosis, resulting in EV degradation without microglial activation139. OPCs cultured in close contact with astrocytes result in increased OPC-sEV release coupled with an increase in OPC proliferation regulated via integrin b4-mediated cell adhesion140. Thus, oligodendrocyte derived EVs perform a gamut of functions in the neural niche and may have important actions on glioma cells as well.

1.3.2.5 Microglia-derived EVs

Microglia are specialized macrophages that mount immune responses in the brain141. Release of sEVs and m/lEVs from microglial cells has been extensively studied97,142-145. Several inducers of microglial EV release have been found, including Wnt3a143, serotonin145, and lipopolysaccharide (LPS), which also induce an increase in inflammatory cytokines (TNF and interleukin-6) in the EV cargo146. Microglial EVs have actions on neuronal cells, with some examples including m/lEVs inducing neuronal activity144 and sEVs clearing degenerating neurites to facilitate synaptic pruning147. Given the importance of microglia and infiltrating peripheral macrophages in facilitating glioma cell proliferation and migration148, it seemed likely that the secretion of EVs by these immune cells has biological consequences. Indeed, EVs isolated from IFN-/LPS-stimulated microglia can reduce tumor size in a glioma mouse

12 model149. The potential of microglia-derived EVs as a nano-therapy for glioma is also now being further investigated150.

1.3.2.6 Schwann cell-derived EVs

Schwann cells (SCs) are the principal glial cells in the peripheral nervous system (PNS), responsible for myelinating peripheral axons and promoting axonal regeneration post injury151,152. SCs release EVs carrying p75NTR/Ngfr, an inhibitor of myelination, and these EVs are taken up by ‘injured’ peripheral nerve axons to allow axonal regeneration to occur, either in cultured dorsal root ganglia (DRG) or in vivo following sciatic nerve injury153-155. Notably, exposure of injured nerves to SC exosomes reduces RhoA GTPase levels in the axonal growth cone, promoting growth cone formation, a necessary step in axonal repair. Strikingly, if exosomes are derived from ‘repair’ SCs isolated from an injured nerve, these exosomes instead promoted neurite outgrowth of sensory neurons in DRG explants153. Exosomes from repair SCs have increased miR-21 loading, compared to SCs from an uninjured nerve, a miR- 21 that targets PTEN to elevate phosphatidylinositol-3-kinase signaling and aid axonal repair153. In summary, EVs represent an untapped avenue for enhancing axonal regeneration post injury as well as central nervous system injury156. SC EVs may also influence peripheral nerve tumors, as discussed further below.

1.3.3 Roles of EVs in glioma

1.3.3.1 Glioma subtypes

Gliomas are primary brain tumors comprised of tumor cells with gene expression profiles and morphological characteristics similar to glial cells (e.g. astrocytes, oligodendrocytes), as recently reviewed157-159. Glioma subtypes include oligodendroglioma, astrocytoma, glioblastoma, ependymoma, schwannoma and neurofibroma12,157. Depending on the speed and extent of tumor growth, histological features, and tumor invasiveness, gliomas have been graded by the World Health Organization (WHO) between Grade I-IV1,3. Low grade gliomas (Grade II-III) include astrocytomas and oligodendrogliomas3,4 while the most common type of higher grade glioma is an astrocytoma known as glioblastoma (GBM)12.

As the most aggressive and common glioma, GBM deserves special mention. To identify clinically relevant subtypes, The Cancer Genome Atlas (TCGA) performed a multi-platform analysis to describe a robust gene expression-based molecular classification of GBM160. This 13 landmark study identified four distinct molecular subtypes for GBM (classical, mesenchymal, pro-neural, and neural) and demonstrated that subtypes correlate with clinical phenotypes and treatment responses. More recently, classification of three distinct forms of GBM of the classical, pro-neural, and mesenchymal subtypes has been proposed, discarding the neural subtype as a signature that was likely associated with contaminating mRNA from non-tumor cells78,161. Classical GBMs frequently have copy number alterations and/or mutations in EGFR that lead to its overexpression and activation160. Pro-neural GBMs harbor high-level amplification and/or rearrangements of Platelet-derived growth factor receptor alpha (PDGFRA) that render it constitutively active160. However, not all pro-neural GBMs harbor PDGFRA alterations; some feature IDH1/2 mutations160. However, IDH mutations are only detected in secondary GBMs (arising from grade II and III gliomas) which constitute ~5% of the GBM cohort5,162,163. Finally, mesenchymal GBMs are characterized by NF1 loss (summarized in164).

While the above profiles suggest uniformity amongst GBM subtypes and the cells contained therein, tumor composition is complex. There is additional heterogeneity as proliferating tumor stem and progenitor cells undergo lineage progression and differentiation, and the molecular identity of tumor cells can evolve over time, with pro-neural signatures often resolving into more mesenchymal, aggressive tumor phenotypes165. Accordingly, single cell analyses revealed that cells with different subtype-specific gene expression signatures are found within individual tumors in different proportions166. Understanding how glioma cells interact with cells in the microenvironment is an essential step in understanding disease progression. Of note, analysis of the vesiculome for each GBM subtype has revealed differences in key EV pathway components165. This section delves into the key studies focused on the role of EVs derived from GBMs, oligodendrogliomas, and schwannomas.

1.3.3.2 EVs in Glioblastomas

GBMs are Grade IV astrocytomas and represent the most lethal brain tumor subtype12. NSCs residing in the SVZ carrying driver mutations (TERT promoter mutation, EGFR, PTEN and TP53 mutation) have been reported as the cell of origin for GBM167. However, other studies have found that any type of neural cell, including hippocampal NSCs 168, adult neurons and astrocytes168,169, OPCs170,171 and OPC intermediates (expressing low Pdgfra and high Olig1/2 levels)172 can also give rise to GBM-like tumors. Interestingly, human GBM cell-derived EVs can drive transformation of human NSCs towards a tumorigenic state in vitro, highlighting the 14 potential importance of GBM EVs in tumorigenesis173. Supporting the importance of cell-cell interactions in glioma, the requirement for two oncogenes to drive tumor formation (i.e. RasV12, scribbled) can be achieved even when these genes are not expressed in the same cells, but rather in neighboring clones174. Since then, several studies have revealed the role that EVs play in mediating cell-cell interactions in glioma, as elegantly reviewed175,176.

To give an overall appreciation for the importance of EV communication in glioma, we highlight a handful of key studies. The bioactive nature of GBM EVs is mediated by the inclusion of tumorigenic proteins (EGFRvIII), DNA (mitochondrial DNA) and RNA (e.g. Annexin A2 mRNA, miRs including miR-10b, miR-21, miR-221) in EV cargo177-184. For instance, a long non-coding RNA (lncRNA) antisense transcript of hypoxia-inducible factor-1α (AHIF) is found in human GBM cell line derived sEV cargo, which increases viability and invasive properties of GBM cells in vitro185. Notably, while RNA transcripts are carried uniformly by both sEVs and lEVs, protein cargo amounts are significantly higher in GBM cell derived lEVs compared to sEVs, including EGFRvIII176,179. Apart from EGFRvIII, EVs released by GBM cells (including GBM cell lines and GBM patient derived glioma/cancer stem cell line (GSC/CSC)) were demonstrated to carry the chloride intracellular channel-1 (CLIC1) protein186. CLIC1 plays a role in cell cycle regulation187 and has previously been implicated in GBM growth, such that high CLIC1 levels correlate to poor prognosis in GBM patients188. It was later demonstrated that treatment of GBM cells with EVs carrying CLIC1 (1µg/ml) resulted in increased GBM cell proliferation in vitro and in a mouse GBM xenograft model system in vivo186.

High expression of miR-21 is a common feature in GBM patient tissues and established GBM cell lines in vitro, and suppression of miR-21 in vitro results in decreased proliferation and increased apoptosis189,190. Additionally, GBM patients with high miR-21 levels exhibited poor prognosis190. Interestingly, miR-21 carried in sEVs in the CSF of GBM patients has also been reported, where the level of miR-21 in sEVs is directly related to the glioma status of the patient184. Thus, recurring GBM patients demonstrate low levels of sEV miR-21 post-surgical resection, compared to higher levels of miR-21 observed prior to surgical resection184. Hence, sEVs carrying miR-21 in the CSF are an excellent biomarker for checking the glioma status of GBM patients. Also, antisense miRNA oligonucleotides against miR-21 loaded onto sEVs have been recently tested as a delivery system in vivo using a mouse xenograft model for assessing their therapeutic potential191. On similar lines, miR-21 inhibition in GBM cells in

15 vitro was introduced using sEVs engineered to carry an miR-21 sponge construct (i.e. three miR-21 complementary sequences joined by linker sequences)192. Using this approach suppression of miR-21 target genes, PDCD4 and RECK, in GBM cells was reverted along with an increase in apoptosis and decrease in cell proliferation, similar to prior studies189,192. Additionally, the introduction of sEV loaded miR-21 sponge constructs in a rat xenograft model of GBM led to a significant reduction in tumor volume compared to the control group192. Thus, miR-21 inhibition via sEVs is an active area of research for developing new therapeutic strategies.

Tumor repressive functions have also recently been attributed to GBM derived EVs. miR-302- 367 cluster was found to repress stemness in GBM patient derived GSC lines193. Fareh et al. engineered GBM patient derived GSC lines to express miR-302-367 cluster, which then released sEVs carrying miR-302-367 as cargo. Transfer of miR-302-367 to recipient GSC lines via sEVs repressed the stem cell-like nature of GSCs, as demonstrated via an inhibition of cell stemness and proliferation marker expression (e.g. Shh, SOX2, Cyclin D, Cyclin A)193. Thus, miR-302-367 can block GBM growth in a paracrine fashion and miR-302-367 delivery via sEVs is currently being assessed as a potent therapeutic approach for GBM patients. Interestingly, human GBM cell-derived sEVs were reported to carry O-methylguanine-DNA methyltransferase (or MGMT) mRNA, which is an indicator of GBM drug resistance status194. Additionally, temozolomide (TMZ) resistant human GBM cells release sEVs carrying the lncRNA SBF2-AS1, which represses miR-151a-3p and XRC44 expression in vitro60. Transfer of lncRNA SBF2-AS1 via sEVs to neighboring GBM cells also endowed TMZ resistance in the recipient GBM cells60. Thus, lncRNA SBF2-AS1 in sEVs can be an excellent readout/biomarker of TMZ resistance in GBM patients. Thus, personalization of chemotherapeutic treatment may be possible using EV cargo as more precise readouts of the current glioma status194.

Numerous studies have reported the non-cell autonomous effects of GBM tumor-derived EVs on neural cells present in the tumor microenvironment, including astrocytes, endothelial cells, and pericytes (Fig. 1.2)175,176. Some key studies are worth mentioning due to their emerging role as biomarkers for clinical staging and treatment response. Al-Nedawi et al. found that oncogenic EGFRvIII secreted by GBM EVs is taken up by endothelial cells, which are reprogrammed to express VEGF, and activate VEGF receptor 2 (VEGFR2)195. GBM sEVs are also enriched in lncRNA-ATB, which is transferred to astrocytes, where it suppresses miR204-

3p to increase GFAP expression, leading to reactive astrocyte activation and enhanced glioma invasiveness196. Additionally, GBM EVs modulate gene expression in astrocytes, reducing expression of the tumor suppressor gene TP53, and elevating the expression of oncogenic proteins such as MYC197. Notably, while most of these studies have used in vitro modeling, in vivo transfer of material between glioma cells and ‘normal’ brain tissue has recently been demonstrated using a Cre/LoxP reporter system198. EV-mediated GBM interactions with cells in the microenvironment are reciprocal in nature. For example, endothelial cell-derived EVs isolated from a GBM mass promote glioma cell migration199. Similarly, fibroblasts associated with GBM secrete EVs that are taken up by tumor cells to promote glycolysis200.

Figure 1.2. Schematic of cell-cell interactions in the brain with and without tumor cells.

Normal brain (left panel) and brain tumor microenvironment (right panel). In a healthy state, EVs released by normal neural cells (neural stem cells (NSCs), neurons, oligodendrocytes, astrocytes), inflammatory cells (microglia) and endothelial cells promote cell–cell interactions, which contribute toward the formation of an intricate neural network (left side; magnified panel). In a tumorigenic state, the tumor microenvironment is a complex ecosystem of tumor cells intricately knit together with ‘normal’ brain cells. Tumor–tumor homotypic interactions and tumor–‘normal’ cell heterotypic interactions can be mediated via EVs secreted in the tumor microenvironment. Cell-specific EVs are represented as small spheres with the same color as the cell that releases them.

Finally, attempts have been made to test the functional role of EV secretion in gliomagenesis by knocking-out/-down various drivers of EV biogenesis and secretion that are expressed in gliomas165. Recent knockdown studies have reported that Rab27a/b regulates tumor growth and EV secretion in a mouse glioma model132,198. Thus, further studies focusing on other drivers of EV biogenesis will aid in elucidating the role of EVs in gliomagenesis.

1.3.3.3 EVs in Oligodendroglioma

Oligodendroglioma (ODG) are slower growing, lower grade gliomas associated with a distinct constellation of mutations, including gain-of-function mutations in isocitrate dehydrogenase 1 (IDH1) or IDH2, and chromosomal co-deletion of 1p/19q12, with mutation of CIC in the retained chromosome also prevalent6,7,22,201. Patient-derived ODG cell lines secrete EVs202, but only a handful of studies have examined their roles in ODG tumorigenicity203,204. Strikingly, instead of the growth promoting effects largely attributed to GBM EVs175,176, EVs derived from a mouse G26/24 ODG cell line have cytotoxic effects on neurons and to a lesser extent, astrocytes in vitro203,204. Two pro-apoptotic proteins were identified in G26/24 ODG-EVs that contribute to their cytotoxic effects - Fas ligand (FasL) and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)203,204. Interestingly, FasL was absent in the EVs derived from an anaplastic ODG patient derived cell line UPN9333202, highlighting the heterogeneity of EV content. In another study, the cytotoxic effects of sEVs derived from Human Oligodendroglioma (HOG) cells on other ODG cells in vitro were attributed to the ceramide that makes up these vesicles205. ODG m/lEVs derived from G26/24 ODG cells were also reported to carry aggrecanases (Adamts1, Adamts4, Adamts5) in their cargo, bringing about degradation of aggrecan rich extracellular matrices to promote tumor cell invasiveness in vitro206. There is thus growing support for ODG EVs exerting a potent cytotoxic effect on neurons, astrocytes (non-cell autonomous interaction) and neighboring ODG cells (cell autonomous interactions) via various means.

1.3.3.4 EVs in Schwannoma

Schwannoma are tumors arising from SCs in the PNS207. EVs isolated from primary schwannoma cells were recently reported to carry tumor specific RNA (miR and ncRNA) in their cargo208. PKH67 dye was used to trace the transfer of EVs from host cells to recipient cells, such that EVs budding from the surface of schwannoma cells would be stained with PKH67 as well, allowing easy detection. Upon co-culturing of PKH-67 dye stained

18 schwannoma cells with murine spiral ganglion cells, internalization of PKH-67 stained schwannoma-EVs by spiral ganglion cells was detected, accompanied with transfer of EV RNA cargo as well. Transfer of schwannoma-EVs was also reported in a cochlear explant setup, where the internalization of EVs by cochlear cells lead to neuronal damage. Similar cochlear cell damage was observed upon treatment of cochlear explants with EVs derived from vestibular schwannoma patient derived cell lines208. Thus, schwannoma cell derived EVs can mediate cochlear damage, and further studies may help identify key targets responsible for mediating this effect. Interestingly, a recent study had reported the use of genetically engineered EVs as a therapeutic messenger system (for the delivery of suicide mRNA/protein) to schwannoma cells, thereby inhibiting schwannoma growth in vivo209. Thus, use of EVs in schwannomas as a diagnostic and therapeutic avenue remains to be explored.

1.4 Schwann Cells: An Introduction

1.4.1 Glial cells of the peripheral nervous system: Schwann cells

The peripheral nervous system (PNS), which acts as a bridge between the central nervous system (CNS) and the rest of the body, has both somatic and autonomic components, including spinal and cranial nerves, as well as sensory and autonomic ganglia210. The sensory ganglia contain neuronal cell bodies of afferent nerves that relay sensory information from the periphery to the CNS. The main sensory ganglia include the dorsal root ganglia (DRG, or spinal ganglia) (Fig. 1.3), trigeminal ganglia, and nodose ganglia211. Autonomic ganglia are classified either as sympathetic or parasympathetic – sympathetic ganglia include the superior cervical ganglia and prevertebral sympathetic ganglia, while parasympathetic ganglia include craniosacral ganglia210. The sympathetic and parasympathetic ganglia innervate several target regions including the eyes, lacrimal glands, sub-maxillary gland, parotid gland, heart, larynx, trachea, abdominal and pelvic region210.

Glial cells are non-neuronal cells in the nervous system that support, maintain, and improve neuronal function. Glial cells in the PNS include Schwann cells, satellite glial cells (SGCs), olfactory ensheathing cells, and enteric glial cells212,213. Schwann cells enwrap the length of both myelinated and non-myelinated axons (Fig. 1.3), while SGCs ‘cap’ neuronal cell bodies in the sensory and autonomic ganglia (Fig. 1.3). Olfactory ensheathing cells, which are similar to Schwann cells, ensheathe olfactory axons213-216. Enteric glial cells, on the other hand, are specialized glia located in the autonomic ganglia of the gut217. Of these PNS glial cell types, Schwann cells are the most abundant212. 19

Figure 1.3. Glial cells in the peripheral nervous system.

Dorsal root ganglion (DRG) receives sensory inputs from the body via the dorsal root while motor neuron processes of the spinal cord are sent out via the ventral root. Left inset presents a magnified view of a myelinating Schwann cell, with a myelin sheath deposited along the length of the axon. Nodes of Ranvier are observed at the meeting point of adjacent internodes. Right inset presents a magnified view of a single DRG neuron with the neuronal soma capped with satellite glial cells and the axon enwrapped along the length with Schwann cells.

Schwann cells have multiple developmental and physiological functions, including myelination, clustering of ion channels at the nodes of Ranvier in myelinated axons218, neuronal survival219, and regulation of axonal diameter220. A typical Schwann cell has an elongated morphology with a heterochromatin-rich nucleus and an overlying basal lamina221. In both myelinating and non-myelinating Schwann cells, the cytoplasm is dense, containing glycogen granules, ribosomes, mitochondria, lysosomes, and peroxisomes221. Schwann cells also contain a large Golgi apparatus and a centromere localized at one pole of the perinuclear region221. In myelinating Schwann cells, a blanket of protective myelin formed by layers of tightly compacted cell membrane forms a sheathe that encircles axons in the internodal segments, which are interspersed with nodes of Ranvier (Fig. 1.3)222. These nodes are responsible for the rapid conduction of nerve impulses222. These distinctive features and functions of myelinated axons are similar in the PNS and CNS where oligodendrocytes are the primary glial cells221,223.

Schwann cells can be broadly classified as myelinating Schwann cells, non-myelinating Schwann cells, or peri-synaptic, terminal Schwann cells in the neuromuscular junction224. During development, Schwann cells that associate with axons greater than 1 µm in diameter differentiate into myelinating Schwann cells, which wrap axons with a myelin sheath225,226. In contrast, Schwann cells that associate with smaller diameter axons (< 1 µm; e.g. C-fiber

20 nociceptors) differentiate into non-myelinating Schwann cells, also termed ‘Remak’ cells227,228. Non-myelinating Schwann cells can ensheathe numerous axons together227,228, while a myelinating Schwann cell typically myelinates only a single segment of axon along its length, with a single long axon myelinated by hundreds to thousands of Schwann cells depending on the internodal length225,226. Finally, peri-synaptic Schwann cells are found at neuromuscular junctions which cover the synapse229. Recently a specialized population of Schwann cells was identified in the skin, termed cutaneous Schwann cells, which relay pain sensation230. Under pathological conditions, Schwann cells also have an essential role in promoting peripheral nerve regeneration and restoration of function post injury151,231,232. We further delve into the role of Schwann cells post injury in the following sections.

1.4.1.1 Neural crest cell development & migration

Schwann cell development is a lengthy process involving a series of transitional steps. Schwann cells originate from a multipotent, migratory population of progenitor cells called neural crest cells (NCCs) (Fig. 1.4A-E)233-235. Post-gastrulation, the ectoderm is divided into neural ectoderm and non-neural surface ectoderm233-235. The CNS and indirectly part of the PNS are formed from the neural ectoderm, while non-neural ectodermal cells are destined to form placodes (some of which give rise to neural tissue), dermis and skin233-236. In the developing CNS, inductive signals from the notochord and the node at the anterior end of the primitive streak induce overlying ectodermal tissue to acquire a neural identity234,237. As a result, a thick columnar epithelial plate-like structure is formed, called the neural plate. The neural plate border is the interface that separates neural ectoderm from non-neural ectoderm (Fig. 1.4B). Changes in cell shape occur at the lateral margins of the neural plate resulting in elevation of the neural plate and the formation of neural folds. These folds begin to close-in and fuse, transforming the plate into the neural tube - the precursor of the CNS234,238-240.

Figure 1.4. Trunk neural crest cell migration.

(A) Neurulation observed in developing embryos (embryonic day 8.5). Inserted circle marks developing trunk neural crest cells elaborated in (B-E). (B) Neural- and non-neural ectoderm is separated by the neural plate border. Neural ectoderm (neural plate) gives rise to the CNS while the non-neural ectoderm forms the placodes, dermis, and skin. (C) Elevation of the neural plate (neural folds) occurs culminating in the fusion, and formation of the neural tube (nt). Pre- migratory neural crest cells are specified parallel to the formation of the neural tube. (D,E) Trunk neural crest cells then undergo EMT, delaminate and migrate through distinct dorsolateral or dorsoventral pathways giving rise to sensory, autonomic, mesenchymal cell fates.

As the neural folds fuse, the neural tube loses contact with the overlying non-neural ectoderm (Fig. 1.4C,D). The cells at the dorsal most limit of the neural tube are destined to become NCCs. NCCs are precursor cells that express “neural crest specifier” genes, including the transcription factors Sox9, Sox10, Snail1 (Snail), Snail2 (Snai2, or Slug), Ets1, and FoxD3 (Fig. 1.5)236,237. Prior to migration out of the neural tube, NCCs cannot be distinguished from neuroectodermal cells that remain in the neural tube241. NCCs then undergo a complete or partial epithelial-to-mesenchymal transition (EMT) coupled with a down-regulation of cell adhesion molecules such as E-cadherin242, which marks NCCs for delamination from the neuroepithelium and their subsequent migration 243. Based on single cell RNA sequencing- spatial transcriptomic data, NCCs in the pre-EMT state express both neural plate border specifier and neural tube-specific genes, such as Zic1/3/5, Msx1, Mafb, Gdf7, Olig3, FoxB1, while delaminating NCCs initiate the expression of a new suite of markers, including Snail1, Dlx5, Hapln, Pak3 and Pdgfra244.

Figure 1.5. Cellular transitions steps in the Schwann cell lineage.

Schwann cell development is marked by three transitory stages: a) Migratory neural crest cells (embryonic (E) day 10.5) fated towards a glial lineage generate Schwann cell precursors (SCPs) associating with axons (E12.5) b) Immature Schwann cells (E14.5) ; and c) Pro- myelinating Schwann cells (E18.5), which differentiate into mature myelinating or non-

23 myelinating Schwann cells post-natally. The reversible arrows represent the plasticity of the mature Schwann cells post-injury, represented by Repair Schwann cells. Key determinants and markers of Schwann cell development expressed during the distinct stages are noted (in blue).

NCCs are divided into four populations depending on their axial positions: cranial, cardiac, trunk, and vagal/sacral. NCCs take up distinct region-specific migratory pathways, which exposes them to different inductive signals depending on the path that they take, such that they differentiate into specific cell types237. Furthermore, multipotent NCCs take up appropriate cell fates (sensory, autonomic, mesenchymal) via sequential fate decisions, culminating in the induction of mutually exclusive gene expression programs specific to their cell identity244. Delaminating trunk NCCs have an initial neuronal fate bias, while cranial NCCs exhibit a mesenchymal fate bias244. Trunk NCCs migrate along either of two pathways: dorso-lateral or -ventral (Fig. 1.4E). Cells migrating along the dorso-lateral pathway move between the dermamyotome and the ectoderm and differentiate into melanocytes240,245. Cells that migrate along the dorsoventral pathway include two lineages; cells that migrate first between the somites and reach the dorsal aorta give rise to sympathetic chain ganglia and aortic plexus, while NCCs that migrate through the rostral half of the somites give rise to Schwann cells and boundary cap cells that populate the DRGs and adrenal medulla240,245,246.

1.4.1.2 Embryonic Schwann cell development - from Neural Crest Cells to mature Schwann cells:

Neural crest cell (NCC) stage: During mouse embryogenesis, migrating NCCs are observed beginning at embryonic day (E) 10.5. At this stage, NCCs express a number of transcription factors, including Sox9 247, Sox10247,248, Etv5249, Tfap2a250, and Pax3251 (Fig. 1.5). Migrating NCCs also give rise to boundary cap (BC) cells, which are multipotent in nature and occupy the dorsal root entry zone and motor exit points (Fig. 1.3)246,252-254. BC cells give rise to the Schwann cells found in the dorsal and ventral roots of the dorsal root ganglia254. Interestingly, BC cells can also give rise to oligodendrocytes when transplanted into the CNS, highlighting their multipotent status255. A subset of satellite glial cells and nociceptive neurons populating the DRG also arise from BC cells254.

Schwann cell precursor (SCP) stage: Egr2 is expressed in BC cells and in Schwann cells in the dorsal and ventral roots from E10.5 until birth254. Other genes expressed by BC cells include

Wif1, L20, Sema6A, Sema6D, Hey2, HeyL, Npr3, Hbb-y, and Hbb-b1 256. At E12.5, migratory NCCs predestined to take up a glial fate give rise to Schwann cell precursors (SCPs) (Fig. 1.5) 257-259. Morphologically, SCPs lack basal lamina but associate directly with growing axon bundles and are thus distinguished from migrating NCCs. SCPs are located proximal to the growing nerve tip, promoting the nerves to become compact in structure and also guiding axons to their targets 152,258. SCPs express several transcription factors, including Sox2, Sox10, Tfap2a, Pax3 and Sox9 258,260, and in the dorsal and ventral roots, also Egr1 and Egr2 152,260,261. Other non-transcription factor markers that are expressed in SCPs include Ngfr, Bfabp, and Gfap 260,262,263.

The survival and proliferation of SCPs depends upon neuregulin 1 (NRG1) – isoform type III, an axon derived survival signal that binds to the NRG1 receptors Erbb2 and Erbb3, which are expressed by SCPs264. Bone morphogenetic proteins (BMPs) also have an important role in Schwann cell development265. BMP2 suppresses expression of mature Schwann cell-specific genes in SCPs by inducing early stage glial genes such as Gfap265. BMPs also have neurogenic potential and direct a subset of NCCs towards the neuronal lineage during PNS development265. Interestingly, SCPs residing in the cranial and trunk nerves are responsive to the neurogenic effects of BMP2 and can switch between neuronal or glial fates, thereby giving rise to neurons in the parasympathetic ganglia266,267. The switch in SCP fate is regulated by BMP2, which induces Ascl1 expression in SCPs that are biased towards a neuronal fate 266. Thus, SCPs, like NCCs, are multipotent, and can give rise to different cell types including Schwann cells, parasympathetic neurons (as described above;266,267), endoneurial fibroblasts, melanocytes, as well as chromaffin cells of the adrenal medulla 268-271. In addition, SGCs found within the DRG have can also competently give rise to SCPs in vitro249, but whether SGCs have this role in vivo has not been addressed.

Immature Schwann cell stage: As development proceeds, SCPs give rise to immature Schwann cells (iSCs), which are detected from E14.5 onwards and are observed up until birth (Fig. 1.5)152. iSCs differ from SCPs in their gene expression profile; S100β expression is initiated in iSCs, while SCP-specific genes such as Tfap2a, Egr1, Pax3, and Bfabp are downregulated in iSCs152,258,261. iSCs secrete autocrine survival factors so that the cells are no longer entirely dependent on axon-derived NRG1152,272. iSCs cluster and form Schwann cell ‘families’273,274 around developing axons, depositing basal lamina around a shared axon (Fig. 1.5) 275-281. The axonal bundle is penetrated by cytoplasmic processes of iSCs, which are then able to

25 distinguish large and small diameter axons, with larger axons rearranged to the periphery of the bundle152,274,282. Similar to SCPs, iSCs proliferate and also undergo apoptosis in order to ensure that correct numbers of Schwann cells develop along with the nerves282-284.

Pro-myelinating/myelinating Schwann cell stage: Just prior to birth, iSCs contact large diameter axons, which produce high levels of NRG1, in a proportional manner (i.e., one iSC associates with a single axon), radial sorting ensues282. Radial sorting refers to the selection of a single axon by an iSC for myelination282. These iSCs become pro-myelinating Schwann cells, a transient cell population that ultimately become myelinating Schwann cells. In contrast, iSCs that pair with smaller diameter axon bundles, which release lower levels of NRG1, become non-myelinating Schwann cells283,285.

The key markers for pro-myelinating Schwann cells include the transcription factors Pou3f1286, Nfatc4287, and Yy196 (Fig. 1.5). Egr2 is also expressed at low levels in pro-myelinating Schwann cells, while Sox2 and Ngfr expression are downregulated from this stage onwards152,260,288. Myelinating Schwann cells express high levels of Egr2 and myelin-related genes such as myelin basic protein (MBP), myelin protein zero (MPZ), Peripheral myelin protein 22 (Pmp22) and others. Schwann cells ensheathe axons with a layer of myelin and these genes are expressed up to two weeks post birth289 . Apart from Sox10, Sox2, Jun, other genes expressed in non- myelinating Remak Schwann cells include Pax3, Egr1, NCAM, Ngfr290-293. Interestingly, Pax3, Egr1, and Ngfr are also upregulated in myelinating Schwann cells post injury, contributing to the proliferative repair Schwann cell phenotype discussed in the next section. Thus, myelinating Schwann cells exhibit a gene expression profile supporting activation of the myelin program, while non-myelinating Remak Schwann cells express genes that support a proliferative phenotype.

Small GTPases of the Rho family (e.g. Cdc42, RhoA, Rac1) and Ras family (RalA and RalB) are also expressed in Schwann cells and play important roles in Schwann cell proliferation, radial sorting and myelination294-297. By acting as molecular switches, shuttling between an activated GTP bound state and inactive GDP bound state, Rho and Ras family GTPases regulate cytoskeletal organization in Schwann cells to modulate critical events such as radial sorting and repair294-297. Indeed, chronic lack of Ral proteins in Schwann cells impairs radial sorting, resulting in the formation of unmyelinated or hypo-myelinated large caliber axons, and abnormalities in the myelin sheath294. Similarly, the large GTPase Dynamin2 (Dnm2) is also required for Schwann cell survival, radial sorting, and myelination 298. Deletion of Dnm2 in 26 developing and/or adult Schwann cells results in Schwann cell apoptosis, radial sorting impairment, and a demyelinating phenotype akin to defects seen in peripheral neuropathies298. Thus, both small and large GTPases play important roles in radial sorting.

1.4.2 Genes regulating Schwann cell specification and differentiation

Several transcription factors regulate the progression from a NCC stage to a myelinating/non- myelinating Schwann cell stage260. Many of these same factors contribute to the induction of a repair Schwann cell phenotype post injury231. The major transcription factors involved in Schwann cell development, maturation, and repair are summarized herein.

1.4.2.1 Sox10

In mouse, Sox10 expression is first detected at E8.5 in the anterior dorsal neural tube, and later, at E10.5, in cranial ganglia, developing DRGs and otic vesicles260,299. Sox10 is expressed in migrating trunk NCCs, which are destined to form glia, neurons, and melanocytes247,260. Sox10 is essential for Schwann cell development 248 and is maintained in Schwann cells and SGCs while Sox10 expression is downregulated in cells taking on a neuronal or non-neural fate from E12.5 onwards263. While some Sox proteins are expressed transiently in glial lineages, such as Sox2300, Sox10 is expressed in Schwann cells and SGCs throughout development, and is required to maintain a glial phenotype248,263. Expression of Egr2 and other myelin associated genes (MBP, MPZ, myelin associated glycoprotein (MAG) and connexin-32 (Cx32)) in Schwann cells is also directly promoted by Sox10301-305. Sox10 acts synergistically with other transcription factors such as Pou3f1 and Nfatc4 to activate Egr2. Sox10 also cooperates with Nfatc4 and Egr2 to transactivate peripheral myelin genes287,303,305-308. Sox10 can transactivate

309 and induce S100 expression in Schwann cells in vitro . In Sox10 mutant mice, DRG neurons differentiate, but Schwann cells and SGCs are not generated, resulting in a secondary degeneration of sensory and motor neurons due to the absence of peripheral glial cells310. Sox10 deletion also results in loss of expression of Egr2 and peripheral myelin genes, leading to myelin sheath degeneration, axonal death and reduced nerve conduction311. In addition, Sox10 loss of function studies have revealed that Sox10 is essential for survival of early migrating trunk and vagal NCCs, but not for the survival of adult Schwann cells311-313.

1.4.2.2 Sox2

Sox2 transcripts can be detected from E9.5 onwards314 in the developing nervous system, branchial arches, sensory placodes, and gut. Sox2 maintains neural progenitor cells in an 27 undifferentiated state such that a decline in Sox2 expression is associated with differentiation314-316. The subset of cells that develop into the PNS show an upregulation of Sox2 expression300. As development progresses, a decline in Sox2 expression is seen associated with neuronal commitment while continued (but low) Sox2 expression is observed in SCPs and iSCs300. Notably, it is the cross-repressive interaction between Sox2, Egr2 and Mitf that regulates the differentiation of SCPs into either myelinating Schwann cells or melanocytes269. A decline in Sox2 expression is associated with differentiation of iSCs into pro-myelinating and myelinating Schwann cells288.

Sox2 is among the key reprogramming factors that have been used to drive somatic cells to adopt an induced pluripotent stem cell (iPSC) fate317,318. Due to functional redundancy with the closely related genes Sox1 and Sox3, Sox2 can be replaced with Sox1 or Sox3 during reprogramming319. Sox2 also has an important role in peripheral nerve regeneration320. Axonal injury is typically followed by Schwann cell de-differentiation, which reverts mature myelinating Schwann cells to a ‘repair state’ that phenotypically is similar to iSCs151. An upregulation in Sox2 expression is characteristically observed post injury, which promotes relocation of N-cadherin molecules to facilitate clustering of Schwann cells320. This clustering effect results in formation of multicellular cords (Bünger bands, alluded to earlier) responsible for guiding the regenerating axon through the injured site320, thereby promoting recovery. Sox2 expression also promotes infiltration of macrophages into the nerve which play a role in clearance of myelin and axonal debris from the site of injury321.

1.4.2.3 Jun

Jun encodes a zinc finger transcription factor that is a part of the Activator protein-1 hetero- dimeric complex along with Fos322. Jun is expressed in late iSCs at E17 and is then downregulated by Egr2 with the onset of myelination323. Jun expression is also characteristically up-regulated post nerve injury, coupled with the rapid activation of the JNK pathway324,325. Overexpression of Jun is associated with a decline in myelination and de- differentiation of Schwann cells, along with a decline in Egr2 and MPZ levels325. Thus, Jun is an inhibitor of myelination and has been aptly deemed as a master regulator of nerve repair in the PNS, resulting in the generation of a repair Schwann cell phenotype that is essential for regeneration151,323-326. Jun also plays a role in clearance of myelin debris post injury, which is crucial for successful nerve regeneration324,326,327. Unexpectedly, Jun expression in Schwann cells also affects motoneuron survival and axonal regeneration through neurotrophin

28 production such as glial derived neurotrophic factor and artemin after nerve injury328. All these point to the importance of Jun in the repair process after nerve injury.

1.4.2.4 Pax3

In the PNS, Pax3 is expressed in NCCs and is maintained in NCCs that are fated for the glial lineage291. Pax3 transcripts are observed in SCPs as well as in iSCs and is important to induce Schwann cell proliferation291. However, Pax3 expression declines as iSCs undergo radial sorting291. An upregulation in Pax3 expression is observed postnatally in non-myelinating Schwann cells290,291. With respect to function, Pax3 is important for inducing Schwann cell proliferation in vivo, while in vitro Pax3 prevents apoptosis of Schwann cells mediated by TGFβ329. Pax3 prevents developing Schwann cells from exiting the cell cycle and differentiating into myelinating cells by inhibiting Egr2 driven MPZ and MBP expression330. On the contrary, ectopic expression of Pax3 results in activation of non-myelinating Schwann cell markers such as L1, GFAP, NGFR and NCAM291. Interestingly, an upregulation in Pax3 expression in vitro is accompanied by a decline in Jun expression, even though both Pax3 and Jun are negative regulators of myelination330.

1.4.2.5 Tfap2a

Transcription factor activator protein 2α (Tfap2a) is expressed from E8.5 onwards in the developing mouse embryo in the cranial NCC as well as ectoderm331. Post NCC migration, NCCs fated for a glial lineage maintain expression of Tfap2a250. In vitro, overexpression of Tfap2a is responsible for impeding the transition of SCPs to the immature stage250. Thus, Tfap2a is one of the few transcription factors that mark the SCP stage and its expression can be used to monitor early transition stages during development.

1.4.2.6 Early growth response genes: Egr1 and Egr2

The zinc finger transcription factors Egr1 (Early growth response1 or Krox24) and Egr2 (or Krox20) are among the key regulators of Schwann cell development. Even though Egr1 and Egr2 bear nearly identical DNA binding domains, these transcription factors have opposing roles with respect to myelination257. Activation of Egr2 is responsible for the terminal differentiation of Schwann cells to a myelinating phenotype while Egr1 is considered a non- myelinating Schwann cell marker261. The two Egr genes are expressed in a mutually exclusive manner; i.e., Egr1 is expressed in SCPs but is downregulated as the cells mature into myelinating Schwann cells261. Post birth, an upregulation in Egr1 expression is seen in non- 29 myelinating Schwann cells261. Egr2, on the other hand, is expressed in developing Schwann cells populating the dorsal and ventral roots from E10.5 onwards but remains conspicuously absent from Schwann cells developing in the DRG and the nerves through these stages254. Egr2 expression is prominently observed only post-birth in mice261. Post nerve injury, rapid loss of Egr2 expression in Schwann cells at the site of injury is observed, which in part promotes suppression of myelin gene expression261,332. In contrast, Egr1 expression is upregulated in Schwann cells post injury, mimicking embryonic SCPs and further highlighting the importance of Egr1 in attaining a proliferative repair Schwann cell phenotype261.

1.4.2.7 Pou3f1

POU domain class 3 transcription factor (Pou3f1), also known as Oct6 or SCIP - suppressed cAMP inducible POU, is well studied for its role as a regulator of Schwann cell development333. Pou3f1 is a pro-myelinating Schwann cell marker that is essential for the terminal differentiation of pro-myelinating to myelinating Schwann cells. Pou3f1 expression peaks at postnatal day (P)1 in mice286 and induces Egr2 expression289. Pou3f1 is expressed until P11 after which expression gradually declines with myelination289. However, Pou3f1 expression is maintained in non-myelinating Schwann cells257. Pou3f1 induces Egr2 expression, which in turn promotes the expression of several myelin proteins334. Pou3f1 can also act in synergy with Sox10 (dimeric Sox10 binding occurs to activate Myelin Schwann cell enhancer of Egr2) to induce Egr2305,333. Pou3f1 deficient mice show a transient but severe arrest at the pro- myelinating stage333. The myelination arrest is eventually overcome by P10 via other compensatory mechanisms, and myelin formation begins. Egr2 expression gets delayed till P7 in Pou3f1 deficient mice333.

Notably, constitutive overexpression of Pou3f1 results in a persistent hypo-myelination phenotype in mice and gradual axonal loss335. Egr2 levels in this case remained unchanged which could possibly indicate that the Pou3f1:Sox10 ratio should be strictly maintained to induce Egr2 expression335. Pou3f1 is a transcriptional repressor of MBP and MPZ promoters336. In line with this function, a severe decline in MPZ, MBP and Pmp22 expression was reported in a Pou3f1 gain-of-function study335. Egr2 is capable of inducing expression of several myelin genes337,338, however, the hypo-myelination phenotype suggests that the repression of myelin genes by Pou3f1 dominates over their induction by Egr2335. Thus, apart from inducing terminal differentiation in pro-myelinating Schwann cells via Egr2 expression, Pou3f1 also represses MBP and MPZ expression and prevents premature Schwann cell myelination. A decline in

Pou3f1 levels with time allows MBP and MPZ to be expressed bringing about correctly timed myelination. Pou3f1 thus has dual roles as a positive and negative regulator of myelination in the PNS305,339,340. Finally, Pou3f1 expression in Schwann cells is upregulated post nerve injury, accompanying the marked decline in Egr2 expression.

1.4.2.8 Pou3f2 and Pou3f3

Pou3f2 (also known as Brn2) is closely related to Pou3f1 and also shares similar biochemical properties341. Pou3f2 is also co-expressed with Pou3f1 during Schwann cell development342. Pou3f3 (also known as Brn1), on the other hand, is more distantly related to Pou3f1343 and is not expressed during normal Schwann cell development. Deletion of Pou3f2 alone does not inhibit the differentiation of Schwann cells, but deletion of both Pou3f2 and Pou3f1 results in a severe and prolonged arrest of Schwann cells in the pro-myelinating state342. Knock-in studies have shown that replacing Pou3f1 with Pou3f3 or Pou3f2 can compensate completely (by Pou3f3)344 or partially (by Pou3f2)342 for the loss of Pou3f1 during Schwann cell development.

1.4.3 Nerve growth factor receptor and Neuregulin signaling: Role in Schwann cell myelination

1.4.3.1 Nerve growth factor receptor

The nerve growth factor receptor (Ngfr; also known as p75 neurotrophin receptor) and neurotrophic receptor tyrosine kinases (Ntrk) 1-3 (also referred to as TrkA, B, C) are the two major classes of receptors that bind to neurotrophins345,346. Of these two classes, Ngfr is a low affinity receptor with functions ranging from promotion of cell survival to induction of cell death, depending on co-receptor binding and the precursor or mature form of the neurotrophins345.With respect to Schwann cells, Ngfr regulates survival and myelination347,348. Ngfr is expressed in migrating NCCs and continues to be expressed in developing Schwann cells, after injury, or after dissociation from nerves and placed in cell culture conditions349. It is important to note that Schwann cells express negligible levels of TrkA and TrkB in a truncated form350,351. As pro-myelinating Schwann cells mature to a myelinating state, a decline in Ngfr expression is observed352. An adaptor protein of Ngfr, Receptor interacting serine/threonine protein kinase 2 or RIP2, modulates the effects of Ngfr signaling353,354. In vitro, Ngfr promotes survival in the presence of RIP2 ; but a decline in RIP2 results in Schwann cell death353. However, myelination ability of Schwann cells is reduced in Ngfr knockout mice - even though Schwann cell death is not elevated355. This diversity in functions is 31 interdependent on the extrinsic factors acting in the Schwann cells simultaneously. For example, activation of nuclear factor kappa B (NFκB) via Brain Derived Neurotrophic Factor (BDNF) is essential for iSCs to develop into myelinating Schwann cells356. In stress conditions, such as serum deprivation, nerve growth factor signaling via Ngfr leads to nuclear translocation of the NFκB subunit, p65, in iSCs356. The activation of p65 initiates anti-apoptotic pathways preventing Schwann cell death356. However, in the absence of NFκB (or p65 knockout) and survival signals from the axon, nerve growth factor signaling via Ngfr leads to activation of the apoptotic pathway in these iSCs356. Thus, Ngfr plays a role in promoting survival as well as apoptosis in Schwann cells.

1.4.3.2 Neuregulins

Neuregulins are axon-derived signaling proteins that are essential during Schwann cell development357. Of the four types of neuregulins, NRG1 can bind to tyrosine kinase receptors ErbB2 and ErbB3 expressed by Schwann cells357. NRG1 mainly exists in either transmembrane and soluble forms and has several isoforms of which NRG1 type II and type III are relevant during Schwann cell development357,358. In vivo, NRG1 type III is essential during the terminal differentiation stage of Schwann cells to bring about successful myelination359. Transmembrane NRG1 type III remains tethered to the axonal membrane by means of a cysteine rich domain359. Upon proteolytic cleavage, the molecule undergoes conformational change resulting in a more accessible Epidermal growth factor (EGF)-like domain to the ErbB3 binding site on Schwann cells, triggering the recruitment of ErbB2 and downstream signaling360. The activation of ErbB heterodimers by the membrane bound NRG1 type III is thought to play a critical role in the early event of myelination, since no myelin formation is observed with co-culture of Schwann cells and neurons from NRG1 type III null mice283.Low level of axonal NRG1 type III in normally non-myelinated sympathetic axons and axons from NRG1 type III+/- neurons result in axon ensheathment but poor myelination. Intriguingly, low level of soluble NRG1 type III is able to rescue the deficit and restore proper myelination, whereas high level of soluble NRG1 inhibits myelination361,362; the dichotomous effects suggest a dose dependent regulation of myelination by NRG1 signaling. Additionally, NRG1 type III activates the PI3K pathway resulting in Akt phosphorylation, which is a key step in the process of myelination283.

To achieve differentiation of Schwann cells in vitro, NRG1 (type III) and high levels of cAMP are required363,364. Under these conditions, cAMP response element-binding protein (CREB)

32 gets phosphorylated at Ser133 indicating that CREB could have a possible role during the myelination step363. On the other hand, NRG1 with low levels of cAMP act as a mitogen to promote proliferation of Schwann cells363. Constitutive activation of cAMP activated pathways and their effect on downstream Schwann cell-specific gene expression could shed more light on the importance of CREB in Schwann cell development and could also prove important from a reprogramming perspective. In contrast to NRG1 type III, soluble NRG1 type II isoform signals in a paracrine fashion to inhibit myelination361. The inhibition is brought to effect via activation of the MEK/ERK pathway and blocking this pathway allows soluble NRG1 type II to promote myelination365. Activation of MEK/ERK pathway also promotes Jun expression – a known inhibitor of myelination324. The activation of MEK/ERK pathway is capable of inhibiting myelination alone as well as in cooperation with low levels of soluble NRG1 type II361,365,366. It is the balance between the activation states of Ras/Raf/ERK and PI3K pathways that guides the decision to myelinate a nerve or not, and NRG1 plays a crucial role in this process. Apart from this, activation of Ras/Raf/ERK pathway also promotes the de- differentiation of Schwann cells in vitro367.

In summary, several aspects of Schwann cell development have been examined and elucidated. However, some aspects remain unclear, and are the topic of investigation in this thesis.

1.4.4 Peripheral nerve injury and role of Schwann cells post injury

A common misconception is that neuronal regeneration occurs in the PNS such that cell-based repair-strategies are not required. The reality is that peripheral nerve injury (PNI), which can be the result of trauma or peripheral neuropathies, frequently results in lifelong disability. Nearly 360,000 people in North America alone suffer from upper extremity PNI annually, resulting in 8,648,000 restricted activity days and 4,916,000 bed/disability days per year368. Recovery from PNI is often suboptimal and life-long functional impairment and neuropathic pain is common369. Autologous nerve grafting has been the standard treatment in cases of severed nerve injuries that create a gap between the proximal and distal nerve370. However, nerve grafting requires that a healthy nerve is harvested from the patient, which can itself lead to donor-site morbidity, including chronic and debilitating neuropathic pain371. Moreover, nerve graft repair yields relatively poor results; only 25% of patients recover full motor function and only 3% regain full sensory function after median nerve repair368. Much of this failure is due to poor engraftment between donor and recipient nerves and delayed surgical intervention, resulting in distal muscle atrophy and fibrosis. Schwann cells, which are required

33 to support axonal regeneration during peripheral nerve repair, also have limited remyelination capacity in chronically denervated distal nerves368,372-377.

Given the inability of the peripheral nerve to self-heal or use therapeutic interventions to repair the injured nerve, there is a growing need to develop alternative therapeutic approaches. One alternative to autologous nerve grafting is the use of nerve conduits, also called axon guidance channels, in which axonal regeneration bridges the nerve ends within the lumen of the tube378,379. Nerve graft materials used in nerve conduits are biocompatible, elicit no or negligible inflammatory response, promote axonal elongation and have an appropriate degradation rate after nerve regrowth380. There are a few reports of successful nerve repair using stand-alone nerve conduits, but in humans, repair is typically limited to small digit nerves and outcomes are similar to nerve autografts381. Nerve conduits also fare poorly compared to autografts when used in ‘critical’ length defects, with the risk of complete regeneration failure owing to the lack of cellular support within the conduit382. Nerve conduits may thus be better suited as a platform for the delivery of growth-enhancing substrates, including various forms of nerve and glial growth factors, denatured muscle, small segments of peripheral nerve, and/or purified Schwann cells383-385.

Grafts of nerve-like conduits containing autologous nerve-derived Schwann cells can promote fast and more efficient nerve regeneration386. Schwann cells are able to promote axonal growth, in part, through the release of neurotrophic factors384,387-392, making them an excellent supplement to conduits. The addition of Schwann cells within nerve conduits has demonstrated enhanced axonal regeneration and improved functional recovery in mice, rats, and in nonhuman primates383,390,393,394.

The first clinical use of autologous Schwann cells to supplement sciatic nerve autograft repair was reported in 2017 and strong motor function of the tibial nerve and partial restoration of sensation was observed in one patient395. However, this approach suffered from similar limitations as nerve grafting: requirement for a nerve biopsy from the patient to isolate Schwann cells, as well as extended expansion time of isolated Schwann cells to create sufficient numbers for transplantation386. Moreover, autologous Schwann cell transplantation strategies are further hampered by the limited proliferative capacity of adult nerve-derived Schwann cells in vitro, induction of senescence in Schwann cells grown ex vivo, and finally, the greatly diminished regenerative capacity of Schwann cells derived from patients of advanced age326,396,397. Thus, other sources of Schwann cells are being explored for transplant 34 purposes398,399. One such approach is to derive Schwann cells using new innovations in the field of cellular reprograming, which will be further elaborated on in the following sections.

1.4.5 Phenotype of a repair Schwann cell

An important property of endogenous Schwann cells is their ability to de-differentiate from a mature myelinating or non-myelinating state to a precursor-like ‘repair’ state after PNI151,352. Repair Schwann cells largely resemble embryonic Schwann cells with respect to their gene expression profiles, but they also express genes involved in EMT as well as unique genes, such as Olig1, Artemin, GDNF, BDNF, and Shh (Fig. 1.5)151,324,400. Additionally, repair Schwann cells morphologically appear longer and thinner than iSCs401. Strikingly, both myelinating and non-myelinating Schwann cells take up a similar repair phenotype post injury (Fig. 1.6A-C)400. Schwann cell de-differentiation post injury is triggered by a loss of axonal contact151. Concurrently, Schwann cells at the site of injury downregulate myelin genes to reduce myelin production since myelin is inhibitory to nerve repair324,402,403. Repair Schwann cells, along with macrophages infiltrating the injury site, carry out myelin debris clearance via autophagy and phagocytosis, respectively, helping create an environment that is less inhibitory and hence permissive for axons to re-grow (Fig. 1.6A,B)403-406. As axon regeneration proceeds, emergence of axonal guidance structures, consisting of basal lamina scaffolds, called Bünger bands, is observed400. However, as the response period progresses, the repair ability of Schwann cells in the injured site declines, resulting in poor functional recovery324,368,372-377, possibly due to deprivation of axonal communication372.

Figure 1.6. Schwann cells in peripheral nerve repair.

(A) Peripheral nerve injury may result in axonal degeneration. Axonal and myelin debris is observed distal to the site of injury. Degeneration of perisynaptic Schwann cells ultimately results in distal tissue atrophy. (B) Repair Schwann cells populate the distal stump and promote axonal regeneration and distal target tissue reinnervation. (C) Repair Schwann cells can successfully myelinate the regenerated axon promoting tissue growth.

Advanced age also greatly diminishes nerve regenerative capacity326. Nerve grafts isolated from younger mice potently promote nerve regeneration compared to nerve grafts from older mice, resulting in increased macrophage infiltration and repair Schwann cell phagocytosis is observed using younger nerve grafts while a hyperinflammatory response is observed in older nerve grafts post injury326,407,408. Notably, a delay in expression of key regulators of a repair phenotype (Notch, Jun) is also observed in aged Schwann cells409. Additionally, other factors 36 including survival, proliferation, differentiation, and myelination potential of Schwann cells also declines with age410. Taken together, these studies further emphasise the importance of developing a renewable Schwann cell source for cell-based therapies to accelerate and improve nerve repair potential in the aging population. Interestingly, there are sexually dimorphic differences in nerve repair with the female cohort exhibiting faster regeneration post injury compared to males411,412. However, further studies are required to identify the molecular network governing these differences.

Several regulators of repair Schwann cells have been identified (reviewed extensively in231,232). Notable among these are ERK1/2367,413,414, Jun N-terminal Kinase (JNK)325,415, and p38 MAPK416,417 signaling pathways, all of which are induced in repair Schwann cells post nerve injury. Both the JNK and p38 MAPK pathways have similar roles in repair Schwann cells. JNK activation results in phosphorylation of Jun, Schwann cell proliferation, and inhibition of the myelination program323,325,415. Similarly, the p38 MAPK pathway induces Jun expression, downregulates the myelination program, and modulates the association of repair Schwann cells with the axons post injury416,417. Finally, induction of ERK1/2 promotes the de-differentiation of Schwann cells at the site of injury367,414, while induction of ERK1/2 activity in Schwann cells under normal physiological conditions promotes a transient demyelinating phenotype418. Strikingly, sustained activation of MAPK signaling in developing Schwann cells results in premature and continuous myelin synthesis, ultimately culminating in a hyper-myelinating phenotype419. However, sustained activation of MAPK/ERK signaling in mature Schwann cells did not result in improved remyelination or increased myelin production by repair Schwann cells post injury420. Conversely, myelin compaction defects and infolding were observed with sustained MAPK/ERK activation in Schwann cells420. Furthermore, formation of Remak bundles post injury was also impaired 420. Thus, sustained induction of MAPK/ERK activity in repair Schwann cells has deleterious effects420.

Schwann cells have an indispensable role in rapidly clearing myelin debris and remyelination post injury, but it has been suggested that this repair function could be hampered if Schwann cells were to myelinate multiple axons. Interestingly, a recent study demonstrated that myelinating Schwann cells carry an inherent plastic potential with respect to axon interactions, similar to oligodendrocytes, and can myelinate multiple axons at a given time421. This study reported that Fbxw7, an E3 ubiquitin ligase component, acts as a gatekeeper, preventing myelination of multiple axons by Schwann cells. Notably, ablation of Fbxw7 in Schwann cells

37 increases Jun and mTOR expression coupled with myelination of multiple axons by a single Schwann cell421. While Fbxw7 mutant Schwann cells morphologically resemble repair Schwann cells post nerve injury, it remains to be determined whether the hyper-myelinating cells could promote repair. Future studies will be required to test the potential for Schwann cells lacking Fbxw7 expression to be exploited for therapeutic applications in PNS disease models such as Charcot-Marie-Tooth disease, a demyelinating hereditary neuropathy (reviewed in422).

1.4.6 Identifying an alternate source for Schwann cells: reprogramming somatic cells to generate Schwann cells

1.4.6.1 Differentiation of stem cells to Schwann cells

To use Schwann cells clinically, it is necessary to culture these cells on a large scale in vitro. In experimental animals, the sciatic nerve is a common source for harvesting and culturing Schwann cells 423. However, nerve-derived Schwann cells are not readily accessible386, and to acquire a large number of cells, long expansion periods are required423. Given these limitations, nerve-derived Schwann cells are not an ideal source for experimental procedures or for clinical applications424. Hence, alternative sources for generating Schwann cells are required. Numerous attempts have been made to generate human Schwann cells from somatic stem or progenitor cells, including skin-derived neural crest stem cells425, hair follicle-derived neural crest stem cells 426, muscle derived stem/progenitor cells426, dental pulp stem cells427, umbilical cord- or bone marrow derived mesenchymal stromal cells428,429, adipose tissue derived mesenchymal stem cells 430, and induced pluripotent stem cells (iPSCs)431,432.

Each of these stem/progenitor cell populations can be induced to differentiate into Schwann cells when exposed to appropriate stimulatory cues. For example, human embryonic stem cells (ESCs) cultured as neurospheres in ‘Schwann cell differentiation media’ containing NRG1 and forskolin differentiate into mature myelinating Schwann cells within ~12 weeks433. Similarly, human ESCs have been directed towards a SCP fate and further differentiated into mature Schwann cells over a 4 week period using a combination of chemical small molecule inhibitors (SB431452; a TGF inhibitor, CT99021; a glycogen synthase kinase 3/GSK3 inhibitor), NRG1, and forskolin431. Generation of SCPs using such an approach has multiple advantages in therapeutic applications: 1) repair Schwann cells mimic an embryonic Schwann cell phenotype, and hence use of a pure SCP population may fare better in a clinical setting

38 compared to mature myelinating Schwann cells, 2) SCPs are an expandable cell population, and thus better suited in a clinical setting compared to mature Schwann cells. However, there are also potential issues with directed differentiation strategies that must be considered. For example, Schwann cells generated by the differentiation of human adipose derived mesenchymal stem cells were found to revert to a stem cell like state upon withdrawal of glial induction factors from the differentiation media 430. Thus, it is imperative that the differentiated Schwann cells be assessed in the absence of stimulatory cues over extended cell passages, to determine the stability of the cell type.

Adult skin dermis, which can be obtained from patients with minimal morbidity, has a reservoir of multipotent mesenchymal progenitor cells similar to embryonic NCCs434. These NCC- related precursor cells, termed skin-derived precursors (SKPs), readily differentiate into NCC progeny, including Schwann cells, in vitro in response to appropriate cues435-439. By removal of basic fibroblast growth factor (bFGF) and EGF followed by addition of N2 supplement, forskolin, and NRG1, SKPs from facial skin differentiate into skin-Schwann cells that can be expanded over multiple passages, and when transplanted, associate with axons and generate myelin435,436,440.

Skin-Schwann cells exhibit a gene expression profile like embryonic Schwann cells; both express Pou3f1, Tfap2a, Sox2, Jun, S100, and Cdh19439. Notably, skin-Schwann cells exhibit better proliferative and myelination abilities in comparison to adult nerve-derived Schwann cells441. Skin-Schwann cells have been studied in both acute and chronic nerve injury settings in rodent models with encouraging results372,392,441,442. Skin-Schwann cells are currently being studied with regards to their therapeutic value in a clinical setting. However, similar to the older approaches, Skin-Schwann cell mediated repair may prove to be sub-optimal due to the extended time required to generate these cells (~6 weeks435). Moreover, incomplete conversion of precursor cells to a Schwann cell fate could result in tumorigenic conditions, given the progenitor-like potential of SKPs443. Nonetheless, these studies show that Schwann cells can be successfully acquired through multiple sources other than a patient’s own nerves.

1.4.6.2 Introduction to cellular reprogramming

Besides generating Schwann cells from stem cell populations, more recent studies have focused on methods to convert somatic cells such as fibroblasts to cell types of choice via a process known as cellular reprogramming or lineage conversion444. A commonly used somatic cell

39 source amenable to lineage conversion are fibroblasts, which have been obtained from human foreskin445,446 or from 3mm skin punch biopsies of the dermis447,448. In general, cellular reprogramming is achieved by the overexpression of lineage-specifying transcription factors and/or by the addition of extrinsic factors such as growth factors or small molecule antagonists or agonists, which specify alternative cell identities, repress the identity of the starting cell type, and remove epigenetic barriers to alter cell state449,450. Initial reprogramming studies focused on first generating iPSCs from fibroblasts by overexpressing the Yamanaka factors (c-Myc- Klf4-Sox2-Oct4)317, and then differentiating iPSCs into the cell type of interest. However, the need to transit through a pluripotent stem cell state poses several problems, including the possibility of generating partially reprogrammed cells that may proliferate and/or differentiate erroneously451. Moreover, iPSCs are self-renewing stem cells, and are thus potentially tumorigenic. Consequently, more recent studies have looked at direct cellular reprogramming as an alternative approach452-454.

1.4.6.3 Direct cellular reprogramming

Examples of direct cellular reprogramming include the trans-differentiation of fibroblasts to a neuronal fate454-456 or oligodendrocyte fate457,458, both involving the use of select sets of developmental transcription factors. Notably, oligodendrocytes, the myelinating glial cells of the CNS have been induced by using an 8 transcription factor cocktail including the core transcription factors Sox10-Olig2-Nkx6.2457, or with a triple transcription factor approach (Sox10-Olig2-Zfp536)458, and finally, by overexpression of Sox10 alone459. In general, transcription factor cocktails include at least one pioneer factor, which aids lineage conversion by binding to and opening sites of closed chromatin. For example, reprogramming of fibroblasts to a neuronal fate have involved Ascl1, which acts as a pioneer factor and accesses closed chromatin sites in fibroblasts460,461. Ascl1 then recruits Brn2 and Myt1 to these sites to aid reprogramming, with Myt1 repressing alternative cell fates and Brn2 activating neuronal lineage genes460,462. Similarly, Olig2 is incorporated in oligodendrocyte reprogramming protocols for its function as a pioneer factor463 similar to Sox family genes464,465.

Cell permeable chemical small molecules have also been used to promote desired, alternative cell fates in fibroblasts455,466,467. For example, reprogramming of human fibroblasts into glutamatergic neurons used a chemical cocktail consisting of a Wnt agonist/GSK3 inhibitor (CHIR99021), TGF inhibitor (RepSox), HDAC inhibitor (Valproic acid;VPA), and PKA agonist (Forskolin)466. Similarly, mouse fibroblasts have been converted into glutamatergic

40 neurons using a similar cocktail involving CHIR99021, Forskolin, ISX9 (promotes neurogenesis) and I-BET151 (bromodomain inhibitor)466. Thus, small molecule and transcription-factor based approaches can be used independently or in conjunction for efficient cell reprogramming.

1.4.6.4 Reprogramming of somatic cells to a Schwann cell fate

Reprogramming of adult fibroblasts to a Schwann cell fate has been achieved using either transcription factors or small molecules (Fig. 1.7)468-472. The first success came from the conversion of fibroblasts first to a NCC state by mis-expressing Sox10 and culturing cells with VPA (HDAC inhibitor), 5-Azacytidine (inhibitor of DNA methylation) and CHIR99021 (Wnt agonist)468. Induced NCCs were then differentiated into mature Schwann cells by providing appropriate environmental cues, including ciliary neurotrophic factor, NRG1, bFGF, and cAMP 468. More recently, Schwann cells were generated from adult human fibroblasts by overexpressing Sox10 and Egr2 in presence of Schwann cell specific cues including forskolin, NRG1, bFGF, platelet derived growth factor470,471.

Investigators have also successfully employed small molecules (VPA and Compound B - undefined) to reprogram fibroblasts to a transient neural precursor state469. These proliferative intermediate cells were then treated with Noggin (BMP inhibitor), SB431542 (TGF inhibitor) and CP21 (GSK3 inhibitor) and differentiated into Schwann cells by culturing in a neural differentiation media enriched with B27 and N2 supplements, BDNF, Glial cell-derived neurotrophic factor, ascorbic acid, and dibutyryl-cAMP. A more recent study has used a cocktail of chemical small molecules with other stimulatory cues to generate Schwann cells from fibroblasts, using all-trans retinoic acid, bFGF, forskolin, platelet-derived growth factor- AA, and NRG1472.

Notably, early reprogramming protocols required ~6 weeks for the first appearance of mature Schwann cell markers468,469, however newer studies observed onset of Schwann cell markers within 9-21 days of treatment470-472. Moreover, these protocols have reported generation of a heterogeneous population of Schwann cells representing an early and late developmental stage471,472, and additional maturation protocols are required to obtain mature myelinating Schwann cells. More recently, induced SCPs were generated from human fibroblasts by misexpressing pluripotency factors (OCT4, SOX2, KLF4, MYCL1, LIN28 and p53 shRNA) in fibroblasts using episomal vectors and using induction medium enriched with NRG1 and a host

41 of small molecules (e.g. Wnt agonist/GSK3 inhibitor CT-99021, DNA methyltransferase inhibitor RG108, adenosine receptor agonist 5′-(N-Ethylcarboxamido) adenosine)473. Induced SCPs of high purity were generated in nearly 3 weeks, and an additional week was required for differentiation of the induced SCPs into mature Schwann cells, significantly reducing the time- period473.

While in vitro cellular reprogramming has dominated reprogramming studies in mammalian systems, we expect ‘directed in vivo cellular reprogramming’ to be the future in cell fate conversion studies. In vivo cellular reprogramming aims towards the conversion of resident target cells in their endogenous environment into the cell type of interest, thereby allowing easy integration of induced cells into the existing cellular micro-environment474. In vivo reprogramming was carried out in rodent models as early as 2008 for the generation of  cells from pancreatic exocrine cells using a transcription factor (Ngn3, Pdx1, Mafa) induction approach475. However the crucial shortcomings of in vivo reprogramming (e.g. optimum targeting of host cell, conversion efficiency, as well as conversion into altered cell fates leading to a tumor-like state476) has limited their widespread use in therapeutic model systems. Despite such inadequacies, numerous studies have demonstrated potency in the use of in vivo reprogramming to generate the cell type of interest477-480. Thus, it remains to be seen whether in vivo reprogramming can ultimately be employed as a therapeutic approach for PNI repair.

We have come a long way towards understanding the molecular and cellular events that underlie Schwann cell development and function; to isolate and purify Schwann cells from multiple sources, and to convert other cell types to Schwann cells. With the emerging cases of Schwann cell transplantation in clinical uses and trials, Schwann cells will certainly remain an important focus for future research. One of the challenges will be to efficiently acquire Schwann cells from accessible sources in a shorter time frame. The identification of Schwann cell subpopulations that exhibit a repair phenotype, and exploitation of this identity to develop biomarker-based selection processes, will further enrich the repair capacity of Schwann cells within the injured nerve to improve recovery. Taken together, the realization of autologous Schwann cell therapy for effective clinical use is anticipated to be within our grasp.

Figure 1.7. Direct cellular reprogramming approach to generate induced Schwann cells.

Skin biopsy samples isolated from individuals can be used for propagating dermal fibroblast cultures in vitro. Fibroblasts are amenable to reprogramming using a 1) transcription factor (TF) mediated conversion approach, 2) small molecule approach, as well as a combinatorial conversion approach. Using appropriate protocols for directed reprogramming, induced Schwann cells can be generated which can then be used in a clinical setting to aid nerve repair and recovery post peripheral nerve injury. Current TF mediated conversion approaches for generation of Schwann cells from fibroblasts include misexpression of Sox10 or Sox10+Egr2. Small molecules used in varying combinations in reprogramming protocols include Valproic acid (VPA), 5-azacytidine (5-Aza), SB431542, Noggin, All trans retinoic acid (ATRA), CHIR/CT/CP99021, RG108, and 5′-(N-ethylcarboxamido) adenosine.

1.5 Hypothesis and specific aims of the thesis

Chapter 2: I hypothesized that EVs play an important part in ODG growth and contribute towards their slow growing phenotype. Thus, in Chapter 2 of my thesis I investigated the role of ODG derived EVs in mediating glioma-neural cell (heterotypic) interactions as well as glioma-glioma cell (homotypic) interactions.

Chapter 3: I hypothesized that complex and dynamic transcriptional networks underlie temporal identity transitions in the Schwann cell lineage. Thus, in Chapter 3 I aimed to conduct a comprehensive survey of Schwann cell development in the murine model focusing on identification of markers, with a focus on transcription factors, that define each stage of Schwann cell development.

Chapter 4: I hypothesized that among the early genes expressed during Schwann cell development, the ets domain transcription factor Etv5 plays an important role in Schwann cell specification and development. Thus, in Chapter 4 I used a loss-of-function approach to determine if Schwann cell development and repair can be regulated via Etv5 transcription factor, expressed upstream in the Schwann cell lineage.

Chapter 5: I hypothesized that lineage conversion of somatic cells into ‘repair-like’ Schwann cells may require re-activation of a precise array of embryonic and repair specific glial-lineage genes. Thus, I applied the concept of direct lineage conversion involving the misexpression of a cocktail of embryonic and repair SC-specific TFs in fibroblasts for inducing a repair SC-like identity using a triple transcription factor expression vector approach. My initial study of the molecular basis of Schwann cell development (Chapter 3) assisted me in identifying key genes essential for peripheral glial development that I applied towards my reprogramming approach that was geared towards generating a readily accessible pool of Schwann cells for therapeutic purposes.

SMPD3-mediated extracellular vesicle biogenesis inhibits oligodendroglioma tumor growth

This chapter has been submitted for publication:

Balakrishnan A, Adnani L, Chinchalongporn V, Vasan L, Prokopchuk O, Chen M, El- Sehemy A, Olender T, Touahri Y, Sheikh T, Islam R, Sujanthan S, Zinyk D, Comanita L, Kan B, Fleming T, Leong HS, Morshead CM, Brand M, Wallace V, Chan J, Schuurmans C. SMPD3-mediated extracellular vesicle biogenesis inhibits oligodendroglioma tumor growth. (bioRxiv pre-print server; MS ID#: BIORXIV/2020/202200).

2.1 Abstract

Oligodendrogliomas are lower-grade, slow-growing gliomas that are ultimately fatal. Although driver mutations are known, the mechanisms underlying their signature slow growth rates are poorly understood. We found evidence for intra-tumoral interactions between neoplastic and non-neoplastic cells in oligodendroglioma tissues. To further study these cell interactions, we used two patient-derived oligodendroglioma cell lines of lower and higher aggressivity. Both oligodendroglioma cell lines released extracellular vesicles that had cytotoxic effects on non- neoplastic and neoplastic cells, but each had distinct vesicular proteomes. Consistent with extracellular vesicles mediating growth inhibitory effects in oligodendrogliomas, higher expression levels of several extracellular vesicle biogenesis genes (SMPD3, TSG101, STAM1) correlates with longer survival in oligodendroglioma patients. Furthermore, SMPD3 overexpression slows oligodendroglioma cell growth in culture. Conversely, SMPD3 knockdown enhances oligodendroglioma proliferation in vitro, in murine xenografts, and in human cerebral organoid co-cultures. Oligodendroglioma-derived extracellular vesicles thus mediate tumor cell microenvironmental interactions that contribute to low aggressivity.

2.2 Introduction

Gliomas are a heterogenous group of primary glial brain tumors that are composed of a mixture of neoplastic glial cells ('tumor cells') and non-neoplastic ‘stromal cells’ that include non- neoplastic glia, neurons, and a variety of other inflammatory and vascular cells481,482. Gliomas are classified based on their molecular and histological features1,2. The most aggressive and most frequently studied is glioblastoma multiforme (GBM), a high-grade astrocytoma with 5- year survival rates of less than 5%483. In contrast, lower-grade gliomas (World Health Organization–WHO stage II/III tumors), which include oligodendroglioma (ODG), are slower growing tumors, with median survival times between 9-14 years4,28. Glioma subtypes have a unique constellation of mutations, that in ODG typically includes isocitrate dehydrogenase 1 (IDH1) or IDH2 mutations, chromosomal 1p/19q co-deletion, and mutation of Capicua (CIC), a transcriptional repressor, in the retained 19q allele6,7,22,201. While tumor cells carry driver mutations that are responsible for the cells neoplastic transformation, non-neoplastic cells in the tumor microenvironment form part of the ecosystem that sustains tumor cell proliferation and growth484. Understanding how glioma cells interact with cells in the microenvironment is thus essential to understanding disease progression.

Extracellular vesicles (EVs) are important mediators of intercellular communication. EVs package lipids, protein, DNA and RNA in a lipid bilayer that protects cargo from degradation in the extracellular space, and facilitates membrane fusion and delivery of bioactive material to neighboring cells40. Small EVs (sEVs; 40-200nm), also known as exosomes52, are generated via different biogenic enzymes, including endosomal sorting complex related transport (ESCRT)-dependent and -independent pathways79. ESCRT-independent biogenesis is mediated by sphingomyelin phosphodiesterase 3 (SMPD3), encoding neutral sphingomyelinase 2 (nSMase2), which produces ceramide required for exosome generation and budding93. EVs are secreted by most if not all cells in the brain, including oligodendrocytes138, astrocytes485, neurons486, as well as cancerous cells487. There are now several examples of glioma cells interacting with cells in the tumor niche via EVs175,176. Strikingly, while EVs from higher-grade gliomas are generally growth promoting, the few studies conducted with ODG suggest that ODG-derived EVs may be cytotoxic203,204.

Here, we further queried how vesicular factors contribute to the slow growth properties of ODG, using patient tissue samples, patient-derived ODG cell lines22, and publicly available molecular and clinical glioma data. We found evidence that ODG cells exert growth inhibitory effects non-cell autonomously in part through the secretion of cytotoxic vesicular factors. Accordingly, higher expression levels of several EV biogenic genes (SMPD3, TSG101, STAM1) correlates positively with longer survival in low-grade glioma patients. We focused on SMPD3 and found that it is a critical regulator of cytotoxic EV biogenesis in ODG. SMPD3 negatively regulates ODG growth in cultured cells, in mouse xenografts, and in human cerebral organoid co-cultures. We conclude that ODG cells negatively modulate tumor growth in part through the secretion of vesicular factors that have homotypic and heterotypic cytotoxic effects on cells in the tumor niche.

2.3 Methods

2.3.1 Patient-derived tumor tissues and cells and study approval

IDH mutant ODG patient biopsies were obtained from the pathology archives at the Calgary Laboratory services and Clark Smith Brain Tumor Bank at the University of Calgary22. Samples were formalin fixed and embedded in paraffin. Approval for use was obtained from Calgary Laboratory Services and the Calgary Health Region Ethics Board (University of Calgary Conjoint Health Research Ethics Board to JAC (HREB #2875 and #24993). BT088

47 and BT054 cell lines were collected under approved protocols from the Health Research Ethics Board of Alberta to JAC (HREBA.CC-16-0762 and HREBBA.CC-16-0154). Culture of BT088 and BT054 cells was approved by the Sunnybrook Research Ethics Board (REB) to CS (PIN: 301-2017). Culture of human ESCs received approval from the Canadian Institutes of Health Research (CIHR) Stem Cell Oversight Committee (SCOC) to CS and was approved by the Sunnybrook REB (PIN: 264-2018).

2.3.2 The Cancer Genome Atlas (TCGA) survey

Kaplan Meier plot was generated from lowest and highest quartiles of all low-grade glioma patients for all assessed genes. Normal brain cortex (GTEX), GBM, low-grade glioma, IDH mutant astrocytoma, and IDH mutant ODG (TCGA) datasets were downloaded from UCSC’s Xena Browser (https://xenabrowser.net/) and analyzed to compare SMPD3 expression levels, which was correlated with overall patient survival.

2.3.3 Animals

CD1 outbred mice used for neurosphere assays were purchased from Charles River Laboratories (Senneville, QC). Embryos were staged using the morning of the vaginal plug as embryonic day (E) 0.5. Sex was not considered due to the difficulty in assigning sex at embryonic stages. Animal procedures were approved by the Sunnybrook Research Institute Animal Care Committee (20-606) in compliance with the Guidelines of the Canadian Council of Animal Care.

2.3.4 BT088 and BT054 cell culture

BT088 and BT054 cells22 were grown in Human Neurocult proliferation media (Stem Cell Tech; # 05751) containing human epidermal growth factor (hEGF, 20 ng/ml, Peprotech; AF- 100-15), human fibroblast growth factor 2 (hFGF2, 20 ng/ml, Wisent; # 511-126-QU), heparin (2 µg/ml; Stem Cell Tech, #07980), and Antibiotic-Antimycotic solution (0.1%; Wisent; # 05751). Tumorspheres were dissociated and passaged using Accutase (Stem Cell Tech; # 07920). Neurosphere media was DMEM (Wisent; #319-005-CL):F12 (ThermoFisher Scientific; #31765-035) (3:1), with hFGF2 (20ng/ml), hEGF (20ng/ml), B27 supplement (2%; ThermoFisher Scientific; #17504044), Antibiotic-Antimycotic solution (0.1%), Cyclopamine (0.5 μM; Sigma; #C4116), and Heparin (2 µg/ml). BT088 cells were cultured either as adherent cells on Poly-D-Lysine:Laminin coated tissue culture plates, or in suspension on non-coated 48 flasks/plates. For BT088 tumorsphere assays, dissociated BT088 cells were seeded at 8000 cells/ml in BT088 media (FM), or in BT088 FM+EVs. BT088 cells grew undisturbed for 5 DIV, and tumorspheres were then imaged.

2.3.5 Small molecule inhibitors

Small molecules were added as follows: GW4869 (stock concentration=1 mM): FM was supplemented with 1 or 18 µl/ml DMSO (control) or 1 µl/ml of 1 µM or 18 µl of 18 µM GW4869 (SigmaAldrich). Foretinib (stock concentration=0.1 mM): FM was supplemented with 5 µl/ml DMSO (control) or 0.5, 1, 2, 4 or 5 µl/ml Foretinib (GSK1363089) at 50, 100, 200, 400, or 500 nM.

2.3.6 Mouse NSC isolation and culture

Dorsal telencephalons (cortices) from E13.5 embryos were dissected and dissociated in 0.125% trypsin (Wisent) at 37°C for 10 mins. Dissociated cells were seeded at 8000 cells/ml in neurosphere media (FM), or in BT088 or BT054 CM, CM-EV, or FM+EVs. NSCs grew undisturbed for 7 DIV, and neurospheres were then imaged.

2.3.7 AnnexinV- PI Apoptosis assay

NSCs were cultured in FM, BT088 CM, CM-EV, and FM+EV for 3 DIV. Cells were stained with FITC-labeled Annexin V and PI using the FITC Annexin V Apoptosis Detection Kit with PI as per the manufacturer’s instructions (BD Biosciences; 640914) for 20 min at 25˚C and analyzed by flow cytometry.

2.3.8 Conditioned media

1x106 BT088 or BT054 cells were seeded in 11 ml of fresh media and CM was collected after 24 hrs. CM was centrifuged at 300 x g for 5 min to remove cells, followed by 2000 x g for 10 min to remove cellular debris and 10,000 x g for 30 min to remove protein aggregates and smaller debris. CM was then sequentially ultra-centrifuged at 100,000 x g at 4 ͦ C for 2 hrs in a Beckman Coulter Optima L-100 XP Ultracentrifuge using an SW41-Ti rotor and polycarbonate centrifugation tubes (Beckman Coulter; #331372). The EV pellet, and supernatant (CM-EV) were used as indicated. The EV pellet was rinsed with Phosphate Buffered Saline (PBS)

(ThermoFisher Scientific; # 14190144) and centrifuged at 100,000 x g, 4 ͦ C for 1 hr and resuspended in 50 µl PBS prior to use.

2.3.9 Incucyte live cell imaging

Cell growth and death were monitored using an Incucyte S3 Live cell imaging system (Essen BioScience). BT088 and BT054 cell growth rates were monitored using NucLight Rapid Red Reagent for nuclear staining of the cells (as per manufacturer’s instructions; IncuCyte; #4717). Cells were suspended in media supplemented with 2 µl/ml of the reagent prior to cell seeding. Cell growth was quantified by monitoring area covered by RFP+ objects. All other cell growth studies monitored total phase area confluence or area covered by GFP+ cells. For cell death assays, the media was supplemented with 0.25 µl/ml of Cytotox dye/well (as per manufacturer’s instructions; red #4632 and green #4633) prior to cell seeding. Phase contrast and Red/Green fluorescent imaging was carried out at designated intervals (cell growth studies: every 12/24 hrs; cell death studies: every 4/12 hrs), and at 10x/20x magnification. A minimum of 9 images were taken per well at each time point. Quantification of cell proliferation and cell death was performed using the analyser algorithm built in the Incucyte application. Mean values of total phase area (normalized to day 0) ratio (U) and Cytotox+ object area (normalized to day 0) ratio (V) were plotted, comparing between days 0 to 7.

2.3.10 CO-BT088 co-cultures

Feeder-free H1 hESCs (WiCell) were cultured on Matrigel in TeSR™-E8™ kit for hESC/hiPSC maintenance (StemCell Tech; #05990). hESCs were used to generate COs using media included in the STEMdiff Cerebral Organoid Kit (StemCell Tech; #08570) and STEMdiff Cerebral Organoid Maturation Kit (StemCell Tech; #08571), with some modifications. Briefly, hESCs were plated in 96-well round-bottom ultra-low attachment plates at 9,000 cells/well in embryoid body (EB) seeding medium. Dual SMAD inhibitors (2μM Dorsomorphin; StemCell Tech; #72102, and 2 μM A83-01; StemCell Tech; #72022) until day 5. Newly formed EBs were transferred to 24-well plates containing StemCell Tech CO induction medium. On day 9, EBs with optically translucent edges were embedded in matrigel and deposited into 6-well ultra-low adherent plate with StemCell Tech expansion medium. From day 5 to day 13, media was supplemented with 1 μM CHIR-99021 (StemCell Tech; #72052) and 1 μM SB-431542 (StemCell Tech; #72232) to support formation of well-defined, polarized neuroepithelia-like structures. On day 13, embedded EBs exhibiting expanded

50 neuroepithelia as budding surfaces were transferred to a 12-well spinning bioreactor (Spin Omega488) containing maturation medium in a 37°C incubator. For BT088 co-culture, on day 30, COs were individually transferred to a 24-well plate containing Neurocult NS-A proliferation media (# 05751, StemCell Tech) with freshly added hFGF2 (20 ng/ml), hEGF (20ng/ml), and heparin (2μg/mL). Subsequently, 10,000 GFP+ BT088 cells were added to each well. Plates were incubated for 24 hrs without agitation and on the next day tumor-bearing COs were washed with PBS once and maintained in maturation media on an orbital shaker at 37°C for 7 more days. On day 8, COs were fixed in 4% paraformaldehyde (PFA) for 45 min, transferred into 30% sucrose overnight, snap frozen in OCT for cryosectioning.

2.3.11 Pellet assay

BT088 cells expressing GFP and Cre were mixed with NIH-3T3 cells transfected with BFP- loxP-dsRed in a 5:1 ratio. Cells were centrifuged, and the pellet was placed on a cell culture membrane. The membrane was floated on DMEM media in a 6-chamber dish. The cells were incubated at 37˚C for 3DIV after which the membrane was embedded in a cryopreservative, OCT, and frozen gradually on dry ice. 10µm thick sections were obtained by sectioning the OCT block.

2.3.12 Density gradient Ultracentrifugation

OptiPrepTM (Iodixanol 60% stock solution; StemCellTech; #07820) was diluted with a homogenization solution (0.25M sucrose in 10mM Tris HCl pH 7.5) to generate a discontinuous density gradient of 40% (2.5ml), 20% (2.5ml), 10% (2.5 ml), and 5% (2 ml). The solutions were carefully pipetted in an ultracentrifuge tube and left undisturbed for >1hr. EVs (filtered through a 0.2 µm filter and resuspended in 500 µl PBS) isolated from BT088 CM by sequential centrifugation after the first 100,000 x g spin were loaded onto the gradient. Samples were centrifuged at 100,000 x g for 18 hrs. Post centrifugation, 1 mL fractions were pipetted out carefully from the top. Fractions were mixed with PBS and centrifuged at 100,000 x g for 4 hrs. EV pellets were resuspended in 50 µl PBS.

2.3.13 Scanning Electron Microscopy

BT088 and BT054 cells were cultured on 13 mm coverslips (EM Biosciences) precoated with Poly-O-Lysine – Laminin in a 24 well plate. The samples were then fixed (2% glutaraldehyde in 0.1M sodium cacodylate buffer, pH 7.3; for >2 hrs), rinsed and dehydrated. Samples were 51 mounted on stubs, gold sputter-coated, and imaged with FEI/Philips XL30 scanning electron microscope at 15 kV. Imaging was performed in the Nanoscale Biomedical Imaging Facility, SickKids Research Institute.

2.3.14 Transmission Electron Microscopy

Before grids were prepared, carbon-coated Cu400 Transmission electron microscopy grids were glow discharged for 30 seconds (Pelco EasiGlow, Ted Pella Inc.). Then 4 µL of BT088 exosome solution was applied to the grid for 60 seconds before wicking away excess solution. The grid was washed three times with 4 µL of distilled water. The grid was stained with 4 µL of 2% uranyl acetate solution for 30 seconds. Excess uranyl acetate solution was wicked away. The grids were air-dried. Imaging was performed on a Thermo Fisher Scientific Talos L120C TEM operated at 120 kV using a LaB6 filament. Imaging was performed in the Microscopy Imaging Laboratory, University of Toronto.

2.3.15 Nanosight tracking analysis (NTA) and Nano-flow cytometry

NTA was performed using the Malvern NanoSight NS300 at the Structural & Biophysical Core Facility, University of Toronto. EV pellets were collected by sequential centrifugation, resuspended in 200 µl PBS, diluted 1:50 and passed through the Nanosight chamber. NTA data acquisition settings were as follows: camera level 13, acquisition time 3 × 30 seconds with detection threshold 12. Data was analyzed with the NTA 3.2 Dev Build 3.2.16 software. For nanoscale flow cytometry, EVs (1 µl in 18µl sterile water) were incubated with CD9 antibody (1 µl; Santa Cruz; #sc9148) for 30 mins at room temperature. Post incubation, EVs were stained with Alexa Fluor 647 Far red secondary antibody (1 µl of 1:20 antibody solution; Invitrogen) for 20 mins. Stained EVs were diluted in 500 µl sterile water and quantified on the Nanoscale Flow Cytometer (Apogee Flow Systems Inc). Representative scatterplot of BT088 EVs plotted for 638-Red (detecting Alexa 647 bound CD9+ particles) and Long angle light scatter (LALS; for size distribution) EVs were defined as size events greater than 100 nm.

2.3.16 Molecular Cloning

SMPD3 Gain of function: Conditional overexpression of SMPD3 was achieved with a doxycycline inducible lentiviral vector pCW57-MCS1-P2A-MCS2 (GFP) (Addgene plasmid # 80924). We cloned SMPD3 (SMPD3 Human Tagged ORF Clone, Origene Cat#: RG218441, RefSeq- NM_018667.2) and, after the P2A site, Luc2 (pcDNA3.1(+)/Luc2=tdT (Addgene 52 plasmid # 32904)) into this vector. SMPD3 Loss of function: For all in vitro SMPD3 knockdown experiments, we used NSMase2 (SMPD3) Human shRNA Plasmid Kit (Origene, Locus ID 55512), which included what we termed shSMPD3 variants A-D and shScr. For xenograft experiments, we used a piggyBac shRNA (GFP) construct (SB #PBSI505A-1) as the backbone. The vector backbone was linearized with BamH1 and EcoR1 and the following annealed oligonucleotides were cloned into the site: shSmpd3:5’pGATCCCCCTCATCTTCCCATGTTACTTCAAGAGAGTAACATGGGAAG ATGAGGGACGCGTG3’(sense), and 5’pAATTCACGCGTCCCTCATCTTCCCATGTTACTCTCTTGAAGTAACATGGGAAG ATGAGGGG 3’ (antisense); and for shScrambled:

5’pGATCCATTCACTTATCCGCCTCTCCTTCAAGAGAGGAGAGGCGGATAAGTGAA TCTCGAGG3’(sense), and 5’pGAATTCCTCGAGATTCACTTATCCGCCTCTCCTCTCTTGAAGGAGAGGCGGAT AAGTGAATG -3’ (antisense). The shSmpd3 construct targeted mouse Smpd3 sequence, such that the mouse shRNA target sequence has a 3 bp mismatch upon alignment with human SMPD3 sequence, but it effectively knocked down human SMPD3 (Fig. 2.12A).

2.3.17 Transduction and transfection

SMPD3 Gain of function: To generate lentiviral particles for SMPD3-P2A-Luc2 GFP and Luc2 GFP the lentiviral vector was packaged in LentiX HEK293T cells using the packaging plasmids psPAX2 (Addgene Plasmid #12260) and pMD2.G (Addgene Plasmid #12259). The resulting lentiviral particles were concentrated by ultra-centrifugation and used to transduce BT088 cells. To induce SMPD3-Luc2 or Luc2 only (control) expression, transduced GFP+ cells were treated with 2µg/mL doxycycline hyclate (D9891, Sigma). SMPD3 Loss of function: BT088 cells were transduced with 4 SMPD3 human shRNA lentiviral particles (A,B,C,D) and Lenti shRNA Scrambled control particles (pGFP-c-shLenti; TL301492V; Origene). Transduced GFP+ cells (shSMPD3-GFP variants B,D and shScrambled-GFP) used in CO co-culture studies were sorted using the BD FACS ARIA III. GFP+ cells were selected and collected in Neurocult proliferation media and plated. For in vitro tumorsphere assays and generating BT088 cells for xenograft studies, BT088 cells were nucleofected with each shSmpd3/shScrambled vector mixed with Super PiggyBac Transposase expression vector (SBI, Cat#PB210PA-1) in 1: 3, PiggyBac: transposase ratio. 2x106 dissociated BT088 cells were suspended in 20µl of P3

53 reagent with the DNA mix (final concentration=12µg). Nucleofection was performed using the 4D nucleofector (Lonza) in nucleofector strips using the program CZ167. For shSmpd3 knockdown experiments, HEK 293 LentiX cells were transfected with Super PiggyBac Transposase expression vector, hSMPD3-GFP construct, and shScr/shSMPD3 (internal control)/ or mouse shSmpd3-GFP construct using Lipofectamine 3000 (ThermoFisher).

2.3.18 Tumor xenografts shSmpd3-GFP and shScr-GFP BT088 were injected into the right cerebral hemispheres of 8- 10-week old female NOD scid Gamma mice (NSG; NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ; N=8) (Princess Margaret animal breeding facility). Coordinates for implantation were AP-1.0, ML 2.0, and DV 3.0. The xenografted cells developed into tumors, for which the mice were monitored daily. Upon development of terminal symptoms, mice were sacrificed at relative end points. Two control mice did not show any terminal symptoms post engraftment and were humanely sacrificed after 180 days.

2.3.19 Western blotting

Cell or EV pellets were lysed in lysis buffer with protease (1X protease inhibitor complete, 1 mM phenylmethylsulfonyl fluoride) and phosphatase (50 mM NaF, 1 mM NaOV) inhibitors. 3 µg of cell lysate and 10 µg of EV lysate was run on 10% SDS-PAGE gels for Western blot analysis. Primary antibodies included: Flotillin1 (Cell signalling; #3253), CD9 (Santa Crutz; #sc9148), nSMase2 (Abcam; #ab85017), Cetp (Abcam; #ab2726), Alix (Cell Signaling; #2171S), GM130 (BD Biosciences; #610822), Calnexin (Abcam; #ab22595), Calreticulin (Abcam; #ab2907), Pex5 (Novus Biologicals; #NBP1-87185), VDAC (Cell Signaling; #4661), and Actin (Abcam; #ab8227). Densitometries were calculated using ImageJ. The average values of normalized expression levels were plotted.

2.3.20 Tissue Processing and Immunostaining

ODG patient tumor sections were stained using the Opal™ 4-Color Manual IHC Kit (NEL820001KT), as per manufacturer’s instructions. In case of tumor xenograft studies, upon reaching end-time mice were sacrificed and perfused with PBS and 4% PFA solution. Mouse brains were collected and fixed at room temperature for 30mins. The samples were then rinsed with PBS and immersed in 20% sucrose overnight at 4°C and embedded in OCT. 10µM cryosections were collected on Superfrost Plus slides (Fisher). For immunostaining, sections 54 were washed and permeabilized in Phosphate Buffered Saline with 0.1% Triton X100 (PBT) followed by blocking with 10% normal horse serum/PBT (blocking solution) for 1 hr at room temperature. Sections were then incubated with primary antibodies overnight at 4C. Sections were washed with PBT and incubated with secondary antibody for 1 hr at room temperature. Sections were then washed and counterstained with DAPI diluted in PBT at room temperature. Sections were washed in PBS and mounted with coverslips using AquaPolymount (Polysciences). Primary antibodies included: pERK (Cell Signaling;#CS4370S); IDH1R132H (Dianova;#DIA-H09) ; Ki67 (Abcam;#ab16667); GFAP (Millipore;#mab360) ; Olig2 (Abcam;#ab109816), Sox2 (Abcam;#ab97959) HNA (Millipore;#MAB1281), nSMase2 (Abcam;#ab85017), CD63 (Abcam; #ab59479); turboGFP (Origene;#TA150041), BrdU (Abcam;#ab6326). Secondary antibodies included: Alexa 568 donkey anti-rabbit, Alexa 488 donkey anti-rabbit, Alexa 488 donkey anti-mouse (all from Invitrogen), and were diluted in PBT.

2.3.21 Mass Spectrometry

Mass spectrometry was performed on BT088 and BT054 EVs isolated by sequential ultracentrifugation at the SPARC Biocentre-Mass Spectrometry facility at SickKids Research Institute. Mass spectrometry analysis information is provided in Table 2.4. Scaffold data analysis was performed by applying NCBI annotations to all proteins, removing proteins that matched the search term “keratin” and which did not have “Homo sapiens” under the taxonomy heading. Protein filtering thresholds were set at 99.0%, with a minimum number of 2 peptides and a peptide threshold of 95%. For analysing the EV samples, Cytoscape program and ClueGO plugin were used.

2.3.22 Image analysis

Images were captured with a Leica DM IL LED or DMRXA2 optical microscope using LasX software. ImageJ software was used for image analysis. For ODG patient tumor analysis, Single channel TIFF images with pERK/Ki-67/OLIG2 staining from all sample sets were transformed to binary format with mean intensity as the selecting parameter. A fixed minimum and maximum threshold value were determined for each set of images to ensure correct thresholding of pERK/Ki-67/OLIG2 staining. Images were analyzed by adjusting the size filter option to count cells with pERK or Ki-67 staining giving the total number of pERK+/Ki-67+/ OLIG2+ cells in each data set. Single channel images with pERK/Ki-67/OLIG2 staining were

55 merged with IDH1 R132H (IDHm) single channel images. Cells displaying co-localization of pERK/Ki-67/OLIG2 with IDHm was manually counted. For primary neurosphere and tumorsphere assays, neurosphere/tumorsphere sizes were measured using the ruler measurement tool in ImageJ. For cerebral organoid-tumor co-culture assessment, zone-wise cumulative GFP intensity and total number of Sox2+ cells were analyzed using ImageJ. The freeline tool was used to mark the periphery of the organoid. Seven zones (width=150 pixels or 50µm) were constructed mapping the shape of each organoid, spanning from organoid periphery (zone1) to the core (zone7). Cumulative GFP intensity and Sox2+ cell counts were assessed per zone.

2.3.23 Statistical analysis

A minimum of three biological replicates were carried out for all assays. Statistical analysis and graphs were generated using GraphPad Prism 6 software. Student’s t-test was used when comparing two groups, while One-way ANOVA with TUKEY post corrections were used when comparing groups of more than two. All data expressed as mean value ± standard error of the mean (s.e.m.). In all experiments, a p value <0.05 was taken as statistically significant, *p < 0.05, **p < 0.01, ***p < 0.005, and ****p < 0.001.

2.4 Results

2.4.1 Oligodendroglioma cells exert non-cell autonomous effects within the tumor niche

To determine whether lower-grade glioma cells interact non-cell autonomously with neighboring non-neoplastic cells (heterotypic interactions) in the tumor microenvironment, we analyzed surgical resection specimens from five IDH-mutant, 1p/19q-codeleted ODG patients (Table 2.1). We imaged tumor sections in the more confluent tumor core and in peritumoral infiltrating regions, where we reasoned that tumor cells could influence non-neoplastic reactive cells. To identify tumor cells, sections were immunolabeled with an antibody to IDHmR132H (hereafter IDHm)489. As expected, co-staining with IDHm and OLIG2, an oligodendrocyte lineage marker highly expressed in ODG490, showed high co-localization of IDHm and OLIG2 in all samples (Fig. 2.1A-E).

Table 2.1. Summary of ODG patient tumors analyzed.

Five ODG patient tumors used for the study.

Figure 2.1. IDH mutant oligodendroglioma cells display non-cell autonomous activation of proliferation.

A-J. ODG patient sections co-immunostained with IDHm and OLIG2 (A-D'') or Ki-67 (F-I''). Blue is DAPI counterstain. Yellow arrows indicate IDHm- cells that are OLIG2+ or Ki-67+. Percentage of OLIG2+/Ki-67+ IDHm- and OLIG2+/Ki-67+ IDHm+ cells (E,J) in ODG patients. Regions marked with white dotted boxes in A,C,G,I,M,O were digitally magnified (4 times) and presented in B-B'',D-D'',G-G'',I-I''. K. Tumor microenvironment is composed of tumor cells and non-neoplastic neural cells (astrocytes, neurons, microglia, oligodendrocyte, endothelial cells). Vesicular and/or non-vesicular factors (represented as small spheres) may mediate inter-cellular communication in the tumor microenvironment. Bars represent means ± s.e.m.. Scale bars: 200 µm.

We then asked whether IDHm-negative non-neoplastic cells in the core and peritumoral regions were influenced by the tumor by assessing proliferation, a rare event in non-cancerous adult brains491. Co-labelling with IDHm and Ki-67, a pan-proliferative marker, showed that a surprisingly large fraction of Ki-67+ cells (83-88%) were IDHm-negative (Fig. 2.1F-J). Proliferation of non-neoplastic cells in the tumor mass suggested that tumor cells could secrete growth factors or other signals that modulate the behavior of neighboring cells. Consistent with the presence of non-cell autonomous signaling in the microenvironment, we detected MEK/ERK pathway activation (as evidenced by pERK staining) in IDHm-negative cells (Fig. 2.2A-E, Table A.2.1).

Although correlative in nature, these findings are consistent with the idea that IDH mutant ODG cells may communicate with neighboring stromal cells (neural, vascular, inflammatory) in the tumor niche in a non-cell autonomous fashion (Fig. 2.1K).

Figure 2.2. IDH mutant oligodendroglioma cells display non-cell autonomous activation of receptor tyrosine kinase signaling

A-E. ODG patient biopsies co-immunostained with IDHm and pERK (A-D''). Blue is DAPI counterstain. Yellow arrows indicate IDHm- cells that are pERK+. Percentage of pERK + IDHm- and pERK + IDHm+ cells (E) in oligodendroglioma patients. Regions marked with white dotted boxes in A,C were digitally magnified (4 times) and presented in B-B'',D-D''. Scale bars: 200 µm.

2.4.2 ODG secretomes have distinct bioactive effects but both include cytotoxic EVs

To mediate contact-independent communication, tumor cells and neighboring non-neoplastic stromal cells secrete a variety of bioactive molecules (e.g. growth factors, cytokines, interleukins) that are enriched in the tumor microenvironment492. In vitro, bioactive factors are found in tumor conditioned media (CM), as soluble factors and/or in EVs40. To further understand how ODG cells interact with cells in the tumor microenvironment, we used two different patient-derived cell lines established from IDH-mutant, 1p/19q co-deleted anaplastic ODGs, termed BT088 and BT054 cells22. Both cell lines can self-renew and grow in vitro, but only BT088 cells, isolated from a higher grade III tumor, retain the ability to form tumors after xenografting in immunocompromised mice22. Consistent with reported differences in tumor growth in vivo, using real-time live cell imaging, we found that BT088 cells have a 1.76-fold shorter doubling time than BT054 cells when grown in vitro (Fig. 2.3A, Table A.2.1).

Figure 2.3. Patient-derived oligodendroglioma cells secrete soluble and EV-enclosed bioactive factors.

A. Live cell imaging of BT088 and BT054 cell growth, monitoring expansion of NucLight Rapid Red-stained cells over 10 days in vitro. B. Experimental setup to assess the bioactivity of the BT088 and BT054 secretome. C-J. E12.5 NSCs grown in fresh media (FM; C), BT088 conditioned media (CM; D), conditioned media without EVs (CM-EV; E), and fresh media with EVs (FM+EV; F) for 7 DIV (G). Quantitation of BT088 neurosphere number (H), neurosphere size (I), and live cell number (J). K-N. E12.5 NSCs grown in FM, BT054 CM, CM-EV, and FM+EVs for 7 DIV (K). Quantitation of BT054 neurosphere number (L), neurosphere size (M), and live cell number (N). Bars represent means ± s.e.m.. *, p < 0.05; **, p < 0.01; ***, p < 0.005. Scale bars: 100 µm in (C-F).

To ask whether BT088 and BT054 ODG cells might influence the growth/survival of neighboring cells in the tumor niche, we tested the bioactivity of secreted factors on two cell types - embryonic neural stem cells (eNSCs) and ODG cells themselves (Fig. 2.3B). While eNSCs are not present in the adult tumor niche, they were chosen as a surrogate cell type as their growth is highly sensitive to external cues493. Embryonic day (E) 12.5 NSCs isolated from the cortex were first plated at clonal density in fresh media (FM), or CM collected from the two ODG cell lines. After 7 days in vitro (DIV), neurosphere number (measure of activated NSCs), neurosphere size (aggregate measure of proliferation and apoptosis) and live cell number (measure of cell survival) were quantitated (Fig. 2.3C-F). Compared to FM, NSCs grown in BT088-CM gave rise to more neurospheres that were larger in size (Fig. 2.3C-I, all counts in Table A.2.1). However, the larger spheres had markedly fewer live cells (Fig. 2.3J, Table A.2.1). In contrast, BT054 CM did not alter neurosphere number, and it had an inhibitory effect on both neurosphere size and live cell number compared to FM (Fig. 2.3K-N). Thus, while BT088 CM initially promotes NSC revival from quiescence and NSC proliferation, cell progeny have decreased viability. In contrast, BT054 CM in aggregate is not pro-proliferative, but like BT088 cells, BT054 cells also secrete cytotoxic factors.

The bioactive effects of ODG CM could be due to soluble and/or EV-enclosed factors. To dissect this activity further, we repeated the neurosphere assay using CM in which EVs were removed by sequential centrifugation (CM-EV). E12.5 NSCs grown in BT088 CM-EV formed the same number of neurospheres as NSCs grown in BT088 CM, however, there was a further increase in neurosphere size, and a striking increase in live cell number (Fig. 2.3C-J; Table A.2.1). Similarly, E12.5 NSCs grown in BT054 CM-EV formed the same number of neurospheres as NSCs grown in BT054 CM, and there was an increase in neurosphere size, although live cell number was not altered (Fig. 2.3K-N; Table A.2.1). Thus, BT088 CM contains pro-proliferative factors that are largely soluble, and cytotoxic factors that are mainly vesicular. In contrast, BT054 CM contains soluble pro-proliferative factors, the activity of which is masked by cytotoxic factors that are both soluble and vesicular.

Next, to directly assess EV bioactivity, EVs were added to FM (FM+EV). BT088 EVs added to FM reduced neurosphere number, neurosphere size and live cell number compared to FM (Fig. 2.3C-J, Table A.2.1). In contrast, BT054 EVs added to FM did not alter neurosphere number, while neurosphere size and live cell number were both reduced compared to FM (Fig. 2.3K-N, Table A.2.1). Thus, BT088 and BT054 cells both produce EVs that carry cytotoxic

62 factors.

Finally, to determine whether the cytotoxic nature of BT088 EVs translated to other cell types, we directly examined BT088 EV effects on BT088 cell growth (i.e. homotypic activity) using a similar tumorsphere assay (Fig. 2.3B). Notably, BT088 EVs had a similar cytotoxic effect on BT088 cells themselves, reducing tumorsphere size and live cell number after 5 DIV (Fig. 2.4A-E), suggesting that ODG cells may limit their own growth/survival by EV-mediated autocrine/paracrine effects.

Taken together, these studies confirm that while ODG CM from both BT088 and BT054 cell lines contains pro-proliferative soluble factors, in addition to EV-enclosed factors that induce cell death, which in BT054 cells, overshadows the pro-proliferative effects. These differences in the bioactive nature of the BT054 and BT088 secretomes may help to explain their different growth rates (Fig. 2.3A) and tumorigenicity22.

Figure 2.4. Patient-derived oligodendroglioma cells secrete EV-enclosed bioactive factors which induce cytotoxic effects.

A-E. BT088 cells grown in BT088 FM (A,B) and BT088 FM+EVs (A,C) for 5 DIV. Quantitation of tumorsphere size (D), and live cell number (E). Bars represent means ± s.e.m.. *, p < 0.05; **, p < 0.01; ***, p < 0.005. Scale bars: 50 µm in (B,C).

2.4.3 ODG EVs induce proliferation followed by apoptosis

The ability of BT088 CM to induce the formation of more and larger neurospheres suggested that pro-proliferative effects could occur first, followed by cell death. To interrogate the timeline of events, we performed live cell imaging using phase area confluence as a surrogate

63 measure of cell number (Fig. 2.5A,B). NSCs plated in either BT088 CM or CM-EVs increased in number after 5 DIV at a faster rate than NSCs grown in FM or FM+EV (Fig. 2.5B,D-O). However, after 5 DIV, while NSCs continued to grow exponentially in CM-EV, cell number started to decline in CM (Fig. 2.5B). There was also a decrease in cell number in FM+EV compared to FM, which supports the notion that EVs carry cytotoxic factors (Fig. 2.5B).

We further confirmed the cytotoxic nature of BT088 EVs by performing live cell imaging after incorporating a fluorescent cytotoxic dye (‘CytoTox’) into NSC cultures. A decline in cellular health increases cell permeability thereby permitting entry and intercalation of CytoTox into affected cells. After 3 DIV, NSCs exposed to CM and FM+EVs incorporated CytoTox at a higher rate than cells grown in FM or CM-EV, validating the cytotoxic nature of the EVs (Fig. 2.5C-O). Finally, to determine whether BT088 EVs exerted a pro-apoptotic effect on NSCs, AnnexinV/Propidium Iodide (PI) co-labelling of late apoptotic cells was assessed via flow cytometry494 (Fig. 2.5P). After 3 DIV, NSCs grown in FM included 3.8% AnnexinV+PI+ late apoptotic cells, whereas nearly 52.7% of NSCs grown in FM+EV were double+ apoptotic cells (Fig. 2.5Q). Similarly, NSCs cultured for 3 DIV in BT088 CM included 55.6% AnnexinV+PI+ late apoptotic cells, and this number declined to 35% when grown in CM-EV (Fig. 2.5Q).

In summary, BT088 CM has an initial pro-proliferative effect that is followed a few days later by an increase in EV-mediated cell death, suggesting that ODG EVs induce apoptosis of NSC progeny.

Figure 2.5. Oligodendroglioma EVs induce apoptosis in neural stem cells.

A-C. Live cell imaging of NSCs grown in FM, BT088 CM, CM-EV, FM+EV (A), monitoring growth (B) and dying cells labeled with CytoTox dye (C). D-O. Representative images of CytoTox incorporation in NSCs exposed to FM, CM, CM-EV, FM+EV at day 0,5,7 from one 65 independent experiment. P. NSCs were cultured in FM, BT088 CM, CM-EV, and FM+EV for 3 DIV. Cells were stained with FITC-labeled Annexin V and propidium iodide (PI) and analyzed by flow cytometry. A typical flow cytometry dot plot from one representative experiment is given for each sample. Cell death was determined by quadrant gating: Q1 (necrotic cells); Q2 (late apoptotic cells); Q3 (early apoptotic cells); Q4 (live cells). Q. Mean values of Annexin V+ PI+ cell percentages (Q2) were plotted. Bars represent means ± s.e.m.. *, p < 0.05; **, p < 0.01; ***, p < 0.005. Scale bars: 200 µm.

2.4.4 ODG cells secrete bioactive EVs mainly in the 40-200 nm exosome size range

To determine whether the isolated ‘EVs’ tested above were truly vesicular, we characterized their morphology and protein content. First, we used scanning electron microscopy to assess cellular topography, revealing the presence of budding vesicles on the surfaces of both BT054 and BT088 cells (Fig. 2.6A,B). Next, to assess vesicular size and composition, EVs were isolated from CM using sequential centrifugation40. Using transmission electron microscopy, we demonstrated that sedimented particles isolated from ODG cells had a lipid bilayer- enclosed nanoparticle morphology characteristic of EVs (BT088 EVs shown, Fig. 2.6C,C'). To calculate the size and number of isolated EVs, we used nanosight tracking analysis (NTA). BT054 and BT088 cells both produced EVs that were largely within the exosome size range (i.e. 40-200 nm44, Fig. 2.6,E, Table A.2.1), but BT088 cells produced twice as many EV particles (Table A.2.1). Finally, as a first assessment of molecular phenotype of ODG EVs, using nanoscale-flow cytometry we revealed that 13.7±2.0% of BT088 EVs expressed CD9, a marker of a subset of EVs495 (Fig. 2.6G).

Figure 2.6. Bioactive exosomes are secreted from oligodendroglioma cell lines.

A,B. Scanning EM of BT054 (A) and BT088 (B) cells. Red arrows mark vesicles. C,C'. Transmission electron microscopy of BT088 EVs. Higher magnification image shown in (C'). D-G. Nanosight tracking analysis of BT054 and BT088 EVs isolated by sequential ultracentrifugation (D,E) and density gradient ultracentrifugation (fraction 4; F). CD9+ EVs analyzed using nanoscale flow cytometry (G). H-J. Western blots of BT088 whole-cell lysates (CL) and EV lysates obtained by sequential ultracentrifugation, analyzed for the expression of EV markers (Alix, CD9, Cetp, Flotillin1) (H) and Calreticulin, Calnexin (ER), GM130 (Golgi body), Vdac (mitochondria), and Pex5 (Peroxisomes) (I). Density gradient ultracentrifugation of BT088 EVs, with 8 fractions analyzed by Western blots for Alix, CD9, Cetp, Calnexin, and Vdac. K-O. Schematic of pellet assay mixing BT088 cells expressing Cre-GFP at 5:1 ratio 67 with NIH-3T3 cells expressing a dual BFP-loxP-dsRed reporter (K). Analysis of GFP (L,M), BFP (L,N) and dsRed (L,O) expression in pellets after 3 DIV. White arrows mark dsRed+BFP- GFP- cells. White arrowheads mark dsRed+BFP+GFP- cells. Scale bars: 2 µm in (A,B); 500 nm in (C); 50 nm in (C').

To further assess the molecular nature of ODG EVs, we focused on BT088 EVs, as they were more numerous. Using Western blotting, we compared protein content in the crude cell lysate (CL) and in the EV pellet. Compared to CL, EVs were enriched in EV-associated markers496, including Alix, CD9 and Cetp (Fig. 2.6H). Flotillin1, a common EV marker, was also detected in EVs, but was also present at high levels in CL due to its association with the cell membrane497 (Fig. 2.4H). Finally, we confirmed the relative purity of the EV preparations by probing for proteins associated with other organelles, including the endoplasmic reticulum (Calreticulin, Calnexin), mitochondria (Vdac), Golgi bodies (GM130) and peroxisomes (Pex5), all of which were detected in the CL, as expected, and at negligible levels in EVs (Fig. 2.6I).

As the sequential centrifugation method of EV isolation is sedimentation-based, it can also isolate non-vesicular components44. To determine whether non-vesicular material was in the ‘EV’ pellet, we used density gradient ultracentrifugation for size fractionation. BT088-EVs isolated by sequential ultracentrifugation were loaded onto a discontinuous OptiprepTM gradient. After ultracentrifugation, eight fractions were collected from the density gradient, and analyzed by Western blotting and NTA. EV associated markers (Alix, CD9, Cetp) were detected in fractions 4 and 5, with a density of 1.08-1.09 g/cm3, while negative EV markers (Vdac, Calnexin) were absent in these fractions (Fig. 2.6J). Using NTA, we confirmed that EV particles in layer 4 were predominantly in the exosome size range (Fig. 2.6F). Of note, Cetp was also detected in low density fractions 1-3 (0.95-1.07 g/cm3), which contain non-vesicular low-density lipoproteins to which Cetp associates498. Thus, the BT088-EV pellet collected by ultracentrifugation includes EVs mainly in the exosome size range but also some microvesicles and non-vesicular material, as previously shown for other cell types44.

EVs deliver molecular content to neighboring cells through membrane fusion, a delivery method not available to non-vesicular material. To determine whether BT088 cells secrete bioactive cargo in EVs that can be taken up by and influence recipient cells, we used a Cre

68 recombinase-based fluorescent reporter assay previously used to report EV-mediated Cre transfer499. BT088 ‘donor’ cells were stably transduced to express Cre recombinase and GFP, while NIH-3T3 ‘recipient’ cells were transduced with a dual BFP-dsRed Cre reporter499. BFP is expressed in recipient cells prior to Cre (mRNA/protein) transfer from donor-to-recipient cells, which undergo Cre mediated excision of a STOP cassette to then allow expression of dsRed. We aggregated BT088-GFP-Cre donor cells with NIH-3T3 BFP-dsRed recipient cells (Fig. 2.6K). After 3 DIV, we detected RFP+ cells that were GFP-, indicating that they did not arise from cell fusion (Fig. 2.6L-O).

BT088 cells thus secrete bioactive cargo to recipient cells, and from our in vitro assays, we suggest that this includes EVs and other soluble factors.

2.4.5 Proteomic profiling reveals ODG cells secrete distinct vesicular cargo

To characterize the potential bioactive content of ODG EVs, we carried out proteomic analyses of BT088 and BT054 vesiculomes by LC-MS/MS (Fig. 2.7A). Given the known differences in biologic behavior of the two cell lines, it was unsurprising that the protein content differed between the two EV pools, with 390 proteins detected in BT088 EVs and 186 proteins in BT054 EVs, of which only 72 proteins were common between the two vesiculomes (minimum 2 out of 3 individual replicates; Fig. 2.7A'). Among the shared proteins, both BT088 and BT054 EVs contained commonly associated EV proteins (e.g. Alix, CD63; Table A.2.244,500), validating the vesicular nature of our preparations.

BT088 and BT054 vesiculomes were enriched in proteins associated with several biological processes, including metabolic, developmental, immune system, growth, and cell death (Fig. 2.7B). In both BT088 and BT054 vesiculomes, several identified proteins were assigned to cell death pathways, some of which were higher in BT088 EVs (e.g. RHOA, RPL11; Fig. 2.7C,D), others elevated in the BT054 vesiculome (e.g. HTRA1 (Fig. 2.7E), and some common to EVs from both cell types (e.g. CLU; Fig. 2.7F). Notably, of the proteins associated with proliferation, such as PKM, HSPB1, HSP90AA1, and HSP90AB1 (Fig. 2.7G-J), all were detected at higher levels in the BT088 versus BT054 vesiculome, in keeping with differences in the growth rates of these two tumor cell lines22.

Figure 2.7. Proteomic profiling reveals distinct BT088 and BT054 vesiculomes and identifies VEGF signaling as a targetable growth-promoting pathway.

A. GO term enrichment network of all proteins enriched in BT088 (green) and BT054 EVs (beige). A'. Commonly expressed proteins in BT088 and BT054 vesiculomes. B. GO terms relating to biological processes plotted against percentage of proteins identified in BT088 vs BT054 EVs. C-L. Relative levels of proteins enriched in BT088 versus BT054 cell derived EVs, comparing RHOA (C), RPL11 (D), HTRA1 (E), CLU(F), PKM (G), HSPB1 (H),

HSP90AA1 (I), HSP90AB1 (J), SRI (K), and MFGE8 (L). M,N. Live cell imaging of BT088 cells treated with Foretinib (50-500nM) and DMSO control. Growth was assessed by calculating total phase area (M) and cell death by calculating total Cytotox+ area (N), both normalized to day 0. Bars represent means ± s.e.m..

While our data suggested that cytotoxic signals largely overshadow pro-proliferative signals in ODG-derived EVs, BT088 and BT054 tumor cells continue to grow and thrive. To understand how ODG cell proliferation is supported, we queried the proteome for growth regulators, as proteins are loaded into EVs non-specifically, and serve as an informative readout of cellular state44. Two proteins detected in both BT088 and BT054 EVs stood out - SRI (Sorcin) and MFGE8 (Lactadherin) (Fig. 2.7K,L). SRI expression is elevated in several cancers, including breast, hepatocellular, and gastric501, and it is required to maintain VEGF expression, which is involved in angiogenesis, tumor invasion, and metastasis501,502. Similarly, MFGE8 promotes VEGF-dependent neovascularization in endothelial cells503. In line with the viewpoint that EVs are non-specific carriers of cellular protein, and that EV content implies signaling status44, we inferred that VEGF signaling may be activated in these two ODG cell lines. To assess the importance of VEGF signaling in supporting ODG tumor growth, we treated BT088 cells with Foretinib, a VEGF-receptor inhibitor504, at five different concentrations (50nM-500nM) and cell growth was assessed over five days using live cell imaging. Strikingly, in all conditions, BT088 cell growth resembled BT054 growth (Fig. 2.3C), with increase in BT088-cell doubling times (from 4-9 fold) at all concentrations, compared to BT088 cells grown in DMSO (Fig. 2.7M). Foretinib treatment of BT088 cells also increased cytotoxicity at all assessed doses and inhibited growth of BT088 cells (Fig. 2.7N).

BT088 and BT054 vesiculomes are thus distinct from each other, revealing the heterogeneity of ODG intracellular signaling and the potential pathways that may trigger cell death in neighboring cells.

2.4.6 Lower SMPD3 expression is associated with poor prognosis in oligodendroglioma patients

Given that BT088 and BT054 EVs both had cytotoxic effects, we queried whether EV generation was itself associated with ODG biologic behavior. We analyzed data from the cancer genome atlas (TCGA) database to compare expression of genes involved in ESCRT- dependent (TSG101, STAM1)79, and ESCRT-independent (SMPD3)93 exosome synthesis in a 71 lower grade glioma patient cohort, correlating gene expression with patient survival. Strikingly, lower TSG101, STAM1, and SMPD3 expression levels correlated with shorter survival times compared to patients with higher expression of these genes (Fig. 2.8A-C; p<0.0001, Table A.2.1), consistent with the notion that exosome synthesis may play a role in limiting lower grade glioma growth. Given the better prognosis of lower grade glioma patients with high SMPD3 expression, we focused further studies on SMPD3, which encodes for nSMase2, the major sphingomyelinase in the brain505.

Figure 2.8. Poor prognosis in oligodendroglioma patients associated with low SMPD3.

A-C. Kaplan–Meier survival curves for correlation between SMPD3 (A), TSG101 (B), STAM1 (C) expression and survival in low-grade glioma patients (Log-rank (Mantel-Cox) test; p<0.0001). D. SMPD3 levels are higher in normal brain cortex compared to low-grade gliomas and GBMs. E. SMPD3 levels are higher in ODG tumors versus astrocytoma. p<0.0001. F. Kaplan–Meier survival curves for correlation between SMPD3 expression and survival of astrocytoma patients. Log-rank (Mantel-Cox) test; χ2=7.601; p=0.0058; high SMPD3>15.81 (n=66); low SMPD3<15.0 (n=65). G. Kaplan–Meier survival curves for correlation between SMPD3 expression and survival of ODG patients. Log-rank (Mantel-Cox) test; χ2=15.27; p<0.0001; high SMPD3≥16.96 (n=66); low SMPD3≤16.28 (n=66). ****, p < 0.0001.

We next used the TCGA data to compare SMPD3 expression levels in normal brain versus lower- and higher-grade gliomas. SMPD3 expression levels were highest in normal brain, followed by lower-grade glioma and higher-grade GBM (Fig. 2.8D). We further dissected the lower-grade glioma dataset into IDH-mutant astrocytoma and IDH-mutant ODG and found that astrocytoma patients, which have a shorter survival time (median survival=9-13 years28,29), presented with lower SMPD3 transcript levels compared to ODG patients, which have a longer survival time (median survival= 12-14 years26,27; Fig. 2.8E; p<0.0001) – suggesting that higher SMPD3 levels may reduce tumor growth. Furthermore, within each disease type, patients with lower SMPD3 expression in their tumors had shorter survival times compared to those with higher SMPD3 expression (Fig. 2.8F, p=0.0058 for astrocytoma; and Fig. 2.8G, p<0.0001 for ODG; Table A.2.1).

These data, together with our findings from our EV cell culture experiments (Fig. 2.3-2.5), suggest that higher SMPD3 expression levels may limit ODG (and potentially also IDH-mutant astrocytoma) growth to a certain degree by triggering an increase in cytotoxic exosome production.

2.4.7 SMPD3 inhibits ODG cell proliferation in vitro

To assess whether SMPD3 might play a role in regulating ODG growth, we first confirmed that the encoded protein, nSMase2, was expressed in BT088 and BT054 ODG cells, demonstrating co-immunolabelling with Sox10, a marker of glial-like tumor cells506 (Fig. 2.9A,B). Previous studies indicated that SMPD3 overexpression enhances EV production507, which our data suggested should inhibit ODG cell growth. To test this prediction, we used a doxycycline- inducible lentiviral system to generate stable SMPD3-GFP and control-GFP BT088 cell lines (Fig. 2.10A,A'). We confirmed that 72hrs post-doxycycline exposure, nSMase2 expression was induced (Fig. 2.10B,B'). We also used nanoscale flow cytometry to confirm that SMPD3 overexpression increased the release of CD9+ EVs from SMPD3-GFP cells, revealing a 2.3-2.9 fold increase compared to control cells, and SMPD3-GFP cells not treated with doxycycline (Fig. 2.10C,C', Table A.2.1).

We next monitored the growth of control-GFP and SMPD3-GFP BT088 cells using live cell fluorescent imaging of adherent cells (Fig. 2.10D-G'). As a proxy measure of cell growth, we monitored the cumulative area covered by GFP+ cells normalized to seeding day (day 0) (Fig. 2.10H). Control-GFP cells and SMPD3-GFP cells not exposed to doxycycline grew

73 exponentially, with doubling times in the range of 78-84 hrs. In contrast, doxycycline dosed SMPD3-GFP cells failed to proliferate in the first 1.5 days, after which there was a decline in the cumulative GFP area, suggestive of cell death (Fig. 2.10H, Table A.2.1). We also monitored the growth of Control-GFP and SMPD3-GFP BT088 cells grown in 3D suspension cultures, which can replicate more of the cell-cell interactions observed in vivo and found that SMPD3 overexpression similarly inhibits tumor cell growth (Fig. 2.9C-G). Finally, to assess the potential cytotoxicity of elevated SMPD3 expression levels, we incorporated the cytotoxic dye ‘CytoTox’ in the culture media, and assessed the cumulative area covered by CytoTox+ cells. Doxycycline dosed SMPD3-GFP cells increased CytoTox accumulation compared to all control conditions (Fig. 2.10I-M, Table A.2.1).

Taken together with the cytotoxic nature of BT088 EVs on NSC and BT088 cell growth, we conclude that elevated SMPD3 expression likely inhibits BT088 cell growth through the production of EVs, although we cannot rule out the potential for SMPD3 having additional cell autonomous effects.

Figure 2.9. High SMPD3 expression inhibits oligodendroglioma cell growth in a suspension system.

A,B. Immunostaining of BT088 (A) and BT054 (B) cells with Sox10 and nSMase2. Merged images of Sox10 (green) and nSMase2 (red) with DAPI as counterstain in blue. C-G. To 74 monitor effects of SMPD3 overexpression in cells in a suspension system, GFP expressing cells (GFP-BT088 cells: Ctrl/Ctrl-GFP; SMPD3-GFP-BT088 cells: SMPD3/ SMPD3-GFP) were seeded on uncoated plates and growth was monitored by monitoring total area covered by GFP+ cells. Representative images of cell growth at day 14, from one independent experiment, is presented (C-F). Mean values of total GFP+ area (normalized to day 0) ratio were plotted comparing between days 0 to 14 (G). Bars represent means ± s.e.m.. Scale bars: 200µm.

Figure 2.10. High SMPD3 expression inhibits oligodendroglioma cell growth.

A-C'. Generation of doxycycline-inducible system to assess the effects of SMPD3 on BT088 cell growth, showing Ctrl-GFP (A) and SMPD3-GFP (B) cells. Western blotting showing relative nSMase2 expression normalized to Actin (B,B'). Analysis of CD9+ EV particle number using nano-flow cytometry in Ctrl-GFP and SMPD3-GFP cells (C,C'). D-H. Live cell imaging to monitor growth of Ctrl-GFP and SMPD3-GFP cells, showing growth with and without doxycycline at day 0 (D-G) and day 5 (D'-G'). Quantitation of cell growth (normalized to day 0) (H). I-M. Live cell imaging of Cytotox dye labeled Ctrl-GFP and SMPD3-GFP cells to monitor induction of cell death, showing growth without (I-L) and with (I'-L') doxycycline at days 1, 2, 3, and 5. Quantitation of Cytotox+ cells (normalized to day 0) (M). Bars represent means ± s.e.m.. *, p < 0.05; **, p < 0.01. p-values for all points in H and M in Table S1. Scale bars: 200 µm.

2.4.8 Knockdown of SMPD3 promotes proliferation of ODG cells in vitro

We next asked the converse question, which is whether reduced SMPD3 expression levels might promote ODG cell growth. We used an shRNA approach to stably knockdown SMPD3 expression in BT088 cells. Lentiviral constructs that expressed GFP and one of four shSMPD3 variants (A-D) or an shScrambled (shScr) control sequence were transduced into BT088 cells to create five GFP-tagged BT088 cell lines (Fig. 2.11A). To assess the efficacy of shSMPD3 knockdown, we performed Western blotting, revealing a >50% decrease in nSMase2 levels for shSMPD3 B-D variants, but not variant A (Fig. 2.11B,B'). shSMPD3 variants B-D also all reduced the number of secreted CD9+ EV particles compared to shScr control (Fig. 2.11C,C', Table A.2.1). In addition, isolated EV pellets from the CM (from shSMPD3 B-D variants) had lower levels of Alix, an EV marker (Fig. 2.12A).

To assess the effects of SMPD3 knockdown on BT088 cell growth, we used live cell fluorescent imaging of adherent cell cultures (Fig. 2.11D-H). Compared to shScr control cells, BT088 cells expressing shSMPD3 variants B-D cells grew more rapidly after 5 DIV, with doubling times of ~160-192hrs compared to ~800hrs for shScr control (Fig. 2.11H, Table A.2.1). The growth stimulatory properties of SMPD3 knockdown was similarly observed when BT088 cells were grown in 3D suspension cultures for 5 DIV (Fig. 2.12B-F). To further validate these data, we also performed manual tumorsphere counts and diameter measurements of BT088 cells expressing control shScr versus shSMPD3-B (Fig. 2.12G-K). After 10 DIV, SMPD3

76 knockdown increased tumorsphere number (Fig. 2.12I), tumorsphere size (Fig. 2.12J), and live cell number (Fig. 2.12K), replicating the live cell imaging data.

The increase in cell number observed after shSMPD3 knockdown could be due to an increase in proliferation and/or a decrease in cell death. Proliferation must have increased to produce more cells after shSMPD3 knockdown and, accordingly, all BT088 cell lines incorporated BrdU, indicative of active DNA synthesis in S-phase of the cell cycle (Fig. 2.12L-O). However, to determine whether SMPD3 knockdown also reduced normal levels of cell death in BT088 cells, we assessed ‘CytoTox’ dye incorporation in BT088 cells expressing shScr versus shSMPD3 knockdown constructs, revealing more dye incorporation in the first days of culture by the shSMPD3 lines (Fig. 2.11I-M). Thus, knockdown of SMPD3 expression in BT088 cells not only enhances cell proliferation, but also reduces cell death rates to increase tumor cell number.

As a final independent measure of the effects of SMPD3 knockdown on BT088 cell growth, we used a pharmacological approach, treating BT088 cells with GW4869, a competitive inhibitor of phosphatidylserine, a phospholipid that binds nSMase2 and is required for enzyme activation83,508. Exposure of BT088 cells to 1M GW4869 for 48h reduced CD9+ EV production by 2-fold compared to DMSO control cells (Fig. 2.12P,Q). Treatment of BT088 cells with GW4869 also increased cell proliferation using live cell imaging but only at later stages (Fig. 2.12R-T). To confirm that the knockdown of SMPD3 was stimulatory for BT088 growth, we also used a higher concentration of GW4869 (18µM), which increased tumorsphere number (Fig. 2.12U-W) and size (Fig. 2.12X) after 5 DIV.

Taken together, we have genetic and pharmacological evidence that lowering SMPD3 expression levels increases ODG cell proliferation and survival, in keeping with the reduced survival times associated with lower SMPD3 expression in ODG patients.

Figure 2.11. Low SMPD3 expression promotes oligodendroglioma cell growth.

A-C'. Use of shRNA constructs to assess the effects of SMPD3 knockdown on BT088 cell growth, depicting shScr and shSMPD3 (variant A,B,C,D) constructs (A). Western blotting showing relative nSMase2 expression normalized to Actin (B,B'). Analysis of CD9+ EV particle number using nano-flow cytometry in shScr and shSMPD3 (variant A,B,C,D) BT088 cells (C,C'). D-H. Live cell imaging to monitor growth of shScr and shSMPD3 (variant A,B,C,D) BT088 cells at day 0 (D-G) and day 5 (D'-G'). Quantitation of cell growth (normalized to day 0) (H). I-M. Live cell imaging of Cytotox dye labeled shScr and shSMPD3 (variant A,B,C,D) BT088 cells to monitor induction of cell death, showing Cytotox+ cells at

78 day 0 (I-L) and day 5 (I'-L'). Quantitation of Cytotox+ cells (normalized to day 0) (M). Bars represent means ± s.e.m.. *, p < 0.05; **, p < 0.01. p-values for all points in H and M in Table S1. Scale bars: 200 µm.

Figure 2.12. SMPD3 knockdown is growth enhancing and recapitulated by GW4869, a pharmacological inhibitor.

A. Western blot for Alix protein of shScr and shSMPD3-A,B,C,D BT088 cell-derived EVs. B- F. BT088 cells transduced with shScr and shSMPD3- B,C,D constructs were cultured for 9 days on uncoated plates. Growth was monitored by monitoring total area covered by GFP+ cells. Representative images of cell growth at day 9, is presented (B-E). Mean values of total GFP+ area normalized to day 0) ratio were plotted comparing between days 0 to 9 (F). G-K. BT088 cells transfected with shScr-GFP and shSMPD3-GFP were cultured for 10 days. Representative images of BT088 shScr-GFP (G) and shSMPD3-GFP (H) tumorspheres after 10 DIV. Quantitation of tumorsphere number (I), tumorsphere size (J), and live cell number (K). L-O. BT088 cells transduced with shScr and shSMPD3- B,C,D constructs were cultured for 48 hrs and fixed post a 30 min BrdU pulse. Merged images of fixed cells immunostained with BrdU (red), GFP (green), and DAPI counterstain in blue. P-T. BT088 cells treated with DMSO control (Ctrl) and with GW4869 (1µM) for 14 DIV. Total CD9+ EV particle determination using nanoscale flow cytometry (P,Q). Representative images of Ctrl (R,S) and GW4869 (1µM) treated cells after 14 DIV. Mean values of total phase object count (relative to day 0) ratio were plotted spanning from day 0 to day 14 (T). U-X. BT088 cells treated with Ctrl (U) and 18µM GW4869 (V) for 5 DIV. Quantitation of the number of tumorspheres (W) and the size of tumorspheres (µm) (X) generated after 5DIV. Bars represent means ± s.e.m.. *, p < 0.05; **, p < 0.01; ***, p < 0.005; Student’s t test, and One-way ANOVA when testing more than two groups. Scale bars=200µm in (B-E); 100µm in (G,H); 50µm in (L-O).

2.4.9 SMPD3 knockdown facilitates ODG growth in vivo

BT088 cells engineered to express GFP form tumors by 6 months after orthotopic xenograft into the cerebral cortices of NOD scid Gamma (NSG) mice (Fig. 2.13A). In the xenografts, the neoplastic (human) cells can be identified by the expression of human nuclear antigen (HNA) whereas mouse cells are negative for HNA (Fig. 2.14A,B). As expected, the main tumor masses were comprised predominantly of densely packed HNA+ tumor cells, while a smaller number of HNA+ tumor cells infiltrated at the periphery (Fig. 2.14A). We further confirmed that that engrafted HNA+ BT088 cells continued to express nSMase2 (Fig. 2.14A,B).

We then used this model to further evaluate the impact of SMPD3 expression on tumor growth in the context of the complex in vivo environment. We made stable SMPD3 knockdown (KD)

80 and shScr control BT088 cell lines (Fig. 2.14C-G). shScr and SMPD3-KD cells were orthotopically xenografted into NSG mice (N=8 per cohort), and animals were monitored over six months (Fig. 2.13A). shScr and SMPD3-KD BT088 survival curves were significantly different from one another (Fig. 2.13B; Mantel-Cox log rank test p=0.0074, Table A.2.1). At endpoint, all mice were confirmed to have tumors in sections consisting of hypercellular masses that disrupted the normal cortical architecture. By day 154, however, all mice xenografted with SMPD3-KD BT088 cells had been sacrificed at the humane endpoint after showing terminal symptoms, whereas two mice xenografted with shScr BT088 cells were still alive after 6 months (181 days), our experimental endpoint.

To further characterize the tumor masses associated with BT088 shScr and SMPD3-KD cells, we co-immunostained sections with HNA along with the oligodendroglial lineage marker Olig2, the astrocytic marker GFAP, the vascular marker isolectin, and the proliferative cell marker Ki-67 (Fig. 2.13C-J). Both BT088 shScr and SMPD3-KD cells co-expressed HNA and Olig2 (Fig. 2.13C,D), as expected, and many HNA+ tumor cells were Ki-67+ in both tumor types (Fig. 2.13E,F). However, the two tumor types differed as SMPD3-KD tumors appeared to be more highly vascularized than control shScr tumors, with dense isolectin staining (Fig. 2.13G,H), phenocopying the increase vascular endothelial proliferation that is associated with higher-grade gliomas in patients509. In addition, GFAP+ cells bearing morphologic features of reactive astrocytes (hypertrophy of soma and processes, resulting in coarser and enlarged processes, and more abundant cytoplasm) were more abundant in SMPD3-KD tumor margins (Fig. 2.13I,J), suggestive of enhanced reactive changes510.

SMPD3 knockdown in BT088 cells thus results in the formation of more biologically aggressive tumors that display some phenotypic features of higher-grade III ODG159.

Figure 2.13. SMPD3 knockdown facilitates oligodendroglioma growth in vivo.

A. Schematic of xenografting protocol, showing NSG mice were xenografted with shScr and SMPD3-KD BT088 cells and tumor growth was monitored. B. Kaplan–Meier survival curves associated with shScr and SMPD3 KD BT088 cell xenografts. Log-rank (Mantel-Cox) test; χ2=7.162; p=0.0074. C-J. Co-immunostaining of shScr (C,E,G,I) and SMPD3 KD (D,F,H,J) BT088 cell xenografted tumors with HNA (red) and Olig2 (green, C,D), Ki-67 (green, E,F), isolectin (green, G,H), and Gfap (green, I,J). Blue is DAPI counterstain. Insets present split 82 channel images of regions marked by a dotted box. Br, normal brain; xe, xenograft. Scale bars: 50µm in (C-J).

Figure 2.14. Confirming SMPD3 knockdown of xenografted oligodendroglioma cells.

A,B. BT088 GFP tumor xenograft sections immunostained with nSMase2 (red) and Human Nuclear Antigen (HNA; green), and DAPI as counterstain in blue (A). Region marked in white

83 box in (A) is magnified (4X) and presented in (B). C. Western blot for whole HEK cell lysates co-transfected with shScr or SMPD3 knockdown (KD) construct and SMPD3 construct. Expression of nSMase2 was assessed for knockdown in SMPD3 expression. Three biological replicates of each sample set were loaded in varying amounts (5X/1X) to show efficiency of knockdown. D-G. Representative images of BT088 cells transfected with shScr (D,D'), and SMPD3 KD (E-E') cultured for 10DIV (BF, Brightfield). Quantitation of GFP+ tumorsphere numbers (F) and sphere diameter (µm) (G) after 10 DIV. H. hESC cerebral organoid generation protocol. 30day cerebral organoids were co-cultured with BT088 cells (shScr, shSMPD3-B/D) for 7 DIV. Bars represent means ± s.e.m.. Scale bars: 100 µm.

2.4.10 SMPD3 knockdown increases ODG invasiveness and growth in human cerebral organoid

Tumor xenografting in immunocompromised mice is a powerful method to examine how genetic or pharmacological manipulations impact tumor burden, but species-specific features of tumor growth cannot be examined. To test whether SMPD3-KD BT088 cells grew faster in a human context, we used cerebral organoids (COs) generated from human embryonic stem cells511,512. Human COs have now been used in a few studies for brain tumor modelling and provide an excellent readout of tumor infiltration and growth progression513. To generate 30- day old COs, we followed a modified Lancaster protocol (Fig. 2.14H, 2.15A)488. After 30 DIV, COs were either cultured alone (Fig. 2.15B-B''') or together with BT088 cells for an additional 7 DIV (Fig. 2.15A,C-E'''). To assess SMPD3 knockdown effects, we co-cultured 30 DIV COs with stable BT088 cell lines engineered to express GFP and shScr (Fig. 2.15A,C-C'''), shSMPD3-B (Fig. 2.15A,D-D''') or shSMPD3-D (Fig. 2.15A,E-E''') (knockdown validated; Fig. 2.11).

Figure 2.15. SMPD3 knockdown facilitates oligodendroglioma growth in human cerebral organoids.

A-A'. Schematic of cerebral organoid (CO)- BT088 co-culture assay, depicting 30-day old hESC-derived COs co-cultured with BT088 cells (shScr, shSMPD3-B/D) for 7 DIV (A). Quantitation method, depicting division of each CO into 7 zones spanning organoid periphery to the core (0-350 µm) (A'). B-E'''. Immunostaining of COs grown alone or in CO-BT088 co- cultures with SOX2 (red, B-E, white, B'-E') and turbo GFP (t-GFP; green, B-E, white, B''-E''). Blue is DAPI counterstain (B-E, white, B'''-E'''). F. Percentage of SOX2 + cells in each of 7 zones. G. Lines of best-fit from (F) plotted for the three conditions. shScr: slope= -0.04789 ± 0.003408; shSMPD3-B: slope= -0.02941 ± 0.007366, p= 0.1234; shSMPD3-D: slope= - 0.03215 ± 0.003388, p= 0.2205 H. Percentage of t-GFP intensity per zone per section. I. Lines of best-fit from (H) plotted for shScr: slope= -0.01259 ± 0.003080; shSMPD3-B: slope= - 0.02711 ± 0.001588, p= 0.003646; shSMPD3-D: slope= -0.03269 ± 0.004513, p= 0.000118. J. Cumulative t-GFP intensity/unit area in shScr and shSMPD3-CO sections. K. Summary of major findings: SMPD3 expression in the brain is high under normal healthy conditions. ODG cells with high SMPD3 expression produce more EVs and grow slower. ODG cells with low SMPD3 expression produce less EVs and grow faster. *, p < 0.05; ***, p < 0.005. Scale bars: 100 µm.

After 7 DIV, we first examined the neural identity of co-cultured COs by performing SOX2 immunostaining, which labeled neural rosettes in all COs (Fig. 2.15B'-E'). Notably, SOX2+ rosettes tended to concentrate in the CO periphery, a positional preference that we validated by counting SOX2+ cells in seven zones from periphery to the core (zone width =50µm; Fig. 2.15A',F). The negative slope of the lines of best-fit suggested that there was a biased distribution of neural rosettes in the CO periphery, and as the slopes were similar for shScr, shSMPD3-B and -D co-cultures, the addition of ODG cells did not alter neural cell organization (Fig. 2.15G; Table A.2.1). Next, we co-stained the COs with turbo-GFP (tGFP) to label infiltrating tumor cells. tGFP+ tumor cells were detected in all CO co-cultures (N=3 for each condition; Fig. 2.15B''-E''). Strikingly, for shSMPD3-B and shSMPD3-D BT088 lines, more tGFP+ cells were detected in the periphery, where SOX2+ neural rosettes localize, compared to shScr controls (Fig. 2.15H), as revealed by the significant difference in slopes of the lines of best-fit (Fig. 2.15I; Table A.2.1). Moreover, SMPD3 KD BT088 lines had more tGFP+ cells in the CO compared to shScr, indicative of a growth advantage (Fig. 2.15J, Table A.2.1).

Thus, the CO-tumor co-culture system recapitulated the more proliferative phenotype exhibited by shSMPD3 BT088 cells in vitro and in mouse xenografts.

2.5 Discussion

ODG tumors are slow growing, but the molecular mechanisms underlying their indolent growth is poorly understood. Here we report that ODG cells produce cytotoxic EVs that could allow them to interact and communicate with other tumor cells and non-neoplastic brain cells in the glioma microenvironment. Furthermore, we identify SMPD3 as a critical regulator of EV biogenesis in ODG tumors, revealing that lower SMPD3 expression levels are associated with a less favourable prognostic outcome both in patients and in murine xenografts (Fig. 2.15K). Prior studies had demonstrated that EVs isolated from G26/24 ODG cells exerted a cytotoxic effect on neurons and astrocytes in vitro, acting via Fas ligand (Fas-L) and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), respectively203,204. However, the in vivo relevance and underlying biogenetic pathways were not addressed. Moreover, it is important to acknowledge that G26/24 is a mouse glioma cell line that was originally classified as ODG based on histological resemblance514. G26/24 cells have not been shown to model the human disease genetically, and their relevance to human ODG remains unclear. Furthermore, while TRAIL and Fas-L mediate the cytotoxic effects of G26/24-EVs203,204, these factors were not present in BT088- and BT054- derived EVs, demonstrating key differences between mouse and human ODG EVs, while also highlighting the heterogenous nature of the ODG vesiculomes.

In aggregate, our data suggests that the secretion of cytotoxic EVs by ODG tumor cells may contribute to the slow growing nature of these tumors. Notably, SMPD3 also exhibits growth inhibitory functions in other cancers, mediated either by its role in EV biogenesis or by its role in ceramide production515. Early studies using a rat C6 glioma cell line found that Smpd3 produces ceramide, which promotes apoptosis by activating PKC signaling516. Homozygous deletion of Smpd3 also induces osteosarcoma in a rodent model517. In line with these studies, we report here that SMPD3 inhibits ODG cell growth, possibly through paracrine and autocrine effects. Along with SMPD3, we also found a correlation between high expression levels of ESCRT-dependent exosome biogenesis genes94 (TSG101, STAM1) and improved patient survival in low-grade glioma patients, suggesting that EV biogenesis via several pathways may play an important role in these tumors. It is important to note, however, that increasing EV biogenesis is not always favorable, and depends on the tumor type. For instance, knockdown

87 of Rab27a/b, which also blocks EV secretion, was shown to inhibit tumor growth in other brain tumor models132,198. Rab27a/b knockdown in astrocytes, which reduces EV secretion, blocks the metastasis of breast cancer cells to the brain132. Moreover, knockdown of Rab27a/b in an astrocyte-derived glioma cell line blocked glioma growth in vivo in mouse xenografts 198. A likely cause of these different effects is that EV cargo is well known to differ based on the cell of origin205. In high-grade GBM tumors secrete EVs that contain many growth factors (e.g. EGFRvIII) that could promote glioma growth and progression179,518.

While we showed that ODG EVs have cytotoxic effects, it is likely not the EVs themselves that are cytotoxic, but rather, their enclosed cargo. Even within two different ODG lines, we found distinct differences between the vesiculomes that may help to explain differences in the growth of BT088 and BT054 cells. The identification of SRI (Sorcin) and MFGE8 (Lactadherin) in ODG cell EVs is of interest, as it suggested that VEGF signaling is increased, and indeed, we could block BT088 cell growth with Foretinib in vitro. VEGF signaling is typically linked to shaping the tumor vasculature, including in GBM519, but our data is suggestive of a potential autocrine role for VEGF in supporting ODG growth, as has been suggested in GBM520 and hepatocellular carcinomas516. Future studies in which ODG cell xenografted animals are treated with Foretinib would help assess the potential of this drug as a therapeutic, since in vitro assays are not always translated in an in vivo setting, as we have highlighted. Furthermore, the analysis of ODG EV-associated factors such as CLU, previously implicated in invasion521, may help to understand how these tumor cells infiltrate the brain in vivo.

In summary, the mechanisms underlying neural-ODG cell interaction are complex with the tumor microenvironment playing a crucial role. Our studies indicate that ODG cells release EVs that serve as messengers of predominantly cytotoxic cargo to neighbouring cells. We further report that SMPD3 expression levels correlate with ODG survival, and that SMPD3 acts in part through its ability to regulate EV generation and secretion. Given that ODG EVs in the microenvironment contribute towards a slow growth phenotype, drugs that increase EV production could have therapeutic potential. While most high-throughput screens have focused on identifying small molecules that block EV secretion, which could be useful for GBM tumors522, there have also been recent screens for small molecules that induce EV secretion523. Interestingly, N-methyldopamine and norepinephrine activate nSMase2 to increase EV production in mesenchymal stem cells (MSCs) without increasing cell number, a strategy that

88 is being developed to enhance the regenerative potential of MSC EVs523. Testing whether these drugs increase ODG EV production and the associated cytotoxicity would be of interest in the future. However, one must keep in mind that it is also possible that the interaction of ODG EVs with non-neoplastic cells in the tumor niche may be reciprocated by EV secretion by brain cells that are not transformed, which together would help to create a unique microenvironment for every tumor. Careful analyses of drug efficacy would thus require 3D culture systems that mimic cell-cell interactions, such as our human-CO/tumor cell co-culture assay, or xenografts into immunocompromised mice, which remains the gold standard for pre-clinical data.

Temporal analysis of gene expression in the murine Schwann cell lineage and the acutely injured postnatal nerve

This chapter has been published:

Balakrishnan A, Stykel M, Touahri Y, Stratton JA, Biernaskie J, Schuurmans C. 2016. Temporal analysis of gene expression in the murine Schwann cell lineage and the acutely injured postnatal nerve. PLoS One. Apr 8;11(4):e0153256. doi: 10.1371/journal. pone.0153256. eCollection 2016. PMID:27058953

3.1 Abstract

Schwann cells (SCs) arise from neural crest cells (NCCs) that first give rise to SC precursors (SCPs), followed by immature SCs, pro-myelinating SCs, and finally, non-myelinating or myelinating SCs. After nerve injury, mature SCs ‘de-differentiate’, downregulating their myelination program while transiently re-activating early glial lineage genes. To better understand molecular parallels between developing and de-differentiated SCs, we characterized the expression profiles of a panel of 12 transcription factors from the onset of NCC migration through postnatal stages, as well as after acute nerve injury. Using Sox10 as a pan-glial marker in co-expression studies, the earliest transcription factors expressed in E9.0 Sox10+ NCCs were Sox9, Pax3, AP2α and Nfatc4. E10.5 Sox10+ NCCs coalescing in the dorsal root ganglia differed slightly, expressing Sox9, Pax3, AP2α and Etv5. E12.5 SCPs continued to express Sox10, Sox9, AP2α and Pax3, as well as initiating Sox2 and Egr1 expression. E14.5 immature SCs were similar to SCPs, except that they lost Pax3 expression. By E18.5, AP2α, Sox2 and Egr1 expression was turned off in the nerve, while Jun, Oct6 and Yy1 expression was initiated in pro-myelinating Sox9+/Sox10+ SCs. Early postnatal and adult SCs continued to express Sox9, Jun, Oct6 and Yy1 and initiated Nfatc4 and Egr2 expression. Notably, at all stages, expression of each marker was observed only in a subset of Sox10+ SCs, highlighting the heterogeneity of the SC pool. Following acute nerve injury, Egr1, Jun, Oct6, and Sox2 expression was upregulated, Egr2 expression was downregulated, while Sox9, Yy1, and Nfatc4 expression was maintained at similar frequencies. Notably, de-differentiated SCs in the injured nerve did not display a transcription factor profile corresponding to a specific stage in the SC lineage. Taken together, we demonstrate that uninjured and injured SCs are heterogeneous and distinct from one another, and de-differentiation recapitulates transcriptional aspects of several different embryonic stages.

3.2 Introduction

The two main glial cell types in the peripheral nervous system (PNS) are Schwann cells (SCs) and satellite glial cells. During development, SCs wrap both myelinated and non-myelinated axons, with SCs coupled with axons greater than 1µm in diameter differentiating into myelinating SCs, while SCs connected to smaller diameter axons (<1µm) become non- myelinating Remak cells. Remak cells can enclose several axons, while myelinating SCs surround and myelinate only a single axonal length. SCs play an essential role in peripheral nerve regeneration and restoration of nervous function post-injury524. One of the most remarkable features of adult SCs is their capacity to transiently revert to an undifferentiated and proliferative state following nerve injury. In doing so, ‘de-differentiated’ SCs are able to assume a variety of critical roles to support nerve repair that include modulating the immune response525,526, secretion of trophic molecules527-531, constructing growth permissive pathways for regenerating axons, and re-myelination of regenerating axons152,532,533. Previous work has collectively suggested that this de-differentiation process recapitulates SC development, such that ‘repair’ SCs resemble an embryonic immature Schwann cell (iSC) phenotype152,534. However, the transcription factors that regulate this reversion to a reparative SC state are only partially understood.

SC development has been studied extensively, leading to the identification of distinct developmental stages. SCs originate from a multipotent, migratory population of neural crest cells (NCC) progenitor cells233. In mouse, NCCs emerge from the dorsal neural tube between embryonic day (E) 9.0 until E10.5, with trunk NCCs migrating along either dorso-lateral or ventral pathways. NCCs in the ventral pathway either migrate between the somites to reach the dorsal aorta, where they give rise to sympathetic chain ganglia and the aortic plexus, or migrate through the rostral somite to give rise to the dorsal root ganglia (DRG), adrenal medulla, and SCs240. NCCs also give rise to a subset of multipotent boundary cap cells that line the dorsal root entry zone and motor exit points, and which differentiate into SCs that populate the dorsal and ventral roots254.

At approximately E12.5, migratory NCCs fated for a glial lineage give rise to SC precursors (SCPs) that populate the dorsal and ventral roots, DRG, and developing nerves257. SCPs are distinguished from migrating NCCs as they begin to express glial lineage markers, and they

92 associate directly with growing axon bundles152,535. As development proceeds, SCPs either give rise to immature SCs (iSCs), or alternately, endoneurial fibroblasts and melanocytes 268,536. iSCs first appear from E14.5 and persist until just before birth152. iSCs cluster to envelop and deposit basal lamina around developing axons. Cytoplasmic processes of the iSCs penetrate axonal bundles, dispersing larger axons to the outside of the bundle. Ultimately, iSCs in contact with large diameter axons will associate with a single axon, resulting in radial sorting282. Post radial sorting, iSCs in association with smaller diameter axon bundles become non-myelinating SCs283. iSCs associated with large diameter axons are a transient population termed pro- myelinating SCs; these are the SCs that will progress towards the myelinating stage. Pro- myelinating SCs generally appear just prior to birth, and rapidly increase in number by postnatal day (P) 1 in mice257.

Several transcription factors have been implicated in regulating the progression from NCC to a mature SC. Importantly, the transcription factors known to be upregulated after injury are presumed to be those expressed at the SCP or iSC stage152,534. However, a comprehensive analysis of the expression profiles of these transcription factors has not yet been conducted. Several of the markers have not been analyzed through development or post-injury, while the expression profiles of other markers have not been studied in an in vivo setting. Instead, parallels between developmental and de-differentiated SC factors are often identified across various injury models in an assortment of species, timeframes, ages, each employing different processing techniques and reagents. Additionally, many of these comparisons are made at the level of mRNA, or in-vitro, making it difficult to accurately appreciate the in vivo de- differentiated phenotype, and how closely it recapitulates developmental SC programs. Here, we have described the transcriptional profile of developing and de-differentiated SC in vivo in a comparable and relevant manner. We conducted an extensive spatio-temporal analysis through five key stages of embryonic mouse development (E9.0, E10.5, E12.5, E14.5, E18.5), postnatal stages P7 and P65, as well as within the P65 nerve following acute injury, when SCs are actively acquiring a reparative state.

To identify expression patterns characteristic of each developmental stage, we examined a panel of 12 transcription factors previously implicated in NCC or SC development, including Sox2269,288,300, Sox9237,247,537 and Sox10247,248,302-305,311,538, members of the SRY (sex determining region Y)-box (Sox) family of HMG-box transcription factors, Egr1/Krox24 and Egr2/Krox20254,261, EGR class zinc finger proteins, Yy1 (Yin Yang 1)539, a Gli-Kruppel zinc

93 finger protein, Oct6/Pou3f1257,286,333, a POU domain class 3 transcription factor, AP2α/Tfap2α 250, activating enhancer binding protein 2 alpha factor, Jun/c-Jun323,325, a component of the AP- 1 early response transcriptional complex, Etv5/Erm249, an ets-domain transcription factor, Pax3290,291, a paired homeodomain protein, and Nfatc4287, nuclear factor of activated T cells.

Through these studies, we identified distinct expression profiles for NCCs, NCC precursors, SCPs, iSCs, pro-myelinating SCs and mature myelinating/non-myelinating SCs, and demonstrated an underlying heterogeneity of the SC pool. Our findings also demonstrated that the de-differentiated SC is a unique SC subtype, distinct from any one developmental stage in the SC lineage. Given that this ‘repair’ SC subtype is the driving force behind efficient regeneration in uncompromised nerve injuries, methods to recapitulate this phenotype could be further investigated as a therapeutic avenue to treat chronic nerve injury and demyelinating disease.

3.3 Material and methods

3.3.1 Animals

CD1, C57/BL6, and Sox2eGFP mice540 mice were purchased from Charles River Laboratories (Senneville, QC) and Jackson Laboratory (ME, United States) and maintained in a 12 hr light cycle. Embryos were staged using the morning of the vaginal plug as embryonic day (E) 0.5. Pregnant females were housed individually after mating and euthanized using cervical dislocation for embryo collection. Adult mice used for peripheral nerve harvesting were group housed before injury, and then singly housed with enrichment post-injury. These animals were euthanized using an overdose (0.1mL) of Sodium Pentobarbital (54.7mg/mL, Ceva Sante Animale). Animal procedures were approved by the University of Calgary Animal Care Committee in compliance with the Guidelines of the Canadian Council of Animal Care.

3.3.2 Embryo processing

Whole embryos were collected for stages E9.0 and E10.5, while only bodies were collected for stages E12.5, E14.5 and E18.5. For postnatal studies, sciatic nerves were harvested from the limbs of P7/P65 pups. The embryos and nerves were fixed in 4% paraformaldehyde (PFA)/1X diethyl-pyrocarbonate (DEPC) treated phosphate-buffered saline (PBS) for ~ 4 – 20 hours at 4C. The embryos and nerves were rinsed in DEPC-PBS and transferred to 20% sucrose/1X DEPC-PBS, and were kept overnight at 4C. The embryos and nerves were then embedded

94 using O.C.T™ (Tissue-Tek®, Sakura Finetek U.S.A. Inc., Torrance, CA) and stored at -80°C. For the injury study, five days after injury (P65) mice were sacrificed by overdose of sodium pentrobarbital (i.p.; CEVA, Sante Animale). Nerves were removed and fixed in 4% PFA for two hours and subsequentally placed in 30% sucrose overnight. The next day, nerves were embedded in O.C.T.™ compound, frozen on dry ice and stored at -80°C before cutting cryosections on a Leica cryostat (Richmond Hill, ON).

3.3.3 Surgery

For crush injury, P60 mice were anesthetized using isofluorane (5% induction and 2% maintenance) and then given a preoperative subcutaneous injection of 0.1 mL (0.03 mg/mL) buprenorphine. Hindlimbs were shaved and then cleaned twice with 70% EtOH followed by 10% providine iodine. On only the right hindlimb, the sciatic nerve was crushed at mid-thigh using #10 forcep for one minute. Muscle and skin were sutured back together (7-0 Prolene and 7-0 Silk, black braided; Ethicon Inc.) and buprenorphine was administered once a day for 4 days following surgery.

3.3.4 Immunohistochemistry

Transverse 10 µm cryosections of the embryonic trunk and longitudinal 13 µm nerve sections were collected on SuperFrost™ Plus slides (Thermo Scientific). Tissue sections were washed and permeabilized in 1x PBS/0.1% TritonX (PBT) followed by blocking with 10% normal horse serum/PBT (blocking solution) for ~1 hour at room temperature. Prior to blocking, antigen epitope retrieval was carried out by heating sections in sodium citrate buffer for 20 minutes in a microwave. Sections were then allowed to cool down, washed with PBT and then blocked as above. Primary antibodies were diluted in blocking solution and were added to the sections and kept overnight at 4C. Sections were washed with PBT and secondary antibody diluted 1/500 in PBT was applied for 1hour at room temperature. Sections were then washed and counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Santa Cruz Biotechnology) diluted in PBT (1/5000) or Hoechst 33258 (Sigma-Aldrich #14530; 1:500) at room temperature. Sections were washed in PBS and mounted with coverslips using AquaPolymount (Polysciences). Primary antibodies included: rabbit anti-AP2α (Abcam ab52222; 1:200), rabbit anti-Egr1 (Aviva Systems Biology ARP32241_P050; 1:200), rabbit anti-Egr2 (Bioss Antibodies bs-8368R; 1:200, Santa Cruz Biotechnology sc-20690; 1:200), mouse anti-Egr2 (Abcam ab168771; 1:50), rabbit anti-Etv5 (Abcam ab102010; 1:300), rabbit anti-Jun (Abcam

95 ab31419; 1:300), rabbit anti-Ki67 (Vector Laboratories #VP-K451), rabbit anti-Nfatc4 (Abcam ab3447; 1:200), mouse anti-NeuN (Millipore MAB377; 1:200), goat anti-Oct6 C-20 (Santa Cruz Biotechnology sc-11661; 1:50), mouse anti-Pax3 (Developmental Studies Hybridoma Bank; 1:5), rabbit anti-Sox2 (Cell Signaling #3728; 1:200), rabbit anti-Sox9 (Millipore AB5535; 1:500), goat anti-Sox10 (Santa Cruz Biotechnology sc-17343; 1:400), rabbit anti- Sox10 (Millipore AB5727; 1:200), and rabbit anti-Yy1 (Abcam ab12132; 1:200). For Pax3, Sox9, and Oct6, serial immunostaining was conducted. Secondary antibodies included: Alexa 568 donkey anti-rabbit, Alexa 488 donkey anti-goat, Alexa 555 donkey anti-mouse, Alexa 647 donkey anti-goat, Alexa 488 donkey anti-rabbit, Alexa 555 donkey anti-rabbit, Alexa 488 doney anti-mouse (Invitrogen), and FluoroMyelin (LifeTechnologies) and were diluted in PBT at 1 in 500.

3.3.5 RNA in situ hybridization

RNA in situ hybridization was performed as previously described541.

3.3.6 Microscopy and image processing

For the embryonic and P7 sections, images were captured with a QImaging RETIGA 2000R or QImaging RETIGA EX digital camera and a Leica DMRXA2 optical microscope using OpenLab5 software (Improvision; Waltham MA). P65 nerve sections (injured and un-injured) were imaged using an inverted epifluorescent microscope (40x objective with oil, z-stack, Axio Observer Research Microscope; Zeiss Observer.Z1). Negative controls were included to distinguish non-specific secondary antibody binding. The captured images were processed using Adobe Photoshop software. Quantification in the embryonic studies was restricted to the Sox10+ NCCs at E9.0, and Sox10+ cells populating the dorsal and ventral root and DRG, and the exiting spinal nerve at the remaining stages. Quantification in the injury studies was conducted using a minimum of three images taken distal to the crush site from three different tissue sections for each animal (n=3 mice per group). Double and single-positive cells (protein of interest co-localizing with Sox10+ SCs and Sox10+ SCs) were manually counted using Adobe Photoshop software. An unpaired student’s t-test was performed using Prism software

(GraphPad) to determine whether there was a difference in protein expression of intact or actuely injured adult nerves (significance p<0.05).

3.4 Results

To perform a temporal assessment of the developmental expression profiles of transcription factors implicated in mouse SC lineage progression, we analyzed five embryonic stages of development (i.e., E9.0, E10.5, E12.5, E14.5, and E18.5) as well as the postnatal nerve (P7 and P65). These stages are associated with distinct phases of SC development: (i) delamination of trunk NCCs from the dorsal neural tube (E9.0); (ii) migration of trunk NCCs along the ventral path to populate the developing DRG and peripheral nerves (E10.5); (iii) association of SCPs with developing axons (E12.5); (iv) maturation of SCPs into iSCs (E14.5); (v) differentiation of a subset of iSCs into pro-myelinating SCs (E18.5) and (vi) terminal differentiation of iSCs into myelinating and non-myelinating SCs (P7/P65)542 (Fig 3.1A-F).

Fig 3.1. Schematic representation of the different phases of Schwann cell development. Schematic images of transverse sections through the trunk of E9.0 (A), E10.5 (B), E12.5 (C), E14.5 (D) and E18.5 (E) embryos, and a longitudinal section through the postnatal sciatic nerve

(F). Insets in C-F show transverse sections through the sciatic nerve. bc, boundary cap; dr, dorsal root; drg, dorsal root ganglia; nt, neural tube; sn, spinal nerve; vr, ventral root.

3.4.1 Expression of Schwann cell lineage markers in neural crest cells

We first assessed the expression of SC markers in delaminating trunk NCCs at E9.0 (Fig 3.1A). To reliably label NCCs, and at later stages, to mark peripheral cells fated for the glial lineage, we used Sox10 as a co-label in all marker studies. Sox10 is continually expressed in NCCs, SCs and satellite glia throughout development247,248, and is required for the differentiation of all peripheral glia248,263,310. At E9.0, Sox10 was expressed in trunk NCCs delaminating from the neural tube, including those following both dorsolateral and ventral migratory routes (Fig 3.2A-D,A''-D''). In the E9.0 ventral migratory pathway, Sox10 was strongly co-expressed with Sox9 (100±0% Sox9+Sox10+/Sox10+ cells; Fig 3.2A-A'',E), AP2α (76.1±4.1% AP2α+Sox10+/Sox10+ cells; Fig 3.2B-B'',E), Pax3 (100±0% Pax3+Sox10+/Sox10+ cells; Fig 3.2C-C',E'), and Nfatc4 (56.0±3.8% Nfatc4+Sox10+/Sox10+ cells; Fig 3.2D-D'',E). These observations were consistent with previous reports documenting the activation of a Nfat transcriptional reporter287 and the expression of Sox9247, AP2α250, and Pax3251 in migrating NCCs. Of these markers, Sox9 induces a NCC phenotype537, and its expression biases migrating NCCs towards glial and melanocyte lineage selection247, whereas essential roles have only been documented at later developmental stages for the remaining transcription factors in the SC lineage: AP2α maintains a SCP fate, impeding the transition to an iSC250, Pax3 regulates SC proliferation329 and Nfatc4 acts synergistically with Sox10 to initiate Egr2 expression in SCs287.

Fig 3.2. Expression of SC lineage markers in E9.0 NCCs.

(A-E,A'-D',A''-D'') Co-expression of Sox10 with Sox9 (A-A''), AP2α (B-B''), Pax3 (C-C''), Nfatc4 (D-D'') in E9.0 NCCs. A-D are merged images of Sox10 (green) and protein of interest

(red). Blue is DAPI counterstain. A'-D' shows expression profiles of the protein of interest, while A''-D'' show Sox10 expression. Arrows indicate co-expression of protein of interest with Sox10 in NCCs taking the ventral migratory pathway. Quantification of percent Sox10+ co- expression with each of the proteins of interest (E). Error bars=S.E.M. nt, neural tube. Scale bars, 40μm.

At E9.0, Etv5 was also expressed, but only in a very small number of Sox10+ NCCs (Fig 3.3A- C), which may be why previous reports have suggested that Etv5 is not expressed in the E9.0 neural crest249. Jun expression was also detected in a subset of NCCs, but instead of labelling cells in the ventral migratory pathway, it was primarily expressed in NCCs following a dorsolateral migratory route, which are pre-destined to a melanocyte fate (Fig 3.3D-F). In contrast, we did not detect the expression of Oct6 (Fig 3.3G-I), Sox2 (Fig 3.3J-L), Yy1 (Fig 3.3M-O), Egr1 (Fig 3.3P-R) or Egr2 (Fig 3.3S-U) in E9.0 trunk NCCs. Thus, Sox9, AP2α, Pax3, and Nfatc4 are widely co-expressed with Sox10 in E9.0 NCCs following a ventral migratory route, albeit not in all SCs for AP2α and Nfatc4, whereas Etv5 is only expressed in a small subset of NCCs, and Jun instead marks dorsolaterally migrating NCCs.

100

101

Fig 3.3 Expression of SC lineage markers in E9.0 NCCs.

(A-U) Co-expression of Etv5 (A-C), Jun (D-F), Sox2 (J-L), Yy1 (M-O), Egr1 (P-R), and Egr2 (S-U) with Sox10, and co-expression of Oct6 with Sox9 (G-I) in transverse sections through the E9.0 trunk. A,D,G,J,M,P,S are merged images of the protein of interest in red and Sox10 (or Sox9) in green. Blue is DAPI counterstain. B,E,H,K,N,Q,T show expression profiles of the protein of interest, while C,F,I,L,O,R,U shows Sox10 (or Sox9) expression. Green asterisk indicates background staining from red blood cells. nt, neural tube. Scale bars, 40μm.

3.4.2 Expression of Schwann cell lineage markers in migratory NCC precursors

By E10.5, NCCs have coalesced to form the DRG, which at this stage, are comprised of sensory neurons and migratory NCC precursors that are destined to become SCPs and satellite glial cells (Fig 3.1B). Emanating from the DRG are the dorsal and ventral roots, which coalesce to form the mixed sensory/motor spinal nerve. NCCs also give rise to a subset of multipotent cells called boundary cap cells that are located at the dorsal root entry zone and motor exit points. SCs populating the dorsal and ventral roots find their origin in these boundary cap cells, as do a few satellite glial cells254.

At E10.5, Sox10 was expressed in boundary cap cells in the dorsal and ventral roots of the DRG, in presumptive SCPs, in satellite glia in the periphery of the DRG, and in migratory NCCs in the spinal nerve (Fig 3.4A-C). Sox10 expression was for the most part excluded from the central DRG, where NeuN+ neuronal cells are located (Fig 3.5D-E). Similar to their co- expression profiles in E9.0 NCCs, Sox10 was largely co-expressed with Sox9 (100±0% Sox9+Sox10+/Sox10+ cells; Fig 3.4A-C',P) and AP2α (100±0% AP2α+Sox10+/Sox10+ cells; Fig 3.4D-F',P). Importantly, not all Sox9+ cells were Sox10+, as Sox9 was expressed in a larger subset of non-glial cells (as seen in other stages as well), confirming that the antibodies are not recognizing epitopes shared between Sox family members. We further confirmed the specificity of the antibodies by demonstrating that Sox9 and Sox10 have distinct staining patterns in the CNS (Fig 3.6A-I). Pax3 also continued to be co-expressed with Sox10, however at reduced levels (50.1±7.4% Pax3+Sox10+/Sox10+ cells; Fig 3.4G-I',P), and a much smaller number of Sox10+ NCCs coalescing in the E10.5 DRG and in the ventral root expressed Nfatc4 (3.7±0.8% Nfatc4+Sox10+/Sox10+ cells; Fig 3.4J-L',P). In addition, Etv5 expression was initiated at E10.5 in Sox10+ NCC precursors in the DRG (83.9±6.3% Etv5+Sox10+/Sox10+ cells; Fig 3.4M-O',P), consistent with previous reports249,543.

102

In vitro studies suggested that a block of Etv5 function in NCCs affects neuronal and not glial fate specification544. Consistent with these findings, Etv5 (Fig 3.5J-L) as well as AP2α (Fig 3.5G-I) were also expressed in the neuronal-rich central part of the DRG, where they were co- labeled with NeuN, a pan-neuronal marker. In contrast, Sox9 (Fig 3.4A-C) was exclusively co- expressed with Sox10 in the DRG periphery, where presumptive peripheral glia are located. Sox9 was also expressed with Sox10 in the ventral root, developing spinal nerve, and in mesenchymal tissue between the DRG and somites (Fig 3.4A-C). In contrast, Oct6 (Fig 3.7A- E), Jun (Fig 3.7F-J), Sox2 (Fig 3.7K-O), Yy1 (Fig 3.7P-T), Egr1 (Fig 3.7U-Y) and Egr2 (Fig 3.7Z-DD) expression was not detectable in the developing PNS at this stage, including in Sox9+ and Sox10+ glial cells. In summary, at E10.5, Sox9, AP2α, Pax3, Nfatc4 and Etv5 are co- expressed with Sox10 in presumptive NCC-derived glial cells in the dorsal and ventral roots, DRG and developing spinal nerve, and a subset of these cells have progressed to a SCP fate based on the co-expression of glial markers (data not shown).

103

104

Fig 3.4. Expression of SC lineage markers in E10.5 NCC precursors.

(A-P) Co-expression of Sox10 with Sox9 (A-C'), AP2α (D-F'), Pax3 (G-I'), Nfatc4 (J-L'), and Etv5 (M-O') in transverse sections through the E10.5 trunk. Merged images of the protein of interest (red) with Sox10 (green) (A,D,G,J,M). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (green) (B,E,H,K,N), and single protein of interest images in white (B',E',H',K',N'). High magnification images of the ventral roots, showing merged images of the protein of interest (red) and Sox10 (green) (C,F,I,L,O), and single protein of interest images in white (C',F',I',L',O'). Arrows mark co-expression of proteins of interest with Sox10 in NCC precursors. Quantification of percentage of Sox10+ cells that co-express each of the proteins of interest (P). Error bars=S.E.M. drg, dorsal root ganglion; nt, neural tube; scg, sympathetic chain ganglion; sn, spinal nerve; vr, ventral root. Scale bars, 60μm.

105

Fig 3.5 Co-labelling of SC lineage markers and NeuN in the E10.5 trunk.

(A-L) Co-labelling of NeuN and Sox9 (A-C), Sox10 (D-F), AP2α (G-I), and Etv5 (J-L) in transverse sections through the E10.5 trunk. Low magnification merged images of the protein of interest (red) with NeuN (green) (A,D,G,J). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and NeuN (green) (B,E,H,K), and single protein of interest images in white (C,F,I,L). drg, dorsal root ganglion; nt, neural tube; scg, sympathetic chain ganglion; sn, spinal nerve; vr, ventral root. Scale bars, 60μm. 106

Fig 3.6. Expression of Sox9 and Sox10 in wild type cortices.

(A-I) Co-labelling of Sox10 with Sox9 in sagittal sections of wild type E13.5 (A-C), E15.5 (D- F), and E18.5 (G-I) cortices. Merged images of Sox9 in red and Sox10 in green (A,D,G). Blue is DAPI counterstain. Expression profiles of Sox9 (B,E,H) and Sox10 (C,F,I). Red arrows indicate co-expression of Sox9 with Sox10, while green arrows mark Sox9+Sox10- cells. Scale bars, 60μm

107

Fig 3.7. Expression of SC lineage markers in E10.5 NCC precursors.

(A-DD) Co-labelling of Sox9 with Oct6 (A-E), and Sox10 with Jun (F-J), Sox2 (K-O), Yy1 (P-T), Egr1 (U-Y), and Egr2 (Z-DD) in transverse sections through the E10.5 trunk. Merged

108 images of the protein of interest (red) with Sox10 or Sox9 (green) (A,F,K,P,U,Z). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (green) (B,G,L,Q,V,AA), and single protein of interest images in white (C,H,M,R,W,BB). High magnification images of the ventral roots, showing merged images of the protein of interest (red) and Sox10 (green) (D,I,N,S,X,CC), and single protein of interest images in white (E,J,O,T,Y,DD). drg, dorsal root ganglion; nt, neural tube; scg, sympathetic chain ganglion; sn, spinal nerve; vr, ventral root. Scale bars, 60μm.

3.4.3 Expression of Schwann cell lineage markers in Schwann cell precursors

By E12.5, the vast majority of NCC precursors destined for a glial lineage have differentiated into either SCPs or satellite glia. Morphologically, SCPs are distinguished from migrating NCCs as they associate directly with growing axon bundles, but they lack the basal lamina secreted by iSCs (Fig 3.1C). SCPs are located proximal to the growing nerve tip, and they participate in compacting the nerves while also guiding axons to their targets152. Satellite glia can be partially distinguished from SCPs based on their location; satellite glia are in the DRG but are excluded from the nerves, whereas SCPs are found in both locations. However, because of the lack of specific markers, satellite glial cells are not easily distinguished from SCPs within the DRG, although they do have a more flattened nuclear morphology545,546. For simplicity, we use the SCP nomenclature for precursor cells for both satellite glia and SCs.

At E12.5, Sox10 expression was slightly more widespread in the DRG compared to E10.5, marking both peripheral and central DRG cells (Fig 3.8A,B). The extension of Sox10 expression into the central DRG did not include sensory neurons, as Sox10 was not co- expressed with NeuN at this stage (Fig 3.9D-F). Instead, Sox10 was expressed exclusively in presumptive peripheral glia, as previously suggested247,248. Sox10 was also expressed in the dorsal (Fig 3.8A,C) and ventral (Fig 3.8A,D) roots, where migrating SCPs are located. In co- expression studies, Sox10 continued to be highly co-expressed with Sox9 (99.8±0.2% Sox9+Sox10+/Sox10+ cells; Fig 3.8A-D',Y) and AP2α (82.3±1.6% AP2α+Sox10+/Sox10+ cells; Fig 3.8E-H',Y), whereas Pax3 co-expression started to decline slightly (79.1±2.8% Pax3+Sox10+/Sox10+ cells; Fig 3.8I-L',Y). A larger drop was seen in the number of Sox10+ precursor cells that co-expressed Etv5 (22.9±1.5% Etv5+Sox10+/Sox10+ cells; Fig 3.8M-P',Y). Most Sox10+Etv5+ cells lined the periphery of the DRG and were likely satellite glial cells based on their flattened nuclei. Etv5 expression was also detected in a few SCPs in the dorsal and ventral root. In addition, Etv5 was co-expressed with NeuN+ in DRG sensory neurons (Fig

109

3.9M-O). This data is consistent with previous reports indicating that Etv5 transcripts are detected in satellite glial cells and DRG sensory neurons249,543, although this previous study did not detect Etv5 transcripts in the sciatic nerve544. In contrast to Etv5, Sox9 (Fig 3.9A-C) and AP2α (Fig 3.9G-I) were not co-expressed with NeuN in DRG sensory neurons and were thus glial-specific.

At E12.5, Sox2 and Egr1 expression was also initiated in a subset of Sox10+ SCPs (71.9±5.8% Sox2+Sox10+/Sox10+ cells; Fig 3.8Q-T',Y, 40.7±0.8% Egr1+Sox10+/Sox10+ cells; Fig 3.8U- X',Y) in both the DRG and roots. In contrast, Nfatc4 (Fig 3.10A-G) was no longer co-expressed with Sox10 at E12.5. Additional transcription factors that were not co-expressed with Sox10 in E12.5 SCPs included Oct6 (Fig 3.10H-N), Jun (Fig 3.10O-U), Yy1 (Fig 3.10V-BB) and Egr2 (Fig 3.10CC-II).

In summary, E12.5 SCPs undergo a temporal shift in their expression profile, retaining the expression of Sox10, Sox9, AP2α and Pax3, as observed in E10.5 NCCs, while losing the expression of Nfatc4 and Etv5, and gaining the expression of Sox2 and Egr1. The initiation of Sox2 (Fig 3.8Q-T') and Egr1 (Fig 3.8U-X') expression in E12.5 peripheral glia is consistent with previous reports documenting the expression of Sox2 in SCPs and iSCs300 and Egr1 in SCPs261. However, Egr1 is also co-expressed with NeuN in the DRG (Fig 3.9S-U), indicating that it also labels sensory neurons. Of the genes that are newly expressed at this stage, Sox2 expression is upregulated in a subset of cells that develop into the PNS300, and within the SC lineage, Sox2 regulates the differentiation of SCPs into myelinating SCs versus melanocytes269, while Egr1 is considered a non-myelinating SC marker261.

110

Fig 3.8. Expression of SC lineage markers in E12.5 SCPs.

(A-Y) Co-expression of Sox10 with Sox9 (A-D'), AP2α (E-H'), Pax3 (I-L'), Etv5 (M-P'), Sox2 (Q-T'), and Egr1 (U-X') in transverse sections through the E12.5 trunk. Low magnification merged images of protein of interest (red) and Sox10 (green) (A,E,I,M,Q,U). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (green) (B,F,J,N,R,V), and single protein of interest images in white

111

(B',F',J',N',R',V'). High magnification images of the dorsal roots, showing merged images of the protein of interest (red) and Sox10 (green) (C,G,K,O,S,W), and single protein of interest images in white (C',G',K',O',S',W'). High magnification images of the ventral roots, showing merged images of the protein of interest (red) and Sox10 (green) (D,H,L,P,T,X), and single protein of interest images in white (D',H',L',P',T',X'). Arrows indicate co-expression of proteins of interest with Sox10 in SCPs. Quantification of percentage of Sox10+ cells that co-express each of the proteins of interest (Y). Error bars=S.E.M. dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord; scg, sympathetic chain ganglion; sn, spinal nerve; vr, ventral root. Scale bars, 60μm.

Fig 3.9. Co-labelling of SC lineage markers and NeuN in the E12.5 trunk.

(A-U) Co-labelling of NeuN with Sox9 (A-C), Sox10 (D-F), AP2α (G-I), Sox2 (J-L), Etv5 (M- O), Jun (P-R), Egr1 (S-U). Low magnification merged images of the protein of interest (red) with NeuN (green) (A,D,G,J,M,P,S). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and NeuN (green) (B,E,H,K,N,Q,T), and single protein of interest images in white (C,F,I,L,O,R,U). Arrows indicate co-expression of proteins of interest with NeuN in DRG sensory neurons. drg, dorsal root ganglion; sc, spinal cord. Scale bars, 60μm. 112

Fig 3.10. Expression of SC lineage markers in E12.5 SCPs.

(A-II) Co-expression of Sox9 with Oct6 (H-N), and Sox10 co-expression with Nfatc4 (A-G), Jun (O-U), Yy1 (V-BB), and Egr2 (CC-II) in transverse sections through the E12.5 trunk. Low magnification merged images of protein of interest (red) and Sox9 (green, H) or Sox10 (A,O,V,CC green). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox9 (I; green) or Sox10 (B,P,W,DD; green), and single protein of interest images in white (C,J,Q,X,EE). High magnification images of the dorsal roots, showing merged images of the protein of interest (red) and Sox9 (K; green) or Sox10 (D,R,Y,FF; green), and single protein of interest images in white (E,L,S,Z,GG). High magnification images of the ventral roots, showing merged images of the protein of interest (red) and Sox9 (M; green) or Sox10 (F,T,AA,HH; green), and single protein of interest images in white (H,N,U,BB,II). dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord; scg, sympathetic chain ganglion; sn, spinal nerve; vr, ventral root. Scale bars, 60μm.

113

3.4.4 Expression of Schwann cell lineage markers in immature Schwann cells

As development proceeds, SCPs can either give rise to iSCs, or alternatively, endoneurial fibroblasts and melanocytes268,536. iSCs appear from E14.5 and persist until just before birth152 (Fig 3.1D). iSCs cluster around several axons and deposit a basal lamina that surrounds both the iSCs and the axonal bundle282. iSCs then penetrate axonal bundles, positioning larger diameter axons in the periphery for radial sorting. A characteristic feature of iSCs is that they secrete autocrine survival factors so that they are no longer entirely dependent on axon-derived Neuregulin1, present on the surface of axons152,272.

By E14.5, Sox10 was expressed in scattered cells throughout the DRG, including in the center and periphery, where iSCs and satellite glia are located (Fig 3.11A,B). Sox10 was not co- expressed with NeuN, confirming that it is exclusively labelling glial precursors at E14.5 (data not shown). In addition, Sox10 was expressed in the dorsal (Fig 3.115A,C) and ventral roots (Fig 3.11A,D) and in the exiting spinal nerve (data not shown). In co-expression studies at E14.5, Sox10 was still highly co-expressed with Sox9 (97.1±1.8% Sox9+Sox10+/Sox10+ cells; Fig 3.11A-D',Q) and AP2α (75.4±6.1% AP2α+Sox10+/Sox10+ cells; Fig 3.11E-H,Q'). In contrast, a decline in Sox2 (54.4±1.6% Sox2+Sox10+/Sox10+ cells; Fig 3.11I-L',Q) and Egr1 co-expression with Sox10 was seen (12.8±0.7% Egr1+Sox10+/Sox10+ cells; Fig 3.11M-P',Q).

Compared to E12.5 SCPs, E14.5 iSCs also lost the expression of Pax3 (Fig 3.12A-G) and failed to express Nfatc4 (Fig 3.12H-N), Etv5 (Fig 3.12O-U), Jun (Fig 3.12V-BB), Oct6 (Fig 3.12CC- II), Yy1 (Fig 3.12JJ-PP) and Egr2 (Fig 3.12QQ-WW). Thus, the major difference between E12.5 SCPs and E14.5 iSCs is the loss of Pax3 expression. Previous studies had detected Pax3 transcripts in SCPs as well as in iSCs, but indicated that Pax3 transcript levels decline in late iSCs undergoing radial sorting, and protein levels were not assessed291. In summary, E14.5 iSCs are characterized by the expression of Sox10, Sox9, AP2α, Sox2 and Egr1, as well as glial lineage markers (data not shown), and they differ from E12.5 SCPs in that they no longer express Pax3.

114

Fig 3.11. Expression of SC lineage markers in E14.5 iSCs.

(A-Q) Co-expression of Sox10 with Sox9 (A-D'), AP2α (E-H'), Sox2 (I-L') and Egr1 (M-P') in transverse sections through the E14.5 trunk. Low magnification merged images of protein of interest (red) and Sox10 (green) (A,E,I,M). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (green) (B,F,J,N), and single protein of interest images in white (B',F',J',N'). High magnification images of the dorsal roots, showing merged images of the protein of interest (red) and Sox10 (green) (C,G,K,O), and single protein of interest images in white (C',G',K',O'). High magnification images of the ventral roots, showing merged images of the protein of interest (red) and Sox10 (green) (D,H,L,P), and single protein of interest images in white (D',H',L',P'). Arrows indicate co-expression of proteins of interest with Sox10 in iSCs. Quantification of percentage of Sox10+ cells that co-express each of the proteins of interest (Q). Error bars=S.E.M. dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord; vr, ventral root. Scale bars, 60μm.

115

Fig 3.12. Expression of SC lineage markers in in E14.5 iSCs.

(A-WW) Co-expression of Sox10 with Pax3 (A-G), Nfatc4 (H-N), Etv5 (O-U) Jun (V-BB), Yy1 (JJ-PP), Egr2 (QQ-WW), and Sox9 with Oct6 (CC-II) in transverse sections through the E14.5 trunk. Low magnification merged images of protein of interest (red) and Sox10 (A,H,O,V,JJ,QQ; green) or Sox9 (CC; green). Blue is DAPI counterstain. High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (B,I,P,W,KK,RR; green) or Sox9 (DD; green), and single protein of interest images in white (C,J,Q,X,EE,LL,SS). High magnification images of the dorsal roots, showing merged images of the protein of interest (red) and Sox10 (D,K,R,Y,MM,TT; green) or Sox9 (FF; green), and single protein of interest images in white (E,L,S,Z,GG,NN,UU). High magnification images of the ventral roots, showing merged images of the protein of interest (red) and Sox10 (F,M,T,AA,OO,VV; green) or Sox9 (HH; green), and single protein of interest images in white (G,N,U,BB,II,PP,WW). dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord; vr, ventral root. Scale bars, 60μm.

116

3.4.5 Expression of Schwann cell lineage markers in pro-myelinating Schwann cells

As iSCs develop, they extend cytoplasmic processes that penetrate axonal bundles, helping to distinguish large and small diameter axons. Larger axons are rearranged to the periphery of the bundle, with iSCs associating in a 1:1 proportional manner with these large diameter axons, resulting in radial sorting282. iSCs that associate with large diameter axons are a transient population termed pro-myelinating SCs; these are the SCs that will progress towards the myelinating stage (Fig 3.1E). In addition, a subset of late iSCs persists at E18.5 in association with multiple smaller axons, which are destined to become non-myelinating SCs. Thus, pro- myelinating SCs represent a transient phase in the SC lineage, first appearing just prior to birth, and expanding greatly on the first postnatal day257.

At E18.5, Sox10 was expressed throughout the DRG, including in the center and periphery (Fig 3.13A,B). In the DRG center, Sox10+ late iSCs and pro-myelinating SCs amalgamated around the growing spinal nerve (Fig 3.13I,J). In dual labelling studies, the only transcription factors expressed at E14.5 that continued to be co-expressed with Sox10 at E18.5 were Sox9 (98.6±1.4% Sox9+Sox10+/Sox10+ cells; Fig 3.13A-B',K) and AP2α (19.4±3.2% AP2α+Sox10+/Sox10+ cells; Fig 3.13C-D',K). The number of AP2α+Sox10+ cells was greatly reduced and they were limited to the DRG and not detected in the nerves. Sox10+ late iSCs/pro- myelinating SCs also initiated the expression of Jun (18.3±3.3% Jun+Sox10+/Sox10+ cells; Fig 3.13E-F',K), Oct6 (24.1±2.1% Oct6+Sox9+/Sox9+ cells; Fig 3.13G-H',K), and Yy1 (1.3±0.3% Yy1+Sox10+/Sox10+ cells; Fig 3.13I-J',K) in a small number of cells.

Thus, three new transcription factors are expressed in E18.5 late iSCs and pro-myelinating SCs: Jun, Oct6 and Yy1. Jun has previously been shown to be expressed in late immature SCs and downregulated with the onset of myelination323. Oct6+ SCs were restricted for the most part to the boundary cap or the ventral root and exiting spinal nerve (Fig 3.13G-H), most likely representing pro-myelinating SCs. Indeed, Oct6 is well studied for its role as a cell autonomous regulator of SC development333 and a pro-myelinating SC marker286, and is also essential for bringing about the pro-myelinating to myelinating SC transition286. Finally, Yy1 is important for attaining the myelination phenotype, such that conditional knockdown of Yy1 in SCs results in hypomyelinated nerves with poor expression of the myelin genes MPZ and Pmp22539.

117

118

Fig 3.13. Expression of SC lineage markers in E18.5 late immature/pro-myelinating SCs.

(A-K) Co-labelling of Sox10 with Sox9 (A-B'), AP2α (C-D'), Jun (E-F') and Yy1 (I-J'), and co- labelling of Sox9 and Oct6 (G-H') in transverse sections through the E18.5 trunk. Low magnification merged images of protein of interest (red) and Sox10 (green) (A,C,E,I) or Sox9 (G). High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (or Sox9) in green (B,D,F,H,J), and single protein of interest images in white (B',D',F',H',J'). Blue is DAPI counterstain in A-I. Arrows indicate co-expression of proteins of interest with Sox10 (or Sox9) in late immature and pro-myelinating SCs. Quantification of percentage of Sox10+ cells that co-express each of the proteins of interest (K). Error bars=S.E.M. drg, dorsal root ganglion; sc, spinal cord; sn, spinal nerve; vr, ventral root. Scale bars, 60μm.

At E18.5, Sox2 (Fig 3.14G-I) and Egr1 (Fig. 3.14M-O) expression was lost, and Pax3 (Fig 3.14A-C), Etv5 (Fig 3.14D-F), Nfatc4 (Fig 3.14J-L), and Egr2 (Fig 3.14P-R) were also not expressed in E18.5 pro-myelinating SCs in either the DRG or nerve. The absence of Egr2 protein was surprising as Egr2 transcripts have been detected in the embryonic SC lineage, most notably, in boundary cap cells, from early embryonic stages (Fig 3.15Q-V)254. One possibility is that Egr2 is not translated. Indeed, all three of the Egr2 antibodies that we tested labeled SCs in adult nerves (Fig 3.15F-H, N-P); whereas Egr2 protein remained undetectable embryonically (Fig 3.15A-E, I-M). In summary, E18.5 late iSCs and pro-myelinating SCs in the DRG and nerve roots are characterized by the expression of Sox10, Sox9, Jun, Oct6, Yy1 and AP2α (in the DRG only), and the loss of expression of Sox2 and Egr1.

119

Fig 3.14. Expression of SC lineage markers in E18.5 late immature/pro-myelinating SCs.

(A-R) Co-labelling of Sox10 with Pax3 (A-C), Etv5 (D-F), Sox2 (G-I), Nfatc4 (J-L), Egr1 (M- O) and Egr2 (P-R). Low magnification merged images of protein of interest (red) and Sox10 (green) (A,D,G,J,M,P). High magnification images of the DRG, showing merged images of the protein of interest (red) and Sox10 (green) (B,E,H,K,N,Q), and single protein of interest

120 images in white (C,F,I,L,O,R). Blue is DAPI counterstain. drg, dorsal root ganglion; sc, spinal cord; sn, spinal nerve; vr, ventral root. Scale bars, 60μm.

Fig 3.15. Expression of SC lineage markers in SCs.

(A-V) Co-labelling of Sox10 with Egr2 (Bioss Antibodies) (A-E & F-H), Egr2 (Abcam) (I-M & N-P) in transverse sections of the E12.5 trunk (A,I) and longitudinal sections of postnatal sciatic nerve (F,N). Merged images of the protein of interest in red and Sox10 in green (A,F,I,N). Blue is DAPI counterstain. RNA in situ hybridisation analysis of Egr2 in transverse trunk sections at E12.5 (Q-S) and E18.5 (T-V). Scale bars, 60μm.

3.4.6 Expression of Schwann cell lineage markers in mature neonatal (P7) and adult (P65) myelinating and non-myelinating Schwann cells

Pro-myelinating SCs that are in contact with larger diameter axons (with high levels of Neuregulin 1) progress to form myelinating SCs, whereas iSCs in association with smaller diameter axon bundles (releasing lower levels of Neuregulin1) post-radial sorting become non- myelinating SCs283 (Fig 3.1F). At P7, Sox10+ myelinating and non-myelinating SCs surrounded the sciatic nerve (Fig 3.16A''-F''). In co-labelling experiments in the P7 nerve, Sox10 was highly co-expressed with Sox9 (100±0% Sox9+Sox10+/Sox10+ cells; Fig 3.16A- A'',G) and Nfatc4 (99.1±0.5% Nfatc4+Sox10+/Sox10+ cells; Fig 3.16B-B'',G), Jun (32.6±9.6% Jun+Sox10+/Sox10+ cells; Fig 3.16C-C'',G), and to a lesser extent with Oct6 (51.3±3.9% Oct6+Sox10+/Sox10+ cells Fig 3.16D-D'',G), Yy1 (27.4±0.7% Yy1+Sox10+/Sox10+ cells; Fig 3.16E-E'',G) and Egr2 (46.1±0.7% Egr2+Sox10+/ Sox10+ cells; Fig 3.16F-F'',G). This expression profile was very similar to that observed in E18.5 pro-myelinating SCs, except that the expression of Nfatc4 was re-initiated and Egr2 protein was now detected at P7. In contrast,

121 we did not detect the expression of AP2α (Fig 3.17A-C), Pax3 (Fig 3.17D-F), Etv5 (Fig 3.17G- I), Sox2 (Fig 3.17J-L), or Egr1 (Fig 3.17M-O) in the P7 nerve.

While Egr2 promotes the terminal differentiation of SCs to a myelinating phenotype, Egr1 and Pax3 are considered non-myelinating SC markers261,290. Sox9 is also expressed later in neonatal myelinating and non-myelinating SCs547, and can cooperatively bind the Myelin protein zero promoter, a mature SC marker538, suggesting that Sox9 may also function in postnatal SCs. Nfatc4 acts synergistically with Sox10 to activate the expression of Egr2 in the embryonic nerve287,548, which suggests a later role for Nfatc4 as Egr2 is required for the terminal differentiation of SCs to a myelinating phenotype261. Similarly, Oct6286 and Yy1539 are required for the myelination of peripheral nerves. In contrast, Jun expression is downregulated by Egr2 upon the onset of myelination, and is thus considered a marker of non-myelinating SCs323.

Thus, the P7 nerve is primarily populated by Sox10+ myelinating SCs that co-express Sox9, Nfatc4, Jun, Oct6, Yy1 and Egr2, differing from E18.5 pro-myelinating SCs by the initiation of Nfatc4 and Egr2 protein expression. Non-myelinating SCs may also be present based on the expression of Jun, but they fail to express Egr1 (at least in the nucleus) and Pax3 at this stage.

At P65, within the Sox10+ SC pool, 90.6±2.2% were associated with fluoromyelin+ myelin segments (Fig 3.18A-A'''), indicative of a mature myelinating SC. A subset of Sox10+ cells were also observed in linear arrays independent of fluoromyelin+ myelin segments (not shown), and were presumed to be non-myelinating SCs associated with Remak bundles542. In addition, another subset of fluoromyelin-negative, Sox10+ SCs were associated with a GFP reporter driven by the Sox2 promoter (Fig 3.18A,A'), presumed to be rare iSCs290 or non-myelinating SCs549. We restricted all marker co-localization analysis to Sox10+ cells that were independently localized (i.e., not associated with linear arrays) and were associated with myelin. The majority of Sox10+ SCs co-expressed Sox9 (88.4±0.3%; Fig 3.18B-B''), Nfatc4 (53.3±6.8%; Fig 3.18C-C''), Yy1 (50.0±0.6%; Fig 3.18H-H''), and Egr2 (91.6±1.3%; Fig 3.18I- I''). Egr1 expression was reinitiated in the Sox10+ SCs at P65 (27.1±13.6%; Fig 3.18E-E''). The expression of both Egr1 and Egr2 in the P65 nerve suggests that both myelinating and non- myelinating SCs are present. Notably, Jun was only rarely expressed in Sox10+ SCs (0.6±0.5%; Fig 3.18G-G''), as was Oct6 (4.3±1.8%; Fig 3.18F-F''). Finally, consistent with a prior study550, very few Sox2+Sox10+ SCs were observed within the adult nerve (4.3±2.9%; Fig 3.18D-D''). Thus, the expression profile at P65 was very similar to that observed in P7 SCs, except for

122 expression of Egr1 and appearance of rare Sox2+Sox10+ SCs. We did not detect the expression of AP2α (Fig 3.19A-C), Pax3 (Fig 3.19G-I), or Etv5 (Fig 3.19M-O) in the uninjured P65 nerve.

In summary, early postnatal SCs are characterized by the continued expression of Sox10, Sox9, Jun, Oct6 and Yy1, the acquisition of Nfatc4 and Egr2 protein expression, and the loss of AP2α expression. In contrast, late postnatal (i.e., adult) SCs initiate Sox2 and Egr1 expression at low levels and begin to downregulate Jun and Oct6 expression.

123

124

Fig 3.16. Expression of SC lineage markers in P7 mature SCs.

(A-F'') Co-labelling of Sox10 with Sox9 (A-A''), Nfatc4 (B-B''), Jun (C-C''), Oct6 (D-D''), Yy1 (E-E''), and Egr2 (F-F'') in longitudinal sections of the P7 sciatic nerve. Merged images of the protein of interest in red and Sox10 in green (A-F). Blue is DAPI counterstain. Expression profiles of the protein of interest (A'-F') and Sox10 (A''-F''). Arrows indicate co-expression of proteins of interest with Sox10 in P7 mature SCs Quantification of percentage of Sox10+ cells that co-express each of the proteins of interest (G). Error bars=S.E.M. Scale bars, 40μm

125

Fig 3.17. Expression of SC lineage markers in P7 myelinating/non-myelinating SCs.

(A-O) Co-labelling of Sox10 with AP2α (A-C), Pax3 (D-F), Etv5 (G-I), Sox2 (J-L), and Egr1 (M-O) in longitudinal sections of the P7 sciatic nerve. Merged images of the protein of interest in red and Sox10 in green (A,D,G,J,M). Blue is DAPI counterstain. Expression profiles of the protein of interest (B,E,H,K,N) and Sox10 (C,F,I,L,O). Scale bars, 40μm.

126

127

Fig 3.18. Expression of SC lineage markers in adult P65 uninjured sciatic nerve.

(A-I'') Co-labelling of Sox10 with fluoromyelin (A-A'''), Sox9 (B-B''), Nfatc4 (C-C''), Sox2 (D- D''), Egr1 (E-E''), and Oct6 (F-F''), Jun (G-G''), Yy1 (H-H''), Egr2 (I-I'') in longitudinal sections of the uninjured P65 sciatic nerve. Merged images of protein of interest (green), Sox10 (red) and Hoechst (blue) (B-I). Yellow arrowheads indicate co-labelled cells. Scale bars, 20μm.

128

Fig 3.19. Expression of SC lineage markers absent in the P65 adult uninjured and injured nerve.

129

(A-R) Co-labelling of Sox10 with AP2α (A-C & D-F), Pax3 (G-I & J-L), and Etv5 (M-O & P- R). Merged images of protein of interest (green), Sox10 (red) and Hoechst (blue) (A,D,G,J,M,P). Scale bars, 20μm.

3.4.7 Nerve injury triggers adult Schwann cells to recapitulate a unique pattern of embryonic glial lineage transcription factors

Peripheral nerve injury has been suggested to induce SC de-differentiation and an iSC phenotype533,535. However, while there are clearly global changes in SC gene expression post- injury324,325,535, whether a specific embryonic SC stage is recapitulated, and which set of gliogenic transcription factors are deregulated, remains poorly understood. We thus assessed our panel of transcription factors in “repair” SCs by performing a crush injury on the P60 sciatic nerve, and assessing gene expression at P65, 5 days post injury (dpi).

Similar to uninjured P65 nerves, Sox9 was expressed in the majority of Sox10+ cells in the distal stump at 5 dpi (81.9±3.6% Sox9+Sox10+/Sox10+ SCs; Fig 3.20A-A'' vs 88.4±0.3% in uninjured nerve; Fig 3.18B-B'', Fig 3.21E). Singly labelled Sox9+ (yellow arrows) and Sox10+ (red arrowheads) SCs are marked to depict specificity of the antibodies (Fig 3.20A-A''). In addition, the ratios of Sox10+ SCs expressing Nfatc4 (53.5±6.2% Nfatc4+Sox10+/Sox10+ SCs; Fig 3.20B-B'' vs. 53.3±6.8% uninjured; Fig 3.18C-C'', Fig 3.21E) and Yy1 (55.9±7.6% Yy1+Sox10+/Sox10+ SCs; Fig 3.20D-D'' vs. 50.0±0.6% in uninjured nerve; Fig 3.18H-H'', Fig 3.21E) were not significantly altered post-injury. Notably, although the frequency of Egr1+Sox10+ cells did not change after injury (35.1±5.3% Egr1+Sox10+/Sox10+ SCs; Fig 3.20C-C'' vs 27.1±13.6% uninjured; Fig 3.18E-E'', Fig 3.21E), the Egr1 protein was distinctively localized to the nucleus and was intensified relative to uninjured nerves (where expression was cytoplasmic), suggestive of an active role in transcription and/or altered function in denervated SCs. In contrast, the frequency of Egr2+Sox10+ cells within the distal stump was significantly diminished at 5dpi (40.17 ± 4.67% Egr2+Sox10+/Sox10+ SCs; Fig 3.20E-E'' vs. 89.22 ± 2.1% in uninjured nerve; Fig 3.18I-I'', Fig 3.21E). However, Egr2 expression remained in nearly 40% of Sox10+ cells, but the intensity was diminished relative to pre-injury levels, consistent with previous reports261, which showed that Egr2 expression declined one week post-injury. Downregulation of Egr2 following compression injuries has been previously reported551, and is expected, since Egr2 is required to initiate a terminal myelination program261, and SCs limit myelin production post-injury to facilitate repair and

130 enable cell cycle re-entry288,305,552,553. Interestingly, although the frequency of Yy1+ Sox10+ and Nfatc4+ Sox10+ cells did not change after injury, many SCs also co-expressed the mitotic indicator Ki67 (Fig 3.22A-C,D-F). Because injured SCs co-express pro- myelinating/myelinating genes, as well as Ki67 and Sox2, which is expressed in nearly all SCs post injury, myelinating genes do not inhibit the proliferative capacity of de-differentiating SCs.

Fig 3.20. Expression of SC lineage markers in P65 sciatic nerve after acute nerve injury.

131

(A-E) Co-labelling of Sox10 (red) with Sox9 (A-A''), Nfatc4 (B-B''), Egr1 (C-C''), Yy1 (D-D''), and Egr2 (E-E'') in longitudinal sections of the injured P65 sciatic nerve. Merged images of protein of interest (green), Sox10 (red) and Hoechst (blue) (A-E). Yellow arrowheads indicate co-labelled cells. Scale bars, 20μm.

Fig 3.21 Developmental glial-lineage genes are up-regulated after acute nerve injury.

132

(A-F) Co-labelling of Sox10 (grey) with fluoromyelin (red) and GFP (green) in the distal stump of Sox2-eGFP mice after 5 dpi (A-A'''). Co-labelling of Sox10 with Sox2 (B-B''), Oct6 (C-C''), and Jun (D-D''). Merged images of protein of interest (green), Sox10 (red) and Hoechst (blue) (A-E). Yellow arrowheads indicate co-labelled cells. Quantification of percentage of Sox10+ cells that co-express each of the proteins of interest (F). Error bars=S.E.M. Scale bars, 20μm.

Fig 3.22. Expression of pro-myelinating genes Yy1 and Nfatc4 is sustained in proliferating de-differentiated SCs.

Images showing de-differentiated Schwann cells expressing the promyelinating genes (A-C) Nfatc4 (red) and (D-F) YY1 (red) are mitotically active as indicated by co-localization with Ki67 (green; arrows). Nuclei are stained with Hoechst (blue). Scale bars, 20μm.

We next assessed the expression of transcription factors that were only rarely observed within the uninjured P65 nerve, including Sox2, Jun, and Oct6. To do this, we performed crush injuries in Sox2-eGFP mice. Distal to the injury, the majority of Sox10+ SCs were associated with GFP (>95%), and were also associated with degraded fluoromyelin+ myelin segments (Fig 3.21A- A'''). Consistent with this observation, 99.5±0.3% of Sox10+ SCs expressed Sox2 after 5 dpi (Fig 3.21B-B'', Fig 3.21E), reflecting a significant increase in Sox2 expression in repair SCs. Similarly, the majority of Sox10+ SCs distal to the injury site initiated Jun expression (93.5±1.9% Jun+Sox10+/Sox10+ SCs; Fig 3.21D-D'', Fig 3.21E), which has been reported to 133 play a pivotal role during nerve repair325,554. We also detected a significant upregulation in Oct6 expression (83.7±6.1% Oct6+Sox10+/Sox10+ SCs; Fig 3.21C-C'', Fig 3.21E). Notably, while Sox2 is expressed in E12.5 SCPs, during embryonic development, Jun and Oct6 are not expressed until E18.5 in late immature/pro-myelinating SCs, indicating that the repair phenotype does not faithfully recapitulate all characteristics of either embryonic stage.

We failed to detect Pax3 protein in Sox10+ SCs 5 dpi (Fig 3.19J-L), even though Pax3 transcripts have been isolated from SCs following acute injury291. Similarly, Etv5, which marks satellite glial cells and is downregulated in maturing SCs249, was not expressed in Sox10+ SCs (Fig 3.19P-R). Finally, AP2α, which is expressed in SCPs and involved in negatively regulating SC maturation250, was also absent in the distal segment at 5 dpi (Fig 3.19D-F). In summary, with the exception of Egr2, SCs retain the expression of the core transcriptional program of a SC identity post-injury, including Sox9, Nfatc4 and Yy1. In addition, a ‘repair’ SC phenotype is characterized by an increase in expression of the SC lineage markers Sox2, Jun and Oct6, but other embryonic SC markers (AP2α, Pax3 and Etv5) are not induced, suggesting that SCs do not de-differentiate to a particular embryonic state.

3.5 Discussion

Generation of SCs from NCCs is a progressive process characterized by at least five transient embryonic stages of development. Here we have defined these developmental stages by examining the expression patterns of 12 transcription factors (Sox2, Sox9, Sox10, AP2α, Pax3, Nfatc4, Etv5, Jun, Yy1, Egr1, Egr2, Oct6) (Fig 3.23). While other studies have performed more global analyses of gene expression in SC lineages between E9.5 and P0555, similarly delineating the dynamic changes that occur over these time points, our focus on protein expression levels of a core set of transcription factors provides a framework for scientists to reliably follow temporal identity transitions in this lineage. Based on this temporal profile, we can begin to delineate the potential diverse functional contributions of these genes during SC development and their recapitulation following peripheral nerve injury.

At E9.0, the earliest NCC stage is characterized by the expression of Sox10, Sox9, AP2α, Pax3 and Nfatc4, a gene expression profile that is maintained at E10.5. However, a distinct feature of NCC precursors coalescing in the DRG is the initiation of Etv5 expression and concomitant loss of Nfatc4 expression. Next, E12.5 SCPs in the DRG and dorsal and ventral roots retain the expression of Sox10, Sox9, AP2α, and Pax3 while also initiating the expression of Sox2 and

134

Egr1 but losing the expression of Etv5. This is followed at E14.5 by the loss of Pax3 expression, while Sox10, Sox9, AP2α, Sox2 and Egr1 continue to be expressed in iSCs. A more dramatic change in gene expression is observed in E18.5 late immature/pro-myelinating SCs, which express Sox10, Sox9, Jun, Oct6 and Yy1, but lose the expression of AP2α, Sox2 and Egr1 in the nerve. At P7 and P65, we observed a very similar expression profile as seen at E18.5 (i.e., Sox10, Sox9, Jun, Oct6, Yy1) with the added expression of Nfatc4 and Egr2 at P7 and P65, and Egr1 and Sox2 expression in the P65 nerve.

Following peripheral nerve injury, SCs assume a transient ‘de-differentiated’ phenotype that is critical for supporting axonal regeneration and nerve repair. Hence, we characterized expression of the transcription factor panel following injury in the adult P65 nerve, comparing it against the embryonic and postnatal profile. We reveal that denervated SCs upregulate the expression of only a subset of early glial-lineage transcription factors. Indeed, expression of genes associated with both SCPs (Sox2, Egr1) and pro-myelinating/late immature SCs (Oct6, Jun) are elevated following denervation of adult SCs, while genes involved in the myelination program are actively lost (Egr2). Importantly, the absence of expression of several embryonic genes in denervated SCs (AP2α, Etv5, Pax3) may provide a strategy to enhance or prolong the repair phenotype to improve recovery of function following PNS injury or neuropathy.

135

Fig 3.23 Summary of temporal expression profiles of key transcription factors in the SC lineage.

Sox9 and Sox10 are expressed throughout SC genesis, beginning at E9.0 in migrating NCCs and persisting until P65 in myelinating and non-myelinating SCs in the sciatic nerve. AP2α, Pax3 and Etv5 are also expressed in NCCs, persisting until E12.5 in SCPs with Nfatc4 expression being restricted to NCCs. Egr1 and Sox2 expression is initiated in SCPs at E12.5, persisting until E14.5 in iSCs. At E18.5, AP2α is expressed in the DRG but is undetectable in the nerves while the expression of Jun, Oct6, and Yy1 is initiated and persists till P65. Rare Sox2+Sox10+ positive cells are also detected in the P65 nerve. Egr2 expression is not detected until P7 in myelinating SCs in the sciatic nerve. Post-injury, high upregulation in expression of Sox2, Oct6, and Jun is observed, while distinct nuclear Egr1 expression is also detected. The straight lines represent continued expression of the markers through the different stages, while the dotted lines represent declining or low expression. The asterisk indicates where expression is restricted to the DRG and is undetectable in the nerves at E18.5. Markers expressed in the nerve after an acute injury have been denoted with ‘√’. The green cells represent the developing axon, while the beige cells represent the NCCs at E9.0-E10.5 and the developing SCs at E12.5 to postnatal stages.

3.5.1 Sustained expression of Sox9 and Sox10 across the Schwann cell lineage

An intricate network of transcription factors controls the timely development of peripheral glia, including several of the transcription factors examined in this study. One of the core regulators of SC development is Sox10, which we used to mark peripheral glial cells in our co-labelling experiments. Indeed, we (this study) and others247,248 found that Sox10 is continually expressed in Schwann and satellite glial cells throughout development and into adulthood. Prior studies have revealed that Sox10 is required for the specification and terminal differentiation of iSCs, and to maintain a peripheral glial phenotype248,263. Mechanistically, Sox10 directly activates Egr2 expression, acting synergistically with Oct6 and Nfatc4302-305,538 to induce the expression of peripheral myelin genes such as myelin basic protein (MBP), myelin protein zero (MPZ), myelin associated glycoproteins and connexin-32 (Cx32)287,303,305. Consequently, deletion of Sox10 results in loss of Egr2 expression, as well as myelin sheath degeneration and axonal death resulting in declined nerve conduction 311. However, conditional Sox10 ablation studies also revealed that Sox10 is essential for survival of early migrating trunk NCCs, but not the

136 survival of adult SCs, indicating that it is a critical player early in the SC lineage311-313 and later to maintain functional myelination.

Interestingly, we also found that Sox9 follows a similar temporal pattern with sustained, overlapping expression with Sox10. Sox9 induces a NCC phenotype537, and its expression biases migrating NCCs towards glial and melanocyte lineage selection247. While previous studies have observed Sox9 expression in the peripheral nerve at E14.5547, earlier stages of Sox9 expression have not been documented. Interestingly, recent work has suggested that isolated human SCs show negligible Sox9 expression, but Sox9 is over-expressed in neurofibromatosis 1 tumor derived SCs556, hinting to a role for Sox9 in promoting SC proliferation. It is possible that sustained Sox9 activation in Schwann cells may enable re- establishment of the immature SC phenotype and re-entry into the cell cycle, and should be addressed in future studies.

3.5.2 Transcriptional regulators expressed at early stages in the Schwann cell lineage

In addition to Sox9 and Sox10, four other transcriptional regulators in our panel were expressed at the NCC stage; AP2α, Pax3, and Etv5, with Etv5 appearing one day later than the others. One novel observation was that Nfatc4, which is a calcium-responsive transcription factor, exhibits transient early expression in NCCs but is rapidly lost by E10.5, before re-initiating expression in maturing SCs at P7. A role for Nfatc4 in early NCCs has not previously been reported, however at later stages, Nfatc4 has been reported to bind a myelin specific enhancer in Egr2, cooperatively with Sox10, to activate Egr2 and other myelin genes during the pro- myelination to myelination transition287.

AP2α expression persists until 18.5, however, AP2α+Sox10+ cells at this stage are confined to the DRG, and AP2α expression is not seen in the Sox10+ SCs lining the nerve. This could be suggestive of a role for AP2α in satellite glial cells in the DRG. Also, AP2α is co-expressed with majority of Sox10+ cells throughout development, except at E18.5 when AP2α expression becomes restricted to the DRG. Interestingly, overexpression of AP2α in vitro blocks the transition of SCPs to iSCs250, even though we found that this transcription factor is expressed in iSCs and pro-myelinating SCs. The in vivo requirement for AP2α in the SC lineage has not yet been determined.

137

Pax3 is co-expressed with most Sox10+ glial lineage cells at early embryonic stages, and may play a role in regulating the proliferation of these early glial cells. Indeed, Pax3 induces proliferation in SCPs, and Pax3 transcript levels decline at the onset of differentiation291. Finally, we also found that Etv5 expression is limited to E10.5 NCC precursors in the DRG and in satellite glia cells at later stages. The function of Etv5 in the SC lineage has not been elucidated, although misexpression of dominant negative Etv5 in NCCs affects neuronal and not glial specification544. The absence of Etv5 at all later time points of the SC lineage (including after injury) suggest that it does not play a role in SC differentiation or myelination.

3.5.3 Transcriptional regulators expressed at late stages in the Schwann cell lineage

Several of the transcription factors in our panel were not expressed in NCCs, but were expressed in definitive cells in the SC lineage, including Sox2, Egr1, Jun, Oct6, Yy1 and Egr2. In contrast to Sox9 and Sox10, Sox2 is expressed in a more limited window of SC development, appearing in E12.5 SCPs and E14.5 iSCs. A decline in number of Sox2+Sox10+ cells is observed at E14.5. This data is consistent with previous studies demonstrating that Sox2 expression declines upon neuronal commitment, and continues at low levels in SCPs and iSCs300. Interestingly, persistent Sox2 expression suppresses myelin-associated genes such as Egr2 and MPZ whilst maintaining cells in an undifferentiated state288. Notably, cross- repressive interactions between Sox2 and Mitf/Egr2 regulate the differentiation of SCPs into either myelinating SCs or melanocytes269. Sox2 is thus considered a negative regulator of myelination.

The zinc finger transcription factors Egr1 (Early growth response 1) and Egr2 have nearly identical DNA binding domains but opposite effects on myelination; Egr2 (Krox-20) promotes the differentiation of SCs to a myelinating phenotype while Egr1 is a non-myelinating SC marker257,261. Hence, we (this study) and others261 found that Egr1 is expressed in SCPs but is downregulated as the cells mature. At postnatal stages, Egr1 expression is also re-initiated in non-myelinating SCs261, where a modest cytoplasmic expression pattern is observed, and is sustained in a subset of SCs within the adult (P65) sciatic nerve. Conversely, Egr2 transcripts are detected in the dorsal and ventral roots from E10.5 onwards but are absent from the SCs in the DRG and peripheral nerves throughout embryogenesis254. Strikingly, we did not detect Egr2 protein in the dorsal and ventral roots, suggesting that it may not be translated until

138 postnatal stages. Indeed, we (this study) and others261 observed Egr2 protein in SCs lining the postnatal peripheral nerve, which is expected considering its requirement for myelination.

We observed Jun expression in late immature/pro-myelinating SCs, P7 and in rare Jun+Sox10+ SCs in the adult nerve (P65). Downregulation of Jun expression is mediated by Egr2 just prior to onset of myelination325. Overexpression of Jun has been associated with a decline in myelination and de-differentiation of SCs, along with a reduction in Egr2 and MPZ levels325. Similarly, we detected Oct6 expression in a subset of Sox10+ pro-myelinating SCs at E18.5, in mature SCs at P7 and in rare cells at P65. Oct6 acts in synergy with Sox10 to induce Egr2 expression, which in turn promotes the expression of several myelin proteins305,333. Oct6 deficient mice show a transient arrest at the pro-myelinating stage, which is overcome by P10, with a late onset of Egr2 expression and myelin formation333. Oct6 not only promotes myelination by promoting terminal differentiation from pro-myelinating to myelinating SC via induction of Egr2, but also prevents premature myelination by repression of MBP and MPZ. A progressive reduction in Oct6 levels allows MBP and MPZ to be activated, thereby initiating a temporally controlled myelination program. Notably, constitutive overexpression of Oct6 results in a persistent hypomyelination phenotype in mice and gradual axonal loss335, suggesting that varied levels of Oct6 allow for diverse functional contributions across the SC lineage.

Yy1, expressed at E18.5 and postnatally at P7/P65, is important for attaining the myelination phenotype, such that conditional knockdown of Yy1 in SCs results in hypomyelinated nerves with deficient expression of MPZ and Pmp22539. The Egr2 promoter and Myelinating Schwann cell Element (MSE) has multiple binding sites for Yy1. Activation of MEK pathway occurs in response to the axonal signaling molecule, Neuregulin1. This in turn results in serine phosphorylation of Yy1. The phosphorylated Yy1 is recruited to binding sites in the Egr2 promoter and MSE, activating Egr2 expression539 and underscoring its important role in SC myelination.

3.5.4 Injury activates SC lineage genes that recapitulate features of both Schwann cell precursors and pro-myelinating Schwann cells

Following a nerve crush injury, we observed continued expression of Sox10 in nerve SCs and co-expression with Sox9, Nfatc4 and Yy1. Previous work showed that Sox9 is expressed in isolated SCs from P3 nerves547 but its presence in adult SCs in vivo or following injury has not

139 been reported. Constitutive expression of Sox9 in adult (uninjured) SCs and following injury in vivo, suggesting that Sox9 may play a continued role in the maintenance of the SC fate or in sustaining SC competence to re-acquire a de-differentiated state, particularly since it has a demonstrated role in both induction and maintenance of self-renewal capacity in subependymal neural stem cells557 and various epithelial stem cell types558,559. Indeed, SC de-differentiation encompasses the hallmark features of a stem cell, exhibiting both the capacity for self-renewal, and the ability to generate mature cell types. Future conditional knockout studies will need to be done to determine the role of Sox9 in adult SCs and its potential contribution to the de- differentiation process.

Acutely injured SCs exhibit robust activation of the myelin-inhibitory gene Sox2, as has been previously reported288, and elevated levels of Egr1, both of which are unique to SCPs and iSCs and absent in late immature/pro-myelinating SCs (Fig 3.21). However, several other transcription factors that are upregulated in denervated SCs are markers of late immature/pro- myelinating SCs, including Jun, and Oct6 (Fig 3.23). SC de-differentiation was also associated with a concomitant loss of mature myelinating genes such as Egr2.

Egr1 is a transcriptional activator that is normally active during cell cycle-re-entry560. Although the frequency of Egr1+Sox10+ SCs did not change, Egr1 protein exhibited a marked increase in intensity and nuclear translocation following injury, suggesting that denervation causes a change in Egr1 function261. Egr1 may thus be an important modulator of SC plasticity. Egr1 and Egr2 appear to play opposing roles in modulating the acquisition of non-myelinating versus myelinating phenotypes261. Sustained Egr1 expression in a subset of SCs at all postnatal stages, and its known role in regulating cell cycle entry, suggests that this factor might also enable SC proliferation after injury and/or activation of other genes that are necessary for de- differentiation, partly through its known cooperation with Egr3561. Future studies using conditional knockout approaches will need to be done in order to determine the ultimate role for Egr1 and Egr3 in acquisition of the de-differentiated SC state.

Interestingly, the frequency of pro-myelinating genes Yy1 and Nfatc4, did not change following injury further indicating the retention of late immature/pro-myelinating SC traits. Despite their putative pro-myelination function, many Yy1+ and Nfatc4+ cells that co-localized with Sox10, were also mitotically active, suggesting that both of these transcription factors are permissive of the proliferative, de-differentiated state.

140

The most notable change after injury was observed with respect to Jun and Oct6 expression. A robust increase in Jun expression has been reported in denervated SCs324,325. Indeed, Jun is a critical regulator of de-differentiation such that loss of Jun results in an inability to downregulate myelination genes and severely impairs regeneration. The POU domain transcription factor Oct6 initiates the transition from pro-myelinating to myelinating SC562. Interestingly, a shift from cytoplasmic to nuclear localization in Oct6 expression has been reported in axonal regeneration in peripheral neuropathic conditions563. It is noteworthy that Oct6 expression is highly upregulated at a time when myelin is being degraded and SCs are repressing their myelin program. Oct6 could be playing multiple roles in governing SC function, a possibility that requires further exploration. Since neither Jun nor Oct6 are expressed in SCPs but only by late immature/pro-myelinating SCs at E18.5, it suggests that at the level of protein expression, the de-differentiated SC state cannot be equated to a single embryonic SC stage, but rather is unique and includes features from multiple developmental stages.

Several key transcription factors including AP2α, Etv5, and Pax3 that are associated with early stages of development were not expressed within denervated SCs. We observed only extremely rare Pax3+ cells in the adult nerve (data not shown). A recent report suggests that Pax3 labels approximately 1% of cells in the adult nerve and marks non-myelinating SCs290. This suggested that possibly only a subset of Sox10+ non-myelinating SCs express Pax3. Notably, we did not observe Pax3 expression following injury either, despite seeing robust expression in NCCs that were immunostained in parallel as a positive control. This is in contrast to a previous study291 that reported Pax3 transcripts were detected in the denervated distal stump at seven days post transection injury. This discrepancy may be due to several factors: 1) the temporal expression profile of Pax3 may be delayed, such that it does not peak until after the 5 day time-point we examined, 2) severity of nerve injury (transection versus crush) may be an important determinant of the SC transcriptional response within denervated SCs, 3) young mice (3 weeks of age) may elicit a different response compared to the adult animals (P65) used in our study, and 4) Pax3 transcripts may not be translated. Future studies using a conditional Pax3 gene deletion could determine its role in establishing the SC repair phenotype after injury.

Our spatio-temporal expression study provides a comprehensive glimpse into the expression profiles of the various transcriptional regulators involved in SC development and in SC injury response. To summarize, the injured peripheral nerve contains a highly dynamic and

141 heterogeneous population of glia that undergoes phenotypic reversion to a de-differentiated state by recapitulating a subset of early glial-associated transcription factors. Taken together, we provide evidence that “repair” SCs retain their core SC transcriptional program, while initiating the expression of a subset of embryonic genes that represent several embryonic SC stages. Since SC function is diminished in the adult aging PNS it may be necessary to artificially activate additional genes, particularly those that fail to be re-activated or are diminished following prolonged denervation, in order to maximize nerve regeneration.

142

Etv5 is not required for Schwann cell development but is required to regulate the Schwann cell response to peripheral nerve injury

Contents of this chapter have been prepared for publication:

Balakrishnan A, Belfiore L, Vasan L, Touahri Y, Stykel M, Fleming T, Midha R, Biernaskie J, Schuurmans C. Etv5 is not required for Schwann cell development but is required to regulate the Schwann cell response to peripheral nerve injury. (Manuscript prepared for submission)

143

4.1 Abstract

Schwann cells are the principal glial cells of the peripheral nervous system, and their development into myelinating glia is critically dependent on MEK/ERK signaling. Ets-domain transcription factors (Etv1, Etv4, Etv5) are common downstream effectors of MEK/ERK signalling, but so far, only Etv1 has been ascribed a role in Schwann cell development, and only in non-myelinating cells. Here, we examined the role of Etv5, which is expressed in Schwann cell precursors, including neural crest cells and satellite glia, in Schwann cell lineage development. We analysed Etv5tm1Kmm mutants (designated Etv5-/-) at embryonic days (E) 12.5, E15.5 and E18.5, focusing on dorsal root ganglia. At these embryonic stages, satellite glia (glutamine synthetase) and Schwann cell markers, including transcriptional regulators (Sox10, Sox9, Tfap2a, Pou3f1) and non-transcription factors (Ngfr, BFABP, GFAP), were expressed in the DRG of wild-type and Etv5-/- embryos. Furthermore, by E18.5, quantification of Sox10+ Schwann cells and NeuN+ neurons revealed that these cells were present in normal numbers in the Etv5-/- dorsal root ganglia. We next performed peripheral nerve injuries at postnatal day 21, revealing that Etv5-/- mice had an enhanced injury response, generating more Sox10+ Schwann cells compared to wild-type animals at five days post-injury. Thus, while Etv5 is not required for Schwann cell development, possibly due to genetic redundancy with Etv1 and/or Etv4, Etv5 is an essential negative regulator of the peripheral nerve injury repair response.

144

4.2 Introduction

Neurons conduct electrical signals responsible for nervous system function, but their activity is critically dependent upon glial cells, which provide trophic, structural and functional support. Macroglial cells in the peripheral nervous system (PNS) include Schwann cells and satellite glial cells, which populate peripheral ganglia, including the dorsal root ganglia (DRG) that flank the developing spinal cord. Schwann cells myelinate peripheral nerves, but also wrap non-myelinated axons to regulate neuronal survival and axonal diameter152. Conversely, satellite glia encircle peripheral neuronal cell bodies, and while their functions are less well studied, they have been likened to central nervous system (CNS) astrocytes564. There is also growing support for a role for satellite glia in regulating pain565-569. Despite their distinct positioning, Schwann cells (wrapping axons) and satellite glia (wrapping neuronal cell bodies) share an embryonic origin, both arising from neural crest cell (NCC) progenitors258. Moreover, satellite glia can transition to a Schwann cell fate, at least in vitro249,570, highlighting their close lineage relationship, and suggesting that satellite glia could replenish Schwann cells when required for repair.

Multiple signaling molecules and transcriptional regulators regulate Schwann cell development571. The ErbB family of receptor tyrosine kinases (RTKs), which are activated by Neuregulin 1 (NRG1), an EGF family ligand, have emerged as central regulators of Schwann cell precursor (SCP) proliferation, migration and myelination530. In addition, the MEK-ERK signal transduction cascade, activated downstream of RTK signaling, is essential for Schwann cell differentiation, as revealed by the analysis of genetic mutants of ERK1 and ERK2572. Conversely, if ERK signaling is ectopically activated in Schwann cells, myelination ensues419,573.

ERK kinases have several downstream effectors that they regulate by phosphorylation, including multiple transcription factors574, such as those of the Ets-domain (helix-turn-helix super family) family575. In Drosophila, Pointed (Pnt) is an ets-domain transcriptional activator that is activated downstream of RTK signaling and which is required for glial cell fate specification576. In vertebrates, critical ets domain factors activated downstream of RTK signaling include Etv1 (ER81), Etv4 (Pea3) and Etv5 (Erm). Activation of MEK-ERK signaling initiates the expression of Etv1 and Etv5 to specify an oligodendrocyte fate, which are myelinating glial cells in the CNS7,577-579. Etv1 is also expressed in Schwann cells580, but it is

145 not required for the development of myelinating Schwann cells581. Instead, Etv1 facilitates interactions between peripheral axons and non-myelinating Schwann cells in Pacinian corpuscles581,582.

Etv5 is also expressed in the developing PNS, beginning at embryonic day (E) 9.0 in NCCs and persisting until ~ E12.5 in SCPs and in satellite glial cells249,260. However, Etv5 expression is not glial specific, as it is also expressed in TrkA+ sensory neurons in peripheral ganglia249,543. In vitro studies indicate that blocking Etv5 function with a dominant negative (dn) construct in NCCs prevents neuronal fate specification whereas Etv5-dn does not impact glial specification, and also does not influence NCC progenitor survival and proliferation544. However, there have been no in vivo studies yet to confirm whether or not Etv5 is involved in Schwann cell development. Additionally, an upregulation of several embryonic genes in Schwann cells post peripheral nerve injury is observed which supports a proliferative repair Schwann cell phenotype 260. However, it remains to be seen whether Etv5 plays a role in maintaining the repair Schwann cell population post injury. Here, we analysed Etv5tm1Kmm mutant mice carrying a deletion of exons 2-5 (hereafter designated Etv5-/-)583 to ask whether Etv5 is required for Schwann cell development and the peripheral nerve injury response. We found no evidence for Schwann cell defects in Etv5-/- embryos, suggesting that Etv5 is not required for Schwann cell development. However, we did observe a striking expansion of the Sox10+ Schwann cell pool in Etv5-/- peripheral nerves post-injury. Thus, Etv5 acts as a negative regulator of the Schwann cell repair response, and in its absence, more Schwann cells are generated. We discuss our findings in the context of important caveats, such as the potential for genetic redundancy with Etv1 and/or Etv4, and our use of a hypomorphic mutant allele, which was necessitated by the early embryonic lethality of Etv5 null mice.

4.3 Material and methods

4.3.1 Animals and genotyping

Animal procedures were approved by the University of Calgary and Sunnybrook Research Institute Animal Care Committees (Primary dissections were approved under animal utilization protocol (AUP) 20-606) in compliance with the Guidelines of the Canadian Council of Animal Care. Etv5tm1Kmm/J mice (Stock No. 022300) from Jackson Laboratory (ME, United States) were maintained on a 129/SvJ background Mice were maintained as heterozygotes in a 12hr light / 12hr dark cycle. Heterozygous intercrosses were set up to generate homozygous mutant

146 embryos, designated Etv5-/-. Pregnancy was determined by detection of a vaginal plug, with the morning of plugging designated as embryonic day (E) 0.5. PCR genotyping was performed with the following primers: wild-type forward primer: TCT GGC TCA CGA TTC TGA AG; mutant forward primer: AAG GTG GCT ACA CAG GCA AG and common reverse primer: CGG AGG TCA AGC TGT TAA GG.

4.3.2 Embryo collection

Embryo trunks or postnatal nerves were dissected in ice-cold phosphate-buffered saline (PBS), and then fixed overnight in 4% paraformaldehyde (PFA)/PBS. Fixed tissue was washed in PBS, immersed in 20% sucrose/PBS overnight, and then blocked in O.C.T™ (Tissue-Tek®, Sakura Finetek U.S.A. Inc., Torrance, CA) before storing at -80C. Blocked tissue was sectioned on a Leica cm3050s cryostat (Richmond Hill, ON) at 10µm and collected on SuperFrost™ Plus slides (Thermo Scientific).

4.3.3 Peripheral nerve crush Peripheral nerve crush injuries were performed on the sciatic nerve of P21 wild-type and Etv5- /- mice as previously described260. Briefly, P21 animals were anesthetized (5% isofluorane for induction and 2% for maintenance), hindlimbs were shaved and sterilized, and a small incision was made to expose the sciatic nerve. A crush injury was performed using #10 forceps for one minute, and then the muscle and skin were sutured back together. Buprenorpine subcutaneous injection of 0.1mL (100μL of 0.03mg/mL) were administered for pain on the day of surgery and for 4 days following, with the nerve harvested on day 5. Surgeries were approved under AUP 16-610.

4.3.4 Immunohistochemistry

Sections were thawed, rinsed in PBS to remove excess O.C.T., permeabilized in PBT (PBS with 0.1% TritonX), and then blocked in 10% normal horse serum/PBT for 1 hour. Primary antibodies were then diluted in blocking solution and incubated on sections overnight at 4C, followed by three PBT washes. Species-specific secondary antibodies, conjugated to Alexa 488 or Alexa 555, were diluted 1/500 in PBT and applied to sections for 1 hour. Sections were washed three times in PBT and stained with 4′,6-diamidino-2-phenylindole (DAPI; Santa Cruz Biotechnology) (1:5000 in PBT). Sections were washed three times in PBS and mounted in AquaPolymount (Polysciences). Primary antibodies included: rabbit anti-Etv5 (Abcam ab102010; 1:300), rabbit anti-Tfap2a (Abcam; ab52222; 1:200), goat anti-Oct6 C-20 (Santa 147

Cruz Biotechnology; sc-11661; 1:50), rabbit anti-Sox9 (Millipore; AB5535; 1:500), goat anti- Sox10 (Santa Cruz Biotechnology; sc-17343; 1:400), rabbit anti-Sox10 (Abcam; AB227680; 1:200), Gfap (Dako Cytomation; #Z0334; 1:500) Ngfr (Millipore; #07-476;1:500); BFABP (Millipore; ABN14; 1:500) and mouse anti-NeuN (Millipore MAB377; 1:200).

4.3.5 RNA in situ hybridization.

A digoxygenin-labeled Etv5 riboprobe was generated as previously described using a 10× labelling mix and following the manufacturer's instructions (Roche) 577. The probe was hybridized overnight and washing and staining procedures were followed as previously described 541.

4.3.6 Microscopy and image processing

Images were captured with a QImaging RETIGA EX digital camera and a Leica DMRXA2 optical microscope using OpenLab5 software (Improvision; Waltham MA). Image processing and analysis was performed using Image J software. Three images per wild-type and Etv5-/- embryo/nerve were assessed. DAPI channel images were converted into 8-bit format and the threshold was set using weighted mean intensity. Images were inverted for binary conversion. In E18.5 sections, the dorsal root ganglionic region was manually selected using free-form selection tool, while in nerve sections the entire area was assessed. The number of DAPI+ cells in the selected region were calculated using particle analysis option. The colocalization of DAPI with the green channel (Sox10+/NeuN+ cells) was calculated manually.

4.3.7 Statistical analysis

A minimum of three biological replicates were carried out for all assays. Statistical analysis and graphs were generated using GraphPad Prism 8 software. Student’s t-test (when comparing two groups) or One-way ANOVA with TUKEY post corrections (when comparing groups of more than two) were used. All data expressed as mean value ± standard error of the mean (S.E.M.). In all experiments, a p value <0.05 was considered statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

148

4.4 Results

4.4.1 Schwann cell precursors develop normally in E12.5 Etv5-/- peripheral ganglia

Previous reports have documented that Etv5 is expressed in the Schwann cell lineage from as early as E9.0, first appearing in NCCs, and persisting until E12.5 in satellite glia and Schwann cell precursors (SCPs) in the developing DRG flanking the spinal cord (Fig. 4.1A,B)249,260. We confirmed the expression of Etv5 in SCPs and satellite glia using RNA in situ hybridization (Fig. 4.1C) and by co-immunolabelling with Sox10 (Fig. 4.1D), which marks the Schwann cell lineage at all developmental stages260. To next assess whether Etv5 is required during this early temporal window for Schwann cell lineage development, we examined E12.5 wild-type and Etv5tm1Kmm mutant embryos (Fig. 4.1B). Notably, we studied Etv5tm1Kmm mutant mice carrying a deletion of exons 2-5 encoding the initiation codon and a transactivation domain (hereafter Etv5-/-) because animals homozygous for a null allele (Etv5tm1Hass), which lack the Etv5 DNA binding domain, die by E8.5583,584, precluding an analysis of Schwann cell lineage development.

We first examined the expression of Sox9 and Sox10, two SRY-box family HMG transcription factors. Sox10 is a specific marker of the Schwann cell lineage, but also marks early migrating NCCs and satellite glia throughout the embryonic period and into postnatal stages and is required for development past the immature Schwann cell (iSC) stage248,260,312. Sox9 is also expressed early on in the Schwann cell lineage, beginning at the NCC stage247,260, and it regulates NCC development247. In E12.5 wild-type embryos, Sox9 was expressed more widely, marking SCPs coalescing in the developing DRG and ventral root, but also labelling neural progenitors in the neural tube, part of the CNS, as well as other migratory NCC populations migrating over the surface ectoderm, and mesenchymal cells (Fig. 4.1E,G). A very similar pattern of Sox9 expression was observed in E12.5 Etv5-/- embryos, including in SCPs in the DRG, ventral root, and the motor root of the spinal nerve (Fig. 4.1F,H). In E12.5 sections through the trunk, Sox9 expression overlapped with Sox10 in the DRG and ventral root, but Sox10 expression was much more restricted to the Schwann cell lineage (Fig. 4.1E,I). A similar pattern of Sox10 expression was seen in E12.5 Etv5-/- embryos, with Sox10+ cells restricted to glial cells in the DRG and ventral root (Fig. 4.1F,J).

149

Figure 4.1. Schwann cell transcriptional regulators are expressed normally in E12.5 Etv5- /- Schwann cell precursors (SCPs).

Etv5+ neural crest cells give rise to Etv5+ sensory neurons, Etv5+ satellite glia, and Etv5- Schwann cell precursors (A). Schematic representation of E12.5 trunk section (B). Distribution

150 of Etv5 transcripts in E12.5 transverse sections of the spinal cord (C). Co-expression of Etv5 (red, D), Sox9 (red, E-G,I) and Sox10 (green, D,E,F,H,J) with DAPI counterstain (blue) in E12.5 wild-type (D,E,G,I) and Etv5-/- (F,H,J) embryos. Inset in (D) presents 3X magnified image of region marked by a dotted box. dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord; vr, ventral root. Scale bars, 60 μm.

Transcription factors initiate developmental programs by turning on the expression of genes with functional roles in fate specification and differentiation. Brain fatty acid binding protein (BFABP) is the earliest ‘glial’ gene turned on in SCPs in a Sox10-dependent manner312. Another early marker is Ngfr (also known as p75NTR), a common receptor for neurotrophins, the deletion of which results in a reduced size of peripheral ganglia and reduction in Schwann cell number585. In E12.5 wild-type embryos, both BFABP (Fig. 4.2A,A’) and Ngfr (Fig. 4.2C,C’) were co-expressed with Sox10 in SCPs in the DRG and also prominently in the dorsal root, the sensory root of the spinal nerve. BFABP also marked the dorsal neural tube and floor plate, while Ngfr was detected more prominently in the ventral neural tube. Very similar patterns of expression were observed in E12.5 Etv5-/- embryos, including prominent expression of BFABP (Fig. 4.2B,B’) and Ngfr (Fig. 4.2D,D’) in Sox10+ SCPs in the DRG and dorsal root.

Thus, at E12.5, there are no notable defects in the general positioning or SCP-specific gene expression in Etv5-/- mutant DRG, ventral, and dorsal roots.

Figure 4.2. Schwann cell non-transcription factor lineage markers are expressed normally in E12.5 Etv5-/- SCPs.

Co-expression of Sox10 (green, A-D) with BFABP (red, A,B, black/white, A’,B’) and Ngfr (red, C,D, black/white, C’,D’), counterstained with DAPI (blue, A-D) in E12.5 wild-type 151

(A,A’,C,C’) and Etv5-/- (B,B’,D,D’) transverse sections through the trunk. dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord ; vr, ventral root. Scale bars, 60 μm.

4.4.2 Immature Schwann cells develop normally in E15.5 Etv5-/- peripheral ganglia

Between E12.5 and E15.5, some SCPs persist while others proceed on to form immature Schwann cells (iSCs) that populate the developing spinal ganglia and nerves152. We focused on the lumbar spinal cord in E15.5 wild-type and Etv5-/- embryos (Fig. 4.3A,B), and again examined the expression of Sox9 (Fig. 4.3C,D) and Sox10 (Fig. 4.3E,F). Both Sox9 and Sox10 were expressed in a characteristic salt-and-pepper expression pattern in scattered SCPs and iSCs throughout the DRG and in the dorsal and ventral roots in E15.5 wild-type and Etv5-/- embryos (Fig. 4.3C-F). We also examined the expression of two additional transcription factors with a later onset of expression in SCPs, including transcription factor AP-2α (Tfap2a)250,260 and Sox2, an inhibitor of Schwann cell myelination that is expressed in SCPs and iSCs, but not in pro-myelinating and myelinating Schwann cells260,269,288. Tfap2a was expressed throughout the E15.5 wild-type DRG (Fig. 4.3G), while Sox2 was only detected in a few Sox10+ SCPs and iSCs (Fig. 4.3I,I’), contrasting to the robust expression of Sox2 in the spinal cord ventricular zone (Fig. 4.3I,I’). Notably, both Tfap2a (Fig. 4.3H) and Sox2 were similarly expressed in the Schwann cell lineage in E15.5 Etv5-/- DRGs (Fig. 4.3J,J’).

Finally, to assess the maturation process of SCPs into iSCs, we examined the expression of non-transcriptional regulators that mark the Schwann cell lineage, including BFABP (Fig. 4.3K,L,K’,L’), the intermediate filament, glial fibrillary acidic protein (GFAP) (Fig. 4.3M,M’,N,N’) and Ngfr (Fig. 4.3O,O’,P,P’). All three of these proteins were expressed in scattered Sox10+ SCPs and iSCs in both E15.5 wild-type (Fig. 4.3K,K’,M,M’,O,O’) and Etv5- /- (Fig. 4.3L,L’,N,N’,P,P’) DRGs. Thus, SCPs and iSCs develop normally in Etv5 hypomorphic mutants.

152

Figure 4.3. Schwan cell lineage markers are expressed normally in E15.5 Etv5-/- immature Schwann cells (iSCs).

Low magnification DAPI-stained images of transverse sections through the lumbar spinal cord of E15.5 wild-type (A) and Etv5-/- (B) embryos. (C-H) Expression of Sox9, Sox10 and Tfap2a in E15.5 wild-type (C,E,G) and Etv5-/- (D,F,H) transverse sections through the lumbar spinal cord. (I-P) Co-expression of Sox10 (green, I-P) with Sox2 (red, I,J, black/white, I’,J’), BFABP (red, K,L, black/white, K’,L’), GFAP (red, M,N, black/white, M’,N’), and Ngfr (red, O,P, black/white, O’,P’), counterstained with DAPI (blue, I-P) in E12.5 wild-type (I,I’,K,K’,M,M’,O,O’) and Etv5-/- (J,J’,L,L’,N,N’,P,P’) transverse sections through the lumbar spinal cord. dr, dorsal root; drg, dorsal root ganglion; sc, spinal cord; vr, ventral root. Scale bars (A,B), 100 µm; (C-P’), 60 μm.

4.4.3 Late immature Schwann cells/pro-myelinating Schwann cells are detected in E18.5 Etv5-/- peripheral ganglia and in the dorsal and ventral roots

By E18.5, some iSCs persist, while other iSCs begin to associate with large-diameter axons to become pro-myelinating Schwann cells, whereas iSCs in contact with smaller diameter axons become non-myelinating Schwann cells282. We first labelled all Schwann cells in the lineage with Sox10 in E18.5 wild-type (Fig. 4.4A) and Etv5-/- (Fig. 4B) transverse sections through the lumbar spinal cord, and detected similar numbers of Sox10+ Schwann cells in in the DRG, as well as in the dorsal and ventral roots in the wild-type and Etv5-/- mutants (Fig. 4.4I). Further, we examined whether the neuronal cells populating the DRG were affected by labelling the 153 cells with NeuN, a neuronal marker. A similar number of NeuN+ neurons were observed in both the wild-type and Etv5-/- mutant DRGs (Fig. 4.4C,D,J). Next, we examined the expression of Pou3f1 (Oct-6), which is a marker of late iSCs/pro-myelinating Schwann cells that is required for the transition to a myelinating phenotype transition286. We detected Pou3f1 expression in the ventral roots in both E18.5 wild-type (Fig. 4.4E) and Etv5-/- (Fig. 4.4F) embryos, suggesting that Sox10 cells mature to a myelinating stage. Finally, to detect satellite glial cells in the developing DRG, we examined the expression of glutamine synthetase586, revealing that this marker was expressed in both wild-type (Fig. 4.4G) and Etv5-/- (Fig. 4.4H) mutant DRGs. Thus, in Etv5 hypomorphic mutants, normal numbers of Schwann cells and DRG neurons are generated during development.

Figure 4.4. Schwan cell lineage markers are expressed normally in E18.5 Etv5-/- late immature/pro-myelinating Schwann cells.

Labelling of Sox10 (A,B), NeuN (C,D), Pou3f1 (E,F), and Glutamine Synthetase (GS; G,H) with DAPI counterstain in E18.5 wild-type (A,C,E,G) and Etv5-/- (B,D.F.H) transverse sections through the lumbar spinal cord. Quantification of percentage of Sox10+/DAPI+ cells (I) in the dorsal root ganglia, dorsal and ventral root, and NeuN+/DAPI+ cells (J) in the dorsal root ganglia. Error bars=s.e.m.. drg, dorsal root ganglion; sc, spinal cord; vr, ventral root; ns, non- significant. Scale bars, (A,B), 60 μm; (C,D,G.H), 40 µm; (E,F), 20 µm.

154

4.4.4 Schwann cells populate the postnatal sciatic nerve in Etv5-/- pups and respond to injury normally with an expansion in number

As a final question, we asked whether there were defects in Schwann cells found in the early postnatal nerve, at postnatal day (P) 21, when most pro-myelinating Schwann cells have converted to a myelinating phenotype260. At P21, we observed expression of Sox10 in scattered cells throughout longitudinal sections of the sciatic nerve in both wild-type (Fig. 4.5A) and Etv5-/- (Fig. 4.5B) animals, and there were no differences in Schwann cell numbers between these groups in the uninjured nerve (Fig. 4.5E). We then subjected both P21 wild-type and Etv5-/- animals to a sciatic nerve crush injury, which induces a de-differentiation and subsequent expansion of Schwann cells in the distal stump by 5 days post-injury (dpi)260. We observed an increase in the number of Sox10+ Schwann cells in both P21 wild-type (p=0.028; Fig. 4.5C,E) and Etv5-/- (p=0.0001; Fig. 4.5D,E) distal stumps at 5 dpi, suggesting that the repair response occurs normally in Etv5-/- nerves. Interestingly, a significant increase in Sox10+ cells was observed in Etv5-/- injured nerves compared to wild-type injured nerves (p=0.0027; Fig. 4.5E), indicating that Etv5 normally inhibits the ability of Schwann cells to expand post injury.

Taken together, these studies suggest that while the Schwann cell lineage develops normally in Etv5-/- embryos, from embryonic to early postnatal stages, the response of Etv5-/- Schwann cells to injury is altered, with an increase in Sox10+ Schwann cells populating near the nerve injury site.

155

Figure 4.5. Increase in Sox10+ Schwann cells post-injury in Etv5-/- sciatic nerve.

Sox10 expression in longitudinal sections of the uninjured (A,B) and injured (C,D) P21 sciatic nerve from wild-type (A,C) and Etv5-/- (B,D) animals. Quantification of number of Sox10+ cells (E) in the entire section. Error bars=s.e.m. *, p < 0.05; **, p < 0.01; ***, p < 0.005. Scale bars, 20 μm.

156

4.5 Discussion

In this study we investigated the role of Etv5 in Schwann cell development using a genetic mutant that deletes exons 2-5, an allele associated with defects in stem cell self-renewal in the spermatogonial lineage583. Using a panel of well annotated Schwann cell markers and quantitative studies at E18.5, we did not observe any notable defects in Schwann cell or neuronal differentiation in Etv5-/- peripheral ganglia at embryonic stages. However, following an acute peripheral nerve injury, which results in the de-differentiation of mature Schwann cells and an increase in Schwann cell proliferation151, more Sox10+ cells were observed in Etv5- /- nerves, implicating Etv5 as a negative regulator of the Schwann cell injury response. ERK1/2 signal transduction is activated downstream of NRG1419,587, which is a critical regulator of myelination, and ERK1/2 induce Schwann cell myelination419,573, including after injury 367,418,588,589. Our findings were therefore surprising as our expectation was that Etv5 would be a positive regulator of the injury response.

Notably, in our study we used an Etv5 mutant allele that has a deletion of exons 2-5 (Etv5tm1Kmm), which results in a very striking reduction in spermatogonial stem cell self- renewal583. In contrast, an Etv5 mutation that removes the DNA binding domain (Etv5tm1Hass) is embryonic lethal at E8.5, suggesting it is a true null allele583, but precluding us from analysing Schwann cells due to the early embryonic lethality. Due to the embryonic lethality associated with ‘more severe’ mutant alleles of Etv5, it is believed that our studied Etv5 allele (exon 2-5 knocked out) is hypomorphic rather than a true null allele. Further support for the designation of Etv5tm1Kmm as a hypomorphic allele comes from the lack of a developmental kidney defect590, whereas the generation of chimeric embryos using Etv5tm1Hass embryonic stem cells revealed that Etv5 is required for kidney development584. Future studies on Etv5 function in the Schwann cell lineage would require either the use of such a chimeric approach, or a genetic approach, such as the use of a floxed allele of Etv5591. However, even using these alternative approaches, we may not observe a Schwann cell developmental phenotype if there is genetic redundancy. Indeed, different members of the Fgf-synexpression group of ets transcription factors (Etv1, Etv4, Etv5) may compensate for one another to some extent, as shown for Etv4 and Etv5 in kidney development584. Thus, we may have not uncovered a role for Etv5 in developing Schwann cells due to issues of genetic redundancy. Nevertheless, despite these caveats, we can conclude that the reduced levels of Etv5 function associated with

157 the Etv5tm1Kmm allele, which has striking phenotypic consequences in other lineages, does not impact early Schwann cell development, but does impact the Schwann cell response to injury.

An important area for future studies will be to further examine the role of Etv5 in mature Schwann cells. Schwann cell transplants have the potential to aid peripheral nerve repair, and efforts are being made to improve the isolation and expansion of these cells to provide an adequate source for repair purposes. In this regard it is interesting that both human nerve- derived and skin derived Schwann cells cultured in vitro express a large number of Schwann cell markers associated with an early developmental phenotype, which includes Etv5592. In our study here we provide the first glimpse that a decline in Etv5 expression leads to more number of Schwann cells following a peripheral nerve crush injury. One possibility is that the knockdown of this factor could thus be exploited for regenerative purposes.

RTK-ERK signaling is crucial for Schwann cell differentiation572 and has been implicated in the myelination process, in part by inducing the expression of pro-myelinating transcription factors such as Yy1539. Other ets-domain transcription factors that are involved in ERK signaling are Etv1 and Etv4. While Etv1 is expressed in myelinating Schwann cells580, it is not required for Schwann cell myelination581. In this regard, it is interesting to note that dominant negative Etv5 misexpression in NCC cultures impacts neuronal fate specification, whereas glial fates are left unperturbed544. Our study similarly indicated that Etv5 is not required for generation of mature Schwann cells. In contrast, the increase in Sox10+ cell population in Etv5- /- sciatic nerve post-injury (compared to injured wild-type nerves) suggested that loss of Etv5 expression may promote Schwann cell de-differentiation post-injury. Since ERK1/2 signaling is involved in both Schwann cell myelination573 as well as in promoting a de-differentiated Schwann cell state589, the exact role of Etv5 needs to be further investigated. Questions to be addressed in the future include whether Schwann cells myelinate axons normally in Etv5-/- animals, and if Etv5 regulates Schwann cell proliferation post-injury in conjunction with Sox2 and Jun activity, which play an important role in promoting the de-differentiated repair Schwann cell phenotype151.

In summary, while our study does not support a critical role for Etv5 in Schwann cell development, we demonstrate that Etv5 is involved in regulating the repair Schwann cell response post-injury.

158

Glial transcription factor-based conversion of mouse embryonic fibroblasts to a repair Schwann cell identity

Contents of this chapter are being prepared as a manuscript:

Balakrishnan A, Belfiore L, Touahri Y, Noman H, Zinyk D, Biernaskie J, Schuurmans C. Glial transcription factor-based conversion of mouse embryonic fibroblasts to a repair Schwann cell identity. (Manuscript in preparation)

159

5.1 Abstract

Schwann cells, the major glial cells in the peripheral nervous system, provide trophic support, myelinate axons, and are required for peripheral nerve repair, but they become progressively dysfunctional post-injury. There is thus a growing need to identify a renewable source of Schwann cells with a repair phenotype to transplant into the injury site. Sox10 is a potent Schwann cell fate determinant, that when combined with small molecules can convert somatic cells to a Schwann cell-like identity. Here we further refined this transcription factor-based lineage conversion strategy with the goal of generating reprogrammed cells with a ‘repair’ Schwann cell phenotype. We established a panel of qPCR primers to assess the molecular phenotypes of mouse embryonic fibroblasts (MEFs) grown in glial supportive media after misexpression of Sox10 versus two triple combinations of transcription factors previously implicated in the repair response: Pax3-Jun-Sox2 (triple 1, or T1) and Sox10-Jun-Sox2 (triple 2, or T2). We found that Sox10-Jun-Sox2 (T2) was more efficient than Sox10 and Pax3-Jun- Sox2 (T1) at inducing the expression of embryonic and repair Schwann cell markers in whole cell cultures. We further confirmed the Schwann cell-like phenotype of ‘reprogrammed’ cells by demonstrating that transfected cells expressed GAP43, Ngfr and O4 proteins, employing the latter two as cell surface markers for FACS sorting. Strikingly, qPCR analyses of Ngfr+O4+ FACS-sorted Schwann-like cells yielded different results, with Pax3-Jun-Sox2 (T1) inducing the highest levels of Schwann cell gene expression. Thus, while Sox10-Jun-Sox2 may generate more cells expressing Schwann cell markers, Pax3-Jun-Sox2 is better able to induce elevated expression of Schwann cell markers in those cells that acquire a Schwann cell-like phenotype. The next step will be to test the functional capacity of ‘reprogrammed’ cells using in vitro and in vivo myelination assays. Towards, this goal, we established a dorsal root ganglia (DRG)- Schwann cell co-culture system for future experiments to assess whether lineage-converted cells are functional and have a true Schwann cell identity with enhanced repair capacity.

160

5.2 Introduction

A commonly held misconception is that because neuronal regeneration occurs in the peripheral nervous system (PNS), cell-based repair-strategies are not required. Contrary to this belief, peripheral nerve injury (PNI), which occurs in nearly 3% of all trauma cases, can lead to permanent deficits in sensorimotor function and cause neuropathic pain593. The current standard of care for PNI is a nerve graft, but ‘success’ rates are low, with only ~25% of patients recovering full motor function and 3% regaining sensory function368. While part of this failure is due to poor engraftment, the ability of Schwann cells to remyelinate nerves is limited, especially in a chronic injury scenario372. The ability to transplant Schwann cells with enhanced repair capacity could thus pave the way for newer treatment options. Indeed, Schwann cell transplants at the site of injury improve functional recovery in animal models441, but a renewable Schwann cell source will be required for clinical applications.

Schwann cells de-differentiate in response to PNI, undergoing a series of state changes that partly mimic developmental stages260, which prepares these glial cells to contribute to axonal repair and regeneration151,352. Loss of axonal contact occurs post-injury and triggers mature myelinating Schwann cells to downregulate myelin genes and acquire an immature Schwann cell-like phenotype260,352,552. However, the repair capacity of these de-differentiated Schwann cells declines with time post injury, increased age, and pre-existing ailments326,524. Thus, other sources of Schwann cells are being explored for transplant purposes. Schwann cells have previously been differentiated from a multipotent precursor cell population residing under the adult skin dermis, known as skin-derived precursor cells (SKPs)435. However, the differentiation protocol for generating SKP-derived Schwann cells requires approximately 5-6 weeks435, which may be too long to be clinically useful for autologous transplant. To hasten the process, direct lineage conversion approaches based on small molecules (e.g. Tgf and GSK3 inhibitors + neuregulin 1, or Nrg1) have also been used to generate Schwann cells from pluripotent stem cells468. In addition, transcription factors (e.g. Sex determining region Y (SRY)- box transcription factor; Sox10) have been combined with small molecules (GSK3 and HDAC inhibitors) to reprogram fibroblasts into neural crest cells (NCCs) that can then be differentiated into Schwann cells in response to Nrg1, cAMP and ciliary neurotrophic factor (CNTF)468. In addition, mouse embryonic fibroblasts (MEFs) have been converted to Schwann cell-like cells by misexpression of Sox10 and Early growth response 2 (Egr2), combined with forskolin and Nrg1, glial growth signals470,471.

161

Our goal was to establish a similar direct reprogramming approach, but rather than making a generic Schwann cell, we set out to devise a strategy that would generate Schwann cells with a repair phenotype. We reasoned that lineage conversion of somatic cells into ‘repair-like’ Schwann cells would require re-activation of a precise array of embryonic and repair specific glial-lineage genes. We focused on four transcription factors; Sox10, based on its role as an essential Schwann cell fate determinant, and three transcription factors that are expressed in repair Schwann cells and implicated in the transient acquisition of a repair phenotype post- injury; Sox2, Jun proto-oncogene (Jun), and Paired box 3 (Pax3)151,260,312,326,327,594,595. Sox10 is expressed in NCCs and in embryonic and mature Schwann cells and is required for Schwann cell development248,263. Sox10 mutant mice fail to generate Schwann cells, leading to dorsal root ganglion (DRG) neuronal degeneration due to the loss of glial support310. Sox2 is expressed embryonically in Schwann cell precursors (SCPs) and immature Schwann cells (iSCs)300, and increased expression of Sox2 inhibits myelination and promotes proliferation of mature Schwann cells, to promote a repair Schwann cell phenotype321. Pax3 is expressed in NCCs, SCPs, and non-myelinating Schwann cells, and Pax3 transcripts are upregulated post injury in repair Schwann cells151,290,291. Pax3 induces Schwann cell proliferation and inhibits myelin gene expression291. Similarly, expression of Jun is observed in late iSCs/pro-myelinating Schwann cells and is rapidly upregulated in Schwann cells post injury324,326. Jun inhibits myelination and induces de-differentiation of mature Schwann cells324,326.

In this study we found that two triple transcription factor combinations, Pax3-Jun-Sox2 (triple 1, or T1) and Sox10-Jun-Sox2 (triple 2, or T2), induced Schwann cell gene expression at higher levels than Sox10 alone when misexpressed in MEFs and cultured in glial supportive media (Nrg1, forskolin). However, while Sox10-Jun-Sox2 (T2) induced the formation of more Schwann-like cells, FACS-purified Ngfr+O4+ cells misexpressing Pax3-Jun-Sox2 (T1) expressed the highest levels of Schwann cell markers. Finally, we also established a DRG- Schwann cell co-culture assay that will be used for future studies to assess the myelination capacity of these altered cells. In sum, we have acquired encouraging data to support the notion that it may be possible to generate repair Schwann cell-like cells in a dish.

162

5.3 Material and methods

5.3.1 Vector Construction

Triple transcription factor expression vectors (triple 1 – T1, and triple 2- T2) were generated by cloning transcription factor coding domains (Sox2, Pax3, Sox10, Jun) into a pCAGGS expression vector (Lablife) modified by the addition of a multiple cloning site (pCAGGS2). In T1 and T2, transcription factors at the first, second, and third position were Myc-, HA-, and FLAG-tagged, respectively. A viral 2A (v2A) self-cleaving peptide sequence was inserted between transcription factor sequences to allow for stoichiometric expression of all three genes 596. Final products were pCAGGS2-Jun-Myc-v2a-Pax3-HA-v2a-Sox2-FLAG and pCAGGS2- Sox10-Myc-v2a-Jun-HA-v2a-Sox2-FLAG. pCIG2-Sox10-IRES-GFP and pCIC:Sox10-IRES- mCherry were generated by cloning Sox10 coding domain into pCIG2 and pCIC expression vectors 597.

5.3.2 Animals and genotyping

Animals used in this study included CD1 mice (Charles River Laboratories, Senneville, QC), and two transgenic mouse lines: Sox10-Cre (B6;CBA-Tg(Sox10-cre)1Wdr/J; #025807), and Rosa-tdTomato (B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J; #007914) (Jackson Laboratory, ME, United States). Sox10-Cre x Rosa-tdTomato intercrosses were used to generate Sox10-Cre;Rosa-tdTomato progeny for experimental purposes. For embryonic staging, the morning of the vaginal plug as embryonic day (E) 0.5. Animal procedures were approved by the Sunnybrook Research Institute Animal Care Committee (protocol #AUP-16- 606) in compliance with the Guidelines of the Canadian Council of Animal Care. PCR genotyping was performed with the following primers: Sox10-Cre; mutant forward primer: ACC AGG TTC GTT CAC TCA TGG; mutant reverse primer: AGG CTA AGT GCC TTC TCT ACA C. Rosa-tdTomato; mutant forward primer: CTG TTC CTG TAC GGC ATG G; mutant reverse primer: GGC ATT AAA GCA GCG TAT CC.

5.3.3 Cell culture and cell isolation

NIH/3T3 cells (ATCC® CRL-1658™) and MEFs were cultured in Dulbecco’s modified Eagle medium (DMEM; Wisent #319-005-CL) containing 10% fetal bovine serum (FBS; Wisent #080-450) and penicillin/streptomycin (Wisent #450-201-EL). MEFs were isolated from E13.5 CD1 embryos. Briefly, embryos were decapitated and internal organs along with the vertebral

163 column were discarded. Remaining tissue was finely minced and dissociated in 2% Collagenase Type 4 (Worthington #LS004189) for 20mins. MEFs were maintained for a maximum of four cell passages. Mouse nerve-derived Schwann cells were isolated as described previously435. Briefly, sciatic nerves from postnatal day (P) 21 CD1 mice were dissected, minced into small pieces and dissociated in 2% collagenase. Schwann cells were maintained in Schwann cell media (called glial media): DMEM: Ham's F-12 Nutrient Mix with GlutaMAX (F12; Invitrogen #31765-035) (3:1), FGF2 (10ng/ml, Miltenyi Biotech GMBH #130-093-842), Forskolin (4 µM, Sigma #F8668), Nrg1 (100ng/ml, R&D Biosystems #377-HB), N2 supplement (1%, Wisent 305-016- IL), 1X L-glutamine (Invitrogen), and penicillin/streptomycin. Cells were cultured in 10cm dishes (Sarstedt) coated with Poly-D-Lysine (PDL; 20µg/mL) and Laminin (Lam; 40µg/mL) (PDL; VWR-BD #354210, Lam; VWR-BD #354232).

5.3.4 Cell Transfection

NIH/3T3 cells were transfected using a Lipofectamine 3000 kit (Thermo Fisher Scientific) as per the manufacturer’s instructions. MEFs were transfected using the 4D-Nucleofector™ (Lonza). Briefly, 2×106 cells were mixed in 20µL P3 reagent and transfected with 13µg plasmid DNA using the CZ-167 program on the nucleofector. Transfected cells were plated on PDL:Lam pre-coated dishes and cultured in MEF media for 24hr and then switched to glial media. Cells were fed with fresh media every 3 days.

5.3.5 Dorsal root ganglia-Schwann cell co-culture assay

The in vitro myelination assay was performed as previously described435. Briefly, DRGs were isolated from P2 CD1 pups and grown as explants on 8-well chamber slides coated with Matrigel (1:20; BD Biosciences), PDL, Lam, Collagen IV (1:20; BD Biosciences). Explants were grown in DRG media: DMEM:F12 (3:1) with Nrg1 (50ng/mL; R&D Biosystems), Forskolin (5µM; Sigma Aldridge) and Nerve Growth Factor (NGF; 50ng/mL; R&D Biosystems). After 2 days in DRG media, cytosine arabinoside (CyaA;Sigma Aldrich) was added for four days (~2-3 mitotic cycles) to kill surviving Schwann cells and fibroblasts (i.e., dividing cells). Fresh DRG media was then added and refreshed every 2 days. One week post- dissection, DRG explants were co-cultured with 20,000-25,000 Schwann cells in myelination media: DMEM:F12 with Nrg1 (5ng/mL), Forskolin (0.5µM), Nerve growth factor(NGF; 5ng/mL) and Ascorbic Acid (50µg/mL; R&D Biosystems).

164

5.3.6 Fluorescence-Activated Cell Sorting (FACs)

To isolate Ngfr negative MEFs, four days after the establishment of primary MEF cultures, dissociated cells were incubated with CD271-VioBright FITC antibody (Miltenyi Biotech; ME20.4-1.H4; 1:500) for 30mins. Stained cells were sorted on BD FACS ARIA III and CD271- cells were isolated and cultured in MEF media. To isolate Ngfr+ O4+ reprogrammed cells, dissociated cells were incubated with CD271-VioBright FITC and O4-APC (Miltenyi Biotech; 130-119-155; 1:500) antibodies. Stained cells were sorted on BD FACS ARIA III and CD271+ O4+ cells were collected and processed for RNA isolation.

5.3.7 Western blotting

Whole cell pellets were collected post transfection and lysed in RIPA buffer (150 mM NaCl, 0.1% sodium dodecyl sulphate (SDS), 1% NP-40, 50 mM Tris-HCl, pH 8.0, 0.5% sodium deoxycholate) with protease (1X protease inhibitor complete, 1mM phenyl methyl sulfonyl fluoride (PMSF) and phosphatase (50mM NaF, 1mM NaOV) inhibitors. 10µg of lysate was run on 10% SDS-PAGE gels for Western blot analysis. Primary antibodies included: Jun (Abcam; #ab31419, 1:1000), Pax3 (DSHB; #p23760, 1:1000), Sox2 (Cell Signaling; #C70B1, 1:1000), Sox10 (Santa Cruz Biotech; #17343, 1:1000), and GAPDH (Cell Signaling #2118, 1:10000). Western blots were developed with an ECL Blotting Substrate (Bio-Rad, Canada) according to the manufacturer’s protocol.

5.3.8 Quantitative reverse transcription polymerase chain reaction (RTqPCR)

RNA extraction was performed using the RNeasy Mini and Micro kit (Qiagen) as per the manufacturer’s instructions. 1µg RNA was reverse transcribed using the RT2 First Strand synthesis kit (Qiagen). Reverse transcribed cDNA was amplified using the RT2 primer assay kit and Sybr Green (Qiagen). Qiagen primers used for gene expression analysis were: Cdh19 (PPM40369A), Col1a1 (PPM03845F), Egr1 (PPM02938C), Egr2 (PPM04478F), Erbb3 (PPM05139A), Gap43 (PPM03303A), Gfap (PPM04716A), Jun (PPM03037A), MPZ (PPM04745F), Olig1 (PPM04503A), p75NTR (or Ngfr), Pax3 (PPM24658A), Pou3f1 (PPM38669B), S100 (PPM04492A), Sox2 (PPM04762E), Sox10 (PPM05134D), Shh (PPM04516C), Tfap2a (PPM30132A), and Thyl (PPM31015A). Gapdh (PPM02946E), B2m (PPM03562A), and Hrpt (PPM03559F) were used as reference genes. The delta-delta Ct method was used to calculate relative expression levels for each gene, using Gapdh, B2M, and Hrpt for normalization. Gene expression levels post-transfection were normalized to levels in

165 untransfected MEFs. A minimum of 3 experimental and 3 technical replicates was performed for each experiment. Statistical comparisons were made using an ANOVA with a post-hoc Tukey’s test using Prism software (GraphPad). Data are represented as mean ± standard error of the mean (s.e.m.). p values are denoted as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001

5.3.9 Tissue processing and Immunofluorescence assay

Transfected cells plated on 6 well chamber slides, and DRG-Schwann cell culture explants were washed with phosphate-buffered saline (PBS), and fixed in 4% paraformaldehyde (PFA)/PBS. Embryos were dissected in cold PBS, and fixed in 4% PFA, overnight. Fixed embryos were then transferred to 20% sucrose/PBS overnight. Embryos were blocked in O.C.T™ (Tissue-Tek®, Sakura Finetek U.S.A. Inc., Torrance, CA) and stored at -80C. 10 µm sections were collected by sectioning tissues on a Leica cm3050s cryostat (Richmond Hill, ON) and collected on SuperFrost™ Plus slides (Thermo Scientific). Samples were washed and permeabilized in PBS /0.1% TritonX-100 (PBT) followed by blocking with 10% normal horse serum/PBT (blocking solution). Primary antibodies were diluted in blocking solution and were incubated on sections and kept overnight at 4C. Sections were washed with PBT and secondary antibody diluted in PBT was applied. Sections were counterstained with 4′,6- diamidino-2-phenylindole (DAPI; Santa Cruz Biotechnology) diluted in PBT and mounted with coverslips using AquaPolymount (Polysciences). Primary antibodies included: NGF Receptor p75 (Ngfr, Millipore #AB1554; 1:500); GAP43 (Novus Biologicals; NB300-143, 1:300), tdTomato (Abcam; #ab62341; 1:500), Tuj1 (BioLegend #801202; 1:200). Species- specific secondary antibodies were conjugated to Alexa 568 or Alexa 488 (Invitrogen). Images were captured with a Leica DFC7000T camera and LASX software and processed using Adobe Photoshop.

5.4 Results

5.4.1 Establishing a panel of markers to assess lineage conversion

The goal of this study was to generate a renewable source of repair Schwann cells that could ultimately be translated to the clinic and aid current practices for the treatment of PNI. To achieve this goal, we devised a lineage conversion strategy that was based on the misexpression of Schwann cell repair-related transcription factors in a heterologous cell type (MEFs) (Fig. 5.1A).

166

167

Figure 5.1. Establishing a qPCR primer panel to test Schwann cell lineage conversion.

(A) Schematic of lineage conversion strategy for converting fibroblasts to Schwann cells that are characterized at the molecular level before being transplanted into a nerve injury model to assess functional recovery. (B) Establishment of a qPCR primer panel for the analysis of Schwann cell reprogramming, tested in adult nerve-derived Schwann cells cultured in vitro (white bars) and in MEFs (grey bars). Relative expression of transcripts measured against housekeeping genes. Bars represent means ± s.e.m. (N=3)

To determine whether our cellular reprogramming strategy was effective, it was necessary to identify Schwann cells at various stages of lineage development. For this purpose, we established a panel of 17 qPCR primers that covered the developmental trajectory of Schwann cells, including markers that were expressed in NCCs, SCPs, iSCs, pro-myelinating, myelinating and non-myelinating Schwann cells260. This panel included Sox10, transcription factor Ap2alpha (Tfap2a), paired homeobox 3 (Pax3), Sox2, Egr1, Jun, glial fibrillary acidic protein (Gfap), S100 protein, beta polypeptide (S100), nerve growth factor receptor (Ngfr), Early growth response 2 (Egr2), Myelin protein zero (MPZ), POU domain, class 3, transcription factor 1 (Pou3f1), Cadherin 19 (Cdh19), Growth associated protein 43 (GAP43), Erb-b2 receptor tyrosine kinase 3 (Erbb3), sonic hedgehog (Shh), and oligodendrocyte transcription factor 1 (Olig1)212,258,260 (Fig. 5.1B). To have a baseline for comparison, we first tested the expression of these Schwann cell lineage genes in adult peripheral nerve-derived Schwann cells that were expanded and cultured in vitro. We also tested the specificity of these markers by examining their expression in MEFs, which was our starting cell population for lineage conversion (Fig. 5.1B).

Of the 17 genes in the Schwann cell panel, only Sox10, Egr1, S100, MPZ, Cdh19 and Erbb3 were expressed at high levels in cultured nerve-derived Schwann cells (Fig. 5.1B). Of note, Schwann cells cultured in vitro lose axonal contacts and revert to a de-differentiated state151, explaining why some mature myelinating genes were either not expressed (Egr2) or expressed at modest levels (MPZ) (Fig. 5.1B). In addition, because adult Schwann cells de-differentiate in culture, some embryonic Schwann cell markers were also expressed in the nerve-derived Schwann cells, including markers of SCPs (Tfap2a, Ngfr), iSCs (Sox2, Jun) and pro- myelinating Schwann cells (Pou3f1). In contrast, other embryonic Schwann cell markers

168

(GFAP, GAP43) and markers associated with a Schwann cell repair phenotype (Pax3, Shh, Olig1) were not expressed (Fig. 5.1B)151,324.

As we were using MEFs for lineage conversion, it was also necessary to determine whether any of these Schwann cell markers were expressed in our starting population of cells. Of the 17 markers selected, only two were detected at detectable levels in MEFs - Egr1 and Jun1 (Fig. 5.1B), which we considered for all our analyses moving forward. In summary, we identified baseline marker expression profile in our starting (MEF) and target (Schwann cell) cell types, and applied this full marker set to assess lineage conversion, and especially the acquisition of a repair phenotype, in the remaining assays.

5.4.2 Mis-expression of Sox10 induces Schwann cell specific marker expression in fibroblasts

For cellular reprogramming, we chose a plasmid-based approach instead of more traditional viral vectors as the eventual expulsion of non-replicating vectors means that plasmid- and virus-free cells could ultimately be used safely for transplantation. E13.5 MEFs were used as the starting cell population for lineage conversion (Fig 5.2A), as MEFs are easily accessible and have been used in several other reprogramming studies, including to make induced pluripotent stem cells (iPSCs)598 and Schwann cells470. Using this system, we first examined whether Sox10, which had previously been demonstrated to induce a NCC fate with Schwann cell differentiation potential468, could directly induce a Schwann cell-like fate when combined with Schwann cell supportive media containing Nrg1 and forskolin (hereafter called glial media435; Fig. 5.2A).

MEF cultures were transfected with mCherry (control) and Sox10-mCherry expression constructs (hereafter, Sox10; Fig. 5.2A). Transfection efficiency was monitored one day post- transfection by imaging mCherry epifluorescence and using phase contrast to outline the cells. We observed a relatively high transfection efficiency (~60-65%) for both control and Sox10 constructs after one day post transfection (Fig 5.2B-E). To monitor plasmid dilution due to cell division, we also imaged cells after 3, 5, and 7 days post-transfection, observing a clear decline in mCherry+ cells from day 3 onwards (Fig 5.2F). The decline in expression plasmid was confirmed by qPCR analysis of Sox10 transcripts, revealing a 12-14-fold decline between 3-7 days post-transfection compared to day 1 post transfection (Fig 5.2G).

169

Even though Sox10 expression declined over time, we reasoned that it could induce a more permanent change in endogenous gene expression that could be indicative of at least initial stages of a fate switch. Control and Sox10 transfected cells were thus cultured in glial media for 14 DIV before harvesting to assess our panel of Schwann cell markers by qPCR (Fig. 5.1B). All qPCR values were normalized to gene expression levels in untransfected MEFs to show the effects of Sox10 versus control transfections in the Schwann cell induction media (using negative control vector). Compared to the control transfected cells, Sox10, Tfap2a, Sox2, S100, Egr2 and Pou3f1 were expressed at higher levels in Sox10-transfected cells after 14 DIV (Fig. 5.2H). Pax3, Sox2 and to a lesser extent Egr2 and Pou3f1 were also induced in the control-transfected cells, suggesting they were induced by Nrg1 and forskolin in the culture media (Fig. 5.2H).

Another measure of the conversion of fibroblasts to a new cell type is the downregulation of fibroblast-specific genes, such as Thymus cell antigen 1, theta (Thy1) and Collagen, type I, alpha 1 (Col1a1)470. After 14 days post-transfection, both Thy1 and Col1a1 were expressed at reduced levels in Sox10- and control-transfected cells grown in Schwann cell media compared to MEFs grown in MEF media (Fig. 5.2I). Moreover, the reduction in fibroblast gene expression was ~ 10-fold in Sox10 transfected MEFs, compared to ~2-fold in control MEFs (Fig 5.2I).

In summary, mis-expression of Sox10 together in MEFs grown in the presence of Schwann cell growth factors induces the expression of Schwann cell markers while reducing fibroblast gene expression. However, before moving forward, it was necessary to optimize our cell source as we could not rule out the possibility that our starting MEF cultures were contaminated with a small number of NCCs and SCPs that could more readily differentiate into Schwann cells in the presence of glial supporting cues (forskolin, Nrg1). Indeed, a previous report demonstrated that ~0.25% of dissociated cells in ‘pure’ MEF cultures were Ngfr+ NCCs and SCPs470.

170

171

Figure 5.2. Overexpression of Sox10 in MEFs induces embryonic Schwann cell markers.

(A) Schematic of the lineage conversion strategy, involving the transfection of E13.5 MEFs mCherry (negative control) and Sox10-mCherry constructs, followed by growth in glial induction media for 14 DIV. (B-E) Comparing the transfection efficiency of MEFs transfected with mCherry (negative control, B,C) and Sox10-mCherry (D,E) after 1 day in vitro. (F) MEFs transfected with Sox10-mCherry (D,E) after 3, 5 and 7 days in vitro, showing the decline in mCherry+ cells. (G) qPCR analysis of Sox10 expression levels in MEFs transfected with Sox10, and analyzed after 1, 3, 5, and 7 days post transfection. (H,I) qPCR analysis of MEFs 14 days post-transfection with negative control (black bars) and Sox10 (green bars), showing relative transcript levels (normalized to reference genes) for Sox10, Tfap2a, Pax3,Sox2, Egr1, Jun, Gfap, S100b, Ngfr, Egr2, MPZ, and Pou3f1 (H) and fibroblast specific genes, Thy1 and Col1a1 (I). Bars represent means ± s.e.m.

5.4.3 Design of triple transcription factor expression vectors for induction of a Schwann cell identity in fibroblasts

To ensure that any Schwann cell gene expression we detected was not due to contamination of primary MEF cultures with residual NCCs or SCPs that picked up our expression vectors, we performed a negative FACS sort, removing all Ngfr+ cells (NCCs, SCPs260) from primary dissociated MEFs (Fig 5.3A). All lineage conversion studies described hereafter were then performed on Ngfr- MEFs.

We previously characterized the molecular signature of ‘repair’ Schwann cells post-injury260, and found that they resemble a mixed early/late embryonic phenotype. We thus reasoned that the regenerative function of reprogrammed Schwann cells could be maximized by re- expressing a set of embryonic glial-lineage genes. We focused on four transcription factors (TFs) expressed in embryonic SCPs, iSCs, and at early stages of repair post-PNI (Sox10, Sox2, Pax3, Jun)151,260,321,324. These factors were combined in two ways to generate triple TF expression vectors; Pax3-Jun-Sox2 (Triple 1, or T1) Sox10-Jun-Sox2 (Triple 2, or T2) (Fig. 5.3B). Sox2 and Jun were kept in common in both triple vectors given their importance in defining a repair Schwann cell identity151,321,326. To drive expression, we used the same CAG promoter/enhancer (as with Sox10 alone), but here we separated each cistron with a viral 2A (v2A) self-cleaving peptide to allow stoichiometric expression of each gene596. We verified

172 that T1 and T2 drove expression of the corresponding three TFs by transfecting NIH/3T3 cells followed by Western blotting 24 hr after transfection (Fig 5.3C). Sox2 and Jun were both detected in T1 and T2 transfected cells, while Pax3 and Sox10 were detected only in T1 and T2 transfected cells, respectively. Notably, Jun was expressed in NIH/3T3 cells at relatively high levels, so we could not conclude definitively that it was also expressed from the triple TF vectors, but given the robust expression of the other proteins in the constructs, we continued on with the idea that Jun was expressed, whether from the construct or endogenously.

We next assessed the longer-term effects of misexpressing Sox10 alone versus T1 and T2 in Ngfr- MEFs after 14 and 21 days, first examining expression of the constituent factors, Sox10,

Sox2, Pax3, and Jun. One day post-transfection, Ngfr- MEFs were cultured in glial media supplemented with Nrg1 and forskolin435 (Fig 5.3A). Compared to empty vector control, Sox10 transcript levels were elevated in MEFs transfected with Sox10 alone or T2 (which contains Sox10) after 14 and 21 DIV (Fig 5.3D). Notably, Sox10 levels declined over time in the Sox10- transfected cultures, likely due to plasmid dilution, whereas Sox10 levels remained elevated in T2-transfected cells, possibly indicative of endogenous Sox10 expression being induced. Similarly, Sox2, contained within T1 and T2 expression vectors, was expressed in T1- and T2- transfected MEFs after 14 and 21 DIV, and generally persisted despite plasmid dilution (Fig. 5.3E). Notably, Sox2 was not induced by the glial media (negative control) or by Sox10 misexpression (Fig. 5.3E). In contrast, Pax3 expression was elevated by glial media alone, especially after 21 days in vitro, with a further elevation in Pax3 transcripts only observed in T1 transfections (contains Pax3) (Fig. 5.3F). Finally, our analyses of Jun expression were complicated by the high levels of Jun in MEFs. Only T2 transfections elevated Jun levels above baseline (set at 1 based on levels in non-transfected MEFs) after 21 DIV, even though Jun was also part of T1 (Fig 5.3G).

In summary, we confirmed that plasmid-based transfection of MEFs can drive the expression of our TF combinations, with the caveat that even though Jun expression could not be directly attributed to the vector, it was expressed, allowing us to examine the fate of cells expressing these transcription factor combinations in more detail.

173

Figure 5.3. Validating triple transcription expression vectors, T1 and T2.

(A) Schematic of the lineage conversion strategy, involving the isolation of E13.5 MEFs, negative FACS sort to remove Ngfr+ cells, transfection with mCherry (negative control), Sox10, T1 and T2 constructs, and growth in glial induction media for 14 DIV and 21 DIV. (B) Triple transcription factor expression vectors: Pax3-Jun-Sox2 (triple 1, or T1) and Sox10-Jun- Sox2 (triple 2, or T2). (C) Expression of transcription factors in T1- and T2-transfected cells assessed using Western blotting. GAPDH served as a loading control. (D-G) Relative transcript expression of Sox10 (D), Sox2 (E), Pax3 (F), and Jun (G) in MEFs transfected with negative control vector (neg), Sox10, T1 and T2 after 14 DIV and 21 DIV. Relative expression of transcripts measured against the expression of transcripts in un-transfected MEFs cultured in MEF media. Bars represent means ± s.e.m. (N=3). *, p < 0.05; **, p < 0.01; ***, p < 0.001 compared to the results obtained in Neg MEFs (one-way ANOVA).

174

5.4.4 Sox10-Jun-Sox2 (T2) combined with glial media induces expression of embryonic and repair specific Schwann cell markers

We next asked whether Ngfr- MEFs transfected with Sox10, T1 or T2 and grown in glial- supportive media for 21 DIV expressed other Schwann cell lineage markers (Fig. 5.4A). The same qPCR marker panel assessed in nerve-derived Schwann cells was examined (Fig. 5.1B). To identify gene expression changes due to the glial media alone, we normalized all expression values to those observed in MEFs cultured in the absence of glial cues (set at 1; (t)). Thus, all bars shown above one indicate that gene expression was induced by glial cues in the media. In addition, we made comparisons between the negative control vector and all other constructs to distinguish transcription factor-induced changes.

We assessed marker expression 21 days post transfection and growth of MEFs in glial media (Fig. 5.4A). Several Schwann cell markers (Tfap2a, Olig1, Ngfr, Erbb3, Shh, Gap43, Gfap, MPZ, Pou3f1), were expressed at relative levels above 1 in the negative control after 21 DIV, indicating that the glial cell media could induce Schwann cell marker expression in MEFs (Fig. 5.4B-P). In addition, seven Schwann cell lineage genes were upregulated even further in T2- transfected MEFs compared to negative controls after 21 DIV, including transcription factors (Tfap2a, Olig1), signaling molecules (Ngfr, Errb3, Shh, Gap43), and a structural molecule (Gfap) (Fig. 5.4B-H). In contrast, Cdh19, S100, Egr1, Egr2, MPZ and Pou3f1 were not induced by T2, and no Schwann cell lineage markers were induced by Sox10 or T1 above baseline levels (negative control) (Fig. 5.4I-N). Notably, for all genes that were induced by the various transcription factors and glial media, most did not reach the levels of expression seen in nerve-derived Schwann cells (Fig. 5.4B-E,I-K,M,N). The exceptions were Olig1, Shh, Gap43, Gfap and Egr2 (Fig. 5.4C,F-H,L), which were expressed in treated MEFs above levels detected in nerve-derived Schwann cells, including in negative control plasmid transfections, highlighting their induction by the glial media. Finally, a significant decline in expression of fibroblast markers Thy1 and Col1a1 was detected in all four tested conditions (negative control, Sox10, T1, T2) compared to MEFs cultured in the absence of glial cues (Fig 5.4O,P). Thus, exposure to glial cues alone can suppress the expression of fibroblast specific genes in MEFs.

175

Figure 5.4. Schwann cell-specific markers induced by Sox10, T1 and T2 misexpression in MEFs. (A) Schematic of the lineage conversion strategy, involving the isolation of E13.5 MEFs, negative FACS sort to remove Ngfr+ cells, transfection with mCherry (negative control) and Sox10, T1 and T2 constructs, and growth in glial induction media for 21 DIV. A schematic of T1 and T2 factors is also shown below. (B-P) Relative expression levels (normalized to MEFs) of Tfap2a (B), Olig1 (C), Ngfr (D), Erbb3 (E), Shh (F), Gap43 (G), Gfap (H), Cdh19 (I), S100 (J), Egr1(K), Egr2 (L), MPZ (M), Pou3f1 (N), Thy1 (O), and Col1a1 (P) in transfected cells after 21 DIV. Bars represent

176 means ± s.e.m. (N=3). *, p < 0.05; **, p < 0.01; ***, p < 0.001 relative to the results obtained in Neg MEFs at day 21 (one-way ANOVA).

While the analysis of transcript levels is a good indication of cell fate changes, true lineage conversion should also be accompanied by a change in protein expression levels. We thus performed immunostaining for a few markers, including GAP43 and Ngfr (Fig 5.5A-Q). Interestingly, compared to negative vector controls, GAP43 protein was detected in more cells in Sox10, T1 and T2-transfected MEFs after 21 DIV, even though Gap43 transcripts were only elevated in T2-transfected MEFs (Fig 5.5B-I). In contrast, Ngfr was expressed in a few scattered cells in all conditions, including negative control, Sox10, T1 and T2-transfected MEFs after 21 DIV, suggestive of induction by the glial media alone (Fig 5.5J-Q).

Our analysis of protein expression thus suggests that misexpression of Schwann cell transcription factors in the presence of glial stimulatory cues may influence marker expression not only at the transcriptional level, but also at the translational level. We therefore studied the fate of transfected cells that had turned on Schwann cell proteins in each condition in more detail.

177

Figure 5.5. Schwann cell-specific proteins induced by Sox10, T1 and T2 misexpression in MEFs.

(A) Schematic of the lineage conversion strategy, involving the isolation of E13.5 MEFs, negative FACS sort to remove Ngfr+ cells, transfection with mCherry (negative control) and Sox10, T1 and T2 constructs, and growth in glial induction media for 21 DIV. Immunostaining performed at 21 DIV. A schematic of T1 and T2 factors is also shown. (B-Q) Immunostaining of GAP43 (B-I) and Ngfr (J-Q) on MEFs transfected with Sox10, T1, and T2 after 21 days in vitro. DAPI counterstain in blue. (Scale bar=100µm).

178

5.4.5 Ngfr+O4+ Schwann-like cells present with improved Schwann cell specific marker transcript expression

We reasoned that the relatively low levels of Schwann cell gene expression induced by Sox10, T1 and T2 in MEFs (Fig. 5.4) might not reflect the true effects of these gene combinations on Schwann cell marker expression, and could arise due to a masking effect of non-transformed cells and cells that diluted out plasmids before initiating endogenous gene expression programs. Hence, to better understand the effects of misexpressed TFs on Schwann cell gene expression, 21 days post-transfection of Ngfr- MEFs with Sox10, T1 and T2, we FACS sorted cells that had entered the Schwann cell lineage based on the expression of Ngfr and O4 (Fig. 5.6A). Notably, the same cell surface markers were used in a previous Schwann cell lineage conversion study470. Ngfr is expressed in NCCs and developing Schwann cells, and declines in expression with the onset of myelination349,352. O4 is expressed in immature and mature Schwann cells, but is absent from SCPs599. Together Ngfr and O4 capture Schwann cell lineage cells at different developmental stages470.

RNA was isolated from Ngfr+ O4+ sorted cells, followed by analysis of the expression of the transcription factors in our vectors; Sox10, Sox2, Pax3, and Jun. In Ngfr+ O4+ sorted cells, Sox10, a marker of NCC and Schwann cells at all developmental stages, was elevated above negative control levels after misexpression of Sox10, T1 and T2, reaching levels seen in nerve- derived SCs (Fig. 5.6B). Surprisingly, Sox10 expression levels were highest in T1-transfected cells compared to Sox10 and T2, even though Sox10 was not part of this triple vector, indicative of a change in endogenous gene expression. Sox2 was also expressed above negative control levels in Ngfr+ O4+ sorted cells isolated from Sox10, T1 and T2 transfected MEFs, above levels of nerve-derived Schwann cells, and again, T1-transfections achieved the highest Sox2 expression levels (Fig. 5.6C). A similar observation was made with Jun, except that Jun was also induced by glial media, and was also high in negative control cells (Fig. 5.6D). Finally, Pax3, only in T1, was expressed at elevated levels in Sox10, T1 and T2 transfected MEFs, above levels seen in negative vector controls, Sox10, T2, and cultured nerve-derived Schwann cells (Fig. 5.6E). Thus, even 21 days post-transfection and after FACS sorting, these critical repair genes were expressed in the transfected MEFs, even when not part of the individual constructs.

179

We next analyzed the remaining markers in our panel of Schwann cell markers in Ngfr+ O4+ sorted cells isolated 21 days after transfection with Sox10, T1 and T2. Our first comparison was against MEFs, to determine whether there was an induction of gene expression above baseline (set at 1), be it from the glial media or TFs. We found that Ngfr, Tfap2a, Cdh19, and Erbb3 were induced by all three constructs above negative control levels, and in the range of normal expression levels observed in nerve-derived Schwann cells (Fig. 5.6F-I). An induction above negative control levels was also seen for S100, MPZ and Pou3f1, but not to the levels observed in nerve-derived Schwann cells (Fig. 5.6J-L). In contrast, genes such as Gap43 were also expressed above baseline and negative control levels, suggesting that their induction was related to both the glial media and TFs (Fig. 5.6M), and Gap43 was expressed in T1 and T2 cells above the levels expressed in nerve-derived Schwann cells. Other markers also induced by glial media included Shh, Gfap, Egr1 and Egr2, but these genes were all expressed at relatively low levels (Fig. 5.6N-Q). Notably, Olig1, a repair Schwann cell-specific gene151,324, was induced in negative control cells by glial media alone, but not due to TF misexpression in Sox10, T1,T2 transfected cells (Fig. 5.6R). In contrast to previous results (Fig. 5.3C), Olig1 expression was repressed and detected at negligible or low levels in Sox10, T1, and T2 transfected cells compared to the negative control cells (Fig 5.6R).

In addition to comparing the induction of Schwann cell markers above MEF levels in all conditions, we also compared levels of induction of the genes above negative vector controls in response to Sox10, T1 or T2. For many genes, T1 was the most efficient at inducing the expression of Schwann cell lineage markers (e.g. Sox10, Ngfr, Tfap2a, Gap43, Cdh19, Erbb3, Shh, Egr2), suggesting that it may have a more mature Schwann cell phenotype with enhanced repair potential. For future analyses of the repair potential of these cells, we set up a functional assay below.

180

Figure 5.6. Ngfr+O4+ Schwann-like cells exhibit improved Schwann cell specific marker gene expression.

(A) Schematic of the lineage conversion strategy, involving the isolation of E13.5 MEFs, negative FACS sort to remove Ngfr+ cells, transfection with mCherry (negative control) and Sox10, T1 and T2 constructs, and growth in glial induction media for 21 DIV, followed by Ngfr+O4+ FACs sorting. A schematic of T1 and T2 factors is also shown. (B-R) Relative expression levels

181

(normalized to MEFs) of Sox10 (B), Sox2 (C), Jun (D), Pax3 (E), Ngfr (F), Tfap2a (G), Cdh19 (H), Erbb3 (I), S100 (J), MPZ (K), Pou3f1 (L), Gap43 (M), Shh (N), Gfap (O), Egr1 (P), Egr2 (Q), and Olig1 (R) in transfected cells after 21 DIV. Bars represent means ± s.e.m. (N=3). *, p < 0.05; **, p < 0.01; ***, p < 0.001 compared to the results obtained in Neg MEFs (one-way ANOVA).

5.4.6 Optimization of Dorsal root ganglia-Schwann cell co-culture assay for Schwann cell association and myelination assessment

Axonal myelination is the primary function of mature myelinating Schwann cells212. To assess the functional capacity of induced repair-like Schwann cells to associate with and myelinate axons, we set up an in vitro DRG-Schwann cell co-culture assay435. DRGs were isolated from P2 pups and cultured in an adherent system in the presence of NGF to promote neurite outgrowth. DRG explants were treated with cytosine arabinoside, which drives apoptosis in dividing cells600, to selectively remove contaminating Schwann cells and fibroblasts in the explant culture. The DRG explants were then co-cultured with Schwann cells to assess for Schwann cell association with outgrowth neurites.

In our control studies, to label Schwann cells, we made use of a Sox10-Cre knock-in (KI) mice601, which were crossed with a Rosa-tdTomato Cre reporter line. We validated the expression of tdTomato in developing Schwann cells by performing immunostaining of E12.5 SCPs and E14.5 iSCs using tdTomato and Sox10 antibodies. Sox10-tdTomato co-expression was observed in the DRG, dorsal root and ventral root, which are lined by iSCs and SCPs, at both E12.5 (Fig. 5.7A-A’’) and E14.5 (Fig. 5.7B-B’’). We then isolated nerve-derived tdTomato+ cells in the Schwann cell lineage from P21 Sox10-Cre;Rosa-tdTomato mice, which were co-cultured with DRG explants for 15 DIV under myelinating conditions (i.e., with Nrg1, forskolin, NGF and ascorbic acid) (Fig. 5.7C). DRG-Schwann cell explants were then immunostained with tdTomato (Schwann cells) and Tuj1, to mark outgrowing neurites (Fig. 5.7D,E). We detected strong Tuj1 staining in the DRG neurites that were associated with tdTomato labelled Schwann cells (Fig. 5.7D,E).

Thus, this assay can now be used to assess axonal association and myelination of the Ngfr+ O4+ ‘reprogrammed’ Schwann cells.

182

Figure 5.7. DRG-Schwann cell co-culture assay.

(A,B) Co-expression of Sox10 and tdTomato in E12.5 (A-A’’) and E14.5 (B-B’’) transverse sections through the trunk of Sox10-cre;Rosa-tdTomato embryos. (C) Schematic of DRG- Schwann cell co-culture assay. P2 wild type DRGs were co-cultured with tdTomato labelled Schwann cells collected from P21 Sox10-cre;Rosa-tdTomato nerves. (D,E). Immunostaining of DRG-Schwann cell co-culture after 15 DIV with tdTomato (labelling Schwann cells in red) and

183

Tuj1 (labelling DRG neurite outgrowth in green) and DAPI counterstain in blue. (Scale bars: 100µm in (A,B,C), 75µm in (A’,A’’,B’,B’’,D), 50µm in (E)).

5.5 Discussion

In this study, we set out to develop a TF based lineage conversion strategy involving a non- integrative, episomal approach for the induction of a repair Schwann cell-like identity in MEFs. We found that mis-expression of two combinations of three key TFs involved in Schwann cell development and repair; Pax3-Jun-Sox2 (T1) and Sox10-Jun-Sox2 (T2), could induce the expression of Schwann cell-specific target gene expression in MEFs more efficiently that Sox10 alone. While some of the embryonic and repair Schwann cell-specific markers were induced by the misexpressed TFs, others were induced by the Schwann cell-fate supportive cues in the glial media, highlighting the added efficacy of a combinatorial strategy. Our study differs from others in the literature as we used a non-integrative, episomal approach, whereas prior Schwann cell lineage conversion protocols involved integrative approaches (i.e. use of lentiviral constructs for gene expression) that were combined with small molecules, or small molecules that were used on their own469,470,472,602. Further comparisons of our approach and others are outlined below.

Past reprogramming strategies relied on the transformation of fibroblasts into an intermediate multipotent, progenitor-like state, followed by directed differentiation of the progenitors into Schwann cells468. Kim et al. first reported conversion of human fibroblasts into an induced NCC state by overexpressing Sox10 in conjunction with a cocktail of small molecules (valproic acid, 5-azacytidine, CHIR99021)468. The multipotent induced NCCs were then differentiated into myelinating Schwann cells by providing appropriate environmental cues468. Similarly, Thoma et al. demonstrated that fibroblasts can be directed towards a transient neural precursor state using small molecules alone, and then differentiated into myelinating Schwann cells469. In contrast, recent studies by Mazzara et al. and Sowa et al. have demonstrated a more potent and directed lineage conversion strategy (overexpression of Sox10 and Egr2) that bypasses the intermediary progenitor step and allows direct induction of a Schwann cell fate in fibroblasts470,471. The induced Schwann cells generated in response to Sox10 and Egr2 misexpression had a mixed phenotype, but tended towards an immature to mature Schwann cell state as identified by staining the cells with S100 and O4. Building on these studies, we based our lineage conversion strategy on the TF mediated directed reprogramming approach.

184

Our triple TF approach was distinct from prior studies as we aimed to generate a homogenous population of Schwann cells with a ‘repair’ and proliferative Schwann cell phenotype compared to the heterogenous population of Schwann cells generated in other studies470,471.

Repair and embryonic Schwann cells broadly share similar gene expression profiles, barring expression of a few markers like Olig1 and Shh, which are detected in repair Schwann cells uniquely151,324. Given the shared expression profiles of embryonic and repair Schwann cells, the two triple vectors in this study were constructed using key TFs expressed in both embryonic and repair Schwann cells260, i.e. Sox2, Jun, and Sox10 or Pax3. We postulated that the cooperative action of the three TFs in each triple vector would induce the expression of Schwann cell-specific candidate genes in MEFs and direct these cells towards acquiring a repair Schwann cell-like phenotype. Prior studies have revealed the individual roles of Sox10, Sox2, Jun, and Pax3 in Schwann cell development or nerve repair. For example, studies in rodent models have shown Sox10, expressed in Schwann cells pan-development, can directly induce expression of myelinating genes like Egr2, MBP, MPZ 151 and Pou3f1303. Sox2, Jun, and Pax3 are expressed in embryonic and repair Schwann cells, and have been postulated to play important roles in inhibiting myelination, and maintaining a de-differentiated, proliferative Schwann cell phenotype post-injury to support nerve regeneration and remyelination151,258,260. The induction of several SCP markers (e.g. Ngfr, Tfap2a, Gap43, Cdh19, Erbb3) in triple T1 and T2 transfected cells, over and above negative control and Sox10 transfected MEFs, indicated that T1 and T2 likely act in a combinatorial fashion to turn on several different downstream genes that together specify a repair Schwann cell-like identity in MEFs. Thus, we have some evidence that triple TFs can induce a repair Schwann cell-like phenotype in MEFs.

While our triple TF mediated lineage conversion strategy was successful in inducing a Schwann cell-like identity in somatic cells, we acknowledge the need for further studies to cement this approach to make it clinically relevant. Single cell and bulk RNA sequencing assessment will greatly aid in determining whether our approach definitively transforms somatic cells into Schwann cells with a repair phenotype. The use of repair Schwann cells151 as a comparator cell population in transcriptomic studies will also help assess lineage conversion efficiency more effectively. The TF combination exhibiting the greatest conversion efficiency can then be validated in vivo using acute and chronic nerve injury studies to assess

185 induced Schwann cell survival and effects of induced Schwann cell on axon regeneration, myelination, integration into existing axonal circuitry, and long-term functional recovery.

In summary, our triple TF based lineage conversion strategy holds the potential of improving upon the recently reported Schwann cell generation protocols and may allow generation of clinically relevant repair-like Schwann cells.

186

Discussion

Some content in this chapter was included in a review article:

Balakrishnan A, Roy S, Fleming T, Leong HS, Schuurmans C. 2020. The Emerging Role of Extracellular Vesicles in the Glioma Microenvironment: Biogenesis and Clinical Relevance. Cancers 2020, 12, 1964. doi: 10.3390/cancers12071964.

Link: https://www.mdpi.com/2072-6694/12/7/1964/htm

187

6.1 Summary of present findings 6.1.1 Chapter 2: SMPD3-mediated extracellular vesicle biogenesis inhibits oligodendroglioma tumor growth

6.1.1.1 Summary

Oligodendroglioma (ODG) is a non-aggressive and slow progressing glioma12. To understand the intricacies of ODG growth and progression, it is important to study how ODG cells interact with neighboring cells in the tumor microenvironment. With this aim, I analyzed resected tumor sections from ODG patients and found evidence suggesting that tumor cells may have non cell-autonomous effects on neighboring non-neoplastic stromal cells. To further study this phenotype, I made use of two ODG patient-derived cell lines, BT088 and BT054, which are fast and slow growing, respectively22. To assay the bioactivity of secreted factors from ODG cells, I used an embryonic neural stem cell (NSC) assay, choosing these cells as they are non- transformed primary cells with a neural identity that are easily isolated, highly responsive to growth signals, and have well developed assays for analysis of stem cell self-renewal, proliferation, differentiation and survival493. I also examined whether BT088 cells influence their own growth in a model of homotypic cell-cell interactions. I found that ODG cell derived EVs had cytotoxic effects on both NSC and ODG cells. Upon proteomic profiling of ODG vesiculomes from BT088 and the slow growing BT054 ODG cell line, I found an enrichment of cell death and cell growth proteins, including components of VEGF-signaling. Strikingly, treatment of BT088 cells with Foretinib, a multi-kinase inhibitor that targets VEGFR amongst other receptor tyrosine kinases, increased cytotoxicity and slowed the growth of BT088 cells.

To further examine the potential importance of EVs in ODG tumor growth in patients, I queried the TCGA dataset and found a positive correlation between high expression of exosome synthesis genes (e.g. SMPD3, TSG101, STAM1) and low grade glioma (astrocytoma and oligodendroglioma, combined) patient survival, such that patients with high gene expression levels exhibited favorable prognosis. Among these three genes, SMPD3 correlations with survival were the highest, and further analyses revealed that this correlation held in a smaller pool of specifically ODG patients. I then showed that the growth of ODG cells in vitro was inhibited by overexpression of SMPD3, while conversely, SMPD3 knockdown reduced EV secretion and increased ODG cell proliferation. Xenografting ODG cells after SMPD3 knockdown also reduced survival of host animals. Finally, a cerebral organoid-tumor co-

188 culture system was developed for modelling tumor behavior in an in vitro ‘humanized’ setting. I observed increased infiltration of SMPD3 knockdown ODG cells into the organoid neural rosettes. In conclusion, I have shown that ODG tumor growth is complexly regulated and highlights the importance of ODG-EV secreted signals in mediating these interactions.

6.1.1.2 Biological implications

Over the years, both high and low-grade glioma cells have been shown to secrete a large variety of signaling molecules and growth factors (e.g. EGF, FGF,VEGF,CXCL12, TGF, CD133, Shh, Wnt), but only recently did investigators begin to examine whether these factors were soluble or vesicular175,603. I began by assessing the bioactivity of conditioned media from ODG cell lines, which includes both soluble and vesicular fractions, and ended by focusing on the very striking cytotoxicity associated with EVs, as I reasoned that this activity may account in part for the slow growing nature of ODG tumors. Of note, a few earlier studies demonstrated that another ODG cell line secretes EVs that have cytotoxic effects on normal neural cells203,204, but these studies did not demonstrate a linkage in patient samples or demonstrate an in vivo role, as I have done. D'Agostino et al. first reported that EVs derived from G26/24 ODG cells have cytotoxic effects on neuronal cultures in vitro204. Upon treatment with ODG EVs, there was a decline in expression of neuronal markers Neurofilament p68, as well as neurite associated proteins MAP-2 and GAP43, indicative of cell death204. Indeed, apoptosis was observed with neurons exhibiting an increase in activated caspase-3 expression. This induction of neuronal apoptosis was a function of Fas ligand protein (FasL) carried in ODG derived exosomes204. Compared to neurons, astrocytes exhibit greater resistance against the apoptotic effects of G26/24 ODG-EVs, with 8-10 times more EVs required to initiate astrocyte apoptosis. Moreover, a different cell death signal mediates astrocyte cell death - tumor necrosis factor- related apoptosis-inducing ligand (TRAIL), which is also found in the G26/24 ODG-EV cargo203. This difference in ODG EV effects on neurons and astrocytes highlights the heterogeneity of the enclosed cargo, and the complex cell-cell communication events that likely occur in the tumor microenvironment203.

In my study, I similarly found that BT088 and BT054 ODG cell derived EVs have cytotoxic effects on embryonic NSCs. However, unlike G26/24 ODG cells203, I did not detect FasL or TRAIL proteins in BT088 and BT054 vesiculomes. One important caveat of studies using the G26/24 line is that it was established as an ODG line based on the histological features of the tumor, but more modern criteria (IDH1/2 mutation, 1p;19q co-deletion, CIC mutation) were 189 not applied to validate the initial diagnosis. In contrast, I used BT088 and BT054 cells, which were derived from genetically validated ODG tumors22. However, despite their similar genetic profiles, the growth properties of BT088 and BT054 cells differ, as do their EV content, further highlighting the heterogeneity of ODG EV cargo proteins (observed in my study).

The demonstration that ODG EVs are cytotoxic is in sharp contrast to previous studies investigating high-grade astrocytoma (GBM) derived EVs, which are growth promoting (reviewed in detail in175,176). Strikingly, GBM derived EVs influence the secretory phenotype of human astrocytes, inducing astrocytes to secrete growth factors (EGF, VEGF, FGF), interleukins and chemokines, as well as enhancing astrocytic migration604. GBM derived EVs also induced endothelial and microglial cell proliferation175. Non-vesicular, soluble factors released by glioma cells also contribute towards paracrine signaling observed in the glioma microenvironment175,603 and can induce cell proliferation and differentiation by initiating growth factor signalling605,606. Venugopal et al. demonstrated that the GBM secretome carries soluble factors (EGF, VEGF) that direct neural precursor cells to acquire a transient proliferative, stem cell-like state607. In my study, I observed similar pro-proliferative effects of the BT088 and BT054 ODG conditioned media on normal NSCs, although these effects were largely masked by the cytotoxicity associated with EVs (Chapter 2, Fig. 2.2). The nature of the BT088 and BT054 ODG cell secreted soluble factors remains to be defined, however, VEGF is a potential growth factor secreted by these tumor cells as described further below. Thus, although EVs were the primary focus of my study, it is important to acknowledge that it is the collaborative effect of both soluble factors and EVs that shape ODG growth.

6.1.1.3 Future perspectives

6.1.1.3.1 Therapeutic role of EVs

Through SMPD3 gain- and loss-of-function studies in ODG cells, I demonstrated that ODG EVs carried cytotoxic factors that influence the survival of other ODG cells and potentially other cells in the tumor cell niche. While my assays of ODG EVs were performed on embryonic NSCs, which are not part of the tumor microenvironment, it is possible that these EVs also have cytotoxic effects on other stromal cells that are in the brain parenchyma that is invaded by the tumor, including neurons, astrocytes, oligodendrocytes, microglia and endothelial cells. In the future, I could perform additional studies in which BT088 and BT054 EVs were added to these other cell types, and effects on survival could be monitored.

190

One prediction of my work is that an increase in ODG-EV synthesis, which I found has cytotoxic effects on ODG cells themselves, could translate into a potential therapeutic strategy for ODG patients. While I showed that SMPD3 overexpression blocks ODG growth in vitro (Chapter 2, Fig. 2.6), I did not test whether these cells form smaller tumors in vivo. However, in a related study, Gao et al. recently demonstrated that 73C glioma cells (astrocyte-derived glioma) with a Rab27a knockdown failed to proliferate in an in vivo xenograft system198. This observation further cemented the idea that astrocytoma-EVs (e.g. GBM EVs) are attributed with growth supportive functions175,176, and a loss in EV synthesis thus translates to poor glioma growth. Conversely, my data suggests that given the cytotoxic nature of ODG EVs, a clinically relevant approach to inhibit ODG growth would involve increased synthesis and release of EVs via genetic or pharmacological approaches. In my study, I showed that GW4869 (nSMase2 inhibitor) can inhibit exosome synthesis in vivo608,609, but for therapeutic benefit, the goal would be to increase EV synthesis.

The production of EVs has therapeutic potential in a large range of disorders in addition to cancer. For example, EVs derived from mesenchymal stem cells (MSCs) have regenerative effects in the heart, lung, liver and brain523. Moreover, EVs are excellent candidates for therapeutic delivery as their lipid bilayer would naturally protect nucleic acid and protein cargo from nucleases and proteases in the extracellular space. The limitation of this approach is that levels of EV secretion may be too low to have the desired therapeutic effect. For this reason, several screens have been conducted for small molecules that could induce EV secretion to increase the therapeutic potential of EVs for cargo delivery. In the context of cancer, EVs are released by many tumor cells, however, like in GBM, EV secretion is often associated with a worsening of prognosis in many tumor types (e.g. pancreatic cancer, melanoma, breast cancer)522. For that reason, most high-throughput screens have focused on identifying small molecules that block EV secretion522. However, because of the regenerative potential of MSC EVs, there have also been recent screens for small molecules that induce EV secretion, which have identified N-methyldopamine and norepinephrine as combined signals that not only induce the formation of more MSC EVs (without changing MSC cell number), but importantly, these EVs maintain their regenerative potential523. Another important outcome of this study was that these small molecules acted by turning on nSMase2, which is the enzyme that I found is associated with ODG growth. Future studies in ODG could involve testing whether these agents have the same capacity to boost EV synthesis by ODG cells in vitro, and then to observe whether there is a therapeutic effect.

191

6.1.1.3.2 Understanding the significance of ODG EV cargo

While I showed that ODG EVs have cytotoxic effects, it is likely not the EVs themselves that are cytotoxic, but rather, their enclosed cargo, which is delivered to neighboring cells, that exerts inhibitory effects. Indeed, EVs carry protein and nucleic acid cargo that could all influence the physiology of recipient cells44. Interestingly, a breast cancer study using a Cre recombinase-based fluorescent reporter assay to track EV-mediated Cre transfer suggested that it is the Cre mRNA that is transferred and bioactive499. In my study, I did not profile nucleic acid cargo, so future studies using mRNAseq would be informative. Instead, I profiled protein cargo, which I cannot rule out has some biological effects. One advantage of examining the proteome is that proteins are thought to be packaged into EVs non-specifically (as opposed to the specific loading of nucleic acids), and thus, the EV proteome can serve as an actual readout of cellular state. Thus, I reasoned that the proteins present in ODG EVs would provide an indication of the signaling status of the cell.

Our proteomic analyses identified a large number of cell death and cell growth associated proteins in the BT088 and BT054 vesiculomes. While the ODG EVs themselves have cytotoxic effects, these tumor cells are actively growing and secreting soluble factors that are growth promoting. Hence if these proteomic profiles as considered as indicators of cell state, they match the complexity of the ODG secretomes. I considered above that increasing the synthesis of cytotoxic EVs might be one therapeutic strategy to reduce ODG growth. Another therapeutic strategy would be to use my vesiculome data to identify growth-promoting signaling pathways that may be active in ODG cells, as potential targets of drug therapies. Two proteins that were detected in the BT088 and BT054 vesiculome were MFGE8 and SRI, which are both involved in VEGF signaling501-503. Interestingly, VEGF binds VEGF receptor 2 (VEGFR2) to induce the proliferation of GBM stem like cells520. Similarly, VEGF acts in an autocrine fashion to promote proliferation of hepatocellular carcinomas, acting through a VEGFR1/R2‐ PLC-ERK pathway610. I thus reasoned that VEGF signalling may be active in ODG cells and act in an autocrine fashion to promote proliferation. Consistent with this hypothesis, exposure of ODG cells to Foretinib, a VEGFR inhibitor611-613, resulted in a marked decline in ODG cell growth, and increase in cell death (Chapter 2, Fig 2.4). One experiment still outstanding would be to perform an ELISA on the ODG secretome to check whether BT088 and BT054 cells secrete VEGF, as Foretinib is a multi-kinase inhibitor that may be blocking other pathways. Nevertheless, regardless of its target, my data suggests that Foretinib could be used to block ODG tumor growth in vivo, as I demonstrated in vitro. Before proceeding to mouse ODG 192 xenografts, I would first test whether Foretinib blocks ODG growth in the human cerebral organoid-ODG co-culture system. However, the mouse xenografts would be important to pursue, as numerous drugs have been identified that kill glioma cells in vitro, but these effects are infrequently seen in vivo. One issue is that in vitro screens may not recapitulate disease complexity. Indeed, gliomas are composed of a mixture of malignant glial cells ('tumor cells') and non-cancerous glial cells, neurons, inflammatory and vascular cells ('stromal cells'), as I have highlighted above66,481.

Another interesting function of VEGF signaling is its role in angiogenesis. A defining feature of higher-grade gliomas, like GBMs, is the proliferation of microvascular endothelial cells, and hyperplasia519. Low-grade ODGs on the other hand present with a chicken-wire like vascular phenotype22. Thus, increase in vascularity is a mark of higher-grade gliomas614. Based on the proteomic data, I inferred that VEGF signaling may be activated in the two ODG cell lines and may promote both autocrine and paracrine signaling. One possibility is that ODG EV cargo can directly/indirectly promote VEGF signalling between ODG cells and neighbouring endothelial cells, thereby promoting vascularization to support ODG growth. Therefore, an inhibition of VEGF signalling by Foretinib in ODG cells may lead to a decline in angiogenesis around the tumor, leading to poor survival of ODG cells. Indeed, inhibition of angiogenesis by inhibiting VEGFR signalling leads to a decline in GBM growth, as demonstrated using VEGF- targeting drugs and VEGFR inhibitors including Bevacizumab, Sorafenib, Cediranib, Foretinib612,615. Thus, VEGF inhibition can be considered as a prospective therapeutic strategy to limit ODG growth.

6.1.1.3.3 Role of EVs as clinical biomarkers

While my studies focused solely on the biological functions of ODG EV cargo, glioma EVs also have the potential to be used for disease diagnosis to follow the status of brain tumors181. Indeed, EV cargo can potentially provide information regarding tumor grade (indolent versus aggressive), stage (size and local spread beyond primary site), tumor biology (increased vs. decreased cell proliferation; e.g. EGFRvIII is associated with increased proliferation of glioma cells), and treatment response information when analyzed before and after a chosen therapy (surgery, radiation therapy, systemic therapy)616,617. Here, I further elaborate on the potential of EVs as biomarkers for glioma diagnosis as well as prognostic indicators.

193

Needle biopsies are invasive procedures that are particularly challenging to obtain from brain tumors, especially when tumors are not immediately accessible at the brain surface as normal brain tissue can be damaged in the needle path. A non-invasive means to understand the underlying biology of the mass in question would significantly benefit the patient, sparing them side-effects from the needle biopsy procedure. EVs released by the tumor that are present within the blood circulation incorporate various macromolecules that could serve as biomarkers of tumor status618,619. The term “liquid biopsy” or “fluid biopsy” is a neologism that refers to an alternative to needle biopsy for obtaining the same information regarding a disease or lesion via analysis of a biofluid616. In oncology, a “liquid biopsy” is typically the analysis of a biomarker (nucleic acid, protein, metabolite) released by a tumor and its abundance in human biofluids. Indeed, several EV-based liquid biopsies have been patented and commercialized (previously reviewed by Urbanelli et al.620) with the goal of improving diagnosis and clinical workup of patients with suspected GBM.

The brain is an immune privileged organ and protected by the blood-brain-barrier (BBB), which is a vascular wall that is impermeable to most cells, drugs, metabolites, and proteins621. Indeed, Zhao et al.622 determined that BBB permeability exists in GBM and can vary with stage of the tumor, but not all GBM tumor vasculature is compromised. Pockets or regions of BBB permeability were found to correlate with intravasation of GBM-EVs and subsequently elevated in peripheral whole blood samples622. Established GBM biomarkers (e.g. mutated EGFRvIII) are commonly used due to their robust expression in the majority of GBM tumors and cell surface expression in GBM cells. Significant literature describing utility of EGFRvIII as a GBM-specific EV marker has led to its use as a clinical diagnostic tool623 and as a prognostic tool624. At the time of initial patient presentation where the biology of the suspicious mass is not yet determined, GBM-EVs may also provide risk-stratifying information625,626.

RNA biomarkers resident within GBM-EVs offer the greatest promise for clinical impact due to their low cost and robust expression within the blood due to their serum stability qualities623. Similarly, cerebrospinal fluid (CSF) may be the most relevant biofluid for a liquid biopsy for brain cancers in order to deliver clinically relevant information for brain cancer patients618,627. CSF is predominantly acellular and is proteomically distinct from whole blood which makes analysis and processing more straightforward628. However, it suffers from one major drawback: it can only be obtained via lumbar puncture, a technically challenging means of biofluid collection and is rarely performed in brain cancer patients. However, various key studies49,627

194 have enumerated brain cancer derived EVs in CSF that express EGFRvIII, resulting in a CSF- based liquid biopsy that exhibits promising performance test characteristics with 61% sensitivity and 98% specificity49. Hence, CSF analysis for EV-based biomarkers has yielded very promising liquid biopsies for GBM and other brain cancer patients that may be used clinically.

In the future, it would be interesting to assess whether ODG EVs in a xenograft model can be detected in the CSF, and furthermore, whether the efficacy of various treatments (e.g. Foretinib) can be assessed by assaying EV content.

6.1.2 Chapters 3-5: Schwann cells in a healthy and injured nerve

6.1.2.1 Summary

Schwann cell development is a long process involving distinct stages beginning with neural crest cells (NCCs) that first become committed to a Schwann cell fate by differentiating into Schwann cell precursors (SCPs), that then become immature Schwann cells (iSCs), pro- myelinating Schwann cells, and finally, myelinating or non-myelinating Schwann cells260. Each of the above Schwann cell developmental stages is marked by the expression of key transcription factors which play an important role in defining and promoting stage-specific phenotype258,629. Importantly, part of this developmental progression is re-iterated upon peripheral nerve injury, with Schwann cells de-differentiating into a repair Schwann cell state, mimicking many aspects of embryonic Schwann cells with respect to their gene expression profile258,629.

In Chapter 3, I evaluated the spatio-temporal expression profile of 12 key transcription factors at key stages of Schwann cell development. I characterized the development of Schwann cells at five embryonic and two postnatal stages and identified key transcription factors upregulated during each developmental stage. I further assessed whether the expression profile of ‘repair’ Schwann cells observed five days post-nerve injury imitated a specific embryonic developmental stage. I reported that post-injury de-differentiated Schwann cells recapitulate some of the transcriptional features of SCPs, iSCs as well as pro-myelinating Schwann cells (e.g. induction of Sox2, Jun, Pou3f1/Oct6). Thus, I demonstrated that repair Schwann cells acquire features of embryonic Schwann cells, which likely contribute towards the regenerative and remyelination functions of these cells.

195

In Chapter 4, I queried whether Etv5, an ets domain transcription factor expressed in NCCs and SCPs, is required for the normal development of Schwann cells using a mutant mouse model. To determine the function of Etv5 in Schwann cell specification, I examined the expression of Schwann cell markers (as established in Chapter 3) in hypomorphic Etv5 mutant mice (Etv5tm1Kmm 583), which are viable embryonically and postnatally. Analyses at E12.5, E15.5 and E18.5 revealed that Etv5 mutants did not exhibit altered expression or a developmental delay in the initiation of expression of the assessed Schwann cell markers. Additionally, no phenotypic differences were observed in repair Schwann cells in Etv5 mutant’s post-injury. While my analyses failed to find a role for Etv5 in Schwann cell development, there are two important caveats to my study: 1) I used a hypomorphic Etv5 mutant allele, and 2) Etv5 may act in a redundant fashion with other ets-domain transcription factors, such as Etv4, as previously shown in the developing kidney584.

In Chapter 5, I developed a transcription factor-based lineage conversion strategy for the generation of repair like Schwann cells from somatic cells. I generated two triple transcription factor vectors (Pax3-Jun-Sox2 and Sox10-Jun-Sox2) to promote the mis-expression of key repair Schwann cell genes in target cells. Using mouse embryonic fibroblasts (MEFs) as the starting cellular source, I mis-expressed the two triple transcription factor combinations and cultured the cells in glial cell supportive media. The transfected cells were assessed for induction of embryonic and repair Schwann cell specific markers. As a control, cells were also transfected with a Sox10 construct, to determine whether MEFs can be directed towards a Schwann cell fate by misexpression a single transcription factor alone. I found that Pax3-Jun- Sox2 and Sox10-Jun-Sox2 induced the expression of several embryonic and repair specific Schwann cell markers in MEFs above levels achieved with glial media alone, or by Sox10 alone. I then isolated ‘reprogrammed’ cells that took up a Schwann cell-like identity based on the co-expression of Ngfr and O4 cell surface markers. Interestingly, upon assessment of the purified induced cell population, I found that Pax3-Jun-Sox2 induces elevated expression of Schwann cell markers in Ngfr+O4+ cells to a higher level than Sox10-Jun-Sox2. Future studies will assess the myelination capacity of the induced cells using the optimized in vitro dorsal root ganglion myelination assay and in vivo nerve injury transplant assays. To the best of my knowledge, this is the first study which aimed at generating repair Schwann like cells using a direct reprogramming approach.

196

6.1.2.2 Biological implications

6.1.2.2.1 Dynamic regulation of Schwann cells in development and post-injury

Development of Schwann cells from a NCC stage to mature myelinating Schwann cell state has been described extensively in the literature (reviewed in258,629). It is the tight regulation of Schwann cell specific gene expression, including genes encoding transcription factors, that allows progenitor cells in the Schwann cell lineage to migrate, proliferate, differentiate and myelinate in a temporally and spatially regulated fashion. The differential expression of these lineage-specific markers makes Schwann cells receptive and responsive to axonal and environmental cues so that developmental transitions occur on time and in the correct order. A number of transcription factors have been identified (e.g. Sox10, Sox2, Egr2, Oct6, Jun152,258) that are expressed at various Schwann cell development stages, however, these markers had not been assessed in a direct comparative and comprehensive manner152. Additionally, several of the markers I focused on were not analyzed throughout Schwann cell development and post- injury, while the expression profiles of some known markers were generated from varying species (e.g. Xenopus537), age groups, or using in vitro studies alone. Thus, it was crucial to define the spatio-temporal onset and offset of Schwann cell transcription factor expression to begin to elucidate the transcriptional cascades that underlie the development of this lineage. Furthermore, to design transcription factor-based cellular reprogramming strategies, it was necessary to identify the transcription factors that are expressed at the appropriate developmental stage to act as Schwann cell fate determinants. Hence, my study was focused on a core set of Schwann cell-specific transcription factors, which would allow reliable identification of distinct stages of Schwann cells during development.

Among the key findings of my study, I documented the temporal expression pattern of Sox9, a NCC marker247,537, which had previously only been studied in the developing nerve from E14.5 onwards6. I reported that Sox9 followed a similar expression pattern as Sox10 throughout embryonic Schwann cell development, as well as post-injury. Sox9 is expressed in subependymal neural stem cells557 and various epithelial stem cell types558,559 where it is predicted to induce and maintain a stem cell like state in the cells. Thus, the maintained expression of Sox9 in adult myelinating Schwann cells may be essential for re-acquiring the proliferative state post-injury. Similarly I detected the expression of Egr1, a non-myelinating Schwann cell marker261, in SCPs, and in postnatal Schwann cells. Further, I report that post- injury, Egr1+ Schwann cells did not increase in number, however increased intensity of Egr1

197 staining and nuclear translocation of Egr1 was observed suggesting a functional change in the role of Egr1 in repair Schwann cells. Of note, Egr1 is normally expressed in cells during cell cycle-re-entry560, and their up-regulation post nerve injury630 indicates that Egr1 may play an important role in repair Schwann cells and nerve regeneration. Finally, another important finding of my study was the rare occurrence of Pax3+ Schwann cells in the injured nerve. Pax3 transcripts are upregulated post injury258, and hence the absence of increased Pax3 protein staining in my assessments indicated that the expression of Pax3 post-injury may require a longer window period (> 5 days), or may depend on the severity of nerve injury. Thus, future studies using Pax3 conditional mutants could help determine role of Pax3 in repair Schwann cells.

Of note, improved imaging and isolation of repair Schwann cells401 and use of single cell RNA sequencing has helped tease out the unique genes upregulated post injury (e.g. Olig1, Shh258). While role of other genes like Sox2, Jun, Egr2, Oct6, Sox10 in repair Schwann cells has been studied in great detail (reviewed in258), the functional role of Egr1, Sox9, Pax3, Tfap2a has not been assessed. Since these transcription factors are expressed (either upregulated or sustained expression observed) post-injury in repair Schwann cells, I predict that they may play an important role in inducing the proliferative state of repair Schwann cells, similar to the role of Sox2 in repair Schwann cells288,321,595. Thus, I propose that the use of conditional gene knockout mouse models will allow us to address this query and help determine the individual function of Egr1, Sox9, Pax3, Tfap2a in repair Schwann cells.

6.1.2.2.2 Role of Etv5 in satellite glial cells and Schwann cell development

Schwann cells and satellite glial cells (SGCs) populate the dorsal root ganglion (DRG), where SGCs surround the neuronal cell body (Chapter 1, Fig. 1.3 A,B). Schwann cells and SGCs share a common embryologic origin631 as well as overlapping cytological characteristics211. Morphologically, SGCs encircle the neuronal cell body such that several rounded nuclei appear surrounding the cellular body in a DRG transverse section632. On the other hand, the flattened nuclei of Schwann cells aggregate in the centre of the DRG where the neuronal processes aggregate632. Apart from functional differences, Schwann cells and SGCs differ in the expression of the Etv5 transcription factor. Etv5 belongs to the ETS (E26 Transformation Specific) superfamily of transcription factors, which are characterized by a conserved DNA binding ets domain (helix-turn-helix super family of domains)633. Etv5 is expressed in NCCs, TrkA+ DRG sensory neurons, SCPs, and SGCs249,543. Neurotrophin NGF induces Etv5 198 expression (via the MAPK pathway) in DRG sensory neurons, thus promoting axonal growth of TrkA+ DRG neurons634. Additionally, Etv5 is a transcriptional activator633 which can promote cell growth and differentiation583,635. Prior in vitro studies had demonstrated that a functional block in Etv5 expression in NCCs affects neuronal specification and glial proliferation544. However, my studies using an Etv5 hypomorphic mutant demonstrated that a partial loss in Etv5 did not result in an inhibitory effect on Schwann cell proliferation and differentiation, or in repair Schwann cells appearance post injury. The conflicting results observed could be since 1) the initial studies were performed in vitro544, and were not followed up by in vivo characterization, 2) due to the hypomorphic nature of the Etv5 mutant mice, 3) or due to genetic redundancy of Etv5 with Etv1 and/or Etv4581,584. However, the compensation of Etv5 activity by Etv1 and/or Etv4 remains to be assessed. Use of a conditional Etv5 knockout may also help definitively tease out the role Etv5 in Schwann cell development and in injury.

Prior in vitro studies have shown that SGCs can transition to an Egr2 expressing Schwann cell state570. Studies have also demonstrated the importance of the paracrine signal Neuregulin 1 isoform typeII (GGF2) to induce Etv5+ progenitors to a Schwann cell phenotype (Pou3f1+Etv5- )249. Hence, it has been proposed that Etv5+ SGCs could act as a secondary source for replenishing the Schwann cell pool in the nerves; this could be accomplished by migration of the SGCs from the ganglia to the nerves resulting in the loss of the microenvironment of the DRG essential to maintain the expression of Etv5 in the SGCs. Alternatively, the peripheral nerves may present with repressive signals that can override the action of Neuregulin1 isoform type II resulting in downregulation of Etv5, thus acquiring a Schwann cell phenotype249. However, there have been no studies yet to answer this query and it remains an open question in the field.

6.1.2.2.3 Lineage conversion strategies for Schwann cell generation: benefits and pitfalls

Schwann cells are crucial regulators of peripheral nerve regeneration258. Attempts at nerve- or nerve graft-repair post peripheral nerve injury often yield poor results, which can be attributed to poor nerve engraftment, length of nerve graft, and limited regenerative potential of endogenous Schwann cells which may eventually acquire a senescent state368,373-376,636. Alternative approaches to enhance nerve repair have been tested including grafting nerve-like conduits containing autologous cultured Schwann cells to aid nerve regeneration386. However,

199 this approach suffered from similar limitations as observed during direct nerve engraftment. Thus, other sources of Schwann cells have been explored for transplant purposes.

Cellular reprogramming, which refers to the ability to directly convert terminally differentiated cell types into another adult somatic cell type, or even back to a pluripotent state, provides the potential for generating Schwann cells for transplant637. Previously reported reprogramming protocols for generation of Schwann cells have utilized viral vectors and/or cell permeable small molecules which make cells amenable to reprogramming469-471,638. While these approaches have been successful at generating functional Schwann cells, the strategies involved may not be therapeutically relevant, i.e. involve use of viral vectors470,471, or involve an intermediate progenitor cell stage which may result in generation of other cell types468. Hence there is growing need for the development of clinically relevant lineage conversion strategies.

The use of non-integrative triple transcription factor expression vector approach in my studies was a step in this direction. The use of triple transcription factor vectors for the induction of repair Schwann cell identity in MEFs induced several known embryonic and repair specific markers in MEFs, however a drawback of my study was the low percentage of Ngfr+O4+ induced repair Schwann-like cells (~0.8-1%) after 21 DIV. Additionally, given the episomal nature of the vectors, the dilution of plasmid from transfected cells may result in loss of induced Schwann cell marker expression. This shortcoming can be avoided using non-viral approaches which allow sustained gene expression over a long period of time (described below in 6.1.2.3.3). Although my approach resulted in induction of a host of Schwann cell specific markers in MEFs at high expression levels, I acknowledge that bulk/single cell RNA sequencing studies470 post all lineage conversion protocols may be essential to effectively estimate how similar or divergent the induced Schwann cells are from the target repair Schwann cell population. While a candidate gene expression analysis is informative, larger gene-set screening is inevitable to be able to use such protocols for therapeutic purposes. Additionally, such screening procedures will also allow identification of cells in transitory stages of reprogramming, and thereby permit isolation of a purified induced repair Schwann cell population. Thus, by addressing these shortcomings generation of induced repair Schwann cells may be possible, which hold the potential to aid current nerve repair therapies.

200

6.1.2.3 Future perspectives

6.1.2.3.1 Skin derived precursor cells: Alternative cellular source for lineage conversion studies?

Fibroblasts remain a popular choice for cell source in several reprogramming studies469-471,639. In Chapter 5, I demonstrated that MEFs can be directed to adopt a Schwann cell-like identity using a transcription factor mediated approach. Since appropriate cellular sources are a crucial aspect of lineage conversion studies, I queried whether the lineage conversion protocol can be applied to other cell types as well. Skin derived precursor cells (SKPs) derived from hair follicle dermis have neural and mesenchymal fate potential, and can be differentiated into Schwann cells upon provision of glial supportive cues including neuregulin and forskolin435,436,440. However, the protocol for generating SKP-Schwann cells is protracted, requiring over 5-6 weeks440.

Since SKPs can be directed towards a Schwann cell fate with media cues alone, I argued that my lineage conversion strategy will be capable of inducing Schwann cell specific genes in SKPs in a shorter time frame than by the action of glial supportive cues alone. SKPs were transfected with Sox10, Pax3-Jun-Sox2 (Triple 1, or T1) Sox10-Jun-Sox2 (Triple 2, or T2) and grown in glial-supportive media for 20 DIV (Fig. 6.1A). Dilution of plasmid was visualized using the mCherry tagged negative control vector transfected cells, with scarce number of cells expressing mCherry after 20 DIV (Fig. 6.1B-D’). The same qPCR marker panel assessed in nerve-derived Schwann cells was examined (Chapter 5, Fig. 5.1B). Exogenously expressed genes (Sox10,Sox2,Pax3,Jun) were detected in transfected SKP cells upto 20DIV (Fig. 6.1E- H). To identify candidate gene expression changes due to the glial media alone, I normalized all expression values to those observed in MEFs cultured in the absence of glial cues. Thus, all bars shown indicate that gene expression was induced by glial cues in the media. In addition, I made comparisons between the negative control vector and all other constructs to distinguish transcription factor-induced changes. Indeed, the preliminary studies suggested that a misexpression of triple transcription factor expression vectors in SKP induces embryonic and mature Schwann cell marker transcripts (Tfap2a, Pou3f1, Egr2, and MPZ) within a 20-day period (Fig. 6.1I-P).

201

202

Figure 6.1 Induction of Schwann cell specific marker transcript expression by TTFEV in skin derived precursor cells.

A. Schematic of TTFEV mediated Schwann cell induction procedure using skin derived precursor cells (SKPs). Cells were transfected and cultured in Schwann cell induction media for 10 and 20 DIV. B-D’. Misexpression of mCherry labelled negative control vector in SKPs. Decline in mCherry expression detected in SKPs after 3, 5, 7 DIV post-transfection, demonstrated dilution of plasmid from transfected cells. E-H. RTqPCR analysis of misexpressed genes after Sox10, TTFEV [T1, T2] expression. Relative transcript expression of Sox10 (E), Sox2 (F), Pax3 (G), Jun (H) in transfected SKPs. I-P. Relative transcript expression of early and late Schwann cell markers Ngfr(I), Tfap2a (J), Gfap (K), S100 (L), Egr1 (M), Egr2 (N), MPZ (O), Pou3f1 (P) in transfected cells after 10 DIV and 20 DIV. Relative expression of transcripts measured against the expression of transcripts in un-transfected SKPs. Bars represent means ± s.e.m.

However, my studies were restricted due to limitations associated with the use of SKPs; long expansion times and low transfection efficiencies in comparison to MEFs. Given these shortcomings, I propose that SKPs may not be a suitable cellular source for non-integrative transcription factor mediated reprogramming approaches.

6.1.2.3.2 Characterization of Schwann cell heterogeneity post nerve-injury

Prior literature and my work in Chapter 3 cemented the idea that Schwann cells in development and post-injury are heterogenous in composition, and differ in their proliferative as well as myelination capacity258,260,401,629. Repair Schwann cells populating the site of injury promote regeneration, by creating an environment supportive of axonal survival, growth and remyelination258,629. Repair Schwann cells may exhibit varying repair potential, depending on the degree of repair program induction in these cells258,629. Additionally, only a small subset of endogenous Schwann cells may be responsive to the nerve-injury cue, and de-differentiate into repair Schwann cells401,629. Thus, it becomes important to examine Schwann cell heterogeneity to be able to isolate Schwann cells which can contribute towards a repair response. Apart from the known regulators and markers of repair Schwann cells (Jun324, Pax3, Shh, Olig1258,629), a lot remains to be determined about repair Schwann cells, and is an area of active interest and research.

203

To assess Schwann cell population complexity, a genetic barcoding approach to trace cell lineage640 may prove informative and allow for determination of novel cell surface biomarkers to identify and isolate repair Schwann cells. Such a barcoding approach was previously used for studying the heterogeneity detected in young and aged hematopoetic stem cell pool, and provides information on how many clones from young/aged hematopoetic stem cell pool contribute towards hematopoiesis640. The genetic barcoding approach involves isolation and individual tagging of Schwann cells with heritable barcodes (CellTracker 50M Lentiviral Barcode Library; Cellecta). These barcodes comprise of non-coding stretches of DNA as well as a GFP reporter for visual readout. The tagged Schwann cells can then be transplanted in a peripheral nerve injury model.

In the likely scenario of nerve repair being accomplished by a subset of Schwann cell clones, I expect an enrichment of a select few barcodes post-transplant. In contrast, if all Schwann cells are equally potent at adaptive reprogramming post-injury and can readily convert to a repair Schwann cell state, then a heterogenous mix of barcodes will be detected post the nerve regeneration period. During analysis, Next Generation DNA Sequencing will be performed using index primer sequences common to all barcodes, to compare the normalized distributions of barcodes in starting and end-stage Schwann cell populations. I expect that only a subset of Schwann cells carry an endogenous repair capacity, and these cells will populate the site of injury such that only a subset of barcodes will be recovered post repair. Thus, using such a genetic barcoding approach will help understand Schwann cell heterogeneity post-nerve injury.

Another area of interest is to perform single cell RNAseq analyses on isolated Schwann cells to stratify the population and to identify genes that distinguish one type of Schwann cell from another. This approach is currently being conducted by our collaborators, Dr. Raj Midha and Dr. Jeff Biernaskie.

6.1.2.3.3 Improving reprogramming protocols for Schwann cell generation

Previous cell reprogramming strategies involving viral vectors used for generating induced Schwann cells have reported a reprogramming efficiency of 2.02-43%468,470,471, while my reprogramming strategy had a ~1% conversion efficiency. Apart from lentiviral and episomal approaches for promoting transcription factor mis-expression in cells, alternate reprogramming approaches are now being tested. Lineage conversion using a self-replicative RNA (srRNA) approach has been used for the generation of hiPSCs from fibroblasts recently with a ~90%

204 reprogramming efficiency641,642. Transfection of srRNA coding for the gene(s) of interest, allows for high sustained protein levels, eliminating the need for multiple transfections. Additionally, similar to episomal vectors, srRNA approach is non-integrative in nature making them clinically relevant641,642. This approach is based on the Venezuelan Equine Encephalitis (VEE) alphavirus, consisting of sequence encoding four non-structural proteins required for the replicative cycle, and a subgenomic T7 promoter which drives the expression of the structural genes that follow. The VEE srRNA structural genes can be replaced with desired genes of interest641,642. Compared to endogenous transcription, the srRNA approach generates 10 times more sub-genomic RNA where the desired genes are compared to starting genome RNA643. Thus, such alternate approaches may prove beneficial in sustaining exogenous gene expression in host cells for an extended period, promoting target gene expression and generating induced Schwann cells in a dish compared to traditional viral/small molecule reprogramming approaches. I predict that srRNA constructs expressing the triple transcription factor combinations (Pax3-Jun-Sox2 and Sox10-Jun-Sox2) will result in higher and sustained expression of exogenously induced genes and promote strong induction of Schwann cell marker expression resulting in efficient induced repair Schwann cell generation.

6.2 Conclusions

Recent advances in the field of neuroscience have allowed for improved approaches to be used for dissecting the roles of glial cells in a healthy and diseased state. EVs exhibit an entire gamut of functions in normal and cancerous cells, ranging from proliferation110, cell survival136, cell death203,204,644, as well as tumor cell metastasis132,206. Advances in EV isolation, visualization, and characterization have allowed for their use as biomarkers in liquid biopsies from tumor patients. Through my thesis work, I have paved the way towards understanding how EVs function in the ODG model system, which will have direct implications not only in the glioma field, but also in cancer biology.

Cellular reprogramming studies have been actively worked upon in the past decade since the first Takahashi et al.318 study. Recent advances in this field have further cemented the hope that in vivo fate conversion of target cells will be the future of regenerative medicine474, which can then be applied to not only nerve repair models, but also other diseased conditions including Amyotrophic Lateral Sclerosis and Alzheimer’s disease. Through my studies on Schwann cell development in a mouse model, I have succeeded at concisely defining the spatio-temporal

205 expression pattern of some of the key regulators of Schwann cell development as well as repair Schwann cells, and used this knowledge to formulate reprogramming strategies for generation of Schwann cells with improved repair phenotype, which will prove beneficial for informing future reprogramming studies.

206

Appendices

Table A.2.1. Tabulation of results

207

208

Table A.2.2. Mass spectrometry analysis protocol settings.

Method of Q Exactive HF-X

OVERALL METHOD SETTINGS

Global Settings

Use lock masses best

Lock mass injection

Chrom. peak width (FWHM) 15 s

Time

Method duration 60.00 min

Customized Tolerances (+/-)

Lock Masses

Inclusion

Exclusion

209

Neutral Loss

Mass Tags

Dynamic Exclusion

Experiment

FULL MS / DD-MS (TOPN)

General

Runtime 0 to 60 min

Polarity positive

In-source CID 0.0 eV

210

Default charge state 2

Inclusion

Exclusion