A Multi-Angled Approach to Discover and Improve Skeletal Muscle Stem Cell Therapies

Sadegh Davoudi

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Institute of Biomaterials and Biomedical Engineering University of Toronto

Sadegh Davoudi

Doctor of Philosophy

Institute of Biomaterials and Biomedical Engineering University of Toronto

2019 Abstract

Skeletal muscle plays an essential role in locomotion, metabolism, and thermoregulation. An intriguing characteristic of skeletal muscle is its remarkable regenerative capacity: a highly orchestrated cellular process, in which the resident muscle stem cells (MuSCs) play a central role. Upon tissue injury, quiescent MuSCs are activated and give rise to a population of primary myoblasts (pMBs). pMBs undergo several rounds of division and ultimately fuse with one another to form multi-nucleated myofibers and repair the muscle. As observed in a wide range of conditions such as muscular dystrophies and aging, disruptions in the repair process can lend to impaired regeneration and progressive muscle wasting. Strategies to restore strength and function to pathological muscle include cell-based therapies to replace defective myogenic cells, and treatments to restore the endogenous repair process. Despite substantial advances, these treatment options are still in the early stages of clinical translation. The specific goal of this thesis is to improve upon currently available strategies as well as to lay the groundwork for the emergence of new therapeutic entry-points for skeletal muscle treatments.

In the area of cell transplantation therapy, challenges remain in producing clinically-relevant numbers of cells that, as a population, possess high regenerative potency to produce skeletal muscle and repopulate the stem cell niche. We demonstrate that by using a bioactive hydrogel ii

(HAMC) as the cell delivery vehicle, it is possible to improve MuSC transplantation outcomes, and thereby reduce the overall number of required MuSCs. Next, we address the issue of producing sufficient numbers of highly regenerative myogenic cells. Using a high-throughput drug screen and in-vivo intramuscular transplantation assay validation, we identify epidermal growth factor receptor (Egfr) and vascular endothelial growth factor receptor 2 (Vegfr2, Kdr) as new druggable targets, that upon inhibition, produce a population of cultured MuSCs with greater regenerative potency than control treated. Finally, using single-cell RNA sequencing, we shed light on the diversity and intercommunication of cells present in skeletal muscle. This dataset serves as a valuable resource through which new regulators of MuSCs and other cells in skeletal muscle can be evaluated with an eye towards skeletal muscle regenerative medicine applications.

iii

Acknowledgments

First and foremost, I would like to thank my supervisor, Dr. Penney Gilbert, for her continuous support, advice, and encouragement throughout the past few years. Penney’s enthusiasm for research and her thoughtful feedback and guidance inspired me to be a better scholar and allowed me to find my place in the field at my own pace while ensuring I develop my capabilities as a researcher. Thank you, Penney, for taking a chance on me, continuing to challenge me, and for being a shining example of a great scientist. I am also indebted to my supervisory committee members: Drs. Craig Simmons and Edmond Young. Your time and constructive scientific critique provided me with valuable feedback on my work and helped sharpen my research and analysis.

Throughout my time in the Gilbert lab, I have watched the lab grow from a handful of enthusiastic students to a full-fledged powerhouse producing state-of-the-art research. I want to convey my gratitude to my lab members, current and past, without whom I could not have completed this journey. Mohsen, we began and completed this journey together; thank you for being there during my highs and my lows, and always willing to lend a helping hand. Gini, thanks being an amazing undergraduate student! Gini, Bella, Mohammad, Majid, Aliyah, Richard, Ben, Min, Olivia, Louise, and all others, thank you for bringing your amazing positive energy to the lab and making it such a collaborative and welcoming environment.

To my parents, Hadi and Monir, thank you for sparking my interest in research and your continuous support throughout both my personal and academic life. I will be forever grateful for all the sacrifices you’ve made for me. And to my siblings, Mohammad and Mahtab, thank you for being the wonderful people you are and the memorable memories we’ve built together. Without all of you, I would not be where I am today.

And lastly, I want to acknowledge the contributions of my wife, Fatemeh. We started our journey together at the same time I began my PhD and throughout this time, you have motivated me, challenged me, made me smile, and lead by example how to live a good life. Your love, support, and encouragement have made this dissertation a reality. Thank you!

Table of Contents

Acknowledgments...... iv

Table of Contents ...... v

List of Abbreviations ...... x

List of Tables ...... xiv

List of Figures ...... xv

Chapter 1 Introduction ...... 1

Introduction ...... 1

1.1 Skeletal muscle ...... 1

1.2 Skeletal muscle homeostasis ...... 3

1.2.1 Muscle stem cells in quiescent muscle ...... 4

1.2.2 Vasculature and endothelial cells ...... 7

1.2.3 Interstitial cells ...... 7

1.2.4 Motor neurons ...... 8

1.3 Skeletal muscle regeneration ...... 9

1.3.1 Muscle stem cells in skeletal muscle regeneration ...... 11

1.3.2 Immune response ...... 17

1.3.3 Fibro-adipogenic progenitors ...... 18

1.3.4 Endothelial cells ...... 19

1.4 Skeletal muscle pathology and degeneration ...... 20

1.4.1 Aging...... 20

1.4.2 Muscular dystrophies ...... 22

1.5 Skeletal muscle treatments ...... 25

1.5.1 Cell based therapies ...... 25

1.5.2 Enabling endogenous repair ...... 28

1.6 Thesis Motivation, aims, and approach ...... 30 v

1.6.1 Thesis motivation and aims ...... 30

1.6.2 Thesis Overview ...... 32

Chapter 2 Muscle stem cell intramuscular delivery within hyaluronan methylcellulose improves engraftment efficiency and dispersion ...... 34

Muscle stem cell intramuscular delivery within hyaluronan methylcellulose improves engraftment efficiency and dispersion ...... 35

2.1 Abstract ...... 35

2.2 Introduction ...... 36

2.3 Results ...... 39

2.3.1 Muscle stem cell delivery within HAMC improves engraftment efficiency and dispersion ...... 39

2.3.2 HAMC improves muscle stem cell ejection efficiency ...... 43

2.3.3 HAMC promotes MuSC proliferation via a CD44-independent mechanism ...... 46

2.3.4 HAMC does not modify the skeletal muscle innate immune response ...... 48

2.3.5 HAMC prevents the active clearance of transplanted myogenic cells ...... 51

2.4 Discussion ...... 53

2.5 Limitations and future work...... 59

2.6 Materials and methods ...... 60

2.6.1 Animals ...... 60

2.6.2 Hyaluronan (HA) and methylcellulose (MC) preparation ...... 60

2.6.3 Primary myoblast isolation ...... 61

2.6.4 Muscle stem cell isolation ...... 61

2.6.5 Myogenic cell culture ...... 62

2.6.6 Myogenic cell transplantation ...... 63

2.6.7 Immunohistochemistry ...... 63

2.6.8 Flow cytometry analysis ...... 64

2.6.9 EdU analysis ...... 65

2.6.10 Injured skeletal muscle extract preparation ...... 65

2.6.11 Cell viability assays ...... 65

2.6.12 Cell ejection efficiency measurement ...... 66

2.6.13 Dispersion analysis ...... 66

2.6.14 Immune response quantification ...... 66

2.6.15 Imaging and microscope ...... 67

2.6.16 Statistical analysis ...... 67

Chapter 3 Small molecule inhibition of EGFR or KDR increases the regenerative potency of the cultured MuSC population ...... 68

Small molecule inhibition of EGFR or KDR increases the regenerative potency of the cultured MuSC population ...... 69

3.1 Abstract ...... 69

3.2 Introduction ...... 70

3.3 Results ...... 72

3.3.1 A high-throughput drug screen to identify small molecular inhibitors for myogenic cell expansion ...... 72

3.3.2 Analysis of drug screen and identification of potential candidates for ex-vivo expansion of muscle stem cells ...... 74

3.3.3 Transplantation assay reveals Egfr and Kdr inhibition as methods to increase regenerative capacity of in-vitro cultured MuSCs ...... 78

3.4 Discussion ...... 88

3.5 Limitations and future work...... 92

3.6 Materials and methods ...... 93

3.6.1 Animals ...... 93

3.6.2 Murine muscle stem cell isolation ...... 94

3.6.3 Murine primary myoblast isolation ...... 95

3.6.4 High-throughput drug screen ...... 95

3.6.5 Myogenic cell culture ...... 95

vii

3.6.6 Myogenic cell transplantation ...... 96

3.6.7 Immunohistochemistry ...... 97

3.6.8 Imaging and microscope ...... 97

3.6.9 Statistical analysis ...... 97

Chapter 4 Single-cell RNA sequencing of mono-nuclear cells in murine skeletal muscle ...... 98

Single-cell RNA sequencing of mono-nuclear cells in murine skeletal muscle ...... 99

4.1 Abstract ...... 99

4.2 Introduction ...... 100

4.3 Results ...... 101

4.3.1 Single cell RNA sequencing of mononuclear cells in murine adult skeletal muscle ...... 101

4.3.2 Characterization of 2 identified endothelial cell populations ...... 104

4.3.3 Gpr37l1 as a marker for Schwann cell isolation directly from skeletal muscle ..108

4.3.4 Cellular interactome in adult skeletal muscle ...... 110

4.4 Discussion ...... 116

4.5 Limitations and future work...... 119

4.6 Materials and methods ...... 121

4.6.1 Skeletal muscle single cell isolation for scRNA seq ...... 121

4.6.2 Staining of single cell preparations and tissue sections ...... 121

4.6.3 Immunohistochemistry and Immunocytochemistry ...... 122

4.6.4 Single cell RNA sequencing analysis ...... 122

Chapter 5 Conclusions, impact and future work ...... 124

Summary and impact...... 124

5.1 Summary of results ...... 124

5.2 Future work ...... 125

5.3 Impact ...... 126

viii

References ...... 128

Appendix I: Summary of drug screen results ...... 167

Appendix II: Supplemental information for Chapter 4 ...... 175

List of Abbreviations

Abcg2 ATP binding cassette subfamily g Member 2 Akt Protein kinase B Alpl Tissue non-specific alkaline phosphatase Angpt1 Angiopoietin 1, Ang1 AurkA Aurora kinase A BaCl2 Barium Chloride BSA Bovine serum albumin Cadm Cell adhesion molecule Cdh5 Cadherin 5 Cdkn2a Cyclin dependent kinase inhibitor 2A Clk Cdc2-like kinase c-Met Tyrosine-protein kinase Met, Hepatocyte growth factor receptor (Hgfr) Col Collagen CSA Cross-sectional area Cspg4 Chondroitin sulfate proteoglycan 4 Dag1 Dystroglycan DAPC Dystrophin-associated protein complex Dgk Diacylglycerol kinase 1 Dhh Desert hedgehog Dll Delta like DMD Duchenne muscular dystrophy Dsh Disheveled ECM Extra cellular matrix ECs Endothelial cells EDL Extensor digitorum longus muscle eEf2k Eukaryotic elongation factor 2 kinase Egf Epidermal growth factor Egfr Epidermal growth factor receptor Ephb4 EPH receptor B4 ER Endoplasmic reticulum

Erk Extracellular signal-regulated kinase FACS Fluorescence activated cell sorting FAP Fibroadipogenic progenitor Fgf2 bFGF, Fibroblast growth factor 2 Fgfr Fibroblast growth factor receptor Fn1 Fibronectin Foxo1 Forkhead Box O1 Fzd Frizzled Gcsf Granulocyte-colony stimulating factor Gdf Growth differentiating factor GFP Green fluorescent protein Gpr37l1 G protein-coupled receptor 37-like 1 Gsk3 Glycogen synthase kinase 3 Gy Gray (Irradiation dose unit) HA Hyaluronan HAMC Hydrogel composed of a physical blend of HA and MC Hes1 Hairy and enhancer of split Hey Hairy and enhancer of split related with YRPW motif 1 HeyL Hairy and enhancer of split related with YRPW motif 3 Hgf Hepatocyte growth factor Igf-1 Insulin-like growth factor 1 Igf1r Insulin-like growth factor 1 receptor IHC Immunohistochemistry Il-1β Interleukin 1 beta Il-6 Interleukin 6 Il-6r Interleukin 6 receptor Ire1 Inositol-requiring enzyme 1 Itga7 α7 Integrin Itgb1 β1 Integrin Jag Jagged Jak Janus kinase Jnk c-Jun N-terminal kinase

Kcna Potassium voltage-gated channel subfamily A Kdr VEGFR2, Vascular endothelial growth factor receptor 2 Lyve1 Lymphatic vessel endothelial hyaluronan receptor MAPK Mitogen-activated protein kinases Mark2 Microtubule affinity regulating kinase 2 MC Methylcellulose Mef2 Myocyte enhancer factor-2 MMP Matrix metalloproteinase Mrf4 Myogenic regulatory factor 4 MSC Mesenchymal stem cell mTOR Mammalian target of rapamycin MuSC Muscle stem cell Myf5 Myogenic factor 5 MyHC Myosin heavy chain MyLC Myosin light chain MyoD Myogenic differentiation 1 Myog Myogenin Ncam1 Neural cell adhesion molecule 1 NF-kB nuclear factor kappa-light-chain-enhancer of activated B cells Ngf Neural growth factor NICD Notch intracellular domain NK cells Natural killer cells NMJ Neuromuscular junction NO Nitric oxide OPMD Oculopharyngeal muscular dystrophy Pabpn1 Poly(A) binding protein nuclear 1 PAR partitioning defective Pax7 Paired box 7 PCA Principal component analysis PCP Planar cell polarity Pdgfr Platelet derived growth factor Pecam1 Platelet and endothelial cell adhesion molecule 1; Cd31

xii

Perk PKR-like endoplasmic reticulum kinase Pkr Protein kinase R Plp1 Proteolipid protein 1 pMB Primary myoblast PNS Peripheral nervous system Rb1 Retinoblastoma tumor suppressor Rbpj Recombining binding protein suppressor of hairless Rhamm Receptor for hyaluronan-mediated-motility ROCK Rho-associated protein kinase Sca1 Stem cells antigen-1 Scarb1 Scavenger receptor class B Member 1 Sdc Syndecan Sox SRY-box Spry1 Sprouty-1 Stat Signal transducer and activator of transcription TA Tibialis anterior muscle Tcf4 Transcription factor 4 Tgfβ Transforming growth factor beta TIMP Tissue inhibitor of metalloproteinases Tnfα Tumor necrosis factor α Tnmd Tenomodulin Tregs Regulatory T cells t-SNE t-distributed stochastic neighbor embedding Vangl2 Van Gogh-like protein 2 Vcam1 Vascular cell adhesion molecule 1 Vegfa Vascular endothelial growth factor A Wisp1 Wnt1-inducible-signaling pathway protein 1

xiii

List of Tables

Table 3.1: Small molecule inhibitors selected from high-throughput drug screen ...... 78

Table A.1: List of compounds designated as hits in drug screen (compound name, targets, and successful doses) ...... 167

Table A.2: Synonyms of compound targets in drug library...... 173

Table A.3: Top 80 differentially expressed genes in identified clusters identified in sc-RNA seq ...... 175

Table A.4: Top 40 differentially expressed genes between EC-1 and EC-2 ...... 177

xiv

List of Figures

Figure 1.1: Schematic illustrating skeletal muscle structure...... 2

Figure 1.2: Cellular dynamics of skeletal muscle regeneration in response to injury ...... 10

Figure 1.3: Myogenic lineage progression...... 11

Figure 1.4: Asymmetric vs symmetric division ...... 14

Figure 2.1: MuSC isolation and transplantation schematic ...... 40

Figure 2.2: Delivering MuSCs within HAMC improves engraftment efficiency and dispersion 42

Figure 2.3: HAMC improves ejection efficiency without altering cultured myogenic cell viability ...... 44

Figure 2.4: HAMC influences MuSC proliferation via a CD44-independent mechanism ...... 47

Figure 2.5: HAMC does not modify Ly6G+ or CD68+ immune cell incidence in the first 24 hours post-transplantation ...... 50

Figure 2.6: HAMC increases retention of transplanted cells post-transplantation ...... 52

Figure 3.1: Schematic of high-throughput drug screen to identify small molecule inhibitors for MuSC expansion ...... 73

Figure 3.2: Clonal analysis of ex-vivo cultured MuSCs in the presence of EGFR inhibitor reveals reduced time between divisions resulting in higher number of cells after 7 days...... 76

Figure 3.3: Dose-response curves of small molecule inhibitors selected for in-vivo functional assays ...... 77

Figure 3.4: Freshly isolated MuSCs have a higher regenerative capacity compared to primary myoblasts (pMBs) ...... 81

Figure 3.5: Schematic of in-vivo transplantation assay to assess effectivity of selected small molecule inhibitors in expanding the myogenic cells with increased regenerative capacity ...... 83

Figure 3.6: Representative images of in-vivo transplantation assay results ...... 83

Figure 3.7: Inhibiting EGFR and KDR activity increases the regenerative capacity of MuSCs cultured for 7 days...... 85

Figure 3.8: EGFR and KDR inhibition of cultured MuSCs does not lead to changes in the number of GFP+ fibers...... 87

Figure 3.9:GFP+ fiber diameter analysis suggests a trend towards fiber hypertrophy in fibers formed from MuSCs exposed to EGFR, KDR inhibitors...... 88

Figure 4.1: Schematic of experimental procedure to obtain mono-nuclear cell isolation for single cell RNA sequencing ...... 102

Figure 4.2: scRNA-seq analysis identifies major cell populations in skeletal muscle ...... 103

Figure 4.3: Dot plot of top 5 differentially expressed genes in each identified cluster ...... 104

Figure 4.4: The two identified endothelial cell clusters cannot be distinguished using venous/arterial or lymphatic/vascular endothelial cell or pericyte markers ...... 106

Figure 4.5: Alpl, Kcna5, and Scarb1 are not suitable markers for separating the two endothelial sub-populations ...... 107

Figure 4.6: Abcg2, muscle side-population marker, is differentially expressed in the EC - 1 cluster ...... 108

Figure 4.7: Gpr37l1 can be used as a marker to isolate Schwann cells from dissociated skeletal muscle ...... 110

Figure 4.8: Putative signaling between major skeletal muscle cell populations...... 112

Figure 4.9: FAPs are the source of key growth factors in skeletal muscle ...... 113

Figure 4.10: Cellular source of ligands presented to MuSCs in homeostatic skeletal muscle ... 115

xvi

Chapter 1 Introduction Introduction 1.1 Skeletal muscle

Skeletal muscle is a highly secretive organ, making up to 45% of the body mass (Dumont et al., 2015b; Pedersen and Febbraio, 2012; Wosczyna and Rando, 2018). It is also an essential component of a variety of different functions such as locomotion, thermoregulation, as well as breathing. Healthy adult skeletal muscle is composed of a dense network of parallel striated muscle fibers held together by a rich extracellular matrix (ECM) (Sanes, 2003). These bundles of muscle fibers are connected to the bone through myotendinous junctions and are responsible for generating the contractile force required for voluntary movement (Lieber, 2002).

Muscle fibers (myofibers) are among the most unique cells in the body, reaching lengths of up to 30 cm in humans (Heron and Richmond, 1993). Myofibers are highly specialized cells, converting chemical energy to mechanical work. They are large syncytial structures formed through the fusion of myogenic progenitor cells, in which the numerous myonuclei are required to meet the high-protein synthesis demands of the myofiber (Dumont et al., 2015b; Judson et al., 2018).

An important feature of skeletal muscle is that it is highly adaptive and exhibits plasticity in response to external and internal stimuli. In addition to regulating the structure, function, and secretory profile of skeletal muscle, exercise and resistance training have been demonstrated to lead to muscle hypertrophy and strength increase (Fluck, 2006; Pedersen and Febbraio, 2012). Alternatively, a host of diseases and conditions, including but not limited to immobility, cancer, aging, denervation, genetic disorders, and critical illness, can lead to muscle wasting and loss of function (Ali and Garcia, 2014; Cosgrove et al., 2009; Dodson et al., 2010; Fitts et al., 2000; Puthucheary et al., 2013).

Finally, skeletal muscle is considered as one of the most regenerative organs in the body. Various studies have demonstrated that skeletal muscle can fully regenerate within a few weeks following major injuries (Ehrhardt and Morgan, 2005; Hardy et al., 2016). Skeletal muscle

regeneration is a highly orchestrated cellular process involving a variety of cells including but not limited to endothelial cells, fibroadipogenic progenitors, nerve cells, and the immune system. The main cell population responsible for the remarkable regenerative capacity of striated muscle are muscle stem cells (MuSCs), also called satellite cells due to their anatomical position with respect to the myofibers (Mauro, 1961; Relaix and Zammit, 2012). Satellite cells are mitotically quiescent under homeostatic conditions. Upon injury, they become activated and give rise to a population of transient amplifying cells also called primary myoblasts (pMBs). pMBs undergo several rounds of division and ultimately fuse either with one another to form new myofibers, or into pre-existing damaged fibers and repair them. A subset of the satellite cells will self-renew and return to quiescence following regeneration (Bentzinger et al., 2013a; Wosczyna and Rando, 2018).

Figure 1.1: Schematic illustrating skeletal muscle structure. Skeletal muscle is composed of a compact network of multinucleated muscle fibers, packed together by the ECM. A network of blood vessels, providing nutrients and oxygen, surround the myofibers, as well as a variety of interstitial cells. Satellite cells are found between the sarcolemma of the muscle fibers and the basal lamina of the basement membrane. The muscle fibers are connected to the bones through myotendinous junctions (Image from Dumont et al., 2015b with permission from John Wiley & Sons Inc.)

1.2 Skeletal muscle homeostasis

Homeostatic skeletal muscle has a relatively low turnover rate. This is due to the fact that MuSCs and the interstitial and vessel-associated cells in homeostatic skeletal muscle are generally mitotically quiescent, showing little or no signs of proliferation. Additionally, the myonuclei in skeletal muscle fibers are post-mitotic as well (Bentzinger et al., 2013a).

In homeostatic skeletal muscle, muscle fibers are encased in a rich ECM. The ECM immediately surrounding the individual fibers can be broken down into the endomysium and the basement membrane. The endomysium is mainly composed of collagens and houses the intricate nervous and vascular systems surrounding the muscle fibers. Between the endomysium and the plasmalemma (myofiber membrane), lies the basement membrane. The basement membrane itself is composed of 2 layers: the external reticular lamina and the internal basal lamina. The reticular lamina is a fibrillar structure mainly composed of collagen fibrils. The main constituents of the basal lamina on the other hand are collagen IV, laminin and proteoglycans. Integrins and dystroglycans on the surface of the fiber plasmalemma ultimately connect the fiber and its cytoskeleton to the ECM via the basal lamina (Sanes, 2003).

The maintenance and survival of stem cells is a highly regulated process, with inputs from the immediate local environment of the stem cells, i.e. the stem cell niche. The stem cell niche is a complex and dynamic structure, specifically adapted to provide structural and trophic support, topographical information, and/or physiological cues (Jones and Wagers, 2008). Most studies in homeostatic skeletal muscle have been focused on MuSCs and factors regulating their quiescence. In addition to the MuSCs and their immediate niche, interstitial cells, motor neurons, blood vessels and the factors associated with these cells are also present in homeostatic skeletal muscle and play a role in regulating MuSC and muscle function. However, their role in skeletal muscle has often been studied in the context of regeneration and MuSC function. As such, further research is required to fully understand the cellular and molecular underpinnings amongst the cells present in skeletal muscle and their role in muscle homeostasis. Here we present a brief review on the cells and the ECM present in homeostatic skeletal muscle.

1.2.1 Muscle stem cells in quiescent muscle

Quiescent MuSCs are located between the fiber sarcolemma and the basal lamina. In addition to their anatomical location, quiescent MuSCs can be identified via their unique markers. Paired box protein Pax7 is the most widely recognized marker of satellite cells since all MuSCs express high levels of Pax7 in their nuclei (Gros et al., 2005; Kassar-Duchossoy, 2005; Seale et al., 2000). However, Pax7 expression is not unique to quiescent MuSCs and is also present in activated MuSCs and primary myoblasts; therefore, it cannot be used on its own to identify quiescent MuSCs. Other muscle specific transcription factors such as Myod1 however, are not expressed at detectable levels in quiescent MuSCs and can therefore be used to distinguish activated from quiescent satellite cells. In addition to transcription factors, various surface markers can be used to identify quiescent MuSCs. M-cadherin, Integrin a7 (Itga7), Syndecan-3 (Sdc3) and Syndecan-4 (Sdc4), Calcitonin receptor, Caveolin-1, Cd34, Vcam1, and Ncam1 are all present at the surface of MuSCs in their quiescent niche (Beauchamp et al., 2000; Cornelison et al., 2001; Dumont et al., 2015b; Fukada et al., 2007; Irintchev et al., 1994).

Studies comparing the transcriptome of quiescent and activated MuSCs have revealed ~500 genes that are upregulated in quiescent MuSCs. These genes were associated with processes such as cell-cell adhesion, cell-ECM interactions, and regulation of cell growth (Farina et al., 2012; Fukada et al., 2007; García-Prat et al., 2016; Liu et al., 2013; Lukjanenko et al., 2016; Pallafacchina et al., 2010). This suggests that MuSC quiescence is an active process regulated by intrinsic and extrinsic factors provided to the MuSC by its surrounding niche. Here, we will present an overview of several of the factors involved in the regulation of MuSC quiescence.

1.2.1.1 Notch signaling

Notch signaling, demonstrated to be active in dormant MuSCs, is one of the most widely studied pathways regulating MuSC quiescence (Bjornson et al., 2012; Philippos et al., 2012). Notch proteins are a family of transmembrane proteins that are expressed by quiescent MuSCs (Verma et al., 2018). Exposure of notch ligands (Jag1, Jag2, Dll1, Dll3, and Dll4) presented by neighboring cells leads to the activation of the Notch signaling pathway (Chillakuri et al., 2012; Koch et al., 2013). In skeletal muscle, myofibers present the Notch ligand Dll1 at their membrane (Conboy and Rando, 2002). Additionally, recent studies have demonstrated that

vascular endothelial cells express Dll4, another Notch ligand (Verma et al., 2018). Given that quiescent MuSCs are located on the myofiber and are found in the proximity of blood vessels, it appears as though the main sources of Notch signaling in quiescent skeletal muscle are the endothelial cells and myofibers (Christov et al., 2007; Verma et al., 2018).

Notch signaling activates various genes (such as Hes1, Hes5, Hey, and HeyL) via the Rbpj transcription factor. Disruption of the Notch signaling pathway leads to impaired myogenesis. Over-expression of constitutively activated Notch 1 intracellular domain (NICD) inhibits MyoD expression and myogenic differentiation in addition to promoting self-renewal of the satellite cells (Wen et al., 2012). Conversely, inhibiting Notch signaling or any of its downstream effectors, leads to the activation and differentiation of MuSCs, and in the long run depletes the muscle stem cell pool (Bjornson et al., 2012; Kuang et al., 2007; Philippos et al., 2012).

1.2.1.2 Post-transcriptional regulation of mRNA

MicroRNAs (miRNA) are small non-coding RNA molecules involved in post-transcriptional regulation of mRNA molecules (Bartel, 2018). The majority of quiescent MuSCs do not express the myogenic regulatory factor Myf5 but are able to immediately upregulate it following injury. Researchers determined that although quiescent MuSCs express the Myf5 gene, miRNA-31 regulates the translation of Myf5 by sequestering it in mRNP granules. Upon receiving activation cues, the granules are dissociated allowing the translation of the Myf5 protein which then leads to the myogenic progression of the MuSCs (Crist et al., 2012). In a separate study, comparing the transcriptome of quiescent and activated satellite cells revealed that 22 miRNAs were upregulated specifically in quiescent MuSCs. miRNA-489 was identified as a regulator of MuSC quiescence, through post-transcriptional suppression of Dek. Upon translation, Dek protein promotes the transient expansion of the primary myoblasts. Therefore by inhibiting Dek translation, miRNA-489 promotes MuSC quiescence (Cheung et al., 2012). More recently, Notch signaling in quiescent MuSCs was demonstrated to induce transcription of miRNA-708 which in turn antagonized Tensin3 activity. Tensin3 is involved in the migration of MuSCs outside of the basement membrane and in MuSC proliferation, differentiation, and fusion. In vivo inhibition of miRNA-708 lead to the disruption of MuSC quiescence and self-renewal (Baghdadi et al., 2018a). These studies demonstrate that although the exact role of many non-coding RNA has yet

to be determined, they can play an important role in regulating MuSC quiescence in the homeostatic niche.

1.2.1.3 Cell-cycle regulators and pro-proliferation pathways

Several cell-cycle regulators are expressed at high levels in MuSCs of homeostatic skeletal muscle. p27kip1 is a cell cycle inhibitor protein that is highly expressed in quiescent MuSCs (Chakkalakal et al., 2014). Retinoblastoma tumor suppressor protein (Rb1) is another cell cycle inhibitor that is detected at elevated levels in dormant MuSCs (Hosoyama et al., 2011). Conditional deletion of Rb1 in Pax7-expressing cells and inhibition of p27kip1 leads to increased proliferation of the satellite cells, suggesting that cell cycle regulators also play a role in maintaining MuSC quiescence.

Additionally, quiescent satellite cells express factors inhibiting pro-proliferation pathways. c- Met, Fgfr1, and Fgfr4 are receptor tyrosine kinases present on the surface of quiescent MuSCs. Rtk ligands, such as Fibroblast growth factor 2 (Fgf2) and hepatocyte growth factor (Hgf), lead to the activation and subsequent proliferation of quiescent MuSCs (Jones et al., 2001; Sheehan and Allen, 1999; Webster and Fan, 2013). Sprouty-1 (Spry1) is a negative regulator of Rtk signaling present in quiescent satellite cells, preventing the activation and proliferation of the MuSCs through Erk1/2 inhibition. MuSCs downregulate Spry1 upon activation, and subsequently upregulate it to return to quiescence (Shea et al., 2010). In a separate study, Angiopoietin 1 (Angpt1), secreted by endothelial and periendothelial cells (endomysial fibroblasts and smooth muscle cells), was demonstrated to prevent the activation and proliferation of MuSCs, as well as cell-cycle arrest in activated progenitors (Abou-Khalil et al., 2009; Mofarrahi et al., 2015).

1.2.1.4 Interactions with the ECM

Quiescent MuSCs are consistently found to be located between the myofiber plasmalemma and the basal lamina, suggesting that the immediate surroundings of the satellite cells plays a role in their fate. The basal lamina is enriched with laminins and MuSCs can interact with them through expression of Itga7, Integrin β1 (Itgb1), and dystroglycan (Dag1) on their surface (Cohn et al., 2002; Mayer et al., 1997). ColV secreted by MuSCs has been shown to be an important regulator of MuSC quiescence through ColV – Calcitonin receptor (Calcr) signaling (Baghdadi et al., 6

2018b). Moreover, it is speculated that MuSCs stabilize the quiescent niche and prevent matrix metalloproteinase (MMP) mediated ECM degradation via expression of tissue inhibitor of metalloproteinases (TIMPs) (Pallafacchina et al., 2010).

1.2.2 Vasculature and endothelial cells

The high demands for oxygen and nutrients of the myofibers is met through an intricate vascular network surrounding the muscle fibers. Additionally, most MuSCs are found in the proximity of the vasculature and endothelial cells in homeostatic skeletal muscle (Christov et al., 2007). This close association between MuSCs and endothelial cells (ECs) hints towards the importance of the vascular system in MuSC quiescence. Indeed, Angpt1 secreted by periendothelial cells plays a role in maintaining MuSCs in a quiescent state while at the same time, stabilizing the vessels (Abou-khalil et al., 2010; Abou-Khalil et al., 2009). Furthermore, recent studies suggested the close proximity of MuSCs and ECs potentially allows for EC-mediated Dll4-Notch signaling required for MuSC quiescence. Interestingly, Dll4 production by ECs might be positively reinforced through Vegfa secreted by MuSCs, hinting towards reciprocal signaling between the two cell types (Verma et al., 2018).

1.2.3 Interstitial cells

The interstitial space between skeletal muscle fibers is rich with a variety of cells, some which have only been identified within the past decade. These cells are one of the main sources of ECM protein deposition, especially during skeletal muscle regeneration. They are also responsible for the production of a variety of growth factors as well.

1.2.3.1 Fibroadipogenic progenitors (FAPs)

Fibroadipogenic progenitors (FAPs) are mesenchymal progenitor cells that have garnered significant attention since their identification in 2010 (Joe et al., 2010; Uezumi et al., 2010). FAPs are considered the mesenchymal stem cells (MSCs) resident in skeletal muscle, due to their shared molecular and functional properties to MSCs: expression of platelet derived growth factor α (Pdgfrα), and the ability to differentiate into multiple mesodermal lineages (Wosczyna and Rando, 2018). FAPs in homeostatic skeletal muscle are quiescent and identified by the expression of Cd34, Sca1, and Pdgfrα and the absence of hematopoietic markers Cd45 and Cd31

(Bentzinger et al., 2013a; Joe et al., 2010). Interestingly, similar to MuSCs, FAPs are found in the vicinity of blood vessels under homeostatic conditions (Pretheeban et al., 2012). These cells are multipotent and in addition to adipo- and fibrogenesis, are capable of chondro- and osteogenesis as well (Wosczyna et al., 2012).

1.2.3.2 Fibroblasts

Fibroblasts are a subtype of interstitial cells that are non-vascular, non-epithelial, and non- inflammatory by nature, and reside in the interstitium between fibers during homeostasis (Murphy et al., 2011). Fibroblasts are major contributors to the ECM and basement membrane through the production of collagen, laminin, fibronectin, tenascin, and Ncam (Bentzinger et al., 2013a; Gatchalian et al., 1989). Despite the numerous studies on muscle resident fibroblasts, these cells do not appear to have a clear molecular identity. Fibroblasts are a heterogeneous cell population and finding markers to precisely isolate them has been a challenge. They were initially identified as cells from digested skeletal muscle that were able to quickly adhere to tissue culture plates (Richler and Yaffe, 1970). Later studies initially identified fibroblasts as Tcf4+ cells, and demonstrated that ablation of Tcf4+ cells resulted in impaired skeletal muscle regeneration (Kardon et al., 2003; Murphy et al., 2011). However, Tcf4 was shown to be expressed in the majority of the cell types in skeletal muscle, suggesting the observations from Tcf4+ ablated muscles were not solely due to fibroblast depletion (Murphy et al., 2011; Schaum et al., 2018). As such, studies on the exact role and function of fibroblasts during homeostasis and regeneration are hindered until the identification of distinguishing markers for fibroblasts.

1.2.4 Motor neurons

The innervating motor neurons in skeletal muscle control the voluntary contractions of muscle fibers through the neuromuscular junction (NMJ): a chemical synapse formed by the contact of a motor neuron and a muscle fiber (Levitan and Kaczmarek, 2015). In both human and murine homeostatic skeletal muscle, a higher incidence of perisynaptic MuSCs have been observed compared to extrasynaptic satellite cells (Kelly, 1978; Wokke et al., 1989). Additionally, it is well-documented that shortly following muscle denervation, the number of satellite cells increase similar to satellite cell proliferation due to muscle injuries. Long-term denervation on the other hand, results in a decline in satellite cell numbers due to decreased mitotic capability and

increased apoptosis of the MuSCs (Rodrigues and Rodrigues, 2010; Schultz, 1978; Viguie et al., 1997). Moreover, in developing muscle, satellite cells are seen close to the motor endplate, in addition to making close plasmalemmal contact with the Schwann cells surrounding the neurons (Kelly, 1978). Despite these observations, the underlying mechanism of neural influence on MuSC function, as well as the role of MuSCs in the regeneration and maintenance of the neuromuscular junction is not yet fully understood.

1.3 Skeletal muscle regeneration

In response to injury, a highly orchestrated regeneration process is activated in healthy skeletal muscle. Any disruption in the temporally defined cellular response, involving both tissue- resident cells and the infiltrating cells from the general circulation, results in impaired or incomplete regeneration (Wosczyna and Rando, 2018; Yin et al., 2013). Muscle regeneration occurs in three distinct, but overlapping stages: 1) the inflammatory response, 2) activation, proliferation, differentiation, and fusion of satellite cells, and 3) maturation of the newly formed fibers (Yin et al., 2013).

The first response to injury is inflammation and the necrosis of the damaged myofiber. Myofiber necrosis is accompanied by activation of calcium-dependent proteolysis as a result of calcium influx or calcium release from the sarcoplasmic reticulum and leads to tissue degeneration. Additionally, myofiber necrosis activates the complement cascade and generates inflammatory responses. Following the inflammatory response, leukocytes begin accumulating in the injured tissue through chemotaxis. Neutrophils arrive at the injured location as early as 1 hour following injury. Subsequently, macrophages infiltrate the damaged muscle in two waves. The first wave involves pro-inflammatory macrophages which are essential for debris clearance. Anti- inflammatory macrophages arrive in the second wave and persist until inflammation comes to an end.

Following muscle degeneration and debris clearance, a highly regulated regeneration process is initiated. Extensive cellular proliferation is a hallmark of this stage. Inhibiting cellular proliferation either through pharmacological inhibition (colchicine) or irradiation severely impairs muscle regeneration (Pietsch, 1961a; Quinlan et al., 1995). In this stage, satellite cells are activated and are able to actively migrate towards the damaged areas (Alfaro et al., 2011). 9

Following activation, MuSCs give rise to a large number of primary myoblasts (pMBs) which undergo several rounds of division, and finally differentiate and either fuse with one another to form new myofibers, or into existing damaged fibers and repair them. Newly formed fibers can often be identified through centrally located nuclei, as well as their relatively small diameters.

In the final stage of regeneration, the newly formed myofibers undergo hypertrophy and the myonuclei migrate to the periphery of the muscle fiber (Yin et al., 2013).

Next, we will review the cellular and molecular dynamics of muscle regeneration with an emphasis on MuSCs.

Figure 1.2: Cellular dynamics of skeletal muscle regeneration in response to injury The relative abundance of fibroadipogenic progenitors (FAPs), muscle stem cells (MuSCs), endothelial cells (ECs), fibroblasts, M1 and M2 macrophages, neutrophils, eosinophils, regulatory T cells (Tregs) in the first 14 days following injury is presented in the diagram above. (Adapted from Wosczyna and Rando, 2018 with permission from Elsevier)

1.3.1 Muscle stem cells in skeletal muscle regeneration

Figure 1.3: Myogenic lineage progression Following muscle injury, quiescent satellite cells (Pax7+/Myf5+/-) are activated and give rise to primary myoblasts (Pax7+/Myf5+/MyoD+). The satellite cell population can also self-renew to maintain the stem cell population. Primary myoblasts undergo several rounds of division and ultimately differentiate into myocytes (MyoD+/Myog+/Mrf4+). Myocytes can then either fuse with one another to form new myofibers (Myog+/Mrf4+/MyHC+).

1.3.1.1 Muscle stem cell activation

The majority of MuSCs respond rapidly to changes within their niche and are activated within a short period following muscle injury. One of the roles of the ECM is to sequester inactive growth factors, and to act as a local reservoir, to allow the rapid activation of the MuSCs in response to injury. Hepatocyte growth factor (Hgf) is one of the factors that is entrapped in the basal lamina and is released upon injury (Tatsumi et al., 1998). Upon its release, Hgf binds to c-Met, a receptor expressed in both quiescent and activated MuSCs (Allen et al., 1995). Activation of the Hgf/c-Met pathway leads to MuSC activation, increased pMB proliferation, and prevents myogenic differentiation. Basic fibroblast growth factor (Fgf2) is another growth factor that is sequestered in the ECM of uninjured muscle and is released following injury (DiMario et al., 1989). Similar to c-Met, Fgf receptors are present in both quiescent and activated MuSCs (Ernfors et al., 1991). Fgf2 activates the p38 mitogen-activated protein kinases (MAPK) pathways which leads to satellite cell activation and proliferation, whereas inhibiting the p38 MAPK pathway prevents MuSC activation (Jones et al., 2005; Maher, 1999). Interestingly, Sdc4

expression on quiescent MuSCs is necessary to facilitate both Hgf and Fgf2 mediated activation of MuSCs (Cornelison et al., 2004).

Igf-1 is another growth factor that can lead to MuSC activation (Musarò et al., 2001). The binding of Igf1 to its receptor, Igf1 receptor (Igf1r), promotes the Akt/mTOR pathway, which in turn downregulates the activity of the Foxo transcription factor. Downregulation of Foxo1 through Igf signaling has been demonstrated to suppress p27kip, a cell cycle inhibitor (previously discussed in 1.2.1.3), expressed in quiescent MuSCs (Machida et al., 2003). Additionally, activation of Igf1r in satellite cells leads to the expression of myogenic regulatory factors (MRFs) and initiates mitogenic and myogenic intracellular signaling cascades (Coolican et al., 1997; Florini et al., 2009; Musarò and Rosenthal, 2015).

Tumor necrosis factor α (Tnfα) is a pro-inflammatory factor rapidly released in the muscle following injury. Injecting Tnfα into uninjured skeletal muscle led to the activation of the quiescent satellite cells (Li, 2013). In a separate study, researchers have demonstrated that Tnfα leads to the activation of NF-kB pathway in satellite cells. NF-kB signaling in turn activates quiescent satellite cells by epigenetic silencing of Notch1, and the suppression of Notch signaling (Acharyya et al., 2010). Nitric oxide (NO) is another molecule that is immediately presented in damaged and injured muscle. NO has been demonstrated to stimulate matrix metalloproteinase (MMP) expression which leads to the degradation of the ECM to both release necessary growth factors as well as allowing the migration of the satellite cells to the injured regions (Kherif et al., 1999; Tatsumi, 2010). Inhibition of the enzyme required for NO production, NO synthase, leads to delayed activation of satellite cells (Anderson, 2000).

1.3.1.2 Primary myoblast proliferation and differentiation

Following activation, satellite cells give rise to a population of transient amplifying cells called primary myoblasts (pMBs). Myoblasts express the myogenic markers Pax7, Myf5, and MyoD (Figure 1.3). pMBs undergo several rounds of division to create sufficient myogenic cells for successful regeneration of damaged muscle and inhibiting this crucial expansion step impairs muscle regeneration (Pietsch, 1961b).

Various signaling pathways are involved in pMB proliferation. These pathways usually act by either repressing premature differentiation or by increasing cell cycle progression. As an example, Jak1/Stat1 signaling has been demonstrated to promote pMB proliferation by inhibiting pro-differentiation genes such as Mef2 (Sun et al., 2007). Interestingly, the initial proliferative phase of pMBs overlaps with the infiltration of pro-inflammatory macrophages (Figure 1.2). Various pro-inflammatory cytokines secreted by this first wave of macrophages, such as Il-6, Il- 1β, and Tnfα, have been demonstrated to increase the proliferation rate of pMBs (Tidball and Villalta, 2010).

Following the proliferative phase, the pMBs exit cell cycle and begin differentiating into myocytes (Figure 1.3). During this process, pMBs downregulate Pax7 and Myf5 expression, and express Myogenin (Myog) and Myogenic regulatory factor 4 (Mrf4). Various cues regulate the transition from pMBs to myocytes. MyoD is one of the most important genes in regulating the switch from proliferating pMBs to myocytes. During the differentiation phase, MyoD promotes cell cycle exit through p21 and p57 inhibitors. Additionally, it is essential for Myog and Mrf4 expression in differentiating myocytes (Cornelison et al., 2000; Halevy et al., 1995; Hollenberg et al., 2006). Myog expression leads to the activation of muscle specific genes involved in the contractile apparatus such as myosin heavy chain (MyHC), myosin light chain (MyLC), and muscle creatine kinase (Davie et al., 2007). Similar to Myog, Mrf4 is essential for later stages of differentiation (Rawls et al., 1998).

Following differentiation, the myocytes will undergo a cell-fusion process which will drastically change the shape and structure of the cells. The myocytes will initially adhere to one another, and subsequently merge into one another to form nascent myotubes. The newly formed myotubes will then undergo a maturation process to become fully mature fibers (Dumont et al., 2015b).

1.3.1.3 Asymmetric and symmetric self-renewal

Figure 1.4: Asymmetric vs symmetric division Satellite cells have the capability of symmetric or asymmetric divisions. In asymmetric division (left panel), the satellite stem cell divides in an apicobasal manner to give rise to a daughter committed progenitor that will take part in muscle regeneration, as well as an uncommitted satellite cell to maintain the stem cell population. Alternatively, symmetric division in a planar orientation gives rise to 2 similar daughter satellite cells to either expand the stem cell population (right panel, left) or expand the committed progenitors (right panel, right). (Figure adapted from Dumont et al., 2015b with permission from John Wiley & Sons Inc.)

In addition to giving rise to pMBs required for muscle regeneration, the satellite cells must also maintain skeletal muscle’s long-term regenerative potential for future injuries by retaining the quiescent MuSC pool. Interestingly, even after multiple injuries, the MuSC population remains relatively constant (Shi and Garry, 2006). This is as a direct result of the self-renewal capabilities of the MuSCs, where transplantation of a single MuSC is able to both contribute to the generation of new myofibers and repopulate the stem cell niche (Sacco et al., 2008).

MuSCs are a heterogeneous cell population and efforts are underway to fully understand the various subpopulations. One of the most studied categorization of satellite cells has been performed using Myf5-cre/R26R-YFP transgenic mice in which cells that have at any point during development or adult regeneration expressed Myf5, will be permanently labeled with

YFP. Using this mouse line, researchers identified that ~10% of the satellite cells have never expressed Myf5 (YFP-). Furthermore, the YFP- population had a higher self-renewal capability whereas the YFP+ satellite cells were primed to transition to committed progenitors. Transplantation experiments confirmed these results with the YFP- cells resulting in long-term engraftment in the transplanted muscle and populating the stem cell niche, whereas the YFP+ cells had a higher capacity in creating donor-derived fibers (Kuang et al., 2007).

Upon activation, the YFP+ committed satellite cells are able to proliferate, differentiate, and give rise to new myofibers. The YFP- subpopulation on the other hand, can either expand the stem cell population through symmetric division, or divide asymmetrically and create a daughter stem cell and a committed progenitor (Figure 1.4). Although the exact mechanisms behind cell fate are not clear, studies throughout the past decade have begun to unravel some of the factors regulating these decisions.

Two major contributors to asymmetric cell divisions have been identified: 1) polarization of proteins and cell fate determinants; 2) mitotic spindle orientation (Dumont et al., 2015a). During asymmetric divisions, various proteins have been demonstrated to preferentially localize in either the YFP- mother or the YFP+ daughter cell (Dumont et al., 2015b). The partitioning defective (PAR) complex is expressed preferentially in the daughter cell, which leads to the asymmetric activation of the p38 MAPK pathway. Activation of p38 MAPK subsequently leads to MyoD expression in the daughter cell and myogenic progression (Troy et al., 2012). Recent studies have identified new regulators of the PAR complex in MuSCs. Dystrophin has been demonstrated to regulate Microtubule affinity regulating kinase 2 (Mark2) expression, which is required for successful redistribution of the PAR complex in the dividing cells. In the absence of functional dystrophin, satellite cells exhibit impaired asymmetric division (Dumont et al., 2015c). A recently published study has identified the Epidermal growth factor (Egf) and Aurora kinase A (AurkA) pathways as another key regulator of asymmetric cell divisions. The researchers demonstrated that Egf stimulation activated Egf receptor (Egfr) at the basal surface of MuSCs which then induces asymmetric expansion through AurkA (Wang et al., 2019). Other studies have demonstrated the role of Notch signaling in asymmetric cell division. Numb, a notch signaling inhibitor, is asymmetrically expressed in the committed daughter cell (Conboy and Rando, 2002). Although two recent studies demonstrated that Numb does not impact on 15

Notch signaling in muscle (George et al., 2013; Le Roux et al., 2015), a separate study confirms the role of Notch signaling by demonstrating the asymmetric expression of Dll1 and its receptor Notch3, in the committed daughter (YFP+) and mother stem (YFP-) cells, respectively (Kuang et al., 2007). Notch signaling plays an essential role in MuSC quiescence (reviewed in 1.2.1.1), and its activation in the YFP- mother stem cell is important in maintaining its stemness.

The main pathway involved in symmetric self-renewal of the YFP- satellite cells is the planar cell polarity (PCP) pathway. The PCP is a non-canonical pathway activated by Wingless (Wnt) ligands. Wnt proteins typically bind to the Frizzled (Fzd) receptors of target cells (Clevers and Nusse, 2012; Sethi and Vidal-Puig, 2010). The intracellular domain of the Fzd receptor then binds Disheveled (Dsh) leading to the activation of Rac/JNK and Rho/ROCK pathways (Katoh and Katoh, 2007). Studies demonstrate Wnt7a (upregulated during early stages of regeneration stages) binds to Fzd7 (enriched in the YFPneg satellite cells) leading to increased expression of the Wnt/PCP downstream effector Vangl2 (Le Grand et al., 2009). Wnt7a stimulation of MuSCs resulted in a higher incidence of symmetric self-renewal as well as a polarized distribution of Vangl2 in the daughter cells.

Physical interaction with the ECM also appears to play a role in MuSC fate. The majority of the symmetric and asymmetric divisions are planar (parallel to myofiber) and apical-basal (perpendicular to myofiber), respectively (Figure 1.4). During the asymmetric apical-basal divisions, the cell that maintains contact with the basal lamina remains YFP-, whereas the cell that is in contact with the myofiber becomes a YFP+ committed progenitor. Based on in vitro studies, it appears as though symmetrical adhesion to the ECM preferentially leads to random DNA segregation in dividing cells, and non-symmetrical ECM adhesion resulted preferentially in non-random DNA segregation (Yennek et al., 2014). On the other hand, both daughter cells are equally exposed to the ECM during symmetric expansion. Researchers have demonstrated that Sdc4 is a co-receptor with Fzd7, and that Wnt7a signaling through Fzd7 is dependent on fibronectin ligation to Sdc4 (Bentzinger et al., 2013b). Collagen VI (ColVI) is another ECM protein that has been demonstrated to play a role in MuSC self-renewal, where ColVI depletion results in the exhaustion and depletion of the MuSC pool (Urciuolo et al., 2013). Furthermore, ColV secreted by MuSCs is an important component of the quiescent MuSC niche. Conditional

deletion of ColV results in depletion of the MuSC pool and abnormal cell cycle entry of the MuSCs (Baghdadi et al., 2018b).

1.3.2 Immune response

The precisely orchestrated and rapid immune response following injury is critical for successful skeletal muscle regeneration (Tidball, 2011). In homeostatic skeletal muscle, resident leukocytes (predominantly mast cells and macrophages) act as sensors for damage. Upon skeletal muscle injury, these cells immediately release chemoattractive molecules such as Tnfα and Il-6, to initiate the inflammatory response (Auffray et al., 2007; Galli et al., 2011). In response to cellular and extracellular contents secreted by damaged tissues as well as the cytokines and chemokines released by the resident leukocytes, granulocytes (mainly neutrophils) accumulate in the injured area (Heredia et al., 2013; Wang and Thorlacius, 2005). Neutrophils can be detected in injured tissue as early as 1-6 hour post-injury (Fielding et al., 1993). The accumulated neutrophils promote the inflammatory environment, and recruit monocytes to the sites of injury through the secretion of chemokines (Kasama et al., 1993; Scapini et al.).

Once monocytes arrive to the location of injury, they differentiate into macrophages. For the purpose of this review, we will be using the simplified terminology of M1 and M2 macrophages. However, various studies have shown that a continuum exists between these 2 states and they cannot be fully represented by distinct phenotypes (Novak and Koh, 2013). The initial wave of macrophages arriving to skeletal muscle are the pro-inflammatory M1 macrophages that peak at 24 hours post-injury and rapidly decline in the following days. The second wave of macrophages are the anti-inflammatory M2 macrophages that peak at 2-4 days post-injury and remain in injured muscle until the end of the inflammation phase (Figure 1.2).

Upon their arrival, M1 macrophages begin clearing cellular debris through phagocytosis. They also secrete pro-inflammatory cytokines such as Igf-1, Il-6, Il-1β and Tnfα which stimulates satellite cell proliferation and preventing premature differentiation (Broussard et al., 2014; Langen et al., 2004; Saclier et al., 2013a; Tidball and Villalta, 2010; Tonkin et al., 2015; Wang et al., 2008). Additionally, Tnfα secretion by these macrophages plays a crucial role in regulating FAP apoptosis and preventing detrimental fibrosis (Lemos et al., 2015) through Tnfα secretion. The arrival of the M2 macrophages marks the beginning of the regenerative phase of skeletal 17

muscle. M2 macrophages initially secrete high levels of Igf-1 to promote satellite cell proliferation (Tonkin et al., 2015). Subsequently, they promote myogenic differentiation through secretion of growth differentiation factor 3 (Gdf3) and low levels of Tnfα and Tgfβ (Juban and Chazaud, 2017; Saclier et al., 2013b; Varga et al., 2016).

Studies have reported eosinophils to also be early responders of the immune system. Interestingly, they appear to play a role in promoting proliferation and preventing adipogenesis in FAPs (Heredia et al., 2013). This cellular interaction indirectly promotes satellite cell expansion by increasing the pro-regenerative properties of FAPs (discussed in 1.3.3).

Any disruption in the immune response impairs skeletal muscle regeneration. Inhibiting the initial reaction to injury by the mast cells has been shown to be prevent monocyte and macrophage accumulation in skeletal muscle (Dumont et al., 2007). Moreover, disrupting the balance in M1 and M2 macrophages also hinders muscle regeneration. Overexpression of M2 macrophages in the regenerative phase of muscle repair results in fibrosis (Wang et al., 2015). Macrophage ablation studies have also revealed increased fibrosis and decreased muscle regeneration following injuries (Lemos et al., 2015; Liu et al., 2017; Shen et al., 2008; Xiao et al., 2016). As such, the temporally regulated immune response is an indispensable step in successful and timely skeletal muscle regeneration.

1.3.3 Fibro-adipogenic progenitors

Similar to MuSCs, tissue insult activates quiescent FAPs. Their proliferation timing mimics those of MuSCs, peaking in numbers at around 3-4 days post-injury while receding back to normal levels by 7-9 days post-injury (Joe et al., 2010; Lemos et al., 2015; Uezumi et al., 2010). FAPs are major contributors to ECM remodeling and deposition of extracellular matrix proteins during regeneration. Interestingly, reciprocal signaling appears to exist between FAPs and myogenic cells. FAPs promote myogenic differentiation of satellite cells through Il-6 secretion (Joe et al., 2010), and myotubes prevent the adipogenic differentiation of FAPs (Uezumi et al., 2010).

Various studies have demonstrated that FAPs are tightly regulated by the immune response. In the earlier stages of skeletal muscle regeneration, eosinophils promote FAP proliferation and

prevent adipogenic differentiation through secretion of Il-4 and Il-13 (Heredia et al., 2013). Following the rapid proliferation of FAPs around day 3-4 post-injury, Tnfα secretion by the inflammatory macrophages leads to FAP apoptosis. Following the switch from M1 to M2, anti- inflammatory macrophages promote ECM deposition by the FAPs by releasing Tgfβ (Lemos et al., 2015). Disturbances in the timely transition from M1 to M2, or prolonged Tgfβ signaling due to overexpression of M2 macrophages has been demonstrated to result in pathological fibrosis (Lemos et al., 2015). And finally, FAPs appear to have an overlapping function with macrophages and are capable at removing cellular debris from the injured sites as well (Heredia et al., 2013).

Further loss-of-function studies are required to fully decipher the cause-and-effect relationships observed between FAPs and the phenomena observed in skeletal muscle. However, due to their molecular heterogeneity and lack of unique molecular identifiers, studying the exact role and necessity of FAPs in skeletal muscle regeneration has proven difficult.

1.3.4 Endothelial cells

Following the inflammatory and at the onset of the regenerative phase of skeletal muscle regeneration, the number of ECs increases dramatically, reaching peak numbers ~7 days post- injury (Figure 1.2). These ECs form the nascent capillaries that are necessary to rebuild the intricate vascular system required for a functional skeletal muscle (Hershey et al., 2001; Scholz et al., 2003). In most types of injury, the vascular network is partially restored by 12 days and returns to normal preinjury conditions by 1 month post-injury (Hardy et al., 2016).

ECs have been demonstrated to influence satellite cell proliferation and differentiation, and at the same time, satellite cells and differentiating myoblasts appear to be proangiogenic. In vitro studies reveal that ECs promote pMB proliferation through the secretion of various growth factors such as Igf-1, Hgf, Fgf, and Vegfa (Christov et al., 2007). Alternatively, differentiating myoblasts as well as MuSCs express Vegfa which has a proangiogenic effect on ECs (Christov et al., 2007; Verma et al., 2018). More recent in vitro studies have demonstrated that ECs promote pMB proliferation through Apelin and Oncostatin M secretion (Latroche et al., 2017). Additionally, ECs in regenerating muscle express Angpt1 for which proliferating pMBs express

the receptor Tie2 (Abou-Khalil et al., 2009). Angpt1-Tie2 signaling is involved in MuSC quiescence, suggesting a role for ECs in regenerating muscle in returning MuSC to quiescence.

Taken together, various studies suggest reciprocal signaling to exist between ECs and satellite cells, implying a functional role for the proximity of MuSCs and microvasculature. Most of the research attempting to decipher the cross-talk between ECs and satellite cells are in vitro studies and further in vivo studies are required to unveil these interactions. However, given that vasculature is present throughout all tissues, distinguish local and global effects following manipulation of the endothelial system has proven to be a challenge.

1.4 Skeletal muscle pathology and degeneration

Healthy skeletal muscle has the remarkable ability to regenerate in response to injuries. The most common muscle injuries are tears and lacerations as a result of physical trauma without significant loss of muscle tissue, such as exercise (Beiner and Jokl, 2001; Garrett, 1996; Järvinen et al., 2000, 2005; Lehto and Järvinen, 1991). However, there are certain situations where the repair capacity of muscle is diminished. Inherited genetic mutations resulting in muscular dystrophy, aging, and muscle wasting associated with chronic diseases such as cancer (cachexia), are examples of conditions where skeletal muscle is unable to successfully regenerate and begins to deteriorate. Additionally, in traumatic injuries where more than 20% of the skeletal muscle is lost (volumetric muscle loss), the natural repair process will fail to repair the missing tissue and will result in scar tissue accumulation and loss of function (Turner and Badylak, 2012).

In this section, we will provide an overview on the factors affecting skeletal muscle regeneration in two of the most frequent causes of skeletal muscle degeneration: aging and muscular dystrophies.

1.4.1 Aging

Aged skeletal muscle exhibits reduced capacity for regeneration. Regeneration in aging muscle often results in more fibrosis, increased fat deposition, and smaller myofiber size (Grounds, 1998; Sadeh, 1988). Additionally, aged skeletal muscle exhibits impaired angiogenesis and innervation (Carlson et al., 2001) and is accompanied with a decrease in the number of satellite cells (Shefer et al., 2006). Although aging is not considered a pathological state, the changes in 20

the extrinsic and intrinsic regulators of MuSCs has been directly linked to the reduced regenerative capacity and function of aged skeletal muscle.

The first evidence of extrinsic factors being a factor in regenerative decline of aged muscle came from parabiotic mice. Engrafting aged muscle in young rats lead to an increase in muscle mass and contractile strength, whereas engrafting young muscle into aged rats provided opposite results (Carlson and Faulkner, 1989). Furthermore, conjoining the circulatory system of young and aged mice leads to increased regenerative capacity of the aged and impaired regeneration in the young skeletal muscle (Conboy et al., 2005).

In healthy skeletal muscle regeneration, MuSCs are exposed to a series of temporally regulated factors from the various cells in the immediate niche and the circulatory system. This carefully orchestrated dynamic response to regeneration is perturbed in aged muscle which manifests itself in loss of MuSC quiescence and impaired MuSC function during regeneration. Aged muscle exhibits decreased levels of Notch ligand Dll1 on the myofiber plasma, which is critical for MuSC self-renewal and quiescence (Bjornson et al., 2012; Conboy et al., 2003; Philippos et al., 2012). Additionally, canonical Wnt signaling is upregulated in aged muscle, antagonizing Notch signaling while at the same time promoting fibrogenic differentiation of satellite cells (Brack et al., 2007, 2008). Furthermore, oxytocin, which has been demonstrated to promote muscle regeneration in young mice, is downregulated in aged muscle (Elabd et al., 2014). All these extrinsic factors lead to impaired MuSC quiescence, self-renewal, and MuSC-mediated myogenesis. Interestingly, many of these adverse effects in aged muscle can be ameliorated through parabiosis and exposure to young serum, or more directly via pharmacological correction of the root cause (Brack et al., 2007, 2008; Conboy et al., 2003; Elabd et al., 2014). Changes in the ECM surrounding the MuSCs also plays a role in satellite cell and muscle regeneration dysfunction. Fibronectin (Fn1) has been demonstrated to be downregulated in the aged niche. In addition to being the preferred adhesion substrate for MuSCs, Fn1 plays an important role in the symmetric expansion of satellite cells (Bentzinger et al., 2013b; Lukjanenko et al., 2016). Loss of Fn1 in the aged niche leads the activation of aging pathways such as ERK and p38 MAPK. Interestingly, aged MuSCs can be rejuvenated upon treatment with Fn1 (Lukjanenko et al., 2016). Recent studies have also demonstrated that changes in the interaction of FAPs and satellite cells in aged muscle also affects successful satellite cell function and 21

skeletal muscle regeneration. Wisp1 is a FAP-derived matricellular signal that is involved in the expansion and asymmetric commitment of MuSCs. In aged skeletal muscle, Wisp1 signaling by the FAPs is lost. However, restoring Wisp1 signaling in regenerating skeletal muscle, either through systemic administration of transplantation of young FAPs, rescues skeletal muscle regeneration (Lukjanenko et al., 2019).

MuSCs in aged skeletal muscle also exhibit intrinsic alterations as well. Transplantation of freshly isolated aged and young MuSCs into young regenerating muscle, revealed an intrinsic loss of reparative and self-renewal capacity in aged MuSCs. Activation of p38α/β MAPK, Jak2/Stat3, and Cdkn2a cell cycle inhibitory pathways are a few examples of signaling axes present in aged skeletal muscle (Bernet et al., 2014; Cosgrove et al., 2014; Price et al., 2014; Sousa-Victor et al., 2014; Tierney et al., 2014). Aged muscle fibers were shown to express higher levels of Fgf2, leading to the increased p38α/β MAPK signaling. Increased p38 MAPK signaling can negatively regulate MuSC self-renewal by promoting myogenic differentiation and restricting cell cycle progression (Bernet et al., 2014; Chakkalakal et al., 2012; Palacios et al., 2010). Similar to p38 MAPK signaling, increased Jak2/Stat3 signaling leads to increased asymmetric division of aged MuSCs and promotes myogenic differentiation through MyoD expression. This increase in asymmetric division of satellite cells has been linked to the exhaustion of the MuSC pool (Price et al., 2014). Furthermore, aged SCs also exhibit defects in Itgb1 activity (Rozo et al., 2016). And finally, a subset of aged MuSCs enter premature senescence by activating Cdkn2a (Sousa-Victor et al., 2014).

In summary, proper satellite cell functions and successful skeletal muscle regeneration rely on an intricately balanced of intrinsic and extrinsic factors that is disrupted during aging.

1.4.2 Muscular dystrophies

Muscle dystrophies are genetically inherited myogenic disorders that result in progressive muscle wasting and weakness (Emery, 2002). Over 30 types of muscular dystrophies have been identified with various degrees of severity. A hallmark of muscular dystrophies is the constant state of regeneration in muscle, where the recurring breakdown of myofibers elicits a constant need for regeneration (Sacco et al., 2010). Some dystrophies are relatively mild with normal life

expectancy and slow progression, whereas others result in severe systematic loss of skeletal muscle function, disability, and respiratory/cardiac failure (Emery, 2002).

In the following, we will provide a brief overview over two types of muscular dystrophy.

1.4.2.1 Duchenne muscular dystrophy

The most common and severe form of muscular dystrophy is Duchenne muscular dystrophy (DMD), caused by mutations in the dystrophin gene. The onset of DMD is early childhood patients and results in muscle weakness and wasting(Hoffman et al., 1987). Dystrophin is a major component of the dystrophin-associated protein complex (DAPC) and connects the cytoskeleton the myofiber to the ECM through the myofiber membrane (Allikian and McNally, 2007). Due to mutations in the dystrophin gene in DMD patients, the myofiber membrane is extremely fragile and is injured as a result of mild stress. Normal skeletal muscle is quiescent under normal circumstances. However, the constant recurring damage in the muscles of DMD patients and mdx mice (mouse model for DMD) results in a continuous cycles of degeneration and regeneration (Serrano et al., 2011).

The dynamics of muscle regeneration in DMD muscle do not mimic the cellular and molecular dynamics of healthy skeletal muscle. The constant degeneration of myofibers promotes an inflammatory phenotype in the muscle. Inflammatory cells are deregulated and encourage the accumulation of M1 macrophages (Villalta et al., 2009). Furthermore, injured myofibers or perhaps the absence of mature fibers, promote adipogenesis and fibrotic deposition in the FAPs (Joe et al., 2010; Uezumi et al., 2010). Some of these symptoms can be temporarily ameliorated through anti-inflammatory corticosteroids (Matthews et al., 2016) and Tgfβ inhibitors to decrease myofiber fibrosis (Cohn et al., 2007; Taniguti et al., 2011).

The chronically degenerative environment in DMD has negative effects on the satellite cells and their function. Human DMD myoblasts have limited in vitro proliferative capacity which is aggravated by aging (Webster and Blau, 1990). Additionally, various studies have suggested that the shorter telomere length in dystrophic satellite cells could be responsible for the decreased regenerative capacity of DMD satellite cells (Decary et al., 2000; Sacco et al., 2010). On the other hand, studies on human DMD patient biopsies and mouse models of DMD have

characterized higher numbers of satellite cells present in DMD muscle (Chakkalakal et al., 2014; Kottlors and Kirschner, 2010; Reimann et al., 2000). Further studies are required to capture the effect of DMD on satellite cells in vivo.

Initially, DMD was thought to be a dystrophic condition indirectly affecting satellite cells through the chronic degeneration/regeneration cycles and the inflammatory milieu of skeletal muscle. However, recent studies have highlighted that mutations in the dystrophin gene directly affects satellite cell fate irrespective of the satellite cell environment as well. During asymmetric satellite cell division (Figure 1.4, reviewed in 1.3.1.3), dystrophin regulates cell polarity. The absence of functional dystrophin in activated satellite cells results in decreased asymmetric cell division, loss of polarity in the dividing cells, impaired mitotic spindle orientation, and prolonged cell divisions. These defects ultimately result in reduced numbers of necessary myogenic progenitors required for muscle regeneration (Dumont et al., 2015c).

1.4.2.2 Oculopharyngeal muscular dystrophy

Oculopharyngeal muscular dystrophy (OPMD) is a late onset muscular dystrophy manifesting itself in the 5th or 6th decade of life and most frequently observed in French Canadians. OPMD is caused by a mutation resulting in a GCG trinucleotide repeat in the coding region of the Pabpn1 gene. OPMD patients exhibit weakness and wasting in head and neck muscles, with ptosis (drooping of the upper eyelid) and dysphagia (difficulty swallowing) being some of the first symptoms (Morgan and Zammit, 2010).

Pabpn1 is a nuclear protein involved in pre-mRNA polyadenylation, mRNA nucleocytoplasmic transport, and transcription regulation (Abu-Baker and Rouleau, 2007). The exact mechanisms behind OPMD muscle wasting are unclear. However, several studies have demonstrated intrinsic defects in myoblasts with perturbed Pabpn1 function. Pabpn1 is expressed in activated satellite cells and myonuclei (Morgan and Zammit, 2010). siRNA-mediated knockdown of Pabpn1 in murine myoblasts resulted in reduced in vitro proliferation and differentiation, in addition to defects in mRNA processing (Apponi et al., 2010). Moreover, pMBs derived from the pharyngeal muscles of OPMD patients exhibit reduced proliferative and myogenic capacity (Périé et al., 2006).

1.5 Skeletal muscle treatments

As the fundamental building blocks of skeletal muscle regeneration many of the symptoms of muscle degenerative diseases or conditions can be traced back to MuSCs and the intrinsic and extrinsic factors regulating MuSC fate and function (Morgan and Zammit, 2010). Therefore, development of skeletal muscle regenerative therapies has mainly focused on two avenues: 1) Cell based therapies to deliver healthy myogenic cells to diseased muscle; 2) Improve and restore the endogenous myogenic potential of satellite cells (Dumont et al., 2015b).

1.5.1 Cell based therapies

Intrinsic defects in skeletal muscle function is a contributor to many skeletal muscle ailments. Gene mutations, faulty cell cycle kinetics, satellite cell exhaustion and senescence, and overall reduced myogenic capacity of MuSCs have been demonstrated to lead to the reduced regenerative capacity of muscle in various muscular dystrophies and degenerative conditions such as aging. Myofibers are multi-nucleated cells that are formed from the fusion of multiple myogenic progenitors, in which the gene products of the nuclei are distributed throughout the fiber. It has been suggested that in the case of genetic mutations, the addition of a small fraction of healthy nuclei into the myofiber can produce sufficient functional copies of the mutated gene to restore proper function (Judson et al., 2018; Watt et al., 1982). Therefore, delivery of healthy myogenic cells has long been envisioned as a potential solution to repair host myofibers and restore the functional satellite cell pool, especially in the case of muscular dystrophies (MDs).

Early studies of myogenic cell transplantation have demonstrated the potential of cell-based therapies. For example, autologous myoblast transplantation into the pharyngeal muscles of patients with oculopharyngeal muscular dystrophy resulted in an improvement in quality of life, as well as a dose-dependent improvement in swallowing (Perie et al., 2014). Furthermore, researchers have long been able to restore functional dystrophin in the muscles of DMD patients or mdx mice through transplantation of healthy primary myoblasts (Mendell et al., 1995; Partridge et al., 1989; Skuk et al., 2004; Tremblay et al., 1993). However, various studies and clinical trials have reported challenges associated with myogenic cell transplantation including but not limited to: poor cell survival, rapid injection site clearance, poor donor cell dispersion, and limited contributions to short- and long-term tissue repair (Skuk, 2004).

1.5.1.1 Clinically relevant numbers of cells with high therapeutic potential

In order to support the long-term regeneration of the donor muscle, the transplanted cells used in cell-based therapies must be able to contribute to MuSC pool repopulation, in addition to the ability to engraft and form new fibers in the donor muscle.

The engraftment potential of in vitro cultured pMBs is relatively low. This is partially due to low pMB survival upon transplantation and cannot be resolved by simply increasing the number of transplanted pMBs(Kang et al., 2015). In addition to low engraftment potential, the surviving myoblasts rapidly differentiate and form myofibers, and are not able to relocate to the stem cell niche for the long-term regenerative requirements of muscle (Huard et al., 1992). In comparison to pMBs, freshly isolated MuSCs have a much higher therapeutic capacity; transplanted MuSCs can both contribute to new fiber formation, as well as efficiently repopulating the stem cell compartment (Cerletti et al., 2008; Cezar and Mooney, 2015; Collins et al., 2005; Crist et al., 2012; Kuang et al., 2007; Sacco et al., 2008).

However, obtaining clinically relevant numbers of MuSCs to treat multiple muscles has proven to be a challenge. In addition to MuSCs being a rare stem cell population, cultured MuSCs rapidly lose their regenerative capacity and differentiate into pMBs. Recent efforts have been underway to maintain the therapeutic potential of ex-vivo cultured MuSCs. Mimicking the in- vivo niche is one method to improve the transplantation efficiency of ex-vivo cultured MuSCs. Recreating an artificial niche (Quarta et al., 2016) or inhibiting protein activity using small molecule inhibitors (Zismanov et al., 2016) have been demonstrated to promote MuSC quiescence in vitro. Small molecule inhibitors have also been used to rejuvenate ‘aged’ MuSCs by inhibiting p38 MAPK and Stat3 activity (Bernet et al., 2014; Tierney et al., 2014). Culturing MuSCs on hydrogels with the same elastic modulus as skeletal muscle (12 kPa) can maintain the regenerative capacity of the expanded MuSCs, and even rejuvenate ‘aged’ MuSCs in combination with pharmacological inhibition of p38 MAPK activity (Cosgrove et al., 2014; Gilbert et al., 2010). Furthermore, culturing MuSCs in the presence of ECM components found in the MuSC niche such as Fn1 (Bentzinger et al., 2013b), ColVI (Urciuolo et al., 2013), and laminins (Ishii et al., 2018) can also maintain the regenerative capacity of the MuSCs. While these studies increase the regenerative capacity of cultured cells compared to their respective

controls, they do not address the challenge in producing sufficient cell numbers for cell-based therapies. More recent studies have attempted to address this issue by identifying novel targets for ex-vivo expansion of cultured MuSCs. Pharmacological inhibition of Setd7 (Judson et al., 2018), Jak2/Stat3 signaling (Price et al., 2014), dephosphorylation of eIF2α (Lean et al., 2019), and p38 MAPK (Charville et al., 2015) in cultured MuSCs have been demonstrated to result in increased numbers of cells with therapeutic ability. Although these advances are encouraging, identification of druggable targets for ex-vivo expansion of MuSCs while retaining their therapeutic potential remains a major focus of skeletal muscle research.

1.5.1.2 Co-transplantation with hydrogels, cells, and growth factors

Another approach to improve cell-based therapies is the co-delivery of myogenic cells with a combination of supporting cells, growth factors, or biomaterials. As an example, short ex-vivo treatment of pMBs with Wnt7a increased both dispersion and engraftment of the transplanted cells (Bentzinger et al., 2014). Furthermore, transplantation of myogenic progenitors alongside pro-inflammatory macrophages (Bencze et al., 2012; Lesault et al., 2012), or growth factors such as Vegfa (Bouchentouf et al., 2008), Fgf2, and Igf-1 (Lafreniere et al., 2009) has been proven to improve the transplantation outcomes through increased dispersion and engraftment.

Synthetic and natural biomaterials are increasingly used in tissue engineering and regenerative medicine (Cezar and Mooney, 2015; Lutolf and Hubbell, 2005; Sionkowska, 2011). Biomaterials are often used to take advantage of their various biochemical and biophysical properties during transplantation. For example, the shear thinning properties of hydrogels can have a protective effect on the cells and increase cell viability throughout the injection process (Aguado et al., 2012). Studies on skeletal muscle regeneration and myogenic cell transplantation have used a wide array of these biomaterials to promote migration, survival, and contribution of the transplanted cells (Beier et al., 2006; Boldrin et al., 2007a; Borselli et al., 2011; Cezar and Mooney, 2015; Cima et al., 1991; Han et al., 2018; Hill et al., 2006; Rossi et al., 2011; Serena et al., 2008). As an example, transplantation of pMBs embedded in a macroporous scaffold designed to release of growth factors (Vegfa, Igf-1) promoted pMB engraftment, reduced fibrosis, and improved the regeneration process in an ischemic muscle injured with myotoxins (Borselli et al., 2011).

Despite the many advantages of these biomaterials, a common issue is the arduous surgical procedure, often involving in situ polymerization of a hydrogel or fixing a previously seeded scaffold in place. For clinical applications, an injectable hydrogel to be used as a carrier for the transplanted cells is clearly more suitable, due to benefits such as reduced surgery time, and a minimally invasive transplantation procedure (Aguado et al., 2012). In fact, researchers in a recent study developed an injectable delivery system using peptide amphiphiles, to co-deliver MuSCs and growth factors into injured skeletal muscle. The MuSCs transplanted using this delivery system demonstrated enhanced engraftment potential compared to MuSCs delivered in saline(Sleep et al., 2017). Despite these advances, development of new biomaterials for myogenic cell delivery remains an important goal in devising new cell-based therapies.

1.5.2 Enabling endogenous repair

Despite the proven advantages and effectivity of cell-based therapies, there are fundamental issues that limit the application of these treatments. Skeletal muscle is not localized to a specific location and is distributed throughout the body. As such, in the case of chronic conditions such as aging or muscular dystrophies, intramuscular injections of cells throughout a patient’s body is not an optimal solution. Therefore, the development of systemic therapies such as gene editing techniques and treatments to enable the endogenous repair of pathological skeletal muscle are required for long-term treatment of muscular dystrophies and chronic conditions.

1.5.2.1 Gene therapies

The advent of adeno-associated vectors (AAVs) has enabled the development of gene therapies to reinstate functional dystrophin expression in muscle. AAV-mediated micro-dystrophin delivery in murine and canine DMD models, has led to the expression of truncated but functional dystrophin protein in the myofibers. Clinical trials are underway to test the efficacy of micro- dystrophin delivery in humans (Chamberlain and Chamberlain, 2017). An alternative approach is to use AAVs to systematically deliver CRISPR-Cas9 to correct mutations in dystrophin (Mendell and Rodino-Klapac, 2016). Recent studies demonstrated functional recovery (increased grip strength, improved force generation, reduced serum creatine kinase) in mdx mice treated with CRISPR-Cas9 (Long et al., 2014; Madhavan et al., 2016; Tabebordbar et al., 2016). More importantly, one of the studies was able to demonstrate efficient editing of Pax7+ MuSCs, and

establishing the potential of this treatment in long-term regeneration of DMD patients (Tabebordbar et al., 2016). Despite being their infancy, recent success in gene editing technologies have highlighted the potential of these therapies for treating genetic disorders.

1.5.2.2 Boosting the endogenous repair capacity through systemic treatments

In chronic conditions such as muscular dystrophies or aging, intrinsic and extrinsic changes in factors regulating MuSC function lead to impaired skeletal muscle regeneration. Altering these factors to boost the endogenous regenerative capacity of skeletal muscle is a promising approach to improve muscle repair. Although these approaches will not correct any genetic deficits, evidence from pre-clinical trials and animal models suggest these interventions can be used to ameliorate some of the symptoms of chronic myopathies. Despite the advancements in this area in recent years, further research is required before a clinically relevant therapy can be devised.

Intramuscular injection of Wnt7a, previously revealed to promote satellite cell expansion through the planar cell polarity pathway (Le Grand et al., 2009), has been demonstrated to be a promising candidate to ameliorate DMD symptoms. Administering Wnt7a to mdx mice muscles led to satellite cell expansion, myofiber hypertrophy, increased muscle strength, as well as reduced contraction induced damage (von Maltzahn et al., 2012). Furthermore, in vivo administration of Granulocyte-colony stimulating factor (Gcsf) improves muscle regeneration in mdx mice by promoting satellite cell expansion (Hayashiji et al., 2015). And finally, pharmacological inhibition of Il6-r (upstream of Stat3 signaling) and has been demonstrated to improve the dystrophic phenotype in mdx mice (Wada et al., 2017).

Scientists have demonstrated that altering systemic factors through parabiotic pairing of young and aged mice, can modulate the age-related decline in myogenic progenitors (Conboy et al., 2005). Aged MuSCs’ impaired regenerative capacity has been linked to elevated levels of Jak/Stat signaling, and direct injection of Jak/Stat inhibitors into the muscles of aged mice improved muscle regeneration, increased satellite numbers, and functionally improved the muscle (Price et al., 2014). In a separate study, researchers determined that reduced p53 activity due to a deficiency in Notch activators in the aged microenvironment, resulted in aged MuSCs’ impaired proliferative expansion. In vivo pharmacological activation of p53 was able to restore

youthful function of aged MuSCs (Liu et al., 2018). And finally, although controversy exists surrounding the upregulation or downregulation of Gdf11 in aged mice, altering this systemic protein has been demonstrated to improve skeletal muscle regeneration in aged animals (Egerman et al., 2015; Sinha et al., 2014).

1.6 Thesis Motivation, aims, and approach 1.6.1 Thesis motivation and aims

Skeletal muscle wasting is a common occurrence in a wide range of conditions including but not limited to cancer (cachexia), HIV infection, muscular dystrophy, aging (sarcopenia), and physical inactivity (Cosgrove et al., 2009). Additionally, the most recent nationwide cost-of- illness study conducted by the Public Health Agency of Canada, estimates the direct and undirect costs of musculoskeletal diseases to be ~$6.7 billion (Public Health Agency of Canada, 2010). Skeletal muscle research aims to relieve the health and economic burdens imposed by skeletal muscle disorders.

The robust regenerative potential of healthy skeletal muscle is attributed to the presence of resident muscle stem cells (MuSCs), as well as the coordinated response of a variety of cells within the skeletal muscle niche (Wosczyna and Rando, 2018). Disruption in any step of the highly orchestrated regeneration process leads to incomplete or delayed regeneration of skeletal muscle. Congenital defects in genes required for proper skeletal muscle function and regeneration (Emery, 2002), and changes in the skeletal muscle niche during aging (Chakkalakal et al., 2012; Gopinath and Rando, 2008), are examples of situations where this delicate balance is disrupted.

Treatments to restore skeletal muscle strength and function to aged or diseased skeletal muscle can be categorized into three main categories: a) cell-based therapy, b) systemic treatments to support endogenous repair, and c) replacement of skeletal muscle tissue. Addressing these issues is at the core of most skeletal muscle research. The goal of this thesis is to contribute to efforts in the skeletal muscle cell-based therapy and endogenous repair fields through the following aims:

1) To improve MuSC transplantation outcomes by implementing a bioactive hydrogel as the cell delivery vehicle. 30

2) To identify signaling pathways whose inhibition improves the regenerative capacity of cultured MuSC populations.

3) To identify new endogenous repair therapeutic entry-points by uncovering the diversity and intercommunication of cells present in the adult skeletal muscle niche.

The first aim attempts to address common hurdles associated with myogenic cell transplantations. Despite clinical evidence of successful restoration of muscle strength and function through transplantation therapies (Perie et al., 2014; Tremblay et al., 1993), low cell survival, rapid injection site clearance, poor donor cell dispersion, and limited contributions to tissue repair are issues that need to be resolved for cell-based therapies to achieve their full potential. Here we attempt to ameliorate some of these issues by utilizing HAMC (a physical blend of hyaluronan (HA) and methylcellulose (MC)) as a MuSC cell delivery vehicle in transplantations.

In the second aim, we address another issue plaguing the development of cell-based therapies in skeletal muscle: the inability to produce clinically-relevant numbers of cells possessing a high regenerative capacity. Given that MuSCs are a relatively rare cell population, efforts are focused on producing sufficient cell numbers for successful cell-therapies. However, MuSCs rapidly lose their regenerative potential during ex-vivo expansion by differentiating into primary myoblasts (pMBs). This reduces the efficiency of cell transplantation results since pMBs have significantly lower regenerative capacity compared to MuSCs, an issue that is difficult to resolve, even by transplanting a higher number of primary myoblasts (Ikemoto et al., 2007; Kang et al., 2015; Montarras et al., 2005a; Sacco et al., 2008). In this aim, we work to identify novel druggable targets that, when manipulated, serve to increase the regenerative capacity of the cultured MuSC population.

In the third aim, we implement single cell RNA sequencing (sc-RNA seq) analysis of skeletal muscle, to shed light on the cellular diversity and intercommunication of cells within skeletal muscle. Skeletal muscle homeostasis and regeneration is a delicate balance of intrinsic and extrinsic signaling of cells present in the skeletal muscle niche. Understanding the mechanisms behind MuSC regulation in vivo is crucial in identifying treatments and therapies for endogenous skeletal muscle repair. By deriving this dataset, we attempt to compile a resource through which 31

novel regulators of MuSC and other factors skeletal muscle quiescence can be identified, and to be applied towards developing therapies for endogenous muscle repair.

1.6.2 Thesis Overview

In the first chapter we provide background information on skeletal muscle and muscle regeneration, as well as the various cellular players in homeostatic and regenerating skeletal muscle homeostasis. We then review conditions that can lead to skeletal muscle wasting, such as aging and muscular dystrophies. Finally, we provide an overview of the efforts underway to develop skeletal muscle therapies and treatments.

In the second chapter, we demonstrate that MuSC transplantation outcomes can be improved by using HAMC hydrogel as the cell delivery vehicle. More specifically, incorporating MuSCs within HAMC prior to in vivo transplantation assays, increased both the number and dispersion of donor derived fibers compared to saline control. Further analysis revealed that this increase in donor-derived fibers can be attributed to three main factors. We demonstrate that HAMC leads to an increase in MuSC proliferation through a CD44-independent mechanism. Additionally, we provide evidence that HAMC protects the transplanted cells from active clearance from the site of injection as well as supporting in vivo expansion by delaying differentiation.

In the third chapter, we identify epidermal growth factor receptor (Egfr) and Kdr as new therapeutic intervention targets, that when inhibited, produce a population of cultured MuSCs with greater regenerative potency than control treated. These targets were identified from a kinase inhibitor drug screen designed to identify small molecule inhibitors that increased the number of cells produced from MuSCs cultured for 7 days. We next used an in vivo intramuscular transplantation assay in mice to determine the regenerative capacity of the cells that were expanded in the presence of a selection of ‘hit’ drugs identified in drug screen. Analyzing the transplantation results revealed that inhibiting EGFR or KDR in ex-vivo cultured MuSCs led to an increase in the total area of recipient muscle containing donor-derived fibers.

In the fourth chapter, we use single cell RNA sequencing (sc-RNA seq) to obtain a holistic view of the cells present in homeostatic adult skeletal muscle, and potential intercellular signaling that exists between the identified cells. More specifically, we use sc-RNA seq to analyze and identify

through an unbiased fashion, the mononuclear cells populations present in adult mouse skeletal muscle. Our analysis revealed 6 different cell types and 9 transcriptionally distinct clusters of skeletal muscle cells. From the transcriptional signature of these cell clusters, we were then able to derive a complex cell-cell communication network, providing a novel resource to decipher the cellular interplay among cells in homeostatic skeletal muscle. Additionally, we identified and validated novel markers for Schwann cell isolation directly from dissociated skeletal muscle. Finally, we revealed the presence of a new subset of endothelial cells through transcriptional analysis.

In the final chapter, we provide a summary of the work covered in the thesis, the impact of the studies, and future directions for the various presented projects.

Chapter 2 Muscle stem cell intramuscular delivery within hyaluronan methylcellulose improves engraftment efficiency and dispersion

A version of this chapter has been published by:

Sadegh Davoudi, Chih-Ying Chin, Michael C. Cooke, Roger Y. Tam, Molly S. Shoichet, and Penney M. Gilbert

In Biomaterials, Volume 173, August 2018, Pages 34-46

Author contributions:

Chih-Ying Chin was involved in the immunohistochemical analysis. Dr. Michael Cooke, Dr. Roger Tam, and Dr. Molly Shoichet contributed to study design and provided hyaluronan and methylcellulose for the study. All other experiments were designed and executed by Sadegh Davoudi. Dr. Penney Gilbert oversaw this work.

Muscle stem cell intramuscular delivery within hyaluronan methylcellulose improves engraftment efficiency and dispersion 2.1 Abstract

Adult skeletal muscle tissue harbors the capacity for self-repair due to the presence of tissue resident muscle stem cells (MuSCs). Advances in the area of prospective MuSC isolation demonstrated the potential of cell transplantation therapy as a regenerative medicine strategy to restore strength and long-term regenerative capacity to aged, injured, or diseased skeletal muscle tissue. However, cell loss during ejection, limits to post-injection proliferation, and poor donor cell dispersion distal to the injection site are amongst hurdles to overcome to maximize MuSC transplant impact. Here, we assess a physical blend of hyaluronan and methylcellulose (HAMC) as a bioactive, shear thinning hydrogel cell delivery system to improve MuSC transplantation efficiency. Using in vivo transplantation studies, we found that the HAMC delivery system results in a >45% increase in the number of donor-derived fibers as compared to saline delivery. We demonstrate that increases in donor-derived fibers when using HAMC are attributed to increased MuSC proliferation via a CD44-independent mechanism, preventing injected cell active clearance, and supporting in vivo expansion by delaying differentiation. Furthermore, we observed a significant improvement in donor fiber dispersion when MuSCs were delivered in HAMC. Our study results suggest that HAMC is a promising muscle stem cell delivery vehicle.

2.2 Introduction

Skeletal muscle is a striated muscle that facilitates voluntary movement, maintains posture, and aids in thermoregulation under the control of the nervous system (Buckingham, 2001; Janssen et al., 2000). A skeletal muscle tissue is comprised of multinucleated muscle fibers that are aligned with one another and organized into packed bundles to maximize contractile force. Amongst the mononucleated cells within a skeletal muscle are the ‘satellite cells’, named according to their anatomic positioning relative to the muscle fibers they reside atop (Mauro, 1961).

Satellite cells are a tissue resident stem cell required for the regenerative potential of healthy adult skeletal muscle tissue (Mauro, 1961; Relaix and Zammit, 2012). They express the paired box transcription factor Pax7 and are mitotically quiescent under homeostatic conditions (Gros et al., 2005; Kassar-Duchossoy, 2005; Relaix et al., 2005; Schultz et al., 1978; Seale et al., 2000). Tissue insult activates satellite cells to divide and give rise to a population of transient amplifying cells co-expressing the myogenic regulatory factors MyoD and Myf5 along with Pax7. Eventually the transient amplifying population downregulates Pax7, MyoD, and Myf5 expression, and upregulates myogenin, thereby committing to exiting the cell cycle and fusing to reform the post-mitotic multinucleated muscle fibers (Bentzinger et al., 2010; Dumont et al., 2015a, 2015b; Tedesco et al., 2010). In culture, the transient amplifying population undergoes a ‘crisis’ phase selecting for a subpopulation of cells referred to as primary myoblasts (pMBs), which can be passaged a limited number of times, and that are competent to fuse together into multinucleated muscle fibers upon mitogen withdrawal.

Myogenic cell transplantation is considered a putative treatment to restore localized strength and function to aged, injured, or diseased skeletal muscle. Indeed, a recent clinical trial focused on patients suffering from oculopharyngeal muscular dystrophy (OPMD) demonstrated an improvement in quality of life, as well as a dose-dependent improvement in swallowing, following autologous myoblast transplantation into the pharyngeal muscles (Perie et al., 2014). Early studies experimenting with pMB transplantation in clinical trials reported challenges including cell survival, rapid injection site clearance, poor donor cell dispersion, and limited contributions to tissue repair (Skuk, 2004). Within the last decade, methods to prospectively isolate mononucleated cells from murine (Liu et al., 2015; Montarras et al., 2005b; Sacco et al.,

2008) and human (Castiglioni et al., 2014; Charville et al., 2015; Xu et al., 2015) skeletal muscle that possess muscle stem cell (MuSC) properties (i.e. self-renewal and differentiation) renewed hope for advocates of myogenic cell transplantation (reviewed in Skuk and Tremblay, 2014). Side-by-side comparisons of the in vivo expansion and regenerative potential of transplanted MuSCs compared to pMBs highlighted the therapeutic potency of MuSCs (Cerletti et al., 2008; Cezar and Mooney, 2015; Collins et al., 2005; Crist et al., 2012; Kuang et al., 2007; Sacco et al., 2008). However, MuSCs are a relatively rare population of cells in skeletal muscle. Despite advances in the area of expanding MuSCs ex vivo while maintaining their regenerative capacity (Charville et al., 2015; Cosgrove et al., 2014; Gilbert et al., 2010), methods to produce clinically relevant numbers of this therapeutic cell population are still under development. Therefore, parallel efforts aimed at optimizing the transplantation procedure to maximize engraftment efficiency promise to synergize with MuSC ex vivo expansion studies to produce a clinically relevant therapy.

Synthetic and natural biomaterials are broadly studied in the context of putative skeletal regenerative medicine applications (Borselli et al., 2011; Cezar and Mooney, 2015; Hill et al., 2006; Lutolf and Hubbell, 2005; Sionkowska, 2011). Biomaterials are an advantageous class of polymers due to the variety of biochemical and biophysical parameters that can be tuned to suit the regenerative medicine application. Surprisingly, the design of injectable biomaterials to improve MuSC transplantation efficiency following a simple intramuscular injection remains understudied despite the clear clinical value. In a recent study, peptide amphiphiles forming an injectable liquid crystalline scaffold were used to encapsulate and deliver murine muscle stem cells intramuscularly (Sleep et al., 2017). The peptide amphiphiles organized to form nanofibers and when extruded through a custom injection device into a tissue, the solution polymerizes via tissue resident divalent ions and the nanofibers align. Notably, MuSC engraftment efficiency was improved when delivered within the synthetic scaffold compared to the saline control. In vitro studies of the C2C12 immortalized cell line and mouse pMBs indicated that scaffold stiffness optimization efforts and the presence of aligned nanofiber maximized cell viability and proliferation, and may account for the observed in vivo benefits. To our knowledge, this was the first study evaluating an injectable cell delivery vehicle to improve MuSC transplantation efficiency.

In this study, we sought to further overcome translational challenges associated with MuSC intramuscular delivery through the study of a biomaterial scaffold shown to improve engraftment of other adult stem cell types (Ballios et al., 2010, 2015; Mothe et al., 2013). HAMC is a hydrogel comprised of two components: hyaluronan (HA) and methylcellulose (MC) (Caicco et al., 2013; Gupta et al., 2006). Hyaluronan is a natural polysaccharide found in all tissues including skeletal muscle extracellular matrix (ECM) (Liang et al., 2016; Toole, 2004). Methylcellulose is a chemical compound derived from cellulose that, when dissolved in water, is liquid at low temperatures but forms a gel at higher temperatures (Ruel-Gariépy and Leroux, 2004). HAMC is an injectable hydrogel due to its shear-thinning properties; while it is a liquid at lower temperatures, HAMC forms a gel at physiological temperatures (Gupta et al., 2006). HAMC is biodegradable and biocompatible in the context of ocular, brain, and spinal cord regeneration, with no observed deleterious effects in vitro or in vivo, and can attenuate the immune response in the brain and spinal cord (Gupta et al., 2006; Wang et al., 2012). Our previous studies demonstrated that using HAMC as a cell delivery vehicle for transplanting neural stem cells into the spinal cord (Mothe et al., 2013), retinal stem progenitor cells to the sub-retinal space (Ballios et al., 2010), retinal stem cell-derived rods to the retina (Ballios et al., 2015), and neural stem cells to the brain (Ballios et al., 2015), improves the survival, distribution, and contribution of the transplanted cells to regeneration.

HA is a naturally occurring ECM ligand (Piehl-Aulin et al., 1991), and since myogenic cells express two of the known HA receptors (Cd44 and Rhamm), it is reasonable to expect that HAMC exerts bioactive effects on encapsulated myogenic cells. For example, myoblasts express CD44 (Lesley et al., 2000), which plays a role in regulating their migration, and differentiation (Mylona et al., 2006). Furthermore, studies of various cell types revealed that CD44-HA interactions promote cell growth and proliferation, and prevent apoptosis (Bourguignon et al., 2005, 2006; Peterson et al., 2000; Slevin et al., 2007; Yu et al., 1997).

In this study, we evaluate HAMC as a myogenic cell delivery vehicle, and investigate the cellular and molecular mechanisms by which it impacts the transplantation outcome. We report that intramuscular delivery within HAMC improves MuSC transplantation efficiency, without the need for a specialized delivery device, by increasing donor fiber numbers and their dispersion in the recipient tissue. From culture and in vivo studies, we conclude that the shear thinning 38

property of HAMC increases the number of MuSCs that emerge during the ejection procedure, and that bioactive properties of HAMC prevent the active clearance of the transplanted cells and delay MuSC differentiation, while promoting MuSC proliferation, and aiding in MuSC migration, thereby culminating in improved engraftment efficiency and dispersion compared to delivery in saline.

2.3 Results

2.3.1 Muscle stem cell delivery within HAMC improves engraftment efficiency and dispersion

To assess the influence of hyaluronan (HA) methylcellulose (MC) as a cell delivery vehicle for murine MuSCs, we performed transplantation assays in mice. Two days prior to transplant, we damaged the tibialis anterior (TA) muscles of recipient mice with a single intramuscular injection of BaCl2 to create a regenerative environment (Figure 2.1B). We prospectively isolated MuSCs from transgenic mice expressing GFP under the control of β-actin (Figure 2.1A). The freshly isolated GFP+ MuSCs were suspended in saline or in 0.75:0.75% w/w HA:MC dissolved in saline and then transplanted intramuscularly into the injured TA muscle of immune-competent wild-type littermates to recapitulate a syngeneic MuSC transplantation therapy. To ensure that MuSCs were consistently transplanted into the center of the muscle, we added fluorescent microbeads into the cell injection mixture (Figure 2.1C-D). 3-4 weeks after transplantation, we euthanized the animals and harvested the TA muscles (Figure 2.1B). The isolated TAs were sectioned and immunostained to visualize the contribution of GFP+ donor cells to the process of regeneration (Figure 2.1C, Figure 2.2A).

Figure 2.1: MuSC isolation and transplantation schematic (A) Representative fluorescence activated cell sorting (FACS) gates for isolation of live (far left plot, PI-), lineage negative (middle left plot, CD31-/CD11b-/CD45-/Sca1-), CD34+/ITGA7+ (middle right plot) MuSCs from GFP+ (far right plot) mice. (B) Schematic of muscle stem cell transplantation protocol. Tibialis anterior (TA) muscles of wild-type (WT) mice were injured with a single intramuscular injection of BaCl2. 48-hours following injury, muscle stem cells (MuSCs) freshly isolated from transgenic mice ubiquitously expressing Green Fluorescent 40

Protein (GFP) were injected into the injured TA muscles of their WT siblings with (+) or without (-) hyaluronan methycellulose (HAMC). TA muscles were harvested 3-4 weeks post- transplantation and analyzed to determine the contribution of the transplanted MuSCs to muscle regeneration. (C) Representative tiled fluorescent image of a transverse section from a tibialis anterior muscle that was harvested and immunostained 4 weeks after transplantation with GFP+ (green) MuSCs. Transplant location was determined by the presence of co-transplanted fluorescent (GFP+) spherical microbeads (yellow arrowheads) that are readily visible in the blown-up image (right). Donor-derived GFP+ fibers (white arrowheads) are viewed in the vicinity as well as distant locations from the transplantation location. (D) Representative epifluorescence images of fluorescent bead distribution one-month following intramuscular transplant into BaCl2 pre-injured TA muscle when delivered within saline (left) compared to HAMC (right). Tissue sections are co-stained to visualize neutrophil (Ly6G, red) and nuclei (Hoechst, blue). Scale bar, 100 µm.

We first performed a limiting dilution assay to assess the ability of transplanted GFP+ donor cells to outcompete endogenous wild-type MuSCs during the regeneration process in the immune-competent recipients. To this end, we transplanted different numbers of freshly isolated MuSCs within saline or HAMC and plotted the number of GFP positive fibers against the number of transplanted cells. We did not detect GFP+ donor fibers following transplantation of 1.5×103 MuSCs, but observed a linear increase in GFP+ fibers following transplant of 5×103 and 10×103 GFP+ MuSCs (Figure 2.2B) within saline (R2=0.81) or HAMC (R2=0.92). Interestingly, MuSC delivery within HAMC reduced transplant variability, an effect also noted in another study where MuSCs were delivered within a synthetic biomimetic scaffold (Sleep et al., 2017). Notably, we observed a >45% increase in the number of donor-derived (GFP+) fibers when MuSCs were delivered in HAMC compared to the saline control (Figure 2.2C). We observed no differences in donor fiber cross-sectional area (Figure 2.2E-F), indicating that HAMC does not induce fiber hypertrophy. Interestingly, delivery within HAMC resulted in a significant increase in donor-derived muscle fiber dispersion (Figure 2.2A,D), suggesting that by optimizing the cell delivery vehicle it is possible to overcome a commonly observed MuSC transplant hurdle: failure of transplanted cells to engraft at sites distal to the injection site (Beauchamp et al., 1999; Marquardt and Heilshorn, 2016). From these results, we conclude that transplanting MuSCs within HAMC improves on two stem cell therapy challenges: engraftment efficiency and cell dispersion.

Figure 2.2: Delivering MuSCs within HAMC improves engraftment efficiency and dispersion (A) Representative tiled phase and epifluorescence composite images of TA and extensor digitorum longus (EDL) muscles immunostained for GFP at 4-weeks following transplantation with 10×103 MuSCs delivered in saline control (left) or HAMC (right). Scale bar, 1 mm. (B) Paired scatter plots comparing the number of GFP+ fibers quantified in recipient animals one- month following injection of 1.5x103 (n = 2), 5x103 (n = 5), and 10×103 (n = 4) freshly isolated GFP+ MuSCs that were delivered intramuscularly within (left) saline or (right) HAMC. Please note that the data in these paired scatter plots were used to generate the Figure 2.2C bar graphs. (C) Graph portraying GFP+ fiber number following injection of 5×103 (left, n = 5) and 10×103

(right, n=4) MuSCs within saline (dark grey) or HAMC (black). (D) Bar graph showing the fold change in the distribution of the GFP+ fibers (dispersion) 1-month following transplantation of 5x103 or 10x103 MuSCs within saline or HAMC control. n = 6. (E) Bar graph showing the average cross-sectional area (CSA) of GFP+ donor-derived fibers arising from GFP+ MuSCs delivered intramuscularly within saline (dark grey) or HAMC (black). n = 9. (F) Histogram displaying donor-derived GFP+ fiber CSA comparing saline (dark grey) to HAMC (black) delivery of GFP+ MuSCs. Error bars indicate SEM. Statistical significance determined by paired (Figure 2.2C) or unpaired (Figure 2.2D-E) student’s t-test where; p < 0.05 2.3.2 HAMC improves muscle stem cell ejection efficiency

Next, we sought to understand how HAMC influenced the transplanted cells using in vitro assays. Our aim was to recapitulate the complex tissue environment and the various conditions the transplanted cells are exposed to prior, during, and after the transplantation process, as well as during regeneration, to assess the effects of HAMC on early myogenic fate post- transplantation. We studied freshly isolated MuSCs and low passage pMBs in culture to determine if HAMC elicits differential effects based on stem cell hierarchy status (i.e. quiescence, activation, transient amplifying).

HAMC was previously shown to improve cell viability (Ballios et al., 2015). Therefore, we tested the hypothesis that HAMC improves myogenic cell viability resulting in a greater number of therapeutic cells engrafting into the host. HA is a natural component of the skeletal muscle microenvironment (Piehl-Aulin et al., 1991) so we first assessed whether passive culture within bioactive HAMC impacts myogenic cell viability. We gently resuspended pMBs in HAMC reconstituted in growth media using a wide-bore tip, to reduce shear stress, and found no significant differences in the viability of the pMBs after 24 and 48 hours of culture as compared to the growth media control (Figure 2.3B-C).

HAMC is a shear thinning hydrogel (Gupta et al., 2006), so we next tested whether the reduced shear stress during ejection through a narrow needle protected the cells from damage during the transplant process (Figure 2.3A). Here we quantified the proportion of viable MuSCs (Figure 2.3D) and pMBs (Figure 2.3E) in the culture well 2 hours or 24 hours after ejection through a 10 µL Hamilton syringe equipped with a 32G needle. Again, we observed no differences in cell viability when comparing myogenic cell delivery in HAMC versus the culture media control (Figure 2.3D-E). Our results are in line with prior pMB studies concluding that passage through

a syringe and needle partially damages the cell membrane, but does not result in cell death in vitro (Baines and Molkentin, 2005; Chazaud, 2003; Guérette et al., 1997).

Figure 2.3: HAMC improves ejection efficiency without altering cultured myogenic cell viability (A) Schematic of cell viability assays comparing HAMC to media control. Primary myoblasts (pMBs) and muscle stem cells (MuSCs) were plated with or without HAMC using a wide-bore pipette tip (top) or a syringe with a 32G needle (bottom) into a 96 well plate. At the time-points indicated, the calcein AM and ethidium homodimer ‘live/deadTM assay’ was used to assess cell viability in each condition. (B) Representative epifluorescence images of pMBs cultured in control media or HAMC and co-stained with calcein AM (green, live) and ethidium homodimer (Ethd-1, red, dead). Scale bar, 200 µm. (C) Bar graph indicating the viability of pMBs 24 and 48

hours (hrs) after plating in media control (dark grey) compared to HAMC (black) culture. n = 3. (D-E) Bar graphs showing the viability of (D) MuSCs and (E) pMBs 2 and 24 hours following ejection through a syringe equipped with a 32G needle when delivered within media (MuSCs, dark grey; pMBs, white) or HAMC (MuSCs, black; pMBs, light grey). n = 3. (F) Bar graph indicating the viability of pMBs 24 hours after ejection through a syringe and 32G needle when delivered within media (white) or HAMC (light grey) into a 1mg / mL solution of skeletal muscle (SKM) extract that was prepared 48 hours after a BaCl2-induced tissue injury. n = 3. (G) Bar graph showing the percentage of MuSCs lost during ejection through a syringe and 32G needle when delivered within media (dark grey) as compared to HAMC (black). n = 3. Error bars indicate SEM. Statistical significance determined by student’s t-test where; p < 0.05.

If the ejection process induces membrane damage, we posited that exposure to biomolecules present in the hostile regenerative environment might push the transplanted cells towards death, and that perhaps HAMC protects against this. To simulate the regenerating environment of injured skeletal muscle that the donor cells are injected into, we injured the TA of C57Bl/6N mice, harvested the injured tissue 48 hours later, and prepared a tissue extract (Chen and Quinn, 1992). We then resuspended primary myoblasts in HAMC or media control and ejected the cell suspensions through a Hamilton syringe equipped with a 31G needle directly into a culture well containing the injured skeletal muscle tissue extract. As expected, ejection and culture in injured tissue extract resulted in an overall reduction in cell viability (Figure 2.3F; ~60%) compared to the growth media (Figure 2.3C,E; ~80%), and delivery in HAMC did not afford protection from death in this setting (Figure 2.3F). These results suggest that HAMC does not improve the viability of myogenic cells that pass through the syringe and needle.

All of our viability studies to this point focused on analyzing the cells after ejection through the syringe and needle. Interestingly, when we quantified the total number of MuSCs that passed through the syringe and needle into the culture dish, we observed a greater number of cells when delivered in HAMC compared to the saline control (Figure 2.3G). This translated to a 6% reduction in stem cell loss during the ejection procedure. Therefore, we conclude that myogenic cell delivery in HAMC does not modify the viability of the cells that emerge from the needle. However, HAMC ultimately increases ejection efficiency by protecting cells from obliteration during the ejection or by preventing cells from becoming lodged in the syringe or needle during ejection, thereby resulting in an increase in the total number of myogenic cells transplanted into the tissue.

2.3.3 HAMC promotes MuSC proliferation via a CD44-independent mechanism

Based on our limiting dilution analysis (Figure 2.2B), the modest improvement in ejection efficiency (6%;Figure 2.3G) does not fully account for the >45% increase in donor derived fibers observed when MuSCs are delivered in HAMC (Figure 2.2C). Therefore, we investigated the possibility that HAMC may also influence myogenic cell proliferation. Freshly isolated MuSCs were cultured in growth media or HAMC reconstituted in growth media for a period of 72 hours and pulsed with 5-ethynyl-2’-deoxyuridine (EdU) during the final 12 hours of culture (Figure 2.4A). Intriguingly, we found that the number of cells that incorporated EdU was ~14% higher when embedded in HAMC compared to culture in the growth media control condition (Figure 2.4B, left). In addition to revealing another benefit of using HAMC for MuSC delivery, these results suggest that HAMC either pushes a greater proportion of MuSCs to activate from quiescence or it increases the proliferation rate of MuSCs following activation.

In the first week of muscle repair, MuSCs activate and proliferate to give rise to a transient amplifying progenitor pool that will ultimately fuse to produce nascent muscle fibers on the third day of repair in the regenerating tissue. Since prior studies showed that HAMC remains at the injection site for as many as 7 days (Ballios et al., 2010), we next investigated the effects of HAMC on pMB (i.e. transient amplifying progenitors) cell cycle entry. pMBs were cultured in growth media or HAMC reconstituted in growth media for a period of 36 hours and pulsed with EdU during the final 12 hours of culture. Interestingly, similar proportions of pMBs incorporated EdU when cultured in HAMC compared to the growth media control (Figure 2.4D, left). We then investigated the effect of HAMC on pMB cell cycle entry in the context of the regenerating environment by ejecting cells into injured skeletal muscle tissue extract. In this context, we observed an overall lower incidence of EdU incorporation, with no significant differences observed between the HAMC and control conditions (Figure 2.4D, right). Together, these data (Figure 2.4B,D) suggest that HAMC-induced effects on myogenic cell cycle entry are limited to MuSCs, and not their downstream progeny, revealing an intriguing hierarchical bias.

Figure 2.4: HAMC influences MuSC proliferation via a CD44-independent mechanism (A) Schematic of 5-ethynyl-2’-deoxyuridine (EdU) assay comparing myogenic cell proliferation in media control as compared to HAMC culture. Primary myoblasts (pMBs) or muscle stem cells (MuSCs) were plated into a 96 well plate with or without HAMC using a wide-bore pipette tip. At the culture time-points indicated, EdU was added to the culture media for 12 hours (hrs). Cells were then fixed, stained, and analyzed for EdU incorporation. (B) Bar graphs showing the percentage of MuSCs freshly isolated from (left) wild-type (WT; n=6) or (right) Cd44 knock-out (Cd44KO; n=3) mice that incorporate EdU after 72 hours of culture in media control (WT, dark grey; Cd44KO, patterned dark grey) or HAMC (WT, black; Cd44KO, patterned black) when pulsed with EdU for 12 hours prior to the analysis time-point. (C) Representative flow cytometric plots analyzing Cd44 cell surface expression on freshly isolated quiescent (bottom left) and activated (bottom right; 24 hours post-BaCl2 injury) MuSCs, and primary myoblasts (top right), as compared to isotype control staining (top left). (D) Bar graphs depicting the percentage of primary myoblasts that incorporated EdU after 24 hours of culture when plated in growth media control (white) or HAMC (light grey) in the presence (right) or absence (left) of injured skeletal muscle (SKM) lysate when pulsed with EdU for 12 hours prior to the analysis

time-point. n=3. Error bars indicate SEM. Statistical significance determined by student’s t-test where; p < 0.05.

Signaling through the HA-Cd44 receptor-ligand axis positively regulates cell cycle entry in a wide variety of cell types (Bourguignon et al., 2005, 2006; Peterson et al., 2000; Slevin et al., 2007; Yu et al., 1997). Therefore, we sought to determine whether the proliferative influence of HAMC on MuSCs was mediated by Cd44. First, we used flow cytometry to determine if myogenic cells express and present the Cd44 receptor. Consistent with recent reports (Porpiglia et al., 2017), we found that Cd44 is not expressed on freshly isolated MuSCs, while activated MuSCs and primary myoblasts both have Cd44 cell surface expression (Figure 2.4C). Next, freshly isolated MuSCs from Cd44-/- mice were cultured in HAMC or the growth media control for 72 hours and pulsed with EdU for the final 12 hours of culture (Figure 2.4A). In contrast with reports for other cell types (Ballios et al., 2015; Bourguignon et al., 2005, 2006; Peterson et al., 2000; Trochon et al., 1996), loss of Cd44 elicited a dramatic increase in the proportion of MuSCs entering cell cycle compared to the wild-type control (Figure 2.4B; grey solid and patterned bars). Furthermore, similar to our wild-type MuSC study (Figure 2.4B; left), culture within HAMC increased the proportion of Cd44KO MuSCs that incorporated EdU during the final 12 hours of the 3-day culture period (Figure 2.4B; right). Taken together, our results suggest that HAMC culture leads to an increase in MuSC, and not pMB proliferation, via a CD44- independent mechanism.

2.3.4 HAMC does not modify the skeletal muscle innate immune response

One of the common causes of cell death post-transplantation is the inflammatory reaction, with the neutrophil response inducing particularly deleterious effects on cell survival (Bouchentouf et al., 2007a; Sammels et al., 2004). Prior studies showed that HAMC attenuates the inflammatory response (i.e. the presence of microglia and astrocytes) in the brain and spinal cord (Gupta et al., 2006; Wang et al., 2012). However, the effect of HAMC on the inflammatory environment in skeletal muscle has not been investigated. To narrow in on additional mechanisms by which HAMC improves the engraftment efficiency of the transplanted cells, we investigated the presence of neutrophils and macrophages following HAMC injection. First, we injected a barium chloride solution intramuscularly into the TA muscles of C57Bl/6N wild-type mice. 48 hours later we injected saline +/- HAMC (and mixed with fluorescent beads) intramuscularly into the

center of the regenerating TA muscle group. At 2 and 24 hours post-injection, we harvested the tissues for immunohistological analysis (Figure 2.5A).

As expected, neutrophils and macrophages of the innate immune system, expressing Ly6G and Cd68, respectively, have a scarce presence in healthy, uninjured TA muscle (Figure 2.5B-C, far left panels). However, their incidence dramatically increased 50 hours (Figure 2.5B; middle left panel) and 72 hours (Figure 2.5C; middle left panel) after a barium chloride-mediated tissue injury. HAMC injection did not induce gross alterations in the incidence or localization of neutrophils 50 hours after injury (Figure 2.5B, D) or the neutrophils (Figure 2.5C; top panels) and macrophages (Figure 2.5C; bottom panels) 72 hours (Figure 2.5C-F) after injury, when compared to saline control injections. Based on these results, we conclude that HAMC addition does not modify days 2 – 3 of the skeletal muscle regeneration innate immune response. However, it remains to be determined whether HAMC influences earlier time-points in the repair process or whether HAMC influences immune cell behavior and / or secretome.

Figure 2.5: HAMC does not modify Ly6G+ or CD68+ immune cell incidence in the first 24 hours post-transplantation (A) Schematic of experiments designed to assess the short-term effects of HAMC transplantation on the early regeneration immune response. Tibialis anterior (TA) muscles of wild-type (WT) 50

mice were injured with BaCl2. 48 hours post-injury, HAMC hydrogel or a control media was injected into injured TA muscles of WT mice. The TA muscles were isolated 2 or 24 hours post- transplantation (equivalent to 50 and 72 hours post-injury) and analyzed immunohistologically to visualize the immune response at early time points post-injection. (B) Representative epifluorescence images of neutrophil (Ly6G, red) distribution in uninjured skeletal muscle (far left) and injured tissue 50 hours post BaCl2-injury (middle left). Images of neutrophil presence 2 hours post-injection of saline control (middle right) or HAMC (far right) into the pre-injured TA are also shown. Scale bar, 100 µm. (C) Representative epifluorescence images of neutrophil (top, Ly6G) and macrophage (bottom, CD68) distribution in uninjured (far left) and injured tissue 72 hours post BaCl2-injury (middle left). Images of neutrophil and macrophage presence 24 hours post-injection of saline control (middle right) or HAMC (far right) into the pre-injured TA are also shown. Scale bar, 100 µm. (D-F) Bar graphs depicting the normalized Ly6G (neutrophil) expression at the site of injection (D) 2 hours and (E) 24 hours following saline or HAMC intramuscular injection. n = 3. (F) Comparison of normalized Cd68 (macrophage) expression at the site of injection 24 hours following saline or HAMC intramuscular injection. n = 3. 2.3.5 HAMC prevents the active clearance of transplanted myogenic cells

Previous studies revealed that a majority of transplanted myoblasts do not withstand the first 4 days post-transplantation (Bouchentouf et al., 2007a). Loss of cells over these early time-points might be due to death but might also be owed to passive clearance or active (e.g. immune cell mediated) clearance mechanisms. Therefore, we designed a series of experiments to determine whether HAMC prevents cell clearance post-transplantation. Since HAMC did not influence pMB proliferation in culture (Figure 2.4D), we utilized pMBs in these studies to avoid proliferation as a confounding parameter.

We transplanted 1x105 GFP+ pMBs into the TA of WT littermate control animals 48 hours after a BaCl2-induced injury. pMBs were delivered within saline or HAMC and after 2, 24, or 48 hours, we harvested and dissociated the TA into a mononucleated cell slurry. With flow cytometry, we quantified the number of retrieved cells (Figure 2.6A-B). No significant differences were observed in the number of GFP+ cells retrieved 2 hours after transplanting GFP+ pMBs in HAMC versus saline (Figure 2.6C-D), suggesting that the HAMC hydrogel did not serve to prevent the passive clearance of the injected cell population. At later time-points post- injection, we expect cell clearance is due to active, immune cell-mediated mechanisms. No differences were observed 24 hours post-injection (Figure 2.6C, E), but at 48 hours we observed a ~1.5-fold increase in the number of cells retrieved when they were delivered in HAMC (Figure 2.6C, F). Since we observed no differences in cell retrieval at 2 and 24 hours post-transplant, we

conclude that HAMC does not influence passive or early active cell clearance. However, since the 48-hour post-transplant analysis time-point coincides with the period when mononucleated cells fuse to form multinucleated muscle fibers during regeneration, we conclude that aside from preventing the active clearance of cells at this timepoint, HAMC may serve to delay MuSC differentiation and instead preserves an extended proliferation window to augment transplanted MuSC engraftment efficiency (Figure 2.6F).

Figure 2.6: HAMC increases retention of transplanted cells post-transplantation

(A) Schematic of experiments aimed at determining the effect of HAMC on mononucleated cell retention within the first two days post-transplant. Tibialis anterior (TA) muscles of wild-type 5 (WT) mice were injured with BaCl2. 48 hours post-injury, 10 actin-GFP primary myoblasts were transplanted intramuscularly into the injured TA muscles within saline or HAMC hydrogel. The TA muscles were isolated 2, 24, or 48 hours post-transplantation, enzymatically digested, and analyzed by flow cytometry to quantify the number of GFP+ mononucleated cells. (B) Representative flow plots of retrieved GFP+ myoblasts 2 hours post-transplantation within saline (left) or HAMC (right). (C) Scatter plot displaying the fold change in GFP+ myoblasts retrieved over time when delivered within HAMC (triangle) compared to saline (square) where data are normalized to the number of cells retrieved at 2 hours post-transplantation within saline. (D-F) Bar graphs replotting the raw data from (C) to indicate the absolute number of GFP+ myoblasts retrieved from enzymatically digested skeletal muscle tissue (D; n = 3) 2 hours, (E; n=7) 24 hours, and (F; n = 3) 48 hours after intramuscular delivery within HAMC (light grey) compared to saline control (white). Error bars indicate SEM. Statistical significance determined by student’s t-test where; p < 0.05.

Based on our culture and in vivo results, we conclude that HAMC improves the engraftment efficiency of transplanted MuSCs by supporting the delivery of an overall greater number of cells, by promoting proliferation, prevention of active clearance, and by delaying differentiation, which in turn serves to augment in vivo expansion. Furthermore, by visualizing the position of the fluorescent beads that were co-delivered with the MuSCs relative to engrafted GFP+ muscle fibers (Figure 2.1C-D), it is clear that passive cell dissemination does not account for the HAMC-mediated improvement in donor fiber dispersion. Since, Cd44 is also known to promote myogenic cell migration, it is enticing to speculate that improved cell dispersion is due to the active engagement of Cd44 receptors mediated by HA-induced Cd44 receptor ligation.

2.4 Discussion

Our results show that the application of HAMC, a shear thinning hydrogel cell delivery vehicle, to MuSC intramuscular transplantation, increases engraftment efficacy and dispersion. Our mechanistic studies suggest that enhanced engraftment effects are owed to both HAMC material and bioactive properties. To our knowledge, this is the first report of a bioactive delivery vehicle improving MuSC engraftment efficiency and the first delivery vehicle of any kind augmenting MuSC post-transplant dispersion.

We sought to explore MuSC transplantation efficacy in the context of an intact immune response since the immune system is known to contribute to the process of skeletal muscle repair (reviewed in Tidball, 2005). In contrast, most studies in the area of MuSC transplantation utilize 53

immuno-compromised and hindlimb irradiated animals as recipients. As such, we first performed limiting dilution analyses to ensure that the number of GFP+ MuSCs transplanted into syngeneic immune competent sibling recipients was in the linear range of engraftment for our saline and HAMC delivery conditions. Plotting the number of histologically detected GFP+ fibers versus the number of transplanted cells revealed that even at our highest transplant number (1×104), we were still within the linear range, as we had not yet reached a fiber engraftment plateau (Figure 2.2B) (Praud et al., 2003; Rando and Blau, 1994; Skuk and Tremblay, 2014; Skuk et al., 2014). Since the transplanted cells compete with endogenous MuSCs, this experimental model and data may be useful to future studies aimed at recapitulating autologous MuSC transfer.

We found that delivering MuSCs intramuscularly within HAMC rather than within saline resulted in an overall greater number of GFP+ donor-derived fibers in transplant recipients (Figure 2.1B, Figure 2.2A-C). While HAMC did not induce fiber hypertrophy that was detectable after 4-weeks of regeneration (Figure 2.2E-F), we cannot rule out the possibility that HAMC may influence muscle fiber hypertrophy at earlier time-points of repair. However, given that skeletal muscle contains copious amounts of hyaluronic acid, it seems unlikely that HAMC would exert an additive effect in this regard.

To identify a cellular mechanism to understand the positive impact of HAMC on MuSC engraftment efficacy, we evaluated several culture scenarios to explore the hypothesis that HAMC may improve transplanted MuSC viability. By simply mixing pMBs in HAMC and quantifying cell viability after 24 and 48 hours of exposure, we detected no differences between the two culture conditions (Figure 2.3A-C). Shear thinning hydrogels can improve cell transplantation outcome by reducing the extensional flow at the entrance of the needle, which otherwise will induce membrane damage during the ejection (Aguado et al., 2012). However, when we assessed the viability of MuSCs or pMBs that emerged from the syringe and needle when encapsulated in HAMC, we observed no differences compared with ejection within saline control (Figure 2.3D-E). This result aligns with that of previous studies (Baines and Molkentin, 2005; Chazaud, 2003; Guérette et al., 1997) showing that pMBs that survive the ejection process maintain high viability in culture. Since it was previously reported that the ejection process compromises membrane integrity (Baines and Molkentin, 2005), we tested the hypothesis that HAMC may protect myogenic cells from death signals they encounter in the hostile transplant 54

environment. While the viability of pMBs ejected into an injured skeletal muscle extract was decreased compared to ejection into regular growth media (Figure 2.3E-F), HAMC did not elicit a protective effect in this context either (Figure 2.3F). We note that our results with myogenic cells conflict with our prior studies demonstrating that HAMC encapsulation increases the proportion of viable cells after incubation (Ballios et al., 2010, 2015; Caicco et al., 2013; Mitrousis et al., 2016), and we expect this is an example of a HAMC cell type specific effect.

Finally, we evaluated the total number of MuSCs that emerged from the syringe and needle when delivered in HAMC compared to the saline control. In the highly controlled setting of culture, we found that more MuSCs emerged from the syringe and needle when delivered in HAMC (~6%). We expect that HAMC either protects MuSCs from being obliterated during ejection or, since hyaluronan is made of polysaccharides, the sugar groups may act as a cellular ‘slip-n-slide’ to prevent cells from becoming lodged in the syringe or needle. Interestingly, when we harvested and dissociated TA muscles to retrieve and quantify myogenic cells shortly after intramuscular injection (2 hours), we saw no difference in the total number of cells retrieved when delivered in HAMC compared to saline (Figure 2.6A-D). However, experimental error introduced during the process of tissue digestion, mononucleate cell retrieval, and flow cytometric analysis may introduce error that effectively masks the small, but statistically significant, 6% difference in transplanted cell number. This observation serves as a cautionary tale when drawing conclusion solely based on in vitro data.

Ultimately a 6% increase in delivered MuSCs does not fully account for the 1.5-fold increase in donor derived fibers that we observed when using HAMC as the delivery vehicle. As such, we explored the possibility that HAMC encapsulation promotes MuSC proliferation potential. Indeed, we observed a greater proportion of HAMC-encapsulated MuSCs in cell cycle (~14%) when we assessed EdU incorporation in the freshly isolated MuSC population after an EdU pulse in the final 12 hours of the 72 hour culture period (Figure 2.4A-B). Interestingly, this effect appears to be specific to activated MuSCs since pMB cell cycle status was insensitive to the presence of HAMC (Figure 2.4A,D). Even applying an additional environmental pressure, such as exposure to lysate from regenerating skeletal muscle, did not uncover HAMC-mediated effects on pMB proliferation (Figure 2.4A,D), thereby uncovering a MuSC-specific effect of HAMC. 55

HA-Cd44 interactions induce Erk1/2 and Pi3k-Akt signaling pathway activation, leading to the growth and proliferation of other cell types (Bourguignon, 2008; Bourguignon et al., 2006, 2007; Slevin et al., 2007; Toole, 2004). Consistently, inhibition of Cd44-HA results in the inhibition of growth (Peterson et al., 2000) and cell apoptosis (Yu et al., 1997) of other cell types. Since it was previously shown that HA-mediated effects on cell proliferation were mediated by Cd44 (Bourguignon et al., 2005, 2006; Peterson et al., 2000; Trochon et al., 1996), we sought to determine whether the Cd44-HA signaling axis was an underlying mechanism of HAMC- mediated MuSC proliferation. We first assessed cell surface expression of Cd44 on myogenic cells (Figure 2.4C) and found that activated MuSCs and pMBs uniformly express the receptor, but, similar to recent studies using Cytof (Porpiglia et al., 2017), only a small subpopulation of freshly isolated MuSCs presented the receptor. When we prospectively isolated MuSCs from Cd44-/- mice, encapsulated and cultured the MuSCs in HAMC for 72 hours, and then assessed EdU incorporation, we still observed a greater proportion of MuSCs in cell cycle compared to the control condition (Figure 2.4A-B). Interestingly, comparing the control conditions (Figure 2.4B grey solid and striped bars), revealed a greater proportion of EdU incorporated MuSCs in the absence of the Cd44 receptor at this time-point. It is possible that loss of Cd44 in MuSCs releases a brake on MuSC proliferation. However, given that a prior study noted a delay in differentiation when studying Cd44-/- pMBs in culture (Mylona et al., 2006), our MuSC results may instead reflect delayed activation. Regardless, the HAMC-mediated boost in MuSC proliferation cannot be accounted for by a Cd44-HA interaction, and instead implicates a role for another HA receptor such as Receptor for hyaluronan-mediated-motility (Rhamm).

Since HAMC forms a gel at physiological temperatures, we hypothesized that delivery in HAMC might protect MuSCs from being passively cleared in the earliest time-points after injection. However, when we collected GFP+ cells 2 hours after intramuscular injection, flow cytometric analysis revealed no difference in the total number of retrieved cells when delivered in HAMC compared to saline.

Next, we designed a series of experiments to determine whether HAMC might modify the skeletal muscle immune environment to support the observed improvements in MuSC engraftment efficiency. High cell death, especially during the first 4 days post-transplantation, is a common reason for transplantation failure (Skuk and Tremblay, 2014). The skeletal muscle 56

inflammatory response after exercise or acute injury includes the invasion of neutrophils, macrophages, and natural killer (NK) cells (Bouchentouf et al., 2007a; Tidball, 2005). Studies showed that neutrophils secrete proteases to break down cellular debris, and in the process, can cause damage and death to bystander healthy cells through release of cytokines and cytotoxins (Sammels et al., 2004; Tidball, 2005; Tiidus, 1998). Macrophages and NK cells, on the other hand, do not appear to be detrimental to the fate of transplanted cells, and are more targeted in their activities (Sammels et al., 2004; Skuk et al., 2002). Since previous studies on HAMC showed it to attenuate inflammation in the brain and spinal cord by reducing the presence of astrocytes and microglia, respectively (Gupta et al., 2006; Wang et al., 2012), we next assessed whether HAMC elicits similar effects in skeletal muscle. In this regard, we mimicked our transplantation regime by injuring immune-competent C57Bl/6N hindlimbs with a single injection of BaCl2 and then 2 days later, injected saline or HAMC, and then harvested the tissue for histological analysis 2 hours or 24 hours later (Figure 2.5). We observed no gross differences in the distribution or accumulation of Ly6G+ neutrophil and Cd68+ macrophage cell populations compared to uninjected and saline injected controls.

Our histological analysis does not out-rule the possibility that perhaps HAMC shields the transplanted cells from deleterious interactions with the immune system that might then manifest as an increase in the total number of transplanted cells present over time. Indeed, previous studies showed that blocking interactions between neutrophils and transplanted cells increases their chance of survival (Guérette et al., 1997). Additionally, treating cells with fibronectin and vitronectin increased the survival of transplanted myoblasts, and reduced anoikis (Bouchentouf et al., 2007a, 2007b). Therefore, the HA in HAMC might prevent anoikis by activating pathways involved in cell survival and growth such as Erk1 or Pi3k-Akt pathways (Toole, 2004). To test this theory, we transplanted GFP+ pMBs into pre-injured tissue and then retrieved and enumerated the cells using flow cytometry 24 hours and 48 hours later (Figure 2.6C,E-F). Since our culture studies concluded that HAMC does not influence pMB proliferation, we used pMBs in our retrieval studies to avoid the confounding influence of proliferation on cell count studies. While no differences in retrieved cell number were uncovered 24 hours after injection, at 48 hours a statistically significant decrease in total retrieved cells was observed in the saline control condition while the HAMC condition seemed to maintain a constant number of cells when

assessed at 2, 24, and 48 hours post-transplant (Figure 2.6A-F). This data supports a protective effect of HAMC in preventing cell death and/or clearance by immune cells. However, 4 days after injury (i.e. 48 hours after transplant in our experimental regime; Figure 2.6A) corresponds to the time-point when myogenic progenitors fuse to form multinucleated muscle fibers. Therefore, we favor the possibility that the reduction in the number of mononucleated cells retrieved at 48 hours can be explained by fusion and that HAMC serves to delay differentiation, similar to previous studies showing biomaterials can delay the differentiation of transplanted myoblasts and thereby increase the contribution of transplanted progenitor cells to host muscle regeneration (Borselli et al., 2011; Hill et al., 2006).

In addition to delivering a greater number of cells and influencing proliferation, our HAMC data indicate that the delivery scaffold improves the migration of transplanted cell resulting in greater dispersion of donor-derived muscle fibers (Figure 2.2D). Studies showed that the interaction between HA and its receptors, including Cd44 and Rhamm, leads to increased cell motility, and is involved in supporting the migration of many cell types including pMBs (Bourguignon et al., 2000; Evanko et al., 1999; Mylona et al., 2006; Zhu et al., 2006). Additionally, a recent study suggested that a biphasic relationship exists between Cd44 expression and survival and migratory behavior of glial cells in that increased Cd44 expression resulted in faster migration rates and lower cell survival whereas lower Cd44 expression lead to higher survival and less migration (Klank et al., 2017). Hence, the increased dispersion of GFP+ donor-derived fibers could be a result of the increased HA presence in the local milieu (Ponta et al., 2003) or HA receptor ligation induced in the process of MuSC encapsulation and delivery. Since pretreating MuSCs with recombinant Wnt7a prior to transplant also influences cellular dispersion (Bentzinger et al., 2014), it would be interesting to determine whether HAMC delivery affords an additive effect on this engraftment metric.

Together, our study aimed at optimizing the MuSC delivery method and exploring the mechanistic underpinnings of our in vivo results pave the way for reducing the number of cells required for cell transplantations to make MuSC transplant therapy more feasible and cost- effective.

2.5 Limitations and future work

Ultimately, an important end goal of MuSC transplantation is improved skeletal muscle function. It is unclear whether the 1.5-fold increase in donor derived fibers reported in our study translates to an increase in muscle force generation. Further experiments are required to determine whether any functional changes are observed in muscles receiving HAMC treatment compared to saline control. Additionally, our results indicated that the HAMC-mediated increases in MuSC proliferation cannot be attributed to Cd44-HA interactions. The mechanism behind the observed increase in MuSC proliferation will need to be uncovered as a next step. Rhamm, an alternative HA receptor mainly implicated in cell migration, might be responsible for the observed phenomenon. Alternatively, MuSCs cultured in HAMC and growth media control are in a 3D and 2D environment, respectively; these different culture conditions could potentially lead to the observed difference in MuSC EdU incorporation.

Since beginning this study, several other studies have demonstrated the beneficial effects of biomaterials in accelerating and improving MuSC transplantation therapies. We predict that by combining the findings of these studies with our observations, it is possible to create enhanced MuSC delivery vehicles. For instance, MuSCs’ preferred binding substrate is fibronectin (Fn1; Lukjanenko et al., 2016), and RGD sequences are the sites of cell attachments to Fn1 (Sechler et al., 2013). A recent study revealed that implanting MuSCs encased in a biodegradable PEG synthetic matrix functionalized with integrin-binding RGD sequences into injured skeletal muscle, promotes MuSC survival and engraftment in aged and dystrophic mice (Han et al., 2018). Given that the RGD sequences were the bioactive components of the non-injectable synthetic matrix, we speculate that by covalently conjugating RGD peptides to methylcellulose (MC) as previously demonstrated (Tam et al., 2012), it is possible to create a modified HAMC hydrogel with increased bioactive properties. In a separate study, researchers demonstrated that an injectable peptide amphiphile-based delivery scaffold improved engraftment efficiency of transplanted MuSCs (Sleep et al., 2017). Since the transplantation benefits we observed in our studies were largely attributed to bioactive influences of HAMC, we predict that modifying the recently reported peptide amphiphile-based MuSC delivery scaffold to include hyaluronan peptide sequences would serve to boost the efficacy of the therapeutic delivery platform. Finally, since Wnt7a pre-transplant treatment improves MuSC engraftment efficiency and donor fiber

dispersion (Bentzinger et al., 2014), we speculate that we might expect additive improvements if the Wnt7a pre-treated therapeutic cell population was delivered in an optimized delivery scaffold.

2.6 Materials and methods

2.6.1 Animals

All animal use protocols were reviewed and approved by the Division of Comparative Medicine (DCM) at University of Toronto. C57Bl/6NCrL (Charles River) and C57BL/6-Tg(CAG- EGFP)1Osb/J (expressing eGFP under the control of Actin, Jackson Laboratories), B6.129(Cg)- Cd44tm1Hbg/J (CD44-/- mice courtesy of Dr. Tak Mak, University of Toronto; Weber et al., 2002), Pax7-zsGreen reporter mice (courtesy of Dr. Michael Kyba, University of Minnesota; Bosnakovski et al., 2008) were used in this project. The C57BL/6-Tg(CAG-EGFP)1Osb/J and Pax7-zsGreen lines were maintained as a heterozygous line by breeding against wild-type C57Bl/6NCrL females, resulting in litters comprised of both wild type and transgenic pups. The B6.129(Cg)-Cd44tm1Hbg/J line was maintained as a homozygous line. 8-12 week old mice (wild type or transgenic) were used in all of the experiments. Injuries to the tibialis anterior (TA) muscle were induced by injecting 30 µL of a 1.2% BaCl2 (Bio Basic, cat. no. BC2020) into the center of the TA muscle of anaesthetized mice using a 100 µL insulin syringe (BD, cat. no. 324702).

2.6.2 Hyaluronan (HA) and methylcellulose (MC) preparation

Sodium Hyaluronate (HA; Kibun Food Chemifa, cat. no. 9004-61-9) and methylcellulose (MC;

Shin Etsu, cat. no. SM-4000) were dissolved in distilled H2O at 0.5 and 1 g/L at 4˚C overnight. HA and MC solutions were then sterile filtered using a 0.2 µm vacuum filter and aliquoted in 35 mL aliquots in 50 mL conical tubes and flash frozen in liquid N2. The caps were then replaced with perforated, filter caps (Corning, cat. no. 431720) and placed in -80˚C for 2 hours. The tubes were then lyophilized for 2 - 3 days. The resulting sterile HA and MC were then weighed out at a 1:1 ratio, depending on the amount of HAMC required, and dissolved in the appropriate solution at a 1:1 % weight/volume ratio overnight at 4˚C on a rocking shaker.

2.6.3 Primary myoblast isolation

Primary myoblast lines were established using a method similar to the one described in Rando et al (Rando and Blau, 1994). Briefly, all of the hind limb muscles were dissected from humanely euthanized 8-12 week old mice and placed in 7 mL DMEM containing Type IA collagenase (Sigma, cat. no. C9891) at 628 Units/mL. The muscles were then dissociated using a gentleMACS Dissociator (Miltenyi Biotec, cat. no. 130-093-235). Tissues were then incubated on a nutating mixer at 37˚C and 5% CO2 for 90 minutes. Dispase II (Life Technologies, cat. no. 17105041) was added at a concentration of 4.8 U/mL and the muscle was incubated for another

30 minutes on the nutating mixer at 37˚C and 5% CO2. After the incubation, the tissues were passed through a 20 G needle equipped with a 10 mL syringe to fully dissociate the tissue. 7 mL of growth media was then added to the dissociated tissue and the entire volume was filtered using a 40 µm cell strainer (Corning, cat. no. 352340). The filtered liquid was centrifuged at 400g for 15 minutes. The supernatant was then discarded, and the pellet resuspended in 1 mL of red blood cell lysis buffer (0.155 M NH4Cl, 0.01 M KHCO3, 0.1 mM EDTA), which was then incubated at room temperature for 7 minutes. 10 mL of FACS buffer (PBS, 2.5% goat serum, 2 mM EDTA) was added and centrifuged at 400g for 15 minutes. The pellet was then resuspended in culture media and transferred to a collagen coated 10 cm tissue culture plate. The culture media was changed the following day. After 2 days, the tissue culture plate was washed once with PBS and 5 mL of PBS is added to the tissue culture plate. The plate is placed at 37˚C and

5% CO2 for ~7-8 minutes after which the mitotic satellite cells are detached from the plate by firmly tapping the plate. The PBS containing the cells is collected, centrifuged, and the cells are transferred in culture media to a new collagen-coated tissue culture plate. This process is repeated every few days until the majority of the cells are myoblasts.

2.6.4 Muscle stem cell isolation

Muscle stem cells were obtained using a similar method described in Sacco et al (Sacco et al., 2008). Briefly, all of the hind limb muscles were dissected from humanely euthanized 8-12 week old mice and placed in 7 mL DMEM containing Type IA collagenase at 628 Units/mL. The muscles were then dissociated using a gentleMACS Dissociator. Tissues were then incubated on a nutating mixer at 37˚C and 5% CO2 for 90 minutes. Dispase II was then added at a concentration of 4.8 U/mL and the muscle was incubated for another 30 minutes on the nutating 61

mixer at 37˚C and 5% CO2. After the incubation, the tissues were passed through a 20 G syringe needle equipped with a 10 mL syringe to fully dissociate the tissue. 7 mL of serum containing media (10% FBS, 90% DMEM) was then added to the dissociated tissue and the entire volume was filtered using a 40 µm cell strainer. The filtered liquid was centrifuged at 400g for 15 minutes. The supernatant was then discarded and the pellet, resuspended in 1 mL of red blood cell lysis buffer, was incubated at room temperature for 7 minutes. 10 mL of FACS buffer was added and the cell slurry centrifuged at 400g for 15 minutes. The pellet was then resuspended in 1 mL of FACS buffer. The cells were incubated with biotinylated Cd31 (1:200, BD, cat. no. 553371), Cd45 (1:500, BD, cat. no. 553078), Cd11b (1:200, BD, cat. no. 553309), and Sca1 (1:200, BD, cat. no. 553334) for 30 minutes on a nutating mixer at 4˚C. After washing the cells and resuspending them in 1 mL FACS buffer, streptavidin microbeads (1:20, Miltenyi Biotec, cat. no. 130-048-101), streptavidin PE-Cy7 (1:200, Life Technologies, cat. no. SA1012), 7 Integrin – PE conjugated (1:500, Ablab, cat. no. 530010-05), and Cd34-eFluor 660 conjugated (1:65, ebioscience, cat. no. 50-0341-82) antibodies were added to the cells. The cells were incubated for 1 hour on the nutating mixer at 4˚C. After the incubation, the cells were once more washed with FACS buffer, and depleted for biotin labeled cells by passing through a magnetic column (Miltenyi Biotec, cat. no. 130-091-051) after resuspension. The cells were washed one final time and resuspended in 1 mL FACS buffer containing Propidium Iodide (Sigma, cat. no. P4864). Muscle stem cells (MuSCs) were then sorted using a BD FACS-Aria II based on 7- AAD-/Cd31-/Cd45-/Cd11b-/Sca1-/Cd34+/α7-Integrin+.

2.6.5 Myogenic cell culture

Muscle stem cells (MuSCs) and primary myoblasts (pMBs) were cultured in tissue culture plates coated with Type 1 rat tail Collagen protein (Life Technologies, cat. no. A1048301). The culture medium was composed of 20% fetal bovine serum (Life Technologies, cat. no. 12483), 79% F- 12, 1% Penicillin-Streptomycin (Life Technologies, cat. no. 11765054) and rh-bFGF (ImmunoTools, cat. no. 11343627) at a final concentration of 2.5 ng/mL. The culture media was changed every other day and the cells were maintained at 37˚C and 5% CO2.

2.6.6 Myogenic cell transplantation

The tibialis anterior muscle of 8-12 week old wild type offspring of crossing C57Bl/6N and actin-eGFP transgenic mice were injured with a single intramuscular injection of BaCl2 (described in animal section). Two days post-injury, muscle stem cells were freshly isolated from the transgenic siblings of the injured mice. The freshly isolated muscle stem cells were then suspended in either saline or 0.75:0.75 HA:MC dissolved in saline.

The injured wild type mice were then anaesthetized in the operating room. The hair was removed from the skin on top of the TA muscle and a small 1” incision was created in the skin to expose the TA muscle. 2 µL of saline or HAMC containing 1.5×103, 5×103, or 10×103 cells was injected into the center of the exposed muscle using a 32G needle and a Hamilton syringe. The skin was then closed and sutured using a 6-0 suture (Covidien, cat. no. SS681). The procedure was performed on both TA muscles of the mice, with one muscle receiving the cells delivered in saline, and the other, cells delivered within HAMC.

3-4 weeks post-transplantation, the mice were euthanized and the TA muscles were harvested and either fixed and frozen for immunohistochemistry or dissociated for flow cytometry analysis.

Similar methods were used to transplant myoblasts, HAMC, and saline into the muscle.

2.6.7 Immunohistochemistry

Post-euthanasia, the TA muscles were extracted and fixed in 0.5% Pierce methanol-free paraformaldehyde (Thermo Fisher, cat. no. 28908) for 2 hours at 4˚ C. The muscles were then transferred to a 20% sucrose (Sigma, cat. no. S9378) solution in ddH2O overnight at 4˚ C. They were then embedded and frozen in OCT (TissueTek, cat. no. 4583) and stored at -80˚ C prior to further processing.

Frozen tissues were sectioned and mounted as 10 µm sections. The tissue sections were rehydrated, blocked, and permeabilized with blocking solution consisting of 20% goat serum (Life Technologies, cat. no. 16210-072), 79% PBS, and 1% Triton-X100 (Bioshop, cat. no. TRX777) for 1 hour at room temperature (RT) or overnight at 4˚C. 1˚ antibody was added to the slides and incubated for 2 hours at RT or overnight at 4˚C. After numerous washes with PBS, 2˚

antibody solutions were added to the slides and incubated for 30 minutes at RT. The slides were then mounted using fluoromount (Sigma, cat. no. F4680) and subsequently imaged.

The primary antibodies used in this project were rabbit anti-GFP (1:500, Invitrogen, cat. no. A11122), rat anti-laminin (1:300, Abcam, cat. no. ab11575), rat anti-Cd68 (1:200, Abcam, ab53444), and rat anti-Ly6G (1:500, Abcam, ab25377). Secondary staining was performed using Alexafluor goat anti-rabbit 488 (1:500, Life Technologies, cat. no. A11008), Alexafluor goat anti-rat 647 (1:500, Life Technologies, cat. no. A21247). The nuclei were visualized using Hoechst 33342 (1:1000, Life Technologies, cat. no. H3570).

2.6.8 Flow cytometry analysis

Flow cytometry analysis was used to enumerate transplanted GFP+ myoblasts 2, 24, and 48 hours post-transplantation into the injured TA. Post-euthanasia, the TA muscles were extracted and placed in 1 mL DMEM containing Type IA collagenase at 628 Units/mL and Dispase II at 4.8 U/mL. The muscle was minced with scissors into small pieces (roughly 2 mm × 2 mm × 2 mm pieces) and then placed on a nutating mixer at 37˚C and 5% CO2 for 120 minutes. After the incubation, the tissues were passed through a 20 G syringe needle equipped with a 1 mL syringe to fully dissociate the tissue. The entire volume was then transferred to a FACS tube equipped with a cell strainer cap (Corning, cat. no. 352235) to further deplete debris. 3 mL of FACS buffer was then added to the sample, which was then centrifuged at 400 g for 15 minutes. The supernatant was discarded and the pellet was resuspended in 0.5 mL FACS buffer. Propidium iodide was added at 1:1000. The sample was then analyzed on a BD-Canto flow cytometer (courtesy of Dr. Peter Zandstra, University of Toronto), to enumerate PI-/GFP+ cells.

For Cd44 expression analysis, a method similar to muscle stem cell isolation protocol was followed. In this case, the cell slurry was incubated with biotinylated Cd31, Cd45, Cd11b, and Sca1 for 60 minutes on the rocking shaker at 4˚C. Either rat anti-mouse Cd44 (1:100, BD, cat. no. 558739) or purified rat IgG1, κ Isotype Control (1:100, BD, cat. no. 559072) was also added in this step. After washing the cells and resuspending them in 1 mL FACS buffer, Streptavidin PE-Cy7, 7 Integrin – PE conjugated, and Cd34-FITC conjugated (1:65, ebioscience, cat. no. 11-0341-82), and Alexafluor goat anti-rat AF647 antibodies were added to the cells. The cells were incubated for 1 hour on the nutating mixer at 4˚C. The cells were washed one final time and 64

resuspended in 1 mL FACS buffer containing propidium iodide (1:1000). The percentage of quiescent or activated MuSCs expressing Cd44 was determined using a BD FACS-Aria II flow cytometer based on 7-AAD-/Cd31-/Cd45-/Cd11b-/Sca1-/Cd34+/7-integrin+/Cd44+.

2.6.9 EdU analysis

Freshly isolated MuSCs were seeded into 96 well plate tissue culture plates in either 0.75:0.75 HAMC or culture media. Cells that had entered cell cycle were visualized using the Click-it EdU kit (Thermo Scientific, cat. no. C10634). The cells were pulsed with 5-ethynyl-2’-deoxyuridine (EdU) for 12 hours, after which they were trypsinized and cytospun (courtesy of Dr. Julie Audet, University of Toronto) onto charged glass slides. The cells were immediately fixed with a 4% paraformaldehyde solution in PBS. The EdU was then visualized using Alexa Fluor 647 Azide, and the nuclei were counter-stained with Hoechst 33342.

2.6.10 Injured skeletal muscle extract preparation

Injured tissue extract was obtained using a modified method described in Chen and Quinn, 1992.

Briefly, the TA of 8-12 week old C57Bl/6N mice were injured with BaCl2. The TA was isolated 48 hours post injury and lysed in a 2 mL tube containing F-12 media and 1X Halt protease inhibitor (Thermofisher, cat. no. 78430). Lysate protein concentration was then measured using a Pierce BCA Protein Assay kit (Thermofisher, 23227). Cell culture media for conditions that required injured skeletal muscle extract were then prepared with a final concentration of 1 mg/mL of the injured tissue extract.

2.6.11 Cell viability assays

Passage 4-10 primary myoblasts were seeded into 96 well plate tissue culture plate in either 0.75:0.75 HAMC or culture media. The Live/Dead® assay (Life Technologies, cat. no. L3224) was used to determine cell viability at desired time points. Calcein AM (1:2000), Ethd-1˚ (1:500), and Hoechst 33342 (1:1000) were used to determine live, dead, and nuclei respectively.

The cells were incubated at 37˚C and 5% CO2 for 15 minutes with the live/dead stain prior to imaging.

The viability of the ejected MuSCs and pMBs was assessed by ejecting 10 µL of the cells using 0.75:0.75 HA:MC or growth media into a collagen coated 384 well plate. The Live/Dead® assay was then used at the desired time points post-ejection to determine cell viability.

2.6.12 Cell ejection efficiency measurement

Ejection efficiency assays were performed by ejecting cells suspended in 0.75:0.75 HA:MC or growth media at 500,000 cells/mL using a 29 G syringe onto a hemocytometer and counting the number of ejected cells.

2.6.13 Dispersion analysis

Dispersion of the GFP+ fibers was determined by calculating the average of the average distance of each GFP+ fiber to all other GFP+ fibers. A higher average would mean the fibers are more dispersed and spread out throughout the tissue. ImageJ analysis and Matlab coding were used to determine dispersion using the following formula:

푛 푛 2 2 ∑푖=1 ∑푗=1 √(푥푖 − 푥푗) + (푦푖 − 푦푗) 퐷푖푠푝푒푟푠푖표푛 = 푛 × (푛 − 1) where n is the number of GFP+ fibers

+ 푥푖 and 푦푖 are the coordinates of the center of the GFP fibers.

2.6.14 Immune response quantification

To quantify the immune response at the injection site during different timepoints, tissue sections were stained for either Ly6G (neutrophils) or Cd68 (macrophages). The nuclei were visualized with Hoechst. The transplantation site for each condition was then imaged using the same microscope settings with a 20x objective. The average fluorescence of Ly6G or CD68 in the entire image was then normalized using the average Hoechst expression to determine a normalized Ly6G (neutrophil) or Cd68 (macrophage) expression, at the site of injection for each transplantation condition.

2.6.15 Imaging and microscope

Tissue sections and cells were visualized using an Olympus IX83 inverted microscope and 10x or 20x objectives. The images were captured with an Olympus DP80 dual CCD color and monochrome camera and CellSens software. Images were adjusted consistently using open source ImageJ software.

2.6.16 Statistical analysis

All experiments were performed with a minimum of three biological replicates. Paired student’s t-test was used in all statistical analysis, unless otherwise specified, with significance set at p < 0.05.

Chapter 3 Small molecule inhibition of EGFR or KDR increases the regenerative potency of the cultured MuSC population

This chapter is a collaboration among the following:

Sadegh Davoudi, Chih-Ying Chin, Dr. Richard Marcellus, and Penney M. Gilbert

Author contributions:

Chih-Ying Chin was involved in the immunohistochemical and drug screen analysis. Dr. Penney Gilbert, and Dr. Richard Marcellus performed the drug screen. All the analysis and other experiments were designed and executed by Sadegh Davoudi. Dr. Penney Gilbert oversaw this work.

Small molecule inhibition of EGFR or KDR increases the regenerative potency of the cultured MuSC population 3.1 Abstract

The remarkable regenerative capacity of adult skeletal muscle is due to the presence of mitotically quiescent resident muscle stem cells (MuSCs). MuSCs ex-vivo culture results in a rapid loss of the regenerative potential of the expanded cell population. This in turn hinders the production of clinically significant numbers of MuSCs for cell-based therapies to restore function and long-term regeneration in aged or diseased skeletal muscle, in addition to presenting difficulties for basic MuSC research. Here, through a small molecule screen, we identify EGFR and KDR as potential targets that can be manipulated to produce a population of myogenic cells with high regenerative capacity in vitro. Through an in vivo transplantation assay, we demonstrated that EGFR or KDR inhibition increased the regenerative capacity of ex-vivo cultured MuSCs. Recipient hindlimb muscles into which cells cultured in the presence of EGFR and KDR inhibitors were injected, contained a larger area of donor-derived fibers compared to control. Further analysis suggests that the observed phenomenon is as a result of large donor fibers and not an increase in the number of donor-derived fibers. Together, our results identify EGFR and KDR as novel targets to increase the regenerative capacity of in vitro cultured MuSCs compared to control.

3.2 Introduction

Muscle stem cells (MuSCs), are resident stem cells in skeletal muscle that play an irreplaceable role in adult skeletal muscle regeneration (Relaix and Zammit, 2012; Sambasivan et al., 2011). MuSCs express the paired box transcription factor Pax7 (Gros et al., 2005; Kassar-Duchossoy, 2005; Relaix et al., 2005; Seale et al., 2000) and are mitotically quiescent in homeostatic muscle (Schultz et al., 1978). Upon activation as a result of tissue damage, MuSCs give rise to a population of transient amplifying cells called primary myoblasts (pMBs) that co-express Pax7, Myf5, and MyoD. pMBs undergo several rounds of division and subsequently downregulate Pax7 and MyoD to terminally differentiate into myocytes. Differentiation-committed myocytes ultimately fuse into preexisting damaged fibers, repairing them in the process, or with one another to form new multi-nucleated muscle fibers (Bentzinger et al., 2010; Dumont et al., 2015a, 2015b; Tedesco et al., 2010).

An important feature of MuSCs is that they harbor stem cell properties, i.e. they can both self- renew to maintain the stem cell pool, as well as give rise to progenitor cells (Collins et al., 2005; Kuang et al., 2007; Sacco et al., 2008; Simons and Clevers, 2011). Upon activation, MuSCs can undergo symmetric or asymmetric divisions. During asymmetric division, MuSCs maintain their population and give rise to a daughter stem cell and a committed progenitor cell. During symmetric division however, the MuSC population is either expanded by creating two daughter stem cells or is depleted by producing two committed progenitor cells.

Many skeletal muscle ailments and dystrophies can be traced back to deficiencies in MuSC fate regulation and function (Judson et al., 2018; Morgan and Zammit, 2010). In Duchenne muscular dystrophy (DMD), the mutations in the dystrophin gene has been demonstrated to result in excessive self-renewal in MuSCs (Dumont et al., 2015c), and the chronic state of regeneration in DMD muscle leads to MuSC exhaustion (Sacco et al., 2010) further exacerbating the dystrophy phenotypes. In aged skeletal muscle, irregular MuSC cell-cycle kinetics (Chakkalakal et al., 2012; Cosgrove et al., 2014) and senescence (García-Prat et al., 2016; Sousa-Victor et al., 2014) are possible reasons for the observed loss in muscle mass (sarcopenia) and regenerative potential.

Skeletal muscle fibers are multinucleated single cells, in which genetic products produced by each of the nuclei is distributed throughout the entire fiber. The current literature suggests that by 70

adding a small fraction of functional myogenic nuclei (either through transplantation of healthy cells (Davies and Grounds, 2007; Rousseau et al., 2010) or through gene editing techniques (Madhavan et al., 2016; Tabebordbar et al., 2016) phenotypes observed in congenital diseases could be ameliorated and functionality restored to the skeletal muscle. Other studies have demonstrated that myogenic cell transplantation can locally restore strength and function to injured and diseased skeletal muscle (Perie et al., 2014; Tremblay et al., 1993).

Numerous studies have demonstrated the regenerative advantage of transplanted MuSCs compared to pMBs. In addition to higher regenerative capacity, transplanted MuSCs are able to repopulate the MuSC niche, allowing for long-term maintenance of skeletal muscle (Cerletti et al., 2008; Cezar and Mooney, 2015; Collins et al., 2005; Crist et al., 2012; Kuang et al., 2007; Sacco et al., 2008). However, ex-vivo expansion of freshly isolated MuSCs results in rapid loss of the highly regenerative MuSCs and yields pMBs instead. In addition to hindering development of cell-based therapies, this loss of function has created a hurdle in in-vitro MuSC research. Therefore, parallel efforts have been in place to both optimize delivery of transplanted cells through biomaterials (Boldrin et al., 2007b; Davoudi et al., 2018; Han et al., 2018; Rossi et al., 2011; Sleep et al., 2017), as well as identifying methods to maintain or increase the regenerative capacity of cultured MuSCs.

Researchers have been able to maintain (to some extent) in-vitro MuSC quiescence by mimicking the in-vivo niche (Quarta et al., 2016) and through small molecule inhibitors (inhibiting eIF2α dephosphorylation; Zismanov et al., 2016). Other studies have demonstrated that by optimizing in-vitro culture conditions, it is not only possible to maintain the regenerative capacity of cultured cells (Gilbert et al., 2010), but restore aged MuSC function using small molecule inhibitors (inhibiting p38 MAPK and Stat3 activity; Bernet et al., 2014; Cosgrove et al., 2014; Tierney et al., 2014). These methods however mainly attempt to ameliorate aging phenotypes or maintain quiescence, and do not solve the problem of limited numbers of highly regenerative MuSC numbers. More recently, researchers have progressed towards this goal through recreating the in-vivo inflammatory response (Fu et al., 2015; Ho et al., 2017) and identifying novel inhibition targets (eIF2α dephosphorylation and Setd7; Judson et al., 2018; Lean et al., 2019). Although these results are encouraging, identification of novel druggable

targets to obtain clinically relevant numbers of MuSCs for transplantation remains a major objective in the field of skeletal muscle research.

In this study, we report novel targets (Egfr, Kdr) that can be manipulated to increase the regenerative capacity of ex-vivo cultured MuSCs. We first assess the effect of a 360-compound library of kinase inhibitors across a 12-point titration curve (doses ranging from 2.6 nM to 10 µM) on the expansion of freshly isolated MuSCs. We report 241 compounds, targeting 139 unique kinases and pathways, that result in a 200% increase in our readout. We identify through an in vivo transplantation assay that inhibiting Egfr or Kdr but not Wee1 kinase, Dgk, Gsk3, Ire1, eEf2k, and Clk1/4 in freshly isolated MuSCs consistently produces a population of cells with greater regenerative capacity when compared to the carrier control treatment. Egfr and Kdr inhibition in cultured cells transplanted in vivo result in a significantly larger area of donor- derived fibers compared to control. Interestingly, in both cases the increased donor-derived fiber area is attributed to the production of large donor fibers rather than to an increase in the absolute number of donor-derived fibers.

3.3 Results 3.3.1 A high-throughput drug screen to identify small molecular inhibitors for myogenic cell expansion

To identify novel pathway targets for ex-vivo expansion of muscle stem cells, in collaboration with Ontario Institute for Cancer Research (OICR), we conducted a 360-compound screen of a kinase inhibitor library (Figure 3.1). One embedded goal of this drug screen was to identify treatments for skeletal muscle maladies that were amenable to systemic delivery. Therefore, identifying drugs that were effective at low doses was favorable. Hence, we tested a 12-point titration curve (0.0026, 0.013, 0.028, 0.0572, 0.1196, 0.2496, 0.524, 1.12, 2.32, 4.8, and 10 µM) of each drug in the screen.

Figure 3.1: Schematic of high-throughput drug screen to identify small molecule inhibitors for MuSC expansion (A) Muscle stem cells (MuSCs) were freshly isolated from 8-12 week old C57Bl/6N mice and seeded in the wells of 384 well plates at a density of 100 cells/well. 1 and 4 days post-seeding, the cells were exposed to 12 different doses (0.0026-10 µM) of a 360-compound drug library. 7 days after seeding the cells, we used an ATPlite luminescence assay kit to determine the cytotoxic and proliferative effects of the drugs on the MuSCs.

We isolated muscle stem cells (MuSCs) from 8-12 week old female C57Bl/6N mice (Sacco et al., 2008) (Figure 2.1A). We seeded the freshly isolated MuSCs into collagen-coated wells of a 384-well plate, at a density of 100 cells/well, and placed the plate in a 37°C incubator overnight. Despite fibronectin being the preferred culture substrate for freshly isolated MuSCs, we used collagen to coat the plates since collagen is the standard ECM protein used in the muscle field (Lukjanenko et al., 2016). Furthermore, by culturing the cells on collagen we would increase the sensitivity of the assay since it is not MuSCs’ preferred ECM protein and the proliferative effect of the tested compounds would not be masked by the normal proliferation of the cells. The following day, we loaded small molecule inhibitors into a Hewlett Packard D300 (HP-D300) digital dispenser. Drug vehicle (DMSO) was loaded onto the HP-D300 as negative control. Additionally, as p38 MAPK inhibition has been previously demonstrated to promote ex-vivo expansion of muscle stem cells (Bernet et al., 2014; Charville et al., 2015; Cosgrove et al., 2014), a p38α/β MAPK inhibitor (SB202190) was used in all experiments as a positive control. Next, we placed the 384-well plate containing the MuSCs into the digital dispenser and the HP-D300 randomly treated the cells in the plate with the loaded drugs at the different doses in a randomized pattern. The dispenser also sprayed DMSO and SB202190 (10 µM) into the negative 73

and positive control wells. A second dose of the drugs was administered using the HP-D300 on the 4th day post-seeding. The outer two rows of wells were filled with water and never used in the screen to avoid edge effects. 7 days after the initial seeding, we used an ATPLite Luminescence Assay kit to determine the cytotoxic/proliferative effect of the small molecule inhibitors on the MuSC populations.

3.3.2 Analysis of drug screen and identification of potential candidates for ex-vivo expansion of muscle stem cells

By monitoring ATP, the ATPLite Luminescence assay indirectly measures the proliferative/cytotoxic effects of the drugs on the cultured MuSCs. Increased ATP readout suggests increased metabolic activity. This could be as a result of increased cell numbers, myogenic cell differentiation (Pala et al., 2018), enhanced metabolic activity, or a combination thereof. Upon obtaining the results of the ATPLite assay from the plate reader, we normalized the readouts for each drug dose compared to the DMSO control (set to 100, Figure 3.3) of that experiment. Drug doses resulting in a 200% increase in ATPlite signal as compared to control were considered as positive hits.

Out of 360 compounds tested in the drug screen, 241 small molecule inhibitors, targeting 139 kinases and cell signaling pathways, resulted in positive hits in at least 1 concentration. 14 different compounds inhibiting the p38 MAPK signaling pathway were among the positive hits, confirming the sensitivity of our assay in detecting myogenic cell expansion. 144 of the drugs caused a 200% in the ATPLite readout in at least 2 consecutive doses. A summary of the drug screen results (hit compounds, their targets, and successful doses) is summarized in Appendix I.

The identified positive compounds often inhibited multiple targets. Additionally, overlapping targets existed between the drugs. As an example, 25 different compounds target PDGFR-β. Therefore, we only considered compounds that were predicted to target a single pathway to avoid manipulating multiple pathways. To narrow down our drug candidate pool, we focused on small molecule inhibitors inhibiting pathways that were either not thoroughly studied in MuSCs, or ambiguity surrounded the role of those specific pathways. Also, we attempted to avoid targets that might be controversial, i.e. were generally known to be positive and negative regulators of cell proliferation and apoptosis, respectively. For example, Cdk2 is required for myoblast cell

cycle progression and its inhibition, using small molecules (Tintignac et al., 2000) or senescence (Chakravarthy et al., 2000), leads to cell cycle arrest. However, in our screen results, inhibition of Cdk2 activity by 2 different compounds (Kenpaullone and SU 9516) resulted in a higher readout.

3.3.2.1 Inhibiting EGFR in cultured MuSCs expands cell numbers by reducing the time between divisions

One of most common targets among the compounds identified as positive hits in the drug screen was epidermal growth factor receptor (Egfr). 24 different compounds targeting Egfr lead to greater than 200% increase in the ATPLite assay readout. At the time of our analysis, previous research had demonstrated that while Egfr is absent in quiescent MuSCs, it can be detected in all MuSCs as early as 24 hours post-activation (Golding et al., 2007). Despite these early studies, the exact function of Egfr in MuSCs had not yet been identified. Given the high number of ‘hit’ compounds targeting this receptor, we were interested in Egfr’s role in regulating MuSC function. Among the positive hits inhibiting Egfr, we chose a compound that specifically targeted this receptor: Erlotinib Hydrochloride (E-HCl). E-HCl is an ATP-competitive inhibitor, which prevents the formation of Egfr homodimers, required for Egfr signaling (Raymond et al., 2000). We then decided to use clonal analysis to determine how inhibiting Egfr in cultured MuSCs using E-HCl leads to cell expansion at a single cell level.

To that end, we fabricated PEG hydrogel microwells (500 µm in diameter) at the bottom of the wells of a 24 well plate. We then functionalized the microwells with laminin. Next, we sorted fresh MuSCs from C57Bl/6N mice and seeded 1200 cells/well into the wells of the 24 well plate, so that the microwells would each be occupied by 1 MuSC. We then exposed each well of the 24 well plate to either Egfr inhibitor (Erlotinib Hydrochloride) or carrier control (DMSO). Beginning at 35 hours post-seeding (at which point the cells would begin dividing for the first time), we used time-lapse microscopy to track the sorted MuSCs for a period of 5 days (Figure 3.2A). We then analyzed the images using the Baxter algorithm to extract information on the fate of each tracked cell (Figure 3.2B): time of first division, time between divisions, time of death, number of cells after 7 days in culture.

Figure 3.2: Clonal analysis of ex-vivo cultured MuSCs in the presence of EGFR inhibitor reveals reduced time between divisions resulting in higher number of cells after 7 days. (A) Schematic of MuSC clonal analysis. Briefly, PEG microwells with a diameter of 500 µm were fabricated in 24 well plates and functionalized with laminin. Freshly isolated MuSCs from C57Bl/6N mice were seeded into the 24 well plate at 1200 cells/well so each microwell was occupied by an average of 1 MuSC. The cells in each well were exposed to either DMSO control or Egfr inhibitor (Erlotinib Hydrochloride; E-HCl). 35 hours after plating, the cells were tracked using time-lapse microscopy for 5 days. The time-lapse images were then analyzed using the Baxter algorithm to determine individual cell fate. (B) Representative image of Baxter algorithm analysis of the fate of a cell and its progeny in one microwell. (C) Graph depicting the time between divisions of the analyzed cells. (D) Scatter plot of the number of progenies derived from each MuSC after 7 days of culture in either E-HCl or DMSO control. Error bars indicate SEM. Statistical significance determined by unpaired student’s t-test where; p < 0.05.

We tracked a total of 40 and 44 cells exposed to DMSO control or E-HCl, respectively. Our analysis revealed that although there was no significant difference in the time of first division between the cells exposed to E-HCl or DMSO control, the time between subsequent divisions 76

was significantly reduced in cells exposed to Egfr inhibitor (Figure 3.2C). Consequently, at the end of 7 days, the MuSCs exposed to E-HCl had given rise to a significantly higher number of progenies compared to control MuSCs (Figure 3.2D).

3.3.2.2 Identification of potential candidates for increasing regenerative capacity of cultured MuSCs

Upon performing the first clonal analysis, we learned that another research group was currently investigating the role of Egfr in MuSC fate decisions. To avoid any potential conflict of interests and overlapping research, we decided to go back to our high-throughput drug screen and use that dataset to identify new druggable targets to increase the regenerative capacity of cultured MuSCs compared to control. Given the high number of positive hits, we ultimately used the following guidelines to choose our candidates:

• Kinases and compounds that had successful hits at sub-µM doses (Clk1/2/4, eEf2k, Dgk, Wee1 kinase, Ire1). • Kinases that were of biological interest (Gsk3) • Kinases that were targets of multiple positive compounds and of biological interest (Egfr, Kdr)

Figure 3.3: Dose-response curves of small molecule inhibitors selected for in-vivo functional assays

(A) Normalized ATPlite luminescence assay readout for the selected small molecule inhibitors at different doses. Bottom (- - -) and top (- ∙ ∙ -) dashed lines represent DMSO and 200% cell growth, respectively.

The dose response curves of the selected drugs are displayed in Figure 3.3. The list of chosen compounds, their targets, and the selected dose is presented in Table 3.1. Of the identified compounds, the Egfr, Clk1/4, Gsk3, and eEf2k inhibitors are ATP-competitive inhibitors. MK- 1775 inhibits Wee1 kinase activity by preventing the phosphorylation of Cdc2 at Tyr15 (Cdc2Y15), a direct substrate of Wee1 kinase. We were unable to obtain the exact targets of the remaining inhibitors (Kdr, Ire1, and Dgk inhibitors) from the available literature.

Table 3.1: Small molecule inhibitors selected from high-throughput drug screen Drug name Target Target dose SB203580 P38 MAPK (+ control) 10 µM Erlotinib hydrochloride Egfr 5 µM ZM 323881 hydrochloride Kdr (Vegfr-2) 250 nM TG 003 Clk1, Clk4, Clk2 200 nM MK 1775 (Adavosertib) Wee1 kinase 250 nM BIO Gsk-3α/β 2 µM NH125 eEf2k 60 nM Dioctanoylglycol Dgk (Dagk) 500 nM ASC-033 Ire1 500 nM

3.3.3 Transplantation assay reveals Egfr and Kdr inhibition as methods to increase regenerative capacity of in-vitro cultured MuSCs

Previous studies highlight the higher regenerative potential of freshly isolated MuSCs compared to cultured MuSCs and pMBs (Cerletti et al., 2008; Cezar and Mooney, 2015; Collins et al., 2005; Crist et al., 2012; Kuang et al., 2007; Sacco et al., 2008). We therefore decided to perform an in-vivo transplantation assay to determine whether our selected compounds (Table 3.1) led to an increase in the number of highly regenerative MuSCs or the less potent pMBs. Prior to those experiments, we first needed to design an experimental setup to discern the different engraftment potential of myogenic cells in various stages of myogenesis.

We designed the transplantation model used in Chapter 2 (HAMC) to determine the effectivity of HAMC as a cell delivery vehicle. In that model, we transplanted MuSCs with HAMC/saline into the injured TA muscle of immunocompetent mice (Figure 2.1.B). We used immunocompetent wild-type mice to observe the effect of HAMC on MuSC transplantation outcomes in the presence of an intact immune system. Furthermore, we injured but did not irradiate the TA muscle to allow the endogenous MuSCs to participate in regeneration and compete with the donor MuSCs, as they would do in a human patient. In summary, we designed the model to examine HAMC’s ability compared to saline, in improving MuSC transplantation outcomes in conditions as similar to human clinical trials as possible. However, this model was not suitable to compare the regenerative capacity of two populations of cells. The immune system and regeneration from the endogenous MuSCs will reduce donor cell contribution to regeneration and reduce the sensitivity of our assay. Therefore, we used a separate transplantation model in this chapter to empower the transplanted cells. Therefore, we performed a series of pilot transplantations in a separate injury model comparing the regenerative capacity of early passage (P4) pMBs and freshly isolated MuSCs (Figure 3.4A). We irradiated the hindlimbs of 8-12 week old recipient NOD scid gamma (NSG) mice with 12 Gy of γ-rays to provide a functional advantage for our transplanted cells compared to the recipient MuSCs. 24 hours after irradiation, we injured the tibialis anterior (TA) muscles of the NSG mice with a single dose of BaCl2 injected intramuscularly to create a regenerative environment. 48 hours after BaCl2 injury, we injected freshly isolated MuSCs (103 or1.5×103 cells) or P4 pMBs (103 or 5×103 cells) derived from transgenic mice (expressing GFP under the control of β-actin), into the injured TA muscles of the recipient mice. 4 weeks after transplantation, we humanely euthanized the mice, and harvested the TA muscles. We next cryo-sectioned, and immunostained the harvested TAs to visualize the GFP+ donor-derived fibers. Representative tiled immunostained images of the TA sections is presented in Figure 3.4B, in which donor-derived fibers are visualized in green.

Analyzing the total GFP+ area (sum of the cross-sectional area of all GFP+ fibers) and GFP+ fiber number revealed the higher regenerative capacity of freshly isolated MuSCs compared to P4 pMBs (Figure 3.4C). Transplanting 103 or 1.5×103 MuSCs yielded both higher total GFP+ area (Figure 3.4C, left) as well as GFP+ numbers (Figure 3.4C, right) compared to 5×103 pMBs. We were unable to identify any donor-derived fibers in the muscle transplanted with 103 pMBs.

Finally, despite detecting a relatively high number of donor-derived fibers from transplanting 103 MuSCs, increasing the number of transplanted cells to 1.5×103 cells increased the donor-derived fiber number and total area, without saturating the entire muscle. Therefore, in our hands 103 fresh MuSCs is a reasonable number of cells to begin the experiments with, which allows for the observation of both increased and reduced contribution to regeneration from the transplanted cells. These pilot experiments confirmed that our experimental setup can distinguish between the regenerative capacity of cell across the myogenic lineage.

Figure 3.4: Freshly isolated MuSCs have a higher regenerative capacity compared to primary myoblasts (pMBs) (A) Schematic of in-vivo transplantation assay. The hindlimbs of recipient NSG mice were exposed to a 12 Gy dose of γ-irradiation. 24-hours following irradiation, the tibialis anterior (TA) muscles of the NSG mice were injured with a single intramuscular injection of BaCl2. 48-hours following injury, freshly isolated muscle stem cells (MuSCs) or passage 4 (P4) primary myoblasts (pMBs) from transgenic mice ubiquitously expressing Green Fluorescent Protein (GFP) were injected into the irradiated and injured TA of the NSG mice. The TA muscles were harvested 4 weeks post-transplantation and analyzed to determine the contribution of the 81

transplanted cells to muscle regeneration. (B) Representative tiled fluorescent image of a transverse section from tibialis anterior (TA) muscle that was harvested and immunostained for GFP (green) and laminin (red) 4 weeks after transplantation with 103 GFP+ MuSCs (left) or 5×103 GFP+ P4 pMBs (right). Scale bar, 500 nm. (C) Plot portraying the total GFP+ fiber area (left, sum of the GFP+ fibers’ area) and GFP+ fiber number (right) of TA muscles injected with freshly isolated MuSCs and pMBs.

Upon optimizing our experimental setup, we performed transplantation assays in mice to assess the ability of the selected small molecule inhibitors in expanding myogenic cells while maintaining their regenerative capacity (Figure 3.5A). We isolated MuSCs from transgenic mice expressing GFP under the control of β-actin (Figure 2.1A). We seeded the freshly isolated GFP+ MuSCs into collagen-coated wells of 96-well plates, at a concentration of 1000 cells/well, and added the selected compounds (Table 3.1) to the cell growth media in each well. We cultured the cells for a period of 7 days, while refreshing the drugs every other day by replacing half of the media with freshly prepared growth media containing the selected inhibitors. On the 4th day of culture, we exposed the hindlimbs of the recipient NSG mice to 12 Gy γ-irradiation. 24 hours after the irradiation (5th day of cell culture), we injured the tibialis anterior (TA) muscles of the th NSG mice through intramuscular injection of BaCl2. 2 days after the BaCl2 injection (7 day of culture), we resuspended all the cells in each well in saline and transplanted them intramuscularly into the irradiated, and injured TA of the NSG mice. 4 weeks after transplantation, we humanely euthanized the recipient mice, and harvested the TA muscles. The isolated TAs were sectioned and immunostained to visualize the contribution of GFP+ donor cells to the regeneration process (Figure 3.5A). Representative images of the immunostaining results for each condition is presented in Figure 3.6A.

Figure 3.5: Schematic of in-vivo transplantation assay to assess effectivity of selected small molecule inhibitors in expanding the myogenic cells with increased regenerative capacity (A) Schematic of in-vivo transplantation assay. Freshly isolated muscle stem cells (MuSCs) from transgenic mice ubiquitously expressing Green Fluorescent Protein (GFP) were seeded into the 96 well plates at a density of 1000 cells/well and treated with the selected small-molecule inhibitors. Half of the media was replaced every other day over a period of 7 days. On the 4th day post-seeding, the hindlimbs of the recipient NSG mice were exposed to a 12 Gy dose of γ- irradiation. 24-hours following irradiation, the tibialis anterior (TA) muscles of the NSG mice were injured with a single intramuscular injection of BaCl2. 48-hours following injury, all the cells in each well of the 96-well plate were collected and injected into the irradiated and injured TA of the NSG mice. The TA muscles were harvested 4 weeks post-transplantation and analyzed to determine the contribution of the transplanted cells to muscle regeneration.

Figure 3.6: Representative images of in-vivo transplantation assay results (A) Representative tiled fluorescent image of a transverse section from a tibialis anterior muscle that was harvested and immunostained 4 weeks after transplantation with GFP+ (green) myogenic cells cultured for 7 days in the presence of small molecule inhibitors. The periphery of the muscle fibers is visualized through the laminin (red) in the basal lamina. Scale bar: 200 µm. 83

To assess the regenerative capacity of the transplanted cells, we analyzed the immunostained muscle sections to determine the total GFP+ area in each section (Figure 3.7A). During our analysis, we noticed that the regenerative capacity of the donor cells (total GFP+ areas of our negative control) would vary between experiments (Figure 3.7A Ctrl). As such, we paired each experimental result with the control from the same donor, visualized through distinct symbols (Figure 3.7A). Analyzing the muscle sections revealed a significant increase in the total GFP+ area of donor cells cultured in the presence of p38 (positive control, 81%, Figure 3.7B), Egfr (113%, Figure 3.7C), and Kdr (105%, Figure 3.7D) inhibitors compared to DMSO control. Inhibiting Wee1 kinase, Dgk, Gsk3α/β, Ire1, eEf2k, and Clk1/2/4 however did not significantly affect the engraftment of donor cells (Figure 3.7E-J). Cells exposed to p38, Egfr, and Kdr inhibitors consistently outperformed their paired control cells (Figure 3.7B-D) and cells cultured with eEf2k, Wee1 kinase, and Dgk inhibitors had a lower engraftment rate compared to their controls. Interestingly however, cells cultured with Clk, Gsk3α/β, and Ire1 inhibitors had mixed effects on the cultured cells, leading to both higher and lower engraftment rates compared to negative control.

Figure 3.7: Inhibiting EGFR and KDR activity increases the regenerative capacity of MuSCs cultured for 7 days. (A) Scatter plot portraying the total GFP+ fiber area (sum of the GFP+ fibers’ area) of tibialis anterior (TA) muscles injected with cells cultured in the presence of inhibitors. Paired experiments have identical symbols. (B-J) Bar graphs indicating the average total GFP+ area from each TA muscle injected with cells treated with p38i (B, positive control, n = 6), Egfri (C, n = 4), Kdri (D, n = 4), Wee1i (E, n = 4), Dgki (F, n = 5), Gsk3i (G, n = 4), Ire1i (H, n = 4), eEf2ki (I, n = 4), Clki (J, n = 4), and their respective controls (DMSO). Each paired dataset (n) represents data analyzed from transplanting control or drug treated MuSCs into 2 separate TA muscles. Paired experiments are indicated by the same symbols. Graphs are replotted from (A). Error bars indicate SEM. Statistical significance determined by ratio paired student’s t-test where; p < 0.05.

Next, we analyzed the number of donor-derived GFP+ fibers in each condition (Figure 3.8A). Similar to total GFP+ area analysis, the number of GFP+ fibers varied greatly between different experiments (Figure 3.8A Ctrl). Therefore, we used the same symbols as the total GFP+ area analysis (Figure 3.7) to present paired experimental data. We identified that despite significantly higher total GFP+ area upon inhibiting p38, Egfr, and Kdr in cultured MuSCs (Figure 3.7B-D), only p38 inhibition led to a significantly higher number of GFP+ fibers (66%, Figure 3.8B-D). Additionally, the number of GFP+ fibers in the Clk inhibitor dataset were significantly lower than their paired controls (34%, Figure 3.8J), despite having similar total GFP+ areas (Figure 3.7J). There were no significant changes in the number of GFP+ fibers in the rest of the conditions (Figure 3.8E-I). However, we observed an opposite trend between the GFP+ fiber number and total GFP+ area in muscles receiving cells exposed to Gsk3α/β inhibitors (Figure 3.8G, Figure 3.7G).

To dissect the discrepancy between total GFP+ area and GFP+ fiber numbers, we plotted the cross-sectional area (CSA) distribution of the GFP+ fibers for each experimental condition (Figure 3.9). We excluded experimental conditions that yielded less than 15 donor-derived (GFP+) fibers to prevent artificial skewing of the data in any direction. Donor-derived fiber CSA distribution of cells exposed to p38 inhibitor closely follows those of the control (Figure 3.9A). Interestingly, larger donor-derived fibers are observed in muscles which received MuSCs cultured with Egfr inhibitors (Figure 3.9B). Despite not being statistically significant, engraftment of GFP+ MuSCs exposed to Kdr, Gsk3α/β, and Clk inhibitors appeared to lead to larger donor fibers as well (Figure 3.9C,F,I). And finally, as expected, the CSA distribution of Wee1, Dgk, Ire1, and eEf2k inhibitor conditions follow those of their DMSO controls (Figure 3.9D-E,G-H).

Figure 3.8: EGFR and KDR inhibition of cultured MuSCs does not lead to changes in the number of GFP+ fibers. (A) Scatter plot portraying the total number of GFP+ fibers of tibialis anterior (TA) muscles injected with cells cultured in the presence of inhibitors. Paired experiments have identical symbols. (B-J) Bar graphs indicating the average number of GFP+ fibers from each TA muscle injected with cells treated with p38i (B, positive control, n = 6), Egfri (C, n = 4), Kdri (D, n = 4), Wee1i (E, n = 4), Dgki (F, n = 5), Gsk3i (G, n = 4), Ire1i (H, n = 4), eEf2ki (I, n = 4), Clki (J, n = 4), and their respective controls (DMSO). Each paired dataset (n) represents data analyzed from transplanting control or drug treated MuSCs into 2 separate TA muscles. Paired experiments are indicated by the same symbols. Graphs are replotted from (A). Error bars indicate SEM. Statistical significance determined by ratio paired student’s t-test where; p < 0.05.

Figure 3.9:GFP+ fiber diameter analysis suggests a trend towards fiber hypertrophy in fibers formed from MuSCs exposed to EGFR, KDR inhibitors. (A-I) Histograms displaying donor-derived GFP+ fiber cross-sectional areas (CSA) comparing fibers derived from cells cultured in the presence of p38i (A, n = 6), Egfri (B, n = 3), Kdri (C, n = 3), Wee1i (D, n = 3), Dgki (E, n = 3), Gsk3i (F, n = 2), Ire1i (G, n = 3), eEf2ki (H, n = 3), Clki (I, n = 3) or DMSO control (datasets are paired with their DMSO controls). Each paired dataset (n) represents data analyzed from transplanting control or drug treated MuSCs into 2 separate TA muscles. Statistical significance determined by student’s t-test where; p < 0.05.

3.4 Discussion

Here, we provide evidence that inhibition of Egfr or Kdr mitigates, to a certain extent, the loss of in-vivo regenerative capacity of cultured MuSCs as a result of ex-vivo culture. We generated a screening data set in which we treated murine MuSCs with compounds in a kinase inhibitor library and identified drugs which were able to expand the myogenic cell population. Based on the results of the drug screen, we selected 8 compounds targeting pathways whose role in MuSC expansion had not been previously studied. We then used an in-vivo transplantation assay to identify the compounds which were successful in increasing the regenerative capacity of ex-vivo 88

cultured MuSCs. These experiments identify Egfr and Kdr as novel inhibition targets to increase the regenerative potency of ex-vivo cultured MuSCs.

To provide sufficient numbers of cells with high regenerative capacity for cell-based therapies, as well as enabling in vitro manipulation and study of MuSCs, we sought to identify novel compounds for ex-vivo expansion of MuSCs through high-throughput screening of a comprehensive library of pharmacological compounds (Figure 3.1A). The readout of the assay (ATPLite Luminescence) measured the metabolic activity of the cells within each well, which we used as a proxy for cell number. Additionally, in anticipation that the identified pathways might be potential systemic treatments for muscle diseases, we administered the drugs as a 12 point-titration ranging from 2.6 nM to 10 µM. We were able to identify 144 compounds that were able to induce a minimum of 200% increase in our readout compared to the vehicle control (Appendix I).

Our screen was able to identify compounds that had previously been shown to induce MuSC self-renewal and expansion, validating its ability to detect compounds capable of ex-vivo expansion of MuSCs. p38 MAPK is one of the most well-studied pathways in satellite cells and inhibiting it can expand MuSCs as well as rejuvenate aged ones (Bernet et al., 2014; Charville et al., 2015; Cosgrove et al., 2014). 14 compounds targeting p38 MAPK were identified as positive hits in our screen.

However, our screen also identified targets that were questionable. p-eIf2α was recently demonstrated to be required for MuSC quiescence, and pharmacological inhibition of eIf2α dephosphorylation leads to in-vitro expansion of MuSCs (Lean et al., 2019; Zismanov et al., 2016). Pkr-like endoplasmic reticulum (ER) kinase (Perk) is believed to be the kinase responsible for eIF2α phosphorylation in MuSCs. Protein kinase R (Pkr) is another kinase responsible for eIF2α phosphorylation (Lemaire et al., 2008), although its role in eIF2α phosphorylation of MuSCs has not been previously investigated. Two compounds, one targeting Perk and the other targeting Pkr, came up as positive hits in our drug screen. This might suggest that promoting eIF2α dephosphorylation (through Pkr and Perk inhibition) will result in a higher number of cells by promoting the activation of MuSCs and expansion of committed progenitor cells. Alternatively, the higher readout could simply be as a result of the higher metabolic

activity of committed progenitor cells compared to quiescent MuSCs and cells that are not fully committed to differentiation (Pala et al., 2018).

We therefore decided to use an in vivo transplantation assay to determine the myogenic potential of the expanded cells in the drug screen. Our hypothesis was that a transplantation assay would be successful in distinguishing cells with a high and low regenerative potential, even despite numerical advantage in favor of committed progenitors (Figure 3.4A). We confirmed our hypothesis through our pilot experiment (Figure 3.4B-C) and determined that 1000 freshly isolated MuSCs would be a suitable starting number for our experiments. Transplantation of 1000 MuSCs would generate many donor-derived GFP+ fibers without saturating the TA muscle, and we would be able to detect both increases and decreases in donor-derived fibers as a result of increased/decreased regenerative capacity of the cultured cells.

Next, we identified 8 compounds and their optimal dose (Figure 3.3, Table 3.1) for the in-vivo transplantation assay (Figure 3.5A), targeting epidermal growth factor receptor (Egfr), vascular endothelial growth factor receptor (Vegfr2, Kdr), Wee1 kinase (Wee1), Glycogen synthase kinase 3 (Gsk3), eukaryotic elongation factor-2 kinase (eEf2k), Diacylglycerol kinases (Dagk, Dgk), CDC like kinase 1/4 (Clk1/4), and Inositol-requiring enzyme 1 (Ire1). All selected compounds resulted in at least 200% increase in the screen readout. If the compounds were successful in either increasing or maintaining the regenerative capacity of the cultured cells compared to the control, we expected to observe a significant increase in donor cell contribution compared. We found that transplanting cells cultured with Egfr and Kdr inhibitors led to a 113% and 105% increase in the donor-derived fiber area compared to control cells, respectively (Figure 3.6, Figure 3.7C-D).

Egfr has been shown to be involved in a variety of cellular functions including proliferation, migration, differentiation, survival, and apoptosis (Holbro and Hynes, 2004). Its role in MuSCs however has not been thoroughly studied. Previous studies revealed that although Egfr is not detected in quiescent satellite cells, it is expressed in all satellite cells 24 hours after activation. It was also suggested that Egfr activation can prevent apoptosis in serum-free conditions (Golding et al., 2007). A recently published paper demonstrated the role of Egfr signaling in asymmetric MuSC division. Inhibiting Egfr resulted in increased symmetric divisions, and Egf treatment led

to an increase in asymmetric divisions in the satellite cells (Wang et al., 2019). The researchers also provided evidence that Egfr signaling in MuSCs was through Aurka activity. Interestingly, Aurka was the target of 5 compounds that were designated as successful hits in our drug screen as well. An independent study showed that inhibiting Stat3, which is downstream of activated Egfr (Gao et al., 2007), also results in increased MuSC self-renewal and expansion (Tierney et al., 2014). Our observations with regards to Egfr inhibition in MuSCs align with these published studies.

Kdr is not detected in muscle fibers or satellite cells of healthy muscle and is expressed in vascular endothelial cells. Kdr is however observed in newly formed murine fibers and myoblasts in the later stages of regeneration (Arsic et al., 2004; Verma et al., 2018). Interestingly, 27 compounds targeting Kdr were identified as positive hits in our screen, the highest successful target hit. The in-vivo functional assay confirms that the cells cultured with Kdr inhibitor have a higher regenerative capacity. Based on the manufacturer’s notes, ZM323881 selectively inhibits Kdr activity over Flt1 (Vegfr-1, receptor with higher affinity for Vegfa compared to Kdr) and a range of other receptor tyrosine kinases such as Pdgfrβ, Fgfr1, Egfr. Therefore, although it is unlikely that the observed effects are as a result of off-target effects of the Kdr inhibitor, the selectivity of the compound should be investigated in independent experiments. Based on the available information, our speculation is that either Kdr is expressed at undetectable levels in MuSCs and activated myoblasts, or inhibiting Kdr is reversing activated myoblasts into a less differentiated state with a higher regenerative potential. Further studies are required to identify the exact mechanism of action behind the increased myogenic capacity of cells exposed to Kdr inhibitor.

Another interesting observation from our transplantation assay was that the increase in donor- derived fiber area in our Egfr and Kdr inhibitor datasets did not accompany a higher number of donor derived fibers (Figure 3.8C-D). This contrasted with cells cultured with our positive control drug (p38 inhibitor) in which both donor-derived total area and fiber number increased by 88% and 66% compared to control, respectively. Additionally, despite having similar total GFP+ fiber area, cells exposed to Clk1/4 inhibitor resulted in a smaller number of GFP+ fibers (Figure 3.8J). To dissect these discrepancies, we plotted the cross-sectional area (CSA) distribution of the donor derived fibers in each condition (Figure 3.9). The CSA distribution of 91

donor-derived fibers from cells cultured in the presence of p38 inhibitor matched those of their control (Figure 3.9A). However, GFP+ fibers derived from cells cultured with EGFR inhibitor appeared to be larger than their DMSO counterparts (Figure 3.9B). Donor-derived fibers from the Kdr and Clk1/4 inhibited cells displayed a trend similar to that of the Egfr inhibitor condition but none of the differences were statistically significant (Figure 3.9C, I). Given that increased cell fusion has been demonstrated to lead to fiber hypertrophy (Rosenblatt et al., 1994), a possible explanation could be that culturing the fresh MuSCs in the presence of the Egfr and Kdr inhibitors results generates more cells compared to the carrier control which then leads to the observed increase in donor-derived fiber diameters. However, further studies are required to determine the exact reasons behind the observed phenomenon.

Together, our results demonstrate that culturing freshly isolated MuSCs in the presence of Egfr and Kdr inhibitors, compared to control, increases their ability to engraft in-vivo. The cultured MuSCs have a higher regenerative capacity with respect to their DMSO counterparts; however, further studies are required to determine the regenerative capacity of the expanded cells compared to freshly isolated MuSCs. Additionally, functional assays are required to determine whether the increased engraftment also leads to increased force generation. Furthermore, the observed trend towards larger donor-derived fibers of muscles injected with Egfr, Kdr, and Clk1/4 inhibited cells hints towards novel treatments for muscle wasting and atrophy. Finally, the presented drug screen results (with a wide range of concentrations) provides a useful resource towards identifying other targets for both in vitro studies of murine MuSCs.

3.5 Limitations and future work

In this chapter, we demonstrate that inhibiting Egfr or Kdr but not Wee1 kinase, Dgk, Clk1/4, Ire1, eEf2k in freshly isolated MuSCs, generates cells with greater myogenic potential compared to the carrier control treatment. However, our work does not explore the number and the myogenic identity (based on Pax7, MyoD, and Myog expression) of the generated cells. Future work is required, especially with regards to the Egfr and Kdr inhibitor conditions, to determine how the inhibitors affected the cultured MuSCs. Additionally, our analysis does not investigate whether the increased engraftment also translates to increased muscle strength and function.

Furthermore, the in vivo transplantation assays suggest that the donor-derived fibers in muscles transplanted with Egfr, Kdr, and Clk1/4 were larger compared to their respective controls. Elucidating how the transplanted cells contribute to this observed phenomenon is of interest. Additionally, the potential of the identified inhibitors to be modified and used as systemic treatments should be investigated in future studies.

Finally, while the design of our drug screen enabled it as a high throughput analysis to analyze 260 drugs over 12 doses, the output of the assay was a metabolic reading of the cultured cells, with no direct information on the number and myogenic identity of the cells. We categorized compounds increasing the ATPLite readout by 200% as ‘hits’. A recently published study performed a drug screen using a similar drug library used in our study. The readout of their assay was fold change in the number of satellite cells compared to carrier control. Due to the complexity of the experiments, they only used 1 dose per drug (1 µM). They identified 29 compounds, to increase the number of satellite cells in their assay by 50%. We had identified 8 out of the 29 compounds as successful hits in our screen as well. BIO used for Gsk3 inhibition in our in vivo transplantation assay was one of these 8 compounds. Furthermore, out of the remaining 21 compounds, 7 of them increased our readout by 100% compared to the carrier control. This suggests that by reducing the threshold we use to identify our positive hits, we can potentially identify further compounds that are able to increase the regenerative capacity of cultured MuSCs.

3.6 Materials and methods

3.6.1 Animals

Animals were housed in the Division of Comparative Medicine (DCM) of University of Toronto. Animal protocols were reviewed and approved by the DCM at University of Toronto. C57Bl6/JNOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) and C57BL/6-Tg(CAG-EGFP)1Osb/J (Actin-eGFP, expressing eGFP under the control of beta-actin promoter) mice used in this project were purchased from The Jackson Laboratory.

The Actin-eGFP and the NSG mice were used at 6-12 and 8-12 weeks old, respectively. Injuries to the tibialis anterior (TA) of the NSG mice were performed under anesthesia (2.5%

Isofluorane) by injecting 30 µL of 1.2% BaCl2 (Bio Basic, cat. no. BC2020) in ddH2O into the center of the TA muscle of the mice. Irradiation was performed by first anaesthetizing the NSG mice with an intra-peritoneal (IP) injection of 100 mg/kg Ketamine (Narketan 10, Vetoquinol) and 10 mg/kg Xylazine (Rompun, Bayer). Once the mice were immobilized, their hindlimbs were exposed to 12 Gy irradiation via a Gammacell 40 Extractor.

3.6.2 Murine muscle stem cell isolation

Murine muscle stem cells (MuSCs) were isolated from skeletal muscle using methods similar to the protocol described in Sacco et al., 2008. The hindlimbs of euthanized 6-12-week-old mice were dissected and transferred to DMEM containing Type 1A collagenase (Sigma, cat. no. C9891) at 628 Units/ml. The muscle was then physically dissociated using a gentleMACS Dissociator (Miltenyi Biotec, cat. no. 130-093-235) and placed on a nutating mixer at 37° C for 1 hour. Dispase II was then added at a concentration of 0.04 U/mL and the muscle was incubated for another 60 minutes on the nutating mixer at 37˚C. After 1 hour, the enzymatically digested muscle was fully broken down by slowly passing through a 20 G needle attached to a 10 mL syringe 5-10 times, after which 7 ml of wash medium (10% FBS, 90% DMEM) was added to the digested muscle. The entire mixture was passed through a 70 µm (Miltenyi Biotec, cat. no. 130- 098-462) cell strainer, and subsequently filtered with a 40 µm cell strainer (Corning, cat. no. 352340). The filtered cell slurry was then centrifuged at 400 g for 15 minutes and the supernatant was aspirated. The red blood cells (RBCs) were lysed by resuspending the pellet in RBC lysis buffer (0.155 M NH4Cl, 0.01 M KHCO3, 0.1 mM EDTA) and incubating at room temperature for 7 minutes. 10 ml FACS buffer ((PBS, 2.5% goat serum, 2 mM EDTA) was added to the cell suspension and centrifuged for 15 minutes at 400g. The supernatant was aspirated, and the pellet was resuspended in 1 mL FACS buffer. The cells were then incubated at 4° C for 1 hour with the following antibodies: Cd34-eFluor 660 (1:65, ebioscience, cat. no. 50-0341-82), Itga7-PE (1:500, AbLab, cat. no. 530010-05), and biotinylated Cd31 (1:200, BD, cat. no. 553371), Cd45 (1:500, BD, cat. no. 553078), Cd11b (1:200, BD, cat. no. 553309), and Sca1 (1:200, BD, cat. no. 553334). The cells were then washed and resuspended in 1 mL FACS buffer. Next, the cells were incubated with Streptavidin-PE-Cy7 (1:200, Life Technologies, cat. no. SA1012) for 30 minutes at 4° C. Following the incubation, the cells were washed one last time and Propidium

Iodide was added to the cell suspension. murine MuSCs were isolated via FACS based on the following gating strategy: PI-/Cd31-/Cd45-/Cd11b−/Sca1-/Cd34+/Itga7+.

3.6.3 Murine primary myoblast isolation

We established primary myoblast (pMB) lines using methods similar to ones described in Rando and Blau, 1994. Briefly, we followed the methods described for MuSC isolation (section 3.6.2) up to the RBC lysis. The RBC lysis buffer was washed with 10 ml washing media. The pellet was then resuspended in myogenic cell culture media and transferred to a collagen coated 10 cm dish. The culture media was changed the next day. After 2 days, we washed the tissue culture plate once with PBS, added 5 mL of fresh PBS to the tissue culture plate, and incubated at 37°C for ~7-8 minutes. The mitotic satellite cells are then detached from the plate by firmly tapping the plate. The PBS containing the cells is then collected, centrifuged, and transferred into a new collagen-coated tissue culture plate. This process is repeated every few days until a stable myoblast line is produced.

3.6.4 High-throughput drug screen

Freshly isolated MuSCs (section 3.6.2) were seeded at a density of 100 cells/well into 384 well plates and placed at 37°C and 5% CO2 overnight. The following day, small molecule inhibitors from the OICR drug library were loaded into a Hewlett Packard D300 (HP-D300) digital dispenser. The plate was then placed into the dispenser and the cells were randomly treated with the drugs at various doses and placed back into the incubator. 3 days later, the cells were treated with another dose of the drugs. On the 7th day of culture, an ATPLite luminescence assay kit (Perkin Elmer, cat. no. 6016941) was used to determine the cytotoxic/proliferative effect of the various doses of the inhibitors on the cultured MuSCs. The entire drug library was tested over the course of 15 experiments (24 drugs/experiment).

3.6.5 Myogenic cell culture

The base media used to culture the cells was 79% F-12, 20% fetal bovine serum (Life Technologies, cat. no. 12483), and 1% Penicillin-Streptomycin (Life Technologies, cat. no. 11765054), and 2.5 ng/mL rh-bFGF (ImmunoTools, cat. no. 11343627). The following inhibitors were used in this study: p38 inhibitor (10µM, SB203580, New England biolabs, cat. no. 5633),

Kdr inhibitor (250 nM, ZM323881, R&D Systems, cat. no. 2475), Dgk inhibitor (500 nM, Dioctanoylglycol, Tocris, cat. no. 0484), Wee1 kinase inhibitor (250 nM, MK-1775, Selleckchem, cat. no. S1525), eEf2k inhibitor (60 nM, NH125, Sigma Aldrich, cat. no. N2162), Gsk3 inhibitor (2 µM, BIO, Sigma Aldrich, cat. no. B1686), Clk inhibitor (200 nM, TG003, Sigma Aldrich, cat. no. T5575), Egfr inhibitor (5 µM, Erlotinib Hydrochloride, Santa Cruz Biotech., cat. no. sc-202154), Ire1 inhibitor (500 nM, ASC-033, courtesy of Ontario Institute for Cancer Research). DMSO (Sigma Aldrich, cat. no. D2650) was used as the negative control.

Freshly isolated murine MuSCs were cultured on collagen coated tissue culture plastic. The culture media was prepared by adding the desired concentration of inhibitors to freshly prepared base medium. The culture media was changed every other day by replacing 50% of the media in the wells with fresh media.

3.6.6 Myogenic cell transplantation

To assess the effectivity of selected inhibitors on cultured MuSCs, 1000 freshly isolated murine MuSCs were seeded into collagen coated wells of 96 well plates. The cells were treated with culture media containing inhibitors for 7 days. To prepare the cells for transplantation, all the cells were removed from the well on the day 7 using Accutase (Life Technologies, cat. no. A1110501), washed, and resuspended in ~10 µL 0.1% BSA (Bioshop, cat. no. ALB001) in PBS. For transplanting freshly isolated MuSCs, the desired number of cells were suspended in ~10 µL 0.1% BSA. Cell suspensions were maintained on ice until transplantation.

3 days prior to transplantation, the hindlimbs of the NSG mice were irradiated. 2 days prior transplantation, the TA of the host mice were injured via BaCl2 injection. Detailed descriptions of the irradiation and BaCl2 injury procedures are provided in section 3.6.1. Throughout the transplantation procedure, the irradiated and injured mice were anaesthetized via isofluorane inhalation. Once the mice were anaesthetized, the skin directly above the TA was opened to expose the TA. Cells were then slowly transplanted into the center of the TA via a 32 G needle (Hamilton, cat. no. 7803-04) and a 25 µL Hamilton Syringe (Hamilton, cat. no. 7654-01). The skin was then sutured using a 6-0 suture (Covidien, cat. no. SS681).

4 weeks after transplantation, the mice were humanely euthanized, and their TA was harvested. The harvested muscles were fixed in 0.5% paraformaldehyde (ThermoFisher, cat. no. 28908) for 2 hours at 4° C. The muscles were then transferred to a 20% sucrose (Sigma Aldrich, cat. no. S9378) solution in ddH2O and stored overnight at 4° C. They were then embedded in OCT (TissueTek, cat. no. 4583), and frozen by being immersed in liquid nitrogen cooled 2- Methylbutane (Sigma, cat. no. 270342). The frozen muscles were then stored at -80° C.

3.6.7 Immunohistochemistry

For immunohistochemistry (IHC) analysis, the frozen TAs were sectioned using a cryostat and mounted as 10µM sections onto charged glass slides. The sections were rehydrated, blocked, and permeabilized using blocking solution: 20% goat serum (Life Technologies, cat. no. 16210-072), 79% PBS, and 1% Triton-X100 (Bioshop, cat. no. TRX777). The sections were then stained with the primary antibody for 2 hours at room temperature, or overnight at 4° C. The samples were then washed 8x with blocking solution and stained with the 2° antibodies for 30 minutes at room temperature. Sections were prepared for imaging by washing and mounting them using fluoromount (Sigma, cat. no. F4680).

The following antibodies were used in this project: rat anti-laminin (1:300, Abcam, cat. no. ab11575), rabbit anti-GFP (1:500, Invitrogen, cat. no. A11122), Alexafluor goat anti-rat 647 (1:500, Life Technologies, cat. no. A21247), and Alexafluor goat anti-rabbit 488 (1:500, Life Technologies, cat. no. A11008). Hoechst 33342 (1:1000, Life Technologies, cat. no. H3570) was used to visualize nuclei.

3.6.8 Imaging and microscope

Tissue sections were imaged using an Olympus IX83 inverted microscope and a 10x objective. The images were capture using CellSens software and an Olympus DP80 dual CCD color and monochrome camera. Images were adjusted consistently using open source ImageJ software.

3.6.9 Statistical analysis

All experiments were completed with a minimum of three biological replicates. Ratio paired and unpaired student’s t-test was used in all statistical analysis. Statistical significance was determined at p < 0.05. 97

Chapter 4 Single-cell RNA sequencing of mono-nuclear cells in murine skeletal muscle

This chapter is a collaboration among the following:

Sadegh Davoudi, Brendan Innes, Alaura Androschuk, Chih-Ying Chin, Gary Bader, and Penney M. Gilbert

Author contributions:

BI extracted the receptor ligand interactions from the scRNA seq. AA performed the immunocytochemistry and immunohistochemical validation of the Schwann cell markers. CYC assisted in data visualization. All other experiments were designed, executed, analyzed by SD. GB and PMG oversaw this work.

Single-cell RNA sequencing of mono-nuclear cells in murine skeletal muscle 4.1 Abstract

Skeletal muscle’s exceptional regenerative capacity is a coordinated response by a variety of muscle resident cells including but not limited to muscle stem cells (MuSCs), fibroadipogenic progenitors (FAPs), and endothelial cells (ECs). Inter-cellular signaling between the cells in the skeletal muscle niche plays an essential role in both muscle homeostasis and timely muscle regeneration. Uncovering this complex cellular interactome is crucial to understanding the mechanisms behind healthy skeletal muscle. Here, using single cell RNA sequencing, we identify 9 populations of mononuclear muscle-resident cells: MuSCs, 2 EC populations, 3 FAP clusters, Schwann cells, Tenocytes, and B cells. We characterize the 2 distinct populations of endothelial cells in addition to uncovering a novel marker for direct Schwann cell isolation from skeletal muscle. Further, using the transcriptional signature of the identified cell clusters, we construct a dense cellular interactome among the identified populations. Our study provides a unique insight into the various cell populations present in adult skeletal muscle and is a useful resource for studying the interplay of cells within homeostatic muscle.

4.2 Introduction

Skeletal muscle has a remarkable regenerative capacity, in which muscle stem cells (MuSCs) play an indispensable role (Mauro, 1961; Relaix and Zammit, 2012). Upon injury, Pax7 expressing MuSCs exit quiescence, and give rise to a population of transient amplifying cells called primary myoblasts (pMBs) expressing the myogenic regulatory factors Myf5 and MyoD (Relaix et al., 2005; Schultz et al., 1978; Seale et al., 2000). pMBs undergo several rounds of division and ultimately either fuse with one another to create new myofibers, or merge into damage pre-existing fibers and repair them (Bentzinger et al., 2010; Dumont et al., 2015a, 2015b; Tedesco et al., 2010).

The majority of skeletal muscle regeneration research to date has been focused on MuSCs and their role in myogenesis (reviewed by Relaix and Zammit, 2012), making MuSCs amongst the best understood adult stem cells. Studied cell populations are isolated using major cell markers through FACS, which results in loss of heterogeneity and rare subpopulations of the sorted cell type. Bulk RNA sequencing and microarray analysis has often been used to provide a detailed description of the transcriptional signature of MuSCs and their progenitors of myogenic cells in various stages of myogenesis (Aguilar et al., 2016; Alonso-Martin et al., 2016; Liu et al., 2013; Machado et al., 2017; Pala et al., 2018; Pallafacchina et al., 2010; Sousa-Victor et al., 2014; van Velthoven et al., 2017). However, bulk RNA sequencing and microarray analysis of isolated cells only provides an average representation of the identified cell type, and results in loss of information through merging the transcriptome of the isolated cells.

Research in the past decade has revealed that extrinsic signaling from a variety of muscle interstitial and immune cells is essential for the successful progression of MuSCs from quiescence to new myofiber formation, as well as MuSC self-renewal. Fibroadipogenic progenitors (Joe et al., 2010; Uezumi et al., 2010), endothelial cells (Christov et al., 2007; Latroche et al., 2017) and macrophages (Arnold et al., 2007; Tidball, 2005) are a subset of cells that have been shown to play an important role in the timely regeneration of skeletal muscle. Additionally, signaling between the different muscle-resident cells has also been implicated in muscle homeostasis. Dll4-Notch signaling between endothelial cells (ECs) and MuSCs has recently been shown to promote MuSC quiescence in skeletal muscle (Verma et al., 2018). And

100

skeletal muscle fibers presumably secrete factors preventing adipogenic progression of fibroadipogenic progenitors (FAPs; Uezumi et al., 2010). Finally, a growing body of literature suggests various ECM proteins deposited by muscle-resident cells are implicated in maintaining MuSC quiescence and promoting timely regeneration (Baghdadi et al., 2018b; Ishii et al., 2018; Lukjanenko et al., 2016; Urciuolo et al., 2013). Therefore, understanding how muscle resident cells communicate with one another and the signals they provide for each other is of utmost importance in understanding the molecular mechanisms behind muscle homeostasis and skeletal muscle regeneration. However, similar issues to MuSC studies exist in other muscle-resident cells, where isolation of cells using specific markers and bulk analyses leads to loss of heterogeneity and a generalized overview of the studied population.

Single cell RNA sequencing (sc-RNA seq) is an increasingly available technology that can provide transcriptomic information of a population of cells at a single cell level (Tang et al., 2009; Zheng et al., 2017). Through an unbiased approach, sc-RNA seq can identify cell populations with similar transcriptional signatures, while maintaining single cell resolution. Previous studies have highlighted the potential of sc-RNA seq in resolving the cellular heterogeneity and characterizing the transcriptome of the cellular components of various tissues such as the heart (Skelly et al., 2018), pancreas (Enge et al., 2017), and liver (MacParland et al., 2018).

In this report, we use sc-RNA seq technology to analyze and through an unbiased approach, identify the mononuclear cell populations present in homeostatic adult skeletal muscle. We identify a novel subset of endothelial cells, in addition to identifying a novel marker for direct Schwann cell isolation from skeletal muscle. And finally, using the transcriptional profile of each identified population, we predict the complex cell-cell communication network present in adult skeletal muscle.

4.3 Results 4.3.1 Single cell RNA sequencing of mononuclear cells in murine adult skeletal muscle

We performed sc-RNA seq to create a high-resolution map of the non-myofiber cells in adult skeletal muscle. We harvested the tibialis anterior (TA) muscle of an 8 week old C57Bl/6N 101

mouse and enzymatically digested the muscle into a cell slurry. We used fluorescence-activated cell sorting (FACS) to acquire a single cell suspension of live (PI-) metabolically active (Calcein AM+) mononuclear cells. We used the 10X Chromium platform for library preparation, which we then sequenced using an Illumina HiSeq 2000. We used the R Seurat package (Butler et al., 2018) to analyze the sequencing results (Figure 4.1).

Figure 4.1: Schematic of experimental procedure to obtain mono-nuclear cell isolation for single cell RNA sequencing (A) The tibialis anterior muscle of an adult C57Bl/6N mouse was harvested and enzymatically digested. Fluorescence-activated cell sorting (FACS) was used to isolate the live (PI-) and metabolically active (Calcein+) cells from the cell slurry. The isolated single cells were then loaded into a 10X chromium system for library preparation. The library was sequenced using an Illumina HiSeq 2000 sequencer. The raw data was processed using Cell Ranger 2.2.0 to obtain the gene expression matrix of the cells and Seurat 2.3.4 R package was used for the remaining analysis.

Our analysis revealed that 2039 cells with an average expression of 1198 genes/cell passed quality control and pre-set filters. We then used principal component analysis (PCA) dimensionality reduction and unsupervised clustering tools from the Seurat package to group the cells into distinct clusters based on their transcriptional identity. We visualized the results using t-distributed stochastic neighbor embedding (t-SNE) and determined the identities of the identified clusters based on commonly known markers of major cell populations (Figure 4.2A- B). Our analysis revealed 8 distinct clusters comprised of 5 known cell types (Figure 4.2B): FAP clusters expressing Pdgfrα (Joe et al., 2010), EC clusters expressing Cdh5 (Breviario et al., 1995) and Pecam1 (Newman et al., 1990), MuSCs expressing Pax7 (Gros et al., 2005; Kassar- Duchossoy, 2005; Relaix et al., 2005; Seale et al., 2000), Schwann cells expressing Plp1

102

(Doerflinger et al., 2003), and Tenocytes expressing Tnmd (Docheva et al., 2005; Giordani et al., 2018). Our analysis also revealed the presence of B cells (expressing Cd79a, data not shown), however we decided to focus on the non-immune cells for the remainder of this study.

Figure 4.2: scRNA-seq analysis identifies major cell populations in skeletal muscle (A) Unsupervised clustering of single cells from scRNA-seq analysis reveals major cell populations (5 cell types, 8 distinct clusters). t-Stochastic neighbor embedding (t-SNE) dimensionality reduction was used to create a 2D plot for visualization. Each dot represents a single cell, and cells within a cluster are represented by identical colors. (B) Heatmap depicting relative expression of known cell markers used to identify the major cell populations. Gene expression values are normalized to the values of the cluster with the highest expression of each gene.

103

Figure 4.3: Dot plot of top 5 differentially expressed genes in each identified cluster (A) Top 5 differentially expressed genes in each identified cluster was determined in an unbiased fashion to identify cluster biomarkers. Biomarkers of each cluster were identified in comparison to all other cells. The size of each circle depicts the percentage of cells within that cluster expressing the gene, and the color depicts the normalized gene expression within each cluster compared to the cluster with the highest expression of that gene.

We next identified genes which were differentially expressed in each cluster (Figure 4.3, Table A.3). To determine the genes that are specifically expressed by each cluster, the transcriptome of that cluster is compared to the transcriptome of all other cells. As such, where only one cluster exists of a cell type (MuSC, Schwann cells, Tenocytes), common identifiers of those cells are found in the top 5 differentially expressed genes of that cluster. Pax7, and Myf5, commonly used to identify MuSCs, Plp1 and Sox10 used for Schwann cell identification, and Tnmd which is a hallmark of Tenocytes are all among the top differentially expressed genes of their respective clusters. However, in the EC and FAP clusters, we cannot expect to see traditional EC and FAP markers in their significantly expressed genes, since the other clusters will be expressing them as well. Instead, genes that are more specifically expressed in each of the subclusters are identified which can be potentially used to determine their distinct roles and processes.

4.3.2 Characterization of 2 identified endothelial cell populations

Our sc-RNA seq analysis of adult skeletal muscle identified 2 cell populations expressing common EC markers: Cdh5 and Pecam1 (Figure 4.2B, Figure 4.4B-C). We next attempted to 104

distinguish the two EC populations based on their transcriptional profile, using markers for various endothelial cell subtypes. Lyve1, a lymphatic EC marker (Kong et al., 2017), was not detected in either EC sub-population (Figure 4.4E). Ephb4, which can be used to detect venous ECs (Paz and Amore, 2014), was expressed in both EC-1 and EC-2 subpopulations (Figure 4.4F). Additionally, arterial EC markers Kdr (Figure 4.4D) and Dll4 (Figure 4.4G) were detected in both EC subpopulations as well. Interestingly, despite being expressed in both clusters, Kdr and Dll4 were differentially expressed in the EC-1 and EC-2 clusters, respectively. Finally, Ng2 and Pdgfrβ, common pericyte markers, were not detected in either of the EC subclusters either (Figure 4.4H-I). We therefore concluded that both EC-1 and EC-2 subpopulations are vascular endothelial cells.

Given that both EC-1 and EC-2 subpopulations were established to be vascular ECs, we next decided to separate them using FACS based on cluster specific surface markers. To this end, we compared the gene expression of EC-1 and EC-2 with each other to obtain the differentially expressed genes in each cluster (Table A.4). It is important to note that the results of this gene comparison is different than the one previously performed (Figure 4.3) as the transcriptional profile of EC-1 and EC-2 is only being compared to one another, and not all of the cells in the scRNA seq analysis. Additionally, in order to identify the uniquely differentially expressed genes, we identified the top 15 marker genes in each cluster that was detected in less than 25% of the cells in the other cluster (Figure 4.5A), excluding genes such as Kdr and Dll4.

We next identified genes in each cluster whose translational product would be expressed at the cell surface: Kcna5 (Snyders, 1993) and Scarb1 (Zanoni et al., 2016) in the EC-1, and Alpl (Orimo, 2010; Sultana et al., 2013) in the EC-2 clusters. We proceeded to use FACS to separate endothelial cells based their EC-1 and EC-2 cells. We harvested and enzymatically digested the TA muscles of adult C57Bl/6N mice. However, we were unable to detect Alpl, Kcna5, or Scarb1 on the surface of PI-/Cd31+ cells in the skeletal muscle digest (Figure 4.5B).

105

Figure 4.4: The two identified endothelial cell clusters cannot be distinguished using venous/arterial or lymphatic/vascular endothelial cell or pericyte markers (A) t-SNE plot depicting the 2 identified endothelial cell clusters. (B-I) Subplots presenting expression of vascular (B, Cdh5; C, Pecam1; D, Kdr), lymphatic (E, Lyve1), venous (F, Ephb4), and arterial (G, Dll4) endothelial, and pericyte (H, Ng2; I, Pdgfrβ) cell markers within cells of the 2 endothelial cell clusters.

106

Figure 4.5: Alpl, Kcna5, and Scarb1 are not suitable markers for separating the two endothelial sub-populations (A) Dot plot identifying differentially expressed genes distinguishing the two endothelial cell populations. In addition to being differentially expressed in one cluster, the genes are expressed in less than 25% of the cells in the other cluster. (B) Flow cytometry plots displaying expression of Alpl, Kcna5, and Scarb1 in PI-/Cd31+ cells from digested skeletal muscle.

107

During our analysis of differentially expressed genes in all of the cell clusters identified in the sc- RNA seq analysis, we noticed that the Abcg2 gene, a member of the ATP-binding cassette (ABC) transporter superfamily, was differentially expressed in our EC-1 subpopulation (Figure 4.6A-B, Table A.3). Abcg2 is also identified as a differentially expressed gene when comparing EC-1 and EC-2 clusters together (Table A.4). Previous reports have identified that ~22% of mononuclear cells in skeletal muscle are Abcg2+ and upon injury, this population gives rise to a large number of vascular associated cells (Cd31+ and Cspg4+; Doyle et al., 2011). The EC-1 cluster identified in our scRNA seq analysis is composed of 401 cells and makes up ~19.7% of the total analyzed cells. We therefore speculate that the EC-1 subpopulation is the same Abcg2+ cells identified in previous reports that give rise to endothelial cell progeny upon muscle injury.

Figure 4.6: Abcg2, muscle side-population marker, is differentially expressed in the EC - 1 cluster (A) Violin plot depicting the expression of the Abcg2 in the clusters of adult skeletal muscle. (B) Gene expression gradient of Abcg2 within analyzed cells.

4.3.3 Gpr37l1 as a marker for Schwann cell isolation directly from skeletal muscle

Schwann cells are the main glial cell in the peripheral nervous system (PNS) and are responsible for forming myelin in the nerves of the PNS (Fields, 2010). Additionally, upon injury, in addition to providing various growth factors, the basal lamina of Schwann cells provides a conduit for axonal regrowth (Gu et al., 2015). As a result, Schwann cells play a crucial role in 108

nerve regeneration. To study Schwann cells, researchers typically isolate them by enzymatically digesting nerves and purifying for p75+ cells. p75 however, is not a suitable candidate for direct Schwann cell isolation from digested skeletal muscle, since p75 expression is not unique to Schwann cells and can be found on other cell types, such as MuSCs (Olguín and Pisconti, 2012). Additionally, common transgenic models used for Schwann cell isolation (Sox2, Dhh, Cadm3, Cadm4) are either via genes that are transcription factors or based on our scRNA seq data, not unique to Schwann cells (Figure 4.7A). As such, to our knowledge, suitable surface markers for direct isolation of Schwann cells from digested skeletal muscle have not been previously reported.

To identify surface markers for Schwann cell isolation, we first extracted the top differentially expressed genes in the Schwann cell cluster compared to other cell types (Figure 4.7B). Next, we identified 2 surface markers that based on the differential gene expression analysis, are uniquely expressed in Schwann cells: Kcna1, Gpr37l1. We first assessed the presence of the predicted surface markers in the IMS32 Schwann cell line. Immunocytochemical analysis revealed that both Kcna1 and Gpr37l1 are expressed in IMS32 cell (Figure 4.7C). We then asked whether we can detect Gpr37l1 and Kcna1 on mononuclear cells of skeletal muscle. To that end, we harvested skeletal muscle from the hindlimbs of adult C5Bl/6N mice. Upon enzymatic digestion, we stained the cell suspension for Gp37l1 and Kcna1. Through flow cytometry, we were able to observe live (PI-) cells expressing Gpr37l1 (~3.3% of PI- events, Figure 4.7E), but not Kcna1 (Figure 4.7F). Finally, to obtain positional information with regards to Gpr37l1+ cells in skeletal muscle, we analyzed cryosections of tibialis anterior (TA) muscles from C57Bl/6N mice. We found several Gpr37l1-rich regions in the muscle section which appear consistent with nerve bundles. Based on our results, we believe Gpr37l1 can be used as a novel marker for direct Schwann cell isolation from skeletal muscle.

109

Figure 4.7: Gpr37l1 can be used as a marker to isolate Schwann cells from dissociated skeletal muscle (A) Dot plot displaying expression of common genes used in murine transgenic models used for Schwann cell isolation, in skeletal muscle scRNA seq clusters (B) Dot plot presenting top 10 differentially expressed genes in the Schwann cell cluster in comparison to other clusters. Expression values are normalized to gene levels in cluster with highest expression. (C) Immunocytochemistry images of known (Sox10, Necl1), and proposed (Gpr37l1, Kcna1) Schwann cell markers in the IMS32 Schwann cell line. Scale bar: 50 µm. (D-E) Flow cytometry plots presenting the expression of the Kcna1 (D) and Gpr37l1 (E) in PI- cells from digested skeletal muscle. (F) Representative immunofluorescent image of Gpr37l1 (red) and Sox10 (blue) co-staining in tibialis anterior (TA) muscle section. Scale bar: 100 µm.

4.3.4 Cellular interactome in adult skeletal muscle

Skeletal muscle is composed a variety of different cell types. Based on our scRNA seq analysis, MuSCs, ECs, FAPs, Schwann cells, and tenocytes are some of the major cell populations present in homeostatic muscle. To elucidate the cellular interactome in skeletal muscle, we crossed the transcriptome of the 8 identified cell clusters in our scRNA seq analysis with a curated reference 110

database of receptor-ligand interactions (Bader; Kirouac et al., 2010; Qiao et al., 2014; Ramilowski et al., 2015; Rieckmann et al., 2017; Yuzwa et al., 2016). The reference receptor- ligand interaction database contained 2593 unique proteins and 38,446 unique interactions. The resulting interactome between the cell populations in skeletal muscle, reveals a dense intercellular communication network present in skeletal muscle (Figure 4.8A).

In order to better understand the cellular interactome in which each cell cluster is connected to all other clusters, we broke down the communication network into 8 distinct graphs in which the ligands presented to each identified cell cluster from all other clusters are presented individually (Figure 4.8B). It appears the FAPs and tenocytes are the most trophic cells, presenting the most ligands to other cell populations. Interestingly, FAPs appear to be the main source for many of the growth factors required by the ECs, neuronal cells, and even myogenic cells (Figure 4.9A). These growth factors include but are not limited to Vegfa (Lohela et al., 2009), Igf-1 (Bach, 2015; Duan et al., 2010), Angpt1 (Papapetropoulos et al., 2000), Ngf (Glebova, 2004), and Fgf2 (Bikfalvi et al., 1997).

111

Figure 4.8: Putative signaling between major skeletal muscle cell populations (A) Predicted receptor-ligand signaling between the major cell types in skeletal muscle. Lines depict ligands presented by the cluster of the same color, for which the receiving cluster is expressing cognate receptors. The line width is correlated with the number of receptor-ligand interactions. Loops indicate autocrine signaling. (B) Detailed view of ligands presented to each

112

cluster by other cell populations. The number of ligands presented by each cluster is indicated next to the source.

Figure 4.9: FAPs are the source of key growth factors in skeletal muscle (A) Heatmap depicting expression of key growth factors in skeletal muscle in different clusters.

We next proceeded to uncover the potential interactions of MuSCs in a healthy quiescent niche. To that end, we extracted signals provided to the MuSCs from the interactome network (Figure 4.8B, top far left). Many of the ligands presented to MuSCs have more than one cell source, and many of the interactions are not unique. The interactome predicted 785 interactions between MuSCs and other clusters (200 unique interactions, 98 presented ligands and 24 cognate receptors on MuSCs). We then created a heatmap presenting the cellular source of the presented ligands to the MuSCs (Figure 4.10).

Several ligands and ECM proteins, previously proven to be required for MuSCs quiescence and proper muscle homeostasis, are identified in the MuSC interactome. As an example, Dll4 presented by ECs (Verma et al., 2018), which has been previously reported to be a positive regulator of MuSC quiescence through Notch receptors (Bjornson et al., 2012; Philippos et al., 2012), is correctly identified in the MuSC interactome. ColV, another ECM protein that is essential for the quiescent niche (Baghdadi et al., 2018b) is also detected in the cell interaction network. Successful endogenous repair in skeletal muscle has been linked to ColVI (Urciuolo et al., 2013; Zou et al., 2008) and Fn1 (Lukjanenko et al., 2016) expression, both of which are identified in our MuSC interactome as well. These results confirm that the constructed cellular

113

interactome can serve as a resource for the identification of novel ligands and ECM proteins that play an essential role in both MuSC quiescence and a healthy skeletal muscle niche.

114

Figure 4.10: Cellular source of ligands presented to MuSCs in homeostatic skeletal muscle

115

(A) Heatmap depicting the cellular source of ligands for which MuSCs express cognate receptors for. Gene expression values are normalized to the values of the cluster with the highest expression of each gene. Red arrows indicate ligands (Dll4) and ECM proteins (ColV, ColVI, Fn1) that have previously been shown to play an essential role in MuSC quiescence and healthy niche. 4.4 Discussion

In this study, we have used sc-RNA seq to identify the various mono-nuclear cell populations present in adult skeletal muscle and determine the molecular signature of each population. Our analysis revealed 9 distinct cell clusters of 6 major cell types: MuSCs, 3 FAP clusters, 2 EC clusters, Tenocytes, Schwann cells, and B cells. In addition to identifying novel biomarkers for Schwann cell isolation from skeletal muscle, our analysis is among the first to detect various subpopulations among the muscle-resident endothelial and fibroadipogenic progenitor cells using sc-RNA seq. Finally, we were able to report for the first time, the predicted intercellular communication network present among muscle-resident cells.

Our scRNA seq analysis of 2039 cells with an average expression of 1198 genes/cell is similar to recent reports of muscle scRNA seq analysis (Dell’Orso et al., 2019; Giordani et al., 2018; Schaum et al., 2018). We identified 6 major cell populations in our analysis (Figure 4.2): MuSCs, FAPs, ECs, Schwann cells, Tenocytes, and B cells. However, our study is the first to report subpopulations of endothelial and fibroadipogenic progenitor cells, through transcriptional analysis of these cells at a single cell level. Through identifying the differentially expressed genes for each of the identified clusters, our dataset presents as a valuable resource for the study of the various cells present in homeostatic adult skeletal muscle.

The major cell types identified in our analysis were similar to recently published sc-RNA seq analyses of skeletal muscle (Dell’Orso et al., 2019; Giordani et al., 2018). In addition to MuSCs, ECs, FAPs, Schwann cells, and B cells, we were able to identify tenocytes in our analysis, a cell type recently characterized through sc-RNA seq (Giordani et al., 2018). However, smooth muscle cells (SMCs), macrophages, and T cells do not form distinct clusters in our analysis. This is possibly due to the difference in the number of analyzed cells, as well as variations in the cell isolation methods in the different studies.

116

The interactions between various cell types through autocrine and/or paracrine signaling has rarely been studied in homeostatic skeletal muscle. In the studies that have studied the effect of one cell type on another in skeletal muscle, either the direct interaction has not been studied, or researchers have mainly focused on the interaction between only 2 cell types, often through in vitro studies (reviewed in Wosczyna and Rando, 2018). By analyzing the ligands and receptors expressed by the identified clusters, we predicted the comprehensive cell-cell communication network in adult skeletal muscle (Figure 4.8A). The identified interactome reveals potential inter-cellular signaling among all the cell populations present in the muscle. The FAPs and tenocytes appear to be the most trophic cells present in skeletal muscle compared to the other cell types. Additionally, many of the essential growth factors such as x, y, z, are expressed at high levels in the FAPs. The tenocytes on the other hand express large amounts of ECM transcripts. Overall, this is a key resource providing an overview of the potential signaling that occurs among the various cells in homeostatic skeletal muscle. It is important to note that our current model is a prediction of the intercellular signaling in skeletal muscle and does not consider anatomical barriers, post-transcriptional regulation of the expressed genes, and the spatiotemporal resolution of cells within adult skeletal muscle. Additionally, mature skeletal muscle fibers are also an important source for various ECM proteins and growth factors and are not represented in the cellular interactome.

Focusing on MuSC signaling, we extracted the ligands presented to MuSCs from the interactome (Figure 4.10). The presence of known regulators of MuSC quiescence and muscle regeneration (Dll4, ColVI, Fn1, and ColV) among the presented ligands affirmed the notion that the derived cellular interactome can predict existing cell-cell interactions necessary for proper skeletal muscle homeostasis and regeneration. As an example, the Notch signaling pathway has been previously demonstrated to promote MuSC quiescence (Bjornson et al., 2012; Philippos et al., 2012). Due to the anatomical proximity of ECs and MuSCs (Christov et al., 2007), Egfl7 produced by ECs and a Notch antagonist (Müller-Esterl et al., 2009) might also play a role in MuSC quiescence. Other examples of potential regulators of MuSC fate are basal lamina components such as Lama4, Lama2, Lamb2 that are identified as ligands presented to MuSCs and might be crucial for MuSC quiescence. Additionally, interesting observations can be made from the ligands presented to MuSCs, discerning new roles for the cells in skeletal muscle. For

117

example, the 3 FAP subpopulations present many similar ligands as the tenocytes to the MuSCs, whereas there are many factors uniquely secreted by tenocytes. Given the recent research identifying tenocytes (Giordani et al., 2018), this data provides potential insight into the role of tenocytes as potentially unique modifiers of the ECM environment.

Through our scRNA seq analysis, we were able to identify 2 clusters of transcriptionally distinct endothelial cells (Pecam1+, Figure 4.4). It appears EC-2 might be a more trophic source of MuSC ligands compared to EC-1, including the MuSC quiescence factor Dll4. Previous studies have reported the presence of various endothelial subpopulations in skeletal muscle including lymphatic (Lyve1+) vs vascular, or venous (EphB4+) vs arterial (Dll4+) endothelial cells (Ieronimakis et al., 2008; Wardrop and Dominov, 2011). Further analysis revealed that no pericytes markers were detected among the clusters and that both EC-1 and EC-2 populations were in fact a mixture of arterial and venous vascular endothelial cells (Figure 4.4). Tissue nonspecific alkaline phosphatase (Alpl) is the only form of alkaline phosphatase (AP) present in skeletal muscle and is solely expressed in pericytes and a small subset of endothelial cells (Dellavalle et al., 2011). Although we were not able to detect Alpl at the cell surface of freshly isolated ECs using flow cytometry, the EC-2 cluster significantly expressed the Alpl gene compared to EC-1 (Figure 4.5). The exact function of Alpl in endothelial cells is not yet determined and perhaps the findings of this study can be a used to determine the role of Alpl and this subset of endothelial cells. Another differentially expressed gene between the EC-1 and EC- 2 populations is Abcg2+. Side-population (SP) cells are a subset of cells found in various tissues including but not limited to muscle (Meeson et al., 2004; Uezumi et al., 2006), bone marrow (Goodell et al., 1996, 1997), and myocardium (Hierlihy et al., 2002; Martin et al., 2004; Oyama et al., 2007; Pfister et al., 2005, 2008), and are enriched for stem cell activity. A unique characteristic of SP cells is their ability to efflux Hoechst 33342 through Abcg2. The majority of SP cells have been shown to express Cd31, are considered to be a subpopulation of ECs, and participate in vascular regeneration (Uezumi et al., 2006). However, SP cells make up less than 1% of the cells in skeletal muscle (Meeson et al., 2004) whereas the identified EC-1 cluster is 19.7% of the total analyzed cells. As a result, the SP cells cannot account for the identity of the EC-1 subpopulation. Abcg2 has been shown to be expressed in ~22% of the mononuclear cells in skeletal muscle, similar to our results. Upon injury, the number of Abcg2+ cells increase, and

118

they give rise to the majority of Cd31+/Sca1+ endothelial cells. Interestingly, vascular regeneration is unperturbed in injured Abcg2-/- mice but muscle regeneration is impeded as a result of impaired immune response (Doyle et al., 2011). Based on the literature and our scRNA seq analysis, we speculate that EC-2 is the small subset of ECs that has been demonstrated to express Alpl in vivo. Furthermore, we speculate that the Abcg2+ EC-1 subpopulation are the endothelial cells that normally give rise to endothelial cell progeny upon injury. However, although a distinguishing factor between the EC-1 and EC-2 populations, Abcg2 appears to be dispensable for vascular regeneration and does not play a key role in endothelial cell function.

Using our scRNA seq data, we were able to identify a novel surface marker for Schwann cell isolation. We were able to detect Gpr37l1 in a small population of mononuclear cells in skeletal muscle. Additionally, IHC analysis of TA sections revealed colocalization of Gpr37l1+ areas with nerve bundles. This provides as a useful marker to isolate Schwann cells directly from digested skeletal muscle, enabling the study of the peripheral nervous system in regions far from the main nerves.

Together, our data presents a novel perspective of the various cells in homeostatic skeletal muscle and how they communicate with one another. Through scRNA seq analysis we uncovered the major cell populations residing in adult skeletal muscle. We characterized for the first time, 2 endothelial cell populations and potential ways to separate them from one another (Abcg2). Additionally, we identified Gpr37l1 as a novel marker for direct Schwann cell isolation from skeletal muscle. Finally, we compiled a dense cell-cell communication network based on the transcriptome of the identified cell clusters. This serves as a resource to decipher the complex cellular interplay regulating skeletal muscle homeostasis and MuSC quiescence.

4.5 Limitations and future work

In this chapter, we use sc-RNA seq to identify the various cell types present in homeostatic skeletal muscle. With the popularization of sc-RNA seq the various limitations of this method in analyzing the transcriptome of different cells are also being increasingly brought to light. Perhaps one of the most important considerations of sc-RNA seq analysis of tissues is the changes to the transcriptome of the cells during the isolation process. The total preparation time of our cells, from euthanizing the mouse to cell lysis for library preparation, is ~5 hours. This 119

lengthy process disrupts the stem cell niche and exposes the cells to stress. Studies on MuSCs have attempted to shed light on the changes of the transcriptome during the isolation process. Although some studies suggest that major transcriptional changes occur in MuSCs during the isolation process (Machado et al., 2017), other studies show that the ex-vivo transcriptome remains largely reflective of the in-vivo transcriptome (van Velthoven et al., 2017). Nevertheless, precaution should be taken when interpreting the results.

Furthermore, the accuracy of our analysis can be further increased through the analysis of additional cells. Unlike recently published manuscripts (Dell’Orso et al., 2019; Giordani et al., 2018; Schaum et al., 2018), our analysis was not able to identify any smooth muscle cells, macrophages, or T cells which could be as a result of sequencing lower numbers compared to other studies. However, the high number of reads in our analysis was able to identify previously unreported subtypes of endothelial cells and fibroadipogenic progenitors. Future work should focus on identifying the functional differences between the various subpopulations of these cell types.

As described in the discussion, the cellular interactome derived from the transcriptome of the identified cell types, is a prediction of the intercellular signaling. It does not consider the spatiotemporal resolution, post-transcriptional regulation, and anatomical barriers of the cells. Nevertheless, it provides as a rich resource to decipher the intercellular communication occurring in homeostatic skeletal muscle.

Finally, this dataset provides new entry points for the development of endogenous repair therapies. Using this resource, we have obtained a better understanding of the interactome of cells present in healthy homeostatic skeletal muscle and the signaling between them. Furthermore, it provides a blueprint for what is required in a healthy skeletal muscle. By comparing the results of this dataset to similar datasets obtained from aged or diseased muscle, we can identify irregularities in both the interactome of the individual cell types (e.g. aged MuSCs vs young MuSCs), as well as disruptions in intercellular signaling (e.g. reduction of Fibronectin-MuSC interactions in aged muscle). Upon identifying these variations from healthy skeletal muscle, we can then design drugs to block signaling pathways that should not be present or administer growth factors to promote missing processes.

120

4.6 Materials and methods 4.6.1 Skeletal muscle single cell isolation for scRNA seq

Mononuclear cells from adult skeletal muscle were isolated using modified methods as described in Sacco et al., 2008. Briefly, the tibialis anterior (TA) muscle of a euthanized 8 week old C57Bl6/N mouse were harvested, minced with surgical scissors, and transferred to 1 ml DMEM containing Type 1A collagenase (Sigma, cat. no. C9891) at 628 Units/ml. The muscle was then placed in an incubator at 37°C and 5% CO2 for 1 hour. The semi-digested muscle was then passed through a 1 mL pipette tip 10 times. Dispase II was then added to the digestion media at a final concentration of 0.04 U/ml, and placed back into the incubator. After 1 hour, the enzymatically digested muscle was physically dissociated by passing the cell slurry through a 20G needle using a 1 ml syringe 5-10 times. 1 ml of FACS buffer (PBS, 2.5% goat serum, 2 mM EDTA) was added to the digested muscle and the entire mixture was passed through a 40 µm cell strainer (Corning, cat. no. 352340). The filtered cell slurry was then centrifuged at 400g for 15 minutes and the supernatant was removed. 1 ml of red blood cell (RBC) lysis buffer (0.155 M NH4Cl, 0.01 M KHCO3, 0.1 mM EDTA) was added to the pellet and allowed to incubate for 7 minutes at room temperature. The cells were washed and resuspended in 200 µl FACS buffer. 0.25 µM Calcein AM (1:8000, ThermoFisher, cat. no. C3100MP) was added to the cells and incubated at room temperature for 30 minutes. PI was then added to the cell suspension and the live metabolically active mononuclear cells (PI-/Calcein+) from skeletal muscle were isolated using FACS.

4.6.2 Staining of single cell preparations and tissue sections

To stain for the endothelial and Schwann cell surface markers identified in the scRNA seq analysis, following RBC lysis as described in 4.6.1, the cells were incubated with the conjugated or non-conjugated antibodies for 1 hour at 4°C. Next, the cells were washed with FACS buffer, and in the case of non-conjugated primary antibodies, 2° antibodies were added to the cells and incubated for 30 minutes at 4°C. The cells were washed for a final time and PI was added to the cell suspension. The cells were then analyzed on a BD FACS Aria II.

The antibodies used in this study were Cd31-APC (1:200, BD, cat. no. 551262), rabbit anti- Kcna5 (1:200, ThermoFisher, cat. no. PA5-77650), rabbit anti-Alpl (1:200, Abcam, cat. no. 121

ab108337), Scarb1-PE (1:200, Novus Biologicals, NB400-104PE), rabbit anti-Gpr37l1 (1:200, abcam, ab151518), rabbit anti-Kcna1 (1:200, ThermoFisher, cat. no. PA5-77654), Alexafluor goat anti-rabbit 488 (1:500, Life Technologies, cat. no. A11008).

4.6.3 Immunohistochemistry and Immunocytochemistry

For obtaining the immunohistochemistry results, the tibialis anterior (TA) of euthanized C57bl/6N mice was harvested and flash frozen in liquid nitrogen cooled isopentane. The TA muscle was then cryosectioned into 10 µm thick sections. The sections were then thawed to room temperature and rehydrated using PBS. The samples were then fixed using 4% paraformaldehyde (Thermo Fisher, cat. no. 28908) for 10 minutes at room temperature, and subsequently washed 8x and incubated for 1 hour with blocking solution: 20% goat serum (Life Technologies, cat. no. 16210-072), 79% PBS, and 1% Triton-X100 (Bioshop, cat. no. TRX777). Following incubation, primary antibody was added onto the slides for 2 hours at room temperature. Following incubation, the slides were washed 8x with blocking solution, and 2° antibody was added for 30 minutes at room temperature. The slides were then washed with blocking solution for a final time. Fluoromount (Sigma, cat. no. F4680) was then added to the slides to prepare them for imaging.

For immunocytochemistry analysis of the IMS32 Schwann cell line, the cells were fixed with 4% PFA for 10 minutes, washed and incubated with blocking solution at 4°C overnight. Primary antibody was then added onto the cells and incubated for 2 hours at room temperature. Following primary antibody incubation, the cells were washed, and 2° antibody was added for 30 minutes at room temperature. The cells were washed for a final time to prepare for imaging.

The antibodies used in this study were rabbit anti-Gpr37l1 (1:200, abcam, ab151518), rabbit anti- Kcna1 (1:200, ThermoFisher, cat. no. PA5-77654), Alexafluor goat anti-rabbit 647 (1:500, Life Technologies, cat. no. A21244). Hoechst 33342 (1:1000, Life Technologies, cat. no. H3570) was used to visualize the nuclei.

4.6.4 Single cell RNA sequencing analysis

To prepare the scRNA seq data, we used FACS to isolate the live metabolically active mononuclear cells from the TA of an adult (8 week old) C57Bl/6N mouse as described in 4.6.1. We

122

loaded ~4000 cells into the 10X Chromium system using the v2 single cell reagent kit (10X Genomics). Following the manufacturer’s protocol, we constructed the libraries which we subsequently sequenced using an Illumina HiSeq 2000. We used Cell Ranger 2.2.0 (10X Genomics) to process the sequencing results and obtain the gene expression matrix of the sequenced cells. We then used R version 3.4.1 and the Seurat 2.4.0 R package(Butler et al., 2018) to carry out the remaining analyses.

From the 4000 cells entered into the Chromium system, only 2039 cells passed both CellRanger quality control filters, as well as the additional filters we placed on the cells: genes must be expressed in at least 1 cell; cells must express between 200 and 5000 genes; cells should contain less than 20000 Unique Molecular Identifiers (UMIs); Mitochondrial genes should map less than 25% of a cell’s reads.

We then used principal component analysis (PCA) to reduce dimensionality in our dataset. We first identified 20 significant principal components (PCs) using methods implemented in Seurat (Macosko et al., 2015). We then used the identified PCs to cluster the cells based on their transcriptional similarity. We used a clustering resolution of 0.6 for the current analysis. Finally, we visualized the clusters through t-distributed stochastic neighbor embedding to obtain a 2D image of the identified clusters. After clustering the cells, we determined the differentially expressed genes in each cluster using built in functions in the Seurat package.

Finally, to obtain the cellular interactome, we obtained a compiled database of ligand-receptor interactions from a list curated in the Bader lab (Bader). We then linked any two cell types where a ligand in the database is expressed in once cell type and a cognate receptor of that ligand is expressed in another. The directionality of this interactome was preserved when plotting the results as observed in Figure 4.8.

123

Chapter 5 Conclusions, impact and future work Summary and impact 5.1 Summary of results

Throughout this thesis, we built upon prior efforts in the development and advancement of therapies for skeletal muscle regeneration. In the area of cell transplantation therapy, challenges remain in producing clinically-relevant numbers of cells that, as a population, possess high regenerative potency to produce skeletal muscle and repopulate the stem cell niche. We implemented a bioactive hydrogel as a cell delivery vehicle to improve MuSC transplantation outcomes (Chapter 2). We next identified signaling pathways whose inhibition improved the regenerative capacity of cultured MuSC populations (Chapter 3). And finally, we uncovered the cellular diversity and intercellular communication present in the adult skeletal muscle (Chapter 4). This dataset serves as a valuable resource that can potentially act as a gateway towards identifying new endogenous repair therapies, in addition to new ways to increase myogenic transplantation efficiency.

We first demonstrate that it is possible to improve the engraftment potential as well as dispersion of transplanted MuSCs by embedding them in a physical blend of hyaluronan (HA) and methylcellulose (MC) prior to injection into injured muscle. From in-vitro and in-vivo experiments we reveal that the bioactive and shear-thinning properties of HAMC increase the number of MuSCs exiting the syringe during the injection procedure, prevent the active clearance of the transplanted cells, and promote MuSC proliferation. Our results suggest HAMC is a useful tool to address common hurdles associated with cell transplantations: cell survival, rapid injection site clearance, poor donor cell dispersion, and limited contributions to tissue repair.

Next, we address another challenge in cell-based therapies: the inability to generate clinically relevant numbers of cells with therapeutic potential. Using a high-throughput drug screen and in- vivo transplantation assay validation, we identified Egfr and Kdr as novel druggable targets for increasing the regenerative capacity of cultured MuSCs. Intramuscular transplantation of MuSCs

124

cultured in the presence of Egfr or Kdr inhibitors, resulted in an increase in the donor-derived fiber area compared to MuSCs treated with carrier control. Furthermore, the results of the high throughput drug screen serve as a resource for the identification of other regulators of MuSC fate.

Finally, through scRNA seq of the mononuclear cells in skeletal muscle, we obtain a holistic and unbiased view of the cells present in homeostatic adult skeletal muscle. We identify 9 transcriptionally distinct cell clusters: MuSCs, 2 EC clusters, 3 FAP clusters, Schwann cells, Tenocytes, and B cells. We report for the first time the presence of 2 distinct EC clusters in skeletal muscle. We also validated Gpr37l1 as a surface marker to directly isolate Schwann cells from digested skeletal muscle. Finally, we predict the cellular interactome in homeostatic skeletal muscle, revealing the numerous cell-cell signaling events that are potentially present and necessary for muscle homeostasis. This analysis serves as a unique resource to study the intercellular communication present in skeletal muscle.

Taken together, our results contribute to efforts in developing and enhancing skeletal muscle treatments. We demonstrate novel methods to improve cell-based therapies. Furthermore, we created a comprehensive and unbiased resource to study cellular regulators of skeletal muscle homeostasis.

5.2 Future work

In vivo success of skeletal muscle therapies relies on understanding the mechanisms governing skeletal muscle regeneration, and developing tools to manipulate and guide this complex cellular process. Each chapter of this thesis focused on either obtaining a better understanding of the underlying cellular mechanics of skeletal muscle regeneration, or developing and identifying tools to control the fate of skeletal muscle stem cells. Although these results have proven to overcome existing hurdles in the development of skeletal muscle therapies to some extent, next iterations should enhance these findings to develop feasible therapies for a clinical setting. Methods to improve the findings of each chapter and future work in those areas was previously discussed in each of the chapters. However, another way to improve on the findings can be achieved by combining the results of this thesis.

125

Our results have demonstrated that by embedding the donor cells in HAMC we are able to increase their regenerative potential. Furthermore, our drug screen has revealed that by inhibiting Egfr in ex-vivo cultured MuSCs, we are able to increase their regenerative capacity (presumably by preventing myogenic progression) as well as expand cell numbers. Upon transplantation into regenerating muscle, donor cells have a limited time prior to committing to differentiation or reverting to a quiescent state. This reduces the effective time in which the transplanted cells can contribute to muscle regeneration. By adding Egfr inhibitor into the HAMC hydrogel and using this modified HAMC as the cell delivery vehicle, we speculate that we can further improve transplantation outcomes. Another approach is upon identification of new regulators of MuSC quiescence using the sc-RNA seq analysis, new compounds inhibiting or presenting the identified regulators will be embedded into the HAMC hydrogel to prevent or promote the quiescence of the transplanted cells depending on the desired outcome. These are just 2 cases of potential work that can be followed up on to develop new skeletal muscle therapies.

5.3 Impact

Canada’s senior population (aged 65 and older) has been steadily increasing over the past years and seniors are expected to represent 23-25% of the total population by 2036 (Statistics Canada, 2011). Age related sarcopenia is estimated to be prevalent in 8-40% of elderly people over the age of 60, and 50% of elderly over 80 (Fielding et al., 2011; Kan et al., 2009; Zembroń-Łacny et al., 2014). Additionally, medical advances have enabled clinicians to prolong the life of patients diagnosed with various muscular dystrophies. As such, there is an increasing need for treatments and therapies to restore skeletal muscle strength and function to pathological muscle.

The work presented in this thesis builds upon prior advancements in the development of skeletal muscle cell and endogenous repair therapies. We demonstrate that it is possible to improve the outcomes of MuSC transplantations using a bioactive hydrogel, and thereby reducing the number of required cells. Furthermore, we reveal novel MuSC culture inhibition targets for the generation of myogenic cells with high therapeutic potential. And finally, we obtain a holistic view of the cellular diversity present in adult skeletal muscle. Moreover, we report the first ever cellular interactome among the cells present in adult skeletal muscle. Our work here aims

126

improve currently available strategies as well as to lay the groundwork for the emergence of new therapeutic entry-points for skeletal muscle treatments.

127

References

Abou-khalil, R., Mounier, R., and Chazaud, B. (2010). The “ ménage à trois ” of satellite cells , periendothelial cells and endothelial cells Regulation of myogenic stem cell behavior by vessel cells. Cell Cycle 9.

Abou-Khalil, R., Le Grand, F., Pallafacchina, G., Valable, S., Authier, F.-J., Rudnicki, M. a, Gherardi, R.K., Germain, S., Chretien, F., Sotiropoulos, A., et al. (2009). Autocrine and paracrine angiopoietin 1/Tie-2 signaling promotes muscle satellite cell self-renewal. Cell Stem Cell 5, 298–309.

Abu-Baker, A., and Rouleau, G.A. (2007). Oculopharyngeal muscular dystrophy: Recent advances in the understanding of the molecular pathogenic mechanisms and treatment strategies. Biochim. Biophys. Acta - Mol. Basis Dis. 1772, 173–185.

Acharyya, S., Sharma, S.M., Cheng, A.S., Ladner, K.J., He, W., Kline, W., Wang, H., Ostrowski, M.C., Huang, T.H., and Guttridge, D.C. (2010). TNF inhibits notch-1 in skeletal muscle cells by Ezh2 and DNA methylation mediated repression: Implications in Duchenne muscular dystrophy. PLoS One 5.

Aguado, B.A., Mulyasasmita, W., Su, J., Lampe, K.J., and Heilshorn, S.C. (2012). Improving Viability of Stem Cells During Syringe Needle Flow Through the Design of Hydrogel Cell Carriers. Tissue Eng. Part A 18, 806–815.

Aguilar, C.A., Pop, R., Shcherbina, A., Watts, A., Matheny, R.W., Cacchiarelli, D., Han, W.M., Shin, E., Nakhai, S.A., Jang, Y.C., et al. (2016). Transcriptional and Chromatin Dynamics of Muscle Regeneration after Severe Trauma. Stem Cell Reports 7, 983–997.

Alfaro, L., Dick, S., Siegel, A., Anonuevo, A., McNagny, K., Megeney, L., Cornelison, D., and Rossi, F. (2011). CD34 promotes satellite cell motility and entry into proliferation to facilitate efficient skeletal muscle regeneration. Stem Cells 29, 2030–2041.

Ali, S., and Garcia, J.M. (2014). Sarcopenia, cachexia and aging: Diagnosis, mechanisms and therapeutic options - A mini-review. Gerontology 60, 294–305.

128

Allen, R.E., Sheehan, S.M., Taylor, R.G., Kendall, T.L., and Rice, G.M. (1995). Hepatocyte growth factor activates quiescent skeletal muscle satellite cells in vitro. J. Cell. Physiol. 165, 307–312.

Allikian, M.J., and McNally, E.M. (2007). Processing and assembly of the dystrophin glycoprotein complex. Traffic 8, 177–183.

Alonso-Martin, S., Rochat, A., Mademtzoglou, D., Morais, J., de Reyniès, A., Auradé, F., Chang, T.H.-T., Zammit, P.S., and Relaix, F. (2016). Gene Expression Profiling of Muscle Stem Cells Identifies Novel Regulators of Postnatal Myogenesis. Front. Cell Dev. Biol. 4, 1–20.

Anderson, J.E. (2000). A Role for Nitric Oxide in Muscle Repair: Nitric Oxide-mediated Activation of Muscle Satellite Cells. Mol. Biol. Cell 11, 1859–1874.

Apponi, L.H., Leung, S.W., Williams, K.R., Valentini, S.R., Corbett, A.H., and Pavlath, G.K. (2010). Loss of nuclear poly(A)-binding protein 1 causes defects in myogenesis and mRNA biogenesis. Hum. Mol. Genet. 19, 1058–1065.

Arnold, L., Henry, A., Poron, F., Baba-Amer, Y., van Rooijen, N., Plonquet, A., Gherardi, R.K., and Chazaud, B. (2007). Inflammatory monocytes recruited after skeletal muscle injury switch into antiinflammatory macrophages to support myogenesis. J. Exp. Med. 204, 1057–1069.

Arsic, N., Zacchigna, S., Zentilin, L., Ramirez-Correa, G., Pattarini, L., Salvi, A., Sinagra, G., and Giacca, M. (2004). Vascular endothelial growth factor stimulates skeletal muscle regeneration in Vivo. Mol. Ther. 10, 844–854.

Auffray, C., Fogg, D., Garfa, M., Elain, G., Join-Lambert, O., Kayal, S., Sarnacki, S., Cumano, A., Lauvau, G., and Geissmann, F. (2007). Monitoring of Blood Vessels and Tissues by a Population of Monocytes with Patrolling Behavior. Science (80-. ). 317, 666–670.

Bach, L.A. (2015). Endothelial cells and the IGF system. J. Mol. Endocrinol. 54, R1–R13.

Bader, G. Cell-Cell Interaction Database.

Baghdadi, M.B., Firmino, J., Soni, K., Evano, B., Di Girolamo, D., Mourikis, P., Castel, D., and

129

Tajbakhsh, S. (2018a). Notch-Induced miR-708 Antagonizes Satellite Cell Migration and Maintains Quiescence. Cell Stem Cell 23, 859-868.e5.

Baghdadi, M.B., Castel, D., Machado, L., Fukada, S., Birk, D.E., Relaix, F., Tajbakhsh, S., and Mourikis, P. (2018b). Reciprocal signalling by Notch–Collagen V–CALCR retains muscle stem cells in their niche. Nature 557, 714–718.

Baines, C., and Molkentin, J. (2005). STRESS signaling pathways that modulate cardiac myocyte apoptosis. J. Mol. Cell. Cardiol. 38, 47–62.

Ballios, B.G., Cooke, M.J., van der Kooy, D., and Shoichet, M.S. (2010). A hydrogel-based stem cell delivery system to treat retinal degenerative diseases. Biomaterials 31, 2555–2564.

Ballios, B.G., Cooke, M.J., Donaldson, L., Coles, B.L.K., Morshead, C.M., van der Kooy, D., and Shoichet, M.S. (2015). A Hyaluronan-Based Injectable Hydrogel Improves the Survival and Integration of Stem Cell Progeny following Transplantation. Stem Cell Reports 4, 1031–1045.

Bartel, D.P. (2018). Metazoan MicroRNAs. Cell 173, 20–51.

Beauchamp, J.R., Morgan, J.E., Pagel, C.N., and Partridge, T.A. (1999). Dynamics of Myoblast Transplantation Reveal a Discrete Minority of Precursors with Stem Cell–like Properties as the Myogenic Source. J. Cell Biol. 144, 1113–1122.

Beauchamp, J.R., Heslop, L., Yu, D.S.W., Tajbakhsh, S., Kelly, R.G., Wernig, A., Buckingham, M.E., Partridge, T.A., and Zammit, P.S. (2000). Expression of CD34 and Myf5 defines the majority of quiescent adult skeletal muscle satellite cells. J. Cell Biol. 151, 1221–1233.

Beier, J.P., Stern-Straeter, J., Foerster, V.T., Kneser, U., Stark, G.B., and Bach, A.D. (2006). Tissue engineering of injectable muscle: three-dimensional myoblast-fibrin injection in the syngeneic rat animal model. Plast. Reconstr. Surg. 118, 1113–1121; discussion 1122-4.

Beiner, J.M., and Jokl, P. (2001). Muscle contusion injuries: current treatment options. J. Am. Acad. Orthop. Surg. 9, 227–237.

Bencze, M., Negroni, E., Vallese, D., Youssef, H.Y., Chaouch, S., Wolff, A., Aamiri, A., Santo,

130

J.P. Di, Chazaud, B., Butler-browne, G., et al. (2012). Proinflammatory Macrophages Enhance the Regenerative Capacity of Human Myoblasts by Modifying Their Kinetics of Proliferation and Differentiation. 20, 2168–2179.

Bentzinger, C.F., von Maltzahn, J., and Rudnicki, M.A. (2010). Extrinsic regulation of satellite cell specification. Stem Cell Res. Ther. 1, 1–27.

Bentzinger, C.F., Wang, Y.X., Dumont, N.A., and Rudnicki, M.A. (2013a). Cellular dynamics in the muscle satellite cell niche. EMBO Rep. 14, 1062–1072.

Bentzinger, C.F., Wang, Y.X., von Maltzahn, J., Soleimani, V.D., Yin, H., and Rudnicki, M.A. (2013b). Fibronectin regulates Wnt7a signaling and satellite cell expansion. Cell Stem Cell 12, 75–87.

Bentzinger, C.F., von Maltzahn, J., Dumont, N.A., Stark, D.A., Wang, Y.X., Nhan, K., Frenette, J., Cornelison, D., and Rudnicki, M.A. (2014). Wnt7a stimulates myogenic stem cell motility and engraftment resulting in improved muscle strength. J. Cell Biol. 205, 97–111.

Bernet, J.D., Doles, J.D., Hall, J.K., Kelly Tanaka, K., Carter, T. a, and Olwin, B.B. (2014). p38 MAPK signaling underlies a cell-autonomous loss of stem cell self-renewal in skeletal muscle of aged mice. Nat. Med. 20, 265–271.

Bikfalvi, A., Klein, S., Pintucci, G., and Rifkin, D.B. (1997). Biological Roles of Fibroblast Growth Factor-2. Endocr. Rev. 18, 26–45.

Bjornson, C.R.R., Cheung, T.H., Liu, L., Tripathi, P. V., Steeper, K.M., and Rando, T.A. (2012). Notch signaling is necessary to maintain quiescence in adult muscle stem cells. Stem Cells 30, 232–242.

Boldrin, L., Elvassore, N., Malerba, A., Flaibani, M., Cimetta, E., Piccoli, M., Baroni, M.D., Gazzola, M. V, Messina, C., Gamba, P., et al. (2007a). Satellite cells delivered by micro- patterned scaffolds: a new strategy for cell transplantation in muscle diseases. Tissue Eng 13, 253–262.

Boldrin, L., Elvassore, N., Malerba, A., Flaibani, M., Cimetta, E., Piccoli, M., Baroni, M.D., 131

Gazzola, M.V., Messina, C., Gamba, P., et al. (2007b). Satellite cells delivered by micro- patterned scaffolds: a new strategy for cell transplantation in muscle diseases. Tissue Eng 13, 253–262.

Borselli, C., Cezar, C.A., Shvartsman, D., Vandenburgh, H.H., and Mooney, D.J. (2011). The role of multifunctional delivery scaffold in the ability of cultured myoblasts to promote muscle regeneration. Biomaterials 32, 8905–8914.

Bosnakovski, D., Xu, Z., Li, W., Thet, S., Cleaver, O., Perlingeiro, R.C.R.R., and Kyba, M. (2008). Prospective isolation of skeletal muscle stem cells with a Pax7 reporter. Stem Cells 26, 3194–3204.

Bouchentouf, M., Skuk, D., and Tremblay, J.P. (2007a). Early and massive death of myoblasts transplanted into skeletal muscle: responsible factors and potential solutions. Curr. Opin. Organ Transplant. 12, 664–667.

Bouchentouf, M., Benabdallah, B.F., Rousseau, J., Schwartz, L.M., and Tremblay, J.P. (2007b). Induction of Anoikis Following Myoblast Transplantation into SCID Mouse Muscles Requires the Bit1 and FADD Pathways. Am. J. Transplant. 7, 1491–1505.

Bouchentouf, M., Benabdallah, B.F., Bigey, P., Yau, T.M., Scherman, D., and Tremblay, J.P. (2008). Vascular endothelial growth factor reduced hypoxia-induced death of human myoblasts and improved their engraftment in mouse muscles. Gene Ther. 15, 404–414.

Bourguignon, L.Y.W. (2008). Hyaluronan-mediated CD44 activation of RhoGTPase signaling and cytoskeleton function promotes tumor progression. Semin. Cancer Biol. 18, 251–259.

Bourguignon, L.Y.W., Zhu, H., Shao, L., and Chen, Y.W. (2000). CD44 Interaction with Tiam1 Promotes Rac1 Signaling and Hyaluronic Acid-mediated Breast Tumor Cell Migration. J. Biol. Chem. 275, 1829–1838.

Bourguignon, L.Y.W., Gilad, E., Rothman, K., and Peyrollier, K. (2005). Hyaluronan-CD44 interaction with IQGAP1 promotes Cdc42 and ERK signaling, leading to actin binding, Elk- 1/estrogen receptor transcriptional activation, and ovarian cancer progression. J. Biol. Chem.

132

280, 11961–11972.

Bourguignon, L.Y.W., Gilad, E., Brightman, A., Diedrich, F., and Singleton, P. (2006). Hyaluronan-CD44 interaction with leukemia-associated RhoGEF and epidermal growth factor receptor promotes Rho/Ras co-activation, phospholipase C epsilon-Ca2+ signaling, and cytoskeleton modification in head and neck squamous cell carcinoma cells. J. Biol. Chem. 281, 14026–14040.

Bourguignon, L.Y.W., Gilad, E., and Peyrollier, K. (2007). Heregulin-mediated ErbB2-ERK signaling activates hyaluronan synthases leading to CD44-dependent ovarian tumor cell growth and migration. J. Biol. Chem. 282, 19426–19441.

Brack, A.S., Conboy, M.J., Roy, S., Lee, M., Kuo, C.J., Keller, C., and Rando, T.A. (2007). Increased Wnt signaling during aging alters muscle stem cell fate and increases fibrosis. Science 317, 807–810.

Brack, A.S., Conboy, I.M., Conboy, M.J., Shen, J., and Rando, T. a (2008). A temporal switch from notch to Wnt signaling in muscle stem cells is necessary for normal adult myogenesis. Cell Stem Cell 2, 50–59.

Breviario, F., Caveda, L., Corada, M., Martin-Padura, I., Navarro, P., Golay, J., Introna, M., Gulino, D., Lampugnani, M.G., and Dejana, E. (1995). Functional properties of human vascular endothelial cadherin (7B4/cadherin-5), an endothelium-specific cadherin. Arterioscler. Thromb. Vasc. Biol. 15, 1229–1239.

Broussard, S.R., McCusker, R.H., Novakofski, J.E., Strle, K., Shen, W.H., Johnson, R.W., Dantzer, R., and Kelley, K.W. (2014). IL-1 Impairs Insulin-Like Growth Factor I-Induced Differentiation and Downstream Activation Signals of the Insulin-Like Growth Factor I Receptor in Myoblasts. J. Immunol. 172, 7713–7720.

Buckingham, M. (2001). Skeletal muscle formation in vertebrates. Curr. Opin. Genet. Dev. 11, 440–448.

Butler, A., Hoffman, P., Smibert, P., Papalexi, E., and Satija, R. (2018). Integrating single-cell

133

transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol.

Caicco, M.J., Zahir, T., Mothe, A.J., Ballios, B.G., Kihm, A.J., Tator, C.H., and Shoichet, M.S. (2013). Characterization of hyaluronan-methylcellulose hydrogels for cell delivery to the injured spinal cord. J. Biomed. Mater. Res. A 101, 1472–1477.

Carlson, B.M., and Faulkner, J.A. (1989). Muscle transplantation between young and old rats: age of host determines recovery. Am. J. Physiol. Physiol. 256, C1262–C1266.

Carlson, B.M., Dedkov, E.I., Borisov, A.B., and Faulkner, J.A. (2001). Skeletal muscle regeneration in very old rats. Journals Gerontol. - Ser. A Biol. Sci. Med. Sci. 56, 224–233.

Castiglioni, A., Hettmer, S., Lynes, M.D., Rao, T.N., Tchessalova, D., Sinha, I., Lee, B.T., Tseng, Y.-H., and Wagers, A.J. (2014). Isolation of progenitors that exhibit myogenic/osteogenic bipotency in vitro by fluorescence-activated cell sorting from human fetal muscle. Stem Cell Reports 2, 92–106.

Cerletti, M., Jurga, S., Witczak, C.A., Hirshman, M.F., Shadrach, J.L., Goodyear, L.J., and Wagers, A.J. (2008). Highly efficient, functional engraftment of skeletal muscle stem cells in dystrophic muscles. Cell 134, 37–47.

Cezar, C.A., and Mooney, D.J. (2015). Biomaterial-based delivery for skeletal muscle repair. Adv. Drug Deliv. Rev. 84, 188–197.

Chakkalakal, J. V., Christensen, J., Xiang, W., Tierney, M.T., Boscolo, F.S., Sacco, A., and Brack, A.S. (2014). Early forming label-retaining muscle stem cells require p27kip1 for maintenance of the primitive state. Development 141, 1649–1659.

Chakkalakal, J. V, Jones, K.M., Basson, M.A., and Brack, A.S. (2012). The aged niche disrupts muscle stem cell quiescence. Nature 490, 355–360.

Chakravarthy, M. V., Abraha, T.W., Schwartz, R.J., Fiorotto, M.L., and Booth, F.W. (2000). Insulin-like growth factor-I extends in vitro replicative life span of skeletal muscle satellite cells by enhancing G1/S cell cycle progression via the activation of phosphatidylinositol 3’- kinase/Akt signaling pathway. J. Biol. Chem. 275, 35942–35952. 134

Chamberlain, J.R., and Chamberlain, J.S. (2017). Progress toward Gene Therapy for Duchenne Muscular Dystrophy. Mol. Ther. 25, 1125–1131.

Charville, G.W., Cheung, T.H., Yoo, B., Santos, P.J., Lee, G.K., Shrager, J.B., and Rando, T.A. (2015). Ex Vivo Expansion and In Vivo Self-Renewal of Human Muscle Stem Cells. Stem Cell Reports 5, 621–632.

Chazaud, B. (2003). Endoventricular porcine autologous myoblast transplantation can be successfully achieved with minor mechanical cell damage. Cardiovasc. Res. 58, 444–450.

Chen, G., and Quinn, L.S. (1992). Partial characterization of skeletal myoblast mitogens in mouse crushed muscle extract. J. Cell. Physiol. 153, 563–574.

Cheung, T.H., Quach, N.L., Charville, G.W., Liu, L., Park, L., Edalati, A., Yoo, B., Hoang, P., and Rando, T. a (2012). Maintenance of muscle stem-cell quiescence by microRNA-489. Nature 482, 524–528.

Chillakuri, C.R., Sheppard, D., Lea, S.M., and Handford, P.A. (2012). Notch receptor-ligand binding and activation: Insights from molecular studies. Semin. Cell Dev. Biol. 23, 421–428.

Christov, C., Chretien, F., Abou-khalil, R., Bassez, G., Vallet, G., Authier, F.-J., Bassaglia, Y., Shinin, V., Tajbakhsh, S., Chazaud, B., et al. (2007). Muscle Satellite Cells and Endothelial Cells : Close Neighbors and Privileged Partners □. Mol. Biol. Cell 18, 1397–1409.

Cima, L.G., Vacanti, J.P., Vacanti, C., Ingber, D., Mooney, D., and Langer, R. (1991). Tissue Engineering by Cell Transplantation Using Degradable Polymer Substrates. J. Biomech. Eng. 113, 143.

Clevers, H., and Nusse, R. (2012). Wnt/β-catenin signaling and disease. Cell 149, 1192–1205.

Cohn, R.D., Henry, M.D., Michele, D.E., Barresi, R., Saito, F., Moore, S.A., Flanagan, J.D., Skwarchuk, M.W., Robbins, M.E., Mendell, J.R., et al. (2002). Disruption of Dag1 in Differentiated Skeletal Muscle Reveals a Role for Dystroglycan in Muscle Regeneration. Cell 110, 639–648.

135

Cohn, R.D., Van Erp, C., Habashi, J.P., Soleimani, A.A., Klein, E.C., Lisi, M.T., Gamradt, M., Ap Rhys, C.M., Holm, T.M., Loeys, B.L., et al. (2007). Angiotensin II type 1 receptor blockade attenuates TGF-β-induced failure of muscle regeneration in multiple myopathic states. Nat. Med. 13, 204–210.

Collins, C.A., Olsen, I., Zammit, P.S., Heslop, L., Petrie, A., Partridge, T.A., and Morgan, J.E. (2005). Stem cell function, self-renewal, and behavioral heterogeneity of cells from the adult muscle satellite cell niche. Cell 122, 289–301.

Conboy, I.M., and Rando, T. a (2002). The regulation of Notch signaling controls satellite cell activation and cell fate determination in postnatal myogenesis. Dev. Cell 3, 397–409.

Conboy, I.M., Conboy, M.J., Smythe, G.M., and Rando, T.A. (2003). Notch-Mediated Restoration of Regenerative Potential to Aged Muscle. Science (80-. ). 302, 1575–1577.

Conboy, I.M., Conboy, M.J., Wagers, A.J., Girma, E.R., Weissman, I.L., and Rando, T. a (2005). Rejuvenation of aged progenitor cells by exposure to a young systemic environment. Nature 433, 760–764.

Coolican, S.A., Samuel, D.S., Ewton, D.Z., McWade, F.J., and Florini, J.R. (1997). The mitogenic and myogenic actions of insulin-like growth factors utilize distinct signaling pathways. J. Biol. Chem. 272, 6653–6662.

Cornelison, D.D., Filla, M.S., Stanley, H.M., Rapraeger, a C., and Olwin, B.B. (2001). Syndecan-3 and syndecan-4 specifically mark skeletal muscle satellite cells and are implicated in satellite cell maintenance and muscle regeneration. Dev. Biol. 239, 79–94.

Cornelison, D.D.W., Olwin, B.B., Rudnicki, M.A., and Wold, B.J. (2000). MyoD(-/-) satellite cells in single-fiber culture are differentiation defective and MRF4 deficient. Dev. Biol. 224, 122–137.

Cornelison, D.D.W., Wilcox-Adelman, S.A., Goetinck, P.F., Rauvala, H., Rapraeger, A.C., and Olwin, B.B. (2004). Essential and separable roles for Syndecan-3 and Syndecan-4 in skeletal muscle development and regeneration. Genes Dev. 18, 2231–2236.

136

Cosgrove, B.D., Sacco, A., Gilbert, P.M., and Blau, H.M. (2009). A home away from home: challenges and opportunities in engineering in vitro muscle satellite cell niches. Differentiation. 78, 185–194.

Cosgrove, B.D., Gilbert, P.M., Porpiglia, E., Mourkioti, F., Lee, S.P., Corbel, S.Y., Llewellyn, M.E., Delp, S.L., and Blau, H.M. (2014). Rejuvenation of the muscle stem cell population restores strength to injured aged muscles. Nat. Med. 20, 255–264.

Crist, C.G., Montarras, D., and Buckingham, M. (2012). Muscle satellite cells are primed for myogenesis but maintain quiescence with sequestration of Myf5 mRNA targeted by microRNA- 31 in mRNP granules. Cell Stem Cell 11, 118–126.

Davie, J.K., Cho, J.H., Meadows, E., Flynn, J.M., Knapp, J.R., and Klein, W.H. (2007). Target gene selectivity of the myogenic basic helix-loop-helix transcription factor myogenin in embryonic muscle. Dev. Biol. 311, 650–664.

Davies, K.E., and Grounds, M.D. (2007). Modified Patient Stem Cells as Prelude to Autologous Treatment of Muscular Dystrophy. Cell Stem Cell 1, 595–596.

Davoudi, S., Chin, C.-Y., Cooke, M.J., Tam, R.Y., Shoichet, M.S., and Gilbert, P.M. (2018). Muscle stem cell intramuscular delivery within hyaluronan methylcellulose improves engraftment efficiency and dispersion. Biomaterials 173, 34–46.

Decary, S., Ben Hamida, C., Mouly, V., Barbet, J.P., Hentati, F., and Butler-Browne, G.S. (2000). Shorter telomeres in dystrophic muscle consistent with extensive regeneration in young children. Neuromuscul. Disord. 10, 113–120.

Dell’Orso, S., Juan, A.H., Ko, K.-D., Naz, F., Gutierrez-Cruz, G., Feng, X., and Sartorelli, V. (2019). Single-cell analysis of adult skeletal muscle stem cells in homeostatic and regenerative conditions. Development dev.174177.

Dellavalle, A., Maroli, G., Covarello, D., Azzoni, E., Innocenzi, A., Perani, L., Antonini, S., Sambasivan, R., Brunelli, S., Tajbakhsh, S., et al. (2011). Pericytes resident in postnatal skeletal muscle differentiate into muscle fibres and generate satellite cells. Nat. Commun. 2, 499.

137

DiMario, J., Buffinger, N., Yamada, S., and Strohman, R. (1989). Fibroblast growth factor in the extracellular matrix of dystrophic (mdx) mouse muscle. Science (80-. ). 244, 688–690.

Docheva, D., Hunziker, E.B., Fassler, R., and Brandau, O. (2005). Tenomodulin Is Necessary for Tenocyte Proliferation and Tendon Maturation. Mol. Cell. Biol. 25, 699–705.

Dodson, S., Baracos, V.E., Jatoi, A., Evans, W.J., Cella, D., Dalton, J.T., and Steiner, M.S. (2010). Muscle Wasting in Cancer Cachexia: Clinical Implications, Diagnosis, and Emerging Treatment Strategies. Annu. Rev. Med. 62, 265–279.

Doerflinger, N.H., Macklin, W.B., and Popko, B. (2003). Inducible site-specific recombination in myelinating cells. Genesis 35, 63–72.

Doyle, M.J., Zhou, S., Tanaka, K.K., Pisconti, A., Farina, N.H., Sorrentino, B.P., and Olwin, B.B. (2011). Abcg2 labels multiple cell types in skeletal muscle and participates in muscle regeneration. J. Cell Biol. 195, 147–163.

Duan, C., Ren, H., and Gao, S. (2010). Insulin-like growth factors (IGFs), IGF receptors, and IGF-binding proteins: roles in skeletal muscle growth and differentiation. Gen. Comp. Endocrinol. 167, 344–351.

Dumont, N., Lepage, K., Côté, C.H., and Frenette, J. (2007). Mast cells can modulate leukocyte accumulation and skeletal muscle function following hindlimb unloading. J. Appl. Physiol. 103, 97–104.

Dumont, N. a., Wang, Y.X., and Rudnicki, M.A. (2015a). Intrinsic and extrinsic mechanisms regulating satellite cell function. Development 142, 1572–1581.

Dumont, N.A., Bentzinger, C.F., Sincennes, M.-C., and Rudnicki, M.A. (2015b). Satellite Cells and Skeletal Muscle Regeneration. Compr. Physiol. 5, 1027–1059.

Dumont, N.A., Wang, Y.X., Von Maltzahn, J., Pasut, A., Bentzinger, C.F., Brun, C.E., and Rudnicki, M.A. (2015c). Dystrophin expression in muscle stem cells regulates their polarity and asymmetric division. Nat. Med. 21, 1455–1463.

138

Egerman, M.A., Cadena, S.M., Gilbert, J.A., Meyer, A., Nelson, H.N., Swalley, S.E., Mallozzi, C., Jacobi, C., Jennings, L.L., Clay, I., et al. (2015). GDF11 Increases with Age and Inhibits Skeletal Muscle Regeneration. Cell Metab. 22, 164–174.

Ehrhardt, J., and Morgan, J. (2005). Regenerative capacity of skeletal muscle. Curr. Opin. Neurol. 18, 548–553.

Elabd, C., Cousin, W., Upadhyayula, P., Chen, R.Y., Chooljian, M.S., Li, J., Kung, S., Jiang, K.P., and Conboy, I.M. (2014). Oxytocin is an age-specific circulating hormone that is necessary for muscle maintenance and regeneration. Nat. Commun. 5, 1–11.

Emery, A.E. (2002). The muscular dystrophies. Lancet 359, 687–695.

Enge, M., Arda, H.E., Mignardi, M., Beausang, J., Bottino, R., Kim, S.K., and Quake, S.R. (2017). Single-Cell Analysis of Human Pancreas Reveals Transcriptional Signatures of Aging and Somatic Mutation Patterns. Cell 171, 1–10.

Ernfors, P., Wetmore, C., Eriksdotter-Nilsson, M., Bygdeman, M., Strömberg, I., Olson, L., and Persson, H. (1991). The nerve growth factor receptor gene is expressed in both neuronal and non-neuronal tissues in the human fetus. Int. J. Dev. Neurosci. 9, 57–66.

Evanko, S.P., Angello, J.C., and Wight, T.N. (1999). Formation of Hyaluronan- and Versican- Rich Pericellular Matrix Is Required for Proliferation and Migration of Vascular Smooth Muscle Cells. Arterioscler. Thromb. Vasc. Biol. 19, 1004–1013.

Farina, N.H., Hausburg, M., Betta, N.D., Pulliam, C., Srivastava, D., Cornelison, D., and Olwin, B.B. (2012). A role for RNA post-transcriptional regulation in satellite cell activation. Skelet. Muscle 2, 21.

Fielding, R.A., Manfredi, T.J., Ding, W., Fiatarone, M.A., Evans, W.J., and Cannon, J.G. (1993). Acute phase response in exercise. III. Neutrophil and IL-1 beta accumulation in skeletal muscle. Am. J. Physiol. Integr. Comp. Physiol. 265, R166–R172.

Fielding, R.A., Vellas, B., Evans, W.J., Bhasin, S., Morley, J.E., Newman, A.B., Abellan van Kan, G., Andrieu, S., Bauer, J., Breuille, D., et al. (2011). Sarcopenia: an undiagnosed condition 139

in older adults. Current consensus definition: prevalence, etiology, and consequences. International working group on sarcopenia. J. Am. Med. Dir. Assoc. 12, 249–256.

Fields, R.D. (2010). Schwann Cells and Axon Relationship. Encycl. Neurosci. 485–489.

Fitts, R.H., Riley, D.R., and Widrick, J.J. (2000). Physiology of a Microgravity Environment Invited Review: Microgravity and skeletal muscle. J. Appl. Physiol. 89, 823–839.

Florini, J.R., Ewton, D.Z., and Roof, S.L. (2009). Insulin-Like Growth Factor-I Stimulates Terminal Myogenic Differentiation by Induction of Myogenin Gene Expression. Mol. Endocrinol. 5, 718–724.

Fluck, M. (2006). Functional, structural and molecular plasticity of mammalian skeletal muscle in response to exercise stimuli. J. Exp. Biol. 209, 2239–2248.

Fu, X., Xiao, J., Wei, Y., Li, S., Liu, Y., Yin, J., Sun, K., Sun, H., Wang, H., Zhang, Z., et al. (2015). Combination of inflammation-related cytokines promotes long-term muscle stem cell expansion. Cell Res. 25, 655–673.

Fukada, S., Uezumi, A., Ikemoto, M., Masuda, S., Segawa, M., Tanimura, N., Yamamoto, H., Miyagoe-Suzuki, Y., and Takeda, S. (2007). Molecular signature of quiescent satellite cells in adult skeletal muscle. Stem Cells 25, 2448–2459.

Galli, S.J., Borregaard, N., and Wynn, T.A. (2011). Phenotypic and functional plasticity of cells of innate immunity: Macrophages, mast cells and neutrophils. Nat. Immunol. 12, 1035–1044.

Gao, S.P., Mark, K.G., Leslie, K., Pao, W., Motoi, N., Gerald, W.L., Travis, W.D., Bornmann, W., Veach, D., Clarkson, B., et al. (2007). Mutations in the EGFR kinase domain mediate STAT3 activation via IL-6 production in human lung adenocarcinomas. J. Clin. Invest. 117, 3846–3856.

García-Prat, L., Martínez-Vicente, M., Perdiguero, E., Ortet, L., Rodríguez-Ubreva, J., Rebollo, E., Ruiz-Bonilla, V., Gutarra, S., Ballestar, E., Serrano, A.L., et al. (2016). Autophagy maintains stemness by preventing senescence. Nature 529, 37–42.

140

Garrett, W.E. (1996). Muscle Strain Injuries. Am. J. Sports Med. 24, S2–S8.

Gatchalian, C.L., Schachner, M., and Sanes, J.R. (1989). Fibroblasts that proliferate near denervated synaptic sites in skeletal muscle synthesize the adhesive molecules tenascin(J1), N- CAM, fibronectin, and a heparan sulfate proteoglycan. J. Cell Biol. 108, 1873–1890.

George, R.M., Biressi, S., Beres, B.J., Rogers, E., Mulia, A.K., Allen, R.E., Rawls, A., Rando, T.A., and Wilson-Rawls, J. (2013). Numb-deficient satellite cells have regeneration and proliferation defects. Proc. Natl. Acad. Sci. 110, 18549–18554.

Gilbert, P.M., Havenstrite, K.L., Magnusson, K.E.G., Sacco, A., Leonardi, N. a, Kraft, P., Nguyen, N.K., Thrun, S., Lutolf, M.P., and Blau, H.M. (2010). Substrate elasticity regulates skeletal muscle stem cell self-renewal in culture. Science 329, 1078–1081.

Giordani, L., He, G.J., Negroni, E., Sakai, H., Law, J.Y.C., Siu, M.M., Wan, R., Corneau, A., Tajbakhsh, S., Cheung, T.H., et al. (2018). High-Dimensional Single-Cell Cartography Reveals Novel Skeletal Muscle Resident Cell Populations. SSRN Electron. J. 1–13.

Glebova, N.O. (2004). Heterogeneous Requirement of NGF for Sympathetic Target Innervation In Vivo. J. Neurosci. 24, 743–751.

Golding, J.P., Calderbank, E., Partridge, T. a, and Beauchamp, J.R. (2007). Skeletal muscle stem cells express anti-apoptotic ErbB receptors during activation from quiescence. Exp. Cell Res. 313, 341–356.

Goodell, M.A., Brose, K., Paradis, G., Conner, A.S., and Mulligan, R.C. (1996). Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J. Exp. Med. 183, 1797–1806.

Goodell, M.A., Rosenzweig, M., Kim, H., Marks, D.F., Demaria, M., Paradis, G., Grupp, S.A., Sieff, C.A., Mulligan, R.C., and Johnson, R.P. (1997). Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species. Nat. Med. 3, 1337–1345.

Gopinath, S.D., and Rando, T.A. (2008). Stem Cell Review Series: Aging of the skeletal muscle 141

stem cell niche. Aging Cell 7, 590–598.

Le Grand, F., Jones, A.E., Seale, V., Scimè, A., and Rudnicki, M. a (2009). Wnt7a activates the planar cell polarity pathway to drive the symmetric expansion of satellite stem cells. Cell Stem Cell 4, 535–547.

Gros, J., Manceau, M., Thomé, V., and Marcelle, C. (2005). A common somitic origin for embryonic muscle progenitors and satellite cells. Nature 435, 954–958.

Grounds, M.D. (1998). Age-associated Changes in the Response of Skeletal Muscle Cells to Exercise and Regenerationa. Ann. N. Y. Acad. Sci. 854, 78–91.

Gu, X., Ding, F., Yang, Y., and Liu, J. (2015). Tissue Engineering in Peripheral Nerve Regeneration (Elsevier Inc.).

Guérette, B., Skuk, D., Célestin, F., Huard, C., Tardif, F., Asselin, I., Roy, B., Goulet, M., Roy, R., Entman, M., et al. (1997). Prevention by anti-LFA-1 of acute myoblast death following transplantation. J. Immunol. 159, 2522–2531.

Gupta, D., Tator, C.H., and Shoichet, M.S. (2006). Fast-gelling injectable blend of hyaluronan and methylcellulose for intrathecal, localized delivery to the injured spinal cord. Biomaterials 27, 2370–2379.

Halevy, O., Novitch, B.G., Spicer, D.B., Skapek, S.X., Hannon, G.J., Beach, D., Lassar, A.B., Skapek, S.X., Rhee, J., Hannon, G.J., et al. (1995). Correlation of Terminal Cell Cycle Arrest of Skeletal Muscle with Induction of p21 by MyoD Published by : American Association for the Advancement of Science Stable URL : http://www.jstor.org/stable/2886294 JSTOR is a not-for- profit service that helps sch. Science (80-. ). 267, 1018–1021.

Han, W.M., Anderson, S.E., Mohiuddin, M., Barros, D., Nakhai, S.A., Shin, E., Amaral, I.F., Pêgo, A.P., García, A.J., and Jang, Y.C. (2018). Synthetic matrix enhances transplanted satellite cell engraftment in dystrophic and aged skeletal muscle with comorbid trauma. Sci. Adv. 4.

Hardy, D., Besnard, A., Latil, M., Jouvion, G., Briand, D., Thépenier, C., Pascal, Q., Guguin, A., Gayraud-Morel, B., Cavaillon, J.-M., et al. (2016). Comparative Study of Injury Models for 142

Studying Muscle Regeneration in Mice. PLoS One 11, e0147198.

Hayashiji, N., Yuasa, S., Miyagoe-Suzuki, Y., Hara, M., Ito, N., Hashimoto, H., Kusumoto, D., Seki, T., Tohyama, S., Kodaira, M., et al. (2015). G-CSF supports long-term muscle regeneration in mouse models of muscular dystrophy. Nat. Commun. 6, 6745.

Heredia, J.E., Mukundan, L., Chen, F.M., Mueller, A.A., Deo, R.C., Locksley, R.M., Rando, T.A., and Chawla, A. (2013). Type 2 innate signals stimulate fibro/adipogenic progenitors to facilitate muscle regeneration. Cell 153, 376–388.

Heron, M.I., and Richmond, F.J.R. (1993). In‐series fiber architecture in long human muscles. J. Morphol. 216, 35–45.

Hershey, J.C., Baskin, E.P., Glass, J.D., Hartman, H.A., Gilberto, D.B., Rogers, I.T., and Cook, J.J. (2001). Revascularization in the rabbit hindlimb: Dissociation between capillary sprouting and arteriogenesis. Cardiovasc. Res. 49, 618–625.

Hierlihy, A.M., Seale, P., Lobe, C.G., Rudnicki, M.A., and Megeney, L.A. (2002). The postnatal heart contains a myocardial stem cell population. FEBS Lett. 530, 239–243.

Hill, E., Boontheekul, T., and Mooney, D.J. (2006). Regulating activation of transplanted cells controls tissue regeneration. Proc. Natl. Acad. Sci. U. S. A. 103, 2494–2499.

Ho, A.T. V., Palla, A.R., Blake, M.R., Yucel, N.D., Wang, Y.X., Magnusson, K.E.G., Holbrook, C.A., Kraft, P.E., Delp, S.L., and Blau, H.M. (2017). Prostaglandin E2 is essential for efficacious skeletal muscle stem-cell function, augmenting regeneration and strength. Proc. Natl. Acad. Sci. 114, 201705420.

Hoffman, E.P., Brown, R.H., and Kunkel, L.M. (1987). Dystrophin: The protein product of the duchenne muscular dystrophy locus. Cell 51, 919–928.

Holbro, T., and Hynes, N.E. (2004). ErbB receptors: directing key signaling networks throughout life. Annu. Rev. Pharmacol. Toxicol. 44, 195–217.

Hollenberg, S.M., Cheng, P.F., and Weintraub, H. (2006). Use of a conditional MyoD

143

transcription factor in studies of MyoD trans-activation and muscle determination. Proc. Natl. Acad. Sci. 90, 8028–8032.

Hosoyama, T., Nishijo, K., Prajapati, S.I., Li, G., and Keller, C. (2011). Rb1 gene inactivation expands satellite cell and postnatal myoblast pools. J. Biol. Chem. 286, 19556–19564.

Huard, J., Bouchard, J.P., Roy, R., Malouin, F., Dansereau, G., Labrecque, C., Albert, N., Richards, C.L., Lemieux, B., and Tremblay, J.P. (1992). Human myoblast transplantation: Preliminary results of 4 cases. Muscle Nerve 15, 550–560.

Ieronimakis, N., Balasundaram, G., and Reyes, M. (2008). Direct isolation, culture and transplant of mouse skeletal muscle derived endothelial cells with angiogenic potential. PLoS One 3.

Ikemoto, M., Fukada, S.I., Uezumi, A., Masuda, S., Miyoshi, H., Yamamoto, H., Wada, M.R., Masubuchi, N., Miyagoe-Suzuki, Y., and Takeda, S. (2007). Autologous transplantation of SM/C-2.6+ satellite cells transduced with micro-dystrophin CS1 cDNA by lentiviral vector into mdx mice. Mol. Ther. 15, 2178–2185.

Irintchev, A., Zeschnigk, M., and Wernig, A. (1994). Expression pattern of M‐cadherin in normal, denervated, and regenerating mouse muscles. Dev. Dyn. 199, 326–337.

Ishii, K., Sakurai, H., Suzuki, N., Mabuchi, Y., Sekiya, I., Sekiguchi, K., and Akazawa, C. (2018). Recapitulation of Extracellular LAMININ Environment Maintains Stemness of Satellite Cells In Vitro. Stem Cell Reports 10, 568–582.

Janssen, I., Heymsfield, S.B., Wang, Z.M., and Ross, R. (2000). Skeletal muscle mass and distribution in 468 men and women aged 18-88 yr. J. Appl. Physiol. 89, 81–88.

Järvinen, T.A., Kaariainen, M., Jarvinen, M., and Kalimo, H. (2000). Muscle strain injuries. [Review]. Curr. Opin. Rheumatol. 12, 155–161.

Järvinen, T.A.H., Järvinen, T.L.N., Kääriäinen, M., Kalimo, H., and Järvinen, M. (2005). Muscle injuries: Biology and treatment. Am. J. Sports Med. 33, 745–764.

Joe, A.W.B.B., Yi, L., Natarajan, A., Le Grand, F., So, L., Wang, J., Rudnicki, M.A., and Rossi,

144

F.M.V. V (2010). Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis. Nat. Cell Biol. 12, 153–163.

Jones, D.L., and Wagers, A.J. (2008). No place like home: Anatomy and function of the stem cell niche. Nat. Rev. Mol. Cell Biol. 9, 11–21.

Jones, N.C., Fedorov, Y. V., Rosenthal, R.S., and Olwin, B.B. (2001). ERK1/2 is required for myoblast proliferation but is dispensable for muscle gene expression and cell fusion. J. Cell. Physiol. 186, 104–115.

Jones, N.C., Tyner, K.J., Nibarger, L., Stanley, H.M., Cornelison, D.D.W.W., Fedorov, Y. V., and Olwin, B.B. (2005). The p38alpha/beta MAPK functions as a molecular switch to activate the quiescent satellite cell. J. Cell Biol. 169, 105–116.

Juban, G., and Chazaud, B. (2017). Metabolic regulation of macrophages during tissue repair: insights from skeletal muscle regeneration. FEBS Lett. 591, 3007–3021.

Judson, R.N., Quarta, M., Oudhoff, M.J., Soliman, H., Yi, L., Chang, C.K., Loi, G., Vander Werff, R., Cait, A., Hamer, M., et al. (2018). Inhibition of Methyltransferase Setd7 Allows the In Vitro Expansion of Myogenic Stem Cells with Improved Therapeutic Potential. Cell Stem Cell 22, 177-190.e7.

Kan, G.A.V.A.N., Rolland, Y., Andrieu, S., Bauer, J., Beauchet, O., Bonnefoy, M., Cesari, M., Donini, L.M., Inzitari, M., Nourhashemi, F., et al. (2009). Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) Task Force. J. Nutr. Health Aging 13, 881–889.

Kang, Y., Tierney, M., Ong, E., Zhang, L., Piermarocchi, C., Sacco, A., and Paternostro, G. (2015). Combinations of kinase inhibitors protecting myoblasts against hypoxia. PLoS One 10, 1–16.

Kardon, G., Harfe, B.D., and Tabin, C.J. (2003). A Tcf4-positive mesodermal population provides a prepattern for vertebrate limb muscle patterning. Dev. Cell 5, 937–944.

Kasama, B.T., Strieter, R.M., Standiford, T.J., Burdick, M.D., and Kunkel, S.L. (1993). From the 145

Departments of *Pathology and ~Internal Medicine, Division of Pulmonary and Critical Care Medicine, The University of Michigan Medical School, Ann Arbor, Michigan 481O9. 178.

Kassar-Duchossoy, L. (2005). Pax3/Pax7 mark a novel population of primitive myogenic cells during development. Genes Dev. 19, 1426–1431.

Katoh, M., and Katoh, M. (2007). WNT signaling pathway and stem cell signaling network. Clin. Cancer Res. 13, 4042–4045.

Kelly, A.M. (1978). Perisynaptic satellite cells in the developing and mature rat soleus muscle. Anat. Rec. 190, 891–903.

Kherif, S., Lafuma, C., Dehaupas, M., Lachkar, S., Fournier, J.G., Verdière-Sahuqué, M., Fardeau, M., and Alameddine, H.S. (1999). Expression of matrix metalloproteinases 2 and 9 in regenerating skeletal muscle: A study in experimentally injured and mdx muscles. Dev. Biol. 205, 158–170.

Kirouac, D.C., Ito, C., Csaszar, E., Roch, A., Yu, M., Sykes, E.A., Bader, G.D., and Zandstra, P.W. (2010). Dynamic interaction networks in a hierarchically organized tissue. Mol. Syst. Biol. 6, 1–16.

Klank, R.L., Decker Grunke, S.A., Bangasser, B.L., Forster, C.L., Price, M.A., Odde, T.J., SantaCruz, K.S., Rosenfeld, S.S., Canoll, P., Turley, E.A., et al. (2017). Biphasic Dependence of Glioma Survival and Cell Migration on CD44 Expression Level. Cell Rep. 18, 23–31.

Koch, U., Lehal, R., and Radtke, F. (2013). Stem cells living with a Notch. Development 140, 689–704.

Kong, L.-L., Yang, N.-Z., Shi, L.-H., Zhao, G.-H., Zhou, W., Ding, Q., Wang, M.-H., and Zhang, Y.-S. (2017). The optimum marker for the detection of lymphatic vessels. Mol. Clin. Oncol. 7, 515–520.

Kottlors, M., and Kirschner, J. (2010). Elevated satellite cell number in Duchenne muscular dystrophy. Cell Tissue Res. 340, 541–548.

146

Kuang, S., Kuroda, K., Le Grand, F., and Rudnicki, M.A. (2007). Asymmetric self-renewal and commitment of satellite stem cells in muscle. Cell 129, 999–1010.

Lafreniere, J.-F., Caron, M.-C., Skuk, D., Goulet, M., Cheikh, A.R., and Tremblay, J.P. (2009). Growth factor coinjection improves the migration potential of monkey myogenic precursors without affecting cell transplantation success. Cell Transplant. 18, 719–730.

Langen, R.C.J., Van Der Velden, J.L.J., Schols, A.M.W.J., Kelders, M.C.J.M., Wouters, E.F.M., and Janssen-Heininger, Y.M.W. (2004). Tumor necrosis factor-alpha inhibits myogenic differentiation through MyoD protein destabilization. FASEB J. 18, 227–237.

Latroche, C., Weiss-Gayet, M., Muller, L., Gitiaux, C., Leblanc, P., Liot, S., Ben-Larbi, S., Abou-Khalil, R., Verger, N., Bardot, P., et al. (2017). Coupling between Myogenesis and Angiogenesis during Skeletal Muscle Regeneration Is Stimulated by Restorative Macrophages. Stem Cell Reports 9, 2018–2033.

Lean, G., Halloran, M., Mariscal, O., Jamet, S., and Lumb, J. (2019). Ex vivo expansion of skeletal muscle stem cells with a novel small compound inhibitor of eIF2 α dephosphorylation.

Lehto, M.U., and Järvinen, M.J. (1991). Muscle injuries, their healing process and treatment. Ann. Chir. Gynaecol. 80, 102–108.

Lemaire, P.A., Anderson, E., Lary, J., and Cole, J.L. (2008). Mechanism of PKR Activation by dsRNA. J. Mol. Biol. 381, 351–360.

Lemos, D.R., Babaeijandaghi, F., Low, M., Chang, C.K., Lee, S.T., Fiore, D., Zhang, R.H., Natarajan, A., Nedospasov, S.A., and Rossi, F.M.V. (2015). Nilotinib reduces muscle fibrosis in chronic muscle injury by promoting TNF-mediated apoptosis of fibro/adipogenic progenitors. Nat. Med. 21, 786–794.

Lesault, P.-F., Theret, M., Magnan, M., Cuvellier, S., Niu, Y., Gherardi, R.K., Tremblay, J.P., Hittinger, L., and Chazaud, B. (2012). Macrophages improve survival, proliferation and migration of engrafted myogenic precursor cells into MDX skeletal muscle. PLoS One 7, 1–10.

Lesley, J., Hascall, V.C., Tammi, M., and Hyman, R. (2000). Hyaluronan binding by cell surface 147

CD44. J. Biol. Chem. 275, 26967–26975.

Levitan, I., and Kaczmarek, L. (2015). The Neuron: Cell and Molecular Biology (Oxford University Press).

Li, Y.-P. (2013). TNF-α is a mitogen in skeletal muscle. Am. J. Physiol. Physiol. 285, C370– C376.

Liang, J., Jiang, D., and Noble, P.W. (2016). Hyaluronan as a therapeutic target in human diseases. Adv. Drug Deliv. Rev. 97, 186–203.

Lieber, R.L. (2002). Skeletal Muscle Structure, Function, and Plasticity (Lippincott Williams & Wilkins).

Liu, L., Cheung, T.H., Charville, G.W., Hurgo, B.M.C., Leavitt, T., Shih, J., Brunet, A., and Rando, T.A. (2013). Chromatin modifications as determinants of muscle stem cell quiescence and chronological aging. Cell Rep. 4, 189–204.

Liu, L., Cheung, T.H., Charville, G.W., and Rando, T.A. (2015). Isolation of skeletal muscle stem cells by fluorescence-activated cell sorting. Nat. Protoc. 10, 1612–1624.

Liu, L., Charville, G.W., Cheung, T.H., Yoo, B., Santos, P.J., Schroeder, M., and Rando, T.A. (2018). Impaired Notch Signaling Leads to a Decrease in p53 Activity and Mitotic Catastrophe in Aged Muscle Stem Cells. Cell Stem Cell 23, 544-556.e4.

Liu, X., Liu, Y., Zhao, L., Zeng, Z., Xiao, W., and Chen, P. (2017). Macrophage depletion impairs skeletal muscle regeneration: The roles of regulatory factors for muscle regeneration. Cell Biol. Int. 41, 228–238.

Lohela, M., Bry, M., Tammela, T., and Alitalo, K. (2009). VEGFs and receptors involved in angiogenesis versus lymphangiogenesis. Curr. Opin. Cell Biol. 21, 154–165.

Long, C., McAnally, J.R., Shelton, J.M., Mireault, A.A., Bassel-Duby, R., and Olson, E.N. (2014). Prevention of muscular dystrophy in mice by CRISPR/Cas9-mediated editing of germline DNA. Science (80-. ). 345, 1184–1188.

148

Lukjanenko, L., Jung, M.J., Hegde, N., Perruisseau-Carrier, C., Migliavacca, E., Rozo, M., Karaz, S., Jacot, G., Schmidt, M., Li, L., et al. (2016). Loss of fibronectin from the aged stem cell niche affects the regenerative capacity of skeletal muscle in mice. Nat. Med. 22, 897–905.

Lukjanenko, L., Karaz, S., Stuelsatz, P., Rudnicki, M.A., Bentzinger, C.F., Feige, J.N., Lukjanenko, L., Karaz, S., Stuelsatz, P., Gurriaran-rodriguez, U., et al. (2019). Aging Disrupts Muscle Stem Cell Function by Impairing Matricellular WISP1 Secretion from Fibro- Adipogenic Progenitors Article Aging Disrupts Muscle Stem Cell Function by Impairing Matricellular WISP1 Secretion from Fibro-Adipogenic Progenitors. Cell Stem Cell Cell 24, 433–446.

Lutolf, M.P., and Hubbell, J. a (2005). Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nat. Biotechnol. 23, 47–55.

Machado, L., Esteves de Lima, J., Fabre, O., Proux, C., Legendre, R., Szegedi, A., Varet, H., Ingerslev, L.R., Barrès, R., Relaix, F., et al. (2017). In Situ Fixation Redefines Quiescence and Early Activation of Skeletal Muscle Stem Cells. Cell Rep. 21, 1982–1993.

Machida, S., Spangenburg, E.E., and Booth, F.W. (2003). Forkhead transcription factor FoxO1 transduces insulin-like growth factor’s signal to p27Kip1 in primary skeletal muscle satellite cells. J. Cell. Physiol. 196, 523–531.

Macosko, E.Z., Basu, A., Satija, R., Nemesh, J., Shekhar, K., Goldman, M., Tirosh, I., Bialas, A.R., Kamitaki, N., Martersteck, E.M., et al. (2015). Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. Cell 161, 1202–1214.

MacParland, S.A., Liu, J.C., Ma, X.Z., Innes, B.T., Bartczak, A.M., Gage, B.K., Manuel, J., Khuu, N., Echeverri, J., Linares, I., et al. (2018). Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 1–21.

Madhavan, S., Moreb, E.A., Gersbach, C.A., Asokan, A., Thakore, P.I., Pan, X., Ousterout, D.G., Yan, W.X., Ran, F.A., Zhang, F., et al. (2016). In vivo genome editing improves muscle function in a mouse model of Duchenne muscular dystrophy. Science (80-. ). 351, 403–407.

Maher, P. (1999). P38 Mitogen-Activated Protein Kinase Activation Is Required for Fibroblast

149

Growth Factor-2-Stimulated Cell Proliferation But Not Differentiation. J. Biol. Chem. 274, 17491–17498. von Maltzahn, J., Chang, N.C., Bentzinger, C.F., and Rudnicki, M. a (2012). Wnt signaling in myogenesis. Trends Cell Biol. 22, 602–609.

Marquardt, L.M., and Heilshorn, S.C. (2016). Design of Injectable Materials to Improve Stem Cell Transplantation. Curr. Stem Cell Reports 2, 207–220.

Martin, C.M., Meeson, A.P., Robertson, S.M., Hawke, T.J., Richardson, J.A., Bates, S., Goetsch, S.C., Gallardo, T.D., and Garry, D.J. (2004). Persistent expression of the ATP-binding cassette transporter, Abcg2, identifies cardiac SP cells in the developing and adult heart. Dev. Biol. 265, 262–275.

Matthews, E., Brassington, R., Kuntzer, T., Jichi, F., and Manzur, A.Y. (2016). Corticosteroids for the treatment of Duchenne muscular dystrophy. Cochrane Database Syst. Rev. CD003725.

Mauro, A. (1961). Satellite cell of skeletal muscle fibers. J. Biophys. Biochem. Cytol. 9, 493– 495.

Mayer, U., Saher, G., Fässler, R., Bornemann, A., Echtermeyer, F., von der Mark, H., Miosge, N., Pöschl, E., and von der Mark, K. (1997). Absence of integrin alpha 7 causes a novel form of muscular dystrophy. Nat. Genet. 17, 318–323.

Meeson, A.P., Hawke, T.J., Graham, S., Jiang, N., Elterman, J., Hutcheson, K., DiMaio, J.M., Gallardo, T.D., and Garry, D.J. (2004). Cellular and Molecular Regulation of Skeletal Muscle Side Population Cells. Stem Cells 22, 1305–1320.

Mendell, J.R., and Rodino-Klapac, L.R. (2016). Duchenne muscular dystrophy: CRISPR/Cas9 treatment. Cell Res. 26, 513–514.

Mendell, J.R., Kissel, J.T., Amato, A.A., King, W., Signore, L., Prior, T.W., Sahenk, Z., Benson, S., McAndrew, P.E., and Rice, R. (1995). Myoblast transfer in the treatment of Duchenne’s muscular dystrophy. N. Engl. J. Med. 333, 832–838.

150

Mitrousis, N., Tam, R.Y., Baker, A.E.G., van der Kooy, D., and Shoichet, M.S. (2016). Hyaluronic Acid-Based Hydrogels Enable Rod Photoreceptor Survival and Maturation In Vitro through Activation of the mTOR Pathway. Adv. Funct. Mater. 26, 1975–1985.

Mofarrahi, M., McClung, J.M., Kontos, C.D., Davis, E.C., Tappuni, B., Moroz, N., Pickett, A.E., Huck, L., Harel, S., Danialou, G., et al. (2015). Angiopoietin-1 enhances skeletal muscle regeneration in mice. Am. J. Physiol. Integr. Comp. Physiol. 308, R576–R589.

Montarras, D., Morgan, J., Collins, C., Relaix, F., Zaffran, S., Cumano, A., Partridge, T., and Buckingham, M. (2005a). Direct isolation of satellite cells for skeletal muscle regeneration. Science 309, 2064–2067.

Montarras, D., Morgan, J., Collins, C., Relaix, F.F., Zaffran, S.S., Cumano, A., Partridge, T., and Buckingham, M. (2005b). Direct isolation of satellite cells for skeletal muscle regeneration. Science 309, 2064–2067.

Morgan, J.E., and Zammit, P.S. (2010). Direct effects of the pathogenic mutation on satellite cell function in muscular dystrophy. Exp. Cell Res. 316, 3100–3108.

Mothe, A.J., Tam, R.Y., Zahir, T., Tator, C.H., and Shoichet, M.S. (2013). Repair of the injured spinal cord by transplantation of neural stem cells in a hyaluronan-based hydrogel. Biomaterials 34, 3775–3783.

Müller-Esterl, W., Dikic, I., Nikolic, I., Babuke, T., Picuric, S., Bicker, F., Plate, K.H., Meister, J., and Schmidt, M.H.H. (2009). Epidermal growth factor-like domain 7 (EGFL7) modulates Notch signalling and affects neural stem cell renewal. Nat. Cell Biol. 11, 873–880.

Murphy, M.M., Lawson, J.A., Mathew, S.J., Hutcheson, D.A., and Kardon, G. (2011). Satellite cells , connective tissue fibroblasts and their interactions are crucial for muscle regeneration. 3637, 3625–3637.

Musarò, A., and Rosenthal, N. (2015). Maturation of the Myogenic Program Is Induced by Postmitotic Expression of Insulin-Like Growth Factor I. Mol. Cell. Biol. 19, 3115–3124.

Musarò, A., McCullagh, K., Paul, A., Houghton, L., Dobrowolny, G., Molinaro, M., Barton, 151

E.R., L Sweeney, H., and Rosenthal, N. (2001). Localized Igf-1 transgene expression sustains hypertrophy and regeneration in senescent skeletal muscle. Nat. Genet. 27, 195–200.

Mylona, E., Jones, K.A., Mills, S.T., and Pavlath, G.K. (2006). CD44 regulates myoblast migration and differentiation. J. Cell. Physiol. 209, 314–321.

Newman, P., Berndt, M., Gorski, J., White, G., Lyman, S., Paddock, C., and Muller, W. (1990). PECAM-1 (CD31) cloning and relation to adhesion molecules of the immunoglobulin gene superfamily. Science (80-. ). 247, 1219–1222.

Novak, M.L., and Koh, T.J. (2013). Macrophage phenotypes during tissue repair. J. Leukoc. Biol. 93, 875–881.

Olguín, H.C., and Pisconti, A. (2012). Marking the tempo for myogenesis: Pax7 and the regulation of muscle stem cell fate decisions. J. Cell. Mol. Med. 16, 1013–1025.

Orimo, H. (2010). The mechanism of mineralization and the role of alkaline phosphatase in health and disease. J. Nippon Med. Sch. 77, 4–12.

Oyama, T., Nagai, T., Wada, H., Naito, A.T., Matsuura, K., Iwanaga, K., Takahashi, T., Goto, M., Mikami, Y., Yasuda, N., et al. (2007). Cardiac side population cells have a potential to migrate and differentiate into cardiomyocytes in vitro and in vivo. J. Cell Biol. 176, 329–341.

Pala, F., Di Girolamo, D., Mella, S., Yennek, S., Chatre, L., Ricchetti, M., and Tajbakhsh, S. (2018). Distinct metabolic states govern skeletal muscle stem cell fates during prenatal and postnatal myogenesis. J. Cell Sci. 131, jcs212977.

Palacios, D., Mozzetta, C., Consalvi, S., Caretti, G., Saccone, V., Proserpio, V., Marquez, V.E., Valente, S., Mai, A., Forcales, S. V, et al. (2010). TNF/p38α/polycomb signaling to Pax7 locus in satellite cells links inflammation to the epigenetic control of muscle regeneration. Cell Stem Cell 7, 455–469.

Pallafacchina, G., François, S., Regnault, B., Czarny, B., Dive, V., Cumano, A., Montarras, D., and Buckingham, M. (2010). An adult tissue-specific stem cell in its niche: a gene profiling analysis of in vivo quiescent and activated muscle satellite cells. Stem Cell Res. 4, 77–91. 152

Papapetropoulos, A., Fulton, D., Mahboubi, K., Kalb, R.G., O’Connor, D.S., Li, F., Altieri, D.C., and Sessa, W.C. (2000). Angiopoietin-1 Inhibits Endothelial Cell Apoptosis via the Akt/Survivin Pathway. J. Biol. Chem. 275, 9102–9105.

Partridge, T.A., Morgan, J.E., Coulton, G.R., Hoffman, E.P., and Kunkel, L.M. (1989). Conversion of mdx myofibres from dystrophin-negative to -positive by injection of normal myoblasts. Nature 337, 176–179.

Paz, N.G., and Amore, P.A.D. (2014). Arterial versus venous endothelial cells Nathaniel. Cell Tissue Res. 335, 5–16.

Pedersen, B.K., and Febbraio, M.A. (2012). Muscles, exercise and obesity: Skeletal muscle as a secretory organ. Nat. Rev. Endocrinol. 8, 457–465.

Perie, S., Trollet, C., Mouly, V., Vanneaux, V., Mamchaoui, K., Bouazza, B., Marolleau, J.P., Laforet, P., Chapon, F., Eymard, B., et al. (2014). Autologous myoblast transplantation for oculopharyngeal muscular dystrophy: a phase I/IIa clinical study. Mol. Ther. 22, 219–225.

Périé, S., Mamchaoui, K., Mouly, V., Blot, S., Bouazza, B., Thornell, L.E., St Guily, J.L., and Butler-Browne, G. (2006). Premature proliferative arrest of cricopharyngeal myoblasts in oculopharyngeal muscular dystrophy: Therapeutic perspectives of autologous myoblast transplantation. Neuromuscul. Disord. 16, 770–781.

Peterson, R.M., Yu, Q., Stamenkovic, I., and Toole, B.P. (2000). Perturbation of hyaluronan interactions by soluble CD44 inhibits growth of murine mammary carcinoma cells in ascites. Am. J. Pathol. 156, 2159–2167.

Pfister, O., Mouquet, F., Jain, M., Summer, R., Helmes, M., Fine, A., Colucci, W.S., and Liao, R. (2005). CD31 − but Not CD31 + Cardiac Side Population Cells Exhibit Functional Cardiomyogenic Differentiation. Circ. Res. 97, 52–61.

Pfister, O., Oikonomopoulos, A., Sereti, K.I., Sohn, R.L., Cullen, D., Fine, G.C., Mouquet, F., Westerman, K., and Liao, R. (2008). Role of the ATP-binding cassette transporter Abcg2 in the phenotype and function of cardiac side population cells. Circ. Res. 103, 825–835.

153

Philippos, M., Sambasivan, R., Castel, D., Rocheteau, P., Bizzarro, V., and Tajbakhsh, S. (2012). A critical requirement for notch signaling in maintenance of the quiescent skeletal muscle stem cell state. Stem Cells 30, 243–252.

Piehl-Aulin, K., Laurent, C., Engström-Laurent, A., Hellström, S., and Henriksson, J. (1991). Hyaluronan in human skeletal muscle of lower extremity: concentration, distribution, and effect of exercise. J. Appl. Physiol. 71, 2493–2498.

Pietsch, P. (1961a). The effects of colchicine on regeneration of mouse skeletal muscle. Anat. Rec. 139, 167–172.

Pietsch, P. (1961b). Differentiation in regeneration I. The development of muscle and cartilage following deplantation of regenerating limb blastemata of Amblystoma larvae. Dev. Biol. 3, 255–264.

Ponta, H., Sherman, L., and Herrlich, P. a. (2003). CD44: From adhesion molecules to signalling regulators. Nat. Rev. Mol. Cell Biol. 4, 33–45.

Porpiglia, E., Samusik, N., Van Ho, A.T., Cosgrove, B.D., Mai, T., Davis, K.L., Jager, A., Nolan, G.P., Bendall, S.C., Fantl, W.J., et al. (2017). High-resolution myogenic lineage mapping by single-cell mass cytometry. Nat. Cell Biol. 19, 558–567.

Praud, C., Montarras, D., Pinset, C., and Sebille, A. (2003). Dose effect relationship between the number of normal progenitor muscle cells grafted in mdx mouse skeletal striated muscle and the number of dystrophin-positive fibres. Neurosci. Lett. 352, 70–72.

Pretheeban, T., Lemos, D.R., Paylor, B., Zhang, R.H., and Rossi, F.M. (2012). Role of stem/progenitor cells in reparative disorders. Fibrogenes. Tissue Repair 5, 1–12.

Price, F.D., von Maltzahn, J., Bentzinger, C.F., Dumont, N.A., Yin, H., Chang, N.C., Wilson, D.H., Frenette, J., and Rudnicki, M.A. (2014). Inhibition of JAK-STAT signaling stimulates adult satellite cell function. Nat. Med. 20, 1174–1181.

Public Health Agency of Canada (2010). Economic Burden of Disease in Canada, 2010.

154

Puthucheary, Z.A., Rawal, J., McPhail, M., Connolly, B., Ratnayake, G., Chan, P., Hopkinson, N.S., Padhke, R., Dew, T., Sidhu, P.S., et al. (2013). Acute skeletal muscle wasting in critical illness. JAMA - J. Am. Med. Assoc. 310, 1591–1600.

Qiao, W., Wang, W., Laurenti, E., Turinsky, A.L., Wodak, S.J., Bader, G.D., Dick, J.E., and Zandstra, P.W. (2014). Intercellular network structure and regulatory motifs in the human hematopoietic system. Mol. Syst. Biol. 10, 741–741.

Quarta, M., Brett, J.O., DiMarco, R., De Morree, A., Boutet, S.C., Chacon, R., Gibbons, M.C., Garcia, V.A., Su, J., Shrager, J.B., et al. (2016). An artificial niche preserves the quiescence of muscle stem cells and enhances their therapeutic efficacy. Nat. Biotechnol. 34, 752–759.

Quinlan, J.G., Lyden, S.P., Cambier, D.M., Johnson, S.R., Michaels, S.E., and Denman, D.L. (1995). Radiation inhibition of mdx mouse muscle regeneration: dose and age factors. Muscle Nerve 18, 201–206.

Ramilowski, J.A., Goldberg, T., Harshbarger, J., Kloppmann, E., Kloppman, E., Lizio, M., Satagopam, V.P., Itoh, M., Kawaji, H., Carninci, P., et al. (2015). A draft network of ligand- receptor-mediated multicellular signalling in human. Nat. Commun. 6, 7866.

Rando, T.A., and Blau, H.M. (1994). Primary mouse myoblast purification, characterization, and transplantation for cell-mediated gene therapy. J. Cell Biol. 125, 1275–1287.

Rawls, A., Valdez, M., Zhang, W., Richardson, J., Klein, W., and Olson, E. (1998). Overlapping functions of the myogenic bHLH genes MRF4 and MyoD revealed in double mutant mice. Development 125, 2349–2358.

Raymond, E., Faivre, S., and Armand, J.P. (2000). Epidermal Growth Factor Receptor Tyrosine Kinase as a Target for Anticancer Therapy. Drugs 60, 15–23.

Reimann, J., Irintchev, A., and Wernig, A. (2000). Regenerative capacity and the number of satellite cells in soleus muscles of normal and mdx mice. Neuromuscul. Disord. 10, 276–282.

Relaix, F., and Zammit, P.S. (2012). Satellite cells are essential for skeletal muscle regeneration: the cell on the edge returns centre stage. Development 139, 2845–2856. 155

Relaix, F., Rocancourt, D., Mansouri, A., and Buckingham, M. (2005). A Pax3/Pax7-dependent population of skeletal muscle progenitor cells. Nature 435, 948–953.

Richler, C., and Yaffe, D. (1970). The in vitro cultivation and differentiation capacities of myogenic cell lines. Dev. Biol. 23, 1–22.

Rieckmann, J.C., Geiger, R., Hornburg, D., Wolf, T., Kveler, K., Jarrossay, D., Sallusto, F., Shen-Orr, S.S., Lanzavecchia, A., Mann, M., et al. (2017). Social network architecture of human immune cells unveiled by quantitative proteomics. Nat. Immunol. 18, 583–593.

Rodrigues, A. de C., and Rodrigues, S.P.D.M. (2010). Myonuclei and satellite cells in denervated rat muscles. Microsurgery 30, 242–248.

Rosenblatt, J.D., Yong, D., and Parry, D.J. (1994). Satellite cell activity is required for hypertrophy of overloaded adult rat muscle. Muscle Nerve 17, 608–613.

Rossi, C.A., Flaibani, M., Blaauw, B., Pozzobon, M., Figallo, E., Reggiani, C., Vitiello, L., Elvassore, N., and De Coppi, P. (2011). In vivo tissue engineering of functional skeletal muscle by freshly isolated satellite cells embedded in a photopolymerizable hydrogel. FASEB J. 25, 2296–2304.

Rousseau, J., Dumont, N., Lebel, C., Quenneville, S.P., Côté, C.H., Frenette, J., and Tremblay, J.P. (2010). Dystrophin expression following the transplantation of normal muscle precursor cells protects mdx muscle from contraction-induced damage. Cell Transplant. 19, 589–596.

Le Roux, I., Konge, J., Le Cam, L., Flamant, P., and Tajbakhsh, S. (2015). Numb is required to prevent p53-dependent senescence following skeletal muscle injury. Nat. Commun. 6, 8528.

Rozo, M., Li, L., and Fan, C.-M. (2016). Targeting β1-integrin signaling enhances regeneration in aged and dystrophic muscle in mice. Nat Med 22, 889–896.

Ruel-Gariépy, E., and Leroux, J.-C. (2004). In situ-forming hydrogels—review of temperature- sensitive systems. Eur. J. Pharm. Biopharm. 58, 409–426.

Sacco, A., Doyonnas, R., Kraft, P., Vitorovic, S., and Blau, H.M. (2008). Self-renewal and

156

expansion of single transplanted muscle stem cells. Nature 456, 502–506.

Sacco, A., Mourkioti, F., Tran, R., Choi, J., Llewellyn, M., Kraft, P., Shkreli, M., Delp, S., Pomerantz, J.H., Artandi, S.E., et al. (2010). Short telomeres and stem cell exhaustion model duchenne muscular dystrophy in mdx/mTR mice. Cell 143, 1059–1071.

Saclier, M., Cuvellier, S., Magnan, M., Mounier, R., and Chazaud, B. (2013a). Monocyte/macrophage interactions with myogenic precursor cells during skeletal muscle regeneration. FEBS J. 280, 4118–4130.

Saclier, M., Yacoub-Youssef, H., Mackey, A.L., Arnold, L., Ardjoune, H., Magnan, M., Sailhan, F., Chelly, J., Pavlath, G.K., Mounier, R., et al. (2013b). Differentially activated macrophages orchestrate myogenic precursor cell fate during human skeletal muscle regeneration. Stem Cells 31, 384–396.

Sadeh, M. (1988). Effects of aging on skeletal muscle regeneration. J. Neurol. Sci. 87, 67–74.

Sambasivan, R., Yao, R., Kissenpfennig, A., Van Wittenberghe, L., Paldi, A., Gayraud-Morel, B., Guenou, H., Malissen, B., Tajbakhsh, S., and Galy, A. (2011). Pax7-expressing satellite cells are indispensable for adult skeletal muscle regeneration. Development 138, 3647–3656.

Sammels, L.M., Bosio, E., Fragall, C.T., Grounds, M.D., van Rooijen, N., and Beilharz, M.W. (2004). Innate Inflammatory Cells are not Responsible for Early Death of Donor Myoblasts after Myoblast Transfer Therapy. Transplantation 77, 1790–1797.

Sanes, J.R. (2003). The Basement Membrane/Basal Lamina of Skeletal Muscle. J. Biol. Chem. 278, 12601–12604.

Scapini, P., Lapinet-vera, J.A., and Cassatella, M.A. The neutrophil as a cellular source of chemokines - Scapini - 2000 - Immunological Reviews - Wiley Online Library. 195–203.

Schaum, N., Karkanias, J., Neff, N.F., May, A.P., Quake, S.R., Wyss-Coray, T., Darmanis, S., Batson, J., Botvinnik, O., Chen, M.B., et al. (2018). Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris. Nature 562, 367–372.

157

Scholz, D., Tomas, S., Sass, S., and Podzuweit, T. (2003). Angiogenesis and myogenesis as two facets of inflammatory post-ischemic tissue regeneration. In Vascular Biochemistry, (Boston, MA: Springer US), pp. 57–67.

Schultz, E. (1978). Changes in the satellite cells of growing muscle following denervation. Anat. Rec. 190, 299–311.

Schultz, E., Gibson, M.C., and Champion, T. (1978). Satellite cells are mitotically quiescent in mature mouse muscle: an EM and radioautographic study. J. Exp. Zool. 206, 451–456.

Seale, P., Sabourin, L. a, Girgis-Gabardo, A., Mansouri, A., Gruss, P., and Rudnicki, M. a (2000). Pax7 is required for the specification of myogenic satellite cells. Cell 102, 777–786.

Sechler, J.L., Corbett, S.A., and Schwarzbauer, J.E. (2013). Modulatory Roles for Integrin Activation and the Synergy Site of Fibronectin during Matrix Assembly. Mol. Biol. Cell 8, 2563–2573.

Serena, E., Flaibani, M., Carnio, S., Boldrin, L., Vitiello, L., De Coppi, P., and Elvassore, N. (2008). Electrophysiologic stimulation improves myogenic potential of muscle precursor cells grown in a 3D collagen scaffold. Neurol Res 30, 207–214.

Serrano, A.L., Mann, C.J., Vidal, B., Ardite, E., Perdiguero, E., and Muñoz-Cánoves, P. (2011). Cellular and molecular mechanisms regulating fibrosis in skeletal muscle repair and disease.

Sethi, J.K., and Vidal-Puig, A. (2010). Wnt signalling and the control of cellular metabolism. Biochem. J. 427, 1–17.

Shea, K.L., Xiang, W., LaPorta, V.S., Licht, J.D., Keller, C., Basson, M.A., and Brack, A.S. (2010). Sprouty1 regulates reversible quiescence of a self-renewing adult muscle stem cell pool during regeneration. Cell Stem Cell 6, 117–129.

Sheehan, S.M., and Allen, R.E. (1999). Skeletal muscle satellite cell proliferation in response to members of the fibroblast growth factor family and hepatocyte growth factor. J. Cell. Physiol. 181, 499–506.

158

Shefer, G., Van de Mark, D.P., Richardson, J.B., and Yablonka-Reuveni, Z. (2006). Satellite-cell pool size does matter: Defining the myogenic potency of aging skeletal muscle. Dev. Biol. 294, 50–66.

Shen, W., Li, Y., Zhu, J., Schwendener, R., and Huard, J. (2008). Interaction between macrophages, TGF-β1, and the COX-2 pathway during the inflammatory phase of skeletal muscle healing after injury. J. Cell. Physiol. 214, 405–412.

Shi, X., and Garry, D.J. (2006). Muscle stem cells in development, regeneration, and disease. Genes Dev. 20, 1692–1708.

Simons, B.D., and Clevers, H. (2011). Strategies for homeostatic stem cell self-renewal in adult tissues. Cell 145, 851–862.

Sinha, M., Jang, Y.C., Oh, J., Khong, D., Wu, E.Y., Manohar, R., Miller, C., Regalado, S.G., Loffredo, F.S., Pancoast, J.R., et al. (2014). Restoring systemic GDF11 levels reverses age- related dysfunction in mouse skeletal muscle. Science 344, 649–652.

Sionkowska, A. (2011). Current research on the blends of natural and synthetic polymers as new biomaterials: Review. Prog. Polym. Sci. 36, 1254–1276.

Skelly, D.A., Squiers, G.T., McLellan, M.A., Bolisetty, M.T., Robson, P., Rosenthal, N.A., and Pinto, A.R. (2018). Single-Cell Transcriptional Profiling Reveals Cellular Diversity and Intercommunication in the Mouse Heart. Cell Rep. 22, 600–610.

Skuk, D. (2004). Myoblast transplantation for inherited myopathies: a clinical approach. Expert Opin. Biol. Ther. 4, 1871–1885.

Skuk, D., and Tremblay, J.P. (2014). Clarifying Misconceptions About Myoblast Transplantation in Myology. Mol. Ther. 22, 897–898.

Skuk, D., Caron, N., Goulet, M., Roy, B., Espinosa, F., and Tremblay, J.P. (2002). Dynamics of the Early Immune Cellular Reactions after Myogenic Cell Transplantation. Cell Transplant. 11, 671–681.

159

Skuk, D., Roy, B., Goulet, M., Chapdelaine, P., Bouchard, J.P., Roy, R., Dugr??, F.J., Lachance, J.G., Desch??nes, L., Senay, H., et al. (2004). Dystrophin expression in myofibers of Duchenne muscular dystrophy patients following intramuscular injections of normal myogenic cells. Mol. Ther. 9, 475–482.

Skuk, D., Goulet, M., and Tremblay, J.P. (2014). Intramuscular transplantation of myogenic cells in primates: importance of needle size, cell number, and injection volume. Cell Transplant. 23, 13–25.

Sleep, E., Cosgrove, B.D., McClendon, M.T., Preslar, A.T., Chen, C.H., Sangji, M.H., Pérez, C.M.R., Haynes, R.D., Meade, T.J., Blau, H.M., et al. (2017). Injectable biomimetic liquid crystalline scaffolds enhance muscle stem cell transplantation. Proc. Natl. Acad. Sci. 114, E7919–E7928.

Slevin, M., Krupinski, J., Gaffney, J., Matou, S., West, D., Delisser, H., Savani, R.C., and Kumar, S. (2007). Hyaluronan-mediated angiogenesis in vascular disease: Uncovering RHAMM and CD44 receptor signaling pathways. Matrix Biol. 26, 58–68.

Snyders, D.J. (1993). A rapidly activating and slowly inactivating potassium channel cloned from human heart. Functional analysis after stable mammalian cell culture expression. J. Gen. Physiol. 101, 513–543.

Sousa-Victor, P., Gutarra, S., García-Prat, L., Rodriguez-Ubreva, J., Ortet, L., Ruiz-Bonilla, V., Jardí, M., Ballestar, E., González, S., Serrano, A.L., et al. (2014). Geriatric muscle stem cells switch reversible quiescence into senescence. Nature 506, 316–321.

Statistics Canada (2011). Canada Year Book. 156–167.

Sultana, S., Al-Shawafi, H.A., Makita, S., Sohda, M., Amizuka, N., Takagi, R., and Oda, K. (2013). An asparagine at position 417 of tissue-nonspecific alkaline phosphatase is essential for its structure and function as revealed by analysis of the N417S mutation associated with severe hypophosphatasia. Mol. Genet. Metab. 109, 282–288.

Sun, L., Ma, K., Wang, H., Xiao, F., Gao, Y., Zhang, W., Wang, K., Gao, X., Ip, N., and Wu, Z.

160

(2007). JAK1-STAT1-STAT3, a key pathway promoting proliferation and preventing premature differentiation of myoblasts. J. Cell Biol. 179, 129–138.

Tabebordbar, M., Zhu, K., Cheng, J.K.W., Chew, W.L., Widrick, J.J., Yan, W.X., Maesner, C., Wu, E.Y., Xiao, R., Ran, F.A., et al. (2016). In vivo gene editing in dystrophic mouse muscle and muscle stem cells. Science (80-. ). 351.

Tam, R.Y., Cooke, M.J., and Shoichet, M.S. (2012). A covalently modified hydrogel blend of hyaluronan-methyl cellulose with peptides and growth factors influences neural stem/progenitor cell fate. J. Mater. Chem. 22, 19402–19411.

Tang, F., Barbacioru, C., Wang, Y., Nordman, E., Lee, C., Xu, N., Wang, X., Bodeau, J., Tuch, B.B., Siddiqui, A., et al. (2009). mRNA-Seq whole-transcriptome analysis of a single cell. Nat. Methods 6, 377–382.

Taniguti, A.P.T., Pertille, A., Matsumura, C.Y., Neto, H.S., and Marques, M.J. (2011). Prevention of muscle fibrosis and myonecrosis in mdxmice by suramin, a TGF-β1 blocker. Muscle and Nerve 43, 82–87.

Tatsumi, R. (2010). Mechano-biology of skeletal muscle hypertrophy and regeneration: possible mechanism of stretch-induced activation of resident myogenic stem cells. Anim. Sci. J. 81, 11– 20.

Tatsumi, R., Anderson, J.E., Nevoret, C.J., Halevy, O., and Allen, R.E. (1998). HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells. Dev. Biol. 194, 114–128.

Tedesco, F.S., Dellavalle, A., Diaz-Manera, J., Messina, G., and Cossu, G. (2010). Repairing skeletal muscle: regenerative potential of skeletal muscle stem cells. J. Clin. Invest. 120, 11–19.

Tidball, J.G. (2005). Inflammatory processes in muscle injury and repair. Am. J. Physiol. Integr. Comp. Physiol. 288, R345–R353.

Tidball, J.G. (2011). Mechanisms of Muscle Injury, Repair, and Regeneration. In Comprehensive Physiology, (Hoboken, NJ, USA: John Wiley & Sons, Inc.), p. 161

Tidball, J.G., and Villalta, S.A. (2010). Regulatory interactions between muscle and the immune system during muscle regeneration. Am. J. Physiol. Integr. Comp. Physiol. 298, R1173–R1187.

Tierney, M.T., Aydogdu, T., Sala, D., Malecova, B., Gatto, S., Puri, P.L., Latella, L., and Sacco, A. (2014). STAT3 signaling controls satellite cell expansion and skeletal muscle repair. Nat. Med. 1–7.

Tiidus, P.M. (1998). Radical species in inflammation and overtraining. Can. J. Physiol. Pharmacol. 76, 533–538.

Tintignac, L.A., Leibovitch, M.P., Kitzmann, M., Fernandez, A., Ducommun, B., Meijer, L., and Leibovitch, S.A. (2000). Cyclin E-Cdk2 phosphorylation promotes late G1-phase degradation of MyoD in muscle cells. Exp. Cell Res. 259, 300–307.

Tonkin, J., Temmerman, L., Sampson, R.D., Gallego-Colon, E., Barberi, L., Bilbao, D., Schneider, M.D., Musarò, A., and Rosenthal, N. (2015). Monocyte/macrophage-derived IGF-1 orchestrates murine skeletal muscle regeneration and modulates autocrine polarization. Mol. Ther. 23, 1189–1200.

Toole, B.P. (2004). Hyaluronan: from extracellular glue to pericellular cue. Nat. Rev. Cancer 4, 528–539.

Tremblay, J.P., Bouchard, J.P., Malouin, F., Théau, D., Cottrell, F., Collin, H., Rouche, A., Gilgenkrantz, S., Abbadi, N., Tremblay, M., et al. (1993). Myoblast transplantation between monozygotic twin girl carriers of Duchenne muscular dystrophy. Neuromuscul. Disord. 3, 583– 592.

Trochon, V., Mabilat, C., Bertrand, P., Legrand, Y., Smadja-Joffe, F., Soria, C., Delpech, B., and Lu, H. (1996). Evidence of involvement of CD44 in endothelial cell proliferation, migration and angiogenesisin vitro. Int. J. Cancer 66, 664–668.

Troy, A., Cadwallader, A.B., Fedorov, Y., Tyner, K., Tanaka, K.K., and Olwin, B.B. (2012). Coordination of satellite cell activation and self-renewal by par-complex-dependent asymmetric activation of p38α/β MAPK. Cell Stem Cell 11, 541–553.

162

Turner, N.J., and Badylak, S.F. (2012). Regeneration of skeletal muscle. 759–774.

Uezumi, A., Ojima, K., Fukada, S.I., Ikemoto, M., Masuda, S., Miyagoe-Suzuki, Y., and Takeda, S. (2006). Functional heterogeneity of side population cells in skeletal muscle. Biochem. Biophys. Res. Commun. 341, 864–873.

Uezumi, A., Fukada, S., Yamamoto, N., Takeda, S., and Tsuchida, K. (2010). Mesenchymal progenitors distinct from satellite cells contribute to ectopic fat cell formation in skeletal muscle. Nat. Cell Biol. 12, 143–152.

Urciuolo, A., Quarta, M., Morbidoni, V., Gattazzo, F., Molon, S., Grumati, P., Montemurro, F., Tedesco, F.S., Blaauw, B., Cossu, G., et al. (2013). Collagen VI regulates satellite cell self- renewal and muscle regeneration. Nat. Commun. 4, 1964.

Varga, T., Mounier, R., Patsalos, A., Gogolák, P., Peloquin, M., Horvath, A., Pap, A., Daniel, B., Nagy, G., Pintye, E., et al. (2016). Macrophage PPARγ, a Lipid Activated Transcription Factor Controls the Growth Factor GDF3 and Skeletal Muscle Regeneration. Immunity 45, 1038–1051. van Velthoven, C.T.J., de Morree, A., Egner, I.M., Brett, J.O., and Rando, T.A. (2017). Transcriptional Profiling of Quiescent Muscle Stem Cells In Vivo. Cell Rep. 21, 1994–2004.

Verma, M., Asakura, Y., Murakonda, B.S.R., Pengo, T., Latroche, C., Chazaud, B., McLoon, L.K., and Asakura, A. (2018). Muscle Satellite Cell Cross-Talk with a Vascular Niche Maintains Quiescence via VEGF and Notch Signaling. Cell Stem Cell 23, 530-543.e9.

Viguie, C.A., Lu, D., Huang, S., and Rengen, H. (1997). Quantitative Study of the Effects of Long-Term Denervation on the Extensor Digitorum Longus Muscle of the Rat. 354, 346–354.

Villalta, S.A., Nguyen, H.X., Deng, B., Gotoh, T., and Tidbal, J.G. (2009). Shifts in macrophage phenotypes and macrophage competition for arginine metabolism affect the severity of muscle pathology in muscular dystrophy. Hum. Mol. Genet. 18, 482–496.

Wada, E., Tanihata, J., Iwamura, A., Takeda, S., Hayashi, Y.K., and Matsuda, R. (2017). Treatment with the anti-IL-6 receptor antibody attenuates muscular dystrophy via promoting skeletal muscle regeneration in dystrophin- / utrophin-deficient mice. 1–15. 163

Wang, Y., and Thorlacius, H. (2005). Mast cell-derived tumour necrosis factor-α mediates macrophage inflammatory protein-2-induced recruitment of neutrophils in mice. Br. J. Pharmacol. 145, 1062–1068.

Wang, X., Wu, H., Zhang, Z., Liu, S., Yang, J., Chen, X., Fan, M., and Wang, X. (2008). Effects of interleukin-6, leukemia inhibitory factor, and ciliary neurotrophic factor on the proliferation and differentiation of adult human myoblasts. Cell. Mol. Neurobiol. 28, 113–124.

Wang, Y., Cooke, M.J., Morshead, C.M., and Shoichet, M.S. (2012). Hydrogel delivery of erythropoietin to the brain for endogenous stem cell stimulation after stroke injury. Biomaterials 33, 2681–2692.

Wang, Y., Wehling-Henricks, M., Samengo, G., and Tidball, J.G. (2015). Increases of M2a macrophages and fibrosis in aging muscle are influenced by bone marrow aging and negatively regulated by muscle-derived nitric oxide. Aging Cell 14, 678–688.

Wang, Y.X., Feige, P., Brun, C.E., Hekmatnejad, B., Dumont, N.A., Renaud, J.M., Faulkes, S., Guindon, D.E., and Rudnicki, M.A. (2019). EGFR-Aurka Signaling Rescues Polarity and Regeneration Defects in Dystrophin-Deficient Muscle Stem Cells by Increasing Asymmetric Divisions. Cell Stem Cell 24, 419-432.e6.

Wardrop, K.E., and Dominov, J.A. (2011). Proinflammatory signals and the loss of lymphatic vessel hyaluronan receptor-1 (LYVE-1) in the early pathogenesis of laminin alpha2-deficient skeletal muscle. J. Histochem. Cytochem. 59, 167–179.

Watt, D.J., Lambert, K., Morgan, J.E., Partridge, T.A., and Sloper, J.C. (1982). Incorporation of donor muscle precursor cells into an area of muscle regeneration in the host mouse. J. Neurol. Sci. 57, 319–331.

Weber, G.F., Bronson, R.T., Ilagan, J., Cantor, H., Schmits, R., and Mak, T.W. (2002). Absence of the CD44 gene prevents sarcoma metastasis. Cancer Res. 62, 2281–2286.

Webster, C., and Blau, H.M. (1990). Accelerated age-related decline in replicative life-span of Duchenne muscular dystrophy myoblasts: Implications for cell and gene therapy. Somat. Cell

164

Mol. Genet. 16, 557–565.

Webster, M.T., and Fan, C.M. (2013). c-MET regulates myoblast motility and myocyte fusion during adult skeletal muscle regeneration. PLoS One 8, 1–16.

Wen, Y., Bi, P., Liu, W., Asakura, A., Keller, C., and Kuang, S. (2012). Constitutive Notch Activation Upregulates Pax7 and Promotes the Self-Renewal of Skeletal Muscle Satellite Cells. Mol. Cell. Biol. 32, 2300–2311.

Wokke, J.H.J., van Den Oord, C.J.M., Leppink, G.J., and Jennekens, F.G.I. (1989). Perisynaptic satellite cells in human external intercostal muscle: A quantitative and qualitative study. Anat. Rec. 223, 174–180.

Wosczyna, M.N., and Rando, T.A. (2018). A Muscle Stem Cell Support Group: Coordinated Cellular Responses in Muscle Regeneration. Dev. Cell 46, 135–143.

Wosczyna, M.N., Biswas, A.A., Cogswell, C.A., and Goldhamer, D.J. (2012). Multipotent progenitors resident in the skeletal muscle interstitium exhibit robust BMP-dependent osteogenic activity and mediate heterotopic ossification. J. Bone Miner. Res. 27, 1004–1017.

Xiao, W., Liu, Y., and Chen, P. (2016). Macrophage Depletion Impairs Skeletal Muscle Regeneration: the Roles of Pro-fibrotic Factors, Inflammation, and Oxidative Stress. Inflammation 39, 2016–2028.

Xu, X., Wilschut, K.J., Kouklis, G., Tian, H., Hesse, R., Garland, C., Sbitany, H., Hansen, S., Seth, R., Knott, P.D., et al. (2015). Human Satellite Cell Transplantation and Regeneration from Diverse Skeletal Muscles. Stem Cell Reports 5, 419–434.

Yennek, S., Burute, M., Théry, M., and Tajbakhsh, S. (2014). Cell adhesion geometry regulates non-random DNA segregation and asymmetric cell fates in mouse skeletal muscle stem cells. Cell Rep. 7, 961–970.

Yin, H., Price, F., and Rudnicki, M. a (2013). Satellite cells and the muscle stem cell niche. Physiol. Rev. 93, 23–67.

165

Yu, Q., Toole, B.P., and Stamenkovic, I. (1997). Induction of Apoptosis of Metastatic Mammary Carcinoma Cells In Vivo by Disruption of Tumor Cell Surface CD44 Function. J. Exp. Med. 186, 1985–1996.

Yuzwa, S.A., Yang, G., Borrett, M.J., Clarke, G., Cancino, G.I., Zahr, S.K., Zandstra, P.W., Kaplan, D.R., and Miller, F.D. (2016). Proneurogenic Ligands Defined by Modeling Developing Cortex Growth Factor Communication Networks. Neuron 91, 988–1004.

Zanoni, P., Khetarpal, S.A., Larach, D.B., Hancock-Cerutti, W.F., Millar, J.S., Cuchel, M., DerOhannessian, S., Kontush, A., Surendran, P., Saleheen, D., et al. (2016). Rare variant in scavenger receptor BI raises HDL cholesterol and increases risk of coronary heart disease. Science (80-. ). 351, 1166–1171.

Zembroń-Łacny, A., Dziubek, W., Rogowski, Ł., Skorupka, E., and Dąbrowska, G. (2014). Sarcopenia: monitoring, molecular mechanisms, and physical intervention. Physiol. Res. 63, 683–691.

Zheng, G.X.Y., Terry, J.M., Belgrader, P., Ryvkin, P., Bent, Z.W., Wilson, R., Ziraldo, S.B., Wheeler, T.D., McDermott, G.P., Zhu, J., et al. (2017). Massively parallel digital transcriptional profiling of single cells. Nat. Commun. 8, 1–12.

Zhu, H., Mitsuhashi, N., Klein, A., Barsky, L.W., Weinberg, K., Barr, M.L., Demetriou, A., and Wu, G.D. (2006). The Role of the Hyaluronan Receptor CD44 in Mesenchymal Stem Cell Migration in the Extracellular Matrix. Stem Cells 24, 928–935.

Zismanov, V., Chichkov, V., Colangelo, V., Jamet, S., Wang, S., Syme, A., Koromilas, A.E., and Crist, C. (2016). Phosphorylation of eIF2α is a Translational Control Mechanism Regulating Muscle Stem Cell Quiescence and Self-Renewal. Cell Stem Cell 18, 79–90.

Zou, Y., Zhang, R.Z., Sabatelli, P., Chu, M.L., and Bönnemann, C.G. (2008). Muscle interstitial fibroblasts are the main source of collagen VI synthesis in skeletal muscle: Implications for congenital muscular dystrophy types Ullrich and Bethlem. J. Neuropathol. Exp. Neurol. 67, 144– 154.

166

Appendix I: Summary of drug screen results

Table A.1: List of compounds designated as hits in drug screen (compound name, targets, and successful doses)

Drug Targets Dose 1-Naphthyl PP1, 1-NA-PP 1 SRC, FYN (p59fyn), ABL1 0.2496 2-(p-Hydroxyanilino)-4-(p-chlorophenyl) SPK1 0.1196, 0.2496, 0.524, 1.12, 2.32, thiazole, HCl, SKI-II, ABC-294640 4.8, 10 5-Iodotubercidin AK, CK1a 0.013, 1.12 7-Cyclopentyl-5-(4-phenoxyphenyl)-7H- LCK (p56lck), SRC 4.8, 10 pyrrolo[2,3-d] pyrimidin-4-ylamine An 83-01 TGFbR1 0.0026, 10 ABT-702 dihydrochloride AK 0.028, 1.12, 4.8, 10 AC220 FLT3, FMS, KIT, PDGFRb, RET 0.2496, 0.524, 1.12 AEG 3482 JNK1 4.8, 10 Afatinib EGFR 1.12, 2.32, 4.8 AG 213, Tyrphostin AG 213 EGFR, PDGFRb 1.12, 10 AG 490 EGFR, JAK2, JAK3/STAT, JAK3/AP-1, JAK3/MAPK 2.32, 4.8, 10 AG 494 EGFR 0.2496, 0.524, 2.32, 4.8, 10 AG 825 ErbB2 0.013, 0.524 AG13958 PDGFRb, KDR 0.524, 2.32, 4.8, 10 Akt-I-1 AKT1 0.013, 1.12, 4.8 Akt-I-1,2 AKT1, AKT2 0.0026, 0.013, 1.12, 2.32, 4.8, 10 Alsterpaullone CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 4.8 CDK2/Cyclin E, CDK5/p35, GSK3 AMG-47a LCK (p56lck), SRC, KDR, p38a 0.0052, 0.524, 1.12, 2.32, 4.8, 10 Amuvatinib AXL, KIT, FLT3, MET, PDGFRa, RET, RAD51 0.524, 1.12, 2.32, 4.8, 10 expression inhibitor Arctigenin, (-)-Arctigenin MAP2K1, DNA Topoisomerase II, HIV Integrase, 0.013, 0.524, 1.12, 2.32, 4.8, 10 NFKB, AP-1 Arcyriaflavin A CDK4/Cyclin D1, CaMK2b 0.013, 10 AS-252424 PI3K 0.0026, 0.1196 AS-601245 JNK1 4.8, 10 ASC-033 IRE1 0.0052, 0.524, 1.12, 2.32 ASC-069 IRE1 0.028 ASC-081 IRE1 0.2496, 2.32 AT-7519 derivative Unknown 0.013, 0.0572, 2.32, 4.8, 10 AV-412, MP-412 EGFR, ErbB2 0.524, 1.12, 2.32, 4.8, 10 AZ 3146 TTK 0.028, 0.0572, 0.524 AZD5438 CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 1.12, 2.32, 4.8 CDK9/Cyclin T1 AZD-7762 hydrochloride CHK1, CHK2 0.524 AZD-7762 hydrochloride CHK1, CHK2 0.0572 AZD-8055 FRAP, mTORC1, mTORC2 0.013 Barasertib AurB 0.1196 Barasertib AurB 0.013, 0.028, 4.8

167

BAY 61-3606 hydrochloride SYK 2.32, 4.8, 10 BI 78D3 JNK1 0.2496, 10 BIBX 1382 dihydrochloride EGFR 4.8, 10 BI-D1870 RSK1 0.0052, 0.2496, 0.524, 1.12, 2.32, 4.8, 10 BIO GSK3 0.524, 1.12, 2.32, 4.8 Bisindolylmaleimide I hydrochloride PKC 0.2496, 2.32, 4.8 Bisindolylmaleimide IX PKC 4.8 Bisindoylmaleimide X, HCl salt PKC 0.0052, 0.2496, 0.524, 1.12, 2.32, 4.8, 10 BIX 02189 MAP2K5, Erk5 0.013, 0.2496, 0.524, 10 BMS 794833 MET, KDR 1.12, 4.8, 10 BMS-2 MET 0.013, 0.2496, 2.32, 10 BMS-3 LIMK1 0.0026, 0.0052, 0.013, 0.1196, 0.524, 1.12 BMS-599626 EGFR, ErbB2, ErbB3, ErbB4 2.32, 4.8, 10 Brivanib KDR, FGFR1, FGFR3, FLT4, FGFR2 0.028 BS-181 hydrochloride CDK7 0.0572 BX912 PDK1 10 C-1 PKC 0.1196 Cabozantinib malate FLT3, MET, KIT, RET, TIE2, KDR 10 Canertinib EGFR, ErbB2 - Irreversible 2.32, 4.8, 10 CC-401 JNK1, JNK2 10 CGK 733 ATR, ATM 0.028, 2.32, 4.8, 10 CGP-57380 MNK1 2.32, 4.8, 10 CGP-57380 MNK1 2.32, 4.8, 10 CGP-74514A hydrochloride CDC2(p34cdc2)/Cyclin B 0.524, 2.32 character (0) character 1.12, 2.32, 4.8, 10 Chelerythrine chloride PKC 4.8 CHIR-124 CHK1 0.0572, 0.1196, 0.2496, 0.524, 1.12 Chk2 Inhibitor II, 339253 CHK2 10 Compound 401 DNAPK, FRAP 0.028 CP466722 ATM 0.2496, 0.524, 1.12, 2.32, 4.8 CP-724714 EGFR, ErbB2 2.32, 4.8, 10 Crizotinib ALK, MET 0.028, 0.0572, 0.2496, 2.32, 4.8, 10 CX-4945 CK2a1 2.32, 4.8, 10 CYC-116 AurA, AurB, KDR 0.0052, 1.12, 2.32, 4.8, 10 Cyclapolin 9 PLK1 4.8, 10 Cyclapolin 9 PLK1 2.32, 4.8, 10 Dasatinib BCR, ABL1, SRC, EphA1, FYN (p59fyn), LCK 0.028, 0.0572, 0.2496, 0.524, (p56lck), PDGFRb, YES, Erk2(p38) 2.32, 4.8, 10 Dioctanoylglycol DAGK 0.524 DMPQ dihydrochloride PDGFRb 0.524, 4.8 Dorsomorphin dihydrochloride AMPK, ALK2, BMPR1A, BMPR1B 1.12, 4.8 Dovitinib KDR, FLT3, KIT, PDGFRb 0.1196, 0.524, 1.12, 2.32

168

EKI-785 EGFR - Irreversible 0.2496, 2.32 EMD-1214063 MET 0.524, 1.12, 2.32, 4.8 ENMD-2076 AurA 0.028, 1.12, 2.32, 4.8, 10 Enzastaurin AKT1, PKCb, p70S6K, GSK3B 0.0026, 0.0052, 0.028, 0.0572, 0.1196, 0.524, 1.12, 2.32, 10 EO-1428 p38a, p38b 0.524, 1.12, 2.32, 4.8, 10 ERK2 inhibitor Erk2 0.028 Erlotinib HCl EGFR, ErbB2 4.8, 10 FAK Inhibitor 14 FAK 4.8 Foretinib MET, KDR, FLT3, TIE2 0.028, 0.524 GDC-0879 BRAF 0.0052, 0.524, 4.8, 10 Gefitinib EGFR, ErbB2 10 GSK-1904529A IGF1R 0.0026, 0.028, 1.12, 2.32, 4.8, 10 GSK690693 AKT1 0.0026, 0.0572, 0.1196, 10 GW2974 EGFR, ErbB2 2.32, 4.8, 10 GW441756 hydrochloride TRKA 4.8, 10 HDS 029 EGFR, ErbB2, ErbB4 0.0572, 10 HMN-214 PLK1 0.0572 IC87114 PI3Kd 0.013 IKK 16 IKKa, I KKB 0.524, 1.12 ILK inhibitor ILK 0.0026, 0.013, 0.0572, 1.12, 2.32 Imatinib (free base) PDGFRb, KIT, ABL1 10 Imatinib Mesylate BCR, ABL1, PDGFRb, KIT 10 IMD 0354, IMD-0354 IKKb 2.32, 4.8 Imidazolo-oxindole PKR inhibitor C16 PKR 0.0052, 0.524, 1.12, 2.32, 4.8 INCB018424 JAK1, JAK2, JAK3, TYK2 1.12, 10 Intedanib KDR, PDGFRb, MET 4.8 IPA 3 PAK1 0.2496, 2.32, 4.8, 10 IRAK4 Inhibitor IRAK4 0.028, 0.1196, 0.2496, 0.524, 1.12, 2.32, 4.8, 10 Janex-1 JAK3 2.32, 4.8, 10 JNJ 28871063 hydrochloride EGFR, ErbB2, ErbB4 4.8, 10 JNJ-10198409 PDGFRb 2.32 JNJ-38877605 MET 4.8 JX-401 p38a 4.8 K-252a TRKA 0.028, 0.0572, 0.1196, 0.2496 K-252c PKC 0.2496, 0.524, 1.12, 2.32 Kenpaullone CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 0.0052, 4.8, 10 CDK2/Cyclin E, CDK5/p35, GSK3, LCK (p56lck) Ki20227 (+/-) FMS, KDR, KIT, PDGFRb 0.0026, 0.013, 0.2496, 1.12, 10 Ki-8751 KDR, KIT, PDGFRa, FGFR2, FLT3 0.0572 KN-62 CAMKK2 4.8, 10 KU-55933 ATM, PI3K 4.8, 10 KU-60019 ATM 0.0026, 0.0052, 0.0572, 0.524, 1.12, 2.32, 4.8, 10 Lapatinib ditosylate EGFR, ErbB2 2.32, 4.8, 10 169

Lavendustin A EGFR, SRC 1.12 Lenvatinib FGFR1, KIT, PDGFRb, FLT1, KDR, FLT4 0.0026, 0.0052 Lestaurtinib FLT3, TRKA, JAK2 0.0572, 0.1196, 0.2496 LFM-A13 BTK 4.8 Linifanib KDR, PDGFRb, FMS, FLT3 0.524, 1.12, 2.32, 4.8, 10 Linsitinib IGF1R 2.32, 4.8, 10 Masitinib mesylate KIT, PDGFRb, FGFR3 0.0572, 2.32, 4.8, 10 MK-1775 WEE1 0.0052, 0.013, 0.0572, 0.1196, 0.2496, 0.524, 1.12, 2.32 MK-2206 AKT1 0.028, 1.12, 2.32, 4.8, 10 MK-8033 MET 0.0052, 1.12, 10 ML 9 hydrochloride smMLCK 4.8, 10 MLN-8237 AurA 0.0052 Momelotinib JAK2 0.0052, 4.8, 10 Motesanib diphosphate salt KIT, PDGFRb, KDR 0.524, 1.12, 2.32, 4.8, 10 Mubritinib ErbB2 0.2496, 0.524, 1.12, 2.32, 10 Necrostatin-1 RIPK1 4.8, 10 Neratinib EGFR, ErbB2 0.2496, 0.524, 1.12, 2.32, 4.8, 10 NH125 eEF2K 0.028, 0.0572, 0.1196 NIK Kinase Inhibitor NIK 4.8, 10 Nilotinib BCR, ABL1, PDGFRb, KIT 4.8, 10 NSC 109555 ditosylate, DDUG CHK2 0.1196, 1.12, 2.32, 4.8, 10 NSC 625987 CDK4/Cyclin D1 4.8, 10 NU-2058 CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 10 CDK2/Cyclin E, CDK5/p35, GSK3 NU-6102 CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A 10 NVP-AEW-541 IGF1R 0.2496, 0.524 OSU-03012 hydrochloride PDK1 4.8, 10 Pazopanib hydrochloride FLT1, KDR, FLT4, KIT, PDGFRa, PDGFRb 4.8 PD-0325901 MAP2K1, MAP2K2 0.013 PD-0332991 CDK4/Cyclin D1, CDK6/Cyclin D3 2.32, 4.8 PD04217903 MET 10 PD-153035 EGFR 4.8, 10 PD-158780 EGFR, ErbB2, ErbB4 0.013, 0.524, 2.32, 10 PD-169316 p38 MAPK 2.32, 10 PD173955 BCR, ABL1, SRC 0.2496, 0.524, 1.12, 2.32, 4.8 PD173955-Analogue 1 BCR, ABL1 0.524, 2.32, 4.8, 10 PD-180970 ABL1, SRC 0.2496, 0.524, 1.12 PD-184161 MAP2K1, MAP2K2 2.32 PD-198306 MAP2K1, MAP2K2 0.013, 0.524, 1.12, 2.32, 10 PD-406976 PKC, JAK2 0.524, 1.12 PD-407824 CHK1, WEE1 0.0026, 1.12, 2.32, 4.8 PERK inhibitor PEK 0.0026, 1.12, 2.32, 4.8 PF-04691502 PI3K, FRAP 0.013

170

PF431396 FAK, PYK2 0.2496 PF562271 FAK, PYK2 0.1196, 0.2496, 1.12 PHA-665752 MET, RON, KDR 2.32 PHA690509 CDK2/Cyclin A 0.1196 PHA-767491 hydrochloride CDC7, CDK9/Cyclin T1, MAPKAPK2 0.028, 1.12 PHA-793887 CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 1.12, 2.32, 10 CDK2/Cyclin E, CDK4/Cyclin D1, CDK5/p25 PI3-Kinase an Inhibitor 2 PI3K 0.1196 PI-93 PI3Ka 0.2496, 0.524, 2.32, 4.8, 10 PIK-294 PI3Ka 0.028, 0.524, 4.8 PIM 1 Inhibitor 2 PIM1, PIM2 0.1196, 10 PKC theta inhibitor PKCt 0.1196, 0.2496, 4.8, 10 PKC-412 FLT3 0.0572 Ponatinib BCR, ABL1, SRC, FLT3, KIT, RET, PDGFRb, FGFR1, 2.32 FGFR2, FGFR3 PP-2 LCK (p56lck), FYN (p59fyn), HCK (p56hck), SRC 0.0052 Purvalanol A CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 10 CDK2/Cyclin E, CDK4/Cyclin D1, CDK5/p35, DYRK1A PV-1019 CHK2 2.32, 4.8 R 59-022 DAGK 10 R1487 p38 MAPK 0.028, 10 R406 FLT3, SYK, production inhibitor 0.2496 RAF265 KIT, PDGFRb, BRAF, RAF1, KDR 0.0026, 0.0572, 0.2496, 0.524, 1.12, 2.32, 4.8, 10 RG-50810 EGFR, PDGFRb 0.2496, 2.32, 4.8, 10 Rigel JAK Compound II JAK1, JAK2, JAK3 0.1196 Rigosertib PLK1 0.028 Ruboxistaurin PKCb 0.524, 1.12, 2.32 RWJ-67657 p38a, p38b 0.028 Ryuvidine CDK4/Cyclin D1 0.013, 0.028, 0.0572, 0.1196, 0.2496, 0.524, 1.12 Saracatinib ABL1, SRC, KIT 1.12, 2.32 SB 216763 GSK3 10 SB 431542 TGFbR1 10 SB-202190 p38 MAPK 0.0052, 10 SB-218078 CHK1, CDC2(p34cdc2)/Cyclin B, PKC 4.8 SB-415286 GSK3 1.12, 4.8, 10 SB590885 BRAF 0.1196, 1.12, 2.32, 4.8, 10 SC-514 IKKb 10 SD 208 TGFbR1 10 Seliciclib CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 10 CDK2/Cyclin E, CDK5/p35 Semaxanib FLT3, MET, KDR 1.12, 4.8, 10 SGI-1776 PIM1, PIM2 4.8 SGX-523 MET 0.0026, 0.013, 1.12, 2.32, 4.8, 10 SIS3 TGFbR1 1.12, 4.8, 10 SKF-86002 dihydrochloride p38 MAPK 10 171

SL327 MAP2K1, MAP2K2 10 SNS-032 CDK2/Cyclin E, CDK7, CDK9/Cyclin T1 0.0026, 0.013, 0.0572 SNS-314 AurA, AurB, AurC 10 SP-600125 JNK1 10 Sphingosine kinase Inhibitor 2 SPK1 2.32 SR3677 ROCK2 0.2496, 10 Src I1 SRC, LCK (p56lck), KDR 0.2496, 0.524, 10 ST638 Tyrosine kinase inhibitor 10 SU 16f PDGFRb, KDR 0.524, 1.12, 2.32, 4.8, 10 SU 4312 KDR 0.2496, 4.8, 10 SU 6656 SRC, YES, LYN, FYN (p59fyn) 1.12, 2.32, 4.8 SU-11274 MET 2.32, 4.8 SU-5402 KDR, FGFR1, PDGFRb 0.028, 2.32, 4.8, 10 SU-5607 CK1a 10 SU-6668 PDGFRb, KDR, FGFR1, AurA, AurB, TBK1 0.2496, 4.8, 10 SU-9516 CDC2(p34cdc2)/Cyclin B, CDK2/Cyclin A, 0.2496, 1.12, 10 CDK4/Cyclin D1 Sunitinib Malate KIT, FLT3, FMS, KDR, PDGFRb 0.0572, 0.1196, 0.2496, 1.12, 2.32, 4.8 TAE-684 ALK 0.013, 0.0572, 1.12, 2.32 TAK-715 p38 MAPK 0.0572 Tandutinib FLT3, KIT, PDGFRb 4.8, 10 TBB, NSC 231634 CK2a1 0.1196, 10 TCS 2312 dihydrochloride CHK1 0.2496, 0.524, 1.12 TCS 359 FLT3 2.32 TG003 CLK1, CLK4 0.0052, 0.028, 0.1196, 0.2496 TG-101348 FLT3, JAK2 0.524, 1.12, 2.32 TGX-221 PI3K 0.524, 4.8, 10 Tivantinib MET 2.32 Triciribine AKT1 0.0572 Tyrphostin AG-1296 PDGFRb 0.028 Tyrphostin B44, (-) enantiomer EGFR, ErbB2 4.8, 10 U0126 MAP2K1, MAP2K2, JAK2, Erk2(p38) 0.2496, 0.524, 1.12, 2.32, 4.8, 10 Vemurafenib BRAF 10 VX-702 p38 MAPK 0.0052, 2.32 WHI-P 154 JAK3, EGFR, SRC, ABL1, KDR, p38 MAPK, PI3K 0.013, 0.2496, 0.524, 1.12, 2.32, 4.8, 10 WZ4002 EGFR 0.0052, 0.524, 1.12, 4.8, 10 XL-147 PI3K 0.013, 0.028, 0.0572, 1.12, 10 XMD8-92 Erk5 0.2496, 0.524, 1.12, 2.32, 4.8, 10 YM201636 PIKfyve 0.028, 0.1196, 0.524, 1.12, 2.32, 4.8, 10 ZM 306416 hydrochloride KDR 1.12, 10 ZM 323881 hydrochloride KDR 0.2496, 4.8, 10

172

Table A.2: Synonyms of compound targets in drug library

Target Name Synonyms ABL1 ABL AK Adenosine Kinase AKT1 PKBa AKT2 PKBb ALKMET c-MET, HGFR BMPR1A ALK3 AurA AuroraA AurB AuroraB AXL Ark BCRABL1 ABL BMPR1B ALK6 BRAF B-raf CAMKK2 CAMKK CDC2 CDK1, p34cdc2, CDC28p CK2a1 CK2, CKII, Casein kinase2 EGFR ERBB1, HER1 EphA1 EPH ErbB2 TKR1, HER2, NEU ErbB3 HER3 ErbB4 HER4 Erk5 BMK1, PRKM7 FAK FAK1 FGFR1 FLT2 FLT1 VEGFR1 FLT3 STK1, FLK2 FLT4 VEGFR3 FMS c-FMS, CSF-1R, CSF1R IGF1R JTK13, IGFIR IKKa IKK1 IKKb IKK2 IRE1 IRE1a, ERN1 JAK2Erk2 ERK, p38 JAK3EGFR ERBB1, HER1 JNK1 JNK KDR VEGFR2, VEGFR, FLK1 KIT c-KIT LIMK1 LIMK MAP2K1 MEK1 MAP2K2 MEK2 MAP2K5 MEK5 173

MET c-MET, HGFR PAK1 PAKa PDGFRa PDGFR2 PDGFRb PDGFR, PDGFR1 PDK1 PDPK1 PEK PERK PIKfyve PIP5K p70S6K S6K, S6K1 PKR EIF2AK1, EIF2AK2 PLK1 PLK, STPK13 PYK2 FAK2, PTK2B, CAKB RAF1 c-Raf RIPK1 RIP RSK1 RSK, S6Ka, MAPKAPK1C smMLCK MLCK, myosin light chain kinase, MYLK SPK1 SPHK, SK1, Sphingosine Kinase SRC c-SRC TGFbR1 ALK5 TRKA TRK TTK ESK, MPS1L1, Monopolar spindle 1 (Mps1) kinase YESErk2 ERK, p38

174

Appendix II: Supplemental information for Chapter 4

Table A.3: Top 80 differentially expressed genes in identified clusters identified in sc-RNA seq

EC-1 EC-2 MuSC FAP-1 FAP-2 FAP-3 Tenocytes Schwann cells Aqp1 Rbp7 Chodl Smoc2 Pcolce2 Spp1 Cilp2 Kcna1 Timp4 Stmn2 Myf5 Ccl11 Tmem100 Shisa3 Tnmd Gpr37l1 Gpihbp1 8430408G22Rik Des Cxcl14 Fn1 Tenm2 Ptx4 Plp1 Tspan13 Sema3g Sdc4 Crispld2 Cd248 Cpe Kera Sox10 Egfl7 Tm4sf1 Pax7 Lum Stmn4 6030408B16Rik Thbs4 Gatm Rgcc Sox17 Fgfr4 Hsd11b1 Efhd1 Vit Col11a1 Fabp7 Flt1 Alpl Gnas Dpep1 Cadm3 Dlk1 Scx Cmtm5 Cdh5 Icam2 Cd82 Htra3 Aspn Gfra2 Fmod Col20a1 Kdr Nepn Rps18 Dcn Sema3c Cp Tnc Fxyd3 Cd36 S1pr1 Rps5 Col6a1 Igfbp6 Angptl7 Mfap4 Mbp Sdpr Hey1 Rpl23 Col6a2 Ecm1 Abca6 Col12a1 Chl1 Gimap6 Eps8l2 Myod1 Col15a1 Col14a1 Gpc3 Cilp Erbb3 Fabp4 Mmrn2 Rps6 Clec3b C3 Bace2 Mkx Acsbg1 Tmsb4x Vegfc Asb5 Gsn Mfap5 Rasgrp2 Chad Bche C1qtnf9 Grrp1 Rps4x Rnase4 Dpp4 Apod Comp Cadm1 Esam Mecom Rps19 Podn Efemp1 Matn2 1500015O10Rik Mpz 1810011O10Rik Slc9a3r2 Rpl13 Col6a3 Pi16 Gstm5 Ptn Mal Cav2 Edn1 Rplp2 Col3a1 Ahnak2 Fam213a Ccdc3 Gjc3 Cav1 Pecam1 Rpl17 Steap4 Clec3b Il33 Col8a2 Gas2l3 Ctla2a Ly6c1 Rpl6 Plxdc2 Myoc Hk2 Wif1 Cpm Emcn Tinagl1 Rpl37 Pcolce Fst Kcnk2 Emb Csmd1 Ccdc85a Ptprr Rpl13a Itih5 Serpine2 Olfml3 Cpxm2 Nrn1 Gng11 Cldn5 Rps28 Mmp2 Anxa1 P2ry14 Gas2 Sfrp5 Cd300lg F11r Rpl39 Nmb Fstl1 Lum Col13a1 Nrxn1 Adgrl4 Fabp4 Rpl26 Rarres2 Ifi27l2a Dcn Insc Mamdc2 Fabp5 Ly6e Rpl14 Serping1 Timp2 Srpx Pdgfa Plekhb1 Ly6c1 Serinc3 Rpl18a Ms4a4d Procr Cpxm2 Wisp1 Pou3f1 Tcf15 Sox18 Rps23 G0s2 Tmeff2 Tgfbi Fibin Cadm4 Pecam1 Cd36 Rps8 Oaf Serping1 Ecm1 Pamr1 Art3 Tspan7 Cdh5 Eef1a1 Cd302 Fbln1 Cpxm1 Fam180a Postn Mllt4 Gata2 Rpl21 Angptl1 Prg4 Efemp1 Cntfr S100b Srgn Podxl Rps7 Entpd2 Igfbp5 Sdc3 Ppfibp2 Ptn Cldn5 Tm6sf1 Errfi1 Nupr1 S100a6 Rab20 Angptl7 Etv1 Car4 Aqp7 Gpx3 Enpp2 Col1a1 Smox Itgbl1 Cnp Slc9a3r2 Pdgfd Rps3 Cyb5a Zfp385a Aspa Fbln1 Lgi4 Nrp1 H2-D1 Rpl36a Vtn Scara5 Tmem100 P3h2 Lyz2 Cyyr1 Flt1 Rpl9 Serpinf1 Dpt Mmp19 Ptgis Cdh19 Serinc3 Tsc22d1 Rps14 Col5a3 Col3a1 Phgdh Tns3 Insc

175

Adgrf5 Esam Rps15a Col1a2 Ccdc80 Ndrg2 Col1a1 Stmn1 Calm1 Tspan13 Rpl38 Cygb Cd34 Enpp2 Hopx Cyp2j9 Rasip1 Epas1 Edn3 C1ra Rnase4 Smoc2 Spon2 Egfl8 Iigp1 Tmem88 Rpl4 Bgn Qpct Csrp1 Ctgf L1cam Myct1 Cav1 Rps27 Prss23 Emilin2 Entpd2 Gnai1 Cd59a Kcna5 Igfbp3 Rpl18 Ifi205 Penk Foxs1 Prg4 Gpm6b Cd200 Slc6a6 Rps21 Abca8a Mustn1 Vwa1 Col1a2 Matn2 Aqp7 Ptprb Rpl12 Fbln2 Fth1 Col18a1 Aebp1 Matn4 Sox18 Ly6a Rps3a1 Sdc2 Gm9780 Sox9 Timp1 Gng2 Cd93 Crip1 Rpl3 Col1a1 Gfpt2 Col8a1 Matn4 Gas7 Jam2 B2m Rps9 Dpt Lrrn4cl Col1a2 Serpinf1 Csrp1 Ushbp1 Srgn Rpl10 Mme Nbl1 Ogfrl1 Ptprd Sfrp1 Slc28a2 Parvb Rpl32 C1s1 Fbn1 Fam20c Lpar1 Klk8 Klf2 Sema7a Rpl11 Mnda Has1 Plin3 Thbs2 Cyp2j6 Hspb1 Eng Rplp0 Inmt Ifi204 Laptm4b Abi3bp Prnp Car8 Uaca Chrdl2 Col4a1 Serpinf1 Col1a1 Mgp Ldhb Gm12002 Prex2 Rpl10a Cmklr1 Scara3 Atf5 Dpysl3 3632451O06Rik Kank3 Apold1 Cdh15 F3 Gsn Ctsb Clec11a Ncam1 Scarb1 Plk2 Rps27a Nid1 Tnxb C1s1 Lox Metrn Palmd Egfl7 Rpl7 Htra1 Metrnl Gas7 Olfml3 Dbi Lims2 Isg15 Rps16 Islr Ugdh Ccl11 Dnm3os Ccnd1 Ybx1 Klf2 Rps26 Mgst1 Axl Mmp2 Lgals1 Col27a1 Mgll Eva1a Rps13 Mfap5 Cd55 Gfra1 Pcsk5 Nr4a2 Arhgap31 Cav2 Gm9493 Ogn C1s1 Ivd Sparc Marcksl1 B2m Ets1 Rpl35a Igfbp4 Col1a2 Slc2a1 Antxr1 Lbh Fli1 Psmb8 Rps2 Mgp Adgrd1 Col15a1 S100a4 Ckb Cdc42ep3 Ucp2 Eef2 Gstt1 Dbn1 F3 S100a6 Limch1 Ecscr Calm1 Rpl8 Col5a2 Ly6a Fez1 Sema3b Negr1 Chek2 Sdpr Crlf1 Ace Anpep Idi1 Meox1 Ccser2 Pde2a Myct1 Rpl23a Cpq Fndc1 Cox6c Aig1 Zeb2 Tie1 Tmsb4x Runx1 Fap Pcolce Slc29a1 Aspn Pdgfa Kitl Tspan7 Rpl19 Mxra8 Pcsk6 Gprc5b Cd9 Slc22a17 Itga6 Ehd1 Rpsa S100a6 Lrp1 Scpep1 Mfap2 Snta1 Apold1 Cyyr1 Rpl37a Plau Emp3 Col3a1 Fxyd6 Fam134b Tinagl1 Efnb2 Rpl5 Fbln7 Rarres2 Crispld2 Igfbp6 Fibin Sparcl1 Cd93 Wdr89 Bmp5 Thbs3 Sh3kbp1 Mmp2 Pmp22 Lnx1 Id1 Rpl31 Lhfp Lsp1 Bin1 Timp2 Dmd Ptrf Ecscr Mt2 Itm2a Adamts2 Zfyve21 Rgs2 Ldlrad3 F11r Rapgef4 Trib1 Prelp Lamb2 Rgs16 Svil Rhobtb3 Abcg2 Fabp5 Pde4b Ctsh Nupr1 Capn1 Ccdc80 Fzd1 Ptprb Mapk3 F2r Tnxb Entpd2 Vimp Prelp Aspa Adgrg1 Dll4 Dhcr24 Gpm6b Cfh Glul Chodl Taldo1

176

Table A.4: Top 40 differentially expressed genes between EC-1 and EC-2

EC-1 EC-2 Aqp1 Rbp7 Timp4 Fbln5 Lpl Stmn2 Tmsb4x Alpl Gng11 Crip1 Kdr Tm4sf1 1810011O10Rik Hey1 Ccdc85a Sema3g Tshz2 Vegfc Gpihbp1 8430408G22Rik Ctla2a Edn1 Car4 Igfbp4 Igfbp7 Glul Car8 Tsc22d1 Gngt2 Rps27 Gimap6 S100a6 Sult1a1 Eps8l2 Gm12002 Nepn Rgcc Bsg Sdpr Fxyd5 Tmsb10 Cst3 Eepd1 Pdcd4 Kcna5 Rps28 Chek2 Igfbp3 C1qtnf9 Gadd45g Kank3 Rps19 Sgk1 Slc6a6 Nrp1 Foxp1 Ppp1r2 Icam2 Scarb1 Rps23 Fmo1 Ptprr Palmd Trim47 Egfl7 Sox17 Meox2 Grrp1 Jam3 Ebf1 Zfos1 Itm2b Slc28a2 Rpl18a Sepp1 Ucp2 Ppp1r10 Pdgfd 177

Figure 1.1 & Figure 1.4 Reprinted by permission from Dumont et al., 2015b; with permission from John Wiley & Sons Inc. Permission conveyed through Copyright Clearance Centre, Inc.

Figure 1.2 Reprinted by permission from Wosczyna and Rando, 2018; with permission from Elsevier. Permission conveyed through Copyright Clearance Centre, Inc.

178